Scientists Probe Roles of Mitochondria in Neurological Disease and Injury

New Orleans—If ATP is the fuel that keeps the cellular engine running, mitochondria are the gas stations open all night for service. But these "powerhouses of the cell" have a dark side. Provoked by stressful conditions, mitochondria—semiautonomous organelles that maintain their own genome and life cycle—can be transformed from beneficent energy suppliers into ticking time bombs capable of triggering events that lead to cell damage and death.

Mitochondria in brain cells are subject to damage from various agents, including toxic stimuli and inflammatory mediators. In some cases, the cell loss that occurs in neurological diseases and injuries can be traced back to harm that befalls mitochondria in brain cells—not just neurons, but also the supporting astrocytes. New findings about mitochondrial function and dysfunction presented at the annual meeting of the Society for Neuroscience are providing insights into how mitochondria influence brain health and disease and pointing to new targets for clinical intervention.

Traditionally, the focus in stroke has been on neurons rather than astrocytes, which are quite resistant to brain ischemia, said Laura Dugan, MD, of Washington University School of Medicine, St Louis. In the last few years, however, researchers have discovered that astrocytes themselves come under stress during a stroke and can negatively affect the course of disease.

Once considered little more than the glue that holds neurons in place, astrocytes are now becoming appreciated for the key roles they play in brain function. These cells perform a number of tasks that are important to neuronal survival, including glutamate uptake, calcium buffering, and transfer of lactate or pyruvate to neurons as energy substrates.

Some of these tasks depend on the mitochondrial membrane potential, the driving force behind cellular respiration and ATP production. In examining the role astrocytes play in the damage wrought by ischemic stroke, Dugan and colleagues have found that mitochondrial metabolic function in astrocytes can be impaired and may render these cells less capable of rescuing neurons during ischemia—work suggesting that astrocytes might be useful therapeutic targets in stroke.

In cell culture conditions mimicking stroke—oxygen-glucose deprivation for 60 minutes, an exposure that is lethal to neurons but not to astrocytes—Dugan's group found that the astrocyte mitochondrial membrane potential became depolarized, an important indicator of mitochondrial dysfunction. The astrocyte mitochondria return to normal after oxygen and glucose are restored, although more slowly than expected, which suggests that damage does occur to mitochondrial function during ischemia. The mitochondria also release free radicals, including superoxide and hyperoxide peroxide, highly reactive chemicals that are linked to tissue damage and which appear to turn on signaling pathways in nearby neurons.

The researchers found two factors that can bring about depolarization of mitochondrial membranes when the cultured astrocytes were deprived of oxygen and glucose: the opening of the mitochondrial permeability transition pore (mPTP)—a channel that releases substances from the mitochondria—and the presence of nitric oxide and its reactive by-products. But in vitro studies indicate that the loss of the mitochondrial membrane potential under these conditions can be partially offset with either of two drugs: the immunosuppressant drug cyclosporine A, which inhibits the mPTP, and ^GN-nitro-arginine, an agent that inhibits nitric oxide synthase, the enzyme that synthesizes nitric oxide.

These results suggest that neuronal injury may in part reflect astrocyte damage, said Dugan. By targeting astrocyte mitochondria with drugs such as cyclosporine A, astrocytes may be better able to protect neurons at risk in neurological disease and trauma, although further work needs to be done to determine whether this is the case.

While nitric oxide can induce mitochondrial membrane depolarization in cultured astrocytes deprived of oxygen and glucose, in other circumstances, it actually brings about hyperpolarization of the astrocyte mitochondrial membrane, a condition that renders the astrocyte better able to withstand injury.

Astrocytes seem to have a fall-back measure to protect themselves when exposure to nitric oxide occurs. This protective mechanism, identified by Juan Bolanos, PhD, of the University of Salamanca, Spain, and colleagues, is the ability to rapidly upregulate glycolysis, the first step of cellular respiration that occurs in the cytosol outside the mitochondria. Glycolysis is able to produce enough ATP to maintain the membrane potential when nitric oxide inhibits cellular respiration. But this compensatory mechanism cannot function during oxygen-glucose deprivation, when glucose is unavailable, said Bolanos. In contrast, neurons do not seem to be able to increase their glycolytic activity after nitric oxide exposure, and mitochondrial membrane depolarization and cell death can result.

Recently, the Bolanos team has found evidence to explain these differing responses to nitric oxide in neurons and astrocytes (Nat Cell Biol. 2003;6:45-51). They identified a signaling pathway involving two key proteins—AMP-activated protein kinase and 6-phosphofructo-2-kinase (PFK-2)—that turns on glycolysis in astrocytes exposed to nitric oxide.

Astrocytes, it seems, have detectable amounts of PFK-2 that they can rely on in times of stress, but neurons do not. The inability of neurons to activate this pathway when they are in jeopardy may explain why neurons are more vulnerable than astrocytes to nitric oxide, and may be a key factor in their susceptibility to neurodegeneration, said Bolanos.

While the clinical implications of these findings for neurological conditions are limited at this point, Bolanos said that they do suggest that administering precise amounts of nitric oxide to upregulate this defense mechanism of the astrocytes might be one approach to improving cell resistance to injury for various neurological conditions.

These findings are also relevant to cancer research, said Bolanos. Because cancer cells consume huge amounts of glucose through glycolysis and have more PFK-2 activity than normal cells, he noted, PFK-2 and AMP-activated protein kinase may provide useful targets in cancer therapy.

Other findings from mitochondrial research suggest that thiazolidinediones (TZDs), a class of drugs used to treat patients with type 2 diabetes, hold promise for the treatment of a variety of neurodegenerative diseases. TZDs regulate insulin sensitivity, and two of these agents—pioglitazone and rosiglitazone—have received approval from the US Food and Drug Administration for diabetes treatment.

But TZDs also influence astrocyte metabolism and mitochondrial function, sparking interest in using these agents against neurological disease and injury. Small clinical trials to examine the safety of using TZDs for patients with Alzheimer disease and multiple sclerosis are planned for 2004, said Douglas Feinstein, an associate professor of anesthesiology at the University of Illinois, Chicago, who studies the effects of these agents in the brain.

TZDs perform in part by binding to proteins called peroxisome proliferator-activated receptors (PPARs), primarily a protein called PPAR-γ. In astrocytes, TZDs are able to reduce the expression of several genes involved in inflammation, such as the genes encoding the inducible form of nitric oxide synthase, interleukin 1-β, and tumor necrosis factor-α—all of which contribute to neurological damage in multiple sclerosis and Alzheimer disease, said Feinstein.

In a mouse model of multiple sclerosis, for example, Feinstein's group has shown that TZDs and other PPAR-γ agonists can prevent disease onset as well as bring about remissions in sick animals (Ann Neurol. 2002;51:694-702). Promising findings have also been demonstrated in animal models of Parkinson disease and Alzheimer disease.

The protective effects of TZDs in these conditions, however, are not limited to their anti-inflammatory action. TZDs also appear to increase glucose metabolism in astrocytes, making them more resistant to damage by noxious agents and also may improve their ability to provide lactate to neurons when blood flow or glucose supplies are limited—effects that are independent of PPAR-γ. The investigators have found that TZDs directly influence the astrocyte mitochondria, eventually leading to hyperpolarization of the mitochondrial membrane potential, thereby making them more resistant to toxic stimuli (J Biol Chem. 2003;278:5828-5836).

In addition, TZDs may have other beneficial effects on the brain, Feinstein and colleagues are finding in animal studies. These drugs apparently can inhibit T-cell proliferation, a key component in multiple sclerosis, as well as elicit a heat-shock response, which has been shown to be protective in several neurological conditions.

Mitochondria also may contribute to brain disease and injury if they are unable to get to where they are needed in the cell, such as when the organelles are immobilized by various neurotoxins, said Ian Reynolds, PhD, professor of pharmacology at the University of Pittsburgh.

Mitochondria are believed to be generated near the nucleus and are then transported to sites within the cell where energy is required. In some of the longest neurons in the body, noted Reynolds, a mitochondrion conceivably could travel as far as a meter from nucleus to synapse.

Mitochondria travel along the highways and byways of the cytoskeleton (online video available), but roadblocks and detours can subvert this mitochondrial trafficking, said Reynolds. "If a mitochondrion is not delivered to a terminal when needed—or, alternatively, if it gets retrieved from a terminal too soon—this could perhaps contribute to the death of a neuron," he speculated. In cell culture studies, Reynolds and colleagues have discovered that several substances that can be released in large amounts in a stroke, including glutamate, zinc, and nitric oxide, can inhibit mitochondrial movement.

However, Reynolds and coworkers found that the neurons could be protected in vitro with the drug wortmannin, which stops zinc from blocking mitochondrial movement. Although preliminary, these findings suggest that targeting substances that inhibit mitochondrial movement may be another useful strategy to investigate for treating neurological disorders.