Identification of a series of tetrahydroisoquinoline derivatives as potential therapeutic agents for breast cancer

Article abstract of DOI:10.1016/j.bmcl.2007.02.002

A series of tetrahydroisoquinoline-N-phenylamide derivatives were designed, synthesized, and tested for their relative binding affinities, and antagonistic activities against estrogen receptor (ER). Compound 1f (relative binding affinity, RBA = 5) showed

Full text of DOI:10.1016/j.bmcl.2007.02.002

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2581–2589
Identification of a series of tetrahydroisoquinoline derivatives
as potential therapeutic agents for breast cancer
Hsiang-Ru Lin,a,* Martin K. Safob and Donald J. Abrahamb
aDepartment of Chemistry, College of Science, National Kaohsiung Normal University, Kaohsiung 824, Taiwan, ROC
bSchool of Pharmacy, Department of Medicinal Chemistry and Institute for Structural Biology and Drug Discovery,
Virginia Commonwealth University, Richmond, VA 23298, USA
Received 8 December 2006; revised 31 January 2007; accepted 2 February 2007
Available online 4 February 2007
Abstract—A series of tetrahydroisoquinoline-N-phenylamide derivatives were designed, synthesized, and tested for their relative
binding affinities, and antagonistic activities against estrogen receptor (ER). Compound 1f (relative binding affinity, RBA = 5)
showed higher binding affinity than tamoxifen (RBA=1), a potent ER antagonist and currently being used for breast cancer therapy.
Compound 1f also exerted optimal antagonistic activity against ER in reporter and cell proliferation assays. Interestingly, com-
pound 1j, which only has a minor agonistic effect against ER, acted as a progesterone receptor (PR) antagonist and exerted agonistic
activity against AP-1 through ER pathway. Our results show that these new compounds can be employed as leading pharmacophore
for further development of potent selective ER and/or PR modulators or antagonists.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Breast cancer is the second leading cause of cancer-relat-
ed deaths in women today and is the most common can-
cer among women, excluding non-melanoma skin
cancers. The nuclear receptors, estrogen receptor (ER)
and progesterone receptor (PR) and their associated ste-
roid hormones, are known to play essential roles in the
growth of breast tumors, and their status is also em-
ployed as diagnostic indicators for endocrine respon-
siveness and tumor recurrence.1
domain is a hinge domain, while the E and F domains
are commonly referred to as the AF-2 domain and make
up the second transactivation domain. The AF-2
domain contains the LBD and its function is highly reg-
ulated by compounds that bind to the LBD.6 In addition
to its ligand dependent properties, the AF-2 domain is
also responsible for the interaction between ERs and
coactivators/corepressors.7 Traditionally, ERs exert
their functions mainly via the genomic pathway where
they serve as transcription factors. Initially, ERs are
inactive and located in the cytoplasm.8 Upon E2 bind-
ing, they form an active complex that translocates to
the nucleus, dimerizes, and binds to specific promoter
or enhancer regions. Finally, this complex turns on tran-
scription action as well as regulates the transcription ac-
tions of genes involved in several essential biological
functions, including Bcl-2, cathepsin D, progesterone
receptor, and VEGF.9–12 Apart from the classic ligand
dependent transactivation mechanism, there is a DNA
binding independent transactivation mechanism for
exerting estrogen’s functions. The hER can mediate gene
transcription through the AP-1 enhancer element that
requires hER ligand and the AP-1 transcription factors,
Fos and Jun, for the transcriptional activation.13
Physiologically, estrogen (E2, Fig. 1a) is involved in cel-
lular proliferation and differentiation of organs such as
breast, uterus, ovary, and bone.2 Its action at the cellu-
lar level is exerted through ER. To date, two subtypes of
human ERs have been cloned—hERa and hERb.3,4
Both proteins contain six functional domains, A to F.5
The A and B domains, collectively referred to as the
AF-1 domain, form the transactivation domain for
ERs. The function of the AF-1 domain is regulated by
promoter regions, as well as influenced or activated by
E2 and several antiestrogenic compounds upon binding
to the ligand binding domain (LBD). The ER’s C
domain is the DNA binding domain (DBD). The D
Keywords: Estrogen receptor; Progesterone receptor; Antiestrogen;
Breast cancer; Ligand binding domain; Antagonist; Agonist.
Progesterone (Fig. 1a) also plays a major role in the
development, differentiation, and function of female
reproductive tissues required for pregnancy, and recent
*
Corresponding author. Tel.: +886 7 7172930; fax: +886 7
6051083; e-mail: t3136@nknucc.nknu.edu.tw
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.002

2582
H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
clinical data indicate that progesterone increases breast
cancer risk and may also have a role in established
breast cancer.14 The biological effects of progesterone
are mediated by PR. Like other nuclear receptors, PR
also acts as a transcription factor and contains six func-
tional domains (A–F), similar to those described for ER.
