key: cord-0882074-dupo77k6
authors: Mutantu, Pierre Nsele; Ngwe Tun, Mya Myat; Nabeshima, Takeshi; Yu, Fuxun; Mukadi, Patrick Kakoni; Tanaka, Takeshi; Tashiro, Masato; Fujita, Ayumi; Kanie, Nobuhiro; Oshiro, Ryosaku; Takazono, Takahiro; Imamura, Yoshifumi; Hirayama, Tatsuro; Moi, Meng Ling; Inoue, Shingo; Izumikawa, Koichi; Yasuda, Jiro; Morita, Kouichi
title: Development and Evaluation of Quantitative Immunoglobulin G Enzyme-Linked Immunosorbent Assay for the Diagnosis of Coronavirus Disease 2019 Using Truncated Recombinant Nucleocapsid Protein as Assay Antigen
date: 2021-09-13
journal: Int J Environ Res Public Health
DOI: 10.3390/ijerph18189630
sha: 6b61937246e414d7d785acec6ff3ccd5d0248836
doc_id: 882074
cord_uid: dupo77k6

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of coronavirus disease 2019 (COVID-19). Real-time RT-PCR is the most commonly used method for COVID-19 diagnosis. However, serological assays are urgently needed as complementary tools to RT-PCR. Hachim et al. 2020 and Burbelo et al. 2020 demonstrated that anti-nucleocapsid(N) SARS-CoV-2 antibodies are higher and appear earlier than the spike antibodies. Additionally, cross-reactive antibodies against N protein are more prevalent than those against spike protein. We developed a less cross-reactive immunoglobulin G (IgG) indirect ELISA by using a truncated recombinant SARS-CoV-2 N protein as assay antigen. A highly conserved region of coronaviruses N protein was deleted and the protein was prepared using an E. coli protein expression system. A total of 177 samples collected from COVID-19 suspected cases and 155 negative control sera collected during the pre-COVID-19 period were applied to evaluate the assay’s performance, with the plaque reduction neutralization test and the commercial SARS-CoV-2 spike protein IgG ELISA as gold standards. The SARS-CoV-2 N truncated protein-based ELISA showed similar sensitivity (91.1% vs. 91.9%) and specificity (93.8% vs. 93.8%) between the PRNT and spike IgG ELISA, as well as also higher specificity compared to the full-length N protein (93.8% vs. 89.9%). Our ELISA can be used for the diagnosis and surveillance of COVID-19.

Coronavirus disease 2019 (COVID- 19) is an emerging viral disease caused by the novel coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The virus was first reported in Wuhan City, China, in late 2019 and rapidly spread in the country and throughout the world [1] . As of 9 March 2020, COVID-19 cases were reported in more than 100 countries [2] . Subsequently, the disease was declared a pandemic by the World Health Organization (WHO) on 11 March 2020 [3] . At the beginning of the COVID-19 outbreak, molecular techniques were first applied to detect the viral genome using conventional reverse transcription-polymerase chain reaction (RT-PCR) and deep sequencing [4] [5] [6] . Currently, real-time RT-PCR is considered the most commonly used technique for diagnosing COVID-19 [7] . Nevertheless, molecular methods rely on the timing of sample collection, adequate sample collection technique, and a sufficient amount of viral RNA from the sample collection site [4] . Reliable serological assays are urgently needed as complementary tools to molecular techniques to enhance the capability of laboratory diagnosis [8, 9] , especially for individuals with viral load undetectable by realtime RT-PCR [10] . Furthermore, the detection of SARS-CoV-2 asymptomatic carriers who have a strong transmission capacity remains critical, especially in resource-limited regions. In such situations, reliable, convenient, and cost-effective diagnosis methods such as serologic tests [11] are highly needed for COVID-19 diagnosis, understanding the immune response to SARS-CoV-2 [12, 13] , and epidemiological surveillance [14] .

SARS-CoV-2 is an enveloped, positive-sense RNA virus with an approximately 30-kilobase genome size that belongs to the Coronaviridae family, in the Betacoronavirus genus [15] . The SARS-CoV-2 genome contains a minimum of six open reading frames (ORFs) [12] . Like other coronaviruses, the SARS-CoV-2 genome encodes four major structural proteins, including spike (S), envelop (E), membrane (M), and nucleocapsid (N) [1] and sixteen non-structural proteins (nsp 1-16) [12] . Among these proteins, N and S are the main antigens used for serological diagnosis of coronaviruses [16] [17] [18] . The structural protein N is smaller than S protein and lacks glycosylation sites; it is involved in assembly by binding to the viral RNA genome [2, 12] . Although its immunological importance is not well known, this protein plays a key function in regulating the viral RNA transcription during the replication stage [19] . The N protein encounters the host immune system after dissociating from the viral genome inside the cell. However, the S protein specifically binds to the receptor of the host cells through its receptor-binding domain (RBD) both before, during, and after the infection has been initiated [19] . These properties make it a good target antigen for the detection of neutralizing antibodies [12] as well as for enzyme-linked immunosorbent assay (ELISA) [14, 20] .

