Discovery of a series of potent and selective human H4 antagonists using ligand efficiency and libraries to explore structure-activity relationship (SAR)

Article abstract of DOI:10.1016/j.bmcl.2011.07.114

We describe the identification of a potent, selective lead series that shows antagonism against the human histamine H4 receptor from thirteen actives identified in an HTS as part of a hit to lead program. By focusing on ligand efficiency and concurrently using a diversity based approach, compounds based around 2,4-diaminopyrimidine were identified with compound 25 being quickly shown to be a good lead. It also had the highest ligand efficiency in the series.

Full text of DOI:10.1016/j.bmcl.2011.07.114

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6591–6595
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a series of potent and selective human H4 antagonists using
ligand efficiency and libraries to explore structure–activity relationship (SAR)
⇑
M. Abid Masood , Matthew D. Selby, Andrew S. Bell, Andrew C. Mansfield, Mark Gardner,
Graham F. Smith , Charlotte Lane, Helen Kenyon-Edwards, Rachel Osborne, Rhys M. Jones, Wai L. Liu,
Christopher D. Brown, Nicholas Clarke, Francesca Perrucio, Charles E. Mowbray
Pfizer Worldwide Research and Development, Sandwich Laboratories, Ramsgate Road, Sandwich CT13 9NJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe the identification of a potent, selective lead series that shows antagonism against the human
histamine H4 receptor from thirteen actives identified in an HTS as part of a hit to lead program. By focus-
ing on ligand efficiency and concurrently using a diversity based approach, compounds based around
2,4-diaminopyrimidine were identified with compound 25 being quickly shown to be a good lead. It also
had the highest ligand efficiency in the series.
Received 21 June 2011
Revised 28 July 2011
Accepted 29 July 2011
Available online 27 August 2011
Published by Elsevier Ltd.
Keywords:
Histamine
Antagonist
Ligand efficiency
Library
Pyrimidine
Histamine 1 is a naturally occurring biological amine that affects
a variety of functions in the human body. There are four subtypes of
histamine receptor, H1–H4, H4 is a 390 amino acid G-protein cou-
pled receptor and the newest member of the histamine receptor
family. It was identified at the turn of the millennium by a number
independent research groups. Its predominant expression on
inflammatory cells and lymphoid tissues, coupled with an active ro-
lein processes ranging from chemotaxis to cytokine release, sug-
gests it may command an important role in immune and
inflammatory responses.1
of the disease, a variety of inflammatory cell types which produce
a range of chemical mediators and cytokines in the airways, are
thought to be most important for its perpetuation. Repeated acute
episodes of inflammation can exacerbate the disease, often in re-
sponse to specific allergens. Histamine 1 is a well-described medi-
ator of this acute inflammation and is involved in vasodilation,
smooth muscle contraction, oedema, mucus hypersecretion, and
adhesion molecule up-regulation. Histamine 1 is produced during
acute asthmatic episodes and its role in many physiological func-
tions associated with asthma is well characterised. However, until
recently it has not been thought to be a significant contributor to
the disease because histamine H1 receptor antagonists, effective
in symptomatic treatment of allergic rhinitis, have little or no effi-
cacy in asthma. More recent evidence suggests that histamine 1
has a role in inflammation and allergy beyond that traditionally
described,3 potentially through activation of the H4 receptor.
Johnson & Johnson were the first company to publish on a selec-
tive histamine H4 antagonist,4 JNJ 10191584 2 and JNJ 7777120 3
(Fig. 1) have been investigated as orally active antagonists. Early
studies into the function of H4 receptors suggested that antago-
nism of these receptors may be of utility for the treatment of aller-
gic diseases with an eosinophilic and/or mast cell component since
histamine binding to this receptor subtype stimulated the
chemotaxis of both cell types.5
N
N
H
NH2
Histamine [1]
Asthma is predominantly characterised by bronchoconstriction
that is largely reversible and chronic inflammation of the airways.2
Although there are many contributing factors to the pathogenesis
⇑
Corresponding author. Tel.: +44 1227 780843; fax: +44 1304 651821.
