Patients with peripheral artery disease demonstrate altered expression of soluble and membrane-bound immune checkpoints by peripheral blood immune cells

Abstract

Cardiovascular events seem to occur more frequently after treatment with immune checkpoint inhibitors. This suggests a possible role of immune checkpoints in accelerating atherosclerosis formation. We therefore aimed to assess the expression of soluble and membrane-bound immune checkpoints in patients with peripheral artery disease (PAD). We assessed the levels of 14 soluble immune checkpoints in fasting blood plasma of PAD patients (intermittent claudication, n= 37) and healthy controls (HCs, n=39) by Multiplex protein assay. We measured the levels of T-cell immunoglobulin and mucin domain 3 (TIM-3), programmed cell death 1 ligand 1 and 2 (PD-L1/2), B-and T-lymphocyte attenuator (BTLA), programmed cell death 1 (PD-1), glucocorticoid-induced TNRF-related protein (GITR), herpesvirus entry mediator (HVEM), cytotoxic T-lymphocyte associated protein 4 (CTLA-4), indoleamine 2,3-dioxygenase (IDO), CD28, CD27, CD80, CD137 and lymphocyte-activation gene 3 (LAG-3). The surface expression of CD28, PD-L2, PD-1, GITR, LAG-3, TIM-3 and BTLA on peripheral blood immune cells was determined by flow cytometry. Cytokine production capacity was measured by flow cytometry in TIM-3+ T cells to determine immune exhaustion. Soluble levels of PD-L2 were decreased in PAD patients, but only in females, whereas soluble levels of TIM-3 showed a trend towards increased concentration in female PAD patients. All monocyte subsets had increased frequencies of PD-L2+ cells in PAD patients. CD4+ T cells had increased frequencies of TIM-3+ cells, which showed little overlap with other immune exhaustion markers. TIM-3+ CD4+ T cells appeared to have a low capacity to produce pro-inflammatory cytokines, but a higher capacity to produce IL-10 compared to TIM-3-CD4+ T cells. We found that PAD patients show differences in expression of both membrane-bound and soluble immune checkpoints. Some of these differences might be caused by prolonged immune activation, although immune exhaustion markers did not always overlap.