key: cord-0999940-f15t1wx1
authors: Fujigaki, Hidetsugu; Inaba, Masato; Osawa, Michiko; Moriyama, Saya; Takahashi, Yoshimasa; Suzuki, Tadaki; Yamase, Kenya; Yoshida, Yukihiro; Yagura, Yo; Oyamada, Takayoshi; Takemura, Masao; Doi, Yohei; Saito, Kuniaki
title: Comparative analysis of antigen-specific anti-SARS-CoV-2 antibody isotypes in COVID-19 patients
date: 2020-12-04
journal: bioRxiv
DOI: 10.1101/2020.12.04.407510
sha: cfaf7560002de004c2f89445abe12276ff90534e
doc_id: 999940
cord_uid: f15t1wx1

Serological tests for detection of anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies in blood are expected to identify individuals who have acquired immunity against SARS-CoV-2 and indication of seroprevalence of SARS-CoV-2 infection. Many serological tests have been developed to detect antibodies against SARS-CoV-2. However, these tests have considerable variations in their specificity and sensitivity, and whether they can predict levels of neutralizing activity is yet to be determined. This study aimed to investigate the kinetics and neutralizing activity of various antigen-specific antibody isotypes against SARS-CoV-2 in serum of coronavirus disease 2019 (COVID-19) patients confirmed via polymerase chain reaction test. We developed IgG, IgM and IgA measurement assays for each antigen, including receptor-binding domain (RBD) of spike (S) protein, S1 domain, full length S protein, S trimer and nucleocapsid (N) domain, based on enzyme-linked immunosorbent assay. The assays of the S protein for all isotypes showed high specificity, while the assays for all isotypes against N protein showed lower specificity. The sensitivity of all antigen-specific antibody isotypes depended on the timing of the serum collection and all of them, except for IgM against N protein, reached more than 90% at 15-21 days post-symptom onset. The best correlation with virus neutralizing activity was found for IgG against RBD (RBD-IgG), and levels of RBD-IgG in sera from four severe COVID-19 patients increased concordantly with neutralizing activity. Our results provide valuable information regarding the selection of serological test for seroprevalence and vaccine evaluation studies.

Samples 103 This study was reviewed and approved by the Ethics Committee for Clinical 104 Research of the Center for Research Promotion and Support in Fujita Health University 105 (authorisation number HM19-493 and HM17-341). 106 We utilized a series of residual serum samples from 41 COVID-19 patients who 107 were admitted to Fujita Health University Hospital from February 28 to May 21, 2020. The 108 demographic and clinical characteristics of the patients are presented in Table 1 One hundred serum samples obtained from 100 healthy human volunteers (mean 118 age, 47; males, 58 and females, 42), collected before the COVID-19 pandemic (July 2012), 119 were used as negative controls to evaluate the specificity and cut-off values for each assay. 120 All serum samples (aliquoted and stored at −80°C) were thawed and evaluated at the same SARS-CoV-2 RBD, S1, S full, S trimer and N protein expressed in HEK293 cells 125 were selected as the target antigens. 96-well plates ( Statistical difference between non-severe and severe COVID-19 patients was determined 160 using two-tailed Mann-Whitney test and p-value less than 0.05 was considered statistically 161 significant.

Antigen-specific antibody isotype responses in COVID-19 patients 165 We evaluated the ELISA designed to detect SARS-CoV-2 antigen-specific IgG, IgM 166 and IgA against RBD, S1, S full, S trimer and N protein. To quantify the antibody responses 167 to each antigen, we tested 169 serums from 41 SARS-CoV-2-infected patients and 100 168 negative control serums from healthy donor collected before SARS-CoV-2 pandemic. We To evaluate the specificity and sensitivity of the developed ELISAs, the optimal cut-off values for each antigen-specific antibody isotype were determined by the Youden's index 186 using the ROC analysis of all samples ( Table 2 ). The specificity of the S protein (RBD, S1, 187 S full and S Trimer) assays for all isotypes showed comparable results, with high specificity presents the Spearman correlation coefficient (r) of each antigen-specific antibody isotypes. 208 Overall, IgG showed good correlation with neutralizing activity in all the antigens. 209 Conversely, IgA showed a relatively lower correlation with neutralizing activity and N-IgA 210 in particular showed no significant correlation.

Relationship between RBD-IgG production and disease severity 213 We categorized the COVID-19 patients into two severity groups (severe and non- reported that S1 or RBD showed better specificity than N-based serological assay and that 261 there was significant cross-reactivity when N protein was used as the antigen for their assay showed lower sensitivity than S1-based assays (23). Our findings support the use of S protein 264 as the antigen for the detection of SARS-CoV-2 specific antibodies. 265 We also determined which antigen-specific antibody isotype assays best represented 266 the virus neutralizing activity (Fig. 3A and B) . Although most of the antibody isotypes IgG levels in convalescent patients can be used to identify appropriate donors with high 286 neutralizing activity for convalescent serum/plasma therapy (7, 10).

There are several limitations in this study. Importantly, there were only four severe 288 COVID-19 patients from whom residual serum samples were available to investigate the 289 association between serum RBD-IgG levels and neutralizing activity. Since these four severe 290 COVID-19 patients all recovered from COVID-19, we could not investigate differences in 291 antibody responses and neutralizing activity between patients who recovered from COVID-292 19 and those who did not.

In summary, our results indicate that the anti-SARS-CoV-2 antibody response in 294 COVID-19 patients varies among the antigen-specific antibody isotypes. Diagnostic performance of ELISAs for detection of anti-SARS-CoV-2 antibodies also varies among 296 antigen-specific antibody isotypes. Among them, serum RBD-IgG levels best correlate with 297 virus neutralizing activity and disease severity, thus may be the optimal assay to track 298 COVID-19 seroconversion responses and use as the basis for COVID-19 serological tests. 

A Novel Coronavirus from Patients with 318 Pneumonia in China

Serology assays to manage COVID-19

Laboratory 322 Diagnosis of COVID-19: Current Issues and Challenges

Will antibody tests for the coronavirus really change everything? 324

Antibody responses to SARS-CoV-2 in patients with COVID-434 19

Serological differentiation between COVID-19 and SARS infections

Serological 442 assays for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)

Convergent antibody responses 459 to SARS-CoV-2 in convalescent individuals

2020. Phase I/II 464 study of COVID-19 RNA vaccine BNT162b1 in adults