The genes involved in cancer survival mechanism and
regulated by progesterone include Bcl-XL and trans-
forming growth factors (TGF).15,16
Structurally, most non-steroid estrogen receptor modu-
lators, including agonists and antagonists, contain three
components: a core scaffold that mimics the A (phenolic
moiety) and B rings of estrogen; a second aromatic ring
that directly connects with the B ring of the core scaf-
fold; and an antiestrogenic side chain that is substituted
on the B ring of the core scaffold (such as found in these
antiestrogen agents: ICI164, 384; ICI182, 782; and
raloxifene; shown in Fig. 1a) and can interrupt the local-
ization of helix 12 of hER/LBD. In this study, we em-
ploy a tetrahydroisoquinoline–N-phenylamide moiety
(shown in Table 1) as a chemical core scaffold. It con-
tains a carbonyl bridge between its core scaffold and a
second aromatic ring to mimic the steroidal structure
of estrogen. It also contains an antiestrogenic side chain,
similar to that of ICI164, 384, the prototype of ICI182,
780 (Fig. 1a). Importantly, the 3-D structure of this new
derivative can superimpose fairly well with that of the
side chain of the pure antiestrogen, ICII64, 384 as
shown in Figure 1b. The primary goal of this study is
to determine whether this new pharmacophore is capa-
ble of serving as potential antiestrogen. The overall goal
of our study is to generate new pure antiestrogens that
can be more potent than ICI182, 780 and act as pro-
drugs but lack its complicated synthetic routes.
The synthesis of compound 1a–1 l is generated by a sim-
ple amidation reaction to produce the core scaffold, tet-
rahydroisoquinoline-N-phenylamide, which is then
followed by addition of hydrophobic antiestrogenic side
chain as shown in Scheme 1.17
Table 1 summarizes the relative binding affinities for
compounds 1a to 1l based on a fluorescence-based com-
petitive binding assay.18 This assay utilizes a synthetic
estrogenic probe with high fluorescence polarization
property following the binding of a compound to
recombinant hERa. Individual binding affinity is further
calculated as relative binding affinity by using tamoxifen
as a standard. Our studies have identified compound 1f
as the most potent antiestrogen among the tested com-
pounds with relative binding affinity (RBA) five times
greater than that of tamoxifen. The structure–activity
relationship from this study indicates that the hydroxyl
group substituted on the D ring enhanced significant
higher binding affinity compared to other substituents
including halogens, methoxy, and hydrophobic groups.
We note that a hydroxyl group at the 30 position (1f,
RBA = 5) of the second aromatic ring exerts better bind-
ing affinity than at the 40 position (1g, RBA = 1.79).
Also, it appears that a larger size connective bridge
(1l, RBA = 0.267; compared to 1b, RBA = 0.287) and
different moieties at the side chain terminus (1k,
Figure 1. (a) Chemical structures of estrogen, progesterone and
novel antiestrogens. (b): Superimposition of 3-D structures of
11-[2-(3-hydroxy-benzoyl)-1,2,3,4-tetrahydroisoquinolin-3-ylmethoxy]-
undecanoic acid butylamide (yellow stick) and ICII64, 384 (green
stick). These structures and the figures were generated with Sybyl
7.0 and InsightII.

H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
Table 1. Relative binding affinity of compounds 1a–1l for hERa
2583
R1
O
C
2'
3'
4'
(CH2)n
O
N
O
H
N
CH2
C
R2
(CH2)10
Compound
N
R1
R2
Relative binding affinity
1a
1b
0
0
0
0
0
0
0
0
0
0
0
1
20-F
Butyl
Butyl
Butyl
Butyl
Butyl
Butyl
Butyl
Butyl
Butyl
Butyl
Octyl
Butyl
0.225
0.287
0.264
0.192
0.14
30-F
1c
1d
40-F
30-OCH3
40-OCH3
30-OH
40-OH
30-CH3
40-CH3
30-Cl
1e
1f
1g
5.0
1.79
1h
1i
0.156
0.132
0.156
0.177
0.267
1
1j
1k
30-F
30-F
1l
Tamoxifen
ER binding: fluorescence polarization competitive binding assay with recombinant ER protein. ER binding: the polarization values versus test
compound concentration curves were analyzed by the graphfit software to generate IC50 value. The IC50 value was further converted to relative
binding affinity (RBA) by using tamoxifen’s IC50 as a standard. The RBA value of each test molecule was quantified as RBA = IC50 of tamoxifen/
IC50 of test molecule.
