In the early 1900s, most researchers believed that DNA was a “stupid molecule,” too simple to be of any value in the transmission of life. Instead, scientists championed proteins, with their variability and complexity, as the key component of heredity. Then in the early 1950s, geneticists Alfred Hershey and Martha Chase, working with bacteria-infecting viruses called bacteriophages, confirmed DNA as the informational unit of the cell. The now-famous Hershey-Chase experiments tracked the movement of viral DNA into bacterial cells, as well as the fate of the proteins on the viral capsid, and demonstrated that viral DNA is necessary and sufficient for phage replication, implicating DNA as the molecule needed for every organism’s reproduction.

In 1953, James Watson, Francis Crick, and Rosalind Franklin solidified Hershey and Chase’s conclusions by elucidating the double-helical structure of DNA. And later, additional work with bacteriophages definitively demonstrated that DNA...

Viruses may be little more than a ball of genetic material with a single functional goal of transmitting pure information in the form of DNA or RNA, but researchers are finding more and more evidence that viruses freely share their genes among diverse ecosystems, making viruses a powerful and large genetic reservoir that challenges how we think about all biological life.1 As pure biological language, however, viruses are inherently difficult to understand.

For more than 130 years, since the first inkling that viruses existed, these submicroscopic entities have continued to occupy a sort of netherworld in both the scientific community and the public consciousness. According to Forest Rohwer of San Diego State University and science writer Merry Youle, bacteriophages are the dark matter of the biological universe—vital elements that don’t seem to behave the way other things do. Every time science poses a biological rule, phages seem to break it.

What is clear, however, is phages’ vital role in human and environmental health, and many laboratories across the globe are looking to uncover how bacteriophages interact with their bacterial hosts to affect the Earth’s ecosystems. Found in virtually every biome on the planet, from coral reefs to the mucus layers of many animals, bacteriophages are the most ubiquitous “organisms” on Earth. We encounter billions of phages daily in what we breathe, eat, drink, and bathe in—shedding an equally large number as we live our lives. Phages powerfully affect genetic change in soils, vegetation, and oceans, regulating nutrient cycling, evolution, and even climate change on a global scale. (See “An Ocean of  Viruses,” The Scientist, July 2013.)

Phages are so integrated into healthy ecosystems that we can often overlook their essential role. Even seemingly explainable interspecies relationships, like that between aphids and their symbiotic bacteria, are proving to be a tripartite commingling, with phages providing vital genes for their bacterial hosts that subsequently benefit the macro-host as well. In the case of pea aphids, a viral gene encodes a toxin, expressed by  a commensal bacteria, that imparts protection to the aphid host against parasitic wasps.2

One can start to envision phages less as discrete entities and more as fluid conveyors of genes in an ecosystem.

Though many tripartite relationships among organisms have been studied, we still don’t know exactly how many exist, and with more than 1031 viruses worldwide, it is likely that phages are a crucial part of all ecological relationships. As Harald Brüssow of the Switzerland-based Nestlé Research Center aptly notes in a recent Current Opinion in Microbiology paper, “[the] molecular biology of higher organisms does not stand on the shoulder of giants, but on the shoulder of dwarfs like phage T4 and lambda.” 3

Dana Willner, a researcher at the Australian Centre for Ecogenomics at the University of Queensland in Brisbane, whimsically compares viruses to mythical spirits that must move from host to host in order to survive, transforming the host in the process. This analogy is not far from the truth. Not dead, yet decidedly not alive by strict biological standards, viruses move between worlds in ways that we are only just beginning to understand.

Viral defenders

THREE-WAY MUTUALISM: Pea aphids (right) can protect themselves against parasitism by female wasps (left)—which lay their eggs in the insect’s body—by harboring a bacterium that carries a toxin-encoding phage.© NIGEL CATTLIN/SCIENCE SOURCEFor many years, the study of viruses was limited to their functional role in disease—from which they derived names like the hepatitis C virus and the tobacco mosaic virus. Indeed, viruses often make their hosts sick, and researchers have generally aimed their studies at how to prevent viral disease. It wasn’t until about 100 years ago, when French-Canadian microbiologist Félix d’Herelle isolated phages and used them to cure bacterial dysentery, that we learned that viruses can also make us well.

