ASK a biologist to explain cancer, and you’re likely to get an earful about the relentless cell division driving the disease, but hardly a peep on the everyday processes that make division possible. For years, the less glamorous workings of cell life, such as energy metabolism and cell growth, haven’t carried much cachet in the field of cancer research. But all that may be changing.
Newly discovered links between cancer and the way cells grow and manage their energy budget are transforming our view of the disease. The findings also point to new types of cancer drugs, such as “energy therapeutics” that trick cancer cells into believing they lack the energy to divide, and growth inhibitors that prevent cells growing big enough to divide. Our pharmacy shelves may even hold such medicines already: the latest discoveries suggest that drugs long prescribed for diabetics and transplant recipients may work for some cancer patients as well.
If existing drugs really can help, it will be welcome news for doctors and patients. In the 15-odd years it takes for a new cancer drug to go from the lab bench to medicine cabinet, an estimated 90 million people die of the disease. The long road from discovery to approved therapy is littered with failures, and along the way pharmaceutical companies spend hundreds of millions of dollars just to develop and market one new medicine.
Advertisement
With hindsight, the hints that energy metabolism and cancer are linked have been with us for years. For one thing, people with metabolic diseases, such as diabetes or obesity, are more prone to getting cancer. People with type 2 diabetes are at a higher risk of getting colorectal cancer, for example.
Researchers also know that molecules responsible for sensing and responding to insulin – the hormone that controls blood sugar levels – can also become subverted to promote cell division. Normally, cells in the pancreas secrete insulin after a meal, when blood sugar levels are high. Insulin binds to a protein receptor found on the surfaces of cells throughout the body, and triggers a set of reactions that tell cells to remove sugars from the bloodstream and convert them to a more complex form for storage.
But the insulin signalling system is no one-trick pony. The body can also use it to tell cells to divide, and in many cancer patients, mutations affect components of the insulin signalling pathway. Cancer cells are then fooled into behaving as if insulin is always there, and the end result is excessive cell division. The fact that so many tumours display such mutations underscores how important this pathway is in cancer.
All this fits in nicely with a well-established model for how many cancers work. In multicellular organisms, cell proliferation is controlled by small soluble proteins released under very specific circumstances to prompt particular populations to multiply. Like insulin, these proteins bind to receptors and tell cells to divide, and mutations that affect this signalling often send cell division haywire. In normal cells, these pathways are only activated during the brief periods that the proteins are around, but in some cancer cells, mutations permanently activate the pathways and result in unrestrained cell division.
A wealth of supporting evidence has encouraged cancer biologists to pursue this model, but they have focused on how the protein signals drive cell division, ignoring more generic factors that influence cell growth and proliferation, such as nutrient availability and cellular energy levels. This has led many to assume that these other factors aren’t important in cancer.
But last year, Graham Hardie and Dario Alessi at the University of Dundee in Scotland challenged the old assumptions with a stunning discovery. Not only did their work suggest that energy is a key factor in cancer, it also linked diabetes and cancer in an entirely new way. “It was a very unexpected finding,” says Hardie. “I would never have guessed it.”
Breakthrough
For many years, Hardie had been studying a protein called AMPK, an important monitor of cellular energy levels. A cell’s energy reserves are like the molecular equivalent of a car battery, and AMPK’s job is to sense whether this battery is charged or depleted and to make sure it does not go flat.
Cells try to keep their batteries fully charged, but to make proteins, copy DNA, grow and divide, the cells must use up energy, draining the battery. To keep it topped up, cells extract energy from food, such as sugars, but when food is scarce, the recharging reactions cannot keep up with energy expenditures. The battery charge falls. In response, AMPK switches off the reactions that discharge the battery, and turns up the ones that recharge it.
This generally explains how cells conserve energy when times get tough, but what is telling AMPK to switch on when needed? Cells often use a chemical reaction called phosphorylation to turn enzymes on or off, and evidence suggested that this was what was happening to AMPK. The big question was: what was the identity of the protein that was phosphorylating AMPK? Despite years of searching, nobody had found it.
To solve the problem, Hardie turned to a much simpler organism. In yeast, researchers had identified 119 proteins that were able to perform phosphorylation. Hardie systematically tested each of these for the ability to phosphorylate the yeast version of AMPK. He found one that fitted the bill, and by comparing its gene sequence with the yeast genome, two others that can do the same job.