Before the COVID-19 outbreak, six members of the Coronaviridae family were known to infect humans. Two of these species, severe acute respiratory coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV), cause severe respiratory illness, whereas four are responsible for common cold symptoms (human coronavirus [HCoV] HKU1, OC43, NL 63, and 229E) [21] . These human coronaviruses (HCoVs) share 90.5%, 46.1%, 27.6%, 26.5%, 20.0%, and 19.1% amino acid homology with SARS-CoV-2 N protein, respectively [4] .

Most serological tests have focused on the S protein antibody detection method [13] , which may not detect asymptomatic SARS-CoV-2 carriers eight weeks after the infection has occurred [22] . In addition, the performance of S protein might be affected by the mutations resulting from the immune pressure [12] . In contrast, the smaller size of N protein, including the absence of glycosylation sites, makes cloning easier by prokaryotic expression systems [12] . Moreover, antibodies to N protein are generated earlier than S protein [12, 23, 24] . However, N protein might generate false-positive results [25] .

This study aimed to develop a less cross-reactive IgG SARS-CoV-2 N protein-based ELISA, which is faster and cheaper than the current RT-PCR. The cross-reactivity with other human coronaviruses was reduced by deleting the first 121 residues of N protein, which contains a highly conserved motif (FYYLGTGP) of all coronaviruses [26, 27] . The reliability of the assay was evaluated using the plaque reduction neutralization test (PRNT) and the commercial S protein-based IgG indirect ELISA.

Over 600 clinical samples were collected from suspected COVID-19 cases in 2020 in Nagasaki, Nagasaki City, Japan, for virological and serological diagnoses. Nasopharyngeal swab samples were used to detect viral genes by real-time RT-PCR [6] and/or reverse transcription loop-mediated isothermal amplification (RT-LAMP) assay [28] . Of these, 149 were laboratory confirmed as either symptomatic or asymptomatic positives cases, and the remainder were diagnosed as negative cases. Among them, EDTA blood samples were collected one month later in 2020 from 177 people for serological diagnosis. In this group, 139 real-time RT-PCR and/or RT-LAMP positive samples were included. In addition, 100 serum samples collected from febrile illness cases and 55 serum samples collected from healthy volunteers in 2007 were used for the pre-COVID-19 group. All 332 plasma/serum samples were heat-treated at 56 • C for 30 min before applying ELISA and PRNT.

SARS-CoV-2 strain 2019-nCoV/Japan/TY/WK-521/2020 (not registered in the Gen-Bank) that was isolated in Tokyo from migrants from Wuhan in January-February 2020 [29] was inoculated to confluent monolayer Vero E6 cells maintained at 37 • C in Eagle's minimum essential medium (MEM) containing 2% fetal calf serum (FCS) and 0.2 mM nonessential amino acids. Infected culture fluid (ICF) was harvested by centrifugation (2000× g for 10 min) on day 4 when strong cytopathic effects (CPE) appeared (80% to 100%). Viral RNA was extracted from the ICF by using a QIAamp viral RNA mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions and stored at −80 • C until use.

SARS-CoV-2 RNA was used as a template to generate the SARS-CoV-2 N truncated gene, in which the first 121 amino acids (aa) were deleted (N ∆121 ) by real-time RT-PCR using Takara PrimeScript one-step RT-PCR kit (Takara Bio, Kusatsu, Japan).

