Transcriptional inactivation of the single X chromosome occurs in spermatogenic cells during male meiosis in mammals and has been shown to be coincident with expression of the Xist gene in spermatogonia and spermatocytes in mice. However, male mice carrying an ablated Xist gene show normal fertility. Here we examined expression from the Xist locus during spermatogenesis in wild-type mice and detected sense (Xist), but not antisense (Tsix) transcripts. In addition, we examined expression and chromatin conformation of X-linked structural genes in meiotic and postmeiotic spermatogenic cells from wild-type and Xist(-) mice and found no differences associated with the absence of a functional Xist gene. These results, along with the formation of a morphologically normal XY body in primary spermatocytes in Xist(-) mice, indicate that a functional Xist gene is not required for X-chromosome inactivation during spermatogenesis and that this process is therefore regulated by a different mechanism than that which regulates X-chromosome inactivation in female embryonic cells.

We examined the involvement of the cyclin-dependent kinase inhibitor 2A (CDKN2A) locus in the pathogenesis of ultraviolet (UV) radiation-induced melanomas in an opossum (Monodelphis domestica) melanoma model in which suckling young were exposed to UVB to produce melanocytic lesions. Monodelphis CDKN2A and alternated reading frame (ARF) cDNAs were cloned and sequenced, and the expression patterns of these genes were determined by reverse transcription-polymerase chain reaction in normal tissues, 39 primary melanocytic skin lesions, and two tumor-derived cell lines, one nonmetastatic and one metastatic. Primary melanocytic lesions, including hyperplasias, benign melanomas, melanomas metastatic to lymph nodes, and melanomas metastatic to nodes and additional visceral organs, were categorized accordingly as types I-IV. Levels of CDKN2A transcripts were most abundant in type III tumor samples and the metastatic cell line but absent in the nonmetastatic cell line. ARF transcripts were expressed in all tumors and cell lines. A UV-signature mutation was detected with the wild-type allele at the CDKN2A locus in type II and III primary tumor samples and in the nonmetastatic cell line. Interestingly, in the metastatic cell line, only the mutant allele was present and expressed. These data suggest dynamic changes in the expression and/or structure of the CDKN2A and ARF genes represent one molecular defect associated with the etiology of melanoma formation and progression in the Monodelphis model system.

Monodelphis domestica, a South American opossum, has been established as a mammalian model for sporadic ultraviolet radiation (UVR)-induced melanoma. Using this model system, we investigated the role of changes in the p53 gene in the development of cutaneous melanocyte-derived lesions. Cutaneous melanocytic hyperplasias, benign melanomas and metastatic primary melanomas, plus affected lymph nodes and visceral organs, were screened for mutations in the Monodelphis p53 gene by single-strand conformation polymorphism analysis and direct sequencing. With the exception of a silent point mutation found in a single benign melanocytic hyperplasia sample, no p53 mutations were detected. Furthermore, a relative quantitative reverse transcriptase-polymerase chain reaction approach was used to analyse p53 gene expression at different stages of primary melanoma progression and revealed no substantial changes in p53 mRNA levels. These results suggest that, as in humans, UVR-induced melanoma in the Monodelphis model is initiated and progresses on the basis of predominantly p53-independent molecular pathways.