Every breath: Peter Ratcliffe and oxygen levels
In 1978, the daring climber Reinhold Messner wrote: "In my state of spiritual abstraction, I no longer belong to myself and to my eyesight. I am nothing more than a single narrow gasping lung, floating over the mists and summits."
This was his poetic verdict on becoming, with his climbing partner Peter Habeler, the first adventurer to reach the summit of Everest without oxygen. Despite suffering bouts of extreme exhaustion, confusion and snowblindness, they had overcome what high-altitude physiologists - and even Sherpas - thought was an impossible challenge for the human body.
In fact, a lot of what physiologists believed about the human body's response to oxygen changed dramatically in the 20th century. Much of the credit for this rethink must go to Professor Peter Ratcliffe and colleagues at the University of Oxford. Ratcliffe's research, funded primarily by the Wellcome Trust, has centred on investigating how individual cells in the body sense and respond to changes in oxygen levels. Not that you would find him clambering up an icy peak to investigate it first-hand: "Last summer my family and I climbed El Misti (a 5800-metre volcano in Peru) and I wasn't too good on the mountain," he confesses. "I did climb it but my wife and children climbed it before me. It involved trying to sleep at 4800 m, which was difficult."
While it has always seemed obvious that oxygen deprivation was something to be avoided - hypoxia (low oxygen levels in the tissues) is associated with heart disease, stroke and cancer - too much oxygen is also toxic. In the 1950s, for example, many premature babies were left blind or partially sighted after hospital treatment; it turned out that supplemental oxygen disrupted normal development of the retina.
But, just as there are direct and indirect ways of gauging an oven's temperature - you can look for secondary evidence that it's hot enough (is my cake rising?) or, more directly, open the oven door and stick your hand in - scientists weren't sure whether cells sensed oxygen directly or via its effects on metabolism. Breathing, for example, is mainly regulated by metabolism. It looked as if metabolic regulators were the physiological middlemen - mediating between oxygen and all its effects. "Even understanding the pathogenesis [of blindness in premature babies exposed to high oxygen levels] didn't clearly implicate oxygen sensing as opposed to metabolic sensing," Ratcliffe explains. "It took a long time in the 20th century for the idea to develop that oxygen itself might be sensed."
During this time, academics were convinced the kidneys contained an oxygen sensor. The reason is that a lack of oxygen prompts the body to make the hormone erythropoietin, which in turn stimulates the production of red blood cells. This all happens in a subset of cells found in the kidneys - which is how Ratcliffe, then a trainee kidney specialist, became involved. In 1989, supported by a Senior Fellowship from the Wellcome Trust, he founded a laboratory specifically to explore the cellular oxygen-sensing enigma.
Erythropoietin's response to oxygen is extremely fast and sensitive, and isn't turned on by metabolic poisons. So, biologists began to wonder whether erythropoietin was, in fact, responding directly to a dedicated oxygen sensor. The erythropoietin gene had just been sequenced, and Ratcliffe had the idea of inserting its control sequence - the piece of DNA that is used to switch production of the hormone on or off - into non-kidney cells. He and his colleagues were in for a shock: the control sequence worked in other cells, not just in erythropoietin cells. He explains: "When we inserted that gene into other cells, it had the same oxygen-regulating effect. That meant our thinking [that oxygen regulation was carried out only by a select group of kidney cells] was wrong. In fact, all cells in the body can do this." The implication was that, as the control sequence didn't specifically make erythropoietin, it must have been doing something else. "That changed the field and changed our programme," he recalls.
His group was later able to show that these cellular oxygen-sensing pathways exist in all animals, even ones that don't produce erythropoietin or red blood cells, such as fruit flies and nematode worms. And if these pathways weren't producing red blood cells, what were they doing? Further studies found that this oxygen-sensing system was actually influencing a host of fundamentally important cellular processes, such as metabolism, cell migration and even cell survival. Professor Gregg Semenza at Johns Hopkins University School of Medicine in Maryland would later discover, characterise and clone HIF, or hypoxia inducible factor, the protein that makes the oxygen-sensing system work.
But Ratcliffe's interest didn't stop there: he then wanted to detail exactly how cells were detecting oxygen. When oxygen levels fall, levels of HIF shoot up. This switches on a multitude of genes (around 1000 are now known to respond to HIF activity) to counteract the hypoxia. Ratcliffe's laboratory, along with Professor William Kaelin's at the Boston-based Dana-Farber/Harvard Cancer Center, turned to the molecular puzzle of precisely how oxygen and HIF interacted. Homing in on the exact piece of the HIF protein that did the interacting was difficult, however: three distinct stretches of the protein appeared to be involved.
