91av

“IT HAS not escaped our notice that the specific pairing we have postulated immediately suggests a copying mechanism for the genetic material.” With this throwaway comment in their famous paper, Watson and Crick heralded a biological revolution by finally explaining how genes could be copied and passed on.

But there was another implication that the enterprising pair apparently didn’t spot. The double-helical structure of DNA also means that all our genes have a back-up copy. If one strand gets damaged, the information on the opposite strand can be used as a template to repair it. And it turns out that this is one of DNA’s most important properties.

To be fair to Watson and Crick, you can’t blame them for missing it. Back in the 1950s, no one gave much thought to DNA damage. “Historically, it was assumed for many years that DNA must be a uniquely stable chemical,” says Phil Hanawalt, a biologist at Stanford University in California who co-discovered one of the first DNA repair systems in bacteria back in 1963. If it weren’t, the argument went, then how could organisms survive? That view still prevailed when Watson and Crick worked out the structure of the DNA molecule. “The idea of repair didn’t occur to them, because it wasn’t thought to be necessary,” says Hanawalt.

We now know just how wrong that view was. DNA in living cells is a perilously unstable molecule that requires constant maintenance to keep it structurally sound. Over the past few years, biologists have painstakingly worked out how organisms exploit DNA’s built-in back-up to keep their genomes in peak condition. In doing so, they’ve realised that DNA repair is one of the fundamental processes of life, and that letting it slip – even a little – can have devastating consequences.

As well as being prone to disintegrate of its own accord, DNA suffers a staggering barrage of chemical attacks from both inside and outside the cell. Each genome in each of your cells endures 30,000 damage events each day, according to Tomas Lindahl, a biochemist at Cancer Research UK in London. The average adult human body contains about 10 trillion cells, so on a typical day your DNA will rack up about 300,000 trillion hits.

Most of the chemical vandals responsible for the carnage are produced within your cells as a result of normal metabolism. “Most of it you can’t do anything about,” says Lindahl. Themain culprits are reactive oxygen species – highly aggressive chemicals such as superoxide anions and hydrogen peroxide. These are produced as a by-product of respiration and can cause more than 100 different types of oxidative damage to the bases in DNA. Add on the assaults from UV light from the Sun, chemicals in cigarette smoke, not to mention background ionising radiation, and you might think it’s a wonder that any of us are alive at all.

If this damage were left unrepaired, the effect on cells would be disastrous. One of the biggest problems is mutations that lead to defective versions of the proteins that DNA codes for. Unrepaired damage can also prevent cells from making RNA copies of their genes to turn into proteins. Crucially, it can also block DNA replication, which is essential to dividing cells. And if a repair job is botched or delayed, itcan lead to every cell’s worst nightmare: adouble-strand break, where the helix snaps in two. Without a back-up strand, the cell has great difficulty repairing such breakages accurately.

Given the dire threat posed by damage, it is no surprise that our cells expend a vast amount of energy seeking out and repairing it. More than 130 human genes are either known or predicted to be involved in DNA repair, and that number is still growing. These genes make up at least four different DNA repair systems, or pathways, each specialising in a particular type of damage (see “The main players”).

The best evidence that DNA repair is central to survival comes from studying people with defects in one of the repair pathways. Many have a higher predisposition to cancer. People with the condition xeroderma pigmentosum, for example, can’t repair UV-induced damage and have a 2000-fold increase in their risk of skin cancer. Other defects result in greatly accelerated ageing, or “progeria”. Children with Cockayne syndrome, for example, develop cataracts, deafness and muscle-wasting, and usually die of infections such as pneumonia before they reach their twenties. Over the past few years, researchers have uncovered the genes responsible for many similar ageing disorders, and found that they are intimately linked to DNA repair.

