MMR vaccination induces trained immunity via functional and metabolic reprogramming of γδ T cells

Abstract

The measles, mumps, and rubella (MMR) vaccine protects against all-cause mortality in children, but the immunological mechanisms mediating these effects are poorly known. We systematically investigated whether MMR can induce long-term functional changes in innate immune cells, a process termed trained immunity, that could at least partially mediate this heterologous protection. In a randomized, placebo-controlled trial, 39 healthy adults received either the MMR vaccine or a placebo. Using single-cell RNA-Seq, we found that MMR caused transcriptomic changes in CD14+ monocytes and NK cells, but most profoundly in γδ T cells. Monocyte function was not altered by MMR vaccination. In contrast, the function of γδ T cells was markedly enhanced by MMR vaccination, with higher production of TNF and IFN-γ, as well as upregulation of cellular metabolic pathways. In conclusion, we describe a trained immunity program characterized by modulation of γδ T cell function induced by MMR vaccination.

Keywords: Cellular immune response; Immunology; Innate immunity.

Figure 1. Study setup, plasma proteomics analysis after MMR vaccination, and WBC counts.

(A) Setup of the present randomized, placebo-controlled trial of MMR vaccination. M/F, males/females. (B) Volcano plot of Olink targeted proteomics (n = 1,289 analyzed proteins in total) in plasma, after MMR vaccination (n = 16). (C–F) Volcano plots of subcategories of the plasma proteome measured by Olink. In the volcano plots for the subpanels, the full panel is depicted in light gray as the background. The side plots in panels B–F show the relative expression values (NPX) of selected proteins in each (sub)panel. (G) Total and differentiated WBC counts, before and after vaccination in placebo- and MMR-vaccinated groups. An unadjusted P value of less than 0.05 is the cutoff for the volcano plots. P values for this figure were calculated using the Wilcoxon signed-rank test. T1, baseline; T2, one month after treatment.

Figure 2. Single-cell analysis of PBMCs following MMR vaccination.

(A) UMAP analysis of scRNA-Seq and snATAC-Seq of PBMCs. (B) Proportions of cell types annotated according to the scRNA-Seq data in placebo and MMR samples. (C) Top: Differentially expressed genes (scRNA seq) per cell type, between baseline (T1) and 1 month after placebo or MMR (T2). Bottom: differentially accessible genes (snATAC-Seq) per cell type, between baseline (T1) and 1 month after placebo or MMR (T2).

Figure 3. Single-cell analysis of monocyte subpopulations and monocyte-associated cytokine production by PBMCs.

(A) UMAP analysis and subpopulation identification of scRNA-Seq and snATAC-Seq, specifically in monocytes. (B) Pathway enrichment of genes that were differentially expressed in monocytes after MMR vaccination. (C) Top 20 differentially expressed genes in monocytes following MMR vaccination, by time point and treatment group. (D–H) Monocyte-associated cytokines produced by PBMCs following diverse stimulations; the data are expressed as log₂ fold changes between baseline and 1 month after treatment. P values for D–H were calculated using the Mann-Whitney U test. Box plots were drawn as follows: lower and upper hinges indicate the 25th and 75th percentiles; whiskers indicate the hinge plus 1.5 times the IQR; and the line in the box indicates the median.

Figure 4. Single-cell analysis of γδ T cell populations.

(A) UMAP analysis and subpopulation identification of scRNA-Seq and snATAC-Seq, specifically in γδ T cells. (B) Pathway enrichment analysis of genes that were differentially expressed in γδ T cells after MMR vaccination. (C) Top 20 differentially expressed genes in γδ T cells following MMR vaccination, by time point and treatment group.

Figure 5. Functional and metabolic characterization of Vδ2 cells following MMR vaccination.

(A) Percentage of Vδ2 T cells in isolated PBMCs. (B) Percentage of Vδ2 T cells that produced TNF or IFN-γ following CD3/CD28 stimulation. (C) Percentage of Vδ2 T cells expressing markers of cytotoxic granule release (CD107a) or production (perforin and GZMB). Metabolic parameters by modified SCENITH (https://www.scenith.com), calculated as in Argüello et al. (24): (D) puromycin incorporation, (E) fatty acid oxidation/amino acid oxidation (FAO/AAO) capacity, (F) glycolytic capacity, (G) mitochondrial dependence, and (H) glucose dependence. All parameters were measured by flow cytometry. P values were calculated using the Wilcoxon signed-rank test. Box plots were drawn as follows: lower and upper hinges indicate the 25th and 75th percentiles; whiskers indicate the hinge plus 1.5 times the IQR; and the line in the box indicates the median.