Novel serotonin type 3 receptor partial agonists for the potential treatment of irritable bowel syndrome

Article abstract of DOI:10.1016/j.bmcl.2010.11.080

Serotonin type 3 (5-HT₃) receptor partial agonists are being targeted as potential new drugs for the treatment of irritable bowel syndrome (IBS). Two new chemical series bearing indazole and indole cores have exhibited nanomolar binding affinity for the h5-HT₃A receptor. A range of partial agonist activities in HEK cells heterologously expressing the h5-HT ₃A receptor were measured for the indazole series. Excellent 5-HT₃ receptor selectivity, favorable in vitro metabolic stability and CYP inhibition properties, and good oral in vivo potency in the murine von Bezold-Jarisch reflex model is exemplified thereby indicating the series to have potential utility as improved IBS agents.

Full text of DOI:10.1016/j.bmcl.2010.11.080

Bioorganic & Medicinal Chemistry Letters 21 (2011) 58–61
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel serotonin type 3 receptor partial agonists for the potential treatment
of irritable bowel syndrome
David D. Manning a, , Christopher L. Cioffi a, Alexander Usyatinsky a, Kevin Fitzpatrick a, Liaqat Masih a,
⇑
Cheng Guo a, Zhenjun Zhang a, Sok Hui Choo a, M. Inthikhab Sikkander a, Kristen N. Ryan a,
Jennifer Naginskaya a, Carla Hassler a, Svetlana Dobritsa a, Jonathan D. Wierschke a, William G. Earley a,
Amy S. Butler b, Catherine A. Brady b, Nicholas M. Barnes b, Marlene L. Cohen c, Peter R. Guzzo a
a Discovery R&D AMRI, 26 Corporate Circle, PO Box 15098, Albany, NY 12212-5098, United States
b Celentyx Ltd, Birmingham Research Park, Vincent Drive, Edgbaston, Birmingham B15 2SQ, UK
c Creative Pharmacology Solutions LLC, 10532 Coppergate, Carmel, IN 46032, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Serotonin type 3 (5-HT3) receptor partial agonists are being targeted as potential new drugs for the treat-
ment of irritable bowel syndrome (IBS). Two new chemical series bearing indazole and indole cores have
exhibited nanomolar binding affinity for the h5-HT3A receptor. A range of partial agonist activities in HEK
cells heterologously expressing the h5-HT3A receptor were measured for the indazole series. Excellent
5-HT3 receptor selectivity, favorable in vitro metabolic stability and CYP inhibition properties, and good
oral in vivo potency in the murine von Bezold–Jarisch reflex model is exemplified thereby indicating the
series to have potential utility as improved IBS agents.
Received 24 September 2010
Revised 12 November 2010
Accepted 16 November 2010
Available online 21 November 2010
Keywords:
5-HT3 receptor partial agonist
Irritable bowel syndrome
IBS
Ó 2010 Published by Elsevier Ltd.
Ligand-gated ion channel
Ramosetron
Alosetron
Indole
Indazole
Irritable bowel syndrome (IBS) affects as many as 10% of the US
adult population, with symptoms ranging from constipation to
diarrhea or a combination of the two, coupled with severe abdom-
inal pain and discomfort.1 Direct medical care costs in the US have
been estimated to be as high as $8 billion per year though only 6%
of the costs are attributed to medication, implying that there is sig-
nificant unmet medical need in IBS therapy.2 While the underlying
cause of IBS is currently unknown, the typical symptoms of pain
and altered bowel habits suggest a potential dysfunction of neural
pathways involved in sensory and/or motor function processing.
