Conditional immortalization establishes a repertoire of mouse melanocyte progenitors with distinct melanogenic differentiation potential.
Ke Yang,Jin Chen,Wei Jiang,Enyi Huang,Jing Cui,Stephanie H Kim,Ning Hu,Hong Liu,Wenwen Zhang,Ruidong Li,Xiang Chen,Yuhan Kong,Jiye Zhang,Jinhua Wang,Linyuan Wang,Jikun Shen,Hue H Luu,Rex C Haydon,Xiaohua Lian,Tian Yang,Tong-Chuan He
DOI: https://doi.org/10.1038/jid.2012.145
2012-01-01
Abstract:TO THE EDITOR Epidermal melanocytes (MCs) are specialized melanin-producing cells (Slominski et al., 2004; Yamaguchi et al., 2007; Thomas and Erickson, 2008; Park et al., 2009) that synthesize melanin within the melanosome (Slominski et al., 2004) and protect individuals from harmful UV rays (Slominski et al., 2004; Yamaguchi et al., 2007; Thomas and Erickson, 2008; Park et al., 2009). Defects in or a lack of MCs can lead to melanoma, pigment disorders, and auditory defects. MC proliferation and differentiation in skin is tightly linked to hair regeneration cycles. MCs in vertebrates are derived from neural crest (Thomas and Erickson, 2008). In hair follicles, melanoblasts are segregated into hair matrix MCs (for hair pigmentation) and melanocyte stem cells (MCSCs). The discovery of MCSCs and induced pluripotent (iPS) cells provides important resources to elucidate mechanisms underlying melanogenesis and pathogenesis of MC-related disorders (Lin and Chuong, 2011; Nishikawa-Torikai et al., 2011; Nishimura, 2011; Ohta et al., 2011; Yang et al., 2011). Although MCSCs and iPS cells are important MC precursors for mechanistic studies, their isolation and expansion are technically challenging. Here we investigate whether MCs can be immortalized without compromising melanogenic potential. By using primary MCs isolated from newborn mouse skin, we engineered >100 clones by introducing SV40 T antigen (SV40T), which is flanked with FRT sites (Supplementary Figure S1A online; Westerman and Leboulch, 1996). We found that although primary MCs grew slower after passage 5 immortalized MC (iMC) cells acquired high proliferative activity (Supplementary Figure S1B online); however, wide variations in proliferation were observed among clones. For instance, iMC23 grew faster than iMC65 (Figure 1a). We analyzed melanogenic markers in iMC clones. More than half of the analyzed clones expressed progenitor marker c-kit, whereas MC progenitor markers Pax3, Sox10, and MITF-m were readily detected in most clones (Supplementary Figure S2A online). Dopachrome tautomerase and tyrosinase-related protein 1 (TRP-1) were highly expressed in most iMC clones, whereas tyrosinase was highly expressed in about half of the clones (Supplementary Figure S2B and C online). It is conceivable that early melanogenic iMCs should express c-kit and early melagonic markers but not late markers (e.g., tyrosinase). Three iMC clones were chosen for further characterization: iMC23 (melanoblast progenitor-like) (Figure 1aA), iMC65 (late-stage melanocyte-like) (Figure 1aB), and iMC37 (intermediately differentiated melanoblast-like). These results indicate that SV40T-mediated immortalization can create a repertoire of iMCs with varied melanogenic potential. We also tested whether immortalization phenotypes were reversible by flippase recombination enzyme (FLP). By using adenovirus FLP (AdFLP) that co-expresses green fluorescent protein (GFP; He et al., 1998; Luo et al., 2007), we found that iMCs were effectively transduced by AdFLP or AdGFP, and SV40T expression was reduced in FLP-transduced iMCs (Figure 1bA and bB). iMC23 infected with AdFLP grew slower (Figure 1bC and bD), suggesting that the proliferative activity of iMCs may be reversed by removing SV40T. By using reporter pTyr-Gluc-expressing Gaussia luciferase (GLuc) driven by a 2.0-kb mouse tyrosinase promoter, we found that GLuc activity was increased by dexamethasone and was potentiated by removing SV40T with FLP (Supplementary Figure S3A and B online). We further analyzed spontaneous differentiation of iMCs by assessing endogenous melanin and tyrosinase activity. Cell pellets from iMCs exhibited varied amounts of melanin. iMC65 and iMC61 exhibited the highest level of melanin and iMC23 produced the lowest level, whereas a majority of iMC clones produced low-to-modest levels of melanin (Figure 1cA). Accordingly, quantitative analysis revealed that iMC65 and iMC61 exhibited high tyrosinase activity, and seven clones including iMC23 had low tyrosinase activity, whereas iMC37 exhibited modest-to-high levels of tyrosinase activity (Figure 1cB), consistent with the qualitative