Investigating Factors That Regulate the Direct Drp1-Mff Interaction

INVESTIGATING FACTORS THAT REGULATE THE DIRECT DRP1-MFF INTERACTION

RYAN W CLINTON

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Dissertation Advisor: Jason A Mears, Ph.D.

Department of Pharmacology

CASE WESTERN RESERVE UNIVERSITY

August 2018

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

Ryan W Clinton

Candidate for the Doctor of Philosophy Degree

(Thesis Advisor) Jason A. Mears, Ph.D.

(Committee Chair) Philip Kiser, Ph.D.

(Committee Member) Rajesh Ramachandran, Ph.D.

(Committee Member) Derek Taylor, Ph.D.

(Committee Member) Edward W. Yu, Ph.D.

Date of Defense

May 31st, 2018

*We also certify that written approval has been obtained for any proprietary material

contained therein.

DEDICATION

This work is dedicated to pursuing curiosity, seeking new knowledge, to the wonderful

friends that I’ve made during my time in Cleveland, and to my ever-supportive family.

iii

TABLE OF CONTENTS

DEDICATION iii

LIST OF TABLES viii

LIST OF FIGURES ix

ACKNOWLEDGEMENTS xii

LIST OF ABBREVIATIONS xiii

ABSTRACT 1

Chapter 1: Introduction 3

1.1 Mitochondrial membrane architecture and organization 4

1.2 Functions of the mitochondrial membranes 6

1.3 Mitochondrial dynamics 8

1.4 Mitochondrial autophagy 10

1.5 Mitochondria in apoptotic signaling 11

1.6 The dynamin family GTPases 12

1.7 Non-mitochondrial dynamin family proteins 14

1.8 Mitochondrial fusion dynamin family proteins 16

1.9 Dynamin-related protein 1 18

1.10 Mitochondrial fission factor 22

1.11 Other Drp1 partner proteins 24

1.12 Post-Translational regulation of Drp1 and Mff 26

1.13 Methods for assessing membrane protein function and assembly in vitro 28

1.14 Foundation and experimental framework 29

Figures and legends 31

Chapter 2: Using Scaffold Liposomes to Reconstitute Lipid-proximal 36 Protein-protein Interactions in vitro

2.1 Abstract 37

2.2 Introduction 37

2.3 Protocol 40

2.3.1 Scaffold liposome preparation 40

2.3.2 Use of scaffold liposomes for protein binding analysis 42

2.3.3 Use of scaffold liposomes for enzymatic assay 44

2.4 Representative results 46

2.5 Discussion 49

2.6 Acknowledgements 52

Figures and Legends 53

Chapter 3: Dynamin-Related Protein 1 Oligomerization in Solution Impairs 59 Functional Interactions with Membrane-Anchored Mitochondrial Fission Factor

3.1 Abstract 60

3.2 Introduction 61

3.3 Results 64

3.4 Discussion 77

3.5 Materials and Methods 83

3.6 Acknowledgements 90

Figures and Legends 91

Chapter 4: Mff Interacts with the Stalk of Drp1 Via a Novel VD-Occluded 104 Interface to Promote Drp1 Assembly and Membrane Constriction

4.1 Abstract 105

4.2 Introduction 105

4.3 Results 108

4.4 Discussion 124

4.5 Materials and Methods 128

4.6 Acknowledgements 138

Figures and Legends 140

Chapter 5: Conclusions and Future Directions 159

5.1 Summary 160

5.2 Recapitulating membrane fission in vitro 162

5.3 Direct effects of Mff post-translational modification on the 166 Drp1-Mff complex

5.4 Structures of a Drp1-Mff complex 167

5.5 Discerning the specific role of Mff in mitochondrial fission 171

5.6 Identifying proteins and lipids in the Mff microenvironment in vivo 174

5.7 Conclusions 175

Figures and Legends 176

References 179

vii

LIST OF TABLES

Table 4.1: Hydroxyl Radical Footprinting Oxidation Rates of Drp1 Peptides 157

Table 4.2: Hydroxyl Radical Footprinting Oxidation Rates of Drp1 Amino 158 Acid Side Chains

viii

LIST OF FIGURES

Figure 1.1: Mitochondrial membrane architecture and dynamics 31

Figure 1.2: Characteristic domains and structural features of 33 dynamin family proteins

Figure 1.3: The mitochondrial fission apparatus differs between 34 yeast and higher eukaryotes

Figure 2.1: Lipid preparation schematic 53

Figure 2.2: Methods to assess protein assembly 54

Figure 2.3: Structural assessment of Drp1 recruitment 56

Figure 2.4: Scaffold Liposome Enzymatic assay 57

Figure 3.1: The variable domain (VD) of Drp1 is a 91 negative-regulator of Mff-induced self-assembly

Figure 3.2: Coupling of Mff to topology-enforcing liposomes 93 enhances Drp1 stimulation

Figure 3.3: The VD is not essential for mitochondrial targeting 95 and subsequent fission in MEF cells

Figure 3.4: Mutations that alter the multimeric equilibrium of 97 Drp1 interfere with Mff-induced self-assembly

Figure 3.5: Removal of the VD rescues the R376E defect in 99 Mff-induced assembly

Figure 3.6: Oligomerization of Mff cytosolic domains promotes 101 Mff-induced Drp1 self-assembly

Figure 3.7: Mff selectively promotes oligomerization of 103 assembly-competent Drp1 dimers

Figure 4.1: Drp1 polymers constrict Mff-decorated 140 liposomes upon addition of GTP

Figure 4.2: Mff builds Drp1 polymers on membranes 142 via a stalk interface

Figure 4.3: Sequence conservation analysis highlights 144 Drp1 stalk loops as potential Mff-interaction sites

Figure 4.4: Mutation of stalk loop 1CS disrupts 146 Drp1-Mff interaction in vitro

Figure 4.5: Mutation of L1CS, not L3S disrupts 148 Drp1-Mff interactions

Figure 4.6: Drp1TSN functions comparably to Drp1WT 150

Figure 4.7: Drp1TSN is deficient in Mff binding and mitochondrial fission in cells 152

Figure 4.8: Hydroxyl radical footprinting reveals a VD 154 occluded surface on the stalk of Drp1

Figure 4.9: Loop 1CS oxidation is comparable between 156 Drp1WT and Drp1G363D, an assembly-defective dimer mutant

Figure 5.1: Mff recruits Drp1 to membranes to form a functional 176

membrane remodeling copolymer

Figure 5.2: AMPK phosphomimetic Mff mutants are deficient in Drp1 177 assembly and stimulation

Figure 5.3: Drp1-Mff complexes for cryo-EM study 178

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Jason Mears, for his advice over the years and for

helping me to develop and cultivate my ability to critically and logically interpret and

assess scientific data. I would also like to thank Dr. Charles Hoppel along with the other

members of the Center for Mitochondrial Diseases for the perspective that I gained at our

weekly meetings. In particular, I would like to thank Dr. Srinivasan Dasarathy for his advice

at one of these meetings which has helped me get the most out of any scientific

presentation: don’t be afraid to ask questions if you don’t understand something, or you will just be wasting your time.

I would also like to thank all of the members of my thesis committee for giving me valuable advice and perspective on my research project over the years to guide it as it developed. Finally, I would like to acknowledge all members of the Mears lab, Dr. Chris

Francy and Dr. Frances Alvarez taught me about Drp1 and mitochondria when I first joined

the lab, and I hope that I’ve been able to pass on a similar foundation to the new members

of Jason’s lab. I would also like to thank my supervisor at Blue Sky Biotech, Dr. Edward

Esposito. Without his guidance so early in my career, and the standards of excellence that

he taught me to expect from myself and others, I wouldn’t be sitting here writing this

dissertation right now.

Finally, I would like to thank my family, especially my wife Jesi who has always inspired

me to improve, and who helped me to keep persevering when it felt like my experiments

were rebelling against me over the years.

xii

LIST OF ABBREVIATIONS

CC Coiled-coil

CL Cardiolipin

Cryo-EM Cryo-electron microscopy

BSE Bundled signaling element

Drp1 Dynamin related Protein 1

ETC Electron Transport Chain

EM Electron microscopy

GC Galactosyl Ceramide

GED GTPase effector domain

GMP-PCP β,γ-Methyleneguanosine 5’-triphosphate

GTP Guanosine triphosphate

GUV Giant unilamellar vesicle

HDX Hydrogen-deuterium exchange

IMS Intermembrane space

MEF Mouse embryonic fibroblast

Mff Mitochondrial fission factor

xiii

Mfn1/2 Mitofusin 1/2

MiD49/51 Mitochondrial dynamics proteins of 49/51 kDa

MIM Mitochondrial inner membrane

MOM Mitochondrial outer membrane mtDNA Mitochondrial DNA

MMP Mitochondrial membrane potential

Opa1 Optic atrophy protein 1

PA Phosphatidic acid

PC Phosphatidylcholine

PE Phosphatidylethanolamine

PH Pleckstrin homology

PRD Proline-rich domain

PS Phosphatidylserine

SL Scaffold liposome

SUPER Supported bilayers with excess membrane reservoir

TM Transmembrane

VD Variable domain

xiv

Investigating Factors that Regulate the Direct Drp1-Mff Interaction

Abstract

RYAN W CLINTON

The complex processes of mitochondrial fission and fusion are opposing functions whose proper balance ensures optimal mitochondrial function in eukaryotic cells.

Aberrant mitochondrial morphology where one of these mitochondria-shaping processes dominates the other are commonly found in diverse pathologies, highlighting the importance of maintaining appropriate rates of both fission and fusion. Both of these competing processes are mediated by members of the dynamin superfamily of membrane-remodeling GTPases. Scission of the mitochondrial membranes is carried out by an ancient member of this protein superfamily called dynamin-related protein 1

(Drp1). While its function is required for fission, Drp1 alone is unable to mediate this complex process, and requires interaction with one or more partner proteins of the mitochondrial outer membrane to ensure fission. Chordates express several such proteins whose genetic interaction with Drp1 has been proven to be crucial for maintenance of

appropriate mitochondrial morphology. These include mitochondrial fission protein 1

(Fis1), mitochondrial fission factor (Mff), and mitochondrial dynamics proteins of 49 and

51 kilodaltons (MiD 49/51). Of these proteins, the first that was proposed to contribute

significantly to the maintenance of mitochondrial morphology in man was Mff. Due to its relatively recent discovery, its specific role(s) in this function remain unclear. To address this lack of knowledge, the primary objective of these studies was to better understand

the various factors that control the association of Drp1 and Mff, and to shed light on the

regulatory mechanisms that underlie this interaction. We have shown that the interaction

between Drp1 and Mff is mediated by the stalk of Drp1, and not the variable domain (VD)

as was previously thought. We also demonstrated the utility of mitochondrial outer

membrane-like scaffolding liposomes as a template for studying the interaction between

Drp1 and its various membrane-bound partner proteins. Finally, we sought to better

understand the interaction between Drp1 and intrinsically disordered polypeptides such

as the VD and Mff, and found that Mff co-assembles with Drp1 to form a membrane

remodeling copolymer. Taken together, this work provides a crucial foundation for

comprehending the role played by Mff in aiding Drp1-mediated mitochondrial

remodeling.

CHAPTER 1: THE ROLE AND REGULATION OF DRP1 IN MITOCHONDRIAL FISSION

1.1 Mitochondrial membrane architecture and organization

Mitochondria are an essential component of cellular metabolism in eukaryotes that participate in multiple pivotal roles including lipid metabolism, generation of cellular energy, and regulating the progression of apoptosis. While many membrane- encompassed organelles within cells function with a degree of specialization, mitochondria are unique among these as a consequence of their origins. Specifically, the mitochondria are hypothesized to have derived from alphaproteobacteria that were engulfed by an anaerobic ancestor of modern eukaryotes, evaded digestion, and entered an endosymbiotic relationship (Andersson et al. 1998; Fitzpatrick, Creevey, and

McInerney 2006; Thrash et al. 2011). While mitochondria supplement the metabolic processes of its host cell, this organelle maintains several features that are characteristic of its bacterial origins such as its self-regulated genome (mtDNA), unique membrane lipid composition, and specialized membrane ultrastructure.

The mitochondrial membrane architecture is one of the most widely recognizable ultrastructural features of the eukaryotic cell and is characterized by a highly convoluted inner membrane (mitochondrial inner membrane or MIM) encompassed by a spheroid outer membrane (mitochondrial outer membrane or MOM). This overall assembly takes on a complex form that give rise to several distinct membrane compartments and microenvironments (Figure 1.1A). First, regions of contact between the inner and outer mitochondrial membranes have been described and are stabilized by the mitochondrial contact site and cristae organizing system (MICOS) that tethers the two membranes in close proximity. This allows for the coupling of transport across the mitochondrial

membranes as well as facilitating lipid transport among these lipid compartments

(Brdiczka, Zorov, and Sheu 2006; Connerth et al. 2012). In addition to facilitating transport

across the mitochondrial membranes, this complex acts as a tether for the cristae, or

invaginations of the inner membrane, that form the characteristic folded morphology of the MIM. At the necks of these cristae, called cristae junctions, both MICOS and other proteins of the mitochondrial inner membrane scaffold and stabilize these assemblies.

These complexes effectively segment the space between the membranes into three distinct compartments: the intermembrane space, the intracristal space, and the matrix.

Thus, these discrete mitochondrial partitions allow for several distinctly specialized functions of the mitochondria to occur in isolation from each other.

The ultrastructural features as well as the membrane composition of these membranes enable distinct biological functions (Figure 1.1A). The MOM is composed of roughly 50% lipids and 50% proteins which is comparable to the makeup of other cellular membranes such as the plasma membrane (Krauss Stefan 2001). The lipids of this membrane are predominantly zwitterionic phospholipids, such as phosphatidylethanolamine (PE) and phosphatidylcholine (PC), but an important minority of these membranes are comprised of anionic lipids and sphingolipids, including phosphatidylserine (PS), phosphatidylinositides (PIPs), phosphatidic acid (PA), cardiolipin

(CL), and sphingomyelin. In contrast, the MIM is enriched for proteins when compared to the MOM, and the atypical dimeric phospholipid, cardiolipin, is much more abundant in this membrane than the outer membrane (Comte, Maǐsterrena, and Gautheron 1976; de

Kroon et al. 1997; Chu et al. 2013). Interestingly, the mitochondrial membranes

(especially MIM) are the only known eukaryotic membranes containing cardiolipin, which

is most commonly found in bacteria (Mileykovskaya and Dowhan 2009). This atypical lipid

has functional significance in the MIM as it is known to interact with and stabilize several

of the protein complexes of the electron transport chain, and disruption of CL synthesis

leads to severe metabolic disorders such as Barth’s syndrome (Gebert et al. 2009;

Schlame and Ren 2006).

While the presence of cardiolipin on the MOM is a controversial topic in mitochondrial

research, it is known that apoptotic stressors enrich cardiolipin in the MOM (Chu et al.

2013). The function of multiple mitochondrial proteins including phospholipid scramblase

3 (PLS3) and nucleoside diphosphate kinase (NDPK-D) are implicated in CL translocation

from the MIM to MOM (J. Liu et al. 2003; Schlattner et al. 2013). In fact, phospholipases

of the MOM rapidly converts cardiolipin into PA (Choi et al. 2006), which suggests that CL is in fact translocated to the MOM, but that its abundance may be limited to specific microdomains due to its rapid degradation. Overall, the distinct segregations and strict

regulation of local lipid composition in the mitochondria allows for the

compartmentalization of function among these many specialized microenvironments.

1.2 Functions of the mitochondrial membranes

The asymmetric distribution of membrane components among the mitochondrial

membranes allows for distinct biological functions within these two membrane compartments. The MIM is the site of one of the most fundamentally important

metabolic functions of the mitochondria: oxidative phosphorylation. This is an aerobic

process where NADH (Hatefi, Haavik, and Griffiths 1961; Lenaz et al. 2006) and succinate

(Ziegler and Doeg 1962; Rutter, Winge, and Schiffman 2010) derived from various cellular

metabolic processes are oxidized. The released electrons are passed through several

multi-protein complexes of the electron transport chain (ETC) which ultimately results in

the reduction of O2 into H2O (Hatefi et al. 1962; Heinemeyer et al. 2007). The ETC complexes undergo conformational changes as a consequence of the electron flow which

pumps protons from the matrix into the intracristal space, resulting in an H+ gradient across the MIM (mitochondrial membrane potential or MMP). A protein called ATP synthase (also referred to as Complex V) is permeable to protons which return to the matrix and induce a conformational change in ATP synthase resulting in the synthesis of adenosine triphosphate (ATP) from adenosine diphosphate (ADP) and inorganic phosphate (Fernández-Morán et al. 1964; MacLennan and Asai 1968; Davies et al. 2012;

Jonckheere, Smeitink, and Rodenburg 2012). The composition of the MIM enables this process because it is impermeable to most solutes, and transport across it requires integral membrane proteins that selectively regulate solute concentrations within the matrix as predicted by the chemiosmotic theory underlying oxidative phosphorylation

(Mitchell 1961).

The MOM, on the other hand, is relatively permeable to a variety of ions, metabolites, and proteins via nonselective channels such as VDAC (Bayrhuber et al. 2008) and protein translocase complexes (Bausewein et al. 2017). Moreover, it serves as a signaling nexus between the bioenergetic core of the mitochondria and its host cell to communicate the

7 bioenergetic status of the organelle (D. P. Narendra et al. 2010) and to interact with cytoskeletal elements. These cytoskeletal interactions are crucial for subcellular distribution of mitochondria (Boldogh et al. 1998; Hollenbeck and Saxton 2005; Glater et al. 2006), and for their recruitment to subcellular locations with high energy demand

(Sajic et al. 2013). Thus, the interaction of the MOM with the cytoskeleton allows for a dynamic regulation of mitochondrial localization and function.

1.3 Mitochondrial dynamics

Mitochondria are not static organelles, but instead are both motile and dynamic within the cell. The morphology of a cell’s mitochondria is dictated by the relative rates of mitochondrial division (fission) and the merging of individual mitochondria (fusion)

(Figure 1.1B). Mitochondrial fusion involves joining the MIM and/or MOM of two distinct mitochondria, which results in mixing of both lipid bilayers as well as the soluble components of the intermembrane spaces and matrix to create a new, larger organelle.

In contrast, mitochondrial fission involves the scission of both MOM and MIM resulting in two independent daughter mitochondria, which is reminiscent of bacterial replication, further reinforcing the bacterial origin of this organelle. Interestingly, sites of mitochondrial fission are marked by contact between the endoplasmic reticulum (ER) and the mitochondria (Friedman et al. 2011). In fact, the ER-mitochondrial contact sites also mark sites of mitochondrial chromosome localization, and the replication of this DNA at

these sites serves as a reliable predictor for sites of mitochondrial fission (Lewis,

Uchiyama, and Nunnari 2016).

The overall morphology of a cell’s mitochondria differs among various tissue lineages, but dysregulation of mitochondrial fusion and fission is often deleterious to the entire organism. Aberrant mitochondrial morphology is a hallmark of many varied human diseases (Waterham et al. 2007; X. Guo et al. 2013; Vanstone et al. 2016; Luo et al. 2015).

Excessive mitochondrial fusion can be characterized by the presence of an excessively

interconnected mitochondrial reticulum, which is most commonly found in patients with

de novo mutations in the proteins responsible for mitochondrial fission that result in their

dysfunction. Cells from these patients exhibit excessively interconnected mitochondrial

networks and the affected individuals often present with severe neurological and

neurodevelopmental defects (Gerber et al. 2017; Vanstone et al. 2016) that can lead to

death (Chang et al. 2010; Waterham et al. 2007; G. Yoon et al. 2016). More prevalently,

human disease is associated with excessive mitochondrial fission, sometimes referred to

as fragmentation, which is characterized by the appearance of small punctate

mitochondria. The mutation of mitochondrial fusion proteins can result in this

dysregulated mitochondrial morphology, and can produce a variety of neurological

disorders (Gerber et al. 2017; Alexander et al. 2000; Züchner Stephan et al. 2006; Züchner

et al. 2004). Recently, an increasing body of research has observed fragmented

mitochondria as a consequence of neurodegenerative diseases, ischemic insult, and many

types of cancer (Xie et al. 2015, 1; Kim, Yun, and Yun 2018, 1; Tanwar et al. 2016). Notably, inhibition of excessive mitochondrial fission in these diseases by genetic (Zhao et al. 2012;

Qian, Wang, and Van Houten 2013) and pharmacologic (Cassidy-Stone et al. 2008; Qi et al. 2013; Mallat et al. 2018) approaches have been shown to attenuate the severity of these diseases. Thus, modulation of mitochondrial morphology represents a novel therapeutic strategy for a broad range of human diseases.

1.4 Mitochondrial autophagy

In addition to motility within the cytoplasm and dynamic morphology, another facet of mitochondrial dynamics is the constitutive process of mitochondrial turnover, called mitophagy. This process is a specialized form of macroautophagy whereby an autophagosome selectively engulfs one or more mitochondria, fuses to a lysosome, and degrades the entire organelle. A well-described mechanism targeting dysfunctional mitochondria for mitophagy involves a protein called PTEN-induced putative kinase 1

(PINK1). This protein is constitutively produced in the cytosol, and targeted to the mitochondria by an N-terminal signal peptide (Zhou et al. 2008, 1; Silvestri et al. 2005). In healthy mitochondria with intact MMP, PINK1 is partially imported through the MOM where it is cleaved by mitochondrial proteases, resulting in its degradation (Becker et al.

2012; Jin et al. 2010, 1; Yamano and Youle 2013). In the absence of sufficient MMP, PINK1 instead inserts into the MOM and is stabilized (D. P. Narendra et al. 2010). Here it autoactivates and recruits a cytosolic E3 ubiqutin ligase called parkin to the MOM. Parkin conjugates ubiquitin to various target proteins of the MOM, ultimately resulting in their degradation (N. C. Chan et al. 2011; Ordureau et al. 2014; Glauser Liliane et al. 2011),

which in turn also leads to recruitment of autophagosome initiating proteins involving

p62 and LC3 (Pickrell and Youle 2015; Yamano, Matsuda, and Tanaka 2016). This process

can be chemically stimulated using ETC uncoupling small molecules including the protonophore, carbonyl cyanide m-chlorophenyl hydrazine (CCCP), which abolishes MMP

(D. Narendra et al. 2008) and activates PINK1/Parkin-mediated mitophagy. Interestingly, the proteins responsible for mitochondrial fission appear to be dispensable for certain mitophagy-driving stressors (Burman et al. 2017) while they are required for others (Y. S.

Park, Choi, and Koh 2018, 5). Thus, more research is clearly required to discern between these two types of mitophagy, and to define the differences that lead to these distinct mechanisms of autophagy-related fission.

1.5 Mitochondria in apoptotic signaling

Apoptosis is a tightly regulated cellular signaling program that results in an

orchestrated and orderly form of cell death that can be initiated by both intrinsic and

extrinsic factors. Extrinsic apoptosis is instigated by the activation of death receptors on

the plasma membrane by extracellular ligands, whereas intrinsic apoptosis is triggered by

a myriad of cell stressors including hypoxia, a lack of nutrients, and toxins. Both intrinsic

and extrinsic signals ultimately lead to the formation of pores in the MOM called

mitochondrial permeability transition pores (MPTP), that allow for the leakage of large

mitochondrial proteins into the cytosol (Tait and Green 2010). Concurrently, the cristae

are remodeled allowing for the rapid release of cytochrome c, an intracristal protein

involved in ETC function, into the cytosol (Goldstein et al. 2000, 2005). This released cytochrome c drives the assembly of the apoptosome, a scaffold for caspase-family proteases, ultimately resulting in breakdown of the entire cell and recruitment of nearby phagocytic cells to remove any remaining debris (Kothakota et al. 1997; Bratton et al.

1997; Fadok et al. 2001). Interestingly, several studies have identified a connection between mitochondrial fission and apoptotic sensitivity. During the formation of MPTP,

Bax proteins translocate from the cytosol to the MOM (Wolter et al. 1997) and along with

Bak proteins undergo conformational rearrangements to forms pores in the MOM

(Westphal et al. 2014). Interestingly, in cells lacking a key regulator of mitochondrial fission, dynamin-related protein 1 (Drp1), apoptotic signaling is delayed (P. Wang et al.

2015; Montessuit et al. 2010, 1). Thus a clear connection between mitochondrial morphology and apoptosis exists, but the mechanistic connection is still unclear.

1.6 The dynamin family GTPases

Dynamin family proteins (DFPs) mediate diverse membrane-remodeling processes throughout the cell. Unlike the classic signaling GTPases (e.g. Ras family members), DFPs are mechanoenzymes. Specifically, the chemical energy of GTP hydrolysis induces dramatic conformational changes throughout the DFP that are transmitted as force to

target membranes. This enzyme family has three characteristic domains (Figure 1.2A) that

represent the minimal mechanoenzymatic core unit conserved among all members

dubbed the GTPase domain, middle domain, and GTPase effector domain (GED). The

catalytic GTP-hydrolyzing domain of DFPs (Figure 1.2 – green) is similar at its core to that of signaling GTPases, but differs in peripheral regulatory elements. In place of the switch helix found in signaling GTPases, DFPs have a 3-helix bundle (i.e. bundled signaling

element or BSE; Figure 1.2 – purple) that undergoes conformational changes in response to the GTP-binding state of the GTPase domain. The common tertiary structures of these proteins (Figure 1.2B) consist of a globular GTPase domain poised atop the BSE that is in turn connected to a structural domain dubbed the ‘stalk’ (Figure 1.2 – light/dark blue).

This domain consists of 4 extended alpha helical stretches from the middle domain and

GED that fold together into a compact, rod-like structure which serves as a platform for self-assembly into a variety of oligomers.

Functionally, DFPs are distinguished from small GTPases based on their rate of catalytic activity and the affinity for nucleotide substrates. While signaling GTPases have a high (picomolar) affinity for guanosine nucleotides (B. Ford et al. 2009), DFPs have a relatively low (micromolar) affinity and readily dissociate from these nucleotides

(Macdonald et al. 2016; Binns et al. 2000). Furthermore, the basal rate at which DFPs hydrolyze GTP (1-2 µM GTP min-1 µM DFP-1) is also much higher than that of the small

GTPases (less than 1 µM GTP min-1 µM enzyme-1). Therefore, unlike signaling GTPases

which require accessory proteins called guanosine nucleotide exchange factors (GEFs)

and GTPase activating proteins (GAPs) to facilitate GDP release and GTP hydrolysis

respectively, DFPs accomplish these functions alone. Overall, these biochemical

differences attest to their distinct primary functions; while signaling GTPases bind and

maintain nucleotide state as on/off switches, DFPs are mechanoenzymes that utilize GTP

hydrolysis to do work.

1.7 Non-mitochondrial dynamin family proteins

Many DFPs, including the protein family’s namesake, do not function at or within the mitochondria. Originally, dynamin (Dyn1) was isolated from cow brain extracts as a microtubule-associated polypeptide (Shpetner and Vallee 1989). The authors found that dynamin formed crosslinks between microtubules and hypothesized that it played a role in mediating microtubule sliding. Dynamin (specifically Dyn1) was soon found to be a key protein involved in the internalization of clathrin coated vesicles (Bliek et al. 1993). In fact, a dynamin mutant identified in Drosophila melanogaster displayed a temperature- dependent paralysis linked to defective synaptic vesicle endocytosis (Grigliatti et al.

1973). Dynamins are targeted to clathrin coated pits via interactions with both lipids and proteins. A pleckstrin homology (PH) domain at the base of the stalk mediates interactions with PIPs, especially PIP2. In addition, a carboxy-terminal proline-rich domain

interacts with the SH3 domains (Src-homology domain 3) of proteins responsible for initial

membrane deforming events (BAR-domain proteins) as well as signaling molecules that

localize to clathrin coated pits. Several structural studies of dynamins focused on

understanding the structure and nucleotide-induced changes in a minimal fragment of

dynamin which consisted of the GTPase domain and BSE dubbed the G-G (GTPase-GED)

(Chappie et al. 2011). These highlighted the dramatic conformational changes that

14 nucleotide binding and hydrolysis elicit in dynamin. Subsequent structural studies revealed the intact structure of dynamin (M. G. Ford, Jenni, and Nunnari 2011; Reubold et al. 2015; Faelber et al. 2011), revealing (in addition to the typical dynamin family domains) a PH domain protruding from the stalk opposite the GTPase domain (Dyn1 is shown in Figure 1.2B – right). Also evident within the stalk were several self-assembly motifs (named Interfaces 1-3) numbered in proximity to the GTPase domain. Cryo-EM studies of dynamin assembled on flexible PS membranes revealed a densely packed helical polymer (Chappie et al. 2011) that constricts upon GTP binding/hydrolysis (Anna

C. Sundborger et al. 2014; P. Zhang and Hinshaw 2001). Taken together with previous crystallographic studies, these cryo-EM studies indicate that membrane constriction to the point of scission is mediated by a combination of polymer constriction, membrane- destabilizing conformational changes of the PH domain, and coincident disassembly of the dynamin polymer.

Another notable and functionally unique branch of the dynamin family is the myxovirus resistance proteins (MxA and MxB). This family of proteins were originally identified in studies of inherited resistance to influenza in mice (J. Lindenmann 1962; Jean

Lindenmann, Lane, and Hobson 1963; Horisberger, Staeheli, and Haller 1983; Staeheli et al. 1986). Unlike other dynamins, the Mx proteins lack any modular protein-protein or protein-lipid interacting domains. In essence, MxA/B are the minimal essential dynamin family member. Interestingly, while Mx proteins lack a modular lipid interacting domain, they are able to bind and deform anionic liposomes (Accola et al. 2002). Also unlike other dynamins, the central function of Mx proteins is not related to cellular membrane

remodeling. Instead, these proteins are thought to interact with viral proteins and

interfere with their intracellular transport (Kochs and Haller 1999a, 1999b). In contrast to other DFPs, the expression of these proteins is transient and is robustly induced by interferon signaling that results from viral infection (Ronni et al. 1993; Horisberger,

Staeheli, and Haller 1983). Interestingly, while the sequences of these proteins are 64% identical, they appear to differ in antiviral activity and subcellular localization. The cytoplasmic isoenzyme, MxA, opposes cellular infection by many viruses including influenza viruses (Zimmermann et al. 2011; Pavlovic et al. 1995). MxB in stark contrast to

MxA instead localizes to the nucleus, but its antiviral function opposing HIV infection remained elusive until much more recently (Z. Liu et al. 2013; Kane et al. 2013). Clearly, the Mx proteins are a crucial component of innate cellular immunity to viral infection, and represent a fundamental dynamin family protein that provides insight into a natural minimal mechanoenzymatic dynamin.

1.8 Mitochondrial fusion dynamin family proteins

The fusion of both the outer and inner mitochondrial membranes is a topologically

unique challenge that is solved using several dynamin family proteins. Fusion of the outer

membrane is mediated by two related isoenzymes called ‘mitofusins’ (Mfn1/2). Contrary

to the cellular distribution of other DFPs, these proteins are anchored via transmembrane

domain(s) that position them in the MOM with the majority of the protein facing the

cytosol. These two proteins were originally identified in drosophila as functionally distinct

enzymes that control mitochondrial morphology in separate cell types (Hales and Fuller

1997; Dorn et al. 2011). On the contrary, these proteins are often coincidently expressed

in mammals (Santel and Fuller 2001), and have some functional redundancies (H. Chen et

al. 2003). Further highlighting the crucial role of these proteins, the loss of these proteins

is embryonic lethal in mice (H. Chen et al. 2003, 2), and even their mutations in humans

cause various neurological disorders (Züchner et al. 2004; Züchner Stephan et al. 2006).

Notably, while ex vivo mitochondrial fusion has been achieved (Brandt et al. 2016), the

fusion activity of purified mitofusins has not yet been observed in vitro. Furthermore, the

tertiary organization and transmembrane architecture of these proteins has recently

been called into question (Mattie et al. 2017). Taken together, these highlight the need

for further structural and functional investigation to truly understand the mechanism

underlying mitofusin-mediated membrane fusion.

The MIM is fused and shaped by a separate protein of the dynamin superfamily named Opa1. This name stems from its connection to autosomal dominant optic atrophy, which can result from one of many mutations in this enzyme. In contrast to the two isoenzyme-coding genes responsible for outer membrane fusion, Opa1 is translated from a single gene and undergoes a myriad of alternative splicing events leading to 8 distinct

splice variants (Delettre et al. 2001). These isoforms differ in alternative exon inclusion in the N-terminus of the protein that alter its proteolytic processing. Under normal conditions the transmembrane domain of Opa1 is anchored in the MOM, and the protein is enriched at the cristae junctions (Figure 1.1A). Oligomerization of Opa1 stabilizes the cristae junctions and precludes premature cytochrome c release (Frezza et al. 2006).

Under stress conditions, Opa1 is cleaved by proteases of the intermembrane space such

as Yme1L and Oma1 (Naotada Ishihara et al. 2006, 1), which produces a soluble form of

Opa1 that is biochemically distinct from intact Opa1. In fact, a recent study (Ban et al.

2017) indicated that membrane contacts that are homotypic (Opa1 interacts with Opa1

on an opposing membrane) are stably tethered whereas heterotypic membrane contacts

(Opa1 interacts with CL on an opposing membrane) undergo full membrane fusion. These distinctions can explain the difference in Opa1 function at cristae junctions and in MIM fusion. Despite this wealth of biochemical information, there is a lack of structural insight to provide context for this functional data, thus highlighting a need for cryo-EM or crystal structures to further complete this scientific foundation.

1.9 Dynamin-related protein 1

The mechanical force required for mitochondrial membrane fission is supplied by the dynamin family protein: Drp1. Originally described as a protein homologous to the yeast proteins Dnm1p and Vps1p (Shin et al. 1997), Drp1 was soon identified as a controller of mitochondrial morphology (Smirnova et al. 1998). It should be noted that Drp1 and its homologues are also crucial for peroxisomal membrane fission, but for the purposes of this dissertation we will only discuss its role in mitochondrial fission. The central role of

Drp1 is reinforced by findings in mammalian cells that lack Drp1 where mitochondrial fission is abolished (Loson et al. 2013; Osellame et al. 2016), and in mice where the knockout of Drp1 was found to be lethal due to defects in several crucial developmental

processes (Wakabayashi et al. 2009; N. Ishihara et al. 2009). Furthermore, several mutations of Drp1 have been identified in human patients that present with encephalopathy, neurodevelopmental defects, optic atrophy, and early neonatal death

(Waterham et al. 2007; G. Yoon et al. 2016; Vanstone et al. 2016; Sheffer et al. 2016;

Gerber et al. 2017).

As a member of the dynamin superfamily, Drp1 maintains the 3 core DFP domains,

and additionally includes a unique intrinsically disordered regulatory domain called the

variable domain or VD (Figure 1.2A). Alternative splicing of Drp1 results in at least 6 distinct splice variants (Strack, Wilson, and Cribbs 2013) that differ by combinations of a

13 amino acid exon (A-insert) in the GTPase domain or two exons in the VD that encode

up to 37 amino acids inserted (B-insert). These alternative splice variants exhibit distinct

tissue distributions and those with the full B-insert tend to be enriched in neuronal

tissues. Further, these splice variants seem to alter the geometry of macromolecular

assemblies formed by Drp1 (Macdonald et al. 2016).

Structurally, Drp1 is similar to many other dynamin family members (Figure 1.2B).

Structures solved with x-ray crystallography of a nearly intact Drp1 construct highlight the

expected architecture of a GTPase domain connected to the stalk (comprised of the

middle and GED) by the BSE (Fröhlich et al. 2013). Additionally, this structure highlighted

interfaces (depicted in Figure 4.8A) that underlie a key feature of Drp1: its quaternary

assemblies. These crystallographic studies highlight the crucial stalk interface 2, which

mediates Drp1 dimerization. An interesting and unexpected finding in these

crystallographic studies was the identification of a novel interface, stalk interface 4, that

19 is absent in all other crystal structures of DFPs. While the role of interface 4 is unclear based on functional assessments (Fröhlich et al. 2013), proteins with mutation in this interface are unable to mediate mitochondrial fission in cells. A potential role for this interface may lie in maintenance of a dynamic equilibrium of Drp1 oligomers, which we propose based on our study in Chapter 4. This equilibrium is based on dimeric building blocks which dynamically assemble into larger oligomers and subsequently disassemble via yet unknown interfaces (Macdonald et al. 2014). This equilibrium serves to generate a dynamic reserve of Drp1, and to alter its affinity for lipid and protein partners of the

MOM thus regulating its function.

While the conformational mechanisms underlying the contractile activity of dynamin have been well-established, similar studies are limited for Drp1. Crystallization of the Drp1

GTPase-GED fusion protein (G-G construct) shed further light on distinctions between

Drp1 and other DFPs. In contrast to other family members, the BSE of the Drp1 G-G construct does not undergo drastic conformational rearrangements upon GTP binding

(Wenger et al. 2013). This correlates well with observations of Drp1 constriction of PS membranes, where GTP binding elicited a conformational stabilization, not constriction

(Francy et al. 2015). In fact, the lipid-associated self-assembly properties of Drp1 are further distinguished from other DFPs by recent cryo-EM studies. Unlike the classic dynamin, Drp1 forms a relatively sparse protein layer enveloping underlying lipid bilayers.

Dynamin assembles via stalk interfaces 1, 2, 3, and GTPase domain dimerization. In contrast, Drp1 assembles via only stalk interface 2 and GTPase domain interfaces (Francy et al. 2017). These contrasting assembly properties highlight other intrinsic differences

that exist between Drp1 and dynamin, and bring into question other activities that have

been assumed to be conserved among all DFPs.

Similar to other DFPs, Drp1 accumulates on the surface of liposomes containing

negatively charged lipids, deforms the liposomes into tubules, and forms a helical polymer

(Francy et al. 2015; Y. Yoon, Pitts, and McNiven 2001; Fröhlich et al. 2013). While PS-

containing membranes are readily deformed by Drp1, PS is not abundant in the MOM and

as such does not represent a relevant lipid template for Drp1. In fact, recent studies have

identified cardiolipin as a more physiologically relevant lipid that stimulates Drp1 activity more robustly than PS (Macdonald et al. 2014; Bustillo-Zabalbeitia et al. 2014).

Interestingly, while the variable domain is the site of lipid interaction with Drp1,

constructs lacking the VD associate with and deform lipid bilayers containing PS (Francy

et al. 2015), but do not form helical oligomers on nanotubes containing CL (Francy et al.

2017).

In the cell Drp1 is predominantly cytosolic, and is reversibly recruited to the mitochondria where it can self-assemble into a fission-competent polymer. In addition to

the mitochondria, Drp1 has been found to interact with various elements of the

cytoskeleton (Strack, Wilson, and Cribbs 2013; W. K. Ji et al. 2015), the endoplasmic

reticulum (W.-K. Ji et al. 2017), and the peroxisome (Koch et al. 2003). While Drp1 is the

mechanochemical component that provides the contractile force for membrane scission,

it cannot act alone. In fact, a variety of membrane-tethered partner proteins have been

identified that directly interact with Drp1 and drive its recruitment and function at target

organelles.

1.10 Mitochondrial fission factor

The study of Drp1 function in mitochondrial fission in mammalian cells was initially

confounded by difficulties in identifying a key partner protein of the MOM. The

mitochondrial fission apparatus in budding yeast was well-established at this point (Figure

1.3A), but key protein factors of this complex were absent in metazoan cells. The key

differences between the yeast and metazoan will be highlighted in the upcoming section

regarding other fission partner proteins (Chapter 1.11, Figure 1.3B). The first metazoan-

specific partner protein was identified in a siRNA screen in D. melanogaster cells and was

dubbed mitochondrial fission factor (Mff). This protein lacks homologues in yeast and lower eukaryotes, but is absolutely conserved in metazoans. In man, Mff is alternatively spliced to give rise to at least 9 distinct splice variants ranging from 25-39 kilodaltons

(Gandre-Babbe and van der Bliek 2008), but the tissue distribution and distinct functions of these isoforms has yet to be described. The crucial role played by Mff in organismal development and health is highlighted by studies demonstrating that its deletion in mice results in early death due to cardiomyopathy (Hsiuchen Chen et al. 2015), and

furthermore in mutations in man are associated with severe neuromuscular defects

(Shamseldin et al. 2012). In complementary cellular studies, a loss of Mff was shown to

decrease or ablate mitochondrial fission, further highlighting the crucial role in fission

played by this protein (Loson et al. 2013; Osellame et al. 2016).

The function of Mff at the mitochondria and peroxisome requires a carboxy-terminal

transmembrane helix that localizes it to the MOM and peroxisomal membrane (Gandre-

Babbe and van der Bliek 2008). In addition to this transmembrane segment, preliminary

examination of the amino acid sequence (Figure. 1.3C) also identified two well-conserved repeated amino acid regions referred to repeats 1 and 2 in addition to a juxtamembrane coiled-coil motif (Gandre-Babbe and van der Bliek 2008) that enables tetramerization of

Mff (R. W. Clinton et al. 2016). There is currently no structural information available for

Mff, which may be due to a lack of a well-defined secondary and tertiary structural

elements. In fact, analysis with bioinformatics prediction software packages to identify

secondary structural elements (JPred) and intrinsically disordered polypeptide segments

(IUPred) suggest that Mff is predominantly intrinsically disordered with several short

regions of alpha-helical propensity. Interestingly, while the initial study that described this

protein demonstrated a genetic linkage between Drp1 and Mff, no physical interaction

was demonstrated. Subsequent studies were able to bridge this gap, and demonstrated

that the interaction between Drp1 and Mff is transient. In fact, covalent linkage of these

proteins with crosslinking agents is required to capture the Drp1-Mff complex (Otera et

al. 2010). In spite of evidence demonstrating a direct interaction between these proteins

in vitro and in vivo, the rate of GTP hydrolysis by Drp1 was unaffected by the presence of

Mff (Otera and Mihara 2011; Koirala et al. 2013). More recently, studies from our group

demonstrated that when the Drp1-Mff complex is reconstituted at a lipid interface, Drp1

undergoes Mff-dependent assembly and exhibits stimulated activity (R. W. Clinton et al.

2016). As is evident, there is a scarcity of information regarding the function of Mff and

its direct interaction with Drp1, this is an area in need of detailed biochemical assessment.

1.11 Other Drp1 partner proteins

In addition to Mff, there are several other proteins of the MOM that play a role in

Drp1 recruitment and play a role in mitochondrial fission in chordates (Figure 1.3B). Of

these, fission protein 1 (Fis1) is conserved from yeast while the mitochondrial dynamics

proteins of 49 and 51 kDa (MiD49 and MiD51) are exclusive to chordates. As previously

discussed, the mitochondrial fission complex in lower eukaryotes, such as unicellular

fungi, is not well conserved in metazoans. The best such example is the case of Fis1 (the

human protein) and Fis1p (the yeast homologue), which are both comprised of an N-

terminal TPR-repeat domain and a C-terminal transmembrane helix (Fig 1.3C). The

subcellular localization of Fis1 is comparable to that of Mff: it localizes to both the MOM

and peroxisomal membrane. In yeast (Figure 1.3A), Fis1p is an essential component of the mitochondrial and peroxisomal fission complexes that interacts with Dnm1p via cytosolic adaptor proteins Mdv1 and Caf4 (Tieu and Nunnari 2000; Naylor et al. 2006;

Griffin, Graumann, and Chan 2005, 40; Q. Guo et al. 2012). Notably, this interaction is mediated by a domain of Dnm1p analogous to the VD of Drp1 (Huyen T. Bui et al. 2012,

1). In mammals (Figure 1.3B), the role of Fis1 is much more widely debated. The human homologue of Fis1 was identified and shown to dramatically regulate mitochondrial morphology in a Drp1-dependent manner (James et al. 2003, 1; Y. Yoon et al. 2003). In subsequent studies, the effects of Fis1 on mitochondrial morphology were shown to be much less pronounced (Loson et al. 2013; Gandre-Babbe and van der Bliek 2008) than that of other Drp1 partner proteins. Thus, the role that Fis1 plays in human mitochondrial dynamics needs further study.

In contrast to Fis1, the most ancient Drp1 partner protein, are the chordate-exclusive

MiD proteins. Originally identified as related candidate genes responsible for Smith-

Magenis syndrome, the proteins encoded by these related genes were found to localize

to mitochondria and have an effect on mitochondrial morphology (Zhao et al. 2011, 1;

Palmer et al. 2011). Interestingly, in contrast to other mitochondrial partner proteins,

overexpression of MiD49/51 results in a hyperfused mitochondrial network. Further

unlike Mff and Fis1, the MiD proteins do not localize to the peroxisome, and have no

effect on peroxisomal morphology. The mitochondrial expression pattern of the MiD

proteins further distinguishes them from Fis1. While Fis1 is uniformly distributed on the

mitochondria, the MiD proteins form punctate foci on the MOM dependent on the

intermembrane space-localized N-terminus of the protein, which appears to be crucial for

apoptotic cristae remodeling (Otera et al. 2016) and cytochrome c release. More recently, crystallographic studies have revealed an unforeseen regulatory role of MiD49 and

MiD51. The carboxy-terminus of both proteins consists of a well-folded globular

nucleotidyltransferase-like domain (Losón et al. 2015; Richter et al. 2014, 51; Loson et al.

2014, 51). Notably, unlike other Drp1 partner proteins, the MiD proteins are able to

dramatically alter Drp1 assembly properties (Loson et al. 2014; Koirala et al. 2013) and

GTPase activity (Osellame et al. 2016). Thus, the MiD proteins are able to rearrange Drp1

polymers to alter functional fission complex formation in response to local energy

demands and represent an additional regulatory element for metazoan mitochondrial

morphology.

1.12 Post-translational regulation of Drp1 and Mff

Since Drp1 is an essential gene, its expression level is not thought to directly correlate with its mitochondrial fission activity in cells. Instead, it appears that the fission activity of Drp1 correlates better with post-translational modifications that have the potential to modulate its stability, activity, and subcellular localization. Consequently, mitochondrial fission is tightly regulated by post-translational modification. Both Drp1 and Mff are known to be regulated by a host of post-translational modifications. Of these two proteins, the modifications of Drp1 have been much more widely studied. While S- nitrosylation (Bossy et al. 2010; T. Nakamura et al. 2010), ubiquitination (H. Wang et al.

2011; N. Nakamura et al. 2006), sumoylation (C. Guo et al. 2017; Figueroa-Romero et al.

2009), and O-GlcNAcylation (Gawlowski et al. 2012) have been identified, the best characterized Drp1 post-translational modifications are its reversible phosphorylations.

Of these, the most extensively studied are two particular serine residues within the VD:

S616 and S637. Serine 616 phosphorylation was the first of these Drp1 modifications identified in studies exploring the changes in mitochondrial morphology as the cell proceeds through mitosis (Taguchi et al. 2007). This initial study identified the cell-cycle dependent CDK1/Cyclin B complex as the kinase responsible for this modification.

Subsequent studies showed that in addition to cell cycle progression, diverse signals including ERK (Kashatus et al. 2015), CDK5 (Xie et al. 2015; Cho et al. 2014), and PKCδ (Qi et al. 2011) can drive Drp1-dependent mitochondrial fission by phosphorylation of S616.

Opposing the stimulatory role of Drp1 phosphorylation on serine 616, a serine at the junction of the VD and GED (serine 637) was identified as a repressive modification (Chang

and Blackstone 2007; Cribbs and Strack 2007). These initial studies found that the phosphorylation of serine 637 blunted the extent of mitochondrial fission in cells, and further that the GTPase activity of this modified Drp1 was slightly diminished. Subsequent studies showed that this phosphorylation precluded mitochondrial translocation and

fission activity of Drp1 regardless of the phosphorylation state of serine 616 (Cereghetti

et al. 2008). Interestingly, in several cancers where the total level of Drp1 is unchanged

but its phosphorylation fluctuates, the mitochondrial morphology appears to be a driving

factor in tumor growth (Xie et al. 2015, 1), migration (Zhao et al. 2012), and disease

severity (Tanwar et al. 2016). Thus, the post-translational regulation of Drp1 function

represents an attractive therapeutic approach for several types of human cancers.

The post-translational control of Mff is a much more recent discovery, and as such the

available knowledge is more limited than that pertaining to Drp1 regulation. The most

widely-understood modification of Mff is its phosphorylation by adenosine

monophosphate activated protein kinase (AMPK) which was initially identified in a global

phosphoproteomic mass spectrometry study seeking new substrates for AMPK

(Ducommun et al. 2015). Soon after this initial study, the functional consequence of Mff

phosphorylation by AMPK was revealed to be enhanced Drp1-mediated mitochondrial

fission (E. Toyama et al. 2016). This finding complements the energy-sensing function of

the MiD proteins, allowing for mitochondrial energy stress to be sensed by all three of the

key Drp1 receptor proteins. Interestingly, in a study that predated the documentation of

an Mff-targeting kinase, its interaction with phosphoprotein binding 14-3-3 proteins was

identified (Johnson et al. 2011). At this time, no functional roles of the interaction

between Mff and 14-3-3 proteins have been found, and the specific kinases that drive this

association are unknown. The only other known post-translational modification of Mff to

be described is its lysine ubiquitination. It was observed that Mff interacts with Parkin on

the MOM following MMP disruption, is ubiquitinated, and aids in formation of autophagic

isolation membranes surrounding dysfunctional mitochondria (Gao, Qin, and Jiang 2015).

Mutation of the key lysine that is modified by parkin totally disrupted autophagosome

formation, highlighting a key role for Mff in parkin-mediated mitophagy. Interestingly,

this indicates a Drp1-independent cellular function for Mff as Drp1 is dispensable for autophagy (Burman et al. 2017). Overall, while Mff ubiquitination has a clear mechanism, the effects of Mff phosphorylation are still vague and require additional mechanistic study.

1.13 Methods for assessing membrane protein function and assembly in vitro

The study of membrane proteins, such as Mff, outside of their native membranes is often complicated by their removal from a native lipid environment. This is a necessary step in the purification of these proteins, but many varied approaches have been developed over the past 20 years to return these proteins to a membrane environment that emulates their native localization. This reconstitution is essential because in the absence of a membrane environment, many proteins lose full function. To examine the function of Mff in vitro throughout this dissertation, membrane scaffold liposomes are used to reconstitute Mff in a lipid-proximal environment.

Membrane scaffolding liposomes are a unique and modular strategy to investigate

membrane protein functions and structural assemblies. Membrane scaffolding refers to the use of functionalized lipids as a membrane anchor in place of a protein’s native transmembrane domain. This includes lipids that have been functionalized with moieties that can interact with proteins or protein tags including biotin-, NTA-(Ni2+)-, and

crosslinker-tethered lipids. Early studies using this type of technique used lipid tethering

to induce crystallization of tethered proteins in 2D planar lattices (Kubalek, Le Grice, and

Brown 1994; Barklis et al. 1997; Vénien-Bryan et al. 1997) or helical lattices (Dang et al.

2005; Wilson-Kubalek et al. 1998) for structural study. Expanding off of these findings,

this technique (also called template-directed assembly) was expanded to reconstitute and

study membrane-anchored signaling complexes (A. Shrout, Montefusco, and Weis 2003;

Esposito, Shrout, and Weis 2008; A. L. Shrout, Esposito, and Weis 2008). By using these

lipid scaffolds, the prevalence of a protein as well as its topology relative to the lipid

interface are easily controlled, allowing for protein-mediated self-assembly. The major

drawback to this technique is fundamental in that the intact protein is not used, so any

functions of transmembrane segments are lost, meaning that the technique cannot be

used for proteins such as ion channels, but it is well suited to systems such as the

interaction with a cytosolic protein with a membrane-anchored partner. Recently, this

technique was used to demonstrate Drp1 interaction with MiD49 in vitro by assessing

Drp1 co-flotation with MiD49-tethered liposomes (Koirala et al. 2013). Thus, scaffold

lipids offer a rapid and modular technique to explore the interaction of Drp1 and Mff at a

lipid membrane.

1.14 Foundation and experimental framework

Cellular and genetic studies have clearly demonstrated the functional relationship

between Drp1 and Mff in cells, but critical features of their direct interaction remain

unknown due to the transient nature of this contact. In Chapter 2, we discuss the methods

used for studying the Drp1-Mff interaction proximal to a lipid bilayer throughout the subsequent chapters. In Chapter 3, we demonstrate the first evidence in vitro for functional and structural consequences of Mff interaction with Drp1, and identify the role of protein oligomerization in regulating this interaction. In Chapter 4, we explored the role of Mff in assembling a contractile Drp1 apparatus and identified how the dynamics of the VD regulate its interaction with Mff. These studies collectively offer unique insight into the crucial interaction between Drp1 and Mff and form a foundation for understanding key regulatory elements that govern mitochondrial fission in mammals.

FIGURE 1.1

FIGURE 1.1 Mitochondrial membrane architecture and dynamics. The mitochondrial

membranes adopt a characteristic complex architecture with a highly convoluted inner

membrane englobed by an outer membrane (A). In addition to the matrix (light orange)

and intermembrane space (orange), several other distinct subregions and compartments

are formed. The mitochondria constantly undergo fission and fusion resulting in an

actively dynamic organellar morphology. Mitochondrial fission (B – left) is mediated by

Drp1 (green) along with partner proteins of the mitochondrial outer membrane (pink).

Mitochondrial membrane fusion (B – right) is mediated by Mfn1/2 (blue) on the outer membrane and Opa1 (red) on the inner membrane.

FIGURE 1.2

FIGURE 1.2 Characteristic domains and structural features of dynamin family proteins.

Soluble human dynamins include MxA/B (MxA PDB ID: 5GTM), Drp1 (PDB ID: 4BEJ), and the endocytic dynamins (Dyn1 PDB ID: 3ZVR). (A) These distantly related proteins have characteristic functional and assembly domains including the GTPase domain (green),

Middle domain (dark blue), and GED (light blue). (B) In addition, these domains fold together into comparable 3D structural elements that are required for the functional and self-assembly properties of these proteins.

FIGURE 1.3

FIGURE 1.3 The mitochondrial fission apparatus differs between yeast and higher

eukaryotes. The components of the mitochondrial fission machinery are not well

conserved among all eukaryotes. In yeast (A) the mechanoenzymes (Dnm1p – green) is

recruited to the MOM via interaction with various partner proteins involving Mdv1p and

Fis1p (pink and purple respectively). In chordates (B), no cytosolic partners are conserved, and instead new adaptor proteins including Mff and MiD49/51 (orange and red

respectively) recruit the mechanoenzymes (Drp1 – green) to the MOM while the role of

Fis1 (purple) remains unclear. Schematic representations of the primary sequence for human partner proteins of Drp1 (C). The known sequence features of these proteins are highlighted including transmembrane segments (TM), tetratricopeptide repeats (TPR),

coiled-coil (CC), Mff repeats 1 and 2 (Rpt1/2), and nucleotide binding domain (NBD).

CHAPTER 2: USING SCAFFOLD LIPOSOMES TO RECONSTITUTE LIPID-PROXIMAL PROTEIN-

PROTEIN INTERACTIONS IN VITRO

This chapter was previously published:

Ryan W. Clinton, and Jason A. Mears

Using Scaffold Liposomes to Reconstitute Lipid-proximal Protein-protein Interactions In

Vitro

J Vis Exp. 2017, 119, e54971-e54971

2.1 ABSTRACT

Studies of integral membrane proteins in vitro are frequently complicated by the presence of a hydrophobic transmembrane domain. Further complicating these studies, reincorporation of detergent-solubilized membrane proteins into liposomes is a stochastic process where protein topology is impossible to enforce. This paper offers an alternative method to these challenging techniques that utilizes a liposome-based scaffold. Protein solubility is enhanced by deletion of the transmembrane domain, and these amino acids are replaced with a tethering moiety, such as a His-tag. This tether interacts with an anchoring group (Ni2+ coordinated by nitrilotriacetic acid (NTA(Ni2+)) for His-tagged proteins), which enforces a uniform protein topology at the surface of the liposome. An example is presented wherein the interaction between dynamin- related protein 1 (Drp1) with an integral membrane protein, mitochondrial fission factor

(Mff), was investigated using this scaffold liposome method. In this work, we have demonstrated the ability of Mff to efficiently recruit soluble Drp1 to the surface of liposomes, which stimulated its GTPase activity. Moreover, Drp1 was able to tubulate the Mff-decorated lipid template in the presence of specific lipids. This example demonstrates the effectiveness of scaffold liposomes using structural and functional assays and highlights the role of Mff in regulating Drp1 activity.

2.2 INTRODUCTION

Studying membrane-proximal protein-protein interactions is a challenging endeavor due to difficulty in recapitulating the native environment of the integral

37 membrane proteins involved (Seddon, Curnow, and Booth 2004). This is due to the necessity of detergent solubilization and the inconsistent orientation of proteins in proteoliposomes. In order to avoid these issues, we have employed a strategy whereby soluble domains of integral membrane proteins are expressed as His-tag fusion proteins, and these soluble fragments are anchored to scaffold liposomes via interactions with NTA(Ni2+) headgroups at the lipid surface. Using these scaffolds, lipid-proximal protein interactions can be investigated over a range of lipid and protein compositions.

We have effectively applied this method to investigate the critical protein-protein interactions that govern assembly of the mitochondrial fission complex and examine lipid interactions that modulate this process (R. W. Clinton et al. 2016). During mitochondrial fission, a conserved membrane remodeling protein, called dynamin- related protein 1 (Drp1) (D. C. Chan 2006), is recruited to the surface of the mitochondrial outer membrane (MOM) in response to cellular signals that regulate energy homeostasis, apoptotic signaling, and several other integral mitochondrial processes. This large, cytosolic GTPase is recruited to the surface of mitochondria through interactions with integral MOM proteins (H. T. Bui and Shaw 2013; Elgass et al. 2013; Loson et al. 2013; Otera et al. 2010; Gandre-Babbe and van der Bliek 2008).

The role of one such protein, mitochondrial fission factor (Mff), has been difficult to elucidate due to an apparent weak interaction with Drp1 in vitro. Nevertheless, genetic studies have clearly demonstrated that Mff is essential for successful mitochondrial fission (Gandre-Babbe and van der Bliek 2008; Otera et al. 2010). The method

38 described in this manuscript was able to overcome previous shortcomings by introducing simultaneous lipid interactions that promote Drp1-Mff interactions.

Overall, this novel assay revealed fundamental interactions guiding assembly of the mitochondrial fission complex and provided a new stage for ongoing structural and functional studies of this essential molecular machine.

To date, examination of interactions between Drp1 and Mff have been complicated by the inherent flexibility of Mff (Koirala et al. 2013), the heterogeneity of Drp1 polymers (Macdonald et al. 2014; R. W. Clinton et al. 2016), and the difficulty in purifying and reconstituting full-length Mff with an intact transmembrane domain

(Macdonald et al. 2016). We addressed these challenges by using NTA(Ni2+) scaffold liposomes to reconstitute His-tagged Mff lacking its transmembrane domain(MffΔTM-

His6). This strategy was advantageous because MffΔTM was extremely soluble when over-expressed in E. coli, and this isolated protein was easily reconstituted on scaffold liposomes. When tethered to these lipid templates, Mff assumed an identical, outward facing orientation on the surface of the membrane. In addition to these advantages, mitochondrial lipids, such as cardiolipin, were added to stabilize Mff folding and association with the membrane (Macdonald et al. 2016). Cardiolipin also interacts with the variable domain of Drp1 (Bustillo-Zabalbeitia et al. 2014; R. W. Clinton et al. 2016) which may stabilize this disordered region and facilitate assembly of the fission machinery.

This robust method is widely applicable for future studies that seek to evaluate membrane-proximal protein interactions. Through the use of additional

tethering/affinity interactions, the sophistication of these membrane reconstitution

studies can be enhanced to mimic additional complexity found at the surface of

membranes within cells. At the same time, lipid compositions can be modified to more

accurately mimic the native environments of these macromolecular complexes. In

summary, this method provides a means to examine the relative contributions of

proteins and lipids in shaping membrane morphologies to during critical cellular

processes.

2.3 PROTOCOL

2.3.1 SCAFFOLD LIPOSOME PREPARATION

Note: Ideally, initial experiments should use a relatively simple and featureless

scaffold (comprised of DOPC (1,2-dioleoyl-sn-glycero-3- phosphocholine or PC) and

DGS-NTA(Ni2+) (1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)

iminodiacetic acid) succinyl] (nickel salt)). Building off of these experiments, lipid

charge, flexibility, and curvature can be introduced as individual factors with the

potential to alter membrane-proximal interactions. These changes can be achieved

by adding defined amounts of specific lipid constituents, including

phosphatidylserine or cardiolipin (CL), phosphatidylethanolamine (DOPE or PE), or

galactosyl(β) ceramide.

1. Combine lipids dissolved in chloroform in a clean glass test tube. Evaporate the

solvent with dry nitrogen gas while rotating the tube to form a thin lipid film. Remove

residual solvent with a centrifugal evaporator for 1 h at 37°C.

Note: Various liposome formulations are used in the protocols described below:

scaffold liposomes (3.3 mol% DGS-NTA(Ni2+) / 96.7 mol% DOPC), scaffold liposomes

with cardiolipin (3.3 mol% DGS-NTA(Ni2+) / 10 mol% cardiolipin / 86.7 mol% DOPC),

flexible scaffold liposomes with cardiolipin (3.3 mol% DGS-NTA(Ni2+) / 10 mol%

cardiolipin / 35 mol% DOPE / 51.7 mol% DOPC), and enriched scaffold liposomes

(10 mol% DGS-NTA(Ni2+) / 15 mol% cardiolipin / 35 mol% DOPE / 40 mol% DOPC).

2. Add Buffer A (25 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid),

150 mM KCl, pH adjusted to 7.5 with KOH) preheated to 37°C such that the final

lipid concentration is 1-2 mM. Incubate 30 min at 37°C with occasional vortexing to

fully resuspend the lipid mixture (Figure 2.1a).

3. Transfer to a plastic test tube, place the tube in liquid nitrogen until completely

frozen (roughly 30 s), and place in a 37 °C water bath until fully thawed (roughly 1-2

min). Repeat for a total of 4 freeze-thaw cycles (Figure 2.1b).

4. Prepare a lipid extruder by soaking 4 filter supports and a polycarbonate filter in

buffer and assembling the extruder according to manufacturer instructions.

Extrude the lipid solution through the filter 21 times. Use gentle, constant pressure

to ensure a homogenous size distribution (Figure 2.1c).

NOTE: For all experiments described in this protocol, a 1.0 µm polycarbonate filter

was used for extrusion. Drp1 interaction with anionic lipids can be observed with a

variety of liposome diameters ranging from 50 nm to 400 nm (Bustillo-Zabalbeitia et

al. 2014) or larger (Francy et al. 2015). Hence, the filter size of 1 µm was chosen to

be ideal for both GTPase activity and for electron microscopy. If other liposome

diameters are desired, preparation of giant unilamellar vesicles (Walde et al. 2010;

Moscho et al. 1996) (GUVs) or small unilamellar vesicles (Klingler et al. 2015) (SUVs)

can be used. Dynamic light scattering can be used to assess liposome size

heterogeneity (Francy et al. 2015).

5. Store extruded liposomes at 4°C and discard after 3-5 days.

2.3.2 USE OF SCAFFOLD LIPOSOMES FOR PROTEIN BINDING ANALYSIS

Sample Preparation

1. Incubate His-tagged MffΔTM (5 µM final) with scaffold liposomes (40 mol% PC / 35

mol% PE / 15 mol% CL / 10 mol% DGS-NTA(Ni2+); 50 µM final) for at least 15 min at

room temperature in Buffer A + BME (25 mM HEPES, 150 mM KCl, 10 mM β-

mercaptoethanol (BME), pH adjusted to 7.5 with KOH). For an Mff-free control,

incubate liposomes with a his-tagged control protein (such as GFP) to bind and shield

exposed NTA(Ni2+).

Note: MffΔTM was expressed and purified as described in a previous study (R. W. Clinton et al. 2016). GFP was purified in a similar manner, but the ion- exchange step was omitted.

BME was required for these experiments because Drp1 is sensitive to oxidation, which can alter its activity and assembly properties.

2. Add Drp1 (2 µM final) and incubate for 1 h at room temperature.

Note: Drp1 was expressed and purified as described in a previous study (R. W. Clinton

et al. 2016). After incubation with Drp1, the effect of nucleotide binding on membrane

deformation can be investigated by incubating one additional hour with 2 mM MgCl2

and either 1 mM GTP, 1 mM GMP- PCP, or Buffer A + BME.

Negative Stain Transmission Electron Microscopy (EM) Analysis

1. Transfer 5 µL of sample to a sheet of laboratory film, and lay a carbon-coated Cu/Rh

grid on the sample. Incubate the grid 1 min on the sample, blot away excess liquid

on filter paper, and transfer to a drop of 2% uranyl acetate. Incubate 1 min, blot

excess stain on filter paper, and transfer to a grid box. Store under vacuum overnight

to ensure full desiccation.

2. Image samples using a transmission electron microscope at 18,500-30,000x

magnification to observe ultrastructural changes in protein and liposome

morphologies (Mears and Hinshaw 2008).

Note: Ultrastructural changes can be quantified using image analysis software, such as

ImageJ (Francy et al. 2015) (http://imagej.nih.gov/ij/). Protein decoration can be

measured when compared with naked lipid templates. Additionally, the diameters of

tubular segments can be measured from the outermost portion of the assemblies (Francy

et al. 2015). A more detailed analysis can be performed using cryo-electron microscopy

(Mears and Hinshaw 2008). This method can be used to image native protein-lipid

assemblies in solvent without the use of heavy metal stains that coat the sample. In this

43 way, detailed structural features not apparent in negative stain, including changes in the underlying lipid morphology, can be examined and quantified.

2.3.3 USE OF SCAFFOLD LIPOSOMES FOR ENZYMATIC ASSAY

Note: A colorimetric GTPase assay (Leonard et al. 2005) was used to measure phosphate liberation via GTP hydrolysis. Alternative GTPase assays are available

(Ingerman et al. 2005) and can be implemented as needed.

1. Incubate His-tagged MffΔTM (Mff), Fis1ΔTM(Fis1), or GFP (5 µM final for all) with

scaffold liposomes (150 µM final) for 15 min at room temperature in Buffer A +

BME (volume = 30 µL). Add Drp1 (500 nM final) and incubate an additional 15 min

at room temperature (volume = 80 µL).

Note: Fis1 was purified in a similar manner to Mff (R. W. Clinton et al. 2016), but the ion- exchange chromatography step was omitted. The purpose of His-tagged GFP is to shield the NTA(Ni2+) headgroups and prevent non-specific charge interactions with other proteins. If no effect is observed in the absence of GFP, then this control may not be required. Alternative blocking proteins (of comparable size to the protein of interest) can be used as well, but GFP allows for direct visualization of the interactions with scaffold liposomes.

2. Transfer tubes to a thermocycler set to 37 °C, and initiate reactions by addition

of GTP and MgCl2 (1 mM and 2 mM final, respectively; volume = 120 µL).

3. At desired time points (i.e. T=5, 10, 20, 40, 60 min), transfer 20 µL of reaction to

wells of a microtiter plate containing 5 µL of 0.5 M EDTA to chelate Mg2+ and stop

the reaction.

4. Prepare a set of phosphate standards by diluting KH2PO4 in Buffer A + BME to

calibrate results. A useful set of standards is 100, 80, 60, 40, 20, 10, 5, and 0 µM.

Add 20 µL of each to wells containing 5 µL of 0.5 M EDTA.

5. Add 150 µL of Malachite green reagent (1 mM malachite green carbinol, 10 mM

ammonium molybdate tetrahydrate, and 1 N HCl) to each well, and read OD640 5

min after addition.

Note: GTP is acid labile, and will hydrolyze in the presence of malachite green reagent.

Ensure that the time between adding malachite reagent and reading is constant to

ensure reproducible results.

6. Generate a standard curve by plotting OD650 of the standards as a function of

phosphate concentration. Use linear regression to determine the relationship

between OD650 and phosphate concentration in a sample.

7. Using the linear regression, convert the OD650 of the protein reaction samples to

µM phosphate. Determine the rate of phosphate generation for each reaction

mixture by plotting phosphate concentration as a function of time, and convert to

kcat by dividing the rate by the Drp1 concentration (0.5 µM).

Note: Only the initial linear rate should be used to determine the rate of phosphate

generation, and a minimum of 3 data points must be used. If the rate of reaction is

sufficiently rapid that the first three data points are not linear (i.e. the r2 of the linear fit

45 is less than 0.9) a significantly shorter time course with at least 3 time points should be performed.

2.4 REPRESENTATIVE RESULTS

While the interaction between Drp1 and Mff has been demonstrated to be important for mitochondrial fission, this interaction has been difficult to recapitulate in vitro. Our goal was to better emulate the cellular environment wherein Drp1 and Mff interact. To this end, liposomes containing limiting concentrations of NTA(Ni2+) headgroups were prepared by rehydrating a lipid film as described above. The lipid solution initially consists of unilamellar and multilamellar vesicles of heterogeneous diameters as evidenced by the opacity of the solution (Figure 2.1a). This opacity is reduced by freeze-thawing (Figure 2.1b), which reduces the prevalence of multilamellar vesicles. The liposome diameters are further homogenized by extrusion through a polycarbonate filter, which results in a clear solution (Figure 2.1c).

In previous studies, we found that Drp1 was able to assemble on Mff-decorated scaffold liposomes, and membrane tubulation was observed when flexible membranes were employed (R. W. Clinton et al. 2016). Building on these findings, we utilized a new template composed of PC, PE, Ni, and CL (called Enriched Scaffold Liposomes or ESL) to promote ordered assembly of a polymeric Drp1-Mff complex capable of inducing membrane deformation. Specifically, increased NTA(Ni2+) and cardiolipin lipids were utilized (10 mol% and 15 mol% respectively) for this application. Then, GFP or Mff was

46 tethered to ESL templates in the presence and absence of Drp1 (Figure 2.2), and the ability of Drp1 to remodel membranes was qualitatively assessed. In the absence of

Drp1, neither Mff nor GFP resulted in membrane deformation (Figures 2.3a, 2.3b), and similarly in the case of GFP-decorated ESL, only featureless liposomes were observed

(Figure 2.3c). However, when Drp1 was added to Mff-decorated ESL templates, remodeling of the liposomes was evident (Figure 2.3d).

While macromolecular complex formation clearly demonstrates an interaction between Drp1 and Mff, this qualitative analysis alone is incapable of determining the functional effects of such an interaction. Therefore, we utilized a malachite green phosphate generation assay (Leonard et al. 2005) to assess alterations in the catalytic activity of Drp1 in response to interaction with Mff. As described previously (R. W.

Clinton et al. 2016), we initially utilized a simple scaffold liposome (SL; 3.3 mol% DGS-

NTA(Ni2+), 96.7 mol% DOPC) to investigate the effect of Mff alone on Drp1 structure and function. Nonspecific interaction of Drp1 with NTA(Ni2+) has previously been described (Fröhlich et al. 2013), so SL was initially designed to contain low concentrations of NTA(Ni2+) to avoid nonspecific activity stimulation of Drp1. With the larger amounts of NTA(Ni2+) in ESL, the use of His-tagged GFP as a control was found to be critical to shield the Ni2+ and prevent non-specific Drp1 interactions. After decoration of SL liposomes by Mff or GFP (as illustrated in Figure 2.2), the extent of self-assembly can be assessed by measuring the GTPase activity of Drp1. In the absence of liposomes,

Drp1 has a relatively low basal GTPase activity, which is slightly enhanced by addition of SL. Decoration of these scaffold liposomes with Mff enhanced GTPase activity (Figure

2.4a, 1.8-fold). Conversely, when the exposed NTA(Ni2+) headgroups were blocked with

His-tagged GFP, this augmented GTPase activity was ablated. We also tested the role of

Fis1, an MOM protein that has been suggested to have a role in mitochondrial fission

(James et al. 2003; Y. Yoon et al. 2003), though this has been challenged in recent studies (Osellame et al. 2016; Otera et al. 2010). Tethering of Fis1 lacking its transmembrane domain to SL also failed to elicit a stimulation of Drp1's GTPase activity

(Figure 2.4a).

We then utilized a slightly more complex lipid scaffold containing a small amount of cardiolipin (SL/CL: SL with 10 mol% cardiolipin replacing DOPC) to determine the role of this mitochondrial lipid in the interaction of Drp1 and Mff. This moderate concentration of cardiolipin was specifically chosen to limit the stimulation of Drp1 by cardiolipin as described previously (Macdonald et al. 2014). Similar to SL, addition of SL/CL to Drp1 resulted in a slight stimulation of GTPase activity that was reversed by tethering His-tagged Fis1 or GFP to the liposomes. A synergy between Mff and cardiolipin was observed as the GTPase activity of Drp1 was stimulated 2.6-fold when it was incubated with Mff-decorated SL/CL (Figure 2.4b). Membrane fluidity and the ability of Drp1 to remodel lipid bilayers have been proposed to enhance its GTPase activity. Therefore, we sought to examine the effect of membrane fluidity/flexibility using a flexible scaffold liposome. This was achieved by replacing 35 mol% of DOPC in SL/ CL with DOPE (SL/PE/CL), which has previously been shown to allow for Drp1- mediated membrane remodeling (Macdonald et al. 2014). Addition of undecorated

SL/ PE/CL scaffold liposomes to Drp1 slightly enhances Drp1 GTPase activity, and

48 decoration of these liposomes with GFP eliminates this effect. When SL/PE/CL templates were decorated with Mff, Drp1 activity was enhanced (Figure 2.4c, 2.4- fold). As we have previously shown, the ability of Drp1 to remodel liposomes into lipid tubules was enhanced by the addition of PE to the scaffold liposomes.

Interestingly, this improved tubulation leading to the formation of a helical Drp1 polymer did not result in any greater stimulation when compared to liposomes that

Drp1 was unable to remodel (R. W. Clinton et al. 2016).

Using these adaptable lipid templates, Mff and Drp1 were found to interact in a more native environment in vitro. This technique has enabled us to control the relative abundance of Drp1, Mff (through NTA(Ni2+) concentration), and specific lipids

(cardiolipin and PE specifically) that appeared to regulate the assembly of this macromolecular complex. As we have demonstrated, this method can be utilized to visualize the membrane remodeling of Mff-recruited Drp1 by electron microscopy, and to determine the effects of Drp1 assembly on its catalytic activity using GTPase activity assay.

2.5 DISCUSSION

This protocol offers a method for investigating protein-protein interactions involving integral membrane proteins. By utilizing a modular liposome scaffold, investigators are capable of assessing the activity of one or more proteins in a lipid-proximal environment. Previous studies have demonstrated a similar method for receptor

49 enzymes of the plasma membrane(Esposito, Shrout, and Weis 2008; A. L. Shrout,

Esposito, and Weis 2008; Celia et al. 1999). We have expanded this method to incorporate lipid cofactors and explore interactions between proteins that make up the mechanoenzymatic core of the mitochondrial fission machinery.

For the model system presented above, we found that Mff-decorated SL enhanced Drp1 self-assembly. Moreover, we now show that Mff-decorated ESL templates were efficiently remodeled by wild-type Drp1 to form extended tubular structures. We also assessed the roles of various mitochondrial lipids, including the negatively charged cardiolipin and the conical lipid PE. Cardiolipin synergizes with

Mff to further enhance Drp1 self-assembly, while the membrane flexibility and fluidity imparted by PE enhances membrane tubulation but does not further augment Mff-induced stimulation.

To assess ultrastructural changes in membrane morphologies, EM analyses were required. Drp1 GTPase activity was elevated through clustering and assembly of filamentous polymers that did not reshape the liposomes to any great extent (R. W.

Clinton et al. 2016). However, membrane deformation was observed when the more mitochondria-like SL/PE/CL template was used. Interestingly, the enzyme activity was not enhanced. Therefore, the EM studies were essential in identifying key differences that would otherwise be missed using the functional assay.

While this technique is powerful for exploring the function and interaction of soluble proteins and soluble protein domains, these lipid scaffolds cannot account for the role of transmembrane domains. This is an important consideration because the

50 transmembrane domain can effect dynamic protein processes such as self-assembly

(Li, Wimley, and Hristova 2012) and lateral diffusion (F. Zhang et al. 1991; Ramadurai et al. 2009; Gambin et al. 2010) in lipid bilayers. If these factors are critical for evaluating protein interactions at the membrane surface, then traditional lipid reconstitution experiments with detergent would be favored. Alternative tethers may also be explored to control the recruitment and mobility of the membrane anchored proteins.

In addition to using His-tagged proteins with NTA(Ni2+) lipid anchors, other tethers such as biotin-conjugated (Wilson-Kubalek et al. 1998) or reactive group-conjugated lipids can be utilized. These covalent modifications would more stably trap proteins at the lipid surface, but mobility and exchange of these factors would likely be diminished.

As such, the tether should be carefully considered in the context of the protein complexes being studied. When considering the use of this method, the mode of tethering proteins to lipid templates have the potential to influence certain assays. For instance, the His-tag tethering to NTA(Ni2+) method may be more appropriate for in situ assays rather than separation experiments, especially in the case of transient protein- protein interactions. This is clearly demonstrated in Figure 2.3 by the discrepancy between the in situ negative stain electron microscopy and the sedimentation assay.

In the future, a combination of two or more of these anchor lipids with distinct target headgroups could be implemented to allow for recruitment of multiple proteins to a scaffold template without competition for a single lipid tether. Moreover, the relative abundance of each component can be managed by altering the lipid

composition. Additional lipid cofactors, such as phosphoinositides, cardiolipin, and

phosphatidylserine, can easily be introduced in these templates to assess the isolated

impact of a variety of factors.

Overall, these lipid scaffolds represent a novel platform for investigating complex

protein interactions near lipid membranes. These templates are easily generated and

are simple to tailor to a range of diverse applications, including enzymatic assays,

electron microscopy, or fluorescence imaging. In addition, the lipid composition can be

formulated to resemble an organelle or membrane microdomain of interest to better

recapitulate protein function at these specific regions. Using these techniques,

biochemists can probe the complex interactions of membrane associated proteins with

their partners and their environment.

2.6 ACKNOWLEDGEMENTS

The authors would like to acknowledge the funding received from the American Heart

Association (SDG12SDG9130039).

FIGURE 2.1

FIGURE 2.1 Lipid preparation schematic. (a) Upon resuspension, liposomes of diverse sizes form and consist of unilamellar and multilamellar vesicles, which results in an opaque solution (inset). (b) Freeze-thawing the solution results in a more unilamellar population of liposomes, which are still heterogeneous in diameter. Freeze-thawing clarifies the solution (inset). (c) Extrusion of the lipid solution homogenizes the liposome diameter (1.0 μm in this example), and results in a clear solution (inset).

FIGURE 2.2

FIGURE 2.2 Methods to assess protein assembly. A schematic depicting partner protein assembly on scaffold liposomes is presented. His-tagged partner proteins or GFP are

54 incubated with scaffold liposomes, and then Drp1 is incubated with decorated or undecorated liposomes. These Drp1-preassembled liposomes can then be analyzed by structural methods (electron microscopy) and functional assays (GTPase assay).

FIGURE 2.3

FIGURE 2.3 Structural assessment of Drp1 recruitment. Negative stain transmission micrographs of GFP or Mff decorated liposomes alone (a and b respectively) or incubated with Drp1 (c and d respectively). Scale bar = 100 nm.

FIGURE 2.4

FIGURE 2.4 Scaffold Liposome Enzymatic assay. (a-c) The generation of phosphate over time was measured (inset), and the kcat was determined. This method was applied to SL- tethered proteins (a), SL/CL-tethered proteins (b), or SL/PE/CL-tethered proteins (c). #: p

< 0.05, *: p < 0.0001, **: p < 0.000001 as determined by unpaired student’s t-test. All error bars represent standard deviation from 3 independent samples.

CHAPTER 3: DYNAMIN-RELATED PROTEIN 1 OLIGOMERIZATION IN SOLUTION IMPAIRS

FUNCTIONAL INTERACTIONS WITH MEMBRANE-ANCHORED MITOCHONDRIAL FISSION

FACTOR

This chapter was previously published:

Ryan W. Clinton, Christopher A. Francy, Rajesh Ramachandran, Xin Qi,

and Jason A. Mears

Dynamin-Related Protein 1 Oligomerization in Solution Impairs Functional Interactions

with Membrane-Anchored Mitochondrial Fission Factor

J Biol Chem. 2016, 291, 478-492

3.1 ABSTRACT

Mitochondrial fission is a crucial cellular process mediated by the mechanoenzymatic

GTPase, dynamin-related protein 1 (Drp1). During mitochondrial division, Drp1 is

recruited from the cytosol to the mitochondrial outer membrane (MOM) by one, or

several, integral membrane proteins. One such Drp1 partner protein, mitochondrial

fission factor (Mff), is essential for mitochondrial division, but its mechanism of action

remains unexplored. Previous studies have been limited by weak interactions between

Drp1 and Mff in vitro. Through refined in vitro reconstitution approaches and multiple independent assays, we show that removal of the regulatory variable domain (VD) in Drp1 enhances formation of a functional Drp1-Mff copolymer. This protein assembly exhibits greatly stimulated cooperative GTPase activity in solution. Moreover, when Mff was anchored to a lipid template, to mimic a more physiologic environment, significant stimulation of GTPase activity was observed with both WT and ∆VD Drp1. Contrary to recent findings, we show that premature Drp1 self-assembly in solution impairs functional interactions with membrane-anchored Mff. Instead, dimeric Drp1 species are selectively recruited by Mff to initiate assembly of a functional fission complex. Correspondingly, we

also found that the coiled-coil (CC) motif in Mff is not essential for Drp1 interactions, but

rather serves to augment cooperative self-assembly of Drp1 proximal to the membrane.

