Synthesis of L-660,631 methyl ester and related compounds

Tetrahedron Letters,Vol.29,No.l9,pp Printed in Great Britain

2279-2282,1988

0040-4039/88 $3.00 + .OO Pergamon Press plc

SYNTHESIS OF L-660,631 METHYL ESTER AND RELATED COMPOUNDS M. D. Lewis*, J. P. Duffy, J. V. Heck, and R. Menes Merck Sharp & Dohme Research Laboratories PO Box 2000, Rahway NJ 07065 Summary: Triyne carbonate L-660631 methyl ester (2) was synthesized in eight steps from cyclooctene. methodology to permit systematic variation of the triyne portion of the molecule has been developed. Triyne carbonate

L-660631

structure determlnation P-ketothlolase, important

(1) is a novel natural product obtained from actinomycefes

of which has been described recently.’

the initial enzyme of cholesterol biosynthesis,

enzymen2

Unfortunately,

such studies

fermentation,

the Isolation and

This compound Is an extremely potent inhibitor of cytosolic

and may be useful in studying the mechanism of action of this

may be rendered

more complex

by the rather unstable

nature of

In order to develop compounds with enhanced stability that still retained inhibitory

concentrated preparations of L-660,631.

activity, a total synthesis of the methyl ester of L-660631 was sought.

Foremost in the design of the synthetic plan was the

choice of a synthetic route that would allow for the systematic replacement of the offending more stable fragments.

Synthetic

1,3,5-hexatriyne subunit with

This report details such a total synthesis of L-660,631 methyl ester (2).

Retrosynthetic analysis suggested that simple disconnection

into triyne 3 and aldehyde 4 subunits might be appropriate.

Problematic in this approach would be the generation and control of the triyne anion with respect to the regioselectivity of the three possible sites of carbonyl addition to an aldehyde precursor such as 4. The carbonate potentially troublesome

by the tendency

of the natural

product L-660,631 (1) to undergo

Worrisome also was the prospect of the aldehyde decomposing conditions.

portion was indicated as

hydrolysis

in pH 10 buffer.

via 6-elimination under the strongly basic addition reaction

Another obstacle was selectivity; an addition reaction as proposed is not Inherently diastereoselective.

Despite

these substantial shortcomings, the disconnection was adopted because of its simplicity. ---

-

-M 3

t of’

COzR oK
L-660.631

2

R = H R = Me

It was intended stereochemistry

that

asymmetric

at C.6 and C.9.

4

Sharpless

kinetic resolution3

Earlier experiments

be used to set both the absolute

had suggested

that no interference

and the relative

by the terminal methyl ester

functionality would occur during epoxidation of substrates such as 6.4 A kinetic resolution such as that intended, whereby the epoxide

is the product

diastereoselectivity

further

transformed,

and enantioselectivity

would

only be practical

in those

cases where

both the desired

of the epoxidatlon process was very high. At the onset of this work, it appeared

that the desired epoxidation of 6 might be expected be In the range >95% by comparison with published examples for both enantiomeric and diastereomeric selection.3 Schreiber workup

of the ozonotysis

product

of cyclooctene

In methanol

provided

methyl 7-formylhepanoate

(6) in

reasonable yield.5*6 Selective addition of vinyl magnesium bromide to aldehyde 6 could be accomplished in THF at -76’C. In this manner, allylic alcohol 7 could be prepared in large quantities from simple and inexpensfve precursors.

2279

2280

j (for

1s)

k (for

14)

Rd

Rmctton

condtttom:

a)l. OS, MeOH,

66%. 9) BFs”n, C$Cr, m)i.

CH9C19, -76‘k:

il. 40,

NE’s

PhH, 77%. b) CH,-CHMgBr,

?HF, -7kC.

~6.

C)

1BuOOH.ti rlew* Cl$C& -20% to Rl, 36% (72?4c4tI~eoraUc@. d) PhNCO. MeCN. RT, I wmk.

TIO'W4. L-(+)-DIPT &O,

AT. 1)t. ItBull.

-2U’, 69%. 0 (RJMPA-CI, THF; 1. CtC09t&.

Pvtdlm,

D CUC=CR

CH9C$

AT, 73%. g) CrOs, H,SO,,

Ut. Et900 ITT. !G) L!BF&XtR.

