SIRISHA BANDARU

DECOUVERTE DES MECANISMES MOLECULAIRES DE REGULATION

DE LA MORPHOLOGIE MITOCHONDRIALE CHEZ LES MAMMIFÈRES : LE

RÔLE CONTRÔLEUR DE LA PROTEASE RHOMBOÏDE PARL

Mémoire présenté à la Faculté des études supérieures de l'Université Laval

dans le cadre du programme de Maîtrise en Neurobiologie pour l'obtention du grade de Maître Es Sciences (M.Se.)

FACULTE DE MEDECINE

UNIVERSITÉ LAVAL QUÉBEC

2007

© Sirisha Bandaru, 2007

Résumé

Pendant l'évolution des métazoaires, les mécanismes moléculaires qui contrôlent le

remodelage de la morphologie des mitochondries sont changés. Ceci a permis le

recrutement de cette organelle dans le développement, la signalisation du calcium, et

l'apoptose. Une protéine centrale au remodelage des mitochondries est la protéase

rhomboïde PARL (presenilin-associated rhomboid-like), mais ces mécanismes de

régulation sont toujours inconnus.

Dans cette étude, nous démontrons que la phosphorylation et le clivage d'un domaine N-

terminal de PARL règlent la morphologie des mitochondries. Dans ce clivage, le domaine

C-terminal de PARL joue un rôle structural essentiel.

Nos résultats démontrent que la phosphorylation et le clivage de PARL ont un impact sur

la dynamique des mitochondries, ce qui nous fournit un modèle pour étudier l'évolution

moléculaire de la morphologie des mitochondries.

Abstract

Remodeling of mitochondria is a dynamic process coordinated by fusion and fission of

the inner and outer membranes of the organelle, mediated by a set of conserved proteins.

In metazoans, the molecular mechanism behind mitochondrial morphology has been

recruited to govern novel functions, such as development, calcium signalling, and

apoptosis, which suggests that novel mechanisms should exist to regulate the conserved

membrane fusion/fission machinery. Hère we show that phosphorylation and cleavage of

the vertebrate-specific Pbeta domain of the mammalian presenilin-associated rhomboid-

like (PARL) protéase can influence mitochondrial morphology. Phosphorylation of three

residues embedded in this domain, Ser-65, Thr-69, and Ser-70, impairs a cleavage at

position Ser(77)-Ala(78) that is required to initiate PARL-induced mitochondrial

fragmentation. Our findings reveal that PARL phosphorylation and cleavage impact

mitochondrial dynamics, providing a blueprint to study the molecular évolution of

mitochondrial morphology.

Avant-propos 1

Authors contributions in the accompanying paper entitled "Phosphorylation and cleavage of presenilin-associated rhomboid-like protein (PARL) promûtes changes in mitochondrial morphology" by Jeyaraju DV, Xu L, Letellier MC, Bandaru S, Zunino R, Berg EA, McBride HM, and Pellegrini L. Sirisha Bandaru:

Provided technical assistance in the Pellegrini lab, including: gênerai lab duties, tissue culture room maintenance, site-directed mutagenesis, compétent cell préparation, plasmid DNA préparation, cloning, PCR screening, preparative western blots, and more. Dannv Jevaraiu:

1. Performed the experiments in Figs. 1, 4, and 5C; 2. Generated the PARL mutants; 3. Purified (/) endogenous PARL from mitochondria isolated from human placenta, (//') wild

type PARL and mutant STS 65-69-70 AAA, from transfected HeLa cells (for subséquent masspec analysis) for Fig. 2.

4. Developed and characterized the anti PARL-C-Term antibody for immunoprecipitation and western blots analysis.

Lia un Xu: 1. Performed the experiments in Fig. 3 and 5A.

Marie Claude Letellier: 1. Generated ail the PARL A mutants shown in Fig. 4B, and the STS 65-69-70 AAA and

STS 65-69-70 DDD shown in Fig. 4A. 2. Isolated MAMP from HEK293 (for subséquent masspectrometry analysis).

Sirisha Bandaru and Rodolfo Zunino: Provided technical assistance in the McBride lab.

Dr. Eric Berg: Performed the mass spectrometric analysis shown in Fig 2.

Dr Heidi McBride: 1. Directed and supervised the experiments shown in Figs. 3, 5A and 5B; 2. Co-wrote the manuscript.

Dr. Luca Pellegrini: 1. Conceived, initiated, and directed the study; 2. Designed the experiments; 3. Wrote the manuscript.

Avant-propos 2

Given the collaborative nature of my research work, the introduction and methods section of this Mémoire hâve been written jointly with my MSc fellow and lab mate, Danny Jeyaraju.

3

Acknowledgements

First, I would like to warmly thank my supervisor, Dr. Luca Pellegrini, for his invaluable support, patience and confidence, for his precious and never lacking enthusiasm for research and for his continuous supervision in preparing and writing this thesis. I also want to thank him for having so strongly believed in me and provided me with insights which helped me solve many of the problems 1 encountered in my research. I also acknowledge him for assisting me financiaily during my masters and for his generosity which gave me an opportunity to attend the International meeting of the Canadian Association for Neuroscience.

I would like to thank Dr. Katalin Toth for helping me in many ways throughout the duration of my research; also little Emma and Léo for their sweet ways which always brings a smile.

I would like to thank my colleague, Danny Jeyaraju, for sharing his research with me, the great support and for always providing me with suggestions about my research. Also, for the constant help and teaching.

To my friends at the CRULRG, a big thank you for everything. Spécial thanks to Sonya, who was so kind as to help with ail the administrative work and for being so cheerful. Thanks to Jacqueline for being very helpful.

1 will always cherish the friendship of Stefania, Albert, and Daniel, without whom I wouldn't hâve had so much fun and who hâve supported me always. Madhuri, Ramachander, and Sidh for providing me with the comforts of a home and helping me to enjoy life in Québec.

Thank you to Jyothi, Deepika, Anil, Ella and Jia for their wonderful company. Colette and Guylaine for being a source of inspiration and for teaching me so much about life. I am also veiy glad to hâve been associated with Kathryn, although for only a short period.

My life in Québec would not be the same without my friends Rashi, Suresh, Murali, Arjun, Mridula, Mohini, Kiran, Venkat, Kalyan, Priya, Renuka and Manju who hâve always been there for me.

Finally, I would like to thank my parents and Sailu, who hâve supported me in every aspect of life and whose unconditional love and affections are most invaluable for me. Our dear little Pranavi has given a whole new meaning to life.

4

Abbreviations

PARL - Presenilin-associated rhomboid-like protein

MAMP - Mature mitochondrial protein

PACT - PARL C-terminal product

PDB - Protein Data Bank

Pcpl - Processing of cytochrome c peroxidase protein 1

GST - Glutathione S transferase

ATP - Adenosine triphosphate

OPA1 - Optic Atrophia 1

GTP - Guanosine triphosphate

DRP1 - Dynamin-related protein 1

hFis - human Mitochondrial fission 1 protein

Rbdlp - Rhomboid protein 1

IMS - Intermembrane Space

PKA - Protein Kinase A

PINK1 - PTEN-induced putative kinase protein 1

GSK - Glycogen synthase kinase

CK - Casein Kinase

MtDNA - Mitochondrial DNA

ROS - Reactive Oxygen Species

Rtg - Rétrograde régulation protein

TCA - Tricarboxylic Acid

CCCP - Carbonyl cyanide m-chlorophenylhydrazone

5

Table of Contents

1 Introduction Page 7

I. Mitochondrial Rhomboids: regulators of mitochondria morphology Page 7 remodelling

> Rhomboids: enzymatic activity, évolution, and structure > Mitochondrial rhomboids: novel components of the mitochondria morphology

remodelling machinery > The mammalian mitochondrial rhomboid protease PARL

II. Régulation of apoptosis by mitochondria morphology Page 11

> Apoptosis: rôle of cytochrome c in the intrinsic pathway > The mitochondria morphology machinery > Mechanisms of mitochondrial morphology régulation integrated to the apoptotic program > Rôle of the mitochondrial rhomboid protease PARL in mitochondria morphology and

apoptosis > Molecular mechanisms regulating PARL activity

2.1 Objectives - part I Page 18

2.2 Results and Discussion - part I Page 18

> Reprint of Jeyaraju, Xu, Letellier, Bandaru, Zunino, Berg, McBride, Pellegrini

PNAS (2006)

3.1 Objectives - part II Page 19

3.2 Results and Discussion - part II Page 19

4 Perspectives Page 21

5 Materials and methods Page 22

6 Bibliography Page 33

6

Introduction

1. Mitochondrial Rhomboids: regulators of mitochondria morphology remodelling

Rhomboids: enzymatic activity, évolution, and structure

Regulated intramembrane proteolysis (RIP), is a new paradigm of signal transduction, which

appears to be prominent in ail forms of life (Brown et al 2000). Under RIP, membrane-bound

proteins undergo site-specifïc proteolysis within one of their transmembrane hélices (TMH). This

processing releases a moiety of the protein that, through intracellular relocation or extracellular

release, typically exécutes the signalling function of the membrane-tethered precursor protein.

This mechanism influences processes as diverse as lipid metabolism, cellular differentiation, the

response to unfolded proteins, mitochondria morphology, and apoptosis.

Enzymatically, RIP requires proteases that, despite the water-excluding environment of the lipid

bilayer, somehow are able to hydrolyze their transmembrane substrates. This processing is thus

executed by a very unique class of proteases, named I-Clips, for intramembrane-cleaving

proteases (Weihofen & Martoglio 2003). To date, only three distinct I-Clip families hâve been

discovered. The first family, whose prototypic member is the human site-two protease (S2P) that

cleaves and activâtes sterol regulatory élément binding proteins (SREBPs), are a group of

metalloproteases (Brown & Goldstein 1999, Rawson et al 1997). A second family, whose

prototypic members are the presenilins (PSs) involved in cleavage of the amyloid-B precursor

protein (ApPP) and Notch, are a group of aspartic proteases (De Strooper et al 1999, De Strooper

et al 1998). This family also includes the signal peptide peptidase (SPP) (Weihofen et al 2002),

which catalyses intramembrane proteolysis of signal séquence remnants and possibly also

membrane proteins in the endoplasmic reticulum (ER) membrane of animal and plant cells

(Weihofen & Martoglio 2003). The third and most recently discovered I-Clip family are the

rhomboids (Urban 2006), which are the most evolutionary conserved one and, by implication, the

first to émerge in life (Koonin et al 2003).

The conserved core of rhomboid family members consists of six conserved TMHs (Koonin et al

2003), with the Ser and His residues required to form the catalytic dyad (Urban & Wolfe 2005)

embedded in TMH-4 and TMH-6, respectively (Wang et al 2006). Consistent with the widely

7

accepted hypothesis that I-Clips catalytic centre is assembled within the plane of the membrane, récent crystallographic studies hâve shown that rhomboid's catalytic dyad is found at the protein interior at a depth below the membrane surface, indicating that, in RIP, the scission of the peptide bonds indeed takes place within the hydrophobic environment of the membrane bilayer (Wang et al 2006).

Fig 1.1 Représentation of the four known families of intramembrane proteases. Catalytic residues are in red, and conserved motifs that are typical of proteases of each mechanistic class are in black. Note that presenilins and Rhomboids cleave only type I proteins (with an extracellular/luminal N terminus), whereas site 2 proteases and signal-peptide peptidases cleave type II tproteins. The approximate sites of cleavage are shown, and the green arrow indicates the direction of domain release: only Rhomboids are involved predominantly in extracellular release of factors. Presenilins and signal-peptide peptidases are aspartyl

proteases that use two aspartates to cleave substrates, the site 2 proteases are metalloproteases that coordinate a zinc ion using two conserved histidines and an aspartate, and Rhomboids are serine proteases that use a catalytic triad to cleave substrates (hydrogen bonds of the triad are indicated as dashed lines). Taken from Urban, S., and Freeman, M. (2002) Intramembrane proteolysis controls diverse signalling pathways throughout évolution. Curr Opin Genêt Dev 12, 512-518

In spite of the présence of rhomboids in the majority of modem life forais from ail three primary superkingdoms, phylogenetic analysis suggests that this family has not been inherited from the last universal common ancestor (LUCA) (Koonin et al 2003). Instead, the tree topology indicates that this family emerged in some bacterial lineage and afterwards had been widely disseminated by horizontal gène transfer (HGT), and then lost in some lineages. Both

s

archaea and eukaryotes hâve acquired rhomboids on several independent occasions. In particular, at least two HGT events hâve contributed to the origin of eukaryotic rhomboids, one of them yielding the RHO subfamily and the other one the PARL subfamily, with a possible additional HGT in plants (Koonin et al 2003), where rhomboids could regulate plastids morphology and activities.

Eukaryotic members of the two subfamilies hâve différent domain architecture organization. Proteins of the RHO subfamily, which is prototyped by the drosophila developmental regulator Rhomboid (Bier et al 1990), typically hâve an extra TMH added carboxy-terminally to the 6-TMH catalytic core, the "6+1" structure, lnstead, members of the PARL subfamily, which is prototyped by the mitochondrial morphology regulator Presenilin-associated rhomboid-like protein (PARL) (Pellegrini et al 2001), hâve an extra TMH added to the amino terminus of the 6-TMH catalytic core, the "1+6" structure (Koonin et al 2003). Such additional TMH in either subfamily implies the existence of a loop that connects it to the 6-TMH catalytic core, which could hâve a regulatory function.

