Parkinson's disease, first described in the early 1800s by British physician James Parkinson as "shaking palsy," is among the most prevalent neurological disorders. According to the United Nations, at least four million people worldwide have it; in North America, estimates run from 500,000 to one million, with about 50,000 diagnosed every year. These figures are expected to double by 2040 as the world's elderly population grows; indeed, Parkinson's and other neurodegenerative illnesses common in the elderly (such as Alzheimer's and amyotrophic lateral sclerosis) are on their way to overtaking cancer as a leading cause of death. But the disease is not entirely one of the aged: 50 percent of patients acquire it after age 60; the other half are affected before then. Furthermore, better diagnosis has made experts increasingly aware that the disorder can attack those younger than 40.

So far researchers and clinicians have found no way to slow, stop or prevent Parkinson's. Although treatments do exist--including drugs and deep-brain stimulation--these therapies alleviate symptoms, not causes. In recent years, however, several promising developments have occurred. In particular, investigators who study the role proteins play have linked miscreant proteins to genetic underpinnings of the disease. Such findings are feeding optimism that fresh angles of attack can be identified.

As its 19th-century name suggests--and as many people know from the educational efforts of prominent Parkinson's sufferers such as Janet Reno, Muhammad Ali and Michael J. Fox--the disease is characterized by movement disorders. Tremor in the hands, arms and elsewhere, limb rigidity, slowness of movement, and impaired balance and coordination are among the disease's hallmarks. In addition, some patients have trouble walking, talking, sleeping, urinating and performing sexually.

Perhaps one day CHAPERONE-TYPE DRUGS can be developed to limit degeneration in people.

These impairments result from neurons dying. Although the victim cells are many and found throughout the brain, those producing the neurotransmitter dopamine in a region called the substantia nigra are particularly hard-hit. These dopaminergic nerve cells are key components of the basal ganglia, a complex circuit deep within the brain that fine-tunes and coordinates movement. Initially the brain can function normally as it loses dopaminergic neurons in the substantia nigra, even though it cannot replace the dead cells. But when half or more of these specialized cells disappear, the brain can no longer cover for them. The deficit then produces the same effect that losing air traffic control does at a major airport. Delays, false starts, cancellations and, ultimately, chaos pervade as parts of the brain involved in motor control--the thalamus, basal ganglia and cerebral cortex--no longer function as an integrated and orchestrated unit.

Proteins Behaving Badly
In many Parkinson's cases, the damage can be seen in autopsies as clumps of proteins within the substantia nigra's dopaminergic neurons. Such protein masses also feature in Alzheimer's and Huntington's--but in Parkinson's they are called Lewy bodies, after the German pathologist who first observed them in 1912. Like researchers studying those other neurodegenerative diseases, Parkinson's investigators heatedly debate whether the protein clusters themselves cause destruction or are protective and endeavoring to remove toxic molecules from the neurons. Regardless of their position, however, most agree that understanding these accumulations is key to understanding Parkinson's.

Two cellular processes occupy a central place in this emerging story: protein folding and protein elimination. Cells synthesize proteins, which are chains of amino acids, based on instructions written in the DNA of genes. As the proteins are produced, molecules called chaperones fold them into the three-dimensional form they are supposed to take. These chaperones also refold proteins that have become unfolded.

If the chaperone system fails for some reason, proteins not properly folded in the first place or those that did not correctly refold become targeted for disposal by what is called the ubiquitin-proteasome system. First, ubiquitin, a small protein, is attached to a misshapen protein in a process called ubiquitinylation. Such tagging is repeated until ubiquitin chains of varying lengths end up draped over the ill-fated protein. These chains become the kiss of death. They alert the nerve cell's proteasome, a garbage disposal system, to the existence of the bedecked protein. The proteasome then digests it into its constituent amino acids. Aaron Ciechanover and Avram Hershko of Technion-Israel Institute of Technology and Irwin Rose of the University of California at Irvine were awarded the 2004 Nobel Prize in Chemistry for their work describing this system.

