Discovery of substituted benzyl tetrazoles as histamine H3 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.07.009

A series of potent and subtype selective H3 receptor antagonists containing a novel tetrazole core and diamine motif is reported. A one-pot multi-component Ugi reaction was utilised to rapidly develop the structure-activity relationships (SAR) of these compounds. Optimisation for liver microsome stability (t_1/2 >60 min), minimal CYP inhibition (IC₅₀ >50 μM) and high cell permeability (Caco-2 P_app >20 × 10^-6 cm/s) identified several compounds with drug-like properties.

Full text of DOI:10.1016/j.bmcl.2010.07.009

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5165–5169
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of substituted benzyl tetrazoles as histamine H3 receptor antagonists
a
a
a
a
Adam J. Davenport a, , Christopher C. Stimson , Massimo Corsi , Darshan Vaidya , Edward Glenn ,
*
Timothy D. Jones a, Sarah Bailey a, Mark J. Gemkow b, Ulrike Fritz b, David J. Hallett a
a Evotec (UK) Ltd, 114 Milton Park, Abingdon, Oxfordshire OX14 4SA, United Kingdom
b Evotec AG, Schnackenburgallee 114, 22525 Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
AseriesofpotentandsubtypeselectiveH3receptorantagonistscontaininganoveltetrazolecoreand diamine
motif is reported. A one-pot multi-component Ugi reaction was utilised to rapidly develop the structure–
activity relationships (SAR) of these compounds. Optimisation for liver microsome stability (t1/2 >60 min),
Received 14 May 2010
Revised 2 July 2010
Accepted 3 July 2010
minimal CYP inhibition (IC50 >50
l
M) and high cell permeability (Caco-2 Papp >20 Â 10À6 cm/s) identified
several compounds with drug-like properties.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Histamine
H3R
Antagonists
CNS
GPCR
Drug discovery
The histamine H3 receptor (H3R) is primarily located in the cen-
tral nervous system and exists as one of four known subtypes
alongside the H1, H2 and H4 receptors. With marketed therapeutic
agents for the H1R and H2R having been highly successful, the
more recently discovered H3R1 and the H4R are attracting signifi-
cant interest from both academia and industry as potential thera-
peutic targets.2,3 Located mainly in the basal ganglia, hippocampus
and cortical areas,4,5 the H3R subtype acts as an inhibitory auto-
and hetero-G-protein coupled receptor. Activation of the receptor
decreases the release of histamine whereas antagonism (or inverse
agonism) increases the release of histamine.1,6 As a heteroreceptor,
the H3R is capable of regulating the release of other important
neurotransmitters, such as acetylcholine, dopamine, serotonin
and norepinephrine.7–10 Preclinical data, as well as publicly avail-
able clinical data,11,12 support the potential utility of H3R antago-
nists in the amelioration of multiple CNS disorders, including
ADHD, AD, schizophrenia and narcolepsy.13–17
Early H3R antagonists based around the imidazole core18 of his-
tamine caused inhibition of cytochrome P450 enzymes and were
probably not progressed into the clinic due to the inherent risk of
undesirable drug–drug interactions (DDI’s).19,20 Various laborato-
ries have subsequently developed non-imidazole based H3 receptor
antagonists including GSK189254, ABT-288, PF-03654746, MK-
0249, JNJ-17216498 and BF2.649 (see Fig. 1 for disclosed struc-
tures).13 These have entered clinical trials for AD, ADHD, schizo-
phrenia, narcolepsy and excessive daytime sleepiness.
Attracted by the potential of the H3R as a drug target, a high
throughput screen (HTS) of Evotec’s screening collection was per-
formed. This resulted in the identification of compound 1a; a mod-
erately potent human H3R antagonist with selectivity versus H1,
O
F
O
H
N
N
O
N
N
H
N
N
Cl
O
F
BF2.649 (Tiprolisant)
GSK189254
PF-03654746
Figure 1. Disclosed structures of clinical compounds.
* Corresponding author. Tel.: +44 (0) 1235 838807; fax: +44 (0) 1235 441509.
E-mail address: Adam.Davenport@evotec.com (A.J. Davenport).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.07.009

5166
A. J. Davenport et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5165–5169
and revealing the preference of the receptor for small sterically
C
O
N N
N+
demanding alkyl groups. Potent examples (2d, 2e and 2i) demon-
strated >500-fold selectivity for the H3R over H1, H2 and H4 recep-
tor subtypes. The in vitro stabilities of 2d and 2i were assessed
using human and rat liver microsome assays (HLM and RLM,
respectively) indicating rapid clearance (t1/2 = <5 min). Metabolite
identification studies on 2d HLM incubate revealed the des-cyclo-
butyl compound as the major product. Formation of this metabo-
lite can be rationalised via hydroxylation of the cyclobutyl group
and subsequent elimination to reveal the secondary amine. Loss
of the tertiary amine environment of 2d had a deleterious effect
on H3R potency.
N
N
R1
a
+
R1
H
N
R2
N
H
R2
F
2a-j
F
Scheme 1. Reagents and condition: (a) TMS azide, MeOH, rt, 18 h. Isolated yields
range from 11% to 64%.
H2 and H4 receptors. In this report, the discovery, SAR and devel-
opment to potent and selective H3R antagonists are described.
Early investigations around the novel tetrazole core were under-
taken to investigate the SAR around this class of compound.
It was rationalised that improved stability could be achieved by
blocking the major site of metabolism. This led to the synthesis of
substituted piperazine analogues 3a–l (Table 2).
Compounds 3a and 3b gave excellent human H3R potency (of
2.2 and 3.3 nM, respectively) but shielding around nitrogen did
not afford an improvement in metabolic stability. Complete re-
moval of the alkyl ring in 3c resulted in a marked loss in potency.
Further attempts at rigidifying the piperazine ring whilst sterically
shielding the distal nitrogen (compounds 3d and 3e) resulted in up
to 20-fold loss of potency with no improvement in stability.
Replacement of the cyclobutyl group of 3a gave a loss in potency
(3f–h), although a minor increase in stability was observed with
the acetylene capped compound 3f. Moving the distal basic amine
to an exocyclic position was accompanied by reduced H3R activity
in all cases (3i–l), although compound 3j gave a significant
improvement in microsomal stability.
Compound 2i was selected for further optimisation on the basis
of a superior early ADME profile, compared to 2d. Retaining the ac-
tive amine of 2i and modifying the aromatic substituents would al-
low further SAR to be elucidated and potentially enhance potency.
It was anticipated that a global change in molecular shape could
also leave the metabolic site less prone to oxidation. Compounds
4a–r (Table 3) were thus synthesised following the route previ-
ously disclosed in Scheme 1.
N N
hH3 IC50 = 0.9 µM
rH3 IC50 >20 µM
hH1 IC50 >20 µM
hH2 IC50 >20 µM
hH4 IC50 >20 µM
N
N
N
N
1a
Initial work was carried out varying the cyclic amine to investigate
the importance and positioning of the distal basic nitrogen. The
multicomponent Ugi reaction21 was utilised to give rapid access to
the desired products in low to high isolated yield as shown in
Scheme 1.22 Reaction of 4-fluorobenzaldehyde with benzylisocya-
nide, TMS-azide and a range of amine reagents gave a convenient
one-pot approach to a series of analogues (2a–j) as shown in Table 1.
The first results highlighted the importance of the basic
nitrogen where replacement with either carbon or oxygen led to
a complete loss of activity (2a and 2b). Increasing the number of
rotatable bonds in 2c also showed a marked decrease in potency.
Retaining the six-membered piperazine ring and investigating dif-
ferent alkyl piperazine substituents in derivatives 2d–f gave im-
proved activity; optimal potency of ꢀ25 nM was observed for
cyclobutyl and cyclopentyl analogues 2d and 2e. A similar trend
was seen for homopiperazine derivatives 2g–j demonstrating the
optimal positioning and environment of the basic nitrogen atom
Table 3 shows the potencies of ortho (4a and 4b), meta (4c and
4d) and para (4e–r) substituted compounds. A significant drop in
potency was seen for ortho examples 4a and 4b alongside a smaller
Table 1
Human H3R activities of compounds 2a–j
N N
N
N
R1
N
R2
F
Compound
2f
a
a
Compound
NR1R2
hH3R IC50 (nM)
20,000 ( 0)
NR1R2
hH3R IC50 (nM)
2a
285 ( 160)
738 ( 297)
N
N
N
N
N
N
O
N
2b
2c
2d
20,000 ( 0)
2742 ( 128)
25 ( 3.5)
2g
2h
2i
N
N
N
94 ( 24.5)
47 ( 16.7)
356 ( 88)
N
N
N
N
N
N
N
N
2e
31 ( 14.3)
2j
See Ref. 22 for complete details of hH3R assay conditions.
a
Values are arithmetic means of two or more experiments, standard deviation is given in parentheses. Differences of <2-fold should not be considered significant.

A. J. Davenport et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5165–5169
5167
Table 2
Human H3R activities for compounds 3a–l
N N
N
N
R1
N
R2
F
NR1R2
hH3R IC50 (nM)
HLM t1/2 (min)
Compound
NR1R2
hH3R IC50 (nM)
HLM t1/2 (min)
a
a
Compound
N
N
3a
2.2 ( 0.8)
3.4 ( 1.7)
8
3
3g
4894 ( 1871)
20,000 ( 0)
3
O
N
N
N
N
N
N
3b
3h
—
N
N
N
N
3c
1761 ( 676)
241 ( 73)
—
7
3i
3j
3612 ( 840)
2431 ( 686)
—
HN
N
N
N
3d
35
N
N
N
N
N
3e
3f
516 ( 192)
184 ( 52)
7
3k
3l
2470 ( 927)
—
—
N
N
N
N
N
N
17
14,215 ( 8181)
See Ref. 22 for complete details of hH3R assay conditions.
a
Values are arithmetic means of two or more experiments, standard deviation is given in parentheses. Differences of <2-fold should not be considered significant. HLM,
human liver microsome.
decrease with meta substituted examples 4c and 4d, compared to
the para analogues 4i and 4k that retained activity. A large range of
para substituted functionalities were tolerated including alkyl (4f
Table 3
Human H3R activities for compounds 4a–r
and 4g), alkoxy (4h–j), halo (4k and 4l), amide (4m and 4p), sul-
fone (4q) and sulfonamide (4r) groups. The comparable potencies
observed with both electron withdrawing and donating groups
suggests that the electron density on the aromatic ring does not
significantly influence the binding. Compared to the unsubstituted
analogue 4e, only amide, sulfone and sulfonamide examples 4n–r
gave significantly improved potencies (up to IC50 = 0.3 nM for
4o). We inferred that further amino acid residues might be avail-
able for additional hydrogen bonding interactions within the
receptor binding site, which presented further opportunity for po-
tency optimisation. A modest increase in human microsomal sta-
bility was observed with alkoxy substituents (4h–j) with half-
lives of up to 21 min (4j). Gratifyingly, amides 4n and 4p as well
as sulfone 4q were more stable still, with half-lives of up to 40 min.
Encouraged by these results, a range of analogues containing
para substituted heterocycles was synthesised to take advantage
of these findings (Table 4).
Morpholine and piperazine analogues 5b–d showed excellent
sub-nanomolar potencies, although 5a was unexpectedly less po-
tent (IC50 = 550 nM). Compound 5d had an improved half-life of
74 min and demonstrated the tolerance of the receptor to an addi-
tional basic centre.
Five-membered aromatic heterocycles 5e–k were potent
(IC50 <10 nM), again demonstrating the tolerance for a wide range
of functionalities. Unfortunately, microsomal stabilities were gen-
erally undesirably low (t1/2 = <25 min). The addition of the methy-
lene spacer in 5m retained potency and gave an increase in half-life
N N
N
N
N
N
R
a
Compound
R substituent
hH3R IC50 (nM)
HLM t1/2 (min)
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4q
4r
2-Cl
2-OCHF2
3-Cl
475 ( 109)
1891 ( 535)
64 ( 34)
58 ( 2.9)
17 ( 3.4)
28 ( 0.7)
42 ( 20)
14 ( 9)
—
—
—
13
—
9
—
12
9
21
21
—
—
35
—
34
40
—
3-OCHF2
4-H
4-iPr
4-CF3
4-OEt
4-OCHF2
4-OCF3
4-Cl
28 ( 4)
51 ( 26)
34 ( 11)
115 ( 26)
19 ( 6)
3.0 ( 2.3)
0.3 ( 0.04)
21 ( 7)
4-Br
4-CONH2
4-CONHMe
4-CONHPh
4-Pyrrolidin-2-One
4-SO2Me
1.6 ( 0.4)
6.8 ( 1.2)
4-SO2NMe2
See Ref. 22 for complete details of hH3R assay conditions.
a
Values are arithmetic means of two or more experiments, standard deviation is
given in parentheses. Differences of <2-fold should not be considered significant.
HLM, human liver microsome.

