Tetrahydroisoquinoline-Derived Urea and 2,5-Diketopiperazine Derivatives as Selective Antagonists of the Transient Receptor Potential Melastatin 8 (TRPM8) Channel Receptor and Antiprostate Cancer Agents

Article abstract of DOI:10.1021/acs.jmedchem.5b01448

Tetrahydroisoquinoline derivatives containing embedded urea functions were identified as selective TRPM8 channel receptor antagonists. Structure-activity relationships were investigated, with the following conclusions: (a) The urea function and the tetrahydroisoquinoline system are necessary for activity. (b) Bis(1-aryl-6,7dimethoxy-1,2,3,4-tetrahydroisoquinolyl)ureas are more active than compounds containing one tetrahydroisoquinoline ring and than an open phenetylamine ureide. (c) Trans compounds are more active than their cis isomers. (d) Aryl substituents are better than alkyls at the isoquinoline C-1 position. (e) Electron-withdrawing substituents lead to higher activities. The most potent compound is the 4-F derivative, with IC₅₀ in the 10^-8 M range and selectivities around 1000:1 for most other TRP receptors. Selected compounds were found to be active in reducing the growth of LNCaP prostate cancer cells. TRPM8 inhibition reduces proliferation in the tumor cells tested but not in nontumor prostate cells, suggesting that the activity against prostate cancer is linked to TRPM8 inhibition.

Full text of DOI:10.1021/acs.jmedchem.5b01448

Subscriber access provided by UCL Library Services
Article
Tetrahydroisoquinoline-derived urea and 2,5-diketopiperazine
derivatives as selective antagonists of the transient receptor potential
melastatin 8 (TRPM8) channel receptor and anti-prostate cancer agents
Luciano De Petrocellis, Francisco José Arroyo, Pierangelo Orlando, Aniello Schiano
Moriello, Rosa Maria Vitale, Pietro Amodeo, Aránzazu Sánchez, Cesáreo Roncero,
Giulia Bianchini, M. Antonia Martín, Pilar López-Alvarado, and J. Carlos Menéndez
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01448 • Publication Date (Web): 27 May 2016
Downloaded from http://pubs.acs.org on May 29, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Tetrahydroisoquinoline-derived urea and 2,5-
diketopiperazine derivatives as selective
antagonists of the transient receptor potential
melastatin 8 (TRPM8) channel receptor and anti-
prostate cancer agents
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Luciano De Petrocellisa,*, Francisco J. Arroyo,b Pierangelo Orlando,c Aniello Schiano
Moriello,a Rosa Maria Vitale,a Pietro Amodeo,a Aránzazu Sánchez,d Cesáreo Roncero,d Giulia
Bianchini,b M. Antonia Martín,e Pilar López-Alvaradob, J. Carlos Menéndezb,*
aEndocannabinoid Research Group, Institute of Biomolecular Chemistry, National Research
Council, Via Campi Flegrei 34, Comprensorio Olivetti, 80078 Pozzuoli, Naples, Italy.
bDepartamento de Química Orgánica y Farmacéutica, Facultad de Farmacia, Universidad
c
Complutense, 28040 Madrid, Spain. Endocannabinoid Research Group, Institute of Protein
Biochemistry, National Research Council, Via P. Castellino 111, 80131 Naples, Italy.
dDepartamento de Bioquímica y Biología Molecular II, Facultad de Farmacia, Universidad
Complutense, 28040 Madrid, Spain.
eS.D. Química Analítica, Facultad de Farmacia, Universidad Complutense, 28040 Madrid,
Spain.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 92
1
2
3
4
5
6
7
8
9
KEYWORDS Tetrahydroisoquinolines, ureas, diketopiperazines, TRPM8 channels, apoptosis,
anti-prostate cancer activity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABSTRACT. Tetrahydroisoquinoline derivatives containing embedded urea functions were
identified as selective TRPM8 channel receptor antagonists. Structure activity relationships
were investigated, with the following conclusions: (a) The urea function and the
tetrahydroisoquinoline system are necessary for activity. (b) Bis(1-aryl-6,7dimethoxy-1,2,3,4-
tetrahydroisoquinolyl)ureas are more active than compounds containing one
tetrahydroisoquinoline ring and than an open phenetylamine ureide. (c) Trans compounds are
more active than their cis isomers. (d) Aryl substituents are better than alkyls at the
isoquinoline C-1 position. (e) Electron-withdrawing substituents lead to higher activities. The
most potent compound is the 4-F derivative, with IC50 in the 10-8 M range and selectivities
around 1000:1 for most other TRP receptors. Selected compounds were found to be active in
reducing the growth of LNCaP prostate cancer cells. TRPM8 inhibition reduces proliferation
in the tumor cells tested, but not in non-tumor prostate cells, suggesting that the activity
against prostate cancer is linked to TRPM8 inhibition.
ACS Paragon Plus Environment

