Background
TRPM8 is a non-selective cation channel of the TRP family that is activated by mild cold temperatures and cooling compounds such as menthol, eucalyptol and icilin [
1,
2]. Like several other TRP channels, TRPM8 is also gated by voltage [
3‐
6]. The voltage dependence of TRPM8 is characterised by a strong outward rectification at depolarized transmembrane potentials, and a rapid and potential-dependent closure at negative membrane potentials. Cooling and menthol application shift the activation curve of TRPM8 towards more negative potentials, thus increasing the probability of channel openings, boosting inward currents at physiological membrane potentials [
6]. Endogenous factors such as phospholipase A
2 products [
7,
8], endocannabinoids [
9] and PIP
2 [
10‐
12] also participate in channel regulation.
TRPM8 is expressed in a subset of small diameter primary sensory neurons and their peripheral terminals [
13,
1,
2]. In addition to its well characterized and critical role in the activation of low threshold thermoreceptors, responsible for the sensations of innocuous cold [
14‐
16], other evidence indicates the possible involvement of TRPM8 channels in normal noxious cold sensations and cold allodynia [reviewed by [
17]]. Notably, in an animal model of neuropathic pain, cold allodynia is significantly attenuated by capsazepine, a TRPM8 blocker [
18], and mice lacking TRPM8 show reduced responses in nerve injury induced models of cold-allodynia [
15,
16]. Also, sensory fibers with high threshold cold-evoked responses are difficult to record from in these mice [
14]. Moreover, TRPM8-positive fibers are prominent in peripheral territories with marked noxious responses to cold [
19]. Many neurons responding to TRPM8 agonists are also activated by capsaicin, a marker of nociceptors [
20‐
22]. These new findings stress the potential use of TRPM8 modulators in the therapeutic management of cold-evoked pain, a characteristic symptom in some patients with neuropathic pain [
23].
Despite its fundamental role in many aspects of cold temperature transduction in mammals, the pharmacology of TRPM8 is still largely unexplored. Only a few studies have so far been dedicated to TRPM8 channel inhibitors and their mechanisms of action [
24‐
30]. Recently, Malkia et al. [
26] showed that several antagonist compounds, including BCTC and SKF96365, act as negative allosteric modulators of channel gating, shifting the voltage activation of TRPM8 towards more positive potentials, suppressing the depolarizing effects of temperature and chemical agonists [
25,
26]. SKF96365 is a non-specific blocker of various calcium-permeable channels, including receptor-operated channels [
31]. BCTC was originally introduced as a highly potent and specific antagonist of the heat-activated vanilloid receptor TRPV1 [
32]. However, later studies showed that it also inhibits the TRPM8 channel, as does another TRPV1 blocker, capsazepine [
24,
26,
29]. These two antagonists bind competitively at the vanilloid binding pocket of the TRPV1 channel, governed by residues in the intracellular parts of the putative transmembrane domains 2, 3 and 4 [
33,
34].
Recently, during a massive random mutagenesis screen, tyrosine 745, located in the middle of putative transmembrane segment 2, was identified as a crucial residue for the menthol sensitivity of mouse TRPM8 [
35]. The generated TRPM8-Y745H mutant channel was insensitive to menthol, but retained the responsiveness to cold and voltage exhibited by the wild-type channel. Because of the significant parallels between TRPM8 and TRPV1 pharmacology [
24,
29], we decided to study the effect of the Y745H mutation on the activity of the best characterized TRPM8 antagonists: BCTC, capsazepine, SKF96365 and clotrimazole, as well as of two new inhibitor candidates: econazole and imidazole. We identify, for the first time, structural elements on the TRPM8 protein that are critical for channel antagonism, and demonstrate an important difference in the way antagonists interact with the menthol binding site of TRPM8.
Discussion
In light of the potential involvement of TRPM8 channels in the pathophysiology of cold nociception and cold allodynia [reviewed by [
17]], there is a strong interest in the pursuit of novel modulators of TRPM8 channels. In this study, we investigated the effect of different compounds at the TRPM8-Y745H mutant channel, focusing on the differential effects of several antagonists on the gating properties.
Cold and menthol activate wild-type TRPM8 by shifting its voltage-dependent activation towards more negative potentials [
3,
6]. We confirmed a previous report showing that the tyrosine at position 745, on the putative S2 transmembrane segment, is essential for the agonist effects of menthol [
35]. Menthol is a small and fairly lipophilic compound which easily partitions in the lipid bilayer and could thus be expected to affect TRPM8 channel function from various interaction sites. However, during menthol application to the menthol-insensitive TRPM8-Y745H mutant, the parameters describing voltage gating: V
1/2 and g/g
ctrl underwent no changes whatsoever, suggesting that menthol exerts its full effect by specific interaction with the binding site governed by the tyrosine residue 745. In the absence of menthol, the apparent gating charge, z
app, was slightly lower in the TRPM8-Y745H mutant, indicating that this site is to some extent communicated with the voltage sensor of the channel. This finding is supported by recent work on TRPM8 voltage sensor mutants [
39] and a fragmental model of the channel structure [
38].
While TRPM8 and TRPV1 share only around 20% protein sequence homology [
2], the channels have several phenotypic characteristics in common. Both are activated by temperature changes, and C-terminal chimeras between the channels exhibit reversed temperature sensitivity [
40]. Furthermore, both TRPM8 and TRPV1 are inhibited by compounds such as BCTC, CTPC, SB-452533, capsazepine and ruthenium red [
24,
29]. All TRPV1 antagonists, with the exception of pore blockers such as ruthenium red, seem to bind at the vanilloid binding pocket [
41]. However, no reports exist on the binding site of TRPM8 antagonists.
In this study, we demonstrate the variable inhibitory effect of several compounds at the TRPM8-Y745H mutant channel. This finding is unexpected since all of them affect gating of the channel in a similar fashion, shifting voltage-dependence to more positive potentials and exerting an allosteric negative modulation of the channel during cold- or chemically evoked activation [
26]. Our results reveal that the tyrosine residue 745 at the menthol binding site is critical for inhibition mediated by SKF96365. In contrast, the inhibition by other antagonists is unaffected (BCTC) or only partially reduced (capsazepine, clotrimazole, econazole) by the mutation, suggesting that at least one other binding site exists on the TRPM8 channel from where the inhibitors exert their negative allosteric modulation. Furthermore, the results imply that not all TRPM8 blockers should be considered competitive antagonists of menthol, even though the allosteric inhibitory effect they mediate lies in competition with the activating effect of cooling agents [
26].
The experimental observations of differential interaction of Y745 with the antagonists SKF96365 and BCTC were further confirmed by molecular docking studies. Interestingly, in these simulations SKF96365 exhibited strong interactions with both Y745 and an asparagine residue, N799, that also interacts with icilin [
42]. Looking at Figure
8C, SKF96365 can be hypothesized to lock the S2 and S3 domains into a fixed position, thereby preventing conformational shifts of the S4 domain induced by menthol (or cooling) that would lead to channel activation. This idea is further supported by our finding that SKF96365 inhibits the TRPM8-wt current at 33°C, a condition in which the channel is only activated by voltage, whereas no effect is seen at the Y745H mutant. BCTC, in contrast, blocks both the wild-type and mutant channels in this condition with similar potency, indicating the presence of an alternative binding site. Overall, our data suggest a sequential model of TRPM8 gating where chemical modulators can favour (e.g. menthol) or hinder (e.g. BCTC or SKF96365) the energetics of subsequent channel opening by cold temperature or voltage, from different binding sites.
Our results provide direct clues for the identification of different structural motifs required for antagonist binding, aiding in the design of new candidate molecules for specific inhibition of TRPM8 channels. They also provide tools for current efforts to resolve the crystal structure of TRP channels [
43‐
45]. Further mutagenesis work is required to identify the remaining binding site(s) of the various TRPM8 blockers. One potential site is formed by residues in the S2-S3 linker region, known to be important for the sensitivity to icilin [
42]. Another plausible location is centred on residues within the TRP domain. This domain is known to be important in the energetics of channel opening, i.e. translating drug binding into channel opening [
35,
46,
47]. In this regard, we consider a very powerful approach to subject random generated libraries of TRPM8 mutant channels [
35] to the inhibitory actions of the different antagonists characterized in this study. In this way, one could obtain unbiased structural information on the action of different inhibitors.
Finally, we also studied for the first time the inhibitory capacity of two members of the imidazole family: the parent compound itself, and the antimycotic agent econazole, at the cooling-activated TRPM8 channel. Imidazole was unable to inhibit neither of the two TRPM8 constructs, while econazole, similarly to its relative clotrimazole, proved to be a potent antagonist of the wild-type channel. Both econazole and clotrimazole lost potency at the Y745H mutant channel. This identification of a novel TRPM8 antagonist prompts further screening of imidazole-based compounds in the quest for new TRPM8 blockers, and offers indications for the design of more potent derivatives.
Methods
Generation of point-mutant TRPM8-Y745H
The full length cDNA encoding mouse TRPM8 (NM_134252) in pcDNA5 (Invitrogen) was kindly provided by Dr. Ardem Patapoutian, (Scripps Research Institute, La Jolla, CA). The TRPM8-Y745H mutant was obtained by site-directed mutagenesis with the following primers: (for) 5'-CGTGGTCTTCCACATCGCC-3'; (rev) 5'-GGCGATGTGGAAGACCACG-3', and Pfu Turbo Taq DNA Polymerase (Roche).
Cell culture and transfection
HEK293 cells were cultured in DMEM containing 10% of foetal bovine serum and antibiotics, and plated in 2 cm2 wells at 400.000-500.000 cells/well. 20-24 h after plating, the cells were co-transfected with eGFP (0.3 μg/well) and TRPM8-wt or TRPM8-Y745H (2 μg/well) by incubating them with a solution containing the plasmid DNA and Lipofectamine 2000 (Invitrogen) for 5 hours. Subsequently, the cells were trypsinized and replated on poly-L-lysine-coated round coverslips (6 mm diameter). 20-48 h post-transfection, GFP(+) cells were selected for calcium imaging or electrophysiology experiments.
Calcium imaging
Prior to experiment, HEK293 cells transfected with the TRPM8-wt/Y745H constructs were incubated with 5 μM Fura-2AM dissolved in the bath solution and 0.02% Pluronic (both from Invitrogen) for 40 min at 37°C in darkness. Fluorescence measurements were made with a Leica DM IRE2 inverted microscope fitted with a CCD camera (Imago QE Sensicam, TillPhotonics, Graefelfing, Germany). Fura-2 was excited at 340 nm and 380 nm with a Polychrome IV monochromator (Till Photonics) and the emitted fluorescence was filtered with a 510 nm longpass filter. Calibrated ratios were displayed online with Till Vision software v4.01 (TillPhotonics). The bath solution in calcium imaging experiments, from here on referred to as "control solution", contained (in mM): 140 NaCl, 3 KCl, 2.4 CaCl2, 1.3 MgCl2, 10 HEPES and 10 glucose, and was adjusted to a pH of 7.4 with NaOH.
Electrophysiology
Whole-cell voltage-clamp recordings of TRPM8-wt/Y745H transfected HEK293 cells were performed simultaneously with temperature recordings. Current development was monitored at a holding potential of -60 mV. For information on voltage-dependent effects, 1.25-s-duration voltage ramps from -100 to +150 mV were delivered at desired time points. Standard patch-pipettes (3-5 MΩ) were made of borosilicate glass capillaries (Harvard Apparatus Ltd, UK), and contained (in mM): 150 NaCl, 3 MgCl2, 5 EGTA, 10 HEPES, (278 mOsm/kg; pH 7.2, adjusted with NaOH). During electrophysiological recordings, a divalent cation -free bath solution was used to minimize desensitization, which contained: 150 NaCl, 10 EDTA, 10 HEPES and 10 glucose (pH 7.4 with NaOH). The divalent-free solution was used in the preparation of all chemical modulator containing solutions. Current signals were recorded with an Axopatch 200B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA, U.S.A.) applying series resistance compensation (60-90%). Stimulus delivery and data acquisition were performed using pClamp 9 software (Molecular Devices). Data were sampled at 5 kHz (ramps) and 1 kHz (timecourse) and filtered online at 2 kHz.
Temperature stimulation
Coverslip pieces with cultured cells were placed in a microchamber and continuously perfused with solutions warmed at 33°C. The temperature was adjusted with a water-cooled Peltier device placed at the inlet of the chamber, and controlled by a feedback device. Cold-sensitivity was investigated with a temperature decrease to 18-20°C (Figure
1B).
