Our studies of Tregs have revealed that TNF by acting on the TNFR2 receptor, which is most highly expressed by Tregs, unexpectedly results in the proliferative expansion and functional activation of Tregs both in mice and in man. In view of the well-known proinflammatory effects of TNF, our data showing that TNF in a more delayed manner can also down-regulate immune responses, is rather surprising. Furthermore, TNF together with IL-2 up-regulates the cell surface expression of TNFR2 and also of 4-1BB and OX-40 receptors on Tregs. Thus, TNF amplifies its stimulatory effect on Tregs by inducing 3 TNFRSF members. TNF interactions with TNFR2 also stabilizes the CD4+FOXP3+ T regulatory cell phenotype at inflammatory sites. One clarification of these unexpected effects of TNF is based on our data showing that TNF by activating Teffector cells also induce them to express more TNFR2 and to become more resistant to the suppressive effects of Tregs. Thus, activated inflammatory cells can prevail over the suppressive effects of Tregs. However, as inflammation subsides in healing wounds or in non-inflamed tumors, Tregs prevail. About 50% tumor infiltrating T cells (TIL's) develop into Tregs and express TNFR2. They are activated by tumor-derived TNF to be even more immunosuppressive than Tregs in peripheral lymphoid tissues. Suppression of Tregs results in more effective host antitumor responses and reduces tumor growth. We hypothesize that TNFR2 can be targeted by checkpoint inhibitors of Tregs. This hypothesis was tested by treating mice bearing small CT26 colon tumors with neutralizing anti-TNFR2 antibodies, which suppresses Tregs, and CpG, a TLR9 ligand, to promote cell mediated immunity. This resulted in 80% of the mice becoming tumor free and selectively resisting a subsequent CT26 tumor challenge. These treated mice developed considerable infiltration of their tumors by IFNgamma producing rather than Treg TIL cells and also reduced the numbers of MDSC in their spleens and tumors. These finding suggest that this therapy enabled the mice to exhibit considerable antitumor immunity. Identification of more potent antagonists of TNFR2 should yield even more effective checkpoint inhibition. The expression of TNFR2 is also upregulated on CD4+ FoxP3- T effector cells by TCR stimulation, but to a lesser extent than on CD4+ FoxP3+ Tregs. Nevertheless, adoptive transfer of CD4+ TNFR2+ Teffs induced full-fledged colitis in recipient Rag 1-1- mice, whereas CD4+ TNFR2-1- Teffs failed to do this. Thus, TNFR2 Teffs when activated by TNF also undergo proliferative expansion and behave as pathogenic Th1 polarized Teffs. Thus, inhibition of the TNF-TNFR2 pathway may provide a more effective treatment of IBD. This motivated us to collaborate with Dr. Dimiter Dimitrov and his colleagues who engineered the production of high affinity anti-human anti-TNFR2 antibodies. These antibodies are cytotoxic by ADCC for human, but not mouse cells expressing TNFR2. The antibody may be able to deplete both highly activated human Tregs and Teff cells. Although we can show this antibody together with NK cells has potent anti-TNFR2 effects in vitro, we can not evaluate its effects on mouse cells in vitro or in vivo because it is active only against human TNFR2. We therefore have to collaborate with a clinical group that would be interested in testing it in patients to determine if it acts as a potent checkpoint-like inhibitor. Although a number of companies have expressed an interest, none have licensed this antibody as yet. We are currently investigating the effect and mechanism of action of reagents that upregulate TNFR2 expressing Tregs to identify therapeutic approaches to inhibiting autoimmune conditions.
Showing the most recent 10 out of 15 publications
