Cooling from noxious heat to normal skin temperatures excites a subpopulation of cutaneous Aβ-fiber low-threshold mechanoreceptors

Abstract

Sensing cooling temperatures is achieved by primary afferent endings located in the skin and is essential for the survival of animals. TRPM8 channels, primarily expressed in cutaneous C-fibers, have been established as receptors for cooling temperatures, sensing innocuous cooling from the normal skin temperature near 30°C to 17°C, and noxious cooling below 17°C. A cooling sensation is also felt when skin temperatures are first elevated to higher temperatures, for example, noxious heat, and then cool down to the normal skin temperature near 30°C. It is currently not clear what types of cutaneous afferent fibers are involved in sensing the cooling from a high heat to the normal skin temperature. Cutaneous Aβ-fiber low-threshold mechanoreceptors (Aβ-LTMRs) are primarily involved in the sense of touch and are thought to play no role in cooling sensation. In the present study, we conducted the opto-electrophysiological recordings from the skin-nerve preparations made from the hindpaw glabrous skin of Nav1.8-ChR2 transgenic mice. In these transgenic mice, Nav1.8-ChR2-negative Aβ-fiber mechanoreceptors are primarily Aβ-LTMRs, and Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors are mainly high-threshold mechanoreceptors (Aβ-HTMRs). Neither Aβ-LTMRs nor Aβ-HTMRs responded to temperature rising from 30°C to the noxious heat of 43°C. However, a subpopulation of Aβ-LTMRs, but not Aβ-HTMRs, robustly fires action potential impulses in response to the temperature drop from 43°C to 30°C. This finding reveals for the first time that a subpopulation of Aβ-LTMRs senses the cooling for a temperature drop from noxious heat to normal skin temperature.

Introduction

The temperatures of animals’ skin change with the environmental temperatures, resulting in the sensation of heating and cooling. This sensation of skin temperatures is essentially important for survival, and it is primarily generated by the endings of thermally sensitive afferent fibers located in the skin of all mammals. Noxious heat on the skin is sensed by subpopulations of C- and Aδ-afferent fibers,^1,2 and TRPV1 has been found to be the primary thermoreceptor for noxious heat.^1,3,4 Cooling temperatures on the skin are also detected by subpopulations of C- and Aδ-afferent fibers, ² and the TRPM8 channel has been found to be a main cooling temperature sensor that detects temperature drop from 28°C to 10°C.^5–7 Psychophysical studies have indicated that cooling skin in the temperature range from 28°C to near 15°C induces an innocuous cooling sensation, and in the temperature range below 15°C results in the sensation of noxious cold. ⁸ A cooling sensation can also be experienced when one is exposed to the hot sun in the summer and then enters a shaded area. How this type of cooling sensation is generated remains to be poorly understood.

Cutaneous Aβ-fiber low-threshold mechanoreceptors (Aβ-LTMRs) transduce tactile stimuli into action potential impulses at their terminals to result in the sense of touch. In the glabrous skin of mice, Aβ-LTMRs are mainly the termini of Meissner’s corpuscles and the Merkel cell-neurite complex.^9–11 Mechanical indentation to the skin can activate the Merkel cell-neurite complex to induce slowly adapting impulses and activate Meissner’s corpuscles to elicit rapidly adapting impulses. ¹¹ Aβ-LTMRs are generally believed to only respond to mechanical stimuli, and are thought to be insensitive to thermal stimuli such as heating and cooling. In addition to Aβ-LTMRs, some Aβ-afferent terminals can be excited by painful mechanical stimuli, and these afferents are termed Aβ-fiber high-threshold mechanoreceptors (Aβ-HTMRs). Thermal sensitivity of Aβ-HTMRs has never been examined previously. In the present study, we aimed to investigate whether Aβ-LTMR and/or Aβ-HTMRs may play a role in sensing cooling temperatures.

