Validation of the removal of alpha 2,6-sialylated glycoproteins from whole saliva by western blot and mass spectrometric identification of alpha 2,6-sialylated glycoproteins. (A) The alpha 2,6-sialylated glycoproteins in human whole saliva bound to SNA lectin-coated beads were dissociated by heating at 100°C in SDS-PAGE sample buffer. Proteins (50 µg) in whole saliva (lane 1), bead-unbound (lane 2) and bound (lane 3) fractions were subjected to a 12 % SDS-PAGE and stained with Coomassie blue. Arrowheads indicated that a protein above 100 KDa in whole saliva was completely removed by SNA lectin-coated beads. Tryptic digests of a protein band indicated in (A) were subjected to Nano-LC-LTQ MS/MS. Two sequenced peptides [AYIFIDEAHITQALIWLSQR (B) and YATTAYVPSEEINLVVK (C)] are presented and assigned as internal peptides of A2M (P01023) and A2M-like protein (A8K2U0), respectively. The m/z value of each “y” and “b” ion in collision-induced dissociation (CID) spectra was indicated. All sequenced peptides were shown in supplementary Table 1.

Scheme of A2M functions as a “double-edged sword” on the blocking and inactivating influenza virus infection. Activation of human influenza virus HA was mediated by host proteases (P) and attachment to cells was via cell-surface alpha 2,6-sialylated glycoprotein receptors (2,6-sialyated). The 2,6-sialyated residues of A2M obstruct influenza virus (V) infection by competing with alpha 2,6-sialylated glycoprotein receptors on the host, and the bait region (BR) of A2M traps and inactivates the host proteases that disrupt the viral infectivity. The binding of A2M to HA may influence the recognition of HA by anti-HA antibody. Both activities of A2M (binding of alpha 2,6-sialylated glycoprotein receptors and inactivation of host proteases) may impact the efficacy of antiviral vaccines.

Blockage of the inhibitory effect of human whole saliva on SD/H1N1-S-OIV-induced erythrocyte hemagglutination after removal of alpha 2,6-sialylated glycoproteins. (A) A representative result from six independent hemagglutination activity inhibition assays was shown. The hemagglutination of 50 µl of 0.5% guinea pig erythrocytes in PBS occurred at room temperature for 40 min (lane 1; PBS). The hemagglutination activity inhibition assay was performed by mixing 50 µl of 0.5% guinea pig erythrocytes in PBS with 25 µl of each dilution (as indicated) of SD/H1N1-S-OIV alone (32 HAU), (lane 2; H1N1-S-OIV) or with 25 µl of each dilution of human whole saliva (1 µg/µl) (lane 3; H1N1-S-OIV+Saliva). The alpha 2,6-sialylated glycoproteins in human whole saliva were removed by using SNA lectin-coated beads. After removal, the bead-unbound fraction of saliva (Unbound Saliva) was used for hemagglutination activity inhibition assay. The 50 µl of 0.5% guinea pig erythrocytes in PBS with 25 µl of two-fold serial dilution (1–128) of SD/H1N1-S-OIV (32 HAU) in the presence of 25 µl of two-fold serial dilution (1–128) of unbound saliva (1 µg/µl) (lane 4; H1N1-S-OIV/Unbound Saliva). The 50 µl of 0.5% guinea pig erythrocytes in PBS mixed with 50 µl of each dilution of human whole saliva in the absence of SD-H1N1-S-OIV served as (lane 5; Saliva) an additional control. (B) The inhibition of hemagglutination activity was expressed as log₂ of HAUs and calculated from six independent experiments. ***, P < 0.001 by Student’s t-test.

Restitution of the diminished anti-hemagglutination activity of bead-unbound fraction of saliva by adding purified A2M and examination of the potency of A2M in reduction of hemagglutination and viral infectivity. (A) The hemagglutination activity inhibition assay was performed as described in Fig. 1. The guinea pig erythrocytes in PBS (lane 1; PBS) were mixed with SD/H1N1-S-OIV (32 HAU) (lane 2; H1N1-S-OIV) in the presence of human whole saliva (lane 3; H1N1-S-OIV+Saliva), unbound saliva (lane 4; H1N1-S-OIV/Unbound Saliva), unbound saliva plus purified A2M (75 µg) (lane 5; H1N1-S-OIV/Unbound Saliva/A2M), purified A2M (75 µg) alone (lane 6; H1N1-S-OIV/A2M) or BSA (75 µg) (lane 7; H1N1-S-OIV/BSA). A representative of three separate experiments with similar results was shown. (B) To test the effect of A2M on viral infectivity, a 1/100 dilution of SD/H1N1-S-OIV (32 HAU) plus 75 µg of A2M or BSA was added into confluent MDCK cells in a 96-well plate for 1 h before adding virus growth medium. After overnight culture, the cell viability was determined by an acid phosphatase (ACP) assay as described in Materials and methods. Data represent the mean ± standard error (SE) of three independent experiments. *, P < 0.05 by Student’s t-test.