Microtubules are involved in melanosome redistribution from dispersed to clustered (centripetal transport) [[18]Evans R.D. Robinson C. Briggs D.A. Tooth D.J. Ramalho J.S. Cantero M. et al.Myosin-Va and dynamic actin oppose microtubules to drive long-range organelle transport.]. We first investigated whether microtubule polymerization could increase melanosome phagocytosis. Tubastatin A, a selective HDAC6 inhibitor, promotes microtubule polymerization by increasing acetylation. Nocodazole induces microtubule depolymerization. HaCaT cells incubated with fluorescent microspheres for 2 h were capable of phagocytizing the microspheres from the culture media. The fluorescence images showed aggregation of phagocytosed microspheres in DiO (green, Beyotime C1038) stained cell membrane background. Tubastatin A increased microsphere phagocytosis but nocodazole did not (Fig. 3a). Similarly, we observed that physiological doses of UVA/B also enhanced α-tubulin acetylation in HEKs, with Tubastatin A treated cells as the positive control (Fig. 3b). These results indicate that acetylation of α-tubulin may contribute to the phagocytic effect. Notably, higher CYLD protein expression was observed in HEKs than in HEMs and HaCaT cells (Fig. 3c). This may be related to the mechanism of melanosome transfer to the keratinocytes. To explore the effects of CYLD expression on keratinocytes, we performed co-immunoprecipitation experiments in HEKs. The results showed that CYLD interacted with HDAC6 (Fig. 3d) and TRPA1 (Fig. 3e). CYLD binding to HDAC6 increases α-tubulin acetylation. To further explore the association of CYLD and TRPA1, immunofluorescence experiments were performed, and we observed that the interaction between TRPA1 and CYLD is dynamic and depends on the activation status of TRPA1. In the absence of UVA/B, CYLD localized throughout the cytoplasm where it strongly colocalized with TRPA1, whereas in the presence of UVA/B and activated TRPA1, CYLD translocated to the perinuclear region (Fig. 3f). The activation of TRPA1 may be related to the role of CYLD and its downstream signaling pathways. To further confirm this effect, we conducted immunoprecipitation experiments. The results showed that in the presence of JT010, a TRPA1 specific agonist, the binding of CYLD to acetylated α-tubulin was more prominent than that of unmodified α-tubulin (Fig. 3g). These data indicate that when TRPA1 is activated, CYLD translocates to the perinuclear region, stabilizes microtubules, and promotes melanosome phagocytosis by competitively binding to acetylated microtubules.

Fig. 3TRPA1 activation indirectly stabilizes microtubules by promoting the competitive binding of CYLD and acetylated α-tubulin. (a) The inverted fluorescence image of HaCaT cells stimulated for 30 min with 100 nM Tubastatin A (deacetylase inhibitor) or 50 nM Nocodazole (microtubule depolymerization), followed by incubating with fluorescent microspheres (red) for 24 h and stained for cell membranes (green) with DiO. Scale bar = 25 µm. (b) HEKs were treated with UVA 225 mJ/cm2, UVB 25 mJ/cm2 or 100 nM Tubastatin A. Expression of AC (acetylation) α-tubulin was tested by western blot using β-Actin as a loading control. The quantification of protein abundance was shown. (c) CYLD protein expressed in HEKs, HEMs and HaCaT cell line by western blot analysis. The protein in line 2 was not denatured (the HEMs lysate was not boiled). GAPDH was used as a loading control. The quantification of protein abundance was shown. (d) HEKs were immunoprecipitated with antibodies against HDAC6 and immunoblotted against CYLD and (e) HEKs were immunoprecipitated with antibodies against TRPA1 and immunoblotted against CYLD. The input shows equal amount of protein used for immunoprecipitation. (f) TRPA1 (red) and CYLD (green) were stained by immunofluorescence staining in HEKs untreated or treated with UVA (225 mJ/cm2) or UVB (25 mJ/cm2) detected by fluorescence confocal microscope. Scale bar = 25 µm. (g) HEKs were untreated or treated with UVA or JT010 in advance, followed by immunoprecipitation of endogenous CYLD and immunoblotting against AC α-microtubule/α-microtubule. The input shows equal amount of protein used for immunoprecipitation. All experiments in this figure were performed three times with comparable results. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.