transcriptional paradigms in mammalian mitochondrial ... · transcriptional paradigms in mammalian...

Transcriptional Paradigms in Mammalian MitochondrialBiogenesis and Function

RICHARD C. SCARPULLA

Department of Cell and Molecular Biology, Northwestern Medical School, Chicago, Illinois

I. Introduction 612A. Overview 612B. Pathophysiology 613

II. The Mitochondrial Genetic System: mtDNA Structure and Inheritance 614III. Transcription of mtDNA 615

A. Mitochondrial transcriptional units 615B. Tfam 616C. mtTFB 617D. Transcription termination and mitochondrial gene expression 617

IV. Nuclear Control of Mitochondrial Function 618A. Nuclear transcription factors governing respiratory gene expression 618B. Nuclear coactivators in mitochondrial biogenesis 623

V. Lessons From Mouse Gene Knockouts 628A. Transcription factor knockouts 628B. Coactivator knockouts 630

VI. Retrograde Pathways 631A. The RTG pathway in yeast 631B. Mammalian retrograde regulation 631

VII. Conclusion 631

Scarpulla RC. Transcriptional Paradigms in Mammalian Mitochondrial Biogenesis and Function. Physiol Rev 88:611–638, 2008; doi:10.1152/physrev.00025.2007.—Mitochondria contain their own genetic system and undergo aunique mode of cytoplasmic inheritance. Each organelle has multiple copies of a covalently closed circular DNAgenome (mtDNA). The entire protein coding capacity of mtDNA is devoted to the synthesis of 13 essential subunitsof the inner membrane complexes of the respiratory apparatus. Thus the majority of respiratory proteins and all ofthe other gene products necessary for the myriad mitochondrial functions are derived from nuclear genes.Transcription of mtDNA requires a small number of nucleus-encoded proteins including a single RNA polymerase(POLRMT), auxiliary factors necessary for promoter recognition (TFB1M, TFB2M) and activation (Tfam), and atermination factor (mTERF). This relatively simple system can account for the bidirectional transcription of mtDNAfrom divergent promoters and key termination events controlling the rRNA/mRNA ratio. Nucleomitochondrialinteractions depend on the interplay between transcription factors (NRF-1, NRF-2, PPAR�, ERR�, Sp1, and others)and members of the PGC-1 family of regulated coactivators (PGC-1�, PGC-1�, and PRC). The transcription factorstarget genes that specify the respiratory chain, the mitochondrial transcription, translation and replication machin-ery, and protein import and assembly apparatus among others. These factors are in turn activated directly orindirectly by PGC-1 family coactivators whose differential expression is controlled by an array of environmentalsignals including temperature, energy deprivation, and availability of nutrients and growth factors. These transcrip-tional paradigms provide a basic framework for understanding the integration of mitochondrial biogenesis andfunction with signaling events that dictate cell- and tissue-specific energetic properties.

Physiol Rev 88: 611–638, 2008;doi:10.1152/physrev.00025.2007.

www.prv.org 6110031-9333/08 $18.00 Copyright © 2008 the American Physiological Society

I. INTRODUCTION

A. Overview

Mitochondria are ubiquitous membrane-bound or-ganelles that are a defining feature of the eukaryotic cell.The organelle is comprised of a soluble matrix sur-rounded by a double membrane, an ion impermeableinner membrane, and a permeable outer membrane. Earlybiochemists recognized the importance of mitochondriaas the sites of aerobic oxidation of metabolic fuels. It isnow well established that they contribute to many impor-tant functions including pyruvate and fatty acid oxidation,nitrogen metabolism, and heme biosynthesis among oth-ers. Most notably, the mitochondrion is the site of theelectron transport chain and oxidative phosphorylationsystem that provides the bulk of cellular energy in theform of ATP (16, 88). Most of the chemical bond energyfrom the oxidation of fats and carbohydrates is convertedto the reducing power of NADH and FADH2 within themitochondrial matrix. The respiratory apparatus consistsof a series of electrogenic proton pumps that convert thisreducing potential to an electrochemical proton gradientacross the inner membrane (Fig. 1). The electrochemicalpotential of this gradient is converted via the ATP syn-thase to the high-energy phosphate bonds of ATP. Inaddition, the gradient can be dissipated in specializedbrown adipocytes by uncoupling proteins to generateheat.

Mitochondria and their chloroplast cousins areunique among eukaryotic extranuclear organelles in thatthey contain their own genetic system. In vertebrates, thissystem is based on a mitochondrial genome consisting ofa circular double-stranded DNA (mtDNA) (Fig. 2). Thegene complement of mtDNA comprises but a small frac-tion of the number of genes necessary to account for the

molecular architecture and biological functions of theorganelle. Because of this limited coding capacity, mito-chondria are genetically semiautonomous in that they relyheavily on the expression of nuclear genes for all of theirbiological functions. For example, the majority of proteinsubunits that comprise the five inner membrane com-plexes of the electron transport chain and oxidative phos-phorylation system are nucleus encoded (Fig. 1). Thelargest of these, complex I, the NADH dehydrogenase, has39 of its 46 subunits specified by nuclear genes, whereasthe smallest, the succinate dehydrogenase (complex II), iscomprised entirely of nucleus-encoded subunits (Fig. 1).Thus nuclear genes must specify most of the structuraland catalytic components directly involved in energy me-tabolism and, in addition, control mitochondrial transcrip-tion, translation, and DNA replication (27, 46, 185, 227).

Mammalian mitochondrial biogenesis is subject tocomplex physiological control. Mitochondrial mass is in-duced in adult muscle in response to exercise (28) orchronic electrical stimulation (232), presumably as anadaptation to facilitate increased oxygen utilization. Cal-cium-dependent signaling has been linked to a transcrip-tional pathway of mitochondrial biogenesis in skeletalmuscle (155, 235). The proliferation of mitochondria oc-curs in the brown fat of rodents during adaptive thermo-genesis coinciding with the activation and induction ofUCP-1, an uncoupling protein that dissipates the protongradient to produce heat as a response to cold exposure(34, 175). During early mouse development, the mitochon-drial DNA (mtDNA) content remains constant from fertil-ization to the blastocyst stage after which DNA replica-tion and organelle division resumes (167). Subsequentmitochondrial proliferation increases continuously through-out development, and tissue-specific factors are thought todictate the final adult complement of mtDNA (90). Therespiratory capacity of mitochondrial membranes is en-

FIG. 1. Summary of protein subunits of the fiverespiratory chain complexes encoded by nuclear andmitochondrial genes. Depicted is a schematic of thefive respiratory complexes (I–V) embedded in thelipid bilayer of the inner mitochondrial membrane.Dissociable electron carriers cytochrome c (Cyt c)and coenzyme Q (Q) are also shown. Arrows (green)show the pathway of electrons from the various elec-tron donors. Broken arrows (blue) show the sites ofproton pumping from the matrix side to the cytosolicside by complexes I, III, and IV. The red arrow showsthe flow of protons through complex V from the cy-tosolic side to the matrix coupled to the synthesis ofATP. Indicated above each complex are the numberof protein subunits encoded by nuclear (nDNA) andmitochondrial (mtDNA) genomes.

612 RICHARD C. SCARPULLA

Physiol Rev • VOL 88 • APRIL 2008 • www.prv.org

hanced postnatally as an adaptation to oxygen exposureoutside the womb (217). These examples bear on a fun-damental question in understanding the biology of eu-karyotic cells, namely, what are the mechanisms of com-munication between physically separated nuclear and mi-tochondrial genetic systems in meeting cellular energydemands?

B. Pathophysiology

In the last 20 years, mitochondrial dysfunction hasbeen recognized as an important contributor to an arrayof human pathologies. Mitochondrial defects play a directrole in certain well-defined neuromuscular diseases andare also thought to contribute indirectly to many degen-erative diseases. Since the identification of the first hu-

man pathological mutation in mtDNA, many such muta-tions have been catalogued. Mutations in mitochondrialgenes for respiratory proteins and translational RNAs,particularly tRNAs, manifest themselves in a wide rangeof clinical syndromes, most of which affect the neuromus-cular system (55, 227). These mtDNA mutations are oftenmaternally inherited, and in some cases, patients withcertain mitochondrial myopathies exhibit excessive pro-liferation of abnormal mitochondria in muscle fibers, theso-called ragged red fiber (226). In addition, a subset ofmitochondrial diseases exhibits a Mendelian inheritancepattern typical of nuclear gene defects (199, 205). Thesecan affect respiratory protein subunits, assembly factors,and gene products required for mtDNA maintenance andstability.

In addition to single gene defects, dispersed lesionsin mtDNA that accumulate over time may play a role inhuman pathologies including neurodegenerative disease(189), diabetes (131), and ageing (57). It has been widelyspeculated that free radical production by the mitochon-drial respiratory chain contributes to the neuropathologyobserved in dementia and other degenerative diseases.Although there is good evidence for oxidative stress as-sociated with neuropathology, it has been difficult toprove whether this is a cause or a consequence of neuro-nal death (5). The accumulation of mutations in mtDNA isalso thought to contribute to human ageing. Documentedchanges in mtDNA, including increased prevalence ofpoint mutations and deletions, have been observed inaged individuals (43). This along with the known age-related decrease in oxidative energy metabolism has ledto the hypothesis that mtDNA mutations impair the respi-ratory chain leading to a decline in oxidative phosphory-lation (57).

Direct experimental support for this model comesfrom recent studies in which a mtDNA polymerase that isdefective in proofreading function was substituted for thenormal polymerase in mice (213). Animals homozygousfor the defective allele had a mutator phenotype withmarkedly increased levels of mtDNA point mutations anddeletions. Although the mutations far exceed those asso-ciated with normal aging, the mutator mice had a reducedlife span and a number of physiological changes related topremature aging. There has also been considerable inter-est in mitochondrial dysfunction as a contributing factorin the onset of type 2 diabetes (161). Insulin resistance inhealthy aged individuals with no family history has beencorrelated with a decline in mitochondrial oxidative phos-phorylation (163). The expression of a number of genesinvolved in oxidative metabolism is reduced in diabeticsubjects as well as in those predisposed to diabetes be-cause of family history (144, 162). Transcriptional activa-tors and coactivators that regulate mitochondrial biogen-esis have been suggested as potential contributors to thisphenomenon. In addition, mitochondrial functional insuf-

FIG. 2. Schematic representation of the human mitochondrial ge-nome. Genomic organization and structural features of human mtDNAare depicted in a circular genomic map showing heavy (blue) and light(black) strands assigned as such based on their buoyant densities.Protein coding and rRNA genes are interspersed with 22 tRNA genes(red bars denoted by the single-letter amino acid code). Duplicate tRNAgenes for leucine (L) and serine (S) are distinguished by their codonrecognition (parentheses). The D-loop regulatory region contains theL- and H-strand promoters (LSP, HSP1, and HSP2), with arrows showingthe direction of transcription. The origin of H-strand replication (OH) iswithin the D-loop, whereas the origin of L-strand replication (OL) isdisplaced by approximately two-thirds of the genome within a cluster offive tRNA genes (W, A, N, C, Y). Protein coding genes include thefollowing: cytochrome oxidase (COX) subunits 1, 2, and 3; NADH dehy-drogenase (ND) subinits 1, 2, 3, 4, 4L, 5, and 6; ATP synthase (ATPS)subunits 6 and 8; cytochrome b (Cyt b). ND6 and the 8 tRNA genestranscribed from the L-strand as template are labeled on the inside of thegenomic map, whereas the remaining protein coding and RNA genestranscribed from the H-strand as template are labeled on the outside.

TRANSCRIPTIONAL PARADIGMS IN MAMMALIAN MITOCHONDRIA 613


ficiency has been found in the insulin-resistant offspringof patients with type 2 diabetes (164). The fact that thisoccurs in healthy individuals that are not diabetic sug-gests that an inherent defect in oxidative phosphorylationmay be a contributing factor. These associations of mito-chondrial dysfunction with human degenerative diseaseraise the basic question of how mammalian cells controlmitochondrial biogenesis. It has become increasingly ap-parent that transcriptional mechanisms contribute to thebiogenesis of mitochondria including the expression ofthe respiratory apparatus. Transcriptional control oper-ates through a subset of well-defined transcription factorsand transcriptional coactivators. These regulators exerttheir influence on the expression of both nuclear andmitochondrial genes required for the maintenance andbiogenesis of the organelle.

II. THE MITOCHONDRIAL GENETIC SYSTEM:

mtDNA STRUCTURE AND INHERITANCE

It is long established from both genetic and biochem-ical studies that the mitochondrial genetic apparatus hasfeatures that are distinct from that of the nucleocytosolicsystem. Early work pointed to a protein synthetic machin-ery in isolated mitochondria that could synthesize a smallnumber of proteins utilizing mitochondrial ribosomes,rRNA, and tRNA (45). A major advance was the discoveryof mitochondrial DNA. Genetic studies in yeast led to theobservation that certain respiratory mutations displayed acytoplasmic inheritance pattern that resulted from therandom segregation of mtDNA molecules during mitosis.This provided genetic evidence that mitochondria hadtheir own DNA before its existence was demonstratedphysically. This conclusion was supported by physicalevidence for the existence of mitochondrial DNA in yeastand in other organisms (148). Since these early discover-ies, much has been done to define the structure and geneorganization of mtDNA. The first complete mtDNA se-quence was obtained from humans and sequences of mi-tochondrial genomes from many organisms have nowbeen catalogued (227). A striking result from this work isthat a similar complement of genes is conserved in mtDNAsfrom all multicellular organisms.

In contrast to the nuclear genome, where repetitivesequence families, introns, and vast intergenic regionsaccount for all but a few percent of the total DNA, themtDNA of mammals and other vertebrates exhibits strik-ing economy of sequence organization. The vertebratemitochondrial genome exists as a closed circular mole-cule of �16.5 kb whose entire protein coding capacity isdevoted to the synthesis of 13 proteins that function asessential subunits for respiratory complexes I, III, IV, andV (Figs. 1 and 2). The genes specifying complex II areentirely nuclear. The mtDNA also encodes the 22 tRNAs

and 2 ribosomal RNAs necessary for the translation ofthese respiratory subunits within the mitochondrial ma-trix. Mitochondrial genes lack introns and are arrangedend on end with little or no intergenic regions. Somerespiratory protein genes overlap, and the adenine nucle-otides of UAA termination codons are not encoded in themtDNA but rather are supplied by polyadenylation follow-ing RNA processing (152). Protein coding and rRNA genesare interspersed with tRNA genes that are thought todemarcate the cleavage sites of RNA processing. The onlysubstantial noncoding region is the D-loop, named afterthe triple-stranded structure or displacement loop that isformed by association of the nascent heavy (H)-strand inthis region (Fig. 2). The D-loop contains the origin ofheavy (H)-strand DNA replication and is also the site ofbidirectional transcription from opposing heavy (H) andlight (L) strand promoters (45). The designation of thesestrands is assigned based on their relative migration upongradient centrifugation. Interestingly, the structural econ-omy found in vertebrate mtDNA is not found in plants andfungi where the mitochondrial genomes are much largerand contain intergenic regions, introns, and multiple pro-moters and transcriptional units (48, 169).

The fact that mtDNA is a compartmentalized extra-chromosomal element contributes to a mode of inheri-tance that differs from that of nuclear genes. Somaticmammalian cells generally have 103-104 copies of mtDNAwith �2–10 genomes per organelle (179). These genomesreplicate in a relaxed fashion that is independent of thecell cycle that is defined by nuclear DNA replication (46,25). Some mtDNA molecules undergo multiple rounds ofreplication while others do not replicate. This, along withrandom sampling during cell division, allows the segrega-tion of sequence variants during mitosis (198).

In mammals, mtDNA is maternally inherited (for ref-erences, see Refs. 101, 198). In general, paternal mtDNA islost during the first few embryonic cell divisions and doesnot contribute mtDNA to the offspring, although there arereports of the presence of the paternal lineage in somatictissues (192). In one case, recombination between mater-nal and paternal genomes has been documented (109). Inaddition, because mtDNA is a multicopy genome, an in-dividual may harbor more than a single sequence, a con-dition referred to as heteroplasmy. A detrimental se-quence variant may be tolerated in low copy because thedefective gene product(s) it encodes do not reach thethreshold for disrupting cellular function. However, se-quence variants are known to segregate rapidly from het-eroplasmy to homoplasmy in passing from one generationto the next (12). This can result in offspring in which thedetrimental variant predominates, leading to a defectivemitochondrial phenotype. The molecular basis for thisrapid meiotic segregation has been ascribed to a bottle-neck or sampling error in the female germ line. A massiveamplification of mtDNA occurs during oogenesis from



�103 copies in the primary oocyte to �105 copies in themature oocyte (198). Replication of mtDNA is halted inthe mature oocyte, and the existing population of mtDNAmolecules is partitioned to the daughter cells during earlycell divisions until the copy number is diluted to approx-imately that found in somatic cells. Embryonic replicationdoes not resume until the blastocyst stage of development(167).

III. TRANSCRIPTION OF mtDNA

A. Mitochondrial Transcriptional Units

In vertebrates, the D-loop regulatory region containsbidirectional promoters for transcribing H- and L-strandsas well as the H-strand replication origin (OH) (Fig. 2).The H- and L-strand transcriptional units differ from mostnuclear genes in that they are polygenic, specifying mul-tiple RNAs (rRNA, tRNA, or mRNA). L-strand transcrip-tion is initiated at a single promoter (LSP), and RNAsynthesis can traverse the entire mtDNA template produc-

ing a transcript that is processed to 1 mRNA and 8 of the22 tRNAs (Fig. 3A). In addition, the preponderance ofevidence indicates that transcription from LSP is coupledto H-strand replication (25). Mapping studies support theconclusion that transcripts truncated in the vicinity ofCSB II, one of three evolutionarily conserved sequenceblocks (CSB I, II, and III), serve as primers for the initia-tion of H-strand replication (Fig. 3A). Notably, RNA withits 5�-end mapped precisely to the LSP initiation site iscovalently linked to newly synthesized H-strand DNA(38). In addition, the nascent H-strand DNA 5�-endsclosely match the 3�-ends of truncated transcripts initi-ated at PL (37). It is widely accepted that primer RNAs aregenerated by cleavage of L-strand transcripts by mito-chondrial RNA processing (MRP) endonuclease at spe-cific sites that match the in vivo priming sites (Fig. 3A).This ribonucleoprotein endonuclease also participates inthe processing of 5.8S rRNA precursors and contains anucleus-encoded RNA that is essential for catalysis (MRPRNA) (158, 197). A recent alternative model, derived fromin vitro experiments using purified components, suggeststhat termination events associated with CSB II can also

FIG. 3. Schematic representation of mtDNA transcription within the D-loop regulatory region. A: transcription initiation complexes areassembled at bidirectional promoters within the D-loop. They are comprised of mitochondrial RNA polymerase (POLRMT), Tfam, a stimulatoryfactor that unwinds DNA, and one of the two TFB isoforms (TFB1M or TFB2M) that function as dissociable specificity factors that contact boththe polymerase and Tfam. The TFBs are related to rRNA dimethyltransferases and thus may also function in RNA modification or processing.Initiation occurring at LSP can traverse the entire template or be cleaved by MRP endonuclease (RNAse MRP) at discrete RNA 3 DNA transitionsites in the vicinity of the conserved sequence blocks (CSB I, II, III; gray shaded bars) demarcating the origin of heavy strand replication (OH). TheRNAse MRP cleavage sites correspond to the heterogeneous 5�-ends of the newly synthesized H-strand during mtDNA replication. Transcriptsinitiating at HSP2 can also traverse the entire template, whereas those initiating at HSP1 terminate at a transcription terminator (TERM, yellowshaded bar) localized to the tRNAL(UUR) gene to produce the 12S and 16S rRNAs. B: the transcription termination factor mTERF bindssimultaneously both HSP1 and TERM resulting in the looping out of the intervening 12S and 16S rDNA. This is thought to promote efficient recyclingof the transcription complexes resulting in a high rate of rDNA transcription. The mechanism likely contributes to maintaining a high ratio of rRNAto mRNA required for operation of the mitochondrial translation machinery.



account for the primer RNA 3�-ends that mark the majortransition sites between RNA and DNA synthesis (165). Incontrast to the LSP, H-strand transcription is initiatedfrom two closely spaced promoters, HSP1 and HSP2(Figs. 2 and 3A). Transcription initiating at HSP2 results inlong polygenic transcripts that are processed to give 14tRNAs, 12 mRNAs, and the 2 rRNAs. HSP1 on the otherhand appears specialized for the production of ribosomalRNAs (Fig. 3A). As discussed below, termination eventscoupled to initiation at HSP1 provide a mechanism forcontrolling the relative abundance of rRNA to mRNA.

