Membrane Microdomains As Platforms for Extra-Cellular Fusions

Loyola University Chicago Loyola eCommons

Dissertations Theses and Dissertations

2019

Membrane Microdomains as Platforms for Extra-Cellular Fusions

Michael Hantak

Follow this and additional works at: https://ecommons.luc.edu/luc_diss

Part of the Virology Commons

Recommended Citation Hantak, Michael, "Membrane Microdomains as Platforms for Extra-Cellular Fusions" (2019). Dissertations. 3337. https://ecommons.luc.edu/luc_diss/3337

This Dissertation is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Dissertations by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

LOYOLA UNIVERSITY CHICAGO

MEMBRANE MICRODOMAINS AS PLATFORMS FOR EXTRA-CELLULAR FUSIONS

A DISSERTATION SUBMITTED TO

THE FACULTY OF THE GRADUATE SCHOOL

IN CANDIDACY FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

PROGRAM IN MICROBIOLOGY AND IMMUNOLOGY

MICHAEL HANTAK

CHICAGO, ILLINOIS

MAY 2019

ACKNOWLEDGEMENTS

I would like to thank my parents, my brother Joey, and my partner Matt, for their unwavering support and encouragement to pursue my dreams. I also owe a tremendous amount of gratitude to my mentor, Dr. Tom Gallagher, for the opportunity to learn in his lab. Tom’s enthusiasm for science has been truly contagious, and his expertise and feedback have made me become a better scientist, communicator, and educator. I would also like to thank my lab mates Dr. Jung-Eun Park, Dr. James

Earnest, Arlene Barlan, Enya Qing, Natalie Nidetz, Dr. Mayukh Sarkar, and Ryan

Sweeney. Their technical expertise was essential to my success in the lab, and my work has been shaped for the better by their creativity and curiosity. I would also like to thank Drs. Stan Perlman, Paul McCray, Kun Li, Keith Jones, and Lauren Haar for their collaboration with my project.

I am very grateful for my experiences working in the labs of Dr. Brian J. Wilkinson

(Illinois State University-Normal, IL) and Dr. Devkumar Mustafi (University of Chicago-

Chicago, IL). Dr. Wilkinson and Dr. Mustafi introduced me to laboratory science, and their mentorship has been an instrumental part of my training.

Furthermore, I would like to thank my dissertation committee, Drs. Ed Campbell,

Francis Alonzo, Seth Robia, and Susan Uprichard, for their continued feedback and generosity with their time. I would also like to thank the staff, students and faculty in the

Department of Microbiology and Immunology and across campus for creating a collaborative learning environment that laid the foundation for my success.

iii

Finally, this work would not be possible without: funding support from the T32

Immunology Training Grant to Dr. Knight; generous donations from the FALK

Foundation and the Doyle Fund; a Fellowship from the Arthur J. Schmitt Foundation; and the expertise of Pat Simms and Ashley Hess at the flow cytometry core.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iii LIST OF FIGURES ...... viii

CHAPTER ONE: REVIEW OF THE LITERATURE 1 Extra-Cellular Membrane Fusion ...... 1 Steps in Membrane Fusion ...... 2 Viral Membrane Fusion ...... 3 Coronavirus Membrane Fusion ...... 4 CoV Entry Routes ...... 5 Extracellular Vesicles ...... 8 EVs in Intercellular Communication ...... 9 Extracellular Vesicle Heterogeneity ...... 10 Extracellular Vesicle Cargo Delivery and Membrane Fusion ...... 10 Cellular Membrane Fusion ...... 12 Monocyte Membrane Fusion ...... 13 Cellular Membrane Fusion Catalysts ...... 14 Facilitators of Membrane Fusion ...... 16 Tetraspanins ...... 17 Tetraspanin-Enriched Microdomains ...... 19 Tetraspanins in Virus Entry ...... 20 Tetraspanins in Non-Viral Membrane Fusion ...... 20 Inhibitors of Membrane Fusion ...... 21 Interferon-Induced Transmembrane Proteins ...... 22 IFITM Activities during Enveloped Virus Entry ...... 23 Non-Viral IFITM Activities ...... 25 Purpose of Dissertation ...... 26

CHAPTER TWO: MATERIALS AND METHODS ...... 29 Cells and Reagents ...... 29 Plasmids and Cloning ...... 30 Transfections ...... 31 Generation of CD9 Knockout Cells ...... 32 Pseudoviruses ...... 32 Isolation of Tetraspanin-Enriched Membranes ...... 33 Recombinant Adenovirus Production ...... 34 Transduction and Infection of Mice ...... 34 EV Isolation ...... 35 Transmission Electron Microscopy ...... 35 Western Blot Analyses ...... 35 Quantifying EV Concentration ...... 36 Quantifying EV Cargo Secretion by Luciferase Assay ...... 37 Quantifying Intercellular Transfer of DSPs ...... 37

Monitoring Intercellular Transfer of DSPs by Fluorescence Microscopy...... 38 Analysis of EV-Cell Binding ...... 39 Production of IFITM3 KO Cells ...... 39 Quantifying IFNβ-Mediated Blockade of EV Cargo Transfer ...... 40 In Vitro EV-Virus and EV-EV Fusion Assays ...... 40 Fluorescence Microscopic Analyses of EVs ...... 41 Production of ifitm5-Encoding Lentiviral Particles and Cell Lines ...... 42 Immunofluorescence Microscopy of RAW Cells ...... 42 Osteoclast Formation Assay ...... 43 Osteoclast Differentiation Assay ...... 43 Viral Syncytial Formation Assay ...... 43 Statistical Analysis ...... 44

CHAPTER THREE: RESULTS ...... 45 SECTION 1: TEMs as Platforms for Virus-Cell Membrane Fusion ...... 45 The Tetraspanin CD9 Facilitates MERS-CoV Entry ...... 45 CD9 Holds DPP4 in TEMs ...... 47 CD9 Brings DPP4 and TMPRSS2 into Close Proximity ...... 48 CD9 and TMPRSS2 Are Critical for MERS-CoV Infection in vivo ...... 50 SECTION 2: TEMs as Platforms for Extracellular Vesicle Membrane Fusions 53 Membrane Fusion Drives EV Content Transfer ...... 53 Tetraspanins Promote EV Release and Enable Cargo Transfer ...... 57 Tetraspanins Act at the Level of EV Cargo Delivery to Enable EV Transfers 61 Additional Enablers of EV-mediated Cargo Transfer ...... 62 Type I Interferons Block EV Cargo Transfer through IFITM3 ...... 64 IFITM3 Blocks EV Cargo Transfer at the Level of Membrane Fusion...... 66 IFITM3 Preferentially Incorporates into Tetraspanin-Containing EVs ...... 71 Tetraspanins and TMPRSS2 Enable EV-Directed Transfers in Mesenchymal Stem Cells ...... 72 SYSTEM 3: TEMs as Platforms for Monocyte Cell-Cell Fusions ...... 74 Disease-causing Mutations Alter IFITM5 Subcellular Localization ...... 74 IFITM5 Blocks Osteoclast Formation ...... 76 IFITM5 Does Not Influence Pre-Osteoclast Differentiation ...... 77 Amphotericin B Inhibits IFITM5...... 78 IFITM5 Restricts Membrane Fusion ...... 81 CD9 Overexpression Rescues Osteoclast Formation in the Presence of IFITM5 ...... 83

CHAPTER FOUR: DISCUSSION ...... 85 Summary of Data ...... 85 Requirement of CD9 during MERS-CoV Early Entry ...... 87 Selective Use of TMPRSS2 by MERS-CoVs to Infect the Lung ...... 88 Tetraspanins in the Entry of Other CoVs ...... 89 Utilizing Recombinant Adenoviruses To For Gene Knockdown ...... 90 TEMs as General Platforms for Viral Entry ...... 90 Mechanisms of EV Cargo Delivery ...... 91

vii

Inhibition of EV Cargo Delivery ...... 92 Potential Implications of IFITM-Mediated Inhibition of EV Fusion ...... 93 Limitations to EV Cargo Transfers ...... 94 Membrane Fusion Inhibition as a Shared Function of IFITMs ...... 95 Functional Relationship between IFITMs and Tetraspanins ...... 96 Amphotericin B and IFITM-Mediated Fusion Inhibition ...... 97 Potential Heterogeneity in OI Type V Pathologies ...... 97 Identifying Additional Membrane Fusion Catalysts and Regulators ...... 98

REFERENCES ...... 99

VITA…………………………………………………………………………………….121

viii

LIST OF FIGURES

Figure 1. Steps in Membrane Fusion ...... 3

Figure 2. Steps in Coronavirus Membrane Fusion ...... 5

Figure 3. Proteases Dictate CoV Entry Routes ...... 7

Figure 4. Steps in EV Cytoplasmic Cargo Transfer ...... 11

Figure 5. Tetraspanin Structure and Protein Interactions ...... 18

Figure 6. Structure of the IFITMs ...... 23

Figure 7. CoV Entry into Tetraspanin KO Cells ...... 46

Figure 8. Association of CoV Entry Factors with CHAPS-resistant Membranes in the Presence or Absence of CD9 or CD81 ...... 48

