Studies on the mechanism of action of prekinamycin, a member of the diazoparaquinone family of natural products: Evidence for both sp2 radical and orthoquinonemethide intermediates

Article abstract of DOI:10.1021/ja0642616

The putative reductive activation chemistry of the diazoparaquinone antibiotics was modeled with Bu₃Sn-H and prekinamycin dimethyl ether along with prekinamycin itself. Reaction in various combinations of aromatic solvents, with and without the

Full text of DOI:10.1021/ja0642616

Published on Web 09/02/2006
Studies on the Mechanism of Action of Prekinamycin, a
Member of the Diazoparaquinone Family of Natural Products:
Evidence for Both sp2 Radical and Orthoquinonemethide
Intermediates
Ken S. Feldman* and Kyle J. Eastman
Contribution from the Department of Chemistry, 104 Chemistry Building,
The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802
Received June 16, 2006; E-mail: ksf@chem.psu.edu
Abstract: The putative reductive activation chemistry of the diazoparaquinone antibiotics was modeled
with Bu3Sn-H and prekinamycin dimethyl ether along with prekinamycin itself. Reaction in various
combinations of aromatic solvents, with and without the nucleophile benzylmercaptan present, led to isolation
of both radical-trapping arene adducts and nucleophilic capture benzyl thioether products. On the basis of
these product distribution studies, the intermediacies of, first, a cyclopentenyl radical and, next, an
orthoquinonemethide electrophile are postulated.
Introduction
The diazoparaquinone family of antibiotics, exemplified by
its founding members kinamycin A-D (1a-1d),1 recently has
been enlarged by the discovery of the potent anticancer marine
isolates lomaiviticins A (4) and B, Figure 1.2 In addition to these
species, at least 16 other kinamycins, all varying as per the
alcohol acylation pattern and/or D-ring oxidation level, and the
aromatic D-ring biosynthetic precursor prekinamycin (5) define
this group of natural products.3 The rarity of the diazo function
in naturally occurring compounds is emphasized by the fact that
only five other entries have been reported to date, as shown in
Figure 2.4 Whereas none of these species possesses the
diazoparaquinone function characteristic of the kinamycins and
lomaiviticins, they do display antitumor and/or antibiotic
activity, presumably as a consequence of a reactive N2 unit.4b-f
However, it is the profound cytotoxic activity of the diazo-
(1) (a) Ito, S.; Matsuya, T.; Omura, S.; Otani, M.; Nakagawa, A.; Takeshima,
H.; Iwai, Y.; Ohtani, M.; Hata, T. J. Antibiot. 1970, 23, 315-316. (b) Hata,
T.; Omura, S.; Iwai, Y.; Nakagawa, A.; Otani, M.; Ito, S.; Matsuya, T. J.
Antibiot. 1971, 24, 353-359. (c) Omura, S.; Nakagawa, A.; Yamada, H.;
Hata, T.; Furusaki, A.; Watanabe, T. Chem. Pharm. Bull. 1973, 21, 931-
940.
(2) He, H.; Ding, W.-D.; Bernan, V. S.; Richardson, A. D.; Ireland, C. M.;
Greenstein, M.; Ellestad, G. A.; Carter, G. T. J. Am. Chem. Soc. 2001,
123, 5362-5363.
(3) (a) Gould, S. J.; Chen, J.; Cone, M. C.; Gore, M. P.; Melville, C. R.;
Tamayo, N. J. Org. Chem. 1996, 61, 5720-5721. (b) Gould, S. J. Chem
ReV. 1997, 97, 2499-2509 and references therein. (c) Marco-Contelles,
J.; Molina, M. T. Curr. Org. Chem. 2003, 7, 1433-1442.
(4) (a) Isoprekinamycin: Proteau, P. J.; Li, Y.; Chen, J.; Williamson, R. T.;
Gould, S. J.; Laufer, R. S.; Dmitrienko, G. I. J. Am. Chem. Soc. 2000,
122, 8325-8326. (b) Lagunamycin: Imae, K.; Nihei, Y.; Oka, M.;
Yamasaki, T.; Konishi, M.; Oki, T. J. Antibiot. 1993, 46, 1031-1033. (c)
Azaserine: Fusari, S. A.; Haskell, T. H.; Frohardt, R. P.; Bartz, Q. R. J.
Am. Chem. Soc. 1954, 76, 2881-2883. (d) 6-Diazo-5-oxo-L-norleucine:
Dion, H. W.; Fusari, S. A.; Jakubowski, Z. L.; Zora, J. G.; Bartz, Q. R. J.
Am. Chem. Soc. 1956, 78, 3075-3077. (e) SF2415A2 and congeners:
Gomi, S.; Ohuchi, S.; Sasaki, T.; Itoh, J.; Sezaki, M. J. Antibiot. 1987, 40,
740-749. (f) Fukuda, D. S.; Mynderse, J. S.; Baker, P. J.; Berry, D. M.;
Boeck, L. D.; Yao, R. C.; Mertz, F. P.; Nakatsukasa, W. M.; Mabe, J.;
Ott, J.; Counter, F. T.; Ensminger, P. W.; Allen, N. E.; Alborn, W. E., Jr.;
Hobbes, J. N., Jr. J. Antibiot. 1990, 43, 623-633.
Figure 1. Diazoparaquinone-containing natural products.
paraquinone-containing compounds that has captivated the
imagination of several groups, and preliminary mechanism-of-
action hypotheses have been generated, Scheme 1.
9
12562
J. AM. CHEM. SOC. 2006, 128, 12562-12573
10.1021/ja0642616 CCC: $33.50 © 2006 American Chemical Society

Mechanism of Action of Prekinamycin
A R T I C L E S
alkylation of isoprekinamycin with â-naphthol (Hypothesis 2).7
Central to this thesis is the observation that the deshydroxyl
control 13 did not function similarly, a result which led to the
proposal that internal hydrogen bonding is a key activator of 6
by virtue of rendering the N2 function more diazonium-like and,
hence, more electrophilic when compared with the diazo-like
moiety of 13 (note the N2 IR absorption for 6 vs that for 13).
Presumably, an extension of this proposition to the diazo-
paraquinones would lead to the expectation that kinamycin F
(1f), for example, would react through the diazonium-like
resonance form 1f′ with biologically relevant nucleophiles.
However, one point of concern that may cloud this interpretation
of diazoparaquinone reactivity emerges upon comparison of the
N2 stretching frequencies for the H-bond capable kinamycin F
(1f) (2120 cm-1) and the H-bond incapable derivative kinamycin
J (1j) (2150 cm-1).1c The fact that the H-bond incapable species
has the higher (more diazonium-like) N2 stretch frequency does
not lend support to the Dmitrienko hypothesis, at least insofar
as it applies to the diazoparaquinones.
Figure 2. Diazo-containing natural products lacking the paraquinone
function.
Scheme 1. Previous Mechanistic Thinking on the Role of the
Diazo Function in Reactions with Model Biological Targets
A significant new clue regarding the biological mechanism
of action of the diazoparaquinones might be found in He et
al.’s lomaiviticin isolation/characterization studies, wherein the
authors note that “An ongoing study showed that lomaiviticin
A cleaved double-stranded DNA under reducing conditions”.2
The role that reductive activation might play in Jebaratnam’s
oxidative hypothesis or in Dmitrienko’s electrophilic activation
proposal remains unclear, especially since neither study included
diazoparaquinone-bearing species. Thus, there may be room for
consideration of alternative schemes through which the appar-
ently reductively activated diazoparaquinone-containing natural
products might elicit their striking cytotoxicity.
Much earlier work in the mitomycin field has led to
formulation of the paradigm that a lone pair of electrons situated
adjacent to a paraquinone function is sequestered by vinylogous
resonance (e.g., 14), but upon 1-electron reduction of the
paraquinone, the lone pair is liberated and can participate in
further chemistry, eq 1.8
Extending this line of reasoning to the diazoparaquinones
leads to a similar proposition that entails formation of the
3-electron system 17a by 1-electron reduction of 16, Scheme
2. By reapportioning the 3-electron array as shown in the
equivalent resonance form 17b, a role for the diazo unit is
revealed. Specifically, â-elimination of N2 gas from the naph-
thalenolate moiety within 17c can provide a new reactive
intermediate, the sp2 radical species 18. The facility of this
transform may in large measure depend on the tradeoff between
the energetic penalty associated with the pyramidalization of
C(11) required for alignment of the π cloud with σ*C-N and
the energetic gain resulting from nitrogen gas expulsion. It is
Jebaratnam and co-workers (Hypothesis 1) offered the first
defined proposal based on the observation that the model
compound diazofluorene (10) nicks plasmid DNA upon expo-
sure to the oxidant Cu(OAc)2.5 They suggested that similar
chemistry with the kinamycins and endogenous oxidants may
lead to a reactive radical intermediate (cf. 11) that could damage
DNA via known oxygen-mediated pathways.6 Later, Dmitrienko
argued for an electrophilic intermediate on the basis of the facile
(6) (a) Stubbe, J.; Kozarich, J. W. Chem. ReV. 1987, 87, 1107-1136. (b)
Murphy, J. A.; Griffiths, J. Nat. Prod. Rep. 1993, 551-564. (c) Dussy, A.;
Meggars, E.; Giese, B. J. Am. Chem. Soc. 1998, 120, 7399-7403.
(7) Laufer, R. S.; Dmitrienko, G. I. J. Am. Chem. Soc. 2002, 124, 1854-1855.
(8) (a) Danishefsky, S. J.; Schkeryantz, J. M. Synlett 1995, 475-490. (b) Rajski,
S. R.; Williams, R. M. Chem. ReV. 1998, 98, 2723-2795.
(5) Arya, D. P.; Jebaratnam, D. J. J. Org. Chem. 1995, 60, 3268-3269.
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12563

