14 Biological Chemistry Part (ii) Organic Peroxides Biological and Synthetic Aspects By W. ADAM" Department of Chemistry University of Puerto Rico Rio Piedras Puerto Rico 00931 USA and A. J. BLOODWORTH Department of Chemistry University College London 20 Gordon Street London WCl HOAJ The current reintensification of interest in the chemistry of organic peroxides stems from the recognition that organic peroxides have an important and varied role in a biological context. The identification of peroxides as key products in biosynthetic pathways has presented organic chemists with new synthetic challenges. These challenges have in turn stimulated the development of new synthetic methodology in the peroxide area. In this review we attempt to highlight these developments.1 Biological Perspective The discovery of ascaridole in 1908 as principal constituent of chenopodium oil and its characterization in 1912 as the endoperoxide structure (1) marks the beginning of (1) the challenging and fruitful era of biologically relevant peroxides.* Thereby it was illustrated that the reaction of biological matter with molecular oxygen or its derivatives (hydrogen peroxide superoxide ion ozone etc.) leads to peroxide- containing products which survive the biological conditions despite the labile nature of the peroxide linkage. a National Institutes of Health Career Development Awardee (1975-80) K.Gollnick and G 0.Schenck in '1,CCycloaddition Reactions' ed. J. Hamer Academic Press 1967 Chap. 10; H. H. Szmant and A.Halpern J. Amer. Chem. Soc. 1949,71,1133. 342 Biologica1 Chemistry Nevertheless it took about half-a-century until full-scale exploration of this important field began. This is indeed rather paradoxical since biological systems must abound with bio-peroxides in view of the fact that molecular oxygen is essential for sustaining life of most organisms. Thus biological matter in cells continuously interacts with molecular oxygen transforming it into hydrogen peroxide superoxide ion etc. all known to degrade cell components by oxidation presumably but not necessarily via the labile peroxides. In part the reason for the delay on the progress of detecting isolating and identifying bio-peroxides can be ascribed to the contro- versial case of ergosterol endoperoxide (2).It was first discovered by Windaus (1928:)’ in connection with his Vitamin D studies but claimed as a ‘natural’ endoperoxide by Wieland and Prelog (1947),3who isolated it from the mycelium Aspergillus fumigatus which was cultured in the dark. However first Schenck’ and more recently Arditti4 suggested that (2) was formed as an adventitious by-product due to photo-sensitized oxygenation of ergosterol. However a careful recent investigation’ clearly established that the endoperoxide (2) is the product of an enzymatic oxygenation. The mechanistic origin of ascaridole (l),which is produced in the leaves of the plant is clearly by way of photosensitized oxygenation in which the chlorophyll serves as sensitizer sunlight as radiation source a-terpinene as substrate and atmospheric oxygen as oxidant.However ergosterol endoperoxide (2) is formed in the dark as the result of enzymatic oxygenation. The biomechanistic details of this interesting process are still not well understood but it is clear that oxygenating enzymes (oxygenases) must be involved. The detection purification and charac- terisation of a number of these important enzymes6 during the last twenty-five years provided impetus in the search for bio-peroxides. The latter must be the initial products of enzymatic oxygenation and consequently their detection isolation and identification form the cornerstone in understanding the action of oxygenases on biological matter. Much progress has been made on bio-peroxides and enzymatic oxygenation especially during the last decade.To summarize this progress exhaus- tively would be difficult since new bio-peroxides are being discovered or postulated daily. However we shall highlight the salient features of this exciting development in the interest of providing an overview and perspective. Stable Peroxides of Marine and Plant Origin.-Let us illustrate the diversity and complexity of stable peroxides that have been isolated and characterized during the A. Windaus and J. Brunken J. Liebig Ann. Chem. 1928,460,225. P. Wieland and V. Prelog Helv. Chim. Acta 1947,30 3272. J. Arditti R. Ernst M. H. Fisch andB. H. Flick J.C.S. Chem. Comm. 1972 1217. M. L. Bates W. W. Reid and J. D. White J.C.S. Chem. Comm. 1976.44. ‘Molecular Mechanisms of Oxygen Activation’ ed.0.Hayaishi Academic Press,1974. W. Adam and A. J. Bloodworth last few years from a variety of natural sources especially of marine origin. An unusual structure is rhodophytin (31 which was isolated from marine algae of the genus Laurencia (Rhodophyta).' The remarkable structural feature of this novel marine natural product is the vinyl peroxide moiety. To date no stable model compound of chemical origin possessing this elusive peroxide functionality is known which underlines the uniqueness of this bio-peroxide. The evidence on which the structure (3) is based is convincing but its thermal stability contradicts chemical experience. An X-ray structure confirmation seems important. From sponge of the Chondrilla genus the stable natural peroxide (4) containing the peroxyketal structure was isolated.8 Again the structural assignment appears convincing except that the HC1-catalysed conversion of (4) into a chloro-enone via a cyclic vinyl peroxide seems unlikely.A simpler and more probable decomposition mode of (4) into chloro-enone would be a Criegee-Hock rearrangement. Another stable bioperoxide derived from the Caribbean sponge Plakortis halichondrioides is plakortin (5).9 Unlike the stable steroidal peroxides possessing the ergosterol endoperoxide structure (2) found in sponges,'o which are derived from adventitious photo-oxygenation (4) and (5) appear to be of enzymatic origin. Both are formed stereospecifically rather than as a mixture of diastereoisomers as in the case of (2).The interesting neoconcinndiol hydroperoxide (6) was recently isolated from the red seaweed Laurencia sugderiae." It is still uncertain whether this unprecedented diterpene hydroperoxide (6) is the result of a dioxygenase biogenic pathway or an artifact of ene-type hydroperoxylation with singlet oxygen. ' W. Fenical J. Amer. Chem. SOC.,1974 96 5580. R. J. Wells Tetrahedron Letters 1976,2637. M. D. Higgs and D. J. Faulkner J. Org. Chem. 1978,43,3454. lo E. Fattorusso S. Magno C. Santacroce and D. Sica Garzetta 1974,104 409; Y. M. Sheikh and C. Djerassi Tetrahedron 1974 30,4095. B. M. Howard W. Fenical J. Finer K. Hirotsu and J. Clardy J. Amer. Chem. SOC.,1977,99,6440. Biological Chemistry (6) Novel natural peroxides of plant origin are (7) and (8),isolated respectively from the leaves of Eucalyptus grandis and from fungi.I3 The hydroxy-1,2-dioxy- cyclohex-4-ene structure (7) represents the ring tautomer of the hydroperoxy isomer (7a) which most likely is the product of a photosensitized ene-type oxygenation in the leaves of Eucalyptus grandis.The fungi peroxides verruculogen (8a) and fumitre- morgin A (8b) are unusual and it is as yet unknown whether they are produced by enzymatic oxygenation. The physiological activity of these stable natural peroxides is as diverse as their varied structural features. Thus the sponge endoperoxide plakortin (5) shows antimicrobial activity,' the Eucalyptus grandis derived peroxide (7) inhibits root formation,'* and the fungus metabolites verruculogen (8a) and fumitremorgin A (8b) are tremorgenic agents.13 (7) CH Ho-o oyo (8b)R= (74 Diepoxide and Furan Metabolites.