Antigenic stimulation in conjunction with cytokine is required for mediating IL-17A production in human MAIT cells

Abstract

Mucosal-associated invariant T (MAIT) cells are donor unrestricted T cells capable of both antigen-specific adaptive responses and cytokine driven innate-like functions. Although human MAIT cells uniformly express RORγt and IL23R, they generally produce IFN-γ, and only a small fraction produces IL-17. Recent studies show that combined TCR and cytokine stimulation can elicit functional heterogeneity in blood-derived MAIT cells. Here, we investigate the role of IL-23/IL-23R signaling in mediating the function and transcriptional profiles of lung MAIT cell clones. We demonstrate that BAL-derived lung MAIT cell clones exhibit distinct cytokine profiles and variable IL23R expression. Short-term IL-23 stimulation triggers clone-specific transcriptional programs and IL23R-dependent upregulation of type 17-associated genes. Prolonged conditioning of lung MAIT cell clones with TCR (5-OP-RU) and cytokine (IL-23) stimulation induces stable IL-17A production along with unique transcriptional changes. TCR + IL-23 conditioning alone upregulates clone-specific and shared cytoskeletal/structural gene programs, whereas subsequent PMA/Ionomycin stimulation further induces IL-12 family signaling and metabolic genes. Together, these findings demonstrate that IL23R expression and TCR signaling are required for IL-17A production, highlighting that these conditions may be met in tissue environments where MR1-specific antigens and proinflammatory cytokines coexist.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-61896-7.

Introduction

Mucosal-associated invariant T (MAIT) cells are a family of T cells, highly prevalent in humans that can mediate antimicrobial defense. They are defined by their use of a semi-invariant T cell receptor (TCR) α-chain (TRAV1-2-TRAJ33/TRAJ12/TRAJ20), and by their dependence on histocompatibility complex (MHC) class I-related protein 1 (MR1), which allows MAIT cells to be categorized as donor unrestricted T cells¹. MAIT cells can recognize small molecule metabolites derived from not only a broad array of microbial pathogens, but also from the host, drugs and drug-like molecules, as well as, synthetic and organic ligands^2–8. MAIT cells represent 1-10% of circulating T cells in the peripheral blood, and are further enriched in airways and mucosal tissues such as the lung, liver, and gut^9–11. As MAIT cells express a semi-invariant TCR α-chain, and rely on a nearly monomorphic presenting molecule, they were considered to be innate. However, through distinct TCR β-chain usage, MAIT cells can discriminate between microbial ligands and initiate pathogen-specific responses^12–14. In addition to TCR-dependent activation, MAIT cells can also be directly activated by cytokines such as IL-12 and IL-18¹⁵. Both TCR-dependent and TCR-independent activation can lead to the production of proinflammatory cytokines such as TNF-α and IFN-γ, as well as cytotoxic molecules such as Granzyme B and Perforin to promote antimicrobial defense^16,17. In certain mucosal tissues and during inflammatory disease states, MAIT cells preferentially produce IL-17A and IL-22 to maintain tissue homeostasis and regulate host-microbe interactions^18–20. These combined features position MAIT cells as a unique T cell subset capable of both antigen-specific adaptive and cytokine driven innate-like effector functions.

Murine MAIT cells have distinct functional IFN-γ (MAIT1) and IL-17 (MAIT17) subsets, which can be distinguished by expression of T-bet and RORγt, respectively^21,22. In humans, despite universal expression of RORγt, fewer than 5% of peripheral blood MAIT cells produce IL-17, while the majority produce IFN-γ^20,23–25. This paradox is further compounded by the observation that tissue-resident MAIT cells are phenotypically distinct from blood MAIT cells and that MAIT cells can differentiate in response to inflammatory stimuli^17,21,25. Garner et al. show that human MAIT cells in blood and liver are transcriptionally distinct, with liver MAIT cells enriched for activation and tissue-resident transcriptional profiles. Moreover, although human MAIT cells are transcriptionally homogeneous at rest, activation through TCR or cytokines (IL-12 and IL-18) induces stimulus-specific transcriptional and phenotypic profiles, such as expansion of IL-17-positive MAIT cells with elevated IL17F expression¹⁷. Recent studies further illustrate this functional plasticity in human blood MAIT cells exposed to distinct polarizing cues^26,27. Wang et al. demonstrate that IL-17A-producing MAIT cell clones derived from peripheral blood mononuclear cells (PBMCs) gain IFN-γ-producing capacity upon stimulation with polarizing cytokines such as IL-12 or IL-23, whereas IFN-γ-producing clones maintain their original functional phenotype even with the same stimuli²⁶. Together, these findings highlight the capacity of human MAIT cells to produce IL-17 through transcriptional and functional plasticity and support the importance of tissue residency and inflammatory milieu in driving these changes.

While human lung MAIT cells express RORC and IL23R transcripts, IL-17 production has not been demonstrated^9,28. For example, lung MAIT cells exhibit significantly higher expression of IL23R, RORC, and IL17A compared to non-MAIT CD8⁺ T cells²⁸. However, fewer than 1% of human lung MAIT cells produce IL-17⁹. Building on recent studies demonstrating that dual TCR and cytokine stimulation drives functional heterogeneity in blood MAIT cells, we sought to determine whether IL-23/IL-23R signaling modulates the function and transcriptional profiles of lung MAIT cell clones. We used IL-23 as the polarizing cytokine since IL23R is highly expressed in both blood and lung MAIT cells^27,28. Here, we demonstrate that conditioning human PBMC-derived MAIT cells with TCR (5-OP-RU) and cytokine (IL-23) induces IL-17A production. In contrast, bronchoalveolar lavage (BAL)-derived lung MAIT cell clones exhibit distinct cytokine profiles and variable IL23R expression across lung MAIT cell clones. Short-term IL-23 stimulation of MAIT cell clones triggers IL23R-dependent upregulation of type 17-associated genes. Conditioning these clones with TCR + IL-23 results in unique transcriptional profiles and stable IL-17A production following PMA/Ionomycin stimulation. Together, these findings demonstrate that IL23R is required for lung MAIT cell clones to induce a specific network of signaling and metabolic genes that reinforce IL-17 production upon TCR + IL-23 conditioning and stimulation.

