Design and evaluation of fragment-like estrogen receptor tetrahydroisoquinoline ligands from a scaffold-detection approach

Article abstract of DOI:10.1021/jm1011116

A library of small tetrahydroisoquinoline ligands, previously identified via structure- and chemistry-based hierarchical organization of library scaffolds in tree-like arrangements, has been generated as novel estrogen receptor agonistic fragments via traditional medicinal chemistry exploration. The approach described has allowed for the rapid evaluation of a structure-activity relationship of the ligands concerning estrogen receptor affinity and estrogen receptor - subtype selectivity. The structural biological insights obtained from the fragments aid the understanding of larger analogues and constitute attractive starting points for further optimization.

Full text of DOI:10.1021/jm1011116

ARTICLE
pubs.acs.org/jmc
Design and Evaluation of Fragment-Like Estrogen Receptor
Tetrahydroisoquinoline Ligands from a Scaffold-Detection Approach†
Sabine M€ocklinghoff,‡,§ Willem A. L. van Otterlo,||,^ Rolf Rose,§ Sascha Fuchs,‡ Tobias J. Zimmermann,||
Marta Dominguez Seoane,§ Herbert Waldmann,|| Christian Ottmann,§ and Luc Brunsveld*,‡,§
‡Laboratory of Chemical Biology, Department of Biomedical Engineering, Technische Universiteit Eindhoven, Den Dolech 2, 5612AZ
Eindhoven, The Netherlands
§Chemical Genomics Centre of the Max-Planck Society, Otto-Hahn-Strasse 15, 44227 Dortmund, Germany
Max-Planck Institute of Molecular Physiology, Department of Chemical Biology, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany
^Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, PO Wits, 2050 Johannesburg, South Africa
S Supporting Information
b
ABSTRACT: A library of small tetrahydroisoquinoline ligands,
previously identified via structure- and chemistry-based hierarchi-
cal organization of library scaffolds in tree-like arrangements, has
been generated as novel estrogen receptor agonistic fragments
via traditional medicinal chemistry exploration. The approach
described has allowed for the rapid evaluation of a structure-
activity relationship of the ligands concerning estrogen receptor
affinity and estrogen receptor β subtype selectivity. The structural
biological insights obtained from the fragments aid the under-
standing of larger analogues and constitute attractive starting
points for further optimization.
’ INTRODUCTION
for estrogen receptor β (ERβ) subtype binding by this class of
scaffolds is identified, thus providing clear starting points for
optimization programs for subtype selective ligand optimization.
The estrogen receptor (ER) belongs to the superfamily of nuclear
receptors.6,7 The primary endogenous ligand for ER is 17β-estradiol
(E2).7 Two subtypes of ER are known (ERR8 and ERβ9), each with
their own unique tissue distribution patterns and transcriptional
properties. While ERR is mainly involved in reproduction events in
the uterus and mammary glands,10 ERβ is more generally expressed
and is not the dominant receptor in uterus and breast tissues. The
challenge of identifying selective ERβ modulators provides an
opportunity to elucidate the exact physiological function of ERβ
and to develop novel, tissue- and cell-selective drug candidates such
as those related to inflammatory diseases.11,12 Fragment-based drug
discovery promises to be a very valuable tool to generate both such
compounds for nuclear receptors13 and help in the elucidation of the
molecular pharmacological mechanism of larger, more potent
compounds.
Drug discovery ideally starts from bioactive molecules with
scaffold structures that allow easy, and structure-based molecular
optimization, thus providing rapid access to a clear structure-
activity relationship (SAR). Fragment-based screening and design
approaches are highly attractive in this respect.1 Small molecules
with moderate affinity allow rapid SAR determination, entries into
molecular diversity and complexity, and frequently, easy access to
structural information. Structure- and chemistry-based hierarchical
organization of library scaffolds in tree-like arrangements provides a
novel and accelerated entry into the identification of such novel
small ligands or fragments of biologically prevalidated relevance.2-5
Brachiation of scaffold trees from complex to simpler, yet still
similar, structures can also be used to generate new and simpler
analogues of known active molecules as new starting points for
molecular diversity and to elucidate the binding mechanism of
more complex analogues. We have shown that this approach allows
for the identification of a new active fragment-like ligand previously
not annotated with estrogen receptor R (ERR) activity but based
on larger more complex scaffolds.4 Here we now show that a small
focused library based on the identified fragment allows for rapid
generation of a clear SAR and for optimization of the binding
affinity. With the aid of structural biology, the modest preference
Using the tree-like arrangement of library scaffolds, we pre-
viously identified the tetrahydroisoquinoline (THIQ) scaffold as a
Received: July 6, 2010
Published: March 07, 2011
r
2011 American Chemical Society
2005
dx.doi.org/10.1021/jm1011116 J. Med. Chem. 2011, 54, 2005–2011
|

Journal of Medicinal Chemistry
ARTICLE
Table 1. ERr and ERβ Activities (EC50) of THIQ Fragments
in a Fluorescence Polarization Cofactor Recruitment Assay
enhancing ER affinity and addressing ER selectivity. A variety of
functional amide groups were examined for this position, includ-
ing electron withdrawing, aromatic and polar N-substituents.
