investigator_user investigator user funding collaborators pending menu bell message arrow_up arrow_down filter layers globe marker add arrow close download edit facebook info linkedin minus plus save share search sort twitter remove user-plus user-minus
  • Project leads
  • Collaborators

Developing New Immunotherapies Based of CD4+ T Cells

Nicholas Restifo

2 Collaborator(s)

Funding source

National Cancer Institute (NIH)
The roles of CD4+ cells in anti-tumor immunity remains controversial and poorly understood. CD4+ T cells can differentiate in diverse subsets, but these T cell subsets have not been comprehensively studied in tumor-bearing mice. Although several mouse models have been previously described, they often involve the prevention of cancer cells modified to express potentially highly-immunogenic foreign or surrogate antigens (e.g. OVA or H-Y in female hosts) or their treatment as unrealistically small, non-vascularized pulmonary metastases. To more closely mimic human disease, we created transgenic mice expressing a class II-restricted TCR recognizing an endogenous melanocyte differentiation antigen called tyrosinase-related protein 1 (TRP-1 or gp75). We had indirect evidence that a recombinant vaccinia virus (rVV) encoding TRP-1 elicited a Th-dependent autoimmune vitiligo, but attempts to clone CD4+ cells reactive to TRP-1 were unsuccessful, perhaps due to tolerance-related mechanisms. We successfully generated B16/CIITA-specific CD4+ cells from antigen-negative Bw mice, identified the TRP-1 epitope they recognize corresponding to amino acids 113-127 (CRPGWRGAACNQKI) and cloned their TCR (Vbeta14/Valpha3.2). A founder with Y-chromosome-linked transgene was designated TRP-1 TCR transgenic and bred onto a RAG1-/- background to eliminate the rearrangement of endogenous TCR. To evaluate the degree of protection against the growth of melanoma conveyed by TRP-1 cells, we inoculated BwRAG1-/-TRP-1 Tg mice with B16 cells. Despite the presence of a large population of melanoma-specific T cells, the growth rate and lethality of the B16 tumor was only minimally delayed in TCR transgenic animals. A similar lack of protection against the tumor has been demonstrated previously in other TCR transgenic models and attributed to immunologic ignorance and lack of co-stimulatory signaling. We also found that adoptive transfer of cells cultured in IL-2 had minimal impact on tumor growth. We generated different TRP-1 T helper subsets (Th1, Th17 and Th0) by culturing the cells under strictly defined polarizing conditions. The degree of expansion of both Th1 and Th17 was similar and higher in comparison to the neutral (Th0) condition. To assess subset commitment, we analyzed the phenotype, gene expression patterns and cytokine secretion profiles. Th1 cells produced high quantities of IFN-gamma, TNF-alpha, IL-10 and lower amounts of IL-2 upon antigen stimulation. As expected, only Th17-skewed cells secreted significant high quantities of IL-17A and CCL20 (MIP3) as well as IL-2, IL-6 and IL-21 and produced smaller but amounts of IFN-gamma and TNF-alpha. Non-polarized Th0 cells secreted IFN-gamma at intermediate levels but did not produce IL-17A. Recognition was stronger upon exposure to B16 engineered to express the MHC class II transactivator CIITA, but there was no release of cytokines upon incubation with TRP-1-negative tumor cell lines MCA205 and EL-4. Intracellular staining upon restimulation demonstrated that virtually all Th1 cells produced IFN-gamma, while only those cells programmed in TGF-beta and IL-6 contained a significant percentage of IL-17A-secreting lymphocytes were also capable of producing IFN-gamma. Microarray analysis of in vitro polarized cells populations showed a striking up-regulation of IL-17A (105-fold difference) and CCL20/MIP3 (95-fold difference). mRNAs encoding IL-17F and IL-22, which are additional markers of Th17 polarization, were also elevated. As suggested by ELISA results, mRNA levels for IL-2 and IL-21 as well as another common gamma-chain cytokine, IL-9, were higher in Th17-polarized population. We treated C57BL/6 mice bearing 10-12-day-established tumors with adoptive transfer of Th0, Th1 or Th17 cells. Surprisingly, only Th17-skewed cells mediated a significant (p=0.001 vs.Th0 and Th1-treated groups) tumor regression leading to a complete cure and the long-term survival (Fig. 39). Despite initial tumor shrinkage, all animals injected with Th0 or Th1 cells relapsed and eventually had to be sacrificed because of melanoma progression. Long-term surviving mice developed vitiligo in both Th1 and Th17-treated groups, but the severity of this autoimmune manifestation was far greater in the Th17-treated animals. In addition, the absolute numbers of Vbeta14+CD4+ splenocytes recovered after transfer from Th17-treated animals were consistently the highest, indicating persistence and/or proliferation advantage of cells polarized with TGF-beta and IL-6 over the other subtypes. To test the roles of cytokines produced by Th17 cells, we used neutralizing antibodies to IL-17A, IFN-gamma or IL-23, which is known to support the survival of Th17 T lymphocytes. Unexpectedly, tumor rejection was completely inhibited only by anti-IFN-gamma treatment while the effects of in vivo neutralization of IL-17A and IL-23 did not reach statistical significance (p>0.05 vs. Th17 isotype control) (Fig. 40). In exploring this further, we found that therapy with Th17-polarized cells was equally effective in both WT (C57BL/6) and IFN-gamma-/- hosts, indicating that host IFN-gamma did not play a major role. In sharp contrast, the ability of hosts to receive the IFN-gamma signal was critical because Th17-skewed cells were essentially ineffective in IFN-gammaR-/- mice. It is possible that Th17 phenotype is not stable and undergoes evolution into type 1 response in vivo after adoptive cell transfer in the TRP-1 model. Many cytokines involved (IL-6, IL-21, IL-23) signal via STAT3. STAT3-mediated signaling has been implicated in cancer development and has been associated with anti-apoptotic and pro-survival effects. While STAT3 may have a role in de novo cancer formation, it is also possible that in mature T cells it might have some effects that are beneficial and allow for better survival upon adoptive cell transfer. Finally, the impact of distinct cytokines on lineage commitment decisions is better established in CD4+ than CD8+ T cells, but it seems likely that similar differentiation plasticity might occur in CD8+ T cells, as shown in a hypothetical model of CD8+ T cell differentiation that we have developed in the lab.

Related projects