Results

Conditioning of Human MAIT cells with TCR and cytokine enables IL-17A production

To determine if conditioning with 5-OP-RU (TCR) and IL-23 can induce IL-17 production, we first established an in vitro conditioning model that was modified from a 5-OP-RU stimulation protocol previously published²⁹. Specifically, we conditioned peripheral blood mononuclear cells (PBMCs) from a healthy donor with 5-OP-RU on Day 1 and 7 to induce TCR-dependent activation, with or without the addition of the polarizing cytokine IL-23 every other day (Fig. 1A). This healthy donor PBMCs expressed both IL12RB1 and IL23R at detectable levels by single-cell RNA-sequencing (scRNA-seq) (Supp. Figure 1A-C). To evaluate the functional capacity of these clones after conditioning, we stimulated them on Day 12 with PMA/Ionomycin and assessed cytokine production. MAIT cells, defined as CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells, conditioned with 5-OP-RU in the absence of IL-23 produced only IFN-γ (Supp. Figure 1D, Fig. 1B, C). In contrast, conditioning with 5-OP-RU and IL-23 expanded both IFN-γ/IL-17A-producing and IL-17A only-producing MAIT cells. Notably, conditioning with TCR + IL-23 induced IL-17A production in a minor proportion of MAIT cells that were previously IFN-γ-only producers, suggesting that MAIT cells can acquire new effector functions in response to polarizing cytokines. These findings demonstrate that human blood MAIT cells can adopt an IL-17A-producing phenotype following IL-23 conditioning.

Fig. 1.

Human MAIT cells can produce IL-17A upon IL-23 stimulation and lung MAIT cell clones exhibit variable IL23R expression. (A) Experimental design: in vitro conditioning of PBMCs with 5-OP-RU and IL-23. Cells were incubated with PMA/Ionomycin for 3 hours on Day 12 for flow cytometry analysis. (B) Representative flow cytometry plot showing production of IFN-γ and IL-17A by MAIT cells conditioned with 5-OP-RU ± IL-23. MAIT cells are defined as live CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells. (C) Percent cytokine production of IFN-γ⁺-only, IFN-γ⁺ IL-17A⁺, and IL-17A⁺-only by MAIT cells. Data plotted as mean±SEM and pooled from four independent experiments (n=1). Two-tailed paired Student’s t-test. ns = not significant, * = p ≤ 0.05. (D) Representative flow cytometry plot showing production of IFN-γ and IL-17A by BAL-derived lung MAIT cell clones (D1004 A3, B1023 A5, B1026 A8) after incubation with PMA/Ionomycin for 3 hours. MAIT cells are defined as live CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells. (E) Relative gene expression levels of IL23R compared to GAPDH in lung MAIT cell clones and normalized to lung non-MAIT CD8 T cell clone (D0033-JE1). Data plotted as mean±SEM and pooled from three independent experiments (n=3). Ordinary One-Way ANOVA Tukey’s multiple comparisons test. * = p ≤ 0.05, **** = p ≤ 0.0001.

Human lung MAIT cell clones can produce IL-17 and exhibit variable IL23R expression

To further explore the functional consequences of IL-23 stimulation, we evaluated MAIT cell clones that were isolated from BAL samples collected from participants in South Africa⁹. As shown in our previous work, MAIT cell clones derived from lung express IL23R, but do not produce IL-17A^9,28. Unexpectedly, we found discrete phenotypes with regards to IL-17A production. Clones D1004 A3 and B1026 A8 predominantly produced IFN-γ following PMA/Ionomycin stimulation. In contrast, the B1023 A5 clone produced both IFN-γ and IL-17A (Fig. 1D). Given that we were now observing stable IL-17 production in a clone, it provided a unique opportunity to explore the mechanisms underlying IL-17A production. Since IL-23 promoted these phenotypic changes in blood MAIT cells, we examined IL23R expression in these lung MAIT cell clones. We found that clones D1004 A3 and B1026 A8 had low and high IL23R expression, respectively (Fig. 1E). In contrast, B1023 A5, a clone with stable IL-17A production, expressed an intermediate level of IL23R (Fig. 1E). The varying levels of IL23R expression and incomplete correlation between IL-23R levels and IL-17 production among these clones led us to investigate the mechanism of IL-23/IL-23R signaling and to identify additional gene programs that regulate IL-17A production.

Increasing levels of IL-23R are associated with distinct expression profiles

To further assess the role of IL-23/IL-23R signaling, we conducted scRNA-seq on clones D1004 A3, B1023 A5, and B1026 A8, hereafter referred to as IL23R-low (A3), IL23R-med (A5), and IL23R-high (A8), that were unstimulated, stimulated with IL-23 overnight, or stimulated with PMA/Ionomycin for 3 hours. Stimulation with PMA/Ionomycin was chosen to ensure that each clone was viable and to confirm the IL-17 expression phenotype described above. The UMAP dimensionality reduction of each clone is shown in Fig. 2A. Stimulation with PMA/Ionomycin for all three clones resulted in a separate cluster reflecting a distinct transcriptional profile as compared to unstimulated or IL-23 stimulated cells. In contrast, stimulation with IL-23 revealed three distinct patterns. IL23R-low (A3) clone did not result in a unique cluster following IL-23 stimulation. In contrast, both IL23R-med (A5) and IL23R-high (A8) clones responded to IL-23 stimulation, with more distinct clusters in A8 than A5. These results suggest that increasing levels of IL23R expression result in transcriptional profiles that were more distinct from unstimulated cells, which nonetheless did not correlate with the ability to produce IL-17 following PMA/Ionomycin stimulation.

Fig. 2.

Increasing levels of IL23R are associated with distinct expression profiles. Single-cell RNA-sequencing was performed on IL23R-low (A3), IL23R-med (A5), and IL23R-high (A8) that were unstimulated, stimulated with IL-23 overnight, or stimulated with PMA/Ionomycin for 3 hours (n=3, one independent experiment). (A) UMAP reductions of IL23R-low (A3), IL23R-med (A5), and IL23R-high (A8) for unstimulated (blue), IL-23 stimulation (red), or PMA/Ionomycin stimulation (green). (B) Dot plot of the of top differentially expressed genes defining each clone and stimulation state. Boxes indicate cluster of genes upregulated with short-term IL-23 stimulation for each clone. (C) IL17A transcript levels across clones with unstimulated, IL-23 stimulation, or PMA/Ionomycin stimulation. Kruskal–Wallis and post-hoc pairwise Wilcoxon with BH correction. **** = p ≤ 0.0001. (D–F) Venn diagrams showing numbers of differentially expressed genes (DEGs) for IL-23 versus Unstimulated (D), PMA/Ionomycin versus Unstimulated (E), and PMA/Ionomycin versus IL-23 (F).

