Evaluation of small-molecule modulators of the luteinizing hormone/choriogonadotropin and thyroid stimulating hormone receptors: Structure-activity relationships and selective binding patterns

Article abstract of DOI:10.1021/jm060247s

The substituted thieno[2,3-d]pyrimidine 3 (Org 41841), a partial agonist for the luteinizing hormone/choriogonadotropin receptor (LHCGR) and the closely related thyroid-stimulating hormone receptor (TSHR), was fundamentally altered, and the resulting anal

Full text of DOI:10.1021/jm060247s

3888
J. Med. Chem. 2006, 49, 3888-3896
Evaluation of Small-Molecule Modulators of the Luteinizing Hormone/Choriogonadotropin and
Thyroid Stimulating Hormone Receptors: Structure-Activity Relationships and Selective
Binding Patterns
,§
Susanna Moore,†,‡,§ Holger Jaeschke,‡,#,§ Gunnar Kleinau,|, Susanne Neumann,‡ Stefano Costanzi,| Jian-kang Jiang,†
John Childress,‡ Bruce M. Raaka,‡ Anny Colson,| Ralf Paschke,# Gerd Krause, Craig J. Thomas,† and
Marvin C. Gershengorn*,‡
Chemical Biology Core Facility, Clinical Endocrinology Branch, and Computational Chemistry Core Laboratory, National Institute of Diabetes
and DigestiVe and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892, III Department of Medicine, UniVersity of
Leipzig, D-04103 Leipzig, Germany, and Structural Bioinformatics and Molecular Design, Leibniz-Institut fu¨r Molekulare Pharmakologie,
D-13125 Berlin, Germany
ReceiVed March 3, 2006
The substituted thieno[2,3-d]pyrimidine 3 (Org 41841), a partial agonist for the luteinizing hormone/
choriogonadotropin receptor (LHCGR) and the closely related thyroid-stimulating hormone receptor (TSHR),
was fundamentally altered, and the resulting analogues were analyzed for their potencies, efficacies, and
specificities at LHCGR and TSHR. Chemical modification of the parent compound combined with prior
mutagenesis of TSHR provided compelling experimental evidence in support of computational models of
3 binding to TSHR and LHCGR within their transmembrane cores. Biochemical analysis of a specific
modification to the chemical structure of 3 provides additional evidence of a H-bond between the ligand
and a glutamate residue in transmembrane helix 3, which is conserved in both receptors. Several key
interactions were surveyed to determine their respective biochemical roles in terms of both van der Waals
dimensions and hydrogen bond capacity and the respective relationship to biological activity.
Introduction
diagnostic screen for thyroid cancer. TSHR agonists and
antagonists may well have utility in the diagnosis and treatment
of thyroid cancer, respectively. The development of small-
molecule modulators of LHCGR and FSHR has also been
pursued with varying degrees of success.5 Notable is the
development of a FSHR antagonist 1,6 a FSHR agonist 2,7 and
the potent LHCGR agonist 38 (Org 41841) (Figure 1).
Luteinizing hormone/choriogonadotropin (LH/CG), follicle-
stimulating hormone (FSH), and thyroid-stimulating hormone
(TSH) are heterodimeric glycoprotein hormones that regulate
reproduction and thyroid homeostasis.1-3 LH is responsible for
ovulation induction in women and controls testosterone produc-
tion in men. FSH causes ovarian follicle maturation in women
and is involved in spermatogenesis in men. TSH is involved in
the growth and function of thyroid follicular cells. Cellular
responses to all three glycoprotein hormones are mediated via
distinct seven transmembrane-spanning receptors (7-TMRs), i.e.,
the LHCGR, FSHR, and TSHR. Each receptor is characterized
by an elongated extracellular domain distinguished by several
leucine-rich motifs that are involved in recognition and binding
of the large glycoprotein hormones. The seven-transmembrane
helices of each receptor are noteworthy because of their high
degree of homology.
Disruption of physiological regulation of LHCGR, FSHR,
and TSHR by diverse pathogenic mutations has been implicated
in a number of human diseases.1,4 The specific and potent control
of these multifunctioning receptors could provide important
therapeutic advancements. LH and FSH are currently used
clinically for the treatment of infertility, and it can be reasoned
that synthetic agonists of LHCGR and FSHR have potential as
infertility therapeutics. Recombinant TSH is used in the
The advancement of low molecular weight modulators of
7-TMR function is well established within medicinal chemistry
and pharmacology. While highly potent small-molecule agonists
and antagonists of a multitude of 7-TMRs are frequently
reported from high-throughput screening, their receptor specific-
ity and binding mechanism often remains unknown. While the
prohibitive number of 7-TMRs and other cellular targets
preclude knowing the absolute specificity of any small molecule,
the competitive interactions of pharmacologically active agents
at highly homologous receptors can and should be explored.
Moreover, when possessing a combination of data regarding
the putative binding site of a pharmacological agent and a
modest degree of structural information of homologous recep-
tors, it may be possible to re-engineer the molecular structure
of a known binding agent to alter the binding preference. Using
a combination of molecular modeling and evaluation of the
binding of 3 within several TSHR/LHCGR chimera mutants,
we have recently reported that 3 was found to bind within the
seven-transmembrane domain of the TSHR and LHCGR.9
Specifically, 3 was shown to bind within a localized pocket
between transmembrane helices 3-7 and extracellular loop 2.
Low molecular weight ligands of the LHCGR and TSHR are
remarkable in that they are not likely to compete with the large
native hormones for binding at the extracellular N-terminal
domain. Herein, we report the exploration of the small-molecule
agonist 3 via several fundamental structural alterations and the
* To whom correspondence should be addressed. Phone and fax: 301-
496-4128. E-mail: marving@intra.niddk.nih.gov.
† Chemical Biology Core Facility, National Institute of Diabetes and
Digestive and Kidney Diseases.
‡ Clinical Endocrinology Branch, National Institute of Diabetes and
Digestive and Kidney Diseases.
§ These authors contributed equally to this work.
# University of Leipzig.
| Computational Chemistry Core Laboratory, National Institute of
Diabetes and Digestive and Kidney Diseases.
Leibniz-Institut fu¨r Molekulare Pharmakologie.
