Total Syntheses of Prostanoids By P. H. Bentley BEECHAM RESEARCH LABORATORIES, BROCKHAM PARK, BETCHWORTH, SURREY 1 Introduction ‘Prostanoids’ is a recently introduced1 term used to designate the family of natural prostaglandins and prostaglandin-like compounds (cf. ‘steroids’). The prostaglandins have been known for about forty years, but not until 1962 was the structure of prostaglandin El announced by Bergstrom and co-workers.2 Since then at least fourteen more have been isolated from various sources. Few other biologically potent, naturally occurring substances have such diverse bio- logical effects, and it is reasonably certain that some clinical use will be found for at least one member or a synthetic analogue. The need for adequate amounts of the materials has led to several total syntheses of these substances and it is with this aspect that the review is concerned.Literature to April, 1972 is included. For other aspects the reader is referred to a recent review by Horton3 as well as to an earlier chemical re vie^.^ 2 Nomenclature and Structures The structures of the prostaglandins (PG’s) are derivable from prostanoic acid (1). They occur in four series, designated by the letters E, I;, A, and B [partial structures (2)--(5)], depending on the substitution within the ring. (1) I I HO HO E Series F Series A Series 8 Series (2) (3) (41 (5) E. J. Corey, T. Ravindranathan, and S. Terashima, J. Amer. Chem. Soc., 1971, 93, 4326 S. Bergstrom, R. Ryhage, B. Samuelsson, and J.Sjovall, Acta Chem. Scad., 1962, 16, 501 E. W. Horton, ‘Prostaglandins’, Vol. 7 of Monographs on Endocrinology, Springer Berlin, 1972. J. E. Pike, Fortsch. Chem. Ore. Naturstoffe, 1970, 28, 313. TotaI Syntheses of Prostanoidr As well as a 15-hydroxy-group, up to three double bonds may be present in the side-chains. Thus El, E2, and E3 have one, two, and three double bonds respectively in the positions and with the configurations shown in (6)-(8). 0 (6) PGE1, no 5,6 -or 17,18-double bond (7) PGf2,5,6 -cis, no 17,18 -dou b[e bond (8) PGE3,5,6 -cis, 17,18-cis -double bond The 13,14-double bond is trans In the F series the 9-hydroxy-group can be either a,and hence cis to the C-8 side-chain, or p, when it is trans. Only the former occurs naturally.Prosta- glandins with a 19a-hydroxy-group also occur naturally. Structure (6) shows the absolute stereochemistry of PGEl as determined by X-ray analysis. PGEl is therefore (-)-1la,l5(S)-dihydroxy-9-oxo-13-trans-prostenoicacid. The six prostaglandins of the E and Fseries are termed the primary prostaglandins. 3 Chemical Syntheses The salient features of PGEl (6)are the four chiral centres, the /?-ketol function in the five-membered ring, the all-trans arrangement of the three ring substituents, and the allylic alcohol group at C-15. The additional cis-double bonds and the configuration of the 9-hydroxy-group in the Fseries add further problems to the design of a general synthesis. Furthermore, prostaglandins of the E and A series both give rise to B-prostaglandins under basic conditions, whereas under acidic conditions E-prostaglandins dehydrate to A-prostaglandins. Final steps in any synthesis must therefore be conducted under mild conditions. Earlier reductions of the 9-keto function of the E series yielded a separable mixture of the Fa and the biologically less active Fp-prostaglandins. However, more sophisticated reducing agents furnish the desired Fa compounds exclu- sively5so that a synthesis of the E-prostaglandins