Recent progress in the synthesis of taxanes A.N. BOA," P.R. JENKINS," and N.J. LAWRENCEb "Department of Chemistry, The University, Leicester LEI 7RH Department of Chemistry, UMIST, PO Box 88, Manchester M60 1 QD Reviewing the literature published between January 1991 and July 1993. Reference to earlier synthetic work is included where this provides additional perspective 1 2 2.1 2.2 2.3 2.4 3 4 4.1 4.2 4.3 4.4 4.5 4.6 5 Introduction Approaches to the total synthesis of taxanes From A-ring precursors From c-ring precursors From A-ring and c-ring precursors Syntheses starting from the Wieland-Miescher ketone Semi-syntheses of taxanes Syntheses of the C-13 side chain of taxol Phenylglycidate synthon method The Staudinger synthesis of p-lactams Lithiobenzylamine synthon method Enzymic syntheses Aldol reaction approaches A chiral pool approach References 1 Introduction The highly complex tetracyclic diterpene taxol 1, first described by Wall and co-workers in 1971,' is proving to be of great potential in the successful treatment of many types of cancer.* Taxol's unique antimitotic action3 and remarkable efficacy as an anti-cancer drug has stimulated great biochemical attention.BzNH 0 OH BzO 1 Nevertheless the true potential of taxol will only be realized when it is more readily available. The problems associated with its isolation from the bark of the Pacific yew tree Taxus brevifolia, have been reported at length." Total5 and semi-synthesis4 are just two of the many proposed solutions7 to increase the supply of taxol without endangering the yew tree.The former approach has proved extremely arduous and, to date: no successful total synthesis of taxol has been published. Nevertheless as a challenging target, taxol has stimulated many elegant synthetic approaches, including the development of new methods that, in addition, have led to the synthesis of many analogues. We have divided this review into two sections, namely (if approaches to the total synthesis of taxanes, and (ii) semi-synthesis of taxanes. We have further divided the first section, somewhat loosely, into three parts. The first part (2.1) describes linear approaches that sequentially build the taxane ring from an A-ring precursor (so-called left to right approach) whereas the second part (2.2) details approaches that construct the taxane ring from a c-ring precursor (the right to left approach).The third part (2.3) of the first section includes approaches to taxanes that construct the B-ring, as the final step. from precursors that contain both the A and c-rings. These approaches are summarized diagrammatically in Figure 1. ~ + A B C Figure 1 Disconnective approaches towards the synthesis of taxanes. 2 Approaches to the total synthesis of taxanes 2.1 From A-ring precursors Pattenden Pattenden and HitchcockY have synthesized a compound with the tricyclic ring system common to Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 47+OAC 79% (1) + -=YH 0 0 k H 2 OH 4 a R = - b R = H 5 6 3a R = Br 1 (v9 bR=l- c R = H Reagents: (i) BF,.OEt,, - 78°C; (ii) 3 eq. CH,CHMgBr; (iii) cat.tetra (n-propy1)ammonium perruthenate, N-methylmorpholine oxide; (iv) 25eq (€)-Bu,SnCH=CH(CH,),Br, Bu"Li, - 75°C; (v) BaMnO,; (vi) Nal, MeCOEt; (vii) Bu,SnH, cat. AIBN, PhH, A Scheme 1 molecules of the taxane group using a powerful tandem radical macrocyclization-transannulation sequence (Scheme 1 ). In this sequence, the functionalized A-hg 2 containing the unsaturation and methyl substitution of the taxane skeleton was obtained from the Diels-Alder reaction between 2,4-dimethyl- 3- (ace t oxy me t hy 1 )-pent a- 2,4-diene and ac r olein, and was then modified to give the radical precursor 3b as depicted. Upon treatment with tributyltin hydride and a,a'-azoisobutyronitrile (AIBN) the iodo trienedione 3b gave the two separable C-1 epimers 5 and 6 of the taxane ring system in a 3 : 1 ratio (25% yield) along with the reduced product 3c (30%).A second product of reduction 4 b (20%) was isolated, resulting from quenching of the intermediate radical 4a produced after the initial and impressive 12-endo radical macrocyclization step. This ring closure fixes the eventual C-1 ratio of epimers 5 and 6 and is most likely controlled by the conformation of the trienedione 3b before cyclization. The 6-ex0 trig (transannular) cyclization of 4a to 5 and 6 led to the desired trans fused BC ring junction, in accord with the predictions made using the Beckwith transition state model. Oishi and Ohtsuka ABC ring structure 12. However, as it stood this approach seemed limited since synthesis of the mesylate 9 was rather lengthy (7-9,23 steps and 6% overall yield), and the Michael addition of nitromethane anion used in the synthesis of 8 was poor yielding.The chances of overall success in this strategy have recently been greatly increased, however, by a much improved, shorter synthesis (Scheme 3).12 Thus, the monobenzylated 1,5-pentanedio113 was converted in five steps into the diene 14, which reacted in a highly stereoselective Diels-Alder reaction with maleic anhydride to give 15 as a single isomer. Direct reduction of 15 with sodium borohydride under various conditions led to mixtures of the isomeric lactones 17 and 18, with the undesired isomer 18 predominating in most cases. Exclusive conversion of 15 to 17 could be achieved via the iodoacid 16, using a hydrolysis-iodolactonization-reduction sequence.Alkylation of the lactone 17 with LDA and methyl iodide next gave 19 in good yield. The lactone 19 was then converted into mesylate-acid 20 in which the two pendant groups in the A-ring have the cis relationship necessary for subsequent macrocyclization. The acid 20 was converted into the AB structure 1 l b following much the same sequence summarized in Scheme 3. Alkylation of the acid 20 was not effected immediately, that is to give 9, but was left until after Oishi and Ohtsuka have developed methodology for the formation of the AB ring system in the taxanes using a strategy based on transannular acylation of sulfone stabilized anion intermediates. Previous reports'O from formation of the marocyclic lacte-sulfide ring. This methylation was stereospecific, leading to a single isomer of 10 with undetermined relative configuration.their research group had shown that the mesylate 9 can be made from a-ionone 7 as briefly depicted in Scheme 2. Macrocyclization of 9 with 0-( methy1amino)thiophenol next gave the 12-membered lactam sulfide 10, which upon oxidation to the corresponding sulfone and treatment with lithium diisopropylamide (LDA) gave the ring- contracted bicyclic structure 1 la. Reductive cleavage of the sulfone group in 1 l a then gave the AB ring system 1 l b which has been converted" to the tricyclic Fetizon Fetizon has investigated several strategies for the synthesis of taxanes. The first strategy13 involved coupling of A and c ring fragments, but the subsequent attempted closure to form the B ring was unsuccessful.This work has been reviewed el~ewhere.~ In their most recent reportI4 Fetizon and co-workers have shown that the a-fenchol derived enols 2 1 and 22 undergo photocycloaddition with vinyl acetate to give the 48 Contemporary Organic Synthesk0 <-*- 7 30% --- 21 % 8 v*H OMS 9 H 0 12 Reagents: (i) (COCI),, PhH; (ii) (a) 2'-cyanoethyi(2-methylamino)phenyl sulfide; (b) K,CO,, NaBH,, DMF; (iii) NalO,; (iv) LDA, THF; (v) Na-Hg, Na,HPO, Scheme 2 HO-OBn 13 14 I 15 16 17 18 Reagents: (i) PCC, DCM; (ii) EtO,CC(=PPh,)Me, PhMe; (iii) LiAIH,, Et,O; (iv) PCC, DCM; (v) Ph,PMel, BuLi, THF; (vi) maleic anhydride, PhMe, A; (vii) 0.5M NaHCO, then I,, KI; (viii) (a) THF-B(OMe),, BH,.Me,S, (b) Zn-AcOH; (ix) LDA, Mel; (x) (a) DIBAL-H, PhMe, (b) NH,NH,, NaOH, diethyiene glycol, (c) Ac,O, pyridine; (xi) (a) H, Raney-Ni, EtOH, (b) 3,4-dihydro-2H-pyran, H +, (c) LiAIH,, (d) MsCI, (e) Jones' oxidation Scheme 3 cyclobutanes 23a,b and 24a,b as a mixture of separable isomers (Scheme 4).The boron trifluoride etherate mediated retroaldol reactions of either 23a or 24a then gave the bicyclic diketone 25a; likewise 24b led to the epimeric diketone 25b. The 0-methoxy isomer 23b on the other hand failed to undergo a retroaldol reaction and was recovered unchanged from the reaction mixture. The products 25 represent contracted AB ring systems in which the A ring is lacking a methylene group. The products 25a,b have been modified to give the new diketone 26, and the authors now hope to eventually attach the taxane c ring to 26 via an annulation procedure.Fetizon and his co-workers describe the photochemical cycloaddition route with more complex vinyl acetate derivatives. Thus, the known vinyl acetate 28 was first reacted photochemically with the enol In a subsequent reportt5 of an AB ring synthesis, Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 49RO 21 R=Me 22R=H 0 H RO '' OAc + a:b 1 :2.6 23a + 23b (R = Me) a:b1:1 24a + 24b (R=H) 23. (85%) 1 24b (80%) (10 ;: (g; 26 25a R = H, R' = OAc 25b R = OAC, W = H Reagents: (i) CH,=CHOAc, hv, DCM (for 21) or MeOH (for 22) (ii) BF,.0Et2, DCM, 0°C; (iii) (CH,OH),, PPTS, PhH, A, 48 h; (iv) NaOMe, MeOH, O"C, 2 h; (v) PDC, DCM, 24 h Scheme 4 tautomer of dimedone 27, leading to cyclobutanol29 (Scheme 5), which was not isolated but instead underwent a spontaneous retroaldol reaction to give the diketone 30.The reaction leading to 30 was found to be regio- and stereo-specific. After various functional group interconversions the triketone 3 1 was next produced. Upon treatment of 3 1 with a range of bases ( e.g. sodium methoxide, potassium-t-butoxide, LDA, and sodium hydride) the dehydrated-cyclized product 32 was then formed. However, treatment of 3 1 with bromomagnesium diisopropylamide (BMDA ) gave the tertiary alcohol 33 in 50% yield. This proved to be very stable under acidic and basic conditions and its structure was determined by X-ray analysis. Fetizon and co-workers have recently reported16 another approach to the taxane AB ring system using a Norrish type I1 photo-fragmentation strategy (Scheme 6).The Diels-Alder cycloaddition of benzoquinone 34 to the diene 35 gave the bicyclic compound 36 as a single regio- and stereo-isomer. This enedione was next reduced to the dione 37, giving a mixture of epimers at C-9. Further reduction of the C-7 ketone in this mixture with lithium t-butoxy aluminium hydride led to the ketoalcohols 38 and 39 which were easily separated by chromatography. The structure of 38 was proven by X-ray crystallography. Treatment of 38 with the non-nucleophilic base sodium hexamethyldisilazide (NaHMDS) produced the hemiacetal4 la, via the lactone 40. Subsequent methylation of 4 l a gave the acetal4 l b in overall 56% yield from 34. Irradiation of 4 1 b resulted in homolysis of the C-4-C- 12 bond, followed by selective hydrogen migration from C-3 to give the aldehyde 42.Molecular models showed that, for steric reasons, the homolysis of the C-1-C-12 bond in 4 l b cannot be followed by a concerted H-migration from either C-1 or C-7. Unfortunately, the reaction was severely hampered by the appearance of [2 + 21 cycloaddition by-products quite soon after the start of photolysis. The reaction was monitored by TLC and stopped when these by- products built up, typically after only 12% conversion (69% yield based on consumed starting material) to 42. The starting material