In this Issue

Abstract

Sentinel lymph node biopsy has been recently advocated as a way to determine whether lymph node dissection would be necessary as part of the surgical treatment of melanoma. This concept is based on the notion that the “sentinel lymph node”, the first lymph node of the draining site of the primary tumor, is never bypassed in malignant melanoma. Thus, lack of metastases in the sentinel lymph node has been considered a good prognostic finding. In this issue, Lukowsky et al. (p. 554) have evaluated this concept by examining the regional lymph nodes, sentinel and nonsentinel, in 24 patients with malignant melanoma by histopathology and immunohistochemistry, and compared these results for the presence of tyrosinase mRNA as determined by RT-PCR. Because the tyrosinase gene is expressed exclusively in melanocytic cells, detection of the corresponding transcripts would signify the presence of metastases in the tissue. As expected, 10 of 20 patients with histologic evidence of metastasis in the sentinel lymph node were also positive for tyrosinase mRNA, as were three additional patients without morphologic evidence, attesting to the sensitivity of the RT-PCR approach. Surprisingly, however, in seven patients RT-PCR was clearly positive for tyrosinase mRNA in nonsentinel lymph nodes, whereas the sentinel lymph node was negative. Assuming the specificity of the PCR for tyrosinase, it appears that the sentinel lymph node biopsy may miss micrometastases that have spread beyond the primary tumor and somehow bypassed the sentinel lymph node. These results clearly warrant further studies to validate the concept of sentinel lymph node biopsy. They also suggest that incorporation of RT-PCR into the armamentoire to examine regional lymph nodes in patients with malignant melanoma increases the sensitivity for detection of metastases Collagens comprise a superfamily of proteins, and as many as 20 different mammalian collagens have been studied thus far to the extent that they have been designated by a Roman numeral. Among these 20 collagens, at least a dozen have been demonstrated in the skin. In this issue, Peltonen et al. (p. 635) have examined the localization of another collagen, type XIII, which has recently been shown to be a transmembrane protein. Using a variety of immunologic and molecular biologic techniques, the authors demonstrate that type XIII collagen is expressed in the epidermis in various contact sites. Specifically, colocalization studies suggested that type XIII collagen is not a member of the desmosome, but is likely to be associated with adherens type junctions. Previously, another collagen, type XVII collagen, has been shown to be a keratinocyte gene product, in type 2 transmembrane orientation and a component of hemidesmosomes. Type XVII collagen clearly confers stability to the dermal–epidermal basement membrane zone, as mutations in the corresponding gene (COL17A1) underlie a non-Herlitz type of junctional epidermolysis bullosa, known as generalized atrophic benign EB (GABEB). The functional role of type XIII collagen within the epidermis is currently unknown, but the authors speculate that, in analogy with type XVII collagen, aberrations in this collagen may be associated with heritable and/or acquired blistering skin diseases. The spectacular progress recently made in understanding the molecular basis of various heritable skin diseases is exemplified by dystrophic epidermolysis bullosa (DEB). At the beginning of this decade nothing was known of the specific mutations resulting in various forms of DEB, and the prevailing notion suggested that somehow the expression of proteolytic enzymes, primarily collagenase, was altered, resulting in dissolution of anchoring fibrils and other collagenous structures. Following the cloning of type VII collagen as the candidate gene for DEB, a number of reports have detailed specific mutations in COL7A1 in different variants of DEB. In this issue, Whittock et al. (p. 673) compare the efficiency of three different mutation detection strategies, i.e., protein truncation test, fluorescence chemical cleavage of mismatch, and conformation sensitive gel electrophoresis, for their sensitivity to detect mutations in COL7A1 in patients with DEB. The results indicate that none of the these methods alone is able to detect all putative mutations, but combined use of these techniques yields an overall sensitivity of 87%. In fact, the authors were able to detect 147 of 169 mutations in a total of 93 patients with DEB, 57 of the mutations being previously unreported. These novel mutations, together with previously published ones, brings the total number of COL7A1 mutations in DEB to over 200. Examination of the mutation database reveals that the majority of the genetic lesions are family specific with a few recurrent mutations being detected. There is thus far no evidence for locus heterogeneity, and the gene for collagenase has been excluded from DEB locus by genetic linkage analyses. These observations have diagnostic importance with prognostic implications, and they form the basis for DNA-based prenatal testing and preimplantation genetic diagnosis, as well as for gene therapy in the future.