viruses, evolution, and net traffic

David Shivak shivakd at fhs.csu.McMaster.CA
Thu Oct 9 16:11:55 EST 1997


In article <61avu6$2j8 at net.bio.net>, Ed Rybicki <ed at molbiol.uct.ac.za> wrote:
>Hey, some traffic!!  Wow!
>
>Now seriously: Paul Keese (now of National Univ of Singapore) stood
>up at the last Int Soc Plant Mol Biol meeting in Singapore (21-27
>Sept) with a very interesting theory / proposition to account for
>why most plant viruses have ssRNA genomes (and mammals have all
>sorts).  He maintains (or so my understanding goes) that organisms
>tend to have the viruses that they have defence mechanisms for
>(presumably because otherwise they would be dead).  Thus bacteria
>have mainly dsDNA viruses as they have restriction systems; mammals 
>have all sorts as the selection is on the basis of the virion outer 
>surface (immune system) - and plants have ssRNA viruses because they 
>have evolved mechanisms like gene silencing to shut off excessive 
>production of mRNA.  Fungi have a prepondeance of dsRNA viruses 
>because they have learned to live with them (vertically transmitted, 
>etc.).  VERY interesting...
>
>However, Thomas Hohn (who asked the question) and I independently 
>realised that precisely the opposite may be true: that what we see in 
>an organism is what it CANNOT handle; that is, that classes of 
>organisms may be "genetically immune" to classes of viruses, and what 
>we see in any organism as an infection is what the organism is 
>learning to handle.

It could, of course, be both processes operating.  Viruses that are too 
successful may wipe out a host species without defenses, leaving only the 
species (or individuals) with defences.  They can also learn to live with 
the host so well that they integrate into the host and form a new defense 
system against other similar viruses.  On the other hand, organisms 
with defenses that are very successful against a type of virus may 
exterminate a class of viruses susceptible to the defense.

My own bias is that viral selection is more likely as all known viruses 
_have_ overcome the defenses of their hosts (examples are easy to come by 
in humans, at least).  Host extermination of a virus would be more likely 
with large genome complex viruses that cannot adapt as rapidly to host 
changes.  

Very different experiments are needed to tell which hypothesis is correct.

I find it hard to believe that plants could have merely not been exposed 
to DNA viruses mutable enough to replicate within them sometime in the 
last few million years.  It may be that the "founder effect" is 
reinforced by an initial defense that forms a large stumbling block to a 
virus trying to jump into a new system.  Thus, a virus not only has to 
"happen" to be able to replicate in plants, but also has to be lucky 
enough to evade hypothetical defense X, making the odds against a successful 
interphyla jump an order of magnitude worse.

<snip>
>Seeing as we are so terrestrially oriented in our search for /
>knowledge of viruses, this collection represents most of what we know 
>about viruses: however, as has been amply demonstrated in recent 
>years, there are a vast number of viruses in every litre of sea 
>water, most of which have never been described - and most virus 
>diversity may in fact still be out there, waiyting for 
>characterisation.  Why do algae have such big dsDNA viruses, while 
>land plants do not?  A case of the one that stayed behind developing 
>a new kind of virus?  Viral "founder effects"?
>
>                     Ed Rybicki, PhD  

My own guess is that someone will find an exception that proves the rule 
in land plants, and any DNA virus that replicates in plants will 
demonstrate a neat ability to jump through a previously unrecognized hoop 
in the defense system of plants.

David Shivak
-- 
David Shivak - email shivakd (at) fhs.mcmaster.ca * 
Check out the web site Careers In (and Out) of Science @  
http://www.freenet.hamilton.on.ca/~ae047/employ7.html
Opinions expressed therein and from this address are my own (unless forged!)






More information about the Virology mailing list