This link is particularly intriguing as it provides long-sought evidence that ordinary ageing is caused by damage to your DNA. The idea goes back to 1956 when Denham Harman of the University of California at Berkeley proposed the “free radical theory” of ageing. Since then, numerous studies have implied that conditions which reduce the production of damaging free radicals, such as an extremely low-calorie diet, are linked with increased lifespan (91av, 25 March 2000, p 20). But concrete evidence linking DNA damage and ageing has been elusive. “People have argued for a link between repair and ageing for many, many years, usually on the basis of very flimsy evidence,” says Alan Lehmann, an expert on DNA repair syndromes at the Genome Damage and Stability Centre at the University of Sussex near Brighton.

But as we discover more about the genes responsible for premature ageing syndromes, the case for a link between genome instability and ageing grows stronger. “All of them point to DNA being the target of ageing,” says Jan Hoeijmakers, a leading DNA repair researcher at Erasmus University in Rotterdam.

It is now clear that patients with Cockayne syndrome have mutations in the genes that coordinate a type of DNA mending called transcription-coupled repair, which prioritises the repair of active genes by summoning repair enzymes to damage in DNA that is being transcribed. As a result the distorted DNA helix goes unrepaired, and RNA polymerases – the enzymes that transcribe genes – get jammed on the strand. The resulting polymerase pile-up permanently blocks the gene. What’s more, stalled RNA polymerases are also a potent trigger of the cell’s suicide mechanism. This accelerated cell death, says Hoeijmakers, means that tissues age and die faster.

The Cockayne work showed that stalled transcription could be a key part of ageing. But there was still no proof that DNA damage was at the root of the problem. Then, in a ground-breaking paper published in May last year, Hoeijmakers and his team provided the first unequivocal evidence that DNA damage is a central cause of ageing (Science, vol 296, p1276).

The team had been studying mouse models of a human disease called trichothiodystrophy, or TTD, which is linked to faulty transcription and bears some of the hallmarks of premature ageing. The mice suffered TTD-like symptoms, ageing prematurely and living only half as long as normal. What was causing the ageing? Problems with transcription could have been involved, but the mice also had problems with a DNA repair pathway called NER. Was DNA damage the key?

To find out, the team took TTD mice and knocked out their NER system completely. Theresults were dramatic. The mice were born small and sickly and died within 3 weeks – just 3 per cent of the normal mouse lifespan. Post-mortems revealed the mice had developed symptoms of ageing at a vastly accelerated rate. There was only one explanation: faulty transcription plus a defective DNA repair system equals fast-forward ageing. That meant that instabilities in the genome caused by DNA damage really do underlie the ageing process. “I’d been very sceptical for a long time, but this has now convinced me that there is really quite a strong link between accumulated DNA damage and ageing,” says Lehmann.

There are still many mysteries to solve, however. One puzzle is why the premature ageing syndromes don’t cause the full range of symptoms of normal ageing. Instead, different syndromes seem to pick and choose different symptoms. People with Werner syndrome, forexample, develop grey hair and wrinkles at a young age, but not senile dementia.

One possibility, suggests Hoeijmakers, is that different body organs have different degrees of sensitivity to defects in specific DNA repair pathways, because each organ produces its own pattern of DNA damage as a result of its unique metabolism. For example, skin cells sustain far more UV-based damage than liver cells, which in turn suffer more damage from metabolic by-products which cause cross links in DNA. This idea is very new, and Hoeijmakers and his team haven’t yet revealed the results of their studies, but the opinion among repair researchers is that they are building a strong case.

“I wouldn’t say this theory is rock-hard fact, but it’s getting pretty compelling,” says Lehmann. Hoeijmakers can’t discuss his results as they are in the process of being published, but he revealed at a cancer-research conference in Cambridge last October that he had discovered a patient with a deficiency in a protein that repairs DNA cross links, and sure enough, the patient showed accelerated ageing predominantly in his liver.

An intriguing idea arising from these findings is that subtle variations in DNA repair genes, as opposed to outright deficiencies, could influence how people age – whether you will be prone to getting diabetes or Alzheimer’s or atherosclerosis, for example. But however appealing this idea may be, there is currently little evidence to back it up.

