April 1, 2009 (Vol. 29, No. 7)

Richard A. A. Stein M.D., Ph.D.

Increasingly Dynamic Field Hones In on the Many Facets of Mitochondrial Biology

As Immo Scheffler so insightfully remarks in his book Mitochondria, “where there is DNA, there must be mutations.” Few topics are so profoundly rooted in the history of science yet remain as actively expanding and full of surprises as mitochondrial biology. Recent advances position this organelle as a key participant in a plethora of cellular-signaling networks and increasingly reveal its involvement in disease pathogenesis.

While historically mitochondria were depicted as individual organelles, this view is challenged by recent findings, which point toward an active interchange of mitochondrial contents as a result of dynamic fusion and fission events. “We found that mitochondrial fusion is necessary to protect the function of the mitochondrial population,” says David C. Chan, M.D., Ph.D., associate professor of biology at Caltech and Howard Hughes Medical Institute investigator.

The Chan group recently generated a conditional mouse knockout for Mfn2, the gene encoding mitofusin 2, a mitochondrial GTPase involved in fusion. “There is something about the long neurons that makes them prone to defects in mitochondrial fusion,” says Dr. Chan. Mitochondria in Mfn2 knockout mice exhibited inner membrane ultrastructural defects, impaired respiratory complex activity, reduced fusion, and increased fragmentation, all of which lead to morphological changes and selective neurodegeneration of cerebellar Purkinje cells.

Impaired localization of mitochondria, which were present in cell bodies, but rarely found in the dendritic branches, emerged as an additional defect. Another transgenic mouse model generated by the Chan group expresses a pathogenic Mfn2 allele and recapitulates several features observed in Charcot-Marie-Tooth 2A, thus becoming a promising model to dissect mitochondrial defects underlying this neurodegenerative condition that selectively affects motor and sensory functions in long peripheral neurons.

A GFP fusion in the outer mitochondrial membrane (Michael Ryan, Ph.D., and Catherine Palmer/ La Trobe University)

Mitochondrial Modifications

Mitochondrial modifications indicative of metabolic impairment appear to represent one of the earliest events in neurodegenerative conditions and precede pathological changes or functional impairment. “Indeed several different groups have reported that in mouse models of Alzheimer’s disease, mitochondrial dysfunction is observable before pronounced pathology appears, indicating early causative mitochondrial defects,” says Gail V.W. Johnson, Ph.D., professor of anesthesiology and of pharmacology and physiology in the mitochondrial research and innovation group at the University of Rochester Medical Center.

The Johnson lab focuses on understanding early cellular modifications in neurodegenerative conditions. In Huntington’s disease, CAG expansions in a ubiquitously expressed protein, huntingtin, cause neuronal loss in the striatum, but the molecular mechanism(s) responsible for this selectivity and the early events causing cellular damage are poorly understood. Research in the Johnson lab supports the hypothesis that mitochondrial dysfunction is a contributing factor to neuronal pathology.

Striatal cells derived from a Huntington’s disease mouse model showed increased sensitivity to mitochondrial complex II inhibition compared to striatal cells from wild type mice, and they also exhibited mitochondrial calcium-handling defects and impaired respiration. Other groups have shown that mitochondria from Huntington’s disease patients and mouse models undergo permeability transition pore opening at significantly lower calcium levels than control mitochondria and, overall, increasing evidence implicates mitochondrial abnormalities as key in the pathogenic process.

In Alzheimer’s disease, accumulation of amyloid beta, the primary component of neuritic plaques that represent the neuropathological hallmark of this condition, was recently shown to increase mitochondrial fragmentation, and there is evidence that tau proteins, which form the neurofibrillary tangles observed in the brain during the course of the disease, mediate amyloid beta toxicity.

Furthermore, a recent report established a link between increased cytosolic calcium levels and mitochondrial fission. “Given that dysregulation of calcium has been associated with Alzheimer’s and Huntington’s diseases, this could be a common contributing factor to the mitochondrial dysfunction that occurs in these diseases; however, further studies are needed,” explains Dr. Johnson.

The combined and integrated use of multiple experimental approaches provides an important driving force that fuels scientific advances. “Genomics and pharmacogenetics are here today. The largest barrier will be the understanding of which genetic variations are important and how they affect protein expression or function. The average person utilizes about 40 kg of ATP in a day.  Understanding allelic differences in any of the 18 genes encoding subunits of the ATP synthase will be challenging, but important,” says David M. Mueller, Ph.D., professor of biochemistry and molecular biology at the Chicago Medical School of Rosalind Franklin University.