Their big leap forward lay, again, in renal science: clinicians had noted that kidney cancers were particularly associated with upregulated angiogenesis, or a growth spurt of new blood vessels. As blood ferries oxygen around, this suggested an abnormally large oxygen supply. Ratcliffe explains: "The fact that these tumour cells, in common with other tumours, have upregulated angiogensis led people to think there might be a link to an abnormal hypoxia pathway. We eventually pinned the mechanism of that link down to the tumour suppressor protein that's mutated in the cancer. In normal circumstances, this tumour suppressor protein is a protein destroyer: it destroys HIF in the presence of oxygen as HIF is not needed when oxygen is plentiful. But when this tumour suppressor protein is abnormal and won't work, as in cancer, then HIF survives and the hypoxia pathway switches on." In this case, HIF is like a fire alarm that can't be switched off, and the body responds by ramping up blood supply to the area and possibly feeding the tumour.
Analogies like this don't do justice to the delicacy and sensitivity of the experiments required to pin down the exact protein interactions involved. These experiments were only possible because the Wellcome Trust funding enabled the Ratcliffe laboratory to buy a range of equipment, including a large anaerobic chamber (which enable experiments to be carried out in an oxygen-free environment). This allowed them to deduce, almost atom by atom, what happened when HIF and the tumour suppressor protein came together in the presence of oxygen.
"We needed to do some quite demanding experiments, but they showed that these two proteins need oxygen to associate," he says. "We deduced that one was being modified by oxygen, and that was HIF. After a series of biochemical experiments, we eventually deduced, along with Bill Kaelin, that the modification was the hydroxylation of proline. It's just an oxygen atom being added to a normal proline [one of the body's 20 amino acids] that enables part of the tumour suppressor protein, called VHL, to recognise HIF and destroy it. And that's it. It's unexpectedly direct and unexpectedly simple, although there's still the question of how it all works in human physiology, and how it's flexible enough to drive all these complicated things that you need to do to live and have just the right amount of oxygen." The added oxygen atom acts like a red flag, signalling that HIF should be destroyed.
Ratcliffe thinks things would have turned out very differently had it not been for the Wellcome Trust's largesse: "The Trust has funded me for over 20 years and it provides the core of our departmental funding. It's set me up and it's kept me going. I probably wouldn't have been able to do the research at all without the Wellcome Trust's Senior Fellowship." The Oxford-Wellcome Trust connection is echoed in the signs that direct visitors to Ratcliffe: he resides in a large, sunlit office in the Henry Wellcome Building for Molecular Physiology, on the Headington hospitals campus.
The environment has allowed him to draw on the wisdom of neighbouring scholars. He says he is exceptionally fortunate to have collaborated with departmental colleague Professor Christopher Pugh (their researchers are known as the Ratcliffe-Pugh group), Professor Patrick Maxwell (a fellow nephrologist who now heads University College London's Division of Medicine), Professor Christopher Schofield (a world expert on oxygenase enzymes) and Professor Peter Robbins (an Oxford physiologist). Unbelievably, in between his research and strategic oversight of one of Europe's largest biomedical research centres (his department is one of the biggest recipients of Trust funding), Ratcliffe still sees patients for two months a year.
WTVM052246`The Khumbu region of Nepal. Nepalese people appear to have oxygen-sensing systems genetically adapted to high altitude. Credit: Carole Reeves/Wellcome Images
While his work has garnered him worldwide recognition - he is a Fellow of the Royal Society, and in the past two years he has picked up both the Louis Jeantet Prize for Medicine and the much-coveted Gairdner Award (worth CAN$100 000) - attention has turned to whether there might be therapeutic applications. Several drug companies are researching the HIF system as a possible target. The broad idea is that, by tweaking the system, the body can be prompted to raise oxygen levels. This would be useful in such conditions as anaemia, poor circulation, ulcers or other non-healing wounds, and heart disease that is not treatable with angioplasty. Interestingly, a paper in March 2011 by researchers from the University of Colorado and Harvard University suggests that people who live at high altitude have a higher life expectancy and a lower risk of dying from heart disease. The researchers are now exploring whether the low-oxygen environment switches on genes that modify the way heart muscles function, and/or encourage the growth of new blood vessels.