Taken together, our findings provide a mechanism wherein the multimeric states of both

Mff and Drp1 regulate their collaborative interaction.

3.2 INTRODUCTION

Mitochondria undergo continuous cycles of fission and fusion to maintain a functional

organelle network within eukaryotic cells. This mitochondrial network is crucial for ATP

generation, apoptotic signaling, and calcium homeostasis. When the proper balance of

mitochondrial dynamics is disrupted, mitochondrial dysfunction is observed (Berman,

Pineda, and Hardwick 2008; D. C. Chan 2006). This insult is associated with increased cell

death in several human diseases, including neurodegenerative disorders (H. Chen and

Chan 2009; Knott et al. 2008), ischemia-reperfusion injury (Ashrafian et al. 2010; Ong and

Hausenloy 2010), and glaucoma (Ju et al. 2007). Therefore, mitochondrial division has developed into a compelling therapeutic target for intervention with small molecule and peptide inhibitors that limit cell death in several of these pathologies (Cassidy-Stone et al.

2008; X. Guo et al. 2013; Lackner and Nunnari 2010; Ong et al. 2010; S. W. Park et al.

2011; Su and Qi 2013).

The master regulator of mitochondrial fission, dynamin-related protein 1 (Drp1), has been targeted in these diseases. Similar to other dynamin family members, Drp1 is a large

GTPase that mediates membrane remodeling. The primary sequence of Drp1 is comprised of four conserved regions (Figure 3.1A): the GTPase domain, middle domain, variable domain (VD), and GTPase effector domain (GED). Hydrolysis of GTP triggers conformational changes in Drp1 oligomers that generate mechanical force needed to promote mitochondrial membrane scission (Francy et al. 2015; Mears et al. 2011), and factors that inhibit Drp1 GTPase activity prevent mitochondrial division (Cassidy-Stone et al. 2008; Qi et al. 2013; Smirnova et al. 1998). The middle and GED domains promote Drp1

61 self-assembly, which is also critical for its role in facilitating mitochondrial fission (Chang et al. 2010; P. P. Zhu et al. 2004). In vitro, addition of negatively charged lipids increases

Drp1 self-assembly to form larger helical assemblies that represent the contractile apparatus of mitochondrial fission (Y. Yoon, Pitts, and McNiven 2001), and these functional polymers exhibit stimulated GTPase activity (Francy et al. 2015; Bustillo-

Zabalbeitia et al. 2014; Fröhlich et al. 2013; Macdonald et al. 2014). The VD has recently been shown to act as a negative regulator of Drp1 self-assembly (Francy et al. 2015) with an inherent ability to interact with cardiolipin (CL) present in mitochondrial membranes

(Bustillo-Zabalbeitia et al. 2014; Macdonald et al. 2014; Stepanyants et al. 2015; Ugarte-

Uribe et al. 2014). Studies in yeast have shown that the VD is required for interactions with a mitochondrial adaptor protein (H. T. Bui and Shaw 2013), but the partner protein identified in that study is not conserved in higher eukaryotes, which suggests that the role of the VD may have evolved in higher organisms to accommodate different regulatory interactions in the cytosol and at the surface of mitochondria.

Drp1 interactions with multiple mitochondrial outer membrane (MOM)-anchored transmembrane proteins have been identified to promote its recruitment to the mitochondria (H. T. Bui and Shaw 2013; Elgass et al. 2013; Loson et al. 2013, 49). One such protein is mitochondrial fission factor (Mff), and genetic studies have unambiguously shown that Mff is critical for Drp1 recruitment to the MOM. In fact, Mff deletion suppresses Drp1 localization to mitochondria (Otera et al. 2010), which results in an excessively interconnected mitochondrial network (Gandre-Babbe and van der Bliek

2008). Concomitantly, over-expression of Mff results in excessive mitochondrial fission

(Otera et al. 2010). Even though Drp1-Mff interactions are crucial for mitochondrial dynamics, the interaction between Drp1 and Mff appears to be transient. Previous studies report that Drp1 GTPase activity is either unaffected (Koirala et al. 2013) or mildly enhanced in vitro in the presence of Mff (Otera and Mihara 2011). Additionally, crosslinking agents are required to capture a stable complex using pull-down experiments

(Otera et al. 2010; Gandre-Babbe and van der Bliek 2008). Given this relatively weak affinity, the molecular basis for Drp1-Mff interactions remains uncharacterized.

Using a combination of biochemical, cellular and electron microscopy (EM) methods, we have examined the structural and functional ramifications of Drp1 and Mff interactions in vitro. The role of the VD in Mff interactions was investigated by examining established assembly mutants and distinct Drp1 isoforms. We find that the Drp1 VD negatively regulates the assembly of a functional fission complex dependent on Drp1 interaction with Mff. Using mutations that alter the oligomeric state of Drp1 in solution, we show that Mff selectively assembles Drp1 dimers into large complexes with greatly stimulated GTPase activity. Our results also show that the conserved coiled-coil (CC) motif in Mff improves the efficiency of Drp1 recruitment and provides a scaffold to coordinate

Drp1 assembly. Therefore, effective, functional interactions within the mitochondrial fission complex are shaped by the oligomeric tendencies of both Drp1 and Mff.

3.3 RESULTS

The variable domain of Drp1 limits productive interaction with Mff

In this study, Drp1 constructs were expressed and isolated as described previously

(Francy et al. 2015), and the affinity tag was removed to examine the properties of the native Drp1 sequence. Initially, the role of the VD was investigated given its proposed role

in interactions with partner proteins (H. T. Bui and Shaw 2013). To study this region, the

previously characterized VD deletion mutant (∆VD) was expressed and isolated (Francy et

al. 2015; Fröhlich et al. 2013). Recent studies have implied that different Drp1 isoforms

may have distinct cellular interactions (Strack, Wilson, and Cribbs 2013), but the

functional role of these sequence changes is not fully understood. Therefore, Drp1 splice

variants with maximal (Isoform 1, or Drp1-1) and minimal (Isoform 3, or Drp1-3) sequence

inclusion in the VD were also used to examine Drp1-Mff interactions (Figs. 2.1 and 2.2).

The basal GTPase activity of each construct was assessed using a colorimetric assay, as

previously described (Leonard et al. 2005), and the kcat of both Drp1-1 and Drp1-3 was

found to be 1.8 min-1 (Figure 3.1B), which is consistent with previous studies using tagless

or his-tagged Drp1 constructs (Bustillo-Zabalbeitia et al. 2014; Macdonald et al. 2014).

The GTPase activity of ∆VD was slightly diminished when compared to WT constructs

(0.98 min-1), which was also observed previously (Francy et al. 2015; Fröhlich et al. 2013).

To characterize the interaction between Mff and Drp1, the GTPase activities of Drp1-

1, Drp1-3, and ΔVD were measured in the absence and presence of MffΔTM-His6 (referred

to as Mff from here on). Consistent with previous studies (Koirala et al. 2013), no

64 difference in the GTPase activity of Drp1-1 and Drp1-3 was observed when Mff was present in solution. However, the GTPase activity of ∆VD was stimulated more than 10- fold when Mff was present (Figure 3.1B). This stimulation of ∆VD by Mff was dependent on concentration (Figure 3.1C), and Mff at 5 μM was shown to elicit the maximal response. Therefore, this concentration of Mff was used in all subsequent GTPase assays.

To further examine interactions between Drp1 and Mff, negative stain electron microscopy (EM) analysis was used to assess the formation of distinct macromolecular complexes. Samples were made that contained Drp1 alone and, separately, Drp1 was incubated with a 5-fold molar excess of the cytosolic portion of Mff lacking its transmembrane (TM) segment (MffΔTM; residues 1-218). When each of the Drp1 constructs were examined alone, small protein complexes were readily observed (Figure

3.1, E-G), and no large polymers were apparent. In the case of Drp1-1 and Drp1-3, these complexes likely represent smaller multimeric Drp1 species that predominate at this concentration (Figure 3.1D) (Macdonald et al. 2014).

When Mff was added to the Drp1-1 and Drp1-3 solutions, the protein complexes appeared to be small and well dispersed (Figure 3.1, H-I), and larger complexes were still absent. Conversely, when Mff was added to ΔVD in solution (Figure 3.1J), large filamentous polymers were observed with an average diameter of 24.2 ± 2.6 nm (n = 384).

This increased assembly of Drp1 into filaments was consistent with the observed stimulation in GTPase activity (Figure 3.1B). Therefore, we demonstrate that removal of the VD favors Drp1-Mff interactions, which parallels recent findings (R. Liu and Chan

2015). In addition, we now show that this interaction nucleates the formation of large, functional protein complexes.

As the VD has been shown to regulate Drp1 self-assembly (Francy et al. 2015), we

examined the oligomeric propensities of WT and ∆VD Drp1 using size exclusion

chromatography coupled to multi-angle light scattering (SEC-MALS) analyses (Figure

3.1D). Consistent with previous studies (Macdonald et al. 2014), Drp1-3 was shown to

equilibrate between dimers and higher-order multimers (Figure 3.1D, black), and a similar

trend was observed for Drp1-1 (see MacDonald et al, 2015). Conversely, ΔVD was found

to be exclusively dimeric (Figure 3.1D, red). The role of the VD in regulating Drp1 self-

assembly was previously demonstrated as assembly-primed, CBP-tagged ∆VD efficiently

formed oligomers in solution and bound to lipid (Francy et al. 2015). However, at the

concentrations used in these studies with untagged protein, assembly competent ∆VD

dimers only formed large polymers in the presence of Mff. Collectively; these results show

that Mff preferentially supports stable assembly of Drp1∆VD dimers in solution.

Tethering of Mff to liposome scaffolds stimulates Drp1-3 and ∆VD GTPase activity

While robust oligomerization and stimulation of ΔVD GTPase activity was observed in

the presence of Mff, no corresponding changes were observed with Drp1-1 and Drp1-3 in

solution. Since Mff is an integral membrane protein, a model system was developed to

mimic Drp1-Mff interactions proximal to a lipid bilayer.

A His6-tag engineered at the C-terminus of Mff was used as a handle to tether the protein on scaffolding liposomes (SL) containing 3.3 mol% of DGS-NTA (Ni2+) lipid (Figure

3.2A). In this way, Mff is oriented at the lipid bilayer with its C-terminus anchored to the

membrane surface akin to the native protein. A similar approach has previously been

employed to study interactions of other proteins at the plasma membrane (Esposito,

Shrout, and Weis 2008; A. L. Shrout, Esposito, and Weis 2008). Using this strategy, Mff

abundance at the membrane surface and the stoichiometry of Mff-Drp1 interactions

could be tightly controlled. Moreover, the lipid composition used was intentionally inert

(96.7% DOPC) to exclusively measure the effect of tethered Mff on Drp1 function, and not

Drp1 stimulation by lipid alone. A separate lipid mixture that incorporated 10 mol% CL,

termed scaffold lipid with cardiolipin (SL/CL), was used to explore the potential role of

this dimeric, negatively charged phospholipid in promoting interactions between Drp1

and Mff. Importantly, this lower CL concentration also limits lipid stimulation of Drp1

GTPase activity and ensures that any changes in enzyme activity were principally due to

interactions with Mff on the lipid template. Consistent with earlier reports (Macdonald et

al. 2014), we found that GTPase activity of Drp1 was not significantly altered in the

presence of either lipid mixture (<20% difference in activity when SL or SL/CL was added).

When these topology-enforced liposomes were used, stimulation of WT Drp1 GTPase activity was observed in the presence of Mff. However, under these conditions, this change was dependent on the Drp1 isoform used. For Drp1-1, GTPase activity was

unaffected by the addition of Mff tethered to either SL or SL/CL (Figure 3.2, B-C). On the

other hand, Drp1-3 exhibited a 1.8-fold stimulation (3.2 min-1) in the presence of Mff-

67 decorated SL (Figure 3.2B). The presence of CL in the lipid template significantly enhanced the stimulation in activity (2.6-fold increase) when Drp1-3 was added to SL/CL with Mff

(5.1 min-1, p < 0.0005; Figure 3.2C).

When compared to WT Drp1, ΔVD GTPase activity was more robustly stimulated in the presence of Mff decorated liposomes, which is consistent with experiments performed in solution (Figure 3.1B). Mff tethering slightly enhanced stimulation of ∆VD

-1 activity (kcat = 15.2 min , a ~15-fold increase compared to a 10-fold increase in solution;

Figure 3.2B), which reflects the increased local concentration of Mff on the membrane.

When ΔVD was added to Mff coupled to SL/CL, a comparable stimulation (13.9 min-1) was observed which was not significantly distinct from the stimulation on SL (p = 0.46, Figure

3.2C). Therefore, the enhanced stimulation observed with Drp1-3 in the presence of limiting amounts of CL was attributed to VD interactions with the lipid template that were otherwise missing in the deletion mutant.

When examined by EM, no discernible remodeling was observed when Drp1-3 or ΔVD were added to lipid templates lacking Mff, and no macromolecular complexes were observed (Figure 3.2, D & F, respectively). This result is consistent with the lack of stimulated GTPase activity. Interestingly, when Drp1-3 was incubated with Mff-decorated liposomes there was no evident membrane remodeling (Figure 3.2E) although GTPase activity stimulation was observed (Figure 3.2, B-C). When these templates were preincubated with Mff, addition of ∆VD led to the formation of filamentous protein structures on the surface of the liposomes (Figure 3.2G), and these structures were similar to those found in solution (Figure 3.1J). The increase in stimulated GTPase activity, when

compared to Drp1-3, is attributed to the greater abundance and enhanced assembly of

Drp1 polymers mediated by Mff interactions at the surface of the liposome.

Alterations within the VD modulate Drp1-mediated mitochondrial fission in MEFs

Since there was a clear distinction in the activities of Drp1 splice variants and the ∆VD

mutant, each of these proteins were expressed in MEFs lacking Drp1 (Drp1-/- MEFs)

(Wakabayashi et al. 2009) and changes in mitochondrial morphology were evaluated.

Over-expression of Myc-tagged Drp1-1, Drp1-3, and ΔVD in Drp1-/-MEFs (Figure 3.3, A &

C) resulted in significant mitochondrial fragmentation (Figure 3.3B). Overexpression of all

Myc-Drp1 constructs were capable of rescuing significant mitochondrial fragmentation within these cells lacking endogenous Drp1, although Drp1-3 overexpression resulted in

a more potent effect than did Drp1-1 (Figure 3.3B). Interestingly, mitochondrial fission

was observed when ΔVD was overexpressed, but its effect was significantly diminished

compared to WT proteins (Figure 3.3B). Thus, the inclusion of the VD results in a more

efficient fission machinery in cells, and alternative splicing can modulate this activity.

To examine the recruitment of ∆VD to the MOM in cells, the same construct was

expressed in WT and Mff-knockout (MffWT and Mff-/- respectively) MEFs (Figure 3.3D).

Little, if any, mitochondrial fragmentation was observed when comparing MffWT and

Drp1-/- MEFs using a control vector (Figure 3.3, B & E). Over-expression of ∆VD led to a

significant increase in fragmentation in both cell lines, and this response was greatly attenuated in Mff-/- MEFs (Figure 3.3E). This reduction in fission was attributed to

decreased recruitment of ∆VD in cells lacking Mff, as ΔVD was efficiently recruited to the

mitochondria in MffWT MEFs. This recruitment was reduced by ~50% in Mff-/- MEFs (Figure

3.3F). Based on these findings, Mff targets ΔVD to mitochondria, but fission appears to be

impeded post-recruitment.

The assembly-incompetent Drp1 G363D mutant is insensitive to Mff interactions

Drp1-Mff interactions with the greatest stimulation in GTPase activity appeared to be dependent on the ability of Drp1 to self-assemble into higher order oligomers. In fact, the

ΔVD mutant has previously been shown to potentiate Drp1 assembly into larger polymers

(Francy et al. 2015). This could explain the enhanced Mff-induced oligomerization of ΔVD

compared to Drp1-3, as well as the increased stimulation in GTPase activity. To test this

idea, the assembly-defective Drp1 G363D mutant was used. This middle domain mutation prevents Drp1 self-assembly into species larger that a dimer (Chang et al. 2010; Tanaka,

Kobayashi, and Fujiki 2006).

To confirm the oligomeric potency of the G363D mutant (Drp1-3G363D) compared to

WT Drp1, SEC-MALS was used. As shown previously (Macdonald et al. 2014), Drp1-3 in solution exists as a mixture of dimers and higher-order multimers (Figure 3.4A, black trace). When the G363D mutation was introduced to the Drp1-3 construct (Drp1-3G363D),

the isolated protein migrated predominantly as a dimer in solution (Figure 3.4A, blue

trace). In solution, the Drp1-3G363D mutant did not exhibit any defect in GTPase activity

when compared to WT (Figure 3.4C) and GTPase activity was unaffected when

undecorated SL or SL/CL were added (Figure 3.5, A & F). Unlike Drp1-3, the assembly- incompetent Drp1-3G363D did not exhibit stimulated GTPase activity when it was added to

Mff-decorated SL or SL/CL templates (Figs. 3.5, A & F, respectively). This result shows that

the Mff-induced stimulation of GTPase activity reflects enhanced Drp1 self-assembly

proximal to the membrane template.

To complement these studies, a double mutant was generated that combined the

G363D and ΔVD mutations (ΔVDG363D). This mutant was designed to restrict the self-

assembly properties of ΔVD that result in high-order oligomers. SEC-MALS analysis

revealed that ΔVDG363D is also predominantly dimeric (Figure 3.4B, blue trace). Therefore,

∆VD and ΔVDG363D are both dimers and only differ in their ability to form higher order

oligomers. This mutant allowed us to directly evaluate the role of Drp1 self-assembly on

Mff induced polymerization and GTPase stimulation.

The GTPase activity of ΔVDG363D was similar to ∆VD, and the double mutant did not

exhibit an increase in activity when assayed in the presence of undecorated SL or SL/CL

(Figure 3.5, B & G). Where addition of Mff to ∆VD in solution yields a ~10-fold stimulation

in activity, no stimulation was observed with ΔVDG363D (Figure 3.4D). Moreover,

stimulation in the presence of Mff coupled to SL or SL/CL was completely abolished using

the ΔVDG363D mutant (Figs. 3.5B and 3.5G, respectively). EM analyses confirmed these findings as the large filamentous structures observed when ΔVD was added to Mff (Figure

3.4E) were not observed using ΔVDG363D (Figure 3.4F). Similarly, the liposome-targeted

filamentous structures observed when ΔVD was added to Mff-decorated SL (Figure 3.5C)

were not detected when ΔVDG363D was incubated with the same template (Figure 3.5D).

Based on these results, Mff stimulates Drp1 activity by supporting cooperative self- assembly into large, filamentous structures.

While the G363D mutation prevents Drp1 self-assembly, it was unclear whether this dimeric mutant could still interact with Mff. Co-sedimentation analysis showed that ∆VD formed stable complexes with Mff, but consistent with our EM observations ΔVDG363D was unable to form large polymers that would sediment (Fig 3.4H). To capture short-lived intermediates of Drp1-Mff in solution a non-specific amine-to-amine chemical crosslinker, glutaraldehyde was utilized. Electrophoretic mobility shifts were visualized by SDS-

PAGE and Western blot analyses to identify covalently linked Drp1-Mff complexes. Both

ΔVD and ΔVDG363D were found to interact with Mff∆TM in solution as unique complexes were observed. We also noticed that MffΔTM was shifted into larger molecular weight

(MW) complexes, which demonstrates an association with both ΔVD and ΔVDG363D (Figure

3.4I). Collectively, these experiments demonstrate that Drp1 dimers clearly interact with

Mff, but Drp1 self-assembly is required for complex stabilization. Consequently, Mff may act as a scaffold for Drp1 self-assembly, which stimulates GTPase activity.

Higher-order self-assembly of Drp1 R376E prohibits functional interactions with Mff

Given that Drp1-3 and ΔVD GTPase activities were stimulated by Mff interactions in vitro, we sought to characterize the putative Mff-binding defective mutant, Drp1 R376E

(Strack and Cribbs 2012). This charge reversal was designed to disrupt an interaction interface between Drp1 and Mff, and Mff immunoprecipitation of Drp1 from HEK293 cells

in the presence of a crosslinker was inhibited (Strack and Cribbs 2012). The recently

solved crystal structure of Drp1 reveals that R376 is situated in close proximity to an

assembly interface, interface 4, unique to Drp1 (Fröhlich et al. 2013). Thus, two

alternative explanations for the observed lack of interaction between Mff and the R376E

mutant are possible. Namely, destabilization of this novel self-assembly interface could

alter the self-assembly properties of Drp1, which in turn affects Mff interactions. On the other hand, this mutation could result in a direct perturbation of the Mff interaction interface as originally interpreted. To distinguish between these possibilities, the R376E

mutation was introduced within the Drp1-3 and ∆VD constructs (termed Drp1-3R376E and

ΔVDR376E respectively) to examine whether this single residue modification could alter

Drp1 interactions with Mff.

Following from the observation that the Mff-induced stimulation of Drp1 activity was

intimately linked to the oligomeric state of Drp1, SEC-MALS was used to assess the R376E

mutation in the contexts of full-length Drp1-3 and ∆VD. When compared to Drp1-3

(Figure 3.4A, black trace) at equivalent concentration, the Drp1-3R376E mutant exhibited a

propensity to form higher-order multimers (Figure 3.4A, red trace). In agreement,

previous studies assessing the assembly competence of Drp1R376E found that this mutant

was enriched in high MW complexes (~ 700 kDa) and depleted of low MW complexes

(~160 kDa) in cell lysates (Strack and Cribbs 2012). Therefore, this mutant provides an

opportunity to assess functional Mff interactions with larger Drp1 multimers.

When compared to WT Drp1 (Drp1-3), the R376E mutant exhibited an equivalent

basal GTPase activity and no stimulation was observed when Mff was present in solution

(Figure 3.4C). Unlike Drp1-3, addition of Mff-tethered liposomes had no stimulatory effect on the activity of the Drp1-3R376E mutant (Figs. 3.5A and 3.5F, respectively). EM studies confirmed that even though Drp1-3R376E is an assembly-competent mutant (i.e. it readily polymerizes in the presence of GTP analogs), it was unable to form large oligomers on

Mff-decorated liposomes (not shown). Thus, the prevalence of higher-order Drp1 multimers in solution impairs functional interaction with membrane-anchored Mff.

Having established that Drp1ΔVD exists predominantly as a dimer in solution that polymerizes in the presence of Mff, the impact of the R376E mutation on cooperative

ΔVD-Mff assembly was assessed. Remarkably, the R376E mutant remained exclusively dimeric in the absence of the VD (Figure 3.4B, red trace). In striking contrast to the prematurely multimerized Drp1-3R376E, addition of Mff to ΔVDR376E in solution enhanced

GTPase activity ~5-fold (2.8 min-1; Figure 3.4D). EM analysis of ΔVDR376E in the presence of

Mff in solution shows an abundance of filamentous oligomers (Figure 3.4G) analogous to those formed by ΔVD in the presence of Mff (Figure 3.4E). This demonstrates that the

R376E mutation does not directly disrupt Drp1-Mff interaction. Yet, the ΔVDR376E -Mff assemblies in solution were not as large or as ordered as those seen with ΔVD and Mff, which agrees with the less prolific stimulation in activity (~5-fold for ΔVDR376E versus ~10- fold for ∆VD).

Functional assembly of ΔVD and ∆VDR376E with Mff tethered to either SL or SL/CL was comparable. GTPase activity of ΔVDR376E was stimulated 20-fold and 16-fold using Mff-

-1 R376E decorated SL and SL/CL templates, respectively (kcat = 12.7 min for ΔVD with SL +

Mff and 13 min-1 for ΔVDR376E with SL/CL + Mff; Figure 3.5 B & G). Moreover, addition of

ΔVDR376E to Mff-tethered lipid templates led to formation of filamentous complexes

(Figure 3.5E) that were similar to ΔVD oligomers under the same conditions (Figure 3.5C).

Collectively, these results demonstrate that the R376E mutation does not directly inhibit

interactions between Drp1 and Mff. Rather the Drp1-3R376E mutant augments the

propensity of the full-length protein to multimerize in solution, and this equilibrium shift

towards higher-order multimers impairs Mff-induced self-assembly on membranes.

Removal of the VD reverts the prematurely multimeric full length Drp1-3R376E to predominantly dimeric species, which rescues functional interactions with Mff.

Mff Multimerization Enhances Drp1 Assembly and GTPase Activity

Having established that the multimeric state of Drp1 potentiates interactions with

Mff, we sought to examine whether sequence variation in Mff would affect Mff-induced

Drp1 assembly. We focused on two domains conserved among all Mff splice variants: a

pair of N-terminal repeats and a conserved coiled-coil (CC) motif immediately preceding

its TM segment (Figure 3.6A). These domains are particularly interesting because each

domain distinctly impacts Mff function. The repeat domains have been shown to be

important for interaction with Drp1, while the CC has been implicated in Mff

multimerization (Gandre-Babbe and van der Bliek 2008; Koirala et al. 2013; Otera and

Mihara 2011).

We first mutated the 4 amino acid core of each individual repeat (VPER or VPEK) to

alanines, and found that either mutation resulted in a great reduction of Mff solubility.

Due to this reduced solubility, these mutants were deemed unusable. On the other hand,

75 deletion of the CC domain yielded soluble protein, so its role in Drp1-Mff interactions was explored.

SEC-MALS analysis revealed that Mff∆TM (26 kDa) is predominantly a tetramer in solution (Figure 3.6B, black trace). By striking contrast, Mff∆CC-TM (22 kDa) was exclusively monomeric (Figure 3.6B, green trace), which clearly demonstrates the role of the CC motif in Mff multimerization. Additionally, the MffΔCC-TM mutant was unable to stimulate ΔVD GTPase activity or self-assembly in solution (Figure 3.6, G & C, respectively). Thus, the CC motif plays an important role in Mff tetramerization, which coordinates stable ∆VD interactions to promote assembly of filamentous polymers.

On the other hand, the tethering of MffΔCC-TM to a membrane template improved its ability to stimulate ΔVD activity (Figure 3.6H). Consistent with this observation, ΔVD polymerization was observed when it was added to liposomes decorated with MffΔCC-

TM (Figure 3.6F). Moreover, membrane deformation was observed leading to the formation of well-ordered helical oligomers of ∆VD on the template. Therefore, the high local concentration of Mff∆CC-TM on the lipid scaffold enhanced stable interactions with

Drp1 to promote oligomerization. This result shows that removal of the CC, and subsequent disruption of the Mff tetramer, allowed the Drp1 oligomers to enforce lipid curvature. Conversely, Mff tetramers, assembled via the CC-motif, provided a less flexible scaffold that interacted with Drp1 to nucleate filamentous assemblies on the lipid surface.

Based on this observation, we concluded that limited mobility of Mff complexes resists

Drp1–induced remodeling of the lipid template.

Since removal of the CC motif from Mff allowed Drp1 to impose curvature on the

SL/CL template, a more malleable lipid template was examined with Mff∆TM tethered.

Previous studies have shown that addition of PE to liposomes results in a more fluid lipid

bilayer that Drp1 can more readily deform (Macdonald et al. 2014). Therefore, a third lipid

mixture (SL/PE/CL) was generated containing 35 mol% PE. When ΔVD was added alone,

no interaction was observed with the SL/PE/CL template (Figure 3.6D). But when the same

template was decorated with Mff∆TM, addition of ∆VD led to the formation of well-

ordered protein-lipid tubules (Figure 3.6E). Interestingly, the Mff-induced stimulation in

GTPase activity was not enhanced by the helical polymerization of ∆VD. Rather, Mff

decoration of SL/CL or SL/PE/CL templates yielded comparable increases in activity

(Figure 3.6I) despite the apparent differences in structure (disconnected filaments versus

a tightly packed helical lattice; Figs. 3.5C and 3.6F, respectively). Nevertheless,

incorporation of PE enhanced the fluidity of the lipid template and allowed the Mff

tetramer complex to recruit Drp1 polymers that deformed the membrane. This

demonstrates the ability of Mff to nucleate Drp1 polymerization at sites of active

membrane remodeling.

3.4 DISCUSSION

Genetic and cellular studies have clearly demonstrated the importance of Mff in recruiting Drp1 to the MOM (Otera et al. 2010; Gandre-Babbe and van der Bliek 2008).

But it remained unclear how Mff directly influences Drp1 function and cellular

localization. These studies reveal the inherent ability of Mff to stimulate Drp1 self-

assembly and GTPase activity in vitro. Previously, this role remained uncharacterized

because Drp1-Mff interactions are transient and strongly influenced by the oligomeric tendencies of both proteins. Factors that alter the assembly properties of either, or both, proteins have the potential to alter interactions within this mitochondrial fission complex.

Initially, the role of the VD was examined based on its apparent proximity to the

membrane as well as its ability to interact with CL (Bustillo-Zabalbeitia et al. 2014;

Macdonald et al. 2014; Ugarte-Uribe et al. 2014). This proposed location would also place

it directly adjacent to receptor proteins on the MOM to promote intermolecular

interactions. Despite this proximity to partner proteins at the surface of the membrane,

our results show that the VD indirectly regulates Drp1 interactions with partner proteins

by modulating its oligomeric propensity. Interestingly, Mff selectively assembles dimeric

Drp1, which represents a subset of Drp1 multimers observed in these studies. Since the

ΔVD mutant yields exclusively dimeric species, an enhanced cooperative interaction with

Mff was observed. Self-assembly of the ΔVD dimers in the presence of Mff led to

formation of filaments with a diameter of ~24 nm. This parallels the ~23 nm width of Drp1

polymers observed in the crystal lattice used to determine the atomic structure of Drp1

(Fröhlich et al. 2013). This geometry suggests that Drp1 middle domain and GED

interfaces are responsible for the formation of these filamentous structures.

Correspondingly, when Drp1 self-assembly is disrupted by the G363D mutation, Mff-

induced self-assembly and stimulated activity is ablated. Thus, Mff guides productive

Drp1 self-assembly, and augmentation of Drp1 activity is not due to Mff interactions alone. Instead, maximal stimulation of ∆VD GTPase activity was achieved upon formation of extended filamentous structures in the presence of Mff in solution and at the membrane. Therefore, the fundamental mechanism of Drp1 activation by Mff is independent of lipid interactions, and is instead a direct result of intermolecular contacts that are enhanced with the ∆VD mutant. This indicates that Mff coordinates Drp1 self- assembly, which enhances its activity.

Unlike ∆VD, WT Drp1 interactions with Mff are regulated by the diversity of multimers formed in solution. In solution Drp1 interchanges among dimers and higher ordered multimers at physiologic salt concentrations (Macdonald et al. 2014), and we propose that larger oligomers are unable to form functional interactions with Mff. While this manuscript was in revision, another study implied that Mff selectively recruits oligomeric

Drp1 (R. Liu and Chan 2015); however, our results are clearly incongruent with this finding. In point of fact, the R376E mutation within Drp1 alters its assembly properties and favors higher-order multimers in solution that impede functional interactions with

Mff. Interestingly, this change is dependent on the VD as deletion of this region rescues the dimeric tendencies of the R376E mutant, and functional interactions with Mff are restored. Therefore, this residue directly influences the conformational sampling of the

VD and its ability to regulate Drp1 oligomerization.

Correspondingly, VD interactions with the membrane can influence cooperative Drp1-

Mff interactions. We determined that when full-length Drp1-3 was incubated with Mff tethered-liposomes containing limited amounts of CL, the stimulation in activity was

79 greater than when Mff was coupled to liposomes lacking CL. Moreover, this effect was abolished when the VD was removed. Thus, interactions between the VD and the membrane have the potential to stabilize Drp1 dimers and promote cooperative interactions with Mff. In fact, complementary studies (see Macdonald et al., 2015) clearly show that Mff-stimulation of Drp1 activity is synergistic with CL-stimulation. These results are congruent with previous reports of VD interactions with CL that promote Drp1 recruitment to lipid bilayers (Bustillo-Zabalbeitia et al. 2014; Macdonald et al. 2014;

Stepanyants et al. 2015; Ugarte-Uribe et al. 2014). Given that CL has been shown to stabilize dimeric Drp1 at the lipid surface to promote Drp1 self-assembly (Macdonald et al. 2014), we propose that CL interactions at the membrane directly promote Drp1 dimer interactions with Mff (Figure 3.7).

Interestingly, the same stimulation was not seen when Drp1-1 was added to Mff- decorated liposomes. Therefore, the presence of the 37-amino acid B-insert within the

VD can further regulate Drp1 interactions with Mff as shown in a companion study

(Macdonald et al., 2015). Given that as many as 8 isoforms of Drp1 (Strack, Wilson, and

Cribbs 2013; C. H. Chen et al. 2000) and at least 9 splice variants of Mff have been identified (Gandre-Babbe and van der Bliek 2008), the potential complexity of interactions that are regulated by sequence changes is vast. Regardless, these results demonstrate how natural sequence modifications alter interactions between Drp1 and one of its receptors. Native sequence changes in this region due to alternative splicing and post-translational modifications have the potential to “tune” Drp1 interactions with partner proteins by altering its assembly properties.

In line with previous studies (Fröhlich et al. 2013; Strack and Cribbs 2012), we also confirmed that disruption of the variable domain sequence altered the efficiency of mitochondrial fission in cells. While we clearly demonstrate that ∆VD is hypomorphic compared to WT when expressed in cells lacking Drp1, significant mitochondrial fission activity is retained. We also show that ∆VD is recruited to mitochondria in a predominantly Mff-dependent manner. Previous studies have differed in their assessments of ∆VD function, as removal of this region has been proposed to both enhance and impair mitochondrial fission. This discrepancy may be due to the design of the VD deletion constructs, the level of over-expression, and the cell lines used in these studies. We find clear mitochondrial localization in our analyses, which indicates that the hypomorphic phenotype is likely due to post-recruitment activity. This agrees with recent experiments showing that ∆VD can tubulate liposomes in vitro, but GTP-induced constriction of these membranes is diminished compared to WT Drp1 (Francy et al. 2015).

Consequently, ∆VD-induced constriction may result in infrequent membrane scission, consistent with the observed decrease in mitochondrial fragmentation. Taken together with our biochemical observations, these results reveal that the fundamental role of Mff is to provide a scaffold for Drp1 self-assembly, and molecular alterations that change the assembly properties of Drp1 or Mff can regulate this interaction.

The present lack of structural information for Mff makes it hard to predict where direct interaction sites would reside, and structural prediction software suggests that the cytosolic portion of Mff is largely disordered. One exception is the CC-motif, which is predicted to form a helical segment adjacent to the C-terminal TM region, which

promotes Mff self-assembly. Consistent with this prediction, Mff was found to exist as a stable tetramer in solution, which enhances the ability of Mff to coordinate Drp1 interactions as evidenced by the formation of Drp1 filaments in solution and proximal to membrane templates. Removal of the CC resulted in Mff monomers that could not stabilize Drp1 interactions in solution. Still, the use of a lipid template enhanced the local concentration of Mff∆CC-TM and provided an adequate scaffold for Drp1 recruitment and

self-assembly. Moreover, membrane tubulation was observed, which suggests that

flexibility within the scaffold and/or lipid template determines the extent to which Drp1

can impart curvature on the membrane. Accordingly, incorporation of PE to the more

rigid SL/CL template enhanced membrane fluidity and/or lateral movement of the Mff

tetramer such that Drp1 oligomerization was able to deform the membrane and generate

tubular structures. Based on these results, Mff provides a platform for the nucleation of

Drp1 oligomers that subsequently impose curvature on the underlying membrane.

Overall, we propose a model wherein Drp1 dimers selectively interact with cytosolic

segments of Mff tetramers (Figure 3.7) to constitute functional copolymers, and

mutations or factors that alter the oligomeric state of Drp1 affect the efficiency of its

recruitment to the MOM. In addition, membrane interactions have the potential to

stabilize Drp1-Mff complexes localized at mitochondrial constriction sites. Moving

forward, the versatile tools employed in this study provide a means to explore how Drp1

interactions with different adaptor proteins on the MOM regulate its structure and

activity.

3.5 MATERIALS AND METHODS

Protein Constructs and Mutagenesis

Drp1 Isoforms 1 and 3 (Drp1-1 and Drp1-3; Uniprot IDs O00429-1 and O00429-4) and

Drp1 Isoform 1 lacking residues 517-639 (ΔVD) were cloned into a pCal-n-EK vector with a human rhinovirus 3C protease (HR3CP) site as previously described (Francy et al. 2015).

Mff lacking its transmembrane (TM) segment (Mff∆TM; Uniprot ID Q9GZY8-5, residues 1-

218) and Mff lacking both its CC and TM (Mff ΔCC-TM; residues 1-186) were cloned into pET28a with a C-terminal 6-His affinity tag using NcoI and HindIII restriction sites introduced during PCR amplification. Site-directed mutagenesis was performed using the

QuikChange Lightning kit (Agilent).

Protein Expression & Purification

All Drp1 constructs were expressed in BL21-(DE3) Star E. coli in LB broth for 24 hours at 18°C after induction with 1 mM IPTG. Cells were harvested by centrifugation, and stored at -60°C until purification. Cells were resuspended in Cal-A Buffer (0.5 M L-Arginine pH 7.5, 0.3 M NaCl, 5 mM MgCl2, 2 mM CaCl2, 1 mM imidazole, and 10 mM BME) with pefabloc-SC (0.5 mM final) and cells were lysed by sonication on ice. Cell lysates were cleared by centrifugation at 150,000 xg for 1 hour at 4°C, and the supernatant was isolated. Affinity capture was performed using gravity filtration with calmodulin agarose

(Agilent) pre-equilibrated with fresh Cal-A buffer. After loading the supernatant, the resin was washed with 25 column volumes (CV) of Cal-A, and protein was eluted with 0.5 CV elutions with Cal-B buffer (0.5 M L-Arginine pH 7.5, 0.3 M NaCl, 2.5 mM EGTA, and 10 mM

BME). Protein-containing fractions were pooled and incubated overnight at 4°C with His-

tagged HR3CP. This solution was concentrated using a 30,000 molecular weight cut-off

(MWCO) centrifugal concentrator for all constructs excluding ΔVD due to its propensity

to polymerize and fall out of solution. Protein samples were further purified by size

exclusion chromatography (SEC) using an AKTA Purifier FPLC (GE Healthcare) and a

Superdex 200 16/600 column equilibrated with a HEPES column buffer containing 150

mM salt (HCB150: 50 mM HEPES(KOH) pH 7.5, 0.15 M KCl, and 10 mM BME) and 5 mM

MgCl2. All Drp1-containing fractions resolved by the column were collected, concentrated, and glycerol was added to 5% final. This isolated protein was aliquoted, frozen, and stored at -60°C until use.

Mff was expressed in BL21-(DE3) E. coli in LB broth for 4 hours at 30°C after induction

with 0.5 mM IPTG. Cells were harvested by centrifugation and stored at -60°C until

purification. Cells were resuspended in immobilized metal affinity chromatography

(IMAC)-A buffer (50 mM Tris-HCl pH 7.5, 0.5 M NaCl, 20 mM imidazole, 10 mM BME) with

pefabloc-SC (0.5 mM final), and cells were lysed by sonication on ice. Cell lysates were

cleared as described above, and affinity capture was performed using FPLC and a

prepacked HiTrap IMAC column (GE Healthcare) charged with Ni2+ and equilibrated with

IMAC-B (50 mM Tris-HCl pH 7.5, 0.1 M NaCl, 20 mM imidazole, 10 mM BME). Clarified lysate was loaded onto the column, and washed to baseline with IMAC-B. Mff was eluted from the column with a linear gradient from 0-100% IMAC-C (50 mM Tris-HCl pH 7.5, 0.1

M NaCl, 250 mM imidazole, 10 mM BME) over 10 CV. Protein-containing fractions were

pooled, diluted 10-fold in ion exchange (IEX)-A (50 mM Tris-HCl pH 7.5, 10 mM BME), and

loaded onto a Q Sepharose anion exchange column (GE Healthcare). The column was

84 washed to baseline with IEX-A, and Mff was eluted by adding of 10% IEX-B (50 mM Tris-

HCl pH 7.5, 1 M NaCl, 10 mM BME). Peak Mff fractions were pooled, concentrated, and subjected to SEC using a Superdex 200 16/600 column with HEPES column buffer containing 300 mM salt (HCB300: 50 mM HEPES (KOH) pH 7.5; 0.3 M NaCl; 10 mM BME).