Serotonin, a key neurotransmitter synthesized and stored in the
GI tract, plays a central role in the normal function of the gut and
has been reported in abnormal levels in IBS patients.3 Acting at
multiple receptor subtypes, serotonin activates both enteric and
central neurons and gastrointestinal tract smooth muscle, which
can ultimately lead to gut movement and/or the perception of pain
and discomfort.4 One type of serotonin receptor, the 5-HT3 recep-
tor, is located on vagal afferents within the intestinal wall and
can be activated by serotonin release induced, for example, by
stress or in response to intracolonic pressure.5
Among its pharmacological actions, the 5-HT3 receptor has
shown that it can affect intestinal transit, small bowel secretion
and the perception of visceral pain.6 In fact, antagonism of the
5-HT3 receptor represents one of the few clinically validated and
effective strategies for the symptomatic treatment of diarrhea pre-
dominant IBS (IBS-D). Unfortunately, broad utilization of 5-HT3
receptor antagonists in IBS therapy has been hampered due to rare
occurrences of severe constipation and ischemic colitis associated
with alosetron, the earliest pharmaceutical product introduced in
this class.7 Ramosetron hydrochloride, a 5-HT3 receptor antagonist
launched in 2008 in Japan, has no reports of the same serious ad-
verse events in IBS patients thereby demonstrating that safer
5-HT3 receptor modulators can be achieved.8
Partial activation of ligand gated ion channels is an established
drug discovery strategy and principally employed to improve side
effect profiles of first generation ligands.9 A high affinity 5-HT3
receptor partial agonist is predicted to attenuate 5-HT3 receptor
function in the presence of excessive endogenous serotonin, yet
maintain a basal level of receptor activity. The preservation of a
⇑
Corresponding author. Tel.: +1 518 512 2129; fax: +1 518 512 2081.
E-mail address: David.Manning@amriglobal.com (D.D. Manning).
0960-894X/$ - see front matter Ó 2010 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2010.11.080

D. D. Manning et al. / Bioorg. Med. Chem. Lett. 21 (2011) 58–61
59
N
CO2Me
MeO2C
R1
NH
MeO2C
CHO
CHO
N
N
R2
a or b
N
c
N
N
R1
N
R1
Figure 1. Preferred functional activity of 5-HT3 receptor ligands for IBS.
H
R2
3a,b
1
2
modicum of 5-HT3 receptor function is predicted to reduce the risk
of constipation and possible other side effects associated with full
receptor inhibition in IBS-D patients. Further, by tuning the intrin-
sic activity of a partial agonist through chemical modification, it
may be possible to identify compounds that will treat a range of
IBS symptom classes (Fig. 1).10
Herein we report the discovery of two new series of heterocy-
clic compounds derived from indole and indazoles, which have
shown early promise toward achieving our goal to identify a
5-HT3 receptor partial agonist drug candidate suitable for the treat-
ment of IBS (Fig. 2).
The general synthesis of indazole analogs, 5–12, is shown in
Scheme 1. The commercial 3-formyl indazoles 1 were either alkyl-
ated or arylated and then subjected to reductive amination condi-
tions using either (R)- or (S)-3-aminoquinuclidine to provide
aminomethyl indazoles 3a,b. The methyl ester was hydrolyzed
and then cyclized using HBTU to provide the desired indazoles
5–12.
N
N
O
N
CO2Li
d
NH
e
N
N
R2
N
N
R1
R1
R2
4a,b
5-12
Scheme 1. General synthesis of indazole series. Reagents and conditions:
(a) Cs2CO3, R2X, DMSO; (b) R2 = Ar; ArB(OH)2, Cu(OAc)2, Et3N, CH2Cl2; (c) 3-
aminoquinuclidine dihydrochloride, NaBH(OAc)3, 1% HOAc in CH2Cl2; (d) LiOHÁH2O,
1:1 THF/H2O; (e) HBTU, DMF.
N
N
CO2Me
CO2Me
CO2Li
NH
NH
a
b
The synthesis of a related indole analog is shown in Scheme 2.
Mannich coupling to methyl 1-methyl-1H-indole-4-carboxylate
(13) using formaldehyde and either (R)- or (S)-3-aminoquinucli-
dine provided aminomethyl indole 14a,b. As in the synthesis of
the indazoles, the methyl ester was hydrolyzed using lithium
hydroxide, followed by cyclization to the desired indoles 16a,b.
Novel indole and indazole-derived heterocycles have been iden-
tified and found to exhibit potent in vitro h5-HT3A receptor bind-
ing affinity (Table 1).11 On the indazole core, substituents at the
R2 position are well tolerated, with only a slight decrease in affinity
observed when substituted phenyl substituents were incorporated
at R2 (12a,b). Fluorine was tolerated at R1. Several pairs of enanti-
omers were examined and, in general, the binding affinities were
equivalent between the two.
A number of analogs were selected for assessment of functional
activity in HEK293 cells expressing the h5-HT3A receptor subunit.
In order to confirm the agonist activity of compounds with small
signals, compounds were tested for their ability to evoke 5-HT3
receptor mediated responses with and without addition of the po-
sitive allosteric modulator, 5-chloroindole, which has been used to
magnify partial agonist responses of the 5-HT3 receptor, but has no
effect on 5-HT3 receptor antagonists.12 The results for select analogs
are shown in Table 2. Two pairs of enantiomers (5a,b and 6a,b) sug-
gest that stereochemistry plays a role in dictating the extent of par-
tial agonist activity of the scaffold. While the (R)-enantiomer 5a
exhibited a strong partial agonist response both with and without
5-chloroindole, (S)-enantiomer (5b) only demonstrated a partial
agonist response when potentiated with 5-chloroindole.
N
N
N
CH3
CH3
CH3
13
14a,b
15a,b
N
O
N
c
N
CH3
16a,b
Scheme 2. Synthesis of indoles 16a,b. Reagents and conditions: (a) 3-aminoqui-
nuclidine dihydrochloride, CH2O, HOAc, rt; (b) LiOHÁH2O, 1:1 THF/H2O, reflux; (c)
HBTU, DMF, 50 °C.
A different outcome was observed for the enantiomeric pair
6a,b. In this case, the (R)-enantiomer 6a proved to be a slightly
weaker partial agonist than its non-alkylated parent 5a, while
the (S)-enantiomer 6b proved to be strictly an antagonist both in
the absence and presence of 5-chloroindole. Analogs from the ser-
ies demonstrated a range of partial agonist activities, falling both
below and above the value measured for DDP733, a reference 5-
HT3 receptor partial agonist.13 In contrast, known 5-HT3 receptor
antagonists showed no effect in this assay, both with and without
the potentiator 5-chloroindole (Table 2).
A control experiment was carried out using the 5-HT3 receptor
antagonist granisetron (Fig. 3) to confirm that the responses were
mediated by the 5-HT3 receptor. As shown, the serotonin-induced
agonist response in the HEK293 cells is sensitive to the addition of
granisetron (5-HT panel). Likewise, addition of granisetron com-
pletely abolished the partial agonist responses attributed to inda-
zole 6a both in the presence and absence of 5-chloroindole.
Compounds 5a and 6a which exhibited Emax values of 35% and
6%, respectively, were selected for further investigation. Subse-
quent in vitro ADME profiling (Table 3) with 5a and 6a demon-
strated a low risk for inhibition of cytochrome P450 enzymes and
a stability in human liver microsomes comparable to alosetron.
In addition, no significant off-target activity (>50% inhibition at
N
O
N
X
N
R2
R1
X = N or CH
Figure 2. Indazole and indole derived chemical series.

