Siglec-1 inhibits RSV-induced interferon gamma production by adult T cells in contrast to newborn T cells

Abstract

Interferon gamma (IFN-γ) plays an important role in the antiviral immune response during respiratory syncytial virus (RSV) infections. Monocytes and T cells are recruited to the site of RSV infection, but it is unclear whether cell-cell interactions between monocytes and T cells regulate IFN-γ production. In this study, micro-array data identified the upregulation of sialic acid-binding immunoglobulin-type lectin 1 (Siglec-1) in human RSV-infected infants. In vitro, RSV increased expression of Siglec-1 on healthy newborn and adult monocytes. RSV-induced Siglec-1 on monocytes inhibited IFN-γ production by adult CD4⁺ T cells. In contrast, IFN-γ production by RSV in newborns was not affected by Siglec-1. The ligand for Siglec-1, CD43, is highly expressed on adult CD4⁺ T cells compared to newborns. Our data show that Siglec-1 reduces IFN-γ release by adult T cells possibly by binding to the highly expressed CD43. The Siglec-1-dependent inhibition of IFN-γ in adults and the low expression of CD43 on newborn T cells provides a better understanding of the immune response against RSV in early life and adulthood.

Keywords: CD43; Interferon gamma; Newborns; Respiratory syncytial virus; Siglec-1.

Figure 1

Induction of Siglec‐1 expression during RSV infection. (A) Transcriptome analysis of unstimulated adult PBMCs (N = 5) (x‐axis) compared to RSV‐stimulated adult PBMCs (N = 5) stimulated with RSV for 24 h (y‐axis). Plots in the upper left corner represent genes that are upregulated during In vitro RSV infection of adult PBMCs. A difference of ± 4‐fold difference between expression levels of unstimulated compared to expression levels of RSV‐stimulated PBMCs was used to determine up‐ and downregulation of genes. Data shown are pooled from two or more independent experiments with two or three samples per experiment (B) Transcriptome analysis of unstimulated adult PBMCs (x‐axis) compared to RSV‐stimulated adult PBMCs (y‐axis) combined with upregulated genes in whole blood from RSV‐infected infants in Table 1 (white dots). (C) Correlation of Siglec‐1 gene expression (y‐axis) and RSV titers in the nasopharyngeal aspirate (x‐axis) during infant RSV infection. (D) Comparison of Siglec‐1 gene expression level in healthy infants (Healthy; N = 14), RSV‐infected infants (Acute; N = 40) and infants 4–6 weeks after RSV infection (Recovery; N = 30). Data points depicted represent individual samples from two or more independent experiments with ten to fifteen samples per experiment Data are represented as median ± interquartile range. Statistical analysis employed Kruskal‐Wallis test and if significant followed by Mann–Whitney U test. Spearman correlation test was used for correlation tests. Dashed lines indicate a 4‐fold difference of expression level between x‐axis and y‐axis. **p<0.01. ***p<0.001.

Figure 2

RSV induces Siglec‐1 on monocytes. (A): Representative gating strategy to determine Siglec‐1 expression on monocytes (CD14⁺ cells) after stimulation of MCs with RSV. (B‐C): Percentage of Siglec‐1⁺ monocytes after stimulation of (B) MCs or (C) adult monocytes with RSV N = 5‐7. (D): Mean fluorescence intensity of Siglec‐1 on CD14⁻ and CD14⁺ cells after stimulation of MCs with RSV (N = 5). Data represent means ± SEM. Statistical analyses employed the Wilcoxon matched‐pairs signed rank test. *p<0.05. **p<0.01. Data are pooled from two or more experiments with two to four samples per experiment.