Mff peak fractions were pooled, concentrated with a 10,000 MWCO centrifugal concentrator, 5% glycerol was added, and aliquots were frozen and stored at -60°C until use.

Liposome preparation

Three distinct lipid formulations were utilized in this study: scaffold liposomes (SL:

96.7% 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), 3.3% 1,2-dioleoyl-sn-glycero-3-

[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] (nickel salt) (DGS-NTA(Ni2+))) and scaffold liposomes with cardiolipin (SL/CL: 86.8% DOPC, 3.3% DGS-NTA(Ni2+), 9.9% bovine heart cardiolipin), and scaffold liposomes with phosphatidylethanolamine (PE) and CL (SL/PE/CL: 51.8% DOPC, 3.3% DGS-NTA(Ni2+), 35% 1,2-dioleoyl-sn-glycero-3- phosphoethanolamine (DOPE), and 9.9% bovine heart cardiolipin). All lipids used in this study were purchased from Avanti Polar Lipids (Alabaster, AL). Lipids were mixed and dried to a thin film with a stream of nitrogen gas. Trace solvent was removed with a speed vac at 37°C for 1 hour immediately after films were prepared. The lipids were rehydrated in HCB150 for 30 minutes in a 37°C water bath with occasional vortexing. Resuspended lipid solutions were freeze-thawed and extruded through a 1.0 μm polycarbonate filter.

Lipid solutions were stored at 4°C or on ice until use.

GTPase Assay

Drp1 GTPase activity was determined using a colorimetric phosphate generation assay

(Leonard et al. 2005) with some modifications. Briefly, Mff (4.95 μM final) was diluted to

4x with HCB150 in the presence or absence of SL or SL/CL (150 μM total lipid final) for 15

minutes at room temperature. This solution was added to 1.2x Drp1 (500 nM final), and

incubated for an additional 15 minutes at room temperature. 3x GTP/Mg2+ (1 mM and 2

mM final, respectively) in HCB150 was added to Drp1/Mff mixtures to start reactions and

samples were incubated at 37 ºC. At designated timepoints, EDTA (0.1 M final) was added

to sample aliquots. Malachite green reagent (1 mM malachite green carbinol, 10 mM

ammonium molybdate tetrahydrate, and 1N HCl) was added to each sample and A650 was

measured using a Versamax microplate reader (Molecular Devices). Using Excel

(Microsoft), the obtained raw phosphate levels were converted into rates using all data

that contributed to a linear trend (minimum of 3 timepoints). These rates were converted

to kcat by accounting for Drp1 concentration, and were plotted in GraphPad Prism 6.

Statistical significance was determined using an unpaired t-test.

The concentration dependence of Mff-induced stimulation of Drp1 GTPase activity

was assayed using varying Mff concentrations. The obtained kcat was plotted as a function of Mff concentration, and the data were fit with the nonlinear log(agonist) vs response fit using GraphPad Prism 6.

Negative Stain Transmission Electron Microscopy

For all microscopy, samples were prepared as indicated, adsorbed to carbon-coated

grids, and stained with 2% uranyl acetate (UA). Samples were visualized on a Technai T12

(FEI Co.) electron microscope at 100 keV and images were acquired using a Gatan 4k x 4k camera at a magnification of 49,000x (in the absence of liposomes) or 18,500x (in the presence of liposomes).

Size exclusion Chromatography with Multi-angle Light Scattering

In order to accurately determine the oligomeric distribution of Drp1 in solution, proteins were fractionated on a Superose 6 10/300 GL SEC column in HCB150 containing

1 mM DTT rather than BME. Column eluate was analyzed by tandem miniDAWN Treos

MALS and Optilab rEX differential refractive index detectors (Wyatt Technologies) as previously described (Macdonald et al. 2014). Molar mass was determined using the

ASTRA 6.1 software package (Wyatt technologies) and was plotted with molar mass (right axes) and normalized refractive index (left axes) as a function of elution volume. Drp1-3

(10 μM), ΔVD (6 μM), and Mff (75 μM) and corresponding mutants were loaded in a total volume of 0.5 mL.

Cosedimentation assay

To identify proteins within oligomeric complexes, an ultracentrifugation sedimentation assay was used. Briefly, Drp1 constructs (2 μM) in the absence and presence of Mff∆TM (10 μM) were combined in HCB150 for 2 hours at room temperature.

Samples were centrifuged 30 minutes at 160,000 x g at 4°C. Supernatant was discarded, and pellets were washed and recentrifuged twice. The final pellet was resuspended in 1x

Laemmli buffer, resolved by SDS-PAGE, and stained with coomassie dye (Expedeon,

Cambridge UK) to identify proteins that sedimented in oligomeric complexes.

Glutaraldehyde crosslinking EMSA

Mff interactions with ΔVD and ΔVDG363D was investigated using a chemical crosslinking electrophoretic mobility shift assay. Briefly, 3 μM ΔVD or ΔVDG363D were incubated 30 minutes in the presence or absence of 15 μM MffΔTM in HCB150. Each protein combination was treated with either HCB150 or 5.5 mM glutaraldehyde diluted in HCB150 for 30 minutes at room temperature. Crosslinking was quenched with Tris-HCl pH 7.5 at a final concentration of 150 mM for at least 15 minutes at room temperature. Samples were dissolved in Laemmli buffer, and resolved by SDS-PAGE. Completed SDS-PAGE gels were either stained for total protein by coomassie staining (Expedeon, Cambridge UK) or transferred to a PVDF membrane and Mff was detected using an anti-His antibody

(1:3,000, Thermo Scientific).

Cell culture and immunocytochemistry

All mouse embryonic fibroblasts (MEFs) were maintained in DMEM supplemented with 10% heat-inactivated fetal calf serum (FBS) and 1% penicillin/streptomycin at 37C in

5% CO2/95% air. Cells were transfected with 1 μg of plasmid DNA encoding either Myc- tagged Drp1 isoform 1, or Myc-Drp1 isoform 3, or Myc-Drp1 ∆VD using TransIT-2020

Transfection Reagent (Mirus Bio, Madison, WI) according to the manufacturer’s protocol.

Cells cultured on coverslips were washed with cold phosphate buffered saline (PBS), fixed in 4% formaldehyde, and permeabilized with 0.1% Triton X-100. After incubation with 2% normal goat serum (to block nonspecific staining), fixed cells were incubated overnight at 4°C with rabbit anti-Tom20 (1:500; Santa Cruz Biotechnology, Santa Cruz, CA)

and mouse anti-Myc (1:500; Santa Cruz Biotechnology) primary antibodies. Cells were

washed with PBS and incubated with Alexa Fluor 488–conjugated anti-mouse and Alexa

Fluor 568–conjugated anti-rabbit secondary antibodies (1:500; Invitrogen, Carlsbad, CA)

for 60 min at room temperature. Coverslips were mounted on glass slides and imaged by

confocal fluorescence microscopy using an Olympus FV1000 IX81 confocal microscope

(Olympus USA).

To quantitate mitochondrial fragmentation, cells were immunostained with anti-

Tom20 and anti-Myc antibodies as described. Mitochondrial morphology was then

examined in Myc-Drp1-expressing cells. The percentage of Myc-Drp1–expressing cells

with fragmented mitochondria relative to the total number of Myc-Drp1-expressing cells

was calculated. To quantitate Drp1 localization on the mitochondria, Pearson’s

coefficienct of ∆VD localization on mitochondria was calculated in cells expressing Drp1

∆VD.

Assessment of Drp1 expression in MEFs

Drp1 KO MEFs were transfected with the indicated plasmids as described. Total

protein was harvested 24 h after transfection, and protein concentration was determined

by Bradford assay. Thirty μg of total protein was resuspended in Laemmli buffer, resolved

by SDS–PAGE, and transferred onto nitrocellulose membranes. Membranes were probed

with anti-Myc and anti-actin antibodies (1:1000 dilution for both, Sigma-Aldrich) followed

by visualization using enhanced chemiluminescence.

3.6 ACKNOWLEDGEMENTS

The authors thank Heather Holdaway and Hisashi Fujioka for their expertise and advice

with electron microscopy studies. We acknowledge Hiromi Sesaki and David Chan for

providing Drp1 and Mff knockout MEFs, respectively. This work was supported by the

American Heart Association (Grant ID: SDG12SDG9130039 for JAM and 13BGIA14810010

for RR), and CAF was supported by a NIH training grant (Grant ID: 2T32GM008803-11A1).

XQ was supported by NIH R01 (NS088192).

FIGURE 3.1

FIGURE 3.1 The variable domain (VD) of Drp1 is a negative-regulator of Mff-induced self- assembly. (A) Schematic representation of the proteins used in this study, including WT

Drp1 isoforms 1 and 3 (Drp1-1 and Drp1-3) and the ∆VD mutant. The GTPase, middle, variable (VD), and GTPase effector (GED) domains are highlighted. The yellow region in

Drp1-1 represents the alternatively spliced B-insert. (B) GTPase activity of Drp1-1, Drp1-3 and ∆VD (0.5 μM final for each) in the absence (white) or presence (black) of Mff (5 μM) in solution. *: p-value < 0.0005. (C) GTPase activity of ∆VD (0.5 μM final) in the presence of Mff∆TM at various concentrations. (D) SEC-MALS was used to assess the solution oligomer state of Drp1-3 (black trace) and ∆VD (red trace). Dotted lines on the right axis correspond to the indicated oligomer based on the predicted molecular weight of indicated Drp1 multimers (E-J) Negative stain EM images of Drp1-1 (E & H), Drp1-3 (F &

I), and ∆VD (G & J) are shown in the absence (E-G) or presence (H-J) of Mff. Scale bar, 100 nm.

FIGURE 3.2

FIGURE 3.2 Coupling of Mff to topology-enforcing liposomes enhances Drp1 stimulation.

(A) Schematic representations of the scaffold liposomes used in this study. Drp1 does not interact with liposomes in the absence of Mff (left), but Drp1 is recruited to Mff-decorated liposomes (right). (B & C) GTPase activities were measured for Drp1-1, Drp1-3 and ∆VD

(0.5 μM for each) in the absence (white) or presence (black) of Mff (5 μM) tethered to SL

(B) and SL templates with 10% CL (C, SL/CL: 150 μM). *: p-value < 0.0005. (D-G) Negative

stain EM images of Drp1-3 (D & E) and ∆VD (F & G) (1 μM) added to scaffold liposomes

(SL: 150 μM) in the absence (D & F) or presence (E & G) of Mff∆TM (5 μM). Scale bar, 100 nm.

FIGURE 3.3

FIGURE 3.3 The VD is not essential for mitochondrial targeting and subsequent fission in MEF cells. (A) Representative fluorescence micrographs of Drp1-/- MEFs transfected with Myc-tagged Drp1-1, Drp1-3, or ΔVD. Confocal imaging analysis was carried out using anti-Myc (green) and anti-Tom20 (a marker of mitochondria, red) antibodies. (B)

Mitochondrial fragmentation within transfected cells was quantitated as the percentage of Myc-Drp1-expressing cells with fragmented mitochondria relative to total Myc- expressing cells. (C) Western blot analysis of Myc-Drp1 expression in Drp1 knockout MEFs

24 hrs post-transfection. Actin was used as a loading control. (D) MffWT (top) and Mff-/-

(bottom) MEFs were transfected with Myc-∆VD. Confocal imaging analysis was carried out using anti-Myc (green) and anti-Tom20 (red) antibodies. (E) Quantitation of mitochondrial fragmentation in MffWT (white) and Mff-/- (black) MEFs expressing Myc-

∆VD. (F) ∆VD/Tom20 co-localization in (D) was determined from confocal images by calculating the Pearson’s coefficient. #: p < 0.05 and *: p < 0.0005 compared with Myc- vector transfected cells.

FIGURE 3.4

FIGURE 3.4 Mutations that alter the multimeric equilibrium of Drp1 interfere with Mff-

induced self-assembly. (A-B) SEC-MALS was used to assess the multimeric distributions

of WT and mutant proteins. WT Drp1 (isoform 3, black trace) is compared to the G363D

(blue) and R376E (red) mutants in (A). ΔVD (black trace) and the corresponding double

mutants ∆VDG363D (blue) and ∆VDR376E (red) are shown in (B). Dotted lines indicate the

predicted molecular weights of Drp1 multimers. (C) GTPase activity was measured for

Drp1-3, Drp1-3G363D and Drp1-3R376E in solution in the absence (white) and presence

(black) of Mff. (D) Similarly, GTPase activity was measured for ∆VD, ∆VDG363D and ∆VDR376E

in solution in the absence (white) and presence (black) of Mff. *: p-value < 0.0005. (E - G)

Negative stain EM images of Mff∆TM (5 μM) incubated with ∆VD (E), ∆VDG363D (F), and

∆VDR376E (G). Scale bar, 100 nm. (H) Mff∆TM (10 μM) was co-sedimented with Drp1-1,

∆VD, and ∆VDG363D (2 μM each) using ultracentrifugation. The input (I) and the final

washed pellet (P) were separated by SDS-PAGE and stained with coomassie dye. (I)

Mff∆TM (15 μM) was incubated alone, or with ∆VD, and ∆VDG363D (3 μM each) in the presence or absence of glutaraldehyde for 30 minutes. Samples were resolved by SDS-

PAGE, and stained with coomassie (left). Mff was detected using an anti-6-His antibody

(right). A high molecular weight band was observed in samples containing Mff crosslinked

with ∆VD and ∆VDG363D (indicated by filled arrowheads). The same complex is not

observed in other samples (indicated by open arrowheads).

FIGURE 3.5

FIGURE 3.5 Removal of the VD rescues the R376E defect in Mff-induced assembly. (A)

GTPase activity of Drp1-3, Drp1-3G363D and Drp1-3R376E (0.5 μM final) was determined in

the presence of undecorated liposomes (white, SL:150 μM final) or Mff-decorated

liposomes (black, SL/Mff∆TM: 150 μM / 5 μM final). (B) GTPase activity of ∆VD, ∆VDG363D

99 and ∆VDR376E (0.5 μM final) was determined in the presence of undecorated liposomes

(white, SL:150 μM final) or Mff-decorated liposomes (black, SL/Mff∆TM: 150 μM / 5 μM final). (C-E) Negative stain EM of ∆VD (C), ∆VDG363D (D), or ∆VDR376E (E) (1µM each) incubated with Mff-decorated liposomes (SL/Mff∆TM: 150µM/5 μM). Scale bar, 100 nm.

(F) GTPase activity of Drp1-3, Drp1-3G363D and Drp1-3R376E (0.5 μM final) in the presence of liposomes with limited CL (white, SL/CL: 150 μM final) or the same liposomes decorated with Mff (black, SL/CL/Mff∆TM: 150 μM / 5 μM final). (G) GTPase activity of ∆VD, ∆VDG363D and ∆VDR376E (0.5 μM final) in the presence of liposomes with limited CL (white, SL/CL:

150 μM final) or the same liposomes decorated with Mff (black, SL/CL/Mff∆TM: 150 μM

/ 5 μM final). *: p<0.0005.

100

FIGURE 3.6

FIGURE 3.6 Oligomerization of Mff cytosolic domains promotes Mff-induced Drp1 self- assembly. (A) Schematic representation of the Mff constructs used in this study, including

Mff∆TM and Mff∆CC-TM. The N-terminal repeat segments (orange), CC-motif (blue), and

C-terminal 6-His tag (yellow) are highlighted. (B) SEC-MALS was used to determine the multimeric state of Mff∆TM (black) and Mff∆CC-TM (green). Dotted lines indicate the predicted molecular weights of Mff multimers. (C-F) Negative stain EM images of ∆VD in the presence of Mff∆CC-TM in solution (C), in the presence of undecorated liposomes

101 containing PE (25%) and CL (10%) (D: SL/PE/CL), in the presence of Mff∆TM tethered to the same liposomes (E: SL/PE/CL/Mff∆TM), or in the presence of Mff∆CC-TM tethered to liposomes lacking PE (F: SL/CL). Scale bars, 100 nm. (G-H) GTPase activity of ΔVD was measured alone and in the presence of Mff∆TM or Mff∆CC-TM in solution (G) or tethered to SL/CL (H). (I) ΔVD activity was also measured in the absence (white) or presence (black) of MffΔTM coupled to SL/CL and SL/PE/CL lipid templates as indicated. *: p-value <

0.0005.

102

FIGURE 3.7

FIGURE 3.7 Mff selectively promotes oligomerization of assembly-competent Drp1 dimers. WT Drp1 exists as a mixture of multimeric species in solution (dimers and larger multimers, black arrows). Coincident interaction of the VD with CL in the membrane (red area of the bilayer) relieves its regulatory effect, which stabilizes Drp1 dimer interactions with Mff to promote assembly of the fission machinery.

103

CHAPTER 4: MFF INTERACTS WITH THE STALK OF DRP1 VIA A NOVEL VD-OCCLUDED

INTERFACE TO BUILD A MEMBRANE REMODELING COPOLYMER

This chapter has been submitted for publication:

Ryan W. Clinton, Rita A. Avelar, Xin Qi, Janna G. Kiselar, Rajesh Ramachandran, and

Jason A. Mears. Mff Interacts with the Stalk of Drp1 via a Novel VD-Occluded Interface to

Build a Membrane Remodeling Copolymer. Submitted in May 2018.

104

4.1 ABSTRACT

Drp1 is an essential mechanoenzyme among eukaryotic organisms that is recruited to

the mitochondria to impart the force required for membrane fission. One of the

membrane-bound Drp1 recruiting factors, Mff, is an integral MOM protein that augments

Drp1 activity. In this study, we find that Drp1 and Mff work in concert to reshape lipid bilayers to narrow diameters in a GTP hydrolysis-dependent manner. This co-polymer

assembles through an interaction between Mff and Loop 1CS in the Drp1 stalk, and a

chimeric mutation replacing residues in this loop with the yeast sequence selectively

disrupt the Drp1-Mff interaction. Using footprinting methods, we show that the regulatory variable domain (VD) of Drp1 occludes Loop 1CS and, thereby, prevents Mff

binding to Drp1. These results explain how deletion of the VD enables Mff interactions.

Collectively, this study identifies key residues driving Drp1-Mff co-assembly to form a

robust membrane remodeling complex.

4.2 INTRODUCTION

Within eukaryotic cells, mitochondria are dynamic organelles that continually

undergo the opposing processes of fission and fusion. Mitochondrial fission is an essential

cellular event that involves the scission of both the mitochondrial outer and inner

membranes (MOM and MIM, respectively). This conserved process is required for the

distribution of mitochondria throughout the cytosol, the isolation of damaged regions of

mitochondria, and the progression of apoptosis, among many other roles. Dynamin-

105

related protein 1 (Drp1), an ancient member of the dynamin family of membrane

remodeling GTPases (Ramachandran and Schmid 2018), is indispensable for mitochondrial fission, and human mutations in this protein are associated with severe neurodevelopmental disorders (Chang et al. 2010; Sheffer et al. 2016), epilepsy (Vanstone et al. 2016), and optic atrophy (Gerber et al. 2017).

As a dynamin family member, Drp1 maintains a conserved domain architecture that includes a catalytic GTPase domain and a self-assembly region called the stalk comprised of the middle and GTPase Effector (GED) domains (Figure 4.2A). In contrast to dynamin, the superfamily’s namesake protein, Drp1 lacks a well-structured lipid interaction motif.

In the place of the pleckstrin homology (PH) domain, which is interposed between the middle and GED in dynamin, Drp1 contains sequence termed the ‘variable domain’ (VD), which is thought to be largely unstructured. Recent studies have demonstrated a strong interaction between the VD and anionic lipids, including the mitochondrial specific cardiolipin (Bustillo-Zabalbeitia et al. 2014; Macdonald et al. 2014; Francy et al. 2017).

This domain also acts as a negative regulator of Drp1 self-assembly (Clinton et al. 2016;

Francy et al. 2015), but the specific mechanism governing this role is unknown.

Drp1 translocation and function at the surface of the mitochondria is a complex

process that involves a myriad of factors including, but not limited to, ER-mitochondrial

contact (Ingerman et al. 2005), Drp1 interaction with lipids of the MOM (Bustillo-

Zabalbeitia et al. 2014; Macdonald et al. 2014; Francy et al. 2017; Adachi et al. 2016;

Stepanyants et al. 2015), and Drp1 interaction with integral receptor proteins in the MOM

(Osellame et al. 2016; Gandre-Babbe and van der Bliek 2008; Otera et al. 2010; Zhao et

106 al. 2011). Of the proteins implicated in recruiting Drp1 to its target membranes, Mff is the most ancient in metazoans (Gandre-Babbe and van der Bliek 2008). Since its discovery, little has been elucidated about the relationship between Mff and Drp1 due to the transient interactions observed in cells and with isolated proteins. In previous studies, the

VD was identified as a negative regulator of Drp1-Mff interactions (Clinton et al. 2016; Liu and Chan 2015). In part, this effect was attributed to changes in the self-assembly properties of Drp1, as premature oligomerization in solution prevented interactions with

Mff. In addition, a coiled-coil motif in Mff was shown to promote self-assembly into a tetramer that preferentially interacted with Drp1 dimers (Clinton et al. 2016).

Dynamin GTPases function as mechanoenzymes whose conformation is dictated by the nucleotide state of the GTPase domain. Liposomes containing cardiolipin or phosphatidylserine (PS) are tubulated by Drp1 and are constricted following hydrolysis of

GTP. Interestingly, GTP binding alone is insufficient to elicit tubule constriction (Francy et al. 2015), which functionally distinguishes Drp1 from classic dynamins (A. C. Sundborger et al. 2014). The mechanoenzymatic core (the GTPase and stalk) of Drp1 is the minimal requirement for constriction of anionic liposomes, but in cells, Drp1 alone is insufficient for maintenance of mitochondrial morphology (Osellame et al. 2016; Loson et al. 2013).

The participation of Drp1 partner proteins in membrane constriction and scission is required, but functional studies that examine these interactions remain an unexplored research area with significant physiologic relevance. The extent to which Drp1 interactions with receptors can impact membrane remodeling must be determined to better understand the functional assemblies driving mitochondrial division.

107

Given the central role of Mff in Drp1-mediated organelle fission, this study set out to

identify whether Mff contributes to the remodeling of lipid bilayers in concert with Drp1.

We found that Mff enhances Drp1 constriction through functional contacts at the base of

the contractile machinery. Specifically, loop 1CS (L1CS) in the stalk of Drp1 was identified

as the Mff binding site, and mutations in this loop selectively and directly destabilize this

interaction. Concurrently, the VD was found to selectively occlude L1CS, which explains the observation that the removal of this domain enhances Drp1 interaction with Mff.

Taken together, these data support a model whereby Mff recruits Drp1 to the MOM via a VD-occluded stalk interface, and this co-polymer drives membrane remodeling from

large tubules to narrow diameters. In this way, Drp1-Mff complex assembly is tightly

regulated to form a potent membrane constriction machinery.

4.3 RESULTS

The minimal Drp1-Mff complex mediates membrane remodeling and constriction

The membrane remodeling activity of Drp1 has typically been investigated in the

context of anionic-lipid-containing membranes. In fact, upon GTP hydrolysis lipids such as cardiolipin are reorganized by Drp1 to form membranes with narrow diameter constrictions (Stepanyants et al. 2015). Still, it has been suggested that Drp1 requires additional factors for membrane fission, including dynamin, leaving the open question: do Drp1’s partner proteins contribute to its membrane-remodeling activity? To this point, in vitro studies of Drp1 membrane remodeling activity have not examined potential

108

contributions by mitochondrial fission partner proteins. Since they are anchored in the

membrane, these partners likely influence functional interactions with the dynamins

driving mitochondrial fission.

Based on this logic, we sought to examine Drp1 interactions with its essential

membrane partner, Mff. While Mff has been shown to augment Drp1 GTPase activity under specific conditions (Clinton et al. 2016; Osellame et al. 2016; Macdonald et al.

2016), its role in Drp1-mediated membrane constriction remained unclear. Previous studies have identified conditions where Mff promotes Drp1 self-assembly on target

membranes. What remains unclear is whether this scaffold prohibits Drp1-driven

membrane remodeling and constriction or whether Mff is able to participate as a partner

in a contractile copolymer. Therefore, the interaction between Drp1 and Mff was

2+ reconstituted on a flexible lipid template (PC55/PE35/Ni 10) in the absence of anionic lipids

(Figure 4.1A). Using negative stain EM, extended protein-decorated lipid tubules (Figure

4.1B - left) were observed. Similar to previous observations of Drp1 tubulation of negatively charged liposomes (Francy et al. 2015; MacDonald et al. 2016), the outer diameter of Drp1-Mff copolymers on lipid tubes was heterogeneous, ranging from ~70-

240 nm with an average of 134 ± 39 nm (Figure 4.1C; n=89). When GTP was added to

these preformed Drp1-Mff-lipid tubules, a rapid constriction was observed (Figure 4.1B –

right), and the observed outer diameters of the protein-lipid tubules narrowed to a

distribution with an average diameter of 58 ± 9 nm (Figure 4.1C; n = 121).

Based on these observations, cryo-electron microscopy (cryo-EM) was used to directly

examine the morphology of the lipid bilayer underlying Drp1-Mff copolymers during the

109

GTP hydrolysis cycle. This method prevents limitations associated with negative stain EM

(i.e. dehydration artifacts, visualization is dependent on the surface adhesion of the

uranyl stain) and can be used to directly visualize the interaction between the protein complex and underlying lipid bilayer (Figure 4.1D – gray box). The outermost diameter of the tubes in the absence of GTP was consistent with those observed in negative stain: 110

± 20 nm (Figure 4.1E – gray; n = 57). This distribution is comparable to diameters observed for Drp1 membrane remodeling of anionic lipid (Macdonald et al. 2016; Francy et al.

2017). However, the underlying lipid tubule was clearly distinct from the outer protein density of the tube, and the Drp1 polymer encircling the tubules had additional spacing at the lipid surface due to Mff-decoration. Two diameters, the protein outer diameter and

the lipid outer diameter, were separately measured to more accurately assess the lipid

constriction by this minimal Drp1-Mff complex. The lipid bilayer itself was considerably

more narrow than the protein density (59 ± 21 nm; n=57). Of note, this displacement of

Drp1 from the membrane is distinct from previously identified helical polymers formed

by Drp1 around lipid nanotubes containing anionic lipid, including cardiolipin or

phosphatidylserine (Francy et al. 2017). Still, a Drp1-Mff complex can actively remodel

flexible liposomes with large diameters (>500 nm) to tubules with diameters averaging

110 nm. This lipid remodeling is specific to the Drp1-Mff complex as previous studies

showed that comparable liposomes decorated with a control protein (GFP) are not

tubulated by Drp1 (Clinton and Mears 2017). Furthermore, the curvature adaptability of

this polymer was apparent from the range of diameters measured, and these geometries

110

were consistent with the measured diameters of Drp1 foci on mitochondria in living cells

(Rosenbloom et al. 2014).

Next, GTP was added to these Drp1-Mff-lipid tubes and early conformational changes

(~5 s) were captured using cryo-EM. Constricted lipid tubules were observed (Figure 4.1F

– red boxes) with protein outer diameters comparable to the tube diameters seen by

negative stain: 67 ± 9 nm (Figure 4.1E – red; n=62). When assessing the constriction of the

lipid tubule, the outer diameter of the membrane was reliably measured (20 ± 3 nm; n=62). With an estimated bilayer thickness of 4 nm, this Drp1-Mff copolymer constriction

results in an average lumen diameter of 11.5 nm at constriction foci. These local

constriction events highlight the initial membrane changes resulting from the

mechanoenzymatic influence of the Drp1-Mff complex. In comparison, the non-

constricted regions (Figure 4.1D – blue box) adjacent to narrow sites retain a diameter

comparable to the tubules in the absence of GTP (Figure 4.1, E-F, blue v. gray). It is unclear

what drives this alternating constriction pattern, but the addition of GTP to Drp1-Mff-lipid

tubules triggered constriction of the underlying lipid bilayer to outer diameters less than

20 nm. Interestingly, it appeared that Mff was not displaced, because the average

thickness of protein surrounding the lipid tubule was not significantly altered when GTP

was added (apo: 26 nm ± 3 nm; GTP constricted: 24 ± 4 nm; GTP non-constricted: 25 nm

± 3 nm). Therefore, an additional layer of Mff circumnavigates the membrane below the

Drp1 oligomer. This added belt of protein beneath the contracting Drp1 polymer further

compressed the membrane, which may enhance constriction of the underlying bilayer

through an undefined mechanism.

111

Based on the experimental design, we know that Mff was tethered at the surface of

the membrane (Figure 4.1A). The Drp1 assembly forming the outer layer of the tubule has

the characteristic arched shape observed with dynamins, which positions the stalk of Drp1

towards the Mff-decorated lipid bilayer. In this way, Mff is juxtaposed between the lipid

bilayer and Drp1 stalk. And because the VD is dispensable for Mff interactions, we

propose that the Drp1 stalk provides an interaction surface for Mff binding.

Stable Drp1-Mff copolymers build a thicker contractile machinery

To confirm that Drp1 interactions with Mff build a helical polymer on lipid bilayers,

this complex was reconstituted on uniform lipid nanotubes to assess the architecture of

the protein-lipid interactions. Specifically, Drp1 was added to Mff-decorated lipid

nanotubes (GC50/PE25/Ni10/CL15) in the presence of GMP-PCP (Figure 4.2B - right). The

polymers observed were comparable to those seen on deformed liposomes. Importantly,

the distance between the outer protein diameter and lipid bilayer of the nanotubes (26.2

± 2.5 nm, n=110) was consistent with that measured for on liposomes even when CL was

added to the template. Therefore, Mff restricts Drp1 interaction with the underlying

membrane, and this observation is reinforced by comparing Drp1 polymers in the

presence (Figure 4.2B – right) and absence of Mff (Figure 4.2B - left) assembled on CL- containing lipid nanotubes (GC40/PE35/CL25; protein thickness = 13.3 ± 1.4 nm, n =114).

Assessment of these distinct polymers using line plots (Figure 4.2B - bottom) highlights

the additional spacing introduced when Mff was tethered on the lipid template.

112

Class averages generated from the Drp1-Mff-nanotube complexes (Figure 4.2C)

revealed a protein-lipid assembly with the nanotube bilayer at the center of the polymer and a T-shaped density at the periphery, consistent with previous Drp1 structures (Francy et al. 2017). While these 2D class averages appear well-ordered, heterogeneity within this

polymer limits our ability to resolve a structure based on cryo-EM reconstruction

methods. Nevertheless, the thickness of the protein density surrounding the nanotube

could be measured and extends 25 nm, of which 15 nm is accounted for by Drp1 density.

Previous studies have shown Mff lacks strong secondary structural elements, so it has

been predicted to be intrinsically disordered (Macdonald et al. 2016). Using DLS with the

stable monomeric MffΔCC-TM mutant, an average hydrodynamic radius of Mff was

measured to be 5.4 nm (10.8 nm diameter average, Figure 4.2D), which correlates well

with the unattributed space between the lipid bilayer and Drp1 polymer in these class

averages. Based on the similarity of these class averages to previously determined Drp1

helical polymers, the GTPase domain of Drp1 was assigned to the peripheral regions of

the protein oligomer. The stalk and VD were positioned adjacent to the Mff-decorated

bilayer, which is consistent with the prediction that the Drp1 stalk resides at the Mff

interface and likely coordinates the interaction between these two proteins.

Mff interactions stabilize the Drp1 Stalk

To further assess the interaction between Drp1 and Mff, a SYPRO-orange thermal

stability assay was used. This assay takes advantage of an environmentally-sensitive dye,

113

SYPRO Orange, whose fluorescence is enhanced upon interaction with hydrophobic

protein patches. Thus, the temperature of protein in the presence of this dye was slowly

increased, and the unfolding of separate protein domains was assessed with factors that

alter their stability. Initial characterization of Drp1 thermal stability identified two discrete

domains that unfold at distinct thermal thresholds (Figure 4.2E – black line). To determine the Drp1 domains associated with each melting event, guanosine nucleotides were introduced to selectively stabilize the GTPase domain during thermal denaturing.

Specifically, GMP-PCP and GDP were pre-incubated with Drp1 at saturating concentrations (Figure 4.2E – red and green lines respectively), and both nucleotides selectively stabilized the GTPase domain of Drp1, which was identified by an increase in the temperature of the first protein unfolding peak (shifted to the right). To further confirm that this peak is indicative of GTPase domain unfolding, a GTPase-GED (GG) fusion protein (Wenger et al. 2013) was subjected to the same thermal stability assay (Figure

4.2E – black dashed line), and the unfolding temperature of this isolated domain overlaid with the first unfolding peak of the intact Drp1 protein. Collectively, these results allowed for the assignment of the unfolding events to the GTPase domain (Figure 4.2A – green

region; unfolding around 40°C) and stalk domain (Figure 4.2A – blue regions; unfolding

around 70°C) of Drp1.

To identify the Drp1 domain(s) stabilized by interactions with Mff, the thermal shift

assay was used to measure the stability of Drp1ΔVD, because it stably interacts with Mff

in solution (Clinton et al. 2016). Similar to full-length Drp1, the unfolding profile of ΔVD

exhibited two peaks corresponding to the GTPase and stalk domains (Figure 4.2F – black

114 line). When Mff was added to this mutant, the peak corresponding to stalk unfolding was specifically stabilized (Figure 4.2F – red line). Therefore, the stalk promotes assembly of the Drp1-Mff complex, and this result reinforced previous findings that the VD is dispensable for Mff interactions.

Sequence conservation analysis identifies Drp1 stalk loops as potential sites of Mff interaction

Based on the observation that the Drp1 stalk is in close proximity to Mff in a lipid- proximal co-complex, amino acids in this region were examined to identify candidate sequences responsible for interactions with Mff. To begin, sequence conservation was compared between Drp1 and the yeast homolog, Dnm1p (Figure 4.3B). Specifically, regions lacking conservation were of interest since no Mff homolog exists in yeast (Figure

4.3A). As an example, this analysis ruled out residues within the stalk loop 1N, such as

G350, since this residue is conserved in Dnm1p and its mutation impairs the self-assembly properties of Drp1 (Chang et al. 2010; Bhar et al. 2006; Ingerman et al. 2005). Second,

Drp1 and metazoan Drp1 homologues were compared (Figure 4.3C) to identify regions of conservation, as Mff is required for mitochondrial recruitment of Drp1 in these species.

Last, Drp1 and other human dynamin proteins (Figure 4.3D) were compared, as Mff has been shown to interact exclusively with Drp1 and not other dynamin-family proteins.

Based on these multifaceted comparisons, two regions of interest were identified: loop

1CS and loop 3S. Importantly, residues in these loops are not buried and are selectively

115

conserved in metazoan Drp1 homologues (not Dnm1p) unlike previous mutants that were

suggested to directly disrupt the Drp1-Mff interaction (R. Liu and Chan 2015).

Loop 1CS of Drp1 is structurally unique when comparing crystal structures of human

dynamins (Figure 4.3E - left). A well-conserved ‘LGG’ motif in the center of this loop was

mutated to the corresponding sequence from Dnm1p (TSN), which resulted in a chimeric

Drp1 mutant termed Drp1TSN. The second region of interest, loop 3S of Drp1, was particularly interesting because a flexible loop does not exist in this region of the stalk in other human dynamins (Figure 4.3E - right). For this region, a mutation was made that

replaced a conserved Drp1 sequence (QE) with a positively charged di-lysine, referred to

as Drp1KK, since these residues introduce a charge reversal that is common in other human dynamins.

Loop 1CS at the base of the Drp1 stalk mediates stable interaction with Mff

To test Mff interactions with the stalk loops identified by sequence alignments, the

ΔVDTSN and ΔVDKK mutants were initially examined. Both proteins exhibited a biphasic

melting curve comparable to the wildtype protein (Figure 4.5 A-B, purple and blue traces),

suggesting that the GTPase and stalk folds remained intact. The ΔVDKK exhibited an

interaction with Mff, because a stabilization of the stalk domain was observed with a

right-shift of the second melting event, similar to that seen with the wildtype protein

(Figure 4.5A – green trace). Conversely, the stabilization of the stalk by Mff was blunted

116 with the ΔVDTSN mutant, as the unfolding of the stalk was not stabilized upon the addition of Mff (Figure 4.5 1B – orange trace).

To further confirm the lack of interaction between the ∆VD stalk and Mff, low speed sedimentation was performed to measure the formation of large co-polymers when these proteins were mixed. Previous studies have shown that ΔVD is an assembly-primed dimer

(Clinton et al. 2016), and this mutant does not pellet when subjected to low speed centrifugation (Figure 4.4A). The addition of Mff induces the rapid assembly of ΔVD into large, filamentous polymers (Figure 4.4B) that are efficiently pelleted (Figure 4.4A).

Similarly, ΔVDTSN does not sediment alone; however, addition of Mff does not induce sedimentation of Drp1 (Figure 4.4A). In agreement with this finding, no filaments or large assemblies were observed by EM when ∆VDTSN and Mff were combined (Figure 4.4C). In contrast, the addition of Mff to ΔVDKK induced robust sedimentation of Drp1 (Figure 4.5C).

Negative stain EM analysis of ΔVDKK with Mff revealed filamentous polymers that are indistinguishable from the wildtype ΔVD (Figure 4.5E); thus Loop 3S was not required for

Mff interaction.

To determine whether the lack of sedimentation for ΔVDTSN was due to a defective

Drp1-Mff interaction or a defect in Drp1 oligomerization, a DSP-crosslinked Mff IP was used to stabilize and isolate ΔVD-Mff complexes. Relative to the wildtype ΔVD, the interaction of mutant ΔVDTSN with Mff was severely blunted (Figure 4.4D, DSP cross-linked

Mff co-IP of ∆VDTSN was roughly 40% as effective as ∆VDWT). Therefore, Drp1 interactions with Mff were disrupted when Loop 1CS was mutated to the homologous sequence found

117

in yeast Dnm1p. This destabilized interaction is sufficient to impede the formation of

larger Mff-induced Drp1 oligomers.

Previous in vitro studies from our group and others (Clinton et al. 2016; Osellame et

al. 2016; Macdonald et al. 2016; Clinton and Mears 2017) have demonstrated that Drp1

exhibits stimulated GTPase activity upon Mff-induced oligomerization. To further assess

the impact of disrupting Drp1-Mff interactions, Mff-stimulated GTPase activity of

wildtype ΔVD was compared to ΔVDTSN, and the latter was found to be insensitive to Mff addition (Figure 4.4E). Conversely, the addition of Mff to the ΔVDKK mutant exhibited

comparable stimulation in GTPase activity to that of the wildtype ΔVD (Figure 4.5D). This

further confirms the specific interaction between Drp1 L1CS and Mff. Finally, to ensure

that this mutant disrupted Mff interactions in the context of the full-length protein, the

stimulation of Drp1WT by Mff-decorated liposomes was assessed. When the TSN mutation

was introduced to the full-length construct, Drp1TSN, the Mff-stimulated GTPase activity

was lost (Figure 4.4F). Collectively, these findings demonstrate that the chimeric Drp1

L1CS mutant, Drp1TSN, disrupts direct Drp1-Mff interaction in vitro.

Chimeric Mutation of Drp1 L1CS retains intrinsic Drp1 assembly and enzymatic

properties

Based on the previous findings that Drp1 mutations can alter intrinsic Drp1 activities

(e.g. self-assembly and GTPase activity) and, thereby, indirectly impact its interaction with

Mff, we wanted to test whether Drp1TSN maintained fundamental Drp1 properties. To

118

begin, SEC-MALS was utilized to examine the multimeric states of Drp1 in solution. The

Drp1TSN (Figure 4.6A – red) mutant exhibited a distribution comparable to that observed

for Drp1WT previously (Figure 4.6A – black), so the assembly properties of Drp1 in solution

are largely unaffected. Moreover, higher-order polymerization of Drp1 can be induced through nucleotide interactions. In the presence of GMP-PCP, Drp1WT and Drp1TSN formed

‘spirals’ observed using negative stain EM (Figure 4.6 B and C, respectively). Using a

sedimentation assay, the quantified extent of protein self-assembly in the presence of

GMP-PCP was comparable in both wildtype and mutant proteins (Figure 4.6D). As a

control, the assembly-defective mutant, Drp1G363D, does not sediment in the presence of

nucleotide. Previous studies have indicated that Drp1 assembly in the presence of GMP-

PCP is altered in the presence of MiD49 (Kalia et al. 2017b; Koirala et al. 2013), and co-

assembly of these proteins results in a filamentous complex (Figure 4.6E). In the presence

of MiD49 and GMP-PCP, Drp1TSN forms a comparable filamentous complex (Figure 4.6F)

rather than the ‘spirals’ formed in the absence of MiD49. Therefore, the TSN mutation

selectively blocks Mff interaction without sacrificing interaction with other partner

proteins, and Loop 1CS is not required for MiD49 interaction.