In the past few years, many scientists have come to believe that Parkinson's emerges when the chaperone and ubiquitin-proteasome systems malfunction. They reason that the disease process might go something like this: some form of injury to neurons of the substantia nigra triggers a cascade of cellular stresses [see "Understanding Parkinson's Disease," by Moussa B. H. Youdim and Peter Riederer; Scientific American, January 1997]. These stresses result in a wealth of misfolded proteins that congregate. This buildup might initially be protective because all the renegade proteins are herded together and thus prevented from causing trouble elsewhere in a cell. Chaperones then set to work refolding, and the disposal system starts eliminating those proteins that cannot be reformed. When the production of poorly folded proteins overwhelms the cell's ability to process them, however, trouble arises: The ubiquitin-proteasome system becomes inhibited, chaperones get depleted, and toxic proteins accumulate. Neuronal cell death follows.

Researchers espousing this hypothesis think it could explain Parkinson's two forms. An estimated 95 percent of patients suffer from sporadic disease--the results of a complex interplay between genes and the environment. When someone with a susceptible genetic background encounters certain environmental factors, such as pesticides or other chemicals, the cells in that individual's substantia nigra suffer more stress and accumulate more misfolded proteins than do the same cells in other people. In the remaining 5 percent of patients, Parkinson's appears to be controlled almost entirely by genetics. Discoveries in the past eight years have revealed a connection between mutations and either the buildup of misshapen proteins or the failure of the cell's protective machinery. These genetic insights have been the most exciting developments in the field in years.

The Genetic Frontier
At the National Institutes of Health in 1997, Mihael H. Polymeropoulos and his colleagues identified a mutation in a gene for a protein called alpha-synuclein in Italian and Greek families with an inherited form of Parkinson's. It is an autosomal dominant mutation, meaning just one copy (from the mother or the father) can trigger the disease. Mutations in the alpha-synuclein gene are extremely rare and insignificant in the worldwide burden of Parkinson's (they account for far less than 1 percent of patients), but identification of the link between the encoded protein and Parkinson's set off an explosion of activity--in part because alpha-synuclein, normal or otherwise, was soon found to be one of the proteins that accumulates in the protein clumps. Investigators reasoned that a better understanding of how the mutation leads to Parkinson's could suggest clues to the mechanism underlying Lewy body formation in dopamine-producing cells of the substantia nigra in patients with sporadic disease.

The alpha-synuclein gene codes for a very small protein, only 144 amino acids long, which is thought to play a role in signaling between neurons. Mutations result in tiny changes in the amino acid sequence of the protein--in fact, several such mutations are now known, and two of them result in the change of a single amino acid in the sequence. Studies of fruit flies, nematodes (roundworms) and mice have shown that if mutated alpha-synuclein is produced in high amounts, it causes the degeneration of dopaminergic neurons and motor deficits. Other studies have revealed that mutated alpha-synuclein does not fold correctly and accumulates within Lewy bodies. Altered alpha-synuclein also inhibits the ubiquitin-proteasome system and resists proteasome degradation. In addition, it has recently become clear that having extra copies of the normal alpha-synuclein gene can cause Parkinson's.

In 1998, one year after the discovery of the alpha-synuclein mutation, Yoshikuni Mizuno of Juntendo University and Nobuyoshi Shimizu of Keio University, both in Japan, identified a second gene, parkin, that is mutated in another familial form of Parkinson's. This mutation appears most often in individuals diagnosed before age 40; the younger the age of onset, the more likely the disease is caused by a parkin mutation. Although people who inherit a defective copy from both parents (that is, when the mutation is autosomal recessive) inevitably develop the disease, those who carry a single copy of the mutated gene are also at greater risk. Parkin mutations appear to be more common than alpha-synuclein gene mutations, but no good figure on incidence is currently available.

The parkin protein contains a number of amino acid sequences, or domains, common to many proteins. Of particular interest are two so-called RING domains; proteins with these RING domains are involved in the protein degradation pathway. Findings now suggest that neuronal death in this form of Parkinson's stems in part from the failure of the ubiquitinylation component of the protein disposal system: parkin attaches ubiquitin to misfolded proteins--without it, there is no tagging and no disposal. Our own work has recently shown that a protein called BAG5, which is found in Lewy bodies, can bind to parkin to inhibit its function and cause the death of dopamine-producing neurons.

Interestingly, some patients with parkin mutations lack Lewy bodies in their nigral neurons. This observation suggests that proteins may not form aggregates unless the ubiquitinylation process is functioning. It also suggests that when harmful proteins are not huddled together within Lewy bodies they create cellular havoc. Because patients with parkin mutations develop the disease early in life, it seems likely that they miss some initial protection conferred by having toxic proteins quarantined in clumps.

Several other recent discoveries highlight further genetically induced muck-ups in the cellular machinery. In 2002 Vincenzo Bonifati and his colleagues at Erasmus Medical Center in Rotterdam identified a mutation in a gene called DJ-1. Like that in parkin, this mutation is responsible for an autosomal recessive form of Parkinson's and has been found in Dutch and Italian families. Investigators have seen mutations in another gene, UCHL1, in patients with familial Parkinson's. A paper in Science just described a mutation in PINK1 that may lead to metabolic failure and cell death in the substantia nigra. And other work has identified a gene called LRRK2, which encodes the protein dardarin (meaning "tremor" in the Basque region, where the affected patients came from). It, too, is involved with metabolism and appears in familial Parkinson's. But researchers are not far along in understanding exactly what all these mutations set wrong.

New Avenues for Treatment
Because the insights just described involve molecules whose activity could potentially be altered or mimicked by drugs in ways that would limit cell death, the discoveries could lead to therapies that would do more than ease symptoms--they would actually limit the neuronal degeneration responsible for disease progression.

This strategy has yielded two intriguing results. Increasing the levels of chaperones in cells of the substantia nigra has been found to protect against the neurodegeneration set in motion by mutated alpha-synuclein in animals. Recent studies using fruit-fly models of Parkinson's have shown that drugs that induce chaperone activity can offer protection against neurotoxicity. Perhaps one day chaperone-type drugs can be developed to limit degeneration in people, or gene therapy could be devised to trigger the production of needed chaperones. In addition, investigators have found that increasing the amount of normal parkin protein in cells protects against the neurodegeneration resulting from noxious, misfolded proteins. Much more study will be needed, however, to determine whether such interventions could be made to work in humans.

In addition to pursuing the preliminary leads that have arisen out of the new protein-related and genetic findings, investigators have begun introducing neurotrophic factors--compounds promoting neuronal growth and differentiation--into the brain. These agents not only alleviate symptoms but also promise to protect neurons from damage or even to restore those already harmed.

One line of research in animals, for instance, suggests that a family of proteins called glial cell line-derived neurotrophic factor (GDNF) can enhance the survival of injured dopamine neurons and dramatically reduce parkinsonian symptoms. Steve Gill and his colleagues at Frenchay Hospital in Bristol, England, have embarked on a pilot study to give Parkinson's patients GDNF. Surgeons insert a catheter into the left and right striatum, the main recipients in the basal ganglia of the dopamine secreted by neurons of the substantia nigra. Minute volumes of GDNF are then continuously infused to the brain from a pump set into the abdomen. The pump holds enough GDNF to last one month and is replenished during an office visit; a syringe pierces the skin and refills the pump reservoir.

Initial results in a handful of patients suggested that symptoms had improved, and PET scans indicated some restoration in dopamine uptake in the striatum and substantia nigra. But the results of a larger, more recent trial have been unconvincing: patients who received saline solution fared no better than those who received GDNF. Nevertheless, many of us who work in this area feel that this approach is still worth pursuing. It is not unusual in medicine for the first forays into a treatment to be negative. Levo-do-pa, for instance, initially showed no benefit and only unwanted side effects; now it is one of the principal treatments for Parkinson's.

