Synthesis of cycloalkane-annelated 3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines and their evaluation as 5-HT6 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.02.046

Synthesis and biological evaluation of 1 ('angular') and 2 ('linear') s{cyrillic}ycloalkane-annelated 3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines as novel ligands of the 5-HT₆ receptors are disclosed. The new compounds 1 and 2 are highly selective antagonists of the receptor with sub-nanomolar affinities (K_i <1 nM). In its structure, this new chemotype lacks a basic ionizable side chain, which is considered as the characteristic feature of the 5-HT₆ receptor antagonists pharmacophore model.

Full text of DOI:10.1016/j.bmcl.2010.02.046

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2133–2136
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of cycloalkane-annelated 3-phenylsulfonyl-pyrazolo[1,5-a]pyrimi-
dines and their evaluation as 5-HT6 receptor antagonists
Alexandre V. Ivachtchenko a,b, Dmitri E. Dmitriev a, Elena S. Golovina c, Elena S. Dubrovskaya b,
Madina G. Kadieva a, Angela G. Koryakova d, Volodymyr M. Kysil b, Oleg D. Mitkin a,
a
Sergey E. Tkachenko b, Ilya M. Okun b, , Anton A. Vorobiov
*
a Department of Organic Chemistry, Chemical Diversity Research Institute, 114401 Khimki, Moscow Region, Russia
b ChemDiv, Inc., 6605 Nancy Ridge Drive, San Diego, CA 92121, USA
c Department of Molecular Pharmacology, Chemical Diversity Research Institute, 114401 Khimki, Moscow Region, Russia
d Bioanalytical Department, Chemical Diversity Research Institute, 114401 Khimki, Moscow Region, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Synthesis and biological evaluation of 1 (‘angular’) and 2 (‘linear’) cycloalkane-annelated 3-phenylsulfo-
Received 12 January 2010
Revised 9 February 2010
Accepted 10 February 2010
Available online 14 February 2010
nyl-pyrazolo[1,5-a]pyrimidines as novel ligands of the 5-HT6 receptors are disclosed. The new com-
pounds
1 and 2 are highly selective antagonists of the receptor with sub-nanomolar affinities
(Ki <1 nM). In its structure, this new chemotype lacks a basic ionizable side chain, which is considered
as the characteristic feature of the 5-HT6 receptor antagonists pharmacophore model.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Phenylsulfonyl-pyrazolo[1,5-a]pyrimidines
5-HT6 receptor
5-HT6 antagonist
5-HT2B receptor
5-HT2B antagonist
hERG channel
Since the discovery of serotonin 5-HT6 receptors (5-HT6R), mul-
tiple selective antagonists were synthesized. Majority of these are
based on heterocyclic structures containing a sulfonyl substitu-
ent.1–3 3-Phenylsulfonyl-8-piperazin-1-yl-quinoline (SB-742457,
GlaxoSmithKline, Fig. 1)4 is in an advanced stage of clinical devel-
opment—currently in phase II of clinical trials for the treatment of
Alzheimer’s disease.5
Another series of 5-HT6R antagonists,6 substituted 8-amino-2-
methylsulfanyl-3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines, has
been developed by Roche. The leader in this series of compounds is
2-methylsulfanyl-3-phenylsulfonyl-8-piperazin-1-yl-6,7-dihydro-
5H-cyclopenta[d]pyrazolo[1,5-a]pyrimidine (Ro-65-7674, Roche,
Fig. 1) with good DMPK properties in preclinical studies. However,
this molecule was shown to exhibit quite a high potency hERG
Figure 1. Structure of sulfonyl-substituted drug candidates currently in clinical
trials.
channel inhibition activity with IC20 of 0.2 l
M.7
It should be noted that the role of 5-HT6R in the etiology of dif-
ferent diseases is still elusive, mainly due to a lack of safe, highly
selective and potent ligands with a good blood–brain barrier per-
meability. Therefore, the quest for discovery of new potent and
selective antagonists of 5-HT6R as potential drugs for the treat-
ment of various diseases of the central nervous system, continues
unabated.1,8,9
The pharmacophore model (Fig. 2A) of 5-HT6R ligands has been
developed based on 3D analysis of 45 structurally diverse antago-
nists.10 The model includes a central aromatic or heterocyclic
scaffold, AR, separating a positive ionizable group, PI (usually sec-
ondary or tertiary amino-group), from a strong multiple hydrogen
* Corresponding author. Tel.: +1 858 794 4860; fax: +1 858 794 4931.