The synthesis of the fragments was based on the in situ
protection of phenols 2 and 3 (Table 1), followed by reaction
with the appropriate anhydride or sulfonyl chloride. The inter-
mediates thus obtained were subsequently desilylated in the
same flask to yield the desired N-substituted THIQ derivatives
5-1917 (Table 1) in varying yields. Finally, 2-(trifluoroacetyl)-
tetrahydro-6-isoquinolinol 8 was methylated with methyl iodide
in the presence of potassium carbonate to afford the methoxy
THIQ derivative 9 in good yield.
’ BIOCHEMICAL EVALUATION
The THIQ fragments were next evaluated in a biochemical
fluorescence polarization peptide recruitment assay (Table 1). In
this assay, a fluorescein-labeled peptide probe binds to ER ligand
binding domain (LBD), in the ligand-bound state, via an LXXLL
recognition motif. Binding of a compound with an agonistic
profile results in the formation of a complex between the ER LBD
and the fluorescent peptide and a subsequent increase in
polarization. By contrast, a ligand with antagonistic properties
would repress the intrinsic binding of the peptide to the ER LBD,
resulting in a decrease in the observed polarization. The calcu-
lated concentrations for half-maximum response (EC50) of the
substituted THIQ derivatives are presented in Table 1.
An initial comparison of the biochemical activity data revealed
that 11 out of the 18 THIQ fragments showed activity for both
ER LBDs and that all 11 induced an increase in LXXLL peptide
binding to the ER LBDs. Results for THIQs bearing hydroxyl
group, at either the 6- (2) or the 7-position (3), showed that an
hydroxyl group alone is not sufficient for ER binding. Further-
more, the introduction of both OH-groups simultaneously did
not facilitate ER recognition (4). To mask the basic amino
functionality and to address the hydrophobic ligand binding
pocket, various N-substituents were introduced. Modification to
the methyl acetamide (5) did not lead to any measurable ER
affinity, even in combination with a 6-hydroxyl group, as in
compound 6. However, N-substitution with electron withdraw-
ing groups resulted in good affinities for both ER isoforms.
Whereas N-substitution in the absence of hydroxyl functionality
did not facilitate ER affinity, as shown for 7, the concomitant
incorporation of a 6-hydroxyl group with an N-trifluoroaceta-
mide group, as in 8, produced a good affinity for the ER LBDs.
The importance of the phenol functionality in terms of hydrogen
bonding was elucidated by masking it as a methoxy group to
afford 9, a ligand with a significantly lower activity. In addition,
relocation of the 6-hydroxy group to the 7-position (10) resulted
in a 4-fold loss in ER affinity. For gain in ER affinity and the
tendency toward ERβ (E2 showed a slight preference of ERR in
this assay), this limited set of THIQ fragments thus already
showed the importance of a hydroxyl group at the 6-position in
combination with an appropriate nitrogen substituent.
a values are means of a least three experiments. n.a.: no activity. n.t.: not
tested.
simple two-ring core chemical fragment for which little ER activity
was known.4 More highly decorated THIQ derivatives, with three14
or more15,16 ring scaffolds, have been previously described as
(antagonistic) ER modulators with selectivity for ERR. In addition,
a small number of THIQ analogues, typically with smaller side
chains, have also been reported to be agonistic and selective for
ERβ.14,15 Here, we describe the development and structural
support of a focused library of even smaller two-ring THIQ
analogues with ERβ selectivity. Using these fragments, molecular
insights into the action and receptor selectivity of the THIQ ligands
could be obtained, offering new starting points for the under-
standing and generation of selective ER modulators.
’ DESIGN AND SYNTHESIS
On the basis of the simple THIQ scaffold (1) (Table 1), a
small focused library of 18 THIQ fragments was designed and
investigated for their affinities toward both ER receptors. Com-
parison of the structure of THIQ, and its published
derivatives,14-16 with the structure of E2 indicates that the
bicyclic ring system typically mimics the A- and B-rings of E2.