T-IF.

GOCI), DMSO, lHF, do to 3soc; II. 12 -60 to 35oc: Iii. NEta, Ssoc to Rl. n) UCICR,

acaton9, RT, 96%. h) (COq2, -76°C.

I) NaSH,,

THF, -7eOc.

Table: Syntheses of propargylic alcohols related to L-660,631.

F+? ~C/_%. Method

R

Yield ”

Ratio (a$)

20”

A

13.2%

1:1

Me,Si~

21

C

46%

1:J

n&Is

22”

A

9.0%

1:l

08UG

22

B

3.4%

1:1

nBue

22

C

59%

1.&l

1::

:

1:

:::,

2622

B

1.8%

1:1

PhG

Ii= Me,6,z

Ii-

M0~s.l =

=

20 23

c

20%

1:2.5

Me=

=

27 ‘=

C

46%

1:l.Z

He,Si

=

=

=

2823

c

39%

1:1.8

Mc,C

=

=

=

3023

c

54%

1:Z.Q

”

MeOH, RT.

0

R = CONHPh

10

R -

(R)-MTI’A

2281

Asymmetric epoxidation of 7 afforded epoxy-alcohol 9 using modified Sharpless conditions.’

Alcohol 9 could bs

transformed into phenyl urethane 0 by extended treatment with phenyi isocyanate in acetonltrile.

The epoxy-urethane 0

was rearranged uneventfully with cold boron trffluonde etherate in the usual manner, affording hydroxycarbonate 11 In good yield.’ Determination of the extent of enantiomeric excess obtained durlng t,he epoxldation step was accomplished by reaction of both racemic and resolved epoxy-alcohols 9 with R-(+)-MTPA chloride.O~10 Examination of the 300 MHz ‘H NMR spectrum of 10 verified that the desired resolved material was >95% ee. In order to check that the asymmetric epoxidation had occurred in the proper absolute sense, hydroxycarbonate 11 was correlated wkh a substance of known absolute configuration. Conversion of 11 to the bromide 17 followed by reductive vicinal elimination produced aliylic alcohol 18.” Ozonolysis, reduction, and acetonide formation provided acetonide 19 which had an optical rotation identical to that of the same compound prepared from R-glyoeraidehyde.‘2

:4 RLco2Me-

Me Me

J--we

-

+

oLco*Me 19

18 11 R=OH 17 R = Br

Cxidation of hydroxycarbonate 11 proved more difficult than orlginaliy anticipated. Oxidation with mild Cr(Vl) reagents such as PCC or PDC produced only unsaturated aldehyde products derived from ~-elimination.

Swem oxidationi was more

successful, but the aldehyde product proved to be lnsufflclently robust to survive aqueous workup and chromatographlc purification. Oxidation with Jones reagent In acetone was successful, and the crude acid product could be convened directly to acid chloride 13 by treatment with oxalyl chloride. Reaction of acid chloride 13 with isolated copper-(l) acetylides led to poor but reproducible yields of acetylenic ketones il.

These ketones could be reduced in a nonspecfflc manner with

sodium borohydride to afford moderate yields of the desired alcohois 16 (method A). Slightly more versatile was treatment of the mixed anhydride 14 with lithium trifluoroborate organoalkyis” followed by nonselective reduction as before (method 6). In order to circumvent the instablllty of aldehyde 12, additions of organolithiums directly to the oxidation reaction mixture were investigated. lnltlal attempts to react aidehyde 12, obtained from Swem oxidation of 11 in diethyl ether, in situ with varkw organometailic reagents were not successful. However, the desired addition products were obtained in reasonable

yields when the Swem oxidation and subsequent In Mu reaction with alkyl lithiums was carried out in THF (method C).” In this manner, 1-lithiohexyne could be reacted with 12 to afford a 59% yield of a 1.&l =:p C.10 mixture (by 300 MHz ‘H NMR) of diastereomers 16 [R = CRC(CH&.le,

(22, see TABLE)].A summary of results for the construction of similar systems by

use of these three methods is shown in the TABLE. Formation of 1-lithlo diynes via methyl lithium-lithium bromide complex desilylation of 1#l&is-trimethylsilyl-1,3butadiyne has been reponed.16 Thus, when

Straightforward application of thls concept was utilized in the formation of 1-lithlo-1,3,5-hexatriynes.