5168
A. J. Davenport et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5165–5169
Table 4
Human H3R activities and microsomal stabilities for 5a–p and 6a–g
N N
N N
N
N
N
N
N
N
R
N
N
R
5a-p
6a-g
a
a
Compound
R substituent
hH3R IC50 (nM)
HLM t1/2 (min)
Compound
R substituent
hH3R IC50 (nM)
HLM t1/2 (min)
N
N
O
N
5a
551 ( 435)
—
5m
4.6 ( 1.9)
32
N
N
5b
5c
0.6 ( 0.3)
14
23
5n
5o
5.9 ( 0.6)
0.9 ( 0.2)
16
—
O
N
N
O
N
N
0.5 ( 0.002)
N
O
N
N
5d
5e
5f
0.4 ( 0.4)
6.5 ( 1.6)
1.2 ( 0.4)
74
5
5p
6a
6b
12 ( 7)
13
20
23
N
N
N
58.5 ( 27.7)
41 ( 7)
O
7
O
N
N
N
NH
5g
5h
0.2 ( 0.04)
1.6 ( 0.1)
25
12
6c
119 ( 33)
98 ( 13)
—
9
N
N
6d
N
N
H
N
N
O
5i
9.2 ( 1.5)
—
6e
32 ( 14.4)
10
N
N
N
S
N
N
5j
0.4 ( 0.1)
3.6 ( 1.7)
0.7 ( 0.2)
11
8
6f
33 ( 3)
27
—
Me
S
N
5k
5l
6g
102 ( 7.6)
N
N
N
11
N
See Ref. 22 for complete details of hH3R assay conditions.
a
Values are arithmetic means of two or more experiments, standard deviation is given in parentheses. Differences of <2-fold should not be considered significant. HLM,
human liver microsome.
to 32 min (relative to 5h). The pyridine and pyrimidine substitu-
ents in examples 5n–p also exhibited good potency but offered
no improvement in microsomal stability relative to the 5-mem-
bered aromatic heterocycles. The introduction of fused heterocy-
cles in 6a–g reduced potency and provided no improvement in
stability. Tetrazole compounds exhibit low species specificity for
human over rat H3R, data for representative examples is shown
in Table 5, with potent compound 5d being equipotent at both hu-
man and rat H3R.
The most promising compounds were profiled in further in vitro
assays to assess their cell permeability properties and CYP450
inhibition.
As depicted in Table 5, the permeability data obtained from
Caco-2 monolayer studies show, with the exception of 5h and
6b, that the compounds tested have a low rate of transport across
the cell membrane. The efflux ratios of >40 indicated that these
compounds are also likely to be substrates for ABC transporters
(such as P-glycoprotein). Gratifyingly, Caco-2 data from com-
pounds 5h and 6b suggested that discrete structural modifications
can circumvent these issues to give highly permeable compounds
with minimal efflux. The CYP450 inhibition data highlighted that
the compounds tested generally have the propensity to inhibit
CYP2D6 (IC50 = 0.6–1.7 lM) and in some cases (5h, 5m and 6b)
moderate inhibition of CYP3A4 was also observed. Methylsulfone
4q had a clean CYP450 profile, highlighting how varying substitu-
ents on the phenyl ring can have a significant effect on the
in vitro profile.
In conclusion, we report here the discovery and development of
a series of potent and subtype selective histamine H3R antagonists.
In vitro testing demonstrated that selected examples have moder-
ate microsomal stability and that structural modifications gave
highly permeable compounds with minimal CYP450 inhibition. Fur-
ther optimisation is ongoing to give the desired balance of proper-
ties and will be reported in follow-up publications.

A. J. Davenport et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5165–5169
5169
Table 5
In vitro data for selected compounds
N N
N
N
N
R
N
a
a
Compound
R substituent
MW
TPSA
79.2
C log P
hH3R IC50
(nM)
rH3R IC50
(nM)
HLM t1/2
(min)
Caco-2
A-Bb
Caco-2
CYP2D6
CYP3A4
B-A/A-Bb
Inhib.c
(
lM)
Inhib.c
(lM)
H
N
4n
459.6
2.2
3.0 ( 2.3)
12 ( 5)
35
0.3
127
1.4
>50
O
4q
5d
5h
480.6
514.7
469.6
84.2
56.6
80.8
1.9
3.8
3.2
1.6 ( 0.4)
0.4 ( 0.4)
1.6 ( 0.1)
17 ( 8)
40
74
12
0.9
0.8
104
62
46
2.5
>50
1.7
0.7
>50
>50
29
S
O
O
N
0.4 ( 0.3)
9.2 ( 3.5)
17 ( 5)
N
N
N
N
N
N
5m
6b
483.6
442.6
80.8
63.2
3.0
4.2
4.6 ( 1.9)
41 ( 7)
32
23
0.5
41
90
0.6
1
12
N
138 ( 46)
0.7
2.4
O
See Ref. 22 for complete details of hH3R assay conditions.
a
Values are arithmetic means of two or more experiments, standard deviation is given in parentheses. Differences of <2-fold should not be considered significant. HLM,
human liver microsome.
b
Caco-2 units: Papp 1 Â 10À6 cm/s.
c
CYP P450 inhibition reported as IC50 values.
