Are Errant Astrocytes the Cause and Therapeutic Target of Epilepsy?

Targeting astrocytes instead of symptoms may become a new strategy for managing--and perhaps curing--epilepsy. Indeed, researchers from the University of Rochester in New York demonstrated that common antiepileptic agents--gabapentin (Neurontin, Pfizer), phenytoin, and valproate--"potently reduce" astrocytic calcium signaling. Their research illustrated how such signaling, which mediates glutamate release, is associated with seizure activity. The findings reinforced the researchers' theory that malfunctioning astrocytes play a primary role in the pathology of epilepsy.

Through experiments performed on both histologic slices of rat hippocampi and craniotomized adult mice, Maiken Nedergaard, MD, DMSc, and a team from the University of Rochester School of Medicine and Dentistry demonstrated that malformed, reactive astrocytes are responsible for inappropriate glutamate release, which in turn triggers paroxysmal depolarization shifts (PDSs)--epileptiform discharges seen on electroencephalograms. The researchers also found that PDSs were preceded by or occurred concomitantly with astrocytic calcium signaling.

Because the role of astrocytes is to release and regulate neurotransmitters as well as to create an environment in which neurons thrive and function, "it's easy to see how serious disease might result" if astrocytes behave abnormally, said Nedergaard, a professor in the Department of Neurological Surgery at the University of Rochester.

To assess whether glutamate plays a role in epilepsy, Nedergaard and colleagues cellularly replicated several models of epileptiform seizures in hippocampal slices by exposing them to the potassium channel blocker 4-aminopyridine (4-AP), exposing them to bicuculline or penicillin (which are g-aminobutyrate_A receptor antagonists), or causing extracellular magnesium inhibition (which activates N-methyl d-aspartate receptors) or extracellular calcium inhibition. They then blocked neuronal firing by treating the histologic material with tetrodotoxin and a calcium channel blocker.

To their surprise, according to Nedergaard, an increase in reactive astrocytes was found in all 5 experimental models. Furthermore, the reactive astrocytes were releasing excessive amounts of glutamate. Nedergaard concluded that "reactive astrocytes trigger epilepsy, and not the other way around."

The researchers then looked at the action of antiepileptic agents on astrocytic calcium signaling in craniotomized mice, the cortexes of which were treated with 4-AP to induce PDSs. They found that calcium signaling preceded seizure activity. The team also found that the antiepileptic agents valproate, gabapentin, and phenytoin reduced astrocytic calcium signaling transmission by 69.7%, 55.6%, and 45.5%, respectively. When a similar experiment was conducted in nonepileptic animals that were pretreated with these antiepileptic agents before astrocytic calcium waves were induced by using adenosine triphosphate, a similar effect was seen.

The research team surmised that the development of therapies aimed at maintaining and correcting the integrity of astrocytes might be a revolutionary way to treat epilepsy. Rather than using drugs to depress or slow brain function--and tolerating side effects as the lesser of 2 evils--by reducing the circumstances that lead to seizure activity, clinicians ultimately may be able to treat the malfunctioning brain cells that mediate epileptic seizures. The study, titled "An astrocytic basis of epilepsy" by GF Tian, H Azmi, T Takano, et al, appears in the August 14 online issue of Nature Medicine. *

Peripheral diabetic neuropathy may originate in bone marrow, according to a team of investigators from the Baylor College of Medicine in Houston and Shiga University in Japan. Lead investigator Lawrence Chan, MD, chief of diabetes, endocrinology, and metabolism at Baylor, used colorful--and timely--imagery to describe the phenomenon in a press release: "These insulin-producing bone marrow cells are like terrorists that infiltrate the nerve-cell populations. They produce proteins that can kill or subvert the purposes of the nerve cells almost like a suicide bomb," he said.

