Reduction of CYP450 inhibition in the 4-[(1H-imidazol-4-yl)methyl] piperidine series of histamine H3 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.10.087

A novel series of histamine H₃ receptor antagonists based on the 4-[(1H-imidazol-4-yl)methyl]piperidine template displaying low CYP2D6 and CYP3A4 inhibitory profiles has been identified. Structural features responsible for the reduction of P450 activity, a typical liability of 4-substituted imidazoles, have been established.

Full text of DOI:10.1016/j.bmcl.2005.10.087

Bioorganic & Medicinal Chemistry Letters 16 (2006) 989–994
Reduction of CYP450 inhibition in the
4-[(1H-imidazol-4-yl)methyl]piperidine series of
histamine H3 receptor antagonists
Michael Berlin,* Pauline C. Ting, Wayne D. Vaccaro, Robert Aslanian,
Kevin D. McCormick, Joe F. Lee, Margaret M. Albanese, Mwangi W. Mutahi,
John J. Piwinski, Neng-Yang Shih, Luli Duguma, Daniel M. Solomon, Wei Zhou,
Rosy Sher, Leonard Favreau, Matthew Bryant, Walter A. Korfmacher,
Cymbelene Nardo, Robert E. West, Jr., John C. Anthes, Shirley M. Williams,
Ren-Long Wu, H. Susan She, Maria A. Rivelli, Michel R. Corboz and John A. Hey
The Schering Plough Research Institute, 2015 Galloping Hill Rd. Kenilworth, NJ 07033, USA
Received 19 August 2005; revised 25 October 2005; accepted 26 October 2005
Available online 15 November 2005
Abstract—A novel series of histamine H3 receptor antagonists based on the 4-[(1H-imidazol-4-yl)methyl]piperidine template
displaying low CYP2D6 and CYP3A4 inhibitory profiles has been identified. Structural features responsible for the reduction of
P450 activity, a typical liability of 4-substituted imidazoles, have been established.
Ó 2005 Elsevier Ltd. All rights reserved.
In the preceding paper, we described a series of novel his-
tamine H3 receptor antagonists based on the 4-[(1H-imi-
dazol-4-yl)methyl]piperidine scaffold. While displaying
high affinity for both guinea pig and human receptors
as well as excellent functional activity, these compounds
simultaneously had a high level of CYP450 enzyme inhi-
bition, in particular CYP2D6, a notorious liability of
4-substituted imidazoles limiting their use.1 Strong inhi-
bition of P450 enzymes may disrupt normal metabolism
of xenobiotics, potentially resulting in undesired drug–
drug interactions.2 We report here SAR investigations
that have led us to establish structural features within
the 1H-imidazol-4-yl series that allow suppression of
CYP2D6 and CYP3A4 inhibitory activity, while main-
taining H3 antagonist activity.
activity.3 While its P450 inhibitory activity was above
30 lM level with respect to both CYP2D6 and CYP3A4,
H3 activity left room for improvement. We decided to
test whether we could advantageously incorporate the
structural features of this analog into the previously
described noticeably more potent 4-[(1H-imidazol-4-
yl)methyl]piperidine series, exemplified by 2.
N
O
HN
N
N
O
N
2
1
Ki (H3, gp) = 4 nM
pA2 (gp ileum) = 8.3
IC50 (2D6) = 0.03 µM
IC50 (3A4) > 20 µM
Ki (H3, gp) = 33 nM
pA2 (gp ileum) = 7.5
IC50 (2D6, 3A4) > 30 µM
Our earlier unpublished investigations in this area iden-
tified 1 as a non-imidazole H3 antagonist of moderate
The compounds prepared for this study are listed in
Table 1. Direct crossover analogs between the two struc-
tural types, exemplified by 1 and 2, have the piperidine
ring in the position of either of the two basic nitrogen
sites in 1. In particular, replacement of the right-side
piperidine ring with the 4-[(1H-imidazol-4-yl)methyl]pi-
peridine fragment gave rise to 3—the first 4-substituted
imidazole analog devoid of 2D6 and 3A4 inhibitory
activity, but possessing only moderate H3 activity.
Keywords: Histamine; H3 antagonists; Imidazole series; 4-[(1H-imi-
dazol-4-yl)methyl]piperidine; Decongestion; Allergic rhinitis; CYP450
inhibition.
*
Corresponding author. Fax: +1 908 740 7152; e-mail:
michael.berlin@spcorp.com
Present address: Bristol-Myers Squibb Pharmaceutical Research
Institute, PO Box 4000, Princeton, NJ 08543-4000, USA.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.10.087

