Catalytic converters compromised
Mitochondria supply cells with heat and energy in the form of adenosine triphosphate (ATP). But as with any powerhouse, the organelles also generate some pollution. Reactive oxygen species (ROS) are byproducts of mitochondrial respiration that damage essential cellular macromolecules, such as nucleic acids and proteins. Energy hogs such as neurons and muscle cells are particularly vulnerable to ROS, and indeed these toxic byproducts have been linked to numerous neurodegenerative diseases including Parkinson and Alzheimer diseases (see ARF related news story). However, most of us seem to cope with this subcellular pollution just fine, thanks to a variety of enzymes that catalyze the conversion of toxic ROS to innocuous molecules, such as water. So is there something different about mitochondria in disease states?

Wondering just that, Spiegelman from Dana-Farber Cancer Institute, and colleagues there and at Beth Israel Deaconess Medical Center, and Brigham and Women’s Hospital, all in Boston, tested the role of PGC-1 proteins in ROS regulation and neurodegeneration. PGC-1α has been linked to Huntington disease and it is a potent stimulator of mitochondrial biogenesis and respiration, suggesting the coactivator might spur increases in ROS. But Spiegelman and coworkers showed that coactivators actually help to detoxify ROS. When they used RNAi to knock down PGC-1α or PGC-1β in mouse embryonic fibroblasts (10T1/2 cell line), first author Julie St-Pierre and colleagues found that a variety of ROS scavengers, including copper/zinc superoxide dismutase (SOD1), manganese SOD (SOD2), glutathione peroxidase (GPx1), and catalase were all reduced by about half. They also found that peroxisomal catalase and SOD1 and 2 were downregulated in heart and brain tissue from PGC-1α-null mice, while in fibroblasts from these animals the response to oxidative challenge, in the form of hydrogen peroxide, was blunted compared to that in wild-type cells.

These findings suggest that loss of PGC-1 transcriptional coactivators may overexpose cells to the ravages of ROS. To test this, St-Pierre and colleagues subjected mice to sub-lethal doses of MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine), a mitochondrial toxin that induces Parkinson disease in mammals. PGC-1α-null mice lost about 60 percent of dopaminergic neurons in the substantia nigra (SN), that part of the brain most seriously damaged in PD patients. In contrast, wild-type animals only lost about 12 percent of SN neurons. These losses seemed directly related to oxidative stress because the PGC-1α-null mice had much higher amounts of nitrotyrosine in their brains. Nitrotyrosine is a stable adduct that forms when proteins react with peroxynitrite, a ROS byproduct.

The findings help explain why loss of the PGC-1α in knockout mice causes symptoms that are reminiscent of Huntington disease (see ARF related news story). Just recently, in fact, Dimitri Krainc and colleagues at Massachusetts General Hospital found that expression of PGC-1α seems to be suppressed by toxic, polyglutamine-expanded huntingtin (see ARF related news story).

If PGC-1α dysregulation might be linked to HD and now PD, what about other neurodegenerative diseases? St-Pierre and colleagues similarly challenged mice with kainic acid (KA), the classic excitotoxin that kills neurons by overstimulation of glutamate receptors. The authors found that KA induced much more apoptosis in the CA1 region of the hippocampus in PGC-1α-null mice than in wild-type animals. Again, oxidative stress was much higher in the nulls, as judged by higher levels of 8-oxoguanine, a nucleic acid derivative. The findings indicate that PGC-1 coactivators are needed to detoxify ROS in regions of the brain linked to different neuropathologies. Of course, the hippocampus is an area of the brain particularly hard hit in AD patients.

The cold shoulder—uncoupling uncoupled
If the PPARγ coactivators can have such a profound effect on ROS scavenging, what about ROS production? Mitochondria are equipped with a mechanism to uncouple the electron transport chain from ATP production (see ARF related news story). This speeds up the flow of electrons, reducing the time to react with oxygen and form toxic byproducts. This process not only reduces ROS, but it is also a principal mechanism for generating heat. It turns out that uncoupling proteins are also regulated by PPARγ coactivators. St-Pierre and colleagues found that when PGC-1α is knocked down in MEF cells, expression of uncouplers UCP2 and UCP3 is downregulated and the upregulation of UCP3 in response to hydrogen peroxide is compromised. In the second paper, Albert La Spada and colleagues at the University of Washington, Seattle, report that aberrant PGC-1α transcriptional coactivation is responsible for a variety of metabolic effects in an HD mouse model, including poor thermoregulation and downregulation of uncoupling protein 1 (UCP-1).

