| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Alicante Institute of Neuroscience, University Miguel Hernández-CSIC
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
140 mV more negative in native channels compared to recombinant channels, with a much higher open probability at negative membrane potentials in the former. In functional terms, this difference translates into a shift in the apparent temperature threshold for activation towards higher temperatures for native currents. This difference in voltage-dependence readily explains the high threshold temperatures characteristic of many cold thermoreceptors. The modulation of TRPM8 activity by different chemical agents unveils an important flexibility in the temperature–response curve of TRPM8 channels and cold thermoreceptors.
(Received 22 November 2006;
accepted after revision 16 February 2007;
first published online 22 February 2007)
Corresponding author A. Mälkiä: Instituto de Neurociencias de Alicante, Universidad Miguel Hernández-Consejo Superior de Investigaciones Cientificas (CSIC), Apartado 18, San Juan de Alicante 03550, Spain. Email: annika.malkia{at}umh.es
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
TRPM8 is a non-selective calcium-permeable TRP channel that is activated by cold and menthol, and is postulated to play a critical role in the transduction of moderate cold stimuli that give rise to cool sensations (McKemy et al. 2002; Peier et al. 2002; Reid, 2005; Voets et al. 2007). TRPM8 has a limited expression profile in the nervous system, restricted to a subpopulation of primary sensory neurons of small diameter in dorsal root and trigeminal ganglia (McKemy et al. 2002; Peier et al. 2002). Most cold-sensitive neurons are excited by both cooling and menthol (Reid & Flonta, 2001; McKemy et al. 2002; Viana et al. 2002; Thut et al. 2003), and also express TRPM8 mRNA transcripts (Nealen et al. 2003). Moreover, many cold-sensitive neurons express a non-selective cation current (Icold) with biophysical and pharmacological properties consistent with the properties of TRPM8-dependent currents in transfected cells (Okazawa et al. 2002; Reid et al. 2002). In addition, TRPM8 has been found in prostate tissue, where its physiological function remains uncertain (Tsavaler et al. 2001; Zhang & Barritt, 2006).
Although TRP channels were originally thought to be voltage independent, it now seems that several of them exhibit weak voltage dependence (Hofmann et al. 2003; Nilius et al. 2003, 2005; Brauchi et al. 2004; Voets et al. 2004). In the case of TRPM8, the voltage dependence manifests as activation upon depolarization to positive transmembrane potentials, and a rapid and voltage-dependent closure at negative potentials (Brauchi et al. 2004; Voets et al. 2004). Cooling shifts the activation curve of TRPM8 towards more negative potentials, and thus increases the probability of channel openings at physiological membrane potentials. A similar shift is induced by the cooling agent menthol, causing the channel to activate at temperatures above 30°C (Voets et al. 2004).
Despite the important physiological functions of TRP channels, knowledge about their biophysical properties is still modest compared to that of other ion channels (Owsianik et al. 2006) and pharmacological tools to study or modulate them are very limited (Desai & Clapham, 2005; Dhaka et al. 2006). A notable exception is the pharmacology of the heat- and vanilloid-activated channel TRPV1, which has expanded significantly in the past few years (Garcia-Martinez et al. 2002; Valenzano et al. 2003; Krause et al. 2005). By contrast, reports on means to regulate TRPM8 activity are scarce and incomplete. Recent studies show that ethanol inhibits TRPM8 function at concentrations of the order of 0.5–3% (Weil et al. 2005; Benedikt et al. 2007). In addition, a number of known TRPV1 antagonists have been tested on heterologously expressed TRPM8 channels, such as the complex between the divalent copper ion and 1,10-phenanthroline (Cu–Phe), capsazepine, 4-(3-chloro-pyridin-2-yl)-piperazine-1-carboxylic acid (4-tert-butyl-phenyl)-amide (BCTC) and the related thio-BCTC and (2R)-4-C3-chloro-2-pyridinyl)-2-methyl- N-[4-(trifluoromethyl)phenyl]-1-piperazinecarboxamide (CTPC), the urea derivative SB-452533 and the cinnamide derivative SB-366791 (Behrendt et al. 2004; Weil et al. 2005; Madrid et al. 2006). With the exception of BCTC and ethanol, studies with these antagonists have been limited to responses evoked by application of menthol at constant room temperature – a mixture of chemical and physiological stimuli – thereby revealing little information about the nature and mechanism of the inhibition. Furthermore, almost nothing is known about the actions of these or other TRPM8 blockers on native thermoreceptors.
