Medicine: New Therapy Strengthens Fight Against Osteoporosis Neurology: Studies Find Caffeine May Counter Parkinson's Pathology: Programmed Death by Mitochondria Health Care Policy 1: Medicare May Restrict Good Care for Dying Patients Health Care Policy 2: In First Seidman Lecture, Keynote Predicts No Drug Benefit as Part of Medicare Workplace Exposure May Be Similar for Smokers, Nonsmokers Breast-feeding May Limit Teenage Obesity Some Medical Errors More Common With Children Than Adults Rod Cells Pay for Light Sensitivity With Reaction Time Faculty Council Proceedings Two from HMS Added to the National Academy of Sciences Roster A Preview of Alumni Week Center for Neurodegeneration and Repair Invites Applications for Grants, Fellowships NAS and AAAS Address Nationwide Postdoc Issues Call for Writers Front Page

NEUROLOGY Studies Find Caffeine May Counter Parkinson's Two Harvard studies—one at HMS and the other at HSPH—demonstrate that caffeine may protect against Parkinson's disease. One provides epidemiological evidence that caffeine consumers have a lower risk for developing the disease, and the other illuminates the molecular basis of this effect, showing in a mouse model that caffeine may shield neurons from damage characteristic of Parkinson's. "The receptor system through which caffeine may be acting is not only a clue to the pathophysiology of Parkinson's but also a potential lead on novel therapies," says Michael Schwarzschild (r), shown with colleagues Kui Xu, Jiang-Fan Chen, and Yue-Hang Xu (l to r). Photo by Steve Gilbert The cause of the disease is still unclear, and there has been growing interest in the idea that environmental and behavioral factors play a role. While these findings may not justify therapeutic coffee drinking, they shed light on the underlying cause of Parkinson's and point toward potential new drugs by identifying the molecular target of caffeine's protection. A Medicinal Brew Experiments published in the May Journal of Neuroscience and led by Michael Schwarzschild and Jiang-Fan Chen, both HMS assistant professors of neurology at Massachusetts General Hospital, revealed that caffeine could protect against dopamine loss in a mouse model of Parkinson's. They pretreated mice with caffeine at levels equivalent to a cup or two of coffee in humans and then exposed them to the chemical MPTP, which causes neurological damage and symptoms similar to Parkinson's. In addition to finding that caffeine had a neuroprotective effect, Schwarzschild's group has linked the activity to a specific target of caffeine, the adenosine receptor subtype A2A. These receptors have been actively studied as potential therapeutic targets in Parkinson's because they can modulate dopamine expression and are expressed only in the striatum, part of the neural system deprived of dopamine in Parkinson's patients. Caffeine is known to bind to and inactivate A2A receptors in addition to other molecular targets. In order to determine whether this interaction was causing the protective effect, the team looked at other A2A antagonists and found that they mimicked the protective effects of caffeine in the same model, while antagonists of other adenosine receptor subtypes did not. A line of A2A knockouts initially developed by Chen also exhibited a similar neuroprotection against MPTP. Caffeine protects from dopamine loss by blocking the adenosine A2A receptor. Above, A2A expression is largely restricted to subcortical neurons of the striatum. Below, treatment with MPTP induces loss of dopaminergic nerve terminals in this same region (bottom left) as opposed to control (top left). Pre-treatment with caffeine attenuates this loss (bottom right). But the finding that caffeine may be protective does not mean its consumption is a viable therapy or preventive measure. "Caffeine is pharmacologically a very dirty drug," said Chen, since it has several targets in addition to A2A and a variety of effects. But the team is interested in the idea that more specific A2A antagonists could help prevent the progression of the disease. Current therapies for Parkinson's help replace dopamine in the brain but do not prevent the underlying cell death of dopaminergic neurons. Adenosine A2A antagonists are currently in clinical trials for their potential to relieve motor symptoms of Parkinson's by promoting dopamine expression, but the discovery of their neuroprotective effect raises another potential role for them in treatment. "In other words, this drug may have a dual function in both slowing down the progression of Parkinson's disease at the same time as it provides symptom relief," said Chen. A2A antagonists are especially attractive as a therapy because the receptors are limited to the very part of the brain affected in Parkinson's disease, decreasing the potential for side effects in other areas. The Epidemiology Angle After completing these experiments, Schwarzschild and Chen were surprised when they saw a flier for a talk on caffeine and Parkinson's disease given by Alberto Ascherio, associate professor of nutrition and epidemiology at HSPH. They attended, and since then the two groups have been working together closely. Ascherio's team has found epidemiological evidence of an inverse association between caffeine consumption and Parkinson's disease. The prospective study, available on the Annals of Neurology website, follows a large cohort of women and men from the Nurses' Health Study and the Heath Professionals Follow-up Study. The team found a strong inverse relationship between caffeine and Parkinson's in men, with those consuming the equivalent of four to five cups of coffee a day having about half the risk for the disease as those who consume little or no caffeine. For women, the association was less clear and requires further study: those who had one to three cups a day had the lowest risk, but the benefit tapered off at higher levels. Consumption of decaffeinated coffee did not lower the risk. The study adds credence to previous reports that caffeine consumers have a lower incidence of the disease. The two papers offer a rare convergence of research attacking the same problem from two different but complementary angles. "The difference between the animal experiments and the epidemiology is that they can prove causality, but they cannot generalize to humans," said Ascherio. "We work directly on humans but we cannot prove causality. That's why the two papers are nicely complementary to each other because the two together strengthen the case for a causality in humans." Both groups emphasize that the causality is far from proven. There has also been an observed, though perhaps less publicized, inverse association between smoking and Parkinson's, leading some researchers to speculate that there is a preexisting condition in the brains of Parkinson's patients that makes them resistant to addiction, since both are linked to the dopamine system. By providing a molecular rationale that caffeine can actually be protective, Schwarzschild and Chen's research bolsters the idea that the epidemiological effects of caffeine are causative. —Courtney Humphries