Response to Comment on “Tim-3 directly enhances CD8 T cell responses to acute Listeria monocytogenes infection”

We thank Dr. Kuchroo and colleagues for their interest in our paper (1) and their commentary. The studies we described consist of two parts. In the first part, we compared CD8 T cell responses by wild-type (WT) C57BL/6J (B6) mice and Tim-3 KO mice to infection with OVA-expressing Listeria monocytogenes (LM-OVA). We found that Tim-3 KO mice had reduced CD8 T cell responses to LM-OVA relative to control mice. In the second part, we performed co-adoptive transfers of OVA-specific WT and Tim-3 KO OT-I CD8 T cells into WT hosts and then analyzed responses by the transferred cells to LM-OVA infection. The advantages of this approach are that the absence of Tim-3 expression is restricted to the transferred Tim-3 KO OT-I cells and that responses by WT and Tim-3 KO OT-I cells within the same WT host can be analyzed simultaneously. We found that, relative to WT cells, responses by Tim-3 KO OT-I cells to LM-OVA were impaired, indicating that Tim-3 directly augments CD8 T cell responses to the infection.

Dr. Kuchroo and colleagues state the concern that our Tim-3 KO mice carry the 129 haplotype for the Tim gene locus and therefore express forms of the Tim proteins that differ from those expressed by control B6 mice due to gene polymorphisms between the two mouse strains. To our knowledge, polymorphisms affecting mouse Tim-2 or Tim-4 have not been described. However, as pointed out in the commentary, polymorphisms affecting mouse Tim-1 have been associated with differences in CD4 T cell responses to antigen challenge (2). To determine whether these polymorphisms may have been a factor in our studies, we performed RT-PCR analysis, which indicated that our Tim-3 KO mice express the 129 form of Tim-1. Thus, we cannot rule out that differences in Tim-1 had some influence on the results from the first part of our studies where we compared CD8 T cell responses by WT B6 and Tim-3 KO mice to LM-OVA infection.

We also asked whether differences in Tim-1 were a factor in the second part of our studies in which responses by WT and Tim-3 KO OT-I CD8 T cells to LM-OVA were analyzed following co-adoptive transfer into WT hosts. Here, the relevant polymorphisms would only be a factor if CD8 T cells expressed Tim-1. To address this possibility, we performed flow cytometric analysis of naïve CD8 T cells and effector CD8 T cells generated by LM-OVA infection. Although we detected Tim-1 expression by non-CD8 T cells, we found no evidence that Tim-1 was expressed by naïve or effector CD8 T cells. These data indicate that differences in Tim-1 were not a confounding factor in our co-adoptive transfer experiments. Thus, our conclusion that Tim-3 can function to directly promote CD8 T cell responses remains valid.

Lastly, we would agree that our studies do not invalidate all the previous work showing that Tim-3 functions as an inhibitory receptor. Nonetheless, our results suggest that the role of Tim-3 in regulating T cell responses is more complex than previously thought.