Evaluation of an E. coli-expressed spike protein-based in-house ELISA system for assessment of antibody responses after COVID-19 infection and vaccination.

Abstract

Evaluating long-term immunity after COVID-19 infection and vaccination is critical for managing potential outbreaks. The aim of this study was to develop a cost-effective in-house enzyme-linked immunosorbent assay (ELISA) based on Escherichia coli-expressed SARS-CoV-2 spike protein (E-S1) for antibody detection and to evaluate its performance. The system was validated by comparing the in-house ELISA results with those obtained using a commercial ELISA with HEK293-expressed spike protein (H-S1). Recombinant SARS-CoV-2 spike protein was produced in E. coli, purified, and validated for antigenicity via ELISA. Indirect ELISAs with both E-S1 and H-S1 antigens were performed on 386 serum samples from COVID-19 survivors, vaccinated individuals, and pre-pandemic controls collected at different time points. The E-S1 ELISA showed a statistically significant but weak correlation with H-S1 ELISA across all samples (r=0.205; p=0.0001). Stronger correlations were observed among vaccinated individuals with prior infection on day 90 (r=0.6017; p<0.001) and in naïve vaccine recipients on day 30 (r=0.5361; p=0.0003). Pre-pandemic sera from a rural population in Sumba Island exhibited high background reactivity in E-S1 ELISA, likely due to anti-E. coli antibodies, while urban pre-pandemic sera from Jakarta showed a stronger correlation with H-S1 ELISA. This suggests potential regional or immune background differences influencing assay performance. Although E-S1 retained antigenic properties, its diagnostic utility is limited by non-specific reactivity and reduced sensitivity compared to H-S1. In conclusion, E. coli expression systems may not be ideal for producing spike protein-based ELISA antigens specific to SARS-CoV-2. Alternative expression systems, such as human or baculovirus, could enhance diagnostic accuracy and specificity for COVID-19 antibody detection.