To further explore the transcriptional profile induced by short-term IL-23 stimulation, we explored the top 10 genes differentially expressed by each clone following stimulation with IL-23 or PMA/Ionomycin (Fig. 2B). In all three clones, stimulation with PMA/Ionomycin induced broad and robust transcriptional activation, marked by high expression of effector genes (IL2, CSF2, and IFNG), chemokines (XCL1, XCL2, CCL3L3, CCL4L2, and CCL4), and activation-associated genes (CRTAM, PDCD1, and CTLA4). Notably, PMA/Ionomycin stimulation induced IL17A expression in IL23R-med (A5) clone, corresponding to its ability to produce IL-17A upon activation, but there was no significant IL17A expression in the other clones (Fig. 2C). Stimulation with IL-23 elicited no differentially expressed genes in the IL23R-low (A3) clone. In contrast, IL-23 triggered a selective, clone-specific transcriptional programs in the IL23R-med (A5) and IL23R-high (A8) clones. Specifically, the IL23R-med clone upregulated interferon-stimulated genes (MX1, IFI44L, IFIT1, IFITM1, ISG15, OAS1, IFI6, IFI44, OAS3, and RSAD2), whereas the IL23R-high clone upregulated genes such as GPR25, FGFBP2, CLU, BCL3, GLUL, CTSW, GZMH, LINC02195, and CD8B2, with BCL3 and CD7 representing type 17-associated genes^17,30. These data elucidate the distinct patterns seen in the UMAP analysis.

To determine both shared and unique differential gene expressions (DEGs), we generated Venn-diagrams to compare clones in response to different stimulations (Fig. 2D-F, Supp. Table 1 and 2). IL-23 stimulation induced transcriptional changes only in the IL23R-med (A5) and IL23R-high (A8) clones. The IL23R-high clone displayed the largest response (361 unique DEGs) and the IL23R-med clone showed a modest response (41 DEGs), while the IL23R-low (A3) clone exhibited no IL-23–responsive genes (Fig. 2D, Supp. Table 1 and 2). In contrast, PMA/Ionomycin stimulation elicited broad transcriptional activation across all clones, with thousands of DEGs and a substantial shared core of 3,496 genes, alongside clone-specific signatures ranging from 151 DEGs unique to A3 to 1,179 DEGs unique to A8 (Fig. 2E, Supp. Table 1 and 2). When directly comparing PMA/Ionomycin to IL-23 stimulation, the three clones again shared a large common set of 3,513 DEGs, with additional clone-restricted gene programs, most pronounced in A8 (1,325 unique DEGs) (Fig. 2F, Supp. Table 1 and 2). These findings highlight the strong, largely overlapping transcriptional program induced by PMA/Ionomycin across MAIT cell clones, contrasted with the more selective and IL-23R-dependent response to IL-23.

Given the observed changes in type 17–associated genes, we next applied MAIT1 and MAIT17 transcriptional scores from Garner et al. to assess whether IL-23 or PMA/Ionomycin stimulation shifted the MAIT cell clones toward a MAIT1 or MAIT17 phenotype¹⁷. Across all three clones, PMA/Ionomycin stimulation exhibited a pronounced increase in MAIT1 scores, whereas short-term IL-23 stimulation had little effect on the MAIT1 score (Supp. Figure 2A). PMA/Ionomycin stimulation consistently reduced MAIT17 scores in all clones, regardless of IL23R expression (Supp. Figure 2B). IL-23 stimulation did not alter MAIT17 scores in the IL23R-low (A3) or in IL23R-med (A5) clones. However, in IL23R-high (A8) clone, IL-23 stimulation increased the MAIT17 score. Together, these findings indicate that PMA/Ionomycin strongly skews MAIT cells toward a MAIT1-like transcriptional profile while suppressing MAIT17 features, whereas IL-23 selectively enhances MAIT17 polarization in the IL23R-high clone.

Conditioning of MAIT cell clones with 5-OP-RU in the presence of IL-23 results in the stable IL-17A production

As shown in Fig. 2C, short-term exposure to IL-23 did not result in the upregulation of IL17A expression. Previous studies and Fig. 1A-C show that prolonged TCR activation in combination with cytokine stimulation is required to elicit upregulation of IL17-associated genes. For example, Garner et al. demonstrated that combined TCR and cytokine stimulation for at least 68 hours strongly induces distinct IL17F upregulation, while Cole et al. showed that 3 days of TCR activation followed by PMA/Ionomycin stimulation enhances both IL-17A and IL-17F production^17,25. Therefore, we modified our rapid expansion protocol to an experimental design using 5-OP-RU (TCR) and IL-23 (Cytokine) co-stimulation to more effectively assess the functional capacity and role of IL-23 on lung MAIT cell clones. Specifically, our conditioning protocol entailed co-culturing individual MAIT cell clones with irradiated lymphoblastoid cell lines (LCLs) and PBMCs, stimulating them with 5-OP-RU on Day 0 and Day 6, and adding IL-23 every other day where indicated (Fig. 3A). To determine the functional profile of conditioned MAIT cells (defined as CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells), cultured in the presence or absence of IL-23, we stimulated with PMA/Ionomycin and measured IL-17 and IFN-γ production (Fig. 3A, Supp. Figure 3). As expected, IL-23 did not alter the phenotype of the IL23R-low (A3) clone (Fig. 3B). In contrast, the IL23R-med (A5) clone showed a significant increase in IL-17A production when conditioned with IL-23 (Fig. 3B, C). The IL23R-high (A8) clone acquired the capacity to co-produce IL-17A in conjunction with IFN-γ following conditioning with IL-23 (Fig. 3B, C). These findings closely parallel the phenotypic changes we observed in human blood MAIT cells and underscore the functional adaptability of lung MAIT cell clones in response to polarizing cytokines.

Fig. 3.

