Tim-3 co-stimulation promotes short-term effector T cells, restricts memory precursors and is dispensable for T cell exhaustion

Tim-3 is highly expressed on a subset of T cells during T cell exhaustion, in settings of chronic viral infection and tumors (1, 2). Using LCMV Clone-13, a model for chronic infection, we have found that Tim-3 is neither necessary nor sufficient for the development of T cell exhaustion. Nonetheless, expression of Tim-3 was sufficient to drive resistance to PD-L1 blockade therapy during chronic infection. Strikingly, expression of Tim-3 promoted development of short-term effector T cells, at the expense of memory precursor development, after acute LCMV infection. These effects were accompanied by increased Akt/mTOR signaling in T cells expressing endogenous or ectopic Tim-3. Conversely, Akt/mTOR signaling was reduced in effector T cells from Tim-3 deficient mice. Thus, Tim-3 is essential for optimal effector T cell responses, but may also contribute to exhaustion, by restricting development of long-lived memory T cells, including PD-1int \"stem-like\" exhausted T cells that expand during PD-1 pathway blockade. Taken together, our results suggest that Tim-3 is actually more similar to co-stimulatory receptors that are upregulated after T cell activation, rather than a dominant inhibitory protein like PD-1. These findings have significant implications for the development of anti-Tim-3 antibodies as therapeutic agents.\n\nSignificanceDuring a chronic viral infection, prolonged exposure to viral antigens leads to dysfunction or \"exhaustion\" of T cells specific to the virus, a condition also observed in T cells that infiltrate tumors. The exhausted state is associated with expression of specific cell-surface proteins, some of which may inhibit T cell activation. Expression of Tim-3 is associated with acquisition of T cell exhaustion, although it is also expressed transiently during acute infection. Here we provide evidence that a major function of Tim-3 is to enhance T cell activation, during either acute or chronic viral infection. However, Tim-3 is not required for development of exhaustion. Thus, we propose that Tim-3 would be better described as a stimulatory, rather than inhibitory, protein.