Cloning and characterization of the expression pattern of a novel splice product MIA (splice) of malignant melanoma-derived growth-inhibiting activity (MIA/CD-RAP) [corrected].

Melanoma-inhibiting activity/cartilage-derived retinoic acid-sensitive protein, a 11 kDa protein, is mainly expressed in cartilage during embryogenesis, and is related to invasion, metastasis, and immunomodulation of melanoma and glioma cells in vivo and in vitro. Here, we describe an alternative splice product of this gene termed melanoma-inhibiting activity (splice), lacking exon 2 of the original protein. A predicted frameshift by alternate splicing results in a unique C-terminal portion of the protein. Consistent with this, a protein migrating at the predicted molecular weight of the splice form (3.5 kDa) was detected using an N-terminal specific antibody. This band was undetectable when using a C-terminal specific antibody. In addition, we describe the expression pattern of melanoma-inhibiting activity (splice) in different human tumors. Expression was shown in tissue samples of five of six primary melanomas, 11 of 12 primary sites of metastatic melanomas, 10 of 10 systemic metastases of melanomas, four of four central nervous system metastases of melanomas, six of eight primary melanoma cultures, and five of five melanoma cell lines. Only a faint signal was obtained in tissue samples of five of six naevi. Interestingly, seven of eight nonmelanocytic tissue samples and five of seven glioma cell lines showed weak expression of melanoma-inhibiting activity (splice). Approaching first functional aspects, reverse transcriptase-polymerase chain reaction showed weak expression of melanoma-inhibiting activity (splice) in relation to melanoma-inhibiting activity in nonmelanocytic and strong expression in melanocytic cells. Staining with a specific anti-serum raised against a synthetic peptide resembling the amino acid sequence of melanoma-inhibiting activity (splice) showed a more nuclear staining pattern in comparison with melanoma-inhibiting activity. Furthermore, incubation of melanoma and glioma cell cultures with transforming growth factor-beta2 showed inverse regulation of the mRNA of melanoma-inhibiting activity and melanoma-inhibiting activity (splice), both suggesting also a different function within the physiologic role of this unique family of proteins. Melanoma-inhibiting activity (splice) has no homology to any other known protein so far. Whereas the biologic function of melanoma-inhibiting activity (splice) is not clear yet, it might provide a relevant diagnostic and therapeutic tool for malignant melanomas.