Target-directed enediynes: Designed estramycins

Article abstract of DOI:10.1021/jo0055842

The goal of selective targeting of enediyne cytotoxins has been investigated using estrogenic delivery vehicles. A series of estrogen-enediyne conjugates were assembled, and affinity for human estrogen receptor [hERα] was determined. The most promising candidate induced receptor degradation following Bergman cycloaromatization and caused inhibition of estrogen-induced transcription in T47-D human breast cancer cells.

Full text of DOI:10.1021/jo0055842

3688
J . Org. Chem. 2001, 66, 3688-3695
Ta r get-Dir ected En ed iyn es: Design ed Estr a m ycin s
Graham B. J ones,*,† George Hynd,† J ustin M. Wright,† Ajay Purohit,† Gary W. Plourde II,†
Robert S. Huber,‡ J ude E. Mathews,‡ Aiwen Li,‡ Michael W. Kilgore,‡ Glenn J . Bubley,§
Molly Yancisin,| and Myles A. Brown|
Department of Chemistry, Northeastern University, Boston, Massachusetts 02115, GHS-CU Biomedical
Cooperative, Clemson University, Clemson, South Carolina, 29634, Beth Israel Deaconess Medical Center,
Harvard Medical School, Boston, Massachusetts 02215, and Dana-Farber Cancer Institute,
Harvard Medical School, Boston, Massachusetts 02115
grjones@lynx.neu.edu
Received J uly 20, 2000
The goal of selective targeting of enediyne cytotoxins has been investigated using estrogenic delivery
vehicles. A series of estrogen-enediyne conjugates were assembled, and affinity for human estrogen
receptor [hERR] was determined. The most promising candidate induced receptor degradation
following Bergman cycloaromatization and caused inhibition of estrogen-induced transcription in
T47-D human breast cancer cells.
In tr od u ction
synthetic estrogens including diethylstilbestrol 2 and its
dihydro derivative, hexestrol.6 It has been noted that an
intraphenolic distance of ∼10-11 Å confers potent es-
trogenic activity. However, introduction of specific (basic)
functionality to these frameworks can confer antiestro-
genic character, as exemplified by raloxifene 3 and even
the tamoxifen metabolite 4, which act through through
a combination of competitive binding and inducing
changes in the receptor topology.7 The recent disclosure
of the X-ray crystal structures of 1 and 3 bound to the
hormone binding domain of ERR suggested that subtle
differences in binding interactions may translate to
profound differences in ER mediated physiological ef-
fects.8
Estrogens regulate cellular events through specific
interactions with intracellular receptors (designated ER
R and â) that function as ligand-inducible transcription
factors.1 In its inactive form, ER is bound to one or more
heat shock proteins,2 which are believed to mediate
folding of the protein during translation and aid in the
stability of the receptor. After hormone binding to the
estrogen receptor, a series of events follow that include
the dissociation of heat shock proteins,3 dimerization,4
recruitment of coactivators,5 and finally recognition and
binding to specific DNA sequences known as estrogen
response elements. Although the exact sequence of these
events is the subject of a great deal of investigation, all
these events precede the interaction of the complex with
the basic transcriptional machinery.5
Modulation of estrogen receptor function is of para-
mount importance for a variety of diseases including
breast cancer and osteoporosis. Indeed, it is well-
recognized that chemotherapeutic management of breast
cancer is most successful when the tumor, which may
contain up to 20 000 ERs per cell, is in the estrogen-
responsive phase. The development of competitive inhibi-
tors as antiestrogens has been influenced by QSAR
analysis of the endogenous agonist estradiol 1 and
† Northeastern University.
‡ Clemson University.
§ Beth Israel Deaconess Medical Center, Harvard Medical School.
| Dana-Farber Cancer Institute, Harvard Medical School.
(1) O’Malley, B. W. Mol. Endocrinol. 1990, 4, 363; Evans, R. M.
Science 1988, 240, 889. Kumar, V.; Green, S.; Stack, G.; Berry, M.;
J in, J .-R.; Chambon, P. Cell 1987, 51, 941.
(2) Dunn, D. K.; Whelan, R. D.; Hill, B.; King, R. J . J . Steroid
Biochem. Mol. Biol. 1993, 46(4), 469; Segnitz, B.; Gehring, U. Proc.
Natl. Acad. Sci. U.S.A. 1995, 92, 2179.
A potential strategy for therapeutic management of
breast cancer involves use of cytotoxic antiestrogens,
where a chemically reactive functional group is attached
(3) J oab, J .; Radanyi, C.; Renoir, J . M.; Buchou, T.; Catelli, M.-G.;
Binart, N.; Mester, J .; Balulieu, E. E. Nature 1984, 308, 850. Sanchez,
E. R.; Faber, L. E.; Henzel, W. J .; Pratt, W. B. Biochemistry 1990, 29,
5145. Sanchez, E. R. J . Biol. Chem. 1990, 265, 22067.
(4) Gordon, M. S.; Notides, A. C. J . Steroid Biochem. 1986, 25, 177.
