All animal experiments were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee at the University of Michigan. Trpv3^G573S mutant animals were obtained from Dr. Takeshi Yoshioka, Shionogi Discovery Research Laboratories, Japan. We re-derived the Trpv3^G573S mutants using 129S1/SvlmJ eggs; the animals obtained were then backcrossed to 129S1/SvlmJ background for six-eight generations. These backcrossed 129S1/SvlmJ Trpv3^G573S mutants and control littermates were used for all the experiments. Animals were housed at room temperature with ad libitum access to standard lab mouse pellet food and water. A 12 h light/12 h dark cycle was maintained in the animal housing area. Littermates of both sexes were used in the experiments. Prior to the behavioral evaluation, animals were habituated in the testing apparatus for 20–30 min for three to 5 days.

Data shown is represented as mean ± SEM. Statistical analysis was performed using the Prism (GraphPad). Statistical significance between different experimental groups was calculated using unpaired Student’s t-test; p < 0.05 was considered statistically significant.

The TRPV3 channel has been documented to contribute to skin and hair architecture. Mice carrying Trpv3^G573S mutation showed abnormal hair coat with dramatically reduced hair length and diameter resulting in the phenotype of nearly naked skin (Figure 1A). The atypical activity of TRPV3 channels leads to greater epidermal thickness in the layer of stratum corneum (hyperkeratosis), granular layer (hypergranulosis), stratum spinosum, and stratum basale (Figure 1B), consistent with previous studies (Asakawa et al., 2006; Yoshioka et al., 2009). To understand the pattern of innervation of the Trpv3^G573S mutant, we tested the skin sections for β-tubulin, a pan-neuronal marker that can also label the sensory nerve terminals in the skin. Following a defined assessment method for counting epidermal nerve fibers (Ebenezer et al., 2007), we evaluated the number of all the nerve terminals in the epidermis (Figure 2A). Interestingly, we observed decreased innervation in the epidermis of the Trpv3^G573S mutants in comparison to control (Figure 2B). The unmyelinated polymodal C-afferents constitute a major population that innervates the skin epidermis and mediates nociceptive, pruriceptive, and temperature-sensitive somatic responses. The C-fibers can majorly be marked by the expression of Calcitonin Gene-Related Peptide (CGRP) and Isolectin B4 (IB4). To apprehend whether the decreased innervation in Trpv3^G573S mutants is due to the loss of small diameter C-subtype sensory neurons or because of mere loss of epidermal C-afferents, we counted the numbers of peptidergic (marked by CGRP) and non-peptidergic neurons (marked by IB4) in Trpv3^G573S mutants. The quantified data from L3 and L4 DRG showed no significant differences in the number of CGRP- and IB4-positive neurons in Trpv3^G573S mutant animals in comparison to control (Figure 2C).

To inquire how such a dramatic increase in skin thickness and decreased density of epidermal nerve innervation in these Trpv3^G573S mutants can affect different somatosensations, we evaluated these mutants for various somatosensory behaviors.

Initially, we tested sensory-motor coordination in the Trpv3^G573S mutant and found it to be unchanged in comparison to control littermates (Figure 3A). Also, we tested the sensation of touch in these mutant animals. A soft brush was stroked from heel to toe direction three times with an interval of 10 s between each trial, and we found no evident difference between control and Trpv3^G573S mutant (Figure 3B). Hence, the hyperactivity of the TRPV3 channel does not have any adverse effect on sensory-motor coordination and touch sensation.

The physiological contribution of TRPV3 in pain signaling is still not completely understood. Using the gain-of-function TRPV3 mice model, we tested mechanical pain behavior in Trpv3^G573S mutants using the von Frey and pin-prick assays. Trpv3^G573S mutants showed an increased mechanical threshold for punctate mechanical stimulation determined by the von Frey assay (Figure 3C). Additionally, we noted a trend of decreased response to a sharp pain in the mutant animals in comparison to control littermates in the pin-prick test (Figure 3D). Hence, we observed that the hyperactive TRPV3 channel could lead to desensitization to sharp and punctate mechanical pain.

Next, we examined the role of TRPV3 in mediating noxious heat. We tested the response threshold by thermally stimulating the glabrous skin of the hind paw by radiant heat (Hargreaves Test). There was no difference in response latency between Trpv3^G573S mutant and control littermates in the Hargreaves test (Figure 4A). At higher temperatures, such as 46, 50, and 54°C, the latency of the hind paw lick was unchanged in the Trpv3^G573S mutant compared to control (Figures 4B–D). We also tested Trpv3^G573S mutants for acetone-induced cooling and 0°C-evoked cold responses. We observed that Trpv3^G573S mutants had increased latency to respond to acetone and 0°C cold plate (Figures 4E,F). To study the molecular mechanism of heat sensing in the skin, we stained TRPV1⁺ fibers in control and mutant mice. We found that TRPV1⁺ fibers were mainly located in the border regions between the epidermis and dermis (Supplementary Figure 1A). We detected rare TRPV1⁺ intraepidermal fibers (Supplementary Figure 1B), consistent with a previous study (Hsieh et al., 2012). Interestingly, there was no change in the innervation densities of TRPV1⁺ fibers in the Trpv3^G573S mutants when compared to control mice (Supplementary Figures 1C,D), which might be the reason why behavioral responses to noxious heat remain unaffected. Our results show that hyperactive TRPV3 channels did not alter the threshold or latency for sensing noxious heat. However, Trpv3^G573S mutants do show deficits in sensing cool and cold sensations.

