MAIT Cells Linked to Cancer Progression

Background Mucosa-associated invariant T (MAIT) cells represent a unique population of innate-like T lymphocytes capable of detecting non-peptide antigens in the context of monomorphic antigen-presenting molecules. Due to their abundance in barrier tissues, reactivity to local inflammatory cues, and cytotoxic and regulatory functions, MAIT cells are poised to shape the dynamics of various tumor microenvironments. Growing evidence suggests that MAIT cells can exert protumor and/or antitumor effects in cancers arising from or metastasizing to mucosal tissues. However, MAIT cell roles in bladder cancer (BCa) remains unclear.

Methods To begin to identify MAIT cells in BCa, we stained bladder tumor biopsies for T cell receptor (TCR) Vα7.2+ cells. We then refined a human MAIT cell signature, which enabled us to interrogate a bulk RNA sequencing dataset and conduct correlation analyses linking intratumoral MAIT cell abundance and mortality from BCa. To extend our work to an in vivo setting, we employed a clinically relevant mouse model in which Mr1 +/+ B6-MAITCAST (MAIT-sufficient) and Mr1 −/− B6-MAITCAST (MAIT-deficient) mice were exposed to N-butyl-N-(4-hydroxybutyl)nitrosamine, a chemical carcinogen associated with tobacco smoke. In additional experiments, MAIT cells were functionally removed through acetyl-6-formylpterin (Ac-6-FP) administration. Effector and regulatory cell types were phenotyped by flow cytometry, and BCa tumor burden and progression were assessed by MRI and/or H&E and Ki67 staining.

Results TCR Vα7.2+ cells were readily detectable in several BCa biopsies, and our bioinformatic analyses correlated heavier MAIT cell presence in BCa tumors with poorer overall survival. Similarly, we found higher tumor burdens in Mr1+/+ B6-MAITCAST mice than in Mr1−/− or Ac-6-FP-treated animals. Bladder MAIT cells from tumor-bearing mice exhibited phenotypic MAIT17 bias based on transcription factors they harbored along with increased interleukin-17A and tumor necrosis factor-α production capacities upon stimulation. Finally, FoxP3+ regulatory T (Treg) cell frequencies were elevated in Mr1+/+ mouse bladder tumors, likely contributing to an immunosuppressive tumor microenvironment, a finding that could be recapitulated in our transcriptomic studies on human BCa.

Conclusions MAIT cells are abundant in BCa tumor microenvironments where they potentiate Treg cell accumulation and play protumor roles.

What is already known on this topic

Mucosa-associated invariant T (MAIT) cells are a unique subset of T cells with powerful immunomodulatory properties that can promote or impede antitumor immunity in different cancers. MAIT cell roles in bladder cancer (BCa) have been essentially unexplored.

MAIT cells are present in bladder tumor microenvironments where they exhibit a MAIT17 phenotype, promote regulatory T cell accumulation, and potentiate cancer progression.

How this study might affect research, practice or policy

Our findings introduce MAIT cells as an attractive target for BCa immunotherapy. Given the monomorphic nature of major histocompatibility complex-related protein 1 (MR1), which presents antigens to MAIT cells, MR1/MAIT cell-based interventions, once optimized, should work in genetically diverse human populations.

In 2023, bladder cancer (BCa) was ranked as the fourth most common malignancy and the eighth leading cause of cancer-related deaths in men in the USA.1 The single most important risk factor for BCa is smoking. Tobacco smokers have a threefold risk increase for developing BCa compared with those who have never smoked,2 with smoking estimated to be responsible for approximately 50% of BCa cases.3

Non-metastatic, organ-confined BCa can be categorized as either non-muscle-invasive BCa (NMIBC) or muscle-invasive BCa (MIBC) depending on the depth of bladder wall invasion. NMIBC and MIBC make up approximately 70–75% and 25–30% of newly diagnosed BCa cases, respectively.4 The management of BCa depends largely on the type of BCa and the stage of the disease. Almost all patients with NMIBC are initially treated with transurethral resection of bladder tumor (TURBT). Patients who are found to have high-risk NMIBC may be treated with intravesical Bacillus Calmette-Guérin (BCG) to stimulate a local immune response, which may in turn prevent tumor recurrence and its progression to MIBC.5 6 Patients diagnosed with MIBC are often treated with neoadjuvant chemotherapy followed by radical cystectomy or with trimodal therapy consisting of maximal TURBT and concurrent chemoradiation.7 The US Food and Drug Administration has approved the use of immune checkpoint inhibitors, such as nivolumab, in the treatment of high-risk MIBC and metastatic BCa.8 The application of immunotherapy in the treatment of both NMIBC and MIBC highlights the need for in-depth investigations into the BCa immune landscape as they may shed light on novel targets with therapeutic potentials.