could be recovered easily using chromatrography. The aldehyde function in 42 was reduced to a methyl group, so giving the interesting acetal43. application of the Haller-Bauer reaction (sodium amide in toluene) to fragment the acetal4 1 b resulted in direct formation of the lactam 45.After formation of the expected product amide 44, the strongly basic conditions presumably promoted a ring closure reaction to the acetal functional group in 44, followed by a reduction of the resulting C-5 ketone. No further elaboration of 45 has been reported, but hydrolysis of the lactam, followed by reduction of the carboxylic acid function to a methyl group could give access to some interesting aza-analogues of the taxane AB ring sy s tem . In an earlier publi~ation,'~ Fetizon et al. showed that '0 27 r o Ac 29 32 Reagents: (i) hv, MeOH, 0°C; (ii) (CH,OH),, PPTS, PhH; (iii) MeONa, MeOH; (iv) PDC, DCM; (v) PPTS, H,O-acetone (1 :9); (vi) BrMgNPr:, THF, -78°C Scheme 5 50 Contemporary Organic Synthesis35 34 36 37 38 39 tviil 6996 (12.5%Conversion) I r 42 R = CHOJ 43 R = CH3 41aR = H 56% from 34 bR=MeJ@') 45 40 44 Reagents: (i) PhH, A, 72 h; (ii) Zn, AcOH,))))), ; (iii) lithium t-butoxyaluminium hydride; (iv) chromatography; (v) NaHMDS; (vi) (MeO)-,CH, p-TsOH; (vii) hv, 12254 nm, MeOH, 0°C; (viii) (a) LiAIH,, (b) MsCI, (c) lithium triethylborohydride; (ix) NaNH,, PhMe Scheme 6 Cha Cha has reported'* a synthesis of the taxane AB ring system based on an initial [4 + 31 diene-oxyallyl cation cycloaddition reaction (Scheme 7).Treatment of 3-chloro-2-pyrrolidinocyclohexene and spiro [2.4] hepta-4,Gdiene with AgBF, yielded, after basic hydrolysis to the ketone, the cycloadduct 46. The stereochemical assignment of this compound was based upon the known preference of oxyallyl cations to react in an endo mode.Reduction of the ketone functionality in 46 with lithium aluminium hydride occurred stereospecifically to give the corresponding endo alcohol which was next protected as its triisopropylsilyl (TIPS) ether. The alkene 47 was then treated with dichlorocarbene generated from ethyl trichloroacetate and sodium methoxide. This reaction gave the ring expanded product 48; which was modified, after the introduction of a methyl substituent using an SN2' substitution, to give the ketone 49. Initially, Cha et al. had hoped that this ketone would undergo a Baeyer-Villiger oxidation in order to gain access to the taxane AB skeleton, but unfortunately both of the ketones 49 and 46 proved resistant to this oxidation.This problem was circumvented by using a Beckman reaction. Thus, treatment of the ketone 49 with hydroxylamine hydrochloride led to a 3 : 2 mixture of oximes, which underwent Beckman rearrangement upon treatment with tosyl chloride in pyridine to give the regioisomeric lactams 50a,b (3 : 2 also). After conversion of 50a,b into the imidates 51a,b treatment with trifluoroacetic acid (TFA) and rn-chloroperbenzoic acid (m-CPBA) then gave the isomeric nitro esters 52a,b, but these were isolated in disappointingly low yields ( 18%). Cha et al. hope eventually to extend this approach to give access to the ABC ring system in the taxanes by using the bicyclo oxyallyl cation synthon 53 derived from the optically active Wieland-Miescher ketone. Fallis FallislY has recently reported an intramolecular Diels-Alder synthesis of the taxane ring system using a suitably functionalized A ring compound.The Diels-Alder precursor 58 was constructed from the aldehyde 54 as shown in Scheme 8. Addition of the diene fragment to the aldehyde 54 was achieved using l-lithio-1,3-butadiene, and led to the diene 55 as the major isomer in 74% yield after hydroxyl group protection. The relative stereochemistry in 55 was determined from an X-ray crystallographic structure determination of the derivative 56. The diene 55 was then taken through to the Diels-Alder precursor 58 via compound 57. It is interesting to note that oxidation of the acetylenic alcohol function in the desilylated derivative of 57 only gave moderate yields in a generally efficient synthesis, and attempts to improve this step by variation of the oxidant were unsuccessful.Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 518 + 0 46 CI * N A0 i-l 5Oa 47 CI + 0 50b v 48 CI + 49 O H i A2 52a R' = C02Me, R2 = NO2 b R' = N02, R2 = C02Me 51 a I 51b OMe Reagents: (i) AgBF,, DCM; (ii) NaOH, MeOH, A; (iii) LiAIH,, Et,O, 0°C; (iv) TIPSOTf, DCM, 2,6-lutidine; (v) CCI,CO,Et, NaOMe, 0°C; (vi) Me,CuLi, Et,O; (vii) TBAF, THF, 0°C; (viii) PDC, DCM, 0°C; (ix) CIH.H,NOH, MeOH, py, 80°C; (x) TsCI, py, r.t.-80°C; (xi) Me,O.BF,, DCM; (xi) mCPBA, TFA, DCM Scheme 7 H 53 (+)-Wieland Miescher Ketone Microwave assisted Diels-Alder cyclization of 58 gave the tricyclic taxane ring structure, and the major adduct 59 arose from the endo transition state 60; here the non-bonded interactions are minimized due to alignment of the dienophile on the opposite face of the diene to the 0-methoxymethyl substituent.An attempted Lewis acid mediated cyclization of 58 proved unsuccessful due to the migration of the cyclohexene double bond into conjugation with the acetylenic ketone. The authors hope that a C- 1 3 carbonyl function, as found in natural taxanes, will suppress this tendency and lead to better yields in the Diels-Alder cycloaddition. Indeed the low yields of 55 57 58 (vii) 3540% I 59 Reagents: (i) (€)-Bu,SnCH==CH-CH-CH,, Bu"Li, THF, - 78°C; (ii) MOM-CI, DIPEA, DCM; (iii) DIBAL-H; (iv) HC = CTMS, Bu"Li, - 78°C; (v) Scheme 8 KOH, MeOH, DCM; (vi) Dess-Martin oxidation; (vii) 005 M in PhMe, microwave, 1 mol% hydroquinone 52 Contemporary Organic Synthesisthe thermal Diels-Alder reaction were also due to this double bond migration and large amounts of uncyclized products were recovered.for introducing the C- 13 ketone by allylic oxidation, and the authors also report that the C-1 hydroxyl function can be introduced via the corresponding C-2 enolate, in accordance with the results of other researchers. Previous studies provided Fallis et al. with a method Wmg Wang's approach20 to the taxanes also involves the construction of the c-ring by a Diels-Alder reaction, but in an intermolecular sense. The A-ring is derived from the mono protected ketone 6 1 (Scheme 9) which was first subjected to the Shapiro reaction followed by trapping with dimethylformamide and hydrolysis to produce the ketoaldehyde 62.Acetal formation and subsequent addition of a substituted diene fragment next gave the alcohol 63 which was then subjected to a protection, functional group interconversion sequence to give the aldehyde 64. Zinc mediated intramolecular cyclization with 64 next provided the AB-rhg fragment 65 which was finally converted into the taxoid 66 by an intermolecular Diels-Alder reaction with dimethyl acetylenedicarboxylate. Sakan Sakan's approach to the taxanes is similar to that described by Fallis, where a functionalized A-ring is cyclized to create the B and c rings together. In 1983 Sakan and Craven2' reported a synthesis of the diene 67 and showed that the thermal Diels-Alder reaction gave the trans fused ketone 68 (7O%), whereas the Lewis acid catalysed cyclization gave the corresponding cis fused product 69 (85%) (Scheme 10).This observation was unusual, as catalysis of Diels-Alder reactions normally enhances stereoselectivity, but does not reverse it! The outcome is presumably a result of endo attack in the Lewis acid mediated reaction giving 69, and ex0 attack in the thermal reaction giving 68. Subsequent to this initial observation the stereodirecting effects of alkyl substituents on the diene and dienophile components in 67 were investigated on model systems,22 and this strategy has now been extended to the taxane system.23 69 Reagents: (i) PhH, 160°C; (ii) MefilCI, PhH, r.t. Scheme 10 The aldehyde 71 was prepared from the enone 70, as previously reported, and was next converted to the diene-enones 67 and 72-74 by standard methods (Scheme 1 1).These compounds were then cyclized under both Lewis acid and thermal conditions, and the ratio of isomers determined. The results are summarized in Table 1. In the nomenclature used by Sakan et al. the four possible stereoisomers are designated either cis or trans depending on the relative stereochemistry of the B-c ring junction, and a or B depending on the C-1 -C-3 relative configuration. Thus 68 is the a-trans isomer (the relative configuration in naturally occurring taxanes) and 69 is the a-czk isomer. Where possible the relative configurations were elucidated by X-ray crystal structures; otherwise comparison of NMR spectra or H Me0 H (vii)-(ix) MEMO &r 62 63 64 61 BzO (xii) M Ehr 66 65 TBDPS Reagents: (i) p-TsNHNH,; (ii) 4 eq., BuLi; (iii) DMF; (iv) HCI, H,O; (v) p-TsOH, MeOH; (vi) Br x , ZnCu; (vii) MEMCI, PrLNEt; Scheme 9 (viii) Bu,NF; (ix) CBr,, Ph,P then silica gel; (x) Zn-Cu; (xi) PhCOCI, pyridine; (xii) Me0,C-C C - CO,Me, A Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 5338.1% 11 steps I CO2Et 70 H 71 t l$f$ 67 R' = R2 = Me (8.8% from 71) 72 R' = Me, R2 = H (13.5%) 73 R' = H, R2 = Me (12.0%) 74 R' = R2 = H (10.3%) \ I R' Scheme 11 chemical correlation methods were used.The authors found that under thermal cyclization conditions a methyl substituent on the diene ( R2 = Me) increases selectivity for the a- over 8-isomers, and approximately doubles the a-trans: a-cis ratio. A methyl substituent on the dienophile (R2 = Me) decreases the a /8 ratio by 40%, but the selectivity for a-trans over a-ck increases significantly (ca.five-fold). In the Lewis acid catalysed reaction only the a-cis isomer was formed in all cases, with the exception of 73 where a minor product, tentatively assigned as the s-cis isomer, was isolated. The preference of the a stereochemistry at C-3 appears to be a unique feature of this carbon skeleton, and Smith and Houk are currently investigating the molecular mechanics of this system. Table 1 67 thermal 72 thermal 73 thermal 74 thermal catalyzed catalyzed catalyzed catalyzed 70 0 36 0 38 0 17 0 0 85 49 97 27 80 64 97 0 0 15 0 12 0 19 0 0 0 0 0 13.5 1 5(h) 0 0 (")Values are in absolute percentage yield (b)Structural assignment is tentative Blechert Blechert has investigated a photochemical [2 + 21 cycloaddition-retroaldol route to form the eight- membered B ring in the taxanes (Scheme 12).Earlier be built via the key dienone 75 and dione 76 intermediates. The ally1 carbonate derivative 77 underwent a stereospecific [2 + 21 cycloaddition with cyclohexene leading to cyclobutane 78. When the had shown how the tricyclic ABC skeleton could analogue of 7 7 lacking the C- 13 ketal protection was subjected to these reaction conditions Blechert found that the reaction occurred without stereoselectivity at the crucial C-8 centre. Another inconvenience was that using 1-methylcyclohexene, with the aim of incorporating the angular c-ring methyl group, gave the wrong regiosiomer, which would have led to a methyl group at C-3 instead of C-8.Blechert is reportedly investigating an alternative route to by-pass this particular p r ~ b l e m . ~ ~ . ~ ~ After deprotecting the tertiary alcohol and ketone functions in 78, followed by stereoselective reduction of the latter, treatment with potassium t-butoxide gave the retroaldol product (4 80) and the ABC tricyclic taxane skeleton. This sequence also effected complete epimerization at C-3 to give the thermodynamically favoured trans B/C ring junction. Blechert had found that if the C- 13 ketal was left in place, enolization occurred towards C- 1, C-9, and C-1 1 and not C-3. Alternatively, selective reduction of the C-10 carbonyl function allowed epimerization at C- 1 as a consequence of altered conformational preferences.through to the cinnamate ester 80 using the same retroaldol-epimerization sequence. Recently27 Blechert had taken the cyclobutanol79 75 CO2H C02Me (Vii) 85% I - 76 78 79 OH 0 .