Pandemic influenza A (H1N1) virus causes abortive infection of primary human T cells

Abstract

Influenza A virus still represents a noticeable epidemic risk to international public health at present, despite the extensive use of vaccines and anti-viral drugs. In the fight against pathogens, the immune defence lines consisting of diverse lymphocytes are indispensable for humans. However, the role of virus infection of lymphocytes and subsequent abnormal immune cell death remains to be explored. Different T cell subpopulations have distinct characterizations and functions, and we reveal the high heterogeneity of susceptibility to viral infection and biological responses such as apoptosis in various CD4⁺ T and CD8⁺ T cell subsets through single-cell transcriptome analyses. Effector memory CD8⁺ T cells (CD8⁺ T_EM) that mediate protective memory are identified as the most susceptible subset to pandemic influenza A virus infection among primary human T cells. Non-productive infection is established in CD8⁺ T_EM and naïve CD8⁺ T cells, which indicate the mechanism of intracellular antiviral activities for inhibition of virus replication such as abnormal viral splicing efficiency, incomplete life cycles and up-regulation of interferon-stimulated genes in human T cells. These findings provide insights into understanding lymphopenia and the infectious mechanisms of pandemic influenza A virus and broad immune host-pathogen interactional atlas in primary human T cells.

Keywords: Influenza A virus; abortive infection; primary T cells; single-cell sequencing; virus-host interactions.

Figure 1.

H1N1 infected different subsets of primary human T cells heterogeneously. (A) Distribution of different subsets in infected human CD4⁺ T cells. Left, 0 h.p.i.; right, 16 h.p.i. Blue, naïve CD4⁺ T; green, follicular helper T (Tfh); wine, CD4⁺ cytotoxic T (CTL); rose pink, regulatory T (Treg). (B) Distribution of different subsets in infected human CD8⁺ T cells. Left, 0 h.p.i.; right, 16 h.p.i. Blue, naïve CD8⁺ T; green, stem cell memory T (T_SCM); wine, effector memory T (T_EM); rose pink, terminally differentiated effector T (T_TE); yellow, exhaust T (T_EX). (C) Infection rates among different primary human CD4⁺ and CD8⁺ T subsets. Blue, 0 h.p.i.; red, 16 h.p.i.; green dotted line, 5%. (D) mRNA expression levels of viral HA in naïve CD8⁺ T and CD8⁺ T_EM from one healthy donor at eight time points. Results are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 through Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = non-significant. (E) Immunofluorescent staining of naïve CD8⁺ T and CD8⁺ T_EM which exposure to H1N1 respectively. Green, α-2,6-linked sialic acid receptors; pink, CD45RO; red, viral HA proteins; blue, DAPI (4′,6-diamidino-2-phenylindole).

Figure 2.

Quantitative analysis of α-2,3- and α-2,6-linked sialic acid receptors in MDCK.2 and primary human T cells. (A) Flow cytometry analysis of broad-spectrum neuraminidases treated MDCK.2 cells or not with the lectin antibody concentrations of 5 and 20 μg/mL. Grey, unstained as controls; blue, α-2,6-linked sialic acid receptors; red, α-2,3-linked sialic acid receptors. (B) The comparison of α-2,6- and α-2,3-linked receptors in MDCK.2 cells from ATCC. (C) The comparison of α-2,6- and α-2,3-linked receptors of human CD14⁺ mononuclear/macrophages. (D) The comparison of α-2,6- and α-2,3-linked receptors of human total CD4⁺ T and total CD8⁺ T cells. (E) Purities of gating and MFI of receptors with the lectin antibody concentrations of 5 μg/mL in CD8⁺ T_EM and naïve CD8⁺ T cells and the quantitative analysis of α-2,6- and α-2,3-linked sialic acid receptors between CD8⁺ T_EM and naïve CD8⁺ T cells from five healthy donors. Both CD8⁺ T_EM and naïve CD8⁺ T cells were stained with identical fluorescent antibodies. Red, naïve CD8⁺ T; blue, CD8⁺ T_EM. Results of (B-E) are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 through two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3.

Abortive infection of pandemic H1N1 was established in primary human T cells. (A) Schematic diagram of supernatant culture, transwell culture and co-culture directly between infected CD8⁺ T_EM and untreated MDCK cells. (B) mRNA expression levels of viral HA in samples of MDCK cells among the above three culture methods. Results are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 through two-tailed Student’s t-test. *p < 0.05, ns = non-significant. (C) Numbers of H1N1 viral genes per cell in different infected CD4⁺ T subsets. Blue, naïve CD4⁺ T; green, follicular helper T (Tfh); wine, CD4⁺ cytotoxic T (CTL); rose pink, regulatory T (Treg). (D) Numbers of H1N1 viral genes per cell in different infected CD8⁺ T subsets. Blue, naïve CD8⁺ T; green, stem cell memory T (T_SCM); wine, effector memory T (T_EM); rose pink, terminally differentiated effector T (T_TE); yellow, exhaust T (T_EX).

Figure 4.

Abnormal splicing efficiency and up-regulation of ISGs dedicate to work for abortive infection in primary human T cells. (A) Frequency of different H1N1 viral genes in infected total CD4⁺ T and follicular helper T cells. Blue, 0h.p.i.; red, 16 h.p.i. (B) Frequency of different H1N1 viral genes in infected total CD8⁺ T and CD8⁺ T_EM. Blue, 0h.p.i.; red, 16 h.p.i. (C) Schematic representation of alternative splicing of M1 and NS1 mRNA and their alternatively spliced product M2 and NEP (NS2) mRNA. The arrowheads show corresponding primer positions for detection. (D) Alternative splicing efficiency of M and NS genes in infected CD8⁺ T_EM and A549 cells post-infection. Red, A549 cells; blue, CD8⁺ T_EM. Results are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 by two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = non-significant. (E) mRNA expression levels of cell host splicing factors ns1-bp, sf2 and hnRNP K in CD8⁺ T_EM post-infection. (F) The gene heatmap of ISGs in infected CD8⁺ T_EM. Blue, down-regulation; red, up-regulation. (G) mRNA expression levels of rsad2, adar1, oas1 and oas2 in CD8⁺ T_EM post-infection. Results of (E and G) are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 by two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = non-significant.

Figure 5.

Biological consequences of non-productive infection in infected effector memory CD8⁺ T cells. (A) Left, subsets partition diagram of human CD8⁺ T cells through single-cell sequencing. Right, subpopulation distribution of infected CD8⁺ T cells (red) and bystander cells (grey). (B) Volcano plot of differentially expressed genes in infected CD8⁺ T_EM over time. Red, up-regulation; blue, down-regulation. (C) Apoptosis of CD8⁺ T_EM. Left, cells with annexin V and PI staining on 2 d.p.i. Right, apoptosis ratios of CD8⁺ T_EM after exposed to H1N1 over time. Red represents for the group with treatment of live viruses while blue represents for the control group of UV-treated inactive viruses. Results are represented as mean fold change ± SD and statistical significances were analysed using GraphPad Prism 8.0 through two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (D) The heatmap of ribosomal protein-related genes in infected CD8⁺ T_EM post-infection over time. (E) GO enrichment pathways change analysis of molecular function (green), cellular component (blue) and biological process (red) among infected and non-infected CD8⁺ T_EM. (F) The gene change heatmap of eEFs in infected CD8⁺ T_EM over time. (G) The gene change heatmap of MHC I-immunoproteasomes in infected CD8⁺ T_EM over time. (H) Circos analysis of interactive relationships between membranous receptors and ligands among different CD8⁺ T subsets after exposed to H1N1 over time.