Viruses that attack bacteria have evolved to collaborate


THAT predators often hunt in packs is a commonplace. Wolves do it. Killer whales do it. Even Velociraptor, a species of dinosaur made famous by “Jurassic Park”, is believed to have done it. These are, or were, all intelligent species, capable of exchanging and interpreting information. But the logic of pack hunting, that many may achieve what one alone cannot, and that individual pack members may perform different roles, does not depend on intelligence. Indeed, evidence has now emerged that this logic applies to viruses, the simplest biological entities of all. It was published this week in Cell, by Edze Westra and Stineke van Houte at the University of Exeter, in England.

The viruses in question are bacteriophages, which “hunt” bacteria. They do not eat their prey. Rather, they take over its genetic apparatus to create replicas of themselves, killing the host as a consequence. To do so they have to penetrate a bacterium’s cell wall and then subvert its internal defences, of which there are several. One of the best known, because it is the basis of an emerging gene-editing technology (see article), is called CRISPR. The CRISPR system detects and cuts up alien DNA. In the wild, such DNA will almost always have come from a virus. To counter this, some bacteriophages have evolved ways of gumming up CRISPR’s cellular machinery. Dr Westra and Dr van Houte have shown that, in essence, such phages collaborate. Some do the gumming. Others hijack the genetic apparatus.

Dr Westra and Dr van Houte were able to deduce what was going on by watching oddities in the rise and fall of bacterial and phage numbers in cultures. On the face of things, a population of CRISPR-armed bacteria would be expected to plummet in the presence of phages counter-armed with anti-CRISPR mechanisms. But this does not always happen. Instead, bacteriologists studying the matter have noticed that phages with anti-CRISPR traits are sometimes unsuccessful in attacking bacteria with CRISPR defences, and die out. Perplexed by this, the two researchers decided to take a closer look.

To do so, they and their colleagues generated a population of CRISPR-armed bacteria and another of phages with anti-CRISPR traits, and monitored exactly what happened when they introduced the one to the other. To start with, the density of viruses always declined. In other words, most of the early anti-CRISPR attacks were unsuccessful. These failed attacks did not leave the bacteria unscathed though. They resulted in the CRISPR defensive systems being weakened, a process the researchers were able to track by stopping an attack in midstream, washing away the phages, and testing the ability of the remaining bacteria to chop up alien DNA.

After this initial fall in viral numbers, if the culture was left long enough—and if there were enough phages in the first place—things eventually turned round. As the number of bacteria with weakened defences increased, more and more of them were subject to subsequent, lethal attacks, resulting in the creation of more phages. Ultimately, as the phages multiplied, the bacteria were overwhelmed and wiped out. Whether bacteria or viruses prevailed thus depended on the initial ratios of the two. Below a certain threshold of phage abundance at the beginning, the bacteria prevailed; above it, the viruses did.

Intriguingly, the evolutionary success of the phages’ approach depends on a second phenomenon—also first studied in social animals—as well. This is kin selection. It relies on the fact that genetically determined behaviour that harms an individual can nevertheless spread if it disproportionately helps kin that carry the same genetic trait. In the case of the phages, the anti-CRISPR mechanism is exactly such a trait. Some viruses carrying it sacrifice themselves so that others may multiply.

Understanding this interaction between phages and bacteria is important, though, for reasons beyond its evolutionary elegance. One such is that phages are under consideration as alternatives to chemical antibiotics, particularly in situations where bugs are immune to those antibiotics. A second is that phages are a crucial, though ill-understood, part of the gut microbiome, the importance of which to human life is becoming clearer by the day. A third is that the interactions of phages and their hosts may be analogous to those of other viruses and other hosts, including human beings. Though animals do not employ CRISPR as part of their defence against viruses, they have a host of other antiviral mechanisms.

Dr Westra and Dr van Houte argue that theories about the spread of disease do not sufficiently take into account the possibility of these defences being damaged and weakened by failed attacks when determining the threats posed by specific sorts of viral pathogen. Monitoring such damage, and the degree to which it makes organisms vulnerable to later attacks, might improve control of the transmission of such diseases, and also the treatment of those who catch them.