Duck CD8+ T Cell Response to H5N1 Highly Pathogenic Avian Influenza Virus Infection In Vivo and In Vitro

Abstract

Domestic ducks are the important host for H5N1 highly pathogenic avian influenza virus (HPAIV) infection and epidemiology, but little is known about the duck T cell response to H5N1 AIV infection. In infection experiments of mallard ducks, we detected significantly increased CD8⁺ cells and augmented expression of cytotoxicity-associated genes, including granzyme A and IFN-γ, in PBMCs from 5 to 9 d postinfection when the virus shedding was clearly decreased, which suggested the importance of the duck cytotoxic T cell response in eliminating H5N1 infection in vivo. Intriguingly, we found that a CD8^high+ population of PBMCs was clearly upregulated in infected ducks from 7 to 9 d postinfection compared with uninfected ducks. Next, we used Smart-Seq2 technology to investigate the heterogeneity and transcriptional differences of the duck CD8⁺ cells. Thus, CD8^high+ cells were likely to be more responsive to H5N1 AIV infection, based on the high level of expression of genes involved in T cell responses, activation, and proliferation, including MALT1, ITK, LCK, CD3E, CD247, CFLAR, IL-18R1, and IL-18RAP. More importantly, we have also successfully cultured H5N1 AIV-specific duck T cells in vitro, to our knowledge, for the first time, and demonstrated that the CD8^high+ population was increased with the duck T cell activation and response in vitro, which was consistent with results in vivo. Thus, the duck CD8^high+ cells represent a potentially effective immune response to H5N1 AIV infection in vivo and in vitro. These findings provide novel insights and direction for developing effective H5N1 AIV vaccines.

FIGURE 1.

Monitor of duck survival, H5N1 AIV shedding, HI Ab level, and T lymphocyte percentage postinfection. (A) Survival rate of the mallard ducks (Sheldrake) infected with A/Duck/Guangdong/383/2008 (DK383) virus with the dose of 10⁶ EID₅₀. Both the infected group and the control group consisted of 15 ducks. Curves are significantly different (p < 0.001) by log-rank and Gehan–Breslow–Wilcoxon analyses. (B) Virus (H5N1 AIV) shedding was monitored via detecting the viral load in oropharyngeal and cloacal swabs. Statistical analyses for virus titer in swabs at various time points were performed using one-way ANOVA. (C) HI Ab level was monitored using 1% chicken RBCs. The value >4 (dotted line) was considered HI Ab-positive. One-way ANOVA was used for statistical comparisons. (D and E) The percentage of CD4⁺ T cells (D) or CD8⁺ T cells (E) between the control group and infection group at various time points was detected. Cells (2 × 10⁵) from each sample were collected for flow cytometric analysis. A two-way ANOVA was used for statistical comparison. H5N1 virus shedding and H5N1 HI Ab in the control group were all negative at various time points (data not shown). Four ducks of infected and control groups were randomly selected for sampling and detection. *p < 0.05, **p < 0.01, ***p < 0.001. ns, not significant.

FIGURE 2.

Analysis of immune-related gene expression in duck PBMCs by qRT-PCR. The total RNA of PBMCs was extracted from three ducks of the infected and control groups, respectively. The data were collected from three biological samples in each group; each sample was performed in triplicate. (A) Cytotoxicity-associated genes in PBMCs. (B) Innate immune genes in PBMCs. (C) Th2 markers in PBMCs. (D) Inflammatory cytokine IL-6 expression. (E) Cell survival–related genes in PBMCs. The results are presented as means ± SEM, and the paired t test was used for statistical comparison. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

Analysis of the duck CD8⁺ cell phenotype. (A) Gating strategy for duck CD8⁺ cells. (B) Analysis of the percentage and phenotype of CD8⁺ cells in PBMCs of H5N1-infected duck no. 16 and control duck no. 3 at various time points. (C) Contour plot and histogram of CD8^low+ and CD8^high+ cell populations in PBMCs of three ducks in both the H5N1-infected group and control group at 9 dpi. Cells (2 × 10⁵) from each sample was collected for flow cytometric analysis.

FIGURE 4.

WGCNA analysis of duck CD4, CD8, CD8^low+, and CD8^high+ cell populations in PBMCs at 9 dpi. The CD4⁺ T or CD8⁺ T cells in the PBMC pool were collected from three control and three H5N1 AIV-infected ducks, respectively, at 9 dpi (control_CD4, H5_CD4, control_CD8, and H5_CD8), and CD8^low+ and CD8^high+ cell populations in the PBMC pool from three H5N1 AIV-infected ducks at 9 dpi (H5_CD8^low and H5_CD8^high) were sorted out for Smart-Seq2. (A) PCA of the six cell populations, each with five replications. The x-axis and y-axis show the principal component (PC)1 and PC2, which explains 42.1 and 31.1% of the total variance. (B) Heatmap of sample expression pattern. Red indicates high-level expression. Green indicates low-level expression. (C) Heatmap of the gene expression pattern of the salmon module. (D) Top 20 KEGG pathways selected for salmon module gene enrichment. (E) Heatmap of the gene expression pattern of the purple module. (F) Top 20 KEGG pathways selected for purple module gene enrichment.

FIGURE 5.

Differential gene expression analysis of CD4 or CD8 T cells between the control group and H5N1-infected group. (A) Statistics of differentially expressed genes (DEGs) of CD4 or CD8 T cells after H5N1 AIV infection. (B) Top 20 KEGG pathways selected for DEG enrichment of CD4 T cells. (C) Top 20 KEGG pathways selected for DEG enrichment of CD8 T cells. (D) Heatmap of selected immune-related DEGs from CD8 T cells.

FIGURE 6.

Trend analysis of CD8^low+, CD8⁺, and CD8^high+ cell populations of H5N1-infected group at 9 dpi. (A) All DEG expression profiles ordered based on the p values in the order of CD8^low+ cells, CD8⁺ cells (containing CD8^low+ and CD8^high+ cells), and CD8^high+ cells. (B) The DEG expression trends in profile 0. (C) Top 10 KEGG pathways were selected for profile 0 after DEG enrichment. (D) The DEG expression trends in profile 7. (E) Top 10 KEGG pathways were selected for profile 7 after DEG enrichment. (F) Heatmap of selected immune genes. (G) Gene interaction network analysis based on the WGCNA analysis and the STRING database. The gray line indicates the interaction relationship based on the WGCNA analysis. The green line indicates the potential interaction based on the STRING database. Red nodes represent hub genes.

FIGURE 7.

In vitro culture of H5N1 AIV-specific duck T cells. (A) Screening of optimal H5N1 AIV infection time points with the maximum living cells. The data were collected from three biological samples. The results are presented as means ± SEM, and the unpaired t test was used for statistical comparison. (B) The percentage of NP-expressing cells with various infection dosages. The data were collected from three biological samples. Statistical analysis was performed using one-way ANOVA. (C) Flow cytometric analysis of the NP-positive rate between the infection group and control group. (D) Morphologic observation of memory PBMCs with or without H5N1 AIV stimulation (DK383 strain). Scale bar, 100 μm. Two independent experiments with two duck memory PBMC donors were performed. (E) The proliferation of CFSE-labeled memory PBMCs was detected by CFSE dilution in H5N1-stimulated cells from H5N1-infected ducks after 2 wk of culturing. Red sample indicates the CFSE-labeled memory PBMCs without stimulation. Yellow, green, and black samples represent CFSE-labeled memory PBMCs with H5N1 stimulation after 7, 8, and 14 d of culturing, respectively. (F) The CD4 or CD8 T cell percentage between H5N1-stimulated and unstimulated cells after 14 d of culturing. The unpaired t test was used for statistical comparison. (G) The CD4 or CD8 T cell numbers between H5N1-stimulated and unstimulated cells after 14 d of culturing. T cell percentage or number data were collected from two replicates in two independent experiments. Statistical analysis was performed by an unpaired t test. *p < 0.05, **p < 0.01. ns, not significant.

FIGURE 8.

H5N1 AIV-specific duck T cell phenotype and response in vitro. (A) Gating and analysis of the percentage and phenotype of CD4⁺ T cells between H5N1-stimulated and unstimulated cells after 14 d of culturing. (B) Gating and analysis of the percentage and phenotype of CD8⁺ T cells between H5N1-stimulated and unstimulated cells after 14 d of culturing. (C) Detection of the H5N1 AIV-specific duck T cell response via qRT-PCR. The data were collected from three replicates in the H5N1-stimulated and unstimulated groups, respectively. The results are presented as means ± SEM, and the paired t test was used for statistical comparison. *p < 0.05, **p < 0.01.