Development of an Affordable ELISA Targeting the SARS-CoV-2 Nucleocapsid and Its Application to Samples from the Ongoing COVID-19 Epidemic in Ghana

Abstract

Introduction: The true nature of the population spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in populations is often not fully known as most cases, particularly in Africa, are asymptomatic. Finding the true magnitude of SARS-CoV-2 spread is crucial to provide actionable data about the epidemiological progress of the disease for researchers and policymakers. This study developed and optimized an antibody enzyme-linked immunosorbent assay (ELISA) using recombinant nucleocapsid antigen expressed in-house using a simple bacterial expression system.

Methods: Nucleocapsid protein from SARS-CoV-2 was expressed and purified from Escherichia coli. Plasma samples used for the assay development were obtained from Ghanaian SARS-CoV-2 seropositive individuals during the pandemic, while seronegative controls were plasma samples collected from blood donors before the coronavirus disease 2019 (COVID-19) pandemic. Another set of seronegative controls was collected during the COVID-19 pandemic. Antibody detection and levels within the samples were validated using commercial kits and Luminex. Analyses were performed using GraphPad Prism, and the sensitivity, specificity and background cut-off were calculated.

Results and discussion: This low-cost ELISA (£0.96/test) assay has a high prediction of 98.9%, and sensitivity and specificity of 97% and 99%, respectively. The assay was subsequently used to screen plasma from SARS-CoV-2 RT-PCR-positive Ghanaians. The assay showed no significant difference in nucleocapsid antibody levels between symptomatic and asymptomatic, with an increase of the levels over time. This is in line with our previous publication.

Conclusion: This study developed a low-cost and transferable assay that enables highly sensitive and specific detection of human anti-SARS-CoV-2 IgG antibodies. This assay can be modified to include additional antigens and used for continuous monitoring of sero-exposure to SARS-CoV-2 in West Africa.

Fig. 1

Purification of nucleocapsid protein from remnant SUMO-tag post ULP-1 cleavage. A A Coomassie-stained gel visualizing the nucleocapsid protein after 30 kDa filter unit-based purification (purity > 70%). B A western blot image that shows the recognition of the expressed, cleaved and purified SARS-CoV-2 nucleocapsid protein by a human anti-SARS-CoV-2 nucleocapsid mAb (ab273168, Abcam). M marker (PagerRuler^TM Plus), CN cleaved and purified nucleocapsid, BSA bovine serum albumin

Fig. 2

Optimization of the coating antigen, human mAb to SARS-CoV-2 nucleocapsid, and goat pAB to human IgG (HRP) using A commercial primary antibody and B pooled plasma. The OD450 of each concentration is shown by the line graph with the 25th and 75th percentiles. The orange arrows indicate the chosen suitable dilutions during optimizations

Fig. 3

Establishing cut-off values and evaluating the ELISA. A A scatter plot showing the distribution of the negative samples (n = 142) relative to the cut-off value (mean + 3 × standard deviation), shown as a dashed line. B A scatter plot showing the distribution of both negative (n = 142) and positive (n = 100) samples along the cut-off value represented by a dashed line. C The ROC curve of the ELISA at a 95% confidence interval of 0.9781–0.9997

Fig. 4

The nucleocapsid antibody concentration levels of patients with COVID-19. A Scatter plots with a bar showing comparisons of antibody concentration levels between symptomatic (n = 295) and asymptomatic patients (n = 75) individuals. B Change in antibody concentration levels over time. The antibody concentration levels were analysed from symptomatic (n = 43) and asymptomatic (n = 10) patients at different timepoints for 4 weeks. The median quantity of the antibody levels for each sampling day is shown by the line graph and the 25th and 75th percentiles are antibody levels. Statistical significance between groups was determined by a two-tailed Mann-Whitney U test (ns: p > 0.05, *p < 0.05, **p < 0.01, ****p < 0.0001)