BOOK REVIEW
The Dark Side of DNA
Fred Gould
Genes in Conflict: The Biology of Selfish Elements. Austin
Burt and Robert Trivers. x + 602 pp. The Belknap Press of Harvard
University Press, 2006. $35.
Although many of us have gotten used to the idea that our bodies
serve the needs of a variety of viruses, bacteria, mites and other
parasitic species, it comes as a surprise to most people when they
hear that their bodies are also hosting alien parasitic DNA.
Analysis of output from the Human Genome Project makes it clear that
just one form of such alien DNA, transposons, makes up about 50
percent of our genome. Every time one of your cells divides, it uses
time and energy to replicate this parasitic DNA. There is even
evidence that the size of your cells is set to accommodate this
extra genetic load. In return, this type of DNA typically does
nothing useful for you or any of the other organisms it inhabits.
So why do humans and the vast majority of other species serve as
homes for parasitic DNA? This is one of many questions about selfish
genetic elements that Austin Burt and Robert Trivers address in
their scholarly, thought-provoking new book, Genes in
Conflict. As can be gleaned from the title, the authors don't
envision an easy alliance between selfish genes and the rest of the genome.
As background, it is worth noting that all specific sequences of DNA
manage to persist over time by causing their host organisms to keep
passing them on to their progeny. There are two basic evolutionary
mechanisms that DNA sequences use to improve their odds of getting
into that next generation. The first method involves increasing the
number of viable offspring produced by the host relative to
competing individuals. This process fits within our typical
understanding of adaptation and natural selection.
The second evolutionary mechanism is for a DNA sequence somehow to
increase the percentage of the host's offspring in which it is
contained. A DNA sequence that is represented by one copy in a
diploid sexual organism is generally expected by Mendelian
principles to be inherited by 50 percent of progeny. If a DNA
sequence can manage to wind up in, say, 90 percent of the offspring,
it has a greater chance of persisting over time—even if that
DNA sequence causes a decrease in fitness. Burt and Trivers consider
genetic elements that use this second strategy to be "selfish
DNA." Of course, there is a balancing act here, because if the
negative impact on fitness is too great, the host species could go
extinct, taking the selfish DNA with it.
Over time, numerous types of genetic elements have evolved diverse
parasitic strategies for proliferation in genomes and, as a result,
have become very abundant in many organisms. But some species appear
to lack selfish DNA of any kind. Whenever selfish DNA reduces
fitness, natural selection will favor any mutant gene that decreases
the probability of that selfish DNA being passed on. Burt and
Trivers provide a long list of examples demonstrating a
coevolutionary battle between selfish genetic elements and genes
that negate their impact. In some species it seems that the selfish
DNA is winning; in others it is taking a beating.
The dynamics of "gamete killer" genetic elements in the
fungus Neurospora, and genes for resistance to these gamete
killers, illustrate this coevolutionary tug-of-war.
Neurospora has a brief diploid phase of its life cycle.
This stage is followed by three cell divisions—two meiotic and
one mitotic—giving rise to eight haploid spores arranged in a
neat row. Researchers found that when they made genetic crosses
between two particular strains of this fungus, four of the resulting
eight spores in a set were always dead. Genetic analysis revealed
that one of the strains, called the gamete-killer strain, has a
tightly linked set of genes that somehow kills the four spores in
which it was not present. The current hypothesis is that during
spore formation, one or more of the gamete-killer genes code for a
toxic substance, which is deposited in all of the forming spores.
Four spores can survive, though, because they inherit genes from the
killer strain that neutralize the toxin.
The important result here is that the gamete-killer genes wind up in
100 percent of the viable spores instead of just 50 percent. If even
one spore in a population had this set of genes, it would in some
situations be expected to increase in frequency until it was carried
by all individuals, despite the substantial decrease in reproductive
capacity experienced when only a minority of spores carry these genes.
A survey of Neurospora populations around the world found
that these gamete-killer genotypes are typically rare but can indeed
reach a frequency of 100 percent in populations of some species.
Genes that confer resistance to the gamete killers are typically
more common than the killer genes. Once a resistance gene enters a
population and negates the ability of the killer genes to be
overrepresented in progeny, the killer genes are selected against if
they cause fitness reduction. So in the case of Neurospora,
the selfish DNA seems to be losing the coevolutionary battle.
These gamete-killer genes, which are also found in invertebrates and
vertebrates, are just one form of selfish genetic element, and they
are far from the most diverse type. Transposons, sometimes called
"jumping genes," are found in a wide array of taxa and
vary dramatically in the mechanisms they use to become
overrepresented in offspring. Burt and Trivers devote an entire
chapter to transposons, which are able to replicate themselves
within a genome and move to new locations on the chromosomes. An
individual organism can start life with one copy of a transposon and
end up with two or more in its germline cells.
Barbara McClintock discovered transposons in 1952, but the first
selfish genetic elements revealed themselves as early as 1906. And
new forms are still being found. A challenge to researchers is to
think creatively enough to anticipate what an as-yet-undescribed
selfish element might look like and how to design experiments for
its recognition. Burt and Trivers discuss a wonderful example of
this creative thinking by the renowned evolutionary biologist George
Williams, who in 1988 wrote,
Perhaps it is just a matter of time before someone
discovers (or invents in the lab) an all-male species. It makes
diploid sperm that inseminate eggs of related species and give rise
to diploid nuclei that exclude the egg pronuclei.
This would seem to be the ultimate in parasitic DNA—an entire
eukaryotic genome that has no body of its own, surviving only by
stealing the eggs of a related species in each generation. Although
it may seem to be the subject matter for a science-fiction novel,
this type of system was recently discovered in a clam, a conifer and
a stick insect. Very cool!
But these examples are only the tip of the iceberg. In their
602-page opus, Burt and Trivers provide a plethora of exciting case
studies. Although there is no lack of data to discuss, the authors
emphasize repeatedly how little we really know about this area of
evolution and biodiversity.
I found the tone of this book to be very engaging. It is full of
details that have been woven together into a very readable,
well-organized package. Of importance for the nonspecialist reader,
Burt and Trivers succeed in conveying complex concepts in population
genetics without using mathematical equations. Detailed mathematical
treatment of these topics is certainly warranted, but that will be
another book. The authors clearly reveal their attitude in the first chapter:
We review the evidence in more detail than we can make
sense of and we describe logic beyond what the evidence will verify.
Our aim is to strike a balance between what is known and what is
not, the better to invite others to join in generating the missing
logic and evidence.
What a gift to graduate students and all researchers who are just
entering this field of evolutionary biology! I found at least a
dozen good projects for Ph.D. theses suggested within the pages of
this book, and I am sure that there are many more.