Discovery through total synthesis: A retrospective on the himastatin problem

FULL PAPER

Discovery through Total Synthesis:

A Retrospective on the Himastatin Problem

Theodore M. Kamenecka^[a]and Samuel J. Danishefsky*^[b]

Dedicated to Albert Eschenmoser

Abstract: A total synthesis of a struc-

ture proposed for himastatin was ac-

complished. The non-identity of the

fully synthetic material with himastatin

necessitated a revision of the assigned

structure. Confirmation of the revised

stereostructure was subsequently con-

firmed through total synthesis. Among

the achievements during this effort were

i) stereospecific routes to both anti ± cis

and syn ± cis pyrrolindoline substruc-

tures; ii) a practical synthesis to 5-hy-

droxypiperazic acid in enantiomerically

pure form; iii) a Stille coupling leading

to a complex bi-indole moiety, and iv)

efficient protecting group management

throughout the evolving depsipeptide

domain. The outlines for a biological

pharmacophore have been delineated.

The alternating d- and l-substituents in

the 6-mer as well as the biaryl linkage

connecting the two identical subunits

are critical for maintaining biological

activity. This pattern is simulated in

another antibiotic, and suggests a possi-

ble structural trend for future SAR

investigations.

Keywords: antibiotics ´ himastatin

´

natural products

´

protecting

groups ´ total synthesis

launched. Extensive hydrolytic studies served to identify the

components of the depsipeptide, which joins to an oxidatively

modified tryptophan. Attention was next directed to the

linkage order of the subunits. Mass spectrometry served to fill

in some of the missing elements in the structural puzzle. Most

critical to the analysis, was the creative use of state of the art

Introduction

Given the growing need for new antibiotic agents in the

ongoing battle against microbial infection, the screening of

multiple potential sources of useful metabolites continues.

Clearly, the increasingly serious consequences associated with

the emergence of new strains of microorganisms which are

resistant to the menu of known bactericides heightens interest

in this important area.^[1]There is a particularly acute need for

antibiotics of novel structure which lend themselves to further

structural diversification for purposes of combating resist-

ance. Undoubtedly, these sorts of considerations loomed large

in directing researchers from the Bristol Myers Laboratories

to the Himachal Pradesh State in India and to the study of an

indigenous actinomycete strain. During the course of this

venture, the isolation team had characterized an apparently

new metabolite of the formula C₇₂H₁₀₄N₁₄O₂.^{[2a, b]}In keeping

with the history and geographic setting of the project, as well

as the antibacterial (gram positive) and antitumor properties

the new agent was termed himastatin.

1

interactive H and ¹³C NMR spectroscopy. At the chemical

level, a major advance was accomplished by treatment of

himastatin with lithium borohydride (Figure 1). The major

H

HN

CH₂OH

N

H

HO

HN

H

O

NH

OH

LiBH₄

N

himastatin

HOH₂C

NH

H

A

Figure 1. Himastatin degradation product.

product resulting from this protocol, C₂₂H₄₄N₆O₆bearing a

single valinol residue per subunit, lent itself to more detailed

analysis. At this stage the presence of a symmetric, oxidatively

dimerized version of a hypothetical monomer (vide infra)

could be formulated, which contains an oxidatively cyclized

tryptophan subunit. The site of oxidative dimerization was in

the benzo region of the indole, while oxidative cyclization had

occurred in its pyrrolo sector. The C terminus of the modified

tryptophan serves to acylate a d-valinol residue (see gross

structure A).

There then ensued a program of strain improvement and

corresponding upgrading of fermentation yield. In parallel, an

effort directed to establishing the structure of himastatin was

[a] Dr. T. M. Kamenecka

Laboratory for Bioorganic Chemistry

Sloan-Kettering Institute for Cancer Research

1275 York Ave., Box 106, New York, NY 10021 (USA)

[b] Prof. S. J. Danishefsky

Department of Chemistry, Columbia University

Havemeyer Hall, New York, NY 10027 (USA)

E-mail: s-danishefsky@ski.mskcc.org

Starting from a secure assignment of the d-valine config-

uration at the side chain in A, ORD measurements in concert

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0041 $ 17.50+.50/0

41

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

with extensive NMR measure-

ments led the Bristol Myers

scientists to assign the pyrro-

loindoline moiety to the d-tryp-

tophan series, and to assert an

anti relationship of the trypto-

phan derived carboxamido

function and the cis-fused junc-

tion substituents (Figure 2).

The N_bof the tryptophan de-

rived pyrroloindole is joined in

an amidic linkage to a d-threo-

iii)

i)

O

iv)

Val

O

Val

H

O

R

O

H

N

H

N

H

Thr

HO

OH

H

S

O

ii)

H

N

O

H

R

O

N

O

HO

N

O

N

H

N

O

H

N

H

N

R

Thr

•

H

OH

H

HO

OH

S

N

H

NH

ii)

H

N

H

Leu

OH

O

3

4

iv)

Figure 3. Key synthetic issues and potential bond disconnections.

nine residue. The summation of these arguments led to the

advancement of structure 1 to himastatin and structure 2 to

product A arising from reduction of the natural product with

lithium borohydride.^[2c]

monomer is apparently not available from fermentation or

from degradation of himastatin, chemical synthesis would be

required for such a study.

Broadly speaking, one could anticipate several key issues

which would require particularly close attention in attempting

a total synthesis of himastatin. In citing these problem areas,

we imply no a priori commitment as to the sequence in which

they would be addressed. Clearly, there is a need, at some

stage, to achieve coupling of the indole moieties (see arrow

i) in structure 4, Figure 3). Another potentially serious

impediment could be the building of the pyrroloindoline with

the angular hydroxyl group at C3a ii). It would be necessary to

correlate the stereochemistry of the cis pyrrolo[2,3-b]indoline

with the acyl function of the modified tryptophan, Figure 3

iii). Workable routes to the amino acids or amino acid

surrogates which are to be interpolated between the ªtrypto-

phylº carboxy and N_bfunctions would be required. A

particularly novel segment of the five-component insert

needed to establish the depsipeptide, is the rather rare

5-hydroxypiperazic acid in the d-amino acid configuration

(Figure 3 iv).^[6]Needless to say, the subtleties of protecting

these building blocks as they are entered into the evolving

depsipeptide domain must be mastered. Moreover, creation

of a setting for successful deprotection of the various subunits

without unraveling fragile functional group accommodations

would pose no small challenge to the enterprise.

OH

Ph

O

H a

HN

N

HO

H

O

HO

N

•

O

N

R

O

N

H

O

R

N

O

b

N

•

a

N

OH

H

OH

N

NH

H

O

N

H

1

OH

H

HN

R

O

N

H

HO

•

HN

O

NH

•

OH

H

N

R

NH

H

HO

2

Figure 2. Proposed structures for himastatin and degradation product.

Results and Discussion

Key issues: Given our longstanding interest in the synthesis of

various pyrroloindole alkaloids,^[3]the promising antibiotic

activity reported for himastatin,^[2a]and its nonconventional

structural features, we were naturally drawn to this molecule

as a focusing target for a study in total synthesis. From the

outset, there was no doubt that such a total synthesis venture

would prove to be difficult. However, we hoped and expected

that there could be correspondingly favorable learning

opportunities. The lessons so garnered could well go beyond

the particulars of himastatin.^{[4, 5]}

Included in the program was the goal of establishing

whether the bacteriocidal activity of himastatin arises from

the core monomer alone, or whether some form of intra-

molecular collaboration between the identical subgroups in

the ªdimerº is critical. We also hoped to undertake a more

subtle, but related inquiry, that is, whether there is a long-

range communication between the two identical domains,

which could be detected at the spectroscopic level. To deal

with these questions, it would be necessary to gain access to

the ªmonomer-likeº system 3 and further evaluation with

regard to activity as well as congruency of its depsipeptide

domains with respect to those of 1 (Figure 3). Since the

Synthesis of anti ± cis (11) and syn ± cis (14) pyrroloindoline

cores: With the potential difficulties well appreciated, we

initiated our quest by focusing on the pyrroloindoline prob-

lem. Even confining our inquiry to the cis [2,3-b] series and

setting aside considerations of absolute configuration, it

would be necessary to achieve decent control in setting the

relation of this fusion relative to the acyl (carboxamido)

center of the modified tryptophyl subunit. Put differently, if

we would commence with a suitably protected tryptophan 5,

shown in the d-configuration, it would be necessary to gain

access to the anti ± cis series (cf. 7) corresponding to the

assigned structure of himastatin (Figure 4). Ideally, at least for

purposes of molecular diversification, it could be helpful if we

could also gain stereoselective access to the syn ± cis series (cf.

6). Indeed as matters unfolded (see below), this capability

proved to be crucial.

The possibility of inducing oxidative cyclization of various

tryptophan derivatives such as 5, followed by aromatization

(loss of HX) which led to 8, had been appreciated since the

pioneering work of Witkop (Figure 5).^[7]Our first thought was

42

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0042 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

groups PG, which were surveyed for N_b, only those of the

sulfonyl type (AnthSO₂, SES, Ts) were appropriate for

subsequent oxidation of the pyrroloindole tetrasubstituted

double bond with DMDO. Other protecting groups at N_b

(Boc, CO₂Me, Ac, Tr) and other oxidants (OsO₄, MCPBA,

MMPP, Pb(OAc)₄) failed to give the desired products in

useful yields.^[9]

H

HO

H

CO₂R

HO

CO₂R

H

NH

NH-PG

NH

N

H

N

H

N

H

6

7

5

syn-cis

anti-cis

Figure 4. Oxidative cyclization of tryptophan.

The anthracene sulfonyl residue had been chosen among

the various sulfonyl groups because of the ease of its removal

under mild reducing conditions (SmI₂, THF, rt, min, or

Al(Hg), THF, H₂0, rt, h). For additional verification, the

transformation of 11 ! 12 was also accomplished. The

correlation of this compound with compounds previously

assigned to be in the ªanti ± cisº series strongly supported our

assignments.^[10]

[O]

CO₂R

X⁺

CO₂R

- HX

NH-PG

N

PG

N

H

N

H

5

8

CO₂R

PG

We next turned to the matter of gaining stereospecific

access to the syn ± cis pyrroloindoline series. At the time we

undertook this study, the idea was to expand our options for

diversifying himastatin congeners. We were mindful that this

type of transformation had not been accomplished by direct

cyclization of a tryptophan derivative with high stereoselec-

tivity.^[11]Many tryptophan-derived congeners were screened

as to their amenability to stereospecific oxidative cyclization

in the desired mode. After considerable trial and error type

research, we found that the N_b-trityl substrate, as its tert-butyl

ester (see compound 13 derived from l-tryptophan as shown),

gave 14 in 55% yield upon oxidation with DMDO and

cleavage of the trityl group apparently unaccompanied by any

anti ± cis diastereomer (Scheme 2).

CO₂R

PG

HO

-

[H ]

N

•

N

H

7

Figure 5. Witkopꢁs oxidative transposition of tryptophan.

to accomplish the required conversion through a substrate

that yields a product which can be hydrated (by oxidation

followed by reduction) with high stereoselection in the anti ±

cis sense, to reach a system of type 7. To bring this result about,

the N_bnitrogen of the tryptophan had to be subject to

deprotection of its blocking group with maintenance of the

aromatic ªhydrateº substructure. These stringent criteria

were met through the use of the anthracene sulfonyl protect-

ing group as the N_bprotecting group (PG),^[8]while a tert-butyl

ester (R) served to protect the carboxyl function (see

compound 9, Scheme 1). Treatment of 9 with NBS and

H

CO₂H

CO₂tBu

HO

a), b)

c), d)

b

NH₂

NH

NHTr

H

N

H

N

H

L - tryptophan

CO₂H

NH₂

CO₂tBu

14

13

CO₂tBu

Scheme 2. Synthesis of pyrroloindoline 14. a) TrCl, TEA, THF; b) tert-

butyl isourea, CH₂Cl₂, rt, 76% (over two steps); c) DMDO, CH₂Cl₂,

ꢀ788C; d) HOAc, MeOH, CH₂Cl₂, 55% (over two steps).

N SO₂Anth

b

H

a), b)

NSO₂Anth

c)

N

H

N

H

N

H

9

a

D-tryptophan

10

CO₂tBu

CO₂Me

At this stage we had accomplished our key early objective

of gaining smooth and highly stereoselective access to both

the anti ± cis and syn ± cis versions of the [2,3-b]-pyrroloindole

series. Our expectation was to use the former as a building

block to obtain himastatin while the latter might be useful for

SAR investigations.

HO

d), e)

f)-i)

NCO₂Me

NSO₂Anth

•

N

H

N

H

11

12

Scheme 1. Synthesis of pyrroloindoline 11. a) AnthSO₂Cl, TEA, THF;

b) tert-butyl isourea, CH₂Cl₂, rt, 70% (over two steps); c) NBS, TEA,

CH₂Cl₂, 08C to rt; d) DMDO, CH₂Cl₂, ꢀ788C; e) NaBH₄, MeOH, 08C to

rt, 75% (over three steps); f) TFA, CH₂Cl₂; g) CH₂N₂, Et₂O; h) Al(Hg),

THF, aq NH₄OAc; i) ClCO₂Me, py, CH₂Cl₂, ꢀ788C to rt, 50% (over four

steps).

Synthesis of dimer 24: Our focus then shifted to paving the

way for the bi-aryl linkage (see i) in structure 4, Figure 3). We

were rather aware of the fact that if the biaryl linkage could be

introduced at an early stage, the building of the depsipeptide

could be conducted concurrently on the two identical subunits

potentially with a notable economy of operations. We came to

favor a Stille coupling for the biaryl bond formation.^[12]This

route was all the more interesting since it had not yet been

reported in the bisindolyl series.

triethylamine generated an unstable dihydropyrroloindole 10.

Subsequent treatment of this compound with 3,3-dimethyl-

dioxirane (DMDO) at ꢀ788C followed by reduction with

sodium borohydride led to 11 as substantially a single

diastereomer. The use of a tert-butyl ester afforded a much

higher anti to syn stereoselectivity (>15:1) in the oxidation

step than the corresponding methyl ester which provided a

more modest 3:1 preference. Interestingly, of the protecting

We had of course benefited from the availability of d-

tryptophan to reach 11. Now it would be necessary to

functionalize the benzo region at the required carbon of the

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0043 $ 17.50+.50/0

43

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

indolenine. We reasoned that N_a, even if modified with a

protecting group, might direct an electrophilic functionalizing

agent to the C-5 position.

Much trial and error was required before N_aas well as N_b

were presented in an optimal way for accomplishing clean

functionalization at the C-5. In practice compound 11 lent

itself to conversion to 16 through a three-step sequence

(Scheme 3). Protecting group interplay was required at this

CO₂tBu

NCbz

CO₂tBu

NCbz

•

TBSO

RO

c), d)

a)

2

•

N

Cbz

19

Cbz

20: R = OH

21: R = TES

b)

CO₂tBu

NFmoc

CO₂Allyl

NFmoc

•

TESO

e), f)

g)

2

•

N

H

N

H

22

23

CO₂tBu

NCbz

CO₂tBu

HO

RO

a), b)

d)

CO₂Allyl

NH

•

NSO₂Anth

TESO

•

N

5-mer

2

H

Cbz

N

11

15: R = H

16: R = TBS

H

c)

24

Cbz

N

Scheme 4. Synthesis of pyrroloindoline dimer 24. a) TBAF, THF, 85%;

b) TESCl, DBU, DMF, 508C, 79%; c) H₂(1 atm), Pd/C, EtOAc; d) Fmoc-

HOSu, py, CH₂Cl₂, 87% (over two steps); e) TESOTf, lutidine, 08C to rt,

CH₂Cl₂; f) allyl alcohol, EDCI, DMAP, CH₂Cl₂, 78% (over two steps);

g) piperidine, CH₃CN, 92%.

CO₂tBu

TBSO

CO₂tBu

NCbz

•

X

•

TBSO

CbzN

tBuO₂C

f)

NCbz

•

OTBS

N

Cbz

N

Cbz

17: X = I

18: X = SnMe₃

19

e)

Scheme 3. Synthesis of pyrroloindoline dimer 19. a) Al(Hg), THF, aq

NH₄OAc; b) CbzCl, py, CH₂Cl₂, rt, 60% (over two steps); c) TBSCl, DBU,

DMF, 508C, 77%; d) ICl, 2,6-di-tert-butylpyridine, CH₂Cl₂, 81%;

e) Me₆Sn₂, [Pd(Ph₃P)₄], THF, 608C, 89%; f) [Pd₂dba₃], Ph₃As, 17, DMF,

458C, 50 ± 70%.

triethylsilyl esters and reprotection of the diacid as its bis-allyl

ester afforded derivative 23. Finally, the Fmoc functions were

cleaved upon reaction of 23 with piperidine, thereby providing

access to 24. In relying on the doubly deprotected diamine

arrangement (N_band N_a), we expected that N_bwould be more

reactive than the ªaniline-likeº N_a.

point since neither the anthracene sulfonyl group nor the free

NH or OH functions were compatible with the iodination

conditions. With the particular set of protecting groups shown,

iodination at C-5 of the indoline occurred as indicated (see

compound 17). The stannyl indole 18 was prepared from 17 as

shown. Happily and seemingly uneventfully, a palladium

mediated Stille coupling of 17 and 18 gave rise to 19. The

critical campaign for a two-fold interpolation of the depsipep-

tide between N_band the tryptophan derived carboxy center

could now commence.

Significant restructuring of this pyrroloindoline domain was

necessary to render it suitable for initial attachment to the

depsipeptide domain, as well as for macrocyclization and

ultimate deprotection (Scheme 4). For instance, parallel

studies in related series suggested that we would not be able

to deprotect an angular TBS group at C-3a when the full

macrocycle was in place. Hence, the TBS protecting group

was cleaved at the stage of 19. Fortunately, a triethylsilyl

function could be introduced at this point through the use of

TESCl in the presence of DBU to provide 21. Additionally, on

the basis of model probe studies, it appeared that reliance of a

tert-butyl ester for carboxy protection would lead to difficul-

ties in late stage deprotection. Accordingly, we replaced this

group with an allyl ester which could be cleaved through

palladium mediated p-allyl formation at a strategic point of

our choosing. Of course, the presence of an allyl group from

the start would not have been compatible with our oxidative

cyclization. The four Cbz functions were discharged under

standard hydrogenation conditions and the resulting tetra-

amine was reprotected as its bis-Fmoc derivative (see com-

pound 22). The tert-butyl ester function was cleaved with

triethylsilyltriflate and 2,6-lutidine. Hydrolysis of the resultant

Synthesis of piperazic acid 32: Our next subgoals involved

gaining access to the five building blocks comprising the

depsipeptide. The individual units would be pre-assembled to

produce the 5-mer which would be interpolated between N_b

and the erstwhile tryptophan carboxyl function. Even casual

inspection of these subunits forecasts that the most complex

of them would be the 5-hydroxypiperazic acid moiety,

properly presented for incorporation into the 5-mer. The

synthesis had to be responsive to the relative and absolute

stereochemistry required for himastatin. This subgoal was

accomplished through two different schemes.^[13]In the first

(Scheme 5), methanolysis of commercially available tosylate

25 at rt gave an epoxy ester. The oxido linkage suffered

opening with LiBr and HOAc, which gave hydroxy ester 26.

Protection of the resulting secondary alcohol as its tert-

O

a), b)

d)

O

MeO

c)

Br

OR

TsO

26: R = H

27: R = TBS

25

O

Boc

N

MeO₂C

NBoc

e)

MeO

Br

OTBS

NBoc

+

BocN

OTBS

H

28

29

30

Scheme 5. Synthesis of piperazic esters 29 and 30. a) NaOMe, MeOH, 08C

to rt; b) LiBr, THF, HOAc, rt, 93% (over two steps); c) TBSOTf, lutidine,

CH₂Cl₂, ꢀ788C to rt, 65%; d) NaHMDS, THF, DBAD, ꢀ788C to rt, 79%;

e) NaH, DMF, 08C, 44% of 29, 37% of 30.

44

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0044 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

butyldimethylsilyl ether led to 27 in good overall yield for the

three steps. Deprotonation with strong base (NaHMDS)

followed by quenching with bis-Boc azodicarboxylate afford-

ed 28 as a roughly 1:1 mixture of diastereomers. Exposure of

this mixture to NaH in DMF at 08C led to smooth cyclization

giving piperazic esters 29 and 30 after a difficult separation.

The cis-ester 30 was advanced to 32 through ester 31 in a

straightforward manner as shown (Scheme 6). Much effort

was expended to make use of the undesired trans piperazic

ester 29 by means of epimerization. Success was ultimately

O

a)

b)

O

N

HO

NBoc

NHBoc

BocN

Bn

NHBoc

35

36

H

O

Boc

N

Boc

HO₂C

N

NTeoc

N

NBoc

O

NBoc

d), e)

f), g)

c)

H

Br

O

34

OTBS

32

37

Scheme 7. Alternate synthesis of piperazic acid 32. a) LiOH, THF, 08C,

71%; b) NBS, toluene, 08C, 65%; c) NaHMDS, DMF, 08C, 64%; d) TFA,

MeOH; e) TeocCl, py, CH₂Cl₂; f) TBSOTf, lutidine, CH₂Cl₂, ꢀ788C to rt;

g) LiOH, THF, 08C, 82% (over four steps).

Boc

H

N

MeO₂C

HO₂C

NBoc

NTeoc

a), b)

c)

OTBS

30

31

32

Troc derivative 44 (Scheme 8). Of these, 39 is commercially

available. The remaining components 38, 42, and 44 were all

derived from well established chemistry.

f), g), h), i)

Boc

N

Boc

N

Boc

N

MeO₂C

NBoc

d)

e)

O

HO₂C

AllylO₂C

TBSO

O

a), b)

c), d)

e)

NH₂

OH

OTBS

NH₂

N

H

HO

TBSO

33

29

34

38

D-threonine

40

Scheme 6. Synthesis of piperazic acid 32. a) TFA, CH₂Cl₂; b) TeocCl,

iPr₂NEt, CH₂Cl₂, 92% (over two steps); c) LiOH, THF, 08C, 100%;

d) TBAF, THF, 08C to rt, 80%; e) DBU, toluene, 1108C, 69%; f) TFA,

CH₂Cl₂; MeOH; g) TeocCl, py, CH₂Cl₂; h) TBSOTf, lutidine, CH₂Cl₂,

ꢀ788C to rt; i) LiOH, THF, 08C, 82% (over four steps).

HO

O

Teoc

N

Teoc

N

HN

O

f), g)

O

AllylO₂C

N

O

AllylO₂C

O

OTBS

NH

N

H

accomplished as seen in Scheme 6. TBAF deprotection of 29

cleanly gave secondary alcohol 33 in 80% yield. Refluxing

this trans-hydroxy ester with DBU in toluene using a Dean ±

Stark trap filled with 4 molecular sieves led to a slow, but

clean conversion to the cis piperazic lactone 34. Acidic

cleavage of both Boc functions, and methanolysis of the

lactone, acylation of the remote nitrogen with TeocCl,

silylation of the alcohol function, and ester hydrolysis led to

32. In theory, a mixture of cis and trans esters (29 and 30)

could be taken on through this sequence thereby obviating the

need for a difficult chromatographic separation. Albeit less

than elegant from the standpoint of stereocontrol, the

approach is amenable to scale-up and requires a minimal

number of purification steps.

A second generation, stereoselective approach to the

piperazic acid was also developed. The route started with

the known pentenoic acid derivative, 35 (Scheme 7).^[14]

Cleavage of the acyl oxazolidinone bond was accomplished

through lithium hydroxide in THF. Bromolactonization of the

resulting acid 36 under the influence of N-bromosuccinimide

gave a ꢁ4.5:1 ratio of 37 relative to its trans counterpart.^{[15, 16]}

The bromine function was displaced upon deprotonation of

the second Boc nitrogen giving rise to lactone 34. As before,

four straightforward steps converted 34 into acid 32.

TBSO

43

41

O

TrocN

H

O

h)-k)

O

N

O

HO₂C

O

OTBS

NH

N

H

TBSO

45

Scheme 8. Synthesis of pentadepsipeptide acid 45. a) allyl alcohol, p-

TsOH, C₆H₆, Dean-Stark, 808C; b) TBSCl, imidazole, CH₂Cl₂, 99% (over

two steps); c) Fmoc-l-leucine (39), EDCI, DMAP, CH₂Cl₂; d) piperidine,

CH₃CN, 88% (over two steps); e) piperazic acid (32), HATU, HOAt,

collidine, CH₂Cl₂, 70 ± 95%; f) Fmoc-OCH(iPr)COCl (42), collidine,

CH₂Cl₂, 08C to rt; g) piperidine, CH₃CN, 92% (over two steps); h) Troc-

d-valine (44), IPCC, Et₃N, DMAP, CH₂Cl₂, ꢀ208C to rt; i) ZnCl₂,

CH₃NO₂; j) TBSOTf, lutidine, CH₂Cl₂, ꢀ788C to rt; k) [Pd(Ph₃P)₄],

PhSiH₃, THF, 08C to rt, 69% (over four steps).

With the requisite building blocks in hand, assembly of the

5-mer domain could commence. Coupling of 38 with 39 was

followed by cleavage of the Fmoc group to release the leucine

amino function (see compound 40). The latter was acylated

with the piperazic acid 32, to provide tripeptide 41. Fortu-

nately, the relatively unreactive NH moiety of the piperazic

acid 41 was subjected to acylation by 42. Following depro-

tection, building block 43 was in hand. Acylation of 43 with 44,

deprotection of the three silyl groups, reprotection of the diol,

and deprotection of the allyl ester gave the required acid 45.

Pentadepsipeptide synthesis: The other four fragments which

were employed in building the peptidal domain were: i) the d-

threonine derivative 38, ii) Fmoc-l-leucine 39, iii) the

hydroxyisovaleryl derivative 42 (prepared from commercially

available (S)-hydroxyisovaleric acid), and iv) d-valine, as its

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0045 $ 17.50+.50/0

45

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

This sequence was required since projected removal of the

Teoc group on the piperazic fragment was highly problematic

in the context of the full himastatin core structure. Hence, this

deprotection had to precede incorporation of the 5-mer into

the sensitive himastatin architecture. Fortunately, the free

piperazic NH group in 45 is relatively unreactive. With careful

management of reaction conditions, it proved to be possible to

carry this NH group forward into the synthesis.

ground work to enable the final deprotection had been

carefully surveyed. In the event, concurrent deprotection of

the angular pyrroloindoline hydroxyl (from its TES deriva-

tive) and the piperazic and threonine based hydroxyl groups

(from their TBS ethers) was possible and the goal structure 1

was in hand.