To mimic the phenolic group on the E2 A-ring, known to be
important for high ER affinity, a hydroxyl group was introduced
at different positions on the aromatic ring of the designed THIQ
fragments. Small side chain groups were then added via N-
substitution of the THIQ scaffold to partially fill the space
normally occupied by the C- and D-rings of E2. These side
chains were expected to provide the greatest opportunity for
On the basis of the inactivity of derivatives presenting the
methyl acetamide group and the strong activity of the fluoro-
methyl acetamide analogue, the importance of the amide group’s
size and electronegativity was evaluated. Replacement of one or
all three fluoro groups in 8 by chloro groups was well tolerated
(11 and 12) with only a small decrease in affinity. An additional
set of compounds, bearing a sulfonamide in place of the
acetamide group as N-substituent, was also investigated. In line
2006
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011

Journal of Medicinal Chemistry
ARTICLE
Next, the transcriptional activity of fragment 19 on ERR and
ERβ in mammalian cells was profiled using a cellular estrogen
response element-luciferase reporter gene assay in U2OS cells.
E2 caused a 33-fold increase in luciferase activity in ERR-
transfected cells and a 7-fold increase in ERβ-transfected cells.
In agreement with the biochemical studies, 19 showed a full
agonistic profile on both receptors, reaching similar levels of
luciferase activity as E2. In ERR-transfected cells, 19 measured an
EC50 value of 1.87 μM and in ERβ-transfected cells an EC50 of
0.31 μM. Even though the EC50 of 19 is five orders of magnitude
lower than the highly potent E2, the strong transcriptional
activity and full agonistic profile are still remarkable for such a
small ligand. In addition, the cellular assay supports the bio-
chemical observation of ERβ selectivity in the case of compound
19 and analogues.
’ STRUCTURAL EVALUATION
To elucidate the binding affinity and selectivity of the frag-
ments in molecular detail, X-ray crystal structures were obtained
on selected fragment-ERβ LBD complexes (Figure 2). Such
structural information is expected to provide the molecular basis
for future fragment enlargement studies and the design of follow-
up libraries. For the X-ray studies, crystal structures of the ERβ
18
LBD bound to the natural agonist E2 and to three relevant
THIQ fragments (8, 10, and 19) were each generated in complex
with a coactivator peptide sequence based on the steroid receptor
coactivator-1 (SRC-1) box 2 sequence. The general structures of
the ERβ LBD-THIQ fragment complexes are similar to pre-
vious E2-bound hERR structures and to reported hERβ struc-
tures with other agonists19-21 (Figure 2). For all three fragments,
the helix 12 conformation of the ERβ corresponds to that
observed for ERβ bound to E2 and other agonists. This is in
turn consistent with the observed functional agonistic activity of
the fragments in the cofactor peptide binding and transcriptional
activation studies.
Overlays of the X-ray structures of the ERβ LBD bound
fragments 8, 10, and 19 with the E2 structure (Figure 2b-d)
show that all three fragments adopt a similar position. The
binding mode of the bicyclic ring system thus mimics the A-
and B-ring of E2. The backbones of all three compounds establish
a number of van der Waals contacts with the residues forming the
binding pocket. Additionally, for compounds 8 and 19, the
6-hydroxyl functionality perfectly matches the phenolic group
Figure 1. Binding profiles of 43 cofactor peptides immobilized on a
membrane, against both ERR and ERβ LBD without ligand and with
compounds E2, 8, and 19 (a complete list of all cofactor peptides tested
and their affinities can be found in the Supporting Information).
with the acetamide derivatives, methyl sulfonamide analogues
lacking hydroxyl substituents on the THIQ scaffold were found
not to be active (13). In contrast to the acetamide series, however
(e.g., 6), the combination of a 6-hydroxyl group with a non-
halogenated sulfonamide instead resulted in a compound (14)
with ER affinity and respectable ERβ selectivity. As well as for the
sulfonamide functionality, insertion of three fluorides, as in 19,
significantly enhanced the EC50 value in the peptide recruitment
assay (around 40-fold to 0.6 μM for ERβ and 3 μM for ERR).
Increasing the steric bulk of the sulfonamide side chain through
the incorporation of a 1-naphthyl group produced less potent
compounds (15 and 16). Reducing the steric bulk by replacing
the 1-naphthyl group with a phenyl ring (17) led to an increase in
ER affinity compared with the naphthyl derivatives. Further-
more, the incorporation of a tert-butyl ester resulted in a
compound (18) with an ERβ binding affinity similar to 17 and
confirms that halogenated N-substituents are not strictly re-
quired for obtaining high affinity.
of E2, forming a hydrogen bonding network with Glu305, Arg346
,
and a water molecule (Figure 2b,d). Fragment 10 instead features
a 7-hydroxyl group, which forms only one hydrogen bond with
Glu305 (Figure 2c). This feature is a likely explanation for the
lower ER affinity measured for 10.