The challenge, Ratcliffe points out, is to develop an HIF-based drug that has a specific, rather than a scattergun, biological action. And he speaks from experience: his laboratory set up a spinout company, ReOx Ltd, which collaborated with Amgen but within a few years realised the size of the challenge and scaled back its ambitions: "HIF has many, many effects on the body. For example, if you go to altitude some of these responses get switched on but not others. If you go to 2000 m, your body might make a few more blood cells but not new blood vessels. If you're developing a medicine, you might want to produce more red blood cells but not new blood vessels, or new blood vessels but not red blood cells, or alter the metabolism without doing either of the first two. Will there be a dose, or duration of treatment, or a mode of application of a specific inhibitor that will work on one part of system and not on others, that would deliver this? We don't even know if it will be difficult or easy.
"This is a major discovery so I don't want to do it down but whether it'll be a major medicine is completely unknown. In my view it's important to be patient. There are large sums at stake - the anaemia market alone is worth £10 billion a year - but there's a concern that wading in could be dangerous." Ratcliffe, now an adviser to the pharmaceutical giant GlaxoSmithKline, contends that the worst possible outcome is that a drug is rushed to market and then withdrawn for being ineffective or even dangerous. Other avenues of research include studying people known to have unusual HIF systems: the Nepalese, for example, appear to have HIF systems genetically adapted to altitude (in the parlance, they show HIF downregulation), while people with Chuvash polycythemia, endemic in the Chuvash region of Russia, have a genetic mutation that pushes their HIF systems into overdrive (upregulation).
For cancer, the therapeutic outlook is similarly uncertain. Although hypoxia appears to be associated with tumour development - in particular, angiogenesis and abnormal metabolism - there are two major problems with producing an oncodrug that affects the HIF system. From a molecular point of view, it is much more difficult to make a drug that dampens HIF down than to make one that turns HIF levels up. Secondly, as scientists don't have a crystal-clear understanding of how tumours develop and flourish, it is possible that interfering with the tumour by tinkering with its oxygen supply might make matters worse. Ratcliffe describes it as a potential "minefield".
"It's very difficult to understand what's helping the cancer and what's hindering it," he points out. "Just because something is a property of cancer doesn't mean that it's helping tumour growth. HIF upregulation and angiogenesis, for example, are all properties of cancer. They might be helping the cancer so turning them down might help, but under other circumstances they might be hindering it. This makes it very complicated to predict the action [of a drug] on the pathway in cancer."
On the optimistic side, however, cancer cells are interlopers: "A cancer cell is living on the edge. If you do anything to it, it's more likely to die than a normal cell. One of our cancer geneticists calls it a 'just right' hypothesis - it is extremely sensitive to dysregulation." That might explain, he says, why such a wide variety of drugs appear to reduce tumour growth, even if their mode of action is not fully elucidated.
Ratcliffe knows first-hand how slow and heartbreaking the search for a cancer cure can be - he lost his mother to cancer - but it isn't the prospect of a miracle drug that gets him up in the morning: "The most important thing for me is that this is fundamental knowledge," he says, with obvious passion. "It's about how our bodies work, and it underpins how most animals work. The satisfying thing for me is that it's true now, it will be true in 100 years and it will still be true in 1000 years. If we are lucky there'll be a drug in it, but we can't say when and we can't say what."
Anjana Ahuja is a former science columnist and feature writer for 'The 'Times'. She now freelances for national newspapers and magazines, and is a co-author of 'Selected: Why we lead, why we follow and why it matters' (Profile Books).
Top image: Peter Ratcliffe. Credit: Robert Taylor Photography.
Further reading
Jaakkola P et al. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 2001;292(5516):468-72.
Epstein AC et al. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 2001;107(1):43-54.
Mandriota SJ et al. HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in thenephron. Cancer Cell 2002;1(5):459-68.
Cockman ME et al. Posttranslational hydroxylation of ankyrin repeats in IkappaB proteins by the hypoxiainducible factor (HIF) asparaginyl hydroxylase, factor inhibiting HIF (FIH). Proc Natl Acad Sci USA 2006;103(40):14767-72.
Coleman ML et al. Asparaginyl hydroxylation of the Notch ankyrin repeat domain by factor inhibiting hypoxiainducible factor. J Biol Chem 2007;282(33):24027-38.