Scientists at the Center for Infection and Immunity (CII) at Columbia University Mailman School of Public Health and SunYat-Sen University in China have set the stage for the development of highly sensitive antibody tests for infection with all known human coronaviruses, including new variants of SARS-CoV-2.
These tests should also allow differentiation of immune responses due to infection and vaccination. The research is published in Communications Biology, a Nature journal.
The HCoV-Peptide array developed by CII scientists consists of 3 million immune markers on a glass chip, covering proteins of all known human coronaviruses, including the SARS-CoV-2. In collaboration with a team at Sun Yat-Sen University, the CII researchers identified 29 immune signatures specific to SARS-CoV-2.
To develop the HCoV-Peptide array, the researchers first analyzed blood samples taken from individuals with asymptomatic, mild, or severe SARS-CoV-2 infections, and controls including healthy individuals and those exposed to SARS-CoV-1 and seasonal coronaviruses.
Next, they validated their test using a second set of blood samples, including those from confirmed cases of SARS-CoV-2, those with antibodies to other human coronaviruses, and healthy individuals.
The new test has a 98 percent specificity and sensitivity.
Immune signatures were present from eight days after onset of COVID-19 symptoms to as long as six to seven months after infection.
“This work will allow us and others to build inexpensive, easy to use blood tests that can provide data for exposure as well as immunity,” says author Nischay Mishra, Ph.D., assistant professor of epidemiology at the Columbia Mailman School.
“This work with our colleagues at SunYat-Sen, led by Professor Jiahai Lu, and with Nimble Therapeutics, underscores the importance to public health of global collaboration and partnerships with industry in addressing the challenges of the COVID-19 pandemic,” says senior and corresponding author W. Ian Lipkin, MD, director of CII.
Previously, the researchers have used similar methods to develop tests for Zika, acute flaccid myelitis, and tick-borne infections.
Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) is a single-stranded RNA virus in the Coronaviridae family that emerged in late 2019 and has caused morbidity, mortality, and economic disruption on a global scale with few precedents.1
The Coronaviridae family includes four species/strains that are endemic in the human population—HCoV-229E, HCoV-NL63, HCoV-HKU1, and HCoV-OC43 (Betacoronavirus 1 species)—and are usually associated with mild, self-limiting upper respiratory tract infections, although they can cause severe illness in immunocompromised patients.2
Two other species, Middle East respiratory syndrome-CoV (MERS-CoV) and SARS-CoV, have recently emerged and cause severe disease in humans. Like the other human-infecting CoVs (HCoVs),3,4 SARS-CoV-2 infection can elicit a robust antibody response in humans,5,6 and this response represents a major focus of widespread efforts to develop accurate diagnostics and strategies for passive and active immunization against infection.7, 8, 9
Existing serological assays for SARS-CoV-2 antibody reactivity generally use full-length viral proteins or domains—Spike (S), Nucleocapsid (N), or the receptor-binding domain (RBD) of S—as antigenic baits, followed by enzyme-linked or fluorescent detection.9
These assays provide a single measure of antibody reactivity, which represents a composite signal across many epitopes, and are able to detect viral exposure with a range of accuracies.10,11 Neutralization assays using either native or pseudotyped viruses have also been developed.12 It remains to be seen how these different assays will perform as diagnostics or correlates of the protection conferred by infection or vaccination.
Relative to protein-based analyses of the humoral response, epitope-level assays have the potential to add several layers of information. First, although SARS-CoV-2 proteins are generally distinct from other HCoVs, some regions of strong conservation exist,1,13 meaning that there is the potential for immune cross-reactivity that can only be resolved at the epitope level.
It was recently demonstrated that a large fraction of non-exposed individuals have T cell reactivity to SARS-CoV-2 peptides, indicating cross-reactivity with existing responses, possibly those generated against homologous peptides from endemic HCoVs.14
In the case of antibody responses, cross-reactivity has been described between the more closely related SARS-CoV and SARS-CoV-2.15,16 Epitope-resolved analyses therefore have the potential to identify antigens that may discriminate related CoVs, leading to more specific diagnostic assays.
High levels of sequence conservation may also indicate functional essentiality; therefore, by highlighting potentially cross-reactive epitopes in conserved regions of the proteome, epitope-level assays can identify antibodies and targets with therapeutic potential, against which viral escape may be more difficult.17
A second rationale for generating epitope-resolved views is that antibody recognition of different protein regions can have divergent functional consequences, including neutralization potential. For CoVs, antibodies binding the surface-exposed, receptor-binding S protein exhibit the greatest neutralizing potential,18,19 but these antibodies can recognize a wide variety of epitopes within the protein, each with the potential for different functional consequences.
This likely accounts for the imperfect correlation between the titers of S-binding antibodies and viral neutralization activity across individuals.20 Due to its interaction with the host entry receptor (the angiotensin-converting enzyme 2 [ACE2]), the RBD of S represents the predominant target of vaccination and monoclonal antibody development strategies, and a growing number of antibodies against this domain have been described.20, 21, 22, 23
However, the RBD is one of the less conserved regions of the CoV proteome, and antibodies against epitopes outside the RBD have also been shown to have neutralizing activity21,24; these may act in various ways, including by preventing important protease cleavage events and/or conformational changes required for successful entry into cells.
However, antibodies that recognize epitopes within the N protein, which coats the viral genome and is contained within mature viral particles, likely provide little or no neutralization potential, but may be useful signatures for differentiating vaccine responses from those resulting from natural virus infection, a strategy already used for other viruses.25,26
In addition to the different neutralization potential, it is possible that unfavorable distributions of epitope reactivity can contribute to immunopathology—for example, through antibody-dependent enhancement,27, 28, 29 although this phenomenon remains to be demonstrated for SARS-CoV-2.30
Peptide sub-sequences have been used for decades as probes to detect antibodies recognizing linear epitopes within the full-length proteins from which they are derived.31,32 Although unable to detect antibodies whose binding depends on elements that are discontinuous in the primary sequence, this strategy has the advantage that it enables the parallel design, synthesis, and assay of thousands of epitope-level antigen baits.
In its simplest format, peptides can be used individually—for example, in separate wells in an ELISA. A recent study used this approach to identify two linear epitopes in S protein that were targeted by neutralizing antibodies in SARS-CoV-2 convalescent donors.24 More powerful assays involve sets of peptides that are assayed in multiplex using either spatial addressing, in the case of peptide arrays,33 or DNA indexing, in the case of phage display libraries.34 Using the latter approach, the highly multiplexed and epitope-resolved detection of antibodies to viruses has been demonstrated with high sensitivity and specificity.35
Here, we present a synthetic biology approach to highly multiplexed peptide-based serological assays (PepSeq) in which libraries of peptide baits, each covalently coupled to a DNA barcode, are synthesized from high-complexity DNA pools using a simple and fully in vitro approach.
Library synthesis takes advantage of in vitro transcription and translation, including an intramolecular coupling mediated by puromycin,36,37 and the DNA-barcoded peptides can then be used to probe antibodies using a high-throughput sequencing readout.
We use this platform to synthesize libraries of overlapping 30-mer peptides covering all HCoV proteomes and assay these against sera from prepandemic and SARS-CoV-2 convalescent donors. Our results demonstrate the accurate detection of SARS-CoV-2 exposure and reveal multiple recurrent antibody epitopes, including two Spike epitopes at which antibody responses cross-react between SARS-CoV-2 and one or more endemic HCoVs. We further demonstrate that these cross-reactive antibodies preferentially bind to endemic HCoV peptides, suggesting that the response to SARS-CoV-2 at these regions is shaped by previous CoV exposure.
reference link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7816965/
More information: Nischay Mishra et al, Immunoreactive peptide maps of SARS-CoV-2, Communications Biology (2021). DOI: 10.1038/s42003-021-01743-9
