Signs of chronic itch in the mouse imiquimod model of psoriasiform dermatitis: sex differences and roles of TRPV1 and TRPA1.

The study was approved by the University of California, Davis Institutional Animal Care and Use Committee. Experiments were conducted using male and female wildtype C57BL6/J (B6) mice, as well as KO mice lacking TRPV1 (purchased from Jackson Labs) or TRPA1 (a generous gift from David Julius, University of California, San Francisco). The rostral back was shaved before the beginning of treatment. Imiquimod cream (5%; Aldara, 50 mg; 3 M Health Care Limited, UK) or vehicle (Vanicream; Pharmaceutical Specialties, Rochester, MN) was applied topically to the shaved area on the rostral back once per day for 5 consecutive days. Treatment groups consisted of age-matched male and female wildtype, TRPV1KO and TRPA1KO mice (n = 6 per group). Imiquimod treatment induced clear signs of skin pathology including skin scaling and erythema. Skin barrier function was assessed by measuring transepidermal water loss (TEWL; Vapometer, Delfin Technologies) before treatment began, and after the fifth day of imiquimod treatment. Spontaneous scratching behavior was measured before the first day of treatment (day 0), and again on treatment days 1, 3, and 5. All videotaping of spontaneous scratching, and assessment of alloknesis, was conducted 23 hours following imiquimod treatment the previous day. Mice were videotaped and tested between 10 AM and 5 PM, with each individual mouse tested at the same time each day. The mice were habituated to glass cylinders for 3 successive days before recording. Animals were videotaped from above for 30 minutes. Behavioral videos were analyzed by 2 blinded observers. Only discrete bouts of hindlimb scratches directed toward the application site were counted, as described previously [ 44 ] . Scratch bouts were measured and summed over the entire 30 minutes period. Alloknesis was assessed as previously described [ 45 ] . The mouse was placed in an enclosed area and a 0.07 g von Frey monofilament was applied to the perimeter of the imiquimod application area 5 consecutive times. The alloknesis score consisted of the number of immediately occurring hindlimb scratch bouts directed to the stimulus site. Hyperknesis was assessed by counting the number of hindlimb scratch bouts elicited by intradermal injection of chloroquine (100 μg/10 μL) in rostral back skin within the treatment area, 1 week before day 1 of imiquimod treatment, and again on day 4 of imiquimod treatment. The animals were habituated to glass cylinders and video recorded from above for 15 minutes before and 60 minutes following chloroquine injection. Scratch bouts were counted as described above. Within-group and between-group comparisons of TEWL, spontaneous scratch bouts, alloknesis scores, and chloroquine-evoked scratch bouts were statistically assessed using analysis of variance followed by a Bonferroni post hoc test, with P <0.05 considered to be significant. To compare sex differences within genotypes a linear mixed model that included an interaction between time, treatment, and sex was used in the lme4 package version 1.1–21 in R version 3.6.0 [ 46 ] .

All genotype groups receiving imiquimod treatment exhibited a significant increase in TEWL score as well as skin scaling ( Figs. 1B , ( D ) compared with vehicle controls ( Figs. 1A , C ). TEWL scores were significantly lower in both TRPV1KO and TRPA1KO mice compared with wildtypes, for both female and male groups ( Figs. 1B , D ). Application of vehicle did not result in any significant increase in TEWL in any genotype ( Figs. 1A , C ). Imiquimod treatment resulted in a significant increase in the number of spontaneous scratch bouts in wildtype males but not females when compared with vehicle controls ( Figs. 2A , B ). However, because of behavioral variability the effect size for imiquimod treatment failed to reach statistical significance when sexes were directly compared ( P = 0.0677). Male and female TRPV1KO mice receiving imiquimod treatment exhibited a significant increase over vehicle-treated controls in the number of spontaneous scratch bouts ( Figs. 2C , D ). There was a significant effect of imiquimod treatment when compared for sex of TRPV1 mice ( P = 0.0257). Curiously, imiquimod treatment resulted in a significant increase in spontaneous scratching in male ( Fig. 2F ), but not in female TRPA1KO groups ( Fig. 2E ), although the latter groups showed a nonsignificant trend toward increased scratching. There was no significant effect of sex on imiquimod treatment in TRPA1 mice when directly compared ( P = 0.3486). Before the start of imiquimod treatment, alloknesis scores were essentially zero in all mice ( Fig. 3 , Pre). All imiquimod-treated mice exhibited a significant increase in alloknesis scores. There were no significant differences in alloknesis scores among imiquimod-treated female wildtype and KO groups ( Fig. 3A ). In contrast, imiquimod-treated male TRPV1KO and TRPA1KO mice exhibited significantly lower alloknesis scores compared with wildtypes ( Fig. 3B ). When effects of imiquimod treatment were directly compared by sex, there was a trending yet nonsignificant effect for TRPA1KO and TRPV1KO ( P = 0.0598 and 0.585, respectively). Wildtype mice showed no significant sex differences in imiquimod-mediated alloknesis ( P = 0.4729). Before imiquimod treatment, intradermal injection of chloroquine elicited significant increases in the number of scratch bouts in all mice, including TRPA1KOs. Following treatment with imiquimod or vehicle for 4 days, chloroquine still elicited a significant increase in scratch bouts in all treatment groups, with no significant between-group differences ( Fig. 4 ). This indicates that hyperknesis did not develop as a result of imiquimod treatment and was independent of sex and genotype.

