Adhesion molecules in lymphoma metastasis

Abstract

Recently, many surface proteins of lymphoid cells that mediate adhesion to other cells and extracellular matrix have been identified. Several of these cellular adhesion molecules (CAM) are also expressed by metastatic lymphoma cells and may mediate adhesion to tissue components during the metastatic process. Correlations observed between expression of certain CAM, like MEL-14 and CD44, and particular patterns of spread, support this notion, but conclusive evidence is scarce. We have used T-cell hybridomas to study the mechanisms of wide-spread lymphoid metastasis. The results obtained with this model are reviewed here. The advantages are that a large number of genetically similar cell lines can be generated, which can be grouped in large panels of highly invasive and non-invasive cells. Invasiveness of these cells in hepatocyte and fibroblast monolayers correlates with experimental metastasis. Lymphoid CAM that are potentially involved in metastasis are reviewed. Several of these CAM are not, or not consistently, expressed by the invasive T-cell hybridomas, indicating that they are not indispensable. Notably, some of the CAM involved in the onset of an immune response or in migration into inflamed tissues, like ICAM-1 and VLA-4, and the 'homing receptors' MEL-14 and LPAM-1 do not seem to be involved. CAM that are consistently expressed by the T-cell hybrids include LFA-1, the beta-1 integrin subunit CD29, CD31 (PECAM-1) and CD44 ('Hermes homing receptor'). We have generated considerable evidence that LFA-1 is required for efficient metastasis of T-cell hybrids, based on the behavior of LFA-1-deficient mutants and revertants. High levels of LFA-1 are required. The relevant counterstructure is probably ICAM-2 rather than ICAM-1. Preliminary results suggest that also a beta-1 integrin, possibly VLA-5, plays a role. Finally, we summarize evidence indicating that CD31 and CD44 are primary candidates for involvement in metastatic spread of T-cell hybridomas.