-Several natural products possess structural units consisting of either cis-diepoxides (10) or furans (11).Some of these are thought to be derived via metabolic transformation of the intermediate endo- peroxides (9) arising from the oxygenation of 1,3-dienes (Scheme 1).Although this biogenetic pathway has not been established for most of the examples that are cited studies of model compounds amply support this supposition. Furthermore bio- mechanistically it constitutes the most expedient rationalization of their formation. One of the earliest cis-diepoxides to be recognized was crotepoxide (12) isolated from the fruits of Croton macrostachys.l4 This novel tumor-inhibitory antibiotic was l2 W.D. Crow W. Nicholls and M. Sterns TetrahedronLetters 1971 1353. 13 (a)J. Fayos. D. Lokensgard,J. Clardy R. J. Cole and J. W. Kirksey J. Amer. Chem. Soc. 1974 % 6785; (b)N. Eickman J. Clardy R. J. Cole and J. W. Kirksey Tetrahedron Letters 1975 1051. 14 S. M. Kupchan R. J. Hemingway P. Coggon A. T. McPhail and G. A. Sim J. Amer. Chem. SOC.,1968 90,2983. W.Adam and A. J. Bloodworth 0 II "OAc recently synthesized" using the pathway in Scheme 1. A structurally related natural diepoxide is the fungal antibiotic (13).16 Also the unusual antileukemic diterpenoid triepoxides triptolide and tripdiolide (14) isolated from the roots of Tripterygium wilfurdii," might have a similar biosynthetic history.The related diterpene diepoxide stemolide (15) from the leaves of Stemodia maritima is not antitumoric.18 (14a) R=H (i4bj R=OH The naturally occurring 3-alkylfurans perillaketone (16) a-clausenane (17) and ipomeamarone (18) were efficiently synthesized from their respective dienes via singlet oxygenation and subsequent dehydration (Scheme l).I9 This constitutes )JJ ,J 0 / 0 (16) (17) (18) Is M. R. Demuth P. E. Garrett and J. D. White J. Arner. Chem. SOC..1976 98,634. l6 D. B. Borders P. Shu and J. E. Lancaster,J. Arner. Chem. Soc. 1972,94,2540; H.-J. Altenbach and E. Vogel Angew. Chern. 1972 84,9. S. M. Kupchan W. A. Court R. G. Dailey jun. C. J. Gilmore and R. F. Bryan J.Amer. Chem. SOC. 1972,94,7194. P. S. Marchand and J. F. Blount Tetrahedron Letters 1976 2489. l9 K. Kondo and M. Matsumoto Tetrahedron Letters 1976,4363. Biological Chemistry therefore a likely pathway for the biogenesis of terpene derived substances. The rather drastic conditions for the dehydrationlg were recently avoided by employing lithium di-isopropylamide and p-toluenesulfonyl chloride” or the biosynthetically more significant ferrous sulfate-catalysed transformation.21 AflatoxinBiogenesis.-These extremely toxic materials produced in peanuts by the fungus Aspergillus fravus are also among the most potent carcinogenic natural products.2’ Their fascinating biosynthetic history involving intermediate peroxide- containing metabolites is still controversial.BUchiz3 first suggested that aflatoxin B1(21) might be derived from the endoperoxide (19) via the established sterig- matocystin (20). At that time no chemical model studies were available to support the proposed rearrangement of endoperoxide (19) and the pyran (22) was proposed as intermediate. This novel biogenetic hypothesis was tested rigorously by experi- ments involving 13C n.m.r. spectroscopy and the results revealed that the naph- thacene endoperoxide (19)could not be the precursor to (21) via (22).23 (Starred positions represent 13C atoms derived from carboxy carbon-labelled acetic acid). OH OH CH ‘OH 0 ‘ti. (22) However recentz4 studies on the decomposition of model endoperoxides (Scheme 2) revive the endoperoxide (19) as a plausible intermediate metabolite in the aflatoxin biogenesis.Application of the mechanistic sequence of Scheme 2 to the endo- peroxide (19) affords aflatoxin B1(21) with the correct labelling pattern. The fact that the Czo-polyketide metabolite averufin (24) leads to aflatoxin B by the action of Aspergillus para~iticus’~ speaks however against the C18-naphthacene 2o B. Harirchian and P. D. Magnus Synthetic Comm. 1977,119. ” J. A. Turner and W. Herz J. Org. Chem. 1977,42 1900. ’’J. A. Miller ‘Toxicants Occurring Naturally in Foods’,National Academy of Sciences Washington D.C. 1973;E. K. Weisburger. Chemistry 1977,50,42 23 P. S. Steyn R. Vleggaar P. L. Wessels and D. B. Scott J.C.S. Chem. Comm. 1975 193. 24 M.K.Logani W.A. Austin and R. E. Davies Tetrahedron Letters 1978,511 ;J. Rigaudy C. Breliere and P. Scribe Tetrahedron Letters 1978 687. 25 D. P. H. Hsieh R. C. Yao D. L. Fitzell and C. A. Reece J. Amer. Chem. SOC.,1976,98 1020. W.Adam and A. J. Bloodworth \ '01 R Scheme 2 endoperoxide (19)intermediate. Since the biosynthetic history of (24)is still obscure and since the ketal (23) which is structurally similar to (24) is an endoperoxide product (Scheme 2) an endoperoxide precursor is still viable in the biogenesis of aflatoxin B1. Prostaglandin Endoperoxide.-The establishment of the intervention of the endo- peroxide (25) in the metabolism of arachidonic acid emphasizes the importance of peroxides in biological oxidations. The labile endoperoxide (25) serves as precursor to the pharmacologically potent hormonal agents PGE (26a) PGF (26b) throm- boxane (26c) and prostacyclin (26d).Fortunately two recent reviews26 spare us the task of summarizing the structural synthetic and physiological aspects of this vast field. R. R= MCOzH; R'= OH 26 K. H. Gibson Chem. SOC.Revs. 1977,6,489;K.C.Nicolaou G. P. Gasic and W. E. Barnette Angew. Chem. Internat. Edn.. 1978,17,293. Biological Chemistry The prostaglandin endoperoxide (25) was first postulated in 1967,27 isolated and characterized in 1973,28 and synthesized from PGF (26b) in 1977.29 Recent work suggests that the prostaglandin endoperoxide synthetase is a glycoprotein with an easily dissociable haem prosthetic group which exhibits both oxygenase and peroxi- dase a~tivity.~’ The ferrous sulphate-catalysed transformation of endoperoxides to y-hydroxyketones was suggested as a chemical model for the biosynthesis of PGE (26a) thromboxane (26c) and prostacyclin (26d).31 Lipid Hydroperoxides.-The pathological consequences especially in ageing of cellular lipid peroxides has been convincingly summarized by Bland32 in a recent review.The labile lipid hydroperoxides (27) formed from linoleic acid by the action of lipoxygenase and molecular oxygen,33 serve as initiators of free radical chain oxidation of cell components. More recently the intriguing hydroperoxy- 1,2-diox- acyclopentanes (28) have been characterized in the enzymatic oxygenation of FH2\ CH3CH2-CH=CH-CH=CH-CH CH-CHfCH&CO,Me \/I 0-0 OOH (28d /CH2\ CH3CH2-CH-CH CH-CH=CH-CH=CHfCH2+C02Me (!)OH ‘0-0’ (28b) methyl lin01enate.