a molecular evolution question re ebola

bhjelle at unm.edu bhjelle at unm.edu
Fri May 26 13:41:42 EST 1995

In article <3q36lt$ieo at agate.berkeley.edu>,
 <salamon at allele5.berkeley.edu> wrote:
>bhjelle at unm.edu wrote:
>: All of the talk about the inherent poor fidelity of RNA-
>: dependent RNA polymerases (1 in 10-3 - 10-4 misincorporation
>: is typical) is true, but in nature we are measuring
>: *predominant* species of viruses, in settings in which the
>: viruses rarely encounter a genetic bottleneck that might
>: favor significant genetic drift. In the absence of such
>: bottlenecks, it is unlikely that RNA quasispecies will
>: change en masse. Only nonsynonymous changes are likely
>Intersting.  So, which is nearer to what you are suggesting:
>The sampling effects on a virus are minimal because
>	a) there are many hosts at one time
>	b) a fair sized population of virus particles enters each new
>	host, and evolves little in a host (not true for HIV)
Can't say that we have enough information to really address
this properly. The explanation I gave is designed to fit
the observation that some RNA viruses remain very stable
when confined to their native ecological niche. However, many
have assumed that the great instability and rapid radiation
of HIV into subtypes is representative of RNA viruses
in general. That is completely wrong.

>A large population of virus (what is an individual? one particle?
>all the virus in one host?) should be less subject drift, but
>should evolve more subtle adaptations.  I am having difficulty
>accepting that viruses do not experience population bottlenecks.
>An outbreak can be due to a single infected individual.  Say the virus
>then returns to a more endemic state somewhere.

Not much evidence that that happens. For recent "outbreak"
viruses (Ebola, morbillivirus, hantavirus) it seems very
unlikely that (1) these viruses have time to change substantially
by virtue of acquisition of a new host (humans), and
(2) they then have a chance to go back into their true hosts,
at least in significant numbers.

There certainly are a large population of different viral
genomes (quasispecies) in each infected host. That is an
inevitable consequence of the low-fidelity polymerases
that are used by RNA viruses. To have one minor variant 
emerge as predominant in another individual host, it
probably would have to go through a bottleneck, or
be advantageous to the virus.

My question
>relates to where it resides such that it does not experience sampling
>effects.  I do understand that you are suggesting it is quite
>normal for there to be a sufficient population (and mixing among
>potential host) that a predominant strain lasts a long time.
>Surely new varieties arise either through strong advantage or
>through some sort of "allopatric" diversification.

There probably are such events going on, and those events
probably encourage genetic drift and instability. But that
instability tends to be greatly overstated.  In
the end, there are a number of viruses that are exquisitely
well-adapted to their hosts. Nonsynonymous changes are
generally poorly-tolerated in those settings. Synonymous
changes are tolerated to a much greater extent, and thus
deer mice from California have FCVs that are much different
than those in Alberta, etc (at the nucleotide level -
at the amino acid level, they are very similar).

>I am wondering why HIV is so diverse (and resides succesfully entirely in
>human populations) while Ebola Zaire (we have had suggested) is
>relatively static (appearing sporadically).  Does the answer lie in different
>molecular/physiologcal mechanisms, or in different population dynamics?
>If it is the latter, it suggests to me that E. Zaire normally resides
>in a large population of extremely well mixed host (over the time of
>several virus "generations").
I think that the virus/host relationship is important. There is
little question that HIV is a newcomer in human populations;
as a result, it must undergo considerable evolutionary selection
before it achieves equilibrium in its new host. By comparison,
HTLV-I probably has achieved equilibrium, after countless
generations of coexistence with man.


More information about the Virology mailing list