Isoxazole derivatives as potent transient receptor potential melastatin type 8 (TRPM8) agonists

Article abstract of DOI:10.1016/j.ejmech.2013.08.056

Modulation of the transient receptor potential melastatin type-8 (TRPM8), the receptor for menthol acting as the major sensor for peripheral innocuous cool temperatures, has several important applications in pharmaceutical, food and cosmetic industries. I

Full text of DOI:10.1016/j.ejmech.2013.08.056

European Journal of Medicinal Chemistry 69 (2013) 659e669
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Isoxazole derivatives as potent transient receptor potential melastatin
type 8 (TRPM8) agonists
,
,
,
,
Carmine Ostacolo a 1, Paolo Ambrosino b 1, Roberto Russo a 1, Matteo Lo Monte c,
*
Maria Virginia Soldovieri b, Sonia Laneri a, Antonia Sacchi a, Giulio Vistoli c,
Maurizio Taglialatela b
, Antonio Calignano
,
d,
a
**
a Department of Pharmacy, University of Naples Federico II, Via D. Montesano 49, 80131 Naples, Italy
b Department of Medicine and Health Sciences, University of Molise, Via F. De Sanctis, 86100 Campobasso, Italy
c Department of Pharmaceutical Science Piero Pratesi, University of Milan, Via L. Mangiagalli 25, 20133 Milan, Italy
d Div. Pharmacology, Department of Neuroscience, University of Naples Federico II, Via Pansini 5, 80131 Naples, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Modulation of the transient receptor potential melastatin type-8 (TRPM8), the receptor for menthol
acting as the major sensor for peripheral innocuous cool temperatures, has several important applica-
tions in pharmaceutical, food and cosmetic industries. In the present study, we designed 12 isoxazole
derivatives and tested their pharmacological properties both in F11 sensory neurons in vitro, and in an
in vivo model of cold allodynia. In F11 sensory neurons, single-cell Ca2þ-imaging experiments revealed
that, when compared to menthol, some newly-synthesized compounds were up to 200-fold more potent,
though none of them showed an increased efficacy. Some isoxazole derivatives potentiated allodynic
responses elicited by acetone when administered to rats subjected to sciatic nerve ligation; when
compared to menthol, these compounds were efficacious at earlier (0e2 min) but not later (7e9 or 14
e16 min) time points. Docking experiments performed in a human TRPM8 receptor model revealed that
newly-synthesized compounds might adopt two possible conformations, thereby allowing to distinguish
“menthol-like” compounds (characterized by high efficacy/low potency), and “icillin-like” compounds
(with high potency/low efficacy). Collectively, these data provide rationale structureeactivity relation-
ships for isoxazole derivatives acting as TRPM8 agonists, and suggest their potential usefulness for cold-
evoked analgesia.
Received 10 May 2013
Received in revised form
20 August 2013
Accepted 25 August 2013
Available online 20 September 2013
Keywords:
Transient receptor potential melastatin
type-8 channels
Isoxazolylamine derivatives
Structureeactivity relationship
Calcium imaging
Chronic constriction injury
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
somatosensory afferents [5]. Among these, transient receptor po-
tential melastatin type 8 (TRPM8) channels act as Ca2þ-permeable
Since the days of Hippocrates and Galen [1,2], cooling agents
have been successfully used to induce analgesia during traumatic
injuries [3,4]. However, little is known about the molecular
mechanisms underlying cooling-induced analgesia, albeit a num-
ber of cool-sensitive ion channels have been identified in
ligand-gated cation channels activated by chemical cooling agents
(such as menthol) or by cold temperatures (<25 ꢀC) [6e8]. Both
peripheral and central activation of TRPM8 induces analgesic ef-
fects that specifically reverse the sensitization of the behavioral
reflexes elicited by peripheral nerve injury [9]. These effects are
produced in a range of very low concentrations of topically applied
TRPM8 activators, whereas high concentrations of menthol were
found to cause both cold and mechanical hyperalgesia in healthy
volunteers [10,11]. These findings suggest the potential relevance
of TRPM8 activators as therapeutic strategy for pain treatment
[12e14].
Abbreviations: TRPM8 channels, transient receptor potential melastatin type-8
channels; Ca2þ
, calcium; GM, growth medium; DM, differentiation medium;
DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; RT-PCR,
reverse-transcription PCR; RT, room temperature; SAR, structureeactivity rela-
tionship; CCI, chronic constriction injury.
Expression of TRPM8 also represents a prognostic marker and a
therapeutic target in prostate cancer (PCa): indeed, the activation of
TRPM8 receptors by menthol enhanced apoptosis of lymph node
carcinoma of prostate cells (LNCaP) [15], pointing to a potential
involvement of TRPM8 channels in cancer cell growth.
* Corresponding author. Department of Pharmacy, University of Naples Federico
II, Via D. Montesano 49, 80131 Naples, Italy. Tel.: þ39 081 6788609.
** Corresponding author. Department of Medicine and Health Sciences, University
of Molise, Via F. De Sanctis, 86100 Campobasso, Italy.
E-mail address: ostacolo@unina.it (C. Ostacolo).
1
These authors equally contributed to this work.
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.08.056

660
C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
Scheme 1. (AeG) Molecular structures of main TRPM8 modulators: menthol (A), icilin (B),
a-ketoenamine (C and D), benzo-fused imidazole (E), oxazole (F) and thiazole derivatives
(G). Generic structure of the newly synthesized compounds of the present study (H).
Given the potential role of TRPM8 agonists for the treatment of
neuropathic pain conditions and cancer, increasing efforts in the
last few years have been dedicated to the design of selective and
potent TRPM8 ligands. To date, the most active cooling compound,
icilin (Scheme 1B), is characterized by the presence of a central
tetrahydropyrimidine-2-one ring substituted by a nitrophenyl and
phenol moieties. Several other analogues, based on the tetrahy-
dropyrimidine-2-one moiety, have been described and patented for
their cooling activity [16].
The need for a heterocyclic ring to elicit efficient interactions
with TRPM8 is also witnessed by the isolation of the potent cooling
agents alpha-ketoenamine (Scheme 1C and D) from roast malt
extract [17]; however, these molecules are penalized by their great
instability to oxidation, probably related to the reactive enamine
functional group. A similar instability was also found for a class of
aminophenols synthesized as hybrid derivatives of icilin and
ketoenamines [17]. Therefore, and in view of these stability issues,
different heterocyclic moieties have been utilized as chemical
scaffolds for the synthesis of new TRPM8 modulators; these include
benzimidazole-based (Scheme 1E), as well as fused oxazole
(Scheme 1F) and thiazole derivatives (Scheme 1G) [18,19]. In
addition, aminoisoxazoles, in particular the 3-amino derivatives,
are resistant to the typical degradation occurring to the isoxazole
rings [20], and are suitable for simple and high-yield chemical
derivatization of the amino group.
modifications on the isoxazole ring. These mainly involved the
introduction of aminoaliphatic or aminoaromatic chains in position
3 and/or the introduction of a methyl group in position 5 of this
scaffold ring.
When tested for their ability to trigger TRPM8-induced [Ca2þ
]
i
responses in sensory neurons in vitro, some of these derivatives
showed higher potency when compared to menthol. Thereafter, the
most potent compounds were tested in vivo in an animal model of
cold allodynia, where they showed strong and rapid, although
short-lasting, allodynic responses. In order to identify the structural
determinants for the higher potency of specific compounds, dock-
ing experiments were performed using the recently described
TRPM8 model [23]. This approach revealed that the more potent
compounds strongly stabilized the agonist binding site of TRPM8
receptors, favoring its active configuration. Altogether these results
suggest that the newly-synthesized isoxazole derivatives act as
potent TRPM8 modulators, and may represent promising drugs for
pain treatment.
