General strategy to analyse coat colour phenotypes in mice

Melanocytes are dendritic cells that produce melanin, the natural pigment responsible for the colour of skin, eyes, hair and skin appendages. Melanin synthesis, as brown/black eumelanin and red/yellow pheomelanin, occurs within a specialized lysosome-like structure called the melanosome (Orlow, 1995). Pigment synthesis by melanocytes not only produces the extensive and complex coloration patterns seen in mammals, but also protects the skin against the potentially harmful effects of light, by absorbing and scattering light and reducing the production of reactive oxygen species (Marks and Seabra, 2001; Miyamura et al., 2007). Laboratory mice present a vast diversity in coat colour, and mutations that affect mouse coat colour were among the first mutant alleles studied (Lamoreux et al., 2010). Understanding the molecular and cellular events in hair follicle biology, including the interaction between melanocytes and keratinocytes, is extremely helpful in characterizing and understanding the coat colour in mice (Nishimura, 2011; Schneider et al., 2009). The ease of observation and identification of a phenotype made mice a valuable and highly used model for studying gene action and interaction (Larue and Beermann, 2007). Different genetic factors may affect a certain phenotype and can be specific for a genetic background strain. Hence, coat colour characterization of genetically engineered mice concerns both the gene of interest and the genetic background on which the mutation is produced and maintained. Frequently used mouse strains, such as C57BL/6, are well characterized, and the effect of genetic background on the gene of interest can be analysed, controlled and better understood (Linder, 2006). The precise description of coat colour (amount and distribution of melanin in hairs, number and localization of melanoblasts and melanocytes) is crucial to define the main affected cellular and molecular mechanisms. The long history of mouse research has enabled the development of many strategies, methods and reagents to analyse coat colour phenotypes in mice, as briefly summarized later. Prior to determining the melanin content in mouse skin and hairs, a proper examination of the mouse coat colour and pelage hairs is required. To evaluate coat colour, (i) a thorough photographic documentation of a minimum of four mice, showing together the mutant and control siblings, can be easily obtained from dorsal, side and ventral views recorded at 2 weeks, 2 months and 1 yr. (ii) A minimum of 50 wild-type (wt) and mutant hairs (including awl, acheyne, zigzag and guard) can be plucked (Sundberg and Hogan, 1994), observed under the microscope and photomicrographed. Depending on the coat colour, hairs have to be plucked from the region(s) of interest and at different time periods to collect spatial and temporal information (Data S1) (Luciani et al., 2011; Russell, 1946). Additional information can be obtained from the study of vibrissae, cilia and hairs of the tail and ears. (iii) The overall amount of melanin in wt and mutant hairs provides information on hyper- or hypopigmentation (Data S2) (Larue et al., 1993). Once this general information is obtained, hypotheses can be raised on the presence/absence of melanocytes, the transfer/transport of melanosome and/or the production and quality of melanin in the mutant. (iv) Histological skin sections will provide information on the number and locations of melanocytes and surrounding cellular partners, including keratinocytes. Melanocytes are recognizable by their shape, location and pigment. Abnormal features may complicate this analysis. Melanin can be visualized using different techniques, including Fontana-Masson or Brissette stainings (Data S3) (Motoi et al., 1978; Weiner et al., 2007). Melanocytes can be visualized using a variety of markers. The marker S100 is sensitive and robust, but somewhat unspecific and should be combined with another marker such as Tyrp1, Dct or tyrosinase (Data S4). They can also be genetically labelled using mice in which a reporter gene is under the control of a lineage-specific melanocyte promoter (e.g. Dct::LacZ mice (Mackenzie et al., 1997)) or a combination of mice allowing lineage tracing using the Cre-LoxP or the rtTA system. Tyr::Cre or Tyr::CreERt2 mice produce the Cre recombinase protein in the melanocyte lineage from embryonic day 10.5, as do reporter mice (Rosa26 [producing LacZ after Cre recombination], Z/EG [producing EGFP after Cre recombination] or Tomato [producing RFP after Cre recombination]) (Delmas et al., 2003; Muzumdar et al., 2007; Novak et al., 2000; Soriano, 1999; Yajima et al., 2006). A parallel and elegant system was