A DIET of ground cow’s brain as a cure for multiple sclerosis; fibres from the sternums of baby chicks to relieve the crippling effects of rheumatoid arthritis; scrapings from the retinas of farmyard animals to ease the swelling eyeballs associated with uveitis. These remedies might smack of medieval quackery but they are all quite contemporary. Each one is being tested by researchers looklng for a dietary approach to treating “autoimmune” diseases, conditions such as rheumatoid arthritis in which a faulty immune system turns on the body it’s meant to be protecting.
The approach is to feed patients certain kinds of proteins extracted from animals as a way of correcting their faulty immune systems and slowing the progress of the disease. And superficially, the results look very promising. A team of researchers in the US claims to have put rheumatoid arthritis into remission simply by feeding people the protein collagen, and multiple sclerosis into remission by feeding patients with a protein found on the protective sheath that covers nerve fibres.
But critics say the findings are too good to be true. Autoimmune diseases are among the most difficult illnesses to treat, and cures have proved elusive. The reasons why immune systems mistakenly turn on the body’s own tissues are complex and ill understood. Can “therapies” based on simple protein meals really reverse this fault? Sceptics point to the fact that the research is funded by a drugs company that stands to profit from the therapies – AutoImmune from Lexington, Massachusetts – and that the team’s leader, Howard Weiner, is a shareholder in that company. Indeed, because of his financial connections Weiner decided to withdraw from any direct involvement in the company’s latest clinical trials of its new drug for treating MS, the results of which will be submitted to the US Food and Drug Administration in an effort to gain approval to market the drug. Last month, AutoImmune announced that it had signed an agreement with the pharmaceuticals giant Eli Lily to develop drugs for the treatment of a form of diabetes caused by an immune disorder.
Advertisement
But the prize of financial gain is part and parcel of most medical research. And Weiner’s academic credentials are impeccable. He holds the Robert L. Kroc chair in neurological diseases at Harvard Medical School and is co-director of the centre for neurologic diseases at the Brigham and Women’s Hospital in Boston. The real problem (if there is one) with therapies based on dietary proteins may have more to do with the complexity of the human immune system than with commercial ambitions. Exactly how its cells and antibodies distinguish between the body’s own cells and those from outside invaders – protecting one set, destroying the other – remains something of a mystery. What we do know is that the consequences can be devastating when the immune system gets it wrong and attacks the body’s own proteins. The results are felt by the millions of people worldwide who suffer from rheumatoid arthritis, in which the immune system attacks collagen in joints, causing pain and swelling, or multiple sclerosis, in which the myelin that normally surrounds nerves fibres is destroyed.
Holding fire
Weiner’s controversial therapy for autoimmune diseases is designed to exploit something most of us take for granted – the fact that our immune system does not normally attack the proteins we eat as food even though they are entirely foreign to our bodies. Immunologists call it “oral tolerance”. In autoimmune diseases, some of the body’s own proteins are attacked by the immune system. Weiner’s hope, like that of other researchers before him, is that if someone with an autoimmune disease is fed proteins that are similar to the ones being attacked, their faulty immune system may be fooled into holding fire. Instead of attacking the proteins, it will start to treat them with the hands-off respect accorded food molecules.
The idea has a long pedigree. Anecdotes dating back to 1829 tell how South American Indians ate the leaves of poison ivy to lessen the itchy rash that normally follows when you touch the plant. The term “oral tolerance” was coined in 1911, after the phenomenon was demonstrated in laboratory experiments on guinea pigs. Within seconds of being injected with the protein albumin from hen’s eggs, guinea pigs normally develop severe, usually fatal, respiratory symptoms. What the experiments showed was that guinea pigs that had first been fed the protein did not react when the protein was injected.
In the early 1980s, Caroline Whitacre, professor of medical microbiology and immunology at Ohio State University at Columbus, was the first to show that oral tolerance could be a way to deal with autoimmune diseases. But it was not until 1986 that two teams independently published research showing that arthritis in rats could be suppressed by feeding them type II collagen, the protein found at the surface of joint tissue. The first team was led by Norman Staines, professor of immunology and head of the infection and immunity research group at King’s College London, and the second by Jeanette Thorbecke, professor of pathology at New York University School of Medicine.
Hopes that oral tolerance therapy might cure human autoimmune diseases were raised further in 1991 when Weiner reported some intriguing findings in the Journal of Experimental Medicine. Immune responses are normally very specific: antibodies and immune cells that respond to one protein don’t necessarily respond to others. So if you feed animals one particular type of protein, you might expect their immune systems to become desensitised to this protein and no other. But what Weiner claimed to have found was evidence of a much less specific immune response. Feed collagen from chickens to a rat, he reported, and the immune system not only refrains from attacking this protein; it also holds fire on similar proteins found in the rat’s tissues. “We discovered a mechanism no one else had described before,” he says.