While using murine models to research gene therapy for diabetes, Chan and his team discovered insulin-producing bone marrow cells that were migrating to different organ systems. They embarked on a series of experiments in animal models to learn more about the activity of these rogue cells, keeping in mind earlier studies by a team from Stanford University that showed how migrating bone marrow cells cross the blood-brain barrier and interact with neurons. "Some neurologists are not aware that bone marrow cells can cross the blood-brain--or blood-nerve--barrier and fuse with nerve cells," Chan noted in a communication with Applied Neurology. He referred to 2 studies, the findings of which were published in 2003 and were conducted by a team led by Helen M. Blau, PhD, director of gene technology therapy and the Baxter Laboratory in Genetic Pharmacology at Stanford University. These studies documented the fusion of bone marrow cells with Purkinje neurons.

Chan and his colleagues discovered that insulin-producing bone marrow cells set the stage for neuropathology by migrating to the sciatic nerve and dorsal root ganglion, where they fuse with neurons. These hybrid, fused cells, which were polyploid, express tumor necrosis factor a, a toxic gene product associated with peripheral diabetic neuropathy. Neuronal dysfunction and apoptosis ensue in the presence of clinical signs of neuropathy. "The difference between our finding and that of Weimann et al [Blau's team] is that we have a pathological situation," Chan explained.

"We speculate that a similar process contributes to some, if not all, of the other chronic complications of diabetes, and we look forward to pursuing [research on] this possibility," Chan said. The researchers believe that their findings are a significant step forward in defining the underlying cause of diabetic neuropathy and hope that they will be useful in developing treatment strategies for this common and debilitating condition.

The research article appears in the August 30 issue of the Proceedings of the National Academy of Sciences USA. The citation is Terashima T, Kojima H, Fujimiya M, et al. The fusion of bone-marrow-derived proinsulin-expressing cells with nerve cells underlies diabetic neuropathy. Proc Natl Acad Sci U S A. 2005;102:12525-12530.

Citations for the studies by Blau and colleagues are:

• Weimann JM, Johansson CB, Trejo A, Blau HM. Stable reprogrammed heterokaryons form spontaneously in Purkinje neurons after bone marrow transplant. Nat Cell Biol. 2003;5:959-966.

• Weimann JM, Charlton CA, Brazelton TR, et al. Contribution of transplanted bone marrow cells to Purkinje neurons in human adult brains. Proc Natl Acad Sci U S A. 2003; 100:2088-2093. *

The clinical utility of the growth hormone-releasing hormone (GHRH) antagonists in the treatment of glioblastomas has hung on whether GHRH antagonists could cross the blood-brain barrier at sufficiently therapeutic levels. The ability of such agents to arrest tumorigenesis in the brain could augur a major advance in the treatment of gliomas by, for example, inhibiting relapse in patients in whom brain tumors have been incompletely excised. Now, a multicenter team of researchers from Missouri and Louisiana, some of whom were the early pioneers of research into the antitumor properties of GHRH antagonists, has recently shown that the GHRH antagonist JV-1-42 indeed crosses the blood-brain barrier at therapeutic levels.

Several studies, dating back to the late 1990s, showed that GHRH antagonists can intercept GHRH-mediated tumorigenesis in various organ systems. Nobel Laureate in Physiology and Medicine (1977) Andrew V. Schally, MD, PhD, MDHC, a senior investigator at the Endocrine, Polypeptide, and Cancer Institute at the Veterans Affairs Medical Center and chief of experimental medicine at Tulane University in New Orleans, has been involved in many of these studies. One such study, published in the May-June 2000 issue of Neoplasia, demonstrated the value of GHRH antagonists, specifically MA-5-156 and JV-1-36, in significantly inhibiting tumor growth in mice that bear extracranial xenografts of human glioblastomas.

The mechanism of antitumor activity of GHRH antagonists is thought to be suppression of insulinlike growth factors I and II. After demonstrating that certain GHRH antagonists could arrest tumorigenesis in human glioblastomas grafted to the flanks of mice, Schally and a group of researchers from the St Louis University School of Medicine; the Geriatric Research, Education, and Clinical Center of the Veterans Affairs Medical Center in St Louis; and Tulane University sought to confirm that a GHRH antagonist could cross the blood-brain barrier at therapeutic levels.