10.1021/jm060247s CCC: $33.50 © 2006 American Chemical Society
Published on Web 05/24/2006

Hormone/Choriogonadotropin Receptor, Org 41841
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13 3889
Figure 1. Structures of FSHR antagonist 1, FSHR agonist 2, and LHCGR agonist 3 (Org 41841).
Scheme 1a
a Reagents and conditions: (i) K2CO3, EtOH, 60 °C, 5 h; (ii) POCl3, dioxane, reflux, 2 h; (iii) NaOEt, EtOH, 50 °C, 3 h; (iv) LiOH, dioxane/H2O; (v)
PyBop, DIPEA, DMF followed by addition of amine (a subsequent benzyl deprotection was needed for the synthesis of analogue 19; see Supporting Information).
resulting effects on efficacy, potency, and specificity between
the highly homologous LHCGR and TSHR.
Biochemistry. Both human TSHR and human LHCGR were
stably expressed in HEK 293 EM cells as previously de-
scribed.13,14 Cell surface expression of TSHR and LHCGR were
determined via FACS analysis.15 Agonism of compounds 3-20
were determined via measurement of intracellular cyclic AMP
accumulation. It should be noted that 3 and the derivatives
described herein were evaluated at the FSHR and no significant
activity was observed.
Docking experiments of the ligand within both LHCGR and
TSHR consistently suggested a potential hydrogen bond between
the amine functionality of 3 and E3.37. We have previously
reported the examination of this prospective interaction via an
E to A mutant at position 3.37 for TSHR.9 Consistent with the
suggested interaction displayed in our models, E3.37A was not
activated by 3. The dimethylated analogue 20 provided a reagent
to test this hypothesis directly by eliminating the needed
hydrogen bond donation by the relevant amine. The resulting
agonism displayed by 20 was shown to be only 7% relative to
that of 3 at both LHCGR and TSHR (Figure 2). These
experiments provide congruent evidence from two unrelated
analyses for a critical H-bond between 3 and TSHR and
LHCGR.
The chemical structure of 3 was further modified in a variety
of ways in an attempt to define novel activity. Table 1 presents
a compilation of the activities of 3-19. Of note is the greater
tolerance toward ligand modification found at the LHCGR
compared to TSHR. To more adequately appreciate the differ-
ences in the receptor-ligand interaction, the full dose-response
curves of 3, 5, and 7 were established (Figure 3). These
analogues were chosen on the basis of the range of activation
that was initially noted for each small molecule at both receptors.
Further, each compound represents single-point alterations to
the potentially important amide pharmacophore. These experi-
ments show that the addition of an N-methyl group to the tert-
Results
Chemistry. The synthesis of 3 and several analogues based
thereupon was accomplished in a manner similar to that
described in the original report by van Boeckel and co-workers.8
A modified Biginelli condensation afforded the substituted
pyrimidone scaffold.10 We found that numerous aldehydes were
tolerated within this system including highly electron-withdrawn
(i.e. polyfluoro and nitro) and electron-rich (polymethoxy and
hydroxyl) aromatic ring systems. Treatment with POCl3 afforded
the 4-chloro-substituted pyrimidines in quantitative yields, and
substitution with either ethyl 2-mercaptoacetate or tert-butyl
2-mercaptoacetate11 afforded several thienopyrimidines, includ-
ing biochemically relevant compounds 4 and 5 (Scheme 1).
Saponification of the ethyl esters with lithium hydroxide in a
dioxane/water mixture provided the thienopyrimidine acids, and
PyBOP catalyzed amide couplings with several amines provided
3 and compounds 6-19.12
Initial docking experiments suggested a potential hydrogen
bond between the amine functionality of 3 and E3.37 in
transmembrane helix 3 of both TSHR and LHCGR. To fully
examine this, we chose to eliminate this potential interaction
via two distinct experimental means. When the small molecule
is used as a point of manipulation, the removal of the aromatic
amine or the protection of the aromatic amine via dimethylation
would accomplish the exclusion of H-bond donation capability.
Unfortunately, all attempts to deaminate the structure of 3 were
unsuccessful. However, direct treatment with methyl iodide in
basic acetonitrile afforded the dimethylamine analogue 20
(Figure 2) along with the monomethylated and trimethylated
analogues that were not characterized fully. Purification via
HPLC was required prior to biological evaluation of 20.

3890 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13
Moore et al.
Figure 2. Analysis of 20 at both the TSHR and LHCGR. Comparison of the activation of TSHR and LHCGR by compounds 3 and 20 relative to
basal activities of both receptors is shown. Intracellular cAMP production was determined in response to 100 µM of each compound and is expressed
as the percent of maximum response of TSHR/LHCGR to TSH (100 mU/mL)/LH (1000 ng/mL). The data are presented as the mean ( SEM of
two independent experiments, each performed in duplicate.
Table 1. Pharmacological Characterization of Selected Analogues of 3 at TSHR and LHCGR Stably Expressed in HEK EM 293 Cellsa
a Agonistic activity of compounds was determined via measurement of intracellular cyclic AMP. The efficacy (maximum response) is expressed as the
percent of maximum response of LHCGR or TSHR to LH (1000 ng/ml) or TSH (100 mU/mL), respectively. EC50 values and 95% conficence intervals (C.I.)
were obtained from dose-response curves (0-100 µM compound) using the GraphPad Prism 4.0 software. Confidence intervals were not calculated in
dose-response curves that did not reach an obvious plateau. n.d. ) not determined. / ) estimated maximum response at 100 µM compound. // ) estimated
EC50 (dose-response curve revealed no plateau).
butylamide in compound 7 precludes an agonistic response at
the TSHR while allowing a response at LHCGR similar to that
of 3. The original report of 3 described the utility of the tert-
butyl moiety as a major determinant of its potency at LHCGR.
We explored the role of the linkage (amide or ester) between
the aromatic and tert-butyl segments of 3. The alteration of the
tert-butylamide (displayed in 3) to a tert-butyl ester (displayed
in 5) had a deleterious effect on both potency and efficacy at
LHCGR, while the activity at TSHR was unaffected.
TSHR and LHCGR (detailed experimental procedures and
descriptions are found in the Supporting Information).9
The similarities and dissimilarities of the amino acid residues
between the receptor subtypes were studied to elucidate positions
that form and cover the binding pocket for the two receptors.
To facilitate the comparative relationship between selected
residues of each receptor, a GPCR residue indexing system was
utilized (i.e., E3.37 for E506 at TSHR and for E451 at
LHCGR).17,18 Several conserved amino acids, such as the key
glutamate (E3.37), as well as nonconserved amino acids, such
as the F f T (position 5.42), Y f F (position 6.54) and L f
F (position 570/515), were deemed relevant toward the binding
of 3 at both LHCGR and TSHR. A complete comparison of
Docking Studies. We have previously reported the construc-
tion of molecular models of TSHR and LHCGR.9 In this study,
by means of the ICM software,16 3 and selected analogues were
docked within the previously described binding pockets of 3 at

Hormone/Choriogonadotropin Receptor, Org 41841
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13 3891
Figure 3. Full concentration analyses of compounds 3, 5, and 7 at TSHR and LHCGR. The data are presented as the mean ( SEM of two
independent experiments, each performed in duplicate.
Table 2. Comparison of Selected Amino Acid Residues of the TSHR
and LHCGR with Relevance to the Org 41841 Binding Pocket
phenyl ring is seen in a position proximal to transmembrane
helixes 4 and 5 and bordering the extracellular loop region. In
region
TMH3
position
TSHR
LHCGR
orientation B, the methoxyphenyl ring is seen in a position
proximal to transmembrane helix 3 and more deeply buried
within the transmembrane core (Figure 4). We chose to amplify
upon these two orientations by docking selected analogues of
3 (namely, compounds 5 and 7) based on the altered hydrogen
bond capacity of the amide and their differing pharmacological
results. We further anticipated that the docking models would
allow us to amplify upon the relevance of the tert-butylamide
moiety in terms of both van der Waals dimensions and hydrogen
bond capacity. Both binding orientations for both receptors
(shown in Figure 4 for 3 and in the Supporting Information for
5 and 7) are shared (with certain dissimilarities) for all three
compounds.
A detailed survey of the resulting 12 models (shown in the
Supporting Information) provided several points of interest but
none more striking than the significant docking shifts that were
noted for analogue 7 (Figure 5). Despite bearing a larger and
more hindered amide substituent because of the presence of an
additional methyl group, compound 7 can be accommodated
within the binding pocket of LHCGR with direct overlap with
compound 3. In contrast, in molecular models of the TSHR,
the larger van der Waals dimensions of compound 7 did not
allow this compound to adopt a docking pose similar to that of
3 in either orientation A or B (Figure 5 details the docking of
7 in orientation A). Because of the additional steric constraints,
the amide moiety of compound 7 could not be accommodated
in the tightly packed region surrounding M6.48 in orientation
A within the TSHR and, importantly, the essential H-bond with
E3.37 was not observed. Moreover, the increased bulkiness of
the amide moiety of compound 7 generated a steric clash within
the TSHR binding pocket of orientation B as well. Further, fewer
docking poses with acceptable docking scores were observed
for 7 within TSHR molecular models presumably because of
the presence of the bulkier amino acid residues that constitute
the binding pocket (i.e., L570, F5.42, and I6.59). Consequently,
the aromatic amine moiety of compound 7 did not engage in a
H-bond with E3.37 (the distance between carboxyl oxygen of
E3.37 and the NH2 group of compound 7 was noted to be 5.7
Å, shown in Supporting Information). These results are signifi-
cant for the advancement of next-generation derivatives of 3.
3.32
3.33
3.36
3.37
3.40
4.56
4.57
4.59
4.60
Thr 501
Val 502
Ser 505
Glu 506
Val 509
Leu 552
Ala 553
Leu 555
Pro 556
Ile 560
Ser 561
Leu 570
Pro 571
Met 572
Asp 573
Pro 577
Ala 579
Leu 580
Ala 581
Tyr 582
Phe 585
Val 586
Leu 589
Asn 590
Met 637
Ile 640
Thr 446
Val 447
Ser 450
Glu 451
Val 454
Ile 497
TMH 4
ECL 2
Ala 498
Leu 500
Pro 501
Val 505
Ser 506
Phe 515
Pro 516
Met 517
Asp 518
Thr 522
Ser 524
Gln 525
Val 526
Tyr 527
Thr 530
Ile 531
Leu 534
Asn 535
Met 582
Ile 585
Ser 586
Phe 588
Ala 589
Ala 593
Val 609
Tyr 612
TMH 5
5.34
5.36
5.37
5.38
5.39
5.42
5.43
5.46
5.47
6.48
6.51
6.52
6.54
6.55
6.59
7.39
7.42
TMH 6
TMH 7
Ser 641
Tyr 643
Ala 644
Ile 648
Val 664
Tyr 667
a Amino acids are identified according to the residue numbering
(including signal peptide) and to the Ballesteros-Weinstein numbering. The
binding pockets of TSHR and LHCGR are composed of 32 amino acids,
11 of which are divergent. TMH ) transmembrane helix. ECL )
extracellular loop. Bolded residues indicate dissimilar amino acid residues.
the amino acid residues that form the putative binding pockets
is shown in Table 2.
The resulting models represent an extension of our previous
work on TSHR9,15 and constitute the first reported docking
models of this class of ligands at LHCGR. In part, the models
confirm our supposition that LHCGR contains a larger binding
pocket than the TSHR. This has numerous consequences with
regard to the binding mode of 3 and selected analogues within
the two related receptors. From an early stage in our modeling
experiments it was apparent that 3 could bind in two distinct
orientations in both receptors with nearly similar docking scores.
Common to both is the pivotal role of E3.37 as a hydrogen
bond acceptor. Further, these two orientations were statistically
more significant than all other docking orientations examined.
They can be briefly explained by using the methoxyphenyl
moiety as a point of reference. In orientation A, the methoxy-
Discussion
The optimization of the binding arrangements of 3 within
the molecular models of LHCGR and TSHR provided numerous
insights that could be experimentally examined. For instance,
we determined whether exploitation of the T f F variation
(T530 and F585 in LHCGR and TSHR, respectively) could
provide an analogue of 3 with altered specificity favoring TSHR
activation. To this end we synthesized several aromatic amides

3892 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13
Moore et al.