The excitement at having ostensibly reached our synthesis

end point was short lived. Unfortunately, it was clear that the

¹H NMR spectrum of the final product, prepared in our

laboratory by the synthesis described above, did not corre-

spond with that recorded for naturally derived himastatin,^[2b,c]

also presumably corresponding to 1. Particularly striking were

two upfield signals at d 0.55 and 0.35 in synthetic 1,

attributable to the nonequivalent methyl centers in the valine

isopropyl moiety. No such upfield resonances are present in

the reported ¹H NMR spectrum

Synthesis of dimer 1 and monomer 3: The next stage of the

venture would involve interpolation of the 5-mer, now

identified as 45, between N_band the acyl group of 24

(Scheme 9). We would be confronting the question of the

differentiability of the two NH groups of 24. We would, at

of himastatin isolated from nat-

ural sources.

Additionally, the gross prop-

O

R'N

H

CO₂R

TrocN

H

O

TESO

N

erties of the synthetic construct

differed from those described

for himastatin. Thus, the natu-

ral product was reported to be

soluble in a variety of organic

CO₂Allyl

+

N

a)

2

N

O

NH

O

•

O

HO₂C

O

OTBS

N

OTBS

NH

H

N

H

N

H

TBSO

24

45

46: R = Allyl, R' = Troc

47: R = H, R' = Troc

b)

solvents

(CH₂Cl₂,

CHCl₃,

EtOAc, MeOH) with an t_R

0.6 (8:1 CHCl₃/MeOH). Al-

though soluble in DMSO, syn-

thetic 1 is insoluble in CH₂Cl₂,

CHCl₃, EtOAc, MeOH, and is

much more polar with an t_R

0.15 (8:1 CHCl₃/MeOH).

OPG'

H

N

H

O

H

HN

N

PG'O

H

O

PGO

N

ï

N

O

c), d)

O

N

•

N

OPG

H

OPG'

H

N

NH

O

H

N

H

PG'O

Moreover, the synthetic logic

to reach dimer 1 was applied to

the synthesis of monomeric 3.

For this goal, we returned to

compound 11 (Scheme 10). The

tert-butyl group of the ester

function was cleaved and an

48: PG = TES, PG' = TBS

1: PG = PG' = H

e)

Scheme 9. Synthesis of isohimastatin 1. a) HATU, HOAt, collidine, CH₂Cl₂, ꢀ108C to rt, 67%; b) [Pd(Ph₃P)₄],

PhSiH₃, THF, 08C to rt, 86%; c) Pb/Cd couple, THF, aq NH₄OAc; d) HATU, HOAt, iPr₂NEt, DMF, 08C to rt;

e) TBAF, HOAc, THF, 48% (over three steps).

allyl ester was built. N_bwas exposed by reductive cleavage

(Al(Hg)) of the sulfonamide (see compound 50). Coupling of

50 specifically at N_bwith acid 45 led to seco-system 51. The

synthesis proceeded in much the same manner as that used to

reach 1. Once again, the ¹H NMR spectrum of this monomer 3

had upfield peaks at d 0.55 and 0.35 which had no counter-

part in the reported spectrum of himastatin.^[2b,c]

every stage, be testing the stability of the pyrroloindoline

moiety, bearing the angular silyl protected tertiary alcohol

and, as discussed above, the accomodatability of the free

NH group of the piperazic acid. Subunit 45 would be

presented as the deprotected carboxyl in the threonine

residue. The valine amino terminus is protected as a Troc-

urethane. Correspondingly, in 24, the tryptophan derived

carboxyl of the pyrroloindole sector is protected as an allyl

ester, leaving N_aand N_bas free NH groups to potentially

compete for acylation by the threonine carboxyl function. As

noted earlier, we hoped that the ªaniline-likeº N_awould be

less reactive than N_b.

Fortunately, coupling of 24 with two units of 45 under the

highly controlled conditions shown, occurred exclusively at

the N_bcenters giving rise to 46. The next phase involved

exposure of the trytophyl carboxyl group (from its allyl ester)

and the valine amino group from its Troc derivative. These

steps were accomplished in the manner shown. In the defining

step of the synthesis, two-fold cyclization of the crude diamino

acid led to the bis-macrolactam 48. As discussed above, the

Correction of stereochemistry: At this stage, two possible

conclusions suggested themselves. It could be argued that

during the course of our total synthesis, an unanticipated

epimerization or even skeletal rearrangement had taken

place. Upon carefully and critically reviewing our method-

ology and data, we came to regard this family of possibilities

as most unlikely. This view led us to the next alternative, that

is that the structure (possibly at the stereochemical level) of

the ªrealº naturally occurring himastatin is not identical to

that which had been assigned. Favoring this accounting for the

mysterious non-congruence of synthetic end product and

naturally derived material, we then directed our efforts

46

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0046 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

CO₂Allyl

NH

to the previously synthesized tryptophan derivative 14. It

will be recalled that this substance had been derived from

l-tryptophan, but is in the syn ± cis series. The methodology

for converting 14 to the corresponding d-valinol derivative

54 was, in principle, well in hand, provided that the protocols

which had been operative in the anti ± cis series leading to 1

and 3 would now be transferable to the syn ± cis regime.

Indeed, this turned out to be the case and compound 54

was synthesized as shown. In this case, the key features of the

high field ¹H NMR spectrum of the synthetic syn ± cis

ªmonomerº version of the bis-valinol degradation product

matched the reported values for the dimer. Thus, the

resonances of the nonequivalent isopropyl methyl groups of

54 are at d 0.92 and 0.88; this matches closely the reported

resonances at 0.94 and 0.89 of the degradation product

previously reported to be 2, but which in fact now would

require re-formulation.

Further experimental evidence gleaned from the literature

bears on the anti ± cis versus syn ± cis stereochemical issue.

Both trans and cis esters 12 and 55 have been characterized by

¹H NMR and by X-ray crystallography (Figure 6).^[10]In 12, the

methyl ester singlet appears at d 3.20 while that in 55 is seen

at d 3.80. From the crystallographically derived structure of

12, the methyl ester is clearly on the concave face of the

tricycle and would be expected to experience the shielding

effect of the aromatic ring. This chemical shift difference

between the trans and cis esters follows the trend we were now

proposing to extend to compounds 53 and 54.

CO₂Allyl

CO₂tBu

TESO

HO

c)

a), b)

NSO₂Anth

•

N

H

11

49

50

O

R'N

O

HN

H

O

HO

TESO

N

O

CO₂R

N

O

d)

f)-h)

N

O

•

O

OH

OTBS

N

NH

H

N

H

N

H

HO

TBSO

51: R = Allyl, R' = Troc

52: R = H, R' = Troc

3

e)

Scheme 10. Synthesis of trans-monomer 3. a) TESOTf, iPr₂NEt, CH₂Cl₂,

08C to rt; b) allyl alcohol, EDCI, DMAP, CH₂Cl₂, 78% (over two steps);

c) Al(Hg), THF, H₂O, 69%; d) acid 45, HATU, HOAt, collidine, CH₂Cl₂,

ꢀ108C to rt, 66%; e) [Pd(Ph₃P)₄], PhSiH₃, THF, 08C to rt, 92%; f) Pb/Cd

couple, THF, aq NH₄OAc; g) HATU, HOAt, iPr₂NEt, DMF, 08C to rt;

h) TBAF, HOAc, THF, 76% (over three steps).

toward ascertaining the nature of the discrepancy. For this

purpose, we focused on opportunities suggested by the

previously described degradation product, assigned to be 2,

containing the sensitive [2,3-b]-pyrroloindoline moiety as well

as the d-valinol side chain. Since the ¹H NMR spectra of 1 and

the monomeric 3 were virtually identical in the depsipeptide

sector, we reasoned that a comparison of the high field spectra

reported for the degradation product presumed to be 2 with a

candidate monomeric version thereof, while imperfect in

terms of intellectual rigor, would be revealing in practice.

Hence, we returned to compound 11 (Scheme 11). It was

CO₂Me

HO

NCO₂Me

•

N

H

OH

HN

12

55

CO₂tBu

HO

Figure 6. Structures of known relative stereochemistry confirmed by X-ray

analysis.

O

a)-c)

NSO₂Anth

NH

•

N

H

11

53

On the basis of these findings and considerations, we were

prepared to postulate that the relationship of the pyrroloindo-

line junction moieties and the tryptophane-carboxamido

group in naturally occurring himastatin is syn rather than

anti as previously proposed.

OH

HN

CO₂tBu

NH

HO

O

d)-h)

NH

•

N

H

N

H

Having reached this stage in our reformulation of the

structure of himastatin, there were still uncertainties. Of

course, we could not be sure that the depsipeptide domain of

the natural product had in fact been correctly formulated.

14

54

Scheme 11. Synthesis of degredation monomers 53 and 54. a) TFA,

CH₂Cl₂; b) d-valinol, EDCI, DMAP, CH₂Cl₂; c) Al(Hg), THF, H₂O, 25%

(over three steps); d) Fmoc-HOSu, pyridine, CH₂Cl₂; e) TMSOTf, lutidine,

CH₂Cl₂, 08C to rt; f) d-valinol, EDCI, DMAP, CH₂Cl₂; g) piperidine,

CH₃CN; h) TFA, CH₂Cl₂, 28% (over five steps).

1

Given the similarity of the high field H NMR spectra of

synthetic 1 with that described for himastatin (neglecting the

difference in the methyl resonances in the isopropyl group)

we started with the assumption that the depsipeptide sector of

natural himastatin is as asserted. However, even if this is

assumed to be the case, there was no evidence with which to

formulate the absolute stereochemistry of the tryptophan

derived pyrroloindoline. In principle, it could indeed be

derived from d-tryptophan with the discrepancy rooted

exclusively in the syn relationship of the cis junction to the

carboxamido group. Alternatively, the tryptophan in question

may be l-configured. In this event, the ªabsolute stereo-

readily converted to 53 as shown. Once again, the chemical

shift data for the isopropyl group in 53 (d 0.78, 0.62) and

those reported for 2 (d 0.94 and 0.89) were not in accord.

Hence, there was strong indication of a discrepancy between

the relative configurations proposed for himastatin in the

pyrroloindole valinol sector and the synthetically derived

structure 53, which we took to be secure.

Reflecting a growing suspicion as to the nature of the

discrepancy between synthetic 1 and himastatin, we returned

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0047 $ 17.50+.50/0

47

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

chemistryº of the pyrroloindoline junction would have been

correctly formulated. However, the l-carboxamido configu-

ration would be S rather than the R in 1, and its relationship to

the junction functions, accordingly, would be syn rather than

anti.

In the event, upon reaction of 56 with ICl, an iodine atom was

introduced at C-5 (see compound 57). A portion of 57 was

converted to aryl stannane 58 as shown. At this stage, we

could hope to take advantage of precedent, demonstrated in

the anti ± cis series, to join sophisticated pyrroloindolines in

their ªbenzoº sectors through carbon to carbon bond

formation by an extension of the Stille reaction. In the event

compounds 57 and 58 did couple smoothly under the

conditions shown to produce 59.

Once again, a restructuring of the protecting groups was

required to facilitate interpolation of the 5-mer and to

anticipate the sensitivity associated with global deprotection

in the last step. The protocols which we employed to convert

the trans-dimer 19 to a suitable coupling partner (i.e., 24) were

equally successful in the syn ± cis series (see 59 ! 64,

Scheme 13). Thus, cleavage of the angular TBS group exposed

the angular alcohol which was quickly reprotected as its

In evaluating the issue of assignment of the tryptophan

derived (d or l) pyrroloindoline and the clear need to reverse

the anti ± cis assignment to a syn ± cis arrangement, the

chiroptical data cited in support of the original assignment is

more readily accommodated in the context of an l-tryptophan

derived insert. By reversing this configuration of the carbox-

amido group (d-tryptophyl ! l-tryptophyl) the junction

substituents of the pyrroloindoline would be the same as that

originally proposed in 1. We also noted with interest that if the

pyrroloindoline were indeed derived from l-tryptophan and

the 5-mer were indeed properly formulated, the components

in the depsipeptide domain are presented in alternating d-

and l-configurations. Recent studies by Ghadiri on the special

properties associated with related systems might explain the

markedly different solubility and polarities described above

for himastatin and 1.^{[17a, b, c]}

Given these considerations we advanced the structure 68

for himastatin and 69 for its bis-valinol degradation product.

We well recognized that our proposal was not secure. In

addition to the unproven assignment of an l-tryptophan

absolute configuration for the pyrroloindoline sector, we had

no verification for supposing that the depsipeptide 5-mer

region of himastatin would correspond in all its detail to that

which had been assigned. Given the practicalities of our

situation, wherein no significant quantities of himastatin were

available, chemical synthesis emerged as the only means to

solve the problem of the structure of himastatin in a rigorous

fashion.

R

N

CO₂R

NR'

TESO

CO₂tBu

NR'

•

TESO

R'N

tBuO₂C

2

•

OTES

N

H

R

a), b)

59

60: R = R' = Cbz

61: R = H; R' = Fmoc

62: R = H; R' = Fmoc

63: R = Allyl; R' = Fmoc

64: R = Allyl; R' = H

f)

c), d)

e)

g)

Scheme 13. Synthesis of pyrroloindoline dimer 64. a) TBAF, THF;

b) TESCl, DBU, DMF, 95% (over two steps); c) H₂(1 atm), Pd/C, EtOAc;

d) Fmoc-HOSu, py, CH₂Cl₂, 96% (over two steps); e) TESOTf, lutidine,

CH₂Cl₂, 08C to rt; f) allyl alcohol, DBAD, Ph₃P, THF, 81% (over two

steps); g) piperidine, CH₃CN, 74%.

triethylsilyl derivative 60. The two Cbz functions were cleaved

by hydrogenolysis and two Fmoc groups (one per domain)

were introduced at N_b(see compound 61). The moment was in

hand for cleaving the tert-butyl ester so that it could be

replaced by an allyl ester function in turn poised for a highly

specific deprotection under mild conditions at a strategic

point of our choosing. This subgoal was accomplished by

treatment of compound 61 with TES triflate, as shown,

resulting in exposure of the free ªtryptophyl-likeº carboxylic

acid 62 in each component of the dimer. Each acid was

converted to its allyl ester as indicated (see compound 63).

There remained only the cleavage of the Fmoc group to allow

for presentation of the pyrroloindoline in a form which could

be acylated by the threonine carboxyl of the previously

encountered 45. The Fmoc groups, protecting the two

identical N_bfunctions, were readily cleaved providing com-

pound 64 for presentation to 5-mer 45.

Synthesis of himastatin (68): Accordingly, we returned to l-

tryptophan and to the previously described 14, hoping that the

methodology carefully developed in closely related models

would prove to be adequate (Scheme 12). The protecting

groups (of 14) were re-organized in a fashion suggested by

previous experiences. Cbz groups were introduced at N_aand

N_b. The junction hydroxyl group was protected as its TBS

ether (see compound 56). We were now poised to undertake

the oxidative dimerization of a suitably presented monomer.

CO₂tBu

NCbz

HO

TBSO

a), b)

c)

NH

H

•

N

H

N

Cbz

14

56

Fortunately, acylation occurred smoothly at both N_b

functions in 64 with the threonine derived carboxyl in 45

providing a 60% yield of 65 (see Scheme 14). We were now

well positioned to take advantage of the highly specific

deprotection of the two allyl esters of the identical pyrro-

loindoline sectors (see compound 66). Once again, cleavage of

the two identical Troc groups served to expose the free amino

groups of the valine residues. Fortunately, the resultant

diamino acid underwent macrolactamization mediated by

HATU to give rise to the lactam ester 67. There remained the

Cbz

CO₂tBu

N

TBSO

CO₂tBu

NCbz

•

X

•

TBSO

CbzN

tBuO₂C

e)

NCbz

•

OTBS

N

Cbz

N

Cbz

59

57: X = I

d)

58: X = SnMe₃

Scheme 12. Synthesis of pyrroloindoline dimer 59. a) CbzCl, py, CH₂Cl₂;

b) TBSCl, DBU, DMF, rt, 61% (over two steps); c) ICl, 2,6-di-tert-

butylpyridine, CH₂Cl₂, 73%; d) Me₆Sn₂, [Pd(Ph₃P)₄], THF, 608C, 87%;

e) [Pd₂dba₃], Ph₃As, 57, DMF, 458C, 83%.

48

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0048 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

and, thence the N-Troc function

afforded the seco acid. Macro-

lactamization followed by de-

protection provided ªhimasta-

tin monomerº 74.

An evaluation of the antibi-

otic properties of four com-

pounds: himastatin (68), isohi-

mastatin (1) (originally pro-

posed to be himastatin), anti-

monomer 3 and syn-monomer

74 was undertaken.^[18]A variety

O

R'N

H

CO₂R

TrocN

CO₂Allyl

H

O

TESO

N

a)

2

+

NH

N

•

O

•

O

HO₂C

O

OTBS

N

OTBS

NH

H

N

H

N

H

TBSO

45

64

65: R = Allyl, R' = Troc

66: R = H, R' = Troc

b)

OPG'

H

N

H

O

HN

N

HN

N

PG'O

H

O

PGO

c), d)

N

ï

N

O

of

microorganisms

were

N

•

N

OPG

H

OPG'

NH

H

N

screened. Indeed, himastatin

(68) as reported strongly inhib-

ited the growth of Gram-Pos-

itive bacteria such as Entero-

cocci faecalis, whereas isohi-

mastatin (1) was inactive.

We then evaluated monomer

compounds 3 and 74. It was

found that 3, which corresponds

NH

O

H

N

H

PG'O

67: PG = TES, PG' = TBS

68: himastatin, PG = PG' = H

e)

Scheme 14. Synthesis of himastatin 68. a) HATU, HOAt, collidine, CH₂Cl₂, ꢀ108C to rt, 60%; b) [Pd(Ph₃P)₄],

PhSiH₃, THF, 08C to rt, 81%; c) Pb/Cd couple, THF, aq NH₄OAc; d) HATU, HOAt, iPr₂NEt, DMF, 08C to rt;

e) TBAF, HOAc, THF, 34% (over three steps).

otherwise nontrivial matter of exposing the six hydroxyl

groups of the final target. It was for the purpose of paving the

way for such a step that we had taken the pains to introduce a

TES group to protect the two equivalent hydroxyl functions at

C-3a. Indeed, it proved possible to cleave the six silyl groups

thereby providing 68.

CO₂Allyl

CO₂tBu

TESO

HO

a)

b)-d)

NH

NFmoc

•

N

H

N

H

14

70

71

Happily, the high field proton spectrum of the fully

synthetic himastatin (68) was identical with that pub-

lished for the natural product. Thus encouraged, we obtained

from the Bristol Myers group, a small amount of natural

O

R'N

H

HN

H

O

HO

N

O

TESO

N

CO₂R

N

O

N

O

g)-i)

e)

N

O

•

O

•

O

OH

N

OTBS

N

1

NH

himastatin. We measured the 500 MHz H NMR spectra of

H

N

H

N

H

HO

TBSO

our fully synthetic material (which we thought to be 68) along

side that of the reference sample. It was clear by an overlay of

these richly detailed spectra that the total synthesis of

himastatin had been accomplished and that its structure is

indeed 68. The bis-valinol degradation product is therefore

69 (Figure 7).

72: R = Allyl, R' = Troc

73: R = H, R' = Troc

74: "himastatin monomer"

f)

Scheme 15. Synthesis of cis-monomer 74. a) Fmoc-HOSu, py, CH₂Cl₂,

55%; b) TESOTf, lutidine, CH₂Cl₂, 08C to rt; c) allyl alcohol, DBAD, Ph₃P,

THF; d) piperidine, CH₃CN, 46% (over three steps); e) acid 45, HATU,

HOAt, collidine, CH₂Cl₂, ꢀ108C to rt, 52%; f) [Pd(Ph₃P)₄], PhSiH₃, THF,

08C to rt, 92%; g) Pb/Cd couple, THF, aq NH₄OAc; h) HATU, HOAt,

iPr₂NEt, DMF, 08C to rt; i) TBAF, HOAc, THF, 78% (over three steps).

OH

H

HN

N

H

HO

•

HN

O

NH

•

OH

to the monomer version of isohimastatin (1), and 74, which

corresponds to the monomer version of himastatin (68), were

both inactive when tested against a variety of Gram-Positive

bacteria. Thus, both the ªmonomerº 3 and dimer 1, in the anti

series based on d-tryptophan and himastatin monomer (74) in

the syn series based on l-tryptophan are all inactive. The role

of the special topographic features of the alternating d and l

subunits in the 6-mer depsipeptide remains to be establish-

ed.^[17]

It is also well to note that our findings in both the chemistry

and biology of himastatin accord well with the very recent

discovery of Umezawa in connection with the novel antibiotic

chloptosin subsequent to our disclosures on himastatin.^[19]

Chloptosin (see Figure 8), like himastatin, is a potent anti-

biotic against Gram-Positive strains. It, too, contains a biaryl

linkage (now with ortho ± orthoꢁ chlorine substituents). As in

H

N

NH

H

HO

69

Figure 7. Corrected structure for himastatin degradation product.

In anticipation of comparative measurements of antibiotic

activity in the himastatin series, we directed our attention to

the synthesis of 74, the ªmonomericº version of himastatin.

The methodology to synthesize this compound was by now

well in hand. For this purpose we returned to compound 14

(Scheme 15). Selective Fmoc urethane formation occurred at

N_b(see compound 70). Protection of the angular hydroxy

group was followed by deprotection of the tert-butyl ester and

reprotection of the carboxyl group as an allyl ester. Ready

discharge of the Fmoc group led to 71, which was acylated

with 45 (see compound 72). Cleavage of the allyl ester gave 73

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0049 $ 17.50+.50/0

49

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

OH

H

N

O

H

HN

a

HN

N

HO

H

O

HO

N

HO

N

•

O

N

R

O

N

b

N

R

N

O

H

R

N

b

N

•

a

N

H

OH

•

H

NH

H

OH

N

H

NH

H

O

N

H

N

H

OH

1

3

OH

O

H

N

HN

H

O

H

HN

HO

N

HN

O

N

HO

H

N

H

N

O

HO

•

O

N

H

N

O

N

H

O

•

N

OH

N

•

N

OH

O

NH

H

N

H

OH

N

NH

H

O

N

H

HO

68 himastatin

74 "himastatin monomer"

H₃CO

HN

OH

N

H

N

CH₃

H₃C CH₃

HN

Cl

H

N

O

•

O

HN

H

O

HO

N

O

N

O

N

OH

Cl

O

H

N

NH

•

O

N

NH

OCH₃

H

N

H

H₃C CH₃

H₃C

OH

chloptosin

Figure 8. Structures of isohimastatin 1, trans-monomer 3, himastatin 68, cis-monomer 74, and chloptosin.

himastatin, there is a syn ± cis pyrroloindoline motif as part of

a hexapeptide ensemble. Once again, the hexapeptide (in-

cluding two oppositely configured piperazic acids) is arranged

with alternating d and l components. Hence, chloptosin

partakes of those structural features of himastatin which we

believe are critical to its function.

isohimastatin monomer (3), are virtually inactive. More

surprising was the finding that the syn-monomer 74, corre-

sponding in detail to the himastatin on all of its stereo-

chemical relationships, is inactive.

For the moment, we assume that antibiotic properties of

himastatin follow closely from the particular alternating d-

and l substituents in the 6-mer and are contingent on the

bidomainal motif. This pattern is simulated in a new antibiotic

chloptosin and suggests a possible structural trend for

orienting future SAR experiments.

Conclusion

At the chemical level, the synthesis provided a framework

for solving the issue of achieving smooth stereospecific access

to both the anti:cis and syn:cis pyrroloindoline subunits. The

himastatin problem also brought with it the challenge of

practical syntheses of 5-hydroxypiperazic acids in enantio-

merically pure form. While improvements in terms of the

stereoselectivity of our synthesis these piperazic acids can

easily be imagined, the optically pure compounds are now

accessible in multigram scale. Finally, it was necessary to sort

out the optimal protecting group arrangements for facilitating

orderly synthesis of himastatin including definition of con-

ditions for enabling global deprotection. We are confident

that the lessons learned from himstatin will transcend the hard

won successful total synthesis shown here. Indeed, we expect

that the methodology employed in reaching himastatin and

ªisohimastatinº will find broad application in the field of

biologically active, structurally complex depsipeptides. It can

well be argued that the himastatin problem is indeed

illustrative of the discovery dimension of the science of total

synthesis.

In retrospect, we had underestimated the scope and complex-

ity of the himastatin problem. As matters transpired, chemical

synthesis was able to demonstrate that the structure initially

proposed for himastatin (i.e., 1) was correct, except for the

absolute configuration of the tryptophyl carboxamido center

and its relationship to the cis-fused [2,3-b]-pyrolloindoline

substructure. We now refer to compound 1 as ªisohimastatin.º

Furthermore, the correct structure of naturally occurring

himastatin was shown to be 68 and its degradation product

retaining a single valinol unit per monomer domain is 69.

In addition, through the medium of total synthesis, the

outlines of a biological pharmacophore have been sketched

out. It is critical to maintain the stereochemistry of himastatin

in the pyrroloindoline ± depsipeptide sphere. The cyclic dep-

sipeptide of the active himastatin is characterized by alter-

nating d- and l subunits in the ª6-mer.º Moreover, it was

ꢀ

found that the presence of a biaryl arrangement with a C C

bond connecting the identical subunits is also necessary for

biological activity. Thus ªisohimastatinº (1), as well as

50

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0050 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

(6.70 mL, 47.9 mmol) was added. The resulting orange solution was

maintained at 08C for 30 min and then concentrated. The resulting residue

was quickly purified by chromatography on silica gel (3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes) to give 10 as an orange-red

solid which was used without further purification.

Experimental Section

General methods: All reactions were run under an atmosphere of N₂or

argon and concentrations were performed under reduced pressure with a

Büchi rotary evaporator, unless stated otherwise. Tetrahydrofuran (THF),

Et₂O, and CH₂Cl₂were degassed with argon and then passed through 4 Â

36 inch columns of anhydrous neutral A-2 alumina (8 Â 14 mesh; LaRoche

Chemicals, activated under a flow of argon at 3508C for 3 h) to remove

H₂O. Toluene was degassed with argon and then passed through one 4 Â

36 inch column of Q-5 reactant (Englehard; activated under a flow of 5%

hydrogen/nitrogen at 2508C for 3 h) to remove O₂then through one 4 Â

36 inch column of anhydrous alumina to remove H₂O. Triethylamine

(Et₃N), hexane, pyridine, and diisopropyethyllamine (iPr₂NEt) were

distilled from CaH₂at atmospheric pressure. Reagents were purchased

from Aldrich Chemical Company unless noted otherwise.

A solution of DMDO (ꢁ0.08m in acetone, 200 mL, 16.0 mmol) was added

in one portion to a ꢀ788C solution of this orange-red solid and CH₂Cl₂

(100 mL). The resulting pale yellow solution was aged at ꢀ788C for 10 min

and then concentrated. NaBH₄(1.80 g, 47.9 mmol) was added to a 08C

solution of this crude residue and MeOH (100 mL). The resulting solution

was allowed to warm to rt over ꢁ4 h and then was diluted with EtOAc

(300 mL) and quenched with 1m HCl (100 mL). The layers were separated

and the organic layer was washed with 1m HCl (3 Â 100 mL), brine (2 Â

100 mL), dried (MgSO₄), and concentrated to give 11 (8.00 g, 97%) as a

pale yellow solid that was used without further purification.

A small portion of this solid was purified for analysis by chromatography on

silica gel (3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes

then Et₂O) to give a light yellow solid that was homogeneous by TLC

analysis: m.p. 188 ± 1918C; ¹H NMR (500 MHz, CDCl₃): d 9.31 (d, J

9.3 Hz, 2H), 8.64 (s, 1H), 7.98 (d, J 8.4 Hz, 2H), 7.68 ± 7.60 (m, 2H), 7.50 ±

7.42 (m, 2H), 7.09 (d, J 7.7 Hz, 1H), 7.04 (t, J 7.7 Hz, 1H), 6.67 (t, J

7.4 Hz, 1H), 6.50 (d, J 7.9 Hz, 1H), 5.08 (s, 1H), 4.81 (d, J 8.8 Hz, 1H),

2.66 (d, J 12.8 Hz, 1H), 2.56 (dd, J 12.8, 9.3 Hz, 1H), 0.86 (s, 9H);

¹³C NMR (125 MHz, CDCl₃): d 169.2, 149.6, 136.4, 131.1, 131.0, 130.7,

129.4, 128.9, 128.5, 127.8, 125.2, 124.8, 123.8, 119.6, 110.5, 88.2, 82.9, 81.7,

62.0, 41.3, 26.9; IR (film): nÄ 3434, 3342, 3019, 1718, 1614, 1335, 1216,

1150 cm^ꢀ1; [a]²_D⁵ ꢀ180 (c 0.65 in CHCl₃).

N_b-Trityl-l-tryptophan tert-butyl ester (13): N,N'-Diisopropyl-O-tert-butyl-

isourea (36.0 g, 179 mmol) was added to a 08C solution of crude N_b-trityl-l-

tryptophan (40.0 g, 89.7 mmol) and

CH₂Cl₂(200 mL). The resulting mix-

ture was allowed to warm to rt over-

night and after 36 h, was filtered

through a pad of silica gel (Et₂O),

and concentrated. This crude residue

Instrumentation and chromatography: 400 MHz ¹H NMR and 125 MHz

¹³C NMR spectra were measured on a Bruker QE 400 FT NMR. ¹H NMR

chemical shifts are reported as d values in ppm. ¹H NMR coupling

constants are reported in Hz and refer to apparent multiplicities and not

true coupling constants. Multiplicity is indicated as follows: s (singlet), d

(doublet), t (triplet), q (quartet), m (multiplet), appd (apparent doublet),

appt (apparent triplet), dd (doublet of doublets); brs (broad singlet), etc.

Mass spectra were measured on a Perkin ± Elmer Sciex Model API-100

spectrometer with electrospray. TLC and column chromatographies were

done with E. Merck silica gel. Acronyms: MMPP Magnesium

monoperoxyphthalate hexahydrate, HATU O-(7-azabenzotriazol-1-yl)-

N,N,N',N'-tetramethyluronium hexafluorophosphate, HOAt 1-hydroxy-

7-azabenzotriazole, EDCI 1-(3-dimethylaminopropyl)-3-ethylcarbodi-

imide hydrochloride, DMDO 3,3-dimethyldioxirane, DBAD di-tert-

butyl azodicarboxylate, IPCC isopropenyl chloroformate, NBS N-bro-

mosuccinimide, Cbz benzyloxycarbonyl, dba 1,5-dibenzylideneace-

tone, DBU 1,8-diazabicyclo[5.4.0]undec-7-ene, Fmoc (9H-fluoren-9-yl-

methoxy)carbonyl, HOSu N-hydroxysuccinimide; TBAF tetrabuty-

lammonium fluoride; TBS tert-butyldimethylsilyl, TES triethylsilyl,

Tf trifluoromethylsulfonyl, Tr triphenylmethyl, Teoc 2-(trimethyl-

silyl)ethyloxycarbonyl, TFA trifluoroacetic acid, HMDS hexamethyl-

disilazane, Troc (2,2,2-trichloroethoxy)carbonyl.

was purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to give 13

(34.0 g, 76%) as a near colorless foam that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 8.02 (s, 1H), 7.58 (d, J 7.6 Hz,

1H), 7.46 (d, J 7.3 Hz, 6H), 7.35 ± 7.04 (m, 12H), 6.91 (s, 1H), 3.67 (brs,

1H), 3.15 ± 3.02 (m, 2H), 2.72 (brs,1H), 1.00 (s, 9H); ¹³C NMR (125 MHz,

CDCl₃): d 174.2, 146.3, 136.0, 128.7, 127.9, 127.6, 126.1, 123.1, 121.6, 119.3,

119.0, 111.6, 110.9, 80.1, 71.1, 57.2, 31.8, 27.7; IR (film): nÄ 3423, 3057, 2977,

2930, 1717, 1940, 1455, 1217, 1152, 747 cm^ꢀ1; ES-MS: m/z: calcd for

N_b-Anthracenesulfonyltryptophan tert-butyl ester (9):

A solution of

anthracenesulfonyl chloride (8.00 g, 29.0 mmol) in THF (50 mL) was

added to a 08C solution of d-trypto-

phan (5.00 g, 24.2 mmol), Et₃N

(10.0 mL, 72.5 mmol) and THF/H₂O

(2:1, 300 mL). After 4 h, the mixture

was diluted with EtOAc (100 mL) and

C₃₄H₃₅N₂O₂[MH] : 503.26, found: 503.26; [a]²_D⁵ 2.98 (c 1.0 in CHCl₃).

acidified with aqueous 1m HCl until

3a-Hydroxypyrroloindoline tert-butyl ester (14): A solution of DMDO

(ꢁ0.08m in acetone, 250 mL, 19.9 mmol) was added in one portion to a

ꢀ788C solution of 13 (10.0 g,

pH ꢁ3.0. The layers were separated and the aqueous layer was extracted

with EtOAc (2 Â 50 mL). The combined organic layers were washed with

aqueous 1m HCl (2 Â 50 mL), brine (1 Â 50 mL), dried (Na₂SO₄), and

concentrated to give the crude acid as a yellow oil.

19.9 mmol) and CH₂Cl₂(100 mL).

The resulting solution was aged at

A mixture of the crude acid, N,N'-diisopropyl-O-tert-butylisourea (19.6 g,

98.0 mmol) and CH₂Cl₂(100 mL) was aged at rt for 36 h. The solids were

removed by filtration (Et₂O) and the filtrate was concentrated. This crude

residue was purified by chromatography on silica gel (3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes) to give 9 (8.60 g, 70%) as a

pale yellow solid that was homogeneous by TLC analysis: m.p. 166 ± 1688C;

¹H NMR (500 MHz, CDCl₃): d 9.21 (d, J 9.2 Hz, 2H), 8.45 (s, 1H), 7.92

(s, 1H), 7.86 (d, J 8.3 Hz, 2H), 7.60 ± 7.55 (m, 2H), 7.45 ± 7.40 (m, 2H), 7.20

(d, J 7.9 Hz, 1H), 7.10 (d, J 8.1 Hz, 1H), 7.03 (t, J 7.4 Hz, 1H), 6.88 (d,

J 2.1 Hz, 1H), 6.85 (t, J 7.2 Hz, 1H), 5.81 (d, J 8.0 Hz, 1H), 4.25 ± 4.18

(m, 1H), 3.07 (dd, J 14.7, 5.6 Hz, 1H), 2.98 (dd, J 14.7, 7.0 Hz, 1H), 0.98

(s, 9H); ¹³C NMR (125 MHz, CDCl₃): d 170.2, 135.6, 130.8, 130.1, 129.2,

128.6, 128.5, 126.7, 125.0, 124.6, 123.1, 121.6, 119.0, 118.0, 111.0, 108.8, 82.0,

ꢀ788C for 30 min and then concen-

trated to give the crude hydroxypyr-

roloindoline as a yellow oil.

A solution of this crude residue, HOAc (50 mL), MeOH (100 mL), and

CH₂Cl₂(200 mL) was maintained at rt for 2 h and then concentrated. The

resulting residue was purified by chromatography on silica gel (8:1

hexanes/Et₂O then 3:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O

then EtOAc then MeOH) to give 14 (3.00 g, 55%) as a tan solid that was

homogeneous by TLC analysis: m.p. 197 ± 1988C; ¹H NMR (500 MHz,

MeOH): d 7.21 (d, J 7.1 Hz, 1H), 7.07 (dt, J 7.9, 1.1 Hz, 1H), 6.72 (t, J

7.2 Hz, 1H), 6.57 (d, J 7.9 Hz, 1H), 4.94 (s, 1H), 3.56 (dd, J 9.4, 6.4 Hz,

1H), 2.67 (dd, J 12.4, 9.5 Hz, 1H), 2.43 (dd, J 12.4, 6.4 Hz, 1H), 1.45 (s,

9H); ¹³C NMR (125 MHz, MeOH): d 173.7, 151.8, 132.5, 130.7, 125.0,

119.8, 110.8, 90.1, 86.1, 82.7, 61.1, 46.3, 28.3; IR (film): nÄ 3442, 3395, 3215,

2994, 2911, 1736, 1610, 1436, 1310, 1051, 953, 698 cm^ꢀ1; ES-MS: m/z: calcd

56.5, 28.9, 27.2; IR (film): nÄ 3477, 3019, 1728, 1399, 1215, 1149 cm^ꢀ1; ES-

MS: m/z: calcd for C₂₉H₂₈N₂O₄SNa [MNa] : 523.17, found: 523.2; [a]_D²⁵

10.5 (c 1.0 in CHCl₃).

for C₁₅H₂₁N₂O₃[MH] : 277.15, found:

N_b-Anthracenesulfonylpyrroloindo-

277.1; [a]²_D⁵ ꢀ89.6 (c 1.0 in MeOH).

line tert-butyl ester (11): Solid NBS

(2.84 g, 16.0 mmol) was added to a

08C solution of 9 (8.00 g, 16.0 mmol)

and CH₂Cl₂(1 L). After 1 min, Et₃N

Pyrroloindoline tert-butyl ester 15: Fresh-

ly prepared Al(Hg) (10 g) was added to a

vigorously stirred mixture of crude 11

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0051 $ 17.50+.50/0

51

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

(6.00 g, 11.6 mmol) and THF/H₂O (10:1, 200 mL) at 08C. The resulting

mixture turned from bright yellow to gray over 2 h indicating consumption

of 11. The mixture was filtered through a pad of celite (EtOAc) and

concentrated. The crude residue was used without further purification.

allowed to cool to rt, and then diluted with EtOAc (100 mL), and poured

into saturated aqueous NaHCO₃(100 mL). The layers were separated and

the organic layer was washed with saturated aqueous NaHCO₃(2 Â

50 mL), brine (3 Â 50 mL), dried (Na₂SO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (8:1 hexanes/Et₂O

then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to

give 18 (3.10 g, 89%) as a pale yellow glass that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 7.70 (brs, 1H), 7.60 ± 7.35 (m,

12H), 6.02 (brs, 1H), 5.50 ± 4.85 (m, 4H), 4.56 (brs, 1H), 2.84 (d, J

12.9 Hz, 1H), 2.80 ± 2.70 (m, 1H), 1.10 (s, 9H), 0.74 (s, 9H), 0.26 (s, 9H),

ꢀ0.37 (s, 3H), ꢀ0.44 (s, 3H); ¹³C NMR (125 MHz, CDCl₃): d 169.4,

153.4, 143.6, 137.9, 136.3, 136.1, 131.7, 131.5, 128.3, 128.0, 117.0, 86.5, 82.0,

81.1, 67.5, 67.2, 59.8, 40.9, 27.3, 25.4, 17.8, ꢀ3.7, ꢀ4.4, ꢀ9.6; IR (film): nÄ

2953, 1718, 1479, 1338, 1302, 1149, 1103, 1020 cm^ꢀ1; ES-MS: m/z: calcd for

Neat pyridine (5.60 mL, 69.8 mmol) was added to a 08C solution of the

crude residue, ClCO₂Bn (5.25 mL, 34.9 mmol) and CH₂Cl₂(40 mL). The

resulting solution was allowed to warm to rt overnight and then

concentrated. The crude residue was dissolved in EtOAc (100 mL) and

washed with 1m HCl (2 Â 50 mL), brine (1 Â 50 mL), dried (MgSO₄), and

concentrated. This residue was purified by chromatography on silica gel

(8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1

Et₂O/hexanes then Et₂O then EtOAc) to give 15 (3.82 g, 60%) as a near

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.67 (brs, 1H), 7.45 ± 7.08 (m, 12H), 7.03 (t, J 7.4 Hz, 1H), 6.09

(s, 1H), 5.31 ± 4.65 (m, 4H), 4.55 (brs, 1H), 4.10 (brs, 1H), 2.75 ± 2.60 (m,

2H), 1.01 (s, 9H); ¹³C NMR (125 MHz, CDCl₃): d 169.2, 153.5, 142.5,

136.0, 135.7, 132.3, 130.4, 128.5, 128.3, 127.9, 127.8, 124.0, 123.7, 117.1, 82.2,

82.0, 81.3, 67.6, 67.1, 60.0, 40.2, 27.2; IR (film): nÄ 3410, 2978, 1716, 1453,

C₄₀H₅₄N₂O₇SiSnNa [MNa] : 843.26, found: 843.4, 135; [a]²_D⁵ ꢀ46.9 (c

1.0 in CHCl₃).

3a-tert-Butyldimethylsiloxy-pyrroloindoline tert-butyl ester dimer 19: A

solution of stannane 18 (900 mg, 1.10 mmol) in DMF (8 mL) was added to a

solution of iodide 17 (781 mg,

1289, 1207, 1018, 735 cm^ꢀ1; ES-MS: m/z: calcd for C₃₁H₃₂N₂O₇Na [MNa] :

567.21, found: 567.3; [a]²_D⁵ ꢀ40.5 (c 1.0 in CHCl₃).

1.00 mmol),

0.100 mmol),

[Pd₂dba₃]

Ph₃As

(96.0 mg,

(61.0 mg,

Silylpyrroloindoline tert-butyl ester 16: A solution of DBU (8.60 mL,

57.4 mmol) and DMF (50 mL) was added to a solution of 15 (7.80 g,

0.199 mmol), and DMF (9.0 mL) at

rt. The resulting mixture was warmed

14.3 mmol),

TBSCl

(4.40 g,

to 458C for 22 h, allowed to cool to rt and then diluted with EtOAc

(100 mL) and poured into saturated aqueous NaHCO₃(100 mL). The

layers were separated and the organic layer was washed with saturated

aqueous NaHCO₃(2 Â 50 mL), brine (3 Â 50 mL), dried (Na₂SO₄), and

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then

3:1 Et₂O/hexanes) to give 19 (722 mg, 55%) as a light yellow oil that was

homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.80 (brs,

2H), 7.43 (d, J 8.1 Hz, 2H), 7.40 ± 7.25 (m, 22H), 6.10 (brs, 2H), 5.50 ± 4.80

(m, 8H), 4.59 (brs, 2H), 2.84 (d, J 12.9 Hz, 2H), 2.80 ± 2.65 (m, 2H), 0.97

(brs, 18H), 0.77 (s, 18H), ꢀ0.35 ± ꢀ 0.40 (m, 12H); ¹³C NMR (125 MHz,

CDCl₃): d 169.2, 153.3, 142.6, 136.6, 136.2, 136.0, 132.6, 129.4, 128.4, 128.2,

128.1, 128.0, 122.9, 117.6, 86.0, 82.3, 81.3, 67.6, 67.3, 59.8, 41.1, 27.3, 25.4, 17.8,

ꢀ3.6, ꢀ4.2; IR (film): nÄ 2953, 1718, 1477, 1303, 1147, 1103, 837 cm^ꢀ1; ES-

28.7 mmol), and DMF (50 mL). The

resulting solution was warmed to

508C for 5 h, cooled, and was then

diluted with EtOAc (400 mL) and

poured into aqueous 1m HCl

(200 mL). The layers were separated and the organic layer was washed

with aqueous 1m HCl (3 Â 100 mL), saturated aqueous NaHCO₃(1 Â

100 mL), brine (2 Â 100 mL), dried (MgSO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (3:1 hexanes/Et₂O

then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to give 16 (7.20 g, 77%) as a

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.78 (brs, 1H), 7.50 ± 7.10 (m, 12H), 7.05 (t, J 7.3 Hz, 1H), 6.09

(brs, 1H), 5.42 ± 4.92 (m, 4H), 5.08 (brs, 1H), 4.59 (brs, 1H), 2.83 (d, J

12.8 Hz, 1H), 2.78 ± 2.72 (m, 1H), 1.03 (s, 9Hs), 0.78 (s, 9H), ꢀ0.32 (s, 3H),

ꢀ0.37 (s, 3H); ¹³C NMR (125 MHz, CDCl₃): d 169.2, 153.3, 143.2, 136.2,

136.0, 131.8, 130.6, 128.3, 128.2, 128.1, 128.0, 127.9, 124.3, 123.6, 117.3, 86.8,

81.9, 81.2, 67.4, 67.1, 59.8, 41.1, 27.3, 25.3, 17.7, ꢀ3.8, ꢀ4.4; IR (film): nÄ

2953, 1718, 1480, 1255, 1142, 1107, 1019 cm^ꢀ1; ES-MS: m/z: calcd for

MS: m/z: calcd for C₇₄H₉₀N₄O₁₄Si₂Na [MNa] : 1337.59, found: 1337.8, 135;

[a]²_D⁵ ꢀ78.0 (c 1.0 in CHCl₃).

3a-Hydroxypyrroloindoline tert-butyl ester dimer 20: A solution of TBAF

(1m in THF, 2.28 mL) was added to a 08C solution of 19 (2.00 g, 1.52 mmol)

and THF (10.0 mL). The resulting

C₃₇H₄₆N₂O₇SiNa [MNa] : 681.30, found: 681.5, 135; [a]²_D⁵ ꢀ38.1 (c 1.0

in CHCl₃).

solution was maintained at 08C for

2 h and was then diluted with EtOAc

(100 mL) and poured into saturated

aqueous NaHCO₃(100 mL). The lay-

5-Iodo-pyrroloindoline tert-butyl ester 17: Neat ICl (1.80 mL, 35.3 mmol)

was added to a 08C solution of 16 (2.30 g, 3.50 mmol), 2,6-di-tert-butyl-4-

methyl pyridine (6.70 g, 35.3 mmol),

ers were separated and the organic layer was washed with 1m HCl (2 Â

50 mL), saturated aqueous NaHCO₃(2 Â 50 mL), brine (3 Â 50 mL), dried

(MgSO₄), and concentrated. The crude residue was purified by chroma-

tography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1

Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give 20

(1.40 g, 85%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 7.71 (brs, 2H), 7.41 (s, 2H), 7.35 ± 7.18 (m,

22H), 6.08 (s, 2H), 5.22 ± 4.73 (m, 8H), 4.54 (d, J 7.8 Hz, 2H), 3.93 (brs,

2H), 2.80 ± 2.60 (m, 4H), 0.93 (s, 18H); ¹³C NMR (125 MHz, CDCl₃): d

169.5, 153.5, 142.0, 136.0, 135.8, 133.0, 129.1, 128.4, 128.1, 128.0, 122.1, 117.3,

85.0, 82.5, 81.6, 67.8, 67.3, 60.2, 40.8, 27.2; IR (film): nÄ 3424, 3018, 1712,

1478, 1410, 1215, 1110, 755 cm^ꢀ1; ES-MS: m/z: calcd for C₆₂H₆₂N₄O₁₄Na

and CH₂Cl₂(40 mL). The resulting

solution was allowed to warm to rt

overnight, and after 24 h, was diluted

with EtOAc (200 mL), and poured

into saturated aqueous Na₂S₂O₃

(100 mL). The layers were separated and the organic layer was washed

with aqueous 1m HCl (3 Â 50 mL), saturated aqueous NaHCO₃(2 Â

50 mL), brine (2 Â 50 mL), dried (MgSO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (8:1 hexanes/Et₂O

then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes) to give 17 (2.24 g, 81%) as a

pale yellow glass that was homogeneous by TLC analysis: ¹H NMR

(500 MHz, CDCl₃): d 7.62 (d, J 1.4 Hz, 1H), 7.60 (s, 1H), 7.39 ± 7.25 (m,

11H), 6.02 (brs, 1H), 5.45 ± 4.97 (m, 4H), 4.56 (brs, 1H), 2.80 ± 2.68 (m,

2H), 1.27 (s, 9H), 0.76 (s, 9H), ꢀ0.31 (s, 3H), ꢀ0.37 (s, 3H); ¹³C NMR

(125 MHz, CDCl₃): d 168.9, 153.1, 143.0, 139.3, 136.1, 135.7, 134.7, 133.2,

128.4, 128.2, 128.0, 119.5, 86.2, 81.9, 81.7, 67.8, 67.4, 59.6, 40.7, 27.5, 25.3, 17.7,

ꢀ3.5, ꢀ4.3; IR (film): nÄ 2953, 2856, 1720, 1473, 1255, 1146, 837 cm^ꢀ1; ES-

[MNa] : 1109.42, found: 1109.7, 135; [a]²_D⁵ ꢀ122 (c 0.6 in CHCl₃).

Pyrroloindoline tert-butyl ester dimer 21: A solution of DBU (0.45 mL,

3.04 mmol) and DMF (5.0 mL) was added to a solution of 20 (1.10 g,

1.01 mmol), TESCl (1.02 mL, 6.08 mmol), and DMF (8.0 mL) at rt. The

resulting solution was warmed to 508C

for 5 h, cooled, and was then diluted

with EtOAc (200 mL) and poured into

MS: m/z: calcd for C₃₇H₄₅N₂O₇SiINa [MNa] : 807.19, found: 807.4, 135;

[a]²_D⁵ ꢀ59.0 (c 1.0 in CHCl₃).

5-Trimethylstannyl-pyrroloindoline tert-butyl ester 18: Neat Me₆Sn₂

(2.76 g, 8.41 mmol) was added to a

aqueous 1m HCl (100 mL). The layers

were separated and the organic layer

solution of 17 (3.30 g, 4.21 mmol),

[Pd(Ph₃P)₄] (0.49 g, 0.421 mmol), and

THF (15.0 mL) at rt. The resulting

solution was warmed to 608C for 5 h,

was washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃(1 Â

50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (8:1 hexanes/Et₂O

then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to

52

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0052 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

give 21 (1.05 g, 79%) as a near colorless oil that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 7.81 (brs, 2H), 7.44 (d, J

8.5 Hz, 2H), 7.38 (s, 2H), 7.36 ± 7.25 (m, 20H), 6.10 (brs, 2H), 5.52 ± 4.90

(m, 8H), 4.58 (brs, 2H), 2.84 (d, J 12.8 Hz, 2H), 2.80 ± 2.63 (m, 2H), 0.97

(brs, 18H), 0.73 (t, J 8.0 Hz, 18H), 0.38 ± 0.20 (m, 12H); ¹³C NMR

(125 MHz, CDCl₃): d 169.2, 153.3, 142.5, 136.4, 136.2, 132.7, 129.3, 128.3,

128.0, 122.5, 117.6, 86.2, 82.4, 81.3, 67.6, 67.2, 59.9, 40.8, 27.2, 6.5, 5.4; IR

(film): nÄ 2955, 1718, 1478, 1340, 1253, 1147, 1102, 1017, 735 cm^ꢀ1; ES-MS:

0.412 mmol) and CH₃CN (4.0 mL). The resulting solution was allowed to

warm to rt over 5 h and then concentrated. The crude residue was purified

by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to

give 24 (281 mg, 92%) as a near colorless oil that was homogeneous by

1

TLC analysis: H NMR (500 MHz, CDCl₃): d 7.30 (s, 2H), 7.23 (dd, J

8.1, 1.4 Hz, 2H), 6.58 (d, J 8.1 Hz, 2H), 5.65 ± 5.50 (m, 2H), 5.08 (d, J

17.2 Hz, 2H), 5.00 (d, J 10.4, 2H), 4.87 (s, 2H), 4.30 ± 4.22 (m, 4H), 4.01

(dd, 7.0, 5.4 Hz, 2H), 2.68 (dd, J 12.7, 7.4 Hz, 2H), 2.52 (dd, J 12.7,

5.3 Hz, 2H), 0.84 (t, J 8.0 Hz, 18H), 0.49 ± 0.33 (m, 12H); ¹³C NMR

(125 MHz, CDCl₃): d 173.0, 148.2, 132.7, 131.6, 131.2, 128.1, 122.8, 118.3,

110.5, 90.2, 84.9, 65.5, 59.5, 44.6, 6.7, 5.6; IR (film): nÄ 3365, 2953, 1735,

1619, 1482, 1240, 1117, 1008, 740 cm^ꢀ1; ES-MS: m/z: calcd for C₄₀H₅₉N₄O₆-

m/z: calcd for C₇₄H₉₀N₄O₁₄Si₂Na [MNa] : 1337.59, found: 1337.9, 135;

[a]²_D⁵ ꢀ74.1 (c 1.0 in CHCl₃).

Pyrroloindoline dimer 22: A solution of 21 (920 mg, 0.700 mmol) and

EtOAc (20.0 mL) was charged with 10% Pd/C (100 mg) at rt. The resulting

mixture was evacuated/purged with

Si₂[MH] : 747.40, found: 747.7, 135; [a]²_D⁵ ꢀ77.8 (c 0.65, CHCl₃).

H₂(3 Â ) and was then stirred vigo-

Methyl-5-bromo-4R-hydroxypentanoate (26): A solution of NaOMe in

MeOH (25% by weight in MeOH, 25.4 mL, 111 mmol) was added to a

solution of 25 (15.0 g, 55.6 mmol) and

rously under 1 atm H₂for 26 h. The

mixture was filtered through a pad of

celite (EtOAc) and concentrated.

