mGluR6 regulates keratinocyte phagocytosis by modulating CaM KII/ERK/MLC signalling pathway

Melanosome transfer from melanocytes to keratinocytes is crucial for the protection of the skin against harmful ultraviolet radiation (UVR). The process of melanosome transfer from melanocytes to keratinocytes has been widely studied. It has been suggested that keratinocyte phagocytosis plays an important role in this process, including direct phagocytosis of the dendritic tips of the melanocytes and endocytosis of extracellular melanosomes released from melanocytes.1 Glutamate is the major excitatory neurotransmitter in the mammalian central nervous system (CNS) via a number of glutamate receptors subdivided into ionotropic receptors and metabotropic receptors. Recently, in peripheral tissues, especially in epidermis, glutamate receptors, are proved to be involved in cutaneous barrier homoeostasis, terminal differentiation, proliferation and inflammation. Moreover, stimulation of mGluR6, one of group III metabotropic glutamate receptors, could alter morphology of melanocyte dendrites.2 In the other hand, previous reports have demonstrated the requirement for melanocyte filopodia in melanosome transport to keratinocytes.3 Keratinocytes engagement with the microsphere occur through association with fine filopodia and membrane ruffles.4 Therefore, we consider that mGluR6 may involve in keratinocyte phagocytic process. There was almost no report showing that mGluR6 was associated with keratinocyte phagocytosis. Therefore, we attempted to explorer weather mGluR6 involved in keratinocyte phagocytic process and its molecular mechanism. Detailed methods are described in Data S1. HaCaT cells were infected by a lentivirus which contained mGluR6 shRNA or an negative vector (NC shRNA). Following stably infection, the cells were collected for protein expression analysis by western blot (Fig. 1a). Compared with uninfected Blank cells and NC shRNA cells, the protein level of mGluR6 was reduced significantly in mGluR6 shRNA cells (Fig. 1b). To explorer weather mGluR6 knockdown leads to cell morphological changes, we tested the number of filopodia in HaCaT cells by scan electron microsphere (SEM) (Fig. 1c). The mGluR6 shRNA cells showed a marked loss of filopodia (29.25 ± 4.92) compared with NC shRNA cells (58.50 ± 2.89) (Fig. 1d). Then, the distribution of cytoskeleton (F-actin) in three groups was visualized by scanning laser confocal microscope. The cytoskeleton was located mainly at cell periphery in Blank or NC shRNA cells, while the mGluR6 shRNA cells represented reorganization of F-actin cytoskeleton (Fig. 1e). NC and mGluR6 shRNA cells were incubated with BSA-coated fluorescent microspheres for 16 h, respectively. The mGluR6 shRNA cells showed significantly decreased fluorescent microsphere uptake, compared with NC shRNA cells (Fig. 2a). Quantitative analysis of fluorescence microsphere uptake by HaCaT cells represented that the rate in mGluR6 shRNA cells was significantly decreased (35.6 ± 2.5)%, compared with NC shRNA cells (43.8 ± 1.3)% (Fig. 2b, c). With these data, we demonstrate the inhibitory effect of mGluR6 on keratinocyte phagocytosis. To explore the molecular mechanism of keratinocyte phagocytosis, we examined the expression of T286/287-phosphorylated CaM KII, total CaM KII, phosphorylated ERK, total ERK, T18/S19-phosphorylated MLC and total MLC by western blot (Fig. 2d). The relative expression of phosphorylated CaM KII, ERK and MLC was significantly decreased in mGluR6 shRNA cells, compared with Blank and NC shRNA cells (Fig. 2e). These data suggested that mGluR6 may regulate keratinocyte phagocytosis via CaM KII/ERK/MLC signalling pathway. mGluR6, one of group III metabotropic glutamate receptor, has been well documented ever since Nakajima et al. investigated its existence in rat retina.5 Then, many studies explicitly demonstrate that mGluR6 is essential for synaptic transmission of ON-bipolar cell responses in both the rod and cone systems. Subsequently, mGluR6 has been proved to exist in non-neural cells, such as prostate cancer cells and cornea endothelium.6 Recently, a report demonstrated that melanocytes expressed mGluR6 and activation of mGluR6 could increase melanin content in human melanocytes.2 But weather mGluR6 plays a role in keratinocytes remains unknown. In this study, we show that mGluR6 is expressed in human keratinocyte cell lines – HaCaT cells, which has not been reported previously. Actin remodelling leads to the formation of plasma membrane extensions, which induces formation of pseudopods that surrounded the particle, and then, fusion of the pseudopods gives rise to phagosomal vesicles, which detaches from the plasma membrane and are internalized. Previous reports have been provided strong support for the involvement of melanocyte filopodia in melanosome transfer, and keratinocyte filopodia are the important conduit for melanin between adjacent keratinocytes.4 Our data suggest that knockdown of mGluR6 leads to cytoskeleton reorganization and reduces the number of keratinocyte filopodia, resulting in inhibition of phagocytic activity. Previous studies have been showed that stimulation of mGluR6 could induce calcium influx in human melanocytes.5 The increase in intracellular calcium-mediated downstream events is via the activation of calmodulin-dependent kinases (CaM kinases), an upstream activator of ERK signalling pathway.7 The inhibition of melanosome uptake in human epidermal keratinocytes is regulated by PI3k/Akt and MEK/ERK signalling pathways.8 The myosin light chain (MLC) protein, as a downstream of ERK, has been proved to be involved in the regulation of phagocytosis.9 In this report, we have been observed that the activity of CaM KII, ERK and MLC is inhibited by mGluR6 knockdown. In conclusion, we hypothesize that mGluR6 may regulate keratinocyte phagocytosis via CaM KII/ERKMLC signalling pathway. It is necessary to understand the mechanism of melanosome transfer in pigmentation disorders. We would like to thank Prof. Vijayasaradhi Setaluri for the generous gift of the plasmid for mGluR6. Siwen Zheng performed in experiments and prepared the manuscript; Zhe Sun, Jing Ni, Zhen Li, Yang Sha and Ting Zhang performed experiments; Song Qiao and Guangming Zhao conceived modelling experiments; Zhiqi Song conceived and designed experiments. This work was funded by National Natural Science Foundation of China, 81472865, 81171491. No conflict of interest to declare. Data S1 Supplementary Materials and Methods Data S2 Supplementary References Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.