Conditioning of MAIT cell clones with 5-OP-RU in the presence of IL-23 results in the stable IL-17A production. (A) Experimental design: in vitro conditioning of lung MAIT cell clones with 5-OP-RU and IL-23. Cells were stimulated with PMA/Ionomycin for 3 hours on Day 11 for flow cytometry analysis. (B) Representative flow cytometry plot showing production of IFN-γ and IL-17A by lung MAIT cell clones conditioned with 5-OP-RU ± IL-23. MAIT cells are defined as live CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells. (C) Percent cytokine production of IFN-γ⁺-only, IFN-γ⁺ IL-17A⁺, and IL-17A⁺-only by MAIT cell clones. Data plotted as mean±SEM and pooled from four independent experiments (n=3). Two-tailed paired Student’s t-test comparing 5-OP-RU-IL-23 versus 5-OP-RU+IL-23 for each clone with cytokine produced. ns = not significant, * = p ≤ 0.05, ** = p ≤ 0.01. (D) Experimental design: Rapid T cell expansion after in vitro conditioning of lung MAIT cell clones with 5-OP-RU and IL-23. On Day 11, cells were rested for 3 days in low-dose IL-2 before the rapid T cell expansion for another 12 days. On Day 26, cells were stimulated with PMA/Ionomycin for 3 hours for flow cytometry analysis. (E) Representative flow cytometry plot showing production of IFN-γ and IL-17A by lung MAIT cell clones that were previously conditioned with or without IL-23 and then rapidly expanded. MAIT cells are defined as live CD8⁺TRAV1-2⁺5-OP-RU-tet⁺ cells. (F) Percent cytokine production of IFN-γ⁺-only, IFN-γ⁺ IL-17A⁺, and IL-17A⁺-only by MAIT cell clones that were previously conditioned with or without IL-23 and then rapidly expanded. Data plotted as mean±SEM and pooled from two independent experiments (n=3).

To evaluate whether this phenotype is stable over time, we expanded these MAIT cell clones again in the absence of the polarizing cytokine IL-23. Specifically, after resting MAIT cell clones for three days in low-dose IL-2, we expanded the clones with TCR stimulation (αCD3) and co-stimulation (LCLs and PBMCs)³¹. At the end of this expansion, we measured cytokine production of lung MAIT cell clones following PMA/Ionomycin stimulation (Fig. 3D). The resulting cytokine profiles remained consistent with those observed immediately following the initial 5-OP-RU and IL-23 conditioning phase (Fig. 3B-C, E-F). The IL23R-low (A3) clone continued to produce only IFN-γ, regardless of prior exposure to IL-23. The IL23R-med (A5) clone, previously conditioned with IL-23, continued to show a trend toward increased IL-17A-only and expansion of IFN-γ/IL-17A-producing cells. Furthermore, the IL23R-high (A8) clone appeared to maintain its capacity to co-produce IFN-γ and IL-17A even in the absence of continued IL-23 exposure. Together, these results demonstrate that human lung MAIT cell clones undergo stable phenotypic changes in response to prolonged TCR and IL-23 co-stimulation.

Single-cell profiling of human lung MAIT cell clones conditioned with 5-OP-RU and IL-23

To explore the mechanisms by which IL-23 conditioning led to the stable production of IL-17A, we performed scRNA-seq on the MAIT cell clones that were conditioned with 5-OP-RU (TCR) ± IL-23 and then stimulated or not with PMA/Ionomycin (Fig. 4A). As shown in Fig. 4B, IFN-γ and IL-17A production were consistent with our earlier observations. Both the IL23R-med (A5) and IL23R-high (A8) clones increased IL-17A secretion without changes in IFN-γ when stimulated with PMA/Ionomycin after IL-23 conditioning (Fig. 4B). Single-cell profiling of these clones also confirmed the protein expression as demonstrated by IL17A mRNA expression (Fig. 4C). Both the IL23R-med (A5) and IL23R-high (A8) clones showed modest upregulation of IL17A expression with IL-23 conditioning, which was enhanced by PMA/Ionomycin stimulation.

Fig. 4.

Functional phenotypes of human lung MAIT cell clones with IL-23 conditioning and PMA/Ionomycin stimulation. (A) Experimental design: in vitro conditioning of lung MAIT cell clones with 5-OP-RU ± IL-23. Cells were either stimulated or not with PMA/Ionomycin for 3 hours on Day 11 for flow cytometry analysis and single-cell RNA-sequencing (scRNA-seq). Labeling of each clone depending on conditioning and stimulation. (B) Validation of cytokine production after conditioning and stimulation following the experimental design in Fig. 4A. Percent cytokine production of IFN-γ⁺-only and IL-17A⁺-only by MAIT cell clones after incubation with PMA/Ionomycin for 3 hours for flow cytometry analysis. Data plotted as mean from one independent experiment (n=3). (C) IL17A transcript levels across clones conditioned with 5-OP-RU ± IL-23 and then stimulated or not with PMA/Ionomycin. Kruskal–Wallis and post-hoc pairwise Wilcoxon with BH correction. **** = p ≤ 0.0001.

We next examined the dimensionality reduction for each conditioning regimen with subsequent stimulation (Fig. 5A). The combination of TCR + IL-23 conditioning (con+) in both stimulated or not with PMA/Ionomycin compared to TCR alone (con-) showed no distinct clustering in the IL23R-low (A3) clone, whereas the IL23R-med (A5) and IL23R-high (A8) clones displayed partial and discrete clustering, respectively (Fig. 5A). These data, therefore, demonstrate that TCR + IL-23 conditioning can alter the gene expression profile in the IL-23R expressing clones and that the degree of unique transcriptional phenotypes correlated with IL23R expression level. To further assess the genes differentially expressed after TCR and IL-23 conditioning from these UMAP clusters, we generated volcano plots and heatmaps comparing con-/stim- versus con+/stim- for each clone. In the IL23R-low (A3) clone, reactome pathway enrichment analysis of genes significantly downregulated following IL-23 conditioning revealed suppression of pathways associated with “interferon stimulated genes” and “adhesion and trafficking” (Supp. Figure 4A, D). In contrast, IL-23 conditioning significantly upregulated genes in both the IL23R-med (A5) and IL23R-high (A8) clones (Supp. Figure 4B, C). Although these two clones shared a subset of genes associated with “Cytoskeletal/Structure,” most upregulated genes were clone-specific, such as genes categorized as “Adhesion and Migration” for the IL23R-med (A5) clone and “Metabolic Activation” and “Lipid/Cholesterol Handling” for the IL23R-high (A8) clone (Fig. 5B, C).

Fig. 5.

Single-cell transcriptional profiling of human lung MAIT cell clones conditioned with 5-OP-RU and IL-23. Single-cell RNA-sequencing was performed on IL23R-low (A3), IL23R-med (A5), and IL23R-high (A8) clones that were conditioned with 5-OP-RU in the absence (con-) or presence (con+) of IL-23 for 10 days. On Day 11, cells were analyzed following no stimulation (stim-) or stimulation (stim+) with PMA/Ionomycin (n=3, one independent experiment). (A) UMAP projections of IL23R-low (A3), IL23R-med (A5), and IL23R-high (A8) clones under con-/stim-, con+/stim-, con-/stim+, and con+/stim+. (B) Heatmap of genes significantly upregulated in IL23R-med (A5) with IL-23 conditioning. Genes are grouped into shared categories and described. (C) Heatmap of genes significantly upregulated in IL23R-high (A8) with IL-23 conditioning. Genes are grouped into shared categories and described. (D) Dot plot of top differentially expressed genes per each clone with IL-23 conditioning and PMA/Ionomycin stimulation. Boxes indicate cluster of genes upregulated with IL-23 conditioning with PMA/Ionomycin stimulation for each clone. (E) Heatmap of specific genes significantly upregulated in IL23R-high (A8) with IL-23 conditioning and PMA/Ionomycin stimulation as part of IL-12 family signaling pathway enrichment analysis.

To further examine the changes in transcriptional profile with TCR + IL-23 conditioning and subsequent PMA/Ionomycin stimulation (con-/stim+ versus con+/stim+), we examined the top differentially expressed genes for each conditioning regimen (Fig. 5D). We observed minimal transcriptional changes in the IL23R-low D1004 A3 clone, whereas both IL23R-med (A5) and IL23R-high (A8) clones upregulated unique gene sets in response to IL-23 conditioning as well as subsequent PMA/Ionomycin stimulation. Across all three clones, PMA/Ionomycin stimulation broadly activated effector molecules such as CSF2 and CD40LG, as well as transcription factors including NFKBID, EGR2, NR4A1 and NR4A2. Moreover, compared to PMA/Ionomycin stimulation alone (con-/stim+), TCR + IL-23 conditioning (con+/stim+) in both IL23R-med and IL23R-high clones suppressed MAIT1-associated genes such as CXCR3, XCL1, and XCL2, as opposed to the upregulation of MAIT17-associated genes, including IL17A and CCL20¹⁷.

Given these transcriptional changes shown in Fig. 5D, we next performed pathway enrichment analysis to understand the distinct biological processes associated with TCR + IL-23 conditioning and subsequent PMA/Ionomycin stimulation (con-/stim+ versus con+/stim+) (Supp. Figure 5A-D). The IL23R-low (A3) clone showed no differentially expressed genes with conditioning and stimulation. In contrast, IL-23 conditioning alone for both the IL23R-med (A5) and IL23R-high (A8) clones was associated with cell cycle–related pathways, including “M phase,” “Mitotic anaphase,” and “RHO GTPase effectors” (Supp. Figure 5A, B), suggesting that the TCR + IL-23 conditioning regimen promotes cell proliferation and activation-related pathways. We next evaluated the resulting phenotype following PMA/Ionomycin stimulation. Following TCR + IL-23 conditioning and subsequent PMA/Ionomycin stimulation, the IL23R-med (A5) and IL23R-high (A8) clones exhibited different pathway enrichments. The IL23R-med clone was enriched for interleukin signaling, TCR signaling, and NF-κB–related pathways (Supp. Figure 5C), whereas the IL23R-high clone uniquely upregulated IL-12 family signaling, nucleotide metabolism-related genes, and reactive oxygen species detoxification (Supp. Figure 5D).

To further identify specific genes that drives IL-23 responsiveness, we generated a heat map of “IL-12 family signaling” from the pathway enrichment analysis and compared IL23R-med (A5) and IL23R-high (A8) clones (Supp. Figure 6). We focused on “IL-12 family signaling” pathway as IL-23 belongs to this cytokine family and shares components such as p40 subunit, IL12RB1 receptor, and downstream JAK-STAT activation. In both clones, IL-23 induced genes implicated in IL-23 responsiveness and IL-17 secretion such as IL12RB1 and MIF^32,33. Interestingly, the IL23R-high (A8) clone uniquely expressed IL6ST and upregulated RAP1B and CDC42 upon IL-23 conditioning, which are genes associated with STAT3 signaling and Th17 differentiation (Fig. 5E)^34–36. Moreover, the IL23R-high (A8) clone further upregulated metabolic/redox regulators such as TALDO1, MTAP, and GSTO1 compared to the IL23R-med (A5) clone. Together, these findings reveal a network of signaling and metabolic genes that allow the IL23R-high (A8) clone to co-produce IFN-γ and IL-17A. Specifically, for IL23R-high (A8) clone, IL23R expression appears to amplify STAT3-driven Th17 pathways and metabolic and redox programs, sustaining IL-17 production in response to IL-23 conditioning.

Discussion

In humans, MAIT cells are largely characterized by their production of IFN-γ. Paradoxically, MAIT cells in both blood and lung express RORγt and IL-23R, raising the question of why IL-17A-producing MAIT cells have been so difficult to identify. Recently, two studies address this paradox by demonstrating that dual TCR and cytokine stimulation can skew the transcriptional and cytokine profiles toward IL-17A production^17,26. In this study, we extend these observations by developing a conditioning protocol that combines TCR (5-OP-RU) with cytokine (IL-23) stimulation. We focused on IL-23 based on its receptor expression on both blood- and lung-derived MAIT cells. When we applied the conditioning regimen to blood-resident MAIT cells, this resulted in the production of IL-17A. Moreover, we identified lung-derived MAIT cell clones that showed heterogeneity in IL23R expression and IL-17A production. Following overnight IL-23 stimulation, we found that IL-23 responsiveness in lung MAIT cell clones was associated with the level of IL23R expression. We then applied the conditioning regimen, together with subsequent PMA/Ionomycin stimulation, and found stable IL-17A production and unique transcriptional changes in IL23R-expressing MAIT cell clones. These findings highlight that the environment can actively shape MAIT cell function and, to some extent, their functional plasticity.

To understand the impact of conditioning regimen on lung MAIT cell clones with variable IL23R expression, scRNA-seq revealed key transcriptional changes. For both IL23R-med A5 and IL23R-high A8 clones, IL-23 conditioning upregulated clone-specific genes, while also inducing a shared subset of genes associated with “Cytoskeletal/Structure.” This potentially suggests that IL-23 conditioning primes MAIT cells for activation and motility to facilitate their effector functions. Notably, the IL23R-med A5 – a clone with stable IL-17A production – upregulated “Adhesion/Migration” genes, such as AQP3 and S100A4, key components of the MAIT17 gene signatures¹⁷. In contrast, the IL23R-high A8 clone induced “Metabolic activation” genes, such as LDHA, a glycolytic enzyme associated with RORγt expression in MAIT17 cells³⁷. These findings support the idea that TCR + IL-23 conditioning biases human lung MAIT cells toward a MAIT17 transcriptional program and IL-17A effector function. Interestingly, we also observed induction of interferon-stimulated genes (ISGs) in the IL23R-med A5 clone following IL-23 stimulation. While stimulation with IL-23 is not classically linked to type I interferon signaling, this may reflect crosstalk between inflammatory cytokine pathways or heightened sensitivity to basal interferon signaling. Notably, type I interferon programs have been associated with enhanced cytotoxic and antimicrobial functions in T cells, suggesting that IL-23–mediated ISG induction may represent an additional mechanism by which inflammatory cues augment MAIT cell effector capacity^38,39.

Pathway enrichment analysis comparing IL23R-med A5 and IL23R-high A8 clones elucidated transcriptional changes that enabled the IL23R-high A8 clone to acquire the capacity to co-produce IL-17A and IFN-γ following conditioning and stimulation. IL-23 conditioning selectively upregulated IL-12 family signaling in IL23R-high clone – a pathway that shares JAK/STAT activation via shared IL12RB1 receptor – and induced genes linked to STAT3 signaling and Th17 differentiation (IL6ST, RAP1B, CDC42), as well as metabolic/redox regulators (TALDO1, MTAP, GSTO1)^34–36. Together, these findings provide a potential explanation for the observed functional plasticity seen in IL23R-high A8 clone, where IL-23/IL-23R signaling is important to engage in STAT3-dependent and metabolic programs, and thereby promote IL-17A production. These data would also argue that the presence of IL-23 in the appropriate environment could shape the effector function of MAIT cells.

Although human MAIT cells universally express RORγt and are abundant in both blood and tissues, only a small fraction – less than 5% in blood and 1% in the lung – produce IL-17^9,25. The role of IL-23/IL-23R signalling, which drives Th17 differentiation and IL-17 secretion in conventional CD4⁺ T cells, has been investigated in MAIT cells but appears to be context dependent. In blood MAIT cells, single-cell studies show that IL-23 enhances cytotoxic function²⁷, and dual stimulation through the TCR and cytokines is required to induce IL-17A production^25,26. Cole et al. reported that IL-17 secretion by TCR+IL-12+IL-18 stimulation is independent of IL-23 signaling, whereas Wang et al. demonstrated that TCR+IL-23 stimulation can drive IL-17A producing MAIT cells to produce IFN-γ^25,26. Our results not only confirm the findings from recent work, but also extend them by elucidating specific attributes of the IL-23 conditioning regimen that promote IL-17A production. We conclude that while IL-23R expression is necessary for IL-17A production, activation through this receptor must be complemented by TCR signaling. These requirements are best met in tissue environments, such as the lung, where both MR1-specific antigens and proinflammatory cytokines (IL-23 or IL-12+IL-18) are present.

Several technical and biological limitations inherent to the study of primary human T cell clones were present. Although we utilized three independently lung-derived MAIT cell clones with variable IL23R expression, we were unable to directly quantify IL-23R protein expression due to the low specificity and sensitivity of commercially available flow cytometry antibodies. Also, our study was limited to a small number of clones. Additional studies incorporating larger number of donors and lung-derived MAIT cell clones will be helpful to determine the generalizability of our findings. Lastly, future studies directly investigating IL-23/IL-23R signaling through receptor blockade, overexpression, or genetic deletion will be necessary to identify the mechanism in regulating IL-17A production. These future studies and validations may help explain pronounced IL-17A-producing phenotype in IL23R-med A5 clone that contribute to MAIT cell plasticity beyond IL-23R expression.

Understanding how IL-23/IL-23R signaling shapes MAIT cell function and plasticity could have important implications for vaccine design. We postulate that adjuvants capable of enhancing IL-23 production or potentiating IL-23R signaling could result in augmented MAIT cell effector responses. For example, Wang et al. demonstrated that co-administration of IL-23 with 5-OP-RU improved control of pulmonary Legionella infection in mice and a 50-fold reduction in bacterial burden²⁹. Moreover, Riffelmacher et al. showed that pulmonary immunization with an attenuated Salmonella strain expands two discrete MAIT cell lineages with distinct protective roles: an IL-17A-producing subset that protects against Streptococcus pneumoniae, and an IFN-γ-producing subset that protects against influenza²². Together with our findings, these studies suggest that targeted activation of IL-23R and the downstream STAT3 and type 17-associated genes can bias MAIT cells toward MAIT17 or mixed MAIT1/17 effector states, enhancing protective immunity against certain infections. Therefore, modulation of IL-23/IL-23R signaling is a potential therapeutic target to be considered when designing MAIT-targeted vaccines or immunotherapies.

Method

Study participants

Oregon Health & Science University, Oregon, USA. This study was conducted according to the Declaration of Helsinki. All samples and informed consent forms were collected, and all experiments were conducted according to protocols approved by the Institutional Review Board at Oregon Health & Science University (OHSU, IRB00000186). All ethical regulations relevant to human research participants were followed. Peripheral blood mononuclear cells (PBMCs) were obtained by apheresis from healthy adults. PBMCs were used for in vitro stimulation and to expand T cell clones as described below. Human serum was obtained from healthy adults.

Durban, South Africa. This study was conducted according to the Declaration of Helsinki. All samples and written informed consents from participants were collected, and all experiments were conducted according to protocols approved by the University of KwaZulu Natal Human Biomedical Research Ethics Committee and the Partners Institutional Review Board. All ethical regulations relevant to human research participants were followed. Bronchoalveolar lavage (BAL) fluid was obtained for isolating T cell clones. Collection of excess fluid from adult patients undergoing clinically indicated diagnostic bronchoscopies was conducted at Inkosi Albert Luthuli Central Hospital. BAL-derived lung MAIT cell clones were isolated from participants, including patients with active tuberculosis and uninfected controls with suspected lung cancer as characterized in the previous study⁹.

