Quantitating T cell cross-reactivity for unrelated peptide antigens

Abstract

Quantitating the frequency of T cell cross-reactivity to unrelated peptides is essential to understanding T cell responses in infectious and autoimmune diseases. Here we used 15 mouse or human CD8+ T cell clones (11 antiviral, 4 anti-self) in conjunction with a large library of defined synthetic peptides to examine nearly 30,000 TCR-peptide MHC class I interactions for cross-reactions. We identified a single cross-reaction consisting of an anti-self TCR recognizing a poxvirus peptide at relatively low sensitivity. We failed to identify any cross-reactions between the synthetic peptides in the panel and polyclonal CD8+ T cells raised to viral or alloantigens. These findings provide the best estimate to date of the frequency of T cell cross-reactivity to unrelated peptides ( approximately 1/30,000), explaining why cross-reactions between unrelated pathogens are infrequently encountered and providing a critical parameter for understanding the scope of self-tolerance.

FIGURE 1

Comparison of detection methods for OT-I activation. OT-I cells activated by the number of EL-4 cells indicated at various concentrations of SIINFEKL were tested for CD69 up-regulation (A), incorporation of [³H]thymidine (B), or synthesis of IFN-γ (C).

FIGURE 2

Effect of peptide pool on sensitivity of T cell activation. Competition for binding class I MHC and for presentation with 15 other peptides reduces the secretion of IFN-γ, as well as the consequent detection of SFU in IFN-γ ELISPOTs. To mimic surveying conditions, 300 T cell clones of Vβ17-bearing HLA-A2 flu-specific clones were mixed with 300 clones specific for the CMV pp65 Ag, NLVPMVATV, and 300 clones specific for the HIV-1 p17 Ag, SLYFNT VATL, as well as 10,000 HLA-A2-matched B cells purified from PBMCs. Data are normalized to 100% of maximum detectable SFU.

FIGURE 3

Surveys of H2-D^b peptide library with transgenic T cells reveal a single cross-reaction. Surveys of four transgenic TCR T cells detected a single, weak cross-reaction between MataHari T cells and an H2-D^b-restricted peptide. Positive controls are indicated by plus signs adjacent to the sample. Surveys shown here are representative of multiple trials conducted at 100 ng/ml/peptide. Not pictured are surveys at 1 ng/ml and surveys of H2-K^b peptides. A, OT-I surveys of H2-D^b peptides (in red) fail to reveal cross-reactions. Although pool 53 appears to produce a slight up-regulation of CD69 in a single repetition of the 100 ng/ml/peptide survey, this result did not repeat in other experiments or in surveys at 1 ng/ml/peptide. B, Surveys of F5 T cells against H2-D^b peptides (in blue). Pool 32 appears to produce a slight up-regulation of CD69, but it failed to replicate this effect in other experiments and in surveys at 1 ng/ml/peptide. C, Surveys of MataHari T cells against H2-D^b peptides (in yellow) reveal a weak cross-reaction. At 100 ng/ml/peptide, pool 61 produces a modest up-regulation of CD69 due to the presence of a cross-reactive peptide (indicated by a red arrow). Notably, this cross-reaction was not reliably detectable in surveys at 1 ng/ml/peptide. D, Surveys of Pmel-1 T cells against H2-D^b peptides (in green) fail to reveal cross-reactions. Pmel-1 T cells are known to cross-react with a human GP100 homolog (28). However, it is significantly more difficult to detect this cross-reaction in competition with other peptides at 1 ng/ml/peptide than at 100 ng/ml/peptide.

FIGURE 4

Binding affinity and CD69 up-regulation in MataHari T cells of HY and cross-reactive peptides. A, The cross-reactive YILCNMALL peptide was estimated have a K_d of ∼40 nM, compared with the agonist peptide K_d of ∼3 nM. B, The agonist WMHHMDLI peptide achieves 50% maximal activation of MataHari T cells at a concentration >2 logs lower than the cross-reactive YILCNMALL peptide (0.025 vs 6 nM).

FIGURE 5

ELISPOT surveys of HLA-A*0201 peptides fail to reveal agonist peptides. Surveys were conducted at 2 μg/ml/peptide. No strong cross-reactions were observed. Promising candidates for partial cross-reactions were further surveyed. SFU values are normalized against the values of at least two negative controls present on the same plate during the assay. Negative control values ranged consistently between 0 and 4. Positive controls are indicated by a plus sign adjacent to the sample. Positive controls too numerous to count are set to 100 SFU. Representative surveys of one group of VV peptides is shown for each group of T cells described below. A, Group 1 consisted of the influenza-specific flu clone, the HIV-1-specific G10 clone, and the CMV-specific 1999 clone. The M1 peptide GILGFVFTL was correctly identified and is located in pool 77. Pools selected for further surveying included pool 363 of the vaccinia peptides, pool 403 of the arenaviruses, and pool 15 of the LCMV peptides. B, Group 2 consisted of the HIV-1-specific T15 clone and the CMV-specific 016 and 005 clones. Pools selected for further surveying included pool 260 of the vaccinia peptides, and pools 77 and 46 of the influenza peptides. C, Group 3 consisted of the HIV-1-specific T16, T19, and T4 clones and the CMV-specific F11 clone. Pools selected for further surveying included pools 293 and 309 of the vaccinia peptides and pools 29 and 41 of the influenza peptides.

FIGURE 6

Partial surveys of H2-K^b library against OT-Is with dendritic cells and LPS fail to reveal further cross-reactions. Partial surveys of H2-K^b peptides with OT-I T cells, dendritic cells, and 5 ng/ml LPS (A) produce stronger IFN-γ signals than do surveys with EL-4 cells alone (B). However, surveys with dendritic cells and LPS also produce greater background noise. Repetitions of this experiment and IFN-γ ICS experiments failed to confirm any cross-reactions.