by Michael Laub
The rise of antibiotic resistance is a frightening reality. In response, many have insisted that we must find and develop new antibiotics. But developing even a single new antibiotic – one that is both safe and effective – is a daunting task that could take decades, if successful at all. Indeed, given the risks and the lack of financial incentives, many pharmaceutical companies have completely abandoned their antibiotic development efforts. This could be grim news with dire consequences for human health. But fortunately there is evidence that we can deploy the current arsenal of antibiotics more effectively to circumvent resistance and the need for new antibiotics.
The article that I chose to highlight for the PLOS Biology XV Collection, “When the Most Potent Combination of Antibiotics Selects for the Greatest Bacterial Load: The Smile-Frown Transition”, tackles this issue. In this article, Robert Beardmore and colleagues focus on understanding how combinations of antibiotics impact bacterial growth, with some rather counterintuitive results.
Combination therapy is a popular and powerful means of combatting many bacterial pathogens, viral infections, and even cancer. The simple, and perhaps simplistic, notion is that treating a bacterial infection with two antibiotics must be better than one. In particular, for two antibiotics that are initially synergistic, i.e. the combination suppresses growth more effectively than either does alone, it would seem logical that more is better. Continued treatment with those two antibiotics seems like the natural course of action to eradicate the bacteria. But what this article demonstrates is the exact opposite! What is initially a powerful combination can, in fact, lead to the highest load of bacteria in the long-run.
How do we make sense of such a counterintuitive result? Beardmore’s group showed that the key is to consider the competition that occurs between drug-susceptible and drug-resistant members of a population. Antibiotic-resistant mutants readily arise within almost any population of bacteria. If a population containing both susceptible and resistant mutants is treated with two antibiotics, the susceptible majority is, as expected, rapidly wiped out. But this also removes any competition for the resistant minority, enabling them to quickly grow and proliferate. In contrast, treating the same mixed population with a single antibiotic can be less effective initially as the susceptible bacteria aren’t eliminated as quickly, and their continued (albeit impeded) growth helps keep the resistant bacteria in check.
This article demonstrates how such a competition-based model can, in principle, complicate the long-term dynamics of antibiotic-treated bacterial populations, leading to what they term a ‘smile’ to ‘frown’ transition. To explain: if one plots the density of bacteria on the y-axis and various combinations of two antibiotics on the x-axis, ranging from a 100:0 split of antibiotic 1:antibiotic 2, to an even 50:50 split, to 0:100, the result is a ‘smile’, at least initially (see Figure). But what the modeling suggests is that over time, this curve is inverted to a ‘frown’ as the 50:50 split goes from being the best combination to the worst.
The article demonstrates that this smile-frown transition occurs in real populations of E. coli. And they show, as predicted from their modeling, that the transition depends on the emergence of antibiotic resistance, frequently through duplication of a drug efflux gene. Eliminating this gene largely eliminated the smile-frown transition.
This work by the Beardmore group is a beautiful demonstration of how easily our initial intuition can lead us astray and how complex dynamics can emerge in seemingly simple systems. Whether the exact principles learned from this article will apply in the context of clinically-relevant infections remains to be seen. But at a minimum, this paper highlights the profound importance of careful and quantitative analyses in thinking about how we deploy antibiotic combinations. Such efforts might just be a better and more practical strategy in the coming era of antibiotic resistance than waiting for a new miracle drug.
Pena-Miller R, Laehnemann D, Jansen G, Fuentes-Hernandez A, Rosenstiel P, Schulenburg H, Beardmore R (2013) When the most potent combination of antibiotics selects for the greatest bacterial load: the smile-frown transition. PLOS Biol 11(4):e1001540. doi: 10.1371/journal.pbio.1001540
Michael Laub works at the Massachusetts Institute of Technology, and is a member of the PLOS Biology Editorial Board
Featured image Credit: Pixabay
Michael Laub image credit: Kalman Zabarsky