摘要:

Natural product chemistry 6 Roger Whitehead The Department of Chemistry University of Manchester Oxford Road Manchester UK M13 9PL 1 Introduction The oceans which cover almost 75% of the Earth’s surface contain a variety of species many of which have no terrestrial counterparts. The marine world thus represents an attractive source of novel natural products which prior to 1980 had been largely unexploited. Over recent years technological advances have aided marine isolation chemists and have fuelled a rapidly growing interest in the hidden secrets of the oceans. This has resulted in the puri.cation and patent protection of a large number of marine natural products many with potential applications as novel therapeutic agents. Prior to 1995 a total of 6500 marine natural products had been isolated; by January of 1999 this .gure had risen to approximately 10 000. When compared with the 150 000 natural products obtained from terrestrial plants this is a relatively small number of compounds however an increase of 50% over a period of four years clearly represents an explosion of interest in natural products obtained from the marine environment.The major thrust of the research e.ort has been centred on slow moving or sessile organisms such as sponges coelenterates (e.g. soft corals) ascidians (sea squirts) molluscs (e.g. sea hares) and bryozoans. Many of the members of these phyla are brightly coloured and the absence of a spine or protective shell means that they require an e.ective means of self-defence; it is not surprising therefore that many of them produce unusual biologically active natural products as a means of defence.About one third of all marine natural products have been isolated from sponges making them currently the most popular source of novel compounds. It is the author’s aim to describe a selection of natural products which have been isolated from marine sponges over the preceding two years (January 1997—December 1998). This review will be far from exhaustive and particular attention will be paid to those compounds which in the author’s opinion are noteworthy with regard to their unusual structures biological activities and/or their likely biogenetic origins. 2 Natural products from marine sponges Sponges (Phylum Porifera) are often described as the most primitive of all multicellular animals.They .rst appeared on the Earth almost 600 million years ago and have 183 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 1 undergone very little structural evolution since that time. They are simple cell aggregates which possess no true organs and although they are composed of a variety of di.erent cells these cells display signi.cant independence. Sponges are sessile and apart from approximately 150 freshwater species they are marine animals which are commonly found in seas where there are rocks coral or other suitable substrata. They are mostly found in relatively shallow waters although there are a number of examples which live in water which is thousands of metres deep. Sponges act as hosts to a variety of symbiotic/parasitic organisms including blue—green algae and bacteria.The question therefore arises as to which organisms are the true producers of the structurally diverse compounds isolated by extraction of whole marine sponges. It is quite widely accepted that many such metabolites are produced by symbiotic microorganisms a conclusion supported by the .nding that the natural products from some classes of sponge are very similar to those known from terrestrial microorganisms. i)‘Manzamine-type’ alkaloids Over recent years marine sponges have become recognised as a rich source of complex polycyclic alkaloids which despite possessing quite diverse structural frameworks are believed to be biogenetically derived from similar building blocks ‘ammonia’ C3 ‘acrolein equivalents’ and long chain dialdehydes.The parent members of this group of alkaloids were manzamines A (1) B (2) and C (3) which were originally isolated by Higa and co-workers from an Okinawan sponge Haliclona sp. (Fig. 1). In the original isolation paper, Higa was prompted to comment that the provenance of manzamine A ‘is problematic as there appears to be no obvious biogenetic path’. In 1992 Baldwin put forward a plausible biogenetic link between the highly complex structures of the manzamines and the three simple building blocks. According to this proposal formal reductive condensation of ammonia with an acrolein equivalent and one symmetrical dialdehyde (4) gave aldehyde 5 which then underwent Pictet—Spengler-type cyclisation with tryptophan or tryptamine to give manzamine C (3) (Scheme 1).With regard to the more complex manzamines Baldwin suggested that reductive 184 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Scheme 1 condensation of two ammonias two acrolein equivalents and two C10 dialdehydes would lead to the bis-dihydropyridinium ion 6 (Scheme 2). A single deprotonation then gave the unsymmetrical species 7 which underwent an endo selective intramolecular Diels—Alder reaction to give the pentacyclic iminium ion 8. Subsequent redox reaction of 8 gave the regioisomeric iminium ion 9 which on hydrolytic ring-opening gave tetracyclic aldehyde 10. Pictet—Spengler-type cyclisation with tryptophan or tryptamine followed by epoxidation then gave manzamine B. Manzamine A was viewed as being derived from manzamine B by a trans-eliminative ring-opening of the epoxide followed by allylic oxidation of one of the macrocyclic double bonds and ring-closure to form the eight-membered azacycle.Since the .rst isolation of manzamine A a large number of sponge derived alkaloids have been found which are apparently derived from the same simple building blocks and several of these bear close structural similarities to the intermediates envisaged in the biosynthetic proposal. In particular ircinal A (11), which was isolated from an Okinawan sponge Ircinia sp. is a compound very similar to aldehyde 10 and keramaphidin B (12), isolated from another Okinawan sponge Amphimedon sp. is the reduced form of cycloadduct 8 (Fig. 2).A number of articles have appeared recently reviewing the recent synthetic and biosynthetic literature on this unusual class of natural products.— The last two years have seen a continuation of interest in manzamine-type alkaloids. Three new manzamine congeners (13—15) from an Okinawan marine sponge Amphimedon sp. were reported by the Kobayashi group (Fig. 3). Manzamine M (13) 3,4-dihydromanzamine J (14) and 3,4-dihydro-6-hydroxymanzamine J (15) all displayed antibacterial activity as well as signi.cant cytotoxicity against murine leukaemia L1210 cells (IC 1.4 0.5 and 0.3 gml respectively). ManzamineMis particularly noteworthy in that it is the only manzamine congener bearing a hydroxy group on the C13—C20 chain. Kobayashi and collaborators have also reported the isolation of Ma’eganedin A (16) from an Okinawan sponge Amphimedon sp.(Fig. 4). This compound is a tetrahydro- -carboline alkaloid with a similar core structure to manzamine B but also possessing the unusual structural features of a methylene bridge between N-2 and N-27 and also a C-11 C-12 vicinal cis-diol. Ma’eganedin A exhibited antibacterial activity against a 185 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Scheme 2 186 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 2 Fig. 3 range of bacteria (MIC values between 2.8 and 5.7 gml) as well as cytotoxicity against murine leukaemia L1210 cells (IC 4.4 gml). A number of manzamine-related alkaloids possessing polycyclic N-containing ring systems but lacking a -carboline unit have also been isolated.Nakadomorin A (17) an unusual alkaloid possessing an ‘unprecedented 8/5/5/5/15/6 ring system’ and incorporating a furan ring was isolated by the Kobayashi group from an Okinawan sponge Amphimedon sp. (Fig. 4). This compound was shown to be cytotoxic against murine leukaemia L1210 cells (IC 1.3 gml) as well as to possess weak antimicrobial activity and inhibitory activity against cyclin dependent kinase 4 (IC 9.9 gml). A plausible biogenesis for nakadomorin A was proposed commencing with ring-opening of the known marine alkaloid ircinal A (11) (Scheme 3). Misenine a novel polycyclic ‘cage-like’ alkaloid was isolated by Cimino and coworkers from a Mediterranean sponge Reniera sp.collected o. Capo Miseno in the Bay of Naples (Fig. 5). The H NMR spectrum of this unusual alkaloid showed signi.cant variations with pH and it was concluded that the dominant species in neutral and basic solutions was 18a whereas under acidic conditions 18b was preferred. A similar transannular N/C——O ‘proximity e.ect’ had previously been observed in saraine A (19) although in this case a lowering of pH enhanced the C—N linkage. 187 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Scheme 3 Fig. 4 Unfortunately it was not possible to deduce the individual lengths of the alkyl side-chains and the relative location of six methylenes still remains unclear. Although misenine appears to be structurally related to the ircinals the unprecedented skeleton prompted Cimino to comment that ‘there is no easy way to explain the biogenetic origin of 18a/18b by analogy with that of ircinal A’.The authors did however put forward a hypothetical biosynthetic pathway which despite lacking 188 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 5 precise mechanistic details invoked many of the principles proposed for the manzamines. The madangamines (Fig. 6) despite lacking any direct resemblance to the manzamines are believed to be derived from the same simple building blocks. They constitute a series of highly complex relatively non-polar polycyclic alkaloids characterised by the key feature of two long alkyl chains possessing varying degrees of unsaturation. Madangamine A (20) was the .rst to be isolated by Andersen and collaborators from a sponge Xestospongia sp.collected o. Madang Papua New Guinea. Since its initial isolation this compound has stimulated interest, not only due to its unprecedented structure but also its in vitro cytotoxicity towards a range of cancer cell lines including murine leukaemia P388 and breast MCF-7 (ED 0.93 gml and 5.7 gml respectively). In 1998 Andersen reported a further four members of this series of pentacyclic alkaloids madangamines B—E (21—24). Unfortunately however no data regarding the biological activity of these new compounds have been reported. Interestingly all .ve of the madangamines isolated to date have identical N-1 to C-3 bridges but di.er in the length and level of unsaturation of the N-7 to C-9 link.Andersen has proposed that the madangamine skeleton may be derived via rearrangement of an intermediate structurally similar to the reduced form of the pentacyclic cycloadduct envisaged in the manzamine biosynthetic proposal (i.e. keramaphidin B-like). Thus reduced cycloadduct 25 undergoes fragmentation to give tetracyclic iminium ion 26 which is proposed to be in equilibrium with regioisomeric iminium ion 27. Subsequent ‘aza-Prins-type’ reaction leads to the madangamine skeleton (Scheme 4). It was also suggested that the enzyme(s) responsible for catalysing the rearrangement may display speci.city for a particular chain length thus explaining the lack of variation in length of the N-1 to C-3 bridge in all of the madangamines. 189 Annu.Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 6 Scheme 4 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 190 Scheme 5 A number of simple azamacrocycles have been isolated from marine sponges which are believed to be derived from the same simple biosynthetic building blocks as the ‘manzamine-type’ alkaloids already described. The .rst of these was keramaphidin C (28) (6Z-azacycloundecene) which is the likely biogenetic precursor to manzamine C. In 1998 the isolation of haliclorensin (29) from a sponge Haliclona sp. collected at Sodwana Bay South Africa was reported. The crude extract of the sponge exhibited strong cytotoxicity against murine leukaemia P388 cells although no biological activity was reported for the puri.ed compound. Haliclorensin is a novel diamino alkaloid possessing a methylated azacyclodecane ring which is believed to be derived via a similar biosynthetic route to manzamine C with ammonia (or its biosynthetic equivalent) replacing tryptophan/tryptamine (Scheme 5).of A related series of alkaloids have been isolated by Andersen and collaborators from a sponge Xestospongia exigua collected in Papua New Guinea. Motuporamines A—C (30—32) were isolated as an inseparable mixture which exhibited modest cytotoxicity against a range of human solid tumour cancer cells with a mean IC 0.6 gml. Treatment of the mixture of alkaloids with acetic anhydride gave the corresponding diacetylated derivatives which could be separated by reversed-phase HPLC thus facilitating structural elucidation.MotuporamineC was the only member of the series possessing unsaturation in the azacycle however it was not possible to distinguish between and locations of the double bond. A plausible biogenesis was put forward which utilized the same simple precursors involved in manzamine biosynthesis with two acrolein equivalents being required for formation of the ‘spermidine-like’ side-chain (Scheme 6). ii)Halogen-containing metabolites Marine sponges are a rich source of highly halogenated compounds. Despite the relative concentrations of chloride bromide and iodide ions in sea water (559 mM 0.86mM and 0.45 M respectively) there exists a marked predominance of bromine containing metabolites. It is believed that this is due to the greater ease with which bromide ions are oxidised to give reactive bromonium species which react readily as electrophiles with unsaturated species.The last two years have seen signi.cant interest in halogen containing heteroaromatic natural products which are a particularly common class of sponge-derived secondary metabolites. In particular brominated pyrroles have been isolated on 191 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Scheme 6 several occasions as major constituents of marine sponges. Many of the simpler members of this class of compounds are structurally related as they comprise two heterocyclic rings linked by a linear chain. In 1997 for example Kobayashi and collaborators reported the isolation and structural elucidation of tauroacidins A (33) and B (34) two novel taurine residue containing bromopyrrole alkaloids from an Okinawan sponge Hymeniacidon sp.(Fig. 7). Degradation studies showed that tauroacidin A was a 3 2 mixture of 9S and 9R isomers whereas tauroacidin B was isolated as a racemic mixture. Both alkaloids were reported to show inhibitory activity against EGF receptor kinase and c-erbB-2 kinase (IC 20gml each). In the same year Fattorusso and co-workers reported two new bromopyrroles dispacamides C (41) and D (42) together with oroidin (35) agelongine (36) clathramides (37) dispacamides A (38) and B (39) and sceptrin (40) (Fig. 8). These structurally related alkaloids were isolated from four Caribbean sponges of the genus Agelas. Both of the new alkaloids were isolated as racemic mixtures and the only feature di.erentiating 42 from 41 was the absence of a bromine atom at C-2 of the pyrrole ring.Prior to this report bromopyrrole alkaloids had been subjected to a variety of 192 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 7 Fig. 8 biological activity tests however they had not been assessed for antihistamine activity. All eight of the compounds isolated from the Agelas sponges were therefore assessed for activity on histamine-induced contractions of guinea pig isolated ileum. Only agelongine (36) and clathramides (37) were found to be inactive towards the histamine receptor even at millimolar concentrations whereas dispacamide A (38) was found to be the most potent non-competitive antagonist (pD 5.52).The presence of a hydroxy group on the linking chain in dispacamides C (41) and D (42) resulted in a decrease in activity (pD 4.48 and 4.34 respectively) indicating that both the length and functionality of the linking chain are important for antihistamine activity. In 1998 Fattorusso reported the isolation of clathramides C (43) and D (44) the N-14 nor-derivatives of clathramides A and B (37) from the Caribbean marine sponge 193 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 9 Agelas dispar collected in the lagoon of Little San Salvador Island Bahamas (Fig. 9). These compounds displayed no antibacterial activity but similar antifungal activity to their methylated counterparts against Aspergillus niger. The cyclic bromopyrrole alkaloid longamide B (45) was also isolated as a racemic mixture from this sponge.In contrast to the clathramides this compound showed weak antibacterial activity against a range of Gram-positive bacteria. Longamide B is the free acid analogue of hanishin (46) which was isolated in 1997 by Pietra and collaborators from an axinellid sponge Acanthella carteri collected from the northern coast of the Hanish Islands Yemen. This latter compound exhibited in vitro activity against NSCLC-N6 human non-small-cell-lung carcinoma (IC 9.7 gml). A plausible biogenesis for hanishin (and therefore longamide B) was put forward whereby cyclisation of an ‘oroidin-like’ aminoimidazolinone from N-1 to C-9 followed by oxidative breakdown of the side-chain gives the cyclic alkaloids (Fig.9). A number of more complicated polycyclic bromopyrrole alkaloids have recently been isolated. The spongiacidins A—D (47—50) are a series of pyrrolo[2,3-c]azepinetype alkaloids which were reported in 1998 by Kobayashi and collaborators (Fig. 10). These compounds isolated from an Okinawan sponge Hymeniacidon sp. can be viewed as cyclic analogues of the bromopyrroles described above and bear close structural similarities to the previously isolated hymenaldisines (51) and (52) and axinohydantoin (53). Spongiacidins A (47) and B (48) which both possess (7E)- geometry displayed inhibitory activity against c-erbB-2 kinase (IC 8.5 and 6.0 gml respectively) and cyclin-dependent kinase 4 (IC 32 and 12 gml respectively).Spongiacidins C (49) and D (50) which both possess (7Z)-geometry and have a hydantoin ring in place of the aminoimidazolinone ring of 47 and 48 showed no inhibition of either kinase. The agelastatins are even more complicated bromopyrrole alkaloids which possess a unique fused tetracyclic skeleton which may formally be derived from an open chain pyrroloaminopropylimidazole precursor similar in structure to oroidin. In 1998 194 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 10 Fig. 11 Molinski and collaborators reported the isolation of the known agelastatin A (54) together with two new alkaloids agelastatinsC (55) andD(56) from a West Australian sponge Cymbastela sp. collected at Muiron Island (Fig 11). It was reported in this article that 54 was highly toxic to brine shrimp with an LC of 1.7ppm (5.0 M) 195 Annu.Rep. Prog. Chem. Sect. B 1999 95 183—205 Scheme 7 whereas 55 was much less toxic (LC 220 ppm). Agelastatin A was also reported to be insecticidal with a biological activity comparable to a commercial preparation of the biopesticide Bacillus thuringiensis. The palau’amines are a series of complex halogen-containing alkaloids which possess six contiguous rings and an unbroken chain of eight chiral centres. In 1998 Scheuer and co-workers presented a full account of the isolation of palau’amine (57) and its 4-bromo (58) and 4,5-dibromo (59) derivatives as well as the three previously reported ring A regioisomers (styloguanidines 60—62) from a sponge Stylotella aurantium collected near Wonder Channel and Rock Islands Republic of Bellau. Palau’amine itself was reported to display a remarkable range of biological activities with an IC 18 ng ml in the mixed lymphocyte reaction and IC ’s of 0.1 and 0.2 gml against P-388 and A-549 cell lines respectively.In contrast the brominated analogues (58 and 59) showed no comparable activity although the 4,5-dibromo compound (59) was selective against a human melanoma cell line with an IC 0.25 gml. The impressive biological pro.le of 57 is re.ected in the fact that preclinical studies are currently in progress into the anti-fungal anti-tumour and immunosuppressive properties of this compound. of The complex structure of 57 prompted the authors to comment that ‘the biogenesis of palau’amine is obscure’.A plausible biosynthetic pathway was put forward however involving combination of pyrrole-2-carboxylic acid (63) and 3-amino-1-(2- aminoimidazol-4-yl)prop-1-ene (AAPE 64) (both of which are known constituents of 196 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 12 marine sponges) to give a 11,12-dehydrophakellin (65). Subsequent Diels—Alder reaction with a second equivalent of 64 followed by electrophilic chlorination bond migration and reaction with water furnished the palau’amines. In addition to bromopyrroles halogen-containing indoles have also been encountered as sponge metabolites. In 1998 Capon and collaborators reported the isolation of the known alkaloid dragmacidin D (66) together with two new bis-indole alkaloids (67 and 68) from a sponge Spongosorites sp.collected at a depth of 90malong the coast of South Australia (Fig. 12). Dragmacidin E (67) is a .uorescent yellow pigment apparently derived via cyclisation of a dragmacidinD precursor. It remains unclear as to whether this compound exists exclusively in the pyrazine form or alternatively in tautomeric equilibrium with the pyrazinone form. The aqueous ethanol extract of the deep-water marine sponge was found to possess antibiotic properties as well as to inhibit serine-threonine protein phosphatases and it was shown that compounds 67 197 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 and 68 were solely responsible for this biological activity. Interestingly dragmacidin E was found to inhibit both PP1 and PP2A whereas initial results suggested that dragmacidin D was a selective inhibitor of PP1.iii)Polyacetylenes Acetylenes are quite common components of terrestrial plants. It is only over the past 20 years or so however that polyacetylenes have been reported from marine organisms. Long-chain polyacetylenes represent a rapidly growing group of marine sponge metabolites which display signi.cant structural diversity with regard to their chainlengths and functionalities. A number of examples of this class of compounds which have been found in relatively few families of marine sponges possess interesting biological properties including anti-microbial anti-tumour and enzyme-inhibitor activities. In 1998 Shin and collaborators reported the isolation of seven novel long-chain polyacetylenes from a sponge of the genus Petrosia collected o.Keomun Island Korea (Fig. 13). The petrocortynes 69—71 represent a series of C46 linear polyacetylenes. Petrocortyne A (69) and its 43,44-dihydro derivative petrocortyne B (70) possess very similar structures to the petroformynes. Petrocortyne C (71) is quite unusual in that it possesses a -pyrone ring and it was suggested that this structural feature may be biogenetically derived via oxidative cyclisation of the diacetylenic carbinol group of petrocortyne A (69). Perhaps surprisingly the petrocortynes were not toxic against brine-shrimp larvae which is in contrast to the reported toxicity of the petroformynes. The four remaining novel components of the sponge were the linear C polyacetylenes petrosiacetylenes A—D (72—75) which in contrast to the petrocortynes display a high level of symmetry.Compounds 72 and 74 exhibited signi.cant toxicity against brine-shrimp larvae (LC ’s 0.22 and 19.9ppm respectively) whereas 73 and 75 were not toxic. Furthermore in an assay against RNA-based reverse transcriptase 72 and 74 totally cleaved the template 16S rRNA from E. coli at a concentration of 20 g/20 l whereas 73 and 75 did not cleave at all. None of the polyacetylenes cleaved a super coiled DNA (pUC 119) at the same concentration leading to the suggestion that the brine-shrimp lethality resulted from the RNA cleaving activity. In 1998 Jung and co-workers reported that the crude extract of a marine sponge genus Petrosia collected o.Komun Island Korea delivered signi.cant lethality to brine shrimp. Guided by biological activity four new polyacetylenes were isolated which were given the trivial names dideoxypetrosynols A—D (76—79) (Fig. 14). Compounds 76 and 78 are essentially identical to petrosiacetylenes A (72) and B (73) the only di.erence being the ill-de.ned absolute stereochemistry at C3 and C28 in the former. All four dideoxypetrosynols showed cytotoxicity against human tumour cells with 76 being the most potent in all cell lines tested. Interestingly the polyacetylenes displayed quite selective cytotoxicity against SK-OV-3 and SK-MEL-2 cells. The crude organic extract of the sponge Haliclona osiris collected from Guam was also found to exhibit potent toxicity against brine shrimp larvae (LC 52 ppm). Bioassay guided fractionation led to the isolation of six new highly oxygenated C linear polyacetylenes osirisynes A—F (80—85) (Fig.15). Highly functionalised polyacetylenes of this type are quite rarely encountered 198 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 13 particularly those containing an -acetylenic carboxylic acid moiety. All six compounds exhibited moderate cytotoxicity against a human leukaemia cell line (K562 LC ’s 25 48 52 25 20 and 22 M respectively) and in addition 82 84 and 85 exhibited inhibitory activity against Na/K-ATPase and reverse transcriptase at concentrations of 1 g/10 l. Finally Pietra and collaborators have reported the isolation of two methyl branched polyacetylenes azte` quynols A (86) and B (87) from a sponge Petrosia sp.collected at the top of the sea mounts on the Banc Azte` que o. New Caledonia (Fig. 16). Initially precise location of the methyl group was not possible however the structure of azte` quynol A (86) was .nally solved with the aid of high energy collisionally-activated decomposition tandem mass spectrometry. Unfortunately it was not possible to con.rm the structure of azte` quynol B (87) with respect to the position of methyl branching and absolute con.guration however the co-occurrence with 86 would suggest the most likely structure to be as depicted. These compounds are particularly noteworthy as methyl branched polyacetylenes are quite rare.The only 199 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 14 other examples of related structures from marine sponges were reported in 1996 by Faulkner and collaborators from a sponge Haliclona sp. iv)Amino acid derivatives Marine sponges are a rich source of biologically active metabolites derived from uncommon amino acids which often occur in the form of cyclic peptides. Over the last two years a number of these compounds have been isolated from which the author has selected three examples. The rare sponge genus Theonella has given rise to a large number of peptides. In 1997 Higa and Scheuer reported the isolation of a novel cyclic heptapeptide named cupolamide A (88) from two samples of this sponge collected in Indonesia and Okinawa (Fig.17). The cyclic peptide contains three uncommon amino acid resi- 200 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 Fig. 16 Fig. 15 dues -homoarginine (-Har) trans-4-hydroxy-proline (-Hyp) and -2,4-diaminobutanoic acid (-Dba). The resemblance of the latter to -aminobutanoic acid (GABA) prompted the suggestion that cupolamide A may possess central nervous system activity however no relevant biological activity data were provided. The cyclic peptide was shown to be cytotoxic with an IC of 7.5 gml against murine leukaemia P388 cells. In the same year Kashman and co-workers reported the isolation of oriamide (89) from a marine sponge Theonella sp. collected in Sodwana Bay. This cytotoxic cyclic peptide contains the unprecedented amino acid 4-propenoyl-2-tyrosylthiazole which is similar to a related thiazole amino acid previously reported as a constituent of the cyclic peptide keramamide F (90).Although sponges of the family Thorectidae are recognised as containing a large 201 Annu. Rep. Prog. Chem. Sect. B 1999 95 183¡X205Fig. 17 Fig. 18 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 202 Fig. 19 Fig. 20 number of terpenoids they have recently been recognised as a source of unusual N-containing compounds. In 1997 Crews and collaborators reported the isolation and cytotoxic properties of cyclocinamide A (91) a minor constituent of a marine sponge Psammocinia sp. collected in Papua New Guinea (Fig. 19). The structure of this cyclic peptide is quite uncommon possessing a tetrapeptide core which comprises a 14-membered ring and an unusual dipeptide side-chain terminating in an N-methylchloropyrrole derived from proline.Cyclocinamide A was reported to display ¡¥robust tumour selectivity¡¦ against colon-38 tumour cells. In general it has been observed that free amino acids do not occur at very high concentrations in sponge extracts although two recent exceptions to this are the new lysine derivative 92 and the proline derivative 93 (Fig. 20). The cyclic amino acid derivative 92 was isolated as a major non-toxic component of the methanol extracts of a marine sponge Axinyssa terpnis de Laubenfels 1954 collected in Micronesia. The new amino acid cis-3-hydroxy-N-methyl-proline (93) was found at high levels in an extract of the southern Australian marine sponge Dendrilla sp. trans-3-Hydroxy-proline has been detected from a variety of sources however the cis-isomer is much less common with the antibiotic telomycin being the only known natural substance to incorporate this substructure.Interestingly although the synthetic ¡¥betaine¡¦ derivative 94 was found to inhibit acetyl cholinesterase the natural product was inactive and consequently ¡¥the ecological signicance if any of high concentrations of 93 in the sponge Dendrilla sp. remains unknown¡¦. During the course of investigations into the neuroexcitatory properties of extracts from Micronesian marine organisms it was found that the water extracts of the sponge Dysidea herbaceae induced toxic chronic convulsions in mice. These symptoms resemble those induced by the algal and microalgal metabolites kainic acid (95) and 203 Annu. Rep. Prog. Chem.Sect. B 1999 95 183¡X205Fig. 21 domoic acid (96) (Fig. 21). The principal active component dysiherbaine (97) was shown to be a novel diamino carboxylic acid and possesses a fused hexahydrofuro[ 3,2-b]pyran ring with an alanyl side-chain. Dysiherbaine is a representative of a new class of amino acids with a unique core skeleton. It was suggested by the isolation chemists that the glutamate structural unit C1—C2—C3—C4—C11 may be an important contributor to the compound’s neurological activity. Investigations into the inhibitory e.ects on rat brain synaptic membrane led to the conclusion that dysiherbaine is a selective inhibitor of non-NMDA type glutamate receptors in the CNS. It was also shown that dysiherbaine displayed no cytotoxic activity against leukaemia L1210 cells or antimicrobial activity against Bacillus subtilis.3 Conclusion The marine environment is an extremely rich source of unusual biologically active natural products. Marine sponges the most primitive of all multicellular animals are probably the most commonly exploited of all marine organisms with respect to the provision of novel structurally diverse compounds many of which have no de.nitive biogenetic origin. Throughout this review it has been the author’s intent to describe a number of recently isolated natural products which serve to exemplify the huge variety of intriguing structures to be isolated from this natural source and to emphasise the range of biological properties possessed by these compounds. Unfortunately it has not been possible to describe any of the isoprenoid or polyketide derived materials that have been reported over the last two years although many of these have equally fascinating structures and associated biological activities.No doubt over the years to come marine organisms and in particular marine sponges will continue to provide us with an expanding selection of fascinating compounds of interest to biologists synthetic chemists and isolation scientists alike. 204 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205 References 1 M. Jaspars Chem. Ind. 1999 51. 2 R. Sakai T. Higa C. W. Je.ord and G. Bernardinelli. J. Am. Chem. Soc. 1986 108 6404. 3 R. Sakai S. Kohmoto T. Higa C. W. Je.ord and G. Bernardinelli Tetrahedron Lett. 1987 28 5493. 4 J.E. Baldwin and R.C. Whitehead Tetrahedron Lett. 1992 33 2059. 5 K. Kondo H. Shigemori Y. Kikuchi M. Ishibashi T. Sasaki and J. Kobayashi J. Org. Chem. 1992 57 2480. 6 J. Kobayashi M. Tsuda N. Kawasakin K. Matsumoto and T. Adachi Tetrahedron Lett. 1994 35 4383. 7 M. Tsuda and J. Kobayashi Heterocycles 1997 46 765. 8 J. Kobayashi and M. Tsuda J. Synth. Org. Chem. Jpn. 1997 55 1114. 9 E. Magnier and Y. Langlois Tetrahedron 1998 54 6201. 10 N. Matzanke R. J. Gregg and S. M. Weinreb Org. Prep. Proc. Int. 1998 30 1. 11 D. Watanabe M. Tsuda and J. Kobayashi J. Nat. Prod. 1998 61 689. 12 M. Tsuda D. Watanabe and J. Kobayashi Tetrahedron Lett. 1998 39 1207. 13 J. Kobayashi D. Watanabe N. Kawasaki and M. Tsuda J. Org. Chem. 1997 62 9236. 14 Y. Guo E. Trivellone G. Scognamiglio and G.Cimino Tetrahedron 1998 54 541. 15 G. Cimino G. Scognamiglio A. Spinella and E. Trivellone J. Nat. Prod. 1990 53 1519. 16 F. Kong R. J. Andersen and T. M. Allen J. Am. Chem. Soc. 1994 116 6007. 17 N. Matzanke R. J. Gregg S. M. Weinreband M. Parvez J. Org. Chem. 1997 62 1920. 18 F. Kong E. I. Graziani and R. J. Andersen J. Nat. Prod. 1998 61 267. 19 M. Tsuda N. Kawasaki and J. Kobayashi Tetrahedron Lett. 1994 35 4387. 20 G. Koren-Goldschlager Y. Kashman and M. Schleyer J. Nat. Prod. 1998 61 282. 21 D. E. Williams P. Lassota and R. J. Andersen J. Org. Chem. 1998 63 4838. 22 D. J. Faulkner Chem. Br. 1995 680. 23 J. Kobayashi K. Inaba and M. Tsuda Tetrahedron 1997 53 16 679. 24 F. Ca.eri R. Carnuccio E. Fattorusso O. Taglialatela-Scafati and T.Vallefuoco BioMed. Chem. Lett. 1997 7 2263. 25 F. Ca.eri E. Fattorusso and O. Taglialatela-Scafati J. Nat. Prod. 1998 61 122. 26 I. Mancini G. Guella P. Amade C. Roussakis and F. Pietra Tetrahedron Lett. 1997 38 6271. 27 K. Inaba H. Sato M. Tsuda and J. Kobayashi J. Nat. Prod. 1998 61 693. 28 T.W. Hong D. R. Jimenez and T. F. Molinski J. Nat. Prod. 1998 61 158. 29 R. B. Kinnel H.-P. Gehrken R. Swali G. Skoropowski and P. J. Scheuer J. Org. Chem. 1998 63 3281. 30 R. J. Capon F. Rooney L. M. Murray E. Collins and A. T. R. Sim J. Nat. Prod. 1998 61 660. 31 Y. Seo K.W. Cho J.-R. Rho J. Shin and C. J. Sim Tetrahedron 1998 54 447. 32 Y. Guo M. Gavagnin C. Salierno and G. Cimino J. Nat. Prod. 1998 61 333 and references cited therein. 33 J. S. Kim K. S. Im J. H. Jung Y.-L. Kim J. Kim C. J. Shim and C.-O. Lee Tetrahedron 1998 54 3151. 34 J. Shin Y. Seo K.W. Cho J.-R. Rho and V. J. Paul Tetrahedron 1998 54 8711. 35 A. Guerriero C. Debitus D. Laurent M. D’Ambrosio and F. Pietra Tetrahedron Lett. 1998 39 6395. 36 D. H. Williams and D. J. Faulkner J. Nat. Prod. 1996 59 1099. 37 L. S. Bonnington J. Tanaka T. Higa J. Kimura Y. Yoshimura Y. Nakao W. Y. Yoshida and P. J. Scheuer J. Org. Chem. 1997 62 7765. 38 L. Chill Y. Kashman and M. Schleyer Tetrahedron 1997 53 16 147. 39 W.D. Clark T. Corbett F. Valeriote and P. Crews J. Am. Chem. Soc. 1997 119 9285. 40 C.-J. Li F. J. Schmitz and M. Kelly-Borges J. Nat. Prod. 1998 61 387. 41 R. J. Capon S. P. B. Ovenden and T. Dargaville Aust. J. Chem. 1998 51 169. 42 R. Sakai H. Kamiya M. Murata and K. Shimamoto J. Am. Chem. Soc. 1997 119 4112. 205 Annu. Rep. Prog. Chem. Sect. B 1999 95 183—205