Transcription of mtDNA is completely dependent onnucleus-encoded gene products. In yeast, transcription isdirected by a single subunit RNA polymerase (RPO41p),which shares sequence similarities with the T7 and T3bacteriophage polymerases (104, 134). Although RPO41pcan initiate transcription nonselectively from syntheticDNA templates, the specificity factor Mtf1p (discussedbelow) is required for promoter recognition and selectiveinitiation (104). Interestingly, RPO41p, through its inter-actions with mitochondrial promoters and nucleosidetriphosphates, may act as a sensor of ATP availability.Mitochondrial transcript abundance is correlated withrespiration-coupled ATP synthesis through a requirementfor a high ATP concentration for transcription initiation.This represents a potentially novel and elegant mecha-nism for linking energy production to the transcription ofrespiratory subunits (4).

Although purification of the human polymerase tohomogeneity has been elusive, a human cDNA that en-codes a protein with sequence similarity to yeast mito-chondrial and T3/T7 bacteriophage polymerases wasidentified in database screenings (211). As expected for abona fide mitochondrial polymerase, the 140-kDa proteinproduct, designated POLRMT, is localized to mitochondriavia an NH2-terminal targeting presequence. The utilization ofsingle subunit bacteriophage-related polymerases seemsnearly universal for mitochondrial transcription. A potentiallink between mitochondrial transcription and translationin mammalian cells is implicated by the finding of a directinteraction between POLRMT and mitochondrial ribosomalprotein L12 (MRPL12) (231). MRPL12 stimulates transcrip-tion from both LSP and HSP promoters in an in vitrotranscription system. Moreover, it exists in a complexwith POLRMT in HeLa cell extracts, and its in vivo over-expression can enhance the steady-state levels of mtDNA-encoded transcripts. It is not yet clear whether this occursthrough effects on transcription or RNA stability. Finally,an alternative splice variant of POLRMT that is missingthe NH2-terminal 262 amino acids including the mitochon-drial targeting presequence has been identified (108). Al-though it is currently unclear whether this protein func-tions as an independent nuclear polymerase, it has beenlocalized to the nucleus and linked to the expression of anumber of nuclear genes. Its precise role in nuclear tran-

scription and its potential functional implications for mi-tochondrial biogenesis await further characterization.

B. Tfam

The region of the D-loop that separates the LSP fromthe two HSPs contains conserved nucleotide sequencemotifs that define the core promoter and mediate bidirec-tional transcription (212). In addition, the promotersshare an upstream enhancer that binds Tfam (previouslymtTF-1 and mtTFA), the first well-characterized transcrip-tion factor from vertebrate mitochondria (Fig. 3A). Tfamwas first identified as a high mobility group (HMG)-boxprotein that stimulates transcription through specificbinding to recognition sites upstream from both LSP andHSP (69). Structurally, it consists of two tandemly ar-ranged HMG motifs and a COOH-terminal tail. Tfam re-sembles other HMG proteins, in that it can bend andunwind DNA, properties potentially linked to its ability tostimulate transcription upon binding DNA (159, 70). Tet-rameric binding of Tfam to its recognition site is thoughtto promote bidirectional transcription by facilitating sym-metrical interactions with other transcriptional compo-nents (7). In addition to specific promoter recognition,Tfam binds nonspecifically to apparently random sites onmtDNA (70, 71). This property, along with its abundancein mitochondria, suggests that it plays a role in the stabi-lization and maintenance of the mitochondrial chromo-some. ABF2p, a related HMG-box factor from yeast, re-sembles Tfam and is required for both mtDNA mainte-nance and respiratory competence (54). Expression ofhuman Tfam in ABF2p-deficient yeast cells can rescueboth phenotypes (160). Thus, although highly divergent inprimary structure, the human and yeast proteins are func-tionally interchangeable in supporting mitochondrial re-spiratory function in yeast. However, despite this func-tional complementation, ABF2p lacks an activation do-main and does not stimulate transcription from yeastmitochondrial promoters in vitro (237, 50). The transcrip-tional activation function of Tfam resides in a COOH-terminal activation domain that is required both for tran-scriptional competence and for specific binding to pro-moter recognition sites (50).

As discussed below, a Tfam knockout mouse dis-plays embryonic lethality and a depletion of mtDNA con-firming an essential role for the protein in mtDNA main-tenance in mammals (113). Interestingly, Tfam levels cor-relate well with increased mtDNA in ragged-red musclefibers and decreased mtDNA levels in mtDNA-depletedcells (168). This is consistent with the observation thattransgenic mice overexpressing human Tfam display in-creased mtDNA copy number (59). The mtDNA abun-dance measured in somatic tissues and in embryos isproportional to the amount of Tfam expressed in each,



suggesting that Tfam can function as a limiting determi-nant of mtDNA copy number. Because human Tfam is apoor activator of mouse LSP and HSP, overexpression ofthe human protein results in an increase in mtDNA with-out affecting respiratory capacity, suggesting that copynumber control by Tfam is independent of its transcrip-tional function. In fact, both endogenous Tfam and amutated derivative that lacks the COOH-terminal activa-tion domain are equally competent in maintaining mtDNAcopy number in cultured cells (103). These observationsare reminiscent of Abf2p, which does not direct transcrip-tion but is required for mtDNA maintenance (54). Abf2p isan abundantly expressed component of mitochondrialnucleoids, protein-DNA complexes thought to be a basicsegregating unit of mtDNA (105). Tfam is also expressedat levels high enough to coat mtDNA and has been recov-ered in vertebrate nucleoids in association with otherproteins required for mtDNA genomic integrity and expres-sion (26, 102, 230). Thus both its in vivo and in vitroproperties implicate Tfam as an ideal target for regulatorypathways that control both mtDNA maintenance and tran-scriptional expression.

C. mtTFB

In yeast, RPO41p associates with a 43-kDa specificityfactor MTF1p also known as sc-mtTFB (196). The primarystructure of sc-mtTFB bears some resemblance to pro-karyotic sigma factors (99), but a crystal structure revealssignificant homology to rRNA methyltransferase (191).Although individually neither RPO41p nor MTF1p caninteract with yeast mitochondrial promoters, togetherthey engage in specific promoter recognition (104). Evi-dence for a vertebrate factor that functions in analogousfashion to the yeast sc-mtTFB came from biochemicalstudies in Xenopus laevis (24, 24). The partially purifiedfactor bound DNA nonspecifically and stimulated tran-scription in vitro by purified mitochondrial RNA polymer-ase. Although purification and molecular cloning of thevertebrate factor has been elusive, this impasse has beenresolved by the identification of a human mtTFB cDNA(138). The encoded protein is localized to mitochondriaand stimulates transcription from an L-strand promoterin vitro, properties consistent with it being a functionalhomolog of sc-mtTFB. The human protein exhibits non-sequence-specific DNA binding and requires Tfam to stim-ulate transcription from mitochondrial templates in vitro.Two isoforms of h-mtTFB, termed TFB1M and TFB2M,have been identified independently (65). TFB1M is iden-tical to the original mtTFB and has about 1/10 the tran-scriptional activity of TFB2M. As depicted in Figure 3A,both proteins work together with Tfam and mtRNA poly-merase to direct proper initiation from H- and L-strandpromoters in an in vitro system comprised of purified

recombinant proteins (65). TFB1M has been detected in acomplex with Tfam and POLRMT in transcriptionallycompetent mitochondrial extracts. Moreover, both TFBisoforms make direct contact with the COOH-terminalactivation domain of Tfam (139). These observations sup-port the notion that mtDNA transcription in vertebratesystems is directed by three essential components (Fig.3A). Interestingly, like the yeast factor, both TFBs arerelated to rRNA methyltransferases. TFB1M can bindS-adenosylmethionine and methylate rRNA at a stem-loopstructure that is conserved between bacterial and mito-chondrial rRNAs (195). However, this activity is not re-quired for transcriptional activation by the factor (139). Ithas yet to be determined whether the proteins are bifunc-tional or whether they evolved a single function from anancestral methyltransferase.

Recent experiments suggest that the two TFB iso-forms are not functionally identical. At early times follow-ing serum stimulation of quiescent fibroblasts, mRNA forthe TFB1M isoform is transiently downregulated relativeto that of the TFB2M isoform, suggesting that the latter isfavored in the transition to proliferative growth (73). InDrosophila cultured cells, RNAi knockdown of the Dro-

sophila B2 isoform results in reduced mtDNA transcrip-tion and copy number (136). This contrasts with RNAiknockdown of the B1 isoform, which has no effect onmtDNA transcription or replication but does result inreduced mitochondrial translation (135). It is also notablethat overexpression of either TFB2M or Tfam in thissystem increases mtDNA copy number, whereas overex-pression of TFB1M fails to do so. The stimulatory effect ofTfam is consistent with the observation that overexpres-sion of human Tfam in mice increases mtDNA copy num-ber (59). Thus both gain- and loss-of-function experimentssupport a role for Tfam and TFB2M in mtDNA copynumber control. The inability of TFB1M to function in asimilar capacity in Drosophila cells is surprising in light ofthe ability of the mammalian protein to bind the transcrip-tion activation domain of Tfam and stimulate transcrip-tion in vitro.

D. Transcription Termination and Mitochondrial

Gene Expression

Specific termination events play an important role ingoverning the steady-state levels of mitochondrial tran-scripts. The high rate of rRNA synthesis relative to mRNAhas been ascribed to more frequent initiation of H-strandtranscription at HSP1. As shown in Figure 3A, HSP1 di-rects transcription from a site within the tRNAPhe gene.Most transcripts initiated at HSP1 terminate downstreamfrom the 16S rRNA gene at a strong bidirectional termi-nator localized within the adjacent tRNALeu gene (44).The 28-nucleotide terminator binds mTERF, a 34-kDa pro-



tein that specifies site-specific transcription terminationin fractionated mitochondrial lysates (49). Although thepurified recombinant protein displays the expected bind-ing specificity, it is not sufficient for termination, suggest-ing that an additional component(s) may be required (66).Interestingly, in addition to binding the termination site,mTERF stimulates transcription from HSP1, suggestingthat transcription initiation and termination are coupled(13, 111). Recent experiments demonstrate that this isindeed the case. Multiple lines of in vitro and in vivoevidence establish that mTERF binds simultaneouslyboth the HSP1 initiation site and the terminator resultingin the looping out of intervening rDNA (Fig. 3B). Theseinteractions are proposed to enhance the reinitiation rateat HSP1 resulting in a higher rate of rRNA synthesis. Thismechanism likely accounts for the stimulatory effects ofmTERF on transcription (133). Additional complexity tomTERF function is suggested by the observation of sev-eral mTERF-related genes in vertebrates (129). The pro-tein products all have putative mitochondrial targetingsignals and in one case mitochondrial localization hasbeen confirmed (40). Interestingly, this isoform, desig-nated as mTERFL, differs from mTERF in that it is down-regulated upon serum induction of serum-starved cells.This is reminiscent of the differential expression ofTFB1M and TFB2M in response to serum stimulation (73).It is possible that functionally distinct transcription com-plexes arise from the differential expression of alternativeisoforms of these transcription initiation and terminationfactors.

IV. NUCLEAR CONTROL OF

MITOCHONDRIAL FUNCTION

A. Nuclear Transcription Factors Governing

Respiratory Gene Expression

1. Cytochrome c and the identification of NRF-1

Isolation of cytochrome c genes in yeast and subse-quently in mammalian cells opened the way to the molec-ular analysis of the control of nuclear genes governingmitochondrial respiratory function. In yeast, transcrip-tional regulation of the two cytochrome c isoforms ismediated through upstream enhancer elements withintheir promoters (77). These enhancers bind specific tran-scriptional activators and repressors that direct dramaticchanges in gene expression in response to oxygen andcarbon source availability (169, 242). Although cyto-chromes c are highly conserved at the functional levelbetween yeast and mammals (186), the mammalian cyto-chrome c promoter contains recognition sites for tran-scription factors that bear no obvious relationship tothose identified in yeast (62). In particular, systematic

analysis of the cytochrome c control region revealed apalindromic recognition site for a transcription factordesignated as nuclear respiratory factor 1 (NRF-1) (63)(Fig. 4). Specific NRF-1 binding sites are present in thepromoters of several nuclear genes required for mito-chondrial respiratory function (39, 64). The protein bindsits recognition site as a homodimer through a unique DNAbinding domain and functions as a positive regulator oftranscription (78, 222). This is consistent with the pres-ence of a COOH-terminal transcriptional activation do-main (Fig. 4) comprised of glutamine-containing clustersof hydrophobic amino acid residues (79). NRF-1 exists asa phosphoprotein in proliferating mammalian cells andserine phosphorylation within a concise NH2-terminal do-main (Fig. 4) enhances both its DNA binding (78) andtrans-activation functions (92). Although other mamma-lian NRF-1 isoforms have not been found, NRF-1 is re-lated, through its DNA binding domain, to developmentalregulatory proteins in sea urchins (32) and Drosophila

(53). In addition, chicken (74), zebrafish (20), and mouse(187) homologs of NRF-1 have been characterized.

NRF-1 has now been linked to the expression ofmany genes required for mitochondrial respiratory func-tion including the vast majority of nuclear genes thatencode subunits of the five respiratory complexes (forrecent compilations, see Refs. 106, 183, 184). In addition,considerable evidence supports a potential integrativefunction for NRF-1 in coordinating respiratory subunitexpression with that of the mitochondrial transcriptional

NRF-1

A

DNA binding

B

1 503

100

100

100

80 80 85 90 95 75 95 100

85

YGCGCAYGCGCRRCGCGTRCGCGY

trans-activation

serinephosphorylation

NLS

FIG. 4. Summary of nuclear respiratory factor (NRF)-1 functionaldomains and DNA recognition site. NRF-1 has a central DNA bindingdomain (stippled box) flanked by a nuclear localization signal (horizon-tally hatched box) and a bipartite transcriptional activation domain(vertically hatched boxes). The protein is phosphorylated in exponen-tially growing cells at multiple serine residues within a concise NH2-terminal domain (cross-hatched box). NRF-1 binds a GC-rich palin-drome (shown below the diagram) as a homodimer and makes guaninenucleotide contacts (filled circles) over a single turn of the DNA helix.Numbers below the NRF-1 consensus binding site represent the percent-age of the time the indicated nucleotide is present at that position in 20functional NRF-1 sites. Y, pyrimidine nucleotide; R, purine nucleotide.



machinery. As illustrated in Figure 5, NRF-1 binds andactivates the promoters of Tfam (224) and both TFBisoform genes (73) whose products (as discussed above)are major regulators of mitochondrial transcription.NRF-1 has also been linked to the expression of 5-amin-olevulinate synthase (30) and uroporphyrinogen III syn-thase (2), key enzymes of the heme biosynthetic pathway.The former is the rate-limiting enzyme in the biosynthesisof heme, a cofactor essential to the function of electroncarriers encoded by both genomes (Fig. 5). Moreover,NRF-1 acts on genes whose functions are not restricted tothe bigenomic expression of the respiratory apparatus(Fig. 5). The outer membrane multisubunit receptor com-plex designated as TOMM is required for the import of thethousands of proteins that contribute to diverse mito-chondrial functions (214). TOMM20 is a key receptorsubunit involved in the initial interaction with precursorproteins targeted to mitochondria. NRF-1 activatesTOMM20 gene transcription through a specific promoterrecognition site and is bound to the TOMM20 promoterregion in vivo (23). Similarly, NRF-1 is implicated in theexpression of mouse COX17, a putative cytochrome oxi-

dase assembly factor (206). NRF-1 control over key com-ponents of the protein import and assembly machinery issuggestive of a broader role for the factor in orchestratingevents in mitochondrial biogenesis beyond the transcrip-tional expression of the respiratory chain.

This hypothesis is reinforced by reports that associ-ate elevations in NRF-1 mRNA or DNA binding activitywith generalized effects on mitochondrial biogenesis.Both NRF-1 and its coactivator PGC-1� (see below) areupregulated during the adaptive response of skeletal mus-cle to exercise training (15, 147). A similar effect whichmimics exercise-induced mitochondrial biogenesis oc-curs in cultured myotubules in response to increasedcalcium (154). Treatment of rats with a creatine analog,which induces muscle adaptations resembling those ob-served during exercise, leads to the activation of AMP-activated protein kinase. This coincides with increasedNRF-1 DNA binding activity, cytochrome c content, andmitochondrial density (21). Both NRF-1 and Tfam mRNAsincrease in cells depleted of mtDNA, presumably as aresponse to increased oxidative stress (141). NRF-1 andNRF-2 (see below) along with Tfam are also upregulated

FIG. 5. Diagrammatic summary of the nuclear control of mitochondrial functions by NRF-1 and NRF-2 (GABP). NRFs contribute both directlyand indirectly to the expression of many genes required for the maintenance and function of the mitochondrial respiratory apparatus. NRFs act ongenes encoding cytochrome c, the majority of nuclear subunits of respiratory complexes I–V, and the rate-limiting heme biosynthetic enzyme5-aminolevulinate synthase. In addition, NRFs promote the expression of key components of the mitochondrial transcription and translationmachinery that are necessary for the production of respiratory subunits encoded by mtDNA. These include Tfam, TFB1M, and TFB2M as well asa number of mitochondrial ribosomal proteins and tRNA synthetases. Recent findings suggest that NRFs are also involved in the expression of keycomponents of the protein import and assembly machinery.



in response to lipopolysaccharide-induced oxidative dam-age to mitochondria, presumably to enhance mtDNA lev-els and OXPHOS activity (203). Exogenous oxidants re-store NRF-1 and normal cell growth to �o hepatoma cellswhere reduced oxidant levels are associated with loss ofNRF-1 and growth delay. Interestingly, NRF-1 occupancyof the Tfam promoter increases under pro-oxidant condi-tions, and the control of Tfam expression by NRF-1 isblocked by inhibition of NRF-1 phosphorylation by Akt(166). This suggests that redox regulation of NRF-1 targetgenes such as Tfam is modulated by redox-regulatedphosphorylation events. Finally, both NRF-1 (DNA bind-ing activity and mRNA) and Tfam (protein and mRNA) areupregulated in the skeletal muscle of aged human sub-jects (120). This may be a compensatory response toage-related reductions in energy metabolism. These asso-ciations are consistent with a general integrative role forNRF-1 in nucleomitochondrial interactions.

It is important to note that NRF-1 targets are notrestricted to genes involved in mitochondrial function. Aninitial search for NRF-1 binding sites in mammalian pro-moters revealed a number of primate and rodent geneswhose functions are not linked directly to mitochondrialbiogenesis (222). Among these are genes encoding meta-bolic enzymes, components of signaling pathways, andgene products necessary for chromosome maintenanceand nucleic acid metabolism among others. Moreover,NRF-1 is among seven identified transcription factorswhose recognition sites are most frequently found in theproximal promoters of ubiquitously expressed genes (72).Recently, chromatin immunoprecipitations (ChIP) coupledwith microarray assay (ChIP-on-chip) were used to identifyhuman promoters that are bound by NRF-1 in vivo (33).Survey of �13,000 human promoters by ChIP-on-chipanalysis (174) identified 691 genes whose promoters areoccupied by NRF-1 in living cells (33). As expected, amajority of these genes are involved in mitochondrialbiogenesis and metabolism, including many that had notbeen previously identified. These include a collection ofmitochondrial ribosomal protein and tRNA synthetasegenes. Notably, a significant subset of the NRF-1 targetgenes was also bound by the growth regulatory transcrip-tion factor E2F, suggesting that NRF-1 participates in theregulation of a subset of E2F-responsive genes. This sub-set was enriched in genes required for DNA replication,mitosis, and cytokinesis. Although NRF-1 was bound to itstarget promoters under conditions of transcriptional re-pression, an NRF-1 siRNA reduced expression of severalE2F target genes along with Tfam and cytochrome c,confirming that it functions as a transcriptional activator.NRF-1 exists in the dephosphorylated state in serum-starved cells and is phosphorylated upon serum addition(78). Since phosphorylation enhances NRF-1 transcrip-tional activity (92), it is possible that phosphorylationcontrols the functional state of the DNA bound factor.

This along with derepression by release of E2F factorsfrom a subset of NRF-1 target genes may help promotecell proliferation. Notably, NRF-1 has also been impli-cated in the transcriptional repression of E2F1 (58) andthe activation of E2F6 (107). These results are consistentwith a role for NRF-1 in regulating cell cycle progressionand may explain the early embryonic lethality of NRF-1knockout embryos (96).

2. NRF-2 (GABP) an activator of cytochrome

oxidase expression

A second nuclear factor designated as NRF-2 wasidentified based on its specific binding to essential cis-acting elements in the cytochrome oxidase subunit IV(COXIV) promoter (181, 182). NRF-2 binding sites con-tained of the GGAA core motif that is characteristic of theETS-domain family of transcription factors. Direct re-peats of this sequence motif are interspersed with multi-ple transcription initiation sites within the mouse COXIV

promoter (223, 36). Human NRF-2 was purified to homo-geneity from HeLa cell nuclear extracts and is comprisedof five subunits (Fig. 6). These include a DNA-binding �subunit and four others (�1, �2, �1, and �2,) that complexwith � but alone do not bind DNA. Molecular cloning ofthe five NRF-2 subunits (80) revealed that NRF-2 is thehuman homolog of mouse GABP (112). The two addi-tional human subunits, �1 and �1, are minor splice vari-ants of GABP subunits �1 and �2 (80). The GABP �1

subunit, corresponding to NRF-2 �1 and �2 (80), has adimerization domain that facilitates cooperative bindingof a heterotetrameric complex to tandem binding sites(210). In solution, GABP exists as a heterodimer but isinduced to form the heterotetramer �2�2 by DNA contain-ing two or more binding sites (41). The crystal structureof the heterotetramer bound to DNA has been determined(19). All of the non-DNA-binding subunits contain a tran-scriptional activation domain (Fig. 6). This domain resem-bles that found in NRF-1 and has been localized to aregion upstream from the homodimerization domain (79).

As observed for COXIV, COXVb transcription is alsoactivated by NRF-2 through directly repeated recognitionsites within the proximal promoter, suggesting that NRF-2is a general activator of cytochrome oxidase subunit geneexpression (225). Several lines of evidence point to adirect role for NRF-2 in the expression of all 10 nucleus-encoded cytochrome oxidase subunits. NRF-2 has beenassociated in vivo with multiple COX subunit promotersby chromatin immunoprecipitation (157). Moreover, ex-pression of a dominant negative NRF-2 allele reducedCOX expression, and an siRNA directed against NRF-2�reduced expression of all 10 nucleus-encoded COX sub-units (156). These findings are consistent with an impor-tant regulatory function for NRF-2 in cytochrome oxidaseexpression.



A strong correlation has been observed betweenNRF-2� (GABP�) and cytochrome oxidase expression inthe visual cortex (149). NRF-2� and cytochrome oxidasewere associated with metabolic state and neuronal activ-ity in mature neurons, suggesting that NRF-2� is impor-tant in maintaining neuronal function. Moreover, the cor-relation was extended to NRF-2� mRNA, indicating thatNRF-2� may be regulated at the transcriptional level byneuronal activity (82). Both NRF-2� and -� subunits areenriched in the nucleus in response to neuronal stimula-tion (241), and both subunits are specifically translocatedto the nucleus in response to neuronal depolarization(238). In fact, the neuronal factor heregulin directs thethreonine phosphorylation of GABP�, stimulating both itstranscriptional activity and its mobilization to the nucleuswhere it activates transcription of acetylcholine receptorsubunits (188, 204).

In addition to the COX promoters, functional NRF-2sites have been identified in a number of other genesrelated to respiratory chain expression (for a recent com-pilation, see Refs. 106, 183). These include genes for Tfam(113, 173) as well as TFB1M and TFB2M (138, 65) (Fig. 7).Three of the four human succinate dehydrogenase (com-plex II) subunit genes also have both NRF-1 and NRF-2sites in their promoters (14, 60, 94). In many cases, NRF-1sites are also present in NRF-2-dependent promoters, butthis is not a general rule. For example, several COXpromoters and the rodent Tfam (42) and TFB (172) pro-moters do not have obvious NRF-1 consensus sites. Thiscontrasts with the human Tfam (224) and TFB (73) pro-moters, which rely on both NRF-1 and NRF-2 for their

activities (Fig. 7). Both NRF-1 and NRF-2� occupy bothTFB1M and -2M promoters in vivo as determined by chro-matin immunoprecipitation (73). Thus, like NRF-1, NRF-2participates in coordinating the expression of essentialrespiratory chain proteins with key components of themitochondrial transcription machinery. Such a mecha-nism may serve to ensure the coordinate bigenomic ex-pression of respiratory subunits.

3. ERR�

The estrogen-related receptor ERR� has been linkedto the regulation of oxidative metabolism. ERR�, one of afamily of orphan nuclear receptors including ERR� andERR�, resembles the estrogen receptor but does not bindestrogen or other ligands. ERR� levels are high in oxida-tive tissues such as kidney, heart, and brown fat, and itacts as a regulator of �-oxidation via its control of themedium-chain acyl-coenzyme A dehydrogenase (MCAD)promoter (97). The �- and �-isoforms increase postnatallyin the heart along with enzymes promoting mitochondrialfatty acid uptake and oxidation (98). ERR� knockoutmice have reduced fat mass and are resistant to diet-induced obesity, suggesting defects in lipid metabolism(132). Recent studies have also implicated ERR� in PGC-1�-induced mitochondrial biogenesis (143, 190). A computa-tional approach coupled with PGC-1�-induced RNA ex-pression profiles identified both ERR� and GABP� in adouble positive-feedback loop with PGC-1� in directingexpression of a subset of mitochondria-related genes(143). An inhibitor of ERR� attenuated PGC-1�-mediated

homo-dimerization

trans-activation

VNLTGLVSSENSSKATDETG

DNA binding

ETS(51.4 kDa)

hetero-dimerization

1(42.5 kDa)

2(41.3 kDa)

1 (38.1kDa)

2 (36.9 kDa)

FIG. 6. Summary of NRF-2 (GABP) functional domains. NRF-2, the human homolog of mouse GABP, is comprised of 5 subunits. The �-subunitcontains the ETS domain (stippled box) and can bind its DNA recognition site in the absence of the other subunits. The �- and �-subunits containthe trans-activation domain (bracketed) and can complex with � through interactions between the ankrin repeats (arrows) and the ETS domain.The �-subunits differ from the �-subunits in that they have an NH2-terminal homodimerization domain (cross-hatched box). This allows theformation of a heterotetrameric complex (�2�2) that binds tandem NRF-2 recognition sites in respiratory gene promoters with high affinity. The��-complexes lacking the homodimerization domain are transcriptionally competent but bind DNA with much weaker affinity. The �1- and�1-subunits are alternative splice variants of �2 and �2 that contain an additional 20-amino acid insertion (solid box with sequence below) ofunknown function.



target gene expression and total cellular respiration. Inthis scheme, NRF-1 was downstream of the ERR�-di-rected changes in expression. It should be noted thatPGC-1� was abundantly overexpressed from an adenovi-rus vector in many of the experiments utilized to validatethis model. This generally results in massive changes inmitochondrial biogenesis that are not observed in geneticknockouts of PGC-1� or ERR� (discussed below). Never-theless, ERR� does mediate the effects of virally ex-pressed PGC-1� on mitochondria-related genes such asTfam and ATP synthase �-subunit (190). It is also ofinterest to note that PGC-1� functions through NRF-2(GABP) in the control of a program of neuromusculargene expression in skeletal muscle (85). Thus these fac-tors likely coordinate broad adaptive changes beyondtheir initially observed functions in respiratory chain ex-pression.

4. PPARs and fatty acid oxidation

A number of mitochondrial oxidative pathways feedinto the respiratory apparatus. Important among these isthe pathway for fatty acid oxidation, which consists of a

series of enzymes within the mitochondrial matrix thatoxidize fatty acids to acetyl CoA. To date, the transcrip-tional expression of the genes encoding this pathway hasnot been linked to the transcription factors that governthe expression of the respiratory chain. Rather, PPAR�and -�, members of the nuclear hormone receptor super-family, have been implicated as transcriptional regulatorsof fatty acid oxidation enzymes in tissues with high ratesof fat oxidation including heart, kidney, and skeletal mus-cle (17, 81, 97). Conversely, the PPARs have not beenassociated with the expression of the respiratory appara-tus. This suggests that higher order integration of diversetranscription factors may be necessary to coordinate thelarge number of genes required for mitochondrial biogen-esis and function. It is notable that ERR� regulates theexpression, during adipocyte differentiation, of MCAD, akey enzyme in fatty acid oxidation (219). Thus ERR�,through its control of both respiratory and fatty acidoxidation genes, may participate in coordinating the ex-pression of the two pathways in certain physiologicalcontexts.

5. Other nuclear factors

Additional nuclear factors have been implicated in theexpression of respiratory genes. The cytochrome c promotercontains cis-elements that recognize transcription factorsof the ATF/CREB family (63, 75). These elements bindCREB both in vitro and in vivo, and phosphorylation ofCREB and NRF-1 is linked to the serum inductionof cytochrome c in quiescent fibroblasts (92, 220). Al-though CREB sites are common to cAMP- and growth-activated genes (137), these elements are not a generalfeature of nuclear respiratory promoters. Cytochrome c

appears to be limiting for mitochondrial respiration earlyin the program of serum-induced entry to the cell cycle(92). CREB activation of cytochrome c expression likelycontributes to its relatively rapid induction in response toserum-stimulated growth. This may be a mechanism forramping up cellular respiration in preparation for cellgrowth and division.

Maximal cytochrome c promoter activity also de-pends on synergy between functional recognition sites forthe general transcription factor Sp1 (63). Sp1 is also in-volved in the activation and/or repression of cytochromec1 (125) and adenine nucleotide translocase 2 genes (124),both of which lack NRF sites (240). Muscle-specific COXsubunit genes, in contrast to their ubiquitously expressedisoforms (193), depend on MEF-2 and/or E-box consensuselements for their tissue-specific expression (228). Theinitiator element transcription factor YY1 has been impli-cated in both positive and negative control of cytochromeoxidase subunit gene expression (18, 194). Interestingly,YY1 knockout mice resemble the NRF-1 and NRF-2� knock-outs in that they exhibit peri-implantation lethality (56).

FIG. 7. Arrangement of NRF-1 and NRF-2 recognition sites in hu-man promoters specifying key components of the mitochondrial tran-scription and replication machinery. Shown are linear representations ofthe 5�-flanking regions proximal to the putative transcription initiationsites (arrows) of human genes encoding Tfam, TFB1M, TFB2M, andPOL�2. Each promoter region bears a strong resemblance to a numberof respiratory subunit promoters in that they contain a single NRF-1 site(blue bars) and two or more tandem NRF-2 sites (red bars). In addition,each promoter has at least one Sp1 recognition site (gray bars).



More recently, c-Myc has been linked to respiratorygene expression and mitochondrial biogenesis. C-Myc caninduce cytochrome c and other NRF-1 target genes bybinding noncanonical Myc/MAX binding sites containedwithin certain NRF-1 sites and can sensitize cells to apop-tosis by dysregulating NRF-1 target genes (146). A domi-nant negative NRF-1 allele can block this c-Myc-inducedapoptosis without affecting c-Myc-induced cell prolifera-tion. In addition, Myc null fibroblasts were found deficientin mitochondrial content, and c-Myc expression in thesecells partially restored mitochondrial mass (121). Thisalong with the observation that many mitochondria-re-lated genes including Tfam are c-Myc targets led to theconclusion that c-Myc can regulate mitochondrial biogen-esis. One caveat to this conclusion is that c-Myc is esti-mated to have an enormous number of genome bindingsites that even exceeds the number of Myc molecules inproliferating cells. Thus it is thought that only a minorityof genes bound by Myc/MAX are actually regulated byc-Myc (1). In addition, because of the extremely largenumber of Myc target genes, it is possible that the ob-served effects on mitochondrial biogenesis result fromindirect effects of Myc on cell growth and metabolism.

6. Dual function factors

There are several reports suggesting that certain nu-clear transcription factors or their derivatives have a dualfunction in directing mitochondrial gene expression.Binding sites for ligand-dependent trans-activators havebeen observed in mtDNA (51, 234). In the best-studiedexample, a 43-kDa protein, closely related to the c-ErbA�1 thyroid hormone receptor, has been localized to themitochondrial matrix and observed to bind the mitochon-drial D-loop in vitro (234). This protein appears to be anNH2-terminally truncated variant of the nuclear �1 thyroidhormone receptor. The overexpressed variant gives a par-ticulate cytosolic immunofluorescence and stimulates mi-tochondrial respiration as measured by rhodamine stain-ing and cytochrome oxidase activity. The functional sig-nificance of this finding is supported by the observation ofa direct effect of thyroid hormone on mitochondrial RNAsynthesis (61). Similarly, the rat liver glucocorticoid re-ceptor (GR) was found to translocate to the mitochondriaupon treatment of animals with dexamethasone (52). Sixglucocorticoid response elements (GREs) were found inmtDNA and were shown to bind GR in vitro (51). Two ofthe six elements are present in the D-loop, but the otherfour are within the COXI and COXIII genes, far removedfrom the in vivo sites of mtDNA transcription initiation.

I�B-� and NF�B have also been localized to mito-chondria (29, 47). In one case, they were found in theintermembrane space associated with the adenine nucle-otide translocator and were released upon induction ofapoptosis (29). In the other, they were localized to the

matrix where they were thought to act as negative regu-lators of mitochondrial mRNA expression (47). Theseseemingly contradictory results illustrate the difficulty inassigning function based on subcellular localization. It isimportant to note that the transcriptional activity andspecificity of these shared nuclear factors using mtRNApolymerase and a defined mtDNA template has not beendemonstrated. Thus it remains inconclusive as to whetherthese nuclear factors are true regulators of mitochondrialtranscription.

B. Nuclear Coactivators in

Mitochondrial Biogenesis

Most of the evidence supports a model whereby arelatively small number of nuclear transcription factorsserve to coordinate the expression of nuclear and mito-chondrial respiratory proteins. Of these, the NRFs, Sp1,and ERR� have been consistently associated with themajority of genes required for respiratory chain expres-sion. These factors exert direct control over transcriptionof the nucleus-encoded respiratory subunits and act indi-rectly on the mitochondrial subunits via their regulationof mitochondrial transcription factors. In addition, othermitochondrial oxidative pathways are controlled by unre-lated factors as exemplified by PPAR� and the fatty acidoxidation pathway (81). This raises the question of howdiverse transcription factors are integrated into a programof mitochondrial biogenesis. The discovery of the PGC-1�family of transcriptional coactivators has provided amechanistic framework for understanding how nuclearregulatory pathways are coupled to the biogenesis ofmitochondria.

1. PGC-1�

PGC-1�, the founding member of a family of tran-scriptional coactivators, was identified in a differentiatedbrown fat cell line on the basis of its interaction withPPAR�, an important regulator of adipocyte differentia-tion (171). Although it apparently lacks histone-modifyingactivities itself, it interacts, through a potent NH2-terminaltranscriptional activation domain, with cofactors contain-ing intrinsic chromatin remodeling activities (Fig. 8). Inaddition to PPAR�, PGC-1� binds several nuclear hor-mone receptors and trans-activates the PPAR�- and thy-roid receptor �-dependent expression of the UCP-1 pro-moter. Nuclear hormone receptor coactivator signaturemotifs (LXXLL) adjacent to the activation domain areessential for coactivation through certain nuclear recep-tors, including PPAR� and thyroid receptor � (Fig. 8).Coupling of transcription and RNA processing by PGC-1�occurs through COOH-terminal arginine/serine rich (R/S)and RNA recognition motifs reminiscent of those found inRNA splicing factors (142). The dramatic induction of



PGC-1� mRNA in brown fat upon cold exposure supportsits involvement in thermogenic regulation (126, 170). Animportant part of the thermogenic program is the induc-tion of mitochondrial biogenesis, and PGC-1� is a potentinducer of this process. Ectopic overexpression of thecoactivator in cultured myoblasts and other cells inducesrespiratory subunit mRNAs and increases COXIV and cy-tochrome c protein levels as well as the steady-state levelof mtDNA (236).

As illustrated in Figure 9, NRF-1 has been identifiedas an important target for the induction of mitochondrialbiogenesis by PGC-1�. The coactivator binds NRF-1 andcan trans-activate NRF-1 target genes involved in mito-chondrial respiration. Moreover, a dominant negative al-lele of NRF-1 blocks the effects of PGC-1� on mitochon-drial biogenesis providing in vivo evidence for a NRF-1-dependent pathway (236). PGC-1� induction of bothNRF-1 and NRF-2 transcripts may also contribute to thebiogenic program. As depicted in Figure 9, PGC-1� maylink nuclear regulatory events to the mitochondrial tran-scriptional machinery through its transcriptional activa-tion of Tfam, TFB1M, and TFB2M expression. As withrespiratory subunit genes, the coactivator targets theNRF-1 and NRF-2 recognition sites within Tfam and TFBpromoters leading to increased mRNA expression (73).ERR� has also been associated with PGC-1�-induced mi-tochondrial biogenesis (143, 190). ERR� and GABP�(NRF-2�) recognition sites are conserved in the promot-ers of a number of oxidative phosphorylation genes, in-cluding cytochrome c and �-ATP synthase, and PGC-1�can drive expression through these sites (143, 190) (Fig. 9).Interestingly, computer modeling indicates that PGC-1� in-

duction of ERR� and GABP� (NRF-2�) is upstream fromNRF-1 in the program of respiratory gene expression. ThePGC-1� induction of mitochondrial biogenesis has beenconfirmed in transgenic mice. PGC-1� expression is ele-vated in the postnatal mouse heart, and cardiac-specificoverexpression of PGC-1� leads to massive increases inmitochondrial content in cardiac myocytes (116). Thisoverexpression is associated with edema and dilated car-diomyopathy and is likely unrelated to normal PGC-1�function in the heart.

In addition to its effects on the respiratory chain andmitochondrial transcription, PGC-1� promotes mitochon-drial oxidative functions by inducing the expression ofgenes of the mitochondrial fatty acid oxidation and hemebiosynthetic pathways (Fig. 9). As we have seen, PPAR�directs the transcriptional expression of fatty acid oxida-tion enzymes and is enriched in brown fat and othertissues with high oxidative energy demands. Ligand-de-pendent binding of PPAR� through the LXXLL motif ofPGC-1� is associated with the trans-activation of PPAR�-dependent promoters (97, 218). In addition, overexpres-sion of PGC-1� induces MCAD gene expression throughits direct interaction with ERR� (98). PGC-1� can alsoutilize both NRF-1 and FOXO1, a forkhead box familymember, to induce transcription of �-aminolevulinate syn-thase, the first and rate-limiting enzyme of the heme bio-synthetic pathway (86). Thus, as summarized in Figure 9,PGC-1� has the potential to integrate the activities of adiverse collection of transcription factors that have beenimplicated in the expression and function of the mito-chondrial oxidative machinery.