Figure 1.2 Structures of GlpG rhomboid intramembrane serine proteases from E. coli from différent laboratories. AH structures reveal six transmembrane hélices and a Ser201-His254 catalytic dyad located near the N termini of TM4 and TM6, respectively. The short TM4 and TM5 hélices ( "15 residues) aid the formation of a central water-filled cavity containing the catalytic dyad, which is located about 10 À below the extracellular surface of the membrane bilayer. The water molécules in the cavity, shown as van der Waals sphères, are based on Wang et al. (Wang et al 2006) (purpie) and Wu. et al. (Wu et al 2006)(green). Thèse structures (PDB 2IC8, 2NRF and 2IRV) crystallized in the PZ2, P3i and P2i space groups with one, two (antiparallel) and two (antiparallel) molécules in the asymmetric unit, respectively. Each of the antiparallel molécules in the 2NRF and 2IRV asymmetric units hâve slightly différent structures, (a) View of GlpG structures from the extracellular surface, based on Wu er ai, showing the alignment of molécules A and B of 2NRF (2.6 A, molécule A red, molécule B yellow) relative to the single molécule of 2IC8 based an Wang ef al., (2.1 A, grey). Ail parts of the three molécules align very well, except for TM5. In chain A, this helix is distorted and displaced away from the central cavity, compared with TM5 of 2IC8. In chain B, TM5 is also displaced, but less than in chain A. Loop L5 blocks access to the cavity in 2IC8 (gray) but not in chain A of 2NRF (red). (b) View of GlpG structures in a direction parallel to the membrane plane. This view toward the catalytic dyad reveals a direct pathway to the dyad via a TM5-TM2 'gâte', (c) View, perpendicular to the membrane, of structures in a, with aligned structures of Ben-Shem et al (Ben-Shem) (2.3 A, 21RV, molécule A cyan, molécule B pink) added. TM5 is displaced away from the cavity in both molécules A and B, but not as much as in 2NRF. (d) View as in b of the five aligned structures. The great variability of TM5 in the structures suggests that TM5 is easily distorted in the crystal structures, consistent with TM5 being easily displaced in vivo to allow entry of the substrate helix. Taken from Stephen H White (2006) Rhomboid intramembrane protease structures galorel Nature Structural & Molecular Biology -13, 1049-1051

9

Mitochondrial rhomboids: novel components of the mitochondria morphology remodelling machinery

A yeast genetic screen aimed at identifying components involved in mitochondrial fusion led to the identification of two mutants, Ugo 1 and Ugo 2, which were defective in this process (Sesaki & Jensen 2001). Ugo 2 was later described as a mutant of Pcpl (Esser et al 2002, Sesaki et al 2003), a gène encoding a rhomboid protease of the PARL subfamily which is targeted to the mitochondria in vitro (Steinmetz et al 2002) and in vivo (Esser et al 2002). Récent évidences hâve shown that mitochondrial rhomboids participate in the régulation of the organelle's morphology. Several groups hâve indeed shown that yeast cells lacking the yeast rhomboid Pcplp are defective in the processing of cytochrome c peroxidase 1 (Ccpl) (Esser et al 2002, McQuibban et al 2003) and of the dynamin-related Mgml protein (Herlan et al 2003, McQuibban et al 2003, Sesaki et al 2003), a key component of the mitochondria fusion machinery (Guan et al 1993, Shepard & Yaffe 1999, Wong et al 2000). While the function of Ccpl cleavage is still unknown, that of Mgmlp is linked to mitochondria morphology régulation (Herlan et al 2003, McQuibban et al 2003). Consistently, APcpl cells contain partially fragmented mitochondria, instead of the long tubular branched mitochondria of wild-type cells (Dimmer et al 2002, Sesaki & Jensen 2001, Sesaki et al 2003). However, this phenotype is rescued by expression of the drosophila and mammalian ortholog of Pcplp, Rhomboid-7 and PARL, respectively (McQuibban et al 2003). Thus, during animal évolution the function of this rhomboid protease in the régulation of mitochondrial morphology has been conserved, which is reflected by their localization in the inner mitochondrial membrane (Jeyaraju et al 2006b) (Dimmer et al 2002) and identical "1+6 " structure, with the catalytic serine and histidine located in TMH5 and TMH7, respectively (Koonin et al 2003).

The mammalian mitochondrial rhomboid protease PARL

PARL, presenilin-associated rhomboid-like, dérives its name from being identifïed in a yeast two hybrid screen for proteins that could interact with Alzheimer's Presenilin-1 protein (Pellegrini et al 2001). Although Presenilin-1 has been found in the mitochondria (Ankarcrona & Hultenby 2002, Gupta et al 2004), its association in vivo with PARL remains to be addressed. The first functions of a PARL protease were inferred from a discovery in the budding yeast: early genome-

10

wide screens revealed that a homolog of PARJL, Pcpl (also called Rbdl and YgrlOlw), is localized to mitochondria (Steinmetz et al 2002), and its deletion results in mitochondrial fragmentation (Dimmer et al 2002). Pcpl was subsequently discovered to be responsible for cleaving the targeting séquence of cytochrome c peroxidase (Ccpl), a nuclear-encoded mitochondrial protein présent in the intermembrane space (Esser et al 2002). This defined a fourth and previously unknown targeting peptide processing pathway in mitochondria. However, pcpl deleted cells had a slow growth phenotype while the ccpl deletion did not, suggesting that the basis of the pcpl phenotype results from failure to process another substrate.

Despite the functional and structural conservation of yeast Pcplp, drosophila Rhomboid-7, and mammalian PARL, their N-terminal domain are unrelated (Sik et al 2004). The N-terminal région of PARL, which is in the matrix (Jeyaraju et al 2006b), spans the first 100 amino acids of the protein and shows no détectable similarity to any other available protein séquences. This domain of PARL, designated Pp domain, is vertebrate-specific, as indicated by the notable conservation among mammals and, to a lesser extent, other vertebrates, but not between vertebrates and insects (Sik et al 2004). Although the fonction of the Pp domain remains unknown, its biological relevance is évident from its séquence conservation. Indeed, in the four available mammalian PARL séquences, 58 of the 62 residues of the Pp domain are invariant, and there are no insertions or deletions, which suggests that at least during mammalian évolution, the N-terminal région of PARL was subject to strong purifying sélection, which can be explained by functional constraints. In unconstrained séquences evolving neutrally, very few, if any, invariant residues would be expected to survive the ~100 million years of évolution separating mammalian orders. This analysis suggests that émergence of the Pp domain at the outset of vertebrate évolution may be associated with the appearance of a new mechanism of régulation of PARL in mitochondria morphology and/or with the recruitment of PARL into novel mitochondrial pathways.

II. Régulation of apoptosis by mitochondria morphology Programmed cell death and its morphologie manifestation of apoptosis is a conserved pathway that in its basic tenets appears operative in ail metazoans. Cell deaths during embryonic development are essential for successful organogenesis and the crafting of complex multicellular tissues. Apoptosis also opérâtes in adult organisms to maintain normal cellular homeostasis. This

i l

is especially critical in long-lived mammals that must integrate multiple physiological as well as

pathological death signais, which for example includes regulating the response to infectious

agents. Gain- and loss-of-function models of gènes in the core apoptotic pathway indicate that the

violation of cellular homeostasis can be a primary pathogenic event that results in disease.

Evidence indicates that insufficient apoptosis can manifest as cancer or autoimmunity, while

accelerated cell death is évident in acute and chronic degenerative diseases, immunodeficiency,

and infertility.

Apoptosis: rôle of cytochrome c in the intrinsic pathway

Apoptosis can be triggered through two major pathways. Extracellular signais such as members

of the tumour necrosis factor family can activate the receptor-mediated extrinsic pathway.

Alternatively, stress signais such as DNA damage, hypoxia, and loss of survival signais may

trigger the mitochondrial intrinsic pathway. Within this pathway, in response to apoptotic stimuli,

cytochrome c, a previously known component of électron transfer chain in mitochondria, gets

released to cytosol where it binds to a partner protein, Apaf-1 (Liu et al 1996, Shi 2002). Apaf-1

consists of three functional domains; the N-terminal caspase recruitment domain (CARD), the

middle nucleotide-binding and oligomerization domain, and the C-terminal regulatory région

composed of 13 WD-40 repeats (Zou et al 1999). This regulatory région normally keeps Apaf-1

in an autoinhibitory state and when cytochrome c binds to this région, Apaf-1 becomes activated

in the présence of dATP or ATP (Acehan et al 2002, Adrain et al 1999, Srinivasula et al 1998).

The activation is accomplished through oligomerization of seven individual Apaf-1/cytochrome c

complexes into a wheel-like heptamer, called apoptosome (Acehan et al 2002, Zou et al 1999).

Once bound to the apoptosome, caspase-9 is activated, and subsequently triggers a cascade of

effector caspases activation and proteolysis, leading to apoptotic cell death.

Under non-apoptotic conditions, cytochrome c is kept confined inside "cisternae", which are

formed by juxtaposition of the cristae (Mannella 2006a, Mannella 2006b). Therefore, a crucial

step in the apoptotic program is to remodel the cristae to open thèse cisternae and liberate

cytochrome c into the intermembrane space (IMS) (Scorrano et al 2002), from where it can be

released to the cystosol through permeabilization of the outer membrane (Forte & Bernardi 2006,

Zamzami & Kroemer 2003). Conversely, under steady-state conditions the integrity of the

12

cisternae is essential to maintain cell viability (Scorrano et al 2002). Although récent work suggested that the BH3-only pro-apoptotic Bcl-2 member Bid, disrupts the structure of the dynamin-related protein OPAl, which is involved in keeping the cristae closed under steady-state conditions, the only molécules known to maintain cristae closed under steady-state conditions are the mitochondrial rhomboid protease PARL and the dynamin-related GTPase OPAl (Cipolat et al 2006, Frezza et al 2006).

The mitochondria morphology machinery

Mitochondria form a functional reticulum whose steady-state morphology is regulated by dynamic membrane fission and fusion events (Shaw & Nunnari 2002, Yaffe 2003). Typically, suppression of the molecular mechanisms coordinating mitochondria membrane fusion, or dominance of those governing membrane fission, cause mitochondria to fragment into short rods or sphères. Conversely, disruption of the molecular processes regulating mitochondria membrane fission, or prevalence of those controlling membrane fusion, generate elongated, interconnected tubules.

Mitochondrial shape is determined by the antagoniste forces of fission and fusion.

Fused WjW type Fragmented

Fig 1.3. The différent states of mitochondria morphology.

Pioneering work in Saccharomyces cerevisiae has shown that mitochondria morphology is governed by a small but growing set of conserved "mitochondria-shaping" proteins that independently regulate membrane fusion and fission (Meeusen & Nunnari 2005, Okamoto &

13

Shaw 2005, Yaffe 2003). In the last years, some of their mammalian orthologues hâve been identified, although their mechanism of régulation is still unexplored (Chen & Chan 2005).

In mammals, mitochondrial fission relies on hFis (Koch et al 2005, Mozdy et al 2000, Yoon et al 2003) and on a member of the dynamin family of GTPases, DRPl (Kuroiwa et al 2006). Mechanistically, it has been suggested that DRP1, like other dynamins (Schrader 2006), oligomerizes into ring-like structures around the fission sites, to constrict the organelle in the points where it divides (Ingerman et al 2005).

Mammalian mitochondrial fusion relies on yet three other GTPases from the dynamin family. Mitofusin 1 (MFN1) and Mitofusin 2 (MFN2) are integrated within the mitochondrial outer membrane, with their GTPase and coiled-coil domains exposed to the cytosol (Chan et al 2006, Chen & Chan 2005, Chen et al 2005, Chen et al 2003). MFN1 and MFN2 exist as homotypic and heterotypic complexes that can form between adjacent organelles. Mechanistically, it has been suggested that the carboxy-terminal coiled coils tether two organelles undergoing fusion, with the GTPase domains probably regulating the fusion reaction (Koshiba et al 2004).

Another dynamin-like GTPase, OPA1 (Alexander et al 2000, Delettre et al 2000), résides in the intermembrane space, where it is associated with the inner membrane (Olichon et al 2002). OPA1 exists as multiple splice and cleavage variants, which ultimately control its activity in membrane fusion (Delettre et al 2001, Frezza et al 2006, lshihara et al 2006). Cleavage of OPA1 and of its yeast orthologue Mgmlp is mediated by the rhomboid protease PARL, which therefore is also part of the machinery regulating mitochondria morpliology remodelling (Cipolat et al 2006, Herlan et al 2004, Herlan et al 2003, McQuibban et al 2003). The mechanism controlling OPA1 processing is still unknown.

Mec lui ni s m s of mitochondrial morpliology régulation integrated to the apoptotic program

It has been long known that in mammalian cells mitochondria undergo fragmentation during apoptosis. However, only recently this process was shown to be an intégral part of the apoptotic program, and not an epiphenomenon. In a pionering study by Frank, Youle and colleagues, this process was found to be inhibited by ectopic expression of the dominant-négative mutant of the pro-fission protein DRP1, DRP1-K38A (Frank et al 2001). More importantly, DRPl-KogA also severely reduced the extent of cell death by delaying cytochrome c release, indicating that

14

apoptosis initiation requires mitochondria fragmentation. Since this pivotai observation, during the past 5 years more évidence emerged indicating that the machinery governing the coordinated fusion and fission of the outer and inner mitochondrial membrane participâtes in the progression of apoptosis (Karbowski & Youle 2003, Youle & Karbowski 2005). For instance, in mammalian cells down-regulation of the pro-fïssion protein hFisl powerfully inhibits cell death to an extent significantly greater than down-regulation of DRP1 and at a stage of apoptosis distinct from that induced by DRP1 inhibition (Alirol et al 2006, Yu et al 2005). In addition, it has been shown that cells depleted of the inner membrane pro-fusion protein OPA1 are extremely sensitive to exogenous apoptosis induction, indicating that OPA1 may normally function as an antiapoptotic protein (Lee et al 2004b). Finally, the outer membrane pro-fusion proteins MFN1 and MFN2, alone or in combination, also prevent death induced by several intrinsic stimuli (Cipolat et al 2004, Neuspiel et al 2005, Sugioka et al 2004), consistent with early inhibition of MFN1-dependent fusion during apoptosis (Karbowski et al 2004).

Although it is now well established that "mitochondria-shaping" proteins can positively and negatively regulate apoptosis, little is known about how their coordinated activity is integrated into the apoptotic program (Chan et al 2006), thereby limiting our understanding on whether altered régulation of thèse proteins participâtes in tumour initiation and anti-cancer drug résistance.

Rôle of the mitochondrial rhomboid protease PARL in mitochondria morphology and apoptosis Pioneering studies of mitochondrial structure by Mannella and Frey using électron tomography hâve shown that the well-known lamellar and tubular structure of the mitochondiial inner membrane générâtes a novel type of compartment, created by juxtaposition of inner-membrane folds (Frey & Mannella 2000). Thèse intra-cràtae structures form cisternae in which cytochrome c is trapped (Mannella 2006a, Mannella 2006b), thereby forming a barrier to the diffusion of the protein to the IMS (Frey & Mannella 2000). A critical step is mus required for apoptosis progression: the opening of the cristae to eliminate the diffusion barrier and free cytochrome c to the IMS, from where it can be released to the cytosol (Scorrano et al 2002).

