Changes in Systemic Regulatory T Cells, Effector T Cells, and Monocyte Populations Associated With Early-Life Stunting

Abstract

Stunting and environmental enteric dysfunction (EED) may be responsible for altered gut and systemic immune responses. However, their impact on circulating immune cell populations remains poorly characterized during early life. A detailed flow cytometry analysis of major systemic immune cell populations in 53 stunted and 52 non-stunted (2 to 5 years old) children living in Antananarivo (Madagascar) was performed. Compared to age-matched non-stunted controls, stunted children aged 2-3 years old had a significantly lower relative proportion of classical monocytes. No significant associations were found between stunting and the percentages of effector T helper cell populations (Th1, Th2, Th17, Th1Th17, and cTfh). However, we found that HLA-DR expression (MFI) on all memory CD4⁺ or CD8⁺ T cell subsets was significantly lower in stunted children compared to non-stunted controls. Interestingly, in stunted children compared to the same age-matched non-stunted controls, we observed statistically significant age-specific differences in regulatory T cells (Treg) subsets. Indeed, in 2- to 3-year-old stunted children, a significantly higher percentage of memory Treg, whilst a significantly lower percentage of naive Treg, was found. Our results revealed that both innate and adaptive systemic cell percentages, as well as activation status, were impacted in an age-related manner during stunting. Our study provides valuable insights into the understanding of systemic immune system changes in stunted children.

Keywords: Madagascar; environmental enteric dysfunction; flow cytometry; monocytes; regulatory T cells; stunting; systemic immune cells.

Figure 1

Characterization of monocyte subsets in non-stunted vs. stunted children. (A) Representative flow cytometry dot plots of CD14 and CD16 expression on monocyte subsets from non-stunted (left) and stunted (right) children. Classical, non-classical and intermediate monocyte subsets were defined as CD14⁺CD16^low/int, CD16^hiCD14^low, and CD16^hiCD14^+, respectively. (B–D) Relationships between stunting status and the percentages of classical, non-classical, and intermediate monocyte populations. NS, non-stunted; S, stunted. Significance (p<0.05) was determined by Benjamini-Hochberg correction after Mann-Whitney test. N total = 98, N non-stunted = 50, N stunted=48.

Figure 2

HLA-DR expression (MFI) on helper T (CD4+) and cytotoxic T cells (CD8+). (A) Representative dot plots of the gating strategy of T cells subsets based on their surface expression of CD27 and CD45RA markers. In each T cell subsets, naïve (CD27⁺CD45RA⁺), central memory (CM, CD27⁺CD45RA^-), effector memory (EM, CD27^-CD45RA^-), and CD45RA⁺ effector memory (EMRA, CD27^-CD45RA⁺) cells were defined. Representative dot plots from a non-stunted child are shown. (B) Representative histograms of HLA-DR expression on surface of naïve, CM, EM, EMRA T cells in stunted (red) versus non-stunted (blue) children. (C, D) Boxplot displaying the MFI values within all the T cell subsets in stunted (red) versus non-stunted (blue) children. Significance (p<0.05) was determined by Benjamini-Hochberg correction after Mann-Whitney test. N total= 101, N non-stunted = 50, N stunted=51.

Figure 3

Characterization of peripheral blood regulatory T cells (Treg) and their subsets in non-stunted vs. stunted children. (A) Representative flow cytometry dot plots showing the differences of Treg cell percentages from one non-stunted vs. one stunted child. Regulatory T cells were gated in CD4+ cells based on their expression of CD25 and CD127 markers. Naïve, memory and activated Treg cell subsets were determined based on their expression of CD127 and HLA-DR. (B) Relationship between Treg cell percentages and stunting status and (C) child age in years. (D) Representative graphs of regulatory T cell percentages by age sub-groups (years) and stunting status. Significance (p<0.05) was determined by Benjamini-Hochberg correction after Mann-Whitney test. N total= 103, N non-stunted = 51, N stunted= 52. ns, not significant.

Figure 4

Representative graphs of the percentages of Treg cells subsets stratified by nutritional status and age sub-groups. (A) Percentages of regulatory T cell subsets children by stunting status. Percentages of naïve (B), memory (C), and activated (D) Treg cells by stunting status and age sub-groups (years). P-values were FDR corrected after Mann-Whitney test. N total= 103, N non-stunted = 51, N stunted= 52. ns, not significant.

Figure 5

Distribution of cell percentages in the T reg panel. Summary of PERMANOVA analysis of the cell percentages in the full dataset, stratified by age and in each age category individually. The outcomes of the PERMANOVA analysis were the cells percentages in the “Treg panel”. As cell percentages are related to their “parents gate,” we analyzed the percentages of the five non-overlapped subpopulations: CD8⁺ Tc cells, CD4⁺CD25^-CD127⁺ Th cells, Treg naïve (CD4⁺CD25⁺CD127^-CD45RA⁺HLA-DR^-), Treg memory (CD4⁺CD25⁺CD127^-CD45RA^-HLA-DR^-), and Treg activated (CD4⁺CD25⁺CD127^-CD45RA^-HLA-DR⁺). The tested variables were as follows: HAZ score (height-for-age z-score); Age (child’s age in months); Anemia (presence or absence of anemia); AAT (fecal alpha-1 antitrypsin in mg/g dry weight); Calprotectin (fecal calprotectin in μg/g dry weight); CRP (serum C-reactive protein in mg/l); and parasite carriage. *Starred variables are significant with an FDR-corrected p<0.05. Each variable was tested individually in the PERMANOVA without other covariates. Coef: the coefficient of variance by PERMANOVA.