Figure 9. Proximity Ligation Assay of DPP4 and TMPRSS2 in CD9KO Cells ..... 50

Figure 10. Analysis of Adenovirus Knockdown of MERS-CoV Entry Factors in Cell Culture and in vivo ...... 52

Figure 11. Diagram of EV Cargo Transfer Assay Using the DSP Reporter System ...... 54

Figure 12. Membrane Fusion Drives EV Cargo Transfers...... 56

Figure 13. Tetraspanins Act in Producer Cells to Enable EV Cargo Transfers ... 58

Figure 14. Tetraspanins Promote EV Secretion and are Enriched in Transfer-competent Vesicle Preps ...... 60

Figure 15. Tetraspanins Drive EV Content Delivery into Target Cells ...... 62

Figure 16. TMPRSS2 and Baf Enhance EV Outputs and Enable EV Content Transfer ...... 64

Figure 17. IFN Inhibits EV-directed Cargo Transfer through IFITM3 ...... 66

viii

Figure 18. IFITM3 Inhibits EV-directed Cargo Transfer ...... 68

Figure 19. IFITM3 Inhibits in vitro EV-virus and EV-EV Fusions ...... 70

Figure 20. IFITM3 Incorporates into a Subset of Tetraspanin-containing EVs .... 72

Figure 21. Tetraspanins and TMPRSS2 Promote EV Cargo Transfer in MSCs .. 73

Figure 22. Localization of IFITM5 in RAW Macrophages...... 75

Figure 23. Effect of ifitm5 Overexpression on Osteoclast Formation ...... 77

Figure 24. Effect of ifitm5 Overexpression on the Expression of Tartrate-resistant Acid Phosphatase...... 78

Figure 25. Effect of Amphotericin B on IFITM5 Activity ...... 80

Figure 26. Effect of ifitm5 Overexpression on Viral Membrane Fusion ...... 82

Figure 27. Effect of CD9 Overexpression on IFITM5 Activity ...... 84

CHAPTER ONE

REVIEW OF LITERATURE

Extra-Cellular Membrane Fusion

Intra- and inter-cellular communications are essential for life. In some circumstances, communications are mediated by the transfer of hydrophilic molecules

(1, 2). The lipid membranes that line our cells and organelles, due to their hydrophobic cores, prevent these hydrophilic molecules from being readily transferred. To successfully transfer hydrophilic cargos, membrane-enclosed compartments separate and unite through dynamic, flexible fission and fusion reactions (2–5). These catalyzed processes are central to organelle and organismal biogenesis and communication.

Intra-cellular membrane fusions are often utilized by cells to traffic macromolecules between organelles. The molecular details of these intracellular fusions are relatively well understood, and rely on membrane proteins extending into the cytoplasm to catalyze membrane mergers, facilitating cargo transfers between compartments (2). Extra-cellular membrane fusions, catalyzed by membrane proteins extending into the extracellular space, are also central to several biological processes.

Enveloped viruses enter cells through membrane fusions (5). Extracellular vesicles

(EVs) released from cells also transfer proteins and nucleic acids between cell cytoplasms, presumably through membrane fusion (3, 4). Even entire cells can fuse together, generating essential biological structures, as discussed below (6). While the ability to mediate extracellular membrane fusion is conserved amongst all eukaryotes

(7), and possibly all life (8), the molecular details of these extracellular fusions are often not well understood. Further elucidation of extracellular membrane fusion mechanisms is required to better control these fusion events in health and disease.

Steps in Membrane Fusion

The main constituents of biological membranes are amphipathic phospholipids, each of which are composed of a hydrophilic head group and two hydrophobic tails. In aqueous environments, hydrophobic interactions between phospholipid tails facilitate lipid aggregation into a semi-permeable bilayer with the hydrophobic tails of each leaflet facing each other. Many hydrophilic molecules cannot penetrate the hydrophobic cores of these membranes. Where hydrophilic molecules are contained exclusively within cell- and organelle-limiting membranes, fusion between distinct lipid bilayers is required to transport these molecules between cells.

Current models of membrane fusion indicate that fusion takes place in several steps (9) (Figure 1). First, membranes are brought into close proximity. Outward,

“positive” curvatures in the bilayers facilitate these close interactions (10). Upon coming into close proximity, the proximal leaflets of the bilayers merge, forming a hemifusion intermediate (11). In this hemifused state, the distal leaflets of the two bilayers remain distinct, preventing the mixing of compartment lumens. Finally, the merger of distal membrane leaflets leads to the formation of an aqueous pore. Pore stabilization and expansion culminates in the mixing of internal contents between the now joined compartments.

While membrane fusion is thermodynamically favorable, there are high kinetic barriers to the fusion process. Cells and enveloped viruses have evolved to utilize a

3 variety of lipids and membrane proteins to reduce these energy barriers and render membranes permissive for fusion (5, 9, 12, 13). In many contexts, however, these membrane fusion catalysts remain to be identified and characterized.

Figure 1. Steps in Membrane Fusion. Fusion takes place in several steps. First, opposing membranes (grey lipids) are brought into close proximity, likely aided by positive membrane curvatures away from compartment lumens (blue and red). Then, outer leaflets of the membrane merge, forming a hemifusion intermediate. Finally, distal leaflets merge, allowing for the mixing of compartment lumens (now purple).

Viral Membrane Fusion

The most well-studied membrane fusion processes occur during enveloped virus entry. These viruses are encased by cell-limiting membranes during their egress from infected cells, and thus must fuse with target cell-limiting membranes to drive further infections. Virus-cell membrane fusions are catalyzed by virus-encoded membrane fusion glycoproteins that are embedded in the virus envelope (5). During their production, virus fusion proteins are folded in such a way to store all of the energy required for fusion catalysis. This energy is stored through intramolecular interactions

(14), or through interactions with other viral proteins (15). To catalyze membrane fusion, fusion proteins are “activated” to undergo enormous conformational changes

4 that release the energy required to drive membrane merger (5, 14, 16, 17). They first unfold into long extended intermediates, exposing hydrophobic fusion peptides (FPs) that embed into the core of the opposing lipid bilayer. These extended intermediates then refold, pulling virus and cell membranes together and catalyzing membrane merger.

All characterized virus membrane fusion proteins are not enzymatic, and thus their activation must be tightly controlled to prevent untimely unfolding. To provide this control, fusion proteins have evolved to utilize factors on target cells as “activators”, thereby isolating the fusion process to target cell membranes (5). “Activators” directly engage membrane fusion proteins, and this engagement “triggers” fusion-catalyzing conformational changes. Activators dictate cellular susceptibility to fusion, and can include cellular receptors, proteases, and cations.

Coronavirus Membrane Fusion.

Coronaviruses (CoVs) are enveloped, positive-sense RNA viruses that can cause mild and severe infections in the enteric and respiratory tracks of many mammals

(18). In humans, severe acute respiratory syndrome (SARS-) and Middle East respiratory syndrome (MERS-) CoVs can cause severe lung pathologies, especially amongst immunocompromised or co-morbid patients (19, 20). CoV infections are mediated by trimeric, fusion-catalyzing spike (S) glycoproteins that extend ~10 nm from the virion surface (21). To catalyze membrane fusion, CoV S proteins must engage multiple activators on the target cell (Figure 2). In the case of MERS-CoV, dipeptidyl peptidase 4 (DPP4) receptors are the first activators that S proteins must encounter on the target cell (22). MERS-CoV S (MERS-S) proteins engage DPP4s via their receptor

5 binding domains (23). This engagement exposes sites on the S protein susceptible to cleavage by cellular proteases (24). S protein proteolysis “triggers” the S protein to relocate its receptor binding domains and unfold into an extended intermediate structure. Unfolding exposes hydrophobic fusion peptides that embed into the target cell membrane (25). MERS-S proteins then refold, pulling virus and cell membranes closely together and facilitating their merger (26).

Figure 2. Steps in Coronavirus Membrane Fusion. To catalyze membrane fusion and virus entry into host cells, MERS-CoV S (MERS-S) proteins (gray) bind dipeptidyl peptidase 4 (DPP4) receptors (purple) via their receptor-binding domains (green). Receptor engagement exposes substrates (blue stars) susceptible to cleavage by cellular proteases (blue). Proteolytically- triggered MERS-S relocates its receptor binding domains and unfolds into an extended intermediate structure that embeds hydrophobic fusion peptides into target cell membranes. Refolding of intermediates then pulls virus and cell membranes together to catalyze membrane fusion. CoV Entry Routes.