2. Results and discussion
2.1. Chemistry
Tested compounds were synthesized in a one-pot procedure, by
reductive alkylation, using aldehyde or ketone and isoxazolyl-
amines as starting materials (Scheme 2). To the proper isoxazolyl-
amine (10.2 mmol), dissolved acetic acid/dichloroethane (1:5 v/v,
20 ml), aldehyde or ketone was added (20.4 mmol). The solution
thus obtained was warmed at 60 ꢀC for 1.5 h, under magnetic
stirring. Sodium triacetoxyborohydride (18.4 mmol) was then
added portionwise, and the reaction was maintained at the same
temperature. After 3.5 h the mixture was cooled and NaOH 1 N
(15 ml) was added. The organic phase was separated and extracted
one more time with the alkaline solution. Then it was dried over
Na2SO4, filtered and concentrated in vacuo. The crude products
were purified by column chromatography using mixtures of n-
hexane/ethyl acetate as eluent. Final products were crystallized, if
Substituted aminoisoxazole derivatives seem to include the
minimum structural requirements for a TRPM8 ligand, as seen for
menthol or icilin, such as hydrogen bonding groups, a compact
(mainly cyclic) hydrocarbon backbone, a correct hydrophobic/hy-
drophilic balance with a log P range between 1 and 5, and a mo-
lecular weight in the range of 150e350 g/mol. Indeed, several
structural similarities can be observed between the alpha-
ketoenamine derivatives and the aminoisoxazoles (Scheme 1H)
[21].
Based on these evidence and on the known SARs for menthol
(Scheme 1A) and its analogues [21,22], in the present study we
synthesized 12 new isoxazole derivatives (Scheme 2) carrying

C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
661
Scheme 2. Representation of the synthesis procedure for all isoxazolylamine derivatives of the present study (see Experimental section for details) and classification of the
substituent groups present in each compound.
possible, by mixtures of n-hexane/ethyl acetate and were easily
obtained in excellent yields (62e73%).
derivatives belonging to the series 3 (3b and 3d) lack the methyl
group in position 5. Finally, the derivatives 2b and 2d are isomers of
1b and 1d, due to inversion of the oxygen and nitrogen
heteroatoms.
2.2. Characterization of TRPM8-mediated menthol-induced [Ca2þ
increase in F11 cells
]
i
To characterize their pharmacological activity as TRPM8 ago-
nists in vitro, these new compounds were tested for their ability to
enhance [Ca2þ]i in differentiated F11 cells, as previously shown for
menthol. All tested molecules increased [Ca2þ]i in a concentration-
dependent manner, although with different efficacy and potency
(Fig. 2AeN). To simplify comparison of the efficacy among tested
compounds, the maximal fluorescent increase prompted by each
compound was normalized to the maximal response produced by
menthol. As reported in Table 1, none of the tested compounds
showed a higher efficacy than menthol. The compounds referred as
1d and 1f showed the same efficacy than menthol (p >0.05);
instead, the other ten tested compounds were less effective than
menthol, showing the following rank order of efficacy:
1g > 3b > 3d > 1h > 1c > 1e > 1b > 1a > 2b > 2d, suggesting that
longer aliphatic chains or aromatic substituents at the position R2
are required to retain high efficacy.
F11 cells, a cloned cell line obtained by the fusion of embryonic
rat dorsal root ganglion (DRG) neurons with mouse neuroblastoma
cells [24], have been widely used to study nociceptive molecular
mechanisms [25]. RT-PCR experiments revealed the expression of
TRPM8 transcripts in undifferentiated F11 cells (expected size of
the band: about 130 bp; insert in Fig.1A). Furthermore, quantitative
real time PCR revealed that differentiated F11 cells showed a 4-fold
increase in TRPM8 transcript expression when compared to un-
differentiated cells (Fig. 1A). Based on these results, following ex-
periments were performed in differentiated cells. As previously
demonstrated in TRPM8-expressing cells [6], exposure of F11 cells
to
L
-menthol (3e300
mM) for 20 s led to a transient, concentration-
dependent increase in intracellular Ca2þ concentrations [Ca2þ
]
i
(Fig. 1B). The calculated EC50 for menthol-induced [Ca2þ]i increase
was 23.4 ꢁ 2.6
m
M (Fig.1C), a value consistent with previous reports
By contrast, comparison of the potency among tested com-
pounds was performed by normalization of the experimental
data to the maximal value of fluorescence ratio prompted by each
compound. As shown in Fig. 3 and summarized in Table 1, eight
of the tested compounds (i.e. 1aee, 1h, 3b, 3d) showed a
significantly lower EC50 when compared to menthol, suggesting
that aliphatic substituents at the R2 and R1 positions are
involved in conferring high potency to the tested compounds.
The compound referred as 1g showed the same potency as
menthol (p >0.05); finally, the other three tested compounds
were less potent than menthol, with the following rank order of
potency: 2b > 2d > 1f, suggesting that the introduction of an
aromatic ring at the position R2 or the inversion of the hetero-
atoms at the position X, Y (see Scheme 2) have deleterious effects
on compound potency.
As later detailed in the Discussion paragraph, the reported
newly-synthesized molecules can be roughly subdivided into two
groups: the first includes compounds characterized by sub-
micromolar potency but with quite low efficacy (as seen for
example in 1aec derivatives), and the second includes compounds
endowed with low potency and high efficacy (as exemplified by 1f
and 1g). The compound 1d appears to be the most promising
compromise since it shows low micromolar potency combined
with a very remarkable efficacy.
[26,27]. Menthol-induced [Ca2þ]i response displayed a very rapid
onset (within about 1 s), and peaked after 10e15 s. By contrast, the
kinetics of [Ca2þ]i decline following menthol removal from the bath
appeared to depend on agonist concentration, being slower at
higher menthol concentrations (range between 50 and 400 s). The
effect of menthol (100 m
M) on [Ca2þ]i levels in F11 cells was fully
prevented by the TRPM8 antagonist BCTC (3 mM) [28] (Fig. 1D) or by
agonist exposure in a Ca2þ-free solution (plus 1 mM EDTA) (Fig.1D).
These observations suggest that menthol-induced enhancement of
[Ca2þ
] was due to its ability to activate TRPM8 channels, and
i
depended on an enhanced flux of extracellular Ca2þ ions across the
plasma membrane.
2.3. Pharmacological properties of newly synthesized
aminoisoxazole derivatives
In the present study, 12 new isoxazole derivatives were syn-
thesized, bearing substituents at position 3 and 5 of the hetero-
aromatic ring (see Scheme 2), to define the best structural
requirements for the pharmacological activity of each compound.
In particular aminoaliphatic (1aee) and aminoaromatic (1feh)
chains were introduced at position 3 for derivatives belonging to
series 1, while a methyl group occupies position 5. By contrast,

662
C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
Fig. 1. Expression of TRPM8 transcripts in F11 cells and characterization of the menthol-induced responses. (A) The insert indicates the RT-PCR obtained by mRNA from undif-
ferentiated F11 cells. The amplicon expected size for TRPM8 was 130 bp (lane 2), as indicated. GAPDH was used as housekeeping gene (amplicon expected size: 300 bp, lane 1).
DDct
Quantitative real-time PCR on the total mRNA of undifferentiated (GM; black bar) or differentiated (DM; white bar) F11 cells. Data are expressed as 2ꢂ
relative to GAPDH (see
Experimental section for details). Each bar is the mean ꢁ S.E.M. of three separate determinations; *p <0.05 versus control (F11 cells in GM). (B) Representative traces of the changes
in fluorescence ratio (F/F0) evoked by 1 M (orange trace), 3 M (sky-blue), 10 M (violet), 30 M (green), 100 M (brown), or 300 M (blue) menthol. The bar length on the top of all
m
m
m
m
m
m
traces corresponds to the duration of drug exposure. (C) Concentrationeresponse curve from the experimental data obtained as in C. The solid line represents the fit of the
experimental data to a standard binding equation of the following form: y ¼ max/(1 þ x/EC50)n, where x is the drug concentration and n the Hill coefficient; n was 1.1 ꢁ 0.1. Each
point is the mean ꢁ S.E.M of 23e28 separate determinations performed in at least 3 experimental sessions. (D) Quantification of the fluorescence ratio (F/F0) peaks measured in
differentiated F11 cells at basal levels or after the exposure to 100 mM menthol in a control solution, in co-perfusion with the TRPM8 antagonist BCTC (3 m
M) or in a Ca2þ-free
solution. Each bar represents the average of 12e23 determinations obtained in at least 3 experimental sessions.