Figure 4. Docking poses of 3 within homology models of TSHR and LHCGR optimized to orientations A and B. The binding pocket is located
within the extracellular half of the transmembrane helical bundle between TMH 3-7 and ECL 2 (backbone ribbons in white). Amino acids that are
not conserved between the two receptors are underlined. Coloring is as follows: green ) hydrophobic side chains; green/blue ) aromatic side
chains; magenta ) hydrophilic side chains; red ) negatively charged side chains. Compound 3 is shown in orange with atoms in color.
(14 and 15) in the hope of achieving a stacking interaction
between the phenyl rings of the altered compounds and F585.
There was, however, no enhancement of activity at TSHR based
on these structural modifications to 3. We also took note of the
specific interactions between 3 and the negatively charged E3.37
and synthesized several analogues with amplified positive charge
and/or enhanced H-bonding capacity in an effort to enhance
the binding of the small molecule. However, none of these
derivatives (8-11) were found to be effective modulators of
either receptor. Several derivatives were synthesized in an effort
to introduce functionality with different H-bonding capacity at
the amide portion of the molecule (12 and 13) in the hopes of
defining novel binding arrangements for receptor activation.
However, neither derivative provided an enhancement of
potency or selectivity. Our studies did confirm that the tert-
butyl moiety was optimal for the small molecule to interact with
one of the two hydrophobic cores (between TMH4 and TMH5
proximal to the extracellular domain and between TMH3 and
TMH6 near the M6.48 region) located within the binding pocket
defined by our models. Further, the necessity of the amide
functionality was confirmed by the distinct activities of 3 and
5. We are unable, on the basis of our current experiments and
models, to categorize the amide pharmacophore from the basis
of the H-bond variance of amides and esters or the associated
van der Waals constraints. A small survey of the methoxyphenyl
ring revealed that modest alterations of this moiety are seemingly
allowed. In particular, the replacement of the methoxy group
with a hydroxyl (19) was noted to have no critical influence on
the potency at the LHCGR but a large enhancement in the
displayed efficacy.
Analysis of the docking models of this class of ligands at
both LHCGR and TSHR provided two distinct docking orienta-
tions. The two orientations have nearly equivalent docking
scores, and yet they represent two vastly divergent ligand
postures. It should be noted that recent structural modifications
of 3 have denoted that the methoxyphenyl ring can be modified
to accommodate larger substitutions (such as phenylamide and
several small peptide chains up to eight atoms) without
consequence toward the potency of the small molecule at the
LHCGR.5 Taking these data into account does not provide any
definitive preference in favor of either orientation A or B, since
the limited spatial extension at the methoxy moiety was

Hormone/Choriogonadotropin Receptor, Org 41841
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13 3893
Figure 5. Docking pose of 7 within homology model of TSHR optimized to orientation A. The binding pocket is located within the extracellular
half of the transmembrane helical bundle between TMH 3-7 and ECL 2 (backbone ribbons in white). Amino acids that are not conserved between
the two receptors are underlined. Coloring is as follows: green ) hydrophobic side chains; green/blue ) aromatic side chains; magenta ) hydrophilic
side chains; red ) negatively charged side chains. Compound 7 is shown in orange with atoms in color.
accommodated in both cases. From Figure 4 (right) one can
comprehend that in orientation A limited spatial extension such
as the phenylamide are spatially feasible at Y5.39 in the LH/
CGR. In orientation B a phenylamide could be accommodated
above Y7.42 between TMH7 and TM3.
a mechanism by which the selectivity of 3 could be optimized.
Several fundamental adjustments were made to the scaffold of
3 in an attempt to take advantage of several key amino acid
residue variances between the two receptors. A critical hydrogen
bond between 3 and a key glutamate residue (E3.37) was
suggested by molecular modeling and subsequently confirmed
by complementary experimental evidence through mutagenesis
of the receptor target9 and modification of the small-molecule
ligand. The selectivity of 3 toward LHCGR was amplified by
taking note of the smaller binding pocket within TSHR and
accordingly increasing the steric bulk of the amide portion of
the molecule (compound 7). Modeling revealed a potential
mechanism for this selectivity through the preclusion of the
required H-bond to E3.37 based on the hindered interaction
between 7 and TSHR. These results will be of key importance
in improving our molecular model of the receptor-small
molecule interaction and further refinement of the ligand.
The predicted interaction best supported by experimental data
is the H-bond observed between the amine functionality of 3
and E3.37. We examined this potential interface by converting
the amine to a dimethylamine moiety in 20, thus eliminating
the H-bond donation capacity of the small molecule (Figure
2). Previously, we mutated TSHR to eliminate the glutamate
residue and thus abolish the H-bond potential between the
receptor and the small molecule.9 We observed that each of
these independent changes markedly reduced agonistic activity
at TSHR and, in terms of the activity of 20, at LHCGR as well.
This provides compelling experimental evidence supporting the
existence of a critical H-bond between the small molecule and
E3.37 and our proposed molecular docking arrangements of 3
within both LHCGR and TSHR.
Experimental Section
Among the most important derivatives of the parent com-
pound was the N-tert-butyl-N-methyl analogue 7. This small
molecule was noted to retain similar potency and efficacy at
LHCGR while, importantly, eliminating the modest activity
displayed by 3 at TSHR. Examination of 7 within our TSHR
docking studies revealed that the slight addition of steric bulk
at the amide was sufficient to eliminate the apparently critical
H-bond between the compound and E3.37. However, docking
of 7 within our LHCGR homology models revealed good
overlap with the parent compound and the apparently critical
hydrogen bond is unaltered. Again, these results provide
persuasive evidence for a putative binding pocket within a cleft
between transmembrane helices 3-6 of both receptors.
General. 1H NMR data were recorded on a Varian Gemini 300
MHz instrument. Spectra were recorded in DMSO-d6, CDCl3-d4,
and acetone-d6 and were referenced to the residual solvent peak at
2.50, 7.26 and 2.04 ppm, respectively. Reverse-phase (C18) HPLC
was carried out using an Agilent HPLC with a Zorbax SP-C18
semiprep column. High-resolution mass spectrometry measurements
were performed on a Micromass/Waters LCT Premier electrospray
TOF mass spectrometer.