Unlike most enveloped viruses, which enter cells in particular subcellular locations, CoVs have remarkable flexibility in the pathways they can take during entry

(27). This is due to the ability of CoVs to utilize a variety of proteases as membrane fusion “triggers” (28–33). Entry routes are thus dictated by the relative abundance of proteases on the cell surface and along the endocytic network. Transmembrane proteases (TMPRs), such as transmembrane type II serine proteases (TTSPs), can cleave S proteins at or near the cell surface (30, 34, 35), while water-soluble cysteine cathepsin proteases can cleave S proteins in acidified endosomes (29, 31). Therefore, in cells with high TMPR levels, S protein proteolysis occurs rapidly, leading to “early” entry at or near the cell surface (Figure 3, left panel). In cells with low TMPR levels,

CoVs are instead endocytosed before cathepsin-mediated proteolysis facilitates “late” entry from endosomes (Figure 3, right panel).

Figure 3. Proteases dictate CoV entry routes. (Left panel) When cell surface proteases, like TMPRs (blue), are present on the cell surface in sufficient abundance, MERS-S proteins bind DPP4 (purple) and quickly encounter triggering cleavages (blue stars). Rapid cleavage facilitates “early” entry at or near the cell surface. (Right panel) In the absence of cell surface proteases, MERS-S proteins are not efficiently triggered on the cell surface. MERS-CoVs are instead endocytosed, where they encounter endosomal cathepsins (brown). At low pH (yellow), cathepsins cleave MERS-S proteins, triggering inefficient “late” entry in the endosomal network.

Serine protease inhibitors significantly reduce CoV titers in the lung, suggesting that early entry routes are also preferred in vivo (33), and that late entry may be relatively inefficient. This may be due to CoVs encountering harsh, degradative environments in endosomes and lysosomes during late entry (36), as well as being exposed to endosome-localized antiviral factors (37). However, there are also additional requirements imposed on CoVs during early entry. “Triggering” proteolytic cleavage sites on MERS-S are only exposed upon DPP4 engagement (28). In addition, it is likely that multiple MERS-S proteins must be triggered to generate the forces required to catalyze membrane fusion (38). DPP4 receptors and proteases must

8 therefore be in close proximity to facilitate efficient MERS-CoV entry into target cells.

During “late” entry, water-soluble cathepsins likely have relatively easy access to DPP4- bound MERS-S proteins. In contrast, it is thought that transmembrane proteases are sequestered in membrane microdomains (39), and mislocalization can cause severe damage to cells and organisms (40). TMPRs could therefore be scarce on most of the cell surface, making it unclear how DPP4-bound MERS-CoVs engage TMPRs efficiently.

Extracellular Vesicles

Virtually all cells can release membrane-enclosed extracellular vesicles (EVs) containing cytoplasmic proteins, nucleic acids, and small molecules (4, 41, 42). EVs are formed by outward budding events at the plasma membrane (43), or by membrane budding into endosomes (44). EVs are abundant in most biological fluids, with concentrations approaching one billion particles/mL in some cases (45). When they were first observed in the early 1980s (43, 44, 46, 47), EVs were thought to function as cellular “garbage bags,” loaded with materials to be either expelled into the extracellular space or degraded in lysosomal compartments (44). It is now clear, however, that EVs also play central roles in intercellular communication (1). They do this by ferrying their cargos to other cells, changing how those cells behave (4, 41). While the field of EV biology is nascent, it is growing exponentially, with thousands of articles documenting

EV-mediated phenotypic changes to cells and organisms. These articles indicate that

EVs can regulate a variety of biological processes, including tumor growth and metastasis (48), lymphocyte activation (49, 50), innate immune responses (51), pathogenic infections (52), skeletal maintenance (53), and development (54).

EVs in Intercellular Communication.

While many studies indicate that EVs play critical roles in health and disease, most investigations have focused on documenting EV phenotypic effects on cells and tissues. Mechanistic insights into how EVs carry out these functions, however, are surprisingly lacking. Several pioneering studies, however, have begun to expand our mechanistic understanding of EV activities. Much of this work has focused on the integral membrane proteins displayed on EV surfaces. For example, integrins laden on

EV membranes can facilitate cell motility through interactions with extracellular matrix components (55). Proteases on EV surfaces can also degrade extracellular matrix, contributing to lung disease (56). In another example, dendritic cell – derived EVs display antigen-MHC complexes that can activate T cells (49, 50). These critical EV activities do not require delivery of internal vesicle cargos.

EV luminal cargos are also known to facilitate critical intercellular communications. Indeed, the ability of EVs to delivery cargo is increasingly recognized for its biological importance and therapeutic potential (57). There has been particular interest in the effects of EV microRNAs. In one of the most well-documented examples, luminal EV microRNAs from breast cancer cells can induce non-tumorigenic cells to form tumors, and this activity is dependent on a component of the RNA induced silencing complex, strongly supporting the hypothesis that EV-delivered microRNAs contribute to these effects (48). In another example, mesenchymal stem cell – derived

EVs can ferry the anti-apoptotic microRNA-21 to surrounding heart tissues following ischemic injury, to prevent cell death and tissue damage (58). Other cytoplasmic factors are also transmitted by EVs. During viral infections, EVs from infected cells can

10 transmit viral genomes to naïve cells, thereby propagating infection (59). Intriguingly, non-enveloped viruses, including paramyxoviruses and hepatitis A virus incorporate into

EVs during their egress (60, 61), potentially to evade antiviral immune responses.

Pathogenic prions and other self-aggregating proteins associated with neurodegenerative disease are also transferred via EVs (62, 63). These studies highlight just some of the biological activities mediated by internal EV cargos.

Extracellular Vesicle Heterogeneity.

EVs are remarkably heterogeneous. They can range in size from 40 to over

1000 nm in diameter, and they can contain a variety cytoplasmic and transmembrane cargos (64). Their heterogeneity stems, in part, from the subcellular location of their biogenesis. Some EVs, termed microvesicles, originate from budding events at the plasma membrane (43). Another subset of EVs, exosomes, are instead derived from membrane budding into endosomes, which must subsequently fuse with the plasma membrane to release their luminal vesicles (44). EVs heterogeneity also extends to the molecular mechanisms of their biogenesis. Some EVs are generated by the endosomal sorting complex required for transfer (ESCRT) machinery (65, 66), with others are generated by ESCRT-independent mechanisms (67–69). These different biogenesis locations and mechanisms may correlate with distinct EV structures, cargos, and biological functions (70–72), though these distinctions remain to be fully elucidated.

Extracellular Vesicle Cargo Delivery and Membrane Fusion.

EV-directed transport of cytoplasmic effectors requires several steps (Figure 4).

First, the effectors must be incorporated into vesicles that bud out from cell-surfaces or endosome membranes (43, 44, 46, 47, 73, 74). For vesicles budding into endosomes,

11 the resultant multi-vesicular bodies (MVBs) must traffic and fuse with plasma membranes to permit their escape as EVs (44). Second, the secreted EVs must disseminate and bind to cells at nearby or distant locations (75–77). Third, the EV payloads must cross the limiting EV and target cell membranes. These EV cargo deliveries have been documented in cell culture (41), and recent work from Dr. Stephen

Momma and colleagues has documented that EVs can transfer Cre recombinases to cell cytosols in vivo (78).

Figure 4. Steps in EV Cytoplasmic Cargo Transfer. To exclusively cytoplasmic molecules between cells. First, cytoplasmic factors, including proteins (red), nucleic acids (blue) and small molecules (green), are loaded into EVs during their biogenesis. EVs then secrete by budding from the plasma membrane, or by budding into endosomes followed by endosome – plasma membrane fusion. Secreted EVs disseminate and bind to target cells. Finally, EVs deliver their cargo into target cell cytoplasms, likely through EV-cell membrane fusion.

There is some evidence that EV cargo delivery can be accomplished through EV

– target cell membrane fusion. The first observations of EV-cell membrane fusion came

12 from Parolini and colleagues in 2009 (3). Using EVs labeled with self-quenching fluorescent dyes, they documented increases in fluorescent signals after incubating these EVs with target cells, suggesting that quenched dyes were dispersed into target cell membranes by membrane fusion events. These investigations were extended by

Montecalvo and colleagues in 2012, who also documented the delivery of EV- encapsulated luciferin into luciferase-expressing target cells (4). It remains unclear, however, whether EVs can deliver their cargos by other mechanisms, such as transmembrane channels (79). Whether EV cargo deliveries rely on EV-cell fusions remains a divisive question in the field (1), and factors facilitating EV-cell membrane fusion and cargo delivery remain to be identified.

Cellular Membrane Fusion

The most extensive form of membrane fusion-mediated communication is cell- cell fusion, which facilitates the complete mixing of cell cytoplasms. Cell-cell fusions are required to form many important biological structures in multicellular organisms. For example, sexual reproduction depends on cell-cell fusions between gametes to generate fertilized zygotes (80). During fetal development, trophoblastic cells fuse together to form the placenta (81). Even after birth, cell-cell fusions continue to occur.

Fusions between myoblasts generate long skeletal muscle fibers (82), and monocyte- monocyte fusions generate osteoclasts that maintain skeletal integrity (83), as well as giant-cell macrophages that protect tissues from foreign pathogens and materials (84).

