12 Alkaloids By D. J. ROBINS Department of Chemistry University of Glasgow Glasgow G12 800 1 Introduction A colossal amount of new work has been reported in the five years since alkaloids were last reviewed in Annual Reports.’ Three more volumes of Specialist Periodical Reports on the alkaloids are available,24 which cover the period July 1979-June 1982. The biosynthesis of alkaloids is included in these Reports and in Volume 7 of the Specialist Periodical Report on biosynthesis which covers a three year period 1979-198 1.5More recent aspects of alkaloid chemistry and biochemistry are treated in the journal Natural Product Reports which has superseded the Specialist Periodical Reports on alkaloids and biosynthesis and which began in 1984. Several new monographs on alkaloid chemistry have appeared.68 The period covered by this selective review is 1980-1984 inclusive.Two particular themes that have become increasingly apparent over the past five years merit special attention. One is the use of chiral starting materials to synthesize alkaloids in optically active form and the other is the shift towards the utilization of stable isotopes and cell-free preparations to elucidate details of biosynthetic pathways to alkaloids. 2 Pyrrolidine Alkaloids The biosynthesis of nicotine continues to generate much attention. The pyrrolidine ring is known to be derived from ornithine. Leete and Yu9 fed DL-[2,3-’3c2,5-’4c]-ornithine (1) to Nicotiana glutinosa plants. The I3C n.m.r. spectrum of the biosyn- thetically derived nicotine (3) exhibited satellites of equal intensity at C-2’ C-3’ and C-4’ C-5’ due to contiguous I3C atoms (Scheme 1).This result confirmed the D. G. Buckley Annu. Rep. hog. Chem. Sect B 1979 76 382. * ‘The Alkaloids’ ed. M. F. Grundon A Specialist Periodical Report The Royal Society‘of Chemistry London 1981 Vol. 11. ‘The Alkaloids’ ed. M. F. Grundon A Specialist Periodical Report The Royal Society of Chemistry London 1982 Vol. 12. ‘The Alkaloids’ ed. M. F. Grundon A Specialist Periodical Report The Royal Society of Chemistry London 1983 Vol. 13. E. Leete in ‘Biosynthesis’ ed. R. B. Herbert and T. J. Simpson A Specialist Periodical Report The Royal Society of Chemistry London 1983 Vol. 7 chapter 4. D. R. Dalton. ‘The Alkaloids. The Fundamental Chemistry.A Biogenetic Approach’ Marcel Dekker New York and Basel 1979. ’ M. Hesse ‘Alkaloid Chemistry’ Wiley-Interscience Somerset New Jersey 1981. G. A. Cordell ‘Introduction to Alkaloids. A Biogenetic Approach’ Wiley-Interscience Somerset New Jersey 1981. E. Leete and M.-L. Yu Phyrochemistry 1980 19 1093. 29 1 292 D. J. Robins Scheme 1 symmetrical labelling of the pyrrolidine ring from ornithine probably via putrescine (2). Furthermore when [ l-I3C methylamino-'5N]-N-methylputrescine(4) was fed by Leete and McDonell to N. tabacum only C-5' of nicotine (6) showed a satellite in its 13C n.m.r. spectrum due to an adjacent '*N (J15N-~3c4.2 Hz)." This labelling pattern is consistent with the incorporation of the precursor (4) into nicotine via the N-methyl-1-pyrrolinium ion (5) formed after oxidation of the primary amine of (4) to the aldehyde (Scheme 2).The stereochemistry of the enzymic processes Scheme 2 involved in transforming putrescine into nicotine has been deduced by Wigle et al." (R)-[l-2H]Putrescine (7) was incorporated into nicotine and the labelling pattern (8) was evident from the 2H n.m.r. spectrum. The presence of 2H at the 2'-and 5'4 R) positions indicates that the pro-S hydrogen is stereospecifically lost from C-1 when N-methylputrescine is oxidized to 4-methylaminobutanal. This stereochemistry is in accord with other reactions catalysed by diamine oxidase. Attack of the pyridine-ring precursor on the N-methyl-1-pyrrolinium ion (5) takes place on the l-si,2-re face to yield (S)-nicotine (8).(7) An unusual feeding technique whereby aqueous solutions of the precursors containing a detergent were painted on the leaves of Erythroxylon coca had to be used to obtain reasonable incorporations of precursors into cocaine (9). The label from DL-[5-'4C]ornithine [cf (l)] was shown by Leete to be equally distributed between the two bridgehead carbons C-1 and C-5 in cocaine (9).12 Thus cocaine is derived from ornithine again via the symmetrical intermediate putrescine (2). It should be noted that the biosynthesis of some other structurally related tropane alkaloids derived from ornithine does not involve any symmetrical intermediates. 