Other researchers are using gene therapy instead of surgery to administer GDNF, hoping the delivered gene will provide a long-term supply of this neurotrophic agent. Jeffrey H. Kordower of Rush Presbyterian-St. Luke's Medical Center in Chicago and Patrick Aebischer of the Neurosciences Institute at the Swiss Federal Institute of Technology and their colleagues engineered a lentivirus to carry the gene for GDNF and deliver it to dopamine-producing striatal cells in four parkinsonian monkeys. The results were impressive: the monkeys' motor problems significantly diminished, and they were unaffected by a subsequent injection of MPTP, a chemical toxic to dopamine neurons of the substantia nigra. The introduced gene induced cells to make the protein for up to six months, after which the experiments were stopped. Based on these studies, scientists at Ceregene in San Diego are using a similar technique to deliver the protein neurturin, a member of the GDNF family. Although the studies are in the preclinical phase, researchers plan to test a gene similar to the gene for neurturin in human patients.

Still other forms of therapy are being investigated. Working with Avigen near San Francisco, Krys Bankiewicz has shown in animals that placing the gene for an enzyme called aromatic amino acid decarboxylase in the striatum can enhance dopamine production in this area of the brain. In rats and monkeys this approach has also ameliorated parkinsonian symptoms. Trials in patients have been approved and will be launched soon.

Michael Kaplitt of Cornell University and his team are taking a different tack, using gene therapy to shut down some of the brain regions that become overactive when dopamine released from the substantia nigra falls too low--including the subthalamic nucleus of the basal ganglia. (The loss of dopamine causes neurons making glutamate, an excitatory neurotransmitter, to act unopposed and thus overstimulate their targets, causing movement disorders.) Kaplitt will begin human trials using a virus to introduce the gene for glutamic acid decarboxylase--which is crucial to the production of the inhibitory neurotransmitter gamma amino butyric acid (GABA)--to these sites. He and his co-workers hope that the GABA will quell the overexcited cells and thus calm parkinsonian movement disorders. In the experiments, they thread a tube about the width of a hair through a hole the size of a quarter on top of a patient's skull. The tube delivers a dose of virus, which ferries copies of the gene into neurons of the subthalamic nucleus. The chemical released from the altered cells should not only quiet the overactive neurons residing in that region but may be dispatched to other overactive brain areas.

Perhaps the most hotly debated potential treatment entails transplanting cells to replace those that have died. The idea has been to implant embryonic stem cells or adult stem cells and to coax these undifferentiated cells into becoming dopamine-producing neurons. Because embryonic stem cells are derived from days-old embryos created during in vitro fertilization, their use is highly controversial. Fewer ethical questions surround the use of adult stem cells, which are harvested from adult tissue, but some scientists believe these cells are more difficult to work with.

Despite important progress in identifying the molecular cues and recipes for pushing undifferentiated cells to produce dopamine, no one yet knows whether transplantation of any kind will be as fruitful a strategy as has been hoped. The clinical trials using the most meaningful protocols have so far been conducted with fetal material. These have shown hundreds of thousands of surviving transplanted dopamine-producing cells in patients, yet the functional benefits have been at best modest and inconsistent, and the treatment has been associated with serious adverse effects, including dyskinesias (uncontrollable writhing and twisting movements). Scientists are trying to determine why transplantation has not been more helpful and why side effects have arisen, but for now they are not conducting human trials of the procedure in the U.S.

Finally, researchers continue to investigate and refine the approach behind deep-brain stimulation: applying electric pulses. Several months ago St¿phane Palfi and his colleagues at the CEA Fr¿d¿ric Joliot Hospital Service in Orsay, France, reported that gently stimulating the brain surface could improve symptoms in baboons with a version of Parkinson's. Clinical trials are under way in France and elsewhere to determine whether this surgical intervention is similarly effective in humans.

Although much remains unknown about Parkinson's, the genetic and cellular insights that have come to light in just the past few years are highly encouraging. They give new hope for treatments that will combine with existing ones to slow disease progression and improve control of this distressing disorder.