MeOH (200 mL) at rt. After 20 min,

Neat pyridine (0.340 mL, 4.20 mmol) was added to a 08C solution of the

crude residue, Fmoc-HOSu (944 mg, 2.80 mmol) and CH₂Cl₂(6.0 mL). The

resulting solution was allowed to warm to rt overnight and then

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then

3:1 Et₂O/hexanes then Et₂O) to give 22 (746 mg, 87%) as a colorless solid

that was homogeneous by TLC analysis: m.p. 210 ± 2128C; ¹H NMR

(500 MHz, CDCl₃): d 7.85 ± 7.09 (m, 20H), 6.64 (dd, J 11.2, 8.1 Hz, 1H),

6.36 (dd, J 13.6, 8.1 Hz, 1H), 5.36 ± 5.26 (m, 2H), 4.75 ± 3.95 (m, 10H),

2.90 ± 2.82 (m, 1.6H), 2.78 ± 2.62 (m, 2.4H), 1.04 (s, 4.5H), 1.02 (s, 4.5H),

1.00 (s, 4.5H), 0.99 (s, 4.5H), 0.94 ± 0.80 (m, 18H), 0.54 ± 0.38 (m, 12H);

¹³C NMR (125 MHz, CDCl₃): d 169.5, 169.4, 169.1, 169.0, 154.9, 153.8,

149.5, 149.4, 148.8, 148.7, 144.1, 143.9, 143.5, 141.7, 141.3, 141.2, 141.1, 132.8,

132.5, 129.2, 128.9, 128.8, 127.8, 127.7, 127.5, 127.3, 127.2, 127.1, 127.0, 125.1,

125.0, 124.6, 124.4, 122.9, 122.8, 120.2, 120.0, 119.9, 110.2, 109.6, 109.5, 88.1,

87.3, 82.3, 81.5, 81.4, 81.1, 67.7, 66.6, 59.6, 59.3, 47.3, 47.2, 42.7, 42.1, 27.4, 27.3,

6.8, 6.7, 5.5, 5.3; IR (film): nÄ 3427, 3019, 1742, 1708, 1420, 1215, 1119,

the reaction was quenched with gla-

cial HOAc until pH ꢁ7 and concen-

trated. The crude residue was dissolved in Et₂O (100 mL) and H₂O

(100 mL). The layers were separated and the aqueous layer was extracted

with CH₂Cl₂(5 Â 75 mL). The combined organic layers were washed with

brine (1 Â 100 mL), dried (MgSO₄), and concentrated.

Glacial HOAc (9.50 mL, 167 mmol) was added to a 08C solution of the

crude epoxy ester, LiBr (14.5 g, 167 mmol) and THF (100 mL). The

reaction was allowed to warm to rt overnight and then diluted with Et₂O

(100 mL) and H₂O (100 mL). The layers were separated and the organic

layer was washed with cold saturated aqueous NaHCO₃(2 Â 50 mL), brine

(1 Â 50 mL), dried (Na₂SO₄), and concentrated. The crude residue was

purified by chromatography on silica gel (3:1 hexanes/Et₂O then 3:1 Et₂O/

hexanes then Et₂O) to give 26 (10.2 g, 93%) as a colorless oil that was

homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 3.90 ±

3.85 (m, 1H), 3.69 (s, 3H), 3.51 (dd, J 10.3, 3.9 Hz, 1H), 3.39 (dd, J 10.4,

6.7 Hz, 1H), 2.51 (t, J 7.1 Hz, 2H), 2.08 ± 1.80 (m, 2H); ¹³C NMR

(125 MHz, CDCl₃): d 173.8, 69.7, 51.5, 38.7, 29.8, 29.6; IR (film): nÄ 3451,

2953, 1734, 1438, 1174, 1092 cm^ꢀ1; ES-MS: m/z: calcd for C₆H₁₁BrO₃Na

747 cm^ꢀ1; ES-MS: m/z: calcd for C₇₂H₈₆N₄O₁₀Si₂Na [MNa] : 1245.58,

found: 1245.9, 135; [a]²_D⁵ ꢀ270 (c 1.0 in CHCl₃).

3a-Triethylsiloxypyrroloindoline allyl ester dimer 23: Neat TESOTf

[MNa] : 232.98, found: 232.9; [a]²_D⁵ 6.32 (c 1.0 in CHCl₃).

(2.60 mL, 11.4 mmol) was added to

a 08C solution of 22 (696 mg,

Methyl-5-bromo-4R-O-tert-butyldimethylsilylpentanoate (27): TBSOTf

(8.50 mL, 37.0 mmol) was added to a ꢀ788C solution of 26 (6.50 g,

30.8 mmol), 2,6-lutidine (7.17 mL,

0.57 mmol), 2,6-lutidine (3.30 mL,

28.5 mmol), and CH₂Cl₂(5.0 mL).

The resulting solution was allowed

to warm to rt overnight, recooled to

08C, quenched with saturated aque-

61.6 mmol), and CH₂Cl₂(50 mL).

The reaction was maintained at

ꢀ788C for 1 h, warmed to rt for 1 h,

ous NaHCO₃(50 mL), and diluted with EtOAc (100 mL). The layers were

separated and the organic layer was washed with aqueous 1m HCl (2 Â

50 mL), brine (1 Â 50 mL), dried (Na₂SO₄), and concentrated to give crude

diacid. This residue residue was azeotroped with toluene (2 Â 100 mL) to

remove TESOH and was used without further purification.

quenched with saturated aqueous NaHCO₃(100 mL) and diluted with

EtOAc (100 mL). The layers were separated and the organic layer was

washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃(1 Â 50 mL),

brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The crude residue

was purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes) to give 27

(6.51 g, 65%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 3.92 ± 3.84 (m, 1H), 3.62 (s, 3H), 3.23

(ddd, J 14.8, 10.3, 4.4 Hz, 2H), 2.33 (t, J 7.4 Hz, 2H), 2.15 ± 1.95 (m,

1H), 1.88 ± 1.75 (m, 1H), 0.84 (s, 9H), 0.046 (s, 3H), 0.018 (s, 3H); ¹³C NMR

(125 MHz, CDCl₃): d 173.5, 70.4, 51.4, 36.8, 30.2, 29.1, 25.6, 17.9, ꢀ4.6,

ꢀ4.9; IR (film): nÄ 2954, 1741, 1437, 1256, 1079, 837, 777 cm^ꢀ1; ES-MS:

EDCI (656 mg, 3.40 mmol) was added to a 08C solution of the crude

residue, allyl alcohol (3.90 mL, 56.9 mmol), DMAP (10 mg), and CH₂Cl₂

(10.0 mL). The resulting solution was allowed to warm to rt overnight and

then concentrated. The crude residue was purified by chromatography on

silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes

then 3:1 Et₂O/hexanes then Et₂O) to give 23 (530 mg, 78%) as a near

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.90 ± 7.15 (m, 20H), 6.63 (t, J 8.0 Hz, 0.8H), 6.39 (t, J

8.0 Hz, 1.2H), 5.55 ± 5.39 (m, 2H), 5.32 ± 5.21 (m, 2H), 5.10 ± 4.92 (m,

4H), 4.79 ± 3.85 (m, 14H), 2.94 ± 2.65 (m, 4H), 0.95 ± 0.81 (m, 18H), 0.55 ±

0.31 (m, 12H); ¹³C NMR (125 MHz, CDCl₃): d 170.2, 169.9, 154.7, 153.9,

149.5, 149.4, 148.7, 148.6, 144.0, 143.8, 143.5, 141.6, 141.2, 141.1, 132.5, 132.3,

131.5, 131.4, 129.0, 128.9, 128.6, 127.8, 127.7, 127.6, 127.5, 127.3, 127.2, 127.0,

126.9, 124.8, 124.7, 124.4, 122.7, 122.6, 120.2, 120.0, 119.9, 118.4, 118.0, 110.2,

110.0, 87.8, 86.9, 82.1, 81.6, 67.5, 66.7, 65.8, 65.6, 58.9, 58.7, 47.1, 42.2, 41.8, 6.7,

5.4, 5.4; IR (film): nÄ 3419, 2954, 1750, 1706, 1481, 1305, 1142, 908,

m/z: calcd for C₁₂H₂₅BrO₃SiNa [MNa] : 347.1, found: 347.2; [a]²_D⁵ 20.4

(c 1.0 in CHCl₃).

Boc-hydrazine methyl esters 28: A solution of 27 (6.20 g, 19.1 mmol) and

THF (15.0 mL) was added to a ꢀ788C solution of NaHMDS (1m in THF,

27.0 mL) and THF (30.0 mL). After

20 min, a solution of DBAD (5.27 g,

22.9 mmol) and THF (20.0 mL) was

added. The resulting dark yellow

solution was maintained at ꢀ788C

741 cm^ꢀ1; ES-MS: m/z: calcd for C₇₀H₇₈N₄O₁₀Si₂Na [MNa] : 1213.51,

for 10 min, quenched with HOAc

found: 1213.8, 135; [a]²_D⁵ ꢀ263 (c

(5.0 mL), warmed to rt and diluted with EtOAc (100 mL) and saturated

aqueous NaHCO₃(100 mL). The layers were separated and the organic

layer was washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃

(2 Â 50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The

crude residue was purified by chromatography on silica gel (8:1 hexanes/

0.55 in CHCl₃).

Pyrroloindoline allyl ester dimer 24:

Piperidine (163 mL, 1.65 mmol) was

added to a 08C solution of 23 (490 g,

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0053 $ 17.50+.50/0

53

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O) to give 28 (5.53 g,

79%) as a 1:1 mixture of diastereomers as judged by ¹H NMR analysis. This

mixture was used without further purification.

hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give 33

(4.06 g, 80%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 5.03 (brs, 0.80H), 4.83 (brs, 0.20H),

4.25 ± 4.20 (m, 0.6H), 4.10 ± 3.92 (m, 1.4H), 3.64 (s, 3H), 2.90 ± 2.40 (m, 2H),

2.34 ± 2.21 (m, 1H), 1.60 ± 1.50 (m, 1H), 1.46 (s, 9H), 1.43 (s, 9H); ¹³C NMR

(125 MHz, CDCl₃): d 171.4, 155.3, 155.1, 83.3, 82.0, 63.1, 55.3, 53.3, 50.3,

A small amount was purified for analysis by chromatography on silica gel

(8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O) to give

the desired (R,R) diastereomer as a colorless oil that was homogeneous by

TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 6.40 (brs, 1H), 4.94 (brm,

1H), 4.25 (brs, 1H), 3.67 (s, 3H), 3.30 ± 3.23 (m, 2H), 2.06 ± 1.96 (m, 2H),

1.40 (brs, 18H), 0.87 (s, 9H), 0.065 (s, 3H), 0.010 (s, 3H); ¹³C NMR

(125 MHz, CDCl₃): d 172.9, 155.7, 154.9, 81.7, 80.5, 67.0, 56.6, 52.3, 38.4,

35.2, 28.1, 28.0, 25.7, 17.8, ꢀ4.5, ꢀ5.3; IR (film): nÄ 3382, 3327, 2977, 1744,

1719, 1473, 1368, 1255, 1153, 1076, 839, 778 cm^ꢀ1; ES-MS: m/z: calcd for

34.2, 29.3; IR (film): nÄ 3460, 2978, 1709, 1395, 1326, 1158, 1087, 993 cm^ꢀ1

;

ES-MS: m/z: calcd for C₁₆H₂₈N₂O₇Na [MNa] : 383.18, found: 383.1, 135;

[a]²_D⁵ ꢀ48.0 (c 1.0 in CHCl₃).

Piperazic lactone (34 from 33): DBU (5.21 mL, 34.9 mmol) was added to 33

(6.30 g, 17.5 mmol) and toluene (200 mL). The reaction was warmed to

reflux using a Dean ± Stark trap filled

with activated 4 molecular sieves.

After 52 h, the reaction was cooled to

C₂₂H₄₃BrN₂O₇SiNa [MNa] : 577.2, found: 577.3; [a]²_D⁵ 8.40 (c 1.0 in

CHCl₃).

rt, diluted with EtOAc (100 mL), and

1m HCl (100 mL). The layers were

trans- and cis-Piperazic methyl esters (29, 30): NaH (60% dispersion in oil,

0.61 g, 15.2 mmol) was added to a 08C solution of 28 (6.50 g, 11.7 mmol)

and DMF (65 mL). After 20 min, the

separated and the organic layer was washed with 1m HCl (2 Â 50 mL),

saturated aqueous NaHCO₃(1 Â 50 mL), brine (1 Â 50 mL), dried

(MgSO₄), and concentrated. The crude residue was purified by chroma-

tography on silica gel (3:1 hexanes/Et₂O then Et₂O) to give an 8:1 mixture

(by ¹H NMR) of 34:33 as a colorless solid. Trituration of this solid with

hexanes/Et₂O 20:1 gave a 15 ± 20:1 mixture (by ¹H NMR) of 34:33 (4.00 g,

69%) as a colorless solid which was used without further purification.

reaction was poured into a mixture of

ice-cold

EtOAc/1m

HCl

(1:1,

300 mL). The layers were separated

and the organic layer was washed with

1m HCl (2 Â 50 mL), saturated aque-

ous NaHCO₃(1 Â 50 mL), brine (1 Â

50 mL), dried (MgSO₄), and concen-

Piperazic ester (31 from 34): Solid 34 (4.70 g, 14.3 mmol) was added to TFA

(50 mL) at 08C. The resulting solution was maintained at 08C for 30 min, at

rt for 4 h, and then recooled to 08C.

trated. The crude residue was purified

by chromatography on silica gel (8:1

hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O) to give 29

(3.0 g, 44%) followed by 30 (2.20 g, 37%), both as colorless oils

homogeneous by TLC analysis: trans-(29): ¹H NMR (500 MHz, CDCl₃):

d 5.07 ± 4.83 (m, 1H), 4.23 ± 3.98 (m, 2H), 3.72 (s, 3H), 3.75 ± 2.50 (m,

1H), 2.35 ± 2.19 (m, 1H), 1.70 ± 1.65 (m, 1H), 1.47 (s, 9H), 1.44 (s, 9H), 0.87

(s, 9H), 0.036 (s, 6H); ¹³C NMR (125 MHz, CDCl₃): d 170.4, 154.1, 153.7,

81.9, 80.5, 63.0, 54.3, 52.1, 49.5, 33.8, 28.2, 25.7, 18.0, ꢀ4.8; IR (film): nÄ

2976, 1736, 1708, 1393, 1254, 1158, 1096, 878 cm^ꢀ1; ES-MS: m/z: calcd for

MeOH (60.0 mL) was added, the

reaction was allowed to come to rt

overnight, and then concentrated.

TeocCl (3.13 mL, 17.2 mmol) was

added to a ꢀ308C solution of this

crude residue, pyridine (11.6 mL, 143 mmol), and CH₂Cl₂(150 mL). The

resulting solution was allowed to come to rt overnight and then

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then

3:1 Et₂O/hexanes then Et₂O then EtOAc) to give the secondary alcohol

(3.60 g, 84%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 4.20 ± 4.17 (m, 2H), 4.1 ± 4.07 (brs, 1H),

3.81 (s, 3H), 3.80 ± 3.75 (m, 2H), 3.56 (dd, J 10.1, 3.0 Hz, 1H), 2.98 (brs,

1H), 2.30 (dt, J 12.8, 3.2 Hz, 1H), 1.64 ± 1.62 (m, 1H), 1.02 ± 0.98 (m, 2H),

0.008 (s, 9H); ¹³C NMR (125 MHz, CDCl₃): d 171.3, 155.8, 64.6, 64.3, 57.0,

52.3, 50.9, 35.9, 17.8, ꢀ1.6; IR (film): nÄ 3418, 2954, 1743, 1698, 1250, 1171,

C₂₂H₄₂N₂O₇SiNa [MNa] : 497.3, found: 497.3; [a]²_D⁵ ꢀ36.9 (c 1.0 in

CHCl₃); cis-(30): ¹H NMR (500 MHz, CDCl₃): d 5.00 ± 4.75 (m, 1H), 4.10

(d, J 13.3 Hz, 0.70H), 3.89 ± 3.84 (m, 1.3H), 3.66 (s, 3H), 3.00 (d, J

13.2 Hz, 0.35H), 2.79 (d, J 13.2 Hz, 0.65H), 1.85 ± 1.79 (m, 1H), 1.46 ± 1.39

(m, 18H), 0.84 (s, 3H), 0.80 (s, 6H), 0.009 (s, 6H); ¹³C NMR (125 MHz,

CDCl₃): d 170.8, 170.5, 154.7, 154.5, 154.2, 152.3, 81.5, 80.0, 79.8, 63.3, 63.2,

52.0, 51.9, 51.7, 51.6, 50.8, 48.6, 32.7, 28.2, 28.1, 25.6, 25.5, 18.0, ꢀ4.9, ꢀ5.2,

ꢀ5.3, ꢀ5.4; IR (film): nÄ 2976, 1768, 1735, 1700, 1418, 1365, 1173, 1133,

1095, 882 cm^ꢀ1; ES-MS: m/z: calcd for C₂₂H₄₂N₂O₇SiNa [MNa] : 497.3,

1073, 858 cm^ꢀ1; ES-MS: m/z: calcd for C₁₂H₂₄N₂O₃SiNa [MNa] : 327.14,

found: 497.2; [a]_D²⁵ 12.8 (c 1.0 in CHCl₃).

found: 327.0; [a]_D²⁵ 11.3 (c 1.0 in CHCl₃).

Piperazic acid methyl ester (31 from 30): TFA (20 mL) was added to 30

(2.30 g, 4.85 mmol) and CH₂Cl₂

TBSOTf (5.87 mL, 25.6 mmol) was added to a ꢀ788C solution of the above

alcohol (4.30 g, 14.1 mmol), 2,6-lutidine (6.63 mL, 56.6 mmol), and CH₂Cl₂

(50 mL). After 4 h, the reaction was quenched with saturated aqueous

NaHCO₃(100 mL) and diluted with EtOAc (300 mL). The layers were

separated and the organic layer was washed with 1m HCl (2 Â 50 mL),

saturated aqueous NaHCO₃(1 Â 50 mL), brine (1 Â 50 mL), dried

(MgSO₄), and concentrated. The crude residue was purified by chroma-

tography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1

hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O) to give 31 (5.80 g, 98%) as

a colorless solid that was homogeneous by TLC analysis: m.p. 71 ± 728C;

¹H NMR (500 MHz, CDCl₃): d 4.42 ± 4.15 (m, 4H), 3.70 (s, 3H), 3.69 ±

3.54 (m, 1H), 3.51 (dd, J 11.7, 3.0 Hz, 1H), 2.75 (brs, 1H), 2.22 ± 2.19 (m,

1H), 1.76 ± 1.55 (m, 1H), 1.10 ± 1.00 (m, 2H), 0.84 (s, 9H), 0.044 (s, 3H),

0.038 (s, 3H), 0.004 (s, 9H); ¹³C NMR (125 MHz, CDCl₃): d 170.7, 155.9,

65.6, 64.7, 57.4, 52.3, 51.4, 37.6, 25.8, 18.1, 17.9, ꢀ1.35, ꢀ1.42, ꢀ4.64; IR

(film): nÄ 2954, 1747, 1699, 1252, 1171, 1094, 838 cm^ꢀ1; ES-MS: m/z: calcd

(20 mL) at 08C. The reaction was

maintained at 08C for 30 min, at rt for

3 h and concentrated.

TeocCl (0.88 mL, 4.85 mmol) was

added to a 08C solution of this crude

residue, pyridine (1.18 mL, 14.6 mmol), and CH₂Cl₂(20 mL). The reaction

was allowed to warm to rt overnight and was then diluted with EtOAc

(100 mL) and 1m HCl (50 mL). The layers were separated and the organic

layer was washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃

(1 Â 50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The

crude residue was purified by chromatography on silica gel (3:1 hexanes/

Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O then EtOAc)

to give 31 (1.73 g, 92%) as a colorless solid homogeneous by TLC analysis.

Hydroxypiperazic acid methyl ester (33): A solution of TBAF (1m in THF,

21.2 mL) was added to a 08C solution of 29 (6.70 g, 14.1 mmol) and THF

(50 mL). The resulting solution was

for C₁₈H₃₈N₂O₅Si₂Na [MNa] : 441.22, found: 441.3; [a]²_D⁵ ꢀ20.5 (c 1.0

in CHCl₃).

maintained at 08C for 2 h and then

poured into saturated aqueous NaH-

CO₃(200 mL) and diluted with

EtOAc (100 mL). The layers were

Penteoic acid (36): 2m LiOH (100 mL) was added to a vigorously stirred

mixture of 35 (19.3 g, 38.9 mmol) and THF (100 mL) at 08C. After 3 h, the

mixture was diluted with Et₂O (100 mL) and the layers were separated. The

organic layer was extracted with 1m

separated and the organic layer was

NaOH (3 Â 50 mL) and the organic

washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃(2 Â 50 mL),

brine (3 Â 50 mL), dried (MgSO₄), and concentrated. The crude residue

was purified by chromatography on silica gel (3:1 hexanes/Et₂O then 1:1

layer was discarded. The combined

aqueous layers were acidified with

conc. HCl until a pH 2 ± 3 and extract-

54

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0054 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

ed with EtOAc (3 Â 100 mL). The combined organic layers were dried

(Na₂SO₄) and concentrated to give 36 (9.10 g, 71%) as a colorless oil which

was used without further purification.

TBSCl (22.0 g, 0.148 mol) was added to a 08C mixture of the crude residue,

imidazole (23.0 g, 0.337 mol) and CH₂Cl₂(150 mL). The mixture was

allowed to come to rt overnight and then concentrated. The crude residue

was dissolved in EtOAc (300 mL), washed with 1m NaOH (4 Â 50 mL),

brine (1 Â 50 mL), dried (Na₂SO₄), and concentrated to give 38 (ꢁ21.0 g,

>100%) as a near colorless oil that was used without further purification.

A

small amount of the crude residue was purified for analysis by

chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O): ¹H NMR

(500 MHz, CDCl₃): d 9.76 (brs, 1H), 6.74 (brs, 1H), 5.80 ± 5.61 (m, 1H),

5.21 ± 5.00 (m, 2H), 2.85 ± 2.49 (m, 2H), 1.60 ± 1.37 (m, 18); ¹³C NMR

(125 MHz, CDCl₃, 558C): d 171.6, 154.1, 134.1, 117.8, 84.0, 64.6, 61.8, 33.5,

28.1; IR (film): nÄ 3582, 3284, 2980, 1717, 1393, 1369, 1151 cm^ꢀ1; ES-MS:

A small amount was purified for analysis by chromatography on silica gel

(3:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give

38 as a colorless oil that was homogeneous by TLC analysis: ¹H NMR

(500 MHz, CDCl₃): d 5.84 ± 5.78 (m, 1H), 5.23 (d, J 17.1 Hz, 1H), 5.13

(d, 10.4 Hz, 1H), 4.52 (dd, J 13.1, 5.9 Hz, 1H), 4.43 (dd, J 13.1, 5.9 Hz,

1H), 4.21 ± 4.18 (m, 1H), 3.18 ± 3.16 (m, 1H), 1.48 (s, 2H), 1.13 (d, J

6.3 Hz, 3H), 0.73 (s, 9H), 0.072 (s, 3H), ꢀ0.13 (s, 3H); ¹³C NMR (125 MHz,

CDCl₃): d 173.9, 131.7, 118.4, 69.3, 65.4, 60.5, 25.4, 20.6, 17.6, ꢀ4.6, ꢀ5.5;

m/z: calcd for C₁₅H₂₇N₂O₆[MH] : 331.19, found: 331.19; [a]²_D⁵ 13.7

(c 0.46, CHCl₃).

cis-Bromo lactone 37: Solid NBS (1.94 g, 11.0 mmol) was added to a 08C

solution of crude acid 36 (3.60 g, 11.0 mmol) and toluene (75.0 mL). After

45 min, the resulting orange solution

IR (film): nÄ 3390, 2955, 2890, 1742, 1253, 1156, 1076, 837 cm^ꢀ1; ES-MS:

m/z: calcd for C₁₃H₂₈NO₃Si [MH] : 274.18, found: 274.2, 135; [a]_D²⁵

was diluted EtOAc (100 mL) and

quenched with saturated aqueous

Na₂S₂O₃(100 mL). The layers were

separated and the organic layer was

16.8 (c 1.0 in CHCl₃).

Dipeptide 40: EDCI (9.80 g, 51.3 mmol) was added to a 08C solution of

Fmoc-l-leucine (39) (18.0 g, 51.3 mmol), DMAP (20 mg), and CH₂Cl₂

(150 mL). After 10 min, a solution of

washed with brine (1 Â 100 mL),

crude 38 (ꢁ14.0 g, 51.3 mmol) and

dried (MgSO₄), and concentrated. The crude residue was purified by

chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes) to give the trans diaster-

eomer (650 mg, 15%) and 37 (2.92 g, 65%) both as colorless solids that

were homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 6.50

(brs, 1H), 5.21 (brs, 0.8H), 4.95 (brs, 0.2H), 4.65 ± 4.57 (m, 1H), 3.58 ± 3.45

(m, 2H), 2.76 ± 2.65 (m, 1H), 2.40 ± 2.21 (m, 1H), 1.50 (s, 18H); ¹³C NMR

(125 MHz, CDCl₃, 558C): d 171.9, 155.5, 154.1, 82.8, 81.8, 75.3, 58.7, 32.7,

30.7, 28.2, 28.1; IR (film): nÄ 3311, 2978, 1798, 1735, 1709, 1368, 1243, 1163,

CH₂Cl₂(100 mL) was added. The

reaction mixture was maintained at

08C for 2 h and then concentrated.

The crude residue was purified by

chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O) to give the Fmoc-dipeptide (26.0 g, 96%) as a

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.74 (d, J 7.5 Hz, 2H), 7.74 ± 7.58 (m, 2H), 7.40 ± 7.37 (m, 2H),

7.31 ± 7.26 (m, 2H), 6.77 (d, J 9.0 Hz, 1H), 5.89 ± 5.84 (m, 1H), 5.44 (d, J

7.9 Hz, 1H), 5.31 (dd, J 17.2, 1.1 Hz, 1H), 5.20 (d, J 10.4 Hz, 1H), 4.62 ±

4.22 (m, 8H), 1.78 ± 1.60 (m, 3H), 1.18 (d, J 5.9 Hz, 3H), 0.98 (s, 6H), 0.84

(s, 9H), 0.045 (s, 3H), 0.16 (s, 3H); ¹³C NMR (125 MHz, CDCl₃): d 172.3,

169.9, 156.0, 143.6, 141.2, 131.3, 127.6, 126.9, 125.0, 119.8, 118.9, 68.7, 66.9,

66.0, 57.6, 53.5, 47.0, 41.4, 25.5, 24.7, 22.8, 21.9, 20.9, 17.6, ꢀ4.5, ꢀ5.4; IR

1015, 758 cm^ꢀ1; ES-MS: m/z: calcd for C₁₅H₂₆BrN₂O₆[MH] : 409.10,

found: 409.10; [a]²_D⁵ ꢀ23.1 (c 0.52, CHCl₃).