990
M. Berlin et al. / Bioorg. Med. Chem. Lett. 16 (2006) 989–994
Table 1. H3 binding affinity, functional activity, and CYP2D6 inhibition profile of H3 antagonists
HN
N
N
R
Ki (nM)a
pA2
IC50 (2D6) (lM)c
b
Compound
R
O
2
4
8.3
0.03
3
4
17
3
7.1
8.6
8.4
7.8
>30
O
N
O
O
14
N
N
5
1
3
O
6
16
5
>20
N
O
7
NDe
8.5
3
N
O
O
8
3
1d
N
9
4
8.8
6
N
O
N
10
11
12
13
14
15
16
17
18
19
20
13
16
3
7.3
>30
O
N
>30d
O
7.6
NH
N
O
O
O
ND
ND
ND
ND
ND
ND
ND
ND
ND
>30
N
N
F
F
4
6
6
0.3
O
O
O
O
3
2
15
15
12
9
1
N
O
O
O
0.9
N
H
>20
N
NH
N
4
N
N
O
45
6
N
(continued on next page)

M. Berlin et al. / Bioorg. Med. Chem. Lett. 16 (2006) 989–994
991
Table 1 (continued)
b
Compound
R
Ki (nM)a
pA2
IC50 (2D6) (lM)c
0.3
O
21
12
ND
N
O
O
22
23
52
12
ND
7.7
2
N
12
N
N
N
NH
O
O
24
25
26
34
7.7
2d
O
ND
5d
N
O
O
O
NH
26
27
28
23
315
45
7.8
>20
N
N
O
ND
ND
1
3
O
O
N
O
NH
29
30
31
51
10
24
ND
8.1
7.7
30
3
O
N
N
O
O
N
N
O
N
2
O
N
S
O
O
N
N
32
33
34
8
4
7.6
7.6
ND
6
2
9
N
H
N
H
H
N
40
O
N
a Inhibition of [3H]-N-a-methylhistamine binding to guinea pig brain receptor.4 H3 binding Ki values are the average of at least two independent
determinations. The assay-to-assay maximum variability in the series was 3-fold. More common variability was 2-fold.
b Antagonist potency in an electrically stimulated guinea pig ileum.5 pA2 values are the average of at least four independent determinations. The
assay-to-assay maximum variability in the series was 0.2.
c Human liver microsome assay,6 unless indicated otherwise. IC50 values are the average of at least three independent determinations. The assay-
to-assay maximum variability in the series was 15%.
d SupersomeTM assay.7 IC50 values are the average of at least three independent determinations. The maximum variability in the series was 15%.
For very good correlation between SupersomeTM and human liver microsome assays with respect to IC50 (CYP2D6) see ref. 7.
e Not determined.
At the same time substitution of the 4-[(1H-imidazol-4-
yl)methyl]piperidine moiety for the left-side nitrogen
center in 1 provided 4, a highly potent H3 antagonist with
significantly among the compounds listed in Table 1,
3A4 inhibitory activity was found not to be an issue in
this series (IC50 > 20 lM). Further studies established
structural features critical for the best combination of
high H3 and low CYP2D6 inhibitory activity.
a
favorably high threshold of 2D6 and 3A4
(IC50 > 30 lM) inhibition. While 2D6 activity varied

992
M. Berlin et al. / Bioorg. Med. Chem. Lett. 16 (2006) 989–994
Scheme 1. Reagents and conditions: (a) Methyl acrylate, toluene, 110 °C; (b) LiAlH4, THF, 80% (2 steps); (c) 4-hydroxybenzaldehyde, 1,10-
(azodicarbonyl)dipiperidine, Bu3P, THF, 90%; (d) R1R2NH, NaBH(OAc)3, CH2Cl2 or dichloroethane, 45–95%; (e) HCl, MeOH, D, 95%; (f) maleic
acid, MeOH, D, 95%.
Scheme 2. Reagents and conditions: (a) 35, EDC, HOBT, CH2Cl2, 60–80%; (b) LiAlH4, THF, 55–80%; (c) HCl, MeOH, D, 95%; (d) 35,
NaBH(OAc)3, CH2Cl2, 91%; (e) piperidine, NaBH(OAc)3, CH2Cl2, 75–95%; (f) ethyl 4-bromobutyrate, K2CO3, acetone–H2O, 63%; (g) NaOH,
MeOH–H2O-THF, 100%; (h) maleic acid, MeOH, D, 60–99%; (i) BrCH2CH2OH, Et3N, DMF, 60 °C, 40%; (j) SOCl2, CH2Cl2, 79%; (k) 1-(4-
hydroxybenzyl)piperidine 38 (see Scheme 3), NaH, KI, DMF, 50 °C, 38%; (l) HOCH2CH2CH2NMe2, 1,10-(azodicarbonyl)dipiperidine, Bu3P, THF,
80%; (m) HOCH2CH2CH2OH, 1,10-(azodicarbonyl)dipiperidine, Bu3P, THF, 70%; (n) CrO3, H2SO4, H2O, 90%; (o) OsO4 (cat.), NaIO4, acetone–
H2O, 95%; (p) RuO2 (cat.), NaIO4, CH3CN-H2O, 30%; (q) benzotriazole, COCl2, toluene, 60 °C; then, 40, Et3N, 35, CH2Cl2, 12%; (r) triphosgene,
Et3N, 35, CH2Cl2, 32%; (s) H2, 10% Pd/C, EtOAc-MeOH, 37%; (t) acryloyl chloride, Et3N, CH2Cl2, 55%; (u) 35, MeOH, 60%.
Within the 4-substituted imidazole series, typified by 4, a
low CYP2D6 inhibition profile seems to be largely
dependent on the structural characteristics of the part
of the molecule most distant from the imidazole ring.
Most compounds appear to benefit from the basic nat-
ure of this region. In the case of aliphatic amines, the
threshold of CYP2D6 inhibition is raised significantly,
compared to 2, often into the double-digit micromolar
range (6, 10, 11, 12, and 18), although the position of
the nitrogen seems to be critical (e.g., 4 vs 8 and 4 vs
5). Aromatic amines, as well as aromatic nitrogen
heterocycles, have generally been stronger CYP2D6
inhibitors (e.g., 16, 17, 190, and 21), possibly due to
the altered position of the nitrogen or reduced basicity
of these compounds.
than 4, while reintroduction of a basic site in 26 reestab-
lishes a much higher inhibition threshold. Modifications
in other parts of the molecule, although attempted, have
not generated clear SAR trends and generally did not
improve the P450 profile (e.g., 27, 29–34). Compound
34 is a crossover analog to the 4-[(1H-imidazol-4-
yl)methyl]piperidine series, described in the previous pa-
per, in which aniline amide was shown to be the most H3
active pharmacophore. However, with the introduction
of piperidinemethyl substituent in the 4-position H3
activity drops dramatically. Overall, 4-alkoxybenzyl
amines appear to have the best combination of H3 activ-
ity and CYP450 profile (e.g., 4, 12). The synthetic
approaches to compounds in Table 1 utilize previously
described trityl-protected imidazolylmethylpiperidine
359 and are shown in Schemes 1–4.
Other attempts to moderate or eliminate basicity of this
part of the molecule, either through selective substitu-
tion or removal of basic nitrogen, produced compounds
with a lower CYP2D6 inhibition threshold compared to
4 (e.g., 13, 14, and 15).8 In agreement with this trend,
amide derivatives 24 and 25 are stronger P450 inhibitors
Having achieved a significant separation between the H3
activity and undesired CYP450 inhibitory activity, we
investigated the pharmacokinetic profiles of several
compounds in this series. The compound with the best
biological profile, 4, showed no plasma levels when