Dr Pellegrini's lab has recently contributed to show that this process is mediated by the rhomboid intramembrane-cleaving protease PARL (Cipolat et al 2006), a conserved regulator of

15

mitochondria morphology remodeling (McQuibban et al 2003). Mechanistically, PARL

coordinates the intramembraneous cleavage of the inner membrane protein OPA1, a single-pass

transmembrane GTPase, to liberate an IMS-soluble form of the protein (IMS-OPAl) that

assembles in macromolecular complexes with PARL and with the uncleaved, membrane bound

form of OPA1. This structure "staple" cristae juxtaposition (Cipolat et al 2006, Frezza et al

2006), thereby precluding cytochrome c access to the IMS. Consistent with this finding, we

showed that lack of PARL blocked IMS-OPAl formation, causing faster cristae remodelling,

cytochrome c mobilization in the IMS, and apoptosis progression. Conversely, increased IMS-

OPAl expression in the IMS protected MEF cells from several intrinsic pro-apoptotic stimuli

(Cipolat et al 2006). Thus, by maintaining cristae juxtaposed PARL "keeps a lid to apoptosis"

(Zhang et al 2003). Whether in some cancer cells PARL activity is upregulated is not known,

although this may be a mechanism tumour cells develop to acquire insensitivity to intrinsic

proapoptotic stimuli or résistance to anti-cancer drugs.

Although the exact components and stoichiometry of the molécules forming the macromolecular

complex that staple cristae juxtaposition are still unknown, évidence indicates the présence of at

least two uncleaved, membrane-bound forms of OPA1 and one IMS-OPAl. Given the fact that

OPA1 and PARL directly interact together via the first IMS loop of PARL, a model has been

proposed where two molécules of PARL on opposite sides of the cristae in juxtaposition each

bind a molécule of OPA1 which, in turn, binds to one molécule of IMS-OPAl (Cipolat et al

2006, Frezza et al 2006). The mechanism that "unstaples" cristae juxtaposition during apoptosis

and release the PARL/OPA1/IMS-OPAl complex is unknown. In addition, whether and how

increased IMS-OPAl expression, which protects cells from apoptosis (Cipolat et al 2006),

stabilizes the PARL/OPA1/IMS-OPAl complex remain undiscovered.

Molecular mechanisms regulating PARL activity

Despite their functional and structural conservation, the yeast mitochondrial rhomboid protease

Pcplp and PARL hâve unrelated N-terminal domains. The N-terminal région of PARL shows no

détectable similarity to any other available protein séquences. This région of PARL, designated

PP (spanning arnino acids 40-101), is vertebrate-specific, as indicated by the notable

conservation among mammals and, to a lesser extent, other vertebrates, but not between

16

vertebrates and insects (Sik et al 2004). Although the fonction of the Pp domain remains unknown, its biological relevance is évident from its séquence conservation. Indeed, in the four available mammalian PARL séquences, 58 of the 62 residues of the PP domain are invariant, and there are no insertions or deletions (Sik et al 2004), which suggest that at least during mammalian évolution, the N-terminal région of PARJL was subject to strong purifying sélection, which can be explained by functional constraints. In unconstrained séquences evolving neutrally, very few, if any, invariant residues would be expected to survive the 100 million years of évolution separating mammalian orders. This analysis suggests that émergence of the Pp domain at the outset of vertebrate évolution may be associated with the appearance of a new mechanism of régulation of PARL. We hâve recently shown that this part of the PARL molécule undergoes two consécutive cleavage events, termed a and p. The proximal a-cleavage is a constitutive processing associated with the protein import in the mitochondria, whereas the distal p-cleavage is regulated through a mechanism of proteolysis requiring PARL activity supplied in tram (Sik et al 2004). Whether this cleavage occurs in vivo is unknown. In addition, its mechanism of régulation and its functional significance remain unexplored.

17

2.1 Objectives - part I

When I started my Master studies in the laboratory of Dr. Pellegrini, I joined an ongoing study that was aiming at:

1) To investigate whether PARL P-cleavage occurs in vivo;

2) To discover the mechanism of régulation of PARL P-cleavage;

3) To address the rôle of PARL P-cleavage in mitochondria dynamics.

2.2 Results and Discussion - part I

Jeyaraju DV, Xu L, Letellier MC, Bandaru S. Zunino R, Berg EA, McBride HM, Pellegrini L.

Phosphorylation and cleavage of presenilin-associated rhomboid-like protein (PARL) promotes changes in mitochondrial morphology.

Proc Natl AcadSci USA 2006 Dec 5; 103(49): 18562-7

Voir pages suivantes.

18

Phosphorylation and cleavage of presenilin-associated rhomboid-like protein (PARL) promotes changes in mitochondrial morphology Oanny V. Jeyaraju*, Liqun Xu*, Marie-Claude Letellier*, Sirisha Bandaru*, Rodolfo Zunino*, Eric A. Berg*, Heidi M. McBridets, and Luca Pellegrini*6

•Centre de Recherche Université Lavai Robert-Giffard, 2601 ch. de la Canardière, Québec City, QC, Canada G1J 2G3; 'University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, ON, Canada Kl Y 4W7; and *21st Century Biochemicals, 33 Locke Drive, Marlboro, MA 01752-1146

Edited by Walter Neupert, Institute fur Physiologische Chemie, Munich, Germany, and accepted by the Editorial Board October 12, 2006 (received for review June 14, 2006)

Remodeling of mitochondria is a dynamic process coordinated by fusion and fission of the inner and outer membranes of the organelle, mediated by a set of conserved proteins. In metazoans, the molecular mechanism behind mitochondrial morphology has been recruited to govern novel f unctions, such as development, calcium signaling, and apoptosis, which suggests that novel mechanisms should exist to regulate the conserved membrane fusion/fission machinery. Hère we show that phosphorylation and cleavage of the vertebrate-specif ic PJ3 domain of the mammalian presenilin-associated rhomboid-like (PARL) protease can influence mitochondrial morphology. Phosphor­ylation of three residues embedded in this domain, Ser-65, Thr-69, and Ser-70, impair a cleavage at position Ser77-Ala78 that is required to initiate PARL-induced mitochondrial fragmentation. Our findings re-veal that PARL phosphorylation and cleavage impact mitochondrial dynamics, providing a blueprint to study the molecular évolution of mitochondrial morphology.

protein évolution | protein phosphorylation j rhomboids | mitochondrial dynamics | intramebrane proteolysis

itochondrial biogenesis is an essential cellular process governed by a small set of proteins with membrane

pro-fusion and pro-fïssion activities which are conserved in ail eukaryotes (1-3). During metazoan évolution, this process has been recruited to coordinate novel mitochondrial functions, such as apoptosis (4-6), thereby suggesting the émergence, in higher eukaryotes, of novel mechanisms of régulation of the fusion and fission machinery of the organelle. Formai, mechanistic évi­dence supporting this hypothesis is, however, still missing.

RecentJy, rhomboid proteases hâve been implicated in the rég­ulation of mitochondrial membrane remodeling. Studios in Sac-chammyces cerevisiae demonstrated that PCPIP is required to cleave Mgmlp, an intermembrane space dynamin family member that participâtes in mitochondrial fusion events (7, 8). The yeast PCPIP protein belongs to a subfamily of mitochondrial rhomboid proteases typified by presenilin-associated rhomboid-like (PARL) protein (9, 10), the human ortholog of PCPIP (8). Despite their functional and structural conservation, PCPIP and PARL hâve unrelated N-tenninal domains. The N-terminal région of PARL shows no détectable similarity to any other available protein stîquences. This région of PARL, designated Pj3 (spanning amino acids 40-100), is vertebrate-specific, as indicated by the notable conservation among mammals and, to a lesser extent, other verte-brates, but not between vertébrales and insects (11). Although the function of the P/3 domain remains unknown, its biological rele-v, tncc is évident from its séquence conservation. Indeed, in the four available mammalian PARL séquences, 58 of the 62 residues of the P/3 domain are invariant, and there aie no insertions or deletions (il), which suggests that at least during mammalian évolution, the N-terminal région of PARL was subject to strong purifying sélec­tion, which can be explained by functional constraints. In uncon-slrained séquences evolving neutrally, very few, if any, invariant

residues would be expected to survive the *»100 million years of évolution separating mammalian orders (12, 13). This analysis suggests that émergence of the P0 domain at the outset of verte-brate évolution niay be associated with the appearance of a new mechanism of régulation of PARL. We hâve recentry shown that this part of the PARL molécule undergoes two consécutive cleav­age events, termed a and fi. The proximal «-cleavage is a consti­tutive processing associated with the protein import in the mito­chondria, whereas the distal j3-cleavage is regulated through a mechanism of proteolysis requiring PARL activity supplice! in trans (11). Whether this cleavage occurs in vivo is unknown. In addition, its mechanism of régulation and its functional significance remain unexplored.

Results Human PARL Is Subjected to /{-Cleavage in Vivo. P A R L iransl'ccted in HEK 293 cells is cteaved at position Ser77-Ala78, which maps within the vertebrate-specific P0 domain (11), suggesting that, in vivo, the rhomboid protease may undergo the same processing. To address this possibilily, we generated a polyclonal antibody against a peptide spanning the C terminus of PARL (anti-PARL-C-Term; Fig. L4). This spécifie antibody (Fig. \B) was used to immunopre-cipitate endogenous PARL froin lysates of mitochondria isolated from human placenta. Using antibodies recognizing the N-terminal and C-terminal régions of PARL (Fig. \À ), we then examinée! the cleavage of the endogenous protein relative to the transfected PARL-FCT by epitope mapping. Data showed two bands whose mobility is, as expected, slightly higher than that of the FLAG-tagged positive contrat (Fig. IC Upper). Although both forms are immunoreactive against anti-PARL-C-Term, only the slower mi­grating band was positive to anti-PARL-N-Term, indicating that the corresponding epitope was absent in the faster migrating band (Fig. IC). Thèse data strongly suggest that endogenous PARL N terminus undergoes /3-cleavage, indicating that this processing may mechanistically coordinate the function of the rhomboid protease in vivo.

Author contributions: D.VJ. and L.X. contributed equally to this work; H.M.M. and L.P. designed research; D.VJ., L.X., M.-C.L, S.B.. R.Z., E.A.B., H.M.M., and L.P. performed research; H.M.M. and L.P. contributed newreagents/analytic tools; E.A.B., H.M.M., and L.P. analyzed data; and H.M.M. and L.P. wrote the paper.

Conflict of interest statement: E.A.B. is an employée and stockholder of Century 21st Biochemicals. This company sells mass spectrometry analysis and antibody production services. This article isa PNAS direct submission. W.N. is a guest editor invited by the Editorial Board. Abbreviations: LOMS, liquid chromatography/mass spectrometry; MAMP, mature mito­chondrial PARL; PACT, PARL C-terminal product (of p-deavage); PARL, presenilin-associ­ated rhomboid-like (protein); PARL-FCT, PARL-FLAG-Cr.n»ir,i«. sTo whom correspondence may be addressed. E-mail: hmcbride©ottawaheart.ca or luca.pellegriniecrulrg.ulaval.ca.

© 2006 by The National Academy of Sciences of the USA

111562-18567 | PNAS | December 5,2006 | vol. 103 | no. 49 www.pnas.org/cgi/doi/10.1073/pnas.0604983103

ja jP ■ I I I I I O J PARL 42kDa

I I I I I I I I I MAMP 36kDa I I I I I I I I ■ PACT 33kDa

PARL-N-Term epitope m PARL-C-Term epitope

B Peptide PARL-C-Term

t P a n t i - P A R L < - T e r m

Transfected

PARL-FLAG-CT ►

MAMP-FLAG-CT »• endogenous

<3MAMP

WBant i -PARL-N- te rm

IP anti-PARL-C-Term

PARL-FCT Human transfected placenta

cells mltochondria

MAMP-FLAG-CT > 35kDa-

PACT-FUfrCT ►

Transfected PARL-FLAG-CT »►

MAMP-FLAG-CT k> 3 5 k D a -

endogenous < ] MAMP

^ PACT

endogenous ^ M A M P

WBanti-PARL-N-Term

Fig. 1. Cleavage of PARL P0 domain in vivo. (A) Schematic représentation of the «- and (i-cleaved forms of PARL, MAMP, and the PARL C-terminal product ot (j-cleavage (PACT). The locations of the epitopes recognized by the anti-PARL-N-Term and anti-PARL-C-Term antibodies are indicated. Small black squares depict the seven transmembrane helixes of PARL. (S) Specif icity of the anti-PARL-C-Term antibody. The anti-PARL-C-Term antibody specifically im-munoprecipitates (IP) PARL-FLAG-CTwminus (PARL-FCT) from HEK 293-transfected cells as well as endogenous PARL from mitochondrial lysates of human placenta. The specif icity was addressed by preadsorbing the antisera with the synthetic peptide PARL-C-Term used to generate the corresponding antibody. WB, Western blotting. (O Endogenous PARL is subjected to N-terminal 0-cleavage as observed for transfected PARL. Transfected PARL-FCT and endogenous PARL were immunoprecipitated with anti-PARL-C-Term and subjected to epitope mapping. (Upper) Both the transfected and endogenous PARL are présent in two forms, MAMP and PACT, which migrate according to the predicted molecular mass shown in A The lower band labeled as PACT-FLAG-CT corresponds to the product of 0-cleavage at residue Ser"-iArg78 (11). Note the slight différence in mobility between transfected and endogenous PACT, which is caused by the présence/absence of the FLAG tag. (Lower) Transfected and endogenous PACT lack the N-Term epitope because of N-te'minai 0-cleavage. Asterisks indicate nonspecific cleavages.

Endogenous and Transfected PARL Are Hyperphosphorylated at the Vertebrate-Specific P/3 Domain. To investigate the mechanism of régulation of PARL /3-cleavage in vivo, we conducted mass spec-trometric studios to identify posttranslational modificatioas on the P/3 domain ( 14,15). PARL was immunoprecipitated from lysates

obtained from 250 mg of solubihzed human placenta nutochondna, it was digested, and the peptides were subjected to LC/MS analysis. Data showed that >35% of the entire protein was covered (Table 1, which is published as supporting information on the PNAS web site), with two molecular ions spanning nearfy the entire P/3 domain of the mitochondrial mature form of PARL, MAMP (Fig. IA). Ion m/z 1072.932* corresponded to a triple-phosphorylated ^VE-PRRSDPGTSGEAYKR76 peptide, which maps between the a-and /3-cleavage sites; ion m/z 1138.13** corresponded instead to an umnoditïcd peptide spanning the /3-cleavage site and ils distal région (77SALIPPVEETVFYPSPYPIR%; Fig. 2 A and£).