T cell clones

Isolation and characterization of T cell clones. For isolation of B1023-A5 and B1026-A8 clones, cells from BAL samples were stained with Propridium Iodide (Miltenyi Biotec), MR1/5-OP-RU tetramer (NIH tetramer core), and an antibody cocktail of surface stains to sort out non-CD3+ cells (PE/Cyanine7-conjuaged anti-CD14 (clone M5E2), anti-CD15 (clone HI98, anti-CD16 (clone B73.1), anti-CD19 (clone HIB19), anti-CD123 (clone 6H6), anti-CD235 (clone HIR2) antibodies from BioLegend). Live MR1/5-OP-RU tetramer positive cells were sorted and then cryopreserved on an Influx cytometer (BD Biosciences). After thawing, the cells were plated in a 96-well plate in a limited dilution assay with 1.5e5 cells irradiated PBMC (3000 cGray) and 3e4 cells irradiated lymphoblastoid cell line (LCL) (6000 cGray) along with αCD3 (30 ng/mL), rhIL-2 (2 ng/mL), rhIL-12 (0.5 ng/mL; BioLegend), rhIL-7 (0.5 ng/mL; BioLegend), and rhIL-15 (0.5 ng/mL; BioLegend) in cytokine-supplemented RPMI 1640 medium (Gibco) containing 10% heat-inactivated human serum (R10HuS) in 96 well round bottom plates. On day 5, αCD3 was washed out by removal of half of the volume from the wells and replaced with additional cytokine-supplemented medium. Medium was changed for all wells every 2–3 days and replaced with cytokine-supplemented medium. T cell “buttons” were evaluated for growth on Day 20 and selected buttons stained with the MR1-5-OP-RU tetramer, PE/Cyanine7 anti-CD3 (clone SK7; BioLegend), FITC anti-CD4 (clone SK3; BioLegend), APC/Cyanine7 anti-CD8 (clone SK1; BioLegend), and BV605 anti-TCR Vα7.2 (clone IP26; BioLegend) antibodies. T cell clones that remained MR1-5-OP-RU tetramer positive were re-expanded in a 24 well non-ULA plate with irradiated 7.5e5 PBMC and 1.5e5 LCL as well as R10HuS supplemented with αCD3 and IL-2 but without additional rhIL-12, rhIL-7, or rhIL-15³¹. Briefly, human lung MAIT cell D1004-A3 clone and human lung non-MAIT clones (D0033-JE1) were previously isolated in the same manner as above, sorted for live MR1/5-OP-RU tetramer positive and live MR1/5-OP-RU tetramer negative CD8 positive cells, and characterized at the same time as other D1004-associated clones⁹.

Rapid expansion protocol. For expansion, T cell clones were co-cultured with a ratio of 5:1 for irradiated allogenic PBMCs and irradiated allogenic lymphoblastoid cell lines in R10HuS and αCD3 antibody (30 ng/mL)⁴⁰. Recombinant interleukin-2 (2 ng/mL) was added the following day and every 2-3 days thereafter. T cell clones were washed on Day 5 to remove anti-CD3 antibody. Clones were frozen down at -80°C for 1-1.5e6 cells/mL after at least 11 days and used from frozen. New freezeback stocks were validated before use by flow cytometry.

Reagents and chemicals

5-(2-oxopropylideneamino)-6-d-ribitylaminouracil (5-OP-RU) was made from 5-A-RU*HCl that was prepared by the OHSU Medicinal Chemistry Core following a previously described procedure⁴¹. To prepare 5-OP-RU, equal volumes of circa 650 mM methylglyoxal (Sigma) and 32 mM 5-A-RU*HCl dissolved in water were combined immediately before addition to the assay to give a stock solution of 16 mM of 5-OP-RU⁴⁰. Recombinant human IL-23 protein (R&D systems) was aliquoted and frozen at -20°C. IL-23 protein was diluted in R10HuS to appropriate conditions before use.

Stimulation with IL-23 cytokine

IL-23 conditioning of PBMC. PBMC from a healthy donor were thawed at Day 0 and rested overnight in R10HuS. On Day 1 and 7, PBMC were stimulated with 5-OP-RU (1 µM). Starting at Day 1, PBMC were added with recombinant IL-2 (20 ng/mL) and with or without recombinant IL-23 (20 ng/mL) every other day until Day 11. Cells were harvested and analyzed for flow cytometry on Day 12.

IL-23 conditioning of MAIT cell clones. T cell clones were thawed and co-cultured with irradiated allogenic PBMCs and irradiated allogenic lymphoblastoid cell lines in R10HuS. On Day 0 and 6, T cell clones were stimulated with 5-OP-RU (1 µM). Starting at Day 0, T cell clones were added with recombinant IL-2 (2 ng/mL) and with or without recombinant IL-23 (20 ng/mL) every other day until Day 10. Cells were harvested and analyzed for flow cytometry assay or single-cell sequencing.

1-day short-term IL-23 stimulation on MAIT cell clones for single-cell sequencing. 5e5 T cell clones were plated and incubated with following conditions: no stimulation, 1-day stimulation with IL-23 (200 ng/mL), 3-hr PMA/Ionomycin stimulation. After stimulation, T cell clones were harvested and prepared for single-cell sequencing.

Flow cytometry assays

The following reagents were obtained through NIH Tetramer Core Facility: MR1/5-OP-RU tetramer and MR1/6-FP tetramer conjugated to R-phycoerythrin (PE). Identification of MAIT cells with MR1 tetramers was described before⁴². To determine the total functional capacity of T cells, PBMC and T cell clones were incubated with PMA/Ionomycin (PMA, 20 ng/mL, Sigma; Ionomycin 1µM, Sigma) for 3 hours in the presence Brefeldin-A (5 µg/mL, BioLegend). For tetramer and intracellular cytokine staining, cells were then harvested and stained with MR1/5-OP-RU tetramer and MR1/6-FP tetramer at a 1:250 dilution in tetramer buffer (PBS containing 2% FBS) for 45 minutes at room temperature. After 45 minutes, cells were stained with Live/Dead Fixable Dead Cell Stain Ki (Thermo Fisher) and surface antibodies (APC anti-CD4 antibody, clone RPA-T4, BD Biosciences #555349; APC/Cyanine7 anti-CD8 antibody, clone SK1, BioLegend #344714; PE/Cyanine7 anti-TCR Vα7.2 antibody, clone 3C10, BioLegend #351711) for 30 minutes at 4°C. Cells were washed then fixed and permeabilized using Cytofix/Cytoperm (BD Biosciences) per manufacturer’s instructions. Cells were then stained FITC anti-IFN-γ antibody (clone B27, BioLegend #506504) and BV421 anti-IL-17A antibody (clone BL168, BioLegend #512322). All data was obtained with LSR II (BD) cytometer at the OHSU Flow Cytometry Shared Resource and analyzed with FlowJo software version 10 (TreeStar).

