Total Synthesis of Epothilones A and B

J. Am. Chem. Soc. 1997, 119, 10073-10092

10073

Total Syntheses of Epothilones A and B

Dongfang Meng,^†,‡Peter Bertinato,^†Aaron Balog,^†Dai-Shi Su,^†Ted Kamenecka,^†

Erik J. Sorensen,^†and Samuel J. Danishefsky*^,†,‡

Contribution from The Laboratory for Bioorganic Chemistry, The Sloan-Kettering Institute for

Cancer Research, 1275 York AVenue, New York, New York 10021, and The Department of

Chemistry, Columbia UniVersity, HaVemeyer Hall, New York, New York 10027

ReceiVed June 12, 1997^X

Abstract: Convergent, stereocontrolled total syntheses of the microtubule-stabilizing macrolides epothilones A (2)

and B (3) have been achieved. Four distinct ring-forming strategies were pursued (see Scheme 1). Of these four,

three were reduced to practice. In one approach, the action of a base on a substance possessing an acetate ester and

a nonenolizable aldehyde brought about a remarkably effective macroaldolization see (89 f 90 + 91; 99 f 100 +

101), simultaneously creating the C2-C3 bond and the hydroxyl-bearing stereocenter at C-3. Alternatively, the

16-membered macrolide of the epothilones could be fashioned through a C12-C13 ring-closing olefin metathesis

(e.g. see 111 f 90 + 117; 122 f 105 + 123) and through macrolactonization of the appropriate hydroxy acid (e.g.

see 88 f 93). The application of a stereospecific B-alkyl Suzuki coupling strategy permitted the establishment of

a cis C12-C13 olefin, thus setting the stage for an eventual site- and diastereoselective epoxidation reaction (see 96

f 2; 106 f 3). The development of a novel cyclopropane solvolysis strategy for incorporating the geminal methyl

groups of the epothilones (see 39 f 40 f 41), and the use of Lewis acid catalyzed diene-aldehyde cyclocondensation

(LACDAC) (see 35 + 36 f 37) and asymmetric allylation (see 10 f 76) methodology are also noteworthy.

The introduction of taxol (paclitaxel) (1) into cancer chemo-

therapy is testimony to the synergism of broadly based contribu-

tions from many areas of scientific expertise en route to the

clinic. The original isolation and structure work, which also

served to identify the cytotoxicity of the drug, was accomplished

by Wall and co-workers.¹In a seminal paper, Horwitz identified

the in Vitro mode of action of taxol, demonstrating its ability to

stabilize microtubule assemblies.²This finding gave impetus

to a wider range of pharmacological investigations of critical

importance. The development of improved methods from

phytochemical sources for obtaining baccatin III, and improved

chemical methods for converting baccatin III to paclitaxel,

Figure 1. Structures of taxol (1) and epothilones A (2) and B (3).

provided the drug in ample quantities for human trials.

the fact that paclitaxel is subject to a significant attenuation of

On the basis of favorable findings that issued from these

therapeutic value through the onset of multiple drug resistance

evaluations, paclitaxel, developed by the Bristol Myers-Squibb

(MDR). Although there has been extensive structure-activity

Co., was approved for chemotherapeutic application against

work in the paclitaxel area, to our knowledge the only modified

ovarian carcinomas. Since then, this drug has been undergoing

compound presently being evaluated in human clinical trials is

extensive evaluations for other indications and is being incor-

the close relative, taxotere.⁴

porated in a variety of clinical contexts. Though it is often not

Given the high interest engendered by taxoids as a conse-

a curative agent, paclitaxel is emerging as a useful main line

quence of their clinical usefulness, the search for new agents

chemotherapeutic resource. There being no evidence to the

that function by a comparable mechanism is of great interest.

contrary, it is assumed that the in ViVo mode of action and

It is in this connection that the recently discovered bacterial

antitumor properties of paclitaxel arise from inhibition of cellular

metabolites epothilone A (2) and B (3) have attracted consider-

mitosis through the Horwitz mechanism. The question of

able attention. These compounds were first identified as

whether or not this mode of action is actually operative in the

antifungal cytotoxic agents by Ho¨fle and co-workers.⁵During

human patient is not easily addressed.

the course of a screening program aimed at the identification

Although having many advantages, paclitaxel is not an ideal

of substances with a paclitaxel-like mode of action, a group

drug.³One major problem with this agent, useful as it is, has

based at Merck found that the epothilones are powerful cytotoxic

to do with difficulties in its formulation. Paclitaxel is a rather

agents which function through stabilization of cellular micro-

insoluble substance in water, thereby necessitating awkward

tubules.⁶The strong implication of the Merck research effort

forms of clinical administration. Perhaps even more serious is

was that the epothilones share in the paclitaxel mode of action.

^†Sloan-Kettering.

^‡Columbia University.

(4) Gueritte-Voegelein, F.; Guenard, D.; Dubois, J.; Wahl, A.; Marder,

R.; Muller, R.; Lund, M.; Bricard, L.; Potier, P. In Taxane Anticancer

Agents; Georg, G. I., Chen, T. T., Ojima, I., Vyas, D. M., Eds.; American

Cancer Society: San Diego, 1995; Chapter 14, p 189.

^XAbstract published in AdVance ACS Abstracts, October 1, 1997.

(1) Wani, M. C.; Taylor, H. L.; Wall, M. E.; Coggon, P.; McPhail, A.

T. J. Am. Chem. Soc. 1971, 93, 2325.

(2) Horwitz, S. B.; Fant, J.; Schiff, P. B. Nature 1979, 277, 665.

(3) Landino, L. M.; MacDonald, T. L. In The Chemistry and Pharmacol-

ogy of Taxol^®and its DeriVatiVes; Farin, V., Ed.; Elsevier: New York,

1995; Chapter 7, p 301.

(5) (a) Gerth, K.; Bedorf, N.; Ho¨fle, G.; Irschik, H.; Reichenbach, H. J.

Antibiot. 1996, 49, 560. (b) Ho¨fle, G.; Bedorf, N.; Steinmetz, H.; Schomburg,

D.; Gerth, K.; Reichenbach, H. Angew. Chem., Int. Ed. Engl. 1996, 35,

1567.

10074 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Indeed, these compounds seem to adhere to the “taxol binding

domains” of the microtubule assemblies. In light of the

possibility that the epothilones, or suitably modified derivatives,

might find a role in cancer chemotherapy, this series also merits

multidisciplinary attention. Among the scientific enterprises

which we felt to be warranted in the case of the epothilones

would be research directed to their total synthesis.

A^8cand epothilone B.^8eIn a concurrent time frame, studies

from other laboratories accomplished these goals.^9,10Moreover,

the field of epothilone synthesis has spawned a host of

interesting disclosures of potential impact on the long term total

synthesis goal.¹¹

Overall Synthetic Strategies

Unlike the situation with paclitaxel, where it was clear from

the outset that total synthesis would be unlikely to impact upon

the actual availability of the drug itself,⁷the simpler structures

of the epothilones invited the hope that chemical synthesis could

improve accessibility to the desired agents. Thus, organic

chemistry could contribute to the development of an epothilone-

based drug effort through a highly efficient total synthesis or

possibly by delivering much simpler structures that still manifest

chemotherapeutically useful biological activity.

In pursuit of this program, a variety of initiatives were

considered for assembling epothilone systems, including the

natural products themselves. It is well to identify the main

themes that were followed, often in parallel. In the earliest

phase, our thinking was much influenced by the presence of

the achiral domain encompassing carbons 9, 10, and 11. As

noted above, this achiral region insulates the chiral “O-alkyl

and acyl” sectors of the molecule from one another. We felt

that the presence of this spacer element would pose a consider-

able difficulty in communicating stereochemical information

from one chiral locus to the other. Rather, it seemed appropriate

to build the two chiral domains independently and to join them

through carbon-carbon bond formation somewhere in the C9-

C11 sector. In this way, the integrity of chiral centers at C8

and C12, that are terminal to their respective chiral enclaves,

would not be placed at risk in the crucial “merger” phase. One

obvious possibility which presented itself in this regard was

that of ring-forming olefin metathesis^12-14(see approach I,

Scheme 1). In this connection we were mindful of a seminal

precedent disclosed by Hoveyda et al. in the context of

synthesizing a macrolactam. Indeed, other laboratories, active

in the epothilone field recognized, as did we, the potential

pertinence of olefin metathesis to the epothilone area.¹³

Moreover, several intrinsically challenging chemical issues

require attention in any total synthesis venture aimed at the

epothilones. Quickly recognized in the case of epothilone A is

the presence of a thiazole moiety, a cis-epoxide (C12-C13)

and, somewhat unusual for a macrolide, the presence of geminal

methyl groups at C4. Not the least noteworthy feature of the

synthesis problem is inherent in the array of three contiguous

methylene groups which serves to insulate the two domains of

the epothilones that bear stereochemical imprints. The acyl

section, numbered from carbons 1-8, presents a constellation

of four chiral centers whose proper emplacement would require

careful management. An agenda dealing with the synthesis of

this domain must also include programs for elaborating and

maintaining a potentially unstable â-hydroxy ester linkage at

C3. The oxidation state at C3 must be cleanly differentiated

from that at C5 where a ketonic group is to emerge. This entire

polypropionate sector is insulated by carbons 9, 10, and 11 from

the chiral O-alkyl domain comprising carbons 12-15. The

already mentioned cis-epoxide, connecting carbons 12 and 13

(disubstituted in the case of epothilone A and trisubstituted in

the case of epothilone B), is insulated by a single methylene

group from carbon 15, which bears an allylic alcohol and a

thiazole-based version of an R-methyl styryl linkage.

(8) The work described herein is based on the following disclosures from

our laboratory: (a) Meng, D.; Sorensen, E. J.; Bertinato, P.; Danishefsky,

S. J. J. Org. Chem. 1996, 61, 7998. (b) Bertinato, P.; Sorensen, E. J.; Meng,

D.; Danishefsky, S. J. J. Org. Chem. 1996, 61, 8000. (c) Balog, A.; Meng,

D.; Kamenecka, T.; Bertinato, P.; Su, D.-S.; Sorensen, E. J.; Danishefsky,

S. J. Angew. Chem., Int. Ed. Engl. 1996, 35, 2801. (d) Meng, D.; Su. D.-

S.; Balog, A.; Bertinato, P.; Sorensen, E. J.; Danishefsky, S. J.; Zheng,

Y.-H.; Chou, T.-C.; He, L.; Horwitz, S. B. J. Am. Chem. Soc. 1997, 119,

2733. (e) Su, D.-S.; Meng, D.; Bertinato, P.; Balog, A.; Sorensen, E. J.;

Danishefsky, S. J.; Zheng, Y.-H.; Chou, T.-C.; He, L.; Horwitz, S. B. Angew.

Chem., Int. Ed. Engl. 1997, 36, 757. (f) Balog, A.; Bertinato, P.; Su, D.-S.;

Meng, D.; Sorensen, E. J.; Danishefsky, S. J.; Zheng, Y.-H.; Chou, T.-C.;

He, L.; Horwitz, S. B. Tetrahedron Lett. 1997, 26, 4529.

(9) (a) Nicolaou, K. C.; He, Y.; Vourloumis, D.; Vallberg, H.; Yang, Z.

Angew. Chem., Int. Ed. Engl. 1996, 35, 2399. (b) Yang, Z.; He, Y.;

Vourloumis, D.; Vallberg, H.; Nicolaou, K. C. Angew. Chem., Int. Ed. Engl.

1997, 36, 166. (c) Nicolaou, K. C.; Sarabia, F.; Ninkovic, S.; Yang, Z.

Angew. Chem., Int. Ed. Engl. 1997, 36, 525. (d) Nicolaou, K. C.; Winssinger,

N.; Pastor, J.; Ninkovic, S.; Sarabia, F.; He, Y.; Vourloumis, D.; Yang, Z.;

Li, T.; Glannakakou, P.; Hamel, E. Nature 1997, 387, 268.

None of these issues in isolation pose an insurmountable

obstacle to the capabilities of contemporary organic chemistry.

However, taken together, they constitute a significant challenge

to the goal of a stereocontrolled total synthesis of the epothilones.

Below, we provide a summary of our activities that eventually

accomplished this goal⁸for the first time for both epothilone

(6) Bollag, D. M.; McQueney, P. A.; Zhu, J.; Hensens, O.; Koupal, L.;

Liesch, J.; Goetz, M.; Lazarides, E.; Woods, C. M. Cancer Res. 1995, 55,

2325.

(10) (a) Schinzer, D.; Limberg, A.; Bo¨hm, O. M. Chem. Eur. J. 1996, 2,

1477. (b) Schinzer, D.; Limberg, A.; Bauer, A.; Bo¨hm, O. M.; Cordes, M.

Angew. Chem., Int. Ed. Engl. 1997, 36, 523.

(7) For total syntheses of paclitaxel (taxol^TM), see: (a) Nicolaou, K. C.;

Yang, Z.; Liu, J. J.; Ueno, H.; Nantermet, P. G.; Guy, R. K.; Claiborne, C.

F.; Renaud, J.; Couladouros, E. A.; Paulvannan, K.; Sorensen, E. J. Nature

(London) 1994, 367, 630. (b) Holton, R. A.; Somoza, C.; Kim, H.-B.; Liang,

F.; Biediger, R. J.; Boatman, P. D.; Shindo, M.; Smith, C. C.; Kim, S.;

Nadizadeh, H.; Suzuki, Y.; Tao, C.; Vu, P.; Tang, S.; Zhang, P.; Murthi,

K. K.; Gentile, L. N.; Liu, J. H. J. Am. Chem. Soc. 1994, 116, 1597. (c)

Holton, R. A.; Kim, H.-B.; Somoza, C.; Liang, F.; Biediger, R. J.; Boatman,

P. D.; Shindo, M.; Smith, C. C.; Kim, S.; Nadizadeh, H.; Suzuki, Y.; Tao,

C.; Vu, P.; Tang, S.; Zhang, P.; Murthi, K. K.; Gentile, L. N.; Liu, J. H. J.

Am. Chem. Soc. 1994, 116, 1599. (d) Masters, J. J.; Link, J. T.; Snyder, L.

B.; Young, W. B.; Danishefsky, S. J. Angew. Chem., Int. Ed. Engl. 1995,

34, 1723. (e) Danishefsky, S. J.; Masters, J. J.; Link, J. T.; Young, W. B.;

Snyder, L. B.; Magee, T. V.; Jung, D. K.; Isaacs, R. C. A.; Bornmann, W.

G.; Alaimo, C. A.; Coburn, C. A.; DiGrandi, M. J. J. Am. Chem. Soc. 1996,

118, 2843. (f) Wender, P. A.; Badham, N. F.; Conway, S. P.; Floreancig,

P. E.; Glass, T. E.; Gra¨nicher, C.; Houze, J. B.; Ja¨nichen, J.; Lee, D.;

Marquess, D. G.; McGrane, P. L.; Meng, W.; Mucciaro, T. P.; Mu¨hlebach,

M.; Natchus, M. G.; Paulsen, H.; Rawlins, D. B.; Satkofsky, J.; Shuker, A.

J.; Sutton, J. C.; Taylor, R. E.; Tomooka, K. J. Am. Chem. Soc. 1997, 119,

2755. (g) Wender, P. A.; Badham, N. F.; Conway, S. P.; Floreancig, P. E.;

Glass, T. E.; Houze, J. B.; Krauss, N. E.; Lee, D.; Marquess, D. G.;

McGrane, P. L.; Meng, W.; Natchus, M. G.; Shuker, A. J.; Sutton, J. C.;

Taylor, R. E. J. Am. Chem. Soc. 1997, 119, 2757.

(11) For studies toward a total synthesis of the epothilones, see: (a)

Mulzer, J.; Mantoulidis, A. Tetrahedron Lett. 1996, 37, 9179. (b) Claus,

E.; Pahl, A.; Jones, P. G.; Meyer, H. M.; Kalesse, M. Tetrahedron Lett.

1997, 38, 1359. (c) Gabriel, T.; Wessjohann, L. Tetrahedron Lett. 1997,

38, 1363. (d) Taylor, R. E.; Haley, J. D. Tetrahedron Lett. 1997, 38, 2061.

(e) De Brabander, J.; Rosset, S.; Bernardinelli, G. Synlett 1997 (July), 824.

(12) (a) Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.;

DiMare, M.; O’Regan, M. J. Am. Chem. Soc. 1990, 112, 3875. (b) Schwab,

P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem., Int. Ed. Engl.

1995, 34, 2039. For reviews of ring-closing olefin metathesis, see: (c)

Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28, 446. (d)

Schmalz, H.-G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1833.

(13) For an outstanding demonstration of ring-closing olefin metathesis

in the context of a natural product total synthesis, see: Houri, A. F.; Xu,

Z.; Cogan, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1995, 117, 2943. For

the first published application on a model basis of the olefin metathesis

concepts to the epothilone problem see refs 9a and 8b disclosed in that

order. For the disclosures of the total synthesis of epothilone A based on

olefin metathesis see refs 9b and 8d disclosed in that order.

(14) For recent examples of ring-closing olefin metathesis, see: (a)

Martin, S. F.; Chen, H.-J.; Courtney, A. K.; Liao, Y.; Pa¨tzel, M.; Ramser,

M. N.; Wagman, A. S. Tetrahedron 1996, 52, 7251. (b) Fu¨rstner, A.;

Langemann, K. J. Org. Chem. 1996, 61, 3942.

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10075

Scheme 1. Bond Construction Strategies for Total Synthesis of Epothilone A (2) and B (3)

Refining the matter still further, we directed our attentions

to a construction in which the olefin metathesis bond would

join carbons 9 and 10 (see 4). An alternative approach wherein

carbons 10 and 11 would be joined seemed riskier. The latter

construction would have involved substrates in which there was

an allylic relationship between the epoxide and the C10-C11

unsaturation in both the starting material and the product.

Accordingly, the olefin metathesis prospectus that we came to

favor called for a precursor of the general type 4 (P )

unspecified protecting group). The resultant olefin would

comprise carbons 9 and 10 of the goal system. Reduction of

the olefin, followed by appropriate functional group manipula-

tions, would then lead to target systems 2 and 3.

As will be shown, some surprising limitations in the ring-

forming olefin metathesis reaction surfaced as we attempted to

reduce this line of thinking to practice. When the full

dimensions of the obstacles associated with a C9-C10 bond

construction through ring-closing olefin metathesis were re-

vealed, we came to focus on a fundamentally different assembly

strategy wherein a double bond would be established between

carbons 12 and 13 through ring-closing olefin metathesis. In

this prospectus, the thiazole bearing chiral domain would display

sp³asymmetry only at C-15. Carbons 12 and 13 would first

be presented in the form of a cis-olefin, hopefully en route to

a properly configured epoxide.

We also had occasion to contemplate an alternative synthetic

logic, i.e. that of cross coupling, wherein a bond would be

fashioned between future carbons 11 and 12. Specifically, we

came to favor a B-alkyl Suzuki motif¹⁵to achieve this goal. In

this line of reasoning, the fragments entering into this Suzuki

coupling would be implied by a structure of the type 5. From

such a seco-compound, the 16-membered macrolide ring could

be established through an intramolecular aldol addition,¹⁶giving

rise to the C2-C3 bond (approach II). Alternatively, one could

envision a post-Suzuki coupling structure of the type 6 which

could set the stage for macrolide construction through macro-

lactonization¹⁷(approach III).

As will eventually be shown, the double bond at C-12 and

C-13, either in the di- or trisubstituted series, could be very

(15) (a) Miyaura, N.; Ishiyama, T.; Sasaki, H.; Ishi-Kawa, M.; Satoh,

M.; Suzuki, A. J. Am. Chem. Soc. 1989, 111, 314. (b) For an excellent

review, see: Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (c)

Johnson, C. R.; Braun, M. P. J. Am. Chem. Soc. 1993, 115, 11014. (d)

Ohba, M.; Kawase, N.; Fujii, T.; Aoe, K.; Okamura, K.; Fathi-Afshar, R.;

Allen, T. M. Tetrahedron Lett. 1995, 36, 6101.

(16) For a prior instance of a keto aldehyde macroaldolization, see:

Hayward, C. M.; Yohannes, D.; Danishefsky, S. J. J. Am. Chem. Soc. 1993,

115, 9345.

(17) For a review of methods for constructing macrolides, see: Paterson,

I.; Mansuri, M. M. Tetrahedron 1985, 41, 3569. The first macrolactonization

in this series was demonstrated (though unknown to us at the time of our

experiments) by Nicolaou and associates (reference 9c).

10076 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

effectively exploited for introduction of the â-epoxide required

for the final targets. As this feasibility was established, we also

directed efforts to establishment of the C12-C13 bond through

a ring-closing olefin metathesis reaction. An assessment of this

interesting possibility would require the construction of a

diolefin of the general type 7 (see approach IV, Scheme 1).

This subgoal was accomplished. We now proceed to describe

the progress attained in pursuing each of these strategies.

Scheme 2^a

The First Generation Ring-Closing Olefin Metathesis

Strategy (Approach I, Scheme 1)

As was alluded to above, several options for constructing the

16-membered macrolide ring of the epothilones were considered

as we contemplated strategies for a total synthesis. Initially,

we favored a strategy wherein the C9-C10 would be fashioned

during the course of the macrocyclization event. It was our

hope that a structure of the general type 4 (see Scheme 1) could

be induced to undergo a ring-closing olefin methathesis (RCM).

Our considerations for favoring such an approach were several.

In synthesizing and merging the subunits leading to 4, all of

the stereochemical problems associated with an epothilone total

synthesis would have been overcome. The unsaturation at C9-

C10 that would emerge from a successful ring-closing olefin

metathesis could conceivably be removed by reduction en route

to the epothilones, or used to introduce new functionality for

the purpose of synthesizing novel analog structures.

