Serological Array-in-Well Multiplex Assay Reveals a High Rate of Respiratory Virus Infections and Reinfections in Young Children

Abstract

Serological assays are used to diagnose and characterize host immune responses against microbial pathogens. Microarray technologies facilitate high-throughput immunoassays of antibody detection against multiple pathogens simultaneously. To improve survey of influenza A virus (IAV), influenza B virus (IBV), respiratory syncytial virus (RSV), and adenovirus (AdV) antibody levels, we developed a microarray consisting of IAV H1N1, IAV H1N1pdm09 (vaccine), IAV H3N2, IBV Victoria, IBV Yamagata, RSV, AdV type 5 hexon protein, and control antigens printed on the bottom of a microtiter plate well. Bound IgG antibodies were detected with anti-human IgG-coated photon-upconverting nanoparticles and measured with a photoluminescence imager. The performance of the microarray immunoassay (MAIA) was evaluated with serum samples (n = 576) collected from children (n = 288) at 1 and 2 years of age and tested by standard enzyme immunoassays (EIAs) for antibodies to IAV vaccine and RSV. EIAs and MAIA showed substantial to almost perfect agreement (Cohen's κ, 0.62 to 0.83). Applying MAIA, we found seroprevalences of 55% for IAV H1N1, 54% for IAV vaccine, 30% for IAV H3N2, 24% for IBV Victoria, 25% for IBV Yamagata, 38% for RSV, and 26% for AdV in 1-year-old children (n = 768). By the age of 2 years, IgG seropositivity rates (n = 714) increased to 74% for IAV H1N1, 71% for IAV vaccine, 49% for IAV H3N2, 47% for IBV Yamagata, 49% for IBV Victoria, 68% for RSV, and 58% for AdV. By analyzing increases in antibody levels not biased by vaccinations, we found a reinfection rate of 40% for RSV and 31% for AdV in children between 1 and 2 years of age.IMPORTANCE The multiplex immunoassay was successfully used to simultaneously detect antibodies against seven different viruses. The developed serological microarray is a new promising tool for diagnostic, epidemiological, and seroprevalence analyses of virus infections.

Keywords: adenoviruses; immunoassays; influenza; influenza vaccines; microarray; multiplex; respiratory syncytial virus; upconversion luminescence.

FIG 1

Multiplex array-in-well assay principle, array layout, and fluorescent array-in-well images. (A) Principle of the assay. (B) Array layout. The antigens (ags; IAV H1N1 vaccine, IAV H1N1, IAV H3N2, IBV Yamagata, IBV Victoria, RSV, and AdV5 hexon protein), HSA (human serum albumin; negative control), hIgG (positive control for the detector), and anti-hIgG (positive control for the sample) were printed on the bottom of the streptavidin-coated 96-well microtiter plate. Binding of serum IgG antibodies to the antigens and positive controls was detected by using anti-hIgG coated photon-upconverting nanoparticles (UCNPs). (C) Fluorescent image of an example of the 96-well plate. (D) Fluorescent array-in-well images. (a) An array well without serum sample (reagent background); (b) negative-control serum; (c) positive-control serum; (d) anti-RSV- and anti-AdV IgG-positive sample. Fluorescence colors are related to signal intensity: green corresponds to signal counts of 7,000 to 35,000, yellow to 35,000 to 45,000, and red to 45,000 to 65,000. (E) Average signal counts from no-sample well, negative control, positive control, and a serum sample positive for IgG antibodies against RSV and AdV. Average signal counts were calculated as a mean signal from 2 duplicate spots and 2 duplicate wells (from 4 spots altogether). The error bars indicate the SD of the upconversion luminescence signals.

FIG 2

Correlation of the multiplexed MAIA with the reference IgG EIAs (n = 288). (A and B) Anti-RSV IgG unit values for serum samples from 1-year-old children (A) and 2-year-old children (B). (C and D) Anti-IAV H1N1 vaccine IgG unit values for serum samples from 1-year-old children (C) and 2-year-old children (D). All 4 correlations were statistically significant (P < 0.001, two-tailed test).

FIG 3

Seropositivity rate comparisons by EIA and MAIA in 1- and 2-year-old children (n = 288). (A) RSV; (B) IAV H1N1 vaccine ag.

FIG 4

Detection of IgG antibodies reactive with RSV, IAV H1N1 vaccine ag, IAV H1N1, IAV H3N2, IBV Victoria, IBV Yamagata, and AdV5 hexon in 1-year (n = 768) and 2-year (n = 714) serum samples by the microarray assay. The solid horizontal lines show the mean unit values of seropositive serum specimens; the dotted horizontal lines indicate the cutoff values for each virus antigen. The values are shown as microarray units (MAIA), calculated in relation to the values for negative-control specimens (MAIA unit value set as 0) and a highly positive serum specimen (MAIA unit value set as 100).

FIG 5

Seropositivity rates for antibodies against all tested antigens (RSV, AdV5 hexon, IAV H1N1 vaccine ag, IAV H1N1, IAV H3N2, IBV Yamagata, and IBV Victoria) in serum specimens collected from 1-year-old (n = 768) and 2-year-old (n = 714) children by the microarray assay.

FIG 6

Follow-up of mean IgG antibody levels for RSV, IAV H1N1 vaccine ag, IAV H1N1, IAV H3N2, IBV Victoria, IBV Yamagata, and AdV5 hexon protein in children who were seropositive at 1 year (among 450 individuals for whom both 1-year and 2-year samples were available). The boxes represent the interquartile range (25 to 75%), the whiskers indicate 1st and 99th percentiles, the horizontal lines inside boxes show a median, the squares inside boxes show a mean, and the triangles show minimum and maximum. n, number of serum specimens.

FIG 7

Levels of RSV-, AdV5 hexon-, and IBV Victoria-reactive IgG in children who were IgG seropositive in the corresponding assay at 1 year (n = 170, n = 118, and n = 108, respectively). Children were separated into two groups based on their likely reinfection. (A, C, and D) Mean IgG levels in children with >25 MAIA units increased at the age of 2 years compared to the 1-year sample (likely reinfection between 1 and 2 years). (B, D, and F) Mean IgG levels in children showing increases of ≤25 MAIA units during follow-up. The boxes indicate the interquartile range (25 to 75%), the whiskers 1st and 99th percentiles, the horizontal lines inside the boxes the median, the squares inside boxes the mean, and the triangles the minimum and the maximum.