They do so by infecting pathogenic bacteria, hijacking the cellular machinery to make more phages, and causing bacterial cells to explode in a cloud of new viral particles. These viral explosions spread exponentially throughout a bacterial colony, eventually destroying all susceptible pathogenic cells. Because phages typically infect specific bacterial strains, a diverse population of phages is needed to effectively destroy all offending bacteria. And the best way to isolate a diverse population of phages is to isolate them directly from the bacteria they infect.

D’Herelle’s rudimentary method of isolation involved culturing bacteria from the feces of spontaneously recovering dysentery patients and removing all of the cells by passage through a porcelain filter. He then exposed new bacteria to this filtrate. As the cells began to die, d’Herelle correctly presumed that virulent bacteriophages had made it through the filter and were killing the new cells as the viral population continued to multiply. After another pass through the filter, he then successfully treated patients with the resulting filtrate. A similar process is used today in parts of Russia and Europe to develop phage therapies against common human pathogens, and in the United States to control food-borne bacterial contaminants.

But as d’Herelle had barely begun to suspect, many phages, probably most, do not actually kill their host. Rather, they integrate into the host genome as prophages, preserving viral DNA for later replication but also giving the bacterial host access to new genes that may confer new properties. For example, the phage that provides the aphid’s symbiotic bacteria with the genes for protection against parasitic wasps isn’t killing its host cell. Rather, it is changing the cell, conferring a superpower of sorts that protects the aphid, keeping the ecosystem in balance between predator and prey.

GOING VIRAL: As little more than carriers of DNA, bacteriophages serve to shuttle genes between diverse ecosystems. The viruses take up genetic material from their bacterial hosts and donate it to future hosts, both near and far. Such genetic movement can spread bacterial traits such as virulence, antibiotic resistance, or adhesion capability—and even introduce novel genes to new environments. While viruses can only move short distances outside of a host, the migration of bacteria, or of the megafauna that bacteria infect, allows phages to traverse the globe.
 View full size JPG | PDF 
However, this virus-induced transformation often results in uncertain outcomes, like a game of genetic Russian roulette, and there is always a possibility that phages could confer enhanced pathogenicity or phage/antibiotic resistance to resident bacteria. Furthermore, research shows that phage-bacteria interactions can be altered in unforeseen ways due to environmental influences, such as the administration of antibiotics4 or even something as seemingly innocuous as the ingestion of soy sauce.5 Couple this possibility with the fact that, even after 100 years of phage research, scientists still cannot definitively describe an overarching mechanism for phage-bacteria interactions nor predict outcomes of those interactions in the wild, and it is clear that the risks of phage therapy can be high.

Indeed, such risks appear to be a primary hurdle in the U.S. for developing phage therapies as human medicines. While nonprescription phage cocktails in the form of pills, liquids, topical applications, and injections are marketed and sold in France, Poland, and Russia for treating illnesses from intestinal troubles to skin infections, the US Food and Drug Administration (FDA) currently does not approve any phage therapies for treatment of human disease, in part because of the uncertainty of phage-host interactions.

As the medical practice of altering our gut microbiota through procedures such as fecal transplants gains popularity, however, this uncertainty may become painfully apparent, as these therapies will no doubt include the phage community associated with the commensal bacteria being transplanted. Another concern about the current phage cocktails on the market is that most of the foreign companies that sell such therapies do not provide any information about the contents of the phage mixtures. A recent analysis from the Brüssow group of one popular Russian phage cocktail found that it contained a very complex phage population with high genetic variation.6 Although the phages in the cocktail did not possess any known pathogenic genes, and in a small human trial the cocktail appeared to be harmless, the researchers cautioned that we still know very little about phage genomes and that there may be hidden virulence-conferring genes yet to be identified.

There are a few medical products in development, however, such as Baltimore, Maryland–based Intralytix’s PhagoBioDerm (still in clinical trials), a biodegradable film that is permeated with phages and can be placed on wounds to clear and prevent bacterial infection. Furthermore, researchers have used phages to treat bacterial infections in the lungs of mice—a result with significant implications for human cystic fibrosis therapy7—and a recent large grant from the European Union funds multicenter research in phage therapy for burn patients. Nevertheless, the interest in this field is growing very slowly. “The main hesitation [is in] invest[ing] 5 to 10 million dollars (which is needed to develop a phage product) in the face of the undocumented clinical benefit of phage therapy and the unclear regulatory and patent situation for a phage product,” Brüssow writes in an e-mail.

In other industries, however, the use of phages is thriving. With the growing failure of antibiotics over the last few decades, researchers are turning to lytic phages for bacterial control in the US food industry, commercial agriculture, and veterinary medicine. The FDA and US Department of Agriculture approved Intralytix’s ListShield in 2007 and EcoShield in 2011 as phage-based sprays for commercially processed foods to prevent the growth of some types of Listeria and E. coli, respectively. And just this year, Intralytix’s SalmoFresh has been approved by the FDA as GRAS (generally recognized as safe) for human use against contamination by highly pathogenic serotypes of Salmonella.

A PHAGE HELPER: Many bacteria harbor viral genes that improve their own survival. In the case of Vibrio cholerae, the causative agent of cholera, bacteriophages deliver genes for the bacterium’s adhesion molecules and tail fibers.© JAMES CAVALLINI/SCIENCE SOURCEIn agriculture, using phages in combination with other crop treatments, such as certain pesticides or nonpathogenic versions of disease-causing bacteria, has the potential to be very effective against numerous bacterial diseases that devastate many plant species, from fruits and vegetables to tobacco and fungi. Utah-based OmniLytics, for example, sells an agricultural treatment called AgriPhage that protects tomatoes and peppers from bacterial spot and bacterial speck, two common infections that can reduce tomato yields by half. The treatment was registered by the Environmental Protection Agency (EPA) in 2006, and the Institute of Food and Agricultural Sciences officially recommends that tomato growers in Florida use this and similar products for phage-based management of bacterial disease.8

Similarly useful phage therapies are employed or are in development in a wide range of industries, from dairy and fish farms to veterinary medicine, where phages that induce an immune response that attacks the animal’s reproductive cells are proposed for the noninvasive sterilization of animals. Also, research into wastewater treatment has suggested that phages could be effectively used to clear away bacterial biofilms.

The chicken and the egg

Researchers agree that safe and controlled use of phages requires detailed information on how the viruses interact with their bacterial hosts, but acquiring such information is difficult in the face of the constant genetic swapping that characterizes phage dynamics. Viruses and bacteria are in a constant evolutionary arms race, modifying their genetics in a madcap attempt to infect or evade infection. Scientists estimate that 2.5 x 1025 viral genomes are replicated every second, and those genomes are packaged and transferred to the next suitable cell. Further, researchers believe that there is at least one “mistake”—a mutation that could lead to a nonfunctioning virion or a virus with different binding abilities, for example, or the inclusion of genetic material that is not essential to the viral life cycle, known as a moron—for every 1,000 genomes replicated. Completing the math, it means that 2.5 x 1022 altered viruses are made every second.9

How can we control something that won’t sit still long enough for characterization? Further, how do we take what we’ve learned in vitro about viral dynamics into the real world, when we are increasingly learning that these dynamics are often abandoned by phages and bacteria alike in vivo?

“Despite decades of research, phage research is very much an emerging field,” says Rohwer. “We still don’t understand how phage-host systems are set up in the environment or the full extent of their interactions between environments.”

Despite their intrinsic lack of motility, phages appear able to move around, with viral lineages moving from soil to the tops of plants or from ocean to freshwater, bringing with them genes that the new environment may have previously lacked. (See illustration above.) Amazingly, despite this travel through various ecosystems, viruses’ own genetic identities stay remarkably stable over time and distance, making them perfect vectors for genetic transmission between organisms and even entire ecosystems.

Viruses shuttle bacterial genes around, but conversely, bacteria can often take up genes from the viruses they harbor. Indeed, one can start to envision phages less as discrete entities and more as fluid conveyors of genes in an ecosystem. In combination with their extreme abundance, phages create countless genetic variations on which natural selection can act; they are the infinite number of monkeys typing randomly for an infinite number of years to produce biological drama that is on par with the best Shakespeare play. In some cases, it may actually be more beneficial to the host to allow phage infection than to develop resistance, and many bacteria freely “borrow” viral genes to better ensure their own survival—such as viral genes encoding photosynthesis proteins in cyanobacteria, which keep the cells alive, allowing them to produce more virions even when their own machinery has begun to fail. Viruses can also carry genes that facilitate cell-to-cell interaction and colonization—such as genes for the adhesion molecules and tail fibers of V. cholerae.10 Furthermore, despite their limited capsid size, phages often carry extra and seemingly irrelevant morons, which can range from various virulence factors to genes that code for defensive proteins that mimic bacterial defense mechanisms.11 Phages can even carry the entire genome of another phage.12

It has become increasingly clear that genes shift back and forth between virus and host in an intricate dance that blurs the line between viral and bacterial genes. After millions of years of freely sharing, stealing, and modifying genes from each other, bacteria and phages often resemble each other in areas of functionality. Many times, the information we gather about phages is a mere snapshot of microbial evolutionary bustle: a frozen moment in the flow of genes.

“We’ve seen how viruses play the game,” Willner says. “They win in numbers and diversity. They carry important genes and rearrange them, and even their genomes as a whole, to become intricate mosaics. Viruses use probability to their advantage.”

Dark matter

Phages should be simple: they are only genetic material and a few proteins that form a hollow ball and a tail. They rely on the metabolic energy of their bacterial hosts to reproduce and spread to new hosts. But they are not simple. Despite their host specificity, phages are not picky. They will squeeze almost any genetic material into their tiny heads whether it is helpful for their life cycle or not, thus changing the nature of their next host, and their interaction with it. Combine this promiscuous movement of genes with their mind-blowing numbers, and bacteriophages are capable of forming and collapsing entire ecosystems.

The study of phages puts us right back into beginning biology, relearning and rethinking tenets that are centuries old. Even in this century, we still argue about viral “species” and genetic propagation and the biological golden rule of a one-way flow of DNA to RNA to protein. As pure movers of genes—coded in DNA or RNA—phages defy these rules.

Mythical language is appropriate for describing phages. They are both Dr. Frankenstein and his chimerical creation. They are powerful spirits looking for a host. To enter the phage world is to enter the theoretical, the unknown, and at times the seemingly nonsensical. Phage research falls into the rabbit hole, and the Wonderland at the end defies our current biological reality. Phages are indeed the hotly contested “dark matter” that somehow makes the biological universe work in ways we have yet to understand. 

Breeann Kirby is a writer who partners with Forest Rohwer’s laboratory at San Diego State University. Jeremy J. Barr is an assistant research professor in the Department of Biology at San Diego State University.


  1. A.D. Hershey, M. Chase, “Independent Functions of Viral Protein and Nucleic Acid in Growth of Bacteriophage,” J Gen Physiol, 36:39-56, 1952.
  2. N.A. Moran et al., “The players in a mutualistic symbiosis: Insects, bacteria, viruses, and virulence genes,” PNAS, 102:16919-26, 2005.
  3. H. Brüssow, “Bacteriophage–host interaction: From splendid isolation into a messy reality,” Curr Opin Microbiol, doi:10.1016/j.mib.2013.04.007, 2013.
  4. S.R. Modi et al., “Antibiotic treatment expands the resistance reservoir and ecological network of the phage metagenome,” Nature, doi:10.1038/nature12212, 2013.
  5. D. Willner et al., “Colloquium paper: Metagenomic detection of phage-encoded platelet-binding factors in the human oral cavity,” PNAS, 108:4547-53, 2010.
  6. S. McCallin et al., “Safety analysis of a Russian phage cocktail: From MetaGenomic analysis to oral application in healthy human subjects.” Virology, 443:187-96, 2013.
  7. L. Debarbieux et al., “Bacteriophages can treat and prevent Pseudomonas aeruginosa lung infections,” J Infect Dis, 201:1096-104, 2010.
  8. J.B. Jones et al., “Bacteriophages for plant disease control,” Annu Rev Phytopathol, 45:245-62, 2007.
  9. M. Youle et al., “Scratching the surface of biology’s dark matter,” Viruses: Essential Agents of Life, ed. Günther Witzany (Springer Netherlands), 61-81, 2012.
  10. A.M. Comeau, H.M Krisch, “War is peace—dispatches from the bacterial and phage killing fields,” Curr Opin Microbiol, 8:488-94, 2005.
  11. K.D. Seed et al., “A bacteriophage encodes its own CRISPR/Cas adaptive response to evade host innate immunity,” Nature, 494:489-91, 2013.
  12. M.M. Swanson et al., “Novel bacteriophages containing a genome of another bacteriophage within their genomes,” PLOS ONE, 7:e40683, 2012.

Correction (September 11): This story has been updated from its original version to correct the names of the first authors on three of the listed references (2, 4, and 10). The Scientist regrets the errors.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?