key: cord-0819395-kdwwdp6v authors: Caceres-Martell, Y.; Fernandez-Soto, D.; Campos-Silva, C.; Garcia-Cuesta, E. M.; Casasnovas, J. M.; Navas-Herrera, D.; Beneitez-Martinez, A.; Martinez-Fleta, P.; Alfranca, A.; Sanchez-Madrid, F.; Escudero-Lopez, G.; Vilches, C.; Jara-Acevedo, R.; Reyburn, H. T.; Rodriguez-Frade, J. M.; Vales-Gomez, M. title: Bead-assisted SARS-CoV-2 multi-antigen serological test allows effective identification of patients date: 2021-04-10 journal: nan DOI: 10.1101/2021.04.08.21254348 sha: 8f8341ee8ea3b7f6cd4ef69b8ba5e0b0aa704cdf doc_id: 819395 cord_uid: kdwwdp6v Many new aspects of COVID-19 disease, including different clinical manifestations, have been identified during the pandemic. The wide array of symptoms and variation in disease severity after SARS-CoV-2 infection might be related to heterogeneity in the immune responses of different patients. Here we describe a new method for a simple multi-antigen serological test that generates a full picture of seroconversion in a single reaction. The assay is based on the detection by flow cytometry of multiple immunoglobulin classes (isotypes) specific for four SARS-CoV-2 antigens: the Spike glycoprotein (one of the highly immunogenic proteins), its RBD fragment (the major target for neutralising antibodies), the nucleocapsid protein and the main cysteine-like protease. Until now, most diagnostic serological tests measured antibodies to only one antigen and some patients seemed to not make any antibody response. Our data reveal that while most patients respond against all the viral antigens tested, others show a marked bias to make antibodies against either proteins exposed on the viral particle or those released after cellular infection. Combining all the four antigens and using machine learning techniques, it was possible to clearly discriminate between patients and healthy controls with 100% confidence. Further, combination of antigens and different immunoglobulin isotypes in this multi-antigen assay improved the classification of patients with mild and severe disease. Introduction of this method will facilitate massive screenings of patients to evaluate their immune response. It could also support vaccination campaigns both to select non-immune individuals and to distinguish infected patients from vaccine responders. The novel single-stranded RNA-enveloped beta-coronavirus, called Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2), causes the respiratory disease referred to as COVID-19 that was recognised by the WHO as a pandemic in 2020 [1, 2] . Average mortality ranges from 0.16 to 20.88% for women and 0.27-34.68% in men [3, 4] , depending on age. Moreover, the virus spreads very quickly and a significant percentage of patients develop an exacerbated immune response with a widespread inflammation and multi-organ failure [5] , which require long-term hospitalization and causes a serious problem to the health systems. Further, many cases are essentially asymptomatic, thus tests for viral infection are needed in order to follow the propagation worldwide and to identify the role of the virus, as new clinical manifestations are described in different patient groups. As with any other infectious disease, antibodies are generated against certain viral proteins and detection of those antibodies can be used in diagnostics, to complement assays for viral nucleic acids and to follow the evolution of the infection. One of the major antigens of SARS-CoV-2 is the envelope Spike (S), which mediates attachment to host cells and virus cell entry via its Receptor Binding Domain (RBD) [6, 7] . Antibodies directed against the RBD can often neutralise the infection [8] . Other viral antigens include the nucleocapsid protein (NP) and the 3CL main protease (Mpro), which are only synthesised once the virus has infected the cell. These viral proteins also generate antibody responses [7, 9] and can be used in serological tests [9, 10] . In fact, antibodies to the viral protease could be detected in plasma and saliva of individuals who had been infected with SARS-CoV-2 [10] . Antibody tests to detect exposure to SARS-CoV-2 are commercially available in several formats, such as ELISA, CLIA and lateral flow devices. These assays are useful tools for epidemiological studies that need to identify infected people. However, the commercial assays usually test for antibodies to only one antigen, generally either Spike or the nucleocapsid protein. As the pandemic advances and new clinical manifestations are described, it is important to evaluate the quality (e.g. antibody isotypes, response to different antigens), quantity (antibody titre) and duration of the immune response in patients with different severity and symptoms. Carrying out multiple ELISA assays, to analyse several antigens and immunoglobulin types over plasma dilutions for each individual, greatly increases the amount of reagents and time needed to evaluate large cohorts of patients. This thorough immune characterization would be of particular interest to follow up vaccination efficacy, as well as to follow up certain population groups such as immunodeficient patients or other high risk individuals. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint To facilitate the implementation of screenings of COVID-19 patient populations, we here report the development of a multiplex bead-based flow cytometry assay that assesses, in a single reaction, for sero-reactivity to four different SARS-CoV-2 antigens: the S protein extracellular region, its RBD, NP and Mpro. This test includes the analysis of IgA, IgG, IgM, which can also be performed simultaneously by using different fluorophores for each antiisotype on any standard flow cytometer (488nm or 633nm excitation), and do not require the use of specialized software. The technique yields results with extremely low background signals and has specificity and sensitivity near 100%, therefore providing a very good tool to have a full view of COVID-19 patient immune response. Differences in the specific Ig responses against the four antigens allows easy discrimination between vaccinated and naturally-infected individuals. Further, machine learning analysis allowed classification of patients and healthy controls, without any error just using IgG data. Thus, a simple multiantigen serological test clearly discriminates between patients and healthy controls with a 100% confidence. Modification of the algorithm to take Ig isotype data into account also allowed high confidence discrimination between mild and severe presentations of COVID-19 disease, at least retrospectively. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 10, 2021. Recombinant SARS-CoV-2 proteins were expressed with a histidine tag. Cys-like protease (Mpro) and nucleocapsid (NP) proteins constructs were expressed in the E. coli strain BL21 Star (DE3) pLysS (ThermoFisher) and purified as described [10] . . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint Recombinant cDNAs coding for soluble S (residues 1 to 1208) and RBD (332 to 534) proteins were cloned in the pcDNA3.1 vector for expression in HEK-293F cells using standard transfection methods. The two constructs contained the S signal sequence at the N-terminus, and a T4 fibritin trimerization sequence, a Flag epitope and an 8xHis-tag at the C-terminus. In the S protein, the furin-recognition motif (RRAR) was replaced by the GSAS sequence and it contained the A942P, K986P and V987P substitutions in the S2 portion. Proteins were purified by Ni-NTA affinity chromatography from transfected cell supernatants and they were transferred to 25 mM Hepes-buffer and 150 mM NaCl, pH 7.5, during concentration. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint ELISA assays for detection of antibodies directed against the four SARS-CoV-2 antigens were carried out as described [10] . To assess the prediction capacity of the new methodology, an algorithm was built using Scikit-learn python package [11] (code available on request). Samples were stratified and randomly spliced into a training and a test set. The training samples were used to fit a random forest classifier which then predicted the healthy vs disease category of unseen test samples (1/7 of total samples). This was repeated n=10,000 times. For each patient, accuracy was calculated as the proportion of correct predictions divided by the number of predictions made. As a complementary approach, a mean Receiver Operating Characteristic (ROC) curve was built for the random forest classifier by stratified 15-fold cross-validation, using the smaller set (2-3 samples) to train the model and then predicting the remaining ones. For heatmap representation, each variable was scaled to a range (0,1) using the MixMaxScaler command from Scikit-learn and visualized using heatmap command from seaborn python packages. For Principal Component Analysis, each variable was scaled as described, and the PCA command from Scikit-learn was used to fit and transform the data. Principal components up to a 95% of accumulated explained variance were saved. Comparison between severe and mild patients in each variable was performed by multiple ttests followed by False Discovery Rate (1%) correction by two-stage step-up method in Graph Pad Prism 8 Software (GraphPad Software, USA, www.graphpad.com). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint The NP, S and RBD proteins of Coronaviruses have been widely used in single-antigen serological assays for SARS and MERS-caused diseases [12] . However, the use of these antigens in combination with the immunogenic Mpro [10] , can more fully describe the magnitude and duration of the immune response in SARS-CoV-2-infected patients. In order to facilitate comprehensive characterization of COVID-19 patients with a high throughput approach, a multi-antigen assay was developed with several viral antigens immobilised on fluorescent beads, to allow flow cytometry detection of the multiple antibodies generated during SARS-CoV-2 infections. As depicted in Figure 1A The anti-His signal obtained for the S protein was lower compared to other viral antigens, but did not affect detection in patient plasma. The lower detection of S by His-tag antibody was likely due to a lower molar amount of S than NP, MPro or RBD bound to the beads. This was expected because S molecular weight (180 KDa) is at least four times higher than the other antigens (25-40 KDa). Sera analysis allowed a very good separation of control and convalescent samples in a wide range of dilutions ( Figure 1B,C) . The multi-antigen assay also . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint three antigens, one isotype detection (IgG) and one dilution results in accurate classification of patients, facilitating large screenings. Using a small training set, ROC curves were generated to compare the sensitivity and specificity of each single-antigen ELISA test and for the multi-antigen FACS technique ( Figure 3A ), and the latter again demonstrated the best performance, highlighting that a multi-antigen approach could be more useful in clinical contexts in which a high number of unknown samples must be classified using a limited amount of known controls. The enhanced efficiency of the multi-antigen test is likely related to the observation that some patients clearly respond preferentially to antigens present in the viral particle (S, RBD), while other patients respond mainly to antigens normally only exposed once cells have been infected (NP, Mpro) ( Figure 3B ) [10] . The existence of this bias was independently confirmed when a Principal Component Analysis (PCA) was performed with data for each antibody isotype. This analysis revealed a clear separation of seropositive and seronegative patients ( Figure 3C ). Inspection of the PCA loadings ( Figure 3D ) showed that, for IgG, the second principal component discriminated between production of antibodies against either NP+MPro or S+RBD. Similar patterns were noted when IgA and IgM responses were analysed (not shown). The detection of this bias when analysing only a limited number of patients suggests that preferential antigen-specific responses are common and makes a strong case for the use of multi-antigen serological assays to avoid false-negative results. As the pandemic has advanced, it has been established that not all the patients respond in the same manner to the infection by SARS-CoV-2. In fact, a large body of clinical manifestations have been described and it has been suggested that different types of immune response may contribute to these different presentations. Therefore, it will likely be relevant to characterise potentially biased antibody responses when exploring the association between SARS CoV 2 infection and different clinical manifestations. In aggregate, the multi-antigen assay produced data that easily and efficiently discriminated between seronegative and COVID seropositive individuals. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint In general, patients with higher antibody titres are more likely to have suffered a severe infection, indicating infection severity is linked to increased antibody titres [13] . However, analysis of only IgG responses did not clearly discriminate between patients who had suffered . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint The development of fast, sensitive serological assays to detect exposure to SARS-CoV-2 is important. Unlike tests based on the detection of viral nucleic acids, serological tests detect antibodies which remain in serum after elimination of the virus and recovery from disease [14] . Current serological tests include classical ELISA techniques, serum chemiluminescence immunoassays (CLIA) against viral proteins and IgG/IgM lateral immunochromatography. Commercial presentations of these tests usually involve the detection of antibodies to only one viral antigen, generally S or NP [9, 15] . Here we report the development of a robust, quantitative, multiplex methodology that provides a much more complete description of the humoral immune response to infection with SARS-CoV-2 with excellent sensitivity and specificity. The method can be easily put into practice in most hospitals and clinical laboratories. The simultaneous detection of antibodies to multiple viral proteins in a single tube greatly facilitates sample handling and comparison of the antibody response to different viral antigens, a consideration that is even more important given the identification of COVID-19-convalescent patients whose antibody responses appear to be strongly biased for specific viral antigens. Indeed, patients that only respond to RBD or Spike would be classified as dubious or seronegative if a test for NP antibodies only was used. Similarly, a test for Spike antibodies would not identify patients that respond only to antigens released from infected cells, such as the NP and Mpro. IgM was only present in certain patients, as it corresponds to earlier disease stages. Thus, the multi-antigen test essentially eliminates the problems of false negatives and positives, even in low-seroprevalence settings [16] . In this study, we confirmed our previous data, describing a strong correlation between the antibody responses against intracellular antigens like NP and Mpro [10] . Further, while the majority of patients produced antibodies against all the antigens tested, we also identified several individuals who made responses with a marked bias for either antigens exposed on the viral particle envelope (S, RBD) or those that are mainly intracellular (P, NP). These data suggest distinct humoral immune responses among individuals, perhaps depending on the cellular damage caused by the virus infection. Importantly, serological assays based on the detection of antibodies to only one viral antigen are unlikely to detect people with strongly biased immune responses. The use of multiplex assays will help to understand the significance of this biased response. Since the test can be automated and performed in a single reaction, the technique described here permits a simple, rapid and complete serological analysis of many patients, as it can be performed in multi-well plate compatible flow cytometers. In addition, the amount of . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint protein required to coat beads is lower than that required for ELISA. All the advantages from this novel method could be of considerable importance to support vaccination campaigns and to select convalescent plasma for therapeutic use. Further, the use of antigens that are not part of the vaccine formulation will allow discrimination between those individuals responding to the vaccine from those with antibodies to other viral antigens, because of infection either shortly before or after vaccination ( Figure 5 ). Finally, accurate serological testing will be critical to monitor the duration of the immune response after vaccination and to discriminate between vaccine-mediated protection from infection or disease. Moreover, this assay has been designed to be easily put into practice in clinical settings and it would be straightforward to include more viral antigens as they are discovered to be immunogenic, and so further optimise the sensitivity and predictive value of the serological analyses in different clinical settings. The simultaneous analysis of immunoglobulin isotypes and multiple antigens in one reaction can be combined with automated data analysis, allowing rapid evaluation of serological status. In this work we also tested the differences in serological responses between patients with mild and severe disease. Using the algorithms described here, patients requiring ICU hospitalization could be discriminated, with high confidence, from patients who experienced mild disease, something that appears difficult to achieve when the antibody response to only one antigen is assayed. It will be interesting to analyse the usefulness of these predictions on a prospective basis in order to evaluate its ability to suggest prognosis. The assay reported here also has great potential to facilitate thorough analyses of serological responses from patients with different SARS-CoV-2 disease manifestations. The multi-antigen test can provide results in large screenings to aid in testing for a correlation between a given antibody response and a clinical aspect. In addition to an impact on early classification of patients, current limitations in the availability of vaccine doses suggest a novel possible application for sensitive multi-antigen assays for SARS-CoV-2 seropositivity. It has been shown that the antibody response to the first vaccine dose in individuals with pre-existing immunity is comparable or greater to that observed in naïve individuals who have been immunized twice [17] . Screening of the unvaccinated population with an assay sufficiently sensitive to identify individuals previously infected despite waning of antibody titres over time, would allow these individuals to be given the vaccine as a single booster, sparing them from possible suffering and complications after a second dose, and freeing up many urgently needed vaccine doses to be given to individuals with no protection. The test described here would also provide comprehensive information to support selection of convalescent sera or plasma for therapeutic use. Our data indicate the importance of this multi-antigen, multi-. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint isotype analysis to detect potential SARS-CoV-2 reinfections in vaccinated individuals and suggest a possible use in establishing alternative vaccine administration routes that may elicit more potent IgA responses. This multi-antigen and multi-Ig assay can be easily modified for detection of antibodies in other fluids as saliva and breast milk. It is also highly tunable to different research needs, including detection of different immunoglobulins, other viral proteins and even other potential antigens present in vaccine formulations. In conclusion, the highly sensitive multi-antigen assay for flow cytometry offers the possibility of performing screenings on antibody response against SARS-CoV-2 both for research and to support patient diagnosis and development of therapeutic approaches based in convalescent plasma. Importantly, it provides a tool for the follow up of vaccinated individuals. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21254348 doi: medRxiv preprint A Novel Coronavirus from Patients with Pneumonia in China Coronaviridae Study Group of the International Committee on Taxonomy of V: The species Severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2 Disparities in Age-specific Morbidity and Mortality From SARS-CoV-2 in China and the Republic of Korea What's Sex Got to Do With COVID-19? Gender-Based Differences in the Host Immune Response to Coronaviruses The trinity of COVID-19: immunity, inflammation and intervention Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor A serological assay to detect SARS-CoV-2 seroconversion in humans Immune response to SARS-CoV-2 and mechanisms of immunopathological changes in COVID-19 Serology for SARS-CoV-2: Apprehensions, opportunities, and the path forward Cysteine-like Protease Antibodies Can Be Detected in Serum and Saliva of COVID-19-Seropositive Individuals Scikit-learn: Machine Learning in Python Comparison of Serologic Assays for Middle East Respiratory Syndrome Coronavirus. Emerg Infect Dis SARS-CoV-2 Infection Severity Is Linked to Superior Humoral Immunity against the Spike Immunology of COVID-19: Current State of the Science Serology assays to manage COVID-19 SARS-CoV-2 Serology: Much Hype, Little Data Robust spike antibody responses and increased reactogenicity in seropositive individuals after a single dose of SARS-CoV-2 mRNA vaccine The authors would like to thank the director of the CNB-CSIC, M. Mellado, for coordination; the donors and the Biobank Hospital Universitario Puerta de Hierro Majadahonda (HUPHM)/Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA) (PT17/0015/0020 in the Spanish National Biobanks Network) for the human specimens used in this study; the Infectious Diseases and Intensive Care Units of HUPHM, for organization of sample and clinical data collection; Sofía Garrido and Miriam García, from the Immunology