Heterogeneous antibodies against SARS-CoV-2 spike receptor binding domain and nucleocapsid with implications for COVID-19 immunity
Kathleen M. McAndrews, … , James P. Allison, Raghu Kalluri
Kathleen M. McAndrews, … , James P. Allison, Raghu Kalluri
Published September 17, 2020; First published August 14, 2020
Citation Information: JCI Insight. 2020;5(18):e142386. https://doi.org/10.1172/jci.insight.142386.
View: Text | PDF
Research Article COVID-19 Immunology

Heterogeneous antibodies against SARS-CoV-2 spike receptor binding domain and nucleocapsid with implications for COVID-19 immunity

Abstract

Evaluation of potential immunity against the novel severe acute respiratory syndrome (SARS) coronavirus that emerged in 2019 (SARS-CoV-2) is essential for health, as well as social and economic recovery. Generation of antibody response to SARS-CoV-2 (seroconversion) may inform on acquired immunity from prior exposure, and antibodies against the SARS-CoV-2 spike protein receptor binding domain (S-RBD) are speculated to neutralize virus infection. Some serology assays rely solely on SARS-CoV-2 nucleocapsid protein (N-protein) as the antibody detection antigen; however, whether such immune responses correlate with S-RBD response and COVID-19 immunity remains unknown. Here, we generated a quantitative serological ELISA using recombinant S-RBD and N-protein for the detection of circulating antibodies in 138 serial serum samples from 30 reverse transcription PCR–confirmed, SARS-CoV-2–hospitalized patients, as well as 464 healthy and non–COVID-19 serum samples that were collected between June 2017 and June 2020. Quantitative detection of IgG antibodies against the 2 different viral proteins showed a moderate correlation. Antibodies against N-protein were detected at a rate of 3.6% in healthy and non–COVID-19 sera collected during the pandemic in 2020, whereas 1.9% of these sera were positive for S-RBD. Approximately 86% of individuals positive for S-RBD–binding antibodies exhibited neutralizing capacity, but only 74% of N-protein–positive individuals exhibited neutralizing capacity. Collectively, our studies show that detection of N-protein–binding antibodies does not always correlate with presence of S-RBD–neutralizing antibodies and caution against the extensive use of N-protein–based serology testing for determination of potential COVID-19 immunity.

Authors

Kathleen M. McAndrews, Dara P. Dowlatshahi, Jianli Dai, Lisa M. Becker, Janine Hensel, Laura M. Snowden, Jennifer M. Leveille, Michael R. Brunner, Kylie W. Holden, Nikolas S. Hopkins, Alexandria M. Harris, Jerusha Kumpati, Michael A. Whitt, J. Jack Lee, Luis L. Ostrosky-Zeichner, Ramesha Papanna, Valerie S. LeBleu, James P. Allison, Raghu Kalluri

×

Figure 1

Detection of serum binding antibodies against SARS-CoV-2 proteins in patients with PCR-confirmed COVID-19 and healthy samples.

Options: View larger image (or click on image) Download as PowerPoint
Detection of serum binding antibodies against SARS-CoV-2 proteins in pat...
(A) Timeline of COVID-19 diagnosis/ICU admittance, serum sample collection, and convalescent plasma (CP) administration. Time 0 is defined as day of COVID-19 diagnosis (PCR positive for SARS-CoV-2) and ICU admittance. Blood collections are denoted in gray and CP administration is denoted in pink. Patients were stratified based on current status (recovered, hospitalized, or deceased). Patient 29 from our cohort had symptoms but was PCR negative for SARS-CoV-2; this sample was not included in figures since there was no proof of disease. (B) Schematic of SARS-CoV-2 viral structure (top panel) and antigens assayed (bottom panel). S-protein, light orange; envelope protein, yellow; membrane glycoprotein, dark orange; RNA, blue; N-protein, green. Absorbance normalized to the respective no antigen control for each sample at 450 nm plotted for S-RBD (left panel), and N-protein (right panel), antigen coating with the most recent (or only) SARS-CoV-2 samples not treated with CP (n = 21) and healthy samples collected in 2017–2019 (HS 2017–2019, n = 104 for S-RBD, n = 103 for N-protein) and 2020 (HS 2020, n = 308). Data are presented with each dot representing the mean normalized absorbance for a given serum sample; the red bar depicts the median ± interquartile range of all samples. HS, healthy sample; NC (line), negative control cutoff (see Methods). Kruskal-Wallis with Dunn’s multiple-comparisons test performed. ****P < 0.0001; ns, not significant.