General Experimental Procedures. 5-Carbonitrile-1,6-dihy-
dro-2-(methylthio)-6-oxo-4-(substituted phenyl)pyrimidines. To
a solution of S-methylisothiourea (1 equiv), the appropriately
substituted benzaldehyde (2 equiv) and ethyl cyanoacetate (2 equiv)
in ethanol was added K2CO3 (2 equiv). The reaction mixture was
heated to 60 °C for 5 h and filtered upon cooling to obtain products.
Purification by flash chromatography (using EtOAc/hexane 1:1)
provided the final products as off-white solids in 30-50% yields.
Conclusion
Herein, we have described the first reported docking model
of a LHCGR-small molecule complex. Further, we have refined
our model of the highly homologous TSHR-small molecule
complex and compared the two models in an attempt to identify
5-Carbonitrile-4-chloro-2-(methylthio)-6-(3-substituted phe-
nyl)pyrimidines. To a mixture of the oxopyrimidines in dioxane
was added POCl3 (excess) in dioxane. The mixture was heated to
reflux for 3 h, and the solvent was removed by reduced pressure.

3894 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13
Moore et al.
1
Saturated NaHCO3 was added to the resulting brown solids, and
the reaction mixtures were extracted with CH2Cl2 (3 × 100 mL).
The organic layers were combined and dried over Na2SO4, and the
solvent was removed under reduced pressure. Purification by silica
plug filtration (using EtOAc/hexane 1:1) provided the final products
as white crystalline solids in 80-90% yields.
Ethyl 5-Amino-2-(methylthio)-4-(substituted phenyl)thieno-
[2,3-d]pyrimidine-6-carboxylates. To a solution of the appropriate
pyrimidine (1 equiv) and ethyl 2-mercaptoacetate or tert-butyl-2-
mercaptoacetate (1.1 equiv) in ethanol was added sodium (0.910
equiv) in ethanol. The yellow reaction mixture was heated to 50
°C for 3 h and cooled, and the ethanol was removed under reduced
pressure. The yellow solids were dissolved in CH2Cl2 (50 mL) and
washed with deionized water (3 × 25 mL). The organic layer was
dried over Na2SO4, and the solvent was removed under reduced
pressure. Purification by flash chromatography (using EtOAc/
hexane 1:1) provided the final products as yellow solids in 70-
90% yields.
N-tert-Butyl-5-amino-2-(methylthio)-4-(substituted phenyl)-
thieno[2,3-d]pyrimidine-6-carboxamides. To a solution of the
appropriate ethyl ester (1 eqiuv) in a dioxane and water mixture
was added lithium hydroxide (2 equiv). The reaction mixture was
heated to 50 °C for 3 h and cooled, and the solvent was removed
under reduced pressure. The crude acid was used without further
purification. The yellow solids were dissolved in a minimal amount
of DMF, followed by the addition of PyBOP (3 equiv), DIPEA
(5.5 equiv), and tert-butylamine (3 equiv), sequentially. Purification
by flash chromatography (using EtOAc/hexane 2:1) provided the
final products as yellow solids in 50-90% yields.
N-tert-Butyl-5-amino-4-(3-methoxyphenyl)-2-(methylthio)-
thieno[2,3-d]pyrimidine-6-carboxamide (3). Analysis by C8 re-
versed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 17 min, 30% f 70% CH3CN
at a flow rate of 1 mL/min, tR ) 13.5 min) indicated greater than
99% purity by peak integration. 1H NMR (CDCl3) δ 1.45 (s, 9H),
2.64 (s, 3H), 3.86 (s, 3H), 5.99 (br s, 2H), 7.07-7.26 (m, 3H),
7.41-7.47 (m, 1H); mass spectrometry (TOF) m/z ) 403.1262 (M
+ H+) (theoretical 403.1257).
Ethyl 5-Amino-4-(3-methoxyphenyl)-2-(methylthio)thieno[2,3-
d]pyrimidine-6-carboxylate (4). Analysis by C8 reversed-phase
LC-MS using a linear gradient of H2O with increasing amounts
of CH3CN (0 f 15 min, 30% f 90% CH3CN at a flow rate of 1
mL/min, tR ) 12.5 min) indicated greater than 92% purity by peak
integration. 1H NMR (DMSO-d6) δ 1.37 (t, J ) 7.2 Hz, 3H), 2.69
(s, 3H), 3.92 (s, 3H), 4.35 (q, J ) 7.2 Hz, 2H), 6.15 (br s, 2H),
7.27-7.31 (m, 3H), 7.59-7.64 (m, 1H); mass spectrometry (TOF)
m/z ) 376.0790 (M + H+) (theoretical 376.0784).
tert-Butyl 5-Amino-4-(3-methoxyphenyl)-2-(methylthio)thieno-
[2,3-d]pyrimidine-6-carboxylate (5). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 15 min, 30% f 90% CH3CN at a flow
rate of 1 mL/min, tR ) 13.2 min) indicated greater than 98% purity
by peak integration. 1H NMR (CDCl3) δ 1.57 (s, 9H), 2.64 (s, 3H),
3.86 (s, 3H), 5.78 (br s, 2H), 7.08-7.16 (m, 3H), 7.42-7.47 (m,
1H); mass spectrometry (TOF) m/z ) 404.1097 (M + H+)
(theoretical 404.1103).
95% purity by peak integration. H NMR (DMSO-d6) δ 1.36 (s,
9H), 2.58 (s, 3H), 3.00 (s, 3H), 3.82 (s, 3H), 5.22 (br s, 2H), 7.17-
7.20 (m, 3H), 7.51 (t, J ) 8 Hz, 1H); mass spectrometry (TOF)
m/z ) 417.1413 (M + H+) (theoretical 417.1419).
5-Amino-4-(3-methoxyphenyl)-2-(methylthio)thieno[2,3-d]py-
rimidine-6-carbohydrazide (8). Analysis by C8 reversed-phase
LC-MS using a linear gradient of H2O with increasing amounts
of CH3CN (0 f 18 min, 40% f 80% CH3CN at a flow rate of 1
mL/min, tR ) 13.7 min) indicated greater than 92% purity by peak
integration. 1H NMR (DMSO-d6) δ 2.59 (s, 3H), 3.82 (s, 3H), 6.18
(br s, 2H), 7.17-7.20 (m, 3H), 7.50 (t, J ) 8.7 Hz, 1H), 9.20 (br
s, 1H); mass spectrometry (TOF) m/z ) 362.074 (M + H+)
(theoretical 362.0745).
