Kin selection, quorum sensing and virulence in pathogenic bacteria

Growth and virulence of pathogenic bacteria is often increased by the production of extracellular factors, released in response to density dependent cell-to-cell signalling, known as quorum sensing (QS). At low cell densities a high production of extracellular factors are not as beneficial and so quorum sensing provides a means of controlling this. However, other cells may exploit these extracellular factors and by doing so, gain the benefits of them without the cost of producing them. High genetic relatedness provides an answer, as the benefits and costs are shared across individuals who share the genes for production of extracellular factors (kin selection).

(Rumbaugh et al., 2012) examined bacterial infection in mice in order to study how relatedness (defined as the genetic similarity to your social partner compared to the whole population) influences selection for QS. They used two bacterial strains, both Pseudomonas aeruginosa, one was a wild type which both produces and responds to QS signal molecules, and second was a mutant strain which does not respond to QS signal molecules. They did this through two sets of experiments testing high relatedness (low strain diversity) and low relatedness (high strain diversity). Mice were infected, allowed 24 hours to incubate the bacteria and then euthanised. Their livers were homogenised and the bacteria from their livers were pooled together. Colonies were randomly chosen from the pool and used to infect the next group of mice; this process was repeated up to six times, effectively creating different subcultures at each pooling event.

In the high relatedness experiments, each mouse was only infected with one strain, which could be either the wild or the mutant strain. During the first infection stage, there was an equal amount of mice infected with both the wild and mutant strain. If one strain within a mouse had a higher growth rate than strains in other mice, this would contribute more proportion of that strain, compared to others, when the bacteria were pooled together. This means that bacterial strain would be more likely to be chosen to initiate the next infection stage.

In the low relatedness experiments, multiple (10) clones were used to infect each mouse. In the first infection stage there was always a 50:50 mix of wild and mutant strains in each mouse. As before, after killing mice, the livers were homogenised and the bacteria were pooled together. 10 clones, at a concentration of 10², were randomly chosen and used to infect the next group of mice. Each mouse was infected by a mixture of clones, this allowed for competition within the host, where the mutant strain could potentially exploit the wild type strain.

The results showed that the wild type strain was favoured under conditions of high relatedness, whereas the mutant was favoured under conditions of low relatedness. The mice of high relatedness which had been infected with the wild type compared to the mutant found that the wild type grew to significantly higher densities. Combining the results of this selection experiment with data they had already collected on the mortality rate in the infections of mice to Pseudomonas aeruginosa, they found that 100% of mice infected with only the wild type died after 5 days, 67% infected with only the mutant strain died after 5 days and 56% infected with a 50:50 mixture died after 5 days.

In the high relatedness experiments, the wild strain and mutant strain occurred in different mice, and so the greater growth of the wild type led to a greater frequency until it eventually reached 100%. In low relatedness experiments the two strains were able to coexist in the same mice, which allowed the mutant to exploit the extracellular factors produced by the wild, and so the mutant increased in frequency.

This is odd as classical virulence theory would predict that higher strain diversity will lead to greater competition for host resources and this select for a higher growth rate, which leads to greater virulence. From this experiment the opposite was found, higher relatedness favoured greater co-operation between bacterial cells, which facilitated bacterial growth and hence leads to host mortality.

I enjoyed this study as it looked at variation in relatedness focusing on quorum sensing in a population, through the use of genetic manipulation. From the results of this study it would seem that quorum sensing is much more beneficial in populations of high relatedness and so would likely have evolved in such populations. However in natural populations the relatedness will be much more similar across the entire genome, instead of a single trait being different – as Spicer says, “it’s the phenotype which is selected for through evolution, not just one gene”. Another problem which is unique to studying evolution in microbes is the potential for horizontal gene transfer, which can cause relatedness to vary across the genome.

If anyone's interested here's a link to the study: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3396913/

Rumbaugh, K. P., Trivedi, U., Watters, C., Burton-Chellew, M. N., Diggle, S. P., & West, S. a. (2012). Kin selection, quorum sensing and virulence in pathogenic bacteria. Proceedings. Biological sciences / The Royal Society, 279(1742), 3584–8. doi:10.1098/rspb.2012.0843