In addition, the association of Drp1 with anionic lipids and subsequent GTPase activity stimulation (especially by cardiolipin) has been well-characterized in previous studies

(Bustillo-Zabalbeitia et al. 2014; Macdonald et al. 2014; Francy et al. 2017, 2015; Adachi

et al. 2016; Stepanyants et al. 2015). Using a FRET-based lipid association assay

(Macdonald et al. 2016) that measures FRET generated between a tryptophan in the VD

of Drp1 and fluorescently labeled lipids incorporated into a membrane template, both

119

Drp1WT and Drp1TSN associated to a similar extent with cardiolipin nanotubes as quantified

by the FRET efficiency (Figure 4.6G). Drp1 binding to CL-containing membranes promotes

a robust lipid-induced helical self-assembly (Francy et al. 2017), and cryo-EM analysis of

Drp1-bound nanotubes revealed similar polymers with both Drp1WT and Drp1TSN (Figure

4.6H, left and right panels, respectively). In the presence of lipid nanotube templates

containing DOPC(GC60/PE25/PC15), wildtype and TSN mutant proteins exhibited similar

non-stimulated GTPase activities (Figure 4.6I), since self-assembly does not occur under

these conditions (Francy et al. 2017). In the presence of nanotubes containing a

WT TSN cardiolipin (GC60/PE25/CL15), the GTPase activities for both Drp1 and Drp1 were stimulated to the same extent (Figure 4.6I), which is consistent with the comparable

helical assembly observed by cryo-EM. When taken together, these results exclude off-

target defects in Drp1TSN function that would alter Mff interaction. Therefore, we

concluded that the loss of Mff interaction and Mff-stimulated GTPase activity of Drp1TSN

are directly due to incompatible changes in the sequence of loop 1CS.

Unstable Drp1-Mff interaction precludes processive mitochondrial fission

Building off of this biochemical evidence, we examined the ability of Drp1TSN to

mediate mitochondrial fission and perform other Mff-dependent cellular processes. To

this end, previously generated Drp1-/- and Mff-/- MEFs were used to assess the function of

Drp1WT and Drp1TSN in a cellular context. Both wildtype and mutant Drp1 were transiently

expressed in MEFs lacking endogenous Drp1 to assess the capacity of both proteins to

120 facilitate mitochondrial fission (Figure 4.7A). While expression of Drp1WT induced excessive mitochondrial fission in these MEFs, the expression of Drp1TSN led to significantly fewer cells having fragmented mitochondria (Figure 4.7B).

Given that this mutant disrupted mitochondrial fission, the interaction between Drp1 and Mff was assessed to determine whether the interaction between these proteins was disrupted in a cellular context. To this end, HeLa cells which lack endogenous Drp1 (Drp1

-/- HeLa) were transduced with Myc-tagged Drp1WT or Drp1TSN. A crosslinking IP was used to isolate Myc-Drp1 and assess the co-immunoprecipitation of Mff. This method directly interrogates the interaction of Drp1 with Mff in cells rather than the overall recruitment of Drp1 to mitochondria. Additionally, this cellular context allows for the interaction of

Drp1 and Mff in the presence of additional factors, including cardiolipin and other protein factors that were not present in the in vitro conditions previously tested (i.e. Figure 4.4D).

Interestingly, Mff co-immunoprecipitation with Myc-Drp1TSN was diminished by ~60% when compared to Mff co-IP with Myc-Drp1WT in a cellular environment (Figure 4.7C), and these results are comparable to the IP efficiency in vitro (Figure 4.4D).

Finally, to exclude any Drp1TSN defects other than disruption of its interaction with

Mff, the fission activity of both Drp1WT and Drp1TSN was assessed in MEFs lacking endogenous Mff (Mff-/- MEF; Figure 4.7D). In these cells lacking Mff, Drp1TSN should function comparably to the WT protein. Mitochondrial fission was diminished in these cells due to the absence of Mff, but a similar level of fission was observed in both Drp1WT and Drp1TSN transfected cells (Figure 4.7E). Taken together, these data demonstrate that disruption of the Drp1-Mff complex, in cells expressing both Drp1 and Mff, is sufficient to

121 prevent formation of a functional Drp1-containing fission complex, and this inhibition impedes mitochondrial fission.

The VD of Drp1 occludes conserved residues throughout the stalk

The regulatory impact of the VD in Drp1 function has been established, but mechanisms regulating the interaction of Drp1 and Mff through Loop 1CS remained unclear. The VD may influence the self-assembly properties of Drp1 to control Mff interactions, but little is known about the structural or conformational dynamics of the

VD and how these features would contribute to its regulatory function. Based on secondary structure analyses, the VD of Drp1 is predicted to be intrinsically disordered, and this region is either removed (Fröhlich et al. 2013) or unresolved (Francy et al. 2017) in existing structures of Drp1 due to intrinsic flexibility. Interestingly, the VD was found to mediate interactions with partner proteins in yeast (Huyen T. Bui et al. 2012), and it has been assumed that Drp1 interactions with its partner proteins would occur through a similar mechanism (C. Guo et al. 2017; Lee et al. 2016). However, the interaction between

Drp1 and Mff in the mammalian mitochondrial fission complex was only observed when the VD was removed. Thus, the VD is dispensable for Drp1-Mff interaction (Clinton et al.

2016; Liu and Chan 2015). Still, the lack of structural information for the VD limits our understanding of the mechanisms regulating interactions with Mff. Based on the negative regulatory role observed for the VD, we hypothesized that it participates in intramolecular contacts that occlude the Mff interaction interface on Drp1 (L1CS).

122

To test this prediction, hydroxyl radical footprinting with mass spectrometry analysis

was used to characterize intramolecular VD interactions within Drp1 and how these

contacts influence Mff recruitment. Specifically, the relative oxidation of Drp1WT and the

Drp1 mutant lacking the VD was compared (∆VD; Figure 4.8B). In this way, enhanced sequence oxidation in the ∆VD mutant relative to Drp1WT was attributed to VD-occluded

regions. This analysis revealed a surface of significantly enhanced oxidation in the absence

of the VD (Drp1WT/∆VD < 0.5) spanning from the base of the stalk to a flexible loop at the

top of the stalk (Loop 1NS). Subsequent assessment of the exposed regions at a single

amino acid level revealed several residues (e.g. R376, V381, L384, R423, and R430) whose

oxidation was enhanced in the absence of the VD (Figure 4.8C).

To ensure that these residues were specifically shielded by the VD, and were not due

to the multimeric changes for the ΔVD mutant (i.e. dimer vs. tetramer interactions)

(Clinton et al. 2016), a comparison was made to Drp1G363D, which is self-assembly limited

to a dimer species like ∆VD (Figure 4.9). Interestingly, several residues (e.g. V381 and

L384) display enhanced oxidation specifically in the absence of the VD, so these regions are selectively occluded in the full-length protein. Additionally, there are several residues

(e.g. R376, R423, and R430) that exhibit enhanced oxidation in both mutant proteins.

These residues represent a potential interface responsible for the formation of Drp1 multimers in solution, since tetramers and larger multimers are reduced or ablated as a consequence of these mutations (∆VD and G363D).

Interestingly, a region (residues 476-497) on the opposing side of the four-helix

bundle spanning the length of the stalk exhibited enhanced protection in the absence of

123

the VD (blue region, Figure 4.8B), which was not observed at the peptide or single amino acid level in Drp1G363D (Figure 4.9). Oxidation of M482 is enhanced in Drp1WT compared

to ∆VD even though this residue is positioned in the middle of a conserved interface in

the crystal structure (Fröhlich et al. 2013). Therefore, the VD may regulate the quaternary

structure of Drp1 dimers through intramolecular stalk interactions.

Simultaneously, L1CS at the base of the stalk region is selectively occluded by the VD based on the hydroxyl radical footprinting (Figure 4.8C). In fact, leucine 384, which we

identified and mutated to ablate Mff interactions (Figures 5.3, 5.4 and 5.7), and

neighboring residues, V381 and D382, are selectively occluded when the VD is intact.

These findings support a mechanism whereby the VD interferes with Drp1-Mff co-

assembly by masking the key site of interaction, L1CS.

4.4 DISCUSSION

The overarching goal of this study was to investigate the functional interaction

between Drp1 and Mff and to define how this interaction is regulated by the VD.

Footprinting studies provided a novel means to examine the conformational dynamics of

the VD and identify intramolecular interactions that have remained a mystery due to the

intrinsic flexibility of this region. We found that the VD docks back and occludes a region

with the Drp1 stalk conserved in higher eukaryotes. Specifically, loop 1CS contains a

unique sequence that is crucial for interaction with Mff. Therefore, Mff-induced assembly

of Drp1 is opposed by intramolecular masking of the stalk by the VD (Figure 4.8D).

124

In addition, a separate interface was exposed when Drp1 mutations were introduced

that limit (Figure 4.8C – ΔVD) or ablate (Figure 4.9B - Drp1G363D) self-assembly in solution.

Under these conditions where dimers are exclusively observed, the region identified as

Interface 4 (Figure 4.8A) in the crystal structure of Drp1 (Fröhlich et al. 2013) showed

increased oxidation at residues R423, E426, and R430. These results suggest a potential

role for interface 4 as a relevant self-assembly motif for solution multimers. Interestingly,

point mutations within this interface blunted mitochondrial fission, even though purified

protein behaves comparably to the wildtype (Fröhlich et al. 2013). Thus, mutations

throughout the stalk have the potential to elicit assembly defects due to direct disruption

of self-assembly motifs or indirectly through altered intramolecular, regulatory

interactions.

Importantly, the bulk of our structural insight for Drp1 comes from the crystal

structure that was solved without the VD present, and a key intermolecular contact within

the core dimer is mediated by interface 2 (Figure 4.8A). The relative oxidation of M482,

which is positioned in the center of interface 2, is increased in Drp1WT compared to ∆VD.

This suggests that the relative orientation of stalks within the Drp1 dimer differs in the presence of the VD. In this way, the VD may regulate the relative orientation of opposing

Drp1 molecules in the dimer unit, which could impact the self-assembly properties

through its inter- and intramolecular contacts.

The footprinting data also lends functional insight into the assembly defects of

previously identified Drp1 stalk mutations. The first identified human Drp1 mutation,

Drp1A395D (Waterham et al. 2007), is immediately adjacent to the VD-occluded region of

125

the stalk. This indicates that the hypomorphic phenotype of this mutant is due to changes

in the Drp1 stalk structure or to disruption of VD-stalk contacts. In agreement with our

findings, this mutant also appears to be deficient in Mff interaction in vitro (Otera et al.

2010). Additionally, mutation of R376, which is immediately adjacent to critical residues

S in L1C , has previously been shown to disrupt Drp1-Mff interactions by evoking a

hyperoligomeric self-assembly tendency. In fact, dimeric Drp1 species were eliminated

and Drp1-Mff interactions were indirectly inhibited (Clinton et al. 2016). Because R376

lies adjacent to a VD-occluded region of Drp1, its mutation to a glutamic acid could

stabilize the VD in a conformation that favors larger solution multimers. Overall, stalk

mutations have been shown to perturb core Drp1 properties, like oligomerization and

GTPase activity (Chang et al. 2010). However, these alterations likely result in a secondary

defect in partner protein interactions that have largely been uncharacterized for Drp1

patient mutations.

Based on this new insight into the conformational dynamics of the VD, we sought to

address an apparent ambiguity in the field: does Mff interact with the stalk or VD of Drp1?

Several studies (Huyen T. Bui et al. 2012; C. Guo et al. 2017; Lee et al. 2016) have

suggested that the VD is required for Mff interactions, while others have suggested that

the stalk is the interaction site (Strack and Cribbs 2012). In fact, we have previously

published studies unambiguously demonstrating that the VD is dispensable for Mff

interactions (Clinton et al. 2016), and so we sought to identify the key Drp1 residues

responsible for Mff binding. Using sequence alignments, we found well-conserved

metazoan-specific regions of Drp1. A chimeric mutation replacing three residues in loop

126

1CS with corresponding sequence from yeast Dnm1p disrupts Mff binding, which

demonstrates that the base of the Drp1 stalk is the site of Mff interaction. Importantly,

this effect was achieved without introducing an off-target defect in other core Drp1

properties. These findings underscore differences that exist between the mammalian and

yeast mitochondrial fission machineries. In yeast, Dnm1p is recruited to the mitochondria

via interactions between the ‘Insert B’ domain (comparable to the Drp1 VD) and a

cytosolic protein, Mdv124. In higher eukaryotes, Mdv1 is not conserved, so the VD cannot participate in an analogous connection. Rather, residues at the base of the stalk directly form partner protein interactions, and the accessibility of region is regulated by the VD.

Finally, a minimal complex of Drp1 and liposome-tethered Mff was found to be sufficient to tubulate and constrict lipid tubules in vitro. Rather than serving as simply a recruitment signal for Drp1, Mff is able to participate in the membrane remodeling process as a part of a co-polymer. These results also allow for a new understanding of the functional and regulatory consequences of Drp1-Mff interaction. While Drp1 is recruited to membranes by Mff, the mechanoenzymatic core is displaced from the lipid interface.

This placement may restrict the physical interaction between Drp1 and the lipid bilayer, but it importantly allows space for other membrane-associated factors to access Drp1.

Specifically, other partner proteins or enzymes that introduce post-translational modifications could have access to directly regulate Drp1 activity. In addition, this displacement of Drp1 from the membrane promotes the assembly of a more efficient fission machinery. This is best highlighted when comparing previously characterized Drp1 assemblies on lipid nanotubes (Francy et al. 2017) (Figure 4.2B - left) with the local

127

constrictions observed when Mff is tethered to the flexible liposomes in this study (Figure

4.1E-F). While both Drp1 assemblies have similar average outer diameters (50 nm on

nanotubes, and 60-70 nm on constricted liposomes), the underlying lipid morphology is much more narrow when Mff is scaffolded beneath Drp1 (30 nm for nanotubes and < 20 nm at local constrictions). Taken together, this indicates that Drp1 constriction is augmented when scaffolded on Mff, which promotes further membrane remodeling to comparatively narrower geometries. Therefore, the cytosolic domain of Mff is sufficient to recruit Drp1 and augment its membrane remodeling capabilities. Collectively, this study offers a new role for Mff in mitochondrial fission whereby Drp1 polymers are dynamically regulated and impart a more efficient contractile force through an Mff- scaffolded assembly.

4.5 MATERIALS AND METHODS

Plasmids and mutagenesis

Bacterial expression vectors for Drp1WT (Drp1 isoform 3; Uniprot ID O00429) and ΔVD

(Drp1 Isoform 1 lacking amino acids 517-639) in the pCal-N-EK backbone were used as

previously described (Clinton et al. 2016). Mff lacking its transmembrane domain (Mff;

Uniprot ID Q9GZY8-5, residues 1-218) was expressed in a pET28a vector as previously

described (Clinton et al. 2016). Mutagenesis of these plasmids was performed using

Quikchange Lightning (Agilent). Transient overexpression of Drp1 in mammalian cells was

128

achieved using a pCMV vector encoding Myc-Drp1 as previously described (Clinton et al.

2016). Lentiviral expression vectors were prepared by subcloning the Myc-Drp1WT and

Myc-Drp1TSN sequences into a pDONR221 gateway vector, and subsequent gateway

cloning into pLX304.

Protein Expression and Purification

Drp1WT (isoform 3 – Uniprot accession number O00429), Drp1ΔVD, and Mff were

expressed and purified as previously described (Clinton et al. 2016). Briefly, Drp1

constructs were overexpressed in BL21-(DE3) Star E. coli, purified by calmodulin affinity

purification, the CBP-tag was removed with HVR3C protease, and size-exclusion

chromatography was used to achieve full sample purity. Mff constructs were

overexpressed in BL21-(DE3) E. coli, purified by NTA(Ni2+) agarose affinity purification,

cleaned up with ion exchange chromatography, and purified to homogeneity with size-

exclusion chromatography. The Drp1 GG construct and ΔN50-MiD49 were kindly provided

by the Ramachandran lab, and were expressed and purified using standard protocols

(Wenger et al. 2013; Osellame et al. 2016).

Liposome and lipid nanotube preparation

All lipids were obtained from Avanti Polar Lipids (Alabaster, AL) and were referred to

throughout this study as PC (DOPC), PE (DOPE), CL (heart cardiolipin), Ni2+ (DGS-

NTA(Ni2+)), and GC (Galactosyl(β) Ceramide (d18:1/24:1(15Z)). Lipids dissolved in

chloroform were combined, dried under a stream of nitrogen gas, and further dried under

vacuum overnight at room temperature. Lipid films were rehydrated in HK buffer (25 mM

129

HEPES (KOH), 150 mM KCL) for 30 minutes at 37°C with occasional agitation. Lipid

nanotubes were sonicated for 1 minute in a bath sonicator. Liposomes were freeze-

thawed 4x by alternate immersion in liquid nitrogen and hot water bath. The freeze-

thawed liposomes were extruded 21x through a 1.0 µm polycarbonate filter, and stored

at 4°C for up to 3 days until use.

GTPase Activity Assay

The GTPase activity of Drp1 was assessed using a well-established colorimetric assay as previously described (Leonard et al. 2005). When scaffolding liposomes (PC86.8/Ni3.3/CL9.9)

were utilized (150 µM final), Mff (5 µM final) was preincubated for 15 minutes with liposomes to allow for scaffolding. Drp1 (500 nM final) was preincubated for 15 minutes at room temperature with any additional assay components (i.e. Mff, scaffold liposomes, nanotubes, etc.). Reactions were initiated by adding GTP/Mg2+ (1 mM and 2 mM final

respectively), and were carried out at 37°C. Aliquots of each reaction were halted by

adding EDTA (100 mM final) at various timepoints, and inorganic phosphate liberation

was assessed using malachite green reagent.

Low Velocity Sedimentation

For ΔVD-Mff filament sedimentation, ΔVD (2 µM final) and Mff (10 µM final) were

incubated for 1 hour at RT prior to centrifugation for 30 minutes at 16,100 xg at 4°C. The

supernatant and pellet were isolated, and separated by SDS-PAGE. The gel was stained

with InstantBlue (Expedeon), and densitometry was analyzed using GelAnalyzer.

Nucleotide-stimulated oligomerization was assessed by incubating Drp1 (5 µM final) in

130

the presence of GMP-PCP and MgCl2 (1 mM and 2 mM final respectively) for 30 minutes.

Samples were centrifuged at 4°C and analyzed as described above.

Dynamic Scanning Fluorimetry

The thermal stability of Drp1 in the presence of nucleotides or Mff was assessed using a

SYPRO orange based protein unfolding assay. Briefly, Drp1 (2.5 µM final) was incubated

with guanosine nucleotides (GDP or GMP-PCP, 1 mM final) in HKMB (Same as HKB with 2

mM MgCl2) with SYPRO orange (10x final) for 15 minutes. After this pre-incubation,

samples were placed in a StepOne Plus thermocycler (Applied Biosciences), incubated 2

minutes at 25°C, and temperature was increased at a rate of 1°C per minute from 26-99°C

while monitoring SYPRO orange fluorescence. Data were presented with normalized

SYPRO orange fluorescence as a function of sample temperature.

Dynamic Light Scattering (DLS)

The hydrodynamic radius of Mff was determined using dynamic light scatter using scatter measurements from a Dynapro Nanostar (Wyatt Technologies) instrument. Briefly, 50 µL of MffΔCC-TM-His6 at 100 µM was analyzed in an Eppendorf UVette at RT.

Autocorrelation curves from a set of 10 acquisitions (10s integration time for each) were

analyzed using Dynamics v7.1.3 software (Wyatt Technologies) to resolve the average

hydrodynamic radius of the sample.

131

Drp1-Lipid FRET

Interaction of Drp1 with lipid nanotubes was investigated by taking advantage of a

tryptophan residue located in the membrane-facing variable domain of Drp1 achieve

FRET with Dansyl-DOPE incorporated into a lipid nanotube as previously described

(Macdonald et al. 2014, 2016). Briefly, GC/PE/Dansyl/CL (100 µM total lipid) was incubated with Drp1 (1 µM) for 30 min @ RT. Emission spectra (310-550 nm) were

obtained at 25°C using a Tecan Infinite M1000 PRO microplate reader using an excitation

wavelength of 295 nm. All spectra were background corrected and corrected for the

direct excitation of Dansyl by the excitation at 295 nm. FRET Efficiency (E) was calculated

using the formula = 1 where FDA is the emission intensity of Tryptophan 𝐹𝐹𝐷𝐷𝐴𝐴 𝐸𝐸 − � � 𝐷𝐷� measured in the presence of dansyl𝐹𝐹 -PE, and FD is the same emission in the absence of dansyl-PE.

Electron Microscopy

Samples for electron microscopy imaging were prepared as follows. For liposome constriction studies, Mff (40 µM for cryo, 10 µM for negative stain) was incubated for 15 minutes at room temperature with flexible liposomes (PC55/PE35/Ni10, 0.4 mM final) in

HKMB, and then incubated with Drp1 (4 µM final) for at least 1 hour at room temperature.

Hydrolysis was initiated by addition of GTP (1 mM final) prior to plunge freezing or

negative staining grids. Drp1 (2 µM final) was assembled with partner proteins (Mff or

MiD49, 10 µM final), lipids (PC40/PE35/CL25, 100 µM final), or nucleotides (MgCl2/GMP-

PCP, 2mM and 1 mM final respectively) for at least 1 hour at room temperature in HKB.

132

For cryo-EM analysis of Drp1 on cardiolipin nanotubes, Drp1 (0.4 µM final) was incubated with nanotubes (GC40/PE35/CL25, 300 µM final) for at least 15 minutes at room temperature in HKMB prior to stabilization with GMP-PCP (1 mM final) for at least 15 minutes at room temperature prior to plunge freezing. Finally, Mff-scaffolded lipid nanotubes were prepared by incubating Mff (25 µM final) with lipid scaffold nanotubes

(GC50/PE25/CL15/Ni10, 250 µM final) for at least 15 minutes at room temperature. Drp1 was added to 5 µM final, incubated at least 1 hour at room temperature, and then stabilized by addition of MgCl2/GMP-PCP (2 mM and 1 mM final respectively) for at least one hour prior to plunge freezing.

For all negative stain electron microscopy, samples were prepared as indicated, incubated on carbon-coated EM grids, and stained with 2% uranyl acetate. Imaging was performed on a TF20 (FEI Co.) electron microscope at 200 keV, and images were acquired using a Tietz 4k x 4k CCD camera at 29,000x magnification. For cryo-electron microscopy, samples were prepared as incubated, briefly incubated on holey carbon Quantifoil EM grids (R3.5/1 or R2/1), and plunge frozen in liquid ethane. Prepared grids were maintained in liquid nitrogen until imaging. Images were acquired using a TF20 and Tietz 4k x 4k CCD camera at 29,000x magnification. Class averages were generated using the 2D classification function of Relion 2.1.

Size-exclusion chromatography with multi-angle light scattering

The oligomeric tendencies of Drp1WT and Drp1TSN were investigated using SEC-MALS as previously described (Macdonald et al. 2014; Clinton et al. 2016; Macdonald et al. 2016).

133

Briefly, Drp1 (loaded in a total volume of 0.5 mL at 10 µM) was fractionated on a

superpose 6 10/300 GL column in HK buffer (20 mM HEPES(KOH) pH 7.5, 150 mM KCl) +

1 mM DTT. Flowthrough from the column was analyzed by tandem miniDAWN TREOS

MALS and Optilab rEX differential refractive index detectors (Wyatt Technologies). Molar

mass was determined using the ASTRA 6.1 software package (Wyatt technologies) and

was plotted with molar mass (right axis) and normalized refractive index (left axis) as a

function of elution volume.

Immunocytochemistry

All mouse embryonic fibroblasts (MEFs) were cultured in DMEM + 10% Heat-Inactivated-

FBS and 1% penicillin/streptomycin. All cells were maintained at 37°C with 5% CO2/95% air. MEFs were transfected with 1 µg of plasmid DNA encoding Myc-Drp1WT or Myc-

Drp1TSN using TransIT-2020 transfection reagent (Mirus Bio, Madison, WI) according to

the manufacturer’s protocol. Transfected cells were cultured on glass coverslips and were

washed with cold PBS, fixed with PBS + 4% formaldehyde, and permeabilized with PBS +

0.1% Triton X-100. Subsequently, the cells were blocked with 2% normal goat serum, and incubated overnight with rabbit anti-Tom20 and mouse anti-Myc (each antibody at 1:500,

Santa Cruz Biotechnology, Santa Cruz, CA). Cells were washed with PBS and incubated with Alexa Fluor 488-conjugated goat anti-mouse and Alexa Fluor 568-conjugated goat

anti-rabbit (1:500; Invitrogen) for 60 min at room temperature. Coverslips were mounted

on glass slides and imaged by confocal fluorescence microscopy using an Olympus FV1000

IX81 confocal microscope (Olympus USA). Mitochondrial fragmentation was examined in

134 these cells, and the percentage of Myc-Drp1 expressing cells with fragmented mitochondria relative to the total Myc-Drp1 expressing cells was calculated.

Crosslinking Immunoprecipitation

HeLa were cultured in DMEM + 10% FBS with 1% penicillin/streptomycin. Drp1 -/- HeLa were transduced with lentivirus carrying pLX304-Myc-Drp1, selected for two weeks with

8 µg/mL blasticidin and subsequently kept under selection by maintenance blasticidin concentration for the remainder of their time in culture (DMEM + 10% FBS + 1%

Penicillin/streptomycin + 4 µg/mL blasticidin). Transduced HeLa cells were crosslinked for

10 min @ RT with 1 mM DSP in PBS. Crosslinking was quenched with 100mM Tris pH 7.5 in PBS for 20 minutes @ RT. Cells were washed with ice cold PBS, scraped cells into PBS +

1% Triton X-100 + protease inhibitor cocktail. Cells were lysed by sonication on ice 3 times

(1s on at 30% amplitude, 1s rest, repeat for 1 minute), and protein content was quantified with BCA assay (Thermo Fisher Scientific). Lysate (0.5 mg of protein) was incubated overnight @ 4°C with 2 µg of normal IgG or anti-Myc (Santa Cruz Biotechnology).

Immunocomplexes were isolated with Protein A/G agarose (Santa Cruz Biotechnology), washed with PBS + 1% Triton X-100 and PBS, and eluted by boiling beads in 2x Laemmli sample buffer. Immunoprecipitates were separated by SDS-PAGE, transferred onto PVDF membranes, and probed with anti-Myc or anti-Mff antibodies (Santa Cruz Biotechnology,

1:1000 for both), followed by visualization using chemiluminescence. Densitometry analysis was performed using Image Lab (BioRad) and relative densities (Mff/Myc) were assessed. Statistical significance was assessed using an unpaired t-test (GraphPad Prism).

135

Hydroxyl radical footprinting

Drp1WT, Drp1ΔVD and Drp1G363D protein samples were exchanged into PBS + DTT (10 mM

Phosphate, 138 mM NaCl, 2.7 mM KCl, 0.01 mM DTT) using Zeba spin desalting columns

(Thermo Fisher Scientific). The protein concentration for all samples was adjusted to 1

µM. Samples were exposed to X-ray white beam for 0-225 microseconds at the 17-BM beamline of the National Synchrotron Light Source II (NSLS-II), Brookhaven National

Laboratory (BNL), and immediately quenched with 10 mM methionine amide to prevent secondary oxidation. Beam parameters were optimized by using an Alexa-488 fluorophore assay as previously described(Klinger et al. 2014). After irradiation, all samples were reduced with 10 mM DTT at 56 oC for 45 minutes and alkylated with 25 mM iodoacetamide at room temperature in the dark for 45 minutes. Subsequently, protein samples were digested with a modified trypsin (Promega) at 37 oC for overnight at an enzyme to protein ratio of 1:10 (w/w). To identify the sites of radiolytic modification and quantify the extent of oxidation on the peptide and single residue levels, the LC-MS analysis of digested samples were carried out on an Orbitrap Elite mass spectrometer

(Thermo Fisher Scientific, CA) interfaced with a Waters nanoAcquity UPLC system

(Waters, MA). A total of 3 pmol of proteolytic peptides were loaded on a trap column (180

μm × 20 mm packed with C18 Symmetry, 5 μm, 100 Å (Waters, MA)) to wash away salts and concentrate peptides. The peptide mixture then was separated on a reverse phase column (75 μm x 250 mm column, packed with C18 BEH130, 1.7 μm, 130 Å (Waters, MA)) using a gradient of 2 to 46% mobile phase B (0.1% formic acid and acetonitrile (ACN)) vs. mobile phase A (100% water/0.1 % formic acid) over a period of 62 minutes at 40 °C with

136

a flow rate of 300 nl/min. Peptides eluting from reverse phase column were introduced

into the nano-electrospray source at a capillary voltage of 2.5 kV. All MS data were

acquired in the positive ion mode. For MS1 analysis, a full scan was recorded for peptides

with m/z ranging from 350 to 1600 and resolution of 120,000 in the Fourier transform

mass analyzer followed by tandem MS of the 20 most intense peptide ions scanned in the

ion trap mass analyzer. To determine the extent of oxidation for each peptide, the

selected ion chromatograms (SIC) were extracted and integrated for the unmodified and

all modified forms of peptide ion. These peak area values were used to characterize

reaction kinetics in the form of dose response (DR) plots, which measure the loss of

unoxidized peptide as a function of the hydroxyl radical exposure (J. G. Kiselar et al. 2002;

Janna G. Kiselar et al. 2003; Takamoto and Chance 2006; Janna G. Kiselar et al. 2011). The

extent of oxidation for each specific residue was calculated as previously described

(Klinger et al. 2014). The first order oxidation rate constant (K) was derived for each

peptide and specific residue from a corresponding DR plots as previously described

(Klinger et al. 2014; J. G. Kiselar et al. 2002; Janna G. Kiselar et al. 2003; Takamoto and

Chance 2006; Janna G. Kiselar et al. 2011). The ‘relative oxidation of peptides’ values were

determined as a ratio of oxidation rate constants for each peptide segment within Drp1

) (KWT/KΔVD and KWT/KG363D , while ‘relative oxidation of amino acids’ were determined as a

ratio of oxidation rate constants for each specific residue. Ratio values for

Drp1WT/Drp1ΔVD and Drp1WT/Drp1G363D were normalized by the average of the mean and median value (0.655 and 1.755 respectively) to correct for the oxidation rate constant variations between dimer and tetramer Drp1 forms. To identify specific sites of

137

modification, the resulting tandem MS spectra were searched against a Drp1WT, Drp1ΔVD

and Drp1G363D protein database using the software MassMatrix (H. Xu and Freitas 2007)

with mass accuracy values of 10 ppm and 0.8 Daltons for MS1 and tandem MS scans,

respectively, and allowed variable modifications including carbamidomethylation for

cysteines and all known oxidative modifications previously documented for amino acid

side-chains (G. Xu and Chance 2007). The peptide segments and the amino acid residues

in each segment for which oxidation rate constants were determined are provided in

Tables 5.1 and 5.2 respectively.

4.6 ACKNOWLEDGEMENTS

We would like to acknowledge Richard Youle for providing both wildtype and Drp1 knockout HeLa and HCT116 cells. We would also like to acknowledge Henrietta Lacks and her family for the contribution of her cells to our study. We also thank the labs of Analisa

Difeo and Goutham Narla for providing plasmids and reagents for the generation of the lentiviral vectors used in this study. This research was supported by grants from the

American Heart Association and the National Institutes of Health (JAM: AHA

16GRNT30950012, NIH R01 CA208516-01A1 and NIH R01 GM125844-01; RR: NIH R01

GM121583; XQ: NIH R01 NS0881920). Jennifer Bohon assisted with sample irradiation at

the National Synchrotron Light Source (NSLS) of Brookhaven National Laboratory (Upton,

NY) which is supported by NIH grant P30EB009998. These footprinting experiments were

supported pilot by funding from the Clinical & Translational Science Collaborative of

138

Cleveland, 4UL1TR000439 from the National Center for Advancing Translational Sciences

(NCATS) component of the National Institutes of Health and NIH roadmap for Medical

Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

139

FIGURE 4.1

140

FIGURE4.1 Drp1 polymers constrict Mff-decorated liposomes upon addition of GTP. A.

Schematic representation of the lipid-tethered Mff liposomes is presented. When incubated with Drp1 (green), these templates deform into tubules (middle). Addition of

GTP results in narrow constriction of the underlying lipid bilayer (bottom). B. Negative stain EM micrographs of Drp1 tubules formed on Mff-tethered liposomes in the absence of GTP (left) and 30 s after addition of 1 mM GTP (right). Scale bar = 100 nm. C. Measured diameters of Drp1-Mff tubules (as seen in Figure 4.1B) in the absence of GTP (black; n=89) and 30 s after addition of 1 mM GTP (red; n=121). ***: p < 0.0001 based on student’s t- test. D. Cryo electron micrographs of Drp1 tubules formed on Mff-tethered liposomes in the absence of GTP (left) and 5 s after addition of 1 mM GTP (right). Insets highlight a tube in the absence of GTP (gray border), a non-constricted region of the tube in the presence of GTP (green border), or locally constricted tubules (red border). *: Undeformed liposome. Scale bar: 100 nm, inset scale bars: 25 nm. E-F. The diameters of Drp1-Mff complexes on scaffold lipids (as observed in Figure 4.1D) were measured in the absence of nucleotide (Apo) and after GTP was added. The outermost protein diameter (E), or the central lipid diameter (F) highlight the constriction of the underlying lipid template. Two distinct diameter classes were observed after GTP was added: a non-constricted regions and adjacent regions of local constriction. ***: P < 0.0001 based on 1-way ANOVA with multiple comparisons.

141

FIGURE 4.2

142

FIGURE 4.2 Mff builds Drp1 polymers on membranes via a stalk interface. A. The domain architecture (top) and the 3D crystal structure (below, PDB ID 4BEJ) of Drp1 highlights distinct domains (GTPase domain (green), bundled signaling element (purple), middle domain (dark blue), variable domain (orange line), and GTPase effector domain (light

blue). B. Representative cryo electron micrographs of GMP-PCP stabilized Drp1

assembled on a CL nanotube (left, GC40/PE35/CL25) and Mff-scaffold nanotubes (right,

GC50/PE25/Ni10/CL15) are shown. Line plot representations of each decorated nanotube is

provided (below) to highlight changes in protein distribution, and the distance from the

edge of the lipid to the outer protein density is indicated. Scale bar = 50 nm. C. A 2D class average of Drp1 assembled on Mff-tethered nanotubes (as in Figure 4.2B) was generated

using Relion 2.1. Densities are assigned to Drp1 (green), Mff (orange), and lipid nanotube

(grey). Scale bar = 50 nm. D. Dynamic light scattering data from MffΔCC-TM (100 µM)

indicated a hydrodynamic radius of 5.4 nm for monomeric Mff. E-F. Dynamic scanning

fluorimetry (DSF) analysis of Drp1 (2.5 µM) is presented for the Drp1 GG-construct

(dashed lines) and full-length Drp1 (solid lines) in the absence of nucleotide (black) and

the presence of GDP (green) and GMP-PCP (red). E. Similar DSF analysis is presented for

Drp1ΔVD alone (black) and in the presence of MffΔTM (red).

143

FIGURE 4.3

FIGURE 4.3 Sequence conservation analysis highlights Drp1 stalk loops as potential Mff- interaction sites. A. Cartoon representation highlighting differences in the proteins responsible for mitochondrial fission in chordates (left) and in yeast (right). B-D. Amino acid sequence alignments generated using T-Coffee (Notredame, Higgins, and Heringa

2000) identified Loop 1CS (left) and Loop 3S (right) as regions of interest based on lack of

conservation when comparing human Drp1 and yeast Dnm1p (B), conservation when

comparing metazoan Drp1 homolgoues (C), and lack of conservations when comparing

144 soluble human dynamin-family proteins (D). E. Structural alignments of the crystal structures of Drp1 (blue ribbons), MxA (red ribbons), and Dyn3 (green ribbons) highlight structural differences in Loop 1CS and Loop 3S (left and right, respectively).

145

FIGURE 4.4

FIGURE 4.4 Mutation of stalk loop 1CS disrupts Drp1-Mff interaction in vitro. A.

Quantification of low velocity sedimentation (top) is shown along with representative images of the supernatant and pellet fractions for each sample (bottom). Bars indicate mean ± SD. ***: p < 0.0005 based on student’s t-test, n > 3. B-C. Negative stain EM

146

micrographs depicting 2 µM ΔVD (B) or ΔVDTSN (C) in the presence of Mff. D. A

representative western blot of in vitro DSP-crosslinked ∆VD and mutant ΔVDTSN

coimmunoprecipitated with Mff (left) along with quantification of the relative intensities of Drp1/Mff (right). ‘<’ indicates IgG light chain. Bars indicate mean ± SD, n=3, *: p < 0.05.

E-F. Mff-induced stimulation of Drp1 GTPase activity was measured. ∆VDWT and ∆VDTSN

(0.5 µM) were incubated with Mff in solution (E), while Drp1WT and Drp1TSN were assayed

with Mff-tethered SL/CL (PC86.8/Ni3.3/CL9.9) (F). ***: p < 0.001, n > 3.

147

FIGURE 4.5

FIGURE 4.5 Mutation of L1CS, not L3S disrupts Drp1-Mff interactions. A-B. Dynamic

scanning fluorimetry was performed with 2.5 µM ΔVDTSN (A) and ΔVDKK (B) in the absence

(purple and green lines) and presence (orange and blue lines) of Mff (as in Figure 4.2). C-

E. Similar to Figure 4.3, ΔVDKK interactions with Mff was assessed using low-velocity

148

sedimentation (C), Mff-stimulation of ΔVD GTPase activity (D), and a negative stain EM, which highlights filaments formed by ΔVDKK in the presence of Mff (E). ***: p < 0.0001

based on student’s t-test, n > 3. Scale bar = 100 nm.

149

FIGURE 4.6

FIGURE 4.6 Drp1TSN functions comparably to Drp1WT. A. SEC-MALS analysis assessed the oligomeric tendencies of Drp1WT (black) and Drp1TSN (red). Protein abundance was

150

measured on the left axis as normalized dRI while molar mass is displayed as Drp1

oligomeric state on the right axis. B-C. Negative stain EM identified ‘spirals’ formed by

Drp1WT (B) and Drp1 TSN (C) in the presence of 1 mM GMP-PCP. Scale bar = 100 nm. D. Low velocity sedimentation examined self-assembly of Drp1WT, Drp1G363D, and Drp1TSN in the

absence (white bars) and presence of 1 mM GMP-PCP (black bars). ***: p < 0.001 based

on student’s t-test, n = 3. E-F. Drp1WT (E) or Drp1TSN (F) filaments were observed by

negative stain EM in the presence of 1 mM GMP-PCP and MiD49ΔN50, scale bar = 200

nm. G. Drp1 affinity for cardiolipin nanotubes (GC60/PE15/Dansyl-PE10/CL15) was assessed

by measuring the FRET between a tryptophan in Drp1 and dansyl-PE. H. Representative

cryo electron micrographs are shown for GMP-PCP stabilized Drp1 assemblies on

WT TSN cardiolipin nanotubes (GC40/PE35/CL25) with Drp1 (left) and Drp1 (right). Scale bar =

25 nm. I. Stimulated GTPase activity of Drp1WT and Drp1TSN was measured using

nanotubes containing phosphatidylcholine (GC60/PE25/PC15; white bars) and cardiolipin

(GC60/PE25/CL15; black bars). ***: p < 0.0001 based on student’s t-test, n=3. J. Negative

stain EM compared Drp1WT or Drp1TSN (left and right, respectively) tubulation of CL- containing liposomes (PC40/PE35/CL25).

151

FIGURE 4.7

FIGURE 4.7 Drp1TSN is deficient in Mff binding and mitochondrial fission in cells. A-B.