1,6-b/s-trimethylsilyl-1,3,bhexatriyne was treated with one equivalent of

methyl lithium-lithium bromide

complex, a solution of the desired monoanian could be obtalned. This monoanion reacts with simple aldehydes in a straightromsrd manner to give the expected addition products in hlgh yields.

Application toward the synthesis of

L-860,331 methyl ester (2) was demonstrated by facile reactfon with aldehyde 12 to provide trimethylsifyl capped triyne carbonate 29. This compound ls reasonably stable at room temperature in concentrated form and has been stored for several weeks at -2@C without loss of activity.

Exposure to tetrabutyl ammonium fluoride-acetic acid In THF afforded

L-660,631methyl ester (2) In 30% yield, identical to naturally derived material In all respects examined.”

2282

Bioiogicai evaluation of compounds 20 to 30 as inhibitors of mammalian @-ketothiolasewill be the subject of a future report from these laboratories. The chemical instability of compounds bearing uncapped polyacetylenes is marked relative to the others, and it should prove interesting to determine whether this property correlates with inhibitory activity. A detailed u~~andi~

of the mech~~m

of action of the unique triyne functional group will have to awa& these further

studies. Reference8 and Notes: I.

2.

3.

4.

:: 7. 8.

Q. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

22. 23.

Kempf, A. J.; Hensens, 0. D.; Schweutt, R. E.; Sykes, Ft.S.; Wichmann, C. F.; Wilson, K E.; Zink, 0. I.; Zitano, L Abstracts of the New YorkAcademy ofSciencef st fn~~~~~ Conference on Drug Research in fmmunofog~c and infectious Diseases; Antifunga/ Drugs: Synthesis, Pn?clinicai and Cllnical EVa/U8fion, Garden City, Long Island, New York, tOlai87 - lOl10/87; Lewis, M. D.; Menes, Ft.TefmhecfmnLetters 1987,28, 5129. of the ~eri~an Societies for Qreenspan, M. 0.; Chen, J.; Yudkovitz, J. 8.; Albert& A. W. F~eration Proddings ~~firnental Biology 72nd Annual Meeting, Las Vegas, Nevada, 5/l/88 - 5/5/a& See also Onlshi, J. C.; Abruuo, Q. K.; Frorntllng, R. A.; Qarrity, Q. M.: Milligan, J. A.; Pelak, S. A.; Rozdilsky,W.Abstracts of the New York Academy of Science 1st ln~m8tional Conference on Drug Research in Immunologic and lnfeetious Diseases; ~~~funga~ Drugs: ~n~es~s~ Preclinical and Clinical Evaluation, Garden City, Long Island, New York, lOJ8/87- 1OliO/67. Martin, V. S.: Woodard, S. S.; Katsukl, T.; Yamada, Y.; Ikeda, M.; Sharpless, K. 8. J, Am. Chem, Sot. 1981, 103, 6237; Hanson, R. M.; Sharpless, K 8. J. Org. Chem. 1986, 57, 1922. Recent reviews: Finn, M. Gi.; Sharpless, K S. in Asymmetric Synthesis: Morrison, J. D., Ed.; Academic Press: New York, 1985: Vol. 5, Chapter 8, 247; Rossiter,S. E. in Asymmetrfc Synthesis; Morrison, J. D., Ed.; Academic Press: New York, 1985; Vol. 5, Chapter 7, 193; Pfenninger,A. Synthesis 1988,89. More proximal ester functionality on an epoxidation substrate has been repotted to interfere with asymmetric epoxidation, Corey, E. J.; Hashlmoto, H.; Barton, A. J. J. Am. Chem. Sot. 1981, 703, 721; Pridgen, L W.; Shilcrat, S. C.; Lantos I. Tetrahed~n Letf. 1984,27,2835. Schreiber, S. L; Claus, R. E.; Reagan, J. Tetrahedron