key: cord-0315435-1alysf24
authors: Chowdhury, Roshni Roy; Valainis, John R.; Kask, Oliver; Ohanyan, Mane; Sun, Meng; Huang, Huang; Dubey, Megha; von Boehmer, Lotta; Sola, Elsa; Huang, Xianxi; Nguyen, Patricia K.; Scriba, Thomas J.; Davis, Mark M.; Bendall, Sean C.; Chien, Yueh-hsiu
title: The role of antigen recognition in the γδ T cell response at the controlled stage of M. tuberculosis infection
date: 2021-09-15
journal: bioRxiv
DOI: 10.1101/2021.09.14.460324
sha: 8c7c283755b70825422e2eb67de3c2d3280c5fff
doc_id: 315435
cord_uid: 1alysf24

γδ T cells contribute to host immune defense uniquely; but how they function in different stages (e.g., acute versus chronic) of a specific infection remains unclear. As the role of γδ T cells in early, active Mycobacterium tuberculosis (Mtb) infection is well documented, we focused on elucidating the γδ T cell response in persistent or controlled Mtb infection. Systems analysis of circulating γδ T cells from a South African adolescent cohort identified a distinct population of CD8+ γδ T cells that expanded in this state. These cells had features indicative of persistent antigenic exposure but were robust cytolytic effectors and cytokine/chemokine producers. While these γδ T cells displayed an attenuated response to TCR-mediated stimulation, they expressed Natural Killer (NK) cell receptors and had robust CD16 (FcγRIIIA)-mediated cytotoxic response, suggesting alternative ways for γδ T cells to control this stage of the infection. Despite this NK- like functionality, the CD8+ γδ T cells consisted of highly expanded clones, which utilized TCRs with different Vγ/δ pairs. Theses TCRs could respond to an Mtb-lysate, but not to phosphoantigens, which are components of Mtb-lysate that activate γδ T cells in acute Mtb infection, indicating that the CD8+ γδ T cells were induced in a stage-specific, antigen-driven manner. Indeed, trajectory analysis showed that these γδ T cells arose from naive cells that had traversed distinct differentiation paths in this infection stage. Importantly, increased levels of CD8+ γδ T cells were also found in other chronic inflammatory conditions, including cardiovascular disease and cancer, suggesting that persistent antigenic exposure may lead to similar γδ T cell responses.

A better understanding of how lymphocytes control both the acute and chronic phase of the same infection is key to the development of new intervention strategies and improved vaccines. The involvement of gd T cells in several acute and chronic infections has been noted for some time. In particular, in individuals with active tuberculosis (TB) disease, and other acute bacterial and parasitic infections, gd T cell frequencies can increase in the peripheral blood from <5% in healthy subjects to >45% in certain patients 1 . These gd T cells primarily express Vg9Vd2 TCRs, which respond to phosphoantigens -metabolites generated in the isoprenoid pathway in bacteria and parasites. Activated gd T cells can eliminate infected cells or pathogens in vitro 2-5 , suggesting that direct killing by these cells is part of how the infection is controlled. More recently, increases in the frequency of circulating IL-17+ Vg9Vd2 T cells were observed in children with acute bacterial meningitis 6 and in patients with active pulmonary tuberculosis 7 , consistent with the findings in mouse models of infection, where gd T cells are the major initial IL-17 producers in acute infection that initiate the inflammatory response 8 .

gd T cells have also been implicated in controlling chronic viral infections 9-11 . Selective gd T cell populations expand in the peripheral blood of patients infected with HIV, CMV, EBV, or HSV 12-14 .

In immunocompromised transplant patients with CMV reactivation, the rise of circulating gd T cells correlates with the resolution of infection, and increased numbers of gd T cells are also associated with improved long-term survival in these patients with high risk of CMV reactivation and tumor development 15 . In these cases, most of the activated gd T cells express TCRs encoded by Vd1 and Vd3, paired with different Vg chains, some of which show reactivity to virus-infected cells and tumor cell lines 15 .

While these studies highlight the ability of gd T cells to respond to pathogenic challenges in a context-dependent manner, it is unclear whether the difference in these responses reflect different pathogenic challenges (bacteria or parasites vs. viruses) and/or the chronicity of the infection.

