A new family of H3 receptor antagonists based on the natural product Conessine

Article abstract of DOI:10.1016/j.bmcl.2007.12.059

A new family of Histamine H₃ receptor antagonists (5a-t) has been prepared based on the structure of the natural product Conessine, a known H₃ antagonist. Several members of the new series are highly potent and selective binders of rat and human H₃ receptors and display inverse agonism at the human H₃ receptor. Compound 5n exhibited promising rat pharmacokinetic properties and demonstrated functional antagonism of the H₃ receptor in an in-vivo pharmacological model.

Full text of DOI:10.1016/j.bmcl.2007.12.059

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1490–1494
A new family of H3 receptor antagonists based on the
natural product Conessine
Vincent J. Santora,a,* Jonathan A. Covel,a Rena Hayashi,a Brian J. Hofilena,a
Jason B. Ibarra,a Michelle D. Pulley,a Michael I. Weinhouse,a Dipanjan Sengupta,a
Jonathan J. Duffield,a Graeme Semple,a Robert R. Webb,a Carleton Sage,a Albert Ren,a
Guilherme Pereira,a Jens Knudsen,a Jeffrey E. Edwards,b Marissa Suarez,c
John Frazer,c William Thomsen,c Erin Hauser,c Kevin Whelanc and Andrew J. Grottickc
aMedicinal Chemistry, Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, USA
bDrug Metabolism and Pharmacokinetics, Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, USA
cDiscovery Biology, Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, USA
Received 20 November 2007; revised 17 December 2007; accepted 21 December 2007
Available online 25 December 2007
Abstract—A new family of Histamine H3 receptor antagonists (5a–t) has been prepared based on the structure of the natural prod-
uct Conessine, a known H3 antagonist. Several members of the new series are highly potent and selective binders of rat and human
H3 receptors and display inverse agonism at the human H3 receptor. Compound 5n exhibited promising rat pharmacokinetic prop-
erties and demonstrated functional antagonism of the H3 receptor in an in-vivo pharmacological model.
Ó 2008 Elsevier Ltd. All rights reserved.
The histamine H3 receptor is one of four distinct G-pro-
tein coupled histamine receptors known as H1, H2, H3,
and H4.1 Histamine H1 and H2 receptors are validated
drug targets and involved in allergic reactions (H1
antagonists) and gastric acid secretion (H2 antagonists).
The more recently discovered H4 receptor is found in
immune cells and is being investigated as a target for
inflammatory disorders. H3 receptors are expressed pri-
marily in the central nervous system (CNS) where they
are located presynaptically, and function as constitu-
tively active regulators of the synthesis and release of
histamine and several other neurotransmitters.2 H3
antagonists have attracted interest recently as potential
treatments for CNS disorders relating to feeding, wake-
fulness, cognition, and pain.3 Many of the early H3
antagonists contained an imidazole group similar to
the natural ligand histamine. The potential for imidaz-
ole-containing ligands to inhibit CYP-450 enzymes4
and the low receptor selectivity exhibited by some early
H3 antagonists5 encouraged the development of second
generation H3 antagonists which lacked this group. Such
‘nonimidazole’ H3 antagonists are now at the forefront
of H3 drug development and several compounds in this
class have recently entered human clinical trials.6
We were interested in developing novel H3 antagonists/
inverse agonists as wake promoting agents, and were
intrigued when a high throughput screen of our com-
pound collection identified the steroidal alkaloid natu-
ral product Conessine (1) as a fairly potent H3
antagonist.7 Compound 1 was found to inhibit [3H]
N-methyl histamine binding in a rat cortex membrane
assay with a Ki of 66 nM. We hypothesized that the
relative positioning of the two amine functionalities
of 1 was crucial to its biological activity, and sought
to construct novel H3 antagonists by designing a series
of compounds that would orient two amines in a man-
ner similar to that of 1. This approach was further sup-
ported by several reports of diamine-based H3
antagonists (e.g., 2,8 3,9 and 4,10 Fig. 1) that appeared
during the course of our work. As outlined in Figure 2,
the initial target compounds consisted of a series of
diamines 5 in which the structure of 1 was simplified
by aromatization of the central six-membered ring,
Keywords: Histamine H3 receptor antagonist; Histamine H3 receptor
inverse agonist; Conessine.
