Fork me on GitHub

Even pathogens hate a cheater

I would like to apologize for the long delay since my last post. The excuse (I keep telling myself) is that, having already written too many computational articles, it was time to prove that I could write about biology too. Unfortunately I'm not nearly as good at reading the literature as I should be. Anyway, it's done. You can stop complaining now.

One barrier to engineering bacteria for biofuel production or any other human endeavor is that evolutionary rates are scaled by population sizes and growth rates. For an organism with massive population sizes (trillions of individuals or more) and doubling times on the order of hours, evolution can occur quite quickly. Genetic variants within the population which are capable of growing faster will quickly take over. For an organism which is, for example, wasting a huge fraction of its energy producing your future gasoline, you can bet this months graduate stipend that mutants which put more resources into growth will sweep the population.

See kids, cheaters do win in the end! In the above example the microbes are "cheating" the metabolic engineers out of their valuable chemical products, but the same sort of evolutionary trend: favoring fitness over (and often at the expense of) any other metric we can think of, has big implications for other bacteria as well. My current research explores the impact of selection on growth rate versus resource use efficiency, but I'll write more about that some other time.

Many microbial populations engage in cooperative behavior. Salmonella Typhimurium, for instance, a human pathogen, uses a type III secretion system (imagine bacteria using tiny syringes to pump toxins into your cells) to induce wide-spread inflamation in the gut. By inducing enough inflamation, a population of Salmonella can make a niche for themselves in which they outcompete the non-pathogenic (commensal) bacteria already present. Those syringes are expensive, but without them the Salmonella can't colonize at all, so everyone works together for the common good.

Unfortunately (for Salmonella at least) evolution doesn't really care about the common good. Every so often a mutant will come along which has lost the type III secretion system genes, or doesn't express them as frequently or at such high levels. Since this new strain doesn't have to waste resources on producing the costly virulence factor, but can still take advantage of the inflamation being induced by all his straight-laced bretheren, he grows faster than all the normal ("wild-type") Salmonella. Even when it's almost entirely cheaters and the population's ability to inflame the host is greatly reduced, even when they're all struggling to outcompete the commensal bacteria, the mutant is still a little better off because it's not wasting resources on that silly syringe.

The cheater wins again! In this example it's not the metabolic engineers that are being deprived of their liquid fuel, but the wild-type Salmonella which is being cheated out of its public resource: inflamation. Of course winning for the cheater also means that the entire population is wiped out. Without sufficient inflamation Salmonella can't survive, but up until that last cell is killed or expelled, at least the cheater was doing better relative to the wild-type. And, in the short-term, that's what winning is in the Game of Evolution.

"But wait just a minute", says the astute reader, "why have I been cooking my eggs all these years if Salmonella can so easily be wiped out by this internal rebellion?" Aha! That's the exact question raised by Médéric Diard and co-authors in their recent paper, "Stabilization of cooperative virulence by the expression of an avirulent phenotype."

What they found was that wild-type Salmonella Typhimurium was playing chess with the mutant, sacrificing a Knight in order to take the King. Diard et al. discovered that a fraction of the wild-type Typhimurium population was not expressing the costly virulence factors. These are not cheaters; their genotype is identical to cooperators which are expressing the type III secretion system. Their phenotype, however, is identical to the cheaters. By choosing to not expend the energy needed for the construction of those microscopic syringes, they achieve the same fitness as the mutant.

It's not a complete equilizer. Their daughters still have some probability of expressing the virulence factors (or else they might as well be real cheaters), but their fitness deficit is not as great. This boost was bought, however at the price of a reduced population size. Since only a fraction of their ranks are inducing host inflamation, the total virulence of the population is reduced.

"But wait again", you might point out, "just because the fitness benefit of cheating has been minimized does not mean that it has been eliminated entirely. The cheaters will still take over the population in the end, no?" You're right, cheaters will still grow faster in the long run, since they never cooperate, but it will require additional time for them to take over.

The rate at which an allele fixes in the population is positively related to the relative fitness of that allele and therefore the cheater will not fix as quickly. In a pathogen like Salmonella, the ultimate goal is to infect a new host. If the cooperators can sustain a sufficiently large population for long enough, their host is likely to pass them on to another unsuspecting (and un-hand-washing) individual. The cheater's success was all for nothing; without the type III secretion system, the mutant will never colonize another host. While they may have won the figurative battle, they lost the war. They've painted themselves into a corner. (Insert additional cliches here.)

Diard et al., in their article, do some fantastic simulations, showing all of the dynamics at play. I encourage you to check it out. One of the most exciting sentences is the second to last (see, a snoodier blogger would have said "penultimate"), the authors point out that this tendency for mutants to out-compete the wild-type pathogen, and therefore reduce the negative effects on the host, suggests a potential therapy: treat those suffering from Salmonella with a large dose of cheaters!