Linstedt, A. D. ; West, N. D.; Brenner, R. M. J . Steroid Biochem. 1986,
24, 677. Kumar, V.; Chambon, P. Cell 1988, 55, 145.
(5) Chen, H.; Lin, R. J .; Xie, W.; Wilpitz, D.; Evans, R. M. Cell 1999,
98, 675. Ing, N. H.; Beekman, J . M.; Tsai, S. Y.; Tsai, M.-J .; O’Malley,
B. W. J . Biol. Chem. 1992, 267, 17617.
(6) Gao, H.; Katzenellenbogen, J . A.; Garg, R.; Hansch, C. Chem.
Rev. 1999, 99, 723. Korach, K. S. In Molecular Structure and Biological
Activities of Steroids; Bohl, W., Duax, W. L., Eds.; CRC Press: Boca
Raton, FL, 1992; p 210.
(7) Kym, P. R.; Anstead, G. M.; Pinney, K. G.; Wilson, S. R.;
Katzenellenbogen, J . A. J . Med. Chem. 1993, 36, 3910.
(8) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G. L.; Gustafsson, J .-A.;
Carlquist, M. Nature 1997, 389, 753.
10.1021/jo0055842 CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/01/2001

Target-Directed Enediynes
J . Org. Chem., Vol. 66, No. 11, 2001 3689
Sch em e 1. Design for En ed iyn e Hybr id s
to a carrier substrate that has affinity for ER and can
thus potentially accumulate the cytotoxin in an ER-rich
cell.9 In most cases reported to date, the principal target
of these agents is the DNA of ER-responsive tumor
cells.9,10 Though clinical candidates have emerged, gener-
ally this has been an unsuccessful strategy, largely
because of independent cytotoxicity and poor accumula-
tion of the drugs in the desired target cells. These
problems are often a consequence of the structural
features of the cytotoxic group, which may reduce the
overall binding affinity of the hybrid for ER. Subtle
changes in lipophilicity often adversely affect ER binding
affinity, and there are few chemically reactive functional
groups that will not have a pronounced effect on this
parameter.6,9 Guided by these constraints, we sought to
incorporate carbocyclic enediynes into cytotoxic estrogen
hybrids.
Enediynes are one of the most recently discovered
classes of antitumor agents and have garnered a great
deal of interest over the past decade due to their
exceptional cytotoxicity, a consequence of their ability to
generate diyl radicals on cycloaromatization of the ene-
diyne core.11 The presumed target of these diyl radicals
is DNA, resulting in single- and double-stranded lesions.
However, protein targets have also been identified for
specific enediynes, resulting in proteolysis,12 protein
agglomeration,13 and protein dimerization,14 and at the
molecular level it has been demonstrated that amino acyl
radicals are generated when amino acids are exposed to
diyl radicals.15 Harnessing the proteolytic/DNA cleaving
capacity of a designed enediyne toward the transcrip-
tional machinery of an ER-rich tumor cell therefore
constituted an attractive proposition.16 A number of
tumors have been shown to possess high concentrations
of ER including breast cancer, prostatic carcinoma,
melanoma, ovarian adenocarcinoma, colon adenocarci-
noma, hypernephroma, and endometrial carcinoma. The
half-lives of unstrained C-10 monocyclic enediynes is
approximately 12-18 h at physiological temperature,
making them ideal as thermally generated cytotoxins.17
Our aim was to assemble conjugates 6 using a method
for enediyne synthesis pioneered in this laboratory, which
involves an in situ coupling elimination from bis prop-
argylic halides 5 (Scheme 1).18 Incubation of substrates
6 with ER-rich cells then offers the opportunity to gauge
the effects of the 1,4-diyl cycloaromatization products 7
on the transcriptional machinery.19
Resu lts a n d Discu ssion
Since a common feature of many molecules with
affinity for ER is a phenolic group (viz. 1-4), we elected
to initially study coupling of the enediyne moiety to the
carrier molecule via a phenolic ether linkage. We recently
developed a late-stage coupling method for the synthesis
of linear and cyclic enediynes18 and, to avoid thermal
decomposition during synthetic manipulation, elected to
attach the cyclic enediyne precursors to the desired
phenol. Using phenol itself as a model study, readily
available diyne 8 was converted to free alcohol 9, coupled
under Mitsonobu conditions, and then transformed into
propargyl bromide 10 (Scheme 2). Direct conversion to
the enediyne 12 was effected using the metallohalocar-
benoid coupling-elimination; however, to enable storage
of the enediyne, conversion to the corresponding dicobalt
hexacarbonyl complex 11 was effected. Liberation of the
unmasked enediyne 12 was effected using TBAF,20 and
in the presence of excess 1,4-cyclohexadiene, cycloaro-
matization to yield 14 took place, presumably via inter-
mediate diyl 13. The identity of 14 was confirmed by
independent synthesis involving Mitsonobu coupling of
phenol with 1,2,3,4-tetrahydro-2-naphthalene methanol.
Kinetics of the Bergman cycloaromatization were inves-
tigated, and it was determined that the t1/2 for enediyne
12 is 18 h at 37 °C.
