News & Views | Published:


A parasite's parasite saves host's neighbours

Nature volume 540, pages 204205 (08 December 2016) | Download Citation

Viruses can be attacked by parasitic viruses, which compete with them for cellular resources. It emerges that one such parasitic virus can defend a host-cell population from a viral attack. See Letter p.288

So nat'ralists observe, a flea

Has smaller fleas that on him prey;

And these have smaller fleas to bite 'em.

And so proceeds ad infinitum.

These words, written by the satirist Jonathan Swift in On Poetry: A Rapsody1, describe a real biological phenomenon whereby a parasite is itself subject to parasitic attack2. The interactions that occur when a parasite is preyed on are poorly understood — yet anything that limits the success of the primary parasite might affect the outcome of infection. On page 288, Fischer and Hackl3 provide insight into the relationship between a host cell, a virus and a parasitic virus that preys on the virus.

Single-celled organisms called protists can be infected by giant viruses of the family Mimiviridae4. Giant viruses reproduce in a membrane-bound structure known as the virus factory, which they establish in the cytoplasm of their protist host. Giant viruses are themselves preyed on by parasitic viruses called virophages5, from the family Lavidaviridae6. Virophages also reproduce in the virus factory, and substantially reduce the reproduction rate of the giant viruses5.

Fischer and Hackl investigated the infection of the marine protist Cafeteria roenbergensis with the C. roenbergensis giant virus CroV and its associated virophage, mavirus. Mavirus can enter the host cell independently of CroV uptake7. The authors co-infected a cell population with CroV and mavirus, and found that the mavirus genome was inserted into the protist genome in approximately one-third of the cells, providing the first direct evidence of virophage DNA integration into the genome of a cellular host. The C. roenbergensis strain analysed carried 11 copies of the mavirus genome, integrated at different chromosomal locations. Mavirus and other virophages encode enzymes known as integrases that enable DNA insertion into a host genome7,8.

Does virophage DNA that has been integrated into the protist genome have any function in the host cell? Fischer and Hackl observed that when protist cells containing integrated mavirus genome were infected with CroV, the mavirus genes were expressed and mavirus replicated to form new viral particles, resulting in reactivation of mavirus in the cell. The authors propose that a CroV-encoded transcription factor is responsible for mavirus reactivation, consistent with a previous observation7 that CroV and mavirus have similar promoter sequences that drive gene expression.

Fischer and Hackl found that the reactivation of mavirus that followed CroV infection did not prevent the replication of CroV, and host cells infected with CroV still died, releasing virus particles of both CroV and mavirus into the environment. Presumably, reactivation of the integrated mavirus occurs late in the reproductive cycle of CroV and therefore cannot block CroV reproduction. However, the released mavirus decreased CroV reproduction in a subsequent round of infection, thereby preventing the spread of CroV in the protist population. Although a cell infected by CroV seems to be doomed, irrespective of the presence of mavirus, the death of this cell can protect related neighbouring cells from destruction by the giant virus (Fig. 1).

Figure 1: A parasitic virus can protect a host-cell population from viral attack.
Figure 1

Fischer and Hackl3 investigated the relationship between the unicellular protist Cafeteria roenbergensis, the C. roenbergensis giant virus CroV, which infects this host, and mavirus, a parasitic virus that targets CroV by suppressing its replication. a, When the authors infected C. roenbergensis with mavirus and CroV, they observed integration of the mavirus genome into the genome of about one-third of the host cells. b, When cells with integrated mavirus were subsequently infected with CroV, mavirus sequences were transcribed. Mavirus particles then formed and replicated in a membrane-bound structure in the cytoplasm known as the virus factory, where CroV also replicates, but did not block the replication of CroV. The cell eventually broke open and released virus particles into the environment. c, The released viruses were taken up by neighbouring cells. Although nearly every cell infected with CroV seems to die, Fischer and Hackl observed increased survival of the host-cell population in these subsequent cellular infections, presumably because mavirus suppresses CroV reproduction in the virus factory. Thus, mavirus can protect host-cell populations from CroV attack, in a similar way to how the memory of a previous attack can provide protection in acquired immunity, as occurs in microbial CRISPR–Cas systems.

What kills host cells co-infected with CroV and mavirus is unknown. Is it CroV, despite its reproduction being limited by the presence of mavirus in the virus factory, or is it mavirus itself? Reactivation of mavirus in response to CroV infection seems to represent a case of an altruistic defence mechanism of the host, in which a cell dies, releasing mavirus that can then protect the neighbouring population of related cells. The existence and role of altruistic defence mechanisms in such a context remain controversial, given the continued debate about the importance of kin selection in evolution, especially in unicellular organisms9. Nevertheless, potential examples in this category are increasing, and include the activation of programmed cell death by infection10.

Mavirus protection of the C. roenbergensis population against CroV spread can also be regarded as a case of adaptive immunity, which involves immunological memory of past infections. In microbes' CRISPR–Cas-mediated adaptive immune system11, viral DNA sequences encountered during an attack are retained by cells and used to prevent subsequent attacks associated with the same or closely related DNA sequences. The twist involved when mavirus is the memory aid is that the infectious agent is remembered indirectly, by host integration of the virophage DNA sequences, which are expressed only during subsequent encounters with the giant virus.

As with CRISPR–Cas-mediated immunity, it is not clear how a host cell survives to retain immunological memory, given that infection by CroV is usually fatal, as Fischer and Hackl demonstrate. In the CRISPR–Cas system, immunological memory is thought to be formed when cells are infected with defective phage particles12, and such abortive infection might also occur in the CroV–mavirus system. Another possibility is that mavirus genome integration into the host genome can occur in the absence of CroV co-infection.

Many questions remain. Perhaps the most pressing one is: how general is this antiviral defence mechanism discovered by Fischer and Hackl? The genome of the unicellular green alga Bigelowiella natans contains numerous integrated virophage genomes13, and this possibly provides another example of a virophage-mediated host defence. The current observations might be only the tip of the iceberg.

The mavirus virophage belongs to a large class of Polinton-like viruses and mobile genetic elements that can integrate into the genomes of diverse organisms8,14. Do some — or even most — of this class that have not, so far, been observed to produce virus particles provide defence against viral infections? The integration of viral DNA sequences, known as endogenous viral elements, is a common outcome when viruses with RNA or DNA genomes infect cells15. Were these endogenous viral elements obtained fortuitously as a result of aberrant genome integration, or might they have been retained by a host to provide immunological memory? The experiments needed to address these intriguing questions have not yet begun in earnest.

The relationship between C. roenbergensis, CroV and mavirus highlights two fundamental biological principles: the pervasiveness of parasites that have evolved to target almost all replicating biological entities, and the use of altruistic modes of protection that are probably intrinsic to cellular life. There is still much to learn about the mechanisms of attack and defence that operate in these fascinating and complex struggles for survival.



  1. 1.

    On Poetry: A Rapsody (Huggonson, 1733).

  2. 2.

    & ISME J. 10, 1815–1822 (2016).

  3. 3.

    & Nature 540, 288–291 (2016).

  4. 4.

    Curr. Opin. Microbiol. 31, 50–57 (2016).

  5. 5.

    , & Adv. Virus Res. 82, 63–89 (2012).

  6. 6.

    , & Arch. Virol. 161, 233–247 (2016).

  7. 7.

    & Science 332, 231–234 (2011).

  8. 8.

    , , , & BMC Biol. 13, 95 (2015).

  9. 9.

    , & Proc. Natl Acad. Sci. USA 110, 20135–20139 (2013).

  10. 10.

    , & Curr. Biol. 26, R587–R593 (2016).

  11. 11.

    et al. Science 353, aad5147 (2016).

  12. 12.

    , & Nature Commun. 5, 4399 (2014).

  13. 13.

    , & Proc. Natl Acad. Sci. USA 112, E5318–E5326 (2015).

  14. 14.

    & Nature Rev. Microbiol. 13, 105–115 (2015).

  15. 15.

    & Curr. Biol. 26, R427–R429 (2016).

Download references

Author information


  1. Eugene V. Koonin is at the National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, Maryland 20894, USA.

    • Eugene V. Koonin
  2. Mart Krupovic is in the Department of Microbiology, Institut Pasteur, Paris 75015, France.

    • Mart Krupovic


  1. Search for Eugene V. Koonin in:

  2. Search for Mart Krupovic in:

About this article

Publication history



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.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing