Analysis of Serologic Cross-Reactivity Between Common Human Coronaviruses and SARS-CoV-2 Using Coronavirus Antigen Microarray

Saahir Khan; Rie Nakajima; Aarti Jain; Rafael Ramiro de Assis; Al Jasinskas; Joshua M. Obiero; Oluwasanmi Adenaiye; Sheldon Tai; Filbert Hong; Donald K. Milton; Huw Davies; Philip L. Felgner

The 2019 novel coronavirus strain (SARS-CoV-2) originating in Wuhan, China has become a worldwide pandemic with significant morbidity and mortality estimates up to 2% of confirmed cases. The current case definition for confirmed COVID-19 due to SARS-CoV-2 infection relies on PCR-positive nasopharyngeal or respiratory specimens, with testing largely determined by presence of fever or respiratory symptoms in an individual at high epidemiologic risk. However, this case definition likely underestimates the prevalence of SARS-CoV-2 infection, as individuals who develop subclinical infection that does not produce fever or respiratory symptoms are unlikely to be tested, and testing by PCR of nasopharyngeal or respiratory specimens is unlikely to be 100% sensitive in detecting subclinical infection. Widespread testing within the United States is also severely limited by the lack of available testing kits and testing capacity limitations of available public and private laboratories. Therefore, the true prevalence of SARS-CoV-2 infection is currently unknown, and the sensitivity of PCR to detect infection is also unknown. Serology can play an important role in defining both the prevalence of and sensitivity of PCR for SARS-CoV-2 infection, particularly for subclinical infection. This point is demonstrated by analogy with influenza virus, for which a meta-analysis of available literature measured the fraction of asymptomatic infections detected by PCR as approximately 16%, while the fraction of asymptomatic infections detected by seroconversion was measured as approximately 75% 1 . The seroprevalence of common human coronaviruses is known to increase throughout childhood to near 100% by adolescence 2 . Thus, any serologic methodology to estimate prevalence of SARS-CoV-2 needs to identify and rule out cross-reactivity with these common human coronavirus strains. One challenge in applying serology to SARS-CoV-2 is that the choice of antigen and choice of assay is less well defined for coronavirus than more well studied viruses such as influenza. However, prior approaches to serologic detection of infection with emerging coronaviruses including SARS and MERS have focused on the S1 domain of the spike (S) glycoprotein and the nucleocapsid (N) protein, which are considered the immunodominant antigens for these viruses 3 . In particular, the S1 domain is strainspecific, while the N protein shows cross-reactivity across strains. The assay methodologies used for serologic detection of coronavirus infection include binding and neutralization assays. These methodologies have been shown to be well correlated 4 . However, neutralization assays require viral culture, which must be performed in high-level biosafety containment units for emerging coronaviruses with high epidemic potential such as SARS-CoV-2. Conversely, binding assays such as ELISA can be readily performed with widely available reagents and equipment so are field deployable and suitable for point of care testing. The protein microarray methodology has been widely used to simultaneously perform binding assays against hundreds of antigens printed onto a nitrocellulose-coated slide for detection of multiple antibody isotypes 5 . This methodology was recently demonstrated for simultaneous measurement of IgG and IgA antibodies against over 250 antigens from diverse strains and subtypes of influenza 6 . This methodology has previously been applied to detect antibodies against the S1 domains of SARS an