Introducing Yesterday's Phage Therapy in Today's Medicine

Jean-Paul Pirnay; Gilbert Verbeken; Thomas Rose; Serge Jennes; Martin Zizi; Isabelle Huys; Rob Lavigne; Maia Merabishvili; Mario Vaneechoutte; Angus Buckling; Daniel De Vos


Future Virology. 2012;7(4):379-390. 

In This Article

Phages: Not Your Regular Medicinal Products

Phages can be seen as bacteria's natural infectious agents. Up to 50% of bacterial mortality is thought to be due to phage-induced lysis;[45] hence, phages impose strong selection for bacteria resistance. However, lytic phages can only propagate by infecting and lysing bacteria, hence there is strong selection to overcome this resistance. This interaction leads to antagonistic coevolution, consisting of the repeated emergence of new phage infectivity and bacterial defense mutations.[46–50] Typically, coevolution results in continual increases in bacteria resistance and phage infectivity ranges, although recent work, including a study following real-time coevolution in soil,[51] suggests that high costs associated with resistance may instead result in different, rather than greater, resistance mechanisms being selected through time.[48,52] In principle, coevolution between bacteria and phages could therefore allow the continual production of highly infectious phages that can overcome common bacterial defense mechanisms. However, it is important to emphasize that not all phages are lytic. Many integrate into bacterial genomes, and are propagated via bacterial reproduction.[53] Such lysogenic phages will themselves coevolve with each other,[54] with bacteria and with other lytic phages, and the consequences of this for phage therapy are currently unclear. A recent study showed that in vitro coinfection of Pseudomonas fluorescens with multiple phages had no net effect of accelerating or slowing down adaptation to the host through between-parasite conflict in the system.[55] It is thus tempting to speculate that phages act as 'evolving antibiotics' during real-time coevolution between therapeutic phages and infecting bacteria within patients. However, while real-time coevolution between bacteria and phages results in continual suppression of bacterial densities to some extent,[51,56] the clinical significance of these relatively modest density directions is still unclear.[57] Phages do, however, play a major role in controlling bacterial densities in natural populations, and it is reasonable to assume that coevolution plays a role in this. For example, phages appear to be key players in ending cholera epidemics. Faruque et al. observed that seasonal epidemics of cholera inversely correlated with the prevalence of environmental cholera phages.[58] The removal of phages by conditions such as severe flooding might contribute to rendering water more conducive to human-to-human transfer of Vibrio cholerae. Phage amplification in cholera patients during a cholera epidemic likely contributed to increased environmental phage abundance, decreased load of environmental V. cholerae and, hence, the collapse of the epidemic. In vivo phage amplification in patients and subsequent phage infection in the environment could thus explain the self-limiting nature of seasonal cholera epidemics in Bangladesh.[59]

It is clear that therapeutic phages are very different from classical (chemical, molecular) medicinal products such as antibiotics. Instead, they are natural biological entities that play an important role in maintaining equilibrium in bacterial populations of ecological environments, including humans. Hence, we should not see them as conventional stable medicinal products, but more as interactive and evolving antibacterial products, which could also be used in combination (synergy) with antibiotics.[60] The coevolutive aspect of the phage–bacterium couplet, which is essential for sustainable phage therapy, is often neglected.

However, there are potential negative consequences of this coevolutionary potential. For example, coevolution has been shown to drive the evolution of bacterial mutation rates in laboratory populations of the bacterium P. fluorescens. A quarter of the bacterial populations coevolving with phages had rapidly (i.e., in less than 200 generations) acquired mutations that resulted in ten- to 100-fold increases in mutation rates, whereas no significant change in mutation rates was observed in the absence of phages.[61] Given the increase in evolvability of mutator bacteria (e.g., elevated rates of resistance evolution to antibiotics), evolvable phages may have unknown net consequences on disease severity. Phage therapy should not be implemented widely and without limitation, without first determining these consequences through real-time experimental evolution studies. In the end, natural phages could prove useful, but maybe only in specific (niche) clinical contexts and under certain conditions (e.g., dosage).

Phage therapy fits well in the emerging field of Darwinian medicine (in contrast to a classical mechanistic – man as a machine – view),[62,63] whereby the insights into evolution are fully taken into account, but it is less compatible with our actual western drug development and marketing model.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.