The total synthesis of dynemicin A leading to development of a fully contained bioreductively activated enediyne prodrug

J. Am. Chem. Soc. 1996, 118, 9509-9525

9509

The Total Synthesis of Dynemicin A Leading to Development

of a Fully Contained Bioreductively Activated Enediyne

Prodrug

Matthew D. Shair,^†,‡,§,Tae Young Yoon,^†,§Karoline K. Mosny,^§T. C. Chou,^|and

Samuel J. Danishefsky*^,†,‡,§

Contribution from the Department of Chemistry, Yale UniVersity, New HaVen, Connecticut 06511,

the Laboratories for Bio-Organic Chemistry and Biochemical Pharmacology, The Sloan-Kettering

Institute for Cancer Research, Memorial Sloan-Kettering Cancer Center, 1275 York AVenue,

New York, New York 10021, and the Department of Chemistry, Columbia UniVersity,

New York, New York 10027

ReceiVed January 4, 1996^X

Abstract: The title compound has been synthesized as its racemate in 33 steps. An intramolecular Diels-Alder

reaction (see Scheme 5, 24 f 25) was used to provide control over the eventual cis C₄-C₇relationship. The installation

of another cis related ethynyl group at C₂arose from transformation 40 f 42 (see Scheme 8) whose directionality

is governed by a benzophenone ketal functioning as a temporary steric control unit. Closure of the enediyne unit

was accomplished on a trimethylsilylethoxycarbonyl (TEOC) protected dihydroquinoline derivative. It involved

use of a novel bis-iodoalkyne/distannylethylene interpolative coupling transformation (61 + 58 f 63, Scheme 12).

In the terminal phase of the synthesis, a novel iminoquinone ketal 74 (Scheme 15) was condensed with homophthalic

anhydride derivative 78 (Scheme 16) as indicated in Scheme 17. The final deprotection involved cleavage of a

methoxymethyl ester and two methoxymethyl phenol ethers. From this work, there arose the concept and demonstration

of p-quinone monoimines 82 and 93 (Scheme 18), as bioreductively activated enediyne prodrugs.

Introduction

itself is attenuated by concerns surrounding its difficult acces-

sibility as well as its insolubility and lability.

In 1989, Konishi and associates disclosed the structure of

dynemicin A (1)^1,2which was then the newest member of the

enediyne family of antibiotics.³This metabolite of Micromono-

spora chersina demonstrated high levels of in vitro antitumor

activity comparable to those which had been registered for two

other enediynes, calicheamicin (2)⁴and esperamicin (3).^4d,5In

addition, dynemicin A was described to prolong the life span

of mice which had been inoculated with leukemia cell lines.¹

Unfortunately, enthusiasm for the clinical use of dynemicin A

During early biological studies, it was discovered that

dynemicin effects single and double stranded DNA cleavage.^6-8

This cleavage is enhanced by the addition of various cofactors

including thiols, reducing agents, and even visible light.⁹The

DNA cuts occasioned by 1 were not nearly as sequence specific

as those observed with calicheamicin or esperamicin.^4,5Unlike

these drugs, dynemicin lacks a carbohydrate domain to provide

high specificity in its DNA contacts. Although the an-

thraquinone section of dynemicin may well furnish drug-DNA

contact through intercalation, there seems to be little or no

sequence specificity associated with this effect.¹⁰

An important proposal concerning the mode of action of

dynemicin A was advanced by Semmelhack (Scheme 2).¹¹He

envisioned that a sequence of transformations would be initiated

by reduction of the anthraquinone system (1 f 4). No longer

delocalized into the electron reducing quinone system, the “lone

pair” nitrogen atom of 4 is now available to assist in the opening

of the epoxide linkage thereby generating quinone methide 5.

Following addition of a nucleophile to 5, a conformational

^†Yale University.

^‡Department of Chemistry, Havemeyer Hall, Columbia University, New

York, NY, 10027. Telefax: Int. code + (212) 854-7142.

^§Sloan-Kettering Institute for Cancer Research, Laboratory for Bioor-

ganic Chemistry, 1275 York Avenue, Box 106, New York, NY 10021.

Telfax: Int. code + (212) 772-8691.

Present Address: Department of Chemistry, Harvard University,

Cambridge, MA 02138.

^|Biochemical Pharmacology, The Sloan-Kettering Institute for Cancer

Research.

^XAbstract published in AdVance ACS Abstracts, August 15, 1996.

(1) Konishi, M.; Ohkuma, H.; Matsumoto, K.; Tsuno, T.; Kamei, H.;

Miyaki, T.; Oki, T.; Kawaguchi, H.; VanDuyne, G. D.; Clardy, J. J. Antibiot.

1989, 42, 1449.

(5) (a) Golik, J.; Clardy, J.; Dubay, G.; Groenwold, G.; Kawaguchi, H.;

Konishi, M.; Krishnan, B.; Ohkum, H.; Saitoh, K.; Doyle, T. W. J. Am.

Chem. Soc. 1987, 109, 3461. (b) Golik, J.; Clardy, J.; Dubay, G.; Groenwold,

G.; Kawaguchi, H.; Konishi, M.; Krishnan, B.; Ohkum, H.; Saitoh, K.;

Doyle, T. W. J. Am. Chem. Soc. 1987, 109, 3462.

(6) Sugiura, Y.; Shiraki, T.; Konishi, M.; Oki, T. Proc. Natl. Acad. Sci.

U. S. A. 1990, 87, 3831.

(7) Langley, D. R.; Doyle, T. W.; Beveridge, D. L. J. Am. Chem. Soc.

1991, 113, 4395.

(8) Wender, P. A.; Kelly, R. C.; Beckham, S.; Miller, B. L. Proc. Natl.

Acad. Sci. U.S.A. 1991, 88, 8835.

(9) Sugiura, Y.; Shiraki, T. Biochemistry 1990, 29, 9795.

(10) For insight into the dynamics of dynemicin-DNA interactions, see:

Myers, A. G.; Cohen, S. B.; Tom, N. J.; Madar, D. J.; Fraley, M. E. J. Am.

Chem. Soc. 1995, 117, 7574.

(11) Semmelhack, M. F.; Gallagher, J.; Cohen, D. Tetrahedron Lett. 1990,

31, 1521.

(2) Konishi, M.; Ohkuma, H.; Tsuno, T.; Oki, T.; VanDuyne, G. D.;

Clardy, J. J. Am. Chem. Soc. 1990, 112, 3715.

(3) For a personalized retrospective treatment of the biology and

chemistry of the enediyne antibiotics, see: (a) Danishefsky, S. J.; Shair,

M. D. J. Org. Chem., 1996, 61, 16. (b) See, also: Yoon, T. Y. Ph.D. Thesis,

Yale University, 1994. (b) Enediyne Antibiotics as Antitumor Agents; Doyle,

T. W., Borders, D. B., Eds.; Marcel-Decker: New York, 1994. (c) Nicolaou,

K. C.; Dai, W.-M. Angew. Chem., Int. Ed. Engl. 1991, 30, 1387.

(4) (a) Lee, M. D.; Dunne, T. S.; Siegel, M. M.; Chang, C. C.; Morton,

G. O.; Borders, D. B. J. Am. Chem. Soc. 1987, 109, 3464. (b) Lee, M. D.;

Dunne, T. S.; Siegel, M. M.; Chang, C. C.; Morton, G. O.; Borders, D. B.

J. Am. Chem. Soc. 1987, 109, 3466. (c) Zein, N.; Sinha, A.; McGahren, W.

J.; Ellestad, G. A. Science 1988, 240, 1198. (d) Casazza, A. M.; Kelley, S.

L. Biological Properties of Esperamicin and Other Enediyne Antibiotics.

In Enediyne Antibiotics as Antitumor Agents; Doyle, T. W., Borders, D.

B., Eds., Marcel-Dekker: New York, 1994; pp 283-299.

9510 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

Scheme 1

was the challenge associated with a total synthesis of dynemicin

A. It seemed likely that this would be a challenging goal. The

solution would predictably be instructive to the strategy and

practice of organic synthesis.

In addition, another goal was undertaken. We hoped to

generate for evaluation, functional but structurally less demand-

ing, equivalents of dynemicin. We were of course sensitive to

the Semmelhack hypothesis, wherein the active DNA cleaving

(and presumably cytotoxic form) of dynemicin is incorporated

in structure 4. It is this moiety in which epoxide solvolysis is

triggered, thereby initiating a cascade which leads to eventual

cytotoxicity. We wondered about the possibility of generating,

in ViVo, a truncated version of 4 which would maintain the

capacity for “intramolecularly assisted epoxide solvolysis,” to

trigger diyl formation. In our construct, we hoped to delete

much of the molecular complexity associated with 4, while

retaining its capacity for DNA cleavage. One possibility for

consolidation would be the napthalenediol domain. It seemed

possible that this section is per se irrelevant to the mode of

action, and that recourse to its precursor anthraquinone in the

natural product arises from its being a biosynthetically accessible

triggering device for Micromonospora cherina to employ.

Another candidate area for molecular simplification would be

the complex functionality found in ring E. Here are encountered

a synthetically awkward combination of secondary methyl and

carboxyl groups, the latter conjugated to an enol ether. Perhaps

such implements could be waived in a quest for a simpler

analog, where function is emphasized over optional structural

frills.

restructuring of the AB ring system of 6 causes the distance

between the terminii of the transannular diyne to contract, hence

favoring the Bergman type cyclization wherein electronic

reorganization leads to 1,4 diradical, 7. This diyl⁶as well as

related diyls from enediynes 2 and 3 have been implicated as

the causative high energy intermediates in effecting single and

double stranded breaks in deoxyribonucleic acid targets (via

hydrogen atom abstraction).^4,5

Owing, no doubt, to the unique mode of action of dynemicin

A and to its high levels of antitumor activity, not to speak of

its novel structure, interest in its synthesis began to grow.¹²

Concurrently, a variety of strategies were adopted to create

synthetically accessible congeners which might improve on the

antitumor performance of the drug.^12a,bThe first accomplish-

ments aimed at a dynemicin total synthesis were disclosed by

Schreiber and associates.¹³These breakthroughs culminated in

a route to various methylated versions of dynemicin.¹⁴In early

1995, full success was realized by Myers and associates in their

elegant total synthesis of (+)-dynemicin A.¹⁵Early demonstra-

tions of dynemicin inspired fully synthetic DNA cleaving agents

were provided by Nicolaou.^12f

Thus, we sought a hypothetical entity 9 (Scheme 3), of as

yet unspecified character, which would, on suitable in ViVo

actuation, lead to a structure of the type 10, to initiate the

cytotoxic cascade. Other laboratories, notably such as that of

Nicolaou and associates,^13a,bconceived of solutions to this

problem in terms of an in Vitro cleavable urethane blocking

group leading to active agent 10. We were drawn to the

possibility of a more compact type of entity, closely related to

known types of drugs which might, on suitable bio-priming,

lead to 10 without concurrently ejecting debris of problematic

biological agendas. In other words, we sought a “fully

contained” prodrug for a system of the type 10.

In defining system 9 with greater specificity, we were not

unmindful of the presumed mode of action of the mitomycins.¹⁶

Here a cascade is initiated by reduction of a quinone (cf.

structure 12). This reduction event initiates a sequence leading

to the active DNA damaging agent aziridinomitosene 13. The

mitomycin logic led us to propose recourse to a quinone imine

as the self-contained, bioreductively activatable type of prodrug.

Eventually, considerations of amenability to gram scale synthesis

led us to identify quinone imine system type 14 as a prodrug.

However, it was through intermediates in our total synthesis

exercise that we first prepared the more complicated quinone

imine (compound 82, Vide infra). Its promise, as established

by in Vitro and in ViVo experiments,^17a,18,19led us back to the

still simpler target 14. Hence, it is with the total synthesis of

dynemicin²⁰that our account begins.

Our laboratory began work in the dynemicin area in 1990.

Two considerations drove our program. A very large impetus

(12) For synthesis of various dynemicin A model systems, see: (a)

Nicolaou, K. C.; Dai, W.-M. J. Am. Chem. Soc. 1992, 114, 8908 and

references cited therein. (b) Wender, P. A.; Zercher, C. K.; Beckham, S.;

Haubold, E.-M. J. Org. Chem. 1993, 58, 5867 and references cited therein.

(c) Nishigawa, T.; Isobe, M.; Goto, T. Synlett 1991, 393. (d) Magnus, P.;

Fortt, S. M. J. Chem. Soc., Chem. Commun. 1991, 544. (e) Takahashi, T.;

Sakamoto, Y.; Yamada, H.; Usui, S.; Fukazawa, Y. Angew. Chem., Int.

Ed. Engl. 1995, 34, 1345. (f) Nicolaou, K. C.; Dai, W.-M.; Tsay, S.-C.;

Estevez, V. A.; Wrasildo, W. Science 1992, 256, 1172.

(13) (a) Porco, J. A., Jr.; Schoenen, F. J.; Stout, T. J.; Clardy, J.;

Schreiber, S. L. J. Am. Chem. Soc. 1990, 112, 7410. (b) Wood, J. L.; Porco,

J. A., Jr.; Taunton, J.; Lee, A. Y.; Clardy, J.; Schreiber, S. L. J. Am. Chem.

Soc. 1992, 114, 5898.

(16) For a recent review of the chemsitry and biomechanistics of the

mitomycins, see: Schkerynatz, J.; Danishefsky, S. J. Synlett #5, 1995, 475.

(17) (a)Shair, M. D.; Yoon, T. Y.; Chou, D.; Danishefsky, S. J. Angew.

Chem., Int. Ed. Engl. 1994, 33, 2477. (b) Shair, M. D. Ph.D. Thesis,

Columbia University, 1995.

(18) Shair, M. D.; Yoon, T. Y.; Chou, D.; Danishefsky, S. J. Patent

Application 08/347,952.

(14) Taunton, J.; Wood, J. L.; Schreiber, S. L. J. Am. Chem. Soc. 1993,

115, 10378.

(15) Myers, A. G.; Fraley, M. E.; Tom. N. J.; Cohen, S. B.; Madar, D.

J. Chem. Biol. 1995, 2, 33.

(19) Chou, D.; Shair, M. D.; Yoon, T. Y.; Zheng, Y. H.; Danishefsky,

S. J. Proc. Am. Assoc. Cancer Res. 1995, 36: 384.

(20) Shair, M. D.; Yoon, T. Y.; Danishefsky, S. J. Angew. Chem., Int.

Ed. Engl. 1995, 34, 1721.

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9511

Scheme 2

Scheme 3

Scheme 4

Synthetic Strategy

bond and an R-configured epoxide. Similarly not formulated

were the functions R₁and R₂whose nature goes to the question

as to the degree to which implements for construction of the

naphthoquinone domain are to be incorporated prior to fashion-

ing the cyclic enediyne.

A second approach (see path II), which also postponed

definition of these critical issues, contemplated the interpolation

of a two carbon ethylene moiety unit into a suitable (syn)

configured diyne construct (see structure 19). Here again, on-

line realities, rather than overarching doctrine would dictate the

way in which the diyne functionality would be organized (see

substructure unit W in structure 19). Similarly, the specifics

by which the ethylene unit (Y) would be presented for

interpolation as well as the staging for introduction of the

epoxide (see substructure Z) were “negotiable”.

Though paths I and II represented, at this stage, rather vaguely

specified perceptions, it was already discernible that the latter

would be the more demanding in terms of the stereochemical

relationships which would be in need of implementation. While

the first program required suitable provisions for the cis

connectivity of the C₄methyl and C₇enediyne functions to be

provided (see structure 18), the C₂stereochemistry would be

We considered many prospective solutions for the installation

of the enediyne bridge needed to reach dynemicin. We hoped

that the challenge posed by the intriguing substructure enediyne

system, and its surrounding functionality, would provide a fertile

testing ground for the development of novel synthetic methodol-

ogy. Two general approaches (Scheme 4) emerged among the

many which were up for consideration. The first approach (see

path I) focused on an intramolecular Reissert²¹coupling of an

enediyne anion (or equivalent thereof) to some type of activated

iminium construct (see structure 18). At this level of the thought

process, the nature of the functionality appended the A and B

regions corresponding to the anthraquinone and vinylogous

carbonate moieties of the goal structure were left unspecified.

Correspondingly, the stage at which the epoxide linkage would

be introduced was to be determined through the unfolding of

events rather than via adherence to an apriori blueprint. Hence,

the expression of Z in structure 18 represents uncertainty on

our part at the planning level between a tetrasubstituted double

(21) Popp, F. D. Chem. Heterocycl. Comp. 1982, 32, 353 and references

cited therein.

9512 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

fixed as a consequence of the cyclization. By contrast, path II

required (see structure 19) the fashioning of a cis relationship

of the C₂, C₄, and C₇substitutents before cyclization could take

place.

Scheme 5^a

It seemed likely that the study of both propositions (path I

and II) might be possible from a common synthetic precursor

of the type 17 (W unspecified). Such a construct could provide

an excellent point of divergence for both plans while maintaining

flexibility as to the issues raised above. It was further

contemplated that 17 might arise from ammonolysis of a formal

quinone aldehyde of the type 16. The latter might conceivably

become available through oxidative disconnection of generic

system 15, via either of two modalities (see alternate dotted

line cleavage sites in structure 15). It is with the pursuit of

such a plan, that our journey began.

Results and Discussion

We set as a more specific goal, a tricyclic system corre-

sponding to 17 where R₁and R₂are hydrogens. We had in

mind several ideas for fashioning the symmetrical dihydrox-

ynaphthaquinone moiety which, if successful, would not require

prebuilt “handles” in the benzene ring. The synthesis com-

menced with alkylation of commercially available benzaldehyde

20, with sorbyl bromide as shown in Scheme 5. Two carbon

homologation of 21 using a Horner-Emmons procedure

provided the ethyl ester 22 which was subsequently converted

to triene aldehyde 24 via allylic alcohol 23. Upon exposure of

24 to the action of ZnCl₂at room temperature (Scheme 5), an

endo selective intramolecular Diels-Alder reaction, leading to

compound 26 took place in 60% yield. Alternatively, the

Diels-Alder reaction could also be conducted by heating 24 in

refluxing benzene. A 3:1 ratio of endo:exo adducts (26 and

25) was obtained under the thermal (uncatalyzed) conditions.

Although the endo:exo ratio was per se of little consequence

with respect to the stereochemical relationship of the emerging

C₄and C₇centers (see starred carbons), its importance arose

from the failure of the exo product 25 to survive the ensuing

transformation. Accordingly, Lewis acid catalysis was used in

practice. These conditions provided the endo product more

cleanly albeit in a lower yield than did the strictly thermal

protocol.

Treatment of endo adduct 26 with ceric ammonium nitrate

(CAN) gave rise to the quinone 27 in high yield. Several critical

transformations had been accomplished in this experiment. The

para dioxygenated aromatic ring had suffered oxidation to a

quinone with presumed concomitant dispatch of methanol in

the usual way. In addition, the hydroxymethyl function,

unveiled in the same oxidation, engaged the proximal aldehyde

in the form of a hemiacetal linkage. The storage of the putative

aldehyde in this protective arrangement stabilized the product

toward the oxidative conditions. For instance, the exo adduct

(25), in which hemiacetal protection is not accessible, was

substantially destroyed through the CAN protocol.

^a(a) sorbol bromide, K₂CO₃, acetone, 96% yield; (b) (EtO)₂-

P(O)CH₂CO₂Et, NaH, THF, 95% yield; (c) DIBAH, CH₂Cl₂, 99% yield;

(d) (COCl)2, DMSO, Et₃N, CH₂Cl₂, -78 °C; 91% yield; (e) ZnCl₂,

CH₂Cl₂, 25 °C, 60% yield of 26; or PhH, 80 °C, 3:1 26/25; (f)

Ce(NH₄)₂(NO₃)₆, MeCN, H₂O, 90% yield; (g) NH₄OAc, HOAc, 100

°C, 89% yield; (h) TBSCl, IMID, CH₂Cl₂, 98% yield.

A brief account of our explorations to proceed to dynemicin

via path I will be described first.²³Since the route turned out

to be unsuccessful, we present here only the highlights of these

pursuits. Fuller accounts of the many initiatives to reduce this

intramolecular Reissert pathway to practice have been provided

elsewhere.^3a,bAlthough this sequence did not lead to a total

synthesis of dynemicin A, information garnered along this

excursion was crucial to the success of path II and to the

eventual total synthesis of dynemicin.

From compound 29, we reached probe substrates 30 and 31

(Scheme 6).^3a,bUnfortunately, attempted cyclization of 30 with

LHMDS led primarily to elimination product 32. Furthermore,

attempted Reissert-like cyclization of 30 utilizing isopropyl-

magnesium bromide and methylchloroformate afforded prima-

rily urethan 33. These results suggested that we had in our

planning exercise overlooked a serious kinetic acidity problem

associated with the C₇-H bond. Given these findings, we

sought to incorporate the C₇-C₈epoxide early on. Pursuant

to that reasoning, compound 29 was converted to the surprisingly

stable oxirane 31.²⁴Unfortunately, several attempts to achieve

palladium mediated elongation (“eneynylation”) were unsuc-

cessful. Apparently, the vinyloxirane moiety of 31 had served

as an insertion site to the palladium based catalyst.

Exposure of compound 27 to the action of ammonium acetate

in acetic acid at 100 °C afforded 28 and, following bis-silylation,

29. Thus, was the desired quinoline structure constructed. It

will not escape the notice of the reader that the doctrine of

suprafaciality in the Diels-Alder reaction (cf. 24 f 26) had

indeed been tapped to dictate the required syn stereochemical

relationship between C₄and C₇of quinolines 28 and 29

(corresponding to the generalized target compound 17, shown

in Scheme 4).²²

We then attempted to study the cyclization in the context of

pentacyclic substrates. Toward this end, we reached substrates

34 and 35. In the case of 34, in which the B ring was

nonquinoidal, we were unable to achieve a Reissert reaction.

We attribute this impasse to difficulties associated with alkoxy-

carbonylation of the quinoline type nitrogen, hindered as it is

(23) Taken in part from the following: Yoon, T. Y. Ph.D. Thesis, Yale

University, 1994.

(22) Yoon, T. Y.; Shair, M. D.; Danishefsky, S. J.; Schulte, G. K. J.

Org. Chem. 1994, 59, 3752.

(24) Yoon, T. Y.; Shair, M. D.; Danishefsky, S. J. Tetrahedron Lett.

1994, 35, 6259.

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9513

Scheme 6^a

Scheme 7

Scheme 8^a

^a(a) (i) Ph₂C(OMe)₂, H₂SO₄, CH₂Cl₂, 40 °C, (ii) TBSCl, imidazole

83% yield (i and ii); (b) (triisopropylsilyl)ethynyl magnesium bromide,

ClCO₂R, THF, -20 °C, 80% yield, R ) Me (41) or allyl (42).

R-disposed vicinal diol at carbons 5 and 6 of compound 37 in

a creative way. The hope was to engage these alcohols in a

protectiVe deVice which would, in fact, render the R face of the

total ensemble more hindered than the â face. If that could be

accomplished, an ethynyl equivalent might be directed to the â

face. We attempt to capture this concept in Scheme 7 wherein

the goal becomes the conversion of diol 37 to the unspecified

construct 38, whose R face would be so shielded as to direct a

nucleophile (possibly an ethynyl silyl anion as shown) to the â

face (see formation of construct 39). Needless to say, this

temporary steric control unit must eventually give way to the

vinylogous carbonate functionality of the natural system.

The 1,2 diol linkage of 37 was engaged as a benzophenone

ketal (see compound 40) under the carefully maintained

conditions specified in Scheme 8. Following previous studies,

and in response to insights provided by molecular modeling, a

theory arose to the effect that one of the diastereotopic phenyl

rings of the ketal might lie beneath the face of the C₂carbon.

Were this the case, a putative nucleophile could well attack from

its â-face.

In the event, when compound 40 was exposed to the action

of methylchloroformate in the presence of the Grignard anion

of triisopropylsilyl acetylene at -20 °C, a 9:1 ratio of diaster-

eomeric products favoring the desired â addition product 41

was isolated in excellent yield.²³Our configurational assign-

ment was confirmed by X-ray crystallographic analysis. Al-

though this chemistry was originally developed with the methyl

carbamate, the use of an allyl carbamate (see compound 42)

proved to be a critical advance for the total synthesis effort (Vide

infra). Thus, an important hurdle had been surmounted. The

problem of required syn stereochemical connectivity issues

relating carbons 2, 4, and 7 had been solved. Construction of

the C₇alkyne unit and conversion of adducts 41 and 42 to an

enediyne cyclization precursor now awaited investigation.

Experiments directed toward this end in the case of compound

42 commenced (Scheme 9) with selective cleavage of the

primary silyl linkage. Under strictly defined acidic conditions,

alcohol 43 became accessible. Oxidation²⁶of compound 43

and exposure of the labile aldehyde to the Corey-Fuchs

^a(a) (Z)-Cl(H)CdCsCtCTMS, Pd(0), many variations.

by the bay region²⁵domain. In the case of intermediate 35, we

were unable to advance to the described enediyne structure 36,

apparently as a consequence of the instability of the vinyloxirane

or quinone moieties to the required palladium based catalysis.

It was in response to these and other setbacks that the path I

program for enediyne annulation was set aside in favor of the

interpolative route contemplated in path II (see Scheme 3). For

this purpose, we returned to compound 29 and its derived diol

37.

The first large challenge to testing the interpolative enediyne

formation contemplated in plan II was that of arranging for the

required relative stereochemistry of the various components. As

noted above, it would be necessary to implement an all cis stereo

relationship between carbons 2, 4, and 7. While the 4,7

relationship had already been secured, via compound 28 through

the Diels-Alder based strategy (Scheme 5) we now had to

squarely face the problem of introducing an ethynyl group on

the â face of C₂, cis to the already existing â disposed functions

at carbons 4 and 7. The problem to be contended with was

that intermolecular Reissert reaction, resulting in placement of

an alkoxycarbonyl group at the nitrogen and an ethynyl

equivalent at C₂, would most likely occur from the R face of

the system, thereby avoiding abutments with the already existing

â face functionality.

It was for the purpose of maneuvering around this very serious

problem, that we were drawn to the possibility of using the

(25) Lakshman, M. K.; Sayer, J. M.; Jerina, D. M. J. Am. Chem. Soc.

1991, 113, 6589.

(26) Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org. Chem. 1979, 44,

4148.

9514 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

Scheme 9^a

Scheme 10^a

^a(a) concentrated HCl, THF, 86% yield; (b) (COCl)₂, DMSO, Et₃N,

CH₂Cl₂, -78 °C; (c) PPh₃, CBr₄, CH₂Cl₂, -78 °C, 84% yield; (d) ⁿBuli,

PhCH₃, -78 °C, 74% yield; (e) TBAF, THF, 0 °C; (f) conc. HCl,

MeOH, 25 °C, 74% yield (two steps); (g) NaH, TBSCl, THF, 85%

yield; (h) Ac₂O, Et₃N, DMAP, CH₂Cl₂, 95% yield.

protocol²⁷unveiled the terminal acetylene 46. At this stage,

both the alkynyl and phenolic functional groups were liberated

by treatment of 46 with TBAF. Additionally, the benzophenone

ketal linkage was cleaved under acidic conditions to afford a

triol which, following selective silylation of the phenolic oxygen,

was smoothly converted to diol 47. Following acylation, diyne

48 was obtained.

In anticipation of the conditions which were to be employed

in the interpolative enediyne construction (i.e., palladium), it

proved advantageous to exchange the carbamate protecting

group to avoid interference with the allyl carbamate. Thus, a

trimethylsilylethoxycarbonyl (TEOC) group was selected to

replace the “alloc” linkage. Unfortunately, the TEOC group

could not be employed from the start of our sequence due to

the requirement of fluoride induced deprotection maneuvers.

Removal of the alloc group from 48 was accomplished

through the action of Pd(PPh₃)₄in combination with morpholine

(Scheme 10).²⁸In situ acylation of 49 with TEOCCl under basic

conditions delivered the carbamate exchanged product 50 in

90% yield. Cleavage of the two acetates with ammonia afforded

compound 51 which, upon stereospecific epoxidation of the

resultant diol with mCPBA, gave rise to epoxide 52. With this

substance in hand, we could initiate an investigation to explore

the feasibility to interpose the cis ethylene unit as discussed

earlier in Scheme 3 under plan II.

^a(a) Pd(PPh₃)₄, morpholine, THF, 0 °C; (b) TEOCCl, NaH, DMAP,

THF, 0-25 °C; (c) NH₃, MeOH, 90% (three steps); (d) mCPBA,

CH₂Cl₂, 87% yield.

Scheme 11^a

The first attempts to accomplish such an insertion involved

a projected Sonogashira variant of the Castro-Stephens reaction

in the methyl carbamate series (Scheme 11).²⁹The epoxydiyne

53 (synthesized in a manner analogous to the above described

TEOC carbamate series) was exposed to the action of Pd(0)/

Cu(I) conditions in the presence of cis-1,2-dichloroethylene (54).

Unfortunately, even after numerous attempts, none of the desired

cyclic enediyne 55 was observed.³⁰

Our attentions then turned to more powerful cross-coupling

conditions. Stille had previously described highly efficient Pd-

catalyzed cross coupling conditions which employed stannyl

acetylide and vinyl iodide.³¹However, the prospects of using

the parent cis-1,2-diiodoethylene, a compound reputed to be

thermally and photolytically unstable,³¹were not reassuring.

Additionally, stannyl acetylides are known to be quite sensitive

functionalities. Therefore, we decided to pursue a coupling

^a(a) Pd(PPh₃)₄, CuI, nBuNH₂, PhH. (b) (catalyst) AgNO₃, NIS,

THF; (c) 57, Pd(PPh₃)₄, DMF, 60 °C, slow addition.

reaction between cis-1,2-distannyl ethylene and a bis-iodoalkyne.

There had been a reported coupling of this type by Liebeskind

and co-workers between an iodoalkyne and a stannyl cy-

clobutenedione.³²A conceptually related interpolative ring

closure, though using totally different chemistry, was described

by Nicolaou and associates in their total synthesis of rapamy-

cin.³³

(27) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 13, 3769.

(28) Four, P.; Guibe, F. Tetrahedron Lett. 1982, 23, 1825.

(29) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975,

16, 4467.

(30) For the only example of this type of construction to produce a cyclic

enediyne in the context of an annulene synthesis, see: Huynh, C.;

Linstrumelle, G. Tetrahedron 1988, 44, 6337.

The possibility was first probed in the case of compound 56.

This substance had been prepared en route to 53 starting with

(32) Liebeskind, L. S.; Fengl, R. W. J. Org. Chem. 1990, 55, 5359.

(33) Nicolaou, K. C.; Chakraborty, T. Y.; Piscopio, A. D.; Monowa, N;

Bertinato, P. J. Am. Chem. Soc. 1993, 115 4419.

(31) Stille, J. K.; Simpson, J. H. J. Am. Chem. Soc. 1987, 109, 2138.

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9515

Scheme 13^a

Scheme 12^a

^a(a) Tf₂O, Pyr, CH₂Cl₂, -20 °C, 95%; (b) Dess-Martin periodinate,

CH₂Cl₂, 95% yield; (c) CrCl₂, THF, 75% yield; (d) MgBr₂, CO₂, Et₃N,

MeCN then MOMCl, iPr₂NEt, THF, 61% yield (two steps); (e) CH₂N₂,

MeOH, 0 °C, 70% yield.

^a(a) AgNO₃(catalyst), NIS, THF, 25 °C, 60: 98%, 61: 91%. (b) 5

mol% Pd(PPh₃)₄, 0.023 M, 75 °C, 62: (78% yield), 63: (81% yield).

methylcarbamate 41, via much the same chemistry employed

in reaching the alloc urethane 48. The two ethynyl linkages of

56 were iodinated with NIS in the presence of silver nitrate to

afford bis-iodo-alkyne 57.^34aExposure of this compound to the

action of Pd(0) and Z-1,2-bis-(trimethylstannyl)ethylene (58)^34b

failed to bring about formation of any of the desired cyclic

enediyne (59). Although rapid cross-coupling had occurred,

only the product of two intermolecular couplings was isolated.

Though the interpolation of the ethylene unit into 56 or 57

was unsuccessful, it was hoped that cyclization could be

accomplished if the epoxide were in place. The C₈-C₉epoxide

might serVe to shorten the approach of the two ethynyl units

while proViding some relief from the projected strain in the

cyclization product. To study this possibility, we returned to

the previously mentioned epoxide 53. Bis-iodination proceeded

smoothly, as described for 56, to provide the bis-iodide 60 in

98% yield. Following extensive experimentation and optimiza-

tion, treatment of 60 with bis-stannyl ethylene reagent 58 and

catalytic Pd(PPh₃)₄under medium dilution conditions, a 78%

yield of the long sought after cyclic enediyne 62 was obtained!³⁵

As anticipated, conduct of the cyclization reaction in the

presence of the central epoxide was critical to the successful

double coupling.

With this success, we returned to the TEOC series, focusing

on the epoxydiol 52. This compound was converted to bis-

iodide 61 in high yield. The transformation which was to be

the hallmark of our total synthesis inVolVed exposure of

compound 61 to Pd(PPh₃)₄and synthon 58 to afford the crucial

cyclic enediyne 63 in 81% isolated yield. With access to cyclic

enediyne 63 accomplished, exploration of chemistry which

would eventually lead to a total synthesis of dynemicin A would

now be taken up in full earnest.

The first stage of the progression of 63 to dynemicin A

involved the further development of the C₅-C₆vinylogous

carbonate region. A crucial requirement in this regard would

be the ability to differentiate the vicinal hydroxyl groups. This

subgoal was indeed realized through selective triflation of the

equatorial C₅hydroxyl with Tf₂O as shown in Scheme 13.

Oxidation of the resultant C₆hydroxyl of 64 with Dess-Martin

periodinane³⁶afforded the labile keto-triflate 65. Reductive

excision of the trifloxy function was accomplished via the action

of CrCl₂to provide ketone 66.³⁷With the intention of installing

the C₅carboxyl function, several attempts to generate a

“stoichiometric” enolate of the ketone in 66 were undertaken.

These initiatives were uniformly unsuccessful. Decomposition

products whose structures were not readily formulated were

produced.

Accordingly, we turned to the use of the mild conditions for

R-carboxylation developed by Rathke.³⁸In the event, ketone

66 was treated with MgBr₂and Et₃N under an atmosphere of

CO₂. This protocol produced a highly labile â-keto acid. It

was not isolated as such, but rather immediately treated with

MOMCl, as shown, to afford enol 67. After careful optimization

of this sequence, a reliable 61% yield could be attained over

the two steps (66 f 67). Finally, methylation of enol 67 with

diazomethane in methanol produced the MOM protected viny-

logous carbonate 68. This compound represents a suitably

protected but fully functionalized version of the ABC ring

system of 1.

Due to the lack of natural material available to us, survey of

a variety of feasible conditions for removal of protecting groups

using probe structures derived from dynemicin would not be

possible. In this context, successful cleavage of the MOM group

from advanced ABC ring systems bolstered our confidence that

this way of containing the vinylogous carbonate and the eventual

anthraquinone hydroxyl functions (Vide infra) would service our

needs. Having developed a synthetic route to the fully func-

tionalized ABC core system of 1, we entered what we hoped

would be the final phase of our total synthesis journey. We

now had to deal with the attachment of the D and E rings to

form the trihydroxyanthraquinone moiety via a deprotectable

precursor of dynemicin A.³⁹

Any DE + ABC annulation scheme which was to be

accomplished, must take note of vulnerabilities which might

result in cleavage of the epoxide linkage. Any intermediate

which lacked nitrogen stabilization would certainly be at serious

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

(37) Hanson, J. R. Synthesis 1974, 1.

(38) Tirpak, R. E.; Olsen, R. S.; Rathke, M. W. J. Org. Chem. 1985, 50,

4877.

(39) For an excellent survey of methods to construct anthraquinone

constructs, see: Krohn, K. Angew.Chem., Int. Ed. Engl. 1986, 25, 790.

(34) (a) Hofmeister, H.; Annen, K.; Laurent, H.; Wiechert, R. Angew.

Chem., Int Ed. Engl. 1984, 23, 727. (b) Mitchell, T. N.; Amamria, A.;

Killing, H.; Rutschow, D. J. Organomet. Chem. 1986, 304, 257.

(35) Shair, M. D.; Yoon, T. Y.; Danishefsky, S. J. J. Org. Chem. 1994,

59, 3755.

9516 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

Scheme 14^a

Scheme 15^a

^a(a) TBAF, THF, 0 °C; (b) PhI(OAc)₂, 0 °C, 60% yield (two steps).

Scheme 16^a

^a(a) TBAF, AcOH, THF, 0 °C; (b) PhI(OAc)₂, MeOH, 0 °C, 65%

(two steps); (c) 72, LDA, THF, 0 °C; (d) various bases and Lewis

acids.

risk with respect to epoxide solVolysis culminating in Bergman

cyclization as discussed earlier in Scheme 2. The resolution

of this problem would require the implementation of delicate

chemical transformations under the most exacting of conditions.

Our first tendencies were to utilize chemistry which we had

developed earlier in the synthesis of a pentacyclic anthraquinone

system in reaching compound 35 discussed earlier. To apply

these lessons to the case at hand, we returned to model

compound 62. In order to fashion a quinone aminal bearing

the cyclic enediyne, the phenolic silyl linkage of 62 was cleaved

(Scheme 14), and the resulting free phenol was oxidized with

PhI(OAc)₂in methanol to afford quinone aminal 69 as a single

diastereomer. Treatment of 69 with the anion of cyanophthalide

72 was undertaken. Unfortunately, in the case at hand, this

reaction did not lead to desired anthraquinone 70. Instead, seco

product 71 was isolated. Apparently, a Michael addition had

indeed occurred. However, the hoped for cyclization by addition

of the resulting enolate to the lactone had not materialized.

Numerous attempts to coax the cyclization of 71 were unsuc-

cessful. Additionally, quinone aminal 69 appeared to be

unresponsive as a potential dienophile in any type of Diels-

Alder process.

Having failed to bring about the needed attachment of the

DE ring system using tested quinone aminal technology, we

hoped to generate a system with increased reactivity through

reduced steric encumberance in the environment of the nitrogen

center. We were concerned that the presence of a carbamate,

or other protective linkage, might inhibit a cyclization event

by placing the N-resident group on the inside contour of the

“bay region”. Previously (Scheme 6, compound 34) we had

rationalized that access to the nitrogen of pentacyclic systems

was thwarted by the steric limits of this concave molecular

surface. The target system we chose, which would deal with

the criteria posed above, was a quinone imine.

^a(a) LiCH(CO₂Me)₂, LiTMP, THF, -78 °C, 71% yield; (b) KOH,

MeOH, H₂O, 95% yield; (c) (trimethylsilyl)ethynyl ether, CH₂Cl₂, 0

°C, quant.

formation of quinone imine 74. Although intermediates similar

to 73 had been observed spectroscopically,^12fthis was the first

instance where a chemical reaction had been conducted on this

unstable system. The quinone imine construct served to stabilize

the epoxide linkage by attenuating the potential nucleophilicity

of the nitrogen from the epoxide linkage. We note in passing

that through recourse to quinone imines, our total synthesis and

prodrug design efforts had been joined. We shall return to this

subject shortly.

At this juncture, we turned to the application of homophthalic

anhydride (HPA) methodology which had been developed by

Tamura and associates as a means of delivering a naphtho-

quinone ring system in a mild and versatile way.⁴⁰Notwith-

standing the numerous examples of HPA systems which had

been previously synthesized, the substitution pattern which was

required for dynemicin had not been reported. The synthesis

of a suitably protected and oxygenated HPA construct is detailed

in Scheme 16.

In a not widely precedented transformation, the bromoarene

75 was condensed with the anion of dimethylmalonate in the

presence of lithium tetramethylpiperidide at low temperature.⁴¹

The homophthalic ester 76 was isolated in 71% yield. Presum-

ably, the product ester arises via the intermediacy of a benzyne

which has been intercepted by the malonate anion. Although

a related transformation had been previously described, the

reported yield was quite low (10%), and the desired reaction

It was not without some trepidation that we undertook

(TBAF) the removal of the TEOC and TBS groups of compound

68 since the resulting product was to be 73 wherein the integrity

of the epoxide linkage could well be threatened. Fortunately,

in situ oxidation PhI(OAc)₂of 73 thus generated led to the

(40) Tamura, Y.; Fukata, F.; Sasho, M.; Tsugoshi, T.; Kita, Y. J. Org.

Chem. 1985, 50, 2273 and references cited therein.

(41) Guyot, M.; Molho, D. Tetrahedron Lett. 1973, 14, 3433.

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9517

Scheme 17^a

^a(a) LHMDS, 78, THF, 0 °C, then 74; (b) PhI(OCOCF₃)₂, THF, 0 °C, 5 min; (c) air, daylight, THF, high concentration; (d) MgBr₂, Et₂O, 15%

yield (four steps).

was accompanied by the formation of numerous side products.

Our optimization of this reaction should prove quite applicable

to the general synthesis of highly substituted homophthalic esters

and related compounds from easily accessible bromo arenes.

Saponification of diester 76 yielded diacid 77 which underwent

cyclodehydration via the action of (trimethylsilyl)ethynyl ether⁴²

thereby affording the target homophthalic anhydride 78.

The successful amalgamation of the DE region of dynemicin,

as homophthalic anhydride 78, and the ABC region, as quinone

imine 74, paving the way for the synthesis to the naturally

occurring product (1) is depicted in Scheme 17. Deprotonation

of 78 with LHMDS generated a bright yellow entity to which

was added quinone imine 74. A product, presumed to be the

highly unstable and unisolable anthrone 79 quickly formed and

was immediately oxidized with PhI(OCOCF₃)₂to deliver

anthracenol 80.⁴³Following the screening of numerous unsuc-

cessful oxidation protocols, it was discovered that exposure of

80 to daylight under aerobic conditions brought about an

oxidation which afforded the long sought after protected goal

structure 81.⁴⁴In the concluding step of the synthesis, treatment

of 81 with MgBr₂in ether, following purification on a sephadex

column, yielded a pure sample of (()-dynemicin A (1).

Comparison of spectral data recorded for synthetic (1) with

literature data and data recorded on a trace specimen supplied

by the Bristol-Myers pharmaceutical company confirmed that

the total synthesis of (1) had indeed been accomplished. The

overall yield of the sequence from 74 to isolated homogenous

1 (four steps) was only 15%. The disappointing efficiency of

the transformations is attributable to the instability of the

intermediates, and the difficulty associated with the isolation

and purification of 1, rather than to failings in the chemical

reactions per se. In fact, the difficulties in the management of

the intermediates in our synthesis due to their marginal stability

cannot be overemphasized. Notwithstanding these problems,

a total synthesis of the enediyne antibiotic dynemicin A (1) has

been achieved.

Synthesis and Biological Evaluation of Enediyne

Containing Quinone Imine Systems

The logic of the emergence of the quinone imine linkage

suitably arranged such that its reduction product might initiate

a sequence leading to diyl formation (see structure 11) was

discussed earlier. Surely, we were drawn to such a system by

analogy to the paradigm for bioreductive activation of mito-

mycins (see structures 12 and 13).¹⁶The recognition that this

type of arrangement could be central to our total synthesis helped

to augment interest in this class of compounds. The first

compound examined was compound 82^17a(Scheme 18). This

compound had been prepared, as previously described, through

a shunt in our total synthesis effort.^17aThe promising biological

activity of compound 82, as regards DNA cutting, as well as in

Vitro cytoxicity and in ViVo experiments in mice have been

described elsewhere.^17,18,19It seemed unlikely that the C₄-

methyl group and the C₅and C₆acetoxyl groups were playing

a role at the mechanistic level in fostering the dynemicin like

activity. We were not unmindful of the synthetic complications

involved in incorporating this apparently noncritical functionality

in the prodrug. Accordingly, we set out to examine a system

which would be more readily synthesized. We defined as our

goal structure compound 93. From a strictly biomechanistic

point of view, there was no a priori reason to include the C₇

hydroxyl group (R₃) OH). In this instance, inclusion of such

an additional group would be likely to simplify the synthesis.

Our route to quinone imine 93 is shown in Scheme 18. It

follows from the enediyne cyclization chemistry used in our

synthesis of calicheamicin and from chemistry developed by

Nicolaou in connection with a prodrug in the dynemicin series

based on a cleavable urethane blocking group.⁴⁵Diacetate 84

was prepared from phenolic acetate 83 via a Polonovski reaction.

A C₂acetylenic unit was then installed by treatment of 84 with

ethynylmagnesium bromide and (trimethylsilyl)ethoxy carbo-

(42) Kita, Y.; Akai, S.; Ajimura, N.; Yoshigi, M.; Yasuda, H.; Tamura,

Y. J. Org. Chem. 1986, 51, 4150.

(43) Although anthracenol 98 could not be purified and hence fully

characterized due to its lability, analysis of its ¹H NMR spectrum matched

well with a related model system.

(44) For the conversion of anthracenols to anthraquinones in the presence

of ground state oxygen and light, see: (a) Cameron, D. W.; Schultz, P. E.

1945, 67, 1721.

(45) Nicolaou, K. C.; Maligres, P.; Suzuki, T.; Wendeborn, W. V.; Dai,

W.-M.; Chadha, R. K. J. Am. Chem. Soc. 1992, 114, 8890.

9518 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

Table 1. In Vitro Cytotoxicity of 93 and Mitomycin C versus

Scheme 18^a

Various Cancer Cell Lines

IC₅₀(µM)

compd

HL-60

MT-2

MT-4

833K

SK-Br-3

93

mitomycin C

0.012

0.08

0.015

0.10

0.0035

0.15

0.020

0.30

0.015

0.30

Table 2. In Vivo Anticancer Activity of 105 and Mitomycin C

(MMC) in B2D6F1 Mice Bearing Lewis Lung Adenocarcinoma

AWC

day 10

ATV

day 10

dose

(mg/kg)

compd

day 7

1.00

control

93

MMC

0

+0.4

-1.2

-3.9

-1.5

-0.8

+1.0

-1.7

-4.8

-0.1

-0.9

1.00

0.5

1.0

0.5

1.0

0.15

0.53

0.38

0.18

0.71

0.45

respect to several cancer cell lines. The results are recorded in

Table 1 which also provides an IC₅₀comparison of the clinically

prescribed anticancer agent mitomycin C. Compound 93

exhibited in Vitro activity between 7 and 43 times greater

potency than mitomycin C.

We turned to a more critical in ViVo evaluation of compound

93 versus clinically prescribed mitomycin C. Listed in Table

2 are the relative life prolongation performances of compound

93 versus mitomycin C with respect to mice bearing subcutane-

ously implanted Lewis lung adenocarcinoma.

As in the in Vitro study, compound 93 has also outperformed

mitomycin C in this screen. After ten days of treatment with a

dosage of 0.5 mg/kg of compound 93 a 50% reduction in tumor

mass resulted. In comparison, mitomycin C treatment has

resulted in a 29% reduction in tumor volume in the same screen.

While more searching preclinical investigations are required and

are, indeed, in progress, the biorationale involved in leading us

to such quinoneimine enediyne prodrugs has been vindicated.

^a(a) mCPBA, CH₂Cl₂, then Ac₂O, 70 °C, 84% yield; (b) THF, -78

°C; (c) mCPBA, CH₂Cl₂, NaHCO₃, 42% yield (two steps); (d) (i) KCN,

MeOH, (ii) TBSCl, imidazole, CH₂Cl₂, 69% yield (two steps); (e)

DIBAH, CH₂Cl₂, -78 °C; (f) PCC, CH₂Cl₂, 25 °C, 46% yield (two

steps); (g) Pd(PPh₃)₄, CuI, nBuNH₂, PhH, 75% yield; (h) AgNO₃,

EtOH-THF, 90% yield; (i) LDA, PhCH₃, -78 °C, 71% yield (j) i)

TBAF, THF, 0 °C, (ii) PhI(OAc)₂, THF, 0 °C, 49% yield (two steps).

Summary

A total synthesis of the enediyne natural product dynemicin

A (1) has been accomplished in 33 chemical steps. Several

notable transformations were employed in this expedition. The

suprafaciality of the Diels-Alder reaction was utilized to control

the stereochemical relationship between the C₄and C₇append-

ages of (1). Additionally, a novel method for the insertion of

an ethylene unit between a cis diyne system was developed for

the synthesis of the strained enediyne bridge. This method has

already found use in another laboratory.^12eAdditionally,

recourse to the highly sensitive 73 (Scheme 15) en route to the

critical quinone imine linkage had been successfully navigated.

A homophthalic anhydride annulation partner for the construc-

tion of the anthraquinone moiety was efficiently constructed

(see compound 78).

As a consequence of the access to enediyne quinone imines

occasioned by these studies, we investigated the use of this

reductively activatible linkage for triggering diyl formation. The

synthesis of two such agents (82 and 93) was accomplished,

and their biological profiles both in Vitro and in ViVo are

suggestive of the emergence of a novel enediyne prodrug,

drawing from the lessons of the mitomycin and dynemicin parent

drugs.

nylchloride to give 85 as a moderately unstable compound.

Stereoselective epoxidation of 85 was accomplished with

mCPBA to furnish 86 in 42% overall yield. The phenolic ester

linkage was then cleaved selectively with KCN in methanol,

and the free phenol was subsequently silylated with TBSCl to

deliver silyl ether 87. It was then necessary to unveil the C₇

hydroxyl group without cleaving the hydrolytically labile

phenolic silyl unit. This was efficiently accomplished by

exposure of 87 to DIBAH at -78 °C which furnished alcohol

88. Ketone 89 was accessed by PCC oxidation of 88 under

standard conditions.

At this juncture, we elected to install the enediyne unit

through a Sonogashira type coupling of 89 with (Z)-1-trimeth-

ylsilyl-4-chloro-but-3-en-1-yne. The intended reaction pro-

ceeded smoothly, and enediyne 90 was in hand.

In order to induce cyclization, the alkynyl bound silyl group

of 90 was cleaved through Ag(I) catalysis, providing 91.

Exposure of this compound to the action of lithium diisopro-

pylamide at -78 °C induced a rapid cyclization which afforded

the cyclic enediyne system 92 in 71% yield. In close analogy

to the chemistry which had been developed during our dyne-

micin A total synthesis effort, compound 92 was treated with

TBAF. The resultant unstable product was then treated im-

mediately with PhI(OAc)₂to produce the goal quinone imine

structure 93 in 49% yield over the two-step procedure.

Experimental Section

General Procedures. ¹H NMR and ¹³C NMR spectra were recorded

on a Bruker AMX-400. Infrared spectra were recorded on a Perkin

Elmer 1600 Series FTIR. Optical rotations were measured on a Jasco

DIP-370 polarimeter. Mass spectra were obtained on a JOEL JMS-

Our first investigation into the antitumor potential of quinone

imine 93 involved evaluation of its in Vitro performance with

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9519

DX-303 HF mass spectrometer. Analytical chromatography was

performed on E. Merck silical gel 60 F₂₅₄plates (0.25 mm). Flash

chromatography was performed on Mallinckrodt silica gel 60 (230-

400 mesh). Tetrahydrofuran (THF) was distilled from sodium metal/

benzophenone ketyl. Dichloromethane (CH₂Cl₂), acetonitrile, benzene

(PhH), triethylamine, and pyridine were distilled from calcium hydride.

N,N-dimethylformamide (DMF) was purchased from Aldrich in sure

seal containers. All other commercially obtained reagents were used

as received.

H3), 6.78 (AB of ABX, ∆ν ) 9.7 Hz, J_AB) 8.9 Hz, J_BX) 2.9 Hz, 2

H, H3′ and 4′), 6.37 (dt, J ) 16.0, 5.9 Hz, 1 H, H2), 6.3-6.0 (m, 2 H,

diene), 5.8-5.7 (m, 2 H, diene), 4.51 (d, J ) 6.0 Hz, 2 H, CH₂OAr),

4.33 (d, J ) 5.9 Hz, 2 H, H1), 3.78 (s, 3H, OCH₃), 1.77 (d, J ) 6.7

Hz, 3H, methyl).

To a stirred solution of oxalyl chloride (5.0 mL, 58 mmol) in 200

mL of anhydrous CH₂Cl₂at -78 °C was slowly added DMSO (9.4

mL, 130 mmol). As soon as gas evolution subsided, a solution of the

above alcohol (12.5 g, 48 mmol) in 30 mL of CH₂Cl₂was added Via

a dropping funnel slowly dropwise over a 30 min period. Further stirred

for 30 min at -78 °C, triethylamine (33 mL, 240 mmol) was added,

and the mixture was allowed to warm to ambient temperature over 15

min before quenched by 300 mL of HCl. The aqueous phase was

extracted twice with CH₂Cl₂, and the combined organic was washed

with saturated aqueous NaHCO₃, dried (MgSO₄), and concentrated to

ca. 50 mL by rotary evaporation. Resulting suspension was diluted

by an equal volume of hexane to precipitate out the product which

was collected and dried to give 9.4 g of 24 as a yellow crystalline

solid. The mother liquor was concentrated and chromatographed on

silica to give an additional 2 g of 24 (92%, combined): mp 82-84 °C

(CH₂Cl₂-hexane); IR (CDCl₃) ν_max2915, 2836, 1672, 1620, 1494,

2-(2,4-Hexadienoxy)-5-methoxybenzaldehyde (21). To a solution

of 5-methoxysalicyl aldehyde 20 (28 g, 0.18 mol) in 300 mL of acetone

was added 100 g (0.72 mol) of finely ground K₂CO₃. To the yellow

suspension was added crude sorbyl bromide (prepared from sorbol (37

mL, 0.33 mol) and PBr₃(15 mL, 0.16 mol) in ice-cold ether), and the

mixture was refluxed for 30 min when the fluorescent yellow color

disappeared. The mixture was allowed to cool to room temperature

and filtered, and the filtrate was concentrated and redissolved in ether.

The ethereal solution was washed twice with 1 N NaOH and once with

saturated brine, dried (Na₂SO₄), and concentrated to a small volume.

Treatment with hexane gave rise to formation of white crystalline solid

which was filtered and dried to give 29 g of 21. The mother liquor

was concentrated and chromatographed on silica to give additional 12

g of the product (96%, combined): mp 74-75 °C (Et₂O-hexane); IR

1

1285, 1210, 1131 cm^-1; H NMR (300 MHz, CDCl₃) δ 9.69 (d, J )

7.8 Hz, 1 H, CHO), 7.85 (d, J ) 16.1 Hz, 1 H, H3), 7.06 (d, J ) 2.9

Hz, 1 H, H6′), 7.10 (dd, J ) 9.0, 2.9 Hz, 1 H, H4′), 6.87 (d, J ) 9.0

Hz, 1 H, H3′), 6.73 (dd, J ) 16.1, 7.8 Hz, 1 H, H2), 6.31 (dd, J )

15.2, 10.5 Hz, 1 H, diene), 6.10 (dd, J ) 15.2, 10.5 Hz, 1 H, diene),

5.8-5.7 (m, 2 H, diene), 4.58 (d, J ) 4.1 Hz, 2 H, CH₂OAr), 3.79 (s,

3 H, OCH₃), 1.77 (d, J ) 6.7 Hz, 3 H, methyl); ¹³C NMR (75 MHz,

CDCl₃) δ 194.5, 153.6, 151.9, 147.9, 134.3, 131.4, 130.5, 129.0, 124.5,

123.9, 118.6, 114.3, 112.6, 69.8, 55.8, 18.2; HRMS (FAB) for C₁₆H₁₈O₃,

calcd 258.1256, found 258.1259. Anal. Calcd for C₁₆H₁₈O₃: C, 74.40;

H, 7.02. Found: C, 74.14; H, 7.05.

1

(neat) ν_max2900, 1679, 1669, 1494, 1275, 1158 cm^-1; H NMR (300

MHz, CDCl₃) δ 10.47 (s, 1 H, CHO), 7.32 (d, J ) 3.2 Hz, 1 H, H6),

7.10 (dd, J ) 9.0, 3.2 Hz, 1 H, H4), 6.93 (d, J ) 9.0 Hz, 1 H, H3),

6.32 (dd, J ) 15.3, 10.4 Hz, 1 H, diene), 6.08 (dd, J ) 15.3, 10.4 Hz,

1 H, diene), 5.8-5.7 (m, 2 H, diene), 4.63 (d, J ) 6.0 Hz, 2 H, CH₂-

OAr), 3.79 (s, 3 H, OCH₃), 1.77 (d, J ) 6.6 Hz, 3 H, methyl); ¹³C

NMR (75 MHz, CDCl₃) δ 189.7, 156.0, 153.8, 134.3, 131.4, 130.4,

125.5, 124.2, 123.6, 115.2, 110.2, 69.9, 55.9, 18.2; MS (20 eV EI) m/e

(rel intensity) 232 (M⁺, 3), 152 (100), 137 (9), 81 (83); HRMS (FAB)

for C₁₄H₁₆O₃, calcd 232.1100, found 232.1098. Anal. Calcd for

C₁₄H₁₆O₃: C, 72.39; H, 6.95. Found: C, 72.34; H, 6.94.

2-(2,4-Hexadienoxy)-5-methoxycinnamaldehyde (24). To a flame-

dried 500 mL flask was placed NaH (60% in mineral oil, 4 g, 0.10

mol). The mineral oil was removed by washing with pentane and

decanting under N₂atmosphere. Charged with 150 mL of anhydrous

THF, to the stirred suspension at 0 °C was added triethyl phospho-

noacetate (18.2 mL, 0.092 mol) slowly dropwise. After the addition

was complete, the mixture was further stirred at room temperature for

30 min to give clear solution, to which was added a solution of 21

(19.5 g, 0.084 mol) in 70 mL THF via a dropping funnel slowly

dropwise over a 30 min period. Further stirred at room temperature

for another 30 min, THF was removed by rotary evaporation. The

residue was diluted by saturated NaHCO₃and extracted 3 times with

Et₂O. Combined ether layer was washed with saturated brine, dried

(MgSO₄), and concentrated to give crude cinnamate 22 as a yellowish

oil which was pure enough to be used directly: IR (neat) ν_max2936,

1710, 1632, 1496, 1215, 1171, 1042 cm^-1; ¹H NMR (300 MHz, CDCl₃)

δ 8.00 (d, J ) 16.2 Hz, 1 H, H2), 7.03 (d, J ) 2.7 Hz, 1 H, H6′), 6.84

(AB of ABX, ∆ν ) 4.3 Hz, J_AB) 9.0 Hz, J_BX) 2.7 Hz, 2 H, H3′ and

4′), 6.48 (d, 16.2 Hz, 1 H, H1), 6.30 (dd, J ) 15.2, 10.4 Hz, 1 H,

diene), 6.08 (dd, J ) 16.2, 10.4 Hz, 1 H, diene), 5.8-5.7 (m, 2 H,

diene), 4.55 (d, J ) 5.9 Hz, 2 H, CH₂OAr), 4.25 (q, J ) 7.1 Hz, OCH₂-

CH₃), 3.77 (s, 3 H, OCH₃), 1.76 (d, J ) 6.6 Hz, 3 H, methyl), 1.33 (t,

J ) 7.1 Hz, 3 H, OCH₂CH₃).

To a solution of the crude 22 from above in 300 mL of anhydrous

CH₂Cl₂at -78 °C was added 1.5 M DiBAL (in toluene, 144 mL, 0.22

mol). Stirred at -78 °C for 1 h, the reaction was quenched by cautious

addition of 5 mL of MeOH before allowed to warm to room

temperature. The mixture was poured into 300 mL of ice-water and

acidified by concentrated HCl with vigorous stirring with external

cooling. The organic layer separated, the aqueous phase was extracted

again with CH₂Cl₂, and the combined organic was dried (MgSO₄) and

concentrated to ca. 75 mL, which was then diluted with an equal volume

of hexane, and resulting slurry was filtered to collect 13.5 g of the

alcohol 23 as a white crystalline solid. Silica gel column chromatog-

raphy of the concentrated mother liquor gave an additional 7 g of the

product (94%, combined): mp 82-84 °C (toluene-hexane); IR

(CDCl₃) ν_max3610, 2938, 1493, 1212, 1042 cm^-1; ¹H NMR (300 MHz,

CDCl₃) δ 7.00 (d, J ) 2.9 Hz, 1 H, H6′), 6.93 (d, J ) 16.0 Hz, 1 H,

6-Methoxy-3-methyl-3,4,4a,9,10,10a-hexahydro-9-oxaphenanthrene-

4-carbaldehyde (26). (A) Thermal Reaction. A solution of 24 (10.0

g, 38.7 mmol) in 250 mL of anhydrous toluene was heated to 80 °C

for 5 days. After cooling to room temperature, the solvent was removed

by rotary-evaporation, and the residue was chromatographed on silica

to give 9.3 g of 25 + 26 (inseparable mixture of endo:exo ) ca. 3:1

by NMR) as a yellowish oil (93%). (B) ZnCl₂-Catalyzed Reaction.

To a solution of 24 (150 mg; 0.58 mmol) in 3 mL of anhydrous CH₂-

Cl₂was added 1 M ZnCl₂(in ether, 0.60 mL, 0.60 mmol). The mixture

was stirred at room temperature for 3 days, diluted by aqueous NH₄Cl,

and extracted twice with CH₂Cl₂. Combined organic was dried (Na₂-

SO₄), concentrated, and chromatographed on silica to give 88 mg of

25 + 26 (endo:exo ) ca. 20:1 by ¹H NMR) as a yellowish oil (59%):

1

IR (neat) ν_max2963, 2721, 1720, 1495, 1210, 1048 cm^-1; H NMR

(300 MHz, CDCl₃) δ 10.01 (d, J ) 4.3 Hz, 1 H, CHO), 6.72 (AB of

ABX, ∆ν ) 13.3 Hz, J_AB) 8.8 Hz, J_BX) 2.7 Hz, 2 H, H12,13), 6.42

(d, J ) 2.7 Hz, 1 H, H10), 5.85 (ddd, J ) 9.8, 4.3, 2.7 Hz, 1 H, H4 or

5), 5.59 (dt, J ) 9.8, 1.8 Hz, 1 H, H4 or 5), 4.41 (dd, J ) 10.1, 5.0 Hz,

1 H, H2), 3.96 (dd, J ) 12.2, 10.1 Hz, 1 H, H2), 3.70 (s, 3 H, OCH₃),

3.12 (app t, J ) 10.9 Hz, 1 H, H8), 2.91 (ddd, J ) 10.9, 6.7, 4.3 Hz,

1 H, H7), 2.78 (m, 1 H, H3), 2.44 (m, 1 H, H6), 1.14 (d, J ) 7.2 Hz,

3 H, 6-CH₃); ¹³C NMR (75 MHz, CDCl₃) δ 206.2, 153.4, 148.7, 134.0,

126.9, 124.5, 117.2, 113.1, 111.4, 71.1, 55.8, 52.2, 36.9, 34.3, 32.2,

16.9; HRMS (FAB) for C₁₆H₁₈O₃, calcd 258.1256, found 258.1238.

Anal. Calcd for C₁₆H₁₈O₃: C, 74.40; H, 7.02. Found: C, 74.60; H,

6.89.

9-(1,4-Benzoquinonyl)-2-hydroxy-8-methyl-3-oxabicyclo[3.3.1]-

non-6-ene (27). To a solution of 25 + 26 (11.0 g, 42.6 mmol, ca. 3:1

diastereomeric mixture) in 100 mL of MeCN at 0 °C was added aqueous

solution of ammonium cerium nitrate (60 g, 109 mmol) in 200 mL of

H₂O rapidly dropwise. The initially black solution became orange as

yellow solid precipitated at the end of the addition. Stirred at 0 °C for

30 min, the solid was collected by filtration and air-dried to give 6.9

g of 27 as a yellow solid (62%): mp >176 °C (dec); IR (CDCl₃) ν_max

3601, 2964, 1657, 1602, 1287, 1080 cm^-1; ¹H NMR (300 MHz, CDCl₃)

δ 6.80 (s, 1 H, H3), 6.73 (AB of ABX, ∆ν ) 18.6 Hz, J_AB) 10.1 Hz,

J_BX) 2.2 Hz, 2 H, H5, 6), 5.73 (AB of ABX, ∆ν ) 30.3 Hz, J_AB

)

10.1 Hz, J_AX) 5.3 Hz, 2 H, H4′, 5′), 5.39 (s, 1 H, acetal), 4.22 (dd,

J ) 10.6, 1.3 Hz, 1 H, -CH₂O-), 3.73 (s, 1 H, H1′), 3.45 (dd, J )

10.6, 1.5 Hz, 1 H, -CH₂O-), 2.74 (br s, 1 H, -OH), 2.25-2.10 (m, 3

9520 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

H, H2′, 3′, 6′), 1.08 (d, J ) 7.2 Hz, 3 H, 3′-CH₃); ¹³C NMR (75 MHz,

CDCl₃) δ 187.9, 187.3, 148.9, 137.4, 135.9, 134.4, 132.6, 123.6, 91.9,

63.8, 41.1, 33.6, 32.8, 30.2, 17.2. Anal. Calcd for C₁₅H₁₆O₄: C, 69.22;

H, 6.20. Found: C, 68.84; H, 6.05.

Purification by column chromatography (SiO₂, 20:1 f 10:1 f 4:1

hexane/EtOAc) yielded 22 g (83%) of the ketal 40 as a sticky white

foam: ¹H NMR (400 MHz, CDCl₃) δ 8.61 (s, 1H), 8.02 (d, J ) 9.0

Hz, 1H), 7.62 (m, 2H), 7.36 (m, 4H), 7.27 (d, J ) 2.5 Hz, 1H), 7.24

(dd, J ) 9.0, 2.5 Hz, 1H), 7.03 (m, 2H), 6.94 (m, 2H), 4.98 (d, J ) 6.9

Hz, 1H), 4.48 (dd, J ) 6.9, 1.4 Hz, 1H), 3.95 (dd, J ) 11.5, 2.1 Hz,

1H), 3.87 (dd, J ) 11.4, 2.1 Hz, 1H), 3.65 (t, J ) 11.4 Hz, 1H), 3.54

(q, J ) 7.8, 1.3 Hz, 1H), 1.35 (d, J ) 7.8 Hz, 3H), 1.06 (s, 9H), 0.89

(s, 9H), 0.29 (s, 6H), 0.05 (s, 3H), 0.02 (s, 3H); ¹³C NMR (100.6 MHz,

CDCl₃) δ 154.1, 149.7, 143.3, 141.6, 141.2, 136.9, 132.4, 131.4, 128.1,

128.0, 127.9, 127.6, 126.6, 126.4, 123.9, 110.1, 108.8, 108.1, 78.3,

75.1, 64.8, 42.9, 35.7, 25.7, 25.6, 22.3, 18.1, -4.3, -4.4, -5.4, -5.5;

FTIR (CDCl₃) ν 2930 (s), 2858 (s), 2247 (m), 1617 (s), 1504 (s), 1449

(s) cm^-1; HRFABMS calcd for C₃₉H₅₃NO₄Si₂655.3513, found 655.3510.

2-(tert-Butyldimethylsilyl)oxy-10-[(tert-butyldimethylsilyl)oxy]-

methyl-7-methyl-7-10-dihydrophenanthridine (29). To a suspension

of 27 (5.8 g, 22.3 mmol) in 20 mL of AcOH was added a solution of

NH₄OAc (12 g, 156 mmol) in 20 mL of water. The mixture was heated

to 100 °C with stirring under N₂atmosphere for 1 h. Resulting dark

solution was cooled to 0 °C and neutralized by NH₄OH until basic

under continuous stream of nitrogen. After further stirring for 1 h at

0 °C with the N₂stream maintained, the precipitate was filtered, washed

3 times with water, and air-dried to give 5 g of greenish tan powder

(28), which was dissolved in 10 mL of DMF, cooled to 0 °C, and 7.1

g of imidazole (91 mmol) and 7.8 g of TBSCl (49 mmol) were added.

After having been stirred for 2 h, diluted by water, and extracted 3

times with Et₂O, the ether layer was washed with saturated brine, dried

(Na₂SO₄), concentrated, and chromatographed to give 9.0 g of 29 as a

slightly tan viscous oil (86%) : IR (neat) ν_max2950, 1617, 1503, 1254,

Alkyne (41). To a solution of (triisopropylsilyl)acetylene (6.35 mL,

28.32 mmol, 3.5 equiv) in THF (100 mL) at 0 °C was added a 3 M

solution of ethylmagnesium bromide (8.09 mL, 24.7 mmol, 3.0 equiv)

in ether. The reaction was allowed to warm to room temperature and

stir for 3 h. After which time, the reaction was cooled to -78 °C, and

a THF (50 mL) solution of quinoline 40 (5.3 g, 8.09 mmol, 1.0 equiv)

was added followed by the slow dropwise addition of methylchloro-

formate (3.12 mL, 40.5 mmol, 5.0 equiv) by syringe pump. After

complete addition of the chloroformate, the reaction was allowed to

slowly warm to -30 °C over 1.5 h and stir at -30 °C for an additional

15 h. The reaction was quenched by the addition of methanol (5 mL)

and allowed to warm to room temperature. The reaction was poured

into a mixture of EtOAc (400 mL) and H₂O (100 mL). The layers

were separated, and the organic layer was washed with saturated brine

(100 mL), dried (MgSO₄), filtered, and concentrated. Purification by

column chromatography (SiO₂, 25:1 f 10:1 hexane/EtOAc) gave 26.9

g (73%) of the alkyne 41 as a clear oil: ¹H NMR (300 MHz, CDCl₃)

δ 7.55 (m, 3H), 7.37 (m, 4H), 7.10 (m, 4H), 6.80 (d, J ) 2.5 Hz, 1H),

6.70 (dd, J ) 8.6 2.5 Hz, 1H), 5.50 (br s, 1 H), 4.70 (d, J ) 7.0 Hz,

1H), 4.22 (d, J ) 7.0 Hz, 1H), 3.84 (m, 1H), 3.80 (s, 3H), 3.43 (t, J )

10.5 Hz, 1H), 3.26 (dd, J ) 10.5, 3.0 Hz, 1H), 2.81 (br s, 1.0 H), 1.34

(d, J ) 8.1 Hz, 3H), 1.05 (s, 9H), 0.99 (s, 9H), 0.85 (m, 21 H), 0.20

(s, 6H), 0.07 (s, 3H), 0.03 (s, 3H); FTIR (CDCl₃) ν 2955 (s), 2930 (s),

2890 (m), 2246 (w), 1698 (s), 1494 (s), 1445 (m), 1257 (s) cm^-1; LRMS

(CI, NH₃) m/z 926 ([M + H + NH₃] 100%); Anal. Calcd for C₅₃H₇₇-

NO₆Si₃: C, 70.12; H, 8.56. Found: C, 70.22; 8.59.