(9) Catane, R.; Bruno, S.; Muggia, F. M. In Cytotoxic Estrogens in
Hormone Receptive Tumors; Raus, J ., Leclercq, G., Martens, H., Eds.;
Academic Press: London, 1980.
(10) Wittliff, J . L.; Weidner, N.; Everson, R. B.; Hall, T. C. Cancer
Treatment Rep. 1978, 62, 1262.
(11) Xi, Z.; Goldberg, I. In Comprehensive Natural Products Chem-
istry; Barton, D. H. R., Nakanishi, K., Eds.; Pergamon: Oxford, 1999;
Vol. 7, p 553. Nicolaou, K. C.; Smith, A. L.; Yue, E. W. Proc. Natl.
Acad. Sci. U.S.A. 1993, 90, 5881. Nicolaou, K. C.; Dai, W. M. Angew.
Chem., Intl. Ed. Engl. 1991, 30, 1387.
(12) Zein, N.; Reiss, P.; Bernatowicz, M.; Bolgar, M. Chem. Biol.
1995, 2, 451. Zein, N.; Casazza, A. M.; Doyle, T. W.; Leet, J . E.;
Schroeder, D. R.; Solomon, W.; Nadler, S. G. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 8009. Zaheer, A.; Zaheer, S.; Montgomery, R. J . Biol.
Chem. 1985, 260, 11787.
(13) Zein, N.; Solomon, W.; Cassazza, A. M.; Kadow, J . F.; Krishnan,
B. S.; Tun, M. M.; Vyas, D. M.; Doyle, T. W. Bioorg. Med. Chem. Lett.
1993, 3, 1351.
(14) J ones, G. B.; Wright, J . M.; Plourde, G. W.; Wyatt, J . K.; Hynd,
G.; Fouad, F.; Purohit, A. J . Am. Chem. Soc. 2000, 122, 9872.
(15) J ones, G. B.; Plourde, G. W.; Wright, J . M. Org. Lett. 2000, 2,
811.
(16) Related efforts: (a) Wang, J .; DeClercq, P. J . Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1749. (b) Py, S.; Harwig, C. W.; Banerjee, S.;
Brown, D. L.; Fallis, A. G. Tetrahedron Lett. 1998, 39, 6139.
(17) Grissom, J . W.; Gunawardena, G. U.; Klingberg, D.; Huang,
D. Tetrahedron 1996, 52, 6453. Smith, A. L.; Nicolaou, K. C. J . Med.
Chem. 1996, 39, 2103. Magnus, P.; Fortt, S.; Pitterna, T.; Snyder, J .
P. J . Am. Chem. Soc. 1990, 112, 4986.
With the model study in place, we turned our efforts
toward bona fide estrogen conjugates and selected the
readily available bis-phenols hexestrol and diethylstil-
bestrol (DES). Since our coupling protocol involves mask-
(18) J ones, G. B.; Wright,J . M.; Plourde, G. W., II; Hynd, G.; Huber,
R. S.; Mathews, J . E. J . Am. Chem. Soc. 2000, 122, 1937. J ones, G. B.;
Huber, R. S.; US Pat. 5, 436, 361.
(19) J ones, G. B.; Huber, R. S.; Mathews, J . E.; Li, A. Tetrahedron
Lett. 1996, 37, 3643. J ones, G. B.; Wright, J . M.; Hynd, G.; Wyatt, J .
K.; Yancisin, M.; Brown, M. A. Org. Lett. 2000, 2, 1863.
(20) J ones, G. B.; Wright, J . M.; Rush, T. M.; Plourde. G. W.; Kelton,
T. F.; Mathews, J . E.; Huber, R. S.; Davidson, J . P. J . Org. Chem. 1997,
62, 9379.

3690 J . Org. Chem., Vol. 66, No. 11, 2001
Sch em e 2. Mod el Stu d ies for P h en ol-Lin k ed En ed iyn es
J ones et al.
Sch em e 3. P r ep a r a tion a n d Cycloa r om a tiza tion of DES a n d Hexestr ol Con ju ga tesa
a
Legend: (i) LiHMDS/HMPA then Co2(CO)8 (16-81%); (ii) TBAF/THF -20 °C then 1,4-CHD, 37 °C/48 h (84-89%).
ing of one of the two primary binding sites of these
enediyne in good yield, isolated as the dicobalt hexacar-
bonyl complex 23. Finally, one-pot decomplexation-
desilation gave the thermally unstable enediyne, which
underwent cycloaromatization and trapping to give arene
24.
estrogens, we also wished to prepare suitable reference
compounds for use as control elements in subsequent
bioassay. Accordingly, coupling DES with diyne 9 fol-
lowed by phenolic protection (methyl or TBDMS ether)
and subsequent propargylic bromination gave enediyne
precursors 15 (Scheme 3). Carbenoid coupling followed
by immediate protection of the enediyne core gave 16,
which when unmasked under controlled conditions and
the enediyne subjected to cycloaromatization in the
presence of 1,4-cyclohexadiene gave adducts 17. In the
case of 16b, the method of deprotection used facilitated
in situ desilation, to give the free phenol 17b. Similarly,
using hexestrol as the phenolic component, bromides 18
were prepared, which gave arenes 20 on cycloaromati-
zation of the derived enediynes (Scheme 3).
Finally, an estradiol conjugate was prepared. This
required coupling of dialkyne 9 with commercially avail-
able estrone, giving the steroidal diyne 21 following
reduction of the 17 keto group (Scheme 4). Protection of
the 17â alcohol, liberation of the propargyl alcohols, and
subsequent bromination then gave carbenoid precursor
22. In situ coupling-elimination gave the corresponding
Recep tor Bin d in g. To assess the ability of the
compounds to recognize and bind to the presumed target,
ERR, a competitive binding assay was employed. The
assay requires incubation with ligand over extended
periods, during which time the enediyne would undergo
cycloaromatization. Accordingly, we elected to assay the
cycloaromatized products rather than the enediynes
themselves. However, since the Bergman cycloaromati-
zation proceeds via a late-stage transition state, the diyl
radical intermediates are “product-like”, suggesting the
arene products are appropriate structural mimics of the
diyls. Accordingly, arenes 17, 20, and 24 were incubated
with recombinant ER in the presence of 3H estradiol, and
affinity for ER was calculated using Scatchard analysis.21
Under the conditions employed (Table 1) affinity for all
compounds was disappointingly low, underscoring the
(21) Scatchard, G. Ann. N. Y. Acad. Sci. 1949, 51, 660.

Target-Directed Enediynes
J . Org. Chem., Vol. 66, No. 11, 2001 3691
Sch em e 4. P r ep a r a tion of 3-Hyd r oxy Estr a d iol Con ju ga te
Sch em e 5. Attem p ted Rou te to 17r-Alk yn yl Estr a d iol Con ju ga te
Ta ble 1. Bin d in g Affin ity for ERra
was prepared and a 17R alkyne moiety introduced
(Scheme 5). A one-pot protection of the 17â hydroxy group
and deprotection of the alkynyl silyl group was achieved,
giving alkyne 26. To introduce convergence to the syn-
thesis, for the enediyne component, we elected to use a
preformed C-10 enediyne, available from our library of
existing structures.18 We opted to couple the steroid and
enediyne via an ester linkage, and conversion of 26 to
carboxylate 27 was thus conducted. Direct coupling of 27
with the shelf-stable masked enediyne complex 28 was
effected, giving the ester hybrid 29 in excellent yield. It
was anticipated that global deprotection using TBAF
would give the free steroidal enediyne (33). However,
despite exhaustive efforts, 30 was the only product
isolated. Indeed, a variety of alternate methods for
deprotection at this (17â) position were attempted with-
out success, confirming earlier reports of the problems
encountered in deprotection at this position when a 17R
substituent is in place.16a
compd
RBA/M
1
1.0 × 10-9
4.3 × 10-4
2.6 × 10-5
6.4 × 10-4
3.1 × 10-5
7.7 × 10-5
5.1 × 10-7
17a
17b
20a
20b
24
35
Candidates/controls + 3H 1 (2.5 nM) incubated with cytosol
at 4 °C. Unbound agents removed (DCC) and bound H 1 measured
by scintillation counter.21 RBA corresponds to concentration
required to reduce ER-bound 3H 1 by 50%.
a
3
importance of not only having one but two hydroxy
functions available for productive binding interactions
with the receptor.
Refin em en t. Though a multitude of substituted ste-
roid templates are available via direct synthesis, we were
particularly interested in commencing from a com-
mercially available building block. Surveying numerous
options, it was decided to utilize estrone, introducing the
enediyne functionality at the 17R position. Through
QSAR analysis it has been shown that 17R alkynyl
derivatives of estradiol show comparable ER affinity to
estradiol,22 and we decided to construct a hybrid with an
alkynyl linker between the enediyne and the steroidal
template. Accordingly, the TBDMS ether of estrone 25
Commencing from commercially available ethynyl es-
tradiol 31, monoprotection followed by carboxylation of
the dianion gave alkynoic acid 32 directly (Scheme 6).
The triethylsilyl derivative was chosen over the tert-
butyldimethylsilyl to permit more rapid deprotection.
(22) Pomper, M,. G.; VanBrocklin, H.; Thieme, A. M.; Thomas, R.
D.; Kiesewetter, D. O.; Carlson, K. E.; Mathias, C. J .; Welch, M. J .;
Katzenellenbogen, J . A. J . Med. Chem. 1990, 33, 3143.

3692 J . Org. Chem., Vol. 66, No. 11, 2001
Sch em e 6. Mod ified Rou te to 17r-Alk yn yl Estr a d iol Ca n d id a tea
J ones et al.
a
Legend: (a) TESCl (94%); (b) BuLi, CO2 (69%); (c) EDCI, 28 then TBAF (78%); (d) 1,4-CHD (69%).
F igu r e 1. Ability of glutathione-bound ERR to recruit SRC1, GRIP, and RAC3 coactivators in the presence of 10-6 M 1, 35, or
ICI 182,780.
