6588
P. T. Weiser et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6587–6590
OH
R
N
R
NH
OH
O
O
Cl
B
O
O
b
R
R
OH
OH
R
R
f, g
d
a
e
4b-e
Br
c
Br
3b-e
R
O
R
O
R
2b-e
O
O
O
O
O
O
R
O
O
O
OH
a
R = H
b
c
d
e
R = iPr
R = sBu
R = tBu
R = Bn
6a-ea
5b-e
7a-e
1a-e
Figure 1. Reagents and conditions: (a) tetrabutylammoniumtribromide, CHCl3, rt, 100%; (b) PdCl2(dppf), B2pin2, dioxane, 80 °C, 95%; (c) 2-Etoxycarbonylethyl bromide, NaH,
THF, rt, 92%; (d) PdCl = (PPh3)2, PPh3, Na2CO3, THF/H20 (4:1), reflux, 65%; (e) 2-chloro-N,N-dimethlethlamine hydrochloride, K2CO3, acetone, reflux, 60%; (f) 1 N NaOH, EtOH or
MeOH, 40 °C; (g) 4 N HCl, EtOAc or dioxane, rt, 85%; a 6a was analogously prepared using commercially available 4-iodophenol and 4-hydroxyphenylboronic acid as starting
material.
nyl proteomimetics (Fig. 1) as novel ER
a
coactivator binding
To complement our previously reported preliminary biological
analysis of the amino-acid series (1a–e), we subjected the same
series to a mammalian two-hybrid competitive binding assay
employing the estrogen receptor (VP16-ERa) and the co-activator
peptide pM-GRIP1 LxxLL2 (Table 1). This assay was used to assess
inhibitors.12 A combinatorial approach was used to allow for syn-
thesis of a large, diverse library from relatively few starting mate-
rials. Commercially available 2-substituted phenols (2b–e),
substituted with isopropyl, sec-butyl, tert-butyl and benzyl, were
selectively brominated using tetrabutylammonium tribromide to
yield the substituted 4-bromophenols (3b–e) in excellent yields.
These important intermediates were then converted to (4b–e)
and (5b–e) via Miyaura and Williamson ether conditions respec-
tively. The biphenyl phenolic-ester series (6a–e) was synthesized
from the coupling of identically substituted 4 bromobenzene and
5 arylboronic ester under Suzuki conditions, utilizing PdCl2(PPh3)2
as a catalyst. A tertiary amine functionality was subsequently
introduced via Williamson ether synthesis of the biphenyl pheno-
lic-esters with N,N-dimethylaminoethylchloride to yield biphenyl
amino-esters 7a–e. Finally, the C-terminal functionality was var-
ied by hydrolyzing the ethyl ester to the corresponding carboxylic
acid and the resultant biphenyl amino-acid was isolated as the
hydrochloride salt (1a–e). Initial biological studies probed the ef-
whether the test compounds interrupt the interaction between
ER
VP16-ER
a
and the LXXLL2 peptide derived from the coactivator GRIP1.
and pM-GRIP1-LXXLL2 were transfected into HepG2
a
cells together with the reporter gene 5xGal4-Luc3 and a normaliza-
tion control pCMV- b -gal as described.13 Cells were then treated
with serial twofold dilutions of test compounds +/ꢀ 1 nM estradiol
and incubated for 40 h before assaying. As these results illustrate,
structural changes on the biphenyl core significantly affect the
activity of the compounds. 1c and 1e, which have the largest sub-
stituents, inhibit the E2-induced association of the peptide with
the receptor, which is consistent with our previously reported
analysis which identified 1c as the most effective inhibitor of
ERa-coactivator interactions, as assessed by both TR-FRET and re-
porter gene assays. Additionally, 1d illustrates agonist activity in
the absence of E2 and inhibitory activity in the presence of E2, indi-
cating that this compound is likely binding inside of the LBD. The
unsubstituted biphenyl amino-ester 1a not only did not inhibit
peptide binding in the presence of E2, but also was observed to
promote association in the absence of E2. This is consistent with
our previous finding that 1a possesses low but observable affinity
fect of these compounds on antagonizing the ER
interaction and demonstrated that some of the final amino-car-
boxy products 1a–e were significant inhibitors at the ER -CBD.
Compound 1c emerged as a lead compound from this initial anal-
ysis, demonstrating low micromolar affinity for ER in TR-FRET
binding assays and the capability to disrupt NR-coactivator bind-
ing interactions when evaluated in ER reporter gene assays in
a-coactivator
a
a
a
for the ERa-LBD. Additionally, compound 1b, where R = isopropyl,
Hec-1 cells. This evaluation clearly demonstrated that the identity
of hydrophobic substituents plays an integral role in influencing
binding affinity, however, further analysis remained necessary
to investigate the significance of the terminal functionalities.
In this study we evaluated the significance of the terminal
groups in competitively inhibiting coactivator binding to the
has little effect on either association or inhibition. The most effec-
tive compounds in the competitive binding assays were the pheno-
lic-esters 6b and 6c, where R = isopropyl and sec-butyl,
respectively. The amino-ester series (7a–e) had little to no effect
on binding.
A subsequent cellular assay was employed to evaluate the
in vitro activity of the amino-acids to inhibit gene transcription
in MCF-7 (ER+) human breast cancer cells (Table 2). MCF-7 cells
ERa-CBD and the AR-CBD. These polar termini were designed to
interact with polar residues in the CBD, mimicking the charge
clamp typically observed between the CBD and backbone residues
surrounding the NR box. Analyzing the biological effect of trun-
cating the amino terminus or esterifying the carboxyl terminus
would provide a basis for determining their contribution to affin-
ity and selectivity. For this reason we evaluated the phenolic-es-
ters 6a–e and the amino-esters 7a–e in addition to the final
amino-acids 1a–e.
were pre-treated with 20 lM solutions of each compound for 2 h,
followed by exposure to either vehicle or estradiol for 14 h. RNA
was harvested and quantitative-PCR was used to determine the ef-
fects of these compounds on the expression of the known ER
a tar-
get genes SDF1, PR, and PS2. IDH3A, which is not an ER target
a
gene, was used as a negative control to demonstrate the selectivity
of these compounds.
Weiser, Patrick T.