established by Merlino’s team (Zaidi et al., 2011a,b). This system allows the conditional and inducible expression of an autofluorescent protein in the melanocyte lineage. To do so, two mouse lines are used: Dct::rtTA and TRE::H2BeGFP. The reverse tetracycline-controlled transactivator gene (rtTA) is under the control of Dct, and its gene product is only active in the presence of tetracycline. The active transactivator binds and induces the TRE promoter, controlling a modified autofluorescent gene (H2BeGFP). (v) Melanocytes are naturally dispersed in the skin and represent a small fraction of the total number of cells. Therefore, it is difficult to generate and interpret chemical, biochemical and molecular information on melanoblasts/melanocytes when using the entire cutaneous organ as starting material. Melanoblasts can be isolated and studied during embryonic development (Data S5) (Colombo et al., 2011). The establishment of a pure population of melanocytes in culture (primary or immortal) allows for such an analysis (Data S6) (Bennett et al., 1989; Larue and Mintz, 1990; Tamura et al., 1987). The amount of melanin (Data S7) and tyrosinase activity (Data S8) in these cells can be determined (Larue and Mintz, 1990; Townsend et al., 1984). Furthermore, mRNA and protein levels of tyrosinase-associated genes (Dct, Tyrp1, M-Mitf and Brn2) can easily be determined by RT-qPCR (Data S9). The level of expression and phosphorylation status of key regulatory proteins, such as Akt, Gsk3β, ERK and β-catenin, can be determined by Western blot analysis and immunofluorescence. Finally, the distribution and migration of melanosomes can be investigated by 2-3D microscopy and time-lapse analysis of melanocyte culture (Data S10). During development, the number and localization of melanoblasts can be determined on whole-mount embryos or on transversal sections. This allows (i) to evaluate the number of melanoblasts in the truncal and vagal regions, (ii) to assess the number of dermal and epidermal melanoblasts and their doubling time and (iii) to determine the cellular mechanisms altered in melanoblasts during development. Using an integrative approach, the main molecular mechanisms affected by the deletion/activation of specific genes in melanoblasts can be determined. Once embryos have been produced, these are genotyped (Data S11), treated immediately to visualize melanoblasts or conserved in paraffin or Tissue-Tek after paraformaldehyde fixation for further experiments (Data S12 and S13). The number and localization of all dermal and epidermal melanoblasts can be determined on whole-mount Dct::LacZ (or any reporter) embryos and on transversal sections during development (Data S14 and S15) (Luciani et al., 2011; Yajima et al., 2006). The growth rates of all dermal and epidermal melanoblasts can be determined (Delmas et al., 2007; Luciani et al., 2011). The proliferative capacity and rate of the melanoblasts can be evaluated using Ki-67 labelling and BrdU incorporation (Data S4 and S16). The rate of apoptosis can be evaluated using caspase-3 antibody. (Data S4). Study of the melanocyte lineage allows a better understanding of the potential loss of differentiation, transdifferentiation or dedifferentiation. The general analyses in melanogenesis described previously help to decipher the main processes involved and provide insights into the affected molecular pathways. We thank all colleagues of the field, specifically Heinz Arnheiter for his help in the establishment of these protocols. Work in the laboratories is supported by grants from the CNRS, the INSERM, the Association pour la Recherche contre le Cancer, the Ligue Nationale contre le Cancer and INCA grant. LL is ‘équipe labellisée’ of the Ligue Nationale contre le Cancer. Data S1. Hair preparation and photography. Data S2. Determination of the amount of eumelanin mouse hairs. Data S3. TTA (Tyramide-based tyrosinase assay) on frozen mouse skin sections. Data S4. Immunostaining on cryosections. Data S5. Isolation of melanoblasts from embryonic skin by FACS and RNA extraction. Data S6. Primary culture of murine skin melanocytes. Data S7. Level of melanin in cells in culture. Data S8. Tyrosinase assay. Data S9. RT qPCR from cultured melanocytes. Data S10. Single cell migration assay using time-lapse video-microscopy. Data S11. Mouse genotyping – Mouse tail quick DNA extraction followed directly by PCR. Data S12. Embryos paraffin embedding. Data S13. Embedding tissue or embryos in Tissue-Tek. Data S14. Eosin/Haematoxylin coloration on skin or embryo paraffin sections. Data S15. X-gal staining of embryos. Data S16. BrdU incorporation followed by tissue-tek embedding and immunostaining. 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.