E-mail address: iokun@chemdiv.com (I.M. Okun).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.02.046

2134
A. V. Ivachtchenko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2133–2136
izable group, PI, is considered a common pharmacophoric element
of most of the ligands that interact with biogenic amine GPCRs.
However, we have found no publications, which would convinc-
ingly indicate the necessity of the basic side chain for effective
interaction with the receptor. Therefore, the purpose of this com-
munication is to establish if the basic side chain is essential for
the high affinity binding of the ligands to the 5-HT6R. For this,
we have synthesized several analogs of the Ro-65-7674. The syn-
thesized compounds represent a family of previously unknown
Figure 2. (A) Simplified schematic representation of a 5-HT6R pharmacophore
model,10 which does not reflect the interatomic distances, vectors, and 3D atom
locations. (B) Representation of a new pharmacophore model. AR—aromatic or
heterocyclic scaffold; PI—positive ionizable group; mHBA—multiple hydrogen bond
acceptor group; HYD—hydrophobic site.
cycloalkane-annelated
3-phenylsulfonyl-pyrazolo[1,5-a]pyrimi-
dines (PSPP), 1 (‘angular’) and 2 (‘linear’) analogs (Table 1) and
examined their affinity to the 5-HT6R. The characteristic feature
of these molecules is the absence of the basic side chain, PI, in their
structures.
Table 1
Structure of ‘angular’, 1, and ‘linear’, 2, sulfonyl-substituted cycloalakane-annelated
3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines
The PSPPs 1, 2 were synthesized by the interaction of the 3-
Ph
Ph
aminopyrazoles
3 with the corresponding b-dicarbonyl com-
O
O
S
S
pounds 4 (Schemes 1 and 2).
O
O
R2
N
N
We found that the interaction of 3-amino-5-methylsulfanyl-4-
phenylsulfonyl-2H-pyrazole 3(1) with 2-formylcyclopentanone
4(1) in acetic acid with or without HCl at either room temperature
or 100 °C led to synthesis of the 2-methylsulfanyl-3-phenylsulfo-
nyl-7,8-dihydro-6H-cyclopenta[e]pyrazolo[1,5-a]pyrimidine 1(1)
only (yield 42%). The reaction of 3(1) with 2-formylcyclohexanone
4(2) in acetic acid, at 100 °C, led to synthesis of both the 2-meth-
ylsulfanyl-3-phenylsulfonyl-6,7,8,9-tetrahydro-pyrazolo[1,5-a]quin-
azoline, 1(2), and 2-methylsulfanyl-3-phenylsulfonyl-5,6,7,8-tetra-
hydro-pyrazolo[5,1-b]quinazoline, 2(2), in approximately equal
proportion. At the same time, this reaction performed in a mixture
of acetic acid and HCl at room temperature, produced PSPP 1(2) with
a yield of 59%. When performed at room temperature in acetic acid
only, the reaction produced PSPP 2(2) with a yield of 53%.
The reaction of 3-amino-2H-pyrazoles 3(1–3) with b-diketones
5(1,2), regardless of the conditions presented in Scheme 1 as well
as in alcohol in presence of triethylamine, led to synthesis of a mix-
ture of angular 1(3–7) and linear 2(3–7) products (Scheme 2).
In reactions with 2-acetylcyclopentanone 5(1), the main prod-
ucts were represented by angular compounds 1(3,5) (>90%). Pure
5-methyl-3-phenylsulfonyl-7,8-dihydro-6H-cyclopenta[e]pyrazol-
o[1,5-a]pyrimidines 1(3,5) were isolated from the mixtures by re-
peated re-crystallization.
R1
R1
N
N
N
N
n
R2
2
1
n
Compd
n
R1
R2
1(1)
1(2)
1(3)
1(4)
1(5)
1(6)
1(7)
2(2)
1(3)*
2(4)
1(5)*
2(6)
2(7)
1
2
1
2
1
2
2
2
1
2
1
2
2
SMe
SMe
Me
Me
SMe
SMe
H
H
Me
Me
Me
Me
Me
H
Me
Me
Me
Me
Me
OCH2CH2OH
SMe
Me
Me
SMe
SMe
OCH2CH2OH
*
The compound was not separated from the mixture.
In reactions with 2-acetylcyclohexanone 5(2), a barely separa-
ble mixture comprising 50–70% of the linear products was formed.
Using HPLC, we were able to separate individual angular 1(6) and
linear 2(6) compounds.