It's the job of the membrane transporter P-glycoprotein to act as a gatekeeper through which molecules, including endogenous molecules, enter the brain. "The blood-brain barrier is set up to very carefully patrol what it lets into the brain and what it keeps out" because most drugs that fight cancer are toxic to normal cells as well as cancer cells, explained one of the study investigators, William A. Banks, MD, professor of geriatrics and of pharmacologic and physiologic science at St Louis University.

The researchers conducted a number of experiments in which groups of mice were injected (in the carotid artery, jugular vein, lateral ventricle, or elsewhere, depending on the primary end point of the particular experiment) with a solution containing radioactively labeled JV-1-42. The level of radioactivity within the brain as well as a series of other issues, including clearance of the GHRH antagonist from serum, uptake in the brain and cerebrospinal fluid, stability in the blood and brain, and capillary depletion, were analyzed.

The researchers reported that JV-1-42 "crosses the blood-brain barrier intact at a rate of 0.8514 microl/g per minute, with a serum half-life of 12.2 minutes." They concluded that other GHRH inhibitors may have the same properties and may prove useful in the treatment of malignant glioblastomas and, possibly, other types of brain cancers.

For more information on this study and the study appearing in the May-June 2000 issue of Neoplasia, see:

• Jaeger LB, Banks WA, Varga JL, Schally AV. Antagonists of growth hormone-releasing hormone cross the blood-brain barrier: a potential applicability to treatment of brain tumors. Proc Natl Acad Sci U S A. 2005;102: 12495-12500.

• Kiaris H, Schally AV, Varga JL. Antagonists of growth hormone-releasing hormone inhibit the growth of U-87MG human glioblastoma in nude mice. Neoplasia. 2000;2: 242-250. *

A German study that compared the value of electric stimulation (ES) using electromyographic biofeedback with that of computerized arm training as therapy for arm paresis in patients recovering from stroke probably shed more light on the intensity of therapy needed than it did on therapeutic technique, according to the study's investigators. Although more patients regained mobility and muscle strength by using the experimental arm trainer (Bi-Manu-Track robotic arm trainer, Reha-Stim) than from ES, patients using the arm trainer performed many more repetitions than those using ES. Also, arm trainer therapy used a bilateral approach.

The study was led by Stefan Hesse, MD, a researcher and specialist in stroke rehabilitation at the Klinik Berlin, and included a team, as well as patients, from his institution in Berlin and the Klinik Bavaria in Kreischa. Twenty-two patients from each center who met study criteria were randomly assigned to one or the other intervention, which they participated in for 20 minutes each day 5 days a week for 6 weeks. Patients also received other standard forms of stroke rehabilitation. Electromyographic biofeedback was performed in accordance with standard recommendations from the literature.

After 6 weeks, the average motor function score for patients training with the computerized arm was 24.6, whereas the score for patients using electromyographic biofeedback was 10.4. The average score for upper limb muscle strength was 21.8 among patients using the computerized arm and 6.8 among those receiving ES.

Although superior results were achieved with the computerized arm trainer, the study investigators pointed out that those using it performed 24,000 repetitions that incorporated 4 movement directions and use of the nonaffected arm, whereas those using ES performed only 1800 to 2400 repetitions of wrist extensions in accordance with the recommendations for use discussed in the scientific and clinical literature. The researchers concluded that, more important than the actual hardware used for therapy, intensity of repetition and a bilateral approach were probably key to the results seen.

The study appears in the September issue of Stroke. The citation is Hesse S, Werner C, Pohl M, et al. Computerized arm training improves the motor control of the severely affected arm after stroke. A single-blinded randomized trial in two centers. Stroke. 2005;36:1960-1966. *