The properties of endothelial cells as antigen-presenting cells are likely to be important during local inflammatory responses and following transplantation of vascularized tissues, where EC are the first alloantigen-bearing target encountered by transiting host T cells. Endothelial cells constitutively express major histocompatibility complex (MHC) class I molecules, which enables them to present antigen to CD8+ (cytotoxic) T lymphocytes. In addition, vascular EC often acquire the potential to present antigen to CD4+ (helper) T cells, since MHC class II molecule expression on small vessels is a common finding in a variety of inflammatory conditions and human capillary ECs have been shown to express MHC class II molecules in heart, kidney, and liver. Triggering of the T-cell antigen receptor (TCR) by the endothelium is likely to occur during the sustained adhesive interactions leading to T-cell extravasation. The functional consequences of antigen presentation by the endothelium, particularly its ability to initiate T-cell responses, are still controversial. A key point in these issues is that ECs (at least in the human system) do not express the B7 family of costimulatory molecules that, when triggered in conjunction with the antigen receptor, allow autocrine production of the T cell growth factor inteleukin-2 and cell division. Other costimulatory molecules that have been found to be expressed by endothelial cells, such as inducible T-cell costimulator-ligand (ICOS-L), are unable to induce IL-2 gene transcription, although they can support the production of other cytokines, such as IL-4. Finally, negative regulators of T-cell activation, such as programmed death-ligand 1 (PD-L1) and PD-L2, have been identified on ECs.
CD4+ (Helper) T Cells
The conclusions from some studies on the functional effect of cognate recognition of the endothelium by CD4+ T cells are to date controversial. Some groups have reported that EC can support mitogen-induced T-cell proliferation, antigen-specific T-cell proliferation, and primary allore-sponses. These authors hypothesized that during antigen presentation by ECs, costimulation can be provided by surface molecules other than B7, such as LFA-3. Other groups, including ourselves, have failed to detect alloproliferation of human and mouse CD4+ T lymphocytes induced by either resting or cytokine-treated endothelium. Possible explanations for these discrepancies include a contamination with bone-marrow-derived antigen presenting cells (such as dendritic cells) of either the EC or the T-cell preparations.
An interesting observation that arose from these studies is that, unlike what has been observed with other costimulation-deficient parenchymal cells, memory T cells were not rendered hyporesponsive by cognate recognition of B7-negative endothelium. It is possible that the availability of accessory molecules displayed on the EC surface and coengaged during cognate recognition might modulate the signal delivered through TCR engagement. Thus antigen presentation in the absence of costimulation, but in the presence of strong cell-cell adhesion, mediated by the large array of accessory molecules displayed on the EC surface, might fall below the threshold required for T-cell activation but above that leading to T-cell silencing.
CD8+ (Effector) T Cells
The few studies that have investigated the effects of cognate recognition of EC by CD8+ T cells have focused on the development of proliferative responses, and, as for CD4+ T cells, contradictory results have been reported. Some studies have concluded that EC can induce human and murine CD8+ T lymphocyte differentiation, although with a low efficiency compared with conventional antigen-presenting cells (APC), and with impaired ability to secrete g-IFN. In contrast, other investigations suggested that human ECs are unable to stimulate a functional response by freshly purified CD8+ T cells.
Similarly to all the in vitro studies described, the few studies analyzing antigen presentation by EC in vivo have led to conflicting conclusions. A caveat that must be borne in mind is that the microvascular endothelium has different phenotypic characteristics in different species. For example, murine EC express the costimulatory molecule CD80 but not CD40, whereas human ECs are totally B7-negative but express functional CD40. A recent study has suggested that EC allorecognition by adoptively transferred CD8+ T cells in the absence of CD4+-mediated T cell help and professional hematopoietic antigen-presenting cells leads to heart allograft rejection in MHC class I recipients. However it has clearly been shown that secondary lymphoid organs are required for transplant rejection, implying that antigen presentation by graft-resident dendritic cells migrated in the host lymphoid tissue is required for alloreactive T cells to be activated. The reason for these discrepancies remains unresolved.
A possible compromise of these opposite views is the recent definition of EC as "semiprofessional" APC, due to their ability to enhance T-cell responsiveness and cytokine production without eliciting full T-cell activation. Given that ECs might display tissue-specific as well as foreign peptides, full competence as APC would be potentially dangerous in terms of self-tolerance, and induction of nonresponsiveness following cognate recognition of ECs would be likewise undesirable for virus-specific T cells entering an infected tissue. Thus the neutral effects on subsequent T-cell behavior, described earlier, could be advantageous to overall immunoregulation. Of course the primary immunological function of EC is to regulate lymphocyte trafficking. Recent developments in vascular immunology suggest that antigen presentation by EC might induce a different activation response, inducing T-cell transendothelial migration.
Was this article helpful?