ISSN:0069-3030    

DOI:10.1039/a808577c

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Natural polymers—chemistry 7 Harri Lo� nnberg Department of Chemistry University of Turku FIN-20014 Turku Finland 1 Introduction The very rapid development of molecular and cell biology during the past two decades has challenged organic chemists. The continuously increasing need for chemically synthesized polypeptides polysaccharides and polynucleotides including their structurally modi.ed analogs and organic conjugates has made the chemistry of biopolymers one of the most intensively studied .elds of organic chemistry. Synthetic biopolymers are not however only valuable tools for the molecular level research of cellular events but they also show promise as candidates for next generation drugs. It hence appears quite obvious that natural polymers also in the near future will retain their position among the central research objects of organic chemistry.2 Oligonucleotides General remarks Oligonucleotides o.er a good example of the interplay between molecular biology and chemistry. Above all the antisense oligonucleotide approach i.e. selective inhibition of the expression of a desired gene by a hybridization arrest of the respective mRNA with a synthetic oligonucleotide has greatly stimulated and steered the chemical research of oligonucleotides. But the opposite is also true. The imagination of chemists has created numerous surrogates of biomolecules to be tested by biologists in vitro and in vivo. General protocols of oligonucleotide synthesis The machine-assisted synthesis of oligonucleotides which is based on the stepwise coupling of appropriately protected nucleoside 3-phosphoramidites or hydrogen phosphonates to the 5-terminus of the nascendent resin anchored chain represents such a state-of-the-art process that only a few improvements to the commonly used protocols have been suggested.Nevertheless alternative protocols have been introduced which allow synthesis of exceptionally labile or otherwise uncommon oligonucleotides. 207 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 Protecting groups and coupling. An interesting novel base moiety protection strategy based on the N-pent-4-enoyl group has been introduced. This group is compatible with the normal phosphoramidite protocol when peroxides instead of iodine are used to oxidize the phosphite ester to phosphate ester after each coupling.The advantage over the conventional acyl protections is that ammonolytic deprotection is avoided;N-pent-4-enoyl groups can be removed with iodine under neutral conditions. Another novel base moiety protecting group of special interest is the phenylacetyl group. For the .rst time an enzyme-cleavable amino protecting group has been used in solid phase oligonucleotide synthesis. The group can be removed with penicillin G acylase. Instead of the conventional acid labile DMTr† group a 2-(levulinyloxymethyl)-5- nitrobenzoyl group which can be removed by successive treatments with hydrazine hydrate in acetic acid—pyridine and with imidazole in acetonitrile has successfully been used as the nucleoside 5-hydroxy protecting group in the phosphoramidite synthesis of oligodeoxyribonucleosides (ODN). 4-(Tri.uoroacetylamino)butyl group has in turn been used as the phosphodiester protecting group instead of the conventional 2-cyanoethyl group. 4,5-Dicyanoimidazole has been shown to be an activator for the coupling of phosphoramidite monomers superior to the normally used tetrazole.On assembling short ODNs on a large scale only a catalytic amount of 5-(5-nitrophenyl)-1H-tetrazole promoter can be used without marked diminution of the coupling yield. Another useful methodological improvement is the use of O-(2- cyanoethyl)-O-octyl-N,N-diisopropylphosphoramidite as a capping reagent. Compared to the normally used acetylation with acetic anhydride this approach o.ers the obvious advantage that the lipophilicity of the introduced capping reagent allows easy separation of the capped failure sequences by RP HPLC.Serious attempts towards the assembly of ODNs from monomeric units that do not bear base protection have been made. The amino groups of nucleobases were found not to be modi- .ed during the coupling of H-phosphonate monomers when BOMP was employed as a condensing agent and the internucleotidic H-phosphonate linkages were oxidized with BNO under anhydrous conditions in the presence of N,O-bis(trimethylsilyl)acetamide. Linkers. While the commonly used succinyl linker anchoring the growing ODN chain through the 3-terminal hydroxy group to the support satisfactorily ful.ls all the requirements of the preparation of unmodi.ed ODNs more labile or stable linkers are needed for special purposes.Hydroquinone-O,O-diacetic acid linker allows increased productivity on automated DNA synthesizers because the time of ammonolysis is reduced to minutes. No other modi.cations to the existing reagents or synthesis and deprotection methods are needed. The linker may also be rapidly cleaved with milder reagents such as potassium carbonate or .uoride ion. The 3,4-dichlorophthaloyl linker which also withstands the normal chain assembly can be cleaved by treatment with 10% DBU in acetonitrile for 5 min leaving the acyl protections of the base moieties even cytosine N-acetyl group untouched. Although all the linkers discussed above and many others introduced earlier are very useful for di.erent purposes they share a common shortcoming special tech- † Abbreviations are de.ned at the end of this chapter.208 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 niques have to be applied to attach the .rst nucleoside to the support and hence this step usually has to be carried out manually before inserting the support in the synthesizer. That is why the observation that several uronium (HBTU HATU HBPyU HBPipU) or phosphonium (BOP PyBOP PyBroP BroP) coupling reagents enable attachment of nucleosides to CPG and polystyrene supports via succinyl or hydroquinone-O,O-diacetic acid linkers in seconds is important. Accordingly automated derivatization of the supports just prior to the ODN synthesis is possible. However it would be even more desirable to have a so-called universal linker which would allow attachment of any nucleoside to the support by using the same coupling chemistry that is applied in the subsequent chain assembly.Some serious attempts to ful.l this requirement have been reported. Immobilization of 1,4-anhydro-3-O-chloroacetyl-2-O-(4,4-dimethoxytrityl)-.-ribitol through its primary 5-hydroxy group to LCAA—CPGhas been proposed to a.ord an applicable universal support.A normal detritylation cycle releases the 2-hydroxy function onto which the chain is assembled. The neighboring 3-hydroxy group is released during the ammonolysis and it displaces the 3-O-linked oligonucleotide chain as an intramolecular nucleophile. The underlying principle of the recently introduced Rainbow Universal support is rather similar.cis-3-Hydroxy-4-DMTrO-tetrahydrofuran is tethered via the succinyl linker to the support. The ODN chain is assembled onto the detritylated 4-hydroxy function. Treatment with aqueous lithium hydroxide cleaves .rst the succinyl linker and the released 3-hydroxy function intramolecularly attacks on the 3-terminal phosphate displacing the ODN in the 3-hydroxy form. A novel method for fast deprotection and release of ODNs from the support has been introduced. Treatment with 1M sodium hydroxide in aqueous methanol under microwave irradiation a.orded complete removal of the acyl protections in 4 min without any side-products. While fully automated methods are available for the solid phase synthesis of ODNs no serious attempt at automation of the liquid-phase synthesis has previously been made.Recent demands for preparation of ODNs in large quantities for clinical trials of potential antisense oligonucleotide drugs have however urged the development of solution synthesis. An apparatus has now been described that performs all the steps required by the liquid-phase approach in a fully automated way. Branched and circular oligonucleotides Fully automatedynthesis of branched ODNs has been accomplished by using 2-Olevulinyl protected 1-(-.-arabinofuranosyl)uracil 3-phosphoramidite as the branching monomer. The key step of the procedure is selective removal of the levulinyl group with hydrazine hydrate after complete assembly of the all-3,5-chain. The branching chain is then assembled onto the exposed 2-hydroxy function.2-OTBDMS-adenosine has also been used as the branching monomer. With this approach the 2-cyanoethyl protections are removed from the phosphate backbone by -elimination without detaching the nascendent ODN from the support to allow .uoride ion assisted removal of the 2-O-TBDMS protection. Subsequent assembly of the branch onto the 2-hydroxy group is then carried out in an inverse direction i.e. by using 5-phosphoramidites as the building blocks. Branched DNA/RNA chimera have been obtained analogously by assembling .rst the ribonucleotide sequence from 209 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 2-O-Fpmp protected monomers with the exception of the desired site of branching where the 2-O-TBDMS monomer was inserted. An alternative strategy for the preparation of branched oligoribonucleotides involves assembly of the chain in the 53 direction using 2-O-Fpmp-3-O-pixylribonucleoside 5-(methyl N,N-diisopropylphosphoramidite) s for normal chain elongation and 2-O-pixyl-3-O-levulinylribonucleoside 5-(2-cyanoethyl N,N-diisopropylphosphoramidite) to create the branching points. Closely related strategies have been applied to obtain highly branched dendrimeric ODNs containing ribonucleosides at the branching points, and comb-type ODNsconsisting of one unique ODN the primary sequence covalently attached through a comb-like branch network to many identical copies of a di.erent OND the secondary sequence. Dendrimeric ODNs have also been synthesized using a non-nucleosidic phosphoramidite derived from tris-2,2,2-(3-DMTrOpropoxymethyl) ethanol as the branching synthon. An elegant solid phase method for the preparation of cyclic ODNs ranging from 2- to 30-mers has been reported. An aminoalkyl support was acylated with (3-chloro-4- hydroxyphenyl)acetic acid and the chain was assembled onto the phenolic hydroxy function by the normal phosphoramidite strategy.Treatment of the 5-detritylated chain with MSNT resulted in cyclization and the cyclized product was released from the support with tetramethylguanidinium syn-pyridine-2-aldoximate. Oligonucleotide arrays Synthesis of ODN arrays attached to various kinds of solid supports has gained increasing interest owing to the numerous applications of such arrays in sequencing of DNA and detection of mutations.The photolithographic ODN synthesis on glass supports based on nucleoside 3-phosphoramidites bearing a photo-labile 5-O-(- methyl-2-nitropiperonyloxycarbonyl) or 5-O-dimethoxybenzoincarbonate protecting group has been carefully optimized. The drawback of the photolithographic methods appears to be relatively low stepwise coupling e.ciency (92—96%) resulting from incomplete photochemical deprotection. Attempts have been made to increase the e.ciency by using conventional acid-labile DMTr protection and a polymer support that releases acid upon exposure to light. Pretreatment of glass with boiling aqueous hydrogen chloride has been shown to be a convenient method for the exposure of the hydroxy groups needed for anchoring of the .rst nucleotide in array synthesis. The attachment can be accomplished by phosphite triester chemistry.Preparation of one dimensional ODN arrays on modi.ed polypropylene .lms has been described. Oligonucleotide conjugates Oligonucleotide conjugates have recently found an increasing number of applications as research tools for molecular biology. The conjugate groups are attached to oligonucleotides for many di.erent purposes. Reporter groups allow sensitive detection chemically reactive groups result in selective cross-linking or cleavage intercalators and groove binding agents enhance hybridization and peptides and hydrophobic groups increase cellular uptake. Accordingly synthetic methods for the preparation of oligonucleotide conjugates are of considerable importance.210 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 Conjugation on a support. An increasing number of oligonucleotide conjugates are prepared entirely on a solid support or the conjugate group is introduced upon the deprotection and cleavage of the oligomer from the solid support. For example several solid phase methods for preparation of oligonucleotide 3-phosphates have been described. Conventional phosphoramidite chemistry has been applied to obtainODN 3-phosphates on photo-cleavable 4-(4-hydroxymethyl-2-methoxy-5-nitrophenyloxy) butanoyl and 2-(nitrophenyl)propane-1,3-diol linkers attached to aminoalkylated supports via amide and phosphoramidate diester linkages respectively.3-Phosphates of rather base labile methyl phosphotriester and methyl phosphonate ODNs have been prepared by normal phosphoramidite strategy on a 2,2- bis(ethoxycarbonyl)propane-1,3-diol handle tethered to the aminoalkyl support via a malonyl linker. An allylic 9-hydroxyundec-10-enoyl linker has been used to obtain oligoribonucleotide 3-phosphates. Aliphatic amino mercapto or carboxy functions are most frequently utilized to attach conjugate groups and hence methods for convenient introduction of tethers bearing these functions are of interest. ODN 3-phosphates bearing a phosphatelinked aminoalkyl or a mercaptoalkyl tail have been prepared on supports obtained by modifying aminopropyl-CPG with trimellitic anhydride chloride and reacting the support with an appropriate bifunctional spacer.The desired amino or mercapto functionality hence becomes bonded to the support as an amide or thioester while the DMTr-protected hydroxy group at the other end of the spacer serves as the starting point for the chain elongation. Upon ammonolytic deprotection the 3-terminal functional groups are released. The Rainbow Universal support mentioned above has also been used for the preparation of 3-amino tailed ODN 3-phosphates. Introduction of the amino tail however requires a non-nucleosidic phosphoramidite building block. Novel methods exploiting H-phosphonate coupling have been developed for the synthesis of 5-aminoalkylated ODNs on a solid support. The 5-hydroxy group of the support bound ODN was either phosphonylated with H-pyrophosphonate and condensed with an amino alcohol or phosphonylated with diphenyl H-phosphonate and transesteri.ed with an amino alcohol. ODNs bearing an aminoalkyl tether at any position within the chain have been prepared by introducing 5-methoxycarbonyl- 2-deoxyuridine at a desired position and treating the protected oligomer with alkanediamines when still bound to the support. Alternatively the treatment with alkanediamine may also be used to cleave the oligomer from the support concurrent with the covalent attachment of the diamine to the modi.ed uracil base. CGlycosidic building blocks containing a reactive thioester bond have been used similarly. Several examples of conjugation to amino groups on a solid support have been published.DBUhas been shown to be e.ective in labeling of aliphatic amino functions of support-bound ODNs with isothiocyanate activated reporter groups. A disulfide linker was used to keep the ODN anchored to the support under the conditions of ammonolytic deprotection. Alternatively the N-pent-4-enoyl group may be used to protect the amino functions.As mentioned above this protecting group allows base moiety deprotection under conditions that do not cleave the succinyl linker. Post-synthetic conjugation. The post-synthetic derivatization of oligonucleotides usually involves insertion of a masked functional group most commonly tethered to 211 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 C5 of a pyrimidine or C8 of a purine nucleoside into the ODN by the normal phosphoramidite strategy and the attachment of the conjugate group to the released and deprotected ODN in solution.For example an active ester method has been employed to derivatize aminoalkyl side-arms attached to C5 of a 2-deoxyuridine unit. Alternatively a conjugate group having a nucleophilic functionality may be introduced by a reaction with an electrophilic side-arm functionality. For example a 5-cyanomethoxycarbonylmethyl-2-deoxyuridine residue has been derivatized in this manner. In principle similar techniques have been applied to attach conjugate groups to functionalized 3- or 5-tails of the ODN chain. Examples include conjugation through thioether, disul.de, peptide and urea linkages.S-Alkylation of 6-thio-6-deoxyhypoxanthine residue and displacement of the chloro substituent of 6-chloropurine with amine nucleophiles serve as illustrative examples of another typical approach of post-synthetic conjugation. Oligonucleotides containing an aldehydic function at any preselected position have been prepared by post-synthetic oxidation of alkene tethers attached to the base moieties. The resulting aldehyde groups may then be e.ciently coupled with aminooxy derivatized conjugate groups. Masked aldehyde groups have also been incorporated into ODNs by using a Cglycosidic building block. Peptide conjugates. Peptide conjugates of ODNs have attracted special interest since it has been hoped that peptides which are known to be actively transported through the cellular membrane could be exploited as carriers of antisense oligonucleotides into the cell.A rather straightforward method for the preparation of such conjugates appears to be acylation of aminopropyl-CPG with S-thiobutyl-N-Fmoccysteine and subsequent ODN chain assembly onto the deprotected amino group. The cysteine mercapto group deprotected upon the cleavage from the resin is then conjugated in solution with the desired synthetic peptide. Another feasible strategy involves insertion of 5-amino-5-deoxythymidine at the 5-terminus of the support anchored ODN and elongation of the chain by peptide synthesis using either stepwise or fragment coupling. Conjugates where the 3-end of the ODN is linked via a phosphodiester bond to the side-chain hydroxy function of theN-terminal amino acid residue have in turn been prepared by subsequent stepwise solid phase assembly of the peptide and ODN sequence. ODN 5,3-bis-peptide conjugates have been obtained by synthesizing the ODN 5-conjugate bearing a 3-amino tail on a solid support and condensing the separately synthesized peptide with the 3-amino function in solution. Chemically reactive oligonucleotides A number of ODNs containing an additional chemically reactive functionality have been prepared to achieve cross-linking with target nucleic acids or proteins.ODN duplexes O3—PO(O )— containing an acylphosphate linkage NH—CH —CO—PO(O )—O5 in one of the strands o.er a highly interesting example of cross-linking reagents for DNA recognizing proteins. The key step in obtaining this conjugate was the coupling of the amino group of glycine methyl ester to the 3- 212 Annu.Rep. Prog. Chem. Sect. B 1999 95 207—234 phosphate of one ODN fragment and template-directed ligation of the carboxy group with the 5¡¦-phosphate of the other ODN fragment. Another novel cross-linking agent for DNA processing enzymes has been obtained by inserting 1-(-galactopyranosyl)thymine to ODN and oxidizing the galactopyranosyl ring with periodate to an acyclic dialdehyde. Incorporation of photo-cross-linking nucleosides to ODNs represents a more conventional approach. Insertion of 5-E-[4-(3-triuoromethyl-3Hdiazirin-3-yl)styryl]-2-deoxyuridine and the thymidine monoadduct of cis-syn furan-side psoralen into ODNs by a solid support synthesis oers recent illustrative examples.Incorporation to ODN and post-synthetic derivatization of 4-thio-2- deoxyuridine with -bromoacetyl diamines and tethering of aryl nitrogen mustards to the 5-terminus of ODNs represent other typical approaches to obtaining sequence selective cross-linking agents. Tethering of diamines or metal ion chelates to ODNs has aorded conjugates that cleave sequence selectively RNA or DNA. Cleavage ofRNAhas been accomplished with conjugates derived from macrocyclic lanthanide complexes and diamines whereas cationic manganese porphyrin conjugates have been shown to be eective towards single stranded DNA. Oligonucleotide analogs exhibiting enhanced duplex or triplex formation Base modied oligonucleotides. The basic requirements of the antisense oligonucleotides aimed at selectively inhibiting gene expression include stability to nucleases good cellular uptake and high anity to the target mRNA.For this reason numerous base sugar and phosphate modied ODNs have been prepared to nd oligomers exhibiting enhanced duplex formation properties. Some novel examples of base modi- cations that enhance hybridization with unmodied oligonucleotides have been reported. These include 5-(3-methoxyprop-1-ynyl)uracil 5-N-[2-(N,N-bis(2- aminoethyl)amino)ethyl]carbamoyluracil 7-deaza-7-methylguanine and 7-propynyl- 7-iodo- or 7-cyano-7-deaza-2-aminoadenine. For the incorporation of the respective modied nucleosides into ODNs appropriately protected building blocks compatible with the solid phase phosphoramidite strategy have been developed.Polypyrimidine ODNs are known to recognize double helical polypurine¡Xpolypyrimidine regions in DNA by binding to the polypurine strand via Hoogsteen base-pairing.Since this kind of triple helix formation could possibly be exploited in recognition of erronous genes attempts have been made to stabilize triple helices and to broaden the variety of double helical structures that can be recognized. For this purpose a C-nucleoside 2-amino-5-(2-deoxy--ribofuranosyl)pyridine and its 3- methyl derivative have been incorporated in homopyrimidine ODNs as protonated cytidine equivalents and shown to increase the anity to ds DNA over the pH range 6¡X8. Introduction of 8-amino-2-deoxyadenosine in the purine tract has also been observed to stabilize the Hoogsteen base-pairing involved in the triplex formation.It has been proposed that the thermal stability of the triplex may simply be increased by increasing the exibility of the carbohydrate portion of the pyrimidine nucleosides involved. Oligonucleotides with alternating or consecutive isoguanine residues form duplexes with parallel chain orientation and also a tetrameric aggregate. Oligonucleotides containing consecutive 7-deazaisoguanine residues self-assemble to quartets indica- 213 Annu. Rep. Prog. Chem. Sect. B 1999 95 207¡X234ting that the purine N7 of isoguanine is not participating in the hydrogen bonding pattern. Sugar modi.ed oligonucleotides. The so-called locked nucleic acids (LNA) i.e. oligomers consisting of 2-O,4-C-methylene bicyclonucleoside units locked to the N type conformation have been observed to exibit markedly enhanced a.nity towards both DNA and RNA. Somewhat unexpectedly oligonucleotide analogs containing 2-O,3-C-linked bicyclo[3.2.0]arabinonucleoside units also show enhanced thermal stability of duplexes although in this case the sugar ring does not adopt N-type ring-puckering. Phosphate modi.ed oligonucleotides.The most straightforward way to obtain oligonucleotide analogs that are resistant towards nucleases is to modify the internucleosidic phosphodiester linkages. This usually ensures enzymatic stability but unfortunately the hybridization is often weakened. Nevertheless some phosphate modi.ed oligomers exhibit enhanced binding a.nity to RNA. Among these N3,P5-phosphoramidate ODNs appear to have several attributes that make them promising candidates for antisense therapeutics. Two new methods for the synthesis of these oligomers have been described one based on the reaction of the support-bound nucleoside 5-O-(2-cyanoethyl-H-phosphonate) with 3-amino-2,3-dideoxynucleoside in the presence of iodine, the other on phosphoramidite chemistry and an amine exchange reaction in the coupling step. Zwitterionic ODN analogs containing alternating negatively charged N3—PO(O )—O5 phosphoramidate bonds and positively charged O3—PO(NH )—O5 phosphoramidate diester groups have been synthesized and shown to exhibit markedly enhanced ionic strength independent a.nity to RNA. Importantly the duplex turned out to be a substrate of RNase H an intracellular enzyme that cleaves the RNA strand engaged in anODN—RNA complex releasing the ODN.ODN boranophosphates constitute another set of potential antisense oligomers. An e.ective solid phase synthesis for theseODNanalogs has been developed. Elongation of the chain by H-phosphonate coupling is followed by boronation which involves intermediate conversion of the H-phosphonate to a phosphite triester group by silylation and subsequent oxidation by a borane—amine complex. Although ODN phosphoromonothioates hybridize with complementary RNA less e.ciently than natural ODNs they are extensively used as .rst generation antisense oligonucleotides since they are able to activate the intracellular RNase H enzyme mentioned above. Consequently improvements in the synthesis of phosphorothioates are of special importance.A new sulfur-transfer reagent bis(ethoxythiocarbonyl) tetrasul.de has been prepared and used for the synthesis of ODN phosphorothioates by the solid phase phosphoramidite approach. It has also been shown that chain assembly from dimeric phosphoramidite synthons on CPG gives a signi.cantly improved impurity pro.le compared to oligomers synthesized by coupling of the standard monomer phosphoramidites. Nucleoside 3-O-(2-thio-‘‘spiro’’-4,4-pentamethylene-1,3,2-oxathiaphospholane)s and nucleoside bicyclic oxazaphospholidines derived from .- or .-prolinol have been introduced as novel synthons for the stereo-controlled solid phase synthesis of ODN phosphorothioates. Two novel methods for the synthesis of ODN phosphorothioates in solution have 214 Annu.Rep. Prog. Chem. Sect. B 1999 95 207—234 been described H-phosphonate coupling at40 °C followed by in situ sulfur-transfer with either N-[(4-chlorophenyl)sulfanyl]phthalimide or 4-[(2-cyanoethyl)sulfanyl] morpholine-3,5-dione, and a phosphotriester approach using S-(2-cyanoethyl) protection for the phosphorothioate internucleotide linkages and MSNT as the coupling agent. To enhance the cellular uptake of ODN phosphorothioates the internucleosidic linkages have been masked with protecting groups aimed at being removed by intracellular enzymes. The protections proposed included S-acyl-2-thioethyl groups, pivaloyloxymethyl group, and acyloxyaryl groups. 3 Peptides General remarks As with oligonucleotides the solid phase synthesis of peptides is based on such a long experience and the commonly used Boc and Fmoc strategies are so well established that the recent contributions though certainly important may rather be regarded as improvements of the existing protocols than entirely new break-throughs.Considerable progress has been achieved in the preparation of structurally modi.ed peptides and peptide conjugates on a solid support. Furthermore several new protecting strategies for the side chain functionalities have been introduced as well as useful methods for the preparation of so-called di.cult sequences. General methodology of solid phase synthesis Coupling. The overall success of a peptide synthesis is to a very large extent dependent on the e.ciency of coupling and avoidance of side reactions such as racemization upon coupling.Accordingly continuous e.orts towards improvements in coupling are of importance. Bis(tetramethylene).uoroformamidinium hexa- .uorophosphate has been reported to be a convenient coupling reagent for the solid phase synthesis of peptides incorporating sensitive amino acids. The applicability of 2-chloro-1,3-dimethylimidazolium hexa.uorophosphate as a coupling reagent in the presence of HOAt and HODhbt has been demonstrated by assembling up to 20 residue long peptides containing sterically hindered Aib residues. Phosphonium and uronium salt based coupling reagents such as BOP PyBrOP PyAOP HBTU and HATU have been used in the preparation of tertiary amide bonds. Interestingly freezing the reaction solution has been shown to accelerate EDAC assisted peptide coupling in aqueous solution and to suppress the formation of the N-dipeptidylurea side-product with DCC-assisted coupling in organic solvents. Racemization of the N,S-protected cysteine residues upon coupling is one of the problems of solid phase peptide synthesis.This racemization has now been shown to be unexpectedly substantial with several common coupling reagents on using conventional S-protecting groups such as acetamidomethyl trityl 2,4,6-trimethoxybenzyl or 9H-xanthen-9-yl. Couplings mediated by phosphonium and aminium salts including BOP HBTU HATU and PyAOP which typically involve preactivation and the presence of suitable additives (HOBt or HOAt and a tertiary amine) were all observed 215 Annu.Rep. Prog. Chem. Sect. B 1999 95 207—234 to be accompanied by extensive racemization. When the preactivation step was omitted a weak base 2,4,6-trimethylpyridine was used as an additive instead of the aliphatic tertiary amine and neatDMFas the solvent was replaced with a 1 1 mixture of CH Cl and DMF the racemization could be supressed below 1%. Di.cult sequences. Sequences containing successive sterically hindered amino acid residues such as Aib or diphenylglycine are known to be di.cult to prepare. The chlorides of -azido acids have been used without racemization to obtain such extremely hindered peptides. N-Arenesulfonyl protected amino acid chlorides have also been suggested to be superior to the commonly used urethane protected chlorides for this purpose. For the preparation of sterically hindered sequences phosphonium derivatives of HOAt such as PyAOP have been recommended as coupling reagents. PyAOPwas shown not to undergo the detrimental side reaction at the amino terminus which blocks further chain assembly.Backbone amide nitrogen protection may be needed to prevent epimerization during the solid support chain assembly of di.cult sequences.N-Fmoc-N-(2-hydroxy- 4-methoxybenzyl) protected amino acids have been recommended for this purpose. This protection has however been argued to retard the coupling and hence a new backbone amide protecting group 5-(6-hydroxy-1,3-benzoxathiolyl)methyl group has been developed. With this protection the backbone epimerization was suppressed without retardation of coupling.Preparation of hydrophobic peptides has also turned problematic. A mixture of NMP and DMSO (4 1 v/v) has now been shown to be superior to DMF for assembling such peptides by the conventional solid support Boc/benzyl chemistry. Optimized conditions for the solid phase synthesis of some other di.cult sequences have also been reported. These include preparation of tryptophan-rich peptides, easily aggregating peptides, and the biologically interesting sequence H—Ala—Arg—(Ala) —Lys—OH that exhibits problems in coupling and N-Fmoc deprotection. Cleavage from the support. Hydrogen .uoride is widely applied to cleave the peptide chains from the support.Under these conditions acylium ions are generated and hence a nucleophilic scavenger is usually added to suppress the unwanted side reactions. The role of p-cresol as a scavenger of acylium ions has been examined. Initially p-cresol esters were shown to be formed and they subsequently rearranged to ketones. Importantly peptides containing p-cresol esters at the glutamyl side chains were observed to be susceptible to amidation and fragmentation at these sites. These side reactions were largely avoided by mild hydrogen peroxide-catalyzed hydrolysis which converted the p-cresol adducts to free carboxylic acids in nearly quantitative yield. Accordingly p-cresol appears to be an appropriate acylium ion scavenger to be used in the cleavage step of Boc/benzyl peptide synthesis.Comparison of several aliphatic thiols has in turn led to recommendation of DTT as a scavenger. Me SnOH or [(n-C H ) Sn] O have been used to selectively and nonacidolytically cleave N-Boc-peptide phenacyl esters linked to a polystyrene resin. The method is compatible with Boc/benzyl protection strategy and yields enantiomerically pure N-Boc-peptides useful for further manipulation. 216 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 Inverse chain assembly. For some special purposes the assembly of the peptide chain in an inverse direction i.e. coupling of the free amino group of the entering amino acid to the carboxy function of N-anchored peptide chain is desirable. This kind of synthesis has been carried out on a modi.ed Merri.eld resin derivatized with a Boc type handle. The coupling was achieved by converting the resin-bound amino acid to activated dinitrophenyl ester and reacting it with an N-unprotected amino acid.This kind of inverse solid phase chain assembly has also been accomplished by repetitive addition of the HOBt salts of the amino acid .uoren-9-ylmethyl esters the .rst amino acid being coupled through its amino group to a support-bound trityl linker. Acylation of the aminomethyl groups of polystyrene or TentaGel resins with [4-(9-hydroxy.uoren-9-yl)phenyl]acetic acid has been shown to give a linker exhibiting improved acid stability compared to the trityl linker. Acetylation of the 9-hydroxy group and subsequent nucleophilic displacement results in anchoring of the amino acid through its amino group and hence a support applicable for the inverse chain assembly is obtained.A new arenesulfonyl linker has been developed for the solid phase synthesis of arginine containing peptides. The arginine group is attached to the support by a nucleophilic attack of its guanidine group on the sulfonyl function and the peptide chains can be assembled onto either the amino or carboxy groups or both of the immobilized arginine. The linker is compatible with both Boc and Fmoc chemistry. Protecting groups. N-Allyloxycarbonyl (Alloc) protection which can be removed under neutral conditions with catalytic amounts of tetrakis(triphenylphosphine)palladium( 0) in the presence of phenylsilane as a scavanger of the allyl unit o.ers an alternative to the N-Boc and N-Fmoc protecting groups. Short peptides have even been synthesized in solution by a stepwise strategy using this protection the key feature of the methodology being selective removal of the Alloc protection catalyzed by a water-soluble palladium catalyst palladium acetate—triphenylphosphinotrisulfonate sodium salt. The hydrophilic (1,1-dioxobenzo[b]thiophen-2-yl)methyloxycarbonyl group o.ers some advantages as a base- and nucleophile-sensitive protecting group for amino functions in peptide coupling. Owing to its hydrophilic character the group can be removed with tris(2-aminoethyl)amine followed by washing with water or saturated salt solution.Cyclohexyloxycarbonyl group may be used for orthogonal protection of the - amino function in solid phase peptide synthesis by Boc chemistry. The protection is stable under TMSOTf-thioanisole—TFA cleavage conditions but it can be removed with anhydrous HF.In combination with cyclohexyl ester protection for the side chain carboxy functions this protection has allowed the preparation of fully protected peptides with multiple amino and carboxy groups. 2-Acetyl-4-nitroindane-1,3-dione is another novel orthogonal protecting group for primary amines. The group withstands well acids and secondary and tertiary amines but it can be removed with 2% hydrazine at ambient temperature. Protected peptides can also be obtained using the conventional Boc or Fmoc protections when the synthesis is carried out on the recently developedHYCRON (hydroxycrotyl-oligoethylene glycol-n-alkanoyl) linker.This linker has been shown to be compatible with both Boc and Fmoc chemistry and the protected peptides can be released under almost neutral conditions making use of Pd(0)-catalyzed allyl transfer to a weakly basic nucleophile. 217 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 1-Ethylpropyl and 1-isopropyl-2-methylpropyloxycarbonyl groups have been used to protect the tyrosine hydroxy function. Both withstand well the piperidine assisted removal of the Fmoc protection and may be completely cleaved by the standard hydrogen .uoride treatment. Alternatively 2-adamantyloxycarbonyl group may be used. This group is stable to tertiary amines TFA and hydrogenation over a Pd catalyst and it may be removed by treatment with tri.ic acid—thioanisole—TFA or hydrogen .uoride. The side chain carboxy groups may in turn be protected as 2-adamantyl esters and deprotected under the conditions indicated above for the removal of the tyrosine 2-adamantyloxycarbonyl group. The photo-cleavable 4- hydroxyphenylacetyl group has been used for the same purpose. 2,4-Dinitrophenyl protection for the histidine imidazole moiety has been recommended to suppress racemization during solid phase coupling. It has however been argued that this protection is to a large extent cleaved under the conditions used to remove the Fmoc protection i.e.with 20% piperidine in DMF. Accordingly allyl protection that can be selectively removed with palladium catalysts, or 2- adamantyloxymethyl group which is compatible with Boc chemistry and can be removed with tri.ic acid—thioanisole—TFA or hydrogen .uoride, may be a better choice for this purpose.5H-Dibenzo[a,d]cyclohept-5-enyl group and its 10,11-didehydro and 2-methoxy- 10,11-didehydro congeners have been suggested to be useful for protection of the guanidino function of arginine on using the Fmoc strategy and BOP or HOBt activation. All these groups are readily removed by mild acid treatment (50% TFA 1 h). Novel S-xanthenyl groups viz. 9H-xanthen-9-yl and its 2-methoxy derivative have been suggested to be superior to the commonly used S-trityl protection of cysteine residues. The group is compatible with Fmoc chemistry and it may be selectively removed with a mixture of TFA CH Cl and Et SiH retaining the commonly used tris(alkoxybenzyl)amide anchoring.Monitoring. Considerable progress has taken place with the on-line monitoring of peptide synthesis on solid supports. The methods applied include MALDI-Tof and Tof-SI mass spectrometry NIR—FT-Raman spectroscopy, EPR spectroscopy, Magic Angle Spinning NMR spectroscopy, and solution NMR spectroscopy of material cleaved from single resin beads. Ligation in solution. Ligation of peptides in solution allows the construction of polypeptides which are too long to be conveniently obtained by a solid phase synthesis. Preparation of a 41 amino acid residues long cysteine-rich peptide corresponding to the C-terminal fragment of the mouse Agouti protein o.ers an illustrative example of the applicability of this methodology. The peptide was assembled from .ve peptides each containing C-terminal glycine or proline.These peptides bearing theN-terminalN-Boc protection were assembled on the 2-chlorotrityl resin by Fmoc chemistry and cleaved with a mixture of CH Cl methanol and acetic acid. The N-terminal Boc protection was removed when desired in neat TFA and stepwise solution phase condensation of the peptides was carried out using EDAC as an activator. 218 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 A useful enzyme assisted method for ligation of peptides in solution has also been described.The 4-guanidinophenyl esters bind to the active site of thrombin trypsin and clostripain in a way that mimics trypsin speci.c side chains. Consequently peptide coupling occurs irreversibly and independent of speci.city. Construction of arti.- cial proteins from -helical peptides via selective disul.de cross-linking also appears to be a method of general interest. Synthesis of cyclic and modi.ed peptides on a solid support Cyclic peptides. A highly useful new strategy for the preparation of cyclic peptides on a solid support has been described. The -amino group of the prospective C-terminal amino acid or its modi.ed analog is coupled to the support-bound 5-(4- formyl-3,5-dimethoxyphenoxy)valeric acid handle by reductive amination and conventional Fmoc solid phase chemistry is then applied to assemble the peptide chain.The cyclization is carried out on the support and the cleavage from the resin takes place upon removal of the side chain protections. In addition to cyclic peptides examples of the syntheses of peptide alcohols aldehydes esters and N,N-dialkylamides by essentially the same methodology have been provided. The active carbonate resin o.ers an alternative method for the preparation of cyclic peptides. N,N-Disuccinimidyl carbonate is used to generate a carbonate (carbamate) bridge between the side chain hydroxy (amino) group of the .rst otherwise protected amino acid and the hydroxymethyl group of polystyrene resin. The peptide chain is assembled onto the deprotected -amino group after which the carboxy function of the .rst support bound amino acid is deprotected and the cyclization is achieved.With a slight modi.cation the method may also be used for preparation of peptide alcohols. On benzhydryl resins peptide carbamates are obtained. C-Terminal modi.ed peptides. C-Terminally modi.ed peptides have been prepared by a versatile method based on the use of two di.erent linkers having markedly di.erent acid stability in the early and late stages of the synthesis. Initially the -amino group of the support-bound N-Alloc protected lysine is derivatized with a highly acid labile linker onto which the peptide chain is assembled by Fmoc chemistry. TheN-terminus of the peptide is then tethered to the deprotected -amino group of the support-bound lysine via a less acid labile linker.Cleavage of the C-terminus of the peptide from the N-linker allows derivatization of the released C-terminus and the modi.ed peptide is .nally released to solution by cleavage of the N-linker. Peptide aldehydes have attracted interest as inhibitors of proteolytic enzymes. Mainly for this reason several novel solid phase approaches have been developed for their preparation. A simple and elegant method involves reaction of the prospective C-terminal aldehyde unit with the support-bound serine or threonine residue to give an oxazolidine. These oxazolidine linkers withstand Fmoc peptide synthesis including TFA treatment but they can be cleaved with mild aqueous acid at slightly elevated temperature.A related approach is the usage of thiazolidine peptides obtained in solution by Mitsunobu cyclization of the -hydroxy thiopeptide as precursors. Copper salts promoted hydrolysis of the terminal heterocycle then furnishes the peptidyl aldehyde without any racemization. The other methods reported are 219 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 based on reductive or oxidative cleavage of the linker. The peptide can be assembled onto the methoxyaminoalkyl linker by Boc or Fmoc chemistry and then released as the peptide aldehyde by a lithium aluminium hydride reduction. Alternatively the chain elongation may take place on an ,-unsaturated aminoalkyl linker and the terminal aldehyde group is created by ozonolysis of the double bond of the linker. It is worth noting that the C-terminal unit is in this case formed from the linker which makes the approach less versatile.In addition to peptide aldehydes peptide esters alcohols amides ketones and hydroxamic acids have been prepared on solid supports. Mercaptoalkyl linkers have been used for the synthesis of support-bound peptide thioesters which upon cleavage with alcohols amines and organometallic reagents are released as peptide esters amides ketones aldehydes and alcohols. Similarly the cleavage of support-bound peptide o-nitrobenzyl esters with alkoxide or amine nucleophiles has been utilized to a.ord peptide esters and amides. Peptides bound to the Kaiser oxime resin have been cleaved with various oxygen nucleophiles in the presence of DBU to obtain fully protected peptide acids and esters. Peptide amides and N-alkylamides have been prepared by several alternative methods.Firstly polystyrene supported 2- methoxybenzaldehyde linker is reductively aminated with the desired primary amine the peptide chain is assembled onto the resulting secondary amine group and the peptide is released as N-alkylamide by treatment with TFA in dichloromethane. Secondly the peptide may be assembled onto an orthogonal aryl hydrazide linker which is then cleaved under mild oxidative conditions to give the amide. Thirdly the peptide amides are assembled by the Fmoc strategy on polystyrene supported [9-(4-methoxyphenyl)xanthen-9-yl]amine linker and on TentaGel supported 9- hydroxy-9-(4-carboxyphenyl)xanthene linker. Peptide hydroxamic acids have been prepared on 4-methylbenzhydrylamine resin. The support bound carboxy function was converted to a hydroxamate group by condensation with O-benzylhydroxylamine.The hydroxamate benzyl protection was removed upon HF cleavage of the peptide from the resin. Peptoids. Peptoids comprisingN-substituted glycine residues have recently attracted interest as peptide mimics. A solid phase procedure for the assembly of such oligomers from Fmoc-protected N-substituted glycines has been described. Solid phase synthesis of oligoureas constitutes another example of convenient preparation of peptide mimics. Preparation of peptide conjugates on a solid support Glycopeptides. Remarkable advances in the solid phase synthesis of peptide glycoconjugates have been achieved.It has been shown that peptides assembled by Fmoc chemistry on a novel polyoxyethylene—polyoxypropylene resin could be glycosylated on-resin with trichloroacetimidate glycosyl donors. More frequently the sugar moieties have however been incorporated in peptides as building blocks derived from amino acid glycoconjugates. Accordingly the Fmoc protected asparagine conjugate of an N-linked pentasaccharide has been incorporated in a peptide using it as a building block in chain assembly on hydroxymethylated polystyrene. 220 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 The Fmoc strategy was applied and the benzyl group was employed as a global protection of the oligosaccharide moiety.The benzylated pentasaccharide moiety did not interfere with the peptide coupling reactions and it turned out to be su.ciently stable to withstand the acidolytic cleavage from the resin. The threonine glycosides of acyl protected di- and trisaccharides have also been used as building blocks in solid phase peptide synthesis by the Fmoc strategy. Glycopeptides related to type II collagen have been prepared by conjugating the sugar moiety bearing silyl isopropylidene and 4-methoxybenzyl protecting groups with 5-hydroxy-.-norvaline and using the resulting building block in the Fmoc protected form in the solid phase synthesis. The carbohydrate moiety was completely deblocked during the acidolytic cleavage from the support while the O-glycosidic bonds remained intact.A similar approach has been utilized to obtain the B-chain of human 2HS glycoprotein a 27-residue peptide carrying a trisaccharide side chain. In this case Fmoc/benzyl protected glycosyl serine building blocks were employed.A sialyl glycoundecapeptide corresponding to the repeating unit of MUC1 has been synthesized on an allylic linker by introducing the glycosyl moiety as a threonine disaccharide glycoside by Fmoc chemistry. Useful chemoenzymic syntheses of complex glycopeptides have also been described. An Fmoc-threonine -glycoside of fully acylated N-acetylglucosamine has been used as a building block in peptide chain assembly on the allyl HYCRON linker by Fmoc chemistry. The acyl protections were removed on the support and glycosyltransferases were employed to extend the glycan.Palladium catalyzed cleavage then released the conjugate in solution. Similarly an N-acetylglucosaminyl peptide was .rst assembled on a solid support using Fmoc protected N-(acetylglucosaminyl)asparagine as a building block and the oligosaccharide of human transferrin glycopeptide was then transferred to the N-acetylglucosaminyl peptide by a microbial endoglycosidase the endo--N-acetylglucosaminidase of Mucor hiemalis. Oxime linked glycopeptides have also been prepared by a chemoenzymic method. A single Nacetylgalactosamine residue was incorporated into the glycopeptide by solid phase methods and the sugar moiety was subsequently oxidized to the corresponding aldehyde with galactose oxidase. Hydroxylamine functionalized sugars were then ligated onto the glycopeptide aldehyde a.ording oxime-linked products with native sugar—peptide linkages.Phosphopeptides. Serine phosphopeptides have been obtained by introducing the diallyl ester of the N-Alloc protected phosphoserine into the N-terminus of a sequence prepared by the standard Fmoc method. Global cleavage of the allyl protections allows subsequent coupling. The protection of the O-phosphono group as the dicyclopentyl ester is compatible with Boc chemistry and the dibenzoyl protection is compatible with Fmoc chemistry. Phosphotyrosine containing peptides have been synthesized using bis(2,2,2-trichloro)ethyl groups for phosphate protection and assembling the chain by Boc chemistry. Enzyme cleavable protecting groups have also been exploited in the synthesis of phosphopeptides.Accordingly the heptyl ester has been used as the C-terminal or the phenylacetamide as the N-terminal protecting group.A phosphorylated glycopeptide fragment of the large subunit of mammalian RNA polymerase II has been prepared by a chemoenzymic approach. The key feature of the synthesis was the application of a 4-(phenylacetyloxy)benzyloxycarbonyl group 221 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 for the N-protection of the amino acid building blocks. This group could be removed enzymatically with penicillin G acylase under very mild conditions which allowed stepwise assembly of the peptide chain from amino acid building blocks including side chain glycosidated and phosphorylated serine.Peptide nucleic acids. Peptide nucleic acids (PNA) constitute an interesting class of biopolymer analogs. They consist of a peptide-like backbone and side chains bearing nucleic acid bases. Accordingly they are able to form highly stable double and triple helical structures with nucleic acids and hence they may be regarded as potential lead compounds for development of chemotherapeutics and diagnostic probes. Numerous approaches for the solid phase synthesis of novel PNAs have been reported. The following oligomers serve as illustrative examples alanyl PNAof alternating con.guration, -homoalanyl PNA, .--homoserinyl PNA, PNAs with a conformationally constrained chiral cyclohexyl-derived backbone and chirally pure ornithine based PNA. Cyclization of peptides in solution Although convenient methods are available for solid phase synthesis of cyclic peptides cyclization in solution may still be useful in many cases.C-Terminal peptide thioesters undergo in the presence of silver(.) ion end-to-end lactam formation in aqueous bu.ers containing DMSO. With appropriate protections side chain-to-end cyclization may also be achieved. When peptide thioesters bearing an unprotected N-terminal cysteine are used for the cyclization the initially formed cyclic thioester is subsequently isomerized to a lactam having a free mercapto group and this functionality may be exploited in disul.de bond formation. Cystine-knot motifs and peptide dendrimers were constructed in this manner.Preparation of cyclic peptides from peptide conjugate precursors bearing a C-terminal thioester bond and N-terminal N-(2-mercaptoethoxy)glycine deserves special attention. These compounds have been shown to undergo intramolecular cyclization to a thioester and subsequent S,N-acyl rearrangement to cyclic N-(2-mercaptoethoxy) lactam. Reductive removal of the N-alkoxy group then gave the unmodi.ed glycine linked cyclic peptide. Esterasesensitive cyclic prodrugs of peptides have been synthesized by tethering the terminal carboxy and amino functions with an O-(hydroxymethyl)formyl or a 3-(2-hydroxy- 4,6-dimethylphenyl)-3-methylbutyryl group. 4 Oligosaccharides General remarks Of the three major classes of biopolymers polysaccharides have proven to be the most di.cult ones to synthesize.While oligonucleotides and peptides are routinely synthesized by machine-assisted solid phase methods polymer-supported synthesis is still an emerging tool for the production of oligosaccharides. The main reason for this is that it is rather complicated to achieve complete stereoselectivity for the whole variety of glycosidations needed in construction of a complex oligosaccharide. Recent ad- 222 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 vances however strongly suggest that methodologies useful for glycosidation in solution are applicable also on solid supports. Synthesis of oligosaccharides on a solid support An intelligent strategy for the solid phase preparation of branched oligosaccharides has been developed and applied to the synthesis of a heptasaccharide phytoalexin elicitor. The .rst monosaccharide unit was attached through its anomeric carbon to the hydroxy group of a photocleavable 4-alkoxy-2-nitrophenylmethanol handle and appropriately protected phenyl thioglycosides bearing a participating protecting group at O-2 were used as glycosyl donors.Reiteration of the glycosidation process enabled stereocontrolled growth of the chain in high yield and anomeric purity. To avoid the presence of both - and -anomers at every cleavage stage as well as the reactivation of the cleavage product prior to its reincorporation to the support the methodology has been further improved by incorporating an appropriate spacer 4-hydroxybenzoic acid between the photolabile handle and the anomeric carbon of the immobilized glycoside. This structure ensured exclusive -stereochemistry which was maintained throughout the synthesis allowed the cleavage as anomerically pure 1-O-(4-hydroxybenzoate) and o.ered the opportunity to cleave from the resin with concomitant activation to anomeric phenyl thioglycosides.In this manner a stereochemically homogeneous dodecasaccharide related to the phytoalexin elicitor family was successfully prepared. These studies convincingly demonstrate the feasibility of the solid phase approach in the construction of complex carbohydrates. The previously described methodology which is based on conversion of the support-bound glycal to a 1,2-anhydro glycosyl donor by dimethyldioxirane activation has been modi.ed. While -galactosyl linkages have been e.ectively obtained by the previous method preparation of -glucosyl bonds has turned problematic.This shortcoming has now been overcome by opening the epoxide ring with ethanethiol in the presence of TFA. Subsequent protection of the 2-hydroxy function with a pivaloyl group then gives a support-bound -thioglucoside that upon activation with methyl tri.ate has been shown to yield -glucosyl (14) and -glucosyl (13) bonds almost free of contaminating side products. Arelated methodology has been applied to synthesize the tetrasaccharide domain of the Lewis blood group determinant by amidoglycosylation of polymer-bound glycals. To achieve this the polymer-supported glycal was treated with iodonium sym-collidine perchlorate and benzenesulfonamide to form an iodosulfonamide intermediate which was transformed into a glycosyl donor via rearrangement and displacement with EtSH.The support-bound ethylsulfanyl 2-amidoglucosyl donor underwent e.cient coupling with glycal acceptors in the presence of methyl tri.ate. A highly useful method for monitoring the glycosidation of support bound glycals by high-resolution magic angle spinningNMR has been developed. This technique was also used to verify that glycosidation of appropriately protected -1,2-anhydro sugars on solid supports is highly stereoselective yielding the -galactoside even more overwhelmingly than in solution. A repetitive solid phase synthesis for -(12)-oligomannopyranosides containing up to six monosaccharide units has been described. 2-O-Acetyl-3,4,6-tri-O-benzyl- -.-mannopyranosyl trichloroacetimidate was used as the glycosyl donor.The .rst such building block was immobilized to a support-bound mercaptoalkyl linker the 223 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 2-O-acetyl protection was removed and the next building block was coupled using TMSOTf as a catalyst. The oligomer was cleaved from the support with NBS in aqueous acetone. It should be however noted that while this glycosidation using the ax 2-hydroxy group of mannose as a nucleophile is stereoselective a similar previous approach by the same authors yielded (16)-glucosides as anomeric mixtures. This strategy has also been applied to the solid phase synthesis of the branched pentasaccharide moiety that occurs in most complex type N-glycan chains.The synthesis was initiated by immobilizing the branching sugar unit -mannopyranoside to the support and assembling the branches onto its 3- and 6-hydroxy groups by two successive double glycosylations with trichloroacetimidate glycosyl donors. By a rather similar approach -(12)-trimannoside has been prepared on mercaptopropyl-functionalized CPG. The applicability of polystyrene Tentagel and CPG to the solid phase synthesis of oligosaccharides by the trichloroacetimidate methodology has been compared. CPG gave the best results with TMSOTf as the glycosylation promoter. Linear trisaccharides have been assembled from n-pentenyl glycosyl donors. The rst unit bearing global benzyl ether and temporary ester protections was immobilized to the support through the pentenyl group.After removal of the base-labile ester protection the next n-pentenyl glycosyl donor was introduced using a mixture of NIS and TESOTf as an activator. On using -glucopyranoside bearing a participating ester protection at O-2 as a glycosyl donor the 1,2-trans () glycoside was obtained similarly to the situation in solution. An interesting solid support modication of the previously reported intramolecular aglycon delivery has been developed to obtain -mannopyranosides i.e. glycosides considered to be most dicult to prepare stereoselectively. A protected methyl 1-thio--mannopyranoside was attached to the polymer support through a 2-O-(4- alkyloxybenzyl) linker. To achieve glycosidation this support-bound glycosyl donor was then subjected to -mannosylation by the method of 4-methoxybenzyl assisted intramolecular aglycon delivery.Accordingly treatment with DDQ aorded acetalization of the support-bound 2-O-benzyl group with the prospective glycosyl acceptor and activation of the anomeric C¡XS bond with MeOTf¡XMeSSMe in the presence of 2,6-di-tert-butyl-4-methylpyridine triggered the actual intramolecular aglycon delivery. This process allows the glycosidation product to be specically released from the polymer to give the mixture highly enriched with the -mannoside while the auxiliary alkoxybenzyl linker and the glycan and reducing sugar side products remain bonded to the support. Synthesis of oligosaccharides in solution As mentioned above oligosaccharides are still synthesized more frequently in solution than on a solid support and the strategies of solution synthesis are continuously improved.A good example of a highly convergent solution synthesis is oered by the total synthesis of the KH-l (adenocarcinoma) antigen a branched nonasaccharide glycosidated with a bulky hydrophobic aglycon. The route utilized involves surprisingly few protecting group manipulations. A linear hexasaccharide was rst constructed by glycal assembly making use of the known reactivity preferences of various sugar hydroxy groups and the stereochemical outcome of glycosidations. A key step 224 Annu. Rep. Prog. Chem. Sect. B 1999 95 207¡X234was the concurrent addition of three -fucosyl residues at required hydroxy acceptor sites. The synthesis was completed with introduction of the aglycon moiety.Another nonasaccharide conjugate viz. a triantennary high mannose type nonasaccharide which is part of the glycoprotein gp 120 of the viral coat of HIV-1 has been prepared in solution by exploiting thioethyl and selenophenyl glycosyl donors and tuning their reactivity by cyclohexane-1,2-diacetal protection of the 3,4-diol unit in mannosides. The tetra- and pentasaccharide moieties of this mannan were hence obtained by consecutive glycosidation steps without the need for any protecting group manipulation. In fact the only essential protecting group manipulation was removal of the TPS group from the tetrameric fragment before the nal assembly of the nonasaccharide. A similar tuning of the reactivity of glycosyl donors by selective introduction of dierent protecting and leaving groups has been applied to obtain tetra- and pentasaccharides in one pot.Linear and branched penta- and heptasaccharides have also been obtained in one pot by using butane-2,3-diacetal protection for reactivity tuning of mannosyl uorides. A series of modular di- and trisaccharide trichloroacetimidate building blocks providing access to tri- and tetraantennaryN-glycan oligosaccharides have been prepared. As examples of the applications of these building blocks a tetraantennary nonasaccharide was synthesized in high yield. Similarly an octasaccharide mimicking the structural elements of type III group B Streptococcus capsular polysaccharide has been assembled in a convergent manner from di- tri- and tetrasaccharide trichloroacetimidate donors. An asparagine conjugate of a biantennary heptasaccharide has been prepared in seven steps from ve dierent monosaccharide building blocks exploiting glycosyl uorides chlorides and bromides as glycosyl donors.A trisaccharide containing the branching mannose unit was rst synthesized and the branches were assembled onto the 3- and 6-hydroxy groups of this unit by two successive double glycosylations. A versatile strategy for the synthesis of complex type N-glycans has been developed. The key elements of the strategy are glycosyl trichloroacetimidates as glycosyl donors exible protecting group patterns and branched tetra- and pentasaccharide core structures for additional antenna attachment. The eciency of the strategy was demonstrated by the synthesis of a nonaand decasaccharide. Several convergent syntheses of hexasaccharides have been reported.The hexasaccharide fragment of landomycin A a linear oligosaccharide of 2,6-dideoxyhexose units containing both - and -(14) linkages was obtained by glycosylating a trisaccharide acceptor with a trisaccharide pinacol phosphite donor. Unfortunately this nal glycosidation was not markedly stereoselective while the trisaccharide units relied on stereocontrolled glycosidations using glycosyl tetrazoles and phosphites as donors to establish and glycosidic linkages respectively. The phytoalexin elicitor hexasaccharide -glucohexatose has been prepared via coupling of a trisaccharide donor and acceptor that were in turn obtained by coupling a common disaccharide precursor with acetobromoglucose and 6-O-chloroacetylated acetobromoglucose respectively.Hexasaccharide sialoglycoconjugates viz. sialyl Lewis ganglioside analogs modied at C-6 of the galactose residue have been assembled from methyl thioglycoside trisaccharide donors and a disaccharide acceptor in the presence of DMTST and the pentasaccharide obtained was glycosylated with a phenyl thioglycoside derivative of -fucose using NIH-TfOH activation. A linear hexasaccharide corresponding to the binding site of dermatan sulfate to heparin cofactor II was assembled from three 225 Annu. Rep. Prog. Chem. Sect. B 1999 95 207¡X234disaccharides exploiting trichloroacetimidate chemistry, and an -(1,2)-linked fucose hexasaccharide from disaccharide thioglycosides by CuBr —Bu NBr activation that gave stereoselectively -glycosidic linkages. A series of di- and trisaccharides which can be employed as acceptors in subsequent glycosylations without protecting group manipulations have been prepared and used in a convergent one-pot multi-step glycosylation yielding 14 and 16 linked hexasaccharides. A branched hexasaccharide derived from group B type III Streptococcus has been assembled from n-pentenyl and 1,2-O-(1-cyanoethylidene) disaccharide donors. The feasibility of a linear block-wise synthesis in solution has been nicely demonstrated by the synthesis of a hexadecasaccharide corresponding to the repeating units of the O-speci.c polysaccharide of Shigella dysenteriae. Two tetrasaccharide donor/ acceptor repeating units were assembled from monosaccharide precursors in a stepwise fashion and used in a linear iterative manner to construct the hexadecasaccharide making use of trichloroacetimidate donors and TMSOTf activation.As an interesting application of soluble polymer supports to oligosaccharide synthesis heparin sulfate-like oligomers have been assembled on PEG-polymers. The key reaction of the synthesis was repetitive coupling of disaccharide trichloroacetimidates to the PEG-supported acceptor that originally was an iduronic acid containing disaccharide. Oligomers up to 12 units in length were prepared and sulfated post-synthetically after deprotection and release from the support. It has been shown that glycosyl tetramethylphosphoramidates having the primary 6-hydroxy function unprotected may safely be used as glycosyl acceptors on using more reactive glycosyl diphenylphosphinimidates or phosphites as glycosyl donors in boron tri.uoride promoted glycosidation.This allows a rather e.cient strategy for oligosaccharide synthesis since the disaccharide obtained may subsequently be used as a glycosyl donor without any protecting group manipulations. A two directional glycosylation strategy for the convergent assembly of oligosaccharides has also been proposed. Primary TBDMS ethers and secondary TES ethers of thioglycosides were observed to be satisfactorily stable on using these molecules as glycosyl donors in NIS—TMSOTfpromoted glycosidations and the products obtained were appropriate acceptors when glycosylated with glycosyl .uorides in the presence of Cp ZrCl —AgOTf.Glycosidation Glycosidation is the key reaction of oligosaccharide synthesis. Further development of synthesis strategies both in solution and on a solid support is largely dependent on the e.ciency and stereoselectivity of glycosidation. Glycosyl donors covering a wide range of reactivity and being dependent on various modes of activation are needed hence novel glycosidation techniques ful.lling these requirements are of importance. A novel glycosidation a.ording -mannopyranosides and other hindered glycosides from thioglycoside donors has been introduced. Phenyl -thiomannosides are converted with benzenesulfenyl tri.ate to -mannosyl tri.ates and the latter may be coupled with a range of secondary and tertiary glycosyl acceptors to give -mannosides in good yield and selectivity.Sugar ortho esters have been used as intermediates in the regio- and stereoselective preparation of 1,2-trans-linked saccharides. Silver( .) tri.ate promoted coupling of acetobromosugar with 4,6-unprotected glycosides in the presence of 2,4-lutidine for example gives the ortho ester intermediate which 226 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 with a catalytic amount of TMSOTf undergoes rearrangement to disaccharide. The ortho esters can also be used to obtain sterically congested glycosides such as mannosyl-(14)-glycosides. These have been stereoselectively synthesized by reduction of glycosidic spiro-orthoesters with lithium aluminium hydride in the presence of aluminium() chloride. A high yielding method for 1,2-cis-mannosylation and 1,2-cis-rhamnosylation has been developed using 1,2-stannylene acetals of unprotected sugars as nucleophilic glycosyl donors and triic esters of carbohydrates as electrophilic glycosyl acceptors.Stereoselective -glycosidation of per-acetylated sugars despite the presence of participating 2-O-acetyl groups has been achieved by using ferric chloride as a Lewis acid catalyst. Direct -selective glycosidations with 1-hydroxy glycosyl donors may be achieved with the aid of triic acid anhydride and diphenyl sulfoxide. An unusual proximity eect has been encountered in sulfonamido glycosylation when C4 on the galactose ring of an ax-glycosylating donor bears a free hydroxy group -glycoside formation predominates whereas protection of this function sends the glycosidation in the direction of -glycoside formation.Protecting groups Some protecting group manipulations of potential value in oligosaccharide synthesis have been introduced. Among these the so called uorous benzyl protection strategy appears most unconventional. The underlying idea is protection of the substrate with a group containing so many uorine substituents that the molecule becomes uorous with respect to simple uorous¡Xorganic phase separations by liquid¡Xliquid or liquid¡Xsolid extractions. This allows facile purication analogously to solid phase synthesis but the advantages of traditional solution phase synthesis are still retained. The principle has been applied to disaccharide synthesis by protecting the three hydroxy groups of a glucal donor with 4-[tris(peruorohexylethyl)silyl]benzyl bromide.The cyclic 4,6-O-phenyl boronate diester protection has been shown to allow convenient attachment of ethyl -galactothiopyranoside to a soluble PEG polymer bearing a 1,4-dioxyxylene linker. This protection was completely stable towards precipitation of the polymer bound glycoside with tert-butyl methyl ether and recrystallization from ethanol. 2,5-Dimethylpyrrole group has been used as an orthogonal amine-protecting group in oligosaccharide synthesis. The group is stable to conditions required for deprotection of the N-phthalimido group. As an alternative to the phthaloyl group the 4,5-dichlorophthaloyl group has been investigated as an amino protecting group. This group when introduced onto N-2 of glucosamine proved to have sucient stability under the standard conditions of protecting group manipulations such as deacetylation benzylation and benzylidenation and it also tolerated Lewis acid silver salt and iodonium ion promoted glycosylations.The group could be smoothly removed with ethylenediamine in alcohol solvent under substantially milder conditions than required for deprotection of the phthaloyl group. The tetrachlorophthaloyl group has been shown to be a viable global amine protecting group in the synthesis of polyglucosamine natural products allowing oligosaccharide assembly from n-pentenyl glycosides. Several tetrabutylammonium nitrite based systems have been developed for ecient deblocking of sugar N-phenylcarbamates. A mixture of guanidine and guanidinium nitrate has been 227 Annu. Rep. Prog. Chem. Sect. B 1999 95 207¡X234reported to result in O-deacetylation of sugars leaving the N-Troc protections intact.Tin() chloride deprotects regioselectively 4-methoxybenzyl ethers of carbohydrates. Enzymic synthesis of oligosaccharides Owing to the complexity of the chemical synthesis of oligosaccharides enzyme assisted methods play in synthetic carbohydrate chemistry a much more important role than in preparation of oligonucleotides or peptides. On the one hand enzymes have been utilized to obtain very complex oligosaccharide structures on the other hand they also oer potential methods for large scale synthesis of oligosaccharides. Glycosynthasecatalyzed transglycosidation of glycosyl uoride for example has been shown to yield the corresponding oligosaccharides in good yields. Trichoderma viride xylanase has been observed to be able to use a non-natural disaccharide 4-O--xylopyranosyl- -glucopyranosyl uoride as a glycosyl donor resulting in smooth polymerization to a cellulose¡Xxylan hybrid polymer.-1,4-Linked mannooligosaccharides have been obtained regioselectively by incubating mannose with -mannosidase from Rhizopus niveus. An improved method for the preparation of low molecular weight water soluble 2-deoxymaltooligosaccharides has been developed. The method uses silica gel immobilized maltopentaose as a recyclable primer in the reaction of glycogen phosphorylase with -glucal. Alternate (1,3)-(1,4)--glucooligosaccharides have been prepared by polymerizing -laminaribiosyl uoride by (1,3)-(1,4)--glucan 4- glucanohydrolase from Bacillus licheniformis. Many oligosaccharides that are recognized by receptors at the cell surface bear a terminal sialic acid residue.Accordingly preparation of sialylated oligosaccharides and oligosaccharides bearing related terminal determinants plays an important role in modern glycobiology. The chemical methods for glycosylation with sialic acid usually suer from side reactions although recently a 2-methylthioethyl ester protection on the sialyl donor has been shown to steer the glycosidation towards the -anomer. Hence enzymatic approaches based on sialyltransferases have been extensively used to obtain sialyl conjugates of oligosaccharides. Accordingly -(23)-sialyltransferase has been applied both in solution and immobilized to transfer the sialyl group from a variety of sialic acid nucleotides onto oligosaccharides. Galactosyl- sialyl- and fucosyl-transferases were used to obtain complex oligosaccharides that possess multiple terminal sialyl determinants.Sialylated and site-specically -(13)- fucosylated polylactosamines have been generated enzymatically. The other enzyme assisted syntheses of particular interest include chemoenzymatic preparation of the partial structure common to all high mannose-type sugar chains and complex-type sugar chains and the partial structure of many glycoproteins a glycosylated asparagine carrying a complex type undecasaccharide N-glycan. 228 Annu. Rep. Prog. Chem. Sect. B 1999 95 207¡X234Abbreviations Aib Alloc Boc BNO BOMP BOP DBU DCC DDQ DMTr DMTST DTT EDAC Fmoc Fpmp HATU HBTU HOAt HOBt HODhbt LCAA-CPG long chain amino alkyl controlled pore glass MeOTf MSNT NIS NMP ODN PEG pixyl PyAOP PyBOP PyBrOP TBDMS TES TESOTf TFA TfOH TMSOTf TPS Troc -aminoisobutyric acid allyloxycarbonyl tert-butoxycarbonyl 2-(phenylsulfonyl)-3-(3-nitrophenyl)oxaziridine 2-(benzotriazol-1-yloxy)-1,1-dimethyl-2-pyrrolidin-1-yl-1,3,2- diazaphospholidinium hexa.uorophosphate (benzotriazolyloxy)tris(dimethylamino)phosphonium hexa.uorophosphate 1,8-diazabicyclo[5.4.0]undec-7-ene 1,3-dicyclohexylcarbodiimide 2,3-dichloro-5,6-dicyanobenzo-1,4-quinone 4,4-dimethoxytrityl dimethyl(methylthio)sulfonium tri.uoromethanesulfonate .,.-dithiothreitol 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (.uoren-9-yl)methoxycarbonyl 1-(2-.uorophenyl)-4-methoxypiperid-4-yl N-(dimethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene- N-methylmethanaminium hexa.uorophosphate N-oxide N-[(1H-benzotriazol-1-yl)(dimethylamino)methylene]-Nmethylmethanaminium hexa.uorophosphate N-oxide 1-hydroxy-7-azabenzotriazole 1-hydroxybenzotriazole 3-hydroxy-3,4-dihydro-4-oxo-1,2,3-benzotriazine methyl tri.uoromethanesulfonate 1-(mesitylene-2-sulfonyl)-3-nitro-1,2,4-triazole N-iodosuccinimide N-methylpyrrolidone oligodeoxyribonucleotide polyethylene glycol 9-phenylxanthen-9-yl (7-azabenzotriazol-1-yloxy)tris(pyrrolidino)phosphonium hexa.uorophosphate (benzotriazol-1-yl)oxytripyrrolidinephosphonium hexa.uorophosphate bromotrispyrrolidinophosphonium hexa.uorophosphate tert-butyldimethylsilyl triethylsilyl triethylsilyl tri.uoromethanesulfonate tri.uoroacetic acid tri.uoromethanesulfonic acid trimethylsilyl tri.uoromethanesulfonate triphenylsilyl 2,2,2-trichloroethoxycarbonyl 229 Annu.Rep. Prog. Chem. Sect. B 1999 95 207—234 References 1 R.P. Iyer D. Yu I. Habus N.-H. Ho S. Johnson T. Devlin Z. Jiang W. Zhou J. Xie and S. Agrawal Tetrahedron 1997 53 2731. 2 H. Waldmann and A. Reidel Angew. Chem. Int. Ed. Engl. 1997 36 647. 3 K. Kamaike H. Takahashi T. Kakinuma K. Morohoshi and Y. Ishido Tetrahedron Lett. 1997 38 6857. 4 A. Wilk K. Srinivasachar and S. L. Beaucage J. Org. Chem. 1997 62 6712. 5 C. Vargeese J. Carter J. Yegge S. Krivjansky A. Settle E. Kropp K. Peterson and W. Pieken Nucleic Acids Res. 1998 26 1046.6 Y. Hayakawa and M. Kataoka J. Am. Chem. Soc. 1997 119 11 758. 7 F. Natt and R. Haner Tetrahedron 1997 53 9629. 8 T. Wada Y. Sato F. Honda S.-i. Kawahara and M. Sekine J. Am. Chem. Soc. 1997 119 12 710. 9 R.T. Pon and S. Yu Nucleic Acids Res. 1997 25 3629. 10 T. Wada A. Kobori S.-i. Kawahara and M. Sekine Tetrahedron Lett. 1998 39 6907. 11 R. T. Pon and S. Yu Tetrahedron Lett. 1997 38 3331. 12 C. Scheuer-Larsen C. Rosenbohm T. J.D. Joergensen and J. Wengel Nucleosides Nucleotides 1997 16 67. 13 P. S. Nelson S. Muthini M. Vierra L. Acosta and T. H. Smith BioTechniques 1997 22 752. 14 P. Kumar and K. C. Gupta Nucleic Acids Res. 1997 25 5127. 15 A. Bagno S. Bicciato C. Di Bello and G.M. Bonora J. Chem. Technol. Biotechnol. 1998 71 77. 16 M. Meldgaard N.K. Nielsen M. Bremner O. S. Pedersen C. E. Olsen and J. Wengel J. Chem. Soc. Perkin Trans. 1 1997 1951. 17 R. S. Braich and M.J. Damha Bioconjugate Chem. 1997 8 370. 18 M.J. Damha and R. S. Braich Tetrahedron Lett. 1998 39 3907. 19 M. Grotli R. Eritja and B. Sproat Tetrahedron 1997 53 11 317. 20 R. H. E. Hudson S. Robidoux and M. J. Damha Tetrahedron Lett. 1998 39 1299. 21 T. Horn C.-A. Chang and M. S. Urdea Nucleic Acids Res. 1997 25 4835. 22 M.S. Shchepinov I. A. Udalova A. J. Bridgman and E. M. Southern Nucleic Acids Res. 1997 25 4447. 23 E. Alazzouzi N. Escaja A. Grandas and E. Pedroso Angew. Chem. Int. Ed. Engl. 1997 36 1506. 24 G. H. McGall A. D. Barone M. Diggelmann S. P. A. Fodor E. Gentalen and N. Ngo J. Am. Chem. Soc. 1997 119 5081.25 M.C. Pirrung L. Fallon and G. H. McGall J. Org. Chem. 1998 63 241. 26 J. E. Beecher G. H. McGall and M. J. Goldberg Polym. Mater. Sci. Eng. 1997 76 597. 27 G. Cohen J. Deutsch J. Fineberg and A. Levine Nucleic Acids Res. 1997 25 911. 28 R. Bader M. Hinz B. Schu and H. Seliger Nucleosides Nucleotides 1997 16 829; R. Bader H. Brugger M. Hinz C. Rembe E. P. Hofer and H. Seliger Nucleosides Nucleotides 1997 16 835; J. Weiler and J. D. Hoheisel Nucleosides Nucleotides 1997 16 1793. 29 D. L. McMinn R. Hirsch and M. M. Greenberg Tetrahedron Lett. 1998 39 4155. 30 C. Dell’Aquila J.-L. Imbach and B. Rayner Tetrahedron Lett. 1997 38 5289. 31 A. Guzaev and H. Lo� nnberg Tetrahedron Lett. 1997 38 3989. 32 X. Zhang B. L. Ga.ney and R. A. Jones Nucleic Acids Res.1997 25 3980. 33 M.H. Lyttle H. Adams D. Hudson and R. M. Cook Bioconjugate Chem. 1997 8 193. 34 P. S. Nelson S. Muthini M. A. Kent and T. H. Smith Nucleosides Nucleotides 1997 16 1951. 35 M. Sobkowski A. Kraszewski and J. Stawinski Nucleosides Nucleotides 1998 17 253. 36 N. Haginoya A. Ono Y. Nomura Y. Ueno and A. Matsuda Bioconjugate Chem. 1997 8 271. 37 K. Shinozuka A. Umeda T. Aoki and H. Sawai Nucleosides Nucleotides 1998 17 291. 38 J. Hovinen and H. Salo J. Chem. Soc. Perkin Trans. 1 1997 3017. 39 H. Salo A. Guzaev and H. Lo� nnberg Bioconjugate Chem. 1998 9 365. 40 I. Habus J. Xie R. P. Iyer W.-Q. Zhou L. X. Shen and S. Agrawal Bioconjugate Chem. 1998 9 283. 41 J. M. Tomkins K. J. Barnes A. J. Blacker W. J. Watkins and C. Abell Tetrahedron Lett. 1997 38 691.42 S. Kohgo K. Shinozuka H. Ozaki and H. Sawai Tetrahedron Lett. 1998 39 4067. 43 J. G. Harrison and S. Balasubramanian Bioorg. Med. Chem. Lett. 1997 7 1041. 44 D. L. McMinn and M. M. Greenberg J. Am. Chem. Soc. 1998 120 3289. 45 D. L. McMinn T. J. Matray and M. M. Greenberg J. Org. Chem. 1997 62 7074. 46 R. S. Coleman J. C. Arthur and J. L. McCary Tetrahedron 1997 53 11 191. 47 H.-Y. Kim L. Nechev L. Zhou P. Tamura C.M. Harris and T.M. Harris Tetrahedron Lett. 1998 39 6803. 48 E. Trevisiol A. Renard E. Defrancqand J. Lhomme Tetrahedron Lett. 1997 38 8687. 230 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 49 M. Dechamps and E. Sonveaux Nucleosides Nucleotides 1998 17 697. 50 S. Soukchareun J. Haralambidis and G. Tregear Bioconjugate Chem. 1998 9 466.51 C. N. Tetzla. I. Schwope C. F. Bleczinski J. A. Steinberg and C. Richert Tetrahedron Lett. 1998 39 4215. 52 J. Robles M. Maseda M. Beltran M. Concernau E. Pedroso and A. Grandas Bioconjugate Chem. 1997 8 785. 53 J. D. Kahl D. L. McMinn and M. M. Greenberg J. Org. Chem. 1998 63 4870. 54 S. A. Kuznetsova I. E. Kanevsky and Z. A. Shabarova FEBS Lett. 1998 431 453. 55 M.G. Brevnov O. M. Gritsenko S. N. Mikhailov E. V. E.mtseva B. S. Ermolinsk A. Van Aerschot P. Herdewijn A. V. Repytk and E. S. Gromova Nucleic Acids Res. 1997 25 3302. 56 T. Yamaguchi K. Suyama K. Narita S. Kohgo A. Tomikawa and M. Saneyoshi Nucleic Acids Res. 1997 25 2352. 57 W.R. Kobertz and J. M. Essigmann J. Am. Chem. Soc. 1997 119 5960. 58 R. Sleman and R. M. Pires Nucleic Acids Res.1997 25 4771. 59 M.W. Reed E. A. Lukhtanov V. Gorn I. Kutyavin A. Gall A. Wald and R. B. Meyer Bioconjugate Chem. 1998 9 64. 60 J. Hall D. Husken and R. Haner Nucleosides Nucleotides 1997 16 1357. 61 M. Endo Y. Azuma Y. Saga A. Kuzuya G. Kawai and M. Komiyama J. Org. Chem. 1997 62 846. 62 B. Mestre M. Pitie C. Loup C. Glaparols G. Pratviel and B. Meunier Nucleic Acids Res. 1997 25 1022. 63 V. V. Tolstikov D. A. Stetsenko V. K. Potapov and E. D. Sverdlov Nucleosides Nucleotides 1997 16 215. 64 Y. Ueno M. Mikawa and A. Matsuda Bioconjugate Chem. 1998 9 33. 65 F. Seela and Y. Chen Helv. Chim. Acta 1997 80 1073. 66 G. Balow V. Mohanj E. A. Lesnik J. F. Johnston B. P. Monia and O. L. Acevedo Nucleic Acids Res. 1998 26 3350. 67 S. Hildbrand A.Blaser S. P. Parel and C. J. Leumann J. Am. Chem. Soc. 1997 119 5499. 68 K. Kawai I. Saito and H. Sugiyama Tetrahedron Lett. 1998 39 5221. 69 G. Xiang and L. W. McLaughlin Tetrahedron 1998 54 375. 70 F. Seela and C. Wei Helv. Chim. Acta 1997 80 73. 71 F. Seela and C. Wei Chem. Commun. 1997 1869. 72 A. A. Koshkin S. K. Singh P. Nielsen V. K. Rajwanshi R. Kumar M. Meldgaard C. E. Olsen and J. Wengel Tetrahedron 1998 54 3607; S. Obika D. Nanbu Y. Hari J.-I. Andoh K.-I. Morio T. Doi and T. Imanishi Tetrahedron Lett. 1998 39 5401. 73 N. K. Christensen M. Petersen P. Nielsen J. P. Jacobsen C. E. Olsen and J. Wengel J. Am. Chem. Soc. 1998 120 5458. 74 S. M. Gryaznov Nucleosides Nucleotides 1997 16 899. 75 J. Baraniak D. Korczynaki R. Kaczmarek and E.Wasilewska Nucleosides Nucleotides 1998 17 1347. 76 K. L. Fearon B. L. Hirschbein J. S. Nelson M. F. Foy M.Q. Nguyen A. Okruszek S. N. McCurdy J. E. Frediani L. A. DeDionisio A.M. Raible E. N. Cagle and V. Boyd Nucleic Acids Res. 1998 26 3813. 77 N. Mignet and S. M. Gryaznov Nucleic Acids Res. 1998 26 431. 78 D. S. Sergueev and B. R. Shaw J. Am. Chem. Soc. 1998 120 9417. 79 Z. Zhang A. Nichols M. Alsbeti J. X. Tang and J. Y. Tang Tetrahedron Lett. 1998 39 2467. 80 A. H. Krotz P. Klopchin D. L. Cole and V. T. Ravikumarf Bioorg. Med. Chem. Lett. 1997 7 73. 81 W.J. Stec B. Karwowski M. Borzkowska P. Guga M. Koziolkiewicz M. Sochacki M. W. Wieczorek and J. Blasczyk J. Am. Chem. Soc. 1998 120 7156. 82 R. P. Iyer M.J. Guo D. Yu and S. Agrawal Tetrahedron Lett.1998 39 2491. 83 C. B. Reese and Q. Song Bioorg. Med. Chem Lett. 1997 7 2787. 84 C. B. Reese Q. Song M.V. Rao and I. Beckett Nucleosides Nucleotides 1998 17 451. 85 G. Tosquellas K. Alvarez C. Dell’aquila F. Morvan J.-J. Vasseur J.-L. Imbach and B. Rayner Nucleic Acids Res. 1998 26 2069. 86 G. Tosquellas F. Morvan B. Rayner and J.-L. Imbach Bioorg. Med. Chem. Lett. 1997 7 263. 87 R. P. Iyer N.-H. Ho D. Yu and S. Agrawal Bioorg. Med. Chem. Lett. 1997 7 871. 88 A. El-Faham Chem. Lett. 1998 671. 89 K. Akaji Y. Tamai and Y. Kiso Tetrahedron 1997 53 567. 90 G. Jou I. Gonzales F. Albericio P. Lloyd-Williams and E. Giralt J. Org. Chem. 1997 62 354. 91 R. Liu and L. E. Orgel J. Am. Chem. Soc. 1997 119 4791. 92 T. Vajda G. Szokan and M. Hollosi J. Pept. Sci.1998 4 300. 93 Y. Han F. Albericio and G. Barany J. Org. Chem. 1997 62 4307. 94 M. Meldal M.A. Juliano and A. M. Jansson Tetrahedron Lett. 1997 38 2531. 95 L. A. Carpino D. Ionescu A. El-Faham P. Henklein H. Wenschuh M. Bienert and M. Beyermann Tetrahedron Lett. 1998 39 241. 96 F. Albericio M. Cases J. Alsina S. A. Triolo L. A. Carpino and S. A. Kates Tetrahedron Lett. 1997 38 4853. 97 E. Nicolas M. Pujades J. Bacardit E. Giralt and F. Albericio Tetrahedron Lett. 1997 38 2317; W. Zeng P.-O. Regamey K. Rose Y. Wang and E. Bayer J. Pept. Res. 1997 49 273. 231 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 98 J. O.er T. Johnson and M. Quibell Tetrahedron Lett. 1997 38 9047. 99 E. Oliveira A. Miranda F. Albericio D. Andreu A. C. M. Paiva C. R. Nakaie and M.Tominaga J. Pept. Res. 1997 49 300. 100 D. Andreu and F. J. Garcia Lett. Pept. Sci. 1997 4 41. 101 L. M. Varanda and M. T. M. Miranda J. Pept. Res. 1997 50 102. 102 M. Dettin S. Pegoraro P. Rovero S. Bicciato A. Bagno and C. Di Bello J. Pept. Res. 1997 49 103. 103 L. P. Miranda A. Jones W. D. F. Meutermans and P. F. Alewood J. Am. Chem. Soc. 1998 120 1410. 104 M. Breslav J. Becker and F. Naider Tetrahedron Lett. 1997 38 2219. 105 R. L. E. Furlan E. G. Mata and O. A. Mascaretti J. Chem. Soc. Perkin Trans. 1 1998 355. 106 R. Leger R. Yen M. W. She V. J. Lee and S. J. Hecker Tetrahedron Lett. 1998 39 4171. 107 B. Henkel L. Zhang and E. Bayer Liebigs Ann./Recl. 1997 2161. 108 K. H. Bleicher and J. R. Wareing Tetrahedron Lett. 1998 39 4591. 109 H.M.Zhong M.N. Greco and B. E. Maryano. J. Org. Chem. 1997 62 9326. 110 N. Thieriet J. Alsina E. Giralt F. Guibe and F. Albericio Tetrahedron Lett. 1997 38 7275. 111 S. Lemaire-Audoire M. Savignac E. Blart J.-M. Bernard and J. P. Genet Tetrahedron Lett. 1997 38 2955. 112 L. A. Carpino M. Philbin M. Ismail G. A. Truran E. M.E. Mansour S. Iguchi D. Ionescu A. El-Faham C. Riemer R. Warrass and M.S. Weiss J. Am. Chem. Soc. 1997 119 9915. 113 G. Mezo N. Mihala G. Koczan and F. Hudecz Tetrahedron 1998 54 6757. 114 B. Kellam B. W. Bycroft W. C. Chan and S. R. Chhabra Tetrahedron 1998 54 6817. 115 O. Seitz and H. Kunz J. Org. Chem. 1997 62 813. 116 J. Bodi Y. Nishiuchi H. Nishio T. Inui and T. Kimura Tetrahedron Lett. 1998 39 7117. 117 K. Rosenthal A. Karlstrom and A.Unden Tetrahedron Lett. 1997 38 1075. 118 Y. Okada N. Shintomi Y. Kondo T. Yokoi S. Joshi and W. Li Chem. Pharm. Bull. 1997 45 1860. 119 Y. Okada and Y. Mu Chem. Pharm. Bull. 1997 45 88. 120 R. S. Givens A. Jung C.-H. Park J. Weber and W. Bartlett J. Am. Chem. Soc. 1997 119 8369. 121 M. Beltran E. Pedroso and A. Grandas Tetrahedron Lett. 1998 39 4115. 122 H. E. Garay L. J. Gonzalez L. J. Cruz R. C. Estrada and O. Reyes Biotecnol. Appl. 1997 14 193. 123 A.M. Kimbonguila S. Boucida F. Guibe and A. Lo.et Tetrahedron 1997 53 12 525. 124 Y. Okada J. Wang T. Yamamoto T. Yokoi and Y. Mu Chem. Pharm. Bull. 1997 45 452. 125 M. Noda and M. Ki.e J. Pept. Res. 1997 50 329. 126 Y. Han and G. Barany J. Org. Chem. 1997 62 3841. 127 G. Talbo J. D. Wade N. Dawson M.Manoussios and G.W. Tregear Lett. Pept. Sci. 1997 4 121; B. J. Egner and M. Bradley Tetrahedron 1997 53 14 021. 128 C. Drouot C. Enjalbal P. Fulcrand J. Martinez J.-L. Aubagnac R. Combarieu and Y. De Puydt Tetrahedron Lett. 1997 38 2455. 129 J. Ryttersgaard B. D. Larsen A. Holm D. H. Christensen and O. F. Nielsen Spectrochim. Acta Part A 1997 53 91. 130 E. M. Cilli R. Marchetto S. Schreier and C. R. Nakaie Tetrahedron Lett. 1997 38 517. 131 C. Dhalluin C. Boutillon A. Tartar and G. Lippens J. Am. Chem. Soc. 1997 119 l0 494. 132 N. J. Wells M. Davies and M. Bradley J. Org. Chem. 1998 63 6430. 133 J. Bodi H. Suli-Vargha K. Laudanyi K. Fekey and G. Orosz Tetrahedron Lett. 1997 38 3293. 134 F. Bordusa D. Ullmann C. Elsner and H.-D. Jakubke Angew. Chem. Int.Ed. Engl. 1997 36 2473. 135 S. Futaki and K. Kitagawa Tetrahedron 1997 53 7479. 136 K. J. Jensen J. Alsina M. F. Songster J. Vagner F. Albericio and G. Barany J. Am. Chem. Soc. 1998 120 5441. 137 J. Alsina F. Rabanal C. Chiva E. Giralt and F. Albericio Tetrahedron 1998 54 10 125. 138 M. Davis and M. Bradley Angew. Chem. Int. Ed. Engl. 1997 36 1097. 139 N. J. Ede and A.M. Bray Tetrahedron Lett. 1997 38 7119. 140 N. Galeotti E. Plagnes and P. Jouin Tetrahedron Lett. 1997 38 2459. 141 J. A. Fehrentz M. Paris A. Heitz J. Velek F. Winternitz and J. Martinez J. Org. Chem. 1997 62 6792. 142 B. J. Hall and J. D. Sutherland Tetrahedron Lett. 1998 39 6593; C. Pothion M. Paris A. Heitz L. Rocheblave F. Rouch J.-A. Fehrentz and J. Martinez Tetrahedron Lett. 1997 38 7749.143 I. Vlattas J. Dellure.cio R. Dunn I. Sytwu and J. Stanton Tetrahedron Lett. 1997 38 7321. 144 E. Nicolas J. Clemente T. Ferrer F. Albericio and E. Giralt Tetrahedron 1997 53 3179. 145 A. Pichette N. Voyer R. Larouche and J.-C. Meillon Tetrahedron Lett. 1997 38 1279. 146 D. Sarantakis and J. J. Bicksler Tetrahedron Lett. 1997 38 7325. 147 C. R. Millington R. Quarrell and G. Lowe Tetrahedron Lett. 1998 39 7201. 148 M. Meisenbach H. Echner and W. Voelter Chem. Commun. 1997 849. 149 B. Henkel W. Zeng and E. Bayer Tetrahedron Lett. 1997 38 3511. 150 J. J. Chen and A. F. Spatola Tetrahedron Lett. 1997 38 1511. 151 J. A.W. Kruijtzer L. J. F. Hohneyer W. Heerma C. Versluis and R. M.J. Liskamp Chem. Eur. J. 1998 4 1570. 152 K. Burgess J. Ibarzo D.S. Linthicum H. Shin A. Shitangkoon R. Totani and A. J. Zhang J. Am. Chem. 232 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 Soc. 1997 119 1556. 153 A. Schleyer M. Meldal R. Manat H. Paulsen and K. Bock Angew. Chem. Int. Ed. Engl. 1997 36 1976. 154 Z.-W. Guo Y. Nakahara Y. Nakahara and T. Ogawa Angew. Chem. Int. Ed. Engl. 1997 36 1464. 155 N. Mathieux H. Paulsen M. Meldal and K. Bock J. Chem. Soc. Perkin Trans. 1 1997 2359. 156 J. Broddefalk K.-E. Bergqvist and J. Kihlberg Tetrahedron 1998 54 12 047. 157 Y. Nakahara Y. Nakahara Y. Ito and T. Ogawa Tetrahedron Lett. 1997 38 7211. 158 B. Liebe and H. Kunz Helv. Chim. Acta 1997 80 1473. 159 O. Seitz and C.-H. Wong J. Am. Chem. Soc. 1997 119 8766. 160 K. Yamamoto K. Fujimori K. Haneda M. Mizuno T. Inazu and H.Kumagai Carbohydr. Res. 1997 305 415. 161 E. C. Rodriguez K. A. Winans D. S. King and C. R. Bertozzi J. Am. Chem. Soc. 1997 119 9905. 162 G. Shapiro D. Buchler C. Dalvit P. Frey M. Del Carmen Fernandez B. Gomez-Lor E. Pombo-Villar U. Stauss R. Swoboda and C. Waridel Bioorg. Med. Chem. 1997 5 147. 163 T. Wakamiya R. Togashi T. Nishida K. Saruta J.-I. Yasuoka S. Kusumoto S. Aimoto Y. K. Kumagaye K. Nakajima and K. Nagata Bioorg. Med. Chem. 1997 5 135. 164 A. Paquet B. Blackwell M. Johns and J. Nikiforuk J. Pept. Res. 1997 50 262. 165 D. Sebastian A. Heuser S. Schulze and H. Waldmann Synthesis 1997 1098. 166 T. Pohl and H. Waldmann J. Am. Chem. Soc. 1997 119 6702. 167 U. Diederichsen Angew. Chem. Int. Ed. Engl. 1997 36 1886. 168 U. Diederichsen and H.W. Schmitt Angew. Chem. Int. Ed. Engl. 1998 37 302. 169 N.M. Howarth and L. P. G. Wakelin J. Org. Chem. 1997 62 5441. 170 P. Lagri.oule P. Wittung M. Eriksson K. K. Jensen B. Norden O. Buchardt and P. E. Nielsen Chem. Eur. J. 1997 3 912. 171 A. C. Van der Laan I. Van Amsterdam G. I. Tesser J. H. Van Boom and E. Kuyl-Yeheskiely Nucleosides Nucleotides 1998 17 219. 172 L. Zhang and J. P. Tam Tetrahedron Lett. 1997 38 4375. 173 J. P. Tam and Y.-A. Lu Tetrahedron Lett. 1997 38 5599. 174 L. Zhang and J. P. Tam J. Am. Chem. Soc. 1997 119 2363. 175 Y. Shao W. Lu and S. B. H. Kent Tetrahedron Lett. 1998 39 3911. 176 S. Gangwar G.M. Pauletti T. J. Siahaan V. J. Stella and R. T. Borchardt J. Org. Chem. 1997 62 1356. 177 B. Wang S. Gangwar G.M. Pauletti T.J. Siahaan and R. T. Borchardt J. Org. Chem. 1997 62 1363. 178 K. C. Nicolaou N. Wissinger J. Pastor and F. DeRoose J. Am. Chem. Soc. 1997 119 449. 179 K. C. Nicolaou N. Watanabe J. Li J. Pastor and N. Winssinger Angew. Chem. Int. Ed. Engl. 1998 37 1559. 180 C. Zheng P. H. Seeberger and S. J. Danishefsky J. Org. Chem. 1998 63 1126. 181 P. H. Seeberger M. Eckhardt C. E. Gutteridge and S. J. Danishefsky J. Am. Chem. Soc. 1997 119 10 064. 182 C. Zheng P. H. Seeberger and S. J. Danishefsky Angew. Chem. Int. Ed. Engl. 1998 37 786. 183 P. H. Seeberger X. Beebe G. D. Sukenick S. Pochapsky and S. J. Danishefsky Angew. Chem. Int. Ed. Engl. 1997 36 491. 184 J. Rademann and R. R. Schmidt J. Org. Chem. 1997 62 3650. 185 J. Rademann A. Geyer and R. R. Schmidt Angew.Chem. Int. Ed. Engl. 1998 37 1241. 186 A. Heckel E. Mross K. H. Jung J. Rademann and R. R. Schmidt Synlett 1998 171. 187 M. Adinol. G. Barone L. De Napoli A. Iadonisi and G. Piccialli Tetrahedron Lett. 1998 39 1953. 188 R. Rodebaugh S. Joshi B. Fraser-Reid and H. M. Geysen J. Org. Chem. 1997 62 5660. 189 Y. Ito and T. Ogawa J. Am. Chem. Soc. 1997 119 5562. 190 P. P. Deshpande H.M. Kim A. Zatorski T.-K. Park G. Ragupathi P. O. Livingston D. Live and S. J. Danishefsky J. Am. Chem Soc. 1998 120 1600. 191 P. Grice S. V. Ley J. Pietruszka H. M. I. Osborn H.W. M. Priepke and S. L. Stuart Chem. Eur. J. 1997 3 431. 192 M. K. Cheung N. L. Douglas B. Hinzen S. V. Ley and X. Pannecoucke Synlett 1997 257. 193 L. Green B. Hinzen S. J. Ince P. Langer S. V. Ley and S.L. Warriner Synlett 1998 440. 194 C. Unverzagt Angew. Chem. Int. Ed. Engl. 1997 36 1989. 195 W. Zou and H. J. Jennings Bioorg. Med. Chem. Lett. 1997 7 647. 196 Z.-W. Guo Y. Nakahara and T. Ogawa Tetrahedron Lett. 1997 38 4799. 197 S. Weiler and R. R. Schmidt Tetrahedron Lett. 1998 39 2299. 198 Y. Guo and G. A. Sulikowski J. Am. Chem. Soc. 1998 120 1392. 199 W. Wang and F. Kong Tetrahedron Lett. 1998 39 1937. 200 N. Otsubo H. Ishida M. Kiso and A. Hasegawa Carbohydr. Res. 1998 306 517. 201 P. Bourhis F. Machetto P. Duchaussoy J.-P. Herault J.-M. Mallet J.-M. Herbert M. Petitou and P. Sinay Bioorg. Med. Chem. Lett. 1997 7 2843. 202 M. Ludewig and J. Thiem Synthesis 1998 56. 203 G. J. Boons and T. Zhu Synlett 1997 809. 204 A. Demchenko and G.-J.Boons Tetrahedron Lett. 1997 38 1629. 233 Annu. Rep. Prog. Chem. Sect. B 1999 95 207—234 205 V. Pozsgay J. Org. Chem. 1998 63 5983. 206 C.M. Dreef-Tromp H. A. M. Willems P. Westerduin P. van Veelen and C. A. A. van Boeckel Bioorg. Med. Chem. Lett. 1997 7 1175. 207 S.-i. Hashimoto H. Sakamoto T. Honda and S. Ikegami Tetrahedron Lett. 1997 38 5181. 208 T. Zhu and G.-J. Boons Tetrahedron Lett. 1998 39 2187. 209 D. Crich and S. Sun J. Am. Chem. Soc. 1998 120 435. 210 W. Wang and F. Kong J. Org. Chem. 1998 63 5744. 211 T. Iimori H. Ohtake and S. Ikegami Tetrahedron Lett. 1997 38 3415. 212 G. Hodosi and P. Kovac J. Am. Chem. Soc. 1997 119 2335. 213 S. K. Chatterjee and P. Nuhn Chem. Commun. 1998 1729. 214 B. A. Garcia J. L. Poole and D. Y. Gin J. Am. Chem.Soc. 1997 119 7597. 215 O. Kwon and S. J. Danischefsky J. Am. Chem. Soc. 1998 120 1588. 216 D. P. Curran R. Ferritto and Y. Hua Tetrahedron Lett. 1998 39 4937; D. P. Curran S. Hadida S. He and M. He J. Org. Chem. 1997 62 6714. 217 G. G. Cross and D. M. Whit.eld Synlett 1998 487. 218 S. G. Bowers D.M. Coe and G.-J. Boons J. Org. Chem. 1998 63 4570. 219 M. Lergenmuller Y. Ito and T. Ogawa Tetrahedron 1998 54 1381. 220 R. Rodebaugh J. S. Debenham and B. Fraser-Reid J. Carbohydr. Chem. 1997 16 1407. 221 S. Akai N. Niahino Y. Iwata J.-I. Hiyama E. Kawashima K.-I. Sato and Y. Ishido Tetrahedron Lett. 1998 39 5583. 222 U. Ellervik and G. Magnusson Tetrahedron Lett. 1997 38 1627. 223 K. P. Kartha M. Kiso A. Hasegawa and H. J. Jennings J. Carbohydr. Chem.1998 17 811. 224 L. F. Mackenzie Q. Wang R. A. J. Warren and S. G. Withers J. Am. Chem. Soc. 1998 120 5583. 225 M. Fujita S.-i. Shoda and S. Kobayashi J. Am. Chem. Soc. 1998 120 6411. 226 H. Fujimoto M. Isomura and K. Ajisaka Biosci. Biotechnol. Biochem. 1997 61 164. 227 B. Evers and J. Thiem Bioorg. Med. Chem. 1997 857. 228 J.-L. Viladot V. Moreau A. Planas and H. Driguez J. Chem. Soc. Perkin Trans. 1 1997 2383. 229 A. Takahashi H. Tsukamoto and H. Yamada Tetrahedron Lett. 1997 38 8223. 230 M. D. Chappell and R. L. Halcomb J. Am. Chem. Soc. 1997 119 3393. 231 A. Lubineau K. Basset-Carpentier and C. Auge Carbohydr. Res. 1997 300 161. 232 V. H. Thomas J. Elhalabi and K. G. Rice Carbohydr. Res. 1998 306 387. 233 J. Rabina J. Natunen R. Niemela H. Salminen K. Ilves O. Aitio H. Maaheimo J. Helin and O. Renkonen Carbohydr. Res. 1997 305 491; H. Salminen K. Ahokas R. Niemela� L. Penttila� H. Maaheimo J. Helin C. E. Costello and O. Renkonen FEBS Lett. 1997 419 220. 234 I. Matsuo M. Isomura T. Miyazaki T. Sakakibara and K. Ajisaka Carbohydr. Res. 1997 305 401. 235 C. Unverzagt Carbohydr. Res. 1997 305 423. 234 Annu. Rep. Prog. Chem. Sect. B 1999 95 207&mdash

ISSN:0069-3030    

DOI:10.1039/a808579j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Synthesis highlights a review of the literature for 1998 8 Peter Quayle Department of Chemistry University of Manchester Manchester UK M13 9PL 1 Introduction This year saw the 170th anniversary of the .rst and often misrepresented synthesis of an organic compound. In his landmark paper of 1828,Wo� hler reported the serendipitous synthesis of urea from ammonium cyanate a reaction which incidentally takes place in both the solution and solid state. The solid state variant of this transformation —another .rst—has been the subject of much investigation but it is only with the advent of sophisticated X-ray di.raction techniques that a detailed picture of this reaction has come to be unravelled. Since Wo� hler’s investigations the practice of organic synthesis has increased exponentially both in terms of the degree of complexity and structural variation of targets under consideration.Such advances have been made possible by the ready availability of advanced analytical techniques (vide supra) and the realisation by Barton and others, that chemical reactivity with the non d-block elements at least can be correlated with constitutional and stereochemical features of a given molecule. That these principles are self evident is now witnessed by the fact that the interplay between mechanism and stereochemistry provides the cornerstone for most modern texts on the subject. Indeeed it is almost impossible to underestimate the in.uence of Barton on the development of contemporary organic chemistry. Because of the central importance of the principles of conformational analysis exceptions to the often quoted textbook guidelines are continually being searched for.A recent example concerning the identi.cation of ‘‘completely stable axial conformers of monosubstituted cyclohexanes at room temperature’’ appears to be fallacious. The remarkable pace at which the total synthesis of even complex structures can be accomplished has continued this year. In addition the application of combinatorial techniques means that the synthesis of analogues of natural products with potentially interesting biological pro.les can be accomplished relatively rapidly in certain cases. This situation is clearly exempli.ed in the synthesis of structural analogues of the epothilones 1 and derivatives 2 as outlined by Nicolaou. In terms of advances in synthetic methodology the development of a family of catalysts which enable the clean metathesis of functionalised ole.ns has elicited numerous applications the synthesis of ole.ns in this manner has started to rival more conventional methodology such as the Wittig reaction and must rank among one of 235 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 O O S S n HO HO N N O O O O O O OH 1 epothilone A OH 2 n = 1 [14]epothilone A n = 2 [15]epothilone A n = 3 epothilone A n = 4 [17]epothilone A n = 5 [18]epothilone A the most important developments of the decade. The observation that readily available yet usually relatively inert aryl chlorides can be coaxed to undergo Heck Stille and Suzuki reactions under mild conditions merely by judicious choice of ligand will serve to widen the use of these reactions further whilst synthetic applications of polymerbound and solid supported reagents and reactions seemingly show no limit. 2 Brevetoxins and related compounds The development of e.cient strategies for the synthesis of medium-ring ethers polyfused ethers and spirocyclic hemiacetals continues to be the focus of much attention.A number of natural products notably toxins of marine origin such as the brevetoxins and ciguatoxins non-terpenoid C -metabolites derived from fatty acid metabolism such as dactoelyne the bryostatins and the spongistatins possess interesting biological pro.les (e.g. modulation of sodium transport or antineoplastic activity) but their low natural abundance precludes a full evaluation of their pharmacology hence the synthesis of the natural products themselves or analogues thereof is pro.ered as a rationale for many of these investigations.Parsons and Ibuka have reported complementary approaches to the synthesis of furanopyrans which utilise di.erent facets of the chemistry of sulfur to e.ect cyclisation. In the .rst example an allenyl sufoxide is used in a tandem intramolecular Michael reaction whilst the second example utilises the redox chemistry of cyclopropyl sul.des to generate a sulfur stabilised cation which su.ers intramolecular capture with a pendent hydroxy moiety (Scheme 1). New approaches to the synthesis of fused polyether motifs—all of which utilise metathesis chemistry—have been disclosed by the groups of Rainier, Hirama and Clark. Rainier’s route employs Danishefsky’s procedure (DMDO) for the stereoselective epoxidation of a suitably functionalised unsaturated sugar derivative followed by regio- and stereospeci.c ring opening with allylmagnesium bromide in THFto a.ord the alcohol 3.Acylation of the free hydroxy group of 3 Takai methylene transfer (CH Br /Zn/TiCl /PbCl ) followed by a ring closure metathesis reaction (RCM) using the molybdenum catalyst 4 a.orded the bicyclic enol ethers in respectable overall yield (ca. 50%) Scheme 2. This procedure can be performed in an iterative sense leading to the synthesis of tricyclic systems in a relatively rapid sequence this 236 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 OTBS HO SPh OTBS Ph(O)S O (ii) (i) • O OTBS H OTBS Cl H SPh OMe O OMe OMe (iii) Cl + O OH O OH O H 33% 34% (iv) 51% Scheme 1 Reagents (i) PhSCl Et N Et O (94%); (ii) HF (aq.) CH CN 60 °C (49%); (iii) CAN Me NCl 3Å MS MeOH; (iv) p-TsOH PhH. Scheme 2 Reagents (i) DMDO CH Cl ; (ii) CH ——CHCH MgBr THF; (iii) Ac O; (iv) CH Br Zn TiCl PbCl ; (v) (RO) Mo(—— NAr)——CHCMe Ph (4); (vi) NBS H O—DMF; (vii) KH. study was restricted to the synthesis of pyran systems only. A potential limitation of this variant lies in the initial Takai ole.nation step in that its e.ciency is dependent in an empirical sense upon the precise reaction conditions employed.Hirama’s study clearly indicates that polycyclic ethers can be prepared in a convergent manner where the .nal carbon—carbon forming step is again an RCM reaction whose e.ciency is relatively insensitive to the ring size being generated (seven- to ten-membered rings in yields ranging from 68—94%) Scheme 3. Clark has also demonstrated that function- 237 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 O OBn O O H OBn O O H H (i) 90% O O H H O H H O (ii) (iv) H H H H O O O O O O H H O H H O (iii) 94% (iii) 81% H H H O O O O H H H O H H H O Scheme 3 Reagents (i) (a) CH O H H H O —— CHMgBr THF,78 °C 80%; (b) Et SiH BF ·OEt CH CN 71%; (c) —— CHCH MgBr THF Et O; (b) Et SiH BF ·OEt ; (ii) (a) Li C H THF 91%; (b) (COCl) Et N DMSO CH Cl 78 °C; (c) Ph PMe Br NaHMDS THF 83%; (iii) (PCy ) Cl Ru——CHPh (cat.) PhH 70 °C; (iv) (a) CH Li C H ; (d) (COCl) Et N DMSO CH Cl ; (e) Ph PMeBr NaN(SiMe ) .(i) PCy3 Cl Ru H H CH3 CH3 O PMB O Cl PCy3 Ph PMB steps O O O O 86% H O O H H H CH3 H H Scheme 4 Reagents (i) CH CH3 Cl (0.01 M) 20 °C 12 h. O alised seven- and eight-membered cyclic ethers can be prepared using RCM reactions on highly oxygenated templates Scheme 4.In a somewhat di.erent approach to the synthesis of linearly-fused polytetrahydropyrans Sasaki and Tachibana have shown for example that the carbohydrate-derived vinyl tri.ate 5 undergoes Suzuki coupling with the alkyl borane 6 to a.ord enol ether 7. Oxidation of 7 (hydroboration —Swern sequence) followed by intramolecular hemiacetal formation and then (stereoselective) ionic reduction (Et SiH BF ·OEt ) a.orded the pentacyclic system 8 as a single diastereoisomer in six steps from 5 and 6 Scheme 5. Endocyclisation of enol 238 Annu. Rep. Prog. Chem. Sect. B 1999 95 235&mdash H OTBS O H H O (i) O + 66% B OTf O O H H H H O 6 5 TBS H H H O O O (ii) (iii) O 82% O O O H H H H 7 O O H H OTBS H H H H H H O O O O (iv) (v) OAc O (vi) OAc O O O O 75% O H H H H H H H H 8 Scheme 5 Reagents (i) Cs CO PdCl (dppf) KBr Ph As DMF rt; (ii) (a) ThexylBH THF 0 °C; (b) H O NaOH; (iii) (COCl) DMSO Et N; (iv) CSA CH Cl MeOH; (v) Ac O py rt; (vi) Et SiH BF ·OEt CH Cl ,10 °C. ethers, cobalt-complexed propargylic alcohols and -alkoxy epoxides also gives relatively easy access to a variety of medium-ring ethers Scheme 6 whereas stereoselective 7-exo-trig radical cyclisations have been put to good e.ect in the synthesis of the oxepane ring system present in ciguatoxin, Scheme 7.A crucial step in Yamamoto’s recent synthesis of hemibrevetoxin B requires the installation of the fused oxepane C and D rings which relies upon two sequential stereospeci.c intramolecular alkylation reactions.Remarkably in both cases cyclisation occurs cleanly upon exposure of the respective stannyl-aldehydes to BF ·OEt at 78 °C .nally a.ording the stereochemically homogenous tetracyclic intermediate 9 (94 and 98% isolated yield respectively for cyclisation steps). Mori has also developed an iterative strategy for the introduction of the C and D rings of hemibrevetoxin B. In this rather unusual example oxiranyl anions of the type 10 serve as synthetic equivalents to hypothetical dipole 11 Scheme 8. The coup de gra� ce in the area this year must be Nicolaou’s synthesis of brevetoxin A a molecule possessing 10 rings (from .ve- to nine-membered) and 22 stereogenic centres on a skeleton comprising 44 carbon atoms.This convergent synthesis uses readily available chiral pool materials—.-glucose and .-mannose—as sca.olds for the elaboration of the ABCDE and GHIJ rings respectively. In the crucial union of these two fragments reaction of the anion derived from the phosphine oxide 12 with the aldehyde 13 a.orded a 1 1 mixture of diastereoisomeric adducts which on exposure to KH in DMF produced the Z-ole.n 14 in 72% yield. Ring closure of the alcohol 14 to 15 (80% yield) was accomplished by treatment with Ag followed by ionic reduction Scheme 9. 239 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 OH Br PF6 N 2 74% (i) (i) Br PF6 N OMe 2 31% Co2(CO)6 AcO O 3 OH AcO O 2 O OH H AcO H H OTBS O O HO O OBn H O Scheme 6 Reagents (i)CH O CH 15 h. Cl ; (ii) (a) 4-O NC H CH OH; (b) CH ——CHCH SiMe BF ·OEt ; (c) CF CO Ag H O—MeNO ; (d) Ph PCH OMe; (iii) TFOH Cl 20 °C; (iv) H 5% Rh/C EtOH 60 °C; (v) Eu(fod) toluene 80 °C OH H 3 Bryostatins The bryostatins present another signi.cant challenge to the synthetic chemist both in terms of advancing the art of synthesis itself and producing compounds for biological testing.Bryostatin 1 has entered Phase II clinical trials for the treatment of melanoma non-Hodgkins lymphoma and renal cancer.However it is likely that the development of therapeutic agents from these compounds will only be possible if much simpli.ed and more accessible semi-synthetic or wholly synthetic analogues can be identi.ed. Encouragingly not only was there a total synthesis (Evans et al.) of bryostatin 2 (16) 240 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 OMe OMe O O (ii) 74% Br OH Br OMe O 2 O OH (iv) (iii) H 66% 52% AcO OAc Co(CO)6 OTBDPS (v) H H OTBS O 38% OTBDPS O O H O OBn H O O steps OHC O A B C HO Hemibrevetoxin B Scheme 7 Reagents (i) i-Bu AlSePh toluene,20 °C; (ii) n-Bu SnH (2 equiv.) Et B (catalyst) benzene rt; (iii) BF ·OEt CH Cl 78 °C. H H O O H O OTBPS H H O O H SePh O O H OMOM CO2Me Me OH H H H O D O H H O Me H TIPSO TIPSO TIPSO TIPSO Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 H (i) 92% O H (ii) 87% O H O H C O H TIPSO 94% H O H C O H TIPSO H H O H C O H TIPSO 98% H O H C O H TIPSO 9 O O H SePh TBPSO OH H H O H O O H H H CO2Me OMOM SnBu3 H O CHO 3 (iii) H H O C OH 3 steps H CHO O C O SnBu3 3 (iii) CH3 H H OH O D C O H 3 241 Me HO H 40 H Me Me O H H O H H O H H H 45 O O I O 30 5 1B0 H 15 H 20 E OH H O H 35 H A O D J H O 2 O H H G O H C O 55 HO OHH 50 F O H H 25 H M HO K H Me L Me OH ciguatoxin (CTX1B) OBn CH3 OTES OTBDPS (i) O + OTf BnO 90% Li SO2Tol O H H OBn OBn H CH3 (ii) O OTES CH3 OTBDPS 90% BnO BnO O O OTBDPS O O H CH3 CH3 O O SO2Tol 10 11 Scheme 8 Reagents (i) THF—HMPA,100 °C; (ii) p-TsOH CHCl 0 °C. Me H Me O B O H A C O Me O H H D H H H O O H OH E F H Me H H O O H H I J G O H O O O H H H brevetoxin A this year but also a report from the Wender group concerning the synthesis of simpli.ed bryostatin analogues which possess signi.cant antitumour activity in vitro.Evans’s synthesis of bryostatin 2 utilises sequential aldol-reduction sequences to set the stereochemistry and oxidation level in the three subunits which comprise the A B C rings. Connection of these fragments was accomplished using Julia—Lythgoe 242 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 (a) Lactonization Me Conjugate addition H Me O Lactonization B H O A C O Me O H H D H Horner-Wittig coupling H H O O H Epoxide opening OH bis- Lactonization E F H Me H H O O H H I J G O H O O O Dithioketal cyclization H H H Epoxide opening OH O Dithioketal cyclization HO OH HO OH D-glucose OH O OH HO Me Me PPh2 OH HO H M e H O H H O B C D O OM e O D-mannose TBDPSO E O O H H H H H TrO + BCDE ring system OHC OTBS 12 H H Me H O O H I J G EtS EtS OTBDPS O O H H H GHIJ ring system (i) 13 72% Scheme 9(a) For reagents see part (b).14 reaction (C16—C17; E:Z95 5) sulfone alkylation (C9—C10) whilst macrolactonisation was achieved in excellent yield (81%) using Yamaguchi’s procedure Scheme 10. Of crucial importance in the synthesis of the bryostatins is the stereocontrolled installation of the exocyclic enoate moieties at C13 and C21. In this approach this problem was tackled relatively late on in the synthesis the enoate moiety at C13 was accomplished in high yield (93%) but with moderate stereoselectivity (Z:E86 14) using a Wittig reaction between ketone 17 and Fuji’s chiral phosphonate 18.The 243 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 Me (b) 12 + 13 H Me TBDPSO B O H (i) 72% C TrO Me O H H D H H O H O H OTBS E EtS H Me H H O O H I J G H EtS OH OTBDPS O O Me H H H 14 H Me TBDPSO B O H (ii) 70% C HO Me O H H D H H H O O H OTBS F H Me H H O O H H I J G O H OTBDPS O O H H H 15 Scheme 9(b) Reagents (i) (a) n-BuLi THF,78 °C; (b) KH DMF; (ii) (a) AgClO ; (b) m-CPBA; (c) Et SiH BF E ·OEt . Me Me HO OR MeO2C O O O OH O OH O H Me Me M e O H O O Pr CO2Me bryostatin 1 R = Ac bryostatin 2 R = H ienoate at C21 was introduced in a two-step aldol dehydration-sequence a.ording the desired product 19 in 54% isolated yield (E:Z12 1).The carbonyl at C20 not only served as a control element for the stereocontrolled introduction of the enoate at C21 but also underwent reagent controlled reduction to a.ord 21 with the correct S 244 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 (a) 7 13 MeO2C O M e M e HO OR7 B A 9 O O 1 O 16 OH O OH O H Me Me C M e M e MeO OPMB B A 9 O 20 1 Me Me O 20 M e O 26 H O Pr CO2Me 16 TBSO 13 16 10 PhO2S (–) B O R3SiO 13 16 H (+) O LG (+) (–) O Me Me C SO2Ph 17 20 Bryostatin retrosynthetic analysis Scheme 10(a) absolute stereochemistry (dr10 1). Completion of the synthesis from this point employed standard deprotection and functional group manipulations.As a corollary to these studies the Wender synthesis of the bryostatin mimic 25 deserves mention. In this investigation extensive molecular modelling studies indicated that the A and B rings of the bryostatin framework could be dispensed with and be replaced with a much simer ‘‘spacer unit’’. Hence the fragment 22 was coupled to the alcohol 23 using the Yamaguchi protocol deprotection of the ketal 24 followed by a rare example of a ‘‘macrotransacetalization’’ followed by deprotection of the C26 hydroxy a.orded the bryostatin analogue 25 in moderate to good yield (56—88% two steps) Scheme 11. Cyclisation in this manner—presumably under conditions of thermodynamic control—a.orded 25 as a single diastereoisomer at C15 in which the side chain adopts an equatorial disposition with respect to the dioxolane ring.4 Acetogenins At this juncture mention should also be made of Marshall’s strategy for the synthesis of bis-tetrahydrofuran acetogenins. In the crucial step of this approach treatment of the aldehyde 26 with the optically pure stannane 27 in the presence of InCl a.orded Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 7 O TBSO O O C Me 26 OPMB M e M e 9 A O OPMB 1 CONHR TBSO OTBS Me 26 OPMB 245 ·SMe . Scheme 10(b) Reagents (i) NaHMDS THF 78 °C then 18; (ii) KHMDS THF 78 °C; then OHCCO Me,78 °C Et NSO NCO Me; (iii) 20 BH the anti-allylic alcohol 28 (86% yield).Conversion of 28 into the aldehyde 29 followed by treatment with the allyl stannane 30 this time in the presence of BF ·OEt a.orded the syn adduct 31 in 92% isolated yield. Double cyclisation of 31 followed by chain extension to 32 using standard methodology completed the synthesis Scheme 12. 5 Altohyrtin (spongistatin 1) A number of complex natural products containing the spiroketal motif have been synthesised this year including calyculin C, oligomycin C, and spongistatin 1 (altohyrtin A). The synthesis of altohyrtin A (spongistatin 1) a potent antitumour agent isolated 246 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 Cl rt; (b) Pd(OH) /C H . Scheme 11 Reagents (i) 2,4,6-(O N) C H COCl Et N PhCH rt; (ii) (a) Amberlyst- 15 CH from marine sponges presents a daunting task.The molecule is made up from a 51-carbon chain which incorporates six pyran rings—four of which are embedded in two isolated spiroketal units a 42-membered lactone ring and a novel chlorodiene moiety. Until this total synthesis was accomplished there was some debate as to a number of stereochemical issues which highlights the (bene.cial) symbiotic relationship between synthetic and natural products chemists alluded to in the introductory section. Approaches to this class of compounds have aroused considerable interest amongst the synthetic community as illustrated by recent publications from the groups of Nakata, Crimmins and Paterson. Nakata’s and Crimmins’ work (Schemes 13 and 14 respectively) is primarily associated with the construction of the spiroketal fragments (C1—C14) from either tri-O-acetyl-.-glucal or readily available pyrone 247 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 Scheme 12 Reagents (i) InCl ; (ii) BF ·OEt . intermediates respectively. Paterson describes an aldol strategy for the union of the A/B and C/D spiroketal moieties Scheme 15 which enables the direct introduction of the stereocentres at C15 and C16 in high yield but with modest stereochemical control (84% yield; 67% ds). Kishi’s synthesis of spongistatin 1 is a masterly execution of synthetic planning and utilises a number of reactions which are of growing importance in contemporary organic synthesis Brown allylations for the synthesis of highly versatile allylic alcohols 248 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 OAc O OTr OBn 10 steps AcO OAc (iii) Ph O S OTBS O O Me 85% S steps OTBS OPiv OBn O Me OPiv OTBS Me O O HO 1 OMP Scheme 13 Reagents (i) 1,3-Dithiane n-BuLi HMPA 20 °C; (ii) IBX DMSO—THF rt; (iii) (a) n-BuLi HMPA THF; (b) epoxide; (iv) (a) H Pd/C EtOAc rt; (b) CSA CH 1 Me Me O 15 16 14 OR3 OMe 19 C O HO O 33 41 M e M e O F E O OH H OH HO R1 R3 R2 Ac Ac Cl Cl —MeOH (1 1) rt. Me HO B O A D 7 O 23 O R2O 5 O HO OH 50 47 R1 spongistatin 1 Annu. Rep.Prog. Chem. Sect. B 1999 95 235—263 (i) (ii) OTr OBn O S O 96% S OBn OMP Ph OTBSS S O O OBn OH OMP Me (iv) OMP OH OBn 71% Me O 14 OTBS Me O O TESO 1 OMP 249 O O CH3 (ii) (i) 55% 73% OTHP OH O O + OTHP OHC O O H (iii) O O O O 82% OTIPS OTIPS Scheme 14 Reagents (i) LiHMDS THF 78 °C; (ii) (a) TIPSOTf CH Cl ; (b) MeOH PPTS; (c) CF CO H PhH; (iii) CH ——CHMgBr CuBr·Me S. Scheme 15 Reagents (i) LiTMP LiBr THF 78 °C 30 min; (ii) (a) 2 min; (b) AcOH. 250 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 Scheme 16(a) in high ee’s the use of indium reagents and the Nozaki—Kishi reaction for the formation of carbon—carbon bonds in highly complex substrates. His retrosynthesis Scheme 16 splits the target into two fragments C1—C28 33 and C29—C51 34.It was envisaged that coupling of C28 and C29 could be achieved by way of a Z-selective Wittig reaction whilst macrolactonisation installing the O41—C1 bond would be tackled towards the end of the synthesis anticipating that competing cyclization routes e.g. via an unprotected—OH at C-38 would not be observed. Fragment 33 was further disconnected into the vinyl iodide 35 and the aldehyde 36. The 251 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 Scheme 16(b) For reagents see part (c). Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 252 42 + 41 (c) (i) O 28 H O D TBSO MeO 1 MeOCH2CO2 H H O A O AcO B Me OH 43 O H O HO HO (iii) H O H Me H O O AcO Me OH 45 Spongistatin assembly of ABCD fragment Scheme 16(c) Reagents (i) (a) NiCl odinane 83%; (ii) (a) PPTS acetone—H HF—py THF 82%.Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 OMPM OMe TBSO HO 19 (ii) H O Me H O O 15 Me AcO OAc Me OH OMPM OMe HO HO H O H O O A Me B AcO OAc Me OH 1:1 —CrCl THF 86%; (b) Dess-Martin peri- O; (b) Triton-B MeOH—MeOAc 50%; (iii) OMPM 27 O H O OMe 21 19 O O H Me Me OAc 44 OMPM O H O D OMe C O O H Me Me OAc 46 253 OHC OTIPS 47 OTIPS OH 48 steps OTIPS OHC 50 Brown allylation 74% O (iii) O 100% 53 Scheme 17 Reagents (i) ()-Ipc -(Z)-crotyl boronate THF 78 °C; (ii) (a) NMO OsO ; (b) NaIO ; (c) diketene—acetone adduct; (d) ; (iii) DMDO CH Cl .254 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 OH (i) OTIPS 65% OTIPS Me (ii) OTIPS 54% O H OTIPS E O H OTIPS OH 51 OTIPS O O CH3 CH3 O O 54 I Construction of ring E OTIPS Cu(Me)(CN)Li2 Spongistatin construction of ring F steps OTIPS steps OTIPS 49 O OTIPS OTIPS 52 OTIPS H O F OTIPS OH 55 (a) Me Me O O (i) 85% 37 I TIPSO 51 Cl steps 51 Cl Scheme 18(a) Reagents (i) In CH ——C(Cl)CH Cl NaI DMF; (ii) (a) Martin’s sulfurane CH Cl ; (b) HF—py; (c) DMSO CH Cl Et N; (iii) LDA THF HMPA Cl —H O; (b) KF MeOH; (c) Et N 2,4,6-Cl C H COCl 5 °C; (iv) (a) DDQ CH PhMe 50%.OTIPS O OTIPS H OH 55 OTIPS Cl OH 57 Introduction of chlorodiene unit (C48-C51) OTIPS Me 29 O H 59 CHO Me H O H OMPM OMPM 58 78% MeO HO TBSO H 62 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 steps TIPSO OHC OTIPS (ii) 51 89% Cl OTIPS O BrMg OTIPS H HO 38 H O TIPSO H Cl OMPM 61 OTBS Me 29 O P+Ph3I– H H Me O OMPM OMPM OTIPS O OMPM H OMPM 56 CHO O F OMPM OMPM 58 OTIPS 60 OTIPS Me O OTIPS H Me OMPM (iii) 63 40% 255 (b) Cl 50 Cl 256 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 62 (iii) 40% OTBS Me MeO HO O H H Me O TBSO H 41 OMPM TBSO 1 OMPM CO2TBDPS H H O O AcO 10 OH Me 63 (iv) OR Me HO HO O 30 H H 40 Me O RO H RO O OH O H H O O AcO 10 OH Me 63 Spongistatin final coupling sequence Scheme 18(b) H O OMe O O H Me Me OAc H O OMe 20 O O H Me Me OAc OH (a) Cl O O 6 2 4 OH HO O O HN X O HN NHMe HN H H O NH NH O O CH2CH(CH3)2 O NH 5 HO2C NH2 7 OH OH HO Oallyl OH O F 6 4 2 HO HO NO O X = Cl vancomycin aglycon X = H eremomycin aglycon 2 HN Cl O HN HN H H O NH O O NMeBoc O NH O O 5 CH2CH(CH3)2 NHDdm MeHN 7 OBn OBn BnO NO Oallyl NH 2 F O 6 6 4 HO X OH HN HN Cl O HN H H NH OH 2 H H O O O NH NH 5 5 HO2C HO2C 7 7 OH OH OH OH HO HO Vancomycin/eremomycin retrosynthetic analysis Scheme 19(a) C1—C12 fragment was prepared from the epoxide 37 using well developed epoxideopening chemistry whilst C13—C17 was prepared from the known alkene 38.These two fragments were joined together by ring opening of the epoxide 39 with the cuprate derived from the iodide 40. Standard functional group manipulations then a.orded the pivotal intermediate 41 which was coupled with the vinyl iodide 42.A key sequence in this synthesis was formation of the C17—C18 bond using the Nozaki—Kishi Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 Oallyl OMs HO 4 O HN NHBoc O 257 (b) X X (X = H Cl) F F 6 6 OH OH O O O2N 2N H NHR O N O i NHR O NH O 65% NH OBn OH NH O MeHN 5 MeHN 5 7 7 OMe MeO OMe OMe OMe MeO 65 66 Biaryl synthesis oxidative coupling sequence BF ·OEt AgBF TFA—CH Cl 0 °C; (b) Scheme 19(b) Reagents (i) (a) VOF NaHB(OAc) . reaction which in this instance a.orded a mixture of diastereoisomeric alcohols (at C17) in 86% yield. Oxidation of this mixture of alcohols to the ketone 43 deprotection of the C23-OH group followed by base-catalysed conjugate addition a.orded the spiroketal fragment 44.It should be pointed out that when cyclization was attempted with a protected hydroxy at C26 then the wrong epimer at C23 was obtained as the sole product (44). However in the case where cyclization was conducted with a free hydroxy group at C26 then a separable 1 1 mixture of spirocycles 45 and 46 was obtained. With theABCD fragment 46 in hand preparation of C29—C51 coupling and cyclisation was attempted. It is tempting as others have done in connection with developing approaches to the spongistatins to prepare the tetrahydropyranyl residues such as the E and F rings from carbohydrate precursors. Oftentimes such a strategy may be cumbersome and a more convergent route from non-carbohydrate precursors should be considered.The key intermediate 48 required for the ring E fragment 34 was in this case available in 64% yield (90% ee) from the aldehyde 47 via two consecutive Brown allylation reactions (Scheme 17). Transformation of 48 to the iodo glucal 49 was accomplished using established chemistry. The precursor 51 to the ring F fragment 55 was also prepared using Brown’s allylation chemistry starting from the aldehyde 50 intermediate 51 was prepared in 74% yield (90%ee) from 50 which then transformed into the glucal 52. Stereospeci.c oxidation of 52 and ring opening with the cuprate 54 a.orded 55 in 70% yield. At this juncture a route to the introduction of the novel chlorodiene fragment (C48—C51) was sought.Conversion of the acetonide 55 to the aldehyde 56 then enabled coupling with the organoindium reagent derived from 2,3-dichloropropene a.ording epimeric alcohols 57 which were dehydrated (Martin’s sulfurane; 85% yield for the two step alkylation dehydration sequence) and converted to the aldehyde 58 Scheme 18. Coupling of 58 with a nucleophilic organometallic reagent derived from the iodide 59 was then investigated an obvious contender (especially from this group) being the the Nozaki—Kishi procedure. Unfortunately although coupling of 58 with 59 using Cr(..)/Ni(..) worked well the undesired epimer was produced as the major product. However treatment of the organomagnesium reagent 258 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 CO DMSO rt; (ii) CsF DMSO rt.Scheme 20 Reagents (i) Na Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 259 Diels-Alder + N N + 70 O N Diels-Alder N NH OMe 72 O (ii) HO 100% CO2Me OAc (iii) 22% HO OCOC6H4NO2-p 75 Scheme 21 Reagents (i) MeOH pH 7.3; (ii) MeOH rt 6 days; (iii) (a) Hg(OTf) CH CN rt; (b) NaCl. 260 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 N (i) 0.3% N 71 HN H OMe N N O steps O O NH HNN O 73 MeO2C H H CO2Me O H HO2C H 74 OAc H O HgCl 76 , 60 with the aldehyde 58 under conditions of chelation control a.orded the desired product 61 in 78% yield. Completion of the synthesis involved Wittig coupling of the aldehyde 33 with the phosphorane derived from 62 which a.orded the Z-alkene 63 (40%) which on deprotection followed by macrolactonization using the now standard Yamaguchi’s procedure selectively a.orded the 41-membered ring lactone 64 Scheme 18.Completion of the total synthesis validated the structure put forward by Kitagawa in preference to that pro.ered by Pettit and Fustani. 6 Vancomycin Arguably the notion that natural products such as the erythromycins are derived biogenetically by way of aldol-type chemistry initiated a mammoth e.ort to replicate these reactions in the laboratory. Progress in recent years indeed has been impressive as illustrated by Evans’ synthesis of 6-deoxyerythronolide B and oleanolide which utilises essentially four bond constructs stereoselective aldol reactions diastereoselective epoxidations and reductions.A somewhat di.erent situation regarding stereochemical control is apparent when planning synthetic routes to targets such as vancomycin and erenomycin members of a large family of antibiotics which have an arylglycine heptapeptide core. These antibiotics are of considerable medicinal value due to their e.cacy against Staphylococcus aureus infections. The added complication here is that in the case of vancomycin for example there are three additional elements of atropoisomerism to be considered due to hindered rotation about the two aryl ether moieties (rings 2—4 and 6—4) and the biphenyl residue (rings 5—7). Hence even if the peptide chain was constructed in an enantiomerically pure state the problem which then arises is how to prepare one of the possible eight atropodiastereoisomers.This problem has been addressed by two groups this year, culminating in the total synthesis of the vancomycin and erenomycin aglycons. The strategic bond constructs in these syntheses rely upon relatively standard oxidative and Suzuki biaryl coupling reactions for the union of rings 5—7 and S Ar chemistry to e.ect formation of the aryl ether links between rings 2—4 and 4—6 Scheme 19. In the case of Evans’ synthesis for example whilst the oxidative coupling (VOF ) of 65 to 66 a.orded the unnatural R atropodiastereoisomer (dr95 5) subsequent formation of the ether linkage between rings 4 and 6 (intramolecular S Ar) and atropoisomerization (MeOH; 55 °C; 24 h) a.orded 69 with the desired S biaryl con- .guration.In this particular synthesis a second diastereoselective synthesis (dr5 1; major isomer having natural R con.guration) completed the synthesis of the core structure. Cyclisation at this stage appeared to proceed with no epimerization at the other centres and the major diastereoisomer possessing the correct natural product stereochemistry could be isolated by column chromatography in 80% yield Scheme 20. It is apparent that the stereochemical course of the biaryl and aryl ether coupling reactions is dependent on a subtle interplay between a number of factors including the thermodynamic bias dictated by the global structure as well as those emanating from proximal stereochemical interactions. 261 Annu.Rep. Prog. Chem. Sect. B 1999 95 235—263 7 Potpourri A number of reports concerning ‘‘biomimetic’’ approaches to natural product synthesis have been disclosed this year (Scheme 21). A biomimetic synthesis of keramaphidin B (71) utilising an intramolecular Diels—Alder reaction of 70 has been reported by Baldwin. Although the yield of the key uncatalysed reaction is low (ca. 0.3%) it clearly validates Baldwin’s earlier suggestion concerning the biosynthesis of the manzamines. Along similar lines Williams has developed a synthesis of brevianamide 73 from the indole derivative 72 whilst Whitehead has reported a ‘‘predisposed’’ approach to the manzamenone 74. A biomimetic approach to the taxane skeleton was less rewarding as treatment of 75 with Hg lead to extensive rearrangements resulting in the isolation of 76. Total syntheses of taxol and taxusin have however been reported using more standard methodology.Taxol in particular continues to generate much synthetic interest as exempli.ed by the number of model studies of the parent ring system which appear at regular intervals in the literature. References 1 F.W o� hler Pogg. Ann. 1828 12 253. 2 J.D. Dunitz K. D. M. Harris R. L. Johnston B. M. Kariuki E. J. MacLean K. Psallidas W. B. Schweizer and R. R. Tykwinski J. Am. Chem. Soc. 1998 120 13 274. 3 D.H.R. Barton and R. C. Cookson J. Chem. Soc. Quart. Rev. 1956 10 44. 4 C.W. Rees Chem. Br. 1998 34(6) 75. 5 F.-A.Kang and C.-L. Yin J. Am. Chem. Soc. 1997 119 8562. 6 B. Cornett M. Davis S. Wu N. Nevins and J. P. Snyder J. Am. Chem. Soc. 1998 120 12 145. 7 e.g. K.C. Nicolaou Y. He F. Roschangar N. P. King D. Vourloumis and T. Li Angew. Chem. Int. Ed. Engl. 1998 37 84; M. Reggelin V. Brenig and R. Welcker Tetrahedron Lett. 1998 39 4801. 8 Total synthesis of epothilone E K. C. Nicolaou Y. He F. Roschangar N. P. King D. Vourloumis and T. Li Angew. Chem. Int. Ed. Engl. 1998 37 84; epothilone B J. Mulzer A. Mantoulidis and E. OQ hler Tetrahedron Lett. 1998 39 8633. Synthesis of analogues K. C. Nicolaou F. Sarabia S. Ninkovic M.R. V. Finlay and C. N. C. Boddy Angew. Chem. Int. Ed. Engl. 1998 37 81. For a review see K. C. Nicolaou F. Roschangar and D. Vourloumis Angew. Chem.Int. Ed. Engl. 1998 37 2015. 9 For reviews see S. K. Armstrong J. Chem. Soc. Perkin Trans 1 1998 371; R. H. Grubbs and S. Chang Tetrahedron 1998 54 4413. 10 D.W. Old J. P. Wolfe and S. L. Buchwald J. Am. Chem. Soc. 1998 120 9722; A. F. Littke and G. C. Fu Angew. Chem. Int. Ed. Engl. 1998 37 3387. For a discussion of the mechanism of the Stille reaction see A. L. Cassado and P. Espinet J. Am. Chem. Soc. 1998 120 8978. 11 e.g. K.C. Nicolaou J. Pastor N. Wissinger and F. Murphy J. Am. Chem. Soc. 1998 120 5132. 12 N. Edwards J. A. Macrichie and P. J. Parsons Tetrahedron Lett. 1998 39 3605. 13 Y. Takemoto and T. Ibuka Tetrahedron Lett. 1998 39 7545. 14 J. D. Rainier and S. P. Allwein J. Org. Chem. 1998 63 5310. 15 T. Oishi Y. Nagumo and M. Hirama Chem. Commun.1998 1041. 16 J. S. Clark O. Hamelin and R. Hufton Tetrahedron Lett. 1998 39 8321. 17 M. Sasaki H. Fuwa M. Inoue and K. Tachibana Tetrahedron Lett. 1998 39 9027. 18 F. Simart Y. Brunel S. Robin and G. Rousseau Tetrahedron 1998 54 13 557. 19 C. Yenjai and M. Isobe Tetrahedron 1998 54 2509. 20 (a) K. Fujiwara H. Mishima A. Amano T. Tokiwano and A. Murai Tetrahedron Lett. 1998 39 393; (b) T. Oka K. Fujiwara and A. Murai Tetrahedron 1998 54 21. 21 M. Sasaki M. Inoue T. Noguchi A. Takeichi and K. Tachibana Tetrahedron Lett. 1998 39 2783. 22 I. Kadota and Y. Yamamoto J. Org. Chem. 1998 63 6597. 23 Y. Mori K. Yaegashi and H. Furukawa J. Org. Chem. 1998 63 6209. 24 K. C. Nicolaou Z. Yang G.-Q. Shi J. L. Gunzner K. A. Agrios and P. Ga� rtner Nature 1998 392 264.262 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263 25 D. A. Evans P. H. Carter E. M. Carreira J. A. Prunet A. B. Charette and M. Lautens Angew. Chem. Int. Ed. Engl. 1998 37 2354. 26 P. A. Wender J. De Brabander P. G. Harran J.-M. Park M. F. T. Koehler B. Lippa C.-M. Park and M. Shiozaki J. Am. Chem. Soc. 1998 120 4534. 27 J. A. Marshall and K. W. Hinkle Tetrahedron Lett. 1998 39 1303. 28 J. S. Panek and N. F. Jain J. Org. Chem. 1998 63 4572. 29 A. K. Ogawa and R. W. Armstrong J. Am. Chem. Soc. 1998 120 12 435. 30 J. S. Guo K. J. Du.y K. L. Stevens P. I. Dalko R. M. Roth M. M. Hayward and Y. Kishi Angew. Chem. Int. Ed. Engl. 1998 37 187. 31 M. Hayward R. M. Roth K. J. Du.y P. I. Dalko K. L. Stevens J. S. Guo and Y. Kishi Angew. Chem. Int. Ed. Engl. 1998 37 192. 32 T. Terauchi and M. Nakata Tetrahedron Lett. 1998 39 3795. 33 M.T. Crimmins and D. G. Washburn Tetrahedron Lett. 1998 39 7487; M.T. Crimmins D. G. Washburn J. D. Katz and F. J. Zawacki Tetrahedron Lett. 1998 39 3439. 34 I. Paterson D. J. Wallace and R. M. Oballa Tetrahedron Lett. 1998 39 8545. 35 J. W. Cornforth in Perspectives in Organic Chemistry A. Todd (ed.) Interscience New York 1956. 36 D. A. Evans A. S. Kim R. Metternich and V. J. Novack J. Am. Chem. Soc. 1998 120 5921. 37 D. A. Evans M. R. Wood B. W. Trotter T. I. Richardson J. C. Barrow and J. L. Katz Angew. Chem. Int. Ed. Engl. 1998 37 2700. 38 K. C. Nicolaou M. Takayanagi N. F. Jain S. Natarajan A. E. Kuombis T. Bando and J. M. Ramanjulu Angew. Chem. Int. Ed. Engl. 1998 37 2717. 39 D. A. Evans C. J. Dinsmore P. S. Watson M. R. Wood T. I. Richardson B. W. Trotter and J. L. Katz Angew. Chem. Int. Ed. Engl. 1998 37 2704. 40 J. E. Baldwin T. D.W. Claridge A. J. Culshaw F. A. Heupal V. Lee D. R. Spring R. C. Whitehead R. J. 41 J. E. Baldwin and R. C. Whitehead Tetrahedron Lett. 1992 33 2059. 42 R.M. Williams J. F. Sanz-Cervera F. Sancenon J. A. Marco and K. Halligan J. Am. Chem. Soc. 1998 120 Bought.ower I. Mutton and R. J. Upton Angew. Chem. Int. Ed. Engl. 1998 37 2661. 1090. 1998 39 389. 43 S. Al-Busa. M.G. B. Drew T. Sanders and R. Whitehead Tetrahedron Lett. 1998 39 1647. 44 M. Nishizawa H. Imagawa I. Hyodo M. Takeji E. Morikuni K. Asah and H. Yamada Tetrahedron Lett. 45 K. Morihira R. Hara S. Kawahara T. Nishimori N. Nakamura H. Kusama and I. Kuwajima J. Am. Chem. Soc. 1998 120 12 980. 46 L. A. Paquette H.-L. Wang Z. Su and M. Zhao J. Am. Chem. Soc. 1998 120 5213. 47 S. Arseniyadis J. I. M. Hernando J. Quilez del Moral D. V. Yashunsky and P. Poitier Tetrahedron Lett. 1998 39 3489; G. Stork K. Manabe and L. Liu J. Am. Chem. Soc. 1998 120 1337. 263 Annu. Rep. Prog. Chem. Sect. B 1999 95 235—263

