16 December 2005. One of the puzzles of Parkinson pathology is why death comes to only a subpopulation of neurons in the midbrain. The dopaminergic neurons in the substantia nigra (SN) and their projections are lost, while neighboring dopaminergic neurons in the ventral tegmental area (VTA) survive. Multiple animal models of PD show this pattern, suggesting that different causative insults converge on a single, as-yet unknown toxic mechanism. A common denominator is mitochondrial dysfunction, and reduced complex I activity and energy depletion are involved in several models. Why these changes culminate in cell death for some neurons but not others has been a mystery (see Alzforum live discussion on this topic), but two recent studies are shedding new light on this issue.
Work from Birgit Liss, Jochen Roeper, and colleagues at Marburg University, Germany, and in Oxford, UK, and Kobe, Japan, lays the blame on selective activation of the ATP-sensitive potassium channel (K-ATP). This voltage-dependent potassium channel effectively acts as a metabolic sensor in that it opens in response to ATP depletion and oxidative stress in cells. Presented at the 35th Annual Conference of the Society for Neuroscience, held last month in Washington, D.C., and published online November 20 in Nature Neuroscience, the study shows that neurotoxic mitochondrial complex I inhibitors activate K-ATP channels in SN dopaminergic neurons, causing the cells to lose their electrical activity. The same toxins have no effect on the channel or the electrophysiology of VTA neurons. Knockout mice lacking the channel are resistant to neuron loss in two different models of PD. This surprising role of the K-ATP channel in neuron death provides a link between energy depletion, oxidative stress, and the selective killing of SN dopaminergic neurons in PD.
Using tissue slices and single cells from adult mice, the researchers established that both SN and VTA dopaminergic neurons express functional K-ATP channels. They have a similar subunit composition, where the Kir6.2 (i.e., the pore-forming) subunit couples with the SUR1 regulatory subunit. Lowering ATP levels activated the channels in both types of neurons, resulting in cell hyperpolarization. But when Liss and colleagues added the mitochondrial complex I inhibitors rotenone or MPP+, the results were strikingly different. In SN neurons, the PD toxins both caused hyperpolarization and a slowing and cessation of spontaneous electrical activity, while the same treatment had no effect on VTA neurons. The effects were due to K-ATP activation, since SN neurons from Kir6.2 knockout mice displayed no sensitivity to the toxins.
The complex I inhibitors were able to activate K-ATP in VTA neurons, provided they were combined with a low concentration of a mitochondrial uncoupler called FCCP. In SN neurons, the opposite occurred: Low concentrations of FCCP prevented channel opening in response to complex I inhibitors. The authors explain this reversal of sensitivity by suggesting that mild uncoupling causes cells to resist K-ATP channel opening in response to complex I inhibition. The data from SN neurons fits with observations that increased mitochondrial uncoupling protects against neurodegeneration in rodent and primate models of PD.
The VTA cells, on the other hand, could possess a mild constitutive uncoupling that protects them from rotenone or MPP+. In support of this idea, Liss et al. show that VTA dopaminergic neurons express higher levels of the uncoupling protein UCP2 than do the SN dopaminergic neurons.
To explore the role of K-ATP channels in cell death in vivo, the researchers turned to Kir6.2 knockout mice. When treated with the PD-inducing regimen of chronic, low-dose MPTP, these mice were completely protected from SN dopaminergic neuron loss. Likewise, crossing the Kir6.2 knockouts with mutant weaver mice, which spontaneously develop a PD-like pathology due to mutations in the Girk2 channel gene, resulted in partial amelioration of SN dopaminergic neuron loss.
Summing up their results, the authors write, “In essence, the convergence of mitochondrial complex I dysfunction and oxidative stress (both of which are present in Parkinson disease) on the activity of K-ATP channels provides a previously unknown candidate mechanism for differential vulnerability of DA neurons that would couple metabolic stress with electrophysiological failure and selective death of SN DA neurons.”
The downstream events that link channel opening to cell death need to be worked out. Meanwhile, the results imply that chronic opening of the K-ATP channel as a result of long-term stress can lead to neurodegenerative disease. By contrast, short-term activation of the channel protects the brain against energy depletion. This channel is known to regulate the energy balance of the organism as a whole via its roles in insulin release and glucose homeostasis. Clinically, the K-ATP channel is an important target in type II diabetes, where incomplete closing impairs insulin secretion. The sulfonylurea channel inhibitors glibenclamide and tolbutamide are widely used to stimulate insulin secretion, and could provide an existing strategy for neuroprotection, the authors note.
The results of this study dovetail with a separate study in another PD mouse model, presented at the conference by Antonio Pisani at Tor Vergata University in Rome, Italy. In an ongoing collaboration, Pisani’s group uses electrophysiology to find out what goes wrong in dopaminergic neurons of DJ-1 knockout mice made in the lab of Jie Shen at Brigham and Women’s Hospital, Boston (see also Goldberg et al., 2005). DJ-1 loss-of-function mutations cause early-onset parkinsonism.
In the latest study, Pisani and colleagues challenged the mice with rotenone and deprived them of oxygen and glucose to mimic ischemia. Dopaminergic neurons in substantia nigra of mice lacking DJ-1 proved to be much more sensitive to these stressors than the same kinds of neurons from wild-type mice. K-ATP channels are known to play a major role in the observed electrophysiological response to rotenone. They are also the preferred candidate for mediating the response to oxygen and glucose deprivation, as they are activated when the ATP content falls in conditions of energetic stress. However, when Pisani et al. looked for the mechanism underlying the heightened DA neuron sensitivity, they found that dopaminergic neurons from DJ-1 knockout mice were much more sensitive to the block of the Na/K pump than were other types of neurons in the striatum.
In contrast to the K-ATP channel, the Na/K pump is not voltage-dependent. It follows an electrochemical gradient in order to balance the amount of sodium and potassium across the cell membrane, importing two potassium ions for every three sodium ions it extrudes. Regulating homeostasis to avoid sodium overload inside the cell, this pump is not specific to dopaminergic neurons at all. It depends on ATP and collapses when levels of this cellular fuel drop.
Taken together, the studies suggest that a drop in ATP content might have two consequences. It could block the activity of the Na/K pump, impairing ion homeostasis, and activate ATP-dependent K channels. It is something about the regional specificity of this pump that could mediate the SN dopamine neurons’ unique sensitivity to a loss of cellular energy stores, Pisani suggested. The full set of this data is submitted for publication.—Pat McCaffrey and Gabrielle Strobel.
Liss B, Haeckel O, Wildmann J, Miki T, Seino S, Roeper J. K-ATP channels promote the differential degeneration of dopaminergic midbrain neurons. Nat Neurosci. 2005 Dec;8(12):1742-1751. Epub 2005 Nov 20. Abstract