Immunofluorescence images of Drp1-/- MEFs transiently transfected with Myc-Drp1WT or

Myc-Drp1TSN (A) are shown, and mitochondrial morphology was quantified (B). C. Co-

immunoprecipitation of Mff with Myc-Drp1 from Drp1-/- HeLa stably expressing Myc-Drp1

or Myc-Drp1TSN was performed and quantified as the relative density of Mff/Myc. **: p <

152

0.01, n = 3. D-E. Immunofluorescence images of Mff-/- MEFs transiently transfected with

Myc-Drp1WT or Myc-Drp1TSN (D) are shown, and mitochondrial morphology was quantified

(E). *: p = 0.01 based on Student’s t-test.

153

FIGURE 4.8

FIGURE 4.8 Hydroxyl radical footprinting reveals a VD occluded surface on the stalk of

Drp1. A. Schematic representation of a Drp1 dimer highlighting known protein-protein

interactions and structural domains is presented (colored as in Figure 4.2A). B. A surface

representation of the Drp1 dimer highlights the relative oxidation of peptides by hydroxyl

radicals (Drp1WT/ ΔVD) ranging from more exposed in ΔVD (red) to more oxidized in

Drp1WT(blue). C. Amino acid side chains in Loop L1CS were exposed when the VD was deleted. Specific residues are highlighted with a ribbon representation and colored to represent relative oxidation similar to Figure 6B. D. A schematic representation of Drp1

154 highlight the proposed mechanism of VD occlusion of loop L1CS, which interferes with

Drp1-Mff interaction.

155

FIGURE 4.9

FIGURE 4.9 Loop 1CS oxidation is comparable between Drp1WT and Drp1G636D, an

assembly-defective dimer mutant. A. A surface representation of the Drp1 dimer is

colored according to the relative hydroxyl radical oxidation of peptides (Drp1WT/

Drp1G363D) ranging from more exposed in Drp1G363D (red) to more exposed in

Drp1WT(blue). B. A ribbon representation of Loop L1CS in the stalk of Drp1 highlights few

differences in the oxidation of amino acid side chains, which confirms that Drp1 species

are largely dimeric for these experiments (colored as in Figure 6C).

156

TABLE 4.1

Drp1WT Peptide Sequence Drp1WT K Drp1ΔVD Drp1ΔVD K Drp1G363D Drp1G363D K Ratio Normalized Ratio Ratio Normalized Ratio peptide (/s) peptide (/s) peptide (/s) (Drp1WT/Drp1ΔVD) (Drp1WT/Drp1ΔVD)/0.655 (Drp1WT/Drp1G363D) (Drp1WT/Drp1G363D)/1.755 (mean+medium)/2=0.655 (mean+medium)/2=1.755 11-38 LQDVFNTVGADIIQLPQIVVVGTQSSGK 336.78±19.97 11-38 587.67±16.68 11-38 165.03±15.23 0.62 1 2.05 1.2 39-48 SSVLESLVGR 17.66±1.50 39-48 36.12±0.51 39-48 15.18±0.88 0.49 0.8 1.16 0.7 61-75 RPLILQLVHVSQEDK 291.90±9.05 61-75 374.84±18.87 61-75 141.75±9.29 0.78 1.2 2.06 1.2 78-92 TTGEENGVEAEEWGK 260.07±40.43 78-92 569.72±15.83 78-92 188.36±19.22 0.46 0.7 1.38 0.8 100-108 LYTDFDEIR 189.30±6.33 100-108 292.87±6.02 100-108 102.47±7.47 0.65 1 1.85 1.1 100-117 LYTDFDEIRQEIENETER 523.33±11.19 100-117 1000±20.13 100-117 300.87±30.08 0.52 0.8 1.74 1 109-117 QEIENETER 123.02±5.86 109-117 212.14±11.02 109-117 72.43±5.20 0.58 0.9 1.7 1 124-133 GVSPEPIHLK 98.62±4.81 124-133 188.05±5.12 124-133 64.47±2.66 0.52 0.8 1.53 0.9 134-152 IFSPNVVNLTLVDLPGMTK 2840±251.7 134-152 2780±184.4 134-152 2340±179.0 1.02 1.6 1.2 0.7 153-160 VPVGDQPK 225.82±16.0 153-160 315.16±17.15 153-160 133.79±9.28 0.72 1.1 1.69 1 161-167 DIELQIR 240.36±9.03 161-167 615.80±39.68 161-167 150.80±7.95 0.39 0.6 1.59 0.9 173-198 FISNPNSIILAVTAANTDMATSEALK 4380±336.96 173-198 3290±126.28 173-198 3350±341.84 1.33 2 1.31 0.8 210-216 TLAVITK 14.28±0.43 210-216 25.49±0.76 210-216 5.97±0.46 0.56 0.9 2.39 1.4 217-233 LDLMDAGTDAMDVLMGR 19270±1250 217-233 14400±671.8 217-233 13280±1030 1.34 2.1 1.45 0.8 239-247 LGIIGVVNR 59.86±4.07 239-247 189.59±11.44 239-247 38.48±1.88 0.32 0.5 1.56 0.9 248-255 SQLDINNK 201.87±11.60 248-255 492.30±16.44 248-255 124.23±6.45 0.41 0.63 1.62 0.9 257-271 SVTDSIRDEYAFLQK 663.32±43.72 257-271 1060±28.32 257-271 345.55±30.66 0.63 1 1.92 1.1 272-279 KYPSLANR 330.08±22.65 272-279 484.15±19.87 272-279 154.02±13.40 0.68 1 2.14 1.2 292-298 LLMHHIR 3760±258.70 292-298 4670±9181 292-298 4250±416.69 0.81 1.2 0.88 0.5 299-305 DCLPELK 646.43±42.58 299-305 183.35±7.39 299-305 70.56±6.03 3.53 5.4 9.16 5.2 308-329 INVLAAQYQSLLNSYGEPVDDK 489.08±20.76 308-329 961.72±34.55 308-329 256.96±15.73 0.51 0.8 1.9 1.1 330-339 SATLLQLITK 138.06±7.26 330-339 465.12±17.75 330-339 106.48±7.93 0.3 0.5 1.3 0.7 340-353 FATEYCNTIEGTAK 124.23±3.35 340-353 228.90±15.31 340-353 60.73±2.45 0.54 0.8 2.05 1.2 354-365 YIETSELCGG(D)AR 45.46±5.0 354-365 197.82±6.49 354-365 38.61±2.14 0.23 0.4 1.18 0.7 366-376 ICYIFHETFGR 421.20±48.96 366-376 431.70±17.95 366-376 179.12±12.60 1 1.5 2.35 1.3 377-397 TLESVDPLGGLNTIDILTAIR 421.04±17.22 377-397 1620±55.96 377-397 241.77±17.80 0.26 0.4 1.74 1 404-418 PALFVPEVSFELLVK 2050±127.27 404-418 4590±412.0 404-418 1160±59.95 0.45 0.7 1.77 1 423-430 RLEEPSLR 43.21±1.92 423-430 155.75±1.98 423-430 42.38±1.88 0.28 0.4 1.02 0.6 442-456 IIQHCSNYSTQELLR 361.22±12.27 442-456 752.72±17.86 442-456 170.65±17.97 0.48 0.7 2.12 1.2 460-473 LHDAIVEVVTCLLR 7950±377.33 460-473 6080±339.51 460-473 2440±185.83 1.31 2 3.14 1.8 476-497 LPVTNEMVHNLVAIELAYINTK 4630±318.14 476-497 2870±132.46 476-497 2860±139.51 1.61 2.5 1.62 0.9 498-516 HPDFADACGLMNNNIEEQR 4960±461.80 498-516 5690±165.29 498-516 4570±337.65 0.87 1.3 1.09 0.6 523-530 ELPSAVSR 220.79±10.21 - - 523-530 155.86±11.87 - - 1.42 0.8 533-553 VASGGGGVGDGVQEPTTGNWR 797.68±26.75 - - 533-553 414.41±17.94 - - 1.92 1.1 561-569 AEELLAEEK 294.67±10.76 - - 561-569 170.21±12.90 - - 1.73 1 570-582 SKPIPIMPASPQK 8130±489.18 - - 570-582 7610±677.46 - - 1.07 0.6 583-597 GHAVNLLDVPVPVAR 1260±55.57 - - 583-597 732.56±60.79 - - 1.72 1 606-612 DCEVIER 31.09±1.76 517-523 161.54±3.83 606-612 25.79±1.16 0.19 0.3 1.21 0.7 616-622 SYFLIVR 0 527-533 0 616-622 0 0 0 0 0 624-631 NIQDSVPK 0 535-542 0 624-631 0 0 0 0 0 632-642 AVMHFLVNHVK 1250±102.37 543-553 2010±68.47 632-642 794.11±79.15 0.62 1 1.57 0. 9 643-655 DTLQSELVGQLYK 187.38±6.54 554-566 476.75±18.84 643-655 126.77±6.85 0.39 0.6 1.48 0.8 656-672 SSLLDDLLTESEDMAQR 4030±344.54 567-583 4200±111.74 656-672 3560±248.09 1 1.5 1.13 0.6 675-681 EAADMLK 21.81±2.59 586-592 31.47±3.37 675-681 24.28±3.15 0.69 1.1 0.9 0.5 682-694 ALQGASQIIAEIR 206.32±5.13 593-605 451.64±8.51 682-694 98.40±7.73 0.46 0.7 2.1 1.2

TABLE 4.1 Hydroxyl radical footprinting oxidation rates of Drp1 peptides.

157

TABLE 4.2

M191 (+16) 3830±268.9 3230±213.1 3520±395.0 1.19 1.8 1.09 0.6 173-198 173-198 173-198 K198 50.43±1.98 88.78±5.73 29.87±3.42 0.57 0.9 1.69 1 217-233 M220/M227/M231 19270±1250 217-233 14400±671.8 217-233 13280±1030 1.34 2.1 1.45 0.8 L239 13.54±0.55 42.90±2.40 9.44±0.93 0.32 0.5 1.43 0.8 239-247 I241 21.44±1.18 239-247 68.39±3.42 239-247 14.11±1.22 0.31 0.5 1.52 0.9 V245 24.46±1.89 67.61±3.01 14.33±0.68 0.36 0.5 1.71 1 C300 885.19±71.52 251.38±8.86 95.0±8.86 3.52 5.4 9.32 5.3 299-305 299-305 299-305 K305 64.28±2.93 172.26±12.06 54.76±3.48 0.37 0.6 1.17 0.7 L334 104.29±5.18 377.40±13.50 79.90±6.89 0.28 0.4 1.31 0.7 330-339 Q335 5.11±0.27 330-339 13.39±1.19 330-339 3.20±0.28 0.38 0.6 1.6 0.9 K339 30.61±1.04 96.78±4.29 22.56±2.10 0.32 0.5 1.36 0.8 F340 9.59±0.095 36.04±3.95 9.86±0.72 0.27 0.4 0.97 0.6 340-353 C345 99.28±2.72 340-353 156.44±8.86 340-353 42.79±1.60 0.63 1 2.32 1.3 K353 14.60±0.089 66.86±4.78 7.96±0.050 0.22 0.3 1.83 1 Y354 14.26±1.33 54.86±0.49 11.39±1.04 0.26 0.4 1.25 0.7 I355 15.98±2.28 80.54±3.08 11.27±0.99 0.2 0.3 1.42 0.8 354-365 E356 13.18±0.82 354-365 28.32±0.69 354-365 6.58±0.56 0.47 0.7 2 1.1 E359 9.86±0.96 14.54±1.41 4.10±0.31 0.68 1 2.4 1.4 L360 13.27±0.37 38.65±1.41 10.93±0.77 0.34 0.5 1.21 0.7 C367 406.54±35.88 417.67±22.14 137.26±5.94 0.97 1.5 2.96 1.7 366-376 F374 97.98±2.68 366-376 142.27±9.81 366-376 81.45±7.29 0.69 1.1 1.2 0.7 R376 21.63±2.46 72.60±4.40 25.53±1.55 0.3 0.5 0.85 0.5 E379/D382 12.69±0.49 41.62±2.46 9.15±0.68 0.3 0.5 1.39 0.8 V381 22.76±0.76 55.83±1.84 12.33±1.09 0.41 0.6 1.85 1.1 L384 106.50±5.71 380.64±14.19 68.61±5.27 0.28 0.4 1.55 0.9 377-397 377-397 377-397 I390 138.80±4.95 551.40±25.10 85.35±5.09 0.25 0.4 1.63 0.9 L393 56.45±3.53 186.51±8.85 23.13±1.58 0.3 0.5 2.44 1.4 R397 28.15±0.83 112.05±2.46 22.64±1.96 0.25 0.4 1.24 0.7 R423 4.58±0.88 53.56±1.23 8.62±1.39 0.09 0.1 0.53 0.3 E426 7.68±0.44 30.81±1.86 7.54±1.43 0.25 0.4 1.02 0.6 423-430 423-430 423-430 L429 34.44±0.63 120.21±0.78 30.70±2.19 0.29 0.4 1.12 0.6 R430 3.83±0.50 27.88±1.55 4.0±0.39 0.14 0.2 0.96 0.5 H461 107.76±13.14 112.70±12.42 64.57±6.21 0.96 1.5 1.67 1 460-473 460-473 460-473 C470 6640±400.56 5110±336.99 2120±174.57 1.3 2 3.13 1.8 M482 3320±137.67 2280±152.30 2430±130.82 1.46 2.2 1.37 0.8 476-497 476-497 476-497 H484 38.27±2.94 66.15±8.93 18.26±2.06 0.58 0.9 2.1 1.2

TABLE 4.2 Hydroxyl radical footprinting oxidation rates of Drp1 amino acid side chains.

158

CHAPTER 5: DISCUSSION AND FUTURE DIRECTIONS

159

5.1 Summary

Taken together, these studies have provided a fundamental understanding of the

interaction between Drp1 and Mff (Figure 5.1). We first established that the VD of Drp1

acts as a regulatory element to preclude interaction with Mff. In fact, upon removal of

this domain we were able to demonstrate the first structural and functional outcomes in

vitro resulting from the interaction of Mff with Drp1. We also revealed that Drp1-Mff co-

assemblies tethered to liposomes, but not in solution, can elicit a quantifiable change in

the functional properties of full-length Drp1. Building upon this finding, we established

that the oligomeric tendencies of both Mff and Drp1 alter the functional copolymers

formed by these proteins. This offers a potential insight into naturally occurring human

mutations in Drp1 and Mff that disrupt the oligomerization of these proteins. Following

the establishment of this foundation, the structural and functional aspects of this

interaction required further study.

We next found that the Drp1-Mff complex assembled on liposomes formed an

atypical dynamin complex. Rather than directly polymerizing on the surface of an Mff- decorated lipid bilayer, the mechanoenzymatic core of Drp1 was displaced from the membrane surface by a substantial (~10 nm) distance. Furthermore, Mff was not displaced during GTP hydrolysis by Drp1 and instead appeared to remain associated with

Drp1 throughout GTP-hydrolysis driven liposome constriction. In fact, Drp1 remained at the periphery of the tubules with the stalk domain facing the underlying lipid bilayer.

Building off of the apparent organization within the polymer, amino acid conservation analysis identified key regions within the stalk of Drp1 that specifically mediate its

160

interaction with Mff. We found that amino acids within L1CS, a flexible loop at the base

of the stalk, were necessary for stabilizing the interaction between Drp1 and Mff both in

vitro and in vivo. Furthermore, we found that the VD makes a specific intramolecular contact occluding this region, yielding mechanistic insight into its opposition of Mff interaction.

Taken together, this research has provided a foundation for understanding the interaction between Drp1 and Mff, but there are still many open questions that need to be addressed in order to fully understand the role played by Mff in Drp1-mediated

regulation of mitochondrial morphology. While no membrane fission was observed in

these studies, all of these experiments were performed with truncated Mff mutants

lacking the transmembrane domain. Thus, as a consequence of our experimental setup,

any contribution made by the transmembrane domain that directly contacts the MOM is

lost. This begs the question: how does full-length Mff aid in Drp1-mediated lipid

remodeling, and can other factors act in concert with Drp1 and Mff to further destabilize

membranes and elicit complete membrane fission? Moreover, while post-translational

regulation of Drp1 and Mff can dramatically alter the subcellular localization and function

of these proteins, what are the direct effects of Mff phosphorylations on the direct

interaction between Drp1 and Mff? What is the structure of Drp1 in complex with Mff,

and what Drp1 assembly interfaces mediate the formation of this complex? While Mff is

one of many Drp1 partner proteins, does it have any unique roles in mitochondrial fission

or is it instead a part of a larger fission complex? Finally, in its native membrane

environment, what lipids and proteins are closely associated with Mff?

161

When the studies described throughout this dissertation were initiated, little was

known concerning the direct interaction between Drp1 and Mff. This research has served to help establish a groundwork for further studies to build on and further elucidate the role(s) played by Mff in mitochondrial fission.

5.2 Recapitulating membrane fission in vitro

Complete membrane fission resulting from Drp1-mediated membrane remodeling, has yet to be recapitulated in vitro. This fact has suggested to some that Drp1 alone may be insufficient for mitochondrial fission and that instead Drp1 acts as a contractile protein that primes mitochondrial membranes for fission by a separate scissile dynamin

(specifically Dynamin 2) (Lee et al. 2016). While this is an attractive proposal, the full membrane remodeling capabilities of Drp1 in vitro have yet to be fully assessed. To

exclude Drp1 as the scissile dynamin first requires a more complete analysis of the

membrane remodeling activity of Drp1 in the presence of relevant factors including the

mitochondrial lipids, partner proteins, and mechanical forces that are all required for

mitochondrial fission in vivo.

Throughout these studies, only Mff lacking its transmembrane was utilized and it was

tethered to membranes rather than being anchored via a transmembrane peptide

segment. This is an important consideration because the oligomerization of full-length

Mff incorporated into membranes has never been assessed, and it is known that Mff self-

assembly plays a role in recruitment of Drp1 and its subsequent assembly (as highlighted

162

in Chapter 2). In fact, multiple groups including ours, have generated Mff mutants that

lack its transmembrane segment and interestingly, the size of the deleted segment seems

to inversely correlate with the oligomeric status (Koirala et al. 2013; R. W. Clinton et al.

2016). The deleted transmembrane segment may further alter the oligomeric tendencies

of Mff, and as such the intact protein may exhibit altered membrane remodeling capacity

relative to the truncated tetrameric mutant protein we used. Furthermore, lacking the

intimate contact between Mff and its associated membrane may fail to elicit any

membrane-deforming forces due specifically to Mff alone (Farsad and De Camilli 2003).

Thus, there exists the potential for additional contractile capacity and fission competence

for the Drp1-Mff complex with intact Mff. To address this limitation, full-length Mff can

be expressed as a soluble protein and can be reincorporated into proteoliposomes

(Macdonald et al. 2016). These liposomes can then be examined in studies that parallel

those in Chapter 4 (Figure 4.1) to compare the function of membrane inserted Mff and

scaffold liposome-tethered Mff. Furthermore, these studies with membrane-

reconstituted Mff remove a potential complication arising from the properties of lipid

components of the scaffolding liposomes used in the studies in Chapters 2-5: the effects

of NTA-Ni2+ functionalized lipids on intrinsic liposome plasticity and flexibility. Due to the inherent charged nature of this functional group, microdomains enriched for these lipids experience charge-repulsion interactions that can destabilize or even deform the lipid bilayer (Stachowiak et al. 2012). While we never observed of these types of artifacts through our experiments, any subtle effects can be completely discounted by using full-

163

length Mff, thus allowing for a completely unbiased assessment of the effect of Mff on

Drp1-mediated membrane remodeling and contractile properties.

Furthermore, the lipid remodeling capability of Drp1 in the presence of partner proteins and lipid cofactors together has not yet been assessed. Considering the

contractile capacity of a Drp1-Mff complex (as in Chapter 4) and the important role played

by cardiolipin in direct Drp1-lipid interactions (Francy et al. 2017) and stimulating the

GTPase activity of Drp1 (Macdonald et al. 2014), these additional factors may drive the

assembly of a more efficient fission complex. In fact, previous studies have provided

strong evidence that cardiolipin itself may be reorganized by Drp1 and undergo phase

transitions to destabilize membranes and ultimately promote membrane fission

intermediates (Stepanyants et al. 2015). Furthermore, a significant factor that is absent

from most in vitro reconstitution experiments that can contribute to ultimately achieving

Drp1-mediated membrane fission is the application of membrane tension. In cells,

mitochondria are connected to various components of the cellular cytoskeleton, and as a

consequence the membrane is under tension. In fact, the application of membrane

tension to dynamin-decorated liposomes dramatically enhances the efficiency of

membrane fission (Roux et al. 2006), and various mechanical forces trigger mitochondrial

fission in cells (Helle et al. 2017). Taken together, these observations indicate that

cardiolipin and membrane tension are potentially critical variables that have been missing

from previous experimental conditions.

Previous studies have not observed membrane fission by Drp1 even in the presence

of cardiolipin and membrane tension (Ugarte-Uribe et al. 2017), but the incorporation of

164

Mff in these experimental conditions may be required to produce membrane fission. To assess whether Mff plays a role in membrane fission under similar conditions, either membrane-scaffolded Mff or full-length incorporated Mff can be utilized. By forming giant unilamellar vesicles (GUVs) or supported bilayers with excess membrane reservoir

(SUPER) templates from Mff-containing (or Mff-tethering) liposomes of specified composition, the capacity of Drp1 remodeling in the presence of Mff and/or membrane elements (i.e. PS, CL, or PA) can be assessed. Using either fluorescently labeled GUVs or

SUPER templates, membrane remodeling and fission can be qualitatively assessed using live confocal microscopy. If membrane fission occurs under these conditions, a simple sedimentation assay that measures the release of fluorescent lipid vesicles from SUPER templates (Neumann, Pucadyil, and Schmid 2013; Pucadyil and Schmid 2008) can be used to quantify the rate of vesicle release and assess the role of the lipid and protein components of the membrane scission machine. Furthermore, the role of membrane tension can be assessed by extruding lipid tubules via pipette aspiration or by using optical tweezer techniques as previously described (Ugarte-Uribe et al. 2017). By incorporating full-length Mff, anionic lipid cofactors, and membrane tension into consideration in our experimental setup, a more physiologically relevant environment can be emulated in vitro to more accurately recapitulate Drp1 function.

165

5.3 Direct effects of Mff post-translational modification on the Drp1-Mff complex

All of the studies described throughout this dissertation have used recombinant proteins expressed in E. coli, and as a consequence these proteins lack post-translational modification. As highlighted in Chapter 1, phosphorylation of Drp1 and Mff dramatically controls the subcellular localization and mitochondrial fission activity of Drp1. Previous studies have described the phosphorylation of Mff by AMPK which enhances Drp1 translocation to the mitochondria and subsequent fission (E. Q. Toyama et al. 2016), but the direct effects on the Drp1-Mff interaction have not been investigated. To address this shortcoming, we generated a mutant of Mff that mimics the AMPK phosphorylated state

(MffS129D/S146D). Interestingly, this mutant does not stimulate the GTPase activity of

Drp1ΔVD to the same degree as non-modified Mff (Figure 5.2A). In agreement, the formation of filaments by ΔVD and Mff appears to be blunted as well (Figure 5.2B), which indicates that this modification alters the Drp1-assembling function of Mff. Remarkably, cells transfected with intact Mff bearing the same mutations exhibit fragmented mitochondria (E. Q. Toyama et al. 2016), indicating that stimulation of the GTPase activity of Drp1 does not necessarily correlate with mitochondrial fission. The functional role of these Mff phosphorylations in cells have yet to be fully elucidated, but may potentially involve the induction of phosphospecific protein-protein interactions between Mff and additional regulatory factors such as the 14-3-3 proteins discussed in Chapter 1. Taken together, these preliminary findings suggest that Mff phosphorylation can alter its assembly of Drp1, and may drive its interaction with yet-unidentified factors to drive mitochondrial fission. While AMPK phosphorylation of Mff to regulate mitochondrial

166

morphology implies an important connection between cellular energy levels and

mitochondrial fission, the role of other signaling pathways through Mff is still an

unknown.

An important and currently unexplored research area involves the phosphorylation of

Mff by other kinases such as those previously implicated in dictating mitochondrial fission

(e.g. PKA, CAMKII, Erk1/2, CDKs, etc.). Initial studies can rapidly address this by in vitro

phosphorylation of purified Mff with purified kinases of interest, followed by either

western blotting (with anti-pY/pS/pT antibodies) or electrophoresis through

functionalized polyacrylamide gels to separate phospho-peptides from unmodified

peptides (i.e. PhosTag PAGE, Wako). By identifying kinases that modify Mff in this way,

mass spectrometry can be used to identify the specific phosphorylated residues. This

knowledge will enable subsequent studies to identify the direct effect on Drp1-Mff

interaction and the consequences of these phosphorylations on mitochondrial

morphology in cells. Thus, much remains unknown about what kinases are responsible

for Mff phosphoregulation, what residues are modified, and the functional outcomes

resulting from these modifications.

5.4 Structures of a Drp1-Mff complex

The studies described in this dissertation provide a foundation for understanding the structural basis of Drp1-Mff interaction, but no high-resolution structural data for this protein complex is currently available. While cryo-EM studies have revealed a structure

167

of the Drp1-MiD49 complex at 4.2 Å resolution (Kalia et al. 2017a), no such studies have

been published with the Drp1-Mff complex. To this end, one of my initial goals was to

establish Drp1-Mff assemblies that could be imaged using cryo-EM to yield mid- to high-

resolution 3D structures. Through the course of my studies, two different potential Drp1-

Mff complexes were identified. First, the filaments formed by Drp1ΔVD and Mff appeared

to be highly ordered, but formed large bundled oligomers (Figure 3.1J and Figure 5.3A)

that can be manipulated by addition of guanosine nucleotides. The next step towards an

understanding of the organization of Drp1 and Mff within this polymer is collection of

high-resolution data using resources available via the national cryo-electron microscopy

facility (NCEF) or other similar facilities; specifically, a 300 keV microscope equipped with direct electron detector camera. Subsequent 3D reconstruction techniques such as Relion

(Scheres 2012), CryoSPARC (Punjani et al. 2017), or helical reconstruction techniques

(Egelman 2000; Desfosses et al. 2014) can be utilized with these filaments depending on the symmetry of the polymer. Preliminary classification of such particles using Relion

(Figure 5.3B) highlights the various views of these filaments. Furthermore, several classes

appear to have multiple filaments within a single class, so the particle selection clearly

needs optimization. The organization of Drp1 and Mff within these filaments remains

unclear though, and previous structural studies of Drp1 suggest two potential

symmetries: linear or helical. Assembly of a similar Drp1 mutant protein (also lacking the

VD) into linear filaments was identified within the crystals used to solve the structure of

Drp1 (Fröhlich et al. 2013). Conversely, the cryo-EM structure of the Drp1-MiD49 co- complex forms a helical filament with an ‘inside-out’ orientation where Drp1 GTPase

168

domains were at the center of the polymer. Clearly, further work is required to elucidate

the organization of Drp1 and Mff within these filaments.

In cells, Drp1 and Mff interact at the surface of the mitochondrial outer membrane, but the Drp1-Mff filaments described above cannot account for the presence of a lipid

environment. To address this, an additional cryo-EM target was designed to parallel

previous work from our lab (Francy et al. 2017). Using scaffolding lipid nanotubes containing cardiolipin (i.e. Figure 4.2B), Mff was tethered and allowed to interact with

full-length Drp1 with subsequent stabilization of the polymer using GMP-PCP. These

tubes were frozen and imaged on a TF20 microscope at 200 keV with a Tietz 4k x 4k CCD

camera, yielding a dataset with moderate resolution potential. What became

immediately apparent was the variability in diameter of these tubes (Figure 5.3C), which

represents a limitation of reconstructions using this target. This heterogeneity can be

avoided by grouping particles by diameter, and treating each diameter group individually.

A consequence of this subdivision of picked particles is that an extremely large volume of

data is required to obtain a sorted dataset with sufficient particles for high-resolution

reconstructions. As with the filaments discussed above, efforts to generate a 3D

reconstruction of these helical polymers require collection of a large high-resolution

dataset at a national microscopy center. An added benefit to these microscopy resources

is the use of automated image acquisition software, specifically Leginon (Carragher et al.

2000) to semi-autonomously capture large datasets of specific sub-grid regions of interest

(i.e. nanotubes). Unfortunately, the identification of particles and their subsequent

169 extraction is still a manual process and as such is the rate limiting step in this type of project.

In addition to the use of cryo-EM to probe the structural basis for the interaction of

Drp1 and Mff, one of several mass-spectrometry footprinting techniques can be utilized to augment cryo-EM studies or as an alternative. Some such techniques include the use of hydroxyl radical footprinting and hydrogen-deuterium exchange footprinting (HDX)

(Katta and Chait 1991; Percy et al. 2012). In an experiment comparable to those described in Chapter 4 (i.e. figures 4.8 and 4.9), the ΔVD-Mff complex as well as each protein individually can be analyzed by using hydroxyl radical footprinting or HDX. Both techniques can provide information about the relative solvent accessibility and/or intrinsic flexibility of the amino acids of each protein as measured by sidechain oxidation

(in the case of hydroxyl radical footprinting) or the exchange of amino acid amide hydrogens for deuterium (HDX). By comparing the relative rates of oxidation or hydrogen- deuterium exchange of each protein in the complex to that of the proteins alone, specific residues of both Drp1 and Mff that participate in their interaction can be identified.

Regardless of the techniques used to probe the Drp1-Mff complex, important information about its assembly will be revealed. First of all, if high resolution structures are achieved (at least ~5 Å) the specific amino acids that participate in the interaction surface(s) can be visualized within the complex. Using this information in combination with sequence conservation analysis (as described in Chapter 4), mutants completely disrupt the Drp1-Mff interaction can be developed as tools for studying Mff-specific Drp1 function in cells. Furthermore, even with more moderate resolution reconstructions the

170 mode of Drp1 assembly in complex with Mff can be assessed. Previous studies of the classic dynamin have identified critical self-assembly interfaces across the molecule that mediate various assembly modes including the dimer interface (stalk Interface 2), GTPase domain dimerization, as well as stalk Interfaces 1 and 3 that are thought to participate in building lipid-deforming polymers (M. G. Ford, Jenni, and Nunnari 2011). In contrast, comparable studies of Drp1 do not find polymers built with the same interfaces, and instead the helical polymer is formed solely by stalk interface 2 and GTPase domain dimerization (Francy et al. 2017), which suggests that Drp1 assembly properties differ from those of the endocytic dynamin. Furthermore, the assemblies formed by Drp1 alone differ greatly from those formed in the presence of Mff (as highlighted in Figure 4.2B).

Thus, the assembly mode of Drp1 in the presence of Mff may reveal key intramolecular interactions that contribute to a functional membrane-fission polymer in vivo. Finally, by combining cryo-EM reconstructions along with footprinting studies, a more complete understanding of the interaction and assembly of the Drp1-Mff complex can be achieved.

5.5 Discerning the specific role of Mff in mitochondrial fission

Genetic studies have clearly established the necessity of Mff for Drp1-mediated membrane recruitment and fission, but its specific contribution to membrane fission in cells is unclear. Our studies have demonstrated that Mff recruits Drp1 to target membranes and augments its assembly into a functional contractile copolymer. However, mitochondria are decorated with several other partner proteins including Fis1, MiD49,

171

and MiD51. The exact role played by each of these proteins is unclear, in fact the loss of

any partner proteins (with the possible exception of Fis1) results in comparable effects on

mitochondrial morphology and responses to stressors (Osellame et al. 2016). This

indicates either a cooperative role played by these proteins in concert to enable

mitochondrial fission, or alternatively that these proteins have some degree of functional

redundancy. Previous studies that have tried to identify specific roles of Mff in

mitochondrial fission are complicated by the fact that the most common approach is

knockdown or genetic ablation of Mff. This complete loss of Mff has the potential to have

off-target effects on other cellular processes than mitochondrial fission including, but not

limited to, mitophagy which may distort the results. To avoid this, mutants or small

molecule agents that selectively interrupt the interaction between Drp1 and Mff are

required.

Specifically, the chimeric ‘TSN’ mutant identified in Chapter 4 represents a novel

approach to identify the mitochondrial fission-specific role of Mff. By replacing

endogenous Drp1 with Mff-interaction incompetent Drp1, cells can be stressed and the

resulting effects on mitochondrial morphology and mitophagy can be quantified. For

instance, fission can be induced by imparting various stresses onto cells such as loss of

MMP (e.g. with CCCP), pan-kinase inhibition (with staurosporine), or intracellular calcium mobilization (e.g. with ionomycin). By monitoring mitochondrial morphology in treated via immunofluorescence, the effects on cells expressing wildtype and Mff-interaction incompetent Drp1 can be assessed. Furthermore, mitophagy can be triggered by extended exposure to CCCP or by overexpression of a mitochondrial-targeted misfolded

172

protein mutant (Jin and Youle 2013). The progression of mitophagy can be monitored using fluorescent dye-based kits (Mtphagy dye, Dojindo Molecular technologies) or by western blotting for outer and inner membrane protein (e.g. Tom20 and Tim23) and

quantifying changes in the levels of these proteins compared to non-mitochondrial

marker proteins (Y. Zhu et al. 2014). These interaction disrupting mutants represent a novel way to assess the specific role of Mff in Drp1-mediated mitochondrial fission

without sacrificing any Drp1-independent Mff functions.

In addition to the use of mutant proteins for studying the Drp1-Mff interaction in cells,

the development of small molecule interaction inhibitors is another potential avenue for

studying the role of Drp1-Mff interaction in a cellular environment. By using an assay

where the activity of Drp1 is stimulated by Mff (i.e. Figure 3.1B), a strategy of high-

throughput screens and specific counter screens can be enacted to identify small

molecules that inhibit Mff stimulation of Drp1 without altering the activity or assembly of

Drp1 alone. Further strengthening this rationale, recently published studies have identified compounds that selectively inhibit the cardiolipin-stimulated activity of Drp1

(Mallat et al. 2018).

Using interaction-disrupting small molecules or interaction-interrupting mutant proteins, the importance of Mff for pathological mitochondrial fission can be assessed.

Previous studies have indicated a therapeutic benefit to inhibiting mitochondrial fission in ischemic insults (e.g. stroke/heart attack), Parkinson’s disease, and Huntingtin’s disease

(Ong et al. 2010; Filichia et al. 2016; Xing Guo et al. 2013). While either strategy offers a potential for studying the role of Mff in these pathologies, small molecule interaction

173 inhibitors offer a potential for translation to treating human diseases. Therefore, drug discovery efforts centered on identifying small molecule modulators of the Drp1-Mff interaction represent a promising strategy for more fully understanding the role played by Mff in Drp1-mediated membrane fission and for development of novel therapeutic agents.

5.6 Identifying proteins and lipids in the Mff microenvironment in vivo

Finally, while the subcellular localization of Mff is known to be punctate and coincides with sites of mitochondria-ER contact, its local membrane environment and the factors dictating this localization are unclear. To address this, isolation of Mff in its native mitochondrial membrane could identify closely associated proteins and lipids. This isolation can be achieved using styrene-maleic acid copolymers (SMAPs). These polymers insert into lipid bilayers, and extract small (roughly 10-15 nm) lipid nanodiscs (Jamshad et al. 2011). These polymers can be added to isolated mitochondria to dissolve the outer and inner membranes into nanodiscs, and subsequent immunoprecipitation with Mff antibodies can enrich for those which contain Mff. Using these Mff-enriched nanodiscs compared to the total mitochondrial sample, comparative lipidomic and proteomic analysis can reveal what proteins and lipids are selectively enriched in close proximity to

Mff in native mitochondrial membranes. Furthermore, by isolating mitochondria from cells treated with mitochondrial fission stimulating small molecules (e.g. CCCP) or apoptotic stimuli (e.g. staurosporine), changes in Mff’s microenvironment in response to

174 these various stressors can be assessed. Thus, by assessing what lipids and proteins closely associate with Mff on mitochondria in vitro as well as those that are specifically recruited upon the induction of mitochondrial fission, new factors can be identified that may interact with Mff to augment recruitment of Drp1.

5.7 Conclusions

While the role of Mff in metazoan mitochondrial fission has been established for exactly 10 years, its direct interaction with Drp1 has been more difficult to isolate and study. The research presented throughout this dissertation provides a basis for understanding the interaction between Drp1 and Mff. We demonstrated that the interaction between these proteins is mediated by a stalk interface that is selectively occluded by the variable domain. Furthermore, Mff recruits and assembles Drp1 dimers to form a membrane remodeling copolymer on target membranes. Taken together, these results suggest a role for Mff whereby it both recruits and facilitates the assembly of a membrane-active copolymer to drive mitochondrial fission. Furthermore, these studies have suggested that the selective disruption of the Drp1-Mff interaction is possible, and the design of disruptive pharmacologic agents may represent a novel strategy to reverse the aberrant mitochondrial morphology found in many human diseases.

175

FIGURE 5.1

FIGURE 5.1 Mff recruits Drp1 to membranes to form a functional membrane remodeling copolymer. Drp1 (colored as in Figure 4.8) dynamically cycles among dimeric and larger oligomeric species in solution, but only the dimeric population interacts with Mff. The interaction between Drp1 and Mff occurs at the base of the stalk of Drp1, and is opposed by the variable domain which selectively occludes this region. Mff tetramerizes via a

coiled-coil motif (yellow region), but is predominantly intrinsically disordered (orange

region). Upon its recruitment to a lipid bilayer, Drp1 and Mff form a helical copolymer

that can tubulates and constricts liposomes.

176

FIGURE 5.2

FIGURE 5.2 AMPK phosphomimetic Mff mutants are deficient in Drp1 assembly and stimulation. Mff with mutations to mimic AMPK phosphorylation (S129D/S146D) does not stimulate Drp1ΔVD GTPase activity as robustly as the wildtype protein (A). *: p < 0.05 based on student’s t-test. Timecourse sedimentation analysis reveals an assembly defect of the phopshomimic protein compared to the wildtype (B).

177

FIGURE 5.2

FIGURE 5.2 Drp1-Mff complexes for Cryo-EM study. (A) Negative stain EM micrographs highlighting filaments formed by Drp1ΔVD in the presence of Mff. Scale bar = 100nm. (B)

Relion unsupervised 2D classification of Drp1-Mff filaments, highlighting a repeating filamentous structure with unknown symmetry. (C) Relion unsupervised 2D classification of Drp1-Mff polymers assembled on lipid nanotubes as initially described in Figure 4.2C.

178

REFERENCES

Accola, Molly A., Bing Huang, Azzah Al Masri, and Mark A. McNiven. 2002. “The Antiviral

Dynamin Family Member, MxA, Tubulates Lipids and Localizes to the Smooth

Endoplasmic Reticulum.” Journal of Biological Chemistry 277 (24): 21829–35.

https://doi.org/10.1074/jbc.M201641200.

Adachi, Yoshihiro, Kie Itoh, Tatsuya Yamada, Kara L. Cerveny, Takamichi L. Suzuki,

Patrick Macdonald, Michael A. Frohman, Rajesh Ramachandran, Miho Iijima, and

Hiromi Sesaki. 2016. “Coincident Phosphatidic Acid Interaction Restrains Drp1 in

Mitochondrial Division.” Molecular Cell 63 (6): 1034–43.

https://doi.org/10.1016/j.molcel.2016.08.013.

Alexander, C., M. Votruba, U. E. Pesch, D. L. Thiselton, S. Mayer, A. Moore, M.

Rodriguez, et al. 2000. “OPA1, Encoding a Dynamin-Related GTPase, Is Mutated

in Autosomal Dominant Optic Atrophy Linked to Chromosome 3q28.” Nat Genet

26 (October): 211–15. https://doi.org/10.1038/79944.

Andersson, S. G., A. Zomorodipour, J. O. Andersson, T. Sicheritz-Pontén, U. C. Alsmark,

R. M. Podowski, A. K. Näslund, A. S. Eriksson, H. H. Winkler, and C. G. Kurland.

1998. “The Genome Sequence of Rickettsia Prowazekii and the Origin of

Mitochondria.” Nature 396 (6707): 133–40. https://doi.org/10.1038/24094.

Ashrafian, H., L. Docherty, V. Leo, C. Towlson, M. Neilan, V. Steeples, C. A. Lygate, et al.

2010. “A Mutation in the Mitochondrial Fission Gene Dnm1l Leads to

Cardiomyopathy.” PLoS Genet 6: e1001000.

https://doi.org/10.1371/journal.pgen.1001000.

179

Ban, Tadato, Takaya Ishihara, Hiroto Kohno, Shotaro Saita, Ayaka Ichimura, Katsumi

Maenaka, Toshihiko Oka, Katsuyoshi Mihara, and Naotada Ishihara. 2017.

