WHEN the Herculean task of identifying every single human gene and working
out exactly what each one does is complete, we will understand just 3 per cent
of our genome. That’s not a misprint. Three per cent. By way of comparison,
flipping through 3 per cent of a dictionary takes you halfway through the As, to
about “analyse”. So what’s all the rest of our DNA for?
A gene, roughly speaking, is a sequence of DNA that tells a cell how make a
protein. But most of the cumbersome coils of DNA in our cells don’t contain any
molecular blueprints for making proteins. Most experts have doggedly maintained
that this non-coding DNA is little more than useless garbage that has
accumulated over time. If they’re right, then the vast majority of the DNA on
Earth is meaningless.
A few dissenters have argued all along that this so-called “junk DNA” must
fill some useful role, or evolution would have eliminated it long ago. And now,
two researchers have come up with a surprising explanation for what that role
might be: stuffing. The huge excesses of non-coding DNA, they say, are the
molecular equivalent of rags and straw. Such an inelegant idea seems out of
place in genetics, given the sophistication and subtlety of most of a cell’s
workings. However, studies of a bizarre group of primitive organisms suggest
that it may actually be true.
Advertisement
The junk DNA debate started in the 1950s, when geneticists began quantifying
the DNA content of cells. Within a species, all individuals have the same amount
of DNA, they found—an amount that came to be known as the “C-value” (C for
constant). But C-values vary wildly from species to species with no apparent
rhyme or reason. Among fish alone, C-values vary 350-fold, while some algae have
genomes 5000 times as large as others. Salamanders are genomic behemoths, with
40 times as much DNA as human cells, while the deadly pufferfish goes to the
other extreme, with a genome 500 times as small as ours.
Clearly, this is not merely a case of simple organisms having less DNA and
complex organisms having more. In fact, the evidence shows that complex life
forms do not need more DNA than simpler ones. This conundrum—known as the
“C-value paradox”—has had researchers scratching their heads ever since.
At first, some biologists took it as proof that DNA was not the hereditary
material after all, but that idea was soon dismissed in favour of the notion
that in most organisms, most DNA does not carry instructions for making
genes.
Why is it there, then? Scientists have plenty of theories, but evidence
remains scanty (Â鶹´«Ă˝, 12 August 1995, p 30). Many researchers
think the excess DNA is little more than useless junk. Because this so-called
secondary DNA does not code for proteins that can help or hinder organisms in
the struggle to survive, it cannot be the handiwork of natural selection, they
argue. If they’re right, then the most important agent shaping the genome was
not Darwinian selection, but mundane copying errors that duplicated stretches of
DNA, creating a genome littered with the abandoned, broken hulks of genes.
At the other end of the spectrum, a few iconoclasts argue that while this
extra DNA does not code for genes, it may serve some other function. It might,
for example, help organise chromosomes so the cell can access the right genes at
the right times—an idea espoused by Emile Zuckerkandl of the Institute of
Molecular Medical Sciences in Palo Alto, California. In “collectivist”
explanations of this sort, what matters is not the sequence of the secondary
DNA—that is, the information it contains—but the physical space it
occupies.
Late last year, two evolutionary biologists—Thomas Cavalier-Smith of
Oxford University and Margaret Beaton of Mount Allison University in New
Brunswick, Canada—offered an even simpler collectivist explanation they
call the skeletal DNA hypothesis. By its sheer bulk, they say, secondary DNA
determines the size of the nucleus, the membrane-bound structure within cells
where the DNA resides. In other words, a large genome physically plumps up the
nucleus from within, which helps the cell itself get larger.
Natural selection often tinkers with cell size and other traits, such as
metabolic rate, that depend on cell size. And when cells get larger, they need
more copies of the enzymes and structural proteins necessary to carry out their
normal activities. That means making more RNA copies of the relevant genes,
along with more of the ribosomes that convert this RNA into proteins. Since this
means squeezing more of the enzymes that do this work into the nucleus, larger
cells require larger nuclei.
“Think of a car factory,” says Cavalier-Smith. “To produce more cars per day,
the factory must either make cars more quickly or add more assembly lines to the
plant.” Natural selection has already optimised the speed with which cellular
products are made, he thinks. That leaves more assembly lines as the only way to
increase production. Like an automobile factory that needs room for more machine
tools, he reckons that larger cells’ increased demands on RNA-synthesising
machinery are best met by increasing the work space.
The cell could do this to some degree by rearranging the way DNA is folded
within chromosomes or perhaps by changing how the DNA attaches itself to the
inside of the nuclear envelope. “Like telescopes, genomes have a range of
possible compactions, from all the way open to completely closed,” says
Cavalier-Smith. However, a simpler way would be to increase the total amount of
DNA and use the telescoping to fine-tune the nucleus size for particular types
of cells, he argues.
C the pattern
Sure enough, there does appear to be a tantalising relationship between
C-value and cell size. Wherever scientists have looked, they’ve found that
eukaryotes (life forms whose cells have a nucleus—unlike, say, bacteria)
with bigger cells have larger genomes. Though only a handful of studies
comparing just a few dozen species have been published so far, they have
nevertheless found this pattern holds true in far-flung reaches of the tree of
life, from protists and algae to plants and animals.
