Direct measurement of T cell receptor affinity and sequence from naïve antiviral T cells

Abstract

T cells recognize and kill a myriad of pathogen-infected or cancer cells using a diverse set of T cell receptors (TCRs). The affinity of TCR to cognate antigen is of high interest in adoptive T cell transfer immunotherapy and antigen-specific T cell repertoire immune profiling because it is widely known to correlate with downstream T cell responses. We introduce the in situ TCR affinity and sequence test (iTAST) for simultaneous measurement of TCR affinity and sequence from single primary CD8(+) T cells in human blood. We demonstrate that the repertoire of primary antigen-specific T cells from pathogen-inexperienced individuals has a surprisingly broad affinity range of 1000-fold composed of diverse TCR sequences. Within this range, samples from older individuals contained a reduced frequency of high-affinity T cells compared to young individuals, demonstrating an age-related effect of T cell attrition that could cause holes in the repertoire. iTAST should enable the rapid selection of high-affinity TCRs ex vivo for adoptive immunotherapy and measurement of T cell response for immune monitoring applications.

Figure 1. Overview of the iTAST method

(A) A human blood sample is enriched for CD8⁺ streptamer⁺ T cells, sorted using flow cytometry (B), and then undergo streptamer dissociation by addition of excess biotin. (C) T cells are affinity-tested using micropipette adhesion assay, (D) and then picked for paired single-cell TCR sequencing.

Figure 2. Verification of single-cell 2D affinity measurement by iTAST

(A) Adhesion curve for an HCV-specific CD8⁺ T cell clone against HLA-A2-CD8mut/HCV. (B) TCR expression of a HCV-specific CD8⁺ T cell, and its expression after streptamer staining and dissociation. (C) TCR expression of CD8⁺ T cells that were stained with phenotypic antibodies (CD27, CD45RA, CD57, CXCR3, CD95, CD45RO, and CCR7) for 3 hours at room temperature to mimic the environment for single-cell 2D affinity measurement. (D) Error analysis of factors associated with single-cell 2D affinity measurement. Factor I, T cell to T cell variation of TCR site density assessed by using one TCR antibody conjugated RBC against multiple primary CD8⁺ T cells, each point indicates one T cell. Factor II, variation of TCR site density in different areas on the same T cell assessed by using one TCR antibody conjugated RBC against different areas of one T cell, each point indicates one area of the same T cell (two T cells tested). Although the absolute adhesion frequencies are different between these two T cells, the variations of different areas of one T cell are similar between these two T cells. Factor III, RBC to RBC variation of MHC site density assessed by using one HCV-specific CD8⁺ T cell clone against multiple HLA-A2-CD8mut/HCV coated RBCs, each point indicates one RBC. (E) Simulation of coefficient of variation at different adhesion frequency values using values measured from Fig. 2D and eq. S2.

Figure 3. Comparison of single-cell 2D affinity by iTAST with conventional 2D affinity and SPR

(A) Correlation of single-cell 2D affinity and conventional 2D affinity from HCV-specific CD8⁺ T cell clones (N = 43, line denotes log-log linear regression, two-tailed t-test on the slope, p < 10⁻⁴) (B) 2D affinity for a primary CD8⁺ T cell and its clone after being picked into culture well for in vitro expansion. Tabular data available in table S1 (C) Correlation of single-cell 2D affinity of the native NY-ESO-1:157–165, denoted in red, and 6 peptide variants on HLA-A2-CD8mut against the 1G4 TCR expressed on the Jurkat cell line, versus 3D affinity by a previously published SPR study (3) (N = 7, line denotes log-log linear regression, two-tailed t-test on the slope, p < 0.01). Tabular data available in table S2.

Figure 4. Correlation of Single-Cell 2D affinity with functional capacity

(A) Single-cell 2D affinity of HCV-specific CD8⁺ T cells versus their lysis capacity, defined as the percent specific lysis of HCV peptide-pulsed JY cells (N = 43, Spearman two-tailed test, p < 10⁻⁴). (B) Single-cell 2D affinity versus peptide potency, defined as the peptide concentration required to induce 10% cell lysis (N = 15, line denotes log-log linear regression, two-tailed t-test on the slope, p < 0.005). HLA-A2-CD8mut/HCV tetramer median fluorescence intensity (MFI) versus (C) lysis capacity (Spearman two-tailed test) and (D) peptide potency (log-log linear regression, two-tailed t-test on the slope). Dashed line denotes limit of detection in tetramer staining.

Figure 5. Analysis of the primary HCV-specific CD8⁺ T cell repertoire in HCV-seronegative blood donors using iTAST

(A) Single-cell 2D affinity distribution of naïve primary HCV-specific CD8⁺ T cells within 12 HCV seronegative samples from 9 donors. Numbers indicate different donors and letters indicate separate blood draws from the same donor. Red bar indicates median value. 2D affinities transformed by log₁₀ for t-test in 4A–B (N = 63), ANOVA in 5A–C (N = 54), and t-test in aggregate 2D affinities from donor 4 and 5 (N = 117, p < 0.005). Two-tailed test is used in all cases. (B) Relation of single-cell 2D affinity with the number of pMHCs required to form one TCR-pMHC bond at 4 seconds of contact. (C) Median single-cell 2D affinity for each unique donor from Fig. 5A. Samples derived from the same donor were aggregated prior to calculation. (N = 9, two-tailed Mann-Whitney U Test, p < 0.05). Tabular data available in table S3 (D) TCRα and TCRβ V-gene usage of T cells from sample 4B, 5A, 6, and 7 in Fig. 5A.