A R T I C L E S
Feldman and Eastman
Scheme 2. Proposal for the Generation of a C(11) sp2 Radical
from Diazoparaquinones via 1-Electron Reduction
shell species 22; a hydroxymethylacylfulvene subunit, which
is the key electrophilic pharmacophore implicated in the potent
cytotoxicity of the eponymous family of promising anticancer
agents (cf. 26),11 is generated. By analogy with the chemistry
of 26, base alkylation by the electrophilic orthoquinonemethide
22, as exemplified by the speculative depiction of a conjugate
addition of guanosine N(7) with C(11),8b can provide an entirely
independent and orthogonal avenue to introduce damage into
DNA, 24 f 25. N(7)-Alkylated guanosines are reported to
undergo spontaneous depurination to furnish an abasic site.12
The relative contributions (if any) of cyclopentenyl radical 18
and/or orthoquinonemethide 22 to the biological activity of the
diazoparaquinones remain a matter of speculation, but it is
possible that this rare diazoparaquinone chemical functionality
has emerged through evolutionary pressure to act as an inducible
(by 1-electron reduction) source of these two potentially lethal
reactive intermediates. Interestingly, the isolation of the shunt
metabolite seongomycin (27) from the fermentation broth of a
prekinamycin-producing organism can be interpreted in terms
of the transformation 16 f 22 (with an aromatic prekinamycin
D-ring), and then trapping of 22 by N-acetyl cysteine and
reoxidation of the derived hydroquinone.13
The exploration of these mechanistic hypotheses was under-
taken using prekinamycin (5) and derived species with Bu3-
Sn-H as a model 1-electron reductant. These abiological
conditions certainly limit the breadth of the conclusions that
can be drawn, but they also arguably can serve to reveal the
intrinsic chemistry of the diazoparaquinone function under
reducing conditions regardless of the environment. Extrapolation
from these studies to the biological chemistry of the lomaivi-
ticins or kinamycins can be no more rigorous than argument-
by-analogy, and a stronger connection will have to await the
results of further experimentation in a relevant biological milieu.
In the investigation to follow, evidence is presented that speaks
to the core transforms of the mechanistic proposition offered
in Schemes 2 and 3: formation of an sp2 radical 18 from
diazoparaquinones under 1-electron reductive conditions, and
its subsequent conversion to a competent electrophile 22 by
H-atom abstraction.
not possible at this juncture to predict whether thermodynamics
should trump kinetics for this process or vice versa, but if the
former scenario pertains, then the net result is the transfer of
the input electron into a carbon-bound radical at C(11). One
significant consequence of this elimination reaction becomes
apparent when considering the fate of the C(11) radical; it must,
out of mechanistic necessity, switch its occupancy from a π
orbital to an sp2 orbital. Thus, the diazoparaquinones may have
evolved to incorporate a transduction mechanism that converts
biologically accessible 1-electron (radical?) reductants into
potentially much more reactive sp2 carbon radicals.
The biological chemistry of DNA with sp2 radicals in
general,6b and cyclopentenyl sp2 radicals in particular,9 has been
well documented, as exemplified in eq 2.10 Given the dimeric
structure of the lomaiviticins, the claim of double strand nicking
may be rationalized by invoking two DNA-damaging events
that each extend from the chemistry of a radical of the type 18
generated independently from each half.
Results and Discussion
Prekinamycin (5) was chosen as the starting point for these
mechanistic studies, as it had the dual benefits of (1) embodying
the key structural features of this family (diazoparaquinone,
hydrogen-bond donors adjacent to the carbonyls) while at the
same time (2) being readily accessible through synthesis.14 The
biological activity of prekinamycin apparently has not been
reported, and so it is not possible to assess the validity of using
these aromatic D-ring species as stand-ins for the oxidized
D-ring kinamycins that do display potent cytotoxicity. In
In contrast to the neocarzinostatin chromophore chemistry
21 + DNA-H, diazoparaquinone/DNA chemistry has the
potential to go beyond hydrogen abstraction/radical reactions
(i.e., DNA• + O2) in its capacity to inflict chemical damage on
DNA. For example, rebound radical addition of DNA• with 22
might afford a covalent adduct 23. An even more intriguing
line of conjecture emerges upon consideration of the structural
consequences of hydrogen abstraction by 18 to furnish the closed
(11) (a) Woynarowski, J. M.; Napier, C.; Koester, S. K.; Chen, S.-F.; Troyer,
D.; Chapman, W.; MacDonald, J. R. Biochem. Pharmacol. 1997, 54, 1181-
1193. (b) Herzig, M. C. S.; Arnett, B.; MacDonald, J. R.; Woynarowski,
J. M. Biochem. Pharmacol. 1999, 58, 217-225. (c) McMorris, T. C. Bioorg.
Med. Chem. 1999, 1, 881-886.
(12) (a) Kushida, T.; Uesugi, M.; Sugiura, Y.; Kigoshi, H.; Tanaka, H.;
Hirokawa, J.; Ojika, M.; Yamada, K. J. Am. Chem. Soc. 1994, 116, 479-
486. (b) Hara, M.; Saitoh, Y.; Nakano, H. Biochemistry 1990, 29, 5676-
5681. (c) Asai, A.; Saito, H.; Saitoh, Y. Bioorg. Med. Chem. 1997, 5, 723-
729. (d) Hara, M.; Yoshida, M.; Nakano, H. Biochemistry 1990, 29, 10449-
10455. (e) Sugiyama, H.; Furiwara, T.; Ura, A.; Tashiro, T.; Yamamoto,
K.; Kawanishi, S.; Saito, I. Chem. Res. Toxicol. 1994, 7, 673-683.
(13) Carney, J. R.; Hong, S.-T.; Gould, S. J. Tetrahedron Lett. 1997, 38, 3139-
3142.
(9) (a) Goldberg, I. H. Acc. Chem. Res. 1991, 24, 191-198. (b) Myers, A. G.
Tetrahedron Lett. 1987, 28, 4493-4496.
(10) Sugiyama, H.; Fujiwara, T.; Saito, I. Tetrahedron Lett. 1994, 35, 8825-
8828.
(14) Hauser, F. M.; Zhou, M. J. Org. Chem. 1996, 61, 5722.
9
12564 J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006

Mechanism of Action of Prekinamycin
A R T I C L E S
Scheme 4. Preliminary Experiments to Probe the Formal
Scheme 3. Speculative Pathways by Which Radical 18 Might
Ultimately Lead to DNA Damage
1-Electron Reduction Chemistry of Prekinamycin and Derivatives
orthoquinonemethide unit of 30/30a, Scheme 4. In fact, a single
compound was formed in good yield, but initial examination
of its 1H NMR spectrum generated more puzzlement than
illumination. All of the peaks of the A-B-C-D ring platform
were identifiable and present, but the identity of the attachment
at C(11) was not apparent. In what was to be the first of several
surprises with the reductive chemistry of 5, further spectroscopic
analysis revealed the unequivocal presence of solvent deuterio-
benzene attached to this reactive center, 34. Given the rather
modest reactivity of benzene, this result implied that the
reductive conditions with 5 were generating some type of highly
reactive intermediate, and at this juncture none of the obvious
candidatesscarbene, carbocation (or electrophilic orthoquinone-
methide equivalent), or radicalscould be ruled out. Control
experiments with substrate 29 as indicated in Table 1 did provide
some guidance in that they (a) demand the presence of all radical
generating ingredients and (b) exclude the possibility that either
a purely thermally or a photochemically generated species is
involved.
addition to the natural product itself, both the diacetate 28 and
the dimethyl ether 29 were prepared in an attempt to both test
the necessity of activation via hydrogen bonding to the carbonyl
as per Dmitrienko’s postulate, and to provide probe molecules
whose solubility properties might make experimentation easier,
Scheme 4. Both the dimethyl ether 29 and the diacetate 28 were
prepared by Hauser as part of his prekinamycin synthesis work.14
The initial experiments were designed to explore the simple
question of whether the formal addition of H2 (as Bu3Sn-H,
H+) would lead to an isolable orthoquinonemethide 30. Cog-
nizance of a report by Kim15 that the diazoketone 31 reacted
with the H•-equivalent reagent Bu3Sn-H to furnish the formal
Sn-H addition product 32 stimulated thinking along similar
lines with prekinamycin, 5 f 30a. However, concerns about
the stability/reactivity of a putative addition product 30a could
not be dismissed readily, especially in light of the disclosure
that the related natural products, monofulvenones A (33a) and
B (33b), decomposed within minutes upon addition of dilute
acid.16 Presumably, the protonated versions of 33a/33b can
express orthoquinonemethide reactivity to the detriment of
isolation attempts. The question, then, is, will the aromatic
D-ring of the desired kinamycin adduct 30a mitigate its
reactivity enough, compared with the potentially activating C(4)
carbonyl of 33a/33b, so that it could be isolated?
The crude product mixtures were purified via SiO2 chroma-
tography to furnish pure 34a from 5, 35 from 28, and 36 from
29. Despite much effort with modified conditions of reaction,
workup, or purification, no tin-bearing species (cf. 34a with
R1 ) Bu3Sn; compare 32) could be isolated. Attempts to prepare
Table 1. Control Experiments with the Dimethyl Ether 29
entry
light
heat
Bu3Sn
−H
AIBN
result
The answer, unfortunately, is no. Treatment of 5 with Bu3-
SnH/AIBN in C6D6 at 80 °C did not lead to detection of any
product, even as a minor component, that possessed the
1
2
3
4
5
X
X
X
X
X
X
X
X
X
X
80% 36
79% 36
recovered 29
recovered 29
recovered 29
X
X
X
X
X
(15) Kim, S.; Cho, J. R. Bull. Korean Chem. Soc. 1993, 14, 664-665.
(16) Volkmann, C.; Ro¨ssner, E.; Metzler, M.; Za¨hner, H.; Zeeck, A. Liebigs
Ann. 1995, 1169-1172.
X
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12565