@ H 0 Ph 81 Ph 80 Reagents: (i) KCN, NH,CI; (ii) HOCH,CH,OH, TsOH; (iii) KOH, H,O,; (iv) DCC; (v) LiMe,Cu; (vi) CH,N,; (vii) KH; (viii) CICO,CH,CH=CH,, NaHCO,; (ix) cyclohexene, hv; (x) Pd(PPh,),, morpholine; (xi) HCI, H,O, THF; (xii) lithium selectride, - 70°C; (xiii) ButOK, Bu'OH; (xiv) cinnamic acid, dicyclohexylcarbodiimide; (xv) NaBH,, citric acid, CH,OH, 3 min.; (xvi) OsO,, N-methylmorpholine-Noxide Scheme 12 54 Contemporary Organic Synthesiscis-Dihydroxylation of the double bond in 80 gave a 1 : 1 mixture of separable diastereoisomers of 8 1. Both isomers were subjected to ,an in vitro tubulin test.The less polar of this pair was shown to inhibit the depolymerization of tubulin. This result is important as it is the first synthetic taxane with action analogous to taxol. More recent work by Blechert26 has focused on the intermediate 78, as a compound to modify in order to introduce oxygen functionalhation at C-9 in the wring. Thus, deprotection and dehydration of 78 (Scheme 13) first gave the dehydro derivative 82. Various functional group interconversions followed by ozonolysis of the double bond then gave the C- 10 epimers 83 and 84, the first taxanes with three oxygen functionalities in the wring. 0 0 70 82 / (iiir/ (vii) I retroaldol methods to form the wring. Kraus, however, has instead employed a fragmentation of the ring system via a bridgehead carbocation.28 The known keto-ester 85 was prepared as a mixture of diastereoisomers, and was allylated to give the derivative 86 (Scheme 14).This step was unexpectedly difficult and low yielding (35%), but the product was isolated as a single isomer with the alkylation assumed to have occurred from the em face. Wacker oxidation and cyclization of 86 next gave the bridgehead alcohol 87 which, after bromination, fragmented in the presence of silver tetrafluoroborate to give the AB model compound 88. 0s 0 H e-co2M* 6 86 (iO,(iiO 55% 1 0 Reagents: (i) K,CO,, MeOH; (ii) H,O+; (iii) LiAIH,, THF -20"-0°C; 84 (iv) (a) MnO,, DCM, (b) LiAIH,; (v) Ac,O, DMAP, Et,N; (vi) 0, DCM, MeOH then DMS; (vii) NaBH,, MeOH; (viii) NaBH,/CeCI,, MeOH Scheme 13 Kraus The approach to the taxane AB ring system adopted by Kraus et al. involves a similar strategy to that of Blechert and by Fetizon.The common link is the [2 + 21 photocyloaddition of a cyclohexane- 1,3-dione enol leading to a [6.4] ring system. The other researchers, as mentioned above, then investigated 0 ij 88 a7 Reagents: (i) 2.2 eq. LDA, ally1 bromide; (ii) PdCI,, 0,; (iii) NaOMe, Scheme 14 0°C; (iv) PBr,; (v) 1.2 eq. AgBF,, 5: 1 MeCN/H,O, 0°C An alternative synthesis was investigated in the light of the poor yields for the conversion of 85 into 86. Thus the enol ether 89 was elaborated as depicted in Scheme 15, eventually yielding the tertiary alcohol 90. The subsequent bromination and fragmentation of this compound has not yet been reported. 89 ,o, 0 H 59 OH 90 Reagents: (i) hv, PhH, 72 h; (ii) LDA, TMS-CI; (iii) Pd(OAc),; (iv) ButOK, Ph,P=CHCOCH,CO,Et, THF, r.t.; (v) 1 1O"C, aq.THF Scheme 15 Boa, Jenkins, and Lawrence: Recent progress in the synthes& of taxanes 55Ghosh In 1990 Ghosh et al. reported29 a Diels-Alder fragmentation sequence in their strategy for making the taxane carbon skeleton. Full details of this work30 have now appeared. The unsaturated anhydride 9 1 first underwent a Diels-Alder reaction with cyclopentadiene leading to an adduct which was then modified to give the diester 92 (Scheme 16). COrrMe o w 0 91 (i)-(iii) 91% 5- C02Me 92 CO&e Cp2Me i/- 97 Reagents: (i) cyciopentadiene, THF, AICi,; (ii) NaHCO, EtOH, H,O, A; (iii) CH,N,, Et,O; (iv) Na, NH,(I), - 55°C; (v) BH, THF, 0°C then NaOH, H,O,; (vi) acetone, Jones reagent; (vii) Et,OBF,, CH,CI,, N,CHCO,Et, 0°C; (viii) cyclopentadiene, PhMe, A Scheme 16 Unfortunately, Ghosh et al.found that the analogous reaction with 5,5-disubstituted cyclopentadienes failed to produce any of the expected adducts, and so prevented direct entry to analogues with the functionalization needed to introduce the C- 15 (taxane numbering) geminal dimethyl group of the taxane skeleton. He is currently addressing this problem in a number of ways.31 Reductive cleavage of the strained tricyclic 1,2-diester 92 with sodium in liquid ammonia led to the ring expanded diester 93. The double bond in 93 was then modified by a hydroboration-oxidation sequence to give the ketone 94, which then underwent a ring expansion when treated with ethyl diazoacetate, giving the AB analogue 95.Following a similar strategy, the aromatic c-ring tricyclic model 97 was made from the anhydride 96; unfortunately the yield of the key C-C bond cleavage was a disappointing 33%. Recently Ghosh has reported two alternative protocols to replace the sodium-liquid ammonia reductive cleavage step 92-93, again making use of the strain in polycylic systems to help the fragmentation. The first3* method is shown in Scheme 17. The diester 92 was first fully reduced and the resulting diol was then protected as the dimesylate 98. Treatment of 98 with zinc and sodium iodide in hexamethylphosphoramide (HMPA) next gave the ring expanded triene 99. Normally this reductive protocol would reduce the rnesyloxy function in 98 to a methyl group, but in the strained polycyclic compound 98 an intermediate carbanion at one of these centres triggered the fragmentation to give 99, in favourable competition with the reduction.In contrast, with the less strained dimesylate 100 the doubly reduced product 101 was isolated from a mixture of products, and no compounds arising from ring cleavage were detected. In a similar fashion the aromatic dimesylate 102 was converted into the diene 103 in 64% yield (qf 33% for the sodium/liquid ammonia mediated fragmentation). Ruthenium tetroxide oxidation of 103 then gave the diketone 104, so showing the synthetic potential of this protocol. CHrrOMs CH20Ms nr R Reagents: (i) LiAIH,, THF, r.t.; (ii) CH,SO,CI, NEt,, DMAP, DCM, 0°C; (iii) Nai, Zn, HMPA, A; (iv) BH,-THF, 0°C then NaOH, H,O,; (v) Jones oxidation, (CH,),CO; (vi) CH,N,, Et,O; (vii) RuCI,.nH,O, CCI,, MeCN, H,O, r.t.Scheme 17 The second alternative cleavage procedure used by Ghosh involved a radical fragmentati~n.~~ Thus, the anhydride 105 was first reduced to the lactone 106 with sodium borohydride (Scheme 18), and 106 was next converted into the chloro ester 107 using thionyl chloride in methanol. Treatment of this chloro ester with tributyltin hydride and catalytic AIBN then initiated a smooth fragmentation to produce the diene 56 Contemporary Organic Synthesis108 in good yield, with only a trace of the directly reduced product detectable in the 'H NMR spectrum. Conversely with the 'strain free' chloro ester 109, the reduced product 110 predominated. Reduction of the benzo analogue 11 1 under these conditions, gave only 25% of the fragmented product 113 and 31% of the reduced product 1 12 even though the tertiary benzylic radical formed after C-C bond cleavage in this case was expected to be more stable than the corresponding debenzo system.Ghosh has speculated that replacement of the hydrogen atoms at C-3 and C-4 with sp2 carbons in the benzo analogue 11 1 reduces the non-bonded interactions with the hydrogen atoms at C- 10 and so decreases the likelihood of strain- assisted fragmentation. Ghosh has successfully applied many useful protocols and with suitably functionalized precursors a range of interesting ABC taxane compounds should be accessible in the near future. (I) 65% HO 11 4 (-)-Bpatchculene oxide 115 117 116 Reagents: (i) BF,.OEt,; (ii) (Pr'O),Ti, Bu'OOH then Me,S Scheme 19 @& 'H 0 4) C02Me CHpCI 118 119 J steps kO2Me 107 @'R "C02Me Reagent: (i) (Pr'O),Ti Scheme 20 CO&le 2.2 From c-ring precursors "Left to Righr c+AK 111 R=CI 112 R = H (31%) 113 Reagents: (i) NaBH,, THF, 0°C; (ii) SOCI,, MeOH, A; (iii) Bu,SnH, Scheme 18 AIBN, PhH, A Shea The Diels-Alder approach to c-aromatic taxoid structures developed by Shea is an example of a Type II Diels-Alder reaction that he and his colleagues had previously developed in a series of elegant studies.38 The key step is the intramolecular Diels-Alder reaction of the triene 12 1 which gives the c-aromatic taxoid 122 under both thermal and Lewis acid conditions.Shea discovered that the product was produced as two atropisomers, endo 123 and ex0 124, and that the ratio depended upon the conditions used.A strong kinetic preference for the endo isomer 123 was observed when the reaction 12 1 -+ 122 was carried out in the presence Of A Q 3 ' The phenomenon of atropisomerism in taxoid structures has been studied in detail by Shea,4O and the results obtained used by all other workers synthesizing c-aromatic taxoid structures. More recent publications from Shea et aZ.4' demonstrate developments in converting c-aromatic compounds into structures with a non-aromatic c-ring Holton No review of taxane systems would be complete without mention of the elegant work of Holton. The key steps in the Holton route were first illustrated by the rearrangement of ( - )-p-patchoulene oxide 114 into the tertiary alcohol 115 (Scheme 19).Subsequent epoxidation and fragmentation of 1 15 via the intermediate 116 then gave the AB-rhg system of the taxane structure 1 1 7.343 35 This result was then extended to produce the functionalized epoxide 118 (Scheme 20).36 Fragmentation of 1 18 next led to the intermediate 1 19 which was then elaborated to the unnatural enantiomer of taxusin 120. At the time of writing this review Holton's approach is the most advanced taxane synthesis. He has recently written an excellent behind-the-scenes account of his 57 Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes121 122 * ex0 124 endo 1 23 (Scheme 2 1). Thus, the dianion of the aromatic acid 125 was alkylated with the chlorodiene 126 to provide the aromatic diene 127 after esterification.Reduction of the ester 127 to the corresponding aldehyde, followed by addition of vinylmagnesium bromide and oxidation next produced the enone 128. A Lewis acid catalysed intramolecular Diels-Alder reaction then gave the c-aromatic taxoid structure 129 as the single endo product 130. The reduction of 130, with DIBAlH, occurred with acceptable stereocontrol to give a 1 : 3.9 mixture of alcohols from which the stereoisomer 131 was isolated in 71% yield. Methylation of the alcohol 13 1 followed by lithium- halogen exchange and reaction with carbon dioxide next provided the acid 132. Reduction of the aromatic c-ring in 132 followed by esterification and hydrogenation then yielded the ester 133. The major step of reduction of the aromatic c-ring has therefore been taken, and clearly further work on this strategy is underway.Jenkins Concurrent with the work of Shea, Jenkins and his group have also investigated the c --* ABC Diels-Alder approach to taxanes. The main difference between the two approaches is that Shea et al. use an aromatic c-ring precursor which is reduced after cyclization while in the Jenkins route the c-ring precursor is alicyclic. The route of Jenkins et al. is illustrated in Scheme 2242 and it starts from the trimethylsilyl enol ether 134, a compound prepared by the Robinson annulation of 2-methylcyclohexanone and methylvinyl ketone. Ozonolysis of 134 and treatment with diazomethane next gave the ester aldehyde 135. Addition of vinylmagnesium bromide to 135 followed by protection of the resulting allylic alcohol then led to the ester 136.Reduction of the ester group in 136 to an aldehyde with DIBAlH followed by addition of trimethylsilylmethylmagnesium chloride next produced the sensitive alcohol 137. Oxidation of 137 to the corresponding ketone, using a very short reaction time to avoid desilylation, followed by addition of vinylmagnesium bromide and Peterson elimination then provided the key triene 138. Diels-Alder reaction was not possible without the presence of an electron-withdrawing group in the dienophile; hence the silyl protecting group in 138 was removed and the resulting alcohol was oxidized to give the enone 139. The intramolecular Diels-Alder reaction with 139 occurred readily with diethylaluminium chloride to produce the tricyclic taxoid structure 140 as a single diastereoisomer. The relative stereochemistry of the three asymmetric centres in the tricycle 140 was shown to be the same as the corresponding centres in the natural taxanes by X-ray crystallography, which also proved that the eight-membered ring was in the boat-chair conformation 14 1.This is the conformation observed in the X-ray crystal structure of a wide range of taxane derivatives. This Diels-Alder route to the taxanes has been adapted to produce an alkylated taxoid structure as shown in Scheme 23.43 Thus, addition of 2-propenylmagnesium bromide to the aldehyde 142 followed by a Collins oxidation first provided the enone 143. The selenoacetal of acetone was next lithiated and the resulting anion ( LiCMe2SePh)44 was then added to the enone 143; subsequent elimination of PhSeOH finally gave the triene 144.Deprotection and oxidation of 144, to produce the enone 145, was + q (igl ~ $.nQ “ib; * 0 / Me02C H02C Br 0 Br Br 127 128 126 125 133 132 endo 131 130 129 Reagents: (i) LDA, -78°C then the chloride; (ii) CH,N,; (iii) DIBAL, C,H,, 0°C; (iv) PCC; (v) CH,CHMgBr; (vi) BaMnO,; (vii) Et,AICI; (viii) DIBAL, CH,CI,, C,H,, - 78°C; (ix), NaH, Me1 then Bu‘Li followed by CO,; (x) Li, NH,, EtOH, THF, - 78°C then CH,N, followed by H,, PtO, Scheme 21 58 Contemporary Organic Synthesis134 135 bTBS 136 OTBS 138 0 139 om 142 0 145 0 om 144 146 Reagents: (i) CH,C(Me) MgBr; (ii) Collins oxidation; (iii) Me,C(SePh)Li; (iv) SOCI,, Et,N; (v) HF, H,O, CH,CN; (vi) BF,.OEt, H3C LJp=(=Jp 0 0 Scheme 23 141 140 Reagents: (i) 0,, Me,S; (ii) CH,N,; (iii) CH,=CHMgBr; (iv) TBDMSOTf, 2,glutidine; (v) DIBAL; (vi) TMSCH,MgCI; (vii) Collins oxidation; (viii) CH,-CHMgBr; (ix) NaOAc, HOAc; (x) HF, H,O, CH,CN; (xi) Et,AICI Scheme 22 148 147 followed by intramolecular Diels-Alder reaction, using BF,.OEt, as a catalyst, to give the alkylated taxoid 146. The product 146 was not crystalline, and so the relative stereochemistry was determined by NOE studies and comparison with the spectra of the unalkylated model compound 14 1. The preference for the formation of the eight-membered ring in 146 in the boat-chair conformation is reflected in the transition state of 147 -* 148 for the Diels-Alder reaction. The products 141 and 148 correspond to the endo isomers observed in the Lewis acid catalysed Diels-Alder cyclization to c-aromatic taxoid structures presented by Shea et al.taxanes by using a chiral pool derived c-ring. Thus, the readily available protected glucose methyl ketone 149 (Scheme 24) was first subjected to a Robinson annulation to produce the tricyclic enone 1 5045-the first reported example of the application of this annulation reaction to the synthesis of annulated sugars. Reduction of the ketone group in 150 with L-Selectride" next provided the allylic alcohol 15 1, the structure of which was determined by X-ray crystallography. The formation of a trans ring junction between the carbocyclic ring and the sugar was achieved using the Stork silylmethylene radical cy~lization~~ as illustrated in 15 1 -, 152 4 1 53.47 The ring junction between the carbocyclic ring and the sugar ring had now been established with the correct absolute configurations.The next task was to cleave Jenkins et al. have extended their approach to the methoxyacetal group in 153 to leave the highly substituted cyclohexane, the future cring.48 The siloxane ring in 153 proved to be unstable to subsequent reactions, and so it was cleaved oxidatively and then protected to yield the bis-silyl ether 154. Reaction between 154 and N-bromosuccinimide caused fragmentation of the benzylidene ring to give the bromoester 1 55.49 A second fragmentation, following the Vasella protocol, was achieved on heating the bromoester 155 with zinc leading to the aldehyde 156. Reduction and protection of the aldehyde 156 next gave the olefin 157 which was treated with ozone to produce the aldehyde 158.The aim of Jenkins et al. is to construct diene and dienophile components onto the aldehyde 158, and then to use the intramolecular Diels-Alder reaction to produce the A and B rings of the taxoid structure. Yadav An interesting variation on the Diels-Alder approach to the taxanes has been published by Yadav et al. (Scheme 25).50 The diol 159 was alkylated selectively with the bromodiene 160 to give the ether 161. Swern oxidation of 16 1, followed by epimerization and addition of vinylmagnesium bromide next gave the alcohol 162. A further Swern oxidation led to the trienone 163, which underwent an intramolecular, Lewis acid catalysed Diels-Alder reaction to produce Boa, Jenkins? and Lawrence: Recent progress in the synthesis of taxanes 59159 Me' h e 153 OTBPS OTBPS 157 158 Reagents: (i) Lithium tetramethylpiperidine, Et,O, O"C, 1 h; (ii) 3-(trimethylsilyl) but-3-en-2-one, - 78°C -, r.t., 1 h; (iii) KOH (03 mot equiv.), MeOH, 80"C, 6 h; (iv) L-Selectride; (v) CISiMe,CH,Br, Et,N; (vi) Bu,SnH, azoisobutyronitrile (AIBN); (vii) H,02, KF; (viii) t-butyldiphenylsilyl chloride, CH,CI,, imidazole, r.t., 72 h; (ix) NBS, BaCO,, CCI,, reflux 3 h; (x) Zn, Pr'OH, reflux, 5 h; (xi) NaBH,, Pr'OH, 60"C, 15 min.; (xii) Et,SiCI, CH,CI,, imidazole, 15 h; (xiii) 0,, CH,CI,, - 78°C then dimethyl sulfide Scheme 24 the tricyclic ether 164.Reduction of the ketone group in 164 and protection of the resulting alcohol then gave the ether 165 which underwent Wittig rearrangement, using BuLi at - 78"C, to produce the 2 ob+e$J the publication m-ring fragment of the 168 conversion by S.F.Martin of the diene et aL5I 167 in into K*- K* / \ tricyclic compound 166. 160 OH OH 159 161 Oxy-Cope routes (i,Mlem The story of the oxy-Cope route to taxanes starts with 1982. 0 OH &OH KH * a0 167 168 Paque tte The oxy-Cope rearrangement is a key step in Paquette's route to the taxanes. Recent progress on this work is illustrated in Scheme 26. The enantiomerically pure ketone 169 is first reacted with the optically enriched cerium reagent 170 to give the alcohol 17 1. [ 3,3] Sigmatropic rearrangement of 17 1 occurred via an endo chair transition state, leading to the 'carbonyl down' atropisomer 172. Deprotection of 172 and oxidation next produced the ketone 173 which was then equilibrated with sodium methoxide to give a 1 : 1 mixture of ketones with the cis and trans ring junctions, 173 and 174 respectively.Separation and recycling the cis ketone 173 gave the trans isomer 174 in 80% yield. Hydroxylation of 174 next provided a 163 162 I; 164 om 1 0 OTBS 166 Reagents: (i) NaH; (ii) (COCI),, Me,SO, Et,N, -78°C; (iii) NaOMe, MeOH; (iv) H,C=CHMgBr, THF; (v) Et,AICI, CH,CI?j. (vi) NaBH,, EtOH; (vii) TBDMS-CI, imidazole, DMF; (viii) BuLi, THF, -78°C Scheme 25 60 Contemporary Organic Synthesis0' 177 Reagents: (i) THF, - 78°C; (ii) KH, 18crown-6; (iii) Bu,NF then PDC; (iv) NaOMe, MeOH, separate and recycle; (v) OsO,, NaHSO,, pyridine, Scheme 26 water; (vi) CH,SO,CI, pyridine; (vii) EtfilCI O H & 0 90% (9 -& TBSOH OTBS 178 179 < \ ) < I I + 186 single diol 175 which was mesylated selectively to yield the secondary mesylate 176.The second step in this route towards the taxanes is the Et,AlCl catalysed 1 ,Zmigration of the C , bridge in the mesylate 176 to produce the functionalized tricyclo [9.3.1.0.378] pentadecane 177.52 The alternative depiction 178 gives a representation that is easier to compare with the other taxanes covered in this review. Further transformations of the B and c rings of the triketone 178 are outlined in Scheme 27.53 The three carbonyl groups in 178 are differentiated by first converting the A and c-ring ketones into silyl enol ethers to produce the bis-silyl ether 179. Steric factors dictate that the c-ring silyl enol ether is more reactive to hydroxylation, which leads to the diol 180.Protection of 180 followed by low temperature reduction with DIBAlH in hexane provides the /?-alcohol 18 1, whereas reduction in benzene at 8°C led to the a-alcohol 182. Dehydration of alcohols 181 and 182 gave the olefins 183 and 184 respectively, which finally produced the respective diacetates 185 and 186. Clearly this approach is very close to synthesizing some taxane natural products. The main problem to be faced is the introduction of the bridgehead double bond into the A-ring. Once this task has been achieved the route has great potential. 2.3 From A-ring and c-ring precursors Kuwajima The key step in the approach to taxanes highhghted by Kuwajima et al. is the formation of the 9- 10 Reagents: (i) TBSOTf, Et,N; (ii) Me,CO,; (iii) MOMCI, PrLEtN; (iv) DIBAL, hexane, - 78°C to - 10°C; (v) DIBAL, benzene, 8°C; (vi) Burgess reagent, benzene, 2545°C; (vii) [C,H,C(CF,),O],SPh,, benzene, 25°C; (viii) OsO,, CH,CI, then NaHSO, pyridine followed by Ac,O, pyridine, DMAP Scheme 27 Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 61carbon-carbon bond by an intramolecular Lewis acid catalysed cyclization of a dienol silyl ether and an a~etal.~, In the unsubstituted case, 187, the cyclized product 190 was obtained in 74% yield.Despite the fact that a 1 : 1 mixture of E and 2 thioethers 188 was used, conditions were varied until a single stereoisomer of 19 1 was obtained. Similarly, a mixture of the vinyl ethers 189 was converted into one product, the bis-methyl ether 192.In all cases NOE studies showed that the endo product was obtained. 187X=H 188 X = SPh EiZ= 1 :I 189 X = OMe EiZ= 1:4.6 TBDMSO OMe lTiCt4 x, p e 190X=H 74% 191 X = SPh 80% 192 X = OMe 84% H Further progress in this approach is directed towards the introduction of oxygen at C-2 and the synthesis of compounds with a non-aromatic c-ring. Extensive studies on the first problem,55 were based on an efficient synthesis of the A-ring synthon 197 (Scheme 28). Addition of the lithiated THP-propargyl ether 194 to propionaldehyde first produced the alcohol 195. Lindlar hydrogenation of 195 and Swern oxidation next gave the a$-unsaturated ketone 196. Michael addition of lithiated ethyl isobutyrate to 196 then led to the ketoester 197.Dieckmann-like cyclization of 197 next gave the 1,3-diketone 198 which was subjected to a sequence of acetylation, deprotection, and oxidation leading to the key intermediate 193 in 43% overall yield for the eight-step synthesis. 