Bat SARS-like coronavirus that infects human cells

30 October 2013

Rhinolophus sinicusThe SARS pandemic of 2002-2003 is believed to have been caused by a bat coronavirus (CoV) that first infected a civet and then was passed on to humans. The isolation of a new SARS-like coronavirus from bats suggests that the virus could have directly infected humans.

A single colony of horseshoe bats (Rhinolophus sinicus) in Kunming, Yunnan Province, China, was sampled for CoV sequences over a one year period. Of a total of 117 anal swabs or fecal samples collected, 27 (23%) were positive for CoV sequences by polymerase chain reaction (PCR). Seven different SARS-like CoV sequences were identified, including two new ones. For the latter the complete genome sequence was determined, which showed a higher nucleotide sequence identity (95%) with SARS-CoV than had been previously observed before among bat viruses.

One of these new viruses was recovered by infecting monkey cell cultures with one of the PCR-positive samples. This virus could infect human cells and could utilize human angiotensin converting enzyme 2 (ACE2) as an entry receptor. The infectivity of this virus could also be neutralized with sera collected from seven different SARS patients.

None of the SARS-like coronaviruses previously isolated from bats are able to infect human cells. The reason for this block in replication is that the spike glycoprotein of these bat viruses do not recognize ACE2, the cell receptor for SARS-CoV. SARs-like CoVs isolated from palm civets during the 2002-2003 outbreak have amino acid changes in the viral spike glycoprotein that improve its interaction with ACE2. The civet was therefore believed to be an intermediate host for adaptation of SARS-CoV to humans. The isolation of bat SARS-like CoVs that can bind human ACE2 and replicate in human cells suggests that the virus might have spread directly from bats to humans.

This finding has implications for public health: if SARS-like CoVs that can infect human cells are currently circulating in bats, they have the potential to infect humans and cause another outbreak of disease. The authors believe that the diversity of bat CoVs is higher than we previously knew:

It would therefore not be surprising if further surveillance reveals a broad diversity of bat SL-CoVs that are able to use ACE2, some of which may have even closer homology to SARS-CoV than SL-CoV-WIV1.

Is there any implication of this work for the recently emerged MERS-CoV? Sequences related to MERS-CoV have been found in bats, and given that bats are known to be hosts of a number of viruses that infect humans, it is reasonable to postulate that MERS-CoV originated in bats. So far a 190 fragment of MERS-CoV nucleic acid has been found in a single bat from Saudi Arabia. Identification of the reservoir of MERS-CoV will require duplicating the methods reported in this paper: finding the complete viral genome, and infectious virus, in bats.

  • stgoldst

    I guess one of the questions is whether ACE-2 tropism is enough for transmission to and between people, or whether other mutations are necessary before that can occur. From what I understand SARS accumulated further adaptation in intermediate host(s) before sparking the human pandemic, and this can be tracked genetically. So it seems to me an open question whether ACE-2 binding is enough to facilitate sustained transmission, given that SARS evidently acquired important mutations in the amplifying host, from which it jumped to people.

  • profvrr

    I fully agree with you; ACE2 binding is clearly needed, but what other changes in addition? I’m not sure we know the role of any of the other changes that the virus accumulated on the way from bats to people. Not sure if we will ever.

  • Niraj Makadiya

    Great post! Is there any specific reason why bats (not other animals) are reservoir for most of these emerging viruses?

  • stgoldst

    Probably not for sure, but identifying, and trying to characterize, sequence differences between these newly identified CoV and SARS-CoV certainly seems like a worthwhile line of inquiry.

  • Alan Dove

    It might be related to bats’ diversity and distribution. Their order (Chiroptera) is the most diverse in the mammalian class, representing about 20% of all mammals, and there are bats everywhere except the polar regions. Bats are to mammals what beetles are to insects. Even if mammalian viruses were distributed purely at random, I’d expect 1/5 of them to be in bats. Many bat species are also highly social and migratory, which makes them excellent reservoirs.

  • NewProf1

    If you look in this paper, ( the previously identify SARS like virus in bats is not able to infect cells until you swap in the ACE2 binding domain of the SARS epidemic strain. With this swap you get infection, so the rest of the previously identified SARS-like bat Coronavirus was able to work in human cells. Whether this is able to transmit, is another question, but importantly, the rest of the genome works in humans.

  • NewProf1

    There is also this really interesting paper “Evidence for ACE2-Utilizing Coronaviruses (CoVs) Related to Severe Acute Respiratory Syndrome CoV in Bat” ( predicting that the ACE2 binding changes in a SARS like bat virus would have to come before it jumped to people, and that variant is pretty much what was found in the new paper.

  • NewProf1

    Also, remember that there are novel Coronaviruses found in the US already!

  • Niraj Makadiya

    Thank you very much Alan! Sounds interesting to me :)

  • stgoldst

    I’d like to hear from an immunologist on this too…after all to be the reservoir for all of these viruses the bat immune system is acting in some way that allows persistent infection in the absence (so far as we can tell) of disease.

  • stgoldst

    Is there a small animal model that mimics the human disease process? It’d be interesting to see if this newly discovered virus is similarly virulent to SARS-CoV in that kind of system. Pretty clear from that PNAS paper (and your lecture) that ACE2 binding by itself facilitates infection of HAEs, wondering if/what other factors might account for virulence, or if efficient replication mediated by ACE2 usage is sufficient.

  • NewProf1

    Yes, we have a very nice animal model for SARS-CoV. I am sure this is being put in mice and hopefully reported soon!

  • Andriy Buzun

    SARS-like isn’t SARS: there is report about new bat virus for which bat is host but not reservoir

  • Pingback: TWiV 258: Hedging our bats

  • Pingback: Influenza A viruses in bats

  • Pingback: Bat SARS-like coronavirus, influenza and NEIDL: ‘What’s New on Virology Blog?’ - The Global Dispatch