Menthol response and adaptation in nociceptive-like and nonnociceptive-like neurons: role of protein kinases

Transient receptor potential M8 (TRPM8) receptor, first cloned by MacKemy and colleagues [1] as well as Peier and colleagues [2] from primary afferent neurons of rats and mice, is a principal sensor for cold temperature and belongs to the transient receptor potential (TRP) protein family. Like most of other members in TRP family, TRPM8 is a membrane ion channel that can allow positively charged ions (Na⁺, Ca²⁺, K⁺) to flow through cell membranes when the channel opens. The TRPM8 channel opens when temperature drops below 26 ± 2°C, resulting in depolarizing membrane currents [1‐3]. Membrane currents flowing through TRPM8 channels increase with decreasing temperature and reach maximum response near 10°C. TRPM8 senses temperature changes in the range of both innocuous cold (28-15°C) and noxious cold (<15°C) [1‐3]. Activation of TRPM8 can result in a large increase of intracellular Ca²⁺ levels due to the high Ca²⁺ permeability of this channel [1, 2, 4, 5]. TRPM8 can also be activated by menthol, an active ingredient of peppermint that produces a cooling sensation [1, 2, 6, 7].

TRPM8 receptors are expressed on 10-15% of the total trigeminal ganglion (TG) neuron population and 5-10% of dorsal root ganglion (DRG) neuron population [1, 2, 7, 8]. Consistently, the percentage of menthol-sensitive cells in acutely dissociated rat DRG neurons is similar to that of TRPM8-expressing DRG neurons [9, 10]. Many TRPM8-expression neurons are found to lack nociceptive markers, suggesting that they are non-nociceptive cold sensing neurons [2]. However, studies have provided anatomical evidence showing TRPM8 immunoreactivity on some TRPV1 (Transient receptor potential V1)-expressing afferent neurons [7, 8]. TRPV1-expressing neurons are believed to be nociceptive afferent neurons that transmit noxious signals to produce burning pain sensations [11‐13]. Using calcium imaging and patch-clamp recording techniques, Xing and colleagues [9] have found that a subpopulation of menthol-sensitive neurons is also sensitive to capsaicin, a noxious stimulant that acts on TRPV1 receptors. Consistent with these observations, co-expression of TRPM8 and TRPV1 have been directly visualized in mice engineered to express enhanced green fluorescent protein (EGFP) driven by a TRPM8 promoter [14, 15]. Thus, menthol-sensitive neurons appear to consist of both non-nociceptive and nociceptive sensory neurons and may play roles in sensing innocuous and noxious cold respectively under physiological conditions [10].

TRPM8 can be regulated through second messenger systems [16‐18]. A role for the PLC/PIP2 (Phospholipase C/phosphatidylinositol (4,5) bisphosphate) second messenger pathway in regulating TRPM8 functions has been well established [16, 17, 19]. It has been suggested that Ca²⁺ influx through TRPM8 channels activates a Ca²⁺-sensitive phospholipase C and the subsequent depletion of PIP2 results in desensitization of TRPM8 channels [16, 17, 19]. Desensitization of TRPM8 channels could also be induced by inflammatory mediators that activate PLC to deplete PIP2 [20]. In comparison with the PLC/PIP2 pathway, the roles of protein kinase pathways in regulating TRPM8 functions remain unclear. Premkumar and colleagues [18] showed in DRG neurons that PKC activators and bradykinin significantly reduced menthol responses. Using HEK293 cells expressing TRPM8, Abe and colleagues [21] also showed that PKC activators reduced menthol responses. Other second message pathways such as PKA have also been suggested to play roles in regulating TRPM8 functions [22, 23]. These previous studies on the regulation of TRPM8 functions were performed either using heterologous expression system or functionally unidentified sensory neurons. Therefore, it is unclear if the reduction of TRPM8 functions occurs in a similar manner across functionally distinct populations of neurons. In addition, previous studies did not test whether TRPM8-mediated responses were affected by different protein kinase inhibitors, a result that is essential for establishing the roles of protein kinases in modulating TRPM8 functions. In the present study, we addressed some of these issues by examining menthol-responsiveness and adaptation in menthol-sensitive/capsaicin-insensitive and menthol-sensitive/capsaicin-sensitive neurons.