RBA = 0.177) do not appear to significantly influence
the binding affinity.
and tamoxifen were tested for their proliferation effects
against MDA-MB-231 cells (ER negative breast cancer
cells). As shown in Figure 5, unlike tamoxifen (relative
cell viability = 0.65 of vehicle effects) that has been
reported to exert significant antiproliferation activity
against MDA-MB-231 cells through ER independent
pathway to induce apoptosis, compound 1f exhibited
no obvious effects against the proliferation of MDA-
MB-231 cells.20 This result proved that compound 1f
can act as a pure antiestrogen and suggests that 1f could
potentially serve as a lead pharmacophore for the devel-
opment of potent pure antiestrogens.
Besides measuring the binding affinities, we tested the
new analogs in transient transfection assays to deter-
mine their agonist/antagonist activity and selectivity
for hERa.19 Figures 2 and 3 outline the agonistic and
antagonistic effects of these compounds, respectively,
targeting hERa. At 2.5 lM, most of the tested com-
pounds, except 1h, 1i, and 1l, did not show any obvious
estrogenic activity. 1f (relative luciferase activity,
RLA = 30% of 1 nM E2 effects) exerted the most
significant antiestrogenic activity among the studied
compounds, but slightly less than that of tamoxifen
(RLA = 20% of 1 nM E2 effects). Compounds 1b, 1d,
and 1h also showed moderate antiestrogenic activity.
Interestingly, 1l with a larger core scaffold exerted minor
estrogenic but no obvious antiestrogenic effects. Similar
results were also observed for another core scaffold, ben-
zofuran-phenylamide, with larger size than that of tetra-
hydroisoquinoline–N-phenylamide developed in our
laboratory.
Based on the new core scaffold as well as the structure–
activity relationship of the derivatives, there seem to be
three essential moieties that could be suitable for design-
ing potent antiestrogens. The first is the size of the core
scaffold. It appears that a core scaffold larger than our
new core scaffold might not serve as antiestrogens but
rather weak estrogens. This is consistent with the fact
that the other different core scaffolds (data not shown
here) with a larger core size than 1f but containing the
same antiestrogenic side chain as compound 1f showed
weak estrogenic but no antiestrogenic effects. Second,
the antiestrogenic side chain is essential for antiestrogen-
ic effects. Crystallographic study of 4-hydroxytamoxifen
bound to hER ligand binding domain shows 4-hydroxy-
tamoxifen’s antiestrogenic side chain bent toward helix 3
to make interaction with Asp351, and as a result the side
chain does not directly interact with helix 12.21 On the
contrary, compound 1f with a significantly longer side
chain than that of 4-hydroxytamoxifen, as suggested
by modeling studies, might protrude out of the ligand
binding pocket and position in such a way as to directly
interact with helix 12. In the modeling studies, we
docked 1f into the hERa LBD/4-hydroxytamoxifen
We also performed a cell proliferation assay using
MCF-7 (ER positive) cells to evaluate the antiprolifera-
tion activities of the compounds, and the result is shown
in Figure 4. At 5 lM, all tested compounds did not en-
hance the growth of MCF-7 cells. In line with the results
observed with transient transfection assays, compound
1f (relative cell viability = 0.48 of vehicle effects) exerted
the most potent antiproliferation activity among the test
compounds, but slightly less than that of tamoxifen (rel-
ative cell viability = 0.43 of vehicle effects).
In order to determine the specificity of pure antiestrogen-
ic effects of compounds 1f, 1b, and 1d, the compounds

2584
H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
Scheme 1. Synthesis of compounds 1a–1l. (a) i—isobutylchloroformate, CH2Cl2. ii—N-butylamine, (or N-octylamine), Et3N; (b) DEC, HOBt,
DMF, RT; (c) NaH, Br(CH2)10CON(CH2CH2CH2CH3)H (or Br(CH2)10CON(n-octyl)H) DMF; (d) BBr3, CH2Cl2.
structure (PDB code: 3ERT). After energy minimiza-
tion, the result shown in Figure 6 suggests that the core
structures of compound 1f and 4-hydroxytamoxifen
could superimpose closely, however as noted above the
mode of binding of the side chain may be different.
The suggested a side chain binding mode in 1f would
prevent helix 12 of the protein from moving toward
the ligand binding pocket, resulting in the compound
acting as a pure antiestrogen. We also point out that un-
like the antiestrogenic side chain of ICI164, 384 which
interacts with the side chain of Trp 290 of rERb LBD,
the longer antiestrogenic side chain of 1f may also pre-
vent such an interaction.22 The antagonistic mode of
binding by GW5638, a tamoxfen derivative in hERa
LBD, would support our hypothetical inference.23 The
main and only structural difference between tamoxifen
and GW5638 is the terminus of their antiestrogenic side
chains. Tamoxifen has a dimethylamino group at the
antiestrogenic side chain terminus, while GW5638 has
a terminus carboxylic acid, which does not interact with
Asp 351, but rather with Leu 536 and Tyr 537 of helix 12
of hERa LBD. This interaction induces capping of helix

H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
2585
Figure 4. Cell viability assay in MCF-7 cells. After cells were
inoculated into 12-well culture plate and attached to the bottom,
5 lM tested compound was added to the respective well and further
incubated with cells for 48 h. The final relative cell viability activity was
quantified as MTT assay value of 5 lM tested compound/MTT assay
value of DMSO. The MTT assay data represent means SD for three
determinations. Tam, tamoxifen.
Figure 2. Transient transfection reporter assay in MCF-7 cells. ER
agonistic effect was determined by the ERE-driven transactivation
luciferase activity. The relative luciferase activity was quantified as
RLA of 2.5 lM tested compound/RLA of DMSO (vehicle). The RLA
data represent means SD for three determinations. Tam, tamoxifen.
Figure 3. Transient transfection reporter assay in MCF-7 cells. ER
antagonistic effect was determined by the ERE-driven transactivation
luciferase activity in the presence of 1 nM estradiol. The relative
luciferase activity was quantified as RLA of 2.5 lM tested compound/
RLA of 1 nM estradiol. The RLA data represent means SD for three
determinations. Tam, tamoxifen.
Figure 5. Cell viability assay in MDA-MB-231 cells. After cells were
inoculated into 12-well culture plate and attached to the bottom, 5 lM
tested compound was added to the respective well and further
incubated with cells for 48 h. The final relative cell viability activity
was quantified as MTT assay value of 5 lM tested compound/MTT
assay value of DMSO. The MTT assay data represent means SD for
three determinations. Tam: tamoxifen.
12 and causes it to shift to an antagonistic conforma-
tion. This antagonistic mode of binding by GW5638 is
consistent with the hypothetic mechanism of the anties-
trogenic side chain of 1f observed in our modeling study.
Further structure–activity relationship study in the mod-
ification of antiestrogenic side chain of 1f is required to
fully demonstrate this hypothesis. Finally, the last moi-
ety that is essential in modulating antiestrogenic activity
is the substituted group on the D ring of the core scaf-
fold. Most novel antiestrogens (such as raloxifene, but
not tamoxifen) contain a 40-hydroxyl group on their D
ring, which acts as an optimal substituent for activity.
The structure–activity relationship study of raloxifene
derivatives shows that the derivative with a 40-hydroxyl
group (raloxifene) exerted two-fold higher binding affin-
ity than the derivative with 30-hydroxyl group. On the
other hand, both derivatives showed 10-fold higher
binding affinity than the derivative with 20-hydroxyl sub-
stituent.24 In MCF-7 cell proliferation assays, raloxifene

2586
H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
Figure 7. Transient transfection reporter assay in T47D cells. hERa/
AP1 agonistic effect was determined by the AP1 promoter-driven
transactivation luciferase activity. The relative luciferase activity was
quantified as RLA of 5 lM tested compound/RLA of DMSO (vehicle).
The RLA data represent means SD for three determinations.
Figure 6. Hypothetical model of compound 1f (in green color;
4-hydroxytamoxifen in red color; helix 12 of hERa LBD in white
color) located in hERa LBD. The 3-D structure of compound 1f was
energy-minimized and generated using Sybyl 7.0 program. The
minimized structure was further docked into hERa LBD crystal
structure (3ERT, protein data bank) and superimposed with
4-hydroxytamoxifen using Insight II program to generate the model.
transfection reporter assays to investigate whether any
of our compounds can act as an estrogenic molecule
to enhance AP-1 transcription actions through hERa.
As shown in Figure 7, ICI 182, 780 at 5 lM significantly
enhance AP-1 driven transcription effects as previously
reported.28 Compound 1j also exerted a slightly less ef-
fect at the same concentration as ICI 182, 780. Impor-
tantly, although 1j does not contain any hydroxyl
group on its A ring like ICI 182, 780, it still can activate
AP-1 dependent reporter action. This result suggests
that the structural requirement of agonists in AP-1
dependent transcription in the presence of hERa/AP-1
pathway is different from the classical antiestrogenic
structures. Especially, the position of hERa helix 12 in
this mechanism could be different from that in the
classical mechanism. It will be intriguing to study the
structure–activity relationship of ER modulators against
AP-1 actions.
showed 16-fold better activity than its derivative with
30-hydroxyl group. Addition of a 40-hydroxyl group on
the D ring of tamoxifen enhanced its higher binding
affinity by six-fold.25 On the other hand, the addition
of a 30-hydroxyl group decreased the binding affinity
by one-third. Nevertheless, and unlike raloxifene deriva-
tives, addition of a hydroxyl group to either the 30 or 40
position of tamoxifen’s D ring did not significantly in-
crease its in vivo antiestrogenic effects. For our new
scaffold, a 30 or 40-hydroxyl substituent on the D ring
also enhanced its binding affinity, with the 30 hydroxyl
substituent increasing binding affinity to five-fold that
of tamoxifen. Interestingly, we observed an increase of
antagonistic effects in the reporter assay. Wilhelm Stark
et al. at Norvatis have reported a new class of antiestro-
gens with N-phenyl-tetrahydroisoquinoline as the core
scaffold.26,27 In their study, the 30 and 40-hydroxyl deri-
vatives exerted similar binding affinity and cellular
antiestrogenic effects. The above studies and results sug-
gest substitution of an optimal substituent on the D ring
varies with different core scaffold structures. In conclu-
sion, the size of the core scaffold, the substitution on
the D ring, and the antiestrogenic side chain play essen-
tial roles in the observed and future design of potent
antiestrogens.
