On the caveats of a multiplex test for SARS-CoV-2 to detect seroconversion after infection or vaccination

Abstract

The Covid-19 pandemic, caused by SARS-CoV-2, has resulted in over 6 million reported deaths worldwide being one of the biggest challenges the world faces today. Here we present optimizations of all steps of an enzyme-linked immunosorbent assay (ELISA)-based test to detect IgG, IgA and IgM against the trimeric spike (S) protein, receptor binding domain (RBD), and N terminal domain of the nucleocapsid (N-NTD) protein of SARS-CoV-2. We discuss how to determine specific thresholds for antibody positivity and its limitations according to the antigen used. We applied the assay to a cohort of 126 individuals from Rio de Janeiro, Brazil, consisting of 23 PCR-positive individuals and 103 individuals without a confirmed diagnosis for SARS-CoV-2 infection. To illustrate the differences in serological responses to vaccinal immunization, we applied the test in 18 individuals from our cohort before and after receiving ChAdOx-1 nCoV-19 or CoronaVac vaccines. Taken together, our results show that the test can be customized at different stages depending on its application, enabling the user to analyze different cohorts, saving time, reagents, or samples. It is also a valuable tool for elucidating the immunological consequences of new viral strains and monitoring vaccination coverage and duration of response to different immunization regimens.

Figure 1

Scheme summarizing the assay steps. Step 1 corresponds to an overnight plate coating with each of the three SARS-CoV-2 antigens used in the assay—S, RBD and N-NTD recombinant proteins. In the step 2, the empty spaces on well surface are filled with 3% BSA solution (blocking solution). In the step 3, serum samples are added to allow the specific binding of sera immunoglobulin to the antigen. Step 4 corresponds to the addition of peroxidase-conjugated antibodies that specifically bind to sera IgG, IgA or IgM bound to the plate. Step 5 correspond to the addition of the chromogenic reagent TMB that is reduced as peroxidase reaction proceeds, developing a blue color. In step 6 peroxidase reaction is stopped by HCl addition, which also converts blue TMB in a yellow compound that is quantified spectrophotometrically at 450 nm. Created with BioRender.com.

Figure 2

Sera reactivity to SARS-CoV-2 antigens in different assay conditions. (A–C) Effect of antigen density: SARS-CoV-2 antigens S (A), RBD (B), or N-NTD (C) were serially diluted in PBS and used as antigens to coat ELISA plates in assays in which sera were used at a 1:50 dilution. (D–F) Effect of sera concentration: sera samples were serially diluted in a 1% BSA solution in PBS-T and assay was conducted using plates previously coated with 4 µg/ml of S (D), RBD (E) or N-NTD (F). (G–I) Effect of secondary antibody dilution: reactivity of serum IgG (G), IgA (H) or IgM (I) to S protein was accessed using the respective detection antibody serially diluted in a 1% BSA solution in PBS-T, in a plate previously coated with 4 µg/ml S, incubated with the sera samples at a 1:50 dilution. For all the conditions tested, the plates were incubated with the antigens overnight, at 4 °C, blocked for 1 h with 3% BSA in PBS-T and incubated for 2 h with 3 sera collected before pandemic (grey symbols) or 4 sera from PCR + individuals (black symbols). Sera reactivity was quantified spectrophotometrically after incubation with the respective detection antibody for 1 h followed by the addition of the chromogenic substrate.

Figure 3

Titration of serum antibodies against SARS-CoV-2 S protein. Sera from 10 pre-pandemic (white symbols) and 23 PCR + (grey symbols) individuals (A) or sera from individuals vaccinated with two doses of ChAdOx1 nCoV-19 who were reactive (PCR +) (colored symbols) or not (black symbols) before vaccination (B) were serially diluted in a 1% BSA solution in PBS-T and incubated for 2 h in plates previously coated with 50 μl solution of S protein at 4 μg/ml, overnight, at 4 °C, and blocked for 1 h with 3% BSA in PBS-T. IgG reactivity was quantified spectrophotometrically after incubation with the respective detection antibody for 1 h followed by the addition of the chromogenic substrate. The values for the area under the curves (AUC) obtained for each sample are represented in (C). * p = 0.004, n.s. = non significative.

Figure 4

Setting the incubation time for each step of the protocol. ELISA plates were coated with 4 µg/ml S and the reactivity to sera in PCR + (black symbols) and pre-pandemic sera (grey symbols) was followed at different periods of sera incubation (A–C), detection antibody incubation (D–F) and TMB incubation (G–I) to IgG, IgA and IgM, respectively. The plates were incubated with the antigens overnight, at 4 °C, blocked for 1 h with 3% BSA in PBS-T and incubated with the sera at a 1:50 dilution. Reactivity was quantified spectrophotometrically after incubation with each detection antibody by the addition of the chromogenic substrate.

Figure 5

Cut-off determination in the optimized in-house ELISA. (A) Reactivity of IgG, IgA and IgM in 42 pre-pandemic sera to S, RBD and N-NTD. (B, C, D) Receiver operating characteristics (ROC) analysis of the optimized in-house ELISA for IgG reactivity to S (B), RBD (C) or N-NTD (D). Forty-two pre-pandemic sera (grey symbol) and 23 PCR + sera were diluted 1:50 in a 1% BSA solution in PBS-T and incubated for 2 h in plates previously coated with 50 μl solution of antigens at 4 μg/ml, overnight, at 4 °C, and blocked for 1 h with 3% BSA in PBS-T. For each antigen, IgG, IgA or IgM reactivity was quantified spectrophotometrically after incubation with the respective detection antibody for 1 h followed by the addition of the chromogenic substrate. AUC represents the area under the curve. In the top panels, the dotted diagonal lines represent the theoretical performance of a test with no discriminatory ability, corresponding to an AUC of 0.5. Horizontal and vertical dotted lines correspond to the sensitivity and 1-specificity, respectively, which indicates the optimal cut point value for each ELISA test, as defined by the Youden index (J) (crossed out square). In the distribution graphs (bottom panels), solid and dotted lines correspond, respectively, to PCR-positive and pre-pandemic serum samples, the vertical dotted lines indicate the optimal cut-off for each ELISA test determined by ROC analysis calculated by Youden method (Table 2).

Figure 6

In-house ELISA application in different groups of samples. Reactivity of sera IgG, IgA and IgM against S, RBD and N-NTD. (A) Sera from individuals who tested positive for SARS-CoV-2 infection (PCR + samples) shown in ascending order of days post confirmed PCR; (B) Vaccinated individuals; (C) Fist sera collected in November 2020 from individuals without diagnosis for SARS-CoV-2 infection shown in ascending order of S IgG OD value and (D) second sera collected in April 2021; (E) Comparison between the results of samples analysis shown in (C) and (D). Sera were diluted 1:50 in a 1% BSA solution in PBS-T and incubated for 2 h in plates previously coated with 50 μl solution of antigens at 4 μg/ml, overnight, at 4 °C, and blocked for 1 h with 3% BSA in PBS-T. For each antigen, IgG, IgA or IgM reactivity was quantified spectrophotometrically after incubation with the respective detection antibody for 1 h followed by the addition of the chromogenic substrate.