Are we *sure* Ebola isn’t airborne?

Since yesterday’s post, several people have asked me on various social media outlets about the airborne nature of Ebola. Didn’t I know about this paper (“Transmission of Ebola virus from pigs to non-human primates“), which clearly showed that Ebola could go airborne?

Indeed I do–I wrote about that paper two years ago, and it in no way changes my assertion that Ebola doesn’t spread between people in an airborne manner.

Let me back up. The paper in question was an experimental study done in the wake of the 2008 finding of the Reston Ebola virus in pigs and a previous study looking at the Zaire virus in pigs. In the air transmission study, they inoculated pigs with Ebola and examined transmission to macaques (who were not in direct contact with the infected pigs). They did find aerosolized Ebola in air samples, and some of the macaques did come down with symptoms of Ebola. So, it looked like pigs could spread Ebola through the air, which is something that had already been suggested by the epidemiology of the 2008 pig Ebola outbreak. It’s always nice when experimental data matches up with that observed during a real-life occurrence of the virus.

*However*, the kicker was not that Ebola is transmitted by air in human outbreaks, but rather that there may be something unique about pig physiology that allows them to generate more infectious aerosols as a general rule–so though aerosols aren’t a transmission route between primates (including humans, as well as non-human primates used experimentally), pigs may be a bigger threat as far as aerosols. Thus, this may be important for transmission of swine influenza and other viruses as well as Ebola.

So unless you’re sitting next to an Ebola-infected pig, seriously, airborne transmission of Ebola viruses isn’t a big concern. (Perhaps this corollary should be added to this handy diagram examining your risk of Ebola).


Find more of my writing on Twitter or Facebook

Pig-to-monkey Ebola: is there something in the air?

Ebola has long been known to be a zoonotic virus–one which jumps between species. Though it took several decades to find evidence of Ebola virus in bats, these animals had previously been associated with human index cases of Ebola disease have worked in bat-infested warehouses or traveled to caves where bats roost. Non-human primates have also become infected with the virus, sometimes transmitting the virus to humans when killed primates are butchered for food. Ebola has also been suggested to infect dogs and other wild animals. However, livestock are a newer angle to Ebola virus ecology.

Ebola was first found in pigs in 2008 in the Philippines. This was the Reston virus, named after its discovery in imported Filipino monkeys in a facility in Reston, Virginia, in 1989. Though this virus spread among the captive monkeys, no humans came down with symptoms. However, follow-up studies showed that some humans did develop an immune response to the Reston virus–suggesting they had been infected, even if they didn’t realize it. At the time, there was suggestion that perhaps Reston might be spread via aerosol, as the virus appeared to spread amongst monkeys in two different rooms who did not come into physical contact with one another. However, this was not proven at the time and alternative explanations were possible.

When Reston resurfaced in swine and swine farmers in 2008, a similar phenomenon was observed. Though it was not known how the pigs initially became infected with the virus, they did appear to be able to spread it to humans working amongst them, even if those farmers didn’t have contact with blood or other secretions (the most efficient way to transmit Ebola viruses). Suggestive of possible transmission from pigs to people via air, but far from conclusive. Since then, two experimental studies have examined airborne transmission of Ebola via pigs.

The first study examined transmission of the Zaire strain of Ebola–the nastiest one, which can kill up to 90% of those infected–within laboratory pigs. Pigs were inoculated with the Zaire virus and housed with uninfected pigs, who were later tested and found to have acquired the virus. Interestingly, when the pigs got sick with Ebola Zaire, the symptoms were mainly respiratory and the virus replicated in the lungs. This was quite unlike what Zaire does in humans and our other primate cousins, where it’s a systemic disease and we can find virus in the blood. This suggests that pigs could be infected with even nasty types of Ebola, and we wouldn’t realize it.