Physiological expression of PGC-1� at the transcrip-tional level can be modulated through cAMP-dependentsignaling. In brown adipocytes, sympathetic productionof norepinephrine acts predominantly through �3-adren-ergic receptors (34). These receptors are coupled to ade-nyl cyclase via the Gs subtype of G proteins. The thermo-genic signaling cascade results in enhanced cyclase activ-ity and the increased production of cAMP. As shown inFigure 9, elevated cAMP levels lead to activation of CREBthrough its phosphorylation by protein kinase A (PKA).CREB or related CREB family members, such as activat-ing transcription factor 2 (ATF-2), mediate both directand indirect effects on the thermogenic program includ-ing the induction of PGC-1� and UCP1 (34, 35). ThePGC-1� promoter has a potent cAMP response element(CRE) that serves as a target for CREB-mediated tran-scriptional activation (93). In addition to its role in thethermogenic program, the coactivator is also markedlyinduced in mouse liver in response to fasting where itactivates the genes encoding gluconeogenic enzymes(239). The induction of gluconeogenesis by PGC-1� infasted liver also involves the activation of CREB by serinephosphorylation in response to the catecholamine andglucagon-mediated elevation of cAMP. CREB induces the

FIG. 8. Summary of PGC-1� functional domains. Depicted is a linearrepresentation of PGC-1� showing key functional domains. The NH2-terminal activation domain (vertical hatched box) is the site of interac-tion with histone-modifying cofactors represented by SRC-1 and CBP/p300. The approximate positions for interaction with transcription fac-tors including nuclear hormone receptors, PPAR� and NRF-1, andMEF2C are indicated above and below the map. Also illustrated is thecentral proline-rich region (cross-hatched box) and the COOH-terminaldomain comprised of the arginine/serine-rich R/S domain (open box)and the RNA recognition motif (horizontal hatched box) proposed tointeract with RNA processing factors. The region containing p38 MAPKphosphorylation sites is bracketed below.



transcriptional expression of PGC-1�, which in turn servesas a coactivator of gluconeogenic gene expression (93).

The cAMP-dependent pathway is one of several thatdirect the induction of PGC-1� in a number of tissues(126). PGC-1� along with Tfam and NRF-1 are induced viacGMP-dependent signaling resulting from elevated levelsof nitric oxide (NO) (150) (Fig. 9). The NO induction ofPGC-1� is correlated with increased mitochondrial bio-genesis in several cell lines. The induced mitochondrialmass is accompanied by increased oxidative phosphory-lation-coupled respiration consistent with an increase infunctional mitochondria (151). These results were ob-tained by pharmacological increases in either NO orcGMP, raising the question of whether physiological fluc-tuations in endogenous NO can regulate mitochondrialcontent (114). Interestingly, tissue mitochondria in micewith a homozygous disruption of the gene encoding en-dothelial NO synthase (eNOS�/�) were somewhat smallerand less densely packed than in wild-type mice (150, 151).These changes were accompanied by reductions in en-ergy expenditure and mRNA levels for PGC-1�, Tfam, andNRF-1, arguing that basal mitochondrial content is af-fected by the loss of eNOS-generated NO. Establishment

of an NO-dependent pathway of mitochondrial biogenesisawaits confirmation of these findings.

A number of reports link the expression of PGC-1� toexercise-induced mitochondrial biogenesis in skeletalmuscle (3, 15, 76, 154, 178, 207–209, 233). In each case,PGC-1� mRNA and/or protein increase as an adaptiveresponse to endurance exercise of varying intensity andduration. These findings are satisfying in the context ofthe long-standing relationship between endurance exer-cise and enhanced mitochondrial respiratory chain ex-pression and function (95). Both NRF-1 and NRF-2 DNAbinding activities were elevated along with the expressionof several NRF target genes in rat skeletal muscle follow-ing an exercise regimen (15). Similar changes in the PGC-1�-NRF pathway were mimicked in cultured myotubulesin response to increased calcium levels, suggesting thatPGC-1� signaling contributes to adaptive changes in mi-tochondrial biogenesis in skeletal muscle cells (154). Ex-ercise and other neuromuscular activity may also lead tothe activation of the p38 mitogen-activated protein kinase(MAPK) pathway resulting in PGC-1� induction throughATF2 and MEF2 (3). There is evidence that depletion ofATP during exercise leads to elevated AMP/ATP ratios,

FIG. 9. Illustration summarizing PGC-1�-mediated pathways governing mitochondrial biogenesis and function. Depicted in the nucleus (shadedsphere) are the key transcription factors (NRF-1, NRF-2, ERR�, PPAR�, and MEF-2) that are PGC-1� targets and act on nuclear genes governingthe indicated mitochondrial functions. Some of the physiological effector pathways mediating changes in the transcriptional expression or functionof PGC-1� are also shown. The CREB activation of PGC-1� gene transcription in response to cold (thermogenesis), fasting (gluconeogenesis), andexercise has been well documented. The physiological mechanisms of PGC-1� induction by nitric oxide are not established but may involve theproduction of endogenous nitric oxide by eNOS. A potential pathway of retrograde signaling through calcium is also included.



which enhances AMP-activated protein kinase (AMPK)activity (Fig. 9). In addition, increased calcium activatesCa2�/calmodulin-dependent protein kinase (CaMK), andthis along with AMPK results in the induction of PGC-1�and activation of NRFs during energy deprivation (153,243). In fact, as seen in other instances of PGC-1� induc-tion, CREB phosphorylation, in this case by CaMK, directsCREB-activated PGC-1� transcription (Fig. 9). This path-way is also driven by calcineurin A through myocyteenhancer factor 2 (MEF2), which is a potent trans-acti-vator of PGC-1� transcription in muscle (87). This samefactor has been linked to the expression of muscle-spe-cific COX subunits (228). Although it is not clear weatherthe NRFs are direct targets of these kinases, the associ-ated increase in NRF-1 DNA binding activity coincideswith elevated expression of cytochrome c and 5-aminole-vulinate synthase and enhanced mitochondrial density(21). It should be noted that PGC-1� mRNA is induced tosimilar levels in wild-type and in both AMPK �1 and �2knockout mice, arguing that these isoforms of AMPK arenot essential to exercise-induced gene expression byPGC-1� (100). This study was based on patterns of mRNAexpression alone, and thus a definitive conclusion awaitsa systematic analysis of mitochondrial biogenesis in thesemice.

Finally, a recent study shows that NRF-1 and NRF-2occupancy of cytochrome c and cytochrome oxidase sub-unit IV promoters, respectively, occurs prior to the exer-cise-induced increase PGC-1� protein levels (233). Exer-cise also leads to the activation of ATF-2 through p38mitogen-activated protein kinase (p38 MAPK). This tran-scription factor, a member of the CREB/ATF family, bindsthe PGC-1� promoter and induces its transcription, per-haps as a secondary phase of the adaptive response. It isunclear whether direct phosphorylation of PGC-1� by p38MAPK is required in this model.

2. PGC-1�

Two mammalian relatives of PGC-1� have been char-acterized. PGC-1�, although somewhat larger than PGC-1�, shares sequence similarity with PGC-1� along its en-tire length (110, 127). Like PGC-1�, PGC-1� has an NH2-terminal activation domain, LXXLL coactivator signatures,and a COOH-terminal RNA recognition motif. AlthoughPGC-1� lacks an R/S domain, it is not clear whether itdiffers from PGC-1� in its ability to couple transcriptionand RNA slicing. Steady-state tissue expression of PGC-1�mRNA parallels that of PGC-1� with the highest levels inbrown fat, heart, and skeletal muscle, tissues high inmitochondrial content and oxidative energy production.However, PGC-1� differs from PGC-1� in that it is notinduced in brown fat upon cold exposure, and it is a poorinducer of gluconeogenic gene expression in hepatocytesand liver (140). This most likely results from the absence

of an interaction between hepatic nuclear receptor 4�(HNF4�) and forkhead transcription factor 01 (FOXO1),which mediate the expression of gluconeogenic genes.PGC-1� also exhibits a marked preference for promotingthe ligand-dependent activity of ER� while having a min-imal effect on ER� (110). These examples illustrate thatdifferences in transcription factor interactions mediatedifferences in functional specificity between members ofthe PGC-1 family.

Despite their differential utilization of nuclear hor-mone receptors, PGC-1� binds NRF-1 and is a potentcoactivator of NRF-1 target genes leading to increasedmitochondrial gene expression (127). Forced expressionof PGC-1� in cultured muscle cells results in increasedmitochondrial biogenesis and oxygen consumption, al-though PGC-1� has been associated with higher protonleak rates than PGC-1� (202). Thus, although PGC-1� and -�are functionally distinct, they both utilize NRF-1 and othertranscription factors to induce mitochondrial biogenesis.This has been confirmed in transgenic mice in whichPGC-1� is abundantly overexpressed in skeletal muscle(10). The coactivator promotes massive increases in mi-tochondrial content that are accompanied by marked el-evations in the expression of respiratory mRNAs andproteins derived from the expression of both nuclear andmitochondrial genes. This mitochondrial biogenesis is as-sociated with the induction of type IIX oxidative musclefibers and increased capacity for aerobic exercise. Thesechanges are presumed to result from activation of a spe-cific program of gene expression by PGC-1� through tran-scription factor targets such as MEF2, ERR�, and PPAR�among others. Notably, the induction of type IIX fiberswas absent in the soleus muscle despite high-level expres-sion of PGC-1� in this tissue, suggesting that abundantexpression of the coactivator alone is not sufficient. It willbe important to confirm the regulation of muscle fibertype by PGC-1� in loss of function experiments.

3. PRC

A database search for sequence similarities to PGC-1�identified the partial sequence of a large cDNA with aCOOH-terminal RS domain and RNA recognition motif(6). Molecular cloning of the full-length cDNA revealedadditional sequence similarities with PGC-1� includingan acidic NH2-terminal region, an LXXLL signature fornuclear receptor coactivators, and a central proline-rich region (Fig. 10). Although significant sequence sim-ilarity with PGC-1� is confined to these distinct regions,their spatial conservation is highly suggestive of relatedfunction. Thus the protein encoded by this cDNA wasdesignated as PGC-1�-related coactivator (PRC) (6).

PRC resembles both PGC-1� and -� in that it bindsNRF-1 both in vitro and in vivo and can utilize NRF-1 forthe trans-activation of NRF-1 target genes (6). PRC has a



potent NH2-terminal activation domain that is highly sim-ilar to that of PGC-1� and is required for NRF-1-depen-dent trans-activation. These properties along with its nu-clear localization and in vivo interaction with NRF-1 attestto a transcriptional function for PRC. As with PGC-1�,binding of NRF-1 occurs through the NRF-1 DNA bindingdomain. In addition to cytochrome c and 5-ALAS promot-ers, PRC can trans-activate the human TFB1M andTFB2M promoters in a manner indistinguishable fromPGC-1� (73). The NRF binding sites within the proximalpromoters of these genes serve as targets for trans-acti-vation by both coactivators.

Despite these similarities, PRC mRNA is not enrichedin brown versus white fat and is only slightly elevated inbrown fat upon cold exposure, arguing against a majorrole for PRC in adaptive thermogenesis (6). In addition,PRC expression is not particularly enriched in highlyoxidative tissues with abundant mitochondria. However,as illustrated in Figure 11, PRC is rapidly induced uponserum treatment of quiescent fibroblasts and is expressedmore abundantly in proliferating cells compared withgrowth-arrested cells. Moreover, serum induction of PRCis accompanied by a pattern of gene expression that isqualitatively similar to that observed in response to ele-vated PGC-1� (73). This includes increased expression ofTfam, TFB1M and -2M mRNAs, as well as those for bothnuclear and mitochondrial respiratory subunits. It is ofinterest in this context that PRC, NRF-1, and Tfam aremarkedly upregulated in thyroid oncocytomas in conjunc-tion with increases in cytochrome oxidase activity andmtDNA content (180). These thyroid tumors are character-ized by dense mitochondrial accumulation and are appar-ently devoid of PGC-1�. Both PRC and PGC-1� mRNAs arerapidly induced in human skeletal muscle in response toendurance exercise (178). PRC mRNA expression is highinitially and persists for �4 h but then modulates to abasal level at 24 h after exercise. Thus PRC may directearly adaptive changes in gene expression in response toexercise.

Serum induction of PRC has the characteristics ofimmediate early genes, those genes whose expression isinduced by serum growth factors in the absence of denovo protein synthesis (91). Their induction is among thevery earliest events in the genetic program of cell growth.

FIG. 10. Diagram comparing sequence motifs conserved betweenPGC-1-related coactivator (PRC) and PGC-1�. Dashed lines connectsequence similarities between the activation domains (verticallyhatched box), the proline-rich region (cross-hatched box), the R/Sdomain (open box), and the RNA recognition motif (horizontallyhatched box).

FIG. 11. Diagram summarizing the putative role of PRC during the transition from quiescence to proliferative growth (G03 G1 transition). Cellsthat exit the cell cycle either by contact inhibition or serum withdrawal (G0) have reduced levels of PRC mRNA and protein expression thatcoincides with reduced occupancy of the NRF-1 and CREB-dependent cytochrome c promoter in vivo. G0 cells stimulated to grow by serum exhibita rapid and cycloheximide-independent induction of PRC mRNA during the transition from G0 to G1. Serum stimulation leads to increased PRCprotein expression and increased PRC occupancy of the cytochrome c promoter. This is accompanied by the rapid mitogenic phosphorylation ofCREB by the pp90RSK family of protein kinases and the phosphorylation of NRF-1 by as yet unidentified kinases. CREB phosphorylation is rapid andtransient, whereas NRF-1 phosphorylation is slower and persists in proliferating cells along with elevated PRC expression. The relationship, if any,between PRC promoter occupancy and these phosphorylation events is not known.



Like other genes in this class, PRC mRNA is increasedrapidly by serum induction of quiescent fibroblasts in thepresence of the protein synthesis inhibitor cycloheximide(220). The mRNA for PRC and other immediate earlygenes is superinduced and also markedly stabilized by thedrug, probably because of a requirement for protein syn-thesis for mRNA turnover. Cytochrome c is also seruminduced in part through the NRF-1 and CREB sites withinits promoter (92). NRF-1 and CREB have been implicatedas regulators of cell growth. As discussed above, NRF-1occupies the promoters of many genes required for DNAreplication, cytokinesis, and mitosis including a numberof those targeted by the E2F family of growth regulators(33). CREB is well known as a target of mitogenic signal-ing pathways (137). Interestingly, PRC binds CREBthrough the same binding sites used for NRF-1, and bothfactors exist in a complex with PRC in cell extracts (220).In addition, NRF-1 and CREB both occupy the cyto-chrome c promoter in vivo, and PRC occupancy of thecytochrome c promoter increases upon serum inductionof quiescent fibroblasts. This is consistent with a modelwhereby PRC targets these transcription factors in re-sponse to mitogenic signals. This is supported by theobservation that expression of the PRC NRF-1/CREBbinding domain in trans leads to dominant negative inhi-bition of cell growth on galactose as a carbon source(220). Growth on galactose requires mitochondrial respi-ration, suggesting that PRC may link the expression ofgenes required for both cell growth and mitochondrialrespiration.

V. LESSONS FROM MOUSE GENE KNOCKOUTS

The complexity of the pathways of mitochondrialbiogenesis is illustrated by the phenotypes of knockout

mice. As summarized in Table 1, these studies reveal thatboth essential and nonessential genes define the mito-chondrial phenotype. The essential genes are generallysingle copy and encode both nuclear and mitochondrialtranscription factors. These genes are not only requiredfor the maintenance of mitochondrial function but alsofor proper embryogenesis and organismal viability. Theyinclude Tfam, the NRFs, and other factors that are fun-damental to the expression of nuclear and mitochondrialgenes. The nuclear transcription factors implicated inmitochondrial biogenesis are not exclusive to this processbut rather integrate the expression of mitochondria withthat of other fundamental cellular functions. The nones-sential genes are members of small gene families andfunction as important transcriptional modulators of mito-chondrial expression and function. These include thetranscription factors ERR� and PPAR� as well as thePGC-1 family of coactivators that target key transcriptionfactors in specifying tissue-specific energetic properties, asubset of which is mitochondrial content. Individuallythey are not required for viability or for maintaining basalfunctional levels of mitochondria (Table 1). However,through their regulated expression and transcriptionalspecificities, these regulatory factors play an importantrole in relaying extracellular signals to the transcriptionmachinery.

A. Transcription Factor Knockouts

A targeted disruption of Tfam confirmed its in vivofunction in mitochondrial transcription and replication(Table 1). A homozygous Tfam knockout results inpostimplantation lethality between embryonic days E8.5and E10.5 (113). The E8.5 embryos are severely depletedof mtDNA and are deficient in both cytochrome oxidase

TABLE 1. Mouse knockouts of nucleus-encoded transcriptional regulators implicated in mitochondrial

biogenesis and function

Gene Knockout Viability Mitochondrial Phenotype

Tfam (germ line) Embryonic lethal between days E8.5 and E11.5 mtDNA depletion, severe respiratory chain deficiencyPolgA Embryonic lethal between days E7.5 and E8.5 mtDNA depletion, cytochrome oxidase deficiencyNRF-1 Embryonic lethal between days E3.5 and E6.5 mtDNA depletion, loss of mitochondrial membrane potential,

blastocyst growth defectNRF-2 (GABP) Preimplantation lethal Not determinedYY1 Embryonic lethal between days E3.5 and E8.5 Not determinedPPAR� Viable and fertile Reduced constitutive expression of mitochondrial fatty acid

oxidation enzymes, defective fatty acid oxidation inresponse to fasting

ERR� Viable and fertile Normal food consumption and energy expenditure, reducedlipogenesis, cold intolerance

PGC-1� (Lin et al.) Viable and fertile, increased postnatal mortality Normal mitochondrial morphology, reduced O2 consumptionin hepatocytes, fasting hypoglycemia, cold intolerance,defective cardiac ATP production

PGC-1� (Leone et al.) Viable and fertile, normal mortality Increased body mass, modest reduction in skeletal musclemitochondria, normal gluconeogenesis

PGC-1� Viable and fertile Normal energy intake and expenditure, reduction in skeletalmuscle and heart mitochondria



and succinate dehydrogenase activities. In addition, tis-sue-specific Tfam knockout animals show a strong corre-lation between Tfam and mtDNA abundance (122, 201).These studies establish an essential role for Tfam in themaintenance of mtDNA copy number in vivo. Interest-ingly, a homozygous knockout of mitochondrial DNApolymerase � also results in developmental arrest at E7.5–E8.5, severe depletion of mtDNA, and loss of cytochromeoxidase activity (83). These similarities in mitochondrialphenotype between Tfam and DNA polymerase � knock-outs (Table 1) illustrate that intact mtDNA transcriptionand replication machinery is required for mtDNA mainte-nance and for development beyond late gastrulation andearly organogenesis. It is also notable that overexpressingTwinkle, a mtDNA helicase, in transgenic mice leads toincreased mtDNA copy number (215). In humans, domi-nant mutations in Twinkle result in the accumulation ofmultiple mtDNA deletions in several somatic tissues.These results reinforce the conclusion that mtDNA tran-scription and replication factors are essential to the ex-pression of the mitochondrial respiratory chain in mam-malian tissues.

Mouse gene knockouts of several of the nuclear tran-scription factors implicated in mitochondrial biogenesishave also been analyzed (Table 1). A homozygous NRF-1mouse knockout results in early embryonic lethality be-tween embryonic days E3.5 and E6.5 (96). Although theyare morphologically normal, the NRF-1 null blastocystsdegenerate rapidly in culture and have severely reducedmtDNA levels accompanied by a deficiency in maintaininga mitochondrial membrane potential. The mtDNA deple-tion in the blastocyst does not result from a defect inmtDNA amplification during oogenesis, arguing that themtDNA depletion occurs between fertilization and theblastocyst stage. The phenotype is consistent with the lossof an NRF-1-dependent pathway of mtDNA maintenance.However, depletion of mtDNA alone does not explain theperi-implantation lethality of the homozygous NRF-1 nullsbecause Tfam knockout embryos are also depleted ofmtDNA but survive to between embryonic days E8.5 andE10.5 (113). Thus the earlier lethality of NRF-1 nulls mayresult from the decreased expression of NRF-1 targetgenes that are required for other cell growth and devel-opmental functions. Among these may be the subset ofgenes identified as targets of both NRF-1 and the E2Ffamily by ChiP on chip assay (33). It is of interest in thiscontext that loss-of-function mutations in NRF-1 have adramatic negative effect on both Drosophila and ze-brafish development. Partial loss-of-function mutations inEWG, a NRF-1 relative in Drosophila, result in aberrantintersegmental axonal projection pathways in the embryo(53). Total loss-of-function mutations in EWG are embry-onic lethal. The NRF-1 gene in zebrafish is expressed inthe eye and central nervous system of developing em-bryos. Its disruption gives a larval-lethal phenotype ac-

companied by defects in the development of the centralnervous system (20).

As observed for NRF-1, GABP� (NRF-2�) homozy-gous null mice also exhibit a peri-implantation lethal phe-notype (176). Although no determination of the state ofmtDNA or mitochondrial function was made in the mu-tant embryos, the early lethality may result from a com-bination of mitochondrial and nonmitochondrial defects.Homozygous knockouts of other Ets family transcriptionfactors also exhibit early embryonic lethality, suggestingthat members of the family are unable to compensate forone another during embryonic development. It is surpris-ing in light of the NRF knockouts that mice homozygousnull for ERR� are viable and fertile and display no defectsin food intake or energy expenditure (Table 1), althoughthey do have reduced fat mass and are resistant to high-fatdiet-induced obesity (132). The ERR� knockouts show amodest reduction in cytochrome c mRNA, but no changesin the global expression of other respiratory subunitgenes have been noted. The absence of a profound mito-chondrial deficiency in these mice may be explained bythe fact that ERR� is one of a small family of relatedfactors whose members may compensate for the loss of asingle isoform. In contrast, Tfam and NRF-1 are the prod-ucts of unique genes. Recent work demonstrates that theERR� knockout mice are deficient in adaptive thermogen-esis because of decreased mitochondrial mass in brownadipose tissue (221). Although PGC-1� and UCP-1 wereinduced normally by cold exposure in the ERR� knock-outs, the decrease in mitochondria was accompanied byreductions in respiratory gene expression, mtDNA copynumber, and the oxidative capacity of the tissue. Theremaining mitochondria were morphologically and func-tionally normal. The results argue that ERR� has a spe-cific function in differentiation-induced mitochondrialbiogenesis in brown adipose tissue without affectingbasal levels of mitochondrial maintenance.

A similar theme is echoed in mice with a targeteddisruption of PPAR� (Table 1). The PPAR� knockouts arealso viable and fertile with no obvious phenotypic abnor-malities (115). However, they do have a markedly bluntedresponse to peroxisome proliferators, demonstrating acrucial role for PPAR� in the pleiotropic response tothese agents. In addition, the mice display lower consti-tutive expression of several mitochondrial fatty acid me-tabolizing enzymes (8) and, in contrast to wild-type mice,the PPAR� knockouts were resistant to the induction ofthese mitochondrial enzymes in both liver and heart inresponse to fasting (119). Thus, although not required fornormal development, viability, or mitochondrial mainte-nance, PPAR� plays a crucial role in orchestrating regu-latory responses necessary for optimal mitochondrial ox-idative functions.



B. Coactivator Knockouts

As we have seen, gain-of-function experiments haveestablished that overexpression of either PGC-1� orPGC-1� can orchestrate massive increases in mitochon-drial biogenesis both in cultured cells and in mouse tis-sues (10, 116, 236). These changes have been ascribed totheir ability to induce the expression of key transcriptionfactors (NRF-1, NRF-2, ERR�) and to interact specificallywith these factors in mediating the activation of genesthat promote the biogenesis of mitochondria. Thus it issurprising that mice with a targeted disruption of PGC-1�are viable and show no changes in mitochondrial abun-dance or morphology in liver or brown fat (128). They doshow reduced oxygen consumption in isolated hepato-cytes and reduced expression of several mRNAs linked tomitochondrial function. The only tissues to display obvi-ous morphological abnormalities were brown adipose tis-sue and the striatum of the brain. Defects in the striatumhave been associated with movement disorders, and thePGC-1� null mice were markedly hyperactive. This hyper-activity correlated with a loss of axons in the striatum aswell as with reductions in nucleus-encoded mRNAs forrespiratory- and brain-specific genes. Subsequent experi-ments with the PGC-1� null mice demonstrated thatPGC-1� is not required for mitochondrial biogenesis (9).However, the mice are deficient in the expression ofgenes necessary for mitochondrial function and in oxida-tive enzyme activities in heart and skeletal muscle. Thesechanges coincide with reduced cardiac ATP productionand a defect in work output in response to physiologicalstimuli. Recently, a skeletal muscle-specific PGC-1� knock-out mouse was characterized (84). These mice clearlyshow multiple muscle defects including reduced exercisetolerance and abnormalities in the maintenance of normalmuscle fiber composition. The results illustrate that thetrue tissue-specific functions of PGC family coactivatorsmay be masked by pleiotropic phenotypes in mice har-boring a generalized gene disruption in all of their tissues.

The phenotype of an independently constructedPGC-1� null mouse confirmed that the coactivator is notrequired for normal development or for the global biogen-esis or maintenance of mitochondria (118) (Table 1).However, these mice had reduced mitochondrial densityin slow-twitch skeletal muscle and modestly reduced re-spiratory capacity in both liver and skeletal muscle. Thereare also a number of significant differences with the pre-viously described knockout including the absence of adefect in gluconeogenesis and a disparity in cardiac phe-notype. These have been ascribed to differences in ge-netic background or targeting strategies (68).

Recently, a targeted disruption of the mouse PGC-1�gene was constructed by removal of exons containing thenuclear hormone receptor coactivator signature motifs(LXXLL) and introduction of a premature stop codon

(117). Mice with a homozygous disruption of the genewere viable and fertile and exhibited no gross changes inwhole body substrate utilization or total energy input oroutput (Table 1). They did however have a higher overallmetabolic rate associated with a lean phenotype com-pared with wild-type littermates. Closer examination re-vealed a number of tissue alterations in mitochondrialgene expression and respiratory function. There wereclear reductions in some genes associated with mitochon-drial electron transport chain function in brown andwhite fat, skeletal muscle, and heart. These included re-spiratory subunits encoded by both nuclear and mito-chondrial genes. Although diminished mRNA levels werenot always accompanied by similar changes in the en-coded protein, mitochondrial volume was reduced in bothheart and skeletal muscle. In these cases, the accompa-nying decreases in electron transport and oxidative phos-phorylation activities resulted from reduced numbers offunctionally normal mitochondria. In brown fat, mito-chondrial content was normal, but the knockout miceexhibited a blunted response to norepinephrine-inducedthermogenesis. These phenotypes clearly point to an im-portant modulatory role for PGC-1� in maintaining opti-mal tissue energetics.

The absence of a dramatic mitochondrial biogenesisdefect in either the PGC-1� or PGC-1� null mice mayresult from mechanisms that compensate for the individ-ual loss of these coactivators. Since the PGC-1 familymembers share several functional motifs and can alltrans-activate NRF target genes, the remaining familymembers may substitute functionally for the ablatedmember. This is seen in the brown fat of the PGC-1�knockout where PGC-1� levels are markedly upregulated,possibly resulting in the maintenance of normal mito-chondrial content in this tissue (117). This issue wasaddressed by knocking down the expression of PGC-1�by shRNA in a preadipocyte cell line derived from PGC-1�knockout mice (216). Although PGC-1� appears not to berequired for adipocyte differentiation, its loss is corre-lated with an inability to induce cAMP-dependent expres-sion of genes necessary for thermogenesis, includingUCP-1 and cytochrome c. UCP-1, however, is induced byinsulin and retinoic acid in the absence of PGC-1�. Re-duced expression of PGC-1� in the PGC-1� null back-ground led to a failure to establish and maintain differen-tiated levels of mitochondrial density and function inmature brown adipocytes. Mitochondrial content is re-duced to a level present in undifferentiated preadipo-cytes. Thus, although neither coactivator is required todirect the program of adipocyte differentiation, one or theother is essential to achieve a differentiated mitochon-drial phenotype. This is reminiscent of the case of ERR�,suggesting that the PGC-ERR� regulatory pathway directsdifferentiation-specific changes in mitochondrial contentin brown fat and possibly other tissues. It remains an



open question as to whether PRC can support basal levelsof mitochondrial biogenesis in the absence of PGC-1� and-�. It is of interest in this context that expression of adominant negative allele of PRC from a lentivirus vectorinhibits respiratory growth in cultured cells (220). Clearly,the PGC-1 coactivators exhibit highly specialized regula-tory functions but, individually, do not appear to be thesole determinants of mitochondrial content in the major-ity of tissues.

VI. RETROGRADE PATHWAYS

A. The RTG Pathway in Yeast

The regulated expression of the PGC-1 family of in-ducible coactivators provides a mechanism for modulat-ing mitochondrial biogenesis and function in response toa variety of extracellular signals. In addition, pathwaysexist for the communication of the functional state ofmitochondria back to the nuclear transcriptional machin-ery. This phenomenon of retrograde signaling is well stud-ied at the transcriptional level in the yeast Saccharomyces

cerevisiae (130). In yeast cells that are depleted of mito-chondrial DNA (�0 cells), the resulting respiratory defi-ciency reduces the supply of glutamate, which is derivedfrom the �-ketoglutarate of the citric acid cycle. Cellscompensate by increasing production of peroxisomes, thesites of fatty acid oxidation in yeast. This generatesacetyl-CoA, which fuels the citric acid cycle. The nuclearCIT2 gene encoding peroxisomal citrate synthase ismarkedly induced in �0 cells. This enzyme is part of theglyoxylate cycle, and its induction in response to a defectin mitochondrial respiration allows cells to generate ci-trate (200). Positive transcriptional control of peroxiso-mal citrate synthase and other peroxisomal proteins in �0

cells is mediated by basic helix-loop-helix-leucine zippertranscription factors, Rtg1p and Rtg3p (177). A third fac-tor Rtg2p facilitates the translocation of Rtg1p and Rtg3pfrom the cytoplasm to the nucleus where they bind totheir target genes as a heterodimer and drive transcrip-tion. Rtg2p has an ATP binding site and is thought tofunction as a sensor of the functional state of mitochon-dria in part through its interaction with the phosphopro-tein Mks1p. The phosphorylated state of Mks1p dictatesits interaction with Rtg2p or negative regulators creatinga molecular switch for control of the RTG pathway (67).

B. Mammalian Retrograde Regulation

The Rtg pathway has not been identified in vertebratecells, but there are a number of examples where theexpression of nuclear genes is altered by deficiencies inmitochondrial function (31). Defective mitochondria that

result from certain mitochondrial disease mutations pro-liferate in diseased muscle fibers giving rise to ragged redfibers (145, 226). Specific nuclear genes involved in ATPproduction also display elevated expression in cells withmtDNA mutations (89). An altered pattern of nuclear geneexpression, involving proteins of the mitochondrial innermembrane, intermediate filaments, and ribosomes, occursin human cells upon depletion of mtDNA (123). Likewise,chicken cells either depleted of mtDNA or treated withthe mitochondrial protein synthesis inhibitor chloram-phenicol have increased levels of mRNAs for elongationfactor 1�, �-actin, v-myc, and GAPDH (229). Although themechanistic bases for these phenomena have not beenelucidated, they are thought to represent nuclear re-sponses to deficiencies in ATP production.

Recent studies suggest that retrograde signaling inanimal cells may be mediated by calcium (22). Inhibitionof mitochondrial respiratory function either by depletionof mtDNA or by metabolic inhibitors results in a stressresponse that coincides with elevated cytosolic calciumlevels. The response includes increased expression ofcalcium-responsive transcription factors and cytochromeoxidase subunit Vb. Calcium signaling has also been as-sociated with a PGC-1-dependent pathway of mitochon-drial biogenesis in skeletal muscle (155, 235). Constitutiveexpression of calcium/calmodulin-dependent protein ki-nase IV (CaMKIV) in the skeletal muscle of transgenicmice leads to increased mtDNA copy number and respi-ratory enzymes as well as elevated PGC-1� levels. Thuscalcium may be an important link between the relay ofextracellular signals to the nucleus and the bidirectionalcommunication between nucleus and mitochondria. Thisnotion is reinforced by the observation that CREB isfound in its transcriptionally active, phosphorylated statein cells exhibiting mitochondrial insufficiency resultingeither from the loss of mtDNA or by pathological muta-tion in mtDNA (11). This coincides with the activation ofCaMKIV, which, as discussed, can lead to CREB phosphor-ylation in response to calcium. It remains to be determinedwhether calcium signaling through CREB and the PGC-1family coactivators represents a physiologically meaningfulpathway of retrograde regulation in vertebrate cells.

VII. CONCLUSION

Much progress has been made in uncovering thetranscriptional mechanisms that govern the function andbiogenesis of mitochondria. The identification and char-acterization of important regulatory factors has led to aframework for understanding the continuity between sig-naling events affecting cellular energetics and the expres-sion of both nuclear and mitochondrial genes that dictatethe mitochondrial phenotype. The preponderance of evi-dence suggests that nucleus-encoded transcription fac-



tors acting on both nuclear and mitochondrial genes serveto coordinate the expression of the gene products re-quired for maintenance of the respiratory apparatus aswell as other essential mitochondrial functions. A subsetof these is also required for proper embryonic develop-ment and organismal viability. Because the organelle isintimately associated with a myriad of cellular functions,it is not surprising to find that the factors governingmitochondrial expression are engaged in other cellularactivities as well. Thus they likely play a fundamentalintegrative role in coordinating mitochondrial functionswith cellular events necessary for growth and develop-ment. This likely accounts for the embryonic lethalityobserved in gene knockouts for several of these factors.Among the nucleus-encoded factors are those that areimported into mitochondria where they direct the tran-scription and replication of mtDNA. Recent progress hasled to the elucidation of all of the major components ofthe vertebrate mitochondrial transcriptional machineryincluding the polymerase and requisite initiation and ter-mination factors. The stage is set for further understand-ing of the regulatory mechanisms by which this elegantsystem governs mitochondrial gene expression andmtDNA abundance. An unresolved issue is whether fac-tors can be shared between nuclear and mitochondrialtranscription systems and, if so, what roles such factorsplay in nucleomitochondrial interactions.

Superimposed on this core of transcription factorsare regulators that modulate mitochondrial biogenesis inresponse to extra- and possibly intracellular signals. Animportant breakthrough has been the discovery of thePGC-1 family of regulated coactivators. Although individ-ually these factors are not essential for embryonic devel-opment or viability in mice, they are important determi-nants of cell- and tissue-specific bioenergetic phenotypes.PGC-1 family members are differentially expressed ormodified in response to a variety of extracellular signalsincluding temperature, energy deprivation, and availabil-ity of nutrients and growth factors. They also exhibitdifferential specificities for transcription factor utiliza-tion, which in turn governs the pattern of gene expressionorchestrated by each coactivator. All three family mem-bers activate gene expression through the nuclear respi-ratory factors (NRF-1, NRF-2) consistent with the factthat their biological responses all encompass changes inmitochondrial expression. It is clear that the PGC-1 coac-tivators play a role in specifying differentiation-inducedmitochondrial content in tissues such as brown fat andskeletal muscle. However, individually they are unlikelyto be the only determinants of basal mitochondrial con-tent in most tissues. Perhaps there is sufficient functionalredundancy among the three family members such thatthe loss of any one is at least partially compensated for bythe others. Additional loss of function experiments com-bined with a better understanding of the biological func-

tions of PRC and PGC-1� should clarify this issue. It willalso be of interest to determine whether retrograde sig-naling is mediated by any of these coactivators.

ACKNOWLEDGMENTS

Address for reprint requests and other correspondence:R. C. Scarpulla, Dept. of Cell and Molecular Biology, Northwest-ern Medical School, 303 East Chicago Ave., Chicago, IL 60611.

GRANTS

Work in the author’s laboratory was supported by NationalInstitute of General Medical Sciences Grant GM-32525–25.

REFERENCES

1. Adhikary S, Eilers M. Transcriptional regulation and transforma-tion by Myc proteins. Nat Rev Mol Cell Biol 6: 635–645, 2005.

2. Aizencang GI, Bishop DF, Forrest D, Astrin KH, Desnick RJ.

Uroporphyrinogen III synthase. An alternative promoter controlserythroid-specific expression in the murine gene. J Biol Chem 275:2295–2304, 2000.

3. Akimoto T, Pohnert SC, Li P, Zhang M, Gumbs C, Rosenberg

PB, Williams RS, Yan Z. Exercise stimulates Pgc-1� transcriptionin skeletal muscle through activation of the p38 MAPK pathway.J Biol Chem 280: 19587–19593, 2005.

4. Amiott EA, Jaehning JA. Mitochondrial transcription is regu-lated via an ATP “sensing” mechanism that couples RNA abun-dance to respiration. Mol Cell 22: 329–338, 2006.

5. Andersen JK. Oxidative stress in neurodegeneration: cause orconsequence. Nature Rev Neurosci 5: S18–S25, 2004.

6. Andersson U, Scarpulla RC. PGC-1-related coactivator, a novel,serum-inducible coactivator of nuclear respiratory factor 1-depen-dent transcription in mammalian cells. Mol Cell Biol 21: 3738–3749,2001.

7. Antoshechkin I, Bogenhagen DF, Mastrangelo IA. The HMG-box mitochondrial transcription factor xl-mtTFA binds DNA as atetramer to activate bidirectional transcription. EMBO J 16: 3198–3206, 1997.

8. Aoyama T, Peters JM, Iritani N, Nakajima T, Furihata K,

Hashimoto T, Gonzalez FJ. Altered constitutive expression offatty acid-metabolizing enzymes in mice lacking the peroxisomeproliferator-activated receptor alpha (PPARalpha). J Biol Chem

273: 5678–5684, 1998.9. Arany Z, He H, Lin J, Hoyer K, Handschin C, Toka O, Ahmad

F, Matsui T, Chin S, Wu PH, Rybkin II, Shelton JM, Manieri

M, Cinti S, Schoen FJ, Bassel-Duby R, Rosenzweig A, Ingwall

JS, Spiegelman BM. Transcriptional coactivator PGC-1� controlsthe energy state and contractile function of cardiac muscle. Cell

Metab 1: 259–271, 2005.10. Arany Z, Lebrasseur N, Morris C, Smith E, Yang W, Ma Y, Chin

S, Spiegelman BM. The transcriptional coactivator PGC-1� drivesthe formation of oxidative type IIX fibers in skeletal muscle. Cell

Metab 5: 35–46, 2007.11. Arnould T, Vankoningsloo S, Renard P, Houbion A, Ninane N,

Demazy C, Remacle J, Raes M. CREB activation induced bymitochondrial dysfunction is a new signaling pathway that impairscell proliferation. EMBO J 21: 53–63, 2002.

12. Ashley MV, Laipis PJ, Hauswirth WW. Rapid segregation ofheteroplasmic bovine mitochondria. Nucleic Acids Res 17: 7325–7331, 1989.

13. Asin-Cayuela J, Helm M, Attardi G. A monomer-to-trimer tran-sition of the human mitochondrial transcription termination factor(mTERF) is associated with a loss of in vitro activity. J Biol Chem

279: 15670–15677, 2004.14. Au HC, Scheffler IE. Promoter analysis of the human succinate

dehydrogenase iron-protein gene. Both nuclear respiratory factorsNRF-1 and NRF-2 are required. Eur J Biochem 251: 164–174, 1998.

15. Baar K, Wende AR, Jones TE, Marison M, Nolte LA, Chen M,

Kelly DP, Holloszy JO. Adaptations of skeletal muscle to exer-



cise: rapid increase in the transcriptional coactivator PGC-1.FASEB J 16: 1879–1886, 2002.

16. Balaban RS. Regulation of oxidative phosphorylation in the mam-malian cell. Am J Physiol Cell Physiol 258: C377–C389, 1990.

17. Barish GD, Narkar VA, Evans RM. PPAR delta: a dagger in theheart of the metabolic syndrome. J Clin Invest 116: 590–597, 2006.

18. Basu A, Lenka N, Mullick J, Avadhani NG. Regulation of murinecytochrome oxidase Vb gene expression in different tissues andduring myogenesis: role of a YY-1 factor-binding negative enhancer.J Biol Chem 272: 5899–5908, 1997.

19. Batchelor AH, Piper DE, De la Brousse FC, McKnight SL,

Wolberger C. The structure of GABP�/�: an ETS domain ankyrinrepeat heterodimer bound to DNA. Science 279: 1037–1041, 1998.

20. Becker TS, Burgess SM, Amsterdam AH, Allende ML, Hop-

kins N. Not really finished is crucial for development of thezebrafish outer retina and encodes a transcription factor highlyhomologous to human nuclear respiratory factor 1 and avian initi-ation binding repressor. Development 124: 4369–4378, 1998.

21. Bergeron R, Ren JM, Cadman KS, Moore IK, Perret P, Py-

paert M, Young LH, Semenkovich CF, Shulman GI. Chronicactivation of AMP kinase results in NRF-1 activation and mitochon-drial biogenesis. Am J Physiol Endocrinol Metab 281: E1340–E1346, 2001.

22. Biswas G, Adebanjo OA, Freedman BD, Anandatheertha-

varada HK, Vijayasarathy C, Zaidi M, Kotlikoff M, Avadhani

NG. Retrograde Ca2� signaling in C2C12 skeletal myocytes inresponse to mitochondrial genetic and metabolic stress: a novelmode of inter-organelle crosstalk. EMBO J 18: 522–533, 1999.

23. Blesa JR, Prieto-Ruiz JA, Hernandez JM, Hernandez-Yago J.

NRF-2 transcription factor is required for human TOMM20 geneexpression. Gene 391: 198–208, 2007.

24. Bogenhagen DF. Interaction of mtTFB and mtRNA polymerase atcore promoters for transcriptionof Xenopus laevis mtDNA. J Biol

Chem 271: 12036–12041, 1996.25. Bogenhagen DF, Clayton DA. The mitochondrial DNA replica-

tion bubble has not burst. Trends Biochem Sci 28: 357–360, 2003.26. Bogenhagen DF, Wang Y, Shen EL, Kobayashi R. Protein com-

ponents of mitochondrial DNA nucleoids in higher eukaryotes. Mol

Cell Proteomics 2: 1205–1216, 2003.27. Bonawitz ND, Clayton DA, Shadel GS. Initiation and beyond:

multiple functions of the human mitochondrial transcription ma-chinery. Mol Cell 24: 813–825, 2006.

28. Booth FW, Thomason DB. Molecular and cellular adaptation ofmuscle in response to exercise: perspectives of various models.Physiol Rev 71: 541–585, 1991.

29. Bottero V, Rossi F, Samson M, Mari M, Hofman P, Peyron JF.

I�B-�, the NF-�B inhibitory subunit, interacts with ANT, the mito-chondrial ATP/ADP translocator. J Biol Chem 276: 21317–21324,2001.

30. Braidotti G, Borthwick IA, May BK. Identification of regulatorysequences in the gene for 5-aminolevulinate synthase from rat.J Biol Chem 268: 1109–1117, 1993.

31. Butow RA, Avadhani NG. Mitochondrial signaling: the retrograderesponse. Mol Cell 14: 1–15, 2004.

32. Calzone FJ, Hoog C, Teplow DB, Cutting AE, Zeller RW,

Britten RJ, Davidson EH. Gene regulatory factors of the seaurchin embryo. I. Purification by affinity chromatography and clon-ing of P3A2, a novel DNA binding protein. Development 112: 335–350, 1991.

33. Cam H, Balciunaite E, Blias A, Spektor A, Scarpulla RC,

Young R, Kluger Y, Dynlacht BD. A common set of gene regula-tory networks links metabolism and growth inhibition. Mol Cell 16:399–411, 2004.

34. Cannon B, Nedergaard J. Brown adipose tissue: function andphysiological significance. Physiol Rev 84: 277–359, 2004.

35. Cao WH, Daniel KW, Robidoux J, Puigserver P, Medvedev AV,

Bai X, Floering LM, Spiegelman BM, Collins S. p38 mitogen-activated protein kinase is the central regulator of cyclic AMP-dependent transcription of the brown fat uncoupling protein 1gene. Mol Cell Biol 24: 3057–3067, 2004.

36. Carter RS, Bhat NK, Basu A, Avadhani NG. The basal promoterelements of murine cytochrome c oxidase subunit IV gene consist

of tandemly duplicated ets motifs that bind to GABP-related tran-scription factors. J Biol Chem 267: 23418–23426, 1992.

37. Chang DD, Clayton DA. Priming of human mitochondrial DNAreplication occurs at the light-strand promoter. Proc Natl Acad Sci

USA 82: 351–355, 1985.38. Chang DD, Hauswirth WW, Clayton DA. Replication priming

and transcription initiate from precisely the same site in mousemitochondrial DNA. EMBO J 4: 1559–1567, 1985.

39. Chau CA, Evans MJ, Scarpulla RC. Nuclear respiratory factor 1activation sites in genes encoding the gamma-subunit of ATP syn-thase, eukaryotic initiation factor 2�, tyrosine aminotransferase.Specific interaction of purified NRF-1 with multiple target genes.J Biol Chem 267: 6999–7006, 1992.

40. Chen Y, Zhou G, Yu M, He Y, Tang W, Lai J, He J, Liu W, Tan

D. Cloning and functional analysis of human mTERFL encoding anovel mitochondrial transcription termination factor-like protein.Biochem Biophys Res Commun 337: 1112–1118, 2005.

41. Chinenov Y, Henzl M, Martin ME. The � and � subunits of theGA-binding protein form a stable heterodimer in solution. J Biol

Chem 275: 7749–7756, 2000.42. Choi YS, Lee HK, Pak YK. Characterization of the 5�-flanking

region of the rat gene for mitochondrial transcription factor A(Tfam). Biochim Biophys Acta 1574: 200–204, 2002.

43. Chomyn A, Attardi G. MtDNA mutations in aging and apoptosis.Biochem Biophys Res Commun 304: 519–529, 2003.

44. Christianson TW, Clayton DA. In vitro transcription of humanmitochondrial DNA: accurate termination requires a region of DNAsequence that can function bidirectionally. Proc Natl Acad Sci USA

83: 6277–6281, 1986.45. Clayton DA. Vertebrate mitochondrial DNA: a circle of surprises.

Exp Cell Res 255: 4–9, 2000.46. Clayton DA. Mitochondrial DNA replication: what we know.

IUBMB Life 55: 213–217, 2003.47. Cogswell PC, Kashatus DF, Keifer JA, Guttridge DC, Reuther

JY, Bristow C, Roy S, Nicholson DW, Baldwin AS Jr. NF-kappaB and IkappaB� are found in the mitochondria: evidence forregulation of mitochondrial gene expression by NF-kappaB. J Biol

Chem 278: 2963–2968, 2003.48. Costanzo MC, Fox TD. Control of mitochondrial gene expression

in Saccharomyces cerevisiae. Annu Rev Genet 24: 91–113, 1990.49. Daga A, Micol V, Hess D, Aebersold R, Attardi G. Molecular

characterization of the transcription termination factor from hu-man mitochondria. J Biol Chem 268: 8123–8130, 1993.

50. Dairaghi DJ, Shadel GS, Clayton DA. Addition of a 29 residuecarboxyl-terminal tail converts a simple HMG box-containing pro-tein into a transcriptional activator. J Mol Biol 249: 11–28, 1995.

51. Demonacos C, Djordjevic-Markovic R, Tsawdaroglou N, Sekeris

CE. The mitochondrion as a primary site of action of glucocorti-coids: the interaction of the glucocorticoid receptor with mitochon-drial DNA sequences showing partial similarity to the nuclearglucocorticoid responsive elements. J Steroid Biochem Mol Biol

55: 43–55, 1995.52. Demonacos C, Tsawdaroglou N, Djordjevic-Markovic R, Pa-

palopoulou M, Galanopoulos V, Papadogeorgaki S, Sekeris

CE. Import of the glucocorticoid receptor into rat liver mitochon-dria in vivo and in vitro. J Steroid Biochem Mol Biol 46: 401–413,1993.

53. Desimone SM, White K. The Drosophila erect wing gene, whichis important for both neuronal and muscle development, encodes aprotein which is similar to the sea urchin P3A2 DNA bindingprotein. Mol Cell Biol 13: 3641–3949, 1993.

54. Diffley JF, Stillman B. A close relative of the nuclear, chromo-somal high-mobility group protein HMG1 in yeast mitochondria.Proc Natl Acad Sci USA 88: 7864–7868, 1991.

55. DiMauro S, Davidzon G. Mitochondrial DNA and disease. Ann

Med 37: 222–232, 2005.56. Donohoe ME, Zhang X, McGinnis L, Biggers J, Li E, Shi Y.

Targeted disruption of mouse yin yang 1 transcription factor resultsin peri-implantation lethality. Mol Cell Biol 19: 7237–7244, 1999.

57. Dufour E, Larsson NG. Understanding aging: revealing order outof chaos. Biochim Biophys Acta 1658: 122–132, 2004.

58. Efiok BJS, Safer B. Transcriptional regulation of E2F-1 and eIF-2genes by �-Pal: a potential mechanism for coordinated regulation



of protein synthesis, growth, the cell cycle. Biochim Biophys Acta

1495: 51–68, 2000.59. Ekstrand MI, Falkenberg M, Rantanen A, Park CB, Gaspari

M, Hultenby K, Rustin P, Gustafsson CM, Larsson NG. Mito-chondrial transcription factor A regulates mtDNA copy number inmammals. Hum Mol Genet 13: 935–944, 2004.

60. Elbehti-Green A, Au HC, Mascarello JT, Ream-Robinson D,

Scheffler IE. Characterization of the human SDHC gene encodingone of the integral membrane proteins of succinate-quinone oxi-doreductase in mitochondria. Gene 213: 133–140, 1998.

61. Enrıquez JA, Fernandez-Silva P, Garrido-Perez N, Montoya

J. Direct regulation of mitochondrial RNA synthesis by thyroidhormone. Mol Cell Biol 19: 657–670, 1999.

62. Evans MJ, Scarpulla RC. Both upstream and intron sequenceelements are required for elevated expression of the rat somaticcytochrome c gene in COS-1 cells. Mol Cell Biol 8: 35–41, 1988.

63. Evans MJ, Scarpulla RC. Interaction of nuclear factors withmultiple sites in the somatic cytochrome c promoter. Characteriza-tion of upstream NRF-1, ATF and intron Sp1 recognition sites.J Biol Chem 264: 14361–14368, 1989.

64. Evans MJ, Scarpulla RC. NRF-1:a trans-activator of nuclear-encoded respiratory genes in animal cells. Genes Dev 4: 1023–1034,1990.

65. Falkenberg M, Gaspari M, Rantanen A, Trifunovic A, Larsson

NG, Gustafsson CM. Mitochondrial transcription factors B1 andB2 activate transcription of human mtDNA. Nat Genet 31: 289–294,2002.

66. Fernandez-Silva P, Martinez-Azorin F, Micol V, Attardi G. Thehuman mitochondrial transcription termination factor (mTERF) isa multizipper protein but binds to DNA as a monomer, with evi-dence pointing to intramolecular leucine zipper interactions.EMBO J 16: 1066–1079, 1997.

67. Ferreira, J Jr, Spirek M, Liu Z, Butow RA. Interaction betweenRtg2p and Mks1p in the regulation of the RTG pathway of Saccha-

romyces cerevisiae. Gene 354: 2–8, 2005.68. Finck BN, Kelly DP. PGC-1 coactivators: inducible regulators of

energy metabolism in health and disease. J Clin Invest 116: 615–622, 2006.

69. Fisher RP, Clayton DA. Purification and characterization of hu-man mitochondrial transcription factor 1. Mol Cell Biol 8: 3496–3509, 1988.

70. Fisher RP, Lisowsky T, Parisi MA, Clayton DA. DNA wrappingand bending by a mitochondrial high mobility group-like transcrip-tional activator protein. J Biol Chem 267: 3358–3367, 1992.

71. Fisher RP, Parisi MA, Clayton DA. Flexible recognition of rap-idly evolving promoter sequences by mitochondrial transcriptionfactor 1. Genes Dev 3: 2202–2217, 1989.

72. FitzGerald PC, Shlyakhtenko A, Mir AA, Vinson C. Clusteringof DNA sequences in human promoters. Genome Res 14: 1562–1574, 2004.

73. Gleyzer N, Vercauteren K, Scarpulla RC. Control of mitochon-drial transcription specificity factors (TFB1M and TFB2M) by nu-clear respiratory factors (NRF-1 and NRF-2) and PGC-1 familycoactivators. Mol Cell Biol 25: 1354–1366, 2005.

74. Gomez-Cuadrado A, Martin M, Noel M, Ruiz-Carrillo A. Initi-ation binding receptor, a factor that binds to the transcriptioninitiation site of the histone h5 gene, is a glycosylated member of afamily of cell growth regulators. Mol Cell Biol 15: 6670–6685, 1995.

75. Gopalakrishnan L, Scarpulla RC. Differential regulation of re-spiratory chain subunits by a CREB-dependent signal transductionpathway. Role of cyclic AMP in cytochrome c and COXIV geneexpression. J Biol Chem 269: 105–113, 1994.

76. Goto M, Terada S, Kato M, Katoh M, Yokozeki T, Tabata I,

Shimokawa T. cDNA cloning and mRNA analysis of PGC-1 inepitrochlearis muscle in swimming-exercised rats. Biochem Bio-

phys Res Commun 274: 350–354, 2000.77. Guarente L, Mason T. Heme regulates transcription of the CYC1

gene of S. cerevisiae via an upstream activation site. Cell 32:1279–1286, 1983.

78. Gugneja S, Scarpulla RC. Serine phosphorylation within a con-cise amino-terminal domain in nuclear respiratory factor 1 en-hances DNA binding. J Biol Chem 272: 18732–18739, 1997.

79. Gugneja S, Virbasius CA, Scarpulla RC. Nuclear respiratoryfactors 1 and 2 utilize similar glutamine-containing clusters ofhydrophobic residues to activate transcription. Mol Cell Biol 16:5708–5716, 1996.

80. Gugneja S, Virbasius JV, Scarpulla RC. Four structurally dis-tinct, non-DNA-binding subunits of human nuclear respiratory fac-tor 2 share a conserved transcriptional activation domain. Mol Cell

Biol 15: 102–111, 1995.81. Gulick T, Cresci S, Caira T, Moore DD, Kelly DP. The peroxi-

some proliferator-activated receptor regulates mitochondrial fattyacid oxidative enzyme gene expression. Proc Natl Acad Sci USA 91:11012–11016, 1994.

82. Guo AL, Nie F, Wong-Riley M. Human nuclear respiratory factor2� subunit cDNA: Isolation, subcloning, sequencing, in situ hybrid-ization of transcripts in normal and monocularly deprived macaquevisual system. J Comp Neurol 417: 221–232, 2000.

83. Hance N, Ekstrand MI, Trifunovic A. Mitochondrial DNA poly-merase gamma is essential for mammalian embryogenesis. Hum

Mol Genet 14: 1775–1783, 2005.84. Handschin C, Chin S, Li P, Liu F, Maratos-Flier E, Lebrasseur

NK, Yan Z, Spiegelman BM. Skeletal muscle fiber-type switching,exercise intolerance and myopathy in PGC-1alpha muscle-specificknockout animals. J Biol Chem 282: 30014–30021, 2007.

85. Handschin C, Kobayashi YM, Chin S, Seale P, Campbell KP,

Spiegelman BM. PGC-1alpha regulates the neuromuscular junc-tion program and ameliorates Duchenne muscular dystrophy.Genes Dev 21: 770–783, 2007.

86. Handschin C, Lin J, Rhee J, Peyer AK, Chin S, Wu PH, Meyer

UA, Spiegelman BM. Nutritional regulation of hepatic heme bio-synthesis and phorphyria through PGC-1�. Cell 122: 505–515, 2005.

87. Handschin C, Rhee J, Lin JD, Tarr PT, Spiegelman BM. Anautoregulatory loop controls peroxisome proliferator-activated re-ceptor gamma coactivator 1� expression in muscle. Proc Natl Acad

Sci USA 100: 7111–7116, 2003.88. Hatefi Y. The mitochondrial electron transport chain and oxidative

phosphorylation system. Annu Rev Biochem 54: 1015–1069, 1985.89. Heddi A, Lestienne P, Wallace DC, Stepien G. Mitochondrial

DNA expression in mitochondrial myopathies and coordinated ex-pression of nuclear genes involved in ATP production. J Biol Chem

268: 12156–12163, 1993.90. Heerdt BG, Augenlicht LH. Changes in the number of mitochon-

drial genomes during human development. Exp Cell Res 186: 54–59,1990.

91. Herschman HR. Primary response genes induced by growth fac-tors and tumor promoters. Annu Rev Biochem 60: 281–319, 1991.

92. Herzig RP, Scacco S, Scarpulla RC. Sequential serum-dependentactivation of CREB and NRF-1 leads to enhanced mitochondrialrespiration through the induction of cytochrome c. J Biol Chem

275: 13134–13141, 2000.93. Herzig S, Long FX, Jhala US, Hedrick S, Quinn R, Bauer A,

Rudolph D, Schutz G, Yoon C, Puigserver P, Spiegelman B,

Montminy M. CREB regulates hepatic gluconeogenesis throughthe coactivator PGC-1. Nature 413: 179–183, 2001.

94. Hirawake H, Taniwaki M, Tamura A, Amino H, Tomitsuka E,

Kita K. Characterization of the human SDHD gene encoding thesmall subunit of cytochrome b (cybS) in mitochondrial succinate-ubiquinone oxidoreductase. Biochim Biophys Acta 1412: 295–300,1999.

95. Holloszy JO. Biochemical adaptations in muscle. Effects of exer-cise on mitochondrial oxygen uptake and respiratory enzyme ac-tivity in skeletal muscle. J Biol Chem 242: 2278–2282, 1967.

96. Huo L, Scarpulla RC. Mitochondrial DNA instability and peri-implantation lethality associated with targeted disruption of nu-clear respiratory factor 1 in mice. Mol Cell Biol 21: 644–654, 2001.

97. Huss JM, Kelly DP. Nuclear receptor signaling and cardiac ener-getics. Circ Res 95: 568–578, 2004.

98. Huss JM, Kopp RP, Kelly DP. Peroxisome proliferator-activatedreceptor coactivator-1� (PGC-1�) coactivates the cardiac-enrichednuclear receptors estrogen-related receptor-� and -�: identificationof novel leucine-rich interaction motif within PGC-1�. J Biol Chem

277: 40265–40274, 2002.



99. Jang SH, Jaehning JA. The yeast mitochondrial RNA polymerasespecificity factor, MTF1, is similar to bacterial sigma factors. J Biol

Chem 266: 22671–22677, 1991.100. Jorgensen SB, Wojtaszewski JFP, Viollet B, Andreelli F, Birk

JB, Hellsten Y, Schjerling P, Vaulont S, Neufer PD, Richter

EA, Pilegaard H. Effects of �-AMPK knockout on exercise-in-duced gene activation in mouse skeletal muscle. FASEB J 19:NIL795–NIL820, 2005.

101. Kaneda H, Hayashi J, Takahama S, Taya C, Lindahl KF,

Yonekawa H. Elimination of paternal mitochondrial DNA in in-traspecific crosses during early mouse embryogenesis. Proc Natl

Acad Sci USA 92: 4542–4546, 1995.102. Kanki T, Nakayama H, Sasaki N, Takio K, Alam TI, Hamasaki

N, Kang D. Mitochondrial nucleoid and transcription factor A. Ann

NY Acad Sci 1011: 61–68, 2004.103. Kanki T, Ohgaki K, Gaspari M, Gustafsson CM, Fukuoh A,

Sasaki N, Hamasaki N, Kang DC. Architectural role of mitochon-drial transcription factor A in maintenance of human mitochondrialDNA. Mol Cell Biol 24: 9823–9834, 2004.

104. Karlok MA, Jang SH, Jaehning JA. Mutations in the yeast mito-chondrial RNA polymerase specificity factor, Mtf1, verify an essen-tial role in promoter utilization. J Biol Chem 277: 28143–28149,2002.

105. Kaufman BA, Newman SM, Hallberg RL, Slaughter CA, Per-

lman PS, Butow RA. In organello formaldehyde crosslinking ofproteins to mtDNA: identification of bifunctional proteins. Proc

Natl Acad Sci USA 97: 7772–7777, 2000.106. Kelly DP, Scarpulla RC. Transcriptional regulatory circuits con-

trolling mitochondrial biogenesis and function. Genes Dev 18: 357–368, 2004.

107. Kherrouche Z, De Launoit Y, Monte D. The NRF-1/�PAL tran-scription factor regulates human E2F6 promoter activity. Biochem

J 383: 529–536, 2004.108. Kravchenko JE, Rogozin IB, Koonin EV, Chumakov PM. Tran-

scription of mammalian messenger RNAs by a nuclear RNA poly-merase of mitochondrial origin. Nature 436: 735–739, 2005.

109. Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS,

Clayton DA, Vissing J, Khrapko K. Recombination of humanmitochondrial DNA. Science 304: 981–981, 2004.

110. Kressler D, Schreiber SN, Knutti D, Kralli A. The PGC-1-related protein PERC is a selective coactivator of estrogen receptor�. J Biol Chem 277: 13918–13925, 2002.

111. Kruse B, Narasimhan N, Attardi G. Termination of transcriptionin human mitochondria: identification and purification of a DNAbinding protein factor that promotes termination. Cell 58: 391–397,1989.

112. LaMarco KL, McKnight SL. Purification of a set of cellularpolypeptides that bind to the purine-rich cis-regulatory element ofherpes simplex virus immediate early genes. Genes Dev 3: 1372–1383, 1989.

113. Larsson NG, Wang JM, Wilhelmsson H, Oldfors A, Rustin P,

Lewandoski M, Barsh GS, Clayton DA. Mitochondrial transcrip-tion factor A is necessary for mtDNA maintenance and embryogen-esis in mice. Nature Genet 18: 231–236, 1998.

114. Leary SC, Shoubridge EA. Mitochondrial biogenesis: which partof “NO” do we understand? Bioessays 25: 538–541, 2003.

115. Lee SS, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL,

Fernandez-Salguero PM, Westphal H, Gonzalez FJ. Targeteddisruption of the alpha isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of thepleiotropic effects of peroxisome proliferators. Mol Cell Biol 15:3012–3022, 1995.

116. Lehman JJ, Barger PM, Kovacs A, Saffitz JE, Medeiros DM,

Kelly DP. Peroxisome proliferator-activated receptor � coactiva-tor-1 promotes cardiac mitochondrial biogenesis. J Clin Invest 106:847–856, 2000.

117. Lelliott CJ, Medina-Gomez G, Petrovic N, Kis A, Feldmann

HM, Bjursell M, Parker N, Curtis K, Campbell M, Hu P, Zhang

D, Litwin SE, Zaha VG, Fountain KT, Boudina S, Jimenez-

Linan M, Blount M, Lopez M, Meirhaeghe A, Bohlooly Y,

Storlien L, Stromstedt M, Snaith M, Oresic M, Abel ED, Can-

non B, Vidal-Puig A. Ablation of PGC-1� results in defective

mitochondrial activity, thermogenesis, hepatic function, cardiacperformance. PLoS Biol 4: 2042–2056, 2006.

118. Leone TC, Lehman JJ, Finck BN, Schaeffer PJ, Wende AR,

Boudina S, Courtois M, Wozniak DF, Sambandam N, Bernal-

Mizrachi C, Chen Z, Holloszy JO, Medeiros DM, Schmidt RE,

Saffitz JE, Abel ED, Semenkovich CF, Kelly DP. PGC1� defi-ciency causes multi-system energy metabolic derangements: mus-cle dysfunction, abnormal weight control and hepatic steatosis.PLoS Biol 3: 0672–0687, 2005.

119. Leone TC, Weinheimer CJ, Kelly DP. A critical role for theperoxisome proliferator-activated receptor alpha (PPARalpha) inthe cellular fasting response: the PPARalpha-null mouse as a modelof fatty acid oxidation disorders. Proc Natl Acad Sci USA 96:7473–7478, 1999.

120. Lezza AMS, Pesce V, Cormio A, Fracasso F, Vecchiet J, Fel-

zani G, Cantatore P, Gadaleta MN. Increased expression ofmitochondrial transcription factor A and nuclear respiratory fac-tor-1 in skeletal muscle from aged human subjects. FEBS Lett 501:74–78, 2001.

121. Li F, Wang Y, Zeller KI, Potter JJ, Wonsey DR, O’Donnell KA,

Kim JW, Yustein JT, Lee LA, Dang CV. Myc stimulates nuclearlyencoded mitochondrial genes and mitochondrial biogenesis. Mol

Cell Biol 25: 6225–6234, 2005.122. Li H, Wang JM, Wilhelmsson H, Hansson A, Thoren P, Duffy J,

Rustin P, Larsson NG. Genetic modification of survival in tissue-specific knockout mice with mitochondrial cardiomyopathy. Proc

Natl Acad Sci USA 97: 3467–3472, 2000.123. Li K, Neufer PD, Williams RS. Nuclear responses to depletion of

mitochondrial DNA in human cells. Am J Physiol Cell Physiol 269:C1265–C1270, 1995.

124. Li R, Hodny Z, Luciakova K, Barath P, Nelson BD. Sp1 activatesand inhibits transcription from separate elements in the proximalpromoter of the human adenine nucleotide translocase 2 (ANT2)gene. J Biol Chem 271: 18925–18930, 1996.

125. Li R, Luciakova K, Nelson BD. Expression of the human cyto-chrome c1 gene is controlled through multiple Sp1-binding sites andan initiator region. Eur J Biochem 241: 649–656, 1996.

126. Lin J, Handschin C, Spiegelman BM. Metabolic control throughthe PGC-1 family of transcriptional coactivators. Cell Metab 1:361–370, 2005.

127. Lin J, Puigserver P, Donovan J, Tarr P, Spiegelman BM.

PGC-1�: a novel PGC-1-related transcription coactivator associatedwith host cell factor. J Biol Chem 277: 1645–1648, 2002.

128. Lin J, Wu P, Tarr PT, Lindenberg KS, St-Pierre J, Zhang C,

Mootha VK, Jager S, Vianna CR, Reznick RM, Cui L, Manieri

M, Donovan MX, Wu Z, Cooper MP, Fan ML, Rohas LM, Za-

vacki AM, Cinti S, Shulman GI, Lowell BB, Krainc D,

Spiegelman BM. Defects in adaptive energy metabolism withCNS-linked hyperactivity in PGC-1� null mice. Cell 119: 121–135,2004.

129. Linder T, Park CB, sin-Cayuela J, Pellegrini M, Larsson NG,

Falkenberg M, Samuelsson T, Gustafsson CM. A family ofputative transcription termination factors shared amongst metazo-ans and plants. Curr Genet 48: 265–269, 2005.

130. Liu Z, Butow RA. Mitochondrial retrograde signaling. Annu Rev

Genet 40: 159–185, 2006.131. Lowell BB, Shulman GI. Mitochondrial dysfunction and type 2

diabetes. Science 307: 384–387, 2005.132. Luo J, Sladek R, Carrier J, Bader J, Richard D, Giguere V.

Reduced fat mass in mice lacking orphan nuclear receptor estro-gen-related receptor �. Mol Biol Cell 23: 7947–7956, 2003.

133. Martin M, Cho J, Cesare AJ, Griffith JD, Attardi G. Termina-tion factor-mediated DNA loop between termination and initiationsites drives mitochondrial rRNA synthesis. Cell 123: 1227–1240,2005.

134. Masters BS, Stohl LL, Clayton DA. Yeast mitochondrial RNApolymerase is homologous to those encoded by bacteriophages T3and T7. Cell 51: 89–99, 1987.

135. Matsushima Y, Adan C, Garesse R, Kaguni LS. Drosophila

mitochondrial transcription factor B1 modulates mitochondrialtranslation but not transcription or DNA copy number in Schneidercells. J Biol Chem 280: 16815–16820, 2005.



136. Matsushima Y, Garesse R, Kaguni LS. Drosophila mitochondrialtranscription factor B2 regulates mitochondrial DNA copy numberand transcription in Schneider cells. J Biol Chem 279: 26900–26905,2004.

137. Mayr B, Montminy M. Transcriptional regulation by the phospho-rylation-dependent factor CREB. Nat Rev Mol Cell Biol 2: 599–609,2001.

138. McCulloch V, Seidel-Rogol BL, Shadel GS. A human mitochon-drial transcription factor is related to RNA adenine methyltrans-ferases and binds S-adenosylmethionine. Mol Cell Biol 22: 1116–1125, 2002.

139. McCulloch V, Shadel GS. Human mitochondrial transcriptionfactor B1 interacts with the C-terminal activation region of h-mtTFA and stimulates transcription independently of its RNAmethyltransferase activity. Mol Cell Biol 23: 5816–5824, 2003.

140. Meirhaeghe A, Crowley V, Lenaghan C, Lelliott C, Green K,

Stewart A, Hart K, Schinner S, Sethi JK, Yeo G, Brand MD,

Cortright RN, O’Rahilly S, Montague C, Vidal-Puig AJ. Char-acterization of the human, mouse and rat PGC1� (peroxisome-proliferator-activated receptor-gamma co-activator 1�) gene invitro and in vivo. Biochem J 373: 155–165, 2003.

141. Miranda S, Foncea R, Guerreo J, Leighton F. Oxidative stressand upregulation of mitochondrial biogenesis genes in mitochon-drial DNA-depleted HeLa cells. Biochem Biophys Res Commun

258: 44–49, 1999.142. Monsalve M, Wu Z, Adelmant G, Puigserver P, Fan M,

Spiegelman BM. Direct coupling of transcription and mRNA pro-cessing through the thermogenic coactivator PGC-1. Mol Cell 6:307–316, 2000.

143. Mootha VK, Handschin C, Arlow D, Xie XH, St Pierre J, Sihag

S, Yang WL, Altshuler D, Puigserver P, Patterson N, Willy PJ,

Schulman IG, Heyman RA, Lander ES, Spiegelman BM. Err�and Gabpa/b specify PGC-1�-dependent oxidative phosphorylationgene expression that is altered in diabetic muscle. Proc Natl Acad

Sci USA 101: 6570–6575, 2004.144. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Si-

hag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M,

Laurila E, Houstis N, Daly MJ, Patterson N, Mesirov JP,

Golub TR, Tamayo P, Spiegelman B, Lander ES, Hirschhorn

JN, Altshuler D, Groop LC. PGC-1�-responsive genes involved inoxidative phosphorylation are coordinately downregulated in hu-man diabetes. Nature Genet 34: 267–273, 2003.

145. Moraes CT, Ricci E, Bonilla E, DiMauro S, Schon EA. Themitochondrial tRNALeu(UUR) mutation in mitochondrial encephalo-myopathy, lactic acidosis, strokelike episodes (MELAS): genetic,biochemical, and morphological correlations in skeletal muscle.Am J Hum Genet 50: 934–949, 1992.

146. Morrish F, Giedt C, Hockenbery D. C-MYC apoptotic function ismediated by NRF-1 target genes. Genes Dev 17: 240–255, 2003.

147. Murakami T, Shimomura Y, Yosimura A, Sokabe M, Fujitsuka

N. Induction of nuclear respiratory factor-1 expression by an acutebout of exercise in rat muscle. Biochim Biophys Acta 1381: 113–122, 1998.

148. Nass MMK, Nass S, Afzelius BA. The general occurence ofmitochondrial DNA. Exp Cell Res 37: 190–200, 1965.

149. Nie F, Wong-Riley, M. Nuclear respiratory factor-2 subunit pro-tein: Correlation with cytochrome oxydase and regulation by func-tional activity in the monkey primary visual cortex. J Comp Neurol

404: 310–320, 1999.150. Nisoli E, Clementi E, Paolucci C, Cozzi V, Tonello C, Sciorati

C, Bracale R, Valerio A, Francolini M, Moncada S, Carruba

MO. Mitochondrial biogenesis in mammals: the role of endogenousnitric oxide. Science 299: 896–899, 2003.

151. Nisoli E, Falcone S, Tonello C, Cozzi V, Palomba L, Fiorani M,

Pisconti A, Brunelli S, Cardile A, Francolini M, Cantoni O,

Carruba MO, Moncada S, Clementi E. Mitochondrial biogenesisby NO yields functionally active mitochondria in mammals. Proc

Natl Acad Sci USA 101: 16507–16512, 2004.152. Ojala D, Montoya J, Attardi G. tRNA punctuation model of RNA

processing in human mitochondria. Nature 290: 470–474, 1981.153. Ojuka EO. Role of calcium and AMP kinase in the regulation of

mitochondrial biogenesis and GLUT4 levels in muscle. Proc Nutr

Soc 63: 275–278, 2004.

154. Ojuka EO, Jones TE, Han DH, Chen M, Holloszy JO. RaisingCa2� in L6 myotubes mimics effects of exercise on mitochondrialbiogenesis in muscle. FASEB J 17: 675–681, 2003.

155. Ojuka EO, Jones TE, Han DH, Chen M, Wamhoff BR, Sturek

M, Holloszy JO. Intermittent increases in cytosolic Ca2� stimulatemitochondrial biogenesis in muscle cells. Am J Physiol Endocrinol

Metab 283: E1040–E1045, 2002.156. Ongwijitwat S, Liang HL, Graboyes EM, Wong-Riley MT. Nu-

clear respiratory factor 2 senses changing cellular energy demandsand its silencing down-regulates cytochrome oxidase and othertarget gene mRNAs. Gene 374: 39–49, 2006.

157. Ongwijitwat S, Wong-Riley MT. Is nuclear respiratory factor 2 amaster transcriptional coordinator for all ten nuclear-encoded cy-tochrome c oxidase subunits in neurons? Gene 360: 65–77, 2005.

158. Paluh JL, Clayton DA. A functional dominant mutation in Schizo-

saccharomyces pombe RNase MRP RNA affects nuclear RNA pro-cessing and requires the mitochondrial-associated nuclear muta-tion ptp1–1 for viability. EMBO J 15: 4723–4733, 1996.

159. Parisi MA, Clayton DA. Similarity of human mitochondrial tran-scription factor 1 to high mobility group proteins. Science 252:965–969, 1991.

160. Parisi MA, Xu B, Clayton DA. A human mitochondrial transcrip-tional activator can functionally replace a yeast mitochondrialHMG-box protein both in vivo and in vitro. Mol Cell Biol 13:1951–1961, 1993.

161. Patti ME. Gene expression in humans with diabetes and predia-betes: what have we learned about diabetes pathophysiology? Curr

Opin Clin Nutr Metab Care 7: 383–390, 2004.162. Patti ME, Butte AJ, Crunkhorn S, Cusi K, Berria R, Kashyap

S, Miyazaki Y, Kohane I, Costello M, Saccone R, Landaker EJ,

Goldfine AB, Mun E, DeFronzo R, Finlayson J, Kahn CR,

Mandarino LJ. Coordinated reduction of genes of oxidative me-tabolism in humans with insulin resistance and diabetes: potentialrole of PGC1 and NRF1. Proc Natl Acad Sci USA 100: 8466–8471,2003.

163. Petersen KF, Befroy D, Dufour S, Dziura J, Ariyan C, Roth-

man DL, DiPietro L, Cline GW, Shulman GI. Mitochondrialdysfunction in the elderly: possible role in insulin resistance. Sci-

ence 300: 1140–1142, 2003.164. Peterson KF, Dufour S, Befroy D, Garcia R, Shulman GI.

Impaired mitochondrial activity in the insulin-resistant offspring ofpatients with type 2 diabetes. N Engl J Med 350: 664–671, 2004.

165. Pham XH, Farge G, Shi Y, Gaspari M, Gustafsson CM, Falk-

enberg M. Conserved sequence box II directs transcription termi-nation and primer formation in mitochondria. J Biol Chem 281:24647–24652, 2006.

166. Piantadosi CA, Suliman HB. Mitochondrial transcription factor Ainduction by redox activation of nuclear respiratory factor 1. J Biol

Chem 281: 324–333, 2006.167. Piko L, Taylor KD. Amounts of mitochondrial DNA and abun-

dance of some mitochondrial gene transcripts in early mouse em-bryos. Dev Biol 123: 364–374, 1987.

168. Poulton J, Morten K, Freeman-Emmerson C, Potter C, Sewry

C, Dubowitz V, Kidd H, Stephenson J, Whitehouse W, Hansen

FJ, Parisi M, Brown G. Deficiency of the human mitochondrialtranscription factor h-mtTFA in infantile mitochondrial myopathyis associated with mtDNA depletion. Hum Mol Genet 3: 1763–1769,1994.

169. Poyton RO, McEwen JE. Crosstalk between nuclear and mito-chondrial genomes. Annu Rev Biochem 65: 563–607, 1996.

170. Puigserver P. Tissue-specific regulation of metabolic pathwaysthrough the transcriptional coactivator PGC1-�. Int J Obesity 29:55–59, 2005.

171. Puigserver P, Wu Z, Park CW, Graves R, Wright M, Spiegelman

BM. A cold-inducible coactivator of nuclear receptors linked to adap-tive thermogenesis. Cell 92: 829–839, 1998.

172. Rantanen A, Gaspari M, Falkenberg M, Gustafsson CM, Lars-

son NG. Characterization of the mouse genes for mitochondrialtranscription factors B1 and B2. Mammalian Genome 14: 1–6,2003.

173. Rantanen A, Jansson M, Oldfors A, Larsson NG. Downregula-tion of Tfam and mtDNA copy number during mammalian spermat-ogenesis. Mammalian Genome 12: 787–792, 2001.



174. Ren B, Dynlacht BD. Use of chromatin immunoprecipitation as-says in genome-wide location analysis of mammalian transcriptionfactors. Methods Enzymol 376: 304–315, 2004.

