Measles Virus Infects and Programs MAIT Cells for Apoptosis

Ethics Statement

Peripheral blood (PB) was drawn from healthy volunteers, 11 men and 8 women (Supplementary Table 1). Tumor-free hepatic tissue samples were harvested from patients undergoing surgical operations listed in Supplementary Table 2. Our protocols were approved by the Western University Research Ethics Board for Health Sciences Research Involving Human Subjects.

Transcriptomic Analysis of Peripheral Blood Mononuclear Cells

The pbmc_10k_v3 dataset, consisting of single-cell RNA sequencing (scRNA-Seq) data from 11 769 peripheral blood mononuclear cells (PBMCs), was created by 10x Genomics (https://support.10xgenomics.com/single-cell-gene-expression/datasets/3.0.0/pbmc_10k_v3). Preprocessing and analysis were performed using Seurat v3.1.1. Features were retained if detected in ≥5 cells. Cells were retained if they contained ≥500 features, between 2000–20 000 unique molecular identifiers, and <20% mitochondrial transcript reads. Scrublet was used to identify and remove cells with doublet scores >0.1. After log-normalizing the count data and identifying the 3000 most variable genes, expression values were scaled, centered, and used for principal component analysis. The top 30 components were used for cluster detection with the spatial linear model algorithm of Seurat v3.1.1 (resolution = 0.4). The resulting clusters were assigned cell type identities based on known marker genes (Supplementary Figure 1).

PBMC and Hepatic Mononuclear Cell Isolation

PBMCs were isolated using SepMate-50 tubes (STEMCELL Technologies) and density gradient centrifugation at 1200 × g. To obtain nonparenchymal hepatic mononuclear cells (HMNCs), liver samples were homogenized and parenchymal cells were removed by centrifugation at 700 × g in 33.75% Percoll PLUS. Exposure to ammonium-chloride-potassium lysis buffer eliminated erythrocytes, and the remaining HMNCs were washed in phosphate-buffered saline (PBS) containing 2% fetal bovine serum (FBS).

Cytokine Stimulation

PBMCs were suspended in RPMI supplemented with 10% FBS, 2 mM GlutaMAX-I, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and 10 mM HEPES. Cells were seeded at a density of 5×10⁵ cells/well of a U-bottom microplate and cultured for 24 hours at 37°C with or without 5 ng/mL each of recombinant human IL-12 (rhIL-12) p70 (PeproTech) and rhIL-18 (MBL International).

MeV Propagation and Infection

WT MeV strains Ichinose-B 323 and Khartoum Sudan were engineered to express enhanced green fluorescent protein (GFP). They are referred to as IC323-eGFP and KS-eGFP, respectively. MeVs were propagated in Vero cells expressing human CD150. PBMCs were seeded at 1×10⁵ cells/well of a microplate containing DMEM with 5% FBS, to which MeVs were added at a multiplicity of infection (MOI) of 1. Cultures were incubated at 37°C for 1, 4, 8, or 24 hours. In several experiments, PBMCs were stimulated with rhIL-12 and rhIL-18 before they were exposed to IC323-eGFP.

Tetramers and Antibodies

Phycoerythrin (PE)-conjugated 5-(2-oxopropylideneamino)-6-D-ribitylaminouracil (5-OP-RU)-loaded MHC-related protein 1 (MR1) tetramers and allophycocyanin (APC)-conjugated PBS-57-loaded CD1d tetramers were provided by the National Institutes of Health Tetramer Facility. 6-formylpterin-loaded MR1 tetramers and unloaded CD1d tetramers served as staining controls.

Fluorochrome-labeled monoclonal antibodies (mAbs) to CD3ε (clone UCHT1), CD4 (RPA-T4), CD19 (SJ25C1), CD38 (HIT2), CD45RO (UCHL1), CD46 (8E2), CD69 (FN50), CD150/SLAM [A12 (7D4)], CD197/CCR7 (3D12), and γδ TCR (B1.1), a mouse IgG1κ isotype control (P3.6.2.8.1), and an Annexin V Apoptosis Detection Kit APC were purchased from Thermo Fisher.