As will be shown (Vide infra), the guiding paradigm, i.e. the

possibility of creating a C9-C10 double bond during the course

of an intramolecular RCM process, was not reducible to practice

in a subststrate having adequate functionality to reach the natural

products. Nonetheless, many of our perceptions pertinent to

the epothilone stereochemical problem and, indeed, some of the

very compounds used in the pursuit of approach I, did find

application to variations for the successful total syntheses.

Hence, we describe here the findings pertinent to approach I,

Scheme 1.

We defined goal system 20 (see Scheme 2) as a milestone

compound under the approach I program. A structure of this

genre would be joined to an acyl fragment (Vide infra) to

establish a precursor of the type hitherto generalized as 4. Our

path commenced with the known aldehyde 8,¹⁸which was

elongated in a Wittig-type construction with the commercially

available phosphorane 9, leading to 10 in 83% yield (see Scheme

2).^8aAt this stage, it was of interest to us to take advantage of

a line of chemistry that our laboratory had innovated in the

1980s.¹⁹Thus, aldehyde 10 served as a “heterodienophile” in

the context of a Lewis acid catalyzed diene-aldehyde cyclo-

condensation (LACDAC) reaction with the synergistic butadiene

11.²⁰The reaction proceeded quite smoothly, giving rise to the

racemic dihydropyrone 12 in a yield of 65%.

^a(a) C₆H₆, reflux (83%); (b) trans-1-methoxy-3-((trimethylsilyl)oxy)-

1,3-butadiene (11), BF₃‚OEt₂, CH₂Cl₂; then CSA (65%); (c) NaBH₄,

CeCl₃‚7H₂O, MeOH, 0 °C f rt (99%); (d) Lipase-30, vinyl acetate,

DME, rt, (-)-15 (45%; 93% ee); (e) (i) K₂CO₃, MeOH, rt; (ii) PMBCl,

NaH, DMF, 0 °C f rt (97% overall); (f) 3,3-dimethyldioxirane, K₂CO₃,

CH₂Cl₂, 0 °C; then NaIO₄, H₂O/THF (92%); (g) allyl triphenylstannane,

SnCl₄, CH₂Cl₂, -78 °C (98% of 18 + epimer (4:1)); (h) (i) MsCl,

Et₃N, CH₂Cl₂, 0 °C; (ii) DDQ, CH₂Cl₂/H₂O (20:1), 0 °C f rt (93%

overall); (i) (i) LiN(SiMe₃)₂, THF, -78 f 0 °C; (ii) K₂CO₃, MeOH/

H₂O (78% of the cis epoxide); PMB ) p-MeOC₆H₄CH₂; Ms ) SO₂CH₃.

through acetylation in the “forward sense”. Following protocols

of Wong,²³the racemic alcohol was used in a transesterification

experiment with vinyl acetate under the influence of Lipase-30

to afford alcohol 14 and acetate 15. Of course, at this stage,

we were in no position to assert the assignments of the absolute

configuration to the antipodal glycal and glycal acetates with

rigorous confidence. Rather, our tentative formulations arose

from extensive precedents that had been garnered in our

laboratory some years ago in this general area.²²The presumed

3S-acetate 15 was subjected to deacetylation, giving rise to ent-

14. The latter was subjected to the action of sodium hydride

and p-methoxybenzyl chloride to afford 16. In principle, it was

initially supposed that compound 14 in the 3R series could be

utilized in the synthesis program. However, as will be shown,

an alternate route not requiring resolution to reach the desired

3S pyranoid series (cf. 16) became available, and the possibility

of recycling the 3R series available from the lipase chemistry

was not pursued.

Reduction of compound 12 via conditions that we had

introduced some years ago for synthesizing artificial glycals

bearing equatorial alcohols at C3 (glucose numbering),²¹led to

racemic 13. Here we were able to take advantage of more

recently introduced methodology, wherein racemic glycals,

derived by total synthesis rather than from carbohydrate sources,

could be effectively resolved by lipase-mediated kinetic resolu-

tion.²²In the event, we chose to carry out a kinetic resolution

The next phase of the program called for disconnection of

the C1-C2 bond of the artificial glycal 16 in such a fashion

that C3 would emerge as C13 of the projected cis-epoxide 20.

We proceeded as follows. Drawing once again on chemistry

that we had introduced in an earlier era for different purposes,²⁴

(18) (a) Shafiee, A.; Mazloumi, A.; Cohen, V. I. J. Heterocycl. Chem.

1979, 16, 1563. (b) Schafiee, A.; Shahocini, S. J. Heterocycl. Chem. 1989,

26, 1627.

(19) (a) Danishefsky, S. J. Aldrichim. Acta 1986, 19, 59. (b) Danishefsky,

S. J. Chemtracts 1989, 2, 273.

(20) Danishefsky, S. J.; Kitahara, T. J. Am. Chem. Soc. 1974, 96, 7807.

(21) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226.

(22) (a) Berkowitz, D. B.; Danishefsky, S. J. Tetrahedron Lett. 1991,

32, 5497. (b) Berkowitz, D. B.; Danishefsky, S. J. Schulte, G. K. J. Am.

Chem. Soc. 1992, 114, 4518.

(23) Hsu, S.-H.; Wu, S.-S.; Wang, Y.-F.; Wong, C.-H. Tetrahedron Lett.

1990, 31, 6403.

(24) Danishefsky, S. J.; Bilodeau, M. T. Angew. Chem., Int. Ed. Engl.

1996, 35, 1381.

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10077

Scheme 3^a

Scheme 4. A C9-C10 Bond Construction through

Ring-Closing Olefin Metathesis

^a(a) trans-1-Methoxy-3-((trimethylsilyl)oxy)-1,3-butadiene (11),

MgBr₂‚OEt₂, THF, -10 °C; then AcOH, H₂O (93%); (b) (i) NaBH₄,

CeCl₃‚7H₂O, MeOH, 0 °C; (ii) TIPSCl, imidazole, DMF, 0 °C f rt

(87% overall); (c) Na⁰, NH₃(l), THF, -78 °C; then MeOH, -78 f

25 °C (92%); (d) Dess-Martin periodinane, pyridine, CH₂Cl₂, rt (98%);

(e) 26, n-BuLi, THF, -78 °C; then 25, THF, -78 °C f rt (87%); (f)

(i) n-Bu₄NF, THF, rt; (ii) PMBCl, NaH, DMF, 0 °C f rt (97% overall);

BOM ) CH₂OCH₂Ph; TIPS ) Sii-Pr₃; PMB ) p-MeOC₆H₄CH₂.

reaction of 16 with 3,3-dimethyldioxirane gave rise to an

intermediate epoxide which, on oxidative solvolysis with sodium

metaperiodate, afforded a 92% yield of aldehyde 17. It was

hoped that the emergence of the future oxygen at C15

(epothilone numbering) in the form of the formate ester would

serve to prevent hemiacetal formation with the aldehyde group.

Such a complicating event would have been anticipated if the

C15 oxygen were free. While there was considerable apprehen-

sion as to whether the formate protecting device would be equal

to the challenges with which it would be confronted, in practice

this group survived during reaction of compound 17 with

allyltriphenylstannane.²⁵There was obtained a 96% yield of a

4:1 mixture of diastereomers at the future C12. The major

product was assumed (and demonstrated on the basis of

subsequent events) to be 18 in the relative configurational sense.

Compound 18 was then subjected to mesylation, followed by

deprotection of the p-methoxybenzyl group to give the hydroxy

mesylate 19. Finally, the sequence was completed by cycliza-

tion of the hydroxy mesylate with lithium hexamethyldisilazide.

In this way, goal system 20 was obtained. Since compound 20

was indeed a cis-epoxide, the assignments of relative stereo-

chemistry to compounds subsequent to the intermediate 12 had

been substantiated. As for the absolute configuration of 20,

this assignment was proven by a sharply modified route (Vide

infra), which was undertaken to avoid the need for any

resolution.

The pursuit of this modified route was also motivated by the

desire to demonstrate still another dimension to the cyclocon-

densation reaction.¹⁹The concept involved the leveraging of

chirality of heterodienophiles in the LACDAC reaction to create

a pyran matrix of defined absolute configurations. This

accomplished, the original asymmetries of the hetereodienophile

can be abrogated, depending on the needs of the synthesis.

To teach this lesson, we proceeded as follows. Cyclocon-

densation of the known lactaldehyde derivative 21²⁶with 11

under mediation by magnesium bromide etherate gave rise to a

93% yield of a dihydropyrone (see Scheme 3). On the basis of

earlier work,²⁷we formulated this pyrone to be structure 22.

The importance of this assignment lies in the statement that it

makes about the absolute stereochemistry of the center destined

to become C15. This center was presumed to be S as a

consequence of R-chelation control in the cyclocondensation

reaction. It will be appreciated that we had introduced an sp³

chiral element which was per se unneeded, at the future trigonal

C16 center (see the indicated carbon atom in 22).

Compound 22 was integrated within the main synthetic

pathway as follows. The ketone function of the dihydropyrone

was reduced, as before, and the resultant alcohol at C3 (glycal

numbering) was protected as its triisopropylsilyl ether (TIPS)

derivative 23. The benzyloxymethyl (BOM) group was dis-

charged through the action of sodium in liquid ammonia and

the alcohol function in the resultant 24 was subjected to

oxidation using the Dess-Martin periodinane procedure.²⁸This

sequence provided 25 in 90% yield. This compound was, in

turn, successfully condensed with phosphine oxide 26^8ain a

Horner reaction,²⁹thereby producing the elongated structure 27.

For purposes of stereochemical correlation, the silyl group was

cleaved (n-Bu₄NF), and the resultant alcohol was reprotected

as its p-methoxybenzyl ether. At this stage, we had achieved

an alternate synthesis of compound 16 in a way that rigorously

defined the relative stereochemistry as well as the absolute

configuration. The conversion of intermediate 16 to 20 has

already been discussed above.

With these successes as a platform, it was appropriate to focus

on the construction of the acyl fragment projected for the olefin

metathesis step. For this purpose, it would be appropriate to

reach a carboxylic acid (cf. 28, Scheme 4) for joining to alcohol

20 to reach system type 4. We further presumed at the planning

level that acid 28 would arise by a two carbon extension of a

generic aldehyde (cf. 29). In this section of the molecule, we

would also be dealing with incorporation of the geminal methyl

groups at the future C4 as well as the implementation of the

appropriate chirality at carbons 6, 7, and 8. The chirality at

C3 would have to be established during the course of the two

carbon homologation alluded to above.

(28) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155. (b)

Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277. (c) Meyer,

S. D.; Schreiber, S. L. J. Org. Chem. 1994, 59, 7549. (d) Ireland, R. E.;

Liu, L. J. Org. Chem. 1993, 58, 2899.

(29) For examples, see: (a) Lythgoe, B.; Nambudiry, M. E. N.; Ruston,

S.; Tideswell, J.; Wright, P. W. Tetrahedron Lett. 1975, 3863. (b) Lythgoe,

B. Chem. Soc. ReV. 1981, 449. (c) Toh, H. T.; Okamura, W. H. J. Org.

Chem. 1983, 48, 1414. (d) Baggiolini, E. G.; Iacobelli, J. A.; Hennessy, B.

M.; Batcho, A. D.; Sereno, J. F.; Uskokovic, M. R. J. Org. Chem. 1986,

51, 3098.

(25) For a review of allylations, see: Yamamoto, Y.; Asao, N. Chem.

ReV. 1993, 93, 2207.

(26) Heathcock, C. H.; Young, S. D.; Hagen, J. P.; Pirrung, M. C.; White,

C. T.; VanDerveer, D. J. Org. Chem. 1980, 45, 3846.

(27) (a) Danishefsky, S. J.; Pearson, W. H.; Harvey, D. F. J. Am. Chem.

Soc. 1984, 106, 2455. (b) Danishefsky, S. J.; Pearson, W. H.; Harvey, D.

F.; Maring, C. J.; Springer, J. P. J. Am. Chem. Soc. 1985, 107, 1256.

10078 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Scheme 5. General Strategy for a Synthesis of the

Scheme 6^a

Polypropionate Domain of the Epothilones

A central goal of our plan was the attainment of stereospeci-

ficity in the management of chirality at carbons 6, 7, and 8. As

our thinking evolved (Vide infra), it seemed that control over

these issues could be facilitated if our scheme included a tem-

porary chirality element at C5. Though C5 is destined to be-

come a ketone, the temporary sp³-associated chirality at this

center would be very valuable in the management of stereo-

chemistry in this region and in the introduction of the geminal

methyl groups at C4. The possibility of using a dihydropyrone

to address this problem presented itself.^8bThe thought was to

translate the C-6, -7, and -8 domain of epothilone to correspond

to dihydropyrone 30 (see Scheme 5). The latter could be ac-

cessed through cyclocondensation chemistry (Vide infra). Hence,

an artificial glycal (cf. 30) was seen to be an exploitable

intermediate en route to subgoal structure 29. More specifically,

the thought was to utilize a C5 alcohol to facilitate and direct

a cyclopropanation of the glycal double bond (see 31 f 32,

Scheme 5).^30,31A regiospecific solvolytic fragmentation of the

cyclopropane ring in 32³²would then provide, in gross terms,

an aldehyde equivalent of the type 33. In the event that the

crucial cyclopropane solvolysis step would be conducted in an

oxidative sense (i.e. E⁺* H), it would then be necessary to

effect a reduction of 33 to reach a compound of the type 34.

The latter would then be advanced, as appropriate, to reach the

desired aldehyde 29. This general thinking is summarized in

Scheme 5.

^a(a) TiCl₄, CH₂Cl₂, -78 °C; then CSA, PhH, rt (87%); (b) LiAlH₄,

Et₂O, -78 °C (91%); (c) Et₂Zn, CH₂I₂, Et₂O, rt (93%); (d) NIS (7

equiv), MeOH, rt; (e) n-Bu₃SnH, AIBN (cat.), PhH, reflux (80% from

39; (f) Ph₃SiCl, imidazole, DMF, rt (97%); (g) 1,3-propanedithiol, TiCl₄,

CH₂Cl₂, -78 f -40 °C (78%); (h) t-BuMe₂SiOTf, 2,6-lutidine,

CH₂Cl₂, 0 °C (98%); (i) (i) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone

(DDQ), CH₂Cl₂/H₂O (19:1), rt (89%); (ii) (COCl)₂, DMSO, CH₂Cl₂,

-78 °C; then Et₃N, -78 °C f 0 °C (90%); (iii) CH₃PPh₃Br,

NaN(SiMe₃)₂, PhCH₃, 0 °C f rt (76%); (iv) PhI(OCOCF₃)₂, CH₂Cl₂/

CH₃CN/H₂O, rt (85%); Bn ) CH₂Ph; TMS ) SiMe₃; TPS ) SiPh₃;

TBS ) Sit-BuMe₂.

to direct a cyclopropanation under modified Conia-Simmons-

Smith conditions^31ato afford cyclopropano derivative 39.

Oxidative solvolytic fragmentation of this cyclopropane was

accomplished through the agency of N-iodosuccinimide in

methanol to provide methyl glycoside 40.^8bReductive deio-

dination of this compound led to the branched artificial methyl

glycoside 41, after which triphenylsilylation afforded the

protected derivative 42.

At this stage, it was timely to cleave the pyran ring with a

view toward liberating the future aldehyde corresponding to C3

of the epothilones. Advancement en route to this goal involved

subjecting compound 42 to the combined action of 1,3-

propanedithiol and titanium(IV) chloride. This protocol led to

the formation of the dithioacetal 43.³⁶Protection of the future

C7 alcohol as shown (see compound 44) was followed by olefin

formation and liberation of the aldehyde function. In this way

the specific compound 45 was in hand.

In practice, titanium-mediated cyclocondensation of the

known and optically pure â-(benzyloxy)isobutyraldehyde 35³³

with diene 36,³⁴following a protocol previously devised in our

laboratory,³⁵gave rise to dihydropyrone 37 (see Scheme 6).^8b

Reduction of this compound with lithium aluminum hydride in

ether provided glycal 38. The hydroxyl group was then used

The next stage in the pursuit of approach I involved a two-

carbon expansion starting with aldehyde 45. The goal was the

production of an acylation partner for alcohol 20 en route to a

competent substrate for ring-closing olefin metathesis. In an

early experiment, aldehyde 45 was treated with the lithium

enolate of tert-butyl acetate (Rathke anion, see Scheme 7).³⁷

(30) (a) Winstein, S.; Sonnenberg, J. J. Am. Chem. Soc. 1961, 83, 3235.

(b) Dauben, W. G.; Berezin, G. H. J. Am. Chem. Soc. 1963, 85, 468. (c)

Furukawa, J.; Kawabata, N.; Nishimura, J. Tetrahedron 1968, 24, 53. (d)

For recent examples of oxygen-directed cyclpropanations of glycals, see:

Hoberg, J. O.; Bozell, J. J. Tetrahedron Lett. 1995, 36, 6831. (e) For an

excellent review of substrate-directable chemical reactions, see: Hoveyda,

A. H.; Evans, D. A.; Fu, G. C. Chem. ReV. 1993, 93, 1307.

(31) For selected examples of glycal cyclopropanations, see: (a) Bo-

eckman, R. K., Jr.; Charette, A. B.; Asberom, T.; Johnston, B. H. J. Am.

Chem. Soc. 1991, 113, 5337. (b) Murali, R.; Ramana, C. V.; Nagarajan,

M. J. Chem. Soc., Chem. Commun. 1995, 217. (c) Henry, K. J., Jr.; Fraser-

Reid, B. Tetrahedron Lett. 1995, 36, 8901. (d) Hoberg, J. O.; Claffey, D.

J. Tetrahedron Lett. 1996, 37, 2533. (e) Timmers, C. M.; Leeuwenburgh,

M. A.; Verheijen, J. C.; van der Marel, G. A.; van Boom, J. H.

Tetrahedron: Asymmetry 1996, 7, 49. (f) Scott, R. W.; Heathcock, C. H.

Carbohydr. Res. 1996, 291, 205. (g) For an excellent review of cyclopro-

panes in synthesis, see: Wong, H. N. C.; Hon, M.-Y.; Tse, C.-H.; Yip,

Y.-C.; Tanko, J.; Hudlicky, T. Chem. ReV. 1989, 89, 165.

(33) (a) Cohen, N.; Eichel, W. F.; Lopresti, R. J.; Neukom, C.; Saucy,

G. J. Org. Chem. 1976, 41, 3505. (b) Nagaoka, H.; Kishi, Y. Tetrahedron

1981, 37, 3873. (c) Meyers, A. I.; Babiak, K. A.; Campbell, H. L.; Comins,

D. L.; Fleming, M. P.; Henning, R.; Heuschmann, M.; Hudspeth, J. P.;

Kane, J. M.; Reider, P. J.; Roland, D. M.; Shimizu, K.; Tomioka, K.;

Walkup, R. D. J. Am. Chem. Soc. 1983, 105, 5015. (d) Roush, W. R.;

Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc. 1990, 112, 6348.

(34) Danishefsky, S. J.; Yan, C.-F.; Singh, R. K.; Gammill, R. B.;

McCurry, P. M., Jr.; Fritsch, N.; Clardy, J. J. Am. Chem. Soc. 1979, 101,

7001.

(32) (a) Collum, D. B.; Mohamadi, F.; Hallock, J. S. J. Am. Chem. Soc.

1983, 105, 6882. (b) Collum, D. B.; Still, W. C.; Mohamadi, F. J. Am.

Chem. Soc. 1986, 108, 2094. (c) For an important early study, see: Wenkert,

E.; Mueller, R. A.; Reardon, E. J., Jr.; Sathe, S. S.; Scharf, D. J.; Tosi, G.

J. Am. Chem. Soc. 1970, 92, 7428.

(35) (a) Danishefsky, S. J.; Pearson, W. H.; Harvey, D. F.; Maring, C.

J.; Springer, J. P. J. Am. Chem. Soc. 1985, 107, 1256. (b) Danishefsky, S.

J.; Myles, D. C.; Harvey, D. F. J. Am. Chem. Soc. 1987, 109, 862.

(36) For an example, see: Egbertson, M.; Danishefsky, S. J. J. Org Chem.

1989, 54, 11.

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10079

Scheme 7^a

We now had in hand substrates 51 and 52. Surprisingly,

when these substrates were submitted to a range of conditions

and catalysts designed to bring about ring-closing olefin

metathesis, no convincing indication could be garnered for the

presence of cyclized systems even to a small extent. While we

could not exclude the possibility that trace amounts of desired

products had been produced, the overwhelming bulk of the

material was clearly of a very complicated nature, suggesting

probable oligomerization. In any case, neither the protected

system 51, nor the structure bearing a free C3 alcohol in 52

served as usable substrates en route to epothilone A.

We did not take the setbacks in this early skirmishing with

substrates 51 and 52 to necessarily establish the nonviability of

the concept. We hoped that the feasibility of the RCM reaction

could be sharply influenced by the nature and stereochemistry

of the “decorating” substituents along the acyl chain. It seemed

possible that certain substrates might be more amenable to

cyclization in that conformational factors (in these variants)

might help to predispose proximity between the terminal vinyl

groups, or that properly selected substituents would be less

conformationally obtrusive in the cyclic products which we were

hoping to form on metathesis. In that spirit, we synthesized a

wide variety of compounds as potential participants in the RCM

reaction.³⁹Unfortunately, none of these candidate substrates

produced workable amounts of cyclized product. At best, mass

spectrometric analysis indicated the possible formation of some

desired materials. However, attempted isolation of traces of

RCM products (assuming they were actually present) from very

complex mixtures proved to be unsuccessful.