175. Ricquier D, Bouillaud F. Mitochondrial uncoupling proteins:from mitochondria to the regulation of energy balance. J Physiol

529: 3–10, 2000.176. Ristevski S, O’Leary DA, Thornell AP, Owen MJ, Kola I,

Hertzog PJ. The ETS transcription factor GABP� is essential forearly embryogenesis. Mol Cell Biol 24: 5844–5849, 2004.

177. Rothermel BA, Thornton JL, Butow RA. Rtg3p, a basic helix-loop-helix/leucine zipper protein that functions in mitochondrial-induced changes in gene expression, contains independent activa-tion domains. J Biol Chem 272: 19801–19807, 1997.

178. Russell AP, Hesselink MKC, Lo SK, Schrauwen P. Regulationof metabolic transcriptional co-activators and transcription factorswith acute exercise. FASEB J 19: NIL467–NIL486, 2005.

179. Satoh M, Kuroiwa T. Organization of multiple nucleoids and DNAmolecules in mitochondria of a human cell. Exp Cell Res 196:137–140, 1991.

180. Savagner F, Mirebeau D, Jacques C, Guyetant S, Morgan C,

Franc B, Reynier P, Malthiery Y. PGC-1-related coactivator andtargets are upregulated in thyroid oncocytoma. Biochem Biophys

Res Commun 310: 779–784, 2003.181. Scarpulla RC. Nuclear control of respiratory chain expression in

mammalian cells. J Bioenerg Biomembr 29: 109–119, 1997.182. Scarpulla RC. Nuclear transcription factors in cytochrome c and

cytochrome oxidase expression. In: Frontiers of Cellular Bioener-

getics: Molecular Biology, Biochemistry, Physiopathology, editedby Papa S, Guerrieri F, Tager JM. New York: Plenum, 1999.

183. Scarpulla RC. Nuclear activators and coactivators in mammalianmitochondrial biogenesis. Biochim Biophys Acta 1576: 1–14, 2002.

184. Scarpulla RC. Transcriptional activators and coactivators in thenuclear control of mitochondrial function in mammalian cells.Gene 286: 81–89, 2002.

185. Scarpulla RC. Nuclear control of respiratory gene expression inmammalian cells. J Cell Biochem 97: 673–683, 2006.

186. Scarpulla RC, Nye SH. Functional expression of rat cytochromec in Saccharomyces cerevisiae. Proc Natl Acad Sci USA 83: 6352–6356, 1986.

187. Schaefer L, Engman H, Miller JB. Coding sequence, chromo-somal localization, expression pattern of Nrf-1: the mouse ho-molog of Drosophila erect wing. Genome 11: 104–110, 2000.

188. Schaeffer L, Duclert N, Huchet-Dymanus M, Changeux JP.

Implication of a multisubunit Ets-related transcription factor insynaptic expression of the nicotinic acetylcholine receptor. EMBO

J 17: 3078–3090, 1998.189. Schon EA, Manfredi G. Neuronal degeneration and mitochondrial

dysfunction. J Clin Invest 111: 303–312, 2003.190. Schreiber SN, Emter R, Hock MB, Knutti D, Cardenas J,

Podvinec M, Oakeley EJ, Kralli A. The estrogen-related receptor� (ERR�) functions in PPARgamma coactivator 1� (PGC-1�)-in-duced mitochondrial biogenesis. Proc Natl Acad Sci USA 101:6472–6477, 2004.

191. Schubot FD, Chen CJ, Rose JP, Dailey TA, Dailey HA, Wang

BC. Crystal structure of the transcription factor sc-mtTFB offersinsights into mitochondrial transcription. Protein Sci 10: 1980–1988, 2001.

192. Schwartz M, Vissing J. Paternal inheritance of mitochondrialDNA. N Engl J Med 347: 576–580, 2002.

193. Seelan RS, Gopalakrishnan L, Scarpulla RC, Grossman LI.

Cytochrome c oxidase subunit VIIa liver isoform: characterizationand identification of promoter elements in the bovine gene. J Biol

Chem 271: 2112–2120, 1996.194. Seelan RS, Grossman LI. Structural organization and promoter

analysis of the bovine cytochrome c oxidase subunit VIIc gene: afunctional role for YY1. J Biol Chem 272: 10175–10181, 1997.

195. Seidel-Rogol BL, McCulloch V, Shadel GS. Human mitochon-drial transcription factor B1 methylates ribosomal RNA at a con-served stem-loop. Nat Genet 33: 23–24, 2003.

196. Shadel GS, Clayton DA. A Saccharomyces cerevisiae mitochon-drial transcription factor, sc-mtTFB, shares features with sigmafactors but is functionally distinct. Mol Cell Biol 15: 2101–2108,1995.

197. Shadel GS, Clayton DA. Mitochondrial DNA maintenance in ver-tebrates. Annu Rev Biochem 66: 409–435, 1997.

198. Shoubridge EA. Mitochondrial DNA segregation in the developingembryo. Hum Reprod 15: 229–234, 2000.

199. Shoubridge EA. Nuclear genetic defects of oxidative phosphory-lation. Hum Mol Genet 10: 2277–2284, 2001.

200. Shyjan AW, Butow RA. Intracellular dialogue. Curr Biol 3: 398–400, 1993.

201. Silva JP, Kohler M, Graff C, Oldfors A, Magnuson MA, Berg-

gren PO, Larsson NG. Impaired insulin secretion and beta-cellloss in tissue-specific knockout mice with mitochondrial diabetes.Nat Genet 26: 336–340, 2000.

202. St. Pierre J, Lin J, Krauss S, Tarr PT, Yang RJ, Newgard CB,

Spiegelman BM. Bioenergetic analysis of peroxisome proliferator-activated receptor gamma coactivators 1� and 1� (PGC-1� andPGC-1�) in muscle cells. J Biol Chem 278: 26597–26603, 2003.

203. Suliman HB, Carraway MS, Welty-Wolf KE, Whorton AR, Pi-

antados CA. Lipopolysaccharide stimulates mitochondrial biogen-esis via activation of nuclear respiratory factor-1. J Biol Chem 278:41510–41518, 2003.

204. Sunesen M, Huchet-Dymanus M, Christensen MO, Changeux

JP. Phosphorylation-elicited quaternary changes of GA bindingprotein in transcriptional activation. Mol Cell Biol 23: 8008–8018,2003.

205. Suomalainen A, Kaukonen J. Diseases caused by nuclear genesaffecting mtDNA stability. Am J Med Genet 106: 53–61, 2001.

206. Takahashi Y, Kako K, Arai H, Oishi T, Inada Y, Tkehara A,

Fukamizu A, Munekata E. Characterization and identification ofpromoter elements in the mouse COX17 gene. Biochim Biophys

Acta 1574: 359–364, 2002.207. Terada S, Goto M, Kato M, Kawanaka K, Shimokawa T, Ta-

bata I. Effects of low-intensity prolonged exercise on PGC-1mRNA expression in rat epitrochlearis muscle. Biochem Biophys

Res Commun 296: 350–354, 2002.208. Terada S, Kawanaka K, Goto M, Shimokawa T, Tabata I.

Effects of high-intensity intermittent swimming on PGC-1alphaprotein expression in rat skeletal muscle. Acta Physiol Scand 184:59–65, 2005.

209. Terada S, Tabata I. Effects of acute bouts of running and swim-ming exercise on PGC-1� protein expression in rat epitrochlearisand soleus muscle. Am J Physiol Endocrinol Metab 286: E208–E216, 2004.

210. Thompson CC, Brown TA, McKnight SL. Convergence of Ets-and notch-related structural motifs in a heteromeric DNA bindingcomplex. Science 253: 762–768, 1991.

211. Tiranti V, Savoia A, Forti F, D’Apolito MF, Centra M, Racchi

M, Zeviani M. Identification of the gene encoding the humanmitochondrial RNA polymerase (h-mtRPOL) by cyberscreening ofthe expressed sequence tags database. Hum Mol Genet 6: 615–625,1997.

212. Topper JN, Clayton DA. Identification of transcriptional regula-tory elements in human mitochondrial DNA by linker substitutionanalysis. Mol Cell Biol 9: 1200–1211, 1989.

213. Trifunovic A, Wredenberg A, Falkenberg M, Spelbrink JN,

Rovio AT, Bruder CE, Bohlooly Y, Gidlof S, Oldfors A, Wibom

R, Tornell J, Jacobs HT, Larsson NG. Premature ageing in miceexpressing defective mitochondrial DNA polymerase. Nature 429:417–423, 2004.

214. Truscott KN, Brandner K, Pfanner N. Mechanisms of proteinimport into mitochondria. Curr Biol 13: R326–R337, 2003.

215. Tyynismaa H, Sembongi H, Bokori-Brown M, Granycome C,

Ashley N, Poulton J, Jalanko A, Spelbrink JN, Holt IJ, Suo-

malainen A. Twinkle helicase is essential for mtDNA maintenanceand regulates mtDNA copy number. Hum Mol Genet 13: 3219–3227,2004.

216. Uldry M, Yang W, St-Pierre J, Lin J, Seale P, Spiegelman BM.

Complementary action of the PGC-1 coactivators in mitochondrialbiogenesis and brown fat differentiation. Cell Metab 3: 333–341,2006.

217. Valcarce C, Navarrete RM, Encabo P, Loeches E, Satrustegui

J, Cuezva JM. Postnatal development of rat liver mitochondrialfunctions. J Biol Chem 263: 7767–7775, 1988.



218. Vega RB, Huss JM, Kelly DP. The coactivator PGC-1 cooperateswith peroxisome proliferator-activated receptor � in transcrip-tional control of nuclear genes encoding mitochondrial fatty acidoxidation enzymes. Mol Cell Biol 20: 1868–1876, 2000.

219. Vega RB, Kelly DP. A role for estrogen-related receptor � in thecontrol of mitochondrial fatty acid �-oxidation during brown adi-pocyte differentiation. J Biol Chem 272: 31693–31699, 2004.

220. Vercauteren K, Pasko RA, Gleyzer N, Marino VM, Scarpulla

RC. PGC-1-related coactivator (PRC): immediate early expressionand characterization of a CREB/NRF-1 binding domain associatedwith cytochrome c promoter occupancy and respiratory growth.Mol Cell Biol 26: 7409–7419, 2006.

221. Villena JA, Hock MB, Chang WY, Barcas JE, Giguere V, Kralli

A. Orphan nuclear receptor estrogen-related receptor alpha is es-sential for adaptive thermogenesis. Proc Natl Acad Sci USA 104:1418–1423, 2007.

222. Virbasius CA, Virbasius JV, Scarpulla RC. NRF-1, an activatorinvolved in nuclear-mitochondrial interactions, utilizes a new DNA-binding domain conserved in a family of developmental regulators.Genes Dev 7: 2431–2445, 1993.

223. Virbasius JV, Scarpulla RC. Transcriptional activation throughETS domain binding sites in the cytochrome c oxidase subunit IVgene. Mol Cell Biol 11: 5631–5638, 1991.

224. Virbasius JV, Scarpulla RC. Activation of the human mitochon-drial transcription factor A gene by nuclear respiratory factors: apotential regulatory link between nuclear and mitochondrial geneexpression in organelle biogenesis. Proc Natl Acad Sci USA 91:1309–1313, 1994.

225. Virbasius JV, Virbasius CA, Scarpulla RC. Identity of GABPwith NRF-2, a multisubunit activator of cytochrome oxidase ex-pression, reveals a cellular role for an ETS domain activator of viralpromoters. Genes Dev 7: 380–392, 1993.

226. Wallace DC. Diseases of the mitochondrial DNA. Annu Rev Bio-

chem 61: 1175–1212, 1992.227. Wallace DC. A mitochondrial paradigm of metabolic and degen-

erative diseases, aging, cancer: a dawn for evolutionary medicine.Annu Rev Genet 39: 359–407, 2005.

228. Wan B, Moreadith RW. Structural characterization and regulatoryelement analysis of the heart isoform of cytochrome c oxidase VIa.J Biol Chem 270: 26433–26440, 1995.

229. Wang H, Morais R. Up-regulation of nuclear genes in response toinhibition of mitochondrial DNA expression in chicken cells. Bio-

chim Biophys Acta Gene Struct Expression 1352: 325–334, 1997.230. Wang Y, Bogenhagen DF. Human mitochondrial DNA nucleoids

are linked to protein folding machinery and metabolic enzymes atthe mitochondrial inner membrane. J Biol Chem 281: 25791–25802,2006.

231. Wang Z, Cotney J, Shadel GS. Human mitochondrial ribosomalprotein MRPL12 interacts directly with mitochondrial RNA poly-

merase to modulate mitochondrial gene expression. J Biol Chem

282: 12610–12618, 2007.232. Williams RS, Garcia-Moll M, Mellor J, Salmons S, Harlan W.

Adaptation of skeletal muscle to increased contractile activity.J Biol Chem 262: 2764–2767, 1987.

233. Wright DC, Han DH, Garcia-Roves PM, Geiger PC, Jones TE,

Holloszy JO. Exercise-induced mitochondrial biogenesis beginsbefore the increase in muscle PGC-1� expression. J Biol Chem 282:194–199, 2007.

234. Wrutniak C, Cassar-Malek I, Marchal S, Rascle A, Heusser S,

Keller JM, Flechon J, Dauca M, Samarut J, Ghysdael J, Ca-

bello G. A 43-kDa protein related to c-Erb A alpha1 is located inthe mitochondrial matrix of rat liver. J Biol Chem 270: 16347–16354, 1995.

235. Wu H, Kanatous SB, Thurmond FA, Gallardo T, Isotani E,

Bassel-Duby R, Williams RS. Regulation of mitochondrial bio-genesis in skeletal muscle by CaMK. Science 296: 349–352, 2002.

236. Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G,

Mootha V, Troy A, Cinti S, Lowell B, Scarpulla RC,

Spiegelman BM. Mechanisms controlling mitochondrial biogene-sis and function through the thermogenic coactivator PGC-1. Cell

98: 115–124, 1999.237. Xu B, Clayton DA. Assignment of a yeast protein necessary for

mitochondrial transcription initiation. Nucleic Acids Res 20: 1053–1059, 1992.

238. Yang SJ, Liang HL, Ning G, Wong-Riley, MTT. Ultrastructuralstudy of depolarization-induced translocation of NRF-2 transcrip-tion factor in cultured rat visual cortical neurons. Eur J Neurosci

19: 1153–1162, 2004.239. Yoon JC, Puigserver P, Chen GX, Donovan J, Wu ZD, Rhee J,

Adelmant G, Stafford J, Kahn CR, Granner DK, Newgard CB,

Spiegelman BM. Control of hepatic gluconeogenesis through thetranscriptional coactivator PGC-1. Nature 413: 131–138, 2001.

240. Zaid A, Li R, Luciakova K, Barath P, Nery S, Nelson BD. On therole of the general transcription factor Sp1 in the activation andrepression of diverse mammalian oxidative phosphorylation genes.J Bioenerg Biomembr 31: 129–135, 1999.

241. Zhang C, Wong-Riley MT. Depolarizing stimulation upregulatesGA-binding protein in neurons: a transcription factor involved inthe bigenomic expression of cytochrome oxidase subunits. Eur

J Neurosci 12: 1013–1023, 2000.242. Zitomer RS, Lowry CV. Regulation of gene expression by oxygen

in Saccharomyces cerevisiae. Microbiol Rev 56: 1–11, 1992.243. Zong HH, Ren JM, Young LH, Pypaert M, Mu J, Birnbaum MJ,

Shulman GI. AMP kinase is required for mitochondrial biogenesisin skeletal muscle in response to chronic energy deprivation. Proc

Natl Acad Sci USA 99: 15983–15987, 2002.



88:611-638, 2008. doi:10.1152/physrev.00025.2007 Physiol RevRichard C. Scarpulla

You might find this additional information useful...

242 articles, 127 of which you can access free at: This article cites http://physrev.physiology.org/cgi/content/full/88/2/611#BIBL

36 other HighWire hosted articles, the first 5 are: This article has been cited by

[PDF] [Full Text] [Abstract]

, February 19, 2010; 285 (8): 5196-5203. J. Biol. Chem.Y. Dang and B. R. Green

TranscriptionLong Transcripts from Dinoflagellate Chloroplast Minicircles Suggest "Rolling Circle"

[PDF] [Full Text] [Abstract], March 15, 2010; 588 (6): 1011-1022. J. Physiol.

J. P. Little, A. Safdar, G. P. Wilkin, M. A. Tarnopolsky and M. J. Gibala biogenesis in human skeletal muscle: potential mechanisms

A practical model of low-volume high-intensity interval training induces mitochondrial

[PDF] [Full Text] [Abstract], March 19, 2010; 106 (5): 911-919. Circ. Res.

Q. Yang P. Wang, J. Liu, Y. Li, S. Wu, J. Luo, H. Yang, R. Subbiah, J. Chatham, O. Zhelyabovska and

Regulator for Mitochondrial Protection and Biogenesis in Adult HeartPeroxisome Proliferator-Activated Receptor {delta} Is an Essential Transcriptional

[PDF] [Full Text] [Abstract], May 1, 2010; 24 (5): 1525-1531. FASEB J

Wenzel and R. Paus B. Poeggeler, J. Knuever, E. Gaspar, T. Biro, M. Klinger, E. Bodo, R. J. Wiesner, B. E.

Thyrotropin powers human mitochondria

[PDF] [Full Text] [Abstract], May 15, 2010; 0 (2010): icq038v2-icq038. Integr. Comp. Biol.

K. M. O'Brien and I. A. Mueller The Unique Mitochondrial Form and Function of Antarctic Channichthyid Icefishes

on the following topics: http://highwire.stanford.edu/lists/artbytopic.dtlcan be found at Medline items on this article's topics

Cell Biology .. Cytoplasm Biochemistry .. Mitochondria Biogenesis Biochemistry .. Circular DNA Biochemistry .. Electron Transport Chain Biochemistry .. Mitochondrial RNA Biophysics .. Mitochondrial DNA

including high-resolution figures, can be found at: Updated information and services http://physrev.physiology.org/cgi/content/full/88/2/611

can be found at: Physiological Reviewsabout Additional material and information http://www.the-aps.org/publications/prv

This information is current as of May 31, 2010 .

http://www.the-aps.org/.website at MD 20814-3991. Copyright © 2005 by the American Physiological Society. ISSN: 0031-9333, ESSN: 1522-1210. Visit ourpublished quarterly in January, April, July, and October by the American Physiological Society, 9650 Rockville Pike, Bethesda

provides state of the art coverage of timely issues in the physiological and biomedical sciences. It isPhysiological Reviews

http://physrev.physiology.org/cgi/content/full/88/2/611#BIBL

http://icb.oxfordjournals.org/cgi/content/abstract/icq038v2

http://icb.oxfordjournals.org/cgi/content/full/icq038v2

http://icb.oxfordjournals.org/cgi/reprint/icq038v2

http://www.fasebj.org/cgi/content/abstract/24/5/1525

http://www.fasebj.org/cgi/content/full/24/5/1525

http://www.fasebj.org/cgi/reprint/24/5/1525

http://circres.ahajournals.org/cgi/content/abstract/106/5/911

http://circres.ahajournals.org/cgi/content/full/106/5/911

http://circres.ahajournals.org/cgi/reprint/106/5/911

http://jp.physoc.org/cgi/content/abstract/588/6/1011

http://jp.physoc.org/cgi/content/full/588/6/1011

http://jp.physoc.org/cgi/reprint/588/6/1011

http://www.jbc.org/cgi/content/abstract/285/8/5196

http://www.jbc.org/cgi/content/full/285/8/5196

http://www.jbc.org/cgi/reprint/285/8/5196

http://highwire.stanford.edu/lists/artbytopic.dtl

http://physrev.physiology.org/cgi/content/full/88/2/611

http://www.the-aps.org/publications/prv

http://www.the-aps.org/

transcriptional paradigms in mammalian mitochondrial ... · transcriptional paradigms in mammalian...

Documents