becomes a more general one.A conversion of a PGAa derivative into a mixture of PGE, and its llp-epimer E. J. Corey and R.K. Varma, J. Amer. Cliem. Soc., 1971, 93, 7319. BentZey by an epoxidation and reduction sequence has also been reported.6 These interconversions are shown in a general form in Scheme 1.0 HO I ___) 42i” QHO +m0 lii\0 0 0“I.;\ ... Ill -A B Reagents: i, LiBHR3 (for Rsee ref. 5); ii, H20 -AcOH or HCI -H20-THF; iii, KOH-MeOH; iv, H202-KOH-MeOH; v, Ct2+ Scheme 1 Commercially it would be desirable for a synthesis to be as adaptable as possible and ideally to provide in high yield all the resolved primary prosta- glandins and desirable analogues from a single intermediate. So far only the latest Harvard synthesis is likely to meet all these requirements. In this review prostanoid syntheses have been divided into those in which the cyclopentane ring is formed from precursors which incorporate at least one of the carbon side-chains and those where both side-chains are elaborated from initially ‘simple’ cyclopentane derivatives.The object of the review is not to be comprehensive but to illustrate the diversity of the schemes and to suggest which are of commercial importance. 4 Prostanoids from Single-chain Precursors It is convenient to subdivide this section according to which carbon-carbon bond of the ring is formed. A. C-8-C-12 Bond Formation.-Historically this section includes the first synthesis of a prostanoid by Samuelsson and Stallberg.’ Base-catalysed cycliza- G. L. Bundy, W. P. Schneider, F. H. Lincoln, and J. E. Pike, J. Amer. Chem. SOC.,1972, 94, 2123. B. Samuelsson and G. Stlllberg, Actu Chem. Scund., 1963,17,810. 31 2 Total Syntheses of Prostanoids tion of (9), followed by ester hydrolysis and decarboxylation, afforded 15-deoxy- 13,14-dihydro-PGB1 (10) and an isomer (1 1) arising from an alternative mode of cyclization.These isomers were critically compared with a degradation product of natural PGE,, the results supporting the earlier proposed structure (6). (9) (10) + Miyano and Darns effected a similar cyclization of (13), available from a condensation of styrylglyoxal and the diacid (12). The product (14) on hydroxyla- tion and cleavage of the side-chain and reduction of the ring double bond afforded the aldehyde (15). This underwent a Wittig condensation with the phosphorane (16) to furnish 15-dehydro-PGE, (17), together with the 118- epimer and small amounts of dehydration products, all separable by chromato- graphy (Scheme 2).Although the selective reduction of the 15-oxo-group in (17) remained to be solved, and recent work@ suggests that this may be possible microbiologically, the small number of steps and good yields make the synthesis otherwise attractive. Syntheses of PGB, are presently of less commercial interest because it has low biological activity and it has not yet been converted into the primary prosta- glandins. A total synthesis was achieved by three groups10-12 utilizing the diketone (19). The latter was prepared either by selective reductionlOJ1 of (18), available from a condensation of 9-oxo-decanoic acid and oxalate, or by cycliza- tion12 of (20) (Scheme 3). Reaction of the corresponding enol ether (21) with the acetylenic Grignard reagent (22), followed by acidic work-up, afforded (23) as a mixture of C-15-epimers.Partial reduction of the latter gave (+)-PGB1 or * M. Miyano and C. R. Dorn, Tetrahedron Letters, 1969, 1615. M. Miyano, C. R. Dorn, F. B. Colton, and W. J. Marsheck, Chem. Comm., 1971,425. lo P. Collins, C. J. Jug, and R. Pappo, Israel J. Chem., 1968, 6, 839. l1 J. Katsube and M. Matsui, Agric. and Biol. Chem. (Japan), 1969, 33, 1078. 