Figure 3

RSV‐induced IFN‐γ is delayed in newborns and dependent on CD4⁺ T cells. (A‐B): Production of IFN‐γ (measured by ELISA performed in duplo) by CBMCs and PBMCs after stimulation with RSV for (A) 48 h or (B) 5 days. N = 5. Data are pooled from two or more experiments with two to four samples per experiment. (C): Representative flow cytometry results of MCs after depletion of CD4⁺, CD8⁺ or CD56⁺ cells. (D‐E: RSV‐induced IFN‐γ release, measured by ELISA performed in duplo, by (D) CBMCs, CD4‐depleted CBMCs, CD8‐depleted CBMCs or CD56‐depleted CBMCs after 5 days and (E) RSV‐induced IFN‐γ release by PBMCs, CD4‐depleted PBMCs, CD8‐depleted PBMCs or CD56‐depleted PBMCs after 48 h. N = 5. Data are pooled from two or more experiments with two to four samples per experiment. Data represent means ± SEM. Statistical analysis employed Wilcoxon matched‐pairs signed rank test for two conditions and repeated measures ANOVA with Bonferroni's Multiple Comparison Test for more than two conditions. Comparative testing between newborns and adults was performed with Mann–Whitney U test. *p<0.05.

Figure 4

RSV‐induced Siglec‐1 inhibits IFN‐γ release by adult T cells but not by T cells from newborns. A–B: (A) Percentage of RSV⁺ monocytes and (B) GFP fluorescence (MFI) of monocytes after stimulation of adult MCs with RSV in the absence (No blocking) or presence of monoclonal antibodies against Siglec‐1 (Anti‐Siglec‐1) or in the presence of matching isotype controls (Isotype). Data were obtained by flow cytometry. N = 5. Data are pooled from two or more experiments with two to four samples per experiment. (C and D): RSV‐induced IFN‐γ release by (C) newborn CBMCs after 5 days or (D) adult PBMCs after 48h in the absence (No blocking) or presence of monoclonal antibodies against Siglec‐1 (anti‐Siglec‐1) or in the presence of matching isotype controls (Isotype). N = 4–6. Data are pooled from two or more experiments with two to four samples per experiment performed in duplo. (E:) Percentage of RSV‐induced IFN‐γ by PBMCs or PBMCs depleted from memory CD4⁺ T cells. RSV‐induced IFN‐γ by PBMCS was set to 100. N = 5. Data are pooled from two or more experiments experiments with two to four samples per experiment performed in duplo. (F): RSV‐induced IFN‐γ release by adult PBMCs depleted from memory CD4⁺ T cells after 48h in the absence (No blocking) or presence of monoclonal antibodies against Siglec‐1 (anti‐Siglec‐1) or in the presence of matching isotype controls (Isotype). N = 5. Data are pooled from two or more experiments experiments with two to four samples per experiment performed in duplo. Data represent means ± SEM. Statistical analyses employed the Wilcoxon matched‐pairs signed rank test for paired analysis between two conditions and repeated measures ANOVA with Bonferroni's Multiple Comparison Test for paired analysis between more than two conditions. *p<0.05. ***p<0.001.

Figure 5

Low CD43 expression on newborn CD4⁺ T cells compared to adults. (A) Representative gating strategy to measure CD43 expression on naïve (CD3⁺CD4⁺CD45RA⁺CD27⁺), central memory (CD3⁺CD4⁺CD45RA⁻CD27⁺), effector memory (CD3⁺CD4⁺CD45RA⁻CD27⁻) and terminally differentiated (CD3⁺CD4⁺CD45RA⁺CD27⁻) CD4⁺ T cells. (B) Expression of CD43 on newborn and adult CD4⁺ T cells. (C) CD43 expression on naïve and central memory CD4⁺ T cells in newborns and adults. N = 5–6. D: Expression of CD43 on adult naïve, central memory, effector memory and terminally differentiated CD4⁺ T cells. Data are pooled from two or more experiments with two to four samples per experiment. Data represent means ± SEM. Statistical analysis employed Repeated measures ANOVA with Bonferroni's Multiple Comparison Test for paired analysis between more than two conditions. Comparative testing between newborns and adults was performed with Mann–Whitney U test. *p: <0.05. **p<0.01; ***p<0.001.