Mechanistic and therapeutic role of the CD137-CD137L axis in Type 1 Diabetes

PROJECT SUMMARY
Our labs have been extensively studying CD137-CD137L immune signaling in T1D. These studies have
uncovered additional critical mechanistic issues which we propose to study in this application. The regulation of
the CD137-CD137L system is complex, mainly because both CD137 and CD137L signal into the cells
expressing them, which must be taken into account in any analysis of CD137 and CD137L signaling. In our
preliminary data we identify many specific instances in which the effect of CD137 or CD137L signaling differs
according to the type of immune cells. FOXP3+ regulatory CD4 T cells (Tregs), antigen-presenting cells
(APCs), and effector T cells all express different levels of these key molecules with distinct kinetics during the
immune response. The details are critical for understanding how interventions in this system, most specifically
with soluble CD137 (sCD137) which down-regulates immunity via CD137L, affect the course and progression
of T1D. We have recently published that treatment with sCD137, which is expressed in vivo mainly by Tregs
that constitutively express CD137, can ameliorate acute T1D. We demonstrate that sCD137 directly
suppresses mouse CD4 and CD8 T cell activation through binding to CD137L expressed on these cells.
Critically, we have recently proven that CD137L expression on T cells is critical for Treg mediated suppression.
Similar to our findings in mice, we found that human CD25hi CD127low Tregs are the primary source of
sCD137. As in mice, hu-sCD137 inhibited proliferation of human peripheral CD4 T cells. Importantly, we found
that serum sCD137 was lower in human T1D patients compared to age-matched unrelated controls, analogous
to our finding that serum sCD137 is decreased in NOD mice compared to NOD mice congenic for the T1D
protective B10-derived Idd9.3 interval. In addition, we have recently published conclusive proof, using allele-
specific knockout mice, that Tnfrsf9 (encoding CD137) is the diabetogenic gene in the NOD Idd9.3 interval.
Overall, these results indicate that the biology of CD137 and the consequence of its interaction with CD137L
are very similar in mouse and human T1D. Importantly, however, in our investigation of the biology of CD137L
(the target of sCD137), we have recently published that T1D could not be transferred into mice lacking
expression of CD137L on myeloid APCs. Moreover, CD137L expressing myeloid APCs are essential for
accumulation of β-cell specific autoreactive CD8 T cells. These findings highlight the increasingly recognized
role of innate immunity in the initiation and perpetuation of autoimmune T1D, and the need for additional
mechanistic studies of CD137L on myeloid APCs, the critical gatekeeper for T cell entry into the islet. Given
these findings we propose the following aims. Aim 1: Mechanism by which the CD137-positive Treg subset
suppresses immune activation. Aim 2: Mechanistic role of CD137L-expressing myeloid APCs in T1D
pathogenesis and as a target for therapeutic effect of sCD137. Aim 3: Use novel human and mouse soluble
CD137-Fc proteins to test efficacy in humanized mouse models and on human monocytes and T cells.

Grant Number: 5R01DK107541-10
NIH Institute/Center: NIH
Principal Investigator: Yi-Guang Chen