Piperazic lactone (34 from 37): NaHMDS (1m in THF, 4.70 mL) was added

to a 08C solution of 37 (1.75 g, 4.27 mmol) and DMF (20.0 mL). After 1.5 h,

the reaction was quenched with satu-

rated aqueous NH₄Cl (30 mL) and

diluted with EtOAc (100 mL). The

layers were separated and the aque-

ous layer was extracted with EtOAc

(film): nÄ 3429, 3301, 2930, 1719, 1672, 1518, 1253, 1095, 1036, 837 cm^ꢀ1

;

ES-MS: m/z: calcd for C₃₄H₄₈N₂O₆SiNa [MNa] : 631.3, found: 631.5, 135;

[a]²_D⁵ ꢀ27.9 (c 1.0 in CHCl₃).

Piperidine (7.50 mL, 76.0 mmol) was added to a 08C solution of the Fmoc-

dipeptide (25.0 g, 41.1 mmol) and CH₃CN (100 mL). The resulting mixture

was allowed to warm to rt over 3 h and then concentrated. The crude

residue was purified by chromatography on silica gel (8:1 hexanes/Et₂O

then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then

Et₂O then EtOAc) to give 40 (14.5 g, 92%) as a colorless oil that was

homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.76 (d,

J 9.4 Hz, 1H), 5.82 ± 5.75 (m, 1H), 5.21 (d, J 28.1 Hz, 1H), 5.12 (d, J

10.4 Hz, 1H), 4.51 ± 4.37 (m, 4H), 3.35 (dd, J 9.5, 4.16 Hz, 1H), 1.66 ± 1.58

(m, 2H), 1.34 ± 1.31 (m, 3H), 1.05 (d, J 6.2 Hz, 3H), 0.84 (d, J 22 Hz,

3H), 0.80 (d, J 6.1 Hz, 3H), 0.75 (s, 9H), 0.073 (s, 3H), ꢀ0.14 (s, 3H);

¹³C NMR (125 MHz, CDCl₃): d 175.9, 170.1, 131.4, 118.5, 68.5, 65.6, 57.3,

53.3, 43.8, 25.4, 24.5, 23.1, 21.2, 20.8, 17.5, ꢀ4.6, ꢀ5.5; IR (film): nÄ 3376,

2955, 2858, 1748, 1673, 1506, 1255, 1156, 1096, 838, 778 cm^ꢀ1; ES-MS: m/z:

(2 Â 25 mL). The combined organic layers were washed with brine (1 Â

50 mL), dried (MgSO₄) and concentrated.

Ac₂O (0.5 mL) was added to the crude residue. After 2 h, the reaction was

concentrated and the crude residue was purified by chromatography on

silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O

then 3:1 Et₂O/hexanes then Et₂O) to give 34 (900 mg, 64%) as a colorless

solid that was homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃):

d 4.87 (brs, 1H), 4.65 (brs, 1H), 4.11 (brd, 1H), 3.16 (brd, 1H), 2.34 ±

2.29 (m, 1H), 2.02 (d, J 12.3 Hz, 1H), 1.49 (s, 9H), 1.45 (s, 9H); ¹³C NMR

(125 MHz, CDCl₃, 558C): d 170.4, 154.3, 153.2, 82.7, 82.0, 74.5, 55.8, 47.2,

36.2, 28.2, 28.0; IR (film): nÄ 2978, 2852, 1797, 1707, 1477, 1368, 1254, 1151,

1023, 936 cm^ꢀ1; ES-MS: m/z: calcd for C₁₅H₂₄N₂O₆Na [MNa] : 351.16,

found: 351.2; [a]_D²⁵ 13.8 (c 0.5 in CHCl₃).

calcd for C₁₉H₃₉N₂O₄Si [MH] : 387.3, found: 387.4, 135; [a]²_D⁵ ꢀ12.5

cis-Piperazic acid (32): 2m LiOH (30 mL) was added to a 08C solution of 31

(5.10 g, 12.2 mmol) and THF (50 mL). After 10 min, the reaction was

diluted with EtOAc (100 mL) and

(c 1.0 in CHCl₃).

Tripeptide 41: HATU (6.10 g, 16.1 mmol) was added to a 08C solution of

amine 40 (9.60 g, 24.5 mmol), crude acid 32 (5.00 g, 12.4 mmol), HOAt

acidified with 2m HCl until a pH 2 ±

3 was reached. The layers were sep-

arated and the aqueous layer was

(3.40 g,

24.8 mmol),

collidine

(4.90 mL, 37.1 mmol), and CH₂Cl₂

(100 mL). The reaction mixture was

allowed to warm to rt overnight and

then concentrated. The crude residue

extracted with EtOAc (3 Â 50 mL).

The combined organic layers were

washed with brine (2 Â 50 mL), dried (Na₂SO₄), and concentrated to give

was taken up in EtOAc (100 mL) and

1

32 (5.0 g, ꢁ100%) as a near colorless oil. Analysis of the crude H NMR

washed with 1m HCl (3 Â 50 mL),

showed the absence of the methyl ester singlet at d 3.64. This crude acid

saturated aqueous NaHCO₃(3 Â

was used without further purification.

50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (3:1 hexanes/Et₂O

then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O) to give 41 (6.82 g,

72%) as a colorless oil that was homogeneous by TLC analysis: ¹H NMR

(500 MHz, CDCl₃): d 7.00 ± 6.98 (m, 1H), 6.73 ± 6.71 (d, J 9.3 Hz, 1H),

5.92 ± 5.86 (m, 1H), 5.33 ± 5.23 (m, 2H), 4.59 ± 4.46 (m, 5H), 4.22 ± 4.19 (m,

2H), 4.19 ± 4.18 (m, 1H), 3.75 ± 3.72 (m, 1H), 3.42 ± 3.40 (m, 1H), 2.75 ± 2.68

d-Threonine allyl ester 38: A mixture of allyl alcohol (40.0 mL, 0.592 mol),

d-threonine (8.00 g, 67.3 mmol), p-TsOH (15.3 g, 80.7 mmol), and C₆H₆

(200 mL) was refluxed for 42 h using a

Dean ± Stark trap and then concen-

trated. The crude salt was used with-

out further purification.

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0055 $ 17.50+.50/0

55

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

(m, 1H), 2.32 ± 2.30 (m, 1H), 1.71 ± 1.56 (m, 4H), 1.16 (d, J 6.3 Hz, 3H),

1.02 ± 0.98 (m, 2H), 0.95 (d, J 6.3 Hz, 3H), 0.92 (d, J 6.2 Hz, 3H), 0.86 (s,

9H), 0.85 (s, 9H), 0.074 (s, 3H), 0.062 (s, 3H), 0.040 (s, 12H), 0.005 (s, 3H);

¹³C NMR (125 MHz, CDCl₃): d 172.2, 170.4, 170.1, 155.7, 131.5, 119.1,

68.5, 66.0, 65.8, 64.5, 58.8, 57.9, 51.4, 50.9, 40.2, 37.4, 25.6, 24.8, 22.8, 22.1,

21.0, 18.0, 17.8, 17.7, ꢀ1.5, ꢀ4.3, ꢀ4.7, ꢀ4.8, ꢀ5.3; IR (film): nÄ 3358, 3285,

d 175.9, 171.9, 170.0, 169.5, 158.0, 131.6, 118.8, 72.9, 69.0, 67.0, 66.1, 65.8,

64.2, 57.5, 56.0, 54.0, 52.1, 40.6, 31.8, 30.8, 25.6, 24.6, 22.9, 21.8, 20.7, 19.7,

18.0, 17.8, 17.6, 15.4, ꢀ1.5, ꢀ4.3, ꢀ5.1, ꢀ5.2, ꢀ5.3; IR (film): nÄ 3443, 3301,

2930, 1733, 1696, 1682, 1507, 1362, 1251, 1126, 837, 777 cm^ꢀ1; ES-MS: m/z:

calcd for C₄₁H₈₀N₄O₁₀Si₃Na [MNa] : 895.51, found 895.7, 135; [a]_D²⁵

1.58 (c 1.0 in CHCl₃).

2955, 2857, 1737, 1698, 1668, 1522, 1252, 1176, 1127, 837 cm^ꢀ1; ES-MS: m/z:

Troc-d-valine (44): TrocCl (7.20 mL, 52.2 mmol) was added to a 08C

solution of d-valine (5.10 g, 43.5 mmol) and aqueous 2m NaOH (50 mL).

The resulting solution was maintained

at 08C for 1 h, at rt for 1.5 h, and then

diluted with H₂O (100 mL) and Et₂O

calcd for C₃₆H₇₂N₄O₈Si₃Na [MNa] : 795.46, found: 795.5, 135; [a]_D²⁵

ꢀ25.4 (c 1.0 in CHCl₃).

Fmoc-l-hydroxyisovaleric acid chloride (42): Neat allyl bromide (22.0 mL,

254 mmol) was added to a 08C solution of (S)-hydroxy isovaleric acid

(10.0 g, 84.8 mmol), Cs₂CO₃(30.3 g,

(100 mL). The layers were separated,

the aqueous layer was extracted with

93.2 mmol), and DMF (200 mL). Af-

Et₂O (2 Â 100 mL), and the combined organic extracts were discarded. The

aqueous layer was acidified with 1m HCl (100 mL) and extracted with

EtOAc (3 Â 100 mL). The combined organic layers were dried (Na₂SO₄)

and concentrated. The crude residue was purified by chromatography on

silica gel (3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes

ter 2 h at 08C, the reaction was

diluted with EtOAc (100 mL) and

1m HCl (100 mL). The layers were separated and the organic layer was

washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃(2 Â 50 mL),

brine (1 Â 50 mL), dried (MgSO₄), and concentrated to give a near colorless

oil (ꢁ13 g) that was used without further purification.

Neat pyridine (7.30 mL, 89.9 mmol) was added to a 08C solution of this

crude residue, Fmoc-Cl (12.2 g, 47.2 mmol) and CH₂Cl₂(100 mL). The

resulting solution was maintained at 08C for 1 h and then concentrated.

The crude residue was dissolved in Et₂O (100 mL) and washed with 1m HCl

(2 Â 50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated to give a

near colorless oil (ꢁ18 g) that was used without further purification.

then Et₂O) to give 44 (8.96 g, 70%) as

a colorless solid that was

homogeneous by TLC analysis: m.p. 147 ± 1508C; ¹H NMR (500 MHz,

CDCl₃): d 11.5 (brs, 1H), 6.25 ± 6.33 (m, 0.3H), 5.51 (d, J 9.1 Hz, 0.7H),

4.83 ± 4.67 (m, 2H), 4.37 (dd, J 9.1, 4.6 Hz, 0.7H), 4.35 ± 4.30 (m, 0.3),

2.40 ± 2.25 (m, 1H), 1.19 ± 0.86 (m, 6H); ¹³C NMR (125 MHz, CDCl₃): d

177.0, 154.6, 95.3, 74.7, 59.0, 31.1, 19.0, 17.3; IR (film): nÄ 3431, 3272, 3020,

1719, 1515, 1216, 1115, 756 cm^ꢀ1; ES-MS: m/z: calcd for C₈H₁₂Cl₃NO₄Na

[MNa] : 393.97, found: 314.0; [a]²_D⁵ ꢀ4.11 (c 1.0 in CHCl₃).

PhSiH₃(6.60 mL, 53.9 mmol) was added to a 08C solution of this crude

residue, [Pd(Ph₃P)₄] (1.04 g,0.898 mmol), and THF (150 mL). The resulting

solution was allowed to warm to rt overnight and then concentrated. The

crude residue was purified by chromatography on silica gel (8:1 hexanes/

Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes

then Et₂O) to give the acid (12.1 g, 78%) as an off-white solid that was

homogeneous by TLC analysis: m.p. 152 ± 1548C; ¹H NMR (500 MHz,

CDCl₃): d 7.77 (d, J 7.5 Hz, 2H), 7.63 (dd, J 10.9, 7.6 Hz, 2H), 7.42 (t,

J 7.5 Hz, 2H), 7.31 (t, J 7.4 Hz, 2H), 4.90 (d, J 4.1 Hz, 1H), 4.52 (dd,

J 9.9, 6.8 Hz, 1H), 4.40 ± 4.29 (m, 2H), 2.49 ± 2.35 (m, 1H), 1.12 (d, J

7.0 Hz, 3H), 0.97 (d, J 7.0 Hz, 3H); ¹³C NMR (125 MHz, CDCl₃): d

175.6, 154.9, 143.4, 143.0, 141.2, 127.9, 127.2, 125.3, 125.1, 120.0, 79.3, 70.4,

46.6, 30.1, 18.7, 17.0; IR (film): nÄ 3065, 3020, 1749, 1719, 1450, 1280, 1217,

Pentadepsipeptide acid 45: IPCC (0.373 mL, 3.41 mmol) was added to a

ꢀ508C solution of Troc-d-valine (44, 1.43 g, 4.87 mmol), TEA (0.476 mL,

3.41 mmol),

DMAP

(178 mg,

1.46 mmol), and CH₂Cl₂(10 mL). Af-

ter 3 min, a solution of 43 (850 mg,

0.975 mmol) and CH₂Cl₂(5 mL) was

added. The reaction mixture was

allowed to come to rt over 6 h and

then diluted with EtOAc (100 mL)

and 1m HCl (20 mL). The layers were

separated and organic layer was wash-

ed with 1m HCl (3 Â 20 mL), saturat-

ed aqueous NaHCO₃(2 Â 30 mL),

996, 757 cm^ꢀ1; ES-MS: m/z: calcd for C₂₀H₂₀O₅Na [MNa] : 363.13, found:

brine (1 Â 20 mL), dried (MgSO₄), and concentrated. The crude residue

was purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes) to give the

fully protected pentadepsipeptide (953 mg, 85%) as a colorless oil that was

homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 8.34 (brs,

1H), 6.76 ± 6.69 (m, 1H), 5.88 ± 5.79 (m, 1H), 5.53 ± 5.41 (m, 1H), 5.27 (d,

J 17.1 Hz, 1H), 5.19 ± 5.10 (m, 2H), 4.90 ± 4.22 (m, 11H), 4.00 ± 3.74 (m,

2H), 3.66 ± 3.33 (m, 1H), 2.42 ± 1.95 (m, 5H), 1.81 ± 1.55 (m, 2H), 1.09 (d,

J 6.2 Hz, 3H), 0.97 ± 0.75 (m, 41H), 0.006 ± ꢀ 0.067 (m, 21H); ¹³C NMR

(125 MHz, CDCl₃): d 171.9, 170.3, 169.8, 154.2, 131.5, 118.8, 95.3, 74.5,

68.8, 65.7, 64.5, 59.1, 57.5, 56.2, 52.0, 39.9, 31.2, 25.5, 24.3, 22.8, 21.8, 20.6,

19.0, 17.8, 17.7, 17.4, 17.1, ꢀ1.6, ꢀ4.4, ꢀ5.3; IR (film): nÄ 3441, 3298, 2956,

1736, 1717, 1697, 1508, 1251, 1126, 838 cm^ꢀ1; ES-MS: m/z: calcd for

363.1; [a]²_D⁵ ꢀ3.18 (c 1.0 in CHCl₃).

N,N-Dimethyl-2-chloro-propenyl amine (2.20 mL, 16.3 mmol) was added

to a 08C solution of this acid (3.70 g, 10.9 mmol) and CH₂Cl₂(30.0 mL).

After 45 min, an aliquot was removed and concentrated. ¹H NMR

inspection of this sample indicated that full conversion to the acid chloride

had occurred. The solution of acid chloride was used without further

manipulation.

Tetrapeptide 43: A solution of 41 (2.60 g, 3.37 mmol), collidine (4.40 mL,

33.5 mmol), and CH₂Cl₂(10.0 mL) was added at 08C to a freshly prepared

solution of 42 (3.70 g, 10.9 mmol) in CH₂Cl₂(30.0 mL). The reaction

mixture was allowed to come to rt overnight and was then quenched with

saturated aqueous NaHCO₃(30 mL)

C₄₉H₉₀Cl₃N₅O₁₃Si₃Na [MNa] : 1168.48, found: 1168.6, 135; [a]²_D⁵ 13.6

and diluted with EtOAc (300 mL).

The layers were separated and the

organic layer was washed with 1m

(c 1.0 in CHCl₃).

A solution of ZnCl₂(1m in Et₂O, 100 mL) was added to this pentadep-

HCl (3 Â 50 mL), saturated aqueous

sipeptide (4.00 g, 3.49 mmol) and CH₃NO₂(10.0 mL). The Et₂O was

NaHCO₃(2 Â 50 mL), brine (1 Â

evaporated with

a stream of argon and the resulting solution was

50 mL), dried (MgSO₄), and concen-

trated.

maintained at rt for 41 h. The reaction was diluted with EtOAc (200 mL)

and quenched with 1m HCl (50 mL). The layers were separated and organic

layer was washed with 1m HCl (3 Â 50 mL), saturated aqueous NaHCO₃

(2 Â 50 mL), brine (1 Â 50 mL), dried (Na₂SO₄), and concentrated.

Piperidine (3.30 mL, 33.5 mmol) was

added to a 08C solution of this crude

residue and CH₃CN (50.0 mL). The resulting mixture was maintained at

08C for 30 min, at rt for 1 h and then concentrated. The crude residue was

purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O) to

give 43 (2.70 g, 92%) as a colorless oil that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 8.32 (d, J 6.7 Hz, 1H), 6.71 (d,

J 8.7 Hz, 1H), 5.91 ± 5.85 (m, 1H), 5.30 (d, J 7.2 Hz, 1H), 5.22 (d, J

10.3 Hz, 1H), 4.75 ± 3.86 (m, 10H), 3.45 ± 3.36 (m, 1H), 3.25 ± 3.15 (m, 1H),

2.15 ± 1.84 (m, 3H), 1.65 (brs, 3H), 1.11 (d, J 6.3 Hz, 3H), 1.06 ± 0.89 (m,

14H), 0.83 (s, 9H), 0.14 ± ꢀ 0.03 (m, 21H); ¹³C NMR (125 MHz, CDCl₃):

TBSOTf (6.40 mL, 27.9 mmol) was added to a ꢀ788C solution of the crude

residue, lutidine (6.50 mL, 55.8 mmol), and CH₂Cl₂(25 mL). The solution

was maintained at ꢀ788C for 4 h, and then poured into a mixture of EtOAc

(200 mL) and saturated aqueous NaHCO₃(50 mL). The layers were

separated and organic layer was washed with 1m HCl (3 Â 50 mL),

saturated aqueous NaHCO₃(2 Â 50 mL), brine (1 Â 50 mL), dried

(MgSO₄), and concentrated. The crude residue was purified by chroma-

tography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1

hexanes/Et₂O then 3:1 Et₂O/hexanes then Et₂O) to give the free piperazic

NH-derivative (3.32 g, 95%) as a colorless oil that was homogeneous by

56

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0056 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 6.88 (d, J 9.3 Hz, 1H),

6.81 (d, J 9.9 Hz, 1H), 6.65 (d, J 9.6 Hz, 1H), 5.95 ± 5.87 (m, 1H), 5.67

(d, J 7.4 Hz, 1H), 5.31 (d, J 17.2 Hz, 1H), 5.23 (d, J 10.4 Hz, 1H), 5.00

(d, J 6.6 Hz, 1H), 4.92 (d, J 12.0 Hz, 1H), 4.85 ± 4.77 (m, 2H), 4.75 ± 4.70

(m, 2H), 4.67 (dd, J 9.6, 4.0 Hz, 1H), 4.64 (dd, J 12.9, 6.2 Hz, 1H),

4.40 ± 4.32 (m, 1H), 4.31 (d, J 12.0 Hz, 1H), 3.83 (brs, 1H), 2.85 ± 2.76 (m,

3H), 2.41 ± 2.33 (m, 1H), 2.29 ± 2.14 (m, 1H), 1.85 ± 1.77 (m, 1H), 1.68 ± 1.51

(m, 3H), 1.12 (d, J 6.2 Hz, 3H), 1.05 (d, J 6.2 Hz, 3H), 1.00 (d, J

6.7 Hz, 3H), 0.99 (d, J 6.7 Hz, 3H), 0.87 (t, J 7.1 Hz, 6H), 0.82 (s, 9H),

0.78 (s, 9H), 0.070 (s, 3H), 0.0012 (s, 3H), ꢀ0.016 (s, 3H), ꢀ0.066 (s, 3H);

¹³C NMR (125 MHz, CDCl₃): d 173.7, 172.6, 171.3, 170.4, 168.1, 154.5,

131.3, 119.3, 95.5, 74.4, 69.3, 66.3, 62.2, 59.0, 57.3, 54.0, 52.5, 49.3, 43.2, 31.9,

30.2, 29.4, 25.9, 25.7, 24.5, 23.2, 21.6, 21.0, 19.1, 18.7, 18.6, 18.3, 17.8, 17.1,

ꢀ4.5, ꢀ5.0, ꢀ5.2, ꢀ5.7; IR (film): nÄ 3388, 2958, 1735, 1717, 1672, 1509,

834 cm^ꢀ1

;

ES-MS: m/z: calcd for C₁₂₀H₂₀₂Cl₆N₁₄O₂₆Si₆Na [MNa] :

2656.16, found: 2656.3, 135; [a]²_D⁵ ꢀ119 (c 0.65 in CHCl₃).

trans-Hexadepsipeptide carboxylic acid dimer 47: PhSiH₃(0.15 mL,

1.23 mmol) was added to a 08C solution of 46 (530 mg, 0.201 mmol),

[Pd(Ph₃P)₄] (20 mg) and

THF (5 mL). The resulting

solution was maintained at

08C for 45 min, warmed to

rt for 15 min, quenched with

H₂O (0.20 mL) and then

concentrated. The crude

residue was purified by

chromatography on silica

gel (8:1 hexanes/Et₂O then

3:1 hexanes/Et₂O then 1:1

1470, 1375, 1160, 1101, 914, 836 cm^ꢀ1

; ES-MS: m/z: calcd for

C₄₃H₇₈Cl₃N₅O₁₁Si₂Na [MNa] : 1024.42, found: 1024.6, 135; [a]²_D⁵ 19.7

Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give 47

(469 mg, 91%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 7.80 ± 6.41 (m, 12H), 5.80 ± 5.52 (m, 2H),

5.35 ± 4.38 (m, 16H), 4.10 ± 3.79 (m, 4H), 3.08 ± 2.17 (m, 16H), 1.82 ± 1.20

(m, 8H), 1.18 ± 0.75 (m, 96H), 0.55 ± 0.31 (m, 12H), 0.28 ± ꢀ 0.10 (m, 24H);

¹³C NMR (125 MHz, CDCl₃): d 174.2, 173.1, 171.3, 171.2, 170.3, 169.8,

168.2, 167.9, 154.5, 154.3, 153.6, 151.3, 134.1, 132.0, 128.5, 128.4, 128.0, 127.5,

127.0, 125.3, 110.5, 95.4, 95.1, 82.7, 76.4, 74.6, 74.5, 74.3, 69.6, 62.1, 59.2, 55.8,

54.0, 53.3, 49.2, 49.1, 42.8, 32.5, 31.7, 31.6, 31.4, 30.4, 30.2, 29.5, 29.4, 26.0,

25.8, 24.4, 24.1, 23.2, 22.7, 22.5, 22.1, 21.9, 21.6, 19.8, 19.3, 19.2, 19.0, 18.8,

18.6, 18.4, 18.3, 18.2, 17.9, 17.3, 17.1, 16.8, 6.7, 6.6, 5.7, 5.5, 5.4, ꢀ4.7, ꢀ4.8,

ꢀ4.9, ꢀ5.0, ꢀ5.2, ꢀ5.5, ꢀ5.6; IR (film): nÄ 3378, 3302, 2956, 1731, 1675,

1642, 1253, 1217, 1116, 834 cm^ꢀ1; ES-MS: m/z: calcd for C₁₁₄H₁₉₄Cl₆N₁₄O₂₆-

(c 1.0 in CHCl₃).

PhSiH₃(0.810 mL, 6.58 mmol) was added to a 08C solution of this

derivative (3.30 g, 3.29 mmol), [Pd(Ph₃P)₄] (50 mg) and THF (20 mL).

After 45 min, the reaction was quenched with H₂O (0.50 mL) and

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then

3:1 Et₂O/hexanes then Et₂O) to give 45 (2.71 g, 86%) as a colorless solid

that was homogeneous by TLC analysis: m.p. 226 ± 2288C; ¹H NMR

(500 MHz, CDCl₃): d 6.87 (d, J 7.3 Hz, 1H), 6.82 (d, J 9.5 Hz, 1H),

6.23 (d, J 9.4 Hz, 1H), 5.69 ± 5.66 (m, 1H), 4.99 (d, J 7.0 Hz, 1H), 4.88

(d, J 12.0 Hz, 1H), 4.84 ± 4.62 (m, 2H), 4.61 ± 4.59 (m, 1H), 4.52 (dd, J

9.5, 4.4 Hz, 1H), 4.45 (d, J 12.0 Hz, 1H), 4.45 ± 4.38 (m, 1H), 3.83 (brs,

1H), 2.85 ± 2.73 (m, 3H), 2.42 ± 2.30 (m, 1H), 2.28 ± 2.18 (m, 1H), 1.89 ± 1.75

(m, 1H), 1.65 ± 1.45 (m, 3H), 1.08 (d, J 6.2 Hz, 3H), 1.04 (d, J 6.8 Hz,

3H), 1.00 ± 0.80 (m, 15H), 0.88 (s, 9H), 0.81 (s, 9H), 0.11 (s, 3H), 0.10 (s,

3H), 0.057 (s, 3H), 0.0064 (s, 3H); ¹³C NMR (125 MHz, CDCl₃): d 173.5,

172.5, 171.9, 171.3, 168.5, 154.5, 95.4, 74.6, 68.4, 62.1, 59.1, 57.1, 54.0, 52.3,

49.2, 42.7, 31.7, 30.4, 29.4, 25.9, 25.7, 24.5, 23.0, 21.8, 19.4, 19.0, 18.7, 18.4, 18.0,

17.8, 17.1, ꢀ4.8, ꢀ5.0, ꢀ5.1, ꢀ5.6; IR (film): nÄ 3382, 3019, 1733, 1671,

Si₆Na [MNa] : 2576.09, found: 2576.4, 135; [a]²_D⁵ ꢀ94.5 (c 0.65 in

CHCl₃).

Isohimastatin (1): Pb/Cd couple (300 mg) was added to a vigorously stirred

mixture of 47 (305 mg, 0.120 mmol), THF (5.0 mL), and aqueous 1m

NH₄OAc (5.0 mL). After

1.5 h, the layers were separat-

ed, and the aqueous layer was

1515, 1255, 1215, 1103, 1036, 836 cm^ꢀ1

;

ES-MS: m/z: calcd for

extracted with EtOAc (4 Â

C₄₀H₇₄Cl₃N₅O₁₁Si₂Na [MNa] : 984.39, found: 984.5, 135; [a]²_D⁵ 19.8

5 mL). The combined organic

(c 1.0 in CHCl₃).

layers were dried (Na₂SO₄)

and concentrated. This crude

trans-Hexadepsipeptide dimer 46: HATU (489 mg, 1.29 mmol) was added

to a ꢀ108C solution of 24 (240 mg, 0.322 mmol), 45 (1.24 g, 1.29 mmol),

HOAt (350 mg, 2.57 mmol),

amino acid (250 mg, ꢁ95%)

was split into two equal batch-

es and used without further

purification.

collidine

(0.51 mL,

3.86 mmol), and CH₂Cl₂

(5.0 mL). The reaction mix-

HATU (86.0 mg, 0.227 mmol) was added to a solution of the crude amino

acid (ꢁ125 mg), HOAt (62.0 mg, 0.454 mmol), iPr₂NEt (0.12 mL,

0.681 mmol), and DMF (40.0 mL) at rt. The resulting solution was

maintained at rt for 28 h and then concentrated. The resulting crude

residue was dissolved in EtOAc (10 mL) and washed with aqueous 1m HCl

(3 Â 5 mL), saturated aqueous NaHCO₃(3 Â 5 mL), brine (1 Â 5 mL), dried

(Na₂SO₄), and concentrated. The two batches of cyclic dimer 48 (ꢁ145 mg

each) were recombined at this stage and used without further purification.

ture was stirred at ꢀ108C

for 19 h, 08C for 22 h, rt for

11 h and then quenched with

1m HCl (5 mL) and diluted

with EtOAc (30 mL). The

layers were separated and

the organic layer was wash-

ed with 1m HCl (3 Â 10 mL), saturated aqueous NaHCO₃(3 Â 10 mL),

brine (1 Â 10 mL), dried (MgSO₄), and concentrated. The crude residue

was purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O

then EtOAc) to yield 46 (565 mg, 67%) as a colorless oil that was

homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.90 ± 7.80

(m, 1H), 7.35 ± 7.15 (m, 5H), 6.80 ± 6.41 (m, 6H), 5.65 ± 4.32 (m, 26H),

4.12 ± 3.78 (m, 6H), 3.10 ± 2.65 (m, 10H), 2.50 ± 2.17 (m, 6H), 1.81 ± 1.33 (m,

8H), 1.15 ± 0.75 (m, 96H), 0.54 ± 0.39 (m, 12H), 0.31 ± 0.01 (m, 24H);

¹³C NMR (125 MHz, CDCl₃): d 174.4, 173.3, 171.4, 171.2, 171.0, 170.9,

170.8, 170.1, 169.7, 169.4, 169.3, 169.2, 169.1, 169.0, 168.3, 168.2, 168.1, 167.8,

166.7, 154.5, 154.4, 154.2, 154.1, 153.6, 153.5, 150.5, 150.4, 149.2, 149.0, 135.8,

135.0, 132.7, 132.2, 131.5, 131.4, 131.2, 129.2, 129.1, 129.0, 128.0, 127.9, 127.2,

127.1, 125.3, 123.0, 118.4, 118.2, 110.1, 109.8, 95.6, 95.1, 86.5, 86.4, 82.2, 82.1,

82.0, 76.5, 74.5, 74.3, 70.0, 69.8, 66.5, 66.2, 62.1, 62.0, 59.5, 59.4, 59.1, 56.6,

54.2, 54.0, 53.4, 52.4, 49.3, 49.2, 46.7, 46.1, 43.2, 43.0, 42.8, 42.6, 42.2, 32.5,

31.5, 30.3, 30.2, 29.5, 29.4, 26.0, 25.9, 25.8, 25.6, 25.5, 24.5, 24.3, 23.2, 22.9,

21.8, 21.6, 19.3, 19.1, 18.8, 18.7, 18.6, 18.4, 18.3, 18.2, 18.0, 17.9, 17.7, 17.5, 17.3,

16.6, 6.7, 6.6, 5.4, 5.3, ꢀ4.7, ꢀ4.8, ꢀ4.9, ꢀ5.0, ꢀ5.2, ꢀ5.3, ꢀ5.5, ꢀ5.6, ꢀ5.7,

ꢀ5.8; IR (film): nÄ 3381, 2932, 1734, 1676, 1644, 1442, 1254, 1102,

A solution of TBAF (1m in THF, 3.40 mL) was added to a 08C solution of

the crude residue 48 (ꢁ290 mg), HOAc (0.583 mL, 10.2 mmol), and THF

(3 mL). The resulting solution was allowed to warm to rt, maintained at rt

for 55 h and then concentrated to a slurry. The reaction mixture was filtered

through a short pad of sand and washed with CH₂Cl₂(200 mL). The filtrate

was discarded. The remaining white solid was washed with hot MeOH

(250 mL) until all the solid had dissolved and the filtrate was concentrated

to give 1 (85.4 mg, 48%) as a white solid that was >95% pure as judged by

¹³C NMR analysis: m.p. >4008C (dec); ¹H NMR (500 MHz, [D₆]DMSO):

d 8.24 (d, J 8.4 Hz, 2H), 7.68 (d, J 8.1 Hz, 2H), 7.35 (s, 2H), 7.20 ± 7.15

(m, 4H), 7.00 (d, J 4.2 Hz, 2H), 6.54 (d, J 8.2 Hz, 2H), 5.94 (s, 2H), 5.75

(d, J 7.3 Hz, 2H), 5.52 (d, J 4.3 Hz, 2H), 5.45 (d, J 6.0 Hz, 2H), 5.29

(d, J 5.2 Hz, 2H), 5.20 (d, J 11.8 Hz, 2H), 4.98 ± 4.90 (m, 4H), 4.61 (dd,

J 8.4, 4.3 Hz, 2H), 4.55 ± 4.48 (m, 2H), 4.30 (dd, J 8.8, 4.0 Hz, 2H),

3.90 ± 3.82 (m, 2H), 3.60 ± 3.50 (m, 2H), 2.90 ± 2.75 (m, 3H), 2.49 (d, J

13.2 Hz, 2H), 2.06 ± 1.90 (m, 6H), 1.71 ± 1.60 (m, 2H), 1.60 ± 1.51 (m, 2H),

1.49 ± 1.39 (m, 2H), 1.05 (d, J 6.3 Hz, 6H), 0.95 ± 0.80 (m, 24H), 0.39 (d,

J 6.9 Hz, 6H), 0.23 (d, J 6.9 Hz, 6H); ¹³C NMR (125 MHz, DMSO):

d 170.6, 170.4, 170.3, 169.9, 169.2, 147.9, 132.2, 127.1, 121.2, 110.8, 86.0,

84.0, 75.1, 67.3, 61.8, 58.9, 55.6, 53.7, 52.7, 50.8, 49.6, 43.8, 42.3, 31.7, 29.1, 28.8,

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0057 $ 17.50+.50/0

57

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

24.1, 23.1, 22.3, 18.7, 18.6, 17.7, 17.4, 17.3; IR (film): nÄ 2957, 2931, 2857, 1689,

19.3, 19.2, 18.8, 18.7, 18.6, 18.5, 18.3, 18.2, 18.0, 17.4, 16.7, 6.7, 6.6, 5.6, 5.4, 5.3,

ꢀ4.7, ꢀ4.8, ꢀ4.9, ꢀ5.0, ꢀ5.1, ꢀ5.4, ꢀ5.5, ꢀ5.6; IR (film): nÄ 3382, 2956,

1733, 1676, 1642, 1255, 1039, 1004, 755 cm^ꢀ1; ES-MS: m/z: calcd for

1253, 1100, 1083 cm^ꢀ1; ES-MS: m/z: calcd for C₇₂H₁₀₄N₁₄O₂₀Na [MNa] :

1507.74, found: 1508.1, 135.

C₆₀H₁₀₂Cl₃N₇O₁₃Si₃Na [MNa] : 1340.58, found: 1340.8, 135; [a]²_D⁵ ꢀ86.6

Allyl ester 49: Neat TESOTf (10.4 mL, 45.1 mmol) was added to a 08C

solution of 11 (2.60 g, 50.4 mmol), 2,6-lutidine (10.5 mL, 90.3 mmol), and

CH₂Cl₂(20 mL). The resulting solu-

(c 0.65, CHCl₃).

trans-Hexadepsipeptide carboxylic acid monomer 52: The same procedure

for the preparation of 47 was used with the following amounts: PhSiH₃

(36.0 mL, 0.295 mmol), 51 (195 mg,

tion was allowed to warm to rt over-

night and then quenched with satu-

rated aqueous NaHCO₃(100 mL) and

0.148 mmol), [Pd(Ph₃P)₄] (20 mg),

diluted with EtOAc (100 mL). The

and THF (3.0 mL) to give 52

layers were separated and the organic layer was washed with aqueous 1m

HCl (2 Â 50 mL), brine (1 Â 50 mL), dried (Na₂SO₄), and concentrated.

(174 mg, 92%) as a colorless oil that

was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d

EDCI (1.70 g, 8.90 mmol) was added to a 08C solution of the crude residue,

allyl alcohol (31.0 mL, 451 mmol), DMAP (20 mg), and CH₂Cl₂(50 mL).

The resulting solution was allowed to warm to rt overnight and then

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then

3:1 Et₂O/hexanes then Et₂O) to give 49 (2.40 g, 78%) as a yellow solid that

was homogeneous by TLC analysis: m.p. 112 ± 1148C; ¹H NMR (500 MHz,

CDCl₃): d 9.41 (d, J 9.3 Hz, 2H), 8.71 (s, 1H), 8.03 (d, J 8.4 Hz, 2H),

7.74 ± 7.62 (m, 2H), 7.53 (t, J 7.5 Hz, 2H), 7.10 ± 7.01 (m, 1H), 6.69 (t, J

7.4 Hz, 1H), 6.51 (t, J 8.2 Hz, 1H), 5.45 ± 5.31 (m, 1H), 5.12 ± 5.01 (m,

2H), 4.98 (s, 1H), 4.91 (d, J 7.8 Hz, 1H), 4.75 (s, 1H), 3.98 (dd, J 13.0,

5.7 Hz, 1H), 3.80 (dd, J 13.0, 5.8 Hz, 1H), 2.78 (d, J 12.6 Hz, 1H), 2.53

(dd, J 12.6, 9.0 Hz), 0.56 (t, J 7.9 Hz, 9H), 0.20 ± 0.08 (m, 6H); ¹³C NMR

(125 MHz, CDCl₃): d 170.0, 150.1, 136.5, 131.3, 131.2, 130.6, 129.5, 129.1,

128.2, 127.6, 125.4, 125.0, 124.5, 119.1, 118.2, 110.3, 89.3, 82.6, 66.0, 61.0, 43.4,

7.75 ± 6.39 (m, 7H), 5.70 ± 5.58 (m,

1H), 5.22 ± 4.37 (m, 10H), 4.16 ± 3.60

(m, 2H), 3.15 ± 2.15 (m, 7H), 1.80 ±

1.30 (m, 4H), 1.16 ± 0.71 (m, 48H),

0.45 ± 0.30 (m, 6H), 0.29 ± 0.01 (m, 12H); ¹³C NMR (125 MHz, CDCl₃): d

174.2, 173.7, 172.3, 172.0, 171.7, 171.5, 171.3, 171.2, 170.1, 169.9, 168.7, 168.3,

166.9, 135.0, 134.3, 132.0, 131.0, 130.7, 128.4, 127.5, 127.4, 125.4, 124.4, 118.3,

117.6, 110.2, 109.9, 95.5, 95.4, 95.2, 87.2, 86.8, 82.0, 81.9, 76.4, 75.8, 74.8, 74.6,

74.4, 69.7, 69.1, 62.3, 59.5, 59.2, 59.1, 55.7, 53.8, 53.2, 52.5, 49.3, 49.2, 46.8,

43.4, 42.9, 42.4, 32.7, 31.6, 31.3, 30.5, 30.2, 29.6, 29.5, 26.0, 25.9, 25.8, 25.5,

24.4, 24.2, 23.4, 23.0, 21.9, 20.0, 19.5, 19.0, 18.7, 18.6, 18.3, 18.0, 17.7, 17.6, 17.1,

ꢀ4.6, ꢀ4.7, ꢀ4.9, ꢀ5.0, ꢀ5.6; IR (film): nÄ 3379, 3298, 2956, 1735, 1675,

1642, 1253, 1108, 1038, 756 cm^ꢀ1; ES-MS: m/z: calcd for C₅₇H₉₈Cl₃N₇O₁₃-

Si₃Na [MNa] : 1300.55, found: 1300.7, 135; [a]²_D⁵ ꢀ76.4 (c 0.4 in

6.4, 5.1; IR (film): nÄ 3408, 3020, 2956, 1727, 1612, 1216, 1162, 908 cm^ꢀ1

;

CHCl₃).

MS: m/z: calcd for C₃₄H₃₈N₂O₅SSiNa [MNa] : 637.22, found: 637.3, 135;

trans-Cyclic monomer 3: The same procedure for the preparation of 1 was

used with the following amounts: Pb/Cd couple (100 mg), 52 (158 mg,

0.12 mmol), THF (3.0 mL), aqueous

1m NH₄OAc (3.0 mL); HATU

(11.0 mg, 0.029 mmol), the crude ami-

[a]²_D⁵ ꢀ156 (c 1.0 in CHCl₃).

Pyrroloindoline 50: Freshly prepared Al(Hg) (1.0 g) was added to a

vigorously stirred mixture of 49 (2.90 g, 4.72 mmol) and THF/H₂O 10:1

(200 mL) at 08C. The resulting mix-

no acid (ꢁ120 mg), HOAt (45.0 mg,

ture turned from bright yellow to gray

over 1 h indicating consumption of 49.

The mixture was filtered through a

pad of celite (EtOAc) and concen-

0.327 mmol),

iPr₂NEt

(57.0 mL,

0.327 mmol), and DMF (75.0 mL);

TBAF (1m in THF, 1.44 mL), the

crude cyclic monomer (ꢁ130 mg),

trated. The crude residue was purified

HOAc (0.24 mL, 4.52 mmol), and

by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to

give 50 (1.17 g, 69%) as a near colorless oil that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 7.20 (d, J 8.1 Hz, 1H), 7.11 (dt,

J 7.5, 1.2 Hz, 1H), 6.73 (dt, J 7.4, 0.7 Hz, 1H), 6.57 (d, J 7.9 Hz, 1H),

5.75 ± 5.60 (m, 1H), 5.25 ± 5.10 (m, 2H), 4.86 (s, 1H), 4.35 ± 4.22 (m, 2H),

4.03 (dd, J 7.3, 5.3 Hz, 1H), 2.67 (dd, J 12.7, 7.4 Hz, 1H), 2.50 (dd, J

12.7, 5.3 Hz, 1H), 0.83 (t, J 3.3 Hz, 9H), 0.50 ± 0.30 (m, 6H); ¹³C NMR

(125 MHz, CDCl₃): d 172.9, 149.4, 131.7, 130.5, 129.6, 124.4, 118.5, 118.1,

110.1, 90.0, 84.4, 65.3, 59.4, 44.5, 6.6, 5.5; IR (film): nÄ 3368, 2875, 1737,

THF (1.0 mL) to give 3 (76.6 mg, 76%) as a colorless solid that was

homogeneous by TLC analysis: m.p. >3208C (dec); H NMR (500 MHz,

1

CDCl₃): d 7.50 (brs, 1H), 7.40 ± 7.20 (m, 2H), 7.01 (t, J 7.6 Hz, 1H), 6.74

(t, J 7.4 Hz, 1H), 6.55 (brs, 1H), 6.45 (d, J 7.8 Hz, 1H), 5.84 (brs, 1H),

5.77 (s, 1H), 5.45 (d, J 4.8 Hz, 1H), 5.17 (d, J 10.0 Hz, 1H), 4.86 (d, J

5.7 Hz, 1H), 4.74 (dd, J 9.0, 3.0 Hz, 1H), 4.74 ± 4.58 (m, 3H), 4.52 (s, 1H),

4.33 (d, J 12.3 Hz, 1H), 4.28 (s, 1H), 3.63 (s, 1H), 3.15 ± 3.00 (m, 1H), 2.72

(d, J 14.0 Hz, 1H), 2.60 (t, J 13.5 Hz, 1H), 2.60 ± 2.55 (m, 1H), 2.14 (d,

J 14.7 Hz, 1H), 2.08 ± 1.97 (m, 1H), 1.95 ± 1.81 (m, 1H), 1.80 ± 1.72 (m,

1H), 1.70 ± 1.58 (m, 3H), 1.23 (d, J 6.2 Hz, 3H), 1.00 ± 0.95 (m, 9H), 0.90

(d, J 6.2 Hz, 3H), 0.51 (d, J 6.5 Hz, 3H), 0.28 (d, J 6.7 Hz, 3H);

¹³C NMR (125 MHz, CDCl₃): d 171.8, 171.7, 171.3, 171.0, 170.9, 170.5,

148.2, 131.6, 129.7, 124.4, 120.4, 111.1, 86.6, 84.9, 75.9, 68.0, 62.2, 58.8, 56.5,

55.7, 53.1, 52.4, 50.0, 43.6, 32.1, 29.4, 28.6, 24.9, 23.0, 22.7, 19.3, 18.8, 18.1,

1612, 1470, 1203, 1008 cm^ꢀ1; ES-MS: m/z: calcd for C₂₀H₃₁N₂O₃Si [MH] :

375.20, found: 375.2, 135; [a]²_D⁵ ꢀ70.3 (c 1.0 in CHCl₃).

trans-Hexadepsipeptide monomer 51: The same procedure for the

preparation of 46 was used with the following amounts: HATU (198 mg,

0.52 mmol), 50 (195 mg, 0.52 mmol),

17.4, 17.1; IR (film): nÄ 3324, 2963, 1736, 1653, 1526, 1428, 1204, 1100 cm^ꢀ1

;

acid 45 (250 mg, 0.26 mmol), HOAt

ES-MS: m/z: calcd for C₃₆H₅₃N₇O₁₀Na [MNa] : 766.38, found: 766.5, 135;

(141 mg,

1.04 mmol),

collidine

[a]²_D⁵ ꢀ48.3 (c 1.0 in CHCl₃).

(273 mL, 2.08 mmol), and CH₂Cl₂

(3.0 mL) to give 51 (224 mg, 66%)

as a colorless oil that was homoge-

neous by TLC analysis: ¹H NMR

trans-Pyrroloindoline-d-valinol: tert-Butyl ester 11 (110 mg, 0.213 mmol)

was added to TFA (1.50 mL) at 08C. The reaction was maintained at 08C

for 30 min, at rt for 3 h and concen-

trated.

(500 MHz, CDCl₃): d 7.80 ± 7.00

(m, 3H), 6.80 ± 6.41 (m, 4H), 5.71 ±

EDCI (82.0 mg, 0.430 mmol) was add-

5.40 (m, 2H), 5.38 ± 4.35 (m, 11H),

ed to a 08C solution of this crude

4.10 ± 3.90 (m, 2H), 3.85 (brs, 1H), 3.05 ± 2.62 (m, 5H), 2.41 ± 2.16 (m, 2H),

1.78 ± 1.35 (m, 4H), 1.18 ± 0.75 (m, 48H), 0.46 ± 0.31 (m, 6H), 0.25 ± 0.20 (m,

12H); ¹³C NMR (125 MHz, CDCl₃): d 174.5, 173.4, 171.4, 171.2, 171.1,

170.9, 170.8, 170.0, 169.7, 169.3, 169.1, 168.2, 168.1, 167.9, 166.8, 154.6, 153.6,

151.7, 150.5, 131.6, 131.5, 131.4, 130.6, 127.4, 126.7, 124.9, 124.8, 118.6, 118.4,

118.3, 117.7, 109.9, 109.6, 95.6, 95.2, 86.5, 86.4, 81.8, 81.7, 76.5, 75.8, 74.5, 74.3,

70.1, 69.8, 66.6, 66.3, 62.2, 62.1, 59.5, 59.3, 59.2, 59.1, 56.6, 55.8, 54.1, 54.0,

53.4, 52.5, 52.4, 49.3, 49.2, 43.0, 42.9, 42.7, 42.5, 32.6, 31.5, 31.3, 30.4, 30.3,

29.6, 29.4, 26.0, 25.9, 25.8, 25.5, 24.5, 24.4, 23.2, 23.0, 22.5, 21.8, 21.7, 19.4,

residue,

d-valinol

(66.0 mg,

0.640 mmol), DMAP (10 mg), and

CH₂Cl₂(10 mL). The reaction was

allowed to warm to rt overnight and then concentrated. The crude residue

was taken up in EtOAc (100 mL) and washed with 1m HCl (2 Â 50 mL),

saturated aqueous NaHCO₃(1 Â 50 mL), brine (1 Â 50 mL), dried

(MgSO₄), and concentrated. The crude residue was purified by chroma-

tography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1

Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give the

58

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0058 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

title compound (58.0 mg, 50%) as a pale yellow solid that was homoge-

neous by TLC analysis: m.p. 201 ± 2048C; ¹H NMR (500 MHz, CDCl₃): d

9.25 (d, J 9.2 Hz, 2H), 8.72 (s, 1H), 8.01 (d, J 8.3 Hz, 2H), 7.69 ± 7.66 (m,

2H), 7.52 ± 7.49 (m, 2H), 7.13 (d, J 7.36 Hz, 1H), 7.08 ± 7.06 (m, 1H), 6.80

(d, J 8.5 Hz, 1H), 6.73 (t, J 7.4 Hz, 1H), 6.60 (d, J 8.0 Hz, 1H), 5.59 (s,

1H), 5.26 (s, 1H), 4.39 (dd, J 9.9, 0.2 Hz, 2H), 3.09 ± 3.03 (m, 2H), 3.03 (s,

1H), 3.00 ± 2.92 (m, 1H), 2.74 (d, J 13.4 Hz, 1H), 2.48 ± 2.44 (m, 2H),

1.24 ± 1.20 (m, 1H), 0.48 (d, J 6.8 Hz, 3H), 0.33 (d, J 6.8 Hz, 3H);

¹³C NMR (125 MHz, CDCl₃): d 170.9, 148.2, 137.4, 131.3, 131.2, 130.8,

129.8, 129.4, 125.6, 125.5, 124.4, 124.2, 120.8, 110.9, 88.0, 85.3, 62.9, 62.8,

57.6, 41.2, 28.6, 18.7, 18.5; IR (film): nÄ 3382, 3278, 3018, 1652, 1216, 909,

18.7; IR (film): nÄ 3300, 2964, 1673, 1566, 1203, 1137, 1027 cm^ꢀ1; ES-MS:

m/z: calcd for C₁₆H₂₃N₃O₃Na [MNa] : 328.16, found: 328.1, 135; [a]_D²⁵

ꢀ59.0 (c 0.5 in MeOH).

Pyrroloindoline tert-butyl ester 56: Neat pyridine (9.30 mL, 116 mmol) was

added to a 08C solution of 14 (4.00 g, 14.5 mmol), ClCO₂Bn (8.20 mL,

58.0 mmol), and CH₂Cl₂(100 mL).

The resulting solution was allowed

to warm to rt overnight and then

concentrated. The crude residue was

dissolved in EtOAc (100 mL) and

washed with 1m HCl (2 Â 50 mL), brine (1 Â 50 mL), dried (MgSO₄), and

concentrated. This residue was purified by chromatography on silica gel

(8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1

Et₂O/hexanes then Et₂O) to give the bis-carbamate (5.00 g, 63%) as a near

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.63 (brs, 1H), 7.36 (d, J 7.4 Hz, 1H), 7.29 ± 7.16 (m, 11H),

7.08 (t, J 7.4 Hz, 1H), 6.08 (s, 1H), 5.21 ± 4.43 (m, 4H), 3.96 (brs, 1H), 3.75

(s, 1H), 2.77 (dd, J 12.9, 7.7 Hz, 1H), 2.40 (dd, J 12.9, 9.1 Hz, 1H), 1.46

(brs, 9H); ¹³C NMR (125 MHz, CDCl₃): d 171.0, 153.9, 153.3, 141.4,

135.8, 135.6, 132.2, 130.5, 128.5, 128.3, 128.2, 128.0, 127.9, 124.1, 123.1, 117.4,

83.7, 81.9, 81.4, 67.6, 67.2, 59.9, 37.5, 27.6; IR (film): nÄ 3426, 2978, 1720,

1395, 1323, 1152, 1079, 735 cm^ꢀ1; ES-MS: m/z: calcd for C₃₁H₃₂N₂O₇Na

756 cm^ꢀ1; ES-MS: m/z:calcd for C₃₀H₃₁N₃O₅SNa [MNa] : 568.19, found:

568.1; [a]²_D⁵ ꢀ69.9 (c 1.0 in CHCl₃).

trans-Degradation product (53): Freshly prepared Al(Hg) (500 mg) was

added to a vigorously stirred mixture of the sulfonylated dipeptide (58.0 g,

0.107 mmol) and THF/H₂O 10:1

(20 mL) at 08C. The resulting mixture

turned from bright yellow to gray

over 1 h indicating consumption of

starting material. The mixture was

filtered through

a pad of celite

(EtOAc) and concentrated. The

crude residue was purified by chro-

[MNa] : 567.21, found: 567.3; [a]²_D⁵ ꢀ79.9 (c 1.0 in CHCl₃).

matography on silica gel (3:1 hexanes/Et₂O then 3:1 Et₂O/hexanes then

Et₂O then EtOAc then 1:1 EtOAc/MeOH) to give 53 (15 mg, 50%) as a

pale yellow solid. Further purification by chromatography on silica gel

(Et₂O then CH₂Cl₂then 8:1 CHCl₃/MeOH then 4:1 CHCl₃/MeOH then

MeOH) gave 53 (10.0 mg, 31%) as a colorless solid that was homogeneous

by TLC analysis: m.p. >3108C (dec); ¹H NMR (500 MHz, MeOH): d

7.12 (dd, J 7.4, 0.7 Hz, 1H), 6.96 (dt, J 7.7, 1.16 Hz, 1H), 6.64 (dt, J 7.4,

0.6 Hz, 1H), 6.52 (d, J 7.9 Hz, 1H), 4.93 (s, 1H), 4.00 (dd, J 8.8, 6.5 Hz,

1H), 3.43 ± 3.41 (m, 3H), 2.57 (dd, J 13.4, 8.9 Hz, 1H), 2.15 (dd, J 13.5,

6.5 Hz, 1H), 1.60 ± 1.58 (m, 1H), 0.66 (d, J 6.8 Hz, 3H), 0.0.57 (d, J

6.8 Hz, 3H); ¹³C NMR (125 MHz, MeOH): d 177.2, 150.4, 133.3, 130.6,

124.6, 120.2, 111.8, 90.5, 88.5, 63.4, 63.2, 57.6, 45.4, 30.2, 19.9, 18.7; IR (film):

A solution of DBU (10.0 mL, 66.2 mmol) and DMF (50 mL) was added to a

solution of the bis-carbamate (9.00 g, 16.5 mmol), TBSCl (14.8 g,

99.2 mmol), and DMF (300 mL). The resulting solution was maintained

at rt for 8 h, and was then diluted with EtOAc (400 mL) and poured into

aqueous 1m HCl (200 mL). The layers were separated and the organic layer

was washed with aqueous 1m HCl (3 Â 100 mL), saturated aqueous

NaHCO₃(1 Â 100 mL), brine (2 Â 100 mL), dried (MgSO₄), and concen-

trated. The crude residue was purified by chromatography on silica gel (3:1

hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to give 56

(10.6 g, 97%) as a colorless oil that was homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d 7.71 (brs, 1H), 7.38 ± 7.22 (m, 12H), 7.12 (t,

J 7.4 Hz, 1H), 6.00 (brs, 1H), 5.40 ± 4.40 (m, 4H), 3.89 (brs, 1H), 2.87 (dd,

J 12.3, 7.1 Hz, 1H), 2.41 (t, J 12.2 Hz, 1H), 1.60 ± 1.22 (brm, 9H), 0.74

(s, 9H), ꢀ0.29 ± ꢀ 0.50 (m, 6H); ¹³C NMR (125 MHz, CDCl₃): d 170.6,

154.4, 153.1, 142.4, 136.0, 132.1, 130.6, 128.3, 128.2, 128.0, 123.9, 118.0, 85.2,

81.5, 81.0, 67.4, 59.6, 38.5, 27.8, 25.3, 17.7, ꢀ3.9, ꢀ4.4; IR (film): nÄ 3032,

2953, 2856, 1717, 1479, 1285, 1151, 837 cm^ꢀ1; ES-MS: m/z: calcd for

nÄ 3352, 2944, 1642, 1467, 1117, 1027 cm^ꢀ1

; ES-MS: m/z: calcd for

C₁₆H₂₃N₃O₃Na [MNa] : 328.16, found: 328.1, 135; [a]²_D⁵ ꢀ2.97 (c

0.35 in MeOH).

cis-Degradation product 54: Neat TMSOTf (0.760 mL, 3.94 mmol) was

added to a 08C solution of 14 (100 mg, 0.197 mmol), 2,6-lutidine (1.40 mL,

11.8 mmol), and CH₂Cl₂(3 mL). The resulting solution was allowed to

warm to rt over 4 h and then

C₃₇H₄₆N₂O₇SiNa [MNa] : 681.30, found: 681.6, 135; [a]²_D⁵ ꢀ64.2 (c 1.0

in CHCl₃).

quenched with saturated aqueous

NaHCO₃(50 mL) and diluted with

EtOAc (100 mL). The layers were

separated and the organic layer was

5-Iodo-pyrroloindoline tert-butyl ester 57: The same procedure for the

preparation of 17 was used with the following amounts: ICl (1.50 mL,

30.4 mmol), 56 (2.00 g, 3.04 mmol),

2,6-di-tert-butyl-4-methyl

pyridine

washed with aqueous 1m HCl (3 Â

(6.23 g, 30.4 mmol), and CH₂Cl₂

(20 mL). The crude residue was puri-

fied by chromatography on silica gel

50 mL), brine (1 Â 50 mL), dried

(Na₂SO₄), and concentrated.

EDCI (76.0 mg, 0.396 mmol) was added to a 08C solution of this crude

residue, d-valinol (61.0 mg, 0.592 mmol), DMAP (10 mg), and CH₂Cl₂

(3 mL). The reaction was allowed to warm to rt overnight and then

concentrated. The crude residue was purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 1:1 hexanes/Et₂O then Et₂O then EtOAc) to

give a mixture of the 3a-OH and the 3a-OTMS ether (50 mg) as a near

colorless oil which was used without further purification.

(8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes) to give 57

(1.75 g, 73%) as a light yellow foam that was homogeneous by TLC

1

analysis: H NMR (500 MHz, CDCl₃): d 7.65 (d, J 7.2 Hz, 1H), 7.62 (s,

1H), 7.48 ± 7.03 (m, 11H), 5.97 (brs, 1H), 5.50 ± 4.40 (m, 4H), 3.92 (brs,

1H), 2.81 (dd, J 12.4, 7.2 Hz, 1H), 2.38 (dd, J 12.4, 10.3 Hz, 1H), 1.62 ±

1.24 (brm, 9H), 0.74 (s, 9H), ꢀ0.32 ± ꢀ 0.45 (m, 6H); ¹³C NMR (125 MHz,

CDCl₃): d 170.4, 154.3, 152.9, 142.1, 139.3, 135.8, 134.8, 132.8, 128.3, 128.2,

128.0, 119.9, 86.3, 84.9, 81.6, 80.8, 67.4, 59.3, 38.3, 27.7, 25.3, 17.6, ꢀ3.8, ꢀ4.3;

Piperidine (19.0 mL, 0.190 mmol) was added to a 08C solution of this

mixture (50 mg) and CH₃CN (2 mL). The resulting solution was allowed to

warm to rt over 3 h and then concentrated. TFA (1.0 mL) was added to a

08C solution of this residue and CH₂Cl₂(1.0 mL). After 10 min, the

reaction was concentrated and the crude residue was purified by

chromatography on TEA treated silica gel (3:1 Et₂O/hexanes then Et₂O

then EtOAc then 3:1 EtOAc/MeOH) to give 54 (30 mg, 50%) as a colorless

solid that was homogeneous by TLC analysis: m.p. >2508C (dec); ¹H NMR

(500 MHz, MeOH): d 7.25 (d, J 7.1 Hz, 1H), 7.09 (dt, J 7.8, 1.10 Hz,

1H), 6.76 (dt, J 7.4, 0.5 Hz, 1H), 6.62 (d, J 7.9 Hz, 1H), 5.06 (s, 1H), 3.73

(dd, J 11.2, 5.9 Hz, 1H), 3.67 ± 3.65 (m, 1H), 3.55 (dd, J 11.4, 4.4 Hz,

1H), 3.48 (dd, J 11.3, 6.4 Hz, 1H), 3.31 (s, 3H), 2.62 (dd, 12.4, 6.0 Hz,

1H), 2.30 (t, J 11.8 Hz, 1H), 1.82 ± 1.81 (m, 1H), 0.90 (d, J 6.8 Hz, 3H),

0.0.85 (d, J 6.8 Hz, 3H); ¹³C NMR (125 MHz, MeOH): d 172.6, 151.5,

131.9, 131.2, 125.2, 120.6, 111.2, 90.0, 85.7, 63.1, 60.8, 58.0, 46.1, 30.0, 20.1,

IR (film): nÄ 2954, 2856, 1723, 1472, 1289, 1153, 1080, 1027, 838 cm^ꢀ1

;

ES-MS: m/z: calcd for C₃₇H₄₅N₂O₇SiINa [MH] : 807.19, found: 807.4, 135;

[a]²_D⁵ ꢀ80.4 (c 1.0 in CHCl₃).

5-Trimethylstannyl-pyrroloindoline tert-butyl ester 58: The same proce-

dure for the preparation of 18 was used with the following amounts: Me₆Sn₂

(2.10 mL, 6.30 mmol), 57 (2.50 g,

3.19 mmol),

[Pd(Ph₃P)₄]

(0.37 g,

0.32 mmol) and THF (10 mL). The

crude residue was purified by chro-

matography on silica gel (8:1 hexanes/

Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes)

to give 58 (2.27 g, 87%) as a light yellow foam that was homogeneous by

TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.71 (brs, 1H), 7.51 (d, J

7.9 Hz, 1H), 7.48 (s, 1H), 7.46 ± 7.18 (m, 10H), 5.98 (brs, 1H), 5.50 ± 4.40 (m,

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0059 $ 17.50+.50/0

59

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

4H), 3.92 (brs, 1H), 2.89 (dd, J 12.3, 7.2 Hz, 1H), 2.40 (dd, J 12.4,

7.3 Hz, 1H), 1.60 ± 1.30 (brm, 9H), 0.74 (s, 9H), 0.31 (s, 9H), ꢀ0.32 to

ꢀ 0.50 (m, 6H); ¹³C NMR (125 MHz, CDCl₃): d 170.8, 153.1, 142.7, 137.9,

136.0, 131.8, 131.1, 128.3, 128.2, 128.0, 117.6, 85.4, 81.6, 80.9, 67.3, 59.6, 38.4,

27.8, 25.3, 17.7, ꢀ3.8, ꢀ4.4, ꢀ9.5; IR (film): nÄ 2953, 2930, 1723, 1477, 1419,

Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O) to give 61 (2.41 g, 96%) as

a light pink oil that was homogeneous by TLC analysis: ¹H NMR

(500 MHz, CDCl₃): d 7.81 ± 7.40 (m, 20H), 6.70 (dd, J 15.3, 7.9 Hz,

0.8H), 6.38 (dd, J 15.3, 7.9 Hz, 1.2H), 5.56 ± 5.48 (m, 1.6H), 4.79 (brs,

4H), 4.55 ± 3.98 (m, 5.2H), 3.58 ± 3.53 (m, 1.2H), 2.85 ± 2.80 (m, 0.8H),

2.75 ± 2.68 (m, 0.8H), 2.60 ± 2.45 (m, 2.4H), 1.48 ± 1.43 (m, 18H), 1.00 ± 0.79

(m, 18H), 0.65 ± 0.30 (m, 12H); ¹³C NMR (125 MHz, CDCl₃): d 170.6,

169.9, 154.8, 153.8, 147.5, 147.4, 146.8, 146.7, 144.0, 143.9, 143.8, 143.1, 141.6,

141.2, 141.0, 140.9, 132.5, 132.2, 130.2, 130.0, 129.9, 128.8, 128.4, 128.1, 128.0,

127.7, 127.6, 127.5, 127.3, 127.2, 126.9, 126.8, 126.7, 125.2, 124.9, 124.7, 124.4,

124.3, 121.9, 121.8, 121.7, 119.8, 119.7, 87.8, 87.4, 83.6, 82.7, 81.5, 81.1, 67.6,

65.9, 59.5, 59.3, 46.9, 43.5, 43.2, 27.8, 6.6, 6.5, 5.4, 5.3; IR (film): nÄ 3419,

2954, 1744, 1712, 1481, 1296, 1151, 1109 cm^ꢀ1; ES-MS: m/z: calcd for

1321, 1261, 1107 cm^ꢀ1; ES-MS: m/z: calcd for C₄₀H₅₄N₂O₇SiSnNa [MNa] :

843.26, found: 845.5, 135; [a]²_D⁵ ꢀ108 (c 1.0 in CHCl₃).

Pyrroloindoline tert-butyl ester dimer 59: The same procedure for the

preparation of 19 was used with the following amounts: 58 (2.10 g,

2.56 mmol) in DMF (8.0 mL), 57

(2.00 g,

2.56 mmol),

[Pd₂dba₃]

(0.25 g, 0.26 mmol), Ph₃As (0.16 g,

0.51 mmol), and DMF (7.0 mL). The

crude residue was purified by chro-

C₇₂H₈₆N₄O₁₀Si₂Na [MNa] : 1245.58, found: 1245.9, 135; [a]²_D⁵ ꢀ216 (c

1.0 in CHCl₃).

matography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1

Et₂O/hexanes then 3:1 Et₂O/hexanes) to give 59 (2.77 g, 83%) as a pale

yellow foam that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.78 (brs, 2H), 7.52 (d, J 8.0 Hz, 2H), 7.44 (s, 2H), 7.40 ± 7.21

(m, 20H), 6.05 (brs, 2H), 5.40 ± 4.50 (m, 8H), 3.96 (brs, 2H), 2.91 (dd, J

12.2, 7.2 Hz, 2H), 2.45 (m, 2H), 1.60 ± 1.26 (brm, 18H), 0.77 (s, 18H),

ꢀ0.31 ± ꢀ 0.40 (m, 12H); ¹³C NMR (125 MHz, CDCl₃): d 170.6, 154.4,

153.1, 141.7, 136.9, 136.0, 132.9, 129.4, 128.3, 128.1, 127.7, 122.7, 118.2, 85.3,

81.7, 81.2, 67.4, 59.6, 38.5, 27.8, 25.3, 17.7, ꢀ3.7, ꢀ4.2; IR (film): nÄ 2954,

2856, 1720, 1477, 1289, 1151, 1024 cm^ꢀ1; ES-MS: m/z: calcd for C₇₄H₉₀N₄O₁₄-

Pyrroloindoline allyl ester dimer 63: Neat TESOTf (3.80 mL, 16.4 mmol)

was added to a 08C solution of 61 (1.00 g, 0.82 mmol), 2,6-lutidine

(4.80 mL, 41.0 mmol), and CH₂Cl₂

(6 mL). The resulting solution was

allowed to warm to rt overnight,

recooled to 08C, quenched with satu-

rated aqueous NaHCO₃(50 mL), and

diluted with EtOAc (100 mL). The layers were separated and the organic

layer was washed with aqueous 1m HCl (2 Â 50 mL), brine (1 Â 50 mL),

dried (Na₂SO₄), and concentrated to give crude diacid 62. This residue

residue was azeotroped with toluene (2 Â 100 mL) to remove TESOH and

was used without further purification.

Si₂Na [MNa] : 1337.59, found: 1338.8, 135; [a]²_D⁵ ꢀ124 (c 1.0 in

CHCl₃).

3a-Hydroxypyrroloindoline tert-butyl ester dimer 60: The same procedure

for the preparation of 20 was used with the following amounts: TBAF (1m

in THF, 5.50 mL), 59 (3.60 g,

A solution of DBAD (1.13 g, 4.91 mmol) and THF (10.0 mL) was added to

a 08C solution of the crude residue, allyl alcohol (1.10 mL, 16.4 mmol),

Ph₃P (1.72 g, 6.60 mmol), and THF (20.0 mL). The resulting solution was

maintained at 08C for 2 h and then concentrated. The crude residue was

purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O) to

give 63 (792 mg, 81%) as a light purple oil that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 7.85 ± 7.21 (m, 20H), 6.65 (dd,

J 13.7, 8.0 Hz, 0.8H), 6.28 (dd, J 13.7, 8.1 Hz, 1.2H), 5.90 ± 5.80 (m, 2H),

5.35 ± 5.19 (m, 4H), 4.80 ± 4.30 (m, 8H), 4.19 ± 4.11 (m, 2H), 2.82 ± 2.65 (m,

1.6H), 2.49 ± 2.45 (m, 2.4H), 0.91 ± 0.71 (m, 18H), 0.52 ± 0.24 (m, 12H);

¹³C NMR (125 MHz, CDCl₃): d 171.2, 170.7, 154.9, 154.0, 147.4, 147.3,

146.9, 146.8, 144.0, 143.9, 143.4, 141.7, 141.2, 141.1, 141.0, 132.8, 132.4, 131.7,

131.6, 130.1, 129.7, 128.6, 128.3, 127.9, 127.7, 127.6, 127.5, 127.3, 127.0, 125.0,

124.9, 124.6, 124.4, 122.1, 122.0, 121.8, 120.0, 119.9, 118.7, 118.2, 110.6, 109.7,

88.1, 87.7, 83.8, 82.8, 67.7, 66.2, 65.9, 65.7, 59.0, 58.8, 47.1, 47.0, 43.8, 43.2, 28.1,

6.7, 6.6, 5.7, 5.5; IR (film): nÄ 3419, 2953, 1751, 1711, 1420, 1296, 1154,

2.74 mmol), and THF (20 mL). The

crude residue was purified by chro-

matography on silica gel (8:1 hexanes/

Et₂O then 3:1 hexanes/Et₂O then 1:1

Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to give the

tertiary alcohol (2.78 g, 96%) as a light tan solid that was homogeneous by

TLC analysis: m.p. >2508C (dec); ¹H NMR (500 MHz, CDCl₃): d 7.57

(brs, 2H), 7.47 (s, 2H), 7.41 (d, J 8.3 Hz, 2H), 7.38 ± 7.16 (m, 20H), 6.09 (s,

2H), 5.38 ± 4.60 (m, 8H), 4.08 (brs, 2H), 3.97 (brs, 2H), 2.78 (dd, J 13.0,

7.8 Hz, 2H), 2.40 (dd, J 12.9, 8.8 Hz, 2H), 1.37 (brs, 18H); ¹³C NMR

(125 MHz, CDCl₃): d 171.3, 153.5, 140.5, 135.9, 135.5, 133.0, 129.1, 128.4,

128.3, 128.1, 128.0, 121.5, 117.5, 83.8, 82.1, 67.9, 67.3, 60.0, 37.8, 27.7; IR

(film): nÄ 3409, 2977, 1719, 1478, 1367, 1258, 1153, 1082 cm^ꢀ1; ES-MS: m/z:

calcd for C₆₂H₆₂N₄O₁₄Na [MNa] : 1109.42, found: 1109.7, 135; [a]_D²⁵

ꢀ133 (c 1.0 in CHCl₃).

1110 cm^ꢀ1; ES-MS: m/z: calcd for C₇₀H₇₈N₄O₁₀Si₂Na [MNa] : 1213.52,

A solution of DBU (0.72 mL, 4.80 mmol) and DMF (5.0 mL) was added to

solution of the above alcohol (2.60 g, 2.40 mmol), TESCl (1.40 g,

found: 1213.9, 135; [a]²_D⁵ ꢀ205 (c 1.0 in CHCl₃).

a

9.60 mmol), and DMF (10 mL) at rt. The resulting solution was maintained

at rt for 1 h, and was then diluted with EtOAc (200 mL) and poured into

aqueous 1m HCl (100 mL). The layers were separated and the organic layer

was washed with 1m HCl (2 Â 50 mL), saturated aqueous NaHCO₃(1 Â

50 mL), brine (1 Â 50 mL), dried (MgSO₄), and concentrated. The crude

residue was purified by chromatography on silica gel (8:1 hexanes/Et₂O

then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes) to

give 60 (3.00 g, 95%) as a near colorless oil that was homogeneous by TLC

analysis: ¹H NMR (500 MHz, CDCl₃): d 7.78 (brs, 2H), 7.53 (d, J

8.2 Hz, 2H), 7.48 (s, 2H), 7.35 ± 7.20 (m, 20H), 6.05 (brs, 2H), 5.40 ± 4.60

(m, 8H), 3.95 (brs, 2H), 2.91 (dd, J 12.3, 7.2 Hz, 2H), 2.46 (dd, J 12.3,

10.6 Hz, 2H), 1.60 ± 1.32 (brm, 18H), 0.74 (t, J 7.8 Hz, 18H), 0.40 ± 0.20

(m, 12H); ¹³C NMR (125 MHz, CDCl₃): d 170.6, 153.0, 141.6, 137.7, 136.7,

136.0, 129.4, 128.3, 128.3, 128.1, 122.3, 118.2, 85.1, 81.6, 81.5, 67.4, 59.6, 38.4,

27.8, 6.5, 5.4; IR (film): nÄ 2955, 2876, 1723, 1478, 1321, 1257, 1152,

Pyrroloindoline dimer 64: The same procedure for the preparation of 24

was used with the following amounts: Piperidine (0.21 mL, 2.15 mmol), 63

(640 mg, 0.54 mmol), and CH₃CN

(5.0 mL). The crude residue was pu-

rified by chromatography on silica gel

(8:1 hexanes/Et₂O then 3:1 hexanes/

Et₂O then 1:1 Et₂O/hexanes then 3:1

Et₂O/hexanes then Et₂O then EtOAc) to give 64 (297 mg, 74%) as a near

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.30 (s, 2H), 7.21 (dd, J 8.3, 1.6 Hz, 2H), 6.54 (d, J 8.2 Hz,

2H), 5.84 ± 5.72 (m, 2H), 5.22 (dd, J 17.1, 1.0 Hz, 2H), 5.13 (dd, J 10.4,

0.9 Hz, 2H), 4.95 (s, 2H), 4.60 ± 4.45 (m, 4H), 3.67 (dd, J 9.3, 6.1 Hz, 2H),

2.50 (dd, J 12.2, 6.0 Hz, 2H), 2.38 (dd, J 12.1, 9.8 Hz, 2H), 0.80 (t, J

7.9 Hz, 18H), 0.51 ± 0.15 (m, 12H); ¹³C NMR (125 MHz, CDCl₃): d 173.0,

148.7, 132.4, 131.5, 130.8, 128.1, 122.7, 118.3, 109.9, 91.1, 84.4, 65.4, 59.0,

46.9, 6.6, 5.5; IR (film): nÄ 3357, 2952, 1737, 1618, 1482, 1123, 1086,

1009 cm^ꢀ1; ES-MS: m/z: calcd for

1107 cm^ꢀ1; ES-MS: m/z: calcd for C₇₄H₉₀N₄O₁₄Si₂Na [MNa] : 1337.59,

found: 1337.8, 135; [a]²_D⁵ ꢀ124 (c 1.0 in CHCl₃).

C₄₀H₅₉N₄O₆Si₂[MH] : 741.38,

3a-Triethylsiloxypyrroloindoline tert-butyl ester dimer 61: The same

procedure for the preparation of 22 was used with the following amounts:

60 (2.70 g, 2.06 mmol) and 10% Pd/C (100 mg) in EtOAc (30.0 mL);

pyridine (1.00 mL, 12.3 mmol), the crude residue, Fmoc-HOSu (2.77 g,

8.20 mmol), and CH₂Cl₂(10 mL). The

found: 747.7, 135; [a]²_D⁵ ꢀ49.0 (c

1.0 in CHCl₃).

cis-Hexadepsipeptide dimer (65):

The same procedure for the prepa-

ration of 46 was used with the

following amounts: HATU (509 mg,

1.34 mmol), 64 (250 mg, 0.34 mmol),

crude residue was purified by chroma-

tography on silica gel (8:1 hexanes/

Et₂O then 3:1 hexanes/Et₂O then 1:1

60

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0060 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

45 (1.29 g, 1.34 mmol), HOAt (365 mg, 2.68 mmol), collidine (0.53 mL,

4.02 mmol), and CH₂Cl₂(5.0 mL). The crude residue was purified by

chromatography on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O

then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O then EtOAc) to

give 65 (535 mg, 60%) as a near colorless oil that was homogeneous by

TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.26 ± 6.41 (m, 12H), 5.95 ±

5.58 (m, 6H), 5.41 ± 4.20 (m, 24H), 3.81 ± 3.35 (m, 2H), 2.85 ± 2.50 (m, 8H),

2.40 ± 2.05 (m, 6H), 1.82 ± 1.48 (m, 8H), 1.21 ± 0.65 (m, 96H), 0.59 ± 0.34 (m,

12H), 0.18 ± ꢀ 0.15 (m, 24H); ¹³C NMR (125 MHz, CDCl₃): d 174.0,

172.3, 172.1, 171.7, 171.4, 170.9, 170.8, 169.8, 168.0, 167.8, 166.6, 154.2, 154.1,

153.5, 146.1, 145.5, 131.9, 131.8, 131.5, 131.3, 130.9, 129.6, 129.2, 127.9, 127.6,

125.3, 120.8, 119.0, 118.6, 118.2, 109.6, 109.3, 95.4, 95.3, 95.1, 88.2, 87.9, 86.1,

85.9, 76.4, 75.8, 74.4, 74.3, 68.0, 67.8, 66.1, 65.8, 65.4, 62.0, 59.9, 59.8, 59.5,

59.1, 53.9, 51.0, 49.2, 48.9, 46.6, 45.9, 45.7, 43.5, 43.3, 43.2, 32.4, 31.4, 30.3,

30.1, 29.3, 29.2, 25.9, 25.8, 25.6, 24.4, 24.3, 23.0, 22.2, 21.9, 21.7, 21.5, 19.2,

18.8, 18.7, 18.4, 18.0, 17.9, 17.8, 17.6, 17.5, 17.4, 17.1, 6.6, 6.5, 5.8, 5.6, ꢀ4.2,

ꢀ4.6, ꢀ4.9, ꢀ5.0, ꢀ5.2, ꢀ5.3, ꢀ5.6, ꢀ5.7; IR (film): nÄ 3382, 2932, 1743,

1668, 1510, 1252, 1118, 834 cm^ꢀ1; ES-MS: m/z: calcd for C₁₂₀H₂₀₂Cl₆N₁₄O₂₆-

aqueous NaHCO₃(8 Â 5 mL), brine (8 Â 5 mL), dried (Na₂SO₄), and

concentrated. The crude residue was purified by chromatography on silica

gel (Et₂O then EtOAc then CHCl₃then 1% MeOH/CHCl₃then 2%

MeOH/CHCl₃then 10% MeOH/CHCl₃) to give 68 (76.7 mg, 34%) as a

colorless solid that was homogeneous by TLC analysis. M.p. >3608C (dec);

¹H NMR (500 MHz, CDCl₃): d 7.58 (s, 2H), 7.43 ± 7.41 (m, 4H), 7.28 (d,

J 10.0 Hz, 2H), 7.11 (d, J 10.4 Hz, 2H), 6.79 (d, J 8.3 Hz, 2H), 5.91 (s,

2H), 5.81 (brs, 2H), 5.63 (d, J 8.6 Hz, 2H), 5.42 (d, J 12.3 Hz, 2H),

5.30 ± 5.20 (m, 4H), 5.14 ± 5.12 (m, 4H), 4.98 (d, J 10.5 Hz, 2H), 4.89 (dd,

J 9.9, 3.0 Hz, 2H), 4.50 ± 4.44 (m, 2H), 4.25 ± 4.21 (m, 2H), 3.82 (s, 2H),

3.62 (s, 2H), 3.06 (d, J 13.1 Hz, 2H), 2.85 (t, J 13.5 Hz, 2H), 2.76 (d, J

14.3 Hz, 2H), 2.60 ± 2.50 (m, 2H), 2.48 (d, J 14.8 Hz, 2H), 2.25 ± 2.10 (m,

4H), 1.98 ± 1.90 (m, 2H), 1.72 ± 1.65 (m, 6H), 1.41 ± 1.35 (m, 2H), 1.15 (d,

J 6.5 Hz, 6H), 1.11 (d, J 6.6 Hz, 6H), 1.02 (d, J 6.5 Hz, 6H), 1.00 (d,

J 6.6 Hz, 6H), 0.92 (d, J 5.7 Hz, 6H), 0.87 (d, J 6.7 Hz, 6H), 0.85 (d,

J 6.5 Hz, 6H); ¹³C NMR (125 MHz, CDCl₃): d 174.1, 173.9, 173.4, 173.2,

173.0, 172.3, 146.6, 134.3, 132.2, 128.4, 121.3, 112.6, 90.8, 86.2, 66.7, 60.8, 58.7,

57.1, 54.3, 53.8, 52.6, 49.9, 40.9, 39.4, 30.0, 29.9, 28.6, 25.3, 23.0, 21.0, 19.3,

18.9, 18.3, 17.3, 16.4; IR (film): nÄ 3392, 2238, 2963, 1725, 1671, 1624, 1535,

Si₆Na [MNa] : 2656.16, found: 2656.5, 135; [a]²_D⁵ ꢀ15.4 (c 1.0 in

CHCl₃).