8. Schlicker, E.; Fink, K.; Hinterthaner, M.; Göthert, M. Naunyn-Schmiedeberg’s
Acknowledgements
Arch. Pharmacol. 1980, 6, 633.
9. Schlicker, E.; Fink, K.; Detzner, M.; Göthert, M. J. Neural Transm–General Section
The authors thank the ADMET group (notably Anna Pedret-
Dunn and Wayne Thomas) for performing ADME studies, Mark
Cobb for chemical syntheses and Andreas F. Kahrs, Christian Kirch-
hoff and the HTS Group for conducting the H3R HTS.
1993, 1, 1.
10. Clapham, J.; Kilpatrick, G. J. Br. J. Pharmacol. 1992, 4, 919.
11. Lin, J.-S.; Dauvilliers, Y.; Arnulf, I.; Bastuji, H.; Anaclet, C.; Parmentier, R.;
Kocher, L.; Yanagisawa, M.; Lehert, P.; Ligneau, X.; Perrin, D.; Robert, P.; Roux,
M.; Lecomte, J.-M.; Schwartz, J.-C. Neurobiol. Dis. 2008, 30, 74.
12. http://www.gsk.com/investors/presentations/2007/neurosciences-seminar-
dec07/jackie-hunter.pdf.
13. Gemkow, M. J.; Davenport, A. J.; Harich, S.; Ellenbroek, B. A.; Cesura, A.; Hallett,
D. Drug Discovery Today 2009, 14, 509.
14. Passani, M. B.; Lin, J.-S.; Hancock, A.; Crochet, S.; Blandina, P. Trends Pharmacol.
Sci. 2004, 25, 618.
15. Wijtmans, M.; Leurs, R.; de Esch, I. Expert Opin. Invest. Drugs 2007, 7, 967.
16. Bongers, G.; Bakker, R. A.; Leurs, R. Biochem. Pharmacol. 2007, 8, 1195.
17. Hancock, A. A. Biochem. Pharmacol. 2006, 8, 1103.
18. Cowart, M.; Altenbach, R.; Black, L.; Faghih, R.; Zhao, C.; Hancock, A. A. Mini Rev.
Med. Chem. 2004, 9, 979.
References and notes
1. Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Nature 1983, 302, 832.
2. Celanire, S.; Wijtmans, M.; Talaga, P.; Leurs, R.; de Esch, I. J. P. Drug Discovery
Today 2005, 10, 1613.
3. Thurmond, R. L.; Gelfand, E. W.; Dunford, P. J. Nat. Rev. Drug Discovery 2008, 7,
41.
4. Drutel, G.; Peitsaro, N.; Karlstedt, K.; Wieland, K.; Smit, M. J.; Timmerman, H.;
Panula, P.; Leurs, R. Mol. Pharmacol. 2001, 59, 1.
5. Martinezmir, M.-I.; Pollard, H.; Moreau, J.; Arrang, J. M.; Ruat, M.; Traiffort, E.;
Schwartz, J. C.; Palacios, J. M. Brain Res. 1990, 526, 322.
6. Morisset, S.; Rouleau, A.; Ligneau, X.; Gbahou, F.; Tardivel-Lacombe, J.; Stark,
H.; Schunack, W.; Ganellin, C. R.; Schwartz, J. C.; Arrang, J. M. Nature 2000, 408,
860.
19. Boxenbaum, H. J. Pharm. Pharm. Sci. 1999, 2, 47.
20. Lin, J. H.; Lu, A. Y. H. Clin. Pharmacokinet. 1998, 35, 361.
21. Ugi, I.; Meyer, R.; Fetzer, U.; Steinbruckner, C. Angew. Chem. 1959, 71, 386.
22. For example, the synthesis of compound 2i is described in Ref. 23.
23. Davenport, A. J.; Hallett, D. J.; Stimson, C. C.; Corsi, M.; Gemkow, M.
WO20095394, 2009.
7. Schlicker, E.; Betz, R.; Göthert, M. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1988, 337, 588.