Total synthesis of (+)-duocarmycin A, epi-(+)-duocarmycin A and their unnatural enantiomers: Assessment of chemical and biological properties

Article abstract of DOI:10.1021/ja962431g

Full details of an enantioselective total synthesis of (+)-duocarmycin A (1) are described in which a solution to the control of the relative and absolute stereochemistry of the remote stereocenters is provided. Catalytic asymmetric dihydroxylation of 15 was employed to introduce the absolute stereochemistry required for the activated cyclopropane, and a diastereoselective Dieckmann-type condensation of 61 was employed to control the absolute stereochemistry of the C6 quaternary center. The complementary diastereoselectivity of a thermodynamic versus kinetic condensation of 61 permitted the divergent synthesis of (+)-duocarmycin A or epi-(+)-duocarmycin A from common intermediates. Final introduction of the reactive cyclopropane was accomplished by transannular spirocyclization of the mesylate 44 upon treatment with base or directly from the corresponding free alcohol itself, duocarmycin D₁ (42), upon Mitsunobu activation. Notably, the asymmetric dihydroxylation of 15 employing (DHQD)₂-PHAL/(DHQ)₂-PHAL was found to proceed with a reverse enantioselectivity than predicted from established models. Employing this approach, the key partial structures (+)-N-BOC-DA (67) and (+)-6-epi-N-BOC-DA (71) and their unnatural enantiomers were also prepared, and a study of their acid-catalyzed solvolysis reactivity, regioselectivity (3:2), and stereochemistry is detailed. Notably, the solvolysis reaction regioselectivity is lower than the characteristic adenine N3 alkylation of duplex DNA, which proceeds with exclusive nucleophilic addition to the least substituted C8 cyclopropane carbon. This may be attributed to the significant destabilizing torsional strain and steric interactions characteristic of the S(N)2 reaction of a large nucleophile that accompany the abnormal addition of adenine when restricted to the minor groove bound orientation of the reactants.

Full text of DOI:10.1021/ja962431g

J. Am. Chem. Soc. 1997, 119, 311-325
311
Total Synthesis of (+)-Duocarmycin A, epi-(+)-Duocarmycin A
and Their Unnatural Enantiomers: Assessment of Chemical and
Biological Properties
Dale L. Boger,* Jeffrey A. McKie, Takahide Nishi, and Tsuyoshi Ogiku
Contribution from the Department of Chemistry, The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, California 92037
ReceiVed July 15, 1996X
Abstract: Full details of an enantioselective total synthesis of (+)-duocarmycin A (1) are described in which a
solution to the control of the relative and absolute stereochemistry of the remote stereocenters is provided. Catalytic
asymmetric dihydroxylation of 15 was employed to introduce the absolute stereochemistry required for the activated
cyclopropane, and a diastereoselective Dieckmann-type condensation of 61 was employed to control the absolute
stereochemistry of the C6 quaternary center. The complementary diastereoselectivity of a thermodynamic versus
kinetic condensation of 61 permitted the divergent synthesis of (+)-duocarmycin A or epi-(+)-duocarmycin A from
common intermediates. Final introduction of the reactive cyclopropane was accomplished by transannular
spirocyclization of the mesylate 44 upon treatment with base or directly from the corresponding free alcohol itself,
duocarmycin D1 (42), upon Mitsunobu activation. Notably, the asymmetric dihydroxylation of 15 employing
(DHQD)2-PHAL/(DHQ)2-PHAL was found to proceed with a reverse enantioselectivity than predicted from established
models. Employing this approach, the key partial structures (+)-N-BOC-DA (67) and (+)-6-epi-N-BOC-DA (71)
and their unnatural enantiomers were also prepared, and a study of their acid-catalyzed solvolysis reactivity,
regioselectivity (3:2), and stereochemistry is detailed. Notably, the solvolysis reaction regioselectivity is lower than
the characteristic adenine N3 alkylation of duplex DNA, which proceeds with exclusive nucleophilic addition to the
least substituted C8 cyclopropane carbon. This may be attributed to the significant destabilizing torsional strain and
steric interactions characteristic of the SN2 reaction of a large nucleophile that accompany the abnormal addition of
adenine when restricted to the minor groove bound orientation of the reactants.
Two independent reports detailed the isolation and structure
determination of the initial members of a new class of
exceptionally potent antitumor antibiotics now including duo-
carmycin A (1), its potent ring open derivatives 2-5, and
duocarmycin SA (6), Figure 1.1-5 The single-crystal X-ray
structure determination4 of duocarmycin C2 (4, pyrindamycin
A) established the relative and absolute configuration of the
remote stereocenters, which through chemical interconver-
X Abstract published in AdVance ACS Abstracts, December 15, 1996.
(1) Yasuzawa, T.; Muroi, K.; Ichimura, M.; Takahashi, I.; Ogawa, T.;
Takahashi, K.; Sano, H.; Saitoh, Y. Chem. Pharm. Bull. 1995, 43, 378.
(2) Takahashi, I.; Takahashi, K.; Ichimura, M.; Morimoto, M.; Asano,
K.; Kawamoto, I.; Tomita, F.; Nakano, H. J. Antibiot. 1988, 41, 1915.
Yasuzawa, T.; Iida, T.; Muroi, K.; Ichimura, M.; Takahashi, K.; Sano, H.
Chem. Pharm. Bull. 1988, 36, 3728. Ichimura, M.; Muroi, K.; Asano, K.;
Kawamoto, I.; Tomita, F.; Morimoto, M.; Nakano, H. J. Antibiot. 1988,
41, 1285. Ogawa, T.; Ichimura, M.; Katsumata, S.; Morimoto, M.;
Takahashi, K. J. Antibiot. 1989, 42, 1299. Ichimura, M.; Ogawa, T.;
Katsumata, S.; Takahashi, K.; Takahashi, I.; Nakano, H. J. Antibiot. 1991,
44, 1045.
(3) Ichimura, M.; Ogawa, T.; Takahashi, K.; Kobayashi, E.; Kawamoto,
I.; Yasuzawa, T.; Takahashi, I.; Nakano, H. J. Antibiot. 1990, 43, 1037.
(4) Ohba, K.; Watabe, H.; Sasaki, T.; Takeuchi, Y.; Kodama, Y.;
Nakazawa, T.; Yamamoto, H.; Shomura, T.; Sezaki, M.; Kondo, S. J.
Antibiot. 1988, 41, 1515.
(5) Boger, D. L.; Johnson, D. S. Angew. Chem., Int. Ed. Engl. 1996, 35,
217. Boger, D. L.; Johnson, D. S. Proc. Natl. Acad. Sci. U.S.A. 1995, 92,
Figure 1.
3642. Boger, D. L. Acc. Chem. Res. 1995, 28, 20. Boger, D. L.
Chemtracts: Org. Chem. 1991, 4, 329. Boger, D. L. Pure Appl. Chem.
1993, 65, 1123. Boger, D. L. Proceed. Robert A. Welch Found. Conf. Chem.
Res., XXXV. Chem. Front. Med. 1991, 35, 137. Boger, D. L. In AdVances
in Heterocyclic Natural Products Synthesis; Pearson, W. H., Ed.; JAI:
Greenwich, CT, 1992; Vol. 2, p 1. For related reviews on CC-1065:
Warpehoski, M. A.; Hurley, L. H. Chem. Res. Toxicol. 1988, 1, 315.
Warpehoski, M. A. In AdVances in DNA Sequence Specific Agents; Hurley,
L. H., Ed.; JAI: Greenwich, 1992; Vol. 1, p 217. Hurley, L. H.; Draves, P.
H. In Molecular Aspects of Anticancer Drug-DNA Interactions; Neidle,
S., Waring, M., Eds.; CRC: Ann Arbor, 1993; Vol. 1, p 89. Aristoff, P. A.
In AdVances in Medicinal Chemistry; JAI: Greenwich, 1993; Vol. 2, p 67.
sions1,10 provided the absolute stereochemistry for 1-5. In the
short time since their disclosure, they have been shown to exert
their biological effects1-4,6-8 through a characteristic sequence-
selective DNA alkylation.5,9-17
Duocarmycin A (1) constitutes the most reactive and chal-
lenging member of this class of agents, and approaches to the
(6) Ishii, S.; Nagasawa, M.; Kariya, Y.; Yamamoto, H.; Inouye, S.;
Kondo, S. J. Antibiot. 1989, 42, 1713.
S0002-7863(96)02431-6 CCC: $14.00 © 1997 American Chemical Society

312 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
control of the relative and absolute stereochemistry of its remote
stereocenters have not been forthcoming. Herein, we report full
details of an enantioselective total synthesis of (+)-duocarmycin
A (1) and epi-(+)-duocarmycin A (7) and their unnatural
enantiomers in which a solution to the control of the remote
stereochemistry is provided.18 The efforts proceeded through
duocarmycin D1 as a key intermediate and complement the
nondiastereoselective synthesis of 1 and 7 described by Terash-
ima and co-workers19 as well as related efforts on duocarmycin
SA.20,21
Scheme 1
Central to the strategy is the ability to prepare both enanti-
omers of the cyclopropane stereochemistry while maintaining
the opportunity to independently control the C6 absolute
stereochemistry, thereby providing access to the natural and
unnatural enantiomers of either 1 or 7. Key elements of the
approach include a diastereoselective Dieckmann-like condensa-
tion for introduction of the C6 quaternary center and a
transannular Ar-3’ spirocyclization1,10,12,21,22 of duocarmycin D1
for introduction of the reactive cyclopropane, a process that has
also been reported in the context of CC-1065 analog synthesis.23
In turn, the absolute stereochemistry of the cyclopropane was
set through use of an asymmetric dihydroxylation reaction23 that
was found to proceed with an enantioselectivity opposite that
predicted from established models (Scheme 1). Employing
Evans’ optically active acyloxazolidinones, a Dieckmann-like
reaction permitted the introduction of either the natural 6(R) or
epi-6(S) stereochemistry through use of the enantiomeric
auxiliaries or, more interestingly, by conducting the reaction
Scheme 2
(7) Nagamura, S.; Kanda, Y.; Kobayashi, E.; Gomi, K.; Saito, H. Chem.
Pharm. Bull. 1995, 43, 1530. Gomi, K.; Kobayashi, E.; Miyoshi, K.;
Ashizawa, T.; Okamoto, A.; Ogawa, T.; Katsumata, S.; Mihara, A.; Okabe,
M.; Hirata, T. Jpn. J. Cancer Res. 1992, 83, 113. Ichimura, M.; Ogawa,
T.; Takahashi, K.; Mihara, A.; Takahashi, I.; Nakano, H. Oncol. Res. 1993,
5, 165. Kobayashi, E.; Okamoto, A.; Asada, M.; Okabe, M.; Nagamura,
S.; Asai, A.; Saito, H.; Gomi, K.; Hirata, T. Cancer Res. 1994, 54, 2404.
(8) Ogasawara, H.; Nishio, K.; Takeda, Y.; Ohmori, T.; Kubota, N.;
Funayoma, Y.; Ohira, T.; Kuraish, Y.; Isogai, Y.; Saijo, N. Jpn. J. Cancer
Res. 1994, 85, 418. Okamoto, A.; Okabe, M.; Gomi, K. Jpn. J. Cancer
Res. 1993, 84, 93.
(9) Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H.; Kitos, P. A.; Suntornwat,
O. J. Org. Chem. 1990, 55, 4499.
(10) Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H.; Munk, S. A.; Kitos, P.
A.; Suntornwat, O. J. Am. Chem. Soc. 1990, 112, 8961
(11) Boger, D. L.; Munk, S. A.; Zarrinmayeh, H.; Ishizaki, T.; Haught,
J.; Bina, M. Tetrahedron 1991, 47, 2661.
(12) Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H. J. Am. Chem. Soc. 1991,
113, 6645.
(13) Boger, D. L.; Yun, W.; Terashima, S.; Fukuda, Y.; Nakatani, K.;
Kitos, P. A.; Jin, Q. Bioorg. Med. Chem. Lett. 1992, 2, 759.
(14) Boger, D. L.; Johnson, D. S.; Yun, W. J. Am. Chem. Soc. 1994,
116, 1635.
(15) Boger, D. L.; Yun, W. J. Am. Chem. Soc. 1993, 115, 9872.
(16) Boger, D. L.; Johnson, D. S. J. Am. Chem. Soc. 1995, 117, 1443.
Boger, D. L.; Johnson, D. S.; Yun, W.; Tarby C. M. Bioorg. Med. Chem.
1994, 2, 115. Boger, D. L.; Zarrinmayeh, H.; Munk, S. A.; Kitos, P. A.;
Suntornwat, O. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 1431. Boger, D. L.;
Munk, S. A.; Zarrinmayeh, H. J. Am. Chem. Soc. 1991, 113, 3980.
(17) Boger, D. L.; Johnson, D. S.; Wrasidlo, W. Bioorg. Med. Chem.
Lett. 1994, 4, 631.
(18) Boger, D. L.; McKie, J. A.; Nishi, T.; Ogiku, T. J. Am. Chem. Soc.
1996, 118, 2301.
(19) Fukuda, Y.; Itoh, Y.; Nakatani, K.; Terashima, S. Tetrahedron 1994,
50, 2793. Fukuda, Y.; Nakatani, K.; Terashima, S. Tetrahedron 1994, 50,
2809. Fukuda, Y.; Nakatani, K.; Terashima, S. Bioorg. Med. Chem. Lett.
1992, 2, 755. Fukuda, Y.; Nakatani, K.; Ito, Y.; Terashima, S. Tetrahedron
Lett. 1990, 31, 6699.
(20) Boger, D. L.; Machiya J. Am. Chem. Soc. 1992, 114, 10056. Boger,
D. L.; Machiya, K.; Hertzog, D. L.; Kitos, P. A.; Holmes, D. J. Am. Chem.
Soc. 1993, 115, 9025. Boger, D. L. In AdVances in Nitrogen Heterocycles;
Moody, C. J., Ed.; JAI Press: Greenwich, CT, 1994; Vol. 1, p 229.
(21) Muratake, H.; Abe, I.; Natsume, M. Tetrahedron Lett. 1994, 35,
2573. Muratake, H.; Matsumura, N.; Natsume, M. Chem. Pharm. Bull. 1995,
43, 1064. Muratake, H.; Abe, I.; Natsume, M. Chem. Pharm. Bull. 1996,
44, 6779.
of a given auxiliary under thermodynamic versus kinetic reaction
conditions for this reversible reaction.
Nondiastereoselective Total Synthesis of Duocarmycin A
and epi-Duocarmycin A. Prior to initiating our efforts on an
asymmetric synthesis of 1 and 7, we first examined the basic
elements of the approach with the nondiastereoselective prepa-
ration of duocarmycin A (1) and epi-duocarmycin A (7).
Conversion of commercially available 2-amino-5-nitrophenol
to 1324,25 preceeded oxidation to the key p-quinodiimide 14
(84%) effected by Pb(OAc)4 (Scheme 2). Regiospecfic Lewis
(22) Brown, R. F.; Matthews, B. R.; Rae, I. D. Tetrahedron Lett. 1981,
22, 2915.
(23) Aristoff, P. A.; Johnson, P. D. J. Org. Chem. 1992, 57, 6234.
(24) Boger, D. L.; Zarrinmayeh, H. J. Org. Chem. 1990, 55, 1379.
(25) Full experimental details and characterization are provided in the
Supporting Information.

