Selection, synthesis, and structure-activity relationship of tetrahydropyrido[4,3-d]pyrimidine-2,4-diones as human GnRH receptor antagonists

Article abstract of DOI:10.1016/j.bmc.2007.05.029

The present article describes a selection of a new class of small molecule antagonists for the h-GnRH receptor, their preparation, and evaluation in vitro. Three computational methods were combined into a consensus score, to rank order virtual templates. The top 5% of templates were further evaluated in silico and assessed for novelty and synthetic accessibility. The tetrahydropyrido[4,3-d]pyrimidine-2,4-dione core was selected for synthesis and evaluated in vitro. Using an array approach for analog design and synthesis, we were able to drive the binding below 10 nM for the h-GnRH receptor after two rounds of optimization.

Full text of DOI:10.1016/j.bmc.2007.05.029

Bioorganic & Medicinal Chemistry 15 (2007) 5590–5603
Selection, synthesis, and structure–activity relationship
of tetrahydropyrido[4,3-d]pyrimidine-2,4-diones
as human GnRH receptor antagonists
Marion C. Lanier,a,* Miklos Feher,b Neil J. Ashweek,a Colin J. Loweth,a
Jaimie K. Rueter,a Deborah H. Slee,a John P. Williams,a Yun-Fei Zhu,a
Susan K. Sullivana and Michael S. Browna
aDepartments of Medicinal Chemistry and Pharmacology, Neurocrine Biosciences, 12790 El Camino Real, San Diego, CA 92130, USA
bThe Campbell Family Institute for Breast Cancer Research, University Health Network, Toronto Medical Discovery Tower,
101 College Street, Suite 5-361, Toronto, Ont., Canada M5G 1L7
Received 15 January 2007; revised 8 May 2007; accepted 10 May 2007
Available online 17 May 2007
Abstract—The present article describes a selection of a new class of small molecule antagonists for the h-GnRH receptor, their prep-
aration, and evaluation in vitro. Three computational methods were combined into a consensus score, to rank order virtual tem-
plates. The top 5% of templates were further evaluated in silico and assessed for novelty and synthetic accessibility. The
tetrahydropyrido[4,3-d]pyrimidine-2,4-dione core was selected for synthesis and evaluated in vitro. Using an array approach for
analog design and synthesis, we were able to drive the binding below 10 nM for the h-GnRH receptor after two rounds of
optimization.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
tion with an agonist ultimately leads to down regulation
3
of the receptor. Peptide agonists such as Leuprolideꢁ
Gonadotropin-releasing hormone (GnRH), also known
as luteinizing hormone-releasing hormone (LHRH), is a
decapeptide (pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-
Gly-NH2), which is produced and secreted by the hypo-
thalamus in a pulsatile manner.1 Through interaction
with specific GnRH receptors in the pituitary, both folli-
cle-stimulating hormone (FSH) and luteinizing hormone
(LH) are released from this site. These hormones, in turn,
regulate the production of steroids and gametes.2
are commercially available. However, treatment with a
GnRH agonist initially leads to overproduction of both
FSH and LH with a concomitant ‘flare effect’, which
tends to initially exacerbate symptoms in patients. In
contrast, GnRH antagonists act immediately at the
receptor, quickly suppressing the release of FSH and
LH. A number of peptide antagonists such as Cetro-
tideꢁ are currently on the market. Due to the very
1b
low oral bioavailability of these peptides, administration
is normally via injection or depot formulation. In
response to the need for a more convenient route of
administration, intensive efforts have been initiated
toward the development of orally bioavailable small-
molecule GnRH antagonists. TAK-0134 was the most
advanced small molecule antagonist as it proceeded to
phase II clinical trials (has since been discontinued),
followed by NBI-429025 and NBI-56418 from
Neurocrine Biosciences Inc. NBI-56418 has entered
Phase II clinical trials.
A number of disease states can be controlled via the reg-
ulation of this pituitary–gonadal hormone axis, in par-
ticular endometriosis, uterine fibroids, and prostate
cancer. Interestingly, suppression of both FSH and
LH production can be achieved via the agonism or
antagonism of the GnRH receptor as continuous activa-
Keywords: Gonadotropin; GnRH; LHRH; Small molecule antagonist;
Tetrahydropyrido[4,3-d]pyrimidine-2,4-dione.
*
Corresponding author. Tel.: +1 858 617 7685; fax: +1 858 617
7619; e-mail addresses: mlanier@neurocrine.con; mfeher@uhnres.
utoronto.ca
Several distinct classes of small molecule GnRH anta-
gonists have been reported in the literature in the past
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.05.029

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5591
few years.6 Takeda has described benzodiazepines, spiroam-
ines, thienopyridinones, and thienopyrimidinones. 2-
Aryltryptamines and 3-arylquinolones apparently based
compounds that were described by Merck. Pfizer/Agou-
ron patents described a series of 2-furancarboxamides.
Neurocrine’s effort has been focused mainly around 5-
aryluracils, and pyrrolo or imidazopyrimidinones.
Although there are exceptions, most of these molecules
have similar requirements: a basic protonatable nitro-
gen, one or two aromatic groups, and an aliphatic lipo-
philic group. Our goal was to find a new class of small
molecule antagonists with a distinct SAR and to use
in-house computational tools to help select the best tem-
plate. Using the consensus scoring method7 which uti-
lized both in-house and published data; we evaluated
roughly one hundred different templates via computa-
tional analysis as potential GnRH antagonists. These
templates were proposed by medicinal chemists based
on their knowledge of the GnRH literature and had to
fill two requirements: have three points of diversity
and be synthetically amenable to combinatorial chemis-
try. These templates ranged from conservative to more
speculative ones. Those with the best scores were pre-
pared at the bench and tested against the human GnRH
receptor. In this paper, we describe one of them, the tet-
rahydropyrido[4,3-d]pyrimidine-2,4-dione, its in silico
evaluation, synthesis, and biological binding with regard
to the human GnRH receptor.
software. The individual scores were combined using
sum ranks (ranking compounds on each property and
adding these ranks) as well as logistic regression with a
logistic curve fitted to the data and coefficients being
determined from a second training set. These methods
were trained on a set containing 100 actives and 1000
inactives (the fingerprint methods only used information
from the active sets) and validated using 200 actives and
1500 inactives. The logistic regression coefficients were
determined on a separate training set, containing 100 ac-
tives and 900 inactives. The training and validation sets
were chosen using diverse subset selection from a collec-
tion of molecules synthesized for GnRH binding within
Neurocrine or described in the literature, and which
contained a diverse set of structural motifs.7
The consensus scoring method was used to select tem-
plates from 101 core structures proposed by medicinal
chemists. Figure 1 includes a few examples of proposed
templates.
For each template, virtual libraries were generated and
the products were scored. In order to maximize the
probability of finding actives, the top scoring templates,
as opposed to the top scoring individual compounds,
were identified. The top 5% of products based on con-
sensus score were examined and five cores with the most
examples in this set were chosen for further evaluation.
In the second stage of the process, libraries around these
five templates using readily available side chains were
generated. The scoring process proceeded as before.
Based on the computational results, as well as intellec-
tual property and synthetic considerations, the tetrahy-
2. Computational methods
Ligand-based consensus scoring of virtual libraries was
applied to select and rank order proposed templates.
The details of this process, such as the description of
the individual methods applied, their combination into
a consensus scoring method, and the evaluation of their
performance, are given elsewhere.7 Briefly, during the
template selection, the following methods were com-
bined into a consensus score: MACCS,8 TGT9, and
MP6110 fingerprints, encapsulating structural, pharma-
cophore, and gross property information of the mole-
cule, respectively, as implemented in the MOE
modeling suite,9 BCUT descriptors,11,12 as implemented
within the diverse solutions (DVS) software13, and 3D
pharmacophores using the Catalyst14 and CombiCode15
dro[4,3-d]pyrimidine-2,4-dione template
selected as the first library for synthesis.
VI
was
In order to establish similarities and differences in bind-
ing in comparison to uracil-containing GnRH antago-
nists, flexible alignments16 were performed, as
implemented in the MOE suite of programs.9 In sum-
mary, different conformations and alignment poses of
the molecules were randomly generated and evaluated
using a scoring function that includes Gaussian distance
dependence for pharmacophore and volume-like terms,
as well as the relative energy of the given conformation.
It was shown that such alignments reproduce well the
O
O
O
O
O
O
R3
R1
O
R1
O
R1
R3
III
R3
N
N
N
N
N
N
N
O
N
R2
R2
R2
II
I
O
O
O
R1
R3
R3
O
R1
N
R3
N
N
N
N
R1
N
N N
R2
R2
R2
VI
IV
V
Figure 1. Examples of proposed templates.

