Faculty Bibliography

Although rare, melanomas do occur in children and adolescents. Pediatricians should be aware of the clinical features of melanoma and the risk factors for developing this malignancy. Children at high risk for melanoma should have at least annual cutaneous examinations in search of suspicious lesions. If a lesion is suspected of being a melanoma, it should be removed surgically and submitted for pathologic examination. Education of parents and children about the deleterious effects of ultraviolet light affords a means of counteracting the increasing incidence of melanoma at the grassroots level. The use of sunscreens, hats, and other protective clothing and the judicious timing of daily solar exposure should serve to prevent sunburning, limit tanning, and decrease the incidence of melanoma and other more common cutaneous malignancies, such as basal cell carcinoma and squamous cell carcinoma

Mice homozygous for the platinum (cp) allele at the albino locus manifest severe oculocutaneous albinism despite the presence in vitro of tyrosinase activity at 25% wild-type levels. We demonstrate that the cp allele results from an A-->T substitution, changing a lysine residue at position 489 to a termination codon, with truncation of tyrosinase's cytoplasmic tail. In choroidal melanocytes of neonatal mutant mice, tyrosinase activity could be detected in the trans Golgi network, but was absent from melanosomes. Instead, it was detected in vesicles in the cell periphery and dendrites, and on the extracellular surface. In the retinal pigment epithelium, activity was present on the extracellular apical and basolateral surfaces. Our results demonstrate misrouting of a mutant tyrosinase lacking its cytoplasmic tail, providing an explanation for the severe effect of this mutation on ocular and cutaneous pigmentation

Cutaneous melanomas of Tyr-SV40E transgenic mice (mice whose transgene consists of the tyrosinase promoter fused to the coding regions of simian virus 40 early genes) strikingly resemble human melanomas in their development and progression. Unlike human melanomas, the mouse tumors all arise in genetically identical individuals, thereby better enabling expression of specific genes to be characterized in relation to advancing malignancy. The products of pigment genes are of particular interest because peptides derived from these proteins have been reported to function as autoantigens with immunotherapeutic potential in some melanoma patients. However, the diminished pigmentation characteristic of many advanced melanomas raises the possibility that some of the relevant products may no longer be expressed in the most malignant cells. We have therefore investigated the contributions of several pigment genes in melanotic vs. relatively amelanotic components of primary and metastatic mouse melanomas. The analyses reveal marked differences within and among tumors in levels of mRNAs and proteins encoded by the wild-type alleles at the albino, brown, slaty, and silver loci. Tyrosinase (the protein encoded by the albino locus) was most often either absent or undetectable as melanization declined. The protein encoded by the slaty locus (tyrosinase-related protein 2) was the only one of those tested that was clearly present in all the tumor samples. These results suggest that sole reliance on targeting tyrosinase-based antigens might selectively favor survival of more malignant cells, whereas targeting the ensemble of the antigens tested might contribute toward a more inclusive and effective antimelanoma strategy

Crouzon syndrome, an autosomal dominant condition characterized by craniosynostosis, ocular proptosis and midface hypoplasia, is associated with mutations in fibroblast growth factor receptor 2 (FGFR2) (refs 1-3). For example, we have identified 10 different mutations in the FGFR2 extracellular immunoglobulin III (IgIII) domain in 50% (16/32) of our Crouzon syndrome patients. All mutations described so far for other craniosynostotic syndromes with associated limb anomalies--Jackson-Weiss, Pfeiffer, and Apert--also occur in the extracellular domain of FGFR2, as well as FGFR1 for Pfeiffer syndrome. In contrast, only FGFR3 mutations have been reported in dwarfing conditions--achondroplasia, thanatophoric dysplasia, and hypochondroplasia. For achondroplasia, greater than 99% of mutations occur in the FGFR3 transmembrane domain. We now report the unexpected observation of a FGFR3 transmembrane domain mutation, Ala391Glu, in three unrelated families with Crouzon syndrome and acanthosis nigricans, a specific skin disorder of hyperkeratosis and hyperpigmentation. The association of non-dwarfing and even non-skeletal conditions with FGFR3 mutations reveals the potential for a wide range of FGFR pleiotropic effects as well as locus heterogeneity in Crouzon syndrome. Our study underscores the biologic complexity of the FGFR gene family