The H4 receptor is primarily expressed on eosinophils, mast cells,
dendritic cells, and other leucocytes. The H3 receptor is found in the
peripheral and central nervous systems and has been implicated in
E-mail address: abid.masood@pfizer.com (M.A. Masood).
Present address: Merck Research Laboratories, 33 Avenue Louis Pasteur, Boston,
MA 02115-5727, USA.
0960-894X/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2011.07.114

6592
M. A. Masood et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6591–6595
The initial hit from the pyrimidone series 7 showed good po-
N
N
tency in the binding assays. This was not surprising as all active
examples tested featured the histamine side chain. The risk of
Cytochrome P450 inhibition by the unsubstituted imidazole group
meant this series was not followed up because of the potential for
drug–drug interactions.6
Cl
Cl
N
O
N
O
N
N
H
N
H
The tertiary alcohol series, exemplified by 6, was active in both
functional and binding assays. Nearest neighbour searches identi-
fied several close analogues for screening and the initial SAR
showed that an ortho-substituent on the phenyl ether was impor-
tant. However, later studies showed low selectivity over H3 and
Herg activity of these compounds and so no further work was
done.
JNJ-10191584 [2]
JNJ-7777120 [3]
hH4R Binding Ki 47.7 nM
hH4R Functional Ki 39.6 nM
Highly selective vs H1/H2/H3
LE : 0.52
hH4R Binding Ki 8.0 nM (lit Ki 4nM)
hH4R Functional Ki 6.8 nM
Highly selective vs H1/H2/H3.
LE : 0.58
Figure 1.
The 2-aminopyrimidine-4-amide 4 was weakly active in the
binding assay and inactive in the functional assay. Nearest
neighbour searches uncovered several close in analogues, but these
compounds were similar in potency.
various conditions including loss of arousal, cognition and vigilance.
The H1 receptor is expressed in smooth muscles, vascular endothe-
lial cells, the CNS and the heart. H2 receptors are coupled to adenyl-
ate cyclase and are highly expressed in gastric cells and cardiac
tissue. Selectivity over H1, H2 and H3 was therefore important to
avoid any significant adverse effect.
Approximately 30% of the full Pfizer compound collection was
screened in a [3H]histamine ligand binding assay. Triage of the ini-
tial hits focused on compounds that had been made in parallel ar-
rays for file enrichment. This gave 13 compounds that were active
in both functional and binding assays.
These 13 actives were classed into four series; aminopyrimidine
amide 4, diaminopyrimidine 5, tertiary alcohol 6 and the pyrimi-
done 7 series (Fig. 2). Since all compounds were made using paral-
lel chemistry file enrichment, an abundance of close analogues to
each of these four series were available.
This Letter looks at the hit to lead optimisation and of library
chemistry around the diaminopyrimidine series and how HTS ac-
tives were transformed into a potent and selective series of H4
antagonists.
The fourth series, substituted 4,6-diaminopyrimidine 5 was po-
tent (human H4 Ki 2.11 l
M) and also had a ligand efficiency (LE)7
of 0.286 which was a good starting place. Ligand efficiency is de-
fined as the binding energy of a ligand per heavy atom.8 So com-
pounds with high ligand efficiency are those where binding has
been optimised.9
Based on the LE and apparent activity and because it had fewer
selectivity issues, the 2,4-diaminopyrimidine series 5 was followed
up. The key pharmacophore for known H4 antagonists is a basic
amine coupled to a heterocycle.
Our strategy for optimising these hit compounds into potent
and selective H4 antagonist were to use diversity based drug de-
sign and file mining based approaches in parallel. Special focus
on improving the ligand efficiency at each design stage was made.