“Molecular Basis of Selective Mitochondrial Fusion by Heterotypic Action

between OPA1 and Cardiolipin.” Nature Cell Biology 19 (7): 856–63.

https://doi.org/10.1038/ncb3560.

Barklis, Eric, Jason McDermott, Stephan Wilkens, Eric Schabtach, M. F. Schmid, Stephen

Fuller, Sonya Karanjia, et al. 1997. “Structural Analysis of Membrane‐bound

Retrovirus Capsid Proteins.” The EMBO Journal 16 (6): 1199–1213.

https://doi.org/10.1093/emboj/16.6.1199.

Bausewein, Thomas, Deryck J. Mills, Julian D. Langer, Beate Nitschke, Stephan

Nussberger, and Werner Kühlbrandt. 2017. “Cryo-EM Structure of the TOM Core

Complex from Neurospora Crassa.” Cell 170 (4): 693-700.e7.

https://doi.org/10.1016/j.cell.2017.07.012.

Bayrhuber, Monika, Thomas Meins, Michael Habeck, Stefan Becker, Karin Giller, Saskia

Villinger, Clemens Vonrhein, Christian Griesinger, Markus Zweckstetter, and

Kornelius Zeth. 2008. “Structure of the Human Voltage-Dependent Anion

Channel.” Proceedings of the National Academy of Sciences of the United States

of America 105 (40): 15370–75. https://doi.org/10.1073/pnas.0808115105.

Becker, Dorothea, Judith Richter, Maja A. Tocilescu, Serge Przedborski, and Wolfgang

Voos. 2012. “Pink1 Kinase and Its Membrane Potential (Δψ)-Dependent Cleavage

Product Both Localize to Outer Mitochondrial Membrane by Unique Targeting

180

Mode.” Journal of Biological Chemistry 287 (27): 22969–87.

https://doi.org/10.1074/jbc.M112.365700.

Berman, S. B., F. J. Pineda, and J. M. Hardwick. 2008. “Mitochondrial Fission and Fusion

Dynamics: The Long and Short of It.” Cell Death Differ 15 (July): 1147–52.

https://doi.org/10.1038/cdd.2008.57.

Bhar, Debjani, Mary Anne Karren, Markus Babst, and Janet M. Shaw. 2006. “Dimeric

Dnm1-G385D Interacts with Mdv1 on Mitochondria and Can Be Stimulated to

Assemble into Fission Complexes Containing Mdv1 and Fis1.” Journal of

Biological Chemistry 281 (25): 17312–20.

https://doi.org/10.1074/jbc.M513530200.

Binns, Derk D., Michael K. Helms, Barbara Barylko, Colin T. Davis, David M. Jameson,

Joseph P. Albanesi, and John F. Eccleston. 2000. “The Mechanism of GTP

Hydrolysis by Dynamin II: A Transient Kinetic Study.” Biochemistry 39 (24):

7188–96. https://doi.org/10.1021/bi000033r.

Bliek, AM van der, T. E. Redelmeier, H. Damke, E. J. Tisdale, E. M. Meyerowitz, and S. L.

Schmid. 1993. “Mutations in Human Dynamin Block an Intermediate Stage in

Coated Vesicle Formation.” The Journal of Cell Biology 122 (3): 553–63.

https://doi.org/10.1083/jcb.122.3.553.

Boldogh, Istvan, Nikola Vojtov, Sharon Karmon, and Liza A. Pon. 1998. “Interaction

between Mitochondria and the Actin Cytoskeleton in Budding Yeast Requires

Two Integral Mitochondrial Outer Membrane Proteins, Mmm1p and Mdm10p.”

The Journal of Cell Biology 141 (6): 1371–81.

181

Bossy, B., A. Petrilli, E. Klinglmayr, J. Chen, U. Lutz-Meindl, A. B. Knott, E. Masliah, R.

Schwarzenbacher, and E. Bossy-Wetzel. 2010. “S-Nitrosylation of DRP1 Does Not

Affect Enzymatic Activity and Is Not Specific to Alzheimer’s Disease.” J Alzheimers

Dis 20 Suppl 2: S513-26. https://doi.org/10.3233/JAD-2010-100552.

Brandt, Tobias, Laetitia Cavellini, Werner Kühlbrandt, and Mickaël M. Cohen. 2016. “A

Mitofusin-Dependent Docking Ring Complex Triggers Mitochondrial Fusion in

Vitro.” ELife 5 (June): e14618. https://doi.org/10.7554/eLife.14618.

Bratton, Donna L., Valerie A. Fadok, Donald A. Richter, Jenai M. Kailey, Lindsay A.

Guthrie, and Peter M. Henson. 1997. “Appearance of Phosphatidylserine on

Apoptotic Cells Requires Calcium-Mediated Nonspecific Flip-Flop and Is

Enhanced by Loss of the Aminophospholipid Translocase.” Journal of Biological

Chemistry 272 (42): 26159–65. https://doi.org/10.1074/jbc.272.42.26159.

Brdiczka, Dieter G., Dmitry B. Zorov, and Shey-Shing Sheu. 2006. “Mitochondrial Contact

Sites: Their Role in Energy Metabolism and Apoptosis.” Biochimica et Biophysica

Acta (BBA) - Molecular Basis of Disease, Mitochondria in Diseases and

Therapeutics, 1762 (2): 148–63. https://doi.org/10.1016/j.bbadis.2005.09.007.

Bui, H. T., and J. M. Shaw. 2013. “Dynamin Assembly Strategies and Adaptor Proteins in

Mitochondrial Fission.” Curr Biol 23 (October): R891-9.

https://doi.org/10.1016/j.cub.2013.08.040.

Bui, Huyen T., Mary A. Karren, Debjani Bhar, and Janet M. Shaw. 2012. “A Novel Motif in

the Yeast Mitochondrial Dynamin Dnm1 Is Essential for Adaptor Binding and

182

Membrane Recruitment.” J Cell Biol 199 (4): 613–22.

https://doi.org/10.1083/jcb.201207079.

Burman, Jonathon L., Sarah Pickles, Chunxin Wang, Shiori Sekine, Jose Norberto S.

Vargas, Zhe Zhang, Alice M. Youle, et al. 2017. “Mitochondrial Fission Facilitates

the Selective Mitophagy of Protein Aggregates.” J Cell Biol, September,

jcb.201612106. https://doi.org/10.1083/jcb.201612106.

Bustillo-Zabalbeitia, Itsasne, Sylvie Montessuit, Etienne Raemy, Gorka Basañez, Oihana

Terrones, and Jean-Claude Martinou. 2014. “Specific Interaction with Cardiolipin

Triggers Functional Activation of Dynamin-Related Protein 1.” PLOS ONE 9 (7):

e102738. https://doi.org/10.1371/journal.pone.0102738.

Carragher, B., N. Kisseberth, D. Kriegman, R. A. Milligan, C. S. Potter, J. Pulokas, and A.

Reilein. 2000. “Leginon: An Automated System for Acquisition of Images from

Vitreous Ice Specimens.” Journal of Structural Biology 132 (1): 33–45.

https://doi.org/10.1006/jsbi.2000.4314.

Cassidy-Stone, A., J. E. Chipuk, E. Ingerman, C. Song, C. Yoo, T. Kuwana, M. J. Kurth, et al.

2008. “Chemical Inhibition of the Mitochondrial Division Dynamin Reveals Its

Role in Bax/Bak-Dependent Mitochondrial Outer Membrane Permeabilization.”

Dev Cell 14 (February): 193–204. https://doi.org/10.1016/j.devcel.2007.11.019.

Celia, Hervé, Elizabeth Wilson-Kubalek, Ronald A. Milligan, and Luc Teyton. 1999.

“Structure and Function of a Membrane-Bound Murine MHC Class I Molecule.”

Proceedings of the National Academy of Sciences 96 (10): 5634–39.

https://doi.org/10.1073/pnas.96.10.5634.

183

Cereghetti, G. M., A. Stangherlin, O. Martins de Brito, C. R. Chang, C. Blackstone, P.

Bernardi, and L. Scorrano. 2008. “Dephosphorylation by Calcineurin Regulates

Translocation of Drp1 to Mitochondria.” Proc Natl Acad Sci U S A 105 (October):

15803–8. https://doi.org/10.1073/pnas.0808249105.

Chan, D. C. 2006. “Mitochondrial Fusion and Fission in Mammals.” Annu Rev Cell Dev

Biol 22: 79–99. https://doi.org/10.1146/annurev.cellbio.22.010305.104638.

Chan, Nickie C., Anna M. Salazar, Anh H. Pham, Michael J. Sweredoski, Natalie J. Kolawa,

Robert L. J. Graham, Sonja Hess, and David C. Chan. 2011. “Broad Activation of

the Ubiquitin–proteasome System by Parkin Is Critical for Mitophagy.” Human

Molecular Genetics 20 (9): 1726–37. https://doi.org/10.1093/hmg/ddr048.

Chang, C. R., and C. Blackstone. 2007. “Cyclic AMP-Dependent Protein Kinase

Phosphorylation of Drp1 Regulates Its GTPase Activity and Mitochondrial

Morphology.” J Biol Chem 282 (July): 21583–87.

https://doi.org/10.1074/jbc.C700083200.

Chang, C. R., C. M. Manlandro, D. Arnoult, J. Stadler, A. E. Posey, R. B. Hill, and C.

Blackstone. 2010. “A Lethal de Novo Mutation in the Middle Domain of the

Dynamin-Related GTPase Drp1 Impairs Higher Order Assembly and

Mitochondrial Division.” J Biol Chem 285 (October): 32494–503.

https://doi.org/10.1074/jbc.M110.142430.

Chappie, J. S., J. A. Mears, S. Fang, M. Leonard, S. L. Schmid, R. A. Milligan, J. E. Hinshaw,

and F. Dyda. 2011. “A Pseudoatomic Model of the Dynamin Polymer Identifies a

184

Hydrolysis-Dependent Powerstroke.” Cell 147 (September): 209–22.

https://doi.org/10.1016/j.cell.2011.09.003.

Chen, C. H., S. L. Howng, S. L. Hwang, C. K. Chou, C. H. Liao, and Y. R. Hong. 2000.

“Differential Expression of Four Human Dynamin-like Protein Variants in Brain

Tumors.” DNA Cell Biol 19 (March): 189–94.

https://doi.org/10.1089/104454900314573.

Chen, H., and D. C. Chan. 2009. “Mitochondrial Dynamics--Fusion, Fission, Movement,

and Mitophagy--in Neurodegenerative Diseases.” Hum Mol Genet 18 (October):

R169-76. https://doi.org/10.1093/hmg/ddp326.

Chen, H., S. A. Detmer, A. J. Ewald, E. E. Griffin, S. E. Fraser, and D. C. Chan. 2003.

“Mitofusins Mfn1 and Mfn2 Coordinately Regulate Mitochondrial Fusion and Are

Essential for Embryonic Development.” J Cell Biol 160 (January): 189–200.

https://doi.org/10.1083/jcb.200211046.

Chen, Hsiuchen, Shuxun Ren, Clary Clish, Mohit Jain, Vamsi Mootha, J. Michael

McCaffery, and David C. Chan. 2015. “Titration of Mitochondrial Fusion Rescues

Mff-Deficient Cardiomyopathy.” The Journal of Cell Biology 211 (4): 795–805.

https://doi.org/10.1083/jcb.201507035.

Cho, Bongki, Hyo Min Cho, Hyun Jung Kim, Jaehoon Jeong, Sang Ki Park, Eun Mi Hwang,

Jae-Yong Park, Woon Ryoung Kim, Hyun Kim, and Woong Sun. 2014. “CDK5-

Dependent Inhibitory Phosphorylation of Drp1 during Neuronal Maturation.”

Experimental & Molecular Medicine 46 (7): e105.

https://doi.org/10.1038/emm.2014.36.

185

Choi, S. Y., P. Huang, G. M. Jenkins, D. C. Chan, J. Schiller, and M. A. Frohman. 2006. “A

Common Lipid Links Mfn-Mediated Mitochondrial Fusion and SNARE-Regulated

Exocytosis.” Nat Cell Biol 8 (November): 1255–62.

https://doi.org/10.1038/ncb1487.

Chu, C. T., J. Ji, R. K. Dagda, J. F. Jiang, Y. Y. Tyurina, A. A. Kapralov, V. A. Tyurin, et al.

2013. “Cardiolipin Externalization to the Outer Mitochondrial Membrane Acts as

an Elimination Signal for Mitophagy in Neuronal Cells.” Nat Cell Biol 15

(October): 1197–1205. https://doi.org/10.1038/ncb2837.

Clinton, R. W., C. A. Francy, R. Ramachandran, X. Qi, and J. A. Mears. 2016. “Dynamin-

Related Protein 1 Oligomerization in Solution Impairs Functional Interactions

with Membrane-Anchored Mitochondrial Fission Factor.” J Biol Chem 291

(January): 478–92. https://doi.org/10.1074/jbc.M115.680025.

Clinton, Ryan W., and Jason A. Mears. 2017. “Using Scaffold Liposomes to Reconstitute

Lipid-Proximal Protein-Protein Interactions In Vitro.” JoVE (Journal of Visualized

Experiments), no. 119 (January): e54971–e54971.

https://doi.org/10.3791/54971.

Comte, Jane, Bernard Maǐsterrena, and Daniéle C. Gautheron. 1976. “Lipid Composition

and Protein Profiles of Outer and Inner Membranes from Pig Heart

Mitochondria. Comparison with Microsomes.” Biochimica et Biophysica Acta

(BBA) - Biomembranes 419 (2): 271–84. https://doi.org/10.1016/0005-

2736(76)90353-9.

186

Connerth, Melanie, Takashi Tatsuta, Mathias Haag, Till Klecker, Benedikt Westermann,

and Thomas Langer. 2012. “Intramitochondrial Transport of Phosphatidic Acid in

Yeast by a Lipid Transfer Protein.” Science 338 (6108): 815–18.

https://doi.org/10.1126/science.1225625.

Cribbs, J. T., and S. Strack. 2007. “Reversible Phosphorylation of Drp1 by Cyclic AMP-

Dependent Protein Kinase and Calcineurin Regulates Mitochondrial Fission and

Cell Death.” EMBO Rep 8 (October): 939–44.

https://doi.org/10.1038/sj.embor.7401062.

Dang, T. X., R. A. Milligan, R. K. Tweten, and E. M. Wilson-Kubalek. 2005. “Helical

Crystallization on Nickel-Lipid Nanotubes: Perfringolysin O as a Model Protein.” J

Struct Biol 152 (November): 129–39. https://doi.org/10.1016/j.jsb.2005.07.010.

Davies, K. M., C. Anselmi, I. Wittig, J. D. Faraldo-Gomez, and W. Kuhlbrandt. 2012.

“Structure of the Yeast F1Fo-ATP Synthase Dimer and Its Role in Shaping the

Mitochondrial Cristae.” Proc Natl Acad Sci U S A 109 (August): 13602–7.

https://doi.org/10.1073/pnas.1204593109.

Delettre, Cécile, Jean-Michel Griffoin, Josseline Kaplan, Hélène Dollfus, Birgit Lorenz,

Laurence Faivre, Guy Lenaers, Pascale Belenguer, and Christian P. Hamel. 2001.

“Mutation Spectrum and Splicing Variants in the

Type="Italic">OPA1 Gene.” Human Genetics 109 (6): 584–91.

https://doi.org/10.1007/s00439-001-0633-y.

Desfosses, Ambroise, Rodolfo Ciuffa, Irina Gutsche, and Carsten Sachse. 2014. “SPRING

– An Image Processing Package for Single-Particle Based Helical Reconstruction

187

from Electron Cryomicrographs.” Journal of Structural Biology 185 (1): 15–26.

https://doi.org/10.1016/j.jsb.2013.11.003.

Dorn, Gerald W, Charles F Clark, William H Eschenbacher, Min-Young Kang, John T

Engelhard, Stephen J. Warner, Scot J Matkovich, and Casey C Jowdy. 2011.

“MARF and Opa1 Control Mitochondrial and Cardiac Function in Drosophila.”

Circulation Research 108 (1): 12–17.

https://doi.org/10.1161/CIRCRESAHA.110.236745.

Ducommun, Serge, Maria Deak, David Sumpton, Rebecca J. Ford, Antonio Núñez

Galindo, Martin Kussmann, Benoit Viollet, et al. 2015. “Motif Affinity and Mass

Spectrometry Proteomic Approach for the Discovery of Cellular AMPK Targets:

Identification of Mitochondrial Fission Factor as a New AMPK Substrate.” Cellular

Signalling 27 (5): 978–88. https://doi.org/10.1016/j.cellsig.2015.02.008.

Egelman, E. H. 2000. “A Robust Algorithm for the Reconstruction of Helical Filaments

Using Single-Particle Methods.” Ultramicroscopy 85 (December): 225–34.

Elgass, K., J. Pakay, M. T. Ryan, and C. S. Palmer. 2013. “Recent Advances into the

Understanding of Mitochondrial Fission.” Biochim Biophys Acta 1833 (January):

150–61. https://doi.org/10.1016/j.bbamcr.2012.05.002.

Esposito, E. A., A. L. Shrout, and R. M. Weis. 2008. “Template-Directed Self-Assembly

Enhances RTK Catalytic Domain Function.” J Biomol Screen 13 (September): 810–

16. https://doi.org/10.1177/1087057108322062.

Fadok, Valerie A., Aimee de Cathelineau, David L. Daleke, Peter M. Henson, and Donna

L. Bratton. 2001. “Loss of Phospholipid Asymmetry and Surface Exposure of

188

Phosphatidylserine Is Required for Phagocytosis of Apoptotic Cells by

Macrophages and Fibroblasts.” Journal of Biological Chemistry 276 (2): 1071–77.

https://doi.org/10.1074/jbc.M003649200.

Faelber, K., Y. Posor, S. Gao, M. Held, Y. Roske, D. Schulze, V. Haucke, F. Noe, and O.

Daumke. 2011. “Crystal Structure of Nucleotide-Free Dynamin.” Nature 477

(September): 556–60. https://doi.org/10.1038/nature10369.

Farsad, Khashayar, and Pietro De Camilli. 2003. “Mechanisms of Membrane

Deformation.” Current Opinion in Cell Biology 15 (4): 372–81.

Fernández-Morán, H., T. Oda, P. V. Blair, and D. E. Green. 1964. “A MACROMOLECULAR

REPEATING UNIT OF MITOCHONDRIAL STRUCTURE AND FUNCTION: Correlated

Electron Microscopic and Biochemical Studies of Isolated Mitochondria and

Submitochondrial Particles of Beef Heart Muscle.” The Journal of Cell Biology 22

(1): 63–100. https://doi.org/10.1083/jcb.22.1.63.

Figueroa-Romero, Claudia, Jorge A. Iñiguez-Lluhí, Julia Stadler, Chuang-Rung Chang,

Damien Arnoult, Peter J. Keller, Yu Hong, Craig Blackstone, and Eva L. Feldman.

2009. “SUMOylation of the Mitochondrial Fission Protein Drp1 Occurs at

Multiple Nonconsensus Sites within the B Domain and Is Linked to Its Activity

Cycle.” The FASEB Journal 23 (11): 3917–27. https://doi.org/10.1096/fj.09-

136630.

Filichia, Emily, Barry Hoffer, Xin Qi, and Yu Luo. 2016. “Inhibition of Drp1 Mitochondrial

Translocation Provides Neural Protection in Dopaminergic System in a

189

Parkinson’s Disease Model Induced by MPTP.” Scientific Reports 6 (September):

32656. https://doi.org/10.1038/srep32656.

Fitzpatrick, David A., Christopher J. Creevey, and James O. McInerney. 2006. “Genome

Phylogenies Indicate a Meaningful Alpha-Proteobacterial Phylogeny and Support

a Grouping of the Mitochondria with the Rickettsiales.” Molecular Biology and

Evolution 23 (1): 74–85. https://doi.org/10.1093/molbev/msj009.

Ford, Bradley, Sean Boykevisch, Chen Zhao, Simone Kunzelmann, Dafna Bar-Sagi,

Christian Herrmann, and Nicolas Nassar. 2009. “Characterization of a Ras Mutant

with Identical GDP- and GTP-Bound Structures,.” Biochemistry 48 (48): 11449–

57. https://doi.org/10.1021/bi901479b.

Ford, M. G., S. Jenni, and J. Nunnari. 2011. “The Crystal Structure of Dynamin.” Nature

477 (September): 561–66. https://doi.org/10.1038/nature10441.

Francy, Christopher A., Frances J. D. Alvarez, Louie Zhou, Rajesh Ramachandran, and

Jason A. Mears. 2015. “The Mechanoenzymatic Core of Dynamin-Related Protein

1 Comprises the Minimal Machinery Required for Membrane Constriction.”

Journal of Biological Chemistry 290 (18): 11692–703.

https://doi.org/10.1074/jbc.M114.610881.

Francy, Christopher A., Ryan W. Clinton, Chris Fröhlich, Colleen Murphy, and Jason A.

Mears. 2017. “Cryo-EM Studies of Drp1 Reveal Cardiolipin Interactions That

Activate the Helical Oligomer.” Scientific Reports 7 (1): 10744.

https://doi.org/10.1038/s41598-017-11008-3.

190

Frezza, C., S. Cipolat, O. Martins de Brito, M. Micaroni, G. V. Beznoussenko, T. Rudka, D.

Bartoli, et al. 2006. “OPA1 Controls Apoptotic Cristae Remodeling Independently

from Mitochondrial Fusion.” Cell 126 (July): 177–89.

https://doi.org/10.1016/j.cell.2006.06.025.

Friedman, J. R., L. L. Lackner, M. West, J. R. DiBenedetto, J. Nunnari, and G. K. Voeltz.

2011. “ER Tubules Mark Sites of Mitochondrial Division.” Science 334 (October):

358–62. https://doi.org/10.1126/science.1207385.

Fröhlich, Chris, Stefan Grabiger, David Schwefel, Katja Faelber, Eva Rosenbaum, Jason

Mears, Oliver Rocks, and Oliver Daumke. 2013. “Structural Insights into

Oligomerization and Mitochondrial Remodelling of Dynamin 1‐like Protein.” The

EMBO Journal 32 (9): 1280–92. https://doi.org/10.1038/emboj.2013.74.

Gambin, Yann, Myriam Reffay, Emma Sierecki, Fabrice Homblé, Robert S. Hodges, Nir S.

Gov, Nicolas Taulier, and Wladimir Urbach. 2010. “Variation of the Lateral

Mobility of Transmembrane Peptides with Hydrophobic Mismatch.” The Journal

of Physical Chemistry B 114 (10): 3559–66. https://doi.org/10.1021/jp911354y.

Gandre-Babbe, S., and A. M. van der Bliek. 2008. “The Novel Tail-Anchored Membrane

Protein Mff Controls Mitochondrial and Peroxisomal Fission in Mammalian

Cells.” Mol Biol Cell 19 (June): 2402–12. https://doi.org/10.1091/mbc.E07-12-

1287.

Gao, Ju, Siyue Qin, and Chang’an Jiang. 2015. “Parkin-Induced Ubiquitination of Mff

Promotes Its Association with P62/SQSTM1 during Mitochondrial

191

Depolarization.” Acta Biochimica et Biophysica Sinica 47 (7): 522–29.

https://doi.org/10.1093/abbs/gmv044.

Gawlowski, T., J. Suarez, B. Scott, M. Torres-Gonzalez, H. Wang, R. Schwappacher, X.

Han, JR 3rd Yates, M. Hoshijima, and W. Dillmann. 2012. “Modulation of

Dynamin-Related Protein 1 (DRP1) Function by Increased O-Linked-Beta-N-

Acetylglucosamine Modification (O-GlcNAc) in Cardiac Myocytes.” J Biol Chem

287 (August): 30024–34. https://doi.org/10.1074/jbc.M112.390682.

Gebert, N., A. S. Joshi, S. Kutik, T. Becker, M. McKenzie, X. L. Guan, V. P. Mooga, et al.

2009. “Mitochondrial Cardiolipin Involved in Outer-Membrane Protein

Biogenesis: Implications for Barth Syndrome.” Curr Biol 19 (December): 2133–39.

https://doi.org/10.1016/j.cub.2009.10.074.

Gerber, Sylvie, Majida Charif, Arnaud Chevrollier, Tanguy Chaumette, Claire Angebault,

Mariame Selma Kane, Aurélien Paris, et al. 2017. “Mutations in DNM1L, as in

OPA1, Result Indominant Optic Atrophy despite Opposite Effectson

Mitochondrial Fusion and Fission.” Brain 140 (10): 2586–96.

https://doi.org/10.1093/brain/awx219.

Glater, Elizabeth E., Laura J. Megeath, R. Steven Stowers, and Thomas L. Schwarz. 2006.

“Axonal Transport of Mitochondria Requires Milton to Recruit Kinesin Heavy

Chain and Is Light Chain Independent.” The Journal of Cell Biology 173 (4): 545–

57. https://doi.org/10.1083/jcb.200601067.

Glauser Liliane, Sonnay Sarah, Stafa Klodjan, and Moore Darren J. 2011. “Parkin

Promotes the Ubiquitination and Degradation of the Mitochondrial Fusion Factor

192

Mitofusin 1.” Journal of Neurochemistry 118 (4): 636–45.

https://doi.org/10.1111/j.1471-4159.2011.07318.x.

Goldstein, J. C., C. Muñoz-Pinedo, J.-E. Ricci, S. R. Adams, A. Kelekar, M. Schuler, R. Y.

Tsien, and D. R. Green. 2005. “Cytochrome c Is Released in a Single Step during

Apoptosis.” Cell Death and Differentiation 12 (5): 453–62.

https://doi.org/10.1038/sj.cdd.4401596.

Goldstein, J. C., N. J. Waterhouse, P. Juin, G. I. Evan, and D. R. Green. 2000. “The

Coordinate Release of Cytochrome c during Apoptosis Is Rapid, Complete and

Kinetically Invariant.” Nat Cell Biol 2 (March): 156–62.

https://doi.org/10.1038/35004029.

Griffin, E. E., J. Graumann, and D. C. Chan. 2005. “The WD40 Protein Caf4p Is a

Component of the Mitochondrial Fission Machinery and Recruits Dnm1p to

Mitochondria.” J Cell Biol 170 (July): 237–48.

https://doi.org/10.1083/jcb.200503148.

Grigliatti, T. A., L. Hall, R. Rosenbluth, and D. T. Suzuki. 1973. “Temperature-Sensitive

Mutations in Drosophila Melanogaster. XIV. A Selection of Immobile Adults.” Mol

Gen Genet 120 (January): 107–14.

Guo, Chun, Kevin A. Wilkinson, Ashley J. Evans, Philip P. Rubin, and Jeremy M. Henley.

2017. “SENP3-Mediated DeSUMOylation of Drp1 Facilitates Interaction with Mff

to Promote Cell Death.” Scientific Reports 7 (March): 43811.

https://doi.org/10.1038/srep43811.

193

Guo, Qian, Sajjan Koirala, Edward M. Perkins, J. Michael McCaffery, and Janet M. Shaw.

2012. “The Mitochondrial Fission Adaptors Caf4 and Mdv1 Are Not Functionally

Equivalent.” PLOS ONE 7 (12): e53523.

https://doi.org/10.1371/journal.pone.0053523.

Guo, X., M. H. Disatnik, M. Monbureau, M. Shamloo, D. Mochly-Rosen, and X. Qi. 2013.

“Inhibition of Mitochondrial Fragmentation Diminishes Huntington’s Disease-

Associated Neurodegeneration.” J Clin Invest 123 (December): 5371–88.

https://doi.org/10.1172/JCI70911.

Guo, Xing, Marie-Helene Disatnik, Marie Monbureau, Mehrdad Shamloo, Daria Mochly-

Rosen, and Xin Qi. 2013. “Inhibition of Mitochondrial Fragmentation Diminishes

Huntington’s Disease-Associated Neurodegeneration.” The Journal of Clinical

Investigation 123 (12): 5371–88. https://doi.org/10.1172/JCI70911.

Hales, Karen G., and Margaret T. Fuller. 1997. “Developmentally Regulated

Mitochondrial Fusion Mediated by a Conserved, Novel, Predicted GTPase.” Cell

90 (1): 121–29. https://doi.org/10.1016/S0092-8674(00)80319-0.

Hatefi, Y., A. G. Haavik, L. R. Fowler, and D. E. Griffiths. 1962. “Studies on the Electron

Transfer System XLII. RECONSTITUTION OF THE ELECTRON TRANSFER SYSTEM.”

Journal of Biological Chemistry 237 (8): 2661–69.

Hatefi, Y., A. G. Haavik, and D. E. Griffiths. 1961. “Reconstitution of the Electron

Transport System: I. Preparation and Properties of the Interacting Enzyme

Complexes.” Biochemical and Biophysical Research Communications 4 (6): 441–

46. https://doi.org/10.1016/0006-291X(61)90305-9.

194

Heinemeyer, Jesco, Hans-Peter Braun, Egbert J. Boekema, and Roman Kouřil. 2007. “A

Structural Model of the Cytochrome c Reductase/Oxidase Supercomplex from

Yeast Mitochondria.” Journal of Biological Chemistry 282 (16): 12240–48.

https://doi.org/10.1074/jbc.M610545200.

Helle, Sebastian Carsten Johannes, Qian Feng, Mathias J. Aebersold, Luca Hirt, Raphael

R. Grüter, Afshin Vahid, Andrea Sirianni, et al. 2017. “Mechanical Force Induces

Mitochondrial Fission.” ELife 6 (November): e30292.

https://doi.org/10.7554/eLife.30292.

Hollenbeck, P. J., and W. M. Saxton. 2005. “The Axonal Transport of Mitochondria.” J

Cell Sci 118 (December): 5411–19. https://doi.org/10.1242/jcs.02745.

Horisberger, M. A., P. Staeheli, and O. Haller. 1983. “Interferon Induces a Unique Protein

in Mouse Cells Bearing a Gene for Resistance to Influenza Virus.” Proceedings of

the National Academy of Sciences 80 (7): 1910–14.

https://doi.org/10.1073/pnas.80.7.1910.

Ingerman, E., E. M. Perkins, M. Marino, J. A. Mears, J. M. McCaffery, J. E. Hinshaw, and J.

Nunnari. 2005. “Dnm1 Forms Spirals That Are Structurally Tailored to Fit

Mitochondria.” J Cell Biol 170 (September): 1021–27.

https://doi.org/10.1083/jcb.200506078.

Ishihara, N., M. Nomura, A. Jofuku, H. Kato, S. O. Suzuki, K. Masuda, H. Otera, et al.

2009. “Mitochondrial Fission Factor Drp1 Is Essential for Embryonic

Development and Synapse Formation in Mice.” Nat Cell Biol 11 (August): 958–66.

https://doi.org/10.1038/ncb1907.

195

Ishihara, Naotada, Yuu Fujita, Toshihiko Oka, and Katsuyoshi Mihara. 2006. “Regulation

of Mitochondrial Morphology through Proteolytic Cleavage of OPA1.” The EMBO

Journal 25 (13): 2966–77. https://doi.org/10.1038/sj.emboj.7601184.

James, D. I., P. A. Parone, Y. Mattenberger, and J. C. Martinou. 2003. “HFis1, a Novel

Component of the Mammalian Mitochondrial Fission Machinery.” J Biol Chem

278 (September): 36373–79. https://doi.org/10.1074/jbc.M303758200.

Jamshad, Mohammed, Yu-Pin Lin, Timothy J. Knowles, Rosemary A. Parslow, Craig

Harris, Mark Wheatley, David R. Poyner, et al. 2011. “Surfactant-Free Purification

of Membrane Proteins with Intact Native Membrane Environment.” Biochemical

Society Transactions 39 (3): 813–18. https://doi.org/10.1042/BST0390813.

Ji, W. K., A. L. Hatch, R. A. Merrill, S. Strack, and H. N. Higgs. 2015. “Actin Filaments

Target the Oligomeric Maturation of the Dynamin GTPase Drp1 to Mitochondrial

Fission Sites.” Elife 4: e11553. https://doi.org/10.7554/eLife.11553.

Ji, Wei-Ke, Rajarshi Chakrabarti, Xintao Fan, Lori Schoenfeld, Stefan Strack, and Henry N.

Higgs. 2017. “Receptor-Mediated Drp1 Oligomerization on Endoplasmic

Reticulum.” J Cell Biol, November, jcb.201610057.

https://doi.org/10.1083/jcb.201610057.

Jin, Seok Min, Michael Lazarou, Chunxin Wang, Lesley A. Kane, Derek P. Narendra, and

Richard J. Youle. 2010. “Mitochondrial Membrane Potential Regulates PINK1

Import and Proteolytic Destabilization by PARL.” The Journal of Cell Biology 191

(5): 933–42. https://doi.org/10.1083/jcb.201008084.

196

Jin, Seok Min, and Richard J Youle. 2013. “The Accumulation of Misfolded Proteins in the

Mitochondrial Matrix Is Sensed by PINK1 to Induce PARK2/Parkin-Mediated

Mitophagy of Polarized Mitochondria.” Autophagy 9 (11): 1750–57.

https://doi.org/10.4161/auto.26122.

Johnson, Catherine, Michele Tinti, Nicola T. Wood, David G. Campbell, Rachel Toth,

Fanny Dubois, Kathryn M. Geraghty, et al. 2011. “Visualization and Biochemical

Analyses of the Emerging Mammalian 14-3-3-Phosphoproteome.” Molecular &

Cellular Proteomics 10 (10): M110.005751.

https://doi.org/10.1074/mcp.M110.005751.

Jonckheere, An I., Jan A. M. Smeitink, and Richard J. T. Rodenburg. 2012. “Mitochondrial

ATP Synthase: Architecture, Function and Pathology.” Journal of Inherited

Metabolic Disease 35 (2): 211–25. https://doi.org/10.1007/s10545-011-9382-9.

Ju, W. K., Q. Liu, K. Y. Kim, J. G. Crowston, J. D. Lindsey, N. Agarwal, M. H. Ellisman, G. A.

Perkins, and R. N. Weinreb. 2007. “Elevated Hydrostatic Pressure Triggers

Mitochondrial Fission and Decreases Cellular ATP in Differentiated RGC-5 Cells.”

Invest Ophthalmol Vis Sci 48 (May): 2145–51. https://doi.org/10.1167/iovs.06-

0573.

Kalia, Raghav, Ray Y.-R. Wang, Ali Yusuf, Paul V. Thomas, David A. Agard, Janet M. Shaw,

and Adam Frost. 2017a. “Structural Basis of Mitochondrial Receptor Binding and

GTP Driven Conformational Constriction by Dynamin-Related Protein 1.” BioRxiv,

August, 172809. https://doi.org/10.1101/172809.

197

———. 2017b. “Structural Basis of Mitochondrial Receptor Binding and GTP Driven

Conformational Constriction by Dynamin-Related Protein 1.” BioRxiv, August,

172809. https://doi.org/10.1101/172809.

Kane, Melissa, Shalini S. Yadav, Julia Bitzegeio, Sebla B. Kutluay, Trinity Zang, Sam J.

Wilson, John W. Schoggins, et al. 2013. “MX2 Is an Interferon-Induced Inhibitor

of HIV-1 Infection.” Nature 502 (7472): 563–66.

https://doi.org/10.1038/nature12653.

Kashatus, J. A., A. Nascimento, L. J. Myers, A. Sher, F. L. Byrne, K. L. Hoehn, C. M.

Counter, and D. F. Kashatus. 2015. “Erk2 Phosphorylation of Drp1 Promotes

Mitochondrial Fission and MAPK-Driven Tumor Growth.” Mol Cell 57 (February):

537–51. https://doi.org/10.1016/j.molcel.2015.01.002.

Katta, V., and B. T. Chait. 1991. “Conformational Changes in Proteins Probed by

Hydrogen-Exchange Electrospray-Ionization Mass Spectrometry.” Rapid

Communications in Mass Spectrometry: RCM 5 (4): 214–17.

https://doi.org/10.1002/rcm.1290050415.

Kim, Young Yeon, Seong-Hoon Yun, and Jeanho Yun. 2018. “Downregulation of Drp1, a

Fission Regulator, Is Associated with Human Lung and Colon Cancers.” Acta

Biochimica Et Biophysica Sinica 50 (2): 209–15.

https://doi.org/10.1093/abbs/gmx137.

Kiselar, J. G., S. D. Maleknia, M. Sullivan, K. M. Downard, and M. R. Chance. 2002.

“Hydroxyl Radical Probe of Protein Surfaces Using Synchrotron X-Ray Radiolysis

198

and Mass Spectrometry.” International Journal of Radiation Biology 78 (2): 101–

14. https://doi.org/10.1080/09553000110094805.

Kiselar, Janna G., Manish Datt, Mark R. Chance, and Michael A. Weiss. 2011. “Structural

Analysis of Proinsulin Hexamer Assembly by Hydroxyl Radical Footprinting and

Computational Modeling.” Journal of Biological Chemistry 286 (51): 43710–16.

https://doi.org/10.1074/jbc.M111.297853.

Kiselar, Janna G., Paul A. Janmey, Steven C. Almo, and Mark R. Chance. 2003. “Structural

Analysis of Gelsolin Using Synchrotron Protein Footprinting.” Molecular &

Cellular Proteomics 2 (10): 1120–32. https://doi.org/10.1074/mcp.M300068-

MCP200.

Klinger, Alexandra L., Janna Kiselar, Serguei Ilchenko, Hiroaki Komatsu, Mark R. Chance,

and Paul H. Axelsen. 2014. “A Synchrotron-Based Hydroxyl Radical Footprinting

Analysis of Amyloid Fibrils and Prefibrillar Intermediates with Residue-Specific

Resolution.” Biochemistry 53 (49): 7724–34. https://doi.org/10.1021/bi5010409.

Klingler, Johannes, Carolyn Vargas, Sebastian Fiedler, and Sandro Keller. 2015.

“Preparation of Ready-to-Use Small Unilamellar Phospholipid Vesicles by

Ultrasonication with a Beaker Resonator.” Analytical Biochemistry 477 (May):

10–12. https://doi.org/10.1016/j.ab.2015.02.015.

Knott, A. B., G. Perkins, R. Schwarzenbacher, and E. Bossy-Wetzel. 2008. “Mitochondrial

Fragmentation in Neurodegeneration.” Nat Rev Neurosci 9 (July): 505–18.

https://doi.org/10.1038/nrn2417.

199

Koch, Annett, Meinolf Thiemann, Markus Grabenbauer, Yisang Yoon, Mark A. McNiven,

and Michael Schrader. 2003. “Dynamin-like Protein 1 Is Involved in Peroxisomal

Fission.” Journal of Biological Chemistry 278 (10): 8597–8605.

https://doi.org/10.1074/jbc.M211761200.

Kochs, G., and O. Haller. 1999a. “GTP-Bound Human MxA Protein Interacts with the

Nucleocapsids of Thogoto Virus (Orthomyxoviridae).” J Biol Chem 274 (February):

4370–76.

———. 1999b. “Interferon-Induced Human MxA GTPase Blocks Nuclear Import of

Thogoto Virus Nucleocapsids.” Proc Natl Acad Sci U S A 96 (March): 2082–86.

Koirala, S., Q. Guo, R. Kalia, H. T. Bui, D. M. Eckert, A. Frost, and J. M. Shaw. 2013.

“Interchangeable Adaptors Regulate Mitochondrial Dynamin Assembly for

Membrane Scission.” Proc Natl Acad Sci U S A 110 (April): E1342-51.

https://doi.org/10.1073/pnas.1300855110.

Kothakota, Srinivas, Toshifumi Azuma, Christoph Reinhard, Anke Klippel, Jay Tang,

Keting Chu, Thomas J. McGarry, et al. 1997. “Caspase-3-Generated Fragment of

Gelsolin: Effector of Morphological Change in Apoptosis.” Science 278 (5336):

294–98. https://doi.org/10.1126/science.278.5336.294.

Krauss Stefan. 2001. “Mitochondria: Structure and Role in Respiration.” ELS, Major

Reference Works, , April. https://doi.org/10.1038/npg.els.0001380.

Kroon, Anton I. P. M de, Danièle Dolis, Andreas Mayer, Roland Lill, and Ben de Kruijff.

1997. “Phospholipid Composition of Highly Purified Mitochondrial Outer

Membranes of Rat Liver and Neurospora Crassa. Is Cardiolipin Present in the

200

Mitochondrial Outer Membrane?” Biochimica et Biophysica Acta (BBA) -

Biomembranes 1325 (1): 108–16. https://doi.org/10.1016/S0005-

2736(96)00240-4.

