Multiple sclerosis (MS) hides its secrets well. It tends to strike the middle-aged and people who are otherwise healthy, it can last for decades and it affects areas in the brain that can be difficult to observe, providing little opportunity for researchers to watch it in action. Until a disease trigger is identified, scientists and physicians are unsure of the best treatment for people with MS, let alone how to cure them. So when Moses Rodriguez had an unprecedented opportunity to peer into the brains of newly diagnosed patients, he took it.

Rodriguez, an immunologist and neurologist at the Mayo Clinic in Rochester, Minnesota, had access to the brains of patients with abnormal magnetic resonance imaging (MRI) scans. Their physicians had ordered biopsies to determine whether the unusual scans indicated brain tumours. But rather than cancer, results from a small subset of patients showed the abnormality to be the first sign of MS.

The biopsies, like tiny ice cores, allowed Rodriguez to look at different layers of a brain in the first stages of MS. “I knew that by studying the very earliest events in the MS lesion we could start to understand what's going on,” he says. The biopsies contained a noticeable lack of immune cells, leading Rodriguez to believe that he had finally pinpointed how the disease progresses. According to his theory, the process starts when something injures a group of brain cells called oligodendrocytes, which normally produce a fatty, protective layer of myelin that coats the neural fibres of the brain and spinal cord. As the oligodendrocytes die, the myelin begins to degrade and the nerves' slender axon fibres are exposed, leaving them vulnerable to damage and less adept at transmitting electrical impulses. The damage triggers inflammation, and it is only then that the body's ever-vigilant immune system sends in a surge of immune cells to repair the damage.

According to one theory, MS begins with the injury of a group of brain cells called oligodendrocytes, which normally deposit a fatty, protective layer of myelin around the neural fibres of the brain and spinal cord, and which are shown here in a section from a cerebral biopsy in a patient 2 weeks after the onset of symptoms. Destruction of the oligodendrocytes leads to degradation of the myelin sheathing the nerve fibres. Credit: ©Dr. Moses Rodriguez

His theory, however, goes against the accepted dogma of the field, which identifies MS as an autoimmune disorder in which the adaptive immune system — the white blood cells that tag, kill and remember the shape of foreign invaders — mistakenly launches an attack on the body's own myelin. Genetic and therapeutic data clearly implicate the immune system. As many as 100 genes have been found to correlate with an increased risk of disease, and all of those with well-documented functions have been linked to the immune system (see A complex code). The only therapies that have been shown to ease MS flare-ups work by suppressing or otherwise modulating the immune system.

Researchers struggled for a long time to find a consistent way to diagnose the disease, and one of the first successful methods identified MS patients by the abnormally high levels of immune molecules in their cerebrospinal fluid. Such a distinct response led the scientists to postulate that a virus, bacterium or environmental toxin might be responsible for the initial insult to the brain, triggering the immune response. Yet despite decades of searching, researchers have yet to find anything that could reliably spark such a reaction.

Inside-out

Rodriguez believes that MS begins, instead, as a disorder of the nervous system — an idea referred to as the 'inside-out' hypothesis. According to his theory, the disease starts in the nerve cells of the brain, and somehow triggers an immune response that spreads out into the peripheral blood and cerebrospinal fluid. “The process seems to begin in the nervous system way before an inflammatory response takes place,” he says.

The inside-out theory begins to lose traction, however, when it describes a degenerative process that should decline at a constant pace. MS acts like a purely progressive disease in only about 15% of cases; everyone else experiences relapsing–remitting MS, which starts with episodic flare-ups occurring at unpredictable frequencies over many years, and only later changes into a more progressive course.

Critics of the inside-out theory instead believe that the disease starts in the peripheral bloodstream and moves into the brain when something triggers a breakdown in the blood–brain barrier. According to this 'outside-in' hypothesis, something — perhaps a virus or a chemical in the environment — sets off the immune system and compromises the blood–brain barrier, allowing an influx of innate immune cells that are primed to attack either myelin or the cells that produce it.

Over the past two decades, drugs that suppress the immune system have quashed flare-ups in about 70% of people with relapsing–remitting MS. This success has further convinced supporters of the outside-in theory that the disease is an inflammatory reaction gone awry. “When you take their immune system away, they no longer have those attacks,” says Jack Antel, a neuroimmunologist at the Montreal Neurological Institute and Hospital of McGill University in Canada. “People are doing better, and it's hard to argue against that.”