In many cases, the lack of specific blockers prevents or impairs a proper functional characterization of the channel in physiological systems. The availability of selective and potent TRPM8 channel antagonists is an essential tool in clarifying the role of different ion channels in thermal responses of intact cold receptors. In addition, modulators of TRPM8 activity have significant therapeutic potential in the treatment of prostate cancer (Zhang & Barritt, 2006). Here, we studied SKF96365, BCTC and 1,10-phenanthroline for their blocking effects on cold-activated TRPM8 responses. BCTC and SKF96365, which exhibited high antagonist potency, were subsequently characterized with respect to their mechanism of inhibition; results suggest a similar but opposite mode of action to that of menthol. The results obtained on recombinant channels were verified on native cold thermoreceptors.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Transfection of HEK293 cells with TRPM8 was carried out using the recombinant bicistronic expression plasmid pcINeo–TRPM8–IRES–GFP, which carries the protein-coding region of rat TRPM8 (accession number, AY072788) and the green fluorescent protein (GFP) coupled by an internal ribosomal entry site (IRES) sequence. GFP fluorescence could thus be used to identify TRPM8-expressing cells. The bicistronic vector pcINeo–IRES–GFP was provided by Jan Eggermont (Katholieke Universiteit Leuven, Belgium) and pcDNA3–TRPM8 was made available by David Julius (University of California, San Francisco, CA, USA). The new construct was verified by automatic sequencing.
Cell culture
HEK293 cells were obtained from the European Collection of Cell Cultures (Salisbury, UK). Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum and antibiotics, and plated in 2 cm x 2 cm wells at 4–5 x 105 cells well–1. Next, 20–24 h after plating, the cells were transfected with the TRPM8–IRES–GFP construct by incubating them with a solution containing the plasmid DNA (2 µg well–1) and Lipofectamine 2000 (Invitrogen; 3 µl well–1) for 4–6 h. Subsequently, the cells were trypsinized and replated on laminin-coated round coverslips (12 mm diameter) at 8–10 x 104 cells coverslip–1. GFP-positive cells were selected for calcium-imaging or electrophysiology experiments 20–72 h after transfection.
HEK293 cells stably expressing rat TRPM8 channels (CR#1 cells) were kindly provided by Ramón Latorre (Center for Scientific Studies, Valdivia, Chile). They were cultured as described in by Brauchi et al. (2004).
All experimental procedures concerning animals were carried out according to the Spanish Royal Decree 223/1988 and the European Community Council directive 86/609/EEC. Trigeminal ganglion neurons from neonatal mice were cultured as previously described (de la Pena et al. 2005). In brief, newborn Swiss OF1 mice (postnatal day 1–5) were anaesthetized with ether and decapitated. The trigeminal ganglia were isolated and incubated with 1 mg ml–1 collagenase type IA, and cultured in a medium containing 45% DMEM, 45% F-12 and 10% fetal calf serum (Invitrogen), supplemented with 4 mML-glutamine (Invitrogen), 200 µg ml–1 streptomycin, 125 µg ml–1 penicillin, 17 mM glucose and nerve growth factor (NGF, mouse 7S, 100 ng ml–1, Sigma-Aldrich, Madrid, Spain). Cells were plated on poly-L-lysine-coated glass coverslips and used after 1–3 days in culture.
Calcium imaging
The calcium imaging experiments were conducted with the fluorescent indicator Fura-2. Prior to each experiment, the cells were incubated with 5 µM acetoxymethylester form of Fura-2 (Molecular Probes Europe, the Netherlands) for 45 min at 37°C. Fluorescence measurements were made with a Zeiss Axioskop FS (Germany) upright microscope fitted with an ORCA ER CCD camera (Hamamatsu, Japan). Fura-2 was excited at 340 and 380 nm (excitation time, 200 or 300 ms) with a rapid switching monochromator (TILL Photonics, Germany), and the emitted fluorescence was filtered with a 510 nm long-pass filter. Mean fluorescence intensity ratios (F340/F380) were displayed on-line with Axon Imaging Workbench or Metafluor software (Molecular Devices, PA, USA). The calcium imaging experiments were performed simultaneously with temperature recordings. The bath solution, referred to as ‘control solution’, contained (mM): NaCl 140, KCl 3, CaCl2 2.4, MgCl2 1.3, Hepes 10 and glucose 10, and was adjusted to pH 7.4 with NaOH.