The two major types of mammalian melanin are pheomelanin (yellow or red pigment) and eumelanin (black or brown). The agouti (A) and extension (E) loci determine whether follicular melanocytes will deposit pheomelanin or eumelanin within their melanosomes. Mutations at the murine pinkeyed-dilution (P) locus cause a striking reduction in deposition of eumelanic, but not pheomelanic, pigment. The mRNA encoded at the P locus is not expressed in skin that exclusively produces pheomelanic pigment as a result of mutation at the agouti locus. We have suggested, based upon both genetic and biochemical evidence, that three key melanogenic proteins--tyrosinase, tyrosinase-related-protein-1 (TRP-1), and TRP-2, encoded at the albino (C), brown (B), and slaty (Slt) loci, respectively--form a high-molecular-weight 'melanogenic complex' within the melanosome. High-molecular-weight forms of tyrosinase, TRP-1 and TRP-2, are absent from eumelanic ocular tissues of p(un)/p(un) mice that fail to produce normal P-locus transcript, even though these mice are genetically normal at the loci that regulate production of the three melanogenic proteins. We have hypothesized that the presence of the p-locus protein is important for the integrity of the melanogenic complex and for the levels of members of the TRP family. We show here that the yellow skins of mice mutant at the agouti or extension loci, as well as the nonyellow skins of pinkeyed-unstable (p(un)/p(un)) mice, demonstrate greatly diminished levels of tyrosinase, TRP-1 and TRP-2, and an absence or markedly decreased proportion of high-molecular-weight forms of melanogenic proteins. We conclude that normal levels of wild-type P-locus protein are necessary for eumelanogenesis and that the absence of this protein may be necessary, but is not sufficient to cause the melanosome to switch to the production of pheomelanin. We discuss the implications of our results in relation to the interacting genetic controls regulating melanogenesis

Mutations have been reported for several craniosynostotic disorders in exon IIIa (exon U or 7) or IIIc (exon B or 9) of the fibroblast growth factor receptor 2 gene (FGFR2). Among the conditions with FGFR2 mutations are two autosomal dominant syndromes, Crouzon and Jackson-Weiss. In this study, 24 Crouzon and one Jackson-Weiss syndrome patients were screened for mutations in the two exons by direct sequencing, and mutations were detected in 28% (7/25) of all cases. Five different mutations were found including two novel (W290G, C342W) and two previously reported, recurrent mutations for Crouzon syndrome (A344A, S354C), and one new mutation for Jackson-Weiss syndrome (C342R). The W290G mutation was found in exon IIIa which is common to both alternatively spliced forms of FGFR2, BEK (expressed predominantly in primordial bones) and KGFR (expressed preferentially in epithelia). Atypical Crouzon syndrome features of epithelial-derived anal and/or external ear anomalies were present in the two affected family members with the mutation. This phenotype possibly reflects the expression of both mutant BEK and KGFR. In addition, the Jackson-Weiss syndrome mutation, C342R, in exon IIIc was observed previously in other craniosynostotic syndromes, Crouzon and Pfeiffer. These results underscore the allelic heterogeneity of these conditions and the complexity of the phenotypic consequences of FGFR2 mutations

PURPOSE. To study melanogenesis by cultured human uveal melanocytes, and the relationship between melanin production by uveal melanocytes in vitro with the degree of iris pigmentation in vivo. METHODS. Melanin content, melanin production, and tyrosinase activity of cultured uveal melanocytes derived from eyes of various iris color were measured at different stages of cultivation. RESULTS. Cultured uveal melanocytes maintained a constant level of melanin content, expressed tyrosinase activity, and produced measurable amounts of melanin in vitro. Melanosomes in different stages were seen ultrastructurally. Melanin production correlated directly with the degree of iris pigmentation of the eyes from which the uveal melanocytes were isolated. Tyrosinase activity of cultured uveal melanocytes from black versus white donors was significantly different, but, among white donors, there was no correlation with iris pigmentation or with melanin production in vitro. CONCLUSION. Cultured uveal melanocytes can produce melanin in vitro. Cultured uveal melanocytes isolated from eyes of different iris color maintained their inherent capacity for melanogenesis. Therefore, cultured uveal melanocytes are an excellent model system for studying melanogenesis in uveal melanocytes in vitro

The pink-eyed dilution (p) locus in the mouse is critical to melanogenesis; mutations in the homologous locus in humans, P, are a cause of type II oculocutaneous albinism. Although a cDNA encoded by the p gene has recently been identified, nothing is known about the protein product of this gene. To characterize the protein encoded by the p gene, we performed immunoblot analysis of extracts of melanocytes cultured from wild-type mice with an antiserum from rabbits immunized with a peptide corresponding to amino acids 285-298 of the predicted protein product of the murine p gene. This antiserum recognized a 110-kDa protein. The protein was absent from extracts of melanocytes cultured from mice with two mutations (pcp and p) in which transcripts of the p gene are absent or greatly reduced. Introduction of the cDNA for the p gene into pcp melanocytes by electroporation resulted in expression of the 3.3-kb mRNA and the 110-kDa protein. Upon subcellular fractionation of cultured melanocytes, the 110-kDa protein was found to be present in melanosomes but absent from the vesicular fraction; phase separation performed with the nonionic detergent Triton X-114 confirmed the predicted hydrophobic nature of the protein. These results demonstrate that the p gene encodes a 110-kDa integral melanosomal membrane protein and establish a framework by which mutations at this locus, which diminish pigmentation, can be analyzed at the cellular and biochemical levels