Thymic generation of mature T cells is deficient in the absence of secondary TCRα rearrangements. (A) Comparison of numbers of thymocytes in 6 week old B6 and TCRα^+/− mice. Each point represents a single mouse, n = 17, 4 independent experiments, mean ± sem, Student’s t test. (B–E) The effect of secondary TCRα rearrangements on T cell development examined by comparing thymi of B6 and TCRα^+/− mice by flow cytometry, n = 10, 3 independent experiments. (B) Representative plots of CD4 and CD8 labeling of thymocytes from individual mice. (C) Percentages of mature CD4⁺ or CD8⁺ single positive (SP), CD4⁺CD8⁺ double positive (DP), or CD4⁻CD8⁻ double negative (DN) thymocytes. Each point represents a single thymus, mean ± sem, Student’s t test. (D) Representative plots of CD3 and CD69 labeling of thymocytes from individual mice. (E) Percentages of post-positive selection CD3^high thymocytes. Each point represents a single thymus, mean ± sem, Student’s t test.

Deficiency of secondary TCRα rearrangements impairs positive selection. (A–B) BrdU (1.6 mg) was administered i.p. on d 0. (A) Thymocyte BrdU incorporation was evaluated at 24 h post-injection (n = 6, 3 independent experiments, mean ± sem, Student’s t test). (B) Thymocyte development in Brd U-pulsed B6 and B6.TCRα^+/− mice compared by analyzing development of CD3^high post-positive selection thymocytyes by flow cytometry at 24 h, 48 h, and 96 h post-BrdU injection. Mean ± sem of 6 mice per group in 3 independent experiments, Student’s t test. (C–E) Pre-selection thymocytes from congenically-marked B6.Ly5.1 and TCRα^+/−.Thy1.1 mice were enriched by magnetic bead selection, combined at a 1:1 ratio, and 10⁷ cells transferred by intrathymic injection into B6 mice to examine development (n = 8, 3 independent experiments). (C) Transferred cells were detected 7 d after injection by flow cytometric analysis of thymocytes. Representative plot with congenically-marked transferred cells expressed as percentage of total thymocytes. (D) Development of post-selection CD3^high thymocytes was compared between B6.Ly5.1 and TCRα^+/−.Thy1.1 thymocytes in individual mice. Transferred populations in a single mouse are linked by the line between points, paired ratio t test. (E) Ratio of B6.Ly5.1 to B6.Cα^+/−.Thy1.1 cells among total thymocyes and post-selection CD3^high thymocytes. Mean ± sem, paired ratio t test. (F) Flow cytometry analysis of thymocytes from 6 wk old B6 and TCRα^+/− mice to assess thymocyte apoptosis. Thymocytes were labeled with CD4, CD8, Annexin V, and stained with 7-AAD. Each point represents a single mouse (n = 7) from 3 independent experiments, mean ± sem, Student’s t test. (G) Intrinsic thymocyte survival was compared between thymocytes from 6 wk old B6 and TCRα^+/− mice by in vitro culture of 10⁷ thymocytes/ml. Cultures were sampled at indicated time points and labeled for CD4, CD8, Annexin V, and stained with 7-AAD. Points are mean ± sem of 6 thymic cultures from 3 independent experiments, data analyzed by ANOVA.

Elimination of secondary TCRα rearrangements does not broadly alter the peripheral T cell compartment. (A) Comparison of numbers of T cells in the spleens of 6 week old B6 and TCRα^+/− mice. Each point represents a single mouse (n = 6) from 3 independent experiments, mean ± sem, Student’s t test. (B) Comparison of CD4⁺ and CD8⁺ T cell composition, measured by flow cytometry of splenocytes. Mean ± sem of 6 mice from 3 independent experiments, Student’s t test. Comparison of (C) TRAV and (D) TRAJ gene segment use by DNA sequence analysis of splenic T cells from 6 week old B6 and TCRα^+/− mice (3 mice each, 23,877 – 34,995 TCRα sequence reads/mouse). Mean ± sd for each group, data analyzed by ANOVA.

Elimination of secondary TCRα rearrangements results in elimination of specific TCRs from the naive T cell repertoire. TRAV14 T cell receptor libraries were generated from TCRα^+/− TCRVα2⁺ T cells and B6 TCRVα2⁺ dual TCR T cells (identified by co-expression of TCRVα3, TCRVα8, or TCRVα11) sorted by flow cytometry. (A) Three independent libraries per group were sequenced and generated 141,353 B6 TRAV14⁺ TCRα sequences and 148,228 TCRα^+/− TRAV14⁺ TCRα sequences yielding 23,039 total unique sequences divided among the groups as indicated. Analysis of (B) CDR3 length, (C) CDR3 amino acid composition, and (D) TRAJ gene segment use was compared between the 2 groups. (E) CDR3 sequence abundance was determined as a proportion of total TRAV14 transcripts for each population. The 10 most abundant CDR3 sequences for both populations are shown with corresponding frequencies. (F) Total TRAV14 CDR3 sequence frequencies for both populations, with correlation and 99% confidence interval determined by linear regression (Pearson correlation analysis). (G) The 5 most abundant TRAV14 CDR3 sequences unique to each population are shown with corresponding frequencies.

Secondary TCRα rearrangements specifically increase the alloreactive and autoreactive T cell repertoire. Splenocytes, axillary, inguinal, and popliteal lymph nodes were collected from B6 and TCRα^+/− mice, incubated with APC- or PE-labeled allogeneic pMHC tetramers, enriched by magnetic selection using anti-APC or anti-PE beads, and singlet T cells were analyzed by flow cytometry. (A) Representative B6 mouse sample labeled with TFR/I-E^k is shown. (B) Reactivity of pMHC tetramer-labeled cells was assessed by sorting tetramer⁺ and tetramer⁻ T cells by flow cytometry and measuring IFN-γ production following 18 h culture with CHO-E^k cells and 10 µM CD22 or 50 µM TFR peptide. IFN-γ production was measured by intracellular cytokine staining, with percent of IFN-γ⁺ cells indicated. Data are representative examples of 3 independent experiments. (C) The frequency of alloantigen-reactive T cells was calculated as the number of tetramer⁺ cells/10⁵ T cells collected from each mouse. Data points are individual mice (n = 8–11) from 5 independent experiments, mean ± sem, Student’s t test. (D) The frequency of autoantigen- and cognate peptide-reactive T cells was calculated as the number of tetramer⁺ cells/10⁵ T cells collected from each mouse. Data points are individual mice (n = 5–6) from 3 independent experiments, mean ± sem, Student’s t test.

Elimination of secondary TCRα chains reduces T cell in vivo alloresponse. Congenically-marked B6.Ly5.1 and TCRα^+/−.Thy1.1 T cells (both H-2^b) were mixed at a 1:1 ratio, pulsed with CFSE, and injected into lethally-irradiated MHC mismatched B6.K (H-2^k) recipients. (A) Pre-injection cells were labeled for congenic markers to confirm ratio of cells. Representative example from 3 independent experiments. Transferred cells were analyzed at 24 h by flow cytometry of splenocytes. (B) Cells were labeled for congenic markers to examine ratio of TCRα^+/− and B6 T cells pre-transplant and 24 h post-transfer. Results shown are mean ± sem of 12 recipient mice from 3 independent experiments, Student’s t test. (C) CFSE dilution was analyzed in congenically marked cells recovered 24 hours post-transfer. Representative example from 1 recipient mouse. (D) In vivo proliferative response to allogeneic stimulation was assessed in recovered cells by comparing the percentage of cells having undergone at least 1 cell division after 24 h. Data are composite of 12 recipient mice from 3 independent experiments, mean + sem, Student’s t test.

Secondary TCRs enable reaction to closely-related allogeneic peptide antigens. CD22/I-E^k alloantigen-reactive T cells were collected from B6 and TCRα^+/− mice by tetramer-labeling magnetic enrichment and flow cytometry sorting strategy and cultured with CHO-E^k cells for 24 h, either in the presence of the CD22 peptide, or CD22 peptide APLs P2A (H to A change at TCR contact P2) and P5S (G to S change at TCR contact P5). PMA (20 ng/ml) and ionomycin (1 µM) was used as a positive control. (A) T cell responses were measured by flow cytometry with intracellular labeling for IFN-γ production. Representative experiment shown. (B) Aggregate data from 3 independent experiments with APL response normalized to the response against wild-type CD22 peptide, mean ± sem, ANOVA.