Electrophysiology
Whole-cell voltage-clamp recordings were performed simultaneously with temperature recordings. Standard patch pipettes (3–5 M
) were made of borosilicate glass capillaries (Harvard Apparatus Ltd, UK) and contained (mM): CsCl 140, MgCl2 0.6, EGTA 1 and Hepes 10; 278 mosmol kg–1, pH adjusted to 7.4 with CsOH. In Icold threshold experiments, the internal solution contained (mM): KCl 140, NaCl 6, MgCl2 0.6, EGTA 1, NaATP 1, NaGTP 0.1 and Hepes 10; 282 mosmol kg–1, pH adjusted to 7.4 with KOH). The bath solution used was the same as in the calcium imaging experiments. For whole-cell recordings in trigeminal neurons, patch pipettes had a resisitance of 7–8 M. To measure the activation of Icold in neurons, the bath solution contained (mM): NaCl 140, KCl 3, MgCl2 1.3, CaCl2 0.1, Hepes 10, glucose 10 and TTX 0.5 x 10–3; pH was adjusted to 7.4 with NaOH. The pipette solution contained (mM): CsCl 140, MgCl2 0.6, EGTA 1, Hepes 10, ATPNa2 1 and GTPNa 0.1; pH was adjusted to 7.4 with CsOH. These modifications were necessary to minimize large voltage-dependent currents. Current signals were recorded with an Axopatch 200B patch-clamp amplifier (Molecular Devices). Stimulus delivery and data acquisition were performed using pCLAMP9 software (Molecular Devices).
Chemical modulators
The chemical substances studied for their modulatory effect on TRPM8 were the cooling agent L-menthol (Scharlau, Spain), and the antagonists 1,10-phenanthroline (Sigma), SKF96365 (Tocris Bioscience, Bristol, UK) and BCTC which was a kind gift from Grünenthal GmbH Aachen (Germany).
Temperature stimulation
Coverslips with cultured cells were placed in a microchamber and continuously perfused with solutions warmed to 32–34°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 drop to 15–18°C (see Fig. 1E).
|
During calcium-imaging experiments, the effects of the antagonist compounds were investigated with a protocol, wherein a first cooling stimulus in control solution was followed by a second one in the presence of a blocking agent (see Fig. 2A and B). To account for possible desensitization of the response during subsequent cooling stimuli, the same protocol was carried out in the absence of antagonists (see Fig. 1H). The observed reduction of the second response peak in control solution was taken into account when quantifying the blocking effects of the antagonists by defining an expected maximum response (Max Response) for the second application:
|
| (1) |
|
| (2) |
|
| (3) |
|
|
| (4) |
T according to the following expression (Hille, 2001):
|
| (5) |
T is the difference between the baseline temperature (34°C) and the temperature of the cold stimulus. On the basis of control experiments in non-transfected HEK293 cells, the value of Q10 was fixed at 1.5, which is a reasonable value for the conductance of voltage-gated channels (Hille, 2001).
To provide information on shifts in the threshold temperatures, a protocol similar to that used for calcium imaging was used, where responses to cold in the presence and absence of antagonists were recorded at a holding potential of –60 mV. To estimate the shifts in the voltage dependence of activation of TRPM8 in HEK293 cells, current–voltage (I–V) relationships obtained from repetitive (0.2 Hz) voltage ramps (–100 to +200 mV, 525 ms duration) were fitted with a function that combines a linear conductance multiplied by a Boltzmann activation term (Nilius et al. 2006):
|
| (6) |
Using voltage ramps instead of steps involves the possibility of not working under steady-state conditions. The benefit of the ramp protocol is, nevertheless, that it is more rapid, thus minimizing the time-dependent rundown of the current. We performed control experiments where we applied voltage ramps at two different speeds (525 ms and 5 s duration) in the same cells. No statistically significant difference was observed between the fitting parameters at the two speeds. However, we do not rule out the possibilty that the absolute values of the parameters may be slightly affected by non-stationary conditions.