Adult Sprague Dawley rats (100-250 g, both genders) were used in all experiments. Animal care and use conformed to National Institutes of Health guidelines for care and use of experimental animals. Experimental protocols were approved by the University of Cincinnati Institutional Animal Care and Use Committee. DRG neuron cultures were prepared as described previously [4]. In brief, rats were deeply anesthetized with isoflurane (Henry Schein, NY.) and sacrificed by decapitation. DRGs were rapidly dissected out bilaterally in Leibovitz L-15 media (Fisher, GA) and incubated for 1 hour at 37°C in minimum essential medium for suspension culture (S-MEM) (Invitrogen, Grand Island, NY) with 2% collagenase and 10% dispase and then triturated to dissociate neurons. The dissociated DRG neurons were then plated on glass coverslips pre-coated with poly-D-lysine (PDL, 12.5 μg/ml in distilled H₂O) and laminin (20 μg/ml in Hank's Buffered Salt Solution HBSS, BD bio-science), and maintained in MEM (Invitrogen) culture medium that also contained nerve growth factor (2.5 S NGF; 10 ng/ml; Roche Molecular Biochemicals, Indianapolis, IN), 5% heat-inactivated horse serum (JRH Biosciences, Lenexa, KS), uridine/5-fluoro-2'-deoxyuridine (10 μM), 8 mg/ml glucose, and 1% vitamin solution (Invitrogen). The cultures were maintained in an incubator at 37°C with a humidified atmosphere of 95% air and 5% CO₂. Unless otherwise indicated, cells were used within 72 hours after plating.

For calcium imaging experiments, the calcium indicator Fluo-3 (Invitrogen) was loaded into DRG neurons on coverslips by incubation of cells with 5 μM Fluo-3-AM in normal bath solution at 37°C for 1 hour. Fluo-3-AM stock solution was made with 20% pluronic acid (Molecular Probes) in dimethyl sulfoxide (DMSO) and the stock solution was diluted 1:200 with bath solution for final use. Normal bath solution contained (in mM) 150 NaCl, 5 KCl, 2 MgCl₂, 2 CaCl₂, 10 glucose, 10 HEPES, pH 7.3 adjusted with NaOH, and osmolarity 320 mOsm adjusted with sucrose. After dye loading, a coverslip was mounted on a 0.5-ml perfusion chamber and the chamber was then placed on the stage of an inverted Olympus IX70 microscope (Lake Success, NY). Cells on the coverslip were continuously perfused with normal bath solution flowing at 1 ml/min. Fluo-3 was excited at 450 nm with a mercury lamp and fluorescence emission was collected at 550 nm, and the wave lengths of excitation and emission were achieved by a fluorescence filter set. Fluo-3 fluorescence in the cells was detected with a peltier-cooled charge-coupled device (CCD) camera (PentaMAX-III System, Roper Scientific, Trenton, NJ) under a 10× objective. Images were acquired at one frame per second, 200 ms exposure time per frame, using the MetaFluor Imaging System software (Molecular Devices, Downingtown, PA). Neurons were tested for their sensitivity to menthol (100 μM), AIT (allyl isothiocyanate, 100 μM), or capsaicin (0.5 μM) by applying these compounds for 10 seconds. Adaptation of menthol responses was examined by a prolonged application of menthol (100 μM) for 5 min. Effects of protein kinases on menthol responses and adaptation were tested with 1 μM PDBu (phorbol 12,13-dibutyrate), a protein kinase C (PKC) activator, 100 μM 8-Br-cAMP (8-bromoadenosine-3', 5'-cyclic monophosphate) and 10 μM forskolin, two protein kinase A (PKA) activators, 0.5 μM staurosporine, a broad spectrum protein kinase inhibitor, 1 μM BIM (bisindolylmaleimide), a specific PKC inhibitor, 25 μM RP-cAMPs (Rp-adenosine-3',5'-cyclic monophosphorothioate), a specific PKA inhibitor, and 25 μM KN-62 (1-[N,O-bis(5-isoquinolinesulfonyl)-N-methyl-L-tyrosyl]-4-phenylpiperazine, Tocris), a specific Ca²⁺/calmodulin-dependent protein kinase (CaMKII) inhibitor. Unless otherwise indicated, chemicals and compounds were purchased from Sigma (St. Louis, MO). Testing solutions were rapidly applied to neurons through a glass tube (~500 μm ID) positioned 1.0 mm away from cells. Unless otherwise indicated, testing compounds were applied at an interval of 10 min and capsaicin was always tested last. All experiments where pharmacological agents were applied were done in separate dishes. All experiments were carried out at room temperature of ~24°C.

For most experiments, relative fluorescence intensity (ΔF/F₀) was used to represent menthol responses and neurons with ΔF/F₀ values of ≥0.2 (i.e., equal or above 20% baseline fluorescence intensity) were considered as responsive cells [9]. Percentages of maximal ΔF/F₀ values were used as changes in menthol responses. In some experiments, intracellular Ca²⁺ concentrations ([Ca²⁺]) in cells were calibrated from the measured fluorescence signals. The following equation was use for the calibration of intracellular Ca²⁺ concentrations [24]:

[Ca²⁺] = Kd[(F-Fmin)/(Fmax-F)] where [Ca²⁺] is the concentration (nM) of intracellular Ca²⁺, Kd is the dissociation constant of the dye, F is the fluorescence intensity, Fmin is the intensity at zero [Ca²⁺] and Fmax is the intensity at saturated [Ca²⁺]. Procedures for obtaining Fmax and Frnin caused damage to cells and were therefore carried out at the end of the experiments. Fmax was obtained first by adding the ionophore ionomycin (10 μM), making the cell membrane permeable to Ca²⁺ and allowing the extracellular and intracellular Ca²⁺ to equilibrate. Following this, Fmin was obtained by adding EGTA [ethylene glycol bis(~-aminoethyl ether)-N,N,N',N'-tetraacetic acid; 20 mM] to chelate all Ca²⁺ inside and outside the cells. Then MnCl₂ (30 mM) was added to quench the residual fluorescent signals due to autofluorescence [25]. Kd value of 404 nM was used according to a previous study [26]. Changes in intracellular Ca²⁺ concentrations are calculated by Δ[Ca²⁺]/[Ca²⁺]_0, where [Ca²⁺]₀ is basal intracellular Ca²⁺ level and Δ[Ca²⁺] is the difference between intracellular Ca²⁺ concentrations at a giving time point and [Ca²⁺]₀. Unless otherwise indicated, data were presented as Mean ± SEM. Analysis of Variance (ANOVA) was applied for statistic analysis of unpaired data sets of multiple groups followed by Student-Newman-Keuls Post Hoc Test. Student's t-test was applied for paired data sets. Statistical significance was considered at the level of the p < 0.05.

In this study we demonstrated that menthol-sensitive/capsaicin-insensitive and menthol-sensitive/capsaicin-sensitive neurons had different degrees of responses and adaptation to menthol, that activation of protein kinase C, but not protein kinase A, resulted in a large reduction of menthol responses, and that protein kinase C and CamKII inhibitors had significant effects on menthol responses in MS/CI neurons but not in MS/CS neurons. These results reveal some new properties of TRPM8-mediated responses in both MS/CI and MS/CS neurons, the two neuron populations that most likely represent non-nociceptive cold-sensing neurons and nociceptive cold-sensing neurons respectively [9, 30].

We used menthol as a TRPM8 agonist in the present study. Menthol has been found to interact with TRPA1 in heterologous expression systems that express mouse and human TRPA1 [29, 31], raising a possibility that menthol responses in some cells may be mediated by TRPA1 in our study. However, using DRG neurons from rats, we found inverse rather than positive correlation between menthol-sensitivity and AIT-sensitivity (Figure 1E, F). Consistent with our results, low incidence of co-activation of menthol and AIT has been previously reported in DRG cells by others [28, 32]. The low incidence of co-sensitivity to menthol and AIT is unlikely due to the failure of mentioned agonists to cross-activate these two receptors most of the times.

Therefore, TRPA1 is unlike to significantly account for menthol-induced responses in our rat DRG neurons. Menthol responsiveness was found to be higher in MS/CI neurons than in MS/CS neurons, a result consistent with our previous study using both the calcium imaging and patch-clamp recoding techniques on acutely dissociated DRG neurons [9, 10]. Decay of menthol responses was found to be faster in MS/CI neurons than in MS/CS neurons, suggesting that MS/CI has faster adaptation and MS/CS has slower adaptation to menthol.

We showed that the PKC activator PDBu reduced menthol responsiveness in DRG neurons, and that this effect was abolished in the presence of staurosporine or BIM. Staurosporine inhibits a number of protein kinases including PKC, CaMKII, and tyrosine kinase (p60v-src), but BIM is a highly selective PKC inhibitor. The effect of these two inhibitors is consistent with a previous study that first suggested the involvement of PKC in regulating TRPM8 function in sensory neurons [18]. We further showed that menthol-induced adaptation could be significantly attenuated by staurosporine and BIM in MS/CI neurons. The inhibitor experiments added an important supplement to strengthen the argument that PKC plays a role in the adaptation of menthol responses in MS/CI neurons. In addition to PKC, we found that the selective CaMKII inhibitor KN62 attenuated adaptation of menthol responses in MS/CI neurons, suggesting that CaMKII may play a role in regulating TRPM8 functions in MS/CI neurons. In contrast to MS/CI neurons, menthol responses were not significantly affected by BIM, staurosporine, or KN62 in MS/CS neurons. We found that MS/CS neurons had smaller menthol response with weaker adaptation. It was initially thought that this property of MS/CS neurons might be a result of post-transcriptional regulation of TRPM8 function by protein kinases. However, the lack of the effects by protein kinase inhibitors on menthol responses in MS/CS neurons does not favor this idea. Alternatively, the smaller menthol responses in MS/CS neurons were due to the relatively lower TRPM8 expression as was proposed in our previous study [9]. The weaker adaptation observed in MS/CS neurons could be due to the smaller menthol responses in these neurons. However, we did not observe a clear co-relation between the degree of adaption and menthol responsiveness. Therefore, other factors might account for the differences in adaption rate between MS/CI and MS/CS neurons. One possible factor is PIP2, because PIP2 plays an important role in regulating TRPM8 functions and PIP2 hydrolysis accounts for menthol-induced desensitization [16, 17]. It would be helpful in future studies to investigate whether PIP2 levels are significantly different between MS/CS and MS/CS neurons.

We explored potential involvement of PKA in regulating TRPM8 functions in rat DRG neurons by using both PKA activators and inhibitors. A previous study performed on HEK cells expressing TRPM8 showed that 8-Br-cAMP and forskolin, two PKA activators, inhibited TRPM8 activity induced by menthol [23]. However, both compounds were not found to have any significant effect on menthol responses in our work as well as in another recent study [22]. The discrepancy could be due to the use of different cell types. Direct inhibition of PKA by Rp-cAMP-S did not significantly affect menthol responses in our study, suggesting that there is no direct connection between PKA activity and TRPM8 functions. However, our result does not exclude the possibility that TRPM8 activity could be regulated indirectly through PKA pathway [22].

The differences in TRPM8-mediated responses and adaptation in nociceptive and non-nociceptive neuron populations may have physiological significances. Behavioral responses to innocuous and noxious cold stimuli are different and TRPM8-mediated adaptation may contribute to the differences. Mammals capably adapt to innocuous cold. On the other hand, noxious cold is poorly adapted, which perhaps is a conserved biological trait of mammalian sensory systems for animals to be aware of a harmful cold environment. TRPM8-mediated responses and adaptation in nociceptive and non-nociceptive neuron populations may only partially contribute to behavioral cold adaptation because other receptor molecules such as TRPA1 have been reported to also serve as cold sensors [27, 33]. It would be interesting in future research to directly demonstrate differences to cold adaptation between nociceptive- and non-nociceptive cold sensing neurons and determine if such differences contribute to behavioral cold responses and adaptation.