A R T I C L E S
Feldman and Eastman
Table 2. Relative Reactivity of Electron-Rich and
Electron-Deficient Arene Solvents with 29 under
Radical-Generating Conditions
Figure 3. Hammett study of the reaction between 29 and arene solvents
under radical-generating conditions.
ArH
yield 36
39(%)a
+
40
The undeniable dogleg exposed by graphing the data from
entries a-d of Table 2 against σ (see Figure 3) provides little
support for a hypothesis that depends on a C(11) cationic
intermediate for C-C bond formation to the arene. That any
non-hydrogen substituent on the arene accelerates adduct
formation is reinforced by the dual substituent entries e-g of
Table 2, where a rough additivity in rates is seen for 1,3-
dimethyl, 1-3-dimethoxy, and 1,3-dicyanobenzene compared
to that for the corresponding monsubstituted species. The facts
that (1) 1,3-dicyanobenzene is the most reactive arene solvent
examined and (2) this electron-deficient arene is closely followed
in relative reactivity by the electron-rich 1,3-dimethoxybenzene
argue perhaps the most persuasively against adduct formation
via an electrophilic orthoquinonemethide intermediate. However,
these observations, in and of themselves, do not permit the
rigorous exclusion of this species from further consideration if
it were perhaps just a minor player with the more electron-rich
arenes (vide infra).
R
R1
rel. ratea,b
o:m:p
rel rateb,d
o:m:p
a
b
c
d
e
f
CH3
Cl
CN
H
H
H
H
43 ( 3 2.2 ( 0.1 62:23:15
64 ( 6 1.5 ( 0.2 48:32:20
67 ( 5 2.1 ( 0.3 43:25:32
68 ( 11 3.1 ( 0.7 72:16:12
52 ( 11 4.0 ( 0.9 50:50:0c
2.1
0.9
2.0
2.2
69:23:17
49:33:19
56:18:26
63:14:23
OCH3
CH3
CH3
OCH3 OCH3 75 ( 10 4.2 ( 1.5 31:69:0c
g
CN
CN
69 ( 7 6.1 ( 0.5e 24:76:0c
a Quintuple measurements with standard deviation. b Rate relative to
benzene. c Ratio of 2,4-:2,6-:3,5- adducts. d Data from ref 20. e Extrapolated
from the results of 1:9 and 1:13 ratios of 1,3-dicyanobenzene to benzene
solvent due to solubility constraints.
a more stable version of 34a (R1 ) SnPh3 or Si(TMS)3) with
Ph3SnH or TMS3SiH did not provide any benefit in this regard.
Both the diacetate 28 and the dimethyl ether 29 performed
similarly under these reductive tin conditions, leading to the
respective phenyl adducts in good yields. The higher yield of
adduct 36 with the dimethyl ether 29 can be attributed to ease
of chromatographic purification of the product compared to 34a
or 35, both of which streaked on SiO2. All of the subsequent
chemistry was conducted with 29 for this reason. The ability to
hydrogen-bond to the carbonyl (5 vs 29) apparently has no
significant bearing of the facility of this chemistry.
Therefore, by process of elimination, the mechanistic picture
for conversion of 29 into 36/39 or, by extension, that for the
similar conversion of 5 into 34a can be refined substantially to
include a prominent role for direct sp2 radical addition to the
arene solvent,18 Scheme 5. Thus, the data presented in Table 2
are consistent with a sequence whereby sp2 radical 42, generated
by formal 1-electron reduction of 29 with loss of N2, adds
directly to the arene ring of the solvent to furnish a transient
cyclohexadienyl radical-containing adduct 43. Despite the fact
that the transformation is run under globally reducing conditions,
the apparent oxidation of this radical by AIBN itself to
reformulate the aryl ring of product 44 is not unexpected, as
Identifying the reactive intermediate, generated upon reduc-
tion of 29, which participates in C-C bond formation with
benzene, became the focus of the next series of experiments. A
priori, a C(11) carbene seemed the least likely candidate given
the type of products formed and the results of the control
experiments; thus, distinguishing between the remaining two
possibilities, a C(11) radical or a C(11) electrophile (ortho-
quinonemethide ≈ secondary carbocation), was of paramount
concern. The use of aryl substituent effects in the context of a
Hammet-type study provided one avenue to probe this question.
Toward this end, the prekinamycin dimethyl ether substrate 29
was exposed to the usual mixture of Bu3SnH and AIBN in a
variety of equimolar mixtures of benzene and other substituted
aromatic solvents, Table 2. Absolute chemical yields of the
benzene adduct 36 and the substituted arene analogues 39 were
recorded, as were ortho:meta:para ratios (or other substituent
ratios, as appropriate). Standard Hammet (para) substituent
constants σ were used for this analysis, which necessitated
normalizing the raw data.17
(17) σ values from: Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96-103. The
raw [39]/[36] values were (1) divided by 1/6 to compensate for the six
equivalent positions of benzene, and (2) multiplied by the % para product
from Table 2.
(18) (a) Dannley, R. L.; Gregg, E. C., Jr.; Phelps, R. E.; Coleman, C. B. J. Am.
Chem. Soc. 1954, 76, 445-448. (b) Dannley, R. L.; Gregg, E. C., Jr. J.
Am. Chem. Soc. 1954, 76, 2997-3000. (c) Kryger, R. G.; Lorand, J. P.;
Stevens, N. R.; Herron, N. R. J. Am. Chem. Soc. 1977, 99, 7589-7600.
(d) Madhavan, V.; Schuler, R. H.; Fessenden, R. W. J. Am. Chem. Soc.
1978, 100, 888-893. (e) Traynham, J. G. Chem. ReV. 1979, 79, 323-330.
(f) Scaiano, J. C.; Stewart, L. C. J. Am. Chem. Soc. 1983, 105, 3609-
3614. (g) Crich, D.; Hwang, J.-T. J. Org. Chem. 1998, 63, 2765-2770.
(h) Mart´ınez-Barrasa, V.; Garc´ıa de Viedma, A.; Burgos, C.; Alvarez-Builla,
J. Org. Lett. 2000, 2, 3933-3935. (i) Nu´n˜ez, A.; Sa´nchez, A.; Burgos, C.;
Alvarez-Builla, J. Tetrahedron 2004, 60, 6217-6224. (j) McLoughlin, P.
T. F.; Clyne, M. A.; Aldabbagh, F. Tetrahedron 2004, 60, 8065-8071.
(k) Liu, J.-a.; Petzold, C. J.; Ramirez-Arizmendi, L. E.; Perez, J.; Kentta¨maa,
H. J. Am. Chem. Soc. 2005, 127, 12758-12759.
9
12566 J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006

Mechanism of Action of Prekinamycin
A R T I C L E S
Scheme 6. Plausible Routes to the C(11)-C(11′) Dimer 49
Scheme 5. Mechanistic Proposal for the Formation of Arene
Adducts 36/39 from Diazoparaquinone 29
some insight into this question about the role of 45 under such
circumstances (vide infra). Further support for the radical-as-
key-intermediate hypothesis in these arene additions can be
garnered from two observations: (1) similar relative rates and
similar ortho:meta:para ratios for thiazole-derived radical 40
addition to substituted arenes have been recorded by Dou (see
Table 2),20 and (2) small and variable amounts of the C(11)
dimer 49 were isolated from many of the runs, Scheme 6. The
formation of 49 is difficult to explain without invoking the
intermediacy of radical 42, either by direct dimerization or
through participation of the derived but undetected orthoqui-
nonemethide, 42 + 45 f 48 or 45 + 45 f 50 f 49.
A few more attempts to trap radical 42 with other radical-
reactive reagents were explored, eq 3. Standard reaction of 29/
Bu3SnH/AIBN in the presence of Ph2Se2 led to formation of
the C(11) selenium adduct 51 along with the benzene addition
product 36. Attempts to perform the tin hydride-mediated
reduction of 29 in non-aromatic solvents were uniformly
unrewarding, as reaction in either THF, 1,4-dioxane, CCl4, CH3-
CN, EtOH, or Cl3CH returned only unreacted starting material.
These observations, taken together, may point to a defined role
for the aromatic solvent in either promoting the formation of,
or possibly stabilizing, the sp2 radical by π-complexation.
per the work of Beckwith et al.19 Workup via SiO2 chroma-
tography then provides the detected product(s) 36/39.
If the orthoquinonemethide 45 is formed from 42 by hydrogen
abstraction from Bu3SnH, it is unclear at this point whether this
electrophile plays any productive (even if only minor) role in
adduct formation, at least with the electron-rich arene solvents.
The overall chemical yields of the arene adducts ranged from
52 to 80%, leaving as an open question the fate of the remainder
of the starting material. The diversion of 42 to the orthoqui-
nonemethide 45 may account for some of the consumed 29, as
this species apparently does not participate in direct adduct
formation with the electron-deficient arenes (vide infra). With
electron-rich arenes, however, some minor involvement of
Friedel-Crafts-type alkylation with 45 may occur to give the
reduced hydroquinone adduct 47, which can suffer air-induced
oxidation to intercept 44 en route to the observed product 39.
If this electrophile mechanistic channel were expressed, the
oxidation/hydrolysis sequence that forms 39 from 47 might just
as reasonably be reversed (i.e., hydrolysis and then oxidation).
The conclusion regarding the prominence of radical addition
chemistry over orthoquinonemethide chemistry only pertains to
reactions with an aromatic solvent nucleophile. Extrapolating
to a biological milieu, where much more potent nucleophilic
partners for putative orthoquinonemethide 45 are present, would
not seem warranted. Studies to be discussed shortly do provide
So, if some orthoquinonemethide 45 is formed from 42, what
is its fate? This question was probed through two different types
of experiments featuring either (1) variation of the Bu3SnH/
substrate ratio or, independently, (2) examination of the product
distribution when nucleophiles that are better than arene rings
are present. The partitioning of radical 42 into either the arene
(19) Beckwith, A. L. J.; Bowry, V. W.; Bowman, W. R.; Mann, E.; Parr, J.;
Storey, J. M. D. Angew. Chem., Int. Ed. 2004, 43, 95-98 and references
therein.
(20) Vernin, G.; Jauffred, R.; Ricard, C.; Dou, H. J. M.; Metzger, J. J. Chem.
Soc., Perkin Trans. 2 1972, 1145-1150.
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12567

A R T I C L E S
Feldman and Eastman
Scheme 7. Thiol Adduct Formation in the Reductive Activation of
Diazoparaquinone 29
Figure 4. Yield and ratio of aromatic trapping products 39c/36 (left) and
39f/36 (right) upon increasing [Bu3SnH]. All data points result from
averaging quadruplicate or pentuplicate measurements.
adduct 43 or the putative orthoquinonemethide 45 should be
responsive to the tin hydride concentration if the mechanistic
model detailed in Scheme 5 is applicable. In the case of an
electron-deficient solvent, any orthoquinonemethide 45 so
formed would not lead to adduct formation, and thus the overall
yield of adducts 36/39, which can only derive from 42, should
decrease as [Bu3SnH] increases. This hypothesis was tested by
exposing a 1:1 molar mixture of PhH and PhCN to 29/AIBN/
(varying) Bu3SnH at 80 °C, Figure 4 (left). The data obtained
from these studies support this interpretation, as the overall yield
of arene adducts decreases linearly with increasing [Bu3SnH]
over the range 1-12 equiv of the tin reagent. Verification of
the expectation that adduct formation derives only from radical
42 and not orthoquinonemethide 45 for these non-electron-rich
arenes can be seen by examining the 39c/36 ratio as a function
of [Bu3SnH]; this ratio remains constant within experimental
error, indicating no change in mechanism as tin hydride
concentration increases.
This situation changes noticeably when an electron-rich
aromatic solvent, 1,3-dimethoxybenzene, is used in place of
PhCN, Figure 4 (right). For these experiments, the overall yield
of arene adducts 39f + 36 decreases roughly monotonically at
lower tin hydride concentration, but then this trend reverses at
the highest [Bu3SnH] examined, where the yield of total arene
adducts actually increases. Could the orthoquinonemethide 45
now be playing a role in adduct formation (45 f 46 f 47 f
39f, cf. Scheme 5) with this more nucleophilic solvent? Evidence
in support of this hypothesis can be found by examining the
ratio of arene adducts 39f/36 as a function of tin hydride
concentration; unlike the electron-deficient PhCN trials, the ratio
favoring the electron-rich adduct 39f increases at high
[Bu3SnH]. These observations are entirely consistent with
incursion of the orthoquinonemethide reaction channel to some
extent in (electron-rich) arene adduct formation. Thus, it appears
that both reactive intermediates, the sp2 radical 42 and the
orthoquinonemethide 45, are generated sequentially upon formal
1-electron reduction of diazoparaquinone 29, and both species
can be trapped if the appropriate complementary reaction
partners are present.
to speak to the question of relevance to the chemistry of
diazoparaquinones in a biological setting, where nucleophiles
such as thiols and amines (cf. Scheme 3) might participate in
adduct formation. Inclusion of benzylmercaptan in the reaction
solution provided the means to test these ideas (Scheme 7).
Treatment of a solution of diazoparaquinone 29, either 1.1 or
12 equiv of Bu3SnH, and 10 equiv of PhCH2SH in benzene
solvent at reflux with 1.1 equiv of AIBN in benzene led to
results which encouraged the point of view that the orthoquino-
nemethide 45 may, in fact, be a major player in the chemistry
of diazoparaquinones under 1-electron reducing conditions as
long as suitable nucleophile partners are present. Thus, the
products of direct radical 42 addition to benzene, 36 and 53,
are formed in a combined yield of 37% with 1.1 equiv of Bu3-
SnH, and in a combined 33% yield with 12 equiv of Bu3SnH.
That 52 precedes 53 is suggested by the independent synthesis
of 53 from 36 and PhCH2SH. So, if approximately one-third of
the starting diazoparaquinone can be accounted for by the direct
radical addition pathway under these experimental conditions,
where is the other two-thirds going? It was gratifying to see
that about half of the starting material was converted into the
C(11) benzylmercaptan adduct 54, a species that retains the
C(11) hydrogen as well. The hydroquinone 54 was moderately
The last series of experiments were designed to probe the
question of whether the orthoquinonemethide 45 might be
coaxed into playing a bigger role in product formation if a
reagent much more nucleophilic than a methoxyaromatic solvent
were included in the reaction mixture. In fact, these trials begin
9
12568 J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006

Mechanism of Action of Prekinamycin
A R T I C L E S
Scheme 8. Deuterium-Labeling Experiments with
Diazoparaquinone 29 and Bu3SnD and/or PhCH2SD
Conclusions
In summary, the results described herein provide a compre-
hensive picture of the chemistry that unfolds when a diazo-
paraquinone-containing species, prekinamycin dimethyl ether
(29), is treated with the formal 1-electron reductant Bu3SnH/
AIBN. Evidence in support of a mechanistic sequence involving
first a C(11) sp2 radical and then an electrophilic orthoqui-
nonemethide has been obtained. The radical intermediate is a
competent partner for additions with either electron-deficient
or electron-rich arenes, whereas the subsequent orthoquinone
species will react with both electron-rich arenes and an
alkanethiol. The relevance of this chemistry to the larger
question of kinamycin/lomaiviticin mechanism of action remains
to be established, but at the very least, these results speak to
the intrinsic chemistry available to diazoparaquinone-containing
species under reducing conditions, and they hint at plausible
reaction chemistry within a biological milieu.
sensitive to quinone formation via air oxidation, but careful
handling under a N2 atmosphere reduced this unwanted side
reaction to almost undetectable levels. The most concise
explanation for the formation of this benzylmercaptan adduct
involves reaction through the orthoquinonemethide electrophile
45, where now the much greater nucleophilicity of the thiol,
compared with that of aromatic solvents, provides a ready trap
for this reactive entity. No evidence for the dimerization product
49 was forthcoming from these thiol-containing runs. In a critical
control experiment, refluxing a solution of diazoparaquinone
29 with 100 equiv of benzylmercaptan in benzene led to
absolutely no evidence for chemical reaction, and 29 could be
recovered unchanged from this test. The incorporation of radical-
generating (reducing) reagents appears to be an inescapable
requirement for reaction. The lack of direct nucleophilic addition
between the thiol and the diazo function of 29 does not lend
support to the nucleophilic activation mechanism of Dmitrienko
(Scheme 1), although the claimed H-bond activation within 6
is absent in 29. To probe this point further, the H-bond capable
species prekinamycin (5) itself was subjected to the same
controlsexposure to 100 equiv of PhCH2SH in refluxing
benzenesto the same end; no evidence for chemical reaction
was detected, and 5 could be recovered unchanged.
Experimental Section
General Procedure 1. Phenylation of Prekinamycin and Deriva-
tives. AIBN (1.1 equiv) in benzene (0.06 M) was added via syringe
pump addition over a period of 1 h to a stirring solution of
diazoparaquinones 5, 28, or 29 (1 equiv) and Bu3SnH (1.1 equiv) in
benzene (0.06 M) at 80 °C. When the addition was complete, the
reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was concentrated in
vacuo. The resulting residue was purified by flash chromatography on
SiO2 using the specified eluent.
General Procedure 2. Aromatic Solvent Competition Experi-
ments. AIBN (1.1 equiv) in an equimolar solution of benzene and the
indicated aromatic solvent (0.06 M) was added via syringe pump
addition over a period of 1 h to a stirring solution of diazoparaquinone
29 (1 equiv) and Bu3SnH (1.1 equiv) also in an equimolar solution of
benzene and the indicated aromatic solvent (0.06 M) at 80 °C. When
the addition was complete, the reaction solution was allowed to cool
to room temperature. After reaching room temperature the reaction
mixture was diluted with CH2Cl2, poured onto a SiO2 column, and
purified by eluting with CH2Cl2 with an increasing percentage of either
EtOAc or acetone from 0% to 10%. Purification furnished the clean
benzene-trapped product 36 and in most cases a mixture of ortho, meta,
and para or the 2,4- and 2,6-regioisomers of the given substituted
aromatic trapped product 39a-g. Relative rates were quantified by
product mass comparison of the substituted aromatic adducts 39 vs
the benzene addition product 36. Isolation of analytical samples of pure
ortho, meta, or para (or the 2,4- or 2,6-regioisomers) from chroma-
tography permitted 1H NMR identification. In cases where isomer
separation was not achieved, the ratios were determined by inspection
The isolation of a reaction product 54 that still retained a
presumably tin hydride-donated hydrogen at C(11) leads to the
obvious labeling experiment with Bu3SnD, Scheme 8. Not so
obvious was the result when diazoparaquinone 29 was combined
with 12 equiv of the tin deuteride and 10 equiv of PhCH2SH in
refluxing benzene; the expected C(11) benzylthiolated hydro-
quinone product 56 was isolated in good yield, but with only
27% deuterium incorporation at C(11) (1H NMR analysis). The
unexpectedly high proton count in 56 was partially explained
by rerunning the labeling experiment with PhCH2SD as the
source of deuterium. In this instance, the hydroquinone 56 was
formed with 46% deuterium incorporation at C(11). As controls,
heating a mixture of either Bu3SnD/PhCH2SH or, independently,
Bu3SnH/PhCH2SD in benzene with AIBN present did not lead
to any deuterium crossover. Finally, it was comforting to note
that when both Bu3SnD and PhCH2SD were used simulta-
neously, the deuterium count in the product 56 (76%) matched
the sum of the deuterium incorporations of the two single-
deuterium source experiments. The origin of the remaining
∼25% of protium in the double-deuterium source experiment
remains a mystery, although PhCH2SH may be a likely
candidate. No further labeling studies were pursued to probe
this point.
1
of H NMR spectra of the mixtures.
General Procedure 3. Aromatic Solvent Competition Experi-
ments with Varying Equivalents of Tin. General Procedure 2 was
used, varying only in the equivalents of Bu3SnH used as indicated in
Figure 4.
4,5,9-Trihydroxy-2-methyl-11-phenyl-benzo[b]fluoren-10-one (34a).
Following General Procedure 1, prekinamycin (5) (18 mg, 0.057 mmol)
was converted into benzo[b]fluorenone 34a (12.3 mg, 59%): mp 260
°C (dec); IR (neat): 3409, 1585 cm-1. Presumably due to the partial
free radical nature of 34a analogous to kinobscurinone,21 34a appears
to be “NMR silent”, exhibiting neither a 1H- nor a 13C NMR signature;
ESI m/z relative intensity 391(MNa+ 100); TOFHRMS (+ESI) Calcd
for C24H16O4Na: 391.0946, Found 391.0947.
Acetic Acid 9-Acetoxy-11-diazo-2-methyl-5,10-dioxo-10,11-dihy-
dro-5H-benzo[b]fluoren-4-yl ester (28). Pyridine (3.6 mL, 28 mmol)
(21) Gould, S. J.; Melville, C. R. Tetrahedron Lett. 1997, 38, 1473-1476.
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12569