1969,17,408.(Japan),Bull.Chem. and Pharm. Ide,J.Y. Yura and 1' Bentley the &-isomer (24), which was isomerized to PGBl by dilute alkali. Use of optically active (22) has also afforded natural PGB1.13 LLPGE, C5H11 HO 0 Reagents; i,PhCH=CHCO-CHO; ii, aq.NaOH; iii, Os04 -NaIOk ;iv, Zn -AcOH; Scheme 2 B.C-ll-C-12 Bond Formation-Two general syntheses by Corey and co-workers were characteristically outstanding in that pure crystalline prostaglandins of the E and Fseries were obtained for the first time and, in the case of PGE1,a resolu-tion step provided both enantiomeric forms. In addition, completely new chemistry of general applicability was devised. The important synthetic concepts of these elegant syntheses were: (i) the use of the 9-nitro-group as a precursor of the 9-keto function, into which it was transformed under mild conditions; (ii) the successful use of the tetrahydropyranyl (THP) protecting group which suggested that the E series might be obtainable from 11,15-bis-THP-prosta-glandin F derivatives, an important concept of the Harvard bicyclic route; and (iii) the use of aldol cyclizations at a late stage of the synthesis, leading directly to the C-11-C-15 ene-diol unit or its equivalent.R. Pappo, P. Collins, and C. J. Jug, Ann. New Yorkhad. Sci., 1971,180,64. 33 Total Syntheses of Prostanoids Reagents: i, Hz-Pd/C-H'; ii,EtCOCl -AIC13; iii, CH2N2 ; iv, Zn -Pb; v, H2 -Lindlar catalyst; vi, aq.NaOH Scheme 3 In the first of these the key intermediate (29) was constructed by a six-stage process commencing with the Diels-Alder addition of (25) and (26), which gave the adduct (27) as the major product. The latter was modified to (28) (Scheme 4), the cyclohexene ring oxidatively cleaved, and the product (29) cyclized using the base (30).Acetylation of the product fwnished (31) along with small amounts of the llb-epimer. Reduction of the ketone function and ketal hydrolysis gave (32), which was dehydrated to (33) under neutral conditions by a new and useful reaction with dicyclohexylcarbodi-imide,catalysed by cupric chloride. On reduction, the enone (33) yielded both C-15-epimeric alcohols, which were further transformed, via (34),to racemic PGE, and its C-15-epimer. After separation PGE, was converted into PGF,,, PGFlP PGA1, and PGBl by reactions shown in Scheme 1. E. J. Corey, N. H. Anderson, R. M. Carlson,J. Paust, E. Vedejs, I. Vlattas, and R. E. K. Winter, J. Amer. Chem. SOC.,1968,90,3245. Bentley HCONH "I ,R vi ,vii C5Hll HCONH HO 0 AcO 0 THPO OTHP (33) x,;i;4'/-PGEl + 150-epimer Reagents: i, A[-Hg (NO2 +NH2); ii, HC02Ac ( NH2 -HCONH); iii, [E: 9 Hg2* (ketal exchange); iv, 0~04;v, Pb(OAc)4; vi, QN (30); vii, Ac20; viii, NaBHb; ix, H'; w x, C6Hl1NCNC6Hl1- CuCI2; xi, ZnBH,; xii, KOH (OAc -OH); xiii, (OH-OTHP); xiv, KOH(CN-CO2H and HCONH-NH,); XY, NBS(NHp NHBr); xvi, OH'(-HBr); xvii, H+( )NH -)O and OTHP+ OH); xviii,chromatography Scheme 4 Total Syntheses of Prostanoids In a second synthesis15 stannic chloride-catalysed cyclization of (39, which reacted as the aldehyde (36), led to the enone (37), essentially free of the llg-epimer (Scheme 5).Other acidic catalysts gave both C-1 l-epimers. Following p2 R IL *C5Hll C5H1 1L L0 HO 0 Reagents: i, SnCl ;ii, ZnBHL;iii, base ; iv, chromatography; v,Al-Hg; vi, HCOzAc, then xiii -xvii of Scheme 4 Scheme 5 reduction of the ketone