Arms race in a drop of sea water

Marine cyanobacteria can shrug off viral assault by inactivating the genes involved in virus attachment. But this strategy has a cost: it may affect cell fitness or even favour infection by other viruses. See Article p.604

The oceans are teeming with viruses — typically, there are 100 billion viral particles per litre of water in the top 50 metres of most marine ecosystems1,2. With an average of ten viruses for each bacterial cell, these parasites impose a tight control over the composition of marine microbial communities. The 'arms race' hypothesis holds that the selective pressure exerted by viruses continuously triggers adaptive mutations in the bacterial genomes, with counteracting genetic adaptations occurring at a similar pace in the parasites3. On page 604 of this issue, Avrani and colleagues4 contribute to a better understanding of the molecular mechanisms underlying this process of co-evolution.

Marine viruses do not affect all bacterial groups to the same extent. They exert the strongest predation pressure on the most abundant species, with scarcer ones being much less affected2,5. Thus, paradoxically, viruses help to maintain bacterial diversity over time. As the dominant phytoplanktonic group in offshore tropical and subtropical oceanic waters, the cyanobacterium Prochlorococcus constitutes a particular target for viruses called cyanophages2,6. From a human perspective, viruses are invisible threats. But for a Prochlorococcus cell, which is only about 0.6–1 micrometres across, cyanophages are deadly predators that can measure up to one-tenth of their own size.

Some cyanophages, such as the short-tailed podoviruses (Fig. 1), have a narrow host range — they specifically infect one Prochlorococcus 'ecotype' (a genetically homogeneous population occupying a particular ocean habitat7), or even a single strain6. Other cyanophages, such as the contractile-tailed myoviruses, are less specialized and can infect not only several Prochlorococcus ecotypes but also members of the closely related genus Synechococcus.

Figure 1: Transmission electron micrograph of a marine podovirus.

The viral particle consists of a capsid, and a basal nozzle that is surrounded by six tail fibres enabling the virus to bind to the host cell. Scale bar, 50 nanometres. The virus comes from a sample of water from the Mediterranean Sea, and is reproduced courtesy of Jennifer R. Brum (Tucson Marine Phage Lab, Univ. Arizona).

In their paper, Avrani and colleagues4 unveil some of the ways in which Prochlorococcus strains resist their viral parasites. The authors show that when a Prochlorococcus culture is exposed to a podovirus, a fraction of the population can escape infection thanks to gene mutations that affect the attachment of this cyanophage to the cyanobacterial cell surface. Modifications of the proteins encoded by these genes — which are either components of, or involved in the biosynthesis of, these highly specific viral 'receptors' — can prevent viruses from recognizing the cell and thus infecting it. Although it is not the only antiviral mechanism3, this strategy constitutes a particularly efficient defence against these parasites. But it is not risk-free: Avrani et al. observe that these mutations can decrease population growth rates, making the mutant cells less competitive than their non-mutant counterparts. Mutations may also increase the susceptibility of Prochlorococcus cells to other viruses. So, to survive, Prochlorococcus cells have to engage in a constant game of hide-and-seek with their viral predators.

Avrani et al.4 also provide insight into the molecular mechanisms eliciting the narrow host specificity of podoviruses. The authors note that most of the genes involved in the biosynthesis of viral attachment proteins are located in a specific part of the genome (the 'virus susceptibility region') that has a particularly variable gene content, even between closely related Prochlorococcus strains. Thus, the highly dynamic nature of this genomic area is probably responsible for the considerable diversity of host-specific podoviruses found in the marine plankton. The picture emerging from Avrani and colleagues' study is that an apparently homogeneous Prochlorococcus population in fact consists of a complex assemblage of genotypes that have various ranges of susceptibility or resistance to viruses, so that only a fraction of that population is killed when attacked by a particular cyanophage.

This work illustrates how subtle and complex the relationships between viruses and their hosts can be. It nicely complements studies that have shown how viruses can manipulate their hosts in various ways8. For example, several cyanobacterial genes — most often those involved in the processes of energy metabolism, such as photosynthesis — are found in viral genomes9,10. Surprisingly, these genes can sometimes be modified by the viruses themselves to change and/or optimize gene function11,12. During the infection, viruses hijack the translational machinery of their hosts, making the bacteria preferentially express the viruses' version of the genes9. This strategy probably helps to maintain host metabolism long enough for the viral replicative cycle to be completed.

Similarly, Avrani and co-workers4 observe that homologues of genes encoding viral attachment proteins are present in the genomes of myoviruses, as well as of a variety of bacteria only distantly related to Prochlorococcus. This suggests that myoviruses, which have a broader host range than podoviruses, could transfer viral receptor genes between phylogenetically remote microorganisms, possibly after modifying their sequences while in between the two hosts, to increase their own attachment efficiency. If confirmed, this strategy might constitute a key asset for viruses in their unending evolutionary arms race with bacteria. Clearly, further deciphering the evolutionary mechanisms at play in this process is an essential task for virologists — not least, such knowledge could spur advances in biotechnology and synthetic biology.


  1. 1

    Suttle, C. A. Nature 437, 356–361 (2005).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Wommack, K. E. & Colwell, R. R. Microbiol. Mol. Biol. Rev. 64, 69–114 (2000).

    CAS  Article  Google Scholar 

  3. 3

    Stern, A. & Sorek, R. BioEssays 33, 43–51 (2011).

    CAS  Article  Google Scholar 

  4. 4

    Avrani, S., Wurtzel, O., Sharon, I., Sorek, R. & Lindell, D. Nature 474, 604–608 (2011).

    CAS  Article  Google Scholar 

  5. 5

    Pedrós-Alió, C. Trends Microbiol. 14, 257–263 (2006).

    Article  Google Scholar 

  6. 6

    Sullivan, M. B., Waterbury, J. B. & Chisholm, S. W. Nature 424, 1047–1051 (2003).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Coleman, M. L. & Chisholm, S. W. Trends Microbiol. 15, 398–407 (2007).

    CAS  Article  Google Scholar 

  8. 8

    Rohwer, F. & Thurber, R. V. Nature 459, 207–212 (2009).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Lindell, D., Jaffe, J. D., Johnson, Z. I., Church, G. M. & Chisholm, S. W. Nature 438, 86–89 (2005).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Sharon, I. et al. ISME J. doi:10.1038/ismej.2011.2 (2011).

  11. 11

    Sharon, I. et al. Nature 461, 258–262 (2009).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Dammeyer, T. et al. Curr. Biol. 18, 442–448 (2008).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Frédéric Partensky.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Partensky, F., Garczarek, L. Arms race in a drop of sea water. Nature 474, 582–583 (2011). https://doi.org/10.1038/474582a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.