Page 3 of 92
Journal of Medicinal Chemistry
1
2
3
4
Introduction
5
6
7
8
Transient receptor potential melastatin 8 (TRPM8) is a member of the transient receptor
potential (TRP) superfamily, a class of non-selective cation channels that are activated by a
variety of physical and chemical stimuli.1 Owing to the fact that menthol and other agents that
elicit the sensation of coolness, such as icilin and eucalyptol,2-4 act as TRPM8 agonists, the
TRPM8 ion channel was initially known as the cold menthol receptor-1 (CMR-1).2 A number
of TRP channels are thermo-responsive, become activated at different ranges of temperature
and seem to act as thermo-sensors in vivo.5 Thus, TRPV1 is required for hot temperature
sensing6 and TRPV37 and TRPV48 for warm temperature sensing; TRPA1 is activated by
noxious cold (<10 °C) in vitro9,10 and in vivo11. Similarly, TRPM8 is activated in vitro by
temperatures ranging from innocuous cold (23–15 °C) to noxious cold (< 15 °C)1,2 and is
required as a neuronal sensor of mildly cold temperatures, as shown by knockout mouse
studies.12-17 As with TRPV1, TRPM8 is involved in thermoregulation: menthol and icilin
cause a transient rise in basal temperature18-21 at least for the former TRPM8 mediated;22 on
the other hand, TRPM8 antagonists induce a decrease in body temperature13,23,24 although the
hypothermic response was less than 1 °C unlike the febrile response toTRPV1 antagonists.
The TRPM8 gene encodes for a 1104-residue transmembrane protein, whose quaternary
structure is a homotetramer channel.25 The transmembrane portion is formed by six helices,
out of which the protonated residues at S4 and the S4-S5 linker have an important role in the
receptor activation and S2 and S3 include the binding sites for menthol and icilin, although
they do not fully overlap since menthol recognition is mostly associated with S2, with tyrosine
745 (Y745) being critical for the interaction, while icilin interacts mainly with residues
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 92
1
2
3
4
5
6
located at S326,27 The last two helices constitute the permeable pore module, the sixth one
being responsible for the ion selectivity of TRPM8.28,29
7
8
9
Correlating with the cold sensing function, TRPM8 has been localized in a subpopulation of
primary small diameter sensory neurons (on nociceptive Aδ and C fiber neurons) of the dorsal
root and trigeminal ganglia.1,2,30-33 TRPM8-positive neurons being different from those
expressing TRPV1 and/or TRPA1.34 In addition, TRPM8 is expressed in a range of tissues,
mainly in the prostate and liver and, in lesser amounts, in the brain, lung, bladder,
gastrointestinal tract, blood vessels, and immune cells.35
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Under some pathological conditions, TRPM8 expression is up-regulated on nociceptive
afferent neurons, resulting in cold allodynia36,37 whose mechanism after chronic constriction
injury (CCI) has been studied.38-40. TRPM8 can have analgesic effects by diminishing
neuropathic and visceral pain36,41,42 or by attenuating cold hypersensitivity in inflammatory
and nerve-injury pain models,26,43 cold hyperalgesia and cold allodynia, which are commonly
associated with certain types of neuropathic and inflammatory pain.44-46
The role of TRPM8 is not confined to the perception of cold: in prostatic tumour tissue the
expression level of TRPM8 correlates with malignancy,47 although the activation of TRPM8
by menthol suppresses cellular viability of human melanoma cells.48 Subsequent
experimentation has shown that modulators of TRPM8 may be useful in the treatment of
prostate cancer49,50 as well as in colorectal cancer.51.
TRPM8 activation has been reported to be involved in overactive bladder and painful bladder
syndrome, playing a role in the symptomatology and pathophysiology of human urinary
bladder disorders.52 Preclinically, an antagonist of TRPM8 has been shown to decrease the
frequency of volume-induced bladder contractions, without reducing the amplitude of
ACS Paragon Plus Environment

Page 5 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
contraction in rats,53 suggesting a role for TRPM8 antagonists in chronic pain conditions as
well as in painful bladder syndromes.45
7
8
9
With regard to agonists, at least eighteen natural and synthetic compounds have been found to
activate the channel, including menthol, icilin, eucalyptol1,2 and the potent synthetic
menthane-derived amide 1 (WS-12).54 However, menthol is not selective for TRPM8, since its
responses were reduced, but not abolished, in TRPM8 null mice.24 Studies on genetically
modified mice demonstrated that the effects of icilin were mediated by TRPM8 and not
TRPA1, while menthol-induced cold hypersensitivity was dependent on expression of TRPA1
and not TRPM8.55
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Most of the antagonists reported in the literature56 lack selectivity for TRPM8, since they also
act on TRPV1 and TRPA1.57 Capsazepine, a well-known TRPV1 antagonist, inhibits
TRPM8,3 but it also has non-specific activity on voltage-gated calcium channels and nicotinic
acetylcholine receptors.58-60 Similarly, 2 (BCTC) inhibits TRPM8-mediated Ca2+ influx,3 but
it also functions as a TRPA1 agonist.61 The antimycotics clotrimazole62 and the even more
potent econazole63 have TRPM8 antagonistic activity, but they also strongly activate TRPV1
and TRPA1 actions. Compound 4 (SKF96365), a non-specific blocker of several types of
calcium channels, receptor-operated channels, and inwardly rectifying potassium channels64,65
also inhibits TRPM8 in vitro.66 Anandamide and N-arachidonoyldopamine were previously
found to inhibit menthol- and icilin-induced response in TRPM8-HEK-293 cells,67 and the
same response was obtained from some plant-derived cannabinoids.68,69 Recently, some
compounds endowed with improved TRPM8 selectivity have been described, such as the 2-
pyridyl-benzensulfonamide 5 (RQ-00203078).70 Some additional potent, small-molecule
TRPM8 antagonists have been reported in the literature as potential therapeutics, although
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 92
1
2
3
4
5
6
7
8
9
these compounds have off-target effects. They include menthylamides,71 tryptamine
derivatives, which are also ligands for 5-benzyloxytryptamine receptors,72
benzyloxybenzamides,73 cinnamamides as open-chain analogs of icilin,74 benzimidazoles75,76
benzothiophene phosphonates,77,78 fused piperidines,79,80 oxazole and thiazole derivatives,81
arylglycine-based analogs,82 spiro[chromene-2,4’-piperidines] 6,83 urea derivatives (e.g.
compound 7),80 2-pyridyl-benzensulfonamides70 and indole alkaloids.84
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The structurally related compounds 8 and 9 (Figure 1) have been studied for their
antinocioceptive activity.43,46 A survey of recent patents reveals reports of several structural
classes containing examples of compounds having potent TRPM8 blockade activity by many
companies such as Amgen, Bayer Health, Glenmark Pharmaceuticals, GW Pharma, Janssen
Pharmaceuticals and Raqualia Pharma (for recent reviews, see references 85-87). Most
attention is being devoted to the use of TRPM8 antagonists as analgesics, and their potential
against prostate cancer remains almost unexplored.
ACS Paragon Plus Environment