*
Corresponding author. Tel.: +1 858 453 7200; fax: +1 858 453
7210; e-mail: vsantora@arenapharm.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.059

V. J. Santora et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1490–1494
1491
Alteration of the bicyclic portion of these molecules was
examined by preparing diamines 5n–t, which incorpo-
rate the phenethyl pyrrolidine fragment of compound
5f. In this case, bicyclic ketone 8 was reacted with the
lithium reagent derived from 4-bromo-phenethylamine
12 to give the alcohol 13. Deprotection of the N-benzyl
group of 13 followed by dehydration with HCl provided
diamine 14, which was then substituted via reductive
amination or arylation to give analogs 5n–t.
Compounds 5a–t were tested in a rat cortex N-[3H]-
methylhistamine binding assay,11 the results of which
are shown in Table 1. Several compounds in the initial
series bound the receptor with high affinity, and it was
apparent that the basicity and substitution patterns of
both nitrogens were important for binding affinity. For
Figure 1. Structure of Conessine and representative diamine H3
antagonists.
removal of several carbon atoms, and the introduction
of a double bond in the remaining bicyclic portion of
the scaffold.
example,
the
phenethylpyrrolidine
analog
5f
(Ki = 0.3 nM) was 10-fold more potent than difluoro-
pyrrolidine analog 5j (2.8 nM) and 1000 times more po-
tent than the amide 5k (360 nM). Compound 5f was also
significantly more potent than the indoline analogs 5l
and 5m, suggesting that small to medium sized substitu-
ents are preferred at this position. Substitution in the
phenethylpyrrolidine ring also influenced receptor affin-
ity. This is demonstrated by the six-fold increase in po-
tency for methylpyrrolidine analog 5g (0.05 nM)
compared to unsubstituted pyrrolidine analog 5f, and
the apparent preference for the S-hydroxymethyl frag-
ment in 5i over the R-isomer in 5h, a trend that has been
observed with other H3 antagonists.12
The initial target compounds were prepared using the
methods shown in Scheme 1. Commercially available
azomethine ylide precursor 6 was treated with cyclo-
pentenone 7 to give bicyclic ketone 8 via 3 + 2 cycload-
dition. Reaction of 8 with the alkyllithium reagent
derived from silyl-ether 9 provided the alcohol 10, which
was dehydrated and deprotected with HCl then acti-
vated with methanesulfonyl chloride to give mesylate
derivative 11. Reaction of 11 with various secondary
amines under basic conditions provided compounds
5a–m, which enabled exploration of phenethylamine
substituent effects on binding.
Figure 2. Design of H3 antagonists based on Conessine.
Scheme 1. Synthesis of target compounds 5. Reagents and conditions: (a) CH3CN, rt, 69%; (b) n-BuLi, THF, ꢀ78 °C, 64–66%; (c) HCl, i-PrOH,
60 °C, 98%; (d) MsCl, Et3N, CH2Cl2, rt, 82%; (e) HNR1R2, Na2CO3, CH3CN, microwave, 120 °C, 25–60%; (f) NH4CHO2, Pd(OH)2/C, MeOH, 98%;
(g) i—RCHO or R2CO, NaBH(OAc)3, AcOH, CH2Cl2, rt, 22–88%; or ii—ArBr, Pd(dba)2, t-BuOK, DMSO, 175 °C, 8–31%.