The diaminopyrimidine 5 had a ligand efficiency of 0.286 and this
was a good starting place, but there was room for improvement
(0.4–0.5 ideally). As both JNJ-10191584 2 and JNJ-7777120 3, con-
tained N-methyl piperazines, it was rationalised that the furan por-
tion did not contribute to binding and the diamino-pyrimidine was
the key pharmacophore. Synthesis of both stereoisomers of the
aminopyrrolidine deriviatives 8 and 9 showed that the (S)-enantio-
mer 8 was 2-fold more potent and had a higher LE than the enan-
tiomer 9 (Fig. 3).
HN
N
N
The (S)-enantiomer 8 was potent in both the functional and
binding assays as well as being selective over H1, H2 and H3. This
compound 8 was a good starting point as it had a high ligand effi-
ciency and selectivity over H3.
CF3
N
HN
N
N
H
N
N
N
O
OH
O
4
5
cLogP 3.3
mw: 379
hH4 func 57%@15µM
hH4 Ki : 34.9µM
LE: <0.22
cLogP 3.8
mw: 374
hH4 func 73%@15µM
hH4 Ki : 2.11µM
LE: 0.286
HN
N
HN
N
N
N
N
O
HN
HN
N
H
N
N
HO
N
O
N
N
N
H
N
O
H
H
O
8
9
6
7
LE: 0.45
LE: 0.43
cLogP 2
mw: 374
cLogP 0.9
mw: 325
hH4 bind Ki: 502nM
hH4 func Ki: 572nM
H3 Ki : 8.22µM
hH4 bind Ki: 939nM
hH4 func 60%@15µM
(ki>2µM)
hH4 func 65%@15µM
(ki>2µM)
H3 Ki 13.6µM
H1 Ki : >40µM
hH4 binding Ki 3.24µM
H3 Ki : 2.5µM
LE: 0.28
hH4 binding Ki 1.6µM
LE: 0.33
H2 Ki : 8.82µM
mw 263, logD7.4 0.6
Figure 2.
Figure 3.

M. A. Masood et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6591–6595
6593
H
N
N
N
N
H
N
H
N
H
N
N
N
N
N
10
N
NH
NH
NH
N
O
N
N
N
N
hH4 bind Ki 212nM
LE: 0.395
HO
F
O
Figure 4.
11
12
hH4 bind Ki 20µmol
13
hH4 bind Ki 20µmol
hH4 bind Ki 593nM
vary amines
investigate
Figure 6. First library modifying pyrrolidine N-substitution.
aminoindan derivative 10 were tolerated. Three alkylated 2-amino,
4-chloro-pyrimidinestemplates were synthesised by reductiveami-
nation of 3-amino-6-chloropyrimidine 14 with three aldehydes
(cyclopropylaldehyde, pivaldehyde and amyl aldehyde) (Fig. 7), then
ipso-displacement of the pyrimidine-chlorine with a series of Boc-
protected diamines.
different heterocycle
HN
N
cores
N
NH
Reductive
N
H
alkylation
All the libraries were designed using Pfizer’s internal design tool
PGVL hub10 and the design was guided using Lipinski’s rule of 5.11
After filtering, the second library consisting of 234 compounds was
designed. The reductive amination step was high yielding and the
three templates (Fig. 7) were then used directly in the ipso substi-
with aldehydes
and ketones
8
vary diamines
Figure 5. Library strategy.
tution reaction. Each reaction was carried out on 100 lmol scale
Nearest neighbour searches of the Pfizer compound collection
around compound 8 uncovered the 2-aminoindan derivative 10
(hH4 binding Ki: 212 nM) which was more potent than the origi-
nal active 5, it also had a much higher ligand efficiency (LE) of
0.395 (Fig. 4).
Concurrently the HTS hit 5 was followed up with library chem-
istry. With four points of diversity around the diaminopyrimidine
5, SAR was probed using parallel chemistry (Fig. 5). Our first library
was designed to explore alternatives to the furan group by reduc-
tive alkylations on the pyrrolidine amine with a set of aldehydes
and ketones.