RNA isolation, cDNA synthesis, and qPCR analysis

To measure IL23R expression, total RNA was isolated from human lung MAIT T cell clones (D1004-A3, B1023-A5, B1026-A8) using the RNeasy Plus Mini Kit (Qiagen). RNA was reverse transcribed into cDNA using the High-Capacity RNA-to-cDNA kit (Applied Biosystems) according to the manufacturer’s instructions. Quantitative RT-qPCR was performed on a Step One Plus Real-Time PCR System (Applied Biosystems) using TaqMan Universal PCR Master Mix (Life Technologies). Samples were run in triplicates and analyzed using Taqman probes (ThermoFisher Scientific) for IL23R (Hs00332759_m1) and for normalization GAPDH (Hs02758991_g1). Relative expression of human lung MAIT T cell clones were compared to human lung non-MAIT CD8 T cell clone (D0033-JE1).

Single-cell sequencing of MAIT cell clones

After stimulation, T cell clones were incubated in Human TruStain FcX (BiolegendTM) for 10 minutes at 4ºC and then stained with TotalSeq-C hashtag antibodies and CITE-seq antibodies (Biolegend^TM) listed in Supplemental Table 3 for 30 minutes at 4ºC. T cell clones were washed and counted to be loaded in 10X Genomics Chromium Next GEM Single Cell 5’ v2. Full protocol details of single-cell GEX and cell surface protein library preparation can be found from 10X genomics website (https://cdn.10xgenomics.com/image/upload/v1722286086/support-documents/CG000330_Chromium_Next_GEM_Single_Cell_5_v2_Cell_Surface_Protein_UserGuide_RevG.pdf). Finished libraries were then sent to Novogene Corporation Inc.TM in Sacramento California for NovaSeq 6000 sequencing.

Single-cell RNA-seq pre-processing

Raw sequence reads were processed using 10X Genomics Cell Ranger software (version 6.1.1). The resulting sequence data were aligned to the GRCh38 human genome. Cell demultiplexing used a combination of algorithms, including GMM-demux, demuxEM and BFF, implemented using the cellhashR package^43–45. Droplets identified as doublets (i.e. the collision of distinct sample barcodes) were removed from downstream analyses. We additionally performed doublet detection using DoubletFinder, and removed doublets from downstream analysis⁴⁶. Next, droplets were filtered based on UMI count (allowing 0-20,000/cell), and unique features (allowing 200-5000/cell). Additionally, we computed a per-cell saturation statistic for both RNA and ADT data, defined as: 1 – (#UMIs / #Counts). This statistic provides a per-cell measurement of the completeness with which unique molecules are sampled per cell and has the benefit of being adaptable across diverse cell types. Data were filtered to require RNA saturation > 0.35. Analyses utilized the Seurat R package, version 4.2⁴⁷. Using standardized methods implemented in the Seurat R package, counts and UMIs were normalized across cells, scaled per 10,000 bases, and converted to log scale using the ‘NormalizeData’ function. These values were then converted to z-scores using the ‘ScaleData’ command. Highly variable genes were selected using the ‘FindVariableGenes’ function with a dispersion cutoff of 0.5. Principal components were calculated for these selected genes and projected onto all other genes using the ‘RunPCA’ and ‘ProjectPCA’ commands. Clusters of similar cells were identified using the Louvain method for community detection, and UMAP projections were calculated. CITE-seq data were CLR-normalized by lane, meaning raw count data are subset per lane, CLR normalization performed (as implemented in the Seurat R package, using margin = 1), using all QC-passing cells/lane. Normalization per-lane was performed to reduce batch effects.

Statistical analysis

Statistical analysis was done in R studio for all differential gene expression of single-cell gene expression data. Wilcoxon test for statistical significance of gene expression between different groups. For all analysis we considered a 2-log threshold change of more than 0.5 and a Bonferroni-corrected p-value of < 0.05 to be significant. For histograms, data were graphed and analyzed with Prism 10 (GraphPad). Two-tailed paired Student’s t-test and one-way ANOVA Tukey’s multiple comparisons test was used for statistical analysis.

Author contributions

The experiments presented were conceptualized by SK, DK, DAL, GS, and DML. SK, DK, GS, MC, EW, and SK performed experiments, as well as, analysis and interpretation of data. DAL and DML supervised the work. DAL, GS, MC, BB, GM, TR, and DML provided advice and technical expertise. All authors contributed to revising and reviewing the manuscript. All authors approved the final version of the manuscript.

Funding

This project was funded in whole or in part by NIH T32HL083808 (SK), Canadian Institutes of Health Research (MFE CIHR-IRSC:0633005491) (DK), AI129980 (DAL, DML), and ImpacTB (75N93019C00070) (DAL, DML). This work was also supported in part by Merit Review Award # BX000533 from the United States (U.S.) Department of Veterans Affairs Biomedical Laboratory Research and Development Service. The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

Data availability

All R studio code used in the generation of this analysis is freely available and uploaded to github (https://github.com/kaindylan/IL23R-MR1T-Paper/tree/main). All raw data was deposited to dbgap (accessions number: phs003810.v1.p1). The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information files. The source data underlying the graphs in the paper can be found in the Supplementary Data. The accession number for the atomic coordinates of CB964 A2 TCR-MR1-5-OP-RU, along with associated structure factors, have been deposited at the protein databank (www.rcsb.org) with accession code 9MS0. Sequence data is available in SRA database referenced as PRJNA1420921(https://dataview.ncbi.nlm.nih.gov/object/PRJNA1420921?reviewer=ll624nk9527rquedgiedosinft). The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Declarations

Competing of interest

The authors declare no competing interests.