2.4. Pharmacological effects of selected aminoisoxazole derivatives
on acetone-induced allodynic responses in rat with chronic
constriction injury (CCI) of the sciatic nerve
induce cold allodynia after the second and third challenge (Fig. 4D
and E). At least for 1aec, allodynic responses were stronger than
those observed with menthol after any acetone challenge. Instead,
1f, the compound showing the lowest potency in in vitro testing,
did not produce cold allodynia after any acetone application
(Fig. 4CeE).
Based on the in vitro results, we compared the effects of the
most potent compounds (1aed) with those of menthol in the CCI
rat model of cold allodynia. To this aim, 50 ml of each compound
(dissolved in 20% w/v in absolute ethanol) were topically (trans-
dermally) applied on the operated paw, and the allodyinc responses
(number of paw retraction) to three consecutive application of
acetone (each lasting 2 min, applied every 7 min, see protocol in
Fig. 4A) were evaluated 14 days after sciatic nerve ligation. When
the total number of allodynic responses occurring during acetone
application were measured, it was observed that local application
of menthol, 1a, 1b, or 1c elicited similar cold allodynic responses,
whereas 1d and 1f were ineffective (p >0.05 versus vehicle;
Fig. 4B). However, when the effects of the TRPM8 activators were
measured after each of the three consecutive acetone applications,
it was found that menthol failed to produce a significant effect of
cold allodynia after the first (Fig. 4C) and the third (Fig. 4E) acetone
application, while significant effects were detected after the second
acetone exposure (Fig. 4D). By contrast, the time-dependence of the
allodynic responses triggered by the newly synthesized com-
pounds was clearly distinct from that of menthol, showing faster
onset and offset kinetics. In fact, compounds 1aed were efficacious
after the first acetone application (Fig. 4C), whereas they failed to
2.5. Computational studies
To identify the structural elements involved in the interactions
of the here reported isoxazole derivatives with the TRPM8 channel
which might provide a plausible explanation for their different
pharmacological properties, docking simulations of these mole-
cules on the recently reported structural model of the human
TRPM8 receptor [23] were performed. As shown in Fig. 5A, the
main interactions stabilizing the complex between TRPM8 and the
most potent 1a derivative occur in a subpocket adjacent, but not
coincident, with that of menthol, roughly corresponding to that
occupied by the nitrophenyl moiety of icilin [23].
This result emphasizes the role of N799, which stabilizes an
extended
pep stacking involving the carbamido function of the
aminoacid side chain and the entire isoxazole ring. Furthermore,
ligand’s amino group is involved in an H-bond with the D802
carboxylate, a further key residue determining the activity of icilin,
as also demonstrated by mutagenesis [26]. Finally, the interaction
between ligand and the D802 seems to be essential for agonism as

C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
663
Fig. 2. Concentration-dependent effects of isoxazole derivatives. (AeN) Concentrationeresponse curves in terms of Fura-2 fluorescence ratio (F/F0) obtained upon perfusion of 1a
(0.001e30 M), 1b (0.01e30 M), 1c (0.01e100 M), 1d (0.01e100 M), 1e (0.3e100 M), 1f (10e1000 M), 1g (0.1e300 M), 1h (0.1e300 M), 2b (1e500 M), 2d (10e1000 M), 3b
(1e100 M), 3d (1e100 M) compounds on differentiated F11 cells. The solid lines represent the fit of the experimental data to a standard binding equation (see above). Each point
m
m
m
m
m
m
m
m
m
m
m
m
is the mean ꢁ S.E.M. of 27e63 separate determinations performed in at least 3 experimental sessions. The effects of all compounds are represented on the same Y scale to facilitate
comparison.
it destroys the H-bond between D802 and Y745, which character-
izes the inactive state of TRPM8 channels, thus triggering its acti-
vation [29]. These two major contacts are further stabilized by a set
of hydrophobic interactions with surrounding apolar side chains
(e.g. L749, A753 and F794).
Notably, the isopropyl group of the 1a derivative approaches
Y745 contacting the apolar residues involved in the interactions
with menthol, while the ligand’s methyl group seems to act as a
pivot which restrains the ligand pose within the binding site, thus
maximizing its interactions (see below). Accordingly, docking re-
sults show that the desmethyl analogue (3b) is characterized by a
slightly different pose, in which the ligand slides towards F794 and
loses its key H-bond with D802. Again, the isopropyl chain affords
the best potency presumably because it balances apolar contacts
and steric hindrance, while bulkier groups (such as in 1b or in 1d)
tend to clash against Y745 or L749. Interestingly, the isopropyl
Table 1
Pharmacological properties of isoxazole derivatives.
Compound
Potency
EC50 M)
Efficacy
(
m
(% Menthol maximal response)
Menthol
1a
1b
1c
1d
1e
1f
1g
1h
2b
2d
3b
3d
23.4 ꢁ 2.6
0.10 ꢁ 0.01##
0.16 ꢁ 0.01##
0.5 ꢁ 0.1##
3.6 ꢁ 3.5##
13.9 ꢁ 1.5#
400.0 ꢁ 1.4y
18.9 ꢁ 3.6
100%
53 ꢁ 4%*
58 ꢁ 4%*
66 ꢁ 4%*
89 ꢁ 8%
64 ꢁ 7%*
82 ꢁ 9%
74 ꢁ 5%*
66 ꢁ 6%*
36 ꢁ 4%**
32 ꢁ 3%**
72 ꢁ 8%*
69 ꢁ 6%*
12.0 ꢁ 2.9#
48.4 ꢁ 3.7y
100.0 ꢁ 5.6y
13.2 ꢁ 1.3#
17.3 ꢁ 1.6#
Fig. 3. Concentrationeresponse curves for all newly synthesized isoxazole derivatives.
F/F0 fluorescent ratios obtained from Fig. 2AeN were divided by the maximal fluo-
rescent increase obtained for each compound and subtracted by the fluorescent ratio
at the lowest drug concentration. The solid lines represent the fit of the normalized
data to a standard binding equation (see above) and are indicated in different colors for
each compound, as indicated.
#Potency values significantly (p <0.05) lower than menthol.
##Potency values significantly (p <0.05) lower than compound with #.
yPotency values significantly (p <0.05) higher than menthol.
*Efficacy values significantly (p <0.05) lower than menthol.
**Efficacy values significantly (p <0.05) lower than compound with *.

664
C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
Fig. 4. Quantification of the cold responses evoked by the newly-synthesized compounds in vivo. (A) Schematization of the time procedure followed for the experiments plotted in
B, C, D and E. (B) Total number of the cold responses measured during the full-time experiment after 20 min of preincubation with the indicated compounds in CCI mice treated
with acetone. (C, D, E) Time-course of the cold responses prompted upon topical application of vehicle (V), menthol (M) or the indicated compounds on the effects evoked by
acetone on CCI mice during the first (C), second (D) or the third (E) phase of the experiment. Data are the mean ꢁ SEM (n ¼ 6). *p <0.05, **p <0.01, ***p <0.001, all versus vehicle; p
<0.05, p <0.01, p <0.001, all versus menthol.
group corresponds to the substituent also inserted in a comparable
pose by menthol, suggesting that this group best fits the TRPM8
cavity in that region.
could explain their poor potency. In detail, their isoxazole ring
stabilizes an H-bond between the intra-annular oxygen atom and
the N799 side chain, losing the extended
pep stacking and, more
By contrast, when the N-linked group is further enlarged, as
seen in the 1feh derivatives, the resulting ligands cannot assume
the above described pose and show a completely different
arrangement in which the isoxazole ring approaches Y745, while
the bulky N-linked moiety approaches N799. Fig. 5B displays the
pose of the 1h derivative showing that the isoxazole ring is unable
importantly, these compounds appear to be unable to elicit the
pivotal contact with D802.