But comparisons involving multicellular organisms can be tricky, because
different kinds of cells—blood cells and neurons, say—face different
demands and thus work best with nuclei of different sizes. To simplify matters,
Cavalier-Smith and Beaton chose to test their hypothesis by looking at an
unusual group of single-celled organisms known as cryptomonads.
Cryptomonads appeared several hundred million years ago, when a red alga set
up house within another unicellular eukaryote. The two formed a permanent
partnership, evolving into a single cell with two nuclei. Today, the nucleus
derived from the original host cell handles most cell functions, while the algal
nucleus—called the “nucleomorph”—produces only proteins involved in
photosynthesis.
The co-evolution of these two separate genomes in the same cell for 500
million years gave Beaton and Cavalier-Smith a rare opportunity to test their
skeletal DNA hypothesis. If their hypothesis is correct, only the main nuclear
genome should vary with cell size. The nucleomorph genome, being occupied only
with photosynthesis and not with the overall maintenance of the cell, should
show no particular relationship with cell size. On the other hand, if secondary
DNA is just junk that piles up over time, then both genomes should exhibit
similarly inflated C-values.
The two researchers compared 17 types of cryptomonad that varied in volume
from just under 89 cubic micrometres for the tiny Plagioselmis prolonga
to 1174 cubic micrometres for a colossal species that is still unnamed. No one
knows much about cryptomonad ecology, but based on what is known about other
algae, Cavalier-Smith believes these different cell sizes reflect different
ecological niches. “Think of cell size as their body sizes,” he says. “Just as
elephants and shrews have different niches, differently sized cryptomonads
probably have distinct ecological roles.”
The results couldn’t have been clearer. For the main nucleus, larger-celled
species indeed had more DNA, as would be expected if it served as stuffing
(see Diagram).
The nucleomorph, by contrast, showed no such pattern (Proceedings
of the Royal Society B, vol 266, p 2053). In fact, evolution has stripped
all the nucleomorph genomes down to their barest bones by eliminating almost all
the secondary DNA. For one species, Guillardia theta, only 9 per cent
of the nucleomorph DNA is non-coding.
Selection ought to favour such a streamlined genome because it is quicker and
cheaper to reproduce during cell division, Cavalier-Smith speculates. And that
provides further proof that the extra DNA of the main nucleus must be more than
just accumulated junk. “The successful elimination of secondary DNA from
nucleomorphs seems to prove that most of it could have been eliminated from the
nucleus also, if it really had no function,” says Beaton.
Cavalier-Smith and Beaton are not the first to suggest that some DNA plays a
strictly structural role. The tips and middles of chromosomes have long strings
of short DNA sequences repeated thousands of times over. This so-called
satellite DNA serves as a tethering spot for the machinery of DNA replication
and cell division.
But the idea that secondary DNA as a whole—the vast majority of most
genomes—is physical stuffing has taken many researchers by surprise. That
doesn’t mean it isn’t right. “It certainly sounds reasonable,” says David Roise,
a cell biologist at the Palo Alto Institute of Molecular Medicine in California.
Fast living
Not everyone is convinced, however. While changes in the amount of DNA could
be a response to natural selection acting on cell size, it could also work the
other way around, argues Dmitri Petrov, a biologist at Harvard University. That
is, the amount of secondary DNA might be determined by other factors. The genome
size, in turn, would limit how fast a cell can reproduce and thus help determine
whether its ecological niche would be that of a small, fast-growing species or a
larger, slower-growing one. “A switch to a lifestyle that demands quick
development would be available only to those lineages that already have a small
genome,” says Petrov.
In particular, Petrov says, the amount of secondary DNA a species has could
be determined by the balance between duplication and insertion mutations, which
add DNA to the genome, and deletion mutations, which remove it. Until recently,
most scientists had assumed that genomes have more of the former than the
latter, and that secondary DNA accumulates steadily over evolutionary time until
the extraneous DNA becomes so costly that natural selection caps the
increase.
But earlier this year, Petrov and his colleagues showed that this need not be
the case. They measured rates of deletion mutations in fruit flies and Hawaiian
crickets. They found that the fruit flies lost DNA 40 times faster than the
crickets. This radically different balance between DNA gain and loss may explain
why the fruit flies’ genomes are 11 times smaller than those of the crickets
(Science, vol 287, p 1060).
Cavalier-Smith is quick to admit that it is possible that in cryptomonads,
the nucleomorph genome might tend toward deletion errors, while the main genome
is more prone to duplications. “But there’s no good reason to believe that this
is actually the case,” he says.
Even if it is, though, Beaton and Cavalier-Smith’s work, together with
Petrov’s, still disproves a central tenet of the “junk DNA”
hypothesis—that genomes necessarily accumulate additional DNA over time.
And for those who have quietly believed all along that the vast majority of DNA
on the planet simply couldn’t amount to functionless trash, this is welcome
news.
Zuckerkandl, for one, has always felt that even if the vast stretches of
secondary DNA are made of junk, it is junk that serves a function. “Picasso
improvised when he made a bull’s head out of a bicycle seat and a steering bar,”
Zuckerkandl muses. “Those who believe that there is a great deal of junk in
genomes have perhaps not considered that nature is a super-Picasso.”
