Collins's proof-of-concept virus is tailored to a particular type of E. coli biofilm. "There are many species and strains of bacteria out there," he says, and a single biofilm might support multiple bacterial species and strains. To a lesser degree, there is also some diversity in the components of the biofilm matrix. However, Collins says that because of the increasing speed and falling price of DNA-sequencing and synthesis technologies, it would not be difficult to develop a virus tailored to each kind of biofilm.

Collins's viral technique appears to overcome some of the problems with chemical techniques. Blackwell, who is designing small molecules to disrupt the bacterial signaling pathways that maintain biofilms, says that delivery of biofilm-disrupting chemicals such as enzymes has been a major hurdle. (See Blackwell's TR35 Young Innovator profile.)

Viruses like those developed by Collins have been used for decades to treat infections in Eastern Europe and Russia. But none have been approved for clinical use in the United States yet. However, the FDA has approved one virus cocktail for use as a food additive.

The risks of such viruses are unclear, but there is some concern that they might provoke a dangerous immune response. One reason they might not have been widely studied for their potential to treat infection, says Collins, is that antibiotics have been sufficient so far. But with the emergence of multi-drug-resistant bacterial strains in hospitals, "a number of companies are looking to viruses," he says.

The viruses are likely to be approved for industrial use, for which regulations are not as strict, before they are brought to the clinic. "For industrial applications where you're not putting them in someone's body, these viruses could have a huge impact" on biofilm control in places like water pipes and drains, says Blackwell.