Sarbecoviruses include the virus behind COVID-19 (aka, SARS-CoV-2) and hundreds of genetically related viruses largely found in bats. Although researchers understand how SARS-CoV-2 can infect cells, most of the other sarbecoviruses have not been grown in a laboratory. Without the ability to grow a virus in the lab, researchers are not able to develop effective treatments for it. Michael Letko, assistant professor at Washington State University’s Paul G. Allen School for Global Health, has spent years researching sarbecoviruses and found that several bat sarbecoviruses might be able to infect human cells differently from SARS-CoV-2.

This early finding was based on experimental approaches that do not involve real viruses, so to confirm these previous findings, Dr. Letko and researchers at the Wuhan Institute of Virology led by Dr. Zheng-Li Shi tested experimental conditions needed for sarbecoviruses to infect cells. Their latest study confirms the previous results that some bat sarbecoviruses can infect cells differently from pathogenic viruses like SARS-CoV-2 and provides new understanding for researchers developing treatments for coronavirus infections.  

Why we did this work:

Coronaviruses have caused significant diseases in people and livestock throughout history, with some of our common cold infections caused by coronaviruses (CoVs) that emerged hundreds of years ago. SARS, MERS, and now COVID-19 are the latest in this line but will likely not be the last. It is therefore critical to identify which of the hundreds of CoVs known to circulate in wildlife are likely to cause the next outbreak, and to develop vaccines and drugs that can stop them. In this work, we were able to culture a bat-CoV that is unable to use the ACE2 receptor that SARS-CoV and SARS-CoV-2 use. We then showed it can infect human cells under certain conditions. While this virus poses little direct threat to human health, our results provide new tools to understand how these viruses may be able to infect human cells, and to help design vaccines and drugs to stop them.

What useful information is gained from this research?

In order to develop any vaccine or drug for any virus, scientists must know how to grow the virus in a lab. This research is about learning what viruses need to grow in culture. For years, many of the (sarbeco) viruses discovered in animals have been considered to pose no threat to humanity because they cannot be isolated in the lab. This is somewhat flawed reasoning because researchers do not know the experimental conditions required to grow every virus in laboratory conditions. For example, the human coronavirus, HKU1, was considered “unculturable” for many years until breakthroughs in cell culture techniques allowed the virus to be studied in the lab.

A working hypothesis behind this research is that viruses adapt not only to the type of cell they are trying to infect, but also the environment that cell is in. Because most virus laboratory protocols exclude the conditions outside of the cell and instead use biochemically neutral conditions, studies trying to work with these viruses may systematically exclude essential viral cofactors required for infection. When provided with a more complementary suite of conditions, some viruses actually can propagate in human cell cultures suggesting they may have potential to infect humans. Therefore, researchers must understand these viruses better if we want to develop measures to protect ourselves from them. We are not giving viruses new abilities in this research, we are simply trying to better understand the abilities they might already possess.

What previous work led to this study?

Sarbecoviruses include SARS-CoV, SARS-CoV-2 and hundreds of uncharacterized, genetically related viruses in wildlife – mostly bats. All sarbecoviruses have a spike gene that interacts with a receptor protein on host cells to infect that host. In 2020, Letko et al. categorized the sarbecoviruses by their spike genes and then used laboratory tools with non-infectious “viral-like” particles, to demonstrate how many of the spike genes from these viruses could potentially enter human cells. A small number of the spike genes could infect human cells using the known, “ACE2-dependent” mechanism for these viruses. Several other bat viruses, however, could infect human cells without ACE2.

The laboratory approaches used in Letko et al. 2020 did not involve any complete sarbecoviruses, and instead relied on “pseudotype” assays that only mimic part of the viral entry process. Because Letko et al. 2020 used more protease during the cell culture experiments than what had previously been used for other viruses to test viral entry, it remained possible that these “harsher” conditions could not support multiple rounds of viral entry and replication.

In 2017, Dr. Zhengli Shi’s laboratory at the Wuhan Institute of Virology (WIV) identified a cave in Southeast China that contained a bat colony with high prevalence of bat sarbecoviruses in circulation. This published study served as a major source of the bat sarbecovirus sequence data employed by the coronavirus field – including most of the viruses studied in Letko et al. 2020. Researchers in the 2017 paper from WIV assembled reverse genetics clones of some of these viruses but were unable to recover infectious viruses from most of these clones, building on the assumption that most sarbecoviruses do not have any potential to infect human cells.

What was done in this current study?

After making several improvements to the Sarbecovirus entry assay that allowed for better detection of inefficient, low-level virus entry, Dr. Letko contacted Dr. Shi to see if these improved experimental procedures could be used to recover existing reverse genetics clones from the 2017 WIV study.