Cytofluorimetric Analyses

PBMCs and HMNCs were stained for 30 minutes at room temperature with tetramer/mAb cocktails prepared in PBS containing 2% FBS, washed, and interrogated using a FACSCanto II cytometer.

MAIT cells were defined as CD3⁺MR1 tetramer⁺ cells, iNKT cells as CD3⁺CD1d tetramer⁺ cells, γδ T cells as CD3⁺γδTCR⁺ cells, central memory T (T_CM) cells as CD3⁺MR1 tetramer⁻CD1d tetramer⁻γδTCR⁻CD45RO⁺CCR7⁺ cells, effector memory T (T_EM) cells as CD3⁺MR1 tetramer⁻CD1d tetramer⁻γδ TCR⁻CD45RO⁺CCR7⁻ cells, and naive T (T_N) cells as CD3⁺MR1 tetramer⁻CD1d tetramer⁻γδTCR⁻CD45RO⁻CCR7⁺ cells. Gates were drawn based on uninfected, unstained, or isotype control-stained conditions as appropriate. Bulk conventional CD4⁺ T cells and B cells were defined as CD3⁺CD4⁺tetramer⁻ and CD3⁻CD19⁺ cells, respectively.

To detect MAIT cells in early and late stages of death, PBMCs from MeV-infected and -uninfected cultures were harvested, washed, and stained at 4°C with Fixable Viability Dye eFluor 780. Cells were then washed and surface-stained with a PE-Cy7-conjugated anti-CD3 mAb and PE-labeled, 5-OP-RU-loaded MR1 tetramers. After 30 minutes, cells were washed and stained with Annexin V-APC for 15 minutes at room temperature.

Statistical Analyses

Statistical comparisons were performed using Student t tests, ANOVA, or Wilcoxon signed-rank tests. Kolmogorov-Smirnov tests were conducted to assess the normal distribution of indicated variables (Supplementary Table 3).

MAIT Cells Are the Foremost Expressors of CD150 Among PBMCs

To determine whether MAIT cells express SLAMF1, the gene encoding CD150 that mediates measles-associated immunosuppression, we analyzed a PBMC scRNA-Seq dataset from 10x Genomics.

Unsupervised clustering identified several PBMC subsets based on canonical marker genes (Supplementary Figure 1A). These included CD14⁺ monocytes, CD16⁺ monocytes, classical dendritic cells, plasmacytoid pre-DCs (pDCs), NK cells, B cells, and three T cell populations, which all expressed CD3D, CD3E, and TRAC and were further classified into conventional CD4⁺ T cells (CD4, MAL), conventional CD8⁺ T cells (CD8A, CD8B), and MAIT cells (NCR3, SLC4A10, RORC, DPP4, ZBTB16) [5] (Figure 1A and Supplementary Figure 1A and 1B). A minor platelet population was also detectable.

As expected, all PBMC subsets, but not the accompanying platelets, expressed CD46 (Supplementary Figure 2A and 2B). Strikingly, SLAMF1 expression was not only evident but most pronounced within the MAIT cluster (Figure 1A). Statistically, SLAMF1 was more highly expressed by MAIT cells than by any other cell types (P < 2.2e-16 by Wilcoxon signed-rank tests). Of note, iNKT cells did not form a distinct cluster due likely to their low frequency in human PB.

Cytofluorimetric staining of MAIT cells by 5-OP-RU-loaded MR1 tetramers confirmed their abundance in the circulation (Supplementary Figure 3A) and their surface expression of CD150 (Figure 1B and 1C). Both CD150⁺ cell frequencies and CD150 expression on a per-cell basis were significantly higher in PB MAIT cells than in bulk CD4⁺ T cells and B cells (Supplementary Figure 4A and 4B), γδ T, T_N, T_CM, and T_EM cells (Figure 1C and Supplementary Figure 5). This was notable because T_CM and T_EM cells are considered the main T cell subsets targeted by MeV, resulting in immune amnesia [2, 6]. The magnitude of CD150 expression by MAIT cells could be matched only by that of iNKT cells (Figure 1C and Supplementary Figure 4A and 4B).

MAIT cells are enriched in the human liver [4] (Supplementary Figure 3B). We asked whether hepatic and PB MAIT cells follow the same CD150/CD46 expression pattern. We found all MAIT cells among HMNCs to copiously express CD150 (Supplementary Figure 6A, 6B and 6C). Furthermore, the expression pattern of CD150 on various hepatic T cell subsets was similar to that of PB T cells (Supplementary Figure 6B and 6C).

MAIT Cells Exposed to IL-12 and IL-18 Retain Their CD150/CD46 Expression

Viruses lack the vitamin B biosynthesis machinery, which supplies prototoypic MR1 ligands [3], and fail to stimulate MAIT cells in a TCR-dependent fashion. However, they can trigger MAIT cell activation through cytokines, primarily through IL-18 and IL-12 whose receptors MAIT cells express [7]. We asked whether exposing MAIT cells to rhIL-12 and rhIL-18 alters their expression levels of MeV receptors. While 24-hour stimulation with these cytokines upregulated CD69, an early activation marker, the frequencies of CD150⁺ and CD46⁺ MAIT cells remained stable (Figure 1D). This treatment also increased, rather than decreased, the geometric mean fluorescence intensity of staining for these receptors (Supplementary Figure 7). Therefore, MAIT cells should maintain their susceptibility to MeV in a cytokine milieu established through infection with MeV or other viruses, or during coinfections.

MeV Infects MAIT Cells

To explore whether MeV could indeed infect MAIT cells, we incubated PBMCs for 24 hours with IC323-eGFP or KS-eGFP. Exposure to MeV virions resulted in efficient MAIT cell infection (Figure 2A, Supplementary Figure 8, and Supplementary Figure 9A). In these cultures, eGFP⁺ (infected) MAIT cells were significantly more frequent than any other infected T cell subsets with the sole exception of iNKT cells (Figure 2B and Supplementary Figure 9B). Importantly, the percentage of infected cells was higher among MAIT cells than in T_CM and T_EM cells, thus closely mirroring the CD150 expression pattern of these populations (Figure 1C). We found MeV infection not to activate MAIT cells because infected cells did not upregulate CD69 or CD38 (Supplementary Figure 10).

MeV Infection Kills MAIT Cells

To ascertain whether MeV infection of MAIT cells triggers their destruction, we costained PBMCs recovered from uninfected and IC323-infected cultures with Annexin V and Fixable Viability Dye. After gating on CD3⁺MR1 tetramer⁺ events, Annexin V⁺Viability Dye⁻ and Annexin V⁺Viability Dye⁺ cells were defined as early-stage apoptotic and late-stage apoptotic/necrotic MAIT cells, respectively. MAIT cells that did not stain positively with Annexin V and Fixable Viability Dye were considered viable. As early as 1 hour after MeV inoculation, the majority of eGFP⁺ (infected) MAIT cells were Annexin V⁺ (Figure 2C and 2D). Moreover, a substantial proportion of this population was in an early stage of the apoptotic process as judged by a lack of staining with Fixable Viability Dye (Figure 2C). In contrast, eGFP⁻ (uninfected) MAIT cells harvested from infected cultures contained smaller Annexin V⁺Viability Dye⁻ and Annexin V⁺Viability Dye⁺ fractions (Figure 2C and 2D). At 4-, 8-, and 24-hour timepoints post-MeV exposure, higher percentages of dying or dead cells were similarly detectable among infected MAIT cells (Figure 2D). MAIT cells from parallel cultures that had not received an MeV inoculum remained fairly viable over time (Figure 2D). Collectively, these results indicate that MeV infection quickly programs MAIT cells for apoptotic death.