Can vaccination increase virulence?

October 5, 2006 6:07 PM

Why can't vaccination encourage the evolution of viral potency?
posted by freebird to science & nature (16 comments total)

Because the virus in a vaccine is dead.
posted by orthogonality at 6:09 PM on October 5, 2006

Well, vaccines can be attenuated rather than dead I think. But even so - I don't mean in the introduced virus. I mean by creating a selection pressure on wild-strain virus in the population at large, why can't it encourage selection for mutations more able to overcome the immune system?
posted by freebird at 6:19 PM on October 5, 2006

Not necessarily, orthogonality.

(Although I'm not suggesting this would impact viral evolution.)
posted by docgonzo at 6:20 PM on October 5, 2006

To understand this question, you have to understand two seperate biological phenomena. The first is how vaccination works:

Normally, when infected with a virus, it takes the immune system some time to recognize the invader and ramp up production so that it can kill the invader. Vaccines stimulate the body's immune system to recognize certain antigens. This means that the body is primed and ready to fight off any invasion by similar viruses.

The second set of processes that you need to understand is mutation natural selection:

Natural selection is a simple concept that can be summed up as 'survival of the fittest', where the fittest are those who produce the most offspring. In terms of this question, it means that viruses that are recognized by the immune system are quickly destroyed and don't get a chance to reproduce.

Now, if a virus was to develop a mutation that made it resistant to the immune response, that virus would have an selective advantage. While the other viruses crashed and burned, it would be happily hijacking your cells to churn out copies of itself.

Now, the final thing you have to understand is that generally speaking, mutations are not directed. That is, a virus can't say "Oh no! I'm under attack! Time to start mutating!" Mutations are mostly the result of random DNA replication errors. So, the development of a mutation that a) confers resistance and b) doesn't compromise the virus's integrity will be very rare.

This is why flu vaccines work for a while. The predominant strain of the virus is unable to spread very far, and until one of these rare mutations occurs and the new strain is able to get a foothold in the population, we're protected.
posted by chrisamiller at 6:30 PM on October 5, 2006 [1 favorite]

I mean by creating a selection pressure on wild-strain virus in the population at large, why can't it encourage selection for mutations more able to overcome the immune system?

Well, maybe it can. But in order for that to happen, there needs to be way for viruses to at least in theory avoid being killed by antibodies. I don't think it's possible for them to have a general defense mechanism against antibodies the way that bacteria can have a general defense against antibiotics.

So assuming there is no defense against antibodies, your question is kind of like asking why autoclaving doesn't cause the evolution of bacteria that can withstand intense heat, because obviously no bacteria could ever do that.

But on the other hand, obviously some viruses can defend themselves against the immune system, and the AIDS virus does this by actually attacking the human immune system itself. Even though people develop antibodies against the aids virus, they can't wipe it out. But I doubt that your average flu virus could suddenly evolve into a new form of AIDS overnight.

Another mark against this theory is that historically, vaccines do not cause more hard-core variations, in fact, it causes the opposite effect. If people get vaccinated, then transmission becomes more difficult, and the only way for the virus to survive is to be less virulent, and let the host live long enough to come into contact with non-vaccinated people. It's the same with any countermeasures.
posted by delmoi at 6:32 PM on October 5, 2006

It's in part because of the mechanism of a vaccination -- a vaccination primes your immune system to recognize and kill viruses. This means that the mechanism that actually kills the virus is not the vaccine but the phalanx of t-cells that virii have been dealing with for millions (or so) of years already. Also, the immune system is pretty robust -- it can adapt to small changes quite quickly. The mechanism of the t-cells are fairly "brutal" (at least to the virus). It encircles it and digests it into parts.

This is in contrast to a antibiotic or other anti-virals, that the effect of the drug is to interrupt some process directly in the bacteria (or virus's lifecycle). This provides a much easier target for random mutations to happen to hit, and also a higher selection pressure -- the antibiotic / antiviral is a very new mechanism, and one that does not adapt to the virus's response.
posted by printdevil at 6:33 PM on October 5, 2006 [1 favorite]

A billion viruses evolve faster than a thousand do. All those juicy replication events in live cells is where the errors are introduced into the virus' genetic code.