Reductive amination of the amine 8 was carried out by reaction
with the aldehyde/ketone and sodium triacetoxyborohydride in
dichloroethane (DCE) at room temperature overnight. After purifi-
cation, 80 out of a possible 81 compounds were isolated giving a
success rate of 98%. Biological testing of the compounds in the
H4 binding screen revealed only 2 actives (Fig. 6) with very weak
with 10 equiv of diamine and excess Hunig’s base in DMSO. The
reactions were heated at 150 °C over 16 h. After work-up and puri-
fication only 25 compounds were isolated in sufficient purity and
quantity (>1 mg) giving a library success rate of 11%. This low suc-
cess rate was attributed to poor reactivity of the diamines and
instability of the Boc-amines at high temperatures. Analysis of
the LC–MS data at each synthetic step showed that the majority
of compounds were lost in synthesis (61%) and the remainder in
purification. Table 1 shows the most active compounds from this
library (Fig. 8).
The N-methyl piperazine derivative 15 was the most potent
compound and also had the highest LE. However, it was active in
the binding assay12 but not in the functional assay and it also
had lower selectivity over H3. The dimethylaminopyrrolidine
derivative 16 was potent in both functional and binding assays
but was almost equipotent against H3. The remaining three ana-
logues were less potent.
potency (Ki 20 lM).
Library 3 was designed to modify the second alkyl amine, keep-
ing the diamine (N-methylpiperizine) constant again using ipso-
displacement chemistry.
After work-up and purification 187 compounds were isolated
(99% success rate; 187 compounds in the design). 32 compounds
This library demonstrated that alkylating the amine with large
groups was not tolerated. From the known SAR around JNJ-
10191584 and JNJ-7777120 as well as the N-methylpyrrolidine
derivative 13, this showed that methyl was the optimum size.
Our second library was designed to look at SAR around the
diamine and also at how small changes to the alkyl chain on the
from this library showed weak potency in the 2–4
lM range with
the bicycloamine derivative 20 (Fig. 9) being the most potent with
O
O
R
N
O
O
H
H
N
H
NBoc
NR
Boc-diamines
N
Cl
N
Cl
N
N
+
N
N
NH
NH
NH2
R1
O
H
R1
ipso displacement
Reductive
amination
14
Aldehydes
Figure 7. Second library exploring diamines.

6594
M. A. Masood et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6591–6595
Table 1
Library 4 was designed to attempt to identify alternative het-
Activity of most potent library compounds
erocyclic cores and consisted of a two step synthetic route involv-
ing ipso diplacement of a chloroheteroaromatic with a Boc-
protected diamine, followed by acid mediated deprotection. There
were 289 compounds in the design and 288 were purified and
tested against H4 (success rate of 99%). From this library two com-
pounds were identified, the aminoquinoxaline drivative 21 and
bicyclo-diamine 22 (Fig. 10). Both compounds were small with
high LE’s and reasonable potencies in the binding assays, but they
were inactive in the functional assay.
Compound hH4 bind Kia
hH3 bind Ki
(nM)
Ligand efficiency
(LE)b
LogD7.4
(nM)
15
16
17
18
19
23.6
58.9
201
235
601
825
54
—
—
—
0.54
0.49
0.48
0.46
0.35
1.75
1.09
1.40
0.96
3.84
a
Displacement of [3H]histamine from the recombinant human histamine H4
receptor. Ki was calculated according to Cheng and Prussoff and are the geometric
Library 5 was designed to vary both the diamines and amines to
explore a wider set of amines and diamines and so it was the largest
in terms of numbers. The synthesis involved sequential ipso-dis-
placement of 4,6-dichloropyrimidine with a series of diamines
and then displacement with a set of low molecular weight amines.