To investigate whether transfected PARL is also phosphorylated, we overexpressed PARL-FCT in HEK 293 cells. The protein was immunoprecipitated with anti-FLAG to isolate the transfected protein. It was unlikely that endogenous PARL was copurified during this step because coimmunoprecipitation studies with PARL constructs harboring différent tags did not reveal ho-modimers or oligomers (data not shown). The a-cleaved form of PARL, MAMP (Fig. \A), was then isolated by gel electrophoresis, digested, and subjected to LC/MS analysis. More than 51% of the protein was covered (Table 2, which is published as supporting information on the PNAS web site). Within this peptide data set we also observed a triple-phosphorylated 56APRKVEPRRSDPGTS-GEAYKR76 peptide, molecular ion m/z 866.39" (Fig. 2B), which overlaps with most of the triple- phosphorylated m/z 1072.932+

identified during the analysis of the endogenous PARL. Molecular ion m/z 1138.132+, 77SALIPPVEETVFYPSPYPIR'*, was also found (Table 2), indicating that sample préparation and analyses were performed under comparable expérimental conditions. Sim-ilar results were also obtained from PARL-FCT isolated from transfected HeLa cells (data not shown).

To refine thèse results, we subjected ion m/z 866.39-1 ' to tandem MS analysis. Data showed a séries of titrée water and phosphoric acid losses as the primary detected fragments (Fig. 2 D and E and Table 3, which is published as supporting information on the PNAS web site), consistent with the fragmentation pattern of a peptide with phosphorylated Ser and Thr, rather than Tyr residues. Addi-tionalfy, the nonphosphorylated y3 ion and the Y immonium ion but not their corresponding phosphorylated species were detected, indicating lack of phosphorylation at Tyr-74 of PARL. Further-more, the N-terminal ion bi2-K»P04, an internai ion séries (GTSG-2H3P04, PGTS-2H3PO4, DPGTSG-2H,P04), and the C-terminal ion yijj-H3P04 indicate phosphorylation at Ser-65, Thr-69, and Ser-70 (Fig. 2 D and £ and Table 3). This conclusion was further supportod by the lack of phosphorylated peptides in the data set obtained from LC/MS analysis of a transfected PARL mutant bearing Ala substitutions at thèse residues (Fig. 2 C and E and data not shown). A monophosphorylated peptide spanning Ser-65, Thr-69, and Ser-70 was detected in vivo as well as in vitro (Table 1); however, its relative abundance was very low compared with the triple- phosphorylated form (data not shown), indicating that most of the a-cleaved form of PARL is phosphorylated. We conclude that the vertebrate-specific P/3 domain of endogenous and transfected PARL is phosphorylated at residues Ser-65, Thr-69, and Ser-70, and, by implication, that this modification has the same function in vivo and in vitro.

The Phosphorylated P/3 Domain Is Exposed to the Matrix. To déter­mine the localization of the phosphorylated, vertebrate-specific P/3 domain of PARL, we investigated the topology of the protein. To this end, we used a PARL construct with a hemagglutinin (HA) tag inserted at ils N terminus, at position 91, and a FLAG tag at its C terminus (PARL-HA-FCT; Fig. 3C). We perme-abilized PARL-HA-FCT-transfected HeLa cells with increasing amounts of digitonin, and we performed immunofluorescence with antibodies against either the FLAG or HA tag. At low concentrations of digitonin, only the plasma membrane was permeabilized, allowing the outer membrane receptor Tom20 to

Jeyaraju e t al. PNAS | December 5 , 2 0 0 6 | vol . 103 | no. 49 | 18563

B"

,11 II • M . » ' *

/

100* M t *

75%

mt

25%

• 0% jL

000 1100 IMHiM 400 000 800 000 nOQnto

[ ) ' 8 6 0 . W _ _

1M+3H]-H,0

821.7»» [M*3H]-H,PO,-2H,0

^ 1J41.07» [M*2H]-H,PO, tndogonouB

MAMP (A)

Transfected MAMP

FUAG-CT (B)

Tfan«(«ct«d MAMP

FLAG-CT mutant ST&AAA

(C)

| a-cleavage , (S-cteavage

S3GFRKAPRKVEPRRSDPGTSGEAYKRSALIPPVEETVFYPSPYPIRSLIK1M PfS-domaln Ion m/z 1138.13 2* 77SALIPPVEETVFYPSPYPIR (unmodified)

Ion mlz 1072.93 2* aoVEPRRSDPGTSGEAYKR (+3 phosphates)

Ion m/z 866.39 3* 58APRKVEPRRSOPGTSGEAYKR (♦ 3 phosphates)

ion m/z 686.71 2* seAPRKVEPRRSDP

ion m * 307.20 * S7PGTS

(-1 phosphate, -1 waler) I Tandem MS 1 analysis of

(-2 phosphate, -2 water)|i0n mlz 866.39**

Ion mlz 603.30 2* 65ADPGAAGEAYK (unmodified) P

Fig. 2. Phosphorylation of PARL P0 domain at Ser-65, Thr-69, and Ser-70 in vivo and in vitro. (A) Phosphorylation of endogenous PARL LOMS analysis of PARL f rom mitochondria pur i f ied f rom human placenta is shown. The protein was immunoprecipitated wi th the anti-PARL-C-Term antibody (Fig. 1), digested, and subjected t o LOMS analysis. (Left) Molecular ion m/z 1072.9324, which corresponds t o a tr iple-phosphorylated 60VEPRRSDPGTSGEAYKR76 peptide mapping botween PARL a- and 0-cleavage sites (Fig. 1 A) (11). (Righf) Ion m/z 1138.132 ' , corresponding t o the unmodif ied "SALIPPVEETVFYPSPYPIR96 peptide, which also maps on the vertebrate-specif ic P/J domain of PARL. More than 35% of the mature fo rm of PARL (MAMP; Fig. 1 A) could be found th rough this analysis; the complète list of the ions is shown in Table 1. The identity of each peptide was determined manually and w i th a Bayesian reconstruction a lgor i thm as wel l as searching against both theoretical peptide and fragmentat ion data f rom the PARL séquence. (S) Phosphorylation of transfected PARL-FCT. LOMS analysis of transfected PARL-FCT pur i f ied f rom HEK 293 cells is shown. MAMP-FLAG-Ci„minut (MAMP-FLAG-CT) was immunoprecipitated w i t h anti-FLAG monoclonal antibody, puri f ied by gel electrophoresis, digested, and analyzed by LOMS analysis. The triple-phosphorylated MAPRKVEPRRSDPGTSGEAYKR76 peptide, ion mlz 8<>6.39J ' , is indicated. More than 51 % of MAMP séquence could be found through this analysis; the complète list of ions is shown in Table 2. (O PARL mutant S65A/T69A/S70A is not phosphorylated. LOMS analysis of transfected PARL-FCT mutant AAA purif ied f rom HEK 293 cells is presented. The data show ion mlz 603.3-1 ' , corresponding t o the unphosphorylated 6bADPGAAGEAYK ''■' peptide. Note that no phosphorylated peptides encompassing the PfJ domain of this mutant protein were found (data not shown). (O) Tandem MS analysis of phosphorylated PARL. Ion mlz 866.393 ' was fragmentée! t o detect peptides that, th rough the loss of phosphate group(s) andtor water (-H3PO4), f inely map phosphorylation at Ser-65, Thr-69, and Ser-70. The N-terminal ion mlz 686.712 ' (5I'APRKVEPRRSDP-H3P04> and m/z 307.2+ ("PGTS-^HjPOa) are shown in (£). The complète list of molecular ions is shown in Table 3. The identity and phosphorylation state of each peptide were determined by both manual interprétat ion of the spectra and a Mascot search of ail o f the enhanced product ion scans. (£) Schematic représentation summarizlng the results showing phosphorylation of endogenous and transfected PARL at residue Ser-65, Thr-69, and Ser-70.

be efficiently reeognized by its antibody in "=«100% of cells (Fig. 3 A and B). As the digitonin concentration increased, first, 70% of cells were efficiently labeled with the FLAG along with the intermembrane space marker cytochromec. In contrast, the HA epitope was efficiently labeled (<**60% of cells) only on treatment with the highest concentrations of digitonin, which appeared concomitantly with the matrix marker peroxidoreductase 3/sp-22 (16). Thèse data indicate that the C-terminal domain of PARL is located in the intermembrane space, whereas the N-terminal phosphorylated P/3 domain is exposed in the mito-chondrial matrix, thereby correcting our previous examination of PARL topology using immunogold-labeled EM sections (II).

Phosphorylation of the P0 Domain Régulâtes 0-Cleavage. Given the pi oximity of the phosphorylated residues to the /3-cleavage site, we investigated whether this modification could hâve a regulatory fonction on the cleavage itself. We therefore mutated residues Ser-65, Thr-69, and Ser-70 to alanine, to abolish phosphorylation, and to aspartic acid, which is commonly used to rnimic phosphor­ylation by introducing a négative charge (17,18). We then tested the effect of thèse mutations on /3-cleavage, and we found that Asp substitutions of ail three phosphorylated amino acids led to strongry reduced levels of /3-cleavage (Fig. 44). Impaired /3-cleavage was also observed with two double Asp mutants, S65DAT69A/S70D and S65A/T69D/S70D (Fig. 4/1) but not with single substitutions (data

not shown), suggesting an additive inhibitory effect of each phos­phorylated amino acid on /3-cleavage. To demonstrate further the inhibitory effect of Asp substitutions on /J-cleavage, we performed a large-scale mutagenesis screening to identify PARL mutants with constitutive /3-cleavage. A mutant carrying the ^EETV87 deletion in the P/3 domain, A84-87, showed dramatically increased /3-cleav­age (Fig. 4 B and C). However, in this mutant, Asp substitutions in ail three phosphorylated residues dominantry reestablished block of /3-cleavage (Fig. 4D). Notably, none of the PARL mutants analyzed in this work with deletions and/or substitutions on the P/3 domain had eompromised protease activity, as indicated by their ability to cleave in trans a catarytically dead (S277G ) PARL protease (Fig. 4£ and data not shown). Therefore, lack of /3-cleavage of the mutant mimicking constitutive phosphorylation of PARL, DDD, is not the resuit of loss of proteolytic activity. This observation also implies that the import and insertion were not eompromised, as further confirmed in digitonin-permeabilization experiments (data not shown). We conclude that the stable phosphorylation of residues Ser-65, Thr-69, and Ser-70 inhibits PARL /3-cleavage.

/i-Cleavage Médiates PARL Activity in Mitochondrial Morphology. We next examined the rôle of /3-cleavage and phosphorylation on mitochondrial morphology. Transient expression of wild-type PARL-FCT in HeLa cells resulted in a dramatic increase in mitochondrial fragmentation (Fig. 5 A and B). Similar results were

18!>64 | www.pnas.org/cgi/doi/10.1073/pnas.0604983103 Jeyaraju et al.

0.012 02 1.0

B Hl B

■ ■ B ■ ■ B ■ ■ ■

Oigltonin (mg/ml)

antl-Tom20

anti-Cytc

anti-PRDX3

anti-FLAG

PABL-FIAG-CÎ

PARL-HA-FCT

S s 80 -

i l ! se to

Π20

antl-HA

■ Tom20

■ FLAG

D HA

0.012 lu

Fig. 3. Localization of PARL P/3 domain in the matrix. (A) HeLa cells were transfected with a construct expressing PARL-HA-FCT (see scheme in O. fixed, and permeabilized with the indicated concentrations of digitonin. For each condition, cells were coimmunostained with anti-FLAG or anti-HA, and anti-Tom20 (to label the outer membrane), or anti-cytochrome c (to label the intermembrane space), or anti-peroxiredoxin 3 (to label the matrix). (B) Quantitation of the experiments shown in A. (O Scheme summarizlng the topology of PARL.

observed with the S65A/T69A/S70A mutant and the constitutivery /3-cleaved PARL protein, A84-87. By eontrast, similar levels of expression of the S65D/T69D/S70D mutant (Fig. 5C and data not shown) did not resuit in signifïcant mitochondrial fragmentation (Fig. 5), suggesting that transient expression of the nonphospho-rylated, /3-cleaved form of PARL leads to fragmented mitochon-diia. To investigate this observation further, we transfected mutants in which /3-cleavage was abolished by removing (PARLA75-79) or mutating (PARLLTSE) the cleavage site. Expression of either pro­tein did not induce mitochondrial fragmentation (Fig. SA and B), further indicating that /3-cleavage is required to initiale this process.

Cleavage of PARL at the /3-site libérâtes the P/3 peptide, a 25-aa peptide that can target the nucleus when released to the cytosol ( 11). To investigate whether the liberated P/3 peptide is functionally implicated in the initiation of mitochondrial fragmentation, we transfected mutants in which we deleted parts of its séquence, A56-59 and A58-61 (Fig. 4B), and we analyzed the morphology of the mitochondria. Data showed that neither deletion impaired /3-cleavage and mitochondrial fragmentation (Fig. 5), indicating that the fonction of the P/3 peptide is independent of the initiation of PARL-induced mitochondrial fragmentation.

Discussion Considérable mechanistic and functional information on the mi­tochondrial rhomboid protease PCP1P in yeast is available (7,8,19,

3 § i % 1 s s s s s

e B Ë g g MAMPa-4 PACT ►

B II IPami-FLAû, WB anti-FLAG

l l l l l -a < <s < «

[ PpV<Jom»ln s délations

«1 < *i

S3PRXAPRKVEL>IiR6DPâTfra^yibtâALÏH'V£STVF'VPSPyPÏReLIK109 I ac leavage

I

I P cleavage

—■» : <*mm wm * PAG

IPanlI-FlAG, WB.ml-FLAG

PAP.L-FLAG-CT PARL-fUG-CT » « M !

hfs aftef transaction

12 24 36 48 60 12 24 36 48 60

4 MAMP P , « M

IPanti-FLAG. WB anti-FLAG

Wltdlype

SJ7/G

STS 65 69-70 AAA

STS 65-69-70OOO

<\ 84-87 (F.ETV)

A 87-90 (VFVP)

PARL Myx:-CT

+ mRL-FLAG-CTS277G

P anti-FLAG, WB anti-FLAG

Fig. 4. Substitutions mimicking phosphorylation at Ser-65, Thr-69, and Ser-70 inhibit PARL /3-cleavage. (/4) Constructs expressing the indicated mu­tant PARL protein were transfected in either HEK 293 or HeLa cells. The effect of mutations abolishing (Ala) or mimicking (Asp) phosphorylation (17, 18) is monitored by the amount of /3-cleaved form of PARLdetected, PACT (Fig. 1 A). Note that Asp but not Ala substitutions block /3-cleavage. IP, immunoprecipi-tatlon; WB, Western blotting. (B) Scheme of the deletions (A) within PARL P/3 domain that hâve been tested in this work. ( O The A84-87 mutant is consti­tutively cleaved at the S-cleavage site. (O) Asp substitutions at positions Ser-65, Thr-69, and Ser-70 in the A84-87 dominantly reestablish the block of /3-cleavage. (£) Ala and Asp substitutionsat the phosphorylated Ser-65, Thr-69, and Ser-70 residues do not affect PARL protease activity. HEK 293 cells were cotransfected with a catalytically dead PARL protein (PARL-Myc-CT S277G) and the indicated FLAG-tagged mutant, whose enzymatic activity is moni­tored by its ablllty to cleave the inactive PARL in trans and produce PACT (11).

20). On the other hand, the rôle of its mammalian ortholog PARL appears less clear. In higher organisms, dynamic changes in mito­chondrial shape hâve been implicated in programmed cell death (4-6), a process that emerged late during metazoan évolution. Therefore, the machinery of mitochondrial fusion and fission is likeh/ to be regulated by mechanisms additional to those for yeast. Our results implicate phosphorylation and cleavage of the P/3 domain of PARL in mitochondrial morphology. Because this domain is vertebrate-specific (11), this processing apparently is a regulatory mechanism that emerged during vertebrate évolution.

Jey araju et al. PNAS | December 5,2006 | vol. 103 | no. 49 | 18565

UgMMgggd WWftWffW AAA

AAA 000 LTSf , j . M M , J J . J , ^ ,

» • « g b M « MAMP

l M .1 icil >,■ IP jnti-H AI ,. WB a n W U G

Fig. 5. Cleavage of the vertebrate-specific Pfi domain is required to médiate PARL activity in mitochondrial morphology. (A) HeLa cells were transfected with the indicated PARL-FCT constructs, fixed, permeabilized, and stained v/ith monoclonal anti-FLAG (green) and polyclonal anti-Tom20 (red). Images v/ere taken with the Olympus FV1000 confocal microscope. (Scale bars, 5 ̂ m.) (S) The mitochondrial morphologies of the wild-type and PARL mutants shown in A were quantified from three independent experiments, counting 100 cells per experiment. (Q Cells in A express similar levels of transfected proteins. Anti-FLAG immunoprecipitation (IP) and Western blot (WB) analysis of the FLAG-tagged PARL constructs used in A are shown. Equal amounts of immunoprecipitated protein were loaded. Note that in separate experiments, we verified that the immunoprecipitation efficiently depleted ail of the transfected protein, validating this comparison (data not shown).

Recendy, it has been shown that deletion of Part in the moine resulted in prématuré postnatal death (21), which correlated with reduced levels of cleaved OPA1. OPA1 has been shown to be involved in the régulation of the so-called "cristae remodeling" pathway of apoptosis (21, 22) and the régulation of mitochondrial fusion (23). Although PARL was shown to cleave OPA1, only minor changes in mitochondrial morphology were observed in Pari"' fibroblasts (21), which indicates that PARL is not direcdy required for mitochondrial fusion (21). Therefore, the expression of a cleaved form of PARL appears to be a gain of fonction, leading to the fragmented morphology we hâve observed in this work. ttecause PARL-induced fragmentation dépends on ils phosphor-ylation and cleavage, the functional outcome of PARL expression on steady-state mitochondrial morphology is likely to be regulated by the abundance and activity of the yet-unidentiiïed PARL kinase phosphatase, and protease. Thus, apparendy disaccording results

may arise from a complex set of regulators that could be expressed at différent levels in différent tissues and cells.

To date, the only réversible phosphorylation/dephosphorylation events known to occur within the mitochondrial intermembrane space or matrix compartment are limited to the El subunits of the pyruvate and branched-chain a-ketoacid dehydrogenase complexes (24, 25). The identification of the kinase/phosphatase couple that régulâtes PARL cleavage will be of cridcal importance to under-stand the régulation of mitochondrial morphology in différent tissues. Moreover, the discovery of a rôle for phosphorylation in mitochondrial dynamics could also provide an explanation for the contrasting reports on the effect of proteins controUing mitochon­drial dynamics. For example, expression of OPA1 increased mito­chondrial fusion in mouse embryonic fibroblasts and in NIH 3T3 fibroblasts whereas it resulted in dramatic fragmentation in COS-7 cells (23, 26, 27). Similarly, expression of DRP1 did not lead to fission in most cell types, but it was reported to fragment organelles in an inducible HeLa cell line (28). Finally, it is tempting to speculate that tissue selectivity of the clinical phenotype of domi­nant optic atrophy and Charcot-Marie-Tooth Ha, caused by mu­tations in OPÀ1 and MFN2 respectively, could similarly be a conséquence of differential phosphorylation (6).

/3-Cleavage libérâtes a 25-aa nuclear-targeted peptide termed P|3 peptide (11). Topology of PARL now shows that this pepdde is generated in the matrix. A récent MS study has demonstrated die existence of a constant efflux of a large number of peptides from the mitochondria (29). Thèse peptides, which originale from die cleavage of proteins localized mainly in the matrix and inner membrane, range in size from 6 to 27 aa, and they are extruded to the cytosoi in an ATP- and ternperature-dependent manner (29). Whether the P/3 peptide can be exported from the matrix to die cytosoi remains to be demonstrated. However, the existence of specialized machinery for the export of matrix peptides of similar size supports diis possibility. The tact that an intégral P/3 séquence is not required for PARL-mediated mitochondria] fragmentation (Fig. 5A) is consistent with the possibility that the P0 peptide médiates mitochondria-to-nucleus signaling (11).

Materials and Methods Cell Lines and Antibodies. HEK 293 and HeLa cells were purchased from American Type Culture Collection (Manassas, VA) and maintained under standard cell culture conditions. Cells were transfected at 40% confluence with FuGENE 6. The antibodies used in this work were: polyclonal anti-GFP (Clontech, Mountain View, CA), monoclonal anti-GFP (Invitrogen, Carlsbad, CA), monoclonal anti-cytochromec and polyclonal anti-DsRed (PharM-ingen, San Diego, CA), monoclonal anti-HA (Covance, Denver, PA), mouse M2 monoclonal and rabbit polyclonal anti-FLAG. Polyclonal antibodies against the matrix inarker peroxiredoxin 3 (30) were raiscd in rabbits against the recombinant human GST-PRDX3 protein following standard immunization protocols. Sé­rum was tested for specificity by preadsorption with the antigen. Anti-PARL-N-Tenn antibody has been already described in réf. 11. Anti-PARL-C-Term was raiscd against a peptide spanning the last 12 aa of PARL conjugated to keyhole limpet hemocyanin, accord-ing to standard immunization protocols. Immunoprecipitations done with this aniisci uni were performed by covalently coupling this antiserum to protein A-agarose beads.

Constructs. The vector used to express the human PARL protein in mammalian cells was pcDNA3. The PARL-FCT and PARL-HA-FCT constructs hâve been described in réf. 11. Mutants of thèse PARL constructs were obtained by site-directed mutagen-esis. Ail mutations were confirmed by DNA sequencing.

PARL Cleavage Analysis. Cells were transfected with the indicated PARL construct, grown for 24-36 h (unless otherwise indi­cated), washed with Dulbecco's PBS, and lysed in RIPA buffer

18566 I www.pnas.org/cgi/doi/10.1073/pnas.0604983103 Jeyaraju et a/.

containing a mixture of protease inhibitors and 1 mM sodium orthovanadate. Immunoprecipitations were performed with anti-FLAG-M2 monoclonal antibody or anti-PARL-C-Term at 4°C overnight as described in réf. 31. Immunocomplexes were washed four limes with RIPA and denatured at 85°C for 4 min in Laemmli buffer. Samples were fractionated by SDS/PAGE on a 4-12% (wt/vol) 2-[bis(2-hydroxyethyl)amino]-2-(hydroxy-methyl)l,3-propanediol gel, blotted on PVDF membranes, im-munodetected by Western blot analysis, and imaged by using the Versadoc System (Bio-Rad, Hercules, CA). For MS analysis, gels were stained with colloïdal Coomassie blue. Stained bands were excised and stored at -70°C until ready for LC/MS analysis.

MS. Samples were prepared as described in réf. 15. For capillary LC/MS, samples were analyzed by directed infusion of chromato-graphically separated components on an MDS Sciex QStar XL mass spectrometer (Applied Biosystems, Foster City, CA) inter-faced with an Ultimate micropump (LC Packings, Sunnyvale, CA). Capillary columns (150 juin X 100 mm) were packed in-house with Majic C-18 reversed-phase packing material. A 150-min continuons gradient was used for séparation with buffer A [2% ACN (vol/vol)/ 0.1 % formic acid/0.01% TFA] followed by buffer B [10% isopropyl alcohol (vol/vol)/80% ACN (vol/vol)/0.1% formic acid/0.01% TFA|. Dried samples were resuspended in A buffer and loaded directty on the capillary column. Samples were sprayed at 4500V and MS along with tandem MS data were acquired on the fly by using the Analyst QS software (Applied Biosystems).

Data Analysis. Résultant data were reconstructed manually and with a Bayesian reconstruction algorithm, and they were searched against hoth theoretical peptidc and fragmentation data from the PARL séquence. Tandem data were used for web-based searches with Mascol (Matiix Science Ltd., London, U.K.). Matching tan­dem data were veriiïed manually.

Mitochondri.il Morplioloyy Analysis. Fluorescence imaging. HeLa cells were transfected as described in réf. 32. To stain mitochondria with potentially sensitive dyes, cells were incubated witli 50 nM Mit­oFluor Red 589 (Invitrogen) al 37°C for 15 min before imaging. For quantification of mitochondrial phenotypes, images were obtained with an Olympus 1X70 microscope (Olympus Canada, Markham, ON, Canada) through a 100 x objective U Plan Apochromat, NA 1.35-0.50 objective, excited at 500 nm (yellow fluorescent protein; YFP), 434 nm (cyan fluorescent protein; CFP), and 589 nm (MitoFluor Red 589) with the Polychrome IV monochrometer. The emitled light was filtered through a triple CFP/YFP/DsRed

1. Meeuscn SI., Nunnari J (2005) Curr Opin Cell HUA 17:389-394. :!. Okainolo K, Shaw JM (2005) Anrrn Rev Genêt 39:503-530. ). Y.lffe MP (2003) Nat Cell lltol 5:497-499. 1. Bossy-Wetzel E,Barsoum MJ, Godzik A, Schwarzenbacher R, Lipton SA(2003) Cuirtytn

Cell llinl 15:706-716. :>. Karbowski M, Youle RJ (2003) Cell Dealh Differ 10:870-880. ii. Chen H, Chan DC (2005) Hum Mol Oenet 14(Suppl 2):R2a3-R289. 7. Herlan M, Vogcl F, Bomhovd C, Ncupert W, Rekrbert AS (2003)/Biol Oient 278:27781-27788. :î. McQuibban GA, Saurya S. Freeman M (2003) Nature 423:537-541. 9. Pellegrini L, Passer BJ, Candies M, Lcfterov I, Ganjei JK, Fowlkcs BJ, Koonin EV,

D'Adamio l. (2001) J Alzheimer's Dis 3:181-190. 11). Koonin EV, Maknrova KS, Rogozin ffi, Davidovic L, Lctellicr MC, Pellegrini L (2003)

Génome Biol 4:1-12. 11. Sik A, Passer BJ, Koonin EV, Pellegrini L (2004) y Biol Chem 279:15323-15329. 12. Waterston RH, Undblad-Toh K, Birney E, Rogers J, Abrll JF, Aganval P, Agarwala R,

Ainscougll R, Alexandersson M. An P, et al. (2002) Nature 420:520-562. 13. Ogurcsov AY, Sunyaev S, Kondrashov AS (2004; Gemme Ka 14:1610-1616. 14. Borchers CH, Thapar R. Petrotchenko EV, 'Foires MP, Speir JP, Easlerling M, Dominski

Z, Marzlui'l' WF (2006) Proc Natl Acad Sci USA 103:3094-3099. 15. Pcrlman DH, Berg EA, O'C'onnor PB, t'oslello CE, Hu J (2005) Proc Natl Acad Sci USA

102:9020-9025. l(i, Watabe S, Kohno H, Kouyania H, Hiroi T, Yago N, Nakazawa T (1W4) JBiocimii (Tokyo)

115:648-654. 17, lyer D, Chang O, Marx J, Wei L, Oison EN, Parrnacek MS, Balasubramatryam A, Schwartz

RJ (2006) Proc Natl Acad Sci USA 103:4516-4521. 18. Lee SY, Wenk MR, Kira Y, Naira AC. De Camilli P (2004) Proc Natl Acad Sci USA

101:546-551.

pass filter. Acquired images and m u Ilich un ncl overlaying were done with TillVision IV software (TILL Photonics, Pleasanton, CA). Quantification of mitochondrial morphology was done according to mitochondrial length to width ratio. When the ratio was >3:1, the mitochondria were classifieil as tubular and rod-shaped, and when the ratio was <3:1, mitochondria were classified as fragmented. Data were obtained from 100 cells from each condition, and standard déviations were caiculated from three independent ex-periments. High-resolution images were obtained from samples transfected with PARL-FCT constructs and costained wilh mono­clonal anti-FLAG and polyclonal anti-Tom20 by using an Olympus FV1000 confocal microscope (Olympus Canada). A 100X U Plan Apochromat objective NA 1.45 was used along with the argon laser to excite the 488 nm secondary Alexa 488 and the Red HeNe laser for Alexa 647 secondary antibody. The images shown are from 10-20 compressed Z stacks taken in 0.12-jum intervais to capture «>2-/Am sections of the cell by using Kalman averaging of two scans each (33). Digitonin permeabilization. Digitonin was recrystallized as described in réf. 34 and resuspended in PBS. Transfected cells were fixed in 4% paraformaldehyde and permeabilized with increasing concen­trations of digitonin before standard immunofluorescence was performed with the indicated antibody as described previously. The data were quantified from 100 cells in three independent experi-ments (35).