If vaccination reduces the amount of "viral habitat" a million-fold, the next successful virus may well avoid antibodies, but the emergence of that virus will take much longer.

Also, there's no simple way for a virus to evolve completely outside the scope of the immune system. Even HIV engenders an immune response before it destroys the host immune system, and researchers are working on ways to boost this response enough to prevent the virus from ever getting a foothold.

The human immune system is a devastatingly clever little biological security network; you might enjoy reading more about it.
posted by ikkyu2 at 6:36 PM on October 5, 2006

I think you're right in that a vaccine would shift the makeup of a population of viruses in a direction (those which are not affected by the vaccine or who have evolved new defense mechanisms will survive).

However, ability to survive a vaccine attack would not necessarily be correlated with virulence. It's possible that for some reason the weakest strain of whatever virus the vaccine is for will survive (because of some mechanism that the strain does or doesn't have).

It's for this reason that evolutionary biologists like say that evolution doesn't have a direction, the population makeup just moves in whatever direction the pressures push it in.

An example that's been played out in recent years is the discovery that sickle-cell anemia gives malaria infected people a greater chance of survival from the malaria. Thus, the people that survive malaria are not necessarily the "strongest" or "most virile", they just happened to survive because of an off chance. I think vaccines and viruses would probably work the same way. See here for more info.
posted by jourman2 at 6:37 PM on October 5, 2006

So is it accurate to say viral evolutionary change can be (and is) induced by boosted immune response, it's just that:

a) it's much harder (since it involves beating the dynamic & complex immune system rather than a static antibiotic), so succesful mutation is rarer hence evolution is slower.

b) it's only a "temporary win" for the virus, since the immune system is able to adapt unlike antibiotics and drugs.

Or is the "energy barrier" to successful evolution just too high, as (I think) ikkyu2 suggests?

On preview - jourman2, that's an excellent point. I don't mean to conflate the ability of the virus to survive with the magnitude of its effects on the host. My question should be clearer - I'm only asking about the virus' ability to survive and propagate, not on its effects.
posted by freebird at 6:47 PM on October 5, 2006

The polio virus is an interesting example of a virus that may or may not have become more virulent at various times, for reasons we do not fully understand. It spreads rapidly, some theorize through water sources, in mysteriously epidemic outbreaks, and then, just as mysterioiusly, "goes to ground" again. Epidemiologists theorize that vaccination program failures are responsible for the continued success of the virus in the wild, but others are not so sure.

There has been an annual goal to eliminate polio from the world, as smallpox has been eliminated, since 1988. Every year, all member nations of the WHO agree that it should be possible to eradicate the last few hundred human cases of the disease, and vaccination programs are conducted regularly (with the notable exception of failures in Africa in 2003 due to Muslim resistance). And yet, the wily old polio virus survives, and this year, to date, has claimed another 1306 victims, slightly more than in 2005, when outbreaks in non-endemic nations like the United Arab Emirates reminded everybody how quickly this old foe can spread, in a jet connected world.

Some viruses like polio are deceptively simple things, which we have thought we knew enough to control, by vaccination and sanitation, for more than 50 years. And yet, it lives, in us, and in the wild, still, even as again, this year, we plan its demise for next year.
posted by paulsc at 6:59 PM on October 5, 2006

I'm only asking about the virus' ability to survive and propagate, not on its effects.

In practice, a virus' replicative potential is closely tied to its pathogenicity, as its effect -- to the host -- is most often tied to the concentration of the virus in the host and its colonisation of host tissues.
posted by docgonzo at 7:01 PM on October 5, 2006

I'm only asking about the virus' ability to survive and propagate, not on its effects.

It's not that easy to seperate the two. Ebola virus is famously (and gruesomely) deadly, but because it kills its hosts so fast, it doesn't have very much time to spread. Ebola is rare and doesn't survive very well as a result. Conversely, AIDS acts very slowly, so it remains in the body and has ample time to spread to new vectors, which is why it's an epidemic.
posted by chrisamiller at 8:07 PM on October 5, 2006

Look at it this way: most viruses are killed by your immune system. There is already pressure on them to spread rapidly and effectively, because their time living in their host's body is limited (if for no other reason, then because the host will die eventually). Your immune system is capable of preventing infection from ever taking hold, but there is no guarantee. A vaccine makes it harder for viruses to meet the pressure that's already on them from your existing immune system. As ikkyu2 points out, it also cuts down on the ecosystem that they have to work with. So if anything, vaccines should slow down the evolution of viruses.
posted by Humanzee at 8:29 PM on October 5, 2006

A vaccine makes it harder for viruses to meet the pressure that's already on them from your existing immune system.

But making things harder for a population (up to the point of actual or near extermination, which seems like sort of the question here) will generally accelerate evolution, right? As will reduction of available ecosystem?
posted by freebird at 10:24 PM on October 5, 2006

It depends on what exactly you define as "evolution". If you mean simply the (non-directed) process of random mutation and selection, then Humanzee is right--vaccines should slow viruses' rate of evolution. The rate of evolution, as measured in absolute time, is critically dependent on mutation rate and generation time, which determines the number of progeny. The mutation rate of a particular virus is unlikely to change, for a variety of reasons. Since I'm procrastinating from writing my dissertation, I'll go into them.

The mutation rate of DNA (or in the case of some viruses, RNA) is due to environmental factors like carcinogens and also to the inherent error rate in the machinery that replicates the genetic information, which are DNA or RNA polymerases. DNA/RNA polymerases aren't perfect, which is actually a good thing because otherwise there would be no mutation, and thus no evolution (discounting environmental factors). Viruses generally encode their own polymerases, and these tend to have a higher error rate than those of their hosts (in addition to the fact that RNA polymerases are already error-prone compared to DNA polymerases).

This is because viruses depend on mutation to respond to their environment, as they can't detect things or move in the animal sense. An immobile, long-lived organism like a tree has particular defense mechanisms (alkaloid poisons, thorns, etc.), but a virus responds to threats by making a whole lot of progeny very quickly, and by relying on their error-prone polymerases to give most of these progeny different (though not necessarily beneficial) characteristics, some of which may help evade the host immune system.

One might think that having a much more error-prone polymerase would increase the rate of evolution, but this is wrong. If the error rate gets too high, you reach what Manfred Eigen called "information catastrophe", which is when the polymerase makes so many mistakes that the mutant proteins produced can't carry out the necessary functions to put together a functional virus particle. So most viruses' polymerases have reached an error rate that is probably just below the brink of information catastrophe. Any more mistakes and the virus can't make progeny at all; any fewer mistakes, and the virus loses its best defense against the immune system, which is mutation so as to no longer be recognizable by the adaptive immune response. So the mutation rate due to error is unlikely to change.

This leaves generation time. A flu vaccine is usually composed of attenuated versions of three or so strains of flu predicted by groups such as the CDC or WHO to be most populous in a given season, on the strength of the previous year's data. That is, the vaccine is letting host immune systems sample a strain before it actually had a chance to infect them. If a vaccinated person is infected by that strain, their immune system will rapidly surround those viruses and/or kill cells infected by them.

But, as your original intuition led you to suspect, the early progeny made in the initial infection may fortuitously possess mutations that allow them to go "under the radar", so to speak. They might have a much slower life cycle, such that the cells they've infected don't display to the immune system that they've been infected, or in the case of some viruses, favor a mode that allows them to integrate into the host genome and lie latent. (We're all infected with some kind of herpes virus.) These viruses have evolved, yes, but in such a way that increases their generation time, thereby lowering the number of progeny they make. In that sense, they have evolved in response to vaccination in a way that lowers their rate of evolution.