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 313
Scheme 3
Scheme 4
acid-catalyzed24 addition of allyltributyltin (0.5 equiv of BF3‚OEt2,
CH2Cl2, -20 °C, 3 h, 83-89%) cleanly provided 15. Catalytic
dihydroxylation of 15 through treatment with OsO4 (0.2 equiv)
in the presence of N-methylmorpholine N-oxide (NMMO, 2
equiv) in aqueous acetone provided diol 16 (95%). Selective
formation of the primary tosylate 17 (1.1 equiv of Bu2SnO,
toluene-THF 10:1, reflux, 6 h; 1.2 equiv of TsCl, cat. Et3N,
25 °C, 12 h, 87-94%) followed by subsequent protection of
the secondary alcohol as its TBDMS ether (67-75%) cleanly
provided 18. Cyclization of 18 to afford the first key intermedi-
ate 19 was accomplished by base-promoted closure (2 equiv of
NaH, THF, 0 °C, 2 h, 92-97%). Deprotection of the two
benzoyl protecting groups (NH2NH2-EtOH 10:1, 58-65%)
followed by immediate reprotection of the free amine 20 with
BOC2O (95%) provided 21. The major byproduct of the
benzoyl deprotection reaction was the free alcohol of the amine
20 resulting from additional desilylation under the vigorous
reaction conditions. Simply subjecting this recovered alcohol
to TBDMSOTf treatment (2,6-lutidine, CH2Cl2) followed by a
brief treatment with 1 M aqueous citric acid afforded additional
20 and overall conversions of 85% for the transformation of
19 to 21. Mild acid-catalyzed removal of the labile-BOC group
of 21 (5-6 equiv TFA, CH2Cl2, 93%) cleanly provided 22. Less
vigorous conditions for acylation of the amine 20 led to mono
BOC formation at the more reactive secondary amine with no
evidence for the generation of 22 and the more vigorous
conditions employing DMAP preferentially provided 21.
conversion in two steps through formation of the mesylate 2525
(94%) followed by base-promoted closure to 19 (2.5 equiv of
NaH, 97%).
A number of alternatives to the conversion of diol 16 to 19
were also examined. Although selective acetylation of the
primary alcohol of 16 could be effectively accomplished (1.1
equiv of Ac2O, 1.1 equiv of Et3N, CH2Cl2, 0 °C, 73%), the
subsequent protection of the secondary alcohol provided re-
covered 2627 or led to competitive acyl migration and a mixture
of reaction products (Scheme 3). Similarly, selective mesylation
of the primary alcohol to provide 2727 followed by attempted
protection of the remaining secondary alcohol as its TBDMS
ether or acetate 2927 analogous to related studies23 failed to
provide a useful route to 19.
One potential significant improvement in the approach was
investigated that utilizes the p-quinodiimine 32 incorporating
the N-BOC protecting groups directly (Scheme 4). Conversion
of 12 to 3125 via 3025 followed by oxidation to the p-
quinodiimine 3225 provided the key substrate for this alternative
approach. Regiospecific Lewis acid-catalyzed addition of
allyltributyltin (0.5 of equiv of BF3‚OEt2, CH2Cl2, -78 °C, 2
h, 71%), catalytic dihydroxylation of 3325 (cat. OsO4, 2.0 equiv
of NMMO, acetone-H2O, 25 °C, 95%), and selective tosylation
of the primary alcohol 3425 (1.1 equiv of Bu2SnO, toluene,
reflux, 4 h; 1.2 equiv of TsCl, cat. Et3N, 25 °C, 10 h, 88%) to
provide 35,25 formation of the TBDMS ether 3625 (1.5 equiv of
TBDMSOTf, 2 equiv of 2,6-lutidine, CH2Cl2, -30 °C, 2 h,
82%), and base-promoted closure25 (2.0 equiv of NaH, 95%)
provided 22. Despite the improvement, the preparative resolu-
tion of 19 proved remarkably simple on a semipreparative
ChiralCel OD HPLC column (R ) 2.30) and much more
effective than that observed with 22 (R ) 1.14). For the
preparation of optically active 1 and 7, both the Sharpless
asymmetric dihydroxylation of 15 versus 33 as well as the
resolution of 19 or the enrichment of the optical purity of (S)-
or (R)-19 versus 22 proved more effective and was adopted.
Employing optically active 19 (>99.9% ee), the nondiaste-
reoselective synthesis of (+)-duocarmycin A (1) and epi-(+)-
The key intermediate 19 was also obtained by exhausive
protection of the diol 16 as its bis TBDMS ether 2325 (3-7
equiv TBDMSCl, cat. DMAP, 2.5-3 equiv Et3N, DMF, 25 °C,
16 h, 100%) followed by selective deprotection of the primary
alcohol (HOAc-H2O-THF, 3:1:1, 25 °C, 48 h) to provide 2425
(58%) along with recovered 23 (17%) and the diol 16 (23%),
Scheme 3. Both 16 and 23 could be recycled, providing
effective conversions to 24 exceeding 90%. Cyclization of 24
to afford the key intermediate 19 was accomplished either in
one-step upon Mitsunobu activation26 of the primary alcohol
(1.5 equiv of DEAD, 1.5 equiv of Ph3P, THF, 72%) or in higher
(27) Diagnostic characterization is provided in the Supporting Informa-
tion.
(26) Mitsunobu, O. Synthesis 1981, 1.

314 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
Scheme 5
Scheme 6
analogous to those described by Natsume28 provided (+)-
duocarmycin A (1) identical in all respects with authentic
material, [R]25 +291 (c 0.01, CH3OH).1 The intermediate
D
alcohol 42, duocarmycin D1, has also been isolated from the
culture broths of Streptomyces sp. DO-89, which produces (+)-
duocarmycin A and constitutes a natural product in its own
right.7 Alternatively, treatment of 41 with MsCl (1.5 equiv,
pyridine, 0 °C, 1 h, 89-92%) followed by hydrogenolysis of
the benzyl ether 43 (H2, 10% Pd-C, 88-100%) and spirocy-
clization effected by treatment of 44 with DBU1 (83%) also
provided (+)-duocarmycin A in conversions that proved more
dependable on a small scale.
Similarly, acid-catalyzed deprotection of 4625 followed by
coupling with 5,6,7-trimethoxyindole-2-carboxylic acid (40)10
and the subsequent conversion of 47 to epi-(+)-duocarmycin
A (7), [R]25 +155 (c 0.02, CH3OH), was accomplished25 as
D
detailed for (+)-1 (Scheme 5). By means of the same approach
but employing (R)-19 (>99.9% ee), the unnatural enantiomers
of 1, [R]25 -284 (c 0.025, CH3OH), and 7, [R]25 -156 (c
duocarmycin A (7) as well as their unnatural enantiomers was
completed as shown in Scheme 5. N-Alkylation of 22 with
methyl 2-bromopropionate (2 equiv of NaH, DMF, 0 °C, 0.5 h,
96%) provided 37. Dieckmann-like condensation analogous to
the efforts of Terashima19 effected by treatment with LDA (3
equiv of, THF, -78 °C, 1 h, 54%, 90% based on recovered 37)
provided a separable 1:1 mixture of 38 and 45 (SiO2, 20%
EtOAc-hexane, R ) 1.25) epimeric at the C6 center. Interest-
ingly, this readily reversible reaction was difficult to drive to
completion and inevitably provided variable amounts of recov-
ered starting material under the conditions examined. In the
subsequent extension of this reaction to the N-formyl derivative
or the Evans’ acyl oxazolidinones, the ring closure product was
more easily captured, presumably due to the decreased steric
congestion of the product or the reduced acidity of the
intermediate enolate, respectively. Mild hydrolysis of the imine
(2-3 equiv of TsOH‚H2O, THF-H2O 8:1, 0 °C, 2-3 h, 76-
85%) conducted independently on the isomers or on the 1:1
mixture followed by chromatographic separation of 39 and 46
(SiO2, 20% EtOAc-hexane, R ) 1.33) provided the key
precursors to 1 and 7 without competitive N-BOC or TBDMS
ether deprotection. Acid-catalyzed deprotection of 39 (4 M
HCl-CH3OH, 0 °C, 1 h), which served to remove both BOC
and the TBDMS protecting groups, followed by coupling with
5,6,7-trimethoxyindole-2-carboxylic acid10 (40, 73%) provided
41. Removal of the benzyl ether (H2, 10% Pd-C, 91-98%)
followed by direct transannular spirocyclization of the alcohol
42 upon Mitsunobu activation (1.5 equiv of ADDP, 1.5 equiv
of Bu3P, C6H6, 50 °C 1 h, 99%) conducted under conditions
D
D
0.025, CH3OH), were also prepared.
Similar efforts were conducted with the N-formyl derivative
52,25 which was obtained from 20 by BOC protection of the
secondary amine (5 equiv of BOC2O, CH3CN, 90 °C, 16 h,
84%) followed by formylation of the primary amine of 5125
(HCO2H-Ac2O, 25 °C, 1 h, 93%). N-Alkylation of 52 with
methyl 2-bromopropionate in the presence of excess NaH (3
equiv, DMF, 0 °C, 7 h, 78%) and subsequent in situ Dieckmann-
like cyclization19 provided 5325 directly (Scheme 6). Hydroly-
sis25 of the imine 53 to the â-keto esters 54 and 55 was
accomplished by exposure to TsOH (3 equiv, 4:1 THF-H2O,
0 °C, 2 h, 90%). Although the intermediate 53 constituted a
1:1 mixture of diastereomers that was not separable, the two
diastereomers 54 and 55 were separable (SiO2, 30% EtOAc-
hexane, R ) 1.1). Exhaustive acid-catalyzed deprotection of
54 or 55 was accomplished by treatment with saturated HCl-
CH3OH (25 °C, 5 h, 91%), and the resulting amine hydrochlo-
ride was immediately coupled with 5,6,7-trimethoxyindole-2-
carboxylic acid (40)10 employing EDCI (3 equiv, DMF, 25 °C,
48 h, 52%) or DEPC29 (1.5 equiv, 1.5 equiv of Et3N, THF, 70
°C, 16 h, 63%) to cleanly provide 41 and 47. A number of
alternative approaches were also examined, and a summary is
provided in the Supporting Information.
Asymmetric Dihydroxylation of 15: Establishment of the
Cyclopropane Absolute Stereochemistry. The overriding
(28) Tsunoda, T.; Yamamiya, Y.; Ito, S. Tetrahedron Lett. 1993, 34,
1639.
(29) Takuma, S.; Hamada, Y.; Shiori, T. Chem. Pharm. Bull. 1982, 30,
3147.

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 315
Table 1. Asymmetric Dihydroxylation of 15
equiv of OsO4
ligand (equiv)
condnsa (solvent, time (h))
% yield
% ee
0.004
0.1
0.1
0.04
0.1
0.1
0.1
0.1
0.1
0.1
DHQD-CLB (0.25)
PHN-DHQ (0.1)
acetone-H2O, 72
THF-H2O, 48
73
60
12
34
69
84
73
71
46
91
62
86
88
87
89
58
59
92
0
0
0
MEQ-DHQ (0.1)
THF-H2O, 48
(DHQD)2-PYR (0.04)
(DHQD)2-PYR (0.1)
(DHQD)2-PYR (1.0)
(DHQ)2-PYR (0.5)
(DHQ)2-PYR(OMe)3 (0.5)
(DHQ)2-PYR (0.1)
(DHQD)2-AQN (0.2)
(DHQD)2-AQN (0.2)
(DHQ)2-PHAL (0.1)
(DHQ)2-PHAL (0.5)
(DHQ)2-PHAL (1.0)
(DHQ)2-PHAL (0.1)
(DHQ)2-PHAL (0.1)
(DHQ)2-PHAL (0.5)
(DHQD)2-PHAL (0.1)
THF-H2O, 72
14 (S)
36 (S)
58 (S)
39 (R)
49 (R)
50 (R)
50 (S)
49 (S)
47 (R)
62 (R)
60 (R)
77 (R)
74 (R)
78 (R)
78 (S)
THF-H2O, 48
THF-H2O, 48
THF-H2O, 48
THF-H2O, 48
EtOAc-H2O, 96
THF-H2O, (4:1), 16
EtOAc-H2O, 16
THF-H2O, 72
0.1
0.1
0.1
THF-H2O, 48
THF-H2O, 48
THF-H2O, (4:1), 12
EtOAc-H2O, 48
EtOAc-H2O, 48
THF-H2O, (4:1), 16
0.1
0.01b
0.1
0.1
0.01b
a 2:1 solvent-H2O unless indicated otherwise. b Preparative scale.
Table 2. Solvent Effect on the Asymmetric Dihydroxylationa
In preliminary studies, the best results were obtained as the
amount of ligand was increased, and because of the insolubility
of 15 in t-BuOH, alternative solvent systems were also
examined. The reaction was found to exhibit a large solvent
dependency effecting the chemical conversion and a modest
solvent dependency on the degree of asymmetric induction.
Consequently, the effect of the solvent was examined in detail,
and representative results are summarized in Table 2. In general,
both the chemical yields and the ee’s were best in the solvent
systems in which 15 was most soluble (i.e., 4:1 > 2:1 THF-
H2O), and the modest chemical conversions in many of the
solvent systems may be attributed to the low solubility of 15
rather than slow osmate ester hydrolysis. Under the best
conditions, dihydroxylation of 15 in the presence of (DHQD)2-
PHAL (0.1 equiv, 0.01 equiv of OsO4, 3 equiv of K3Fe(CN)6,
3 equiv of K2CO3, 0.4 equiv of CH3SO2NH2, 4:1 THF-H2O,
16 h, 25 °C, 89-92%, 78% ee) provided (S)-16, and the
complementary ligand (DHQ)2-PHAL provided the correspond-
ing enantiomer (89%, 77% ee).
0.1 equiv of OsO4
0.1 equiv of (DHQ)2-PHAL
sssssssssssssf
15
(R)-16
K3Fe(CN)6
K2CO3, 0 °C
solvent
% yield
% ee
t-BuOH-THF-H2O
CH3CN-H2O
50
42
86
66
76
46
51
59
41
68
58
59
89
25
41
47
62
64
66
69
69
70
73
74
78b
77c
THF-H2O
EtOAc-THF-H2O
EtOAc-dioxane-H2O
EtOAc-t-BuOMe-H2O
EtOAc-toluene-H2O
ClCH2CH2Cl-H2O
EtOAc-acetone-H2O
CH2Cl2-H2O
EtOAc-H2O
EtOAc-H2O
THF-H2O (4:1)
Moreover, the absolute configuration of the product derived
from the asymmetric dihydroxylation of 15 proved to be
opposite that predicted from established models.30 This was
first recognized upon taking the major enantiomer derived from
the (DHQ)2-PHAL-catalyzed reaction through the synthesis that
provided the unexpected unnatural enantiomer about the cy-
clopropane. Similarly, the major enantiomer derived from the
complementary (DHQD)2-PHAL-catalyzed reaction unexpect-
edly provided the natural enantiomer series, and these observa-
tions were repeated several times, ensuring that the samples were
not inadvertently switched. Since the absolute configuration
of natural 1 was established by X-ray analysis of a heavy atom
derivative (4)4 and firmly supported by its DNA alkylation
properties, this suggested that it was the sense of asymmetric
induction in the dihydroxylation reaction, and not the natural
product absolute configuration, that was in question. Conse-
quently, the (S)-Mosher ester of the primary alcohol of the major
enantiomer of 16 derived from the (DHQD)2-PHAL-catalyzed
reaction was prepared and subjected to X-ray structure analysis
(Scheme 7).34 Comparison of the relative configuration unam-
biguously established that it possesses the S absolute configu-
a 2:1 solvent-H2O. b 0.5 equiv of ligand. c 4:1 THF-H2O, 0.01 equiv
of OsO4, 0 °C, 16 h.
consideration in the adopted approach was its ability to provide
either enantiomer of the key cyclopropane stereochemistry.
Thus, complementary to the resolution of a prochiral diol such
that all material could be employed to provide either or both
enantiomers, our examination of methods of asymmetric
synthesis was restricted to approaches that could provide access
to both enantiomers. Since these elements are embodied in the
Sharpless asymmetric dihydroxylation reaction,23,30 it was
especially attractive. Related efforts have been disclosed by
Aristoff and conducted in the context of the synthesis of CBI-
based analogs of CC-1065.23 Representative results of the study
of the asymmetric dihydroxylation of 15 are summarized in
Tables 1 and 2 and proved far more interesting than anticipated
or previously discussed.23 The degree of asymmetric induction
was assessed upon conversion of the diol 16 to the TBDMS
ether 23 and analysis of the resulting enantiomers on an
analytical ChiralCel OD HPLC column (0.46 × 25 cm, 13%
i-PrOH-hexane, 0.3 mL/min flow rate, tR(S) ) 31.5 min, tR(R)
) 27.5 min, R ) 1.15). Without optimization of the conver-
sions, an initial survey of the available ligands revealed that
(DHQ)2-PHAL/(DHQD)2-PHAL31 performed better than the
(DHQ)2-PYR/(DHQD)2-PYR ligands,32 which in turn were
superior to their PHN, MEQ, or CLB33 predecessors, Table 1.
(31) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispin, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.; Xu,
D.; Zhang, X.-L. J. Org. Chem. 1992, 57, 2768.
(32) Crispino, G. A.; Jeong, K.-S.; Kolb, H. C.; Wang, Z.-M.; Xu, D.;
Sharpless, K. B. J. Org. Chem. 1993, 58, 3785.
(33) Sharpless, K. B.; Amberg, W.; Beller, M.; Chen, H.; Hartung, J.;
Kawanami, Y.; Lubben, D.; Manoury, E.; Ogino, Y.; Shibata, T.; Ukita, T.
J. Org. Chem. 1991, 56, 4585.
(30) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483.