1

1101, 837 cm^-1; H NMR (300 MHz, CDCl₃) δ 8.60 (s, 1 H, H2),

7.96 (d, J ) 8.9 Hz, 1 H, H13), 7.38 (d, J ) 2.5 Hz, 1 H, H10), 7.23

(dd, J ) 8.9, 2.5 Hz, 1 H, H12), 6.18 (dd, J ) 9.8, 4.4 Hz, 1 H, H6),

6.08 (dd, J ) 9.8, 4.5 Hz, 1 H, H5), 4.2-4.0 (m, 2 H, H7, one of

CH₂OTBS), 3.7-3.5 (m, 2 H, H4, one of CH₂OTBS), 1.41 (d, J ) 7.2

Hz, 3 H, 4-CH₃), 1.03 (s, 9 H), 0.85 (s, 9 H), 0.27 (s, 6 H), -0.1 (s,

3 H), -0.3 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 154.0, 149.2, 142.8,

137.8, 133.8, 131.4, 131.1, 127.5, 127.1, 123.9, 110.8, 69.4, 39.2, 33.5,

26.0, 25.8, 25.2, 18.5, 18.3, -4.2, -5.3; HRMS (FAB) for C₂₇H₄₄-

NO₂Si₂(M + H), calcd 470.2911, found 470.2913. Anal. Calcd for

C₂₇H₄₃NO₂Si₂: C, 69.03; H, 9.23; N, 2.98. Found: C, 68.08; H, 9.49;

N, 3.00.

2-(tert-Butyldimethylsilyl)oxy-10-[(tert-butyldimethylsilyl)oxy]-

methyl-8,9-dihydroxy-7-methyl-7,8,9,10-tetrahydrophenanthri-

dine (37). To a solution of 29 (9.0 g, 19 mmol) in 30 mL of THF and

12 mL of t-BuOH was added 0.2 M OsO₄(in THF, 0.5 mL, 0.1 mmol)

and 60% aqueous NMO (5 mL, 48 mmol). The mixture was stirred at

room temperature for 4 h, concentrated to a small volume (ca. 5 mL),

and diluted by 5 mL of 10% NaHSO₃and 50 mL of saturated NaHCO₃.

Resulting precipitate was collected, washed thoroughly with water, air-

dried, and finally washed with hexane to give 8.93 g of 37 as a slightly

tan powder (90%) which was pure enough to be used in subsequent

reactions. An analytical sample was prepared by recrystallization from

hexane-ethyl acetate: mp 206-209 °C (hexane-EtOAc); IR (neat)

Alkyne (42). To a solution of (triisopropylsilyl)acetylene (14.8 mL,

66 mmol, 2.0 equiv) in THF (220 mL) at 0 °C was added a 3 M solution

of ethylmagnesium bromide (22 mL, 66 mmol, 2.0 equiv) in ether.

The reaction was allowed to warm to room temperature and stir for 3

h. After which time, the reaction was cooled to -78 °C, and a THF

(50 mL) solution of quinoline 40 (22 g, 32.98 mmol, 1.0 equiv) was

added followed by the slow dropwise addition of allylchloroformate

(10.83 mL, 102 mmol, 3.1 equiv) via syringe pump. After complete

addition of the chloroformate, the reaction was allowed to slowly warm

to -30 °C over 1.5 h and stir at -30 °C for an additional 15 h. The

reaction was quenched by the addition of methanol (10 mL) and allowed

to warm to room temperature. The reaction was poured into a mixture

of EtOAc (500 mL) and H₂O (200 mL). The layers were separated,

and the organic layer was washed with saturated brine (100 mL), dried

(MgSO₄), filtered, and concentrated. Purification by column chroma-

tography (SiO₂, 25:1 f 10:1 hexane/EtOAc) gave 26.9 g (80%) of the

alkyne 42 as a clear oil: ¹H NMR (400 MHz, CDCl₃) (330 K) δ 7.82

(d, J ) 5.3 Hz, 2H), 7.51-7.62 (m, 2H), 7.47 (t, J ) 5.3 Hz, 2H),

7.41-7.35 (m, 2H), 7.14-7.07 (m, 2H), 6.82 (d, J ) 2.6 Hz, 1H),

6.72 (dd, J ) 6.7, 2.6 Hz, 1H), 5.94 (m, 1H), 5.55 (br s, 1H), 5.35 (dd,

J ) 17.1, 1.4 Hz, 1H), 5.21 (d, J ) 10.5 Hz, 1H), 4.73 (d, J ) 7.0 Hz,

1H), 4.70 (br s, 2H), 4.23 (d, J ) 7.0 Hz, 1H), 3.84 (dd, J ) 10.7, 3.8

Hz, 1H), 3.44 (t, J ) 10.7 Hz, 1H), 3.27 (dd, J ) 10.7, 3.8 Hz, 1H),

2.84 (br s, 1H), 1.36 (d, J ) 7.7 Hz, 3H), 1.00 (s, 9H), 0.91 (s, 9H),

0.88 (s, 18H), 0.22 (s, 9H), 0.09 (s, 3H), 0.05 (s, 3H); ¹³C NMR (100.6

MHz, CDCl₃) δ 152.4, 141.4, 137.4, 132.2, 132.1, 129.9, 129.4, 128.1,

128.0, 127.9, 127.6, 126.9, 126.7, 124.8, 118.2, 117.6, 113.9, 108.2,

104.7, 84.7, 78.7, 75.2, 66.5, 64.2, 49.7, 42.7, 39.9, 25.7, 25.6, 20.2,

18.3, 18.2, 18.1, 18.0, 11.0, 10.8, 10.5, -4.5, -5.3, -5.6; IR (CDCl₃)

ν 2955 (s), 2247 (w), 1697 (s), 1657 (m), 1448 (s), 1279 (s), 1258 (s)

cm^-1; HRFABMS calcd for C₅₂H₇₉NO₆Si₃897.5215, found 897.5210.

1

ν_max3600, 2931, 1617, 1505, 1260, 1238, 1110 cm^-1; H NMR (300

MHz, CDCl₃) δ 8.63 (s, 1 H, H2), 7.94 (d, J ) 8.9 Hz, 1 H, H13),

7.28 (d, J ) 2.5 Hz, 1 H, H10), 7.21 (dd, J ) 9.0, 2.4 Hz, 1 H, H12),

4.49 (app t, J ) 2.8 Hz, 1 H, H6), 4.13 (dd, J ) 8.9, 2.5 Hz, 1 H,

CH₂OTBS), 3.99 (dd, J ) 7.2, 2.5 Hz, 1 H, H5), 3.72 (app dt, J ) 8.9,

2.8 Hz, 1 H, H7), 3.65 (app t, J ) 8.9 Hz, 1 H, CH₂OTBS), 3.10 (app

quint, J ) 7.2 Hz, 1 H, H4), 1.55 (d, J ) 7.2 Hz, 3 H, 4-CH₃), 1.02 (s,

9 H), 0.83 (s, 9 H), 0.25 (s, 6 H), -0.02 (s, 3 H), -0.07 (s, 3 H); ¹³C

NMR (75 MHz, CDCl₃) δ 154.2, 148.5, 142.7, 136.2, 132.6, 131.4,

127.9, 127.1, 124.1, 110.5, 72.4, 70.9, 65.3, 45.6, 36.0, 25.9, 25.8, 18.5,

18.3, -4.2, -5.4; MS (20 eV EI) m/e (rel intensity) 446 (40, M-^tBu),

428 (39), 336 (23), 324 (25), 312 (100); HRMS (FAB) for C₂₇H₄₆-

NO₄Si₂(M + H), calcd 504.2965, found 504.2979. Anal. Calcd for

C₂₇H₄₅NO₄Si₂: C, 64.37; H, 9.00; N, 2.78. Found: C, 64.18; H, 9.01;

N, 2.64.

Quinoline (40). The diol 37 (20 g, 39.8 mmol, 1.0 equiv) was

dissolved in CH₂Cl₂(300 mL) and cooled to 0 °C. To this solution

was added Ph₂C(OMe)₂(23 g, 99 mmol, 2.5 equiv) and concentrated

H₂SO₄(3.3 mL, 59.7 mmol, 1.5 equiv), and the reaction was then

allowed to warm to 25 °C followed by warming to 40 °C. After 2 h

at 40 °C, the reaction was allowed to cool to 25 °C, diluted with Et₂O

(750 mL), washed with saturated NaHCO₃(2 × 150 mL), and saturated

brine (200 mL). The organic layer was dried (MgSO₄), filtered, and

concentrated. The crude ketal was then resubjected to silylative

conditions by dissolving in CH₂Cl₂(150 mL), followed by addition of

imidazole (3.25 g, 48 mmol, 1.2 equiv) and TBSCl (6.0 g, 40 mmol,

1.0 equiv) after 30 min, and the reaction was diluted with EtOAc (600

mL) and washed with H₂O (250 mL) and saturated brine (200 mL).

The organic layer was dried (MgSO₄), filtered, and concentrated.

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9521

Alcohol (43). The alkyne 42 (26.9 g, 28.3 mmol, 1 equiv) was

dissolved in THF (200 mL) and cooled to 0 °C. Concentrated HCl

(5.1 mL) was added dropwise, and the reaction was allowed to warm

to room temperature. After stirring 1.2 h, the reaction was quenched

by the addition of saturated NaHCO₃(200 mL) and diluted with EtOAc

(200 mL). The organic layer was separated and washed with saturated

brine (100 mL), dried (MgSO₄), filtered, and concentrated. Purification

by column chromatography (SiO₂, 25:1 f 15:1 f 8:1 hexane/EtOAc)

provided the alcohol 43, 20 g (86%) as a white foam: mp ) 72-77

1497 (s), 1393 (s), 1245 (s) cm^-1; HRFABMS calcd for C₃₂H₃₉NO₇Si

577.2496, found 577.2498.

Diol (51). The allylcarbamate 48 (2.40 g, 4.03 mmol, 1.0 equiv)

was dissolved in THF (60 mL) and cooled to 0 °C. To this solution

was added Pd(PPh₃)₄(140 mg, 0.121 mmol, 0.03 equiv) followed by

morpholine (843 µL, 9.67 mmol, 2.4 equiv) dropwise, and the reaction

was allowed to stir at 0 °C for 2 h. Then, the reaction was warmed to

25 °C for 30 min and cooled back to 0 °C. To the reaction was added

a 60% dispersion of NaH (806 mg, 20.13 mmol, 5.0 equiv) and TeocCl

(3.5 mL, 20.13 mmol, 5.0 equiv), and the reaction was allowed to warm

to 25 °C and stir for 13 h. At which point, the reaction was quenched

by the addition of H₂O (50 mL) and diluted with EtOAc (300 mL).

The layers were separated, and the organic layer washed with saturated

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

diacetate 50 was then dissolved in dry MeOH (30 mL), and 7 M NH₃

(20 mL, 20 equiv) was added. After 20 h, the reaction was concentrated

to a small volume and purified by column chromatography (SiO₂, 3:2

hexane/EtOAc) to yield 2.02 g (90%) of the diol 51 as a white foam:

1

°C; H NMR (400 MHz, CDCl₃) (330 K) δ 7.57 (dd, J ) 8, 1.4 Hz,

2H), 7.39 (m, 3H), 7.16 (m, 5H), 6.84 (d, J ) 2.6 Hz, 1H), 6.75 (dd,

J ) 8.7, 2.6 Hz, 1H), 5.95 (m, 1H), 5.61 (br s, 1H), 5.35 (d, J ) 17.2,

1.4 Hz, 1H), 5.21 (d, 10.4 Hz), 4.72 (br s, 1H), 4.64 (d, J ) 6.7 Hz,

1H), 4.31 (d, J ) 6.7 Hz), 3.86 (d, J ) 11.1 Hz, 1H), 3.68 (m, 1H),

3.31 (d, J ) 5.1 Hz, 1H), 2.80 (br s, 1H), 1.39 (d, J ) 7.6 Hz), 1.07

(s, 3H), 1.02 (s, 9H), 0.89 (s, 18H), 0.23 (s, 6H); ¹³C NMR (100.6

MHz, CDCl₃) δ 152.4, 141.7, 141.6, 132.1, 129.2, 128.9, 128.0, 127.9,

127.6, 126.7, 126.6, 125.5, 125.2, 124.1, 118.5, 117.6, 113.8, 108.2,

104.7, 84.7, 78.9, 76.0, 66.6, 63.9, 49.5, 42.3, 40.2, 25.6, 25.5, 19.9,

18.3, 18.2, 18.1, 17.6, 12.2, 11.1, 10.9, 10.8, 10.5, -3.7, -4.5, -4.5;

FTIR (CDCl₃) ν 3670 (m), 3619 (m), 2943 (s), 2248 (m), 1698 (s),

1493 (s), 1281 (s) cm^-1; HRFABMS calcd for C₄₆H₆₅NO₆Si₂783.4350,

found 819.4354.

1

mp 127-130 °C.; H NMR (400 MHz, CDCl₃) (330 K) δ 7.45 (br s,

1H), 7.21 (d, J ) 2.7 Hz, 1H), 6.74 (dd, J ) 8.8, 2.7 Hz, 1H), 5.68 (br

s, 1H), 4.30 (m, 2H), 4.21 (s, 1H), 3.91 (t, J ) 7.2 Hz, 1H), 3.53 (s,

1H), 2.50 (m, 1H), 2.33 (d, J ) 2.6Hz, 1H), 2.15 (d, J ) 4.5Hz, 1H),

2.11 (d, J ) 2.3 Hz, 1 H), 2.05 (d, J ) 2.3 Hz, 1H), 1.42 (J ) 7.0 Hz,

3H), 1.06 (t, J ) 7.5 Hz, 2H), 0.98 (s, 9H), 0.22 (s, 6H), 0.03 (s, 9H);

¹³C NMR (100.6 MHz, CDCl₃) δ 152.2, 128.8, 124.9, 122.3, 118.9,

114.3, 82.2, 80.3, 72.4, 72.3, 72.2, 71.2, 64.8, 45.5, 37.2, 35.5, 25.5,

18.0, 17.5, 14.7, -1.65, -4.6; FTIR (CDCl₃) ν 3614 (m), 3577 (m),

Acetylene (46). The dibromoolefin 45 (20 g, 20.51 mmol, 1.0 equiv)

n

was dissolved in toluene (450 mL) and cooled to -78 °C. Then, -

Buli (17.23 mL, 43.07 mmol, 2.1 equiv) was added dropwise via syringe

pump. Following addition, the reaction was stirred at -78 °C for 4.5

h and then quenched at -78 °C by the addition of AcOH (3 mL). The

reaction was allowed to warm to room temperature and then diluted

with EtOAc (500 mL) and H₂O (300 mL). The layers were separated,

and the organic layer was dried (MgSO₄), filtered, and concentrated.

Purification by column chromatography (SiO₂, 20:1 hexane/EtOAc)

yielded 12.3 g (74%) of the acetylene 46 as a white foam: mp ) 121-

125 °C; ¹H NMR (400 MHz, CDCl₃) δ 7.55 (dd, J ) 8.0, 1.4 Hz, 2H),

7.42 (m, 3H), 7.09 (m, 6H), 6.94 (d, J ) 2.6 Hz, 1H), 6.73 (d, J ) 7.3

Hz, 1H), 5.94 (m, 1H), 5.70 (br s, 1H), 5.31 (d, J ) 17.2 Hz, 1H),

5.19 (d, J ) 5.19 Hz, 1H), 4.71 (br s, 1H), 4.70 (d, J ) 7.2 Hz, 1H),

4.32 (d, J ) 7.2 Hz, 1H), 3.90 (s, 1H), 2.92 (br s, 1H), 2.13 (d, J ) 2.8

Hz, 1H), 1.49 (d, J ) 7.6 Hz, 3H), 1.00 (s, 9H), 0.89 (s, 21H), 0.23 (s,

3H), 0.21 (s, 3H); ¹³C NMR (100.6 MHz, CDCl₃) δ 152.5, 140.5, 132.2,

129.9, 128.4, 127.9, 127.6, 126.9, 126.3, 118.7, 117.6, 114.0, 108.5,

103.7, 85.6, 83.2, 78.6, 77.5, 77.2, 76.9, 76.6, 70.9, 66.6, 50.2, 39.1,

30.1, 25.6, 18.3, 18.2, 18.2, 17.5, 10.7, -4.5, -4.7; IR (CDCl₃) 3306

(m), 2943 (s), 1698 (s), 1495 (s), 1449 (s), 1281 (s) cm^-1; HRFABMS

calcd for C₄₇H₆₃NO₅Si₂777.4244, found 777.4241.

3306 (s), 2956 (s), 2249 (m), 1691 (s), 1495 (s), 1202 (s) cm^-1

HRFABMS calcd for C₃₀H₄₃NO₅Si₂553.2680, found 553.2677.

;

Epoxide (52). The diol 51 (1.5 g, 2.75 mmol, 1.0 equiv) was

dissolved in CH₂Cl₂(20 mL). To this solution was added 95% mCPBA

(2.11 g, 11.01 mmol, 4.0 equiv) at once as a solid. The reaction was

allowed to stir for 8.5 h at 25 °C. After which time, saturated NaHCO₃

(50 mL) was added to the reaction followed by the addition of Me₂S

(710 µL). After stirring for 20 min, the reaction was diluted with EtOAc

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

with saturated brine (100 mL), dried (MgSO₄), filtered, and concen-

trated. Purification by column chromatography (SiO₂, 2:1 hexane/

EtOAc) provided 1.31 g (87%) of the epoxide 52 as a white foam:

1

mp 95-97 °C; H NMR (400 MHz, CDCl₃) δ 7.26 (d, J ) 2.7 Hz,

1H), 7.18 (br s, 1H), 6.83 (dd, J ) 8.7, 2.7 Hz, 1H), 5.66 (br s, 1H),

4.26 (m, 1H), 4.18 (br s, 1H), 4.12 (dt, J ) 12.1, 3.1 Hz, 1H), 3.85 (t,

J ) 3.7 Hz, 1H), 3.78 (ddd, J ) 10.5, 3.1 Hz, 1H), 3.43 (d, J ) 12.1

Hz, 1H), 2.35 (m, buried, 1H), 2.35 (d, J ) 2.5 Hz, 1H), 2.20 (d, J )

10.7 Hz, 1H), 2.13 (s, 1H), 1.58 (d, J ) 7.3 Hz, 3 H), 0.99 (s, 9H),

0.23 (s, 3H), 0.22 (s, 3H), -0.02 (br s, 9H); ¹³C NMR (100.6 MHz,

CDCl₃) δ 152.7, 130.2, 120.03, 118.9, 79.7, 79.00, 74.8, 73.8, 73.5,

72.7, 70.8, 64.6, 62.2, 45.5, 37.9, 32.9, 25.4, 17.9, 17.3, 15.6, -2.1,

-4.8, -4.9; FTIR (CDCl₃) ν 3562 (w), 3307 (m), 2956 (s), 2257 (m),

1697 (s), 1501 (s), 1252 (s), 1208 (s) cm^-1; HRFABMS calcd for

C₃₀H₄₃NO₆Si₂569.2629, found 569.2633.

Diacetate (48). The triol (2.44 g, 6.404 mmol, 1.0 equiv) was

azeotroped from benzene (3 × 10 mL) and dissolved in THF (35 mL).

A 60% dispersion of NaH (307 mg, 7.69 mmol, 1.2 equiv) was added

at once to this solution. After 5 min, TBSCl (1.17 g, 8.00 mmol, 1.25

mmol) was added at once as a solid. After an additional 15 min, the

reaction was quenched by the addition of H₂O (100 mL) and EtOAc

(500 mL). Separation of the layers was followed by washing the

organic layer with saturated brine (100 mL), drying (MgSO4), filtering,

and concentration. The crude product was directly acylated by

dissolving the diol in CH₂Cl₂(40 mL) and adding Et₃N (4.5 mL, 32

mmol, 5.0 equiv) and Ac₂O (3.0 mL, 32 mmol, 5.0 equiv) followed by

DMAP (50 mg). After 12 h of being stirred, the reaction was diluted

with EtOAc (500 mL) and washed with H₂O (150 mL) and saturated

brine (100 mL). The product was dried (MgSO₄), filtered, and

concentrated. Purification by column chromatography (SiO₂, 5:1 f

3:1 hexane/EtOAc) provided 3.15 g (83%, two steps) of the diacetate

48 as a white foam, mp 72-77 °C: ¹H NMR (400 MHz, CDCl₃) (330

K) δ 7.45 (br s, 1H), 7.21 (d, J ) 2.7 Hz, 1H), 6.77 (dd, J ) 8.8, 2.7

Hz, 1H), 5.95 (m, 1H), 5.72 (br s, 1H), 5.50 (t, J ) 2.4 Hz, 1H), 5.30

(m, 3H), 4.76 (dd, J ) 6.7, 2.5 Hz, 1H), 4.64 (br s, 1H), 3.51 (s, 1H),

2.79 (m, 1H), 2.46 (d, J ) 2.6 Hz, 1H), 2.14 (d, J ) 2.5 Hz, 1H), 2.08

(s, 3H), 2.02 (s, 3H), 1.30 (d, J ) 7.0 Hz, 1H), 0.98 (s, 9H), 0.22 (s,

3H), 0.21 (s, 3H); ¹³C NMR (100.6 MHz, CDCl₃) δ 170.2, 170.1, 152.5,

131.9, 128.5, 119.2, 117.9, 114.5, 80.54, 79.8, 73.3, 72.3, 71.7, 70.1,

66.9, 45.4, 34.4, 33.4, 25.5, 20.9, 20.8, 18.0, 14.2, -4.5, -4.6; IR

(CDCl₃) ν 3306 (s), 2956 (s), 2258 (m), 2248 (m), 1744 (s), 1699 (s),

Bisiodide (60). The epoxide 53 (150 mg, 0.268 mmol, 1.0 equiv)

was dissolved in THF (4.0 mL). To this solution was added AgNO₃

(4.6 mg, 0.027 mmol, 0.1 equiv) followed by N-iodosuccinimide (150

mg, 0.67 mmol, 2.5 equiv), and the reaction was stirred in the dark for

3.5 h. After which time, the reaction was diluted with H₂O (50 mL)

and EtOAc (150 mL). The layers were separated, and the organic layer

was washed saturated brine (50 mL), dried (MgSO₄), filtered, and

concentrated. Purification by column chromatography (SiO₂, 2.5:1

hexane/EtOAc) provided 215 mg (98%) of the bis-iodoalkyne 60 as a

very pale yellow glass: ¹H NMR (400 MHz, CDCl₃) δ 7.20 (d, J )

2.6 Hz, 1H), 7.11 (br s, 1H), 6.79 (dd, J ) 8.7, 2.6 Hz, 1H), 5.78 (br

s, 1H), 5.36 (t, J ) 2.8 Hz, 1H), 5.23 (dd, J ) 11, 2.8 Hz, 1H), 3.95

(d, J ) 2.8 Hz, 1H), 3.72 (s, 1H), 2.70 (m, 1H), 2.07 (s, 3H), 2.06 (s,

3H), 1.39 (d, J ) 7.3 Hz, 3H), 1.00 (s, 9H), 0.25 (s, 3H), 0.24 (s, 3H);

FTIR (CDCl₃) ν 2956 (s), 2257 (m), 2190 (w), 1741 (s), 1703 (s), 1502

(s), 1373 (s), 1246 (s), 1151 (m) cm^-1; LRMS (CI, NH₃) 838 ([M +

NH₃+ H] 73%), 837 ([M + NH₃] 90%), 818, ([M + H] 12%). Anal.

Calcd for C₃₀H₃₅I₂NO₈Si: C, 44.06; H, 4.28. Found: C, 43.96; H,

4.12.

Enediyne (62). The diiodide 60 (197 mg, 0.241 mmol, 1.0 equiv)

was dissolved in DMF (22 mL), and dry argon was passed through the

9522 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

solution for 20 min. The reaction was then heated to 75 °C, at which

temperature Pd(PPh₃)₄(14 mg, 0.012 mmol, 0.05 equiv) was added

under a stream of argon. A 0.023 M solution of cis-1,2-bis-

(trimethylstannyl)ethylene (12.0 mL, 0.277 mmol, 1.15 equiv) in

degassed DMF was then added dropwise to the reaction via syringe

pump over 1.2 h. After complete addition, the yellow-tan reaction

mixture was allowed to warm to room temperature and then poured

into H₂O (100 mL) and Et₂O (250 mL). The layers were separated,

and the organic layer was washed with H₂O (5 × 30 mL). The organic

layer was then washed with saturated brine, dried (MgSO₄), filtered,

and concentrated in vacuo. Purification by column chromatography

(SiO₂, 4:1 EtOAc/hexane) gave 111 mg (78%) of the enediyne 62 as

an off-white foam: mp 77-78 °C; ¹H NMR (300 MHz, CDCl₃) δ 7.16

(br s, 1H), 6.94 (d, J ) 2.0 Hz, 1H), 6.75 (dd, J ) 8.0, 2.0 Hz, 1H),

5.77 (dd, J ) 10.1, 1.0 Hz, 1H), 5.69 (dd, J ) 10.1, 1.0 Hz, 1H), 5.49

(t, J ) 3 Hz, 1H), 5.32 (dd, J ) 11.0, 3.0 Hz, 1H), 3.99 (d, J ) 3.0

Hz, 1H), 3.76 (s, 3H), 2.80 (m, 1H), 2.09 (s, 3H), 2.06 (s, 3H), 1.40

(d, J ) 8.0 Hz, 3H), 0.96 (s, 9H), 0.19 (s, 3H), 0.18 (s, 3H); FTIR

(CDCl₃) 2959 (s), 1743 (s), 1701 (s), 1492 (s), 1223 (w); LRMS (EI)

m/z 591 ([M⁺] 100). Anal. Calcd for C₃₂H₃₇NO₈Si: C, 64.97; H, 6.26.

Found: C, 65.09; H, 6.31.

Bisiodide (61). The epoxide 52 (560 mg, 1.005 mmol, 1.0 equiv)

was dissolved in THF (20 mL). To this solution was added AgNO₃

(18 mg, 0.10 mmol, 0.1 equiv) followed by N-iodosuccinimide (610

mg, 2.71 mmol, 2.7 equiv), and the reaction was stirred in the dark for

3.5 h. After which time, the reaction was diluted with H₂O (100 mL)

and EtOAc (300 mL). The layers were separated, and the organic layer

was washed saturated brine (100 mL), dried (MgSO₄), filtered, and

concentrated. Purification by column chromatography (SiO₂, 2:1

hexane/EtOAc) provided the bis-iodoalkyne 61 740 mg (91%) as a

very pale yellow glass: ¹H NMR (400 MHz, CDCl₃) δ 7.20 (d, J )

2.6 Hz, 1H), 7.18 (br s, 1H), 6.84 (dd, J ) 8.7, 2.4 Hz, 1H), 5.72 (s,

1H), 4.35 (m, 1H), 4.16 (br s, 1H), 4.06 (dt, J ) 12.1, 3.2 Hz, 1H),

3.95 (d, J ) 3.0 Hz, 1H), 3.76 (m, 1H), 3.41 (d, J ) 12.1 Hz, 1H),

2.45 (m, 1H), 2.22 (d, J ) 10.4 Hz, 1H), 1.55 (d, J ) 7.3 Hz, 3H),

1.00 (s, 9H), 0.26 (s, 3H), 0.24 (s, 3H), -0.02 (s, 9H); ¹³C NMR (100.6

MHz, CDCl₃) δ 152.5, 130.1, 120.1, 118.8, 79.5, 79.0, 74.3, 73.3, 73.1,

73.0, 72.5, 71.1, 65.9, 62.0, 51.1, 45.5, 37.8, 34.1, 25.3, 17.8, 15.2,

-2.0, -4.8, -4.9; FTIR (CDCl₃) ν 3560 (w), 3301 (m), 2953 (s), 2261

(m), 1699 (s), 1501 (s), 1251 (s) cm^-1; HRFABMS calcd for C₃₀H₄₁I₂-

NO₆Si₂821.0565, found 821.0559.

Enediyne (63). The diiodide 61 (480 mg, 0.592 mmol, 1.0 equiv)

was dissolved in DMF (56 mL) and dry argon was passed through the

solution for 20 min. The reaction was then heated to 75 °C, at which

temperature Pd(PPh₃)₄(34 mg, 0.0296 mmol, 0.05 equiv) was added

under a stream of argon. A 0.023 M solution of cis-1,2-bis-

(trimethylstannyl)ethylene (33 mL, 0.769 mmol, 1.3 equiv) in degassed

DMF was then added dropwise to the reaction via syringe pump over

1.2 h. After complete addition, the yellow-tan reaction mixture was

allowed to warm to room temperature and then poured into H₂O (100

mL) and Et₂O (250 mL). The layers were separated, and the organic

layer was washed with H₂O (5 × 60 mL). The organic layer was then

washed with saturated brine, dried (MgSO₄), filtered, and concentrated

in vacuo. Purification by column chromatography (SiO₂, 3:1 EtOAc/

hexane) gave 284 mg (81%) of the enediyne 63 as an off-white foam:

mp 77-78 °C; ¹H NMR (400 MHz, CDCl₃) (330 K) δ 7.20 (br s, 1H),

7.00 (d, J ) 2.6 Hz, 1H), 6.79 (dd, J ) 8.7, 2.5 Hz, 1H), 5.75 (br s,

buried, 1H), 5.74 (dd, J ) 9.9, 1.4 Hz, 1H), 6.67 (dd, J ) 9.9, 1.4 Hz,

1H), 4.29 (m, 1H), 4.22 (br s, 1H), 4.09 (dt, J ) 12.5, 3.7 Hz, 1H),

4.03 (d, J ) 2.9 Hz, 1H), 3.79 (dt, J ) 12.5, 3.7 Hz, 1H), 3.34 (d, J

) 12.1 Hz, 1H), 2.47 (m, 1H), 2.27 (d, J ) 12.5 Hz, 1H), 1.54 (d, J

) 6.5 Hz, 1H), 1.02 (bt, buried, 3H), 0.97 (s, 9H), 0.19 (s, 3H), 0.18

(s, 3H), 0.00 (br s, 9H); ¹³C NMR (CDCl₃) δ 152.6, 129.4, 128.6, 127.4,

124.0, 122.8, 120.3, 118.4, 97.4, 94.2, 92.2, 90.5, 73.0, 71.9, 71.7, 65.1,

63.4, 46.6, 38.5, 34.4, 25.6, 25.5, 18.1, 17.4, 15.0, -1.6, -4.4, -4.6;

FTIR (CDCl₃) ν 3559 (m), 3491 (m), 2956 (s), 2255 (m), 1697 (s),

1502 (s), 1395 (s), 1314 (s) cm^-1; HRFABMS calcd for C₃₂H₄₃NO₆Si₂

593.2629, found 593.2631.

was then warmed to -20 °C (CCl₄/CO₂) and stirred for 1.5 h. After

which time, the reaction was diluted with H₂O (30 mL) and EtOAc

(100 mL). The organic layer was washed with 1.0 N HCl (1 × 20

mL), saturated NaHCO₃(1 × 20 mL), and saturated brine (1 × 20

mL). The organics were then dried (Na₂SO₄), filtered, and concentrated

in vacuo. Purification by column chromatography (SiO₂, 6:1 hexanes/

EtOAc) yielded 232 mg (95%) of the triflate 64 as an amorphous white

1

solid: mp 71-75 °C dec; H NMR (400 MHz, C₆D₆) (330 K) δ 7.30

(br s, 1H), 7.04 (d, 2.6 Hz, 1H), 6.79 (dd, J ) 8.7, 2.5 Hz, 1H), 6.15

(br s, 1H), 5.30 (d, J ) 7.1 Hz, 1H), 5.02 (s, 2H), 4.35(m, 1H), 4.25

(m, 1H), 3.64 (d, J ) 4.2 Hz, 1H), 3.38 (d, J ) 12.6 Hz, 1H), 2.41 (br

s, 1H), 1.28 (d, J ) 6.4 Hz, 3H), 1.02 (s, 9H), 0.89 (t, J ) 5.8 Hz,

2H), 0.15 (s, 3H), 0.14 (s, 3H), -0.31 (br s, 9H); ¹³C NMR (C₆D₆) δ

152.7, 129.9, 128.4, 127.7, 123.5, 122.8, 120.2, 118.6, 117.0, 96.6,

94.2, 92.6, 90.7, 89.9, 72.6, 69.6, 64.8, 63.0, 46.6, 34.7, 34.2, 25.4,

25.3, 17.9, 17.3, 13.9, -2.12, -4.7, -4.9; FTIR (C₆D₆) ν 3502 (s),

3300 (w), 2956 (s), 1702 (s), 1502 (s), 1411 (s), 1314 (s)1251 (s) cm^-1

HRFABMS calcd for C₃₃H₄₂F₃NO₈SSi₂725.2121, found 725.2126.

;

Keto-Triflate (65). The triflate 64 (232 mg, 0.326 mmol, 1 equiv)

was dissolved in CH₂Cl₂(12 mL), and Dess-Martin periodinane (415

mg, 0.978 mmol, 3.0 equiv) was added. After 3 h, the reaction was

diluted with saturated NaHCO₃(30 mL) and EtOAc (100 mL) and

stirred vigorously until the layers became clear. The layers were then

separated, and the organic layer was washed with saturated brine, dried

(Na₂SO₄), filtered, and concentrated. Purification by column chroma-

tography (SiO₂, 10:1 hexanes/EtOAc) yielded 220 mg (95%) of the

keto-triflate 65 as a yellow foam. The instability of 65 required that

the compound be stored frozen in benzene for prolonged intervals, and

extensive characterization was not accomplished: ¹H NMR (400 MHz,

CDCl₃) δ 6.81 (d, J ) 2.5 Hz, 1H), 5.90 (dd, J ) 9.1, 1.1 Hz, 1H),

5.83 (dd, J ) 9.0, 1.1 Hz, 1H), 5.55 (d, J ) 7.5 Hz, 1H), 4.42 (s, 1H),

4.25 (m, 2H), 3.10 (q, J ) 7.5, 2.1 Hz, 1H), 1.69 (d, J ) 7.5 Hz, 1H),

0.95 (s, 9H), 0.19 (s, 6H), 0.00 (br s, 9H); CI (NH₃) m/z ) 724 [M +

H] (45%).

Ketone (66). A dry flask was charged with anhydrous CrCl₂(206

mg, 1.69 mmol, 6.0 equiv) and THF (5 mL). To this suspension was

added keto-triflate 65 (200 mg, 0.281 mmol, 1 equiv) dropwise as a

solution in THF (10 mL). The reaction turned from green to a dark

green-grey color, and after 2 h, the reaction was poured into a vigorously

stirring solution of saturated NH₄Cl (10 mL) and Et₂O (70 mL). After

5 min, the reaction layers were separated, and the organic layer was

washed with saturated NH₄Cl (10 mL), H₂O (2 × 20 mL), saturated

brine (10 mL), then dried (Na₂SO₄), filtered, and concentrated.

Purification by column chromatography (SiO₂, 12:1 hexane/EtOAc)

yielded 121 mg (75%) of the ketone 66 as an amorphous solid: mp

1

70-72 °C dec; H NMR (400 MHz, CDCl₃) δ 6.87 (d, J ) 2.6 Hz,

1H), 6.78 (dd, J ) 8.8, 2.6 Hz, 1H), 5.75 (dd, J ) 9.7, 1.3 Hz, 1H),

5.69 (dd, J ) 9.7, 1.4 Hz, 1H), 5.68 (br s, buried, 1H), 4.25 (m, 1H),

4.19 (s, 1H), 3.14 (m, 1H), 2.84 (dd, J ) 11.1, 6.5 Hz, 1H), 2.48 (dd,

J ) 11.1, 3.0 Hz, 1H), 1.50 (d, J ) 6.6 Hz, 1H), 1.03 (bt, buried, 1H),

0.97 (s, 9H), 0.18 (s, 3H), 0.16 (s, 1H), 0.00 (br s, 3H); FTIR (C₆D₆)

ν 2957 (s), 1700 (s), 1611 (m), 1501 (s), 1313 (s), 1250 (s) cm^-1

;

HRFABMS calcd for C₃₂H₄₁NO₅Si₂575.2523, found 575.2518.

Enol (67). A recovery flask was charged with anhydrous MgBr₂

(114 mg) and anhydrous CH₃CN (2 mL). Then, the flask was charged

with ketone 66 (64 mg, 0.113 mmol, 1 equiv) in CH₃CN. The reaction

was purged with dry CO₂for 15 min, then, under an atmosphere of

CO₂, Et₃N (473 µL, 3.39 mmol, 30.0 equiv) was added at once. The

reaction was stirred under an atmosphere of dry CO₂for 2.5 h and

concentrated in vacuo. The crude â-keto acid product was diluted with

Et₂O (40 mL) and 1 N HCl (15 mL), the layers were separated, and

the organic layer was washed with H₂O (1 × 20 mL) and saturated

brine (1 × 10 mL), dried (Na₂SO₄), filtered, and concentrated. The

crude acid was azetroped from benzene (3×) and dissolved in THF (7

i

mL). To the reaction was added PrNEt₂(1.72 mL, 22.6 mmol, 200

equiv) followed by MOMCl (200 mL, 1.13 mmol, 10 equiv) dropwise

slowly. After 3 min, TLC indicated that the starting material acid was

fully consumed, and the reaction was diluted with saturated NaHCO₃

(30 mL) and EtOAc (100 mL). The layers were separated, and the

organic layer was washed with saturated brine (20 mL), dried (MgSO₄),

filtered, and concentrated. Purification by column chromatography

(12:1 hexane/EtOAc) yielded 45 mg (61%) of the â-keto ester 67 as

Triflate (64). The diol 63 (200 mg, 0.341 mmol, 1.0 equiv) was

dissolved in CH₂Cl₂(15 mL) and subsequently cooled to -78 °C.

Anhydrous pyridine (1.11 mL, 13.64 mmol, 40.0 equiv) was then added,

followed by Tf₂O (103 µL, 0.613 mmol, 1.80 mmol). The reaction

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9523

an amorphous solid: mp 71-75 °C dec; ¹H NMR (400 MHz, C₆D₆) δ

12.86 (br s, 1H), 7.34 (br s, 1H), 7.09 (d, J ) 2.5 Hz, 1H), 6.80 (dd,

J ) 8.7, 2.6 Hz, 1H), 6.18 (br s, 1H), 5.12 (d, J ) 5.9 Hz, 1H), 5.09

(dd, J ) 6.0, 1.1 Hz, 1H), 5.06 (dd, J ) 6.0, 1.1 Hz, 1H), 4.77 (d, J

) 5.9 Hz, 1H), 4.27 (s, 1H), 4.18 (m, 2H), 3.77 (br s, 1H), 2.96 (s,

3H), 1.41 (br s, 3H), 0.96 (s, 9H), 0.82 (t, J ) 8.0 Hz, 2H), 0.10 (s,

3H), 0.09 (s, 3H), -0.21 (br s, 3H); ¹³C NMR (100.6 MHz, C₆D₆) δ

171.3, 165.6, 152.8, 130.8, 128.3, 123.5, 123.2, 119.7, 118.2, 102.0,

98.1, 96.0, 91.1, 90.6, 89.3, 71.6, 64.6, 63.2, 56.9, 47.8, 35.3, 33.8,

25.5, 18.7, 18.0, 17.2, -2.1, -4.7, -4.9; FTIR (C₆D₆) ν 2957 (s), 1699

(s), 1666 (s), 1613 (m), 1500 (s), 1391 (s), 1278 (s) cm^-1; HRFABMS

calcd for C₃₅H₄₅NO₈Si₂663.2684, found 663.2690.

Vinylogous Carbonate (68). The enol 67 (47 mg, 0.0718 mmol, 1

equiv) was dissolved in dry MeOH (5 mL) and cooled to 0 °C. To

this solution was added CH₂N₂(0.35 M in Et₂O) until the yellow color

persisted for 1.2 h. After 1.2 h, the excess CH₂N₂was quenched with

a few drops of an AcOH/Et₂O solution, and the reaction was

concentrated. Purification by column chromatography (SiO₂, 7:1 f

4:1 hexane/EtOAc) yielded 34 mg (70%) of the vinylogous carbonate

68 as an amorphous solid, mp 70-71 °C. ¹H NMR (400 MHz, C₆D₆)

δ 7.35 (br s, 1H), 7.16 (d, J ) 2.5 Hz, 1H), 6.80 (dd, J ) 8.7, 2.4 Hz,

1H), 6.17 (br s, 1H), 5.24 (J ) 5.9 Hz, 1H), 5.12 (dd, J ) 10.1, 1.0

Hz, 1H), 5.07 (dd, J ) 10.0, 1.1 Hz, 1H), 5.07 (br s, buried, 1H), 4.19

(m, 2H), 4.10 (s, 1H), 3.95 (br s, 1H), 3.42 (s, 3H), 3.10 (s, 3H), 1.47

(br s, 3H), 0.96 (s, 9H), 0.81 (t, J ) 8.4 Hz, 2H), 0.13 (s, 3H), 0.12 (s,

3H), -0.19 (br s, 3H); ¹³C NMR (100.6 MHz, C₆D₆) δ 165.1, 155.9,

152.6, 130.8, 128.3, 123.4, 123.1, 119.5, 118.2, 115.2, 98.8, 96.0, 90.9,

90.2, 89.6, 71.3, 64.6, 63.6, 58.3, 56.8, 47.6, 36.5, 33.7, 25.5, 18.4,

18.0, 17.2, -2.1, -4.6, -4.8; FTIR (C₆D₆) ν 2934 (s), 1703 (s), 1500

(s), 1277 (s); HRFABMS calcd for C₃₆H₄₇NO₈Si₂677.2840, found

677.2835.

Quinone Aminal (69). The enediyne 62 (66 mg, 0.112 mmol, 1.0

equiv) was dissolved in THF (3 mL) and cooled to 0 °C. To this

solution was added HOAc (223 µL, 0.223 mmol, 2.0 equiv) as a 1.0

M solution in THF. Then, 1.0 M TBAF in THF (123 µL, 0.123 mmol,

1.1 equiv) was added dropwise. After 20 min, the reaction was poured

into EtOAc and H₂O, and the organic layer was washed with brine,

dried (Na₂SO₄), filtered, and concentrated. The crude phenol was then

diluted in dry MeOH (4 mL) and cooled to 0 °C, and PhI(OAc)₂(43

mg, 0.134 mmol, 1.2 equiv) was added. After 30 min, the reaction

was poured into EtOAc and H₂O, the organic layer was washed with

brine, dried (MgSO₄), filtered, and concentrated. Purification by column

chromatography (SiO₂, 1:1 EtOAc/hexane) provided 37 mg (65%) of

the quinone aminal 69 as a clear glass. ¹H NMR (400 MHz, CDCl₃)

δ 6.61 (d, J ) 1.9 Hz, 1H), 6.32 (dd, J ) 10.1, 1.9 Hz, 1H), 5.87 (d,

J ) 9.8 Hz, 1H), 5.79 (d, J ) 9.8 Hz), 5.42 (t, J ) 3.4 Hz, 1H), 5.28

(dd, J ) 10.0, 3.4 Hz, 1H), 3.82 (s, 3H), 3.71 (d, J ) 3.5 Hz, 1H),

3.07 (s, 3H), 2.77 (m, 1H), 2.10 (s, 3H), 2.03 (s, 3H), 1.33 (d, J ) 7.3

Hz, 1H); FTIR (film) ν 2931 (m), 1742 (s), 1714 (s), 1673 (s), 1641

(s), 1247 (s) cm^-1; EI [M⁺] ) 509 (100%).

Ketone (71). The cyanophthalide 72 (58 mg, 0.266 mmol, 4.1 equiv)

was dissolved in THF (4 mL) and cooled to 0 °C. Then, 0.3 M LDA

(850 µL, 0.26 mmol, 4.0 equiv) was added dropwise. To the yellow

solution was added quinone aminal 69 (33 mg, 0.65 mmol, 1.0 equiv)

as a solid at once. After 3 h at room temperature, the reaction was

poured into EtOAc and H₂O, and the organic layer washed with brine,

dried (MgSO₄), filtered, and concentrated. Purification by column

chromatography (SiO₂, 1:1 EtOAc/hexane) provided 12 mg (35%) of

the adduct 71 as a clear oil. The product was isolated as an inseperable

mixture of diastereomers. The ¹H NMR spectra is shown in the

supporting information. EI [M⁺] ) 539 (55%).

6.42 (d, J ) 2.0 Hz, 1H), 6.04 (dd, J ) 10.0, 2.0 Hz, 1H), 5.31 (d, J

) 5.9 Hz, 1H), 5.15 (m, 3H), 5.04 (s, 1H), 3.81 (q, J ) 7.3, 2.5 Hz,

1H), 3.61 (s, 1H), 3.48 (s, 3H), 3.20 (s, 3H), 1.38 (d, J ) 7.3 Hz, 3H);

¹³C NMR (100.6 MHz, CDCl₃) δ 185.3, 165.2, 161.0, 154.7, 140.7,

136.8, 131.3, 128.3, 127.5, 127.0, 126.7, 123.5, 122.8, 115.5, 108.5,

96.9, 90.5, 88.1, 63.5, 58.2, 56.8, 35.99, 31.5, 18.5; FTIR (CDCl₃) ν

1703 (s), 1652 (s), 1450 (m) cm^-1; HRFABMS calcd for C₂₄H₁₉NO₆

417.1212, found 417.1218.

Homophthalic Ester (76). A solution of 2,2,6,6-tetramethylpip-

eridine (8.78 mL, 52.0 mmol, 2.45 equiv) in THF (200 mL) was cooled

to -78 °C. To this solution was added 2.48 M nBuLi (21 mL, 51.1

mmol, 2.4 equiv) dropwise. After 10 min, dimethylmalonate (2.91 mL,

25.5 mmol, 1.2 equiv) was added slowly dropwise. After 10 min, the

bromoarene 75 (5.9 g, 21.3 mmol, 1.0 equiv) was added dropwise via

syringe pump as a solution in THF (100 mL). The reaction took on a

purple color as the bromide was added. After complete addition of

the bromide and an additional 10 min at -78 °C, the reaction was

quenched by the addition of H₂O (100 mL) and the reaction was diluted

with EtOAc (700 mL). The layers were separated, and the organic

layer was washed with saturated brine (100 mL), dried (MgSO₄),

filtered, and concentrated. Purification by column chromatography

(SiO₂, 3:1 f 2:1 hexane/EtOAc) provided 4.96 g (71%) of the

homophthalic ester 76 as a clear oil. ¹H NMR (400 MHz, CDCl₃) δ

7.05 (d, J ) 8.1 Hz, 1H), 7.03 (d, J ) 8.1 Hz, 1H), 5.08 (d, J ) 5.8

Hz, 4H), 3.85 (s, 3H), 3.64 (s, 2H), 3.63 (s, 3H), 3.42 (s, 3H), 3.40 (s,

3H); ¹³C NMR (100.6 MHz, CDCl₃) δ 170.9, 167.3, 150.1, 148.7, 126.0,

122.5, 116.4, 115.5, 95.4, 94.8, 55.8, 52.0, 38.4, 32.8, 31.1; FTIR

(CDCl₃) ν 2953 (s), 2256 (s), 1728 (s), 1599 (s), 1481 (s) cm^-1

;

HRFABMS calcd for C₁₅H₂₀O₈328.1158, found 328.1159.

Homophthalic Acid (77). The homophthalic ester 76 (2.80 g, 10.45

mmol, 1 equiv) was dissolved in MeOH (100 mL), and to this was

added solid KOH (13.80 g, 209 mmol, 20 equiv) and H₂O (50 mL).

The reaction was warmed to reflux for 2 h. After which time, the

reaction was allowed to cool to room temperature and concentrated.

The resulting oil was partitioned between Et₂O (200 mL) and H₂O (200

mL), and the pH was adjusted to 1 by addition of 1 N HCl. The layers

were separated, and the aqueous layer was extracted again with Et₂O

(100 mL). The organic layer was washed with brine (50 mL), dried

(MgSO₄), filtered, and concentrated. The homophthalic acid 77

precipitated from Et₂O as a white powder to provide 2.6 g (99%) mp

) 116 °C. ¹H NMR (400 MHz, acetone-d₆) δ 7.08 (d, J ) 8.7 Hz,

1H), 7.02 (d, J ) 8.7 Hz, 1H), 5.11 (s, 2H), 5.09 (s, 2H), 3.66 (s, 2H),

3.39 (s, 3H), 3.37 (s, 3H); ¹³C NMR (100.6 MHz, acetone-d₆) δ 175.8,

172.4, 155.5, 153.5, 128.0, 121.0, 120.3, 100.4, 99.9, 60.4, 60.2, 37.38;

FTIR (KBr) ν 3150 (s), 1705 (s), 1485 (s) cm^-1; HRFABMS calcd for

C₁₃H₁₆O₈300.0845, found 300.0841.

Homophthalic Anhydride (78). The homophthalic acid 77 (313

mg, 1.05 mmol, 1.0 equiv) was dissolved in CH₂Cl₂(5 mL), and to

this solution was added (trimethylsilyl)ethoxyacetylene (208 mL, 1.47

mmol, 1.4 equiv). After having been stirred for 2 h, the reaction was

concentrated to yield 294 mg (100%) of the anhydride 78 as a white

1

solid: mp ) 70-73 °C; H NMR (400 MHz, CDCl₃) δ 7.39 (d, J )

8.5 Hz, 1H), 7.18 (d, J ) 8.5 Hz, 1H), 5.31 (s, 2H), 5.18 (s, 2H), 4.01

(s, 2H), 3.53 (s, 3H), 3.47 (s, 3H); FTIR (CH₂Cl₂) ν 3060 (s), 2960

(s), 1799 (s), 1759 (s), 1592 (m), 1487 (s), 1092 (s) cm^-1; HRFABMS

calcd for C₁₃H₁₄O₇282.0739, found 282.0743.

Dynemicin A (1). The homophthalic anhydride 78 (49 mg, 0.173

mmol, 6.0 equiv) was dissolved in THF (2.5 mL) and cooled to 0 °C.

Then, a 1.0 M solution of LHMDS (171 µL, 0.171 mmol, 5.9 equiv)

was added dropwise, and the solution immediately became bright

yellow. After 35 min, the quinone imine 74 (12 mg, 0.029 mmol, 1.0

equiv) was added as a solution in THF (1 mL). The reaction slowly

became a dark red-brown color, and, after 35 min, TLC indicated that

no starting material remained. At this time, PhI(OCOCF₃)₂(93 mg,

0.218 mmol, 7.5 equiv) was added as a solid at once. Upon addition,

the reaction turned a red-violet color, and, after 5 min, the reaction

was poured into saturated NaHCO₃(15 mL) and EtOAc (30 mL). The

organic layer was separated and washed with saturated NaHCO₃(15

mL), brine (10 mL), dried (Na₂SO₄), filtered, and concentrated. The

crude napthalenol 80 was dissolved in THF (6 drops) and exposed to

daylight and air. After 20 h, TLC indicated that no napthalenol was

present, and a new more polar spot had appeared. At this time, the

Quinone Imine (74). The vinylogous carbonate 68 (41 mg, 0.0613

mmol, 1.0 equiv) was dissolved in THF (4 mL) and cooled to 0 °C.