Dysregulated cell-cell fusions are linked to disease. Overabundant monocyte fusions can significantly enhance bone degradation, and contribute to both osteoporosis

(85) and Paget’s Disease (86). Reduced myoblast fusions have been linked with some

13 forms of muscular dystrophy (87), while inhibition of trophoblast fusions is associated with preeclampsia (88). Tumor cells are also known to fuse with stromal, epithelial, and hematopoietic cells, and these cell-cell fusions may enhance tumor metastasis and drug resistance (89). Strict control of cell-cell fusions is thus required to maintain homeostatic conditions in the body and prevent disease.

Monocyte Membrane Fusion.

Monocytes and macrophages have inherent phagocytic and degradative capacities, and these activities are often used to combat pathogens and remove debris from the body. Under conditions where large amounts of material must be degraded or phagocytosed, monocytic cells fuse together to form giant multinuclear macrophages

(GMMs) (83, 90–92). These syncytial GMMs have a higher phagocytic and degradative capacities than their mononuclear counterparts, in part due to their ability to create large, hydrolytic extracellular compartments to mediate endocytic internalization of degraded materials (93).

One form of GMMs, giant lung cell macrophages, form at sites of tuberculosis granulomas, and are thought to aid in degrading these large masses of material (91,

92). Similar GMMs also form around non-infectious foreign materials, presumably to facilitate their removal (90). The monocyte fusions that generate giant lung macrophages can be induced by inflammatory cytokines, including IL-4 and IL-13 (84).

Osteoclasts, another GMM, are highly specialized cells that adhere to bone and release degradative enzymes to resorb and remodel bone surfaces (83). The process of osteoclast formation begins with the recruitment of peripheral monocytes to sites of skeletal repair (94, 95). Receptor activator of NF-κB (RANK) proteins displayed on

14 these monocytes interact with RANK ligands (RANKL) on bone-depositing mesenchymal osteoblasts. RANK:RANKL interactions induce monocyte differentiation into pre-osteoclasts (96). During differentiation, numerous factors are upregulated, including tetraspanins (97), integrins (98), immunoglobulin-like proteins (99, 100), and seven transmembrane receptors (101). Some of these proteins likely form complexes on the pre-osteoclast cell surface, and then go on to catalyze monocyte-monocyte fusions to form multinucleated osteoclasts (102, 103).

Cellular Membrane Fusion Catalysts.

While the fusion proteins of enveloped viruses are relatively well studied, it is surprising that eukaryotic membrane fusion catalysts have remained elusive. Recent investigations, however, has made significant headway in identifying some endogenous membrane fusion catalysts. Remarkably, these investigations suggest that eukaryotic cells have “captured” several syncytial virus fusion proteins, and have “domesticated” these proteins to catalyze essential membrane fusions in the absence of viral infection.

The first bona-fide endogenous fusion catalysts discovered were two glycoproteins, aptly named syncytins 1 and 2, which catalyze trophoblast cell fusions

(12, 104, 105). These syncytins are encoded by endogenous retroviral elements that likely integrated into mammalian genomes during ancient retroviral infections (106).

Syncytins retain the essential structures of viral fusion proteins (106), and likely catalyze membrane fusions by similar mechanisms. In mice, each of the two expressed syncytins contributes to trophoblast fusion. Disruption of both syncytins renders mice infertile (107), indicating the essential role these fusion proteins play in placental

15 development. In humans, mutations in the syncytins correlate with preeclampsia, adding further support to this notion (88).

It is notable that, while syncytins are found in a wide variety of placental mammals, individual syncytin genes are not orthologous, and thus arose from independent “domestication” events (106). It is, however, conceivable that the individually acquired syncytins displaced the function of a more ancient, common trophoblast fusogen. Nevertheless, evolutionary studies suggest that the syncytins were acquired between 20 and 80 million years ago, roughly coinciding with the appearance of placental mammals (106).

While the syncytins were first thought to be exclusively expressed in trophoblasts, recent studies suggest that also play critical roles in the formation of skeletal muscle and giant cell macrophages. Syncytin – targeting antibodies and peptides suppress both myoblast-myoblast and monocyte-monocyte membrane mergers (108, 109), suggesting that these endogenous fusion proteins may play central roles in many cell-cell fusion events. The evolution of the musculoskeletal system, however, far preceded the acquisition of syncytin proteins, and whether syncytins are typically expressed in non-placental tissues is a matter of debate (12).

A second endogenous fusion protein, hapless 2 (HAP2), catalyzes fusion between gametes, and is essential for sexual reproduction in many organisms (13, 110–

112). Unlike the syncytins, HAP2 is highly conserved across most eukaryotes, with orthologs expressed by some of the most evolutionarily distant eukaryotic lineages, suggesting that this protein was present in the last eukaryotic common ancestor (113).

Structural modeling indicates that HAP2 proteins are highly similar to flavivirus and

16 alphavirus fusion proteins (13). When HAP2 is disabled, gametes adhere closely together, but do not fuse (111), indicating this protein is essential for many gamete fusion events. HAP2 is typically expressed in the “male” gamete (112), suggesting that human sperm may also express HAP2-like fusion proteins. Intriguingly, however, HAP2 has been lost in numerous eukaryotic organisms, including C. elegans, mice, and humans (113). This suggests that other membrane fusion catalysts may have supplanted HAP2 in some eukaryotes. Nonetheless, the conserved structures and origins of both endogenous and viral fusion proteins suggests that these glycoproteins are fine-tuned to drive membrane megers.

Facilitators of Membrane Fusion

In addition to activators that directly engage fusion proteins, target cells also contain factors that indirectly facilitate virus-cell membrane fusion. For example, some lipids spontaneously generate positive membrane curvatures, likely due to the large size of their polar heads relative to their hydrophobic tails (9). These lipids, including lysophosphatidylcholine and polyphosphoinositides, enhance membrane fusogenicity.

Other fusion-promoting lipids, such as cholesterol and surface-exposed phosphatidylserines, do not directly alter membrane curvatures, but instead may control membrane fusion by facilitating protein-lipid and protein-protein interactions (114–116).

Proteins can also act as indirect fusion facilitators. In many cases, these proteins induce signaling pathways that likely induce the expression of membrane fusion catalysts and their essential cofactors (96, 101). Integrins and adhesion proteins can also facilitate fusions, by holding opposing membranes in close proximity (117). During cell-cell fusion, actin and its interacting partners also facilitate fusion by forming highly

17 curved cellular protrusions (118). These protrusions subsequently interact with membranes reinforced by actin “sheaths” in the opposing cells (118). Both of these activities likely enhance tension in cell membranes, and could thus promote membrane disruption and lipid mixing. Another family of proteins, the tetraspanins, facilitate a wide variety of virus-cell and cell-cell fusion events, as described below. This dissertation focuses on identifying and characterizing tetraspanin roles in extracellular membrane fusions.

Tetraspanins

The tetraspanins are highly conserved membrane proteins ubiquitously present in all eukaryotes (119). They are composed of four (tetra) transmembrane spans connected by one small and one large extracellular loop (SEL and LEL, respectively)

(Figure 5, left). Their transmembrane domains are alpha-helical coils that interact to form a cone-shaped structure that tapers in the inner membrane leaflet (116). Their conserved structures and activities suggest that they play a critical role in controlling membrane architecture and composition. They induce positive membrane curvatures

(120) (Figure 5, left), potentially through the cone-shaped structure formed by their transmembrane spans (116), which may displace lipids in the outer membrane leaflet.

Tetrapsanins are also highly enriched in EVs (121), and facilitate EV production (122,

123). This activity may trace to their membrane-curving activities, as EV membranes acquire positive curvatures as they bud into endosomes or the extracellular space.

Tetraspanins also interact with other integral membrane proteins and lipids, controlling their subcellular localization. The tetraspanin CD81, for example, contains a pocket between its transmembrane domains that binds cholesterol, thereby altering LEL

18 conformations (116). Other tetraspanins associate with palmitic acid through covalent interactions between this lipid and the tetraspanin cytoplasmic domains (124, 125). In conjunction with these lipids, tetraspanins also interact with “partner” proteins.

Tetraspanin transmembrane domains likely facilitate homotypic tetraspanin-tetraspanin interactions (126, 127), while palmitolyation facilitates hetero-typic interactions with other tetraspanins. Tetraspanin LELs and cytoplasmic tails also interact with adhesion molecules (128), integrins (129, 130), and proteases (131), giving rise to “webs” termed tetraspanin-enriched microdomains (TEMs) on cell membranes (132, 133) (Figure 5, right). The LELs are the most variable regions of tetraspanin proteins (133). Individual tetraspanins, therefore, often have several distinct protein partners.

Figure 5. Tetraspanin Structure and Protein Interactions. (Left Panel) Tetraspanins are composed of four transmembrane spans connected by one large extracellular loop (LEL) and one small extracellular loop (SEL), with short N- and C-terminal tails protruding into the cytosol. They induce positive membrane curvatures in cell membranes, potentially due to their “cone-shaped” transmembrane domains. (Right Panel) Tetraspanins (red and black) act as scaffolds for transmembrane proteins. The form homo- and hetero-typic interactions with other tetraspanins through interactions between their transmembrane domains. They also interact with other integral membrane proteins (green and yellow) through their LELs and cytoplasmic tails. By interacting with multiple protein partners simultaneously, tetraspanins can “construct” protein complexes.

Tetraspanin-Enriched Microdomains.

Early characterizations of TEMs suggest that multiple homo- and hetero-typic interactions between tetraspanins and their partner proteins lead to the formation of semi-stable membrane microdomains that are highly complex (134). These studies relied on the isolation of tetraspanin-enriched membranes following detergent-mediated cell solubilization. TEMs are similar to lipid rafts, but can be distinguished by their relative sensitivity to particular detergents (135). Zwitterionic detergents, such as Brij99 and CHAPs, do not disrupt membranes containing tetraspanins and TEM resident proteins. It is likely that these detergents do not disrupt tetraspanin-mediated protein- protein interactions, and also could maintain hydrophobic interactions between tetraspanin transmembrane domains (127). Non-ionic detergents like Triton X-100, however, disrupt tetraspanin interactions, but do not disrupt canonical lipid rafts (135).

Thus, TEM and lipid raft – resident proteins can be distinguished by isolating detergent resistant membranes after differential solubilization.

These biochemical approaches, however, do not necessarily reflect the composition of intact TEMs found on the cell surface. Indeed, while analysis of detergent resistant membranes indicates that heterotypic tetraspanin-tetraspanin interactions occur (134, 136), high-resolution stimulated emission depletion microscopy indicates that most TEMs form from homotypic tetraspanin-tetraspanin interactions, with

TEMs containing approximately 3-4 tetraspanins (137). Therefore, while dynamic heterotypic tetraspanin-tetraspanin interactions may have important biological activities, these heterotypic interactions may be more transient that initially thought.

Tetraspanins in Virus Entry.

Accumulating evidence indicates that TEMs act as platforms for virus entry. The strongest evidence for this is in the context of hepatitis C virus (HCV) entry (138). Here, the tetraspanin CD81 acts as a direct receptor for the HCV E2 membrane fusion protein. In other cases, tetraspanins and TEM resident proteins likely promote entry without directly engaging viral fusion proteins. These viruses include human cytomegalovirus (HCMV) (139), lujo virus (140), human papilloma virus (HPV) (141), influenza A virus (IAV) (28, 142), and several alphaviruses (143, 144). For example, the tetraspanin CD151 coalesces receptor-bound HPVs and their integrin co-receptors.

These co-receptors contain endocytic motifs, and thus these CD151-mediated interactions facilitate virus endocytosis (141, 145). In addition, while CD81 binds directing to the HCV E2 fusion protein, it also coalesces the HCV co-receptors syndecan-1, scavenger receptor B1, and claudin-1 into a complex that drives HCV endocytosis (146, 147). Intriguingly, there is evidence that the MERS-CoV receptor

DPP4 binds the tetraspanin CD9 (148), suggesting that MERS-S receptor binding, and subsequent proteolytic priming, may also take place in TEMs. Overall, these data suggest that many viruses have evolved to utilize closely interacting entry factors on target cells, and that tetraspanins aggregate these factors into TEMs.

Tetraspanins in Non-Viral Membrane Fusion.

Outside of viral infection, tetraspanins and TEMs control a wide variety of important endogenous biological activities, including cell metabolism and signal transductions. They are also known to play critical roles in cellular membrane fusion events. During cell-cell fusion, tetraspanins may form complexes with other proteins

21 essential for these membrane fusion processes (102, 103), and they are often found on the tips of cellular protrusions where cell-cell membrane mergers are thought to occur

(149). The evidence that tetraspanins facilitate cell-cell fusion is most striking in the context of fertilization, where CD9 is absolutely required for gamete fusion. During gamete fusion, CD9 can be found on the tips of oocyte cellular extensions (149). Here,

CD9 plays a critical role in maintaining these extensions, potentially through their membrane-curving properties. CD9 knockout mice are infertile, with oocytes lacking

CD9 are entirely unable to fuse with adhered sperm cells (150). Intriguingly, CD9- containing EVs can compensate for this loss of CD9 by linking sperm and egg cells and subsequently catalyzing their fusion in a process known as “fusion-from-without” (151).

Tetraspanins also control myoblast and monocyte fusion events. CD9 overexpression facilitates osteoclast formation (97), and CD9 is an integral component of the multiprotein complexes that likely drive monocyte membrane mergers (83). CD9- targeting antibodies also suppress the fusion of myoblasts (152), suggesting that the tetraspanin protein family may generally regulate cell-cell fusion events.

Inhibitors of Membrane Fusion

Cells do not just rely on membrane fusion facilitators to tightly control membrane fusion processes. They also rely on fusion inhibitors to restrict membrane fusion potential. The roles of fusion-inhibiting lipids have been relatively well studied. For example, the lipids phosphatidylethanolamine and diacylglycerol contain relatively bulky hydrophobic tails and small hydrophilic head groups. These lipids therefore spontaneously generate negative membrane curvatures that suppress interactions between opposing membranes and decrease membrane fusogenicity (9). In addition,

22 cells can also express fusion-inhibiting proteins. B cell – produced antibodies can engage virus fusion proteins to “neutralize” their fusion potential (153). In addition, cells can express interferon-induced transmembrane (IFITM) proteins, which act to inhibit enveloped virus membrane fusion. Below, we will discuss the IFITM protein family in more detail.

Interferon-Induced Transmembrane Proteins

IFITM proteins are type II transmembrane proteins approximately 130 residues in length (154). They are constituents of the larger dispanin protein family, members of which are found across all domains of life (155). Dispanins are composed of two alphahelical domains within the cell membrane linked by an intracellular loop. In the case of the IFITMs, the first of these alphahelical domains is amphipathic (156), and intercalates into the inner leaflet of the lipid bilayer (157). It is connected to a long cytosolic N-terminal tail. The second domain spans the entire bilayer to expose a short

C-terminal tail into the extracellular space (154) (Figure 6, Front View). Phenylalanine residues in the first membrane span facilitate homo- and hetero-typic IFITM mutimerization (158). The human genome contains five ifitm genes. Some ifitms, ifitm1, 2, and 3, are expressed abundantly in response to type 1 interferons (159, 160).

These interferon-induced IFITMs have been most extensively studied, particularly in the context of enveloped virus infection.

Figure 6. Structure of the IFITMs. (Front View) IFITMs are short type II transmembrane proteins with two alphahelical transmembrane domains. A relatively long N-terminal tail precedes the first helix. This helix is amphipathic, and intercalates into the inner leaflet of the lipid bilayer. The second helix is truly transmembrane, and exposes a short C-terminal tail to the extracellular space. (Side View) By intercalating into the inner leaflet of the lipid bilayer, the amphipathic helix may displace lipids, thereby generative negative membrane curvatures. This could be why the amphipathic helix is absolutely required for IFITM antiviral activities.

IFITM Activities during Enveloped Virus Entry.

During viral infection, viral gene products induce the production of type I interferons, which upon secretion and subsequent receptor binding induce the expression of hundreds of antiviral interferon-stimulated genes (ISGs), including the

IFITMs (161). Once elevated, IFITM family members 1, 2 and 3 line the limiting membranes of the cell to restrict enveloped virus entry, preventing fusion between viral and host cell membranes (162–164). When endogenously expressed, the localization of each IFITM is tightly regulated, with IFITM1 present at the cell surface, IFITM2 in early endosomes, and IFITM3 in late endosomes and lysosomes (164, 165). The susceptibility of enveloped viruses to individual IFITMs is therefore dictated by the subcellular sites of virus-cell membrane fusions (165). In addition, IFITMs can incorporate into the membranes of budding enveloped viruses, reducing their fusion

24 potential in a process termed “negative-imprinting” (166, 167). Virus-incorporated

IFITMs appear to be more suppressive than IFITMs in target cell membranes (166), possibly because viruses-incorporated IFITMs cannot be bypassed by alternative entry routes. In addition to being present on cell membranes and viruses, notably IFITMs also incorporate into EVs during viral infections (168). IFITM-containing EVs bind to cells, reducing their permissiveness to viral infection by unclear mechanisms.

While mechanisms of IFITM-mediated fusion inhibition are not fully elucidated, it is known that the first alphahelical domain is essential for its antiviral activities. This domain consists of an amphipathic helix that intercalates into the inner leaflet of the lipid bilayer (156), and may induce fusion-inhibiting negative membrane curvatures (Figure

6, Side View). In addition, there is some evidence that IFITMs reduce membrane fluidity (169), which could restrict membrane fusogenicity. IFITM multimerization is also critical for fusion inhibition (158), potentially because such interactions aid in concentrating IFITMs on cell membranes.