Isolation of Mitochondria. Fresh human placenta was obtained with appropriate permission, cul into fragments, and washed with PBS before decanting into 2 volumes per volume of mitochondrial isolation buffer (220 mM mannitol/68 mM sucrose/20 mM Hepes, pH 7.4/80 mM KC1/0.5 mM EGTA/2 mM MgOAc/protease inhib­itors). The tissue was homogenized by using a Waring blender (Cole Palmer, Ansou, QC, Canada), and mitochondria were isolated by standard differential œntrifugation. The mitochondrial pellet was resuspended in 100 ml of mitochondrial isolation buffer with 10% glycerol, then it was snap frozen and stored al -80°C.

We thank Dr. Gordon Shore (McGill University, Montréal, QC, Canada) for the anti-Tom20 antibody and Dr. Jordan Fishman (21st Century Biochemicals) for assistance in the génération of antibodies. Tliis work was supported by grants from the Natural Sciences and Engineering Research Council of Canada, the Canada Foundation for Innovation, and the Centre de Recherche Université Laval Robert-Giffard (to L.P.); grants from the Canadian Institutes of Health Research and the Canada Foundation for Innovation (to the H.M.M. laboratory); a Fonds de la Recherche en Santé du Québec Junior-2 Scholarship (to L.P.); and a Canadian Institutes of Health Research New Investigator Award (to H.M.M.).

19. Herlan M, Bornhovd C, Hell K, Neupert W, Reichcrl AS (2004) / Cell Biol 165:167-173. 20. Sesaki H. Southard SM. Hoblw Ali, Jensen RE (2003) Biochem Bioithys lies Commun

308:276-283. 21. Cipolat S, Rudka T, Hartmann D, Costa V, Semecls L, Craessaerts K, Meizger K, Frezza

C, Annaerl W, D'Adamio L, et al. (2006) Cell 126:163-175. 22. Frezza C, Cipolat S, Martins de Brlto O, Micaroni M, Beznousscnko GV, Rudka T, Bartoli

D, Polisbuck RS, Danial NN. De Strooper B, Scorrano L (2006) Cell 126:177-189. 23. Cipolat S, Martins de Brito O, Dal ZUio B, Scorrano L (2004) Proc Natl Acad Sci USA

101:15927-15932. 24. Haï ris RA, Havres JW. Popov KM, Zhao Y, Shlmomura Y, Sato J, Jaskicwicz J, Hurley TD

(1997) Adv Enzyme Regul 37:271-293. 25. Roche TE, Baker JC. Yan X, Hiromasn Y, Gong X, Peng T, Dong J, Turkan A, Kasten SA

(2001) Prog Nucleic Acid AVi Mol Biol 70:33-75. 26. Misaku T, Miyashlta T, Kubo Y (2002) / Biol Cliem 277:15834-15842. 27. Olicbon A, Emorine U . Descoins E, Pelioquin L, Brichese I., Gas N, Gnillon E, Delettre

C, Valette A, Hauwl CT. et al. (2002) FEBS Un 523:171-176. 28. Szabadkai G. Sirnoni AM, Chami M, Wieckowski MR, Youle RJ, Rizzuto R (2004) Mol Cell

16:59-68. 29. Augustin S, Nolden M, Millier S, Hardt O, Arnold 1, Langer T (2005) J Biol Chem

280:2691-2699. 30. Wonsey DR, Zeller Kl, Dang CV (2002) Proc Natl Acad Sci USA 99:6649-6654. 31. Pellegrini L, Passer BJ, Tabaton M, Ganjei JK D'Adamio L (1999) J Biol Chem 274:21011-21016. 32. Hartler Z. Zunino R, McBrldc H (2004) Curr Biol 14:340-345. 33. Neuspiel M, Zunino R, Gangarnjti S, Rippsiein P, McBride HM (2005) / Biol Clwm

280:25060-25070. 34. Kun E, Kirslcn E. Piper WN (1979) Methods Enzymol 55:115-118. 35. Otera H, Ohsakaya S, Nagauru Z, Ishihara N, Mihaia K (2005) EMBO J 24:1375-1386.

Jeyaraju e t al. PNAS | December S, 2006 | vol. 103 | no. 49 | 18567

3.1 Objectives - part II

Following up the data published in the article mentioned above, my research work aimed at:

4) To identify structural déterminants of PARL functionally implicated in the régulation of

P-cleavage.

3.2 Results and Discussion - part II

PARL contains 7 transmembrane hélices, which embed the protein in the mitochondrial inner membrane. The portion of PARL encompassing TMH 2-7 constitutes the catalytic rhomboid domain (amino acid 121-167). This implies that the loop Connecting TMH 1 to 2, the N-terminus and the C-terminus of PARL hâve unrelated, non-enzymatic fonctions, which might be associated to a regulatory activity over PARL p-cleavage. Our previous study has shown that PARL N-terminus (amino acid 53-101), in addition to being in itself the domain subjected to P-cleavage at position Ser77,|,Ala78, has also a regulatory fonction over this processing by means of an inhibitory phosphorylation activity of amino acid S6s, T69, and S70 (Jeyaraju et al 2006a). In an attempt to find évidences suggesting the existence of additional P-cleavage regulatory domains in PARL, we performed a large-scale mutagenesis screening focusing on the other two remaining domains of the protease, the loop-1 and the C-terminus. To this goal, small deletion (3 amino acids), large deletion (8 amino acids), and single and double amino acid substitutions were introduced in loop-1 (spanning amino acid 120-168) and in the C-terminus of PARL (spanning amino acid 350-379. A total of over 80 mutants were generated. Each PARL mutant was transfected in HEK 293 cells and P-cleavage monitored by PARL immunoprecipitation and immunodetection with anti-Flag (in collaboration with my lab mate Danny Jeyaraju). An example of the results obtained by this analysis is shown in the figure hère.

19

Ce«(ysatesofHi!k293ŒlstransR»«edwkh««^-«igCT

WHd L79E A I7U74 R357L A

IPamMtag; WBanti-FI»g

Figure 2 identification of mutants at the C-terminal domain of PARL that biock b-cleavage. PARL Flag-CT is the positive control, which therefore undergoes to -cleavage; mutant L79E is the négative control, where -cleavage is blocked because the p-cleavage site of has been mutated (Sik et ai 2004).

Our data show that none of the tested mutations in loop-lof PARL, such as A141-143 and RI52A, affect P-cleavage, indicating that this domain per se has no regulatory function on this processing. This finding is consistent with previous reports showing that loop-1 binds to OPAl in a yeast two-hybrid system (Cipolat et al 2006). However, several mutations in the C-terminus of PARL impair, at différent extent, P-cleavage: A371-374, A363-366, and R357L. This indicates that the integrity and conservation of this domain, which we hâve shown to be exposed to the intermembrane space (IMS) (Jeyaraju et al 2006a), is required for the exécution of the processing of PARL N -terminus in the matrix, thereby suggesting that the régulation of this cleavage impinges on the interaction of this domain with still uncharacterized IMS factors. We conclude that the signal triggering PARL P-cleavage is transduced from the IMS to the matrix via a mechanism requiring the structural participation of PARL C-terminal domain.

20

4 Perspectives

Identification of factors binding to PARL C-terminus

Our data indicate that PARL C-terminus is a structural déterminant of P-cleavage. One mechanistic function of this domain could be to bind to a IMS factor and, by doing so, allow the processing. Studies are currently ongoing in the PelleLAB to identify such factor. If so, it would be expected that the R357L and A363-366 mutations are a loss of function in p-cleavage because they do not bind with this IMS factor. Also, it would be expected that, functionally, thèse PARL mutants will abolished mitochondria fragmentation, just like the uncleavabe L79E mutant (Jeyaraju et al 2006a). The identification of the protein binding to PARL C-term in the IMS will provide a new tool to study the mechanisms of mitochondria membranes remodelling in the mammalian System, a process that is central to the life and death of the cell.

21

5 Materials and methods

DNA préparation

Plasmid DNA préparations can be done by miniprep or PEGprep. Miniprep allows us to purify

small quantities of DNA ( 5 - 1 0 ug) from small volumes (1-2 ml) of bacterial culture in about 30

minutes. This DNA can be used for transformation and cloning purposes. PEGprep allows us to

obtain transfection grade plasmid in large quantities (~5 mg) from 250 ml of bacterial culture.

DNA miniprep

Plasmid minipreps were done using the high pure plasmid isolation kit from Roche applied

science. This procédure is used to extract plasmid DNA from bacterial cell suspensions and is

based on the alkaline lysis procédure.. The procédure takes advantage of the fact that plasmids

are relatively small supercoiled DNA molécules and bacterial chromosomal DNA is much larger

and less supercoiled. This différence in topology allows for sélective précipitation of the

chromosomal DNA and cellular proteins from plasmids and RNA molécules. The cells are lysed

under alkaline conditions, which dénature both nucleic acids and proteins, and when the solution

is neutralized by the addition of potassium acétate, chromosomal DNA and proteins precipitate

because it is impossible for them to renature correctly owing to their large size. Plasmids on the

other hand, renature and stay in solution, effectively separating them from chromosomal DNA

and proteins. The process does not require DNA précipitation, organic solvent extractions, or

extensive handling of the DNA.

Protocol: Grow bacterial cultures overnight in a rotary shaker at 200 rpm and 37° C in LB

médium (lOg/L Tryptone, 5g/L Yeast extract, and lOg/L NaCl) with the required antibiotic based

on the vector used. Centrifuge the cells at 13000 rpm for 2 minutes. Resuspend the pellet in 250

(il of suspension buffer containing RNase A followed by the addition of 250 ul of lysis buffer.

Invert the tube 2 or 3 times to facilitate even lysis of cells in the suspension. In about 2 minutes,

add 350 ul of neutralizing solution followed by gentle inversions of the tube. Place the tube in ice

for 10 minutes followed by centrifugation for 10 minutes at 13000 rpm. Remove the supernatant

carefully and transfer to the purification column followed by a minute of centrifugation. Discard

the flow through and add 750 ul of wash buffer followed by two centrifugations where the flow

22

through has to be discarded in both. Purified plasmid DNA can be eluted by adding 50 ul of

elution buffer (TE; 10 mM Tris, 1 mM EDTA) followed by centrifugation at maximal speed for a

minute.

DNA maxiprep for cell transfection (PEG prep)

Protocol: Incuba te a single bacterial colony in 150 ml of Terrifie Broth with the appropriate

antibiotic at 37°C in a rotary shaker at 135 rpm overnight. Centrifuge at 4000 rpm in swing

bucket rotor for 10 minutes. Resuspend pellet in 4ml of Solution I (50mM glucose, 25mM Tris-

HCL pH 7.6, lOmM EDTA), transfer to corning tube and add 5 ul of RNase A (lOmg/ml,

Fermentas). Add 8 ml of Solution II (0.2M NaOH and 1%SDS). Invert gently 2-3 times and add

4ml of Solution III (3M KAc, 11% v/v glacial acetic acid). Invert gently 2-3 times and place in

ice for 10 minutes. Centrifuge for 10 minutes at 4°C 14000 rpm and if supematant is still turbid,

transfer to a new tube and recentrifuge. Transfer the clear supematant to a new tube and note the

volume. Add 0.6 volumes of lsopropanol followed by inversions of the tube and incubation at -

20°C for an hour. Centrifuge at 14000 rpm at 4°C for 20 minutes. Discard supematant. Air dry

the pellet for 5 minutes and resuspend it in 600 ul of TE (lOmM Tris-HCL pH 7.6 and ImM

EDTA) and transfer to an eppendorf tube. Add 3 ul of RNase A and incubate for 30 minutes at

37°C. Centrifuge at full speed for 10 minutes and transfer the supematant to a new eppendorf

tube and note the volume. Add an equal volume of PEG (20% PEG and 2.5M NaCl) and invert

the tube few times. Incubate in ice for 30 minutes.

Centrifuge for at 14000 rpm for 10 minutes and resuspend pellet in 500 ul of TE, add one volume

of phenol7chloroform (Phenol:Chloroform:Isoamyl alcohol 25:24:1 saturated with lOmM Tris pH

8.0 and ImM EDTA; Sigma), vortex for 10 seconds, centrifuge for 5 minutes at 14000 rpm and

collect the upper layer and note the quantity. Repeat the above phenol/chloroform step once again

with the collected upper layer. Collect the upper layer again and add an equal volume of

chloroform (Sigma), vortex for 30 seconds, and centrifuge at 14000 rpm for 5 minutes and collect

the supematant. Add 0.1 volume of Na Acétate pH 5.3 followed by three volumes of 100%

ethanol and invert the tubes a few times. Centrifuge at full speed for 10 minutes and discard the

supematant. Add 600 ni of 70% ethanol and centrifuge for 10 minutes at 14000 rpm and discard

the supematant. Resuspend the pellet in 1 ml of TE. Quantify the DNA by measuring the optical

density at 260 nm.

23

Mutagenesis

Mutagenesis was performed based on the oligonucleotide based site-directed mutagenesis methodology. In this strategy, two primers are used. A mutagenic primer that introduces the desired mutation and a sélection primer containing a mutation in the récognition séquence for a unique restriction enzyme site. The two primers simultaneously anneal to one strand of the denatured double-stranded plasmid under conditions favoring the formation of hybrids between the primers and the DNA template. After standard DNA elongation, ligation, and a primary sélection by restriction digest, the mixture of mutated and unmutated plasmids is transformed into a mutS E. coli strain defective in mismatch repair. Transformants are pooled, and plasmid DNA is prepared from the mixed bacterial population. The isolated DNA is then subjected to a second sélective restriction enzyme digestion. Since the mutated DNA lacks the restriction enzyme récognition site, it is résistant to digestion. The parental DNA, however, is sensitive to digestion and will be linearized, rendering it at least 100 times less efficient in transformation of bacterial cells. A final transformation with the selectively digested DNA results in highly efficient and spécifie recovery of the desired mutated plasmid.

The advantages of this strategy are as follows:

> Does not require single-stranded vectors or specialized double-stranded plasmids.

> Does not require viral transductions.

> Does not require subcloning.