316 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
Table 3. Resolution on a ChiralCel OD Columna
Scheme 7
agent
solvent
R
19
20% i-PrOH-hexane
25% i-PrOH-hexane
3% i-PrOH-hexane
13% i-PrOH-hexane
2.30b
1.94b
1.14b
1.15
22
23
a 0.46 × 25 cm ChiralCel OD analytical HPLC column. b Semi-
preparative ChiralCel OD HPLCL column (2 × 25 cm).
may be drawn from their examination. This was apparent in
our initial studies conducted in collaboration with Terashima13
on samples of the unnatural enantiomers of 1 and 7 derived
from a diastereomeric derivatization and chromatographic
resolution of racemic intermediates.19 In these studies, the
behavior of the unnatural enantiomers were masked by the
dominant properties of the more potent and contaminating
natural enantiomer (ca. 1%), and their characterization was
limited to the establishment that they were g100× less effective
than the corresponding natural enantiomer. Consequently, we
examined protocols that could efficiently resolve our racemic
intermediates or further enrich the enantiomeric purity of our
optically active intermediates in a manner that could dependably
provide materials that were >99.9% ee. This was effectively
accomplished by direct chromatographic resolution36-38 on a
semipreparative ChiralCel OD HPLC column (2 × 25 cm). A
series of intermediates were examined for potential resolution,
and the results are summarized in Table 3. The resolution of
the key intermediate 19 on a ChiralCel OD HPLC column
proved remarkable, R ) 2.30-1.94, and using even a semi-
preparative HPLC column (2 × 25 cm, 20-25% i-PrOH-
hexane, 5 mL/min) 150 mg of racemic or 250 mg of optically
enriched 19 (78% ee) could be resolved with a single injection
and provided the enantiomers in >99.9% ee.
ration contrary to expectations based on the established ADH
models35 but consistent with that required for synthesis of the
natural product.
Development of a Divergent Approach for the Diastereo-
selective Introduction of the Quaternary C6 Center. One
key element of the approach to 1 or its epimer 7 rests with the
introduction of the C6 quaternary center with control of its
relative and absolute stereochemistry. Given our interest in both
diastereomers, this suggested that its introduction on a late stage
intermediate might prove most effective. In preliminary stud-
ies,39 we had examined a number of approaches and found that
a diastereoselective Dieckmann-like condensation employing
Evans’ optically active N-acyloxazolidinones was especially
effective. Moreover, the Evans auxiliaries proved to be the only
substrates in the series examined in which the diastereomers
were readily separable by chromatography.
Similar observations were made with the substrate 33,
indicating that the degree and sense of asymmetric induction
in the (DHQ)2-PHAL-catalyzed reaction (0.02 equiv of OsO4,
3 equiv of K2CO3, 3 equiv of K3Fe(CN)6, 2:1 t-BuOH-H2O, 0
°C, 24 h, 93%, 56% ee, (R)-34) were not unique to the substrate
15 bearing the N-benzoyl protecting groups. In addition, the
substrate containing a tertiary N-methylamide in place of the
secondary amide of 15 exhibited the same sense and degree of
asymmetric induction, indicating that the potential hydrogen
bonding capabilities of the secondary amide 15 or carbamate
34 were not responsible for this reversal of enantioselectivity,
eq 1.25 Thus, the unusual reversal of the AD enantioselectivity
In initial studies, we found that slow addition of 57 to a
solution of LDA (1.2 equiv) at -78 °C over a period of 30 min
provided 58 and 59 in excellent yield and diastereoselection
(7:1), Scheme 8. Moreover, the major and minor diastereomers
(34) The atomic coordinates have been deposited with the Cambridge
X-ray Data Centre and may be obtained, on request, from the Director,
Cambridge X-ray Data Centre, University Chemical Lab, Lensfield Road,
Cambridge, CB2 1EW, UK.
(35) For additional exceptions, see: Krysan, D. J. Tetrahedron Lett. 1996,
37, 1375. Hale, K. J.; Manaviazar, S.; Peak, S. A. Tetrahedron Lett.1994,
35, 425. Vanhessche, K.; Sharpless, K. B. Unpublished results. With
optically active substrates and double diastereoselection: Carreira, E. M.;
DuBois, J. J. Am. Chem. Soc. 1994, 116, 10825. Iwashima, M.; Kinsho,
T.; Smith, A. B., III. Tetrahedron Lett. 1995, 36, 2199. Krysan, D. J.;
Rockway, T. W.; Haight, A. R. Tetrahedron: Asymmetry 1994, 5, 625.
(36) Boger, D. L.; Yun, W. J. Am. Chem. Soc. 1994, 116, 7996. Boger,
D. L.; Yun, W.; Han, N. Bioorg. Med. Chem. 1995, 3, 1429.
(37) Boger, D. L.; McKie, J. A.; Cai, H.; Cacciari, B.; Baraldi, P. G. J.
Org. Chem. 1996, 61, 1710.
does not appear to be influenced by the obvious proximal olefin
substituents of 15, and such substrates may represent a more
general class of olefins for which the AD enantioselectivity is
reversed or more difficult to predict.
Resolution of Racemic 19 or Enrichment of the Enantio-
meric Purity of Optically Active 19. The sensitivity of the
biological assays to the contaminant enantiomers is exceptional,
and even 1.0-0.1% of contaminant natural enantiomer in the
unnatural enantiomier series complicates the conclusions that
(38) Boger, D. L.; McKie, J. A.; Han, N.; Tarby, C. M.; Riggs, H. W.;
Kitos, P. A. Bioorg. Med. Chem. Lett. 1996, 6, 659. Boger, D. L.; Han, N.;
Tarby, C. M.; Boyce, C. W.; Cai, H.; Jin, Q.; Kitos, P. A. J. Org. Chem.
1996, 61, 4894.
(39) Boger, D. L.; Nishi, T. Bioorg. Med. Chem. 1995, 3, 67.

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 317
Scheme 8
proved to be easily separable by chromatography (Rf ) 0.40
and 0.78, R ) 1.95; SiO2, 50% EtOAc-hexane), and the desired
diastereomer (2R)-58 was isolated in 86% yield (>99.9% de).
A single-crystal X-ray structure determination of the minor
diastereomer (2S)-59 established the relative and absolute
stereochemistry. We also noted in this work that the diaste-
reoselectivity of the reaction was dependent on the reaction
conditions. Both reverse addition or addition over shorter time
periods and at higher reaction temperatures led to diminished
or even reversed diastereoselection. We now report that under
kinetically controlled conditions, the reaction of 57 provides
predominately 58 (7:1) as previously detailed, whereas under
the thermodynamic conditions (1.5 equiv of LDA, THF, -40
°C, 3 h, 92-94%) of a reversible reaction it provides near-
exclusive generation of 59 (14:1).25 Establishment that 58
constitutes the kinetic product and 59 is the thermodynamic
product of a reversible ring closure was obtained by resubjecting
58 to the reaction conditions (1.1 equiv of LDA, -40 °C, 2 h)
with equilibration to 59 (10:1).25 The stereochemical outcome
is consistent with the derivation of the kinetic product from the
reaction of a chelated (Z)-enolate, while the latter equilibration
provides the most stable of the two possible diastereomers (∆E
) 0.76 and 0.41 kcal/mol, MM2 and AM1). The latter result
is dependent on the size of the adjacent N-protecting group and
its interaction with the chiral auxiliary substituent upon adoption
of a nonchelated anti-carbonyl conformation with the acyl
carbonyl eclipsed with the C-6 methyl group, and this is
highlighted nicely in the X-ray structure of 59 (Scheme 8).
Divergent, Enantioselective Total Syntheses of (+)-Duo-
carmycin A and (+)-epi-Duocarmycin A and Their Un-
natural Enantiomers. With the technology in hand to control
the absolute configuration of the C6 stereocenter, we directed
our efforts to the completion of an asymmetric total synthesis
of the natural product. The complemenary stereochemistry of
the kinetic versus thermodynamic Dieckmann-like condensation
provided the option of utilizing a single intermediate bearing
the same Evans oxazolidinone for the preparation of both natural
(+)-duocarmycin A and (+)-epi-duocarmycin A in addition to
the anticipated use of intermediates bearing the enantiomeric
oxazolidinone auxiliaries. Alkylation of (3S)-22 with the
oxazolidone 60a39 (1.5 equiv of NaH, DMF, 0 °C, 1 h, 88%)
followed by Dieckmann-like condensation of 61a under ther-
modynamically-controlled reaction conditions best achieved by
addition of base to a cooled solution of the substrate (6 equiv
of LDA, THF, -78 to -50 °C, 0.5 h) cleanly provided the 2R,8S
Scheme 9
diastereomer 62a (78%) as the near-exclusive closure product
(g8-10:1) and a small amount of recovered 61a (7%), Scheme
9. Alternatively, condensation of 61b incorporating the enan-
tiomer of the acyl oxazolidinone under kinetically controlled
conditions achieved by slow addition of substrate to base
rigorously maintained at -78 °C (4-6 equiv of LDA, THF,
-78 °C, 10 min) provided predominately 62b (51%) also
possessing the desired 2R,8S stereochemistry. That the latter
constitutes the kinetic product and the former the thermodynamic

318 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
Scheme 10
Figure 2. Solvolysis study (UV spectra) of N-BOC-DA in 50% CH3-
OH-aqueous buffer (pH 3.0, 4:1:20 (v/v/v) 0.1 M citric acid, 0.2 M
Na2HPO4, and H2O, respectively). The spectra were recorded at regular
intervals, and only a few are shown for clarity. Time intervals: 1, 0 h;
2, 3 h; 3, 6 h; 4, 9 h; 5, 12 h; 6, 24 h; 7, 36 h; 7, 48 h.
nular spirocyclization provided (+)-N-BOC-DA (67), [R]25
D
+183 (c 0.01, THF), and 6-epi-(+)-N-BOC-DA (71), [R]25
D
+73 (c 0.02, THF). In a similar fashion, the unnatural
enantiomers of 67, [R]25D -190 (c 0.02, THF), and 71,25 [R]25
D
-67 (c 0.01, THF), were also prepared.
Finally, in conjunction with efforts to establish the stereo-
chemistry of ring expansion solvolysis products obtained with
67, both enantiomers of the diastereomeric alcohols 72 and 73
were prepared by hydrogenolysis of 64 and 68.
Solvolysis Reactivity. Two characteristics of the alkylation
subunits have proven important in the studies to date. The first
is the stereoelectronically-controlled acid-catalyzed ring opening
of the activated cyclopropane, which dictates preferential
addition of a nucleophile to the least substituted cyclopropane
carbon. The second is the relative rate of acid-catalyzed
solvolysis, which has been found to accurately reflect the
functional reactivity of the agents and to follow a fundamental,
direct relationship between solvolysis stability and in vitro
cytotoxic potency.36-42
The reactivity of 67 was assessed by following the solvolysis
spectrophotometrically by UV at pH 3 (50% CH3OH-buffer,
buffer ) 4:1:20 (v/v/v) 0.1 M citric acid, 0.2 M Na2HPO4, H2O),
measuring both the disappearance of the long-wavelength
absorption band at 336 nm and the appearance of a short
wavelength band attributable to the solvolysis products (Figure
2). As expected, N-BOC-DA (67, t1/2 ) 11 h, k ) 1.75 × 10-5
s-1) proved to be more reactive toward chemical solvolysis at
pH 3 than the alkylation subunit of CC-1065, N-BOC-CPI (t1/2
) 37 h), or duocarmycin SA, N-BOC-DSA (t1/2 ) 177 h), Table
4. Thus, N-BOC-DA exhibits a stability at pH 3 that is not so
distinct from that of N-BOC-CPI but that is substantially less
than that of N-BOC-DSA.
Thus, the rate of acid-catalyzed solvolysis of N-BOC-DA was
found to be substantially faster than that of N-BOC-DSA,
presumably due to the enhanced gain in delocalization energy
upon aromatization as well as the electronic deactivation by
the conjugated C6 methoxycarbonyl substituent with duocar-
mycin SA. The magnitude of these effects is significant (16×)
and revealing. This closely follows the trends established for
the relative biological potency and DNA alkylation efficiency
of the agents where the chemically more stable agents exhibit
the more effective properties. In addition, the observation of a
product of a reversible reaction was established by resubjecting
pure 62b and 63b to LDA treatment with equilibration to the
thermodynamic product 63b. Methanolysis of 62a or 62b (30
equiv of LiOCH3, THF-CH3OH, 0 °C, 0.5 h, 78-84%)
provided (2R,8S)-38, and its conversion to (+)-duocarmycin A
(1) was accomplished as detailed in Scheme 5.
Similarly, closure of 61b under conditions of thermodynamic
control or condensation of 61a under kinetic control where
product equilibration is minimized provided 63b (65%) or 63a
(56%), respectively, possessing the 3S,6S stereochemistry. Their
methanolysis and conversion to epi-(+)-duocarmycin A (7) was
accomplished as detailed in Scheme 5. Notably, the diastere-
omeric pairs 62-63a (SiO2, R ) 1.40, 20% EtOAc-hexane)
and 62-63b (SiO2, R ) 2.30, 30% EtOAc-hexane) were
readily separable by chromatography, thereby providing the pure
diastereomers (>99.9% de). Thus, the diastereoselective prepa-
ration of 1 or its epimer 7 was achieved based on a divergent
Dieckmann-like condensation in which either C6 absolute
configuration could be introduced through choice of kinetic or
thermodynamic reaction conditions on the same intermediate
or through choice of complementary auxiliaries. By means of
the same approach but employing (3R)-22, the unnatural
enantiomers of 1 and 7 were also prepared.
Preparation of (+)-N-BOC-DA, (+)-epi-N-BOC-DA and
Their Unnatural Enantiomers. In order to permit the
examination of key partial structures of 1 and their unnatural
enantiomers as well as to provide samples key to the examina-
tion of their functional reactivity and reaction regioselectivity,
(+)-N-BOC-DA (67) and 6-epi-(+)-N-BOC-DA (71)25 were
prepared as detailed in Scheme 10. Exhaustive deprotection
of 39 or 4625 (5 M HCl-CH3OH, 0 °C, 1 h), which served to
remove both N-BOC and TBDMS protecting groups, followed
by immediate reprotection of the more reactive secondary amine
(3 equiv of BOC2O, THF, reflux, 2 h, 61-57%) provided 64
and 68,25 respectively. Analogous to the final steps in the
preparation of 1 and 7, activation of the secondary alcohol
toward spirocyclization through formation of the mesylate,
hydrogenolysis removal of the benzyl ether, and final transan-
(40) Boger, D. L.; Ishizaki, T.; Kitos, P. A.; Suntornwat, O. J. Org. Chem.
1990, 55, 5823. Boger, D. L.; Ishizaki, T. Tetrahedron Lett. 1990, 31, 793.
Boger, D. L.; Ishizaki, T.; Sakya, S. M.; Munk, S. A.; Kitos, P. A.; Jin, Q.;
Besterman, J. M. Bioorg. Med. Chem. Lett. 1991, 1, 115. Boger, D. L.;
Ishizaki, T.; Wysocki, R. J., Jr.; Munk, S. A.; Kitos, P. A.; Suntornwat, O.
J. Am. Chem. Soc. 1989, 111, 6461.
(41) Boger, D. L.; Yun, W. J. Am. Chem. Soc. 1994, 116, 5523.
(42) Boger, D. L.; Me´sini, P.; Tarby, C. M. J. Am. Chem. Soc. 1994,
116, 6461. Boger, D. L.; Me´sini, P. J. Am. Chem. Soc. 1994, 116, 11335.
Boger, D. L.; Me´sini, P. J. Am. Chem. Soc. 1995, 117, 11647.
(43) Boger, D. L.; Goldberg, J. A.; McKie, J. A. Bioorg. Med. Chem.
Lett. 1996, 6, 1955.