These facts, however disheartening, forced us to conclude

that the row of substituents projecting from C3 through C8 had

created an unmanageable problem of steric hindrance with

respect to participation of the C9 double bond in the metathesis

process. To probe this matter further, we went so far as to

synthesize a compound which would, in itself, not constitute a

promising intermediate for reaching epothilone. However, we

thought that the study of the olefin metathesis possibility with

this substrate could shed more light on the failures of the

previously described entries. Accordingly, we synthesized

compound 53⁴⁰in which the vinyl group on the acyl side was

insulated from the secondary methyl group at C8 by another

methylene group. Unhappily, homologous olefin 53 also failed

to undergo ring-closing olefin metathesis.⁴⁰

^a(a) t-BuOC(O)CH₂Li, THF, 0 °C (90%; ca. 2.5:1 mixture of C-3

epimers in favor of 46); (b) t-BuMe₂SiOTf, 2,6-lutidine, CH₂Cl₂, rt;

(c) TESOTf, 2,6-lutidine, CH₂Cl₂, rt (90% overall); (d) Ac₂O, Et₃N,

4-DMAP, CH₂Cl₂, rt (94%); (e) 48, EDC, CH₂Cl₂, 4-DMAP, rt; then

20 (78%); (f) 45 + 49, LDA, THF, -78 °C (2-6:1 mixture of C3

epimers in favor of 52; 85%); TBS ) Sit-BuMe₂; TPS ) SiPh₃.

There was thus generated a mixture of diastereomeric alcohols

at the carbon destined to become C3 of epothilone. The major

product was shown to have the required 3S configuration. The

alcohol function in compound 46 was successfully protected

as the tert-butyldimethylsilyl (TBS) ether derivative (see

compound 47). At this stage, it was possible to cleave the tert-

butyl ester function to generate the acid 48.

The coupling of the previously described alcohol 20 with acid

48 was conducted under the influence of 1-(3-(dimethylamino)-

propyl)-3-ethylcarbodiimide hydrochloride (EDC) to produce,

at long last, the proposed metathesis substrate 51 (see Scheme

7). It was also of interest to investigate the possibility of a

more concise construction of a possible metathesis substrate.

This approach started with acetylation of compound 20 to

provide 49. This acetate, when treated with lithium diisopro-

pylamide, generated a presumed lithium ester enolate (see

proposed structure 50). This enolate underwent successful union

with aldehyde 45 to produce a mixture of C3 epimers with the

desired compound 52 predominating. A particularly interesting

and efficient method for conducting the coupling of 49 and 45

involved merger of the two units in a “Barbier” sense.³⁸In

this mode, the aldehyde and the ester would be concurrently

exposed to the action of lithium diisopropylamide. It was felt

that such a treatment might be successful since the aldehyde is

nonenolizable. In the event, an 85% yield of a ca. 5:1 mixture

of C3 epimers was obtained with the major product being the

3S compound shown. This experiment was to have significant

implications for our intramolecular aldol addition strategy which

will be discussed under approach II.

As described elsewhere,^8bwe went on to demonstrate to our

satisfaction that the “culprit” in preventing the ring forming

olefin metathesis reaction was the network of functionality

between C3 and C8. There was nothing inherently unworkable

(39) For specific substrates tested in ring-closing olefin metathesis

reactions, see Supporting Information.

(40) Compound 53 was not a successful substrate for ring-closing olefin

metathesis.

(37) Rathke, M. W.; Sullivan, D. F. J. Am. Chem. Soc. 1973, 95, 3050.

(38) For a prior instance of a Barbier aldol reaction, see: Linde, R. G.,

II; Jeroncic, L. O.; Danishefsky, S. J. J. Org. Chem. 1991, 56, 2534.

10080 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Scheme 8. Construction of the Compounds 56 and 57

through Ring-Closing Olefin Methathesis

Scheme 10. C11-C12 Suzuki Bond Construction

Scheme 11^a

Scheme 9. Intermolecular Carbonyl Addition Strategy for

the Construction of the C11-C12 Bond

^a(a) (i) DDQ, CH₂Cl₂/H₂O (89%); (ii) (COCl)₂, DMSO, CH₂Cl₂,

-78 °C; then Et₃N, -78 f 0 °C (90%); (b) MeOCH₂PPh₃Cl, t-BuOK,

THF, 0 °C f rt (86%); (c) (i) p-TSOH, dioxane/H₂O, 50 °C (99%);

(ii) CH₃PPh₃Br, NaHMDS, PhCH₃, 0 °C f rt (76%); (d) PhI(O-

COCF₃)₂, MeOH/THF, rt, 0.25 h (92%); Bn ) CH₂Ph; TPS ) SiPh₃;

TBS ) Sit-BuMe₂.

about inclusion of the homoallylic epoxide or the thiazolyl

function, which were situated in the alkyl sector of the proposed

premacrolide substrate. This “probable cause” argument was

demonstrated by successful ring-closing metathesis reactions of

substrates 54 and 55 (see Scheme 8). Unfortunately, while

products 56 and 57, respectively, could be obtained from such

reactions in excellent yields, they did not display significant

biological activity, either in tubulin binding or in cell-culture

cytotoxicity studies.

We were now in the horns of what seemed to be an

unsolvable dilemma. Those substrates in which olefin metath-

esis, leading to a C9-C10 cycloalkene (see Scheme 8), could

be conducted provided products with nonuseful biological

profiles. Conversely, those substrates which were functionalized

in the spirit of serious potential epothilone precursors failed to

undergo ring-closing olefin metathesis. It was when the full

scope of this conundrum became clear that approach I was set

aside in favor of other options.

Since we had demonstrated the ability to generate a protected

aldehyde of the type 17 (Scheme 2), we turned to the possibility

of a convergent coupling of this sort of system (generalized as

58) with a conventional nucleophile in the form of a metallo

derivative (generalized as 60) (see Scheme 9). In the ideal

scenario, 60 would be derived from 59, in which case a product

such as 61 could be anticipated.

In practice, it proved possible to synthesize potential probe

compounds for such metalation reactions (i.e. systems of the

type 59). Surprisingly, at no point were we able to accomplish

the metalation of any such derivative to produce a competent

organometallic nucleophile corresponding to 60. In all cases,

either unreacted aldehyde was recovered with the protonated

metal species or decomposition took place. These failures were

documented, not only in attempted couplings to the relatively

sophisticated electrophile 17, but even with much simpler

electrophiles. We could garner little evidence to show that we

had achieved metalation either in the series n ) 0 or n ) 1.

These failures, suggesting problems in accessing external agents

to the terminus of 59, mirrored some of the difficulties which

were soon to be encountered in the Suzuki coupling scheme

(Vide infra). Fortunately, this problem could be overcome in a

novel way.

B-Alkyl Suzuki Strategy (Approaches II and III, Scheme

1)

We now report the results of the first successful syntheses of

epothilones A and B which were achieved by the alkyl Suzuki

method. The concept is generalized in Scheme 10 anticipating

a synthesis of epothilones A and B. Through some as yet

unspecified method, we envisioned a route to a Z-haloalkene

62. Furthermore, it was expected that the chemistry would lend

itself to construction of a terminal vinyl system in the context

of a protected C3 substructure, generalized as system 63.

Construction of the C11-C12 bond would be the hallmark of

the scheme and would be accomplished through a B-alkyl

Suzuki coupling¹⁵

(Vide infra). Also to be dealt with would be

a two carbon insert corresponding to carbons 1 and 2 of

epothilone (see 64). The appendage 64 could be incorporated

in the scheme at several stages. If these carbon-carbon bond

producing maneuvers were to be realized, there would also

remain the need for introduction of the C12-C13 â-epoxide

through the action of a suitable oxidizing agent. It was from

these perceptions that our overall strategy for reaching epothilone

A emerged. With suitable modification, a route to epothilone

B was also embraced under this paradigm.

In practice, we turned to aldehyde 65 (Scheme 11). Com-

pound 65 had been previously described as arising from 44 and

had been converted to terminal vinyl compound 45 (Scheme

6). Now, the same aldehyde was successfully coupled to

(methoxymethyl)triphenylphosphorane to give rise to 66.^8cThe

latter was then subjected to sequential hydrolysis and Wittig

reactions to afford 67. Finally, it proved possible to convert

the dithiane linkage protecting C3 to the dimethyl acetal 68.

Here it was envisioned that the future C3 aldehyde could be

revealed from the dimethyl acetal even in a mutlifunctionalized

substrate.

With this chemistry in hand, we turned our attentions to the

other projected Suzuki coupling partner. The specific version

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10081

Scheme 13^a

Scheme 12^a

^a(a) Allyltri-n-butylstannane, (S)-(-)-BINOL, Ti(Oi-Pr)₄, CH₂Cl₂,

-20 °C (60%; >95% ee); (b) [(-)-Ipc]₂BCH₂CHCH₂, Et₂O, -100 °C;

then 3 N NaOH, 30% H₂O₂(83%; >95% ee); (c) Ac₂O, 4-DMAP,

Et₃N, CH₂Cl₂(96%); (d) (i) OsO₄, NMO, 0 °C; (ii) NaIO₄, THF/H₂O,

rt (iii) 79, THF, -78 f 0 °C (50% overall).

^a(a) Dihydropyran, PPTS, CH₂Cl₂, rt (73%); (b) (i) Me₃SiCCLi,

BF₃‚OEt₂, THF, -78 °C (76%); (ii) MOMCl, i-Pr₂NEt, Cl(CH₂)₂Cl,

55 °C (85%); (iii) PPTS, MeOH rt (95%); (c) (i) (COCl)₂, DMSO,

CH₂Cl₂, -78 °C; then Et₃N, -78 f rt; (ii) MeMgBr, Et₂O, 0 °C f rt

(85% for two steps); (iii) TPAP, NMO, 4 Å mol sieves, CH₂Cl₂, 0 °C

f rt (93%); (d) 26, n-BuLi, THF, -78 °C; then 72, THF, -78 °C f

rt (97%); (e) (i) N-iodosuccinimide, AgNO₃, (CH₃)₂CO (64%); (ii)

dicyclohexylborane, Et₂O, AcOH (65%); (f) (i) PhSH, BF₃‚OEt₂,

CH₂Cl₂, rt (86%); (ii) Ac₂O, pyr, 4-DMAP, CH₂Cl₂, rt (99%); PPTS

) pyridinium p-toluenesulfonate; MOMCl ) methoxymethyl chloride;

TPAP ) tetra-n-propylammonium perruthenate; NMO ) N-methyl-

morpholine N-oxide.

95% enantiomeric excess (see Scheme 13). Alternatively, we

could effect an asymmetric allylation of enal 10 through the

use of Brown’s procedure.^44,9bThese transformations were

generally faster and higher yielding but, of course, lacked the

feature of implementation of chirality through catalytic means.

The optical purity of allylic alcohol 76 prepared by allyl

boration, was established by formation of the Mosher ester and

subsequent analysis by ¹H and ¹⁹F NMR spectroscopy. Eventu-

ally, the assignment and extent of optical purity were cor-

roborated by interfacing this product with compound 75 derived

from the glycidol route. Protection of carbinol 76 afforded

acetate 77. In a very delicate set of transformations, this

homoallylic acetate was subjected to oxidative cleavage, as

shown, to generate the putative â-acetoxy aldehyde 78. That

we had resorted, in the first instance, to the glycidol route

reflected our fears that such a structure would be nonviable given

its projected vulnerability to â-elimination of the cinnamyl-like

acetoxy function. However, in practice, this difficulty could

be managed. Wittig-type reaction with the known phosphorane

79⁴⁵gave rise to 75. Certainly, this route proved to be more

concise for reaching compound 75 than the R-glycidol based

route shown in Scheme 12. The actual practicality of the

process will depend on the feasibility of the scale up of the

conversion of 77 to the vulnerable 78 en route to 75.

Even while this work was in progress, we were investigating

a variation wherein the hypothetical Suzuki coupling would be

conducted with a more advanced coupling partner, better

positioned to reach epothilone itself. For this purpose, we

returned to the acetal 68 which was deprotected to give rise to

aldehyde 80 (see Scheme 14). This compound was condensed

with lithio tert-butyl acetate³⁷(i.e. the Rathke anion). There was

produced a 63% yield of 81, as well as its C3 stereoisomer

(82, not shown here) in a ratio of approximately 2:1. The

undesired C3 epimer could be oxidized to the corresponding

ketone with Dess-Martin periodinane²⁸and subsequently

reduced in a stereoselective fashion to give the desired C3

alcohol exclusively. The major product 81 was subjected to

the action of buffered HF‚pyridine, whereupon the triphenylsilyl

function was selectively cleaved from the C5 oxygen. It was

further possible to selectively silylate the C3 alcohol through

the combined action of TBS triflate and 2,6-lutidine, providing

compound 83. At this point, the C5 alcohol in compound 83

was oxidized to produce the ketone 84. Finally, the tert-butyl

of the generic structure 62 which was settled upon was the

iodoacetate (see 75, Scheme 12). Needless to say, it would be

necessary to “deliver” this compound with the appropriate olefin

geometry and in optically pure form for melding into the

epothilone synthesis. In theory, depending on the coupling

modality, we could utilize either enantiomeric version at

C15; each enantiomer could be interfaced into the synthesis,

subject to whether inversion or retention would be required at

C15.

In our first route,^8cwe anticipated retention of configuration

at this center (see Scheme 12). Accordingly, our program

started with the commercially available R-(+)-glycidol (69).⁴¹

The hydroxyl group was protected in the form of a THP ether

(see compound 70). In a defining step, the epoxide linkage

was used to alkylate the lithium salt of (trimethylsilyl)acetylene,

under the conditions described, to give rise to compound 71. It

will be recognized that the chiral center of glycidol is retained

en route to coupling partner 75.

The next phase involved the classical transformation of the

primary alcohol linkage to a methyl ketone. This was ac-

complished, as indicated, to provide ketone 72. Drawing from

an important precedent (see 25 + 26 f 27, Scheme 3),^8awe

could accomplish the introduction of the thiazolyl nucleus

through an Emmons reaction²⁹of phosphine oxide 26 with

methyl ketone 72. In the concluding phase of this synthesis,

silyl acetylene 73 was converted to the corresponding iodoalkyne

which, upon reduction,⁴²gave rise to the cis-iodoalkene 74.

Finally, cleavage of the MOM protecting group, as shown,

followed by acetylation, produced the desired cis-vinyl iodide

75.

When the general concept of the alkyl Suzuki coupling proved

to be fruitful (Vide infra), more concise syntheses of 75 were

achieved. For this purpose, we returned to the enal 10, a

substance employed in an earlier stage of the synthesis.^8a

Allylation of this compound with tri-n-butylstannane in the

presence of the (S)-BINOL enantiodirecting ligand, as described

by Keck,⁴³gave rise to the allylated product 76 in greater than

(43) Keck, G. E. Tarbet, K. H.; Geraci, L. S. J. Am. Chem. Soc. 1993,

115, 8467.

(41) For an excellent review of the chemistry of glycidol, see: Hanson,

R. M. Chem. ReV. 1991, 91, 437.

(42) Corey, E. J.; Cashman, J. R.; Eckrich, T. M.; Corey, D. R. J. Am.

Chem. Soc. 1985, 107, 713.

(44) Racherla, U. S.; Brown, H. C. J. Org. Chem. 1991, 56, 401.

(45) (a) Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 2173. (b) Stork,

G.; Zhao, K. J. Am. Chem. Soc. 1990, 112, 5875. (c) Chen, J.; Wang, T.;

Zhao, K. Tetrahedron Lett. 1994, 35, 2827.

10082 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Scheme 14^a

stereointegrity of the C11-C12 double bond. The remarkable

versatility of the B-alkyl Suzuki reaction¹⁵was further demon-

strated by successful coupling of keto ester 85 with 75. Under

these circumstances, the TBS ester was cleaved during the

course of the reaction, and the acetate was subsequently cleaved

through the action of K₂CO₃in aqueous methanol to give rise

to the hydroxy acid 88.

Our attentions would next be directed to the construction of

the 16-membered ring. It was felt that our chances for achieving

a stereoselective epoxidation would be better if the framework

of the ring system were already in place when the oxidation of

the C12-C13 double bond would be conducted. In our first

attempt at macrocyclization, we favored a bold possibility. The

thought was to close the ring by connecting the methyl group

of the acetate ester (C2) with the aldehyde center (C3) in a

construct to be deriVed from compound 86. That such a prospect

could be even considered, arose from the fact that the gem-

dimethyl substitution at C4 blocks the possibility of deproto-

nation of the aldehyde function. It will be recalled that earlier,

in converting compound 45 to 52 (Scheme 7), we had exploited

this principle by conducting an ester enolate aldol coupling under

Barbier-type conditions.³⁸Here, we would be drawing from

the same concept in a macroaldolization step. To set the stage

for this interesting ring-forming possibility, the acetal function

in compound 86 was cleaved, thus revealing the electrophilic

C3 aldehyde (see 86 f 89, Scheme 15). In the crucial event,

deprotonation of compound 89 (see Scheme 16) was accom-

plished through the action of potassium hexamethyldisilazide

in THF at -78 °C. Remarkably, these conditions allow a

stereoselective macroaldolization, resulting in the selective

formation (6:1) of the desired (S)-C3 alcohol 90. In some small

scale experiments, compound 90 was the only product noted at

the analytical level. However, optimal results from the stand-

point of yield were actually obtained when the aldolate

intermediate derived from cyclization was quenched at 0 °C or

even at room temperature. When the quenching experiment

was conducted at lower temperature, greater amounts of the

undesired epimer 91 were obtained with an increase in mass

recovery. Apparently, aldolate equilibration favors the forma-

tion of the desired 3S alcohol, whereas conditions more nearly

approximating kinetic control give rise to lower degrees of

stereoselectivity. At higher temperatures, it would seem that

there is virtually no kinetic control in the stereochemistry of

the macroaldolization step; the diastereomeric ratios observed

indicate that equilibration occurs. While it is interesting to

ponder and sort out methods to control stereoselectivity, in

practice, the undesired epimer 91 could be utilized in our

synthesis. Thus, oxidation of 91 to the ketone 92 set the stage

for a diastereoselective reduction with NaBH₄to provide the

desired epimer 90 in high yield. Presumably, this outcome

reflects the directing effects of the C5-OTPS function.

^a(a) p-TsOH, dioxane/H₂O (5:1), 50 °C (81% overall); (b) tert-butyl

acetate, LDA, THF, -78 °C, (95%; ca. 2:1 mixture of C-3 epimers);

(c) HF‚pyr, pyr, THF, rt (98%); (d) TBSOTf, 2,6-lutidine, CH₂Cl₂, -30

°C (96%); (e) Dess-Martin periodinane, CH₂Cl₂, rt (89%); (f) TBSOTf,

2,6-lutidine, CH₂Cl₂, rt (93%); TPS ) SiPh₃; TBS ) Sit-BuMe₂.

Scheme 15^a

^a(a) 68, 9-BBN, THF, rt; then 75, PdCl₂(dppf)₂, Cs₂CO₃, Ph₃As,

H₂O/DMF, rt (75%); (b) 85, 9-BBN, THF, rt; then 75, PdCl₂(dppf)₂,

Cs₂CO₃, Ph₃As, H₂O/DMF, rt (56%); (c) p-TsOH, dioxane/H₂O, 50

°C (85%); (d) K₂CO₃, MeOH/H₂O (84%); 9-BBN ) 9-borabicyclo-

[3.3.1]nonane; dppf ) 1,1′-bis(diphenylphosphino)ferrocene; TPS )

SiPh₃; TBS ) Sit-BuMe₂.

ester could be converted to the tert-butyldimethylsilyl ester (see

compound 85) through the agency of TBS triflate and 2,6-

lutidine.

We were now in a position to probe the feasibility of

establishing the C11-C12 bond by Suzuki coupling in several

contexts. Compound 68 was subjected to the action of 9-BBN

(see Scheme 15). Remarkably, even the hydroboration of this

terminal olefin required surprisingly coercive conditions. This

slowness of hydroboration is reminiscent of the difficulties

previously discussed in Scheme 9 for metalation of systems of

the type 59 as a route to 60. Fortunately, the hydroboration of

68 could be conducted under more vigorous conditions (as

demonstrated by independent oxidative quenching experiments

that revealed the anti-Markovnikov hydration product).

Having convinced ourselves that compound 68 had indeed

been successfully hydroborated, we now conducted the last

phase of the Suzuki reaction under mediation by palladium(II)

chloride-1,1′-bis(diphenylphosphino)ferrocene (dppf) in the

presence of the (Z)-vinyl iodide 75 as shown (Scheme 15).

Gratifyingly, there was obtained a 72% yield of the acetate 86.^8c

Within the limits of our detection, there had been no loss of

Cleavage of the triphenylsilyl ether in 90 could be conducted

selectively, producing the C3-C5 diol (see compound 93).

Selective protection of the C3 hydroxyl in this compound was

readily achieved, thus exposing the C5 alcohol in 94 for

oxidation to a ketone. There was then produced di-TBS C12-

C13 desoxyepothilone (95). Cleavage of the two silyl protecting

groups could be accomplished, giving rise to desoxyepothilone

A (96).

The whole scheme was now at considerable risk as we

approached the matter of epoxidation of the C12-C13 double

bond. We had hoped that oxidation would occur from the

desired â-face on the basis of local conformational preferences

that rendered this face of the molecule more accessible (see

substructure 96 in Figure 2).⁴⁶We further hoped to maximize

our opportunities for stereocontrol by conducting the reaction

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10083

Figure 2. Macromodel minimized stereoview of desoxyepothilone A (96).