The structures of the synthesized compounds 1, 2 were con-
firmed by LC–MS and NMR data. Structural assignments, angular
versus linear, of the synthesized compounds 1, 2 were made based
on 2D NMR experiments (NOESY and HMBC) using published
data11–13 for the NMR spectra and quantum-mechanical calcula-
tions of such heterocyclic compounds.
Scheme 1. Synthesis of cycloalkane-annelated 2-methylsulfanyl-PSPP 1(1,2) and
2(2).
We investigated the effects of the substitutions in the R1 and R2
positions as well as those of the cycloalkane size on the affinity of
the compounds 1 and 2 to the 5-HT6R. The compound affinities
bond acceptor group, mHBA, (usually sulfonyl or sulfamide group),
and a hydrophobic site, HYD (for example, phenyl). The basic ion-
Scheme 2. Synthesis of cycloalkane-annelated 5-methyl-PSPP 1(3–7) and 7-methyl-PSPP 2(3–7).

A. V. Ivachtchenko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2133–2136
2135
Table 2
compounds of a mixture 1(4) + 2(4) also led to almost 10-fold reduc-
tion in the affinity as compared with the respective 1(6) + 2(6)
product.
Comparison of binding affinities (Ki) and antagonistic potencies (IC50) of angular, 1,
and linear, 2, compounds
b
Compd #
Ki,a nM (binding)
IC50
5-HT6
,
nM (functional)
5-HT2B
Similar SAR trends were observed in a cell-based functional as-
say14 where the potency, IC50, of the compounds of series 1 and 2
to block serotonin-induced increase of cAMP level in HEK293 cells
expressing 5-HT6 receptor was measured. For example, the linear
ligand 2(2) was order of magnitude less potent than its angular
analog 1(2) (IC50 = 98.5 nM and 7.5 nM, respectively, Table 2).
The ring expansion from cyclopentyl to cyclohexyl in the series
of angular compounds (1(1) versus 1(2)) affected neither binding
affinity nor functional potency of the compounds.
The product 1(6) + 2(6) was tested in radioligand competition
binding assays (as described in the Supplementary data) on a panel
of therapeutic targets comprising enzyme, GPCR, ion channel, and
transporter families (Fig. 3). This allowed us to simultaneously as-
sess specificity of both the linear and angular forms. The 1(6) + 2(6)
combination was highly specific and selective towards 5-HT6R. At
1(1)
1(2)
2(2)
1(4) + 2(4)
1(6)
0.458
0.549
7.97
6.0
7.5
98.5
26.8
16.8
7.8
118.0
411.0
4.42
0.440
0.680
0.479
161.0
127.0
2(6)
1(6) + 2(6)
16.9
a
Values are averages from single experiment performed in duplicates.
Values are geometric means of three independent experiments. There was less
b
than twofold deviation of IC50 values from the geometric means of the experiments.
(Table 2) were measured in a competitive radioligand binding assay
as described in the Supplementary data. Affinities of the angular
molecules were not affected by either ring expansion from cyclo-
pentyl (1(1)) to cyclohexyl (1(2)) or by substitution of proton
(1(2)) for CH3 (1(6)) in R2 position. Contrary, linear molecules
seemed to be quite sensitive to the substitution in R2 position. Inclu-
sion of a methyl group in 2(6) led to 10-fold increase in the affinity
compared to 2(2) (Table 2). The affinity of a product mixture
1(6) + 2(6), obtained from the reaction with 2-acetylcyclohexanone
5(2), was similar (Ki = 0.48 nM) to those of the individual compo-
nents. Substitution of SCH3 group in R1 position with CH3 group in
1 lM (approx. 0.5 lM each), the product 1(6) + 2(6) besides 5-
HT6R (100 7%) strongly interacted, with only two other targets,
5-HT2BR (85 3%) and peripheral benzodiazepine receptor (PBR)
(66 7%).
Ligands, 1(1,2,6), and 2(6), were also examined for their 5-HT2B
R
functional activity15 (Table 2). Stimulation of this receptor is
thought to be potentially involved in cardiac valvulopathy16 and
represents potential liability. Effect of the compounds was assessed
in
a cell-based functional assay in both the agonistic and
Figure 3. Target-specific profile determined for
a 1(6) + 2(6) product consisting of 1:1 mixture of 5-methyl-5-methylsulfanyl-3-phenylsulfonyl-6,7,8,9-tetrahydro-
pyrazolo[1.5-a]quinazoline , 1(6), and 5-methyl-5-methylsulfanyl-3-phenylsulfonyl-9-methyl-5,6,7,8-tetrahydro-pyrazolo[5.1-b]quinazoline, 2(6). Displacement of corre-
sponding radio-labeled ligands was measured at the compound concentration of 1
displacement values SD are presented.
lM. Shown is a typical experiment performed in duplicates, where the radio-labeled ligand

2136
A. V. Ivachtchenko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2133–2136
References and notes
1. Holenz, J.; Pauwels, P. J.; Diaz, J. L.; Merce, R.; Codony, X.; Buschmann, H. Drug
Discovery Today 2006, 11, 283.