ISSN:0069-3030    

DOI:10.1039/a808600a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

9 Reaction mechanisms Part (i) Polar reactions James C. Fishbein† Department of Chemistry,Wake Forest University,Winston-Salem NC 27109 USA 1 Introduction and caveat An attempt has been made to feature the diversity of chemistry that has been reported with some sacri.ce as to detail. In general the original authors’ conclusions are presented in some cases employing the original sometimes ambiguous phraseology. No endorsement is implied. 2 Hydrogen transfer reactions Proton transfer The formation of the dications of nitro-polycyclic aromatic hydrocarbons has been observed in magic acid by NMR spectroscopy. Mono-N-hydroxy dications (from dehydration) and di-N-hydroxy dications are formed and their distribution is substituent dependent. The pK’s of some dithiole-3-thione carboxylic acids relatives of the chemoprotective phase II enzyme inducer Oltipraz and their oxo analogues have been reported to range between pK 1.14 and 3.4. The .rst determination of a thioenol pK for 1,8.49 is reported and is 2.88 units lower than the analogous oxygen acid. The thioenol decomposes with reversible protonation followed by hydrolysis to the (oxygen) ketone competitive with cyclization to a benzothiophene.Reaction inDMSO of butylamine with N-(2,6-dinitro-4-substituted)phenyl-n-butylamine partitions between formation of the sigma adduct and proton abstraction to give the amide anion. Equilibrium and rate constants for the two reactions are determined and decrease in the order 4-XNO CNCO Me.Kinetics of the approximately thermoneutral proton transfer between pyridine and the chloride and bromide salts of protonated porphyrin have been measured in chloroform by stopped-.ow tech- †Present address Department of Chemistry and Biochemistry University of Maryland Baltimore County 1000 Hilltop Circle Baltimore MD USA 21250. 265 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 niques. The rate law includes terms that are .rst order and second order in H O having values of 310M s and 210M s respectively with values for the bromide salt exceeding those for the chloride salt by a factor of 3 in both cases. The tri.uoromethylsulfonyl group and a sulfur atom acidify adjacent C—H bonds mainly by e.ects other than resonance delocalization on the basis of e.ects of solvent on pK in the case of the tri.uoromethylsulfonyl group and in the case of the sulfur atom larger intrinsic rate constants for phenylthionitromethane relative to phenylnitromethane nitromethane and 2-phenylnitromethane.A methyl group acidi.es the adjacent C—H of a chromium carbene complex 2 presumably through hyperconjugative stabilization of the resulting alkene resonance form of the conjugate base and concomitantly raises the intrinsic barrier for proton transfer. The acid base properties and tautomerism of 3-phenylcoumarin-2-one 3 have been studied. These data contribute to a plot of log k for hydroxide ion catalyzed proton abstraction against carbon acid pK which spans 19 log units in pK has a slope of 0.43 and does not clearly exhibit curvature.Rate constants and primary isotope e.ects for proton transfer involving 1-methylindenes in dichlorobenzene and nitroalkanes in tetrahydrofuran, with an emphasis on the e.ects of steric hindrance in the latter case have been determined. The third order term for the enolization of aldehydes and ketones that is .rst order in both bu.er acid and base has been examined in detail. It is deduced that the third order term is most signi.cant where both the second order terms for the base and acid catalysis are most signi.cant — there is an apparently linear correlation between log k is largest and C——O (the third order rate constant) and log (k *k ) (the product of the second order terms).The third order term is also most signi.cant where the K group basicity is smallest. Hydride transfer Nicotinamide-type reductants with hindered rotation of the amide moiety due to covalent restraint reduce benzoquinone with preferential transfer of one of the two prochiral hydrogens. The degree of preference varies with quinone reactivity with the more reactive quinones being less selective.A signi.cant contribution to the overall product isotope e.ect on hydride transfer is believed to lie in the C—L bond weakening that occurs on formation of the initial electron transfer complex. 3 Mostly carbocationic substitutions Glycosylation of the sulfoxide 4 is inhibited by the sulfenate which is believed to form when the nucleophilic sulfoxide starting material reacts with the carbocation obtained by tri.uoroacylation of a second molecule of the starting material. 266 Annu.Rep. Prog. Chem. Sect. B 1999 95 265—282 The decomposition of benzo[a]pyrene diol epoxide in 1 9 dioxane—H O is stimulated by amine bu.ers. For bu.er acids of pK8 both the amine base and the ammonium ion stimulate the reaction. Plots of k against mole fraction base are curved consistent with a change in rate limiting step. At low [bu.er] the rate limiting step involves general acid catalyzed ring opening. At high bu.er concentration capture of the cation by amine base or by water in a reaction catalyzed by the amine is rate limiting. Capture by azide ion occurs with a rate constant ratio of k /k 260 s.This is consistent with di.usion limited trapping the assumption of which gives k 210 s for the cation. The oxocarbocation from the phosphate ester 5 is believed to be formed from the is speci.c acid catalyzed reaction. The solvent deuterium isotope e.ect on this reaction is k /k 0.45. At 0.9MN the increase in k is20%. This is insu.cient to account for the amount of azide substitution product observed. The increase in k claimed to be due to the trapping of a tight ion pair the remainder of azide product arising from trapping of the solvent-separated ion pair and the free ion whose lifetime is estimated to be310 s. Adamantylideneadamantyl halides 6 solvolyze with sensitivity to solvent ionizing power m0.74 0.90 0.88 for the chloride bromide and iodide respectively. Common ion inhibition of solvolysis is observed and the value of Swain—Scott s0.2 was obtained.Azide ion and iodide ion trapping relative to solvent yield estimates of 3—610 s for the lifetime of the homoallylic cation. Solvolysis of the c-propyl derivative 7 in 80% aqueous acetone has been studied. With electron withdrawing substituents X the main product is the unrearranged tertiary alcohol whereas with electron donating groups the major product is the benzylic alcohol 8. The plot of log k against is linear for meta substituents and the p-electron donating groups deviate positively from this correlation. This and the substituent e.ects on products give rise to the notion that with electron withdrawing groups the substitution occurs via the cyclopropylmethyl cation whereas with electron withdrawing groups a homoallylic cation intervenes.Solvolysis of diphenylchloromethane and bis(4-methoxyphenyl)chloromethane in .uorinated solvents is faster than predicted by reactions in solvents of similar Y value for non.uorinated solvents. This e.ect is limited to reactions that give extensively delocalized cations. A special delocalized-cation-stabilizing-e.ect of .uorinated solvents is proposed. Solvolysis of sterically congested 3 alkyl chlorides has been investigated. Reactivity orders are governed by competing e.ects of strain relief and 267 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 solvent participation.It remains unclear as to whether many of these reactions are stepwise or concerted. The solvolysis of bicyclic tri.ates has been studied with a view to understanding the nature of the e.ect of the -thiocarbonyl group. The ratio of reactivity between the -thiocarbonyl substituted compound and the unsubstituted compound increases from 10 to 10 with increasing ring size from 9 to 10. This is consistent with stabilization of the cationic transition state for solvolyis by conjugation that is more facile for the larger ring system. Ab initio computations concur. The -thioamide cation 11 is 10 -fold less reactive toward 50 50MeOH—H O solvent than the 1-(4-methoxyphenyl)ethyl cation indicating strong stabilizing interactions of the -thiocarbonyl group.— The cation is unreactive toward weaker carbon nucleophiles reacting only with the stronger nucleophiles 2- methoxythiophene and pyrrole.A stereochemical and isotope scrambling investigation of the LiClO stimulated rearrangement of 12 in diethyl ether has been carried out. Either enantiomer of the carbon attached to the vinyl ether gives a racemic mixture of the aldehyde products. -Deuterium labeling at the same carbon shows the acetaldehyde fragment becomes attached to equal extents to either of the equivalent carbons of the allyllic cation. Thus a free carbocation is proposed. Hydrolysis of the cyclic acylals 14 and 15 in H Substituted anilines react with endo and exo norborn-2-yl arenesulfonates 13 in methanol with 1.5 and 1.0. The reactions are kinetically .rst order in aniline with 0.02 and 0.15.The cross-correlation coe.- cient is deemed statistically signi.cantly di.erent from zero. It is concluded that the reaction occurs by a stepwise preassociation mechanism with a carbocation intermediate. O—20% EtOH is H dependent and pH independent. Isotopically labeled solvent H O ends up in the furanyl and pyranyl products and not in the carboxylic acids for both the H dependent and pH independent reactions. This result leaving group e.ects activation parameters and the 268 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 absence of bu.er catalysis are interpreted to indicate rate limiting formation of the oxocarbocation from the protonated acylal.Deamination of 4,4-dimethoxytritylammonium ion occurs by bothpH independent and hydrogen ion catalyzed reactions. It is concluded that rate increases observed by increasing concentrations of oxygen acids are not due to general acid catalysis. The solvent deuterium isotope e.ects are k /k 1.51 and 1.2 for the uncatalyzed and hydrogen ion catalyzed reactions respectively. The uncatalyzed reaction is believed to involve rate limiting di.usional separation of the cation—ammonia pair whereas the the hydrogen ion catalyzed path involves hydrogen ion assisted separation. Acid catalyzed substitution on chiral benzaldehyde methyl isopropyl acetals by methyllithium cuprates with BF -etherate in ether 78 °C has been investigated. The substitution occurs exclusively with displacement of MeO and the 34—40% product enantiomeric excess leads to the conclusion that the reaction occurs to the extent of 56—66% via a free carbenium ion intermediate.Solvolysis of adamant-2-yl azoxytosylate occurs with a dependence upon solvent ionizing power of m0.46 the smallest known value for a pure S 1 reaction that occurs via formation of an ion pair. Photolysis of the chloride 16 gave rise to the allylic cation in a two component process that indicated the intermediacy of a relatively long-lived species believed to be the chloronium ion 17. Rate constants for capture of the cations were determined. The selectivities of the 2,4,6-trimethylphenyl cation formed from the diazonium ion toward the oxygen and nitrogen of amides relative to water in concentrated amide solutions have been determined. The amide-O/H O (unitless) selectivities are slightly less than 1 whereas the amide-N/H O selectivities are slightly less than 0.1.The selectivities are interpreted as arising from the di.erences in the extent to which the O compared to N solvates the diazonium ion prior to its dissociation to the cation which is trapped faster than it can di.usionally separate from the solvent cage. Essentially this entails a preassociation mechanism in which preferential solvation determines the product distribution. The nitrenium ion 18 generated from the N-sulfonatooxy precursor partitions between capture by azide ion and water with a ratio k /k 280M. Solvent capture occurs at the two exocyclic carbon atoms.Azide ion reacts similarly with 13% of the products at high azide ion concentration also arising from attachment at the carbons ortho to the N-acetyl group. Exocyclic trapping generates selective quinone imide methides that could give rise to additional electrophilic damage in vivo. Compounds 19 and 20 undergo solvolysis in their pyridine free base form and yield nitrenium ion-derived substitution products including products of ion-pair return. They do so with rate constants that are 10—10 fold smaller than their carbocyclic analogues and the resultant nitrenium ions are 40—300 fold shorter lived on the basis of azide ion/H Opartitioning studies and the assumption that the azide ion reaction is d.usion limited.Products of reactions of the nitrenium ions with deoxyguanosine have been characterized. 4 Mainly aliphatic substitutions Some methods have been suggested to distinguish between concerted and stepwise 269 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 substitutions. A study of salt e.ects on substitutions in acetone—water mixtures led to the proposal that the e.ect of 1MLiClO on reaction rates in mixed aqueous solvents be employed to distinguish between S 1 and S 2 mechanisms. Classical S 1 substrates (t-BuBr and 2-Ad-OTs) show rate accelerations whereas classical S 2 substrates (MeOTs and EtBr) show inhibition; secondary and benzylic substrates show no e.ect.Employment of ‘anti-hydrophobic’ e.ects has been suggested to distinguish between SET (single electron transfer) and S 2 mechanisms. As conceived addition of organic cosolvent (EtOH) to water will result in a rate acceleration if the TS is more hydrophobic and charge di.use than the ground state. Thus relative to water 20% EtOH—80%H O increases the rate constants for thiophenolate (SET) substitution on a primary alkyl iodide by 80% whereas it only accelerates the (S 2) substitution with a primary mesylate by 6%. As noted the validity of this method is not established. Solvolytic substitution at benzyl sulfonium ions is suggested to be S 2 in most nucleophilic solvents on the basis of the importance of the solvent nucleophilicity sensitivity parameter l in multiparameter correlations which increases with increasing electron withdrawing ability of the benzyl substituent. Positive deviations for the p-Me substituent are accounted for by emergence of a new mechanism — ‘ionization with nucleophilic solvation’.Substitution on 21 in pyridine is proposed to occur by S 2 displacement by most substituted anilines for which 2.5 while positive deviations for nucleophiles with the strongest electron withdrawing groups are proposed to involve the aniline anions. A number of reactions involving carbanion nucleophiles were reported. The carbanion 22 displaces Br from chiral 2-bromobutane with 99.7% inversion and on the basis of a number of previous studies it is claimed that there is a correlation between the extent of substitution by SET and the rate constant for thermoneutral one electron transfer of the radical anion of the nucleophile to an alkyl halide. A mechanistic with alkyl halides span 10 but are insensitive to steric considerations of the dichotomy for nucleophilic C species is claimed. The reaction rate constants of C alkyl halide — k- k .Contrastingly t-BuC reacts with alkyl halides with rate constants that are highly sensitive to steric considerations — k 30k- and t-BuI is unreactive. A SET mechanism is proposed for the dianion whereas the classical S 2 mechanism is invoked for the monoanion. For 23 the alcohol is formed with loss of NO in aqueous base. The OH is derived extensively from solvent on the basis of 270 Annu.Rep. Prog. Chem. Sect. B 1999 95 265—282 isotope incorporation analysis. The reaction shows stereochemical retention and a double displacement mechanism initiated by carbanionic attack to generate the cyclopropanone intermediate is suggested. Evidence of concerted displacements in ring closing reactions has been reported. The value of 1.93 for bi-cyclization of 24 is markedly decreased from a value of 6.3 for solvolysis of an analogous saturated system and the value of S‡ 100 JK mol is substantially more negative consistent with the highly constrained transition state expected for concerted bi-cyclization. Similarly the chloroacetic acid catalyzed cyclization of 25 in CHCl is 15 times faster than that of the saturated analogue 26 and in contrast to 26 pinacolic keto product analogous to 27 is not formed by 25. The chloroacetic acid catalyzed cyclization of the allyl silane 28 is 800 times faster than the cleavage of 26.The e.ects of ion pairing of nucleophiles upon reactivity and secondary isotope e.ects have been analyzed. Rate constants have been measured for reaction of methyl tosylate with chloride ion nucleophile that is to varying degrees ion-paired with a number of counterions exhibiting di.ering extents of occlusion. The reaction of free chloride ion could also be measured in each of the reactions and was the same independent of ion pair reagent. Rate constants for the reactions of di.erently ion paired chloride ion were signi.cantly di.erent and were qualitatively correlated with the ion pair dissociation constants.The e.ect of ion pairing on secondary -deuterium kinetic isotope e.ects is nucleophile dependent. Reaction of phenoxide sodium salt with or without the presence of 18-crown-6 gives identical values of k /k 1.02 and 1.03 in DMF and DMSO respectively. With thiophenoxide sodium salt the values of k /k change from 1.13 (free nucleophile) to 1.03 (ion-paired nucleophile) and 1.13 (free nucleophile) to 1.09 (ion-paired nucleophile) in DMF and DMSO respectively. 5 C––O and related substitutions A number of reports for reactions in acetonitrile have appeared. Aminolysis of arylcyclopropanecarboxylic acid esters with substituted benzylamines shows signi.cant dependence of the rate constants on substituents in the leaving group ( 2—3) and amine ( 2— 1.4) with a signi.cant interaction coe.cient for the two types of substituents. Isotope e.ects of k /k 1.2—1.5 arise on replacement of ArCH NH by ArCH ND in the nucleophile.Rate limiting breakdown of the zwitterionic tetrahedral reaction intermediate T with proton transfer through a four center transition state is proposed. A similar conclusion has been reached in the related reaction of benzylamines with aryl ethyl carbonates in which a reaction that is second order in amine is also observed. Hydrolysis in acetonitrile of the p-nitrophenyl tri- .uoroacetate ester by added H O exhibits a third order dependence on [H O] which is characterized by an isotope e.ect of k /k 2.9. Proton inventory is consistent with the involvement of eight hydrogens in the transition state and the rate limiting step is proposed to involve concerted formation of the neutral tetrahedral species T with proton shuttling by a chain of three water molecules.In protic solvents solvolysis and group transfer to amines has been examined. Solvolysis of ethyl choroformate and ethyl thiochloroformate has been studied in a number of solvents using the two-term Grunwald—Winstein analysis. In the former 271 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 case the dominant mechanism involves addition/elimination while in the most ionizing media there is claimed to be nucleophilic solvation of a developing acylium ion.For the thio analogue the acylium ion mechanism predominates while the most nucleophilic solvents are believed to react by addition/elimination. The lactone 29 ionizes to the enolate with a pK of 8.39. Above this pH the mechanism involves water mediated tautomerization to the lactone followed by rate limiting attack of hydroxide ion on the basis of the pH dependence the observed 0.6 and an analysis of the solvent deuterium isotope e.ect of k /k 1.41 which gives the expected inverse isotope e.ect for lyoxide attack after factoring out the pre-equilibrium. The tetrahedral ‘intermediate’ 30 in the hydrolysis of 1-azaadamantan-2-one is stable at low pH has been crystallized and subjected to X-ray structure analysis. The reverse reaction formation from the zwitterionic amino acid takes place with an e.ective molarity estimated to be 10—10M.Dinitrophenyl 4-cyanobenzoate esters react with cyclic secondary amines in 44wt% ethanol in water and the reaction is characterized by 0.63 from which it is concluded that the reaction is concerted bypassing tetrahedral intemediates. It was earlier demonstated that pyridinolysis occurred with 0.96 for which rate-limiting breakdown of T was proposed. The cause of the di.erences in mechanism for the two types of nucleophiles is ascribed to the greater nucleofugality of the secondary amines compared to equally basic pyridines. A chemiluminescence reaction has been studied that involves the reaction of two moles of imidazole with each mole of oxalate phenyl ester.Subsequent reaction of hydrogen peroxide is proposed to yield the oxetane 31 that may be responsible for the chemiluminescence. Phenyl and 4-nitrophenyl chlorothionoformates 32 react with phenolates of conjugate acid pK 10—5 in 3% dioxane—water with 0.55 and 0.47 respectively. It is thus concluded that the reaction is concerted. A comparison with the aminolysis reaction for which the reaction is apparently stepwise leads to the conclusion that the tetrahedral intermediate in the case of transesteri.cation is too unstable to exist. Substitution on the sulfamate 33 by phenolates has been studied in detail and occurs by a three term rate law with two pH independent terms and a hydroxide ion dependent term. The pH independent reaction at low pH occurs with competitive O—C and O—S cleavage.At higher pH sulfonyl group substitution dominates with a pH independent reaction for which there is a change in mechanism with phenolate basicity from an associative reaction with more basic phenolates to a dissociative reaction with less basic phenolates. The -e.ect has been assessed in these reactions again. Hydrazine is a factor of 100 more powerful than the dipeptide Gly—Gly toward phenyl benzoate esters and the values of 0.76 and 0.9 and 1.08 and 1.25 for N H and Gly-Gly respectively indicate similar mechanisms with di.erent transition states. The -e.ect is ascribed to ground state destabilization and transition state or T stabilization through hydrogen bonding in the case of hydrazine.Reactions of anionic -e.ect 272 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 nucleophiles with phenyl acetates reveal little or no di.erences compared to phenolate nucleophiles in the values of Le.er parameters and it is concluded in the light of this and the small magnitude of -e.ects in general that a special name for this group of nucleophiles is unwarranted. Stimulation/inhibition of hydrolysis reactions other than by simple bu.ers has been examined. Peptide helix—loop—helix motifs containing two histidines or two histidines and two arginines have been constructed. They prove to be e.cient catalysts for hydrolysis of p-nitrophenyl esters.The His pK’s were measured by H-NMR and the pH dependence of the hydrolysis reactions is consistent with one of the His residues acting as a nucleophile and the other as a general acid. Peptides containing the two arginines were superior consistent with these residues acting to form a transition state stabilizing ‘anion hole’. Trans-esteri.cation in weakly polar media is catalyzed by metal-tert-butoxides and the e.ciency is in the order LiNaKRbCs consistent with the larger cations having more favorable interaction with the charge-disperse transition state relative to the localized alkoxide ion. Zinc dication complexed to a cyclic triamine stimulates aryl ester hydrolysis in mostly aqueous media. Three models for the mechanism are considered and two are rejected on the basis of kinetic incompetence — the derived rate constants for the water and hydroxide ion are similar in magnitude.The third model involves complexation and activation of the ester (only) to Zn followed by reaction with solvent nucleophiles. The of 0.17 supports rate-limiting nucleophilic attack. Ca.eine and theophilline-7-acetate (hosts) decrease the rate constants for hydroxide ion reaction with aryl benzoate esters and this is suggested to be due to the formation of 1 1 host—ester complexes. The association is characterized by 0.2 which is consistent with net electron donation from the esters to the host. The rate constantdepressing e.ect is surmised to be due to the hydrophobic repulsion of the anionic hydroxide ion nucleophile by the complex.The Dimroth rearrangement has been reviewed. O depending on the degree of N-methyl- 6 Aromatic substitution Nucleophilic 2 vs. SET substitution in the reactions of aromatic radical anions and alkyl halides An electrochemical technique has been employed to distinguish between the extent of S and sulfonates. For a given halide the percent S 2 increased with increasing standard potential of the radical anion. For a given radical anion the percent S 2 decreased with increasing ability of the alkylating agent to act as a one electron acceptor. The nitrogen acid pK’s of protonated 3-aminothiophenes have been measured and range from 3.4—3.7 in 50% DMSO—H ation. Though they are similar in basicity to anilines they appear to add to dinitrobenzofuran exclusively as carbon nucleophiles attaching at the C2 position of the thiophene.In contrast anilines give the kinetic N-adduct initially which subsequently rearranges to the more stable C-adduct. Addition of aniline to 2,4,6-trinitrophenyl ethyl ether occurs with the non-steady 273 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 state intermediacy of the Meisenheimer complex while 2,4,6-trinitrophenyl phenyl ether exhibits clean .rst order conversion to the N-(2,4,6-trinitrophenyl)aniline product. The partitioning of O and C nucleophilic atoms of alkyl substituted phenoxides to the two electrophilic sites (denoted by asterisks) of 34 has been studied in D DMSO by H-NMR. The chemical competence of 35 to participate in the tertiary amine mediated amine acylation e.ected by chlorotriazenes has been established by synthesis structural and chemical characterization. The e.ect of bulk of the tertiary amine on rates of formation of 35 from the chlorotriazene rules out an S 1 mechanism and the zwitterionic intermediate 36 was directly observed in CDCl by H-NMR.Chlorine and nitrogen isotope e.ects are not inconsistent with a stepwise mechanism. - Nitroarenes react with alkyl hydroperoxide anions to yield nitrophenols via the intermediate like 37. Under some conditions large isotope discrimination e.ects indicate that proton removal from 37 is the product determining step.Electrophilic The Gatterman—Koch aromatic formylation of diaryl compounds in acidic media yields aldehydes typically with high para selectivity. But dibenzyl gives predominant (72%) ortho selectivity under conditions of low SbF /dibenzyl and high para selectivity (93%) at high SbF /dibenzyl. The di.erences are ascribed to the relative proportions of di.erent complexes the ortho product is claimed to arise from a sandwich type complex. Arenediazonium ions decompose in tri.uoroethanol to yield 10—43% .uoroarenes. The absence of a radical process is indicated by the absence (1%) of ArH. The ArH becomes a major product in EtOH with good electron withdrawing groups. An HPLCmethod has been developed for monitoring reaction progress in the decay of arene diazonium ions. On the basis of small selectivity toward chloride ion it is concluded that the p-methylphenyl cation reacts as a free carbocation in aqueous solution.The compounds 38 decompose in tri.uoroacetic acid to give the parent anilines and ortho substituted anilines 39 presumably through hydride or carbanionic migrations respectively. Aromatic nitrations by ‘unconventional’ pathways have been reviewed. 274 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 7 Substitutions on heteroatoms Substitution/addition at the N––O group The nitrosation of substituted aryl sulfonamides has been examined in the forward and reverse directions. The nitrosation reaction is general base catalyzed and conversely denitrosation is general acid catalyzed.The latter reaction is characterized by solvent deuterium isotope e.ects for the lyonium ion and chloroacetic acid catalyzed reactions of k /k 1.2—2.0 and 1.5—2.3 respectively depending on the aryl group. The value of Brønsted 0.7 and the conclusion that there is little bond cleavage to the departing NO group lead to the further conclusion that there is considerable imbalance in the TS between the degree of bonding to H and NO and consequent charge build-up on the sulfonamide nitrogen. The acid catalyzed decomposition of 1-methyl-1-nitroso-3- (arylsulfonyl)guanidines 40 occurs with a solvent isotope e.ect on the lyonium ion reaction of k /k 1.6 and no catalysis by NCS ion leading to the deduction that the reaction is either stepwise or concerted involving rate limiting proton transfer. The decay of 1-methyl-1-nitroso-3-benzoylguanidines 41 has also been characterized.Plots of k against [H] exhibit downward curvature and at low [H] the reaction is stimulated by NCS ion. These observations combined with the observation of k /k 1.6 over the full range of acid studied lead to the conclusion that there are parallel pathways both involving rate limiting proton switches in the protonated substrate one concerted with nucleophilic attack on the nitroso group. S-Nitroso-2- mercaptopyridine has been prepared in mildly acidic aqueous media. Its formation is catalyzed by halide ions. It is moderately stable decomposing to give NO and the disul.de and is a potent nitrosating agent to solvent at neutral pH and to Nmethylaniline.Transnitrosation from nitrite esters to phenolate gives a 90% yield of p-nitrophenol (a nitration of the phenolate) and a small amount of nitrite anion. Both ortho substitution of the phenol and blocking the para position with a methyl group yield increased amounts of nitrite 80% in the case of p-MePhOH and the latter compound reacts faster than phenol. The value of for the reaction is 0.75 and it is concluded that the formation of p-NO -phenols occurs by initial O-nitrosation. The yield of nitrite in the reaction of phenol is not a.ected by the concentration of hydroxide ion so it is concluded that the nitrite arises through homolytic scission and the formation and hydrolysis of N O .Cyclodextrins interact with nitrite esters in contrasting ways in acidic and basic 275 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 media. In acid the formation of 1 1 complexes results in protection of the nitrite ester from hydrolysis. In base cyclodextrins accelerate hydrolysis presumably due to the reactivity of ionized hydroxyl groups proximate to the bound esters. That nitrosothiols react in the presence of high concentration of the parent thiols to yield ammonium ion and disul.des has been con.rmed and generalized. The reaction is overall second order and is una.ected by copper ions and EDTA in marked contrast to observation of the reactions at low thiol concentrations. The mechanism is believed to involve formation of an RSN(OH)SR intermediate with further reduction by thiolate attack at sulfur to give the observed products.Ascorbate also reacts as a reducing agent for nitrosothiols by pathways that are copper ion dependent (low ascorbate) and copper ion independent (high ascorbate). Formation of the azoxy compound 42 from N-methylhydroxylamine and nitrosobenzene is characterized by a pH rate pro.le that has a downward break to a hydrogen ion dependent reaction between pH 0.5 and 3. This is taken to indicate a change in rate limiting step with increasing pH from amine attack to breakdown of an addition intermediate. The higher pH reaction shows general acid (0.34) and base (0.2) catalysis. A preassociation mechanism with some speci.c requirements is invoked.Phosphorus and sulfur atoms Isomerization and competing cyclization of uridinyl-3-phosphate di- and tri-esters 43 have been investigated. The kinetic rate laws and e.ects of bu.er catalysts and leaving group and remaining group structure have been detailed and aspects of the mechanisms of the numerous reactions have been clari.ed. The solvolysis of aryl phosphate monoanions and dianions has been studied in tert-butyl alcohol and tert-amyl alcohol. The dianion of p-nitrophenyl phosphate solvolyses 7500 and 8750 times faster in these solvents respectively than in water while the monoanion is 14 and 16 times less reactive in these solvents respectively than in water. The value of 1.1 was determined for the dianion reactions.Combined with previous results that indicate that racemization is faster than solvolysis a reaction involving a metaphosphate intermediate is proposed for the reaction of the dianions. Substitution in CHCl of the chlorine by secondary amines in 44 is 10 fold faster than the compound in which the .uorenyl moiety of 44 is replaced by isopropyl. 1,8-Diazabicyclo[5.4.0]undec-7-ene(DBU) catalyzes the substitution and the reaction with diethylamine occurs as rapidly as with dimethylamine. Fluorenyl H exchange occurs in the presence of Et ND. These facts lead to the mechanism involving E1cb elimination to give a phosphene intermediate that subsequently undergoes amine addition. 276 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 The .rst hydrolytic reaction in aqueous base of phosphonic acid dimethyl ester hydrolysis involves P—Obond cleavage by hydroxide ion attack at phosphorus to give the monomethyl ester.This is indicated by P-NMR analysis of products formed in 50 50 O/O-containing water. The structure—reactivity relationships in the aminolysis of sulfamyl chorides and sulfamate esters in organic solvents have been investigated. The chloride 45 reacts 10 faster than when the H (attached to N) is replaced by Ph. The values of for aminolysis by substituted anilines varies from 4.8— 2.63 and primary isotope e.ects of k /k 2.6—5.3 are observed. An E2 elimination is proposed followed by addition of aniline to the sulfonylamine intermediate. A similar mechanism with a mobile E2 transition state is proposed for aminolysis by secondary amine sulfamate esters.Optically pure sulfuranes 46 have been synthesized. They undergo hydrolysis to sulfoxides in acidic and basic media with opposite stereoselectivities. Isotopically labelled oxygen from solvent is attached to sulfur in the product in the course of the hydrolysis. A number of thermodynamic quantities have been measured or calculated for methanesulfonic acid and derivatives. 8 Vinylic substitutions Halide ion substitution on vinyl iodonium ions has been studied in polar organic media.— Spectroscopic evidence has been obtained for inital formation of hypervalent iodine by chloride ion attachment. At high chloride ion concentration two chloride ion additions occur.The three iodine containing species partition between substitution and elimination to form the alkyne. Substitution occurs with complete inversion and is claimed to proceed by an in-line concerted S 2 mechanism. Nucleophilic substitution at 2-phenylhaloalkenes by methanethiolate in HMPA and DMSO occurs with e.ectively complete retention which is considered insu.cient to warrant the invocation of a concerted displacement. The element e.ect k /k 1 is used to deduce the conclusion that the mechanism is stepwise with rate limiting attack. The anionic intermediate from the addition of some alkoxides and thiolates to a substituted Meldrum’s acid has been observed and the three rate constants for formation decay to products and reactants have been measured in several cases. Structure —reactivity parameters for all three reactions are a.orded.Intrinsic rate constants for these reactions are all substantially greater than in the case of the analogous nitrostilbene system. 277 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 9 Eliminations The solvent promoted elimination of hydrogen halides and brosylates from indene derivatives 47 occurs mainly with anti speci.city and primary deuterium isotope e.ects of k /k 5. This reaction is in competition with solvolysis that occurs through homoallylic cations. The extent of anti elimination increases with the basicity of general bases and the catalysis is also characterized by k /k 4—6. An E2 mechanism is favored for the anti eliminations based on a consideration of the instability of the hypothetical carbanion intemediates.A number of reports appeared concerning the nitrile-forming eliminations of imines. — Substituent e.ects in the leaving group remaining group and in one case the base catalyst primary deuterium isotope e.ects of k /k 2—9 and cross correlations between some of these parameters support the conclusion of an E2 type elimination with a mobile transition state. Predominant elimination is observed in the speci.c acid catalyzed solvolysis of 1- and 2-methoxydihydronaphthalenes 48. Azide trapping experiments yield /k 210 (unitless) and k 10 s based on the assumption of di.usion limited trapping by azide ion.The predominant elimination results from an extemely rapid deprotonation of the cation by solvent k 1.610 s for which aromatization to the product naphthalene is the driving force. k A number of substitution reactions at sulfonic acid groups in polar and unpolar solvents appear to occur by eliminations which exhibit E1cb character. In chloroform the imidazole catalyzed reaction of 49 shows saturation with an equilibrium constant that is dependent on the nature of Ar ( 1.64) and a limiting rate constant for leaving group expulsion k that also has a signi.cant dependence on substituents in Ar ( 1.78). Deuterium incorporation into 50 in D SO solutions and a curved Brønsted plot of log k against amine nucleophile conjugate acid pK for substitution on 50 underpin the mechanistic conclusions. In the case of 51 the downward break in the pH rate pro.le in aqueous solutions above pH 8 the large 1.86 and S‡5 calK mol contrast with the benzene analogue (replace p-OH by H) and indicate that the elimination reaction of the phenol occurs with the intermediacy of a quinone. 10 Additions Br adds to bisketene 52 to give the E-product identi.ed by X-ray crystallographic 278 Annu.Rep. Prog. Chem. Sect. B 1999 95 265—282 analysis. Warming of this compound permits rearrangement to the -dibromolactone that is presumed to occur by ionization on the basis of solvent e.ects. Bromination of 52 is kinetically .rst order in Br in ClCH CH Cl solvent and this is suggested to be the result of intramolecular assistance by one of the ketene functionalities.Consistent with this bromination of some monoketenes is second order in Br . The bromonium ion 53 adds Br to alkenes and the mechanism appears to involve initial dissociation of a collidine on the basis of inhibition of rates with the addition of free collidine. The secondary deuterium kinetic isotope e.ect on formation of bromonium ion from C L and Br has been determined k /k 0.664 (in MeOH) and 0.572 (in dichloroethane). Computational analysis indicates an equilibrium isotope e.ect on this reaction of K /K 0.63. There is little in the way of rehybridization of C atoms in the bromonium ion compared to ethylene the inverse isotope e.ect is ascribed to a new vibrational mode in the bromonium ion — a ‘CH symmetric twist’.A number of reports concerning H—Nuc addition have appeared. Added halide ion enhances the rate of HX addition to alkenes and alkynes in nonpolar organic solvents and stepwise and concerted mechanisms are proposed in the former and latter cases respectively on the basis of the number and nature of products. Acid promoted addition of water in water across the norbornene 54 has been suggested to occur with rate limiting protonation on the basis of the absolute magnitude of the rate constants activation parameters and solvent deuterium isotope e.ects of k /k 1.1—1.3. Addition of water and alcohols to isocyanates is suggested to occur through clusters of two or three such nucleophiles in dilute solutions of water or alcohols. The reactions of alcohols are second and third order in alcohol depending on alcohol concentration.Computational results consistent with these pathways are presented and a concerted mechanism is suggested. Stereochemical e.ects of directing groups have been investigated. Addition of metal alkyls and allyls in ethereal solvents to 55 shows little syn/anti preference with respect to the sulfur atom but metal vinyl aryl and acetylenyl additions show syn/anti preference of between 1.6—7 ascribed to the polarizability of S toward the anionic 279 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 nucleophile. Lithium and magnesium ethyl vinyl and ethynyl additions to 56 in ethereal solvents gave predominantly the cis alcohol — cis/trans6—40. Computational results point to steric and torsional considerations in the TS as the determining factors for preference.The kinetics of equilibrium addition of oxyanions and thiolate ions to benzylidene Meldrum’s acid 57 have been studied. The equilibrium constants for thiolates are 10—10 larger than those for alkoxides. Some of the rate constants for thiolate addition exceed 10 M s. Intrinsic rate constants for thiolate addition are as much as 10 larger for thiolates compared to oxyanions. The di.erences are ascribed to stronger solvation and lower intrinsic carbon basicity of oxyanions compared to thiolates.The mechanism of an aldol condensation has been quantitatively investigated by kinetic study of the reverse reaction. The rate constant for enol addition is within 0.5 log units of that predicted from a correlation based on the Marcus theory. Hydride addition from a dihydronicotinamide to the .uorene 58 yields the thermodynamically disfavored .uorenyl anion instead of the anion of the substituted malononitrile. A number of novel systems have been explored. The disul.de dication 59 shows stereospeci.c addition with aryl but not alkyl alkenes. Addition of pmethoxyphenol across the disilene 60 has been studied in benzene and THF. The stereochemical preference is independent of p-MeO-phenol concentration.The syn/anti ratio is 9 in benzene and 0.25 in THF and this leads to the proposition that the reaction occurs by formation of the zwitterionic intermediate whose lifetime and ability to rotate and form the more thermodynamically favored anti product is greater in THF than benzene. Photochemically generated silene 61 reacts with acetone in acetonitrile and hydrocarbon solvents. Values of 1—1.5 are solvent dependent and reactions with D -acetone give k /k 1.3—3.1. It is proposed that in most cases the rate limiting step involves intramolecular proton transfer in the addition intermediate. S-Alkoxythiazine 62 formation from halosul.limines and basic alcohol was studied as a function of ROH with RCH giving the highest yields. The S-alkoxythiazines are unstable to the addition of mineral acids but are stable in alkaline protic media and organic solvents.References 1 K.K. Laali and P. E. Hansen J. Chem. Soc. Perkin Trans. 2 1998 1167. 2 M. Chollet and J. L. Burgot J. Chem. Soc. Perkin Trans. 2 1998 387. 3 Y. Chiang A. J. Kresge N. P. Schepp V. V. Popik Z. Rappoport and T. Selzer Can. J. Chem. 1998 76 657. 4 Y. Hasegawa J. Chem. Soc. Perkin Trans. 2 1998 1561. 5 J. Nishimoto M. Tabata T. Eguchi and J. Takauchi J. Chem. Soc. Perkin Trans. 2 1998 1539. 6 F. Terrier E. Kizilian R. Goumont N. Faucher and C. Wakselman J. Am. Chem. Soc. 1998 120 9496. 7 C.F. Bernasconi and K. W. Kittredge J. Org. Chem. 1998 63 1944. 8 C.F. Bernasconi A. E. Leyes and L. Garcia-Rio Organometallics 1998 17 4940.9 D.M. Heathcote J. H. Atherton G. A. De Boos and M.I. Page J. Chem. Soc. Perkin Trans. 2 1998 541. 10 M. Aune R. Danielsson A. Hussenius P. Ryberg A. Guorun Kristjansdottir and O. Matsson Acta Chem. Scand. 1998 52 911. 11 W. Galezowski I. Grzeskowiak and A. Jarczewski J. Chem. Soc. Perkin Trans. 2 1998 1607. 12 A. F. Hegarty J. P. Dowling S. J. Eustace and M. McGarraghy J. Am. Chem. Soc. 1998 120 2290. 13 A. Ohno Y. Ishikawa N. Yamazaki M. Okamura and Y. Kawai J. Am. Chem. Soc. 1998 120 1186. 14 J. Gildersleeve R. A. Pascal and D. Kahne J. Am. Chem. Soc. 1998 120 5961. 280 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 15 B. Lin N. Islam S. Friedman H. Yagi D.M. Jerina and D. L. Whalen J. Am. Chem. Soc. 1998 120 4327. 16 B. A. Horenstein and M. Bruner J.Am. Chem. Soc. 1998 120 1357. 17 X. Huang K. S. E. Tanaka and A. J. Bennet J. Am. Chem. Soc. 1998 120 1405. 18 Y. Kusuyama Bull. Chem. Soc. Jpn. 1998 71 685. 19 T.W. Bentley G. Llewellyn and Z. H. Ryu J. Org. Chem. 1998 63 4654. 20 K. T. Liu S. J. Hou and M. L. Tsao J. Org. Chem. 1998 63 1360. 21 K. Tokunaga T. Ohtsu Y. Ohga and K. Takeuchi J. Org. Chem. 1998 63 2209. 22 J. P. Richard P. Szymanski and K. B. Williams J. Am. Chem. Soc. 1998 120 10 372. 23 R. A. McClelland V. E. Licence J. P. Richard S. S. Lin and K. B. Williams Can. J. Chem. 1998 76 1910. 24 K. B. Williams and J. P. Richard J. Phys. Org. Chem. 1998 11 701. 25 N. Palani A. Chadha and K. K. Balasubramanian J. Org. Chem. 1998 63 5318. 26 H. K. Oh E. M. Joung I. H. Cho Y. S. Park and I.Lee J. Chem. Soc. Perkin Trans. 2 1998 2027. 27 C. D. Hall and V. T. Le J. Chem. Soc. Perkin Trans. 2 1998 1483. 28 J. Crugeiras and H. Maskill J. Chem. Soc. Perkin Trans. 2 1998 1901. 29 S. D. Bull L. M.A.R.B. Correira and S. G. Davies J. Chem. Soc. Perkin Trans. 1 1998 2231. 30 J. K. Conner J. Haider M.N. S. Hill H. Maskill and M. Pestman Can. J. Chem. 1998 76 862. 31 M.A. Miranda J. Perez-Prieto E. Font-Sanchis K. Konya and J. C. Scaiano J. Phys. Chem. A 1998 102 5724. 32 L. S. Romsted J. Zhang and L. Zhuang J. Am. Chem. Soc. 1998 120 10 046. 33 M. Novak K. J. Kayser and M. E. Brooks J. Org. Chem. 1998 63 5489. 34 M. Novak L. Xu and R. A. Wolf J. Am. Chem. Soc. 1998 120 1643. 35 L. C. Manege T. Ueda M. Hojo and M. Fujio J. Chem. Soc. Perkin Trans.2 1998 1961. 36 M. Uljana Mayer and R. Breslow J. Am. Chem. Soc. 1998 120 9098. 37 D. N. Kevill and N. H. J. Ismail J. Chem. Soc. Perkin Trans. 2 1998 1865. 38 J. Oszczapowicz and J. Osek J. Chem. Soc. Perkin Trans. 2 1998 1919. 39 T. Lund and K. Bay Jacobsen Acta Chem. Scand. 1998 52 778. 40 S. Fukuzumi T. Suenobu T. Hirasaka R. Arakawa and K. M. Kadish J. Am. Chem. Soc. 1998 120 9220. 41 A. Yurdakul C. Gurtner E.-S. Jung A. Lorenzi-Riatsch A. Linden A. Guggisberg S. Bienz and M. Hesse Helv. Chim. Acta 1998 81 1373. 42 I. Malnar K. Humski and O. Kronja J. Org. Chem. 1998 63 3041. 43 E. J. Corey and D. Daley Staas J. Am. Chem. Soc. 1998 120 3526. 44 S. Alunni A. Pero and G. Reichenbach J. Chem. Soc. Perkin Trans. 2 1998 1747. 45 Y.-R. Fang S.-G. Lai and K.C. Westaway Can. J. Chem. 1998 76 758. 46 H. J. Koh C. H. Shin H. W. Lee and I. Lee J. Chem. Soc. Perkin Trans. 2 1998 1329. 47 H. J. Koh J.-W. Lee H. W. Lee and I. Lee Can. J. Chem. 1998 76 710. 48 K. Venkatasubban M. Bush E. Ross M. Schultz and O. Garza J. Org. Chem. 1998 63 6115. 49 D. N. Kevill and M. J. D’Souza J. Org. Chem. 1998 63 2120. 50 D.M. Heathcote G. A. De Boos J. H. Atherton and M.I. Page J. Chem. Soc. Perkin Trans. 2 1998 535. 51 A. J. Kirby I. V. Komarov and N. Feeder J. Am. Chem. Soc. 1998 120 7101. 52 E. A. Castro A. Hormazabal and J. G. Santos Int. J. Chem. Kinet. 1998 30 267. 53 A. G. Hadd A. L. Robinson K. L. Rowlen and J. W. Birks J. Org. Chem. 1998 63 3023. 54 E. A. Castro M. Cubillos and J. G. Santos J. Org. Chem. 1998 63 6820.55 P. Blans and A. Vigroux J. Am. Chem. Soc. 1998 120 9574. 56 I.-H. Um E.-K. Chung and S.-M. Lee Can. J. Chem. 1998 76 729. 57 M.J. Colthurst A. J. S. S. Kanagasooriam M. S.O. Wong C. Contini and A. Williams Can. J. Chem. 1998 76 678. 58 K. S. Broo H. Nilsson J. Nilsson A. Flodberg and L. Baltzer J. Am. Chem. Soc. 1998 120 4063. 59 M.G. Stanton C. B. Allen R. M. Kissling A. L. Lincoln and M.R. Gagne� J. Am. Chem. Soc. 1998 120 5981. 60 J. Sun S. J. Son and M. P. Suh Inorg. Chem. 1998 37 4872. 61 N. Pirinccioglu and A. Williams J. Chem. Soc. Perkin Trans. 2 1998 37. 62 T. Fuji and T. Itaya Heterocycles 1998 48 359. 63 H. S. Sorensen and K. Daasberg Acta. Chem. Scand. 1998 52 51. 64 F. Terrier M.-J. Pouet K. Gzouli J.-C. Halle� F. Outurquin and C.Paulmier Can. J. Chem. 1998 76 937. 65 M.R. Crampton and I. A. Robatham Can. J. Chem. 1998 76 627. 66 J. M. Dust and R. A. Manderville Can. J. Chem. 1998 76 662. 67 Z. J. Kaminski P. Paneth and J. Rudzinski J. Org. Chem. 1998 63 4248. 68 M. Makosza and K. Sienkiewicz J. Org. Chem. 1998 63 4199. 69 M. Tanaka M. Fujiwara Q. Xu H. Ando and T. J. Raeker J. Org. Chem. 1998 63 4408. 70 P. S. J. Canning K. McCrudden H. Maskill and B. Sexton Chem. Commun. 1998 1971. 71 M.C. Garcia-Meijide C. Bravo-Diaz and L. S. Romsted Int. J. Chem. Kinet 1998 30 31. 72 H. Takeuchi T. Taniguchi M. Masuzawa and K. Isoda J. Chem. Soc. Perkin Trans. 2 1998 1743. 73 J. H. Ridd Acta Chem. Scand. 1998 52 11. 74 L. Garcia-Rio J. R. Leis J. A. Moreirs and F. Norberto J. Chem. Soc.Perkin Trans. 2 1998 1613. 281 Annu. Rep. Prog. Chem. Sect. B 1999 95 265—282 75 J. R. Leis J. A. Moreira F. Norberto J. Iley and L. Garcia-Rio J. Chem. Soc. Perkin Trans. 2 1998 655. 76 S. Amado A. P. Dicks and D. L. H. Williams J. Chem. Soc. Perkin Trans. 2 1998 1869. 77 J. R. Leis A. Rios and L. Rodriguez-Sanchez J. Chem. Soc. Perkin Trans. 2 1998 2729. 78 E. Iglesias and A. Fernandez J. Chem. Soc. Perkin Trans. 2 1998 1691. 79 A. P. Dicks E. Li A. P. Munro H. R. Swift and D. L. H. Williams Can. J. Chem. 1998 76 789. 80 A. J. Holmes and D. L. H. Williams Chem. Commun. 1998 1711. 81 J. Dal Magro and R. A. Yunes Bull. Chem. Soc. Jpn. 1998 71 2381. 82 M. Kosonen E. Youse.-Salakdeh R. Stromberg and H. Lonnberg J. Chem. Soc. Perkin Trans. 2 1998 1589. 83 M.Kosonen K. Hakala and H. Lonnberg J. Chem. Soc. Perkin Trans. 2 1998 663. 84 R. H. Ho. and A. C. Hengge J. Org. Chem. 1998 63 6680. 85 M. J.P. Harger and B. T. Hurman J. Chem. Soc. Perkin Trans. 1 1998 1383. 86 M.C. Mitchell R. J. Taylor and T. P. Kee Polyhedron 1998 17 433. 87 W. J. Spillane F. A. McHugh and P. O. Burke J. Chem. Soc. Perkin Trans. 2 1998 13. 88 W. J. Spillane and C. Brack J. Chem. Soc. Perkin Trans. 2 1998 2381. 89 J. Zhang S. Saito and T. Koizumi J. Am. Chem. Soc. 1998 120 1631. 90 J. P. Guthrie A. R. Stein and A. P. Huntington Can. J. Chem. 1998 76 929. 91 T. Okuyama T. Takino K. Sato and M. Ochiai J. Am. Chem. Soc. 1998 120 2275. 92 T. Okuyama T. Takino K. Sato K. Oshima S. Imamura H. Yamataka T. Asano and M. Ochiai Bull. Chem.Soc. Jpn. 1998 71 243. 93 T. Okuyama H. Oka and M. Ochiai Bull. Chem. Soc. Jpn. 1998 71 1915. 94 X. Chen and Z. Rappoport J. Org. Chem. 1998 63 5684. 95 C. F. Bernasconi R. J. Ketner X. Chen and Z. Rappoport J. Am. Chem. Soc. 1998 120 7461. 96 Q. Meng and A. Thibblin J. Chem. Soc. Perkin Trans. 2 1998 583. 97 B. R. Cho H. S. Chung and N. S. Cho J. Org. Chem. 1998 63 4685. 98 B. R. Cho N. S. Cho S. H. Song and S. K. Lee J. Org. Chem. 1998 63 8304. 99 B. R. Cho N. S. Cho K. S. Song K. N. Son and Y. K. Kim J. Org. Chem. 1998 63 3006. 100 N. Pirinccioglu and A. Thibblin J. Am. Chem. Soc. 1998 120 6512. 101 W. J. Spillane G. Hogan P. McGrath and J. King J. Chem. Soc. Perkin Trans. 2 1998 309. 102 W. J. Spillane P. McGrath C. Brack and K. Barry Chem. Commun.1998 1017. 103 J. F. King and M. S. Gill J. Org. Chem. 1998 63 808. 104 G. Cevasco and S. Thea J. Org. Chem. 1998 63 2125. 105 R. S. Brown M. Christl A. J. Lough J. Ma E. M. Peters K. Peters F. Sammtleben H. Slebocka-Tilk K. Sung and T. T. Tidwell J. Org. Chem. 1998 63 6000. 106 A. A. Neverov and R. S. Brown J. Org. Chem. 1998 63 5977. 107 T. Koerner R. S. Brown J. L. Gainsforth and M. Globukowski J. Am. Chem. Soc. 1998 120 5628. 108 H.M. Weiss and K. M. Touchette J. Chem. Soc. Perkin Trans. 2 1998 1517. 109 H.M. Weiss and K. M. Touchette J. Chem. Soc. Perkin Trans. 2 1998 1523. 110 M. Lajunen J. Jantunen and P. Koiranen Acta Chem. Scand. 1998 52 728. 111 G. Raspoet and M.T. Nguyen J. Org. Chem. 1998 63 6878. 112 G. Raspoet and M.T. Nguyen J. Org. Chem. 1998 63 6867. 113 D. A. Jeyaraj V. K. Yadav M. Parvez and H. M. Gauniyal J. Org. Chem. 1998 63 3831. 114 K. Ando K. R. Condroski and K. N. Houk J. Org. Chem. 1998 63 3196. 115 C. F. Bernasconi and R. J. Ketner J. Org. Chem. 1998 63 6266. 116 J. P. Guthrie and J. A. Barker J. Am. Chem. Soc. 1998 120 6698. 117 X. Q. Zhu and Y. C. Liu J. Org. Chem. 1998 63 2786. 118 V. G. Nenajdenko N. E. Shevchenko and E. S. Balenkova J. Org. Chem. 1998 63 2168. 119 Y. Apeloig and M. Nakash Organometallics 1998 17 1260. 120 C. J. Bradaric and W. J. Leigh Organometallics 1998 17 645. 121 T. Yoshimura M. Ohkubo T. Fujii H. Kita Y. Wakai S. Ono H. Morita C. Shimasaki and E. Horn Bull. Chem. Soc. Jpn. 1998 71 1629. 282 Annu. Rep. Prog. Chem. Sect. B 1999 95 265&mdash

ISSN:0069-3030    

DOI:10.1039/a808580c

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

9 Reaction mechanisms Part (ii) Pericyclic reactions Ian D. Cunningham Department of Chemistry School of Physical Sciences University of Surrey Guildford UK GU2 5XH 1 Introduction During 1998 advances in pericyclic chemistry appear to have been predominantly on synthetic and computational fronts. However a review of pericyclic transition state aromaticity has appeared. It concentrates on magnetic aspects in particular calculated changes in magnetic susceptibility and nucleus-independent chemical shift. Reactions considered include acetylene trimerisation and the triquinacene—diademane rearrangement along with simple Diels—Alder hydrogen shift and electrocyclisation reactions. 2 Cycloaddition reactions Theoretical studies Computational studies are increasingly being directed to more subtle predictions of pericyclic chemistry including stereoselectivity.For example an ab initio study of the [22] cycloaddition of ketenes 1 (RCl H Me) to the chiral aldehyde (S)-2 (RMe CH Ph) predicts a concerted but asynchronous [2 (2 2)] mechanism along with high syn product 3 stereoselectivity (Scheme 1). The prediction is supported by an experiment using 2 (RCH Ph) which yielded syn 3. While ab initio and Density Functional Theory (DFT) calculations are applicable to increasingly complex systems the Frontier Molecular Orbital (FMO) approach is still attractive to organic chemists for prediction and rationalisation of trends in reactivity and regioselectivity. A nice example of this approach using AM1 calculated HOMO/LUMO energies and polarisations has been used to rationalise the normal and inverse electron demand Diels—Alder reaction of 2-azadienes 4 (Raryl alkyl ketone ester) with alkenes such as 5 (R RH cyano ketone ester) (Scheme 2). 283 Annu.Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 1 Kinetics and mechanism The rate equations for both the cheleotropic and hetero-Diels—Alder reactions of 1,2-dimethylidenecyclohexane 6 with SO (Scheme 3) were found to be second order in Novel dienes and dienophiles Activation parameters have been measured for the [42] cycloaddition of benzyne generated within a hemicarcerand across an aromatic ring of the hemicarcerand. The values of G‡ H‡ and S‡ at 61 kJ mol 48 kJ mol and 45 J molK respectively along with the observed .rst-order kinetics are (allowing for statistical factors) typical of an intramolecular reaction although no covalent bond links the reactants.The author proposes the term innermolecular for this type of reaction. Scheme 2 [SO ] suggesting involvement of a second SO molecule in the TS. Ab initio calculations (e.g. MP2/6-31G*) support this with low calculated energies for TS structures (e.g. 7) involving two SO molecules. In contrast a computational study of the reactions of SO with 1-substituted butadienes gives calculated TS structures and activation energies but these models do not involve a second SO molecule in the TS. Experimental kinetic studies still provide important mechanistic information.Values of K and k for complexation and subsequent [42] cycloaddition of 2- alkenylnaphthalenes (alkenyl—CR——CH ) and tetracyanoethene (TCNE) have been determined using UV—vis spectroscopy. The values of K increase from 0.27 to 13.3dm mol across the series RH Pr Pr Et Me neopentyl while a similar pattern is found for the increase in k from 0.012 to 0.92 s (27.1 °C). The similarity in the trend for the equilibrium and rate constants supports the proposal of the complex as an intermediate on the pathway 2-alkenylnaphthaleneTCNEcycloadduct 284 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 3 although adjustment of the geometry of the complex is required to attain the TS. The RBu derivative deviates from the above trend (k is low relative to K) suggesting that this complex must dissociate not merely ‘adjust’ before the correct TS geometry can be attained. The reactivity of in situ-generated 2-(methoxycarbonyl)buta-1,3-diene 8 as a diene towards dimerisation has been assessed (Scheme 4).The ester group appears to activate 8 as a diene towards electron-de.cient dienophiles (normal electron demand). The authors argue that this is contrary to what would be expected from FMO theory and attribute the unexpected reactivity to favourable ‘paralocalisation’ energy (the ‘portion’ of the enthalpy brought about by re-organisation of the bonds only) in the TS. This assumes that bond re-organisation precedes bond formation to a signi.cant degree and calculated TS geometries (generated at the RHF/3-21G level) are quoted in support of this. The kinetics of the intramolecular Diels—Alder cycloaddition (Scheme 5) using ‘strain-activated’ dienophile components 9 (RH MeO) have been investigated.Values of H‡ and S‡ vary little for the compounds shown but V‡ values vary from 28.4 to40 cm mol. These values are similar to those obtained for intermolecular Diels—Alder reactions and therefore suggest that substituent and steric factors dominate V‡ rather than intra- or intermolecularity. Kinetic data for cycloaddition of cyclic and acyclic dienes to a range of thio- and seleno-carbonyl compounds (e.g. Scheme 6) have been published. The magnitude of the activation parameters and the lack of signi.cant solvent e.ect support a concerted mechanism.Noteworthy points include evidence of neutral electron demand for reaction of 10 (Arsubstituted phenyl) and 11 (other reactions showed normal and inverse electron demand) and a 10-fold greater reactivity for the C——Se analogues of 11. Likewise a large amount of kinetic data has been published for the inverse electron demand cycloaddition of 1,2,4,5-tetrazine to alkenes ketene acetals aminals and enol ethers. 285 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 4 Scheme 5 Scheme 6 A transannular Diels—Alder reaction has been studied as a probe for asynchronicity. It is proposed that the asynchronous transition states 12 and 13 (RCO Et) derived from regioisomeric starting materials di.er in stability.This is because the more advanced bond to the enal -C (heavy broken lines) results in a tendency towards a [6,11] bicyclic system in 12 but a less stable [7,10] system in 13. This is supported by experimental reactivities which di.er by a factor of 9 (toluene 120 °C) rising to 740 for catalysis by BF —OEt (CH Cl 25 °C). The latter di.erence is large enough to imply a stepwise zwitterionic mechanism. Evidence for quantum mechanical tunnelling in the [42] dyotropy of 14 (Ar4-ClPh) has been reviewed (Scheme 7). Calculations indicate barrier penetration at 13.5 kJ below that calculated in the absence of tunnelling. 286 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 7 1,3-Dipolar cycloadditions Rate accelerations for Diels—Alder reactions in water are well-known.However 1,3-dipolar cycloadditions of benzonitrile oxide to electron-rich dipolarophiles (cyclopentene and dihydrofuran) show inverse electron demand and are also accelerated in aqueous solvent. For example for benzonitrile oxide to cyclopentene cycloaddition the second-order rate constant k increases from 0.026dm mol s in hexane to 0.085dm mol s in water at 25 °C. This is attributed to a large hydrogen bondinginduced LUMO lowering but a smaller HOMO lowering for electronrich dipolarophiles. With electron-de.cient dipolarophiles (methyl vinyl ketone acrylonitrile and N-methylmaleimide) the cycloaddition shows normal electron demand. Since hydrogen bonding lowers both HOMO and LUMO similarly little or no acceleration is seen.An additional rate enhancement in water compared to tri.uoroethanol is taken as evidence of a hydrophobic e.ect. The kinetics of the 1,3-dipolar cycloaddition of -aryl-N-tert-butylnitrones 15 (Arsubstituted phenyl) to Fischer carbene complexes 16 (MCr W; ArPh 4-MePh) have been studied (Scheme 8). LFER analysis gives a (log k vs. ) value of 1.19 consistent with a concerted [32] mechanism; the high S‡ value of ca. 160 J molK also supports a structured TS. The results indicate a HOMO—LUMO controlled reaction a view supported by theoretical calculations. The reactivity of 1,3-dipoles -phenyl- and -(4-nitrophenyl)-N-tert-butylnitrone towards the cyano group dipolarophile inN—— — C—R (RCBr CN CCl CN CCl(CN) C(CN) CCl ) has been quanti.ed with values for the second order rate constants k that range from 4.210 to 0.30dm mol s.This classi.es the more reactive N—— — C—R as ‘superdipolarophiles’. The normally sluggish reaction of N-methylindole 17 with the 1,3-dipole 3,5- dichloromesitonitrile oxide 18 (Ar3,5-dichloro-2,4,6-trimethylphenyl) to yield cycloadduct 19 (Scheme 9) is greatly accelerated by addition of a Grignard reagent e.g. ethylmagnesium bromide. Involvement of an indol-2-yl anion appears to be ruled out in light of the product regioselectivity and by the lack of deuterium incorporation upon D O quenching.A mechanism involving double co-ordination of the Mg with both the nitrogen of the indole and the electron rich moiety of the dipole is favoured by the authors. The 1,3-dipolar cycloaddition of 20 (XH RPh) to 21 to give 22 (Scheme 10) is accelerated by pressures of up to 1 GPa. The reaction was monitored by FTIR spectroscopy and the rate shows a logarithmic dependence on pressure in the 0—0.3GParange; qualitatively similar behavior is found for other 20 (XNO Cl OMe; RH). 287 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 8 Scheme 9 Scheme 10 Scheme 11 Reactive azomethine ylides 23 (RH Bu 4-MeOPh; RH Ph) were generated by laser .ash photolysis of aziridines 24. Rate constants for reaction of 23 with acrylonitrile (Scheme 11) dimethyl acetylenedicarboxylate and 1,1-dicyanoethene were found to be in the range 2.6 to 140010dm mol s. Solvent e.ects The reactivities of cyclopentadiene 2,3-dimethylbuta-1,3-diene and cyclohexadiene towards N-alkylmaleimides have been analysed to probe the hydrophobic e.ect in the aqueous acceleration of the Diels—Alder reaction.In cases where the maleimide N-alkyl groups are particularly close to hydrophobic groups on the diene as in the endo TS 25 for reaction of 2,3-dimethylbuta-1,3-diene an increase in the length of this alkyl group increases the acceleration due to a hydrophobic e.ect. Likewise increased hydrophobicity near the reaction centre as in reactions of 2,3-dimethylbuta-1,3-diene 288 Annu. Rep. Prog. Chem.Sect. B 1999 95 283—297 and cyclohexadiene when compared to cyclopentadiene also enhances the rate acceleration. A consideration of the Gibbs energies of transfer (propanol to water) for starting materials and TS suggests that the latter e.ect is due to destabilisation of the starting materials with little or no e.ect on the TS. In the former case while the e.ect on the starting materials is the dominant factor some destabilisation remains in the TS. The reactions of 1-heteroaryl-3-arylenones (as substituted C—— C dienophiles) with cyclopentadiene in aqueous micellar solutions (e.g. sodium dodecyl sulfate cetyltrimethylammonium bromide dodecyl heptaoxyethylene ether zinc and copper didodecyl sulfate) have been investigated. Most non metal-containing micelles retard the rate of the Diels—Alder reaction due to localisation of the diene in the micellar hydrophobic region and the dienophile in the outer region (H NMR).Using copper didodecyl sulfate however the reaction is greatly accelerated due to Lewis-acid complexation of enone to copper ions at the micellar surface. Catalysis Catalysis by an antibody generated against a model for the TS 26 (Ar4-carboxyphenyl) of the [3,3] rearrangement (oxy-Cope) of 2-(4-carboxyphenyl)-3(S)- hydroxy-5-phenylhexa-1,5-diene gives a 5300-fold rate enhancement. Temperaturedependent kinetics gave H‡ and S‡ (derived from k determined under conditions of ‘saturation kinetics’) values of 64.4 kJ mol and 96 J molK respectively for the catalysed reaction compared to values of 114.5 kJ mol and 12.5 J molK respectively for the uncatalysed reaction.The observation that catalysis is due more to enthalpic than entropic factors and the contrasting results of NMR experiments showing pre-organisation of the substrate into a cyclic conformation by the antibody are noted as surprising by the authors. Secondary Kinetic Isotope E.ects (KIEs) (k (H)/k (D)0.61 for tetra-deuteriation at both terminals) are interpreted to indicate signi.cant C1—C6 bond formation in the TS. Furthermore the KIEs provide evidence that the rearrangement (as opposed to product release) is rate limiting. The Diels—Alder reaction of cyclohexadiene and benzo-1,4-quinone (Scheme 12) is accelerated ca. 170-fold by addition of 27 (R4-n-heptylphenyl) which is known to form a dimeric capsule 27 (illustrated).NMR evidence shows that the 27 capsule binds two molecules of benzo-1,4-quinone or one of the adduct and although cyc- 289 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 12 lohexadiene is not encapsulated and encapsulated cyclohexadienebenzo-1,4- quinone is not observed the acceleration of the Diels—Alder reaction suggests catalysis by encapsulation. No catalysis was observed with derivatives of 27 which cannot form capsules. This rules out catalysis by formation of hydrogen bonds that lower the dienophile LUMO. Further evidence of encapsulation prior to reaction is provided by observations of a trend to saturation kinetics and of product inhibition. Asymmetric cycloadditions Studies of TADDOLate-TiCl -catalysed (TADDOL,,,-tetraaryl-1,3-dioxolane-4,5-dimethanol) cycloadditions continue.Computational analysis using various HF and DFT methods of three possible TS models 28 29 and 30 predicts a greater stability for 28 but greater reactivity for 29 due to a lower LUMO (localised on the dienophile). This contrasts with earlier reports (K. V. Gothelf and K. A. Jorgensen J. Chem. Soc. Perkin Trans. 2 1997 111) suggesting that complexes with axial Cl (e.g. 28) are more reactive. 290 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 13 Scheme 14 The value of log (dr) (drproduct diastereomeric ratio) for reaction of 31 with cyclopentadienes (Scheme 13 only one diastereomeric product shown) correlates poorly with the solvent parameter E (30) but well when the multi-parameter About—Abraham—Kamlet—Taft model is used.The increased dr in more polar solvents is rationalised by assuming a preference for the greater dipole of the syn-s-cis conformers of 31 and a C()-re TS. The dr in supercritical CO was also determined but no clear trend is evident. Cl 70 °C) despite overall positive volumes of 3 Electrocyclic reactions DFT and ab initio calculations indicate a low energy pseudopericyclic electrocyclisation for the iminodiazomethane E-32 (XNH E geometry about C——N) (Scheme 14) involving an in-plane interaction of the imine N lone pair and the in-plane N—N orbital. However the TS for the Z-32 shows twisting of the imino N—H along with electronic reorganisation (e.g.formation of a C3—C4 bond) similar to a ‘conventional’ electrocyclisation. For the related 32 (XCH ) a TS similar to that for Z-32 is indicated while results for 32 (XO) are inconclusive. A stereoselective thermal electrocyclisation of s-trans-diallene 33 (Rp-tolyl mesoisomer shown) to the dimethylenecyclobutene 34 in the crystalline state has been reported (Scheme 15). The s-trans to s-cis isomerisation that must precede electrocyclisation (though not directly observed) must also preserve crystallinity since no liquid state was observed during the reaction. The reaction was followed by DSC IR spectroscopy and microscopy and is stereoselective as expected for a pericyclic process. Related compounds 33 (RPh) and the rac diastereomer of 33 (Rp-tolyl) showed similar behavior. Volumes of activation V‡ for the conrotatory ring-opening of some cyclobutenes (e.g.trans-3,4-diethyl-1-(methoxycarbonyl)cyclobutene) were found to be small and negative (e.g. 2.2 cm mol CH 291 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 15 Scheme 16 reaction. The result is interpreted as indicative of a particularly ordered and compact TS. It is widely accepted that electrocyclisation of diarylmethyl cations can be accelerated by a further cationic substituent to the 4 .ve-carbon system. Evidence for this in the case of the conversion of 35 to 36 via 37 (RPh 4-MeOPh OMe OH Me H; RH OMe Me Br F) in tri.uoromethanesulfonic acid—tri.uoroacetic acid (Scheme 16) is provided by evidence of increased .uorene product yield as acidity (H ) increases from 10.9 to 13.2 and by the linearity of logk vs.H plots.DFT calculations support this with calculated H‡ values close to experimental ones (ca. 50 kJ mol) which are derived from k a composite of k and K . 4 Sigmatropic rearrangements [1,n] Shifts In contrast to previous years studies of the ‘truly pericyclic’ ene reaction are few. However a possible novel ene reaction has been reported. Reaction of the titanocene sul.do complex Cp* Ti(pyridine)——S (Cp*pentamethylcyclopentadienyl) with allyl chloride (Scheme 17) appears to be via an S 2 route (i.e. bonding of the S to the C3 of allyl chloride).Kinetic evidence is consistent with reversible loss of pyridine to give a reactive 16-electron complex 38 which then reacts with allyl chloride. The authors propose a cyclic TS (see Scheme 17) for this reaction which appears analogous to that for an ene reaction although the availability of lone pair orbitals on the Cl and S along with vacant orbitals on the Ti might suggest a pseudopericyclic mechanism. [n,n] Shifts An intriguing approach to [n,n] shifts is that of the ‘twin-excited state’ as a probe for TS properties. The concept is illustrated by reference to the semibullvalene degenerate 292 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 17 Scheme 18 rearrangement 39 to 40 (Scheme 18). The TS and an associated twin-excited state are considered as in-phase and out-of-phase respectively combinations of the ground state wave functions for 39 and 40 at the point of the TS along the reaction co-ordinate.The theory predicts similar properties for the two states and is supported by the similarity of quantum mechanically computed geometries for the two states. The mechanism of the [3,3] rearrangement in compounds such as semibullvalenes and barbaralanes continues to be investigated. Activation parameters have been measured using variable temperature C NMR for the degenerate Cope rearrangement of a wide range of substituted barbaralanes 41 and semibullvalenes 42 (Rvarious substituents). Conjugating substituents in the 2,6 positions lower the activation barrier but inductively electron-withdrawing substituents raise it; 3,7 disubstitution has relatively little e.ect. Following on from earlier work the Cope rearrangement of chloro- bromo- and iodobullvalene in solution and in the solid state was studied by X-ray crystallography and C NMR.Solution NMR studies con.rmed the main processes interconversion of 2- (i.e. 43) and 3-substituted isomers (see Scheme 19 for arbitrary numbering) degenerate 2-substituted isomer interconversion and a degenerate 3-substituted isomer interconversion via the 1-substituted isomer. X-Ray crystallographic studies and solid state C NMR experiments for the bromo- and iodo-substituted compounds showed the predominance of the 2-isomer in the solid state. However there was also evidence of a degenerate interconversion which despite the bond reorganisa- 293 Annu.Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 19 Scheme 20 tion requires relatively little displacement of the atoms within the crystal lattice (illustrated in Scheme 19 where dashed lines indicate bonds being broken). A stepwise mechanism has been proposed for the Cope rearrangement of 44 to 45 (Scheme 20) since activation parameters (132 kJ mol) are found to be similar to those for reaction of 46 with O via rate-limiting formation of 47. Furthermore there is evidence that O trapping of an analogous diradical competes with the Cope rearrangement of 44 to 45. The conclusions are supported by the results of CASPT2 calculations. Calculations using restricted and unrestricted DFT (Becke3LYP/6-31G*) predict a diradical TS (and intermediate) for the [5 ,5 ] sigmatropic rearrangement of 5,5a,10,10a-tetrahydroheptalene 48.The intermediate is a true one with the 1,1 bond fully formed before breaking of the 5,5 bond. Likewise diradical transition states are predicted for the rearrangement of (Z,Z)-deca-1,3,7,9-tetraene 49 via both [5,5] and [5 ,5 ] routes although steric factors favour the former. Even the [3,3] (Cope) rearrangement of 2,5-divinylhexa-1,5-diene appears to proceed via a diradical TS and intermediate. A ketyl radical-anion ‘triggered’ [3,3] sigmatropic shift involving -allyloxyenones has been reported. For example the rearrangement of 50 to 51 on treatment with BuSnH—AIBN under mild conditions (benzene 80 °C) (Scheme 21) is proposed to occur via [3,3] rearrangement of the radical-anion 52.An intermediate where the 294 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 21 radical picks upH before the [3,3] rearrangement would also be expected to rearrange readily but the preliminary evidence favours rearrangement via the radical-anion. For example no -allyloxycyclohexanone is detected on water-quenching of the reaction and competition experiments suggest that the [3,3] rearrangement of 52 is faster than reaction with further BuSnH. In contrast to the facile anionic oxy-Cope [3,3] rearrangement attempted anionic amino-Cope rearrangement gave no [3,3] product (e.g. on treatment of 53 with BuLi—THF at 78 °C) but in many cases the allyl group is lost from the reactant.While accepting the theory that an adjacent anion weakens the ‘breaking’ bond the authors quote DFT-calculated bond dissociation energies to show that while homolytic cleavage (favouring [3,3]) is enhanced in the amino case heterolytic cleavage (favouring deallylation) is enhanced to an even greater degree. Houk has published a computational study of the [3,3] rearrangement of compounds 54 55 and 56. RHF MP2//RHF BLYP and Becke3LYP methods using a 6-31G* basis set predict a concerted mechanism. Furthermore the MP2 DFT and hybrid methods give activation enthalpies (111—143 kJ mol) closest to experimental values. The CASSCF method favours a stepwise biradical mechanism. The apparent constancy of the activation parameters across the diverse range of substrates is noted. 295 Annu.Rep. Prog. Chem. Sect. B 1999 95 283—297 Scheme 22 [n,m] Shifts A novel [2,3] rearrangement of N-benzyl-O-allylhydroxylamines 57 (R RH Me Ph) has been described (Scheme 22). Although mechanistic detail is as yet sparse the product regioselectivity and the lack of ‘crossover’ products in mixed reactions suggest an intramolecular pericyclic process and a TS as shown. References 1 H. Jiao and P. von R. Schleyer J. Phys. Org. Chem. 1998 11 655. 2 B. Lecea A. Arrieta I. Arrastia and F. P. Cossý� o J. Org. Chem. 1998 63 5216. 3 T.M.V.D. Pinho e Melo R. Fausto and A.M. d’A. Rocha Gonsalves J. Org. Chem. 1998 63 5350. 4 R. Warmuth Chem. Commun. 1998 59. 5 T. Fernandez J. A. Sordo F. Monnat B.Deguin and P. Vogel J. Am. Chem. Soc. 1998 120 13 276. 6 T. Ferna� ndez D. Sua� rez J. A. Sordo F. Monnat E. Roversi A. Estrella de Castro K. Schenk and P. Vogel J. Org. Chem. 1998 63 9490. 7 L.H. Klemm W. C. Solomon and A. P. Tamiz J. Org. Chem. 1998 63 6503. 8 C. Spino J. Crawford Y. Cui and M. Gugelchuk J. Chem. Soc. Perkin Trans. 2 1998 1499. 9 M. Buback T. Heiner B. Hermans C. Kowollik S. I. Kozhushkov and A. de Meijere Eur. J. Org. Chem. 1998 107. 10 U. Rohr J. Schatz and J. Sauer Eur. J. Org. Chem. 1998 2875. 11 J. Sauer D. K. Heldmann J. Hertr J. Krauthan H. Sichert and J. Schuster Eur. J. Org. Chem. 1998 2885. 12 Y. L. Dory D. G. Hall and P. Deslongchamps Tetrahedron 1998 54 12 279. 13 K. Mackenzie K. B. Astin E. C. Gravett R. J.Gregory J. A. K. Howard and C. Wilson J. Phys. Org. Chem. 1998 11 879. 14 D. van Mersbergen J. W. Wijnen and J. B. F. N. Engberts J. Org. Chem. 1998 63 8801. 15 M.L. Yeung W.-K. Li H.-J. Liu Y. Wang and K. S. Chan J. Org. Chem. 1998 63 7670. 16 L. Eberson J. J. McCullough C. M. Hartshorn and M.P. Hartshorn J. Chem. Soc. Perkin Trans. 2 1998 41. 17 T. Benincori F. Sannicolo` L. Trimarco L. Bonati S. Grandi D. Pitea and C. Gatti J. Phys. Org. Chem. 1998 11 455. 18 V. Melai A. Brillante and P. Zanirato J. Chem. Soc. Perkin Trans. 2 1998 2447. 19 C. Gaebert C. Siegner J. Mattay M. Toubartz and S. Steenken J. Chem. Soc. Perkin Trans. 2 1998 2735. 20 A. Meijer S. Otto and J. B. F. N. Engberts J. Org. Chem. 1998 63 8989. 21 S. Otto J. B. F. N. Engberts and J.C. T. Kwak J. Am. Chem. Soc. 1998 120 9517. 296 Annu. Rep. Prog. Chem. Sect. B 1999 95 283—297 22 E.M. Driggers H. S. Cho C. W. Liu C. P. Katzka A. C. Braisted H. D. Ulrich D. E. Wemmer and P. G. Schultz J. Am. Chem. Soc. 1998 120 1945. 23 J. Kang G. Hilmersson J. Santamarý� a and J. Rebek Jr. J. Am. Chem. Soc. 1998 120 3650. 24 J. I. Garcý� a V. Martý� nez-Merino and J. A. Mayoral J. Org. Chem. 1998 63 2321. 25 C. Chapuis A. Kucharska P. Rzepecki and J. Jurczak Helv. Chim. Acta 1998 81 2314. 26 W.M. F. Fabian V. A. Bakulev and C. O. Kappe J. Org. Chem. 1998 63 5801. 27 F. Toda K. Tanaka T. Tamashima and M. Kato Angew. Chem. Int. Ed. 1998 37 2724. 28 G. Jenner Tetrahedron 1998 54 2771. 29 T. Ohwada T. Suzuki and K. Shudo J. Am. Chem. Soc. 1998 120 4629. 30 Z. K. Sweeney J. L. Polse R. A. Andersen and R. G. Bergman J. Am. Chem. Soc. 1998 120 7825. 31 S. Zilberg Y. Haas D. Danovich and S. Shaik Angew. Chem. Int. Ed. 1998 37 1394. 32 L.M. Jackman E. Fernandes M. Heubes and H. Quast Eur. J. Org. Chem. 1998 2209. 33 Z. Luz L. Olivier R. Poupko K. Mu� ller C. Krieger and H. Zimmermann J. Am. Chem. Soc. 1998 120 5526. 34 W.R. Roth R. Gleiter V. Paschmann U. E. Hackler G. Fritzsche and H. Lange Eur. J. Org. Chem. 1998 961. 35 B. R. Beno J. Fennen K. N. Houk H. J. Lindner and K. Hafner J. Am. Chem. Soc. 1998 120 10 490. 36 E. J. Enholm K.M. Moran P. E. Whitley and M. E. Battiste J. Am. Chem. Soc. 1998 120 3807. 37 H. Y. Yoo K. N. Houk J. K. Lee M. A. Scialdone and A. I. Meyers J. Am. Chem. Soc. 1998 120 205. 38 K. A. Black S. Wilsey and K. N. Houk J. Am. Chem. Soc. 1998 120 5622. 39 S. G. Davies S. Jones M. A. Sanz F. C. Teixeira and J. F. Fox Chem. Commun. 1998 2235. 297 Annu. Rep. Prog. Chem. Sect. B 1999 95 283&m