3. Conclusion
In the present study, the pharmacological properties of newly
synthesized aminoisoxazole derivatives with respect to their
ability to act as novel TRPM8 agonists have been studied using
[Ca2þ]i-imaging experiments in sensory neurons in vitro, and an
in vivo model of cold allodynia. The use of an isoxazole ring as
key scaffold for these molecules was chosen because of many
chemical considerations (see below); modifications on this
scaffold mainly concerned the introduction of a secondary amine
group linked to different aliphatic chains in the position 3 of the
isoxazole ring and/or the introduction of a methyl group in the
position 5.
to elicit both significant
pep stacking and the key H-bond with
Y745. The nitrophenyl moiety mimics the pose assumed by the
same group of icilin stabilizing a key H-bond which involves the
nitro function and the side chain of N799, while the amino group of
1h interacts with D802. However, the inability to contact Y745 as
well as the steric clashes exerted with surrounding residues could
explain its poor potency, an effect which appears to be dramatically
exacerbated when the ligand cannot elicit H-bonds with N799 (as
seen for 1f). Finally, docking simulations revealed that the inverted
3-methyl derivatives (2b and 2d) assume a different pose which

C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
665
Fig. 5. Magnification of the binding site in TRPM8 channel. (A, B) Main interactions stabilizing the complex between the TRPM8 binding site and 1a (A) or 1h (B). The cartoon for the
transmembrane helices is colored according to their orientation within the membrane bilayer (gray ¼ extracellular side; pink ¼ intracellular side).
When tested in vitro, the compounds containing modifications
only at the 3-position, particularly a secondary amine linked to an
isobutyl (3b) or a butyl (3d) group, displayed a slightly higher po-
tency when compared to menthol, whereas their efficacy was
slightly reduced. The introduction of a methyl group at the 5-
position in both these compounds (generating 1b and 1d, respec-
tively), lead to a further improvement of the potency, without
altering the efficacy in the case of 1b or inducing a small, but sig-
nificant efficacy increase in the case of 1d when compared to 3d.
Further modifications of the aliphatic chain connected with the
secondary amine (consisting in an isopropyl or an isobutyl group in
the case of 1a or 1c, respectively) prompted to similar improvement
in the potency, but still not in the efficacy, when compared to the
parent compounds (3b or 3d).
the second acetone application, and the 1f compound failed to
display allodynic effects at any time point after stimulus challenge.
These observation suggests the existence of a good correspondence
between in vitro and in vivo data, with those compounds showing
the highest in vitro potency as TRPM8 agonists being provided of
faster onset and offset in vivo effects when compared to menthol,
and the lower potency TRPM8 agonist showing no efficacy in in vivo
testing. As mentioned in the Results paragraph, the in vitro phar-
macological profile shown by the newly-synthesized molecules
allows to classify them in menthol-like derivatives, characterized
by high efficacy but quite low potency, and icilin-like derivatives,
the latter showing remarkable submicromolar potencies but char-
acterized by an efficacy lower than that of menthol. The results
achieved in our docking experiments suggest valid structural hy-
pothesis to rationalize these biological evidence. Indeed, the
computed complexes suggest that the TRPM8 binding site can be
roughly subdivided into two regions. The first region is more hy-
drophobic, and contains the Y745 residue playing a key role in
TRPM8 activation [34]. Given the nature of the flaking residues, a
ligand can occupy such a region by stabilizing a set of precise hy-
drophobic contacts plus key H-bonds with Y745,as seen for the
prototypical example of menthol. The second region is more polar
as it is lined by N799 and D802 which can elicit reinforced H-bonds
The increase in potency showed by the described compounds is
instead prevented by the introduction of an aromatic function on
the groups linked to the amine at the 3-position of the isoxazole
ring. In fact, when a benzyl group was linked to the secondary
amine (1f),
a dramatic decrease in potency was observed;
furthermore, when this high electron density group was depleted
through the introduction of the known deactivator eNO2 (at the
ortho- or para-position, as in the 1h and 1g derivatives, respec-
tively) or replaced by a cyclohexyl group (1e), the potency
increased by about 20-fold.
and extended
pep stacking with the more potent isoxazole de-
To test in vivo the effects of the compounds showing the highest
TRPM8-activating potencies in vitro (1a, 1b, 1c or 1d), we used the
acetone-induced cold allodynia model in CCI mice [30,31]. In this
test, evaporative cooling of locally applied acetone is used to evoke
nociceptive behavior; pharmacological [32] and genetic [14] evi-
dence suggest that the activation of TRPM8 channels plays a central
role in this in vivo model. Moreover, in addition to acetone, menthol
can trigger TRPM8-dependent allodynic responses in mice with CCI
[33]. In accordance with this, we observed that, similarly to
menthol, the newly-synthesized 1aed compounds, also potenti-
ated allodynic responses prompted by acetone exposure, suggest-
ing that they also acted as TRPM8 agonists. However, when
compared to menthol, 1aed showed a distinct time-dependent
profile. In fact, allodynic responses triggered by these compounds
showed a shorter latency, being already evident after the first
triggering stimulus was applied; in addition, 1ae1d compounds
ceased to elicit allodynic responses after the second or third
acetone application. By contrast, menthol was only effective after
rivatives. When considering these possible polar interactions, it
comes as no surprise that suitable ligands which occupy this second
region can afford markedly higher potencies compared to ligands
which are positioned in the first subcavity where at most they
stabilize H-bonds with Y745. Yet, the ability of molecules to interact
with residues of the first region appears to be an essential requisite
to achieve a good efficacy profile, regardless of the polar contacts
stabilized within the second subpocket. Notably, icilin confirms
that ligands able to suitably occupy both subpockets can success-
fully combine potency and efficacy even though such a double
occupation appears to be vastly constrained by hampering steric
clashes exerted by surrounding apolar residues.
Taken together, the results obtained emphasize that the newly
synthesized compounds (particularly 1aec) can induce TRPM8
activation in vivo. Although the TRPM8-dependent acetone test in
CCI animals utilized in our study is a model of cold allodynia
(indicative of an increased pain sensitivity), it should be reminded
that the activation of TRPM8 channels can also mediate analgesic

666
C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
effects [35]. As a matter of fact, menthol itself, though acting as a
reference compound for TRPM8 activators, has well-known anal-
gesic properties [36]. Concentration-dependent effects of menthol
at TRPM8 receptors might provide a plausible explanation for this
apparently paradoxical behavior; in fact, while, as indicated pre-
viously, elevated concentration of menthol trigger TRPM8-
dependent nociceptive behavior in CCI mice [14,33], lower
menthol concentrations induced marked analgesia in the same rat
model of neuropathic pain [9]. However, whether additional vari-
ables such as changes in TRPM8 kinetics of activation and/or
desensitization, differential treatment modalities (topical vs sys-
temic administration), or model-dependent effects on TRPM8
expression levels at various sites contribute to the distinct phar-
macological profile shown in vivo by menthol or other TRPM8
modulators remains to be investigated.
Despite all these uncertainties, the present results clearly sug-
gest that presently-described aminoisoxazole-based derivatives
display a pharmacological profile, both in vitro and in vivo, consis-
tent with that of TRPM8 agonists. Thus, we believe that these
molecules might be of considerable interest for further structural
optimization and functional analysis, in order to validate their
clinical utility as novel analgesics.
mobile phase. Final product was crystallized from n-hexane as a
white solid (1.0 g, yield 67%). M.p. ¼ 97e99 ꢀC; 1H NMR (CDCl3):
d
5.51 (s, 1H), 3.85 (bs, 1H, D2O exchangeable), 3.00 (t, 2H,
J ¼ 13.0 Hz), 2.31 (s, 3H), 1.95e1.85 (m, 1H, J ¼ 25.7 Hz), 0.97 (d, 6H,
J ¼ 6.6 Hz); 13C NMR (CDCl3):
d 168.5, 164.6, 92.9, 48.4, 26.7, 20.9,
12.4. MS (m/z): 155.61 (M þ H)þ.
4.2.3. N-(3-Methyl)butyl-N-(5-methyl-3-isoxazolyl)amine (1c)
The product was synthesized starting from 3-methylbutanal.
