COVID-19: ORF9b peptide may be an ideal target for next-generation vaccines


Peptides encoded by unexplored regions of the SARS-CoV-2 genome provoke strong immune responses compared to other known peptides.

Current COVID-19 vaccines are effective at preventing severe disease, including infection caused by known variants of concern. But new variants of the SARS-CoV-2 virus could potentially evade immunity, so vaccine makers have already started developing next-generation vaccines.

Recent research has suggested that new vaccines that more potently stimulate the immune system’s T cells may provide longer-lasting protection against the virus, particularly against new variants.

A new study published in Cell has revealed new ways next-generation vaccines could potentially stimulate T cells against the virus. Scientists analyzed previously overlooked parts of the virus’s genome and uncovered a surprisingly large fraction of key viral protein fragments, or peptides, that triggered stronger T cell responses than other known peptides.

The findings, from researchers at the Broad Institute of MIT and Harvard, Boston University, and others, could help vaccine makers identify better viral targets for future vaccines that stimulate durable immunity against the evolving COVID-19 virus.

“Scientists are thinking about incorporating components invoking T cell responses into the next generation of vaccines because it seems like they might provide prolonged infection against new emerging variants,” said Shira Weingarten-Gabbay, a postdoctoral researcher in the lab of institute member Pardis Sabeti at the Broad and co-first author of the study.

“It’s becoming increasingly clear that when it comes to fighting off SARS-CoV-2, T cell immunity has a very important role,” said Mohsan Saeed, co-senior author and virologist at Boston University’s National Emerging Infectious Diseases Laboratories (NEIDL).

Two arms of the immune system

When the body encounters a virus, it raises two kinds of immune response. In B cell-mediated immunity, immune cells make antibodies that neutralize the virus. In T cell-mediated immunity, infected cells chop up viral proteins and present fragments of those proteins on the cell surface using the cell’s human leukocyte antigen (HLA) proteins.

These peptides act like beacons for cell-killing T cells, which launch an attack on infected cells and eliminate them from the body.

How these T cells interact with the SARS-CoV-2 virus is an active area of research. “In the past few months, there have been more and more studies showing that the T-cell response against the new variants is pretty much the same as the T-cell response to the parent virus,” said Weingarten-Gabbay.

This is a critical finding because it means that vaccines designed to stimulate T-cell immunity may not need to be updated as often when new and concerning viral variants emerge.

One possible explanation for this more consistent T-cell response is that vaccine-induced antibodies generally target the virus’s spike protein. This protein is among the most variable regions of the virus, and antibodies might not detect spike proteins that are highly mutated in new viral variants.

By contrast, viral peptides, or epitopes, that generate T-cell responses originate from a number of viral proteins, which are generally more genetically stable than the spike protein alone.

SARS-CoV-2 genome exploration

Knowing the importance of T cell-mediated immunity, the team quickly launched into action in April 2020. Weingarten-Gabbay had long been interested in viral antigens, and worked with co-first author Susan Klaeger of Broad’s Proteomics Platform to characterize peptides from SARS-CoV-2 and other viruses presented on HLA.

Weingarten-Gabbay had also collaborated previously with Saeed, who had developed specialized cell lines to study SARS-CoV-2 infection and had the training and biosafety level 3 facilities necessary to perform experiments with a highly infectious virus.

At the time, Saeed’s lab, only a year old, was already overwhelmed by requests for collaborations. “But I decided to join in with Shira and Susan because the proposed study was very exciting to me,” he said. “I knew together we would have all the expertise and resources to get this work done quickly and analyze the data in a meaningful way.”

Together with Sisi Sarkizova, a computational scientist in the Cell Circuits Program at the Broad and co-first author; and Jennifer Abelin, also of the Proteomics Platform and co-senior author; the team set out to study the peptides presented by SARS-CoV-2-infected cells using mass spectrometry.

Mass spectrometry was important for the study because it allowed the scientists to look for viral peptides in an untargeted way. Previous studies of T cell immunity in SARS-CoV-2 used more conventional methods and largely focused on peptides derived from specific regions of the virus’s genome called canonical open reading frames (ORFs).

But the team looked at other parts of the genome called noncanonical ORFs and directly identified peptides derived from these regions as well. “The beauty of mass spectrometry is that it is high throughput, and that it allows us to discover new epitopes without prior knowledge of a target’s sequence,” said Klaeger.

The team was surprised to find that as many as one out of four peptides uncovered by their mass spectrometry experiments were derived from non-canonical ORFs, indicating that many potential targets for vaccines had been overlooked.