In their studies on transgenic mice (HD N17182Q) that express human huntingtin protein with 82 glutamine repeats, La Spada and colleagues noted that the animals were profoundly hypothermic. Brown adipose tissue is mainly responsible for thermoregulation in mice, a process that is mediated by PPARγ, PGC-1 and the retinoic acid receptor (RxR1), which cooperate to upregulate UCP-1. To test if the transcriptional activation process is normal in these animals, joint first authors Patrick Weydt and Victor Pineda compared gene expression in mice that were given a cold challenge (3 hours at 4 degrees centigrade). They found that while wild-type and transgenic mice had comparable expression of PGC-1α before and after the challenge, only the wild-type animals had elevated (threefold) UCP-1 afterwards. UCP-1 in the transgenic animals remained constant.

To tease out why the mice failed to upregulate UCP-1, the authors challenged pre-adipocyte cells with troglitazone, a PPARγ agonist. In cells transfected with PPARγ, RxR1, and huntingtin protein with 24 glutamine repeats (htt 24Q), troglitazone doubled the activity of a UCP-1 reporter. However, when htt 24Q was replaced with a toxic form of huntingtin, htt108Q, the drug had no effect unless exogenous PGC-1α was also added. These results support Krainc’s previous work linking polyglutamine-expanded huntingtin to loss of PGC-1α activity.

In an interesting twist, Weydt, Pineda, and colleagues found that the thermoregulation and survival of the HD mice are linked. During the cold challenge, mice that failed to regulate their body temperature developed motor problems that strikingly resemble those seen in HD patients. Keeping these mice at higher than normal temperatures (30 degrees) extended their lifespan by 24 days, or 15 percent. Of course, it should be emphasized that this may not be directly relevant to humans. “As humans have very little brown fat and therefore do not regulate body temperature as rodents do, there is little reason to expect that HD patients will display hypothermia,” write the authors.

But the basic finding that PGC-1α transcriptional activation is compromised in the presence of mutant huntingtin does appear very relevant to the human condition. La Spada and colleagues tested for transcriptional interference in both mouse and human striatum. They found that in 20-week-old HD mice, expression of PGC-1α target genes was significantly reduced, while microarray analysis of human brain tissue showed a similar effect. Out of 26 genes chosen that rely on PGC-1α for expression, 24 were significantly reduced in the patient samples.

“All these studies suggest that reduced expression of PGC-1α and its targets contributes to HD striatal degeneration and support a role for mutant htt-mediated transcription interference upon PGC-1α,” write La Spada and colleagues, who also suggest that “PGC-1α deserves consideration as a prime therapeutic target.” PPARγ agonists have been tested for AD (see Watson et al., 2005), but given the pre-adipocyte response to troglitazone, induction of PGC-1α might be more beneficial. In fact, Spiegelman and colleagues found that overexpressing PGC-1α by 40 percent protected murine striatal cells and human SH-SY5Y cells from the toxic effects of paraquat, another mitochondrial toxin. The next challenge might be to find ways to upregulate PGC-1α transcription for therapeutic purposes. Spiegelman and colleagues report that the transcription factor CREB activates the gene, but as Krainc and colleagues recently reported, CREB is displaced from the coactivator’s promoter by mutant htt, which might complicate efforts to therapeutically induce PGC-1α production. It would be ironic if the very disease that helped uncover the role of the coactivator in toxicity might be the least likely to benefit from the discovery.—Tom Fagan.

References:
St-Pierre J, Drori S, Uldry M, Silvaggi JM, Rhee J, Jager S, Handschin C, Zheng K, Lin J, Yang W, Simon DK, Bachoo R, Spiegelman BM. Suppression of reactive oxygen species and neurodegeneration by the PGC-1 transcriptional coactivators. Cell. October 20, 2006;127:397-408. Abstract

Weydt P, Pineda VV, Torrence AE, Libby RT, Satterfield TF, Lazarowski ER, Gilbert ML, Morton GJ, Bammler TK, Strand AD, Cui L, Beyer RP, Easley CN, Smith AC, Krainc D, Luguet S, Sweet IR, Schwartz MW, La Spada AR. Thermoregulatory and metabolic defects in Huntington’s disease transgenic mice implicate PGC-1alpha in Huntington’s disease neurodegeneration. Cell Metabolism in press October 19, 2006. Abstract