Human MAIT and CD8αα cells develop from a pool of type-17 precommitted CD8+ T cells

Abstract

Human mucosal associated invariant T (MAIT) CD8(+) and Tc17 cells are important tissue-homing cell populations, characterized by high expression of CD161 ((++)) and type-17 differentiation, but their origins and relationships remain poorly defined. By transcriptional and functional analyses, we demonstrate that a pool of polyclonal, precommitted type-17 CD161(++)CD8αβ(+) T cells exist in cord blood, from which a prominent MAIT cell (TCR Vα7.2(+)) population emerges post-natally. During this expansion, CD8αα T cells appear exclusively within a CD161(++)CD8(+)/MAIT subset, sharing cytokine production, chemokine-receptor expression, TCR-usage, and transcriptional profiles with their CD161(++)CD8αβ(+) counterparts. Our data demonstrate the origin and differentiation pathway of MAIT-cells from a naive type-17 precommitted CD161(++)CD8(+) T-cell pool and the distinct phenotype and function of CD8αα cells in man.

Figure 1

MAIT cell and CD161⁺⁺CD8⁺ populations are overlapping subsets. (A) CD3⁺CD161⁺⁺Vα7.2⁺ T cells are predominantly CD8αα⁺ and CD8αβ⁺ cells. Left hand panel shows representative FACS plots of coreceptor usage by Vα7.2⁺ T cells. Cumulative data shown in right hand panel. ***P < .001 (1-way ANOVA). (B) Representative FACS plots show a prominent CD8α⁺CD8β⁻ population exclusive to the CD161⁺⁺CD8α⁺ subset in peripheral blood. Data representative of 20 healthy donors. (C) CD161⁺⁺CD8α⁺ T cells are predominantly Vα7.2⁺ (mean 87% Vα7.2⁺) compared with the CD161⁺ and CD161⁻ CD8α⁺ subsets. **P < .001 (1-way ANOVA).

Figure 2

A naive CD161⁺⁺CD8α⁺ T-cell population can be identified in cord blood. (A) Representative FACS plot shown of cord blood showing the CD161⁺⁺CD8α⁺ cell population. Data representative of 22 cord samples. (B) CD161⁺⁺CD8α⁺ T cells in cord blood are CD45RA⁺, CD45RO⁻, CCR7_low, and CD62L_low. Representative FACS plots and cumulative data shown. **P < .001, *P < .05 (1-way ANOVA).

Figure 3

CD161⁺⁺CD8α⁺ T cells in cord blood are a preprogrammed type-17 subset. (A) Comparison of log fold change in gene expression for selected genes between CD161⁺⁺CD8α+ versus CD161⁻CD8α⁺ T cells in adult and cord blood. Data obtained from 3 cord samples and 3 adult healthy donors. (B) Gene set enrichment summary plots comparing gene expression on microarray analysis between CD161⁺⁺CD8α⁺ versus CD161⁻CD8α⁺ subsets from adult and cord blood. Left panel shows the plot for up-regulated genes, right hand panel shows plot for down-regulated genes. Normalized enrichment score (NES), up-regulated genes, = 4.21, P < .00001. NES, down regulated genes, = -3.25, P < .0001. (C) Representative FACS plots and cumulative date showing CD161⁺⁺ CD8α⁺ cells in cord blood express high levels of IL-18R and CCR6. **P < .001 (1-way ANOVA). (D) Relative fold change in gene expression for RORC, ZBTB16, IL-18RAP, and IL-23R in CD161⁺⁺CD8α⁺ compared with the CD161⁺ and CD161⁻CD8α⁺ subsets in cord blood by real time PCR analysis.

Figure 4

Distinct phenotypic and functional features of the CD161⁺⁺CD8α⁺ population are shared between the CD161⁺⁺ CD8αα⁺ and CD8αβ subsets. (A) Representative histograms showing CD161⁺⁺ CD8αα⁺ and CD8αβ cells to express CCR6, CCR2, and CXCR6. Cumulative data shown. ns = non significant (Mann-Whitney test). (B) Cumulative data showing expression of IFN-γ, IL-17 and IL-22 by CD161⁺⁺ CD8αα and CD8αβ cells. *P = .0467, ns = non significant (Mann- Whitney). (C) Heat-map showing up (red) and down (green) regulated gene expression of CD161⁺⁺CD8αα (light blue) compared with CD8αβ (dark blue) cells on microarray.

Figure 5

CD161⁺⁺CD8α⁺ T cells in cord blood have broad T-cell receptor usage compared with adults. (A) Comparison of proportional TCR Vβ usage by CD161⁺⁺CD8α⁺ T cells between cord and adult blood using TCR Vβ antibody panel. Data obtained from 8 cord blood samples and 10 adult healthy donors. (B) Representative FACS plot of Vα7.2 usage by CD161^++/+/− subsets in cord blood (C) Comparison of TCR Vα 7.2 usage between the CD161⁺⁺CD8α⁺ subsets in cord and adult blood. ***P = .002 (Mann-Whitney).

Figure 6

The CD161⁺⁺ CD8αα population emerges on expansion of the CD161⁺⁺CD8αβ subset with age. (A) CD161⁺⁺ CD8αα cells are not found in cord or infant blood. (B) Comparison of CD161⁺⁺CD8α⁺ and CD161⁺⁺CD8αα populations in cord, infant, and adult blood. **P < .001, ***P < .0001 (1-way ANOVA). (C) Comparison of TCR Vβ and Vα usage between CD161⁺⁺ CD8αβ and CD161⁺⁺CD8αα subsets in adult blood.

Figure 7

CD161⁺⁺ CD8αα T cells can be derived from CD161⁺⁺CD8αβ cells in culture. (A) CD161⁺⁺, CD161⁺ and CD161⁻ populations of the CD8 enriched/CD8β⁺ selected PBMCs remain stable over the 7 days of culture in R10 media alone. (B) Representative FACS plots demonstrating change in CD8β expression of CD161⁺⁺, CD161⁺ and CD161⁻ populations of the CD8 enriched/CD8β⁺ selected PBMCs in media alone over 7 days of culture. CD8αα population seen only in CD161⁺⁺ population. (C) Representative histograms showing change in CD8β expression of the CD161⁺⁺CD8α⁺ population of the CD8 enriched/CD8β⁺ selected PBMCs over 7 days of culture (top panel). Cumulative data shown (bottom panel), *P < .05, **P < .001, and ***P < .0001 (1-way ANOVA). Data representative of 4 experiments. (D) Cumulative data showing no significant change in the CD8α⁺ T-cell population of cultures over course of experiment (1-way ANOVA). (E) Ki-67 proliferation data for sorted CD8β⁺ cells left in culture media alone or with CD3/CD28 beads for 48 hours. Baseline Ki-67 staining (gray), after culture Ki-67 staining (black). Data representative of 2 experiments.