ISSN:0069-3030    

DOI:10.1039/a808609e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

9 Reaction mechanisms Part (iii) Bioorganic enzyme-catalyzed reactions Nigel G. J. Richards Department of Chemistry University of Florida Gainesville FL 32611 USA 1 Introduction Progress in structural and mechanistic enzymology continued to be rapid in 1998 and so precludes a comprehensive description of recent developments in the understanding of the myriad catalytic mechanisms employed in primary and secondary metabolism. Continuing the philosophy of the last report in this series, this review seeks to provide an overview of enzyme mechanisms that have not been the subject of extensive literature reviews and in which chemically interesting strategies are employed in catalyzing the overall transformation. In addition two examples of mechanism-based enzyme inhibitors are discussed for which the chemical basis of inhibition has only recently been established.2 Fundamental ideas in enzyme catalysis Debate continues about the fundamental energetic factors underlying the large rate enhancements in enzyme catalyzed reactions. A superb introduction to many current conceptual problems has been provided by the publication of a series of minireview articles on the role of low-barrier hydrogen bonds, reorganization energy and electrostatic e.ects in transition state stabilization and catalysis. In particular the role of low-barrier hydrogen bonds (LBHBs) in catalysis as pointed out in an earlier review, remains controversial with di.ering interpretations of theoretical and experimental results.— The primary issue in the debate concerns the amount of covalent character that is present in the LBHB within the active site relative to that in the cognate hydrogen bond in solution. Opponents argue that LBHBs can only be catalytically e.ective when formed in non-polar environments but that such hydrogen bonds will be less e.ective at stabilizing charged transition states than regular hydrogen bonds acting via electrostatic interactions. An additional energetic penalty must be paid when polar substrates are desolvated on entering the non-polar environment from aqueous solution which might also increase the activation energy for the reaction.Attention has therefore turned to establishing the existence of very short hydrogen bonding interactions using structural methods.Nuclear magnetic resonance (NMR) spectroscopy is often employed to identify protons that participate in LBHBs 299 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 and an excellent review of the advantages in using this technique has been published. Very high-resolution X-ray crystal structures also provide experimental insight into the existence and role of LBHBs. Thus electron density from hydrogen atoms can be observed in the structure of subtilisin solved at 0.78Å resolution, and a very short hydrogen bond between the histidine and aspartic acid residues in the Ser—His—Asp catalytic triad originally assigned by NMR measurements, appears to be present. The hydrogen is shared equally between the heteroatoms in this unusual hydrogen bond. With the exception of reactions involving electron transfer, discussions of enzyme catalysis are couched in terms of transition state theory and describe reacting nuclei using classical models.Contributions to catalysis from the quantum mechanical (QM) properties of particles are usually ignored due to the relatively short de Broglie wavelengths associated with heavy atoms of su.cient energy to surmount the kinetic energy barriers present in enzyme-catalyzed reactions. The situation is less clear in the case of hydrogen transfer from donor to acceptor atoms in the transition state. For a protium atom (H) with an energy of 10 kJ mol the de Broglie wavelength is 0.5Å which is similar in magnitude to distances anticipated in many biological hydrogen transfer reactions.This is also true for enzymes employing radical intermediates in order to overcome chemical problems associated with heterolytic bond cleavage. The extent of the contribution ofQMnuclear tunneling to catalysis has therefore been explored for a number of enzymes including dehydrogenases, amine oxidases and lipoxygenase. Evidence for tunneling in hydrogen transfer reactions is provided by the observation of anomalous kinetic isotope e.ects (KIEs) on substitution of a proton in the substrate by deuterium or tritium. As a result not only must the height of the energy barrier for a chemical step be considered in rationalizing enzyme catalysis but also the width given that this parameter determines the tunneling probability of the hydrogen nucleus.The motional properties of the residues within the active site also modulate tunneling since the barrier to reaction links the potential energy wells describing reactants and products. Two models have been proposed for the coupling of enzyme dynamics and hydrogen tunneling e.ects. In the .rst classical thermal .uctuations of the enzyme mediate hydrogen transfer by shortening the tunneling distance. In the alternate model a .uctuating double well system is employed in which there is a speci.c thermally excited protein mode that alters barrier width and shape allowing the particle to tunnel to the well describing the product state. The wavefunctions’ coherency (which enables tunneling) is then destroyed by active site dynamic motions that trap the system in the product state. Both models .t the known KIE behavior in bovine serum amine oxidase but predict di.erent dependence of the KIEs on temperature.Experiments to test which model (if any) correctly describes experimental observations over a range of temperatures are underway. On the other hand the contribution of tunneling e.ects to catalysis remains controversial because the extent of enzyme catalysis is evaluated by comparing the rates of the enzymecatalyzed and uncatalyzed reactions. If extensive nuclear tunneling takes place in the uncatalyzed reaction then its e.ects will cancel out when k and k are compared. Recent calculations on carbonic anhydrase suggest that the tunneling contribution to the reaction rate is almost identical in the reaction catalyzed by carbonic anhydrase and the non-catalyzed process in aqueous solution. Good experimental data to support such calculations for non-catalyzed systems are however hard to .nd.In this 300 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 regard additional studies of KIEs associated with well-chosen model chemical reactions will be necessary to resolve this issue. 3 Co-factors biosynthesis catabolism and model systems Enzymes catalyzing the biosynthetic transformations leading to redoxactive cofactors continue to be characterized. A recent interesting example has been provided by studies on 4-(-ribofuranosyl)aminobenzene 5-phosphate (-RFAP) synthase which mediates a ce ntral step in the cellular synthesis of methanopterin 1 a modifed folate that is employed in C-1 metabolism by methanogenic archea.This reaction is unique in that the phosphoribosyl transfer yields a C-glycoside with concomitant loss of carbon dioxide. There is evidence that PLP may be present within the active site although the co-factor does not apparently play a catalytic role in the condensation reaction. In common with other PRTases,¡X the kinetic mechanism involves sequential binding of the two substrates to the enzyme to form a ternary complex. The proposed mechanism for the reaction involves an interesting modulation of reactivity in which the aromatic -electrons attack PRPP 2 via either an S1 or S2 mechanism (Scheme 1). Loss of CO from the resulting intermediate 3 then yields the nal product in a step that has chemical precedent. -Alanine 4 is an essential component in the biosynthesis of pantothenate 5 and is produced in Escherichia coli by decarboxylation of aspartic acid.This reaction is catalyzed by the enzyme aspartate decarboxylase but does not require pyridoxal phosphate as a co-factor. Instead there is a pyruvoyl residue at the active site that participates in imine formation with aspartate. Loss of CO from the resulting intermediate 6 then takes place to give -alanine after protonation and hydrolysis (Scheme 2). The catalytic pyruvoyl residue is produced in an interesting intramolecular cleavage reaction of the inactive pro-protein between residues Gly-24 and Ser-25 an example of N¡XO acyl transfer (Scheme 3). The mechanism for the activation step probably involves attack of the Ser-25 side chain on the adjacent peptide bond to yield an ester intermediate 7.-Elimination of the ester gives an enamine 8 which is then hydrolyzed to the catalytically active pyruvoyl moiety. Indirect evidence for this protein splicing reaction has been reported but the likely existence of the ester has now been demonstrated from the crystal structure of aspartate decarboxylase at 2.2 301 Annu. Rep. Prog. Chem. Sect. B 1999 95 299¡X334Scheme 1 Scheme 2 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 302 Scheme 3 resolution. Chemical models suggest that protonation of the amine is necessary to favor formation of the ester 7 at equilibrium. Signi.cant progress continues to be made in characterizing the pathways for formation and degradation of the essential co-factor thiamin 9. In prokaryotes 9 is hydrolyzed to yield the thiazole 11 and the pyrimidine 12 in a reaction catalyzed by the enzyme thiaminase-1 (Scheme 4).The biological function of this activity remains unknown. At present there are no protein sequences that are homologous to that deduced for thiaminase-1 on the basis of the sequence cloned for the gene encoding the enzyme in Bacillus thiaminolyticus. Thiaminase-1 will accept a number of nucleophiles as substrates including amines (aniline quinoline pyridine) thiols (cysteine dithiothreitol) and water. Although co-factor breakdown could proceed via direct attack of the nucleophile on 9 several lines of evidence suggest that the cleavage reaction involves a double displacement mechanism.— Site-directed mutagenesis studies implicate the Cys-113 thiolate as the enzyme-based nucleophile, a hypothesis that is supported by the X-ray crystal structure of thiaminase-1 as its covalently modi.ed form after reaction with 13. The mechanism of co-factor breakdown is therefore proposed to proceed by thiolate attack on the pyrimidine ring to yield a zwitterion that can expel the thiazole leaving group to give the modi.ed enzyme intermediate 10 (Scheme 4).Generation of the thiolate may involve general base catalyzed deprotonation of Cys-113 by the carboxylate of Glu-241. Subsequent attack of a nucleophile then proceeds to regenerate an anion from which the thiolate can be released. This proposal is consistent with the kinetic mechanism (ping-pong) and the stereochemistry of the overall reaction.303 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 4 4 Free radicals in enzyme-catalyzed reactions There is an increasing appreciation of the role of free radical intermediates in extending the repertoire of bioorganic transformations, despite the need to control the inherent reactivity of such species in order to prevent side reactions. Enzymes demonstrated to employ transient protein-based radicals in their catalytic mechanism generally perform reactions that proceed only under extreme conditions in model systems. For example heterolytic cleavage of the bond between two adjacent carbonyl groups does not proceed readily in the absence of covalent catalysis presumably because an acyl anion must be formed as an intermediate. Pyruvate formate lyase (PFL) an enzyme that plays a central role in anaerobic glucose fermentation can catalyze the interconversion of pyruvate and coenzyme A into acetyl-CoA and formate (Scheme 5). This reaction is reversible and proceeds by a ping-pong kinetic mechanism evidence for which has been provided by the isolation of the acetyl-PFL intermediate. C—C bond cleavage is accomplished using a homolytic mechanism although the details of the reaction remain controversial.PFL is synthesized as an inactive enzyme that is activated by another enzyme in an oxygen-dependent reaction. The active form of PFL has been shown to possess a protein-based radical located at Gly-734. The oxidation of glycine to a glycyl radical is chemically unprecedented although one model study has been reported. No structure is yet available for PFL but mutagenesis studies have implicated both Cys-418 and Cys-419 as residues that participate in 304 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 5 catalysis. Two mechanisms have been proposed for the reaction both of which are not completely consistent with all experimental observations, and which lack chemical precedent for many of the individual steps. In the simplest of these, the glycyl radical generates a thiyl radical in the Cys-419 side chain that can then react with pyruvate to yield a radical intermediate 14 (Scheme 6A). Homolytic bond cleavage then gives a thioester intermediate and formate radical. The protonation state of the carboxylic acid in the enzyme has not been established although the cleavage of -ketoesters by Fenton’s reagent suggests that reaction takes place via the neutral acid. Hydrogen abstraction from Gly-734 regenerates the protein-based radical with concomitant production of formate.A series of acetyl transfers in which Cys-418 has been shown to participate then give the free enzyme and acetyl-CoA. Gly-734 and Cys-419 must therefore be in close proximity if hydrogen abstraction is to yield the thiyl radical. Evidence to support the proximity of Gly-734 and Cys-419 has recently been provided however by interesting EPR studies of the mechanism by which PFL is inactivated by oxygen. Remarkably oxygen inactivates PFL by causing breakdown of the protein into two fragments of 82 and 3 kDa the cleavage site being located at Gly-734. By monitoring the radical species formed by incubation of wild type PFL with oxygen at 77 K a long-lived sul.nyl radical (RSO·) was detected that was formed by oxidation of Cys-419.Assignment of Cys-419 as the radical site was accomplished by demonstrating that the same radical was formed by oxygen treatment of the C418A PFL mutant. Incubation of the C419A PFL mutant in which Cys-419 was replaced by an alanine residue with oxygen at 77 K however gave a di.erent species that was assigned as a peroxyl radical (ROO·) located at C of Gly-734. One plausible mechanism that is consistent with these data involves initial addition of oxygen to the Gly-734 radical in the free enzyme followed by reaction of the resulting peroxyl radical with the Cys-419 thiol and cleavage of the resulting adduct to yield the hydroxylated Gly-734 and the sul.nyl radical 17 (Scheme 7). C—Nbond cleavage can then take place to give the C-amidated 82 kDa fragment and the aldehyde 18.Although alternative mechanisms can be proposed that yield 17, all of these require that Gly-734 and Cys-419 be separated approximately by the length of molecular oxygen. A second mechanistic proposal however suggests that the key thiyl radical that initiates C—C bond cleavage is located on Cys-418 and avoids the formation of formate radical anion as an intermediate. Thus after the initial addition of the Cys-419 thiolate anion radical to the protonated form of pyruvate thiyl radical attack yields a cyclic intermediate that can undergo cleavage to the acetylated enzyme derivative 15 (Scheme 6B).Hydrogen migration followed by loss of formate then regenerates the thiyl radical which abstracts a hydrogen atom from Gly-734. Subsequent reaction of coenzyme A with the thioester gives acetyl-CoA and the free enzyme. This proposal is based on the observation that treatment of PFL with hypophosphite a formate analog in the presence of acetyl-CoA inactivates the enzyme by formation of the dead end product 16. This result is not easily accommodated by the .rst mechanism. 305 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 6A Galactose oxidase (GAO) is another enzyme for which protein-based radicals appear to play an important catalytic role. At .rst glance this is a suprising observation as the enzyme catalyzes the two-electron oxidation of the sugar substrate to the aldehyde 19 with the concomitant reduction of oxygen to hydrogen peroxide (Scheme 8).The involvement of a protein-based radical in the catalytic mechanism was .rst postulated on the basis of the unusual EPR spectrum of the redox-activated form 306 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 6B of GAO. As the enzyme also contains a Cu() metal center these spectroscopic properties suggested that a protein-based radical was anti-ferromagnetically coupled to the metal center. Subsequent EPR analysis of the protein radical obtained by oxidation of the metal-free apoenzyme showed that the radical resided on an aromatic ring.The most likely candidate is an unprecedented Cu() ligand containing a thioether bond between the sulfur atom of Cys-228 and the aromatic moiety of 307 Annu. Rep. Prog. Chem. Sect. B 1999 95 299¡X334Scheme 7 Scheme 8 Tyr-272 20 that was revealed in the recent crystal structure of GAO at 1.7Å resolution. Whether the observed structure represents the active form of the enzyme or inactive one-electron reduced GAO remains controversial. Kinetic isotope e.ect (KIE) measurements have therefore been employed in pre-steady state experiments to probe the mechanism by which galactose is oxidized. The presence of deuterium at C-6 of galactose gives rise to a dramatic H/D KIE of approximately 22 on anaerobic reduction of the enzyme by substrate. An H/D KIE of 8 persists in the oxygendependent reaction suggesting that the isotope is involved in a rate-limiting step for hydrogen peroxide formation.In an interesting observation the temperature dependence of this large KIE is consistent with the hypothesis that hydrogen tunneling takes place during catalysis. A mechanism that is consistent with these and other steadystate data has been proposed in which proton transfer from the substrate bound to the Cu(..) center occurs followed by hydrogen atom and electron transfer to give the reduced catalytic center (Scheme 9). As suggested in early studies using heterogeneous enzyme preparations containing protein in a variety of oxidation states, recent work has con.rmed that GAO exhibits ping-pong kinetics. Therefore oxygen-binding 308 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 9 takes place after release of the aldehyde product from the active site and subsequent release of hydrogen peroxide restores the active form of the free enzyme. There is considerable evidence to support the existence of catalytically important protein-based radicals especially from studies of ribonucleotide reductase (RNR). — Amine mono-oxygenases also use protein-based radicals,— but a variety of complexco-factors including pyrroloquinoline quinone (PQQ) 21, and lysine tyrosylquinone (LTQ) 22, are present in the catalytic sites. Direct detection of substrate-based radical intermediates during RNR catalysis however has remained an elusive goal. New studies on the mechanism of RNR inhibition by the .uorinated 309 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 nucleotide analog 23 (Scheme 10) have now provided the .rst indirect evidence for the validity of the hypothesis that the initial step in the chemistry catalyzed by RNR is hydrogen abstraction at C-3 of the ribose ring. Incubation of 23 the phosphorylated derivative of a potential new anti-cancer agent, with Escherichia coli RNR yields an inactivated enzyme species in which the tyrosyl radical is lost. EPR spectroscopic analysis however indicates that a new radical species is formed the signal for which is altered if the inhibitor is deuterated at C-6. Two structures for the radical intermediate 24 and 25 in which .uoride has been lost from the molecule are consistent with the EPR observations. These structures are formed by initial transformation of 23 into the unsaturated ketone 26 (Scheme 10), which can then react with an enzymebased nucleophile.Hydrogen abstraction by one of two possible thiyl radicals then yields inactivated RNR and either 24 or 25. Regardless of which of these alternative structures is actually formed during RNR inactivation the initial step leading to their formation must involve hydrogen abstraction at C-3 of the nucleoside analog providing further evidence in support of the generally accepted mechanism of nucleotide reduction. 5 Nitrogen metabolism The oxygen-dependent oxidation of urate 27 is important in the nitrogen metabolism of plants animals and bacteria. Indeed the urate oxidase from Aspergillus .avus has been used as a drug in the treatment of hyperuricemic disorders. The high nitrogen content of bicyclic ring systems such as that of 27 gives rise to unexpected patterns of reactivity as observed in recent studies on the enzyme urate oxidase. Although early work that is still cited in many text books suggested that urate oxidase converts 27 to allantoin 28 and carbon dioxide, recent NMR studies have shown that the true products of the enzyme-catalyzed conversion are 5-hydroxyisourate 29 and hydrogen peroxide. — The product of the reaction then undergoes a series of nonenzymatic transformations that yield CO and allantoin 28 (Scheme 11).Urate oxidase is of mechanistic interest given that it catalyzes the redox reaction without the involvement of transition metals or co-factors such as .avins. The similarity of the urate ring-system to that of .avins has lead to the proposal that the reaction may proceed via similar hydroperoxide intermediates to those that have been demonstrated in .avin-dependent enzymes and recent studies using O-labeled oxygen and water have shown that the oxygen atoms in the hydrogen peroxide formed during the reaction are both derived from molecular oxygen. Furthermore the hydroxy substituent at C-5 is derived from water. These observations are consistent with a mechanism in which hydrogen peroxide is eliminated from a hydroperoxide intermediate 30 to yield an unstable intermediate 31 that can react with water to give the .nal 310 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 10 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 311 Scheme 11 product (Scheme 12). This proposal is supported by stopped .ow spectroscopic studies in which a series of intermediates can be observed whose absorbance characteristics are consistent with those predicted by the mechanistic scheme. In the absence of detailed computational studies it is interesting to speculate on the unusually high reactivity at C-5 in the proposed intermediate 31 which may be associated with the antiaromatic character of the 8-electron system evident in one of the resonance structures 32 that can be drawn for 31. Progress in characterizing the structures and mechanisms of glutamine-dependent amidotransferases continues to be made.— As discussed previously, these enzymes usually possess two active sites that participate in (i) catalyzing the hydrolysis of glutamine to ammonia and glutamate and (ii) binding and/or activating an electrophilic species that can react with ammonia.Structural evidence suggests that all amidotransferases possess a ‘‘channel’’ through which ammonia can pass from one 312 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 12 active site to the other sequestering this critical intermediate from solvent during the nitrogen transfer reaction. Carbamoyl phosphate synthetase (CPS) which catalyzes the formation of carbamoyl phosphate 33 from glutamine bicarbonate and ATP is a remarkable amidotransferase possessing three active sites. CPS from Escherichia coli is composed of two subunits the smaller of which contains a Cys—His—Glu catalytic triad of residues responsible for catalyzing glutamine hydrolysis, placing this enzyme in the family of Class I amidotransferases. Chemical modi.cation and site-directed mutagenesis experiments identi.ed the Cys-269 thiolate as the active site nucleophile. This prediction was con.rmed by the X-ray crystal structure of CPS, which also revealed that the fold of this glutaminase domain was similar to that observed in / hydrolases. The mechanism of glutamine hydrolysis was therefore proposed to be similar to that in thiol proteases in which histidine acts to protonate the nitrogen leaving group yielding a thioester intermediate after attack of the thiolate anion on the side chain amide of the substrate.Support for the existence of a thioester as an intermediate in the activity of the CPD small subunit was obtained in biochemical studies of CPS and the C269S CPS mutant in which the cysteine residue is replaced by serine. In an unexpected result the thioester 34 (Fig. 1) has now been observed in the X-ray structure of the H353N CPS mutant crystallized in the presence of glutamine. This mutant in which His-353 is substituted by asparagine had been shown to bind glutamine although it exhibited no detectable glutaminase activity. The original crystallography experiments were therefore aimed at determining the mode of glutamine-binding but on solving the structure it was evident that the glutamylated enzyme had been crystallized. The hydrolytic stability of the thioester was proposed to result from the inability of the mutant enzyme to activate water su.ciently for nucleophilic addition using general base catalysis.In contrast to the mechanism of glutamine activation which is apparently identical in known amidotransferases even though the glutaminase active site structures of the Class I and Class II families di.er in important respects, — there is substantial 313 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Fig. 1 variation in the reactions used in the synthetase domains of these enzymes. Structural characterization of the synthetase domains throughout Class I and Class II amidotransferases is of interest given the issue of whether structures for channeling ammonia from the glutaminase sites have evolved from a common precursor or many times.New insights into the structural basis for synthesis and activation of substrates for reaction with ammonia have been obtained in recent studies especially for NAD synthetase and glutamine fructose-6-phosphate amidotransferase (GFAT), which are both Class II amidotransferases. 314 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 13 NAD synthetase catalyzes the .nal step in NAD biosynthesis converting the deamido-NAD 35 into the .nal form of the co-factor 36. Mutants of Bacillus subtilis with severely a.ected cellular metabolism were shown to possess NAD synthetase with reduced activity. Evidence supporting a role for this enzyme in stress-response has also been reported. Interest in NAD synthetase has recently increased with the observation that there is a decrease in NAD in tubercule bacilli grown in the presence of isoniazid, opening a new avenue for developing drugs against strains of Mycobacterium tuberculosis that are multi-drug resistant. In this regard the gene encoding NAD synthetase in this organism has recently been cloned and expressed.In common with several other amidotransferases, — peptide synthetases, luciferase and tRNA synthetases, NAD synthetase catalyzes the initial formation of an NAD-adenylate 37 which is then attacked by either free ammonia as in Bacillus subtilis, or ammonia released from glutamine in an adjacent GAT-domain. The key chemical issue in employing such a strategy is to prevent hydrolysis of the adenylated intermediate before its reaction with ammonia thereby avoiding futile ATP breakdown.Insight into the mechanisms by which this problem is solved in Bacillus subtilis NAD synthetase has been provided by a recent crystal structure of the enzyme complexed to the NAD-adenylate intermediate 37. On the basis of structural homology NAD synthetase appears to be evolutionarily 315 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 related to GMP synthetase and phosphadenyl sulfate (PAPS) reductase. Since structures for the free enzyme and its complexwith ATP are available, conformational changes in the protein structure on formation of 37 are evident from this new structure. Cleavage of the ATP — bond appears to be mediated solely by two Mg ions that (i) increase the electrophilic character of the AMP and (ii) neutralize the negative charges on the pyrophosphate leaving group.No side chains from any of the protein residues are positioned so as to stabilize the transition state leading to the pentacovalent intermediate in the .rst step of NAD-adenylate formation. This catalytic strategy is in sharp contrast to the results of previous studies on Class II aminoacyl tRNA synthetases and asparagine synthetase in which evidence for the involvement of arginine side chains in catalyzing the formation of acyl-AMP intermediates has been obtained.— The crystal structure of the E·ATP·Tl complex obtained in the same study, has revealed an ammonium ion binding site adjacent to the carboxylate side chain of Asp-173.It is proposed that this site is occupied prior to formation of 37 allowing in situ delivery of nitrogen after deprotonation of the ammonium ion by Asp-173. As a result the nitrogen transfer reaction can take place in the presence of the aqueous environment. In the case of GFAT the synthetase active site must not only activate fructose-6- phosphate 38 to nucleophilic attack by ammonia released from glutamine in the N-terminal GAT-domain, by opening the furanose ring of the sugar phosphate to the corresponding linear form 39 but also subsequently act as an isomerase to convert the adduct 40 to glucosamine 6-phosphate 41 (Scheme 13). The crystal structures of a C-terminal domain proteolytic fragment of Escherichia coli GFAT which exhibits isomerase activity, complexed to 41 and a transition state analog 2-amino-2-glucitol-6-phosphate 42 have de.ned the catalytic role of key active site residues (Scheme 13).After ring-opening to yield the linear form of the substrate a step that is proposed to involve general acid/base catalysis by His-504 Lys-603 activates C-2 to attack by ammonia released in the GAT-domain of the enzyme. Independent evidence for the catalytic importance of Lys-603 was obtained in earlier site-directed mutagenesis studies. Nucleophilic attack on the Schi.s base intermediate then yields an imine from which the carboxylate of Glu-488 can abstract the C-1 proton to give the hydroxyenamine 43 an analog of the presumed enediolate intermediate in the reaction catalyzed by triose phosphate isomerase. Reprotonation at C-2 by Glu-488 then yields the linear form of 41 that can cyclize in a reaction that is also likely catalyzed by His-504.It has also been claimed that the structure of the C-terminal fragment is consistent with the existence of an ‘‘ammonia channel’’ linking the synthetase and glutaminase sites as observed in CPS and the ‘‘active’’ conformation of glutamine 5-phosphoribosylpyrophosphate amidotransferase (GPA). Direct sup- 316 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 14 port for this assertion awaits the determination of the structure of the intact enzyme by X-ray crystallography.There seems little doubt however that all glutamine-dependent amidotransferases will employ some form of channel structure for sequestering ammonia released from glutamine. The observation that indole another neutral small molecule intermediate is channeled between active sites in tryptophan synthase raises the question of whether this molecular solution is used more generally especially for passing reactive intermediates between di.erent enzymes. A particularly good example of such a situation is provided in fact by 5-phosphoribosylamine 44 which is the product of the reaction catalyzed by GPA and is the .rst intermediate in the biosynthetic pathway used for the de novo synthesis of purines. In the second step of the pathway 44 reacts with the acyl-phosphate 45 to yield glycinamide ribonucleotide 46 in a reaction catalyzed by the enzyme glycinamide ribonucleotide synthetase (GRS) (Scheme 14). In vitro kinetic studies have shown that 44 has a half-life of 5 seconds at 37 °C in aqueous solution suggesting that this reactive intermediate is passed directly from GPA to GRS in a multi-enzyme complex.E.orts to obtain kinetic evidence for the 317 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 15 existence of such a complex did not provide de.nitive support for this hypothesis. The three-dimensional structure of GRS therefore represents a major step in resolving this question. Molecular modeling studies indicate that GRS and the active conformation of GPA have complementary surfaces and hence transfer of 44 between the two enzymes via a substrate channel is structurally feasible.Kinetic studies which con.rm this hypothesis are eagerly awaited. 6 -Lactam synthesis and degradation The continued rise in bacterial resistance to -lactam antibiotics has stimulated 318 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 interest in delineating the biosynthetic pathways involved in the production of such compounds,— and the mechanisms of enzymes underlying the observed drug resistance particularly -lactamases. Clavulanic acid 49 continues to be a widely used -lactamase inhibitor and signi.cant e.orts have been made to de.ne the enzymes involved in its biosynthesis. As a result the central transformations in the pathway leading to clavulanate appear well understood.One surprising feature in the production of 49 is that clavaminate acid synthase (CAS) an -ketoglutarate-dependent (-KG) iron(..)-containing monooxygenase, catalyzes three separate transformations on the pathway each involving a di.erent substrate. Another intriguing observation and in contrast to the construction of the bicyclic system in penicillins, is that the synthesis of the -lactam moiety in 49 is not catalyzed by CAS. Instead the -lactam is formed by cyclization of 47 to 48 in an ATP-dependent reaction (Scheme 15). The details of the transformations that yield the acyclic precursor 47 remain to be established although arginine and pyruvate are almost certainly used in its construction.In mechanistic terms the chemical activation of 47 by ATP and subsequent intramolecular cyclization with the -amino group are identical to the reaction catalyzed by asparagine synthetase (AS). It was therefore predicted that the enzyme responsible for mediating the conversion of 47 to 48 -lactam synthetase (BLS) would be homologous to AS assuming that chemistry is the principal determinant of active site structure. Sequence analysis of the gene cluster encoding the enzymes involved in clavulanate biosynthesis suggested that one open reading frame ORF3 displayed approximately 25% identity over 434 of 513 residues including a motif associated with pyrophosphatase release from ATP. Cloning and expression of the enzyme encoded by the ORF gene in Saccharomyces clavuligerus then yielded pure BLS which was demonstrated to catalyze the ATP-dependent cyclization of 47 in the presence of Mg using an HPLC-based assay. A subsequent report con.rmed this assignment of enzyme function and made the intriguing mechanistic speculation that the -lactam 50 might be an intermediate in AS-catalyzed asparagine synthesis. In the absence of structural data on AS and BLS however the evolutionary relationship between the two enzymes remains to be established.319 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 16 The emergence of plasmid encoded metallo-lactamases in some bacteria has conferred simultaneous resistance to available -lactam antibiotics on these pathogens.Crystal structures,¡X in combination with metal binding and other spectroscopic studies have indicated that the active site contains two Zn() ions bridged by a deprotonated water molecule. It has been proposed that the metals perform two roles in catalysis. First coordination of the carbonyl group of the -lactam to Zn() activates the substrate to nucleophilic attack by water. Second the metals facilitate deprotonation of the water. Recent work employing nitrocen 51 has allowed the direct spectroscopic observation of an intermediate in the -lactamase catalyzed hydrolysis reaction the properties of which are consistent with the novel acyl-enzyme intermediate 52. Breakdown of 52 is rate-limiting with this substrate the anionic intermediate being stabilized to an unusual degree.In addition to extensive charge delocalization throughout the conjugated double bond system the histidinerich Zn() ion likely stabilizes the intermediate electrostatically together with residues from the substrate-binding pocket such as Lys-171 and Trp-36. The direct interaction of the second active site Zn() with the nitrogen anion excludes an apical water ligand eliminating a potential proton source from the local environment. 7 Saccharide biosynthesis Enzymes that utilize phosphoenolpyruvate continue to be of medicinal and mechanistic interest. These systems are also of mechanistic interest because reactions involving this high-energy metabolite occur through the chemically unusual scission of the C¡XO bond to give phosphate.Two types of pathways are known for PEP-utilizing enzymes in which (a) substitution occurs to give an enol ether or (b) condensation with an electrophile such as an aldehyde takes place with concomitant hydrolysis to give a keto-acid product (Scheme 16). Reactions proceeding by the second pathway are observed for DAHP synthase a key enzyme in the shikimate pathway and Kdo8P synthase which catalyzes the formation of the complexsaccharide 3-deoxy- manno-2-octulosonate-8-phosphate 53 (Kdo8P) (Scheme 17). This eight-carbon sugar is an important constituent of the cell wall in most gram negative bacteria and therefore represents an important target for the development of novel antibac- 320 Annu. Rep. Prog. Chem. Sect. B 1999 95 299¡X334Scheme 17 terial agents. Two mechanisms can be envisaged for the catalytic mechanism of Kdo8P synthase. The .rst of these involves the formation of a carbocationic intermediate that is trapped by water to form 54 (Scheme 17A).Elimination of phosphate from 54 then yields an -ketoacid which can cyclize to give the product (Scheme 17A). This mechanism is therefore analogous to that determined for EPSP synthase now that an alternative proposal (discussed in the previous review in this series) has been invalidated by new solid-state NMR experiments. The second possible catalytic mechanism for Kdo8P formation (Scheme 17B) involves a concerted step in which a cyclic intermediate is formed in the .rst step from which phosphate is removed through nucleophilic attack by a water molecule.Transient kinetic experiments using P- or C-radiolabeled aldehyde demonstrated (i) that the reaction is not reversible and (ii) the absence of covalent intermediates in which substrates become attached to 321 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Kdo8P synthase. No evidence was obtained to support the existence of cyclic intermediate 55 in the reaction mechanism an authentic sample of which was only a modest inhibitor of Kdo8P synthase activity. While acyclic bisphosphate 54 is most likely a reaction intermediate as proposed in early studies, the inherent instability of the hemiacetal moiety in 54 probably precludes its isolation and/or chemical synthesis. Indirect evidence for this hypothesis is provided however by the observation that acyclic tertiary amine 56 is a potent Kdo8P synthase inhibitor. Thus the intrinsic reactivity of PEP remains unchanged in the two mechanistic pathways employed for C—O bond cleavage by both classes of enzymes that employ this substrate.Branched chain sugars are an important class of carbohydrates many of which are present in molecules involved in cell signalling and antibiotics. The biosynthesis of these materials is thought to involve the modi.cation of simple sugar derivatives. Yersiniose 57 is found in the O-antigen of Yersinia pseudotuberculosis. Although simple methyl substituents are derived from S-adenosylmethionine pyruvate has been proposed to be the origin of the two-carbon unit present in 57. Early experiments employing cell-free extracts of the organism however failed to yield evidence supporting this hypothesis. Sequence analysis of the gene cluster encoding enzymes in the biosynthetic pathway leading to 57 has now resolved this issue.Thus the deduced amino acid sequence of the enzyme encoded by the YerE gene appeared homologous to acetohydroxyacid synthase (AHAS) a TPP-dependent enzyme catalyzing the condensation of two pyruvate molecules in the initial step of branched chain amino acid biosynthesis. AHAS is a .avoprotein in which the bound .avin is catalytically inactive apparently playing a structural role as a vestigial co-factor due to the evolution of the enzyme from pyruvate oxidase. Expression puri.cation and characterization of the YerE gene product con.rmed the ability of the enzyme to catalyze the addition of pyruvate to the ketosugar 58 using TPP as a co-factor (Scheme 18).In accordance with the predictions from sequence analysis the recombinant enzyme possesses a bound .avin that is not employed in catalysis. 322 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 18 8 Glycosyl hydrolases Glycosyl hydrolases have functions that range from modulating viral invasion and the control of cell—cell interactions to the simple hydrolysis of polysaccharides. Structural diversity in substrates for these enzymes is mirrored by the number of sequencedistinct families of glycosyl hydrolase. Their reaction mechanisms have been the focus of numerous studies given the potential of inhibitors in the treatment of viral infections and their usefulness as model systems for elucidating the role of general acids and bases in enzyme catalysis. In particular studies on -galactosidase which hydrolyzes a wide range of -galactopyranosyl derivatives 59 via a covalent intermediate in a two-step mechanism, have provided new data on the hypothesis that enzymatic Brønsted acid/base catalysis must be enhanced relative to that in aqueous solution. New structural insights into the conformational changes in the enzyme that take place during glycosyl hydrolysis and the relocation of key catalytic side chains have recently been provided in an elegant study of all of the stable states of Bacillus 323 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 19 agaradhaerens Cel5A along the reaction coordinate. Cel5A catalyzed glycoside hydrolysis proceeds with identical stereochemistry as observed for the reaction mediated by -galactosidase and the active site structures of the two enzymes are similar, suggesting an identical catalytic mechanism (Scheme 19).In addition to that of the free enzyme structures are reported for the Michaelis complexwith unhydrolyzed substrate 60 a catalytically competent glycosyl-enzyme intermediate 61 and the enzyme—product complex 62. These structures yield insight not only into the spatial location of groups involved in acid/base catalysis and stabilization of positively charged transition states but also into the use of binding energy in distorting substrate structure as a mechanism for lowering activation energy barriers in the reaction.9 Nucleotide hydrolysis and repair Enzymes catalyzing the speci.c cleavage and/or repair of phosphoester linkages in DNA and RNA have important cellular functions — and are of chemical interest for the development of synthetic catalysts for phosphoester hydrolysis. Topoisomerases are a particularly important class of DNA-manipulating enzyme, that solve the topological problems that are associated with DNA replication recombination and chromosome segregation. These enzymes are important clinical targets of several antibacterial and anti-cancer agents including camptothecin 63. An important step in understanding the mechanism by which these enzymes create reversible scissions in DNA using a conserved tyrosine residue, has been the recent 324 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 20 determination of the structures of Y723F human topoisomerase I mutant in covalent and non-covalent complexes with DNA. This work has therefore a.orded snapshots of the active site region before and afterDNAcleavage. Molecular modeling indicates that the tyrosine oxygen is positioned for nucleophilic attack on the phosphodiester group and that the pentacovalent intermediate 64 is probably stabilized by two proximal arginine residues. P—O bond cleavage is then promoted by general acid catalysis involving the His-632 side chain to yield the covalently modi.ed enzyme 65 325 Annu. Rep. Prog.Chem. Sect. B 1999 95 299—334 (Scheme 20). One interesting feature of the crystal structure is the lack of protein residues that might act to facilitate nucleophilic attack by removing the proton from the Tyr-723 hydroxy group. It is possible that this role is played by an active site water molecule but the presence of the phenylalanine residue in place of Tyr-523 may cause substantial conformational rearrangement in the active site. The general base therefore remains to be identi.ed. Reversing the mechanism of phosphoester cleavage repairs the strand scission after topological rearrangement of the DNA. The molecular mechanism by which the DNA is rotated before product release remains ill-de.ned. DNA-photolyases represent another class of important DNA-manipulating enzymes which repair pyrimidine dimer lesions such as 66 that are formed on irradiation of double-strandedDNAwith UV light. Repair of the lesion requires electron Scheme 21 326 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 transfer from a reduced deprotonated .avin co-factor and an 8-hydroxy-5-deaza.avin as a second co-factor that functions within the photolyase as a photoantenna transferring energy to the reduced .avin. Unexpectedly the separation of the two cofactors in the enzyme is approximately 17Å raising questions concerning the mechanism by which energy transfer takes place. The interesting chemical models 67 and 68 have been synthesized in an e.ort to study interactions between these .avin co-factors as a function of co-factor redoxand protonation state. Unfortunately an alternate mechanism of electron transfer interfered with the cleavage reaction in these model compounds presumably due to lack of control in the separation of the 327 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 Scheme 22 .avin and deaza.avin moieties. Preventing such side-reactions may account for the large separation of these components in the DNA-photolyase structure. 10 Mechanism-based inhibitors Isoniazid 69 is one of the most widely employed and e.ective drugs for treating tuberculosis, but its mechanism of action and enzyme target were until recently unknown. Due to increasing isoniazid resistance exhibited by strains of Mycobacterium tuberculosis the molecular mechanism by which 69 exerts its e.ects has been delineated as an aid in developing new therapeutic agents. The biological target of 69 is an enoyl-acyl carrier protein reductase (InhA) that catalyzes the NADH-dependent reduction of unsaturated precursors of mycolic acids such as 70 and 71 which are C —C long chain -branched fatty acid components of the mycobacterial cell wall. Isoniazid 69 does not interact directly with InhA however but must be oxidized by a catalase-peroxidase katG before inhibiting the enzyme. Recent crystallographic studies have now established that the activation process yields isonicotinic acyl-NADH 73 that is bound tightly within the co-factor binding site of 328 Annu.Rep. Prog. Chem. Sect. B 1999 95 299—334 InhA. The potent inhibitor 73 is probably formed by reaction of isonicotinic acyl radical 72 obtained in katG-catalyzed isoniazid oxidation, and the NAD radical (Scheme 21).An ionic pathway involving modi.cation of NAD appears less likely as isoniazid-dependent InhA inhibition occurs fastest in the presence of NADH. Cycloserine 74 is a broad-spectrum antibiotic that has also been used in the treatment of tuberculosis and has been shown to inhibit pyridoxal phosphate (PLP) dependent enzymes such as -aminobutyric acid (GABA) aminotransferase and .-amino acid aminotransferase (.-aAT). The latter enzyme is involved in formation of the .-alanine and .-glutamate components required for cell wall biosynthesis. A number of mechanisms have been proposed for the process by which cycloserine inactivates PLP-dependent enzymes in all of which the initial adduct between 74 and PLP is transformed in the active site to an electrophilic species capable of covalently modifying the enzyme.— Recent studies on GABA aminotransferase employing radiolabeled cycloserine and isolation of radioactive inactivated adducts revealed an inhibition mechanism in which enzyme-catalyzed aromatization takes place to give a covalently modi.ed PLP co-factor 75 that is tightly bound within the active site (Scheme 22). The crystal structure of the cycloserine-inactivated form of .-aAT con.rms this proposal and clearly shows the key non-covalent interactions between 75 and the enzyme. Delineation of the unexpected mechanism underlying cycloserine inactivation coupled with the structural observations should aid the development of selective .-aAT inhibitors as new antibiotics.11 Concluding remarks Although recent developments in enzyme mechanisms have been reviewed here progress in characterizing the structure and mechanism of catalysts (ribozymes) constructed from nucleic acids continues to be rapid. This work will be discussed in the next review in this series. Perhaps the greatest surprise is the range of chemical reactions that can be catalyzed by ribozymes despite the more restricted nature of the functional groups present in this class of macromolecular structure. Identi.cation of similarities and di.erences in the chemical mechanisms employed by these two classes of biological catalyst is yielding new insight into the chemical principles of catalysis. References 1 N.G.J. Richards Annu. Rep. Prog. Chem. Sect. B Org. Chem. 1998 94 289. 2 W.W. Cleland P. A. Frey and J. A. Gerlt J. Biol. Chem. 1998 273 25 529. 3 W.R. Cannon and S. J. Benkovic J. Biol. Chem. 1998 273 26 257. 4 A. Warshel J. Biol. Chem. 1998 273 27035. 5 Y. Pan and M. A. McAllister J. Am. Chem. Soc. 1998 120 166. 6 Y. Pan and M. A. McAllister J. Org. Chem. 1997 62 8171. 7 A. Warshel and A. Papazyan Proc. Natl. Acad. Sci. USA 1996 93 13 665. 8 J.A. Gerlt and P. G. Gassman Biochemistry 1993 32 11 943. 9 A. Warshel A. Papazyan and P. A. Kollman Science 1996 269 102. 10 T. K. Harris and A. S. Mildvan Proteins Struct. Funct. Genet.1999 35 275. 11 P. Kuhn M. Knapp S.M. Soltis G. Ganshaw M. Thoene and R. Bott Biochemistry 1998 37 13 446. 12 P. A. Frey S. Whitt and J. Tobin Science 1994 264 1927. 13 R. A. Marcus and N. Sutin Biochim. Biophys. Acta 1985 811 265. 14 J. A. Stubbe Annu. Rev. Biochem. 1989 58 257. 329 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 15 B. J. Bahnson T. D. Colby K. J. Chin B. M. Goldstein and J. P. Klinman Proc. Natl. Acad. Sci. USA 1997 94 12 797. 16 J. P. Klinman Chem. Rev. 1996 96 2541. 17 M.H. Glickman and J. P. Klinman Biochemistry 1996 35 12 882. 18 A. Kohen and J. P. Klinman Acc. Chem. Res. 1998 31 397. 19 W.J. Bruno and W. Bialek Biophys. J. 1992 63 689. 20 D. Antoniou and S. D. Schwartz Proc. Natl. Acad. Sci. USA 1997 94 12 360. 21 M.M. Palcic and J.P. Klinman Biochemistry 1983 22 5957. 22 J. K. Hwang Z. T. Chu A. Yadav and A. Warshel J. Phys. Chem. 1991 95 8445. 23 J. K. Hwang and A. Warshel J. Am. Chem. Soc. 1996 118 11 745. 24 M.E. Rasche and R. H. White Biochemistry 1998 37 11 343. 25 R. H. White Biochemistry 1996 35 3447. 26 A. A. DiMarco T. A. Bobik and R. S. Wolfe Annu. Rev. Biochem. 1990 59 355. 27 W.D. L. Musick CRC Crit. Rev. Biochem. 1981 11 1. 28 J. L. Smith and H. Zalkin Adv. Enzymol. Relat. Areas Mol. Biol. 1998 72 87. 29 G. Scapin D. H. Ozturk C. Grubmeyer and J. C. Sacchettini Biochemistry 1995 34 10 744. 30 S. Danishefsky C.-F. Yan K. S. Rajendra R. B. Gammill P. M. McCurry N. Fritsch and J. Clardy J. Am. Chem. Soc. 1979 101 7001. 31 J. E. Cronan J. Bacteriol. 1980 141 1291.32 P. D. van Poelje and E. E. Snell Annu. Rev. Biochem. 1990 59 29. 33 M.K. Ramjee U. Genschel C. Abell and A. G. Smith Biochem. J. 1997 323 661. 34 Y. Shao M.-Q. Xu and H. Paulus Biochemistry 1996 35 3810. 35 M.-Q. Xu and F. B. Perler EMBO J. 1996 15 5146. 36 A. Albert V. Dhanaraj U. Genschel G. Khan M.K. Ramjee R. Pulido B. L. Sibanda F. von Delft M. Witty T. L. Blundell A. G. Smith and C. Abell Nat. Struct. Biol. 1998 5 289. 37 K. Iwai and T. Endo Methods Enzymol. 1967 11 262. 38 C. A. Costello N. L. Kelleher M. Abe F. W. McLa.erty and T. P. Begley J. Biol. Chem. 1996 271 3445. 39 G. Lienhard Biochemistry 1970 9 3011. 40 R. Nicewonger A. Rammelsberg C. A. Costello and T. P. Begley Bioorg. Chem. 1995 23 512. 41 R. Nicewonger C. A. Costello and T.P. Begley J. Org. Chem. 1996 61 4172. 42 J. A. Hutter and J. T. Slama Biochemistry 1987 26 1969. 43 N. Campobosso C. A. Costello C. Kinsland T. P. Begley and S. E. Ealick Biochemistry 1998 37 15 981. 44 J. Stubbe and W. A. van der Donk Chem. Rev. 1998 98 705. 45 J. Knappe and G. Sawers FEMS Microbiol. Rev. 1990 75 383. 46 J. Knappe H. P. Blaschkowski P. Grobner and T. Schmitt Eur. J. Biochem. 1974 50 253. 47 H. Conradt M. Hohmann-Berger H. Hohmann H. P. Blaschkowski and J. Knappe Arch. Biochem. Biophys. 1984 228 133. 48 V. Unkrig F. A. Neugebauer and J. Knappe Eur. J. Biochem. 1989 184 723. 49 J. E. Baldwin D. Brown P. H. Scudder and M. E. Wood Tetrahedron Lett. 1995 36 2105. 50 W. Plaga R. Frank and J. Knappe Eur. J. Biochem. 1988 178 445. 51 C.V. Parast K. K. Wong S. A. Lewisch J. W. Kozarich J. Peisach and R. S. Magliozzo Biochemistry 1995 34 2393. 52 E. J. Brush K. A. Lipsett and J. W. Kozarich Biochemistry 1988 27 2217. 53 J. Knappe S. Elbert M. Frey and A. F. V. Wagner Biochem. Soc. Trans. 1993 21 731. 54 G. Heinisch and G. Lotsch Angew. Chem. Int. Ed. Engl. 1985 24 692. 55 G. Heinisch and G. Lotsch Tetrahedron Lett. 1985 41 1199. 56 J. Knappe F. A. Neugebauer H. P. Blaschkowski and M. Ga� nzler Proc. Natl. Acad. Sci. USA 1984 81 1332. 57 S. G. Reddy K. K. Wong C. V. Parast J. Peisach R. S. Magliozzo and J. W. Kozarich Biochemistry 1998 37 558. 58 A. F. V. Wagner M. Frey F. A. Neugebauer W. Schafer and J. Knappe Proc. Natl. Acad. Sci. USA 1992 89 996. 59 M.M. Whittaker D. P. Ballou and J.W. Whittaker Biophys. J. 1993 64 762. 60 M.M. Whittaker and J. W. Whittaker J. Biol. Chem. 1988 263 6074. 61 M.M. Whittaker and J. W. Whittaker J. Biol. Chem. 1990 265 9610. 62 N. Ito S. E. V. Phillips K. D. S. Yadav and P. F. Knowles J. Mol. Biol. 1994 238 794. 63 N. Ito S. E. V. Phillips C. Stevens Z. B. Ogel M. J. McPherson J. N. Keen K. D. S. Yadav and P. F. Knowles Nature 1991 350 87. 64 J. W. Whittaker Metal Ions in Biological Systems ed. H. Sigel and A. Sigel Marcel Dekker New York 1994 Vol. 30 p. 315. 65 A. J. Baron C. Stevens C. Wilmot K. D. Seneviratne V. Blakeley D.M. Dooley S. E. V. Phillips P. F. Knowles and M. J. McPherson J. Biol. Chem. 1994 269 25 095. 66 M.M. Whittaker D. P. Ballou and J. W. Whittaker Biochemistry 1998 37 8426.67 Y. Cha C. J. Murray and J. P. Klinman Science 1989 243 1325. 330 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 68 G. A. Hamilton P. K. Adolf J. de Jersey G. C. DuBois G. R. Dyrkacz and R. D. Libby J. Am. Chem. Soc. 1978 100 1899. 69 C. D. Borman C. G. Saysell and A. G. Sykes J. Biol. Inorg. Chem. 1997 2 480. 70 J. Stubbe J. Biol. Chem. 1990 265 5329. 71 J. Stubbe Adv. Enzymol. Relat. Areas Mol. Biol. 1990 63 349. 72 P. Reichard Science 1993 260 1773. 73 J. Stubbe and W. van der Donk Chem. Biol. 1995 2 793. 74 J. P. Klinman and D. Mu Annu. Rev. Biochem. 1994 63 299. 75 J. P. Klinman Chem. Rev. 1996 96 2541. 76 M. R. Parsons M. A. Convery C. M. Wilmot K. D. S. Yadav V. Blakeley A. S. Corner S. E. V. Phillips M. J. McPherson and P. F. Knowles Structure 1995 3 1171.77 J. Westerling J. Frank and J. A. Duine Biochem. Biophys. Res. Commun. 1979 87 719. 78 S. A. Salisbury H. S. Forrest W. B. T. Cruse and O. Kennard Nature 1979 280 843. 79 S. X. Wang M. Mure KF. Medzihradszky A. L. Burlingame D. E. Brown D.M. Dooley A. J. Smith 80 C. H. Baker J. Banzon J. M. Bollinger Jr. J. Stubbe V. Samano M. J. Robins B. Lippert E. Jarvi and R. 81 G. J. Gerfen W. A. van der Donk G. Yu J. R. McCarthy E. T. Jarvi D. P. Matthews C. Farrar R. G. 82 A. J. Bitonti J. A. Dumont T. L. Bush E. A. Cashman D. E. Cross-Doersen P. S. Wright D. P. Matthews H.M. Kagan and J. P. Klinman Science 1996 273 1078. Resvick J. Med. Chem. 1991 34 1879. Gri.n and J. Stubbe J. Am. Chem. Soc. 1998 120 3823. J. R. McCarthy and D. A.Kaplan Cancer Res. 1994 54 1485. 83 W. A. van der Donk G. Yu D. J. Silva J. Stubbe J. R. McCarthy E. T. Jarvi D. P. Matthews R. J. Resvick and E. Wagner Biochemistry 1996 35 8381. 84 S. L. Licht G. J. Gerfen and J. Stubbe Science 1996 271 477. 85 P. H. S. Reynolds M. J. Boland D. G. Blevins D. D. Randall and K. R. Schubert Trends Biochem. Sci. 1982 7 366. 86 B. Frei L. England and B. N. Ames Proc. Natl. Acad. Sci. USA 1989 86 6377. 87 S. Rozenberg B. Roche A. C. Koeger C. Borget N. Wrona and P. Bourgeois Rev. Rhum Eng. Ed. 1995 62 392. 88 K. Kahn and P. A. Tipton Biochemistry 1998 37 11 651. 89 E. S. Canellakis and P. P. Cohen J. Biol. Chem. 1955 213 385. 90 L. Sokolic N. Modric and M. Poje Tetrahedron Lett. 1991 32 7477. 91 N. Modric A. E. Derome S.J. H. Ashcroft and M. Poje Tetrahedron Lett. 1992 33 6691. 92 K. Kahn P. Serfozo and P. A. Tipton J. Am. Chem. Soc. 1997 119 5435. 93 K. Kahn and P. A. Tipton Biochemistry 1997 36 4731. 94 N. Colloc’h M. E. Hajji B. Bachet G. L’Hermite M. Schlitz T. Prange� B. Castro and J.-P. Mornon Nat. Struct. Biol. 1997 4 947. 95 H. Zalkin and J. L. Smith Adv. Enzymol. Relat. Areas Mol. Biol. 1998 72 87. 96 H.M. Holden J. B. Thoden and F. M. Raushel Curr. Opin. Struct. Biol. 1998 8 679. 97 J. L. Smith Curr. Opin. Struct. Biol. 1998 8 686. 98 N. G. J. Richards and S. M. Schuster Adv. Enzymol. Relat. Areas Mol. Biol. 1998 72 145. 99 A. Meister Adv. Enzymol. Relat. Areas Mol. Biol. 1989 62 315. 100 J. Piette H. Nyunoya C. J. Lusty R. Cunin G. Weyens M. Crabeel D.Charlier N. Glansdor. and A. Pie� rard Proc. Natl. Acad. Sci. USA 1984 81 4134. 101 H. Zalkin Adv. Enzymol. Relat. Areas Mol. Biol. 1993 66 203. 102 L. M. Pinkus and A. Meister J. Biol. Chem. 1972 247 6119. 103 S. D. Rubino H. Nyunoya and C. J. Lusty J. Biol. Chem. 1986 261 11 320. 104 J. B. Thoden H. M. Holden G. Wesenberg F. M. Raushel and I. Rayment Biochemistry 1997 36 6305. 105 D. L. Ollis E. Cheah M. Cygler B. Dijkstra F. Frolow S. M. Franken M. Harel S. J. Remington I. Silman J. Schrag J.L Sussman K. H. G. Verschuren and A. Goldman Protein Eng. 1992 5 197. 106 A.C Storer and R. Me� nard Methods Enzymol. 1994 244 486. 107 C. J. Lusty FEBS Lett. 1992 314 135. 108 J. B. Thoden S. G. Miran J. C. Phillips A. J. Howard F. M. Raushel and H.M. Holden Biochemistry 1998 37 8825.109 S. G. Miran S. H. Chang and F. M. Raushel Biochemistry 1991 30 7901. 110 R. L. Spencer and J. Preiss J. Biol. Chem. 1967 242 385. 111 F. Massie` re and M.-A. Badet-Denisot Cell Mol. Life Sci. 1998 54 205. 112 C. Nessi A. M. Albertini M.L. Speranza and A. Galizzi J. Biol. Chem. 1995 270 6181. 113 H. Antelmann R. Schmid and M. Hecker FEMS Microbiol. Lett. 1997 153 405. 114 A. Belierkunst Science 1966 152 525. 115 B. R. Bloom and C. J. Murray Science 1992 257 1055. 116 R. Cantoni M. Branzoni M. Labo` M. Rizzi and G. Riccardi J. Bacteriol. 1998 180 3218. 117 S. K. Boehlein J. D. Stewart E. S. Walworth R. Thirumoorthy N. G. J. Richards and S. M. Schuster Biochemistry 1998 37 13 230. 331 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 118 N.Nakatsu H. Kato and J. Oda Nat. Struct. Biol. 1998 5 15. 119 J. J. G. Tesmer M.L. Deras V. J. Davisson and J. L. Smith Nat. Struct. Biol. 1996 3 74. 120 A.W. Curnow K.W. Hong R. Yuan S. Kim. O. Martens W. Winkler T. M. Henkin and D. So� ll Proc. Natl. Acad. Sci. USA 1997 94 11 819. 121 E. Conti T. Stachelhaus M. A. Marahiel and P. Brick EMBO J. 1997 16 4174. 122 E. Conti N. P. Franks and P. Brick Structure 1996 4 287. 123 M. Delarue Curr. Opin. Struct. Biol. 1995 5 48. 124 C. R. Zerez M.D. Wong and K. R. Tanaka Blood 1990 75 1576. 125 R. Wagner and K. G. Wagner Planta 1985 165 532. 126 M. Rizzi M. Bolognesi and A. Coda Structure 1998 6 1129. 127 M. Rizzi C. Nessi A. Mattevi A. Coda M. Bolognesi and A. Galizzi EMBO J. 1996 15 5125. 128 H.Savage G. Montoya C. Svensson J. D. Schwenn and I. Sinning Structure 1997 5 895. 129 J. G. Arnez J. G. Augustine D. Moras and C. S. Francklin Proc. Natl. Acad. Sci. USA 1997 94 7144. 130 H. Behrhali A. Yaremchuk M. Tukalo C. Bethetcolominas B. Rasmussen P. Bosecke O. Diat and S. Cusack Structure 1995 3 341. 131 S. K. Boehlein E. S. Walworth N. G. J. Richards and S. M. Schuster J. Biol. Chem. 1997 272 12 384. 132 M. N. Isupov G. Obmolova S. Butterworth M.-A. Badet-Denisot B. Badet I. Polikarpov J. A. Littlechild and A. Teplyakov Structure 1996 4 801. 133 M.-A. Badet-Denisot L. Rene� and B. Badet Bull. Soc. Chim. Fr. 1993 130 249. 134 C. Leriche M.-A. Badet-Denisot and B. Badet J. Am. Chem. Soc. 1996 118 1797. 135 A. Teplyakov G. Obmolova M.-A. Badet-Denisot B.Badet and I. Polikarpov Structure 1998 6 1047. 136 A. Teplyakov G. Obmolova M.-A. Badet-Denisot and B. Badet Protein Sci. 1999 8 596. 137 J. R. Knowles Nature 1991 350 121. 138 E. B. Nickbarg R. C. Davenport G. A. Petsko and J. R. Knowles Biochemistry 1988 27 5948. 139 J. M. Krahn J. H. Kim M.R. Burns R. Parry H. Zalkin and J. L. Smith Biochemistry 1997 36 11 061. 140 M. F. Dunn V. Aguilar P. Brzovic F. W. Drewe Jr. K. F. Houben C. A. Leja and M. Roy Biochemistry 1990 29 8598. 141 E. W. Miles S. Rhee and D. R. Davies J. Biol. Chem. 1999 274 12 193. 142 J. Ovadi J. Theor. Biol. 1991 152 1. 143 J. L. Smith Biochem. Soc. Trans. 1995 23 894. 144 Y. S. Cheng J. Rudolph M. Stern J. Stubbe K. A. Flannigan and J.J. Smith Biochemistry 1990 29 218. 145 J.Rudolph and J. Stubbe Biochemistry 1995 34 2241. 146 W. R. Wang T. J. Kappock J. Stubbe and S. E. Ealick Biochemistry 1998 37 15 647. 147 D. Konz and M. A. Marahiel Chem. Biol. 1999 9 R39. 148 S. Donadio M.J. Staver J. B. McAlpine S. J. Swanson and L. Katz Science 1991 252 675. 149 J. F. Martin and P. Liras Annu. Rev. Microbiol. 1989 43 173. 150 J. Handelsman M.R. Rondon S. F. Brady J. Clardy and R.M. Goodman Chem. Biol. 1998 5 R245. 151 M. B. Schmid Curr. Opin. Chem. Biol. 1998 2 529. 152 A. Matagne A. Dubus M. Galleni and J.-M. Fre` re Nat. Prod. Rep. 1999 16 1. 153 K. H. Baggaley A. G. Brown and C. J. Scho.eld Nat. Prod. Rep. 1997 14 309. 154 R.W. Busby M.D.-T. Chang R. C. Busby J. Wimp and C. A. Townsend J. Biol. Chem. 1995 270 4262. 155 J. E.Baldwin and E. Abraham Nat. Prod. Rep. 1998 5 129. 156 J. E. Thirkettle J. E. Baldwin J. Edwards J. P. Gri.n and C. J. Scho.eld Chem. Commun. 1997 1025. 157 J. A. Gerlt and P. C. Babbitt Curr. Opin. Chem. Biol. 1998 2 607. 158 P. C. Babbitt and J. A. Gerlt J. Biol. Chem. 1997 272 30 591. 159 B. O. Bachmann R. Li and C. A. Townsend Proc. Natl. Acad. Sci. USA 1998 95 9082. 160 H. J. McNaughton J. E. Thirkettle Z. Zhang C. J. Scho.eld S. E. Jensen B. Barton and P. Greaves Chem. Commun. 1998 2325. 161 P. Bork and E. V. Koonin Proteins Struct. Funct. Genet. 1994 20 347. 162 P. Bork and E. V. Koonin Curr. Opin. Struct. Biol. 1996 6 366. 163 D. J. Payne J. Med. Microbiol. 1993 39 93. 164 N. O. Concha B. A. Rasmussen K. Bush and O. Herzberg Structure 1996 4 823.165 N. O. Concha B. A. Rasmussen K. Bush and O. Herzberg Protein Sci. 1997 6 2671. 166 A. Car. E. Due� e R. Paul-Soto M. Galleni J.-M. Fre` re and O. Deideberg Acta Crystallogr. Sect. D 1998 54 47. 167 M. W. Crowder Z. Wang S. L. Franklin C. P. Zovinka and S. J. Benkovic Biochemistry 1996 35 12 126. 168 Z. Wang and S. J. Benkovic J. Biol. Chem. 1998 273 22 402. 169 C. H. O’Callaghan A. Morris S. M. Kirby and A. H. Shingler Antimicrob. Agents Chemother. 1972 1 283. 170 Z. Wang W. Fast and S. J. Benkovic J. Am. Chem. Soc. 1998 120 10 788. 171 D.W. Christiansen and C. A. Fierke Acc. Chem. Res. 1996 29 331. 172 J. H. Toney P. M.D. Fitzgerald N. Grover-Sharma W.J. May J. G. Sundelof D. E. Vanderwall K. A. Cleary S. K. Grant J. K. Wu J. W. Kozarich D.L. Pompliano and G. G. Hammond Chem. Biol. 1998 5 185. 173 C. T. Walsh T. E. Benson D. H. Kim and W. J. Lees Curr. Biol. 1996 3 1074. 332 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 174 I. A. Shumilin R. H. Kretsinger and R. Bauerle Proteins Struct. Funct. Genet. 1996 24 404. 175 R. Bentley CRC Crit. Rev. Biochem. Mol. Biol. 1990 25 307. 176 G. D. Dotson R. Dua J. Clemens E. M. D. Wooten and R. W. Woodard J. Biol. Chem. 1995 270 13 698. 177 Bacterial Lipopolysaccharides ACS Symp Ser. 231 ed. L. Anderson and F.M. Unger American Chemical Society Washington DC 1983. 178 P.-H. Liang J. Lewis K. S. Anderson A. Kohen F. Wallace D’Souza Y. Benenson and T. Baasov Biochemistry 1998 37 16 390. 179 K. S. Anderson and K. A. Johnson Chem. Rev. 1990 90 1131.180 D. R. Studelska L. M. McDowell M.P. Espe C. A. Klug and J. Schaefer Biochemistry 1997 36 15 555. 181 D. L. Jakeman D. J. Mitchell W.A. Shuttleworth and J. N. S. Evans Biochemistry 1998 37 12 012. 182 L. Hedstrom and R. H. Abeles Biochem. Biophys. Res. Commun. 1988 157 816. 183 S. Du H. Tsipori and T. Baasov Bioorg. Med. Chem. Lett. 1997 7 2469. 184 A. C. Weymouth-Wilson Nat. Prod. Rep. 1997 14 99. 185 D. A. Johnson and H.-W. Liu Curr. Opin. Chem. Biol. 1998 2 642. 186 R. Schmid H. Grisebach and W. Karl Eur. J. Biochem. 1970 14 243. 187 U. Matern and H. Grisebach Z. Naturforsch. 1974 29 407. 188 H. Chen Z. Guo and H.-W. Liu J. Am. Chem. Soc. 1998 120 11 796. 189 D. Chipman Z. Barak and J. V. Schloss Biophys. Biochim. Acta 1998 1385 401. 190 K.H. Ott J. G. Kwagh G. W. Stockton V. Sidorov and G. Kakefuda J. Mol. Biol. 1996 263 359. 191 Y. Y. Chang and J. E. Cronan J. Bacteriol. 1988 170 3937. 192 G. Davies and B. Henrissat Structure 1995 3 853. 193 B. Henrissat and G. J. Davies Curr. Opin. Struct. Biol. 1997 7 637. 194 G. Davies M. L. Sinnott and S. G. Withers Comprehensive Biological Catalysis ed. M. L. Sinnott Academic Press London 1997 p. 119. 195 G. S. Jacob Curr. Opin. Struct. Biol. 1995 5 605. 196 J. P. Richard Biochemistry 1998 37 4305. 197 M. L. Sinnott Chem. Rev. 1990 90 1171. 198 S. G. Withers I. P. Street P. Bird and D. H. Dolphin J. Am. Chem. Soc. 1987 109 7530. 199 J. P. Richard R. E. Huber S. Lin C. Heo and T. L. Amyes Biochemistry 1996 35 12 377. 200 G. J. Davies L. Mackenzie A.Varrot M. Dauter A. M. Brzozowski M. Schu� lein and S. G. Withers Biochemistry 1998 37 11 707. 201 R. H. Jacobson X.-J. Zhang R. F. Dubose and B. W. Matthews Nature 1994 369 761. 202 G. J. Davies M. Dauter A. M. Brzozowski M. E. Bjornvad K. V. Anderson and M. Schu� lein Biochemistry 1998 37 1926. 203 J. C. Wang Annu. Rev. Biochem. 1996 65 635. 204 G. Slupphaug C. D. Mol B. Kavli A. S. Arvai H. E. Kroken and J. A. Tainer Nature 1996 384 87. 205 F. Guo D. N. Gopaul and G. D. van Duyne Nature 1997 389 40. 206 A. G. Bodnar M. Oulette M. Folkis S. E. Holt C.-P. Chiu G. D. Morin C. B. Harley J. W. Shay S. Lichtsteiner and W.E. Wright Science 1998 279 349. 207 M. O’Reilly S. A. Teichmann and D. Rhodes Curr. Opin. Struct. Biol. 1999 9 56. 208 H.W. Park S.-T.Kim A. Sancar and J. Deisenhofer Science 1995 268 1866. 209 M. Komiyama and J. Sumaoka Curr. Opin. Chem. Biol. 1998 2 751. 210 M. R. Redinbo J. J. Champouxand W. G. J. Hol Curr. Opin. Struct. Biol. 1999 9 29. 211 J. J. Champoux Prog. Nucleic Acid Res. Mol. Biol. 1998 60 111. 212 H.-K. Wang S. L. Morris-Natschke and K.-H. Lee Med. Res. Rev. 1997 17 367. 213 J. M. Berger S. J. Gamblin S. C. Harrison and J. C. Wang Nature 1996 379 225. 214 M. R. Redinbo L. Stewart P. Kuhn J. J. Champouxand W. G. J. Hol Science 1998 279 1504. 215 L. Stewart M. R. Redinbo X. Qiu W. G. J. Hol and J. J. Champoux Science 1998 279 1534. 216 C. T. Walsh Acc. Chem. Res. 1986 19 216. 217 A. Yasui M. Takao A. Oikawa A. Kiener C. T. Walsh and A. P. M. Eker Nucleic Acids Res. 1988 16 4447.218 T. Tamada K. Kitadokoro Y. Higuchi K. Inaka A. Yasui P. E. de Ruiter A. P. M. Eker and K. Miki Nat. Struct. Biol. 1997 11 887. 219 R. Epple and T. Carell Angew. Chem. Int. Ed. Engl. 1998 37 938. 220 T. Carell and R. Epple Helv. Chim. Acta 1997 80 2191. 221 B. N. Trawick A. T. Daniher and J. K. Bashkin Chem. Rev. 1998 98 939. 222 J. Bernstein W.A. Lott B. A. Steinberg and H. L. Yale Am. Rev. Tuberc. 1952 65 357. 223 J. S. Blanchard Annu. Rev. Biochem. 1996 65 215. 224 D. B. Young and K. Duncan Annu. Rev. Microbiol. 1995 49 641. 225 A. Dessen A. Quemard J. S. Blanchard W. R. Jacobs Jr. and J. C. Sacchettini Science 1995 267 1638. 226 A. Quemard J. Sacchettini A. Dessen C. Vilcheze R. Bittmen W. R. Jacobs Jr. and J. S. Blanchard Biochemistry 1995 34 8235.227 P. J. Brennan and H. Mikaido Annu. Rev. Biochem. 1995 64 29. 228 R. F. Zabrinski and J. S. Blanchard J. Am. Chem. Soc. 1997 119 2331. 333 Annu. Rep. Prog. Chem. Sect. B 1999 95 299—334 229 D. A. Rozwarski G. A. Grant D. H. R. Barton W.R. Jacobs Jr. and J. C. Sacchettini Science 1998 279 98. 230 B. K. Sinha J. Biol. Chem. 1983 258 796. 231 L. A. Basso R. J. Zheng and J. S. Blanchard J. Am. Chem. Soc. 1996 118 11 301. 232 O. T. Dann and C. E. Carter Biochem. Pharmacol. 1964 13 677. 233 T. S. Soper and J. M. Manning J. Biol. Chem. 1981 256 4263. 234 T. D. H. Bugg and C. T. Walsh Nat. Prod. Rep. 1992 9 199. 235 J. E. J. Churchich J. Biol. Chem. 1967 242 4414. 236 R. R. Rando Biochem. Pharmacol. 1975 24 1153. 237 H. Ueno J. J. Likos and D. E. Metzler Biochemistry 1982 21 4387. 238 G. T. Olson M. Fu S. Lau K. L. Rinehart and R. B. Silverman J. Am. Chem. Soc. 1998 120 2256. 239 D. Peisach D. M. Chipman P. W. Van Ophem J. M. Manning and D. Ringe J. Am. Chem. Soc. 1998 120 2268. 240 B. L. Golden A. R. Gooding E. R. Podell and T. R. Cech Science 1998 282 259. 241 H. Suga P. A. Lohse and J. W. Szostak J. Am. Chem. Soc. 1998 120 1151. 242 G. J. Narlikar and D. Herschlag Annu. Rev. Biochem. 1997 66 19. 334 Annu. Rep. Prog. Chem. Sect. B 1999 95 299&mdash