Crude product was purified using n-hexane/ethyl acetate (9.5/0.5) as
mobile phase and was isolated as a white solid after crystallization
from hexane (1.18 g, yield 69%). M.p. ¼ 87e89 ꢀC. 1H NMR (CDCl3):
d
5.52 (s, 1H), 3.68 (bs, 1H, D2O exchangeable), 3.25e3.19 (m, 2H,
J ¼ 15.0 Hz), 2.30 (s, 3H),1.62e1.58 (m, 1H, J ¼ 19.6 Hz), 1.50e1.45 (m,
2H, J ¼ 22.4 Hz), 0.95 (d, 6H, J ¼ 6.8 Hz); 13C NMR (CDCl3):
d 168.7,
166.1, 92.4, 48.2, 36.6, 26.2, 22.8, 12.9. MS (m/z): 155.73 (M þ H)þ.
4.2.4. N-Butyl-N-(5-methyl-3-isoxazolyl)amine (1d)
The product was synthesized starting from n-butanal. Crude
product was purified using n-hexane/ethyl acetate (9/1) as mobile
phase. Final product was isolated as a colorless oil that did not
crystallize (0.95 g yield 62%). 1H NMR (CDCl3):
d 5.50 (s, 1H), 3.74
(bs, 1H, D2O exchangeable), 3.21e3.18 (m, 2H, J ¼ 20.1 Hz), 2.28 (s,
4. Experimental protocols
3H), 1.63e1.53 (m, 2H, J ¼ 30.5 Hz), 1.37e1.29 (m, 2H, J ¼ 37.4 Hz),
0.96 (t, 3H, J ¼ 14.7 Hz); 13C NMR (CDCl3):
d 168.8, 165.3, 92.8, 44.0,
4.1. General
32.1, 22.0, 14.0, 12.7. MS (m/z): 169.91 (M þ H)þ.
L
-Menthol, as well all other chemicals and solvents were fur-
4.2.5. N-Cyclohexylmethyl-N-(5-methyl-3-isoxazolyl)amine (1e)
The product was synthesized starting from cyclo-
hexanecarbaldehyde. Crude product was purified using n-hexane/
ethyl acetate (9.8/0.2) as mobile phase. Final product was obtained
nished by SigmaeAldrich (Milan, Italy). Course of reactions and
purity of finished product was estimated by TLC, using precoated
silica gel glass plates (Macherey Nagel, Duren, Germany). Prepara-
tive separations were performed in glass columns packed with
silica gel (Macherey Nagel; Ø 0.063:0.200 mm). All other solvents
and reagents were of analytical grade. Melting points were deter-
mined using a Büchi apparatus B 540 and are uncorrected. 1H NMR
and 13C NMR were recorded using a Varian Mercury 400 instru-
ment (Varian inc., Palo Alto, USA). Chemical shifts are reported in
as colorless oil (1.36 g, yield 70%). 1H NMR (CDCl3):
d 5.50 (s, 1H),
3.99 (bs, 1H, D2O exchangeable), 3.05 (t, 2H, J ¼ 13.4 Hz), 2.27 (s,
3H), 1.75e1.65 (m, 5H), 1.26e1.12 (m, 4H, J ¼ 54 Hz), 0.93e0.84 (m,
2H, J ¼ 34 Hz); 13C NMR (CDCl3):
d 168.3, 166.5, 92.7, 57.6, 36.6, 31.1,
26.7, 26.0, 12.8. MS (m/z): 195.95 (M þ H)þ.
d
units (ppm) relative to tetramethylsilane used as internal stan-
4.2.6. N-Benzyl-N-(5-methyl-3-isoxazolyl)amine (1f)
dard. The following symbols were used to describe the NMR peaks:
The product was synthesized starting from benzaldehyde. Crude
product was purified using n-hexane/ethyl acetate (9/1) as mobile
phase. Final product was obtained as a pale yellow solid after
crystallization in diethyl ether (1.39 g, yield 73%). Spectral data and
melting point were in accordance with literature [37]. MS (m/z):
189.40 (M þ H)þ.
s ¼ singlet, bs ¼ broad singlet, d ¼ doublet, dd ¼ double doublet,
pd
¼
pseudo-doublet,
t
¼
triplet, pt
¼
pseudo-triplet,
m ¼ multiplet. Mass spectra were recorded using an API 2000
spectrometer (Applied Biosystem, Monza, Italy). Elemental analysis
was carried on using a 2400 PerkineElmer CHN Analyzer (Perkin
Elmer Italia, Monza, Italy). Results obtained were within ꢁ0.4% of
theoretical values. Derivatives belonging to series 1 were synthe-
sized starting from 3-amino-5-methyl isoxazole, while derivatives
belonging to series 2 and 3 were synthesized from 5-amino-3-
methyl isoxazole and 3-aminoisoxazole respectively.
4.2.7. N-(40-Nitro)benzyl-N-(5-methyl-3-isoxazolyl)amine (1g)
The product was synthesized starting from 4-nitrobenzaldehyde.
Crude product was purified using n-hexane/ethyl acetate (7/3) as
mobile phase. Final product was obtained as a yellowish solid after
crystallization in diethyl ether (1.62 g, yield 69%). M.p. ¼108e109 ꢀC.
4.2. General procedure for the synthesis of isoxazolylamine
derivatives
1H NMR (CDCl3):
d
8.22 (pd, 2H, J ¼ 8.3 Hz), 7.56 (pd, 2H, J ¼ 8.1 Hz),
5.53 (s,1H), 4.55 (d, 2H, J ¼ 6.0 Hz), 4.32 (bs, 1H, D2O exchangeable),
2.32 (s, 3H); 13C NMR (CDCl3):
d 169.5,164.4,147.4,146.8,128.2,124.1,
4.2.1. N-(Methyl)ethyl-N-(5-methyl-3-isoxazolyl)amine (1a)
Compound 1a was obtained starting from acetone. Crude
product was purified using n-hexane/ethyl acetate (9/1) as eluent.
Final product was crystallized from n-hexane as a white solid (1.0 g,
93.2, 47.5, 12.8. MS (m/z): 234.72 (M þ H)þ.
4.2.8. N-(20-Nitro)benzyl-N-(5-methyl-3-isoxazolyl)amine (1h)
The product was synthesized starting from 2-nitrobenzaldehyde.
Crude product was purified using n-hexane/ethyl acetate (7/3) as
mobile phase. Final product was obtained as a gray solid after
crystallization in diethyl ether (1.5 g, yield 63%). M.p. ¼ 98e99 ꢀC.1H
yield 73%). M.p. ¼ 93e95 ꢀC. 1H NMR (CDCl3):
d 5.49 (s, 1H), 3.73e
3.63 (m, 1H, J ¼ 25.0 Hz), 3.25 (bs, 1H, D2O exchangeable), 2.30 (s,
3H), 1.23 (d, 6H, J ¼ 6.2 Hz); 13C NMR (CDCl3):
45.5, 23.1, 12.5. MS (m/z): 140.95 (M þ H)þ.
d 168.4, 164.4, 93.2,
NMR (CDCl3):
d
8.10 (pd,1H, J ¼ 8.1 Hz), 7.77 (pd,1H, J ¼ 7.3 Hz), 7.64
(pt, 2H, J ¼ 15.0 Hz), 7.47 (pt, 2H, J ¼ 14.8 Hz); 5.51 (s,1H), 4.73 (d, 2H,
4.2.2. N-(2-Methyl)propyl-N-(5-methyl-3-isoxazolyl)amine (1b)
The product was synthesized starting from 2-methylpropanal.
Crude product was purified using n-hexane/ethyl acetate (9/1) as
J ¼ 6.6 Hz), 4.65 (bs, 1H, D2O exchangeable), 2.30 (s, 3H); 13C NMR
(CDCl3):
d 169.2, 164.4, 148.5, 134.7, 134.1, 132.0, 128.7, 125.4, 93.5,
45.7, 12.7. MS (m/z): 235.01 (M þ H)þ.

C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
667
4.2.9. N-(2-Methyl)propyl-N-(3-methyl-5-isoxazolyl)amine (2b)
Derivative 2b was synthesized as previously described for 1b.
Final product was crystallized from petroleum ether as a gray solid
annealing at 60 ꢀC for 1 min, and extension at 72 ꢀC for 1 min. The
expression of both mouse and rat TRPM8 mRNAs, both presumably
expressed in F11 cells, was tested by using specific primers adapted
on the known mouse TRPM8 specific primers (sense: 50-ACA-
GACGTGTCCTACAGTGA-30; reverse: 50-GCTCTGGGCATAACCA-
CACTT-30; http://pga.mgh.harvard.edu/primerbank/). In detail,
while the TRPM8 sense primer sequence was complementary with
the TRPM8 mRNAs of both species, in order to obtain a reverse
primer sequence complementary with rat TRPM8 mRNA, it was
necessary to remove the last nucleotide from the above mentioned
sequence. GAPDH expression was used as internal control. PCR
products were analyzed by gel electrophoresis and visualized by
using ethidium bromide staining.
Quantitative real-time PCR was carried out in an Eppendorf
MastercyclerÒeprealplex Thermal Cyclers with the TRPM8 primers
by using SYBR Green detection. Samples were amplified simulta-
neously in triplicate in one-assay run, and the ct (threshold cycle)
value for each experimental group was determined. Data normal-
ization was performed by using the ct for the amplification of
(1.1 g yield 72%). M.p. ¼ 110e112 ꢀC. 1H NMR (CDCl3):
d 4.82 (s, 1H),
4.62 (bs, 1H, D2O exchangeable), 2.97 (t, 2H, J ¼ 13.0 Hz), 2.17 (s,
3H), 1.91e1.86 (m, 1H, J ¼ 25.8), 0.97 (d, 6H, J ¼ 6.5 Hz); 13C NMR
(CDCl3):
d 170.3, 161.6, 77.8, 52.4, 28.5, 20.2, 12.0. MS (m/z): 155.41
(M þ H)þ.
4.2.10. N-(Butyl)-N-(3-methyl-5-isoxazolyl)amine (2d)
The product was synthesized according to the procedure
adopted for 1d. Final Product was crystallized from petroleum ether
as a gray solid (1.0 g) yield 64%. M.p. ¼ 84e86 ꢀC. 1H NMR (CDCl3):
d
4.85 (s, 1H), 4.62 (bs, 1H, D2O exchangeable), 3.17e3.14 (m, 2H,
J ¼ 19.9 Hz), 2.19 (s, 3H), 1.52e1.41 (m, 2H, J ¼ 30.8 Hz), 1.38e1.32
(m, 2H, J ¼ 37.5 Hz), 0.95 (t, 3H, J ¼ 14.7 Hz); 13C NMR (CDCl3):
d
170.1, 162.8, 77.9, 43.2, 31.3, 19.9, 13.4, 12.1. MS (m/z): 170.01
(M þ H)þ.
4.2.11. N-(2-Methyl)propyl-N-(3-isoxazolyl)amine (3b)
GAPDH mRNA. Differences in mRNA content between groups were
DDct
The product was synthesized according to the procedure
adopted for 1b. Purification was carried out using n-hexane/ethyl
acetate (6/4) as mobile phase. Final product was crystallized from
hexane and was isolated as a white solid (1.1 g yield 74%).
calculated as normalized values by using the 2ꢂ
formula [39].
4.3.3. Calcium imaging experiments
Differentiated F11 cells plated on glass coverslips were incu-
bated with 3 M Fura-2 acetoxymethylesther (Fura-2 AM) for 1 h at
M.p. ¼ 46e48 ꢀC; 1H NMR (CDCl3):
d
8.00 (pd, 1H, J ¼ 1.6 Hz), 5.84
m
(pd, 1H, J ¼ 1.6 Hz), 3.94 (bs, 1H, D2O exchangeable), 3.05 (t, 2H,
J ¼ 13.1 Hz), 1.93e1.85 (m, 1H, J ¼ 25.4 Hz), 0.95 (d, 6H, J ¼ 6.6 Hz);
room temperature in darkness in a standard Normal Kreb’s solu-
tion, containing (in mM): NaCl 160, KCl 5.5, CaCl2 1.5, MgSO4 1.2,
HEPES 10, glucose 10, pH 7.4 adjusted with NaOH. Fluorescence
images, taken at both 340 nm and 380 nm excitation wavelengths,
were obtained by means of a monochromator-based imaging sys-
tem consisting of a DeltaRAMXÔ Microscope Illuminator (Optical
Building Blocks Corporation, Birmingham, NJ) and a coolSNAP ES
camera (Princeton Instruments, Trenton, NJ) connected to an
inverted microscope (DM IRB, Leica, Wetzlar, Germany) equipped
with an oil immersion objective (HCX PL APO, 40ꢃ/1.25, Leica).
After correction for background fluorescence, the ratio of the
fluorescence intensity of images taken at both 340 nm and 380 nm
excitation wavelengths every 3 s was calculated. Calibrated ratios
were displayed online using MetaFluor Imaging System software
(Molecular Devices Corporation, Downingtown, PA).
13C NMR (CDCl3):
d 164.2, 157.8, 95.8, 51.9, 28.4, 20.1. MS (m/z):
141.55 (M þ H)þ.
4.2.12. N-Butyl-N-(3-isoxazolyl)amine (3d)
The product was synthesized according to the procedure
adopted for 1d. Mobile phase consisted in n-hexane/ethyl acetate
(8/2). Final product was isolated as pale yellow oil (1.0 g, yield 70%).
1H NMR (CDCl3):
d
8.07 (pd, 1H, J ¼ 1.5), 5.83 (pd, 1H, J ¼ 1.5), 3.90
(bs, 1H, D2O exchangeable), 3.25e3.21 (m, 2H, J ¼ 20.0 Hz), 1.69e
1.58 (m, 2H, J ¼ 30.7 Hz), 1.47e1.39 (m, 2H, J ¼ 37.6), 0.96 (t, 3H,
J ¼ 14.7 Hz); 13C NMR (CDCl3):
d 164.2, 157.7, 96.0, 43.8, 31.0, 19.9,
13.7. MS (m/z): 142.41 (M þ H)þ.
4.3. Pharmacology
4.3.4. In vivo tests
4.3.1. Cell cultures
F11 cells were grown and differentiated as recently reported
[38]. Briefly, these cells were grown in DMEM medium supple-
Male Swiss CD1 mice weighing 30e35 g were purchased from
Harlan (Udine, Italy). They were housed in cages in a room kept at
22 ꢁ 1 ꢀC and with 12/12-h light/dark cycle. The animals were
acclimated to their environment for 1 week, and had ad libitum
access to tap water and standard rodent chow. Animal care was in
compliance with Italian (D.M. 116192) and European Economic
Community regulations (O.J. of E.C. L 358/112/18/1986) for animals
used for scientific purposes. Sciatic nerve ligation was performed
following the method of Bennett and Xie [40] modified for mice
[41]. Briefly, mice were first anesthetized with xylazine (10 mg/kg
i.p.) and ketamine (100 mg/kg i.p.), the left thigh was shaved and
scrubbed with Betadine, and a small incision (2 cm in length) was
made in the middle of the left thigh to expose the sciatic nerve. The
nerve was loosely ligated at two distinct sites (spaced at a 2-mm
interval) around the entire diameter of the nerve using silk su-
tures (7e0). The surgical area was closed and finally scrubbed with
Betadine. The animals were placed under a heat lamp until they
mented with 10% FBS, 1% penicillin/streptomycin (10 U/ml) and 1% L-
glutamine (200 mM) (referred as GM). The cells were kept in
a humidified atmosphere at 37 ꢀC with 5% CO2 in 100-mm plastic
Petri dishes F11 cell differentiation was achieved upon at least 72 h
cell exposure to a differentiation medium (referred as DM) con-
taining a lower FBS concentration (2%) and 10 mM retinoic acid. For
calcium-imaging experiments, F11 cells were plated on glass cov-
erslips (Carolina Biological Supply Co., Burlington, NC) coated with
poly-L-lysine (SIGMA, Milan, Italy); 24 h after plating, F11 cells were
exposed to the DM.
