Memory T cells established by seasonal human influenza A infection cross-react with avian influenza A (H5N1) in healthy individuals

Abstract

The threat of avian influenza A (H5N1) infection in humans remains a global health concern. Current influenza vaccines stimulate antibody responses against the surface glycoproteins but are ineffective against strains that have undergone significant antigenic variation. An alternative approach is to stimulate pre-existing memory T cells established by seasonal human influenza A infection that could cross-react with H5N1 by targeting highly conserved internal proteins. To determine how common cross-reactive T cells are, we performed a comprehensive ex vivo analysis of cross-reactive CD4+ and CD8+ memory T cell responses to overlapping peptides spanning the full proteome of influenza A/Viet Nam/CL26/2005 (H5N1) and influenza A/New York/232/2004 (H3N2) in healthy individuals from the United Kingdom and Viet Nam. Memory CD4+ and CD8+ T cells isolated from the majority of participants exhibited human influenza-specific responses and showed cross-recognition of at least one H5N1 internal protein. Participant CD4+ and CD8+ T cells recognized multiple synthesized influenza peptides, including peptides from the H5N1 strain. Matrix protein 1 (M1) and nucleoprotein (NP) were the immunodominant targets of cross-recognition. In addition, cross-reactive CD4+ and CD8+ T cells recognized target cells infected with recombinant vaccinia viruses expressing either H5N1 M1 or NP. Thus, vaccine formulas inducing heterosubtypic T cell-mediated immunity may confer broad protection against avian and human influenza A viruses.

Figure 1. Cross-reactive memory T cell responses targeted to the internal proteins of avian influenza A (H5N1) virus in healthy individuals.

All participants (including responders and non-responders) from the UK (A; n = 48) and Viet Nam (B; n = 42) are represented on the x axis. The total magnitudes of ex vivo ELISpot IFN-γ responses to the overlapping peptide pools covering all H5N1 internal proteins are represented on the y axis. Each colored segment represents the source protein corresponding to peptide pools eliciting H5N1 cross-reactive T cell responses. M, matrix protein; PA, polymerase acidic protein.

Figure 2. Ex vivo recognition of the HA and NA of the H3N2 and H5N1 influenza A strains by healthy Vietnamese individuals.

Shown are the magnitudes of ex vivo ELISpot IFN-γ responses to the overlapping peptide pools representing HA (black) and NA (gray) of the H3N2 (A) or H5N1 (B) strains from H5-seronegative healthy Vietnamese participants (n = 20). These participants are shown on the x axis in the same order for A and B.

Figure 3. Relative dominance of cross-reactive memory T cell responses to the internal proteins of avian influenza A (H5N1) virus in healthy individuals.

Gray bars represent the percentage of participants from the UK (A) and Viet Nam (B) who exhibited a detectable IFN-γ T cell response to at least one peptide pool corresponding to the specified source protein on ex vivo IFN-γ ELISpot assay. The average magnitude of IFN-γ T cell responses directed to each protein is represented by black bars. Data are mean ± SD. UK, n = 48; Viet Nam, n = 42.

Figure 4. Distribution of influenza A–specific CD4⁺ and CD8⁺ memory T cell responses across the virus proteome.

The total frequency of CD4⁺ (gray bars) and CD8⁺ (black bars) T cell recognition of positive peptide pools of each viral protein is shown for participants from the UK (A; n = 34) and Viet Nam (B; n = 27). The analysis involved ex vivo IFN-γ responses against the peptide pools of the H3N2 HA and NA and all H5N1 internal proteins. Responders who were unavailable to supply fresh blood samples for the depletion study were excluded from this analysis.

Figure 5. Effector functions displayed by CD4⁺ and CD8⁺ T cell clones upon recognition of target cells infected with the rVACVs expressing H5N1 M1 or NP.

Representative examples. (A and C) ⁵¹Cr release assay. Cross-reactive cytolytic activities against target cells infected with H5N1 NP-VACV (A) or H5N1 M1-VACV (C) displayed by CD8⁺ T cell clones specific for NP 258–273 (donor HUK21) (A) or for M1 58–66 (donor HUK01) (B). E:T, effector to target ratio. (B and D) ICS for effector cytokine secretion or upregulation of degranulation marker by CD8⁺ T cell clones specific for NP 258–273 (donor HUK21) (B) or for M1 58–66 (donor HUK01) (D) in recognition of peptide-pulsed or VACV-infected target cells. (E and F) ICS for effector cytokine secretion by CD4⁺ T cell clones specific for M1 241–252 (donor HUK01) (E) or for M1 95–112 (donor HUK21) (F).

Figure 6. Cross-recognition of target cells infected with rVACVs expressing H5N1 M1 or NP by 7-day polyclonal T cell cultures established with human influenza A viruses.

Representative examples. (A and B) ⁵¹Cr release assay. Cross-reactive cytolytic activities displayed against target cells infected with H5N1 M1-VACV (donor HUK01) (A) or H5N1 NP-VACV (donor HUK36) (B) by the T cell cultures established with A/New Caledonia/20/99 (H1N1). (C) ICS for effector cytokine secretion (IFN-γ, TNF-α, and IL-2) by separately gated CD8⁺ or CD4⁺ population of the T cell culture established with A/New Caledonia/20/99 (H1N1) (donor HUK36). Numbers represent percentages of cells.