To gain structural and functional insight into the mitochondrial ATP synthase and the coupling mechanism that converts proton translocation energy into conformational changes resulting in ATP synthesis, the Mueller lab uses a combination of genetics, biochemistry, and structural biology. Two approaches, one that introduces specific mutations known to affect coupling and examines their effect on enzyme biochemistry and structure, and its reverse, which makes structure-based predictions and subsequently examines the impact of individual mutations on coupling, both emerged as important instruments in dissecting mitochondrial biology.

“At this time, crystallizing the entire ATP synthase is an important goal, and we are making progress on that,” says Dr. Mueller. Previously, the Mueller group provided a high-resolution crystal of the yeast mitochondrial F1-ATP synthase which, in combination with the only other previously solved structure, that of the bovine enzyme, explained mechanistic details involved in the conformational changes during ATP synthesis.

“Understanding mitochondrial biology in higher organisms is important,” says Michael Ryan, Ph.D., associate professor of biochemistry at La Trobe University in Melbourne. “Most of the early work has used yeast, which is a great model organism, to look at basic processes. However, in humans, mitochondria appear to be involved in so many more processes that can affect neighboring cells. This includes apoptosis and damage by a generation of reactive oxygen species.”

At Gordon Research Conference’s “Molecular and Cellular Bioenergetics”  meeting to be held in June, Dr. Ryan will present recent advances his group made in characterizing new complex I assembly factors. Human complex I, or NADH: ubiquinone oxidoreductase, located in the mitochondrial outer membrane, comprises 45 different subunits, of which only seven are encoded by the mitochondrial DNA. The remaining 38 are encoded by the nuclear genome and imported into the mitochondria for assembly.

The lack of a crystal structure for the ~1 MDa complex I, along with its absence in yeast, explain why relatively little is known about its biology. However, isolated complex I defects were estimated to represent the most common group of mitochondrial diseases, and many are lethal shortly after birth or in early childhood. The Ryan group characterized the intermediates and assembly pathways as different complex I subunits, and in collaboration with David Thorburn, Ph.D., associate professor at the Murdoch Children’s Research Institute in Melbourne, examined almost 100 different cell lines originating from patients with isolated complex I deficiency, half of which did not harbor mutations in any complex I subunit-encoding genes.

Screening these cell lines with antibodies against CIA30 (complex I intermediate associated protein), an assembly factor initially identified in the fungus Neurospora crassa, previously led to the identification of the first patient with CIA30 deficiency, exhibiting an ~70% decrease in complex I levels that were restored to near normal after wild type CIA30 was expressed in the patient’s cultured cells. “I think we are going to find a lot of new proteins involved in complex I assembly,” predicts Dr. Ryan. “Defects in complex I are thought to generate damaging ROS and cause cell death. This area needs to be looked at in closer detail.”

Atomic Detail

“I think the challenge for the next decade will be to link atomic detail with physiological function and disease. At this time, this is possible only in a few cases, if at all,” says Ulrich Brandt, Ph.D., professor of molecular bioenergetics at the center of biological chemistry of the Goethe-University Frankfurt. Complex I represents the point of entry for the majority of electrons into the respiratory chain, and a hydrophobic segment embedded in the membrane together with a hydrophilic arm lying orthogonal to it, toward the mitochondrial matrix, give it an L-shaped appearance.

“Most people make the obvious assumption that electrons have to be transferred close to the membrane because the substrate, ubiquinone, is extremely hydrophobic,” explains Dr. Brandt. At the “Molecular and Cellular Bioenergetics” meeting he will present recent findings that suggest that ubiquinone reduction takes place not in the membrane, but approximately 60Å away, in the peripheral arm, and this can be explained by envisioning ubiquinone motion occurring along a channel within the protein complex.

While the recently solved partial x-ray structure of the peripheral, hydrophilic arm provides clues about the electron transfer process, it does not reveal details about proton pumping, thought to occur in the membrane segment. Electron microscopy single particle analysis using complex I mutants with selective proton pumping defects, performed in collaboration with Michael Radermacher, Ph.D., associate professor of molecular physiology and biophysics at the University of Vermont, revealed the existence of two proton pumping sites.

Experimental approaches in the Brandt lab recently answered another long-standing question in mitochondrial biology. During recovery from ischemic episodes, reperfusion itself may cause organ damage. But in what has become known as ischemic or pharmacological preconditioning, brief episodes of ischemia and reperfusion, or certain compounds such as diazoxide, exert a protective effect if administered prior to a longer ischemic event.

