Human Sera Collected between 1979 and 2010 Possess Blocking-Antibody Titers to Pandemic GII.4 Noroviruses Isolated over Three Decades

Abstract

The emergence of pandemic GII.4 norovirus (NoV) strains has been proposed to occur due to changes in receptor usage and thereby to lead to immune evasion. To address this hypothesis, we measured the ability of human sera collected between 1979 and 2010 to block glycan binding of four pandemic GII.4 noroviruses isolated in the last 4 decades. In total, 268 sera were investigated for 50% blocking titer (BT₅₀) values of virus-like particles (VLPs) against pig gastric mucin (PGM) using 4 VLPs that represent different GII.4 norovirus variants identified between 1987 and 2012. Pre- and postpandemic sera (sera collected before and after isolation of the reference NoV strain) efficiently prevented binding of VLP strains MD145 (1987), Grimsby (1995), and Houston (2002), but not the Sydney (2012) strain, to PGM. No statistically significant difference in virus-blocking titers was observed between pre- and postpandemic sera. Moreover, paired sera showed that blocking titers of ≥160 were maintained over a 6-year period against MD145, Grimsby, and Houston VLPs. Significantly higher serum blocking titers (geometric mean titer [GMT], 1,704) were found among IgA-deficient individuals than among healthy blood donors (GMT, 90.9) (P < 0.0001). The observation that prepandemic sera possess robust blocking capacity for viruses identified decades later suggests a common attachment factor, at least until 2002. Our results indicate that serum IgG possesses antibody-blocking capacity and that blocking titers can be maintained for at least 6 years against 3 decades of pandemic GII.4 NoV.IMPORTANCE Human noroviruses (NoVs) are the major cause of acute gastroenteritis worldwide. Histo-blood group antigens (HBGAs) in saliva and gut recognize NoV and are the proposed ligands that facilitate infection. Polymorphisms in HBGA genes, and in particular a nonsense mutation in FUT2 (G428A), result in resistance to global dominating GII.4 NoV. The emergence of new pandemic GII.4 strains occurs at intervals of several years and is proposed to be attributable to epochal evolution, including amino acid changes and immune evasion. However, it remains unclear whether exposure to a previous pandemic strain stimulates immunity to a pandemic strain identified decades later. We found that prepandemic sera possess robust virus-blocking capacity against viruses identified several decades later. We also show that serum lacking IgA antibodies is sufficient to block NoV VLP binding to HBGAs. This is essential, considering that 1 in every 600 Caucasian children is IgA deficient.

Keywords: VLP; norovirus; pandemic.

FIG 1

Serum samples and study design. In total, 268 serum samples collected between 1979 and 2010 were included in the study. Four different GII.4 recombinant VLPs, each representative of a different lineage and having a time-ordered emergence globally, were investigated: MD145, isolated in 1987 (AY032605); Grimsby, isolated in 1995 (AJ004864); Houston, isolated in 2002 (EU310927); and Sydney, isolated in 2012 (JX459908). The years shown in parentheses are the years of isolation of the reference GII.4 norovirus strains. The phylogenetic tree, based on nucleotide sequences of GII.4 NoV strains generated using the neighbor-joining method, shows that VLPs of GII.4 NoVs used in this study belong to different lineages. The numbers under the branches represent bootstrap values from 1,000 replicates.

FIG 2

Blocking titers in sera collected in 1998 from healthy secretor-positive individuals against 3 different NoV VLPs. The graph shows BT₅₀s for each individual and GMTs for each group. The horizontal lines represent the geometric means, with 95% confidence intervals, for the groups. In total, 76 serum samples were analyzed. Dunn's multiple-comparison test showed significant differences between the groups. *, P < 0.05; **, P < 0.01.

FIG 3

Blocking-antibody titers to MD145, Grimsby, and Houston VLPs among paired serum samples. Serum BT₅₀ values against MD145 (A), Grimsby (B), and Houston (C) VLPs among 33 tick-borne encephalitis-vaccinated individuals. From each individual, three samples were collected between 1989 and 1992, 1992 and 1996, and 1995 and 1999. The graph shows BT₅₀s for each individual and GMTs for each group. The horizontal lines represent the geometric means, with 95% confidence intervals, for the groups. A Kruskal-Wallis test showed no significant differences in blocking titers against any VLPs among sera collected at 3 different time points.

FIG 4

Prepandemic sera from 1979 contain blocking antibodies against pandemic GII.4 viruses. The graph shows BT₅₀s for each individual (n = 10) and GMTs for each group. The horizontal lines represent the geometric means, with 95% confidence intervals, for the groups. A Mann-Whitney U test was used to compare differences in blocking titers between MD145 and Houston VLPs (P > 0.05).

FIG 5

IgA-deficient secretor-positive individuals have significantly higher blocking antibody titers than IgA-competent secretor-positive individuals. The graph shows BT₅₀s for each individual and geometric mean titers for each group. The horizontal lines represent the geometric means, with 95% confidence intervals, for the groups. A Mann-Whitney U test was used to compare differences in blocking titers. ***; P < 0.0001.