~~ Steroids such as cholesterol afford the hydroperoxides (29) as intermediate metabolites on enzymatic ~xygenation.~’ ’’ M.Hamberg and B. Samuelsson J. Biol. Chem. 1967 242 5336. 28 M. Hamberg and B. Samuelsson Proc. Nut. Acad. Sci.. USA 1974,71 345; D.H. Nugteren and E. Hazelhof Biochem. Biophys. Acta 1973 326,448. 29 R. A. Johnson E. G. Nidy L. Baczynskyj and R. R. Gorman J. Amer. Chem. Soc. 1977,99,7738. 30 F. J. van der Ouderaa M. Buytenhek D. H. Nugteren and D. A. van Dorp Biochim. Biophys. Acta 1977,487,315. 31 J. A. Turner and W. Herz Experentia 1977,33 1133. 32 J. Bland J. Chem. Edn. 1978,55 151.33 H. W.-S. Chan V. K. Newby and G. Levett J.C.S. Chem. Comm. 1978,82. 34 M.Roza and A. Francke Biochim. Biophys. Acta 1978,528 119. 35 L.L.Smith M. J. Kulig D. Muller and G. A. S. Ansari J. Amer. Chem. Soc. 1978,100,6206. W.Adam and A. J. Bloodworth c,H17 HOLYP (29) Homogentisic Acid Biogenesis.-On the basis of “0labelling experiments it was that the enzymatic oxygenation of p-hydroxyphenylpyruvic acid leading to homogentisic acid (Scheme 3) trespassed through the quinol hydroperoxide (30). Although this biosynthetic pathway was ~hallenged,~’ more recent studies” of model systems make the intervention of (30) plausible. CO,H CO,H o=c I O=C I 6H HO \/CO H HO‘cI’o* * I II c.o* CO:H* I ?H O* cG* H2cfJ 6 Hz8 0 OH 0 0 OH (30) Scheme 3 Plastoquinone Peroxides.-The photochemical destruction of electron transport quinones such as the plastoquinones and menaquinones has been rationalized in terms of peroxide intermediate^.^^ For example model studies4’ on the photo- oxygenationof plastoquinone-1 afforded the stable 1,2,4-trioxacyclohexane product (31).Similarly the menaquinone-9 hydroperoxide (32) which appears to be responsible for the photo-oxidized degradation of menaq~inone,~~ could originate from the intermediate 1,2,4-trioxacyclohexane (32a).39 36 B. Lindblad G. Linstedt and S. Linstedt J. Amer. Chem. SOC., 1970,92 7446. 37 A.S.Widman A. H. Soloway R. L. Stern and M. M. Bursey Bioorg. Chem. 1973 2 176. 38 I. Saito Y. Chujo H. Shimazu M. Yamane T.Matsuura and H. J. Cahnmann J. Amer. Chem. SOC. 1975,97,5272. 39 R.M.Wilson S. W. Wunderly J. G. Kalmbacher and W. Brabender Ann. New York Acad. Sci.,1976 267,201. 40 D.Creed H. Werbin and E. T. Strom J. Amer. Chem. SOC. 1971 93 502. Biologica1 Chemistry 351 Bilirubin Photo-oxygenation.-The lack of hepatic catabolism of bilirubin and consequently its accumulation in tissue are the cause for jaundice in premature babies. To avoid brain damage the common treatment for neonatal jaundice is near-u.v. irradiation of the infant which promotes bleaching of the bilirubin a yellow pigment presumably through autosensitized photo-oxygenation and subsequent breakdown of the intermediate peroxides into smaller soluble fragments. Chemical model studies suggest that photo-oxygenation of bilirubin probably proceeds through the peroxides (33),which represents an efficient and convenient mechanistic rati~nalization.~~ In fact Similar peroxide products are formed with bili~erdin.~~ the catabolic conversion of haem into biliverdin has been interpreted to proceed via peroxide intermediate^.^^ Me R' Me R2 RZ Me Me R' I H I H I H I H (334 Me R' Me R2 R2 Me Me R' H H H H Formylkynurenine Biosynthesis.-The important class of oxygenase enzymes were discovered in connection with tryptophan metabolism.Without this discovery it would have been difficult to understand biological oxygenations of organic matter.6 Even so it took about a quarter of a century to confirm through studies of model compounds that the tryptophan hydroperoxide (34) is the precursor to formylky- nurenine in the oxygenation of tryptophan (Scheme 4).45 Similar model studies on other model indole derivatives in support of tryptophan 2,3-dioxygenase action have recently been reviewed.46 Aromatic Hydroxylation by mavine Monoxygenase.-Conflicting mechanistic rationalizations of biological hydroxylation of aromatic rings by molecular oxygen 41 C.D. Snyder and H. Rapoport J. Amer. Chem. SOC. 1969,91,731. 42 D. A. Lightner and G. B. Quistad F.E.B.S.Letters 1972,25,94; D. A. Lightner,G. S. Bisacchi and R. D. Norris J. Amer. Chem. SOC. 1976,98 802. 43 I. B. C. Matheson and M. M. Toledo Photochem. Photobiol. 1977,25,243; D. A. Lightner and D. C. Crandall Tetrahedron Letters 1973 953.44 J.-H.Fuhrhop S. Besecke J. Subramanian Chr. Mengersen and D. Riesner J. Amer. Chem. Soc. 1975 97,7141. 45 M. Nakagawa H. Watanabe S. Kodato H. Okajiura T. Hino J. L. Flippen and B. Witkop Proc. Nat. Acad. Sci. USA,1977.744730. 46 I. Saito T. Matsuura M. Nakagawa and T. Hino Accounts Chem. Res. 1977 10 346. W.Adam and A. J. Bloodworth NH-C-H II 0 Scheme 4 with the help of flavine monoxygenase have been presented. A number of inter-mediates derived from the dihydroflavin which are capable of transferring a single oxygen to the aromatic substrate have been So far none have been rigorously established through chemical model studies but all seem to be derived from the dihydroflavin hydroperoxide (35). Recent chemical work lends support to the intervention of 4a-hydroperoxyflavin (35)and similar structures have been suggested in bacterial biolumine~cence.~~ Firefly Bioluminescence.-Another class of bio-peroxides namely the a-peroxy- lactones are the active intermediates in biolumines~ence.~~ The a-peroxylactone structure (36)of firefly luciferin was postulated over ten years ago,” but it took a decade of controversy to confirm through isotopic labelling experiment^,^' its intervention.Although these bio-peroxides are too unstable to permit isolation model compounds have been prepared’* and shown to emit light on thermal decomposition.53 0 (36) 47 G. I. Dmitrienko V. Snieckus and T. Viswanatha Biorg. Chem. 1977 6 421; H. W. Orf and D. Dolphin Proc.Nut. Acad. Sci. U.S.A.,1974 71 2646. 48 C. Kemal T. W. Chan and T. C. Bruice J. Amer. Chem. SOC. 1977 99 7272; C. Kemal and T. C. Bruice J. Amer. Chem. Soc. 1977,99,7064;J. W. Hastings and K. H. Nealson Ann.Rev. Microbiol. 1977,31 549. 49 W. Adam J. Chem. Educ. 1975,51,138;J. W. Hastingsand T. Wilson Photochem. Photobiol. 1976,23 461. T. A. Hopkins H. H. Seliger E. H. White and M. W. Cass J. Amer. Chem. SOC.,1967,89 7148; F. McCapra Y. C. Chang and V. P. Francois J.C.S. Chem. Comm. 1968,22. 0. Shimomura T. Goto,and F. H. Johnson Proc. Nut. Acad. Sci. U.S.A.,1977,74,2799;J. Wannlund M. DeLuca K. Stempel and P. D. Boyer Biochem. Biophys. Res. Comm. 1978 81,987. 52 W. Adam A. AlzCrreca J.-C. Liu and F. Yany J. Amer. Chem. SOC.,1977 99 5768.53 W. Adam 0.Cueto andF. Yany J. Amer. Chem. SOC.,1978,100,2587. Biological Chemistry 353 The multitude and diversity of bio-peroxides that intervene as stable or labile intermediates in the metabolic pathways of all sorts of biological matter is evident from the examples sited here. Undoubtedly future work in this new and exciting field should uncover and confirm many more biologically important peroxides and lead to a greater understanding of biological oxygenations and their physiological consequences. 2 Synthetic Methods Organic peroxides have always presented a considerable synthetic challenge in view of their marked thermal instability and high chemical reactivity. The labile inter- mediates postulated in biological oxidations are thus particularly formidable pre- parative targets that have demanded the development of new synthetic methods of exceptional mildness.The methodology of organic peroxide synthesis as it stood in 1960 is summarized with great clarity in Davies’ m~nograph.’~ Preparative developments during the next decade are well covered in three volumes edited by Swern.” Subsequently a number of significant advances have been made and it is the aim of this Section to highlight these. For the purposes of discussion we shall divide the methods into three categories according to the nature of the reagent that introduces the O2moiety. Thus new modifications to the alkylation and acylation of hydrogen peroxide and related nucleophiles are treated first. Next we summarize the synthetically useful aspects of singlet oxygenation and finally we present some preparatively important develop- ments in triplet oxygenation.Nucleophilic Displacements by Hydrogen Peroxide and Related Species.-The reaction of hydrogen peroxide at electrophilic carbon atoms provides the main route to many simple organic peroxides. Thus alkyl hydroperoxides and dialkyl peroxides are obtained by SNldisplacements at tertiary carbon atoms or SN2displacements at primary and secondary carbon atoms. Strongly acidic or basic media and long reaction times at elevated temperatures are often required and can cause extensive decomposition of the desired product. Modifications aimed at engendering reac- tivity under milder conditions primarily involve using more powerful alkylating agents but include the use of alternative nucleophiles such as the superoxide ion.Silver-salt-assisted Alkylation by Alkyl Halides. The first successful application of this technique appears to have been Kopecky’s well known tetra-alkyldioxetane synthe~is.~~ hydroperoxide upon shaking with Thus 3-bromo-2,3-dimethyl-2-butyl a suspension of silver acetate or benzoate in dichloromethane gave a mixture (Scheme 5) from which tetramethyldioxetane (37)was isolated in 30% yield. The dioxetane of 1,2-dimethylcyclohexene was similarly prepared but use of P-iodoalkyl hydroperoxides was found to give better yields and to be essential in the preparation of the h9~’0-octalin dioxetane. 54 ‘Organic Peroxides’ A. G. Davies Butterworths London 1961.55 ‘Organic Peroxides’ ed. D. Swern Wiley-Interscience 1970 Vol. 1; 1971 Vol. 2; 1972 Vol. 3. 56 K. R. Kopecky J. E. Filby C. Mumford P. A. Lockwood and J. Y. Ding Cunud.J. Chem. 1975,53 1103. W.Adam and A. J. Bloodworth HOO Me I 1 Me-C-C-Me I I Me Br AgOAc ___+ 0-0 I1 Me-C-C-Me 1 1 Me Me OOH I //CH2+ Me-C-C he ‘Me 0 II+ Me,C-C-Me (37) Scheme 5 The intermolecular reaction with silver trifluoroacetate (Scheme 6)57provides a convenient and general preparation of alkyl hydroperoxides and dialkyl peroxides. AgOzC.CF R’OOH +R2X R’OOR2 +AgX+ H02C.CF3 Scheme 6 Alkyl bromides are suitable for making tertiary and secondary derivatives but iodides are required in the preparation of primary peroxides. Seventeen examples with yields of 27-93% were reported.Some di-t-alkyl peroxides (e.g. Me2PriCOOMe2Pri)were obtained for which the conventional method of alcohol sulphuric acid and hydrogen peroxide fails because dehydration of the alcohol or formation of rearranged products prevail. It appears that chlorides are sufficiently reactive to permit the synthesis of allylic hydro peroxide^.^^ Silver nitrate was used in the preparation of cyclohexenyl hydroperoxide but it was necessary to employ the trifluoromethanesulphonate with 3-chloro-2-methoxycyclohexene(Scheme 7) to avoid substantial competitive incorporation of the silver salt’s anion. Ag03SCF3 GOMe + H2°z -Scheme 7 The intramolecular reaction has been extended to y-bromoalkyl hydroperoxides thereby providing syntheses of 1,2-dioxacyclopentanes (Scheme 8)” and of bicyclic peroxides (Scheme 9)60including 2,3-dioxabicyclo[2,2 l]heptane61 (see Section 3).Results to date suggest that the silver salt method is excellent for making dialkyl peroxides both acyclic and (bi)cyclic but is less successful as a hydroperoxide R NBS R 0-0 OOH Scheme 8 57 P. G. Cookson A. G. Davies and B. P. Roberts Chem. Comm. 1976 1022. A. A. Frimer J. Org. Chem. 1977,42 3194. 59 W. Adam A. Birke C. Cidiz S.Diaz and A. Rodriguez J. Org. Chem. 1978,43 1154. 6o A. J. Bloodworth and B. P. Leddy Tetrahedron Letters 1979,729. 61 N. A. Porter and D. W. Gilmore J. Amer. Chem. SOC.,1977,99,3503. 355 Biologica1 Chemistry __+ J3r2 oooH Scheme 9 synthesis. High yields of allylic hydroperoxides were reporteds8 but the method utilized a large excess (10-14 fold) of 98% H202which is potentialty hazardous and should be avoided if possible.It is advisable to carry out the reactions in the dark to avoid possible formation of metallic silver which can catalyse peroxide decom- position; silver trifluoroacetate appears to be the reagent of choice. The alkylation proceeds with predominant inversion of configuration,62 a factor of considerable importance in the design of reactions leading to bicyclic peroxides.60*61 Alkylation by Alkyl Trifuoromethanesulphonates and N-Alkyl-N'-tosylhydrazines. Alkyl trifluoromethanesulphonates are sufficiently powerful to alkylate t-butyl hydroperoxide under non-alkaline conditions. Acceptable yields (33-56%) of secondary alkyl peroxides have been obtained without accompanying elimination and under conditions where the corresponding alkyl methanesulphonates do not react (Scheme Scheme 10 The method has been extended to the synthesis of 1,2-dio~acycloalkanes~~ (Scheme 11; n = 1-4) and 2,3-dioxabicyc10[2,2,l]heptane~~(see Section 3) by using bis(trimethylstanny1) peroxide as the nucleophile; bis(trimethylsily1) peroxide 0-0 Scheme 11 A very promising new method for the preparation of primary and secondary alkyl hydroperoxides involves the oxidation of N-alkyl-N'- tosylhydrazines (Scheme 12).66 H202/Na202 R-NH-NH-TS AROOH Scheme 12 Neopentyl hexadecyl cyclohexyl cis-and trans-2 -methylcyclohexyl and some steroidal hydroperoxides were obtained in yields (by h.p.1.c.analysis) of 87--95% ; " A. G. Davies and A. J. Sotowicz personal communication. 63 M. F. Salomon R. G. Salomon and R. D. Gleim J. Org. Chem. 1976 41 3983. '* M. F. Salomon and R. G. Salomon J. Amer. Chem. SOC.,1977,99,3500. " R. G. Salomon and M. F. Salomon J. Amer. Chem. SOC.,1977,99,3501. '' L. Caglioti F. Gasparrini D. Misiti and G. Palmieri Tefiuhedron 1978 34 135. W.Adam and A. J. Bloodworth yields of isolated products were not quoted. The N-alkyl-N'- tosylhydrazines are obtained by reducing the corresponding tosylhydrazones (from aldehydes or ketones) or N-acyl-N'-tosylhydrazines (from carboxylic acids) and isolation prior to oxidation is unnecessary. A notable advantage of the method is that it employs low strength (30%)H202.Alkylation by Alkenes and Epoxides. As a route to organic peroxides the alkylation of hydrogen peroxide by carbonium ions generated by protonation of alkenes is very limited in scope. Reactions induced by other electrophiles (Scheme 13) are more \/ c-c I c=c \/E+ /\+/\ROOH -H+ 'ROO-C-C-E 1 /\ E I1 Scheme 13 successful. Thus sources of positive halogen such as N-bromosuccinimide and 1,3-di-iodo-5,5-dimethylhydantoinprovide the P-halogenoalkyl hydroperoxides that are used in dioxetane synthesis (Scheme 5).56 Except for the recent demon- stration of a related intramolecular reaction6' and of the extension (with difficulty) to cyclopr~panes,~~~~~ there have been no developments of note since Kopecky's original introduction of the use of N-haloamides a decade ago.68 Peroxymercuration (Scheme 14) where the electrophile is a mercury(I1) salt has been extensively developed since its discovery in 1969 and has proved to be R'CH=CHR2 +ROOH +HgX2 +R1CH(OOR)CH(HgX)R2+HX (38) Scheme 14 extremely versatile.The P-mercurioalkyl peroxides (38)are obtained in high yield and can often be demercurated with sodium borohydride (Scheme 15)or halogens R1CH(OOR)CH(HgX)R2+NaBH,/OH R'CH(OOR)CH2R2 (39) Scheme 15 (Scheme 16) without substantial cleavage of the 0-0 bond; both peroxymer- curation and demercurations occur rapidly under mild conditions. R'CH(OOR)CH(HgX)R2+ R'CH(OOR)CH(Hal)R2 Halz --XHgHal (40) Scheme 16 That secondary alkyl compounds can be obtained in much higher yields than those achieved through conventional nucleophilic displacements and that the method is applicable to the synthesis of acylic cyclic and bicyclic peroxides are the facts of prime importance."L. A. Paquette R. V. C. Carr andF. Bellamy J. Amer. Chem. SOC.,1978,100,6764. K. R. Kopecky J. H. van de Sande and C. Mumford Canad J. Chem. 1968,4625. Biological Chemistry Thus secondary alkyl t-butyl peroxides (39)69 and P-halogenoalkyl t-butyl perox- ides (40)70 have been obtained in yields (based on alkene) of 60-75%. Ketone and ester functions can be tolerated and a wide range of new functionally substituted peroxides of types (41)-(44)7’ have also been prepared. R1R2C(OOBu‘)CH2CO2Me RCH(OOBu‘)CH(Hal)COY (41) (42) (Y = Ph OMe) Me2C(OOBu‘)COY HalCH2CMe(00Bu‘)COY (43)(Y = Me OMe) (44) (Y= Me OMe) By using hydrogen peroxide and suitable dienes mercury-free cyclic secondary alkyl peroxides containing 5-and 6-membered rings have been obtained for the first time and in yields of 50-75% (Scheme 17; n = 1or 2,Z = H or Br).72 Scheme 17 Several other examples some with tertiary carbons next to the peroxide linkage have been obtained similarly and the four diastereoisomers of compound (45) have been isolated by h.p.1~~~ (45) Scheme 18 In these diene reactions the initial peroxymercuration generates an unsaturated hydroperoxide which then cyclizes via a second intramolecular addition.The cycloperoxymercuration can of course be carried out on alkenyl hydroperoxides prepared in other ways.Interestingly such reactions have provided synthesis of 4- and 7-membered peroxide rings (Schemes 1974 and 2075),albeit in low yield. Scheme 19 69 (a) D. H. Ballard and A. J. Bloodworth J. Chem. SOC.C. 1971 945; (6)A. J. Bloodworth and G. S. Bylina J.C.S. Perkin I 1972 2433; (c) A.J. Bloodworth and I. M. Griffin J.C.S. Perkin I 1975 195. 70 A. J. Bloodworth and I. M. Griffin J.C.S. Perkin I 1975 695. 71 (a)A. J. Bloodworth and R. J. Bunce J.C.S. PerkinI 1972,2787;(6)A. J. Bloodworth and I. M. Griffin J.C.S. Perkin I 1974,688. ” A. J. Bloodworth and M. E. Loveitt J.C.S.Perkin I 1978 522. 73 A. J. Bloodworth and J. A. Khan unpublished work. 74 W. Adam and K. Sakanishi J. Amer. Chem. Soc.,1978,100 3935. 75 J. R.Nixon. M.A. Cudd and N. A. Porter J. Org. Chem. 1978,43,4048. W. Adam and A. J. Bloodworth HgX NaBH OH-nOOH 0-0 Scheme 20 To date the synthesis of bicyclic peroxides uia peroxymercuration has been restricted to the [3,3,2]- and [5,2,1]-dioxabicyclodecanes obtained from 1,s-cyclo- octadiene (Scheme 21),76 1,4-cyclo-octadiene (Scheme 22),77 and cyclo-octenyl hydroperoxide.60 H,02 *-NaBH,/OH-2HgX2 01 Br2 Z Z=HorBr Scheme 21 -BZ (23 Z Z =H or Br Scheme 22 Each reaction is regiospecific and it is important to note that the products obtained are isomeric with the [4,2,2]-peroxide available via photo-oxygenation of 1,3-cyclo-octadiene (see later). The choice of mercury(I1) salt can be crucial. Thus whereas mercury(I1) nitrate is extremely good for preparing monocyclic peroxides it fails completely with 1,s-cyclo-octadiene.Mercury(I1) acetate can be used but the trifluoroacetate usually gives cleaner reactions. Reductive demercuration is usually accompanied by some epoxidation or deoxymercuration but bromodemercuration is generally very clean. Related to peroxybromination and peroxymercuration is the perhydrolysis of epoxides under acid conditions (Scheme 13; E = OH). The preparation of three 6-hydroalkyl hydroperoxides by this route (Scheme 23) was recently described'* by RLR3 RLwR3 RZ 0 H + H20 R2+H HOO OH Scheme 23 '' W. Adam A. J. Bloodworth H. J. Eggelte and M. E. Loveitt Angew Chem. Internat Edn. 1978,17 209. 77 A. J. Bloodworth and J. A. Khan Tetrahedron Letters 1978 3075.78 V. Subramanyam C. L. Brizuela and A. H. Soloway J.C.S. Chem. Comm. 1976,508. Biological Chemistry 359 authors who were apparently unaware of an earlier of similar uncatalysed reactions which proceed much more slowly. As expected by analogy with peroxymercuration intramolecular variations of the reaction (e.g. Scheme 24) afford good yields of cyclic peroxides.80 Cat. CCl,CO,H Scheme 24 Acylation (see following Section also). The imidazolide technology developed in the 1960's for mild anhydrous acylations appears to be the currer?t method of choice for preparing diacyl peroxides and has been extended to the synthesis of peroxycar- bonates and peroxycarbamates (Scheme 25; Z = Bu'OO RO or RZN).'* Scheme 25 Rather surprisingly the rival carbodi-imide approach of similar vintage has not been used as widely but the cyclization of a-hydroperoxy acids (Scheme 26; R=cyclahe~yl)~~ is a powerful demonstration of its capabilities and it can be expected to gain in popularity.+ RN=C=NR -780c '.FfH -"W+ (RNH),CO HOO 0-0 Scheme 26 Use of Superoxide Ion. The advent of crown ethers has made available solutions of potassium superoxide in organic solvents. A benzene solution of this reagent has been used to provide dialkyl peroxides in 42-77'/0 yield from primary and secon- dary alkyl bromides or sulphonates.82 Both alkylations proceed with inversion of configuration and the reaction is believed to follow the pathway shown in Scheme 27. Oi-+RX + ROO+X-ROO+O; -+ ROO-+02 ROO-+RX -+ ROOR+X-Scheme 27 '' W.Adam and A. Rios Chem. Comm. 1971,822. N. A. Porter M. 0.Funk,D. Gilmore R. Isaac and J. Nixon J. Amer. Chem. Soc. 1976,98,6000. M. J. Bourgeois C. Filliatre R.Lalande B. Maillard and J. J. Villenave Tetrahedron Letters 1978 3355. (a) R.A. Johnson and E. G. Nidy J. Org. Chem. 1975,40 1680; (6)R. A. Johnson E. G. Nidy and M. V. Merritt J. Amer. Chem. SOC..1978 100,7960. W. Adam and A. J. Bloodworth Choice of solvent is crucial for a similar reaction in dimethyl sulphoxide gave mainly alcohols and no peroxides.83 Subsequently it has been that dimethylsulphoxide is oxidized extremely rapidly by ROO-and thus it seems that this process competes effectively with the desired alkylati~n.~~~ However use of dimethylsulphoxide can be tolerated when the second SN2displacement is an intramolecular process (Scheme 28) for a 1,2-dioxacyclopentane was isolated in 35% yield.” Crown Wph MeSO 0,SMe 0-0 Scheme 28 Good yields of diacyl peroxides can be obtained with superoxide in benzene even in the absence of a crown ether (Scheme 29).86 2RCOCI + 2K02 -+ (RC00)z+2KC1+ 02 Scheme 29 This avoids using the anhydrous ether solutions of hydrogen peroxide that are employed in other non-aqueous routes to diacyl peroxides.Singlet Oxygenation.-The area of peroxide synthesis which has received the greatest attention during the 1970’sis unquestionably that of singlet o~ygenation.~~ In its readily accessible ‘Agstate molecular oxygen reacts with a wide range of unsaturated substrates by one or more of the three modes illustrated in Scheme 30.(48) Scheme 30 83 J. S. Filippo C. I. Chern and J. A. Valentine J. Org. Chem.. 1975 40 1678. 84 (a)M. J. Gibian and T. Ungermann J. Org. Chem. 1976,41,2500; (b)However see C. I. Chern R. Di Cosimo R. De Jesus and J. S. Filippo J. Amer. Chem. SOC.,1978,100,7317. 85 E. J. Corey K. C. Nicolaou M. Shibasaki Y. Machida and C. S. Shiner TetrahedronLetters 1975,3 183. 86 R. A. Johnson Tetrahedron Letters 1976 331. ” (a)D. R. Kearns Chem. Rev. 1971,71,395; (b)R. W. Dennyand A. Nickon Org. Reactions 1973,20 133; (c) W. Adam Chem-Zeit 1975,!39,142; (d)‘Singlet Oxygen. Reaction with organic compounds and polymers’ ed. R. Ranby and J. F. Rabek Wiley-Interscience 1978.Biological Chemistry These equations reveal the minimum structural requirements for each process though it should be added that in route a (Scheme 30) one or both of the double bonds can form part of an aromatic system. Structures (46)-(48) thus represent the basic types of peroxide that can be obtained via singlet oxygenation and an extension to type (49) can be achieved by the recently developed technique of di-imide reduction. There are a variety of ways of carrying out singlet oxygenations but for experi- mental convenience the technique of dye-sensitized photo-~xygenation~~’ is usually chosen. Here a solution of the substrate to be peroxidized together with a small quantity to 10-3M) of a coloured sensitizer such as Rose Bengal or tetra- phenylporphine is irradiated with visible light and simultaneously saturated with oxygen.The dye absorbs light to become electronically excited and the excitation is transferred to oxygen to produce the singlet species. Use of sodium lamp largely eliminates the problem of thermal and u.v.-induced decomposition of the peroxidic products without recourse to elaborate filtering devices. An important feature of the reaction is that it often proceeds satisfactorily at temperatures as low as -78°C thereby facilitating the preservation of sensitive products. Labile peroxides have been postulated as intermediates in the photo-oxygenation of a vast number of substrates including biologically important heterocyclic compo~nds,~~*~~ but we shall be concerned only with systems from which peroxides have been isolated and even here we shall of necessity be highly selective in the examples we quote.Where appropriate we shall try to illustrate the discussion with examples that postdate the many excellent review but we shall not be concerned with the mechanistic controversies such as the question of perepoxide intermediates that continue to stimulate much of the current work. 1,2-DioxacycZohex-4-enes (46). Singlet oxygen reacts with many (a)cyclic con- jugated dienes and aromatic substrates by the Diels-Alder mode of addition. Reactions with various substituted butadienes have been known since 1972 and conditions for making the adduct of butadiene itself in 20%yield have recently been reported.89 Vitamin Dz contains an s-cis diene function and affords a 1 1mixture of the expected epimeric peroxides (50) in 35% yield.” Products from endocyclic dienes date from the 1940’s but the bicyclic peroxide (51) is a recent additi~n.~’ (50) (51) Polycyclic aromatics such as rubrene 9,lO-disubstituted anthracenes and activated naphthalenes afford 1,4-adducts that release singlet oxygen on ther- molysis.Recent interest in aromatic substrates has centred on vinylarenes where the ’* T. Matsuura and I. Saito in ‘Photochemistry of Heterocyclic Compounds’ ed. 0.Buchardt Wiley- Interscience 1976,p. 456. 89 T. Kondo M. Matsumoto and M. Tanimoto Tetrahedron Letters 1978 3819. 90 S. Yomada K. Nakayama and H. Takayama Tetrahedron Letters 1978,4895. 91 Y.Kayama M.Oda and Y. Kitahara Chem. Letters 1974,345. W.Adam and A. J. Bloodworth generated peroxide linkage bridges the olefinic &carbon and the ortho-carbon of the aromatic ring. In reactions with stilbenes P-methylstyrenes and &Pdimethyl- styrenes the first-formed product is trapped by further singlet oxygenation (Scheme 31).92 Related monoperoxides from methoxystyrenes where a cis-methoxy group causes a marked acceleration in the rate of addition were isolated as Diels-Alder adducts with l-pheny1-1,3,4-tria~oline-2,5-dione.~~ The aromatic ring can be heterocyclic and in fact addition onto a thiophene ring is preferred to that onto a naphthalene ring in an internal competition (Scheme 32).94 Scheme 32 AZlyt Hydroperoxides (47).Formation of allylic hydroperoxides is usually the preferred mode of reaction for monoalkenes. Synthetic usefulness is reduced because mixtures of hydroperoxides are commonly formed and so there is much interest in regio- and stereo-selective reactions. It has become apparent that there is a very strong preference for abstraction of hydrogens from a group that is cis to a methoxy substituent. Thus (52)affords a mixture containing 72% of the thermo- dynamically less stable hydroperoxide (54) whereas only (53)is formed from the isomer in which methoxy and methyl groups are cis.95 MeO<2 ";o) (D + Me0Hog 8 HOO (52) (53) (54) Scheme 33 92 (a)M. Matsumoto S. Dobashi and K. Kondo Tetrahedron Letters 1977,2329; (b)M. Matsumoto S. Dobashi and K.Kuroda Tetrahedron Letters 1977 3361. 93 D. Lerdal and C. S. Foote Tetrahedron Letters 1978 3227. 94 M. Matsumoto S. Dobashi and K. Kondo Tetrahedron Letters 1975 4471. 95 G. Rousseau P. Le Perchec and J. M. Conia Tetrahedron Letters 1977 2517. Biological Chemistry Where 1,4-addition can compete with the ene reaction only that process involving the group cis to methoxy takes ~lace.~~.~’ This is illustrated in Scheme 34; the alkene with cis phenyl and methoxy groups affords products derived from 1,4-additi0n.~’ Ph Me0 Ph Scheme 34 The earlier discovery that the Me3Si group can take the place of an allylic hydrogen in the ene reaction has been used to advantage in the preparation of a-hydroperoxy acidss2 and (Scheme 35; R = SiMe or Me).Scheme 35 Other products are also formed if R’and R2contain allylic hydrogens or are phenyl groups. 