194 OMP 195 OMP 196 OHC Hgo 193 Reagents: (i) H,, Pd, BaSO,; (ii) (CF,CO),O, DMSO; (iii) (CH,),CLiCO,Et; (iv) Bu'OK; (v) (CH,CO),O, Et,N; (vi), pTsOH, MeOH The substituted phenyllithium 199 was now added to the aldehyde 193 in the presence of CeCl,, and a 3 : 1 ratio of Cram to anti-Cram products was obtained. This mixture was separated and converted into the four products 200-203 as illustrated (Scheme 29); the vinyl ethers 201 and 203 were obtained as a mixture of E and Z isomers as in previous cases. The stereochemistry of the C-2 silyloxy group plays a crucial role in the cyclizations of compounds 200-203. Cyclization of 201 (Scheme 30) at - 78°C with TiCl, gave the endo product 204; on separate treatment of 204 with TiCl, at 0°C epimerization at C-10 produced the endo isomer 205.An unfavourable steric interaction involving the silyloxy group at C-2 causes 203 (Scheme 31) to cyclize via an ex0 transition state leading to the product 206. OMe I 200 X = H 201 X = OMe 193 + Meox - + Y 202X=H 203 X = OMe Reagents: (i) CeCI,; (ii) pyrrolidine, Bu'Me,SiCl, Et,N, separation; (iii) Et,SiCI, Et,N; (iv) Me,SiCH,Li, Bu'OK, for 200 and 202; Me,Si CH (OMe) Li, Bu'OK for 202 and 203 Scheme 29 Isomerization of 206 ex0 to endo was achieved by deprotection and heating; epimerization to the desired isomer 207 was realized on acetylation of the OH at C-2 and treatment with TiC1,.205 endo Reagents: (i) TiCI,, - 78°C; (ii) TiCI,, WC, 30 min. Scheme 30 62 Scheme 28 Contemporary Organic Synthesis203 206 ex0 0 * OAC OMe 207 Reagents: (i) NBu,F; (ii) heat, 30 min.; (iii) Ac,O, Et,N; (iv) TiCI,, Scheme 31 - 45T, 45 min. Two recent publication^^^^^^ have given further details of these cyclization and isomerhation reactions, and studies on the synthesis of a non-aromatic c-ring have been reviewed.5g The precursor 208 has been synthesized and cyclized to the tricyclic compound 209, whose stereochemistry was confirmed by X-ray crystallography. The objective now is to introduce the c-ring methyl group via conjugate addition and to further elaborate the structure to that of taxusin 120.TIPS0 0 =t@? H 0 208 209 120 Taxusin Frejd Frejd's approach to the taxane is a convergent strategy in which separate A and cring fragments are first synthesized, then coupled to give an A[B]C structure; a final cyclization to form the B ring completes the tricyclic structure. In a recent report5' Frejd used an enzymic resolution in the synthesis of an optically active c-ring unit. The racemic acetoxy enone 2 10 was converted into the enantiopure alcohol 2 1 1 ( > 99% e.e.) using an enzymic resolution/chemical hydrolysis sequence (Scheme 32). Elaboration of enone 2 11 next gave the silyl enol ether 2 12, a homochiral taxane c-ring analogue. The enol ether 2 12 was coupled successfully to the cyclohexane carboxaldehyde derived acetal2 13 giving the axially substituted product 2 14, which had the incorrect relative configuration at C-2 (C-3 in the eventual taxane skeleton).It is hoped that this can be altered at a later stage. Progressing to a functionalized cyclohexane carboxyaldehyde acetal ( A-ring fragment) in place of 2 13, Frejd naturally chose the acetal2 16, a derivative of compound 2 15 and an optically active A-ring unit synthesized from L-arabinose. Unfortunately, attempts to form the coupled product 2 17 have till now met with failure. 21 0 21 1 I II oms 0 21 3 I 212 0 .+o 03 21 5 21 6 I HO 214 OH 21 7 Reagents: (i) PLE; (ii) Na,CO,, MeOH; (iii) TBDPSCI, imidazole; (iv) CH,=CHMgBr, CuBr.Me,S; then TMSCI, TMEDA; (v) TiCI,, -75°C Scheme 32 The sequence from L-arabinose to 2 18 was somewhat arduous (23 steps), and is considered too lengthy to be of practical use.Nevertheless full details have just been reported,6O and another publication has revealed details of a much improved synthesis of the diol215:' outlined in Scheme 33. The ene reaction between the allylic ether 2 19 and ethyl glyoxylate 220a yielded none of the desired allylic alcohol when the reaction was catalysed by the chiral Lewis acid derived from (S)-1,1 '-binaphthalene-2,2'-diol and C1,Ti(OPri),. The reaction of 219 and the phenylmenthyl ester 220b using SnC1, as Lewis acid was successful though, and this auxiliary controlled reaction gave yields of the allylic alcohol 22 1 in excess of 90% with diasteroisomeric excesses greater than 95%. The protected alcohol 222 was homologated by a Claisen ester condensation.Subsequent nickel catalysed coupling of a silyl Grignard reagent and an Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 6321 s b R = 8-phenylmenthyl (v)*(vi) 21 8 6296 4 PMBO PMBO - 215 - COpEt 78% TMS TMS 223 Reagents: (i) SnCI,, DCM, -78°C; (ii) 1M NaOH, THF, MeOH; (iii) DBU, Et8r. PhH, A; (iv) TBSCI, imidazole, DCM; (v) LiHMDS-EtOAc, TMEDA, THF; (vi) ButOK, CIP0,Eh. THF; (vii) TMSCH,MgCI, Ni(acac),, EQO; (viii) DDQ DCM, H,O, 0°C; (ix) ButO,H, TiOPr',, (-)-diethy1 tartrate; (x) BF,.OEt, Scheme 33 22s 32:l ais:Erans OHC 227 0 0 228 230 229 Reagents: (i), Bu'OK, Bu'OH, Mel; (ii) H, 10% Pd-C; (iii) BF,.OE%, DCM; (iv) HO(CH,),OH, p-TsOH. PhH, A; (v) DMSO. (COCI),, Et3N, DCM; (vi) TMSOTf, collidine, DCM; (vii) 0,.DCM, pyridine, - 78°C then PPh,; (viii) CH,N, Et,O; (ix) 1 N HCI, THF, r.t., (x) Sml,, THF-MeOH, - 25°C; (xi) NalO,, THF, MeOH Scheme 34 enol phosphate gave, after deprotection, the allylsilane 223, a precursor to the diol215. The stage is now set for coupling the A and c-rings. Arseniy adas Arseniyadas62 has used the derivative 224 of the known lower analogue of the Wieland-Miescher ketone as a precursor in his synthesis of an A-ring equivalent (Scheme 34). The homochiral compound 224 was modified, as depicted, in a highly efficient, stereoselective process. The cis ring junction in 225 was introduced by catalytic hydrogenation, and only a small percentage of the undesired trans isomer was detected. Another interesting point to note in this sequence is the conversion of the silyl enol ether 226 into both the acyloin 227 and the required ester- aldehyde 228 products.The by-product 227 was produced in sigdicant amounts (23%), but could easily be converted into the desired product 228 by periodate cleavage and esteacation, so conveniently increasing the yield of the required compound. Conversion of the aldehyde-ketone 229 into bicyclic ketone 230 by a samarium diiodide-mediated reductive coupling followed by oxidation occurred stereospecifically as a consequence of the cis ring junction in the hydrindanone 225, and this importantly fixed the absolute configuration of the C-1 centre. The authors aim to couple this A-ring fragment, 230, to a c-ring equivalent using enolate chemistry, and then complete the B-ring to form a complete taxane skeleton (Scheme 35).64 Contemporary Organic Synthesis230 0 0 R' I' Scheme 35 Wendec A very efficient synthesis of a c-aromatic taxane structure has been published by Wender.63 The key steps of (i) rearrangement to give a quaternary centre in the A-ring precursor and (ii) an hydroxy epoxide fragmentation to produce the B-ring, are related to the Holton synthesis. The starting material for the Wender route (Scheme 36) is pinene which is available in both enantiomeric forms and contains ten of the twenty carbon atoms of the taxol skeleton. Pinene was first subjected to air oxidation to give verbenone 23 1; deprotonation followed by alkylation next produced the enone 232. Irradiation of 232 achieved the crucial rearrangement to the ketone 233.The stereochemistry of the cyclization of ketone 233 is determined by its bicyclic structure, and leads to a single alcohol 234. Epoxidation of 234 at C-1 led to a single epoxide 235, which fragmented to the taxoid 236. Oxidation of 236 Q to the carbonyl group occurred under basic conditions and reduction of the resulting hydroxyketone 237 led to the c-aromatic taxoid 238 in enantiomerically pure form. The efficiency of this route shows great potential for further elaboration to taxol and related compounds. Clearly the key question is whether the c-aromatic ring can be fashioned into the functionalized c-ring of taxol. Nicolaou Nicolaou has reported an enantioselective synthesis of the fully functionalized A-ring of tax01,~~ together with the synthesis of the c- and D-rings in racemic form.65 The diene 239 (Scheme 37) was first prepared from the appropriate ester,"6 and then subjected to thermal Diels-Alder reaction with 2-chloroacrylonitrile.The adduct 240 was next treated with base to introduce the carbonyl group which was then converted into the ketal24 1 after reacetylation of the alcohol. Regioselective allylic oxidation of 24 1 with Se02 was followed by pyridinium chlorochromate (PCC) oxidation to produce the enone 242, which with the oxazaborolidine procedure developed by core^^^ gave the corresponding allylic alcohol 243 in greater than 98% e.e. Removal of the ketal group in 243 and protection of the alcohol function then gave the fully functionalized taxol A-ring 244 in essentially optically pure form. Nicolaou's synthesis of the taxane c, D-rings is yet again based on the Diels-Alder reaction (Scheme 38).The dienophile is the unsaturated ester-alcohol 245, the diene is 3-hydroxy-2-pyrone 246, and the reaction is made intramolecular using phenylboronic acid. The presumed intermediate 247, where the two components are temporarily tethered together, undergoes regioselective cyclization to give 248 as an initial product which rearranges under the reaction conditions to the lactone 249. Rearrangement back to a bicyclo[2.2.2] lactone 250 occurred under the influence of potassium hydride during the benzylation of 249. Both the ester and the lactone groups in 250 were reduced with Red-A1 to give the trio1 25 1. Acetal formation, hydroboration, and acetylation of both the primary and the secondary alcohols in 25 1 next produced the triply protected diacetate 252.Reorganization of these protecting groups by acetal removal, silylation, and acetate hydrolysis then gave the diol253 which was converted into the mesylate 254. The crucial oxetane ring forming reaction proceeded well, and a final desilyation yielded the fully functionalized, racemic taxol c, D-ring fragment 255. 238 237 236 Reagents: (i) Bu'OK; (ii) hv; (iii) Bu'Li, TMEDA; (iv) (a) Ti(OPr'),, Bu'OOH; (b) DABCO, heat; (v) Bu'Me,SiCI, imidazole; (vi) KOBu', 0,, 60°C; Scheme 36 (vii) Na, EtOH Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 65A ring c and 0 rings x 2Y' CN (i) - 85% CN Cl 0d 240 241 243 242 (vii), (viii) 244 Reagents: (i) 135"C, 96 h, 85%; (ii) KOH, Bu'OH; (iii) Ac,O, DMAP; (iv) HOCH,CH,OH, CSA; (v) SeO,, then PCC; (vi) (R)-oxazaborolidine, catecholborane; (vii) TsOH, acetone, H,O; (viii) Bu'Me,SiOTf, 2,6-lutidine Scheme 37 Having completed effective routes to both A and c-ring units Nicolaou has now joined a functionalized A-ring fragment to a simplified c-ring component (Scheme 39).68 The ketone 256 was converted into the vinyl lithium 257 using the Shapiro reaction, and the aldehyde 258 was then added to produce the alcohol 259.A 2 : 1 mixture of diastereoisomers of 259 was formed from which the required alcohol 259 was separated by chromotography. Vanadium catalysed epoxidation of the allylic alcohol 259 next led to the epoxide 260 which was then reduced to the diol26 1 with LiAlH4.Protection of the diol26 1 leading to the acetonide 262 was followed by a sequence of selective deprotections and oxidations to form the di-aldehyde 263. McMurry coupling of 263 gave the diol264 as a 1 : 1 mixture of diastereoisomers which was then oxidized to the enediol265 with MnO,. 2.4 Syntheses starting from the Wieland-Miescher ketone Danishefsky The Wieland-Miescher ketone 266 is an important commercially available, enantiomerically pure, starting material. The c and D-rings in the taxanes have been prepared from the Wieland-Miescher ketone as shown in Scheme 40.69 The alcohol 267 was prepared by the method of Heathc~ck,~~ and protection followed by stereoselective hydroboration and oxidation produced the ketone 268. Conversion of 268 to the corresponding enol triflate was followed by a palladium-catalysed carbonylation reaction in the 245 246 L EtO2C.. $JH 249 (ii) 80% J h EtO2C 250 247 \ TBDMSO OSMDBT (vii)-(ix) 4 86% OH OAc 253 (x) 80% J TBDMSO OSMDBT L..M 252 HO OH (xi), (xii) 86% 254 255 Reagents: (i) PhB(OH),, 90°C.48 h then 2,2-dirnethylpropane-1,3-diol; (ii) KH, PhCH,Br; (iii) Red-Al; (iv) 2,2-dimethoxypropane, CSA; (v) BH,.THF then H,O, NaOH; (vi) Ac,O, DMAP; (vii) CSA, MeOH; (viii) Bu'Me,SiOTf, 2,6-lutidine; (ix) NaOMe, MeOH; (x) MeSO,CI, DMAP; (xi) NaH, 45", 12 h; (xii) Bu,NF Scheme 38 presence of methanol to give the ester 269. Reduction of 269 to the allylic alcohol, then hydroxylation to the olefin led to the trio1 270 as the major product. Formation of the D-ring from 270 was achieved by the selective silylation of the primary alcohol group and then conversion into the secondary triflate.Heating the triflate with ethylene glycol caused desilylation and cyclization to the oxetane which was then hydrolysed to the ketone 27 1. Deprotonation of 2 7 1 with LDA followed by reaction with trimethylsilyl chloride gave the corresponding trimethylsilyl enol ether, which was treated with Pd( OAc), according to the method of Ito7I giving rise to the enone 272. Formation of a 66 Contemporary Organic Synthesis256 257 258 137% MEMO, ,OBn MEMO, ,OBn - 0J-J OH HC 261 * 264 263 265 Reagents: (i) 2,4,6-triisopropylbenzenesuIfonyI hydrazine; (ii) BuLi, THF, -78°C then 0°C; (iii) Bu'OOH, VO(acac),; (iv) LiAIH,; (v) 2,2-dimethoxypropane, camphor sulfonic acid; (vi) H,, Pd/C; (vii) Ac,O, 4-DMAP; (viii) TiCI,; (ix) K,CO,, MeOH; (x) tetrapropylammonium perruthenate, 4-methylmorpholine-N-oxide; (xi) TiCI,-(DME),.,, Zn-Cu; (xii) MnO, Scheme 39 trimethylsilyloxy diene from 272 followed by ozonolysis finally gave the dialdehyde 273; alternatively the enone 272 was hydroxylated to give the hydroxyketone 274.have reported the synthesis of other taxane intermediates containing the ring.^^ Thus, reaction between 2-methylpentane-3-one 275 and acryloyl chloride was carried out by a known procedure to first give the ketone 276.73 Conversion of 276 into the enol triflate 2 77 and reaction with vinyltributylstannane, with Pdo catalysis, followed by the hydroboration next produced the alcohol 278. Silylation of the alcohol In a separate publication the Sloan-Kettering group 278, and regioselective allylic oxidation with chromium trioxide- 3,5 -dimet hylp yrazole then gave the A-ring synthon 279.The enone 280 was prepared by Swern oxidation of the alcohol 278 followed by addition of 2-propenylmagnesium bromide and a second Swern oxidation. Regioselective Diels-Alder reaction of the enone 280 with the Danishefsky diene next yielded the taxol A , c-ring synthon 28 1. Finally, the enolate from 280 was hydroxylated with the Davis ~xaziridine,~~ and the product was oxidized to the diketone 282 which led to the ~ , c - r i n g synthon 283. Clearly the Sloan-Kettering group are now poised to combine the work described in Schemes 40 and 4 1. Watt The Wieland-Miescher ketone 266 has also been used in an A-ring synthesis (Scheme 42).75 Protection of the a,#?-unsaturated carbonyl group in 266 with 1,2- ethanedithiol gave a thioacetal and, despite the fact that the saturated carbonyl group is hindered, addition of t-butyldimethylsilyl cyanide proceeded stereoselectively to produce the protected cyanohydrin 284.Selective removal of the thioacetal group in 284 occurred with Tl(NO,), leading to the enone 285. The a-acetoxy ketone corresponding to 286 was prepared by the reaction of 285 with Pb( OAc), and this reacted with methanol and potassium carbonate to give the a-hydroxy ketone 286. Periodate cleavage of 286, followed by treatment with diazomethane then yielded the ester aldehyde 287. Finally, decarbonylation with Wilkinson's catalyst provided the A-ring synthon 288.3 Semi-syntheses of taxanes This approach to taxanes has, to date, been the most successful way of making taxol and biologically active analogues. Potential starting materials for semi- synthesis must be easy to obtain, renewable, and require as little elaboration as possible.76 10-Deacetylbaccatin 111 289, first described as a degradation product of taxol,' and isolated from needles of the widely distributed Taxus baccata (ca. 1 g/kg dry leaves)77 nicely meets these criteria. Synthetic routes to taxanes utilizing 289 have been developed to exploit the differing reactivity of the free hydroxyl groups; 7-OH > 1 0-OH % 13-OH (the low nucleophilicity of the 13-OH, is due to H-bonding to the C-4 acetyl C = 0 group and is also on the endo-convex face). Sharpless ~xyamination~~ of 290-obtained by sequential protection of 289 and formation of the C- 13 cinnamate-gave a mixture of regio- and stereo-isomers with little control (Scheme 43).The reaction was later improved7Y by the addition of dihydroquinine p-chlorobenzoate and although regiocontrol was again poor the required (2'R, 3's) stereoisomer 29 1 was now the major product (d.e. - 60%). This isomer was converted into taxol by removal of the t-butyl amido group, followed by benzoylation and removal of the trichloroethoxycarbonyl group. Although the poor control in the oxyamination reaction renders the Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 670 d OTBS e o @- oT, 266 267 268 0 do (viii) c--- 66% 'OH HO 272 \OH 270 273 274 KHMDS, THF, - 78°C then PhNTf,; (vi) Pd (OAc),, PPh,, CO, MeOH; (vii) DIBAL, - 78°C; (viii) 5 mol% OsO,, NMMO; (ix) TMSCI, pyridine, - 78°C then Tf,O - 78°C --L r.t.followed by ethyleneglycol, 40"C, 12 h; (x) collidinium tosylate, acetone, H,O; (xi) 2 equiv. LDA, -78°C then TMSCI; (xii) Pd(OAc), then MeOH, K,CO,; (xiii) TBSCI, imidazole; (xiv) LDA, THF, -78°C then TMSCI then 0,, CH,CI,, - 78°C then Ph,P; (xv) TMSCI, pyridine; (xvi) KHMDS, THF, - 78°C then 2-(phenylsulfonyl)-3-phenyloxaziridine then H,O Reagents: (i) steps reference 70; (ii) TBSOTf, 2,g-lutidine; (iii) BH,-THF then H,O, NaOH; (iv) tetrapropylammonium perruthenate; (v) Scheme 40 C P 0 275 276 , 277 266 284 c- 75% 70% TBDMSO TBDMSO 288 287 286 Reagents: (i) HSCH,CH,SH, p-TsOH then TBSCN, Znl,; (ii) TI (NO,),, MeOH, H,O; (iii) Pb(OAc),; (iv) K,CO,, MeOH; (v) NalO,, H,O, Bu'OH; (vi) CH,N,; (vii) RhCI(PPh,),, 80°C Scheme 42 279 approach impractical it has nevertheless given access to many analoguesso including RP 56976 and taxotere,81 which has similar pharmacological activity to taxol.A more direct approach to taxol was first described by the research groups of Potier and Greene,82 involving the esterification of 7-triethylsilylbaccatin I11 292 with the taxol side-chain acid, Scheme 44. The use of a large excess of the protected acid 293 (6 eq.), 1,3-dicyclohexylcarbodiirnide (DCC), and N,N-dimethyl-4-aminopyridine (DMAP) to effect the esterification was followed by deprotection of the silyl and ethylethoxy protecting groups to give taxol(36% from 10-DAB 111).Other acyl-activated side chain equivalents have been used in attempts to overcome the problems associated with the low reactivity of the 281 Po 283 O b J w 282 O d Reagents: (i) steps, reference 73; (ii) KHMDS, PhNTf,; (iii) Bu,SnCH=CH,, cat. Pd (PPh,),; (iv) 9-BBN; (v) TBDMSCI, Et,N, DMAP; (vi) Cr03-3,5-DMP; (vii) Swern oxidation; (viii) BrMgC(Me)=CH,; (ix) Danishefsky diene, 125"C, then HCI, H,O; (x) KHMDS, F. Davis oxaziridine; (xi) Danishefsky diene, 80°C, then HCI, H,O Scheme 41 68 Contemporary Organic Synthesis289 1 0-Deacetyl baccatin 111 290 AgNcg, Os04 :kmNNacl n I 291 (I) TMSl (ii) PnCOCl (iii) WAcOH 4 Tax011 Scheme 43 C- 13 OH (epimerization of the 2' centre, and generally low yield~)-Ojima~~ and Holtons4 have independently used the P-lactam derivatives 294 to directly couple with 7-TES-10-DAB I11 292.OjimaS5 has further reported a significant improvement to the P-lactam method that avoids large excesses of the p-lactam. A near quantitative coupling can be achieved by sequential treatment of the 7,lO-ditroc- 10- deacetylbaccatin I11 (troc = 2,2,2- trichloroethoxylcarbonyl) with sodium hexamethyldisilazide (2.5 equiv.) and the lactam 294, providing an efficient route to taxotkre. Holtons6 has developed another method that utilizes the oxazinone 295 as the acyl equivalent. It is interesting to note that Swindell has also invoked intermediate oxazinone derivatives as coupling agents.87 4 Syntheses of the C-13 side chain of taxol The C-13 side chain in taxol, the (2'R,3'S)-3'- phenylisoserine unit, presents an interesting and manageable sub-target for asymmetric syntheses, and has consequently seen many elegant approaches.In addition, since the binding of taxol to microtubulesS7 is particularly sensitive to changes in the structure of the side-chain, many active analogues of taxol have been made by semi-synthesis. 4.1 Phenylglycidate synthon method In their first synthesis Greene8* (Scheme 45) and his group started with methyl phenylglycidate 296a which HO- 292 (7-Triethylsilyl)-l O-deacetybaccatin 111 1 (i) AcCI, PY (ii) 283,294 or 285 Taxol BzNH 0 Ph 4 0 H 6eq. OR 36% to taxol 293 Ph -OR 92% to tax01 294 phT?J;H 4eq. Ph 69% to taxol 295 Scheme 44 was made by way of Sharpless epoxidation of cis-cinnamyl alcohol, followed by oxidation and esterification. The amine group was next introduced stepwise following ring-opening of the epoxide with trimethylsilyl a i d e {to give a hydroxy azide which was esterified to give the benzoate 297), and reduction of the azide.The azide reduction was accompanied by 0 -, N migration of the benzoyl group, a procedure followed subsequently by other workers, to give the desired product 298. 0 (i) Bu'OOH. Ti(OPr'), L-(+)-DET, (65%. 80% e.e.1 Ph (ii) RuCS. Nal04; H+ * PhACO,R CH&2 (a%) - - . . 296a R = Me bR=Et (i) MeSiNs ZGI2 (90%) (ii) PhCOCI, NEts DMAP (94%) 1 N3 0 =OM* ~ H2 lO%Pd/C 89% 298 Scheme 45 297 Greene's second and improvedsY synthesis (Scheme 46) is essentially a refinement of the synthesis of 296a. Asymmetric Sharpless dihydroxylation of methyl cinnamate, using dihydroquinidine 4-chlorobenzoate first gave the diol299; a new Sharpless procedureg0 gives 299 (ethyl ester) with even higher selectivity (97% e.e.).The diol299 was next tosylated selectively 69 Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanesto give the 2-tosylate 300 where the high selectivity is thought to be a consequence of strong C-3-OH to ester hydrogen bonding. The epoxide 296a was then obtained from 300 by treatment with potassium carbonate and elaborated to 293 and the taxotkre methyl ester side-chain as described before. 