While alternative explanations involving speculative IFITM-induced, fusion- inhibitory cell signaling pathways have not been entirely ruled out, the current literature suggests that IFITMs act locally at sites of virus-cell membrane fusion to block membrane mergers. Local IFITM activities may involve interactions with the platforms that catalyze membrane fusions. IFITMs are post-translationally palmitoylated, which facilitates their incorporation into TEMs (170, 171). There, they can alter or displace contacts between tetraspanins and their partner proteins. For example, IFITM5 disrupts interactions between the tetraspanin CD9 and CD81 (136). In addition, IFITM1

25 dysregulates HCV co-receptor interactions, and this correlates with inhibition of HCV- cell membrane fusion (172).

Non-Viral IFITM Activities.

In the absence of viral infection, IFITM family members are also linked to a diverse array of biological activities, including cell fate and development, tumor malignancy, and maintenance of bone morphology (136, 173–177). The importance of

IFITMs outside of viral infection is exemplified in the autosomal – dominant severe brittle bone disease Osteogenesis Imperfecta (OI) type V, which is caused by polymorphisms in the ifitm5 locus (178–180). OI type V patients have fragile bones that are prone to fracture, even in the absence of trauma (180, 181). IFITM5 is expressed in the skeletal system (177), as well as in some immune cells, including monocytes (182).

In bone-depositing, mesenchymal osteoblasts, it is known to suppress bone mineralization, which intriguingly correlates with the ability of IFITM5 to disrupt CD9 and

CD81 interactions (136). While interferon-induced IFITMs control membrane fusion,

IFITM5 is not responsive to interferons, and whether this proteins has diverged functionally is unclear. Given the homology of IFITM5 to its interferon-induced counterparts (183, 184) however, it is possible that this protein acts to control membrane fusions in the skeletal system, such as monocyte cell-cell fusions.

Elucidating the molecular mechanism of OI type V is critical, as canonical treatments are ineffective against this form of OI, and may even exacerbate disease (185).

Two mutations in IFITM5 are known to cause OI. One mutation, (c.-14C > T), generates a new translational start site that adds five residues (MALEP) to the N- terminus of the protein (178, 179). Murine models of MALEP-IFITM5 – induced OI

26 indicate that this mutation significantly disrupts fetal skeletal development and leads to perinatal lethality (186). Expression of MALEP-IFITM5 in bone-depositing osteoblasts decreases bone mineralization and collagen secretion (186–188), suggesting that this mutant disrupts bone deposition. MALEP-IFITM5 patients, however, also accumulate large bone deposits that are not resorbed, particularly at sites of fracture, suggesting that bone formation is not simply reduced, but also dysregulated. A second mutation,

(c.119C > T), generates a serine to leucine mutation in the predicted second transmembrane span of the protein (IFITM5(S40L)) (180). IFITM5(S40L) also leads to

OI, but symptoms appear distinct from the MALEP-IFITM5 mutation. Rather than accumulating large deposits of bone, patients harboring IFITM5(S40L) have thin bones with no observable deposits. Increased bone resorption in these patients suggests that bone-resorbing osteoclasts may contribute to disease. Overall, these phenotypes suggest that IFITM5 plays a central role in controlling skeletal maintenance, and also suggest that the two identified IFITM5 mutations may cause distinct forms of disease

(189).

Purpose of Dissertation

A review of the literature suggests that tetraspanins and TEMs may act generally to regulate extracellular membrane fusion events. However, the mechanisms by which

TEMs control these fusion events have not been fully explored. We aimed to further understand extracellular membrane fusions by focusing on these potential fusion platforms. We hypothesized that TEMs are platforms for virus, EV and cell membrane fusions, and that TEM composition regulates these fusion events. We tested these hypotheses using three distinct experimental systems.

The first system evaluated virus-cell fusions, using MERS-CoV as a model. The tetraspanins clearly promote interactions between the cofactors that many enveloped viruses use to fuse with cells in culture (141, 146, 147). However, it is unclear whether

TEMs are critical for viral infections in vivo, and their roles during CoV entry remain unexplored. We predicted that tetraspanins and TEMs would control CoV membrane fusions both in cell culture and in vivo.

The second system evaluated vesicle – cell fusions, using EVs as a model. EVs ferry their luminal contents between cells, but whether membrane fusions facilitate EV cargo deliveries is contentious. In addition, while tetraspanins and IFITMs incorporate into EVs (121, 168), potential mechanisms of EV-cell fusion remain unexplored, and it is unknown whether they control EV cargo transfer. We speculated that EV-cell fusions drive cargo deliveries and intercellular communications, and that tetraspanins and TEM resident proteins regulate these fusion events.

The third system evaluated cell – cell fusions, using monocytes as a model. We hypothesized that IFITM5 restricts monocyte cells fusions, and that restriction is dysregulated by disease – causing IFITM5 mutations. Given the ability of IFITM5 to disrupt CD9 and CD81 interactions (136), we also speculated that IFITM5 may act through tetraspanins to exert these effects.

In all experimental systems, we monitored the transfer of cargos between cells, to gain insights into the membrane platforms regulating vesicle – cell, virus – cell, and cell – cell fusion events. Our results identify tetraspanins and TEM resident proteins as central components in a striking variety of membrane fusion processes, illustrating how

28 membrane dynamics in diverse environments are controlled by a limited set of common elements.

CHAPTER TWO

MATERIALS AND METHODS

Cells and Reagents

293T, HeLa, and murine mesenchymal stem cells (MSCs) were maintained in

Dulbecco's Modified Eagle Media (DMEM) supplemented with 10% fetal bovine serum

(FBS, Atlanta Biologicals), 10 mM HEPES, 100 mM sodium pyruvate, 0.1 mM non- essential amino acids,100 U/ml penicillin G, and 100 μg/ml streptomycin (HyClone).

Mouse astrocytoma DBT cells were maintained in MEM-7.5 (minimal essential medium

[MEM] supplemented with 7.5% FBS, 10% tryptose phosphate broth [Difco], 100 IU/ml penicillin, 1 mg/ml streptomycin, and 2 mM L-glutamine). Murine macrophage

RAW264.7 cells were maintained in Minimum Essential Alpha Medium (α-MEM) (Life

Technologies) supplemented with 10% fetal bovine serum (FBS) (Atlanta Biologicals) and 1% penicillin-streptomycin solution (Thermo Scientific). Cells were maintained in a

o humidified environment at 37 C and 5% CO2.

Polyclonal rabbit antibodies directed against GFP were obtained from Katherine

Knight (Loyola University Chicago). Monoclonal mouse antibodies against CD9 (clone

M-L13), CD63 (clone H5C6), and CD81 (clone JS-81) were obtained from BD

Pharmingen. Mouse anti-KDEL antibodies were obtained from Stressgen Bioreagents.

Polyclonal rabbit IFITM3 antisera (ab74699) were obtained from Abcam. Rabbit anti-

FLAG antibodies were obtained from Sigma Aldrich. Hoechst 33258, goat-anti-mouse-

AlexaFluor 647, goat-anti-rabbit-AlexaFluor 488, goat-anti-mouse-AlexaFluor 488 and goat-anti-rabbit-AlexaFluor 568 antibodies were purchased from Invitrogen.

AlexaFluor 647-conjugated wheat germ agglutinin was obtained from Molecular Probes.

Actin was detected with anti-β-actin peroxidase (Sigma). Goat anti-mouse and goat anti-rabbit HRP conjugated antibodies were purchased from Thermo Scientific. Murine recombinant IFN-β was purchased from R&D Systems. 50 mM Amphotericin B (Fischer

Scientific) stock solutions were prepared in DMSO. Murine sRANKL was obtained from

Peprotech.

Plasmids and Cloning

Codon-optimized MERS-CoV S containing a C9 tag was purchased from

Genscript and subsequently cloned into pcDNA3.1+ between the EcoRI and NotI restriction sites. C-terminal Flag-tagged human DPP4 (hDPP4) plasmid pCMV6-Entry- hDPP4 (catalog no. RC209466) (CMV stands for cytomegalovirus) was purchased from

OriGene. pLVX-CD63 was kindly provided by Adarsh Dharan (Loyola University

Chicago). Rluc8155-156DSP(1-7) and Rluc8155-156DSP(8-11) were kindly provided by

Zene Matsuda (University of Tokyo). pEGFP-C1 was a gift from Chris Wiethoff (Loyola

University Chicago). psPax2 was a gift from Didier Trono (Addgene plasmid #12260). pLentiCRISPR v2 was a gift from Feng Zhang (Addgene plasmid #52961)(190). To generate membrane tethered DSPs, sequences encoding the N-terminal 15 residues of c-Src (sequence:

ATGGGGAGCAGCAAGAGCAAGCCCAAGGACCCCAGCCAGCGCCGG) were introduced at the 5’ ends of DSP genes. The DSPs were subsequently cloned into pCAGGs-MCS (191) between ClaI and NheI restriction sites. S15-mCherry was

31 previously constructed (192). The pHEF-VSV-G plasmid was obtained from BEI

Resources. pCMVSport6-hCD9 was purchased from Open Biosystems and subsequently cloned into pLVX between EcoRI and ApaI restriction sites. The CD81

ORF was cloned from pEE6-CD81 into pLVX between EcoRI and ApaI restriction sites. pCMV-Sport6-hIFITM3 was purchased from open Biosystems and subsequently cloned into pCAGGS between SacI and XhoI restriction sites. The pCAGGS vector encoding for the spike (S) fusion protein of murine hepatitis virus (MHV) strain A59 (pCAGGS-

MHV-A59S) was previously constructed (193), as was pCAGGS-TMPRSS2-FLAG (34).