This question is particularly relevant to Mtb infection, where most infections manifest as a clinically asymptomatic state, presumably held in check, or cleared by the host immune system. This state, previously known as latent TB, and referred to as controlled Mtb infection here, may be best defined as a state of persistent immune response to Mtb antigens detected by an interferon g (IFNg)-release assay, but without signs or symptoms of active disease. Previously, we have characterized the immune state associated with this controlled infection stage in a cohort of South African adolescents aged 13-18 years 16 . This cohort is from a highly endemic area but has a lower rate of active TB disease than is seen in young children and adults 17 , indicating a wellcontrolled Mtb infection. Our results indicated that gd T cells contribute to the concerted effort toward Mtb infection control. However, the gd T cell response in these subjects remains largely uncharacterized.

Here, we employed mass cytometric analysis, transcriptomic profiling, and functional assays, together with a newly developed Cytoskel algorithm to construct pseudotime trajectories, which reflect cellular progression through differentiation stages to study gd T cells from peripheral blood mononuclear cells (PBMCs) from the same South African adolescent cohort. Our results identify unique features of gd T cells in this stage of the infection and underscore the importance of antigen recognition in this phase of the gd T cell response.

To identify the gd T cell subpopulation(s) that associate with controlled Mtb infection, we analyzed our previously generated high-dimensional cytometry by time-of-flight (CyTOF) datasets, which examined cell surface markers, intracellular cytokines/cytolytic effectors, and signaling capacity (Fig. 1a) .

viSNE dimensionality reduction of the CyTOF analysis showed that the CD8+ gd T cells clustered together and displayed features of terminally differentiated effectors, which regain CD45RA expression (CD45RA+CD45RO-CCR7-CD27-) (Extended Data Figs. 2a, b) . These cells had a high expression of cytotoxic molecules (GZMB+PRF1+) and significantly higher proportions of CD16+ cells relative to the CD8-gd T cell subpopulation (Fig. 1b, Extended Data Figs. 2a, b) , suggesting enhanced cytolytic ability, which could be mediated through CD16. The CD8+ subset also comprised of significantly higher frequencies of GZMB+, PRF1+, and IFNg+ cells, as well as polyfunctional cells (Fig. 1b) . Interestingly, these CD8+ gd T cells were distinguished by their significantly increased expression of two classical myeloid cell markers, CD11c and CD70 (Fig.   1c, Extended Data Fig. 2a) , whose expression on ab T cells has been linked to antigen-driven responses and chronic immune activation 19, 20 . Consistently, CD8+ gd T cells showed small but significantly lower cell surface CD3 expression as compared to CD8-gd T cells (Extended Data Fig. 2c ).

The CD8+ gd T cells also differed in terms of their activation state and signaling capacity, displaying enhanced AKT phosphorylation at baseline (unstimulated condition) and post stimulation with phorbol myristate acetate (PMA) and ionomycin in comparison to CD8-gd T cells ( Fig. 1d, Extended Data Fig. 2d) . In cytotoxic CD8+ ab T cells, AKT activation has been shown to play an important role in driving TCR-and IL-2-induced transcriptional programs that control the expression of cytokines/chemokines and cytolytic effectors, regulate trafficking responses, and cell fate determination 21 . Relative to CD8-gd T cells, the CD8+ gd T cell subset displayed significantly lower phosphorylation levels of the mitogen-activated protein kinases (MAPKs), including ERK-1/2, p38, and MAPKAPK2, a substrate for p38, at baseline and upon stimulation with PMA and ionomycin (Fig. 1d, Extended Data Fig. 2d) . Defects in the MAPK pathway and induction of ERK phosphorylation have been associated with reduced proliferation 22 . These observations are consistent with these cells being terminally differentiated effectors.

