The BALB/cByJ mouse strain displays an immunodominant T cell response directed at the same CD4+ T cell epitope peptide region in human IFN-β, as detected in a human population-based assay. BALB/cByJ mice also recognize a second region of the protein with a lesser magnitude proliferative response. Critical residue testing of the immunodominant peptide showed that both BALB/cByJ mice and the human population response were dependent on an isoleucine residue at position 129. A variant IFN-β molecule was constructed containing the single amino acid modification, I129V, in the immunodominant epitope. The variant displayed 100% of control antiproliferation activity. Mice immunized with unmodified IFN-β responded weakly in vitro to the I129V variant. However, BALB/cByJ mice immunized with the I129V variant were unable to respond to either the I129V variant or the unmodified IFN-β molecule by either T cell proliferation or Ag-specific IgG1 Ab production. This demonstrates that a single amino acid change in an immunodominant epitope can eliminate an immune response to an otherwise intact therapeutic protein. The elimination of the immunodominant epitope response also eliminated the response to the subdominant epitope in the protein. Modifying functionally immunodominant T cell epitopes within proteins may obviate the need for additional subdominant epitope modifications.

Material

Evidence suggests natural killer (NK) cells can mediate antiviral activity in HIV-infected humans. Here, we sought to investigate whether the level of NK cell activation in primary HIV infection not only influences the NK cell mediated elimination of HIV infected cells, but also affects the longevity and effectiveness of HIV-specific T cell responses. To address this hypothesis, we monitored HIV viral loads and HIV-specific adaptive immune responses in a humanized BLT mouse model of HIV infection, comparing NK cell-depleted and control mice. Depletion was achieved through the use of an anti-NKp46 antibody prior to HIV infection. Proportions of immune cell subsets and viremia were assessed at baseline and then once a week for at least 5 weeks post-infection using flow cytometry and qRT-PCR. We observed that the impact of NK cells on plasma viral loads varied between batches of mice reconstituted with different tissues. One set of NK cell depleted mice displayed a 4-fold increase in viremia as well as decreased levels of T cell immune activation at 3 weeks post-infection. In a second group of mice reconstituted with independent human tissue, NK cell depletion had no significant effect on viral loads. These results suggest that further investigation is warranted to identify factors accounting for an increased control of HIV replication associated with NK cells in some individuals and not others, such as expression of a protective combination of KIR and HLA alleles.