Scheme 16^a

at low temperature. Given our many successful applications

of the oxidizing powers of 3,3-dimethyldioxirane²⁴it was not

unnatural that we would turn to this reagent.

In practice, the major product of this reaction was the long

sought after epothilone A (see 96 f 2, Scheme 16), confirmed

by spectral and chromatographic comparisons of this material

with authentic epothilone A kindly proVided by Professor Ho¨fle.

The first total synthesis of epothilone A had thus been

accomplished. Two very minor products in the epoxidation

were also isolated. One is the corresponding R-epoxide between

C12 and C13. Still another product is one in which this epoxide

linkage is present, but is accompanied by an epoxide functional-

ity at C16-C17.

Although we had achieved the epoxidation reaction with high

stereoselectivity, our reasoning based on model 96 can be

questioned. Nicolaou and co-workers studied the use of the

more conventional m-chloroperoxybenzoic acid.^9b,cThe use of

this reagent produced mixtures of R and â epoxides. Hence,

the factors controlling the facial sense of epoxidation are rather

more subtle than is reflected in modeling of gross steric

accessibility as suggested in 96, since the course of oxygen

delivery is strongly reagent dependent. However, we did have

at our disposal a protocol for highly stereoselective epoxidation

in the desired sense.

Having demonstrated the macroaldol route, we turned to the

possibility of macrolactonization. This goal brought us back

to compound 88 which, under Yamaguchi conditions,⁴⁷led to

the previously encountered 95. Thus, we now had available to

us two routes to enter the desoxyepothilone series in the form

of compound 95 and, shortly thereafter, epothilone itself.

We next turn to our total synthesis of epothilone B (3) (see

Scheme 17).^8eWe had hoped that this synthesis could be

accomplished using, as much as possible, the chemistry that

had served so well for the synthesis of epothilone A (2).^8c

Indeed, our route started with compound 77, which was cleaved

to the corresponding aldehyde 78. Condensation of this

aldehyde with the appropriate Wittig reagent^45cgave rise to

compound 97, albeit in only 43% yield. Fortunately, the

reaction was highly stereoselective, giving rise to the required

Z-isomer as the only product. The stage was now set for the

key Suzuki coupling. In this instance, we confined ourselves

to acetal olefin 68. Once again, hydroboration of 68 as before

was followed by coupling of the resultant borane with (Z)-vinyl

iodide 97, giving rise to compound 98 in 77% yield. Cleavage

of the acetal linkage led to aldehyde 99. Once again, aldol

^a(a) KHMDS, THF, -78 °C, 0.001M (51%, 6:1 R/â); (b) Dess-

Martin periodinane, CH₂Cl₂, rt; (c) NaBH₄, MeOH, THF, -78 °C f

rt (80% for two steps); (d) HF‚pyridine, pyridine, THF, rt (99%); (e)

TBSOTf, 2,6-lutidine, CH₂Cl₂, -30 °C (93%); (f) Dess-Martin

periodinane, CH₂Cl₂, rt (84%); (g) 2,4,6-trichlorobenzoyl chloride, TEA,

4-DMAP, toluene, rt (88%); (h) HF‚pyridine, THF, rt (99%); (i) 3,3-

dimethyldioxirane, CH₂Cl₂, -35 °C (49%; g16:1 mixture of diaster-

eomers in favor of 2); TPS ) SiPh₃; TBS ) Sit-BuMe₂.

condensation occurred in much the same manner as previously

noted for compound 89, thus allowing us to enter into the

desoxyepothilone B series.

The protocols to reach desoxyepothilone B from cyclized

material were already in hand from our synthesis of the A

compound, 90. Thus, in the case at hand, cleavage of the C5

TPS ether generated the diol 103, which upon resilylation of

(46) Molecular modeling was performed with MacroModel version 5.5;

The MMZ force field was used with a Monte Carlo random walk

conformational search.

(47) (a) Yamaguchi, M.; Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.

Bull. Chem. Soc. Jpn. 1979, 52, 1989. (b) Mulzer, J.; Mareski, P. A.;

Buschmann, J.; Luger, P. Synthesis 1992, 215.

10084 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Scheme 17^a

Scheme 18^a

^a(a) 3-Butenylmagnesium bromide, Et₂O, -78 f 0 °C; (b) 4-iodo-

2-methyl-1-butene, t-BuLi (2.1 equiv), Et₂O, -78 f -50 °C; then 65,

Et₂O, -78 f 0 °C; (c) (i) thiocarbonyl diimidazole, 4-DMAP, 95 °C;

(ii) n-Bu₃SnH, AIBN, C₆H₆, 80 °C; (d) (i) (CF₃CO₂)₂IC₆H₅, MeOH/

THF, rt; (ii) p-TsOH, dioxane/H₂O, 50 °C.

Scheme 19^a

^a(a) (i) OsO₄, NMO, (CH₃)₂CO/H₂O, 0 °C; (ii) Pb(OAc)₄, Na₂CO₃,

C₆H₆, 0 °C f rt; (iii) Ph₃PdC(I)CH₃, THF, -20 °C (43% from 77; Z

geometrical isomer only); (b) 68, 9-BBN, THF, rt; then 97, PdCl₂(dppf)₂,

Cs₂CO₃, Ph₃As, DMF/H₂O, rt (77%); (c) p-TsOH, dioxane/H₂O, 55

°C (71%); (d) KHMDS, THF, -78 °C (60%; 100:101/2.1:1); (e) Dess-

Martin periodinane, CH₂Cl₂, rt; (f) NaBH₄, MeOH, rt (67% for two

steps); (g) HF‚pyridine, pyridine, THF, rt (94%); (h) TBSOTf, 2,6-

lutidine, CH₂Cl₂, -30 °C (89%); (i) Dess-Martin periodinane, CH₂Cl₂,

rt (87%); (j) HF‚pyridine, THF, rt (92%); (k) 3,3-dimethyldioxirane,

CH₂Cl₂, -50 °C (97%; g20:1 mixture of diastereomeric cis-epoxides

in favor of 3); NMO ) N-methylmorpholine N-oxide; 9-BBN )

9-borabicyclo[3.3.1]nonane; dppf ) 1,1′-bis(diphenylphosphino)fer-

rocene; KHMDS ) KN(SiMe₃)₂; TPS ) SiPh₃; TBS ) Sit-BuMe₂.

^a(a) LDA, THF, -78 °C (65%; 111:112/1:1); (b) LDA, THF, -78

°C (70%; 114:115/ca. 1:1); (c) Dess-Martin periodinane, CH₂Cl₂, rt;

(d) NaBH₄, MeOH, THF, -78 °C f rt (ca. 92% for two steps).

the exposed C3 hydroxyl group gave the product 104. Oxidation

of the C5 alcohol led to desoxyepothilone B bis (TBS) ether

(compound 105). Cleavage of the silyl blocking groups at C3

and C7 was accomplished, as shown, thereby allowing us to

reach desoxyepothilone B (106). For obvious reasons, we

turned to the use of dimethyl dioxirane in the oxidation of this

compound. Happily, this reaction was eVen more regio- and

stereoselectiVe, producing epothilone B (3) identical in all

respects with an authentic sample, kindly proVided by Professor

Ho¨fle.

In order to probe the applicability of such a construction to

the total synthesis of epothilone B, we returned to aldehyde

dithiane 65. The latter was converted to compound 109 by

isobutenylation and deoxygenation. Once again, cleavage of

the dithiane linkage provided aldehyde 110.

With compounds 77 as well as 108 and 110 in hand, assembly

of substrates for RCM were possible. To simplify the initial

merger step, we turned, once again, to an intermolecular aldol

condensation of the ester enolate, derived from 77, with

aldehydes 108 and 110 (see Scheme 19). In practice, the

coupling could be readily conducted to give a mixture of

stereoisomers at C3. The product, bearing the S configured

alcohol, could be separated. The R-alcohol, in each case, was

recycled through an oxidation/reduction sequence as shown.

Indeed, with increased spacing between the C12 olefin and

the branched positions of the polypropionate domain (see

compound 111), olefin metathesis chemistry proved to be

successful.^8dCyclizations were conducted as described in

Scheme 20 for compounds 111 and 112. In these studies, we

took recourse to both the ruthenium-based catalyst of Grubbs^12b

and the molybdenum-based catalyst of Schrock^12ato mediate

metathesis. In our work, the ruthenium-based system proved

to be generally more efficacious for constructing the disubsti-

tuted double bonds with the properly configured (3S) alcohol.

The Second Generation Ring-Closing Olefin Metathesis

Strategy (Approach IV, Scheme 1)

Even though we had accomplished our primary goals of

synthesizing epothilones A and and B, it was still of interest to

reinvestigate the possibility of intramolecular olefin metathesis.^8d

However, in this case we would be focusing on elaborating the

C11-C12 double bond in the course of the metathesis reaction.

Since we had known that epoxidation of this double bond could

be conducted in a highly stereoselective and favorable direction,

we were obviously more disposed to consider syntheses where

this double bond would be elaborated in the decisive cyclization

step. Toward this end, we returned to thioacetal aldehyde 65.

This compound was itself converted to product 107 through

butenylation and deoxygenation at C9 (see Scheme 18). The

C3 aldehyde function could be liberated by cleavage of the

dithiane, as indicated, to provide aldehyde 108.

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10085

Scheme 22^a

Scheme 20^a

^a(a) RuBnCl₂(PCy₃)₂, 50 mol %, C₆H₆, 0.001 M rt, 24 h.

Scheme 21^a

^a(a) RuBnCl₂(PCy₃)₂, 50 mol %, C₆H₆, 0.001 M, rt, 4 h; (b)

Mo(CHMe₂Ph)(N(2,6-(i-pr)₂C₆H₃))(OCMe(CF₃)₂)₂20 mol %, C₆H₆,

0.001 M, rt, 1 h, 86%.

ously encountered Z-desoxyepothilone B (106) and E-desoxy-

epothilone B (124).

^a(a) HF‚pyr, pyr, THF, rt (93%); (b) TBSOTf, 2,6-lutidine, CH₂Cl₂,

-30 °C (85%); (c) Dess-Martin periodinane, CH₂Cl₂(94%); (d)

Mo(CHMe₂Ph)(N(2,6-(i-pr)₂C₆H₃))(OCMe(CF₃)₂)₂20 mol %, C₆H₆,

0.001 M, 55 °C, 2 h, 86%, 105/123 1:1; (e) HF‚pyr, THF, rt, 2 h,

90%.

As seen with substrates 111 and 112, a point mutation of

stereochemistry at C3 had tilted the process toward substantial

stereoselectivity. Unfortunately, it was the 3-epi substrate (112)

in which a high stereochemical margin was obtained, and, in

fact, the unnatural E-double bond isomer was favored. We then

mounted a considerable effort toward improving the stereose-

lectivity of the olefin metathesis reaction with the goal of

reaching the natural Z series from the “natural” 3S carbinol

precursor.

Unfortunately, no hypothesis emerged to guide our experi-

ments in crafting the remote functions in the C3-C7 sector.

Accordingly, we took recourse to intermediates that were

accessible from the synthetic studies already in place. In this

respect, we had occasion to prepare compounds 125 to 128 and

to study their olefin metathesis.^8dUsing our collection of

substrates, we were able to observe effects on the stereochemical

course of olefin metatheses as a function of the nature of the

substituents along the acyl chain. However, the only decisive

perturbations favoring significant stereoselectivity were in the

transformations of 112 to 118 and 119 (Scheme 20) as well as

125 to 129 and 130 (Scheme 22), each of which resulted in the

predominant formation of the unnatural E compounds.

Although the yields of cyclized products were generally quite

good, it was unfortunate that the resultant C12-C13 olefins

were produced as a serious mixtures of E/Z isomers (90:117).

The Z compounds could be correlated with earlier intermediates

arising from the previously described B-alkyl Suzuki pathway.

The E compounds were independently deprotected and con-

verted to the corresponding E desoxyepothilone systems, as

shown in Scheme 23 (Vide infra).

Olefin metathesis was also attempted on compound 114 (see

Scheme 21).^8eIn the event, this substrate failed to cyclize with

the ruthenium-based system described by Grubbs,^12bor the

molybdenum based catalyst described by Schrock.^12aHowever,

when compound 122, derived from 124 as shown, was treated

with the Schrock catalyst, cyclization was successful producing

a 1:1 mixture of Z and E isomers 105 and 123.^8eThese products

were independently processed, as shown, leading to the previ-

10086 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

Scheme 23^a

as A, through adaptations of the B-alkyl Suzuki reaction.

Recourse to the separation of E:Z isomer mixtures arising from

olefin metathesis is, at least in our hands, seriously disabling

in terms of throughput of significant amounts of material.

A stereoselective conversion of desoxyepothilone A and B

to the epoxides in the natural series has been accomplished with

the use of dimethyldioxirane (see conversion of 96 f 2 and

106 f 3). Also illustrated in these studies were the power of

catalytic asymmetric allylation (10 f 76), the flexibility of

glycidol as a multifacated member of the chiral pool (see 69 f

75), and the power of the LACDAC reaction in assembling

polypropionate frameworks (see 35 + 36 f 45, 68 and 85).

The rather novel cyclopropanation of a glycal (38 f 39)

followed by oxidative solvolysis (39 f 40) and reduction (40

f 41) as a route to the introduction of quaternary branching is

also deserving of attention. Given the generality of the issues

which have been addressed, it is likely that the lessons garnered

here would find application to other problems in organic

synthesis.

Experimental Section

General. All commercial materials were used without further

purifications unless otherwise noted. The following solvents were

distilled under positive pressure of dry nitrogen immediately before

use: THF and diethyl ether from sodium/potassium-benzophenone

ketyl, CH₂Cl₂, toluene, and benzene from CaH₂. All the reactions were

performed under N₂atmosphere. NMR (¹H, ¹³C) spectra were recorded

on Bruker AMX-400 MHz, Bruker Avance DRX-500 MHz, referenced

to TMS (¹H-NMR, δ 0.00) or CDCl₃(¹³C-NMR, δ 77.0) peaks unless

otherwise stated. LB ) 1.0 Hz was used before Fourier transformation

for all of the ¹³C-NMR. IR spectra were recorded with a Perkin-Elmer

1600 series-FTIR spectrometer, and optical rotations were measured

with a Jasco DIP-370 digital polarimeter using 10 cm pathlength cell.

Low-resolution mass spectral analysis were performed with a JEOL

JMS-DX-303 HF mass spectrometer. Analytical thin-layer chroma-

tography was performed on E. Merck silica gel 60 F₂₅₄plates (0.25

mm). Compounds were visualized by dipping the plates in a cerium

sulfate-ammonium molybdate solution followed by heating. Flash

column chromatography was performed using the indicated solvent on

E. Merck silica gel 60 (40-63 mm) or Sigma H-Type silica gel (10-

40 mm). Melting points are obtained with Electrothermal melting point

apparatus (series no. 9100) and are uncorrected.

Preparation of Compound 68. A solution of (methoxymethyl)-

triphenylphosphonium chloride (2.97 g, 8.55 mmol) in THF (25 mL)

at 0 °C was treated with KO^tBu (8.21 mL, 1 M in THF, 8.1 mmol).

The mixture was stirred at 0 °C for 30 min. Aldehyde 65 (3.10 g,

4.07 mmol) in THF (10 mL) was added, and the resulting solution

was allowed to warm to rt and stirred at this temperature for 2 h. The

reaction was quenched with saturated aqueous NH₄Cl (40 mL), and

the resulting solution was extracted with Et₂O (3 × 30 mL). The

combined Et₂O fractions were washed with brine (20 mL), dried over

MgSO₄, filtered, and concentrated. The residue was purified by flash

chromatography on silica gel eluting with 5% Et₂O in hexanes to yield

the methyl enol ether 66 (2.83 g, 86%) as a colorless foam.

^a(a) 3,3-Dimethyldioxirane, CH₂Cl₂, -30 °C, 70%, 3:1 â/R; (b)

HF‚pyr, THF, rt, 93%; (c) HF‚pyr, pyr, THF, rt, 92%.

As is seen from the data, there is a sensitivity of olefin

geometry to variation of substituents at even some distance from

the terminal vinyl groups. Presumably, the consequence of these

structural permutations reflect subtle effects on the sense of

presentation of the vinyl group of one sector to the terminal

metal-carbene complex¹²derived from the other sector. In this

connection, it is also interesting that the olefin geometry ratio

is also sensitive to the catalyst employed (see ratio of 95:131)

when this question was probed.

Compounds 118, 129, and 131 were processed as shown to

afford 3-epi-epothilone A (136), epothilone A, and (E)-12,13-

desoxyepothilone A (132) (Scheme 23). The results of evalu-

ations of the biological profiles of these epothilone analogs have

been published elsewhere^8d,eand provide a basis for the

development of new classes of structurally simpler and syntheti-

cally more accessible agents.

Summary

Several total syntheses of epothilones A and B have been

described herein. These constructions involved union at either

the C11-C12 bond (B-alkyl Suzuki coupling) or the C12-C13

bond (ring-forming olefin metathesis). Our routes differ from

all others in several important respects. First, the stereochem-

istry of the polypropionate region accrues from the LACDAC

reaction¹⁹and cyclic matrices elaborated therefrom. Ultimately,

all sterochemistry in this domain is induced from the single

chiral center, readily deriVable “Roche aldehyde” deriVatiVe

35 using sound principles formulated in our group many years

ago.¹⁹

Of particular importance is that C8 is incorporated in this

dynamic. By contrast, the other syntheses have taken recourse

to separate syntheses of R-methyl aldehydes, using chiral

auxiliaries, to “deliver” the C8 chiral center to the synthetic

pool. Furthermore, our syntheses uniquely provide strict control

over the geometry of the double bonds of epothilone B as well

To a solution of the methyl enol ether 66 (2.83 g, 3.50 mmol) in

dioxane/H₂O (9:1, 28 mL) was added pTSA‚H₂O (1.0 g, 5.30 mmol),

and the resulting mixture was heated to 50 °C for 2 h. After cooling

to rt, the mixture was diluted with Et₂O (50 mL) and washed

successively with saturated aqueous NaHCO₃(15 mL) and brine (20

mL), dried over MgSO₄, filtered, and concentrated to provide the

corresponding aldehyde (2.75 g, 99%) as a colorless foam: [R]_D

)

+1.74 (c ) 0.77, CHCl₃); IR (film) 2929, 1725, 1428, 1253, 1115,

1039 cm^-1; ¹H NMR (CDCl₃, 500 MHz) δ 9.31 (d, J ) 3.4 Hz, 1 H),

7.68 (dd, J ) 7.8, 1.4 Hz, 6 H), 7.45-7.37 (band, 9 H), 4.60 (d, J )

6.8 Hz, 1 H), 4.20 (s, 1 H), 3.51 (d, J ) 6.7 Hz, 1 H), 2.68 (d, J )

14.0 Hz, 1 H), 2.60 (d, J ) 13.5 Hz, 1 H), 2.37 (m, 1 H), 2.24 (m, 1

H), 1.90 (m, 1 H), 1.81 (m, 2 H), 1.68 (m, 2 H), 1.52 (m, 1 H), 1.32

(s, 3 H), 1.14-1.03 (band, 6 H), 0.86 (s, 9 H), 0.75 (d, J ) 6.9 Hz, 3

H), -0.03 (s, 3 H), -0.06 (s, 3 H); ¹³C NMR (CDCl₃, 125 MHz) δ

202.8, 136.0, 134.6, 130.1, 128.0, 77.6, 76.2, 59.5, 45.1, 44.7, 43.8,

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10087

31.8, 30.9, 30.5, 26.3, 25.9, 22.5, 20.9, 18.6, 17.9, 14.7, -3.0, -3.5;

HRMS calcd for C₃₉H₅₆O₃S₂Si₂: 692.3210; found: 731.2828 (M +

K).

SO₃solution (10 mL). The solution was poured into H₂O (10 mL)

and extracted with EtOAc (8 × 10 mL). The combined organic layer

was dried over MgSO₄, filtered, and concentrated.

Methyltriphenylphosphonium bromide (1.98 g, 5.54 mmol) in THF

(50 mL) at 0 °C was treated with lithium bis(trimethylsilyl)amide (5.04

mL, 1 M in THF, 5.04 mmol), and the resulting solution was stirred at

0 °C for 30 min. The aldehyde (2.00 g, 2.52 mmol), prepared above,

in THF (5.0 mL) was added, and the mixture was allowed to warm to

room temperature and stirred at this temperature for 1 h. The reaction

was quenched with saturated aqueous NH₄Cl (15 mL) and extracted

with Et₂O (3 × 20 mL). The combined Et₂O fractions were washed

with brine (15 mL), dried over MgSO₄, filtered, and concentrated. The

residue was purified by flash chromatography on silica gel eluting with

5% Et₂O in hexanes to afford compound 67 (1.42 g, 76%) as a colorless

foam.

To a solution of the crude diol in THF/H₂O (4 mL, 3:1) was added

NaIO₄(0.260 g; 1.22 mmol). After 1.25 h, the reaction mixture was

quenched with H₂O (10 mL) and concentrated. The aqueous layer was

extracted with EtOAc (5 × 10 mL) and dried over MgSO₄. Flash

chromatography (SiO₂, EtOAc/hexanes, 1:1) on a short pad of silica

gave the crude aldehyde 78 as a yellow oil (0.080 g) which contained

unidentified byproduct(s). This mixture was used without further

purification.