The present results confirm those of Sakai et al [ 15 ] in that imiquimod treatment of wildtype B6 mice resulted in a significant increase in spontaneous scratching, at least in males, as well as increased alloknesis. We extend these findings in several ways. We observed a sex difference for spontaneous scratching, and a trend toward a sex difference for alloknesis, following imiquimod treatment. Neither sex exhibited any significant hyperknesis. The roles of TRPV1 and TRPA1 were somewhat complex. Male and female TRPV1KO and male TRPA1KO mice exhibited significant increases in spontaneous scratching, and female TRPA1KO mice exhibited a tendency toward increased scratching which, however, did not reach statistical significance. Alloknesis scores were significantly lower in male, but not female, TRPV1KO and TRPA1KO mice, indicating that TRPV1 and TRPA1 contribute to the development of mechanical psoriatic itch in male, but not female, mice. Finally, following imiquimod treatment TEWL was significantly lower in both male and female TRPV1 and TRPA1KO mice compared with wildtypes, suggesting that these ion channels participate in imiquimod-induced degradation of skin barrier function. Male but not female wildtype mice exhibited a significant increase in spontaneous scratching following imiquimod treatment. TRPV1KO and TRPA1KO mice of both sexes showed increased spontaneous scratching following imiquimod treatment, although this did not reach statistical significance in female TPA1KO mice. These data, at least for male wildtypes, confirm the findings of Sakai et al [ 15 ] . However, others reported that male and female B6 wildtype and TRPA1KO mice exhibited variable hindlimb scratching following 4 days of imiquimod treatment that did not significantly differ from vehicle-treated mice [ 27 ] . This partly agrees with our present findings that female imiquimod-treated TRPA1KO mice did not exhibit a significant difference in scratching compared with vehicle-treated group, although the male TRPA1KO mice did. Kemény et al [ 27 ] also reported a significant increase in other nocifensive behaviors (biting, licking, flinching) in TRPA1KO mice, suggesting that TRPA1 negatively regulates the development of these behaviors in psoriasiform dermatitis. Both sexes exhibited significant increases in alloknesis scores following imiquimod treatment. Female TRPV1KO and TRPA1KO mice showed imiquimod-induced increases in alloknesis scores that did not differ from wildtypes, whereas male TRPV1KO and TRPA1KO mice showed significantly lower alloknesis scores following imiquimod treatment. These data represent a sex difference, and suggest that TRPV1 and TRPA1 are required for the development of alloknesis, i.e. are positive regulators, in this model for males but not females. In contrast to the fairly minor differences in itch behavior between the wildtype and KO mice observed presently, we have observed significant reductions in key markers such as skin thickness, transepithelial water loss (TEWL) and pathogenic cytokines characteristic of psoriasiform inflammation [ 47 , 48 ] . The significant reduction in imiquimod-induced skin changes implies a positive regulatory role of TRPV1 and TRPA1 in the manifestation of imiquimid-induced psoriasiform dermatitis, but less so in the behavioral signs of chronic itch. Our findings contrast with a recent report that signs of imiquimod-induced psoriasiform dermatitis (dorsal skin blood flow, skin thickness, Munro microabsesses, inflammatory cytokine levels) were significantly enhanced in TRPA1KO mice [ 27 ] . Part of the apparent dis-crepancy between our data and those of Kemény et al [ 27 ] may be attributed to methodological differences between the 2 studies, as discussed elsewhere [ 47 ] . Surprisingly, we did not observe an increase in chloroquine-evoked scratching following the imiquimod treatment. This is at odds with previous studies using mouse models of chronic dry skin itch [ 49 ] or contact dermatitis [ 50 ] , which both reported that nonhistaminergic mediators (serotonin, SLIGRL, BAM8–22), but not histamine, elicited significantly enhanced scratching behavior in the treated animals. Our results suggest that hyperknesis is not a major symptom of psoriasiform dermatitis, at least in mice. While itch from light tactile stimulation such as clothing is reported by psoriatic patients to be bothersome [ 19 ] , it would be interesting to test if mildly itchy stimuli elicit stronger itch in psoriatic patients using quantitative sensory testing. Previous studies have emphasized a role for TRPA1 in mediating acute nonhistaminergic itch [ 22 ] as well as certain types of chronic itch [ 23 ] in mice. However, our present data indicate that the nonhistaminergic itch mediator, chloroquine, elicited equivalent scratching in TRPA1KO mice compared with wildtypes of both sexes ( Fig. 4 ). It was recently reported that cutaneous afferent nerve fiber responses, as well as scratching behavior, elicited by intradermal injection of chloroquine, was not different between wildtype and TRPA1KO mice [ 51 ] , consistent with our present data. The latter study speculated that TRPA1 may play a role in itch at a site central to the primary afferent pruriceptive nerve endings in the skin whose response to chloroquine does not require TRPA1 [ 51 ] .