A New Study Reveals How Protein-to-Protein Interaction Mapping Decodes the Molecular Origins of Pandemic Viruses
Horseshoe bats have coexisted with SARS-like coronaviruses for millions of years without ever getting sick. Humans, as the world learned in 2020, are not so fortunate.
In a study published May 13, 2026, in Cell Host & Microbe, researchers at the Quantitative Biosciences Institute (QBI) at UCSF, the Gladstone Institutes, and the Icahn School of Medicine at Mount Sinai report a discovery that helps explain why the same family of viruses causes devastating disease in humans yet leaves bats entirely unaffected and what that biological divide could mean for our ability to predict and prevent the next pandemic.
Led by co-first authors Jyoti Batra, Ph.D., of the Krogan Lab at QBI UCSF/Gladstone and Magdalena Rutkowska of the Garcia-Sastre Lab at the Icahn School of Medicine at Mount Sinai and drawing on collaborators at Institut Pasteur in Paris and Fred Hutchinson Cancer Center in Seattle, the team catalogued hundreds of protein-protein interactions, mapping how SARS-CoV-2 and its closest bat coronavirus relative, RaTG13, engage with host proteins in both humans and bats simultaneously. By distinguishing which molecular handshakes are conserved across hosts and which are unique to each, the researchers achieved a level of resolution not previously possible for any coronavirus.
Bats can carry many viruses without getting sick, but scientists have never fully understood how these viruses are controlled inside bat cells. This study offers a compelling answer. By comparing how closely related coronaviruses interact with human and bat cells at the protein level, the team found that even small genetic differences lead to dramatically different outcomes depending on the host. A single position in a small viral protein called Orf9b makes all the difference. In human cells, it grabs onto Tom70 — a critical gatekeeper of the immune response — and silences it. In bat cells, the same protein reaches for MTARC2, a protein that fights back and limits infection. One switch, two completely different outcomes.
"What this study shows is that the difference between a virus that stays in bats and one that spills over into humans and causes catastrophic disease can come down to remarkably small genetic changes. By mapping these interactions at the protein level — across two viruses and two species simultaneously — we can begin to read the molecular signatures that predict spillover risk. That is exactly the kind of early warning system the world needs."
The study also required the team to build something that didn't previously exist. Working with lung tissue from a greater horseshoe bat, they created the first functional bat cell line of its kind, an essential new tool for studying how coronaviruses behave in their natural reservoir host. Along the way they discovered that a single mutation in one viral protein is enough to unlock SARS-CoV-2 replication in bat cells entirely.
"We were excited to discover that a single change in the virus can act like a switch, rewiring how it interacts with host cells and evades immunity across species. These findings reveal how viruses use different strategies across species and provide new clues into how they adapt — and what we might do to stop them."
“Our data highlight sarbecovirus host adaptations associated with zoonotic transmission between animal reservoirs and humans. The approaches described in this publication may also be applied to investigate whether other zoonotic viruses, including hantaviruses, rapidly develop adaptive changes during host switching that facilitate human outbreaks."
“Viruses provide a unique opportunity to observe evolution in real time. SARS-CoV-2 rapidly adapted to bat cells in our study, exposing post-entry barriers which have to be crossed by virus to jump to a new species. Integrating cross-species infection models with systems-level approaches can uncover conserved and host-specific determinants of viral replication and improve our ability to predict zoonotic emergence.”
"We found that a single residue difference in one viral protein was sufficient to dramatically alter the ability of SARS-CoV-2 to replicate in bat cells. This finding highlights how finely tuned virus-host interactions can shape adaptation across species and how critical it is to understand these mechanisms in order to better recognize viruses with potential for zoonotic transmission"
This research builds directly on QBI's landmark mapping of the SARS-CoV-2 human protein interaction network during the COVID-19 pandemic — work that identified 69 drug candidates and advanced 27 into clinical trials. Applying that same PPI mapping platform across bat and human cells, and across two related viruses, demonstrates the power of QBI's approach not just for understanding disease after it emerges, but for anticipating it before it does. In pandemic preparedness, the difference between response and readiness is knowing where to look. QBI now has a clearer map.
This research was supported by the National Institutes of Health (NIH/NIAID), the Howard Hughes Medical Institute, the Gladstone Institutes, the James B. Pendleton Charitable Trust, the Roddenberry Foundation, the Innovative Genomics Institute, and Fast Grants. Additional support was provided by the French National Research Agency (ANR) and the National Institute of Allergy and Infectious Diseases (NIAID) intramural research program.