Recognition of the major histocompatibility complex restriction element modulates CD8(+) T cell specificity and compensates for loss of T cell receptor contacts with the specific peptide

Abstract

Triggering of a T cell requires interaction between its specific receptor (TCR) and a peptide antigen presented by a self-major histocompatibility complex (MHC) molecule. TCR recognition of self-MHC by itself falls below the threshold of detection in most systems due to low affinity. To study this interaction, we have used a read-out system in which antigen-specific effector T cells are confronted with targets expressing high levels of MHC compared with the selecting and priming environment. More specifically, the system is based on CD8(+) T cells selected in an environment with subnormal levels of MHC class I in the absence of beta2-microglobulin. We observe that the MHC restriction element can trigger viral peptide-specific T cells independently of the peptide ligand, provided there is an increase in self-MHC density. Peptide-independent triggering required at least four times the natural in vivo level of MHC expression. Furthermore, recognition of the restriction element at expression levels below this threshold was still enough to compensate for lack of affinity to peptides carrying alanine substitutions in major TCR contact residues. Thus, the specificity in TCR recognition and T cell activation is fine tuned by the avidity for self-MHC, and TCR avidities for peptide and MHC may substitute for each other. These results demonstrate a functional role for TCR avidity for self-MHC in tuning of T cell specificity, and support a role for cross-reactivity on "self" during T cell selection and activation.

Figure 1

H-2D^b expression in the absence of β₂m is TAP dependent. FACS^® analysis of H-2D^b and H-2K^b expression on Con A–activated blasts from B6 (A), β₂m^−/− (B), and TAP1/β₂m^−/− (C) mice using the B22-249.1 and Y3 mAbs, respectively. Bold and dotted lines represent H-2D^b and H-2K^b expression, respectively. Solid thin line represents staining with secondary antibody only.

Figure 2

TCR and CD8 coreceptor expression in β₂m^−/− and B6 CTLs. FACS^® analysis of TCR-α/β and CD8α in GP33-specific CTLs, using the PE-conjugated anti–TCR-α/β mAb H57-597 and the FITC-conjugated anti-CD8α mAb 53-6.7 (PharMingen), respectively. Solid lines represent B6 CTLs; dotted lines represent β₂m^−/− CTLs. In staining of CTL clones, the representative clones 2C10 (B6) and 3C5 (β₂m^−/−) were used. The histograms show CD8 expression and TCR expression in TCR⁺ and CD8⁺ cells, respectively.

Figure 3

Elevated self-MHC reactivity in β₂m^−/− CTLs. CTLs generated by synthetic LCMV GP33 peptide immunization (as described in Materials and Methods) in B6 (A) and β₂m^−/− (B) mice were tested against RMA-S, RMA, and RMA-S loaded with GP33 in a ⁵¹Cr-release CTL assay. (C) β₂m^−/− GP33-specific CTL lysis of ⁵¹Cr-labeled (hot) RMA target cells was subjected to cold target competition using cold (unlabeled) RMA-S, RMA, and RMA-S loaded with GP33. β₂m^−/− CD8⁺ CTL clones C10 (D), 27/30 (E), and 3C5 (F), and the B6 CTL clone 2C10 (G) were obtained by limiting dilution and tested against RMA-S, RMA, and RMA-S loaded with GP33.

Figure 4

Recognition of self-MHC at high ligand density by the CTL clone 27/30 is largely independent of peptide. (A) The β₂m^−/− CTL clone 27/30 was tested in a ⁵¹Cr-release CTL assay against RMA-S target cells loaded with LCMV GP33 peptide, influenza NP366 peptide, Yop249 peptide, unloaded RMA-S, or RMA-S incubated for 12 h at 26°C without peptide to stabilize “empty” MHC class I on the cell surface. (B) Clone 27/30 was tested against T2D^b target cells loaded with GP33, NP366, Yop249, and unloaded T2D^b. The T2 cell line was included as a D^b negative control.

Figure 4

Recognition of self-MHC at high ligand density by the CTL clone 27/30 is largely independent of peptide. (A) The β₂m^−/− CTL clone 27/30 was tested in a ⁵¹Cr-release CTL assay against RMA-S target cells loaded with LCMV GP33 peptide, influenza NP366 peptide, Yop249 peptide, unloaded RMA-S, or RMA-S incubated for 12 h at 26°C without peptide to stabilize “empty” MHC class I on the cell surface. (B) Clone 27/30 was tested against T2D^b target cells loaded with GP33, NP366, Yop249, and unloaded T2D^b. The T2 cell line was included as a D^b negative control.

Figure 5

CTL triggering by self-MHC is a low-avidity event that requires interaction with a high density of ligands. (A) Recognition by β₂m^−/− LCMV GP33-specific polyclonal CTLs of RMA in the absence or presence of GP33 peptide was blocked using the anti–H-2D^b conformation-specific mAb B22-249.1. (B) FACS^® analysis of B22-249.1 titration on RMA cells. (C) Recognition by the GP33-specific β₂m^−/− CTL clone 27/30 of T2D^b loaded with GP33, Yop249, NP366, or T2D^b without peptide was blocked using B22-249.1. (D) FACS^® analysis of B22-249.1 titration on T2D^b. The preincubation of target cells with peptide at 37°C did not affect expression of H-2D^b in these experiments (data not shown).

Figure 6

β₂m^−/− CTLs display increased peptide promiscuity in recognition of MHC–peptide ligands. Groups of three B6 and three β₂m^−/− mice were immunized with LCMV GP33 peptide as described in Materials and Methods. After 12 d, immune splenocytes were restimulated with GP33 for 6 d and tested in a ⁵¹Cr-release assay against RMA-S target cells loaded with titrated amounts of GP33 and GP33 variants: GP33 KAVYNFATM; GP33-4A KAVANFATM; GP33-34A KAAANFATM; and GP33-348A KAAANFAAM.

Figure 7

The TCR avidity threshold for T cell recognition. Hypothetical diagrams illustrating how the TCR affinities for peptide and MHC are interchangeable and mutually compensatory (A), and how the number of MHC–peptide ligands necessary for triggering inversely correlates with the TCR affinity for the ligand (B). (C) A and B are combined to generate a threshold composed of TCR affinities for MHC and peptide and the number of MHC–peptide complexes. In this threshold, all three contributing components can compensate for each other to generate the avidity necessary for T cell activation. The curve in A is presented as a straight line because a constant combined value of MHC–peptide avidity required for CTL triggering is subdivided into two parts, representing avidity for either MHC (x-axis) or peptide (y-axis). The shape of the curve in B is hypothetical. However, the hyperbolic shape will result if the total avidity required for triggering is constant, and if this avidity is a product of the number of ligands and the TCR affinity for each of these ligands (i.e., total triggering avidity = no. of ligands × TCR affinity for each ligand).