Data analysis
Data are reported as mean ± standard error of the mean. The apparent threshold temperatures were estimated as the first point at which the measured signal (F340/F380 or current) deviated by at least four times the standard deviation of its baseline. Data were analysed with WinASCD written by Dr Guy Droogmans (ftp://ftp.cc.kuleuven.ac.be/pub/droogmans/winascd.zip) and Origin 7.0 (OriginLab Corporation). Fitting was carried out with the Levenberg–Marquardt method implemented in Origin 7.0 software. In dose–response fits, the standard errors of the mean were used as weights. When comparing two means, statistical significance (P < 0.05) was assessed by Student's two-tailed t test. For multiple comparison of means obtained in the same subjects, one-way repeated-measures ANOVA was performed using GraphPad Prism version 4.00 for Windows (San Diego, CA, USA, http://www.graphpad.com).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Analysis of 25 separate calcium-imaging experiments, with a total of 107 GFP-expressing HEK293 cells, revealed that 88% of these responded to a cold stimulus (Fig. 1A–D). As non-transfected HEK293 cells (Fig. 1B and D) or cells transfected with the empty vector pcINeo/IRES–GFP (data not shown) do not respond to cooling, this was taken as proof of the presence of functional TRPM8 channels in the cells. The time that the cells spent in culture between transfection and experiment seemed to influence the correlation between GFP fluorescence and TRPM8 expression. Cells that were used during days 1 and 2 (i.e. 20–58 h after transfection) exhibited the highest co-expression values: 94% of the GFP-positive cells were also TRPM8-positive at day 1 (n = 34) and 96% at day 2 (n = 55). The lowest percentage (72%, n = 18) was found in cells studied during day 3 (64–82 h) post transfection. A plausible explanation for this finding may be that by day 3, the amount of GFP in the cells reaches a toxic level (Liu et al. 1999).
Cold-evoked responses in TRPM8-transfected HEK293 cells
From electrophysiology experiments, we determined a mean apparent threshold temperature of TRPM8 in HEK293 cells in control solution of 27.6 ± 0.5°C (n = 19), when the temperature was lowered from a base value of 32–33°C at a holding potential of –60 mV. In the calcium-imaging experiments, the apparent threshold temperature was 26.8 ± 0.1°C (n = 641). The threshold distribution in the calcium-imaging experiments exhibited a range of 13°C and a median value of 27.0°C (see Fig. 1F). The wide range of observed thresholds cannot be attributed to experimental error, as cells in the same field exhibited largely different threshold temperatures (see Fig. 1E). The mean Ca2+ elevation was 480 ± 40 nM (n = 95). As seen in Fig. 1G, a low but statistically significant correlation could be established between response threshold and amplitude of Ca2+ increase, the trend being that large calcium responses are more likely to occur in cells with high threshold temperatures.
When two subsequent cooling stimuli were carried out, as shown in Fig. 1H, the amplitude of the second response was found to be 23 ± 2% smaller than the first one (n = 39, P < 0.001, Student's paired t test; Fig. 1I). This mean desensitization of the response was taken into consideration when estimating the blocking effects of drugs, by assigning a desensitization correction (DS, 0.77) to the expected response in eqn (1). Whereas DS affected the response amplitude, it did not affect the threshold temperatures of subsequent cold-evoked responses (Fig. 1I).
SKF96365, BCTC and 1,10-phenanthroline block cold-evoked responses in TRPM8-expressing HEK293 cells
SKF96365 is a non-specific blocker of various calcium-permeable channels, including both receptor-operated and voltage-gated types, as well as those activated by internal calcium store depletion (Merritt et al. 1990). At higher concentrations (IC50, 40 µM), SKF96365 has also been reported to block an inwardly rectifying K+ current in endothelial cells (Schwarz et al. 1994). In dorsal root ganglion neurons, a cold-activated current was reduced by almost 70% in the presence of 100 µM SKF96365 (Reid et al. 2002).
