I am a
Home I AM A Search Login

A Paradox of Thermosensation in Mice: Perceiving Warmth Requires a Cool-Activated Channel

The menthol receptor TRPM8 is necessary for mice to perceive very small skin temperature increases as warming; results support pattern theory of thermosensation

by Francie Moehring


3 June 2020


PRF News

ThermosensationFeatured

The menthol receptor TRPM8 is necessary for mice to perceive very small skin temperature increases as warming; results support pattern theory of thermosensation

Can mice perceive slight rises in temperature as warming? A new study says yes, and asserts that patterns of sensory afferent activity, rather than dedicated cold and warm pathways known as labeled lines, drive thermosensation.

 

Researchers led by Gary Lewin and James Poulet, Max Delbrück Center for Molecular Medicine, Berlin, Germany, show that mice, just like humans, can perceive 1°C warming of the skin and discriminate warm from cool. Further, a combination of electrophysiological, pharmacological, and genetic knockout experiments show that, paradoxically, mice that lacked the cool-sensitive ion channel TRPM8 were unable to perceive warming.

 

“What is really impressive is that the authors were able to train these mice to discriminate warm and cool temperatures, as this is a very difficult behavior assay,” said Mark Hoon, National Institutes of Health (NIH), Bethesda, US. “Together with the electrophysiological recordings, this is a really nice study that builds on previous work and sheds further light on temperature perception,” according to Hoon, who was not involved with the new research.

 

The work was published online June 3, 2020, in Neuron, along with an accompanying Preview by Ana Gómez del Campo and Félix Viana, Universidad Miguel Hernández-CSIC, San Juan de Alicante, Spain.

 

A learning process: mice can report warming stimuli

In a previous collaboration, Lewin and Poulet detailed some of the neural circuitry underlying cooling perception in mice (Milenkovic et al., 2014). When Poulet’s graduate student, Ricardo Paricio-Montesinos joined the lab, he was interested in investigating whether mice could also be trained to perceive warm stimuli.

 

To find out, Paricio-Montesinos, along with fellow co-first author Fred Schwaller and colleagues, used a goal-directed thermal perception task they had employed in their past work. During this task, mice are trained to respond to a thermal stimulus by licking a sensor. The stimulus is applied to the forepaw via a Peltier element (a device that can produce heating or cooling), and the animals receive water as a reward. In the current study, the mice received water if they licked the sensor between the onset of warming stimuli and a re-cooling phase.

 

The authors first used the same size Peltier device (3 millimeter (mm) x 3 mm) as in their previous cold perception study in which mice had learned to discriminate a cooling stimulus within two training sessions. But, “to our surprise, we found it very difficult to train mice to discriminate warm stimuli as compared to cold stimuli. It was not until we increased the area of the stimulus that the animals were able to detect the warming of the Peltier device,” said Paricio-Montesinos.

 

Indeed, when the researchers used a Peltier device more than double the size (8 x 8 mm), which covered nearly all of the animal’s forepaw, they were able to train the mice to report warming within three to four training sessions. That a larger surface area was needed suggested that spatial summation was required for warm perception in the animals, as is the case in people (Stevens et al., 1974).

 

After initially using warming stimuli of 10°C (the baseline temperature of the Peltier element was 32°C), the authors next saw that the mice were able to report a warming stimulus of just 1°C (from 32°C to 33°C). This showed that the animals had a perceptual threshold for warm that is similar to that of people (Frenzel et al., 2012Stevens and Choo, 1998).

 

“This is a great example of what our behavior assays should look like in the future for the pain research field, which currently is focused mainly on investigating simple withdrawal responses,” said Hoon.

 

Afferents activated by warm and cool

Following these behavior assays, Schwaller, a postdoc in the Lewin lab, was interested in determining which populations of cutaneous sensory neurons conveyed the warming information to the central nervous system. To do so, he used a forepaw ex vivo skin-nerve preparation previously developed in the Lewin lab (Walcher et al., 2018) which allowed the authors to directly record from sensory nerve afferent terminals in the glabrous (smooth) skin, the site tested behaviorally in the animals.

 

Receptive fields were stimulated using 1°C per second heating (32°C to 48°C) and cooling (32°C to 12°C) ramps. Fibers activated by warm increased their firing rate as the temperature increased, which the authors called warm-activated afferents. Most of these fibers could be classified as polymodal C-fibers, exhibiting a slow conduction velocity, having little to no spontaneous ongoing activity, and showing strong activation in response to mechanical stimuli.

 

Sensory afferent fibers in the skin can be classified by the type of stimuli that activate them. The authors found that most of the fibers they recorded from were activated by mechanical and heat stimuli, termed C-mechanoheat (C-MH) fibers. The second most abundant fiber type responded to mechanical and cold stimuli (C-mechanocold, or C-MC), closely followed by fibers that responded to mechanical, heat, and cold stimuli (C-mechanoheatcold, or C-MHC). Two C-fibers had no mechanosensitivity and were classified as C-cold fibers. Most of the C-fibers responded to non-noxious temperatures.

 

Warm-inhibited afferents

The above experiments had all been performed with a bath temperature of 32°C, which is the temperature of the skin of a human hand. However, thermal imaging of awake mice revealed something the investigators hadn’t anticipated: The forepaw skin temperature was between 26°C and 28°C. So the authors again made recordings from the forepaw afferents, this time keeping the bath temperature at 27°C. Once again, most heat- and cold-sensitive fibers were polymodal C-fibers. However, to their surprise, the group also saw a new population of C-fibers.

 

“These polymodal C-fibers exhibited ongoing spike activity in the absence of any externally applied thermal stimuli. And, to our surprise, their activity actually decreased when we applied warming stimuli, and increased when we applied cooling stimuli. We termed those fibers warm-inhibited afferents,” explained Schwaller.

 

Interestingly, the group also found that the animals were better at perceiving warm when starting from lower baseline temperatures (22°C instead of 32°C). In this instance, the mice learned the thermal perception task more quickly and could detect warming of just 0.5°C. Further, at 22°C, the group saw only weak and sparse excitation of warm-excited C-MH and C-MHC afferents.

 

A surprise: warm-sensitive TRP channels contribute to warm sensation but are not absolutely required

Considering the known role of individual TRP channels in thermosensitivity, the authors next performed the same goal-directed thermal perception task as in their previous experiments, this time in mice lacking specific TRP channels.

 

First, remarkably, knockout animals missing TRPV1, which is activated by temperatures greater than 43°C, perceived warm similarly to wild-type animals; they were able to learn to report non-painful warm stimulation, and the time it took for them to show a licking response was similar.

 

Next, the authors used knockouts missing TRPM2, a channel that is activated by temperatures of about 34-42°C (Tan and McNaughton, 2016). These mice also learned to report non-painful warm temperatures, but their learning performance was impaired compared to wild-type animals. Furthermore, the TRPM2 knockouts exhibited 2°C higher warm perceptual thresholds compared to wild-type.

 

Third, the authors took advantage of triple knockout animals they had created for a previous study (Vandewauw et al., 2018, and also see PRF related news). These animals lacked TRPV1, TRPA1, and TRPM3, and were unable to sense acute noxious heat in that prior work. But in the current study, another unexpected finding appeared.

 

“To our surprise, these animals learned to report warming stimuli, and most can sense small amplitudes of warming, similar to our wild-type animals. This suggests to us that these warm-sensitive TRP channels are involved in, but are not essential for, warm perception,” said Poulet.

 

A paradox: warm perception requires a cold-sensitive TRP channel

Lastly, the authors turned to TRPM8, which responds to cooling. They were in for another unexpected finding when they looked at knockouts missing this channel.

 

“These were supposed to be our control experiments, so you can probably fathom how surprised we were by the fact that these animals could not report the warming stimulus,” Schwaller said.

 

The authors also used a TRPM8 antagonist to acutely inhibit the channel after the animals had learned to report the warming stimulus. Animals treated with the inhibitor had impaired warm detection responses, which was reversible, as those responses returned to baseline a day later. Overall, the findings suggest that TRPM8 is required for warm sensation in mice.

 

“I think the addition of the pharmacological approach to genetic knockout studies is very important, because the knockout studies are not as conclusive to me, since they are like asking a blind person to perform a visual task. If a mouse is very thirsty, it will find a way to get the water reward, even if it cannot necessarily discriminate the temperature. To me, the pharmacological approach is more convincing because one can more easily make the link between a correct response to the thermal task to get the water and the role the channel plays in the detection of the thermal stimulus,” said Hoon.

 

Hoon also pointed to both consistencies and inconsistencies of the current data with previous data from his group.

 

“This data matches some previous data from my lab, where we showed that TRPM8 was required for warm preference, as was TRPV1,” Hoon said (Pogorzala et al., 2013). “In another study we used a similar behavior assay to what the authors presented here, and we showed that pharmacological inhibition of TRPV1 impairs warmth discrimination,” Hoon added (Yarmolinsky et al., 2016). “I think the difference between our studies is that we used a pharmacological inhibition approach, and they utilized a TRPV1 knockout. Therefore, I would like to see the pharmacological inhibition of TRPV1 by the authors in order to support their conclusion that TRPV1 is not necessary for warm perception.”

 

Patterns or labeled lines?

To make sense of their puzzling behavioral results, the current study authors returned to their ex vivo recordings, this time in the triple (TRPV1, TRPA1, and TRPM3) knockouts as well as in the TRPM8 knockout mice. The proportions of thermosensory fiber subtypes were similar between wild-type and the triple knockout animals. Meanwhile, as anticipated, the TRPM8 preparation showed a loss of fibers responding to cool.

 

“In the afferent recordings from TRPM8 knockout animals, we saw the expected loss of cool-sensitive fibers, an absence of warm-inhibited afferents ─ and warm-activated afferents were present,” said Lewin. “This suggests that warm perception requires the silencing of warm-inhibited afferents and the activation of warm-activated afferents, although the activation of the warm-activated afferents alone is insufficient for warm perception.”

 

“Overall, our data shows that the thermal sensory system might not encode information based on the labeled line theory, but rather that innocuous temperature changes might be encoded via a patterned system of warm-activated and warm-inhibited afferents,” said Paricio-Montesinos.

 

As for future studies, according to Poulet, “When you touch an object, its temperature is really important in creating an accurate sensation. My lab is interested in understanding how the sense of temperature is combined with touch in the brain. However, as a first step, we need to find out where and how warm and cold are represented in higher-order areas like the thalamus and cortex, as these pathways are less clear than for touch.”

 

Meanwhile, Lewin’s lab wants to further understand polymodality. “I would love to design an experimental paradigm that allows us to investigate polymodality – to explain why it exists,” Lewin said. “For example, why do some C-fibers respond to heat but also to painful stimuli? At the moment we do not understand why the same primary afferent would respond to these different stimuli.”

 

Francie Moehring is a freelance writer based in Milwaukee, US.

Share This