Left. 1982,23,3867. SatiSfaCtOrySpectral data were obtained for all isolated synthetic intermediates and products described in thls report. GaO, V; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.; Sharpless, K. S. J. Am. Chem. Sot, 1987, 109,5766. Minami, N.; Ko, S. S.; Kishi, Y. J. Am. Chem. Sot. 1982, 704,1109; Roush, W. FL; Brown, Ft.J. J. Org Chem. 1982,47, 1371; Katsukl, T.; Lee, A. W. M.; Ma, P.; Martin, V. S.; Masamune, S.; Sharpless, K S.; Tuddenham, D.; Walker,F. J. ibid. 1982, 47,1373; Roush,W. R.; Brown, R. J.; DiMare, M. Ibid. 1983,48,5083; Roush, W. R.; Adam, M. A. Ibid. 1985, 50,3752. Dale, J. A.; Dull, 0. L.; Mosher, H. S. J. Org Chem. 1969,34,2543. Racemic erythro epoxy-alcohol 8 was prepared by treatment of racemic allylic alcohol 7 as follows: a) MCPBA, 44% overall. CHzCl , reflux, 95%; b)l. PDC,& molecular sieves, HOAc, CH,CI,, RT; ii. ZnSH,, E 0,O“C, Alcoha 11 was treated with Ph P and CBr in CH Cb to afford bromide 17 (92% yt eld) and was reductively eliminated with zinc metal, HCI, and Nal ii dioxane f~ll~~~ ~~~l~tlon of demethyiated compound wlth diazomethane to afford 18 in quantitative yield. Synthetic 19: [ CCX]~ = +11.2” (~3.44, CH Cl& natural 19: (a] - +ll.O5o (~3.86, CH,Ci& Seeref 1. Mancuso, A. J.; Juang, S.-L; Swern, D; J!Org Chem. 1978,4$2480. Normant, J. F.; Sourgain, M. TeVehedron Lefters 1970, 2659; Brown, H. C.; Racherla, U. 5.; Singh, S. M. Tetrahedron Letters 1984,25,2411 I Ireland, R. E.; Norbeck, 0. W. J. Org. Chem. 1@85,60,2198. Holmes, A B.; Jennings-White, C. L 0.; Schilthess, A. H.; Akinde, B.; Walton, D. R. M. J. C. S. Chemm. Comm. 1979, 840; Holmes, A. B.; Jones, 0. E. Tetrahedron Letters 1980,2 I, 3111, Selected data for 2: 300 MHz ‘H NMR (CDC$): 6 2.207 (IH, s), 6 2.321 (2H, t, J = 7.5 Hz), 6 2.862 (lH, d, J = 5.4 Hz), 6 3.675 (3H, s), 6 4.323 (IH, t, J I 4.9 Hz), 6 4.628 (lH, q, J = 5.8 Hz), 6 4.639 (lH, t, J = 5.0 Hz); IR (CDCI$2250,1605, 1726 cm“; [=I0 = 41 .f (c 0.23, CDC$). All yields are those of isolated products. Ratios determined by 300 MHz ‘H NMR or by quantities of isolated products. Copper acetylides were synthesized by adding the appropriate aikyne in EtOH to a solution of freshly prepared Cu(l) SO, (from CuSO, 5 H 0, NH,OH, 7 0 and HfN OH’HCi). The precipitate was filtered and washed well with HP, Et08 and EtaO,and drleci i! vacua. Caut on is adv sed as copper acetylides are known to be explosive. The lithium acetylide was formed by add&on of 2 autos of nSuLi to 1-(2,2-dib~~he~~~-trim~~lsi~l~~~l benzene In THF at -78’C. This compound was prepared from 4-iodobenzaldehyde as follows: a) trimethylsilyl acetylene, cat, (Ph,P),PdCI , ciiethylamine, cat. Cul, RT, 87%. b) PhsP, CSr,, CH,C$, O’C, 87%. See Takahashl,S.; Kuroyama, Y.; Sonogashira,k; Haglhara, N. Symhes& 1980,627 for related work. The boron reagent was derived from l-lithio-trimethylsiiylbutadiyne (see ref 14). The terminal trlmethylsityl group is cleaved by NaBH during the reduction step. The lithium acetyifdes were prepared by treatment with MeLi LiSr complex in THF at -78 to O°Cfor triynes. See ref 14 for the diyne cases. (Received

in

USA 26 February

1988)