4.3.2. PCR analyses
Total RNA from undifferentiated or differentiated F11 cells using
TRI-Reagent (SigmaeAldrich, Milan, Italy) was treated with DNAse
1 U/
m
l for 15 min at room temperature and quantified by spectro-
awakened.
dissolved in absolute ethanol at a final concentration of 20% w/v. 14
days after Chronic Constriction Injury (CCI) 50 l of solutions con-
taining these compounds were applied on operated paw at each
experimental group (n ¼ 6 for each experimental group). Control
group (V) was only treated with absolute ethanol. After 20 min
L
-menthol (M), as well as compounds 1aed and 1f, were
photometry (260 nm/280 nm ratio >1.7). Thereafter, 5
mg of the
isolated RNA was used as template for the synthesis of cDNA by
m
reverse transcription at 37 ꢀC for 2 h. PCR reactions were carried out
on an AmpliTaq Gold 0.25 U/
ml (Applied Biosystem, Monza, Italy)
using the following parameters: denaturation at 95 ꢀC for 1 min,

668
C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
from topical application of tested drugs, a drop (25
ml) of acetone
References
was applied to the dorsal surface of the ligated paw with a syringe
connected to a thin polyethylene tube while the mice were
standing on a metal mesh, in order to evaluate cold allodynia. In
particular, a brisk foot withdrawal response was considered as a
sign of cold allodynia and express as cold responses. Three mea-
sures of 2 min every approximately 5 min were done. The results
measured in each group were expressed both as full time, when the
values obtained in all the experimental procedure were measured,
and as time-course, when the values obtained during the first, the
second and the third phase of each experiment were separately
expressed.
[1] C. Sprengell, Hippocrates: the Aphorisms of Hippocrates, and the Sentences of
Celsus; with Explanations and References to the Most Considerable Writers to
Which Are Added, Aphorisms upon Several Distempers, Not Well Distin-
guished by the Ancients, second ed., Sine Nomine, London, 1735.
[2] R.E. Siegel, Galen on Sense Perception: His Doctrines, Observations and Ex-
periments on Vision, Hearing, Smell, Taste, Touch and Pain, and Their His-
torical Sources, Karger, Basel, 1970.
[3] G. Bini, G. Cruccu, K.E. Hagbarth, W. Schady, E. Torebjork, Analgesic effect of
vibration and cooling on pain induced by intraneural electrical stimulation,
Pain 18 (1984) 239e248.
[4] J. Sauls, Efficacy of cold for pain: fact or fallacy? Online J. Knowl. Synth. Nurs. 6
(1999) 103e111.
[5] D.D. McKemy, How cold is it? TRPM8 and TRPA1 in the molecular logic of cold
sensation, Mol. Pain 1 (2005) 16e22.
[6] A.M. Peier, A. Moqrich, A.C. Hergarden, A.J. Reeve, D.A. Andersson, G.M. Story,
T.J. Earley, I. Dragoni, P. McIntyre, S. Bevan, A. Patapoutian, A Trp channel that
senses cold stimuli and menthol, Cell 108 (5) (2002) 705e715.
[7] D.D. McKemy, W.M. Neuhauser, D. Julius, Identification of a cold receptor
reveals a general role for TRP channels in thermosensation, Nature 416 (6876)
(2002) 52e58.
[8] G. Reid, A. Babes, F. Pluteanu, A cold- and menthol-activated current in rat
dorsal root ganglion neurones: properties and role in cold transduction,
J. Physiol. 545 (2002) 595e614.
[9] C.J. Proudfoot, E.M. Garry, D.F. Cottrell, R. Rosie, H. Anderson, D.C. Robertson,
S.M. Fleetwood-Walker, R. Mitchell, Analgesia mediated by the TRPM8 cold
receptor in chronic neuropathic pain, Curr. Biol. 16 (16) (2006) 1591e1605.
[10] G. Wasner, J. Schattschneider, A. Binder, R. Baron, Topical menthol e a human
model for cold pain by activation and sensitization of C nociceptors, Brain 127
(2004) 1159e1171.
[11] B. Namer, F. Seifert, H.O. Handwerker, C. Maihöfner, TRPA1 and TRPM8 acti-
vation in humans: effects of cinnamaldehyde and menthol, Neuroreport 16
(9) (2005) 955e959.
4.4. Computational methods
The ligands were simulated considering their neutral forms
since they are the favored ones at the physiological pH. After a
preliminary optimization using the PM6 semi-empirical method, as
implemented in MOPAC 2009, to discard high energy geometries
and to calculate precise atomic charges, the conformational
behavior of the compounds was investigated using VEGAZZ by a
clustered Monte Carlo procedure which generated 1000 con-
formers by randomly rotating the rotors. All geometries obtained
were stored and optimized to avoid high-energy rotamers. For each
ligand, the lowest energy structure was then exploited in the
docking simulations performed by AutoDock 4.0 using the TRPM8
model recently obtained by homology techniques [23]. In detail, the
calculations involved a single TRPM8 monomer and the grid box
[12] D.D. McKemy, Therapeutic potential of TRPM8 modulators, Open Access Drug
Discovery J. 2 (2010) 80e87.
[13] A. Dhaka, A.N. Murray, J. Mathur, T.J. Earley, M.J. Petrus, A. Patapoutian, TRPM8
is required for cold sensation in mice, Neuron 54 (3) (2007) 371e378.
[14] R.W. Colburn, M.L. Lubin, D.J. Stone Jr., Y. Wang, D. Lawrence, M.R. D’Andrea,
M.R. Brandt, Y. Liu, C.M. Flores, N. Qin, Attenuated cold sensitivity in TRPM8
null mice, Neuron 54 (3) (2007) 379e386.
ꢀ
was set to include all residues within a 15 A radius sphere around
Y745, whose role in ligand recognition has been confirmed by
mutational studies [32], in order to comprise the entire binding
cavity. The resolution of the grid was 68 ꢃ 77 ꢃ 100 points with a
[15] L. Zhang, G.J. Barritt, Evidence that TRPM8 is an androgen-dependent Ca2þ
channel required for the survival of prostate cancer cells, Cancer Res. 64
(2004) 8365e8373.
ꢀ
grid spacing of 0.450 A. For docking simulations, the flexible bonds
of the ligand were left free to rotate to account for ligand flexibility
within the binding cavity. Each substrate was docked by using the
Lamarckian algorithm as implemented in AutoDock. The genetic-
based algorithm ran 30 simulations per substrate with 2,000,000
energy evaluations and up to 27,000 generations. The crossover rate
was increased to 0.8, and the number of individuals in each pop-
ulation to 150. All other parameters were left at the AutoDock
default settings. The best complexes were finally minimized to
favor the mutual adaptability between ligand and receptor and the
optimized complexes were then used to re-calculate AutoDock
docking scores, VEGA energy scores and X Score values. All
mentioned minimizations were performed using the conjugated
gradients algorithm as implemented in the Namd 2.51 package
with the force-field CHARMm v22 and Gasteiger’s atomic charges.
[16] A.J. Foster, C.P.E. Van Der Logt, W. Erwin, Tetrahydropyrimidine-2-one de-
rivatives and their uses, Unilever plc, WO 2004026840, 2004.