results shown in Figure 1cA. We characterized the early and late markers c-kit and HMB45, respectively, in iMC23, iMC37, and iMC65. HMB45 is a widely used mAb detecting melanocytic tumors (Gown et al., 1986). iMC23 expressed a high level of c-kit, whereas iMC37 and iMC65 were strongly stained with HMB45 (Figure 1dA). Fontana–Masson staining revealed the highest level of melanin in iMC65 and a modest level in iMC37, whereas a negligent level of melanin was observed in iMC32 (Figure 1dB). Thus, these clones may represent different stages of melanogenic differentiation. We further analyzed the melanogenic potential of iMC23. When iMC23 was treated with dexamethasone, tyrosinase activity increased in a dose-dependent manner (Figure 2aA). Dexamethasone induced tyrosinase and TRP-1 expression (Figure 2aB). Dose-dependent melanin production was visible in culture and cell pellet (Figure 2bA and bB), which was further confirmed by Fontana–Masson staining (Figure 2bC). These results indicate that iMC23 exhibits melanogenic potential. Finally, we determined whether iMC immortalization phenotypes were reversible in vivo. iMC23 and iMC65 tagged with firefly luciferase were infected with AdFLP or AdGFP, collected for subcutaneous injection into athymic mice, and monitored by Xenogen bioluminescence imaging (Caliper Life Sciences, Hopkinton, MA). AdGFP-transduced iMC23 yielded a stronger signal than that of AdFLP-transduced iMC65, whereas within the same lines AdFLP-transduced cells produced no signal compared with the AdGF-transduced iMC65 (Figure 2cA), suggesting that removing SV40T may reduce proliferation and survival of iMCs in vivo. Pigmentation of injected iMCs was visible through skin, as iMC65 exhibited a higher level of pigmentation than iMC23. AdFLP-transduced iMC23 yielded more pigmentation than AdGFP-transduced iMC23 (Figure 2cB). Histologically, iMC23 injection sites had higher cellularity than iMC65 sites, whereas in both lines cellularity was lower in AdFLP-transduced cells (Supplementary Figure S3C online). Production of melanin was more pronounced in AdFLP-transduced iMCs as indicated by HMB45 immunostaining and Fontana–Masson staining (Figure 2dA and dB). Overall, the in vivo results are consistent with these clones’ in vitro features. SV40T-immortalized MCs are non-tumorigenic under our experimental conditions. Interestingly, melanoma phenotypes have been reported by enabling neoplastic transformation of primary human MCs with SV40 early region, which encodes both SV40 T and t antigens, in conjunction with hTERT (Gupta et al., 2005). Nevertheless, we established a repertoire of conditionally immortalized MCs with varied melanogenic potential, ranging from melanoblast-like to well-differentiated phenotypes. Such repertoire of iMCs should be useful for understanding MC biology and raveling molecular pathogenesis of pigment cell disorders, including melanoma. An efficient method to isolate and expand cutaneous MCs is needed, as it enables us to better understand melanogenesis. Here we demonstrate that SV40T-mediated immortalization of MCs is simplistic, effective, and reversible. This approach should be considered an important alternative to the isolation and characterization of melanogenic stem cells. The authors state no conflict of interest. We thank Dr Philippe Leboulch of Harvard Medical School for the provision of SSR #41 vector. The reported work was supported in part by research grants from the National Institutes of Health (HHL and T-CH) and the Natural Science Foundation of China (81101205 to KY and 30901711 to J Chen). Author contributions KY: conception and design; manuscript writing; data analysis and interpretation; collection and/or assembly of data. J Chen and WJ: provision of study materials/reagents; collection and/or assembly of data. EH: provision of study materials/reagents; other (animal experiments and histology). J Cui: provision of study materials/reagents; collection and/or assembly of data. SHK: provision of study materials/reagents; data analysis and interpretation. NH, HL, WZ, RL, XC, YK, JZ, JW, and LW: provision of study materials/reagents. JS: others (histology); data analysis and interpretation. HHL and RCH: conception and design; manuscript writing. XL: conception and design; manuscript writing; data analysis and interpretation. TY: conception and design; manuscript writing; data analysis and interpretation; administrative support. T-CH: conception and design; manuscript writing; data analysis and interpretation. SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper