Viruses and the Tree of Life

We offer here a post by Vincent Racaniello, originally published on his Virology Blog—a way to bring to your attention both a thoughtful review article and a fine blog.

by Vincent Racaniello

A highly resolved Tree Of Life, based on completely
sequenced genomes and generated including the
"Genome size" dataset. Source.

Are viruses living entities? I don’t believe so, but it’s an engaging question for debate. On a recent episode of This Week in Virology we concluded that it’s somewhat of a futile argument because everyone has their own view; perhaps our time is better spent studying viruses than discussing whether or not they are alive. The authors of Ten reasons to exclude viruses from the tree of life cogently disagree: We believe that considering viruses alive or not is not just a matter of opinion, contrary to a commonly held view, but rather is a matter of inference and logic starting from any given definition of life.

Following is my distillation of why the authors believe that viruses are not alive, and why they do not represent ancient lineages of the tree of life.

1. Viruses are not alive
Throughout history there have been many definitions of life. Viruses do not meet the criteria for any of them. They lack any form of energy, carbon metabolism, and cannot replicate or evolve. They are reproduced only within cells, and they also evolve within cells. Without cells, viruses are “inanimate complex organic matter”.

2. Viruses are polyphyletic
In a phylogenetic tree, the characteristics of members of taxa are inherited from previous ancestors. Viruses cannot be included in the tree of life because they do not share characteristics with cells, and no single gene is shared by all viruses or viral lineages. While cellular life has a single, common origin, viruses are polyphyletic─they have many evolutionary origins.

3. There are no ancestral viral lineages
No single gene has been identified that is shared by all viruses. There are common protein motifs in viral capsids, but these have likely come about through convergent evolution or horizontal gene transfer.

4. Because today’s viruses infect phylogenetically distant hosts doesn’t mean that they are ancient
It cannot be proven that early viruses appeared along with the first cells. The problem is that viruses move readily between diverse hosts. Consequently the ability of a virus to infect a particular species could lead to false conclusions about the ancient origin of viruses.

5. Viruses don’t have a structure derived from a common ancestor
Cells obtain membranes from other cells during cell division. According to this concept of ‘membrane heredity’, today’s cells have inherited membranes from the first cells that evolved, and provides evidence that cells are derived from a common ancestor. Viruses have no such inherited structure.

6. Viral metabolic genes originate from cells
Many viral genomes encode proteins involved in energy, carbon, and cellular metabolism. It has been argued that the presence of these genes indicates that viruses are ancestral to cells. Unfortunately, metabolic genes are not present in the ancestors of these viruses. This finding makes it difficult to argue that viruses predate cells.

7. Viral translation genes originate from cells
The mimivirus genome harbors genes encoding elements of the protein synthesis machinery. This observation has been interpreted to mean that viruses did not always rely on cells for translation. However, sequence analysis indicates that these genes are derived from cells by horizontal gene transfer.

8. Viruses steal genes from cells
Viral genomes encode many genes that have no homologues in cells. It has been suggested that viruses influence evolution of cells by donating new genes. Sequence analysis has failed to provide support for this hypothesis. Viruses are ‘gene robbers, not gene inventors and massive gene suppliers’.

9. Most gene transfer goes from viruses to cells
The movement of genes is mainly in the direction of cells to virus. Transfer in the reverse direction is minimal. Therefore viruses have not had a significant role in shaping the gene content of cells. This conclusion is paradoxical because viral genomes are the most abundant on earth.

10. Just because viruses are simple doesn’t mean that they are old
One view of evolution is that it is a process by which simple organisms become more complex. The simplicity of many viruses lead to their placement at the origin of life. This long-standing hypothesis ignores the fact that viral genomes are subject to selective pressure to maintain minimal size to ensure rapid replication rates. The authors conclude that viral simplicity is a consequence of parasitism, not antiquity. Even though viruses are not living and should not be included in the tree of life, they play an important role in evolution of their cellular hosts by regulating populations and biodiversity.

Are you convinced by these arguments? Post a comment and let us know whether you think viruses are living or not.

Vincent Racaniello is professor of Microbiology at Columbia University Medical Center, virus aficionado, host of the Virology Blog, and co-host of a weekly podcast, This Week in Virology

Comments

[Joe The Anthropologist]

If Viruses themselves are not living, then they must me controlled by something that is, that is the ultimate solution. other then that I would like to say there are many flaws in this conclusion, some which have already been stated.
Now this question may seem childish, but I must ask:
If Viruses are not living how then can such things do what they do, with no form of intelligence or lifeleness what so ever? If their purpose is to kill then why is this their purpose? The only explanation is that something smaller or bigger that is living controls the virus.

[theDude]

I think this would be more powerful if you picked a common definition of life and analyzed with respect to that, rather than using the "throughout history, many definitions" approach.

The reader will not exhaust all historically significant definitions of life, and naturally, would suspect you have not done so either - rendering any conclusions drawn here as suspect as well.

Not to take away from the content, which was interesting. But I still am not convinced one way or the other (that "life" does or does not, should or should not, include "viruses".)

thats my unqualified 2 cents

Elio's reply:
We would pick a definition of life if we could. I'm not aware of a commonly agreed one. Different ones continue to appear. The funny thing is that everyone knows what "life" is, even though it escapes definition. That's true for "viruses." You know what they are, and so do I, but defining them is another matter.

There is a gap between "knowing" and "defining." I'm no philosopher, so I wonder about why defining things is so difficult.

[Nathan Myers]

I wonder if viruses originated as organ structures of bacteria carrying what amount to messages, and only later acquired the genes to code for themselves. If that's true, it might be that most virus particles in certain environments would be found not to contain those genes, which would be found instead in the genomes of the bacteria that use the virus particles for their own ends.

We know that anthrax depends for survival in soil on being "infected" with viruses it finds there that carry the genes it needs. Can we think of the population of such viruses as a library left behind in each micro-environment by the anthrax, to be visited by its sister-descendants?

Merry replies:

An insightful idea that speaks to the close inter-relationship between viruses and hosts. A simple pathogen/host model doesn't cover it. You might find---as I did---the story of the "polydnaviruses" especially intriguing.

http://schaechter.asmblog.org/schaechter/2009/03/30000-parasitoids-cant-be-wrong.html

[Eric J. Johnson]

Obviously in philosophy there is room for views at variance. But I've always thought this question about viruses pretty clearly just doesn't matter. What changes if we judge they are alive or not alive? Nil, since we already know on a more fundamental and precise level, very clearly and with reasonable certainty, how life, or {life + viruses} if you prefer, works. Aliveness is not an inbuilt trait of the objective universe, so it's no shock that it's a fuzzy predicate. In contrast, the fact that there are atoms, which can form macromolecules that self-replicate with rare errors via elaborate electromagnetic interactions, is an inbuilt trait of the objective universe - and it just seems to be a clearer and better idea than "aliveness" in every possible respect. Likewise, the universal phylogenetic tree is a fine thing but it's just something we decided to make up: it's only the brute historical facts of the origins of the kingdoms, though they are not totally knowable, that are fundamental and real.

It's not that I think no philosophical questions matter. The mind-body problem, for example, seems to matter a lot, as the answers to it could really change things: are all our commonsense intuitions about personhood and freedom really false, at least within the framework of science (I tend to think so, being a reductionist)? Science obviously is the incomparably superior explanation for everything I perceive outside my own self, but does that really imply I must accept scientific truth as the truth of the whole universe, unquestionably truer than my intuition of the freedom of my will? Why? Just because science is so impressive? - other things in life have impressed me too... Etc.

Another semi-intriguing biophilosophy question - is a highly non-cooperative gene, such as a meiotic drive gene, a part of the organism it's in? In a way, no - it's almost more like a virus, very loosely. But really, nothing whatsoever will change based on one's answer to this question. We already fully know or at least already are fully ready to discover what this particular meiotic drive gene is, what it does, what it's made of, and where it came from, and we can measure and understand its conflict with the organism.

[Welkin Johnson]

I can accept that viruses are not living, but why would polyphyletic origins exclude something from a phylogenetic tree? All this means is that each novel viral lineage has to be considered a distinct branch - technically, a separate kingdom! (yee haw, viruses rule!!). The problem here is only terminology, in that the working classification ("virus") is not a phylogenetic classification, but rather a biochemical one. If all flying critters were routinely called "Wingoids", one might pointlessly argue that Wingoids are polyphyletic (e.g., bats, birds and insects emerged on different parts of the Tree of Life). Virions, like wings, are parallelisms.

As an obligatory intracellular parasite, the evolutionary history of a viral lineage is inseparable from the evolution of its hosts. If you are a virologist, to ignore all those interesting tidbits (items 2-10) would be ridiculous, yet they are conclusions that can be inferred from, or studied by reference to, host phylogeny. The problem with including them is only ideological - if you do, you can't call it the Tree of Life (if you accept that they are not living). Perhaps the tree needs a toggle switch so you can flip to the more inclusive (and more interesting) Tree of Biology.

[Mathgon]

I read the Moreira-Garcia publication with attention last month. Before this reading, I agreed the P.Forterre theories (same lab as Moreira-Garcia) about the origin of Life and proposal of a dichotomic tree with two branches: one for cellular organisms and the other for capsid organisms. Now I'm a little bit disappointed, especially by Mimivirus and Mamavirus that may be in affected by another virus!

That's to stimulate our curiosity and for the discovery of the Viroblog which is now on my RSS aggregator.

[Dale Hoyt]

No. 1 states that viruses do not evolve, yet no. 10 states that "...viral genomes are subject to selective pressure...". You can't have it both ways. No. 3 also states that common protein motifs in viral proteins have evolved convergently, again invoking some form of evolution that no. 1 says that viruses are incapable of.
No. 9 doesn't seem relevant to the issue, unless you think that only living things can transfer genes. Similarly for no. 8.

Comment on Comment
We hope that others will pipe in. The main issues to you seems to be that viruses evolve. The authors don't dispute viral evolution, but say: "viruses neither replicate nor evolve, they are evolved by cells."
Elio