Sunday, October 5, 2014

Why Ebola Airborne Mutation(s) are "Highly Unlikely": Let's Talk Mutation and Natural Selection

Note: This post is not an exhaustive treatise on the evolution or adaptation of filoviruses, but an applicable discussion on certain principles.

When Ebola experts say that mutation to become airborne is “highly unlikely”, they are basing this on what they know, as experts, about Ebola molecular virology, cumulative transmission data and natural selection. While they’ve all said it, I think Heinz Feldmann, one of the foremost experts in Ebola molecular virology, said it best in his interview with Kai Kupferschmidt (17 September 2014):

Q: "There has been speculation that the virus could mutate or has already mutated to spread more easily. How likely is that?"

A: "I don’t think there is any data right now to support that. If you look at the virus sequence, it falls within the normal range of Zaire Ebola strains. Of course any of these mutations could have a dramatic effect, which we don't know right now. But there is nothing obvious that would point to a more transmissible, more virulent virus, or a change of transmission route.

You can speculate in every direction, of course, but I think it should be fact-based, it should be data-based, and I think it makes absolutely no sense to bring in aerosol transmissibility as a potential. I think this is really not helpful, unless you have data to support that."

What he is saying there is that he knows Ebola genetics and he doesn't see anything in the new genetic data that indicate the virus is mutating in that direction. Being an excellent scientist, he also acknowledges that we don’t know what all changes mean yet, but based on the data he sees, there’s no evidence to speculate that this Ebola variant is becoming airborne.

Aside from the genetic data they have, Ebola scientists consider the way in which Ebola is already successfully transmitted, as well as Natural Selection.

So let’s talk about Natural Selection for a little bit. In his 1930 book, “The Genetical Theory of Natural Selection”, Ronald A. Fisher made this important distinction, “Natural Selection is not evolution.” It’s “…a convenient abbreviation for the Theory of Evolution by means of Natural Selection…” We will be coming back to Fisher soon, but I wanted to make sure we were all on the same page. So as far as Natural Selection goes, I think Dr. Laurence Loewe described it really well in his educational piece for Nature (1) when he said, “…selection constantly sorts through the variation that is produced by mutations to select the fit and remove the unfit, while ignoring neutral changes.”

Fitness refers to an organism’s ability to compete in its environment in so far as survival or reproductive ability. Can it eat? Protect itself? Have fit offspring? During replication of any organism’s genetic material mutations are possible. For viruses, the speed of replication makes the rate of mutations higher than for a human.(2)  And mutations can be classified with regard to how they affect the organism’s fitness. When a mutation provides a selective advantage in this regard, it will be selected for; the frequency of the genotype (genetic profile) containing that mutation will increase. If a mutation is deleterious, it will be selected against and the frequency of that genotype will decrease.

So what drives mutation? Well, Natural Selection moves an organism toward an adaptive optimum; the point at which the organism would be most fit in its environment. This doesn't mean that Natural Selection causes mutations, but that it determines whether or not a mutation would be kept or discarded. And Fisher thought this occurred in small steps rather than in large leaps. He thought that small advantageous mutations would happen more often than large ones, because large mutations would be more likely to have negative side effects. Well, a group studying RNA viruses actually found evidence to support this theory: they showed “…that adaptation proceeds by multiple mutations, but that among the set of all possible mutations, adaptation proceeds by the subset of mutations with small effects.”(2)

This supports what Ebola experts have said, that it’s not likely that a single mutation would cause this virus to go airborne, and the likelihood of the virus acquiring all the needed mutations at the same time is extremely low.

In addition, there’s no selective advantage to becoming airborne for Ebola. It is clear that Ebola’s current mode of transmission is working well. The virus is spreading and reproducing successfully. This means there is no selective pressure on the virus to keep any mutations that might confer airborne transmission, especially when this kind of significant change would most likely bring with it some serious deleterious side effects. And while some argue that this is the first time Ebola has been replicating to this extent in humans, we need to keep in mind that it’s been doing this for far longer in the wild in the primate species it infects and in its bat reservoir host. And yet, this mutation hasn't happened.

[edit 10/6/14]: WHO agrees with me and has also pointed out today, that viruses entirely changing their transmission mode is not something we've seen before. They must have read my blog. ;-)]  

So, can we say it's impossible? No. But I'm not losing any sleep over it.





  1. What traits does a virus have to possess in order to achieve airborne transmission? Does it have to be able to enter an organism through the linings of air passages or the lungs? Whatever it is, is there some known feature of Ebola that would have to change in order for this capability to develop?