Chemical modulators
The chemical substances studied for their modulatory effect on the TRPM8 constructs were the cooling agent
L-menthol (Scharlau, Spain), and the antagonists SKF96365 (1- [2-(4-methoxyphenyl)-2-[3-(4-methoxyphenyl)propoxy]ethyl-1H-imidazole; Tocris Bioscience, Bristol, UK); clotrimazole (1- [(2-chlorophenyl)diphenylmethyl]-1H-imidazole, Sigma), econazole (1- [2-[(4-chlorophenyl)methoxy]-2-(2,4-dichlorophenyl)ethyl]-1H-imidazole, Sigma), imidazole (1,3-diazacyclopenta-2,4-diene, Sigma), capsazepine (N- [2-(4-Chlorophenyl)ethyl]-1,3,4,5-tetrahydro-7,8-dihydroxy-2H-2-benzazepine-2-carbothioamide, Sigma); and BCTC (4- [3-chloro-pyridin-2-yl]-piperazine-1-carboxylic acid [4-tert-butyl-phenyl]-amide), which was a kind gift from Grünenthal GmbH Aachen, Germany. ACD/ChemSketch Freeware, version 8.0 (Advanced Chemistry Development, Inc., Toronto, ON, Canada,
http://www.acdlabs.com) was used to draw the structures of the various modulators shown in Figure
3.
Experimental protocols and interpretation of results
Experiments were always performed in parallel with cells expressing TRPM8-wt and TRPM8-Y745H to minimize any day-to-day variability in experimental conditions. To estimate the shifts in the voltage-dependence of activation of TRPM8-wt/Y745H in HEK293 cells, current-voltage (I-V) relationships from voltage ramps were fitted with an equation that combines a linear conductance multiplied by a Boltzmann activation term [
48]:
where g is the maximum whole-cell conductance, E
rev
is the reversal potential, V
1/2 is the potential for half maximal activation and s is the slope factor. For each cell, fitting was started by analyzing a condition exhibiting inward current. The value obtained for the parameter E
rev
at this condition was used when fitting I-V curves that lacked inward current.
Molecular modelling
TRPM8-ligand complexes were built using the TRPM8 model constructed by Pedretti et al. [
38]. Ligand structures were downloaded from the PubChem compound NCBI component database
http://pubchem.ncbi.nlm.nih.gov/ and used with no additional optimization. The protein-ligand docking and the analysis of interactions was accomplished with Autodock [
49] implemented in the general purpose molecular modelling software Yasara [
50], and optimized with AMBER 99 force field [
51]. Docking trials were optimized and clustered to remove redundancy and sorted by binding energy. Figures were drawn with Pymol (DeLano Scientific, Palo Alto, California, USA.
http://www.pymol.org).
Western blot
A TRPM8 antibody was generated in the laboratory. Briefly, whole serum anti-TRPM8 was obtained by rabbit immunization with a peptide corresponding to amino acids 1-10 of mouse TRPM8 (MSFEGARLSM) conjugated to keyhole limpet hemocyanin. For immunoblot analysis of TRPM8-wt/Y745H transfected HEK293 cells, 48-h post transfection, cells were mixed with sample buffer (4% SDS (w/v); 10% glycerol (v/v); 4% 2-mercaptoethanol; 0.2% bromophenol blue (w/v); 50 mM Tris-HCl; pH 6.8), heated at 95°C for 5 min, and then subjected to SDS-PAGE gel electrophoresis (10% acrylamide gel). Proteins were transferred onto a nitro-cellulose membrane Hybond™ ECL (Amersham Biosciences). Membranes were blocked with 10% non-fat milk in TBS (Tris-buffered saline), and subsequently incubated with the primary antibody for TRPM8 at 1:500 dilution for 1 h at room temperature and at 4°C overnight. Blots were treated with peroxidase-conjugated goat anti-rabbit IgG at 1:2000 dilution for 1 h, and protein signals were revealed using the ECL Advance™ Western Blotting Detection Kit (Amersham Biosciences) and visualized with the LAS-1000 imaging system by Fujifilm.
Data analysis
Data are reported as mean ± standard error of the mean. All fitting was carried out with the Levenberg-Marquardt method implemented in the Origin 7.0 software. In the dose-inhibition curves shown in Figure
5, error bars were used as weights in fits to the Hill equation. When comparing two means, statistical significance (p < 0.05) was assessed by Student's two-tailed
t-test. For multiple comparisons of means, one-way or two-way ANOVA was performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego California USA,
http://www.graphpad.com). Paired or repeated-measures tests were used where appropriate to account for inter-individual variability.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AM designed and carried out the electrophysiological studies, the statistical analyses, and drafted the manuscript. MP generated the point-mutant TRPM8-Y745H; conducted the calcium imaging studies and their analysis; as well as the immunoassays. GFB carried out the molecular modelling. AFM participated in the design of the molecular modelling and helped to draft the manuscript. FV conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.