If by "evolution" you mean a particular outcome, such as the emergence of virulent strains, then it's a tougher call. Let's take the bird flu: the event we fear most is that an already virulent strain of avian influenza will become zoonotic; that is, will acquire some set of mutations that allows it to infect humans. The chance of a virus acquiring that set of mutations is quite rare as it is, but in order for it to jump to humans before it dies, there has to be a human present. This introduces a purely external, environmental variable into the equation. A bird flu virus particle capable of infecting humans may have already come into existence and passed out of it without ever coming into contact with a human. Obviously, if chickens are housed in close quarters with humans, the chances of zoonosis occurring is much greater. However, if one were to vaccinate all poultry against bird flu, the rate of evolution would be reduced, and there would certainly be much less chance of a zoonotic bird flu. It's just that if this vaccination were combined with more dense housing with humans, it's hard to say what the final chances of a specific outcome (zoonosis) occurring are.

And now I must really return to writing my dissertation. Which is not about viruses.
posted by ObeyScient at 10:52 PM on October 5, 2006 [2 favorites]

mutations are not directed. That is, a virus can't say "Oh no! I'm under attack! Time to start mutating!" Mutations are mostly the result of random DNA replication errors.

A common myth. There's evidence organisms can 'save' accumulated mutations and deploy them during periods of high stress. Also, gene activation states are heritable. I don't know if either of these things have been observed in viruses, but pedantic descriptions of evolution seem unwarranted here, when a good model of heredity, and the vast majority of viruses, are unknown.

in order for that to happen, there needs to be way for viruses to at least in theory avoid being killed by antibodies

Evolution is one way. But also this assumes 100% of the herd has effective immunity 100% of the time. This is never the case, even if 100% of the population has been immunized. In fact the duration of immunity from vaccines into adulthood is not well understood (apparently adult vaccinations are not as easy to sell as childhood ones).

But I doubt that your average flu virus could suddenly evolve into a new form of AIDS overnight.

Interesting example -- influenza evolution does outpace vaccine development.

Another mark against this theory is that historically, vaccines do not cause more hard-core variations

What history? How many vaccinations did your parents get? Though smallpox was obviously a success, the story with the other vaccines isn't so clear. Several vaccines currently recommended for newborns weren't available in the '80s.

Meanwhile, history shows that viruses can become less virulent on their own. Herpes is thought to have been fairly fatal as recently as the Middle Ages.

If people get vaccinated, then transmission becomes more difficult, and the only way for the virus to survive is to be less virulent, and let the host live long enough to come into contact with non-vaccinated people.

Possibly. For viruses that are already minimally virulent, doing anything to disturb the situation might not be a good idea (chickenpox for example). Getting shingles because your vaccine wore off is going to suck.

A billion viruses evolve faster than a thousand do. All those juicy replication events in live cells is where the errors are introduced into the virus' genetic code.

A lot of host evolution takes place there too.

The human immune system is a devastatingly clever little biological security network; you might enjoy reading more about it.

Then why is the minimal ammount of information in a shot so crucial to it?

The polio virus is an interesting example of a virus that may or may not have become more virulent at various times, for reasons we do not fully understand.

A recent case in Italy may be a strain that 'escaped' from vaccine.

But making things harder for a population (up to the point of actual or near extermination, which seemslike sort of the question here) will generallyaccelerate evolution, right? As will reduction of available ecosystem?

As a general rule, yes.

The rate of evolution, as measured in absolute time, is critically dependent on mutation rate and generation time

And selection pressure.

In that sense, they have evolved in response to vaccination in a way that lowers their rate of evolution.

However the survivors share an adaptive trait -- the one in question. In the normal case, there are more mutations but no selective pressure in the direction you're measuring (which is the only one you can really speak of... even if the viral genome is unchanged, it could still have evolved with respect to changes in the host).

The mutation rate of DNA (or in the case of some viruses, RNA) is due to environmental factors like carcinogens and also to the inherent error rate in the machinery that replicates the genetic information

And cosmic rays, uv, etc.

-Carl
posted by clumma at 9:43 PM on October 12, 2006