Treatment of mice with anti-CD3 mAb induces endothelial vascular cell adhesion molecule-1 expression.

In vivo treatment of mice with anti-CD3 mAb causes polyclonal T cell activation and cytokine release. Since several cytokines are known to alter expression of MHC molecules and adhesion molecules on endothelia, we hypothesized that anti-CD3 mAb treatment should result in activation of vascular endothelia. In previous studies, we established that vascular endothelial cells in murine heterotopic cardiac grafts can develop at least 2 stable inflammatory phenotypes: cardiac allograft endothelia characteristically develop reactivity with MECA-32 mAb (undefined endothelial epitope) and M/K-2 mAb (murine vascular cell adhesion molecule-1 [VCAM-1]), whereas cardiac isografts develop reactivity with MECA-32, but not M/K-2 mAb. We now report that a single treatment of cardiac isograft recipients with the anti-CD3 mAb 145-2C11 caused expression of VCAM-1 on all cardiac isograft endothelia, including the microvascular endothelia. In contrast, expression of endothelial VCAM-1 in the native heart of the isograft recipient was limited to patchy areas of larger arteries. This patchy, arterial pattern of VCAM-1 expression was observed in lung, liver, kidney, and thymus of all mice treated with 145-2C11, whether or not they were implanted with a cardiac isograft, and was dissociated from expression of MECA-32 mAb reactivity. Hence, treatment of mice with anti-CD3 mAb causes systemic endothelial activation (VCAM-1 expression), and endothelial cells of recently implanted cardiac isografts appear to be hypersensitive to induction of VCAM-1 by anti-CD3 mAb treatment. Further studies showed that (1) treatment with 145-2C11 F(ab)'2 fragments did not induce endothelial activation, (2) intravenous pretreatment with pentoxifylline eliminated all endothelial effects of 145-2C11 treatment, (3) induction of endothelial activation by 145-2C11 mAb always paralleled the expression of adverse physiologic symptoms, and (4) mice exhibit strain-specific differences in endothelial responses to 145-2C11 treatment. We propose that anti-CD3 mAb treatment causes simultaneous activation of circulating T cells and systemic vascular endothelial cells which may facilitate systemic lymphocyte-endothelial interactions, and may explain the rapid disappearance of T cells from the circulation that is associated with anti-CD3 treatment in mouse and man.