Does p53 regulate skin pigmentation by controlling proopiomelanocortin gene transcription?

Abstract

A recent issue of Cell highlights an intriguing, unexpected role for p53 as the central regulator of pigmentary responses in tanning and pathologic hyperpigmentation (Cui et al., 2007). Furthermore, in claiming to have provided biochemical and genetic evidence that the UV induction of intracutaneous proopiomelanocortin (POMC) (and its enzymatic cleavage products, a-MSH (melanocyte stimulating hormone) and b-endorphin) is directly controlled by p53, the authors propose that UV-induced tanning mainly follows a linear sequence of events that starts with a p53-mediated activation of the POMC promoter (Cui et al., 2007). The potential clinical significance and the biological implications of the reported findings and concepts necessitate further critical discussion of the experimental evidence, and the conclusions drawn from it, in the context of the available information on the subject. First, a significant part of the authors’ conclusion is based on animal experiments performed with C57BL ⁄6 mice (wild-type and p53 null) mice. However, POMCknockout mice on the same C57BL6 background are well known to show normal melanin pigment production (Slominski et al., 2005a). This observation may weaken the authors’ basic concept and also questions the validity of the proposed main regulatory axis for the suntan response and pathological hyperpigmentation (i.e. UV fi p53 fi POMC fi melanin pigmentation). This central issue requires serious explanation, and should have been carefully discussed by the authors. In our view, published evidence supports the hypothesis that it may not be POMC and its products but rather the melanocortin-1 receptor (MC1-R) that is the key regulator of pigmentation (at least in mice) (Slominski et al., 2004, 2005b). Moreover, in the absence of POMC-derived melanocortins, the MC1-R can regulate melanogenesis through an intrinsic, ligand-independent activity, or can be activated by ligands other than melanocortins (Sanchez-Mas et al., 2004; Slominski et al., 2005a,b). Furthermore, the authors do not appear to acknowledge that the regulation of melanocyte behavior, including melanogenesis control, is very heterogeneous and multi-directional; this further underscores the complexity of microenvironment – specific controls in pigmentation, which cannot be satisfactorily explained by an organization consisting of simple linear sequences (Nordlund et al., 2006; Slominski et al., 2004). On this background, we consider it more likely that p53 acts as one important coordinator (among others), but not as the main or sole regulator of pigmentation in the suntan response and pathological hyperpigmentation. Second, the authors provide evidence for direct induction of POMC transcription by p53, yet show only for a moderate increase of POMC precursor immunoreactivity (no evidence is presented that the selected immunoreactive band in their Figures 1 and 2 actually represents POMC). In this context, the surprising, highly unconventional sequence of reported events (e.g. p53 induction by UV is maximal by 3 h, POMC immunoreactivity reportedly peaks at 6 h, but its mRNA transcription at 24 h) requires explanation. Moreover, the authors fail to show conclusive evidence that the detected immunoreactivities (which, again, are rather unconventional: e.g. nuclear location of a-MSH in keratinocytes) indeed represent a-MSH or b-endorphin peptides, as claimed (confirmatory HPLC (high performance liquid chromatography) or mass spectrometry evaluations of representative extracts were not performed). Also, after transcription POMC mRNA is generated by splicing, which then is translated into a protein precursor of approximately 30 kDa. This precursor is further processed in a cell type-specific and tissue microenvironment-specific manner through complex, sequential post-translational modifications (including endoand exopeptidase cleavage, amidation and acetylation), which end with the regulated secretion of the final product (Slominski et al., 2000; Smith and Funder, 1988). The complexity of these events, in fact, renders it rather unlikely that p53 would directly regulate production and release of a-MSH or b-endorphin (however, indirect effects are quite possible).