We tested the effects of SKF96365 on cold-evoked responses in TRPM8-transfected HEK293 cells. As shown in Fig. 2A, 20 µM SKF96365 produced a robust and reversible inhibition of cold-evoked [Ca2+]i elevation in transfected cells. To quantify the potency of SKF96365 as a TRPM8 channel antagonist, we constructed dose–inhibition curves (Fig. 2C) from the responses at 17.5°C: a temperature at which the average response in control solution had reached its maximum value. Hill fits to these data yielded an IC50 of 1.0 ± 0.2 µM. At the end of each experiment, the cells were left in control solution for 15 min to remove the accumulated antagonist. Washing out the effects of SKF96365 was concentration dependent and amounted to 56–85% of the initial signal, for decreasing concentration of the blocker.
In the whole-cell electrophysiology experiments, various concentrations of SKF96365 (0.1–20 µM) were applied during an extended cooling stimulus (Fig. 2D). Currents were evoked by consecutive voltage ramps from –100 to +100 mV, delivered every 5 s, and Icold at +80 mV was plotted versus time. SKF96365 produced a concentration-dependent, but incomplete inhibition of this current, as seen in Fig. 2D. This effect was highly reversible upon washing: the initial signal recovered by > 80% for concentrations below 10 µM, and > 70% for 10 and 20 µM SKF96365. Figure 2F shows the complete dose–response curve of block of cold-evoked currents by SKF96365 at +80 mV and 17–18°C. The solid line represents the fit to the Hill equation, which yielded an IC50 of 0.8 ± 0.1 µM.
BCTC was recently introduced as an orally bioavailable antagonist agent of the TRPV1 channel, with a high selectivity in an extensive radioligand screen against other ion channels (Valenzano et al. 2003). More recent studies demonstrate that BCTC also readily blocks menthol-evoked (Behrendt et al. 2004; Weil et al. 2005) and cold-evoked responses of TRPM8 (Madrid et al. 2006). We tried to establish BCTC as a reference compound of TRPM8 block, comparing its effects on TRPM8 with those of other drugs. A typical calcium-imaging recording is shown in Fig. 2B. The Hill fit (Fig. 2C) of the blockade at 17.5°C yielded an IC50 of 0.68 ± 0.06 µM and full block at high concentrations. The recovery of the initial signal after a 5 min washing period depended on the BCTC concentration used, varying between 21% and 63% for the highest and lowest concentrations tested. Whole-cell electrophysiology experiments (Fig. 2E), confirmed that BCTC potently blocks TRPM8-mediated cold responses at +80 mV with an IC50 of 0.54 ± 0.04 µM and Blockmax of 1.00 ± 0.01 (Fig. 2F).
Cu–Phe is an oxidizing agent capable of inducing formation of disulphide bridges between appropriately located thiol groups, and has been widely employed in studies of the gating motion of voltage-gated channels (Liu et al. 1996). The Cu–Phe complex is also an antagonist of the TRPV1 channel; however, in this case it acts as an open-channel blocker instead of inducing cysteine cross-linking (Tousova et al. 2004). Meanwhile, both the free 1,10-phenanthroline and its Cu–Phe complex are equally potent open-channel blockers of the human skeletal muscle Na+ channel (Popa & Lerche, 2006).
In our studies, the free 1,10-phenanthroline acted as an antagonist of the TRPM8-mediated cold-evoked responses in HEK293 cells. The dose–response data of 1,10-phenanthroline block from calcium-imaging and whole-cell electrophysiology experiments were fitted to the Hill equation (Fig. 2C and F), and yielded IC50 values of 100 ± 20 and 180 ± 20 µM, respectively. We subsequently investigated the inhibitory capacity of the Cu–Phe complex on cold-evoked TRPM8 responses. In the presence of 100 µM Cu2+ and 400 µM 1,10-phenanthroline, 78 ± 4% (n = 24) of the calcium response was blocked at 17.5°C (Fig. 2C, grey square), a value almost identical to that observed in the presence of 400 µM free 1,10-phenanthroline alone (77 ± 4%, n = 26, P = 0.83, Student's unpaired t test). Thus, in the case of TRPM8, complex formation with the divalent copper ion does not seem to be necessary for 1,10-phenanthroline antagonism. The inhibition in response to free and complexed 1,10-phenanthroline was reversible which strongly indicates that the mechanism of inhibition does not involve cross-linking of cysteines (Liu et al. 1996; Aziz et al. 2002).
Blockade of Icold by BCTC and SKF96365 is voltage dependent
The nature of the antagonism of cold-evoked currents of TRPM8 by BCTC and SKF96365 was studied in more detail with whole-cell electrophysiology. As seen in Fig. 3A and D, cooling of the control bath solution from around 35°C to 18°C activated a current characterized by a reversal potential near 0 mV and strong outward rectification, as reported before for TRPM8 currents (McKemy et al. 2002; Voets et al. 2004). The concentration-dependent inhibition of this current by SKF96365 and BCTC is clearly seen in the I–V plots. More detailed analysis of the I–V curves revealed that blockade of Icold by the two antagonists is voltage dependent. Figure 3B and E shows the blockade induced by concentrations around the IC50 of BCTC and SKF96365 over a wide range of potentials, demonstrating that Icold is strongly blocked at negative potentials by these submaximal concentrations of antagonist, while a significant fraction of the current remains at more positive potentials. The statistical significance of the observed differences in degree of block with potential was assessed using a one-way repeated-measures ANOVA test combined with a post test for a linear trend. The results (BCTC, P < 0.01 for one-way repeated-measures ANOVA and P < 0.001 for linear regression; SKF96365, P < 0.001 for both) indicate that the mean blockade at different potentials is statistically different, and that the voltage dependence of the block is not an artefact caused by random error.
|
The inhibition of the cold-evoked current by 1,10-phenanthroline was also voltage dependent (not shown), exhibiting weaker inhibition at more positive potentials. The effect was quantified for 600 µM 1,10-phenanthroline, which blocked 92 ± 2% of the cold-evoked current at +40 mV, and 70 ± 2% at +100 mV (P < 0.001, Student's paired t test, n = 3).
BCTC and SKF96365 shift the activation curve of TRPM8 towards more positive potentials
The voltage-dependent antagonism by BCTC and SKF96365, together with the fact that both compounds are electroneutral at pH 7.4, led us to think that they are not acting as typical pore blockers driven by the transmembrane voltage (Hille, 2001). Recently it was shown that low temperature and menthol activate TRPM8 channels by producing a shift in the voltage-dependence of activation towards more negative potentials (Voets et al. 2004; Brauchi et al. 2004). We hypothesized that the mechanism of inhibition exerted by the studied TRPM8 antagonists involved an opposite effect on the voltage dependence of activation (i.e. a shift in gating towards more depolarized potentials). Consequently, TRPM8 activation was probed with 525 ms duration voltage ramps from –100 to +200 mV in whole-cell voltage-clamp mode both in CR#1 HEK293 cells stably expressing rat TRPM8 channels (Brauchi et al. 2004) and in transiently transfected HEK293 cells (Fig. 4A, B, D and E).
|
A summary of the mean values obtained for the fitting variables under different experimental conditions is shown in Table 1. As reported previously, cooling and menthol produced marked leftward shifts in V1/2 (Voets et al. 2004; Brauchi et al. 2004). By contrast, application of 3 µM BCTC produced a positive shift in V1/2 and a reduction in g but no apparent change in the slope factor s (P > 0.05 when comparing conditions with and without BCTC). For conditions of 33°C, 20°C, 100 µM menthol at 33°C, and 100 µM menthol at 20°C, application of 3 µM BCTC shifted V1/2 by 34 ± 9 mV (n = 6); 67 ± 11 mV (n = 8); 78 ± 6 mV (n = 21), and 97 ± 11 mV (n = 15), respectively (P < 0.05 for all shifts, paired t test). These data indicate that the inhibitory effects of BCTC increase under conditions of high open probability of the TRPM8 channel.
|
We subsequently studied whether a similar effect on the voltage activation of TRPM8 was induced by SKF96365 and 1,10-phenanthroline. Figure 4D and E shows the time course of whole-cell current development and I–V data of a TRPM8-positive cell where 3 µM SKF96365 was applied during a cold stimulation. As summarized in Fig. 4F, application of 3 µM SKF96365 during cooling induced a positive shift on V1/2 that averaged 24 ± 3 mV (n = 6; P < 0.001, paired t test). For 300 µM 1,10-phenanthroline, the mean value of the shift was 35 ± 5 mV (n = 5, P < 0.01, paired t test).
The functional consequence of the bidirectional shift in channel gating is a displacement in the apparent temperature-response threshold of TRPM8-expressing cells
The observation that the antagonists and menthol shift the activation curve of TRPM8 in opposite directions prompted us to investigate whether the same holds true for the apparent response threshold during a temperature stimulus.
Figure 5A shows the whole-cell current (holding potential, –60 mV) in a TRPM8-expressing cell as the bath temperature was cooled in the absence and presence of 0.6 µM BCTC. Comparing the current responses during the cooling ramp (Fig. 5B), we found that the apparent response threshold was shifted by an average of –4.3 ± 0.5°C (P < 0.01, n = 4, paired t test). The shifts observed in four individual cells are shown in Fig. 5C. The effects of BCTC on apparent TRPM8 response thresholds obtained from calcium-imaging experiments (Fig. 5D) were analysed in terms of relative response curves (Fig. 5E). Briefly, for each cell, the relative response at a temperature T was calculated as the fluorescence response normalized to the expected maximum response of the cell (i.e. the response to cold in control solution corrected with the desensitization coefficient (eqn (1)). The relative responses of the individual cells (n = 16–19 for each condition) were averaged, yielding the curves seen in Fig. 5E. Whereas 10 µM BCTC produced a complete inhibition of the current, lower concentrations of the drug reduced the peak amplitude and shifted the response towards lower temperatures. The control curve was obtained in an identical manner from experiments where the second cooling application was also performed in the absence of drug; this confirmed that the observed shifts in the response curves is not an artefact of applying subsequent cooling applications.
|
|
These results indicate that, with regard to cold sensing, the main functional effect of the antagonists is a dose-dependent shift in the apparent temperature activation threshold of the cell.
The modulation of cold-evoked responses of TRPM8-expressing cells by menthol and antagonists is additive
Considering the opposite shifts in activation curves and apparent temperature thresholds induced by menthol and the antagonists, we were curious to investigate the effects of joint applications of agonists (thermal and chemical) and antagonists on TRPM8 channel activity. Menthol-evoked responses of TRPM8 have previously been shown to be blocked by BCTC (Behrendt et al. 2004; Weil et al. 2005; Madrid et al. 2006). We confirmed the same to be true for SKF96365 (not shown), with 20 µM of the antagonist blocking 99.5 ± 3% of the calcium response evoked by 100 µM menthol (n = 12, P < 0.001, paired t test). To obtain quantitative information on the menthol-induced shifts in temperature sensitivity of TRPM8, we constructed a dose–threshold shift curve (Fig. 7A and B), which exhibited a maximum threshold shift of 11.6 ± 0.9°C and a half-maximum concentration of 37 ± 9 µM. Remarkably, this EC50 value is very similar to the dose-dependent shift in V1/2 produced by menthol (Voets et al. 2004). The cooling ramps in these experiments were initiated from 39–45°C to allow for the detection of the menthol-modified temperature thresholds in the absence of a baseline response (see Fig. 7A). It should be noted that the briefly applied higher baseline temperature did not modify per se the sensitivity of the channel to temperature changes. This was confirmed by an experiment carried out in control solution, where the thresholds to the first cold application (base temperature 32–33°C) and the second cold application (after 200 s of base temperature 39°C) were not statistically different (n = 5, P = 0.7, paired t test).
|
We subsequently searched our data for menthol and antagonist concentrations for which matching threshold temperature shifts, but in the opposite direction, had been observed during individual applications, to see whether these shifts could be cancelled out during the combined application of both agents. Figure 7E shows a calcium-imaging experiment, where the joint effects of 1 µM BCTC and 20 µM menthol on apparent response-threshold temperatures were investigated. Notably, analysis of the results (Fig. 7F) revealed that the individually observed shifts (i.e. the opposite shifts produced by menthol and BCTC) indeed cancel each other out when the compounds are co-applied during cooling. Another antagonist with similar cancelling effects when combined with 20 µM menthol was 4 µM SKF96365 (not shown). The fact that simple algebraic operations can explain the opposite effects of agonists (cold and menthol) and antagonists on the TRPM8 temperature-response threshold suggests that a common mechanistic principle underlies the actions of the various modulators.
Bidirectional modulation by menthol and blocking agents on cold-evoked responses are maintained in trigeminal thermoreceptors
We investigated whether similar bidirectional shifts in channel function can be observed in native cold thermoreceptors, the sensory neurons responsible for the transduction and coding of temperature signals in the peripheral nervous system (Hensel, 1981). Cold-sensitive trigeminal ganglion neurons were identified by calcium imaging as previously described (de la Pena et al. 2005). During rapid reductions in bath temperature from a baseline of 34°C to approximately 18°C, cold-sensitive neurons responded with an average [Ca2+]i elevation of 247 ± 44 nM (n
= 20) exhibiting a mean apparent threshold of 30.2 ± 0.9°C with a range of 10°C. This threshold temperature is significantly higher than the one we obtained in TRPM8-expressing HEK293 cells (P < 0.001, Student's unpaired t test). All cold-sensitive neurons identified in this particular calcium-imaging screen were also activated by menthol, which suggests they all expressed endogenous TRPM8 channels. We note here that in a previous similar screen with a higher number of neurons, we also identified a cold-sensitive but menthol-insensitive population that represented 8% of the total number of cold-sensitive neurons (Madrid et al. 2006).
In these neurons, 3 µM BCTC strongly suppressed [Ca2+]i increases evoked by cold and completely abolished the responses to 100 µM menthol at 34°C (Fig. 8A and C). As was the case with recombinant TRPM8 channels, applying a cold stimulus in the presence of both menthol and BCTC provoked [Ca2+]i increases with very similar amplitude to the control response(change in [Ca2+]I, 262 ± 54 nM, P = 0.29, n = 20). As seen previously for heterologously expressed TRPM8, BCTC application shifted the apparent threshold temperatures of the cold-sensitive neurons towards lower values. In 10 neurons, the cold-induced response was completely abolished; in the other 10 cold-sensitive cells, the effect was partial and the threshold was shifted by an average of –7.6 ± 0.6°C (P < 0.01). In addition, co-application of menthol and cold completely reversed the negative shift in threshold temperature produced by BCTC, yielding a mean apparent threshold temperature of 31.9 ± 0.7°C. The mean shifts in threshold temperature for the 10 neurons in which some amount of response remained in the presence of BCTC are shown in Fig. 8B.
|
The effects of BCTC on the voltage activation of Icold were further studied with application of –100 to +200 mV ramps from a holding potential of –60 mV (Fig. 9A). An I–V plot of the cold-evoked current in the absence and presence of chemical modulators is shown in Fig. 9C. Icold displays strong outward rectification and reversal close to 0 mV (0.5 ± 1.0 mV; n = 8), which is similar to the properties of current conducted by TRPM8 (ITRPM8) in HEK293 cells. As observed for recombinant TRPM8 channels above, application of a cooling stimulus in the presence of BCTC produced an incomplete inhibition of outward currents but a full inhibition of inward currents in trigeminal neurons. This is illustrated in Fig. 9D, where IBCTC represents the part of Icold that was blocked by 3 µM BCTC (i.e. the difference between traces Icold and of the cold current of trigeminal neurons in the presence of 3µM BCTC (Icold+3 BCTC in Fig. 9C). Note that IBCTC only begins to deviate from Icold at potentials more positive than +70 mV, indicating a complete block by 3 µM BCTC below this value. A similar result was obtained by analysing the blockade of Icold in cold-sensitive trigeminal neurons by 3 µM BCTC over various potentials (Fig. 9B).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study, we provide a thorough characterization of the responses of recombinant TRPM8 channels to their physiological stimulus, cold temperature, and describe in mechanistic terms the effects of substances that increase and decrease the temperature sensitivity of the channel. We identified a common modulatory action of three chemical antagonists (BCTC, SKF96365 and 1,10-phenanthroline) on TRPM8 function that involves marked shifts in their voltage-dependent gating. Furthermore, we describe important differences between the properties of recombinant TRPM8 channels and native currents in trigeminal cold thermoreceptors that underlie their high thermal sensitivity.
Cold-induced responses in TRPM8-expressing HEK293 cells and trigeminal neurons
In our experiments, rat TRPM8-transfected HEK293 cells responded to cooling with apparent threshold temperatures of 26.8°C and 27.6°C as measured by calcium imaging and electrophysiology, respectively. These values, although among the highest reported for heterologously expressed rodent T
