The Holton Taxol total synthesis is a good example of a linear synthesis. The synthesis starts from patchoulene oxide, a commercially available natural compound .[3]
This epoxide can be obtained in two steps from the terpene patchoulol and also from borneol.[4][5] The reaction sequence is also enantioselective, synthesizing (+)-Taxol from (−)-patchoulene oxide or (−)-Taxol from (−)-borneol with a reported specific rotation of +- 47° (c=0.19 / MeOH). The Holton sequence to Taxol is relatively short compared to that of the other groups (46 linear steps from patchoulene oxide). One of the reasons is that patchoulene oxide already contains 15 of the 20 carbon atoms required for the Taxol ABCD ring framework.
It was envisaged that Taxol (51) could be accessed through tail addition of the Ojima lactam48 to alcohol 47. Of the four rings of Taxol, the D ring was formed last, the result of a simple intramolecular SN2 reaction of hydroxytosylate 38, which could be synthesized from hydroxyketone 27. Formation of the six-membered C ring took place through a Dieckmann condensation of lactone 23, which could be obtained through a Chan rearrangement of carbonate ester15. Substrate 15 could be derived from ketone 6, which, after several oxidations and rearrangements, could be furnished from commercially available patchoulene oxide 1.
As shown in Scheme 2, the next phase involved addition of the carbon atoms required for the formation of the C ring. Ketone 7 was treated with magnesium bromide diisopropylamide and underwent an aldol reaction with 4-pentanal (8) to give β-hydroxyketone 9. The hydroxyl group was protected as the asymmetric carbonate ester(10). Oxidation of the enolate of ketone 10 with (-)-camphorsulfonyl oxaziridine (11) gave α-hydroxyketone 12. Reduction of the ketone group with 20 equivalents of sodium bis(2-methoxyethoxy)aluminumhydride (Red-Al) gave triol 13, which was immediately converted to carbonate 14 by treatment with phosgene. Swern oxidation of alcohol14 gave ketone15. The next step set the final carbon-carbon bond between the B and C rings. This was achieved through a Chan rearrangement of 15 using lithium tetramethylpiperidide to give α-hydroxylactone 16 in 90% yield. The hydroxyl group was reductively removed using samarium(II) iodide to give an enol, and chromatography of this enol on silica gel gave the separable diastereomers cis 17c (77%) and trans 17t (15%), which could be recycled to 17c through treatment with potassium tert-butoxide. Treatment of pure 17c with lithium tetramethylpiperidide and (±)-camphorsulfonyl oxaziridine gave separable α-hydroxyketones 18c (88%) and 18t (8%) in addition to some recovered starting material (3%). Reduction of pure ketone 18c using Red-Al followed by basic work-up resulted in epimerization to give the required trans-fused diol19 in 88% yield.
C ring synthesis
As shown in Scheme 3, diol19 was protected with phosgene as a carbonate ester (20). The terminal alkene group of 20 was next converted to a methyl ester using ozonolysis followed by oxidation with potassium permanganate and esterification with diazomethane. Ring expansion to give the cyclohexane C ring 24 was achieved using a Dieckman condensation of lactone 23 with lithium diisopropylamide as a base at -78 °C. Decarboxylation of 24 required protection of the hydroxyl group as the 2-methoxy-2-propyl (MOP) ether (25). With the protecting group in place, decarboxylation was effected with potassium thiophenolate in dimethylformamide to give protected hydroxy ketone 26. In the next two steps the MOP protecting group was removed under acidic conditions, and alcohol 27 was reprotected as the more robust benzyloxymethyl ether28. The ketone was converted to the trimethylsilyl enol ether 29, which was subsequently oxidized in a Rubottom oxidation using m-chloroperbezoic acid to give the trimethylsilyl protected acyloin30. At this stage the final missing carbon atom in the Taxol ring framework was introduced in a Grignard reaction of ketone 30 using a 10-fold excess of methylmagnesium bromide to give tertiary alcohol 31. Treatment of this tertiary alcohol with the Burgess reagent (32) gave exocyclic alkene 33.