5592
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
relative orientation and conformations of molecules in
co-crystallized ligands.16
3. Chemistry
There are two regioisomers for the selected template, the
5,6,7,8-tetrahydropyrido[3,4-d]pyrimidine-2,4-dione VI
and the 5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine-2,4-
dione VII (Fig. 3). Both were prepared using similar
routes and evaluated.
The alignments proceeded in the following manner.
First a stochastic conformational search was performed
on the template molecule, NBI-42902,17 using the Merck
Force field and continuum solvation, as implemented in
MOE. The conformations of this molecule were then
clustered and since the lowest energy cluster was well
separated from any other (>3 kcal/mol), only the lowest
energy conformer was used as a template to which other
GnRH compounds were aligned. This lowest energy
conformer had very similar geometry to a crystallized
close analog of NBI-42902 (unpublished results).
Although based on the validation16 the top scoring
alignment solution is expected to reproduce the experi-
mental binding mode best, for 10 randomly selected
5,6,7,8-tetrahydro[4,3-d]pyrimidine-2,4-diones the top-
scoring 50 solutions were visually clustered employing
metric scaling.18 Using this method it was shown that
all top scoring alignments represented similar binding
modes regarding the relative position of the major
pharmacophoric groups, common among the majority
of GnRH antagonists. The alignment results are
exemplified by 6-benzyl-1-(2,6-difluoro-benzyl)-3-{2-
[methyl-(2-pyridin-2-yl-ethyl)-amino]-ethyl}-5,6,7,8-tetrahy-
dro-1H-pyrido[4,3-d]pyrimidine-2,4-dione (34), shown
in Figure 2.
For regioisomer VI, when R3 was a benzyl or a phenyl,
the commercially available 1-benzyl-4-ethoxycarbonyl-
3-piperidone hydrochloride 1 was treated with ammo-
nium acetate for an hour at room temperature to give
the corresponding enamine 2. Reaction of 2 with various
isocyanates followed by a ring-closure under basic con-
ditions gave the 3-substituted-7-benzyl-5,6,7,8-tetrahy-
dro-1H-pyrido[3,4-d]pyrimidine-2,4-dione 3. 3 was then
alkylated at the N-1 position with various halides under
basic conditions to give 4 which underwent hydrogena-
tion to deprotect the amine at the 7 position. Treatment
of 5 with various halides, acid chlorides, and aldehydes
gave the final compounds 6 (Scheme 1). Similar syn-
thetic schemes were used to prepare regioisomer VII,
starting from the commercially available 1-benzyl-3-car-
bethoxy-4-piperidone hydrochloride (not shown).
When R3 was a substituted ethyl amine, the 4-oxo-piper-
idine-3-carboxylic acid ethyl ester 7 was Boc-protected on
nitrogen. 8 underwent the same sequence of steps
described in Scheme 1 using allyl isocyanate to give 11.
Boc-protection instead of benzyl was necessary to keep
the integrity of the amine at the 7 position during the allyl
oxidation step. The allyl group of 11 was oxidized by
treatment with OsO4/2,6-lutidine followed by NaIO4 to
form the intermediate aldehyde 12. Reductive amination
of 12 using standard conditions gave 13. Boc-deprotec-
tion followed by alkylation, acylation or reductive amina-
tion gave the final compounds 15 (Scheme 2). To allow for
more diversity at the 3 position, the p-methoxybenzyl
(PMB) isocyanate was used as a protecting group. Chem-
istry proceeded as previously to give compound 17. After
switching protecting groups at the 6 position, the PMB
group was removed in the presence of aluminum chloride
in DCM and R3 was installed via a Mitsunobu displace-
ment. Finally, the trifluoroacetate was removed from 19
and R1 installed at the 6 position by alkylation, acylation
or reductive amination. The order of steps was modified
depending on the desired point of diversity (Scheme 3).
4. Results and discussion
Initially, 8 libraries were prepared, 4 with each regio-
isomer VI and VII (Table 1). Within each library, R3
and R2 were kept constant and R1 was varied. Because
O
O
4
5
5
8
4
R3
R3
O
R1
N
6
N
N
3
3
2
6
and
2
O
N
7
R1
N
7
N
1
1
Figure 2. Flexible alignment of a tetrahydro[4,3-d]pyrimidine-2,4-
dione (compound 34) with a uracil-containing GnRH antagonist
(NBI-42902). The five major pharmacophoric groups, at least four of
which are found in the majority of GnRH antagonists, are also
displayed.
8
R2
R2
VI
VII
Figure 3. Templates VI and VII.

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5593
O
O
O
R3
O
R2-X
NH4OAc
HCl
Ph
1. R3-NCO/toluene
OEt
NH2
OEt
N
Ph
N
N
Ph
N
NaH, DMF, r.t., 16h
2. NaOMe/MeOH
r.t., 45min.
r.t., 1h, EtOH
O
O
N
H
3
2
1
O
O
N
R3
R3
O
R3
O
RX, RCHO
H2, Pd/C
N
N
N
Ph
N
HN
N
or
RCOCl
N
O
N
R1
iPrOH, 18h
R2
R2
R2
4
5
6
Scheme 1. Synthesis of template VI when R3 is a phenyl or benzyl.
1. CH2=CHCH2NCO
toluene, 55oC, 20h
HCl
CO2Et
NH2
CO2Et
CO2Et
O
NH4OAc
Boc-N
Boc2O, NaHCO3
60oC, 4h
Boc-N
HN
2. NaOMe, MeOH
r.t., 1h
O
EtOH, r.t., 2.5h
9
8
7
O
H
O
O
N
1. OsO4, NMO
dioxane/water
N
O
R5
R2-X
NaH, DMF, r.t., 16h
R4
Boc-N
N
Boc-N
N
Boc-N
NaBH(OAc)3
r.t, 16h
N
O
2. NaIO4
N
O
N
H
O
R2
R2
12
10
11
O
R5
N
O
R5
N
O
R5
N
R1
RX, RCHO
TFA/DCM
Boc-N
R4
N
N
N
R4
HN
N
R4
N
O
or
RCOCl
N
O
N
O
R2
R2
R2
14
13
15
Scheme 2. Synthesis of template VII when R3 is an ethylamine.
O
O
O
1. PMB-NCO
1. TFA
R2-X
NaH, DMF
Boc-N
N
toluene, 80oC
Boc-N
N
Boc-N
OEt
NH2
O
N
O
2. TFAA, DIEA
O
O
N
O
2. NaOMe, MeOH
H
R2
9
16
17
O
O
O
O
1. AlCl3, CH2Cl2
R3
O
R1
1. K2CO3, MeOH
2. R1-X
R3
F
F
N
N
N
N
N
N
F
F
F
F
N
2. PPh3, DtBAD, R3-OH
O
N
O
N
O
R2
18
19
20
R2
R2
1. K2CO3, MeOH
2. R1-X
O
1. AlCl3, CH2Cl2
2. PPh3, DtBAD, R3-OH
R1
N
N
O
N
O
R2
21
Scheme 3. Synthesis of template VII via a PMB protecting group.
of the good overlap between the alignment models and
the 5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine-2,4-dione
core, R2 and R3 at position 1 and 3, respectively, were
selected from groups that had shown binding in previ-
ous uracil GnRH series.6 R1 was a diversity set of 36
reagents leading to aliphatic and aromatic substituted
alkyl and acyl groups (Fig. 4). The diversity set was gen-
erated by computational methods (using Tanimoto sim-
ilarity on MACCS fingerprints) to avoid bias toward
former known SAR.

5594
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
Table 1. Eight libraries prepared in the first round
Since the benzylic substitution gave the most active
compounds in the first round, a library of 30 compounds
around template VIA was prepared with various benz-
ylic, heteroaromatic, and lipophilic groups. The substi-
tuted benzyls or benzyl replacements did not help to
gain additional interactions with the receptor when com-
pared to the unsubstituted benzyl 23. The overall bind-
ing of the molecule was not significantly affected by
the position and/or the nature of the benzyl substituents.
VII
VI
O
O
R
N
N
N
O
O
A
B
N
R
N
O
N
O
F
F
F
F
4.2. Library VIID
O
O
R
N
N
N
O
O
Analogs in the VIID library showed a similar trend to
library VIA. Small or polar alkyl groups and acyl
groups at the 6 position failed to generate tight binders
at the human GnRH receptor but aromatic groups im-
proved the binding substantially. Adding a para-substi-
tuent such as a methyl group to the benzyl moiety
(compound 37) boosted the binding more than 20-fold
(35 nM). Following the same approach used for Library
VIA, a more focused library was prepared around the
R1 region of VIID, using systematic screening of substi-
tuted benzyls, heterocycles, and non-aromatic lipophilic
groups. A total of 45 compounds were prepared. Some
key examples are shown in Table 3.
N
R
N
O
N
O
N
N
N
O
N
F
O
N
F
N
R
N
C
D
F
O
N
R
F
O
N
O
N
F
N
O
N
R
N
N
N
N
N
N
Regarding the benzylic substitution, para and/or ortho
groups (compounds 37, 40–47, 49) improved the bind-
ing. The same substitution at the meta position had no
effect on the binding compared to the unsubstituted ben-
zyl (48, 850 nM). The 4-chloro benzyl compound (42)
had a Ki of 17 nM compared to 990 and 850 nM,
respectively, for the corresponding unsubstituted phenyl
compound (34) and the 3-chloro compound (48). The
4-methyl substitution was twofold better than the 2-
methyl (37: 35 nM vs 49: 85 nM, respectively). The
2,4-disubstitution pattern was found optimal with the
2-methyl-4-chloro (51) and 2,4-dimethyl (52) both hav-
ing a binding affinity of 5 nM against the h-GnRH
receptor. The 4-methyl (37) and 4-ethyl (40) compounds
had similar binding with 35 and 20 nM, respectively.
When the steric bulk was increased more, the binding
started to decrease as illustrated with 4-tert-butyl com-
pound (41, 300 nM). Increased polarity at the para posi-
tion also had a negative impact on the binding as shown
with the 4-methylsulfone (46, 805 nM) and the 4-meth-
oxy (43, 100 nM), although almost any substituent at
the para position showed improved affinity over the par-
ent benzyl compound. Strongly electron-deficient
groups such as 4-trifluoromethyl (45, 30 nM) or elec-
tron-donating groups such as 4-methoxy (43, 100 nM)
did not have a critical impact on binding to the receptor.
The best substitutions were weakly activating and deac-
tivating groups (methyl and chloro) strongly suggesting
that a favored ring orientation or shape (2 and/or 4 sub-
stitution pattern) rather than a p–p interaction was crit-
ical to a good receptor pocket fit. Heterocycles and
bicyclic systems tested were in most cases not as tight
binding as the best substituted benzyls. The 3-methyl
benzothiophene 53 (15 nM) was the exception and was
comparable to the 2,4-dimethyl benzyl 52 (5 nM) and
4-chloro-2-methyl benzyl 51 (5 nM). These results were
much more encouraging than for library VIA. We
N
R
N
F
O
O
F
F
The compounds synthesized were evaluated for their
ability to inhibit [125I-His5, D-Tyr6]GnRH agonist bind-
ing to the cloned human GnRH-receptor as previously
described.19 Out of the eight libraries prepared and
tested, VIA and VIID produced active compounds.
SAR of libraries VIA and VIID were different suggest-
ing a possible difference in binding modes. Both tem-
plates VIA and VIID have three features aligning
perfectly with NBI-42902: the two acceptors and the
aromatic interactions. Template VIID has a fourth
interaction (positive/donor) compared to VIA which
should be reflected in the binding data. Indeed in the
first set prepared, library VIA showed at best binding
in the micromolar range but library VIID had com-
pound 37 with a binding Ki of 35 nM. Some examples
are shown in Table 2.
4.1. Library VIA
The unsubstituted core (24) showed modest binding at
970 nM. None of the small or polar alkyl and acyl side
chains (not shown) showed binding at the h-GnRH
receptor above 30% at a concentration of 10À6 M. When
R1 contained an aromatic group, the binding ranged
from 500 to 2200 nM or worse. These data suggested
that the core amine was essential to maintain the inter-
action with the receptor and a lipophilic and/or aro-
matic interaction was tolerated (Table 2). To optimize
the series, further exploration was done around R1.