ISSN:0069-3030    

DOI:10.1039/a808581a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

9 Reaction mechanisms Part (iv) Free radical reactions Daniel E. Falvey Department of Chemistry and Biochemistry University of Maryland College Park MD 20742 USA 1 Free radical substitution reactions Transfer of hydrogen and other atoms to free radicals continues to attract interest. Much current work has emphasized either experimental or theoretical approaches to determining bond dissociation energies (BDEs). Recent theoretical studies have also emphasized the modeling of reaction mechanisms and transition states rather than simply predicting the structures of these intermediates. Dakternicks et al. undertook an abinitio (QCISD)study of the transition state for H atom transfer from trialkylsilanes and trialkylgermanes to various alkyl radicals. The structure of the transition state and predicted barrier height were found to be relatively insensitive to the number of methyl groups on the Si or Ge atom but these factors were found to be sensitive to the degree of alkyl substitution at the carbon center tertiary alkyl radicals exhibiting the highest barriers and methyl radical the lowest.Rauk et al. used DFT to compute barriers for H atom abstraction from model cysteine-containing peptides 1. Interestingly they found that captodative stabilization makes the -C—H bond weaker than a typical S—H bond (77 vs. 88 kcal mol). However the barrier to intramolecularH atom transfer from the S—H bond in cysteine to the -radical center was considered to be so high as to preclude any reaction (Scheme 1). Zhang computed the BDEs of C—Hbonds that are - to carbon radical centers.For saturated hydrocarbon radicals these were all predicted to fall in the range 32—39 kcal mol. Conjugation of the radical center with a vinyl group or phenyl ring raises the BDE to 563 kcal mol. Zhang used abinitio CBS-4 model calculations to predict the e.ects of .uorination on the strength of C—H bonds. One or two .uorines alpha to the C—H bond are found to weaken it slightly relative to the un.uorinated hydrocarbon. However in trifluoromethane inductive e.ects predominate and the C—H bond strengthens. The e.ect of .uorination on larger radicals is also considered. Ro� mer used gas phase ion cyclotron resonance mass spectrometry methods to 335 Annu. Rep. Prog. Chem. Sect.B 1999 95 335—348 Scheme 1 examine the e.ect of -silyl and -phosphino groups on C—H BDEs. The BDEs for the -silyl and -phosphino C—H bonds are 923 and 953 kcal mol respectively. Elford and Roberts used EPR to study the regiochemistry of H atom transfer from dimethylamine to either tert-butoxy or ((CH ) Si) N· radicals in cyclopropane solution. It was found that tBuO· preferentially attacks the N—H bond but the more sterically hindered ((CH ) Si) N· shows a preference for the less hindered C—H bond. At high concentrations of (CH ) NH the propensity for reaction at N—H decreases presumably due to H-bonding. Kochi’s group considered the longstanding problem of distinguishing concertedH atom transfer from pathways involving reversible electron transfer followed by proton transfer.Using nanosecond LFP they examined the reaction of excited state quinones and alkylbenzenes derivatives. Substantial kinetic isotope e.ects (k /k )are observed. However the addition of salts stabilizes ion pair intermediates and allows for their detection. Fletcher et al. examined the e.ects of supercriticalCO on the free radical chlorination of hydrocarbons (e.g. cyclohexane). It was found that the ratio of in-cage to cage-escape products was dependent only on the viscosity of the .uid. No evidence for a special clustering e.ect of the CO molecules around the solute was found. Dneprovskii et al. examined the selectivity of free radical chlorination in various halogenated solvents.It was found that Cl· forms complexes even with chlorinated solvents and that these complexes show the greatest selectivity for reaction with tertiary C—H bonds. Increasing chlorination of the solvent molecules tends to decrease selectivity presumably because this increases the ionization potential of the solvent. Brominated solvents give the highest selectivity leading these authors to conclude that the complexes were of a primarily charge transfer character. Struder examined unimolecular homolytic substitution reactions between aryl radicals and silicon centers (4 Scheme 2). The best leaving groups were found to be either stannyl radicals or silyl radicals. 2 Free radical rearrangements Because of its use as a chemical probe for radical intermediates the cyclopropylcarbinyl rearrangement and analogous processes have continued to attract the attention of both experimentalists and theorists.Hex-5-enyl radicals have long been known to undergo a 5-exo cyclization to give cyclopentyl carbinyl radicals and the reaction has been widely adopted in organic synthesis. Recent mechanistic work in this area emphasizes accurate theoretical modeling of cyclization and the e.ects of heteroatoms on its rate. Finally several other interesting rearrangements have been characterized. 336 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 2 Scheme 3 Scheme 4 Martinez et al. used abinitio methods (PMP2/6-31G*//UHF/6-31G*)to calculate barrier heights for cyclopropylcarbinyl radical ring opening as a function of substitution on the ring and radical center.As expected ring substituents that can stabilize the product radical center are predicted to reduce the activation barrier and substituents that stabilize the reactant radical center increase the barrier. Choi et al. used phenylselenol competitive trapping kinetics to determine the ring opening rate constants for several secondary and primary 2-phenylcyclopropyl carbinyl radicals (7 Scheme 3). A high yield of ring opened products formed at high concentrations of phenylselenol led to the conclusion that these rearrangements occur on a picosecond timescale. Horner et al. report the measurement by LFP of the rate of a cyclopropylcarbinyl rearrangement (Scheme 4).This was accomplished by means of a reporter group substituted on the 2-position (relative to the carbinyl radical)of the cyclopropane ring. The reporter group itself was a cyclopropyl moiety that was further substituted with a 2-aryl group. Generation of radical 9 by a laser pulse results in a rate-determining ring opening to give radical 10. The latter rapidly rearranges to give radical 11 which is benzylic and thus detectable by UV—vis transient absorbance. Assuming that the initial rearrangement was rate determining its rate constant can be determined by monitoring the appearance of radical 11. A variety of derivatives were examined and rate constants for ring opening of 10 to 10 s were reported. Grossi et al. employed spin trapping and EPR spectroscopy to characterize the 3-exo cyclization of allyloxyl radicals (13 Scheme 5).The latter were generated photolytically and they were found to be in equilibrium with the higher energy oxiranylcarbinyl radical form (14). Smith et al. undertook extensive DFT calculations on the kinetics and ther- 337 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 5 (1) Scheme 6 modynamics of the cyclopropylcarbinyl radical ring opening and those of the O and N-heterocyclic analogs. Heteroatom substitution at the 1-position had relatively little e.ect on the barrier. However heteroatom substitution at the 2-position greatly accelerates the ring opening process. Interestingly the kinetically favored product of these ring-openings is the less stable heteroatom-centered radical.Wang and Tanko were interested in developing reactive probes for radical ion chemistry. These workers showed that one-electron oxidation of phenylcyclopropanes by ceric ammonium nitrate (CAN)gives cyclopropyl rearranged products. Naphthylcyclopropane radical cations also rearrange but more slowly. Anthrylcyclopropanes were not observed to rearrange at all but instead the anthryl ring is attacked by nucophiles such as methanol. Phillips et al. were interested in developing probes for radical ion chemistry. Spirocyclohexa-2,5-dienone derivatives (e.g. 15)were studied with electrochemical methods (Scheme 6). It was found that these anion radicals (e.g. 16)isomerized to the distonic forms (e.g. 17)with rate constants in excess of 10 s.Walton has proposed a general method for the determination of activation energies for reversible unimolecular processes such as the ring-opening/ring closing. This is carried out by determining the midpoint temperature (T)which is the temperature at which both isomers exist in equal quantities. The activation energy for the process is found to obey the empirical law (1). E (kcal mol)0.044T 0.22 Dolbier et al. examined rate constants for cyclization of partially .uorinated hex-5-enyl radicals using competitive trapping experiments. Fluorination of the vinylic position was found to have little e.ect on the rates of cyclization. However .uorination at the radical center and adjacent carbons was found to accelerate this reaction up to 50-fold.This was attributed to pyramidalization of the radical center which places it in a geometry that favors addition. Rhodes et al. examined the behavior of benzylidine imine cation radicals using EPR spectroscopy (18 Scheme 7). In cases where the nitrogen is substituted by an alkyl chain of at least three carbon atoms,H atom transfer from either - or -carbon atoms is observed the latter is preferred when it is possible. Horner et al. used LFP and competition studies to determine rate constants for 5-exo cyclization of amidyl radicals (20 Scheme 8). In the simplest cases these cycliz- 338 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 7 Scheme 8 Scheme 10 Scheme 9 ations are several orders of magnitude faster than observed for the corresponding hydrocarbon radicals.For example radical 20 cyclizes with a rate constant of 210 s. In the case of 6-exo cyclizations H atom transfer was found to be the predominant pathway. Musa et al. used laser .ash photolysis and competition kinetics to determine rate constants for the reaction of -amido radicals (22 Scheme 9). It was found thatHatom transfer rate constants were similar to those observed for -ester radicals. However the rate of 5-exo radical cyclization was slower presumably due to steric e.ects. Iserloh and Curran report an interesting variation on the venerable 5-exo radical cyclization reaction mechanism (Scheme 10). Speci.cally alkyl radicals add to the C——Nof acyl germane oxime ethers 25 and hydrazones with concomitant elimination of the germyl radical.The cyclization rate constants (2526)were determined by competition kinetics to be ca. 10 s. Montevecchi et al. have expanded on earlier studies of the cyclization of vinyl radicals onto azido groups. In the case of ortho-substituted phenyl azides e.cient cyclization ultimately giving indole derivatives was observed. However in the case of alkyl azides a number of competing processes were observed. Chatgilialoglu et al. report that the one carbon ring expansion reactions of cyclopentan-2-one methyl radicals (28 Scheme 11)can be used as free radical clocks. On the basis of competitive trapping they estimate that the rate constant for the rearrangement 2829 is 5.210 s. 339 Annu. Rep. Prog.Chem. Sect. B 1999 95 335—348 Scheme 11 Scheme 12 Scheme 13 Scheme 14 Crich and Mo studied rearrangements of some oxetenon-4-yl carbinyl radicals such as 30 (Scheme 12). Decarboxylation (via 32)and an interesting ring expansion reaction (via 31)are observed. Assuming typical rate constants for H atom transfer from Bu SnH it is possible to derive rate constants and Arrhenius parameters for fragmentation and rearrangement. Amii et al. report an unusual rearrangement/rearrangement reaction of ethyl 3,3- di.uoroalanin-3-yl benzophenone Schi. base derivatives (33 Scheme 13). Ipso attack of the radical center on one of the phenyl rings is followed by expulsion of ArCN from the resulting spiro radical (34)to give radical 35. Laco� te and Renaud examined the e.ect of Lewis acids on the 1,2-shift of acyloxy groups to radical centers (36 Scheme 14).It was found that precomplexation of the precursor with a Lewis acid (Sc(OTf) )enhances the rate of migration relative to that for reduction. Miura and Momoki have reported a new thermal rearrangement of persistent 340 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 15 Scheme 16 Scheme 17 radicals N-(2,6-diphenyl)phenylthioaminyl radicals (38 Scheme 15)that involves the addition of the nitrogen to one of the substituent phenyl rings. The rearranged radical 39 then disproportionates to give the N-thiocarbazole derivative 40. 3 Elimination reactions Elimination of small neutral molecules (e.g. CO OCS)from free radicals has been studied by time resolved spectroscopic methods (LFP and pulse radiolysis).Additionally the elimination of anionic leaving groups from carbons adjacent to free radical centers was discovered several years ago and continues to attract attention. Zhao et al. generated the benzoylthiyl radical (PhCOS·)using pulse radiolysis. Like the analogous benzoyloxy radical (PhCO ·)the thiyl analog absorbs in the visible region of the spectrum ( 460 nm)and undergoes -scission to give Ph· along with OCS (k8.510 s). Simakov et al. used PTOC ester methodology to generate oxycarbonyl radicals (41 Scheme 16). Using LFP they measured the rate constants for decarboxylation of a series of such radicals. The values ranged from 10 to 10 s depending on the stability of the incipient alkyl radical (R·).Bonifacic et al. used pulse radiolysis to study the reactions of glycine cation radicals (42 Scheme 17). It was found that the major decay pathway was C—C bond scission to produce ·CH NH and CO . Meggers et al. have extended their studies on the dissociation of -phosphatoxy radicals (such as 43)by examining the subsequent electron transfer reactions of the resulting cation radicals (Scheme 18). In rigid double-stranded DNA it was found that cation radical 44 could abstract electrons from neighboring guanine bases (G)provided they were7Å distant. The Crich group have continued their mechanistic studies of the biochemically relevant -phosphatoxy alkyl radicals and have obtained interesting results from LFP studies on PTOC esters precursors of radical 46 (Scheme 19).The major radical product was found to be the isomer 48 which results from a 1,2-shift of a phosphatoxy 341 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 18 Scheme 19 Scheme 20 group. Radical cation 47 which would have resulted from an elimination of the phosphatoxy group was not observed in the organic solvents employed in these studies. The authors point out that a radical cation—anion pair may form in the geminate solvent cage in which case its collapse to radical 48 must be fast relative to cage escape. 4Free radical additions Cle� ment et al. have developed a new nitrone spin trap 49 which is capable of distinguishingHO· addition from processes involving one-electron oxidation followed by nucleophilic attack of water (Scheme 20).When HO· is generated in the presence of the spin trap 49 the expected OH adduct is formed. On the other hand one-electron oxidation of this substrate leads to a net transfer of an ethyl group across the ring (via 51). The two products (50 and 52)are readily distinguished by EPR through di.erences in their hyper.ne coupling constants. Wahl et al. also considered the problem of speci.city in spin trapping reactions. Again 49 is shown to selectively trap alkoxyl radicals in preference to alkylperoxyl radicals. 342 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 21 Scheme 22 Scheme 23 Ryu et al. report an interesting variation on the venerable 5-exo-trig cyclization.In this case various primary alkyl radicals such as 53 are generated in the presence of CO. Addition of the latter to give 54 followed by cyclizatioto an imino group and subsequent reduction was found to give good yields of the corresponding pyrrolidone derivatives 56 (Scheme 21). Kim and Jon report that free radical carboxylation occurs when alkyl radical is generated using hexabutylstannane in the presence of PhSCOCl. The PhSCO group is transfered to the alkyl radical and the Cl atom is transfered to the tin-radical (Scheme 22). Chatgilialoglu and Gimisis prepared precursors suitable for generation of the 1-peroxyl radical of 2-deoxyuridine (61 in Scheme 23)in order to determine the mechanism of its decay along separate pathways.On the basis of O labeling experiments it was concluded that the peroxyl radical partitions between heterolytic scission to give the 2-cation (63)which in turn is hydrolyzed to give the 2-ribolactone (64) and reaction with another peroxide radical to give the 2-alkoxy radical (62). The latter eliminates a uracil-l-yl radical also giving the ribolactone 64. Swansburg et al. report further studies on nucleophilic additions to alkene cation radicals. Speci.cally one-electron oxidation of 2-methyleneadamantane derivatives in the presence of cyanide ions was found to give the corresponding 2-cyano-2-methyladamantane derivatives along with the isomeric 2-cyanomethyladamantane derivatives. Hudson et al. used EPR spectroscopy to examine the addition of silyl and germyl radicals to cyclopentenone and cyclohexenone derivatives.It was found that the 343 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 24 Scheme 25 3-position is the preferred site for addition. When that site is blocked by substitution the radical adds to the carbonyl oxygen. Guillemin et al. examined addition of NH and PH to simple alkenes. Generally it was found that NH · from low wavelength photolysis adds to the more substituted carbon of electron poor alkenes such as acrylonitrile. These reactions are thought to have been important in the atmosphere of primitive Earth. Tripathi used pulse radiolysis combined with time resolved resonance Raman spectroscopy to examine the reaction of ·OH radicals with several donor substituted benzenes.These studies established that there is a direct one-electron transfer pathway that competes with an alternative addition/elimination pathway that gives the same intermediates (Scheme 24). The former is favored with more easily oxidized arenes and the latter with less easily oxidized arenes. Sonntag’s group examined the reaction of ·OH with benzoquinone 65 using pulse radiolysis methods (Scheme 25). It is concluded that the initially formed adduct 66 rapidly tautomerizes (2.510 s)to give the 2,4-dihydroxyphenoxyl radical 67. Br and I with a variety of arenes in a Eberson et al. studied the reaction of Cl solvent of hexa.uoroisopropyl alcohol. EPR experiments indicate that arene cation radicals are formed in these reactions.5 Initiators and free radical formation The search for new methods for creating free radicals remains an active area of research. Novel methods involve the use of photochemical generation and of ion radical dissociation. Although it is not exactly a free radical intitiator radical reactions of the new oxidant dimethyldioxirane (DMD)have attracted considerable interest in the past year. Gregory and Jenks applied G2 abinitio calculations to the study of the thermal dissociation of sulfenic esters (e.g. CH —S—O—CH ). Scission of the C—O bond rather than the O—S bond was found to be the most favorable pathway. Alam et al. carried out LFP experiments on N-hydroxypyridine-4-thione 68.Photolysis of this compound gives hydroxyl radicals and 69 from N—OH bond scission an aminoxyl species 70 from O—H bond scission and the excited triplet state 344 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 26 Scheme 27 Scheme 28 of the precursor. At high pH direct photooxidation of the conjugate base of the precursor also produces aminoxyl radical 70 (Scheme 26). Jockusch and Turro examined novel phosphinoyl radicals (73 Scheme 27)using LFP and time-resolved EPR. The P coupling constants and rate constants for addition to alkenes were characterized. Neckers et al. report the generation of amines from photolysis of tetraalkylammonium tetraphenylborate salts. The mechanism involves single electron transfer from the borate to the ammonium ion followed by rapid (ca.200 ns)elimination of the diphenylmethanol radical. This reaction was studied by LFP. Preliminary applications of this process to photolithography are discussed. Nakamura et al. used spin trap methodology to investigate the decay mechanism of the initiator 1-cyclohexyl-1-methylethyl peroxypivalate (74 Scheme 28). Thermolysis of the latter in the presence of the spin trap gave only the cyclohexyl (76)and tert-butyl adducts (75) indicating that -scission of the initially formed radicals is fast relative to the rate of trapping. Anne et al. used cyclic voltammetry to determine the rate constants for proton transfer from the cation radical of 9-cyanomethylacridane 77 to proton acceptors of varying basicities (Scheme 29).The values for these rate constants were compared to those for proton transfer reactions of other acridane derivatives. As expected the intrinsic barriers derived from these Brønsted plots were found to correlate with the C—H BDE of the parent acridane. Deviations from this behavior were attributed to steric e.ects. Freccaro et al. undertook extensive kinetic studies of radical generation through the photooxidation of various benzylic species. The groups listed in Scheme 30 were found to eliminate with rate constants ranging from 10 to 10 s in CH CN. The 345 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Scheme 29 Scheme 30 Scheme 31 e.ects of thermodynamics and counterions on the rates of these processes are discussed. One general topic that has engaged both theoretical and experimental research groups is the mechanism by which dimethyldioxirane (DMD)and similar compounds oxygenate hydrocarbons (Scheme 31).The central question is whether free radical intermediates are formed on the pathway to products. Most of the recent experimental and theoretical work seems to suggest that these reactions are concerted. But some controversy remains. Shustov and Rauk used various abinitio methods to examine the oxidation of hydrocarbons by DMD with the goal of reconciling apparently contradictory experimental .ndings. These calculations predict a highly polar transition state with substantial carbocationic character. This transition state bifurcates between a lower energy concerted reaction channel and a higher energy radical channel.Du and Houk likewise calculate a concerted asynchronous transition state for DMD oxidation of hydrocarbons. Their computed transition state shows considerably greaterOH than CO bond formation and a considerable diradical character. Glukhovtsev carried out high-level DFT calculations on the oxidation of hydrocarbons by DMD. The results of these calculations also support a concerted insertion of an O atom into the C—H bond rather than a stepwise process involving radical intermediates. 346 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 Liu et al. present a combined abinitio and experimental study on the epoxidation of alkenes by DMD. First these workers found none of the allylic oxidation products that were earlier advanced as evidence for a free radical pathway.Finally DFTcalculated transition states were consistent with the concerted pathways. Dinoi et al. examined the reactions of the spin label TEMPOwith dioxiranes. The major product from the reaction of TEMPO with dimethyldioxirane is the methyl—TEMPO adduct. The pathways to this product are complex but kinetic studies indicate that TEMPOitself induces decomposition ofDMDrather than acting as a ‘passive’ radical scavenger. Bravo et al. examined the mechanism for oxidation of hydrocarbons by dimethyldioxirane. Contrary to the widely held view these workers propose a free radical pathway for these reactions. These conclusions were based on chemical trapping studies with O quinolonium ion and CBrCl .Am. Chem. Soc. 1998 120 211. References 1 D. Dakternicks D. J. Henry and C. H. Schiesser J. Chem. Soc. Perkin Trans. 2 1998 591. 2 A. Rauk D. Yu and D. A. J. Armstrong J. Am. Chem. Soc. 1998 120 8848. 3 X.M. Zhang J. Org. Chem. 1998 63 3590. 4 X.M. Zhang J. Org. Chem. 1998 63 1872. 5 B.Ro� mer G. G. Gatev M. L. Zhong and J. I. Brauman J. Am. Chem. Soc. 1998 120 2919. 6 P.E. Elford and B. P. Roberts J. Chem. Soc. Perkin Trans. 2 1998 1413. 7 T.M. Bockman S. M. Hubig and J. K. Kochi J. Am. Chem. Soc. 1998 120 2826. 8 B. Fletcher N. K. Seleman and J. M. Tanko J. Am. Chem. Soc. 1998 120 11 839. 9 A. S. Dneprovskii D. V. Kuznetsov E. V. Eliseenkov B. Fletcher and J. M. Tanko J. Org. Chem. 1998 63 8860. 10 A. Struder Angew.Chem. Int. Ed. Engl. 1998 37 462. 11 F. N. Martinez H. B. Schlegel and M. Newcomb J. Org. Chem. 1998 63 3618. 12 S. Y. Choi P. H. Toy and M. Newcomb J. Org. Chem. 1998 63 8609. 13 J. H. Horner O.M. Musa A. Bouvier and M. Newcomb J. Am. Chem. Soc. 1998 120 7738. 14 L. Grossi S. Strazzari B. C. Gilbert and A. C. Whitwood J. Org. Chem. 1998 63 8366. 15 D.M. Smith A. Nicolaides B. T. Golding and L. Radom J. Am. Chem. Soc. 1998 120 10 223. 16 Y. H. Wang K. H. McLean and J. M. Tanko J. Org. Chem. 1998 63 628. 17 J. P. Phillips J. G. Gillmore P. Schwartz L. E. Brammer D. J. Berger and J. M. Tanko J. Am. Chem. Soc. 1998 120 195. 18 J. C. Walton J. Chem. Soc. Perkin Trans. 2 1998 603. 19 W.R. Dolbier X. X. Rong M. D. Bartberger H. Koroniak B. E. Smart and Z.Y. Yang J. Chem. Soc. Perkin Trans. 2 1998 219. 20 C. J. Rhodes H. Morris H. Agirbas M. Standing and Y. M. Zhang J. Chem. Soc. Perkin Trans. 2 1998 1375. 21 J. H. Horner N. Tanaka and M. Newcomb J. Am. Chem. Soc. 1998 120 10 379. 22 O.M. Musa S. Y. Choi J. H. Horner and M. Newcomb J. Org. Chem. 1998 63 786. 23 U. Iserloh and D. P. Curran J. Org. Chem. 1998 63 4711. 24 P. C. Montevecchi M. L. Navacchia and P. Spagnolo Eur. J. Org. Chem. 1998 1161. 25 C. Chatgilialoglu V. I. Timokin and M. Ballestri J. Org. Chem. 1998 63 1327. 26 D. Crich and X. S. Mo J. Am. Chem. Soc. 1998 120 8298. 27 H. Amii S. Kondo and K. Uneyama Chem. Commun. 1998 1845. 28 E. Laco� te and P. Renaud Angew. Chem. Int. Ed. Engl. 1998 37 2259. 29 Y. Miura and M. Momoki J. Chem. Soc.Perkin Trans. 2 1998 1185. 30 R. Zhao J. Lind G. Merenyi and T. E. Eriksen J. Am. Chem. Soc. 1998 120 2811. 31 P. A. Simakov F. N. Martinez J. H. Horner and M. Newcomb J. Org. Chem. 1998 63 1226. 32 M. Bonifacic I. Stefanic and G. L. Hug J. Am. Chem. Soc. 1998 120 9930. 33 E. D. K. Meggers M. Spichty U. Wille and B. Giese Angew. Chem. Int. Ed. Engl. 1998 37 460. 34 S. Y. Choi D. Crich J. H. Horner X. H. Huang F. N. Martinez M. Newcomb D. J. Wink and Q.W. Yao J. 35 J. L. Cle� ment B. C. Gilbert W.F. Ho N. D. Jackson M. S. Newton S. Silvester G. S. Timmins P. Tordo and A. C. Whitwood J. Chem. Soc. Perkin Trans. 2 1998 1715. 347 Annu. Rep. Prog. Chem. Sect. B 1999 95 335—348 36 R. U. R. Wahl L. S. Zeng S. A. Madison R. L. DePinto and B. J. Shay J. Chem. Soc.Perkin Trans. 2 1998 2009. 37 I. Ryu K. Matsu S. Minakata and M. Komatsu J. Am. Chem. Soc. 1998 120 5838. 38 S. Kim and S. Y. Jon Chem. Commun. 1998 815. 39 C. Chatgilialoglu and T. Gimisis Chem. Commun. 1998 1413. 40 S. Swansburg K. Janz G. Jocys A. Pincock and J. Pincock Can. J. Chem. 1998 76 35. 41 A. Hudson D. Waterman M.A. Della Bona A. Alberti M. Benaglia and D. Macciantelli J. Chem. Soc. Perkin Trans. 2 1998 2255. 42 J. C. Guillemin C.M. Breneman J. C. Joseph and J. P. Ferris Chem. Eur. J. 1998 4 1074. 43 G. N. R. Tripathi J. Am. Chem. Soc. 1998 120 4161. 44 M.N. Schuchmann E. Bothe J. von Sonntag and C. von Sonntag J. Chem. Soc. Perkin Trans. 2 1998 791. 45 L. Eberson M. P. Hartshorn F. Radner and O. Persson J. Chem. Soc. Perkin Trans. 2 1998 59. 46 D. D. Gregory and W. S. Jenks J. Org. Chem. 1998 63 3859. 47 M.M. Alam O. Ito G. N. Grimm and W. Adam J. Chem. Soc. Perkin Trans. 2 1998 2471. 48 S. Jockusch and N. J. Turro J. Am. Chem. Soc. 1998 120 11 773. 49 A.M. Sarker A. Lungu A. Mejiritski Y. Kaneko and D. C. Neckers J. Chem. Soc. Perkin Trans. 2 1998 2315. 50 T. Nakamura W. K. Bus.eld I. D. Jenkins E. Rizzardo S. H. Thang and S. Suyama Chem. Lett. 1998 965. 51 A. Anne S. Fraoua V. Grass J. Moiroux and J. M. Save� ant J. Am. Chem. Soc. 1998 120 2951. 52 M. Freccaro A. Pratt A. Albini and C. Long J. Am. Chem. Soc. 1998 120 284. 53 G. Shustov and A. Rauk J. Org. Chem. 1998 63 5413. 54 X. H. Du and K. N. Houk J. Org. Chem. 1998 63 6480. 55 M.N. Glukhovtsev C. Canepa and R. D. Bach J. Am. Chem. Soc. 1998 120 10 528. 56 J. Liu K. N. Houk A. Dinoni C. Fusco and R. Curci J. Org. Chem. 1998 63 8366. 57 A. Dinoi R. Curci P. Carloni E. Damianni P. Stipa and L. Greci Eur. J. Org. Chem. 1998 871. 58 A. Bravo F. Fontana G. Fronza F. Minisci and L. H. Zhao J. Org. Chem. 1998 63 254. 348 Annu. Rep. Prog. Chem. Sect. B 1999 95 335&m

ISSN:0069-3030    

DOI:10.1039/a808607i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

10 Gas phase organic ion–molecule reaction chemistry Wan Yong Feng and Scott Gronert* Department of Chemistry and Biochemistry San Francisco State University San Francisco CA 94132 USA 1 Introduction Mass spectrometry has become an exceptionally versatile tool for studying molecular structure reactivity and thermochemistry. An important application for mass spectrometry is the study of ion—molecule reactions because the results can give deep insights into the fundamental physical and chemical behavior of gas phase ions. Although most organic chemistry is done in the condensed phase gas phase studies can provide exceptionally valuable information about reaction mechanisms because in the absence of solvation the intrinsic reactivity of the reaction partners can be probed in exquisite detail.Moreover comparison of gas phase results with those from solution gives important insights into the role of solvation and ion pairing in condensed phase reactions. In addition to product distributions and rate constants ion—molecule reactions can be used to determine physical properties such as gas phase acidities basicities and bond energies. Ion—molecule reactions are also a promising tool for solving di.cult analytical problems because the reactions are often highly selective and extremely fast. Instruments for the study of ion—molecule reactions include the .owing afterglow selected ion .ow tubes (SIFT) Fourier transform ion cyclotron resonance mass spectrometers (FT-ICR or FT-MS) quadrupole ion trap mass spectrometers (QITMS) high-pressure mass spectrometers (HPMS) and ion beam systems.The basic operating principles and limitations of each of the experimental methods will not be discussed here. The present paper reviews studies of gas phase organic ion—molecule reactions that were published in 1998. In developing this review it was necessary to set arbitrary boundaries on the chemistry that would be included. First the focus is on bimolecular reactions. Unimolecular reactions initiated by ionization or collisional activation are ubiquitous in mass spectrometry and provide a foundation for chemical analysis by mass spectrometry; however they are outside the scope of the review. Second organometallic reactions have been excluded. In recent years gas phase organometallic chemistry has become an extensive .eld with a strong focus on the bonding properties of bare or partially coordinated metals.It truly represents a distinct .eld and deserves 349 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 a separate review. Third the reactions of large biomolecules have not been included because the great majority of these studies have been limited to simple proton transfer processes rather than functional group transformations. Finally the present review is generally limited to experimental studies; however theoretical studies that are intimately associated with experimental work have been included. In recent years a number of reviews concerned fully or partly with ion—molecule reactions have appeared. Numerous topics including the following have been reviewed organic ions, electrophilic reactions with aromatic substrates, nucleophilic displacement reactions, radical anions, molecular cluster reactions, fullerene carbon clusters, sulfur and silicon containing ions gas phase ion—atom reactions, atmospheric ion chemistry, and the analytical applications of ion—molecule reactions.— These reviews provide a foundation for the material presented here. 2 Reactions ofcarbocations and related cations Carbocations still play a central role in gas phase organic chemistry. During 1998 a number of aspects of carbocation reactivity were probed. Studies of electrophilic aromatic substitution continued and the factors that a.ect carbocation stability were examined.In addition a range of reactions were studied including some with possible analytical applications. A. Reactions the simplest carbocation. Using the SIFT-Drift technique they studied the Several studies were reported involving the electrophilic reactivity of a range of carbocations. Glosik and Lindinger presented a detailed study of the reactivity of CH energy dependence of the reaction rates of CH with N O and NH . With N O HCO is the major product and with NH CH —— NH is the major product. Both rates decrease with increasing energy due to the shorter lifetime of the encounter complexes. Freitas and O’Hair used the .owing-afterglow technique to complete a detailed study of the reactions of the CH OCH cation with 21 neutral nucleophiles.This cation is particularly interesting because it is an ambident electrophile and can react at oxygen or carbon. The neutral compounds studied include alcohols ethers thiols thioethers aliphatic ketones acids esters and amides as well as simple aromatics (benzene pyridine and aniline). Three general types of reaction channels were observed addition at carbon followed by elimination of methanol to give [M CH] ions [eqn. (1)] S 2 methylation giving [M CH ] ions [eqn. (2)] and adduct formation giving [M CH OCH ]. Adduct formation dominates for most substrates but high yields of [M CH] ions are observed with amines thiols and benzene. As an additional product channel the amides give products arising from the cleavage of the amide bond by CH OCH [eqn.(3)]. 350 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 Van der Rest et al. have investigated the gas phase reactions of CH OCH with ketones and aldehydes. In reactions with ketones adduct formation and methylation of the carbonyl group are observed.A third product type is observed in the reactions of CH OCH acylium ions (RCO) via a hydride abstraction pathway. Bortolini et al. reported the ion—molecule reactions of CH (C H N and C H N) with carbon suboxide C O . N—— — C—CH with aldehydes. With the exception of CH ——O aldehydes also give CN and related ions O——C——C—— C——O N—— — C—CH—— CH—C—— — O CO (4) The reactions mainly proceed via electrophilic addition of the ions to the electron-rich central carbon of C O followed by CO loss [eqn.(4)]. Tsuji and Nishimura have investigated the reactions ofCH C H and C H with a series of phenyl ketones (PhCOX XH CH C H Ph COPh). Proton transfer is a common pathway and is followed by fragmentation in some cases. With the allyl cation adducts and their decomposition products are also seen. Chen and co-workers have reported an interesting gas phase analog of the Ziegler—Natta ole.n polymerization. Using an electrospray source they formed an alkyl zirconocene cation (Cp ZrCH Cpcyclopentadienyl) and allowed it to react with several alkenes.Addition products corresponding to the polymerization of the alkene were observed. Williamson and Creaser have used a quadrupole ion trap to complete a series of studies involving the reactions of the benzoyl cation [C H CO] and in 1998 reported its reactivity with amines. They found that the benzoyl cation reacts with compounds containing an amine group (2-methylprop-2-ylamine but-2-ylamine but- 1-ylamine 4-chlorobenzylamine and aniline) to yield characteristic ions [M C H CO] [MNH ] and C H CONH . These ions come from association followed by the loss of C H CONH or (MNH ). The branching ratios are sensitive to the amine’s environment and it is possible that the benzoyl cation could be used as a probe for determining the presence or nature of amines in unknown substrates.Polycondensation of benzyl methyl ether C H CH OCH with gaseous benzyl cation C H CH was studied by Gru� tzmacher and Dohmeier-Fischer in an FTICR. The benzyl cation reacts with up to three C H CH OCH molecules generating condensation product ions with m/z181 271 and 361 [eqn. (5)]. The process involves an electrophilic aromatic substitution by the benzyl cation llowed by loss of methanol to give a benzylated benzyl cation which can react further. Other pathways are observed and by careful labeling studies an array of di.erent ion—molecule reaction mechanisms was discovered.In another study of 351 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 electrophilic aromatic substitution Holman et al. used the reactions of alkyl carbenium ions with .uoro- chloro- and bromobenzene furan thiophene and pyrrole to investigate the intermediacy of and complexes on the potential energy surface. On the basis of collision-activated dissociation (CAD) studies and molecular orbital calculations it was concluded that the extent of complex formation decreases as the aromatic ring becomes more electron (C indicated that alkyl cation rearrangements could occur within the loose complexes. rich H FC H ClC H BrC H OC H SC H N).Other experiments Electrophilic aromatic substitution was studied by radiolytic methods at relatively high pressures (50—4560 torr) by three groups. Aschi et al. investigated the lifetimes of ion—neutral complexes in the tert-butylation of toluene. By a process involving the reaction of protonated toluene with a pro-electrophile (e.g. tert-butyl chloride) they deduced lifetimes on the order of 10 seconds. Angelini et al. studied the competition between protonation and alkylation in the reactions of ethyl cations with N-methylpyrrole and thiophene. Protonation is favored over alkylation under all conditions. With pyrrole alkylation occurs mainly at the -carbon but with thiophene there is a slight preference for alkylation at the -carbon.Using an FT-ICR and radiolytic techniques Crestoni studied the reactions of p-Me Si-substituted 1,3-diphenylpropane with a series of electrophiles. Proton and Me Si shifts were examined and it was con.rmed that intra-annular proton migrations are faster than inter-annular proton migrations. Finally a review of the gas phase chemistry of arenium ions was presented by Fornarini and Crestoni. An interesting aspect of this work is the description of internal solvation by a phenyl group in protonated ,- diphenylalkanes. B. Isomerizations Kuck et al. have investigated hydrogen scrambling reactions in protonated polyarenes such as tribenzylmethane and 2-benzylindane. Even on short time scales (1 s) intra- and inter-annular proton exchange leads to complete equilibration.As an extreme 21 protons are randomized in protonated tetrabenzylmethane prior to fragmentation. Although all the aromatic hydrogens are randomized the hydrogens of the aliphatic or alicyclic spacer group do not participate in the exchange process. Using an FT-ICR mass spectrometer Vanderrest et al. have observed that the keto—enol isomerization of the H COC(O)CH CO cation to an enol ion H COC(OH)CHCO can be catalyzed by methanol [eqn. (6)]. Although the reaction is exothermic unimolecular isomerization by a 1,3-hydrogen shift has a high energy barrier and is not observed. Reactions with labeled methanol and ligand exchange experiments indicate that the isomerization occurs within the methanol encounter complex.Finally Matsumoto et al. reported that the protonation of diazirenes in the gas phase leads to ring opening and the formation of protonated diazomethanes. 352 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 (7) R—X BH R HX B C. Stability Abboud et al. have used dissociative proton attachment reactions to probe the stability of carbocations. Using an FT-ICR the following reactions were studied where RCCl CBr CI CBr F and 1-adamantyl [eqn. (7)]. Thermochemical data were obtained by determining the weakest acid (BH) that would give a dissociative proton attachment reaction at an observable rate. The results indicate that the trihalocarbocations are substantially less stable than typical 2° alkyl carbocations such as the prop-2-yl cation.The same group used a similar approach to determine the energetics of the halide transfer reactions between tert-butyl chloride (CH ) CCl and the trichlorocyclopropenyl cation C Cl . They found that the trichlorocyclopropenyl cation is substantially more stable than the tert-butyl and 1-adamantyl cations. Using the following homodesmotic reaction the stabilization due to -delocalization in the trichlorocyclopropenyl cation was estimated to be over 50 kcal mol [eqn. (8)]. This is comparable to the stabilization estimated in an analogous way for the cyclopropenyl cation itself. 3 Nucleophilic substitution and addition Nucleophilic substitution and addition are two of the most important processes in organic chemistry and have been investigated in great detail in the gas phase.Over the years many groups have used experiment and theory to study gas phase S 2 reactions and there is now a major e.ort to understand the detailed dynamics of substitution reactions in the gas phase. In addition there is a continuing e.ort to understand solvation e.ects by probing the di.erences between gas phase and condensed phase nucleophilic reactions. A. SN2 Reactions One of the most widely studied reactions in organic chemistry is bimolecular nucleophilic substitution (S 2) and over the years there have been numerous studies of gas phase S 2 reactions. The potential energy surface for gas phase S 2 reactions is believed to be a double-well potential and statistical theories have been used to explain the single-collision kinetics of many S 2 reactions.Although many systems can be treated statistically some systems behave nonstatistically to the extent that direct dynamics need to be considered when interpreting experimental observations.— In 1998 both Hase and Brauman presented important overviews focusing on the dynamics and mechanisms of gas phase S 2 reactions. In addition several papers 353 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 Scheme 1 addressed this issue. Craig et al. reported nonstatistical reactivity in a vibrationally excited S 2 intermediate. They found that the deposition of a large amount of vibrational energy into an intermediate complex in the gas phase S 2 reaction of chloride with methyl tri.uoroacetate leads to a dramatic and nonstatistical increase in the rate.The intermediate a complex between CF CO CH and Cl was formed either directly or through the reaction of CH O with CF COCl. The latter process leads to the same complex [CF CO CH ·Cl] but is about 55 kcal mol more exothermic (Scheme 1). The direct reaction gives an S 2 e.ciency of 2% whereas theCH O initiated route produces CF CO with an e.ciency of 8—15%. The results are not compatible with an RRKM analysis and suggest that the intermediate’s lifetime is too short for the statistical redistribution of the energy. Direct dynamics ab initio calculations and trajectory calculations have been applied to the S 2 reaction of .uoride with methyl chloride.The trajectory calculations give values of rate constants as a function of reactant relative translational energy E and CH Cl temperature that are in good overall agreement with the experimental rate constants and those calculated using an ion—molecule capture/RRKM statistical model. At low energies there is evidence for formation of an ion—dipole complex; however the lifetime for this complex is too short for complete energy randomization to occur. The results indicate that the reaction exothermicity is primarily deposited in product vibrational modes. It was also found that excess reactant translational energy is almost exclusively converted to product translational energy.Hase and co-workers also presented a similar trajectory study for the S 2 reaction of Cl CH Cl. For reactions with 20—80 kcal mol of translational activation the results indicate that the reactive trajectories are direct with little trapping in ion—dipole complexes. Excess product energy is primarily partitioned into relative translation with small amounts of energy partitioned into vibration and rotation. This is in contrast to the reaction of Cl with CH Br where considerable energy is deposited in product vibrations. Graul Bowers and co-workers have completed a careful study of the energy released in the dissociation of metastable species corresponding to intermediate complexes in gas phase S 2 reactions.They examined a set of S 2 reactions with widely varying reaction partners and exothermicities (Cl and CN with methyl trifluoroacetate F with anisole and Cl with ethyl iodide). The kinetic energy release pro.les cannot be .tted by phase space theory unless an adjustable parameter is included to account for the non-statistical behavior. One conclusion from the study is that there is relatively little randomization of the internal energy after passing through the S 2 transition state. In continuing work related to the atmospheric chemistry of .uorocarbons Lee et al. used a SIFT apparatus to study the reaction of HO with CH process is proton transfer to give CHF presumably through an S F .The dominant (86%) but some F (11%) is formed 2 pathway. The rate constant is relatively small at 300K 354 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 (2.410cm molecules) and exhibits a minor negative temperature dependence. In contrast the same workers had observed a much higher rate constant (1.810cm molecule s) for the reaction of O with CH F in an earlier study. The work was complemented by high level ab initio calculations. The proton transfer to F is endothermic but can compete with the exothermic substitution reaction because there is a signi.cant S 2 barrier on the potential energy surface. The e.ect of solvation on nucleophilic displacement reactions was reviewed by Viggiano and co-workers in 1998.Overall the results illustrate the transition from gas phase to condensed phase S 2 reactions. In the temperature range of 163—500 K they .nd that the rate of the S 2 reactions of CH Br withOH(H O) Cl(H O) and F(H O) decreases as the level of hydration increases. The studies also demonstrate that in the absence of a fast S 2 reaction channel other mechanisms such as association and ligand switching can become important. The S 2 reactions of cationic substrates have been the subject of two studies. Speranza and Troiani used their radiolytic technique to study the regio- and stereochemistry of the reactions of chiral allylic alcohols with the protonated methyl ethers of the same alcohols.A detailed analysis of the product mixtures indicates that both S 2 and S 2 pathways are active. As for stereochemistry the S 2 reaction involves anti attack by the nucleophile whereas the S 2 reaction involves retention of con.guration (front-side attack). Although inversion is expected in S 2 processes the authors have observed retention in other studies of cationic S 2 reactions. Apparently prior to reaction there is hydrogen bonding between the nucleophile and the protonated ether which holds the nucleophile on the front-side of the substrate. As the bond to the leaving group breaks the nucleophile slips into its place in a process reminiscent of a ligand switching reaction. In another cationic system Katritzky and co-workers investigated the competition between intermolecular and intramolecular gas phase substitution reactions.Cationic substrates were used with pyridine derivatives acting as the leaving groups. Although substrates with internal nucleophiles (amines) readily underwent substitution reactions during collision activated dissociation (CAD) there was no evidence of bimolecular S 2 reactions with analogous cationic substrates. An example leading to a spiro-ammonium ion is given below [eqn. (9)]. In these systems the S 2 displacements have large intrinsic barriers and require the addition of considerable energy (i.e. CAD) to reach the transition state. Semi-empirical and ab initio calculations on model systems support the conclusion that the S 2 reactions of amines with alkylpyridinium ions are not viable under thermal conditions.B. Addition/elimination reactions An unusual substitution process in radical cations has been reported by Gru� tzmacher and co-workers. The reactions of ammonia with the radical cations of vinylic halides were studied by FT-ICR mass spectrometry and it was discovered that substitution is able to compete with highly exothermic proton transfer reactions. The substitution process involves ammonia addition to the alkene followed by elimination of a halogen 355 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 Scheme 2 Scheme 3 atom (Scheme 2). Relatively e.cient substitutions are observed for the reactions of ammonia with 2-halopropenes and 3,3,3-tri.uoro-2-halopropenes (halochloro bromo and iodo).The work was supported by ab initio calculations and evidence was presented that suggests that the deprotonation process also involves an addition/ elimination pathway. An extensive study of nucleophilic acyl substitution in esters was completed by Gross and co-workers using FT-ICR mass spectrometry. They found that the reaction of ester enolates with neutral alcohols led to a transesteri.cation process with rate constants ranging from approximately 10—10cm molecules. An abbreviated reaction mechanism is given in Scheme 3 where R and R are methyl ethyl n-propyl n-butyl or isopropyl. In the .rst step proton transfer within the encounter complex gives the ester and an alkoxide RO. Attack of RO at the carbonyl followed by expulsion of RO gives the new ester which can react with RO to yield the transesteri.ed enolate.These reactions give little or no detectable production of either alkoxide anion and their rate constants increase with increasing acidity in the primary straight-chained alkyl alcohols whereas steric e.ects associated with branched alcohols lead to lower rate constants. Equilibrium constants were measured directly and also by taking the ratio of rate constants in the forward and reverse directions. Interestingly the trends in the equilibrium constants are opposite to those found in solution. The gas phase experiments show a signi.cant preference for having the larger of the alkoxy groups as part of the ester enolate but the formation of methyl ester enolates is preferred in the condensed phase.The authors point out that polarizability plays a larger role in stabilizing gas phase ions and this could be the underlying factor in the preference for the formation of the larger ester. 356 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 4 Reactions ofdistonic radical cations and anions C. Other processes Bondybey and co-workers have investigated the reactions of acetone and acetaldehyde in acidic (H(H O) ) and basic (OH(H O)) water clusters. In the ‘acid’ clusters only ligand exchanges are observed but in the ‘base’ clusters OH catalyzed addition of two molecules of the carbonyl compound takes place to give aldol condensation products. Although aldol condensations are often not favorable for ketones in solution the evaporation that accompanies the aldol process in the cluster makes the reaction e.ectively irreversible.Finally Bachrach Hare and Kass have reported that alkali metal salts of doubly deprotonated benzene can be formed by sequential decarboxylation of metal salts of phthalic acids [eqn. (10)]. Computational work supports the experimental studies of these intriguing species. Distonic radical ions are chemically fascinating species that contain two potentially reactive sites a radical and an ionic site that are formally localized on di.erent atoms in a molecule. As a result the chemical behavior of distonic ions may be driven by either the radical site the ionic site or both. In recent years there has been tremendous activity in this area with work focusing on the identi.cation of distonic species characterization of their reactivity and measurements of their stability.Several studies on distonic radical cations and anions were presented in 1998. A. Cations Kentta� maa has been very active in this area in recent years and presented four papers on the subject in 1998o studies focused on the dimethylene ketene distonic cation. This ion can be generated by electron impact on -methylene--butyrolactone and behaves like a typical carbon-centered radical in the gas phase. Reactions with benzeneselenol dimethyl diselenide and allyl iodide lead to the expected products of hydrogen abstraction CH Se abstraction and iodine abstraction respectively.The reactions of this distonic ion with alkyl disul.des was the subject of a second study. Rapid cleavage of the disul.de bond was seen in all cases and it was argued that the ion could provide a useful analytical tool for identifying and locating disul.des in neutral substrates [eqn. (11)]. In addition to the product from direct RS abstraction a product corresponding to 357 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 RS abstraction and loss of a hydrogen atom was observed in some cases. It is believed that this ion is the result of a subsequent hydrogen atom transfer within the product complex to give a new distonic radical cation. Kentta� maa also reported on the reactions of the distonic biradical cations of pyridinium derivatives. Striking di.erences in reactivity are observed for the 2,6-dimethylenepyridinium and 3,5- dimethylenepyridinium isomers. The former is highly reactive with reagents such as tert-butyl isocyanide (HCN abstraction) while the latter exhibits only a slow association process [eqn.(12) and (13)]. The di.erences may be related to the electronic states of the ions. Density functional theory (DFT) calculations indicate that the 2,6 isomer is a ground state singlet biradical whereas the 3,5 isomer is a triplet. Finally the Kentta� maa group reported improved approaches for forming distonic cations incorporating a phenyl radical site. Moraes and Eberlin presented a study of the gas phase reactions of the ortho meta and para dehydrobenzoyl distonic radical cations C H O·.These species can be formed by electron impact on nitroacetophenones and they display a strong duality of chemical behavior acting as either free radicals or acylium ions depending on the choice of the neutral reaction partner. With dimethyl disul.de CH S abstraction is observed a characteristic pathway for carbon centered radicals. In reactions with 2-methyl-1,3-dioxolane 2-methoxyethanol and chloromethyloxirane this distonic ion exhibits typical acylium reactivity with the formation of cyclic ionic ketals [eqn. (14)]. With the ortho isomer the close proximity of the radical to the acylium ion allows for combined behavior in which a hydrogen atom transfer is included in the process and a closed shell cation is formed as the product.When allowed to react with a combination of reagents (dimethyl disul.de and chloromethyloxirane) the meta and para isomers exhibit radical and acylium reactivity in a pair of sequential reactions. Flammang Bouchoux and co-workers have formed isomeric distonic radical cations CH CNCH · and CH NCCH · from the reactions of the ketene radical cation with acetonitrile and methyl isocyanide respectively. Reactions with pyridined and dimethyl disul.de con.rm that the ions are distonic and that they do not easily 358 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 Scheme 4 interconvert under the reaction conditions. Flammang and co-workers have also studied a sulfur containing distonic radical cation HC(OH)SCH ·.This radical cation is formed during the ionization of S-isobutyl thioformate. Reaction of this ion withNO leads to a signal whose mass corresponds to the replacement of CO with NO in the cation. No other structural information was obtained on the product. B. Anions Squires and co-workers— presented several studies involving distonic radical anions. Most importantly they reported a remarkable synthetic pathway for forming gas phase distonic radical anions. Treatment of bis-trimethylsilyl compounds with a combination of .uoride and molecular .uorine leads to the sequential loss of the trimethylsilyl groups (as (CH ) SiF) and the regioselective generation of a radical and an anionic site. The mechanism is initiated by .uoride-induced desilylation to give a carbanion (Scheme 4).Reaction of the carbanion with molecular .uorine leads to dissociative electron transfer and the formation of a carbon-centered radical. A second .uorideinduced desilylation gives the distonic radical anion. The method is general and a wide range of distonic ions were reported including the negative ions of trimethylenemethane benzynes oxyallyl and acetoxyl biradicals and ,3-dehydrotoluene. Substrates with three trimethylsilyl groups lead to distonic biradicals. Using this method for the regioselective formation of distonic radical anions Wenthold et al. completed a comparative study of the reactions of ortho- meta- and para-benzyne anions with a series of neutral reagents. The meta and para isomers are the most reactive and mainly exhibit nucleophilic rather than radical behavior.These ions can be converted to dehydrobenzoates by treatment with CO . The proton a.nities of the dehydrobenzoates were determined and when combined with values for other benzoates a linear free energy relationship was established. Analysis of the results indicates that the phenyl radical site can be classi.ed as a strong inductive withdrawing and strong resonance donating ‘substituent’. Hill and Squires reported a .owing afterglow/triple quadrupole study of the radical anion derived from 2,4- dimethylenecyclobutane-1,3-diyl (non-Kekule� benzene) which can be formed by the reaction of O· with 1,3-dimethylenecyclobutane [eqn. (15)]. From a variety of bracketing and kinetic measurements acidities electron a.nities and bond strengths 359 Annu.Rep. Prog. Chem. Sect. B 1999 95 349—372 were determined for species related to non-Kekule� benzene. A striking result is that in the conversion of 1,3-dimethylenecyclobutane to non-Kekule� benzene via sequential C—H bond cleavages the second C—H bond is over 16 kcal mol stronger than the .rst. This can be explained by the intrinsic antiaromatic character of the -system in non-Kekule� benzene. Bierbaum and co-workers have used a SIFT apparatus to investigate the properties of the radical anion of cyclooctatetraene (C H ·) and its singly and doubly dehydro-derivatives (C H and C H ·). C H · and C H · undergo electron detachment upon collisional activation with helium but C H undergoes a rearrangement to an isomeric species with markedly di.erent properties.It was concluded that the initial C H ion is not a vinylic species but rather is a -delocalized anion with an allenic moiety incorporated into the eight-membered ring. Rearrangement leads to a relatively stable [3.3.0] bicyclic species with a proton a.nity and electron binding energy very similar to those of the cyclopentadienide ion [eqn. (16)]. In addition to the collisional activation studies the reactions of the hydrocarbon anions with a series of neutral reagents were examined. Through bracketing and equilibrium measurements electron binding energies proton a.nities and heats of formation were determined for each of the anions.The most notable result is the adiabatic electron a.nity of cyclooctatetraene (0.55 eV 0.02 eV). H· C CH · and C CHC H·). Deuterium labeling charge Finally Schwarz Bowie and co-workers found routes to three isomeric C H · radical anions (HC reversal and neutralization reionization experiments were used to con.rm the connectivity of atoms within these species. 5 Reactions ofcarbon clusters Cluster ion chemistry has grown into a large and varied .eld encompassing many disciplines. In recent years fullerene and other carbon clusters have become an area of intense interest and detailed information about them can be found in recent review articles. During 1998 there were several studies that focused on the gas phase ion chemistry of fullerenes.As part of a long series of studies Bohme and co-worked a paper on the reactions of fullerene cations with nitrogen heterocycles. Using a SIFT apparatus they found that C C and C react with pyridine to give addition products in which the total number of pyridine additions equals the charge of the ion. From equilibrium measurements a bond dissociation enthalpy of 360 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 OCH . Proton transfer as well as addition was observed. 19.5 kcal mol was determined for C with one pyridine. In multiply derivatized products evidence for both ‘ball-and-chain’-like as well as ‘spindle’-like structures was presented.Reactions of fullerene cations with pyrrole lead simply to electron transfer. In another study Petrie and Bohme reported a study of the proton transfer reactions of a series of triply protonated fullerene derivatives. Gru� tzmacher and co-workers have presented a study of the reactions of pyridine with C clusters. Ionization of perchlorocoronene appears to lead to C ions with two di.erent structures. One form of ion undergoes a single addition of pyridine and the resulting product clusters lose pyridine during CAD. This suggests that these C ions retain the polycyclic aromatic framework of coronene. The other form of the C ion adds up to .ve molecules of pyridine and undergoes loss of molecular and atomic hydrogen during CAD.These C clusters are likely to be monocyclic and during the addition process the pyridines are incorporated into the ring (followed by rearrangements) to give more complex structures. Liu and co-workers have investigated the reactions of neutral C with the methoxymethyl cation,CH The authors suggest that addition occurs along a 6—6 bond in the fullerene and leads to a [3 2] cycloadduct (tetrahydrofuran derivative). (17) 6 Reactions oforganic ions containing heteroatoms (B Si Ge P S Cl) The gas phase chemistry of boron- silicon- germanium- phosphorus- sulfur- and halogen-containing organic ions has been an active area of research in recent years because mass spectrometry provides a relatively simple way to generate and study unusual ionic species that are unknown or di.cult to investigate in the condensed phase.As a result ion chemistry provides a powerful tool for probing the reactivity of novel heteroatom-containing species. The following section focuses on systems where the heteroatom carries the charge or plays a pivotal role in stabilizing the ionic site. DePuy Damrauer and co-workers presented an extensive SIFT study of the reactions of BH with hydrogen simple hydrides and small hydrocarbons. Most of the reactions follow the same general route and produce a boron insertion product with the loss of H [eqn. (17)]. BH R—H R—B—H H These results also provide an unusual example of insertion into a C—C bond.In the reaction of BH with C H CH BH is formed in addition to CH CH BH. Squires and co-workers have used a .owing afterglow/triple-quadrupole mass spectrometer and ab initio MO calculations to investigate the gas-phase negative ion chemistry of a series of Lewis acid—base complexes of BH and BF with neutral ligands (Me SBH Me NBH Me PBH Me SBF Me OBF Et NBH and Et OBF ). They found that the addition of the Lewis acid (BH or BF ) greatly enhances the reactivity of the ligand. For example Me SBH Me NBH and Me PBH are from 18—20 kcal mol more acidic than Me S Me N and Me P respectively.The authors also found that the carbanions formed by deprotonation of the complexes are stable at thermal energies and do not rearrange to more stable borate isomers. The acidity enhancements were shown to be mainly due to the 361 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 increased electron binding energies of the carbanions in the deprotonated complexes. The complexes also exhibit enhanced reactivity with nucleophiles because complexation to BH or BF leads to a better leaving group. For example F and NH react with Me SBH Me NBH and Me OBF by nucleophilic substitution at carbon. Under the same conditions no such reactions occur for the uncomplexed ligands Me S Me Nand Me O.In addition the ethylated complexes (e.g. (CH CH ) NBH ) give E2 reactions that are not observed with the uncomplexed ligands. Following up on earlier experimental work Damrauer and co-workers reported a computational study on the proton transfer behavior of the silaacetylide anion (HCSi). For reactions with HCSiH and H CSi reaction paths and energetics were computed using high levels of ab initio theory. Mishima and co-workers have used an FT-ICR to study the e.ects of substituents on the stability of dimethylphenylsilyl cations XC H (CH ) Si (X4-MeO 3,5-Me 4-Me 3-Me 3-Cl-4-MeO 3-F-4- MeO H 3-Cl-4-Me 4-Cl 4-F 3-Cl 3-F 3-CF ). Application of the Yukawa—Tsuno linear free energy relationship gives a of5.87 and an r of 0.29.Both the r and values are signi.cantly smaller than those observed for carbon analogs -cumyl cations. On the basis of these results the authors concluded that in the dimethylphenylsilyl cation there is no signi.cant -delocalization of the positive charge into the benzene ring. Cooks and co-workers reported two studies involving silicon cations. In one the binding energies of a series of pyridines to SiF and SiF were determined using a kinetic method. For meta- and para-substituted pyridines good correlations could be found with the pyridine proton a.nities but evidence of steric and agostic interactions was observed in the ortho-substituted systems. Cooks has also investigated the reactions of translationally excited Si and SiCl ions with per- .uorohexane. The former abstracts one .uorine and the latter two or three.The study also included a variety of other small cations including PCl PCl· Cl Br I CBr andW. Xavier and Riveros have used an FT-ICR to complete a detailed investigation of the gas phase ion chemistry of Ge(OMe) . The ion—molecule reactions of HGe(OMe) · Ge(OMe) H Ge(OMe) · HGe(OMe) and GeOMe resulting from fragmentation of Ge(OMe) · were studied in the presence of the neutral substrate. Aside from the formation of protonated Ge(OMe) species of the general structure Ge (OMe) (n3—7) are essentially the only products observed.Although the structures of the product ions could not be determined unequivocally there is evidence that some of the product ions contain Ge—Ge bonds. Castiglioni et al. have used ion trap mass spectrometry to study ion—molecule reactions in CH GeH and in CH GeH —SiH mixtures. Rate constants and e.ciencies were reported for a number of reaction processes. It appears that chain propagation of ions containing Ge Si and possibly C occurs through species such as GeSiCH (n4 6) and GeSi CH (n6 7). These ions are formed from the reaction of Si H (n2 4) and Si H (n4 5) respectively with methylgermane. Morizur and co-workers have reported an interesting example of a [4 2] cycloaddition reaction between a phosphenium ion (CH O) P and 2,3-dimethylbuta-1,3-diene [eqn.(18)]. The CAD spectrum of the adduct ion that is formed in this reaction shows two major fragments corresponding to loss of 2,3-dimethylbuta-1,3- diene and methanol respectively. The loss of methanol from the addition product along with a comparison to the analogous condensed phase reaction of ((CH ) N) P 362 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 with 2,3-dimethylbuta-1,3-diene was used as evidence for formation of a cyclic adduct. An interesting aspect of sulfur chemistry was probed by Illies using high pressure mass spectrometry. It is known that organosulfur cations can associate with neutral organosulfur species to yield products with two center three electron bonds (2c—3e).In these 2c—3e interactions two electrons occupy a orbital and one electron occupies a * orbital both of which are localized between the two sulfur atoms. Determination of the enthales and entropies for the association reactions of (c-C H S)· and (c- C H S)· with c-C H S and c-C H S [eqn. (19)] provided insight into the nature of the bonding interaction. For the four possible combinations there is relatively little variation in the enthalpy of association (27.5—25.3 kcal mol). In the reaction of (c-C H S)· with c-C H S (thiirane) ring expansion was observed with the formation of (c-C H S )·. Ring strain appears to play a large role in this process and expansions were not observed when larger cations (i.e.c-C H S·) or neutrals (i.e. c-C H S) were used in analogous reactions. Nibbering and co-workers also reported thermochemistry for sulfur —sulfur three electron bonds in cationic dialkyl sul.de dimer complexes. The two center three electron bonds of alkyl halide radical cation dimers have been studied by the same two research groups. From CAD studies Illies and co-workers found strong evidence for two center three electron bonds in [CH BrBrCH ]· and [CH IICH ]·. In the radical cation dimer complexes of CH F and CH Cl other connectivities are possible and it appears that the latter adopts two forms [CH ClClCH ]· and [CH Cl·H—CH Cl]·.The situation with CH F is more complicated and an additional structural type is possible [FCH —H·H—CH F]·. Nibbering and co-workers also studied the CH Cl radical cation dimer complex. With bracketing and CAD threshold experiments they were able to determine a number of the thermochemical properties of the complex including the heat of formation and the bond dissociation energy of the two center—three electron bond 26.2 1.2 kcal mol. 7 Reactions oforganic compounds with small inorganic ions The reactions of small inorganic ions with organic substrates play important roles in a number of .elds. During the last year there was a high level of activity in three major areas.First several groups continued investigating the synthesis and fate of organic compounds in extraterrestrial atmospheres and interstellar clouds. Second a number of studies focused on using small inorganic ions as chemical ionization reagents for trace analysis. Finally a few studies focused on using O· as a reagent for chemical analysis as well as for producing unusual organic ions. 363 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 (20) A. Atmospheric and interstellar chemistry In an e.ort to better understand the ionosphere of Titan McEwan and co-workers used a SIFT to study a number of reactions ofN· andN · with simple saturated and unsaturated hydrocarbons and nitriles. Additional reactions relevant to Titan’s ionosphere (involving hydrocarbon and nitrile ions) were also presented.It was observed that several of the reactions of N· andN · with hydrocarbons led to nitrile ions and that the secondary reactions of these species led to more complex nitriles [eqn. (20)]. N· CH ——CH HCNH HCN H· other products It was concluded that these processes as well as hydrocarbon ion—nitrile association reactions could be important in the chemistry of nitrogen rich atmospheres such as Titan’s. Related reactions of nitrogen-containing species were also reported in another paper. In a complementary study McEwan examined the reactions of molecular nitrogen and nitrogen atoms with numerous small cations including HCN· HCO HCO C H · andHC NH. The reactions with nitrogen atoms are particularly noteworthy because problems in producing reactive atoms and monitoring their concentrations have severely limited the study of these species.However .ow tube techniques have proven to be useful for ion—atom reactions since titration methods can be used to monitor the atom concentrations. Using their SIFT apparatus McEwan and co-workers found no characteristic pathway for the reactions of the cations with N atoms. Ion—atom interchange in which the N atom is substituted for an atom in the ion and atom abstraction in which an atom is transferred from the ion to the N atom are the most common processes. Overall the results indicate that spin-conservation is not rigorously obeyed in the reactions of the quartet nitrogen atoms.Snow Bierbaum and co-workers have also used a SIFT to study the reactions of atoms in the gas phase. Rate constants were determined for the reactions of the radical cations of benzene naphthalene and pyrene with H O and N atoms. Relatively large rate constants were observed for the majority of these reactions. The results suggest that these ions are too reactive to survive in high abundances in interstellar clouds and that it is more likely that protonated polycyclic aromatic hydrocarbons (PAH) are responsible for the di.use interstellar bands seen in the optical spectra of stars. B. Trace analysis A systematic SIFT study of the reactions of H O NO· and O · with a range of organic species including carboxylic acids and esters, ethers, sul.des and other sulfur-containing molecules, amines and other nitrogen-containing molecules, and aromatic and aliphatic hydrocarbons was reported by Spane� l and Smith in a series of papers.The neutral reagents used in these studies were chosen because they are known constituents in the atmosphere human breath or emissions from fruit and food products. A goal of this work is the development of an analytical method for the detection and quanti.cation of trace gases using the SIFT technique. For the great majority of these species H O reacts via proton transfer to give theM 1 cation of the substrate. In some cases such as thioacetic acid fragmentation accompanies the 364 Annu. Rep. Prog. Chem.Sect. B 1999 95 349—372 proton transfer. With the hydrocarbons slow reactions mainly ion—molecule association are observed. With NO· as the reagent ion the acids and esters undergo either association reactions or transfer of an alkoxide (or hydroxide) fragment to produce acyl cations and neutral nitrites. The ethers mainly react via hydride transfer to give alkoxy-stabilized carbocations. With nitrogen- and sulfur-containing species charge transfer sometimes accompanied by fragmentation is the most common pathway. An exception are species whose ionization potential is greater than that of NO. For example charge transfer to pyridine or nitriles is endothermic so the preferred reaction is ion—molecule association. In reactions with aliphatic hydrocarbons hydride transfer (at widely varying rates) to give carbocations was observed and with aromatic hydrocarbons charge transfer dominated.Since O · has a higher recombination energy than NO· its reactions led to greater fragmentation. For most of the species dissociative charge transfer was the dominant process although some of the more robust aromatic species reacted via non-dissociative charge transfer. Using their SIFT apparatus Spane� l and Smith applied this chemical ionization approach to the analysis of trace components in the vapor space above chopped onions and crushed garlic. In a related study Arnold and co-workers have determined the rate constants for the reactions of H O and NO· with a series of alkanes (C —C ) at temperatures from 300 to 500 K.There is a correlation between the rate constant and the size of the alkane and thresholds are observed near the point where the reactions become exothermic. The threshold size for observing measurable rates with H O occurs at hexane with association being the predominant reaction channel. Hexane is also the smallest alkane that gives reasonable rates for a (mainly) hydride transfer reaction with NO·. These results taken together with earlier data from the same lab suggest that H O NO· and O · are not suitable chemical ionization agents for the general analysis of hydrocarbon (alkane) emissions because they either do not react su.ciently dly across the range of possible alkanes (H O and NO·) or they react to form complex product mixtures (NO· and O ·).Lindinger and co-workers — presented a series of papers detailing their development of proton-transfer reaction mass spectrometry (PTR-MS) as a tool for on-line monitoring of volatile organic compounds (VOCs) at ppt levels. The PTR-MS system is based on rapid proton transfer reactions from H O to a substrate within a .ow-drift tube. H O was chosen as a reactant ion because it does not react with any of the natural components of air and because most of the common VOCs have proton a.nities larger than water so that fast proton transfer reactions are likely. This method has found applications in medicine food science and environmental research. A variety of endogenous volatile organic compounds were analyzed including acetone methanol ethanol propan-2-ol and isoprene from human breath; diallyl sul.de allyl methyl disul.de diallyl disul.de diallyl trisul.de and dimethyl sul.de from garlic; methanethiol dimethyl sul.de acetaldehyde dimethylamine and methanol from deteriorating beef; acetaldehyde acetone propanal methyl ethyl ketone 2-methylpropanal ethyl formate and methyl acetate from freshly ground co.ee; and xylene toluene benzene and alkylbenzenes from air samples.Overall the technique appears to be very promising for on-line analysis of trace constituents. 365 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 O and an electron as products [eqn. (22)]. C. O· Reactions The reactions of O· continue to interest organic chemists and several studies appeared in 1998.Arnold et al. reported rate constants and branching ratios for the reactions of O· with a series of simple alkanes (CH CD C H n-C H n-C D n-C H i-C H n-C H i-C H neo-C H n-C H n-C H n-C H n- C H and n-C H ). Reaction e.ciencies steadily increase with size for the small alkanes reaching the collision rate with butane at room temperature. In contrast to a previous study, two signi.cant reaction channels were observed a hydrogen abstraction channel yieldingOH [eqn. (21)] and a reactive detachment channel yielding neutral products and a free electron [eqn.(22)]. For the branched and unbranched alkanes the fraction ofOH produced exponentially decreases with increasing alkane polarizability. Two notable exceptions to this trend are neopentane and isooctane which produce signi.cantly more OH than expected. In fact reactive detachment is absent with neopentane. These results suggest that the reactive detachment channel requires the abstraction of two hydrogen atoms from adjacent carbons in the alkane yielding an alkene H In another study employing O· Nibbering and co-workers generated the distonic acetate radical anion (·CH CO ) as a minor product in reactions with -lactones (propiolactone and butyrolactone). The major reactions are proton transfer and proton transfer followed by loss of H .For the formation of the distonic anions a mechanism was proposed that involves nucleophilic attack at the -carbon leading to ring opening [eqn. (23)]. This is followed by -scission to give an aldehyde and the distonic radical anion. The mechanism is supported by experiments with O· which show no incorporation of O in the distonic anion. Finally rate constants and branching ratios were reported for the reactions of O· and O · with per.uorobenzene. The major pathway in the reaction with O· involves a substitution to give .uoride and the per.uorophenoxy radical [eqn. (24)] whereas reaction with O · primarily leads to ring cleavage and the formation of a radical anion C F · [eqn.(25)]. (24) C O· C F O· (25) F O· F C F· 2CFO C F 366 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 (26) (27) (28) (29) 8 Thermochemical determinations via ion–molecule reactions Along with giving information about reaction products and mechanisms organic ion—molecule reactions can yield valuable thermochemical data for organic compounds. Bond strengths acidities basicities and complexation energies can be determined often with very high accuracy. Progress in each of these areas was made in 1998. A—H AH H(A—H) EA(A·) IP (H·) H(A—H) EA(A·) IP (H·) A A· e H e H· A—H A· H· The necessary acidities can be derived from a variety of experiments involving mass spectrometry and the electron a.nity of the radical can be measured by negative ion photoelectron spectroscopy.In a large collaborative e.ort, the bond strengths and acidities of H CNN and HCNN were determined. First SIFT experiments led to A. Bond energies In recent years a number of groups have used a thermochemical cycle involving gas phase acidities and electron a.nities to determine homolytic bond strengths. The associated equations are shown below [eqn. (26)—(29)]. H values of 372 and 352 kcal mol for H CNN and HCNN respectively. These values in combination with the electron a.nities of HCNN (1.685 eV) and CNN (1.771 eV) lead to .rst and second C—H bond strengths of 97 and 79 kcal mol for H CNN.The same approach was used to determine the O—H bond strength in (CH ) COO—H (85 kcal mol). Brauman and co-workers used acidities and electron a.nities to derive C—H bond strengths for bis(trimethylsilyl)methane and bis(dimethylphosphino)methane. The acidities indicate that silyl and phosphino groups provide stabilization to carbanions that is comparable to that of other third period substituents (thio and chloro). The derived C—H bond strengths (central carbon) are 95 and 92 kcal mol for bis(trimethylsilyl)methane and bis(dimethylphosphino) methane respectively. DeTuri and Ervin used guided ion beam tandem mass spectrometry techniques to accurately measure the gas phase acidity of phenol (351.7 kcal mol) and by combining this value with known thermochemical quantities derived an O—H bond strength of 90.1 kcal mol.B. Acidities and basicities Not all of the studies in 1998 that led to the determination of the gas phase acidity or proton a.nity of organic compounds will be listed here. This information can generally be obtained from the extensive databases maintained by the National Institute of Standards and Technology (NIST). Below a few studies are highlighted that have particular relevance to mechanistic or structural physical organic chemistry. 367 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 Koppel and co-workers reported a comparative study of the acidity of amines in the gas phase dimethyl sulfoxide (DMSO) and water. The study includes acidities for a large number of anilines heterocycles imides and amines.Within families of compounds good correlations are found between acidities in the three mediums; however substituent e.ects are greatly attenuated along the progression from the gas phase to DMSO to water. Higgins and Bartmess have used equilibrium measurements in an ICR mass spectrometer to determine the acidities of a series of long chain alcohols. The values obtained for the alkan-1-ols are consistently smaller than those obtained by Cooks’ kinetic method indicating that some e.ect is altering the structure or dynamics of the transition states in the kinetic method. Finally Schlosser et al. reported the gas phase acidities of a large number of trifluoromethyl-substituted benzenes.The substituent e.ects appear to be additive and each tri.uoromethyl group provides approximately a 13 kcal mol increase in acidity when it is ortho and a 10 kcal mol increase when it is meta or para to the acidic hydrogen. Substituent e.ects on the gas-phase basicities of 1-arylpropynes and 1-aryl-3,3- dimethylbutynes were studied by Mishima and co-workers using an FT-ICR mass spectrometer. The 1-aryl-2-methylvinyl and 1-aryl-2-tert-butylvinyl cations were found to be more stable than 1-arylvinyl cations by 1.8 and 5.5 kcal mol respectively. The e.ects of substituents on the stability of these vinyl cations was investigated with the Yukawa—Tsuno equation. The results indicate that as the stability of the unsubstituted cation increases (i.e.addition of alkyl substituents to the beta carbon) the amount of resonance delocalization into the phenyl ring decreases. The authors also concluded that the -delocalization mechanism in these vinyl cations is not unique and is representative of typical benzylic carbocations. Bouchoux and coworkers used an FT-ICR and high pressure chemical ionization techniques to measure the gas phase basicities of acetic anhydride and several cyclic anhydrides. Acetic anhydride has a higher basicity than the cyclic anhydrides because after protonation it is capable of forming a strong intramolecular hydrogen bond. The authors also note that protonation of the cyclic anhydrides leads to acyl bond cleavage. Cooks and coworkers reported both the proton a.nity (208.8 1.2 kcal mol) and H (361.9 2.9 kcal mol) of urea.In addition Cooks’ group studied the proton a.nity of peroxyacetyl nitrate. They took advantage of a membrane introduction system to handle this fragile highly reactive species and obtained a proton a.nity of 191 3 kcal mol. In a series of papers Raczynska— examined the gas phase basicity of amidines. Substituent e.ects were considered and comparisons were made to condensed phase systems. In two studies Brodbelt and coworkers have evaluated the proton-binding capabilities of a series of multidentate ligands including polyamines and polyethers. In addition Szulejko et al. have used FT-ICR and high-pressure mass spectrometry to determine the proton a.nities of -methoxy alcohols.Finally Decouzon et al. have used an FT-ICR to measure the proton a.nities of ethyl ethenyl and ethynyl phosphines and arsines. The basicity decreases in the order ethylethenylethynyl for both series of compounds and the phosphines are systematically stronger bases than the arsines. 368 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 C. Complexation energies The thermodynamics of association reactions can also be determined with high accuracy by mass spectrometry. Several studies explored the energetics of hydrogen bonding in small to medium size clusters. Chabinyc and Brauman have studied the hydrogen bonding interaction between HCF and a series of small alkoxides. Complexation energies of approximately 20 kcal mol are observed and even when the alcohol is a weaker acid than HCF the complex contains an alkoxide unit (RO HCF ) rather than a CF unit (ROH CF ).The authors argue that the greater localization of charge in the alkoxide allows it to be a stronger hydrogen bond acceptor than CF and therefore enhanced hydrogen bonding in the (RO HCF ) complexes can overcome the di.erence in acidity. McMahon and co-workers used FT-ICR and high pressure mass spectrometers (HPMS) to study deuterium fractionation in complexes of alcohols with alkoxide and chloride ions. Fractionation factors of about 0.5 are observed in the alkoxide ion complexes (deuterium is disfavored in the complex) but values near unity are found for the chloride ion complexes.The di.erences in behavior were explained by the di.ering hydrogen bond strengths and the ability of chloride to partake in multiple site coordination. Stone and Carter have used high pressure mass spectrometry to investigate the complexation of ammonium ions to a series of ,-alkanediols. The diols have larger complexation energies than simple alcohols and it appears that they lead to cyclic complexes with two hydrogen bonding interactions. Wincel has studied the clustering reactions of C H N and C H N withCH CN. The results suggest that addition of the .rstCH CNprovides the core of the cluster and that additional CH CN ligands are held by mainly electrostatic interactions. Finally Meot-Ner Scheiner and Yu have used a combination of high pressure mass spectrometry and theory to study the formation and structure of protonated clusters of ketones and water.The results with the clusters provide insights into the energetic contributions of ionic hydrogen bonds in biological systems. Acknowledgement Support for W.-Y. F. from the National Center for Research Resources (Research Infrastructure in Minority Institutions P20 RR11805) of the National Institutes of Health is gratefully acknowledged. References 1 N.M.M. Nibbering S. Ingemann and L. J. de Koning in The Structure Energetics and Dynamics of Organic Ions ed. T. Baet C. Y. Ng and I. Powis Wiley & Sons New York 1996. 2 S. Fornarini Mass Spectrom. Rev. 1996 15 365. 3 M.L. Chabinyc S.L. Craig C. K. Regan and J. I. Brauman Science 1998 279 1882. 4 M. Born S. Ingemann and N.M. M. Nibbering Mass Spectrom. Rev. 1997 16 181. 5 A.W. Castleman Jr. and S. Wei Annu. Rev. Phys. Chem. 1994 45 685. 6 A.A. Viggiano S. T. Arnold and R. A. Morris Int. Rev. Phys. Chem. 1998 17 147. 7 C. Lifshitz Mass Spectrom. Rev. 1993 12 261. 8 O. Bortolini and M. Fogagnolo Mass Spectrom. Rev. 1995 14 117. 9 J.A. Stone Mass Spectrom. Rev. 1997 16 25. 10 M. Sablier and C. Rolando Mass Spectrom. Rev. 1993 12 285. 369 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 11 D. Smith and P. Spane� l Mass Spectrom. Rev. 1995 14 255. 12 J. S. Brodbelt Mass Spectrom. Rev. 1997 16 91. 13 M.K. Green and C. B. Lebrilla Mass Spectrom. Rev. 1997 16 53. 14 W. Lindinger A.Hansel and A. Jordan Chem. Soc. Rev. 1998 27 347. 15 J. Glosik and W. Lindinger Czech. J. Phys. 1998 48 1241. 16 M.A. Freitas and R. A. J. O’Hair Int. J. Mass Spectrom. Ion Processes 1998 175 107. 17 G. van der Rest G. Bouchoux H. E. Audier and T. B. McMahon Eur. Mass Spectrom. 1998 4 339. 18 O. Bortolini L. Pandolfo C. Tommaselli and P. Traldi Rapid Commun. Mass Spectrom. 1998 12 1425. 19 M. Tsuji and Y. Nishimura Bull. Chem. Soc. Jpn. 1998 71 273. 20 D. Feichtinger D. A. Plattner and P. Chen J. Am. Chem. Soc. 1998 120 7125. 21 B. L. Williamson and C. S. Creaser Eur. Mass Spectrom. 1998 4 103. 22 H. F. Gru� tzmacher and S. Dohmeier-Fischer Int. J. Mass Spectrom. 1998 180 207. 23 R.W. Holman T. Eary E. Whittle and M. L. Gross J. Chem. Soc. Perkin Trans. 2 1998 2187.24 M. Aschi M. Attina F. Cacace and G. D’Arcangelo J. Am. Chem. Soc. 1998 120 3982. 25 M. Aschi M. Attina and G. D’Arcangelo Chem. Phys. Lett. 1998 283 307. 26 G. Angelini R. Bucci G. Laguzzi C. Siciliano and A. L. Segre J. Phys. Chem. A 1998 102 6464. 27 M.E. Crestoni Chem. Eur. J. 1998 4 993. 28 S. Fornarini and M. E. Crestoni Acc. Chem. Res. 1998 31 827. 29 D. Kuck A. Petersen and U. Fastabend Int. J. Mass Spectrom. 1998 180 129. 30 G. Vanderrest P. Mourgues J. Tortajada and H. E. Audier Int. J. Mass Spectrom. 1998 180 293. 31 T. Matsumoto S. Kobayashi M. Mishima G. V. Shustov and M. T. H. Liu Chem. Commun. 1998 1655. 32 J. L. M. Abboud O. Castano J. Elguero M. Herreros N. Jagerovic R. Notario and K. Sak Int. J. Mass Spectrom. Ion Processes 1998 175 35.33 J. L. M. Abboud O. Castano M. Herreros I. Leito R. Notario and K. Sak J. Org. Chem. 1998 63 8995. 34 H. Wang and W.L. Hase J. Am. Chem. Soc. 1997 119 3093. 35 W.L. Hase Science 1994 266 998. 36 V. Deturi P. A. Hintz and K.M. Ervin J. Phys. Chem. A 1997 101 5969. 37 A. A. Viggiano R. A. Morris J. S. Paschkewitz and J. F. Paulson J. Am. Chem. Soc. 1992 114 10 477. 38 W.L. Hase H. Wang and G. H. Peslherbe Adv. Gas Phase Ion Chem. 1998 3 125. 39 S. L. Craig M. Zhong and J. I. Brauman J. Am. Chem. Soc. 1998 120 12 125. 40 M. Igarashi and H. Tachikawa Int. J. Mass Spectrom. 1998 181 151. 41 T. Su H. B. Wang and W. L. Hase J. Phys. Chem. A 1998 102 9819. 42 D. J. Mann and W.L. Hase J. Phys. Chem. A 1998 102 6208. 43 G. H. Peslherbe H. Wang and W.L.Hase J. Am. Chem. Soc. 1996 118 2257. 44 S. T. Graul C. J. Carpenter J. E. Bushnell P. A.M. van Koppen and M.T. Bowers J. Am. Chem. Soc. 1998 120 6785. 45 E. P. F. Lee J. M. Dyke and C. A. Mayhew J. Phys. Chem. A 1998 102 8349. 46 E. P. F. Lee and J. M. Dyke Mol. Phys. 1996 88 143. 47 A. A. Viggiano S. T. Arnold a. A. Morris Int. Rev. Phys. Chem. 1998 17 147. 48 M. Speranza and A. Troiani J. Org. Chem. 1998 63 1020. 49 A. R. Katritzky R. D. Burton M. Qi P. A. Shipkova C. H. Watson Z. Dega-Szafran J. R. Eyler M. Karelson U. Maran and M. C. Zerner J. Chem. Soc. Perkin Trans. 2 1998 825. 50 M. Buchner A. Nixdorf and H.-F. Gru� tzmacher Chem. Eur. J. 1998 4 1799. 51 G.W. Haas D. E. Giblin and M.L. Gross Int. J. Mass Spectrom. Ion Processes 1998 172 25.52 U. Achatz S. Joos C. Berg T. Schindler M. Beyer G. Albert G. Niedner-Schatteburg and V. E. Bondybey J. Am. Chem. Soc. 1998 120 1876. 53 S. M. Bachrach M. Hare and S. R. Kass J. Am. Chem. Soc. 1998 120 12 646. 54 P. K. Chou K. K. Thoen and H. I. Kentta� maa J. Org. Chem. 1998 63 4470. 55 K. K. Thoen D. Tutko J. Perez R. L. Smith and H. I. Kentta� maa Int. J. Mass Spectrom. Ion Processes 1998 175 173. 56 E. D. Nelson K. K. Thoen and H. I. Kentta� maa J. Am. Chem. Soc. 1998 120 3792. 57 K. K. Thoen J. Perez J. J. Ferra Jr. and H. I. Kentta� maa J. Am. Soc. Mass Spectrom. 1998 9 1135. 58 L. A. B. Moraes and M. N. Eberlin J. Am. Chem. Soc. 1998 120 11 136. 59 P. Gerbaux R. Flammang M. T. Nguyen J.-Y. Salpin and G. Bouchoux J. Phys. Chem. A 1998 102 861.60 D. Lahem R. Flammang H. T. Le and M. T. Nguyen Rapid Commun. Mass Spectrom. 1998 12 1972. 61 P. G. Wenthold J. Hu B. T. Hill and R. R. Squires Int. J. Mass Spectrom. 1998 179 173. 62 P. G. Wenthold J. Hu and R. R. Squires J. Mass Spectrom. 1998 33 796. 63 P. G. Wenthold and R. R. Squires Int. J. Mass Spectrom. Ion Processes 1998 175 215. 64 R. R. Squires and C. J. Cramer J. Phys. Chem. A 1998 102 9072. 65 P. G. Wenthold R. R. Squires and W.C. Lineberger J. Am. Chem. Soc. 1998 120 5279. 66 B. T. Hill and R. R. Squires J. Chem. Soc. Perkin Trans. 2 1998 1027. 67 S. Kato R. Gareyev C. H. DePuy and V. M. Bierbaum J. Am. Chem. Soc. 1998 120 5033. 68 S. J. Blanksby S. Dua J. H. Bowie D. Schroeder and H. Schwarz J. Phys. Chem. A 1998 102 9949. 69 A.W.Castleman Jr. and K. H. Bowen Jr. J. Phys. Chem. 1996 100 12 911. 370 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 70 J. Sun and D. K. Bohme Int. J. Mass Spectrom. 1998 179 267. 71 S. Petrie and D. K. Bohme J. Am. Soc. Mass Spectrom. 1998 9 114. 72 J. Sun S. Caltapanides and H. F. Gru� tzmacher J. Phys. Chem. A 1998 102 2408. 73 L. Ma Z. Liu Y. Xu W. Wang X. Guo and S. Liu Rapid Commun. Mass Spectrom. 1998 12 790. 74 C. H. DePuy R. Gareyev J. Hankin G. E. Davico M. Krempp and R. Damrauer J. Am. Chem. Soc. 1998 120 5086. 75 J. Ren D. B. Workman and R. R. Squires J. Am. Chem. Soc. 1998 120 10 511. 76 J. A. Rusho M. S. Gordon N. H. Damrauer and R. Damrauer Organometallics 1998 17 3401. 77 H. Tashiro K. Kikukawa K. Ikenaga N. Shimizu and M. Mishima J.Chem. Soc. Perkin Trans. 2 1998 2435. 78 Y. Tsuno and M. Fujio Chem. Soc. Rev. 1996 25 129. 79 F. Wang S. Ma P. Wong R. G. Cooks F. C. Gozzo and M.N. Eberlin Int. J. Mass Spectrom. 1998 180 195. 80 J. S. Patrick T. Pradeep H. Luo S. Ma and R. G. Cooks J. Am. Soc. Mass Spectrom. 1998 9 1158. 81 L. A. Xavier and J. M. Riveros Int. J. Mass Spectrom. 1998 179 223. 82 M. Castiglioni L. Operti R. Rabezzana G. A. Vaglio and P. Volpe Int. J. Mass Spectrom. 1998 179 277. 83 S. Gevrey M.H. Taphanel and J. P. Morizur J. Mass Spectrom. 1998 33 399. 84 A. J. Illies J. Phys. Chem. A 1998 102 8774. 85 S. P. De Vissera F. M. Bickelhaut L. J. De Koning and N. M. M. Nibbering Int. J. Mass Spectrom. 1998 179 43. 86 L. S. Nichols M. L. McKee and A. J. Illies J. Am. Chem.Soc. 1998 120 1538. 87 S. P. de Visser L. J. de Koning and N. M.M. Nibbering J. Am. Chem. Soc. 1998 120 1517. 88 M. J. McEwan G. B. I. Scott and V. G. Anicich Int. J. Mass Spectrom. Ion Processes 1998 172 209. 89 D. B. Milligan D. A. Fairley M. Meot-Ner and M.J. McEwan Int. J. Mass Spectrom. 1998 179 285. 90 G. B. I. Scott D. A. Fairley C. G. Freeman M. J. McEwan and V. G. Anicich J. Chem. Phys. 1998 109 9010. 91 T. P. Snow V. Le Page Y. Keheyan and V. M. Bierbaum Nature 1998 391 259. 92 P. Spane� l and D. Smith Int. J. Mass Spectrom. Ion Processes 1998 172 137. 93 P. Spane� l and D. Smith Int. J. Mass Spectrom. Ion Processes 1998 172 239. 94 P. Spane� l and D. Smith Int. J. Mass Spectrom. 1998 176 167. 95 P. Spane� l and D. Smith Int. J. Mass Spectrom.1998 176 203. 96 P. Spane� l and D. Smith Int. J. Mass Spectrom. 1998 181 1. 97 S. T. Arnold A. A. Viggiano and R. A. Morris J. Phys. Chem. A 1998 102 8881. 98 W. Lindinger A. Hansel and A. Jordan Int. J. Mass Spectrom. Ion Processes 1998 173 191. 99 A. Hansel A. Jordan C. Warneke R. Holzinger and W. Lindinger Rapid Commun. Mass Spectrom. 1998 12 871. 100 P. Prazeller T. Karl A. Jordan R. Holzinger A. Hansel and W. Lindinger Int. J. Mass Spectrom. 1998 178 L1. 101 S. T. Arnold R. A. Morris and A. A. Viggiano J. Phys. Chem. A 1998 102 1345. 102 D. K. Bohme and F. C. Fehesenfeld Can. J. Chem. 1969 47 2717. 103 M. Born J. Chamot-Rooke S. Ingemann and N.M. M. Nibbering J. Mass Spectrom. 1998 33 677. 104 W. B. Knighton T. M. Miller A. A. Viggiano R.A. Morris and J. M. Van Doren J. Chem. Phys. 1998 109 9632. 105 J. Berkowitz G. B. Ellison and D. Gutman J. Phys. Chem. 1994 98 2744. 106 E. P. Cli.ord P. G. Wenthold W. C. Lineberger G. A. Petersson K.M. Broadus S. R. Kass S. Kato C. H. DePuy V.M. Bierbaum and G. B. Ellison J. Phys. Chem. A 1998 102 7100. 107 E. P. Cli.ord P. G. Wenthold R. Gareyev W. C. Lineberger C. H. DePuy V. M. Bierbaum and G. B. Ellison J. Chem. Phys. 1998 109 10 293. 108 B. Roemer G. G. Gatev M. Zhong and J. I. Brauman J. Am. Chem. Soc. 1998 120 2919. 109 V. F. DeTuri and K. M. Ervin Int. J. Mass Spectrom. Ion Processes 1998 175 123. 110 Values are available on-line at the following internet address webbook.nist.gov. 111 I. Koppel J. Koppel P.-C. Maria J.-F. Gal R. Notario V. M. Vlasov and R. W. Taft Int. J. Mass Spectrom. Ion Processes 1998 175 61. 112 P. R. Higgins and J. E. Bartmess Int. J. Mass Spectrom. Ion Processes 1998 175 71. 113 M. Schlosser F. Mongin J. Porwisiak W. Dmowski H. H. Buker and N. M. M. Nibbering Chem. Eur. J. 1998 4 1281. 114 T. Matsumoto S. Kobayashi M. Mishima and Y. Tsuno Int. J. Mass Spectrom. Ion Processes 1998 175 41. 115 G. Bouchoux J. F. Gal P. C. Maria J. E. Szulejko T. B. McMahon J. Tortajada A. Luna M. Yanez and O. Mo J. Phys. Chem. A 1998 102 9183. 116 F. Wang S. Ma D. Zhang and R. G. Cooks J. Phys. Chem. A 1998 102 2988. 117 S. Ma F. Wang and R. G. Cooks J. Mass Spectrom. 1998 33 943. 118 N. Srinivasan R. G. Cooks and P. B. Shepson Rapid Commun. Mass Spectrom. 1998 12 328. 119 E. D. Raczynska Pol. J. Chem. 1998 72 1215. 371 Annu. Rep. Prog. Chem. Sect. B 1999 95 349—372 120 E. D. Raczynska P.-C. Maria and J.-F. Gal Analyst 1998 123 621. 121 E. D. Raczynska and M. Mishima Bull. Chem. Soc. Jpn. 1998 71 2175. 122 M. L. Reyzer and J. S. Brodbelt J. Am. Soc. Mass Spectrom. 1998 9 1043. 123 J. Shen and J. Brodbelt J. Mass Spectrom. 1998 33 118. 124 J. E. Szulejko T. B. McMahon V. Troude G. Bouchoux and H. E. Audier J. Phys. Chem. A 1998 102 1879. 125 M. Decouzon J. F. Gal J. C. Guillemin and P. C. Maria Int. J. Mass Spectrom. Ion Processes 1998 175 27. 126 M. L. Chabinyc and J. I. Brauman J. Am. Chem. Soc. 1998 120 10 863. 127 F. E. Wilkinson M. Peschke J. E. Szulejko and T. B. McMahon Int. J. Mass Spectrom. Ion Processes 1998 175 225. 128 J. A. Stone and M.D. Carter Int. J. Mass Spectrom. 1998 180 1. 129 H. Wincel Int. J. Mass Spectrom. Ion Processes 1998 175 283. 130 M. Meot-Ner S. Scheiner and W. O. Yu J. Am. Chem. Soc. 1998 120 6980. 372 Annu. Rep. Prog. Chem. Sect. B

ISSN:0069-3030    

DOI:10.1039/a806685j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Theoretical organic chemistry 11 Ian L. Alberts EMBL Outstation European Bioinformatics Institute,Wellcome Trust Genome Campus Hinxton Cambridge UK CB10 1SDand Department of Biological Sciences University of Stirling Stirling UK FK9 4LA 1 Introduction The .eld of theoretical organic chemistry has progressed rapidly in recent years due in particular to signi.cant advances in the underlying theoretical methodology and computing speeds. The number of publications appearing in the theoretical organic chemistry literature is ever increasing each year and 1998 was certainly no exception. This report provides a highly selective review of the literature which is no doubt in.uenced by the interests of the author and concentrates on two general areas theoretical advances and applications of quantum chemistry methods.Before proceeding it is important to recognise the Encyclopedia of Computational Chemistry that was published in 1998. This .ve volume set provides an excellent and comprehensive treatment of the .eld that should be of use to the general chemist as well as to practitioners of the subject. 2 Theoretical advances Quantum mechanical (QM) methods are used for calculating the electronic structure of small molecular systems including stable molecules reactive intermediates and transition states with very high accuracy. The study of large hydrated systems particularly biological molecules where the aim is enhanced insight rather than quantitative accuracy has been the realm of classical force-.eld based simulations.However this distinction is no longer so clearly pronounced. With the advent of methods that scale more favourably with system size and the combination of QM potentials with empirical force .elds and continuum based protocols it is possible to use QM based procedures to determine thermodynamic quantities and structural information for larger complex systems. This section reviews recent advances in quantum chemistry methodology particularly the evolution of density functional theory the feasibility of ‘large’ molecule studies and the incorporation of solvation e.ects. 373 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Density functional theory Density functional theory (DFT) has emerged as perhaps the method of choice for the solution of many chemical problems.This is becauseDFT methods have relatively low computational cost include a signi.cant (though as yet unspeci.ed) amount of the dynamic electron correlation they are universally applicable to all molecular systems including transition metal complexes and give results of at least MP2 quality and sometimes as accurate as CCSD(T). Further details of the DFT approach can be seen in a recent edited book which highlights many of the achievements of the method and suggests future directions. Nevertheless DFT should not be oversold. It does have problems. One of the main di.culties of DFT is that there is currently no clear approach to systematically improve the exchange-correlation energy functionals however this is an area of signi.cant ongoing research.— DFT may also give a poor description of certain reactions such as hydrogen abstraction reactions. Considering recent advances in DFT Gra� fenstein et al.describe the development of a DFT scheme for open shell singlet (OSS) states within the restricted open shell formalism. The restricted open-shell singlet (ROSS) DFT method provides the correct form of the wavefunction for OSS states and accounts for dynamical electron correlation at the density functional level. ROSS-DFT gives reasonable geometries vibrational frequencies and singlet—triplet splittings for simple systems. However the method does not account properly for spin polarisation and in conjugated systems where this is important ROSS-DFT overstabilises the triplet state relative to the OSS by 2—4 kcal mol.DFT has also been extended for the determination of single photon vertical excitation energies and other electronic response properties. Conventional DFT schemes often give a poor description of these response properties however newer functionals with a better asymptotic form can yield more accurate results for small molecules. Large conjugated systems are again problematic. Most current functionals fail to compute reliable polarisabilities and hyperpolarisabilities for such systems, although hybrid HF—DFT methods can give reasonable values and further developments in this area are underway in a number of laboratories.DFT is also useful on the conceptual side as it can provide insights into chemical behaviour and functionality through the determination of physically important quantities such as hardness softness Fukui functions and chemical potential that arise readily out of the density functional formulation. Large scale systems Computational schemes have recently been developed that overcome the scalability problem in electronic structure calculations of large systems.The algorithms devised for linear scaling of semi-empirical HF and DFT methods with system size avoid the bottlenecks of the most computationally intensive steps construction of the Fock matrix (both Coulomb and exchange terms) and Fock matrix diagonalisation. For a more comprehensive review of these methods see last year’s report and the references therein. Hierarchical multipole based methods can be used to compute the Coulomb matrix for large scale systems without signi.cantly a.ecting the numerical accuracy.Schweg- 374 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 ler et al. describe a new multipole acceptability criterion to estimate the errors incurred in truncating the multipole expansion at .nite order. The error estimates are important for assessing the success of linear scaling techniques. The fast multipole method has also been implemented for periodic systems to speed up computation of the Coulomb term as an alternative to the Ewald summation method. The linear scaling capabilities of this scheme were established in calculations on periodic NaCl systems and carbon nanotubes. Bates et al. compare the computational e.ciency of two alternative methods to the traditional diagonalisation procedure the Chebyshev expansion method and the conjugate gradient density matrix search method. These procedures are implemented for a tight-binding Hamiltonian and applied to very large icosahedral fullerene systems ranging from C to C .The methods are found to have similar computational requirements for attaining comparable accuracy in calculations of large molecular systems and con.rm the practicability of linear scaling in terms of memory and CPU requisites. To achieve linear scaling these methods rely on the sparsity of the density matrix (DM) in order to screen out elements below a certain threshold. However there are interesting cases that have a nonsparse DM such as metals and semiconductors where the band gap is small and nonlocal DM elements may not be negligible.For these situations Baer and Head-Gordon describe a method for coping with a nonsparse DMwithin a tight-binding model. The scheme involves decomposing theDMinto a sum of terms such that each term in the series describes a smaller energy interval than the previous term and corresponds to a sparse matrix. Tests on a carbon nanotube and a 2D-puckered sheet polysilane semiconductor which have nonlocal DMs show that near linear scaling can be achieved. Pan et al. report the implementation of a parallel version of the divide and conquer scheme for semiempirical QM calculations. This scheme which divides a large system into a series of small subsystems and requires independent diagonalisation of the small Fock matrices of the subsystems has an appropriate structure for parallelisation.Each step in a geometry optimisation procedure usually has negligible cost compared to that of computing the energy and gradient with quantum mechanical methods. However with the advent of linear scaling procedures for large molecules the geometry optimisation process could become the limiting factor oan overall structure determination. To overcome this problem Farkas and Schlegel describe a method that reduces the scaling from N to N where N is the number of atoms or electrons by speeding up the required co-ordinate transformations. Local MP2 (LMP2) methods have been devised in a basis of nonorthogonal orbitals in which the cost of the calculation is reduced to about N (compared to N for traditional MP2) by neglecting excitations to distant virtual orbitals. In the method implemented by Maslen and Head-Gordon the storage of local excitation amplitudes is not required and so their approach is suitable for extension to MP4- SDTQ and CCSD(T). Relatively large scale LMP2 calculations with small errors have been reported involving up to 1000 basis functions. Local MP2 energy gradient schemes have also been implemented and shown to yield geometries very close to those from conventional MP2 calculations. Beachy et al.describe a parallel version of LMP2 using pseudospectral numerical methods which should be applicable to even larger systems. Another approach that is primarily designed for large scale calculations the ‘‘resolution of the identity’’ MP2 (RI-MP2) has been applied to several test 375 Annu.Rep. Prog. Chem. Sect. B 1999 95 373—394 cases. In the RI-MP2 procedure the AO pairwise product space is described by a linear combination of .tting basis functions thus fewer integrals need to be calculated leading to memory and time savings. The method has been used to compute reaction energies electrostatic properties H-bonding and van der Waals interactions and is found to yield results that are within 1% on average of those from conventional MP2. Finally the performance of a parallel version of the Gaussian program suite for large scale quantum chemistry calculations has also been reported. Solvation Solvation e.ects can have a dramatic in.uence on chemical reactivity therefore much e.ort in recent years has focused on the development and application of solvent models.These can typically be divided into two classes continuum and discrete solvent approaches. Continuum or reaction .eld models are computationally more e.cient they readily include electrostatic polarisation e.ects but lack explicit interactions between solvent and solute such as H-bonding. The alternative discrete models include speci.c solvent—solute interactions and hydrophobic e.ects explicitly however they are computationally much more demanding and polarisation e.ects are not so conveniently incorporated. As well as solvation the catalytic e.ect of enzymes and the in.uence of the surrounding enzymatic structure can also be explored with discrete models and examples of such studies are increasingly appearing in the literature.In this section recent advances in both continuum and discrete treatments of the condensed phase will be addressed. Continuum models. In this approach the solvent is treated as a continuous polarizable medium which completely immerses a cavity containing the solute. The solute possibly supplemented by a small number of explicit solvent molecules can be treated quantum mechanically and solute—solvent interactions are included through an interaction potential which perturbs the solute QM Hamiltonian. Self-consistent reaction .eld (SCRF) methods have been devised to treat the solvent electrostatic response including apparent surface charge models (ASC), multipole expansion methods (MPE) and methods based on the generalised Born approximation (GB). The nonelectrostatic contributions (cavitation dispersion and repulsion) can also be included to give solvation free energies.It should be noted that the entropic contribution to these free energies due to thermal motions of the molecules is not explicitly included in continuum based models although their e.ect is believed to be limited. Solvation free energies can be calculated at HF post-HF and DFT levels of theory. Several procedures have been devised for generating a molecular-shaped cavity in reaction .eld theory and the important issue of the cavity size has also been addressed. The solute cavity can be de.ned by the union of spheres associated with all solute atoms exposed to solvent and schemes have been introduced to assign the radii for exposed solute atoms based on chemical considerations. The e.ect of varying the radii on the predicted electrostatic solvation energy of neutral and ionic species has been shown to be signi.cant and so it is important to use suitable values. Cavities have also been de.ned by a self-consistent isodensity surface and the optimal value of the solute isodensity contour has been explored by Zhan and Chipman by calibra- 376 Annu.Rep. Prog. Chem. Sect. B 1999 95 373—394 ting SCRF calculations against experiment for the electrostatic contribution to conformational free energy di.erences in solution. However a single cut-o.is probably not appropriate for all molecular species and di.erent values may be appropriate for di.erentQM methods and basis sets. Amovilli et al. provide a comprehensive review of the polarizable continuum model (PCM) and its modi.cations which describe solvation within the ASC framework and are applicable for dielectrics of very di.erent nature. The PCM scheme has been extended to treat errors associated with ASC methods including discretization errors that depend on the cavity surface partitioning and arise from the use of discrete point charges to describe the solvent reaction .eld. Methods have also been described to treat the tail of the QM solute charge distribution that penetrates outside the cavity. Klamt et al.report re.nements of the conductor-like screening model (COSMO) and its extension to real solvents for the calculation of solvation free energies and chemical potentials. In test calculations using the DFT version of COSMO the chemical potentials of 217 neutral molecules were predicted to an accuracy of about 0.4 kcal mol. Truhlar and co-workers describe further developments of the SM5 set of models for solvation free energy computations. These models use the generalised Born approximation to evaluate the reaction .eld and nonelectrostatic contributions are given by parametrised atomic surface tension terms. The latest generation of models is based on HF and DFT SCRF calculations and uses class IV charges derived from semiempirical mapping to the QM electrostatic potential to model electrostatic solute —solvent interactions.The e.ect of di.erent partial charge assignments on the calculation of aqueous solvation free energies using the GB/SA continuum model has been studied by Reddy et al. Charge sets derived from electrostatic potential .tting to ab initio HF wavefunctions yielded better solvation free energies than Mulliken charge sets. Further improvements were found by using charge sets derived from MP2 orGVBcorrelated methods. For the semiempirical methods examined class IV charge sets derived from AM1 computations yielded the best solvation free energies. Charge sets ascribed by the OPLS force .eld gave the best values of all the standard force .elds employed in the study (perhaps because the GB/SA model was parametrised for the OPLS force .eld).Analytic gradients and second derivatives of the solvation free energy with respect to nuclear displacements have been implemented within the continuum approach for both closed and open-shell systems. These procedures allow reaction pro.les to be studied in solution. For example solvent e.ects on the ClCH ClClCH Cl type I S 2 reaction were examined at the DFT level of theory. The ion—dipole complex Cl· · ·CH Cl and the transition state for the reaction were located in vacuo and in solution and all stationary points were characterised by diagonalising the analytical Hessian matrices. The reaction pro.le for the aqueous phase has a unimodal shape in contrast to the double-well potential energy pro.le of the gas phase reaction.The di.erence between the calculated activation barriers in water and in vacuo is found to be in excellent agreement with experiment. The polarisable continuum model has also been combined with molecular dynamics. Using a molecular mechanics (MM) treatment of the solute dynamics simulations were conducted to analyse the conformational distribution of an alanine dipeptide system in aqueous solution. This procedure could be useful for locating the global 377 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 minimum conformation of solvated molecules. Another well known approach for modelling electrostatic interactions involves solution of the Poisson—Boltzmann (PB) equation. This method has been used for many years to calculate solvation energies and pK shifts in biological macromolecules.The PB equation however is known to be inapplicable beyond monovalent ion solutions. To overcome this limitation Tomac and Gra� slund describe a modi.ed PB approach to facilitate the treatment of higher valency ion solutions that are important in biological systems. Discrete solvent models. Discrete solvent models provide an explicit description of solvent—solute interactions and they can often be combined with molecular dynamics (MD) or Monte Carlo (MC) simulations for the calculation of thermodynamic quantities. In these models the solvent molecules are described by empirical force .elds whereas the solute can either be treated classically or quantum mechanically.A recent example of the classical solute approach involves an MD study of the conformational preferences of dicarboxylic suberic acid HOOC(CH ) COOH in water and methanol solution. This study also found good agreement between theoretical C chemical shielding shifts determined at the B3LYP/6-311G(2d,p) level of theory using a timeaveraged conformation from the MD simulations and corresponding values derived from NMR experiments. Combined quantum mechanics—molecular mechanics (QM—MM) methods have proved to be useful for the study of complex phenomena including enzymatic reactions and organometallic catalysis as well as solvation. In this procedure the system is partitioned into a smallQMregion which could be the solute or reacting species with the rest of the system comprising the solvent or surrounding protein atoms treated classically.The interactions between the QM and MM atoms consisting of electrostatic and van der Waals terms are included in the e.ective Hamiltonian of the system. Several excellent reviews of theQM—MMapproach can be found in the Encyclopedia of Computational Chemistry. For the treatment of a small molecule in solution the partitioning is straightforward the solute is in the QM region and the solvent molecules in the MM region. However if the reactive species is part of a large molecule such as the active site of an enzyme then parts of the same molecule will be in di.erent regions. This leads to a partitioning problem since certain QM and MM atoms will be joined by covalent bonds and the electron density along these bonds has to be terminated.Several methods have been devised to overcome this problem including the link atom approach and schemes that utilise hybrid orbitals and local bond orbitals. In the Encyclopedia of Computational Chemistry, Froese and Morokuma review the ONIOMmethod. This is an embedding scheme in which the system can be partitioned into superposed layers and lower level potentials are employed as the distance of the layer from the core increases. These ‘onion’ models can combine QM and MM potentials as well as higher and lower level QM methods. Appropriate use of capping atoms and junctions atoms in the di.erent calculations can overcome the partitioning problem. Classical force .elds only include polarisation in an implicit sense however for the solvation of charged or polar systems the explicit inclusion of polarisation may be important.Several models have been developed for introducing polarisation e.ects 378 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 including the .uctuating charge (FQ) model of Rappe� and Goddard and Rick et al. Liu et al. describe a new polarisable force .eld for water which uses the FQ approach to incorporate electrostatic polarisation and potential parameters for describing many body and pairwise interactions are chosen to give the optimal .t to ab initio QM energies. In hybrid QM—MMmodels polarisation of the solute charge is readily included in the e.ective Hamiltonian but the incorporation of solvent polarisation is more di.cult computationally.Gao describes a molecular orbital derived polarisation (MP) potential for treating solvent polarisation e.ects. In this scheme each liquid molecule is described by a quantum mechanical wavefunction which is coupled to the surrounding classical multipoles of the other liquid molecules in the manner of hybrid QM—MM procedures. This is analogous to a QM version of the FQ model. The potential of Liu et al. and the MP model using the semiempirical AM1 method both give accurate gas and liquid phase properties for water and may be useful for the study of biomolecular systems in solution. Bryce et al. describe a very similar model to the MP approach of Gao in which the FQ scheme is used to model the solvent molecules and a QMwavefunction describes the solute. A full ab initioQM—MMFQ potential was implemented and used to study the dynamics of small .uoride—water clusters.The results emphasised the importance of solvent polarisation for such charged species. The incorporation of polarisation e.ects in the MM region of QM—MM models is currently limited to solvation studies due to the computational cost and remains one of the main de.ciencies of the approach for the study of enzymatic reactions. MD/MC simulations and free energy calculations using QM—MM potentials in discrete solvent studies can be very expensive computationally (particularly when polarisation e.ects are considered) and thus a semiempirical Hamiltonian is often the only choice for the QMsolute in such calculations. The advent of new linear-scaling techniques may overcome this limitation in the future.The generalised molecular interaction potential (GMIP) described by Luque and Orozco may also be useful in this respect. The GMIP accounts for electrostatic interactions between theQMand MM atoms dispersion—repulsion interactions and a perturbative treatment of polarisation and may be convenient as an e.ective Hamiltonian for the QM—MM interaction energy in combined simulations. Inclusion of polarisation e.ects with this potential requires only one-electron integrals and thus may allow a signi.cant computational saving compared to SCF-based procedures which require the computation of two-electron integrals for the polarisation terms.The QM—MM model has been invoked in several investigations of enzymatic reactions. For example Ridder et al. used the semiempirical AM1QM—MMpotential to determine the activation energy of the hydroxylation reaction of p-hydroxybenzoate and several .uorinated analogues within the active site of the enzyme phydroxybenzoate hydroxylase (PHBH). The use of semiempirical QM methods is known to signi.cantly overestimate absolute energy barriers although relative barriers may reveal correct trends. In this study Ridder et al. observed a linear correlation between the calculated barriers and experimental log k values. Glennon and Warshel conducted simulations of the catalytic reaction of ribonuclease A using the empirical valence bond method (EVB) combined with the free energy perturbation (FEP) technique. This study illustrates the catalytic power of the enzyme and suggests that it originates in electrostatic stabilisation of the transition states.Hybrid QM—MM 379 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 studies of enzymatic reactions are usually limited to minimisations or explorations of reaction mechanisms. The EVB approach however can be used for free energy calculations as it invokes the cheaper EVB Hamiltonian to describe the QM moiety but it it as straightforward to employ as the Hamiltonian parameters need to be adjusted for each distinct application. 3 Applications The chemical applications reviewed below range from accurate electronic structure calculations of small molecules to the simulation of complex solvated systems.Several of the applications involve radical species which often require relatively high levels of electron correlation for a reliable theoretical treatment due to problems associated with symmetry breaking and spin contamination. Thermochemistry The calculation of accurate thermochemical quantities is an important aspect of electronic structure theory and developments in this area have recently been reviewed. Gaussian-2 (G2) theory and the complete basis set (CBS) methods were developed in order to obtain accurate enthalpies of formation. G2 assumes certain additivity approximations to yield .nal energies e.ectively at the QCISD(T)/6- 311G(3df,2p) level of theory and an empirical correction is applied to account for systematic errors.The CBS methods involve an extrapolation of the MP2 energy to the complete basis set limit. The e.ect of high order electron correlation is determined by conducting calculations with smaller basis sets and empirical corrections are applied to account for zero point energy spin contamination in UHF wavefunctions and other correlation e.ects. QCISD(T) is the highest correlated wavefunction used. Three CBS methods have been proposed CBS-Q which is the most accurate and CBS-qand CBS-4 which involve calculations with smaller basis sets and lower levels of theory respectively. Petersson et al. assessed the CBS G2 and B3LYP methods for determining the enthalpies of formation of the G2 neutral test set of 148 molecules.After introduction of corrections for spin-orbit interactions the average absolute deviations from experiment for the G2 G2(MP2) CBS-Q G2(MP2,SVP) CBS-q CBS-4 and B3LYP/6-311G(3df,2p)//B3LYP/6-31G* methods (in order of increasing speed) are 1.43 1.76 1.19 1.64 2.34 2.66 and 3.43 kcal mol respectively. The maximum deviations are 10.6 8.8 8.1 9.4 11.4 12.9 and 24.1 kcal mol respectively and in each case a tetrahalide species is responsible for the maximum error. Anomalously large errors were originally reported for but-2-yne and naphthalene using the CBS models, however these were found to be due to basis set near linear dependencies which can be removed by consideration of the eigenvalues of the overlap matrix. Furthermore the latest DFT functionals give a mean average error of 1.60 kcal mol and a maximum deviation of 8.65 kcal mol which are close to the G2 values of 1.58 and 8.2 kcal mol respectively (without the spin-orbit correction).These results again demonstrate the usefulness of the computationally simpler density functional approaches. In general the errors tend to scale with the size of the molecule and assuming 380 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 constant errors in calculated bond energies for each bond type empirical bond additivity corrections could be used to improve the models. These empirical parameters were obtained for several bond types and their inclusion leads to a reduction in the average deviation of all the models by a factor of 2—3 and reduces maximum errors to about 3 kcal mol.The bond additivity corrections are particularly important in calculations of larger molecules for example the CBS-4 errors for the formation enthalpies of benzene and naphthalene are reduced from 10.5 and 17.5 kcal mol to 2.1 and 1.6 kcal mol respectively upon their inclusion. It is clear that the accuracy of these models varies much less than the computational speed and applicability as the size of system increases. The best method to choose is the most accurate one that is practical to use for the speci.c application. In this respect Feller and Peterson have compiled theoretical predictions of thermochemical quantities into a database to assist in the choice of basis set and method to achieve the required accuracy for a particular application. Mayer et al. performed a detailed analysis of the calculation of thermochemical properties of free radicals using Martin’s in.nite basis set extrapolation techniques, CBS and G2 methods with unrestricted reference wavefunctions.The best heats of formation are provided by Martins techniques since they extrapolate the CCSD(T) energy to the in.nite basis set limit. Several corrections are also added to account for core-correlation zero point energy spin-orbit coupling and the e.ect of triple bonds or cumulated double bonds. These techniques are computationally too demanding for most practical purposes but they provide benchmark heats of formation for the four radicals examined in this paper. Using the three families of procedures the calculated heats of formation are in good agreement when the spin contamination is low but G2 is inferior to the CBS models in comparison to Martin’s techniques when the spin contamination increases.A modi.cation of CBS-Q termed CBS-RAD in which single point energies are determined at the CCSD(T) level of theory rather than QCISD(T) gives the best heats of formation relative to Martins techniques. CBS-RAD is the most practical procedure recommended by the authors for studying free radicals. The combination of B3LYP/6-31G* geometries and vibrational frequencies with CBSRADenergies has been applied to several radical additions to alkenes and to the ring opening reaction of the cyclopropylcarbinyl radical and its nitrogen- and oxygensubstituted analogues. Predicted reaction energies and activation barriers were found to be in good agreement with available experimental data.The authors suggest the use of B3LYP with a large basis set for the study of systems which are not feasible with the CBS-RAD procedure. It is useful to report some applications of other model chemistries and extrapolation techniques. Murphy and Friesner applied the J1 method to examine the relative energetics of C carbon cluster isomers. J1 is based on the localised pseudospectral methods LMP2 and a multireference LMP2 using a generalised valence bond (GVB) reference wavefunction GVB-LMP2. The J1 procedure involves a GVB-LMP2/ccpVTZ(-f) energy calculation which is amended for basis set e.ects by an LMP2 computation with the cc-pVTZbasis set.Several empirical corrections similar to those employed in Martin’s methods are also added to the resultant energies. J1 gives an average absolute deviation of 1.2 kcal mol for the atomisation energies of the 55 small molecules from the original G2 test set which is the same as that given by G2 theory and better than the error of 2 kcal mol given by B3LYP. G2 is not a viable 381 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 method for a system as large as C and B3LYP and BLYP DFT methods give an incorrect energy ordering for the bowl ring and cage isomers of C . In contrast LPMP2 and GVB-LMP2 are successful in predicting the correct isomeric energy ordering and the J1 procedure gives accurate relative energies.Csa� sza� r et al. conducted a systematic examination of the conformation energy pro.les of several small molecules by extrapolating basis set and electron correlation energy contributions to the ab initio limit. Empirical corrections were also appended to relative energies to account for core-correlation relativistic e.ects and the Born—Oppenheimer approximation. Using this focal point scheme the .nal predictions for the classical inversion barriers ofNH H Oand HNCO,the torsional barrier of ethane and the E—Z isomerisation energy of formic acid are in excellent agreement with experiment. The dissociation barrier for A ketene has also been studied with the focal point method and the .nal prediction was about 200 cm (0.6 kcal mol) above experiment. The work details useful strategies for achieving high accuracy in the prediction of conformational energy di.erences and activation barriers.(1) (2) Isomerisation and hydrogen bonding Mineva et al. examined the e.ect of solvent he cis—trans isomerisation of 1,2- diazene using the PCM method in the framework of DFT. The gas phase calculations predicted that the trans-isomer was 4.7 kcal mol more stable than the cis form. Solvent e.ects may be expected to decrease this energy gap by preferentially stabilising the cis form as a result of its dipole moment. However the energy di.erence increased to 7.3 kcal mol on the inclusion of solvation free energies probably due to the greater quadrupole moment of the trans-form.In terms of the cis—trans isomerisation process the two processes shown in (1) and (2) were studied. trans-N H H ON H OHcis-N H H O trans-N H H ON H H Ocis-N H H O Despite the large solvation energy of the charged species the formation of N H OH in reaction (1) was found to be endothermic in solution by 43.5 kcal mol and thus the isomerisation is not expected to take place via this process. Considering reaction (2) in the gas phase proton transfer from H trans-N H was predicted to be endothermic by H is known to decompose rapidly. In solution the H is almost an isoenergetic O to H to form N H was found to be exothermic by 19.2 kcal mol however subsequent deprotonation to give cis-N 23.9 kcal mol.Therefore this is not a feasible isomerisation process in the gaseous state and in fact gaseous trans-N initial proton transfer to give N H from trans-N H is endothermic by 7.1 kcal mol and the deprotonation to give cis-N process. Thus it is feasible that the isomerisation of diazene can take place in solution by proton exchange. Amino acid radicals are important in biochemical reactions and their properties have been explored in detail. Rega et al. used DFTto study the structural and energetic features of glycine radicals derived by homolytic .ssion of the C—H bond. The zwitterionic form of the glycine radical (NH CHCOO) is found to be 50.4 kcal mol above the neutral form (NH CHCOOH) in vacuo but in solution the 382 Annu.Rep. Prog. Chem. Sect. B 1999 95 373—394 energy di.erence is reduced to 24.5 kcal mol. The zwitterion is therefore not the preferred form in solution unlike the situation for the parent amino acid. This is due to captodative stabilisation present in the neutral form of the radical but not the zwitterionic form involving the coupled e.ect of the electron-withdrawing carboxy and electron-donating amino groups that leads to enhanced delocalisation of the unpaired electron. In fact as a result of the captodative e.ect the neutral form of the radical predominates over the cationic radical (NH CHCOOH)even at very low pH.Above pH 10 the glycine anion radical (NH CHCOO) is the prevalent species. This radical has a pyramidal structure and the inversion barrier is predicted to lower from 3.3 to 0.5 kcal mol in aqueous solution. Short strongHbonds have caused much controversy in recent years concerning the role they play in enzymatic reactions and several studies have been conducted to gain further insights into their characteristics. The existence of short strongHbonds is well established in the gas phase. For example at the B3LYP/6-31G(d,p) level of theory the H bond between formic acid and formate ion has an O· · ·O distance r(O · · · O) of 2.434Å and an H bond energy E of 27 kcal mol compared to an r(O · · ·O) of 2.821Å and E of 5 kcal mol between two neutral formic acid molecules.On the introduction of substituents in a series of enol—enolate complexes 1 where R and R NH H F and CN a linear relationship was found between the calculated H bond energy and the proton a.nity di.erence between the monomer donor and acceptor (PA) i.e. E decreases as PA increases. E is sensitive to geometry changes and an inverse linear relationship is observed between E and r(O · · ·O) for these complexes. The strength of theHbond can also be in.uenced by the environment and via SCRF calculations the energy di.erence between the ionic H bonds and traditional neutral H bonds (E ) was found to fall o. quickly as the solvent relative permittivity () was increased.For example increasing from 1 to 35 leads to a 13 kcal mol decrease in E for the formic acid—formate system. For short strong H bonds to have an important stabilising role in enzymatic catalysis it is clear that the active site must provide a low dielectric medium. These studies do not resolve the controversy regarding short strong H bonds but they provide some indication of the conditions which should be met if such hydrogen bonds are to be of signi.cance in enzymatic catalysis. Reaction pathways The following sections focus on theoretical studies of organic reactivity via exploration of potential energy surfaces and the calculation of transition states and associated activation barriers.Cycloaddition reactions The classic Diels—Alder (DA) reaction between butadiene and ethylene was revisited 383 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 by Karadakov et al. using a spin-coupled form of the valence-bond method in combination with a CASSCF wavefunction. This study con.rmed the concerted synchronous nature of the reaction via an aromatic-like transition state. DA reactions involving unsymmetrical dienophiles and/or dienes continue to be the focus of much theoretical investigation. For example the BF Lewis acid catalysed DA reaction of butadiene and acrolein has been examined using the B3LYP/6-31G* method. The calculations predict a concerted asynchronous mechanism with a preference for the endo s-cis transition state. The catalyst increases the asynchronicity of the process and signi.cantly lowers the activation barrier due to stabilising BF —acrolein interactions in the transition state.Interestingly for the catalysed reaction the inclusion of electron correlation in the calculations leads to the transition state for the standard [42] DA reaction rather than for the alternative [24] hetero-DA reaction that had been suggested in earlier studies. Salvatella et al. employed a combinedAM1—MM3method to study theDAreaction between cyclopentadiene (CP) 2 and an acrylate with bulky substituents 3,3- dimethyl-2-butyl acrylate (DBA) 3. DBAhas the formCH ——CHCOORand theO—R bond was chosen to partition the system into QM and MM regions such that the diene and the CH ——CHCOO group of the ester are described quantum mechanically by AM1 while the steric e.ects of the bulky substituent groups are modelled classically.Assuming endo approach of CP to the s-trans conformation of DBA (Scheme 1) the pure AM1 method gives a slight preference for attack at the hindered face of the dienophile in discord with steric considerations and experimental results. In contrast the combined semiempirical AM1—MMapproach which provides a better treatment of steric e.ects gives the correct diastereofacial selectivity for this reaction involving attack at the free face. Substituents can in.uence the concerted/stepwise mechanistic preference for DA reactions. For example at the B3LYP/6-31G* level of theory the DA reaction of s-cis butadiene and triazolinedione (TAD) was found to prefer the asynchronous concerted pathway involving the endo transition state leading to stereospeci.c DA products. This is in agreement with experimental observations for less substituted dienes.The preference for the endo transition state is due to relief of electrostatic repulsions between the butadiene system and the TAD nitrogen lone pairs that destabilise the exo transition state. The asynchronous concerted pathway is predicted to be only 2.9—4.0 kcal mol below the alternative stepwise mechanism. Thus for highly substituted butadienes which are too sterically hindered to adopt the s-cis conformation the stepwise mechanism is expected to be favoured which can lead to a mixture of products again in accord with experimental observations.The 1,3-dipolar cycloaddition reaction of methyl azide and ethene was predicted to follow a concerted synchronous process according to HF and B3LYP calculations. Solvation e.ects for this reaction were modelled by MC and FEP calculations and it was found that rate variations in several solvents were due to a combination of polar hydrogen bonding and hydrophobic e.ects. Liu et al. assessed the stereoselectivities in addition processes by examining the DA and 1,3-cycloaddition reactions of several systems (butadiene acrolein nitrosoethylene and methylenenitrone) with methyl vinyl ether. In all cases calculations at the B3LYP/6-31G* level of theory show a dramatic change in the conformation of the enol ether from s-cis in the reactant to s-trans in the transition state (Scheme 2).This conformational switch was con.rmed by related 384 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 1 Scheme 2 experimental studies. The vinyl ether reactant adopts the s-cis form to relieve electrostatic repulsions between the oxygen lone pairs and the alkene system. However in the transition state the s-trans form is preferred primarily due to favourable electrostatic interactions between the vinyl ether oxygen lone pairs and the partial positive charge that develops at the alkene moiety. Polar solvents signi.cantly reduce this preference since the s-cis transition state has a larger dipole moment. [22] Cycloaddition reactions leading to four-membered rings are forbidden according to orbital symmetry rules and usually follow a stepwise mechanism.The [22] dimerisation of silaethylene was found to proceed in this fashion involving the formation of CH —SiH —SiH —CH and CH —SiH —CH —SiH biradicals via headto-head and head-to-tail mechanisms respectively (Scheme 3). The former process is almost barrierless and favoured over the latter by 2.7 kcal mol at the CASPT2/6- 31G*//CASSCF/6-31G* level of theory. The CH —SiH —SiH —CH biradical leads to 1,2-disilacyclobutane 4 by a fast rotation—recoupling however a competitive [1,2] silyl sigmatropic shift with a similar activation barrier via a rhomboidal transition state was identi.ed that can lead to the 1,3-product 5.In contrast the cycloaddition reactions of SO with alkenes (Scheme 4) are predicted by B3LYP/6-31G* calculations to follow a concerted asynchronous pathway with zwitterionic character in the transition state 6 rather than a stepwise mechanism via a zwitterionic intermediate 7. Polar solvents reduce the activation barrier as a result of the zwitterionic character of the transition state. This preference for a pathway that is normally forbidden is attributed to the high polarity in SO which localises the HOMO on oxygen and the LUMO on sulfur. Frontier orbital energy level crossing during the concerted process is therefore e.ectively removed and the pathway is no longer forbidden. An alternative formally allowed [32] cycloaddition pathway that is favoured for the reaction of OsO with alkenes was found to have an activation barrier signi.cantly above that for the concerted [22] process.This di.erence between alkene sulfonation and osmylation 385 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 3 Scheme 4 is supported by experimental evidence and can again be explained on the basis of frontier orbital interactions. Rearrangement reactions After many years of experimental and theoretical study it is now established that the Cope [3,3] rearrangement of hexa-1,5-diene has a concerted mechanism involving an aromatic transition state which is slightly favoured over a diradical mechanism. The anionic oxy-Cope reaction of deprotonated 3-hydroxyhexa-1,5-diene (Scheme 5) was also predicted to be a concerted process at the B3LYP/6-31G* level of theory.This reaction experiences an enormous rate acceleration compared to the protonated species due to weakening of the C—C bond adjacent to the oxy group. In contrast the anionic amino system rearranges in a stepwise mechanism via an intermediate allyl anion—acrolein imine complex 8. In solution the intermediate complex is expected to be stabilised leading to dissociation products rather than the Cope product in accordance with experimental observations. Homolytic and heterolytic C—C bond cleavage models were invoked to understand the reactivity di.erences between these anionic oxy- and amino-systems. It is clear that most allowed pericyclic reactions involving simple hydrocarbons follow a concerted pathway rather than a stepwise diradical mechanism and it is usually substituent e.ects that are responsible for changes in the mechanistic preference.An example of a formally allowed pericyclic reaction that may favour the 386 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 5 R O NH. Scheme 6 stepwise process even without substituent e.ects is the [5,5] sigmatropic shift. This preference was con.rmed by B3LYP/6-31G* calculations for the [5,5] shift of (Z,Z)- deca-1,3,7,9-tetraene (Scheme 6). A stepwise diradical mechanism was predicted with an activation barrier of 34.3 kcal mol. Interestingly in order to compare isoelectronic [3,3] Cope and [5,5] reactions the Cope rearrangement of 3,6-bismethyleneocta-1,7-diene was explored.This Cope shift was also predicted to prefer a stepwise pathway with a lower activation barrier than for the [5,5] shift of (Z,Z)- deca-1,3,7,9-tetraene due to reduced steric interactions in the transition state. Makita et al. report investigations of the Stevens rearrangement of phosphorus and arsenic ylides [reaction (3)] ZH MCH H MCH Z (Z H CH CH—— CH SiH and GeH ; M P and As) (3) using Mo� ller—Plesset theory up to fourth order. In particular the concerted migration mechanism was compared with the competing radical dissociation—recombination pathway. For the SiH and GeH reactions the concerted migration activation energies are smaller than the radical dissociation energies due to the hypervalent capability of Si and Ge atoms and thus the former pathway is expected to predominate.For the methyl migration the dissociation energy is smaller than the activation barrier of the concerted process indicating the preference for the radical dissociation pathway. For the H and vinyl migrations the activation energies for the concerted process are slightly lower than the radical dissociation energies. The authors suggest however that entropy considerations would lead to a preference for the radical dissociation pathway. Bettinger et al. examined various fascinating rearrangement processes on the complex potential energy hypersurface of benzene using high level quantum mechanical methods. Several mechanisms were identi.ed for the intramolecular topomerisation reactions or label scrambling of [1,2-C ]- [1,3-C ]- and [1,4-C ]benzene 9.The major pathways include Scheme 7 which involves rearrangements through the intermediates benzvalene 10 and 10 and prefulvane 11 and Scheme 8 which involves a [1,2]-H shift and passes through cyclohexa-2,4-dienylidene 12 and 12 and fulvene 13 387 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 7 Scheme 8 intermediates. These pathways are energetically competitive and likely to contribute to benzene topomerisation at high temperature. Ring opening reactions The ring opening reactions of cycloalkanes have long been a topic of intense theoretical and experimental investigation. The simplest system cyclopropane is thought to have a slight preference for the coupled conrotation process.Skancke et al. used highly correlated (10/10)CASPT2N/6-31G* wavefunctions to investigate the e.ects of geminal silyl substituents on the ring opening process. It was found that replacing the hydrogens at C(1) by electron donating silyl groups leads to a stronger preference for conrotatory ring opening relative to disrotation or monorotation of a single methylene group. Therefore stereomutation of 1,1-disilylcyclopropane is expected to take place primarily via conrotation of methylene groups at C(2) and C(3). In contrast substition of the C—H bonds at C(1) with electron withdrawing C—F bonds leads to a strong preference for the disrotatory pathway in the stereomutation of 1,1-di.uorocyclopropane.This change in the preferred coupled rotation pathway is a consequence of the change of symmetry of the highest occupied MO (HOMO) in the corresponding diradical intermediates. The ring opening reaction and other transformations on the cyclobutene radical cation (CB·) potential energy surface were examined by Sastry et al. using several theoretical methods including DFT QCISD and CCSD(T). Two asynchronous concerted mechanisms both of which involve conrotation of the CH groups were located from CB· 14 to cis- and trans-butadiene radical cation (BD·) 15 and 16 and they are within 1 kcal mol of each other at the RCCSD(T)/cc-pVTZ//QCISD/6- 31G* level of theory. Transition states for the rearrangements cis-BD·bicyclobutane radical cation (BCB·) 17 cis-trans-BD· and CB·BCB· were also located although the latter process is not competitive with ring opening of CB· to butadiene cations.All the transformations considered including symmetry designations are depicted in Scheme 9. The CB·trans-BD· and cis-BD·BCB· rearrangements pass over a .at region of the potential energy surface termed the 388 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 9 ‘Bauld plateau’ in which transition state searches are di.cult. The characters of the stationary points located in this region are sensitive to the level of theory employed but they all seem to consist of a three-membered ring bearing the unpaired electron and an exocyclicCH group carrying the positive charge. The BLYPand B3LYPDFT methods encountered di.culties in locating the cis- to trans-BD· isomerisation transition state since spin and charge localisation in the HOMO takes place during this process which DFT is known to incorrectly model. The transition state is only located with DFT using the ‘half and half’ exchange functional which includes 50% HF exchange density.Nevertheless overall B3LYP performs well for this system. Catalysis The Ziegler—Natta polymerisation of -ole.ns is an important industrial process however the mechanism of the process is still not completely understood. Boero et al. conducted ab initio molecular dynamics simulations within the DFT based Carr—Parrinello approach (CPMD) to investigate the titanium catalysed polymerisation of ethylene on an MgCl surface. The favoured reaction pathway (Scheme 10) was found to involve formation of a metal—ethylene complex 18 followed by ethylene insertion via a four-membered ring transition state 19.The ethylene insertion process is supported by a Ti—H—C agostic interaction which reduces steric repulsions. Insertion of a second ethylene molecule was found to follow a similar pathway. These calculations depict a possible pathway for the chain initiation and subsequent propagation steps of the polymerisation process. Froese et al. studied the diimine—M (M Ni Pd) catalysed polymerisation of ethylene using the B3LYPMM3 combination of the integrated molecular orbital—molecular mechanics (IMOMM) procedure. This study focused on the role of substituents in the catalysed polymerisation reaction.Steric e.ects exerted by bulky substituents were found to lower the migratory insertion activation barriers and raise the energy of the termination steps. The formation of branched polythenes was found to be more likely in the presence of bulky substituents for the palladium case whereas for nickel the branching process was not prominent for the substituted and unsubstituted catalyst. The results 389 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 Scheme 10 (4) and the bidentate ligand Scheme 11 are very similar to those obtained by Ziegler and co-workers using a related DFT—MMapproach. Morokuma and co-workers continued their studies of organometallic reactions in 1998. For example Matsubara et al.examined the .rst step in the highly selective Ru catalysed addition of aromatic carbonyl compounds to alkenes (Scheme 11). B3LYP calculations showed that the selectivity for ortho C—H bond activation is due to coordination of the carbonyl oxygen to the Ru complex which leads to 21 via a .ve-coordinate metallacyclic intermediate 20. The ortho C—H bond cleavage in the latter step is assisted by an agostic interaction in the intermediate. Therefore the rate determining step of the addition reaction is expected to be C—C bond formation with the alkene rather than ortho C—H bond activation. Nakamura et al. examined the S 2 reaction of MeBr with lithium organocuprate reagents including the e.ects of a coordinating (CH ) O solvent molecule. An oxidative addition/reduction elimination pathway was followed involving co-operative copper and lithium interactions and solvent stabilisation of a Cu(...) intermediate.The activation ofO—Hbonds by transition metal catalysts has also been analysed in several studies. For example Su and Chu modelled the activation of the methanol O—H bond with platinum(0) complexes. Speci.cally the oxidative addition reactions (4) PtL H-OCH PtL (H)(OCH ) where the monodentate ligand L CO PH L H PCH PH H PCH CH PH were studied. B3LYP calculations with the LANL2DZ basis set which includes e.ective core potentials show that the oxidative 390 Annu. Rep. Prog. Chem. Sect.B 1999 95 373—394 addition reaction is promoted by a small L—Pt—L angle (allowing the O—Hbond ready access to the metal HOMO) and a good electron-donating ligand (since the reaction involves Ptmethanol charge transfer). In fact the activation barriers for the oxidative addition reactions with Pt(H PCH PH ) and Pt(H PCH CH PH ) which involve electron donating bidentate ligands are very low 3.2 and 1.7 kcal mol respectively leading to favourable reactions. Conversely the barrier height with electron withdrawing ligands in a linear PtL complex is very high for example 40.1 kcal mol for Pt(CO) and such complexes would be more suitable for methanol elimination. Photochemistry Robb and co-workers have conducted a series of studies concerning ultrafast photochemical processes and a comprehensive review of this work has recently been presented.In the latest studies the cis—trans photoisomerisation pathway of the protonated Schi. base (PSB) 4-cis--methylnona-2,4,6,8-tetraeniminium cation 22 a realistic model for a retinal chromophore has been examined by CASSCF computations. Evolution of the excited state from the Franck—Condon region leads to an energy plateau on the S surface involving skeletal deformations. The minimum energy pathway progresses to a conical intersection where S and S are degenerate and then decays rapidly to the trans-photoproduct on S . It was postulated that the energy plateau may be responsible for the relatively slow picosecond timescale observed for the cis—trans photoisomerisation of retinal PSBs in solution.When the retinal chromophore is embedded in the rhodopsin cavity the energy plateau on S may be absent and thus leading to the faster photoisomerisation rate on the femtosecond timescale observed for retinal PSBs in rhodopsin. The photochemical reactions of 2,3-diazabicyclo[2.2.1]hept-2-ene 23 have been studied by CASSCF/MP2 calculations with a 6-31G* basis set. Reaction paths encompassing the potential energy surfaces for the S S T and T states were identi.ed and found to involve intermediate structures transition states and funnels (passages) for internal conversion and intersystem crossing. Finally Freund and Klessinger examined the photochemical processes of singlet ethylene by conducting CASSCF calculations. An S /S conical intersection was located at an unsymmetrical structure which can lead to the products of cis—trans isomerisation and [1,2] hydrogen shift processes as well as the initial reactant depending on the detailed dynamics.Electron transfer processes The study of electron transfer processes in biological systems received much attention 391 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 or imidazole (to model the histidine ligand) YSH or SCH in the 1998 literature. For example Siegbahn et al. used B3LYP to model the electron transfer processes in ribonucleotide reductase (RNR) and the bacterial photosynthetic reaction centre.Interestingly the electron transfer process in RNR is predicted to be endothermic and an alternativeHatom transfer mechanism which does not involve charge separation has been proposed. Roos and co-workers conducted a series of calculations to explore the spectroscopic properties and electron transfer processes in blue copper proteins. Models of the copper() active site including the coordinating ligands were constructed in the form CuX YZ XNH (cysteine thiolate) and ZSH or S(CH) (methionine thioether). B3LYP geometry optimisations were performed and CASPT2 was used to compute the spectra. A simple point charge model was also invoked to simulate the eects of the surrounding protein and solvent on the spectroscopic predictions. Distorted trigonal and tetragonal structures were located for these models and they are energetically very close.Both structures are observed in type I blue copper proteins for example the trigonal structure is found in plastocyanin and azurin (axial proteins) whereas pseudoazurin cucumber basic protein and nitrite reductase promote the tetragonal structure (rhombic proteins). The main dierence between the structures is electronic the axial proteins have a Cu¡XS bond and the rhombic proteins mainly a Cu¡XS bond. Structurally both forms are distorted towards a tetrahedron which is the preferred structure for reduced Cu() complexes. Therefore the change in geometry accompanying reduction of Cu() to Cu() is small leading to a small reorganisation energy and a fast electron transfer process. In spectroscopic terms the axial proteins show an intense peak at 600 nm which is responsible for the blue colour and a low intensity peak at 460 nm.For the rhombic proteins the 460nm band gains intensity at the expense of the 600nm peak. The model calculations show that these are SCu charge transfer bands and the relative intensity is dependent on the character of the Cu¡XS bond. In the trigonal structures it is a highly covalent bond and thus the band at 600nm (*) is prominent as observed for the axial proteins. In contrast the Cu¡XS bonding in the tetragonal structures involves a mixture of and interactions and thus the 460nm band (* * combination) gains intensity as observed for the rhombic proteins. References 1 The Encyclopedia of Computational Chemistry ed. P. v. R. Schleyer H. F. Schaefer and P. R. Scheiner John Wiley and Sons Chichester 1998.2 Electronic Density Functional Theory Recent Progress and New Directions ed. J. Dobson G. Vignale and M.P. Das Plenum New York 1998. 3 F.A. Hamprecht A. J. Cohen D. J. Tozer and N. C. Handy J. Chem. Phys. 1998 109 6264. 4 S. Ivanov R. Lopez-Boada A. GoÆØ rling and M. Levy J. Chem. Phys. 1998 109 6280. 5 A.D. Becke J. Chem. Phys. 1998 109 2092. 6 J. GraÆØ fenstein E. Kraka and D. Cremer Chem. Phys. Lett. 1998 288 593. 7 I. Andrejkovics and AT . Nagy Chem. Phys. Lett. 1998. 296 489. 8 M.E. Casida C. Jamorski K. C. Casida and D. R. Salahub J. Chem. Phys. 1998 108 4439. 9 S. J.A. van Gisbergen J. G. Snijders and E. J. Baerends J. Chem. Phys. 1998 109 10657. 10 E. R. Stratmann G. E. Scuseria and M. J. Frisch J. Chem. Phys. 1998 109 8218. 11 D. J. Tozer and N.C. Handy J. Chem. Phys. 1998 109 10 180. 12 C. Van Caillie and R. D. Amos Chem. Phys. Lett. 1998 291 71. 13 B. Champagne E. A. Perpe` te S. J. A. van Gisbergen E.-J. Baerends J. G. Snijders C. Soubra-Ghaoui K. A. 392 Annu. Rep. Prog. Chem. Sect. B 1999 95 373¡X394Robins and B. Kirtman J. Chem. Phys. 1998 109 10 489. 14 P. Geerlings F. De Proft and W. Langenaeker Adv. Quantum Chem. 1998 33 303. 15 T. Mineva N. Neshev N. Russo E. Sicilia and M. Toscano Adv. Quantum Chem. 1998 33 273. 16 I. Alberts Annu. Rep. Prog. Chem. Sect. B Org. Chem. 1998 94 337. 17 E. Schwegler M. Challacombe and M. Head-Gordon J. Chem. Phys. 1998 109 8764. 18 K. N. Kudin and G. E. Scuseria Chem. Phys. Lett. 1998 289 611. 19 K. R. Bates A. D. Daniels and G. E. Scuseria J. Chem. Phys. 1998 109 3308.20 R. Baer and M. Head-Gordon J. Chem. Phys. 1998 109 10 159. 21 W. Pan T.-S. Lee and W. Yang J. Comput. Chem. 1998 19 1101. 22 O. Farkas and H. B. Schlegel J. Chem. Phys. 1998 109 7100. 23 G. Rauhut P. Puland and H. J. Werner J. Comput. Chem. 1998 19 1241. 24 P. E. Maslen and M. Head-Gordon J. Chem. Phys. 1998 109 7093. 25 A. Al Azhary G. Rauhut P. Pulay and H.-J. Werner J. Chem. Phys. 1998 108 5185. 26 M.D. Beachy D. Chasman R. A. Friesner and R. B. Murphy J. Comput. Chem. 1998 19 1030. 27 D. E. Bernholdt and R. J. Harrison J. Chem. Phys. 1998 109 1593. 28 C. P. Sosa J. Ochterski J. Carpenter and M. J. Frisch J. Comput. Chem. 1998 19 1053. 29 C. Amovilli V. Barone R. Cammi E. Cance` s M. Cossi B. Mennucci C. S. Pomelli and J. Tomasi Adv. Quantum Chem.1998 32 227. 30 V. Dillet D. Rinaldi J. Bertra� n and J. L. Rivaldi J. Chem. Phys. 1996 104 9437. 31 G. D. Hawkins C. J. Cramer and D. G. Truhlar J. Phys. Chem. B 1998 102 3257. 32 B. Mennucci E. Cance` s and J. Tomasi J. Phys. Chem. B 1997 101 10 506. 33 V. Barone and M. Cossi J. Phys. Chem. A 1998 102 1995. 34 M. Cossi B. Mennucci J. Pitarch and J. Tomasi J. Comput. Chem. 1998 19 833. 35 J. B. Foresman T. A. Kieth K. B. Wiberg J. Snoonian and M. J. Frisch J. Phys. Chem. 1996 100 16 098. 36 C.-G. Zhan and D. M. Chipman J. Chem. Phys. 1998 109 10 543. 37 C.-G. Zhan J. Bentley and D. M. Chipman J. Chem. Phys. 1998 108 177. 38 A. Klamt V. Jonas T. Bu� rger and J. C. W. Lohrenz J. Phys. Chem. A 1998 102 5074. 39 T. Zhu J. L. Gregory D. Hawkins C.J. Cramer and D. G. Truhlar J. Chem. Phys. 1998 109 9117. 40 M.R. Reddy M.D. Erion A. Agarwal V. N. Viswanadhan Q. McDonald and W.C. Still J. Comput. Chem. 1998 19 769. 41 M. Cossi and V. Barone J. Chem. Phys. 1998 109 6246. 42 M. Cossi C. Adamo and V. Barone Chem. Phys. Lett. 1998 297 1. 43 T. Yamane Y. Inoue and M. Sakurai Chem. Phys. Lett. 1998 291 137. 44 A. H. Elcock and J. A. McCammon J. Mol. Biol. 1998 280 731. 45 S. Tomac and A. Gra� slund J. Comput. Chem. 1998 19 893. 46 J. A. Nilsson A. Laaksonen and L. A. Eriksson J. Chem. Phys. 1998 109 2403. 47 J. Gao P. Amara C. Alhambra and M.J. Field J. Phys. Chem. A 1998 102 4714. 48 L. Salvatella A. Mokrane A. Cartier and M.F. Ruiz-Lo� pez J. Org. Chem. 1998 63 4664. 49 J. C. Corchado and D. G.Truhlar J. Phys. Chem. A 1998 102 1895. 50 A. K. Rappe� and W. A. Goddard J. Phys. Chem. 1991 95 3358. 51 S. E. Rick S. J. Stuart and B. J. Berne J. Chem. Phys. 1994 101 6141. 52 Y.-P. Liu K. Kim B. J. Berne R. A. Friesner and S. W. Rick J. Chem. Phys. 1998 108 4739. 53 J. Gao J. Chem. Phys. 1998 109 2346. 54 R. A. Bryce M.A. Vincent N. O. J. Malcolm I. H. Hillier and N. A. Burton J. Chem. Phys. 1998 109 3077. 55 P. L. Cummins and J. E. Gready J. Comput. Chem. 1998 19 977. 56 J. F. Luque and M. Orozco J. Comput. Chem. 1998 19 866. 57 L. Ridder A. J. Mulholland J. Vervoort and I. M.C. M. Rietjens J. Am. Chem. Soc. 1998 120 7641. 58 T.M. Glennon and A. Warshel J. Am. Chem. Soc. 1998 120 10 234. 59 T. Bally and W. T. Borden in Reviews in Computational Chemistry ed.K. B. Lipkowitz and D. B. Boyd VCH-Wiley New York 1998 vol. 13. 60 Computational Thermochemistry ed. K. K. Irikura and D. J. Frurip ACS Symposium Series No. 677 Washington DC 1998. 61 K. Raghavachari and L. A. Curtiss in Modern Electronic Structure Theory ed. D. R. Yarkony World Scienti.c Singapore 1995. 62 J. W. Ochterski G. A. Petersson and J. A. Montgomery J. Chem. Phys. 1996 104 2598. 63 G. A. Petersson D. K. Malick W. G. Wilson J. W. Ochterski J. A. Montgomery Jr. and M. J. Frisch J. Chem. Phys. 1998 109 10570. 64 L. A. Curtiss K. Raghavachari P. C. Redfern and B. B. Stefanov Jys. 1998 108 692. 65 J. A. Montgomery Jr. M. J. Frisch J. W. Ochterski G. A. Petersson K. Raghavachari and V. G. Zakrzewski J. Chem.Phys. 1998 109 6505.66 H. L. Schmider and A. D. Becke J. Chem. Phys. 1998 109 8188. 67 D. Feller and K. A. Peterson J. Chem. Phys. 1998 108 154. 68 P.M. Mayer C. J. Parkinson D. M. Smith and L. Radom J. Chem. Phys. 1998 108 604. 69 J. M.L. Martin Chem. Phys. Lett. 1996 259 669. 393 Annu. Rep. Prog. Chem. Sect. B 1999 95 373—394 70 M. W. Wong and L. Radom J. Phys. Chem. A 1998 102 2237. 71 D.M. Smith A. Nicolaides B. T. Golding and L. Radom J. Am. Chem. Soc. 1998 120 10 223. 72 R. B.Murphy and R. A. Friesner Chem. Phys. Lett. 1998 288 403. 73 A. G. Csa� sza� r,W.D. Allen and H. F. Schaefer J. Chem. Phys. 1998 108 9751. 74 R. A. King W. D. Allen B. Ma and H. F. Schaefer Faraday Discuss. 1998 110 23. 75 T. Mineva N. Russo and E. Sicilla J. Comput. Chem. 1998 19 290. 76 N.Rega M. Cossi and V. Barone J. Am. Chem. Soc. 1998 120 5723. 77 F. R. Tortonda J.-L. Pascual-Ahuir E. Silla and I. Tun o� n J. Chem. Phys. 1998 109 592. 78 G. A. Kumar Y. Pan C. J. Smallwood and M. A. McAllister J. Comput. Chem. 1998 19 1345. 79 J. Chen M. A. McAllister J. K. Lee and K. N. Houk J. Org. Chem. 1998 63 4611. 80 P. B. Karadakov D. L. Cooper and J. Gerratt J. Am. Chem. Soc. 1998 120 3975. 81 J. I. Garcia V. Martý� nez-Merino J. A. Mayoral and L. Salvatella J. Am. Chem. Soc. 1998 120 2415. 82 L. Salvatella A. Mokrane A. Cartier and M.F. Ruiz-Lo� pez Chem. Phys. Lett. 1998 296 239. 83 J. S. Chen K. N. Houk and C. S. Foote J. Am. Chem. Soc. 1998 120 12303. 84 M. P. Repasky and W.L. Jorgensen Faraday Discuss. 1998 110 379. 85 J. Liu S. Niwayama Y.You and K. N. Houk J. Org. Chem. 1998 63 1064. 86 A. Venturini F. Bernardi M. Olivucci M. A. Robb and I. Rossi J. Am. Chem. Soc. 1998 120 1912. 87 J. Haller B. R. Beno and K. N. Houk J. Am. Chem. Soc. 1998 120 6468. 88 M. Torrent L. Deng M. Duran M. Sola and T. Ziegler Organometallics 1997 16 13. 89 H. Y. Yoo K. N. Houk J. K. Lee M. A. Scialdone and A. I. Meyers J. Am. Chem. Soc. 1998 120 205. 90 B. R. Beno J. Fennen K. N. Houk H. J. Lindner and K. Hafner. J. Am. Chem. Soc. 1998 120 10 490. 91 K. Makita J. Koketsu F. Ando Y. Ninomiya and N. Koga J. Am. Chem. Soc. 1998 120 5764. 92 H. F. Bettinger P. R. Schreiner H. F. Schaefer and P. v. R. Schleyer J. Am. Chem. Soc. 1998 120 5741. 93 A. Skancke D. A. Hrovat and W. T. Borden J. Am. Chem. Soc. 1998 120 7079. 94 G. N. Sastry T. Bally V. Hrouda and P. Ca� rsky J. Am. Chem. Soc. 1998 120 9323. 95 T. Bally and G. N. Sastry J. Phys. Chem. A 1997 101 7923. 96 R. Carr and M. Parrinello Phys. Rev. Lett. 1985 55 2471. 97 M. Boero M. Parrinello and K. Terakura J. Am. Chem. Soc. 1998 120 2746. 98 R. D. J. Froese D. G. Musaev and K. Morokuma J. Am. Chem. Soc. 1998 120 1581. 99 L. Deng T. K. Woo L. Cavallo P.M. Margl and T. Ziegler J. Am. Chem. Soc. 1997 119 6177. 100 T. Matsubara N. Koga D. G. Musaev and K. Morokuma J. Am. Chem. Soc. 1998 120 12 692. 101 E. Nakamura S. Mori and K. Morokuma J. Am. Chem. Soc. 1998 120 8273. 102 M.-D. Su and S.-Y. Chu Chem. Phys. Lett. 1998 282 25. 103 M. Garavelli F. Bernardi M. Olivucci T. Vreven S. Klein P. Celani and M.A. Robb Faraday Discuss. 1998 110 51. 104 M. Garavelli T. Vreven P. Celani F. Bernardi M. A. Robb and M. Olivucci J. Am. Chem. Soc. 1998 120 1285. 105 N. Yamamoto. M. Olivucci P. Celani F. Bernardi and M. A. Robb J. Am. Chem. Soc. 1998 120 2391. 106 L. Freund and M. Klessinger Int. J. Quantum Chem. 1998 70 1023. 107 P. E. M. Siegbahn M. R. A. Blomberg and M. Pavlov Chem. Phys. Lett. 1998 292 421. 108 P. E. M. Siegbahn L. Eriksson F. Himo and M. Pavlov J. Phys. Chem. A 1998 102 10622. 109 K. Pierloot J. O. A. De Kerpel U. Ryde M. H.M. Ollson and B. O. Roos J. Am. Chem. Soc. 1998 120 12 156. 110 M. H. M. Olsson U. Ryde and B. O. Roos Protein Sci. 1998 7 2659. 394 Annu. Rep. Prog. Chem. Sect. B 1999 95 3