To further evaluate the profile of the fragments in comparison
with the natural ligand E2, chip-based ligand-induced cofactor
binding studies were performed on a library of 53 cofactor
peptides, each one representing important binding epitopes of
both coactivator and corepressor proteins (Figure 1). This study
showed that the peptide binding profiles for E2 and fragments 8
and 19 were very similar. For both ER subtypes E2, 8, and 19 all
enhance to the same extent the binding affinity to peptides with a
coactivator motif (LXXLL). In addition, binding to the tested
corepressor motifs was not enhanced by any of the three ligands
(Figure 1). On the basis of these findings, it could be concluded
that the potent fragments have an agonistic pharmacological
profile.
The N-substituents of the three crystallized fragments occupy
part of the positional space usually adopted by the C- and D-rings
of E2 (Figure 2b-d). Because of the lack of a second hydroxyl
group to mimic the 17β-OH in the D-ring of E2, the fragments do
not form a hydrogen bond with the His475 side chain in the ERβ
cavity. The absence of this interaction most likely explains to a
significant extent the lower ER affinity of the fragments and offers
opportunities for ligand optimization, along with addressing the
hydrophobic space normally occupied by the C and D ring of E2.
Of interest is the observation that the N-substituents produce a
specific effect on ER affinity and selectivity. In particular, the
utilization of electron-withdrawing groups (e.g., 8 and 19) turned
out to be most suitable. The affinity enhancement for both ER
2007
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011

Journal of Medicinal Chemistry
ARTICLE
Figure 2. (a) General structural fold of the ERβ LBD bound to fragment 8 in complex with a SRC-1 box 2 peptide (red) and with helix 12 highlighted in
blue; overlays of the X-ray structures of the ERβ LBD cocrystallized with E2 (gray) with that of fragments 8, 10, or 19. (b) ERβ LBD-8 (green). (c) ERβ
LBD-10 (yellow). (d) ERβ LBD-19 (blue). Only key residues are shown for simplicity. Hydrogen bonds to key residues are shown as dotted lines.
interactions with amino acids in the cavity of ERβ than with
residues in the ERR.22 Two amino acid residues in close
proximity to the ligands are responsible for the differences in
the sizes of the ERR and ERβ cavity. The ERβ Leu384 is replaced
by Met336 in ERR, and the ERR Met421 is replaced by Ile373 in
ERβ (Figure 3).19,21,23 Ligands capable of interacting differently
with ERβ Ile373, in comparison to ERR Met421, have been
proposed as reason for selectivity toward ERβ.20,23,24 Specific
functional groups could thus be very important for the stereo-
electronic differentiation between the two receptor subtypes. As
shown in Figure 3, the trifluoromethyl sulfonamide moiety of
fragment 19 is held in close proximity to the Met421/Ile373
residues. Furthermore, the sulfonamide addresses a pocket in
ERβ not explored by E2. In particular, it appears that the small
distance between one of the fluoro atoms and the sulfur of ERR
Met421 would represent a repulsive interaction, which may limit
the conformational space occupied by the methionine side chain,
thus resulting in an unfavorable electrostatic interaction.20 In
contrast, the distance between the electron-withdrawing halogen
atom and ERβ Ile373 is greater, which is expected therefore not to
result in an unfavorable interaction (Figure 3). These observa-
tions could provide an explanation for why these THIQ ligands
show a preference for ERβ over ERR.
Figure 3. Crystal structure of compound 19 (a) with the ERβ LBD
(gray) and overlaid with ERR LBD-E2 (magenta; PBD 3DT3). Only
key residues are shown for simplicity. Distance monitors (black dotted
lines) show that the fluoro group is in close proximity to Met421/Ile373.
isoforms after incorporation of halogen groups can be attributed
to a favorable net hydrophobic effect due to the bulky size of the
substituent, as well as to the reduced polarity of the THIQ core
induced by the amide group. Additionally, van der Waals inter-
actions between the halogen atoms and surrounding residues
further stabilize the interaction of the ligand within the ligand
binding pocket.
Fragments 8 and 19 featured not only the highest ER potency
of this series but as well gave the most promising ERβ selectivity.
Weaker binding fragments such 15 and 16 also displayed a
modest selectivity for ERβ. It has been reported that aromatic
moieties are typically capable of making more favorable
’ CONCLUSIONS
Structure- and chemistry-based hierarchical organization of
library scaffolds in tree-like arrangements has enabled the
identification of the bicyclic THIQ scaffold as an originator of
novel small agonistic ER ligands with ERβ selectivity.4 The
fragment-based character of the ligands greatly facilitated their
synthesis. The affinity toward both ERR and ERβ has been
2008
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011

Journal of Medicinal Chemistry
ARTICLE
evaluated and a clear SAR discerned. The incorporation of a
trifluoromethyl acetamide (as in 8) or trifluoromethyl sulfon-
amide group (as in 19) produced the best ER binders in the series
with ca. 5-fold ERβ selectivity. Importantly, the fragment-like
character of the ligands enabled the ready generation of X-ray
structures and the elucidation of their binding affinity and ERβ
selectivity. The amide functionalities appear to address a hydro-
phobic subpocket in ERβ but clash with the respective methio-
nine at the same position in ERR. The information from the SAR,
together with the crystal structures of the fragments, aids in the
understanding of the binding profile of more decorated THIQ
analogues.14,15 Additionally, the fragments provide numerous
opportunities to develop more highly decorated compounds
with enhanced ERβ affinity and selectivity. In particular, the
affinity of fragment 19 is notable with respect to the small size of
the ligand and the ease of synthesis. Because fragment 19 does
not address His475 in the ERβ binding pocket, whereas E2 does,
further incorporation of polar functionalities branching out from
the amide group could enhance the affinity, for example, through
hydrogen bonding. Similarly, addressing the hydrophobic space
normally occupied by the C and D ring of E2 could also lead to
more potent ligands.