The molecular basis of the diazoxide activity has been debated for a long time, and the Brandt group recently revealed that two opposing effects of this compound, an increase in complex III reactive oxygen species (ROS) and a reduction in complex I ROS, can both be explained by an inhibitory action on mitochondrial complex II.

While in vitro systems provide great contributions to neuronal biology, they do not accurately revisit the complexity of interactions in which live organisms participate. Peter J. Hollenbeck, Ph.D., professor of neurobiology at Purdue University, will present at the meeting in vivo work that his group performed using a Drosophila strain with GFP-tagged mitochondria generated by William Saxton, Ph.D., at the University of California Santa Cruz.

“With this construct, under the confocal microscope, we can look at mitochondrial movement up and down the axon, and the axons are part of an intact neuron; it is kind of like a dream, after years of looking at neurons that divide in a culture dish,” says Dr. Hollenbeck.

Research in the Hollenbeck group used this system to examine cellular changes in Friedrich’s ataxia, and the results were surprising. While excessive oxidative damage is reported for most mitochondrial disorders, no evidence of ROS was found up to the moment of neuronal death in this disease model. Instead, large numbers of depolarized mitochondria, with a low membrane potential and inability to generate ATP, accumulated at the synapse. “This is quite counterintuitive,” says Dr. Hollenbeck. “Instead of oxidative damage, relatively inactive mitochondria are piling up at the end of the axon, and that seems to underlie neuronal death.”

The movement of GFP-mitochondria in an intact segmental nerve of a live Drosophila larva (Peter J. Hollenbeck, Ph.D., Purdue University)

Signaling Pathways

Increasingly, mitochondria have surfaced as essential players in many signaling pathways. “In recent years, the widely held view that mitochondria are simply stewards of energy metabolism and apoptosis is giving way to the realization that they have multiple, additional functions in cells. These include roles in oxygen sensing, signal transduction, and immune system function, to name just a few,” says Gerald S. Shadel, Ph.D., professor of pathology and genetics at Yale University.

“It is, therefore, becoming quite clear,” explains Dr. Shadel, “that mitochondria contribute to human disease and age-related pathology to a much greater degree than previously thought and this involves mechanisms that not only include, but transcend their accepted roles in cellular metabolism and cell death.”

The Shadel lab cloned h-mtTFB1, a mitochondrial transcription factor that concomitantly functions as an rRNA methyltransferase, and linked it to maternally inherited deafness, possibly via its methylation of a conserved 12S rRNA stem-and-loop structure.

In collaboration with Akiko Iwasaki, Ph.D., the Shadel group recently established a connection between mitochondrial ROS and antiviral signaling, and revealed that in the absence of autophagy, cells accumulate dysfunctional mitochondria, which produce more reactive oxygen species and activate antiviral signaling pathways.

In another exciting development, the Shadel lab revealed that inhibition of TOR signaling affects the mitochondrial proteome dynamics by increasing both mitochondrial- and nuclear- encoded protein subunits involved in oxidative phosphorylation, along with other proteins localized to the mitochondria.

A few years ago, Vamsi Mootha, M.D., associate professor of systems biology at the Broad Institute, used a DNA microarray-based approach to reveal that mitochondrial gene expression is reduced in prediabetic and diabetic muscle, and subsequent work from other groups revealed increased ROS in certain prediabetic tissues. A recent chemical screen performed in the Mootha lab that relied on four types of cell-based assays, including multiplexed gene-expression analysis, tested almost 2,500 compounds—40% of them FDA-approved drugs—for their effects on mitochondrial physiology, and unveiled a set of compounds that increase mitochondrial gene expression and reduce ROS.

“We are excited about this result because we are able to reverse, at least in cell culture, two of the signatures of the diabetic muscle,” says Dr. Mootha. One class of compounds is a group of 7–8 microtubule stabilizers and destabilizers that, although structurally diverse, share the ability to reduce mitochondrial ROS and stimulate the transcriptional regulators of mitochondrial gene expression. This freely available compendium provides a valuable discovery tool for exploring mitochondrial signaling, mitochondrial drug toxicity, and mitochondrial therapeutics.

Mitochondrial biology illustrates how multiple disciplines converge to define new concepts, which sometimes challenge old views, making cellular bioenergetics an increasingly dynamic field. Insights into novel cellular pathways and promises of new therapeutic targets provide a testimony of the reputation that this organelle righteously enjoys, a reputation that Nick Lane, in his book Power, Sex, Suicide: Mitochondria and the Meaning of Life, so relevantly refers to as “a badly kept secret.”