Researchers in the Blavatnik Institute at Harvard Medical School and at Brigham and Women’s Hospital are adapting an antibody-detection tool to study the aftermath of infections by the novel coronavirus that is causing the current global pandemic.
The tool, called VirScan, detects antibodies in people’s blood that indicate active and past infections by viruses and bacteria.
It was developed in 2015 by Stephen Elledge, the Gregor Mendel Professor of Genetics and of Medicine at HMS and Brigham and Women’s, and two PhD candidates in the lab, George Xu and Tomasz Kula.
Because it takes 5 to 10 days for a person to develop antibodies, Elledge emphasized that VirScan would not be used to provide real-time diagnoses of infection with SARS-CoV-2, the virus that causes COVID-19.
Rather, the goal is to analyze blood samples from people who recover from infection to learn about how the virus affects the immune system and the epidemiology of the disease.
Once underway, the effort will join others around the world attempting to study post-infection blood samples.
The results could lead to better estimates of true infection and lethality rates by capturing cases that may have gone undetected and could inform the development of vaccines. They could also reveal new insights into the fundamentals of human immunity.
“The situation right now is extremely difficult, but it’s great to be in a position to apply all these new methods to an important human health problem,” said Elledge.
How does VirScan work? How is it different from diagnostic tests?
From a single drop of blood, VirScan tests for antibodies against more than 1,000 different strains of viruses and bacteria that may have infected a person, whether around the time of testing or decades earlier.
This differs from typical blood tests known as ELISA assays, which look for one pathogen at a time.
It also differs from the tests currently used to diagnose COVID-19.
Those tests rely on mucus swabs from the nose and throat and look for nucleic acids that signal that the SARS-CoV-2 virus is contained in the sample.
“The CDC and other testing facilities are looking for the presence of the virus, which is critical,” said Elledge.
“Our assay can detect whether someone’s immune system has engaged the virus. We can tell when someone has harbored the virus but doesn’t have it anymore.”
To create VirScan, Elledge, Kula and Xu built a library of epitopes: short protein fragments derived from the surfaces of viruses.
If a person has encountered a particular viral strain, their immune system has generated antibodies against it.
Those antibodies will then recognize the epitope in the VirScan library and bind to it, giving a positive result.
Elledge’s lab included epitopes from several different coronaviruses in the original VirScan collection. The team is now adding epitopes from the new coronavirus as well as all other known coronaviruses not already included.
VirScan serological profiling of the human virome
The human virome, or the full spectrum of viruses that infect humans, is still not completely known. However, many different human viruses, along with their corresponding DNA and protein sequences, have already been characterized.
Despite this available information, current antibody-based methods for documenting viral infections typically involve testing a few viral proteins from a limited number of viruses at a time.
One new provisional profiling tool, VirScan, has the capability of testing for hundreds of viral proteins at once. Rather than being used as a diagnostic tool, VirScan can look broadly at an individual’s current and historical viral infection history .
This method is based on a T7 bacteriophage system, in which relatively short polypeptide sequences are displayed on bacteriophage surfaces and captured by immobilized antibodies (Figure 1). For VirScan, 206 different viruses known to infect human cells were first bioinformatically analyzed and used to generate a reference viral protein sequence dataset.
Using a programmable, DNA microsynthetic platform, 93,904 200-mer oligonucleotides were synthesized, encoding 56 amino acid peptides that encompassed every protein sequence of all 206 reference viruses.
This large pool of oligonucleotides was then subcloned into a phage T7 vector, allowing this massive library of viral peptides to be expressed individually on the surface of phage particles.
To analyze serologic responses, the viral phage library is used in an immunoprecipitation format with human serum samples. Antiviral antibodies present in serum samples are incubated with phage expressing the viral peptides, allowing the antibody–antigen immune complexes to be captured by immobilized protein A/G magnetic beads.
Following stringent washing, the corresponding bacteriophages are eluted and analyzed directly by high-throughput DNA sequencing.
Subsequent bioinformatics analysis is then used to match and assemble the sequences for each virus and ultimately elucidate the viruses present in a given individual. This technique provides the advantage of possibly identifying humoral responses to hundreds of different viruses at one time to determine the full range of viruses that have infected a given individual.
In one application, VirScan showed high detection sensitivity and specificity for identifying HCV- and HIV- infected individuals .
This is not unexpected, as both viruses are known to induce robust humoral responses. However, VirScan also revealed that individuals with untreated HIV infection showed a higher number of enriched HIV peptides than subjects with treated HIV infection, demonstrating additional information regarding the immunoreactive peptides.
Other insights from VirScan came from cataloging and comparing the seroprevalence of multiple viruses present in different human populations. Comparison of uninfected individuals to HIV-infected subjects revealed a much higher seroprevalence of antibodies in HIV patients against KSHV, HSV-2, CMV and other viruses, consistent with previous studies [47,48].
VirScan serologic analysis of children compared with adults demonstrated that adults had higher seroprevalence of viral antibodies for EBV, Influenza A and Influenza B, HSV-1 and HSV-2, while children had higher seroprevalence for Rhinovirus-A .
The lower seroprevalence of Rhinovirus-A in adults likely reflects the loss of B-cell responses over time in the absence of stimulation by these viral antigens. VirScan also provided evidence that the prevalence of different viruses was geographically different between Peru, South Africa, Thailand and the USA.
For example, almost 100% of individuals in Peru had HSV-1 infection, verses 85% in South Africa and only 50% in Thailand and the USA. Similarly, antibodies against the related herpes virus, CMV were present in almost 100% of individuals in Peru and South Africa, but only 50% of individuals in the USA.
Comparison of VirScan with existing technologies showed that it had high, but not perfect sensitivity for detecting HIV1, HSV1 and HSV2-infected individuals .
VirScan showed limited sensitivity for detecting other viral infections. For example, it underestimated the prevalence of VZV infection (only 25% prevalence vs the near 100% actual prevalence), which was likely due to the inability of the technique to detect antibodies directed against conformational epitopes of viral proteins.
It is also important to point out that there has only been one published report with VirScan, so additional studies are needed to validate the results and technology. However, the widespread use of VirScan technology will likely be hampered by the requirement for the highly complex components comprising hundreds of virus targets (i.e., derived from 93,904 200-mers) making it difficult to manufacture and thereby reproduce independently by other labs.
Additionally, VirScan is a complex procedure immunoprecipitation, DNA sequencing and bioinformatic analysis, which will likely need to be performed in a dedicated laboratory. Despite these issues, VirScan clearly has the potential to be a potent tool for untangling the roles of viral infection in overall heath and complex diseases.
One of the most interesting aspects of VirScan technology is its ability to determine the number and types of viral infections found in a given person. This line of inquiry studying 569 human donors revealed quite heterogeneous rates of virus exposure and infection.
On average, VirScan detected serum antibodies against ten different viruses per person . Antibody responses were detected against 62 of the 206 species of virus in the library in at least five individuals, highlighting the complexity of viral infection observed in some individuals. Remarkably, two individuals showed antibodies against 84 species of virus in the library.
It would be highly informative in future studies to use VirScan to study individuals with complex diseases to determine whether specific viral infections or the number of infectious agents may be associated with or are drivers of disease. The analysis of the role of nonpathogenic viruses in human health, like that of the mutualistic bacteria profile, may even reveal beneficial interactions .
By cataloging viral exposure over time in disease subsets, VirScan may yield novel insights into viral infections that are associated with protection against or susceptibility with many chronic illnesses including cancer, neurodegenerative and autoimmune diseases, where the risk factors and pathoetiology are currently not known.
For several reasons, including the fact that it takes at least a week to generate results, VirScan can’t be used as a real-time diagnostic test.
“It’s not feasible as a point-of-care test,” said Elledge. However, he added that his team might be able to use the information gained to generate a faster version of VirScan.
When the project gets up and running, it will be critical to ensure that blood samples are taken only from people who have recovered fully from SARS-CoV-2 infections, so that vials do not contain active coronavirus particles when they enter the lab.
“We don’t want to infect our researchers,” said Elledge.
How can the work improve estimates of infection and fatality rates?
So far, limited testing has meant that an unknown number of people in the U.S. and beyond have been infected with SARS-CoV-2 but remain uncounted.
Some may not have had symptoms. Some symptoms may have been attributed to other causes. This not only leaves individuals wondering about their infection and immunity status but also obscures the true infection rate across the population.
And without knowing how many people have been infected, it’s impossible to calculate the fatality rate—how likely the new coronavirus is to kill a person it infects.
Running VirScan analyses on blood, or serum, from a sizeable segment of the population can provide “a reliable estimate” of how many people were infected in a given geographic area, said Elledge.
Cross-referenced with medical records of those who tested positive and died, that information can illuminate the virus’s true lethality rate.
“Because right now they say, ‘this many people came in and tested positive,’ and ‘this many died,’ but if there are a lot of people who are not sick enough to go to the hospital and who don’t get tested, it makes the virus look more lethal than it might be,” said Elledge.
How can VirScan inform vaccine development?
VirScan promises to help Elledge and colleagues identify which parts of the virus the immune system responds to.
Recent work from his group suggests that people all over world infected with a particular virus make antibodies against the same proteins—”even the same amino acids”—on that virus, Elledge said.
That’s surprising, considering how many epitopes viruses have and how many antibodies are in the body’s arsenal, said Elledge.
The findings led him to suspect that some epitopes are, in effect, decoys, and therefore, that not all antibodies have the desired neutralizing effect.
“The immune system may be sending out all these antibodies like shooting a shotgun and hoping some of the spray will hit the target, neutralizing some critical part of the virus,” he said.
In principle, said Elledge, VirScan could indicate which epitopes are useful targets against the new coronavirus and which are just noise.
Then researchers could eliminate the useless ones from vaccines they’re developing.
How else is the lab working to assist vaccine efforts?
Antibodies aren’t the only objects in the body that attack invaders. Immune cells called T cells also react to specific epitopes—not on viruses, but on the surfaces of infected cells.
Alerted to danger by these epitopes, T cells can kill virus-infected cells and limit the number of viruses made in the body.
In 2005, Elledge’s lab built a tool, T-Scan, that can detect these epitopes. He would now like to teach T-Scan to detect the epitopes made when cells are infected by the new coronavirus.
But since cells infected by different viruses and bacteria sprout different epitopes, he first needs to know what the epitopes look like for infections with this coronavirus.
That would require obtaining not only blood but also T cells from people who recover from SARS-CoV-2 infection, he said.
The new coronavirus.. The image is credited to NIH/NIAID.
The goal: to identify the epitopes that trigger T-cell attacks so researchers laboring to develop COVID-19 vaccines can include them in the mix.
“T-cell epitopes are often important players in vaccines and in preventing viral infections,” said Elledge. “You want to encourage T cells to kill the infected cells.”
How can VirScan illuminate what SARS-CoV-2 does to the immune system?
Last year, the team used VirScan to help reveal how measles infection wipes out the immune system’s memory of past infections by other viruses and bacteria.
VirScan could similarly illuminate whether people develop immunity to the new coronavirus, how long they remain immune and whether infection causes more widespread damage to the immune system like measles does.
Or the virus may have other surprises in store, said Elledge.
What about the likelihood that there are different strains of the new coronavirus?
Although many mutations in the virus have been documented around the world to date, these variations “wouldn’t affect antibodies much,” so VirScan’s results should still apply, said Elledge.