General procedure: Chlorotriethylsilane (TESCl, 0.044 mL, 0.26
mmol) was added dropwise, by way of syringe, to a cooled (0 ꢀC)
solution of 1,2,3,4-tetrahydroisoquinolinol (2 or 3) (0.050 g, 0.22
mmol) and NEt3 (0.15 mL, 1.1 mmol) in CH2Cl2 (2.5 mL) under Ar.
The reaction was then stirred at rt for the specified time, after which the
anhydride or sulfonyl chloride (0.28 mmol), and if required DMAP (cat.,
0.010 g), were added to the mixture. The reaction was then stirred under
Ar for a further 16 h. An additional volume of CH2Cl2 (5 mL) was then
added, followed by TBAF 3H2O (0.27 g, 0.87 mmol), and the reaction
3
was stirred for another 3 h. Water (10 mL) was then added, and the
reaction mixture was extracted with CH2Cl2 (3 ꢀ 10 mL). The CH2Cl2
fraction was dried (MgSO4) and evaporated under reduced pressure,
resulting in a viscous yellow oil which was purified by silica gel column
chromatography (eluent: ethyl acetate and hexanes as specified) to
afford the desired product. Purity was controlled by HPLC and NMR
spectroscopy and was, when not otherwise noted, g95%.
2-[(Trifluoromethyl)sulfonyl]-1,2,3,4-tetrahydro-6-isoqui-
nolinol 19.
Overall, the described approach for the generation of novel
fragments via scaffold identification in tree-like arrangements
provides validated starting points for the development of novel
ligands with specific profiles, as demonstrated here for the ERs
and ERβ selectivity. In addition, it allows us to generate relatively
potent small ligands that can be used to structurally explain the
binding of other more complex analogues in a rapid and easy
manner, as such significantly enriching the molecular insights in
drug discovery.
According to the general procedure, 1,2,3,4-tetrahydro-6-isoquinolinol
1 was silylated as described for 10 min and then treated with triflic
anhydride (0.048 mL, 0.081 g) for another 18 h. Addition of TBAF,
followed by workup of the dark solution and column chromatography
(30% EtOAc/cyclohexane), then afforded the desired product 19 as a
clear oil, which slowly became an opaque white oily solid (0.010 g, 17%,
purity 91%). 1H NMR (400 MHz, CDCl3) δ 6.95 (d, J = 8.4, 1H), 6.72
(dd, J = 8.3, 2.6, 1H), 6.65 (d, J = 2.6, 1H), 4.87 (br s, 1H), 4.59 (br s,
2H), 3.74 (br s, 2H), 2.93 (br t, J = 5.9, 2H). 13C NMR (101 MHz,
CDCl3) δ 154.6, 134.2, 127.3, 122.9, 120.0 (q, J = 322), 115.3, 114.3,
47.2, 44.2, 28.9. MS (ESI) 561 (30%), 280 (100). HRMS (ESI) m/e
calcd for C10H9O3NF3S [M - H]- 280.02607, found 280.02602.
’ EXPERIMENTAL SECTION
Analytical Techniques and Used Instrumentation. 1H and
13C NMR spectra were recorded on a Varian Mercury 400 (400 MHz,
1H NMR; 100.6 MHz, 13C NMR) at room temperature. NMR spectra
were calibrated to the solvent signals of CDCl3 (δ = 7.26 and 77.00
ppm), CD3OD (δ = 3.31 and 49.05 ppm), or DMSO-d6 (δ = 2.50 and
39.43 ppm). The chemical shifts are provided in ppm and the coupling
constants in Hz. The following abbreviations for multiplicities are used:
s, singlet; d, doublet; dd, double doublet; t, triplet; q, quadruplet; m,
multiplet; br, broad.
