Expression of the transferrin receptor on target cell lines has recently been implicated as a determinant of susceptibility to cytolysis by natural killer (NK) lymphocytes. We have examined this proposed relationship in several ways. First, K562 (a cell line highly vulnerable to NK lysis) cells were grown for 24 h in the iron chelator desferrioxamine. Under these conditions, the cells doubled their surface transferrin receptor expression as determined both by radioligand binding and surface binding of the OK-T9 monoclonal anti-transferrin receptor antibody. In contrast, cells grown for the same period of time in hemin halved their receptor expression. This fourfold change in transferrin receptor expression between the desferrioxamine-treated and hemin-treated cells produced no change in susceptibility to NK cytolysis. Second, HeLa (a cell line which in its native state is very resistant to NK cytolysis) cells were compared with K562 cells with respect to surface transferrin receptor expression. The difference in NK susceptibility of the two cell lines was not reflected in differences in transferrin receptor expression: the K562 cells expressed approximately 1.5 X 10(5) receptors per cell while HeLa cells expressed 2.0 X 10(5) receptors/cell. Third, infection of HeLa cells by measles virus greatly increased their susceptibility to NK lysis but produced no change in surface transferrin receptor expression. Furthermore, when measles-infected HeLa cells were grown for 6 d in medium supplemented with iron-saturated human transferrin they underwent a 50% reduction in receptor expression but no change in NK susceptibility. Finally, possible alterations in the surface expression of NK target antigens on modified cells were further assayed by their ability to serve as cold-target inhibitors of cytolysis of NK-sensitive target cells. We examined two groups of cells in which transferrin receptor expression was reduced. These were the transferrin-treated, measles-infected HeLa cells with the 50% receptor reduction, and K562 cells grown in medium containing hemin and iron salts where the reduction was five- to sixfold relative to control. In neither case was there a change in the apparent expression of NK target antigen(s). We conclude that there is a discordance between transferrin receptor expression and susceptibility to NK cytolysis in the model systems examined. Therefore, it is unlikely that the transferrin receptor per se is the target recognition structure for human NK cells, although a role in concert with other, as yet undefined molecules, cannot be excluded.
K R Bridges, B R Smith