10 E. Leete and J. A. McDonell J. Am. Chem. Soc. 1981 103 658. 1. D. Wigle L.J. J. Mestichelli and I. D. Spenser J. Chem. Soc. Chem. Commun. 1982 662. 12 E. Leete J. Chem. Soc. Chem. Cornmun. 1980,1170; J. Am. Chem. Soc. 1982 104 1403. Alkaloids 293 Leete has also demonstrated that C-2 C-3 C-4 and C-9 of cocaine (9) are derived as expected from acetic acid.13 After feeding [l-'4C]acetic acid 48% of the 14C label was found to be located at C-3 and 38% was at C-9. The higher activity at C-3 is attributed to C-4 and C-3 constituting the starter unit and therefore containing a higher level of activity than the second unit (C-2 and C-9) which is added via malonate. [4-3H]Phenylalanine was also incorporated specifically into the benzoic acid moiety of cocaine (9) confirming previous res~1ts.l~ A tropane alkaloid subhirsine with a most unusual structure (10) has been isolated by Russian workers from Convoluulus ~ubhirsutus.'~ 3 Piperidine and mridine Alkaloids A novel sulphur-containing monoterpenoid glucosidic alkaloid xylostosidine (1 1 ) has been isolated from aqueous extracts of Loniceru xylosteum L." Lysine is known to be a specific precursor for the piperidine ring of anabasine (12).Leete has confirmed these findings by feeding ~~-[4,5-'~C,,6-'~C]lysine (13) to Nicotiana glauca.I6 Satellites for contiguous I3C atoms at (2-4' and C-5' in anabasine (12) were observed in its I3C n.m.r. spectrum. The 14C radioactivity was located almost entirely (98% ) at C-6' of anabasine. Lysine is therefore incorporated into anabasine without going through a symmetrical intermediate such as cadaverine (14) ( CJ nicotine and cocaine biosynthesis).(1 1) (12) (13) (14) Muscopyridine (16) is one of the rare mammalian alkaloids known. It is present in the musk deer and plays a role in communication when used for territorial marking. Two syntheses of (R)-(+):muscopyridine (16) have been reported." In one shown in Scheme 3 the desired enantiomer was formed by reduction of the double bond in (15) using a chiral borane reagent derived from (-)-a-pinene. l3 E. Leete Phytochemistry 1983 22 699. 14 S. F. Aripova E. G. Sharova and S. Yu. Yunusov Khim. Prir. Soedin. 1982 640 (Chem. Abstr. 1983 98 160979). l5 R. K. Chaudhuri 0. Sticker and T. Winkler Helu. Chim. Actu 1980 63 1045. 16 E. Leete J. Nut. Rod. 1982 45 197. 17 K.Utimoto S. Kato M.Tanaka. Y.Hoshino S. Fujikura and H. Nozaki Heterocycles 1982 18 149. 294 D. J. Robins / iii J *-(16) (15) Reagents i AIC13 high dilution; ii NH,; iii tetrachloro-o-benzoquinone;iv MeLi Scheme 3 4 Pyrrolizidine Alkaloids Significant progress has been made in the study of the synthesis and the biosynthesis of pyrrolizidine alkaloids in the past five years. In particular the synthesis of a few natural macrocyclic pyrrolizidine alkaloids has been accomplished and the biosyn- thetic pathway to retronecine (17) has been clarified by use of I3C- I5N- and 'H-labelled precursors. The structures of the 200+ known pyrrolizidine alkaloids are depicted in a review.'* Macrocyclic diester pyrrolizidine alkaloids occur with 1 1- 12- or 13-membered rings.Five new alkaloids which are triesters each containing a 14-membered ring have been isolated by Edgar et all9 from Parsonsia species (fam. Apocynaceae). The novel structure (18) proposed for one of these compounds parsonsine has been confirmed by X-ray diffraction studies on two crystalline modifications of (18) which exhibit different folding patterns for the macrocyclic portion of the molecule.20 HO H 6H20H (17) (18) (19) '' D. J. Robins Forschr. Chem. Org. Natursr. 1982 41 115. I9 J. A. Edgar N. J. Eggers A. J. Jones and G. B. Russell Tetrahedron Lett. 1980 21 2657. N. J. Eggers and G. J. Gainsford Cryst. Struct. Commun. 1979 8 597; 1980. 