Human TRAV1-2-negative MR1-restricted T cells detect S. pyogenes and alternatives to MAIT riboflavin-based antigens

Abstract

Mucosal-associated invariant T (MAIT) cells are thought to detect microbial antigens presented by the HLA-Ib molecule MR1 through the exclusive use of a TRAV1-2-containing TCRα. Here we use MR1 tetramer staining and ex vivo analysis with mycobacteria-infected MR1-deficient cells to demonstrate the presence of functional human MR1-restricted T cells that lack TRAV1-2. We characterize an MR1-restricted clone that expresses the TRAV12-2 TCRα, which lacks residues previously shown to be critical for MR1-antigen recognition. In contrast to TRAV1-2(+) MAIT cells, this TRAV12-2-expressing clone displays a distinct pattern of microbial recognition by detecting infection with the riboflavin auxotroph Streptococcus pyogenes. As known MAIT antigens are derived from riboflavin metabolites, this suggests that TRAV12-2(+) clone recognizes unique antigens. Thus, MR1-restricted T cells can discriminate between microbes in a TCR-dependent manner. We postulate that additional MR1-restricted T-cell subsets may play a unique role in defence against infection by broadening the recognition of microbial metabolites.

Figure 1. M. smegmatis infection elicits a response from MAIT cells and TRAV1-2⁻ HLA-Ib-restricted CD8⁺ T cells.

PBMC-derived CD8⁺ T cells from adult donor (D462) were stimulated ex vivo by A549 cells infected with M. smegmatis to identify mycobacteria-reactive HLA-Ib T cells. Functional MAIT cells were identified by production of IFN-γ and expression of TRAV1-2 TCR. Numbers in quadrants indicate the % of live CD3⁺ cells. Experiments were performed three independent times with similar results. Representative results are shown.

Figure 2. MR1-restricted microbial-reactive CD8⁺ T cells from blood do not exclusively express TRAV1-2.

(a) IFN-γ production by T-cell clones D160-1-23 (restricted by HLA-E) and D426-B1 (restricted by MR1) in response to mycobacteria-infected WT and MR1^−/− A549 cell line. (b) Positively selected CD8⁺ T cells from PBMC were tested for ex vivo IFN-γ responses to M. smegmatis-infected WT or MR1^−/− A549 cell line. Events are gated on live CD3⁺CD4⁻ cells. IFN-γ and TRAV1-2 expressions are shown on the x and y axes, respectively. To the right is a summary of the TRAV1-2-negative response from each donor across experiments. (c) Frequency of IFN-γ⁺ CD4⁻ cells from each donor (represented by one dot) when stimulated by M. smegmatis-infected WT or MR1^−\− A549s, n=5 biological replicates, with n=3 technical replicates. Statistical significance of difference between groups was determined using the nonparametric Mann–Whitney U-test. Error bars are the s.e.m. of triplicates in a,b. Experiments in this figure were performed at least twice with similar results. Representative results are shown. *P value>0.05 was considered significant.

Figure 3. Isolation of an MR1-restricted T-cell clone that reacts to bacteria independently of the TRAV1-2 TCR.

Generation of a CD8⁺ T-cell line and then T-cell clone, D462-E4, through FACS sorting on CD8⁺, γδ TCR⁻, TRAV1-2-negative T cells from donor in (Fig. 1) and co-culturing for 7 days with M. smegmatis-infected autologous DCs. The line was stained with CFSE and co-cultured for 7 days with M. smegmatis-infected autologous macrophages. (a) Staining of TRAV1-2 and co-receptor CD8 on T-cell clone D462-E4. (b) Comparison of T-cell surface marker expression on TRAV1-2-negative clone D462-E4 (black) and TRAV1-2⁺ MAIT clone D426-G11 (grey) and (c) of TRBV 29-1 on D462-E4 and D481-A9 TRAV1-2⁺ MAIT clone. TCR αβ CDR3 sequences and gene names of D462-E4 are shown, right. (d) BEAS2B cell line infected with mycobacteria and tested for their ability to stimulate T-cell clones in the presence of blocking antibodies or isotype by IFN-γ ELISPOT assays, D462-E4, D481-F12 (TRAV1-2⁺ MAIT) and D466-A10 (HLA-B45). (e) Flow cytometry of clones with MR1 tetramer loaded with 6-formyl pterin (grey) or 5-OP-RU (clear). D462-E4, D481-A9 MAIT clone and D466-A10 HLA-B45-restricted T-cell clone. Error bars represent the s.e.m. of triplicates. Experiments were performed at least two times with similar results. Representative results are shown.

Figure 4. T-cell clone D462-E4 displays ligand and microbial infection selectivity.

(a) DCs were infected with M. smegmatis at a range of MOI (x axis) and then co-incubated with indicated MR1-restricted T-cell clones. (b) A549 cells were pulsed for 2 h with 100 μM RL-6,7-diMe (top) or RL-6-Me-OH (bottom) and tested for their ability to stimulate MR1-restricted T-cell clones in the presence of α-MR1 or isotype antibody by IFN-γ ELISPOT. (c) A549 cells were pulsed for 2 h with a range of concentrations of RL-6,7-diMe (top) or RL-6-Me-OH (bottom) and tested for their ability to stimulate MR1-restricted T-cell clones by IFN-γ ELISPOT. (d) DCs were infected at optimized MOIs with pathogens listed on the y axis, except for Yersinia and Shigella where the A549 cell line was used, and then co-incubated overnight with the indicated MR1-restricted T-cell clones. IFN-γ production was quantified by ELISPOT. Detection of less than 25 spot-forming unit per well was considered no response by the T-cell clone. Results are grouped by comparative response by MR1-restricted T-cell clones. Pattern of recognition was maintained over a range of MOI. Error bars are the s.e.m. of triplicates. Assays were performed twice, with similar results. Representative results are shown.

Figure 5. Selective recognition of S. pyogenes by a TRAV1-2-negative T-cell clone D462-E4.

(a) S. pyogenes or E. coli was cultured in minimal broth without riboflavin or with a titration of riboflavin over the course of 96 h (S. pyogenes) or 12 h (E. coli), and growth was monitored by optical density readings. (b) DCs were infected with S. pyogenes at a range of MOI (x axis) and then co-incubated with indicated MR1-restricted T-cell clones. (c) T-cell clones D462-E4 and D426-G11 were incubated with (left to right) S. pyogenes (MOI=10), S. pyogenes culture supernatant (SN) 30 μl, unloaded DCs, DCs with S. pyogenes SN or pulsed DC-conditioned media overnight. On the right, D462-E4 and D426-G11 were incubated with paraformaldehyde-fixed DCs that had been pulsed with S. pyogenes SN. IFN-γ production was quantified by ELISPOT. (d) T-cell clone D462-E4 was blocked with anti-pan TCR αβ, isotype control or not treated and then co-incubated with DCs loaded with S. pyogenes supernatant (30 μl). IFN-γ production was quantified by ELISPOT. (e) Phosphorylation of the CD3-ζ chain of the TCR/CD3 complex at tyrosine 142 or ZAP-70 at tyrosine 319 was quantified by flow cytometry after T-cell clone D462-E4 was co-incubated with DCs infected with S. pyogenes or M.smeg MOI=10, PHA (20 μg ml⁻¹) or left untreated (unstimulated condition). Numbers on the overlay indicate the geometric mean fluorescence intensity of at least 30,000 clones. (f) DCs were infected with S. pyogenes at MOI=3, blocked with anti-MR1 or isotype control (10 μg ml⁻¹) and then co-incubated with D462-E4 T-cell clone. IFN-γ production was quantified by ELISPOT. (g) DCs were either blocked with 6-formyl pterin (50 μg ml⁻¹) or 0.01 M NaOH vehicle control or nothing, and then loaded with S. pyogenes or M.smeg supernatant (15 μl) or PHA at 10 μg ml⁻¹. The DCs were then used to stimulate T-cell clone D462-E4 and IFN-γ production was quantified by ELISPOT. Error bars represent the s.e.m. of at least duplicates. Assays were performed three times, with similar results. Representative results are shown.

Figure 6. MR1-Ag tetramer⁺ CD8⁺ T cells from peripheral blood do not exclusively express TRAV1-2.

PBMCs were stained with antibodies to CD3, CD4, CD8, TRAV1-2, CD26, CD161, viability stain and human MR1-Ag (5-OP-RU) tetramer. All plots are gated on live, CD3⁺, CD4-negative cells. The x axis represents CD161 expression, and the y axis represents MR1 tetramer (top row), TRAV1-2 (middle row) and CD26 (bottom row). MR1 tetramer⁺ cells were gated as indicated by the gate in the top row. MR1 tetramer⁺ cells were subgated based on TRAV1-2 expression where TRAV1-2-negative events and their frequencies are shown in red. The second and third rows of plots are overlays of TRAV1-2⁺ in black and TRAV1-2-negative in red. (b) PBMCs were depleted of TRAV1-2⁺ cells by FACS and then stained with MR1-Ag tetramer and T-cell lineage markers (red bars). The black bars and red-patterned bars indicate frequencies presented in (a) for comparison. This experiment was performed once and (a) was performed twice with similar results. Representative results are shown.