2. Kim, H.-J.; Doddareddy, M. R.; Choo, H.; Cho, Y. S.; No, K. T.; Park, W.-K.; Pae, A.
N. J. Chem. Inf. Model. 2008, 48, 197.
3. Geldenhuys, W. J.; Schyf, C. J. V. Curr. Top. Med. Chem. 2008, 8, 1035.
4. Upton, N.; Chuang, T. T.; Hunter, A. J.; Virley, D. J. Neurotherapeutics 2008, 5,
458.
5. http://clinicaltrials.gov/ct2/show/NCT00348192.
6. Boes, M.; Riemer, C.; Stadler. H. Eur. Pat. Appl. EP941994, 1999.
7. Zhao, S.-H.; Berger, J.; Clark, R. D.; Sethofer, S. G.; Krauss, N. E.; Brothers, J. M.;
Martin, R. S.; Misner, D. L.; Schwab, D.; Alexandrova, L. Bioorg. Med. Chem. Lett.
2007, 17, 3504.
8. http://www.biospace.com/news_story.aspx?NewsEntityId=141148.
9. Tkachenko, S.; Ivachtchenko, A.; Khvat, A.; Okun, I.; Lavrovsky, Y.; Salimov R.
nineth Intern. conf. AD/PD 2009, Prague, Czech Republic, 2009, Poster
Presentations. http://www.kenes.com/adpd2009/posters/Abstract490.htm.
10. López-Rodróguez, M. L.; Bellinda Benhamú, B.; De la Fuente, T.; Sanz, A.; Pardo,
L.; Campillo, M. J. Med. Chem. 2005, 48, 4216.
Figure 4. HERG potassium channel blocking activity of ligands 1(6) and 2(6) as
measured in Patch clamp assay. Shown are averages of three measurements SD.
11. Pugmire, R. J.; Robins, M. J.; Grant, D. M.; Robins, R. K. J. Am. Chem. Soc. 1971, 93,
1887.
12. Vartanyan, M. A.; Eliseev, O. L.; Skov, Kh. R.; Karakhanov, R. A. Russ. Chem. Bull.
1997, 46, 1178.
antagonistic mode.17 In short, the compounds were added into a
1 cm cuvette containing HEK293 cells exogenously expressing h5-
HT2B receptor, 5-HT2BR-HEK293 cells. The cells were loaded with a
ratiometric fluorescent dye, Fura-2 AM, and the fluorescent signals
were registered with the spectrofluorometer RF-5301PC (Shimadzu,
Japan) at 510 nm upon alternative excitation at 340 nm and 380 nm.
The ratios were expressed as free intracellular calcium ion concen-
trations,18 [Ca2+]i, with a built in Super Ion Probe Software. For
13. Bajwa, J. S.; Sykes, P. J. J. Chem. Soc., Perkin Trans. 1 1979, 3085.
14. 5-HT6 receptor functional assay. The 5-HT6-HEK cells were grown in Corning
384-well plates (Lowell, MA) at 37 °C in atmosphere of air:CO2 (95%:5%) in
DMEM supplemented with 10% FBS, 1% AAS, blasticidine S, and phleomycin
(Invitrogen, Carlsbad, CA). T-Rex/5-HT6 receptor expression was activated by
addition of tetracycline, as recommended by the manufacturer, a day before
the experiments. On the day of the experiment, the medium in the wells was
substituted with phenol red-, calcium- and magnesium-free HBSS, (Invitrogen,
Carlsbad, CA), supplemented with 1 mM MgCl2, 1 mM CaCl2, 5 mM HEPES, pH
antagonistic activity,
ette 15 sec after the compound addition and the compound effect
is assessed as the inhibition of the
Me-5-HT-induced [Ca2+]i mobi-
lization. The compounds tested, 1(1,2,6) and 2(6), did not cause any
noticeable activation of the 5-HT2B receptor. As the 5-HT2B receptor
blockers, these compounds were 10–55-fold less potent compared
to the 5-HT6 receptor (Table 2).
aMe-5-HT (5 nM final) was added to the cuv-
7.4, and 100 lM IBMX. The test compounds were added at different
concentrations while maintaining constant final DMSO concentration of 0.1%.