A gain-of-function mutation in Trpv3 is implicated in the development of dermatitis and spontaneous itch in conventional settings. However, in the absence of a specific pathogen, S. aureus, these mutant animals neither develop dermatitis nor show spontaneous scratching (Haraguchi et al., 1997; Asakawa et al., 2006; Imura et al., 2009). In our experiments, Trpv3^G573S mutants were raised in S. aureus-free environment, and we did not observe any dermatitis or spontaneous scratching in the Trpv3^G573S mutants (Figure 5A). To further examine the role of overactive TRPV3 channels in itch sensations, we tested a range of pruritogens-induced acute chemical itch behaviors in Trpv3^G573S mutation mice. Compound 48/80, chloroquine, PAR2 agonist SLIGRL-NH2, or Me-5-HT was injected subcutaneously in the nape of the animals, and scratching bouts were counted in control and Trpv3^G573S mutants. With the administration of different pruritogens, Trpv3 mutants responded with dramatically reduced scratching bouts in comparison to control (Figures 5B–E). To test mechanical itch, a weak mechanical force was delivered on the shaved skin region behind the ear, and resultant scratching was counted. No significant difference was noted in Trpv3 mutants in light touch-evoked pruritus when compared to control (Figure 5F). Hence, we found that the excessive activity of the TRPV3 channel is not sufficient to cause dermatitis, spontaneous scratching, or any difference in touch-evoked itch. However, it can result in a dramatic reduction in acute chemical itch. Our results suggest that overactive TRPV3 channels might not exert the primary role in itch transmission.

We observed that Trpv3^G573S mutant animals showed a deficit in sharp and punctate mechanical pain but no deficit in touch or sensory-motor coordination. In the thermosensation assays, we found that the hyperactivity of the TRPV3 channel did not change the latency to respond to radiant heat or the ability to sense intense noxious heat. Intriguingly, we observed a drastic degeneration of sensory nerve fibers in the epidermis in Trpv3^G573S mice. The observed attenuation in sensing cool, cold, acute mechanical pain and itch could be a result of decreased epidermal sensory innervations that mediates these modalities in naïve animals, whereas the sensory fibers involved in touch or heat could be the ones that remain functional in these mutants. Consistent with our behavioral readouts to noxious heat in Trpv3^G573S, we did observe that innervation densities of TRPV1⁺ fibers in the border regions between epidermis and dermis remain unchanged in Trpv3^G573S animals in comparison to control mice.

A recent study shows that keratinocytes are involved in nerve fiber degeneration in small nerve neuropathy patients. In patients suffering from small fiber neuropathy, axon guidance cue netrin-1 is highly expressed in patients’ keratinocytes which can reduce sensory neurite outgrowth (Kress et al., 2021). Hence, the hypoinnervation could be an indirect effect of TRPV3 hyperactivity in non-neuronal skin cells such as keratinocytes that closely monitor or affect the epidermal sensory terminals. Hence, it would be interesting to examine whether hyperactive TRPV3 channels in keratinocytes can also deregulate the expression of axon guidance cues or other molecules that could affect the milieu of the epidermal nerve terminals to cause sensory fiber degeneration. Another possible mechanism for nerve degeneration could be that the hyperactivation of TRPV3 channels in the nerve terminal may prime the nerve terminal to release cues to cause nerve retraction or depletion. Additionally, Trpv3^G573S mutants also show drastically increased levels of CCL11 (∼250-fold) (Yoshioka et al., 2009), and elevated levels of this chemokine have been reported in many neurodegenerative disorders (Huber et al., 2018). Thus, upregulated cytokine release might also contribute to degeneration of nerve terminals in the epidermis.

Our results suggest that epidermal nerve degeneration in Trpv3^G573S mice might contribute to the deficits in acute itch transmission. However, epidermal innervation of nerve fibers is increased in the AEW-model of chronic itch conditions in control, but no such hyperinnervation is observed in Trpv3^–/– upon AEW treatment (Yamamoto-Kasai et al., 2012). The authors did not find any difference in nerve fiber density in control and Trpv3^–/– mice under normal conditions (Yamamoto-Kasai et al., 2012). In chronic itch induced by AEW, various pathways come into action that precipitates to induce hyperinnervation of the epidermis and might require the participation of TRPV3 channels to induce nerve sprouting in the epidermis in chronic itch conditions. How activation of TRPV3 channels controls the growth of epidermal nerve terminals in control, and chronic itch conditions is a question that remains to be answered.

