Emergent variant modeling of the serological repertoire to norovirus in young children

Abstract

Human norovirus is the leading cause of acute gastroenteritis. Young children and the elderly bear the greatest burden of disease, representing more than 200,000 deaths annually. Infection prevalence peaks at younger than 2 years and is driven by novel GII.4 variants that emerge and spread globally. Using a surrogate neutralization assay, we characterize the evolution of the serological neutralizing antibody (nAb) landscape in young children as they transition between sequential GII.4 pandemic variants. Following upsurge of the replacement variant, antigenic cartography illustrates remodeling of the nAb landscape to the new variant accompanied by improved nAb titer. However, nAb relative avidity remains focused on the preceding variant. These data support immune imprinting as a mechanism of immune evasion and GII.4 virus persistence across a population. Understanding the complexities of immunity to rapidly evolving and co-circulating viral variants, like those of norovirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), and dengue viruses, will fundamentally inform vaccine design for emerging pathogens.

Keywords: antigenic cartography; antigenic seniority; blockade antibodies; immune imprinting; neutralizing antibodies; norovirus; variants of concern.

Graphical abstract

Figure 1

GII.4 human norovirus pandemic variants Shown are serial pandemic GII.4 variants (colored curves) as a percentage of documented norovirus outbreaks over time in England (GOV.UK official statistics: Summary of Norovirus surveillance 2018 to 2019). In this study, sera were collected between 2008 and 2012 (blue shading) from 686 children aged one to six years and tested against a panel of virus-like particles (VLP) representing the time-ordered GII.4 pandemic variants FH 2002, DH 2006, NO 2009, and SY 2012. See also Figure S1.

Figure 2

Serological evidence of GII.4 exposure in young children The proportion of participants with titer (ID50 ≥ 10) to a specific GII.4 variant was determined by age, year, and variant in sera collected from children aged one to six years between 2008 and 2012. Only sera tested against all four variants were included in the analyses (n = 656). The dashed line represents the proportion of responders in all years of sample collection combined. See also Figure S2 and Table S1.

Figure 3

GII.4 nAb by year of sample collection, age of child, and GII.4 variant (A–D) Each marker represents the GMT of the log₁₀-transformed ID50 for children of that age (see also Table S2) for pandemic GII.4 variants FH 2002 (A), DH 2006 (B), NO 2009 (C), and SY 2012 (D). Sera were tested once per VLP with internal and plate controls. Tests were repeated when a control failed. Only sera from children who had an nAb titer (ID50 ≥ 10) to at least one GII.4 variant were included in the analyses (n = 423, 62% total). Log₁₀ GMT are color coded from lowest (teal) to highest (purple).

Figure 4

NO 2009 nAb is most abundant, while the high-avidity DH 2006 nAb persists post NO 2009 emergence (A) The GMT ID50 of GII.4 nAbs in sera from children with ID50 above the limit of detection (≥10) to any GII.4 variant (n = 423) was compiled by year of sample collection and GII.4 variant. (B) GMT relative Ab avidity, as measured by the slope of the Ab neutralization curve of positive titer samples. Sera were tested once per VLP with internal and plate controls. Tests were repeated when a control failed. Error bars, 95% CI. ∗, significantly different from 2008 for each virus, Mann-Whitney test. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. See also Table S2.

Figure 5

The serological nAb repertoire titer is remodeled over time in response to a new dominant variant (A) The antigenic relationship between GII.4 variants (triangles) relative to the nAb response in each child’s serum sample with a titer to at least two GII.4 variants reported (squares, n = 413; Table S3), color coded by child year of age (color of squares: 1, dark red; 2, red; 3, pink; 4, light pink; 5, light blue; 6, dark blue), was mapped via antigenic cartography for 2008 and 2012. One grid box (antigenic distance [AD] unit) corresponds to a two-fold change in titer. As a visual marker, NO 2009 is circled. (B) Summary AD between variants over time, color coded from most (teal) to least (purple) closely related. (C) Mean AD between sera and variants by year. Error bars, SD. ∗p ≤ 0.05, ∗∗ p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001 compared with NO 2009 for each year of serum collection by one-way ANOVA Dunnett multiple comparisons test. (D) Summary AD between variants by age of child at serum collection for all years combined, color coded from most (teal) to least (purple) closely related. See also Figure S3 and Table S3.

Figure 6

nAb relative avidity is shaped by immune imprinting from previous norovirus variants (A) The antigenic relationship between GII.4 variants (triangles) relative to the nAb relative avidity in serum from each child with a measurable titer to at least two GII.4 variants (n = 394; Table S4) (squares), color coded by year of age (color of squares: 1, dark red; 2, red; 3, pink; 4, light pink; 5, light blue; 6, dark blue) was mapped via antigenic cartography for 2008 and 2012. One grid box (AD unit) corresponds to a two-fold change in avidity. As a visual marker, DH 2006 is circled. (B) Relative avidity mean AD between variants and the serum mean in (A) and (B) and Figure S4. Error bars, SD. ∗, different from DH 2006 by one-way ANOVA Dunnett multiple comparisons test. ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. See also Figure S4 and Table S4.