function in (37) with zinc borohydride, mild base treat- ment of the products placed the nitro-group and C-8 side-chains in the more stable trans-orientation, enabling the C-15-epimers (38) to be separated by chro- matography.The synthesis of PGEl was completed essentially as before, but in addition the amine (39) was resolved and each enantiomer reacted separately. The natural forms of the prostaglandins were thereby obtained for the first time. The later bicyclic route was the result of the need for a general synthesis, particu- larly of the higher prostaglandins (PGE,, PGEs etc.), to which the above routes were not easily adaptable.Several potential PGEl syntheses were described by Morin and co-workers.16 Thus it was envisaged that dialdehyde (40), which had been obtained from l6 E. J. Corey, I. Vlattas, and K. Harding, J. Amer. Chem. Soc., 1969, 91,535. l6 R. B. Morin, D.0.Spry, K. L. Hauser, and R. A. Mueller, Tetrahedron Letters, 1968, 6023. 36 Bentley aromatic precursors, might undergo intramolecular aldol cyclization to furnish (41). In the event no aldols were isolable. Moreover, the alternative mode of cyclization was preferred, so that at best a separable mixture of (42) and (43) was produced. Further reactions of (42) afforded PGB1, but added difficulties were encountered in removing the protecting ketal group, and only low yields were obtained.OR ,,,an nm OGpI" o(y b0 R CHO CHOCHo CHO OH (40)R=(CH2)6CO2Me (41) (421 (43) C.C-10-C-11 Bond Formation.-Strike and Smith1' synthesized a stereoisomeric mixture (47), containing 13,14-dihydro-PGEI, providing the only example in this section. Thus cyclization of the aldehyde (44)furnished the enone (45), which was epoxidized and hydrogenated to (46).Mild acid treatment gave (47).w:;::,e:;:,,-K 0 2 HC5Hll CHO OTHP OTHP OH OR (44) (451 (46)R=THP (47)R=H D. C-9-C-10 Bond Formation.-A route to PGE, methoxime by Finch and FitP will be only briefly discussed. Dieckmann cyclization of the tetraester (48) followed by five more steps afforded the enone (49).Introduction of the 11-hydroxy-group was achieved by allylic bromination to (50), treatment with silver acetate to (51), and methanolysis to (52). Following silylation, which gave (53), hydrogenation afforded the all-cis derivative (54), arising from cis-addition of hydrogen to the face opposite the bulky silyloxy-group. After protecting the 9-keto-group with methoxyamine and subsequent base treatment to place the side-chains in the more stable trans-orientation, eight further steps furnished PGEl methoxime. The protecting group was not successfully removed. E. C-8-C-9 Bond Formation.-No examples in this section have appeared. l7 D. P. Strike and H. Smith, Tetrahedron Letters, 1970, 4393. l* N.Finch and J. J. Fitt, Tetrahedron Letters, 1969, 4639.37 Total Syntheses of Prostanoids ** C02 Me (48) R=(CH2)6C02H X or ester (49) X = H (54) (50) X = Br (51) X = OAc (52) X = OH (53) X = OSiMes 5 Prostanoids from 'Simple' Cyclopentanes A. Monocyclic Cyc1opentanes.