A R T I C L E S
Feldman and Eastman
and Ac2O (3.6 mL, 44 mmol) were sequentially added to a mixture of
5 (150 mg, 0.471 mmol) and DMAP (6.0 mg, 0.05 mmol) in CH2Cl2
(15.0 mL) at room temperature. The dark purple mixture was stirred
at room temperature for 2 days, turning to a dark red color. At this
time, the reaction solution was diluted with saturated aqueous NaHCO3
(50 mL) and CH2Cl2 (30 mL). The layers were separated, and the
aqueous layer was extracted with CH2Cl2 (3 × 10 mL). The combined
organic layers were washed with water and brine, dried over Na2SO4,
filtered, and concentrated. The resulting dark red solid was purified by
flash chromatography on silica gel eluting with CH2Cl2/EtOAc (99:1)
to yield the di-acetate (28) as a red solid (156 mg, 85%): mp 205 °C
washed with water and brine, dried over Na2SO4, filtered, and
concentrated. The resulting dark red solid was purified by flash
chromatography on silica gel (CH2Cl2). The triacetate 37 was obtained
as a orange solid (4.4 mg, 99%): mp 195 °C (dec); IR (neat): 1769,
1
1635 cm-1; H NMR (500 MHz, CDCl3) δ 7.59 (d, J ) 7.8 Hz, 1H),
7.58 (d, J ) 7.3 Hz, 1H), 7.47-7.52 (m, 4H), 7.39 (d, J ) 8.3 Hz,
1H), 7.01 (d, J ) 7.7 Hz, 1H), 6.99 (s, 1H), 6.80 (s, 1H), 2.53 (s, 3H),
2.43 (s, 3H), 2.31 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 179.4, 169.6,
168.7, 167.6, 154.6, 151.4, 145.9, 145.2, 142.0, 141.1, 136.5, 133.8,
132.8, 129.4, 129.3, 129.3, 128.9, 128.0, 125.9, 125.3, 124.8, 124.2,
123.7, 121.8, 21.3, 21.2, 21.1, 20.9; ESI m/z relative intensity 517
(MNa+ 100); TOFHRMS (+ESI) Calcd for C30H22O7Na: 517.1263,
Found 517.1245.
1
(dec); IR (neat): 3365, 2924, 2102, 1766 cm-1; H NMR (300 MHz,
CDCl3) δ 8.17 (d, J ) 7.8 Hz, 1H), 7.72 (t, J ) 7.9 Hz, 1H), 7.30 (d,
J ) 7.9 Hz, 1H), 7.20 (s, 1H), 6.89 (s, 1H), 2.57 (s, 3H), 2.47 (s, 3H),
2.46 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 180.0, 177.6, 171.0, 170.0,
150.0, 147.6, 139.7, 137.7, 137.2, 136.7, 136.4, 135.2, 129.0, 126.4,
126.0, 124.3, 122.3, 121.9, 116.9, 22.1, 21.9, 21.6; ESI m/z relative
intensity 425 (MNa+ 72); TOFHRMS (+ESI) Calcd for C22H15N2O6-
Na: 425.0750, Found 425.0738.
4,5,9-Trimethoxy-2-methyl-11-phenyl-benzo[b]fluoren-10-one (38).
From 34a. Methyl iodide (0.042 mL, 0.68 mmol) was added to a
mixture of 34a (25.0 mg, 0.068 mmol) and K2CO3 (93.0 mg, 0.068
mmol) in DMF (2.0 mL) at room temperature. The dark purple mixture
was stirred at room temperature for 12 h, turning to a dark red color.
The reaction solution was then diluted with saturated aqueous NH4Cl
(10 mL) and Et2O (10 mL). The layers were separated, and the aqueous
layer was extracted with Et2O (3 × 10 mL). The combined organic
layers were washed with water and brine, dried over Na2SO4, filtered,
and concentrated. The resulting light red solid was purified by flash
chromatography on silica gel (CH2Cl2). The trimethyl ether 38 was
obtained as a bright red solid (11.5 mg, 41%).
Acetic Acid 9-Acetoxy-5-hydroxy-2-methyl-10-oxo-11-phenyl-
10H-benzo[b]fluoren-4-yl Ester (35). Following General Procedure
1, diazoparaquinone 28 (17.5 mg, 0.044 mmol) was converted into
benzo[b]fluorenone 35 (9.8 mg, 50%): mp 200 °C (dec); IR (neat):
1
3330, 1789, 1766 cm-1; H NMR (400 MHz, CDCl3) δ 9.39 (s, 1H),
7.88 (d, J ) 7.9 Hz, 1H), 7.56 (t, J ) 8.0 Hz, 2H), 7.55 (d, J ) 7.7
Hz, 1H), 7.41-7.50 (m, 3H), 7.06 (d, J ) 8.0 Hz, 1H), 7.06 (s, 1H),
7.05 (s, 1H), 2.51 (s, 3H), 2.36 (s, 3H), 2.32 (s, 3H), 13C NMR (100
MHz, CDCl3) δ 180.2, 170.2, 166.9, 151.3, 149.9, 147.9, 144.4, 143.3,
138.5, 135.7, 133.9, 133.8, 129.7, 128.8, 128.7, 128.2, 126.5, 125.4,
123.3, 123.2, 123.0, 122.6, 114.6, 21.7, 21.6, 21.3; ESI m/z relative
intensity 475 (MNa+ 100); TOFHRMS (+ESI) Calcd for C28H20O6-
Na: 475.1158, Found 475.1142.
From 36. Methyl iodide (0.031 mL, 0.50 mmol) was added to a
mixture of 36 (20.0 mg, 0.050 mmol) and K2CO3 (69.0 mg, 0.050
mmol) in DMF (1.0 mL) at room temperature. The dark red mixture
was stirred at room temperature for 12 h, turning to a light red color.
The reaction was then diluted with saturated aqueous NH4Cl (10 mL)
and Et2O (10 mL). The layers were separated, and the aqueous layer
was extracted with Et2O (3 × 10 mL). The combined organic layers
were washed with water and brine, dried over Na2SO4, filtered, and
concentrated. The resulting light red solid was purified by flash
chromatography on silica gel (CH2Cl2). The trimethyl ether 38 was
obtained as a bright red solid (12.0 mg, 59%): mp 210 °C (dec); IR
(neat): 1643 cm-1; 1H NMR (300 MHz, CDCl3) δ 7.50-7.60 (m, 2H),
7.40-7.50 (m, 5H), 6.96 (dd, J ) 6.7, 2.7 Hz, 1H), 6.76 (s, 1H), 6.59
(s, 1H), 4.02 (s, 3H), 3.96 (s, 3H), 3.89 (s, 3H), 2.32 (s, 3H); 13C NMR
(75 MHz, CDCl3) δ 181.5, 161.6, 155.2, 153.0, 151.7, 145.3, 140.3,
138.7, 134.3, 134.2, 130.1, 129.2, 128.3, 127.8, 126.8, 121.4, 118.3,
118.2, 117.4, 113.9, 113.6, 63.9, 56.1, 55.9, 21.7; ESI m/z relative
intensity 411 (MH+ 100); TOFHRMS (+ESI) Calcd for C27H22O4:
411.1596, Found 411.1596.
5-Hydroxy-4,9-dimethoxy-2-methyl-11-phenyl-benzo[b]fluoren-
10-one (36). Following General Procedure 1, diazoquinone 29 (10.5
mg, 0.032 mmol) was converted into benzo[b]fluorenone 36 (10.0 mg,
1
79%): mp 220 °C (dec); IR (neat): 3181, 1633 cm-1; H NMR (400
MHz, CDCl3) δ 10.76 (s, 1H), 7.58 (d, J ) 7.6 Hz, 1H), 7.56 (d, J )
6.8 Hz, 2H), 7.48 (t, J ) 8.2 Hz, 1H), 7.45 (t, J ) 7.1 Hz, 2H), 7.39
(d, J ) 7.2 Hz, 1H), 7.00 (d, J ) 8.4 Hz, 1H), 6.81 (s, 1H), 6.67 (s,
1H), 4.09 (s, 3H), 3.90 (s, 3H), 2.32 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 182.0, 161.8, 151.9, 151.1, 146.6, 144.3, 138.7, 136.9, 135.0,
134.6, 130.0, 129.4, 128.4, 128.2, 128.1, 120.0, 119.2, 117.8, 115.0,
114.9, 115.2, 56.9, 56.6, 22.2; ESI m/z relative intensity 497 (MH+
50); TOFHRMS (+ESI) Calcd for C26H21O4: 397.1440, Found
397.1443.
5-Hydroxy-4,9-dimethoxy-2-methyl-11-p-tolyl-benzo[b]fluoren-
10-one (39a). Following General Procedure 2, diazoparaquinone 29
(20.0 mg, 0.060 mmol) was converted into a 2:1 mixture of benzo[b]-
fluorenone 39a (ortho:meta:para ) 62:23:15) (7.4 mg, 30%) and benzo-
[b]fluorenone 36 (3.6 mg, 15%). Note that the 2.2:1 ratio of 39a to 36
that is reported in Table 2 reflects the average of five independent runs.
(Ortho:meta:para mixture) IR (neat): 3190, 1633 cm-1; ortho isomer:
1H NMR (500 MHz, CDCl3) δ 10.64 (s, 1H), 7.60 (d, J ) 7.9 Hz,
1H), 7.50 (t, J ) 8.1 Hz, 1H), 7.22-7.36 (m, 3H), 7.19 (d, J ) 7.3
Hz, 1H), 7.01 (d, J ) 8.3 Hz, 1H), 6.69 (s, 1H), 6.54 (s, 1H), 4.11 (s,
3H), 3.89 (s, 3H), 2.31 (s, 3H), 2.16 (s, 3H); (ortho:meta:para mixture)
1H NMR (500 MHz, CDCl3) δ 10.76 (s, 0.4H, meta, para), 10.68 (s,
0.6H, o) 7.60 (d, J ) 7.7 Hz, 0.5H), 7.59 (d, J ) 7.4 Hz, 0.5H), 7.47-
7.51 (m, 1H), 7.33-7.36 (m, 1H), 7.22-7.30 (m, 2H), 7.19 (d, J )
6.8 Hz, 1H), 7.01 (d, J ) 8.3 Hz, 1H), 6.85 (s, 0.25H), 6.79 (s, 0.25H),
6.69 (s, 1H), 6.54 (s, 0.5H), 4.11 (s, 1.8H, o), 4.10 (s, 1.2H, meta,
para), 3.90 (s, 1.2H, meta, para) 3.89 (s, 1.8H, o), 2.35 (s, 1.8H, meta,
para), 2.31 (s, 1.8H, o), 2.16 (s, 3H); (ortho:meta:para mixture) 13C
NMR (125 MHz, CDCl3) δ 181.5, 161.3, 151.5, 150.6, 145.9, 144.2,
138.5, 136.7, 136.5, 135.1, 134.0, 129.9, 129.5, 128.5, 127.6, 126.6,
125.4, 121.3, 119.4, 118.9, 118.6, 117.4, 114.6, 114.4, 111.1, 56.53,
Acetic Acid 4,5-Diacetoxy-2-methyl-10-oxo-11-phenyl-10H-benzo-
[b]fluoren-9-yl Ester (37). From 34a. Pyridine (0.26 mL, 3.1 mmol)
and Ac2O (0.31 mL, 3.1 mmol) were added sequentially to a mixture
of 34a (11.0 mg, 0.031 mmol) and DMAP (1.0 mg, 0.008 mmol) in
CH2Cl2 (1.0 mL) at room temperature. The dark purple mixture was
stirred at room temperature for 30 min, turning to a dark red/orange
color. At this time, the reaction solution was diluted with saturated
aqueous NaHCO3 (10 mL) and CH2Cl2 (10 mL). The layers were
separated, and the aqueous layer was extracted with CH2Cl2 (3 × 10
mL). The combined organic layers were washed with water and brine,
dried over Na2SO4, filtered, and concentrated. The resulting dark red
solid was purified by flash chromatography on silica gel (CH2Cl2) to
yield the triacetate 37 as a bright orange solid (8.1 mg, 52%).
From 35. Pyridine (0.070 mL, 0.90 mmol) and Ac2O (0.85 mL,
0.90 mmol) were sequentially added to a mixture of 35 (4.0 mg, 0.90
mmol) and DMAP (1.0 mg, 0.008 mmol) in CH2Cl2 (1.0 mL) at room
temperature. The dark red mixture was stirred at room temperature for
30 min, changing to a dark red/orange color. At this time, the reaction
solution was diluted with saturated aqueous NaHCO3 (10 mL) and CH2-
Cl2 (10 mL). The layers were separated, and the aqueous layer was
extracted with CH2Cl2 (3 × 10 mL). The combined organic layers were
9
12570 J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006