To broaden our analysis, we performed bulk RNA sequencing (bulk-RNA-seq) on FACS sorted CD8+ and CD8-gd T cells from peripheral blood samples of five donors with controlled Mtb infection and identified 1,273 significantly (Adjusted P<0.05) differentially expressed genes (Fig.   2a, Supplementary Table 1 ). Compared to CD8-gd T cells, the CD8+ gd T cells showed significantly higher levels of some genes associated with prior antigenic exposure and chronic TCR activation in ab T cells, such as TIGIT, LAG3, CD244 [2B4], B3GAT1 [CD57], PRDM1

[Blimp-1], NR4A2, and TOX (Fig. 2b) , and displayed cytolytic features characterized by high expression of GNLY (Granulysin), PRF-1 (Perforin-1), granzymes (GZMA, GZMH), and NKG7 (a regulator of exocytosis of cytotoxic granules) (Fig. 2b, Supplementary Table 1 ). The CD8+ gd T cells also expressed higher levels of both FCGR3A (CD16a) (did not reach statistical significance) and FCGR3B (CD16b) (Fig. 2b, Supplementary Table 2) . Furthermore, they showed a preferential upregulation of NK-cell-associated activating and inhibitory receptor genes, including

CD244 (2B4), and the CD158 family of KIR receptors: KIR2DS4, KIR2DL1, KIR2DL3, KIR3DL1, KIR3DL2, and KIR3DL3 (Fig. 2b, Supplementary Table 2 ). Although not statistically significant, other NK-related genes like KLRF1 (NKp80), and KLRK1 (NKG2D) also showed higher expression levels in CD8+ compared to CD8-gd T cells (Supplementary Table 2) . Taken together, these findings were consistent with the CyTOF analysis, which indicated that CD8+ gd T cells are highly cytolytic effectors and that their response may be modulated by NK cell receptors. These cells may also deploy NK cell-like functions through CD16.

Although the abundance of cytokine mRNAs with a short half-life cannot be detected accurately in gene expression assays without re-stimulation (to re-induce their expressions after cell isolation) as we did here, we found very high levels of IL-32 expression, which has been implicated in various chronic inflammatory diseases 23 . Furthermore, both the CD8+ and the CD8-gd T cell subsets expressed high levels of the chemokine CCL5 (RANTES) (Supplementary Table 2) .

Nonetheless, the CD8+ gd T cells expressed significantly higher levels of XCL1, XCL2, CSF3, CCL4L1(MIP1b), CCL4L2, CCL3L1 (MIP1AP), and CCL19 (Fig. 2b) , which target a wide range of immune cells, including lymphocytes, dendritic cells, and other myeloid cells, suggesting that mobilizing leukocytes to the site of the infection may be one of the ways that CD8+ gd T cells regulate the inflammatory response. In fact, XCL1-producing activated CD8+ ab T cells were reported to orchestrate the trafficking of IFNg+ CD4 T cells in a mouse model of chronic tuberculosis infection 24 .

Compared to the CD8+ gd T cells, the CD8-subset showed higher expression of RORC, SOX4, ID3, and LEF1 -transcription factors associated with IL-17-committed gd T cells, but relatively lower production of lytic effector molecules (Fig. 2b) . While all gd T cells expressed high levels of ITGB2 (LFA-1), a suite of integrin genes -ITGAD (CD11D), ITGA10, ITGAL (CD11A), ITGA4 (CD49D), ITGB1 (VLAB) and the chemokine receptor CX3CR1 were preferentially upregulated in CD8+ gd T cells (Fig. 2b, Supplementary Table 2) , suggesting specific tissue homing properties to extra-intestinal sites, including the central nervous system, lungs and salivary glands 25 . The CD8-gd T cells, in contrast, expressed higher levels of the gut-homing receptor ITGAE (aE).

KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment pathway analysis showed that relative to CD8-gd T cells, CD8+ gd T cells had a significant enrichment for genes associated with FcgR-mediated functions, and those encoding components of the TCR, PI3K-AKT, and Rap1 signaling machinery (Fig. 2c) . Signaling via Rap1, whose activation by TCR ligation and chemokines is known to regulate integrin-(specifically, ITGAL and ITGA4)-mediated cell adhesion 26 , may reflect the mechanistic underpinnings of the migratory behavior of peripheral CD8+ gd T cells. In addition, CD8+ gd T cells had a significant enrichment for genes involved in immune response elements associated with chronic viral infections, such as CMV, EBV, and HIV ( Fig. 2c) , highlighting the similarity between gd T cell responses in controlled Mtb and chronic viral infections.

Taken together, these analyses indicated that CD8+ gd T cells in individuals with controlled Mtb infection were highly cytolytic, multi-functional effectors with altered response to TCR-mediated activation, but with the ability to mount CD16-mediated responses. To test these suppositions, we first assayed the intake of MitoTracker-Green by CD8-and CD8+ gd T cells to compare their mitochondrial mass. Different states of activated ab T cells are known to have different patterns of metabolism, but all effector ab T cells have higher mitochondrial mass when compared with naïve or exhausted T cells 27 . We found that compared to CD8-gd T cells, the CD8+ gd T cells showed significantly higher mitochondrial mass (Fig. 3a) . To evaluate these cells' response to TCR-mediated activation, we stimulated PBMCs in vitro with either anti-CD3 antibody or an Mtblysate and used the upregulation of CD69 as a read-out for T cell activation. We also demonstrated that the response to Mtb-lysate was inhibited by Cyclosporine A, indicating that the activation was mediated through the TCR. We found that both the percentage of cells, which upregulated CD69, and the CD69 expression level, were lower on the CD8+ compared to the CD8-gd T cells (Fig. 3b) . This indicated that these CD8+ gd T cells were hypo-responsive to antigenic challenge. Nonetheless, by measuring antibody-dependent CD107a degranulation from isolated gd T cells and from total PBMCs, we found that the antibody-dependent cellular cytotoxicity (ADCC) potential of the CD8+ gd T cells was significantly higher than the CD8-gd T cells and was comparable to that of NK cells (Figs. 3c, d, Extended Data Fig. 2e ).

Given the observation that the CD8+ gd T cells were hypo-responsive to antigenic challenge, we sought to determine if the increased frequency of peripheral CD8+ gd T cells in controlled Mtb infection reflected antigen-driven responses. To this end, we performed direct ex vivo single-cell TCR sequencing (scTCR-seq) 28 on CD8+ and CD8-gd T cells isolated from peripheral blood samples of four donors with controlled Mtb infection. We found that in all donors, the CD8+ gd TCRs were composed of a few dominant expanded clonotypes (multiplets). In contrast, most of the CD8-gd TCR clonotypes appeared only once (singletons) (Fig. 4a) . The percent clonality (calculated as the proportion that the multiplets occupy in the total repertoire) across all donors showed a significant difference between the CD8+ and CD8-gd T cells (P=0.0045) (Fig. 4b) .

While no identical TCRs in the CD8+ gd TCR repertoire were found between different donors, greater sequencing coverage will be required to determine whether there are shared TCR sequences among individuals. Regardless, these observations suggest that the peripheral CD8+ To determine the antigen specificity of the gd T cell clones identified in controlled Mtb-infection, four clonally expanded TCRs with different gd chain-pairings were selected for expression in Jurkat a-bcells (Fig. 4d) . Three of these clones responded to Mtb-lysate, but none of them to (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP) (Fig. 4d) , a metabolite from the isoprenoid pathway in Mtb, which stimulates Vg9Vd2 T cells 31 . This observation indicated that gd