> Results in mutation efficiencies of >70-90%.

In addition, the use of T4 DNA polymerase instead of PCR reduces the risk of introducing spurious mutations. The following figure (Fig 5.1) explains schematically the entire procédure.

24

Unique Restriction -Situ

Clarmdgene

1. Dénature dsDNA

2. Re-anneal with Pnmers

SELECTION PRIMER

3. Synthesize second strand with T4 DNA Polymerase and seal gaps with T4 DNA Ligase; primary digestion with sélection restriction enzyme.

4. Transform mutS E. coli FIRST TRANSFORMATION

5. Isolate DNA from transformant pool / \

6 Secondary digestion with sélection enzyme

7. Transform £ . coli SECOND (FINAL) TRANSFORMATION

8 Isolate DNA from individual transformants to conflrm présence of deslred mutation

Fig. 3 S c h e m e of the mutagenesis protocol.

Primers design

Sélection Primer: The fonction of the sélection primer is to eliminate the original unique

restriction enzyme site. The sélection primer can be designed by incorporating one or more base

pair changes within the targeted unique restriction site. Since restriction enzymes recognize an

exact DNA séquence, one or more base pair change(s) within the récognition séquence should be

sufficient to abolish the restriction digestion. If the sélection restriction site is located within a

gène, avoid using a

sélection primer that will introduce change that could interfère with the expression of that gène

(e.g., by causing a reading-frame shift or a prématuré termination codon).

Mutagenic Primers: The function of the mutagenic primer is to introduce the desired mutation. In

our experiments on phosphorylation studies, we substituted phosphorylable amino acids (Serine,

Threonine and Tyrosine) with amino acids that either mimick phosphorylation (Asp; (Lee et al

2004a)) or are unphosphorylable (Ala).

Important:

a) The mutagenic and sélection primers must anneal to the same strand of the plasmid.

b) The distance between the sélection primer and the mutagenic primer is not critical. If

possible, however, design the sélection and mutagenic primers so that they will be evenly

spaced after annealing to the DNA template; this will allow the DNA polymerase to

extend both primers an équivalent distance. In rare cases, where the unique restriction site

and the targeted mutagenic site are very close to each other, one single primer can be

designed to introduce both mutations simultaneously.

c) In most cases, 10 nucleotides of uninterrupted matched séquences on both ends of the

primer (flanking the mismatch site) should give sufficient annealing stability, provided

that the GC content of the primer is greater than 50%. If the GC content is less than 50%,

the lengths of

d) the primer arms should be extended accordingly. The mismatch bases should be placed in

the center of the primer séquence. For optimum primer annealing, the oligonucleotides

should start and end with a G or C. The annealing strength of the mutagenic primer

should always be

e) equal to or greater than that of the sélection primer.

26

f) There can be more titan one base mismatch in the mutagenic primer. However, the

success rate for targeted mutagenesis of a plasmid is optimized when there are only one or

a few mismatches in the primer

Primers phosphorvlation

The primers must be phosphorylated at the 5' end, so that they can be ligated to the 3' end of the

newly synthesized strand.

Protocol: Resuspend primers in water at a concentration of 1 ug/ul. For each primer prépare a

reaction mix of 20 ul (final volume) composed of 2 ul of lOmM ATP (Fermentas), 2 ul of 10X

PNK Buffer A (500 mM Tris-HCL pH 7.6, 100 mM MgCl2, 50 mM DTT, 1 mM Spermidine, 1

mM EDTA), 1 ul of primer, and 1 (il of T4 Polynucleotide Kinase (PNK). Incubate for 30

minutes at 37° C and inactivate PNK and 75° C for 10 minutes. Briefly spin the tubes after this

step.

Plasmid denaturation and primers annealing

Protocol: To a final volume of 30 ul, add 1 ul of the phosphorylated sélection primer, 1 ul of the

phosphorylated mutagenic primer, 1 ul of the DNA template (100 ng/ul concentration), 3 ul of

Fermenta's Tango 10X buffer (330 mM Tris-acetate pH 7.9, lOOmM magnésium acétate, 66 mM

Potassium acétate, lmg/ml BSA). Incubate at 100 ° C for 3 minutes and immediately place in ice

for 5 minutes.

Svnthesis of the mutant DNA strand and restriction digestion sélection

Protocol: To each tube of the the primer/plasmid annealed mix, add 1 ul of T4 DNA Ligase (10

U/ul; Fermentas), 1 ul of T4 DNA Polymerase (10 U/ ul, Fermentas), 1 ul of dNTP and ATP mix

(2.5 mM dNTP and 1 mM ATP). Mix well and centrifuge briefly. Incubate at 37°C for 60

minutes. Stop the reaction by heating at 75°C for 10 minutes to inactivate the enzymes. After

allowing cooling, add 1 |xl of the required restriction enzyme to digest the unmodifïed plasmids

and incubate at 37°C for 60 minutes.

27

Transformation into MutS cells

The purpose of this step is to amplify the mutated strand (as well as the parental strand) in heat

shock-competent mutS (repair-deficient) E. coli cells (BMH 71-18 MutS; Clontech or Promega)

Protocol: Preheat a water bath to 42°C. Add 2 ul of the digested plasmid DNA digestion (see

above) to 50 ul of MutS cells and incubate on ice for 10 minutes. Heat shock cells of 42°C for 40

seconds, incubate in ice for 2 minutes, and add 500 ul of LB média. Incubate at 37°C for 45

minutes and add to a Falcon tube containing 5 ml of LB média with the required antibiotic. Shake

overnight at 37°C at 200 rpm.

Isolation of the mixed plasmid pool bv miniprep

Protocol: Do a DNA miniprep from 2 ml of E.coli MutS cells.

Sélection of the Mutant Plasmid

This procédure sélects for the mutant plasmid by complète digestion of parental (unmutated)

plasmids.

Protocol: Digest 15 ul of the miniprep done above (~1 ug) in 20 ul final volume; transform 2 ul

in 50 ul of compétent E.coli Kl 2 cells; plate 1/10 of the transformed cells in LB agar média

containing the appropriate antibiotic.

Mutants screening

Mutants are identified by sequencing. In some cases the mutagenic primer can be designed in

such a way that along with the mutation, a restriction site can be introduced or eliminated. In such

cases, restriction digestion should be performed and potential mutants analysed by agarose gel

electrophoresis.

DNA sequencing

Plasmid sequencing was done using the sequencing service from Université Laval.

Chromatogram files were analysed using Chromas Pro from Technelysium Ltd.

2S

Mammalian cell culture and transfection

HEK 293, HeLa and Cos-7 cells were obtained from American Type Culture Collection (ATCC). DMEM (Dulbecco's Modified Eagle Médium, Invitrogen) was used for HEK 293 and Cos-7 cells while MEM (Minimum Essential Médium, Sigma) was used for HeLa cells. In ail cases, the média was supplemented with 10% fetal bovine sérum, 2mM glutamine, 10 ug/ul penicillin and 10 ug/ul streptomycin. Maintain cells in culture by growing them in pétri dishes. When the cells are 100% confluent, remove and discard culture médium. Briefly rinse the cell layer with 0.25% (w/v) Trypsin- 0.53 mM EDTA solution to remove ail traces of sérum that contains trypsin inhibitor. Add 2.0 to 3.0 ml of Trypsin-EDTA solution to flask and observe cells under an inverted microscope until cell layer is dispersed (usually within 5 to 15 minutes). Note: To avoid dumping do not agitate the cells by hitting or shaking the flask while waiting for the cells to detach. Cells that are difficult to detach may be placed at 37°C to facilitate dispersai. Add 6.0 to 8.0 ml of complète growth médium and aspirate cells by gently pipetting. Add appropriate aliquots of the cell suspension to new culture vessels. Incubate cultures at 37°C. Subcultivation ratio: A subcultivation ratio of 1:2 to 1:4 is recommended Médium renewal: Every 2 to 3 days

For transfection, grow cells in 6 well plates until they reach 50% of confluence. Per every well of cells to transfert, add 1 ug of DNA (PEG prep) to 100 ul of sérum free média and add 3 ul of Fugene 6 (Roche Applied Science). Lncubate for 30 minutes and add the mix to the cells. To check transfection efficiency, every experiment included a control transfection with a plasmid expressing GFP (pEGFP-Nl, Clontech).

Cell lysates préparation

Protocol: When transfected cells reach 90-100% of confluence (24-48 hours after transfection, depending on the cell type), add 800 ul of RIPA (65 mM tris base, 0.15 M NaCl, 1% NP40, 0.25% sodium deoxycholate, ImM EDTA pH 7.4) to each well. Centrifuge cell lysate at 14000 îpm at 4°C for 30 minutes. Collect the supematant and continue with immunoprecipitation or freeze at -30°C for further use.

29

Protein gel electrophoresis

Protocol: Prépare IX SDS Running Buffer by adding 50 ml 20X NuPAGE (Invitrogen) MES or MOPS SDS running buffer to 950 ml of deionized water. Prépare the electrophoresis setup and install the precast 4-12% Bis Tris gel and add the running buffer. Add the appropriate amount of NuPAGE LDS 4X buffer (Invitrogen) and DTT (1M, 10X) and incubate at 85°C for 4 minutes. During this incubation period, remove the comb from the gel and wash the gel pockets with the running buffer. Add the appropriate amount of sample (-12 ul) per well. Add 3 ul of Seeblue Plus Prestained Ladder (Invitrogen) to one of the wells (usually the first or the last). Attach electrical supply and run at 200 Volts until the blue dye front reaches the bottom of the gel.

Western blotting

Protocol: Prépare 1 liter of IX NuPAGE Transfer Buffer by adding 50 ml 20X NuPAGE Transfer Buffer and 100 ml methanol to 850 ml deionized water. Soak blotting pads in 700 ml of IX NuPAGE Transfer Buffer. Prépare transfer membranes by moistening in methanol and then immediately adding transfer buffer. Soak the filter paper briefiy in Transfer Buffer. Place a pièce of pre-soaked filter paper on top of the gel (adhered to the bottom plate) and remove any trapped air bubbles. Tuni the plate over so the gel and filter paper are facing downwards over a gloved

hand or clean fiât surface. Place a pre-soaked transfer membrane on the gel and remove trapped air bubbles. Place another pre-soaked filter paper on top of the membrane and remove any air

bubbles. Place two soaked blotting pads into the cathode (-) core of the blot module. Carefully pick up the gel/membrane assembly and place on blotting pad in the correct orientation, so the gel is closest to the cathode core. Add enough pre-soaked blotting pads to rise to 0.5 cm over rim of cathode core. Place the anode (+) core on top of the pads. Hold the blot module together firmly and slide it into the guide rails on the lower buffer chamber. Insert the Gel Tension Wedge into the Lower Buffer Chamber and lock the Wedge into position. Fill the blot module with Transfer Buffer prepared until the gel/membrane assembly is covered. Fill the Outer Buffer Chamber with 650 ml deionized water. Place the lid on the unit and connect the electrical leads to the power supply. Perform transfer for nitrocellulose or PVDF membranes using 100 V constant for 1 hour.

30

Remove the cassette and rinse the membrane with PBST (137mM NaCl, 27 mM KO, 43 mM

Na2HP04, 14 mM KH2P04, 1% Tween-20). Proceed with blocking the membrane with 5% milk

solution made in PBST. Add the required antibody at its optimal dilution and incubate. The

following table gives the dilution and concentration of the various antibodies used:

Primary Antibody Dilution Incubation time

Anti-Flag Rabbit 1:1000 1 hour, room température

Anti-Flag Mouse 1:1000 1 hour, room température

Anti-PARL N-Term 1:1000 2 hours, room température

Anti- PARL C-Term 1:1000 Overnight 4°C

Anti- Myc Rabbit 1:1000 Overnight 4°C

Anti-GFP Mouse 1:10000 2 hours, room température

After the incubation with the primary antibody, wash the membranes with PBST thrice and add

HRP conjugated secondary antibody (1:2000 dilution; Chemicon). Incubate at room température

for 1 lir and wash for 15 minutes with PBST, changing washing solution from time to time.

Reveal blotted proteins by chemiluminescence with Super Signal West Femto (Pierce) using the

Versadoc imaging station (Biorad).

Other antibodies used during the course of the study were: polyclonal anti-GFP (Clontech,

Mountain View, CA), monoclonal anti-GFP (lnvitrogen, Carlsbad, CA), monoclonal anti-

cytochrome c and monoclonal anti-HA (Covance, Denver, PA), mouse M2 monoclonal and

rabbit polyclonal anti-FLAG.

Cloning

For PCR based cloning, KOD hot start DNA polymerase (Novagen) was used following

manufacurer's instructions. In ail cases, linearized and dephosphorylated vectors were ligated to

the phosphorylated PCR product and positive clones were identified by PCR screening DNA

sequencing. Standard cloning procédures were used throughout the course of this study.

31

Vectors The two mammalian expression vectors used in this study are pcDNA3 (for expression of FLAG/HA/cMyc-tagged proteins; Invitrogen), and pEGFPNl (for expression of GFP tagged proteins; Novagen).

