Virulence: A positive or negative trait for evolution?

10 June 2009

1856663523_cffa76bfbc_mWith just 141 confirmed deaths so far, an interesting question is whether the 2009 H1N1 influenza virus could mutate into something more lethal (“How a Mild Virus Might Turn Vicious“). Of course it could – but is it beneficial for the virus?

A fundamental principle of viral evolution is that viruses must spread from host to host to maintain the viral population. A virus spreads only if an infected individual passes the virus on to more than one new host. Furthermore, infection can spread only if population density exceeds a minimal value.

Some scientists believe that increased viral virulence reduces transmissibility. When infected hosts die faster, exposure to uninfected hosts is reduced. According to Ian Lipkin:

“A really aggressive flu that quickly kills its host” – like SARS and H5N1 avian flu – “gives itself a problem”.

According to this hypothesis, virulence is selected against as the virus spreads in humans. This idea leads to statements like this one:

In the last year, dozens of H5N1 cases have been confirmed in toddlers, almost all of whom have survived – which led some experts to speculate that those are cases of a less lethal version of H5N1 that is better adapted to humans.

Why is reduced lethality equated with being better adapted to humans? And how could the virus become better adapted to humans when human to human transmission has been minimal?

There is insufficient evidence to conclude that increased viral virulence leads to reduced transmission. For example, the 1918 influenza virus strain was extremely virulent, yet spread very efficiently among humans.  SARS and H5N1 influenza aren’t good examples – SARS transmission was probably stopped by containment efforts, and H5N1 influenza virus hasn’t transmitted well among humans, if at all.

In today’s highly crowded and mobile society, even a very lethal virus can be transmitted well. Acute viral infections are preceded by an incubation period, during which virus is shed but symptoms are not yet severe enough to lead to hospitalization. And even a highly pathogenic virus will cause mild or no disease in some individuals – further increasing the chances of spreading infection.

It seems more likely that increased viral virulence could lead to better transmission. For example, a more virulent influenza virus might cause more coughing and sneezing, which would be more effective in transmitting infection. Perhaps we should focus on transmissibility, not virulence, as the property that drives viral evolution. Viruses evolve so they can be efficiently transmitted to other hosts. According to this hypothesis, any other properties that accompany transmissibility, such as virulence, are secondary effects. If this idea were true, then all viruses would evolve to be maximally infectious and avirulent. But this is not the case. Perhaps, as Peter Palese said, viral virulence has unknown benefits:

“Look, I believe in Darwin. Yes, the fittest virus survives. But it’s not clear what the ultimate selection parameter is.” A mutation that confers lethality, he explained, may confer another advantage scientists have not pinned down.

  • http://viromag.wordpress.com ElDean

    IS THERE ANY CERTAINTY IN SCIENCE !, with all this long history with influenza, and with the virus in our hands, we cannot expect what is to come next, I think this is not under any human capability, though I think we should focus on surveillance, my question is: if the virus shifted to a more virulent one, should we just wait and see people die to conclude that it became more virulent !!! aren't there any very early – and I mean VERY EARLY – signs for this kind of shift ?, mmm pig farms surveillance, wild birds, or even if we can make experimental infections with isolates of early next winter and compare virulence with the current circulating one !? … what do you think ?

  • http://scienceblogs.com/effectmeasure revere

    I agree completely with this post. There is a tendency to automatically assume pathogens must moderate their virulence as they establish a “working relationship” with their hosts. We have a number of unfortunate examples where this isn't the case, including HIV and smallpox. I looked at this from the standpoint of viral dynamics here: http://scienceblogs.com/effectmeasure/2006/09/m

    I enjoy your blog. Good work.

  • http://www.virology.ca/ Chris Upton

    Given that the 1918 flu spread rather well, I don't think we have to waste time debating whether a more severe flu can spread efficiently. Something as pathogenic as 1918 would be bad enough; it would be a big problem. I suppose someone may be worrying about a 50% CFR for H5N1.
    As to what happens to the current swineflu, I think if this is regarded as a very low CFR then if it changes, then it's more likely to become more pathogenic than less pathogenic. There's just more space at that end of the variability spectrum.
    I'm concerned about the development of a swine-h1n1 – human (old)-h1n1 reassortment.

  • Ilipkin

    Controversy makes good copy but can do disservice to the parties involved and the audience. No one knows how this story will unfold. The question posed was whether I agreed it was likely that this virus would become more virulent. I presented both sides of the argument.
    Ian Lipkin

  • phytosleuth

    Or infectivity versus lethality:

    Lethality of H1N1 Influenza Virus Increasing According to Latest Analysis of Virus Peptide Genomic Data

    http://news.prnewswire.com/DisplayReleaseConten

  • http://www.virology.ws profvrr

    I don't see why there is more space at the more virulent end of the
    spectrum. And why are you concerned about reassortment with Russian
    H1N1? Both vintages of H1N1 appear to have about the same lethality.
    Unless you are thinking that a reassortant could have greater
    virulence than either virus?

    On Wed, Jun 10, 2009 at 6:54 PM,

  • Phillip Huggan

    For mutations negative. But for antigenic shift positive, in retrospect from our observational PoV. If we see a novel shifted virus killing members of a species it is good for the virus because the crappy immunity probably signals the virus has successfully colonized a new species. IDK which forcing is stronger.

  • http://www.setbb.com/fluwiki2/viewforum.php?f=7&sid=c1408b5cf8a576350c571da5192e08b5&mforum=fluwiki2 gsgs

    yes, the reassorted virus can be worse than each of the parts, why not ?
    Virulence is often just only a small change. H5N1 had reassorted in birds to become what it is.
    Fujian flu(H3N2) 2003 was worse than its parts. 1957,1968, reassorted with harmless avian flu.
    I feel that this Mexflu is not yet optimal, it had not had much opportunity yet
    to check what's out there…let's hope that by the time it starts to reassort
    (e.g. with other swine H1N1) we have some immunity already

  • Pingback: Is virulence good or bad for viruses? | Swine Flu Daily Update

  • MBM

    Thanks for your excellent blog and podcasts. I’ve been thinking about this whole virulence thing on and off all day and I’m afraid I’ve ended up with a bit of an epic (my first post too). Not a biologist, so please tell me off if I’m being dumb.

    I know we should be wary of talking about what a virus does or does not ‘want’ but I find this way of looking at things quite helpful when thinking about evolutionary problems (‘selfish’ genes and all that) so I would say that *all* the virus wants to do is transmit, and it will follow any nearby evolutionary gradient that improves its transmissibility. Although virulence is a big issue for us (the hosts) it is only really a secondary consideration for the virus- it only cares about whether or not it kills or otherwise debilitates us in as far as this may or may not improve its chances of transmission.
    I would guess that the holy grail for a virus would be to infect its host, for its host to stay alive for as long as possible, and for the virus to remain highly transmissible throughout the rest of the host’s long and fruitful life. However, this is generally not going to be achievable.
    As a virus you are going to have to debilitate your host to some degree in order to get transmitted. At the very least you have to appropriate some of your host’s resources to get copies of yourself made, and this will generally have damaging side effects. You may also make your host do other costly things that are beneficial to your spread like inducing infectious secretions or even behavioural changes, and you may well be required to sabotage or otherwise knacker your host’s immune system even to stand a chance of transmitting before you get shut down. The more of this sort of thing you do the more you increase your transmissibility, but if you go too far down this road- or at least too far too soon- you may debilitate your host to the point of death and bring your infectious reign to an untimely end.
    Now, I am arguing that transmissibility is the main feature here (Nature will select viruses based primarily on their transmissibility), so I would expect that over time a virus will evolve to a local optimum of transmissibility and will then probably just sit there. For a ‘new’ virus parachuted in from another species the chances are that it will initially be fairly far from the nearest local optimum of transmissibility (in its new host) and may therefore have quite a lot of potential for evolutionary change before it gets there. However, the side effects of this change need not necessarily be increased virulence. For example, if you are initially very pathogenic in your new host and kill it very quickly then I would guess that one obvious way to increase your transmissibility (all other things being equal) would be to kill your host less quickly. On the other hand, if you start out being fairly benign (compared to other similar infections in your new host) then I would guess that there would be quite a bit of unexplored potential for increasing your transmissibility through various strategies that would probably be bad news for your new host?