A strong immune response

To probe the functional role of the peptides, the team worked with Massachusetts General Hospital, the Dana-Farber Cancer Institute, the La Jolla Institute for Immunology, and Repertoire, a biotech company that has developed specialized T-cell assays, to test HLA binding and T cell responses.

The team found that some of the hidden peptides provoked a stronger immune response than other peptides in the study, derived from canonical regions, in both mice and blood samples from COVID-19 patients. Perhaps most surprising, the team found that one hidden peptide – from a noncanonical ORF called ORF9b – exhibited stronger responses in patients than some of the most immunologically dominant epitopes described to date.

The results suggest the peptide may be an ideal target for next-generation vaccines; the research team has already shared their findings with scientists working on vaccine development.

The team also discovered other potential targets for therapeutic intervention, including proteins from a cellular proteasomal pathway that, when inhibited by SARS-CoV-2, can help the virus evade the immune system. The authors say that improved understanding of the virus’s immune evasion tactics could help scientists devise new strategies to interfere with infection.

Overall, the results suggest a careful reorientation of how scientists study the body’s immune response to SARS-CoV-2. “It’s quite remarkable,” said Weingarten-Gabbay. “There are these very strong signals coming from the virus that we’re blind to because we were looking in what we thought were the most important regions of the virus.”

The ongoing Coronavirus Disease 2019 (COVID-19) pandemic has caused a Once-in-a-Century global crisis1. Despite scientists worldwide racing to develop antiviral drugs, curative treatments are unavailable at the time of writing. The global economy is experiencing the worst plunge in recent history amid fears of further deterioration of the COVID-19 situation.

The International Committee on Taxonomy of Viruses (ICTV) officially named the causative agent of COVID-19, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), based on its similarity to SARS-CoV2. While SARS-CoV-2 and SARS-CoV share many proteins common in other CoVs, including 4 major structural proteins (S, E, M, and N proteins) and 16 nonstructural proteins (nsp1-16), they possess a unique set of proteins, namely orf3a, 3b, 6, 7a, 7b, 8a, 8b, and 9b3,4.

Since these proteins were believed non-essential for virus replication, they have named the accessory proteins. However, this name is misleading. Many studies have demonstrated the accessory proteins are critical to the virus’s survival in the host and contribute significantly to pathogenesis4,5.

They participate in a variety of virus-host interactions ranging from cell proliferation, programmed cell death, cytokine production to antiviral immunity evasion. Disrupting virus-host interaction critical to the viral life-cycle represents a good strategy for drug design because it can avoid resistance commonly induced by direct-acting antiviral drugs6,7,8. Elucidating virus-host interaction at the molecular level is therefore fundamental to identify drug targets in the host.

The accessory protein orf9b is present in both SARS-CoV-2 and SARS-CoV. This 98-amino acid (aa) protein is encoded by an alternative open reading frame (ORF) within the N gene and is translated via a leaky scanning mechanism during translation9. Crystal structures of orf9b alone revealed a homodimeric β-strand-rich structure (PDB id: 2CME, 6Z4U) with a hydrophobic central tunnel for lipid binding, consistent with the role of orf9b in the mature virion assembly10.

In hosts, SARS-CoV orf9b targets the MAVS/TRAF3/TRAF6 signalosome on mitochondria during infection and suppresses innate immunity11. A recent comparative viral-host protein-protein interaction analysis revealed that SARS-CoV-2 orf9b interacts with the mitochondrial outer membrane protein TOM7012, a 70-kDa membrane-anchored adapter implicated in preprotein import into mitochondria, endoplasmic reticulum (ER)-mitochondria contacts, and the activation of antiviral signaling cascade13. The binding of SARS-CoV-2 orf9b to human TOM70 can lead to the suppression of interferon responses14.

TOM70 is a multifunctional protein anchored to the mitochondrial outer membrane. It is a surface receptor of the translocase of the outer membrane (TOM) complex, the gateway of protein import in mitochondria15. TOM70 recognizes a group of preproteins with an internal targeting sequence and cooperates with the molecular chaperone heat shock protein (Hsp) 90 to transfer preproteins to mitochondria16,17.

Intriguingly, TOM70 also plays a crucial role in the activation of antiviral immune responses in hosts. It is a key adapter that relays antiviral signaling from the mitochondrial antiviral signaling protein (MAVS) to TANK-binding kinase 1 (TBK1)/interferon regulatory factor 3 (IRF3)13. Specifically, virus infection triggers the interaction between TOM70 and MAVS.

The N-terminal clamp domain of TOM70 (also referred to as the N-terminal tetratricopeptide repeats, TPRs) binds the C-terminal EEVD motif of Hsp90, resulting in the recruitment of the Hsp90/TBK1/IRF3 complex to mitochondria. Ultimately, IRF3 is phosphorylated and translocated to the nucleus activating the antiviral gene transcription.

Gordon and colleagues provided convincing evidence that human TOM70 binds both SARS-CoV and SARS-CoV-2 orf9b12. They showed that orf9b colocalized with mitochondria and TOM70 was identified as a high-confidence interactor of orf9b during the mapping of virus-host protein interactions. Furthermore, the authors conducted co-immunoprecipitation experiments to demonstrate that endogenous TOM70 was precipitated in the presence of orf9b. Finally, they co-purified both proteins following overexpression in Escherichia coli and obtained a stable TOM70-orf9b complex, which provided the foundation for structural characterization.

The cryo-electron microscopy (EM) structure of human TOM70 in complex with SARS-CoV-2 orf9b was recently determined to ~3.1 Å resolution12. The structure revealed that orf9b occupies the hydrophobic pocket on the C-terminal TPRs of TOM70. Given that this pocket is responsible for recognizing the internal mitochondrial targeting signal (MTS) of preproteins, occupying this site may undermine the function of TOM70.

Notably, when bound to TOM70, orf9b exhibited a helical conformation in stark contrast to the β-strand-rich structure of the orf9b homodimer in the absence of protein binding partner10. This unusual helical conformation of orf9b is reminiscent of a study demonstrating that most mitochondrial targeting sequence tends to form amphiphilic helices18.

TOM70 harbors two distinct sites for protein-protein interaction; while the N-terminal TPRs of TOM70 associate with the heat shock protein family molecular chaperones, the C-terminal TPRs bind mitochondrial preproteins for import. Whether substrate-binding at one site affects the binding at another site remains to be fully elucidated.

A fluorescence anisotropy experiment showed that yeast TOM70 could bind the peptide corresponding to the extreme C-terminal segment of Hsp70 (last eight residues) and the peptide derived from a precursor protein (or preprotein) MTS sequence (181–193aa of the yeast mitochondrial phosphate carrier protein)19. Intriguingly, the binding affinity of yeast TOM70 to the precursor protein-peptide was unchanged in the presence of the Hsp70 peptide, suggesting that binding at the N-terminal TPRs has little effect on binding at the C-terminal TPRs.

Li and colleagues determined crystal structures of TOM71 (a TOM70 homolog that also mediates preprotein transfer) complexed with the EEVD motif of the molecular chaperones Hsp70/Hsp90. Structural characterization suggested that upon binding to the EEVD motif, TOM71 was fixed in an “open state” in which the pocket receiving preprotein adopted a favorable conformation for the loading of preproteins20.

Although a cryo-EM structure of the human TOM70/SARS-CoV-2 orf9b complex has been determined to a reasonable resolution, it is challenging to further improve the resolution using this approach. By contrast, crystallographic methods can often achieve higher resolution structures and provide further details. In addition, robust crystallization of human TOM70 is important to structure-based drug design targeting this host protein. For example, fragment-based screening (FBS) has been widely applied for identifying novel lead compounds against SARS-CoV-2 since the outbreak of Covid-1921,22. High-quality crystals of drug target are usually prerequisite for the FBS method.

In this study, we determined the crystal structure of the human TOM70/SARS-CoV-2 orf9b complex to 2.2 Å resolution. Orf9b occupies the hydrophobic pocket on the TOM70 C-terminal domain (CTD). Owing to the high resolution of this structure, we identified 12 hydrogen bonds and 7 salt bridges between orf9b and TOM70, as well as 23 non-polar residues of TOM70 that recognize orf9b.

We assessed the interaction between TOM70 and orf9b using isothermal titration calorimetry (ITC) and found that while the orf9b homodimer did not bind TOM70, a synthetic peptide corresponding to a central segment of orf9b (denoted the C-peptide) bound TOM70 with nanomolar KD.

Furthermore, we demonstrate that the binding affinity between the Hsp90 C-terminal EEVD motif and TOM70 was greatly reduced when TOM70 was associated with orf9b. Similarly, the C-peptide also blocked the binding of the EEVD motif to TOM70, although to a lesser extent. Conversely, pre-incubating TOM70 with the EEVD motif has little effect on the binding affinity of the C-peptide to TOM70. In summary, our findings support the hypothesis that orf9b allosterically inhibits binding of the molecular chaperon Hsp90 to TOM70.


More information: Profiling SARS-CoV-2 HLA-I peptidome reveals T cell epitopes from out-of-frame ORFs. Cell. Online June 1, 2021. DOI: 10.1016/j.cell.2021.05.046


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.