1492
V. J. Santora et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1490–1494
Table 1. Binding Affinities (Ki and pKi) of compounds 5a–t at rat H3 receptora
R1
N
R2
N
5a-t
R3
Compound
R1
Bn
NR2R3
Rat H3 Kib (nM)
Rat pKib
H
N
5a
2.3
8.64 0.24
H
N
5b
5c
Bn
Bn
2.4
0.4
8.62 0.29
9.43 0.09
H
N
H
N
5d
5e
5f
Bn
Bn
Bn
Bn
Bn
Bn
1.7
8.77 0.27
7.23 0.24
9.51 0.25
10.23 0.25
9.29 0.28
9.80 0.12
O
H
N
58
N
H
N
0.3
H
N
5g
5h
5i
0.05 (h Antag. = 0.1)d
0.51 (h Antag. = 8.5)d
0.16 (h Antag. = 1.5)d
H
N
OH
OH
H
N
H
N
5j
Bn
Bn
Bn
Bn
2.8
8.56 0.15
6.88 0.14
NAc
F
F
H
N
5k
5l
310
O
H
N
NAc
5m
5n
5o
256
6.59 0.20
9.95 0.09
9.38 0.30
HN
H
N
0.1
H
N
0.4 (h Antag. = 1.7)d

V. J. Santora et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1490–1494
1493
Table 1 (continued)
Compound
R1
NR2R3
Rat H3 Kib (nM)
Rat pKib
H
N
5p
0.9
9.02 0.04
7.17 0.39
8.63 0.19
7.45 0.06
8.17 0.19
9.12 0.18
10.35 0.10
10.17 0.21
10.11 0.44
9.53 0.13
H
N
5q
67
H
N
5r
2
N
H
N
5s
36
N
N
N
N
H
N
5t
7
H
N
(3aS, 6aS)-5f
(3aR, 6aR)-5f
(3aR, 6aR)-5n
(3aR, 6aR)-5o
(3aR, 6aR)-5p
Bn
Bn
0.75
H
N
0.05 (h Antag. = 0.6, Inverse Ag. = 2.0)d
H
N
0.07 (h Antag. = 0.9, Inverse Ag. = 1.0)d
H
N
0.08
0.29
H
N
a Displacement of N-[3H]-methylhistamine from rat cortex membranes.11
b Values are reported as average of n P 3 independent measurements for all compounds. Errors are log SD.
c NA, no activity below 5 lM.
d Antagonism determined by displacement of [3H] R(ꢀ)-a-methylhistamine from recombinant CHO-K1 cells expressing the human H3 receptor
(n = 1). Inverse Agonism determined by GTPcS binding in CHO-K1 cells expressing the human H3 receptor in the absence of agonist (n = 1).
Testing performed at MDS Pharma Services (Taiwan).14
The basicity and substitution of the bicyclic nitrogen
was also important in determining potency. Substitution
of the bicyclic nitrogen with simple alkyl groups pro-
duced compounds 5n–p, which exhibited sub-nanomolar
potency similar to that seen for the N-benzyl derivative
5f. In contrast, the less basic N-aryl analogs 5q–t were
significantly less potent than the N-alkyl analogs
although some still bound with affinities in the low
nM range.
on the stereochemistry of (3aS, 6aR)-8, which was as-
signed by X-ray crystallography.13 The same isomer of
8 was then used to prepare the (3aR, 6aR) isomers of
5n–p, which were significantly more potent than their
respective racemic mixtures.
In order to broadly assess binding affinity at the human
H3 receptor, six compounds from the new series were
tested in a human H3 receptor binding assay (Table
1).14 In each case there was a shift toward lower potency
at the human receptor that ranged from 3- to 20-fold
compared to that at the rat receptor. Compounds
(3aR, 6aR)-5f and (3aR, 6aR)-5n also displayed inverse
agonist activity at the human receptor in the low nano-
molar range when tested in an H3 GTPcS—functional
assay in the absence of agonist (Table 1) and displayed
Having confirmed that the new scaffold could be used to
identify potent H3 antagonists, the preferred stereo-
chemistry of the bicyclic moiety was examined. To this
end, both isomers of 5f were prepared from the resolved
isomers of ketone 8. The more potent isomer of 5f was
identified as having the 3aR, 6aR configuration based

1494
V. J. Santora et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1490–1494
Morisset, S. A.; Rouleau, A.; Ligneau, X.; Gbahou, F.;
2.5
2.0
1.5
1.0
0.5
0.0
Tardivel-Lacombe, J.; Stark, H.; Schunack, W.; Ganellin,
C. R.; Schwartz, J.-C.; Arrang, J.-M. High Nat. 2000, 408,
860; (d) Garduno-Torres, B.; Trevino, M.; Gutierrez, R.;
Arias-Montano, J.-A. Neuropharmacology 2007, 52, 527;
(e) Threlfell, S.; Cragg, S.; Kallo, I.; Turi, G.; Coen, C.;
Greenfield, S. J. Neurosci. 2004, 24, 8704.