9 173. Alkaloids 295 The synthesis of natural macrocyclic alkaloids has been a long-standing challenge in this area.The first step forward was taken by Robins and co-workers when they achieved the preparation of unnatural 11 -membered macrocyclic pyrrolizidine diesters exemplified by (19).21 Treatment of (+)-retronecine (17) with 3,3-dimethyl- glutaric anhydride gave a mixture of the 7-and 9-monoesters. These were lactonized uia their corresponding pyridine-2-thiolesters to give the pyrrolizidine aikaloid analogue (19) in 75% yield. The large difference in chemical shift of 1.24 p.p.m. for the C-9 protons of (19) supports the formation of a macrocycle and suggests that the conformation of (19) may be different from those of 1 1-membered macrocyclic pyrrolizidine alkaloids where lower values have been observed." The hydrobromide of (19) was readily metabolized by liver oxidase enzymes to the corresponding toxic pyrrole derivative and showed hepatotoxic effects similar to those of the common macrocyclic pyrrolizidine alkaloid monocrotaline.22 Similar strategy was used by Robins and co-workers to prepare (+)-dicrotaline (20) and its C-13 e~imer.~~ One of the separated products was identical with natural dicrotaline isolated from Crotalaria dura seeds.The absolute configuration at C-13 in both compounds was established by a series of selective reactions on each epimer to yield optically active mevalonolactone (21) (Scheme 4). Dicrotaline (20) has the largest value so far recorded (1.24 p.p.m.) for the chemical shift difference between the C-9 protons for an 11-membered pyrrolizidine alkaloid containing retronecine.HO Me I 11 + t-lo (20) Reagents i H2/Pt02 AcOH; ii Na liq.NH3 Scheme 4 A total synthesis of the 1 1-membered macrocyclic alkaloids (*)-crispathe (27) and (*)-fulvine has been achieved by Vedejs and Lar~en.~~ Crispatic anhydride was prepared and the tertiary hydroxy function was protected as its methoxymethyl ether (22) (Scheme 5). Coupling of the mixed phosphoric anhydride (23) was carried out with the lithium alkoxide of protected (*)-retronecine (24)2' to afford (25) and a diastereoisomer. Lactonization was effected by displacement of the methanesul- phonate group in (26) with the liberated carboxylate anion. The diastereoisomeric products were separated to yield (*)-crispathe (27).An analogous series of reactions " D. J. Robins and S. Sakdarat J. Chem. SOC.,Chem. Commun. 1980 282; J. A. Devlin D. J. Robins and S. Sakdarat J. Chem. SOC. Perkin Trans. I 1982 1117. 22 A. R. Mattocks Chem.-Biol. Interact. 1981 35 301. 23 J. A. Devlin and D. J. Robins J. Chem. SOC. Chem. Commun. 1982 1272; K. Brown J. A. Devlin and D. J. Robins 1. Chem. SOC. Perkin Trans. I 1983 1819. 24 E. Vedejs and S. D. Larsen J. Am. Chem. SOC.,1984 106 3030. 25 E. Vedejs and G. R. Martinez J. Am. Chem. SOC. 1980 102 7993. 296 D. J. Robins M eMe A,OCH,OMe..Me "08 ,OSiMe,Bu' d i ii 0 + 0 (EtO),PO CO,CH,CH,SiMe N Me ?CH20Me M~ OCH,OMe iv e-- Me* 0 v1 VII \ 0 ($ Reagents i Me2A10CH2CH,SiMe3 ; ii ( Et+02POCl; iii Bu"Li 4-dimethylaminopyridine; iv HF; v MeS02CI Et3N vi MeCN.Bu",NF vii BF3.Et20,EtSH Scheme 5 from fulvinic anhydride yielded (*)-fulvine [(27) with opposite stereochemistry at c-131. The first synthesis of a 12-membered pyrrolizidine alkaloid was carried out by Japanese workers.26 They started with a stereoselective synthesis of (*)-integerrinecic acid (28).27A series of steps was necessary to protect the free carboxyl group in (28) cleave the &lactone protect the tertiary hydroxy group and form the linear anhydride (29). This anhydride was coupled with the lithium alkoxide of protected (*)-retronecine (24) to produce a monoester mixture (30). Lactonization was achieved by nucleophilic displacement of the methanesulphonylmethyl group by alkoxide to yield the cyclized product and a diastereoisomer (Scheme 6).These products were separated and removal of the protecting group from one racemate under acidic conditions afforded (*)-integertimine (31). K. Narasaka T. Sakakura T. Uchimaru K. Monmoto and T. Mukaiyama Chem. Lett. 1982 445. 2'7 K. Narasaka and T. Uchimaru Chem. Lett. 1982 57. Alkaloids 297 Y li CO,CH,SMe ll-v -(31) (30) Reagents i Bu"Li 4-dimethylaminopyridine; ii hH,F; iii H202 (NH4)6M~7024; iv Bu"Li; V Zn H2S04 Scheme 6 Retronecine (17) was first synthesized by Geissman and Waiss in 1962.28There has been a long gap until further syntheses have been reported.29 Rueger and Benn3' have made the (+)-lactone (32) from natural (-)-4-hydroxy-~-proline in 12 steps.This is an intermediate in the original route to (*)-retronecine,28 and has been converted into (+)-retronecine by an improved pr0cedu1-e.