Kubalek, Elizabeth W., Stuart F. J. Le Grice, and Patrick O. Brown. 1994. “Two-

Dimensional Crystallization of Histidine-Tagged, HIV-1 Reverse Transcriptase

Promoted by a Novel Nickel-Chelating Lipid.” Journal of Structural Biology 113

(2): 117–23. https://doi.org/10.1006/jsbi.1994.1039.

Lackner, L. L., and J. Nunnari. 2010. “Small Molecule Inhibitors of Mitochondrial Division:

Tools That Translate Basic Biological Research into Medicine.” Chem Biol 17

(June): 578–83. https://doi.org/10.1016/j.chembiol.2010.05.016.

Lee, Jason E., Laura M. Westrate, Haoxi Wu, Cynthia Page, and Gia K. Voeltz. 2016.

“Multiple Dynamin Family Members Collaborate to Drive Mitochondrial

Division.” Nature 540 (7631): 139–43. https://doi.org/10.1038/nature20555.

Lenaz, Giorgio, Romana Fato, Maria Luisa Genova, Christian Bergamini, Cristina Bianchi,

and Annalisa Biondi. 2006. “Mitochondrial Complex I: Structural and Functional

Aspects.” Biochimica et Biophysica Acta (BBA) - Bioenergetics, Mitochondria:

from Molecular Insight to Physiology and Pathology, 1757 (9): 1406–20.

https://doi.org/10.1016/j.bbabio.2006.05.007.

Leonard, M., B. D. Song, R. Ramachandran, and S. L. Schmid. 2005. “Robust Colorimetric

Assays for Dynamin’s Basal and Stimulated GTPase Activities.” Methods Enzymol

404: 490–503. https://doi.org/10.1016/S0076-6879(05)04043-7.

201

Lewis, Samantha C., Lauren F. Uchiyama, and Jodi Nunnari. 2016. “ER-Mitochondria

Contacts Couple MtDNA Synthesis with Mitochondrial Division in Human Cells.”

Science 353 (6296): aaf5549. https://doi.org/10.1126/science.aaf5549.

Li, Edwin, William C. Wimley, and Kalina Hristova. 2012. “Transmembrane Helix

Dimerization: Beyond the Search for Sequence Motifs.” Biochimica et Biophysica

Acta (BBA) - Biomembranes, Membrane protein structure and function, 1818 (2):

183–93. https://doi.org/10.1016/j.bbamem.2011.08.031.

Lindenmann, J. 1962. “Resistance of Mice to Mouse-Adapted Influenza A Virus.”

Virology 16 (February): 203–4.

Lindenmann, Jean, Christine A. Lane, and Derek Hobson. 1963. “The Resistance of A2G

Mice to Myxoviruses.” The Journal of Immunology 90 (6): 942–51.

Liu, J., Q. Dai, J. Chen, D. Durrant, A. Freeman, T. Liu, D. Grossman, and R. M. Lee. 2003.

“Phospholipid Scramblase 3 Controls Mitochondrial Structure, Function, and

Apoptotic Response.” Mol Cancer Res 1 (October): 892–902.

Liu, Raymond, and David C. Chan. 2015. “The Mitochondrial Fission Receptor Mff

Selectively Recruits Oligomerized Drp1.” Molecular Biology of the Cell 26 (24):

4466–77. https://doi.org/10.1091/mbc.E15-08-0591.

Liu, Zhenlong, Qinghua Pan, Shilei Ding, Jin Qian, Fengwen Xu, Jinming Zhou, Shan Cen,

Fei Guo, and Chen Liang. 2013. “The Interferon-Inducible MxB Protein Inhibits

HIV-1 Infection.” Cell Host & Microbe 14 (4): 398–410.

https://doi.org/10.1016/j.chom.2013.08.015.

202

Loson, O. C., R. Liu, M. E. Rome, S. Meng, J. T. Kaiser, S. O. Shan, and D. C. Chan. 2014.

“The Mitochondrial Fission Receptor MiD51 Requires ADP as a Cofactor.”

Structure 22 (March): 367–77. https://doi.org/10.1016/j.str.2014.01.001.

Loson, O. C., Z. Song, H. Chen, and D. C. Chan. 2013. “Fis1, Mff, MiD49, and MiD51

Mediate Drp1 Recruitment in Mitochondrial Fission.” Mol Biol Cell 24 (March):

659–67. https://doi.org/10.1091/mbc.E12-10-0721.

Losón Oliver C., Meng Shuxia, Ngo Huu, Liu Raymond, Kaiser Jens T., and Chan David C.

2015. “Crystal Structure and Functional Analysis of MiD49, a Receptor for the

Mitochondrial Fission Protein Drp1.” Protein Science 24 (3): 386–94.

https://doi.org/10.1002/pro.2629.

Luo, Yu, Alan Hoffer, Barry Hoffer, and Xin Qi. 2015. “Mitochondria: A Therapeutic

Target for Parkinson’s Disease?” International Journal of Molecular Sciences 16

(9): 20704–30. https://doi.org/10.3390/ijms160920704.

Macdonald, Patrick J., Christopher A. Francy, Natalia Stepanyants, Lance Lehman,

Anthony Baglio, Jason A. Mears, Xin Qi, and Rajesh Ramachandran. 2016.

“Distinct Splice Variants of Dynamin-Related Protein 1 Differentially Utilize

Mitochondrial Fission Factor as an Effector of Cooperative GTPase Activity.”

Journal of Biological Chemistry 291 (1): 493–507.

https://doi.org/10.1074/jbc.M115.680181.

Macdonald, Patrick J., Natalia Stepanyants, Niharika Mehrotra, Jason A. Mears, Xin Qi,

Hiromi Sesaki, and Rajesh Ramachandran. 2014. “A Dimeric Equilibrium

Intermediate Nucleates Drp1 Reassembly on Mitochondrial Membranes for

203

Fission.” Molecular Biology of the Cell 25 (12): 1905–15.

https://doi.org/10.1091/mbc.E14-02-0728.

MacLennan, David H., and Junpei Asai. 1968. “Studies on the Mitochondrial Adenosine

Triphosphatase System V. Localization of the Oligomycin-Sensitivity Conferring

Protein.” Biochemical and Biophysical Research Communications 33 (3): 441–47.

https://doi.org/10.1016/0006-291X(68)90592-5.

Mallat, A., L. F. Uchiyama, S. C. Lewis, R. A. Fredenburg, Y. Terada, N. Ji, J. Nunnari, and

C. C. Tseng. 2018. “Discovery and Characterization of Selective Small Molecule

Inhibitors of the Mammalian Mitochondrial Division Dynamin, DRP1.”

Biochemical and Biophysical Research Communications 499 (3): 556–62.

https://doi.org/10.1016/j.bbrc.2018.03.189.

Mattie, Sevan, Jan Riemer, Jeremy G. Wideman, and Heidi M. McBride. 2017. “A New

Mitofusin Topology Places the Redox-Regulated C Terminus in the Mitochondrial

Intermembrane Space.” J Cell Biol, December, jcb.201611194.

https://doi.org/10.1083/jcb.201611194.

Mears, J. A., and J. E. Hinshaw. 2008. “Visualization of Dynamins.” Methods Cell Biol 88:

237–56. https://doi.org/10.1016/S0091-679X(08)00413-5.

Mears, J. A., L. L. Lackner, S. Fang, E. Ingerman, J. Nunnari, and J. E. Hinshaw. 2011.

“Conformational Changes in Dnm1 Support a Contractile Mechanism for

Mitochondrial Fission.” Nat Struct Mol Biol 18 (January): 20–26.

https://doi.org/10.1038/nsmb.1949.

204

Mileykovskaya, Eugenia, and William Dowhan. 2009. “Cardiolipin Membrane Domains in

Prokaryotes and Eukaryotes.” Biochimica et Biophysica Acta 1788 (10): 2084–91.

https://doi.org/10.1016/j.bbamem.2009.04.003.

Mitchell, Peter. 1961. “Coupling of Phosphorylation to Electron and Hydrogen Transfer

by a Chemi-Osmotic Type of Mechanism.” Nature 191 (4784): 144–48.

https://doi.org/10.1038/191144a0.

Montessuit, S., S. P. Somasekharan, O. Terrones, S. Lucken-Ardjomande, S. Herzig, R.

Schwarzenbacher, D. J. Manstein, et al. 2010. “Membrane Remodeling Induced

by the Dynamin-Related Protein Drp1 Stimulates Bax Oligomerization.” Cell 142

(September): 889–901. https://doi.org/10.1016/j.cell.2010.08.017.

Moscho, A., O. Orwar, D. T. Chiu, B. P. Modi, and R. N. Zare. 1996. “Rapid Preparation of

Giant Unilamellar Vesicles.” Proceedings of the National Academy of Sciences 93

(21): 11443–47.

Nakamura, N., Y. Kimura, M. Tokuda, S. Honda, and S. Hirose. 2006. “MARCH-V Is a

Novel Mitofusin 2- and Drp1-Binding Protein Able to Change Mitochondrial

Morphology.” EMBO Rep 7 (October): 1019–22.

https://doi.org/10.1038/sj.embor.7400790.

Nakamura, T., P. Cieplak, D. H. Cho, A. Godzik, and S. A. Lipton. 2010. “S-Nitrosylation of

Drp1 Links Excessive Mitochondrial Fission to Neuronal Injury in

Neurodegeneration.” Mitochondrion 10 (August): 573–78.

https://doi.org/10.1016/j.mito.2010.04.007.

205

Narendra, D. P., S. M. Jin, A. Tanaka, D. F. Suen, C. A. Gautier, J. Shen, M. R. Cookson,

and R. J. Youle. 2010. “PINK1 Is Selectively Stabilized on Impaired Mitochondria

to Activate Parkin.” PLoS Biol 8 (January): e1000298.

https://doi.org/10.1371/journal.pbio.1000298.

Narendra, Derek, Atsushi Tanaka, Der-Fen Suen, and Richard J. Youle. 2008. “Parkin Is

Recruited Selectively to Impaired Mitochondria and Promotes Their Autophagy.”

The Journal of Cell Biology 183 (5): 795–803.

https://doi.org/10.1083/jcb.200809125.

Naylor, K., E. Ingerman, V. Okreglak, M. Marino, J. E. Hinshaw, and J. Nunnari. 2006.

“Mdv1 Interacts with Assembled Dnm1 to Promote Mitochondrial Division.” J

Biol Chem 281 (January): 2177–83. https://doi.org/10.1074/jbc.M507943200.

Neumann, Sylvia, Thomas J. Pucadyil, and Sandra L. Schmid. 2013. “Analyzing

Membrane Remodeling and Fission Using Supported Bilayers with Excess

Membrane Reservoir.” Nature Protocols 8 (1): 213–22.

https://doi.org/10.1038/nprot.2012.152.

Notredame, C., D. G. Higgins, and J. Heringa. 2000. “T-Coffee: A Novel Method for Fast

and Accurate Multiple Sequence Alignment.” Journal of Molecular Biology 302

(1): 205–17. https://doi.org/10.1006/jmbi.2000.4042.

Ong, S. B., and D. J. Hausenloy. 2010. “Mitochondrial Morphology and Cardiovascular

Disease.” Cardiovasc Res 88 (October): 16–29.

https://doi.org/10.1093/cvr/cvq237.

206

Ong, S. B., S. Subrayan, S. Y. Lim, D. M. Yellon, S. M. Davidson, and D. J. Hausenloy. 2010.

“Inhibiting Mitochondrial Fission Protects the Heart against

Ischemia/Reperfusion Injury.” Circulation 121 (May): 2012–22.

https://doi.org/10.1161/CIRCULATIONAHA.109.906610.

Ordureau, Alban, Shireen A. Sarraf, David M. Duda, Jin-Mi Heo, Mark P. Jedrychowski,

Vladislav O. Sviderskiy, Jennifer L. Olszewski, et al. 2014. “Quantitative

Proteomics Reveal a Feedforward Mechanism for Mitochondrial PARKIN

Translocation and Ubiquitin Chain Synthesis.” Molecular Cell 56 (3): 360–75.

https://doi.org/10.1016/j.molcel.2014.09.007.

Osellame, Laura D., Abeer P. Singh, David A. Stroud, Catherine S. Palmer, Diana

Stojanovski, Rajesh Ramachandran, and Michael T. Ryan. 2016. “Cooperative and

Independent Roles of Drp1 Adaptors Mff and MiD49/51 in Mitochondrial

Fission.” J Cell Sci, January, jcs.185165. https://doi.org/10.1242/jcs.185165.

Otera, H., and K. Mihara. 2011. “Discovery of the Membrane Receptor for Mitochondrial

Fission GTPase Drp1.” Small GTPases 2 (May): 167–72.

https://doi.org/10.4161/sgtp.2.3.16486.

Otera, H, N Miyata, O Kuge, and K Mihara. 2016. “Drp1-Dependent Mitochondrial

Fission via MiD49/51 Is Essential for Apoptotic Cristae Remodeling.” J Cell Biol

212 (5): 531–44. https://doi.org/10.1083/jcb.201508099.

Otera, H., C. Wang, M. M. Cleland, K. Setoguchi, S. Yokota, R. J. Youle, and K. Mihara.

2010. “Mff Is an Essential Factor for Mitochondrial Recruitment of Drp1 during

207

Mitochondrial Fission in Mammalian Cells.” J Cell Biol 191 (December): 1141–58.

https://doi.org/10.1083/jcb.201007152.

Palmer, C. S., L. D. Osellame, D. Laine, O. S. Koutsopoulos, A. E. Frazier, and M. T. Ryan.

2011. “MiD49 and MiD51, New Components of the Mitochondrial Fission

Machinery.” EMBO Rep 12 (June): 565–73.

https://doi.org/10.1038/embor.2011.54.

Park, S. W., K. Y. Kim, J. D. Lindsey, Y. Dai, H. Heo, D. H. Nguyen, M. H. Ellisman, R. N.

Weinreb, and W. K. Ju. 2011. “A Selective Inhibitor of Drp1, Mdivi-1, Increases

Retinal Ganglion Cell Survival in Acute Ischemic Mouse Retina.” Invest

Ophthalmol Vis Sci 52 (April): 2837–43. https://doi.org/10.1167/iovs.09-5010.

Park, Yun Sun, Su Eun Choi, and Hyun Chul Koh. 2018. “PGAM5 Regulates PINK1/Parkin-

Mediated Mitophagy via DRP1 in CCCP-Induced Mitochondrial Dysfunction.”

Toxicology Letters 284 (March): 120–28.

https://doi.org/10.1016/j.toxlet.2017.12.004.

Pavlovic, J., H. A. Arzet, H. P. Hefti, M. Frese, D. Rost, B. Ernst, E. Kolb, P. Staeheli, and O.

Haller. 1995. “Enhanced Virus Resistance of Transgenic Mice Expressing the

Human MxA Protein.” Journal of Virology 69 (7): 4506–10.

Percy, Andrew J., Martial Rey, Kyle M. Burns, and David C. Schriemer. 2012. “Probing

Protein Interactions with Hydrogen/Deuterium Exchange and Mass

Spectrometry—A Review.” Analytica Chimica Acta 721 (April): 7–21.

https://doi.org/10.1016/j.aca.2012.01.037.

208

Pickrell, A. M., and R. J. Youle. 2015. “The Roles of PINK1, Parkin, and Mitochondrial

Fidelity in Parkinson’s Disease.” Neuron 85 (January): 257–73.

https://doi.org/10.1016/j.neuron.2014.12.007.

Pucadyil, T. J., and S. L. Schmid. 2008. “Real-Time Visualization of Dynamin-Catalyzed

Membrane Fission and Vesicle Release.” Cell 135 (December): 1263–75.

https://doi.org/10.1016/j.cell.2008.11.020.

Punjani, Ali, John L. Rubinstein, David J. Fleet, and Marcus A. Brubaker. 2017.

“CryoSPARC: Algorithms for Rapid Unsupervised Cryo-EM Structure

Determination.” Nature Methods 14 (3): 290–96.

https://doi.org/10.1038/nmeth.4169.

Qi, X., M. H. Disatnik, N. Shen, R. A. Sobel, and D. Mochly-Rosen. 2011. “Aberrant

Mitochondrial Fission in Neurons Induced by Protein Kinase C{delta} under

Oxidative Stress Conditions in Vivo.” Mol Biol Cell 22 (January): 256–65.

https://doi.org/10.1091/mbc.E10-06-0551.

Qi, X., N. Qvit, Y. C. Su, and D. Mochly-Rosen. 2013. “A Novel Drp1 Inhibitor Diminishes

Aberrant Mitochondrial Fission and Neurotoxicity.” J Cell Sci 126 (February):

789–802. https://doi.org/10.1242/jcs.114439.

Qian, Wei, Jingnan Wang, and Bennett Van Houten. 2013. “The Role of Dynamin-Related

Protein 1 in Cancer Growth: A Promising Therapeutic Target?” Expert Opinion on

Therapeutic Targets 17 (9): 997–1001.

https://doi.org/10.1517/14728222.2013.823160.

209

Ramachandran, Rajesh, and Sandra L. Schmid. 2018. “The Dynamin Superfamily.”

Current Biology 28 (8): R411–16. https://doi.org/10.1016/j.cub.2017.12.013.

Ramadurai, Sivaramakrishnan, Andrea Holt, Victor Krasnikov, Geert van den Bogaart, J.

Antoinette Killian, and Bert Poolman. 2009. “Lateral Diffusion of Membrane

Proteins.” Journal of the American Chemical Society 131 (35): 12650–56.

https://doi.org/10.1021/ja902853g.

Reubold, Thomas F., Katja Faelber, Nuria Plattner, York Posor, Katharina Ketel, Ute

Curth, Jeanette Schlegel, et al. 2015. “Crystal Structure of the Dynamin

Tetramer.” Nature 525 (7569): 404–8. https://doi.org/10.1038/nature14880.

Richter, V., C. S. Palmer, L. D. Osellame, A. P. Singh, K. Elgass, D. A. Stroud, H. Sesaki, M.

Kvansakul, and M. T. Ryan. 2014. “Structural and Functional Analysis of MiD51, a

Dynamin Receptor Required for Mitochondrial Fission.” J Cell Biol 204 (February):

477–86. https://doi.org/10.1083/jcb.201311014.

Ronni, T., K. Melén, A. Malygin, and I. Julkunen. 1993. “Control of IFN-Inducible MxA

Gene Expression in Human Cells.” The Journal of Immunology 150 (5): 1715–26.

Rosenbloom, Alyssa B., Sang-Hyuk Lee, Milton To, Antony Lee, Jae Yen Shin, and Carlos

Bustamante. 2014. “Optimized Two-Color Super Resolution Imaging of Drp1

during Mitochondrial Fission with a Slow-Switching Dronpa Variant.” Proceedings

of the National Academy of Sciences 111 (36): 13093–98.

https://doi.org/10.1073/pnas.1320044111.

210

Roux, A., K. Uyhazi, A. Frost, and P. De Camilli. 2006. “GTP-Dependent Twisting of

Dynamin Implicates Constriction and Tension in Membrane Fission.” Nature 441

(May): 528–31. https://doi.org/10.1038/nature04718.

Rutter, Jared, Dennis R. Winge, and Joshua D. Schiffman. 2010. “Succinate

Dehydrogenase—Assembly, Regulation and Role in Human Disease.”

Mitochondrion 10 (4): 393–401. https://doi.org/10.1016/j.mito.2010.03.001.

Sajic, Marija, Vincenzo Mastrolia, Chao Yu Lee, Diogo Trigo, Mona Sadeghian, Angelina J.

Mosley, Norman A. Gregson, Michael R. Duchen, and Kenneth J. Smith. 2013.

“Impulse Conduction Increases Mitochondrial Transport in Adult Mammalian

Peripheral Nerves In Vivo.” PLOS Biology 11 (12): e1001754.

https://doi.org/10.1371/journal.pbio.1001754.

Santel, A., and M. T. Fuller. 2001. “Control of Mitochondrial Morphology by a Human

Mitofusin.” Journal of Cell Science 114 (5): 867–74.

Scheres, S. H. 2012. “RELION: Implementation of a Bayesian Approach to Cryo-EM

Structure Determination.” J Struct Biol 180 (December): 519–30.

https://doi.org/10.1016/j.jsb.2012.09.006.

Schlame, Michael, and Mindong Ren. 2006. “Barth Syndrome, a Human Disorder of

Cardiolipin Metabolism.” FEBS Letters 580 (23): 5450–55.

https://doi.org/10.1016/j.febslet.2006.07.022.

Schlattner, U., M. Tokarska-Schlattner, S. Ramirez, Y. Y. Tyurina, A. A. Amoscato, D.

Mohammadyani, Z. Huang, et al. 2013. “Dual Function of Mitochondrial Nm23-

H4 Protein in Phosphotransfer and Intermembrane Lipid Transfer: A Cardiolipin-

211

Dependent Switch.” J Biol Chem 288 (January): 111–21.

https://doi.org/10.1074/jbc.M112.408633.

Seddon, Annela M., Paul Curnow, and Paula J. Booth. 2004. “Membrane Proteins, Lipids

and Detergents: Not Just a Soap Opera.” Biochimica et Biophysica Acta (BBA) -

Biomembranes, Lipid-Protein Interactions, 1666 (1): 105–17.

https://doi.org/10.1016/j.bbamem.2004.04.011.

Shamseldin, Hanan E., Muneera Alshammari, Tarfa Al-Sheddi, Mustafa A. Salih, Hisham

Alkhalidi, Amal Kentab, Gabriela M. Repetto, Mais Hashem, and Fowzan S.

Alkuraya. 2012. “Genomic Analysis of Mitochondrial Diseases in a

Consanguineous Population Reveals Novel Candidate Disease Genes.” Journal of

Medical Genetics 49 (4): 234–41. https://doi.org/10.1136/jmedgenet-2012-

100836.

Sheffer, Ruth, Liza Douiev, Simon Edvardson, Avraham Shaag, Khaled Tamimi, Devorah

Soiferman, Vardiella Meiner, and Ann Saada. 2016. “Postnatal Microcephaly and

Pain Insensitivity Due to a de Novo Heterozygous DNM1L Mutation Causing

Impaired Mitochondrial Fission and Function.” American Journal of Medical

Genetics Part A 170 (6): 1603–7. https://doi.org/10.1002/ajmg.a.37624.

Shin, Hye-Won, Chisa Shinotsuka, Seiji Torii, Kazuo Murakami, and Kazuhisa Nakayama.

1997. “Identification and Subcellular Localization of a Novel Mammalian

Dynamin-Related Protein Homologous to Yeast Vps1p and Dnm1p.” The Journal

of Biochemistry 122 (3): 525–30.

212

Shpetner, H. S., and R. B. Vallee. 1989. “Identification of Dynamin, a Novel

Mechanochemical Enzyme That Mediates Interactions between Microtubules.”

Cell 59 (November): 421–32.

Shrout, A. L., EA 3rd Esposito, and R. M. Weis. 2008. “Template-Directed Assembly of

Signaling Proteins: A Novel Drug Screening and Research Tool.” Chem Biol Drug

Des 71 (March): 278–81. https://doi.org/10.1111/j.1747-0285.2008.00627.x.

Shrout, AL, DJ Montefusco, and RM Weis. 2003. “Template-Directed Assembly of

Receptor Signaling Complexes†.” Rapid-communication. October 28, 2003.

https://doi.org/10.1021/bi0352769.

Silvestri, Laura, Viviana Caputo, Emanuele Bellacchio, Luigia Atorino, Bruno Dallapiccola,

Enza Maria Valente, and Giorgio Casari. 2005. “Mitochondrial Import and

Enzymatic Activity of PINK1 Mutants Associated to Recessive Parkinsonism.”

Human Molecular Genetics 14 (22): 3477–92.

https://doi.org/10.1093/hmg/ddi377.

Smirnova, E., D. L. Shurland, S. N. Ryazantsev, and A. M. van der Bliek. 1998. “A Human

Dynamin-Related Protein Controls the Distribution of Mitochondria.” J Cell Biol

143 (October): 351–58.

Stachowiak, Jeanne C., Eva M. Schmid, Christopher J. Ryan, Hyoung Sook Ann, Darryl Y.

Sasaki, Michael B. Sherman, Phillip L. Geissler, Daniel A. Fletcher, and Carl C.

Hayden. 2012. “Membrane Bending by Protein–protein Crowding.” Nature Cell

Biology 14 (9): 944–49. https://doi.org/10.1038/ncb2561.

213

Staeheli, P., O. Haller, W. Boll, J. Lindenmann, and C. Weissmann. 1986. “Mx Protein:

Constitutive Expression in 3T3 Cells Transformed with Cloned Mx CDNA Confers

Selective Resistance to Influenza Virus.” Cell 44 (1): 147–58.

Stepanyants, Natalia, Patrick J. Macdonald, Christopher A. Francy, Jason A. Mears, Xin

Qi, and Rajesh Ramachandran. 2015. “Cardiolipin’s Propensity for Phase

Transition and Its Reorganization by Dynamin-Related Protein 1 Form a Basis for

Mitochondrial Membrane Fission.” Molecular Biology of the Cell 26 (17): 3104–

16. https://doi.org/10.1091/mbc.E15-06-0330.

Strack, S., and J. T. Cribbs. 2012. “Allosteric Modulation of Drp1 Mechanoenzyme

Assembly and Mitochondrial Fission by the Variable Domain.” J Biol Chem 287

(March): 10990–1. https://doi.org/10.1074/jbc.M112.342105.

Strack, S., T. J. Wilson, and J. T. Cribbs. 2013. “Cyclin-Dependent Kinases Regulate Splice-

Specific Targeting of Dynamin-Related Protein 1 to Microtubules.” J Cell Biol 201

(June): 1037–51. https://doi.org/10.1083/jcb.201210045.

Su, Y. C., and X. Qi. 2013. “Inhibition of Excessive Mitochondrial Fission Reduced

Aberrant Autophagy and Neuronal Damage Caused by LRRK2 G2019S Mutation.”

Hum Mol Genet 22 (November): 4545–61. https://doi.org/10.1093/hmg/ddt301.

Sundborger, A. C., S. Fang, J. A. Heymann, P. Ray, J. S. Chappie, and J. E. Hinshaw. 2014.

“A Dynamin Mutant Defines a Superconstricted Prefission State.” Cell Rep 8

(August): 734–42. https://doi.org/10.1016/j.celrep.2014.06.054.

Sundborger, Anna C., Shunming Fang, Jürgen A. Heymann, Pampa Ray, Joshua S.

Chappie, and Jenny E. Hinshaw. 2014. “A Dynamin Mutant Defines a

214

Superconstricted Prefission State.” Cell Reports 8 (3): 734–42.

https://doi.org/10.1016/j.celrep.2014.06.054.

Taguchi, N., N. Ishihara, A. Jofuku, T. Oka, and K. Mihara. 2007. “Mitotic Phosphorylation

of Dynamin-Related GTPase Drp1 Participates in Mitochondrial Fission.” J Biol

Chem 282 (April): 11521–29. https://doi.org/10.1074/jbc.M607279200.

Tait, Stephen W. G., and Douglas R. Green. 2010. “Mitochondria and Cell Death: Outer

Membrane Permeabilization and Beyond.” Nature Reviews Molecular Cell

Biology 11 (9): 621–32. https://doi.org/10.1038/nrm2952.

Takamoto, Keiji, and Mark R. Chance. 2006. “Radiolytic Protein Footprinting with Mass

Spectrometry to Probe the Structure of Macromolecular Complexes.” Annual

Review of Biophysics and Biomolecular Structure 35 (1): 251–76.

https://doi.org/10.1146/annurev.biophys.35.040405.102050.

Tanaka, A., S. Kobayashi, and Y. Fujiki. 2006. “Peroxisome Division Is Impaired in a CHO

Cell Mutant with an Inactivating Point-Mutation in Dynamin-like Protein 1

Gene.” Exp Cell Res 312 (May): 1671–84.

https://doi.org/10.1016/j.yexcr.2006.01.028.

Tanwar, Deepak Kumar, Danitra J. Parker, Priyanka Gupta, Brian Spurlock, Ronald D.

Alvarez, Malay Kumar Basu, Kasturi Mitra, et al. 2016. “Crosstalk between the

Mitochondrial Fission Protein, Drp1, and the Cell Cycle Is Identified across

Various Cancer Types and Can Impact Survival of Epithelial Ovarian Cancer

Patients.” Oncotarget 7 (37): 60021–37.

https://doi.org/10.18632/oncotarget.11047.

215

Thrash, J. Cameron, Alex Boyd, Megan J. Huggett, Jana Grote, Paul Carini, Ryan J. Yoder,

Barbara Robbertse, Joseph W. Spatafora, Michael S. Rappé, and Stephen J.

Giovannoni. 2011. “Phylogenomic Evidence for a Common Ancestor of

Mitochondria and the SAR11 Clade.” Scientific Reports 1 (June): 13.

https://doi.org/10.1038/srep00013.

Tieu, Q., and J. Nunnari. 2000. “Mdv1p Is a WD Repeat Protein That Interacts with the

Dynamin-Related GTPase, Dnm1p, to Trigger Mitochondrial Division.” J Cell Biol

151 (October): 353–66.

Toyama, E, S Herzig, J Courchet, T Lewis, O Losón, Kristina Hellberg, Nathan P. Young, et

al. 2016. “AMP-Activated Protein Kinase Mediates Mitochondrial Fission in

Response to Energy Stress.” Science 351 (6270): 275–81.

https://doi.org/10.1126/science.aab4138.

Toyama, E. Q., S. Herzig, J. Courchet, TL Jr Lewis, O. C. Loson, K. Hellberg, N. P. Young, et

al. 2016. “Metabolism. AMP-Activated Protein Kinase Mediates Mitochondrial

Fission in Response to Energy Stress.” Science 351 (January): 275–81.

https://doi.org/10.1126/science.aab4138.

Ugarte-Uribe, B., H. M. Muller, M. Otsuki, W. Nickel, and A. J. Garcia-Saez. 2014.

“Dynamin-Related Protein 1 (Drp1) Promotes Structural Intermediates of

Membrane Division.” J Biol Chem 289 (October): 30645–56.

https://doi.org/10.1074/jbc.M114.575779.

Ugarte-Uribe, B., Coline Prévost, Kushal Kumar Das, Patricia Bassereau, and Ana J.

García-Sáez. 2017. “Drp1 Polymerization Stabilizes Curved Tubular Membranes

216

Similar to Those of Constricted Mitochondria.” J Cell Sci, January, jcs.208603.

https://doi.org/10.1242/jcs.208603.

Vanstone, J. R., A. M. Smith, S. McBride, T. Naas, M. Holcik, G. Antoun, M. E. Harper, et

al. 2016. “DNM1L-Related Mitochondrial Fission Defect Presenting as Refractory

Epilepsy.” Eur J Hum Genet 24 (July): 1084–88.

https://doi.org/10.1038/ejhg.2015.243.

Vénien-Bryan, Catherine, Fabrice Balavoine, Bertrand Toussaint, Charles Mioskowski,

Elizabeth A Hewat, Brigitte Helme, and Paulette M Vignais. 1997. “Structural

Study of the Response Regulator HupR from Rhodobacter Capsulatus. Electron

Microscopy of Two-Dimensional Crystals on a Nickel-Chelating Lipid11Edited by

W. Baumeister.” Journal of Molecular Biology 274 (5): 687–92.

https://doi.org/10.1006/jmbi.1997.1431.

Wakabayashi, J., Z. Zhang, N. Wakabayashi, Y. Tamura, M. Fukaya, T. W. Kensler, M.

Iijima, and H. Sesaki. 2009. “The Dynamin-Related GTPase Drp1 Is Required for

Embryonic and Brain Development in Mice.” J Cell Biol 186 (September): 805–16.

https://doi.org/10.1083/jcb.200903065.

Walde, Peter, Katia Cosentino, Helen Engel, and Pasquale Stano. 2010. “Giant Vesicles:

Preparations and Applications.” ChemBioChem 11 (7): 848–65.

https://doi.org/10.1002/cbic.201000010.

Wang, Hongxia, Pingping Song, Lei Du, Weili Tian, Wen Yue, Min Liu, Dengwen Li, et al.

2011. “Parkin Ubiquitinates Drp1 for Proteasome-Dependent Degradation:

Implication of Dysregulated Mitochondrial Dynamics in Parkinson Disease.” The

217

Journal of Biological Chemistry 286 (13): 11649–58.

https://doi.org/10.1074/jbc.M110.144238.

Wang, Ping, Peiguo Wang, Becky Liu, Jing Zhao, Qingsong Pang, Samir G. Agrawal, Li Jia,

et al. 2015. “Dynamin-Related Protein Drp1 Is Required for Bax Translocation to

Mitochondria in Response to Irradiation-Induced Apoptosis.” Oncotarget 6 (26):

22598–612. https://doi.org/10.18632/oncotarget.4200.

Waterham, H. R., J. Koster, C. W. van Roermund, P. A. Mooyer, R. J. Wanders, and J. V.

Leonard. 2007. “A Lethal Defect of Mitochondrial and Peroxisomal Fission.” N

Engl J Med 356 (April): 1736–41. https://doi.org/10.1056/NEJMoa064436.

Wenger, Julia, Eva Klinglmayr, Chris Fröhlich, Clarissa Eibl, Ana Gimeno, Manuel

Hessenberger, Sandra Puehringer, Oliver Daumke, and Peter Goettig. 2013.

“Functional Mapping of Human Dynamin-1-Like GTPase Domain Based on X-Ray

Structure Analyses.” PLOS ONE 8 (8): e71835.

https://doi.org/10.1371/journal.pone.0071835.

Westphal, Dana, Grant Dewson, Marie Menard, Paul Frederick, Sweta Iyer, Ray Bartolo,

Leonie Gibson, et al. 2014. “Apoptotic Pore Formation Is Associated with In-

Plane Insertion of Bak or Bax Central Helices into the Mitochondrial Outer

Membrane.” Proceedings of the National Academy of Sciences 111 (39): E4076–

85. https://doi.org/10.1073/pnas.1415142111.

Wilson-Kubalek, E. M., R. E. Brown, H. Celia, and R. A. Milligan. 1998. “Lipid Nanotubes

as Substrates for Helical Crystallization of Macromolecules.” Proc Natl Acad Sci U

S A 95 (July): 8040–45.

218

Wolter, Keith G., Yi-Te Hsu, Carolyn L. Smith, Amotz Nechushtan, Xu-Guang Xi, and

Richard J. Youle. 1997. “Movement of Bax from the Cytosol to Mitochondria

during Apoptosis.” The Journal of Cell Biology 139 (5): 1281–92.

https://doi.org/10.1083/jcb.139.5.1281.

Xie, Q., Q. Wu, C. M. Horbinski, W. A. Flavahan, K. Yang, W. Zhou, S. M. Dombrowski, et

al. 2015. “Mitochondrial Control by DRP1 in Brain Tumor Initiating Cells.” Nat

Neurosci 18 (April): 501–10. https://doi.org/10.1038/nn.3960.

Xu, Guozhong, and Mark R. Chance. 2007. “Hydroxyl Radical-Mediated Modification of

Proteins as Probes for Structural Proteomics.” Chemical Reviews 107 (8): 3514–

43. https://doi.org/10.1021/cr0682047.

Xu, Hua, and Michael A. Freitas. 2007. “A Mass Accuracy Sensitive Probability Based

Scoring Algorithm for Database Searching of Tandem Mass Spectrometry Data.”

BMC Bioinformatics 8 (April): 133. https://doi.org/10.1186/1471-2105-8-133.

Yamano, Koji, Noriyuki Matsuda, and Keiji Tanaka. 2016. “The Ubiquitin Signal and

Autophagy: An Orchestrated Dance Leading to Mitochondrial Degradation.”

EMBO Reports 17 (3): 300–316. https://doi.org/10.15252/embr.201541486.

Yamano, Koji, and Richard J Youle. 2013. “PINK1 Is Degraded through the N-End Rule

Pathway.” Autophagy 9 (11): 1758–69. https://doi.org/10.4161/auto.24633.

Yoon, G., Z. Malam, T. Paton, C. R. Marshall, E. Hyatt, Z. Ivakine, S. W. Scherer, et al.

2016. “Lethal Disorder of Mitochondrial Fission Caused by Mutations in DNM1L.”

J Pediatr 171 (April): 313-6.e1-2. https://doi.org/10.1016/j.jpeds.2015.12.060.

219

Yoon, Y., E. W. Krueger, B. J. Oswald, and M. A. McNiven. 2003. “The Mitochondrial

Protein HFis1 Regulates Mitochondrial Fission in Mammalian Cells through an

Interaction with the Dynamin-like Protein DLP1.” Mol Cell Biol 23 (August): 5409–

20.

Yoon, Y., K. R. Pitts, and M. A. McNiven. 2001. “Mammalian Dynamin-like Protein DLP1

Tubulates Membranes.” Mol Biol Cell 12 (September): 2894–2905.

Zhang, F., B. Crise, B. Su, Y. Hou, J. K. Rose, A. Bothwell, and K. Jacobson. 1991. “Lateral

Diffusion of Membrane-Spanning and Glycosylphosphatidylinositol-Linked

Proteins: Toward Establishing Rules Governing the Lateral Mobility of Membrane

Proteins.” The Journal of Cell Biology 115 (1): 75–84.

https://doi.org/10.1083/jcb.115.1.75.

Zhang, P., and J. E. Hinshaw. 2001. “Three-Dimensional Reconstruction of Dynamin in

the Constricted State.” Nat Cell Biol 3 (October): 922–26.

https://doi.org/10.1038/ncb1001-922.

Zhao, J., T. Liu, S. Jin, X. Wang, M. Qu, P. Uhlen, N. Tomilin, O. Shupliakov, U. Lendahl,

and M. Nister. 2011. “Human MIEF1 Recruits Drp1 to Mitochondrial Outer

Membranes and Promotes Mitochondrial Fusion Rather than Fission.” EMBO J 30

(July): 2762–78. https://doi.org/10.1038/emboj.2011.198.

Zhao, J., J. Zhang, M. Yu, Y. Xie, Y. Huang, D. W. Wolff, P. W. Abel, and Y. Tu. 2012.

“Mitochondrial Dynamics Regulates Migration and Invasion of Breast Cancer

Cells.” Oncogene, November. https://doi.org/10.1038/onc.2012.494.

220

Zhou, Chun, Yong Huang, Yufang Shao, Jessica May, Delphine Prou, Celine Perier,

William Dauer, Eric A. Schon, and Serge Przedborski. 2008. “The Kinase Domain

of Mitochondrial PINK1 Faces the Cytoplasm.” Proceedings of the National

Academy of Sciences 105 (33): 12022–27.

https://doi.org/10.1073/pnas.0802814105.

Zhu, P. P., A. Patterson, J. Stadler, D. P. Seeburg, M. Sheng, and C. Blackstone. 2004.

“Intra- and Intermolecular Domain Interactions of the C-Terminal GTPase

Effector Domain of the Multimeric Dynamin-like GTPase Drp1.” J Biol Chem 279

(August): 35967–74. https://doi.org/10.1074/jbc.M404105200.

Zhu, Yushan, Guo Chen, Linbo Chen, Weiling Zhang, Du Feng, Lei Liu, and Quan Chen.

2014. “Monitoring Mitophagy in Mammalian Cells.” Methods in Enzymology 547:

39–55. https://doi.org/10.1016/B978-0-12-801415-8.00003-5.

Ziegler, Daniel M., and K. A. Doeg. 1962. “Studies on the Electron Transport System.

XLIII. The Isolation of a Succinic-Coenzyme Q Reductase from Beef Heart

Mitochondria.” Archives of Biochemistry and Biophysics 97 (1): 41–50.

https://doi.org/10.1016/0003-9861(62)90042-5.

Zimmermann, Petra, Benjamin Mänz, Otto Haller, Martin Schwemmle, and Georg Kochs.

2011. “The Viral Nucleoprotein Determines Mx Sensitivity of Influenza A

Viruses.” Journal of Virology 85 (16): 8133–40.

https://doi.org/10.1128/JVI.00712-11.

Züchner Stephan, De Jonghe Peter, Jordanova Albena, Claeys Kristl G., Guergueltcheva

Velina, Cherninkova Sylvia, Hamilton Steven R., et al. 2006. “Axonal Neuropathy

221

with Optic Atrophy Is Caused by Mutations in Mitofusin 2.” Annals of Neurology

59 (2): 276–81. https://doi.org/10.1002/ana.20797.

Züchner, Stephan, Irina V. Mersiyanova, Maria Muglia, Nisrine Bissar-Tadmouri, Julie

Rochelle, Elena L. Dadali, Mario Zappia, et al. 2004. “Mutations in the

Mitochondrial GTPase Mitofusin 2 Cause Charcot-Marie-Tooth Neuropathy Type

2A.” Nature Genetics 36 (5): 449–51. https://doi.org/10.1038/ng1341.

222