It is plausible that these gd T cells are induced at the site of infection and/or the draining lymph nodes. Given the difficulty in acquiring human lung and draining lymph node samples, to test this supposition, we utilized human tonsil organoids 32 to evaluate Mtb-specific, lymphoid organ gd T cell responses. We found that stimulation of tonsil organoid cultures with the Mtb-lysate resulted in increased frequencies and numbers of CD8+ gd T cells in all 3 independent tonsil organoid cultures established from 3 healthy donors (Fig. 4e) . Directly ex vivo single cell TCR sequencing of CD8+ gd T cells isolated from a stimulated organoid culture demonstrated clonal expansion ( Fig. 4f) . host antigens. We therefore tested whether the TCRs expressed on gd T cells induced in controlled Mtb infection have similar reactivities. We found that Clones GD8-1 and GD8-5 showed robust CD69 expression upon coculture with T2, but not K562 or Daudi cells. Clone GD8-3 was reactive to both T2 and Daudi, but not K562 cells, while clone GD8-2 recognized Daudi cells, but not T2 or K562 cells (Fig. 4d) . These results indicate that while CD8+ gd T cells as a group respond to antigens expressed by tumor cell lines, the TCRs show different fine specificities, which is expected from adaptive antigen recognition. Fig. 3 a, b) . We found that ~10% of the cells had a naïve phenotype (defined as CD45RA+CD62L+CCR7+) (indicated as X on the map). These cells navigated a trajectory with a branch point (Branch Point 1) that led to two alternative developmental fates -trajectory 1 and trajectory 2, with similar numbers of cells in each trajectory. Trajectory 1 further bifurcated (Branch point 2) into two separate paths which ended at A and B, while trajectory 2 showed an early branching event (Branch Point 3) leading to two divergent cellular paths, eventually resulting in three branch termini (identified as C, D, and E). Additionally, some of the naïve gd T cells appeared to travel a short distinct path (identified as F) that ran parallel to paths A and B and was considered a part of trajectory 1 (Fig. 5a) . Notably, nearly all of the cells along trajectory 1 (comprised of Cell Clusters 2 and 3) appeared to be programmed to differentiate into cytolytic effectors with the expression of NK cell receptors, including CD16. While cells along paths A and B were CD8+, those along path F were CD8- (Fig. 5b, Extended Data Fig. 4a ). In contrast, ~85% of the cells in trajectory 2 appeared to be in a transitory state, trailing around Branch Point 3 (Cell Cluster 8) ( Fig. 5a) . Only ~10% of the cells in trajectory 2 differentiated into cytolytic NK-like effectors, that also expressed CD16, but lacked CD8 expression (path C) (Fig. 5b, Extended Data Fig. 4a) .

The remaining ~5% of cells followed the path from Branch Point 3 to terminus D and E via Branch Point 4 (Cell Clusters 7, 9, and 10). All three of these cell clusters expressed high levels of various cytokine receptors, including IL3RA, IL4R, IL23R, and IFNGR1 (Extended Data Fig. 4b) . The end point E was very similar to D and appeared to be a side excursion of cells retracing their trails forward and backwards between these two paths. gd T cells in Cell Cluster 7 expressed cytolytic effectors NKG7, GNLY, and the chemokine CCL5. gd T cells in Cluster 9 developed into effectors that were distinguished by their expression of the cytokine IL-1b and chemokines, including CXCL2, CXCL3, CXCL5, and CXCL8 (IL-8), that target neutrophils and other myeloid cells (Fig.   5c ). Clusters 7 and 9 also contained cells that expressed CD8 and CD16 (Fig. 5b) .

These divergent paths were marked by the expression of different transcription factors/related genes (Fig. 5d) . The naïve point X, as expected, showed high expression of LEF1 and TCF7. In trajectory 1, the highly cytolytic endpoint B (Cell Cluster 2) was characterized by high levels of PRDM1 (BLIMP-1), TBX21 (T-bet), IRF4, and RUNX3 expression (Fig. 5d) . A recent study has shown that RUNX3 synergizes with TCR signaling to establish IRF4-dependent transcriptional circuits that promote memory cytotoxic T lymphocyte formation 36 . These cells also expressed high levels of IKZF2 (Helios), BCL11B, FOXO1, and BACH2, transcription factors that function in promoting and stabilizing lineage commitment. In Trajectory 2, the branches leading to the end points D and E (Cell Clusters 7, 9, and 10), displayed enhanced AP-1 transcription factor associated FOSB, JUN, and JUNB expression, which are essential for the functional development of T cells, including IL-17-producing helper ab T cells 37 . Consistent with the observation that some cells in Cluster 7 expressed cytolytic effector genes, they expressed many of the factors identified for cells in Cell Cluster 2, including PRDM1 (BLIMP-1), TBX21 (T-bet), EOMES, RUNX3, as well as ZNF683 (HOBIT), which is noted for promoting BLIMP-1 expression.