To a solution of (Ph₃P⁺CH₂I)I^-(0.100 g; 0.19 mmol) in THF (0.25

mL) at rt was added sodium bis(trimethylsilyl)amide (1 M soln in THF,

0.15 mL, 0.15 mmol). To the resulting solution at -78 °C were added

HMPA (0.022 mL; 0.13 mmol) and the crude aldehyde 78 from the

previous step (0.016 g) in THF (0.25 mL). The reaction mixture was

then stirred at rt for 30 min. After the addition of saturated aqueous

NH₄Cl (10 mL), and the solution was extracted with EtOAc (4 × 10

mL). The combined organic extracts were dried (MgSO₄), filtered,

and concentrated. The residue was purified by preparative thin-layer

chromatography (prep-TLC) (EtOAc/hexanes, 2:3) to give the vinyl

iodide 75 as a yellow oil (0.014 g; 50% for three steps).

A solution of the dithiane 67 (1.0 g, 1.34 mmol), prepared above,

in MeOH/THF (2:1, 13 mL) was treated with [bis(trifluoroacetoxy)-

iodobenzene] (0.865 g, 2.01 mmol) at rt. After 15 min, the reaction

was quenched with saturated aqueous NaHCO₃(25 mL). The mixture

was extracted with Et₂O (3 × 25 mL), and the combined Et₂O fractions

were washed once with brine (20 mL), dried over MgSO₄, filtered,

and concentrated. Purification of the residue by flash chromatography

on silica gel eluting with 5% Et₂O in hexanes provided compound 68

(0.865 g, 92%) as a colorless foam: [R]_D) +1.74 (c ) 0.77, CHCl₃);

Preparation of Compound 80. The acetal 68 (0.930 g, 1.44 mmol)

was dissolved in dioxane/H₂O (9:1, 20 mL), and pTSA‚H₂O (0.820 g,

4.32 mL) was added. The mixture was heated at 55 °C for 2 h. After

cooling to rt, the solution was poured into Et₂O (200 mL) and washed

once with saturated aqueous NaHCO₃solution (30 mL) and once with

brine (30 mL) and dried over anhydrous MgSO₄. Purification by flash

chromatography on silica gel eluting with hexanes/ethyl acetate (9:1)

gave 0.702 g (81%) of the aldehyde 80 as a white foam: [R]_D) -12.8

(c ) 3.4, CHCl₃); IR (film) 2929, 1722, 1472, 1429, 1256, 1115, 1059

1

IR (film) 1428, 1252, 1114, 1075, 1046 cm^-1; H NMR (CDCl₃, 500

MHz) δ 7.61 (dd, J ) 7.9, 1.4 Hz, 6 H), 7.38 (s, 9 H), 5.47 (m, 1 H),

4.87 (d, J ) 10.0 Hz, 1 H), 4.76 (d, J ) 15.9 Hz, 1 H), 4.30 (d, J )

3.7 Hz, 1 H), 3.95 (s, 1 H), 3.56 (dd, J ) 7.5, 1.4 Hz, 1 H), 3.39 (s,

3 H), 2.84 (s, 3 H), 2.02 (m, 1 H), 1.64 (m, 2 H), 1.34 (m, 1 H), 1.11

(s, 3 H), 1.02 (d, J ) 7.4 Hz, 3 H), 0.90 (s, 3 H), 0.85 (s, 9 H), 0.62

(d, J ) 6.8 Hz, 3 H), -0.04 (s, 3 H), -0.05 (s, 3 H); ¹³C NMR (CDCl₃,

125 MHz) δ 138.3, 135.8, 135.0, 129.9, 127.8, 114.9, 110.5, 60.1, 55.6,

46.5, 43.9, 36.8, 34.2, 26.3, 19.6, 18.6, 17.1, 16.16, 13.9, -2.9, -3.8;

HRMS calcd for C₃₉H₅₈O₄Si₂: 646.3873; found: 685.3491 (M + K).

1

cm^-1; H NMR (CDCl₃, 400 MHz) δ 9.76 (s, 1 H), 7.71 (d, J ) 6.4

Hz, 6 H), 7.42 (m, 9 H), 5.49 (m, 1 H), 4.87 (m, 1 H), 4.75 (m, 1 H),

4.08 (d, J ) 1.6 Hz, 1 H), 3.56 (dd, J ) 1.6, 8.7 Hz, 1 H), 2.18 (m, 1

H), 1.70 (m, 1 H), 1.46 (m, 2 H), 1.10 (s, 3 H), 0.89 (d, J ) 6.4 Hz,

3 H), 0.79 (s 9 H), 0.60 (d, J ) 6.5 Hz, 3 H), -0.8 (s, 3 H), -0.83 (s,

3 H); ¹³C NMR (CDCl₃, 100 MHz) δ 205.5, 137.6, 135.6, 134.2, 130.0,

127.9, 115.4, 80.7, 76.4, 51.7, 43.5, 38.2, 34.9, 26.1, 20.9, 20.1, 18.5,

15.5, 12.9, -3.3, -3.9; HRMS calcd for C₃₇H₅₂O₃S₂: 600.3455;

found: 623.3333 (M + Na).

Preparation of Compound 76. A mixture of (S)-(-)-1,1′-bi-2-

naphthol (0.259 g, 0.91 mmol), Ti(O-i-Pr)₄(261 mL; 0.90 mmol), and

4 Å sieves (3.23 g) in CH₂Cl₂(16 mL) was heated at reflux for 1 h.

The mixture was cooled to rt, and aldehyde 10 was added. After 10

min, the suspension was cooled to -78 °C, and allyltri-n-butyltin (3.60

mL, 11.6 mmol) was added. The reaction mixture was stirred for 10

min at -78 °C and then placed in a -20 °C freezer for 70 h. Saturated

aqueous NaHCO₃solution (2 mL) was added, and the mixture was

stirred for 1 h, poured over Na₂SO₄, and then filtered through a pad of

MgSO₄and Celite. The crude material was purified by flash chroma-

tography (hexanes/ethyl acetate, 1:1) to give alcohol 76 as a yellow

oil (1.11 g, 60%): [R]_D) -15.9 (c 4.9, CHCl₃); IR (film) 3360, 1641,

Preparation of Compound 81. The aldehyde 80 (0.702 g, 1.17

mmol) was dissolved in THF (50 mL), and tert-butyl acetate (1.26 mL,

9.36 mmol) was added. The solution was cooled to -78 °C, and LDA

(2.0 M soln, 3.51 mL, 7.02 mmol) was added. After 20 min, the

reaction was quenched with MeOH (10 mL) and H₂O (100 mL). The

mixture was extracted with Et₂O (3 × 100 mL). The combined organics

were washed once with brine (30 mL) and dried over anhydrous MgSO₄.

The crude mixture contained a 2:1 ratio (81:82) of diastereomers.

Purification was done by flash chromatography on silica gel eluting

with hexanes/ethyl acetate (19:1) to give 0.527 g (63%) of the desired

isomer 81 as a white foam: [R]_D) -11.2 (c ) 1.4, CHCl₃); IR (film)

1

1509, 1434, 1188, 1017, 914 cm^-1; H NMR (500 MHz, CDCl₃, 25

°C) δ 6.92 (s, 1 H), 6.55 (s, 1 H), 5.82 (m, 1 H), 5.13 (dd, J ) 17.1,

1.3 Hz, 1 H), 5.09 (d, J ) 10.2 Hz, 1 H), 4.21 (t, J ) 6.0 Hz, 1 H),

2.76 (br s, 1 H), 2.69 (s, 3 H), 2.40 (m, 2 H), 2.02 (s, 3 H); ¹³C NMR

(125 MHz, CDCl₃, 25 °C) δ 164.5, 152.6, 141.5, 134.6, 119.2, 117.6,

115.3, 76.4, 39.9, 19.0, 14.2; HRMS calcd for C₁₁H₁₅NOS: 209.0874

found: 209.0872 (M + H).

1

3493, 2929, 1710, 1429, 1153, 1115, 1045 cm^-1; H NMR (CDCl₃,

400 MHz) δ 7.35 (dd, J ) 1.2, 7.6 Hz, 6 H), 7.38 (m, 9 H), 5.50 (m,

1 H), 4.86 (d, J ) 9.5 Hz, 1 H), 4.73 (d, J ) 17.1 Hz, 1 H), 4.09 (d,

J ) 3.5 Hz, 1 H), 3.93 (br d, J ) 10.0 Hz, 1 H), 3.75 (d, J ) 7.1 Hz,

1 H), 3.34 (d, J ) 2.6 Hz, 1 H), 2.29 (dd, J ) 2.5, 16.4 Hz, 1 H), 2.18

(m, 2 H), 1.67 (m, 2 H), 1.44 (m, 1 H), 1.41 (s, 9 H), 1.05 (d, J ) 7.4

Hz, 3 H), 0.95 (s, 3 H), 0.86 (s, 12 H), 0.64 (d, J ) 6.7 Hz, 3 H),

-0.03 (s, 3 H), -0.05 (s, 3 H); ¹³C NMR (CDCl₃, 100 MHz) δ 173.2,

138.0, 136.7, 134.8, 129.8, 127.6, 115.0, 81.2, 78.8, 75.9, 71.3, 44.4,

43.5, 38.0, 37.2, 34.9, 28.0, 26.3, 21.2, 20.5, 18.8, 16.0, 14.1, -3.0,

-4.1; HRMS calcd for C₄₃H₆₄O₅Si₂: 572.4293; found: 573.4390 (M

+ H).

Preparation of Compound 83. The alcohol 81 (0.110 g, 0.0153

mmol) was treated with pyridine buffered HF‚pyridine solution (3.0

mL) (stock solution was prepared from 20 mL of THF, 11.4 mL of

pyridine, and 4.2 g of hydrogen fluoride-pyridne (Aldrich Co.)) at rt

and stirred for 2 h. The reaction mixure was poured into saturated

aqueous NaHCO₃(50 mL) and extracted with ether (3 × 50 mL). The

organic layer was washed in sequence with saturated aqueous CuSO₄

(3 × 10 mL) and saturated aqueous NaHCO₃(10 mL) and then dried

over Na₂SO₄and concentrated in vacuo. The residue was purified by

Preparation of Compound 77. To a solution of alcohol 76 (0.264

g; 1.26 mmol) in CH₂Cl₂(12 mL) were added 4-DMAP (0.015 g, 0.098

mmol), Et₃N (0.45 mL; 3.22 mmol), and Ac₂O (0.18 mL; 1.90 mmol).

After 2 h, the reaction was quenched by the addition of H₂O (20 mL)

and extracted with EtOAc (4 × 20 mL). The combined organic extracts

were dried with MgSO₄, filtered, and concentrated. Flash chromatog-

raphy on SiO₂(EtOAc/hexanes, 1:3) afforded acetate 77 as a yellow

oil (0.302 g; 96%): [R]_D) -40.0 (c 7.3, CHCl₃); IR (film) 1738,

1

1505, 1436, 1370, 1236, 1019 cm^-1; H NMR (500 MHz, CDCl₃, 25

°C) δ 6.95 (s, 1 H), 6.52 (s, 1 H), 5.72 (m, 1 H), 5.33 (t, J ) 6.6 Hz,

1 H), 5.10 (ddd, J ) 17.1, 3.1, 1.5 Hz, 1 H), 5.07 (ddd, J ) 10.2, 3.3,

1.7 Hz, 1 H), 2.70 (s, 3 H), 2.48 (dt, J ) 5.9, 1.3 Hz, 2 H), 2.08 (s, 3

H), 2.07 (s, 3 H); ¹³C NMR (125 MHz, CDCl₃, 25 °C) δ 170.1, 164.5,

152.4, 137.0, 133.4, 120.6, 117.6, 116.2, 77.9, 37.5, 21.1, 19.1, 14.7;

HRMS calcd for C₁₃H₁₇O₂NS: 251.0980; found: 251.0983 (M⁺).

Preparation of Compound 75. To a solution of acetate 77 (0.099

g; 0.39 mmol) in acetone (10 mL) at 0 °C were added H₂O (4 drops),

OsO₄(2.5% wt in butyl alcohol; 0.175 mL; 0.018 mmol), and

N-methylmorpholine N-oxide (0.069 g; 0.59 mmol). The mixture was

stirred at 0 °C for 2 h and then quenched with saturated aqueous Na₂-

10088 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

flash chromatography on silica gel eluting with hexanes/ethyl acetate

(7:1) to give the corresponding diol (0.070 g, 98%) as a white foam.

23.5, 21.9, 21.2, 19.2, 18.9, 18.4, 18.1, 17.5, 15.7, 14.9, -3.7, -3.9,

-4.3, -4.7; HRMS calcd for C₄₀H₇₁O₆NSSi₂: 765.4490; found:

766.4571 (M + H).

The diol (0.231 g, 0.48 mmol) was dissolved in CH₂Cl₂(5.0 mL)

and cooled to -30 °C. 2,6-Lutidine (0.168 mL, 1.44 mmol) was added

followed by TBSOTf (0.131 mL, 0.570 mmol). After 1 h at -30 °C,

the reaction was poured into Et₂O (300 mL), washed once with 1 N

HCl (50 mL), once with saturated aqueous NaHCO₃(50 mL), and once

with brine (30 mL), and dried over anhydrous MgSO₄. Purification

by flash chromatography on silica gel eluting with hexanes/ethyl acetate

(20:1) gave alcohol 83 (0.276 g, 96%) as a clear oil: [R]_D) -5.4 (c

) 0.9, CHCl₃); IR (film) 3466, 2930, 1728, 1462, 1369, 1254, 1156

Preparation of Compound 88. The acetate 87 (0.035 g, 0.044

mmol) was dissolved in MeOH/H₂O (2:1, 1.5 mL), and K₂CO₃(0.050

g) was added. After 3 h, the reaction was diluted with saturated aqueous

NH₄Cl (5.0 mL) and extracted with CHCl₃(5 × 10 mL). The hydroxy

acid 88 was purified by flash chromatography on silica gel eluting with

hexanes/ethyl acetate (4:1 f 2:1) to give the pure hydroxy acid 88

(0.030 g, 84%): [R]_D) -19.8 (c ) 16.5, CHCl₃); IR (film) 3600-

2450, 1710, 1700, 1472, 1253, 988 cm^-1; ¹H NMR (CDCl₃, 400 MHz)

δ 6.95 (s, 1 H), 6.64 (s, 1 H), 5.55 (m, 1 H), 5.42 (m, 1 H), 4.39 (dd,

J ) 3.6, 6.4 Hz, 1 H), 4.15 (t, J ) 6.7 Hz, 1 H), 3.79 (dd, J ) 1.7, 6.9

Hz, 1 H), 3.13 (m, 1 H), 2.71 (s, 3 H), 2.49 (dd, J ) 3.6, 16.4 Hz, 1

H), 2.40 (m, 2 H), 2.31 (dd, J ) 6.5, 16.4 Hz, 1 H), 2.11 (m, 1 H),

2.01 (s, 3 H), 1.38 (m, 3 H), 1.20 (s, 3 H), 1.15 (m, 2 H), 1.13 (m, 3

H), 1.12 (s, 3 H), 1.05 (d, J ) 6.8 Hz, 3 H), 0.91 (d, J ) 6.7 Hz, 3 H),

0.87 (s, 9 H), 0.10 (s, 3 H), 0.06 (s, 6 H), 0.05 (s, 3 H); ¹³C NMR

(CDCl₃, 100 MHz) δ 218.0, 176.2, 165.0, 152.5, 141.8, 133.2, 124.9,

118.9, 115.2, 73.5, 53.6, 44.9, 40.1, 38.9, 33.3, 30.8, 28.0, 27.9, 26.2,

26.0, 23.5, 19.2, 18.9, 18.5, 18.2, 17.4, 15.9, 14.5, -3.7, -3.8, -4.2,

-4.6; HRMS calcd for C₃₈H₆₉O₆NSSi₂: 723.4384; found: 746.4285

(M + Na).

1

cm^-1; H NMR (CDCl₃, 400 MHz) δ 5.76 (m, 1 H), 4.99 (m, 2 H),

4.15 (t, J ) 4.4 Hz, 1 H), 3.91 (d, J ) 4.0 Hz, 1 H), 3.60 (bs, 1 H),

3.75 (dd, J ) 3.3, 7.8 Hz, 1 H), 2.81 (dd, J ) 4.7, 17.3 Hz, 1 H), 2.41

(m, 1 H), 2.22 (dd, J ) 4.2, 17.3 Hz, 1 H), 1.88 (m, 1 H), 1.73 (m, 2

H), 1.45 (s, 9 H), 0.94 (d, J ) 7.0 Hz, 3 H), 0.92 (s, 3 H), 0.90 (s, 9

H), 0.89 (s, 12 H), 0.85 (s, 3 H), 0.08 (s, 3 H), 0.07 (s, 3 H), 0.06 (s,

3 H), 0.05 (s, 3H, s); ¹³C NMR (CDCl₃, 100 MHz) δ 172.1, 138.2,

115.5, 80.8, 77.7, 74.4, 46.3, 43.7, 40.9, 38.7, 37.9, 39.7, 28.1, 26.2,

26.0, 19.5, 19.2, 18.5, 18.1, 16.9, 15.1, -3.9, -4.1, -4.2, -4.9; HRMS

calcd for C₃₁H₆₄O₅Si₂: 572.4293; found: 573.4390 (M + H).

Preparation of Compound 85. The alcohol 83 (0.275 g, 0.46

mmol) was dissolved in CH₂Cl₂(5.0 mL), and Dess-Martin periodinane

(0.292 g, 0.690 mmol) was added. After 2 h, a 1:1 mixture of saturated

aqueous NaHCO₃/saturated aqueous Na₂S₂O₃(2.0 mL) was added. After

10 min, the mixture was poured into Et₂O (40 mL), and the organic

layer was washed with brine (3.0 mL) and dried over anhydrous MgSO₄.

Purification by flash chromatography on silica gel eluting with hexanes/

ethyl acetate (19:1) gave ketone 84 (0.244 g, 89%) as a clear oil.

Preparation of Compound 89. To a solution of the the olefin 68

(0.680 g, 1.07 mmol) in THF (8.0 mL) was added 9-BBN (0.5 M soln

in THF, 2.99 mL, 1.50 mmol). In a separate flask, the iodide 75 (0.478

g, 1.284 mmol) was dissolved in DMF (10.0 mL). Cs₂CO₃(0.696 g,

2.14 mmol) was then added with vigorous stirring followed by

sequential addition of Ph₃As (0.034 g, 0.111 mmol), PdCl₂(dppf)₂(0.091

g, 0.111 mmol), and H₂O (0.693 mL, 38.5 mmol). After 4 h, the borane

solution was added to the iodide mixture in DMF. The mixture quickly

turned dark brown in color and slowly became pale yellow after 2 h.

The reaction was then poured into H₂O (100 mL) and extracted with

Et₂O (3 × 50 mL). The combined organics were washed with H₂O (2

× 50 mL) and once with brine (50 mL) and dried over anhydrous

MgSO₄. Purification by flash chromatography on silica gel eluting with

hexanes/EtOAc (7:1) gave 0.630 g (75%) of the coupled product 86 as

a pale yellow oil. This compound could not be separated completely

from residual borane impurities and was taken forward in impure form.

The acetate 86 (0.610 g, 0.770 mmol) was dissolved in dioxane/

H₂O (9:1, 15 mL), and pTSA‚H₂O (0.442 g, 2.32 mmol) was added.

The mixture was then heated to 55 °C. After 3 h, the mixture was

cooled to rt and poured into Et₂O. This solution was washed once

with saturated NaHCO₃(30 mL) and once with brine (30 mL) and

dried over anhydrous MgSO₄. Purification by flash chromatography

on silica gel eluting with hexanes/EtOAc (7:1) gave 0.486 g (85%) of

the aldehyde 89 as a pale yellow oil: [R]_D) -18.7 (c ) 0.53, CHCl₃);

The olefin 84 (0.420 g, 0.76 mmol) was dissolved in CH₂Cl₂(10

mL) and treated successively with 2,6-lutidine (1.75 mL, 15 mmol)

and TBSOTf (1.72 mL, 7.5 mmol). After 7 h, the reaction was poured

into Et₂O (150 mL), washed successively with 0.2 N HCl (25 mL) and

brine (20 mL), and dried over anhydrous MgSO₄. The residue was

purified by flash chromatography on a short pad of silica gel with fast

elution with hexanes/ethyl acetate (20:1) to give TBS ester 85 (0.611

g, 93%) as a clear oil. The purification must be done quickly to avoid

hydrolysis of the silyl ester: [R]_D) -35.4 (c ) 0.4, CHCl₃); IR (film)

2930, 1730, 1692, 1472, 1367, 1253, 1155, 1084 cm^-^{1 1}H NMR

;

(CDCl₃, 400 MHz) δ 5.73 (m, 1 H), 4.97 (m, 2 H), 4.33 (dd, J ) 3.8,

5.5 Hz, 1 H), 3.78 (dd, J ) 1.9, 7.3 Hz, 1 H), 3.19 (m, 1 H), 2.53 (dd,

J ) 3.8, 17.3 Hz, 1 H), 2.25 (m, 2 H), 1.85 (m, 1 H), 1.40 (m, 1 H),

1.24 (s, 3 H), 1.07 (s, 3 H), 1.04 (d, J ) 6.8 Hz, 3 H), 0.92 (s, 12 H),

0.91 (s, 9 H), 0.87 (s, 9 H), 0.26 (s, 3 H), 0.26 (s, 3 H), 0.10 (s, 3 H),

0.06 (s, 6 H), 0.05 (s, 3 H); ¹³C NMR (CDCl₃, 100 MHz) δ 217.1,

170.6, 137.2, 115.0, 79.8, 77.1, 73.2, 52.9, 45.0, 40.8, 37.5, 34.7, 27.5,

25.6, 25.4, 22.4, 19.7, 17.9, 17.6, 17.3, 14.9, -4.2, -4.3, -4.9, -5.4.

1

IR (film) 1737, 1429, 1237, 1115, 1053 cm^-1; H NMR (CDCl₃, 500

Preparation of Compound 87. To a solution of olefin 85 (0.053

g, 0.081 mmol) in THF (0.8 mL) was added 9-BBN (0.5 M soln in

THF, 0.323 mL, 0.162 mmol). In a separate flask, the iodide 75 (0.036

g, 0.097 mmol) was dissolved in DMF (1.0 mL). Cs₂CO₃(0.053 g,

0.162 mmol) was then added with vigorous stirring followed by

sequential addition of Ph₃As (0.0025 g, 0.0081 mmol), PdCl₂(dppf)₂

(0.0067 g, 0.0081 mmol), and H₂O (0.052 mL, 2.91 mmol). After 4

h, the borane in THF was added to the iodide mixture in DMF. The

reaction quickly turned dark brown in color and slowly became pale

yellow after 2 h. The reaction was then poured into saturated aqueous

NH₄Cl (10.0 mL) and extracted with CHCl₃(3 × 30 mL). The

combined organics were washed with H₂O (2 × 50 mL) and once with

brine (50 mL) and dried over anhydrous MgSO₄. Purification by flash

chromatography on silica gel eluting with hexanes/ethyl acetate (4:1

f 3:1) gave 0.036 g (56%) of the coupled product 87 as a pale yellow

oil: [R]_D) -29.2 (c ) 0.3, CHCl₃); IR (film) 3500-2600, 1738, 1710,

MHz) δ 9.74 (s, 1 H), 7.61 (dd, J ) 7.8, 1.2 Hz, 6 H), 7.38 (m, 9 H),

6.94 (s, 1 H), 6.53 (s, 1 H), 5.39 (m, 1 H), 5.31 (m, 1 H), 5.29 (t, J )

6.9 Hz, 1 H), 4.61 (d, J ) 4.3 Hz, 1 H), 3.50 (dd, J ) 5.2, 2.6 Hz, 1

H), 2.70 (s, 3 H), 2.48 (m, 2 H), 2.14 (m, 1 H), 2.09 (s, 3 H), 2.07 (s,

3 H), 1.83 (m, 2 H), 1.41 (m, 1 H), 1.18 (m, 1 H), 1.01 (s, 3 H), 0.99

(s, 3 H), 0.91 (d, J ) 7.4 Hz, 3 H), 0.85 (s, 9 H), 0.69 (m, 1 H), 0.58

(d, J ) 6.8 Hz, 3 H), -0.05 (s, 3 H), -0.06 (s, 3 H); ¹³C NMR (CDCl₃,

125 MHz) δ 205.46, 170.01, 164.49, 152.46, 137.10, 135.60, 134.22,

132.55, 130.65, 127.84, 123.82, 120.66, 116.19, 81.09, 78.47, 76.73,

51.66, 43.14, 38.98, 30.99, 30.42, 27.63, 26.10, 21.15, 20.92, 20.05,

19.15, 18.49, 15.12, 14.70, 12.75, -3.25, -4.08; HRMS calcd for

C₅₀H₆₉O₅NSSi₂: 851.4435; found: 890.4100 (M + K).

Preparation of Compounds 90 and 91. To a solution of the

acetate-aldehyde 89 (0.084 g, 0.099 mmol) in THF (99 mL) at -78

°C was added potassium bis(trimethylsilyl)amide (0.5 M in toluene,

1.0 mL, 0.5 mmol) dropwise. The resulting solution was stirred at

-78 °C for 30 min. Then the reaction mixure was transfered via

cannula to a short pad of silica gel and washed with Et₂O. The residue

was purified by flash chromatography (silica, 12% EtOAc in hexane)

to give the 3S product 90 and the 3R product 91 in a 6:1 ratio in 51%

combined yield:

1

1691, 1236, 988 cm^-1; H NMR (CDCl₃, 400 MHz) δ 6.95 (s, 1 H),

6.59 (s, 1 H), 5.49 (m, 1 H), 5.31 (m, 1 H), 5.25 (dd, J ) 5.7, 7.7 Hz,

1 H), 4.38 (dd, J ) 3.8, 6.5 Hz, 1 H), 3.78 (dd, J ) 1.9, 6.7 Hz, 1 H),

3.13 (m, 1 H), 2.71 (s, 3 H), 2.49 (m, 2 H), 2.42 (m, 1 H), 2.31 (dd,

J ) 6.5, 16.3 Hz, 1 H), 2.07 (s, 3 H), 2.04 (s, 3 H), 1.95 (m, 1 H), 1.89

(m, 2 H), 1.48 (m, 3 H), 1.21 (s, 3 H), 1.15 (m, 2 H), 1.12 (s, 3 H),

1.06 (d, J ) 6.9 Hz, 3 H), 0.90 (s, 12 H), 0.88 (s, 9 H), 0.10 (s, 3 H),

0.07 (s, 3 H), 0.06 (s, 3 H), 0.05 (s, 3 H); ¹³C NMR (CDCl₃, 100 MHz)

δ 217.9, 176.7, 170.2, 164.9, 152.2, 137.6, 132.7, 123.9, 120.3, 115.9,

78.4, 73.6, 53.5, 45.0, 40.1, 38.8, 31.0, 30.6, 27.9, 27.7, 26.2, 26.0,

Compound 90: [R]_D) -39.4 (c 0.52, CHCl₃); IR (film) 3508,

1

1733, 1428, 1254, 1113, 1034 cm^-1; H NMR (500 MHz, C₆D₆, 60

°C) δ 7.85 (dd, J ) 7.3, 2.1 Hz, 6 H), 7.22 (m, 9 H), 6.54 (s, 1 H),

6.49 (s, 1 H), 5.53 (d, J ) 6.0 Hz, 1 H), 5.42 (m, 2 H), 4.22 (d, J )

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10089

5.9 Hz, 1 H), 4.19 (d, J ) 3.5 Hz, 1 H), 4.17 (br s, 1 H), 2.58 (m, 1

H), 2.45 (m, 2 H), 2.31 (s, 3 H), 2.29 (m, 2 H), 2.13 (s, 3 H), 2.00 (m,

2 H), 1.81 (m, 1 H), 1.56 (m, 1 H), 1.35-1.28 (band, 2 H), 1.24 (d, J

) 7.1 Hz, 3 H), 1.20 (m, 1 H), 1.11 (s, 3 H), 1.07-0.92 (band, 13 H),

0.88 (d, J ) 6.7 Hz, 3 H), 0.12 (s, 6 H); ¹³C NMR (125 MHz, C₆D₆,

60 °C) δ 171.1, 164.3, 153.7, 137.4, 136.6, 136.0, 130.1, 128.5, 125.5,

120.2, 116.7, 78.5, 75.8, 72.8, 44.6, 41.6, 38.3, 32.7, 31.8, 30.1, 28.5,

27.9, 26.6, 22.2, 21.3, 18.9, 18.8, 15.8, 15.7, 1.30, -2.76, -3.66; HRMS

calcd for C₅₀H₆₉O₅NSSi₂: 851.4435; found: 852.4513 (M + H).

500 MHz) δ 6.97 (s, 1 H), 6.57 (s, 1 H), 5.53 (dt, J ) 3.4, 11.1 Hz, 1

H), 5.37 (dd, J ) 16.4, 9.9 Hz, 1 H), 5.00 (d, J ) 10.3 Hz, 1 H), 4.02

(d, J ) 9.7 Hz, 1 H), 3.89 (d, J ) 8.7 Hz, 1 H), 3.00 (m, 1 H), 2.82

(d, J ) 6.5 Hz, 1 H), 2.71 (m, 5H), 2.36 (q, J ) 10.7 Hz, 1 H), 2.12

(, 3 H), 2.07 (dd, J ) 8.2 Hz, 1 H), 1.87 (bs, 1 H), 1.49 (m, 3 H), 1.19

(m, 5H), 1.14 (s, 3 H), 1.08 (d, J ) 6.8 Hz, 3 H), 0.94 (m, 12 H), 0.84

(s, 9 H), 0.12 (s, 3 H), 0.10 (s, 3 H), 0.07 (s, 3 H), -0.09 (s, 3 H); ¹³

C

NMR (CDCl₃, 125 MHz) δ 218.7, 170.1, 164.5, 152.6, 137.9, 133.9,

124.8, 119.6, 115.9, 72.7, 53.2, 43.9, 41.0, 40.3, 32.9, 32.3, 28.4, 27.1,

26.3, 26.1, 26.0, 19.2, 19.1, 18.3, 18.2, 17.1, 16.0, 15.2, 14.3, -4.2,

-4.4, -4.6, -4.8; HRMS calcd for C₃₈H₆₇O₅NSSi₂: 705.4315,

found: 706.4357 (M + H).

Macrolactonization To Produce Compound 95. To a solution of

hydroxy acid 88 (0.094 g, 0.133 mmol) in THF (1 mL) were added

Et₃N (0.11 mL, 0.79 mmol) and 2,4,6-trichlorobenzoyl chloride (0.104

mL, 0.66 mmol). The mixture was stirred at rt for 0.25 h, diluted with

toluene (15 mL), and added dropwise over a period of 3 h to a solution

of DMAP (0.167 mg, 1.37 mmol) in toluene (50 mL). After complete

addition, the mixture was stirred for additional 0.5 h and concentrated

in vacuo. Purification of the residue by flash chromatography on silica

gel eluting with hexanes/ethyl acetate (9:1) gave 0.081 g (88%) of the

previously described lactone 95.

Compound 91: [R]_D) -53.9 (c 0.37, CHCl₃); IR (film) 2927,

1

1734, 1428, 1114, 1036 cm^-1; H NMR (500 MHz, C₆D₆, 60 °C) δ

7.82 (m, 6 H), 7.22-7.19 (band, 9 H), 6.59 (s, 1 H), 6.53 (s, 1 H),

5.58-5.53 (band, 2 H), 5.49 (m, 1 H), 4.39 (d, J ) 9.7 Hz, 1 H), 3.98

(d, J ) 4.7 Hz, 1 H), 3.86 (dd, J ) 1.9, 5.6 Hz, 1 H), 3.59 (br s, 1 H),

2.49-2.43 (band, 3 H), 2.34-2.26 (band, 6 H), 2.18 (s, 3 H), 1.98 (m,

2 H), 1.51 (m, 1 H), 1.50-1.30 (band, 2 H), 1.21-1.19 (band, 6 H),

1.03-0.99 (band, 10H), 0.85 (d, J ) 5.4 Hz, 3 H), 0.78 (s, 3 H), 0.10

(s, 3 H), 0.07 (s, 3 H); HRMS calcd for C₅₀H₆₉O₅NSSi₂: 851.4435;

found: 852.4489 (M + H).

Preparation of Compound 93. The lactone 90 (0.032 g, 0.0376

mmol) was treated with pyridine-buffered HF‚pyridine solution (1 mL)

(stock solution was prepared from 20 mL of THF, 11.4 mL of pyridine,

and 4.2 g of hydrogen fluoride-pyridne (Aldrich Co.)) at room

temperature for 2 h. The reaction mixure was poured into saturated

aqueous NaHCO₃(15 mL) and extracted with Et₂O (3 × 30 mL). The

organic layer was washed in sequence with saturated aqueous CuSO₄

(3 × 10 mL) and saturated aqueous NaHCO₃(10 mL) and then dried

over Na₂SO₄and concentrated in vacuo. The residue was purified by

flash chromatography (silica, 25% EtOAc in hexane) to give diol 93

(0.022 g, 99%) as a white foam: [R]_D) -111.7 (c ) 0.7, CHCl₃); IR

Preparation of Compound 96. To a solution of TBS ether 95

(0.027 g, 0.038 mmol) in THF (1.0 mL) at 25 °C in a plastic vial was

added dropwise HF‚pyridine (0.5 mL). The resulting solution was

allowed to stir at 25 °C for 2 h. The reaction mixture was diluted

with chloroform (2 mL) and very slowly added to saturated aqueous

NaHCO₃(20 mL). The mixture was extracted with CHCl₃(20 mL ×

3). The organic layer was dried (Na₂SO₄), filtered, and concentrated

in vacuo. Purification of the residue by flash chromatography (silica,

30% EtOAc in hexane) provided diol 96 (0.018 g, 99%) as white

foam: [R]_D) -84.7 (c ) 0.85, CHCl₃); IR (film): 3493, 2925, 1728,

1

(film) 3463, 2928, 1729, 1253 cm^-1; H NMR (C₆D₆, 500 MHz, 60

°C) δ 6.65 (s, 1 H), 6.57 (s, 1 H), 5.50 (dd, J ) 2.58, 9.42 Hz, 1 H),

5.39 (m, 2 H), 4.34 (s, 1 H), 4.07 (dd, J ) 2.2, 9.9 Hz, 1 H), 3.47 (t,

J ) 4.5 Hz, 1 H), 3.11 (br s, 1 H), 2.69 (m, 2 H), 2.57 (m, 2 H), 2.31

(s, 3 H), 2.18 (m, 6 H), 2.0 (m, 1 H), 1.78 (m, 1 H), 1.53 (m, 2 H),

1.27 (m, 1 H), 1.09 (d, J ) 4.5 Hz, 3 H), 0.98 (m, 15H), 0.91 (s, 3 H),

0.14 (s, 3 H), 0.13 (s, 3 H); ¹³C NMR (C₆D₆, 125 MHz, 60 °C) δ

171.4, 164.3, 153.8, 137.8, 133.2, 128.5, 125.8, 120.3, 116.6, 83.2,

78.7, 75.8, 73.9, 42.6, 40.9, 39.4, 35.5, 34.4, 32.3, 28.1, 26.3, 22.5,

22.4, 18.8, 18.6, 17.0, 16.5, 15.5, -3.6, -4.4; LRMS calcd for

C₃₂H₅₅O₅NSSi: 593.4; found: 616.3 (M + Na).

1

1689, 1249 cm^-1; H NMR (CDCl₃, 500 MHz) δ 6.96 (s, 1 H), 6.59

(s, 1 H), 5.44 (dt, J ) 4.3, 10.4 Hz, 1 H), 5.36 (dt, J ) 5.1, 10.2 Hz,

1 H), 5.28 (dd, J ) 1.7, 9.8 Hz, 1 H), 4.11 (d, J ) 7.2 Hz, 1 H), 3.74

(s, 1 H), 3.20 (d, J ) 4.5 Hz, 1 H), 3.14 (dd, J ) 2.2, 6.8 Hz, 1 H),

3.00 (s, 1 H), 2.69 (m, 4 H), 2.49 (dd, J ) 11.3, 15.1 Hz, 1 H), 2.35

(dd, J ) 2.5, 15.1 Hz, 1 H), 2.27 (m, 1 H), 2.05 (m, 1 H), 2.04 (s, 3

H), 2.01 (m, 1 H) 1.75 (m, 1 H), 1.67 (m, 1 H), 1.33 (m, 4 H), 1.21 (s,

1 H), 1.19 (m, 2 H), 1.08 (d, J ) 7.0 Hz, 3 H), 1.00 (s, 3 H), 0.93 (d,

J ) 7.1 Hz, 3 H); ¹³C NMR (CDCl₃, 125 MHz) δ 226.5, 176.5, 171.1,

158.2, 144.7, 139.6, 131.1, 125.7, 122.0, 84.6, 80.2, 78.6, 59.4, 47.9,

45.4, 44.6, 38.5, 37.9, 33.7, 33.6, 28.7, 25.1, 25.0, 21.9, 21.7, 19.6;

HRMS calcd for C₂₆H₃₉O₅NS 477.2549, found 478.2627 (M + H).

Preparation of Epothilone A (2). To a cooled (-50 °C) solution

of desoxyepothilone A (96) (0.009 g, 0.0189 mmol) in dry CH₂Cl₂(1

mL) was added freshly prepared 3,3-dimethyldioxirane (0.95 mL, 0.1

M in acetone). The resulting solution was allowed to warm to -30

°C for 2 h. A stream of nitrogen was then bubbled through the solution

to remove excess dimethyldioxirane. The residue was purified by flash

chromatography (silica, 40% EtOAc in hexane) and afforded epothilone

A (2) (0.0046 g, 49%) as colorless solid and 0.0003 g of the cis-epoxide

diastereomer: [R]_D) -41.6 (c ) 0.51, MeOH); IR (film): 3464, 2926,

1737, 1689, 978, 755 cm^-1; ¹H NMR (CD₂Cl₂, 500 MHz) δ 7.00 (s, 1

H), 6.56 (s, 1 H), 5.39 (dd, J ) 9.2, 2.0 Hz, 1 H), 4.16 (br d, J ) 10.0

Hz, 1 H), 3.73 (dd, J ) 8.6, 4.2 Hz, 1 H), 3.59 (br s, 1 H), 3.21 (m, 1

H), 2.99 (m, 1 H), 2.87 (m, 1 H), 2.68 (s, 3 H), 2.50-2.44 (band, 2

H), 2.38 (dd, J ) 15.0, 3.2 Hz, 1 H), 2.14-2.08 (band, 4 H), 1.86 (m,

1 H), 1.75-1.66 (band, 3 H), 1.55 (m, 1 H), 1.41 (m, 4 H), 1.35 (s, 3

H), 1.14 (d, J ) 6.9 Hz, 3 H), 1.06 (s, 3 H), 0.98 (d, J ) 7.0 Hz, 3 H);

¹³C NMR (CD₂Cl₂, 125 MHz) δ 220.2, 170.9, 165.6, 152.4, 138.1,

120.2, 116.7, 77.2, 74.9, 73.4, 57.8, 55.1, 53.7, 39.5, 36.8, 32.1, 30.9,

30.1, 27.7, 23.8, 22.0, 20.2, 19.3, 17.3, 15.6, 14.3; HRMS calcd for

C₂₆H₃₉O₆NS: 493.2498, found: 494.2578 (M + H).

Preparation of Compound 94. To a cooled (-30 °C) solution of

diol 93 (0.029 g, 0.048 mmol) and 2,6-lutidine (0.017 mL, 0.147 mmol)

in anhydrous CH₂Cl₂(1.0 mL) was added TBSOTf (0.015 mL, 0.065

mmol). The resulting solution was then stirred at -30 °C for 30 min.

The reaction was quenched with 0.5 M HCl (10 mL) and extracted

with Et₂O (15 mL). The ether layer was washed with saturated aqueous

NaHCO₃(5 mL), dried (Na₂SO₄) and concentrated in vacuo. Purifica-

tion of the residue by flash chromatogrphy (silica, 8% EtOAc in hexane)

afforded TBS ether 94 (0.032 g, 93%) as white foam: [R]_D) -21.7

1

(c 0.35, CHCl₃); IR (film) 3471, 2928, 1742, 1253, 1076 cm^-1; H

NMR (500 MHz, C₆D₆, 38 °C) δ 6.62 (s, 1 H), 6.53 (s, 1 H) 5.49-

5.46 (band, 3 H), 4.41 (br s, 1 H), 4.10 (br s, 1 H), 3.49 (br s, 1 H),

2.70-2.64 (band, 2 H), 2.44 (dd, J ) 16.1, 6.5 Hz, 1 H), 2.34 (d, J )

15.5 Hz, 1 H), 2.28 (s, 3 H), 2.22-2.15 (band, 5H), 2.02 (m, 1 H),

1.81 (m, 1 H), 1.68 (m, 1 H), 1.50 (m, 1 H), 1.34 (m, 1 H), 1.18 (s, 3

H), 1.14 (d, J ) 20.9 Hz, 3 H), 1.02-0.98 (band, 23 H), 0.90 (s, 3 H),

0.16-0.15 (band, 9 H), 0.10 (s, 3 H); ¹³C NMR (125 MHz, C₆D₆, 43

°C) δ 171.3, 164.2, 153.9, 137.9, 133.0, 127.9, 120.1, 116.6, 78.9, 75.1,

44.7, 41.0, 32.8, 31.9, 27.9, 27.6, 26.4, 26.2, 25.9, 21.7, 18.9, 18.51,

18.49, 17.4, 15.53, -3.4, -3.6, -4.3, -4.4; HRMS calcd for

C₃₈H₆₉O₅NSSi₂: 707.4435; found: 746.4062 (M + K).