The library design was filtered using a clogP range of 0.5–3 and
clogD range of À1 to 3, then with in-silico HLM and RLM models
to help filter down the library size to a few hundred compounds
(586). Synthesis was carried out and the first ipso displacement
with 5 equiv of amine at 70 °C and then 10 equiv of the diamine
at 150 °C for 30 h.13 After purification, 585 compounds were iso-
lated giving a success rate of 99% reflecting how well the chemistry
has worked. It is noteworthy that Boc-protected diamines were not
used in this library, hence the higher success rate as compared with
library 2. In total 15 compounds were found to be active against hu-
man H4 in this library. Two key compounds, 23 and 24 were iden-
tified as they were the most potent and had very good LE’s. 23 had a
LE of 0.6, the highest of all compounds but it was inactive in the
mean of three or more independent determinations.
b
LE calculated using Ki binding.
HN
N
N
N
N
N
N
N
N
N
N
N
NH
NH
NH
15
16
17
N
N
N
N
N
N
N
N
functional assay and the microsomal stability (RLM Clint: 143
min/s) needed improving (Fig. 11).
ll/
NH
NH
RLM was measured because for the majority of compounds
within the pyrimidine series, RLM had been higher than HLM
and it was the aim to get enough exposure in animal species to test
the safety of the compounds.
18
19
The library work and file mining efforts were followed up by
synthesis of discrete single compounds. Based on the emerging
SAR, N-methyl piperazine and 2-methylbutylamine appeared to
be the optimum groups and so the compound 25 with these opti-
Figure 8. Most potent diamino pyrimidines from library 2.
N
N
N
N
H
N
H
20
NH
N
NH
N
N
hH4 Ki : 860nM
LE : 0.39
N
N
N
N
H
N
Figure 9.
24
23
hH4 binding Ki: 207nM
LE : 0.43
hH4 binding Ki: 32.4nM
LE : 0.60
N
N
N
N
Figure 11. Displacement of [3H]histamine from the recombinant human histamine
H4 receptor. Ki was calculated according to Cheng and Prussoff and are the
geometric mean of three or more independent determinations.
N
N
N
N
22
21
hH4 bind Ki: 0.93µM
hH4 bind Ki: 1.06µM
N
H4 bind Ki : 82nM
H4 Func Ki : 37nM
H3 bind Ki : 646nM
H2 bind Ki : 3.56uM
H1 bind Ki : >10uM
Dofetolide IC50 >10uM
logD7.4: -0.86
logD7.4: 1.76
N
N
LE : 0.48
LE : 0.46
N
Figure 10.
NH
LE: 0.48
MW 277, LogD 2.6
HLM 16uL/min/mg
RLM 147uL/min/mg
a Ki of 860 nM in the binding assay but there was divergent activity
in the functional assay. Biological testing showed that, in general
the SAR was very tight and large groups were not tolerated on
the amine.
25
Figure 12.

M. A. Masood et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6591–6595
6595
S.; Gladue, R. P.; Foo, S.; Trim, S.; Peter, B.; Trevethick, M.; Fidock, M. J.
Recept. Signal Transduct. Res. 2002, 22, 431.
2. Cohn, L.; Elias, J. A.; Chupp, G. L. Annu. Rev. Immunol. 2004, 22, 789–815.
3. Dy, M.; Schneider, E. Cytokine Growth Factor Rev. 2004, 15, 393; Jutel, M.; Blaser,
K.; Akdis, C. A. J. Investig. Allergol. Clin. Immunol. 2005, 15, 1.
4. Jablonowski, J. A.; Grice, C. A.; Chai, W.; Dvorak, C. A.; Venable, J. D.; Kwok, A.
K.; Ly, K. S.; Wei, J.; Baker, S. M.; Desai, P. J.; Jiang, W.; Wilson, S. J.; Thurmond,
R. L.; Karlsson, L.; Edwards, J. P.; Lovenberg, T. W.; Nguyen, S.; Ling, P. J. Med.
Chem 2003, 46, 3957; Coruzzi, G.; Adami, M.; Guaita, E.; de Esch, I. J.; Leurs, R. E.