60
D. D. Manning et al. / Bioorg. Med. Chem. Lett. 21 (2011) 58–61
Table 1
In vitro 5-HT3 receptor binding data
N
N
O
N
O
N
N
N
CH3
N
R1
R2
R2
5-12
16
Compound
R1
Stereochemistry
h5-HT3A Kia (nM)
5a
5b
6a
6b
7
8a
8b
9a
9b
10
H
H
H
H
F
H
H
H
H
H
H
H
H
H
H
H
H
CH3
CH3
CH3
(R)
(S)
(R)
(S)
(R)
(R)
(S)
(R)
(S)
(S)
(R)
(R)
(S)
(R)
(S)
4.4 0.5
4.9 1.5
1.7 0.6
0.7 0.1
2.0 0.2
2.0 0.7
1.6 0.9
3.6 0.8
2.3 1.0
2.4 0.6
7.1 1.4
23.4 4.8
7.9 1.9
0.9 0.6
0.7 0.1
0.5 0.1
0.06 0.01
Figure 3. Ability of the selective 5-HT3 receptor antagonist granisetron to block
agonist effects of 5-HT or indazole 6a in the absence and presence of 5-chloroindole
induced increase in [Ca2+]i in HEK293 cells expressing h5-HT3A receptor.
Ethyl
Ethyl
Isopropyl
Isopropyl
Isobutyl
Benzyl
4-Fluorophenyl
4-Fluorophenyl
CH3
CH3
Alosetron
Ramosetron
11
Table 3
Selected drug properties
12a
12b
16a
16b
a
c
Drug
CYP inhibition IC50
M)
Human
CLint
Bezold–Jarisch p.o. ED50
(mpk)
b
(
l
5a
6a
>50
>50
9.3 (3A4)
7.3
3.8
24
3
0.96
0.64
0.25
0.02
DDP733
Alosetron 0.6 (3A4)
a
Mean Ki, n P3.
a
Human CYP inhibition IC50
(
l
M), six CYP isoforms tested: 1A2, 2B6, 2C9, 2C19,
M are specified.
2D6, 3A4; CYP isoforms showing IC50 <10
l
Table 2
b
Compounds were incubated with human liver microsomes; data reported in
L/min/mg.
Comparison of the functional responses of select indazoles with selective 5-HT3
receptor literature reference standards
l
c
Compounds were dosed p.o. to mice 1 h prior to 0.1 mg/kg 5-HT challenge.
Druga
Drug aloneb
Drug + 10 l
M 5-chloroindolec
Approximate ED50 determined by 4-point dose–response, n = 5 mice per dose.
EC50 (nM) % max responsed EC50 (nM) % max responsed
5-HT
DDP733
5a
5b
6a
178 20 100
4
4
3
1
2
1
3
1
1
1
1
141 13
9.2 1.0
120
7
agonists. Given that the series exhibited a range of functional activ-
ities, it may be possible to target a number of different IBS symp-
tom classes. 5-HT3 receptor partial agonists may avoid the
unwanted side effects associated with the first 5-HT3 receptor
antagonist alosetron. Compounds 5a and 6a exhibited favorable
in vitro functional activity, in vitro metabolic stability and CYP
inhibition profiles, target receptor selectivity,14 and in vivo activity
in the von Bezold–Jarisch model, thereby well positioning this ser-
ies of compounds toward the achievement of our goal. Further
chemistry efforts and biological studies will be reported in due
course.
5.4 1.4
14.5 5.8
22
35
0
71 11
97 1*
84 3*
12.6 6.1
6
6.9 1.5
23
2
6b
À2
À2 2**
Alosetron
Ramosetron
Palonosetron
Ondansetron
Granisetron
2
À1
À1
À2
0
2
1
2
1
0
À1
À2
À1
À2
0
a
5-HT3 reference antagonists tested @ 10
l
M.
Functional response in HEK293 cells expressing h5-HT3A (n = 4–6), compounds
M, respectively.
Functional response in HEK293 cells expressing h5-HT3A (n = 4–6).
b
5b and 6b tested at 1 and 10
l
c
Acknowledgments
d
*
% max response is normalized to 5-HT at 3
Compounds 5a and 5b tested at 1 M with 100 lM 5-chloroindole.
Compound 6b tested at 10 lM.
lM.
l
**
The authors gratefully acknowledge Yuri Khmelnitsky, Ph.D.
and the AMRI Metabolism and Biotransformation Department.
1 lM) was observed in a 65-membered selectivity panel or in an
11-membered serotonin receptor subtype panel.14 In addition, sol-
References and notes
ubility of compound 6a in phosphate buffered saline (pH 7.4)
proved to be quite high (>500 lM).