This opened the way for Weiner to set up the trials with human patients that AutoImmune funded. For the first trial, he fed a group of MS sufferers with a protein from cattle found in the myelin coating around nerve fibres. The fibres came from cows’ brain and were presented in the form of a capsule. A similar protein in humans, called “myelin basic protein”, is thought to be involved in MS. In the second trial, he fed rheumatoid arthritis patients with a collagen protein from chicks in a bid to restrain the patients’ immune systems from attacking the human collagen in their joints. In this case, the collagen from a chick’s sternum was liquidised and mixed with fruit juice. Weiner claims that both trials were a success, but critics – and even some enthusiasts – say the effectiveness of therapies based on oral tolerance is still unproven. Some question the way AutoImmune interpreted its data and say its trials, which involved no more than 30 people, were too small to produce any firm conclusions.
But nobody doubts that oral tolerance is a real biological phenomenon; the question is whether it works as a treatment for autoimmune diseases in humans. At best, differences in opinion about just how the immune system responds to the proteins we eat make it difficult to predict the best dose of protein for therapy. At worst, there is still doubt about which, if any, protein is under attack in diseases such as MS and rheumatoid arthritis.
Trial fever
Even so, Weiner’s research for AutoImmune seems to have triggered a dash to test the therapy. Already trials have started in the US on people with uveitis, an inflammatory disease of the eye that can lead to blindness. Here, tissue from the retina of a cow’s eye is presented to patients in the form of a capsule. At the German Rheumatology Research Centre in Berlin, Aviron Mitchison is coordinating trials at the city’s rheumatology clinics. His 90 patients include a group being fed 10 milligrams of collagen a day – more than ten times as much as Weiner used. Among other trials being planned is one due to begin this month at Guy’s Hospital in London, also involving rheumatoid arthritis patients. Meanwhile, AutoImmune is well into larger-scale trials of Weiner’s therapies on hundreds of patients with rheumatoid arthritis and MS throughout the US and Canada.
In short, what was once an obscure immunological theory is now sweating it out under medicine’s spotlight. Inevitably, the hopes of many sufferers are being raised as a consequence. Whether they will be dashed, only time will tell.
The key to the success of the present round of clinical trials of oral tolerance lies in the relatively nonspecific effects that Weiner claims to have found, and which he has dubbed “bystander suppression”. Most autoimmune diseases are triggered by a complex combination of environmental and genetic factors, and in many, including MS and rheumatoid arthritis, nobody knows exactly which proteins are under attack. But if bystander suppression does take place, then oral tolerance therapy could work without it being necessary to know precisely what causes the autoimmune diseases.
The problem is that Weiner’s findings remain controversial. The debate turns on two alternative theories about how oral tolerance might work. Central to both is the mucosal immune system, a specialised line of defence found in the gut, lungs and other boundaries between the body and the outside world. Such tissues produce a class of antibodies known as immunoglobulin A (IgA), which are the main antibodies in secretions such as saliva, milk and tears. The tissues also share an ability to screen the proteins that come their way, letting “friendly” proteins pass into the body while rejecting others that might be harmful.
Crucial to the distinction is the form an invader takes as it enters the body. Bacteria, viruses and other infectious agents usually come as particles, and these trigger attacks from cells in the mucosal immune system. Proteins in foods, on the other hand, usually dissolve into the bloodstream, and that seems to pacify any immune responses. Nobody knows how this happens, but there are two theories in the running. One is that cells that normally mobilise the immune response may become disabled or destroyed. The alternative is that other immune cells may act to suppress the immune response.
Here’s how the first of these responses could happen. Any proteins you eat are absorbed through the lining of the intestine, either whole or as fragments. From here they pass into the bloodstream, where they could switch off, or even destroy, certain helper T cells. These cells normally produce chemicals called lymphokines that help to kick the immune system into action. If that kick never comes, absorbed proteins may be able to slip past the immune system unnoticed. So too might cells in the body that happen to produce proteins identical to the ones that were eaten and that might otherwise have been attacked by the faulty immune system, resulting in some autoimmune disease or other.
Hordes of cells
The alternative theory envisages the proteins you’ve eaten tripping a switch, unleashing hordes of suppressor T cells. These cells are the immunological equivalent of peacekeeping forces: they flow through the body looking for tissues and cells being attacked by the immune system. Suppressor cells can only intervene in certain cases, and this is where bystander suppression comes in. According to Weiner, if the protein under attack is similar, but not necessarily identical, to the one that flipped the switch, the suppressor cells can quell the violence. They do this by pumping out anti-inflammatory lymphokines that “sedate” aggressive immune cells.
This is not all pie-in-the-sky theorising. Many researchers have found evidence in animal experiments of the first kind of tolerance, involving just helper T cells. And Weiner and another immunologist, Dale Gregorson from the University of Minnesota, claim to have detected enhanced levels of suppressor T cells in rats, indicating that bystander suppression may be occurring. Weiner has found that when mice are fed myelin basic protein, their suppressor T cells produce the anti-inflammatory lymphokines TGF-β, IL-4 and IL-10. And when these mice are given injections of cells affected by MS, they develop only a mild form of the disease.