The study of Aβ-LTMRs and Aβ-HTMRs has been facilitated by using Nav1.8-Cre mice. This transgenic mouse line expresses Cre recombinase under the control of the promoter of Nav1.8, a sensory neuron-specific voltage-gated Na⁺ channel primarily expressed in nociceptors.^12–14 Nav1.8-Cre mouse line could be used to genetically tag nociceptive C-, Aδ-, and Aβ-fibers. ¹⁵ For example, by crossing Nav1.8-Cre mice with Ai32 (RCL-ChR2(H134R)/EYFP) mice, we have generated mice that express channel rhodopsin 2 (ChR2) under the control of Nav1.8-Cre (Nav1.8-ChR2 mice). These transgenic mice enable opto-electrophysiological studies on the properties of various mechanoreceptors.^16,17 Utilizing Nav1.8-ChR2 mice, we have investigated the properties of mechanoreceptors in Nav1.8-ChR2-positive and Nav1.8-ChR2-negative afferent fibers innervating the hindpaws of these mice. ¹⁶ We have found that Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors are mainly Aβ-HTMRs, ¹⁶ whereas Nav1.8-ChR2-negative Aβ-mechanoreceptors are primarily Aβ-LTMRs. ¹⁶ Thus, Nav1.8-ChR2 mice provide a valuable transgenic model for investigating the properties of Aβ-LTMRs and Aβ-HTMRs. In the present study, we investigated whether Nav1.8-ChR2-positive Aβ-mechanoreceptors (Aβ-HTMRs) and Nav1.8-ChR2-negative Aβ-mechanoreceptors (Aβ-LTMRs) may respond to cooling temperatures.

Material and methods

Animals

Nav1.8-ChR2 mice were generated by crossing Nav1.8-Cre mice with Ai32 (RCL-ChR2(H134R)/EYFP) mice. Nav1.8-Cre mice were gifts from Dr. John Wood at University College London and transferred to us from Dr. Stephen Waxman’s lab at Yale University. Ai32 mice were purchased from Jackson Labs. We crossed Nav1.8-Cre mice with Ai32 (RCL-ChR2(H134R)/EYFP) mice to generate Nav1.8^cre+; ChR2-EYFP^loxP/+ mouse line, hereafter termed Nav1.8-ChR2. Animal care and use conformed to NIH guidelines for care and use of experimental animals. Experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Alabama at Birmingham.

Ex vivo skin-nerve preparations

Nav1.8-ChR2 mice of both males and females were used at the age of 8–11 weeks. Skin-nerve preparations were made from the glabrous skin of the hindpaw and tibial nerves in a manner described in our previous studies.^16,18 In brief, the animals were anesthetized with 5% isoflurane and then sacrificed by decapitation. Glabrous skin of the hindpaw, including plantar and finger regions, together with the medial plantar nerve and tibial nerve before the branch of the sciatic nerve was dissected. The skin-nerve preparation was then placed in a 60 mm recording chamber that was coated with Sylgard Silicone on the bottom of the chamber. Fat, muscle, and connective tissues on the nerves and skin were carefully removed with a pair of forceps. The skin, with the epidermis side facing up, was affixed to the bottom of the chamber by tissue pins. The nerve bundle was affixed by a tissue anchor in the same recording chamber. The cutting end of the nerve bundle was briefly exposed to a mixture of 0.05% dispase II plus 0.05% collagenase for 30–60 s, and the enzyme was then washed off with the normal Krebs solution (see below). The recording chamber was then mounted on the stage of the Olympus BX51WI upright microscope. The skin-nerve preparation was perfused with a normal Krebs bath solution that contained (in mM): 117 NaCl, 3.5 KCl, 2.5 CaCl₂, 1.2 MgCl₂, 1.2 NaH₂PO₄, 25 NaHCO₃, and 11 glucose (pH 7.3 and osmolarity 325 mOsm) and was saturated with 95% O₂ and 5% CO₂. The Krebs bath solution in the recording chamber was maintained at approximately 30°C during experiments.