To find the human equivalent, Hardie compared the yeast genes that coded for his three enzymes with those of the human genome. He was dumbfounded to discover that their closest relative in humans was LKB1, a cancer-related enzyme that his colleague, Alessi, had been studying for years. “It was an amazing coincidence,” says Hardie. “I suddenly realised that the guy who could help me was just down the corridor.”
All that time, while Hardie had been searching for his protein, Alessi had the opposite problem. He was already working on LKB1 and knew it phosphorylated something – he just didn’t know what. What was clear was that LKB1 was important in cancer. It is a “tumour suppressor”, a protein that somehow stops cells from becoming cancerous. Inherited mutations in the gene for LKB1 cause Peutz-Jeghers syndrome, a rare disorder affecting about 1 in 120,000 people.
PJS patients develop benign tumours called hamartomas in their gastrointestinal tracts, but they also have a much higher than normal risk of getting cancer. An estimated 93 per cent will develop malignant tumours, with the cancer appearing at an average age of 43. And LKB1-related cancers are not just confined to PJS patients. About 30 per cent of sporadic lung cancers (usually smoking-related) show non-inherited mutations in LKB1.
Despite LKB1’s obvious role in preventing cancer, the details of its function had been murky. But Hardie’s results now suggested that LKB1’s job might be to switch on AMPK. He and Alessi quickly did the biochemical experiments to test the idea. They found that in a test tube, LKB1 could phosphorylate and activate AMPK, but only under chemical conditions that mimicked a run-down cellular battery. Their results suggested that LKB1 is always active in the cell, but only switches on AMPK when the cell’s battery needs charging.
Fooling cancer
People with Peutz-Jeghers produce only half as much functional LKB1 as normal. This means that their cells are less able or unable to switch on AMPK when their batteries are low, and so are more likely to divide under low-energy conditions that would stop most normal cells in their tracks. Suddenly, LKB1 showed that a cell’s ability to sense energy is an important feature in the development of cancer.
This immediately suggests a new way of tackling cancers of all kinds. What if you could artificially switch on AMPK and fool a cancer cell into thinking that it was starving and so unable to spare the energy to grow and divide? It might not necessarily rid a patient of a tumour, but it would theoretically help to stop it getting any bigger or spreading. This is an especially attractive idea because a drug already exists that could do just that.
Metformin, which is used to treat people with type 2 diabetes, works at least in part by artificially turning on AMPK (see “Treatments to go”). By turning up reactions that recharge the battery, activated AMPK encourages sugar uptake and metabolism, helping to alleviate the high blood sugar levels typical of diabetes. But in cancer patients with normal blood sugar, using metformin to activate AMPK would make the cell switch off the power-hungry cell division process as if it were facing low-energy conditions. So could metformin – an approved drug already commercially available – be used to halt cancerous growth, or at least slow it down?
It certainly makes sense on paper. “It is possible that this would keep some cancers at bay,” Alessi speculates. In fact, since metformin has been widely used for a decade, the experiment has effectively already been done, and presumably the results are lurking in the medical records of people with diabetes. By comparing patients who have taken metformin with those who have not, researchers should be able to figure out if those who took the drug got cancer less often than those who didn’t.
“I’d really love it if somebody did that type of study,” says Reuben Shaw, a researcher at Harvard Medical School who, along with Lewis Cantley, confirmed Hardie and Alessi’s results. “I’m sure we’ll get the answers in the next five years or so.”
The LKB1/AMPK link might also help explain why exercise is known to help prevent and treat both diabetes and cancer. Hardie and others have shown that exercise activates AMPK in human muscles, which presumably increases sugar uptake and metabolism and might prevent or counteract the effects of diabetes. And exercise-induced AMPK activity might also help prevent cancer.
The big question is how AMPK affects the cell’s decision to grow and divide. Last year, researchers in the US uncovered a link between cancer and the way cells budget energy for growth. Their findings not only strengthen the notion that AMPK stops cells turning cancerous, but they also hint at how it might do it. And it seems that for cancer cells, size really does matter.
Size matters
Kun-Liang Guan and his team at the University of Michigan at Ann Arbor were studying two genes that in their faulty mutant forms are involved in another inherited tumour syndrome called tuberous sclerosis, which affects about 1 in 6000 people. They found that the two proteins made by the healthy versions of the genes limit the activity of an enzyme called mTOR, which controls how fast a cell makes proteins and how fast it grows.