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5595
O
N
H
O
O
N
N
O
O
O
O
O
O
O
N
O
O
N
N
O
N
O
O
O
O
O
O
N
O
N
O
O
N
N
O
N
N
O
O
O
O
O
S
Figure 4. R1 diversity set generated by computational methods.
Table 2. Modification of the R1 group
decided to focus our efforts on library VIID and explore
the substitutions at the 1 and 3 positions (R2 and R3).
O
O
R1
N
N
N
Based on flexible alignments (Fig. 1), it was assumed that
the R2 region of VIID should contain an aromatic or
hydrophobic group. Over 50 compounds were prepared
using the methodology described in Scheme 2 to test this
hypothesis. Representatives of this exploration are dis-
closed in Table 4. When R2 was hydrogen (55) or methyl
(56), no binding was detected. Only the benzyl (58), pref-
erably ortho substituted (53, 59, 60), and the cyclohexyl
(57) showed competitive binding against the human
GnRH receptor, highlighting an important lipophilic
interaction. 2,6-Disubstitutions gave the best binding
with a 20-fold increase in binding compared to the unsub-
stituted phenyl (58, 305 nM compared to 60, 10 nM).
From these results, it appeared that in this region of the
molecule, the SAR of the VIID series mostly paralleled
that of the uracil antagonist series.20 Not only was lipo-
philicity important but the shape, orientation, and elec-
tron-deficiency were key factors for the interaction with
the binding pocket of the receptor. This supported the
alignment suggested by our computational studies where
the uracil moieties overlap and the piperidine ring was
flanking the pyridinedione core. It was worth noting that
cyclohexyl was well tolerated and equipotent to benzyl in
the present series. In the uracil series however, there was a
fourfold binding difference in favor of the benzyl.20
N
O
N
N
N
F
O
R1
N
F
O
F
F
VIID
VIA
R1
Compound Ki
(nM)
Compound Ki
(nM)
23
24
25
26
1290
970
34
33
35
36
990
H
N
>10,000
>10,000
515
1580
2200
H
N
27
28
>10,000 37
35
540
980
38
39
>10,000
N
N
Our last focus was to modify the R3 group on the right
hand side of the templates (Table 5). Based on the align-
ment, the positive/donor interaction was in the correct re-
gion but the ethyl pyridine did not align well with the
reference model and was not in a specific interaction with
the receptor. A library was generated with two points of
diversity: an amine linker and a substituent on that amine.
29
NA
O

5596
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
Table 3. Modification of the R1 group of template VIID
For the linker, substituted and unsubstituted ethyl
amines, methyl 2-, 3-, and 4-piperidines (R and S), 2-,
3-, and 4-piperidines, 2-methyl morpholine (RS), methyl
2- and 3-pyrrolidines (R and S), 2- and 3-pyrrolidines
(S, RS), 2-methyl azetidine (RS) and 1-tetrahydroiso-
quinolines (R) were used. The amine substitutions were
H, methyl, isobutyl, cyclopentyl, benzyl, 2-methyl pyri-
dine, and 2-ethyl pyridine. Amine functionality replace-
ments such as amide, ester, cyanoguanidine, or amidine
moieties were also considered. Over 100 compounds
(not shown) were prepared and tested. Replacing the
amine with any other functionality did not prove to be a
successful approach. The ethyl was still the best linker.
The 2-methyl piperidine (RS) and 2-methyl pyrrolidine
(R) were the next best with a loss of at least twofold bind-
ing. The optimal distance between the core and the basic
center was the two-carbon linker. For example, extending
to three carbons as in 71c, the binding dropped 20-fold
compared to the two-carbon spaced analog 71b. The ste-
reochemistry was also crucial. The S isomer of 71d was
1370 nM compared to 20 nM for the R isomer. This
observation was found to be general and independent of
the ring size. The R-phenylglycine compound was also
made as it was one of the best side chains in the uracil ser-
ies.21 In the present series, its binding was a modest 45 nM
(66e). Regarding the amine substitutions, a lipophilic
interaction helped substantially and in most cases benzyl
and pyridines were best. Compound 72d was tested at
10 nM against the human GnRH receptor, only twofold
less than 52.
O
R1
N
N
N
N
N
F
O
F
VIID
Compound
R1
Ki (nM)
40
20
41
42
43
44
45
46
47
48
49
50
51
52
53
54
300
17
Cl
100
10
O
S
F
F
30
F
In summary, we successfully identified and synthesized a
novel series of h-GnRH antagonists, the 5,6,7,8-tetrahy-
dropyrido[4,3-d]pyrimidine-2,4-dione, using a computa-
tionally driven approach. We have demonstrated that
using a consensus scoring method, we could successfully
identify a new template which included key features
present in most GnRH series, but had only partial cor-
relation with the SAR of other known small molecule
GnRH antagonists. Template VIA had only three key
features out of the four usually required. The binding
for this series plateaued around one micromolar. When
an additional feature was present like in series VIID, the
binding was improved substantially. Using an array
approach, we were able to explore all three points of
diversity of the template and rapidly produce low
nanomolar binding compounds (below 10 nM).
805
15
S
O
O
S
850
85
Cl
255
5
Cl
Cl
Cl
5. Experimental
5.1. General methods and materials
5
Reagents, starting materials, and solvents were purchased
from commercial suppliers and used as received. Concen-
tration refers to evaporation under vacuum using a Buchi
¨
rotary evaporator. Reaction products were purified, when
necessary, by chromatography on silica gel (40–63 lm)
with the solvent system indicated. Nuclear magnetic reso-
nance data were recorded on a Varian Mercury 300 MHz
Spectrometer using TMS as the internal standard. Com-
pounds that were prepared in a library format were puri-
fied by HPLC using mass collection and their purities
15
S
N
>10,000
N

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5597
Table 4. Modification of the R2 group
F
CF3
O
N
N
N
H
Me
S
N
N
O
F
F
F
R2
Compound
55
56
57
58
59
53
60
Ki (nM)
>10,000
>10,000
380
305
80
15
10
Table 5. Modification of the R3 group
N
R
H
Me
N
O
N
N/A
67a*
68a
N/A
N/A
71a
52
N
N
N
R
O
F
Ki
1400
1800
180
5
F
O
R
N
66b
>10000
66c
67b
1710
67c
68b
2590
68c
69b
520
70b
945
70c
160
N/A
71b
25
72b**
N
N
N
O
F
(R) 30
Ki
Ki
F
(S) 2410
O
N
69c
2580
71c
490
72c
R
N
N
N
O
F
2450
>10000 >10000
600
F
Cl
O
66d
N/A
N/A
69d
1370
N/A
71d
72d
10
R
N
N
N
N
O
F
(R) 20
Ki
Ki
>10000
F
(S) 1370
Cl
O
66e
67e
N/A
N/A
N/A
N/A
N
N
N
O
N
R
45 nM
70 nM
F
F
*4-Ethyl benzyl instead of 2,4-dimethyl benzyl.
**4-Chlorobenzyl instead of 2,4-dimethyl benzyl.
checked by LC–MS. All the compounds having a UV pur-
ity above 85% at two different wavelengths (220 and
254 nM) were tested in a biological assay. Analytical
HPLC-MS Method 1 was run on an Agilent 1100 series
platform equipped with an auto-sampler, a UV detector
(220 and 254 nM), an MS detector (APCI); HPLC
column: YMC ODS AQ, S-5, 5 l, 2.0 · 50 mm cartridge;
HPLC gradient: 1.0 mL/min, from 10% acetonitrile in
water to 90% acetonitrile in water in 2.5 min, maintaining
90% for 1 min. Both acetonitrile and water have 0.025%
TFA. Analytical HPLC-MS Method 2 was run on an Agi-
lent 1100 equipped with an auto-sampler, an UV detector