-The 1 1 -deoxy-E and -F prostaglandins are currently of interest as possible substrates for microbiological oxidation at C-11 as well as for their intrinsic biological properties. A recent synthesis by Caton Reagents: i, cyclopentanone enamine ; ii, HCI -BuOH; iii, MezC(0H)CN -Na2CO3; iv, Bu\ AIH; v, phosphorane (IS); vi, Cr03-H'; vii, NaBH4 Scheme 6 Bentfey and co-worker~~~ illustrates a relatively simple route to them (Scheme 6). The condensation product of cyclopentanone and the aldehyde (55) was dehydrated by acid to the enone (56; R = CH20H).Reaction of the latter with cyanide ion, provided by acetone cyanhydrin, afforded the nitrile (57; R = CH80H). The alkaline conditions of the reaction favour the formation of the more stable trans-isomer. Di-isobutylaluminium hydride reduced both the keto and nitrile functions of (57), and the product (58), after Wittig condensation, was oxidized at both alcohol groups and finally reduced to give a mixture (60), containing 1 1 -deoxy-PGF,,. The corresponding PGFIBderivative had been prepared earlier by Bagli and BogrPO using (56; R = C02H), which had been synthesized from Z-ethoxy- carbonylcyclopentanone. Thus diacid (61; R = H),obtained by hydrolysis of the corresponding nitrile (57; R = C02H),was selectively esterified to the half-ester (61; R = Me).The elaboration of the C-12 side-chain was then achieved as follows: the acid chloride of (61; R = Me) reacted with hept-l-yne in the presence of aluminium chloride to provide (62) ;methanolic sodium hydroxide afforded the enol ether (63; R = Me), and hydrolysis, the acid (63; R = €3); reduction of the latter by sodium borohydride followed by acid treatment led to the isolation of the enone (64); the configuration of the 9-hydroxy-group was established by other evidence; finally, a second borohydride reduction gave 1 l-deoxy-PGFlp (65) and its 15p-epimer. The enone (56; R = C0,H) was also utilized in two similar syntheses of PGBl by Hardegger and co-workers,21 and by Klok and associates.22 OH t(64)X = CH=CHCO.CSH,l t(65) X = CH=CH$HC,H,I.I I OH M.P. L. Caton, E. C. J. Coffee, and G. L. Watkins, Tetrahedron Letters, 1972, 773. J. F. Bagli and T. Bogri, Tetrahedron Letters, 1967, 5. a1 E. Hardegger, H. P. Schenk, and E. Broger, Helv. Chim.Acta, 1967,50,2501. R. Klok, H. J. J. Pabon, and D. A. Van Dorp, Rec. Truv. chirn., 1968, 87, 813, and for extension to El series: ibid., 1970,89, 1043. Total Syntheses of Prostarioih vii 0 / ? -o-d( viii ~ I 2 OH Q,R 6H dH (67)(73) (74) COCl Reagents : R (69) i, PhCH20CH2CI ;ii, CH;!=C(CI)COCI j iii, NaN3 (COCL-CON3); CI CIiY,heat(CON3-NCO); v, Ht+(>(NCo-xNH2 +)=NH-+O); vi, m -CIC6H4 C03 H; vii, NaOH; viii, K13-NaHC03 Scheme 7 B.Bicyclic Cyc1opentanes.-This section includes three ~yntheses~~-~~ using bicyclic intermediates. The Harvard synthesisz3 has provided intermediates from a3 See references 26-30 inclusive. 24 H. L. Slates, Z. S. Zelawski, D. Taub, and N. L. Wendler,J.C.S. Chem. Comm., 1972,304. 25 U.Axen, J. L. Thompson, and J. E. Pike, Chem. Comw., 1970, 602, and earlier references cited. Bentley which all the primary prostaglandins in resolved form have been obtained. Prostaglandin El is produced in twenty-one steps from cyclopentadiene, and yields are now such that the prostaglandins could soon be available on a sub- stantial scale. Of the other two schemes, the one by Slates and co-worke~s~~ affords natural PGEl in twenty-nine steps from penta-lY3-diene, but yields, although probably not yet optimum, are less good.The Upjohn synthesis,25 which has been discussed in the review by Pike,* has also led to most of the primary prostaglandins and their isomers. It has the advantage of being short, PGEl being produced in thirteen steps from norbornadiene. However, it suffers from poor yields at several stages. Only the Harvard scheme will be discussed in detail. One important intermediate in Corey’s route is the bicyclic iodolactone (74), the preferred synthesis26 of which is shown in Scheme 7. The average yield per step is 85 %. Because the alkylcyclopentadiene (66) is prone to isomerization to (67), it was found essential to carry out the alkylation and Diels-Alder reactions well below 0°C and with as short an isolation time as possible.For large-scale operations the route shown had definite advantages over an earlier process.27 The reaction of (66) with the dienophile led to (68), where the benzyl- oxymethyl substituent is exclusively anti to the carbon bearing the chloro and chloroformyl groups. In particular, no syn or 2 + 2 addition products [e.g.(69)] were formed. Replacement of the two groups by oxygen was achieved as shown without isolation of intermediates. The oxidation of the product (70) afforded (71) exclusively, no epoxide or isomeric lactone being produced. The hydrolysis product (72) was purified as an ammonium salt and if required. Thus the (+)-salt formed with (+)-amphetamine provided the natural prostaglandins, whereas the enantiomeric series would be obtained from the corresponding (-)-salt.It is apparent from the numbering in (72) that the carboxymethyl group ultimately provides the C-8-substituent and the benzyloxymethyl group the C-12-substituent. Reaction of (72) with potassium tri-iodide in aqueous bi- carbonate introduced the 9a-hydroxy-group at (74) via the intermediate (73). The iodolactone therefore contained the desired stereochemistry at all the nuclear centres. Its conversion into the various prostaglandins will now be described. PGF2,.28 Acylation of (74) and deiodination provided (73, which was trans- formed into (78) by the sequence: catalytic hydrogenation to (76), oxidation with chromic oxide-pyridine complex to the aldehyde (77), and Wittig condensation with phosphonate (79) (Scheme 8).In this sequence the p-phenylbenzoyl (PB) group was chosen over simple acetyl since it allowed for easier crystallization and chromatographic separation of the subsequent reduction product (80) from its 15p-epimer. This reduction of (78) was extensively studied. Zinc borohydride produced a 1 : 1 mixture of the epimers. Other reducing agents yielded lesser z6 See ref. 1 and ref. 28. E. J. Corey, U. Koelliker, and J. Neuffer, J. Amer. Chem. SOC.,1971, 93, 1489, and cited references to earlier work. E. J. Corey, S. M. Albonico, U. Koelliker, T. K. Schaaf, and R. K. Varma, J. Amer. Chem. Suc., 1971, 93, 1491. 41 Total Syntheses of Prostanoids C5H11 PBO I I I I PBO HC) OTHP ATHP (75)R= CHzOCH2Ph (80) (76) R= CH,OH (77)R = CHO HO J Me I C-Pri Reagents: i, p-C&C6H&OCI; ii, 8u: SnH; iii, Hz-Pd/C-H+; iv, CrO3-py ;v, (79); vi, (81)-Bu'Li -OP( NMQ~)~- vii,K2C03; viii, 0; ix, Bu~AIH;x, (83);xi,Hf THF; Scheme 8 Bentley amounts of (80) and products arising from a 1,4-reduction of the enone system.However, of several optically active borohydride ions (Li+BR,H), the reagent derived from the borane (81) and t-butyl-lithium in hexamethylphosphoramide-tetrahydrofuran at -120"C,was shown to reduce (78) to (80) and the 15p-epimer in the ratio 4.5 : 1. Commercially it may be more convenient to effect this reaction with less sophisticated reagents since, following separation, the unwanted epimer can be reoxidized to (78) for recycling.An alternative approach to the stereoselective introduction of the 15-hydroxy-group is described later. The ester group of (80) was next hydrolysed and the two hydroxy-groups of the product reprotected as THP-ethers. Reduction to the lactol(82) and Wittig condensation with phosphorane (83) provided (84) with the desired 5,6-cis-double bond. Mild hydrolysis furnished PGF,,. PGF,,, PGE2, PGE,. Scheme 9 shows how intermediate (84) was further HO .OHI OTHP OTHP OTHP OTHP (84) .. ...i, iii Ii II, Ill iPGEZ PGf1 Reagents: i, H,-Pd /C;ii, CrO,; iii, Ht Scheme 9 tran~formed.~~The selective reduction of the 5,6-double bond depended on the screening of the 13,14double bond by the two tetrahydropyranyloxy-groups.More recent work6 has shown that the dimethylisopropylsilyloxy-groupis even more effective in this respect. PGF,,, PGE,. The desired C-12side-chainof these prostaglandins was introduced stereo~pecifically~~by condensation of optically active aldehyde (85) with the ylide derived from the phosphonium salt (86) (Scheme 10). The latter had been E. J. Corey, R. Noyori, and T. K. Schaaf, J. Amer. Chem. SOC.,1970,92,2586. E. J. Corey, H. Shirahama, H. Yamamoto, S. Terashima, A. Venkateswarlu, and T. K. Schaaf, J. Amer. Chem. SOC.,1971,93, 1490. 43 Total Syntheses of Prostanoids 04 04(75) i-iv @ V @ av-THP~aCHO THPO 6H (85) (87) OH /ii \ A H ____) PGF3aIX THPO OTHP (88) VIII,IX PGf31".* I' P h 3 b- w I OH Reagents : i, K2CO3 ;ii, I ;iii, H2-Pd /C -H';0 (86) iv, CrO3-py ; v, (86) -MeLi ;vi, Bu~ALH;vii,(83); viii, Cr03;ix, Ht Scheme 10 prepared from S-( -)-malic acid. The product (87) was transformed as before to a bis-THP derivative (88) and thence to natural PGF,, and PGE,. PGA2. A novel elimination reaction of the iodolactone (74) provided the basis of a high-yielding synthesis31 of the PGA intermediate (94) (Scheme 11). Al-though the A series is available from the E series (Scheme l), this attractive synthesis is of considerable interest. The elimination to (89) was followed by a one-step ether-cleavage and esterification to (go), ester and lactone reduction to (91), methyl acetal formation to (92), and transformation of the latter to (93) and (94) as before.Further reactions of (94) were not described, but a conversion into (94a) would be envisaged, followed by oxidation and hydrolysis to PGA2. a1 E. J. Corey and P. A. Grieco, Tetrahedron Letters, 1972, 107. BentZey OXMQ i-iv (74) -pi -QCHO (89)%*ORR= CHZPh, X= 0 (93) (90) R=Ac,X= 0 (91) R=H, X= OH (92) R=H, X=OMe OH 0 OTHP (94) (94a) Reagents : i, MeS02CI ;ii,Ac20-BF3; iii, BuhAIH; iv, BF3-MeOH; vJr03-p~; vi, (79); vii, ZnBHi; viii, 0;ix,(83); x,Cr03;xi,H+ 0 Scheme 11 Other Prostanoids. Reference has already been made to the importance of the 11-deoxy-prostaglandins. An efficient route to them was developed by Crabb6 and G~zmin~~ using racemic (89).Direct reduction of the double bond in (89) with Raney Nickel produced (99, through simultaneous hydrogenolysis of the aIlylic oxygen33 (Scheme 12). However, the corresponding hydroxy-acid (961, after reduction and re-lactonization, furnished the desired lactone (97). The P. Crabbe and A. Guzmin, Tetrahedron Letters, 1972, 1IS. 33 (95) was subsequently transformed to a 9,ll-bisdeoxyprostaglandin(ref. 34). s4 P. CrabbC, A. Cervantes, and A. Guzmhn, Tetrahedron Letters, 1972, 1123. Total Syntheses of Prostanoids 9",CHZCO,H ,,CH,CO,H qCHpOCH.Ph CH2OCH2Ph CH2OCH2Ph (95) (96) (97) (100) x= CI (98) (101) X= H 04 xii -or 11 -deoxy -PGF2a or Q+/C*Hl1 xii,xiii,xv,xiv) 11 -deoxy-PGE2 I OTHP (99) Reagents: i, Raney Ni; ii, H'; iii, H2-Pd/C-Ht;*iv, Cr03-py; v,C12CHCOCl -Et3N; vi, Zn -AcOH; vii, H202-OH7 viii, Tl (NO&-H'; ix, (79); x, ZnBHA ;xi, xii, Bu~ALH ;xiii, (83); xiv, H'; xv, CrO3 Scheme 12 synthesis was completed along now familiar lines as shown, e.g.(97) -+ (98) -+ (9%-11-deoxy-PGE, or -POF,,. However, this route had probably already been superseded by a novel and Bentley -b0OAcOAc .. ... I 11,111 + &ir fH OAc &/ (107) R= O-OH(103) R = 0 (106) (104) R= a -OH (108) R = 0 (105) R = a -OAc OAc ?Ac v-viii .KC02Me I:H OAc 6Ac (109) (110) R= CHq (111) R= CHqCHCOC5H11 (112) R = CH-CHCHC5H11 3 OH C5Hll xi-xiii,xv,xit OH OH (114) R= a -OH (11 5) R = 0 Reagents: i, On(105): m-Cl*C6H4C03H; ii, HBr; iii, oxidize; iv,CaCO3; v, Os04;vi, Pb(0Ac)A; vii, (79); viii,.ZnBH~,; ix, KOH;x, HCl -MeOH; xi, 0;xii, BuiAlH; xiii, (83);xiv, H+; xv, CrO3 0 Scheme 13 Total Syntheses of Prostanoids simple synthesis by the Harvard who obtained (98) in four steps from cyclohexadiene with an average yield of > 85 %.Thus the adduct (100) from the diene and dichlorketen was dechlorinated to (101) (Scheme 12). Careful oxida- tion provided the desired lactone (1 02), which underwent oxidative ring- contraction with thallium trinitrate to furnish (98). The later stages of the Harvard syntheses of PGF,, and PGEe were also used by Cro~sley~~ to prepare the cyclohexane analogues (1 14) and (115) (Scheme 13).Thus the dione (103), available from a Diels-Alder addition of p-benzoquinone and butadiene, was reduced with sodium borohydride, and the isolated cis-diol (104) was acetylated to (105) and epoxidized. After removing the a-epoxide, the /%isomer (106) was treated with hydrogen bromide to give the bromohydrin (107), oxidized to the bromoketone (108), and dehydrobrominated to (109). Hydroxylation and cleavage of the double bond furnished the aldehyde (110), which underwent Wittig condensation to the enone (1 11). The latter was reduced to the diols (1 12), the three ester groups hydrolysed, and the product lactonized to (113). The remaining steps closely followed the Harvard syntheses and furnished the analogues (1 14) and (115) and their C-15-e~imers.~' An unexpected source of the prostaglandins, which might compete with the syntheses discussed, is the coral Plexaura homomalla, which has been shown to contain PGA2 and its C-15-e~imer;~~ these have been converted into PGE2 and PGF2,.6 Note added in proof: The rapid developments in prostanoid chemistry since April 1972 have seen the following innovations. 1.Conversion of PGF's into PGE's (E. W. Yankee, C. H. Lin, and J. Fried, J.C.S. Chem. Comm., 1972, 1120). 2. Further stereoselective syntheses based on bicyclic intermediates (G. Jones, K. A. Raphael, and S. Wright, J.C.S.Chem. Comm., 1972, 609; and D. Brewster, M. Myers, J. Ormerod, M. E. Spinner, S. Turner, and A. C. B. Smith, ibid, p. 1235). 3. Synthesis from (82), (74), and (96), respectively, of novel prostanoids con- taining: (i) a propadiene group at C-4,5 (P.CrabbC and H. Carpio, J.C.S. Chem. Comm., 1972, 904). (ii) a 10a-hydroxy-group (P.CrabbC, A. Guman, and E. Velarde, J.C.S. Chem. Comm.,1972, 1126). (iii) photoadducts at C-10,ll (P. CrabbC, G. A. Garcia, and C. Rihs, Tetrahedron Letters, 1972, 295 1). The author thanks colleagues at the Beecham and Wellcome Research Laboratories for helpful discussions. 36 E. J. Corey and T. Ravindranathan, Tetrahedron Letters, 1971, 4753. 36 N. S. Crossley, Tetrahedron Letters, 1971, 3327. ~3' Other true analogues include the 7-oxo-PG's. For these and others see: J. Fried, C. Lin, M. Mehra, K. Kao, and P. Dalven, Ann. New York Acad. Sci., 1971,180,38. s8 W.P. Schneider, R. D. Hamilton, and L. E. Rhuland, J. Amer. Chem. SOC.,1972,94,2122.