Evidence that lymphocyte traffic into rejecting cardiac allografts is CD11a- and CD49d-dependent.

Acute cardiac allograft rejection is characterized by infiltration of leukocytes into tissue parenchyma, but the site of entry and endothelial adhesion molecules involved are not yet defined. Lymphocyte binding to frozen sections prepared from day-3 rejecting cardiac allografts was significantly increased compared with sections made from normal hearts (number of bound lymphocytes, 983 +/- 216 per mm2 vs. 309 +/- 121, respectively, P < 0.001) or syngeneic grafts. The bound lymphocytes were located exclusively only on the top of the capillary structures and not on any other sites on the heart vasculature. We further wanted to analyze which of the cloned endothelial adhesion molecules and their counterreceptors would be involved in the increased lymphocyte binding. Lymphocyte pretreatment with mAb anti-CD11a or anti-CD49d inhibited this binding more than 50%. This inhibition on lymphocyte binding could not be increased by combining these two antibodies. Lymphocyte binding to endothelium has been shown to be at least partly organ specific; therefore, we asked whether increased lymphocytes adhere to cardiac allografts could be organ specific. Lymphocyte binding to lymph node high endothelial venules (HEV) has been shown to be inhibited by mannose-6-phosphate (M6P) and to kidney peritubular capillaries by mannose-1-phosphate (M1P). In the present study neither of these carbohydrates had any effect on lymphocyte binding to cardiac allograft endothelium. Monosaccharide inhibition studies demonstrate that the mechanism of lymphocyte adhesion to cardiac capillary endothelium differs from adhesion to kidney allografts or peripheral lymph node high endothelium.