32

6 Bibliography

Acehan D, Jiang X, Morgan DG, Heuser JE, Wang X, Akey CW. 2002. Three-dimensional structure of the apoptosome: implications for assembly, procaspase-9 binding, and activation. Mol Cell 9: 423-32

Adrain C, SIee EA, Harte MT, Martin SJ. 1999. Régulation of apoptotic protease activating factor-l oligomenzation and apoptosis by the WD-40 repeat région. J Biol Chem 21 A: 20855-60

Alexander C, Votruba M, Pesch UE, Thiselton DL, Mayer S, et al. 2000. OPA1, encoding a dynamin-related GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nat. Genêt. 26: 211 -5

Alirol E, James D, Huber D, Marchetto A, Vergani L, et al. 2006. The Mitochondrial Fission Protein hFisl Requires the Endoplasmic Reticulum Gateway to Induce Apoptosis. Mol Biol Cell

Ankarcrona M, Hultenby K. 2002. Presenilin-1 is located in rat mitochondria. Biochem Biophys Res Commun 295: 766-70

Ben-Shem A, Fass, D. & Bibi, E. Proc. Natl. Acad. Sci. USA (in thepress). Bier E, Jan LY, Jan YN. 1990. rhomboid, a gène required for dorsoventral axis establishment and peripheral nervous

System development in Drosophila melanogaster. Gènes Dev 4: 190-203 Brown M, Goldstein J. 1999. A proteolytic pathway that controls the cholestérol content of membranes, cells, and

blood. Proc Natl Acad Sci USA 96: 11041-8 Brown MS, Ye J, Rawson RB, Goldstein JL. 2000. Regulated intramembrane proteolysis: a control mechanism

conserved from bacteria to humans. Cell 100: 391-8 Chan D, Frank S, Rojo M. 2006. Mitochondrial dynamics in cell life and death. Cell Death Diff'er Chen H, Chan DC. 2005. Emerging functions of mammalian mitochondrial fusion and fission. Hum Mol Genêt 14

Spec No. 2: R283-9 Chen H, Chomyn A, Chan DC. 2005. Disruption of fusion results in mitochondrial heterogeneity and dysfunction. J

Biol Chem 280: 26185-92. Epub 2005 May 17. Chen H, Detmer SA, Ewald AJ, Griffin EE, Fraser SE, Chan DC. 2003. Mitofusins Mfhl and Mfn2 coordinately

regulate mitochondrial fusion and aie essential for embryonic development. J Cell Biol 160: 189-200 Cipolat S, Martins de Brito O, Dal Zilio B, Scorrano L. 2004. OPA1 requires mitofusin 1 to promote mitochondrial

fusion. Proc Natl Acad Sci USA 101: 15927-32. Epub 2004 Oct 27. Cipolat S, Rudka T, Hartmann D, Costa V, Serneels L, et al. 2006. Mitochondrial rhomboid PARL régulâtes

cytochrome c release during apoptosis via OPA1-dépendent cristae remodeling. Cell 126: 163-75 De Strooper B, Annaert W, Cupers P, Saftig P, Craessaerts K, et al. 1999. A presenilin-1-dépendent gamma-

secretase-like protease médiates release of Notch intracellular domain. Nature 398: 518-22 De Strooper B, Saftig P, Craessaerts K., Vanderstichele H, Guhde G, et al. 1998. Deficiency of presenilin-1 inhibits

the normal cleavage of amyloid precursor protein. Nature 391: 387-90 Delettre C, Griffoin JM, Kaplan J, Dollflis H, Lorenz B, et al. 2001. Mutation spectrum and splicing variants in the

OPA1 gène. Hum Genêt 109: 584-91. Delettre C, Lenaers G, Griffoin JM, Gigarel N, Lorenzo C, et al. 2000. Nuclear gène OPA1, encoding a

mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat. Genêt. 26: 207-10 Dimmer K.S, Fritz S, Fuchs F, Messerschmitt M, Weinbach N, et al. 2002. Genetic Basis of Mitochondrial Function

and Morphology in Saccharomyces cerevisiae. Mol Biol Cell 13: 847-53. Esser K, Tursun B, Ingenhoven M, Michaelis G, Pratje E. 2002. A novel two-step mechanism for removal of a

mitochondrial signal séquence in volves the mAAA complex and the putative rhomboid protease Pcpl. J Mol Biol 323: 835-43

Forte M, Bernardi P. 2006. The permeability transition and BCL-2 family proteins in apoptosis: co-conspirators or independent agents? Cell Death Diff'er 13: 1287-90

Frank S, Gaume B, Bergmann-Leitner ES, Leitner WW, Robert EG, et al. 2001. The rôle of dynamin-related protein 1, a mediator of mitochondrial fission, in apoptosis. Dev Cell 1: 515-25

Frey TG, Mannella CA. 2000. The internai structure of mitochondria. Trends Biochem. Sci. 25: 319-24 Frezza C, Cipolat S, Martins de Brito O, Micaroni M, Beznoussenko GV, et al. 2006. OPA1 controls apoptotic

cristae remodeling independently from mitochondrial fusion. Cell 126: 177-89 Guan K, Farh L, Marshall TK, Deschenes RJ. 1993. Normal mitochondrial structure and génome maintenance in

yeast requires the dynamin-like product of the MGM1 gène. Curr Genêt 24: 141-8. Gupta S, Singh R, Datta P, Zhang Z, Orr C, et al. 2004. The C-terminal tail of presenilin régulâtes Omi/HtrA2

protease activity. J Biol Chem 279: 45844-54 Herlan M, Bornhovd C, Hell K, Neupert W, Reichert AS. 2004. Alternative topogenesis of Mgml and mitochondrial

morphology dépend on ATP and a functional import motor. J Cell Biol 165: 167-73. Epub 2004 Apr 19.

33

Herlan M, Vogel F, Bornhovd C, Neupert W, Reichert AS. 2003. Processing ot'Mgml by the Rhomboid-type protease Pcpl is required for maintenance of mitochondrial morphology and of mitochondrial DN A../ Biol Chem

Ingerman E, Perkins EM, Marino M, Mears JA, McCaffery JM, et al. 2005. Dnml forms spirals that are structurally tailored to fit mitochondria. J CelIBiol 170: 1021-7

Ishihara N, Fujita Y, Oka T, Mihara K. 2006. Régulation of mitochondrial morphology through proteolytic cleavage of OPAl. EmboJ 25: 2966-77

Jeyaraju D, Xu L, Letellier MC, Bandaru S, Zunino R, et al. 2006a. Phosphorylation and cleavage of a vertebrate-specific domain of the rhomboid protease PARL regulate mitochondrial morphology. Proc Natl Acad Sci U SA. 2006 103(49):18562-7

Jeyaraju DV, Xu L, Letellier MC, Bandaru S, Zunino R, et al. 2006b. Phosphorylation and cleavage of presenilin-associated rhomboid-like protein (PARL) promûtes changes in mitochondrial morphology. Proc Natl Acad Sci USA 103:18562-7

Karbowski M, Arnoult D, Chen H, Chan DC, Smith CL, Youle RJ. 2004. Quantitation of mitochondrial dynamics by photolabeling of individual organelles shows that mitochondrial fusion is blocked during the Bax activation phase of apoptosis. J Cell Biol 164: 493-9. Epub 2004 Feb 09.

Karbowski M, Youle RJ. 2003. Dynamics of mitochondrial morphology in healthy cells and during apoptosis. Cell DeathDiffèr 10: 870-80.

Koch A, Yoon Y, Bonekamp NA, McNiven MA, Schrader M. 2005. A rôle for Fisl in both mitochondrial and peroxisomal fission in mammalian cells. Mol Biol Cell 16: 5077-86

Koonin EV, Makarova KS, Rogozin IB, Davidovic L, Letellier MC, Pellegrini L. 2003. The rhomboids: a nearly ubiquitous family of intramembrane serine proteases that probably evolved by multiple ancient horizontal gène transfers. Génome Biol 4: RI 9

Koshiba T, Detmer SA, Kaiser JT, Chen H, McCaffery JM, Chan DC. 2004. Strucmral basis of mitochondrial tetheringby mitofusin complexes. Science 305: 858-62.

Kuroiwa T, Nishida K, Yoshida Y, Fujiwara T, Mori T, et al. 2006. Structure, function and évolution of the mitochondrial division apparatus. Biochim Biophys Acta 1763: 510-21

Lee SY, Wenk MR, Kim Y, Nairn AC, De Camilli P. 2004a. Régulation of synaptojanin 1 by cyclin-dependent kinase 5 at synapses. Proc Natl AcadSci USA 101: 546-51

Lee Y.I, Jeong SY, Karbowski M, Smith CL, Youle RJ. 2004b. Rôles of the mammalian mitochondrial fission and fusion mediators Fisl, Drpl, and Opal in apoptosis. Mol Biol Cell 15: 5001-11

Liu X, Kim CN, Yang J, Jemmerson R, Wang X. 1996. Induction of apoptotic program in cell-free extracts: requiretnent for dATP and cytochrome c. Cell 86: 147-57

Mannella CA. 2006a. The relevance of mitochondrial membrane topology to mitochondrial function. Biochim Biophys Acta 1762: 140-7

Mannella CA. 2006b. Structure and dynamics of the mitochondrial inner membrane cristae. Biochim Biophys Acta 1763: 542-8

McQuibban G A, Saurya S, Freeman M. 2003. Mitochondrial membrane remodelling regulated by a conserved rhomboid protease. Nature 423: 537-41

Meeusen SL. Nunnari J. 2005. How mitochondria fuse. Curr Opin CelIBiol 17: 389-94 Mozdy AD, McCaffery JM, Shaw JM. 2000. Dnmlp GTPase-mediated mitochondrial fission is a multi-step process

requiring the novel intégral membrane component Fisl p. J Cell Biol 151: 367-80 Neuspiel M, Zunino R, Gangaraju S, Rippstein P, McBride HM. 2005. Activated Mfn2 signais mitochondrial fusion,

interfères with Bax activation and reduces susceptibility to radical induced depolarization. J Biol Chem 280: 25060-70

Okamoto K, Shaw JM. 2005. Mitochondrial morphology and dynamics in yeast and multicellular eukaryotes. Annu Rev Genêt 39: 503-36

Olichon A, Eniorine LJ, Descoins E, Pelloquin L, Brichese L, et al. 2002. The human dynamin-related protein OPAl is anchored to the mitochondrial inner membrane facing the inter-membrane space. FEBSLett 523: 171-6.

Pellegrini L, Passer BJ, Canelles M, Lefterov 1, Ganjei JK, et al. 2001. PAMP and PARL, two novel putative inetalloproteases interacting with the COOH-terminus of Presenilin-1 and -2. JAlzheimers Dis 3: 181-90

Rawson RB, Zelenski NG, Nijhawan D, Ye J, Sakai J, et al. 1997. Complémentation cloning of S2P, a gène encoding a putative métalloprotease required for intramembrane cleavage of SREBPs. Mol Cell 1: 47-57

Schrader M. 2006. Shared components of mitochondrial and peroxisomal division. Biochim Biophys Acta 1763: 531-41

34

Scorrano L, Ashiya M, Buttle K., Weiler S, Oakes SA, et al. 2002. A Distinct Pathway Remodels Mitochondrial Cristae and Mobilizes Cytochrome c during Apoptosis. Dev Cell 2: 55-67

Sesaki H, Jensen RE. 2001. UGOl encodes an outer membrane protein required for mitochondrial fusion. J Cell Biol 152: 1123-34.

Sesaki H, Soudiard SM, Hobbs AE, Jensen RE. 2003. Cells lacking Pcplp/Ugo2p, a rhomboid-like protease required for Mgmlp processing, lose mtDNA and mitochondrial structure in a Dnmlp-dependent manner, but remain compétent for mitochondrial fusion. Biochem Biophys Res Commun 308: 276-83.

Shaw JM, Nunnari J. 2002. Mitochondrial dynamics and division in budding yeast. Trends Cell Biol 12: 178-84. Shepard KA, Yaffe MP. 1999. The yeast dynamin-like protein, Mgmlp, fonctions on the mitochondrial outer

membrane to médiate mitochondrial inheritance. J. Cell Biol. 144: 711-20 Shi Y. 2002. Mechanisms of caspase activation and inhibition during apoptosis. Mol Cell 9: 459-70 Sik A, Passer BJ, Koonin EV, Pellegrini L. 2004. Self-regulated cleavage of the mitochondrial intramembrane-

cleaving protease PARL yields Pbeta, a nuclear-targeted peptide. J Biol Chem 279: 15323-9. Srinivasula SM, Ahmad M, Fernandes-Alnemri T, Alnemri ES. 1998. Autoactivation of procaspase-9 by Apaf-1-

mediated oligomerization. Mol Cell 1: 949-57 Steinmetz LM, Scharfe C, Deutschbauer AM, Mokranjac D, Herman ZS, et al. 2002. Systematic screen for human

disease gènes in yeast. Nat Genêt 31: 400-4 Sugioka R, Shimizu S, Tsujimoto Y. 2004. Fzol, a protein involved in mitochondrial fusion, inhibits apoptosis. J

Biol Chem 279: 52726-34. Epub 2004 Sep 30. Urban S. 2006. Rhomboid proteins: conserved membrane proteases with divergent biological functions. Gènes Dev

20: 3054-68 Urban S, Wolfe MS. 2005. Reconstitution of intramembrane proteolysis in vitro reveals that pure rhomboid is

sufficient for catalysis and specificity. ProcNatlAcaclSci USA 102: 1883-8 Wang Y, Zhang Y, Ha Y. 2006. Crystal structure of a rhomboid family intramembrane protease. Nature 444: 179-80 Weihofen A, Binns K, Lemberg MK, Ashman K, Martoglio B. 2002. Identification of signal peptide peptidase, a

presenilin-type aspartic protease. Science 296: 2215-8 Weihofen A, Martoglio B. 2003. Intramembrane-cleaving proteases: controlled libération of proteins and bioactive

peptides. Trends Cell Biol 13: 71-8 Wong ED, Wagner JA, Gorsich SW, McCaffery JM, Shaw JM, Nunnari J. 2000. The dynamin-related GTPase,

Mgmlp, is an intermembrane space protein required for maintenance of fusion compétent mitochondria../. Cell Biol 151:341-52

Wu Z, Yan N, Feng L, Oberstein A, Yan H, et al. 2006. Structural analysis of a rhomboid family intramembrane protease reveals a gating mechanism for substrate entry. Nat Struct Mol Biol 13: 1084-91

Yaffe MP. 2003. The cutting edge of mitochondrial fusion. Nat Cell Biol 5: 497-9 Yoon Y, Krueger EW, Oswald BJ, McNiven MA. 2003. The mitochondrial protein hFisl régulâtes mitochondrial

fission in mammalian cells through an interaction with the dynamin-like protein DLP l. Mol Cell Biol 23: 5409-20

Youle RJ, Karbowski M. 2005. Mitochondrial fission in apoptosis. NatRev Mol Cell Biol 6: 657-63 Yu T, Fox RJ, Burwell LS, Yoon Y. 2005. Régulation of mitochondrial fission and apoptosis by the mitochondrial

outer membrane protein hFis 1../ Cell Sci 118:4141-51 Zamzami N, KroemerG. 2003. Apoptosis: mitochondrial membrane permeabilization--the (w)hole story? CurrBiol

13:R71-3. Zhang J, Zheng N, Zhou P. 2003. Exploring the functional complexity of cellular proteins by protein knockout. Proc

Natl Acad Sci U S A 100: 14127-32 Zou H, Li Y, Liu X, Wang X. 1999. An APAF-1.cytochrome c multimeric complex is a functional apoptosome that

activâtes procaspase-9. J Biol Chem 274: 11549-56

35