key: cord-0917789-8h8oeh5u
authors: Kim, Hye-Yeon; Lee, Jong-Hwan; Kim, Mi Jeong; Park, Sun Cheol; Choi, Minsuk; Lee, Wonbin; Ku, Keun Bon; Kim, Bum Tae; Changkyun Park, Edmond; Kim, Hong Gi; Kim, Seung Il
title: Development of a SARS-CoV-2-specific biosensor for antigen detection using scFv-Fc fusion proteins
date: 2020-11-30
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2020.112868
sha: 1c7eb80be248ff2cfb424d472349f0779ebbda06
doc_id: 917789
cord_uid: 8h8oeh5u

Coronavirus disease 2019 (COVID-19) is a newly emerged human infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). In a global pandemic, development of a cheap, rapid, accurate, and easy-to-use diagnostic test is necessary if we are to mount an immediate response to this emerging threat. Here, we report the development of a specific lateral flow immunoassay (LFIA)-based biosensor for COVID-19. We used phage display technology to generate four SARS-CoV-2 nucleocapsid protein (NP)-specific single-chain variable fragment-crystallizable fragment (scFv-Fc) fusion antibodies. The scFv-Fc antibodies bind specifically and with high affinity to the SARS-CoV-2 NP antigen, but not to NPs of other coronaviruses. Using these scFv-Fc antibodies, we screened three diagnostic antibody pairs for use on a cellulose nanobead (CNB)-based LFIA platform. The detection limits of the best scFv-Fc antibody pair, 12H1 as the capture probe and 12H8 as the CNB-conjugated detection probe, were 2 ng antigen protein and 2.5 × 10(4) pfu cultured virus. This LFIA platform detected only SARS-CoV-2 NP, not NPs from MERS-CoV, SARS-CoV, or influenza H1N1. Thus, we have successfully developed a SARS-CoV-2 NP-specific rapid diagnostic test, which is expected to be a simple and rapid diagnostic test for COVID-19.

Coronavirus disease 2019 , which was first reported in Wuhan city, Hubei province, 40 China, in December 2019 (Wang et al. 2020; WHO 2020c) , is caused by the novel virus acute 41 respiratory syndrome coronavirus 2 (SARS-CoV-2) (Wu et al. 2020; Zhou et al. 2020; Zhu et al. 42 2020). Infection with SARS-CoV-2 primarily causes pneumonia-like symptoms, including fever, 43 cough, and fatigue (Le Bert et al. 2020; Long et al. 2020) . Human-to-human transmission is rapid; 44 therefore, the World Health Organization declared the COVID-19 outbreak a global pandemic (WHO 45 2020a). As of August 2020, more than 24,000,000 cases of COVID-19 have been confirmed world-46 wide, and 838,924 people have died (WHO 2020b) . No COVID-19-specific drugs or vaccines are 47 available at present; therefore, accurate diagnosis is crucial if we are to manage the COVID-19 48 pandemic. Rapid screening and isolation of COVID-19 patients prevent super-spreading events and 49 enables patients to receive treatment at the early stage of the illness. Here, we used phage display technology (Ledsgaard et al. 2018 ) to generate single-chain 78 variable fragment (scFv)-crystallizable fragment (Fc) fusion proteins (scFv-Fcs) for use as antibodies 79 for specific detection of SARS-CoV-2 nucleocapsid protein (NP). The interaction between SARS-80 CoV-2 NP and scFv-Fc antibodies was examined by western blotting, enzyme-linked immunosorbent 81 assay (ELISA), and biolayer interferometry (BLI) to measure affinity (K D ). Extensive LFIA screening 82 identified three SARS-CoV-2-specific diagnostic antibody pairs. The LFIA-based biosensors based on 83 Expression of recombinant NP was induced by adding 0.5 mM isopropyl β -D-1-93 thiogalactopyronoside to the bacterial culture. Bacterial cells were lysed by sonication, and soluble 94 NP was purified using Ni 2+ affinity chromatography (HisTrap FF; GE Healthcare, IL, USA) and a 95 size-exclusion column (HiLoad 16/60 Superdex 200 PG, GE Healthcare). The purity and homogeneity 96 of the recombinant protein were purified by sodium dodecyl sulfate-polyacrylamide gel 97 electrophoresis (SDS-PAGE). Finally, the protein was concentrated to 2 mg/ml in phosphate-buffered 98 saline (PBS) using a 30 kDa molecular-weight cutoff centrifugal filter (Millipore, MA, USA). 101 The scFv clones were isolated from a chicken naïve phage library (YntoAb, South Korea) by 102 biopanning using purified recombinant SARS-CoV-2 NP. Monoclonal binders were selected using an 103 ELISA based on SARS-CoV-2 NP (40588-V08B; Sino Biological, Inc., China); clones that were 104 cross-reactive for SARS-CoV NP (40143-V08B, Sino Biological, Inc.) and MERS-CoV NP (40068-105 V08B, Sino Biological, Inc.) were eliminated. SARS-CoV-2-specific scFv binders were inserted to 106 scFv-Fc vectors and expressed in 293F cells (YntoAb). Monoclonal scFv-Fc antibodies were purified 107 by protein A affinity chromatography. After dialysis with PBS, the antibody concentration was 108 measured using a NanoDrop 2000 spectrophotometer (ThermoFisher Scientific, MA, USA). 120 For Western blotting, SARS-CoV-2 NP was separated by SDS-PAGE and transferred onto a 121 polyvinylidene fluoride (PVDF) membrane. For the dot blot, SARS-CoV-2 NP was spotted directly 122 onto a nitrocellulose (NC) membrane, which was then dried for 1 h at 37°C. The membrane was 123 blocked for 1 h at room temperature with 5% skim milk in TBST, followed by incubation with 124 purified scFv-Fc antibodies (1:1000 dilution) for 2 h at 37°C. Afterward, membranes were washed 125 with TBST and incubated for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated 126 goat anti-human IgG (1:5000 dilution, ThermoFisher Scientific). Next, the membranes were washed 127 in TBST, and blotted proteins were visualized using ECL western blot substrate reagents (Millipore).

Finally, the chemiluminescent signals were analyzed using a ChemiDoc MP imaging system (Bio-Rad, 129 CA, USA). 

The binding kinetics of the scFv-Fc antibodies and SARS-CoV-2 NP antigen were analyzed by BLI 133 using an Octet QK384 instrument (ForteBio, CA, USA). First, Ni-NTA biosensors were hydrated for 134 at least 10 min prior to measurement. Next, SARS-CoV-2 NP antigen (10 µg/ml) and four different 135 antibodies (31.3, 15.6, 7.8, or 3.9 nM) were prepared in 1×PBS and 0.09% Tween 20. The association 136 and dissociation steps were adjusted to 900 and 1200 sec, respectively. After each step, the biosensor 137 tip was equilibrated in 1×PBS and 0.09% Tween 20 for 60 sec. K D values were calculated by ForteBio 138 data analysis software using a 1:1 binding model. 141 Protein A and scFv-Fc antibodies were conjugated to Nano Act TM CNBs (Asahi Kasei, Japan) using a 142 CNB conjugation kit (DCN Diagnostics, CA, USA). Briefly, protein A and antibodies (0.5 mg/ml in 143 120 μl conjugation buffer) were mixed with CNB (0.5% CNB in 120 μl conjugation buffer) and conjugants were washed with 7.5 ml wash buffer by centrifugation at 14,400 g for 20 min at 4°C.

Finally, the pellets were suspended gently in 300 μl wash buffer. The concentration of the protein A 148 and antibody-CNB conjugants was measured using UV-vis spectrophotometry (BioTek Instruments, 149 Inc., VT, USA) at an absorption intensity of 554 nm. 

To detect SARS-CoV-2 using LFIA, each scFv-Fc antibody (1 mg/ml) was dispensed onto 168 the test line, and the anti-human IgG antibodies (50 ng/ml in sample diluent buffer) were applied to 169 the control line. The conjugate pad was treated with each scFv-Fc antibody-CNB conjugate (final 170 concentration of CNB, 0.5%) in stabilizing buffer and dried for 1 h at 37°C in a vacuum drying oven.

applied to the sample pad of the assembled LFIA test kit. After 20 min, the density of the test and 173 control lines was measured using the LFIA reader. 185 To develop the SARS-CoV-2 NP-specific LFIA-based biosensor, we first used phage display 186 technology. The phage library contains a considerable number (> 10 12 ) of phages displaying different 187 single-chain variable fragments (scFv). Through three rounds of biopanning and ELISA screening, the 188 specific clones for the SARS-CoV-2 NP were primarily selected. To generate scFv-Fc, the sequences 189 of selected clones were confirmed, and scFv-Fc fusion proteins were generated based on this sequence 190 information. Finally, SARS-CoV-2 NP-specific scFv-Fc fusion proteins (12H1, 12H8, 12B3, and 1G5) mounted onto the portable LFIA reader, the image obtained from the LFIA is analyzed automatically 200 using LED and CMOS sensors, and a quantitative result is obtained within 10 seconds (Scheme 1b). 203 The NP was selected as the target antigen for diagnosis. The SARS-CoV-2 NP is an RNA-binding 204 protein that forms a helical ribonucleoprotein necessary for viral RNA transcription and replication. It 