Mechanism of Action of Prekinamycin
A R T I C L E S
56.5, 56.5, 21.8, 21.7, 19.9; ESI m/z relative intensity 411 (MH+ 30);
TOFHRMS (+ESI) Calcd for C27H22O4: 411.1596, Found 411.1611.
of five independent runs. Ortho isomer: mp 210 °C (dec); IR (neat):
1
3182, 1632 cm-1; H NMR (500 MHz, CDCl3) δ 10.70 (s, 1H), 7.58,
(d, J ) 7.9 Hz, 1H), 7.48 (t, J ) 8.1 Hz, 1H), 7.36 (t, J ) 7.6 Hz, 1H),
7.34 (d, J ) 7.7 Hz, 1H), 7.03 (t, J ) 7.5 Hz, 1H), 7.01 (d, J ) 8.2
Hz, 1H), 7.0 (d, J ) 8.3 Hz, 1H), 6.66 (s, 1H), 6.65 (s, 1H), 4.10 (s,
3H), 3.89 (s, 3H), 3.73 (s, 3H), 2.32 (s, 3H); 13C NMR (125 MHz,
CDCl3) δ 181.3, 161.2, 157.4, 151.4, 150.4, 144.0, 142.8, 138.1, 136.5,
133.8, 130.7, 130.0, 129.2, 124.3, 121.6, 120.4, 119.4, 118.7, 117.3,
114.7, 114.6, 111.3, 110.9, 56.5, 56.1, 55.7, 21.8. Meta:para isomer:
11-(4-Chlorophenyl)-5-hydroxy-4,9-dimethoxy-2-methyl-benzo-
[b]fluoren-10-one (39b). Following General Procedure 2, diazo-
paraquinone 29 (20.0 mg, 0.060 mmol) was converted into a 1.8:1
mixture of benzo[b]fluorenone 39b (ortho:meta:para ) 48:32:20) (9.5
mg, 37%) and benzo[b]fluorenone 36 (5.0 mg, 21%). Note that the
1.5:1 ratio of 39b to 36 that is reported in Table 2 reflects the average
of five independent runs. 39b isomers were separated via preparative
TLC (20% EtOAc in benzene). Ortho isomer: mp 280 °C (dec); IR
1
IR (neat): 3178, 2216, 1622 cm-1; H NMR (500 MHz, CDCl3) δ
1
(neat): 3178, 1630 cm-1; H NMR (500 MHz, CDCl3) δ 10.76 (s,
10.78 (s, 0.43H, p), 10.76 (s, 0.57H, m), 7.59 (d, J ) 7.3 Hz, 1H),
7.58 (d, J ) 8.4 Hz, 1H), 7.49 (t, J ) 8.0 Hz, 1H), 7.37 (t, J ) 8.0 Hz,
1H), 7.14 (d, J ) 7.6 Hz, 0.5H), 6.98-7.06 (m, 2H), 6.93 (d, J ) 8.6
Hz, 0.5H), 6.88 (s, 0.5H), 6.82 (s, 0.5H), 6.69 (s, 1H), 4.10 (s, 3H),
3.91 (s, 3H), 3.88 (s, 1.5H), 3.88 (s, 1.5H), 2.35 (s, 1.5H), 2.34 (s,
1H), 7.60 (d, J ) 7.8 Hz, 1H), 7.50 (m, 2H), 7.33 (m, 3H), 7.02 (d,
J ) 8.2 Hz, 1H), 6.69 (s, 1H), 6.58 (s, 1H), 4.11 (s, 3H), 3.90 (s, 3H),
2.33 (s, 3H). Meta isomer: mp 260 °C (dec); IR (neat): 3378, 1630
1
cm-1; H NMR (500 MHz, CDCl3) δ 10.82 (s, 1H), 7.60 (d, J ) 8.2
1.5H); IR (neat): 3412, 2223, 1631 cm-1 13C NMR (125 MHz, CDCl3)
;
Hz, 1H), 7.51 (m, 2H), 7.44 (d, J ) 6.8 Hz, 1H), 7.38 (t, J ) 7.8 Hz,
1H), 7.37 (s, 1H), 7.30 (d, J ) 8.3 Hz, 1H), 6.75 (s, 1H), 6.71 (s, 1H),
4.12 (s, 3H), 3.91 (s, 3H), 2.36 (s, 3H). Para isomer: mp 240 °C (dec);
IR (neat): 3166, 1631 cm-1; 1H NMR (500 MHz, CDCl3) δ 10.81 (s,
1H), 7.60 (d, J ) 8.4 Hz, 1H), 7.52 (d, J ) 8.6 Hz, 2H), 7.50 (t, J )
8.2 Hz, 1H), 7.42 (d, 8.6 Hz, 2H), 7.30 (d, J ) 8.4 Hz, 1H), 6.79 (s,
1H), 6.71 (s, 1H), 4.12 (s, 3H), 3.92 (s, 3H), 2.35 (s, 3H); (ortho:meta:
para mixture) 13C NMR (125 MHz, CDCl3) δ 181.6, 181.5, 181.3,
161.3, 151.5, 151.4, 151.3, 151.1, 144.5, 143.5, 143.3, 142.3, 138.5,
136.6, 136.5, 136.4, 134.1, 133.8, 133.7, 133.2, 131.1, 130.7, 130.3,
129.5, 129.2, 129.1, 128.9, 128.3, 128.1, 127.9, 126.5, 121.1, 119.6,
119.5, 119.3, 118.5, 118.4, 118.3, 117.5, 114.7, 114.4, 114.2, 111.2,
111.1, 56.6, 56.5, 56.2, 56.1, 21.8, 21.7; ESI m/z relative intensity 431
(MH+ 100); TOFHRMS (+ESI) Calcd for C26H20O4Cl: 431.1050,
Found 431.1060.
δ 181.6, 181.3, 161.4, 161.2, 159.6, 159.2, 157.4, 151.4, 150.8, 150.4,
150.2, 149.4, 146.2, 145.8, 144.0, 142.8, 138.6, 138.2, 136.5, 136.4,
136.2, 136.1, 134.0, 133.8, 133.7, 131.2, 130.7, 130.0, 129.2, 128.9,
128.7, 128.3, 126.6, 124.3, 122.1, 122.0, 121.7, 121.6, 120.4, 119.8,
119.4, 118.9, 118.8, 118.7, 117.4, 117.3, 114.8, 114.7, 114.6, 114.4,
113.7, 113.6, 113.2, 111.3, 111.1, 110.9, 110.5, 56.5, 56.4, 56.1, 56.0,
55.7, 55.3, 55.2, 21.8, 21.7; ESI m/z relative intensity 449 (MNa+ 85);
TOFHRMS (+ESI) Calcd for C27H22O5Na: 449.1365, Found 449.1346.
11-(3,5-Dimethylphenyl)-5-hydroxy-4,9-dimethoxy-2-methyl-ben-
zo[b]fluoren-10-one (39e). Following General Procedure 2, diazo-
paraquinone 29 (20.0 mg, 0.060 mmol) was converted into a 3.1:1
mixture of benzo[b]fluorenone 39e (2:4:5 ) 50:50:0) (11.5 mg, 44%)
and benzo[b]fluorenone 36 (3.4 mg, 14%). Note that the 4.0:1 ratio of
39e to 36 that is reported in Table 2 reflects the average of five
independent runs. 39e isomers were separated via preparative TLC (1%
EtOAc in CH2Cl2). 2,4-Dimethyl isomer: mp 220 °C (dec); IR (neat):
4-(5-Hydroxy-4,9-dimethoxy-2-methyl-10-oxo-10H-benzo[b]fluo-
ren-11-yl)benzonitrile (39c). Following General Procedure 2, diazo-
paraquinone 29 (20.0 mg, 0.060 mmol) was converted into a 2.2:1
mixture of benzo[b]fluorenone 39c (ortho:meta:para ) 43:25:32) (13.0
mg, 51%) and benzo[b]fluorenone 36 (5.6 mg, 23%). Note that the
2.1:1 ratio of 39c to 36 that is reported in Table 2 reflects the average
of five independent runs. 39c isomers were separated via preparative
TLC (20% EtOAc in benzene). Ortho isomer: mp 265 °C (dec); IR
(neat): 3394, 2232, 1631 cm-1; 1H NMR (500 MHz, CDCl3) δ 10.87
(s, 1H), 7.77 (d, J ) 7.6 Hz, 1H), 7.66 (t, J ) 7.7 Hz, 1H), 7.61 (d,
J ) 7.9 Hz, 1H), 7.52 (t, J ) 8.2 Hz, 1H), 7.49 (d, J ) 7.8 Hz, 1H),
7.46 (t, J ) 7.7 Hz, 1H), 7.03 (d, J ) 8.2 Hz, 1H), 6.71 (s, 1H), 6.58
(s, 1H), 4.13 (s, 3H), 3.92 (s, 3H), 2.34 (s, 3H). Meta isomer: mp 265
1
3195, 1633 cm-1; H NMR (500 MHz, CDCl3) δ 10.67 (s, 1H), 7.59
(d, J ) 7.3 Hz, 1H), 7.49 (t, J ) 8.0 Hz, 1H), 7.12 (s, 1H), 7.09 (d,
J ) 7.7 Hz, 1H), 7.05 (d, J ) 7.7 Hz, 1H), 7.00 (d, J ) 8.2 Hz, 1H),
6.68 (s, 1H), 6.58 (s, 1H), 4.11 (s, 3H), 3.89 (s, 3H), 2.37 (s, 3H), 2.30
(s, 3H), 2.13 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 181.5, 161.3,
151.4, 150.4, 146.1, 144.3, 138.4, 137.1, 136.5, 136.5, 133.9, 131.9,
130.8, 129.9, 128.4, 126.1, 121.4, 119.4, 118.6, 117.3, 114.5, 114.4,
111.1, 56.5, 56.1, 21.7, 21.3, 19.8; 2,6-dimethyl isomer: mp 250 °C
1
(dec); IR (neat): 3194, 1633 cm-1; H NMR (500 MHz, CDCl3) δ
10.64 (s, 1H), 7.60 (d, J ) 7.5 Hz, 1H), 7.49 (t, J ) 7.8 Hz, 1H), 7.18
(t, J ) 8.1 Hz, 1H), 7.10 (d, J ) 7.6 Hz, 2H), 7.01 (d, J ) 8.1 Hz,
1H), 6.69 (s, 1H), 6.45 (s, 1H), 4.12 (s, 3H), 3.89 (s, 3H), 2.30 (s, 3H),
2.04 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 181.5, 161.3, 151.5, 150.3,
145.9, 143.3, 138.6, 136.6, 135.9, 134.8, 134.0, 130.8, 130.0, 127.1,
127.0, 121.1, 119.5, 118.0, 117.3, 114.5, 111.2, 56.4, 56.1, 20.7 20.1;
ESI m/z relative intensity 425 (MH+ 50); TOFHRMS (+ESI) Calcd
for C28H25O4: 425.1753, Found 425.1740.
°C (dec); IR (neat): 3412, 2216, 1633 cm-1 1H NMR (500 MHz,
;
CDCl3) δ 10.88 (s, 1H), 7.82 (d, J ) 8.0 Hz, 1H), 7.81 (s, 1H), 7.67
(d, J ) 7.7 Hz, 1H), 7.61 (d, J ) 7.8 Hz, 1H), 7.54 (t, J ) 8.0 Hz, 1H)
7.47 (t, J ) 8.1 Hz, 1H), 7.04 (d, J ) 7.7 Hz, 1H), 6.72 (s, 1H), 6.70
(s, 1H), 4.13 (s, 3H), 3.93 (s, 3H), 2.37 (s, 3H). Para isomer: mp 265
°C (dec); IR (neat): 3412, 2223, 1631 cm-1 1H NMR (500 MHz,
;
CDCl3) δ 10.92 (s, 1H), 7.73 (d, J ) 8.4 Hz, 2H), 7.66 (d, J ) 8.5 Hz,
2H), 7.61 (d, J ) 7.8 Hz, 1H), 7.52 (t, J ) 8.2 Hz, 1H), 7.04 (d, J )
8.3 Hz, 1H), 6.72 (s, 1H), 6.71 (s, 1H), 4.13 (s, 3H), 3.92 (s, 3H), 2.36
(s, 3H); (ortho:meta:para mixture) 13C NMR (125 MHz, CDCl3) δ 181.7,
181.6, 181.4, 161.5, 161.4, 152.4, 152.1, 151.6, 151.5, 143.1, 143.0,
142.9, 139.9, 139.6, 138.7, 138.6, 138.5, 136.4, 136.3, 136.2, 134.4,
133.0, 132.9, 132.3, 131.7, 131.5, 130.8, 130.4, 130.0, 129.7, 128.8,
127.9, 121.0, 119.5, 118.3, 118.1, 118.0, 117.9, 117.8, 115.2, 114.4,
114.3, 114.2, 113.1, 112.1, 111.4, 111.3, 111.3, 56.6, 56.5, 56.3, 56.3,
56.2, 21.8; ESI m/z relative intensity 444 (MNa+ 100); TOFHRMS
(+ESI) Calcd for C27H19NO4Na: 444.1212, Found 444.1215.
5-Hydroxy-4,9-dimethoxy-11-(4-methoxyphenyl)-2-methyl-benzo-
[b]fluoren-10-one (39d). Following General Procedure 2, diazo-
paraquinone 29 (20.0 mg, 0.060 mmol) was converted into a 3.2:1
mixture of benzo[b]fluorenone 39d (ortho:meta:para ) 76:16:12) (14.5
mg, 57%) and benzo[b]fluorenone 36 (4.3 mg, 18%). Note that the
3.1:1 ratio of 39d to 36 that is reported in Table 2 reflects the average
11-(3,5-Dimethoxyphenyl)-5-hydroxy-4,9-dimethoxy-2-methyl-
benzo[b]fluoren-10-one (39f). Following General Procedure 2, diazo-
paraquinone 29 (20.0 mg, 0.060 mmol) was converted into a 4.2:1
mixture of benzo[b]fluorenone 39f (2,4-:2,6-;3,5- ) 31:69:0) (16.2 mg,
59%) and benzo[b]fluorenone 36 (3.4 mg, 14%). Note that the 4.2:1
ratio of 39f to 36 that is reported in Table 2 reflects the average of five
independent runs. 2,4-Dimethoxy isomer: mp 190 °C (dec); IR (neat):
1
3195, 1633 cm-1; H NMR (500 MHz, CDCl3) δ 10.68 (s, 1H), 7.57
(d, J ) 7.6 Hz, 1H), 7.46 (t, J ) 8.2 Hz, 1H), 7.31 (d, J ) 8.9 Hz,
1H), 6.99 (d, J ) 8.3 Hz, 1H), 6.70 (s, 1H), 6.65 (s, 1H), 6.59 (m,
2H), 4.08 (s, 3H), 3.89 (s, 3H), 3.87 (s, 3H), 3.72 (s, 3H), 2.32 (s, 3H);
13C NMR (125 MHz, CDCl3) δ 181.3, 161.2, 160.9, 158.9, 151.4, 150.0,
144.1, 142.8, 138.0, 136.5, 133.6, 131.5, 128.3, 121.8, 119.5, 118.8,
117.2, 116.8, 114.7, 114.5, 110.9, 104.5, 99.0, 56.5, 56.1, 55.6, 55.4,
21.8; 2,6-dimethoxy isomer: mp 180 °C (dec); IR (neat): 3190, 1633
1
cm-1; H NMR (500 MHz, CDCl3) δ 10.80 (s, 1H), 7.58 (d, J ) 8.1
Hz, 1H), 7.47 (t, J ) 8.1 Hz, 1H), 7.30 (t, J ) 8.3 Hz, 1H), 6.99 (d,
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12571