ISSN:0069-3030    

DOI:10.1039/a808606k

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

12 Physical methods and techniques. NMR Spectroscopy Brian A. Salvatore Department of Chemistry and Biochemistry University of South Carolina Columbia SC 29208 USA 1 Introduction The purpose of this review is to pinpoint the important applications of NMRfrom the past year within the .eld of organic chemistry. Key developments include the e.cient and high-throughput characterization of mixtures with attention given to exciting new di.usion-based applications as well as the broader use of LC—NMR. Other highlights pertain to new derivatives solvating agents and techniques for chiral discrimination by NMR including new chiral liquid crystalline media. We will follow essentially the same format as that employed last year with coverage devoted mainly to small organic molecules.Again the author has adopted a selective rather than comprehensive approach in presenting some of the most signi.cant developments. 2 Spin polarization and chemical exchange NMRis useful for monitoring the progress of exchange processes in conformationally mobile systems particularly where the mobility is slow on the NMR time scale. Kelly et al. used selective spin polarization to assay for unidirectional rotation about a single bond in a synthetic molecular ratchet (1). This molecule was studied under conditions 395 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 where the half-life of a single rotamer is 0.17 s (60 °C). In this molecule H H and H each exist in distinct chemical environments. Polarization of any one of these protons (using a selective 180° pulse) followed by an appropriate time delay results in a chemical exchange process whereby the excited proton can go on to assume the environment previously occupied by one of the other two protons.If this molecule truly behaved as a ‘‘molecular ratchet’’ unimolecular rotation would occur and H would always pass .rst through the environment originally occupied by H before reaching the environment originally occupied by H. Thus states B and C would have unequal populations. However this was shown not to be the case as spin polarization originally placed on H always ended up equally on H and H. Therefore it was concluded that this molecule could not function as a molecular ratchet. The authors went on to point out that the ability of any molecule to function as a molecular ratchet would in fact represent a violation of the principle of microscopic reversibility and thus contradict the Second Law of Thermodynamics.3 Measurement and analysis of couplingconstants Longrang e 13C–13C scalar couplings Continued investigations are being made into the use of long-range coupling constants in conformational studies. This is particularly useful in carbohydrate chemistry where conformational analysis is largely based on the measurement of inter-residue proton NOE’s. However much structural information is also available from C—C scalar coupling constants. Long range C—C couplings through carbon—oxygen bonds are proving to be one such valuable source of information. New synthetic methodology for preparing C-labeled oligosaccharides is opening up vast new opportunities in this area. When used in conjunction with NOEinformation two- and three-bond C—C coupling constants can greatly improve the conformational analysis of oligosaccharides.Serianni et al. measured long-range C—C couplings for a variety of C-labeled carbohydrates having di.erent structures and dihedral angles. These workers also extended this technique to the analysis of glycosidic bonds in order to obtain torsion angle information across glycosidic bonds. Since it is not always easy to measure the signs of long-range coupling constants a new application of the C—C COSY was implemented to accomplish this whereby the orientation of a cross-peak could be used to determine the sign of the coupling constant (Fig.1). Advancements have also been made in determining conformation from scalar couplings. The reliable projection-resultant method has recently been applied to threebond C—C scalar couplings. Speci.cally J COCC coupling constants were analyzed as they relate to torsion angles across glycosidic linkages and this information was used to construct a Karplus curve. This illustrated the dependence of coupling constant magnitude on dihedral angle. The results indicated that in addition to the dihedral angle other factors a.ect the sign and magnitude of these coupling constants the most prominent of these being the disposition of terminal electronegative substituents. Although the authors’ Karplus curves remain under-parameterized at this stage they are con.dent that further re.nement of these curves can be accomplished by additional computational studies of J behavior.396 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 1 Crosspeaks from the C¡XC COSY-45 spectrum of -(1,3,6-C )allopyranose. The displacement of the C(3)¡XC(6) crosspeak (A) is opposite to that observed for the C(1)¡XC(3) crosspeak (B). Thus one can infer that these two C¡XC scalar couplings are opposite in sign. Since the orientation in (A) was correlated with a positive scalar coupling. J¡X must be negative in sign. (Spectra reprinted with permission from Carbohydr. Res. 309 p. 145 Copyright 1998 with permission from Elsevier Science). The determination of two-bond and three-bond C¡XC coupling constants in oligosaccharides and polysaccharides was also facilitated by the application of a novel two-dimensional C¡XH heteronuclear experiment that was originally used in proteins.Long range C¡XC coupling constants are related to C¡XH crosspeak intensities for pairs involving a specic proton and a pair of scalar coupled C atoms according to the relationship given in eqn. (1) 397 Annu. Rep. Prog. Chem. Sect. B 1999 95 395¡X417 Fig. 2 The magnitude of the three-bond C—C scalar coupling constant across the glycosidic linkage between C and C can be determined from the ratio of the cross peak intensities for H—C and H—C within the quantitative coherence transfer spectrum according to the arctan dependance described within the text.(1) (2) I — /I — tan(2J — T) where I — and I — represent the two measured C—Hcross-peak intensities for a common proton J — represents the long range C—C coupling constant of interest and T is the delay time in the pulse sequence during which that scalar coupling evolves. Thus the C—C coupling constant can be determined using eqn. (2). J — arctan[(I — /I — )]/2T This pulse sequence is particularly useful for measuring long-range C—C scalar couplings across glycosidic bonds (Fig. 2). Measurement of 13C–1H scalar couplings Continued advances are being made in the detection of long-range H—C coupling constants. The recently devised SIMBA technique uses selective C-excitation and proton detection to provide excellent suppression of protons not coupled to a C. This facilitates the measurement of couplings from atoms appearing in crowded regions of both the proton and carbon spectra.In this technique a reference proton spectrum is acquired to remove anti-phase components resulting from passivelycoupled protons. Detection of hydrogen bonds The direct observation of hydrogen bonds represents a powerful new application of NMR in the .eld of molecular recognition. In a novel adaptation of correlation spectroscopy scalar two-bond N—N couplings established the basis for the detection of hydrogen bonds between an amine (donor) and an imino group (acceptor) in nucleic acids (e.g. Watson—Crick base pairs). The HNN COSY experiment is a combination of a refocused INEPT and the modern TROSY sequence.Use of the TROSY segment for magnetization transfer from N back to H is advantageous for studies on large molecules where T is very short. Though not described in this report the alternative use of a standard reverse INEPT segment instead of the TROSY sequence would likely su.ce for characterizing hydrogen bonds in molecular recognition studies on small organic molecules. Despite the powerful analytical capabilities of this new technique it may not be generally applicable outside of peptide and nucleic 398 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 acid chemistry since N-labeling of both the H-bond donor and acceptor nitrogen atoms is necessary.Analyzingcuprate structure NMR has proven to be a very powerful technique for analyzing organometallic structures in solution. Scalar coupling measurements on dimethylcuprate and higher order cuprates generated from C-labeled methyllithium have provided important new information about cuprate structure. Scalar C—C H—Cwere identi.ed in solution for a variety of cuprates. For dimethylcuprate these results showed that both carbon atoms coordinate to a single central copper atom. Interestingly however higher-order cuprates (e.g. Me CuLi ) displayed solution structures resembling dissociated lower-order cuprates and methyllithium. Similarly no correlations were seen between the cyano and methyl groups in C-labeled Me Cu(CN)Li .This indicated that the cyano group was not bonded to the copper center. These are very interesting results which are highly relevant to the structures of cuprate species in solution and further NMR studies are warranted. 4 Di.usion-ordered and a.nity NMR Di.usion-ordered spectroscopy (DOSY) seeks to separateNMRsignals from di.erent chemical species based on their di.usion coe.cients. In a standard 2-D DOSY experiment the decay of the amplitude of each peak is .tted to an exponential as a function of the square of the square of the .eld gradient pulse area. A prerequisite for high resolution DOSY spectra is the existence of well-resolved NMR spectra for all of the compounds in a given mixture as it is di.cult to extract clean di.usion parameters from overlapping signals particularly if they result from di.erent compounds.Barjat et al. reported three variations on common di.usion pulse sequences. The authors advise against the use of bipolar gradient-based sequences for small molecules since these sequences which sandwich 180° refocusing pulses between each pair of bipolar gradients require more elaborate phase-cycling than gradient-compensated routines lacking refocusing pulses. However di.usion studies involving large molecules where spin di.usion is frequently a problem require the bipolar gradient-based sequences. Thus the authors explored the elimination of common systematic errors in di.usion spectroscopy. Fig. 3 illustrates a 2-D DOSY spectrum of quinine (2) geraniol (3) and camphene (4).The boxed region represents an inaccuracy in the measurement of the quinine di.usion coe.cient due to its overlap with the water (solvent) signal. This problem was alleviated by processing the individual spectra using reference deconvolution and baseline correction. The 2-D sequence was adapted to three dimensions resulting in a 3-D DOSYHMQCsequence. This implementation consists of a stimulated echo sequence using bipolar .eld gradient pulses in combination with a gradient-enhanced HMQC. This experiment provides standard heteronuclear correlation information with the individual 2-D planes sorted by di.usion coe.cient in the third dimension (Fig. 4). It is particularly e.ective in alleviating spectral overlap that often occurs in di.usion measurements of the components of complexmix tures.Millet and Pons developed a 399 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 3 400MHz proton DOSY spectrum of a mixture of quinine (Q) geraniol (G) camphene (C) and solvent (T) deuteromethanol and TMS. The horizontal axis represents the proton chemical shift and the vertical axis represents the di.usion coe.cient. The normal 1D-proton spectrum is shown at the top and the integral projection of the DOSY-spectrum is shown to the left. The dotted boxhighlights the region around 4.9ppm referred to in the text. (Spectrum reproduced with permission from M.D. Pelta H. Barjat G. A. Morris A. L. Davis and S. J. Hammond Magn. Reson. Chem. 1998 36 706. Copyright John Wiley & Sons Limited). simpler alternative to the above 3-D experiment by using accordion spectroscopy. The technique termedGAUDI (Gradient Accordion Used for DI.usion) involved the acquisition of a single 2-D experiment with the concerted incrementation of the evolution and di.usion delays.It allowed the di.usion coe.cients to be derived from the lineshape of the 2-D diagonal and crosspeaks. A.nity NMR—analysis of mixtures Reliable di.usion-basedNMRmethods for screening mixtures of compounds for their binding a.nities to receptors are among the most sought after techniques today. A.nityNMRis based on the well-accepted principle that a small molecule binding to a receptor in solution has a substantially di.erent di.usion coe.cient to that of the small molecule alone. Thus molecules can be edited from the spectrum based on 400 Annu.Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 4 3D-DOSY-HMQC spectrum of a mixture of quinine (Q) camphene (C) geraniol (G) andTMS(T) in deuteromethanol. The four components of the mixture are clearly resolved vertically according to their di.usion coe.cients. Crosspeaks that partially overlap in both the H and C dimensions give rise to di.usion coe.cients which di.er slightly from those for well-resolved signals. These crosspeaks are responsible for the satellite peaks in the di.usion dimension (left axis). (Spectrum reproduced with permission from J. Magn. Reson. 1998 131 131). di.erences in their di.usion coe.cients. A key advantage of a.nity NMR over competing techniques is that binding is detected by the observation of the ligand’s NMRspectrum and not by the receptor’s spectrum.This makes it possible to tune the binding a.nity of the ligand by changing the relative concentration of the receptor. It also makes it unnecessary to work with isotopically labeled proteins. Binding constant measurements between small peptides and the glycopeptide antibiotic vancomycin were determined using the PFG-NMR technique. Since the free and bound ligands are in fast exchange on the NMR time scale in their binding with vancomycin the observed di.usion coe.cients can be approximated as the weighted average of the free and bound species [eqn. (3)]. (3) D FD F D When the concentration of bound ligand [L·R] is much less than the concentration of free receptor [R] the dissociation constant can be determined by eqn.(4) K (4) [L][R]/[L·R][(D D)/(D D)][R] where D and D are the di.usion coe.cients of the free ligand and D is the observed di.usion coe.cient. For most systems where the total amount of receptor that binds is relatively small [R] can be approximated as being equal to the total concentration of receptor. A.nity NMR studies are not limited to ligand—receptor interactions involving proteins. Larive et al. have employed similar di.usion methods in the measurement of SDS micelle—peptide association constants with two tripeptides. The di.usion technique in fact proved to be far superior to standard chemical shift di.erence measure- 401 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 ments which are commonly used for molecules in fast exchange.Technical challenges in di.usion spectroscopy When the receptor is much larger than the ligands it is not di.cult to choose gradient strengths and conditions where the non-binding ligands are not observed. It is more challenging however when the molecular weights of the receptor and peptide lie within the same order of magnitude. These technical limits were recently tested in a study of the antibiotic vancomycin with a mixture of ten di.erent tetrapeptides. To meet this challenge the standard 1-D pulse .eld gradient (PFG) NMR spectrum was extended to two dimensions using di.usion encoded spectroscopy (DECODES). This experiment is a combination of PFG NMR and total correlation spectroscopy (TOCSY).It simpli.es the identi.cation of the relative strength of receptor—ligand interactions to the measurement of relative intensities of selected crosspeaks. This technique successfully identi.ed the two strongest binding ligands for vancomycin out of a mixture of ten tetrapeptides. A potentially serious problem with a.nity NMR is that signals from the receptor are always present in the di.usion-edited spectrum and this can hamper the interpretation of the data. Isotope-.ltered di.usion-edited spectroscopy can be used to suppress this problem. Use of a pulse scheme along with an isotopically-labeled protein receptor results in the elimination of protein signal in addition to those from the non-binding ligands. Serious problems in di.usion-basedNMR experiments can also result when chemical exchange occurs during the di.usion experiment.This can occur with the commonly employed stimulated echo and longitudinal eddy current delay experiments. In a.nity NMR for example chemical exchange resulting from a di.erence in chemical shifts between the bound and unbound species can produce serious inaccuracies in di.usion-ordered and di.usion-edited spectra. This e.ect is most serious in experiments when the gradient delay time () is varied during the course of the experiment but serious intensity distortions may also result in experiments where is held constant depending on the choice of . These problems have been solved with a bipolar gradient pulse sequence where each monopolar gradient is replaced by a pair of gradients of opposite polarity and separated by a 180° rf pulse. Thus chemical shift evolution is refocused by the 180° pulses during the gradient-encode and decode delays.This completely removes the intensity modulation that would otherwise result from chemical exchange. 5 Assignment of absolute stereochemistry The use of stereospeci.c or highly stereoselective reactions generates organic products enriched in one enantiomer but rarely are the products enantiopure. Thus it is important to be able to conveniently assess the enantiomeric purity of these products. The successful application of NMR in this regard generally involves the conversion of a pair of enantiotopic (i.e. NMR-equivalent) protons to a pair of diastereotopic protons whose NMRsignals are distinguishable.Recent studies have produced many new chiral derivatizing agents and chiral solvating agents the former requiring the 402 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 formation of a covalent adduct the latter simply serving as chiral shift agents without reacting chemically with the species of interest. Chiral derivatizingag ents Generally deducing the absolute stereochemistry of a secondary alcohol has required the formation and comparison of spectra of a pair of diastereomeric derivatives. The two most commonly employed derivatives are methoxy(tri.uoromethyl)phenylacetic acid (MTPA Mosher’s acid) esters and methoxyphenylacetic acid (MPA mandelate) esters. MTPA is often unreliable for secondary alcohols because it generally fails to produce adequate NMR chemical shift di.erences for diastereomeric pairs of ester derivatives.This is apparently due to the variety of conformations sampled in Mosher’s esters and it has lead to the popularity of the mandelate ester. The situation however is more .exible for amines since both MTPA and MPA amides are conformationally rigid and have been shown to be satisfactory derivatives for NMR characterization. Although both of these agents are satisfactory for amines research continues to focus on the development of improved derivatizing agents for establishing the absolute stereochemistry of alcohols. In a systematic study Lebreton et al. demonstrated that the O-acetylmandelate esters are superior to the traditionally employed —OMe mandelate esters for assessing enantiomeric excesses and the absolute con.guration of secondary alcohols. It is somewhat surprising that the commonly used O-methylmandelate ester is not the optimal mandelate derivative.These same authors demonstrated that the mandelate ester with a free —OH shows the most superior results of all. That derivative is easily obtainable by hydrolytic removal of the O-acetate group. Other investigators have reported the development of completely new CDA’s. An e.cient reagent for determining absolute stereochemistry uses the chiral chloromethyl pyrrolidone 5. This chiral auxillary contains an isolated pair of protons which produce an AB quartet in the NMR spectra. Stereochemical information is extracted by comparing the chemical shift di.erences between the diastereotopic protons (H and H) for a diastereomeric pair of aminal derivatives.The authors’ model indicates that the largest di.erence in chemical shift will result in derivatives of R alcohols assuming that the sterically most demanding group is also assigned the highest Cahn—Ingold priority. In this and the previous examples the preparation of a pair of diastereomeric derivatives is absolutely necessary but it would be desirable to assign stereochemistry by making just one derivative. Latypov et al. have developed a method through which the absolute stereochemistry of secondary alcohols can be determined from the proton NMR spectrum of a single diastereomer of the methoxyphenylacetic ester. This model is based on the predictable predominance of a low-energy molecular conformation as the temperature of the 403 Annu.Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 5 (R)- and (S)-MPA esters of secondary alcohols exist in two main conformers. Above a Newman projection is used to sight down two bonds within an (R)-MPA ester. The sp conformer is lower in energy than the ap conformer. As the temperature is lowered the sp conformer becomes more populated. Since the phenyl ring is gauche to L it induces an up.eld shift (toward lower ppm) of protons in L . Simultaneously the protons in L should experience a down.eld shift as ap becomes less populated. Thus L and L can be sterospeci.cally assigned from the direction that certain resonances move as the temperature is lowered using just a single MPA ester derivative.NMR sample is lowered (Fig. 5). Thus one group is eclipsed by the phenyl ring and it experiences anisotropic shielding thus undergoing an up.eld shift. Similarly protons within the other group that are eclipsed by the phenyl ring in the higher energy conformer shift down.eld. The methoxy-(1- and 2-naphthyl) acetic acids have also served as useful chiral derivatizing agents whereby the stereochemistry of a variety of straight-chain secondary alcohols was determined from just one derivative. These methods facilitate the determination of absolute stereochemistry since the preparation of just one derivative consumes only one-half of the usual amount of sample. Such methods are particularly useful when the sample requiring derivatization is an advanced synthetic intermediate whose availability is limited.Nearly all methods for determining the absolute stereochemistry of alcohols have been developed for secondary alcohols. Primary alcohols represent a greater challenge since the longer distance between the auxillary and the stereogenic center reduces not only the conformational preference but also the in.uence of the auxillary itself on NMR chemical shifts of groups bound to the stereogenic center. Thus the traditionalMPAand MTPA esters are unsuitable for determining the stereochemistry of primary alcohols. Recently however a novel method has been developed for primary alcohols which are located - to stereogenic centers using esters of both (R)- and (S)-anthrylmethoxyacetic acid (6 and 7). This technique is reliable as long as the -stereogenic center is not part of a ring since drastic di.erences in conformational preferences apparently exist in such systems.The development of methods for determining the absolute stereochemistry of terti- 404 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 6 The fucofuranoside method is useful in determining the absolute stereochemistry of tertiary alcohols. The basis of this technique is the strong and uneven paramagnetic (deshielding) e.ect of the solvent (d -pyridine). In the preferred conformation this shift occurs in the R (methylene) group proximal to the furanose ring. Thus an -proton or carbon in R will have a higher chemical shift value within the .-derivative than in the .-derivative.The opposite is true for the R group. The preparation of both derivatives allows one to stereospeci.cally assign R and R. ary alcohols represents an even greater challenge that was unmet until recently. Kobayashi devised a novel strategy using -glycoside derivatives from .- and .- fucofuranose. Although the glycosidation of tertiary alcohols is very di.cult su.- cient yields were obtained for fucofuranoside derivatives of sixtertiary alcohols. This method is only suitable for tertiary alcohols which are substituted with two methylene groups and one methyl group. The exo-anomeric e.ect helps to establish a preferred gauche conformation of the fucofuranoside derivative about the glycosidic bond (Fig.6). Di.erential solvation of the furanose moiety in a strongly anisotropic solvent like pyridine leads to a down.eld shift for one of the two methylene groups in the tertiary alcohol. Comparison of either the H or C NMR spectra for both the .- and .-fucofuranose derivatives of a given tertiary alcohol generally provides an unambiguous stereochemical assignment for the alcohol. Chiral solvatingag ents Chiral solvating agents (CSAs) di.er from chiral derivatizing agents in that no covalent bonds between the molecule and the auxillary are needed. Rather CSAs exert their in.uence intermolecularly. Recently many new CSAs have been developed for the determination of enantiomeric excess by NMR decreasing the traditional reliance on lanthanide-based chiral shift reagents.Triazine 8 is well-suited for many classes of alcohols carboxylic acids amides and esters with the added advantage that less than an equimolar amount can sometimes be used. Although a full molar equivalent generally produces superior results it is not always necessary. The use of 25 mol% was shown to be su.cient for obtaining substantial separations (16 Hz) between corresponding protons in two enantiomeric dinitrophenyl amide derivatives. Some CSAs rely on the formation of diastereomeric salts. Mandelic acid (2—3 equivs.) for example was shown to be a good CSA for -amino esters and amides producing di.erences in chemical shift by as much as 0.06 ppm in the methyl ester 405 Annu. Rep. Prog. Chem.Sect. B 1999 95 395—417 signals from a racemic pair of -dimethylamino esters. The TRISPHAT anion (9) has been shown to be an excellent CSA for cationic compounds including ruthenium tris[bisamine] complexes including Ru(bipy) and Ru(phen). It also works well with chiral phosphonium salts. Other CSAs rely primarily on hydrogen bonding. Chiral amide 10 (1.0 equiv.) was found to be an excellent agent for -substituted carboxylic acids presumably via the formation of diastereomeric host¡Xguest complexes 11. Liquid crystalline chiral solvatingag ents Courtieu et al. have adopted a very novel approach for assessing stereochemical purity. Instead of performing studies in conventional isotropic media they employed a liquid crystalline solvent poly-(-benzyl-glutamate) (PBLG) in various organic 406 Annu.Rep. Prog. Chem. Sect. B 1999 95 395¡X417 Fig. 7 Schematic proton-decoupled deuterium NMR spectra of a mono-deuterated racemic molecule dissolved in various solvents (a) isotropic solvent; (b) nonchiral nematic solvent; (c) chiral nematic or cholesteric solvent in which the order parameters for the two enantiomers are di.erent. v and v are the quadrupolar splittings for each enantiomer. (Reproduced with permission from Chem. Eur. J. 1998 4 1142). solvents. The small di.erential ordering of enantiomers within this medium serves as the basis for chiral discrimination. Chemical shift anisotropy of dipolar nuclei (e.g. C) has been used as a basis for determining molecular order and thus enantiomeric purity. An alternative method for chiral discrimination involves deuterium NMR. It is well-known that in an achiral ordered medium the interaction of a deuterium quadrupole with the electric .eld gradient at the nucleus produces a doublet in the deuterium NMR spectrum.Within a chiral ordered medium however a pair of doublets is seen for each enantiomer due to di.erential ordering of the two enantiomers. Thus identi.cation of corresponding pairs of quadrupolar doublets in a proton-decoupled deuteriumNMRspectrum can serve as the basis for chiral discrimination (Fig. 7). A detailed study with a series of 22 linear chiral secondary alcohols indicated that the measured di.erence in H quadrupolar splittings depends on the degree of dissimilarity of the two alkyl chains attached to the stereogenic center as well as the degree of molecular ordering within the liquid crystalline medium. The requisite one-dimensional deuterium spectra have generally been acquired with synthetically-labeled molecules since it is often di.cult to simultaneously correlate all 407 Annu.Rep. Prog. Chem. Sect. B 1999 95 395—417 of the quadrupolar doublets in natural abundance deuterium spectra. However that limitation has recently been overcome with the development of a two-dimensional auto-correlation deuterium NMR experiment extending the technique to more general applications with unlabeled samples. The pulse sequence termed QUOSY (QUadrupolar Ordered SpectroscopY) is derived from the COSY sequence but it incorporates a 180° pulse in place of the second 90° pulse. This modi.cation removes all of the diagonal peaks and some cross-peaks except those representing the quadrupolar doublets.A WALTZ-16 sequence is employed to accomplish proton—deuterium scalar and dipolar decoupling. Thus it is possible to identify the individual pairs of quadrupolar doublets discriminated in the liquid crystalline medium from the 2-D spectrum and the corresponding signals from each quadrupolar doublet within the 1-D deuterium spectrum can be .tted individually. Courtieu et al. recently provided a quantitative description of the facial discrimination for molecules oriented within PBLG in various organic solvents. They proved that this technique can be extended to distinguish any enantiotopic pair of magnetically equivalent atoms in prochiral molecules because such atoms become diastereotopic in the PBLG liquid crystalline phase. The measurement of proton dipolar couplings in this media further extends this technique for chiral discrimination.Studies on a series of linear primary alcohols formed the basis for a new method of determining the sign and magnitude of two-bond H—H scalar couplings between a pair of magnetically equivalent enantiotopic protons. Although such couplings normally exist they cannot ordinarily be obtained from isotropic spectra and this new method represents the most reliable technique to date for their determination. 6 Solid phase combinatorial chemistry Analytical NMR methods One principal goal of solid phase organic synthesis is to prepare relatively pure materials without the need for extensive puri.cation.Traditionally direct cleavage of resin-bound materials and measurement by H NMR using an internal standard has been employed to establish loading of resin-bound synthetic intermediates. Although this method is still popular new techniques are extending our ability to perform on-resin NMR analysis rendering the prior cleavage step unnecessary. Armstrong et al. devised a systematic approach for the on-resin analysis of solid phase reaction products using existing analytical methods. Although single bead infrared (IR) spectroscopy requires less time than NMR it is unable to characterize reactions when the products show no changes in functional groups with characteristic infrared absorptions.Thus a combination of IR and H magic angle spinning (MAS) NMR is far more e.cient than IR alone in the analysis of solid phase synthetic products. Although IR and HMASNMRare generally su.cient for monitoring small molecular products H MAS COSY C MAS NMR AND H—C MAS-HSQC are well-suited for more complexresin-bound molecules. The true power of these techniques is best achieved when they are used in combination. 408 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fluorine derivatives Fluorine NMR is .nding application in monitoring the progress of solid phase chemical reactions. Three new .uorinated analogs of three commonly used linkers have been employed in the synthesis of small molecules on TentaGel resin. The .uorine atom on each linker’s aromatic moiety was shown to be sensitive to changes in the local environment corresponding to solid phase synthetic transformations.Coupling cleavage acylation and cyclization reactions were among the processes e.ectively monitored by F NMR. Moreover these NMR assays could be performed under gel-phase conditions with a standard NMR tube and solutions probe. In a separate study p-.uorophenolate was shown to be an e.ective capping reagent for assessing the degree of loading of Merri.eld’s resin. The use of .uorobenzene as an internalNMR standard facilitated the quantitative determination of the amount of unreacted sites requiring just a small portion of resin sample. Spin-echo correlation spectroscopy (SECSY) TentaGel (polyethylene glycol polystyrene) is the preferred support for NMR analysis in solid phase synthesis.Due to poor magnetic susceptibility properties,NMRsamples with resins other than TentaGel display broad featureless spectra. Spin-Echo Correlation SpectroscopY (SECSY) has proved advantageous in the analysis of organic products on resins other than TentaGel. In the early days of 2-D correlation spectroscopy SECSY proved advantageous due to its conservative use of disk storage space (ca. 50% of a COSY experiment). Although that particular advantage is no longer an issue today SECSY provides other advantages in the analysis of resinbound materials. Since it is a spin-echo based experiment sample inhomogeneities are substantially reduced producing much narrower lines. The 2-D SECSY sequence o.ers enhanced resolution similar to J-resolved spectroscopy while additionally providing spin system connectivities.Although SECSY data are presented in a rather unconventional way with the ‘‘diagonal’’ lying horizontal at the center of the indirect dimension the increased resolution and availability of cross-peak correlations make it a very useful technique for the analysis of on-resin products. 7 LC–NMR The integration of high performance liquid chromatography and NMR instrumentation has proven to be a very powerful combination (LC—NMR) especially for the automated analysis of impure samples containing small amounts of material. Although this technique has been known for over 15 years only recently has it gained signi.cant acceptance in the pharmaceutical industry as well as the natural products and polymer communities.— Since LC—NMR links puri.cation with NMR data acquisition we should expect di.erences in the way NMR data are gathered.Traditionally NMR samples are handled in ‘‘batch mode’’ where we are assured that the entire sample is in the NMR tube during the whole experiment. LC—NMR data however are acquired in either ‘‘on-.ow’’ or ‘‘stopped-.ow’’ acquisition modes. On- .ow measurements expectedly deliver far lower sensitivity than their stopped-.ow 409 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 8 A typical two-dimensional contour plot from an ‘‘on-.ow’’ LC—NMR experiment. This is an SEC—NMR analysis of isotactic poly(methyl methacrylate) (PMMA). (Reproduced with permission from Polym.J. 1998 30 439). counterparts and thus most sophisticated 2-D NMR experiments must be performed in stopped-.ow mode. Yet on-.ow analysis can rapidly provide structural information about the major components of a mixture and the sensitivity limitations are gradually being overcome. The increasing popularity of LC—NMR stems from advances in the construction of high .eld magnets probe technology pulsed .eld gradients and new solvent suppression routines. These advances have enhanced the sensitivity and dynamic range in LC—NMR making it a viable analytical technique. Enhancingsensitivity In its most basic implementation LC—NMR su.ers from intrinsically low sensitivity. The sample concentrations typically eluted from an analytical HPLC column are only ca.10gmL. Tailing peaks elute in even smaller concentrations and it becomes di.cult to match the chromatographic peak volume with the volume of theNMR.ow cell. In continuous .ow mode the chromatographic and spectroscopic parameters must be especially carefully matched and optimized to achieve optimal sensitivity. A systematic study of sensitivity vs. .ow rate showed a severe drop-o. in the signal to noise ratio for .ow rates exceeding 0.5mLmin using a 60 l NMR .ow cell. This fallo. at higher .ow rates is presumably caused by the sample transfer rate through the cell being too fast to achieve adequate equilibration of nuclear polarization prior to the R pulse. A common form of output in continuous .ow HPLC—NMR is the 2-D retention time vs.chemical shift contour plot (Fig. 8). Here a new spectrum was added to the contour plot every 18 s averaging the signal from eight scans. The NMR community welcomes new techniques which can alleviate potential problems and thus facilitate the characterization of small concentrations of samples and impurities. E.orts are being made to better match the sample volume with that of the .ow cell. Gri.ths and Horton achieved signi.cant gains in LC—NMR signal to noise ratio by selective concentration of speci.c chromatographic peaks on a chromatographic guard column and a subsequent back.ush of the sample into the NMR .ow cell. This technique begins with standard reverse-phase HPLC. After 410 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 elution from the column in a binary solvent gradient the puri.ed sample is diverted past a T-piece from which D O is pumped by a second pump and onto a short chromatographic guard column.The protonated solvent from the original pump is then cut o. while continued D O .ow serves to trap the compound on the guard column while also .ushing all of the protic solvents out of the downstream lines. When this is accomplished a third pump is activated that back.ushes the guard column with a deuterated solvent of high elutropic strength (e.g. CD CN). This solvent elutes the pure compound o. the guard column and into an NMR .ow cell with a chromatographic peak volume close to that of the NMR .ow cell facilitating stopped-.ow spectral acquisition with optimal sensitivity.Additionally this set-up eliminates the need for solvent suppression routines in the NMR experiment. Solvent suppression Adequate solvent suppression can generally be achieved in LC—NMR with the WET sequence which entails four selective 20 ms R pulses each followed by a pulsed .eld gradient to dephase the excited magnetization. C-decoupling is usually applied during the selective proton pulses to remove carbon satellites from the solvent resonance as well. This is the most commonly-used solvent suppression in LC—NMR and it becomes particularly crucial when column-trapping or comparable techniques are unavailable. Pasch et al. employedWETunder isocratic on-.ow conditions to analyze the structure and tacticity of a series of oligostyrenes. This method facilitated the use of protonated HPLC-grade acetonitrile as the eluent (spectra were acquired unlocked) which amounted to a huge cost saving over the use of a deuterated organic chromatography solvent.Due to its .exibility and speed WETis also well-suited for handling LC solvent gradients. This aspect represents a challenge to most conventional solvent suppression routines because an ever-changing solvent composition continually shifts NMRsolvent resonances presenting the spectrometer with a ‘‘moving target’’. In such cases the commercially-available Scout-Scan software interfaces with WET automatically tracking the solvent composition with a single dummy scan. It then tailors the suppression routine to the composition at each time-point during the HPLC run.This combination solves an otherwise very serious obstacle to LC—NMR. Variations in LC–NMR Various improvements and modi.cations are currently being been made to the standard LC—NMR set-up. Simply removing the chromatography column increases throughput while still permitting the use of very small sample volumes in stopped-.ow mode. This technique is called ‘‘.ow-injection analysis’’. Alternatively ‘‘direct injection NMR’’ involves a further simpli.cation entailing removal of the pump UV detector and the mobile phase leaving only the injector and connective tubing. This system o.ers higher sensitivity because sample dilution does not occur in the lines and UV .ow cell. These techniques are particularly suitable for the automated NMR analysis of combinatorial libraries where sample puri.cation is not as important as highthroughput analysis.411 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Applications of LC–NMR The .rst fruitful applications of LC—NMR were directed at understanding drug metabolism. Recently however this technique has found a lot of use in natural products and polymer chemistry. In complexnatural products solvent suppression occasionally bleaches out portions of the spectrum which contain peaks of interest. The comparison of spectra acquired in two di.erent solvent systems is one solution to this problem. In some cases however more convenient alternatives exist. Kraus et al. used selective 1-D NOESY experiments as an alternative to 2-D WETNOESY. Due to the selective excitation in the 1-D technique no solvent suppression was required in these experiments.Moreover the 1-D experiment can be acquired in much less time than its 2-D counterpart enabling the spectroscopist to do a rapid analysis of all important protons. Perhaps the most sophisticated LC—NMR work in natural products chemistry to date involves the identi.cation of a new prenylated .avone monotesone A. This compound was identi.ed from 1mg of enriched plant extract using on-.ow 1-D techniques as well as 2-D NOESY TOCSY HSQC and HMBC in stopped-.ow mode. Moreover all of this spectroscopy could be performed in a single day! In polymer chemistry size exclusion chromatography—NMR (SEC—NMR) is proving to be particularly useful. This technique is informative in determining the molecular weight dependence of polymers on tacticity and co-monomer composition.Hatada et al. used SEC—NMR to separate and analyze the molecular weight distribution of isotactic poly(methyl methacrylate). Since this polymer has a tert-Bu group at the end of the chain the degree of polymerization can be determined from the relative intensity of the tert-Bu and —OMe signals in the HNMRspectrum. Using a 750MHz spectrometer and isocratic eluent .ow these authors achieved reliable molecular weight data on polymers exceeding 22 000 Da. (5) Advances in LC–NMR technology The eventual widespread and practical application of LC—NMR is contingent upon further technological improvements. Reductions in sample volume are continually being made with the most challenging limitations lying in probe design.The recent development of a 1.7mmgradient inverse-detection probe for a 0.5 mol sample will likely be superseded providing access samples as low as 0.1 mol (requiring just 5 L of solvent) by the time this article gets to press. Further reductions in scale are on the horizon which could provide high-resolution capillary NMR systems that will permit the analysis of a few nanomoles of sample. A useful technique which is not limited to LC—NMR involves the use of Ernst angle pulses. This technique can also speed up acquisitions of NMR spectra in both on-.ow and stopped-.ow LC—NMR. The Ernst angle is the optimal angular pulse width for a particular interscan delay and longitudinal relaxation rate.This angle is de.ned by eqn. (5) where T represents the interscan delay and T is the longitudinal relaxation time constant. cos( )e In 2D-NMR 90° pulse widths and 1 s interscan delays are used. Ross et al. showed 412 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 that the use of excitation pulses at the Ernst angle facilitates data acquisition for a limited variety of NMR experiments where net Z-magnetization exists at the beginning of the acquisition time (e.g.,HMQCand J-spectroscopy but not COSY NOESY or TOCSY). Application of these short optimized pulses in HMQC for an ‘‘SAR by NMR’’ application allowed data to be acquired at a much higher rate. In practice this produced signal levels comparable to those obtained from conventional 90° excitation pulses but in far less time.An upperbound in the pulse repetition rate is established by the spectrometer’s duty cycle (ca. 200 ms). The ability to render 2-D data quickly makes Ernst angle pulses highly useful in high-throughput NMR screening methods. Although this particular report focused on probing protein—ligand interactions the authors commented on more current studies directed toward small organic molecules where larger T values should lead to even more substantial sensitivity gains and time-economy. 8 NMR in supramolecular chemistry and molecular recognition Monitoringencapsulated reactions The use of NMR in analyzing host—guest systems is being extended at a rapid pace. Rebek et al.used H NMR to study the reversible encapsulation of small molecules within self-assembling molecular capsules. This proved to be a very powerful tool in studying the Diels—Alder reaction for molecules entrapped within the capsule. Large changes in the proton chemical shifts for the encapsulated species (both reactants and product) permitted measurement of encapsulation equilibrium constants by NMR. The kinetics of the Diels—Alder reaction was also studied. Sizeable rate accelerations were measured for this bimolecular process on the encapsulated species although product inhibition was found to prevent catalytic turnover. NMR and supramolecular association NMR is often indispensable in assessing the nature of supramolecular complexation. Cloninger and Whitlock used proton NMR to probe multiple edge—face interactions for aromatic guests which bind to a synthetic receptor (Fig.9). These interactions resulted from the abutment of positively polarized hydrogen atoms in the aromatic guests with the -electrons of an anthracene ring within the host. For a series of guests the protons that were positioned over the anthracene ring shifted an average of 4.7 ppm whereas the other aromatic protons within the guest only shifted an average of 1.3ppm upon complexation with the host. It was concluded that such interactions stabilized the supramolecular complexes formed with this particular host. NMR is a very powerful tool for assessing the binding strength and the nature of association in supramolecular complexes.Such studies are typically performed by plotting chemical shift changes for a guest against its concentration in a particular host—guest system. For example Matsui et al. used proton NMR to measure the association constants of a series of naphthalenecarboxylic acid regioisomers with -cyclodextrin. More speci.c structural information about the complexes was obtained from H-ROESY spectra. Steric hindrance resulting from complexation with 413 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 Fig. 9 Structure of an indolecarboxylate bound within a synthetic receptor. This receptor contains an anthracene unit that is capable of forming multiple edge—face interactions. Here two edge—face interactions were assigned based on very large up.eld shifts for two protons (N—H and H ) in the solution NMR spectrum of the host—guest complex.Note that one hydrogen bond also stabilizes this complex. (Reprinted in part with permission from J. Org. Chem. 63 p. 6153. Copyright 1998 American Chemical Society). some guests accounted for a substantial variation in the relationship of the symmetry axes between the host and guests. Despite the analytical power that NMR provides in supramolecular chemistry its use is not free from caveats. Mitra et al. have reminded us about the large proton chemical shift dependence upon the concentration of small molecules. Interestingly the proton chemical shift appears to be uniquely prone to this concentration e.ect as the C chemical shift scalar coupling and T values show minimal variation with concentration.These results suggest that it is necessary to verify any changes in the chemical shift as a function of concentration for guest species alone. Otherwise conclusions drawn about binding strength and nature of association in supramolecular systems are liable to be incorrect. 9 Improvements in existingNMR methods Continued improvements on a variety of well-known NMR techniques continue to breathe new life into old pulse sequences. The INADEQUATE and HMBC experiments are among those which have been enhanced recently. Inadequate The double quantum .ltered C—C correlation experiment (INADEQUATE) is one experiment that has failed to ful.l its true potential due to a severe lack of sensitivity with natural abundance samples.Freeman et al. have found that the introduction of oxygen into INADEQUATE samples greatly enhances the sensitivity by reducing the spin-lattice relaxation times (T ) from 20 s down to 6 s. This is easily 414 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 accomplished by injecting oxygen gas with a syringe into a standard ‘‘high pressure’’ sample tube (Wilmad 528-TR-7). Alternatively the use of per.uoro-tert-butyl alcohol as the solvent can reduce relaxation times for quaternary sites down to 4 s or less. These vast decreases in the spin-lattice relaxation times are accompanied by only modest reductions in the nuclear Overhauser enhancements (NOE’s). These authors also demonstrated that the inclusion of an INEPT pulse sequence in front of the INADEQUATEsequence provides an additional sensitivity enhancement in the form of polarization transfer from attached protons.This will increase sensitivity for all coupled C-pairs in a molecule if at least one of them has an attached proton. Routine use of these facile techniques can result in a .ve-fold reduction in the length of a typical INADEQUATE experiment. HMBC The Heteronuclear Multiple Bond Correlation (HMBC) has been used extensively in the structure determination of organic molecules. This proton-detected experiment is one of the most important techniques for identifying long range H—C correlations. For example it is highly useful in .nding correlations across glycosidic bonds in oligosaccharides. One problem with the HMBC experiment however is that H—H J-modulation during the evolution period causes line broadening of the carbon signals in the indirect dimension.A common solution to this problem in other NMR experiments is to employ a .xed evolution period and vary the locations of the pulses within that period. Seto et al. have developed two new constant-time HMBC experiments. Incorporation of a .xed-length evolution period prevents H—HJ-modulation of the carbon chemical shifts in the indirect dimension. An alternative sequence includes an additional 180° proton pulse during the constant time evolution period to suppress long range C—H scalar coupling as well as proton J-modulation in the indirect dimension. Wagner and Berger took a di.erent approach to the same problem with the HMBC experiment. They implemented the aforementioned ‘‘accordion’’ principle by varying the length of the delay time and the t evolution period simultaneously.This experiment facilitates sampling of a broad range of scalar coupling constants which provides a complete set of long-range C,H connectivities. The authors also implemented an improved two-stage low-pass .lter removing one-bond H—C scalar coupling. This made broadband (GARP) C-decoupling feasible providing better sensitivity for the HMBC experiment. Further improvements in HMBC have been obtained from post-processing cleanup of t noise. Doss developed a new technique whereby the individual f spectra are di.erentially scaled down by a factor proportional to their respective levels of rms noise. This technique capitalizes on the fact that intense t noise is usually associated with very intense peaks which can be scaled down without signi.cant loss of information content.Moreover it is easily implemented in speci.c regions of the spectrum. 10 Miscellaneous Bain et al. have developed a useful internal standard for use in quantitative NOE 415 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 measurements. The compound 1,5-dichloro-2,4-dimethoxybenzene (12) is very useful because it provides an essentially full NOE (i.e. 50% enhancement) when the methoxy protons are saturated. References 1 T.R. Kelly J. P. Sestelo and I. Tellitu J. Org. Chem. 1998 63 3655. 2 L.B. Krivdin and S. V. Zinchenko Curr.Org. Chem. 1998 2 173. 3 S.W. Homans Biochem. Soc. Trans. 1998 26 551. 4 H. Shimizu J. M. Brown S. W. Homans and R. A. Field Tetrahedron 1998 54 9489. 5 B.A. Salvatore and J. H. Prestegard Tetrahedron Lett. 1998 39 9319. 6 S. Zhao G. Bondo J. Zajicek and A. S. Serianni Carbohydr. Res. 1998 309 145. 7 B. Bose S. Zhao R. Stenutz F. Cloran P. B. Bondo G. Bondo B. Hertz I. Carmichael and A. S. Serianni J. Am. Chem. Soc. 1998 120 11 158. 8 Q. Xu and C. A. Bush Carbohydr. Res. 1998 306 335. 9 F.G. Vogt and A. J. Benesi J. Magn. Reson. 1998 132 214. 10 A. J. Dingley and S. Grzesiek J. Am. Chem. Soc. 1998 120 8293. 11 K. Pervushin R. Riek G. Wider and K. Wuthrich Proc. Natl. Acad. Sci. 1997 94 12 366. 12 T. A. Mobley F. Muller and S. Berger J. Am. Chem.Soc. 1998 120 1333. 13 M.D. Pelta H. Barjat G. A. Morris A. L. Davis and S. J. Hammond Magn. Reson. Chem. 1998 36 706. 14 H. Barjat G. A. Morris and A. G. Swanson J. Magn. Reson. 1998 131 131. 15 O. Millet and M. Pons J. Magn. Reson. 1998 131 166. 16 M.J. Shapiro Am. Lab. 1998 30 20. 17 M. Lin M. J. Shapiro and J. R. Wareing J. Org. Chem. 1997 62 8930. 18 L. Or. M. Lin and C. K. Larive Anal. Chem. 1998 70 1339. 19 K. Bleicher M. Lin M.J. Shapiro and J. R. Wareing J. Org. Chem. 1998 63 8486. 20 N. Gonnella M. Lin M.J. Shapiro J. R. Wareing and X. Zhang J. Magn. Reson. 1998 131 336. 21 A. Chen C. S. Johnson M. Lin and M. J. Shapiro J. Am. Chem. Soc. 1998 120 9094. 22 J. M. Seco S. K. Latypov E. Quinoa and R. Riguera J. Org. Chem. 1997 62 7569. 23 I. Chataigner J.Labreton D. Durand A. Guingant and J. Villieras Tetrahedron Lett. 1998 39 1759. 24 S. K. Latypov R. Riguera M.B. Smith and J. Polivkova J. Org. Chem. 1998 63 8682. 25 S. K. Latypov J. M. Seco E. Quinoa and R. Riguera J. Am. Chem. Soc. 1998 120 877. 26 H. Yamase T. Ooi and T. Kusumi Tetrahedron Lett. 1998 39 8113. 27 S. K. Latypov M. J. Ferreiro E. Quinoa and R. Riguera J. Am. Chem. Soc. 1998 120 4741. 28 M. Kobayashi Tetrahedron 1998 54 10 987. 29 G. Uccello-Barretta A. Iuliano E. Franchi F. Balzano and P. Salvadori J. Org. Chem. 1998 63 9197. 30 G. Prestat A. Marchand J. Lebreton A. Guingant and J.-P. Pradere Tetrahedron Asymmetry 1998 9 197. 31 J. Lacour C. Ginglinger F. Favarger and S. Torche-Haldimann Chem. Commun. 1997 2285. 32 C. Ginglinger D.Jeannerat J. Lacour S. Juge� and J. Uziel Tetrahedron Lett. 1998 39 7495. 33 A. Bilz T. Stork and G. Helmchen Tetrahedron Asymmetry 1997 8 3999. 34 D. Merlet A. Loewenstein W. Smadja J. Courtieu and P. Lesot J. Am. Chem. Soc. 1998 120 963. 35 A. Meddour D. Atkinson A. Loewenstein and J. Courtieu Chem. Eur. J. 1998 4 1142. 36 P. Lesot D. Merlet A. Loewenstn and J. Courtieu Tetrahedron Asymmetry 1998 9 1871. 37 D. Merlet B. Ancian W. Smadja J. Cortieu and P. Lesot Chem. Commun. 1998 2301. 38 B. C. Hamper S. A. Kolodziej A. M. Scates R. G. Smith and E. Cortez J. Org. Chem. 1998 63 708. 39 Y. Luo X. H. Ouyang R. W. Armstrong and M. M. Murphy J. Org. Chem. 1998 63 8719. 40 A. Svensson K. E. Bergquist T. Fexand J. Kihlberg Tetrahedron Lett. 1998 39 7193. 41 D.Stones D. J. Miller M.W. Beaton T. J. Rutherford and D. Gani Tetrahedron Lett. 1998 39 4875. 42 J. Chin B. Fell S. Pochapsky M. J. Shapiro and J. R. Wareing J. Org. Chem. 1998 63 1309. 416 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 43 M.J. Shapiro and J. R. Wareing Curr. Opin. Chem. Biol. 1998 2 372. 44 J.-L. Wolfender K. Ndjoko and K. Hostettmann Curr. Org. Chem. 1998 2 575. 45 P. A. Keifer Drugs Future 1998 23 301. 46 K. Ute R. Nimi S. Hongo and K. Hatada Polym. J. 1998 30 439. 47 L. Gri.ths and R. Horton Magn. Reson. Chem. 1998 36 104. 48 H. Pasch W. Hiller and R. Haner Polymer 1998 39 1515. 49 B. Vogler I. Klaiber G. Roos C. U. Walter W. Hiller P. Sandor and W. Kraus J. Nat. Prod. 1998 61 175. 50 K. Stott J. Keeler Q. N. Van and A. J. Shaka J. Magn. Reson. 1997 125 302. 51 E. Garo J.-L. Wolfender K. Hostettmann W. Hiller S. Antus and S. Mavi Helv. Chim. Acta 1998 81 754. 52 G. E. Martin R. C. Crouch and A. P. Zens Magn. Reson. Chem. 1998 36 551. 53 A. Ross M. Salzmann and H. Senn J. Biomol. NMR 1997 10 389. 54 J. Kang G. Hilmersson J. Santamaria and J. Rebek J. Am. Chem. Soc. 1998 120 3650. 55 M.J. Cloninger and H. W. Whitlock J. Org. Chem. 1998 63 6153. 56 W.H. Tan N. Niino T. Ishikura A. Maruta T. Yamamoto and Y. Matsui Bull. Chem. Soc. Jpn. 1998 71 1285. 57 A. Mitra P. J. Seaton R. A. Assarpour and T. Williamson Tetrahedron 1998 54 15 489. 58 D. L. Mattiello and R. Freeman J. Magn. Reson. 1998 135 514. 59 K. Furihata and H. Seto Tetrahedron Lett. 1998 39 7337. 60 R. Wagner and S. Berger Magn. Reson. Chem. 1998 36 S44. 61 G. A. Doss Magn. Reson. Chem. 1998 36 135. 62 A. D. Bain E. P. Mazzola and S. W. Page Magn. Reson. Chem. 1998 36 403. 417 Annu. Rep. Prog. Chem. Sect. B 1999 95 395—417 MMMM

ISSN:0069-3030    

DOI:10.1039/a805983g

出版商:RSC    

年代:1999

数据来源: RSC