One disease where a link with DNA damage has been known for a long time is cancer. “Virtually every cancer that has ever arisen has done so as a result of genome instability,” says Stephen Jackson, who studies double-strand break repair at the Wellcome Trust/Cancer Research UK Institute of Cancer and Developmental Biology in Cambridge. And there is still much to learn in this area.

Recent work on cancer and DNA stability has revealed new levels of complexity in the DNA maintenance system. Studies on a number of human syndromes associated with increased cancer risk point to the idea that in addition to the four main repair systems, cells contain several “DNA guardians”, large proteins whose job it is to keep a watchful eye over the genome. These sound the alarm when they spot persistent damage, summoning the correct repair systems and telling the cell to hold its horses until the damage is fixed. This is especially important for dividing cells, which can wreck their chromosomes if they plough through division with damaged DNA.

If repair attempts fail, these guardians tell the cell to stop dividing, or even order it to commit suicide in case cancer-causing mutations are festering in the corrupted genome. Losing one of these genome guardians takes the brakes off genome instability, and can lead to the multiple mutations that may turn a cell cancerous. About half of all cancers, for example, have a crippled genome guardian called p53. The hereditary breast cancer genes BRCA1 and BRCA2 also function as genome guardians, helping the cell perform recombinational repair.

As well as revealing more about the underlying causes of cancer, studies of these repair pathways are opening up ways of improving treatments such as chemotherapy. Most chemotherapy drugs are potent DNA damaging chemicals, which is what makes them so good at killing rapidly dividing tumour cells. But cancer cells have an additional weakness: they can be defective in one or more DNA repair pathways, so are particularly sensitive to drugs that produce particular forms of damage. BRCA1-deficient cells, for example, will have trouble repairing cross links and should be sensitive to drugs that cause this damage. “Therapeutically, DNA repair has a lot to offer,” says Jackson, whose biotech company KuDOS is one of several exploring this area.

Although researchers are making rapid advances in understanding how our genome defenders protect us against harm, their findings are generating as many questions as answers. Why, for example, do mutations in damage-sensing genes such as BRCA1, which are important to all cells, predispose people to cancer in just one or two tissues? “That’s a mystery nobody’s come up with a good explanation for yet,” says Lehmann. Do subtle deficiencies in our DNA repair genes predispose us to particular cancers? Preliminary evidence suggests that they might. Epidemiological studies reveal that people with the most common version of the base excision repair gene XRCC1, for example, have an increased risk of a number of cancers including bladder, prostate and colon cancer, and the risk is increased even more if they smoke, according to Jack Taylor, a molecular epidemiologist at the National Institute of Environmental Health Sciences in North Carolina. But we need to find out how these variations affect the efficiency of DNA repair enzymes, as well as studying the effects of combinations of genes to really understand how they relate to a cancer risk, he adds.

Perhaps the biggest question is, if repairing the genome is so important, why aren’t our cells better at it? Why will 1 in 3 of us get cancer? Like all things in life, DNA repair involves making compromises. For one thing, a sophisticated repair system demands a lot of energy. “The cell has to strike a compromise between how much damage it can tolerate and the cost of repairing damage,” says Hanawalt.

Repairing damage can also be a laborious process, and if your body spent too long on it, you would be faced with the problem of cells dividing too slowly. As a trade-off, cells have developed a number of ways of tolerating a small amount of damage. One of these is a group of “sloppy” DNA polymerases, which can read through a damaged section of DNA during DNA replication. The price we pay for their tolerance is that they often make mistakes. It may be that they are an important cause of the build-up of mutations that can lead to cancer.

But there’s a silver lining. Hanawalt points out that if repair were absolutely perfect, evolution would be stopped in its tracks. “Without mutation, there can be no evolution,” he says. We may suffer for its faults, but if DNA were perfectly stable, we wouldn’t even be here.