  • http://www.virology.ca/ Chris Upton

    What if a reassortment put PB1-F2 into the swineflu?
    Or “human” H1N1 NS1 from segment 8?

    Ian, sorry if it came across as a rant.
    I totally agree that we can't guess what's going to happen.

  • http://www.virology.ws profvrr

    Chris, the predecessor to 2009 H1N1 in pigs does make PB1-F2.

  • http://www.virology.ws profvrr

    Your ideas are on the right track. I agree that, from our viewpoint,
    transmissibility appears to be the main driver for viral evolution.
    But there is a lot we don't know. For example, increased (or
    decreased) virulence might confer secondary advantages that are not
    obvious to us. As I wrote, all viruses do not evolve to be maximally
    infectious and minimally virulent.

  • http://www.virologyj.com/content/6/1/69 gg

    You may want to check this out:
    Protein intrinsic disorder and influenza virulence: the 1918 H1N1 and H5N1 viruses
    http://www.virologyj.com/content/6/1/69

  • Pingback: Pandémie, mais pas de panique | Le blogue de Valérie Borde

  • ARR_3

    Agreed. Viruses (as for all organsims) have evolved strategies that maximise their basic reproductive rate and this might be achieved with either a low, a medium or a high virulence strategy – depending on the virus, the host environment, mode of transmission, and external selection pressures. For example, some viruses (eg. baculoviruses) actually need to kill their host in order to be transmitted by breaking down the host cuticle by expressing chitinase enzymes to release progeny virions for uptake by susceptible hosts (eg. insect larvae feeding on virus contaminated foliage).

    One possible fitness advantage for influenza viruses with high virulence (lethality) is to minimise the potential for aquired immunity (antibodies) to be spread within the host population (from mother to offspring)?? Indeed, it seems pregnancy is a notable risk factor in the current (highly transmissible) swine H1N1 pandemic…

    Possibly Macabre but also an interesting possibility don't you think?

  • ARR_3

    Agreed. Viruses (as for all organsims) have evolved strategies that maximise their basic reproductive rate and this might be achieved with either a low, a medium or a high virulence strategy – depending on the virus, the host environment, mode of transmission, and external selection pressures. For example, some viruses (eg. baculoviruses) actually need to kill their host in order to be transmitted by breaking down the host cuticle by expressing chitinase enzymes to release progeny virions for uptake by susceptible hosts (eg. insect larvae feeding on virus contaminated foliage).

    One possible fitness advantage for influenza viruses with high virulence (lethality) is to minimise the potential for aquired immunity (antibodies) to be spread within the host population (from mother to offspring)?? Indeed, it seems pregnancy is a notable risk factor in the current (highly transmissible) swine H1N1 pandemic…

    Possibly Macabre but also an interesting possibility don't you think?

  • Pingback: The problems with Barry’s “The Great Influenza”

  • Pingback: The D225G change in 2009 H1N1 influenza virus is not a concern « FluTrackers

  • Rohannarayanm

    the recombinant strain of AH1N1 was formed as a result of mixing up of  rna segments of influenza viruses in 2 diff hosts. dont u think that was  a chance event rather than a part of evolution?

  • Mikegreen200691

    what is important to undeerstand is that a virus has no consious agenda when entering a new host, it is incapable of not having a specific effect on the hosts health whether it is a particularly virulent effect or not, the benfit for a virus in being more aggressive when infecting a host is that it reduces the population to a selection of particular organisms who are able to deal with the effects of the virus. This means the population can then expand only with the genetic capacity to resist the effect of the virus. This means the virus can then replicate without causing any real harm to the population. The advantage for the host is that when it comes into contact with a species who have not encountered this particular virus, the virus aggressively kills off the potentially dangerous species.