A R T I C L E S
Feldman and Eastman
J ) 8.2 Hz, 1H), 6.65 (d, J ) 8.5 Hz, 2H) 6.64 (s, 1H), 6.57 (s, 1H),
4.08 (s, 3H), 3.89 (s, 3H), 3.68 (s, 6H), 2.30 (s, 3H); 13C NMR (125
MHz, CDCl3) δ 181.1, 161.1, 158.5, 151.5, 150.0. 144.0, 130.4, 138.1,
136.7, 136.3, 133.6, 129.1, 121.6, 119.5, 118.5, 117.2, 114.9, 114.5,
113.7, 113.3, 110.9, 110.6, 104.3, 56.2, 56.1, 56.0, 21.8; ESI m/z relative
intensity 457 (MH+ 30); TOFHRMS (+ESI) Calcd for C28H25O6:
457.1651, Found 457.1653.
was insufficient for separation) and purified by eluting with CH2Cl2
with an increasing percentage of EtOAc from 0% to 10%. Purification
furnished the benzene-trapped adduct 36 (5.0 mg, 21%), and the
phenylselenide-trapped adduct 51 (10 mg, 35%): mp 162 °C (dec); IR
(neat): 3414, 2919, 1631, 1607, 1583 cm-1 1H NMR (500 MHz,
;
CDCl3) δ 10.42 (s, 1H), 7.80 (d, J ) 7.0 Hz, 1H), 7.62 (d, J ) 8.3 Hz,
2H), 7.52 (t, J ) 8.1 Hz, 1H), 7.44 (t, J ) 7.2 Hz, 1H), 7.35 (t, J )
7.1 Hz, 2H), 7.06 (d, J ) 8.2 Hz, 1H), 6.53 (s, 1H), 5.75 (s, 1H), 4.03
(s, 3H), 4.02 (s, 3H), 1.96 (s, 3H); 13C NMR (125 MHz, CDCl3) δ
180.6, 161.3, 151.1, 147.6, 143.8, 142.4, 137.4, 136.4, 136.3, 133.7,
131.7, 129.1, 128.9, 128.7, 121.3, 120.1, 117.2, 113.6, 113.5, 113.4,
110.9, 56.4, 56.1, 21.7; ESI m/z relative intensity 477 (MH+ 100);
TOFHRMS (+ESI) Calcd for C26H21O4Se: 477.0605, Found 477.0615.
5-(5-Hydroxy-4,9-dimethoxy-2-methyl-10-oxo-10H-benzo[b]fluo-
ren-11-yl)isophthalonitrile (39g). Due to the insolubility of 1,3-
dicyanobenzene in benzene at 80 °C, the following experiment was
run at a 13:1 ratio of benzene:1,3-dicyanobenzene. Solid AIBN (11.0
mg, 0.066 mmol) was added portionwise over a period of 1 h to a
stirring solution of diazoparaquinone 29 (20.0 mg, 0.060 mmol), 1,3-
dicyanobenzene (0.11 g, 0.86 mmol), and Bu3SnH (0.018 mL, 0.066
mmol) in benzene (1.0 mL, 11 mmol) at 80 °C. When the addition
was complete, the reaction solution was allowed to cool to room
temperature. After reaching room temperature the reaction mixture was
diluted with CH2Cl2 and purified with flash column chromatography,
eluting with an increasing percentage of EtOAc from 0% to 10% in
CH2Cl2. Purification furnished a 1:2.2 mixture of benzo[b]fluorenone
39g (2,4-:2,6-;3,5- ) 24:76:0) (6.5 mg, 24%) and benzo[b]fluorenone
36 (12.4 mg, 52%). The ratio of 39g:36, extrapolated to equimolar
amounts of benzene and 1,3-dicyanobenzene, is calculated to be 6.0:1.
Note that the 6.1:1 ratio of 39g to 36 that is reported in Table 2 reflects
the average of five independent runs. 39g isomers were separated via
preparative TLC (20% EtOAc in benzene). 2,4-Dicyano isomer: mp
270 °C (dec); IR (neat): 3213, 2526, 1633 cm-1; 1H NMR (500 MHz,
CDCl3) δ 10.97 (s, 1H), 8.04 (s, 1H), 7.91 (d, J ) 8.1 Hz, 1H), 7.63
(d, J ) 8.1 Hz, 1H), 7.61 (d, J ) 7.3 Hz, 1H), 7.53 (t, J ) 8.1 Hz,
1H), 7.06 (d, J ) 8.1 Hz, 1H), 6.72 (s, 1H), 6.54 (s, 1H), 4.13 (s, 3H),
3.93 (s, 3H), 2.35 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 181.4, 161.5,
153.7, 151.7, 144.6, 141.8, 138.9, 137.4, 136.2, 136.2, 135.3, 134.8,
131.4, 131.3, 120.58, 119.6, 118.1, 117.2, 117.1, 116.2, 115.6, 115.0,
114.0, 112.3, 111.5, 56.6, 56.3, 21.8; 2,6-isomer: mp 290 °C (dec);
11-Benzylsulfanyl-4,9-dimethoxy-2-methyl-11-phenyl-benzo[b]-
fluorene-5,10-diol (53), and 11-Benzylsulfanyl-4,9-dimethoxy-2-
methyl-11H-benzo[b]fluorene-5,10-diol (54). A Solution of AIBN
(11.0 mg, 0.067 mmol) in benzene (1 mL) was added over the period
of 1 h to a stirring solution of diazoparaquinone 29 (20 mg, 0.06 mmol),
Bu3SnH (0.018 mL, 0.067 mmol, 1.1 equiv) (or 12 equiv), and benzyl
mercaptan (0.077 mL, 0.6 mmol) at 80 °C. The light red mixture turned
to a dark red color upon addition of AIBN. When the addition was
complete, the reaction solution was allowed to cool to room temperature.
After reaching room temperature the reaction mixture was diluted with
CH2Cl2 and poured onto a silica gel column and purified by eluting
with hexanes/CH2Cl2 (1:1) followed by CH2Cl2 with an increasing
percentage of EtOAc from 0% to 5%. Oxidation of the air-sensitive
hydroquinone 54 was minimized during purification by employing N2
purged solvents and using low-pressure N2 gas for the chromatography.
Purification furnished the benzene-trapped adduct 36 (3.5 mg, 15%),
the benzene/benzyl mercaptan-trapped adduct 53 (6.7 mg, 22%) and
the benzyl mercaptan-trapped adduct 54 (12.5 mg, 46%). 53: mp 260
1
°C (dec); IR (neat): 3378, 3307, 2922, 1644, 1610 cm-1; H NMR
(500 MHz, CDCl3) δ 9.85 (s, 1H), 9.12 (s, 1H) 7.95 (d, J ) 8.5 Hz,
1H), 7.50 (d, J ) 8.6 Hz, 2H), 7.33 (t, J ) 7.8 Hz, 1H), 7.19-7.27
(m, 3H), 6.94-7.04 (m, 5H), 6.91 (s, 1H), 6.79 (d, J ) 7.7 Hz, 1H),
6.65 (s, 1H), 4.16 (s, 3H), 3.96 (s, 3H), 3.25 (d, J ) 12.6 Hz, 1H),
3.14 (d, J ) 12.6 Hz, 1H), 2.30 (s, 3H); 13C NMR (125 MHz, CDCl3)
δ 156.5, 152.6, 151.0, 143.5, 141.4, 139.8, 138.9, 137.6, 128.8, 128.3,
128.1, 127.7, 126.9, 126.8, 126.2, 126.0, 125.0, 123.8, 120.4, 120.2,
116.7, 115.8, 111.3, 104.9, 64.0, 56.5, 55.8, 34.4, 29.7; ESI m/z relative
intensity 543 (MNa+ 40); TOFHRMS (+ESI) Calcd for C33H28O4-
SNa: 543.1606, Found 543.1647; 54: mp 160 °C (dec); IR (neat):
1
IR (neat): 3213, 2238, 1631 cm-1; H NMR (500 MHz, CDCl3) δ
10.99 (s, 1H), 7.96 (d, J ) 8.0 Hz, 1H), 7.62 (d, J ) 8.2 Hz, 1H), 7.60
(t, J ) 7.9 Hz, 1H), 7.53 (d, J ) 8.2 Hz, 1H), 7.06 (d, J ) 8.4 Hz, 2H)
6.72 (s, 1H), 6.48 (s, 1H), 4.13 (s, 3H), 3.93 (s, 3H), 2.34 (s, 3H); 13
C
NMR (125 MHz, CDCl3) δ 161.5, 153.9, 151.7, 144.4, 141.5, 138.8,
136.5, 136.3, 135.0, 134.8, 132.2, 131.9, 128.4, 120.5, 119.6, 118.2,
117.1, 116.4, 115.5, 115.0, 114.1, 111.6, 56.6, 56.4, 21.9; ESI m/z
relative intensity 469 (MNa+ 100); TOFHRMS (+ESI) Calcd for
C28H18N2O4Na: 469.1164, Found 469.1156.
1
3389, 3311, 2921, 1611, 1580 cm-1; H NMR (500 MHz, CDCl3) δ
9.64 (s, 1H), 9.26 (s, 1H), 7.86 (d, J ) 8.5 Hz, 1H), 7.26 (t, J ) 7.8
Hz, 1H), 6.89-7.03 (m, 6H), 6.76 (d, J ) 7.6 Hz, 1H), 6.62 (s, 1H),
5.05 (s, 1H), 4.02 (s, 3H), 3.99 (s, 3H), 3.40 (d, J ) 12.9 Hz, 1H),
3.22 (d, J ) 12.9 Hz, 1H), 2.32 (s, 3H); 13C NMR (125 MHz, CDCl3)
δ 156.3, 151.3, 147.4, 143.5, 139.3, 139.1, 138.4, 128.9, 128.0, 127.9,
126.3, 125.1, 124.8, 122.6, 120.8, 120.3, 116.8, 115.3, 111.3, 104.9,
56.5, 56.0, 48.0, 33.5, 29.7; ESI m/z relative intensity 467 (MNa+ 95);
TOFHRMS (+ESI) Calcd for C27H24O4SNa: 467.1293, Found 467.1301.
Control Experiment: Benzyl Mercaptan Addition to Diazo-
paraquinone 29 with No Radical-Generating Ingredients. Benzyl
mercaptan (0.019 mL, 0.15 mmol) was added to a stirring solution of
diazoparaquinone 29 (5.0 mg, 0.015 mmol) at 80 °C. The light red
mixture was stirred at 80 °C for 1 h. Following 1 h of stirring, the
reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was diluted with CH2-
Cl2 and poured onto a silica gel column and purified by eluting with
hexanes/CH2Cl2 (1:1), followed by CH2Cl2 with an increasing percent-
age of EtOAc from 0% to 5%. Purification provided diazoparaquinone
29 (4.7 mg, 94%) unchanged.
C(11) Dimer of 5-Hydroxy-4,9-dimethoxy-2-methyl-benzo[b]-
fluoren-10-one (49). Following General Procedure 2, in many of the
solvent competition experiments aimed at converting diazoparaquinone
29 (20 mg, 0.06 mmol) to the aromatic solvent-trapped adducts (36 +
39a-g), the dimeric species (49) was isolated as a minor product (0-
3.3 mg, 0-15%): mp 210 °C (dec); IR (neat): 3414, 2921, 1626 cm-1
;
1H NMR (300 MHz, CDCl3) δ 10.64 (s, 1H), 7.59 (d, J ) 7.9 Hz,
1H), 7.47 (t, J ) 8.17 Hz, 1H), 6.96 (d, J ) 7.7 Hz, 1H), 6.62 (d, J )
8.1 Hz, 1H), 4.10 (s, 3H), 3.83 (s, 3H), 2.23 (s, 3H); 13C NMR (75
MHz, CDCl3) δ 180.8, 161.2, 151.6, 150.2, 142.7, 141.2, 138.3, 136.7,
133.8, 131.1, 121.3, 120.2, 118.3, 117.3, 115.0, 114.4, 111.1, 56.5,
56.1, 21.8; ESI m/z relative intensity 639 (MH+ 100); TOFHRMS
(+ESI) Calcd for C40H31O8: 639.2019, Found 639.2047.
5-Hydroxy-4,9-dimethoxy-2-methyl-11-phenylselanyl-benzo[b]-
fluoren-10-one (51). A solution of AIBN (11.0 mg, 0.067 mmol) in
benzene (1 mL) was added over the period of 1 h to a stirring solution
of diazoparaquinone 29 (20 mg, 0.06 mmol), Bu3SnH (0.018 mL, 0.067
mmol), and diphenyl diselenide (0.112 mg, 0.36 mmol) at 80 °C. The
light red mixture turned to a dark red/purple color upon addition of
AIBN. When the addition was complete, the reaction solution was
allowed to cool to room temperature. After reaching room temperature
the reaction mixture was diluted with CH2Cl2 and poured onto a AgNO3
impregnated silica gel column (silica gel column chromatography alone
Control Experiment: Benzyl Mercaptan Addition to Prekina-
mycin (5) with No Radical-Generating Ingredients. Benzyl mercap-
tan (0.015 mL, 0.12 mmol) was added to a stirring solution of
prekinamycin (5) (3.7 mg, 0.012 mmol) at 80 °C. The dark purple/
9
12572 J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006