1,Z-Dioxetunes (48). Where a monoalkene lacks allylic hydrogens but carries alkoxy or amino substituents [e.g. (55)] or where an ene reaction would lead to an allylic hydroperoxide with a strained double bond [e.g. (56)] and in a few other cases singlet oxygenation provides a route to 1,2-dio~efanes.~’ EtO/-OEt -+ (55) EtO OEt The formation of a-peroxylactones from ketenes (Scheme 37)98is an important extension of this technique. Acceptable yields could not be obtained by photo-oxygenation and it was neces- sary to use triphenyl phosphite ozonide as the singlet oxygen source. % W. Adam and J. del Fierro J. Org. Chem. 1978,43 1159.97 W. Adam Adv. Heterocyclic Chem. 1977,21,437. N. J. Turro Y. Ito M.-F. Chow W. Adam 0.Rodriguez and F. Yany J. Amer. Chem. Soc. 1977,99 5836. W.Adam and A. J. Bloodworth Scheme 37 1,2-Dioxacyclohexanes (48). The discovery99 that di-imide reduces the double bond of singlet oxygen-diene adducts (46) while leaving the peroxide linkage intact opened up a general route to saturated bicyclic peroxides that incorporate the 1,2-dioxacyclohexane ring. There seems no reason why the technique should not be extended to the monocyclic series. Normally the reduction is carried out in methanol where the di-imide is generated in situ from dipotassium azodicarboxylate and acetic acid. By this method the compounds (57),99(58) epidioxyergosteryl acetate and 1,2,3,4-tetrahydro-l,4- dimethyl-1,4-epidoxynaphthalene,100and (59)76have been obtained.By changing the reduction medium to dichloromethane and using a deficiency of acetic acid the more labile peroxides (60) (61),1°1a (62),1°1b (63),lo1' and 2,3-dioxabi- cyclo[2,2 l]heptane102 (see Section 3) have been prepared. Triplet Oxygenation.-Ground state triplet oxygenation of organic substrates occurs widely but is often an unattractive preparative route to peroxides because complex mixtures of products are obtained. Hydroperoxides result from hydrogen abstrac- tion by intermediate peroxy radicals but if the substrate contains unsaturation in a suitable position cycloaddition can occur. Such a process is believed to take place during the biosynthesis of prostaglandin endoperoxides.By generating specific peroxy radicals from unsaturated alkyl hydroperoxides controlled cycloaddition has been achieved (Scheme 38);" subsequent reduction afforded cyclic peroxides iden- tical with those obtained (Scheme 24) via the corresponding epoxides. 99 D. J. Coughlin and R. G. Salomon J. Amer. Chem. Soc. 1977,99,655. loo W.Adam and H. J. Eggelte Angew. Chem. Internat. Edn. 1977,16,713. lo' (a)W.Adam and I. Erden Angew. Chem. Internat. Edn. 1978,17,210,211; (b)W.Adam and I. Erden .K Org. Chem. 1978,43,2737;(c)W.Adam and H. J. Eggelte Angew. Chem. Internat. Edn. 1978,17 765. lo* W. Adam and H. J. Eggelte J. Ore. Chem. 1977,42 3987. Biological Chemistry Scheme 38 One of the ways in which regiospecificity can be introduced into autoxidation is to use an organometallic derivative or carbanion.Good yields of hydroperoxd escan be obtained in this way provided that reaction conditions are such (e.g. inverse addition) that there is always a high oxygen-to-substrate ratio. This methodology has been exploited recently in the synthesis of a-hydroperoxy ketones,lo3 and acids (Scheme 39).52v104 Scheme 39 Lithium di-isopropylamide (R =NPr'J was used to lithiate aliphatic acids but was unsuitable for arylacetic acids because the corresponding peroxides are decomposed by even traces of di-isopropylamine. However the greater carbon-acidity of aryl- acetic acids permitted the use of butyl-lithium (R =Bu) without complications from attack on the carbonyl group.Finally we must draw attention to some new triplet oxygenations that are catalyzed by Lewis acids and present an alternative to singlet oxygenation for converting some cyclic conjugated dienes into bicyclic peroxides. lo5 The oxygenation of ergosteryl acetate (Scheme 40) at -78°C was used as a model for testing catalyst efficiency. l7 catalyst + 30 -AcO AcO Scheme 40 The compounds BF3 SnCI, SnBr4 SbF5 SnCL WF6 12 and Phb3F4 require simultaneous irradiation to be effective whereas with VOCI, FeCI, MoCI, WC16 and (4-BrCsH4),gBF4 the reaction proceeds satisfactorily in the dark. The mechanism currently involves electron transfer from the alkene to generate intermediate radical cations. On the basis of this hypothesis the reaction lo3 Y. Sawaki and Y.Ogata J. Amer. Chem. SOC.,1975,97,6983. '04 (a)D. A. Konen L. S. Silbert and P. E. Pfeffer J. Org. Chem. 1975,40,3253;(b)W. Adam 0.Cueto and V. Ehrig J. Org. Chem. 1976,41,370;(c) W. Adam and 0.Cueto I. Org. Chem. 1977,42 38. (a)D. H. R.Barton,R. K. Haynes G. Leclerc P. D. Magnus and I. D. Menzies J.C.S. Perkin I 1975 2055; (6) R. K. Haynes Austral. J. Chem. 1978,31 121 131. lo' R.Tang,H. J. Yue J. F. Wolf and F. Mares J. Amer. Chem. SOC.,1978 100 5248. W:Adam and A. J. Bloodworth has been extended to the synthesis of 3,3,6,6-tetra-aryl-1,2-dioxacyclohexanes (Scheme 41) where yields of 80-90°/~ have been achieved."' A related trans- formation has been carried out intramolecularly with 1,l'-bis( 1-phenylviny1)ferro- cene.'" 2 Ar\ ,C=CH + 30 Ar 0-0 Scheme 41 3 Biologically Significant Syntheses The implication of novel organic peroxides in key biological roles demanded the study of simpler analogues to enable the postulated chemistry to be put on a firm experimental basis.From a synthetic viewpoint some simple but unknown peroxides thus assumed the role of biologically significant target molecules. In this section we describe how the new synthetic methodology has been applied to the synthesis of 2,3-dioxabicyclo[2,2 llheptane the peroxidic nucleus of prostaglandin endoperox- ides and to a-peroxylactones models for the chemienergizers of bioluminescence. The successful preparation of these model compounds has been closely followed by the first synthesis of an actual prostaglandin endoperoxide.2,3-DioxabHcyclo[ 2,2,l]heptane and Prostaglandin Endoperoxides.-In 1977 three independent syntheses of 2,3-dioxabicyclo[2,2,l]heptane(64)were reported (Reac- tions i-iii in Scheme 42) and later a fourth route was added. Scheme 42 lo' R. K. Haynes M. K.S.Probert and I. D. Wilmot Austral. J. Chem. 1978,31,1737. M. Hisatome T. Namiki and K.Yamakawa J. Organometallic Chem. 1976 117 C23. Biologica1 Chemistry 367 Salomon and Sa10mon~~ utilized their combination of trifluoromethanesulphonate leaving group and bis(tributy1stannyl)peroxide nucleophile (Reaction i) but even then it was necessary to carry out the reaction in vacua with rapid transfer of the volatile products to a cold trap to avoid decomposition. Purification was effected by t.1.c.on silica gel at -20 "C and the yield was 13%. The product was characterized by 'H n.m.r. spectroscopy and by catalytic hydrogenation to cis-1,3-cyclo-pentanediol. Porter and Gilmore's method6' was to ring-close trans- 3-bromocyclopentyl hydroperoxide with silver (trifluor0)acetate (Reaction ii). Reaction of a bicyclo- pentane with 98% H202and N-bromosuccinimide in ether at -41 "C afforded a 1:1 mixture of cis- and