Activation of T cells recognizing self 60-kD heat shock protein can protect against experimental arthritis.

Abstract

Lewis rats are susceptible to several forms of experimental arthritis-induced using heat-killed Mycobacterium tuberculosis (adjuvant arthritis, or AA), streptococcal cell walls, collagen type II, and the lipoidal amine CP20961. Prior immunization with the mycobacterial 65-kD heat shock protein (hsp65) was reported to protect against AA, and other athritis models not using M. tuberculosis, via a T cell-mediated mechanism. Hsp65 shares 48% amino acid identity with mammalian hsp60, which is expressed at elevated levels in inflamed synovia. Several studies have reported cross-reactive T cell recognition of mycobacterial hsp65 and self hsp60 in arthritic and normal individuals. We previously described nine major histocompatibility complex class II-restricted epitopes in mycobacterial hsp65 recognized by Lewis rat T cells. Of these only one, covering the 256-270 sequence, primed for cross-reactive T cell responses to the corresponding region of rat hsp60. Here we have tested each hsp65 epitope for protective activity by immunizing rats with synthetic peptides. A peptide containing the 256-270 epitope, which induced cross-reactive T cells, was the only one able to confer protection against AA. Similarly, administration of a T cell line specific for this epitope protected against AA. Preimmunization with the 256-270 epitope induced T cells that responded to heat-shocked syngeneic antigen-presenting cells, and also protected against CP20961-induced arthritis, indicating that activation of T cells, recognizing an epitope in self hsp60 can protect against arthritis induced without mycobacteria. Therefore, in contrast to the accepted concept that cross-reactive T cell recognition of foreign and self antigens might induce aggressive autoimmune disease, we propose that cross-reactivity between bacterial and self hsp60 might also be used to maintain a protective self-reactive T cell population. This discovery might have important implications for understanding T cell-mediated regulation of inflammation.