Page 7 of 92
Journal of Medicinal Chemistry
1
2
3
(a) Agonists
4
5
CH3
CH3
O
H3C
CH3
CH3
6
7
8
9
H
N
N
OH
HO
NO2
O
N
O
Eucalyptol
H
H3C
CH3
H3C
CH3
OCH3
Icillin
Menthol
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1 (WS-12)
(b) Antagonists
tBu
N
HO
Z
S
OMe
N
HO
N
H
N
N
N
H
O
N
N
Cl
Capsazepine
Cl
2 Z = O BCTC
4 (SKF 96365)
OMe
3 Z = S Thio-BCTC
R
F3C
Cl
H
N
N
CF3
CH3
O
N
N
N
SO2
CF3
O
F
H
N
O
O
N
HO2C
R1
O
7
6
CF3
5 (RQ-00203078)
O
N
N
N
CH3
CH3
H
F
N
HO2C
HO2C
O
F3C
H3C
8
F
9
Figure 1. Representative small-molecule TRPM8 agonists (a) and antagonists (b)
The basis for TRPM8 antagonism are not fully understood at the atomic level, but much
significant information has been collected over the years. The quaternary structure of human
TRPM8 involves a symmetrical arrangement of four subunits, with the TRPM8 putative
ligand-binding site encompassing the S1-S4 transmembrane helices and the TRP domain of
each subunit. The TRP domain is known to be critical for signal transduction88 and has been
used recently to dock a series of TRPM8 ligands.89 Mutation studies show that the Tyr745
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 92
1
2
3
4
5
6
7
8
9
residue is critical for the inhibition of cold- and voltage-activated TRPM8 currents by menthol
and compound 4. However, the activity of other antagonists, such as compound 2, is not
affected by the mutation. Other antagonists (clotrimazole, econazole, capsazepine), are only
partially active in the mutated channel. Taken in the aggregate, these observations suggest the
existence in the binding site of several independent pockets able to interact with inhibitors.63
Molecular docking studies support these conclusions, and show that the known inhibitors
interact only partially with the amino acids present at the binding site. For instance, compound
4 interacts at a region close to the icilin site, stabilizing the positions of the S2 and S3 domains
by binding to Tyr745 and Asn799.63 In another example, docking of tryptamine-based
derivatives on the rat TRPM8 protein shows electrostatic interactions with Glu-1004, stacking
with Tyr754 and several hydrophobic contacts.89 In such a situation, in which several potential
binding sites are in close spatial proximity, the use of bivalent ligands may be of assistance in
the discovery of compounds with improved potencies, as often exemplified in the literature on
several therapeutic areas.90,91 We reasoned that the use of symmetrical compounds to probe
the full pocket would take advantage of the presence of two arginine residues (Arg842 and
Arg1008) at both ends of the cavity. Furthermore, symmetrical molecules have the advantage
of allowing for simpler synthetic procedures. Thus, in efforts to identify new types of TRPM8
antagonists and in view of the prevalence of urea and thiourea structural fragments in known
TRPM8 antagonists, we planned the synthesis of a library of urea derivatives inspired on the
structure of capsazepine but having the symmetrical structures 10 and 11. We were further
attracted to this idea by the fact that the impact of molecular symmetry in the discovery of
TRPM8 antagonists has never been examined. For comparison purposes, we planned the
preparation of non-symmetrical analogues 12. We also studied briefly a family of related
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 9 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds derived from a 2,5-diketopiperazine core (compounds 13 and 14) and finally,
again for comparison purposes, we examined the activity of some compounds corresponding
only to the tetrahydroisoquinoline moiety of the previously mentioned molecules (15) (Figure
2). The 6,7-dimethoxy substitution was inspired on the structure of capsazepine, whose
cathecol moiety was modified by methylation in order to achieve a better metabolic stability.
Furthermore, subsequent studies (see below) showed that these oxygen functions are crucial
for the interaction of our compounds with the Arg residues at both ends of the channel active
site.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R
R
R
R
N
NH
N
N
R1
O
R1
O
R1
10 (trans) or 11 (cis)
12
O
R1
O
N
NH
R
R
R
R
N
N
R1
O
R1
O
14
13
MeO
MeO
NH
R
15
Figure 2. Frameworks of the compounds studied in this article
Results and discussion
Chemistry
While we acknowledged that the fastest approach to our planned symmetrical urea derivatives
would involve the use of double Pictet-Spengler reactions from the known92 urea derivative
16, this method had the disadvantage that amide-type compounds are challenging Pictet-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 92
1
2
3
4
5
6
7
8
9
Spengler substrates; indeed, ureas have rarely been studied in this regard and usually only for
the case of very reactive aldehydes, such as formaldehyde.93 We found, nevertheless, that our
recently developed method based on the generation of α-amidosulfone intermediates94
provided a general solution to our problem, allowing the preparation of compounds 10 and 11
via a pseudo three-component reaction between 16 and two molecules of aliphatic or aromatic
aldehydes, in refluxing toluene containing one equivalent of toluenesulfinic acid. Under these
conditions, the reaction was fully diastereoselective and afforded exclusively the trans
isomers 10, confirming our previous findings94 (Scheme 1 and Table 1). In some cases, we
employed strongly acidic conditions (H2SO4, AcOH, 120 °C) as an alternative, which led to
the cis derivatives 11, corresponding to the other possible diastereomer of structure 10. The
formation of 10 avoids steric repulsions between the substituent R at the α’ position of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
intermediate 17 and the carbonyl and α-R substituents during the formation of the second
isoquinoline ring, thus explaining the observed diastereoselectivity under our usual conditions.
This cyclization process is altered under strongly acidic conditions, which may protonate the
urea carbonyl, although a precise reason for the formation of the cis compounds 11 in these
cases remains elusive at present.
The choice of substituents in the design of the urea library was based on the presence of aryl
substituents bearing electron-withdrawing groups on most known TRPM8 antagonists (Figure
1). Thus, after preparing the alkyl derivative 10a for comparison purposes, we obtained a
series of α,α'-diaryl derivatives bearing as substituents hydrogen (compound 10b), halogens at
various positions (compounds 10c, 10e-10g and 11g) and nitro at various positions
(compounds 10b, 10h and 10i). Because of the high potency of the compound bearing a 4-
trifluoromethyl substituent in the tetrahydrothienopyridine family of TRPM8 inhibitors,79 we
ACS Paragon Plus Environment

Page 11 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
also prepared compound 10j, which was isolated together with a small amount of its
diastereomer 11j. Some more complex aromatic substituents were also introduced, including
(p-nitrophenyloxy)phenyl (compounds 10k and 11k), 2-naphthyl (compound 10l) and 6-(3-
nitrophenyl)-2-pyridyl (compound 10m).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MeO
MeO
OMe
OMe
MeO
MeO
OMe
OMe
i or ii
+
R-CHO
!
HN
NH
N
N
N
N
!'
R
O
16
R
O
17
MeO
MeO
OMe
OMe
MeO
MeO
OMe
OMe
+
!
!'
N
N
R
O
10
R
R
O
11
R
Reagents and conditions: i. TolSO2H, toluene, 120 °C, 12 h; ii. H2SO4, AcOH, 65-120 °C, 12 h
Scheme 1. Synthesis of symmetrical urea derivatives
Table 1. Reaction conditions and results obtained in the synthesis of symmetrical urea
derivatives
R
Cmpd. a
10a
Conditions
trans/cis
1/0
Yield, %
60
n-Pr
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
Ph
10b
1/0
73
2-ClC6H4
3-O2N-C6H4
4-ClC6H4
4-BrC6H4
10c
1/0
40 b
82
10d
1/0
10e
1/0
37 c
41 d
10f
1/0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 92
1
2
3
4
5
6
7
8
9
4-FC6H4
10g
TolSO2H, toluene, 120 °C
H2SO4, AcOH, 65 °C
1/0
30
10g + 11g
11g
0.85/0.15
0/1
34
H2SO4, AcOH, 120 °C
TolSO2H, toluene, 120 °C
H2SO4, AcOH, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
TolSO2H, toluene, 120 °C
H2SO4, AcOH, 80 °C
37
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4-O2N-C6H4
10h
1/0
50
10i
1/0
20 e
16 (21)
78
NO2
O2N
---
4-F3C-C6H4
10j + 11j
10k
4/1
1/0
21
11k
0/1
36
O
O2N
10l
TolSO2H, toluene, 120 °C
1/0
40
10m
H2SO4, AcOH, 80 °C
H2SO4, AcOH, 120 °C
0/1
---
66
N
20 (29)
30 (30)
NO2
aAll compounds have the trans structure 10, except where otherwise noted. Described in
b
c
d
reference.94 Together with 37% of compound 27. Together with 54% of the monocyclic
compound 23. eTogether with 50% of compound 28
The known compounds 18-24 (Figure 3) were prepared by treatment of the urea derivative 16
with a variety of aldehydes in the presence of toluenesulfinic acid in refluxing
dichloromethane,94 in order to extend the structure-activity relationship study to non-
symmetrical, monocyclic urea derivatives. Compound 25, which was isolated during the
preparation of 19 and arose from the alkylation of the latter compound by the toluenesulfinic
ACS Paragon Plus Environment

Page 13 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
acid-propanal adduct,94 belongs to the same family of monocyclic ureas and was also studied
biologically. In order to further probe the role of molecular symmetry in TRP8 inhibition, we
also prepared the known N-acetyl compound 26,95 which can be viewed as a “half” of
compounds 10 and 11. Furthermore, side products 27-30, isolated in small amounts during
some of the experiments summarized in Table 1, were considered useful for comparison
purposes in connection with the role of a second isoquinoline ring and were also studied
biologically.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MeO
MeO
OMe
OMe
O
O
S
N
NH
H3C
NH
MeO
MeO
R
O
CH3
Cmpd.
18
19
Cmpd.
22
23
R
Et
Pr
R
N
OMe
3-NO2C6H4
4-ClC6H4
4-BrC6H4
O
OMe
H3C
25
20
21
Ph
2-ClC6H4
24
MeO
MeO
MeO
MeO
MeO
MeO
MeO
NH
N
N
O
CH3
NH
MeO
R
N
N
Cmpd.
27
R
4-ClC6H4
2,4-(NO2)2C6H3
F
26
28
30
29
NO2
NO2
Figure 3. Structures of some non-symmetrical, monocyclic urea derivatives studied in this
work, together with some additional compounds studied for comparison purposes
We also studied briefly a family of related compounds derived from a 2,5-diketopiperazine
core fused to a tetrahydroisoquinoline ring (structures 13 and 14 in the Introduction). Besides
some known94 arylmethylene derivatives of the pyrazino[1,2-b]isoquinolin-1-one system
(compounds 31-35 in Figure 4), we also examined the activity of their arylmethyl analogues
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 92
1
2
3
4
5
6
7
8
9
37 and 38, which were prepared by Pictet-Spengler cyclization of precursors 36, either using
our α-amidosulfone-mediated protocol (compound 37) or under more classical acidic
conditions (compound 38). Interestingly, both methods led to different diastereomers,
allowing us to explore the influence of this parameter on activity. The related linear
pentacyclic compound 39, arising from a double Pictet-Spengler cyclization, was also studied.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R3
O
O
R1
H
N
R2
H3CO
OCH3
R2
NH
R1
R
R3
R2
Cmpd.
R
R1
H
R3
H
31
32
Ph
4-BrC6H4
OMe
OMe
OMe
OMe
H
H
33 3-NO2C6H4
H
H
34
35
2-ClC6H4
Ph
H
H
OMe Me
OMe
Figure 4. 3-Arylmethylenepyrazino[1,2-b]isoquinolin-1-one derivatives studied in this work
ACS Paragon Plus Environment

Page 15 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
MeO
MeO
O
O
OMe
MeO
MeO
O
O
OMe
OMe
H
N
H
N
Me
OMe
Me
NH
NH
H
MeO
7
OMe
8
9
37
Ph
38
Ph
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.1 eq aldehyde
TolSO2H (1.05 eq),
Toluene, 115 °C, 5.5 h
1.6 eq aldehyde
TolSO2H (1.5 eq),
Ph2O, 120 °C, 12 h
R3
O
R1
R2
MeO
OMe
NH
HN
Ph
R2
R1
O
+
R3
36
CHO
5 eq aldehyde
TolSO2H (1.5 eq),
Ph2O, 140 °C, 6 h
Ph
MeO
O
OMe
H
Me
OMe
Me
N
N
MeO
H
MeO
O
OMe
Ph
39
Scheme 2. Synthesis of two 3-arylmethylpyrazino[1,2-b]isoquinolin-1-ones 37 and 38 and a
pyrazino[1,2-b:4,5-b']diisoquinoline-7,15(5H,13H)-dione 39
Channel inhibition results and structure-activity relationship (SAR) analysis
The synthesized TRPM8 ligands were evaluated in vitro for antagonism, as measured by
inhibition of receptor activation by the well-known TRPM8 agonist icilin. The assays were
conducted using a fluorometric assay with rat TRPM8 heterologously expressed in human
embryonic kidney (HEK-293) cells,66 and the results obtained are collected in Table 2. The
compounds tested did not significantly activate TRPM8-mediated Ca2+ elevation in
transfected HEK-293 cells (efficacy lower than 10%), with the exception of compounds 10h,
25, 31 and 38 but with a very low potency (EC50 values higher than 10 µM). On the other
hand, preincubation (5 min) of TRPM8-HEK-293 cells with different doses of the tested
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 92
1
2
3
4
5
6
7
8
9
compounds, followed by incubation with icilin (0.25 µM), caused an inhibition of the Ca2+
elevation due to TRPM8 response to icilin. The IC50 values (against icilin at 0.25 µM) are
reported in Table 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. TRPM8 activation and inhibition data
Compound
TRPM8 efficacy
TRPM8
potency
(EC50, µM)
IC50 inhibition
TRPM8, µM
(icilin 0.25 µM)
(% ionomycin, 4 µM)
MeO
MeO
OMe
OMe
HN
NH
< 10
< 10
< 10
< 10
NA
NA
NA
NA
> 50
O
16
MeO
MeO
OMe
OMe
N
N
8.5 ± 2.0
0.50 ± 0.02
19.8 ± 1.7
nPr
O
nPr
10a
MeO
MeO
OMe
OMe
N
N
Ph
O
Ph
10b
MeO
MeO
OMe
OMe
N
N
2-ClC6H4
3-NO2C6H4
4-ClC6H4
4-BrC6H4
4-FC6H4
O
2-ClC6H4
10c
MeO
MeO
OMe
OMe
N
N
< 10
< 10
< 10
< 10
NA
NA
NA
NA
44.4 ± 3.9
0.43 ± 0.10
2.2 ± 0.5
O
3-NO2C6H4
10d
MeO
MeO
OMe
N
N
OMe
O
4-ClC6H4
4-BrC6H4
4-FC6H4
10e
MeO
MeO
OMe
OMe
N
N
O
10f
MeO
MeO
OMe
OMe
N
N
0.072 ± 0.004
O
10g
ACS Paragon Plus Environment