Esterification with cobalt complex 28 followed by mild
deprotection gave desired enediyne 33 in good yield,
Recep tor Degr a d a tion . Our initial assumption was
that 33, when bound to the hERR, may induce proteolysis
of either the receptor or its ternary complexes, following
Bergman cycloaromatization to 34. To address this
possibility, a freshly prepared sample of 33 was incubated
with 35S-labeled full-length hERR at various concentra-
tions for 2 half-lives (36 h) at either 4 or 37 °C, then the
protein was separated using SDS-PAGE and visualized
using fluorography. The results indicate that the ene-
diyne induces degradation of the receptor (Figure 2, lanes
13-14) and that the process has concentration and
temperature-dependent components (lanes 6-7). Control
reactions with either estradiol, the antiestrogens 4-hy-
droxytamoxifen or ICI 182,780,23 or arene 35 indicate the
enediyne-estrogen conjugate is responsible for the deg-
radation, which may imply a proteolytic mechanism. The
which underwent cycloaromatization to arene 35 t1/2
)
14 h/37 °C. Gratifyingly, the binding affinity of arene 35
was in the sub-micromolar range, finally allowing bio-
logical evaluation of a relevant estrogen-enediyne con-
jugate.
Co-Activa tor Recr u itm en t Assa ys. To determine
the nature of the interaction of 35 (and therefore 33) with
ERR, a recruitment assay was conducted, which assessed
the ability of candidate compounds to allow the hormone
binding domain of hERR to bind key co-activators. The
assay revealed that 35, like â-estradiol, is capable of
promoting recruitment of the SRC1, GRIP1, and RAC3
estrogen receptor co-activator proteins, which are impor-
tant for formation of estrogenic complexes competent for
transcription (Figure 1). This suggests that 35 has
estrogen-like properties in contrast to antiestrogens (e.g.,
ICI 182,780), which inhibit co-activator recruitment.
(23) Wakeling, A. E.; Bowler, J . J . Steroid Biochem. Mol. Biol. 1992,
43, 173.

Target-Directed Enediynes
J . Org. Chem., Vol. 66, No. 11, 2001 3693
F igu r e 2. ERR degradation mediated by enediyne 33‚35S-methionine-labeled full-length hERR incubated with either ethanol
alone (lanes 1 and 8), estradiol (lanes 2 and 9, 10 µM), 4-OH-tamoxifen (lanes 3 and 10, 10 µM), ICI182,780 (lanes 4 and 11, 10
µM) or 33 (lanes 5-7 and 12-14 at concentrations indicated) at either 37 or 4 °C for 36 h.
observation of receptor degradation at 10 µM 33 is
encouraging since the binding affinity of 35 for hERR was
only 0.51 µM. The efficiency of atom transfer from
enediyne-derived 1,4-diyls (e.g., 34) is often low, typically
requiring multiple (>30) equivalents of donor species.
Thus, within the context of receptor affinity, the observed
degradation is relatively efficient and may encourage
other applications of this strategy to be developed.
Though no specific fragments were isolated, it is entirely
possible that improved analogues can be developed that
approach the binding affinity levels observed for natural
ligands, including â estradiol, which in turn may improve
both the specificity and selectivity of the degradation
events. Incubation of the enediyne derived from 16a
(deprotection with TBAF) showed no receptor degrada-
tion at 10-5 M. Since the RBA of its cycloaromatization
product (17a ) is lower than that of 35 (Table 1), this
implies a tentative correlation between receptor affinity
and cleavage potential.24 Intriguingly, receptor degrada-
tion induced by 33 was partially attenuated when
conducted in the presence of estradiol, though the mech-
anism of this inhibition is unresolved.25
Indeed, this is logical given the fact that the diyl, when
generated, is presumably positioned several angstroms
remote from the primary binding interactions in the
receptor-steroid complex. We thus probed estrogenic
activity under the influence of enediyne 33 and the
endogenous ligand for hER, estradiol, by employing a
luminometric transcriptional activation assay.26 Specif-
ically, 33 was analyzed for its ability to limit estrogen-
induced transcription in T47-D breast cancer cells using
an assay procedure that relies on an estrogen response
element (ERE), coupled with a luciferase gene, to allow
luminometric quantitation of transcription. Controls were
employed using cells lacking the response element. In
preliminary studies, transfected cells were pretreated for
16 h with enediyne 33 (10-7 M) or media only and then
treated for 6 h with estradiol (E2, 10-9 M). These analyses
(Figure 3) demonstrate that in just 6 h estradiol clearly
activates transcription of its target sequence between 2-
and 3-fold. Significantly, pretreating for 16 h with ene-
diyne 33 results in a marked and statistically significant
(25) The degradation of hER using 33 (10-5 M, 37 °C) is partially
suppressed when assays are conducted in the presence of estradiol (up
to 10-5 M). At estradiol concentrations of 10-6 M, this apparent
inhibitory effect is negligible, and no effect is observed at lower
concentrations, suggesting the observation is not directly related to
displacement. Intriguingly, a similar observation was found with
phenol (10-4 M), tempting speculation that an electron-transfer process
involving phenoxy radicals may be involved.
