Revived interest in bacteriophages

Abstract

Phages have a track record of being left behind whenever science moves on to other things. But will the inexorable spread of antibiotic resistance lead to a revival of their role as antibacterial therapeutics? Michael Gross investigates. Phages have a track record of being left behind whenever science moves on to other things. But will the inexorable spread of antibiotic resistance lead to a revival of their role as antibacterial therapeutics? Michael Gross investigates. In the unlikely setting of Oxford University's Zoology Department, NERC research fellow Britt Koskella studies horse chestnut trees, their bacterial pathogens and, most importantly, the bacteriophages that infect those bacteria. It's all ecology, she says, which fits in with the interests of the department. But why horse chestnuts? “A lot of work on phages has been done in water,” Koskella explains, “but in the leaves of the chestnut tree we have a well-defined environment where we can pin down the interactions of phages and their hosts.” The bacterial hosts are Pseudomonas syringae, which cause bleeding canker disease in the trees. Koskella and her colleagues wanted to find out about the length scale on which bacteria and phages adapt to each other. In other words, is their frame of reference the border of the leaf, the entire leaf, or perhaps the entire tree? They found out that the tree is an important unit defining the relationship between phage and host. Phages were highly efficient at infecting bacteria sampled from elsewhere in the same tree, but much less so for bacteria from other trees (Am. Nat. (2011), 177, 440-451). Building on these findings, Koskella is now further exploring the ecology of phages in the chestnut tree to see whether their presence actually helps the tree fight off the disease. She is also interested in agricultural applications of phages, for instance in cultivation of tomatoes. “One of our motivating factors for this work is determining whether phages would be good candidates for control of plant pathogens in naturally variable populations,” Koskella explains. There is, she says, a revived interest in bacteriophages which can build on a long but frequently interrupted history of achievements. Phages as natural antibacterial agents were discovered several times, but among the discoverers, French-Canadian microbiologist Felix d'Herelle was the only one who ardently believed that these were viruses. He coined the word ‘bacteriophage’ and propagated this view since 1917, so he can take credit for the concept of viruses infecting bacteria. Losing no time in proceeding to the clinic, d'Herelle tested phage cocktails against dysentery in 1919, and set up commercial production of phage products marketed by the company that later became L'Oréal. Following the introduction of penicillin (in the 1940s) and other antibiotics, phages fell into oblivion in Western Europe and in North America, but their use continued in the Soviet Union. Specifically, the capital of Georgia, Tbilisi, hosted a dedicated phage institute founded in 1923 by Georgi Eliava, with support from d'Herelle. Even though Eliava was executed in 1937, the institute survived and is today named after him. Research into phage therapy and indeed treatment using phage cocktails continued to thrive there and in Poland. However, much of the work was published in Russian journals only and thus never came to the attention of western scientists. Phages had another spell in the spotlight beginning in the 1940s when physicist turned geneticist Max Delbrück started studying their genetics, and the phages lambda and T4 became widely known and well-studied models of molecular biology. In 1967, for instance, research with lambda phages led to the first isolation of a DNA-binding protein (lambda repressor) — independently achieved by Mark Ptashne and Walter Gilbert at Harvard. Phage research explicitly focused on a small number of model systems and thus didn't expand into a broader interest in the evolution, diversity, or usefulness of phages. Molecular biology moved on to more complex systems as soon as it had the tools to do so. Similarly, genome sequencing started out with phages MS2 (RNA) and ϕ-X174 (the first DNA genome), but then swiftly moved on to cellular organisms, leaving the world of phages mostly unchartered territory. Only in recent years, following the application of environmental genomics to various biotopes, have researchers started to realise that phages are by far the most numerous biological entity on our planet and play important roles in ecological networks. For instance, research into deep sea samples published in 2008 showed that in nutrient-poor biotopes phages account for over 80% of bacterial mortality, and that the release of nutrients from bacterial cells after lysis by phages creates a microbes-only food web that excludes higher organisms and enables the bacterial community to survive in spite of the poor food supply. The picture emerging from genomic studies is that there may be ten times more phages than cellular organisms on our planet. Even in our own bodies, colonised by many more bacteria than we have cells, there are 100 phages for every bacterium. Horizontal gene transfer operated by phages has probably shaped evolution in many ways, which we are only beginning to understand, adding further complexity to the ‘tree of life’ image based on Darwinian evolution, as phage researcher Harald Brüssow from the Nestlé Research Centre in Lausanne, Switzerland, has recently outlined in a review (Phil. Trans. R. Soc. B (2009), 364, 2263-2274). If we are already infested by billions of phages without ever noticing, adding a few more to fight disease should be quite safe, the proponents of phage therapy argue. Indeed, the long history of phage therapy, while a bit short on the double-blind, placebo-controlled studies that would live up to modern standards, doesn't contain any records of adverse effects. Biologically, that does make sense, as phages are so highly specialised for a specific species or even strain of bacteria that they have no way of interacting with eukaryotic cells. However, one would be well advised to use lytic phages — the kind that instantly lyses and thus kills the bacterial host — as opposed to lysogenic phages, which have the habit of inserting their DNA (known as a prophage) into the host genome, which can lead to the transfer of virulence traits and all kinds of mischief. The epidemiology of cholera illustrates this crucial difference. The gene for cholera toxin, one of several virulence factors required for Vibrio cholerae to become pathogenic, is transmitted by a lysogenic phage. However, among the ∼200 bacteriophages that infect Vibrio (collectively known as vibriophages), there are also lytic phages that can control the spread of the bacterium and don't alter its genome. Under the auspices of the so-called Bacteriophage Inquiry held in India