Mucosa-associated invariant T (MAIT) cells are a subset of innate-like T cells with remarkably potent immunomodulatory properties. Unlike mainstream T cells, MAIT cells express highly conserved, semi-invariant T cell receptors (TCR) (TRAV1-2-TRAJ33/12/20 in humans, and Trav1-Traj33 in mice), and are not major histocompatibility complex (MHC)-restricted.9 MAIT cells recognize vitamin B2 metabolites of microbial origin presented by the monomorphic MHC-related protein 1 (MR1).10 They also respond to inflammatory cytokine signals in the presence or absence of TCR triggering.11 Following activation, MAIT cells promptly release T helper (TH)1-type and/or TH17-type cytokines as well as cytotoxic effector molecules. Found abundantly in various mucosal tissues, as well as in the human liver,12 MAIT cells occur in a poised state to contribute to the outcome of various malignancies.

The presence of MAIT cells in primary and metastatic tumors and their roles in cancer progression or anticancer immune surveillance have been a subject of recent investigations. To this end, several groups, including ours, have reported the significance of MAIT cells in liver,13 14 lung,15 16 colorectal17 18 and ovarian19 cancers. Depending on the tumor microenvironment (TME) in which MAIT cells operate, these cells can serve protumor or antitumor functions.20 MAIT cells are able to release potent cytokines, such as tumor necrosis factor (TNF)-α, interferon (IFN)-γ, and interleukin (IL)-17, which in turn modulate the activity of surrounding immune cells.12 IL-17-producing MAIT cells are often associated with tumor-promoting effects, while IFN-γ-producing MAIT cells are usually expected to promote antitumor immunity.20 MAIT cells can also exhibit cytotoxic effector functions via cell surface-bound receptors, such as NKG2D,21 22 and/or through degranulation to release perforin, granzymes and granulysin.23 24 Despite extensive research on MAIT cells in numerous cancers, their biological behavior in BCa remains far from clearly understood. For instance, Zhang et al 25 found TCRVα7.2+CD161+ T cells, the vast majority of which should be MAIT cells, in BCa tumors. However, they did not report any possible relationships between the frequency of these cells and disease severity.

In this work, we used bladder tumor biopsies, a publicly available human BCa dataset, and a clinically relevant mouse model to investigate MAIT cell roles and characteristics in BCa.

Bulk RNA dataset analysis

Bulk transcriptomic data and corresponding clinical data for The Cancer Genome Atlas (TCGA) urothelial BCa dataset were retrieved from UCSC Xena (http://xena.ucsc.edu/) and cBioPortal (https://www.cbioportal.org/), respectively.26 To define MAIT cells, we used an eleven-gene MAIT cell signature, which we established before,27 along with TRAV1-2, which encodes the human MAIT cell TCR α-chain. We then applied variance stabilizing transformation on these genes (SLC4A10, KLRB1, ME1, TMIGD2, IL23R, NCR3, LST1, COLQ, RORC, ZBTB16, TLE1, TRAV1-2) using the R package DESeq2. Non-parametric and tie-robust Kendall’s rank correlation coefficients (Kendall’s τ) were determined for each of these genes to produce a correlation matrix. Of the above-listed 12 genes, we selected only those that (1) correlated with T cell markers CD3E, CD3G and CD3D; (2) correlated with well-documented MAIT cell markers SLC4A10, ZBTB16 and TRAV1-2; and (3) correlated with one another.27 A refined six-gene signature (SLC4A10, KLRB1, ZBTB16, TRAV1-2, NCR3, LST1) emerged as a result and was used for downstream analyses.

The above genes, as well as CD3E, CD3G and CD3D, were z-standardized to their mean and variance across all individuals. For each tumor sample, we averaged the standardized gene expression values to calculate a temporary MAIT cell score. Similarly, we averaged the standardized gene expression values of the three T cell genes for a T cell score. As the final normalization step, we fitted a linear regression model, with the MAIT cell score regressed on the T cell score (MAIT cell score ~ T cell score). The final MAIT cell scores were defined by the residuals of this model—the difference between modeled values and actual data points. Tumor samples above the 66th quantile were classified as “High MAIT”, and samples below the 33rd quantile as “Low MAIT” for survival analyses, which were visualized with a Kaplan-Meier curve.

Samples were similarly analyzed to assign a regulatory T (Treg) cell score. Using a signature of five genes typically expressed by Treg cells (CTLA4, FOXP3, IL2RA, TNFRSF18, CD4) and a linear regression model (Treg cell score~T cell score), samples were assigned a Treg cell score based on the residuals of this model. Tumor samples above the 66th quantile were classified as “High Treg”, and samples below the 33rd quantile as “Low Treg”.

Based on the above-described scores, samples above the 66th quantile and below the 33rd quantile for both MAIT cell and Treg cell scores were classified as “High MAIT, High Treg” and “Low MAIT, Low Treg,” respectively, for survival analyses.

Immunofluorescence staining of bladder tumor biopsies