Tr a n scr ip tion In h ibition . Since the degradation
assay revealed partial decomposition of receptor at micro-
molar levels, we opted to study the effect of the diyl
radical on receptor mediated transcriptional activation.
(24) Similar results were obtained with the enediyne derived from
19a .
(26) Dana, S. L.; Hoener, P. A.; Wheeler, D. A.; Lawrence, C. B.;
McDonnell, D. P. Mol. Endocrinol. 1994, 8, 1193.

3694 J . Org. Chem., Vol. 66, No. 11, 2001
J ones et al.
either the estradiol displaces the enediyne from the
receptor or interacts with the enediyne to diminish its
potency.25
Con clu sion s
The goal of selective targeting of enediyne cytotoxins
has been applied to estrogenic delivery vehicles. A
general strategy for the synthesis of 17R substituted
enediyne derivatives of estradiol has been developed.
Binding affinity, transcription inhibition, and estrogen
receptor degradation show some correlations, suggesting
the possibility of a mode of action involving the trans-
criptional machinery. On the basis of these preliminary
findings, it is now of interest to perform in-depth biologi-
cal screens and to develop optimized candidates with
nanomolar level affinity. The commercial availability of
steroid 31 and easy access to cyclic enediynes including
28 will expedite this quest. Refinement of these hybrids
may pave the way for more specific cytotoxins, including
photoactivated enediyne cores, which may function as
photodynamic therapies,29 or enzyme inhibitors.30 Per-
haps the greatest potential of such enediyne-hormone
conjugates is as dynamic probes of transcription factor
assembly, an application that presumably could be
extended to other members of the nuclear receptor
superfamily.
F igu r e 3. ER transcription in T47-D cells. Data reported as
fold induction over transfected control cells (no steroid treat-
ment): 1 ) control cells, 16 h; 2 ) control + 10-7 M enediyne/
16 h; 3 ) as 1 then + 10-9 M 1 /6 h; 4 ) as 2 then + 10-9
M
1/6 h. Responses calculated as raw luminometric units (RLU).
Ta ble 2. Gr ow th In h ibition of 33a
cells
IC50 (µM)
MCF-7
MDB-231
LNCaP
31
48
33
10
HEK-293
a
Cells split, grown to 50% confluence, then treated with 33
(triplicate, 10-3-10-9 M) and 3H thymidine.28
Exp er im en ta l P r oced u r es18
3-Tr ieth ylsilyloxy-17-r-eth yn yl-â-estr adiol. Et3SiCl (1.68
g, 11.13 mmol, 1.87 mL) and imidazole (1.03 g, 15.2 mmol)
were added to a solution of â-ethynyl estradiol 31 (3.0 g, 10.2
mmol) in anhydrous DMF (25 mL). The resulting solution was
stirred for 3 h at 25 °C and then diluted with EtOAc (100 mL)
and poured onto cold HCl (1%, 100 mL). The phases were
separated and the aqueous layer further extracted with EtOAc
(3 × 50 mL). The combined organic extracts were washed with
water (3 × 100 mL) followed by saturated NaHCO3 (2 × 100
mL) and saturated aqueous NaCl (1 × 100 mL), dried over
Na2SO4, and concentrated in vacuo. Chromatography (90:10
hexanes/EtOAc) of the residual oil afforded the title compound
(3.9 g, 94%) as a colorless oil: 1H NMR (300 MHz, CDCl3) δ
7.13 (d, J ) 8.7 Hz, 1H), 6.63 (dd, J ) 8.1, 2.4 Hz, 1H), 6.57
(d, J ) 3.0 Hz, 1H) 2.80 (m, 2H), 2.60 (s, 1H), 2.36 (m, 2H),
2.22 (m, 1H), 1.86 (m, 5H), 1.44 (m, 3H), 1.10 (t, J ) 7.9 Hz,
9H), 0.99 (s, 3H), 0.74 (m, 6H); 13C NMR (75 MHz, CDCl3) δ
153.2, 137.8, 132.8, 126.1, 119.8. 117.0, 87.5, 79.9, 74.0, 49.5,
47.1, 43.5, 39.3, 38.9, 32.7, 29.6, 27.2, 26.3, 22.8, 12.7, 6.8, 5.0.
Anal. Calcd for C26H38O2Si: C, 76.04; H, 9.33. Found: C, 75.71;
H, 9.21.
3-Tr ieth ylsilyloxy-17-r-(2-p r op yn oic)-â-estr a d iol (32).
n-BuLi (19.5 mmol, 7.8 mL of 2.5 M in hexanes) was added
dropwise over 10 min at -15 °C to a solution of 3-O-
triethylsilyloxy-17-R-ethynyl-â-estradiol (1.60 g, 3.9 mmol) in
THF (20 mL). The mixture was stirred at -15 °C for 0.5 h,
and then anhydrous CO2 (dried by passage through 4 Å
molecular sieves and then concentrated H2SO4) was bubbled
through the solution for 1 h. The mixture was warmed to 25
°C, and CO2 bubbling continued until the reaction mixture
reached dryness. The residue was dissolved in EtOAc (50 mL)
and the resulting solution poured onto HCl (1%, 30 mL). The
layers were separated, and the aqueous layer was extracted
reduction in the ability of the cells to respond to subse-
quent estrogen treatment to initiate transcription. This
result indicates that alteration in the ability of the
receptor to bind estrogen, recognize and bind response
elements on the DNA, or interact with the transcriptional
machinery has been significantly impaired by enediyne
treatment. While these data do not allow us to interpret
the molecular mechanism of this effect, it clearly indi-
cates that the cells ability to transduce the estrogen
signal has been altered by preincubation with 33. Thus,
while it is possible that the diyl 34 degrades the actual
receptor via a proteolytic mechanism (Figure 2), it may
instead impair the fidelity of the transcriptionally active
complex via a more subtle mechanism.
Gr ow th In h ibition . The antiproliferative effect of
enediyne 33 was assessed using a hERR-rich cell line
(MCF-7 human breast cancer), a hERR-deficient cell line
(MDB-231), an androgen receptor (AR)-positive cell line
(LNCaP), and an AR-negative cell line which is sensitive
to cytotoxins (HEK-293).27 A standard thymidine uptake
assay was employed,28 with a 56 h incubation time,
corresponding to 4 half-lives of 33. Though the IC50 values
show elevated growth inhibition toward the ER and AR
positive cells relative to the ER negative (Table 2), the
marked inhibition of the HEK-293 cells suggests an
independent mechanism for cytotoxicity. Complicating
the issue is the fact that the cycloaromatization product
(35) is a demonstrated estrogen (Figure 1), which could
be expected to stimulate growth of the ER rich cells.
Indeed, addition of estradiol (10-6 M) followed by 33 (10-6
M) resulted in net growth of MCF-7 cells, suggesting that
(29) Funk, R. L.; Young, E. R. R.; Williams, R. M.; Flanagan, M. F.;
Cecil, T. L. J . Am. Chem. Soc. 1996, 118, 3291. Turro, N. J .; Evenzahav,
A.; Nicolaou, K. C. Tetrahedron Lett. 1994, 35, 8089. Beijersbergen,
G. M. J . In Drugs: Photochemistry and Photostability; Albini, A.,
Fasani, E., Eds.; Royal Society of Chemistry: London, 1998; pp 74-
86.
(27) J ones, G. B.; Mitchell, M. O.; Weinberg, J . S.; D’Amico, A. V.;
Bubley, G. A. Bioorg. Med. Chem. Lett. 2000, 10, 1987.
(28) McConnuaghie, A. W.; J enkins, T. C. J . Med. Chem. 1995, 38,
3488.
(30) Woo, L. W.; Purohit, A.; Reed, M. J .; Potter, B. V. L. Bioorg.
Med. Chem. Lett. 1997, 7, 3075.

Target-Directed Enediynes
J . Org. Chem., Vol. 66, No. 11, 2001 3695
with EtOAc (3 × 30 mL). The combined organic extracts were
washed with HCl (1%, 2 × 50 mL) and saturated aqueous
NaCl, (2 × 50 mL), dried over Na2SO4, and concentrated in
vacuo. The residual oil was purified via chromatography (30:
70-100:0 EtOAc/hexanes) to yield 32 (1.22 g, 69%) as a thick
oil: 1H NMR (300 MHz, CD3OD) δ 7.10 (d, J ) 8.1 Hz, 1H),
6.57 (dd, J ) 8.1, 2.4 Hz, 1H), 6.51 (d, J ) 2.4 Hz, 1H) 2.75
(m, 2H), 2.26 (m, 3H), 1.86 (m, 8H), 1.39 (m, 4H), 0.99 (t, J )
8.1 Hz, 9H), 0.86 (s, 3H), 0.71 (m, 6H); 13C NMR (75 MHz,
CD3OD) δ 156.2, 152.2, 137.5, 133.1, 125.8, 118.9. 118.1, 87.8,
80.8, 74.2, 50.1, 48.2, 44.3, 38.4, 38.2, 32.8, 30.1, 28.2, 27.5,
23.9, 13.7, 7.0, 4.9. Anal. Calcd for C27H38O4Si: C, 71.32; H,
8.42. Found: C, 71.49; H, 8.56.
17r-([5-Cyclod ecen -3,7-d iyn ylm eth yl]-2-p r op yn oa te)-
â-estr a d iol (33). A solution of 32 (0.409 g, 0.899 mmol), EDCI
(0.172 g, 0.899 mmol), 28 (0.692 g, 0.899 mmol),18 and DMAP
(0.012 g, 0.0899 mmol) in DMF (10 mL) was stirred at 0 °C
for 5 h. The mixture was then poured onto a biphasic mixture
of water (15 mL) and EtOAc (15 mL). The layers were
separated, and the aqueous layer was extracted with EtOAc
(3 × 25 mL). The combined organic extracts were washed with
HCl (2%, 2 × 30 mL) and saturated aqueous NaCl (2 × 30
mL), dried over Na2SO4, and concentrated in vacuo. The
residual oil was immediately dissolved in THF (3 mL) and the
solution cooled to 0 °C. Bu4NF (8.99 mmol, 8.99 mL of 1.0 M
in THF) was added dropwise over 10 min, and the mixture
was stirred at 0 °C for an additional 1 h and then poured onto
a saturated solution of NH4Cl (3.0 mL). The layers were
separated, and the aqueous layer was extracted with Et2O (3
× 5 mL). The combined organic extracts were washed with
HCl (1%, 2 × 5 mL), NaHCO3 (2 × 5 mL), and saturated
aqueous NaCl (1 × 5 mL) dried over MgSO4, and concentrated
in vacuo. The residual oil was purified via radial chromatog-
raphy (10:90-50:50 EtOAc/hexanes) to yield 33 (0.339 g, 78%)
as a yellow oil: 1H NMR (300 MHz, CDCl3) δ 7.12 (d, J ) 8.4
Hz, 1H), 6.62 (dd, J ) 8.4, 2.7 Hz, 1H), 6.55 (d, J ) 2.4 Hz,
1H), 5.82 (s, 2H), 4.11 (m, 2H), 2.79 (m, 3H), 2.19 (m, 7H),
1.46 (m, 13H), 0.93 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 153.8,
153.5, 138.0, 132.0, 126.4, 123.3, 123.0, 115.1, 112.6, 103.3,
101.4, 91.6, 83.1, 83.0, 79.9, 69.4, 49.7, 47.8, 43.1, 40.6, 39.2,
38.5, 32.9, 32.5, 31.5, 29.5, 27.0, 26.1, 24.7, 22.9, 22.6, 12.6.
17r-[(1,2,3,4-Tet r a h yd r o-2-n a p h t h a len en ylm et h yl)-2-
p r op yn oa t e]-â-est r a d iol (35). A solution of enediyne 33
(0.097 g, 0.2 mmol) in 1,4-cyclohexadiene (5 mL) was degassed
and then sealed in a 10 mm NMR tube and incubated for 48
h at 37 °C in a constant temperature heating bath. On cooling,
the solution was filtered through a silica gel plug and then
condensed in vacuo, and the residual oil was purified via radial
chromatography (10:90-50:50 EtOAc/hexanes) to yield 35
(0.067 g, 69%) as a thick foam: 1H NMR (300 MHz, CDCl3) δ
7.10 (m, 5H), 6.63 (dd, J ) 8.1, 2.4 Hz, 1H), 6.56 (d, J ) 3.0
Hz, 1H), 4.20 (dd, J ) 6.6, 1.2, 2H), 2.86 (m, 4H), 2.31 (m,
6H), 1.57 (m, 13H), 0.93 (s, 3H); 13C NMR (75 MHz, CDCl3) δ
154.0, 153.4, 138.1, 136.2, 135.1, 132.3, 129.1, 128.9, 126.5,
125.8, 125.7, 115.2, 112.6, 91.0, 80.0, 78.0, 70.2, 55.1, 54.9, 49.8,
47.9, 43.2, 39.3, 33.6, 32.3, 29.5, 28.4, 27.0, 26.2, 25.9, 22.9,
12.7. Anal. Calcd for C32H36O4: C, 79.31, H, 7.49. Found: C,
79.89, H, 7.56. Authenticity of the material was confirmed by
independent synthesis, coupling 32 with 1,2,3,4-tetrahydro-
2-naphthalenylmethanol followed by deprotection (HF/Py).34
Ack n ow led gm en t. We thank the NIH for financial
support (RO1GM57123) and Dr. Franc¸ois Nique (Rous-
sel-Uclaf) and Professor Philip Magnus for useful dis-
cussions.
Su p p or tin g In for m a tion Ava ila ble: Synthetic proce-
dures and spectroscopic data for the preparation of 8-30 and
procedures used for bioassay of 33. This material is available
free of charge via the Internet at http://pubs.acs.org.
J O0055842
(31) Li, X.; Singh, S. M.; Labrie, F. Tetrahedron Lett. 1994, 35, 1157;
Poirier, D.; Merand, Y.; Labrie, C.; Labrie, F. Bioorg. Med. Chem. Lett.
1996, 6, 2537.
(32) Katzenellenbogen, J . A.; Heiman, D. F.; Carlson, K. E.; Payne,
D. W.; Lloyd, J . E. In Cytotoxic Estrogens in Hormone Receptive
Tumors; Raus, J ., Leclercq, G., Martens, H., Eds.: Academic Press:
London, 1980. Rhee, C. K.; Chae, K.; Levy, L. A.; Korach, K. S. Bioorg.
Med. Chem. Lett. 1995, 5, 133.
(33) Berthois, Y.; Katzenellenbogen, J . A.; Katzenellenbogen, B. S.
Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 2496.
(34) Beckwith, A. L. J .; Gerba, S. Aust. J . Chem. 1992, 45, 289.