I was recently called by an editor at NewScientist asking for some background on the field of fitness in bacteria, and particularly the issue of multi-resistant bacteria persisting in the environment (or clinic) in the absence of antibiotic selection. The reason for the question arose due to the upcoming publishing of an interesting paper in PLoS Genetics:
Silva RF, Mendonça SCM, Carvalho LM, Reis AM, Gordo I, et al. 2011 Pervasive Sign Epistasis between Conjugative Plasmids and Drug-Resistance Chromosomal Mutations. PLoS Genet 7(7): e1002181. doi: “10.1371/journal.pgen.1002181 “:http://dx.doi.org/10.1371/journal.pgen.1002181
Following this, I have been quoted in a NewScientist Health News item, and as I don’t feel my response is quite in the context I gave it, I thought I would give a more detailed account. I spoke to NewScientist last Wednesday (the paper in question wasn’t due to be published until the following day), but I was told that in the study the authors had observed that antibiotic resistance can have a positive effect on bacterial fitness, even in the absence of selection. I was asked whether this was a surprise to me, and more generally about research in bacterial fitness. What I perhaps should have done was ask specifically whether the paper was still embargoed and whether I could have more particulars of the study, because I could not have anticipated the particular nature of the study in question.
The paper describes an observation that the fitness costs of chromosomal mutations conferring resistance to antibiotics, whilst initially deleterious, can become beneficial through the acquisition of a transferable antibiotic resistant plasmid. The outcome of this might be cause for concern given the widespread occurrence of conjugative resistance plasmids in clinical strains; the idea that these may be stabilising otherwise costly resistance mutations is worrying.
We know from many studies that some mutations that give rise to antibiotic resistance come at a cost to bacteria. A bacterium may for example overcome an antibiotic by having a mutation in the key protein that the antibiotic targets, but the mutation may also make that protein less capable at its day-to-day function, and this presents a fitness cost to the cell. Obviously while there is antibiotic around, such mutations bring a strong fitness advantage, but when the antibiotic is no longer around, and these cells are forced to compete with their antibiotic-sensitive brethren, they can become a fitness cost.
The phenomenon in which the effect of one set of genes (say those encoded on a conjugative plasmid) may have a synergistic effect on another set of genes (i.e. those on the chromosome) that results in a more beneficial outcome is called ‘positive epistasis’. Whilst I’m familiar with this phenomenon more generally in bacterial metabolism, as it has been documented by the inimitable Richard Lenski over the years, I would be hard pressed to cite examples where two negative fitness costs (one chromosomal, and one plasmid-borne) might cancel each other out, or indeed even result in a beneficial fitness advantage that is more than the sum of its constituent costs.
However, as I often quote, Crichton’s Jurassic Park character Dr Ian Malcolm would say, ‘Life finds a way…” Indeed, bacteria that encounter a fitness cost may under go ’mitigating’ mutations that alleviate the fitness costs of the original mutation, restoring their fitness to that of their non-resistant brethren. This way the bacteria get to remain resistant at no extra cost, kind of like having your cake, and eating it too. It was this much that I described in my conversation with NewScientist. I have previously described the means by which bacteria may mitigate fitness costs (second half of post).
The study in question is more observational than mechanistic, so we are not clear why it is that acquisition of a conjugative resistance plasmid can be beneficial to cells already encoding chromosomal resistance to an antibiotic. This will likely have to await further investigation, however, the question inevitably turns to how to address this issue. As I mentioned to NewScientist, one aspect of my research is to identify those antibiotics against which resistance imposes the greatest fitness costs, the idea being that when resistance occurs, it may persist for less time once the use of that antibiotic is curtailed. However, this is not a one-stop solution, it would form one of many small changes in practise to prevent antibiotic persistence.
I don’t disagree with Dionisio’s idea of targeting the plasmids, and looking for ways to inhibit plasmid conjugation and replication, thus addressing one of the major causes of antibiotic resistance spread. Indeed, such approaches have been reviewed previously (Williams & Hergenrother, 2008, Exposing plasmids as the Achilles’ heel of drug-resistant bacteria. Current Opinion in Chemical Biology 12: 389-399 doi: 10.1016/j.cbpa.2008.06.015), and some interesting chemical agents found.
In fact, I wonder if the presence of conjugative plasmids reduce the fitness costs of chromosomal mutations to the extent that the typical mitigating mutations seen to stabilise other chromsomal mutations don’t occur. This at least presents the opportunity that a combined attack on the plasmid and the cell may be successful in eradicating particular resistance determinants before they become independently stable.
I’m glad that NewScientist took an interest in this paper as I think we need to keep the issue of antibiotic resistance, and what we learn about the associated problems and the solutions, in the public eye. I’m also open for comments on this area, but perhaps next time I will clarify what the interviewer has understood from my responses.