International Immunology Advance Access originally published online on May 25, 2006
International Immunology 2006 18(7):1067-1077; doi:10.1093/intimm/dxl040
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Modeling competition among autoreactive CD8+ T cells in autoimmune diabetes: implications for antigen-specific therapy
1 Department of Mathematics and Institute of Applied Mathematics, University of British Columbia, 1984 Mathematics Road, Vancouver, British Columbia, V6T 1Z2, Canada
2 Department of Microbiology and Infectious Diseases and Julia McFarlane Diabetes Research Centre, Faculty of Medicine, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta, T2N 4N1, Canada
3 Present address: Theoretical Biology/Bioinformatics, Utrecht University, Padualaan 8, 3584 CH Utrecht, the Netherlands
Correspondence to: A. F. M. Marée; E-mail: a.f.m.maree{at}bio.uu.nl
Antigen therapy remains a promising strategy for prevention and treatment of autoimmune diseases, but translating this strategy to clinical therapy has been largely unsuccessful. We have shown that development of autoimmune diabetes in non-obese diabetic (NOD) mice involves prevalent recruitment of CD8+ T cells recognizing epitopes of islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP). Administration of peptide analogs of IGRP206-214, the dominant epitope, reduced disease incidence but only under conditions that led to selective deletion of high-avidity T-cell clones. Peptide types or doses that resulted in elimination of all IGRP206-214-reactive T cells, regardless of avidity, promoted the recruitment of sub-dominant epitope-specific T cells and failed to prevent disease development. Here, we mathematically model competition of IGRP-reactive T-cell clones during spontaneous disease, and in response to peptide treatment. Based on realistic T-cell activation, proliferation and differentiation parameter values, our model shows that progression of spontaneous disease is characterized by (i) initial expansion of all IGRP206-214-reactive T-cell clones (irrespective of avidity) and (ii) slow replacement of T-cell clones recognizing peptide/MHC with low avidity by their high-avidity counterparts. This model helps understand the paradoxical outcomes of IGRP-based peptide treatment experiments. Furthermore, it predicts that slight deviations in dose or peptide affinity can lead to treatment failure or disease progression. This will occur if the treatment (i) increases the imbalance between competing IGRP206-214-reactive T-cell clones such that it favors rapid takeover of high-avidity clones or (ii) deletes all IGRP206-214-reactive clones, thereby creating a vacuum that promotes the recruitment of pathogenic sub-dominant specificities. Our data and model urge caution in the application of peptide therapy in autoimmunity.
Keywords: antigen therapy, autoimmunity, CTL, diabetes, mathematical modeling