trans-hydroperoxides which was separated by silica chromato- graphy at -10 "C.Stirring the trans-isomer with silver acetate for 30 min. then gave a quantitative (n.m.r.) yield of (64). Reaction with the cis-isomer was much slower and not clean reflecting the preference for an SN2type of transition state for bromide displacement.62 The peroxide was isolated as a white crystalline solid (m.pt.4243.5"C) after purification by bulb-to-bulb distillation low temperature crystal- lization or sublimation and 13C n.m.r. spectroscopic data were provided as addi- tional characterization. An attractive variation of the method uses the more readily available cis-1,3-~yclopentanediol as starting material. Conversion into the cor- responding dibromide followed by treatment with hydrogen peroxide and silver salt afforded (64) in 3040%yield Adam and Eggelte'02 supplied the third route to (64) by applying their low temperature di-imide reduction in dichloromethane to the singlet oxygen adduct of cyclopentadiene (Reaction iii). After purification by chromatography on silica gel at -20 "C,the yield was 30%.Finally Wilson and Geiserlo9 obtained (64) in an unspecified yield (<25%) by benzophenone-s_ensitized photodecomposition of the corresponding diazo compound and trapping of the resultant triplet biradical with oxygen (Reaction iv). Correct experimental conditions are vital. Inparticular it is essential to irradiate only the benzophenone chromophore for direct excitation of the diazo compound produces a singlet biradical that collapses to bicyclopentane. Oxygen pressure and reaction time must also be carefully regulated to obtain optimum yields of peroxide. The singlet oxygenation and reduction route to 2,3-dioxabicyclo[2,2,l]heptane has the advantage of using the cheapest and most readily available starting material and is for this reason probably the most attractive of the four methods.However if the cis- 1,3-diol is to hand then the silver salt route becomes the method of choice. Furthermore a diol-based synthesis obviously provides an excellent model for the transformation of a prostaglandin F into its endoperoxide and this conversion has now been achieved with the synthesis of prostaglandin H2 methyl ester (66).29,110 First Johnson and his c~-workers*~ reported the preparation of SP,llP-dibrorno- 9,11-dideoxy-PGF2 methyl ester(65) and its conversion into (66) by reaction with crown ether-complexed potassium superoxide in DMF. The yield was only 3% with limited prospects for improvement. Then Porter's group' lo achieved the same conversion with a sevenfold increase in yield by using the silver trifluoroacetate and hydrogen peroxide reaction and isolating the prostaglandin endoperoxide by h.p.1.c.log R. M. Wilson and F. Geiser J. Amer. Chem. SOC.,1978 100 2225. N.A. Porter J. D. Byers,R. C. Mebane D. W. Gilmore and J. R. Nixon J. Org. Chem. 1978,43.2088. W.Adam and A. J. Bloodworth H Br e ZJMe Br OH OH (65) (66) a-Peroxylactones Chemienergizers of Bioluminescence.-The difficulty with these target molecules is their extreme thermal instability. In fact to date no natural a-peroxylactone has been isolated. As already pointed these hyperenergetic molecules (ca. 400 kJ mol-' is released during thermal decarboxylation which is sufficient to generate an electronically excited product) are inferred as reaction intermediates in the enzymatic oxygenation of luciferins through oxygen-1 8 labelling experiments.Their biogenesis is outlined in Scheme 43. Thus a-hydroperoxy acids 0 0 Scheme 43 (67) are key synthetic intermediates. These are chemically very labile substances which readily decarboxylate in the presence of acids and bases to give ketones and water. It was therefore not surprising that the classical methods of hydro-peroxylation under acidic or basic conditions failed to produce (67). Mild and neutral conditions were required. R2 R*0-0 OH 2 H Li Scheme 44 Biological Chemistry In Scheme 44 the successful synthetic routes to a-hydroperoxy acids (67) are A ill~strated.~~ key discovery"' was the reaction between ketene bis(tri-methylsily1)acetals and singlet oxygen which affords the stable trimethylsilylperoxy derivative (Reaction i).This unusual silatropic shift introduces a peroxide function alpha to the carbonyl group under completely neutral and mild conditions and at the same time the trimethylsilyl groups protect the sensitive a-hydroperoxy acid (67) during purification. However the a-hydroperoxy acid (67) can be released quan- titatively by desilylation with methanol at low temperature. An alternative strategy utilized the reactive a-lactones as synthons (reaction ii).'2 In their open dipolar form a-lactones add hydrogen peroxide to produce the desired a-hydroperoxy acids (67). In this way the stable di-t-butylacetolactone prepared by ozonization of the respective ketene afforded the corresponding a-hydroperoxy acid (67)in high yield on treatment with hydrogen peroxide at -70 "C.A convenient method for generating a-lactones in situ is via photo-decarboxylation of malonyl peroxides.' l2 The third method (Reaction iii) employed base-catalysed oxygenation of car- boxylic acid.lo4 This drastic but most direct method can be quite effective if strict control of the critical reaction conditions is exercised. Both the oxygenation of the lithium a-lithiocarboxylate (readily available by direct a-lithiation of the carboxylic acid) and the protonation of the oxygen adduct must be performed at -78 "Cand the product must be isolated swiftly. These three preparative methods for a-hydroperoxy acids (67) are complemen- tary and a large variety of derivatives have been synthesized.The singlet oxy- genation route (Reaction i) has the limitation that the ketene acetal must not have allylic hydrogens to avoid competitive prototropic ene-reactions. The a-lactone route (Reaction ii) is limited to disubstituted derivatives since monosubstituted ones are unstable. The base-catalyzed oxygenation (Reaction iii) potentially the most general method cannot be employed for base-sensitive substrates. The dehydration of the a-hydroperoxy acids (67) to the a-peroxylactones (68) presented a formidable synthetic challenge. Most dehydrating agents failed because the reagents are too basic or acidic or too nucleophilic or electrophilic and thus cause decomposition of the a-hydroperoxy acid and/or a-peroxylactone or they do not exhibit sufficient subambient reactivity To date the most effective and convenient reagents for the conversion of (67) into (68) are the carb~di-imides.~~ Typically dichloromethane solutions of the substrate and the reagent are mixed at -78 "C allowed to warm up to -40°C when the urea is precipitated.Filtration affords a solution of the a-peroxylactone (68) which is then isolated and purified. This review was conceived and written during the tenure of a NATO Research Grant the receipt of which is gratefully acknowledged. '" W. Adam and J.-C. Liu J. Amer. Chem. SOC.,1972,94,2894;G. M. Rubottom and M. Lopez Nieves Tetrahedron Letters 1972 2423. 'I2 W. Adam and R.Rucktaschel J. Org. Chem. 1978 43 3886.