By Julianna Kettlewell
BBC News Online science staff
A collection of mystery DNA segments, which seem to be critical for the survival of many animals, are causing great interest among scientists.
Humans and rats share large amounts of DNA
Researchers inspecting the genetic code of rats, mice and humans were surprised to find they shared many identical chunks of apparently "junk" DNA.
This implies the code is so vital that even 75 million years of evolution in these mammals could not tinker with it.
But what the DNA does, and how, is a puzzle, the journal Science reports.
Before scientists began laboriously mapping several animal life-codes, they had a rather narrow opinion about which parts of the genome were important.
According to the traditional viewpoint, the really crucial things were genes, which code for proteins - the "building blocks of life". A few other sections that regulate gene function were also considered useful.
The rest was thought to be excess baggage - or "junk" DNA.
But the new findings suggest this interpretation was somewhat wanting.
David Haussler of the University of California, Santa Cruz, US, and his team compared the genome sequences of man, mouse and rat. They found - to their astonishment - that several great stretches of DNA were identical across the three species.
To guard against this happening by coincidence, they looked for sequences that were at least 200 base-pairs (the molecules that make up DNA) in length. Statistically, a sequence of this length would almost never appear in all three by chance.
Not only did one sequence of this length appear in all three - 480 did.
The regions largely matched up with chicken, dog and fish sequences, too; but are absent from sea squirt and fruit flies.
"It absolutely knocked me off my chair," said Professor Haussler. "It's extraordinarily exciting to think that there are these ultra-conserved elements that weren't noticed by the scientific community before."
The really interesting thing is that many of these "ultra-conserved" regions do not appear to code for protein. If it was not for the fact that they popped up in so many different species, they might have been dismissed as useless "padding".
DNA: THE CODE OF LIFE
The double-stranded DNA molecule is held together by chemical components called bases
Adenine (A) bonds with thymine (T); cytosine (C) bonds with guanine (G)
Groupings of these letters form the "code of life"; there are close to 3 billion base pairs in mammals such as humans and rodents
Written in the DNA of these animals are 25,000-30,000 genes which cells use as templates to start the production of proteins; these sophisticated molecules build and maintain the body
But whatever their function is, it is clearly of great importance.
We know this because ever since rodents, humans, chickens and fish shared an ancestor - about 400 million years ago - these sequences have resisted change. This strongly suggests that any alteration would have damaged the animals' ability to survive.
"These initial findings tell us quite a lot of the genome was doing something important other than coding for proteins," Professor Haussler said.
He thinks the most likely scenario is that they control the activity of indispensable genes and embryo development.
Nearly a quarter of the sequences overlap with genes and may help slice RNA - the chemical cousin of DNA involved in protein production - into different forms, Professor Haussler believes.
The conserved elements that do not actually overlap with genes tend to cluster next to genes that play a role in embryonic development.
"The fact that the conserved elements are hanging around the most important development genes, suggests they have some role in regulating the process of development and differentiation," said Professor Haussler.
Rethinking "junk" DNA
The next step is to pin down a conclusive function for these chunks of genetic material.
One method could be to produce genetically engineered mice that have bits of the sequences "knocked out". By comparing their development with that of normal mice, scientists might be able to work out the DNA's purpose.
Despite all the questions that this research has raised, one thing is clear: scientists need to review their ideas about junk DNA.
Professor Chris Ponting, from the UK Medical Research Council's Functional Genetics Unit, told BBC News Online: "Amazingly, there were calls from some sections to only map the bits of genome that coded for protein - mapping the rest was thought to be a waste of time.
"It is very lucky that entire genomes were mapped, as this work is showing."
He added: "I think other bits of 'junk' DNA will turn out not to be junk. I think this is the tip of the iceberg, and that there will be many more similar findings."