The SARS-CoV-2/human/CHN/HS_194/2020 MT081068.1 strain was used to design the In-fusion primer sets 5 -TCACCATCACGGATCCCTGCCGTACGGTGCTAA-3 and 5 -TTGGCTGCAGGTCGACTCAAGCCTGGGTAGAGT-3 , which were used to amplify the upstream sense region and the downstream anti-sense region of the SARS-CoV-2 N ∆121 gene, respectively. For the SARS-CoV-2 full-length rN gene construct, the forward primer 5 -TCACCATCACGGATCCATGTCTGATAATGGCCCC-3 with the reverse rN ∆121 primer was used. The expected size of the N ∆121 and full-length N RT-PCR product was confirmed after electrophoresis on the agarose gel. The PCR product was then purified from the agarose gel using a QIAEX II Gel Extraction Kit (Qiagen GmbH, Hilden, Germany). The pQE-30 plasmid DNA was extracted from the Lauria-Bertani medium (LB broth) containing µg/mL ampicillin E. coli culture fluid by using a Qiagen plasmid mini kit (Qiagen GmbH, Hilden, Germany) and digested with BamHI (Takara Bio, Kusatsu, Japan) and SalI (Takara Bio, Kusatsu, Japan) restriction enzymes. Finally, the plasmid backbone was purified from the agarose gel after gel electrophoresis using a QIAEX II Gel Extraction Kit (Qiagen GmbH, Hilden, Germany) and treated with calf intestinal alkaline phosphatase (CIAP) (Takara Bio, Kusatsu, Japan).

The purified SARS-CoV-2 N ∆121 or full-length N gene product was then cloned into the pQE-30 plasmid vector using the In-fusion method (5x In-Fusion HD Enzyme Premix, Takara Bio, Kusatsu, Japan) and then transformed into E. coli XL-1 Blue. The selection of positive clones containing the SARS-CoV-2 N ∆121 or full-length N gene from the E. coli colonies that appeared on the LB agar plate was performed by PCR. The nucleotide sequence of the plasmid DNA extracted from PCR-positive clones was verified by Sanger sequencing to confirm that the sequence was in frame and had no mutations.

The expression of N ∆121 and full-length N proteins were conducted using the E. coli protein expression system. Briefly, a large-scale culture of E. coli (1 L) was performed in a shaker incubator (37 • C, 130 rpm) in LB broth containing 50 µg/mL ampicillin. When the optical density (OD) at 595 nm reached 1.0, expression of rN ∆121 and full-length N protein was induced with 0.1 M isopropyl-β-D-thiogalactopyranoside (IPTG). Three hours after induction, the E. coli culture fluid was centrifuged at 7000× g rpm at 4 • C for 30 min. The pellet was resuspended in denature-lysis-binding buffer containing 8 M urea, 20 mM sodium phosphate, and 30 mM imidazole and sonicated on ice for 5 min (5 s plateau, 5 s interval between pulses). The sonicated mixture was centrifuged at 14,000× g rpm at 4 • C for 30 min, and the supernatant was filtered with a 0.45 µm pore size filter before applying it to the histidine (His) tag affinity column. The expressed protein was double-purified using the His-tag affinity nickel (Ni 2+ ) column (GE Healthcare Biosciences AB, Uppsala, Sweden) under 8 M urea in a denatured condition and His tag affinity cobalt (Co 2+ ) column (GE Healthcare Biosciences AB, Uppsala, Sweden) without urea. Finally, the rN ∆121 and full-length N protein were eluted with 50 mM sodium phosphate, 300 mM sodium chloride, and 150 mM imidazole pH 7.4 elution buffer. The eluted true-peak fractions were pooled and stored at −30 • C before use. In total, 1 to 1.5 mg of purified protein was obtained per 1 L of E. coli culture, and its final concentration was 0.5 mg/mL.

The expected sizes of SARS-CoV-2 rN ∆121 protein and full-length N protein were confirmed by Western blotting after SDS-PAGE. Immunostaining was performed using anti-His IgG (anti-6x His mouse monoclonal antibody [mAb] 4A12EU, Thermo Fisher Scientific, Waltham, MA, USA) and horseradish peroxidase (HRP)-conjugated goat antimouse IgG (American Qualex, CA, USA). The reaction was visualized by diaminobenzidine (DAB) substrate (Dojindo, Kumamoto, Japan). The purity was checked by silver staining (Cosmo Bio Co., Ltd., Tokyo, Japan) according to the manufacturer's instructions.

As described above, 177 plasma samples from the COVID-19 suspected cases and 155 pre-COVID-19 group serum samples were applied to evaluate our in-house IgG indirect ELISA. The optimal concentration of Co 2+ purified SARS-CoV-2 rN ∆121 was determined by checkerboard titration, and 0.13 µg/µL showed low background. Hence, the Co 2+ purified protein was diluted at 0.13 µg/100 µL/well [27] in carbonate buffer pH 9.6, then coated on a 96-well flat-bottom ELISA plate at 4 • C overnight. The same concentration was applied for SARS-CoV-2 full-length N protein-based IgG ELISA. Immulon 1B plate (Thermo Scientific, Rochester, NY, USA) was applied for these ELISAs to reduce the background color reaction. The following day, 100 µL/well of Block Ace (UK-1 B 80, Yukijirushi, Sapporo, Japan) was added to the plate, excluding the blank wells, and the plate was incubated at 37 • C for 1 h for blocking. After washing three times with PBS-Tween 20 (PBS-T), 100 µL/well of 1:200 diluted plasma/serum samples in Block Ace were applied in duplicate. The plate was incubated at 37 • C for 1 h. After washing three times with PBS-T, 100 µL/well of 1:10,000-diluted HRP-conjugated anti-human IgG (American Qualex, CA, USA) in PBS-T with a one-tenth volume of Block Ace was added. The plate was then incubated at 37 • C for 1 h and washed three times with PBS-T. Finally, 100 µL/well of o-phenylenediamine dihydrochloride (OPD) substrate (0.5 mg/mL) (Sigma Chemical, St. Louis, MO, USA) was added. The plate was incubated in the dark at room temperature for 30 min to 1 h before adding 100 µL/well of the stop solution (1N hydrochloric acid). The plate OD was read at 492 nm (Multiscan JX, Thermolab System, Tokyo, Japan). The mean OD of duplicate sample wells was calculated, including the SN (sample/negative control [NC]) ratio (mean sample OD ÷ mean NC OD). The negative control serum was selected among yellow fever vaccinated healthy volunteers collected in 2018. The endpoint IgG titers of the samples were also calculated using the standard curve of the high titer positive control serum. The standard curve was prepared using the 492 nm OD values of the positive control serum starting with a 200-fold dilution and followed by serial two-fold dilutions up to 1:2 11 in PBS-T + 10% Block Ace.

To evaluate the reliability of our in-house rN ∆121 IgG indirect ELISA, a commercial S protein-based IgG ELISA (Cell Signaling Technology, Danvers, MA, USA) was applied to the same samples following the manufacturer's instructions. Briefly, the assay was a solidphase ELISA that detects the binding of human IgG to full-length SARS-CoV-2 S protein.

The mean absorbance at 450 nm of each duplicate sample was calculated. Sample values greater than 4.1 × NC absorbance, less than 3 × NC absorbance, and between 3 × NC and 4.1 × NC absorbance were considered positive, negative, and inconclusive, respectively.

A 50% plaque reduction neutralization test (PRNT 50 ) was performed in biosafety level 3 laboratory (BSL-3) conditions. Briefly, human sera were heat-inactivated at 56 • C for 30 min and two-fold serially diluted (from 1:10 to 1:10,240) in 2% FCS MEM containing 0.2 mM nonessential amino acids. An equal volume of 100 PFU/200 µL/well virus (SARS-CoV-2 virus strain 2019-nCoV/Japan/TY/WK-521/2020) diluted in the same diluent as the serum sample was added with each diluted serum and incubated at 37 • C for 1 h in 5% CO 2 . Subsequently, 200 µL of the mixture was added in duplicate on confluent Vero E6 cells growing in a 24-well plate and incubated at 37 • C for 1 h in 5% CO 2 . Finally, 500 µL of 1.25% methylcellulose overlay medium prepared in 1% FCS with 2 times concentrated MEM containing 0.2 mM nonessential amino acids was added per well and incubated at 37 • C in 5% CO 2 . When CPE appeared on day 5, fixation with 4% paraformaldehyde solution was performed, and the plate was stained with 0.25% crystal violet. The reciprocal of the endpoint serum dilution that provided a 50% or greater reduction in the mean number of plaques relative to the control wells that contained no serum was defined as the PRNT 50 .

Statistical analysis was performed using STATA software version 15.1. The cut-off value of the SN ratio was determined using the receiver operating characteristic (ROC) curve analysis at a 95% confidence interval (CI). The area under the ROC curve (AUC) and the sensitivity and specificity of the IgG indirect ELISA SN ratio were calculated based on the PRNT50 result. The SN ratio with the highest AUC was selected as the optimal cut-off point. Pearson's chi-square test was used to compare the SN ratio to the PRNT50 titer using GraphPad Prism 9.1.1.