5-Amino-4-(3-methoxyphenyl)-N′,N′-dimethyl-2-(methylthio)-
thieno[2,3-d]pyrimidine-6-carbohydrazide (9). Analysis by C8
reversed-phase LC-MS using a linear gradient of 0.1% TFA in
H2O with increasing amounts of CH3CN (0 f 18 min, 30% f
80% CH3CN at a flow rate of 1 mL/min, tR ) 8.7 min) indicated
greater than 93% purity by peak integration. 1H NMR (DMSO-d6)
δ 2.55 (s, 6H), 2.58 (s, 3H), 3.82 (s, 3H), 6.45 (br s, 2H), 7.16-
7.18 (m, 3H), 7.50 (t, J ) 8.7 Hz, 1H), 8.72 (s, 1H); mass
spectrometry (TOF) m/z ) 390.1053 (M + H+) (theoretical
390.1058).
N′-tert-Butyl-5-amino-4-(3-methoxyphenyl)-2-(methylthio)-
thieno[2,3-d]pyrimidine-6-carbohydrazide (10). Analysis by C8
reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 10 min, 25% f 90% CH3-
CN, 10 f 15 min, 90% f 25% CH3CN at a flow rate of 1 mL/
min, tR ) 12.0 min) indicated greater than 95% purity by peak
1
integration. H NMR (DMSO-d6) 1.08 (s, 9H), 2.59 (s, 3H), 3.82
(s, 3H), 6.45 (br s, 2H), 7.16-7.18 (m, 3H), 7.50 (t, J ) 8 Hz,
1H); mass spectrometry (TOF) m/z ) 418.1366 (M + H+)
(theoretical 418.1371).
N′-Boc-5-amino-4-(3-methoxyphenyl)-2-(methylthio)thieno-
[2,3-d]pyrimidine-6-carbohydrazide (11). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 10 min, 25% f 90% CH3CN, 10 f 15
min, 90% f 25% CH3CN at a flow rate of 1 mL/min, tR ) 11.0
min) indicated greater than 97% purity by peak integration. 1H NMR
(DMSO-d6) 1.42 (s, 9H), 2.60 (s, 3H), 3.82 (s, 3H), 6.18 (br s,
2H), 7.16-7.21 (m, 3H), 7.51 (t, J ) 8 Hz, 1H), 8.83 (bs, 1H),
9.62 (bs, 1H); mass spectrometry (TOF) m/z ) 462.1264 (M +
H+) (theoretical 462.127).
5-Amino-4-(3-methoxyphenyl)-N-(2-hydroxyethyl)-2-(meth-
ylthio)thieno[2,3-d]pyrimidine-6-carboxamide (12). Analysis by
C8 reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 18 min, 30% f 60% CH3CN
at a flow rate of 1 mL/min, tR ) 7.4 min) indicated greater than
1
92% purity by peak integration. H NMR (DMSO-d6) δ 2.59 (s,
3H), 3.20-3.40 (m, 2H), 3.41-3.55 (m, 2H), 3.82 (s, 3H), 4.71
(m, 1H), 6.10 (br s, 2H), 7.17-7.22 (m, 3H), 7.50 (m, 1H), 7.80
(m, 1H); mass spectrometry (TOF) m/z ) 391.0893 (M + H+)
(theoretical 391.0899).
5-Amino-N-(cyanomethyl)-4-(3-methoxyphenyl)-2-(methylth-
io)thieno[2,3-d]pyrimidine-6-carboxamide (13). Analysis by C8
reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 10 min, 25% f 90% CH3-
CN, 10 f 15 min, 90% f 25% CH3CN at a flow rate of 1 mL/
min, tR ) 10.5 min) indicated greater than 91% purity by peak
integration. 1H NMR (DMSO-d6) δ 2.60 (s, 3H), 3.82 (s, 3H), 4.22
(d, J ) 5.4 Hz, 2H), 6.23 (br s, 2H), 7.18-7.20 (m, 3H), 7.51 (t,
J ) 8.1 Hz, 1H), 8.53 (t, J ) 5.4 Hz, 1H); mass spectrometry (TOF)
m/z ) 386.074 (M + H+) (theoretical 386.0745).
5-Amino-N-benzyl-4-(3-methoxyphenyl)-2-(methylthio)thieno-
[2,3-d]pyrimidine-6-carboxamide (14). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 18 min, 40% f 80% CH3CN at a flow
rate of 1 mL/min, tR ) 12.6 min) indicated greater than 99% purity
by peak integration. 1H NMR (acetone-d6) δ 2.61 (s, 3H), 3.89 (s,
3H), 4.55 (d, J ) 6 Hz, 2H), 6.29 (br s, 2H), 7.18-7.37 (m, 8H),
5-Amino-N-(ethyl)-4-(3-methoxyphenyl)-2-(methylthio)thieno-
[2,3-d]pyrimidine-6-carboxamide (6). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 18 min, 40% f 80% CH3CN at a flow
rate of 1 mL/min, tR ) 9.8 min) indicated greater than 99% purity
1
by peak integration. H NMR (DMSO-d6) δ 1.08 (t, J ) 7.2 Hz,
3H), 2.59 (s, 3H), 3.22 (p, J ) 7.2 Hz, 2H), 3.82 (s, 3H), 5.75 (s,
1H), 6.10 (br s, 2H), 7.15-7.19 (m, 2H), 7.50 (t, J ) 8.0 Hz, 1H),
7.87 (t, J ) 8.0 Hz, 1H); mass spectrometry (TOF) m/z ) 375.0944
(M + H+) (theoretical 375.0949).
N-tert-Butyl-5-amino-4-(3-methoxyphenyl)-N-methyl-2-(me-
thylthio)thieno[2,3-d]pyrimidine-6-carboxamide (7). Analysis by
C8 reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 18 min, 40% f 80% CH3CN
at a flow rate of 1 mL/min, tR ) 14.6 min) indicated greater than

Hormone/Choriogonadotropin Receptor, Org 41841
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13 3895
7.49 (t, J ) 8.4 Hz, 1H), 7.64 (t, J ) 3 Hz, 1H); mass spectrometry
(TOF) m/z ) 437.1100 (M + H+) (theoretical 437.1106).
potency (EC50) was obtained from dose-response curves (0-100
µM compound) by data analysis with GraphPad Prism 4 for
Windows.