HPLC-ESI-MS analyses were performed on an HPLC system from
LCQ Advantage ESI (column: 125/4 Nucleodur C18 Gravity 3 μm;
Macherey and Nagel); eluent, H2O/0.1% HCO2H (A) and MeCN/
0.1% HCO2H (B) or MeOH/5% THF/0.1% HCO2H (B); gradient:
0-1 min with 10% B, 1-10 min with 10-100% B, 10-12 min with
100% B. HRMS (ESI): LTQ Orbitrap with flow injection (H2O/MeCN
= 1:1, 0.1% HCO2H, flow rate: 250 μL/min). The majority of
compounds possessed a purity of >95%, unless otherwise noted, as
determined by HPLC coupled with mass spectroscopy (HPLC-ESI-
MS).
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental and supporting
b
details to analytical techniques, chemical synthesis, protein
expression, and crystallography, biochemical and cell biological
assays. This material is available free of charge via the Internet at
http://pubs.acs.org.
Accession Codes
†The coordinates of the protein-ligand crystal structures have
been deposited at the Protein Data Bank with the following PDB
IDs: ERβ-8 (3OMO), ERβ-10 (3OMP), ERβ-19 (3OMQ).
’ AUTHOR INFORMATION
Corresponding Author
*Phone: þ31 40 2473737. Fax: þ31 40 2478367. E-mail: l.
High resolution mass spectra (HR-MS) were measured on a Thermo
Orbitrap coupled to a Thermo Accela HPLC system using electrospray
ionization (ESI).
brunsveld@tue.nl.
Chemical Synthesis. The N-protected 1,2,3,4-tetrahydro-6-iso-
quinolinols were synthesized according to a strategy published by Hoye
and co-workers, which utilized an in situ protection of the phenolic
group with a TES group.17 Syntheses of compounds 5-8, 13, and 14
have been published previously,4 and compounds 2-4 were purchased.
Syntheses of compounds 9-12 and 15-18 can be found in the
Supporting Information.
’ ACKNOWLEDGMENT
W.A.L.vO. thanks the Alexander von Humboldt Foundation
for a Georg Forster Research Fellowship for Experienced Re-
searchers and the University of the Witwatersrand for sabbatical
leave. Merck-Serono, MSD, and Bayer-Schering Pharma are
thanked for financial support.
2009
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011

Journal of Medicinal Chemistry
ARTICLE
’ ABBREVIATIONS USED
(14) Barlaam, B.; Dantzman, C. Preparation of isoquinolines and
isoindolines as selective estrogen receptor-β ligand. (AstraZeneca,
PLC). Patent WO 2002/46164, 2002.
(15) Chesworth, R.; Wessel, M. D.; Heyden, L.; Mangano, F. M.;
Zawistoski, M.; Gegnas, L.; Galluzzo, D.; Lefker, B.; Cameron, K. O.;
Tickner, J.; Lu, B. H.; Castleberry, T. A.; Petersen, D. N.; Brault, A.;
Perry, P.; Ng, O.; Owen, T. A.; Pan, L.; Ke, H. Z.; Brown, T. A.;
Thompson, D. D.; DaSilva-Jardine, P. Tetrahydroisoquinolines as
subtype selective estrogen agonists/antagonists. Bioorg. Med. Chem. Lett.
2004, 14, 2729–2733.
ER, estrogen receptor; LBD, ligand binding domain; SAR, struc-
ture-activity relationship; E2, 17β-estradiol; THIQ, tetrahydroi-
soquinoline; SRC-1, steroid receptor coactivator-1
’ REFERENCES
(1) (a) Hajduk, P. J.; Greer, J. A decade of fragment-based drug
design: strategic advances and lessons learned. Nature Rev. Drug
Discovery 2007, 6, 211–219. (b) Siegel, M. G.; Vieth, M. Drugs in other
drugs: a new look at drugs as fragments. Drug Discovery Today 2007, 12,
71–79. Congreve, M.; Chessari, G.; Tisi, D.; Woodhead, A. J. Recent
developments in fragment-based drug discovery. J. Med. Chem. 2008, 51,
3661–3680.
(2) Koch, M. A.; Schuffenhauer, A.; Scheck, M.; Wetzel, S.; Casaulta,
M.; Odermatt, A.; Ertl, P.; Waldmann, H. Charting biologically relevant
chemical space: a structural classification of natural products (SCONP).
Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17272–17277.
(3) Clark, A. M.; Labute, P. Detection and assignment of common
scaffolds in project databases of lead molecules. J. Med. Chem. 2009, 52,
469–483.
(4) Renner, S.; van Otterlo, W. A. L.; Dominguez Seoane, M.;
Mocklinghoff, S.; Hofmann, B.; Wetzel, S.; Schuffenhauer, A.; Ertl, P.;
Oprea, T. I.; Steinhilber, D.; Brunsveld, L.; Rauh, D.; Waldmann, H.
Bioactivity-guided mapping and navigation of chemical space. Nature
Chem. Biol. 2009, 5, 585–592.