Last week, Ed Yong reported on a new paper examining transmission of Zaire virus from experimentally-infected pigs to co-housed macaques. Like the previous paper, they observed that Ebola in pigs was a respiratory disease, and that it could spread to other animals (in this case, non-human primates). The macaques they tested developed the symptoms of Ebola that were expected–a systemic disease, with virus isolated from the blood. In this study, they also added in an air sampling component, and were able to detect evidence of virus (via PCR) in the air. However, the authors do note that this could have been aerosolized in other manners than directly from the exhaling pigs (such as during the floor-cleaning process). Finally, even if it does become aerosolized and spread in this manner, as noted in Ed’s article, Ebola is not “suddenly an airborne virus, like influenza.” Certainly more efficient transmission takes place via close contact with infected secretions during hospital procedures and funeral rites.

Interestingly, the authors note that other experimental studies that have looked specifically at airborne, primate-to-primate transmission of Ebola have come up negative, and that swine are known to generate “infectious short range large aerosol droplets more efficiently then other species.” Is there something specific about pig physiology that may make them better respiratory virus shedders? We know that pigs can be intermediate hosts for other viral pathogens as well, such as Nipah virus and of course influenza.  Are pigs playing any role in Ebola ecology, either in Asia or Africa? Might Ebola have more airborne potential than we previously thought? According to Ed, the authors of the second study are currently working on field studies in Africa to examine the pig question outside of the laboratory. The timing may be good for them, as Uganda is currently experiencing another Ebola outbreak;–the country’s third Filovirus outbreak in five months.


Weingartl, H., Embury-Hyatt, C., Nfon, C., Leung, A., Smith, G., & Kobinger, G. (2012). Transmission of Ebola virus from pigs to non-human primates Scientific Reports, 2 DOI: 10.1038/srep00811

The human origins of “pig” Staph ST398

I recently gave a talk to a group here in Iowa City, emphasizing just how frequently we share microbes. It was a noontime talk over a nice lunch, and of course I discussed how basically we humans are hosts to all kinds of organisms, and analysis of our “extended microbiome” shows that we share not only with each other, but also with a large number of other species. We certainly do this with my particular organism of interest, Staphylococcus aureus. There are many reports in the literature showing where humans have apparently spread their strains of S. aureus to their pets (dogs, cats, hamsters)–and sometimes the pets have been nice enough to share it right back. My own research looks at S. aureus in pigs and the humans who care for them, and many studies have shown that a “pig” type of MRSA, dubbed sequence type 398 (ST398), can be transmitted from pig carriers to their human caretakers. The assumption has been that this is truly a “pig” strain, originating in swine, and has spread to humans (and other animals, including cattle, poultry, dogs and horses) from pig hosts, either directly or indirectly via contaminated meat products.

According to a new study (open access in mBio), it seems that there has been more sharing of ST398 than we’d realized. Led by Lance Price at TGEN (full disclosure–I’m a coauthor on the paper), his group analyzed 89 ST398 isolates from China, Europe, and North America, including isolates from humans and animals as well as both methicillin-susceptible and -resistant strains. Using whole genomic sequence typing, the evolutionary history of these isolates was reconstructed.

The findings throw the ST398 story a bit on its head. Instead of being a true pig strain, ST398 appears to have originated as a methicillin-susceptible human strain which was transferred into the pig population, picked up antibiotic resistance genes (including resistance to methicillin and tetracyclines), and then has been passed back to farmers as more resistant organisms. Some prophages were also gained or lost along the way, probably due to selection by host factors.

This also suggests that there is still likely a low level of “native human” ST398 circulating in people. There have been a few case reports of ST398 colonization and/or infection in people without any known livestock contact. Some of these have been resistant to methicillin and/or tetracycline, which are more frequently associated with livestock-adapted strains. Are these truly “human” strains which aren’t involved in livestock at all, or are these ST398 findings in people lacking livestock contact still due to some livestock exposure along the chain of transmission (farmer neighbors? Transmission via food?) We still don’t know, but carrying out more of this WGST will give us better targets in order to be able to differentiate true “human” ST398 strains from those that have been hanging out in animals, and then transmitted back to people.

Now, for long-time science blog readers, this story may sound a bit familiar. Indeed, it looks like ST398 has taken a very similar path to that of another animal-associated S. aureus strain, ST5. As Ed Yong described back in 2009, humans are also the ultimate origin of a “chicken” type of S. aureus ST5, which spread around the world in broiler chicken flocks. In Ed’s article, the first author of the chicken ST5 paper, Bethan Lowder, notes that the change in chicken farming from small farms to multinational corporations likely aided the spread of this organism–and the exact same thing has happened with pig farming.