Using the adjustments to the cell-culture infection protocol, researchers at WIV were able to replicate viruses in cultures of both human and bat cells. The two laboratories then collaborated on this study, with molecular characterization experiments performed at WSU and work with whole viruses performed at WIV.

The main finding of this study is that the conditions determined in Letko et al. 2020, Khaledian et al. 2022, and refined in this study, allow for some bat sareboviruses to replicate in mammalian cell cultures independent of the known entry route via ACE2 that is used by human sarbecoviruses, SARS-CoV and SARS-CoV-2. Given all that researchers know about coronaviruses, their receptors and the host species, this study ends with a discussion that these ACE2-independent coronaviruses may be primarily gastro-intestinal in their natural hosts.

What viruses were used?

Recombinant molecular clones recovered by reverse genetics techniques.

  • In 2013, WIV researchers discovered a bat Sarbecovirus that was genetically similar to human SARS-CoV and could infect human cells using the same ACE2 entry route. 
  • In 2015, WIV and University of North Carolina used molecular biology techniques to synthesize a bat Sarbecovirus genome that can produce infectious viruses in cell culture. They use the viruses discovered in 2013 to develop this system. 
  • In 2016, UNC demonstrated that the bat Sarbecovirus molecular clone, called WIV1, can infect cells like SARS-CoV, but is attenuated in a small animal model. The authors speculate that WIV1 is less pathogenic because only part of its genome is similar to SARS-CoV while some of the Sarbecovirus genes thought to be involved in pathogenesis are noticeably different in WIV1.

In 2017, WIV generated several derivatives of the recombinant WIV1 virus genome by replacing the WIV1 spike gene with the spike gene from other bat sarbecoviruses. Most of these genomes failed to produce infectious viruses in culture and were not studied further.

In 2021, Dr. Letko contacted WIV to recover the failed molecular clones from the 2017 study.

These viruses are the attenuated molecular clone, WIV1, with the spike gene from bat sarbecoviruses that have demonstrated inferior human cell entry capabilities compared to human coronaviruses. Laboratory experiments demonstrating ACE2-independent spikes have human cell entry that is inferior to ACE2-dependent spikes can be found in:

What cells were used?

The human cell lines used in our study are 293T, Huh-7 and Caco2 which have been used in molecular virology research for over 50 years. The primary bat cell lines used in this study were developed for previous studies at WIV and are derived from kidney and small intestine tissues of bats.

What biosafety level was used for this study?

All experiments for this study were performed at biosafety level 2. Work with single-cycle, pseudotyped viral-like particles was performed at both WSU and WIV and was approved by the respective biosafety committees at both institutions. Work with the recombinant viral clones was performed at WIV under biosafety level 2 conditions and was approved by the biosafety committees at WIV.

Is this gain of function research?

No. Gain of function research applies to work with an infectious agent that aims to either deliberately introduce mutations or apply selective pressure for the acquisition of genetic mutations that enhance transmissibility, virulence, tropism or therapeutic resistance. We do not mutate any viruses in this study.

In this study, we worked only on bat coronaviruses that are not known to infect people and we did not introduce any mutations or apply selective pressures to adapt the viruses. We do not describe any mutations or demonstrate any improvements in transmissibility, virulence, tropism or therapeutic resistance. While the viruses do replicate in human cells in our study, these viruses only replicate in cell culture under specific conditions, requiring high levels of trypsin and are most efficient with the inclusion of high-speed cell culture centrifugation (1200 times the normal gravitational pull of the earth). The cell culture techniques that we used to grow these recombinant bat sarbecoviruses in the lab are very similar to what has been published for many other virus species over the past 70 years.

In addition, there has never been a single documented case of an ACE2-independent sarbecovirus infection in humans. The ACE2-independent viruses used in this study are demonstrably inferior in their ability to infect human cells compared to ACE2-dependent viruses, which may explain why these viruses have never been seen in humans. Seventy years of examples showing trypsin enhancement of viral entry in cell culture:

  1. A study from 1952 demonstrating exogenous trypsin is needed to for Feline Pneumonitis Virus propagation in cell culture: 
  2. A study from 1962 showing exogenous trypsin allows for influenza virus propagation in culture: 
  3. A study from 1975 again showing influenza viruses must be propagated in the presence of trypsin when grown in cell culture: 
  4. A study from 1981 showing that trypsin is needed to propagate rotaviruses in cell culture: ttps://
  5. A 2019 study with results very similar to ours but with bat viruses related to MERS-CoV. Many of the bat-derived merbeco-group of coronaviruses also only propagate in culture from reverse genetics systems in the presence of trypsin: 
  6. A 2020 study showing that, very similar to the trypsin requirement for some viruses, sapoviruses have been shown to require bile acids for viral replication in culture: