An electron micrograph showing a portion of a bacterium covered with viruses
A horde of viruses have deployed their payloads into an unfortunate bacterium. By Dr Graham Beards - en:Image:Phage.jpg, CC BY-SA 3.0,

You may think that a virus is a virus, and they all afflict living things more or less the same. Yet while it is true that viruses attack most everything on Earth, the viruses that do the vexing differ widely in their makeup and their distribution among the domains and kingdoms of life.

Have a look at these pie charts, for instance, which show the proportion of various virus types found amongst the major groups of life, but don't worry about the details yet.

"The abundance and diversity of viral lineages in the domains of life." Protista includes animal-like protists and brown algae; IP = Invertebrates and plants, who often share viruses because invertebrates feed on plants; Fungi includes all fungi and fungi-like protists; Plants include all plants, green algae, and diatoms; and Metazoa are the animals. Fig. 1A from Nasir et al. 2014.

To understand them, it’s necessary to say a little bit about DNA and RNA, the nucleic acid information storage molecules that run life on Earth.

You are assuredly familiar with DNA, which constitutes our own genomes and that of all other cellular life. RNA – close kin chemically -- is the material by which DNA converts its gene database into active proteins with the help of translation machines called RNA polymerases and ribosomes.

Normally, DNA is double stranded, and RNA is single stranded.

Difference DNA RNA-EN BW.svg
By RNA-comparedto-DNA thymineAndUracilCorrected.png: Users Antilived, Fabiolib, Turnstep, Westcairo on en.wikipedia Difference DNA RNA-EN.svg: Sponk derivative work: Leyo - RNA-comparedto-DNA thymineAndUracilCorrected.png Difference DNA RNA-EN.svg, CC BY-SA 3.0,

But in viruses, those rules are chucked out the window.

Viruses can use double- or single-stranded DNA, or double- or single-stranded RNA as their genetic material. In the case of retroviruses, they can convert a single-stranded RNA genome back into double-stranded DNA with the help of a special enzyme called reverse transcriptase. They then insert this new DNA into our own genomes, where it gets copied each time our own cells divide as a genetic freeloader. The sequences can re-emerge from hibernation many years later to plague us once more; chickenpox acquired in youth can become shingles in old age. We are plagued by other retroviruses, including HIV, and so are many other plants and animals.

The varied nucleic acids viruses employ can be arranged inside their capsules in many ways. Among cellular life, bacteria and archaea – similarly shaped and sized microbes that have radically different genetics than bacteria but are just as prevalent in the environment -- have circular genomes, while the rest of us divide our genomes into many linear pieces called chromosomes. Virus genomes can be linear, circular, or segmented into pieces just as ours are.

Although viruses are unabashedly selfish creatures that seek only to propagate their genes, the *effect* they have on other organisms can have nothing to do with this single-minded drive.

For instance, let’s look at this chart again.

"The abundance and diversity of viral lineages in the domains of life." Protista includes animal-like protists and brown algae; IP = Invertebrates and plants, who often share viruses because invertebrates feed on plants; Fungi includes all fungi and fungi-like protists; Plants include all plants, green algae, and diatoms; and Metazoa are the animals. Fig. 1A from Nasir et al. 2014.

You’ll notice a few striking things. Archaea and bacteria are almost solely affected by double-stranded DNA viruses, and bacteria only rarely by the RNA viruses that plague the rest of life. Fungi and plants, on the other hand, are seemingly almost immune to negative-sense RNA and double-stranded DNA viruses. Retroviruses afflict multicellular organisms but nowhere to be found in microbes. Some of this may be due to undersampling, especially in groups like archaea that are undersampled in general, but vast biases in virus distribution undoubtedly remain. What does it mean?

The authors of a provocative 2014 opinion article in Recent Discoveries in Evolutionary and Genomic Microbiology from which this graphic was taken argue that it means that viruses can be significant drivers of evolution – significant enough to cause major evolutionary shifts in groups of host organisms.

The authors mined the Viral Genomes Resource at the National Center for Biotechnology Information to come to these conclusions. They examined the host preferences of viruses with various replication strategies, as well as the physical shapes of the different virus families affecting different kingdoms and domains to see if any patterns emerged. They did.

Although few archaeal viruses were found, that was clearly due to under-sampling, because very few arcahaea have been screened for viruses. In at least one case, four different viruses from four different families have been isolated from a single archaeum, Aeropyrum pernix, so clearly much more diversity remains to be found.

Even based on what little we know, archaeal viruses come in a much wider variety of shapes than bacterial viruses do. In the diagram below, also reproduced from the opinion article, the various viral shapes unique to given groups or shared by two groups are shown. I apologize for the small type, but given the design of the blogs here at Sci Am, I can't make them any bigger.

Viral shapes unique to a domain or shared between domains. Shared shape does not necessarily imply the viruses belonging to the different groups are related, thanks to convergent evolution. Fig. 1B from Nasir et al. 2014.

As you can see at the top of the page, archaea have four virus shapes that are unique to their domain (droplet-, bottle-, coil-, and spindle/lemon-shaped), whereas bacteria have none. In terms of total morphologies, 16 viral morphotypes attack archaea, while just nine attack bacteria. Archaeal viral diversity is only expected to grow as our ability to isolate the microbes from extreme or unusual environments improves, and we test more of those organisms for viruses.

RT8-4 scale.jpg
A virus shape unique to archaea: the archaeon Sulfolobus infected with Sulfolobus tengchongensis Spindle-shaped Virus 1 (STSV1), isolated by Xiaoyu Xiang and colleagues in an acidic hot spring in Yunnan Province, China. Public Domain,

Yet in spite of all this viral diversity among archaea, to our knowledge, archaea completely lack RNA viruses. RNA is inherently less stable at high temperatures than DNA (part of the reason that it was selected as the genetic material for life). Many archaea also thrive in high temperatures, and it is likely that at some point in the past archaea passed through a phase where they lived exclusively in hot water.

The authors posit that this may have been an archaeal attempt, at least in part, to escape RNA viruses. If so, the gambit proved not only hugely successful, but so far permanent. RNA viruses are also present in both eukaryotes and bacteria (though in extremely limited numbers in the latter), and a single loss from the archaea seems more likely than two independent gains in both bacteria and eukaryotes, further bolstering the case. RNA viruses may thus have been “one major trigger” for the evolution of modern archaea.

Bacteria teem with DNA viruses (in the ocean, bacterial viruses outnumber bacteria), but unlike archaeal viruses, there is little taxonomic or physical diversity among them. Ninety-five percent of double-stranded DNA viruses that attack bacteria belong to a single order split into just three families. The relatively small viral diversity in bacteria may be due to the bacterial invention of the peptidoglycan cell wall, a barrier built of cross-linked sugars and amino acids. The inability to get across this fortification may have eliminated many viral taxa from bacteria, the authors suggest.

On the other hand, no double-stranded DNA viruses are known to attack fungi, and they are rare in plants (only green algae have them). Why? Both fungi and plants possess cell walls. Like the bacterial cell wall, is it possible these walls originally evolved, at least in part, as a way of keeping these viruses out? Cell covering with various layers and rigid cell walls “greatly limit means of viral entry” and effectively shut viruses out, the authors write.

Unfortunately, the authors note, a niche rarely stays vacant long, and the loss of the double-stranded DNA viruses from plants and fungi seems to have spurred the evolution of a cornucopia of RNA viruses in their stead.

Retroviruses may also have inadvertently tinkered with evolution in ways that had profound effects on other organisms. As you'll recall, retroviruses have a habit of plonking their DNA into our own genome and then evack-ing it later -- a process that is not always carried out with surgical precision or much care for *where* the virus does the inserting.

Such an insertion (and later extraction) may change our own DNA. It may change how genes are expressed, or otherwise rearrange the genome, producing new genes and cellular machinery in the process. Sometimes viral genes get left behind and never extracted. For example, telomerase, the enzyme that cells use to repair the caps on their chromosomes from the natural shortening that occurs with aging, seems to be descended from retroviral proteins that were incorporated into host genomes and then much later co-opted by the host.

Because viruses can inadvertently incorporate part of a host's genome into their own, and then reintegrate themselves into another organism, they can also increase biodiversity and complexity by providing a source of inter-species genetic diversity on which evolution can act that might otherwise be impossible.

Looking at all these examples, and the pie chart showing that more physically complex organisms host more different kinds of viruses, one is tempted to suggest that co-evolution between viruses and hosts might have helped generate physical complexity on Earth. Indeed, in addition to retroviruses, negative-sense single-stranded RNA viruses and many DNA virus families are found only in eukaryotes. And though only plants, animals, and fungi possess retroviruses, yeast, which are fungi but live like single-celled microbes, intriguingly lack them.

In this way, viruses might very well present a double-edged sword: potentially devastating to individuals, but beneficial to the evolution of diversity and complexity, if you like that sort of thing, as I do. The next time I have norovirus, I am sure I will feel differently.

This post inspired by a short post over at Small Things Considered. Thanks again for the idea, guys!


Nasir, Arshan, Patrick Forterre, Kyung Mo Kim, and Gustavo Caetano-Anollés. "The distribution and impact of viral lineages in domains of life." Recent Discoveries in Evolutionary and Genomic Microbiology (2015): 26.