Weiner says that his experiments indicate that low oral doses of protein are likely to result in the production of suppressor T cells, leading to the bystander effect, whereas high doses act more on helper T cells, so suppressing the immune response directly. If this is correct, it could have implications for treating autoimmune diseases. Where it is unclear which of the body’s proteins are under attack, doctors could try using low doses of proteins to trigger bystander suppression. Two trials on humans, funded by AutoImmune and reported in Science in February and September 1993, set out to test this idea.
Staines, whose work at King’s College London has been sponsored by the Arthritis and Rheumatism Council, a medical charity, agrees that the therapy is likely to work in different ways, depending on dose and also on the disease being treated. “I suspect Weiner is right that different doses lead to different mechanisms,” he says. But he adds that this has yet to be proved.
Weiner’s first trial for AutoImmune involved 30 patients with MS. Every day for a year, half the group were fed a capsule containing 300 milligrams of the bovine equivalent of myelin basic protein while the other half took a placebo. In MS, the tissue under attack is the fatty myelin sheath that surrounds nerve fibres, and normally insulates them so that their electrical signals do not leak out. Blood tests on patients receiving the bovine protein revealed fewer immune cells reacting against the nerve sheath. The researchers also found that 12 of the patients in the control group suffered attacks, compared with only 6 of those taking oral doses of protein.
Weiner’s trial for AutoImmune on patients with rheumatoid arthritis involved feeding type Il collagen from chickens in a solution of fruit juice to 10 of the patients, and giving another 10 a placebo. Type II collagen, the most abundant protein in cartilage, is similar to the human protein thought to be under attack in the disease.
Weiner found that four patients receiving collagen showed complete remission, with no swelling of their joints during the trial period. Three of these relapsed when the drug was withdrawn but improved again following further treatment. In the other six patients receiving collagen, the swelling was reduced by about 30 per cent compared with the matched control group.
These results sound promising but critics question their significance. For a start, they say, the patients continued to receive other therapies right up to the beginning of the trials. Without the usual month-long “washout period”, any apparent benefits could have come from medium-term effects of other drugs. David Wraith, a researcher at Addenbrooke’s Hospital in Cambridge and an enthusiast regarding the therapy, admits that “the MS trial really didn’t show any significant effect”. This view is echoed by Alastair Compston, head of the Cambridge Centre for Brain Repair, where the British Multiple Sclerosis Society is funding research into the disease. Compston believes the AutoImmune trial was too small for its results to be meaningful. “MS is a very variable disease in its behaviour,” he says. “You need a lot of people to iron out all the noise.” Allan Mowat, who studies oral tolerance in the department of bacteriology and immunology at the Western Infirmary, Glasgow, sees no reason why the therapy should not work in humans, but his response to Weiner’s MS trial for AutoImmune is also cautious. “If we’re dealing with a disease proven to be antigen specific I’d be more convinced,” he says. For the moment, people can’t even agree which protein is under attack in diseases such as MS and rheumatoid arthritis.
In the soup
On the face of it, AutoImmune’s rheumatoid arthritis studies proved more successful than the MS trial. Wraith describes the 50 per cent decrease in the number of patients suffering attacks as “encouraging”. But not everyone is as enthusiastic. “You’d need to know what your patients were eating,” says Marc Feldmann, head of the cytokine and immunology division at the Kennedy Institute of Rheumatology in London. He points out that Weiner’s doses of collagen are no more than could be found in a bowl of chicken soup.
But Weiner’s theory that you need low doses of protein to trigger bystander suppression puts the researchers in a bind. When low doses of protein are prescribed it is difficult to distinguish the effects they may be having from the effect of proteins in the diet. And there is a second problem which is potentially much more serious. Researchers have found that the immune system of several species of laboratory animals can actually become more sensitive to a protein after the animals have repeatedly been fed it in small doses. The reason for this effect remains unknown, and nothing similar has been recorded in humans. But if it ever were, the safety of therapies based on oral tolerance might be put in doubt.
Despite the flak, Weiner is not deterred. “It’s clear that oral tolerance exists,” he says. “The question is, how well will it do in humans? That’s why we need to do large trials.” To this end, after trials on animals and small groups of people – the first ones to text toxicity, the second ones to gauge efficacy – AutoImmune has begun phase-three trials of its MS drug. Such trials normally mark the last step before seeking approval from the US Food and Drug Administration to market the drug. Last March, more than 500 MS patients at nine sites in the US and four in Canada began a two-year course of doses of myelin basic protein prepared by AutoImmune as part of the company’s continuing phase-two trials.
Meanwhile, some 300 Americans with severe rheumatoid arthritis have also begun oral tolerance therapy at six sites in the US. Results from this six-month trial to test collagen at doses ranging from 0.02 milligrams to 2.5 milligrams are due to be published soon. This time, both trials have been preceded by a month-long washout period.
The stomach may not be the only way that proteins can enter the body and desensitise the immune system. Evidence from laboratories such as Wraith’s suggests that inhaling proteins or protein fragments desensitises the immune system even more effectively than eating them. Whatever the route followed in administering the proteins, if tolerance therapy does live up to expectations, it could offer a real alternative to immunosuppressants. In the words of one eminent researcher, it would be like using “the body’s own power to heal itself”.