Preparation of Compound 95. To a solution of alcohol 94 (0.030

g, 0.0424 mmol) in CH₂Cl₂(2.0 mL) at 25 °C was added Dess-Martin

periodinane (0.036 g, 0.0848 mmol) in one portion. The resulting

solution was then allowed to stir at 25 °C for 1.5 h. The reaction was

quenched by the addition of 1:1 saturated aqueous NaHCO₃:Na₂S₂O₃

(10 mL) and stirred for 5 min. The mixture was then extracted with

Et₂O (3 × 15 mL). The organic layer was dried (Na₂SO₄), filtered,

and concentrated in vacuo. Purification of the residue by flash

chromatography (silica, 8% EtOAc in hexane) provided ketone 95

(0.025 g, 84%) as white foam: [R]_D) -21.93 (c ) 1.4, CHCl₃); IR

Preparation of Compound 97. To a suspension of ethyltriphen-

ylphosphonium iodide (0.250 g, 0.60 mmol) in THF (6 mL) was added

nBuLi (2.5 M soln in hexanes, 0.24 mL, 0.60 mmol) at rt. After

disappearance of the solid material, the solution was added to a mixture

of iodine (0.152 g, 0.60 mmol) in THF (4 mL) at -78 °C. The resulting

suspension was vigorously stirred for 5 min at -78 °C and then warmed

to -20 °C and treated with sodium bis(trimethylsilyl)amide (1 M soln

in THF, 0.56 mL, 0.56 mmol). The resulting red solution was stirred

for 5 min followed by the slow addition of aldehyde 78 (0.074 g, 0.30

1

(film): 2928, 1745, 1692, 1254, 1175, 836 cm^-1; H NMR (CDCl₃,

10090 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

mmol). The mixture was stirred at -20 °C for 40 min, diluted with

pentane (50 mL), filtered through a pad of Celite, and concentrated in

vacuo. Purification of the residue by flash column chromatography

(hexanes/ethyl acetate, 85:15) gave 0.141 g (43% overall from acetate

77) of the vinyl iodide 97 as a yellow oil: [R]_D) -20.7 (c 2.45,

CHCl₃); IR (film) 2920, 1738, 1369, 1234 cm^-1; ¹H NMR (400 MHz,

CDCl₃) δ 6.98 (s, 1 H), 6.53 (s, 1 H), 5.43 (t, J ) 6.5, 5.4 Hz, 1 H),

5.35 (t, J ) 6.6, 6.5 Hz, 1 H), 2.71 (s, 3 H), 2.58-2.50 (band, 5H),

2.08 (s, 6 H); ¹³C NMR (100 MHz, CDCl₃) δ 170.0, 164.6, 152.4,

136.9, 130.3, 120.7, 120.6, 116.4, 103.6, 40.3, 33.7, 19.2, 19.1, 14.9;

HRMS calcd for C₁₄H₁₈O₂NIS: 391.0103; found: 392.0181 (M + H).

mL), and extracted with Et₂O (4 × 10 mL). The combined ether

solutions were washed with H₂O (1 × 30 mL), brine (1 × 30 mL),

dried with MgSO₄, filtered, and concentrated in vacuo. To a solution

of crude ketone 102 in MeOH/THF (2 mL, 1:1) at -78 °C was added

NaBH₄(0.017 g, 0.447 mmol). The resulting solution was stirred at rt

for 1 h, quenched with saturated aqueous NH₄Cl (10 mL), and extracted

with Et₂O (3 × 15 mL). The organic layers were dried with MgSO₄,

filtered, and concentrated. Flash column chromatography (hexanes/

ethyl acetate, 9:1) gave 0.052 g (67%) of the R-alcohol 100 as a pale

yellow oil and 0.009 g of â-alcohol 101.

Compound 100: [R]_D) -45.4 (c 1.0, CHCl₃); IR (film) 3490, 2929,

1

Preparation of Compound 98. To a solution of olefin 68 (0.977

g, 1.55 mmol) in THF (3 mL) was added 9-BBN (0.5 M soln in THF,

3.4 mL, 1.7 mmol). In a separate flask, iodide 97 (0.749 g, 1.92 mmol)

was dissolved in DMF (5 mL). Cs₂CO₃(1.154 g, 3.54 mmol) was

then added with vigorous stirring followed by sequential addition of

PdCl₂(dppf)₂(0.162 g, 0.198 mmol), Ph₃As (0.061 g, 0.20 mmol), and

H₂O (0.42 mL, 23.4 mmol). After 5 h, the borane solution was added

to the iodide mixture in DMF. The reaction quickly turned dark brown

in color and slowly became pale yellow after 3 h. The solution was

then poured into H₂O (10 mL) and extracted with Et₂O (3 × 15 mL).

The combined organic layers were washed with H₂O (3 × 15 mL),

brine (1 × 20 mL), dried over MgSO₄, filtered, and concentrated in

vacuo. Flash column chromatography (hexanes/ethyl acetate, 9:1) gave

the coupled product 98 (1.073 g; 77%) as a yellow oil: IR (film) 2931,

2856, 1729, 1428, 1114, 1037, 708 cm^-1; H NMR (500 MHz, C₆D₆,

60 °C) δ 7.85 (m, 6 H), 7.22 (m, 9 H), 6.55 (s, 1 H), 6.49 (s, 1 H),

5.56 (br d, J ) 6.9 Hz, 1 H), 5.25 (m, 1 H), 4.24 (br s, 1 H), 4.17 (m,

2 H), 2.60 (m, 1 H), 2.47 (m, 2 H), 2.29-2.22 (band, 6 H), 2.15 (s, 3

H), 2.08 (m, 2 H), 1.96 (m, 1 H), 1.84 (br s, 1 H), 1.68 (s, 3 H), 1.43

(m, 1 H), 1.32 (m, 2 H), 1.24 (d, J ) 7.1 Hz, 3 H), 1.11 (s, 6 H), 1.04

(s, 9 H), 0.91 (d, J ) 6.3 Hz, 3 H), 0.12 (s, 6 H); HRMS calcd for

C₅₁H₇₁O₅NSSi₂: 865.4591; found: 866.4716 (M + H).

Preparation of Compound 103. The silyl ether 100 (0.210 g, 0.247

mmol) was dissolved in HF‚pyridine/pyridine/THF (15 mL). The

solution was stirred at rt for 2 h and then diluted with Et₂O (1 mL),

poured into a mixture of Et₂O/saturated NaHCO₃(20 mL, 1:1), and

extracted with Et₂O (4 × 10 mL). The Et₂O extracts were washed

with saturated aqueous CuSO₄(3 × 30 mL), saturated aqueous NaHCO₃

(1 × 30 mL), and brine (1 × 30 mL), dried with MgSO₄, filtered, and

concentrated in vacuo. Flash column chromatography (hexanes/EtOAc,

9:1) gave diol 103 (0.141 g, 94%) as a pale yellow oil: [R]_D) -41.7

1

1738, 1429, 1239, 1072, 709 cm^-1; H NMR (400 MHz, CDCl₃) δ

7.41 (m, 15 H), 6.98 (s, 1 H), 6.59 (s, 1 H), 5.30 (t, J ) 6.9 Hz, 1 H),

5.15 (t, J ) 6.7 Hz, 1 H), 4.38 (d, J ) 3.5 Hz, 1 H), 4.02 (s, 1 H), 3.60

(m, 2 H), 3.41 (t, J ) 6.9 Hz, 2 H), 2.88 (s, 3 H), 2.79 (s, 3 H), 2.41

(m, 2 H), 2.16-2.10 (band, 6 H), 1.80 (m, 2 H), 1.70 (s, 3 H), 1.30-

1.01 (band, 6 H), 0.85 (s, 15H), 0.73 (d, J ) 6.8 Hz, 3 H), 0.62 (d, J

) 6.7 Hz, 3 H), 0.00 (br s, 6 H); ¹³C NMR (125 MHz, CDCl₃) δ 170.2,

164.5, 152.7, 138.6, 137.5, 135.8, 135.3, 129.9, 127.8, 120.6, 119.1,

116.2, 110.6, 110.5, 78.9, 77.8, 63.3, 60.4, 60.2, 55.5, 46.6, 43.9, 43.4,

38.0, 32.3, 31.7, 30.4, 26.2, 26.1, 26.0, 23.6, 21.3, 19.6, 19.4, 19.2,

18.6, 17.1, 15.8, 15.4, 14.8, 14.2, 13.6, -2.9, -3.9, -4.1; HRMS calcd

for C₅₃H₇₇O₆NSSi₂911.5010, found 950.4613 (M + K).

(c 0.65, CHCl₃); IR (film) 3460, 2928, 1728, 1252, 1037, 757 cm^-1

;

¹H NMR (500 MHz, C₆D₆, 58 °C) δ 6.69 (s, 1 H), 6.56 (s, 1 H), 5.51

(dd, J ) 9.4, 3.0 Hz, 1 H), 5.20 (m, 1 H), 4.30 (br s, 1 H), 4.10 (dd,

J ) 9.7, 2.6 Hz, 1 H), 3.50 (br s, 1 H), 3.07 (br s, 1 H), 2.74 (t, J )

9.6, 5.7 Hz, 1 H), 2.70 (dd, J ) 16.0, 3.5 Hz, 1 H), 2.52 (dd, J ) 16.0,

9.7 Hz, 1 H), 2.31-2.25 (band, 4 H), 2.24 (s, 3 H), 2.22-2.18 (band,

3 H), 1.91 (m, 1 H), 1.81 (m, 1 H), 1.61-1.58 (band, 4 H), 1.50 (m,

1 H), 1.37 (m, 2 H), 1.11 (d, J ) 6.9 Hz, 3 H), 1.04 (d, J ) 7.0 Hz,

3 H), 0.99 (s, 12 H), 0.93 (s, 3 H), 0.16 (s, 3 H), 0.14 (s, 3 H); ¹³C

NMR (125 MHz, C₆D₆, 58 °C) δ 171.3, 164.3, 153.9, 138.2, 137.6,

122.1, 120.4, 116.6, 82.8, 79.2, 73.8, 42.7, 39.5, 34.0, 32.9, 31.8, 26.3,

22.4, 21.9, 18.8, 18.6, 17.0, 15.5, -3.5, -4.5; LRMS calcd for

C₃₃H₅₇O₅NSSi 607.4, found 608.4 (M + H).

Preparation of Compound 99. The acetal 98 (0.069 g, 0.077

mmol) was dissolved in dioxane/H₂O (9:1, 1 mL), and pTSA‚H₂O

(0.045 g, 0.237 mmol) was added. The mixture was then heated to 55

°C. After 3 h, the mixture was cooled to rt, poured into saturated

aqueous NaHCO₃, and extracted with Et₂O (4 × 15 mL). The combined

ether extracts were washed with saturated aqueous NaHCO₃(1 × 30

mL), brine (1 × 30 mL), dried over MgSO₄, filtered, and concentrated.

Flash column chromatography (hexanes/EtOAc, 3:1) gave aldehyde 99

(0.046 g, 71%) as a pale yellow oil: [R]_D) -13.3 (c 0.95, CHCl₃);

Preparation of Compound 104. To a solution of diol 103 (0.0066

g, 0.011 mmol) in 0.5 mL of CH₂Cl₂at -78 °C were added 2,6-lutidine

(7 µL, 0.060 mmol) and TBSOTf (5 µL, 0.022 mmol). The resulting

solution was stirred at -30 °C for 0.5 h and then quenched with H₂O

(5 mL) and extracted with Et₂O (4 × 10 mL). The ether solutions

were washed with 0.5 M HCl (1 × 10 mL) and saturated aqueous

NaHCO₃(1 × 10 mL), dried over MgSO₄, filtered, and concentrated.

Flash column chromatography (hexanes/EtOAc, 93:7) gave alcohol 104

(0.0070 g, 89%) as a pale yellow oil: [R]_D) -2.6 (c 3.15, CHCl₃);

1

IR (film) 3070, 2929, 2856, 1737, 1429, 1238, 1116, 1056 cm^-1; H

NMR (500 MHz, CDCl₃) δ 9.79 (s, 1 H), 7.71-7.69 (band, 6 H), 7.52-

7.44 (band, 9 H), 7.01 (s, 1 H), 6.60 (s, 1 H), 5.31 (t, J ) 6.6 Hz, 1 H),

5.12 (t, J ) 6.9 Hz, 1 H), 4.85 (d, J ) 4.5 Hz, 1 H), 3.58 (br s, 1 H),

2.77 (s, 3 H), 2.47 (m, 2 H), 2.14 (s, 3 H), 2.13 (s, 3 H), 1.87 (m, 2 H),

1.72 (s, 3 H), 1.49 (m, 1 H), 1.30 (m, 2 H), 1.10 (m, 1 H), 1.09 (s, 3

H), 1.06 (s, 3 H), 0.98 (d, J ) 1.5 Hz, 3 H), 0.92 (s, 9 H), 0.80 (m, 2

H), 0.65 (d, J ) 6.7 Hz, 3 H), 0.02 (s, 3 H), 0.00 (s, 3 H); ¹³C NMR

(125 MHz, CDCl₃) δ 205.7, 170.2, 164.6, 152.6, 138.4, 137.4, 135.7,

135.1, 134.9, 134.3, 130.1, 127.9, 120.6, 119.2, 116.2, 81.2, 78.9, 51.7,

43.1, 39.1, 32.3, 31.7, 30.9, 26.1, 26.0, 23.6, 21.3, 20.8, 20.2, 19.2,

18.6, 14.9, 14.8, 12.8, 0.00, -3.2, -4.0; HRMS calcd for C₅₁H₇₁O₅-

NSSi₂865.4592, found 904.4201 (M + K).

Preparation of Compounds 100 and 101. To a solution of

aldehyde 99 (0.290 g, 0.341 mmol) in THF (34 mL) at -78 °C was

added potassium bis(trimethylsilyl)amide (0.5 M soln in toluene, 3.4

mL, 1.7 mmol). The solution was stirred at -78 °C for 1 h and then

quenched with saturated aqueous NH₄Cl and extracted with Et₂O (3 ×

15 mL). The combined organic layers were dried with MgSO₄, filtered,

and concentrated. Flash column chromatography (hexanes/EtOAc, 7:1)

gave 0.120 g of the desired R-alcohol 100 and 0.055 g of â-alcohol

101 (60% combined yield) as pale yellow oils.

1

IR (film) 3453, 2929, 1726, 1252, 1030 cm^-1; H NMR (500 MHz,

C₆D₆, 60 °C) δ 7.15 (s, 1 H), 6.66 (s, 1 H), 6.58 (s, 1 H), 5.54 (t, J )

6.8, 6.7 Hz, 1 H), 5.21 (m, 1 H), 4.64 (br s, 1 H), 3.80 (br s, 1 H), 2.99

(m, 1 H), 2.84 (dd, J ) 14.0, 6.7 Hz, 1 H), 2,45 (dd, J ) 14.1, 1.5 Hz,

1 H), 2.37 (s, 3 H), 2.29 (s, 3 H), 2.17-2.08 (band, 4 H), 1.80 (m, 1

H), 1.70-1.60 (band, 4 H), 1.49 (m, 1 H), 1.12-1.08 (band, 5H), 1.04

(d, J ) 6.8 Hz, 3 H), 1.00-0.90 (band, 25H), 0.18 (s, 3 H), 0.12 (s, 3

H), 0.11 (s, 3 H), 0.06 (s, 3 H); ¹³C NMR (125 MHz, C₆D₆) δ 171.2,

164.1, 154.2, 138.8, 121.4, 120.8, 116.6, 80.1, 77.6, 74.7, 44.7, 41.8,

33.5, 32.5, 31.9, 27.1, 26.4, 26.8, 23.3, 20.2, 18.9, 18.4, 18.3, 17.4,

15.1, -3.8, -4.3, -4.4; LRMS calcd for C₃₉H₇₁NO₅SSi₂: 721.5;

found: 722.7 (M + H).

Preparation of Compound 105. To a solution of alcohol 104 (0.107

g, 0.148 mmol) in CH₂Cl₂(3 mL) at rt was added Dess-Martin

periodinane (0.315 g, 0.743 mmol). The resulting solution was stirred

at rt for 2 h, treated with Et₂O (1 mL) and saturated aqueous Na₂S₂O₃/

saturated aqueous NaHCO₃(2 mL, 1:1), poured into H₂O (20 mL),

and extracted with Et₂O (4 × 10 mL). The ether solution was washed

with saturated aqueous NaHCO₃(1 × 20 mL), dried with MgSO₄,

filtered, and concentrated. Flash column chromatography (hexanes/

EtOAc, 9:1) gave ketone 105 (0.092 g, 87%) as a pale yellow oil: [R]_D

) -18.7 (c 3.65, CHCl₃); IR (film) 2931, 2856, 1742, 1696, 1463,

1255, 835 cm^-1; ¹H NMR (500 MHz, CDCl₃) δ 6.85 (s, 1 H), 6.46 (s,

Conversion of Compound 101 to Compound 100. To a solution

of â-alcohol 101 (0.075 g, 0.088 mmol) in 0.5 mL of CH₂Cl₂at rt was

added Dess-Martin periodinane (0.188 g, 0.44 mmol). The resulting

solution was stirred at rt for 1 h and then treated with Et₂O (2 mL) and

saturated aqueous NaHCO₃:Na₂S₂O₃(3 mL, 1:1), poured into H₂O (20

Total Syntheses of Epothilones A and B

J. Am. Chem. Soc., Vol. 119, No. 42, 1997 10091

1 H), 5.06 (t, J ) 8.4, 7.7 Hz, 1 H), 4.86 (d, J ) 10.0 Hz, 1 H), 3.92

(d, J ) 9.3 Hz, 1 H), 3.78 (d, J ) 8.9 Hz, 1 H), 2.92 (m, 2 H), 2.69

(d, J ) 15.5 Hz, 1 H), 2.60-2.55 (band, 1 H), 2.36 (m, 1 H), 2.06 (d,

J ) 3.5 Hz, 1 H), 2.00-1.93 (band, 5H), 1.63-1.58 (band, 6 H), 1.44

(m, 2 H), 1.15 (s, 3 H), 1.08 (s, 3 H), 1.03 (s, 3 H), 0.99 (d, J ) 6.8

Hz, 3 H), 0.87 (d, J ) 6.9 Hz, 3 H), 0.85-0.73 (band, 12 H), 0.00-

0.06 (band, 9 H), -0.21 (s, 3 H); ¹³C NMR (125 MHz, CDCl₃) δ 215.1,

171.2, 164.5, 152.5, 140.6, 138.8, 119.3, 119.1, 115.9, 79.8, 76.2, 53.4,

39.1, 32.5, 31.9, 31.4, 29.2, 27.4, 26.3, 26.1, 24.5, 24.3, 23.0, 19.1,

18.7, 18.6, 17.8, 15.3, -3.3, -3.7, -5.6; LRMS calcd for C₃₉H₆₉O₅-

NSSi₂: 719.4; found: 720.6 (M + H).

A solution of the mixture of thionoimidazolides prepared above (1.25

g, 1.34 mmol), n-Bu₃SnH (0.66 mL, 2.01 mmol), and AIBN (0.022 g,

0.13 mmol) in benzene (14 mL) was refluxed for 1 h. The reaction

mixture was cooled to rt and diluted with Et₂O (14 mL). To this

solution, DBU (0.32 mL, 2.01 mmol) was added, and the resulting

mixture was titrated with a solution of iodine in Et₂O until the solution

color turned yellow and white precipitate formed. This slurry was

filtered through a 5 cm thick pad of silica gel and concentrated in vacuo.

The residue was purified by flash chromatography (SiO₂, 25% CH₂-

Cl₂:hexanes) to give the dithiane 109 (0.823 g, 84%) as a white foam:

[R]_D) 5.97 (c 15.9, CHCl₃); IR (film) 2931, 1428, 1252, 1114, 1057

cm^-1; ¹H NMR (500 MHz, CDCl₃) δ 7.70 (dd, J ) 7.8, 1.3 Hz, 6 H),

7.45-7.38 (band, 9 H), 4.69 (s, 1 H), 4.64 (s, 1 H), 4.56 (d, J ) 3.9

Hz, 1 H), 4.16 (s, 1 H), 3.53 (d, J ) 5.2 Hz, 1 H), 2.71 (d, J ) 13.4

Hz, 1 H), 2.54-2.43 (band, 2 H), 2.08 (m, 1 H), 1.79-1.73 (band, 4

H), 1.73 (s, 3 H), 1.66-1.61 (band, 2 H), 1.35 (s, 3 H), 1.32 (m, 1 H),

1.16 (s, 1 H), 1.03 (d, J ) 7.3 Hz, 3 H), 0.91-0.85 (band, 11 H), 0.65

(d, J ) 6.8 Hz, 3 H), -0.03 (s, 3 H), -0.05 (s, 3 H); ¹³C NMR (125

MHz, CDCl₃) δ 146.3, 135.9, 134.9, 129.8, 127.8, 109.3, 77.4, 77.1,

38.0, 37.8, 31.0, 30.4, 29.6, 26.2, 25.9, 25.5, 22.6, 22.4, 20.9, -3.1,

-4.0; HRMS calcd for C₄₃H₆₄O₂S₂Si₂: 732.3886; found: 755.3784

(M + Na).

Preparation of Compound 106. To a solution of ketone 105 (0.092

g, 0.128 mmol) in THF (4.5 mL) at 0 °C was added HF‚pyridine (2.25

mL) dropwise. The solution was stirred at rt for 2 h, diluted with CHCl₃

(2 mL), poured into saturated aqueous NaHCO₃/CHCl₃(20 mL, 1:1)

slowly, and extracted with CHCl₃(4 × 10 mL). The combined CHCl₃

layers were dried with MgSO₄, filtered, and concentrated. Flash column

chromatography (hexanes/EtOAc, 3:1) gave desoxyepothilone B (106)-

(0.057 g, 92%) as a pale yellow oil: [R]_D) -61.4 (c 2.85, CHCl₃);

1

IR (film) 3465, 2969, 1735, 1691, 1377, 1181, 1148 cm^-1; H NMR

(500 MHz, CDCl₃) δ 6.95 (s, 1 H), 6.58 (s, 1 H), 5.21 (d, J ) 9.3 Hz,

1 H), 5.15 (dd, J ) 10.7, 5.5 Hz, 1 H), 4.28 (br d, J ) 9.2 Hz, 1 H),

3.80-3.50 (band, 3 H), 3.17 (m, 1 H), 3.16 (br s, 1 H), 2.69 (s, 3 H),

2.66-2.61 (band, 3 H), 2.46 (dd, J ) 14.6, 3.4 Hz, 1 H), 2.34-2.22

(band, 3 H), 2.18 (s, 3 H), 2.07 (s, 3 H), 1.88 (m, 1 H), 1.75 (m, 1 H),

1.34 (s, 3 H), 1.02 (d, J ) 7.0 Hz, 3 H); ¹³C NMR (125 MHz, CDCl₃)

δ 220.6, 210.7, 170.4, 164.9, 151.9, 139.1, 138.4, 120.8, 119.3, 115.6,

78.9, 74.1, 72.3, 69.4, 53.7, 53.4, 41.7, 39.6, 38.3, 32.5, 31.7, 29.2,

22.9, 19.0, 18.0, 15.7; HRMS calcd for C₂₇H₄₁NO₅NS: 491.2705; found

C₂₇H₄₂NO₅S: 492.2782 (M + H).