5-Amino-4-(3-methoxyphenyl)-2-(methylthio)-N-phenethylth-
ieno[2,3-d]pyrimidine-6-carboxamide (15). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 18 min, 40% f 80% CH3CN at a flow
rate of 1 mL/min, tR ) 14.7 min) indicated greater than 98% purity
Acknowledgment. The authors thank Dr. John Lloyd and
Dr. Sonya Hess for assistance in obtaining HRMS data. Special
thanks are given to Amanda Skoumbordis for assistance in
editing this manuscript. This work was supported by the
Intramural Research Program of the National Institute of
Diabetes and Digestive and Kidney Diseases, National Institutes
of Health.
1
by peak integration. H NMR (DMSO-d6) δ 2.59 (s, 3H), 2.81 (t,
J ) 7.2 Hz, 2H), 3.43 (q, J ) 8.4 Hz, 2H), 3.82 (s, 3H), 6.11 (br
s, 2H), 7.16-7.32 (m, 8H), 7.50 (t, J ) 7.8 Hz, 1H), 7.96 (t, J )
3 Hz, 1H); mass spectrometry (TOF) m/z ) 451.1257 (M + H+)
(theoretical 451.1262).
N-tert-Butyl-5-amino-4-(2,3-dimethoxyphenyl)-2-(methylthio)-
thieno[2,3-d]pyrimidine-6-carboxamide (16). Analysis by C8
reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 16 min, 35% f 95% CH3CN
at a flow rate of 1 mL/min, tR ) 14.3 min) indicated greater than
93% purity by peak integration. 1H NMR (CDCl3) δ 1.45 (s, 9H),
2.66 (s, 3H), 3.76 (s, 3H), 3.95 (s, 3H), 5.77 (br s, 2H), 6.91 (dd,
J ) 1.3, 7.5 Hz, 1H), 7.21 (t, J ) 8.2 Hz, 1H); mass spectrometry
(TOF) m/z ) 433.1363 (M + H+) (theoretical 433.1368).
N-tert-Butyl-5-amino-4-(2-fluoro-3-methoxyphenyl)-2-(meth-
ylthio)thieno[2,3-d]pyrimidine-6-carboxamide (17). Analysis by
C8 reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 15 min, 35% f 90% CH3CN
at a flow rate of 1 mL/min, tR ) 11.0 min) indicated greater than
92% purity by peak integration. 1H NMR (CDCl3) δ 1.44 (s, 9H),
2.63 (s, 3H), 3.95 (s, 3H), 5.79 (br s, 2H), 6.98 (dt, J ) 7.5, 1.8
Hz, 1H), 7.15 (dt, J ) 8.1, 1.8 Hz, 1H), 7.24 (t, J ) 7.5 Hz, 1H);
mass spectrometry (TOF) m/z ) 421.1163 (M + H+) (theoretical
421.1168).
N-tert-Butyl-5-amino-4-(3-fluorophenyl)-2-(methylthio)thieno-
[2,3-d]pyrimidine-6-carboxamide (18). Analysis by C8 reversed-
phase LC-MS using a linear gradient of H2O with increasing
amounts of CH3CN (0 f 15 min, 45% f 90% CH3CN at a flow
rate of 1 mL/min, tR ) 11.4 min) indicated greater than 92% purity
by peak integration. 1H NMR (CDCl3) δ 1.54 (s, 9H), 2.68 (s, 3H),
7.19-7.50 (m, 4H); mass spectrometry (TOF) m/z ) 391.1072 (M
+ H+) (theoretical 391.1057).
Supporting Information Available: Characterization of all
synthetic compounds and biological constructs, docking procedures,
detailed descriptions of the docking results, and full-page pictures
of 3, 5, and 7 at the TSHR and LHCGR. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) de Roux, N.; Doeker, B.; Milgrom, E. In Hormone Signaling; Goffin,
V., Kelly, P. A., Eds.; Kluwer Academic Publishers: Norwell, MA,
2002; p 199.
(2) Millar, R. P.; Lu, Z.-L.; Pawson, A. J.; Flanagan, C. A.; Morgan,
K.; Maudsley, S. R. Gonadotropin-Releasing Hormone Receptors.
Endocr. ReV. 2004, 25, 235-275.
(3) Szkudlinski, M. W.; Fremont, V.; Ronin, C.; Weintraub, B. D.
Thyroid-Stimulating Hormone and Thyroid-Stimulating Hormone
Receptor Structure-Function Relationships. Physiol. ReV. 2002, 82,
473-502.
(4) Cooper, D. S. Drug Therapy: Antithyroid Drugs. N. Engl. J. Med.
2005, 352, 905-917.
(5) Guo, T. Small Molecule Agonists and Antagonists for the LH and
FSH Receptors. Expert Opin. Ther. Pat. 2005, 15, 1555-1564.
(6) van Straten, N. C. R.; van Berkel, T. H. J.; Demont, D. R.; Karstens,
W.-J. F.; Merkx, R.; Oosterom, J.; Schulz, J.; van Someren, R. G.;
Timmers, C. M.; van Zandvoort, P. M. Identification of Substituted
6-Amino-4-phenyltetrahydroquinoline Derivatives: Potent Antago-
nists for the Follicle-Stimulating Hormone Receptor. J. Med. Chem.
2005, 48, 1697-1700.
(7) Maclean, D.; Holden, F.; Davis, A. M.; Scheuerman, R. A.; Yanofsky,
S.; Holmes, C. P.; Fitch, W. L.; Tsutsui, K.; Barrett, R. W.; Gallop,
M. A. Agonists of the Follicle Stimulating Hormone Receptor from
an Encoded Thiazolidinone Library. J. Comb. Chem. 2004, 6, 196-
206.
(8) van Straten, N. C. R.; Schoonus-Gerritsma, G. G.; van Someren, R.
G.; Draaijer, J.; Adang, A. E. P.; Timmers, C. M.; Hanseen, R. G. J.
M.; van Boeckel, C. A. A. The First Orally Active Low Molecular
Weight Agonists for the LH Receptor: Thienopyr(im)idines with
Therapeutic Potential for Ovulation Induction. ChemBioChem 2002,
3, 1023-1026.