żěè¶ĚĘÓƵs already knew that AMPK was somehow involved in the same processes. Might mTOR and AMPK be connected in some way? Patients with tuberous sclerosis suffer similar kinds of tumours to those in PJS patients, hinting that these diseases indeed share a mechanism in common.
Such a link would not be so surprising, says Guan: “Cell growth has to be coordinated to the cellular energy response.” Otherwise, a starving cell might try to grow and divide when it lacks the energy to finish the job, resulting in its death. So it certainly seemed logical that AMPK, which directs the cell’s response to energy levels, and the cell growth promoter mTOR, might be tied together in some way. But no one knew how.
Guan and his team discovered that the tumour-suppressor proteins apparently work together to switch off mTOR under certain conditions, including times when cellular energy levels are low. The thing that tells the tuberous sclerosis proteins that energy levels are low is AMPK. And mTOR responds by shutting down protein synthesis and growth – an apt response to near starvation. In tuberous sclerosis patients, the proteins are mutated and can’t ferry the signal to mTOR, so growth continues when it shouldn’t.
With Guan’s findings, AMPK went from being a modest metabolic regulator to a key factor controlling cell growth and division in response to cellular energy levels, with not one, but two clear connections to cancer. “Everything came together all at the same time,” Guan notes. When crucial components of the pathway are missing, as in PJS and tuberous sclerosis, the cell’s inability to check growth during lean times encourages tumour formation.
As the downstream target of this pathway, mTOR is also emerging from all these discoveries as a star player in the fight against cancer, though some are cautious in their enthusiasm. It may be that mTOR is not the only target of the systems that monitor a cell’s energy. “There could be many hundreds of others,” says Alessi. He has found that LKB1 can phosphorylate 11 other proteins in addition to AMPK. And AMPK is known to regulate a host of other processes besides cell growth. Still, says Guan, “there’s a lot of interest and excitement in the cancer field about mTOR”.
One reason that cancer biologists are excited about mTOR is that a powerful inhibitor of the protein already exists, and may work as an anti-cancer drug. First isolated in the 1970s, rapamycin is used as an immunosuppressant in organ transplant recipients. One form of the drug has already shown promise as a potential cancer therapy in small studies, and it is now being tested in a large-scale clinical trial. Meanwhile, Diane Fingar from the University of Michigan hopes all the interest in mTOR will serve to draw attention to what she believes is its main function as a regulator of cell growth. “It’s an area that has definitely been under-appreciated,” she says.
Historically, researchers looking for cancer treatments have focused on the proteins that control how a cell divides and multiplies, rather than on how cells grow bigger. This is understandable given that cancer is, at least superficially, a numbers problem. One rogue cell cheats death and begins dividing uncontrollably, replacing vital tissues with an expanding horde of equally selfish and prolific clones. Block division and you stop the cancer. “But if you think about it,” says Guan, “without cell growth, you will have no proliferation.” A cell must increase in mass and size some minimum amount before it can divide, as otherwise it would get smaller at each division.
Fingar and Guan believe mTOR encourages proliferation indirectly by promoting cell growth, though there is no real proof for this idea. If they are right, drugs like metformin and rapamycin should work mainly as growth inhibitors and slow the growth of a wide range of tumours. However, it seems unlikely that drugs that affect a cell’s energy balance or growth stop a cancer from becoming invasive and spreading through the body.
This means that such drugs will probably work best when combined with other anti-cancer treatments. Still, drugs like metformin, whose side effects are minimal and far milder than those of conventional cancer therapies, will be welcome in the anti-cancer arsenal.
Treatments to go
Metformin: In use since the 1970s as a drug to treat type 2 diabetes, metformin works by bypassing a cell’s faulty response to insulin and switching on the mechanisms that prompt it to absorb and store sugar from the blood. Now, researchers think that metformin could affect these same mechanisms to fool cancer cells into behaving as if they were starving and lacking the resources to divide.
Rapamycin: Isolated from a soil bacterium from Easter Island, rapamycin was originally identified as an antifungal agent in the 1970s. It was later found to be a potent immunosuppressant and was approved as a treatment for transplant patients in 1999/2000. Synthetic versions of the drug are now going through clinical trials in cancer patients, with encouraging results. The drugs are thought to work at least in part by blocking a cancer cell’s ability to increase in size and divide, and have few serious side-effects.