Then, 1.0 M TBAF (171 mL, 0.171 mmol, 3.0 equiv) was added

dropwise. After 3.5 h at 0 °C, PhI(OAc)₂(102 mg, 0.319 mmol, 5.2

equiv) was added as a solid at once. After an additional 30 min at 0

°C, the reaction was poured into saturated NaHCO₃(20 mL) and

extracted with EtOAc (30 mL). The organic layer was washed with

brine (10 mL), dried (Na₂SO₄), filtered, and concentrated. Purification

by column chromatography (SiO₂, 2:1 hexane/EtOAc) provided 15.5

mg (60%) of the quinone imine 74 as a pale yellow powder: mp 75-

78 °C dec; ¹H NMR (400 MHz, CDCl₃) δ 6.82 (d, J ) 10.0 Hz, 1H),

9524 J. Am. Chem. Soc., Vol. 118, No. 40, 1996

Shair et al.

protected anthraquinone 81 was dried over Na₂SO₄, filtered, and

concentrated. The crude violet-red colored product was then dissolved

in Et₂O (3 mL) and cooled to 0 °C. A 0.3 M solution of MgBr₂(96

µL, 0.29 mmol, 10 equiv) was added dropwise. The reaction became

a deep blue color after 2 h, and the reaction was allowed to warm to

room temperature for 10 h. After which time, the reaction was poured

into EtOAc (100 mL) and H₂O (20 mL), and the aqueous layer was

extracted (3×) with EtOAc (20 mL). The organic layer was washed

with brine (15 mL), dried (Na₂SO₄), filtered, and concentrated.

Purification by column chromatography (Sephadex LH-20 20% CH₃-

CN/MeOH) yielded 2.3 mg (15%) of (()-dynemicin A (1) as a deep

blue-violet solid, mp ) 212-215 °C (dec): ¹H NMR (400 MHz, d₆

DMSO) δ 13.12 (s, 1H, aryl OH), 12.71 (br s, 1H, aryl OH), 12.15 (br

s, 1H, aryl OH), 9.85 (s, 1H, H-1, NH), 8.04 (s, 1H, H-10, aryl H),

7.39 (d, J ) 9.2 Hz, 1H, H-16 or 17, aryl H), 7.34 (d, J ) 9.2 Hz, 1H,

H-16 or 17, aryl H), 6.08 (d, J ) 9.9 Hz, 1H, H-25 or 26,

CtCsCHdC), 6.06 (d, J ) 9.9 Hz, 1H, H-25 or 26, CtCsCHdC),

5.08 (d, J ) 3.6 Hz, 1H, H-2, NCH), 4.89 (s, 1H, H-7, CtCsCH),

3.81 (s, 3H, OCH₃), 3.57 (q, J ) 7.3 Hz, 1H, H-4, CHCH₃), 1.25 (d,

J ) 7.1 Hz, 3H, H-29, CHCH₃); FTIR (neat) ν 3683-2733 (br, m-s),

3415 (m), 2926 (w), 1663 (s), 1628 (s), 1470 (s), 1270 (s), 1038 (m);

HRFABMS calcd for C₃₀H₁₉NO₉537.1058, found 537.1060; R_f(0.45)

(vis) 10% MeOH/CH₂Cl₂.

Diacetate (84). A recovery flask was charged with the acetate 83⁴⁵

(4.36 mmol, 1.0 g) in CH₂Cl₂(20 mL). The solution was cooled to 0

°C, and mCPBA (42%) (6.1 mmol, 2.51 g) was added. The solution

was stirred for 3 h under Ar and was slowly warmed to room

temperature. Upon completion of the reaction (CH₃)₂S (0.4 mL) was

added, and the resulting solution was stirred for 30 min. NaHCO₃was

then added, and the organic layer was extracted with CH₂Cl₂. The

organic layer was separated, dried (MgSO₄), filtered, and concentrated.

The N-oxide was obtained as a crude oil.

The N-oxide was then subjected to the action of Ac₂O (5 mL), and

this solution was stirred at 70 °C for 1 h. The reaction was cooled to

room temperature and was redissolved in CH₂Cl₂. Saturated NaHCO₃

was added, and the organic layer was separated. The solution was

washed with brine, dried (MgSO₄), filtered, and concentrated. Purifica-

tion by column chromatography (SiO₂, 1:1 EtOAc/hexanes) provided

the diacetate 84 as a yellow solid (1.32 g, 84%): mp 95.4 °C; ¹H NMR

(400 MHz, CDCl₃) δ 8.71 (s, 1H), 8.09 (d, 1H, J ) 9.0 Hz), 7.46 (d,

1H, J ) 2.4 Hz), 7.39 (d, 1H, J ) 2.4 Hz, 9.0 Hz), 6.48 (d, 1H, J )

3.5 Hz), 2.8-3.2 (m, 2H), 2.0-2.3 (m, 4H), 2.36 (s, 3H), 2.08 (s, 3H);

FTIR (CDCl3) ν 2951 (m), 1760 (s), 1730 (s), 1507 (s), 1370 (s) 1014

(s) cm^-1; HRFABMS calcd for C₁₇H₁₇NO₄200.1157, found 200.1013.

Epoxide (86). A recovery flask was charged with a solution of the

diacetate 84 (6.5 mmol, 2.0 g) in THF (30 mL). The solution was

cooled to -78 °C and was treated with ethynylmagnesium bromide

(0.5 M in THF, 6.5 mmol, 13.0 mL). The solution was stirred for 30

min after which time TeocCl (7.8 mmol, 1.38 mL) was added at -78

°C, and the solution was gradually warmed to 0 °C. An additional 5

equiv of ethynylmagnesium bromide and an additional 2 equiv of

TeocCl were added over the course of 5 h until the reaction was

completed. Water and saturated NaHCO₃were then added, and the

organic layer was extracted with EtOAc, separated, dried (Na₂SO₄),

filtered, and concentrated to yield a crude yellow oil.

This crude acetylide 85 was redissolved in CH₂Cl₂(15 mL) and

was treated with mCPBA (95%, 18.5 mmol, 3.36 g) and was stirred

for 2-3 days at room temperature, monitoring the reaction periodically

by ¹H NMR. Upon completion, (CH₃)₂S (0.8 mL) was added, and the

solution was stirred for 30 min. Saturated NaHCO₃was added, and

the organic layer was extracted with CH₂Cl₂, dried, filtered, and

concentrated in vacuo. Purification by column chromatography (SiO₂,

5:1 hexanes/EtOAc) yielded the amber colored epoxide 86 as an oil

(1.24 g, 42% over two steps): ¹H NMR (400 MHz, CDCl₃) δ 7.35 (br

s, 1H), 7.05 (dd, 1H), 7.05 (buried d, 1H), 5.75 (dd, 1H, J ) 6.1 Hz,

9.0 Hz), 5.55 (br s, 1H), 4.25 (m, 2H), 2.19 (s, 3H), 2.17 (s, 3H), 1.4-

2.1 (m, 6H), 0.0 (br s, 3H); FTIR (CH2Cl2) ν 3450 (m), 2955 (s),

1738 (s), 1508 (s), 1426 (s), 1398 (s), 1315 (s), 1287 (s), 1234 (s),

1188 (s), 934 (s), 861 (s), 839 (s) cm^-1; HRFABMS calcd for C₂₃H₃₁-

NO₇Si 461.1869, found 461.1799.

(14.0 mmol, 0.91 g). The solution was stirred initially at -78 °C and

was then gradually warmed to 0 °C continuing to stir for 2 h. H₂O

was then added, and the product was extracted with EtOAc, separated,

dried (Na₂SO₄), filtered, concentrated, and azeotroped (3×) from

benzene. Column chromatography (SiO₂, 10:1 hexanes/EtOAc) yielded

the phenolic alcohol as an oily residue.

The phenolic alcohol was immediately redissolved in dry CH₂Cl₂

(100 mL), and TBSCl (70 mmol, 10.5 g) and imidazole (70 mmol,

4.76 g) were added, and the reaction was stirred at room temperature

for 1 h. Upon completion of the reaction H₂O and EtOAc were added,

the organic layer was separated, dried (Na₂SO₄), filtered, and concen-

trated. The resulting oil was azeotroped (3×) from benzene and was

subsequently chromatographed (SiO₂, 10:1 hexanes/EtOAc) to yield

the siloxy-acetate 87 as a yellow foam (5.2 g, 69% over two steps):

¹H NMR (CDCl₃) δ 7.18 (br s, 1H), 6.78 (d, 1H, J ) 8.8 Hz), 6.75 (d,

1H, J ) 2.5 Hz), 5.76 (dd, 1H, J ) 6.01 Hz, 8.6 Hz), 5.56 (br s, 1H),

4.26 (t, 2H, 8.3 Hz), 2.10 (s, 3H), 1.6-2.75 (m, 6H), 1.0 (s, 9H), 0.18

(s, 3H), 0.17 (s, 3H), 0.0 (s, 9H); FTIR (CH2Cl₂) ν 3299 (w), 2955

(s), 1732 (s), 1658 (sh), 1500 (s), 1372 (s), 1315 (s), 1287 (s), 1234

(s), 1180 (s), 1064 (s), 1045 (s), 934 (s), 861 (s) cm^-1; HRFABMS

calcd for C₂₇H₄₃NO₆Si₂533.2628, found 533.2645.

Ketone (89). To a solution of the siloxy acetate 87 (7.7 mmol, 15.4

mL) in CH₂Cl₂(100 mL) at -78 °C was added a solution of DIBAL-H

(15.4 mmol, 15.4 mL). The solution was stirred at -78 °C for 2 h,

and MeOH was then added dropwise at -78 °C until the bubbling had

stopped. The solution was then warmed to 25 °C, and ether and

Rochelle’s salt were added. The solution was stirred for 30 min until

the layers became clear. The organic layer was separated, washed with

brine (2×), dried (Na₂SO₄), filtered, concentrated, and azeotroped (3×)

from benzene to yield the oily yellow secondary alcohol 88.

The crude oil 88 was redissolved immediately in CH₂Cl₂(100 mL),

and 4Å dry molecular sieves (3.4 g) and pyridinium chlorochromate

(20.4 mmol, 4.4 g) were added. The solution was then stirred under

Ar at 25 °C for 3 h. The solution was then diluted with ether and

filtered through Celite and SiO₂, flushing with generous amounts of

ether. The solution was concentrated to a dark oil and chromatographed

(SiO₂10:1 hexanes/EtOAc) to yield the yellow siloxy ketone 89 as a

foam (1.78 g, 46% over two steps). ¹H NMR (CDCl₃) δ 7.86 (d, 1H,

J ) 2.5 Hz), 7.25 (1H), 7.16 (br s, 1H), 6.81 (d, 1H, J ) 2.5 Hz, 8.6

Hz), 5.65 (br s, 1H), 4.24 (t, 2H, J ) 8.3 Hz), 2.6-2.7 (m, 1H), 2.51-

2.58 (m, 1H), 2.21-2.29 (m, 2H), 1.88-2.04 (m, 2H), 1.0 (s, 9H),

0.24 (s, 3H), 0.18 (s, 3H), 0.0 (s, 9H); FTIR (CH₂Cl₂) ν 3300 (m),

2956 (s), 2931 (m), 2879 (m), 2858 (m), 1715 (s), 1494 (s), 1313 (s),

1288 (s), 1247 (s) cm^-1; HRFABMS calcd for C₂₆H₃₉NO₅Si₂501.2366,

found 501.2371.

Enediyne (90). In a 25 mL recovery flask was placed Pd(PPh₃)₄

(0.0095 mmol, 0.010 g) in dry degassed (3 h) benzene (1 mL). nBuNH₂

(0.76 mmol, 0.076 mL) was then added followed by the chloroenyne

(0.62 mmol, 0.098 mL). A solution of the acetylene 89 (0.19 mmol,

0.10 g) in benzene (3 mL) was then added to the solution followed by

freshly purified CuI (0.038 mmol, 0.0072 g), and the solution was stirred

at 25 °C overnight. The solution was then poured into saturated

NaHCO₃and was extracted with CH₂Cl₂. The organic layer was

separated, dried (Na₂SO₄), and concentrated. Column chromatography

(SiO₂, 100% hexanes, 20:1 hexanes/EtOAc, 15:1 hexanes/EtOAc)

yielded the foamy yellow solid 90 (0.090 g, 75%): ¹H NMR (CDCl₃)

δ 7.85 (d, 1H, J ) 2.74 Hz), 7.15 (br s, 1H), 6.80 (d, 1H, J ) 6.14

Hz), 5.78 (d, 1H, J ) 11.1 Hz), 5.63 (d, 1H, J ) 11.1 Hz), 4.24 (t, 2H,

J ) 8.3 Hz), 2.6-2.7 (m, 2H), 2.2-2.3 (m, 2H), 1.8-2.0 (m, 2H),

1.02 (s, 9H), 0.24 (s, 3H), 0.23 (s, 3H), 0.21 (s, 9H), 0.0 (s, 9H); FTIR

(CH₂Cl₂) ν 2957 (s), 2931 (s), 2897 (s), 2144 (w), 1696 (s), 1609 (m),

1578 (m), 1495 (s), 1391 (s), 1117 (s), 938 (s), 860 (s) cm^-1

HRFABMS calcd for C₃₄H₄₉NO₅Si₃636.0298, found 636.0289.

;

Cyclic Enediyne (92). In a 100 mL recovery flask was placed 90

(2.2 mmol, 1.4 g) in 1:1:1 THF/EtOH/H

₂O (50 mL). AgNO₃(1.1

mmol, 0.19 g) was then added, and the reaction was stirred for 2 h at

25 °C. H₂O was then added, and the solution was extracted with ether.

The organic layer was dried (Na₂SO₄), filtered, concentrated, and

chromatographed to yield the yellow oil 91 (1.24 g, 90%).

The deprotected enediyne 91 (1.96 mmol, 1.11 g) was then dissolved

immediately in freshly distilled toluene (200 mL), and the solution was

cooled to -78 °C. Freshly prepared LDA (2.9 mmol, 4.99 mL) was

Acetate (87). To a solution of the epoxide 86 (14.0 mmol, 6.8 g)

in freshly distilled MeOH (100 mL) at -78 °C was added dry KCN

Synthesis of Dynemicin A

J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9525

then added slowly dropwise. The reaction was stirred under Ar at -78

°C for 1 h. NH₄Cl was then added, the solution was warmed to room

temperature, and H₂O was added. The solution was extracted with

EtOAc, and the organic layer was separated, dried, and concentrated.

Column chromatography (SiO₂, 10:1 hexanes/EtOAc) yielded the cyclic

enediyne 92 as a yellow oil (0.78 g, 71%): ¹H NMR (CDCl₃) δ 8.10

(d, 1H, J ) 2.7 Hz), 6.75 (dd, 1H, J ) 2.7 Hz, 8.7 Hz), 5.80 (d, 1H,

J ) 10 Hz), 5.65 (dd, 1H, J ) 1.6 Hz, 10 Hz), 4.25 (t, 2H, J ) 8.1

Hz), 2.27 (m 1H), 2.15 (m, 3H), 1.88 (m, 1H), 1.73 (m, 1H), 0.96 (s,

9H), 0.20 (s, 3H), 0.19 (s, 3H), 0.0 (s, 9H); FTIR (CH₂Cl₂) ν 3574

(w), 2957 (m), 2927 (m), 2872 (w), 1798 (m), 1650 (s), 1271 (s), 1106

(m) cm^-1; HRFABMS calcd for C₃₁H₄₁NO₅Si₂563.8471, found

563.8475.

Quinone Imine (93). To a solution of the cyclic enediyne 92 (0.35

mmol, 0.20 g) in THF (10 mL) cooled to 0 °C was added TBAF (1.0

M, 1.76 mmol, 1.76 mL). Immediately the color of the solution

changed from light yellow to bright red. The solution was then stirred

at O °C for 2 h. After removal of the Teoc and TBS groups had been

completed as monitored by thin layer chromatography, PhI(OAc)₂(2.47

mmol, 0.796 g) was added, and the solution was stirred at 0 °C for 1

h. A solution of saturated NaHCO₃was then added, and the solution

was extracted with EtOAc. The organic layer was dried (Na₂SO₄) and

concentrated. Column chromatography (SiO₂4:1 hexanes/EtOAc, 3:2

EtOAc/hexanes) yielded the quinone imine as a yellow oil. Recrys-

tallization from ether/hexanes yielded the quinone imine 93 as a

powdery yellow solid (0.49 g, 49%), mp >125 C: ¹H NMR (CDCl₃)

δ 7.69 (d, 1H, J ) 2.0 Hz), 7.12 (d, 1H, J ) 10.0 Hz), 6.51 (dd, 1H,

J ) 2.0 Hz, 10.0 Hz), 5.94 (d, 1H, J ) 10.0 Hz), 5.88 (d, 1H, J ) 10.0

Hz), 5.13 (s, 1H), 1.3-2.2 (m, 6H); FTIR (CH₂Cl₂) ν 3577 (w), 2966

(s), 2930 (m), 2858 (w), 1698 (s), 1496 (s), 1471 (m), 1392 (s), 1314

(s), 1289 (s), 1249 (s) 1199 (s) cm^-1; HRFABMS calcd for C₁₉H₁₃-

NO₂287.3209, found 287.3210.

Acknowledgment. This work was supported by NIH Grant

CA 28824. A predoctoral fellowship from Memorial Sloan-

Kettering Cancer Center to M.D.S. is gratefully acknowledged.

We acknowledge the spectral laboratory at Columbia University

as well as the spectral, pharmacology, and preparative core

laboratories of the Sloan-Kettering Institute for their assistance.

We also acknowledge Dr. Dolatrai M. Vyas of Bristol-Myers

Squibb Pharmaceutical Research Institute for supplying us with

an authentic sample of dynemicin A. We dedicate this paper

to Professor Harold Moore of the University of California, Irvine

for his early and insightful studies in the field of bioreductive

alkylation as well as to Professors Satoru Masamune of the

Massachusetts Institute of Technology and Robert Bergman of

the University of California, Berkeley for their pioneering

studies in the chemistry of cyclic enediynes.

JA960040W

Article Doi

The total synthesis of dynemicin A leading to development of a fully contained bioreductively activated enediyne prodrug

DOI: 10.1021/ja960040w

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja960040w

Full text of DOI:10.1021/ja960040w

Products guided by the article

R&D Labs maybe for 168421-50-9

Relevant to this article

Hot Product