Page 17 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
MeO
MeO
OMe
OMe
N
N
< 10
22.7 ± 1.7
< 10
NA
14.2 ± 4.2
NA
> 50
> 50
4-FC6H4
O
4-FC6H4
7
8
9
11g
MeO
MeO
OMe
OMe
N
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4-NO2C6H4
O
4-NO2C6H4
10h
MeO
MeO
OMe
N
N
OMe
> 50
(2,4-NO2)2C6H3
O
(2,4-NO2)2C6H3
10i
MeO
OMe
N
N
MeO
OMe
< 10
NA
0.18 ± 0.11
10.6 ± 1.4
4-F3CC6H4
O
4-CF3C6H4
10j
MeO
MeO
OMe
N
N
OMe
< 10
NA
4-F3CC6H4
O
4-CF3C6H4
11j
MeO
MeO
OMe
OMe
N
N
O
< 10
NA
> 50
O
O
O2N
NO2
10k
MeO
MeO
OMe
OMe
N
N
O
< 10
< 10
NA
NA
> 50
> 50
O
O
O2N
NO2
11k
MeO
MeO
OMe
OMe
N
N
O
10l
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 92
1
2
3
4
5
6
MeO
MeO
OMe
OMe
N
N
O
N
N
< 10
NA
43.7 ± 2.1
7
O2N
NO2
8
9
10m
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MeO
MeO
OMe
OMe
N
NH
< 10
< 10
< 10
< 10
NA
NA
NA
NA
> 50
> 50
Et
O
18
MeO
MeO
OMe
OMe
N
NH
NH
NH
nPr
O
19
MeO
MeO
OMe
OMe
N
1.2 ± 0.3
46.1 ± 1.5
Ph
O
20
MeO
MeO
OMe
OMe
N
N
N
N
2-ClC6H4
3-NO2C6H4
4-ClC6H4
O
21
MeO
MeO
OMe
OMe
NH
NH
NH
< 10
< 10
< 10
NA
NA
NA
> 50
O
22
MeO
MeO
OMe
OMe
31.7 ± 1.3
48.2 ± 3.0
O
23
MeO
MeO
OMe
OMe
4-BrC6H4
O
24
Et
Ts
MeO
MeO
N
NH
15.3 ± 0.1
26.1 ± 0.1
> 50
OMe
Et
O
OMe
25
MeO
MeO
N
CH3
< 10
< 10
NA
NA
> 50
> 50
4-F3CC6H4
O
26
MeO
MeO
NH
4-ClC6H4
ACS Paragon Plus Environment

Page 19 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
27
MeO
MeO
NH
< 10
< 10
NA
NA
> 50
> 50
7
8
9
(2,4-NO2)2C6H3
28
MeO
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NH
MeO
N
O2N
29
MeO
MeO
N
N
< 10
NA
22.9 ± 3.2
O2N
30
O
H
MeO
MeO
OMe
OMe
NH
N
14.7 ± 0.1
< 10
15.9 ± 0.1
NA
12.8 ± 0.3
42.4 ± 0.4
35.2 ± 2.3
Ph
O
31
O
H
N
MeO
MeO
OMe
OMe
NH
4-BrC6H4
O
32
O
H
N
MeO
MeO
OMe
OMe
NH
< 10
NA
3-NO2C6H4
O
33
O
H
N
MeO
MeO
OMe
OMe
NH
< 10
< 10
NA
NA
NA
NA
39.3 ± 0.3
0.62 ± 0.06
> 50
2-ClC6H4
O
34
O
OMe
OMe
H
N
Me
OMe
Me
NH
MeO
OMe Ph
O
OMe
35
O
OMe
OMe
Me
OMe
Me
NH
HN
MeO
OMe
O
OMe
36
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 92
1
2
3
4
5
6
7
8
9
MeO
MeO
MeO
MeO
O
O
OMe
OMe
OMe
OMe
H
N
NH
< 10
NA
30.6 ± 3.7
3.2 ± 0.1
Ph
37
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
H
N
Me
OMe
Me
NH
H
MeO
26.6 ± 0.4
20.8 ± 3.1
O
38
O
Ph
Ph
MeO
MeO
OMe
H
N
Me
OMe
Me
N
H
MeO
< 10
NA
8.8 ± 0.7
O
OMe
Ph
39
As shown previously, icilin dose-dependently elevates intracellular Ca2+ (EC50 of 0.19 ± 0.03
µM) in TRPM8-HEK-293 cells, but not in non-transfected cells.66 This assay allowed us to
profile a number of TRPM8 antagonists. The order of decreasing potency (IC50 from lower to
higher values) of the most active compounds (i.e. those showing IC50 < 10 µM against icilin),
is as follows:
10g > 10j > 10e > 10b > 35 > 20 > 10f > 38 > 39 (IC50 range from 72 nM to 8.8 µM)
Although some of the compounds have low potency and therefore reliable SAR analysis is
difficult in these cases, the data summarized in Table 2 have led to a number of useful
conclusions, which are summarized below for the two families of compounds studied.
For urea derivatives:
ACS Paragon Plus Environment

Page 21 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(a) The urea function is not sufficient for activity and the presence of at least one
tetrahydroquinoline ring is required, as shown by the comparison of the activities of
compound 16 (no tetrahydroisoquinoline ring) with 20, 21, 22 and 23 (one
tetrahydroisoquinoline ring).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(b) The isoquinoline ring, on its own, is also insufficient for activity. This is deduced from
comparison of the activities of isoquinoline derivatives 27 and 29 with the
corresponding bis- tetrahydroisoquinoline compounds generated by linking two
isoquinoline units with a carbonyl, leading to an urea moiety (10e and 10m,
respectively).
(c) Symmetric systems containing two tetrahydroisoquinoline rings linked by a carbonyl
unit are more potent than their analogues bearing a single THQ ring. This conclusion
holds true for all cases in which a direct comparison was possible (10a > 19; 10b > 20;
10c > 21; 10d > 22; 10e > 23; 10f > 24). The N-acetyl derivative 26, related to 10g but
having a single tetrahydroisoquinoline ring and a carbonyl group on nitrogen, was
inactive, giving further support to the above conclusion.
(d) C-1 phenyl substituents lead to higher potency than C-1 alkyl groups, both in the mono-
tetrahydroisoquinoline and the bis-tetrahydroisoquinoline series. Thus, 18, 19 and 25
(ethyl-substituted) were less potent than 20 (phenyl-substituted) and similarly 10a (n-
propyl-substituted) was less potent than 10b (phenyl-substituted).
(e)
The nature of substituents on the aromatic ring does not have a clear-cut influence on
activity, although a p-halogen substituent is generally favourable in the case of bis-
tetrahydroisoquinoline derivatives (cf. compounds 10e, 10f and 10g vs. the unsubstituted
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 92
1
2
3
4
5
6
compound 10b). Fluorinated substituents, which include p-fluoro (10g) and p-
trifluoromethyl (10j) lead to the highest potencies.
7
8
9
(f) A cis arrangement between the C-1 and C-1’ substituents in bis-tetrahydroisoquinoline
systems seems detrimental to activity, judging by the comparison between compounds
10g-11g and 10j-11j.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For pyrazino[1,2-b]isoquinolin-1-one derivatives:
(a) As in the previous case, the presence of an isoquinoline ring is essential for activity.
This can be deduced from the comparison of the activities of the open diketopiperazine
derivative 36 and compounds 37 and 38. The presence of a second isoquinoline ring, as
in 39, did not further improve the activity.
(b) A high degree of substitution in the isoquinoline ring increases activity, as shown by the
comparison of the activities of compounds 31 and 35, and also 37 and 38, although in
the latter case there is a stereochemical difference that makes the conclusion less clear-
cut.
In order to rationalize our results on the basis of modeling data, the most active compound
within our series, i.e. 10g, was docked into the putative ligand binding site encompassing the
S1-S4 transmembrane helices and the TRP domain domain site of the rat TRPM8, built as
described in the Experimental section. One of the noteworthy features of this model is the
position of both Asn799 and Asp802 residues, lying on S3 and known to affect icilin binding,3
which are located inside the pocket. The study of 10g by a docking protocol requires, as a
prerequisite, a detailed conformational analysis, since it can assume different, interdependent
ACS Paragon Plus Environment

Page 23 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
conformational states of its two non-aromatic rings. Unfortunately, the implementation of ring
conformational flexibility in most docking programs is at best problematic, and thus molecular
conformations corresponding to all the energetically-accessible combinations of ring dihedral
angles were generated by a conformational search, followed by ab-initio optimization, and
separately docked for both enantiomers of 10g. A total of 28 best poses, selected as described
in the Experimental Section from docking runs carried out on twelve (i.e. six for each
enantiomer) conformers, were energy minimized and scored by ligand-protein interaction
energy (EI) (see Table S.1 of the Supporting Information). Figure 5 shows the most stable
pose for each enantiomer, while Table 3 lists interaction energies and the main polar
interactions for these poses, along with those of the other four exhibiting EI below -95 kcal
mol-1. The emerging pattern of ligand-protein interactions involves: the arginine residues at
both ends of the site, i.e. Arg842, Arg1008, with the dimethoxytetrahydroisoquinoline
moieties (H bonds) and, in some cases, with the fluorobenzene groups (ionic-π interactions);
Tyr754 and Tyr1005 with either fluorophenyl or tetrahydroisoquinoline groups (aromatic
interactions); Arg851 and Asn692 with the urea group (H-bonds). Ligand fluorine atoms also
engage in favorable polar interactions with protein backbone carbonyl groups, a kind of
interactions frequently occurring in X-ray structures of proteins with fluorinated ligands.96
The peculiar interactions engaged by the fluorine atoms with both backbone and arginine
residues help to rationalize the higher activity of this derivative in comparison with those
bearing other halogens (compounds 10e and 10f) and/or hydrogen (compound 3) in place of
fluorine. The ability of compound 10g to engage multiple favorable polar interactions
validates our starting assumption that symmetrical molecules containing two
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 92
1
2
3
4
5
6
tetrahydroisoquinolines linked by a urea moiety would provide strong interactions with the
TRPM8 binding site.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. The best docked complexes of rat TRPM8 monomer model with the RR (A) and SS
(B) enantiomers of compound 10g are shown, using a ribbon representation for the protein
backbone and sticks for the ligands and the side chains of protein residues involved in contacts
with the ligand, which is colored in dark magenta. Atoms are colored with the following code:
O=red, N=blue, H=white, F=green. Ligand-protein H-bonds are depicted with a green spring.
The main contacts and interaction energies associated to docking poses A and B are
summarized in Table 3.
ACS Paragon Plus Environment

Page 25 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3: Interaction energy (EI) and main interactions in the docking binding poses of
compound 10g with EI < 95 kcal mol-1. Poses are numbered with roman numerals by
increasing EI and the corresponding ligand chiralities are reported in parenthesis (in bold for
the stablest pose for both RR and SS enantiomers, corresponding to panels A and B in Figure
5, respectively).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ligand pose Interaction
H-Bondsa
Polar/aromatic interactionsa
(chirality)
Energy
(kcal mol-1)
I (RR)
-121.586
Urea CO-
FB2-Arg1008(sc)/Tyr754,
Asn692(sc)/Arg851(sc)
MeO-THIQ1-Arg1008(sc)
MeO-THIQ2-Arg842(sc)
THIQ2-Arg842(sc)/His845(sc)
II (RR)
-108.579
Urea CO-Asn692(sc),
THIQ2-Arg842(sc)/His845(sc),
FB2-His845(sc)/Asn799(bb)
FB1-Tyr754(sc)/Asn692(sc)
MeO-THIQ1-Arg1008(sc)
III (SS) -107.264
Urea CO-
FB1-
Tyr1005(sc)/Arg851(sc)
MeO-THIQ1-Arg1008(sc)
MeO-THIQ2-Arg842(sc)
Tyr754(sc)/Asn692(sc)/Tyr1005
THIQ2-Arg842(sc)/His845(sc),
FB2-Asn799(bb)/His845(sc)
IV (SS)
V (RR)
-103.850
-97.725
Urea CO-Asn692(sc)
FB2-Arg842(sc)
MeO-THIQ1-Arg1008(sc)
FB1-His845/Asn799(bb)
THIQ1-Tyr1005, THIQ2-
Tyr754
Urea CO-Tyr754(sc)
FB1-Tyr1005(sc)/Asn692(sc)
FB2-Arg842(sc)
MeO-THIQ2-Arg851(sc)
VI (RR) -95.080
Urea CO-Asn692(sc)
FB1-Arg842(sc)/His845(sc),
THIQ2-Tyr754
MeO-THIQ1-Arg851(sc)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 92
1
2
3
4
FB2-
5
6
7
8
Arg851(sc)/Val849(bb),THIQ1-
Arg851
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a THIQ: tetrahydroisoquinoline moiety; FB=Fluorobenzene ring; the 1 and 2 indices refer to
symmetric units of molecule (R1-CO-R2); sc= side chain; bb=CO backbone
Some of the more active compounds were also tested as potential inhibitors of other thermo
TRP channels. The data displayed in Table 4 indicate IC50 values higher than 10 µM for all the
compounds and for all the receptors tested, showing a good degree of selectivity for TRPM8
for all these compounds, but particularly in the case of compound 10g.
Table 4. Inhibition of TRPM8 and other thermo TRP channels by compounds 10b, 10e, 10g
and 35 (IC50 µM). The values in parentheses correspond to the TRPM8/TRPx selectivities
Channel
TRPM8
10b
10e
10g
35
0.50 ± 0.02
0.43 ± 0.10
0.072 ± 0.004
0.62 ± 0.06
(icilin 0.25 µM)
TRPA1
10.0 ± 2.0
(20:1)
23.9 ± 0.7
(56:1)
16.4 ± 2.1
(228:1)
14.2 ± 0.6
(23:1)
(AITC 100 µM)
TRPV1
53.0 ± 1.0
(106:1)
49.7 ± 1.7
(116:1)
68.5 ± 2.1
(951:1)
68.9 ± 1.5
(111:1)
(capsaicin 0.1 µM)
TRPV2
51.9 ± 8.0
(104:1)
54.9 ± 9.4
(128:1)
94.3 ± 10.0
(1309:1)
72.9 ± 4.0
12.4 ± 1.4
(20:1)
(LPC 3 µM)
TRPV3
> 100
> 100
34.3 ± 4.0
ACS Paragon Plus Environment

Page 27 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(thymol 100 µM)
TRPV4
(> 200:1)
(> 233:1)
81.5 ± 0.1
(189:1)
(1012:1)
96.2 ± 16.3
(1336:1)
(55:1)
> 100
82.6 ± 0.1
(194:1)
(4αPDD 1µM)
(> 161:1)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Anti-proliferative activity
TRPM8 mRNA has been detected in malignant cells and this channel has been extensively
studied in prostate cancer,47 Our previous study showed that TRPM8 mRNA is highly
expressed in LNCaP epithelial prostate cancer cells.50 It has been reported that TRPM8 might
be a useful marker for prostate cancer, since loss of TRPM8 expression appears to be
associated to transition to androgen independence and poor prognosis.97,98
.
Against this backdrop, we have studied the anti-proliferative activity on LNCaP prostate
cancer cell lines of a selection of the most active TRPM8 inhibitors, namely compounds 10b,
10e, 10g and 35. We found that at all these compounds, at 10 µM concentration, exert a
marked anti-proliferative activity, inhibiting after three days of treatment the cellular growth
from a minimum of 50.4 % (for 10e) to a maximum of 64.3 % (for 10g). We employed
compound 11g, a diastereomer of 10g that does not inhibit TRPM8, as a negative control. As
positive control, we used compound 40, an antagonist of TRPM8 whose effectiveness has
been previously tested on LNCaP cells71 (Figure 6). On the other hand, cell viability was only
moderately affected, as evaluated by the viability-dye CBNF (see Figure 6 below). At the
concentrations used, we did not observe any cytotoxic effects of the drugs, as measured by
trypan blue.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 92
1
2
3
Me
4
5
O
6
7
N
H
Me
Me
8
9
40
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
75
50
25
0
Figure 6. Evaluation of compounds 10b, 10e, 10g, 11g, 35 and 40 on LNCaP cell growth.
LNCaP cells were cultured for 72 hours and treated with the compounds under assay at the
indicated concentrations, as described in the Experimental Section. The cell number was
evaluated and the statistical analysis performed as described in the Experimental Section (p <
0.001 10g vs LNCaP).
As shown in Figure 7 and in Table 5, the effect of compound 10g on cell viability of prostatic
tumor cell LNCaP and 22RV1 and of human normal prostate epithelium cells PNT1A, was
evaluated by the viability dye CBFN. The cells were gated by selecting the threshold of red-
fluorescence at ≥ 10 fluorescence units. A different response of the three cell lines was
observed. In fact, in the case of LNCaP cells the percent of gated cells was 83.3 % vs 54.4 %
ACS Paragon Plus Environment

Page 29 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and the fluorescence mean value in the gated regions was 215.97 vs 176.01 for the vehicle and
cells treated with compound 10g, respectively. These data indicated a different viability
degree in LNCaP treated by compound 10g, thus confirming the slowdown of cell growth. On
the contrary, for the 22RV1 and PNT1A cells, both the percent of gated cells and the
fluorescence mean values were unaffected by the treatment. In order to evaluate if these
effects on the cell growth and viability are mediated by the TRPM8 receptor, the
transcriptional TRPM8 levels were evaluated by qRT-PCR in the three prostatic cell lines and
in the human kidney embryonic cells HEK-293.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment