Developmental Regulation of Effector and Resident Memory T Cell Generation during Pediatric Viral Respiratory Tract Infection

Abstract

Viral respiratory tract infections (VRTI) remain a leading cause of morbidity and mortality among infants and young children. In mice, optimal protection to VRTI is mediated by recruitment of effector T cells to the lungs and respiratory tract, and subsequent establishment of tissue resident memory T cells (Trm), which provide long-term protection. These critical processes of T cell recruitment to the respiratory tract, their role in disease pathogenesis, and establishment of local protective immunity remain undefined in pediatric VRTI. In this study, we investigated T cell responses in the upper respiratory tract (URT) and lower respiratory tract (LRT) of infants and young children with VRTI, revealing developmental regulation of T cell differentiation and Trm generation in situ. We show a direct concurrence between T cell responses in the URT and LRT, including a preponderance of effector CD8⁺ T cells that was associated with disease severity. During infant VRTI, there was an accumulation of terminally differentiated effector cells (effector memory RA⁺ T cells) in the URT and LRT with reduced Trm in the early neonatal period, and decreased effector memory RA⁺ T cell and increased Trm formation with age during the early years of childhood. Moreover, human infant T cells exhibit increased expression of the transcription factor T-bet compared with adult T cells, suggesting a mechanism for preferential generation of effector over Trm. The developmental regulation of respiratory T cell responses as revealed in the present study is important for diagnosing, monitoring, and treating VRTI in the critical early life stages.

Figure 1. T cell composition in pediatric respiratory and blood sampls

Representative flow cytometry plots of T cell populations from matched blood (top row), upper respiratory tract (URT, middle row), and lower respiratory tract (LRT, bottom row) samples of representative VRTI subjects aged 6 weeks (left), 4 months (middle), and 1 year (right). (A) Gating strategy for analysis of T cells with single cells selected based on forward scatter properties, then lymphocytes based on forward and side scatter, followed by cells expressing CD3. (B) CD4⁺ and CD8⁺T cell composition of CD3⁺ T cells in each site.

Figure 2. Inverse predominance of CD4⁺ and CD8⁺ T cells in the blood and respiratory tract of VRTI subjects

(A) Plots shows CD8:CD4 T cell content of all samples obtained for Blood (n=21), URT (n=113) and LRT (n=82). Median CD8:CD4 T cell ratio for Blood was 0.34 (range; 0.09-1.04), URT was 0.96 (range; 0.12-8.42) and LRT was 1.40 (range; 0.17-8.42). Individual data points depicted with median and interquartile range. Dashed line represents a CD8:CD4 ratio of 1. (* = p=0.02, ****= p<0.0001). (B) Correlation analysis of the CD8:CD4 T cell ratio for all paired URT and LRT samples. P values obtained by Wilcoxon matched-pairs signed rank test and R² values obtained by linear regression analysis.

Figure 3. Respiratory CD8:CD4 T cell ratio associated with disease severity during VRTI

(A) Peak CD8:CD4 T cell ratio within an individual during the course of disease from the URT (left, n=31) and LRT (right, n=13) stratified by subjects diagnosed as having PARDS or not have PARDS. (LRT; PARDS median 2.94; range 0.8 – 8.4 vs no PARDS median 0.56; range 0.31 – 2.82, p < 0.0001; URT; PARDS median 2.58; range 0.8-8.21 vs no PARDS median 1; range 0.38 to 2.79, p = 0.002) (B) Peak CD8:CD4 T cell ratio in URT (left) and LRT (right) as in (A) stratified by viral pathogen. Bars in graphs represent medians. Filled data points represent subjects without PARDS while open points depict those with PARDS. Subjects who contributed both URT and LRT samples are denoted by unique color/shape combinations; as an example, the open red square is a subject with RSV who had samples analyzed from both the URT and LRT. Black data points represent individual subjects who contributed samples from only one site (URT or LRT). P values obtained by Mann-Whitney testing.