One difference between the two is that ST5 causes disease in chickens, whereas ST398 seems to be a very rare cause of illness in pigs. This is likely one reason that ST398 in pigs went undetected until relatively recently–it’s simply not much of an economic issue for pig producers, whereas in chickens, S. aureus can cause several nasty diseases (such as bumblefoot and BCO) leading to animal loss (and thus, less money for the farmer).

So, where do we go from here? Clearly studies like this show the utility of using WGST to examine the evolution and spread of these strains. If you look at how spa types are distributed throughout the tree, you can see that those alone don’t tell you much about where the strain came from, or if it’s fully “human” or a pig-adapted lineage. Ideally, a set of simple markers could be found to distinguish ancestral human strains from livestock strains (as methicillin-sensitive ST398 can also be found in pigs, so methicillin resistance alone isn’t enough of an indicator that it’s a “pig” strain). We’ll be working on this in ST398 and other strains we see being shared between animals and humans, in order to better understand this generous sharing we’re doing amongst species.


Lance B. Price, Marc Stegger, Henrik Hasman, Maliha Aziz, Jesper Larsen, Paal Skytt Andersen, Talima Pearson, Andrew E. Waters, Jeffrey T. Foster, James Schupp, John Gillece, Elizabeth Driebe, Cindy M. Liua, Burkhard Springer, Irena Zdovc, Antonio Battisti, Alessia Franco, Jacek Żmudzki, Stefan Schwarz, Patrick Butayej, Eric Jouy, Constanca Pomba, M. Concepción Porrero, Raymond Ruimy, Tara C. Smith, D. Ashley Robinson, J. Scott Weese, Carmen Sofia Arriola, Fangyou Yu, Frederic Laurent, Paul Keima,, Robert Skov, & Frank M. Aarestrup (2012). Staphylococcus aureus CC398: Host Adaptation and Emergence of Methicillin Resistance in Livestock mBio, 3 (1), 305-311 : 10.1128/mBio.00305-11

Pigs with Ebola Zaire: a whole new can o’ worms

Ebola has long been associated with wildlife. From the early days, bats were viewed as a potential reservoir (though it wasn’t confirmed that they actually harbored the virus until 2005). Contact with wild animals–particularly primates which were butchered for food–was also long thought to be a risk factor, and now we know that primates can become ill with Ebola and pass the virus to humans.

What hadn’t been examined until 2008 were pigs. I mean, it’s not exactly the animal you associate with central Africa, where many of the Ebola cases have been concentrated. However, pigs are much more plentiful in the Philippines, where another Ebola subtype–Ebola Reston–is thought to lurk. The Reston strain actually was first documented in the United States, where twice it was associated with outbreaks in primates originating from the Philippines. When the facility in the Philippines was closed down in 1997, Reston disappeared for 11 years–until it surfaced in pigs in 2008.

The ecology of Ebola Reston in the Philippines isn’t known–unlike African Ebola strains (and their cousin, Marburg), no bats have been caught in that country and tested positive for the virus, though they probably serve as a reservoir of the virus in the Philippines just as they do in Africa. So it was a huge surprise when pigs from that country tested positive for Ebola Reston–and so did 6 of their human caretakers, suggesting cross-species transmission. (I should note here that the Reston strain has yet to be linked to any symptomatic infections in humans–the pig farmers who tested positive probably had no idea they’d been infected and did not show any clinical signs of illness). Pigs hadn’t previously been linked to any Ebola infection, so this brought in a whole other wrinkle when it came to Ebola transmission–the possibility of being exposed to Ebola via contaminated food, and the potential for pig populations to harbor the filovirus (and transmit it to their caretakers, as we have seen with outbreaks of Nipah and Hendra viruses).

