Variation at Extra-epitopic Amino Acid Residues Influences Suppression of Influenza Virus Replication by M158-66 Epitope-Specific CD8+ T Lymphocytes

Abstract

Influenza virus-specific CD8⁺ T lymphocytes (CTLs) contribute to clearance of influenza virus infections and reduce disease severity. Variation at amino acid residues located in or outside CTL epitopes has been shown to affect viral recognition by virus-specific CTLs. In the present study, we investigated the effect of naturally occurring variation at residues outside the conserved immunodominant and HLA*0201-restricted M1_58-66 epitope, located in the influenza virus M1 protein, on the extent of virus replication in the presence of CTLs specific for the epitope. To this end, we used isogenic viruses with an M1 gene segment derived from either an avian or a human influenza virus, HLA-transgenic human epithelial cells, human T cell clones specific for the M1_58-66 epitope or a control epitope, and a novel, purposely developed in vitro system to coculture influenza virus-infected cells with T cells. We found that the M gene segment of a human influenza A/H3N2 virus afforded the virus the capacity to replicate better in the presence of M1_58-66-specific CTLs than the M gene segment of avian viruses. These findings are in concordance with previously observed differential CTL activation, caused by variation at extra-epitopic residues, and may reflect an immune adaptation strategy of human influenza viruses that allows them to cope with potent CTL immunity to the M1_58-66 epitope in HLA-A*0201-positive individuals, resulting in increased virus replication and shedding and possibly increasing disease severity.IMPORTANCE Influenza viruses are among the leading causes of acute respiratory tract infections. CD8⁺ T lymphocytes display a high degree of cross-reactivity with influenza A viruses of various subtypes and are considered an important correlate of protection. Unraveling viral immune evasion strategies and identifying signs of immune adaptation are important for defining the role of CD8⁺ T lymphocytes in affording protection more accurately. Improving our insight into the interaction between influenza viruses and virus-specific CD8⁺ T lymphocyte immunity may help to advance our understanding of influenza virus epidemiology, aid in risk assessment of potentially pandemic influenza virus strains, and benefit the design of vaccines that induce more broadly protective immunity.

Keywords: CD8+ T lymphocytes; evasion; extra-epitopic amino acid residues; influenza A virus; replication.

FIG 1

Optimizing influenza virus-infected target cell–CD8⁺ T lymphocyte coculture conditions. (A) Expression of HLA-A*0201 and HLA-B*2705 on A549A2 and A549B27 cells, respectively. Unstained cells are shown in gray, and stained cells are represented by a black line. (B) Amino acid sequences of avian IAV A/Vietnam/1194/2004 (H5N1) and human IAV A/Netherlands/018/1994 (H3N2) M1 proteins. The location of the M1_58-66 (GILGFVFTL) epitope is highlighted in gray. Extra-epitopic amino acids are indicated in bold, and bold underlined amino acids are of the previously described human or avian signature (27). (C) Multistep replication curves were generated after inoculation of A549A2 cells with the WSN-M-hH3N2 and WSN-M-aH5N1 viruses at an MOI of 0.01 in H10F medium supplemented with TPCK (filled symbols) or H10F medium supplemented with TPCK, IL-2, IL-7, and IL-15 (open symbols). (D) Survival of CD8⁺ T lymphocytes after 1, 24, and 48 h of culture under various medium conditions. (E) IFN-γ production of CD8⁺ T lymphocytes after activation with M1_58-66-pulsed BLCLs under different medium conditions.

FIG 2

Reduction of WSN-M-hH3N2 or WSN-M-aH5N1 influenza virus replication by M1_58-66-specific CD8⁺ T lymphocyte is HLA dependent. The graphs show percent reductions of the number of spots (top) and percent reductions of the well area covered by spots (bottom) for WSN-M-aH5N1 (black) and WSN-M-hH3N2 (white) virus-infected target cells (A549A2 or A549) in the presence of M1_58-66-specific CD8⁺ T lymphocytes, as measured by ViroSpot assay, relative to the values for replication without T lymphocytes. Experiments represent effects on virus replication measured at 12 h p.i. Bars for experiments I and II represent percent reductions for triplicates (n = 3), and bars for experiment III represent percent reductions of the means for quadruplicates (n = 4).

FIG 3

Reduction of WSN-M-hH3N2 or WSN-M-aH5N1 influenza virus replication by CD8⁺ T lymphocytes is dependent on epitope specificity. The graphs show percent reductions of the number of spots (top) and percent reductions of the well area covered by spots (bottom) for WSN-M-aH5N1 (black) and WSN-M-hH3N2 (white) virus-infected target cells (A549A2) in the presence of M1_58-66-specific CD8⁺ T lymphocytes (M1) or NP_174-184-specific CD8⁺ T lymphocytes (NP) relative to the values for virus replication in the absence of CD8⁺ T lymphocytes. M1/NP*, percentage of inhibition calculated according to the following formula: 100 × [(T_NP − T_M1)/T_NP], where T_NP is the average number of spots or % WAC for wells in the presence of NP_174-184-specific CD8⁺ T lymphocytes and T_M1 is the average number of spots or % WAC for wells in the presence of M1_58-66-specific CD8⁺ T lymphocytes. Experiments represent effects on virus replication measured at 12 h p.i. Bars represent percent reductions of the means for quadruplicates (n = 4).

FIG 4

WSN-M-hH3N2 or WSN-M-aH5N1 virus replication is differentially reduced by M1_58-66- but not NP_174-184-specific CD8⁺ T lymphocytes at 12 h p.i. The graphs show percent reductions of the number of spots (top) and the well area covered by spots (bottom) for WSN-M-aH5N1 (black) and WSN-M-hH3N2 (white) virus-infected target cells (A549A2 and A549B27) in the presence of M1_58-66- and NP_174-184-specific CD8⁺ T lymphocytes, respectively, relative to the values for virus replication in the absence of CD8⁺ T lymphocytes. Experiments represent effects on virus replication measured at 12 h p.i. Bars represent percent reductions of the means for quadruplicates (n = 4).

FIG 5

WSN-M-hH3N2 or WSN-M-aH5N1 virus replication is differentially reduced by M1_58-66- but not NP_174-184-specific CD8⁺ T lymphocytes at 24 h p.i. The graphs show percent reductions of the number of spots (top) and the well area covered by spots (bottom) for WSN-M-aH5N1 (black) and WSN-M-hH3N2 (white) virus-infected target cells (A549A2 and A549B27) in the presence of M1_58-66- and NP_174-184-specific CD8⁺ T lymphocytes, respectively, relative to the values for virus replication in the absence of CD8⁺ T lymphocytes. Experiments represent effects on virus replication measured at 24 h p.i. Bars represent percent reductions of the means for quadruplicates (n = 4).