Background: Chronic low-grade inflammation and alterations in innate and adaptive immunity were reported in Type 2 diabetes (T2D). Here, we investigated the abundance and activation of T cells in the bone marrow (BM) of patients with T2D. We then verified the human data in a murine model and tested if the activation of T cells can be rescued by treating mice with abatacept, an immunomodulatory drug employed for the treatment of rheumatoid arthritis. Clinical evidence indicated abatacept can slow the decline in beta-cell function.Methods: A cohort of 24 patients (12 with T2D) undergoing hip replacement surgery was enrolled in the study. Flow cytometry and cytokine analyses were performed on BM leftovers from surgery. We next compared the immune profile of db/db and control wt/db mice. In an additional study, db/db mice were randomized to receive abatacept or vehicle for 4 weeks, with endpoints being immune cell profile, indices of insulin sensitivity, and heart performance.Results: Patients with T2D showed increased frequencies of BM CD4+ (2.8-fold, p = 0.001) and CD8+ T cells (1.8-fold, p = 0.01), with the upregulation of the activation marker CD69 and the homing receptor CCR7 in CD4+ (1.64-fold, p = 0.003 and 2.27-fold, p = 0.01, respectively) and CD8+ fractions (1.79-fold, p = 0.05 and 1.69-fold, p = 0.02, respectively). These differences were confirmed in a multivariable regression model. CCL19 (CCR7 receptor ligand) and CXCL10/11 (CXCR3 receptor ligands), implicated in T-cell migration and activation, were the most differentially modulated chemokines. Studies in mice confirmed the activation of adaptive immunity in T2D. Abatacept reduced the activation of T cells and the levels of proinflammatory cytokines and improved cardiac function but not insulin sensitivity.Conclusions: Results provide proof-of-concept evidence for the activation of BM adaptive immunity in T2D. In mice, treatment with abatacept dampens the activation of adaptive immunity and protects from cardiac damage.