(5) Bon, R. S.; Waldmann, H. Bioactivity-Guided Navigation of
Chemical Space. Acc. Chem. Res. 2010, 43, 1103–1114.
(6) Sato, M.; Grese, T. A.; Dodge, J. A.; Bryant, H. U.; Turner, C. H.
Emerging therapies for the prevention or treatment of postmenopausal
osteoporosis. J. Med. Chem. 1999, 42, 1–24.
(7) Kuiper, G. G.; Carlsson, B.; Grandien, K.; Enmark, E.; Haggblad,
J.; Nilsson, S.; Gustafsson, J. A. Comparison of the ligand binding
specificity and transcript tissue distribution of estrogen receptors alpha
and beta. Endocrinology 1997, 138, 863–870.
(8) Green, S.; Walter, P.; Kumar, V.; Krust, A.; Bornert, J. M.; Argos,
P.; Chambon, P. Human estrogen-receptor cDNA—sequence, expres-
sion and homology to v-erb-A. Nature 1986, 320, 134–139.
(9) (a) Greene, G. L.; Gilna, P.; Waterfield, M.; Baker, A.; Hort, Y.;
Shine, J. Sequence and expression of human estrogen-receptor com-
plementary DNA. Science 1986, 231, 1150–1154. (b) Kuiper, G. G. J. M.;
Enmark, E.; Pelto-Huikko, M.; Nilsson, S.; Gustafsson, J. A. Cloning of a
novel estrogen receptor expressed in rat prostate and ovary. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 5925–5930. (c) Mosselman, S.; Polman, J.;
Dijkema, R. ER beta: identification and characterization of a novel
human estrogen receptor. FEBS Lett. 1996, 392, 49–53.
(10) Zhang, W. H.; Andersson, S.; Cheng, G. J.; Simpson, E. R.;
Warner, M.; Gustafsson, J. A. Update on estrogen signaling. FEBS Lett.
2003, 546, 17–24.
(11) Veeneman, G. H. Subtype-selective Estrogens. In Nuclear
Receptors as Durg Targets; Ottow, E., Weinmann, H., Eds.; Wiley-
VCH: Weinheim, Germany, 2008; pp 65-126.
(12) (a) Nilsson, S.; Kuiper, G.; Gustafsson, J. A. ERbeta: a novel
estrogen receptor offers the potential for new drug development. Trends
Endocrinol. Metab. 1998, 9, 387–395. (b) Miller, C. P. SERMs: evolu-
tionary chemistry, revolutionary biology. Curr. Pharm. Des. 2002, 8,
2089–2111. (c) Wallace, O. B.; Richardson, T. I.; Dodge, J. A. Estrogen
receptor modulators: relationships of ligand structure, receptor affinity
and functional activity. Curr. Top. Med. Chem. 2003, 3, 1663–1680. (d)
Harris, H. A.; Albert, L. M.; Leathurby, Y.; Malamas, M. S.; Mewshaw,
R. E.; Miller, C. P.; Kharode, Y. P.; Marzolf, J.; Komm, B. S.; Winneker,
R. C.; Frail, D. E.; Henderson, R. A.; Zhu, Y.; Keith, J. C., Jr. Evaluation of
an estrogen receptor-beta agonist in animal models of human disease.
Endocrinology 2003, 144, 4241–4249.
(16) See for example: (a) Paul, R.; Coppola, J. A.; Cohen, E. 1-
Phenyl-2-phenylethyl-1,2,3,4-tetrahydroisoquinolines—new series of
nonsteroidal female antifertility agents. J. Med. Chem. 1972, 15, 720–
726. (b) Nagarajan, K.; Talwalker, P. K.; Kulkarni, C. L.; Shah, R. K.;
Shenoy, S. J.; Prabhu, S. S. Antiimplantation agents 2. 1,2-Diaryl-1,2,3,4-
tetrahydroisoquinolines. Ind. J. Chem., Sect. B: Org. Chem. Incl. Med.
Chem. 1985, 24, 83–97. (c) Renaud, J.; Bischoff, S. F.; Buhl, T.;
Floersheim, P.; Fournier, B.; Geiser, M.; Halleux, C.; Kallen, J.; Keller,
H.; Ramage, P. Selective estrogen receptor modulators with conforma-
tionally restricted side chains. Synthesis and structure-activity relation-
ship of ERalpha-selective tetrahydroisoquinoline ligands. J. Med. Chem.
2005, 48, 364–379. (d) Lin, H. R.; Safo, M. K.; Abraham, D. J.
Identification of a series of tetrahydroisoquinoline derivatives as poten-
tial therapeutic agents for breast cancer. Bioorg. Med. Chem. Lett. 2007,
17, 2581–2589.