N-tert-Butyl-5-amino-4-(3-hydroxyphenyl)-2-(methylthio)-
thieno[2,3-d]pyrimidine-6-carboxamide (19). Analysis by C8
reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 10 min, 25% f 90% CH3CN
at a flow rate of 1 mL/min, tR ) 10.1 min) indicated greater than
92% purity by peak integration. 1H NMR (CDCl3) δ 1.45 (s, 9H),
2.64 (s, 3H), 5.98 (br s, 2H), 7.02 (d, J ) 7.2 Hz, 2H), 7.12 (d, J
) 7.5 Hz, 1H), 7.39 (t, J ) 7.8 Hz, 1H); mass spectrometry (TOF)
m/z ) 389.110 (M + H+) (theoretical 389.1106).
N-tert-Butyl-5-(-dimethylamino)-4-(3-methoxyphenyl)-2-(me-
thylthio)thieno[2,3-d]pyrimidine-6-carboxamide (20). Analysis
by C8 reversed-phase LC-MS using a linear gradient of H2O with
increasing amounts of CH3CN (0 f 5 min, 50% f 90% CH3CN,
5 f 15 min, 90% CH3CN at a flow rate of 1 mL/min, tR ) 7.4
min) indicated greater than 93% purity by peak integration. 1H NMR
(CDCl3) δ 1.43 (s, 9H), 2.37 (s, 6H), 2.63 (s, 3H), 3.84 (s, 3H),
7.03 (d, J ) 8.4 Hz, 1H), 7.09-7.11 (m, 2H), 7.38 (t, J ) 8.1 Hz,
1H), 7.48 (br s, 1H); mass spectrometry (TOF) m/z ) 431.1553
(M + H+) (theoretical 431.1575).
Tissue Culture and cAMP Assay. Cells were cultured for 48 h
in 24-well plates before incubation for 1 h in serum-free DMEM
containing 1 mM 3-isobutyl-1-methylxanthine (IBMX) (Sigma) and
bovine TSH (1.8 µM) (Sigma) or human LH (1000 ng/mL) (Dr.
A. Parlow, NIDDK National Hormone and Pituitary Program) or
compounds 3-19 (0-100 µM) in a humidified 5% CO2 incubator.
Following aspiration of the medium after incubation with com-
pounds, cells were lysed using lysis buffer 1 of the cAMP Biotrak
enzyme immunoassay (EIA) system (Amersham Biosciences). The
cAMP content of the cell lysate was determined using the
manufacturer’s protocol. The efficacy of receptor activation by
small-molecule modulators is expressed as the percent of maximum
response of LHCGR or TSHR to LH or TSH, respectively. The
(9) Ja¨schke, H.; Neumann, S.; Moore, S.; Thomas, C. J.; Colson, A.-O.;
Costanzi, S.; Kleinau, G.; Jiang, J.-K,; Paschke, R.; Raaka, B. M.;
Krause, G.; Gershengorn, M. C. A Low Molecular Weight Agonist
Signals by Binding to the Trnasmembrane Domain of Thyroid-
Stimulating Hormone Receptor (TSHR) and Luteinizing Hormone/
Chorionic Gonadotropin Receptor (LHCGR). J. Biol. Chem. 2006,
281, 9841-9844.
(10) Kappe, C. O. Recent Advances in the Biginelli Dihydropyrimidine
Synthesis. New Tricks from an Old Dog. Acc. Chem. Res. 2000, 33,
879-888.
(11) The synthesis of tert-butyl-2-mercaptoacetate was accomplished via
procedures reported by Woulfe and Miller: Woulfe, S. R.; Miller,
M. J. The synthesis of substituted [[3(S)-(acylamino)-2-oxo-1-
azetidinyl]thio]acetic acids. J. Org. Chem. 1986, 51, 3133-3139.
The general procedure involves the addition of potassium ethylxan-
thate to tert-butyl chloroacetate in acetone to afford O-ethyl S-(tert-
butoxycarbonyl)methyl dithiocarbonate in 97% yield. Subsequent
treatment with ethanolamine provided tert-butyl-2-mercaptoacetate
in 34% yield.
(12) General experimental procedures and spectroscopic characterization
for all intermediates are reported in Supporting Information. A benzyl
deprotection was necessary for the complete synthesis of 19.
Additional evaluation of the purity of 3-20 was judged by HPLC
analysis of individual compounds within two dissimilar mobile phase
gradients. A table describing the results is reported in Supporting
Information.

3896 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 13
Moore et al.
(13) Libert, F.; Lefort, A.; Gerard, C.; Parmentier, M.; Perret, J.; Ludgate,
M.; Dumont, J. E.; Vassart, G. Cloning, sequencing and expression
of the human thyrotropin (TSH) receptor: Evidence for binding of
autoantibodies. Biochem. Biophys. Res. Commun. 1989, 165, 1250-
1255.
(14) Schulz, A.; Schoneberg, T.; Paschke, R.; Schultz, G.; Gudermann,
T. Role of the Third Intracellular Loop for the Activation of
Gonadotropin Receptors. Mol. Endocrinol. 1999, 13, 181-190.
(15) Kleinau, G.; Ja¨schke, H.; Neumann, S.; La¨ttig, S.; Paschke, R.;
Krause, G. Identification of a Novel Epitope in the Thyroid-
Stimulating Hormone Receptor Ectodomain Acting as Intramolecular
Signaling Interface. J. Biol. Chem. 2004, 279, 51590-51600.
(16) Abagyan, R.; Totrov, M.; Kuznetsov, D. ICM. A New Method For
Protein Modeling and Design: Applications to Docking and Structure
Prediction from the Distorted Native Conformation. J. Comput. Chem.
1994, 15, 488-506.
(17) Ballesteros, J. A.; Weinstein, H. Integrated Methods for Modeling
G-Protein Coupled Receptors. Methods Neurosci. 1995, 25, 336-
428.
(18) Costanzi, S.; Mamedova, L.; Gao, Z.-G.; Jacobson, K. A. Architecture
of P2Y Nucleotide Receptors: Structural Comparison Based on
Sequence Analysis, Mutagenesis, and Homology Modeling. J. Med.
Chem. 2004, 47, 5393-5404.
JM060247S