from 1928 to 1934, researchers studied possible treatment and prevention of cholera using vibriophages. The study reported a 15-fold reduction in mortality, but the approach was abandoned after the advent of antibiotics and rehydration therapy. The recent cholera outbreak in Zimbabwe has revived interest in research into the dynamics of the relationship between phages, Vibrio and human hosts, based on the hypothesis that vibriophages, while unable to clear Vibrio infections from human intestines, may help to block transmission (Nat. Rev. Microbiol. (2009), 7, 693-702). A complicating factor is that natural phage populations seem to be playing a large role in shaping the evolution of Vibrio cholerae by driving major changes in serotype prevalences. This implies that understanding the natural dynamics will be critical to using phage therapy at the population scale. Additional interest in phage therapy is based on the current problems surrounding the spread of antibiotic resistance genes. In a development that experts have been predicting since the mid-1990s, inconsiderate use of available antibiotics has led to widespread occurrence of resistant strains such as the multidrug-resistant Staphylococcus aureus, also known as MRSA or, in the popular press, as “hospital superbug”. So far, the standard response has been to introduce new antibiotics, but at the moment the pipelines of the pharma industry aren't exactly overflowing with new antibiotics, and there are fears that the infectious diseases may make a comeback. This developing crisis has generated fresh interest in everything that fights bacteria. One of the alternative approaches, so called pilicides, consists of stopping pathogenic bacteria from growing the ‘hairs’ that they need to attach themselves to their hosts. The unique selling point is that this treatment doesn't kill bacteria in the environment (like any leaked antibiotic would), so there is less selection pressure in favour of resistance. However, this approach is still very much in the domain of fundamental research. So why not use phages, which have been known to mankind for almost a century now, as an alternative to antibiotics? Sceptics point to the unwieldy size of a phage (compared with small-molecule drugs), which bars them from entering human organs and cells. “The phage therapy field has not been able to move to phase I clinical trials for any systemic application, and I do not see this happening any time soon,” says Kim Lewis, director of the Antimicrobial Discovery Center at Northeastern University in Boston, Massachusetts. “There is little evidence that phage therapy will help us solve the antibiotic resistance crisis,” he concludes. The size issue may indeed limit phage therapy to the more accessible parts of a patient (but these are also more accessible to bacteria, so more likely to need the phage treatment). Critics also like to emphasize the fact that bacteria will develop resistance against phages. This is where ecological research like Koskella's work on chestnut trees comes in handy. While it is clear that phages are intrinsically unable to wipe out their hosts, a better understanding of their ecology may enable researchers to find out which additional intervention might do that once the phage has reached an equilibrium with the bacterium. “I think that combining treatments is the way forward,” says Koskella. In the best case, the immune system might get the upper hand once the bacterial infection is weakened by fighting its own disease. There is a simple economic reason why pharma and biotech companies in the western world haven't been keen on phages. Something that is (to a certain extent) alive and co-evolves with the pathogen is much harder to patent and exploit economically than a well-defined small molecule made industrially — a problem very familiar to all who work in medical applications of stem cell research. Oxford phage researcher Angus Buckling, who published a recent report describing phage–host coevolution in soil (Science (2011), 332, 106-109) and was also involved with the chestnut paper, says he'd rather not patent phage cocktails, because “this could then stop the very successful cottage industry of phage therapy in the Eastern Europe and other places around the world”. Nevertheless, there are now a few western companies testing the waters. Omnilytics, based in Salt Lake City, Utah, USA, is developing phage cocktails to treat bacterial diseases of agricultural crops that have become resistant to traditional treatments. In 2006, US company Intralytix, based in Baltimore, Maryland, gained FDA (Food and Drug Administration) approval for a cocktail of six phages that can be sprayed onto ready-to-eat food products, in order to control the food-borne bacterium Listeria monocytogenes. This February, Intralytix received regulatory clearance from the FDA for a phage-based food safety product called EcoShield, which is effective against Escherichia coli O157:H7. Alexander Sulakvelidze, Chief Scientist at Intralytix and one of its co-founders, says phages are a very safe and natural way of protecting food against bacteria. “The concept of using phages for improving food safety is based on simply applying phages back onto the foods in the right concentration, at the right place, and at the right time,” Sulakvelidze explains. “Thus, applying lytic phages onto foods to eliminate or significantly reduce the levels of pathogenic bacteria in those foods is one of the most, if not the most, environmentally-friendly, safe, and effective approaches available today,” he concludes. In the UK, the company Targeted Genetics has just changed its name to AmpliPhi Biosciences to emphasize its bacteriophage technology. Earlier this year, the company had acquired Biocontrol, whose lead product, a phage treatment for chronic ear infections, has just completed a clinical phase I and II trial. The company is also developing a phage treatment to combat Pseudomonas lung infections in patients with cystic fibrosis. Meanwhile, international corporation Nestlé is conducting a clinical trial using a cocktail of T4 phages against childhood diarrhoea in Bangladesh. The randomized, double-blind, placebo-controlled trial overseen by Shafiqul Sarker in Dhaka, Bangladesh, enrols around 450 boys aged six months to two years with diarrhoea, provided they test negative for cholera and rotavirus. Equal numbers of the children will randomly receive either a new T4 phage cocktail or an established phage product from the Russian company Microgen, or a rehydration solution as placebo. The trial is due to be completed in June this year. Even if phage cocktails turn out to be no satisfactory solution, there is still the opportunity to learn from phages and to borrow some of their molecular arsenal, such as the lysins (FEMS Microbiol. Rev. (2009), 33, 801-819). In any case, it is clear, concludes Koskella, that “in our fight against pathogenic bacteria we have viruses on our side”.