Mechanism of Action of Prekinamycin
A R T I C L E S
brown mixture was stirred at 80 °C for 1 h. Following 1 h of stirring,
the reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was diluted with CH2-
Cl2 and poured onto a silica gel column and purified by eluting with
hexanes/CH2Cl2 (1:1), followed by CH2Cl2 with an increasing percent-
age of EtOAc from 0% to 5%. Purification provided prekinamycin (5)
(3.7 mg, 100%) unchanged.
11-Benzylsulfanyl-4,9-dimethoxy-2-methyl-11-phenyl-benzo[b]-
fluorene-5,10-diol (53) from Benzene-Trapped Adduct (36). Benzyl
mercaptan (0.100 mL, 0.80 mmol) was added to a stirring solution of
benzene-trapped adduct 36 (3.0 mg, 0.008 mmol) at 80 °C. The bright
red solution turned to a dark red solution over the 2 h of stirring at
80 °C. Upon consumption of the starting material by TLC, the reaction
solution was allowed to cool to room temperature. After reaching room
temperature the reaction mixture was diluted with CH2Cl2 and poured
onto a silica gel column and purified by eluting with hexanes/CH2Cl2
(1:1), followed by CH2Cl2 with an increasing percentage of EtOAc
from 0% to 5%. Purification afforded benzyl mercaptan addition product
53 (3.7 mg, 94%).
3H), 3.99 (s, 3H), 3.40 (d, J ) 12.9 Hz, 1H), 3.22 (d, J ) 12.9 Hz,
1H), 2.32 (s, 3H).
Deuterium-Labeling Experiment: PhCH2SD/Bu3SnD/AIBN. A
solution of AIBN (11.0 mg, 0.067 mmol) in benzene (1 mL) was added
over the period of 1 h to a stirring solution of diazoparaquinone 29
(20 mg, 0.06 mmol), Bu3SnD (0.018 mL, 0.067 mmol), and PhCH2SD
(0.077 mL, 0.6 mmol) at 80 °C. The light red mixture turned to a dark
brown color upon addition of AIBN. When the addition was complete,
the reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was diluted with CH2-
Cl2 and poured onto a silica gel column and purified by eluting with
hexanes/CH2Cl2 (1:1) followed CH2Cl2 with an increasing percentage
of EtOAc from 0% to 5%. Purification furnished benzyl mercaptan-
trapped adduct 54 (46% by 1H NMR) with 76% deuterium incorporation
1
at C(11). H NMR (400 MHz, CDCl3) δ 9.64 (s, 1H), 9.26 (s, 1H),
7.86 (d, J ) 8.5 Hz, 1H), 7.26 (t, J ) 7.8 Hz, 1H), 6.89-7.03 (m,
6H), 6.76 (d, J ) 7.6 Hz, 1H), 6.62 (s, 1H), 5.05 (s, 0.24H), 4.02 (s,
3H), 3.99 (s, 3H), 3.40 (d, J ) 12.9 Hz, 1H), 3.22 (d, J ) 12.9 Hz,
1H), 2.32 (s, 3H).
Deuterium-Labeling Experiment: PhCH2SD/Bu3SnH/AIBN. A
solution of AIBN (11.0 mg, 0.067 mmol) in benzene (1 mL) was added
over the period of 1 h to a stirring solution of diazoparaquinone 29
(20 mg, 0.06 mmol), Bu3SnH (0.018 mL, 0.067 mmol), and PhCH2SD
(0.077 mL, 0.6 mmol) at 80 °C. The light red mixture turned to a dark
red color upon addition of AIBN. When the addition was complete,
the reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was diluted with CH2-
Cl2 and poured onto a silica gel column and purified by eluting with
hexanes/CH2Cl2 (1:1), followed by CH2Cl2 with an increasing percent-
age of EtOAc from 0% to 5%. Purification furnished benzyl mercaptan-
trapped adduct 54 (33% by 1H NMR) with 46% deuterium incorporation
Deuterium-Crossover Control Experiment: PhCH2SD/Bu3SnH/
AIBN. AIBN (12.0 mg, 0.073 mmol), PhCH2SD (0.010 mL, 0.073
mmol), and Bu3SnH (0.020 mL, 0.073 mmol) were dissolved in d6-
1
benzene (1 mL) and inspected by H NMR, and it was noted that a
triplet at 1.43 ppm (PhCH2SH) was absent. The reaction solution was
then heated at 80 °C for 1 h. After 1 h at 80 °C, close inspection of the
1H NMR spectrum did not provide any evidence for a signal at 1.43
ppm.
Deuterium-Crossover Control Experiment: Bu3SnD/PhCH2SH/
AIBN. AIBN (12.0 mg, 0.073 mmol), Bu3SnD (0.020 mL, 0.073
mmol), and PhCH2SH (0.010 mL, 0.073 mmol) were dissolved in d6-
benzene (1 mL) and inspected by 1H NMR, and the presence of a triplet
at 1.43 ppm (1H) (PhCH2SH) was noted. The reaction solution was
then heated at 80 °C for 1 h. After 1 h at 80 °C, close inspection of the
1H NMR spectrum revealed that the triplet at 1.43 ppm remained and
had not diminished in intensity.
1
at C(11). H NMR (400 MHz, CDCl3) δ 9.64 (s, 1H), 9.26 (s, 1H),
7.86 (d, J ) 8.5 Hz, 1H), 7.26 (t, J ) 7.8 Hz, 1H), 6.89-7.03 (m,
6H), 6.76 (d, J ) 7.6 Hz, 1H), 6.62 (s, 1H), 5.05 (s, 0.54H), 4.02 (s,
3H), 3.99 (s, 3H), 3.40 (d, J ) 12.9 Hz, 1H), 3.22 (d, J ) 12.9 Hz,
1H), 2.32 (s, 3H).
Deuterium-Labeling Experiment: PhCH2SH/Bu3SnD/AIBN. A
solution of AIBN (11.0 mg, 0.067 mmol) in benzene (1 mL) was added
over the period of 1 h to a stirring solution of diazoparaquinone 29
(20 mg, 0.06 mmol), Bu3SnD (0.018 mL, 0.067 mmol), and PhCH2SH
(0.077 mL, 0.6 mmol) at 80 °C. The light red mixture turned to a dark
brown color upon addition of AIBN. When the addition was complete,
the reaction solution was allowed to cool to room temperature. After
reaching room temperature the reaction mixture was diluted with CH2-
Cl2 and poured onto a silica gel column and purified by eluting with
hexanes/CH2Cl2 (1:1), followed by CH2Cl2 with an increasing percent-
age of EtOAc from 0% to 5%. Purification furnished benzyl mercaptan-
trapped adduct 54 (40% by 1H NMR) with 27% deuterium incorporation
Acknowledgment. Financial support from the National
Institutes of Health, General Medical Sciences Division (GM
37861) is gratefully acknowledged.
Note Added after ASAP Publication. There were missing
nitrogen, hydrogen, and complete compound labels in Scheme
2 published ASAP August 29, 2006; the corrected version was
published ASAP September 6, 2006.
1
Supporting Information Available: Copies of H and 13C
NMR spectra for 34a, 35, 37, 38, 39a-g, 49, 51, 53, and 54 as
proof of purity. This material is available free of charge via the
Internet at http://pubs.acs.org.
1
at C(11). H NMR (400 MHz, CDCl3) δ 9.64 (s, 1H), 9.26 (s, 1H),
7.86 (d, J ) 8.5 Hz, 1H), 7.26 (t, J ) 7.8 Hz, 1H), 6.89-7.03 (m,
6H), 6.76 (d, J ) 7.6 Hz, 1H), 6.62 (s, 1H), 5.05 (s, 0.73H), 4.02 (s,
JA0642616
9
J. AM. CHEM. SOC. VOL. 128, NO. 38, 2006 12573