After 15 min incubation, serotonin hydrochloride (Sigma, MO) was added to a
final concentration of 10 nM and incubation continued for additional 30 min at
rt. The cells were treated as described in cAMP LANCE assay kit protocol
(Perkin–Elmer, Waltham, MA) as recommended by the manufacturer. The
LANCE signal was measured in white 384-well plates (Corning, MA) using
multimode plate reader VICTOR2V (Perkin–Elmer, Waltham, MA) with built-in
settings for the LANCE detection.
a
15. 5-HT2B receptor functional assay. The 5-HT2B-HEK cells were grown in T-175
flasks at 37 °C in atmosphere of air:CO2 (95%:5%) in DMEM (Sigma, MO)
supplemented with 10% FBS, 1% AAS, blasticidine S, and phleomycin
(Invitrogen, Carlsbad, CA). The T-Rex/5-HT2B receptor expression was
activated by addition of tetracycline, as recommended by the manufacturer,
a day before the experiments. The cells were dissociated with TrypLE™ Express
(Invitrogen, Carlsbad, CA), washed twice with PBS and loaded at room
The ligands 1(6) and 2(6) at 1 lM did not noticeably interact
with another target of potential liability, hERG channel (Fig. 3).
In a functional patch clamp assay (CHO cells with exogenously ex-
pressed hERG channel)16, the compounds showed very low chan-
nel blocking activity (Fig. 4). The angular ligand 1(6) blocked
hERG channel with IC50 = 8.4
the channel permeability only by approx. 25% at 10.0
l
M. Its linear analog, 2(6), reduced
M.
temperature with
4 lM calcium-sensitive dye, Fura-2AM (Invitrogen,
l
Carlsbad, CA) for 30 min. After the loading, the cells were washed once with
PBS, re-suspended into protein free Hybridoma media without phenol red
(Sigma, St. Louis, MO) and allowed to incubate for additional 30 min with
gentle shaking at room temperature. All loading procedures were performed in
dark conditions. The loaded cells were washed twice with PBS and re-
suspended into the Hybridoma media at a cell density of 3–4 Â 106 cells/mL for
subsequent experiments. Fura-2 ratiometric fluorescence signal was registered
at 510 nm upon alternate excitation at 340 nm and 380 nm using
spectrofluorometer RF-5301PC (Shimadzu, Columbia, MD). In a square (1 cm)
In conclusion, to the best of our knowledge, the cycloalkane-
annelated 3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines 1 and 2
are the first in class 5-HT6R antagonists lacking the basic ionizable
side chain. This new chemotype of potent (binding Ki in a pM range
and antagonistic IC50 in low nM range) and highly selective 5-HT6R
antagonists represent a simpler pharmacophore model than the
one suggested earlier.10 The lack of the basic ionizable amino group
shows a promise for a development of 5-HT6R ligands with re-
duced bias towards other biogenic amine receptors.
optical cuvette with a magnetic stirring bar, 100
were diluted into 2.4 mL buffer containing (mM): NaCl (145), KCl (5.4), MgSO4
(0.8), CaCl2 (1.8), HEPES (30), -glucose (11.2). The fluorescence signal was
lL aliquots of the loaded cells
D
allowed to stabilize for 20–30 s before addition of a test compound or vehicle
to assess potential agonistic activity of the compounds, with subsequent
addition of serotonin (2.5
activity.
lL, 10 mM) to assess the compounds’ blocking
Supplementary data
16. Rothman, R. B.; Baumann, M. H.; Savage, J. E.; Rauser, L.; McBride, A.; Hufeisen,
S. J.; Roth, B. L. Circulation 2000, 102, 2836.
17. Okun, I..; Tkachenko, S. E.; Khvat, A.; Mitkin, O.; Kazey, V.; Ivachtchenko, A. V.
Curr. Alzheimer Res., 2010, 7, doi:10.2174/1567210199392222050.
18. Grynkiewicz, G.; Poenie, M.; Tsien, R. Y. J. Biol. Chem. 1985, 260, 3440.
Supplementary data (description of compound syntheses and
biological assay methods) associated with this article can be found,
in the online version, at doi:10.1016/j.bmcl.2010.02.046.