(17) Hoye, T. R.; Renner, M. K.; Vos-DiNardo, T. J. Practical
method for chemoselective formation of MTPA amide derivatives from
amino alcohols and phenols. J. Org. Chem. 1997, 62, 4168–4170.
(18) M€ocklinghoff, S.; Rose, R.; Carraz, M.; Visser, A.; Ottmann, C.;
Brunsveld, L. Synthesis and crystal structure of phosphorylated estrogen
receptor ligand binding domain. ChemBioChem 2010, 11, 2251–2254.
(19) (a) Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E.; Bonn, T.;
Thorsell, A. G.; Engstrom, O.; Ljunggren, J.; Gustafsson, J. K.; Carlquist,
M. Structure of the ligand-binding domain of oestrogen receptor beta in
the presence of a partial agonist and a full antagonist. EMBO J. 1999, 18,
4608–4618. (b) 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. Molecular basis of agonism and antagonism in the
oestrogen receptor. Nature 1997, 389, 753–758. (c) Norman, B. H.;
Dodge, J. A.; Richardson, T. I.; Borromeo, P. S.; Lugar, C. W.; Jones,
S. A.; Chen, K.; Wang, Y.; Durst, G. L.; Barr, R. J.; Montrose-Rafizadeh,
C.; Osborne, H. E.; Amos, R. M.; Guo, S.; Boodhoo, A.; Krishnan, V.
Benzopyrans are selective estrogen receptor beta agonists with novel
activity in models of benign prostatic hyperplasia. J. Med. Chem. 2006,
49, 6155–6157. (d) Richardson, T. I.; Dodge, J. A.; Wang, Y.; Durbin,
J. D.; Krishnan, V.; Norman, B. H. Benzopyrans as selective estrogen
receptor beta agonists (SERBAs). Part 5: Combined A- and C-ring
structure-activity relationship studies. Bioorg. Med. Chem. Lett. 2007,
17, 5563–5566.
(20) Manas, E. S.; Unwalla, R. J.; Xu, Z. B.; Malamas, M. S.; Miller,
C. P.; Harris, H. A.; Hsiao, C.; Akopian, T.; Hum, W. T.; Malakian, K.;
Wolfrom, S.; Bapat, A.; Bhat, R. A.; Stahl, M. L.; Somers, W. S.; Alvarez,
J. C. Structure-based design of estrogen receptor-beta selective ligands. J.
Am. Chem. Soc. 2004, 126, 15106–15119.
(21) (a) Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E. A
structural biologist’s view of the oestrogen receptor. J. Steroid Biochem.
Mol. Biol. 2000, 74, 261–268. (b) Warnmark, A.; Treuter, E.; Gustafsson,
J. A.; Hubbard, R. E.; Brzozowski, A. M.; Pike, A. C. W. Interaction of
transcriptional intermediary factor 2 nuclear receptor box peptides with
the coactivator binding site of estrogen receptor alpha. J. Biol. Chem.
2002, 277, 21862–21868.
(22) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P. J.;
Agard, D. A.; Greene, G. L. The structural basis of estrogen receptor/
coactivator recognition and the antagonism of this interaction by
tamoxifen. Cell 1998, 95, 927–937.
(13) Artis, D. R.; et al. Scaffold-based discovery of indeglitazar, a
PPAR pan-active anti-diabetic agent. Proc. Natl. Acad. Sci. U.S.A. 2009,
106, 262–267.
(23) Malamas, M. S.; Manas, E. S.; McDevitt, R. E.; Gunawan, I.; Xu,
Z. B.; Collini, M. D.; Miller, C. P.; Dinh, T.; Henderson, R. A.; Keith,
J. C.; Harris, H. A. Design and synthesis of aryl diphenolic azoles as
2010
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011

Journal of Medicinal Chemistry
ARTICLE
potent and selective estrogen receptor-beta ligands. J. Med. Chem. 2004,
47, 5021–5040.
(24) (a) Yang, C. J.; Edsall, R.; Harris, H. A.; Zhang, X. C.; Manas,
E. S.; Mewshaw, R. E. ER beta ligands. Part 2: Synthesis and structure-
activity relationships of a series of 4-hydroxy-biphenyl-carbaldehyde
oxime derivatives. Bioorg. Med. Chem. 2004, 12, 2553–2570. (b)
Mewshaw, R. E.; Edsall, R. J.; Yang, C. J.; Manas, E. S.; Xu, Z. B.;
Henderson, R. A.; Keith., J. C.; Harris, H. A. ER beta ligands. 3.
Exploiting two binding orientations of the 2-phenylnaphthalene scaffold
to achieve ER beta selectivity. J. Med. Chem. 2005, 48, 3953–3979.
2011
dx.doi.org/10.1021/jm1011116 |J. Med. Chem. 2011, 54, 2005–2011