1. Brandt, L. J.; Chey, W. D.; Foxx-Orenstein, A. E.; Schiller, L. R.; Schoenfeld, P. S.;
Spiegel, B. M.; Talley, N. J.; Quigley, E. M. M.; Moayyedi, P. Am. J. Gastroenterol.
2009, 104, S1.
2. Ashburn, T. T.; Gupta, M. S. Nat. Rev. Drug Disc. 2006, 5, 99.
3. Camilleri, M. Curr. Opin. Endocrinol. Diabetes Obes. 2009, 16, 53.
4. Baker, D. E. Am. J. Health Syst. Pharm. 2005, 62, 700.
5. Spiller, R.; Aziz, Q.; Creed, F.; Emmanuel, A.; Houghton, L.; Hungin, P.; Jones, R.;
Kumar, D.; Rubin, G.; Trudgill, N.; Whorwell, P. Gut 2007, 56, 1770.
6. Gershon, M. D.; Tack, J. Gastroenterology 2007, 132, 397.
7. (a) Clayton, N. M.; Sargent, R.; Butler, A.; Gale, J.; Maxwell, M. P.; Hunt, A. A. E.;
Barrett, V. J.; Cambridge, D.; Bountra, C.; Humphrey, P. P. A. Neurogastroenterol.
Motil. 1999, 11, 207; (b) Mayer, E. A.; Bradesi, S. Expert Opin. Pharmacother.
2003, 4, 2089.
The von Bezold–Jarisch reflex bradycardia model15 in mice has
been used to characterize all commercial 5-HT3 receptor inhibitors
irrespective of indication and was used to investigate the in vivo
oral activity of 5a and 6a, as compared to DDP733 and alosetron.
Transient bradycardia induced by iv administration of 5-HT can
be blocked by oral pre-treatment with both 5-HT3 receptor antag-
onists and partial agonists. As shown in Table 3, compounds 5a and
6a demonstrated potency similar to DDP733 in this model.
In an effort to develop a new therapeutic treatment for IBS, we
have discovered new indole and indazole-derived heterocycles and
evaluated the compounds for activity as 5-HT3 receptor partial
8. Matsueda, K.; Harasawa, S.; Hongo, M.; Hiwatashi, N.; Sasaki, D. Scand. J.
Gastroenterol. 2008, 43, 1202.
9. Hogg, R. C.; Bertrand, D. Biochem. Pharmacol. 2007, 73, 459.

D. D. Manning et al. / Bioorg. Med. Chem. Lett. 21 (2011) 58–61
61
10. IBS patients are categorized into
a
number of subgroups based on the
13. Coleman, N. S.; Marciani, L.; Blackshaw, E.; Wright, J.; Parker, M.; Yano, T.;
Yamazaki, S.; Chan, P. Q.; Wilde, K.; Gowland, P. A.; Perkins, A. C.; Spiller, R. C.
Aliment. Pharmacol. Ther. 2003, 18, 1039.
predominant symptom: diarrhea (IBS-D), constipation (IBS-C), mixed bowel
patterns (IBS-M).
11. U.S. Patent Application No. 12/473,940, Publication No. 20090298809
(published Dec 3, 2009) (David D. Manning and Christopher L. Cioffi,
applicants).
12. Butler, A. S.; Barnes, N. M. 5-Chloroindole Potentiates Human 5-HT3A Receptor
Function Assessed by [Ca2+]i, Presented at Serotonin Club Meeting, Oxford, UK,
July 17–20, 2008; Poster SCP052.
14. Selectivity data was obtained at MDS Pharma and Cerep for 6a only. The panel
included 5-HT1a and 5-HT4 receptor subtypes.
15. Animal data was obtained at MDS Pharma according to: Saxena, P. R.; Lawang,
A. Arch. Int. Pharmacodyn. Ther. 1985, 277, 235.