I am hoping those of you interested in junk DNA would be so kind to
reply to me with comments.
Thanks!
Thom Quinn
------------------
The Mutation Buffer Hypothesis and The Selective Advantage of Non-Coding
DNA
Thomas Quinn
Evolutionary Theory Research Group
P.O. Box 510373
New Berlin, WI 53151
quinn@evolutionary.org
swo@execpc.com
Abstract
All eukaryote cells posses extremely large amount of non-coding DNA,
which does not appear to be directly or indirectly involved in
transcription. A number of models (i.e. junk DNA, selfish DNA, skeletal
DNA, and nucleotypic DNA) have been proposed to explain the existence of
these non-informational sequences. Genomes are complex entities and it
should be noted that the division between coding and non-coding
sequences is an artificial simplification of the real-world situation.
It is likely that the various kinds of non-coding sequences such as
introns, pseudogenes, transpons, and teleomeres all have different
evolutionary histories. Since there is a great diversity of non-coding
sequences, the above hypotheses are not necessarily mutually exclusive.
I would like to offer a new model that might explain the maintenance of
non-coding sequences in the genome. The mutation buffer hypothesis
suggests that non-informational DNA has a passive functionality and can
confer a selective advantage to an organism. This model proposes that
the non-coding sequences act as a huge buffer in order to absorb the
majority of mutations to the genome. Since most of the mutations within
coding sequences would be deleterious and potentially lethal to progeny,
the existence of "buffer DNA" could prevent decreases in fitness. With a
large buffer, many mutations would not have any negative effects on the
next generation because they would not alter coding sequences. Although
the mutation rate would remain constant, the number of mutations that
actually influenced the fitness of offspring would be reduced in
proportion to the non-coding:coding ratio. The mutation buffer
hypothesis makes the following assumptions:
1) Point mutations are random events that can happen anywhere
within the genome.
2) The probability of one nucleotide being altered from "X" to
"Y" via mutation is identical for coding and non-coding
regions. The bases A, T, C, and G can be substituted for any
X or Y where X is not the same base as Y.
3) If we compare two genomes with identical loci, the one with
the higher percentage of non-coding sequences will also act,
on average, as a stronger mutation buffer.
If the mutation buffer hypothesis can explain the C-value paradox, it
would be expected that K-selected species would have more non-coding DNA
(and a larger buffer) than r-selected species. This is the general
trend. Prokaryote species possess very little, if any, non-coding
sequences while multicellular eukaryotes have extremely high levels of
non-coding sequences. The high cost of replicating, maintaining, and
storing huge genomes would regulate the upper limit of genome size and
the amount of buffer DNA.