Interestingly, gd T cells in the transition state, Cluster 8, expressed (1) high levels of ZBTB16 (PLZF), which is commonly associated with gd T cells that encounter ligands in the thymus, (2)

FOXP3, a lineage specification factor for T regulatory cells. The physiological relevance of this gene expression pattern will require further study. In addition, as our analysis was carried out with targeted scRNA-seq, it is conceivable that additional factors not listed in Fig. 5d , could be important for the cellular processes performed by these cells. Regardless, these data identify transcriptional regulators that may play instructive roles in gd T cell effector fate decisions and maintenance.

Among the cells whose TCRs were successfully sequenced (N=12,706), 39%, 27%, and 26% comprised of Vd1+, Vg9-Vd2+, and Vg9+Vd2+ TCRs, respectively (Extended data Fig. 5a) . The Vd1+ T cells made up 60% of cells in the naïve state (between X and Branch Point 1, Cell Cluster 4), and predominately (86% of the remaining Vd1+ cells) appeared in Trajectory 1. ~90% of these Vd1+ cells traversed distinct differentiation paths (from Branch Point 2 to A and B, and X to F, Cell Cluster 3) leading to cytolytic effector cell fates. In sharp contrast, ~73% of all Vg9+Vd2+ T cells (except those between X and Branch Point 1) belonged to trajectory 2, and ~90% appeared to be in transitional states, gathered around branch points (between Branch Points 1 to 2, 1 to 3, and 3 to 4, Cell Cluster 8), with less than 10% progressing to differentiate into effectors at the various branch termini (C, D, and E). In this context, we found fetal-derived gd T cells, defined as cells expressing TCRs without N-nucleotide additions in the CDR3 regions [annotated by IMGT/V-QUEST] in the PBMCs of all six donors (Supplementary Table 5 ). ~97% of these cells expressed Vg9+Vd2+ TCRs, frequently with the Vd2-Dd3-Jd3 rearrangement, and largely belonged to trajectory 2, located in Cell Cluster 8, a developmental stage distribution similar to other Vg9+Vd2+ cells (Extended data Fig. 5b-e) . It is important to note that gd T cells with different Vg/Vd TCRs showed preferential, but not exclusive distribution in any of the transition nodes and effector end points. This observation is consistent with the TCR usage described for cell clustering analysis from scRNA-seq data of gd T cells from two adult PBMCs and two cord blood samples 38 .

Unlike the Vg9+Vd2+ T cells, the Vg9-Vd2+ T cells were equally abundant in trajectories 1 and 2, and both at branch points and branch termini. This is in line with reports that Vg9-Vd2+ and Vg9+Vd2+ T cells have different antigenic specificities and participate in different pathological conditions 11 . It also indicated that these cells would contribute to the cytolytic effector cell pool in controlled Mtb infection. Indeed, Vg9-Vd2+ cells comprised ~30% of cells with cytolytic effector fate in trajectory 1 (from Branch Point 2 to A and B, and X to F), and approximately one third of cells with cytolytic effector fate in trajectory 2 (from Branch Point 3 to C).

In terms of clonal expansion of gd T cells (based on gd paired chain sequences) along the various branches, we found gradual increases in clonal expansion along trajectory 1 (Branch point 2 to A and B, and X to F) (Extended Data Fig. 6, Supplementary Table 6 ). In contrast, the cells in trajectory 2 showed limited clonal expansion, with no association between clonality and the pseudotemporal ordering of cells. In fact, BTG1 and LGALS1, two of the ten most abundantly expressed genes in Cell Clusters 7, 9, and 10 ( Fig. 5c ) are noted for their anti-proliferative function 39, 40 . These results were consistent with the TCR analysis of CD8+ and CD8-gd T cells from the 4 donors described above (Fig. 4a) .

Taken together, these results indicated that in the state of controlled Mtb infection, the responding gd T cell population (predominantly the Vd1+ T cells) traversed distinct differentiation paths leading to cytolytic effector cell fates. By contrast, most of the Vg9Vd2 T cells were in transitory states, indicating that they are poised to develop into the highly cytolytic effectors reportedly found in acute Mtb infection 1 . These observations suggest that the cytolytic functional fate of gd T cells in a given infection stage is acquired in the periphery as a consequence of antigen and environment driven events.