0 P h q O M e OH 299 dOMe (i) DCQB. NMNO, oso, (cat.) Ph 51% TsCI, Et3N t U O M e 91% Ph K&%. H#. DMF 296a Scheme 46 6TS 300 Commergon et al. have also made the ethyl ester 296b using Evans chiral-enolate chemistry (Scheme 4 7).91 Reaction of the boron enolate of the bromoacetyl30 1 with benzaldehyde first gave the bromoalcohol302, and formation of the epoxide and concomitant removal of the auxiliary then gave the ester 29613.OH 0 0 P h v N A O FNl0 (ii) (i) Et3N, PhCHO BUaBOTf Br w. 5870 Br ,w- / 'Ph *' .Ph 30 1 302 EtOLi 81% I 0 PhAC02Et 296b Scheme 47 JacobsenV2 has reported a similar approach to 306 starting from the epoxide derived from ethyl cis-cinnamate (Scheme 48). The catalytic epoxidation of ethyl cis-cinnamate with 6 mol% ( salen)MnllT complex93 303 and commercial bleach gave rise to the epoxide 29613 in excellent (95-97%) e.e. A modified procedure whereby the epoxide 296b was treated with ethanolic ammonia to give the amide 304 followed by hydrolysis with barium hydroxide then gave the acid PhmC02Et NaOCl -b (R,R) 303 (6 mol%) 305 without epimerization.Generation of the side- chain 306 from 305 was effected by simple treatment with benzoyl chloride. 4.2 The Staudinger synthesis of #blactams A chiral-pool approach to the C- 13 side-chain in taxol is described by Farina and shown in Scheme 49.94 Thus, a highly selective Staudinger reaction between the L-threonine derived imine 307 and acetoxyacetyl chloride first led to the cis p-lactam 308 (PI%, 84% d.e.). This P-lactam was subsequently converted into the #I-lactam 309, by removal of the silyl group, elimination of water, and ozonolysis to give the corresponding mixed oxalic acid derivative, which was simply hydrolysed to the required p-lactam 310. Georg et aL9' have also described a chiral-pool Staudinger reaction (Scheme 50) in which the galactose imine 3 1 1 and the acid chloride 3 12 gave a 2 : 3 mixture of the diastereoisomeric P-lactams 313.Hydrolysis of both the monosaccharide and the p-lactam groups in 313, followed by benzoylation then led to the amide 3 14. Removal of the hydroxyl protecting group in 314 gave the unnatural (2'S,3'R) enantiomer 315. A more detailed study has shown that this poor selectivity is observed with other galactose imines.'6 Aco(hc' Ad,. ,Ph 0 p h l l ?SiPh20But NEt3 ?SiPh20But 74%. d.9.U% - 0 Gp N-P C02Me C02Me 307 308 (I) TBAF (ii) MsCI. NEb I (iii) 0, HO, ,Ph Ad., ,Ph NaHC03 0 f i H 66~frorn * 308 0 J $ O 31 0 C02Me 309 Scheme 49 HoltonY7 has used the p-lactam 3 16 to make taxol from 7-triethylsilyl- 1 O-deacetylbaccatin I11 as described earlier (Scheme 5 1).He made the lactam using the Staudinger reaction between a-acyloxy acetyl chloride and the imine 31 7 as the key step to 0 NH3. EtOH b PhAC02Et 100% 296b 56%,95-97?4 e.e. PhCONH 0 : II 306 (74%) 304 65% &OH 305 (92%) Scheme 48 70 Contemporary Organic SynthesisAcO OAc Ad-!&&N-Ph OAc 31 1 (i) PMPOCH,COCI 312, NEb 1 Ph 313 (4:6,75”!0) bh \ PhCONH 0 PhCONH 0 (NHl)&e(N03)6 phvo Me OPMP Ph+&Me * OH 315 (95%) 314 (84%) Scheme 50 0 NEt3 68% OMe 31 7 318 (45% e.e. 98%) EEO,. ,Ph HO, ,Ph (i) H2C=CHOEt, pTsOH ph (ii) MeLi; BzCl 0 0 31 6 Scheme 51 319 (92oJo) give the p-lactam 3 18, which was converted into 3 16 using standard transformations. The alcohol 3 19 was obtained enantiomerically pure by resolution of its 2-methoxy-2-( trifluoromethyl)phenylacetic ester.Ojirna’sg8 strategy, shown in Scheme 52 is also based on the use of P-lactams, made by a highly selective ester enolate-imine condensation. Thus, deprotonation of the triisopropylsilyl protected ester 320 with lithium diisopropylamide followed by condensation with the imine 32 1 gave exclusively the cis B-lactam 322 (97%). The lactam 322 was then converted into the hydrochloric salt of 305 by treatment with HCl. Palomo99 has made the related P-lactam derivative 324 by cis-selective reduction of 323 (Scheme 53).loo Protection of the hydroxyl group in 323 and N-dearylation led to the P-lactam 324 which by standard treatment gave the ester 298. 4.3 Lithiobenzylamine syn thon method Two approaches that introduce the 3’ carbon in the C- 13 side chain of taxanes, from benzylamine have been developed.Thus, Greene et aZ.lol amongst their many reports have described an alternative synthesis of the side chain of taxotkre (Scheme 54). In this approach dilithiation of BOC-benzylamine with s-butyllithium first gives the dianion 325 which adds to 320 322 321 (97%) ~ G N H C I 305 Scheme 52 323 324 (70%) 298 Scheme 53 acrolein to produce the hydroxycarbamate 326 with reasonable selectivity (syn:anti 6 : 1). The syn preference observed here is consistent with a chelated transition state of the type 327. Protection of the 2’ OH in the syn alcohol 326 as its (trichloroethoxy )methyl ether, followed by oxidative cleavage and resolution using ( + )-ephedrine finally gave the protected taxotkre side-chain 328.r t i 1 325 HNBOC (i) BrAOACCls (70%) HYBOC , 326 (syn :antiB:l) 328 ?OC Scheme 54 Davies et aZ.lo2 have reported a strategy towards 333 involving conjugate addition of the homochiral lithio ( R )-( a-methylbenzy1)benzylamide 329 to t-butyl cinnamate followed by hydroxylation of the intermediate enolate with ( + )-( camphorsulfony1)oxaziridine 330 leading to the anti hydroxy arnine 33 1 with excellent selectivity (92% d-e.). When 33 1 was subjected to hydrogenolysis followed by methanolysis and benzoylation the anti hydroxy amide 332 was produced which could be converted into the corresponding syn (2’S,3’R) isomer 333 (the enantiomer of 298) via Mitsunobu inversion (Scheme 55). Since ( S)-( a-methylbenzy1)benzylamide is readily available, this method will also produce the taxol side chain with the natural (2‘R,3’S) configuration.Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 7172 Ph (i) Ph 0 . .. - (f7)-329 Ph v OH (ii) &y 86% (92% d.e.) 331 $,\ 0 0 0 (+)-330 (i) H2 (7 atm.). PdK: (ii) HCI. MeOH (iii) PhCOCI, NEt3 I PhCONH (i) Diethyl azodicarboxylate. PhCONH PPha phv OMe (iii) ii) HCI NaHC4 P h v O M e 6H 333 Scheme 55 Ph V 0 E t OH 332 (92%) + P h v O E t Ph VOEt 0 334 (35%, e.e. >98%) 4.4 Enzymic syntheses The first example of an enzyme-assisted taxol side-chain synthesis came from the group of H0nig,lo3 in which the racemic butyryl ester ( f )-334, obtained from ( f )-ethyl cis-P-phenylglycidate, was resolved by selective hydrolysis of the (2'S,3'R) isomer with Pseudornonas fluorescens, leaving the required (2'R,3'S) ester 334 unreacted (e.e.> 98%). transesterification of methyl trans-p-phenylglycidate ( k )-335 has been described by Chen.lo4 The best result was obtained with Mucor miehei lipase MAP- 10 An enzymatic resolution involving lipase-mediated Muwr meihei MAP- 10 lipase Ph 0 - \L1, kobutanol C02Me using isobutanol as the acyl acceptor. The ( - )-methyl ester 335 (42%,95% e.e.) and the ( + )-isobutyl ether 336 (43%, 95% e-e.) could be separated by chromatography or fractional distillation. Interestingly, both 335 and 336 can be converted into the azide 337 in 40% and 38% yields respectively. The route from 335, illustrated in Scheme 56, involves epoxide ring-opening by bromide ion and subsequent displacement with sodium wide with overall retention of configuration at the 3 ' position; the same sequence from ( + )-336 results in both the 3' and 2' positions being inverted leading to 337.Sihlo5 has reported a comprehensive study of the lipase-mediated kinetic resolution of P-lactam derivatives. For example, the racemic B-lactam ( k )-338 gives ( + )-338 in high yield, with impressive stereoselection, on treatment with immobilized lipase P-30 (from Pseudornonas cepacia). AcO, ,Ph -m 0 * N K P h 0 (f)-338 Pseudomonas Cq0ack-i (P-30 J 50%.48h J A d , ,Ph 'I7 0 * N K P h 0 (+)-338 (45% e.e.98%) HO Ph 0 0 43% e.e. 99.5% 4.5 Aldol reaction approaches Hanoaka et aZ.lo6 have used an asymmetric aldol reaction between the homochiral chromium complex 339 of o-trimethylsilyl benzaldehyde and the titanium enolate of 340 (Scheme 57).The reaction is highly anti selective yielding only the alcohol 34 1; interestingly reaction of the corresponding lithium enolate was syn selective (syn: anti 4 : 1). Sequential decomplexation of the chromium from 34 1, followed by deprotection of the silyl group and Mtsunobu 0 + PhP8kO2E3ui (*)-335 (-)-Xi5 (42%, 95% e.e.) (+)-336 (43%, 95% e.e.) PhCONH 0 (ij Et2NH#r, Et#l N3 0 -------+ Ph40Me (ib NaN3 * P h d O M e (-)-335 OH 0 (ii) (i) EtCl NaN3 PhCONH 0 SOCI, NAO Php'k02Bui (lv) (iii) MeOH, H,. PdlC Na2C& * Ph+Otvle - P6 C02Me (+)-336 OH Scheme 56 Contemporary Organic Synthesisdisplacement of the hydroxyl group then led to the aide 342. Reduction of 342 with wet triphenylphosphine, followed by benzylation, thallium ( 111) assisted thioester-to-ester interconversion, and deprotection of the benzyl ether finally completed the synthesis of 298.added to vinylmagnesium bromide to give predominantly the syn aUylic alcohol 348 where the selectivity is explained by chelation controlled addition to 350. The alcohol 348 was then protected and the alkene group oxidized to give the required acid 349. 339 341 93% (anti :syn 935) 1 (i) hv, E W (ii) TBAF-HF (6!5%) (iii) HN3. DEAD, Pph, PhCONH 0 . . 298 342 Scheme 57 Yamamoto and his colleaguesio7 have described an efficient enolate-imine condensation involving the imine 343 and the Z-enolate 344, catdysed by the phenylborate (S)-345, leading to the (2'R,3'S) amino alcohol 346 (syn: anti 99 : 1; syn 98% d.e.).The observed selectivity is only slightly reduced if instead the enantiomer (R)-345 is used (syn: anti 94 : 6; syn 94% d.e.). Removal of the a-methylbenzylamine group in 346, followed by selective hydrogenolysis, and Schotten-Baumann benzoylation then produced 306 (Scheme 58). OMe 306(68%) 346 (>9OY0, syn :anti99:1; d.e. 98?/o) (S)-345 Scheme 58 4.6 A chiral pool approach Greene et allo8 have reported a synthesis of 349 starting from (S)-phenylglycine (Scheme 59). Thus, the protected amino aldehyde intermediate 347 was . . 347 (74%) (i) H&=CHMgBr 349 Scheme 59 5 References 1 M.C. Wani, H.L. Taylor, M.E. Wall, P. Coggon, and A.T. McPhail, J. Am. Chem. SOC., 1971,93,2325. 2 WJ. Slichenmyer and D.D. Van Hoff, Anti-Cancer Drugs, 1991,2,519; E.K.Rowinsky and R.C. Donehower, J. Natl. Cancer Inst., 1991,83,1778; F.A. Holmes, R.S. Walters, R.L. Theriault, A.D. Forman, L.K. Newton, M.N. Raber, A.U. Buzdar, D.K. Frye, and G.N. Hortobagyi, J. Natl. CancerInst., 1991,83,1797. 277,665; S.B. Horwitz, Trends Pharm. Sci., 1992,13, 134. 4 S. Borman, Chern. Eng. News, 1991,69,11. 5 S. Blechert and D. Guinard in The Alkaloids, ed. A. Brossi, Academic Press, 1990,39,195; L.A. Paquette in 'Stud. Nat. Proc. Chem'., 1992,ll [Stereoselective synthesis pt. (G)], 3; see also Ref. 8. 6 An overview of semi-synthesis is included in Kingston's review of the chemistry of taxol, D.G.I. Kingston, Pharmac. Ther., 1991,52,1. 3 P.B. Schiff, J. Fant, and S.B. Horwitz, Nature, 1979, 7 S.M. Edgington, Biotechnology, 1991,933.8 For a comprehensive review of approaches towards the total synthesis of taxanes published up to 1990 see: C.S. Swindell, Org. Prep. Proc. Znt., 199 1,4,465. 9 S.A. Hitchcock and G. Pattenden, Tetrahedron Lett., 1992,33,4843. 10 (a) Y. Ohtsuka and T. Oishi, Chem. Pharm. Bull., 1988, 36,4722; (b) Y. Ohtsuka and T. Oishi, Tetrahedron Lett., 1986,27,203. 11 Y. Ohtsuka and T. Oishi, in 'Studies in Natural Products Chemistry', ed. Atta-ur-Rahman, Vol. 3 [Stereoselective Synthesis (Pt B)], p. 73,1989, Elsevier, Amsterdam. 12 Y. Ohtsuka and T. Oishi, Chem. Pharm. Bull., 1991,39, 1359. 13 R. Andriamilisoa, M. Fetizon, I. Hanna, C. Pascard, and T. Prange, Tetrahedron, 1984,40,4285. Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 7314 M.Benchikh le-Hocine, D. Do Khac, and M. Fetizon, 15 M. Benchikh le-Hocine, D. Do Khac, M. Fetizon, and T. 16 M. Benchikh le-Hocine, D. Do. Khac, M. Fetizon, F. Synth. Commun., 1992,22,245. Prange, Synth. Commun., 1992,22,1871. Guir, Y. Guo, and T. Prange, Tetrahedron Lett., 1992, 33,1443. Fetizon, A. Neuman, and T. Prange, Acta Crystallogr., Sect. C., 1991,47,2109. 18 J. Oh, J.-R. Choi, and J.K. Cha, J. 0%. Chem., 1992,57, 6664. 19 Y.-F. Lu and A.G. Fallis, Tetrahedron Lett., 1993,34, 3367. 20 Z. Wang, S.E. Warder, H. Perrier, E.L. Grimm, M.A. Bernstein, and R.G. Ball, J. 0%. Chem., 1993,58, 2931. 105,3732; (b) B.M. Craven and K. Sakan, Acta Crystallogr., Sect. C, 1983,39, 1556. 22 K. Sakan and D.A. Smith, Tetrahedron Lett., 1984,25, 2081. 23 K. Sakan, D.A.Smith, S.A. Babirad, F.R. Fronczek, and K.N.Houk, J. Org. Chem., 1991,56,2311. 24 (a) H. Neh, A. Kuhling, and S. Blechert, Helv. Chim. Acta, 1989,72,101; (b) H. Neh, S. Blechert, W. Schnick, and M. Jansen, Angew. Chem., Int. Ed. Engl., 1984,23,905. 27,2845. 1992,48,6953. Ed. Engl., 1991,30,412. 17 F. Guir, D. Do Khac, M. Benchikh le-Hocine, M. 21 (a) K. Sakan and B.M. Craven, J. Am. Chem. SOC., 1983, 25 R. Kaczmarek and S. Blechert, Tetrahedron Lett., 1986, 26 S . Blechert, R. Muller, and M. Beitzel, Tetrahedron, 27 S. Blechert and A. Kleine-Klausing, Angew. Chem., Int. 28 G.A. Kraus and D. Zheng, Synlett, 1993,7 1. 29 G. Saha, A. Battacharya, S.S. Roy, and S. Ghosh, Tetrahedron Lett., 1990,31,1483. 30 G. Saha, A. Karpha, S.S. Roy, and S. Ghosh, J. Chem. SOC., Perkin Trans. I , 1992,1587.3 1 (a) G. Saha and S. Ghosh, Synth. Commun., 1991,21, 2129; (b) S. Ghosh, G. Saha, G. Mostafa, and S. Ray, J. 0%. Chem., 1992,57,7344. Tetrahedron Lett., 1992,33,2363. 2987. 32 S. Ghosh, A. Karpha, G. Saha, and D. Patra, 33 S. Sarkar and S. Ghosh, Tetrahedron Lett., 1993,34, 34 R.A.Holton, J. Am. Chem. SOC., 1984,106,5731. 35 R.A. Holton and R.M. Kennedy, Tetrahedron Lett., 36 R.A. Holton, R.R. Juo, H.B. Kim, A.D. Williams, S. 1984,25,4455. Harusawa, R.E. Lowenthal, and S. Yogai, J. Am. Chem. SOC., 1988,110,6558. 37 R.A. Holton in ‘Strategies and Tactics in Organic Synthesis’, Academic Press Inc., 1991,3,165. 38 K.J. Shea and S. Wise, J. Am. Chem. SOC., 1978,100, 6519. 39 K.J. Shea and P.D. Davis, Angew. Chem., Znt. Ed. Engl., 1983,22,419.40 K.J. Shea and J.W. Gilman, Tetrahedron Lett., 1983,24, 657; 1984,24,2451; J. Am. Chem. Soc., 1985,107, 4791; KJ. Shea and S.T. Sakata, Tetrahedron Lett., 1992,33,426 1. 41 R.W. Jackson, R.G. Higby, and KJ. Shea, Tetrahedron Lett., 1992,33,4695; R.W. Jackson, R.G. Higby, J.W. Gilman, and KJ. Shea, Tetrahedron, 1992,48,70 13. 42 P.A. Brown, P.R. Jenkins, J. Fawcett, and D.R. Russeli, J. Chem. SOC., Chem. Commun., 1984,253; P.A. Brown and P.R. Jenkins, J. Chem. SOC., Perkin Trans. I , 1986, 1303. 43 R.V. Bonnert and P.R. Jenkins, J. Chem. SOC., Chem. Commun., 1987,1540: J. Chem. SOC., Perkin Trans. 1, 1989,413. 44 D. Van Ende, W. Dumont, and A. Krief, Angew, Chem., Int. Ed. Engl., 1975,14,700; A. Krief, Tetrahedron, 1980,36,2632. 45 R.V. Bonnert and P.R.Jenkins, J. Chem. SOC., Chem. Commun., 1987,6; R.V. Bonnert, J. Howarth, P.R. Jenkins, and NJ. Lawrence, J. Chem. SOC., Perkin Trans. I , 1991,1225. 46 G. Stork and MJ. Sofia, J. Am. Chem. SOC., 1986,108, 6826. 47 R.V. Bonnert, MJ. Davies, J. Howarth, and P.R. Jenkins, J. Chem. SOC., Chern. Commun., 1990,148; R.V. Bonnert, M.J. Davies, J. Howarth, P.R. Jenkins, and N.J. Lawrence, J. Chem. SOC., Perkin Trans. 1,1992,27. 48 A.N. Boa, J. Clark, P.R. Jenkins, and N.J. Lawrence, J. Chem. SOC., Chem. Commun., 1993,15 1. 49 S. Hanessian and N.R. Plessas, J. 0%. Chem., 1969,34, 1035. 50 J.S. Yadav and R. Ravishankar, Tetrahedron Lett., 1991, 32,2629; J.S. Yadav, Pure App. Chem., 1993,65,1349, 51 S.F. Martin, J.B. White, and R. Wagner, J. 0%. Chem., 1982,47,3190; V.M. Lynch, J.-M.Assercq, A.F. Martin, and B.E. Davis, Acta Crystallogr. Sect. C., 1990, 46,2472; ibid., 1991,47,468. Hickey, and R.D. Rogers, Helv. Chim. Acta, 1992,75, 1755; L.A. Paquette, K.D. Combrink, S.W. Elmore, and M. Zhao, ibid., 1992,75,1772; S.W. Elmore, K.D. Combrink, and L.A. Paquette, Tetrahedron Lett., 199 1, 32,6679; L.A. Paquette, M. Zhao, and D. Friedrich, ibid., 1992,33,7311. 53 L.A. Paquette and M. Zhao, J. Am. Chem. SOC., 1993, 115,354. 54 T. Furukawa, K. Morihira, Y. Horiguchi, and I. Kuwajima, Tetrahedron, 1992,48,6975. 55 M. Seto, K. Morihira, S. Katagiri, T. Furukawa, Y. Horiguchi, and I. Kuwajima, Chem. Lett., 1993,133. 56 Y. Kataoka, Y. Nakamura, K. Morihira, H. Arai, Y. Horiguchi, and I. Kuwajima, Tetrahedron Lett., 1992, 33,6979. 57 K. Morihira, M.Seto, T. Furukawa, Y. Horiguchi, and I. Kuwajima, Tetrahedron Lett., 1993,34,345. 58 I. Kuwajima in ‘Organic Synthesis in Japan’, ed. R. Noyori, 1992, Tokyo Kagaku Dozin, Tokyo, page 205. 59 M. Polla and T. Frejd, Tetrahedron, 1991,47,5883. 60 (a) L. Pettersson, T. Frejd, and G. Magnusson, 52 L.A. Paquette, S.W. Elmore, K.D. Combrink, E.R. Tetrahedron Lett., 1987,28,2753; (b) L. Pettersson, G. Magnusson, and T. Frejd, Acta Chem. Scand., 1993,47, 196. 61 M. Polla and T. Frejd, Tetrahedron, 1993,49,2701. 62 S. Arseniyadas, D.V. Yashunsky, R. Pereira de Freitas, M. Munoz Dorado, E. Toromanoff, and P. Potier, Tetrahedron Lett., 1993,34,1137. 63 P.A. Wender and T.P. Mucciaro, J. Am. Chem. SOC., 1992,114,5878. 64 K.C. Nicolaou, C.-K. Hwang, EJ. Sorensen, and C.F.Clairborne, J. Chem. SOC., Chem. Commun., 1992, 1117. 65 K.C. Nicolaou, J.J. Liu, C.-K. Hwang, W.-M. Dai, and R.K. Guy, J. Chem. SOC., Chem. Commun., 1992,1118. 66 I. Alkonyi and D. Szabo, Chem. Ber., 1967,100,2773. 67 E.J. Corey and R.K. Bakshi, Tetrahedron Lett., 1990, 31,611; EJ. Corey, R.K. Bakshi, and S. Shibata, J. Am. Chem. SOC., 1987,109,5551. 68 K.C. Nicolaou, Z. Yang, EJ. Sorensen, and M. Nakada, J. Chem. SOC., Chem. Commun., 1993,1024. 69 T.V. Magee, W.G. Bornmann, R.C.A. Isaacs, and S.J. Danishefsky, J. Org. Chem., 1992,57,3274. 74 Contemporary Organic Synthesis70 C.H. Heathcock and R. Ratcliffe, J. Am. Chem. SOC., 1971,93,1746. 71 Y. Ito, T. Hirao, and T. Saegusa, J. Org. Chem., 1978, 72 Y. Queneau, W.J. Krol, W.G. Bornmann, and S. J. Danishefsky, J.0%. Chem., 1992,57,4043. 73 J. Detering and H.-D. Martin, Angew. Chem., Znt. Ed. Engl., 1988,27,695; J.R. Hargreaves,P.W.Hickmott, and B.J. Hopkins, J. Chem. SOC., C 1968,2559. 74 L.C. Vishwakarama, O.D. Stringer, and F.A. Davis, Org. Synth., 1987,88,203. 75 M. Golinski, S. Vasuden, R. Floresca, C.P. Brock, and D.S. Watt, Tetrahedron Lett., 1993,34,55; M. Golinski, C.P. Brock, and D.S. Watt, J. 0%. Chem., 1993,58, 159. 76 For recent reviews listing natural taxanes see D.G.I. Kingston, A.A. Molinero, and J.M. Rimoldi, in ‘Progress in the Chemistry of Organic Natural Products’. ed. by W. Herz, G.W. Kirby, and C. Tamm, 1993, Springer-Verlag, New York, 59, pp. 1-188. 77 G. Chauviere, D. Guinard, F, Picot, V. Senilh, and P. Potier, C. R. Acad. Sci. Ser. 2,1981,293,501. 78 F. Gueritte-Voegelein, B. David, D. Guinard, V. Sknilh, and P. Potier, Tetrahedron, 1986,42,4451; M. Colin, D. Guenard, F. Gueritte-Voegelein, and P. Potier, Eur. Pat. Appl. EP 253739A1,20 Jan., 1988; V. Sinilh, F. Guiritte, D. Guinard, M. Colin, and P. Potier, C. R. Acad. Sci. Ser. 2, 1984,299,1039. Gueritte-Voegelein, and P. Potier, Tetrahedron, 1989, 45,4177. Goff, L. Mangatal, and P. Potier, J. Med. Chem., 1991, 34,992. 8 1 D. Guenard, F. Guiritte-Voegelein, and P. Potier, Ace. Chem. Res., 1993,26,160. 82 J.N. Denis, A. E. Greene, D. Guenard, F. Gueritte-Voegelein, L. Mangatal, and P. Potier, J. Am. Chem. SOC., 1988,110,591 7; M. Colin, D. Guenard, F. Gueritte-Voegelein, and P. Potier, Eur. Pat. Appl. EP 253738A1,20 Jan., 1988; J.-N. Denis,A. Greene, D. Guinard, and F. Guiritte-Voegelein, Eur. Pat. Appl. EP 336840A1,5 April, 1989. 83 I. Ojima, I. Habus, M. Zhao, G. I. Georg, and L.R. Jayasinghe, J. Org. Chem., 1991,56,1681. 84 R.A. Holton, Eur. Pat. Appl. EP 40097 1Al,30 May, 1990, Chem. Abstr., 114, P164568q. 85 I. Ojima, C.M. Sun, M. Zucco, Y.H. Park, 0. Ducios, and S. Kuduk, Tetrahedron Lett., 1993,34,4149. 86 R.A. Holton, US Patent Appl. US 5 015 744, Nov. 14, 1989; R.A. Holton, Eur, Pat. Appl., EP 428376A1, 13 Nov, 1990, Chem. Abstr., 1 15, P 1 148 17v. 87 C.S. Swindell, N.E. Krauss, S.B. Horwitz, and I. Ringel, J. Med. Chem., 1991,34,1176. 43,1011. 79 L. Mangatal, M.T. Adeline, D. Guinard, F. 80 F. Guiritte-Voegelein, D. Guinard, F. Lavelle, M.-T. Le 88 J.N. Denis,A.E. Greene, A.A. Serra, and M.-J. Luche, J. Org. Chem., 1986,5 1,46; Analogues: A. Chatterjee, J.S. Williamson, J.K. Zjawiony, and J.R. Peterson, Biorg. Med. Chem. Lett., 1992,2,91. 89 J.N. Denis, A. Correa, and A.E. Greene, J. Org. Chem., 1990,55,1957. 90 K.B. Sharpless, W. Amberg, Y.L. Bennani, G.A. Crispino, J. Hartung, K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-M. Wang, D. Xu, and X.-L. Zhang, J. 0%. Chem., 1992,57,2768. 91 A. CommerCon, D. Bizard, F. Bernard, and J.D. Bourzat, Tetrahedron Lett., 1992,33,5185. 92 L. Deng and E.N. Jacobsen, J. Org. Chem., 1992,57, 4320. 93 E.N. Jacobsen, W. Zhang, A.R. Muci, J.R. Ecker, and L. Deng,J. Am. Chem. SOC., 1991,113,7063. 94 V. Farina, S.I. Hauck, and D.G. Walker, Synlett., 1992, 761. 95 G.I. Georg, P.M. Mashava, E. Akgun, and M.W. Milstead, Tetrahedron Lett., 1991,32,3151. In this paper the diastereoselectivity was incorrectly judged to be 100 : 0-see corrigendum, Tetrahedron Lett., 1992, 33,1524. 96 G.I. Georg, E. Akgiin, P.M. Mashava, M.W. Milstead, H. Ping, Z. Wu, and D. Vander Velde, Tetrahedron Lett., 1992,33,2111. 97 R.A. Holton, EP 90305845.1, May 1990; Chem Abstr., 114, P164568q. 98 I. Ojima, I. Habus, M. Zucco, Y.H. Park, C.M. Sun, and T. Brigaud, Tetrahedron, 1992,48,6985; Analogues: G.I. Georg, Z.S. Cheruvallath, R.H. Himes, and M.R. Mejillano, Biorg. Med. Chem. Lett., 1992,2,295. 99 C. Palomo, A. Arrieta, F.P. Cossio, J.M. Aizpurua, A. Mielgo, and N. Aurrekoetxea, Tetrahedron Lett., 1990, 31,6429. Rubiales, and C. Palomo, Tetrahedron Lett., 1988,28, 3133. 101 A. Correa, J.N. Denis, M.-J. Luche, and A.E. Greene, J. 0%. Chem., 1993,58,255. 102 M.E. Bunnage, S.G. Davies, and C.J. Goodwin, J. Chem. SOC., Perkin Trans. 1, 1993, 1375. 103 H. Honig, P. Seufer-Wasserthal, and H. Weber, Tetrahedron, 1990,46,3841. 104 D.-M. Gou, Y.-C. Liu, and C.-S. Chen, J. 0%. Chem., 1993,58,1287. 105 R. Breiva, J.Z. Crich, and CJ. Sih, J. Org. Chem., 1993, 58,1068. 106 C. Mukai, I.J. Kim, and M. Hanoaka, Tetrahedron Asymmetry, 1992,3,1007. 107 K. Hattori, M. Miyata, and H. Yamamoto, J. Am. Chem. SOC., 1993,115,1151. 108 J.N. Denis, A. Correa, and A.E. Greene, J. Org. Chem., 1991,56,6939. 100 F.P. Cossio, C. Lbpez, M. Oiarbide, D. Aparicio, G. Boa, Jenkins, and Lawrence: Recent progress in the synthesis of taxanes 75