1417, 1246, 1154, 914 cm^ꢀ1

;

ES-MS: m/z: calcd for C₇₂H₁₀₄N₁₄O₂₀Na

[MNa] : 1507.74, found: 1508.2, 135; [a]²_D⁵ ꢀ33.8 (c 0.35 in MeOH).

cis-Hexadepsipeptide carboxylic acid dimer 66: The same procedure for

the preparation of 47 was used with the following amounts: PhSiH₃

Fmoc-Pyrroloindoline 70: Neat pyridine (5.30 mL, 65.2 mmol) was added

(0.18 mL,

1.48 mmol),

65

to a 08C solution of 14 (3.00 g, 10.9 mmol), Fmoc-HOSu (7.30 g,

(650 mg, 0.25 mmol), [Pd(Ph₃P)₄]

(50 mg), and THF (5 mL). The

crude residue was purified by

chromatography on silica gel (8:1

hexanes/Et₂O then 3:1 hexanes/

Et₂O then 1:1 Et₂O/hexanes then

3:1 Et₂O/hexanes then Et₂O then

EtOAc) to give 66 (510 mg, 81%)

as a colorless oil that was homo-

21.7 mmol), and CH₂Cl₂(75 mL).

The resulting solution was allowed

to warm to rt overnight and then

concentrated. The crude residue was

purified by chromatography on silica

gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/hexanes then

3:1 Et₂O/hexanes) to give 70 (3.00 g, 55%) as a colorless solid that was

homogeneous by TLC analysis: m.p. 157 ± 158.58C; ¹H NMR (500 MHz,

CDCl₃): d 7.81 ± 7.10 (m, 10H), 6.80 (t, J 7.4 Hz, 0.30H), 6.75 (t, J

7.4 Hz, 0.70H), 6.63 (d, J 7.9 Hz, 0.30H), 6.36 (d, J 7.9 Hz, 0.70H), 5.60

(s, 0.30H), 4.94 (s, 0.70H), 4.78 (dd, J 10.6, 4.4 Hz, 0.70H), 4.72 (dd, J

10.6, 3.9, 0.70H), 4.57 ± 4.40 (m, 0.60H), 4.38 ± 4.24 (m, 1H), 4.21 ± 4.15 (m,

1H), 3.22 (s, 0.30H), 3.17 (s, 0.70H), 2.63 (dd, J 13.8, 9.0 Hz, 0.35H), 2.54

(dd, J 13.8, 3.2 Hz, 0.35H), 2.35 ± 2.24 (m, 1.3H), 1.48 (s, 3H), 1.47 (s,

6H); ¹³C NMR (125 MHz, CDCl₃): d 172.9, 172.7, 155.0, 154.2, 147.9,

147.5, 144.0, 143.9, 143.8, 143.2, 141.6, 141.2, 141.2, 141.1, 130.2, 129.9, 129.8,

129.5, 127.8, 127.5, 127.3, 127.1, 127.0, 125.3, 125.0, 124.7, 124.4, 123.0, 122.8,

120.0, 119.9, 119.5, 119.2, 110.5, 109.9, 87.7, 86.9, 85.8, 85.1, 82.7, 82.6, 68.1,

66.2, 60.7, 60.6, 47.1, 42.2, 41.3, 27.9, 27.8; IR (film): nÄ 3417, 2978, 1708,

1

geneous by TLC analysis: H NMR (500 MHz, CDCl₃): d 7.41 ± 6.36 (m,

12H), 5.80 ± 5.42 (m, 4H), 5.30 ± 4.20 (m, 16H), 3.87 ± 3.75 (m, 2H), 2.90 ±

2.10 (m, 14H), 1.80 ± 1.41 (m, 8H), 1.20 ± 0.62 (m, 96H), 0.59 ± 0.39 (m,

12H), 0.16 ± ꢀ 0.18 (m, 24H); ¹³C NMR (125 MHz, CDCl₃): d 173.6,

173.1, 172.5, 172.1, 171.7, 171.3, 171.1, 171.0, 170.0, 168.8, 168.1, 155.0, 154.2,

153.6, 145.0, 131.9, 129.5, 129.3, 127.5, 121.1, 109.7, 95.3, 95.1, 95.0, 94.5, 89.5,

88.4, 88.2, 86.3, 83.2, 74.9, 74.6, 74.5, 67.8, 62.2, 60.3, 59.4, 58.9, 54.6, 54.0,

49.2, 48.9, 45.7, 43.4, 34.5, 34.1, 32.6, 32.1, 31.6, 31.2, 30.2, 29.9, 29.5, 29.3,

29.2, 26.0, 25.8, 25.8, 25.6, 25.5, 24.4, 23.2, 23.0, 22.8, 22.4, 21.6, 21.3, 20.8,

19.4, 19.2, 18.9, 18.8, 18.6, 18.3, 18.0, 17.9, 17.8, 17.5, 17.4, 17.1, 16.8, 14.0, 13.9,

6.7, 6.6, 5.9, 5.8, 5.7, 5.6, ꢀ4.9, ꢀ5.0, ꢀ5.1, ꢀ5.7; IR (film): nÄ 3381, 2956,

1612, 1154, 755 cm^ꢀ1; ES-MS: m/z: calcd for C₃₀H₃₀N₂O₅Na [MNa] :

1730, 1667, 1650, 1514, 1253, 1120, 756 cm^ꢀ1; ES-MS: m/z: calcd for

521.21, found: 521.4, 135; [a]²_D⁵ ꢀ111 (c 1.0 in CHCl₃).

C₁₁₄H₁₉₄Cl₆N₁₄O₂₆Si₆Na [MNa] : 2576.09, found: 2576.4, 135; [a]_D²⁵

Pyrroloindoline allyl ester 71: Neat TESOTf (8.30 mL, 36.1 mmol) was

added to a 08C solution of 70 (1.80 g, 3.61 mmol), 2,6-lutidine (8.40 mL,

72.2 mmol), and CH₂Cl₂(12 mL). The

ꢀ28.9 (c 1.0 in CHCl₃).

Himastatin (68): Pb/Cd couple (300 mg) was added to a vigorously stirred

mixture of 66 (385 mg, 0.15 mmol), THF (5.0 mL), and aqueous 1m

NH₄OAc (5.0 mL). After 1.5 h, the

resulting solution was allowed to

warm to rt overnight and then

layers were separated and the aque-

quenched with saturated aqueous

ous layer was extracted with EtOAc

NaHCO₃(100 mL) and diluted with

(4 Â 5 mL). The combined organic

EtOAc (100 mL). The layers were separated and the organic layer was

washed with aqueous 1m HCl (2 Â 50 mL), brine (1 Â 50 mL), dried

(Na₂SO₄), and concentrated.

layers were dried (Na₂SO₄) and

concentrated. This crude amino

acid (330 mg, ꢁ100%) was split

A solution of DBAD (1.60 g, 7.20 mol) and THF (5 mL) was added to a 08C

solution of the crude carboxylic acid, allyl alcohol (2.50 mL, 36.1 mmol),

Ph₃P (1.90 g, 7.20 mmol), and THF (15 mL). The resulting solution was

maintained at 08C for 1.5 h and then concentrated. The crude residue was

purified by chromatography on silica gel (8:1 hexanes/Et₂O then 3:1

hexanes/Et₂O then 1:1 Et₂O/hexanes then 3:1 Et₂O/hexanes then Et₂O) to

give the allyl ester as a colorless oil contaminated by bis-Boc hydrazine.

into two equal batches and used

without further purification.

HATU (110 mg, 0.29 mmol) was added to a solution of the crude amino

acid (ꢁ165 mg), HOAt (79.0 mg, 0.58 mmol), iPr₂NEt (0.15 mL,

0.87 mmol), and DMF (50 mL) at rt. The resulting solution was maintained

at rt for 28 h and then concentrated. The resulting crude residue was

dissolved in EtOAc (10 mL) and washed with aqueous 1m HCl (3 Â 5 mL),

saturated aqueous NaHCO₃(3 Â 5 mL), brine (1 Â 5 mL), dried (Na₂SO₄),

and concentrated. The two batches of cyclic dimer (ꢁ165 mg each) were

recombined at this stage and used without further purification.

Piperidine (1.43 mL, 14.4 mmol) was added to a solution of this semi-

purified residue and CH₃CN (15 mL) at rt. After 1.5 h, the resulting

mixture was concentrated. The crude residue was purified by chromatog-

raphy on silica gel (8:1 hexanes/Et₂O then 3:1 hexanes/Et₂O then 1:1 Et₂O/

hexanes then 3:1 Et₂O/hexanes then Et₂O) to give 71 (603 mg, 46%) as a

colorless oil that was homogeneous by TLC analysis: ¹H NMR (500 MHz,

CDCl₃): d 7.21 (d, J 7.5 Hz, 1H), 7.12 (dt, J 7.8, 1.2 Hz, 1H), 6.76 (dt,

J 7.4, 0.8 Hz, 1H), 6.58 (d, J 7.9 Hz, 1H), 5.92 ± 5.84 (m, 1H), 5.30 (dd,

J 17.2, 1.4 Hz, 1H), 5.22 (dd, J 10.4, 1.2 Hz, 1H), 5.22 (s, 1H), 4.65 ± 4.61

(m, 2H), 4.32 (brs, 1H), 3.71 (dd, J 9.6, 6.1 Hz, 1H), 2.58 (dd, J 12.4,

A solution of TBAF (1m in THF, 5.80 mL) was added to a 08C solution of

the crude residue (ꢁ330 mg), HOAc (1.00 mL, 17.4 mmol), and THF

(3.0 mL). The resulting solution was allowed to warm to rt, maintained at rt

for 55 h, and then diluted with EtOAc (10 mL) and quenched with

saturated aqueous NaHCO₃(10 mL). The layers were separated and the

organic layer was washed with aqueous 1m HCl (8 Â 5 mL), saturated

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0061 $ 17.50+.50/0

61

S. J. Danishefsky and T. M. Kamenecka

FULL PAPER

6.1 Hz, 1H), 2.44 (dd, J 12.6, 9.6 Hz, 1H), 0.82 (t, J 3.5 Hz, 9H), 0.46 ±

0.30 (m, 6H); ¹³C NMR (125 MHz, CDCl₃): d 172.9, 150.0, 131.5, 130.1,

129.6, 124.3, 118.3, 118.2, 109.5, 90.9, 83.9, 65.2, 58.9, 46.7, 6.5, 5.4; IR (film):

nÄ 3349, 2911, 1737, 1611, 1470, 1132, 742 cm^ꢀ1; ES-MS: m/z: calcd for

(25.0 mg, 78%) as a colorless solid that was homogeneous by TLC analysis:

m.p. >2608C (dec); ¹H NMR (500 MHz, CDCl₃): d 7.41 (d, J 4.8 Hz,

1H), 7.33 (d, J 7.4 Hz, 1H), 7.29 (d, J 10.0 Hz, 1H), 7.19 (t, J 7.6 Hz,

1H), 7.10 (d, J 10.4 Hz, 1H), 6.89 (t, J 7.4 Hz, 1H), 6.76 (d, J 7.9 Hz,

1H), 5.85 (s, 1H), 5.80 (brs, 1H), 5.63 (d, J 8.6 Hz, 1H), 5.41 (d, J

12.6 Hz, 1H), 5.20 ± 5.18 (m, 2H), 5.13 ± 5.10 (m, 2H), 4.96 (d, J 10.5 Hz,

1H), 4.89 (dd, J 9.9, 2.9 Hz, 1H), 4.45 ± 4.40 (m, 1H), 4.28 ± 4.23 (m, 1H),

3.81 (s, 1H), 3.61 (s, 1H), 3.06 (d, J 14.0 Hz, 1H), 2.83 (t, J 13.2 Hz,

1H), 2.74 (d, J 14.3 Hz, 1H), 2.60 ± 2.50 (m, 1H), 2.48 (d, J 15.1 Hz,

1H), 2.25 ± 2.10 (m, 2H), 2.00 ± 1.88 (m, 1H), 1.80 ± 1.62 (m, 3H), 1.42 ± 1.35

(m, 1H), 1.15 (d, J 6.5 Hz, 3H), 1.11 (d, J 6.6 Hz, 3H), 1.00 (d, J

6.8 Hz, 6H), 0.91 (d, J 5.6 Hz, 3H), 0.86 (d, J 6.5 Hz, 3H), 0.85 (d, J

6.5 Hz, 3H); ¹³C NMR (125 MHz, CDCl₃): d 174.1, 173.9, 173.4, 173.2,

173.0, 172.3, 147.7, 131.7, 129.8, 123.2, 121.0, 112.3, 90.8, 85.9, 66.6, 60.7, 58.6,

57.2, 54.2, 53.8, 52.6, 49.9, 40.9, 39.4, 30.0, 29.9, 28.6, 25.2, 22.9, 20.9, 19.3,

18.9, 18.2, 17.3, 16.4; IR (film): nÄ 3390, 3328, 2963, 1724, 1673, 1642, 1530,

C₂₀H₃₁N₂O₃Si [MNa] : 375.19, found: 375.3, 135; [a]²_D⁵ ꢀ98.2 (c 1.0 in

CHCl₃).

cis-Hexadepsipeptide monomer 72: The same procedure for the prepara-

tion of 46 was used with the following amounts: HATU (101 mg,

0.27 mmol), 71 (122 mg, 0.13 mmol),

acid 45 (75 mg, 0.065 mmol), HOAt

(72.0 mg,

0.53 mmol),

collidine

(0.11 mL, 0.80 mmol), and CH₂Cl₂

(1.0 mL). The crude residue was pu-

rified by chromatography on silica gel

(8:1 hexanes/Et₂O then 3:1 hexanes/

Et₂O then 1:1 Et₂O/hexanes then 3:1

1250, 1153, 753 cm^ꢀ1; ES-MS: m/z: calcd for C₃₆H₅₃N₇O₁₀Na [MNa] :

Et₂O/hexanes then Et₂O) to give 72

(85.0 mg, 52%) as a colorless oil that

766.38, found: 766.5, 135; [a]²_D⁵ 37.5 (c 0.7 in CHCl₃).

was homogeneous by TLC analysis: ¹H NMR (500 MHz, CDCl₃): d 7.24 ±

6.20 (m, 7H), 5.98 ± 5.80 (m, 1H), 5.65 ± 4.20 (m, 15H), 3.82 ± 3.70 (m, 1H),

2.85 ± 2.60 (m, 4H), 2.38 ± 2.04 (m, 4H), 1.80 ± 1.55 (m, 3H), 1.21 ± 0.60 (m,

48H), 0.45 ± 0.30 (m, 6H), 0.18 ± ꢀ 0.10 (m, 12H); ¹³C NMR (125 MHz,

CDCl₃): d 172.4, 172.2, 171.9, 171.5, 171.0, 170.9, 168.0, 167.9, 154.3, 146.8,

131.6, 129.5, 122.7, 122.5, 119.2, 118.8, 118.2, 117.3, 109.5, 95.4, 88.0, 86.1,

76.5, 74.5, 68.1, 67.9, 66.2, 66.0, 65.8, 65.6, 62.1, 60.1, 59.9, 59.6, 59.2, 55.8,

54.1, 54.0, 51.1, 49.3, 49.0, 46.1, 45.9, 43.5, 32.5, 31.5, 30.4, 30.2, 29.5, 29.3,

28.1, 26.0, 25.9, 25.7, 25.5, 24.6, 24.5, 23.2, 22.8, 22.3, 22.0, 21.8, 21.7, 19.3,

19.0, 18.8, 18.2, 18.0, 17.9, 17.7, 17.6, 17.2, 15.2, ꢀ4.2, ꢀ4.8, ꢀ4.9, ꢀ5.0, ꢀ5.2,

ꢀ5.6; IR (film): nÄ 3382, 2932, 1738, 1671, 1112, 834, 753 cm^ꢀ1; ES-MS:

Acknowledgement

This work was supported by the National Institutes of Health (Grant

Number: CA28824). T.M.K. gratefully acknowledges the NIH for post-

doctoral fellowship support (Grant Number: AI09355). We thank Bristol

Myers Squibb for providing us with an authentic specimen of himastatin.

We thank Dr. George Sukenick and the NMR Core Facility Laboratory,

Sloan-Kettering Institute for Cancer Research, for mass spectral and NMR

analyses.

m/z: calcd for C₆₀H₁₀₂Cl₃N₇O₁₃Si₃Na [MNa] : 1340.58, found: 1340.8, 135;

[a]²_D⁵ ꢀ31.0 (c 0.7 in CHCl₃).

cis-Hexadepsipeptide carboxylic acid monomer 73: The same procedure

for the preparation of 47 was used with the following amounts: PhSiH₃

(15.0 mL, 0.12 mmol), 72 (80.0 mg,

[1] a) Y.-T. Tan, D. J. Tillett, I. A. McKay, Mol. Med. Today 2000, 6, 309;

b) D. Guillemot, Curr. Opin. Microbiol. 1999, 2, 494; c) J. A.

Heinemann, Drug Discovery Today 1999, 4, 72.

[2] a) K. S. Lam, G. A. Hesler, J. M. Mattel, S. W. Mamber, S. Forenza, J.

Antibiot. 1990, 43, 956; b) J. E. Leet, D. R. Schroeder, B. S. Krishnan,

J. A. Matson, J. Antibiot. 1990, 961; c) J. E. Leet, D. R. Schroeder, J.

Golik, J. A. Matson, T. W. Doyle, K. S. Lam, S. E. Hill, M. S. Lee, J. L.

Whitney, B. S. Krishnan, J. Antibiot. 1996, 49, 299.

[3] a) S. P. Marsden, K. M. Depew, S. J. Danishefsky, J. Am. Chem. Soc.

1994, 116, 11143; b) J. M. Schkeryantz, J. C. G. Woo, S. J. Danishefsky,

J. Am. Chem. Soc. 1995, 117, 7025.

[4] For a preliminary communication of these results, see T. M. Kame-

necka, S. J. Danishefsky, Angew. Chem. 1998, 110, 3164; Angew.

Chem. Int. Ed. 1998, 37, 2993; Angew. Chem. 1998, 110, 3166; Angew.

Chem. Int. Ed. 1998, 37, 2995.

0.061 mmol), [Pd(Ph₃P)₄] (10 mg),

and THF (1.50 mL). The crude resi-

due was purified by chromatography

on silica gel (3:1 hexanes/Et₂O then

1:1 Et₂O/hexanes then 3:1 Et₂O/hex-

anes then Et₂O) to give 73 (71.1 mg,

92%) as

a colorless oil that was

homogeneous by TLC analysis:

¹H NMR (500 MHz, CDCl₃): d

7.40 ± 6.40 (m, 7H), 5.78 ± 5.48 (m, 2H), 5.10 ± 4.05 (m, 8H), 3.91 ± 3.74

(m, 1H), 2.95 ± 2.51 (m, 5H), 2.40 ± 2.08 (m, 3H), 1.85 ± 1.42 (m, 3H), 1.20 ±

0.75 (m, 48H), 0.58 ± 0.39 (m, 6H), 0.32 ± ꢀ 0.11 (m, 12H); ¹³C NMR

(125 MHz, CDCl₃): d 173.8, 173.2, 172.4, 172.2, 172.1, 172.0, 171.8, 171.5,

171.4, 171.1, 170.8, 169.9, 169.0, 169.2, 168.2, 134.2, 131.5, 130.5, 130.4, 130.2,

129.7, 129.0, 127.7, 124.0, 123.5, 122.8, 121.1, 119.5, 118.3, 112.7, 110.6, 109.6,

95.4, 95.1, 94.6, 89.6, 88.1, 86.4, 83.4, 83.0, 75.0, 74.9, 74.6, 71.0, 69.7, 68.1,

67.9, 65.8, 62.3, 62.2, 60.7, 59.9, 59.7, 59.4, 59.2, 58.2, 55.3, 54.7, 54.2, 54.0,

53.4, 52.0, 51.3, 49.4, 49.2, 45.9, 43.3, 43.1, 42.0, 41.8, 39.5, 32.2, 31.6, 30.6,

30.3, 29.6, 29.3, 29.2, 28.1, 26.2, 26.1, 26.0, 25.9, 25.6, 25.5, 24.7, 23.1, 23.0,

21.8, 21.0, 19.3, 19.1, 19.0, 18.8, 18.3, 18.2, 18.1, 18.0, 17.7, 17.5, 17.3, 16.8, 6.7,

5.9, 5.8, 5.7, 5.5, ꢀ4.7, ꢀ4.8, ꢀ4.9, ꢀ5.1, ꢀ5.4, ꢀ5.6; IR (film): nÄ 3380,

[5] S. J. Danishefsky, Tetrahedron 1997, 53, 8689.

[6] a) C. H. Hassall, R. B. Morton, Y. Ogihara, W. A. Thomas, J. Chem.

Soc. D 1969, 18, 1079; b) K. Bevan, J. S. Davies, M. J. Hall, C. H.

Hassall, R. B. Morton, D. A. S. Phillips, Y. Ogihara, W. A. Thomas,

Experientia 1970, 26, 122.

[7] M. Ohno, T. F. Spande, B. Witkop, J. Am. Chem. Soc. 1968, 90, 6521.

[8] a) A. J. Robinson, P. B. Wyatt, Tetrahedron, 1993, 49, 11329; b) H. B.

Arzeno, D. S. Kemp, Synthesis 1988, 32.

[9] Citations to reagents described by acronyms can be found in the

Experimental Section.

2956, 1732, 1668, 1254, 1116, 835 cm^ꢀ1

; ES-MS: m/z: calcd for

C₅₇H₉₈Cl₃N₇O₁₃Si₃Na [MNa] : 1300.55, found: 1300.7, 135; [a]²_D⁵ ꢀ39.3

[10] M. Nakagawa, S. Kato, S. Kataoka, S. Kodato, H. Watanabe, H.

Okajima, T. Hino, B. Witkop, Chem. Pharm. Bull. 1981, 29, 1013.

[11] a) W. E. Savige, Aust. J. Chem. 1975, 28, 2275; b) M. Nakagawa, T.

Hino, Heterocycles 1977, 6, 1675; c) M. Nakagawa, S. Kato, S. Kataoka,

T. Hino, J. Am. Chem. Soc. 1979, 101, 3136.

(c 0.35 in CHCl₃).

cis-Cyclic monomer 74: The same procedure for the preparation of 68 was

used with the following amounts: Pb/Cd couple (100 mg), 73 (55.0 mg,

0.043 mmol), THF (3 mL), and aqueous 1m NH₄OAc (3 mL); HATU

(24.0 mg, 0.064), the crude amino acid (47.0 mg), HOAt (17.0 mg,

[12] J. K. Stille, Angew. Chem. 1986, 98, 504; Angew. Chem. Int. Ed. Engl.

1986, 25, 508.

0.13 mmol),

iPr₂NEt

(22.0 mL,

0.13 mmol), and DMF (26 mL); TBAF

(1m in THF, 0.84 mL), the crude cyclic

monomer (ꢁ50 mg), HOAc (0.14 mL,

0.24 mmol), and THF (3.0 mL). The

crude residue was purified by chroma-

tography on silica gel (3:1 hexanes/

Et₂O then 1:1 Et₂O/hexanes then 3:1

Et₂O/hexanes then Et₂O) to give 74

[13] For a preliminary account, see K. M. Depew, T. M. Kamenecka, S. J.

Danishefsky, Tetrahedron Lett. 2000, 41, 289.

[14] D. A. Evans, T. C. Britton, R. L. Dorow, J. F. Dellaria, J. Am. Chem.

Soc. 1986, 108, 6395.

[15] a) E. J. Corey, R. L. Danheiser, Tetrahedron Lett. 1973, 45, 4477;

b) E. J. Corey, E. J. Trybulski, L. S. Melvin, Jr., K. C. Nicolaou, J. A.

Secrist, R. Lett, P. W. Sheldrake, J. R. Falck, D. J. Brunelle, M. F.

Haslanger, S. Kim, S.-e. Yoo, J. Am. Chem. Soc. 1978, 100, 4618.

62

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0062 $ 17.50+.50/0

Chem. Eur. J. 2001, 7, No. 1

Himastatin

41±63

[16] a) Y. Ohfune, K. Hori, M. Sakaitani, Tetrahedron Lett. 1986, 50, 6079;

b) S. Matsunaga, N. Fusetani, K. Hashimoto, M. Walchli, J. Am. Chem.

Soc. 1989, 111, 2582; c) C. Hutton, J. M. White, Tetrahedron Lett. 1997,

38, 1643.

[17] a) M. R. Ghadiri, R. J. Granja, R. A. Milligan, D. E. McRee, N.

Khazanovich, Nature 1993, 366, 324; b) H. S. Kim, J. D. Hartgerink,

M. R. Ghadiri, J. Am. Chem. Soc. 1998, 120, 4417; c) J. D. Hartgerink,

T. D. Clark, M. R. Ghadiri, Chem. Eur. J. 1998, 4, 1367.

[18] Biological data to be reported elsewhere.

[19] K. Umezawa, Y. Ikeda, Y. Uchihata, H. Naganawa, S. Kondo, J. Org.

Chem. 2000, 65, 459 ± 463.

Received: September 15, 2000 [F2733]

Chem. Eur. J. 2001, 7, No. 1

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

0947-6539/01/0701-0063 $ 17.50+.50/0

63

Article Doi

Discovery through total synthesis: A retrospective on the himastatin problem

DOI: 10.1002/1521-3765(20010105)7:1<41::AID-CHEM41>3.0.CO;2-D

Source and publish data:

Authors:

Article abstract of DOI:10.1002/1521-3765(20010105)7:1<41::AID-CHEM41>3.0.CO;2-D

Full text of DOI:10.1002/1521-3765(20010105)7:1<41::AID-CHEM41>3.0.CO;2-D

Products guided by the article

R&D Labs maybe for 329786-94-9

Relevant to this article

Hot Product