Pressure-clamped single-fiber recordings

The pressure-clamped single-fiber recording was performed to record impulses evoked by blue LED light, mechanical indentation, and electrical stimulation in a manner described in our previous studies. ¹⁹ Briefly, the recording electrodes were made with thin-walled borosilicate glass tubing without filament (inner diameter 1.12 mm, outer diameter 1.5 mm, World Precision Instruments, Sarasota, FL). They were fabricated by using P-97 Flaming/Brown Micropipette Puller (Sutter Instrument Co., Novato, CA), and the tip of each electrode was fire polished with a microforge (MF-900, Narishige) to a final size of 4–10 μm in diameter. The recording electrode was filled with Krebs bath solution, mounted onto an electrode holder which was connected to a high-speed pressure-clamp (HSPC) device (ALA Scientific Instruments, Farmingdale, NY) for fine controls of intra-electrode pressures. Under a 40× objective, the end of the individual afferent nerve was visualized and separated by applying a low positive pressure (~10 mmHg or 0.19 Psi) to the recording electrode. The end of a single nerve fiber was then aspirated into the recording electrode under a negative pressure of approximately 10 mmHg. Once the end of the nerve fiber entered the recording electrode in approximately 10 µm, the electrode pressure was readjusted to −3 ± 2 mmHg and maintained at the same pressure throughout the experiment. Nerve impulses in the single afferent fiber were recorded under the I₀ configuration and amplified using a Multiclamp 700B amplifier (Molecular Devices, Sunnyvale, CA). Electrical signals were amplified 500 times and sampled at 25 kHz with an AC filter at 0.1 Hz and a Bessel filter at 3 kHz under an AC membrane mode (Digidata 1550B, Molecular Devices).

To determine the conduction velocity of the recorded afferent fibers, a bipolar stimulation electrode was placed in the tibial nerve bundle, and electrical stimulation was applied to the nerve bundle to evoke AP impulses. The distance between the electrical stimulation and recording sites was approximately 10–14 mm. Electrical stimuli were monophasic square pulses that were generated by an electronic stimulator (Master-9, A.M.P.I, Israel) with a stimulation isolator (ISO-Flex, A.M.P.I, Israel) and delivered to the stimulation electrode. The duration of each stimulation pulse was 200 μs, and the stimulation intensities for evoking impulses were 0.3–2.0 mA.

Mechanical, optical, and thermal stimulation

For an afferent fiber being recorded, its mechanosensitive receptive field in the hindpaw glabrous skin was first searched using a glass rod. Poking the glass rod at the mechanosensitive receptive field of the recorded afferent fiber would result in the APs that were detected by the recording electrode. Once a mechanoreceptor was identified, mechanical stimulation was applied to the same receptive field by von Frey filaments to determine von Frey mechanical thresholds.

To determine whether a mechanoreceptor was from Nav1.8-ChR2-positive or Nav1.8-ChR2-negative afferent fibers, the same mechanosensitive receptive field was stimulated by a blue LED light (Thorlab; M455L4, 455 nm) to test opto-sensitivity. A mechanoreceptor was from Nav1.8-ChR2-positive afferent fibers if light stimulation evoked impulses. Otherwise, the mechanoreceptor was from opto-insensitive or Nav1.8-ChR2-negative afferent fibers. Blue LED light was applied through a 40× objective to a mechanoreceptor with a 1-s pulse of light stimulation at an intensity of 50 mW. Afferent impulses evoked by mechanical and light stimulation were recorded using the pressure-clamped single-fiber recordings described above.

To determine whether a mechanoreceptor could be excited by heating and cooling, the recording chamber was first perfused with the Krebs solution at 30 ± 2°C. A heated Krebs solution was then bath-applied to the recording chamber to raise the temperature to 43°C. Once the temperature reached 43°C, the heated Krebs solution was stopped and the 30°C Krebs solution was applied to bring the temperature at the recording site back to 30°C. To determine whether a mechanoreceptor could be excited by cooling alone, a cold Krebs solution was bath-applied to the recording chamber to reduce the temperature to 13.5°C. Once the temperature reached 13.5°C, the cold Krebs solution was stopped, and the 30°C Krebs solution was applied to bring the temperature at the recording site back to 30°C

Data analysis

Electrophysiological data were analyzed using Clampfit 11 (Molecular Devices, Sunnyvale, CA, USA). Data were collected from 17 male and 26 female animals and aggregated for data analysis. The conduction velocity (CV) was calculated as the distance between the stimulation site and the recording site divided by the latency of the AP impulse elicited by the electrical stimulation. All data analyses were performed using GraphPad Prism (version 8). Unless otherwise indicated, all data were reported as individual observations and/or mean ± SEM of n independent observations.

Results

We made the skin-nerve preparations from Nav1.8-ChR2 mice and employed pressure-clamped single-fiber recordings to investigate the properties of Aβ-fiber mechanoreceptors in the hindpaw glabrous skin (Figure 1(a)). The cutaneous Aβ-fiber mechanoreceptors could be classified as Nav1.8-ChR2-positive (opto-sensitive, Figure 1(b)) and Nav1.8-ChR2-negative (opto-insensitive, Figure 1(c)) Aβ-fiber mechanoreceptors based on their sensitivity to blue LED light stimulation. In the present study, Aδ- and C-fiber mechanoreceptors were excluded, and only the Aβ-fiber mechanoreceptors with conduction velocity CV ≥ 9 m/s^16,20,21 were included. The CV was 14.2 ± 0.6 m/s (n = 21, Figure 1(d)) in Nav1.8-ChR2-negative Aβ-fiber mechanoreceptors, and 11.1 ± 0.3 m/s (n = 20, Figure 1(d)) in Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors. The von Frey threshold was 1.3 ± 0.2 mN (n = 21) in Nav1.8-ChR2-negative Aβ-fiber mechanoreceptors, indicating that they were mainly Aβ-LTMRs (Figure 1(e)). In contrast, Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors had the von Frey threshold of 11.6 ± 2.1 mN (n = 20), indicating that they were primarily Aβ-HTMRs (Figure 1(e)).

Figure 1.

Exciting Aβ-fiber low-threshold mechanoreceptors by cooling temperatures: (a) Experiment settings: The skin-never preparation was made from the hindpaw and the tibial nerve. Receptive field was stimulated by LED light, von Frey filaments, and heating and cooling bath solution. (b) Left panel, sample trace shows impulses elicited in an opto-sensitive (Nav1.8-ChR2-positive) Aβ-fiber mechanoreceptor by a von Frey filament (19.6 mN). Right, light stimulation (50 mW) evokes an impulse in the Aβ-fiber mechanoreceptor. (c) Similar to (b), except that an opto-insensitive (Nav1.8-ChR2-negative) Aβ-fiber mechanoreceptor was tested. Impulses were elicited with a 0.68-mN force von Frey filament. (d) Conduction velocity of Nav1.8-ChR2-negative and Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors. (e) von Frey threshold of Nav1.8-ChR2-negative and Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors. (f) Left panel, sample trace of impulses elicited by a von Frey filament (13.7 mN) in an Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors (Aβ-LTMR). Middle, no impulse (top) was elicited by the heating-and-cooling from 30°C to 43°C and returning to 30°C (bottom). Right, no impulse (top) was elicited by the cooling from 30°C to 13.5°C (bottom). (g) Left panel, sample trace of impulses elicited by a von Frey filament (0.39 mN) in an Nav1.8-ChR2-negative Aβ-fiber mechanoreceptors (Aβ-LTMR). Middle, impulse (top) was elicited by the heating-and-cooling in the cooling phase (43°C to 30°C, bottom). Right, no impulse (top) was elicited by the cooling from 30°C to 13.5°C (bottom). (h) Fraction of Nav1.8-ChR2-negative Aβ-LTMRs showing no thermal response (gray), and response to heating-and-cooling with impulses at cooling phase of 43°C to 30°C. (i) Summary data of the impulse frequency elicited by the cooling from 43°C to 40°C, 40°C to 35°C, and 35°C to 30°C in the Nav1.8-ChR2-negative Aβ-LTMRs. Data present individual observations and/or mean ± SEM, *p < 0.05, ***p < 0.001, Student’s t-test or one-way ANOVA with the Tukey’s post-hoc test.

We examined the effects of heating and cooling temperatures on these Aβ-LTMRs and Aβ-HTMRs. All Aβ-HTMRs (n = 20/20, Figure 1(f)) and the majority of Aβ-LTMRs (15/21, not illustrated) showed no response to temperature changes from 30°C to 43°C and then returned to 30°C. These Aβ-HTMRs (20/20) and Aβ-LTMRs (15/21) also showed no response to the cooling from 30°C to 13.5°C (15/21). However, a subpopulation of Aβ-LTMRs (6/21, from five different animals) fired robust impulses during the cooling phase of the temperature changes from 30°C to 43°C and then returned to 30°C (Figure 1(g) and (h)). For these cooling-responsive Aβ-LTMRs, no impulses were elicited in the heating phase from 30°C to 43°C, and impulses were evoked only when the temperature dropped from 43°C to 30°C (Figure 1(g)). The Aβ-LTMRs fired robust impulses during the initial cooling phase, and then the firing frequency was gradually reduced during the later cooling phase (Figure 1(g)). Overall, for the Aβ-LTMRs that responded to the cooling from 43°C to 30°C, the impulse frequency was 2.7 ± 0.8 Hz (n = 6) from 43°C to 40°C, 0.7 ± 0.2 Hz (n = 6) from 40°C to 35°C, and 0.3 ± 0.2 (n = 6) from 35°C to 30°C. On the other hand, no impulse was evoked during the cooling from 30°C to 13.5°C in these cooling-responsive Aβ-LTMRs (n = 6, Figure 1(g)).

Discussion

In the present study, we used the hindpaw glabrous skin-tibial nerve preparations made from Nav1.8-ChR2 mice to investigate how Nav1.8-ChR2-negative and Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors responded to heating and cooling temperatures. We show that Nav1.8-ChR2-negative and Nav1.8-ChR2-positive Aβ-fiber mechanoreceptors are primarily Aβ-LTMRs and Aβ-HTMRs, respectively, consistent with our previous studies.^16,17 We have found that a subpopulation of cutaneous Nav1.8-ChR2-negative Aβ-fiber mechanoreceptors responded to the cooling from noxious heat of 43°C to 30°C with robust firing of AP impulses. Our finding indicates the presence of cooling-responsive Aβ-LTMRs in the glabrous skin, which may contribute to the perception of cooling for temperature drops from noxious heat to normal skin temperatures.

Cutaneous Aβ-LTMRs in the hindpaws of mice are mainly Meissner’s corpuscles and Merkel cell-neurite complexes.^9–11,18 Meissner’s corpuscles respond to mechanical indentation with rapidly adapting (RA) impulses, whereas Merkel cell-neurite complexes respond to mechanical indentation with slowly adapting (SA) impulses.^9–11 In the present study, we used von Frey filaments to measure the mechanical threshold of cutaneous Aβ-fiber mechanoreceptors. We did not examine whether the cooling-sensitive Aβ-LTMRs are Meissner’s corpuscles (RA type) or Merkel cell-neurite complexes (SA type). However, the slowly adapting response to the cooling in the Aβ-LTMRs shown in our study may suggest that the cooling responses are mediated by Merkel cell-neurite complexes. Further study is needed to determine whether the cooling-responsive Aβ-LTMRs are Merkel cell-neurite complexes or Meissner’s corpuscles.

Previous studies have shown that sensing cooling temperatures in the skin is mainly mediated by the TRPM8 channel, a sensor of cooling temperatures expressed in a subpopulation of primary afferent fibers.^5,6 TRPM8 channels are usually activated by a temperature drop below 28°C, and are believed to be involved in the perception of both pleasant cooling and noxious cold. ⁷ Our study showed that the Aβ-LTMRs responded to the cooling temperatures from 43°C to 30°C. In this cooling temperature range, TRPM8 should not be activated. Furthermore, TRPM8 channels are predominantly expressed in cutaneous C-fibers,^5–7 and were not found to be expressed in Aβ-fibers. Therefore, the response of Aβ-LTMRs to the cooling from 43°C to 30°C in the present study is unlikely to be mediated by TRPM8 channels. Although the cooling-responsive Aβ-LTMRs have never been reported previously, previous studies have reported that some C-LTMRs and Aδ-LTMRs could respond to cooling temperatures. For example, it has been reported that some C-LTMRs responded to cooling temperatures from 52°C to 30°C. ²² It has also been reported that some Aδ-LTMRs could respond to cooling with a temperature drop from 40°C to 32°C. ²³ Our study on cooling-responsive Aβ-LTMRs and the previously reported cooling-responsive C-LTMRs and Aδ-LTMRs raises a question about the molecular mechanisms underlying the sense of cooling for the temperature drop from high temperatures, such as noxious heat, to normal skin temperatures. Future studies are needed to address this question and help us fully understand the sense of cooling.

The physiological function of cooling-responsive Aβ-LTMRs in the context of behavioral impact in animals is currently not clear. A previous study on cooling-responsive C-LTMRs has suggested that their cooling sensitivity plays a significant role in thermotaxis. ²⁴ It is possible that cooling-responsive Aβ-LTMRs are also involved in thermotaxis in animals. The ability of sensing cooling temperature gradients, for example, from 43°C to 30°C, by Aβ-LTMRs and other LTMRs may promote animals to move from a highly stressful heating environment to a relatively cooling temperature that is thermally comfortable.