M. Berlin et al. / Bioorg. Med. Chem. Lett. 16 (2006) 989–994
993
Scheme 3. Reagents and conditions: (a) Piperidine, NaBH(OAc)3, CH2Cl2, 95%; (b) 36, 1,10-(azodicarbonyl)dipiperidine, Bu3P, THF, 40–90%; (c)
HCl, MeOH, D, 80–95%; (d) piperidine, EDC, HOBT, CH2Cl2, 88%; (e) LiAlH4, THF; (f) cyclohexene, HgSO4, I2, CHCl3, 34%; (g) Bu3SnH, AIBN,
toluene, D, 39%; (h) NaOH, MeOH, 90%; (i) maleic acid, MeOH, D, 90%; (j) 1,5-dibromopentane, Et3N, CH2Cl2, 5%; (k) cyclohexanone,
NaBH(OAc)3, dichloroethane, 64%; (l) TFAA, pyridine, CH2Cl2, 93%; (m) H2, Pd(OH)2 (cat.), EtOAc, 100%; (n) NaBH4, EtOH, 100%; (o) 6-
hydroxyquinoline, 1,10-(azodicarbonyl)dipiperidine, Bu3P, THF, 75%; (p) 4-methoxyphenylboronic acid, Pd[dppf]2Cl2 (cat.), K3PO4, DME, 100 °C,
92–99%; (q) BBr3, CH2Cl2, 75–90%; (r) N-BOC-piperazine, EDC, HOBT, CH2Cl2; (s) K2CO3, MeOH, 50–65% (2 steps); (t) 2-piperidone, KHMDS,
THF, 17%; (u) PhSO2Cl, Et3N, CH2Cl2, 100%; (v) NBS, AIBN, CCl4, D, 47%; (w) piperidine, BuLi, THF, 38%; (x) NaOH, MeOH–H2O, D, 64%; (y)
ClCH2CN, Et3N, acetone–H2O, D, 30%; (z) piperidine, D, 100%.
O
tral piperidine ring seem to trend in the right direction.
F
F
a, b
c, d
We observed, for example, measurable plasma levels in
the rat with urea 33 (AUC0–6h = 1030 nM h (po,
10 mpk)) and even higher levels with amide 27
(AUC0–6h = 1830 nM h (po, 10 mpk)), however in the
latter case H3 activity was significantly diminished.
13
N
N
H
Bn
Scheme 4. Reagents: (a) DAST, CH2Cl2, 79%; (b) H2, Pd(OH)2,
EtOAc, 94%; (c) 37, NaBH(OAc)3, EtOH, dichloroethane, 81%; (d)
maleic acid, MeOH–H2O, 92%.
In conclusion, introduction of basic groups into the
right side of 4-[(1H-imidazol-4-yl)methyl]piperidine ser-
ies of potent H3 antagonists generally suppressed
CYP450 inhibition, usually observed with imidazole
compounds, while preserving H3 activity. The central
piperidine ring of the series appears to be the main
source of the poor pharmacokinetic profile in rats. Its
derivatives with improved pharmacokinetic profiles
have been identified. Any further developments in this
area will be reported in due course.
dosed orally in rats, and low levels when dosed orally in
monkeys (AUC0–24h = 3200 nM h (po, 3 mpk)). A study
with H-labeled 4 identified no major metabolites and
3
indicated an extremely low (<2%) level of excreted
radioactivity over the 24 h period, both in the case of
oral and intravenous administration. This experiment
was then followed up with a whole body autoradiogra-
phy study in the rat. Upon oral administration poor
absorption of compound from the GI tract was ob-
served, possibly due to the highly polar nature of 4.
Intravenous administration, however, did not improve
plasma levels and led to extensive tissue distribution
without any particular area of concentration. No com-
pound clearance over the 24 h period was observed.
Poor plasma levels in rats appear to be a general feature
of this series of compounds. For example, it was also
observed with 16, 19, 23, 24, 30, and 31. While modifi-
cations of the right side of the molecule do not seem to
have any effect on plasma levels, derivatives of the cen-
Acknowledgment
The authors thank the Radiochemistry group for the
synthesis of radioactively labeled materials.
References and notes
1. (a) Murray, M. Drug Metab. Rev. 1987, 18, 55; (b) Yang,
R.; Hey, J.; Aslanian, R.; Rizzo, C. Pharmacology 2002, 66,

994
M. Berlin et al. / Bioorg. Med. Chem. Lett. 16 (2006) 989–994
128; (c) Halpert, J. R.; Guengerich, F. P.; Bend, J. R.;
Correia, M. A. Toxicol. Appl. Pharmacol. 1994, 125, 163.
2. (a) Shimada, T.; Yamazaki, H.; Mimura, M.; Inui, Y.;
Guengerich, F. J. Pharmacol. Exp. Ther. 1994, 270, 414; (b)
Guengerich, F. J. Pharmacokinet. Biopharm. 1996, 24, 521;
(c) Lin, J.; Lu, A. Pharmacol. Rev. 1997, 49, 403.
3. During the preparation of this manuscript, a non-imdazole
series of H3 antagonists structurally analogous to com-
pound 1 was described Apodaca, R.; Dvorak, C.; Xiao, W.;
Barbier, A.; Boggs, J.; Wilson, S.; Lovenberg, T.; Carru-
thers, N. J. Med. Chem. 2003, 46, 3938.
5. McLeod, R.; Rizzo, C.; West, R.; Aslanian, R.; McCor-
mick, K.; Bryant, M.; Hsieh, Y.; Korfmacher, W.; Mingo,
G.; Varty, L.; Williams, S.; Shih, N.-Y.; Egan, R.; Hey, J.
J. Pharm. Exp. Therap. 2003, 305, 1037.
6. Favreau, L.; Palamanda, J.; Lin, C.; Nomeir, A. Drug
Metab. Disp. 1999, 27, 436, and references therein.
7. Palamanda, J.; Favreau, L.; Lin, C.; Nomeir, A. Drug Disc.
Today 1998, 10, 466.
8. Interestingly, CYP2D6 profile of meta-substituted ana-
logs, while inferior to that of 4, appears to be much
less sensitive to the presence of basic nitrogen or lack
thereof (see 15 vs 5 compared to 14 vs 4).
9. See preceding paper.
4. West, R.; Zweig, A.; Shih, N.-Y.; Siegel, M.; Egan, R.;
Clark, M. Mol. Pharmacol. 1990, 38, 610.