A new study delves further into Ebola in pigs. Instead of using the Reston strain, they use the much-more-deadly Zaire strain. This is the one that movies are made about; the one which can cause outbreaks so nasty that they kill up to 90% of those who are infected. Why use Zaire instead of the Reston strain–the one which has actually infected pigs in nature? Well, the researchers wanted to find an animal that’s easier to work with than primates (there are all kinds of very strict regulations when it comes to working with non-human primates), so if pigs could work as a good model for human Ebola disease, that would make studying the virus just a bit easier. (In any case, for any live Ebola work, it still needs to be done in a biosafety level 4 environment, meaning complete spacesuits and the whole works).

The authors did 2 studies. In the first, they inoculated 6 pigs with Ebola Zaire, via a combination of intranasal, intraocular, and oral routes of infection. (Interestingly, no injection, which can be a key way Ebola is spread). They had an additional 2 pigs that they inoculated the same way with a saline solution, and housed them separately from the Ebola-inoculated animals. The goal of this experiment was to look at the pathogenesis of a virulent Ebola strains in the pig model. The infected animals all developed fevers and respiratory disease, with some internal hemorrhaging and evidence of airway replication by Ebola. Infectious virus was found at low levels in nasal washes and oral and rectal swabs; one animal also had a low level of virus in the blood. Higher levels of virus were found in various organs, including the heart and bladder, while the highest levels were found in lung tissue.

In the second experiment, they inoculated 3 new pigs in the same fashion, but then added in 4 additional (uninoculated) animals to stay with them, and kept 2 additional control animals in a separate area so that they could investigate pig-to-pig transmission of the virus. They did find viral RNA from the mucosa of all contact animals, and infectious virus was detected from 2 of 4, demonstrating that the virus can be passed among pigs. Not stated in the article was if the authors thought this was due to direct contact with respiratory secretions among the pigs, or via airborne transmission (a much more concerning route of transmission, as in humans, Ebola Zaire doesn’t seem to transmit well via air–typically it’s spread via close direct contact and bodily fluids).

Notably, pigs didn’t seem to develop severe systemic disease from Ebola, as primates do–the main symptoms exhibited were respiratory, which the pathology supports (finding little virus in the blood, but a lot in the lungs). This suggests that even for Ebola Zaire, infection in a pig could be mistaken for other respiratory diseases, such as influenza or PRRS virus (porcine respiratory and reproductive syndrome virus, which the initial pigs in the Phillipenes were co-infected with). So, Ebola may be circulating even more than we realize in the pig population, disguised by its commonplace symptoms.

A commentary published in tandem with the research article ponders the issue of foodborne Ebola, suggesting that this is a remote possibility and noting that butchering infected animals in the wild in Africa has certainly spread the virus. However, solely eating meat as a means of infection hasn’t been reported, and cooking likely destroys any risk (similar to influenza viruses). Like influenza virus, Ebola doesn’t seem to survive long in most environments, but it’s also noted that differences in African food storage (with little refrigeration) versus more typical cold storage may affect that as a risk factor, possibly prolonging the life of the virus when held in the cold. I think foodborne transmission is unlikely, but it can’t be completely ruled out right now.

Because of the respiratory symptoms, does this mean Ebola could enter the population via meat from animals that farmers don’t consider very ill, or put butchers at a heightened risk of infection during slaughter? This to me is more concerning than simple foodborne transmission. With Reston, at least no human symptoms have been observed, but if pigs (and potentially other animals?) can present with Ebola Zaire as a rather generic respiratory infection…well, that could spell trouble in a lot of different ways. It means that telling individuals to simply avoid sick-looking primates (and bats) is going to be even more woefully inadequate than it already is. Plus, it raises the remote-but-not-completely-outside-the-realm-of-possibility of someone intentionally spreading the virus via animals that are infected in this manner.

Science fiction? Maybe. Probably. Hopefully. But this research opens the door on many new lines of investigation and once again, raises even more questions.

Kobinger GP, Leung A, Neufeld J, Richardson JS, Falzarano D, Smith G, Tierney K, Patel A, & Weingartl HM (2011). Replication, Pathogenicity, Shedding, and Transmission of Zaire ebolavirus in Pigs. The Journal of infectious diseases PMID: 21571728