~~ (+)-Croalbinecine (33) and (-)-platynecine (34) were also produced.31 An alternative route to the (+)-lactone (32) has been reported by Buchanan et al?2 from a carbohydrate precursor. Synthetic routes to optically active pyrrolizidine bases have blossomed and a further selection includes (-)-hastanecine (35) from (R)-malic (+)-heliotridine (36) from (S)-malic acid,34 and (-)-rosmarinecine (37) from D-glUCOSamine.35 H@ 0 'KHHw20H .-H OH NH N N 28 T. A. Geissman and A. C. Waiss 1. Org. Chem. 1962 27 139. 29 J. J. Tufariello and G. E. Lee J. Am. Chem. Soc. 1980 102 373; G. E. Keck and D.G. Nickell ibid. p. 3634; T. Ohsawa M. Ihara K. Fukumoto and T. Kametani Heterocycles 1982 19 2075; H. Niwa A. Kuroda and K. Yamada Chem. Leu. 1983 125. 30 H. Riieger and M. Benn Heterocycles 1982 19 23. 31 H. Riieger and M. Benn Heterocycles 1983 20 1331. 32 J. G. Buchanan G. Singh and R. H. Wightman J. Chem. SOC. Chem. Commun.,1984 1299. 33 D. J. Hart and T.-K. Yang J. Chem Soc. Chem. Commun. 1983 135. 34 A. R. Charnberlin and J. Y. L. Chung J. Am. Chem. SOC.,1983 105 3653. 35 K. Tatsuta H. Takahashi Y. Arnerniya and M. Kinoshita J. Am. Chem. SOC. 1983 105 4096. 298 D. J. Robins Significant progress has been made in the past few years in the understanding of the biosynthesis of retronecine (17) chiefly by use of precursors containing stable isotopes.This work began with the demonstration by Khan and Robins of the first complete labelling patterns in retronecine by I3C n.m.r. spectroscopy after feeding [1 ,4-13C2]- and [2,3-13C2]-putrescine.36The labelling pattern obtained with [1,2-CJputrescine is illustrated (38).3' These results confirmed that retronecine is derived from two molecules of putrescine. Evidence for the involvement of a later symmetrical C4- N-C4 intermediate in the biosynthetic pathway to retronecine was provided by the use of a l3C-I5N doubly labelled precursor. Retronecine derived from [l-amin~-'~N,l-~~C]purescine (39) showed enhanced 13C signals for the peaks due to C-3 -5 -8 and -9 in the I3C n.m.r. spectrum.38s39 In addition the signals for C-3 and C-5 of retronecine showed satellites due to 13C-15N coupling.The presence of equal amounts of the two labelled species (40) and (41) indicates that a later symmetrical intermediate is involved in retronecine biosynthesis. This intermediate was shown to be homospermidine by use of 14C-38 and I3C-labelled h~mospermidine.~' Thus [1 ,9-13C2]homospermidine (42) was fed to Senecioisatideus plants and the I3C n.m.r. spectrum of the derived retronecine displayed doublets around the natural abundance signals for C-8 and C-9 with a geminal coupling constant of ca. 6 Hz,indicating that homospermidine is incorporated intact into retr~necine.~' Further support for the role of homospermidine in pyrrolizidine alkaloid biosynthesis was provided by Robins.41 [1 ,9-14C]Homospermidine (42) was converted into [''C]-trachelanthamidine (44) using enzymes under physiological conditions.Treatment of homospermidine with diamine oxidase and reduction of the cyclized aldehyde (43) with a dehydrogenase gave the saturated pyrrolizidine base (44). Degradation of the ''C-labelled base (44)yielded methylamine (Scheme 7) containing 51 % of the total activity of the base indicating that homospermidine is not broken down (e.g. to putrescine) before formation of trachelanthamidine. 36 H. A. Khan and D. J. Robins J. Chem. SOC. Chem. Commun. 1981 146; J. Chem. Soc. Perkin Trans. I 1985 101. 37 D. J. Robins J. Chem. Res. (S) 1983 326. 38 H. A. Khan and D. J. Robins J. Chem. SOC.,Chem. Commun. 1981 554. 39 G. Grue-Sorensen and I.D. Spenser 1.Am. Chem. Soc. 1981 103 3208. 40 J. Rana and D. J. Robins J. Chem. Res. (S),1983 146. 4' D. J. Robins J. Chem. SOC.,Chem. Commun. 1982. 1289. Alkaloids 299 (43) (44) iii iv I Reagents i pea seedling diamine oxidase + catalase; ii liver alcohol dehydrogenase; iii SOCl,; iv LiAIH,; v CrO, H,S04; vi NaN3 H2S04 Scheme 7 Information about the stereochemistry of the enzymic processes involved in retronecine biosynthesis has been obtained using ’H-labelled precursors. The label- ling patterns in retrorsine (49) derived biosynthetically from [1 ,4-’H4]- and [2,3-’H4]- putrescine were established by ’H n.m.r. spectro~copy.~’ The formation of (9S)-[9- ’Hlretrorsine from the former precursor is consistent with stereospecific reduction of an aldehyde precursor as for a normal coupled dehydrogenase enzyme system.This work has been extended by the use of chiral [l-’H]putres~ine~~*~ to establish a number of the stereochemical details in retronecine biosynthesis (Scheme 8). Initial &:+7 __* H2NLNJNH2 NH OHC (45) /CHO CHO 1 (47) CHzOH 1 do N (48) Scheme 8 42 J. Rana and D. J. Robins J. Chem. SOC.,Chem. Commun. 1983 1222. 43 G. Grue-Sorensen and 1. D. Spenser J. Am. Chem SOC.,1983 105 7401. 44 J. Rana and D. J. Robins J. Chem. Soc. Chem. Commun. 1984 517. 300 D. J. Robins oxidation of putrescine to 4-aminobutanal takes place with loss of the pro-S hydro-gen. Reduction of the imine (49 formed by coupling of putrescine with 4-aminobutanal occurs by hydride attack on the si-face of the imine to yield homosper- midine (42).Two further oxidation steps each take place with removal of the pro-S hydrogens to afford a dialdehyde (46). Cyclization of the corresponding iminium ion (47) occurs by attack on its re-face to give the 8a-pyrrolizidine (48). Reduction of the aldehyde takes place on the re-face of the carbonyl group. Further insight into the stereochemistry of pyrrolizidine alkaloid biosynthesis is likely to result from the use of other 2H-labelled precursors. 5 Indolizidine Alkaloids The occurrence and synthesis of indolizidine alkaloids has been re~iewed.~’ Swain-sonine (57) has been isolated from Swainsona canescens,46 the spotted locoweed (AstragaZus lentiginosus),4’ and the fungus Rhizoctonia Zeguminicol~.~~ It is a potent inhibitor of the enzyme a-mannosidase.This disruption of the processing of gly- coproteins may. cause locoism a chronic neurological disorder of grazing animals. Three syntheses of (-) -swainsonine from carbohydrate precursors have been recently The route developed by Richardson and co-workers is out- lined in Scheme 9.49Selective protection of the 3-amino-3-deoxy-a-~-mannopyrano-side (50) gave the crystalline diol (51). The free amine formed on hydrogenolysis of (51) cyclized when heated at reflux in ethanol containing sodium acetate. The product was isolated as its N-benzyloxycarbonyl derivative (52). Acid hydrolysis of (52) yielded the furanose (53) which was condensed with ethanethiol under acidic conditions to give the protected aldehyde (54).The aldehyde group was liberated from the corresponding triacetate and condensation with ethoxycarbonylmethyl- enetriphenylphosphorane in a Wittig reaction yielded the a,@-unsaturated ester (55). Hydrogenation of the double bond in (55) also removed the protecting group and generated the indolizidinone (56).Reduction of the lactam with borane and deacetyl- ation with sodium methoxide afforded (-) -swainsonine (57) identical with natural material and obtained in 2.7% overall yield from (50). All of the chiral centres of the aminohexose (50) are incorporated intact into swainsonine (57). Slaframine (61) is another toxin produced by Rhizoctonia Zeguminicola. Swain- sonine and slaframine are both formed from L-lysine via pipecolic acid (58) and the remaining two carbon atoms are derived from mal~nate.’~ The indolizidinone (59) has been shown to be an intermediate in the biosynthetic pathway to sla- frami~~e.’~ In further studies [2,3,4,5,6-2H9]pipecolic acid was incorporated into slaframine (61) with the loss of two deuterium atoms from C-6 suggesting that the ketone (60) is involved in the biosynthetic pathway (Scheme The same feeding 45 E.Gellert J. Not. Prod. 1982 45 50. 46 S. M. Colegate P. R. Dorling and C. R.Huxtable Aust. J. Chem. 1979 32 2257. 47 R.J. Molyneux and L. F. James Science 1982 216 190. 48 M. J. Schneider F. S. Ungemach H. P. Broquist and T. M. Harris Tetrahedron 1983 39 29. 49 M. H. Mi L. Hough and A. C. Richardson J.Chem. SOC.,Chem. Commr:n. 1984,447. so G. W.J. Fleet M. J. Gough and P. W. Smith Tetrahedron Lett. 1984 25 1853. 51 T. Suami K. Tadano and Y.Iimura Chem. Lett. 1984 513. 52 E. C. Clevenstine H. P.Broquist and T. M. Hams Biochemistry 1979 18 3659. 53 F. P. Guengerich S. J. DiMari and H. P. Broquist J. Am. Chem. SOC.,1973 95 2055. 54 M. J. Schneider F. S. Ungemach H.P. Broquist and T. M. Hams J. Am. Chem. SOC.,1982 104,6863. Alkaloids 301 "O? Tso> HoI OH -ZN*OH t (54) (53) 1 Et0,CCH =Ckts AcO--OAcAcO 0 (55) 2 =COOCHzPh (56) Scheme 9 Scheme 10 experiment yielded swainsonine (57) which had lost two deuterium atoms from C-8 and C-8a. The hydroxylation at C-8 accounts for the loss of one of these atoms; loss of the other (from C-8a) can be attributed to the formation of an iminium ion (62).Hydride attack on this iminium ion would then lead to the inversion of configuration at C-8a necessary in the formation of swainsonine (57). 302 D. J. Robins Castanospermine (63) is a toxic alkaloid isolated from seeds of the Australian legume Castanosperrnurn austr~le.~~ The structure and relative configuration of castanospermine were deduced from an X-ray crystal determination. 6 Quinolizidine Alkaloids A short stereoselective route to a-isosparteine (65) involving nitrones as intermedi- ates has been developed by Oinuma et al. (Scheme 11).56Sequential reaction of two molecules of 1-piperideine 1-oxide with pyran gave the intermediate (64).Reductive cleavage of the N-0 bonds in (64) followed by iminium ion formation and reduction of the iminium ions led to a-isosparteine (65). H H iii c- (64) 1 Reagents i 140 "C; ii 190 "C iii H2 PdO.H,O MeOH Scheme 11 Major contributions have been made to our understanding of the biosynthesis of quinolizidine alkaloids. The biosynthetic pathway to these alkaloids is known to proceed from L-lysine via cadaverine (66). Crude enzyme preparations from cell 55 L. D. Hohenschutz E. A. Bell P. J. Jewess D. P. Leworthy R. J. Pryce E. Arnold and P. J. Clardy Phytochemistry 1981 20 811. 56 H. Oinuma S. Dan and H. Kakisawa J. Chem. Soc. Chem. Commun. 1983 654. Alkaloids 303 suspension cultures of Lupinus polyphyllus have been shown to catalyse the conver- sion of cadaverine into 17-oxosparteine (67) in the presence of pyruvic acid.s7 This suggests that transamination reactions are occurring with the pyruvic acid acting as a receptor for the amino groups in cadaverine.Since no intermediates were detected during the biosynthetic process a series of enzyme-linked intermediates on an enzyme complex was postulated by Wink et uZ.,~~and 17-oxosparteine (67) was proposed as a key intermediate in the biosynthesis of tetracyclic quinolizidine alkaloids. Further evidence has been provided by the use of stable isotopes. Three [l-arnino-”N l-’3C]cadaverine units (68) are incorporated to about the same extent into sparteine (69).58,59 Two 13C-”N doublets were observed in the 13C n.m.r.spectrum of sparteine indicating that two of these units are incorporated into the outer rings of sparteine in a specific fashion (69) (Scheme 12). In a similar manner two Scheme 12 cadaverine units (68) were shown to be incorporated into the bicyclic alkaloid lupinine (70) in L. luteus but only one 13C-”N doublet was observed in the I3C n.m.r. spectrum.59B60 This finding demonstrates that a later C5-N-C5 symmetrical intermediate is not involved in lupinine biosynthesis and thus provides an interesting contrast to pyrrolizidine alkaloid biosynthesis. The stereochemical courses of a number of the enzymic reactions involved in lupinine biosynthesis have been established by feeding chiral [1 -2H]cadaverines to 57 M. Wink T. Hartmann and H.-M.Schiebel Z. Naturforsch. Teil C 1979 34 704. 58 J. Rana and D. J. Robins J. Chem. Soc. Chem. Commun. 1983 1335. 59 W. M. Golebiewski and I. D. Spenser J. Chem. SOC.,Chem. Commun. 1983 1509. 60 J. Rana and D. J. Robins J. Chem. Soc. Chem. Commun.. 1984 81. 304 D. J. Robins L. For example formation of (1 1 S)-lupinine from (I?)-[ l-2H]cadaverine is consistent with attack of hydride at the re-face of the carbonyl of an aldehyde intermediate in the biosynthetic pathway. This result is analogous to that obtained in retronecine biosynthesis. Complete labelling patterns in sparteine and other tetracyclic quinolizidine alkaloids derived from (R)-and (S)-[l-2H]cadaverines have been establi~hed.6~~~~ In particular the presence of 2H at C-17 in all the alkaloids derived from (R)-[1 -2H]cadaverine clearly demonstrates that 17-oxosparteine cannot be an intermediate in the biosynthesis of the tetracyclic quinolizidine alkaloids as suggested by Wink et aLS7 7 p-Phenylethylamine Alkaloids A number of new alkaloids with unusual structural features have been discovered.Bharatamine (72) isolated from Alangium Zamarckii has a novel substitution pat- tern.64 The structure of bharatamine was confirmed by its synthesis from the enamine (71) involving cyclization and debenzylation. Karachine (73) is present in Berberis aristata and has a more typical substitution pattern on both aromatic rings but somewhat strangely appears to incorporate two acetone residues6’ Another new alkaloid (*)-chilenine (79 was isolated from a different species in the same genus 61 W.M. Golebiewski and I. D. Spenser J. Am. Chem. SOC,1984 106 1441. 62 A. M. Fraser and D. J. Robins J. Chem. SOC.,Chem. Commun. 1984 1477. 63 W. M. Golebiewski and I. D. Spenser J. Am. Chem. SOC.,1984 106 7925. 64 S. C. Pakrashi R. Mukhopadhyay P. P. G. Dastidar A. Bhattacharya and E. Mi Tetrahedron Lett 1983 24 291. 65 G. Blasko N. Murugesan A. J. Freyer M. Shamma A. A. Ansari and Atta-ur-Rahman J. Am. Chem. SOC.,1982 104 2039. Alkaloids 305 B. empetrifoZia.66Spectroscopic studies showed that (*)-chilenine is an isoindolo- benzazepine; this is a ring system that has not previously been encountered in Nature. The alkaloid (75)was identical to material that had been prepared earlier by oxidation of berberine to the hydroxyketone (74)followed by rearrangement of (74)in base.67 Norlaudanosoline (79)is widely believed to be derived biosynthetically from dopamine (76)and 3,4-dihydroxyphenylpyruvicacid (77)by condensation to give (78),followed by decarboxylation and reduction of the imine produced.68 Some data that contradict these findings have been presented by Zenk and co-~orkers.~~ They isolated an enzyme that catalyses the formation of laudanosoline from several plant species that are known to produce isoquinoline alkaloids.The substrates for the enzyme were dopamine and 3,4-dihydroxyphenylacetaldehyde. 3,4-Dihy-droxyphenylpyruvic acid was not a substrate. The product from a large scale incubation with the enzyme from Eschscholtzia tenufolia was (S)-norlaudanosoline (79),but the material isolated had only 25% optical purity.The racemic material is presumably formed by a non-enzymic process. It is clear however that the role of a-ketoacids like (77),and the amino acid (78)in the biosynthesis of isoquinoline alkaloids should be subject to further scrutiny. Hn HO "q CO,H HO O q 'HH HO o r (76) Hz "H " Hoe' HO 8 Indole Alkaloids Much progress has been made in the synthesis of indole alkaloids in optically active form. A few examples have been selected. Kozikowski introduced the use of an intramolecular [3 + 21cycloaddition reaction with the nitrile oxide derived from a nitroethylindole [cJ (82)]in his synthesis of (*)-chano~lavine.~~ A similar strategy was employed for his first total synthesis of an ergot alkaloid in optically active form." The route to (+)-paliclavine (86)is outlined in Scheme 13.Wittig reaction of the N-protected indole aldehyde (80) with the optically active phosphorane (81) yielded an unsaturated alcohol which 66 V. Fajardo V. Elango B. K. Cassels and M. Shamma Tetrahedron Lett. 1982 23 39. 67 J. L. Moniot D. M. Hindenlang and M. Shamma J. Org. Chem. 1979 44 4343; C. Manikumar and M. Shamma Heterocycles 1980 14 827. 60 A. I. Scott S.-L. Lee and T. Hirata Heterocycles 1978 11 159. 69 M. Rueffer H. El-Shagi N. Nagakura and M. H. Zenk FEBSLett. 1981 129 5. 70 A. P. Kozikowski and H. Ishida J. Am. Chem. SOC.,1980 102 4865. 71 A. P. Kozikowski and Y. Y. Chen J.Org. Chem. 1981 46 5248; A. P. Kozikowski Acc. Chem. Res. 1984 17 410. 306 D. J. Robins CHO MsO 0-N i-iv v-viii (80) Ts + ___ ' NH 0-NMe xii t- \ Reagents i heat; ii dihydropyran pyridine p-TsOH; iii KOH MeOH; iv H,C=CHNO,; v PhNCO Et3N; vi Ac,O 4-dimethylaminopyridi$ejvii Dowex ion exchange resin H+ form; viii MsCl Et3N; ix PhSeNa then NaIO,; x Me30BF,; xi LiAlH,; xii Hg/AI; xiii MeCHO Scheme 13 was protected. Michael addition of this indole to nitroethene led to the nitro compound (82). The corresponding nitrile oxide was generated and underwent an intramolecular 1,3-dipolar cycloaddition reaction to afford after mesylation (83). The cycloaddition was not stereospecific but the mixture of diastereoisomeric mesylates could be separated.The double bond in (84) was introduced by displace- ment of the mesylate with a phenylseleno group followed by oxidative elimination. The final chiral centre was introduced by methylation of the isoxazoline (84) followed by hydride reduction. This led to a mixture of (85) and its C-5 epimer which had to be separated. Reductive cleavage of the N-0 bond in the minor product (85) gave (+)-paliclavine (86) while further condensation of (86) with ethanal afforded (+)-paspaclavine (87). Alkaloids 307 The first synthesis of (-)-antirhine has been reported by Takano and co-~orkers.~~ The chiral starting material was L-glutamic acid and this was converted into the optically active lactone (88).73The preparation of the key lactone aldehyde (89) then required 16 stages and proceeded in 14% overall yield.Condensation of (89) with tryptamine gave the amide (90),and reductive cyclization then led to (-)-antirhine (91)(Scheme 14).The intermediate (89)may be useful in the preparation of other alkaloids of the corynantheine-yohimbine type. II Reagents i tryptamine NaBH3CN aq.MeOH pH 6; ii DIBAH -78 "C; iii H@+ Scheme 14 The optically active lactone (88)used in the preceding synthesis has been converted into the (+)-and (-)-forms of quebrachamine by Takano et uL74*75The route to (+)-quebrachamine (93)is shown in Scheme 15.74The formation of epimers (92) was not a handicap in the synthesis as this chiral centre is not present in the final product (93). In continuation of their work on biomimetic alkaloid synthesis Brown and Pratt have reported the first biomimetic synthesis of the ring E carbocyclic yohimbine alkaloids (Scheme 16).76Treatment of the methyl ester (94)of secoxyloganin with P-glucosidase at pH 7 gave two epimeric aldehyde diesters (95).The synthesis was completed by acetylation and condensation with tryptamine leading to the yohim- bine acetate epimers (96),and their 19,20-didehydroanalogues.A rapidly growing feature in the study of the biosynthesis of terpenoid indole alkaloids is the use of enzyme preparations from plant tissue cultures. Further work on the biosynthetic pathway to the heteroyohimbine alkaloids has been published. Geissoschizine dehydrogenase has been isolated and partially purified from cell suspension cultures of Cutharunthus rose~s.~~ This enzyme catalyses the conversion 72 S.Takano N. Tamura and K. Ogasawara J. Chem. SOC.,Chem. Commun. 1981 1155. 73 S. Takano M. Yonaga and K. Ogasawara Synthesis 1981 265. 74 S. Takano K. Chiba M. Yonaga and K. Ogasawara J. Chem. SOC.,Chem. Commun. 1980 616. 75 S. Takano M. Yonaga and K. Ogasawara J. Chem SOC.,Chem. Commun. 1981 1153. 76 R. T. Brown and S. B. Pratt J. Chem. SOC.,Chem. Commun. 1980 165. 77 A. Pfitzner and J. Stockigt Phytochemisrry 1982 21 1585. 308 D. J. Robins -iii-v "Oyy (88) . .. CH,CH=CH Et CH,CH=CH H (93) Reagents i LiNPr', CH,=CHCH,Br -78°C; ii LiNPr', EtBr -78°C; iii HCl EtOH; iv NaOH v. NaIO,; vi tryptamine AcOH; vii B,H, DMS; viii NaOH H202;ix LiAIH,; x MsCl; xi Na NH3 Scheme 15 OGlc + OH (94) (95) Scheme 16 of geissoschizine (97) a shunt metabolite into 4,21-didehydrogeissoschizine (98) which is an important intermediate at a branch point in the biosynthetic pathway to ajmalicine (99).The enzyme was shown to remove the 21-pro-R hydrogen in geissoschizine in a reaction dependent on NDAP+ (Scheme 17). The ergot alkaloid elymoclavine (102) is believed to be formed biosynthetically from chanoclavine I (100) via the aldehyde (101). Hassan and Floss78 have used chirally tritiated (100) to demonstrate that the 17-pro-R hydrogen is lost and the 17-pro-S hydrogen is retained in the conversion of (100) into (102) (Scheme 18). The steric course of this reaction is the same as for reactions catalysed by yeast and '13 S.B. Hassan and H. G. Floss J. Nor. Prod. 1981. 44,756. Alkaloids 309 Me0,C 'YMe OH OH (97) (98) J (99) Scheme 17 ofic HOH,C __* __* liver alcohol dehydrogenases and indicates that (101) is an intermediate in the conversion of chanoclavine I into elymoclavine. This conclusion is supported by the isolation of the aldehyde (101) from a Claviceps mutant that is unable to make the tetracyclic ergot alkaloids.79 The first cell-free preparations that are able to make ergot alkaloids have been reported by Groger and co-workers." 79 W. Maier D. Erge and D. Groger Planta Med. 1980,45 104. W. Maier D. Erge B. Schumann and D. Groger Biochem. Biophys. Rex Commun. 1981 99 155 W.Maier D. Erge and D. Groger FEMS Microbiol. Lett. 1981 12 143.