Figure 4. Developmental differences in T cell subset composition in the URT and LRT

(A) Representative flow cytometry plots from a single subject infected with RSV showing CCR7 and CD45RA expression by CD4⁺ T cells (top) and CD8⁺ T cells (bottom) in the indicated sites. T cell subsets are identified as; naïve (CCR7⁺/CD45RA⁺), Temra (CCR7⁻/CD45RA⁺), Tcm (CCR7⁺/CD45RA⁻), and Tem (CCR7⁻/CD45RA⁻). (B) Compiled data showing T cell CD4⁺ T cells (left) or CD8⁺ T cells (right) in URT (filled circles, n=22) and LRT (open squares, n=15). For URT CD4⁺ T cell subsets; Tcm median 12.32% (range; 3.31-26%), naïve median 4.02% (range; 0.89-30.71%), Temra median 4.59% (range; 1.13-30.97%) and Tem median 75.77% (range; 43.53-91.05%) and CD8⁺ subsets; Tcm median 2.18% (range; 0.24-9.29%), naïve median 4.34% (range; 0.96-17.32%), Temra median 49.61% (range; 9.33-80.62%) and Tem median 47.32% (range; 13.6-84.84%). For LRT CD4⁺T cell subsets; Tcm median 12.29% (range; 4.61-26.63%), naïve median 1.3% (range; 0.37-15.28%), Temra median 2.91% (range; 0.29-18.77%) and Tem median 81.81% (range; 57.79-92.37%), and CD8⁺ subsets; Tcm median 1.72% (range; 0.57-12.9%), naïve median 3.29% (range; 0.43-15.38%), Temra median 36.66% (range; 6.94-74.77%) and Tem median 56.64% (range; 11.37-91.32%). (C) Correlation of CD4⁺ T cell (left) or CD8⁺ T cell (right) subset frequency in paired URT (filled circles) and LRT (open square) samples from five subjects. (D) Stratification of CD4⁺ T cell (top row) and CD8⁺ T cell (bottom row) subset data obtained from URT samples based on age (<2 months; n =6 and >2 months; n=16). P values obtained by Mann-Whitney t-test and represented by *; (* = p<0.05, ** = p<0.01, ***= p<0.001, ****= p<0.0001).

Figure 5. Trm phenotype cells in respiratory samples and variation with age

(A) Representative flow cytometry plots showing CD69 and CD103 expression by CD4⁺ (top row) and CD8⁺ (bottom row) memory T cells (Tem) in blood and URT samples from infected subjects of indicated ages, including paired blood and URT from the 6 week old subject. (B) CD69 expression by CD4⁺ (left) and CD8⁺ (right) T cells in the URT (closed circles, n=14) and LRT (open squares, n=4) as a function of age. Statistical analysis performed only on URT samples for consistency. (C) CD69/CD103 co-expression on CD4⁺ (left) and CD8⁺ (right) memory T cells in the URT and LRT with age as in (B). R and p values obtained by linear regression.

Figure 6. Increased T-bet expression by Tem cells in early life and locally during VRTI

(A) T-bet expression by CD8⁺ T cell subsets obtained from infant and adult peripheral blood shown as representative flow cytometry plots showing naïve (top), Temra (middle), and Tem (bottom) with the MFI indicated. (B) Relative differences in T-bet expression by CD8⁺ Temra and Tem cells in adult (n=6) and pediatric (n=5) blood samples. Left: T-bet expression (MFI) by Temra and Tem cells with line connecting paired samples from individual subjects. Right: Ratio of T-bet expression on Temra to Tem cells (right) from individual samples shown on left; adult (median 1.18; range 0.89-1.55) and pediatric (median 0.79; range 0.75-1.07). P value obtained by Mann-Whitney (* = p=0.02). (C) Flow cytometry plots depict T-bet (x axis) expression by indicated T cell subsets from URT samples of three subjects with VRTI. Samples are arranged as stated in (A).