J. Pharmacol. 2007, 563, 240.
5. Thurmond, R. L.; Gelfand, E. W.; Dunford, P. J. Nat. Rev. Drug Disc. 2008, 7, 41;
Jutel, M.; Blaser, K.; Akdis, C. A. Chem. Immunol. Allergy 2006, 91, 174.
6. Zhou, Z. W.; Zhou, S. F. Expert Opin. Drug Metab. Toxicol. 2009, 5, 579.
mum groups was synthesised. The diaminopyrimidines were nom-
inated as the lead series after biological, drug metabolism and
safety studies had been carried out on the lead compound 25. This
series was taken into lead optimisation. The diaminopyrimidine 25
showed good selectivity against H1, H2. It was 5-fold selective over
H3 and exhibited encouraging metabolic stability in the human li-
ver microsomal assay (HLM), and moderate stability in the rat liver
microsomal assay (RLM). So the key goals in lead optimisation
were to improve metabolic stability, selectivity over H3 and po-
tency. In addition, the LE was comparable with the J&J H4 antago-
nists (LE 0.52 and 0.58) (Fig. 12).
7. Ligand efficiencies (LE) were calculated using the formula: LE = (DG)/N where
D
G = ÀRTlnKi. Ki is the binding constant and Ki was calculated according to
When tested against a panel of over 40 receptor targets repre-
senting the major classes of biogenic amine receptors, ion channels
binding sites, neuropeptide receptors and transporters, these com-
pounds showed no off-target activity.
Cheng and Prussoff and are the geometric mean of three or more independent
determinations. N is the number of non-hydrogen atoms.
8. Kuntz, I. D.; Chen, K.; Sharp, K. A.; Kollman, P. A. Proc. Natl. Acad Sci. U.S.A. 1999,
96, 9997; Reynolds, L. H.; Bembenek, S. D.; Touge, B. A. Biorg. Med. Chem. Lett.
2007, 17, 4258.
By focusing on the average contribution to binding that each
atom makes (LE) in this program, we were able to take an active
5 from the HTS, remove excess molecular weight and look at the
key binding elements and hence find a leaner more potent com-
pound (8) by file mining of the Pfizer compound collection. With
the help of parallel chemistry to make analogues of 8 and hence
generate data for novel SAR, we were able to identify that the
2,6-diaminopyrimidine series had the potential to deliver a potent
and selective human H4 antagonist. At each stage LE increased and
25 which had a LE of 0.48 is a good example of the lead series
which was taken into lead optimisation.
9. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
10. Peng, Z.; Yang, B.; Mattaparti, S.; Shulok, T.; Thacher, T.; Kong, J.; Kostrowicki,
J.; Hu, Q.; Na, J.; Zhou, J. Z.; Klatte, D.; Chao, B.; Ito, S.; Clark, J.; Sciammetta, N.;
Coner, B.; Waller, C.; Kuki, A. Methods Mol. Biol. 2011, 685, 295; Peng, Z.; Hu, Q.
Methods Mol. Biol. 2011, 685, 321.
11. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Del. Rev.
2001, 46, 3.
12. H4 binding assay: Cell pellets from CHO cells expressing the histamine H4
receptor were homogenised in ice-cold 50 mM Tris–HCI/0.5 lM CaCI2 buffer
containing a protease inhibitor cocktail (RocheÒ, United Kingdom) using a
ground glass homogeniser. Homogenates were centrifuged at 48,000 rpm for
30 min at 4 °C. The membrane pellet was resuspended in fresh buffer and the
centrifugation step was repeated as described above. The membrane pellet was
resuspended in 50 mM Tris–HCI in the same volume as the original cell pellet.