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 319
Table 4.
reached in a study of the CPI solvolysis where a free carbocation
has been invoked to explain the observation of minor ring
expansion products.45 The results herein and those of the related
studies42,43 suggest that this later work should be reexamined
using a more definitive experimental basis for assessing the
stereochemical course of the reaction.
This proved consistent with kinetic studies of the acid-
catalyzed nucleophilic addition conducted on related sys-
tems37,38,45 where the rate of reaction exhibits a first-order
dependence on both the acid concentration (pH) as well as the
nucleophile, indicative of a mechanism involving rapid and
reversible C4 carbonyl protonation followed by a slow, rate-
determining SN2 nucleophilic attack on the activated cyclopro-
pane (eq 3).
74, (+)-N-BOC-DSA
t1/2 ) 177 h, pH 3
75, (+)-N-BOC-CPI
t1/2 ) 37 h, pH 3
67, (+)-N-BOC-DA
t1/2 ) 11 h, pH 3
regioselectivity: 1.5:1
regioselectivity: 6-4:143 regioselectivity: 4:1
IC50 (L1210): 6 nM
IC50 (L1210): 300 nM IC50 (L1210): 2 µM
minor or trace guanine N3 alkylation with duocarmycin A that
is not detected even under forcing conditions with duocarmycin
SA may be attributed to this greater reactivity of duocarmycin
A. Similarly, the greater ease of reversibility of the duocarmycin
SA DNA alkylation reaction may be attributed to the lower
degree of adduct stability resulting from its lower reactivity.
Thus, consistent with this established relative reactivity, the
duocarmycin A DNA alkylation reaction is less efficient, less
selective among the available sites, and less reversible than that
of duocarmycin SA.14
Solvolysis Regioselectivity and Mechanism of Acid-
Catalyzed Nucleophilic Addition. As evidenced by the
isolation and characterization of duocarmycins D1 and D2 and
as observed in the studies of Saito,44 the nucleophilic additions
to the duocarmycin A system occur with both normal and
abnormal ring expansion cleavage of the cyclopropane. An
assessment of the acid-catalyzed additions to 67 was conducted
and established that solvolysis preferentially occurs with cleav-
age of the C7b-C8 bond with addition of a nucleophile to the
least substituted C8 cyclopropane carbon versus cleavage of the
C7b-C8a bond with ring expansion and addition of a nucleo-
phile to C8a. The latter cleavage would place a developing
partial positive charge on a preferred secondary versus primary
center and in preceding agents, this preference was overridden
by the inherent stereoelectronic control of the reaction regiose-
lectivity. However, this preference is greatly diminished with
N-BOC-DA. Preparative solvolysis of 67 (20% H2O-THF, 0.1
equiv of CF3SO3H, 25 °C, 24 h) cleanly provided both the
normal solvolysis product 76 (46%) and the abnormal ring
expansion solvolysis product (+)-72 (30%) in a 1.5:1 ratio (eq
2). Moreover, no trace of the diastereomer (+)-73 was detected,
To date, agents incorporating the CBI nucleus have exhibited
the greatest regioselectivity (g20:1)37,38,40 for the cyclopropane
ring-opening reaction including the exceptionally reactive F2-
CBI (g9:1),46 and more modest selectivity has been observed
with CPI derivatives (ca. 4:1),45 duocarmycin A (1.5-1:1),7,44
duocarmycin SA (6-4:1),43 or CBQ derivatives (3:2).42 The
distinguishing features controlling the regioselectivity of addition
appear to be the stereoelectronic alignment of the two cyclo-
propane bonds available for cleavage and the relative reactivity
of the agent. Within a class of agents whose cyclopropane
alignment with the π-system would be expected to be very
similar due to simple structural constraints, the solvolysis
regioselectivity nicely follows the relative reactivity with the
more stable agents providing the more selective reaction: e.g.,
N-BOC-DSA (6-4:1)43 > CPI (4:1)45 > N-BOC-DA (3:2).
However, this fails to hold true when comparing between classes
of agents: e.g., N-BOC-CBI (g20:1) versus N-BOC-DSA (6-
4:1). Thus, additional important factors contribute to this
reaction regioselectivity. Like the comparisons made in the
available structural studies of CPI,47 DSA,43 CBI,40 MCBI,37
CBQ,42 and F2CBI,46 both the diminished regioselectivity of
DSA relative to CBI and the very modest regioselectivity
observed with the CBQ-based agents but maintained with the
exceptionally reactive F2CBI-based agents may be attributed
to the relative extent of stereoelectronic alignment of the two
possible cyclopropane bonds. In the cases where this structural
information is available,37,40,42,43,46,47 the degree of selectivity
also reflects the relative degree of stereoelectronic alignment
of the two available cyclopropane bonds, and this alone could
account for the reaction regioselectivity.
indicating that under these conditions the ring expansion
solvolysis occurs with clean inversion of the reacting center
stereochemistry indicative of SN2 addition to the activated
cyclopropane. This is in complete agreement with the similarly
unambiguous observations made in studies of the CBQ-based
agents42 and DSA-based agents43 but contrasts the conclusions
DNA Alkylation Regioselectivity. The observation of
exclusive adenine N3 addition to the C8 cyclopropane carbon
in the DNA alkylation studies of duocarmycin A9-13,46 is not
consistent with expectations that the inherent acid-catalyzed
(45) Warpehoski, M. A.; Harper, D. E. J. Am. Chem. Soc. 1994, 116,
7573. Warpehoski, M. A.; Harper, D. E. J. Am. Chem. Soc. 1995, 117,
2951.
(46) Boger, D. L.; Jenkins, T. J. J. Am. Chem. Soc. 1996, 118, 8860.
(47) Chidester, C. G.; Krueger, W. C.; Mizsak, S. A.; Duchamp, D. J.;
Martin, D. G. J. Am. Chem. Soc. 1981, 103, 7629. Martin, D. G.; Kelly, R.
C.; Watt, W.; Wienienski, N.; Mizsak, S. A.; Nielsen, J. W.; Prairie, M. D.
J. Org. Chem. 1988, 53, 4610.
(44) Asai, A.; Nagamura, S.; Saito, H. J. Am. Chem. Soc. 1994, 116,
4171. Sugiyama, H.; Ohmori, K.; Chan, K. L.; Hosada, M.; Asai, A.; Saito,
H.; Saito, I. Tetrahedron Lett. 1993, 34, 2179. Sugiyama, H.; Hosada, M.;
Saito, I.; Asai, A.; Saito, H. Tetrahedron Lett. 1990, 31, 7197. Lin, C. H.;
Patel, D. J. J. Am. Chem. Soc. 1992, 114, 10658. Lin, C. H.; Patel, D. J. J.
Mol. Biol. 1995, 248, 162.

320 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
Figure 3. Top: top and side views of the nucleophile approach for
abnormal ring expansion addition to the duocarmycin A activated
cyclopropane illustrating the destabilizing torsional strain. Bottom:
Abnormal adenine N3 addition confined to the reactant orientations
taken from a model complex of duocarmycin A bound within the w794
DNA high affinity alkylation site illustrating the additional destabilizing
steric interactions: C1-H2/AdC2-H.
Figure 4.
Table 5. Cytotoxic Activity
nucleophilic addition regioselectivity solely controls the DNA
alkylation regioselectivity. This exclusive DNA alkylation
regioselectivity was not only observed in our preceding studies
with duocarmycin A (1), C2, C1,9-12 or 7 and their enantiomers13
but is general with all agents examined to date that undergo
solvolysis with a mixed regioselectivity including duocarmycin
SA (6:1),14 the CPI-based agents (4:1 regioselectivity),16,48,49
and the CBQ-based agents (3:2 regioselectivity).46 Examination
of each of these classes of agents has led only to detection of
adducts derived from adenine N3 addition to the least substituted
cyclopropane carbon. Moreover, each of these studies also
quantitated the adduct formation and, in the case of duocarmycin
A (86-92%),12 duocarmycin SA (95-100%),14 CC-1065 (>
85%),48 and the CBQ-based agents (>75%),42 established that
the regioselectivity of the DNA alkylation reaction is greater
than that of solvolysis. Although several explanations may be
advanced for these observations,42 the two most prominent are
preferential adoption of binding orientations that favor normal
adenine N3 addition (proximity effects) and the significant
destabilizing torsional strain and steric interactions that ac-
company the abnormal addition especially when the reactants
are restricted to the relative orientations found when bound in
the minor groove. Figures illustrating these effects have been
disclosed in our prior work,42 and we would suggest that this
latter subtle effect is most substantial (Figure 3). Consequently,
the clean regioselectivity of the characteristic adenine N3
alkylation reaction benefits not only from stereoelectronic
control but additional important subtle effects characteristic of
a SN2 reaction of a hindered nucleophile that complement the
normally observed regioselectivity as well.
agent
IC50 (L1210, nM)
(+)-1, (+)-duocarmycin A
0.2 nM
1.6 nM
23 nM
14 nM
2 µM
9 µM
(+)-7, epi-(+)-duocarmycin A
ent-(-)-1, ent-(-)-duocarmycin A
ent-(-)-7, ent, epi-(-)-duocarmycin A
(+)-67, (+)-N-BOC-DA
(+)-71, epi-(+)-N-BOC-DA
(-)-67, ent-(-)-N-BOC-DA
(-)-71, ent,epi-(-)-N-BOC-DA
100 µM
>100 µM
hand and of a high stereochemical purity (g99.9% ee and de),
their cytotoxic potency was established, Table 5. Consistent
with the results of our prior studies with the Terashima
samples,13 (+)-duocarmycin was found to be 8× more potent
than epi-(+)-duocarmycin A, and the unnatural enantiomers
were 70-110× less potent than (+)-1. Similarly, (+)-N-BOC-
DA was determined to be 4-5× more potent than epi-(+)-N-
BOC-DA (71), and their unnatural enantiomers were g50× less
potent than (+)-67.
Despite the small differences in cytotoxic potency of the
natural enantiomers but consistent with a well-established
relationship,5 the natural enantiomers follow the trends observed
in prior studies that cover a large range of reactivities with the
chemically more stable agents exhibiting the most potent activity
(Figure 4).
Experimental Section
N,N’-Dibenzoyl-2-(benzyloxy)-6-cyano-1,4-benzoquinone Diimine
(14). Pb(OAc)4 (15.49 g, 33.2 mmol) was added to a solution of 1325
(14.85 g, 33.2 mmol) at 0 °C, and the mixture was stirred for 4 h at 25
°C. The mixture was filtered through a Celite pad, and the filtrate
was washed with saturated aqueous NaCl, dried (MgSO4), and
concentrated. Trituration of the residue with hexane gave 14 (12.4 g,
14.8 g theoretical, 84%) as yellow crystals: mp 118-120 °C (EtOAc,
Cytotoxic Activity. With samples of the natural enantiomers
of 1, 7, 67, and 71 as well as their unnatural enantiomers in
(48) Reynolds, V. L.; Molineaux, I. J.; Kaplan, D. J.; Swenson, D. H.;
Hurley, L. H. Biochemistry 1985, 24, 6228. Hurley, L. H.; Lee, C.-S.;
McGovren, J. P.; Warpehoski, M. A.; Mitchell, M. A.; Kelly, R. C.; Aristoff,
P. A. Biochemistry 1988, 27, 3886. Hurley, L. H.; Warpehoski, M. A.; Lee,
C.-S.; McGovren, J. P.; Scahill, T. A.; Kelly, R. C.; Mitchell, M. A.;
Wicnienski, N. A.; Gebhard, I.; Johnson, P. D.; Bradford, V. S. J. Am.
Chem. Soc. 1990, 112, 4633.
(49) Boger, D. L.; Coleman, R. S.; Invergo, B. J.; Sakya, S. M.; Ishizaki,
T.; Munk, S. A.; Zarrinmayeh, H.; Kitos, P. A.; Thompson, S. C. J. Am.
Chem. Soc. 1990, 112, 4623.
1
yellow plates); H NMR (DMSO-d6, 400 MHz) δ 8.00 (s, 1H), 7.89
(d, 1H, J ) 0.8 Hz), 7.86 (d, 1H, J ) 1.3 Hz), 7.70 (d, 1H, J ) 7.5
Hz), 7.62-7.51 (m, 5H), 7.40 (t, 2H, J ) 7.9 Hz), 7.22 (t, 1H, J ) 7.4
Hz), 7.11 (t, 2H, J ) 7.7 Hz), 6.80 (d, 2H, J ) 7.4 Hz), 6.27 (br s,
1H), 4.90 (br s, 1H), 4.83 (br s, 1H); 13C NMR (DMSO-d6, 100 MHz)
δ 178.5, 176.5, 155.9, 152.1, 146.3, 143.0, 134.1, 133.4, 133.3, 131.8,
131.2, 129.4, 129.0, 128.6, 128.5, 128.24, 128.15, 128.0, 120.3, 113.9,

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 321
104.5, 71.1; IR (KBr) νmax 3435, 3056, 2236, 1667, 1625, 1580, 1446,
1379, 1312, 1246, 1174, 1056, 1030, 841, 748, 728, 697 cm-1. Anal.
Calcd for C28H19N3O3: C, 75.49; H, 4.30; N, 9.43. Found: C, 75.39;
H, 4.35; N, 9.47.
product. Recrystallization from CH2Cl2-hexane gave enantiomericallly
pure material (>99% ee). Colorless plates for X-ray analysis were
obtained by slow recrystallization from CH2Cl2-hexane: 1H NMR
(CDCl3, 400 MHz) δ 9.12 (s, 1H), 8.14 (s, 1H), 7.95-7.88 (m, 5H),
7.58-7.53 (m, 2H), 7.52-7.42 (m, 6H), 7.40-7.28 (m, 8H), 5.15 (d,
1H, J ) 11.4 Hz), 5.10 (d, 1H, J ) 11.4 Hz), 4.69 (dd, 1H, J ) 2.2
Hz, 11.1 Hz), 4.40-4.30 (m, 2H), 3.500 (s, 3H); 3.498 (s, 1H), 3.15-
3.01 (m, 2H); IR (neat) νmax 3286, 2359, 1748, 1659, 1581, 1494, 1422,
1271, 1184, 1104, 1025 cm-1; FABHRMS (NBA-NaI) m/z 738.7301
(M + H+, C41H34F3N3O7 requires 737.7293). The X-ray structure
determination established the relative and S,S absolute stereochemistry.34
6-Allyl-3-(benzyloxy)-2,5-bis(benzoylamino)benzonitrile (15). A
solution of 14 (5.50 g, 12.3 mmol) in anhydrous CH2Cl2 (55 mL) at
-20 °C was treated with BF3‚Et2O (0.76 mL, 6.18 mmol, 0.5 equiv).
The mixture was stirred for 30 min and treated with allyltributyltin
(4.21 mL, 13.6 mmol, 1.1 equiv). The reaction mixture was stirred
for 3 h at -20 °C. Acetonitrile (250 mL) was added to the reaction
mixture, the mixture was washed with hexane (2 × 100 mL), and the
acetonitrile extract was concentrated in Vacuo to afford crude 15. The
residue was crystallized from EtOAc to provide 15 (5.14 g, 6.02 g
theoretical, 85%, typically 83-89%) as a white solid: mp 247.5-249.0
°C (CH3OH, white needles); 1H NMR (DMSO-d6, 400 MHz) δ 10.28
(s, 1H), 10.17 (s, 1H), 7.90-8.00 (m, 4H), 7.42-7.65 (m, 9H), 7.25-
7.35 (m, 3H), 5.85 (m, 1H), 5.17 (s, 2H), 5.00 (dd, 1H, J ) 10.1, 1.5
Hz), 4.95 (dd, 1H, J ) 17.1, 1.5 Hz), 3.58 (d, 2H, J ) 6.0 Hz); 13C
NMR (DMSO-d6, 100 MHz) δ 166.0, 153.0, 137.0, 136.7, 136.4, 134.6,
133.9, 133.7, 132.0, 131.9, 131.1, 128.6, 128.5, 128.4, 128.2, 127.9,
127.7, 127.2, 127.0, 118.1, 116.4, 115.5, 114.2, 70.2, 33.9; IR (KBr)
νmax 3251, 2226, 1651, 1600, 1580, 1523, 1492, 1451, 1415, 1344,
1308, 1272, 1241, 1144, 1103, 1026 cm-1; FABHRMS (NBA-CsI) m/z
620.0955 (M + Cs+, C31H25N3O3 requires 620.0950). Anal. Calcd
for C31H25N3O3: C, 76.37; H, 5.17; N, 8.62. Found: C, 76.31; H,
5.24; N, 8.51.
3-(Benzyloxy)-2,5-bis(benzoylamino)-6-[2-hydroxy-3-[(p-toluene-
sulfonyl)oxy]propyl]benzonitrile (17). A solution of 16 (2.20 g, 4.22
mmol) and Bu2SnO (1.14 g, 4.60 mmol, 1.1 equiv) in 150 mL of
toluene-THF (10:1) was warmed at reflux with azeotropic removal
of H2O with a Dean-Stark trap. After 6 h, 50 mL of solvent was
distilled from the reaction vessel, and the solution was cooled to 0 °C.
Et3N (0.02 g, 0.2 mmol, 0.05 equiv) and p-TsCl (0.95 g, 5.0 mmol,
1.2 equiv) were added, and the reaction mixture was warmed to 25 °C
and stirred overnight. The reaction mixture was quenched with the
addition of 50 mL of saturated aqueous NaCl, and the aqueous layer
was extracted with EtOAc (3 × 60 mL). The combined organic layer
was dried (Na2SO4) and concentrated under reduced pressure. Flash
chromatography (SiO2, 50% EtOAc-hexane) afforded 17 (2.54 g, 2.85
g theoretical, 89%; typically 89-94%) as white crystals: mp 165-
1
167 °C (CH3OH, white needles); H NMR (CDCl3, 400 MHz) δ 9.91
(s, 1H), 8.12 (s, 1H), 7.96 (d, 2H, J ) 7.0 Hz), 7.88-7.86 (m, 3H),
7.79 (d, 2H, J ) 8.4 Hz), 7.59-7.43 (m, 6H), 7.33-7.28 (m, 7H),
5.13 (d, 1H, J ) 11.3 Hz), 5.08 (d, 1H, J ) 11.3 Hz), 4.31-4.29 (m,
1H), 4.24 (dd, 1H, J ) 10.6, 2.6 Hz), 4.10 (dd, 1H, J ) 10.6, 7.9 Hz),
3.56 (s, 1H), 3.06-2.96 (m, 2H), 2.43 (s, 3H); 13C NMR (CDCl3, 100
MHz) δ 166.3, 165.8, 151.6, 145.2, 137.2, 135.5, 134.3, 134.1, 133.6,
132.5, 130.0, 129.1, 128.8, 128.7, 128.6, 128.2, 127.9, 127.6, 127.4,
125.6, 125.1, 115.5, 115.1, 113.0, 112.7, 73.0, 71.0, 70.9, 32.8, 25.4;
IR (film) νmax 3280, 2929, 1667, 1601, 1515, 1422, 1360, 1272, 1176,
1096 cm-1; FABHRMS (NBA-CsI) m/z 808.1105 (M + Cs+,
C38H33N3O7S requires 808.1094).
3-(Benzyloxy)-2,5-bis(benzoylamino)-6-(2,3-dihydroxypropyl)-
benzonitrile (16). A solution of 15 (5.14 g, 10.5 mmol) and
N-methylmorpholine N-oxide (2.47 g, 21.1 mmol, 2 equiv) in 4:1
acetone-H2O (300 mL) was treated with OsO4 (0.393 M toluene
solution, 5.36 mL, 2.11 mmol, 0.2 equiv) and stirred overnight at 25
°C under N2 in the dark. Sodium sulfite (1.0 g) was added, and the
mixture was stirred for an additional 1 h. The mixture was diluted
with saturated aqueous NaCl (300 mL) and extracted with EtOAc (3
× 150 mL). The combined organic extract was dried (MgSO4) and
concentrated in Vacuo. The residue was crystallized from CH3OH to
afford 9 (5.22 g, 5.50 g theoretical, 95%) as a white solid: mp 231-
1
232 °C (CH3OH, white needles); H NMR (DMSO-d6, 400 MHz) δ
3-(Benzyloxy)-2,5-bis(benzoylamino)-6-[2-[(tert-butyldimethylsi-
lyl)oxy]-3-[(p-toluenesulfonyl)oxy]propyl]benzonitrile (18). A solu-
tion of 17 (2.25 g, 3.33 mmol) in 100 mL of CH2Cl2 at 0 °C was treated
with 2,6-lutidine (0.71 g, 6.6 mmol, 2 equiv) and TBDMSOTf (1.32
g, 5.00 mmol, 1.5 equiv), and the solution was stirred at 0 °C for 3 h.
The reaction mixture was quenched with the addition of saturated
aqueous NaHCO3 and extracted with EtOAc (3 × 40 mL). The
combined organic layer was dried (Na2SO4) and concentrated under
reduced pressure. Flash chromatography (SiO2, 17% EtOAc-hexane)
afforded 18 (1.92 g, 2.60 g theoretical, 73%; typically 67-75%) as
10.69 (s, 1H), 10.25 (s, 1H), 7.95-8.05 (m, 5H), 7.52-7.67 (m, 6H),
7.43 (br d, 2H, J ) 6.6 Hz), 7.23-7.34 (m, 3H), 5.99 (br s, 1H), 5.17
(s, 2H), 5.05 (br s, 1H), 3.78 (m, 1H), 3.41 (m, 2H), 3.06 (dd, 1H, J )
14.4, 4.7 Hz), 2.89 (dd, 1H, J ) 14.4, 8.8 Hz); 13C NMR (DMSO-d6,
100 MHz) δ 166.1, 165.0, 152.8, 137.6, 136.5, 134.0, 133.8, 132.2,
132.0, 128.8, 128.6, 128.4, 127.94, 127.91, 127.7, 127.4, 127.2, 126.8,
115.7, 114.7, 114.5, 72.9, 70.2, 65.3, 34.0; IR (KBr) νmax 3421, 2933,
2226, 1656, 1605, 1580, 1518, 1415, 1349, 1308, 1272, 1241, 1149,
1097, 1072, 1021 cm-1; FABHRMS (NBA) m/z 522.2029 (M + H+,
C31H27N3O5 requires 522.2029). Anal. Calcd for C31H27N3O5: C,
71.39; H, 5.22; N, 8.06. Found: C, 71.35; H, 5.52; N, 7.89.
1
white needles: mp 143-146 °C (EtOAc, white crystals); H NMR
(CDCl3, 400 MHz) δ 9.18 (s, 1H), 8.20 (s, 1H), 8.05-8.02 (m, 2H),
7.96 (s, 1H), 7.89 (d, 2H, J ) 7.9 Hz), 7.84 (s, 1H), 7.79 (d, 2H, J )
8.3 Hz), 7.59-7.48 (m, 4H), 7.43 (t, 2H, J ) 7.4 Hz), 7.33-7.27 (m,
6H), 5.19 (1H, d, J ) 11.6 Hz), 5.15 (d, 1H, J ) 11.6 Hz), 4.31-4.28
(m, 1H), 4.25 (dd, 1H, J ) 9.1, 4.0 Hz), 4.06 (dd, 1H, J ) 9.1, 2.1
Hz), 3.31 (dd, 1H, J ) 14.7, 11.8 Hz), 2.91 (dd, 1H, J ) 14.7, 2.8
Hz), 2.42 (s, 3H), 0.60 (s, 9H), -0.11 (s, 3H), -0.41 (s, 3H); 13C NMR
(CDCl3, 100 MHz) δ 165.9, 165.4, 151.1, 145.3, 136.9, 135.4, 132.5,
132.3, 132.2, 130.1, 130.0, 129.1, 128.7, 128.4, 128.0, 127.6, 127.5,
127.4, 127.3, 126.0, 125.0, 124.9, 115.4, 112.7, 112.2, 72.5, 71.3, 70.6,
33.2, 25.5, 21.6, 17.7, -5.4, -5.5; IR (film) νmax 3045, 2928, 1665,
1600, 1520, 1424, 1363, 1266, 1176, 1096 cm-1; FABHRMS (NBA-
CsI) m/z 922.1933 (M + Cs+, C44H47N3O7SSi requires 922.1958).
Asymmetric Dihydroxylation of 15. A solution of OsO4 (0.4 M
in toluene, 0.32 mL, 0.127 mmol, 0.01 equiv), K3Fe(CN)6 (12.6 g, 38.1
mmol), K2CO3 (7 g), CH3SO2NH2 (0.5 g, 5.0 mmol, 0.4 equiv), and
(DHQD)2-PHAL (643 mg, 1.27 mmol, 0.1 equiv) was added to a
solution of 15 (6.2 g, 12.7 mmol) in THF-H2O (4:1, 400 mL) at 0 °C,
and the reaction mixture was stirred overnight at 25 °C. The reaction
mixture was poured into EtOAc, and the organic layer was washed
with aqueous 1 N HCl (2 × 100 mL), saturated aqueous NaHCO3, and
saturated aqueous NaCl, dried (MgSO4), and concentrated. Trituration
of the residue with Et2O gave 6.1 g (92%) of 16: mp 228-230 °C.
The enantiomeric excess of 16 was determined upon conversion to 23
and analysis on a ChiralCel OD HPLC column (Daicel, 0.46 × 25 cm,
eluent ) 13% 2-propanol-hexane, flow rate ) 0.3 mL/min). The
effluent was monitored at 254 nm, and the enantiomers (3R)-23 and
(3S)-23 were eluted with retention times of 27.5 and 31.5 min,
respectively.
1-Benzoyl-6-(benzoylamino)-7-(benzyloxy)-3-[(tert-butyldimeth-
ylsilyl)oxy]-5-cyano-1,2,3,4-tetrahydroquinoline (19). From 18. A
solution of 18 (4.65 g, 5.89 mmol) in THF (130 mL) at 0 °C was treated
with NaH (60% dispersion in oil, 0.47 g, 11.8 mmol, 2.0 equiv) in
several portions over 15 min. After 2 h, the reaction mixture was
quenched by the addition of saturated aqueous NaHCO3 (50 mL), and
the aqueous layer was extracted with EtOAc (4 × 50 mL). The
combined organic layer was dried (Na2SO4) and concentrated in Vacuo.
Chromatography (SiO2, 2 × 10 cm, 30% EtOAc-hexane) provided
19 (3.37 g, 3.63 g theoretical, 93%; typically 92-97%) as a white
(S)-Mosher Ester of 16. (R)-MTPA-Cl (72 µL, 0.38 mmol) was
added to a mixture of (S)-16 (200 mg, 0.38 mmol, >90% ee derived
from the (DHQD)2-PHAL ADH) and Et3N (54 µL, 0.57 mmol) in CH2-
Cl2 at 0 °C. After being stirred overnight at 25 °C, the reaction mixture
was washed with 5% aqueous HCl, saturated aqueous NaHCO3, and
saturated aqueous NaCl, dried (MgSO4), and concentrated. Chroma-
tography (SiO2, 1 × 20 cm, 33% EtOAc-hexane) followed by
crystallization from EtOAc-hexane gave 226 mg (81%, >95% ee) of
1
solid: mp 145-147 °C (CH3OH, white needles); H NMR (CDCl3,
400 MHz) δ 7.86 (br d, 2H, J ) 7.2 Hz), 7.74 (s, 1H), 7.35-7.59 (m,

322 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
8H), 7.27-7.32 (m, 3H), 7.13-7.18 (m, 2H), 6.71 (br s, 1H), 4.53 (s,
2H), 4.40 (m, 1H), 4.28 (br d, 1H, J ) 12.8 Hz), 3.54 (dd, 1H, J )
12.8, 2.0 Hz), 3.24 (dd, 1H, J ) 17.6, 5.1 Hz), 3.05 (dd, 1H, J ) 17.6,
3.4 Hz), 0.84 (s, 9H), 0.10 (s, 3H), 0.03 (s, 3H); 13C NMR (CDCl3,
100 MHz) δ 171.1, 165.7, 149.5, 137.6, 135.6, 135.4, 133.5, 132.3,
130.7, 128.7, 128.5, 128.4, 127.6, 127.1, 126.3, 123.2, 115.0, 114.5,
111.5, 71.4, 64.6, 50.5, 35.5, 25.7, 18.1, -4.9; IR (KBr) νmax 3292,
2933, 2851, 2215, 1656, 1600, 1580, 1492, 1390, 1328, 1256, 1210,
1174, 1092 cm-1; FABHRMS (NBA-CsI) m/z 750.1764 (M + Cs+,
C37H39N3O4Si requires 750.1764). Anal. Calcd for C37H39N3O4Si: C,
71.93; H, 6.36; N, 6.80. Found: C, 71.96; H, 6.31; N, 7.04.
From 25. A suspension of NaH (387 mg, 60%, 9.68 mmol, 2.5
equiv) in THF (20 mL) at 0 °C under N2 was treated with a solution
of 2525 (2.76 g, 3.87 mmol) in THF (20 mL). The reaction mixture
was stirred for 4 h at 25 °C. The reaction mixture was poured into
ice-cold 10% aqueous HCl (100 mL) and extracted with EtOAc (3 ×
80 mL). The combined organic extract was washed with saturated
aqueous NaCl, dried (MgSO4), and concentrated in Vacuo. Chroma-
tography (SiO2, 15-30% EtOAc-hexane gradient elution) afforded
19 (2.32 g, 2.39 g theoretical, 97%) as a white solid identical in all
respects to that described above.
1368, 1331, 1253, 1149, 1104, 1064, 1004 cm-1; FABHRMS (NBA-
CsI) m/z 742.2279 (M + Cs+, C33H47N3O6Si requires 742.2288).
ChiralCel OD HPLC (2 × 25 cm, 5 mL/min, 3% i-PrOH-hexane): tR
) 42 min for (+)-(3R)-22 and tR ) 48 min for (-)-(3S)-22.
(3S)-22: [R]25 -12.4 (c 0.5, CH3OH).
D
(3R)-22: [R]25 +13.0 (c 0.1, CH3OH).
D
From 36. A solution of 3625 (2.75 g, 3.52 mmol) in THF (100 mL)
at 0 °C was treated with NaH (60% dispersion in oil, 0.28 g, 7.04
mmol, 2 equiv) in several portions over 15 min. After 1 h, the reaction
mixture was quenched by the addition of saturated aqueous NaHCO3
and the aqueous layer was extracted with EtOAc (4 × 30 mL). The
combined organic layer was dried (Na2SO4) and concentrated in Vacuo.
PCTLC (SiO2, 4 mm plate, 20% EtOAc-hexane) provided 22 (2.04
g, 2.15 g theoretical, 95%) as an amorphous solid identical in all respects
to that described above.
Methyl 2-[N-(tert-Butyloxycarbonyl)-N-[7-(benzyloxy)-3(S)-[(tert-
butyldimethylsilyl)oxy]-1-(tert-butyloxycarbonyl)-5-cyano-1,2,3,4-
tetrahydroquinolin-6-yl]amino]propionate ((-)-37). A solution of
(-)-22 (105 mg, 0.17 mmol) in DMF (5 mL) at 0 °C was treated with
NaH (60% dispersion in oil, 14 mg, 2 equiv), and the solution was
stirred for 0.5 h. 2-Bromopropionate (57 mg, 2 equiv) was added, and
the solution was stirred for an additional 0.5 h at 0 °C. The reaction
mixture was quenched by the addition of saturated aqueous NaHCO3
(5 mL), and the aqueous layer was extracted with EtOAc (4 × 10 mL).
The combined organic extract was washed with H2O (2 × 10 mL),
dried (Na2SO4), and concentrated under reduced pressure. PCTLC
(SiO2, 2 mm plate, 15% EtOAc-hexane) provided (-)-37 (115 mg,
120 mg theoretical, 96%): 1H NMR (DMSO-d6, 400 MHz) δ 7.75-
7.68 (m, 1H), 7.43-7.33 (m, 5H), 5.14-5.02 (m, 2H), 4.34-4.23 (m,
2H), 3.89-3.77 (m, 1H), 3.56-3.29 (m, 4H), 3.12-3.05 (m, 1H), 2.73-
2.62 (m, 2H), 1.49-1.46 (m, 9H), 1.37-1.13 (m, 12H), 0.85-0.79
(m, 9H), 0.10-0.04 (m, 6H); IR (film) νmax 2930, 1709, 1488, 1368,
1317, 1253, 1154, 1101 cm-1; FABHRMS (NBA-CsI) m/z 828.2625
(M + Cs+, C37H53N3O8Si requires 828.2656).
Resolution of 19. A solution of 19 (150 mg of racemic or 250 mg
of 89:11 3S:3R-19 in 1 mL of CH3OH) was separated on a ChiralCel
OD (Daicel) semipreparative column (2 cm × 25 cm) using 20-25%
2-propanol-hexane eluent at a flow rate of 5 mL/min. The effluent
was monitored at 254 nm, and the enantiomers (3R)-19 and (3S)-19
eluted with retention times of 32-48 and 74-93 min, respectively.
HPLC analysis of the separated enantiomers (>95% recovery) indicated
that both were >99.9% ee.
For (3S)-19: [R]20 -116.0 (c 0.50, CH3OH).
D
For (3R)-19:[R]20 +116 (c 0.5, CH3OH).
D
Anal. Calcd for C37H39N3O4Si: C, 71.93; H, 6.36; N, 6.80.
Found: C, 71.76; H, 6.70; N, 6.55.
7-(Benzyloxy)-3-[(tert-butyldimethylsilyl)oxy]-1-(tert-butyloxycar-
bonyl)-6-[bis(tert-butyloxycarbonyl)amino]-5-cyano-1,2,3,4-tetrahy-
droquinoline (21). A solution of 19 (4.44 g, 7.21 mmol) in EtOH (28
mL) was treated with 280 mL of 98% NH2NH2 under Ar and the
reaction mixture warmed at 140 °C in a sealed vessel for 20 h. The
mixture was cooled, poured onto H2O (600 mL), and extracted with
EtOAc (3 × 200 mL). The organic phase was washed with saturated
aqueous NaCl (150 mL), dried (MgSO4), and concentrated in Vacuo.
Chromatography (SiO2, 10-20% EtOAc-hexane) afforded 20 (1.92
g, 2.95 g theoretical, 65%), which was used immediately in the
following reaction, and variable amounts of the corresponding free
alcohol (45-30%). A solution of 20 (1.85 g, 4.52 mmol), BOC2O
(5.92 g, 27.1 mmol, 6 equiv), and DMAP (0.05 g, 0.45 mmol, 0.1 equiv)
in THF (100 mL) was warmed at reflux for 2 h. The reaction mixture
was cooled to 25 °C and concentrated in Vacuo. PCTLC (SiO2, 4 mm
plate, 0-20% EtOAc-hexane gradient elution) provided 21 (3.05 g,
(-)-(3’S)-37: [R]25 -17 (c 0.3, CH3OH).
D
ent-(+)-(3’R)-37: [R]25 +15 (c 0.2, CH3OH).
D
Dieckmann Closure of 37. A solution of (-)-37 (63 mg, 0.09
mmol) in THF (5 mL) at -78 °C was treated with LDA (1.35 mL of
a 0.2 M solution in THF, 3 equiv). After 1 h at -78 °C, the reaction
mixture was quenched by the addition of saturated aqueous NaHCO3
(1 mL), and the aqueous layer was extracted with EtOAc (4 × 2 mL).
The combined organic extract was dried (Na2SO4) and concentrated in
Vacuo. PCTLC (SiO2, 2 mm plate, 15% EtOAc-hexane, R ) 1.25)
provided (+)-38 (18 mg, 29% Rf ) 0.4) and (-)-45 (16 mg, 25% Rf )
0.5) as colorless oils and recovered starting material (25 mg, 40%).
(3S)-7-(Benzyloxy)-3-[(tert-butyldimethylsilyl)oxy]-6-[N-(tert-bu-
tyloxycarbonyl)-N-[1-[(4R)-2-oxo-4-isopropyloxazolidin-3-yl)carbo-
nyl]ethyl]amino]-5-cyano-1,2,3,4-tetrahydroquinoline (61a). Pre-
pared following the procedure detailed below: amorphous solid, 4:1
diastereomeric mixture: 1H NMR (CDCl3, 400 MHz) δ 7.74 (s, 0.8
H), 7.67 (s, 0.2H), 7.46-7.25 (m, 5H), 5.80-5.70 (m, 1H), 5.09 (s,
1.6H), 5.08 (s, 0.4H), 4.48-4.08 (m, 3H), 3.88-3.70 (m, 1H), 3.50-
3.38 (m, 1.8H), 3.16-3.10 (m, 1H), 2.84 (dd, 0.2H, J ) 6.8 Hz, 16.3
Hz), 2.40-2.23 (m, 1H), 1.57 (s, 9H), 1.51 (s, 9H), 1.40-1.30 (m,
3H), 0.89 (s, 9H), 0.88 (br s, 6H), 0.12 (s, 3H), 0.11 (s, 3H); IR (neat)
νmax 3497, 3415, 2923, 2226, 1779, 1709, 1598, 1482, 1455, 1389,
1364, 1251, 1205, 1154, 748, 702 cm-1; FABHRMS (NBA-CsI) m/z
925.3163 (M + Cs+, C42H60N4O9Si requires 925.3184); [R]25D -12 (c
0.5, CH3OH).
1
3.21 g theoretical, 95%): white amorphous solid; H NMR (CDCl3,
400 MHz) δ 7.71 (s, 1H), 7.26-7.36 (m, 5H), 5.08 (s, 2H), 4.14-4.11
(m, 1H), 3.79 (dd, 1H, J ) 2.8, 12.9 Hz), 3.49 (dd, 1H, J ) 7.6, 12.9
Hz), 3.11 (dd, 1H, J ) 5.6, 17.1 Hz), 2.80 (dd, 1H, J ) 6.2, 17.1 Hz),
1.56 (s, 9H), 1.40 (s, 9H), 1.38 (s, 9H), 0.86 (s, 9H), 0.10 (s, 6H); IR
(neat) νmax 2978, 2931, 2229, 1801, 1763, 1708, 1599, 1489, 1369,
1253, 1154, 1102 cm-1; FABHRMS (NBA-CsI) m/z 842.2848 (M +
Cs+, C38H55N3O8Si requires 842.2813).
(3S)-21: [R]25 -16.4 (c 0.15, CH3OH).
D
(3R)-21: [R]25 +16.0 (c 0.33, CH3OH).
D
7-(Benzyloxy)-3-[(tert-butyldimethylsilyl)oxy]-1-(tert-butyloxycar-
bonyl)-6-[(tert-butyloxycarbonyl)amino]-5-cyano-1,2,3,4-tetrahyd-
roquinoline (22). From 21. A solution of 21 (1.05 g, 1.48 mmol) in
CH2Cl2 (120 mL) at 0 °C was treated with TFA (0.84 g, 7.39 mmol,
5 equiv), and the solution was stirred for 2 h. The reaction was
quenched by the slow addition of saturated aqueous NaHCO3 (10 mL).
The organic layer was separated, dried (Na2SO4), and concentrated in
Vacuo. PCTLC (SiO2, 4 mm plate, 30% EtOAc-hexane) provided 22
(3S)-7-(Benzyloxy)-3-[(tert-butyldimethylsilyl)oxy]-6-[N-(tert-bu-
tyloxycarbonyl)-N-[1-[((4S)-2-oxo-4-isopropyloxazolidin-3-yl)carbo-
nyl]ethyl]amino]-5-cyano-1,2,3,4-tetrahydroquinoline (61b). A sol-
ution of (-)-(3S)-22 (80 mg, 0.13 mmol) in DMF (5 mL) at 0 °C was
treated with NaH (60% dispersion in oil, 10.4 mg, 0.26 mmol, 2 equiv).
After 0.5 h at 0 °C, 60b39 (69 mg, 0.26 mmol, 2 equiv) was added, and
the solution was stirred an additional 0.5 h. The reaction mixture was
quenched by the addition of saturated aqueous NaHCO3 (5 mL), and
the aqueous layer was extracted with EtOAc (4 × 5 mL). The
combined organic extract was washed with H2O (2 × 5 mL), dried
(Na2SO4), and concentrated in Vacuo. PCTLC (SiO2, 2 mm plate, 20%
EtOAc-hexane) provided (-)-(3S,4’S)-61b (89 mg, 104 mg theoretical,
86%): amorphous solid, 3:2 diastereomeric mixture; 1H NMR (CDCl3,
400 MHz) δ 8.07 (s, 0.6H), 7.67 (s, 0.4H), 7.45-7.25 (m, 5H), 6.30-
1
(0.84 g, 0.93 g theoretical, 93%): white amorphous solid; H NMR
(CDCl3, 400 MHz) δ 7.66 (s, 1H), 7.25-7.45 (m, 5H), 6.34 (s, 1H),
5.08 (s, 2H), 4.15-4.16 (m, 1H), 3.74 (d, 1H, J ) 12.7 Hz), 3.50 (dd,
1H, J ) 7.6, 12.7 Hz), 3.15 (dd, 1H, J ) 5.6, 17.2 Hz), 2.83 (dd, 1H,
J ) 6.2, 17.2 Hz), 1.54 (s, 9H), 1.52 (s, 9H), 0.89 (s, 9H), 0.12 (s,
6H); IR (KBr) νmax 3272, 2930, 2256, 1691, 1598, 1501, 1461, 1392,

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 323
5.98 (m, 0.2H), 5.85-5.75 (m, 0.8H), 5.20-4.95 (m, 2H), 4.52-3.81
(m, 4H), 3.51-3.00 (m, 3H), 2.92-2.78 (m, 1H), 2.44-2.22 (m, 1H),
1.53 (s, 9H), 1.51 (s, 5.4H), 1.41 (s, 3.6H), 1.20-1.10 (m, 3H), 1.85-
1.72 (m, 15H), 0.11 (s, 3H), 0.10 (s, 3H); IR (neat) νmax 3497, 3416,
2922, 2222, 1782, 1709, 1599, 1490, 1456, 1389, 1364, 1251, 1205,
1155, 744, 701 cm-1; FABHRMS (NBA-CsI) m/z 925.3152 (M + Cs+,
C42H60N4O9Si requires 925.3184).
Vacuo. PCTLC (SiO2, 0.3 mm plate, 25% EtOAc-hexane) provided
(+)-62b (1.3 mg, 17%) and (+)-63b (5.6 mg, 70%).
(2R,8S)-4-(Benzyloxy)-3,6-bis(tert-butyloxycarbonyl)-8-[(tert-bu-
tyldimethylsilyl)oxy]-1-imino-2-methyl-2-[[(4S)-2-oxo-4-isopropylox-
azolidin-3-yl]carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-f]-
quinoline ((+)-62b): 1H NMR (CDCl3, 400 MHz) δ 9.15 (br s, 1H),
7.50 (br s, 1H), 7.39 (d, 2H, J ) 6.9 Hz), 7.33-7.25 (m, 3H), 5.04 (s,
2H), 4.49-4.46 (m, 1H), 4.25 (t, 1H, J ) 8.8 Hz), 4.16 (dd, 1H, J )
9.0, 2.8 Hz), 4.14-4.10 (m, 1H), 3.88 (br d, 1H, J ) 11.5 Hz), 3.44-
3.40 (m, 1H), 3.33 (dd, 1H, J ) 16.9, 6.0 Hz), 2.86 (dd, 1H, J ) 16.9,
5.3 Hz), 2.55-2.48 (m, 1H), 1.70 (s, 3H), 1.51 (s, 9H), 1.27 (s, 9H),
0.91 (apparent t, 6H, J ) 6.9 Hz), 0.86 (s, 9H), 0.11 (s, 3H), 0.10 (s,
3H); IR (film) νmax 2956, 2917, 1784, 1697, 1362, 1311, 1249, 1151
cm-1; FABHRMS (NBA-CsI) m/z 925.3162 (M + Cs+, C42H60N4O9Si
requires 925.3184).
(3S,4’S)-61b: [R]25 -17 (c 0.3, CH3OH).
D
(3R,4’R)-61b: [R]25 +15 (c 0.2, CH3OH).
D
Dieckmann Reaction of (-)-(3S,4’R)-61a. Kinetic conditions: a
precooled (-78 °C) solution of (-)-61a (10 mg, 0.0126 mmol) in THF
(0.2 mL) was added to a solution of LDA (0.5 M, 150 µL, 0.075 mmol)
in THF (1 mL) at -78 °C over a period of 10 min. The reaction
mixture was quenched with the addition of saturated aqueous NH4Cl
and extracted with EtOAc. The organic layer was washed with
saturated aqueous NaCl, dried (MgSO4), and concentrated. Purification
of the residue by PTLC (SiO2, 30% EtOAc-hexane) gave 62a (1.2
mg, 12%) and 63a (5.6 mg, 56%).
(2R,8S,4’S)-62b: [R]26 +90 (c 0.08, THF).
D
ent-(2S,8R,4’R)-62b: [R]25 -87 (c 0.1, THF).
D
(2S,8S)-4-(Benzyloxy)-3,6-bis(tert-butyloxycarbonyl)-8-((tert-bu-
tyldimethylsilyl)oxy)-1-imino-2-methyl-2-[[(4S)-2-oxo-4-isopropylox-
azolidin-3-yl]carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-f]-
quinoline ((+)-63b): 1H NMR (CDCl3, 400 MHz) δ 9.08 (br s, 1H),
7.57 (br s, 1H), 7.44 (d, 2H, J ) 7.0 Hz), 7.34-7.27 (m, 3H), 5.05 (s,
2H), 4.43-4.40 (m, 1H), 4.27-4.15 (m, 3H), 4.03-4.00 (m, 1H), 3.33-
3.30 (m, 1H), 3.08-3.06 (m, 1H), 2.92 (dd, 1H, J ) 16.8, 8.2 Hz),
2.57-2.53 (m, 1H), 1.69 (s, 3H), 1.51 (s, 9H), 1.30 (s, 9H), 0.90
(apparent t, 6H, J ) 7.2 Hz), 0.89 (s, 9H), 0.11 (s, 3H), 0.10 (s, 3H);
Thermodynamic conditions: LDA (0.5 M, 150 µL, 0.075 mmol) in
THF was added to a solution of 61a (10 mg, 0.0126 mmol) in THF
(0.2 mL) at -78 °C. The mixture was stirred for 30 min below -50
°C before being quenched with the addition of saturated aqueous
NaHCO3 and extracted with EtOAc. The organic layer was washed
with saturated aqueous NaCl, dried (MgSO4), and concentrated. PTLC
(SiO2, 30% EtOAc-hexane) gave 62a (7.8-8.1 mg, 78-81%).
(2R,8S)-4-(Benzyloxy)-3,6-bis(tert-butyloxycarbonyl)-8-[(tert-bu-
tyldimethylsilyl)oxy]-1-imino-2-methyl-2-[[(4R)-2-oxo-4-isopro-
pyloxazolidin-3-yl]carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-
IR (film) νmax 2964, 2923, 1785, 1698, 1364, 1308, 1251, 1149 cm-1
;
FABHRMS (NBA-CsI) m/z 925.3179 (M + Cs+, C42H60N4O9Si requires
1
925.3184).
f]quinoline ((-)-62a): H NMR (CDCl3, 400 MHz) δ 9.07 (s, 1H),
(2S,8S,4’S)-63b: [R]26 +23 (c 0.02, THF).
D
7.55 (s, 1H), 7.41-7.25 (m, 5H), 5.00 (s, 2H), 4.45-4.20 (m, 4H),
4.12-4.01 (m, 1H), 3.23-3.18 (m, 1H), 3.01-2.96 (m, 1H), 2.90-
2.86 (m, 1H), 2.56-2.55 (m, 1H), 1.66 (s, 3H), 1.52 (s, 9H), 1.31 (s,
9H), 0.92-0.88 (m, 6H), 0.88 (s, 9H), 0.08 (s, 3H), 0.07 (s, 3H); IR
ent-(2R,8R,4’R)-63b: [R]25 -20 (c 0.01, THF).
D
Methyl (2R,8S)-4-(Benzyloxy)-8-[(tert-butyldimethylsilyl)oxy]-3,6-
bis(tert-butyloxycarbonyl)-1-imino-2-methyl-2,3,6,7,8,9-hexahydro-
1H-pyrrolo[3,2-f]quinoline-2-carboxylate ((+)-38).50 From 62a. A
solution of 62a (11 mg, 0.014 mmol) in THF-CH3OH (4:1, 2 mL) at
0 °C was treated with LiOCH3 (2 M in CH3OH, 0.21 mL, 0.42 mmol,
3.0 equiv). After 1 h at 0 °C, the reaction mixture was quenched by
the addition of saturated aqueous NaHCO3 (2 mL). The aqueous layer
was extracted with EtOAc (4 × 2 mL), and the combined organic
extract was dried (Na2SO4) and concentrated in vacuo. PCTLC (SiO2,
0.3 mm plate, 30% EtOAc-hexane) provided (+)-38 (7.9 mg, 9.6 mg
theoretical, 82%): 1H NMR (CDCl3, 400 MHz) δ 9.34 (br s, 1H), 7.52
(br s, 1H), 7.49 (d, 1H, J ) 6.9 Hz), 7.38-7.26 (m, 3H), 5.12 (d, 1H,
J ) 11.2 Hz), 5.08 (d, 1H, J ) 11.2 Hz), 4.15-4.09 (m, 1H), 3.98-
3.95 (m, 1H), 3.68 (s, 3H), 3.40 (dd, 1H, J ) 18.8, 6.2 Hz), 3.37-
3.32 (m, 1H), 3.00-2.96 (m, 1H), 2.01 (br s, 1H), 1.72 (s, 3H), 1.51
(s, 9H), 1.35 (s, 9H), 0.87 (s, 9H), 0.10 (s, 3H), 0.09 (s, 3H); IR (film)
(neat) νmax 2966, 2924, 1788, 1700, 1366, 1310, 1252, 1149 cm-1
;
FABHRMS (NBA-CsI) m/z 925.3181 (M + Cs+, C42H60N4O9Si requires
925.3184).
(2S,8S,4’R)-62a: [R]25 -35 (c 0.10, THF).
D
(2S,8S)-4-(Benzyloxy)-3,6-bis(tert-butyloxycarbonyl)-8-[(tert-bu-
tyldimethylsilyl)oxy]-1-imino-2-methyl-2-[[(4R)-2-oxo-4-isopro-
pyloxazolidin-3-yl]carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-
f]quinoline ((-)-63a): 1H NMR (CDCl3, 400 MHz) δ 9.07 (s, 1H),
7.60 (s, 1H), 7.46-7.26 (m, 5H), 5.01 (s, 2H), 4.50-4.45 (m, 1H),
4.30-4.20 (m, 3H), 4.00-3.89 (m, 1H), 3.35-3.31 (m, 1H), 3.07-
3.05 (m, 1H), 2.93-2.91 (m, 1H), 2.71-2.66 (m, 1H), 1.66 (s, 3H),
1.52 (s, 9H), 1.31 (s, 9H), 0.93-0.88 (m, 6H), 0.87 (s, 9H), 0.10 (s,
3H), 0.09 (s, 3H); IR (neat) νmax 2966, 2924, 1785, 1699, 1365, 1309,
1251, 1148 cm-1; FABHRMS (NBA-CsI) m/z 925.3179 (M + Cs+,
C42H60N4O9Si requires 925.3184).
ν
max 2928, 2845, 1743, 1697, 1501, 1367, 1249, 1146 cm-1; FABHRMS
(NBA-CsI) m/z 828.2689 (M + Cs+, C37H53N3O8Si requires 828.2656).
(2S,8S,4’R)-63a: [R]25 -68 (c 0.05, THF).
D
(2R,8S)-38: [R]25 +75 (c 0.06, THF).
D
Dieckmann Reaction of (-)-(3S,4’S)-61b. Kinetic Conditions.
A solution of (-)-61b (42 mg, 0.053 mmol) in THF (0.5 mL) was
added slowly via cannula to a solution of LDA (1 mL of a 0.2 M
solution in THF, 4 equiv) at -78 °C. After 2 h, the reaction mixture
was quenched by the addition of saturated aqueous NaHCO3 (0.5 mL),
and the aqueous layer was extracted with EtOAc (3 × 2 mL). The
combined organic extract was dried (Na2SO4) and concentrated in
Vacuo. PCTLC (SiO2, 1 mm plate, 25% EtOAc-hexane, R ) 1.7)
provided (+)-62b (21.4 mg, 51%, Rf ) 0.35) and (+)-63b (5.9 mg,
14%, Rf ) 0.60) as colorless oils.
Thermodynamic Conditions. A solution of (-)-61b (58 mg, 0.073
mmol) in THF (0.5 mL) at -78 °C was treated with LDA (1.1 mL of
a 0.2 M solution in THF, 3 equiv). After 8 h at -78 °C, the reaction
mixture was quenched by the addition of saturated aqueous NaHCO3
(0.5 mL), and the aqueous layer was extracted with EtOAc (3 × 2
mL). The combined organic extract was dried (Na2SO4) and concen-
trated in Vacuo. PCTLC (SiO2, 1 mm plate, 25% EtOAc-hexane)
provided (+)-62b (5.8 mg, 10%) and (+)-63b (37.7 mg, 65%).
Reequilibration. A solution of (+)-62b (8 mg, 0.01 mmol) in THF
(0.2 mL) at -78 °C was treated with LDA (0.10 mL of a 0.2 M solution
in THF, 1.5 equiv). After 8 h at -78 °C, the reaction mixture was
quenched by the addition of saturated aqueous NaHCO3 (0.5 mL), and
the aqueous layer was extracted with EtOAc (3 × 2 mL). The
combined organic extract was dried (Na2SO4) and concentrated in
ent-(2S,8R)-38: [R]25 -76 (c 0.05, THF).
D
Methyl (2R,8S)-4-(Benzyloxy)-8-[(tert-butyldimethylsilyl)oxy]-3,6-
bis(tert-butyloxycarbonyl)-2-methyl-1-oxo-2,3,6,7,8,9-hexahydro-1H-
pyrrolo[3,2-f]quinoline-2-carboxylate ((+)-39).50 A solution of
(+)-38 (10 mg, 0.014 mmol) in THF-H2O (8:1, 0.5 mL) at 0 °C was
treated with TsOH‚H2O (8.0 mg, 0.042 mmol, 3 equiv). After 2 h at
0 °C, the reaction mixture was quenched by the addition of saturated
aqueous NaHCO3 (0.5 mL), and the aqueous layer was extracted with
EtOAc. The combined organic extract was dried (Na2SO4) and
concentrated in Vacuo. PCTLC (SiO2, 0.3 mm plate, 15% EtOAc-
hexane) provided (+)-39 (8.3 mg, 83%) as a white foam: 1H NMR
(CDCl3, 400 MHz) δ 7.68 (br s, 1H), 7.51 (d, 2H, J ) 6.9 Hz), 7.36-
7.25 (m, 3H), 5.14 (d, 1H, J ) 12.4 Hz), 5.10 (d, 1H, J ) 12.4 Hz),
4.13-4.07 (m, 1H), 3.91 (dd, 1H, J ) 12.7, 2.8 Hz), 3.69 (s, 3H),
3.36 (dd, 1H, J ) 12.7, 6.8 Hz), 3.34 (dd, 1H, J ) 18.4, 5.4 Hz), 2.97
(dd, 1H, J ) 18.4, 6.6 Hz), 1.71 (s, 3H), 1.52 (s, 9H), 1.40 (s, 9H),
0.86 (s, 9H), 0.10 (s, 3H), 0.09 (s, 3H); IR (film) νmax 2928, 2860,
1762, 1746, 1707, 1503, 1363, 1333, 1253, 1146 cm-1; FABHRMS
(NBA-CsI) m/z 829.2464 (M + Cs+, C37H52N2O9Si requires 829.2496).
(2R,8S)-39: [R]25 +103 (c 0.10, THF).
D
ent-(2S,8R)-39: [R]25 -107 (c 0.03, THF).
D
(50) Characterization for the diastereomeric series 45-50 and 68-70 is
provided in the Supporting Information.

324 J. Am. Chem. Soc., Vol. 119, No. 2, 1997
Boger et al.
Methyl (2R,8S)-4-(Benzyloxy)-8-hydroxy-2-methyl-1-oxo-6-[(5,6,7-
trimethoxyindol-2-yl)carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo-
[3,2-f]quinoline-2-carboxylate ((+)-41).50 From 39. (+)-39 (4.0 mg,
5.7 µmol) was treated with 4 M HCl-CH3OH (0.5 mL) at 0 °C for 1
h. The volatiles were removed under a stream of N2, and the residue
was dissolved in DMF (0.5 mL). EDCI (3.3 mg, 3 equiv) and 4010
(2.2 mg, 1.5 equiv) were added, and the solution was stirred for 4 h at
25 °C. PTLC (SiO2, 10 × 20 cm × 0.25 mm, 80% EtOAc-hexane)
provided (+)-41 (2.6 mg, 73%): 1H NMR (CDCl3, 400 MHz) δ 9.04
(s, 1H), 7.29-7.25 (m, 5H), 7.10 (s, 1H), 6.72 (s, 1H), 6.50 (s, 1H),
5.22 (s, 1H), 4.89 (s, 2H), 4.38-4.30 (m, 2H), 4.05 (s, 3H), 3.91 (s,
3H), 3.88-3.85 (m, 1H), 3.86 (s, 3H), 3.76 (s, 3H), 3.36 (dd, 1H, J )
19.2, 5.4 Hz), 3.19 (dd, 1H, J ) 19.2, 2.9 Hz), 1.57 (s, 3H); IR (film)
1512, 1459, 1310, 1236, 1171, 1104 cm-1; FABHRMS (NBA-CsI) m/z
(M + H+, C34H35N3O11S requires 694.2071).
(2R,8S)-43: [R]25 +11 (c 0.01, THF).
D
ent-(2S,8R)-43: [R]25 -11 (c 0.01, THF).
D
Methyl (2R,8S)-4-Hydroxy-8-[(methanesulfonyl)oxy]-2-methyl-
1-oxo-6-[5,6,7-trimethoxyindol-2-yl)carbonyl]-2,3,6,7,8,9-hexahydro-
1H-pyrrolo[3,2-f]quinoline-2-carboxylate ((-)-44).50 A solution of
(+)-43 (3.3 mg, 5 µmol) in CH3OH (0.5 mL) under H2 was treated
with 10% Pd-C (1 mg), and the suspension was stirred for 3 h at 25
°C. The reaction mixture was filtered through Celite and concentrated
in Vacuo. Chromatography (SiO2, 30% EtOAc-hexane) provided
(-)-44 (2.9 mg, 100%) as a yellow solid: 1H NMR (CDCl3, 400 MHz)
δ 9.09 (s, 1H), 6.94 (s, 1H), 6.68 (s, 1H), 6.47 (s, 1H), 6.36 (br s, 1H),
5.29-5.23 (m, 2H), 4.71 (d, 1H, J ) 11.7 Hz), 4.02 (s, 3H), 3.89 (s,
3H), 3.84-3.79 (m, 1H), 3.81 (s, 3H), 3.77 (s, 3H), 3.49-3.46 (m,
2H), 2.90 (s, 3H), 1.67 (s, 3H); IR (film) νmax 3337, 2926, 2849, 1739,
1685, 1617, 1516, 1310, 1226, 1172, 1104 cm-1; FABHRMS (NBA-
CsI) m/z 603.1535 (M+, C27H29N3O11S requires 603.1520).
ν
max 3364, 2923, 1738, 1687, 1610, 1513, 1426, 1303, 1241, 1108 cm-1
;
FABHRMS (NBA-CsI) m/z 748.1275 (M + Cs+, C33H33N3O9 requires
748.1271).
(2R,8S)-41: [R]25 +17 (c 0.10, THF).
D
ent-(2S,8R)-41: [R]25 -17 (c 0.05, THF).
D
(2R,8S)-44: [R]26 -7 (c 0.05, THF).
D
From 54. A sample of 5435 (40.4 mg, 0.079 mmol) was treated
with saturated HCl-CH3OH (10 mL) at 25 °C for 5 h. The solvent
was removed in Vacuo, and the residue was diluted with EtOAc (30
mL), and the mixture was washed with saturated aqueous NaHCO3
(20 mL), saturated aqueous NaCl (20 mL) and dried (MgSO4), and
concentrated in Vacuo. Chromatography (SiO2, 20-30% EtOAc-
hexane gradient elution) afforded the free amine (27.4 mg, 30.1 mg
theoretical, 91%) as a pale yellow solid. A solution of the free amine
(27.4 mg, 0.072 mmol) and 5,6,7-trimethoxyindole-2-carboxylic acid
(40, 19.8 mg, 0.079 mmol) in anhydrous THF (3 mL) was treated with
diethyl cyanophosphonate (16.3 µL, 0.11 mmol, 1.5 equiv) and Et3N
(15.0 µL, 0.11 mmol, 1.5 equiv) at 0 °C under N2. The reaction mixture
was warmed at 70 °C for 16 h. The solvent was removed in Vacuo.
Chromatography (SiO2, 50-80% EtOAc-hexane gradient elution)
afforded 41 (27.8 mg, 44.1 mg theoretical, 63%) as a pale yellow solid
identical in all respects to that described above.
ent-(2S,8R)-44: [R]26 +6 (c 0.01, THF).
D
(+)-Duocarmycin A ((+)-1). From 44. A solution of (-)-44 (8.5
mg, 14 µmol) in CH3CN (1.0 mL) at 0 °C under Ar was treated with
DBU (2 equiv, 4.2 µL, 28 µmol), and the solution was warmed to 25
°C and stirred for 1 h. The reaction mixture was diluted with EtOAc
(10 mL), washed with cold 10% aqueous citric acid and saturated
aqueous NaCl, dried (MgSO4), and concentrated in Vacuo. PTLC (SiO2,
90% EtOAc-hexane, Rf ) 0.65) provided 1 (5.3 mg, 7.1 mg theoretical,
75%; typically 75-83%): 1H NMR (C6D6, 400 MHz) δ 9.22 (s, 1H),
7.65 (s, 1H), 6.68 (s, 1H), 6.43 (d, 1H, J ) 2.3 Hz), 5.58 (s, 1H), 3.76
(s, 3H), 3.68 (s, 3H), 3.51 (s, 3H), 3.45 (d, 1H, J ) 10.1 Hz), 3.22 (dd,
1H, J ) 10.1, 5.2 Hz), 3.19 (s, 3H), 2.40-2.36 (m, 1H), 1.83 (dd, 1H,
J ) 7.6, 3.7 Hz), 1.41 (s, 3H), 0.42 (t, 1H, J ) 4.6 Hz, partially obscured
by H2O); IR (film) νmax 3321, 2935, 1743, 1683, 1622, 1386, 1303,
1274, 1235, 1199, 1110 cm-1; FABHRMS (NBA-CsI) m/z 508.1740
(M + H+, C26H25N3O8 requires 508.1720).
This coupling reaction was also accomplished as follows. A solution
of the free amine (3.9 mg, 0.01 mmol) and 40 (3.1 mg, 0.012 mmol,
1.2 equiv) in anhydrous DMF (0.5 mL) was treated with EDCI (5.9
mg, 0.03 mmol, 3 equiv) at 25 °C for 48 h under N2. The solvent was
removed in Vacuo, and flash chromatography (SiO2, 50-80% EtOAc-
hexane gradient elution) afforded 41 (3.3 mg, 6.3 mg theoretical, 52%).
Natural (+)-1: [R]26 +291 (c 0.01, CH3OH).
D
ent-(-)-1: [R]26 -286 (c 0.025, CH3OH).
D
From 42. A solution of 42 (9.7 mg, 0.018 mmol) in C6H6 (1 mL)
was treated with Bu3P (5.5 mg, 7 µL, 0.027 mmol, 1.5 equiv), and
ADDP (7.0 mg, 0.027 mmol, 1.5 equiv) and the solution was warmed
at 50 °C for 1 h under N2. The volatiles were removed, and the residue
was purified by chromatography (SiO2, 1 × 4 cm, 80% EtOAc-hexane)
to provide (+)-1 (9.3 mg, 9.4 mg theoretical, 99%).
Methyl (2R,8S)-4,8-Dihydroxy-2-methyl-1-oxo-6-[(5,6,7-trimeth-
oxyindol-2-yl)carbonyl]-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-f]-
quinoline-2-carboxylate ((+)-42, Duocarmycin D1).50 A solution of
(+)-41 (11.8 mg, 19 µmol) in CH3OH (1.0 mL) under H2 was treated
with 10% Pd-C (1.7 mg), and the suspension was stirred for 3 h at 25
°C. The reaction mixture was filtered through Celite and concentrated
in Vacuo. Chromatography (SiO2, 30% EtOAc-hexane) provided (+)-
42 (9.7 mg, 10.1 mg theoretical, 96%; typically 92-98%) as a yellow
solid: 1H NMR (CDCl3, 400 MHz) δ 9.28 (s, 1H), 7.12 (s, 1H), 6.98
(s, 1H), 6.71 (s, 1H), 6.59 (s, 1H), 5.23 (s, 1H), 4.40-4.34 (m, 2H),
4.02 (s, 3H), 3.89 (s, 3H), 3.82 (s, 3H), 3.81-3.77 (m, 1H), 3.75 (s,
3H), 3.33 (dd, 1H, J ) 19.0, 5.5 Hz), 3.19 (dd, 1H, J ) 19.0, 2.8 Hz),
1.63 (s, 3H); IR (film) νmax 3350, 2928, 1733, 1682, 1604, 1517, 1306,
1249, 1105 cm-1; FABHRMS (NBA-CsI) m/z 526.1846 (M + H+,
C26H27N3O9 requires 526.1826).
(+)-epi-Duocarmycin A ((+)-7). A solution of (-)-5027 (1.40 mg,
2.32 µmol) in 0.5 mL of CH3CN at 0 °C under Ar was treated with
DBU (2 equiv, 23 µL of 0.20 M in CH3CN) and the solution was
warmed to 25 °C and stirred for 2 h. The reaction mixture was
concentrated, and the crude product was purified by PTLC (5 × 20
cm × 0.25 mm, 90% EtOAc-hexane, Rf ) 0.65) to provide 7 (0.84
mg, 71%): 1H NMR (C6D6, 400 MHz) δ 9.16 (s, 1H), 6.95 (s, 1H),
6.66 (s, 1H), 6.41 (d, 1H, J ) 2.1 Hz), 5.62 (s, 1H), 3.77 (s, 3H), 3.68
(s, 3H), 3.52 (s, 3H), 3.46 (d, 1H, J ) 10.2 Hz), 3.34 (d, 1H, J ) 10.2,
5.3 Hz), 3.20 (s, 3H), 2.41-2.39 (m, 1H), 1.86 (dd, 1H, J ) 7.4, 3.8
Hz), 1.46 (s, 3H), 0.44 (t, 1H, J ) 5.4 Hz, partially obscured by H2O);
IR (film) νmax 3323, 2933, 1738, 1683, 1385, 1302, 1236, 1107 cm-1
;
FABHRMS (NBA-CsI) m/z 508.1744 (M + H+, C26H25N3O8 requires
(2R,8S)-42: [R]25 +8 (c 0.04, THF).
D
508.1720).
ent-(2S,8R)-42: [R]25 -9 (c 0.04, THF).
D
(+)-7: [R]26 +155 (c 0.015, CH3OH).
D
ent-(-)-7: [R]26 -156 (c 0.025, CH3OH).
Methyl (2R,8S)-4-(Benzyloxy)-8-[(methanesulfonyl)oxy]-2-meth-
yl-1-oxo-6-[(5,6,7-trimethoxyindol-2-yl)carbonyl]-2,3,6,7,8,9-hexahy-
dro-1H-pyrrolo[3,2-f]quinoline-2-carboxylate ((+)-43).50 A solution
of (+)-41 (3.2 mg, 5.0 µmol) in pyridine (0.5 mL) was treated with
MsCl (0.6 µL, 1.5 equiv), and the reaction mixture was stirred 1 h at
0 °C. The reaction mixture was quenched by the addition of saturated
aqueous NaHCO3 (0.5 mL), and the aqueous layer was extracted with
EtOAc. The combined organic extract was dried (Na2SO4) and
concentrated in Vacuo. Chromatography (SiO2, 50-80% EtOAc-
hexane gradient elution) provided (+)-43 (3.3 mg, 92%): 1H NMR
(CDCl3, 400 MHz) δ 9.02 (s, 1H), 7.27-7.23 (m, 5H), 6.98 (s, 1H),
6.69 (s, 1H), 6.40 (s, 1H), 5.31-5.26 (m, 2H), 5.28 (s, 1H), 4.91 (d,
1H, J ) 11.4 Hz), 4.87 (d, 1H, J ) 11.4 Hz), 4.67-4.65 (m, 1H), 4.06
(s, 3H), 3.91 (s, 3H), 3.86 (s, 3H), 3.77 (s, 3H), 3.50-3.47 (m, 2H),
2.91 (s, 3H), 1.67 (s, 3H); IR (film) νmax 3336, 292, 1740, 1692, 1618,
D
Methyl (2R,8S)-4-(Benzyloxy)-6-(tert-butyloxycarbonyl)-8-hydroxy-
2-methyl-1-oxo-2,3,6,7,8,9-hexahydro-1H-pyrrolo[3,2-f]quinoline-2-
carboxylate ((+)-64).50 A sample of (+)-39 (4.0 mg, 5.74 µmol) was
treated with 5 M HCl-CH3OH (0.5 mL), and the solution was stirred
for 1 h at 0 °C. The volatiles were removed under a stream of N2, and
the residue was quenched by the addition of saturated aqueous NaHCO3
(0.5 mL). The aqueous solution was extracted with EtOAc (3 × 1
mL), and the combined organic extract was dried (Na2SO4) and
concentrated in Vacuo. The crude product55 and BOC2O (3.8 mg, 17
µmol, 3 equiv) were dissolved in THF (0.5 mL), and the solution was
warmed at reflux for 2 h. The reaction mixture was concentrated, and
the crude product was purified by PTLC (SiO2, 10 × 20 cm × 0.25
mm, 50% EtOAc-hexane) to provide (+)-64 (1.6 mg, 2.8 mg
theoretical, 57%): 1H NMR (CDCl3, 400 MHz) δ 7.45-7.33 (m, 6H),

Synthesis of (+)-Duocarmycin A
J. Am. Chem. Soc., Vol. 119, No. 2, 1997 325
5.28 (s, 1H), 5.09 (s, 3H), 4.27-4.24 (m, 1H), 3.89 (dd, 1H, J ) 6.0,
13.0 Hz), 3.72 (s, 3H), 3.54 (dd, 1H, J ) 2.0, 13.0 Hz), 3.27 (dd, 1H,
J ) 5.7, 18.8 Hz), 3.09 (dd, 1H, J ) 4.2, 18.8 Hz), 1.72 (br s, 1H),
1.60 (s, 3H), 1.50 (s, 9H); IR (film) νmax 3374, 2923, 2851, 1738, 1687,
1512, 1451, 1395, 1251, 1149 cm-1; FABHRMS (NBA-CsI) m/z
615.1134 (M + Cs+, C26H30N2O7 requires 615.1107).
(+)-6-epi-N-BOC-DA ((+)-71). Following the procedure described
above, (-)-7025 (0.85 mg, 1.8 µmol) provided (+)-71 (0.51 mg, 0.68
mg theoretical, 71%): 1H NMR (C6D6, 400 MHz) δ 5.59 (s, 1H), 3.51
(br s, 1H), 3.25 (d, 1H, J ) 11.6 Hz), 3.16 (s, 3H), 2.98 (dd, 1H, J )
4.4, 11.6 Hz), 2.32-2.29 (m, 1H), 1.73 (dd, 1H, J ) 3.7, 7.6 Hz), 1.43
(s, 3H), 1.25 (s, 9H), 0.31 (t, 1H, J ) 4.3 Hz): IR (film) νmax 2921,
2848, 1726, 1677, 1562, 1448, 1386, 1249, 1150 cm-1; FABHRMS
(NBA-CsI) m/z 507.0510 (M + Cs+, C19H22N2O6 requires 507.0532).
(2R,8S)-64: [R]25 +106 (c 0.05, CH3OH).
D
ent-(2S,8R)-64: [R]25 -110 (c 0.05, CH3OH).
D
(+)-71: [R]25 +73 (c 0.02, THF).
D
Methyl (2R,8S)-4-(Benzyloxy)-6-(tert-butyloxycarbonyl)-8-[(meth-
anesulfonyl)oxy]-2-methyl-1-oxo-2,3,6,7,8,9-hexahydro-1H-pyrrolo-
[3,2-f]quinoline-2-carboxylate ((+)-65).50 A solution of (+)-64 (6.0
mg, 12 µmol) in pyridine (0.1 mL) at 0 °C was treated with MsCl (1.9
µL, 25 µmol, 2 equiv). After 2 h at 0 °C, the reaction mixture was
quenched by the addition of saturated aqueous NaHCO3 (0.5 mL), and
the aqueous layer was extracted with EtOAc (3 × 1 mL). The
combined organic extract was dried (Na2SO4) and concentrated in
Vacuo. PTLC (SiO2, 20 × 20 cm × 0.25 mm, 50% EtOAc-hexane)
provided (+)-65 (5.7 mg, 6.7 mg theoretical, 85%; 85-87%) as a light
yellow solid: 1H NMR (CDCl3, 400 MHz) δ 7.45-7.34 (m, 6H), 5.16-
5.13 (m, 1H), 5.14 (s, 1H), 5.11 (d, 1H, J ) 11.2 Hz), 5.07 (d, 1H, J
) 11.2 Hz), 4.31 (dd, 1H, J ) 4.4, 12.9 Hz), 3.72 (s, 3H), 3.56 (d, 1H,
J ) 12.9 Hz), 3.38 (d, 2H, J ) 6.8 Hz), 3.03 (s, 3H), 1.61 (s, 3H),
1.51 (s, 9H); IR (film) νmax 2923, 1741, 1691, 1513, 1352, 1247, 1169
cm-1; FABHRMS (NBA-CsI) m/z 693.0851 (M + Cs+, C27H32N2O9S
requires 693.0883).
ent-71: [R]25 -67 (c 0.01, THF).
D
Aqueous Solvolysis Reactivity of 67. pH 3. N-BOC-DA (67, 100
µg) was dissolved in CH3OH (1.5 mL) and mixed with pH 3 aqueous
buffer (1.5 mL). The buffer contained 4:1:20 (v:v:v) 0.1 M citric acid,
0.2 M Na2HPO4, and H2O, respectively. The solvolysis solution was
sealed and kept at 25 °C protected from light. The UV spectrum was
measured at regular intervals every 1 h during the first day and every
12 h for another week. The decrease in the long-wavelength absorption
at 336 nm and the increase in the short-wavelength absorption were
monitored. The solvolysis rate constant (k ) 1.75 × 10-5 s-1) and
half-life (t1/2 ) 11 h) were calculated from data recorded at the long
wavelength from the least-squares treatment (r ) 0.98) of the slope of
the plot of time versus 1n[(Af - Ai)/(Af - A)].
Preparative Solvolysis of (+)-N-BOC-DA (67). A solution of (+)-
67 (0.35 mg, 0.93 µmol) in THF-H2O (4:1, 0.2 mL) was cooled to 0
°C and treated with CF3SO3H (0.014 mg, 0.093 µmol, 0.1 equiv). The
reaction mixture was warmed to 25 °C and stirred for 24 h. Saturated
aqueous NaHCO3 (0.5 mL) was added, and the aqueous layer was
extracted with EtOAc (4 × 0.5 mL). The organic extract was dried
(Na2SO4) and concentrated in Vacuo. PTLC (SiO2, 5 × 20 cm × 0.25
mm, 75% EtOAc) provided (+)-72 (0.11 mg, 30%) identical in all
respects with authentic material and 76 (0.17 mg, 46%). HPLC analysis
(0.46 × 25 cm ChiralCel OD column, 10% i-PrOH-hexane, 1 mL/
min) of the crude reaction mixture indicated the presence of 76 (tR )
(2R,8S)-65: [R]25 +29 (c 0.07, CH3OH).
D
ent-(2S,8R)-65: [R]25 -28 (c 0.05, CH3OH).
D
Methyl (2R,8S)-6-(tert-Butyloxycarbonyl)-4-hydroxy-8-[(meth-
anesulfonyl)oxy]-2-methyl-1-oxo-2,3,6,7,8,9-hexahydro-1H-pyrrolo-
[3,2-f]quinoline-2-carboxylate ((+)-66).50 A solution of (+)-65 (6.3
mg, 11.2 µmol) in CH3OH (1.0 mL) under H2 was treated with 10%
Pd-C (1 mg), and the suspension was stirred for 2 h at 25 °C. The
reaction mixture was filtered through Celite and concentrated in Vacuo.
PTLC (SiO2, 20 × 20 cm × 0.25 mm, 75% EtOAc-hexane) provided
(+)-66 (5.3 mg, 5.3 mg theoretical, 100%): 1H NMR (CDCl3, 400
MHz) δ 7.52 (s, 1H), 4.63 (s, 1H), 4.51-4.48 (m, 1H), 4.29 (dd, 1H,
J ) 4.0, 13.5 Hz), 3.55 (br s, 1H), 3.29 (d, 1H, J ) 19.0 Hz), 3.18 (s,
3H), 2.97 (dd, 1H, J ) 5.8, 19.0 Hz), 2.87 (d, 1H, J ) 13.5 Hz), 1.99
(s, 3H), 1.54 (s, 9H), 1.45 (s, 3H); IR (film) νmax 3350, 2926, 1739,
1690, 1513, 1359, 1248, 1169 cm-1; FABHRMS (NBA-CsI) m/z
603.0401 (M + Cs+, C20H26N2O9S requires 603.0413).
3.9 min, 60.5%), (+)-72 (tR ) 7.8 min, 39.5%) and no (+)-73 (tR
)
14.6 min, 0%) requiring that the ring expansion solvolysis reaction
proceed by clean SN2 addition of H2O. HPLC analysis of authentic
(+)-72 and (+)-73 under identicaconditions unambiguously established
the tR of the two possible ring expansion solvolysis products.
Acknowledgment. We gratefully acknowledge the financial
support of the National Institutes of Health (CA55276), The
Skaggs Institute for Chemical Biology, the award of an NIH
postdoctoral fellowship (J.A.M., CA62589) and the sabbatical
leaves of T.O. (Tanabe Seiyaku Co., Ltd., Japan) and T.N.
(Sankyo Co. Ltd., Tokyo, Japan). We thank Professor P. A.
Kitos, University of Kansas, for the cytotoxic testing sum-
marized in Table 5.
(2R,8S)-66: [R]25 +52 (c 0.03, CH3OH).
D
ent-(2S,8R)-66: [R]25 -56 (c 0.03, CH3OH).
D
(+)-N-BOC-DA ((+)-67). A solution of (+)-66 (3.6 mg, 7.6 µmol)
in CH3CN (1.0 mL) cooled to 0 °C was treated with DBU (1.4 µL, 9.4
µmol), and the reaction mixture was warmed to 25 °C and stirred for
3 h. The mixture was diluted with EtOAc (10 mL), washed with 10%
aqueous citric acid and saturated aqueous NaCl, and dried (MgSO4).
The solvent was removed in Vacuo. PTLC (SiO2, 20 × 20 cm × 0.25
mm, 80% EtOAc-hexane, or 33% acetone-CHCl3) provided (+)-67
(2.5 mg, 2.9 mg theoretical, 86%) as a pale yellow solid: 1H NMR
(C6D6, 400 MHz) δ 5.52 (s, 1H), 3.54 (br s, 1H), 3.22 (d, 1H, J ) 11.3
Hz), 3.17 (s, 3H), 2.87 (dd, 1H, J ) 4.7, 11.3 Hz), 2.29-2.25 (m,
1H), 1.71 (dd, 1H, J ) 3.6, 7.6 Hz), 1.36 (s, 3H), 1.27 (s, 9H), 0.31 (t,
1H, J ) 4.7 Hz); IR (film) νmax 2923, 2851, 1725, 1677, 1559, 1446,
1385, 1251, 1148 cm-1; FABHRMS (NBA-CsI) m/z 507.0515 (M +
Cs+, C19H22N2O6 requires 507.0532).
Supporting Information Available: Diagnostic character-
ization of 26, 27, and 29, a summary of alternative approaches
explored, full experimental details and characterization for
9-13, 23-25, 30-36, the diastereomeric series 45-50, 51-
55, 56 (routes A and B), 58, 59 and the diastereomeric series
68-70 and 72-73, details of the X-ray structure determination
of the (S)-Mosher ester of 16 (Scheme 7), and data employed
in constructing Figure 4 (40 pages). See any current masthead
page for ordering and Internet access instructions.
Natural 67: [R]25 +183 (c 0.01, THF).
D
ent-67: [R]25 -190 (c 0.02, THF).
JA962431G
D