Phage therapy is an alternative therapeutic strategy consisting on fighting a bacterial infection with bacteriophages. During the cold war, while the western block was developping the use of antibiotics, The phage therapy was developped in the eastern block, where it is still used today.

This specificity can be very advantageous since it protects the patient from a dysbiosis that often happens at least transitorily with the antibiotic therapy. However, bacteriophages cannot always be used. Phages cannot access all infected sites in the human body (i.e. a kidney infection) and can also be recognized be the human immune system. Thus, phage therapy is mostly efficient on topical (i.e. infected burn wounds) and gastrointestinal infections.

Contrary to classic laboratory conditions, bacterial populations are never homogenous in nature. Part of the population will grow will the other will not, some bacteria will be free-swimming in a liquid medium while most of the population will remain in biofilms, structures composed of microorganisms embedded within a slimy extracellular matrix in which cells stick to each other and often also to a surface. To improve our understanding of the phage therapy efficiency, it is important to understand how viral populations cope with bacterial populations under these complex ecological conditions, and how they evolve in response to these conditions.

During this post-doctoral experience at SUNY Albany, I worked on the adsorption rate the bacteriophage lambda. Adsorption describe the ability of the virus to “stick” to the host cell membrane and is therefore the first step during the viral life cycle. It is often believed that bacteriophages (phage) should evolve toward a high adsorption rate to maximize the number of hosts they infect and thus their fitness. During these 2 years of research I showed that this is not always the case.

Viral adsorption rate and the bacterial biofilms - Gallet et al. (2009)

In liquid cultures, a high adsorption rate enables the virus to infect many host cells which of course is beneficial to its fitness. However, in nature, more than 99% of bacteria are in biofilms. If we know that phages can infect bacteria imbedded in biofilms, we have no idea how phages deal such populations structures or if phages infecting biofilms and similar to those infecting free-swimming bacteria. I asked this question in a project in which I performed an evolution experiment in which lambda phages could migrate from biofilm to biofilm by diffusing in a liquid medium. With this simple set up, we could estimate several key parameters like the ability of different lambda strains to settle on a biofilm, the number of phages per plaque, the migration rate (outside of the biofilm), etc. Our results showed that lambda quickly evolve toward a low adsorption rate, by losing its side tail fibers (viral organs stabilizing the virus-bacteria complex). Losing its fibers enabled the phage to enter in the biofilm more easily, make larger and more productive plaques (due to easier migration within the matrix) and migrate faster out of the biofilm. We can conclude from this work that viral traits such as adsorption rate are strongly dependent on the structure of their host population.

Viral adsorption rate and the bacterial stationary phase - Gallet et al. (2012)

For many phages, not all hosts are equally worth infecting. While there are phages, such as T7, whose productive infection is independent of host physiological state, most phages rely on the exponentially growing bacterial cells for productive infections. The bacteriophage lambda belongs to this latter category, and when it infects a cell in stationary phase, it can either integrate its genome into its hosts genome (lysogeny) or hang out in the bacterial cytoplasm until it is recognized by endonuclease which will eventually destroy it. Lambda can induce lysogeny but how do strictly lytic phages deal with the bacterial stationary phase? For these viruses, infecting a non-growing cell is a suicidal act.

I used a strictly lytic lambda mutant to address this question. And what we observed was that while most viruses will infect host cells and died, a small fraction (the residual fraction) exhibiting extremely low adsorption rates, protected phage populations from extinction. We showed that this phenotype was not heritable since the progenies of these low-absorption phage had a “normal” adsorption rate. Whether this cryptic phenotypic variation is an adaptation (diversified bet hedging) or merely reflecting unavoidable defects during protein synthesis remains an open question.