    1. Hi and thanks for stopping by! We don't know the answer to this question. Maybe there needs to be more infectious virus expelled from respiratory cells, maybe a higher infectious dose is required to infect respiratory cells, or maybe the virus particle needs to be more stable physically while in an aerosol form to remain infectious over time and distance?

  2. I am not losing any sleep over it either but it is not an experiment I want to leave running for too long. I agree with everything you have said and think along the same lines but I have one niggling doubt.
    If you want to test a bacteria for drug resistance you apply extreme selection pressure by giving it the perfect growing medium and then try and kill it with increasing dosages of the drug. Any mutations that improve its chances of survival is selected for and you keep refining this group.
    Are we doing the same with Ebola? A patient with late stage EVD is going to packed full of virions looking for a host. As an RNA virus with very sloppy quality control this is going to be a genetically diverse population. Ebola needs a very low viral load to cause an infection. Good quality, correctly fitted PPE is going to act as a total barrier these virions. Normally mutations with major changes, even when they are viable are out performed by the wild type. In this case, if it occurs outside West Africa, there is no competition and it has a totally naive population to refine itself on before it meets the original strain.
    Put all those pieces together and those billions or trillions of virions that never found a host died but if at some point a mutation did occur that conferred the ability to get past PPE, due to improved droplet transmission or anything else, we don't have a second line of defense. There aren't enough PAPR with HIPA filters to go around.
    Heather if it does happen could I book a room inside your BSL4 lab, something ensuite with a good view.
    It was just a thought about how it could happen rather than something I expect to happen.

    1. Hi! And thanks for giving me a smile at the end of your comment. :) I agree that it's possible, but also that it's unlikely. If it did happen, it would probably cost the virus something as well.

  3. Are we being misled by language here? A less-viable strain is not going to disappear when it is "out-performed" by the competition, it's just going to reproduce at a lower rate - still enough, one assumes (otherwise it's a dead end we don't have to worry about in any case) to burst its host cell and send a trillion copies out into the patient's blood stream and mix randomly with all other strains. Then if a substantial amount of virus is communicated by fluids to the next patient it contains some portion of the less-viable (but not extinct) mutation.

    And the first time this is transferred through the air successfully it will be the only strain present in that victim, and from that point in we will have an airborne epidemic in parallel with the fluid-based.

    Of course, if this mutation leaves the new strain just barely able to explode out of a single cell over the course of a couple weeks, or some other length of time that gives the patient time to develop a successful defense before dying then we won't have any airborne epidemic. But that is the only assumption that extinguishes this strain - unless when fluid transfer takes place it does not contain a truly random selection of all the virus types in the patients body.

    This seems to make the situation a little more dangerous.

    1. Hi and thank you for stopping by! You are right in that there are many possible scenarios that could happen if the right set of mutations occurred simultaneously, and we have no way of knowing what the ultimate cost for the virus would be. The bottom line for me, and the reason I wrote this post, was not that it couldn't happen, but that virus mutations that change the entire transmission strategy of that virus are rare (I can't name one), and for a reason, so focusing our attention here isn't all that productive.

  4. I do have a question, I just left ICU to attend grad school in Fort Worth and therefore still have a lot of friends in the area still working the units. I don't fear catching Ebola, I don't think it will mutate suddenly, but I am very against the CDC idea that any hospital can care for an Ebola patient by throwing together training. ICU nurses are incredible and some of the brightest nurses there are, but I think the fluidity of the environment is such that mistakes can be easily made. In most instances those mistakes will never amount to anything, but with Ebola, I feel that the risk is a little higher. I just feel that there should be a certain level of training to be able to care for a patient and avoid self contaminating. I don't believe the Dallas nurse was careless, but I think there was a slip up that she was unaware of. I was wondering what your thoughts were?

    1. Hi and thank you for stopping by! I agree with you. Training is critical and not everyone is trained to handle Ebola patients.That is why the patients deliberately brought to the US were taken to very specific hospitals. I don't think she was careless either. Anthony Fauci, in an interview with NBC just said the same thing, they need better training and "'ve gotta have drills." I think people who are going to be working with an Ebola patient need to be able to demonstrate the correct way to don, use and doff PPE. I think each facility needs a very small team trained specifically for this and if a suspected patient comes in, no one handles them except this team, with on-call and hazard pay included. Fodder for a post I think. Thanks for the question.

    2. She was careless. It's been documented she breached protocol and had extensive contact with the patient outside of her suit.

    3. Hi John and thank you for the information. If documentation does in fact support that, then it's a real shame. There's very little room for error with this virus and people need to follow established protocols. Thank you for letting us know.

    4. I just saw the CBS story that shows the documentation. Thanks again John. This is a tragic case and I hope there are lessons truly learned for other healthcare professionals.