Whose samples would be analyzed?
For the initial study, Elledge envisions collecting samples from about 100 volunteers who’ve recovered from COVID-19. The best-case scenario would be having samples from people before and after infection, he said, although he recognizes that that would be hard to arrange.
“Many people in the lab have given samples in the past, so if anyone gets sick, we’ll have a before and after, but of course we hope that doesn’t happen,” he said.
When will all of this be ready?
Most HMS labs have transitioned to remote work following institutional guidance aimed at containing the spread of the virus, but some have been granted permission to continue on-site work for COVID-19-related projects, including a portion of Elledge’s lab.
Elledge anticipates that VirScan could be deployed to analyze samples in mid-April. Then it would be a matter of obtaining institutional review board approvals for human research and arranging the logistics of collecting the samples.
Elledge is currently in talks with contacts across the HMS and broader Boston communities.
What else is in the works?
At the same time, Elledge’s team is working to detect antibodies against the new coronavirus with even greater sensitivity using a tool they developed in 2014 called PLATO.
Whereas VirScan uses short, linear protein fragments, PLATO uses full-length proteins known as open reading frames, or ORFs, which have a more developed 3D structure. (PLATO stands for ParalleL Analysis of Translated ORFs.)
Who is funding this work?
Elledge is an investigator of the Howard Hughes Medical Institute.
- The diagnosis of viral infections and the study of their clinical impact on overall human health and disease have been hampered by suboptimal detection methods.
- New emerging technologies have potential to reduce testing time, limit the need for sophisticated equipment, increase the scope of viral targets examined and/or increase the quality and quantity of information produced.
CRISPR-based viral detection
- CRISPR-based technology, such as SHERLOCKv2, allows for rapid detection of viral nucleic acids with attomolar sensitivity in a paper-based, field-deployable format.
- There is potential for a battery of CRISPR-based diagnostic tools to detect a wide range of viral infectious agents in clinical samples.
Portable DNA sequencing
- The MinION, a portable Nanopore DNA sequencer, provides real-time sequence information in a field-deployable format for targeted virus detection.
- Future improvements are still required for unbiased viral detection applications.
Luciferase immunoprecipitation systems antibody-based detection
- Luciferase immunoprecipitation systems provide quantitative detection of antiviral antibodies with high sensitivity and specificity.
- The wide dynamic range of antiviral antibody detection by luciferase immunoprecipitation systems is particularly useful for exploring the role viruses play in chronic human diseases.
VirScan analysis of the human virome
- VirScan can simultaneously screen antibody responses against hundreds of viruses.
- VirScan shows great potential for use as a profiling tool to catalog an individual’s viral infection history.
Several methods have been developed to characterize the specificities of protein-binding molecules. Display technologies are typically limited to shorter polypeptides and cDNA-based libraries suffer from highly non uniform clonal abundance distributions and incorrect reading frames.1
Two-hybrid and split-reporter techniques,2 are limited to analyses of bait molecules that can be presented within the cell, and are not suitable for drug or antibody target identification.
More recently, protein microarrays have been used for these purposes,3 but their construction typically requires individual proteins to be purified and arrayed, resulting in substantial costs and various degrees of protein denaturation.
To address these limitations, we developed PLATO (ParalleL Analysis of Translated ORFs), a method that combines in vitro display of full-length proteins with analysis by high-throughput DNA sequencing.
We demonstrate the utility of PLATO by performing diverse interaction screens against the human ORFeome, a normalized collection of 15,483 cDNAs in the Gateway cloning system.4
To express an ORF library in vitro, PLATO employs ribosome display, a technique used to prepare a library of mRNA molecules that remain tethered to the proteins they encode by lacking a stop codon.5
Ribosome display imposes minimal constraints on the length or composition of proteins that can be efficiently displayed.
We constructed a Gateway cloning–compatible ribosome display ‘destination’ vector (pRD-DEST; Supplementary Fig. 1), to be be used as a recipient for a normalized pool of ORF ‘entry’ clones.
After recombination, DNA is amplified by PCR, yielding linear templates lacking stop codons. Following in vitro transcription and translation, the ribosome-displayed ORFeome can be screened for binding to immobilized bait(s).
Enrichment of candidate binding proteins can be rapidly assessed using quantitative real-time PCR (qPCR) with ORF-specific primers, or en masse by deep sequencing of the enriched mRNAs (Fig. 1a).
Sequencing libraries can additionally be highly multiplexed, thereby reducing the cost of each screen. All steps required for PLATO are compatible with automation using standard liquid handling robotics.
Our strategy for deep sequencing of enriched display libraries employs recovery of the ORF 3′ termini, which minimizes interference from RNA degradation and ensures stoichiometric correlation between tag counts and transcript abundance.
To this end, we adopted the following protocol:
(i) chemically fragment enriched mRNAs;
(ii) reverse transcribe fragments using a common primer;
(iii) polyadenylate cDNAs;
(iv) add sample barcodes and sequencing adapters using two-stage PCR amplification (Fig. 1b).
Subsequent multiplex deep sequencing analysis of pooled display libraries is reproducible and quantitative (Supplementary Fig. 2). Sequencing a sample of unenriched human pRD-ORFeome mRNA (input) detected the transcripts of 14,582 unique ORFs out of 15,483 total cDNAs in the entry clone library (94%, Fig. 1c).
To test the ability of PLATO to identify protein-protein interactions, we used LYN, which contains common structural components of the SRC family, including SH3, SH2 and kinase domains,6 and has been extensively characterized for its interaction partners.
After affinity enrichment of the human ORFeome using GST-LYN, GST alone or an unrelated GST-fused protein (GST-Muted), we used Illumina sequencing to identify proteins specifically bound by GST-LYN (Fig. 2a, Supplementary Table 1, Supplementary Fig. 3a).
A number of established LYN binding partners were among those identified, and we validated two by qPCR (Fig. 2b).7, 8 We ranked candidate LYN interactors by their degree of enrichment on GST-LYN, and confirmed five of seven tested by western blot analysis (Fig. 2c).
Of the two candidates not validated, one bound nonspecifically to GST, whereas the other was a true negative. Among the highly enriched ORFs, SH2 domain-containing proteins were overrepresented (P < 0.01, Fisher’s test).
Consistent with a role for LYN autophosphorylation in mediating these interactions, phosphatase treatment of immobilized GST-LYN abolished binding of SH2D1A and SH2D4A, but only partly diminished PIK3R3 binding, suggesting the presence of an additional interaction domain (Supplementary Fig. 3b). These proteins have not previously been reported to interact with LYN.
We next asked whether PLATO could be used to identify protein targets of antibodies from patients with autoimmune disease. We first examined target enrichment using to affinity purified P53 and PDCD4 antibodies immobilized on protein A/G beads for library immunoprecipitation.
By qPCR, P53 and PDCD4 transcripts were robustly enriched by their cognate antibodies, but not by control antibodies (Supplementary Fig. 4).
In previous work, we synthesized an oligonucleotide library encoding a 36-residue overlapping human peptidome for display on bacteriophage T7 (T7-Pep). Deep sequencing of affinity-enriched T7-Pep using autoimmune cerebrospinal fluid (CSF) from three individuals with paraneoplastic neurological disorder (PND) uncovered known and novel autoantigens.9
We screened these samples using PLATO. Unlike T7-Pep, the human ORFeome is an incomplete collection of full-length proteins, and our findings reflect the inherent complementarity of these libraries. For example, neuro-oncological ventral antigen 1 (NOVA1) is absent from the human ORFeome v5.1, and so PLATO was unable to detect this known autoreactivity in patient A, whereas it was robustly identified with T7-Pep. Conversely, PLATO identified numerous autoantigens for each patient that were missed in our peptidome screens (Supplementary Table 2).
For example, PLATO analysis of patients A and B revealed immunoreactivity with known cancer autoantigens not detected with T7-Pep.
Several of these reactive antigens were selected for confirmation via immunoprecipitation and western blotting (Fig. 2d, Supplementary Fig. 5a–d). In addition, we had previously established that antibodies from patient C recognized the tripartite motif containing proteins TRIM9 and TRIM67.
PLATO considerably expanded the members of the TRIM family recognized by antibodies in this patient’s CSF to include TRIM1/MID2, TRIM18/MID1, TRIM54 and TRIM55 (Fig. 2e). Notably, multiple sequence alignment results in tight clustering of this precise subset of the extended TRIM family, suggesting the presence of shared, conformational epitopes not represented in T7-Pep.10
As an alternative PLATO readout, hybridization of autoantibody-enriched libraries to custom oligonucleotide microarrays revealed a similar list of autoantigens (Supplementary Fig. 6).
Discovering the targets of small molecules typically involves the use of cell extracts containing a wide distribution of protein abundances, which limits the accuracy of detection by mass spectrometry.
Normalized ORF libraries and quantitative DNA sequencing might therefore offer greater power to detect protein-small molecule interactions. We tested this idea with gefitinib, an inhibitor of epidermal growth factor receptor’s (EGFR) tyrosine kinase domain.
Gefitinib interacts with the ATP-binding pocket of EGFR and additional tyrosine kinases.11 Analysis after ORFeome affinity enrichment on gefitinib-coupled beads revealed significant enrichment of 10 out of the 17 predicted targets tested (Fig. 2f).
This experiment demonstrates the relative ease by which candidate protein interactions can be assayed with PLATO; the binding of any ORF can be rapidly assessed using qPCR without the need for cloning or western blotting.
ORFeome libraries affinity enriched by the Src family tyrosine kinase inhibitor dasatinib exhibited overrepresentation of protein kinases (P = 0.0003; Fisher’s test), including the known target LCK and several targets not previously associated with this compound (Supplementary Table 3).
PLATO’s limitations include incomplete ORFeome collections and a lack of protein post-translational modifications. However, the quality, completeness and availability of these libraries will continue to improve over time.
In addition, very large ORF proteins may be displayed less efficiently and proteins containing membrane-spanning or aggregation-prone domains that normally require host cellular machinery for proper folding may aggregate; these factors may complicate data analysis.
Finally, ribosome display imposes certain limitations on the conditions under which affinity enrichments can be performed (e.g. low temperature and absence of RNAse contamination are essential), and using proteins containing nucleic acid-binding domains as baits may result in non-specific binding.
When the required conditions for PLATO are met, however, this method provides three main advantages as a tool for proteomic investigations. First, it has minimal protein size and composition bias. Second, it has low cost and minimal instrument requirements.
Finally, the rapidly declining cost of DNA sequencing will make PLATO an ideal platform for projects involving large numbers of samples, such as cohort-scale autoantibody profiling or structure-activity relationship analyses of small-molecule compounds.
43. Zhou P, Fan H, Lan T et al. Fatal swine acute diarrhoea syndrome caused by an HKU2-related coronavirus of bat origin. Nature 556(7700), 255–258 (2018).
•• Demonstrates luciferase immunoprecipitation systems’s potential for rapid development of a serologic test for monitoring antibody responses to a newly discovered virus.
46. Xu GJ, Kula T, Xu Q et al. Viral immunology. Comprehensive serological profiling of human populations using a synthetic human virome. Science 348(6239), aaa0698 (2015).
•• Publication describes the details of the VirScan technology for simultaneously studying the infection history of hundreds of viruses at one time.
49. Roossinck MJ. The good viruses: viral mutualistic symbioses. Nat. Rev. Microbiol. 9(2), 99–108 (2011).