##
**
3. (a) Medhurst, A.; Briggs, M.; Bruton, G.; Calver, A.;
Chessell, I.; Crook, B.; Davis, J.; Davis, R.; Foley, A.;
Heslop, T.; Hirst, W.; Medhurst, S.; Ociepka, S.; Ray, A.;
Regan, C. M.; Sargent, B.; Schogger, J.; Stean, T.; Trail,
B.; Upton, N.; White, T.; Orlek, B.; Wilson, D. Biochem.
Pharmacol. 2007, 73, 1182; (b) Passani, M.; Lin, J.-S.;
Hancock, A.; Crochet, S.; Blandina, P. Trends Pharmacol.
Sci. 2004, 25, 618; (c) Hancock, A.; Brune, M. Expert
Opin. Investig. Drugs 2005, 14, 223; (d) Bernaerts, P.;
Lamberty, Y.; Tirelli, E. Behav. Brain Res. 2004, 154, 211;
(e) Parmentier, R.; Anaclet, C.; Guhennec, C.; Brousseau,
E.; Bricout, D.; Giboulot, T.; Bozyczko-Coyne, D.;
Spiegel, K.; Ohtsu, H.; Williams, M.; Lin, J. S. Biochem.
Pharmacol. 2007, 73, 11571; (f) Esbenshade, T.; Fox, G.;
Cowart, M. Mol. Interv. 2006, 59, 77.
VEH VEH 0.3
1
3
+RAMH (mg/kg PO)
Chart 1. Inhibition of (R)-a-methylhistamine induced drinking after
oral administration of compound (3aR, 6aR)-5n.
greater than 1000-fold selectivity for the H3 receptor
when screened against a panel of over 100 human
GPCR’s, including H1, H2, and H4.
4. (a) Murray, M. Drug Metab. Rev. 1987, 18, 55; (b)
Yang, R.; Hey, J.; Aslanian, R.; Rizzo, C. Pharma-
cology 2002, 66, 128; (c) Halpert, J. R.; Guengerich,
F. P.; Bend, J. R.; Correia, M. A. Toxicol. Appl.
Pharmacol. 1994, 125, 163.
5. Leurs, R.; Tulp, M.; Menge, W.; Adolfs, M.; Zuiderveld,
O.; Timmerman, H. Br. J. Pharmacol. 1995, 116, 2315.
6. Celanire, S.; Wijtmans, M.; Talaga, P.; Leurs, R.; de Esch,
I. J. P. Drug Discovery Today 2005, 10, 1613.
7. Conessine’s activity at the H3 receptor has been reported
by others, see: Zhao, Chen; Bennani, Y. L.; Gopalakrish-
nan, S.; Sun, M.; Esbenshade, T. A.; Krueger, K. M.;
Miller, T. R.; Witte, D. G.; Marsh, Kennan C.; Cowart,
M. D.; Hancock, A. A. Abstracts of Papers, 230th ACS
National Meeting, Washington, DC, United States, Aug.
28–Sept. 1, 2005, MEDI-104.
8. Dvorak, C.; Apodaca, R.; Barbier, A.; Berridge, C.;
Wilson, S.; Boggs, J.; Xiao, W.; Lovenberg, T.; Carru-
thers, N. J. Med. Chem. 2005, 48, 2229.
9. Jesudason, C.; Beavers, L.; Cramer, J.; Dill, J.; Finley, D.;
Lindsley, C.; Stevens, F. C.; Gadski, R.; Oldham, S.;
Pickard, R.; Siedem, S.; Sindelar, D.; Singh, A.; Watsona,
B.; Hipskind, P. Bioorg. Med. Chem. Lett. 2006, 16, 3415.
10. Sun, M.; Zhao, C.; Gfesser, G.; Thiffault, C.; Miller, T.;
Marsh, K.; Wetter, J.; Curtis, M.; Faghih, R.; Esbenshade,
T.; Hancock, A. A.; Cowart, M. J. Med. Chem. 2005, 48,
6482.
11. Esbenshade, T.; Fox, G.; Krueger, K.; Miller, T.; Kang,
C.; Denny, L.; Witte, D.; Yao, B.; Pan, L.; Wetter, J.;
Marsh, K.; Bennani, Y.; Cowart, M.; Sullivan, J.; Han-
cock, A. J. Pharmacol. Exp. Ther. 2005, 313, 165.
12. Cowart, M.; Pratt, J. K.; Stewart, A. O.; Bennani, Y. L.;
Esbenshade, T. A.; Hancock, A. A. Bioorg. Med. Chem.
Lett. 2004, 14, 689.
Preliminary evaluation of the rat pharmacokinetics of
selected compounds from the new series was used to se-
lect a compound for in-vivo study. In general, the dia-
mine analogs exhibited good to excellent oral
bioavailability, long half-lives, and extensive tissue dis-
tribution. The pharmacokinetic profile of (3aR, 6aR)-
5n (F = 78%, t1/2 = 6 h) was considered the most promis-
ing and this compound was chosen for further testing.
Functional antagonism of the H3 receptor in vivo was
demonstrated for (3aR, 6aR)-5n in a rat dipsogenia
model, in which an acute drinking response induced
by an H3 agonist ((R)-a-methylhistamine) is attenuated
by pre-administration of an H3 antagonist.15 As shown
in Chart 1, compound (3aR, 6aR)-5n (0.3, 1 and 3 mg/
kg) or vehicle was administered orally to rats prior to
the subcutaneous administration of 10 mg (R)-a-methyl-
histamine. Compound (3aR, 6aR)-5n was found to inhi-
bit agonist induced drinking in a dose dependent
manner with a minimum effective dose of 3 mg/kg.
In summary, we have developed a new family of H3
antagonists based on the natural product Conessine.
Several members of this family exhibit sub- to low-nano-
molar potency at rat and human H3 receptors. Prelimin-
ary in-vitro and in-vivo screening results suggest that the
new compounds possess promising selectivity and phar-
macokinetic profiles. Results for compound (3aR, 6aR)-
5n, which displays in-vivo activity consistent with that of
a potent, orally available H3 antagonist, are particularly
noteworthy.
13. Single isomers of 8 (>99.5%ee) were isolated by diaste-
reomeric salt formation, starting from racemic 8 and L- or
D- di-p-toluoyl-tartaric acid, followed by repetitive recrys-
tallization from isopropanol. Crystallographic data for
(3aS, 6aR)-8ÆL-di-p-toluoyl-tartaric acid have been depos-
ited with the Cambridge Crystallographic Data Centre
(CCDC 670453). Optical rotation for (3aS, 6aR)-8ÆL-di-p-
References and notes
1. (a) Hill, S.; Ganellin, C.; Timmerman, H.; Schwartz, J.;
Shankley, N.; Young, J.; Schunack, W.; Levi, R.; Haas, H.
Pharmacol. Rev. 1997, 49, 253; (b) Hough, L. B. Mol.
Pharmacol. 2001, 59, 415.
2. (a) Schwartz, J. C.; Arrang, J. M.; Garbarg, M.; Pollard,
H.; Ruat, M. Physiol. Rev. 1991, 71, 1; (b) Leurs, R.; Smit,
M. J.; Timmerman, H. Pharmacol. Ther. 1995, 66, 413; (c)
22
toluoyl-tartaric acid: ½aꢁD ¼ ꢀ113:8 (c 0.0016, EtOH).
14. Human H3 binding and functional (GTPcS) assays were
performed by MDS Pharma Services, Taiwan (http://
www.mdsps.com).
15. Clapham, J.; Kilpatrick, G. J. Eur. J. Pharmacol. 1993,
232, 99.