In addition to ER, PR is also an essential diagnostic fac-
tor for breast cancer status, and its antagonists are
known to show significant antiproliferation effects
against ER+/PR+ mediated breast cancers. Even
though PR antagonists have different structural require-
ment from ER antagonists, ICI182, 780 has been report-
ed to contain antiprogestin activity. The antiprogestin
activity of ICI182, 780 might be a contributing factor
for its significant anti-breast cancer effects.28 Since our
compounds can structurally superimpose with ICI182,
780, we also evaluated compound 1j against hPR by
employing transient transfection reporter assays in
T47D cells (ER positive/PR positive breast cancer cells).
As shown in Figure 8, at 5 lM, 1j (RLA = 54% of 1 nM
progesterone effects) exerted similar antiprogestin activ-
ity as ICI182, 780 (RLA = 48% of 1 nM progesterone ef-
fects). Compound 1j did exert moderate antiprogestin
effects in PRB/MMTV dependent transcription actions
although it has no obvious effects on ER. ER LBD
and PR LBD have quite different structural require-
Although most antiestrogens can antagonize the actions
of hER in many human organs, they are also found to
exert estrogen-like actions in some tissues via hER.
The AP-1 mechanism has been hypothesized to be the
main pathway for estrogen-like actions of antiestrogens.
Thus, we used AP-1 luciferase reporter in transient

H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
2587
ough to run an NMR to identify these compounds.
More study will be carried out in our laboratory to
determine the possibility of our prodrug hypothesis.
The results discussed above indicate that this new phar-
macophore is capable of being further developed as po-
tent pure antiestrogens or antiprogestins.
Acknowledgments
This work was supported by NIH/NCI Grant, 1 RO1
CA100121-01A1. We generously thank Dr. B. S.
Katzenellenbogen at U. of Illinois, Urbana, for offering
us the pCMV5hER vector, Dr. D. P. McDonnell at
Duke U., for offering Dr. Lin H.R. the 3x-ERE-
TATA-Luc vector, Dr. S. Y. Tsai at Baylor College of
Medicine, for offering Dr. Lin H.R. the hPR expression
vector, and Dr. E. P. Gelmann, for offering Dr. Lin
H.R. the pMMTV-luc vector.
Figure 8. Transient transfection reporter assay in T47D cells. PR
antagonistic effect was determined by the MMTV promoter-driven
transactivation luciferase activity in the presence of 1 nM progester-
one. The relative luciferase activity was quantified as RLA of 5 lM
tested compound/RLA of 1 nM progesterone. The RLA data represent
means SD for three determinations.
References and notes
1. Osborne, C. K.; Schiff, R.; Arpino, G.; Lee, A. S.;
Hilsenbeck, V. G. Breast 2005, 14, 458.
2. Macgregor, J. I.; Jordan, V. C. Pharmacol. Rev. 1998, 50,
151.
3. Greene, G. L.; Gilna, P.; Waterfield, M.; Baker, A.; Hort,
Y. Shine J. Science 1986, 231, 1150.
4. Mosselman, S.; Polman, J.; Dijkema, R. FEBS Lett. 1996,
392, 49.
5. Kumar, V.; Green, S.; Stack, G.; Berry, M.; Jin, J. R.;
Chambon, P. Cell 1987, 51, 941.
6. Kuiper, G. G.; Carlsson, B.; Grandien, K. Endocrinology
1997, 138, 863.
7. McKenna, N. J.; Lanz, R. B.; O’Malley, B. W. Endocr.
Rev. 1999, 20, 321.
8. Katzenellenbogen, B. S.; Choi, I.; Delage-Mourroux, R.;
Ediger, T. R.; Martini, P. G. V.; Montano, M.; Sun, J.;
Weis, K.; Katzenellenbogen, J. A. J. Steroid Biochem.
Mol. Biol. 2000, 74, 279.
9. Perillo, B.; Sasso, A.; Abbondanza, C.; Palumbo, G. Mol.
Cell Biol. 2000, 20, 2890.
ments for their ligands due to structural differences in
their ligand binding domains. In the crystal structure
of PR LBD/progesterone complex, the carbonyl group
on the 3 position of progesterone makes hydrogen-bond
contact with Gln 725 and Arg 766.29 Although, the car-
bonyl group on the 20 position of progesterone is in
proximity to Thr 894 and Cys 891, it does not form
hydrogen-bond interactions with any of the residues.
