Our laboratory recently identified a mechanism of immunosuppression by human regulatory T cells (Tregs) that implies transmembrane protein GARP and the production of the potently immunosuppressive cytokine TGF-β1. We showed that production of active TGF-β1 and immunosuppression by Tregs can be inhibited with blocking anti-GARP monoclonal antibodies. Because Tregs inhibit anti-tumor immune responses and thus play detrimental roles in cancer patients, blocking anti-GARP monoclonal antibodies are currently investigated for their potential use in cancer immunotherapy. The goal of this study was to determine whether cells other than Tregs also express GARP and produce active TGF-β1 in a GARP-dependent manner. Our objective was to predict potential toxicities of the anti-GARP antibodies that could be used in the clinics, and to gain new insights into the biology of GARP and GARPdependent TGF-β1 activation by non-Treg cells. We discovered that stimulated human B lymphocytes express GARP and produce active TGF-β1 in a GARP-dependent manner, and that the autocrine activity of the B cell-derived TGF-β1 increases class switch recombination and production of IgA1. These findings clarify the source of active TGF-β1 needed for IgA production in humans. They also suggest that the toxicity of anti-GARP antibodies due to the blockade of B cell function should be limited, as the large majority of individuals with genetic deficiencies in IgA are asymptomatic. Finally, they open new avenues for the analyses of GARP functions in mammals, and may indicate potentially interesting additional applications for the use of anti-GARP antibodies as therapeutic reagents.