For antibody development, we used phage display technology. Generally, the phage disaply 209 technology is faster than traditional hybridoma technology to discover antibodies and identify 210 sequences of paratope, the antigen binding site (Winter et al. 1994) . To screen scFv binders that 211 interact specifically with SARS-CoV-2 NP, we performed phage display screening using a chicken 212 naïve scFv antibody library (Fig. 1a) . After three rounds of biopanning, we isolated 157 positive 213 clones with strong positive binding signals. Next, to isolate SARS-CoV-2 NP-specific scFv binders, 214 non-specific scFv binders showing high background signals and cross-reactivity with NPs from 215 MERS-CoV and SARS-CoV were eliminated. After removing non-specific scFv binders, 22 clones 216 specific for SARS-CoV-2 NP were isolated (Fig. 1b) , and four unique clones (12H1, 12H8, 12H3, and 217 1G5) with different complementary-determining region sequences of heavy and light chains were 218 identified by DNA sequencing (Fig. S2) . To generate specific antibodies for diagnosis, the four scFv 219 binders were cloned into a scFv-Fc plasmid, and scFv-Fc antibodies were expressed and purified (Fig.   220 S1b). The scFv-Fc format allows for characterization of scFvs before conversion into a full-length 221 IgG and also can be used itself as antibody. 224 The virus that causes COVID-19, SARS-CoV-2, is a beta-CoV, which has a genome similar to that of diagnosis of COVID-19. To determine whether the scFv-Fc antibodies are specific for SARS-CoV-2, 227 the binding of purified scFv-Fc antibodies to NP antigens from three known pathogenic beta-CoVs 228 was investigated by biochemical analyses. ELISA experiments revealed that the four scFv-Fc 229 antibodies bound to SARS-CoV-2 NP, but not to NPs from SARS-CoV or MERS-CoV (Fig. 1c) . This 230 indicates that the scFvs isolated from phage display screening are specific for SARS-CoV-2. The 231 interaction between scFv-Fc antibodies and SARS-CoV-2 NP was also analyzed in western and dot 232 blot experiments (Fig. 1d) . The results showed that 2H1, 12H8, and 1G5 scFv-Fcs bound strongly to 239 Next, to determine whether the scFv-Fc antibodies are sensitive enough for use as diagnostic 240 antibodies, we measured their binding affinity of scFv-Fc antibodies for SARS-CoV-2 NP using a 241 real-time label-free BLI biosensor. SARS-CoV-2 NP with a 6× His tag (ligand) was immobilized onto 242 a Ni-NTA biosensor, and the binding affinity of the scFv-Fc antibodies (analyte) was measured. 249 To confirm whether the scFv-Fc antibodies are suitable for use in the LFIA platform, we performed 250 an indirect LFIA. We used CNB as a detection probe because it is more stable and sensitive than gold 269 To develop the LFIA-based RDT, two different antibodies (one for capture and the other for detection) 270 are needed for a sandwich immunoassay. To select the optimum antibody pair for RDT, all possible 271 12H8-12B3, 12H8-1G5 pairs were selected for specific detection of SARS-CoV-2. The common 281 feature of the selected three diagnostic antibody pairs is that they share 12H8 as the capture antibody.

The 12H8 scFv-Fc shows the highest affintity for the target antigen ( Fig. 3b and c) . When 12H8 was 283 used as a detection antibody, however, the antibody pairs showed very weak positive signals (Fig. 4b   284 and Fig. S3 ). This may suggest that for the development of a sandwich diagnostic pair, the antibody 285 with higher affinity is better employed as the capture antibody rather than the detection antibody. In 286 fact, the duration of antigen contact is much shorter for the captrure antibody than that for the 287 detection antibody. Thus, using an antibody with higher affinity as the capture antiobody would 288 improve the sensitivity of the LFIA. 291 Finally, we used the three optimal combinations of scFv-Fc antibodies to ascertain the detection limit 292 for antigen protein and cultured virus. The results showed that the detection limit of the 12H8-12H1,

12H8-12B3, 12H8-1G5 pairs for NP antigen was 2 ng, 5 ng, and 10 ng, respectively ( Fig. 5a and b) .

Using the best pair of scFv-Fc antibodies, we then analyzed the limit of detection for SARS-CoV-2 295 virus. The results showed that LFIA using the 12H8-12H1 scFv-Fc antibody pair could detect SARS-296 CoV-2 virus at levels as low as 2.5 × 10 4 pfu/reaction ( Fig. 5c and d) . Moreover, there was no cross-297 reactivity with NPs of SARS-Co-V, MERS-CoV, influenza virus, or negative control nasal swab 298 specimens ( Fig. 5c and e) . These results indicate that the LFIA biosensor developed herein can 299 successfully distinguish between SARS-CoV-2-positive and -negative samples. Moreover, the The authors declare that they have no known competing financial interests or personal relationships 325 that could have appeared to influence the work reported in this paper. 

A pneumonia outbreak associated with a new coronavirus of probable bat 397 origin

A Novel Coronavirus from Patients with Pneumonia in China

 328 We thank the National Culture Collection for Pathogens of the Korean CDC for providing the clinical 329