Nevertheless, structure–activity relationship studies for
progesterone derivatives suggest that the carbonyl group
at the 20 position could make contact with PR LBD.30
Like ICI182, 780, compound 1j showed no agonistic
but antagonistic effects against PR. Although both com-
pounds do not contain a carbonyl group on the A ring,
like progesterone, it is possible their long hydrophobic
side chains could be responsible for their antiprogestin
effects by directly interfering with helix 12. Unlike ICI
182, 780 that involves difficult synthetic routes, 1j could
easily be derivatized to generate potent antiprogestins.
10. Augerean, P.; Miralles, F.; Cavailles, V.; Gaudelet, C.;
Parker, M.; Rochefort, H.; Wodak, S. J. Mol. Endocrinol.
1994, 8, 693.
11. Petz, L. N.; Nardulli, A. M. Mol. Endocrinol. 2000, 14, 972.
12. Mueller, M. D.; Vigne, J. L.; Minchenko, A.; Lebovic, D.
I.; Leitman, D. C.; Taylor, R. N. Proc. Natl. Acad. Sci.
USA 2000, 97, 10972.
13. Paech, K.;Webb, P.;Kuiper, G. G.;Nilsson, S.;Gustafsson,
J.; Kushner, P. J.; Scanlan, T. S. Science 1997, 277, 1508.
14. Moore, M. R. Curr. Cancer Drug Targets 2004, 4, 183.
15. Moore, M. R.; Conover, J. L.; Franks, K. M. Biochem.
Biophys. Res. Commun. 2000, 277, 650.
16. Gong, Y.; Anzai, Y.; Murphy, L. C.; Ballejo, G.; Holinka,
C. F.; Gurpide, E.; Murphy, L. J. Cancer Res. 1991, 51,
5476.
17. Synthesis of the hydrophobic antiestrogenic side chain A1
and A2
In this study, we have identified a new type of pharma-
cophore and lead compound, 1f, as a pure antiestrogen.
Compound 1f showed higher binding affinity than
tamoxifen, and exerted no estrogenic but significant
antiestrogenic effects in transient transfection reporter
assays against hERa. It also selectively inhibited the
proliferation of ER+ but not ERꢀ breast cancer cells.
Additionally, 1f was also found to interfere with the
dimerization of hERa in yeast two hybrid assays (data
not shown). Apart from 1f, 1j exhibited selective action
against PR and this discovery might lead to a new series
of selective antiprogestins. Finally, we evaluated if these
compounds can be employed as prodrugs by incubating
compound 1k with CYP2D6, a P450 enzyme that has
been reported to transform tamoxifen to 4-hydroxytam-
oxifen.31 HPLC analysis of the reaction products re-
vealed a mixture of new products different from 1k.
However, the amount of mixture generated was not en-
Synthesis of N-butyl-11-bromoundecanamide (A1) and
N-octyl-11-bromoundecanamide (A2). 11-Bromoundeca-
noic acid (12 g, 45 mmol) or 11-bromoundecanoic acid
(12 g, 45 mmol) was dissolved in dry CH2Cl2 (200 ml) and
tributylamine (10 g, 54 mmol). After cooling the mixture
to –10 ꢁC, isobutylchloroformate (8.2 g, 60 mmol) was
added and stirred for 2 h. Excess N-butylamine (4.43 g,

2588
H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
60.8 mmol) was then added, and later the cooling bath was
19. Cell culture, transient transfection reporter, and cell
viability assay: Human breast cancer cells, MCF-7,
MDA-MB-231, and T47D, were purchased from ATCC
(Manassas, VA). The cells routinely were cultured as
monolayer in Dulbecco’s modified minimal essential
medium for MCF-7 and MDA-MB-231 cells, and RPMI
1640 (Gibco/BRL, Grand Island, NY) for T47D cells
supplemented with 10% fetal bovine serum (Hyclone,
Logan, UT), Penicillin (100 U/ml)/streptomycin (100 lg/
ml) (GIBCO/BRL, Grand Island, NY), and incubated at
37 ꢁC in a humidified atmosphere of 5% CO2/air. For the
transient transfection reporter assays, the MCF-7 or T47D
cells were plated in triplicate in 12-well plates at a density
of 3 · 105 cells/well in the phenol red-free DMEM or
RPMI 1640 supplemented with 10% charcoal-stripped
fetal bovine serum (Hyclone, Logan, UT), Penicillin
(100 U/ml)/streptomycin (100 lg/ml). Twenty-four hours
later, the cells were transfected with three plasmids by
using the Superfect transfection kit (Qiagen, Valencia,
CA). For the detection of wild type hERa activity, cells
were transfected with 2 lg wild type hERa expression
plasmid (pCMV-hERa), 6 lg luciferase reporter plasmid
containing estrogen receptor response element (3x-ERE-
TATA-Luc) or luciferase reporter plasmid containing AP-1
response element (AP1-Luc, Clontech, Palo Alto, CA) ,
and 1 lg normalization control, b-galactosidase reporter
plasmid (pCMV-b, Clontech, Palo Alto, CA). For the
activity against hPR, cells were transfected with 2 lg wild
type hPR expression plasmid (pCMV-hPR), 6 lg lucifer-
ase reporter plasmid containing progesterone receptor
response element (pMMTV-Luc), and 1 lg normalization
control, b-galactosidase reporter plasmid (pCMV-b).
After transfection, the cells were treated with test com-
pounds and one positive control (vehicle, DMSO) in
phenol red-free culture medium. After incubation for
further 24 h, the cells were washed with PBS and lysed
with lysis buffer (Pierce, Rockford, IL). The lysate was
used to determine the luciferase activity for ER’s activity
and the b-galactosidase activity for the normalization of
transfection efficiency. For the luciferase activity assay,
20 ll of lysate and 100 ll luciferase assay buffer (Promega,
Madison, WI) were added into a well of 96-well plate. The
luminescence was detected by using luminescence micro-
plate reader, LumiCount (Packard, Boston, MA). For the
b-galactosidase activity assay, 20 ll of lysate and 180 ll
b-galactosidase assay buffer (Clontech, Palo Alto, CA)
were added into a well of 96-well plate. The b-galactosi-
dase activity was measured as luminescence strength by
using LumiCount (Packard, Boston, MA). The normal-
ized luciferase activity was calculated by the equation,
normalized luciferase activity =luciferase activity/b-galac-
tosidase activity. For the estrogenic, antiestrogenic or
antiprogestin effects of test molecules, the normalized
luciferase activity value was further converted to relative
normalized luciferase activity by using vehicle (DMSO)
for estrogenic effects, 1 nM E2’s value for antiestrogenic
effects or 1 nM progesterone’s value for antiprogestin
effects as a standard that was set to 1. In cell viability
assays, MCF-7 or MDA-MB-231 cells were inoculated
into 12-well culture plates at 2 · 105 cells in 2 ml
maintained medium per well. Cells were allowed to attach
to the bottom for 24 h incubation, then the seeding
medium was removed and replaced by the experimental
medium (phenol red-free DMEM supplemented with 5%
charcoal-stripped fetal bovine serum and penicillin (100 U/
ml)/Streptomycin (100 lg/ml). After further 24 h incuba-
tion, the test molecules dissolved in DMSO were added in
the wells. The final concentration of DMSO in the culture
medium did not exceed 0.1%. The culture was continued
removed. After 3 h, CH2Cl2 was added and the organic
phase was washed with 1 N HCl, saturated NaHCO3, and
water. After drying with MgSO4, the solvent was removed
and the crude product purified by flash column chroma-
tography to generate compounds A1 and A2.
Synthesis of compounds 1a–1l. Benzoic acid (7 mmol) and
(s)-(ꢀ)-1,2,3,4-tetrahydro-3-isoquinoline methanol (1 g,
6.7 mmol) were dissolved in DMF (30 mL) under nitrogen
gas. 1-Hydroxybenzotriazole hydrate (HOBT) (1 g,
7.4 mmol) and 1-(3-(dimethylamino) propyl)-3-ethyl-car-
bodiimide) hydrochloride (DEC) (1.5 g, 7.8 mmol) were
added to the solution and the reaction mixture stirred
overnight. EtOAc was added to the reaction mixture and
subsequently washed with 10% KHSO4 (30 mL), saturated
NaHCO3(30 mL), and brine (30 mL). After drying with
MgSO4, the solvent was removed and the crude product
was purified by flash column chromatography. The
purified compound was dissolved in DMF and mixed
with sodium hydride (0.17 g, 7 mmol) and N-butyl-11-
bromoundecanamide (A1, 3.2 g, 10 mmol) or N-octyl-11-
bromoundecanamide (A2, 3.76 g, 10 mmol). The reaction
mixture was refluxed under nitrogen overnight. EtOAc
(100 mL) was added to the reaction mixture and the
solution was subsequently washed with water (30 mL) and
brine (30 mL). After drying with MgSO4, the solvent was
removed and the crude product was further purified by
flash column chromatography to generate the final pure
product.
Synthesis of compounds 1f and 1g. Compound 1f. A
solution of compound 1d (0.54 g, 1 mmol) in dry CH2Cl2
(50 mL) was cooled to ꢀ60 ꢁC under a nitrogen atmo-
sphere. Following, BBr3(1 mL, 10 mmol) was added. After
30 min, the cooling bath was removed and the mixture
stirred overnight. After cooling, the mixture was poured
into an aqueous solution of NaHCO3.The organic layer
was separated, and the aqueous layer was extracted with
EtOAc (50 · 3 ml). The combined organic layers were
washed with water and dried with MgSO4. After evapo-
ration of the solvent, the remaining residue was purified by
flash column chromatography (SiO2, hexane/EtOAc) to
generate compound 1f.
Compound 1 g. Using compound 1e (0.54 g, 1 mmol) in
dry CH2Cl2 (50 mL) and then BBr3(1 mL, 10 mmol),
compound 1g was synthesized following the above proce-
dure for compound 1f.
18. Fluorescence polarization competitive binding assay: the
estrogen receptor competitor assay kit (Panvera, Madison,
WI) was used to determine the ability of test molecules to
displace the synthetic estrogenic probe, ES2, with high
fluorescence polarization property when it binds to hERa,
from hERa-ES2 complex. Serial dilutions of each test
molecule were prepared in DMSO. The recombinant
hERa (7 nM) was preincubated with ES2 (1 nM) in
screening buffer. After preincubation, the test molecules
and ER/ES2 complex solution were added into a 96-well
microplate to produce a final volume of 100 ll per well.
The reaction mixture was incubated at room temperature
for 1 hr and the polarization values were measured by
using fluorescence micro plate reader, Polarion (Tecan,
Research Triangle Park, NC) with excitation wavelength
495 nm and emission wavelength 535 nm. The polarization
values versus test compound concentration curves were
analyzed by the graphfit software to generate IC50 value.
The IC50 value was further converted to relative binding
affinity (RBA) by using tamoxifen’s IC50 as a standard
that was set to 1. The RBA value of each test molecule was
calculated by using the equation, RBA =IC50 of tamox-
ifen/IC50 of test compound.

H.-R. Lin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2581–2589
2589
for 2 days. The final relative cell viability was estimated
with Cell Titer proliferation assay kit (Promega, Madison,
WI) by measuring the absorbance at 570 nm, which is
directly proportional to the number of living cells in the
culture. The testing was performed by following the
manufacturer’s protocol. Experiments were triplicated
for each compound.
L. L.; Glasebrook, A. L.; Bryant, H. U. J. Med. Chem.
1997, 40, 146.
25. Jordan, V. C.; Lieberman, M. E.; Cormier, E.; Koch, R.;
Bagley, J. R.; Ruenitz, P. C. Mol. Pharm. 1984, 26, 272.
26. Renaud, J.; Bischoff, S. F.; Buhl, T.; Floerscheim, P.;
Fournier, B.; Halleux, C.; Kallen, J.; Keller, H.; Schlaeppi,
J.-M.; Stark, W. J. Med. Chem. 2003, 46, 2945.
27. Renaud, J.; Bischoff, S. F.; Buhl, T.; Floersheim, P.;
Fournier, B.; Geiser, M.; Halleux, C.; Kallen, J.; Keller,
H.; Ramage, P. J. Med. Chem. 2005, 48, 364.
28. Nawaz, Z.; Stancel, G. M.; Hyder, S. M. Cancer Res.
1999, 59, 372.
29. Williams, S. P.; Sigler, P. B. Nature 1998, 393, 392.
30. Hillisch, A.; Von Langen, J.; Menzenbach, B.; Droescher,
P.; Kauffman, G.; Schneider, B.; Elger, W. Steroids 2003,
68, 869.
31. Coller, J. K.; Krebsfaenger, N.; Klein, K.; Endrizzi, K.;
Wolbold, R.; Lang, T.; Nussler, A.; Neuhaus, P.; Zanger,
U. M.; Eichelbaum, M.; Murdter, T. E. Br. J. Clin.
Pharmacol. 2002, 54, 157.
20. Perry, R. R.; Kang, Y.; Greaves, B. Ann. Surg. Oncol.
1995, 2, 238.
21. Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner,
P. J.; Agard, D. A.; Greene, G. L. Cell 1998, 95, 927.
22. Pike, A. C. W.; Brzozowski, A. M.; Walton, J.; Hubbard,
R. E.; Thorsell, A. G.; Li, Y. L.; Gustafsson, J. A.;
Carlquist, M. Structure 2001, 9, 145.
23. Wu, Y. L.; Yang, X.; Ren, Z.; McDonnell, D. P.; Norris, J.
D.; Willson, T. M.; Greene, G. L. Mol. Cell 2005, 18, 413.
24. Grese, T.; Cho, S.; Finley, D. R.; Godfrey, A. G.; Jones,
C. D.; Lugar, C. W., III; Martin, M. J.; Matsumoto, K.;
Pennington, L. D.; Winter, M. A.; Adrian, M. D.; Cole,
H. W.; Magee, D. E.; Phillips, D. L.; Rowley, E. R.; Short,