In this study, we deleted the first 121 aa of SARS-CoV-2 N protein to reduce crossreactivity with others HCoVs. Both truncated and full-length SARS-CoV-2 N proteins were successfully expressed in the E. coli protein expression system, purified by His-tag affinity chromatography column, and used as assay antigens to detect IgG against SARS-CoV-2. The reliability of SARS-CoV-2 N ∆121 protein-based IgG ELISA was compared with PRNT 50 , which is the gold standard method of coronaviruses serology [30, 31] . Given that SARS-CoV-2 serological methods have focused on S protein [13, 20] , a commercial SARS-CoV-2 full-length S protein-based IgG indirect ELISA was used as the second gold standard. As a monomer and an oligomer, the SARS-CoV-2 N protein has a molecular mass of 51.38 kDa and >670 kDa, respectively [32, 33] . However, the SDS-PAGE silver-stained gel ( Figure 1A ) and the Western blot ( Figure 1B ) revealed a 47 kDa and 33 kDa band sizes of full-length (lane 2) and truncated (lane 3) N proteins, respectively. This might be due to the 3D structure of N proteins affecting the faster migration. Additionally, additional bands were detected by anti-His below the 33 kDa band ( Figure 1A, lane 3) , which might be due to the degradation of rN protein [34] . reactivity with others HCoVs. Both truncated and full-length SARS-CoV-2 N proteins were successfully expressed in the E. coli protein expression system, purified by His-tag affinity chromatography column, and used as assay antigens to detect IgG against SARS-CoV-2. The reliability of SARS-CoV-2 NΔ121 protein-based IgG ELISA was compared with PRNT50, which is the gold standard method of coronaviruses serology [30, 31] . Given that SARS-CoV-2 serological methods have focused on S protein [13, 20] , a commercial SARS-CoV-2 full-length S protein-based IgG indirect ELISA was used as the second gold standard. As a monomer and an oligomer, the SARS-CoV-2 N protein has a molecular mass of 51.38 kDa and >670 kDa, respectively [32, 33] . However, the SDS-PAGE silver-stained gel ( Figure 1A ) and the Western blot ( Figure 1B ) revealed a 47 kDa and 33 kDa band sizes of full-length (lane 2) and truncated (lane 3) N proteins, respectively. This might be due to the 3D structure of N proteins affecting the faster migration. Additionally, additional bands were detected by anti-His below the 33 kDa band ( Figure 1A, lane 3) , which might be due to the degradation of rN protein [34] . 

The purified rNΔ121 protein was used as an assay antigen for the anti-SARS-CoV-2 IgG indirect ELISA at the optimal concentration of 0.13 µg/100 µL/well. An SN ratio of 2.4 showed the highest AUC (0.9244), and was selected as the positive cut-off criterion (Figures 2 and 3) . 

The purified rN ∆121 protein was used as an assay antigen for the anti-SARS-CoV-2 IgG indirect ELISA at the optimal concentration of 0.13 µg/100 µL/well. An SN ratio of 2.4 showed the highest AUC (0.9244), and was selected as the positive cut-off criterion (Figures 2 and 3) . (Table 1) . (Table 1) . 

To compare the relative performance between the SARS-CoV-2 N ∆121 and the fulllength N protein-based IgG ELISA, the same concentration of the assay antigen (0.13 µG/100µL) and SN ratio cut-off criterion (2.4) were applied for SARS-CoV-2 full-length N protein-based IgG ELISA. With reference to the PRNT gold standard, 124 of 177 COVID-19 suspected case samples were IgG positive, while 17 of 155 pre-COVID-19 samples were positive and 138 were negative ( Table 2 ). 

To compare the reactivity between the commercial SARS-CoV-2 S protein-based IgG indirect ELISA and our in-house rN ∆121 IgG indirect ELISA, all 332 samples were also assessed by S IgG ELISA. Among 177 COVID-19 suspected case samples, 121 were IgG positive, and 56 were IgG negative (Table 4 ). By contrast, only 2 of 155 pre-COVID-19 were IgG positive (Table 4) , and 153 were IgG negative. No inconclusive results were found. 

The PRNT 50 titer ≥ 1:10 was applied as the positive cut-off criterion [30] . Of 177 COVID-19 suspected case samples, 124 samples were positive, with a PRNT 50 titer ranging from 1:10 to 1:5120 (mode titer of positive samples, 1:160), and 53 samples were negative (<1:10 PRNT 50 titer) ( Table 5 ). As expected, the PRNT 50 titers of all 155 pre-COVID-19 samples were below 1:10 ( Figure 4 and Table 5 ). 

The performance of our in-house NΔ121 IgG ELISA was evaluated with the PRNT gold standard. Of 118 rNΔ121 IgG ELISA positive samples, 113 were confirmed positive by PRNT50, and 48 of 59 negative rNΔ121 IgG ELISA samples were confirmed negative by PRNT50 (Table 1) . Compared with PRNT50 results, our in-house NΔ121 protein-based IgG ELISA showed 91.1% sensitivity, 93.8% specificity, 89.7% positive predictive value (PPV), 94.7% negative predictive value (NPV), and 92.8% concordance ( Table 1 ). The correlation between the SN ratio of rNΔ121 IgG indirect ELISA and PRNT50 titer was statistically significant (R 2 = 0.4642, 95% CI: 0.6191-0.7350, p < 0.0001) ( Figure 5 ). 

The performance of our in-house N ∆121 IgG ELISA was evaluated with the PRNT gold standard. Of 118 rN ∆121 IgG ELISA positive samples, 113 were confirmed positive by PRNT 50 , and 48 of 59 negative rN ∆121 IgG ELISA samples were confirmed negative by PRNT 50 (Table 1) . Compared with PRNT 50 results, our in-house N ∆121 protein-based IgG ELISA showed 91.1% sensitivity, 93.8% specificity, 89.7% positive predictive value (PPV), 94.7% negative predictive value (NPV), and 92.8% concordance ( Table 1 ). The correlation between the SN ratio of rN ∆121 IgG indirect ELISA and PRNT 50 titer was statistically significant (R 2 = 0.4642, 95% CI: 0.6191-0.7350, p < 0.0001) ( Figure 5 ). PRNT50, and 48 of 59 negative rNΔ121 IgG ELISA samples were confirmed negative by PRNT50 (Table 1) . Compared with PRNT50 results, our in-house NΔ121 protein-based IgG ELISA showed 91.1% sensitivity, 93.8% specificity, 89.7% positive predictive value (PPV), 94.7% negative predictive value (NPV), and 92.8% concordance ( Table 1 ). The correlation between the SN ratio of rNΔ121 IgG indirect ELISA and PRNT50 titer was statistically significant (R 2 = 0.4642, 95% CI: 0.6191-0.7350, p < 0.0001) ( Figure 5 ). (Table 6 ). Of the 155 pre-COVID-19 samples, 153 samples were confirmed negative by S IgG ELISA (Table 6 ). Compared with the commercial S protein-based IgG ELISA, our in-house N ∆121 protein-based IgG ELISA showed 91.9% sensitivity, 93.8% specificity, 89.7% PPV, 95.2% NPV, and 93.1% concordance ( Table 6 ).

As for the rN ∆121 IgG ELISA, we evaluated the performance of the commercial S IgG ELISA with PRNT 50 as the gold standard. Among 121 COVID-19 suspected IgG-positive samples, 119 were confirmed positive by PRNT, and 51 of 56 IgG-negative samples were confirmed negative by PRNT (Table 4 ). In comparison with the PRNT 50 gold standard, S protein-based IgG ELISA showed 96.0% sensitivity, 98.1% specificity, 96.8% PPV, 97.6% NPV, and 97.3% concordance (Table 4 ).

Recombinant nucleocapsid protein has been used for the serodiagnosis of several viruses, such as SARS-CoV [27, 35] , MERS-CoV [36] , severe fever with thrombocytopenia syndrome virus (SFTSV) [37, 38] , Rift Valley fever virus [39] , Nipah virus [40] , and SARS-CoV-2 [41] [42] [43] [44] . Although SARS-CoV-2 immunoassays and point-of-care assays are appearing in the market, comparative performance data are urgently needed for laboratory diagnosis and the public health response to COVID-19 [45] .

In this study, we developed and evaluated an in-house SARS-CoV-2 rN ∆121 proteinbased IgG ELISA. As indicated in the table below, our in-house SARS-CoV-2 N ∆121 proteinbased IgG ELISA showed similar performance to the PRNT and S IgG ELISA (Table 7) . Xiang et al. evaluated the performance of an in-house SARS-CoV-2 full N protein-based IgG indirect ELISA using 85 RT-PCR-confirmed COVID-19 samples and 60 negative control samples [46] . A similar study was conducted by Tehrani et al. with a larger sample size (100 RT-PCR confirmed COVID-19 cases vs. 300 pre-COVID-19 samples) [43] . Although the size of the assay antigen (truncated vs. full-length) and the gold standards are different (PRNT vs. RT-PCR), the sensitivity of our in-house rN ∆121 IgG indirect ELISA appeared much higher than that reported by Xiang et al. (Table 7 ). This could be due to the difference in ELISA sampling time, which was conducted earlier in the Xiang et al. study (13 days after RT-PCR) than in our study (1 month after RT-PCR/RT-LAMP). The differences in sensitivity, specificity, and concordance between Tehrani's study and our results (Table 7) might be due to the disparity in the number of negative samples (300 vs. 155). Compared to SARS-CoV-2 full-length N protein, the rN ∆121 detected a low number of false-positive samples (13 vs. 21) (Tables 1 and 2) . Overall, the truncation improved the specificity of our in-house ELISA. Previous reports demonstrated that antibody levels remain low or undetectable in mild or asymptomatic COVID-19 cases [47, 48] . This study used plasma samples from mild, asymptomatic, and negative cases of COVID-19 collected one month after RT-PCR/RT-LAMP. Our results showed very few samples with high PRNT 50 titers ( Figure 4 and Table 3 ) or SN ratios of rN ∆121 IgG indirect ELISA (Figure 2 ), which could be explained by the weak correlation between the SN ratio and PRNT 50 titer observed in our study ( Figure 5 ).

To date, cross-reactivity with other human coronaviruses remains a significant concern of SARS-CoV-2 serological tests, including ELISA [13, 49] . We deleted the first 121 aa of the N protein, which contain highly conserved N protein regions among coronaviruses [26, 27] . Our in-house IgG rN ∆121 IgG ELISA detected a lower number of false-positive IgG sera than the full-length IgG ELISA (13 vs. 21) (Tables 1 and 2) . These false-positive IgG results could be due to pre-existing cross-reactive antibodies with common cold HCoVs [13] , non-specific binding antibodies [50] , or specific antibodies without neutralizing ability [51] . Our study has one limitation. We could not confirm whether false-positive IgG samples were cross-reactive antibodies to common cold HCoVs.

Population testing plays a key role in controlling the COVID-19 pandemic. We have developed a highly sensitive and specific SARS-CoV-2 N ∆121 protein-based IgG ELISA, which matches at 93% as concordances with two gold standard tests. Our in-house ELISA is cheaper, faster, and simpler than molecular tests. It can be scaled up and made into immunochromatographic kits, representing a valid option for routine diagnosis and surveillance of COVID-19. Informed Consent Statement: Part of this research uses left-over specimens after preceding laboratory tests. The usage of these specimens is in accordance with the Japanese ethical guidelines and Nagasaki University IRB guidelines (IRB 200409233-3) that provide a waiver for prior informed consent for the research use of existing or stored specimens, whereby the information of the research is made public with the opportunity to opt-out from the study.

The data presented in this study are available on request from the corresponding author.

Coronavirus Disease 2019-COVID-19

COVID-19: A New Challenge for Human Beings

SARS-CoV-2 and the Pandemic of COVID-19

Profiling Early Humoral Response to Diagnose Novel Coronavirus Disease (COVID-19)

Genomic Characterisation and Epidemiology of 2019 Novel Coronavirus: Implications for Virus Origins and Receptor Binding

Detection of 2019 Novel Coronavirus (2019-NCoV) by Real-Time RT-PCR. Eurosurveillance 2020

A Comparative Study of Real-Time RT-PCR Based SARS-CoV-2 Detection Methods and Its Application to Human Derived and Surface Swabbed Material

Humoral Responses and Serological Assays in SARS-CoV-2 Infections

Evaluation of Three Commercial SARS-CoV-2 Serologic Assays and Their Performance in Two-Test Algorithms

Temporal Profiles of Viral Load in Posterior Oropharyngeal Saliva Samples and Serum Antibody Responses during Infection by SARS-CoV-2: An Observational Cohort Study

Antibody Seroconversion in Asymptomatic and Symptomatic Patients Infected with Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)

SARS-CoV-2 S1 and N-Based Serological Assays Reveal Rapid Seroconversion and Induction of Specific Antibody Response in COVID-19 Patients

The Main Molecular and Serological Methods for Diagnosing COVID-19: An Overview Based on the Literature

Recent SARS-CoV-2 Seroconversion in a National

In Vivo Structural Characterization of the SARS-CoV-2 RNA Genome Identifies Host Proteins Vulnerable to Repurposed Drugs

Evaluation of Serologic and Antigenic Relationships Between Middle Eastern Respiratory Syndrome Coronavirus and Other Coronaviruses to Develop Vaccine Platforms for the Rapid Response to Emerging Coronaviruses

Evaluation of the EDI Enzyme Linked Immunosorbent Assays for the Detection of SARS-CoV-2 IgM and IgG Antibodies in Human Plasma

The Receptor Binding Domain of the Viral Spike Protein Is an Immunodominant and Highly Specific Target of Antibodies in SARS-CoV-2 Patients

Immunogenicity and Crossreactivity of Antibodies to the Nucleocapsid Protein of SARS-CoV-2: Utility and Limitations in Seroprevalence and Immunity Studies

Laboratory Diagnosis of COVID-19: Current Issues and Challenges

Evolutionary Perspectives on Novel Coronaviruses Identified in Pneumonia Cases in China

Beyond the Spike: Identification of Viral Targets of the Antibody Responses to SARS-CoV-2 in COVID-19 Patients

Detection of Nucleocapsid Antibody to SARS-CoV-2 Is More Sensitive than Antibody to Spike Protein in COVID-19 Patients

Whole Nucleocapsid Protein of Severe Acute Respiratory Syndrome Coronavirus 2 May Cause False-Positive Results in Serological Assays

Characterization of a Novel Coronavirus Associated with Severe Acute Respiratory Syndrome

Evaluation of Inapparent Nosocomial Severe Acute Respiratory Syndrome Coronavirus Infection in Vietnam by Use of Highly Specific Recombinant Truncated Nucleocapsid Protein-Based Enzyme-Linked Immunosorbent Assay

Development and Evaluation of a Rapid and Simple Diagnostic Assay for COVID-19 Based on Loop-Mediated Isothermal Amplification

Performance Evaluation of Real-Time RT-PCR Assays for Detection of Severe Acute Respiratory Syndrome Coronavirus-2 Developed by the National Institute of Infectious Diseases

SARS-CoV-2 Specific Antibody Responses in COVID-19 Patients

A Nanoluciferase SARS-CoV-2 for Rapid Neutralization Testing and Screening of Anti-Infective Drugs for COVID-19

Architecture and Self-Assembly of the SARS-CoV-2 Nucleocapsid Protein

Dimerization of SARS-CoV-2 Nucleocapsid Protein Affects Sensitivity of ELISA Based Diagnostics of COVID-19

Structures of the Nand C-Terminal Domains of a Coronavirus Nucleocapsid Protein: Implications for Nucleocapsid Formation

Characterization of Monoclonal Antibody against SARS Coronavirus Nucleocapsid Antigen and Development of an Antigen Capture ELISA

Sensitive and Specific Detection of Low-Level Antibody Responses in Mild Middle East Respiratory Syndrome Coronavirus Infections

Preparation and Evaluation of Recombinant Severe Fever with Thrombocytopenia Syndrome Virus Nucleocapsid Protein for Detection of Total Antibodies in Human and Animal Sera by Double-Antigen Sandwich Enzyme-Linked Immunosorbent Assay

Application of Recombinant Severe Fever with Thrombocytopenia Syndrome Virus Nucleocapsid Protein for the Detection of SFTSV-Specific Human IgG and IgM Antibodies by Indirect ELISA

Comparison of Enzyme-Linked Immunosorbent Assay Systems Using Rift Valley Fever Virus Nucleocapsid Protein and Inactivated Virus as Antigens

Serodiagnosis Using Recombinant Nipah Virus Nucleocapsid Protein Expressed in Escherichia coli

Kinetics of SARS-CoV-2 Specific IgM and IgG Responses in COVID-19 Patients

Detection of Serum IgM and IgG for COVID-19

Performance of Nucleocapsid and Spikebased SARS-CoV-2 Serologic Assays

Performance Characteristics of Five Immunoassays for SARS-CoV-2: A Head-to-Head Benchmark Comparison

Antibody Detection and Dynamic Characteristics in Patients With Coronavirus Disease

Profile of Immunoglobulin G and IgM Antibodies Against Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)

Serology Characteristics of SARS-CoV-2 Infection after Exposure and Post-Symptom Onset

Development and Characterization of an Indirect ELISA to Detect SARS-CoV-2 Spike Protein-Specific Antibodies

Evaluating SARS-CoV-2 Spike and Nucleocapsid Proteins as Targets for Antibody Detection in Severe and Mild COVID-19 Cases Using a Luminex Bead-Based Assay

SARS-CoV-2 Serologic Assays in Control and Unknown Populations Demonstrate the Necessity of Virus Neutralization Testing

We would like to thank the National Institute of Infectious Diseases (NIID), Japan, for providing the 2019-nCoV/Japan/TY/WK-521/2020 strain. We also thank members of the Virology Department, Institute of Tropical Medicine, Nagasaki University for their technical assistance and helpful discussions, and Masanobu Agoh for his technical assistance. We also acknowledge the Program of Nurturing Global Leaders in Tropical and Emerging Communicable Diseases, Graduate School of Biomedical Sciences, Nagasaki University, for the training and support.

The authors declare no conflict of interest.