[17] A. Bassoli, G. Borgonovo, G. Busnelli, G. Morini, Synthesis of a new family of N-
aryl lactams active on chemesthesis and taste, Eur. J. Org. Chem. 7 (2006)
1656e1663.
[18] R.R. Calvo, S.K. Meegalla, M.R. Player, Benzimidazole derivatives useful as
TRPM8 receptor modulators, Jansennpharmaceutica, N.V., WO 2010132247,
2010.
[19] N.R. IrlapatiI, A. Thomas, D.K. Kurhe, S.Y. Shelke, N. Khairatkar Joshi, S. Vis-
wanadha, I. Mukhopadhyay, Fused oxazole and thiazole derivatives as TRPMs
modulators, Glen-Mark Pharmaceuticals, SA, WO 2010010435, 2010.
[20] N. Murugesan, Z. Gu, S. Spergel, M. Young, P. Chen, A. Mathur, L. Leith,
M. Hermsmeier, E.C. Liu, R. Zhang, E. Bird, T. Waldron, A. Marino, B. Koplowitz,
W.G. Humphreys, S. Chong, R.A. Morrison, M.L. Webb, S. Moreland,
N. Trippodo, J.C. Barrish, Biphenylsulfonamide endothelin receptor antago-
nists. 4. Discovery of N-[[20-[[(4,5-dimethyl-3-isoxazolyl)amino]sulfonyl]-4-
(2-oxazolyl)[1,10-biphenyl]-2-yl]methyl]-N,3,3-trimethylbutanamide (BMS-
207940), a highly potent and orally active ET(A) selective antagonist, J. Med.
Chem. 46 (1) (2003) 125e137.
[21] A. Amano, Cooling and pungent agents, J. Jap. Soc. Cut. Health 17 (1984) 77e
86.
4.5. Statistical analysis
[22] H.R. Watson, R. Hems, D.G. Rosweel, D.J. Spring, New compounds with the
menthol cooling effect, J. Soc. Cosmet. Chem. 9 (1978) 185e200.
[23] A. Pedretti, C. Marconi, I. Bettinelli, G. Vistoli, Comparative modeling of the
quaternary structure for the human TRPM8 channel and analysis of its binding
features, Biochim. Biophys. Acta 1788 (5) (2009) 973e982.
[24] D. Platika, M.H. Boulos, L. Baizer, M.C. Fishman, Neuronal traits of clonal cell
lines derived by fusion of dorsal root ganglia neurons with neuroblastoma
cells, Proc. Natl. Acad. Sci. U. S. A. 82 (10) (1985) 3499e3503.
[25] S.F. Fan, K.F. Shen, M.A. Scheideler, S.M. Crain, F11 neuroblastoma x DRG
neuron hybrid cells express inhibitory mu- and delta-opioid receptors which
increase voltage-dependent Kþ currents upon activation, Brain Res. 590
(1992) 329e333.
[26] H.H. Chuang, W.M. Neuhausser, D. Julius, The super-cooling agent icilin re-
veals a mechanism of coincidence detection by a temperature-sensitive TRP
channel, Neuron 43 (6) (2004) 859e869.
[27] H.J. Behrendt, T. Germann, C. Gillen, H. Hatt, R. Jostock, Characterization of the
mouse cold-menthol receptor TRPM8 and vanilloid receptor type-1 VR1 using
a fluorometric imaging plate reader (FLIPR) assay, Br. J. Pharmacol. 141 (4)
(2004) 737e745.
Data are expressed as the mean ꢁ SEM. The analysis of the data
obtained from in vitro experiments was performed using Graph-
Pad Prism (GraphPad Software Inc., San Diego, CA). Statistically
significant differences between two sets of data were evaluated
with the Student’s t test (p <0.05). Statistical comparisons between
multiple (>2) experimental groups was performed by use of the
one-way analysis of variance (ANOVA), followed by Newmane
Keuls test. Values with a p <0.05 was considered statistically
significant.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.08.056.

C. Ostacolo et al. / European Journal of Medicinal Chemistry 69 (2013) 659e669
669
[28] R. Madrid, T. Donovan-Rodríguez, V. Meseguer, M.C. Acosta, C. Belmonte,
F. Viana, Contribution of TRPM8 channels to cold transduction in primary
sensory neurons and peripheral nerve terminals, J. Neurosci. 26 (48) (2006)
12512e12525.
[29] A. Pedretti, M. Labozzetta, A.R. Lo Monte, A. Beccari, A. Moriconi, G. Vistoli,
Exploring the activation mechanism of TRPM8 channel by targeted MD sim-
ulations, Biochem. Biophys. Res. Commun. 414 (1) (2011) 14e19.
[30] J.S. Walczak, P. Beaulieu, Comparison of three models of neuropathic pain in
mice using a new method to assess cold allodynia: the double plate technique,
Neurosci. Lett. 399 (3) (2006) 240e244.
[35] S. Mandadi, B.D. Roufogalis, ThermoTRP channels in nociceptors: taking a lead
from capsaicin receptor TRPV1, Curr. Neuropharmacol. 6 (1) (2008) 21e38.
[36] R. Pan, Y. Tian, R. Gao, H. Li, X. Zhao, J.E. Barrett, H. Hu, Central mechanisms of
menthol-induced analgesia, J. Pharmacol. Exp. Ther. 343 (3) (2012) 661e672.
[37] R. Eligeti, S.R. ReddyAtthunuri, R. Samala, F.P. Shaik, G. ReddyKundur, A fast
and highly efficient protocol for reductive amination of aromatic aldehydes
using NaBH4 and isoxazoleamines in a ionic liquid medium, Chin. J. Chem. 29
(4) (2011) 769e772.
[38] P. Ambrosino, M.V. Soldovieri, C. Russo, M. Taglialatela, Activation and
desensitization of TRPV1 channels in sensory neurons by the PPAR
a agonist
[31] K. Vissers, T. Meert, A behavioral and pharmacological validation of the
acetone spray test in gerbils with a chronic constriction injury, Anesth. Analg.
101 (2) (2005) 457e464.
[32] H. Xing, M. Chen, J. Ling, W. Tan, J.G. Gu, TRPM8 mechanism of cold allodynia
after chronic nerve injury, J. Neurosci. 27 (50) (2007) 13680e13690.
[33] O. Caspani, S. Zurborg, D. Labuz, P.A. Heppenstall, The contribution of TRPM8
and TRPA1 channels to cold allodynia and neuropathic pain, PLoS One 4 (10)
(2009) e7383.
[34] M. Bandell, A.E. Dubin, M.J. Petrus, A. Orth, J. Mathur, S.W. Hwang,
A. Patapoutian, High-throughput random mutagenesis screen reveals TRPM8
residues specifically required for activation by menthol, Nat. Neurosci. 9 (4)
(2006) 493e500.
palmitoylethanolamide, Br. J. Pharmacol. 168 (6) (2013) 1430e1444.
[39] F.A. Iannotti, E. Panza, V. Barrese, D. Viggiano, M.V. Soldovieri, M. Taglialatela,
Expression, localization, and pharmacological role of Kv7 potassium channels
in skeletal muscle proliferation, differentiation, and survival after myotoxic
insults, J. Pharmacol. Exp. Ther. 332 (3) (2010) 811e820.
[40] G.J. Bennett, Y.K. Xie, A peripheral mononeuropathy in rat that produces
disorders of pain sensation like those seen in man, Pain 33 (1) (2008) 87e107.
[41] R. Russo, J. LoVerme, G. La Rana, T.R. Crompton, J. Parrott, A. Duranti,
A. Tontini, M. Mor, G. Tarzia, A. Calignano, D. Piomelli, The fatty acid amide
hydrolase inhibitor URB597 (cyclohexylcarbamic acid 30-carbamoylbiphenyl-
3-yl ester) reduces neuropathic pain after oral administration in mice,
J. Pharmacol. Exp. Ther. 322 (1) (2007) 236e242.