The trajectory analysis indicated that gd T cells traverse distinct effector differentiation paths to provide infection stage-specific response. To test whether increased circulating CD8+ gd T cell frequencies may be a common feature of persistent or chronic infectious and inflammatory conditions, we analyzed flow cytometry data of PBMCs from cohorts of chronic HIV (adults), chronic cardiovascular disease (older adults) and acute influenza (adults) infection. We also analyzed publicly available datasets (https://flowrepository.org) from cohorts of melanoma (chronic) 41 and COVID-19 (acute) patients. With striking consistency, circulating CD8+ gd T cell frequency was found to increase across all cohorts presenting chronic or persistent inflammatory conditions, even the frequency of total gd T cell remained unchanged (Fig. 6a) . This trend was absent in donors with acute infections (Fig. 6b) . Consistent with this finding, higher levels of circulating CD8+ gd T cells have been reported for a small group of HIV seropositive subjects 42 , and in immunocompromised patients after allogeneic stem cell transplantation with CMV reactivation 15 . Nonetheless, this is unlikely to be the molecular basis for the tumor cell line recognition by the CD8+ gd T cells identified in controlled Mtb infection since they do not respond to HMBPP. Given the similarity between gd TCRs and immunoglobulins in antigen recognition, and antigen specific repertoires 55 , it is possible that some of these gd TCRs are cross-reactive, and like cross-reactive antibodies, they may respond to a broad range of antigens. In this respect, increased circulating CD8+ gd T cells appears to be a common feature of persistent or chronic infectious and inflammatory conditions. Defining what triggers this subset of gd T cells, not only in chronic infection but also in atherosclerosis and cancer will be an important task going forward. Cohort Study was study was also supported by Aeras and BMGF GC6-74 (grant 37772) and

BMGF GC 12 (grant37885) for QuantiFERON testing.

South African adolescent cohort study (ACS) -As previously described 16 

The antibody panels, staining protocols, and analysis methods used here have been thoroughly described in a previous study 16 . Briefly, PBMCs from the South African adolescent cohort were stained with two panels -one measuring 25 surface markers and 12 cytokines/effector molecules Human genome (UCSC Genome Browser) using star RNASeq aligner version 2.5.4b. Resulting alignments were processed with the feature Counts software version 2.0.0 (subread) to obtain raw counts for each gene. The raw counts were then analyzed using DESeq2 version 3.10 (Bioconductor) to get differential expression data as well as normalized counts.

For the assessment of mitochondrial mass, we purchased the MitoTracker Green reagent from

Invitrogen. The samples were stained according to manufacturer's instructions. All cells were subsequently analyzed by flow cytometry.

PBMCs were thawed in complete RPMI 1640 medium at 2×10 6 cells per ml and recovered TCR transfectants in media only was used as negative control. HMB-PP and Mtb-lysate were used at 5µM and 10µg/ml concentrations, respectively. After 14-hour incubation, cells were collected and CD69 expression was measured using flow cytometry.

Tonsil organoids were established as previously described 32 . Briefly, whole tonsils (overall healthy, without obvious signs of inflammation) were collected in saline after surgery and then 

The data from Rhapsody single-cell analysis was transformed by replacing each count x by arcsinh(x/5). We refer to the D = 411 transformed coordinates (12 cell surface markers and 399

transcripts) for each cell as the feature coordinates and the D dimensional space as feature space.

The data cells form a cloud of points in the feature space. Branching trajectories were then constructed using the cytoskel package. Briefly cytoskel constructs a k-nearest neighbor graph 

We performed FlowSOM 61 meta-clustering on the data cells. FlowSOM first constructs a selforganizing map (SOM) which is a set of cell groups arranged on a (for example) 2D 10 by 10 grid such that the cells in each group are similar to each other and nearby groups are more similar than distant groups. The groups are then joined by a minimum spanning tree between groups.

Finally, higher level clustering is performed on the groups forming meta-clusters each of which contains one or more low level groups. In the following we refer to the meta-clusters simply as clusters. The data cells were clustered into 10 meta-clusters using the FlowSOM R package. The clustering used the same distance calculation as the trajectory algorithm. For each cluster, the cluster mean was calculated for each gene. For the subset of markers of interest, a dataframe was constructed with each row containing the average values for each marker of interest for a given cluster. This dataframe was passed to the clustermap function of the Python seaborn plotting package with scaling set so that the maximum value in each column was scaled to a value of 1.0 to make the gene expression differences and similarities between the clusters clearer. Clonality is defined as the total number of sequences that appear more than once relative to the total number sequenced per sample. P-value was determined using the paired t test. Error bars represent mean and 95% confidence intervals. 

XCL1 CSF3 GZMH GNLY NKG7 CCL4L2 CCL4L1 XCL2 IL32 CCL3L1 GZMA PRF1 CCL19 GZMK

CD8+ γδ T cells B3GAT1  LAG3  ZEB2  TBX21  CD8B  TOX  FCGR3B  RUNX3  EOMES  IKZF3  CD8A  TIGIT  NR4A2  PRDM1  DPP4  TCF7  FOS  ZBTB16  CD28 CD27 RORC ITGAD  ITGA10  ITGB1  ITGA4  ITGAL  IL2RB  ITGAE  CCR4  CCR8  IL18R1  IFNGR1  CCR2  CCR7  CXCR6  IL7R  IL23R  ITGA6  CCR9  IL12RB2 Clonal expansion based on paired γ/δ chains GD8-1 TRDV3  TRDJ1  CAFLRAYGGTGGFLLTDKLIF  TRGV2 TRGJ2  CATWDGRPIYYKKLF  Donor 1  14  20%   GD8-2 TRDV4  TRDJ1  CAMFRRWGITDKLIF  TRGV3 TRGJP1 CATWDRNTTGWFKIF Donor 1  14  20%   GD8-3 TRDV1  TRDJ1  CALGELPSFLYWGIRYTDKLIF  TRGV3 TRGJ2  CATWAGYYKKLF  Donor 2  2  2.5%   GD8-5 TRDV1  TRDJ1  CALGVPRGEPSPISKLIF  TRGV4 TRGJ2  CATWAFYYKKLF  Donor 2  54  68%   TB5  TRDV2  TRDJ1  CACDTVSRRDTSYTDKLIF  TRGV9 TRGJP  CALWEVRRELGKKIKVF   AT1G9  TRDV2  TRDJ1  CACDTWGITDKLIF  TRGV9 TRGJP  CALWEVQGELKKIKVF   TB5  AT1G9  GD8- EGR3  FOSL1  YBX3  IRF8  STAT1  TCF4  FOSB  XBP1  CNOT2  STAT3  MITF  ATF6B  STAT6  BCL6  BTG1  JUNB  EGR1  JUN  FOXP3  LEF1  BCL11B  IKZF2  EBF1  RORA  STAT4  RORC  ZBTB16  FOXO1  STAT5A  MYC  BACH2  TCF7  FOXP1  RUNX3  PAX5  ZBED2  PRDM1  IRF4  ZNF683  EOMES  TBX21  5 A   IL1B  CXCL5  CXCL3  CXCL2  CXCL8  BTG1  GAPDH  S100A10  LGALS1  HLA-DRA  FCER1G  THBS1  KLRB1  HLA-A  CXCR4  CD74  CST7  NKG7  CCL5  GNLY  PIK3IP1 KLRK1 IL32 

Regulatory Landscapes that Drive Memory Cytotoxic T Lymphocyte Formation

The AP-1 transcription factor JunB is required for Th17 cell differentiation

A fetal wave of human type 3 effector gammadelta cells with restricted TCR diversity persists into adulthood

BTG1, a member of a new family of antiproliferative genes

β-Galactoside-binding protein secreted by activated T cells inhibits antigen-induced proliferation of T cells

High-dimensional single-cell analysis predicts response to anti-PD-1 immunotherapy

A subset of gamma delta lymphocytes is increased during HIV-1 infection

Latent Tuberculosis: Two Centuries of Confusion

Cell Exhaustion During Chronic Viral Infection and Cancer

Stratifying subsets (FDR<0.01) Non-stratifying subsets