Preparation of Compound 110. A solution of compound 109 (0.95

g, 1.3 mmol) in MeOH/THF (2:1, 18 mL) was treated with [bis-

(trifluoroacetoxy)iodobenzene] (1.118 g, 2.6 mmol) at rt. After 15 min,

the reaction was quenched with saturated aqueous NaHCO₃(25 mL).

The mixture was extracted with Et₂O (3 × 25 mL), and the organic

layer was dried (MgSO₄) and concentrated in vacuo. The residue was

dissolved in dioxane/water (5:1, 12 mL) and treated with pTSA‚H₂O

(0.74 g, 3.9 mmol), and the resulting mixture was heated at 50 °C for

2 h. After cooling to rt, the mixture was diluted with Et₂O (50 mL)

and washed successively with aqueous NaHCO₃(15 mL) and brine

(20 mL). The organic layer was then dried over MgSO₄, filtered, and

concentrated. Purification of the residue by flash chromatography

(SiO₂, 25% CH₂Cl₂:hexanes) afforded aldehyde 110 (0.79 g, 95%) as

a white foam: [R]_D) -15.2 (c 11.8, CDCl₃); IR (film) 2931, 1722,

Preparation of Epothilone B (3). To a solution of desoxyepothilone

B (106) (0.090 g, 0.183 mmol) in CH₂Cl₂(1.8 mL) at -78 °C was

added freshly prepared dimethyldioxirane (0.087 M soln in acetone,

3.60 mL, 0.313 mmol) dropwise. The resulting solution was warmed

to -50 °C for 1 h, and another portion of dimethyldioxirane (1.0 mL,

0.087 mmol) was added. After stirring at -50 °C for additional 1.5 h,

any excess dimethyldioxirane and solvent were removed by a stream

of N₂at -50 °C. The crude reaction mixture was determined to be

>20:1 ratio of diastereomeric cis-epoxides by ¹H NMR spectroscopy.

The resulting residue was purified by flash column chromatography

(hexanes/EtOAc, 1:1) to give epothilone B (3) (0.090 g, 97%) as a

white solid: [R]_D) -31.0 (c 0.045, CHCl₃); mp 93.6-94.7 °C; IR

1

1472, 1429 cm^-1; H NMR (400 MHz, CDCl₃) δ 9.75 (s, 1 H), 7.64

(d, J ) 6.6 Hz, 6 H), 7.46-7.37 (band, 9 H), 4.69 (s, 1 H), 4.59 (s, 1

H), 4.04 (d, J ) 4.2 Hz, 1 H), 3.52 (dd, J ) 5.3, 2.5 Hz, 1 H), 2.16 (m,

1 H), 1.77 (m, 2 H), 1.70 (s, 3 H), 1.54 (m, 1 H), 1.44-1.26 (band, 3

H), 1.00 (s, 6 H), 0.92-0.80 (band, 15H), 0.62 (d, J ) 6.8 Hz, 3 H),

0.08 (s, 3 H), 0.02 (s, 3 H); ¹³C NMR (125 MHz, CDCl₃) δ 205.6,

146.0, 135.7, 134.4, 130.1, 127.9, 109.6, 81.1, 76.9, 51.7, 43.3, 38.9,

38.0, 30.4, 26.2, 25.6, 22.4, 21.1, 20.1, 18.6, 15.1, 12.8, -3.2, -4.0;

HRMS calcd for C₄₀H₅₈O₃Si₂642.3925, found 665.3842 (M + Na).

Preparation of Compound 114. To a solution of acetate 77 (0.285

g, 1.16 mmol) and aldehyde 110 (0.49 g, 0.772 mmol) in THF (1 mL)

at -78 °C was added LDA (2 M soln in THF, 0.772 mL, 1.54 mmol).

The yellow mixture was stirred at -78 °C for 40 min and then quenched

with saturated aqueous NH₄Cl (10 mL). The mixture was extracted

with Et₂O (3 × 15 mL), dried (MgSO₄), filtered, and concentrated in

vacuo. Purification of the residue by flash chromatography (SiO₂, 6%

EtOAc:hexanes) provided R-alcohol 114 (0.239 g, 35%) and â-alcohol

115 (0.238 g, 35%).

1

(film) 3454, 2962, 1727, 1690, 1263, 978 cm^-1; H NMR (400 MHz,

CDCl₃) δ 6.98 (s, 1 H), 6.52 (s, 1 H), 5.43 (dd, J ) 7.7, 2.4 Hz, 1 H),

4.22 (m, 2 H), 3.78 (t, J ) 4.3 Hz, 1 H), 3.30 (m, 1 H), 2.81 (dd, J )

7.5, 4.6 Hz, 1 H), 2.70 (s, 3 H), 2.54 (m, 1 H), 2.37 (d, J ) 12.7 Hz,

1 H), 2.09 (s, 3 H), 1.93 (m, 1 H), 1.72 (m, 2 H), 1.49 (m, 2 H), 1.43

(m, 3 H), 1.37 (s, 3 H), 1.32 (s, 3 H), 1.17 (d, J ) 6.8 Hz, 3 H), 1.08

(s, 3 H), 1.01 (d, J ) 6.9 Hz, 3 H); ¹³C NMR (100 MHz, CDCl₃) δ

220.5, 170.5, 165.1, 151.7, 137.5, 119.6, 116.1, 74.1, 72.8, 69.5, 61.6,

61.3, 53.7, 53.0, 42.9, 39.1, 36.3, 32.2, 32.0, 31.7, 30.6, 29.2, 22.7,

22.3, 19.6, 19.0, 17.1, 15.7, 13.7; LRMS calcd for C₂₇H₄₁O₅NS 507.3,

found 508.4 (M + H).

Preparation of Compound 109. To a solution of tert-butyllithium

(3.6 mL of a 1.7 M solution in Et₂O; 6.08 mmol) in Et₂O (5 mL) at

-78 °C was added a solution of 4-iodo-2-methyl-1-butene (0.596 g;

3.04 mmol) in Et₂O (20 mL). After 0.5 h, a solution of aldehyde 65

(1.03 g; 1.52 mmol) in Et₂O (5 mL) was added. After 5 min at -78

°C, the cooling bath was removed, and the solution was allowed to

warm to 0 °C and stirred for 0.5 h. The reaction mixture was then

poured into saturated aqueous NH₄Cl solution (100 mL) and extracted

with Et₂O (2 × 100 mL). The organic layer was dried (MgSO₄),

filtered, and concentrated. The residue was purified by flash chroma-

tography (silica, 5-7% Et₂O/hexanes) to give a 3:1 mixture of

diastereomeric alcohols (1.01 g; 89%) as a white foam.

To a solution of â-alcohol 115 (0.238 g, 0.27 mmol) in CH₂Cl₂(3

mL) was added Dess-Martin periodinane (0.689 g, 1.62 mmol) at rt.

The resulting solution was stirred for 1 h and then quenched by the

addition of 1:1 saturated aqueous NaHCO₃:Na₂S₂O₃(10 mL). The

mixture was extracted with Et₂O (3 × 10 mL). The organic layer was

dried (MgSO₄), filtered, and concentrated in Vacuo. The resulting C-3

ketone 116 was immediately dissolved in a mixture of MeOH (4 mL)

and THF (2 mL) and cooled to -78 °C. Sodium borohydride (0.102

g, 2.7 mmol) was then added, and the mixture was allowed to warm to

rt and stirred for 1 h. The reaction was then quenched with saturated

aqueous NH₄Cl (10 mL). The mixture was extracted with ether (3 ×

15 mL), dried (MgSO₄), filtered, and concentrated in Vacuo. Purifica-

tion of the residue by flash chromatography (SiO₂, 8% EtOAc:hexanes)

provided R-alcohol 114 (0.230 g, 92%): [R]_D) -45.3 (c 0.29, CHCl₃);

A solution of the mixture of alcohols prepared above (1.01 g, 1.35

mmol), thiocarbonyl diimidazole (0.801 g, 4.05 mmol), and 4-DMAP

(0.164 g, 1.35 mmol) in THF (10 mL) was heated to 90 °C. A stream

of N₂was then used to evaporate the THF completely. The sticky

residue was maintained at 90 °C for 2 h, cooled to rt, and diluted with

CH₂Cl₂(5 mL). Purification of the residue by flash chromatography

(SiO₂, 20% EtOAc:hexane) provided an epimeric mixture of thiono-

imidazolides (1.25 g, 99%) as a pale yellow foam.

1

IR (film) 3502, 2927, 1715, 1428 cm^-1; H NMR (500 MHz, CDCl₃)

δ 7.64 (d, J ) 6.5 Hz, 6 H), 7.45-7.30 (band, 9 H), 6.93 (s, 1 H), 6.50

(s, 1 H), 5.65 (m, 1 H), 5.31 (t, J ) 6.7 Hz, 1 H), 5.05 (m, 2 H), 4.69

(s, 1 H), 4.63 (s, 1 H), 4.13 (d, J ) 3.5 Hz, 1 H), 4.00 (m, 1 H), 3.73

(d, J ) 4.5 Hz, 1 H), 3.25 (d, J ) 2.5 Hz, 1 H), 2.71 (s, 3 H), 2.42 (m,

10092 J. Am. Chem. Soc., Vol. 119, No. 42, 1997

Meng et al.

1

4 H), 2.06 (m, 1 H), 2.02 (s, 3 H), 1.87-1.78 (band, 2 H), 1.75 (s, 3

H), 1.70 (m, 1 H), 1.31-1.20 (band, 2 H), 1.12 (m, 1 H), 1.02 (d, J )

8.0 Hz, 3 H), 0.96 (s, 3 H), 0.91 (s, 3 H), 0.79 (s, 9 H), 0.60 (d, J )

7.0 Hz, 3 H), -0.06 (s, 3 H), -0.07 (s, 3 H); ¹³C NMR (125 MHz,

CDCl₃) δ 131.1, 164.6, 152.4, 146.2, 136.4, 135.7, 134.9, 133.2, 129.8,

127.7, 121.4, 117.9, 116.5, 109.4, 79.0, 78.7, 76.3, 71.8, 44.2, 43.0,

38.7, 38.0, 37.4, 36.5, 30.4, 26.2, 26.1, 25.7, 22.5, 21.0, 20.5, 19.2,

18.5, 15.2, 14.5, 13.4, -3.1, -4.3; HRMS calcd for C₅₃H₇₅O₅NSSi₂:

893.4904, found: 932.4529 (M + K).

(film) 2929, 1737, 1695, 1253 cm^-1; H NMR (500 MHz, CDCl₃) δ

6.95 (s, 1 H), 6.49 (s, 1 H), 5.70 (m, 1 H), 5.29 (t, J ) 6.8 Hz, 1 H),

5.05 (m, 2 H), 4.68 (s, 1 H), 4.65 (s, 1 H), 4.34 (dd, J ) 5.9, 3.4 Hz,

1 H), 3.72 (d, J ) 5.4 Hz, 1 H), 3.15 (m, 1 H), 2.70 (s, 3 H), 2.53-

2.42 (band, 3 H), 2.28 (dd, J ) 17.0, 6.1 Hz, 1 H), 2.06 (s, 3 H), 1.99

(m, 2 H), 1.69 (s, 3 H), 1.41 (m, 1 H), 1.33-1.29 (band, 3 H), 1.25-

1.21 (band, 4 H), 1.04-1.02 (band, 6 H), 0.92-0.83 (band, 21 H),

0.10 (s, 3 H), 0.05 (s, 3 H), 0.03 (s, 6 H); ¹³C NMR (125 MHz, CDCl₃)

δ 171.2, 164.5, 152.5, 146.0, 136.7, 133.4, 121.1, 117.8, 116.4, 109.8,

78.6, 77.7, 74.0, 53.3, 45.2, 40.3, 38.7, 38.3, 37.5, 31.6, 26.2, 26.0,

25.6, 23.1, 22.4, 20.3, 19.3, 18.5, 18.2, 17.1, 15.5, 14.5, -3.6, -3.8,

-4.3, -4.7; HRMS calcd for C₄₁H₇₃O₅NSSi₂: 747.4748; found:

786.4362 (M + K).

Preparation of Compound 120. The aldol product 114 (0.219 g,

0.27 mmol) was treated with buffered HF‚pyridine in THF (8.0 mL)

at rt (the stock solution was prepared from 20 mL THF, 11.4 mL

pyridine and 4.2 g hydrogen fluoride-pyridne (Aldrich Co.)). After 2

h, the reaction was poured into saturated aqueous NaHCO₃and extracted

with Et₂O. The organic layer was washed in sequence with saturated

aqueous CuSO₄(3 × 30 mL) and saturated aqueous NaHCO₃(50 mL),

dried over Na₂SO₄, and concentrated in vacuo. The residue was purified

by flash chromatography (silica, 10 f 20% EtOAc in hexane) to give

diol 120 (0.15 g, 93%) as a white foam: [R]_D) -40.5 (c 3.8, CDCl₃);

Preparation of Compound 124. To a solution of diene 122 (0.015

g; 0.02 mmol) in dry, degassed benzene (20 mL) at rt was added

Schrock’s metathesis catalyst [Mo(CHMe₂Ph)(N-(2,6-(i-Pr)₂C₆H₃))-

(OCMe(CF₃)₂)₂] (0.0038 g; 0.005 mmol) in a dry box. The reaction

mixture was then warmed to 55 °C and stirred for 2 h. The reaction

mixture was concentrated in vacuo, and the residue was purified by

flash chromatography (silica, 4% EtOAc/hexanes) to give an inseparable

1:1 mixture of stereoisomeric alkenes 105 and 123 (0.013 g; 86%).

To a solution of the mixture of alkenes prepared above (0.013 g;

0.018 mmol) in THF (1 mL) was added HF‚pyridine (0.5 mL), and

the resulting solution was stirred for 1.5 h. The reaction mixture was

then poured into saturated aqueous NaHCO₃solution (50 mL) and

extracted with chloroform (3 × 30 mL). The organic layer was dried

(MgSO₄), filtered, and concentrated. The residue was purified by flash

chromatography (silica, 30% EtOAc/hexanes) to give an equimolar

mixture of stereoisomeric alkenes 106 and 124 (0.008 g; 90%). The

mixture was separated by preparative thin-layer chromatography (2%

MeOH/CH₂Cl₂, four elutions). 124: IR (film) 3484, 2922, 1732, 1693,

1

IR (film) 3457, 2930, 1732, 1472, 1386, 1252 cm^-1; H NMR (500

MHz, CDCl₃) δ 6.93 (s, 1 H), 6.51 (s, 1 H), 5.72 (m, 1 H), 5.35 (t, J

) 6.7 Hz, 1 H), 5.08 (m, 2 H), 4.68 (s, 1 H), 4.65 (s, 1 H), 4.07 (d, J

) 10.0 Hz, 1 H), 3.92 (br s, 1 H), 3.80 (br s, 1 H), 3.49 (br s, 1 H),

2.68 (s, 3 H), 2.61-2.45 (band, 4 H), 2.07 (d, J ) 1.2 Hz, 3 H), 2.01

(br s, 1 H), 1.69 (s, 3 H), 1.55 (m, 1 H), 1.36 (m, 1 H), 0.99 (d, J )

6.9 Hz, 3 H), 0.88 (s, 9 H), 0.80 (s, 3 H), 0.09 (s, 3 H), 0.07 (s, 3 H);

¹³C NMR (125 MHz, CDCl₃) δ 172.7, 164.5, 152.5, 145.8, 136.8, 133.3,

120.9, 117.7, 116.3, 109.8, 78.5, 41.5, 38.0, 37.5, 37.1, 32.9, 25.9, 25.5,

22.3, 21.1, 19.1, 18.2, 16.1, 14.6, -4.4; LRMS calcd for C₃₅H₆₁O₅-

NSSi: 635.4; found: 658.5 (M + Na).

Preparation of Compound 121. To a cooled (-30 °C) solution

of diol 120 (0.110 g, 0.173 mmol) and 2,6-lutidine (0.121 mL, 1.04

mmol) in anhydrous CH₂Cl₂(2 mL) was added TBSOTf (0.119 mL,

0.519 mmol). The resulting solution was then stirred at -30 °C for

30 min. The reaction was quenched with 0.5 M HCl (50 mL) and

extracted with Et₂O (150 mL). The Et₂O layer was washed with

saturated aqueous NaHCO₃(50 mL), dried (Na₂SO₄), and concentrated

in vacuo. Purification of the residue by flash chromatogrphy (silica, 5

f 8% EtOAc in hexane) afforded TBS ether 121 (0.112 g, 85%) as

white foam: [R]_D) -33.7 (c 1.6, CHCl₃); IR (film) 3478, 2929, 1737,

1464, 1260 cm^-1; H NMR (500 MHz, CDCl₃) δ 6.98 (s, 1 H), 6.57

1

(s, 1 H), 5.30 (m, 1 H), 5.11 (t, J ) 6.7 Hz, 1 H), 4.32 (d, J ) 10.0 Hz,

1 H), 3.67 (m, 1 H), 3.29-3.25 (band, 3 H), 2.70 (s, 3 H), 2.65-2.45

(band, 5 H), 2.16 (m, 1 H), 2.07 (s, 3 H), 1.98 (m, 1 H), 1.73-1.61

(band, 3 H), 1.60 (s, 3 H), 1.31 (m, 1 H), 1.28 (s, 3 H), 1.17 (d, J )

6.8 Hz, 3 H), 1.05 (s, 3 H), 0.98 (d, J ) 7.0 Hz, 3 H); HRMS calcd for

C₂₇H₄₁O₅NS: 491.2705; found: 492.2795 (M + H).

Acknowledgment. This research was supported by the

National Institutes of Health (grant number: CA-28824).

Postdoctoral Fellowship support is gratefully acknowledged by

E.J.S. (NSF, CHE-9504805), A.B. (NIH, CA-GM 72231), P.B.

(NIH, CA 62948), and T.K. (NIH AI-09355). A U.S. Depart-

ment of Defense Predoctoral Fellowship is gratefully acknowl-

edged by D.M. (F33 US ARMY DAMD 17-97-1-7146). We

gratefully acknowledge Dr. George Sukenick (NMR Core

Facility, Sloan-Kettering Institute) for NMR and mass spectral

analyses. Professor Dr. G. Ho¨fle of the Gesellschaft fu¨r

Biotechnologische Forschung is gratefully acknowledge for

providing natural epothilone A and epothilone B for comparative

analysis. We also thank Professor Gunda Georg of the Univer-

sity of Kansas for bringing this problem to our attention.

1

1471, 1253 cm^-1; H NMR (500 MHz, CDCl₃) δ 6.93 (s, 1 H), 6.50

(s, 1 H), 5.71 (m, 1 H), 5.30 (t, J ) 6.8 Hz, 1 H), 5.08 (m, 2 H), 4.67

(s, 1 H), 4.65 (s, 1 H), 4.20 (m, 1 H), 3.81 (br s, 1 H), 3.40 (br s, 1 H),

2.90 (dd, J ) 17.2, 3.9 Hz, 1 H), 2.69 (s, 3 H), 2.47 (m, 2 H), 2.35

(dd, J ) 17.2, 5.4 Hz, 1 H), 2.07 (s, 3 H), 2.01 (m, 2 H), 1.99 (m, 1

H), 1.69-1.60 (band, 4 H), 1.54-1.48 (band, 2 H), 1.25 (m, 1 H),

0.94 (d, J ) 7.0 Hz, 3 H), 0.91 (s, 3 H), 0.89-0.87 (band, 12 H), 0.86

(s, 3 H), 0.83 (s, 9 H), 0.07 (s, 3 H), 0.02 (s, 3 H), 0.01 (s, 3 H), -0.05

(s, 3 H); ¹³C NMR (125 MHz, CDCl₃) δ 172.7, 164.5, 152.5, 146.2,

136.7, 133.4, 121.2, 117.8, 116.4, 109.7, 78.7, 78.1, 74.4, 43.4, 39.8,

38.1, 37.5, 32.9, 26.0, 25.9, 25.6, 22.4, 19.9, 19.2, 18.2, 18.1, 14.2,

-4.0, -4.2, -4.22, -4.9; HRMS calcd for C₄₁H₇₅O₅NSSi₂: 749.4904,

found: 788.4518 (M + K).

Preparation of Compound 122. To a solution of alcohol 121 (0.110

g, 0.147 mmol) in CH₂Cl₂(2.0 mL) at rt was added Dess-Martin

periodinane (0.249 g, 0.583 mmol) in one portion. The resulting

solution was then allowed to stir at 25 °C for 1.5 h. The reaction was

quenched by the addition of 1:1 saturated aqueous NaHCO₃:Na₂S₂O₃

(10 mL) and stirred for 5 min. The mixture was then extracted with

Et₂O (3 × 15 mL). The organic layer was dried (Na₂SO₄), filtered,

and concentrated in vacuo. Purification of the residue by flash

chromatography (silica, 6% EtOAc in hexane) provided ketone 122

(0.099 g, 94%) as white foam: [R]_D) -55.1 (c 0.85, CHCl₃); IR

Supporting Information Available: Experimental proce-

dures for the preparation of compounds 10, 12, 13, 15, 16-20,

22, 24, 27, 37-44, 55-57, 65, 70-75, 107, 108, 111, 112,

117-119, and 125-136 as well as full characterization data

are included (60 pages). See any current masthead page for

ordering and Internet access instructions.

JA971946K

Article Doi

Total Synthesis of Epothilones A and B

DOI: 10.1021/ja971946k

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja971946k

Full text of DOI:10.1021/ja971946k

Products guided by the article

R&D Labs maybe for 188259-84-9

Relevant to this article

Hot Product