Expression of CD44 and the pattern of CD44 alternative splicing in uveal melanoma.

In cutaneous melanoma, the standard CD44 molecule is abundantly expressed, whereas the expression of certain splice variants is related to tumour progression and to the metastatic potential of the cell line. In the present study we have investigated the expression of CD44 and the pattern of CD44 alternative splicing in uveal melanoma in relation to the cell type, diameter and invasiveness of the tumour. All uveal melanomas strongly stained with antibodies to the standard portion of CD44. No expression of the CD44 variant (v) exon CD44v7 was found, whereas v5, v6 and v10 were expressed (in 2/12, 5/12 and 8/12 cases, respectively). No correlation was observed between expression of particular splice variants and cell type, tumour diameter or invasion of the sclera or Bruch's membrane. All three uveal melanoma cell lines tested were strongly CD44 positive and expressed low levels of v6-containing isoforms at the cell surface, but v5, v7 and v10 were absent. Our results show that CD44 is strongly expressed in uveal melanoma and that the pattern of CD44 alternative splicing is similar to that observed in cutaneous melanoma. However, in uveal melanoma this alternative splicing does not appear to be related to prognostic parameters.

Emergence of alpha 5 beta 1 fibronectin- and alpha v beta 3 vitronectin-receptor expression in melanocytic tumour progression.

Cell adhesion is crucial in the process of tumour progression. As integrins are important receptor molecules involved in cell adhesion, we studied the distribution of the alpha 1-6, alpha v, alpha IIb, beta 1, beta 3, and beta 4 integrin subunits in tissue sections of common naevocellular naevi (n = 22), dysplastic naevi (16), thin (24) and thick primary cutaneous melanomas (28), and melanoma metastases (25). We found correlated expression of alpha 1/alpha 2, of alpha 4/alpha 5/beta 3, and of alpha 6/beta 4. Decrease of alpha 6 and beta 4, and increase of alpha 4 and alpha v were found to be correlated with melanoma progression. Furthermore, expression of alpha 5 and beta 3 was detected only in primary melanoma and melanoma metastasis. Our findings indicate that during melanoma progression alterations in integrin expression occur, the most striking being emergence of alpha 5 beta 1 fibronectin and alpha v beta 3 vitonectin receptor.

Integrin expression in uveal melanoma differs from cutaneous melanoma.

PURPOSE: During the process of metastasis, changes in cell-cell and cell-matrix contacts occur; therefore, expression of integrins, a superfamily of adhesion molecules, may be important. Expression of integrins has been studied extensively in cutaneous melanoma. Because it is known that uveal melanoma has a metastatic behavior different from cutaneous melanoma, the authors investigated integrin expression in uveal melanoma. METHODS: The authors used monoclonal antibodies recognizing integrin subunits alpha 1-6, alpha v, beta 1, and beta 4 and integrins alpha v beta 3 and alpha v beta 5 on frozen sections of 32 human primary uveal melanomas and 4 metastases, followed by an avidin-biotin-peroxidase complex-immunoperoxidase technique. RESULTS: As in cutaneous melanoma, alpha 4 expression was rare, but most lesions expressed alpha 3 and alpha 6. In contrast to cutaneous melanoma, in which alpha 2 is well expressed in most lesions and alpha 5 is expressed only in a small percentage of lesions, alpha 2 expression was rare in uveal melanoma and alpha 5 expression was found in all lesions. A major difference was observed with regard to the alpha v beta 3 vitronectin receptor. In contrast to cutaneous melanoma, in which alpha v beta 3 is expressed in advanced primary melanomas and metastases, alpha v beta 3 was not detected in any of the primary uveal melanomas, but all lesions strongly expressed alpha v beta 5. CONCLUSIONS: Integrin expression in uveal melanoma cannot be correlated with cell type or invasiveness. In contrast to cutaneous melanoma, it seems that determination of the integrin expression profile is not suitable for categorizing uveal melanomas as less malignant and highly malignant lesions.