Aliquots of membrane preparations were stored at À8 °C and were used for
[3H]-Histamine binding experiments.
Acknowledgments
Cell membranes (20–35 lg/well) were incubated for 90 min shaking at room
temperature with 3 nM [2,5-3H]Histamine dihydrochloride (30–60 Ci/mmol)
in 50 mM Tris–HCI (pH 7.4), with or without competing H4 ligands. The
reaction was terminated by rapid filtration through 0.5% polyethylenimine-
soaked Unifilter GF/B plates (Packard) followed by three washes with 1 ml ice-
cold 50 mM Tris–HCI. Filters were dried for 45 min at 45 °C and bound
radiolabel was determined using scintillation counting techniques. Non-
The authors are greatly indebted to the Separation and Structural
Sciences group, the Department of Pharmacokinetics, Dynamics and
Metabolism and the High Throughput Screening Group for their
efforts in this project.
specific binding was defined with 10 lM JNJ7777120. For competition
binding studies, Ki values were calculated from the IC50 value based on an
experimentally determined ligand Kd of 5.2 nl and a ligand concentration of
5 nM according to the Cheng–Prussoff equation where; Ki = (IC50)/(1 + ([L]/Kd)).
13. The amines were dissolved in dimethylsulfoxide (DMSO) to a concentration of
0.5 M. The heteroaryl chlorides were similarly dissolved in dichloromethane
References and notes
1. Liu, C.; Ma, X.-J.; Jiang, X.; Wilson, S. J.; Hofstra, C. L.; Blevitt, J.; Pyati, J.; Li,
X.; Chai, W.; Carruthers, N. I.; Lovenberg, T. W. Mol. Pharmacol. 2001, 59,
420; Morse, K. L.; Behan, J.; Laz, T. M.; West, R. E., Jr.; Greenfeder, S. A.;
Anthes, J. C.; Umland, S.; Wan, Y.; Hipkin, R. W.; Gonsiorek, W.; Shin, N.;
Gustafson, E. L.; Qiao, X.; Wang, S.; Hendrick, J. A.; Greene, J.; Bayne, M.;
Monsma, F. J., Jr. J. Pharmacol. Exp. Ther. 2001, 293, 1058; Nguyen, T.;
Shapiro, D. A.; George, S. R.; Setola, V.; Lee, D. K.; Cheng, R.; Rauser, L.; Lee,
S. P.; Lynch, K. R.; Roth, B. L.; O’Dowd, B. F. Mol. Pharmacol. 2001, 59, 427;
Zhu, Y.; Michalovich, D.; Wu, H.; Tan, K. B.; Dytko, G. M.; Mannan, I. J.;
Boyce, R.; Alston, J.; Tierney, L. A.; Li, X.; Herrity, N. C.; Vawter, L.; Sarau, H.
M.; Ames, R. S.; Davenport, C. M.; Hieble, J. P.; Wilson, S.; Bergsma, D. J.;
Fitzgerald, L. R. Mol. Pharmacol. 2001, 59; O’Reilly, M.; Alpert, R.; Jenkinson,
(DCM) to a concentration of 0.5 M. An aliquot of the chloride (0.15 ml, 75
1 equiv) was dispensed to a Teflon 96 well plate and the solvent was allowed to
evaporate. An aliquot of the amine (0.45 ml, 150 mol, 3 equiv) and N,N-
lmol,
l
diisopropylethylamine (DIPEA, 0.1 ml) were added to each well and the plates
were sealed with a Teflon sheet using a Combiclamp and heated to 150° for
16 h. The solvent was removed in vacuo (Genevac). The samples were
redissolved in
a 1:1 mixture of dichloromethane:methanol (0.8 ml). A
solution of HCl in 1,4-dioxane (4.0 M, 0.25 ml, 13 equiv) was added and the
reaction mixtures were shaken overnight at room temperature. A sample
(30 ll) was retained for analysis and the solvent removed in vacuo (Genevac).