5598
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
(220 and 254 nM), a MS detector (APCI); HPLC column:
Phenomenex Synergi-Max RP, 2.0 · 50 mm column;
HPLC gradient: 1.0 mL/min, from 5% acetonitrile in
water to 95% acetonitrile in water in 13.5 min, maintain-
ing 95% for 2 min. Both acetonitrile and water have
0.025% TFA. Analytical HPLC-MS Method 3 was run
on an Agilent 1100 series platform equipped with an
auto-sampler, a UV detector (220 and 254 nM), an MS
detector (APCI); HPLC column: Phenomenex Synergi-
Max RP, 2.0 · 50 mm column; HPLC gradient: 1.0 mL/
min, from 5% acetonitrile in water to 95% acetonitrile in
water in 13.5 min, maintaining 95% for 2 min. Both aceto-
nitrile and water have 0.025% TFA. Analytical HPLC-
MS Method 4 was run on an Agilent 1100 series platform
equipped with an auto-sampler, a UV detector (220 and
254 nM), an MS detector (APCI), and Berger FCM
1200 CO2 pump module; HPLC column: Berger Pyridine,
PYR 60A, 6 l, 4.6 · 150 mm column; HPLC gradient:
4.0 mL/min, 120 bar; from 10% methanol in supercritical
CO2 to 60% methanol in supercritical CO2 in 1.67 min,
maintaining 60% for 1 min. Methanol has 1.5% water.
Backpressure regulated at 140 bar. Analytical HPLC-
MS Method 5 was run on a Dionex platform equipped
with an autosampler, a UV detector (220 and 254 nM),
an MS detector (APCI); HPLC column: Phenomenex
CX18 4.6 · 150 mm; HPLC gradient: 95% 0.04%
NH4OH/H2O to 90% 0.04% NH4OH/ACN over
9.86 min, 12.30 min run. The binding assay was per-
formed following the method described in reference 19.
Each compound was tested at least twice.
triturated with methanol (80 mL) to yield 1-benzyl-5-
[3-(3-methoxy-phenyl)-ureido]-1,2,3,6-tetrahydro-pyri-
dine-4-carboxylic acid ethyl ester as a white solid
(8.0 g, 75%). To this intermediate were added methanol
(100 mL) and 30% sodium methoxide in methanol
(10 mL, 60 mmol). The resulting mixture was stirred
at room temperature overnight, after which it was con-
centrated in vacuo, resuspended in water, and precipi-
tated via pH adjustment to 7 with 1 N HCl. The
resulting white solid was filtered, washed with water,
and dried over anhydrous magnesium sulfate to yield
the title compound 22 (6.2 g, 88%). 1H NMR
(300 MHz, CDCl3) d: 7.40–7.30, m, 6H; 6.96, dd, 1H,
(J = 8.4 Hz and 0.9 Hz); 3.79, s, 3H; 3.70, br s, 2H;
3.24, br s, 2H; 2.76, br s, 2H; 2.52, br s, 2H. LCMS-
2: tR = 0.49 (100%); MS: m/z 363.9 [M+H]+, expected
364 [M+H]+.
5.2.3.
7-Benzyl-1-(2,6-difluoro-benzyl)-3-(3-methoxy-phe-
nyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-2,4-dione
(23). To 7-benzyl-3-(3-methoxy-phenyl)-5,6,7,8-tetrahy-
dro-1H-pyrido[3,4-d]pyrimidine-2,4-dione 22 (3.2 g,
8.8 mmol) in dimethylformamide (20 mL) was added so-
dium hydride (0.38 g, 9.7 mmol), followed by 2,6-difluo-
robenzyl bromide (2.0 g, 9.7 mmol). The resulting
solution was stirred at room temperature for 2 h, after
which it was poured into a solution of saturated aqueous
sodium bicarbonate. The resulting solids were filtered and
triturated with hexanes to yield compound 23 (3.7 g,
86%). 1H NMR (300 MHz, CDCl3) d: 7.36, t, 1H,
(J = 7.8 Hz); 7.32–7.19, m, 6H; 6.93, dd, 1H, (J = 8.4
and 2.4 Hz); 6.84, t, 2H, (J = 8.4 Hz); 6.79, d, 1H,
(J = 7.8 Hz); 6.73, t, 1H, (J = 1.5 Hz); 5.10, s, 2H; 3.79,
s, 3H; 3.70, br s, 2H; 3.41, br s, 2H; 2.74–2.70, m, 2H;
2.60–2.57, m, 2H. LCMS-3: tR = 4.88 (93%); MS: m/z
489.9 [M+H]+, expected 490 [M+H]+.
5.2. Synthesis of library VIA via Scheme 1
5.2.1. 5-Amino-1-benzyl-1,2,3,6-tetrahydro-pyridine-4-car-
boxylic acid ethyl ester (2). To a solution of 1-benz-
yl-3-oxo-piperidine-4-carboxylic acid ethyl ester
hydrochloride 1 (20 g, 67.2 mmol) in ethanol (200 mL)
was added ammonium acetate (51.8 g, 672 mmol). The
mixture was stirred for 1 h at room temperature, after
which the TLC (5% methanol in dichloromethane)
showed complete consumption of starting material. The
solvent was removed in vacuo, and the resulting residue
was partitioned between ethyl acetate and 1 N NaOH.
The aqueous layer was further extracted with dichloro-
methane and the combined organic layers were washed
with brine and dried over anhydrous magnesium sulfate,
filtered and evaporated to give a beige solid, which was
recrystallized from diethyl ether and hexanes to give the
product as an off-white solid (11.9 g, 68%). A second crop
of product (3.41 g, 19.5%) was obtained from the mother
liquor. 1H NMR (300 MHz, DMSO-d6) d: 7.20–7.40, m,
5H; 4.01, q, 2H, (J = 6.9 Hz); 3.51, s, 2H; 3.32, s, 2H;
2.93, s, 2H; 2.46–2.51, bm, 2H; 2.21, t, 2H, (J = 5.1 Hz);
1.16, t, 3H, (J = 6.9 Hz). LCMS-1: tR = 0.32 (100%);
MS: m/z 261 [M+H]+, expected 261 [M+H]+.
5.2.4. 1-(2,6-Difluoro-benzyl)-3-(3-methoxy-phenyl)-5,6,7,
8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-2,4-dione (24).
To 7-benzyl-1-(2,6-difluoro-benzyl)-3-(3-methoxy-phenyl)-
5,6,7,8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-2,4-dione
23 (3.0 g, 6.1 mmol) in isopropanol (60 mL) was added
10% palladium hydroxide on carbon (1.0 g). The resulting
mixture was placed under an atmosphere of hydrogen
(1 atm) and stirred for 12 h at room temperature. It was
then filtered through Celite and concentrated in vacuo
to yield 24 as a white solid (2.0 g, 85%). 1H NMR
(300 MHz, CDCl3) d: 7.37, t, 1H, (J = 8.4 Hz); 7.29–
7.26, m, 1H; 6.96–6.86, m, 3H; 6.79, d, 1H, (J = 7.5 Hz);
6.73, br s, 1H; 5.15, br s, 2H; 3.86, br s, 2H; 3.79, s, 3H;
3.04, br s, 2H; 2.51, br s, 2H. LCMS-4: tR = 1.82 (89%);
MS: m/z 399.8 [M+H]+, expected 400 [M+H]+.
5.3. General procedure for the alkylation of 24
To 1-(2,6-difluoro-benzyl)-3-(3-methoxy-phenyl)-5,6,7,
8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-2,4-dione 24
(50 mg, 0.13 mmol) in dimethylformamide (1 mL) were
added the appropriate alkyl halide (0.20 mmol), diisopro-
pylethylamine (50 mg, 0.39 mmol), and tetrabutylammo-
nium iodide (5.0 mg, 0.014 mmol). The resulting mixture
was stirred at 50 ꢂC for 12 h. Compounds were purified di-
rectly by preparative HPLC yielding the TFA salts of the
5.2.2. 7-Benzyl-3-(3-methoxy-phenyl)-5,6,7,8-tetrahydro-
1H-pyrido[3,4-d]pyrimidine-2,4-dione (22). To 3-amino-
1-benzyl-1,2,5,6-tetrahydro-pyridine-4-carboxylic acid
ethyl ester 2 (7.0 g, 26 mmol) in toluene (60 mL) was
added 3-methoxyphenyl isocyanate (4.6 g, 32 mmol).
The resulting mixture was stirred for 12 h at room tem-
perature after which it was concentrated in vacuo and

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5599
title compounds. Compounds 25–27 were prepared using
this procedure.
and extracted with dichloroethane (3· 150 mL). The
combined organic layers were washed with brine and
dried over anhydrous magnesium sulfate, filtered, and
evaporated to give a colorless oil which crystallized
upon standing (40.1 g, 99%). 1H NMR (300 MHz,
CDCl3) d: 4.23, q, 2H, (J = 6.9 Hz); 4.07, br s, 2H;
3.57, t, 2H, (J = 5.9 Hz); 2.37, t, 2H, (J = 5.9 Hz);
1.48, s, 9H; 1.31, t, 3H, (J = 6.9 Hz).
5.3.1. 1-(2,6-Difluoro-benzyl)-7-[pyridin-4-yl-methyl]-3-(3-
methoxy-phenyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-d]pyrim-
idine-2,4-dione (25). LCMS-3: tR = 4.12 (90%): m/z 491
[M+H]+, expected 491 [M+H]+.
5.3.2. 1-(2,6-Difluoro-benzyl)-7-[2-(1H-indol-3-yl)-ethyl]-
3-(3-methoxy-phenyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-
d]pyrimidine-2,4-dione (26). 1H NMR (300 MHz, CDCl3)
d: 8.10, br s, 1H; 7.58, d, 1H, (J = 7.8 Hz); 7.40–7.34, m,
2H; 7.28–7.11, m, 3H; 7.05, br s, 1H; 6.94, dd, 1H,
(J = 9.0 and 2.7 Hz); 6.89, t, 2H, (J = 8.4 Hz); 6.79, d,
1H, (J = 6.9 Hz); 6.74, br s, 1H; 5.11, s, 2H; 3.79, s, 3H;
3.68, br s, 2H; 3.08–2.96, m, 4H; 2.89, br s, 2H; 2.64, br
s, 2H. LCMS-3 tR = 5.36 (94%): m/z 543 [M+H]+, ex-
pected 543 [M+H]+.
5.5.2. Ethyl-4-amino-1-tert-butoxycarbonyl-1,2,5,6-tetra-
hydropyridine-3-carboxylate (9). To a solution of 1-tert-
butoxycarbonyl-3-carbethoxy-4-piperidone
8
(12.9 g,
47.9 mmol) in ethanol was added ammonium acetate
(36.9 g, 479 mmol). The mixture was stirred for 2.5 h at
room temperature, after which the TLC (50% ethyl acetate
in hexanes) showed complete consumption of starting
material. The solvent was removed in vacuo, and the result-
ing residue was partitioned between dichloroethane
(300 mL) and 1 N sodium hydroxide (100 mL). The aque-
ous layer was further extracted with dichloroethane (3·
100 mL) and the combined organic layers were washed with
brine and dried over anhydrous magnesium sulfate, filtered,
and evaporated to give ethyl-4-amino-1-tert-butoxycar-
bonyl-1,2,5,6-tetrahydropyridine-3-carboxylate 9, which
was used without purification in the following reactions
5.3.3. 1-(2,6-Difluoro-benzyl)-7-[4-methylbenzyl]-3-(3-meth-
oxy-phenyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-
2,4-dione (27). LCMS-3: tR = 5.20 (93%); MS: m/z 504
[M+H]+, expected 504 [M+H]+.
5.4. General procedure for the reductive amination of 24
1
(11.4 g, 88%). H NMR (300 MHz, CDCl3) d: 4.15, q,
To 1-(2,6-difluoro-benzyl)-3-(3-methoxy-phenyl)-5,6,7,
8-tetrahydro-1H-pyrido[3,4-d]pyrimidine-2,4-dione 24
(50 mg, 0.13 mmol) in dichloroethane (1 mL) were
added the appropriate aldehyde (0.15 mmol) and so-
dium triacetoxyborohydride (37 mg, 0.17 mmol). The
resulting mixture was stirred at room temperature
overnight, after which it was concentrated in vacuo
and redissolved in methanol (1 mL). Preparative
HPLC yielded the TFA salts of the title compounds.
Compounds 28 and 29 were prepared using this
procedure.
2H, (J = 7.2 Hz); 4.07, br s, 2H; 3.81, bm, 1H; 3.69, br s,
2H; 3.52, t, 2H, (J = 6.0 Hz); 2.28, t, 2H, (J = 6.0 Hz):
1.48, s, 9H; 1.27, t, 3H, (J = 7.2 Hz). LCMS-1: tR = 2.62
(100%); MS: m/z 271 [M+H]+, expected 271 [M+H]+.
5.5.3. 3-Allyl-2,4-dioxo-1,3,4,5,7,8-hexahydro-2H-pyri-
do[4,3-d]pyrimidine-6-carboxylic acid tert-butyl ester
(10). A solution of ethyl-4-amino-1-tert-butoxycarbonyl-
1,2,5,6-tetrahydropyridine-3-carboxylate
9
(8.7 g,
32.4 mmol), allyl isocyanate (3.6 mL, 33.7 mmol), and
diisopropylethylamine (0.9 mL, 5.6 mmol) in toluene
(50 mL) was warmed to 55 ꢂC. An additional volume of al-
lyl isocyanate (3.6 mL, 33.7 mmol) was added and stirring
at 55 ꢂC was continued for 18 h. The reaction was moni-
tored by TLC (20% ethyl acetate in hexanes) for comple-
tion. After cooling to room temperature, solvents were
evaporated, the residue was diluted with methanol
(50 mL), and a 30% weight solution of sodium methoxide
in methanol (17.8 mL, 97.2 mmol) was added carefully.
Thesolutionwasstirredatroomtemperaturefor1 h. Upon
completion, the solvent was removed in vacuo. The result-
ing residuewasdissolvedin water(200 mL), acidifiedtopH
4 with 1 N aqueous HCl. Product 10 precipitated as a pale
yellow solid (8.0 g, 93%). 1H NMR (300 MHz, CDCl3) d:
5.80–5.95, m, 1H; 5.10–5.30, dt, 2H, (J = 14.7 and 9 Hz);
4.53, d, 1H, (J = 5.7 Hz); 4.22, br s, 1H; 3.82, t, 2H,
(J = 5.7 Hz); 3.67, t, 2H, (J = 5.7 Hz); 1.48, s, 9H.
5.4.1. 1-(2,6-Difluoro-benzyl)-7-[4-imidazol-1-yl-benzyl]-
3-(3-methoxy-phenyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-
d]pyrimidine-2,4-dione (28). LCMS-3: tR = 3.50 (98%);
MS: m/z 556 [M+H]+, expected 556 [M+H]+.
5.4.2. 1-(2,6-Difluoro-benzyl)-7-(2-methoxy-benzyl)-3-(3-
methoxy-phenyl)-5,6,7,8-tetrahydro-1H-pyrido[3,4-d]pyrim-
idine-2,4-dione (29). 1H NMR (300 MHz, CDCl3) d: 7.37, t,
1H, (J = 7.8 Hz); 7.38–7.24, m, 3H; 6.98–6.90, m, 3H; 6.88,
t, 2H, (J = 8.4 Hz); 6.78, d, 1H, (J = 7.5 Hz); 6.73, t, 1H
(J = 2.1 Hz); 5.12, s, 2H; 4.03, s, 2H; 3.84, s, 3H; 3.81, br
s, 2H; 3.79, s, 3H; 2.99, br s, 2H; 2.69, bt, 2H (J = 10.5
Hz). LCMS-3: tR = 4.68 (100%); MS: m/z 520 [M+H]+, ex-
pected 520 [M+H]+.
5.5. Synthesis of library VIID via Scheme 2
5.5.4. 3-Allyl-1-(2,6-difluoro-benzyl)-2,4-dioxo-1,3,4,5,7,8-
hexahydro-2H-pyrido[4,3-d]pyrimidine-6-carboxylic acid
tert-butyl ester (30). A solution of 10 (5.5 g, 17.9 mmol) in
anhydrous dimethylformamide (50 mL) was treated with
sodium hydride (60% suspension in oil, 0.64 g, 16.0 mmol).
After stirring at room temperature for 10 min, 2,6-difluo-
robenzyl bromide (3.3 g, 16.0 mmol) was added and stir-
ring continued overnight at room temperature. The
solvent was removed and the residue was dissolved in ethyl
5.5.1. 1-tert-Butoxycarbonyl-3-carbethoxy-4-piperidone
(8). Di-tert-butyl dicarbonate (39.3 g, 0.18 mol) was
added in one portion to 3-carbethoxy-4-piperidone
hydrochloride 7 (31.83 g, 0.15 mol), sodium bicarbon-
ate (13.9 g, 0.16 mol), sodium chloride (26.3 g,
0.45 mol) in water/chloroform (100 mL/200 mL). The
mixture was heated to 60 ꢂC for 4 h. After cooling to
room temperature the aqueous layer was separated

5600
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
acetate (75 mL) and washed with water (50 mL). The two
layers were separated. The aqueous layer was extracted
withethylacetate(3·50 mL).Theorganiclayerswerecom-
bined, washed with brine (50 mL), dried over anhydrous
magnesium sulfate, filtered, and evaporated to give a yellow
oil. After trituration in hexanes, the expected product 30
was obtained (5.8 g, 75%). 1H NMR (300 MHz, CDCl3)
d: 7.25–7.30, m, 1H; 6.80–6.95, m, 2H; 5.80–5.95, m, 1H;
5.09–5.29, m, 2H; 5.20, s, 2H; 4.57, d, 2H, (J = 5.7 Hz);
4.24, s, 2H; 3.80, t, 2H, (J = 5.7 Hz); 3.63, t, 1H,
(J = 5.7 Hz); 2.60, t, 2H, (J = 5.7 Hz); 1.47, s, 9H.
was used without further purification. A sample was
purified for biological testing. LCMS-3: tR = 1.72
(100%); MS: m/z 455.8 [M+H]+, expected 456 [M+H]+.
5.6. General procedure for the alkylation of 33
To 33 (45 mg, 0.1 mmol) in dimethylformamide (1 mL)
were added the appropriate alkyl or benzyl halide
(0.15 mmol) and
diisopropylethylamine
(52 mg,
0.4 mmol). The resulting mixture was shaken at 45 ꢂC
overnight, after which the mixture was purified by pre-
parative HPLC, affording the TFA salts of the title com-
pounds. Compounds 34–36 were prepared using this
method.
5.5.5. 1-(2,6-Difluoro-benzyl)-2,4-dioxo-3-(2-oxo-ethyl)-
1,3,4,5,7,8-hexahydro-2H-pyrido[4,3-d]pyrimidine-6-car-
boxylic acid tert-butyl ester (31). To a solution of 30
(5.2 g, 12.0 mmol) in dioxane/water (90 mL/30 mL) were
added 2,6-lutidine (2.8 mL, 24.0 mmol), osmium tetrox-
5.7. General procedure for the reductive amination of 33
ide (2.4 mL of
a 2.5% solution in isopropanol,
To 33 (32 mg, 0.07 mmol) in dichloroethane (1 mL) were
added the appropriate aldehyde (0.11 mmol) and sodium
triacetoxyborohydride (22 mg, 0.11 mmol). The resulting
mixture was stirred at room temperature overnight, after
which it was concentrated in vacuo and redissolved in
methanol (1 mL). Preparative HPLC purification yielded
the TFA salts of the title compounds. Compounds 37–54
were prepared using this method.
0.24 mmol), and sodium periodate (10.3 g, 48.0 mmol).
The mixture was followed by TLC (50% ethyl acetate
in hexanes). When the reaction was complete, the mix-
ture was treated with water (150 mL), then extracted
with dichloromethane (3· 50 mL). The organic layers
were combined, washed with water (50 mL), then brine
(50 mL), dried over anhydrous magnesium sulfate, fil-
tered, and evaporated. The crude aldehyde 31 was used
immediately without further purification.
Compound LCMS tR (%
m/z
Expected
method purity at [M+H]+ [M+H]+
220 nM)
5.5.6. 1-(2,6-Difluoro-benzyl)-3-{2-[methyl-(2-pyridin-2-
yl-ethyl)-amino]-ethyl}-2,4-dioxo-1,3,4,5,7,8-hexahydro-2H-
pyrido[4,3-d]pyrimidine-6-carboxylic acid tert-butyl ester
(32). The crude aldehyde 31 (5.55 g, 12.8 mmol) was
dissolved in 100 mL of dichloroethane with 2-(2-methyl-
aminoethyl)pyridine (1.91 g, 14.0 mmol). Sodium triacet-
oxyborohydride (4.06 g, 19.1 mmol) was added and the
mixture was stirred overnight at room temperature. The
solvent was removed and the residue was extracted with
ethyl acetate, washed with water (50 mL) and then brine
(50 mL), dried over anhydrous magnesium sulfate, fil-
tered, and evaporated. The mixture was purified by silica
gel chromatography (5% methanol in dichloromethane)
34
35
36
37
38
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
3
5
5
3
5
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3.06 (100) 546.1
3.03 (91) 547.2
4.11 (97) 599.2
3.65 (99) 560
2.70 (100) 612.2
3.97 (100) 574.1
4.70 (100) 602.1
3.80 (99) 580
3.37 (92) 576.0
3.89 (100) 592.0
4.26 (100) 614
3.05 (100) 624.0
4.09 (100) 606.1
4.00 (99) 580.0
3.65 (98) 560
4.47 (100) 614
4.27 (100) 594.2
4.14 (100) 574.2
4.74 (100) 616.0
3.50 (94) 600.1
546
547
599
560
612
574
602
580
576
592
614
624
606
580
560
614
594
574
616
600
1
to give product 32 (5.17 g, 73%). H NMR (300 MHz,
CDCl3) d: 8.52 d, 1H, (J = 4.8 Hz); 7.60, dt, 1H,
(J = 7.8 Hz and 1.8 Hz); 7.20–7.29, m, 2H; 7.10–7.15, m,
1H; 6.85–6.96, m, 2H; 5.20, s, 2H; 4.23, s, 2H; 4.15, t,
2H, (J = 6.9 Hz); 3.63, t, 2H, (J = 5.1 Hz); 2.78–3.05, m,
4H; 2.80, t, 2H, (J = 6.9 Hz); 2.55–2.63, m, 2H; 2.46, s,
3H; 1.47, s, 9H. LCMS-5: tR = 5.44 (91%); MS: m/z
556.3 [M+H]+, expected 556 [M+H]+.
5.5.7. 1-(2,6-Difluoro-benzyl)-3-{2-[methyl-(2-pyridin-2-
yl-ethyl)-amino]-ethyl}-5,6,7,8-tetrahydro-1H-pyrido[4,3-
d]pyrimidine-2,4-dione (33). Compound 32 (4.49 g,
8.09 mmol) was dissolved in dichloromethane (20 mL)
and trifluoroacetic acid (15.5 mL, 202.3 mmol) was
added. After 2-h stirring at room temperature the sol-
vent was removed and a portion of the residue was ex-
tracted with ethyl acetate and a small amount of 1 N
aqueous sodium hydroxide. The aqueous layer was fur-
ther extracted with ethyl acetate and the combined or-
ganic layers were dried over anhydrous magnesium
sulfate, filtered, and evaporated to give the free base as
a slightly tacky foaming solid 33 (2.2 g, 59.8%) which
Compound 51: 1H NMR of TFA salt (300 MHz, DMSO-
d6) d: 8.46 d, 1H, (J = 4.8 Hz); 7.80, t, 1H, (J = 6.6 Hz);
7.23–7.46, m, 6H; 7.08, t, 2H, (J = 8.1 Hz); 5.13, s, 2H;
4.16, s, 2H; 3.54, t, 2H, (J = 7.2 Hz); 3.34, t, 2H,
(J = 7.2 Hz); 3.14, t, 2H, (J = 7.2 Hz); 2.90, s, 4H; 2.48,
s, 8H; 2.35, s, 2H.
Compound 52: 1H NMR of TFA salt (300 MHz, DMSO-
d6) d: 8.47 d, 1H, (J = 4.8 Hz); 7.80, t, 1H, (J = 7.5 Hz);
7.26–7.46, m, 4H; 7.05–7.16, m, 4H; 5.13, s, 2H; 4.17, s,
2H; 3.55, s, 2H; 3.34, s, 2H; 3.14, s, 2H; 2.90, s, 4H;
2.48, s, 6H; 2.33, s, 2H; 2.72, s, 2H.

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5601
Compound 53: 1H NMR of TFA salt (300 MHz,
DMSO-d6) d: 8.45 d, 1H, (J = 4.5 Hz); 7.93, d, 1H,
(J = 7.2 Hz); 7.74–7.84, m, 2H; 7.32–7.46, m, 4H; 7.29,
t, 1H, (J = 6.0 Hz); 7.07, t, 2H, (J = 7.5 Hz); 5.14, s,
2H; 4.16, s, 2H; 3.53, t, 2H, (J = 6.9 Hz); 3.34, s, 2H;
3.14, t, (J = 6.9 Hz), 2H; 2.89, s, 4H; 2.48, s, 8H; 2.41,
s, 2H.
ide in methanol (2.0 mL, 11 mmol). The resulting mixture
was stirred at room temperature for 30 min, after which it
was concentrated in vacuo and partitioned between
dichloromethane (50 mL) and 0.5 M citric acid (50 mL).
The organic layer was dried over MgSO4 and concen-
trated in vacuo. The resulting residue was triturated with
hexanes/dichloromethane 1:1 (30 mL) and dried to yield
the title compound as an off-white solid (1.1 g, 39%). 1H
NMR (300 MHz, DMSO-d6) d: 7.22, d, 2H,
(J = 9.0 Hz); 6.83, d, 2H, (J = 9.0 Hz); 4.85, s, 2H; 3.98,
s, 2H; 3.69, s, 3H; 3.52–3.46, m, 2H; 2.42–2.38, m, 2H;
1.39, s, 9H. LCMS-1: tR = 2.56 (100%); MS: m/z 301.2
[M+H]+, expected 301 [M+H]+.
5.7.1. 6-(3-Methyl-benzo[b]thiophen-2-ylmethyl)-3-{2-[meth-
yl-(2-pyridin-2-yl-ethyl)-amino]-ethyl}-5,6,7,8-tetrahy-
dro-1H-pyrido[4,3-d]pyrimidine-2,4-dione (55). To
a
solution of 10 (400 mg, 1.3 mmol) in dioxane/water
(10 mL/2 mL) wereadded 2,6-lutidine (279 mg, 2.6 mmol),
OsO4 (0.26 mL of a 2.5% solution in isopropylalcohol), and
NaIO4 (1.1 g, 5.2 mmol). The mixture was stirred at room
temperature and followed by TLC. After 4 h, the reaction
was complete. Solvents were evaporated, a mixture of ethyl
acetate/water 20 mL/20 mL was added, and the aldehyde
extracted with ethyl acetate (2· 20 mL). The organic layers
were combined, washed with water, then brine, dried over
MgSO4, filtered, and evaporated. The crude aldehyde
(293 mg) was used immediately without further purifica-
tion. It was dissolved in 5 mL of 1,2-dichloroethane and
2-methylamino ethyl pyridine (0.14 mL, 1 mmol) was
added followed by sodium triacetoxyborohydride
(290 mg, 1.6 mmol). The mixture was stirred overnight at
room temperature. The solvent was removed and the resi-
due was partitioned between dichloromethane and water,
the organics were washed with brine, dried over MgSO4,
filtered, and evaporated. The integrity of the intermediate
was checked by LC/MS and then used without further
purification. The deprotection step and reduction
amination steps were performed as described earlier for
the library via Scheme 2 to give compound 55.
5.8.2. 1-(2,6-Difluoro-benzyl)-3-(4-methoxy-benzyl)-2,4-
dioxo-1,3,4,5,7,8-hexahydro-2H-pyrido[4,3-d]pyrimidine-
6-carboxylic acid tert-butyl ester (61). To 16 (4.3 g,
11.0 mmol) in dimethylformamide (30 mL) was added
NaH (0.44 g, 11.0 mmol), followed by 2,6-difluorobenzyl
bromide (2.3 g, 11.0 mmol). The resulting solution was
stirred at room temperature for 2 h, after which it was par-
titioned between diethyl ether (50 mL) and 1 N NaOH
(50 mL). The organic layer was dried over MgSO4 and
concentrated in vacuo to yield the title compound as a yel-
low oil (5.3 g, 95%). 1H NMR (300 MHz, CDCl3) d: 7.43,
d, 2H, (J = 9.0 Hz); 7.29, dd, 1H, (J = 8 and 9 Hz); 6.89, t,
2H (J = 8.4 Hz); 6.82, d, 2H (J = 9.0 Hz); 5.21, s, 2H;
5.08, s, 2H; 4.21, s, 2H; 3.77, s, 3H; 3.59, t, 2H,
(J = 5.7 Hz); 2.54, br t, 2H (J = 5.4 Hz); 1.45, s, 9H.
LCMS-6: tR = 2.17 (100%); MS: m/z 414.0 [M+H]+, ex-
pected 413 [M+H]+.
5.8.3. 1-(2,6-Difluoro-benzyl)-3-(4-methoxy-benzyl)-6-(2,
2,2-trifluoro-acetyl)-5,6,7,8-tetrahydro-1H-pyrido[4,3-d]py-
rimidine-2,4-dione (62). To 61 (2.7 g, 5.2 mmol) in dichloro-
methane (10 mL) was added trifluoroacetic acid (7.5 g,
65 mmol). The resulting mixture was stirred for 1 hour at
room temperatureandthenconcentratedinvacuo. Theres-
idue was partitioned between dichloromethane (50 mL)
and 1 M NaOH (50 mL), after which the organic layer
was dried over MgSO4 and concentrated in vacuo to give
the free amine as a yellow oil (1.6 g, 74%). It was used with-
out further purification in the next step. To this intermedi-
ate (0.80 g, 1.9 mmol) in dichloromethane (10 mL) was
added diisopropylethylamine (0.50 g, 3.9 mmol) followed
by trifluoroacetic anhydride (0.60 g, 2.9 mmol). The result-
ing mixture was stirred at room temperature for 48 h. The
resulting mixture was partitioned between dichlorometh-
ane (50 mL) and 0.5 M citric acid (50 mL), after which
the organic layer was dried over MgSO4 and concentrated
in vacuo to yield the title compound as a yellow oil (0.96 g,
99%). 1H NMR (300 MHz, CDCl3) d: 7.43, d, 1H
(J = 9.0 Hz); 7.42, d, 1H, (J = 9.0 Hz); 7.34–7.26, m, 1H;
6.91, t, 2H, (J = 8.4 Hz); 6.82, d, 1H (J = 8.7 Hz); 6.81, d,
1H (J = 8.7 Hz); 5.20, s, 1H; 5.18, s, 1H; 5.07, s, 2H; 4.47,
s, 1H; 4.44, s, 1H; 3.87–3.79, m, 2H; 3.77, s, 3H; 2.72-
2.65, m, 2H. LCMS-1: tR = 2.72 (100%); MS: m/z 510.0
[M+H]+, expected 510 [M+H]+.
Compounds 56–60: were made using synthetic Scheme 2.
In the alkylation step of 10, the 2,6-difluorobenzyl bro-
mide was replaced by methyl iodide, bromomethylcyc-
lohxane, benzyl bromide, 2-fluorobenzyl bromide, and
2-fluoro-6-(trifluoromethyl)benzyl bromide, respectively.
Compound LCMS tR (%
m/z
Expected
method purity at [M+H]+ [M+H]+
220 nM)
55
56
57
58
59
60
3
3
5
3
3
3
3.13 (100) 489.8
3.22 (100) 504
5.43 (100) 586.3
4.77 (100) 579.8
5.46 (100) 598.1
4.68 (98) 666.2
490
504
586
580
598
666
5.8. Library made via Scheme 3
5.8.1. 3-(4-Methoxy-benzyl)-2,4-dioxo-1,3,4,5,7,8-hexa-
hydro-2H-pyrido[4,3-d]pyrimidine-6-carboxylic acid tert-
butyl ester (16). To 4-amino-5,6-dihydro-2H-pyridine-
1,3-dicarboxylic acid 1-tert-butyl ester 3-ethyl ester 8
(2.0 g, 7.3 mmol) in dry toluene (15 mL) was added 4-
methoxybenzyl isocyanate (1.3 g, 8.1 mmol). The result-
ing mixture was stirred for 12 h at 80 ꢂC, after which it
was concentrated in vacuo and resuspended in methanol
(30 mL). To this solution was added 30% sodium methox-
5.8.4. 1-(2,6-Difluoro-benzyl)-6-(2,2,2-trifluoro-acetyl)-
5,6,7,8-tetrahydro-1H-pyrido[4,3-d]pyrimidine-2,4-dione
(63). To 62 (0.96 g, 1.9 mmol) in dichloromethane
(20 mL) was added anhydrous aluminum chloride

5602
M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
(0.51 g, 3.8 mmol). The resulting mixture was stirred at
room temperature for 5 min, after which it was parti-
tioned between dichloromethane (50 mL) and 0.5 M
citric acid (50 mL). The organic layer was dried over
MgSO4 and concentrated in vacuo to yield the title
compound as a beige solid (0.70 g, 95%). LCMS-1:
tR = 2.52 (100%); MS: m/z 388.0 [M+H]+, expected
388 [M+H]+.
was added benzaldehyde (14 mg, 0.15 mmol) followed
by NaBH(OAc)3 (50 mg, 0.25 mmol) and stirred at room
temperature for 3 h. The solvent was removed in vacuo
and redissolved in methanol. Purification was by HPLC
to give the TFA salt of the title compound 70c.
The same series of steps was used to synthesize com-
pounds 66a–e, 67a–e, 68a–e, 69a–e, 70a–e, and 71a–e
from Table 5.
5.8.5. 3-[1-(2,6-Difluoro-benzyl)-2,4-dioxo-6-(2,2,2-triflu-
oro-acetyl)-1,4,5,6,7,8-hexahydro-2H-pyrido[4,3-d]pyrimi-
din-3-ylmethyl]-piperidine-1-carboxylic acid tert-butyl ester
(64). To 63 (0.65 g, 1.7 mmol) in dry tetrahydrofuran
(20 mL) were added triphenylphosphine (0.66 g, 2.5 mmol)
and 3-hydroxymethyl-N-Boc-piperidine (0.54 g, 2.5 mmol)
followed by di-tert-butylazodicarboxylate (0.58 g,
2.5 mmol) at 0 ꢂC. The resulting mixture was warmed to
room temperature and stirred for 16 h, after which it
was absorbed onto silica gel and purified by silica gel chroma-
tographyelutingwith40% ethyl acetate inhexanestoyieldthe
title compound with several impurities present (0.97 g) and
was used in the next step without further purification.
5.8.9. 1-(2,6-Difluoro-benzyl)-6-(2,4-dimethyl-benzyl)-3-[1-
(2-pyridin-2-yl-ethyl)-piperidin-3-ylmethyl]-5,6,7,8-tetrahy-
dro-1H-pyrido[4,3-d]pyrimidine-2,4-dione (72c). To
a
solution of the crude amine 67c (50 mg, ꢀ0.1 mmol) in
1,2-dichloroethane (1 mL) was added 2-vinyl pyridine
(38 mg, 0.4 mmol) followed by acetic acid (24 mg,
0.4 mmol) and the reaction mixture was heated at 80 ꢂC
for 3 h. The solvent was removed in vacuo and redissolved
in methanol. Purification was by HPLC to give the TFA
salt of the title compound. 1H NMR (300 MHz, CDCl3) d:
8.60, d, 1H, (J = 5.1 Hz); 8.01, t, 1H, (J = 7.2 Hz); 7.62, d,
1H, (J = 7.8 Hz); 7.50, dd, 1H, (J = 7.5 and 5.4 Hz); 7.28–
7.18, m, 2H; 7.06–7.03, m, 2H; 6.86, t, 2H, (J = 8.2 Hz);
5.13, d, 1H, (J = 16.5 Hz); 5.09, d, 1H, (J = 16.8 Hz); 4.40,
septet, 1H, (J = 6.4 Hz); 4.32, s, 2H; 4.98–3.82, m, 4H;
3.66–3.36, m, 8H; 3.10–2.96, m, 2H; 2.80–2.54, m, 2H; 2.41–
2.30, m, 1H; 2.36, s, 3H; 2.31, s, 3H; 2.01–1.88, m, 3H.
5.8.6.
3-[1-(2,6-Difluoro-benzyl)-2,4-dioxo-1,4,5,6,7,8-
hexahydro-2H-pyrido[4,3-d]pyrimidin-3-ylmethyl]-piperidine-
1-carboxylic acid tert-butyl ester (65). Intermediate 64
(0.97 g, assume 1.7 mmol) was dissolved in methanol
(25 mL) and K2CO3 (0.325 mesh, 0.47 g, 3.4 mmol) was
added and stirred at room temperature for 3 h. Solids
were filtered off washing with methanol (25 mL) and the
solvent was removed in vacuo. Diethyl ether (75 mL)
and 1 M NaOH (10 mL) were added and the aqueous
phase was extracted with diethyl ether (75 mL). The com-
bined organic extracts were washed with brine, dried over
magnesium sulfate, and concentrated. The intermediate
was used without further purification.
The same series of steps was used to synthesize com-
pounds 72b and 72d from Table 5.
Compound LCMS tR (%
m/z
Expected
method purity at [M+H]+ [M+H]+
220 nM)
66b
66c
66d
66e
67a
67b
67c
67e
68a
68b
68c
69b
69c
69d
70b
70c
71a
71b
71c
(R)71d
(S)71d
(R)72b
(S)72b
72c
3
5
3
4
3
3
3
3
5
3
5
3
5
3
3
5
3
3
5
3
3
3
3
5
3
3.96 (100) 509.1
4.25 (100) 509.2
3.56 (100) 501.1
2.08 (94) 551.0
3.96 (98) 483.0
3.99 (100) 523.2
3.92 (100) 523
509
509
501
551
483
523
523
565
525
565
565
577
577
569
599
599
560
600
600
592
592
620
620
614
606
5.8.7. 1-(2,6-Difluoro-benzyl)-6-(2,4-dimethyl-benzyl)-3-
piperidin-3-ylmethyl-5,6,7,8-tetrahydro-1H-pyrido[4,3-
d]pyrimidine-2,4-dione (66c). The crude compound 65
(assume 1.7 mmol) was dissolved in 1,2-dichloroethane
4.33 (100) 565.0
4.89 (100) 525.2
4.34 (100) 565.2
4.71 (97) 565.3
4.24 (100) 577.2
4.81 (94) 577.3
4.38 (98) 569.2
4.49 (89) 599.2
5.01 (87) 599.4
4.06 (96) 560.2
4.20 (100) 600.2
4.50 (98) 600.4
4.20 (100) 592.1
4.25 (98) 592.1
3.95 (95) 620.2
4.07 (100) 620.2
4.11 (100) 614.4
4.01 (100) 606.2
(15 mL)
and
2,4-dimethylbenzaldehyde
(0.27 g,
0.28 mL, 2.0 mmol) was added followed by NaB-
H(OAc)3 (0.57 g, 2.7 mmol) and stirred at room tem-
perature for 16 h. Polymer supported tosylhydrazine
(2.0 g, ꢀ8 mmol) was added and stirred at room tem-
perature for 6 h to scavenge any remaining aldehyde.
The polymer supported reagent was filtered off washing
with dichloromethane (100 mL) and then TFA (4 mL)
was added and the reaction mixture was stirred at
room temperature for 16 h. The solvent and excess
TFA were removed in vacuo and the mixture was basi-
fied with 2 M NaOH and the aqueous extracted with
dichloromethane (3· 100 mL). The combined organic
extracts were dried over magnesium sulfate and con-
centrated to give the crude amine which was used with-
out further purification. An aliquot was purified by
HPLC for biological testing.
72d
5.8.8. 3-(1-Benzyl-piperidin-3-ylmethyl)-1-(2,6-difluoro-
benzyl)-6-(2,4-dimethyl-benzyl)-5,6,7,8-tetrahydro-1H-
pyrido[4,3-d]pyrimidine-2,4-dione (70c). To a solution of
66c (50 mg, ꢀ0.1 mmol) in 1,2-dichloroethane (1 mL)
Compound 71b: 1H NMR (300 MHz, CDCl3) d: 8.55, d,
1H, (J = 3.9 Hz); 7.76, t, 1H, (J = 7.2 Hz); 7.67, d, 1H,
(J = 7.8 Hz); 7.33, dd, 1H, (J = 7.5 and 4.8 Hz); 7.32–

M. C. Lanier et al. / Bioorg. Med. Chem. 15 (2007) 5590–5603
5603
5. Struthers, R. S.; Chen, T.; Campbell, B.; Jimenez, R.; Pan,
H.; Yen, S. S.; Bozigian, H. P. J. Clin. Endocrinol. Metab.
2006, 10, 3903–3907.
6. (a) Zhu, Y.-F.; Chen, C. Expert Opin. Ther. Patents 2004,
14, 187–199; (b) Tucci, F. C.; Chen, C. Curr. Opin. Drug
Disc. Dev. 2004, 7, 832–847; (c) Armer, R. E.; Smelt, K. H.
Curr. Med. Chem. 2004, 11, 3017–3028; (d) Zhu, Y.-F.;
Chen, C.; Struthers, R. S. Ann. Rep. Med. Chem. 2004, 39,
99–110.
7.21, m, 2H; 7.07–7.03, m, 2H; 6.86, t, 2H, (J = 5.4 Hz);
5.12, br s, 2H; 4.60, dd, 1H, (J = 15.0 and 8.1 Hz); 4.44–
4.26, m, 4H; 4.03–3.84, m, 4H; 3.58–3.34, m, 2H; 3.24–
2.96, m, 4H; 2.37, s, 3H; 2.31, s, 3H; 2.16–1.56, m, 3H.
1
Compound 69d: H NMR (300 MHz, CDCl3) d: 7.39, s,
4H; 7.31, dd, 1H, (J = 8.4 and 6.6 Hz); 6.89, t, 2H,
(J = 8.4 Hz); 5.16, br s, 2H; 4.45–4.32, m, 2H; 4.26, br
s, 2H; 3.92–3.64, m, 4H; 3.51–3.32, m, 2H; 3.07–2.93,
m, 4H; 2.25–1.59, m, 12H.
7. Baber, J. C.; Shirley, W. A.; Gao, Y.; Feher, M. J. Chem.
Inf. Model. 2006, 46, 277–288.
8. McGregor, M. J.; Pallai, P. V. J. Chem. Inf. Comput. Sci.
1997, 37, 443–448.
9. MOE (Molecular Operating Environment) version
2004.03; Chemical Computing Group Inc.: 1010 Sher-
brooke St. West, Suite 910, Montreal, Quebec, H3A 2R7,
Canada.
10. Xue, L.; Godden, J. W.; Stahura, F. L.; Bajorath, J.
J. Chem. Inf. Comput. Sci. 2003, 43, 1151–1157.
11. Pearlman, R. S.; Smith, K. M. J. Chem. Inf. Comput. Sci.
1999, 39, 28–35.
Compound 66e: 1H NMR (300 MHz, CDCl3) d: 7.46–7.24,
m, 10H; 6.86, t, 2H, (J = 8.4 Hz); 5.14, bd, 1H,
(J = 16.2 Hz); 4.94, bd, 1H, (J = 16.2 Hz); 4.60, t, 1H,
(J = 10.8 Hz); 4.52, t, 1H, (J = 10.8 Hz); 4.38, qn, 1H,
(J = 5.7 Hz); 4.14, br s, 2H; 3.98, d, 1H, (J = 12.9 Hz);
3.84, br s, 1H; 3.40–3.16, m, 2H; 2.93, bq, 2H, (J = 15.9 Hz).
Acknowledgments
12. Pearlman, R. S.; Smith, K. M. Perspect. Drug Discovery
Des. 1998, 9/10/11, 339–353.
13. DiverseSolutions, version 6.2; Optive Research, Inc.,
12331-A Riata Trace Parkway, Suite 110, Austin, TX
78727, USA.
14. Catalyst, version 9.1; Accelrys Inc., 9685 Scranton Road,
San Diego, CA 92121, USA.
15. CombiCode is a suite of program and libraries in Python
provided in an interface known as Pyxie. It is available
from Deltagen Research Laboratories.
We thank Dr. Bill Shirley, Dr. Christian Baber, and
Yinghong Gao (Computational chemistry group, Neu-
rocrine Biosciences) for performing some of the compu-
tational tasks, Shawn Ayube (Medicinal chemistry
group, Neurocrine Biosciences) for the analytical sup-
port, and Dr. Stephen Betz (Endocrinology group, Neu-
rocrine Biosciences) for countless discussions.
16. Labute, P.; Williams, C.; Feher, M.; Sourial, E.; Schmidt,
J. M. J. Med. Chem. 2001, 44, 1483–1490.
17. Tucci, F. C.; Zhu, Y.-F.; Struthers, R. S.; Guo, Z.; Gross,
T. D.; Rowbottom, M. W.; Acevedo, O.; Gao, Y.;
Saunders, J.; Xie, Q.; Reinhart, G. J.; Liu, Z.; Ling, N.;
Bonneville, A. K. L.; Chen, T.; Bozigian, H.; Chen, C.
J. Med. Chem. 2005, 48, 1169–1178.
References and notes
1. (a) Matsuo, H.; Baba, Y.; Nair, R. M.; Arimura, A. V.
Biochem. Biophys. Res. Commun. 1971, 43, 1334–1339; (b)
Huirne, J. A.; Lambalk, C. B. Lancet 2001, 358, 1793–
1803; (c) Amoss, M.; Burgus, R.; Blackwell, R.; Vale, W.;
Fellows, R.; Guillemin, R. Biochem. Biophys. Res. Com-
mun. 1971, 44, 205–210.
2. Cheng, K. W.; Leung, P. C. K. Can. J. Physiol. Pharma-
col. 2000, 78, 1029–1052.
3. Fujino, M.; Fukuda, T.; Shinagawa, S.; Kobayashi, S.;
Yamazaki, I.; Nakayama, R.; Seely, J. H.; White, W. F.;
Rippel, R. H. Biochem. Biophys. Res. Commun. 1974, 60,
406–413.
18. Feher, M.; Schmidt, J. M. J. Comput. Aided Mol. Des.
2001, 15, 1065–1083.
19. Betz, S. F.; Reinhart, G. J.; Lio, F. M.; Chen, C.;
Struthers, R. S. J. Med. Chem. 2006, 49, 637–647.
20. Rowbottom, M. R.; Tucci, F. C.; Zhu, Y.-F.; Guo, Z.;
Gross, T. D.; Reinhart, G. J.; Xie, Q.; Struthers, R. S.;
Saunders, J.; Chen, C. Bioorg. Med. Chem. Lett. 2004, 14,
2269–2274.
21. Tucci, F. C.; Zhu, Y.-F.; Struthers, R. S.; Guo, Z.; Gross,
T. D.; Rowbottom, M. W.; Acevedo, O.; Gao, Y.;
Saunders, J.; Xie, Q.; Reinhart, G. J.; Liu, X.-J.; Ling,
N.; Bonneville, A. K. L.; Chen, T.; Bozigian, H.; Chen, C.
J. Med. Chem. 2005, 48, 1169–1178.
4. Clark, E.; Boyce, M.; Warrington, S.; Johntson, A.;
Suzuki, N.; Cho, N.; Dole, N.; Nishizawa, A.; Furuya, S.
Proc. Br. Pharmacol. Soc. Clin. Pharmacol. Section Br.
J. Clin. Pharmacol. 2003, 55, 443.