News and Publications

Mechanism and Consequences of Tolerance to Self-Epitopes Expressed on Tumors

L. Sherman, D.J. Morgan, J. Lustgarten, J. Hernandez, T. Nugent, H. Kreuwel, J. Biggs, K. Holst

Tolerance results in elimination of T-cell clones that have high affinity for self-epitopes. This elimination is one way the immune system prevents autoimmunity. However, it is also a significant barrier to eliciting an immune response against tumor-associated antigens, which are also self-epitopes. We are studying the mechanism and consequences of tolerance to self-epitopes expressed on tumors. Two different types of self-antigens are being investigated in murine models. One is the p53 tumor-suppressor protein, which is available as a target of central and peripheral tolerance. The other is influenza hemagglutinin, which has been introduced as a transgene protein under the control of the rat insulin promoter (Ins-HA). It represents a target for peripheral tolerance. We are studying the mechanisms of tolerance to each of these proteins, their recognition by CD8⁺ T cells, and the ability of CD8⁺ T cells to eliminate tumors in vivo that express these antigens.

TUMOR-SUPPRESSOR PROTEIN

The p53 tumor-suppressor protein is inactivated in more than one half of all human tumors. Most often, this inactivation is accomplished through mutation that also leads to high levels of accumulation of the protein in transformed cells. Using HLA-A2.1/K^b transgenic mice deficient in p53, we have identified 2 peptide epitopes from the murine p53 protein that are recognized in association with A2.1. Conventional A2.1/K^b mice that express p53 cannot respond to 1 of these epitopes (residues 187--197) but do respond to the other (residues 261--269). This response is typically 10-fold lower than that of p53-deficient mice, indicating that tolerance has markedly dampened the response.

To determine the mechanism responsible for this reduced response, we compared the CD8⁺ T cells specific for the 261--269 epitope from p53-sufficient and p53-deficient mice. In collaboration with P. Lee and M. Davis, Stanford University, we are using tetramers of the A2.1 molecule containing the 261--269 peptide to determine if T-cell populations and CD8⁺ T-cell clones from the tolerant and nontolerant mice differ in the ability to bind antigen. The binding of such tetramers by murine cytotoxic T lymphocytes relies entirely on the affinity of the T-cell receptors (TCRs), because murine CD8 cannot bind the HLA class I molecule.

We observed that in contrast to findings in p53-deficient mice, cytotoxic T lymphocytes from tolerant mice that express p53 either cannot bind the tetramers or, if they do bind tetramers, express 10-fold lower levels of the CD8 molecule. Thus, the T cells appear to have reduced their avidity for the A2-peptide complex to escape deletion in 2 ways: either by reducing their ability to bind antigen or by reducing their level of expression of CD8. Future experiments will investigate the mechanisms responsible for these 2 ways to escape deletion.

HEMAGGLUTININ

Mice that express the Ins-HA gene are tolerant of hemagglutinin and have only low-avidity cells specific for the dominant hemagglutinin epitope recognized by CD8⁺ T cells. We are using a TCR transgenic line that recognizes a hemagglutinin peptide presented by the K^d molecule (clone 4 TCR) to investigate the mechanism of tolerance in these mice. We found no indication of thymic deletion of cells bearing clone 4 TCR in (clone 4 TCR x Ins-HA) F(1) mice, because CD8⁺ T cells expressing the clone 4 TCR mature normally. Indeed, these mice die of diabetes at a young age.

Why then, do Ins-HA mice with a normal TCR repertoire become tolerant? We are testing the hypothesis that when small numbers of hemagglutinin-specific T cells come out of the thymus, the cells may not be sufficiently activated by the presence of hemagglutinin in the pancreas to cause diabetes. Rather, the cells go through several rounds of proliferation and die by apoptosis. Clone 4 TCR cells were labeled with an internal fluorescent marker and then injected into Ins-HA mice. Fluorescence-activated cell sorting revealed that the injected cells become activated to divide in the pancreatic lymph nodes. Diabetes does not occur under these conditions, and the cells gradually become deleted. This finding suggests that peripheral tolerance can result in elimination of T cells with high avidity for a self-epitope. However, the number of T cells that can be dealt with by this mechanism is relatively small and may be overwhelmed by a TCR transgenic repertoire.

PUBLICATIONS

Kafri, T., Morgan, D., Krahl, T., Sarvetnick, N., Sherman, L.A., Verma, I. Cellular immune response to adenoviral vector infected cells does not require de novo viral gene expression: Implications for gene therapy. Proc. Natl. Acad. Sci. U.S.A., in press.

Lustgarten, J., Marks, J., Sherman, L.A. Redirecting effector T cells through their IL-2 receptors. J. Immunol., in press.

Morgan, D., Kreuwel, H.C.T., Fleck, S., Levitsky, H.I., Pardoll, D.M., Sherman, L.A. The effect of tolerance on rejection of tumor cells expressing self-antigen. J. Immunol. 160:643, 1998.

Sherman, L.A., Theobald, M., Morgan, D., Hernandez, J., Bacik, I., Yewdell, J. Strategies for tumor elimination by cytotoxic T lymphocytes. Crit. Rev. Immunol. 18:47, 1997.

Theobald, M., Ruppert, T., Kuckelkorn, U., Haubler, A., Biggs, J., Levine, A.J., Huber, C., Koszinowski, U.H., Kloetzel, P.-M., Sherman, L.A. A mutational hot spot in p53 protects cells from lysis by CTL specific for a flanking epitope. J. Exp. Med., in press.