Each camp is quick to point out holes in the other's hypothesis, throwing criticisms back and forth like schoolyard rivals. Proponents of the outside-in theory ask how can MS be a neurodegenerative disease when immunosuppressants reduce or completely prevent flare-ups? After all, if the immune reaction were geared towards repairing damaged myelin, why wouldn't the disease get worse when someone took drugs that damped their immune system? What is more, Richard Ransohoff, a neuroinflammation researcher at the Cleveland Clinic in Ohio, has recently shown that innate immune cells are present in the biopsies of early MS lesions, directly contradicting Rodriguez's conclusions.

But the outside-in camp remains firm. How can MS be an autoimmune disease if immunosuppressants work for only a subset of patients and have no impact on progressive MS? It is a fair point — people who experience relief from relapses while on immune-modulating drugs get no benefit when the disease moves into the progressive phase. There is some evidence that the disease is progressing even as the immune-modulating drugs are tackling the flare-up. “We're struggling with this question — which one is the real, true mechanism — because both immunological and neurodegenerative phenomena can be easy to detect in any MS patient, and we don't know which one starts the process,” says Jorge Oksenberg, an immunogeneticist at the University of California, San Francisco.

“Five or six years ago, I probably would have felt more certain that I had some idea of the disease process.” says Henry McFarland, scientist emeritus at the neuroimmunology branch of the US National Institute of Neurological Disorders and Stroke in Bethesda, Maryland. “But over the years there's been an accumulation of data that suggests it may be more complicated.”

It is a classic scientific problem. The harder researchers look at what triggers MS, the less they understand about its origin and progression. “We'd like to know what triggers the initial attack, what triggers a relapse, what causes people to have a mild case, what causes a severe case, and is there something different going on in relapsing–remitting versus progressive MS,” says Lawrence Steinman, head of immunology at Stanford University's School of Medicine. “None of those questions are answered even close to adequately.”

Solving this mystery will require drug studies that follow patients for a decade or more. It will also require more elaborate imaging and post-mortem data from patients in the middle of their disease course, rather than just the beginning or the end. And it will require an as yet elusive animal model that can reliably mirror the human disease (see Not close enough).

Solving the puzzle

Despite their differences, MS researchers share some important common ground. They generally agree that, whether or not the disease starts with a malfunction of the immune system, it clearly has an immunological component. Levels of white blood cells and antibodies, which are produced in response to substances the body does not recognize as its own, are higher than normal in the cerebrospinal fluid of MS patients — a clear sign that the immune system is hyperactive.

They also agree that certain genes increase susceptibility to MS, and that environmental factors play a role (people who live closer to the equator, get more sunlight and have higher blood levels of vitamin D have lower rates of MS than those who live in cooler regions). And they agree that infection with the Epstein–Barr virus in older patients is strongly linked, at least epidemiologically, to the onset of MS.

No single risk factor can predict the onset of MS, says Oksenberg. “You can have the genetic profile but, if you're lucky and don't encounter the trigger, you will be safe. On the other hand, you can encounter the trigger and, if your genetics don't predispose you to the disease, you may also be safe.” Rather, he says, “it's a perfect storm”, in which multiple factors combine to initiate the onset of disease.

Many researchers, including Oksenberg, believe that MS might really be a collection of different diseases that share the same clinical presentation. This might explain why it has been so difficult to piece together the disparate factors to create a single explanation of what triggers and perpetuates the disease. “We are starting to recognize that our simplifications have been going too far, and we need to start to grasp and deal with the complexity of disease,” Oksenberg says.

For now, however, the full picture is still unclear, leaving questions to be answered and theories to be verified. “MS is one of those year-long jigsaw puzzles that's spread out on the table, and everyone in the family is involved, and different people are good at different sections of the puzzle,” says Ransohoff. “Very slowly some definable elements are coming into view. We no longer think the puzzle shows a picture of Van Gogh's Starry Night; now we realize it's the shoreline in Hong Kong.” Before they are finished, the image might have morphed yet again. But while different family members might squabble over where the pieces fit, they are working together — bit by bit — to sort it out.