In the acute itch assays, we observed a decrease in scratching responses to various pruritogens in the Trpv3^G573S mutant in comparison to the control. Gain-of-function mutations in the Trpv3 gene, including the G573S point mutation, cause a form of palmoplantar keratoderma known as Olmsted syndrome, an autosomal dominant genodermatoses phenotypically characterized by palmoplantar keratoderma and periorificial keratotic plaques. In addition to these symptoms, not all, but some cases of Olmsted syndrome exhibit lesion-associated pain, and or itch, with varying degrees of intensity (Poulin et al., 1984; Lucker and Steijlen, 1994; Nofal et al., 2010; Lai-Cheong et al., 2012). However, only ∼22% of patients suffering from Olmsted syndrome suffer from pruritus, and ∼52% of patients suffer from pain (Mevorah et al., 2005; Tao et al., 2008). Hence a direct link between a gain-of-function mutation in Trpv3 and itch cannot be established.

The environmental pathogens and the background of the animal can play a very crucial role in constructing a unique immune profile that dictates the somatosensory responses, including itch. The notion that G573S gain-of-function mutation in Trpv3 does not unswervingly result in pruritus is also supported by the findings where DS-Nh animals harboring a spontaneous Trpv3^G573S mutation when raised in the S. aureus-free environment do not develop itch and/or dermatitis (Haraguchi et al., 1997; Asakawa et al., 2006; Imura et al., 2009). Consistently, we observed that Trpv3^G573S mice housed in the specific-pathogen free (SPF) conditions did not show spontaneous scratching behavior or develop any dermatitis at any time during their lifetime of a year (data not shown). Interestingly, when Nh mutant is introduced in a different background such as NC/Nga, animals harboring gain-of-function mutation in the Trpv3 gene do not develop any scratching or dermatitis even in the presence of S. aureus (Imura et al., 2009). Under conventional conditions, DS-Trpv3^G573S mutant animals exhibit an increased infiltration of mast cells in the skin, increased levels of IL2, IL-13, IL-17, eotaxin, G-CSF, GM-CSF, IFN-y, MCP-I, MIP-1α, IL-1α, IL-6, IL-9, IL-10, and IL-18 in comparison to Trpv3^G573S mutant animals raised in the SPF conditions (Imura et al., 2009). Contrastingly, Yoshioka et al. (2009) found that a transgenic Trpv3^G573S mutant on either C57BL or DS background shows increased scratching, but whether the mutants are raised in the SPF or conventional conditions is not specified. Additionally, TRPV3 antagonists such as forsythoside B and citrusinine-II are shown to attenuate acute and chronic itch and pain; however, these compounds have known (Jiang et al., 2012; Liu et al., 2019), or could have unknown off-targets that might result in suppression of itch (Zhang et al., 2019; Han et al., 2021). Also, no conclusive evidence is available that can indicate how an overactive TRPV3 channel can collaborate with genetic, environmental, and/or inflammatory triggers to induce itch in the patients harboring gain-of-function mutations in the Trpv3 gene. Additionally, patients suffering from Olmsted syndrome display genetic heterogeneity where different naturally occurring mutations in the TRPV3 gene (Wilson et al., 2015) show similar phenotypes along with clinical heterogeneity. The correlation between allelic and phenotypic heterogeneity in Olmsted syndrome is yet to be explored.

Patients suffering from Olmstead syndrome also show diversity in the severity of alopecia. Some human patients do show alopecia totalis, others show patches of hair loss, hypotrichosis, and/or brittle hair quality, and in rare cases, patients have normal hair (Lin et al., 2012; Wilson et al., 2015; Ni et al., 2016). TRPV3 is shown to control hair development both in humans and rodents. Agonist-mediated activation of TRPV3 in vitro by eugenol or 2-APB results in dose-dependent inhibition of human hair shaft elongation and facilitates the premature entry into catagen stage (Borbiro et al., 2011). In rodents, a hyperactive TRPV3 channel can induce an inefficient differentiation of follicular keratinocytes (Song et al., 2021), and the anaphase is recorded to be more persistent, whereas the telophase is absent in DS-Nh mice at P21 (Imura et al., 2007). In our mouse model, Trpv3^G573S mutants did have brittle hair that had drastically reduced hair diameter and length and were not completely hairless (Figure 1A).

In relation to skin thickening due to gain-of-function mutation, while Olmsted patients show diffuse palmoplantar keratoderma, which is an abnormal thickening of the skin, and Trpv3^G573S mutant mice also showed thickened skin. Ambient activation of TRPV3 can result in increased proliferation in keratinocytes in mice (Wang et al., 2021).

Thus, diverse phenotypic spectrum observed in humans and mice due to hyperactive TRPV3 channels observed could be attributable to species-specific differences, allelic or non-allelic genetic diversity, and/or environmental triggers, which could strongly influence the immune responses resulting in inter-and intra-species differences and the precise mechanism of phenotypic diversity in hair and skin between mice and humans is yet to be explored.