In the final stages of the synthesis (Scheme 5), the hydroxyl group in 46 was deprotected to give alcohol 47. Reaction of the lithium alkoxide of 47 with the Ojima lactam48 adds the tail in 49. Deprotection of the triethylsilyl ether with hydrofluoric acid and removal of the BOM group under reductive conditions gave (−)-Taxol 51 in 46 steps.
The secondary alcohol in the 4-pentenal product of the aldol reaction, 9, was protected as an asymmetric carbonate ester. This group was removed in conjunction with the Red-Al reduction of ketone 12 (Scheme 2).
Carbonate (cyclic) [1]
phosgene, pyridine, dichloromethane, -78 °C to room temperature, 1 h
deprotected through Chan rearrangement (treatment with lithiumtetramethylpiperidide)
The cyclic carbonate ester was removed as a result of the Chan rearrangement in 15, which created a carbon-carbon bond that was part of the Taxol framework (Scheme 2).
Diol 19 (Scheme 3) was protected as a cyclic carbonate ester. This carbonate ester was cleaved by phenyllithium in tetrahydrofuran at -78 °C to give hydroxybenzoate 42 (Scheme 4).
After Grob fragmentation (Scheme 1), the resultant alcohol 6 was protected as a TBS ether 7, which is kept in place until the final addition of the tail (Scheme 5).
hydrogen fluoride/pyridine complex in acetonitrile
The secondary hydroxyl group in diol 4 (Scheme 1) was protected as a TES ether in order to prevent its participation in the Grob fragmentation. The TES was cleaved in 37 (Scheme 4) and returned to the alcohol.
Ketone 25 (Scheme 3) was protected as the TMS enol ether and subsequently was oxidized with m-chloroperoxybenzoic acid. In the process the TMS group migrated to the 2-hydroxyl group.
The primary hydroxyl group in triol 35 (Scheme 4) was protected as a TMS ether allowing activation of the secondary hydroxyl group as a tosylate leaving group.
^Robert A. Holton; Carmen Somoza; Hyeong Baik Kim; Feng Liang; Ronald J. Biediger; P. Douglas Boatman; Mitsuru Shindo; Chase C. Smith; Soekchan Kim; Hossain Nadizadeh; Yukio Suzuki; Chunlin Tao; Phong Vu; Suhan Tang; Pingsheng Zhang; Krishna K. Murthi; Lisa N. Gentile; Jyanwei H. Liu (1994). "First total synthesis of taxol. 1. Functionalization of the B ring". J. Am. Chem. Soc. 116 (4): 1597–1598. doi:10.1021/ja00083a066.
^Robert A. Holton; Hyeong-Baik Kim; Carmen Somoza; Feng Liang; Ronald J. Biediger; P. Douglas Boatman; Mitsuru Shindo; Chase C. Smith; Soekchan Kim; Hossain Nadizadeh; Yukio Suzuki; Chunlin Tao; Phong Vu; Suhan Tang; Pingsheng Zhang; Krishna K. Murthi; Lisa N. Gentile; Jyanwei H. Liu (1994). "First Total Synthesis of Taxol. 2. Completion of the C and D Rings". J. Am. Chem. Soc. 116 (4): 1599–1600. doi:10.1021/ja00083a067.
^Robert A. Holton; R. R. Juo; Hyeong B. Kim; Andrew D. Williams; Shinya. Harusawa; Richard E. Lowenthal; Sadamu. Yogai (1988). "A synthesis of taxusin". J. Am. Chem. Soc. 110 (19): 6558–6560. doi:10.1021/ja00227a043.
^Buchi, G.; MacLeod, William D.; Padilla, J. (1964-10-01). "Terpenes. XIX.1 Synthesis of Patchouli Alcohol2". Journal of the American Chemical Society. 86 (20): 4438–4444. doi:10.1021/ja01074a041. ISSN0002-7863.
^Büchi, G.; Erickson, R. E.; Wakabayashi, Nobel (1961-02-01). "Terpenes. XVI.1,2 Constitution of Patchouli Alcohol and Absolute Configuration of Cedrene". Journal of the American Chemical Society. 83 (4): 927–938. doi:10.1021/ja01465a042. ISSN0002-7863.