The cognate receptor of MHV-A59S, CEACAM, was cloned into pcDNA3.1(+) between

HindIII and XhoI restriction sites. pLVX and ΔNRF were kindly provided by Edward

Campbell, Loyola University Chicago. Human natural and MALEP-IFITM5 were purchased from GenScript. A C-terminal Flag tag was added to both IFITM5s during subsequent cloning into pLVX between XhoI and EcoRI restriction sites, generating pLVX-IFITM5-FLAG and pLVX-MALEP-IFITM5-FLAG. pLVX-IFITM5(S40L)-FLAG was generated using NEBuilder HiFi DNA Assembly Cloning Kit (New England BioLabs), template pLVX-IFITM5-FLAG and the following primers: forward, 5’–

ACCACTTGATCTGGTTGGTGTTCAGCACCC – 3’; reverse 5’ –

GGGTGCTGAACACCAACCAGATCAAGTGGT – 3’; forward, 5’ –

CCTGTGTGAAATTGTTATCCGCTCAC – 3’; reverse, 5’ –

GTGAGCGGATAACAATTTCACACAGG – 3’.

Transfections

Plasmid DNA transfections were performed with polyethylenimine (PEI;

Polysciences) at a ratio of 1 μg DNA:3 μg PEI. All DNAs were transfected into cells at a

32 concentration of 0.5 μg / million cells.DNA/PEI complexes were incubated in Opti-MEM

(Life Technologies) for 20 min at room temperature before they were added dropwise to adherent cells. 18 h post-transfection, cells were rinsed with PBS and incubated in cell culture media. 293T cell transfection efficiencies were measured using pEGFP-C1.

Transfection conditions typically resulted in >90% EGFP+ cells.

Generation of CD9 Knockout Cells

pSpCas9-BB-2A-puro was digested with Esp3I (Fermentas) for 4h at 37°C. The digested plasmid was purified and ligated with annealed guide DNAs specific for CD9 or

CD81. Tetraspanin-specific pSpCas9-BB-2A-puro plasmids were transfected into 293T cells. After 72h, cells were selected with 4 μg/ml puromycin for 96h. Selected cells were serially-diluted to isolate clonal populations and clones were selected by western blot.

Pseudoviruses

HIV pseudoviruses were produced as previously described [84]. Briefly, 293T cells were co-transfected with pNL4.3-HIV-luc and pcDNAs encoding appropriate glycoproteins. After two days, supernatants were collected, centrifuged at 10,000 x g at

4°C for 10 minutes to remove debris, and stored in aliquots at -80°C. VSV pseudoviruses were produced by the methods of Whitt, 2010 [40]. Briefly, 293T cells were transfected with plasmids encoding viral glycoproteins. Two days later, cells were inoculated for 2h with VSVΔG-luciferase, rinsed extensively and incubated for one day.

Supernatants were collected, centrifuged at 10,000 x g at 4°C for 10 minutes to remove cellular debris, and stored in aliquots at -80°C.

Pseudovirus transductions were carried out by incubating target cells with pseudoviruses for 1h at 37°C. Following initial incubation, unadsorbed viruses were

33 removed by washing thrice with PBS. Complete media was placed on the cells and incubated for 18h for VSV or 48h for HIV at 37°C. At the end of transduction periods, cells were dissolved into cell culture lysis buffer (25 mM Tris-phosphate [pH 7.8], 2 mM

DTT, 2 mM 1,2-diaminocyclohexane-N,N,N ′,N ′-tetraacetic acid, 10% glycerol, 1%

Triton X-100) and luciferase levels were measured by addition of firefly luciferase substrate (1 mM D-luciferin, 3 mM ATP, 15 mM MgSO4·H2O, 30 mM HEPES [pH 7.8]) using a Veritas microplate luminometer (Turner BioSystems, Sunnyvale, CA).

Isolation of Tetraspanin-Enriched Membranes

Adherent 293β5 cells (~105 / cm2 ) were rinsed with ice-cold PBS, incubated for

30 min at 4o C with 1 mg / ml EZ-Link Sulfo-NHS-LC-Biotin (Pierce) in PBS, rinsed, then incubated for 20 min at 4o C with 100 mM glycine in PBS. Cells were rinsed with

PBS, then incubated for 20 min at 4o C in MES buffer (25 mM MES [pH 6.0], 125 mM

NaCl, 1 mM CaCl2, 1 mM MgCl2) containing 1% 3-[(3- Cholamidopropyl) dimethylammonio]-1-propanesulfonate (CHAPS) detergent (Calbiochem Cat # 220201) or 1% TritonX-100 detergent (Sigma). Cell lysates (107 / ml) were removed from plates and emulsified by 20 cycles of extrusion through 27G needles. Nuclei were removed by centrifugation, lysates mixed with equal volumes of 80% w/v sucrose in MES buffer, placed into Beckman SW60 tubes, and overlaid with 3 ml of 30% w/v sucrose, then with

0.5 ml of 5% w/v sucrose, both in MES buffer. Samples were centrifuged with a

Beckman SW60 rotor at 370 K x g for 18 h at 4o C. Fractions were collected from air- gradient interfaces. Biotinylated proteins in gradient fractions were bound to streptavidin agarose beads (Pierce). Non-reducing dot- and western-blotting procedures were used

34 to identify the distributions of proteins in gradient fractions, as described previously

[162].

Recombinant Adenovirus Production

Recombinant adenovirus vectors were produced by the University of Iowa Gene

Transfer Vector Core. To generate TMPRSS2-expressing adenoviruses, hTMPRSS2 containing a C-terminal FLAG tag was cloned into the pAd5CMV shuttle vector between

XhoI and EcoRI restriction sites. To generate shRNA—expressing adenoviruses, gene blocks containing an shRNA targeting either the coding region of CD9 (target sequence:

CCGATTCGACTCTCAGACCAA) or the 3' UTR of TMPRSS2 (target sequence:

ACACTAGAGTGGATGAATGTCTGGA), flanked by the U6 promoter and RNApolIII termination sequence, were purchased from GenScript. These gene blocks were subcloned into the pacAd5k-NpA E1 shuttle vector between the KpnI and EcoRI restriction sites. Shuttle vectors were linearized and transfected into HEK 293 cells along with a linearized Ad5 backbone containing an RSV promoter -driven hDPP4 in the

E3 region. Homologous recombination in HEK 293 cells yielded recombinant adenovirus encoding both the shRNA and hDPP4. Titers of purified recombinant adenoviruses ranged from 1010−1011 pfu/ml.

Transduction and Infection of Mice

Isoflurane-anesthetized mice were transduced intranasally with 2.5 x 108 pfu of the indicated Ad5 virus in 75 μl of DMEM. 5 days posttransduction, mice were infected intranasally with 105pfu of MERS-CoV in a total volume of 50 μl DMEM. At 2 d.p.i., mice were euthanized by isoflurate inhalation followed by cervical dislocation. Lungs were removed into PBS and manually homogenized. Virus was plaqued on Vero81 cells.

Cells were fixed with 10% formaldehyde and stained with crystal violet 3 d.p.i. All work was performed in the University of Iowa Biosafety Level 3 (BSL3) Laboratory.

EV Isolation

EV producer cells were co-transfected with the indicated DNAs by PEI transfection.

24-36h post-transfection, cells were incubated with serum free DMEM (1mL / million cells) for 12 h at 37°C. Where indicated, 100nM Bafilomycin A1 (Sigma-Aldrich) or DMSO vehicle control were added to media before application to cells. Conditioned media were subjected to sequential centrifugation at 300xg for 10 min, 2000xg for 10 min, and

10,000xg for 30 min to remove cells, cell debris, and apoptotic bodies, respectively, to leave EVs in the 10,000xg supernatants. For microscopic analyses, EVs were collected in serum-free DMEM lacking phenol red.

Transmission Electron Microscopy

EVs were pelleted by ultracentrifugation and resuspended in PBS containing 2%

PFA and processed by the methods of Thery, 2006 (194). Briefly, EVs were deposited onto Formvar-carbon coated EM grids and allowed to adsorb for 20 min. Grids were rinsed once in PBS, fixed in 1% glutaraldehyde for 5 min, and subsequently rinsed 8 times in distilled water. Fixed EVs were then contrasted using uranyl-oxalate for 5 min, and embedded in 2% methyl cellulose containing 4% uranyl acetate for 10 min on ice.

Excess methyl cellulose was removed and grids allowed to air dry. Grids were subsequently observed using a Philips CM120 Transmission Electron Microscope.

Western Blot Analyses

EVs in 10,000xg supernatants were subjected to ultracentrifugation under conditions that pellet all particles with sedimentation coefficients >100 S. Where

36 indicated, sodium dodecyl sulfate (SDS) was added to 10,000xg supernatants to a final concentration of 0.1% prior to ultracentrifugation. Pelleted vesicles were rinsed by resuspending in PBS and re-pelleting by ultracentrifugation. EVs and cells were dissolved on ice for 15 min with TX-100 buffer (1% Triton X-100, 25 mM Tris-HCl [pH

7.4], 125 mM NaCl, 0.1% protease inhibitors, 1 mM MgCl2, 1 mM CaCl2) and debris removed by centrifugation at 1000xg for 10 min at 4°C. The TX-100 extracts were then mixed with 1/6th volume of 6x SDS buffer (0.375 M Tris HCl [pH 6.8], 10% SDS, 6% 2- mercaptoethanol, 30% glycerol) and heated to 95°C for 5 min. When probing for tetraspanin proteins, 2-mercaptoethanol was not added to SDS buffer. A total of 50,000 cell equivalents, in parallel with the indicated cell equivalents of EV lysates, were subjected to sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE).

Separated proteins were transferred to nitrocellulose membranes (BioRad), the membranes incubated in Tris-buffered saline with tween (20mM Tris-HCL, 150mM

NaCl, 0.1% Tween-20 [pH 7.6]) (TBST) containing 5% milk for 1 hour at room temperature, and subsequently incubated with the appropriate primary antibodies at a

1:1000 dilution in TBST-1% milk for 1 hour at room temperature. Membranes were then rinsed thrice in TBST, incubated with the appropriate secondary antibodies at a 1:5000 dilution in TBST-1% milk. Finally membranes were rinsed, treated with ECL substrate

(Pierce), and detected with FluorChem E and analyzed with Alphaview software

(Protein Simple).

Quantifying EV Concentration

293T cells were transfected with either DSP8 or S15-DSP8, along with

TMPRSS2, CD9, CD63, CD81, or a vector control. 24-36 h post-transfection, EVs were

37 collected during a 12 h of incubation period in serum free DMEM, in the presence or absence of 100 nM bafilomycin A1. After media were clarified, EVs were observed by nanoparticle tracking analysis using a NanoSight LM10-HS (Malvern Instruments) at the

Keck Biophysics Facility (Northwestern University), or using a NanoSight 300 (Malvern

Instruments) at Capricor Therapeutics (Beverly Hills, CA). 3 samples were analyzed to determine vesicle concentration and size distribution for each condition. Videos were processed identically post-acquisition.

Quantifying EV Cargo Secretion by Luciferase Assay

293T-DSP1/8 cells were modified through addition of S15-tethers to the DSPs, or through co-transfection with TMPRSS2, CD63, CD9, or CD81. 24-36 h post- transfection, EVs were collected from the cells over a 12 h period with or without bafilomycin A1. EV-containing media were mixed with 1/5th volume of 5x Passive Lysis

Buffer, EV producer cells lysed in 1x Passive Lysis Buffer, and luminescence measured by addition of Renilla luciferase substrate, and quantified as RLU using a Veritas microplate luminometer. Rluc RLU in EV-containing media and producer cells were compared to quantify the proportion of total Rluc in the media.

Quantifying Intercellular Transfer of DSPs

293T-DSP1, DBT-DSP1 or MSC-DSP1 cells were inoculated with conditioned media from corresponding DSP8 cells. After 2 h, DSP1 cells were rinsed once in PBS, lysed in Passive Lysis Buffer, and luminescence was measured by addition of Renilla luciferase substrate, and quantified as RLU using a Veritas microplate luminometer.

Modification of DSP8 cells was required to yield detectable DSP8 transfers into DSP1 cells. To yield detectable DSP8 transfers, unless otherwise indicated, 293T EV

“producer” cells were routinely co-transfected with pCAGGS-hTMPRSS2-FLAG andRluc8155-156DSP(8-11). In some experiments, as indicated, EV-containing media were normalized to Rluc RLU content as described above. Also where indicated, DSP8

EVs were pre-bound to cells as described above.

Monitoring Intercellular Transfer of DSPs by Fluorescence Microscopy

EV-containing media were isolated from 293T cells transfected with S15DSP8 and VSVG. After removing cells and cellular debris by centrifugation, these EV containing media were then incubated with 293T-S15DSP1 cells for 2 hours at 37°C.

Cells were then rinsed thrice in PBS and fixed for 30 minutes at 37oC with 3.7% paraformaldehyde (PFA) in 0.1 M piperazine-N,N′-bis(2-ethanesulfonic acid) [pH 6.8]

(PIPES) buffer. Cells were blocked for 1 hour in PBS containing 2% FBS, and subsequently stained with AlexaFluor 647-conjugated wheat-germ agglutinin (Molecular

Probes) and Hoescht for 20 minutes to label cell surface lectins and nuclei, respectively.

Cells were then rinsed and mounted onto slides. Images were captured using a

DeltaVision microscope (Applied Precision) equipped with a digital camera (CoolSNAP

HQ; Photometrics), using a 1.4-numerical aperture 60X objective lens. Images were deconvoluted with SoftWoRx deconvolution software (Applied Precision). Deconvolved imaged were analyzed for transferred and complemented DSPs (GFP+ puncta) usingthe Spots Mode of the Imaris 6.3.1 software package (Bitplane Scientific

Solutions). Algorithms were designed that detected local fluorescent signals with a quality above background fluorescence.

Analysis of EV-Cell Binding

24-36 h post-transfection, EVs were collected in serum free media over 12 h from

293T-DSP1/8 cells transfected with the indicated proteins. A portion of the supernatant was mixed with 1/5th volume of 5x Passive Lysis buffer (Promega), and luminescence measured by addition of Renilla luciferase substrate and quantified as RLU using a

Veritas microplate luminometer. Serum free DMEM was then used to normalize Rluc levels in supernatants across all conditions. Normalized EV-containing media were then bound to 200,000 target cells plated in 24 well plates, at a multiplicity of 150 RLUs per cell, for 2 h, using 13°C to prevent EV internalization and cargo delivery. Cells were then rinsed once in PBS, lysed in 1x Passive Lysis Buffer, and luminescence was measured by addition of Renilla luciferase substrate, and quantified as RLU using a

Veritas microplate luminometer. The percent of Rluc-containing EVs bound to cells was inferred by comparing Rluc RLU signals on cells to those in media.

Production of IFITM3 KO Cells

To generate IFITM3 – targeting CRISPR-Cas9 lentiviral vectors, the pLentiCRISPRv2 plasmid was digested with BsmB1 for 30 min at 37°C. Oligonucleotides containing guide DNA targeting IFITM3 (target sequence: CTGCCAGTGGAGGACAGCCC) were subsequently ligated into the digested plasmid. To produce CRISPR-Cas9 – containing pseudoviruses to knockout IFITM3, 293T cells were transfected withpHEF-VSV G, psPax2, and the IFITM3 – targeting pLentiCRISPRv2. 2 days post-transfection, pseudovirus – containing media were collected and centrifuged at 1000xg for 10 min to remove cell debris. Pseudoviruses were then used to transduce DBT cells. 2 days

40 post-transduction, DBT cells were selected with puromycin (4 µg/mL) for 2 weeks.

IFITM3 knockout was confirmed by Western blot analysis.

Quantifying IFNβ – Mediated Blockade of EV Cargo Transfer

Serum free DMEM containing the indicated doses of murine recombinant IFN-β (R&D

Systems) were applied to WT or IFITM3 KO DBT-DSP8 cells that had been co- transfected with TMPRSS2 at 24-36 h post-transfection. 12 h later, conditioned media were collected, clarified, and applied to corresponding WT or IFITM3 KO DBT-DSP1 cells. 2 h later, media were removed, cells rinsed in PBS, and lysed in Passive Lysis

Buffer. Luminescence in target cells was then measured by addition of Renilla luciferase substrate and quantified as RLU using a Veritas microplate luminometer.

In Vitro EV-Virus and EV-EV Fusion Assays

Pseudoviruses for EV-virus fusion assays were generated by co-transfecting 293T cells with HEV-VSV G, psPax2, and S15-DSP8. EVs for EV-virus fusion assays were generated by transfecting 293T cells with either S15-DSP8, and by co-transfecting separate 293T cells with S15-DSP1, S15-mCherry, and IFITM3 or vector control. 24-

36 h post-transfection, all cells were incubated with serum free, phenol red – free

DMEM for 12 h at 37°C. Conditioned media were then collected and clarified. EVs containing S15-DSP1 and mCherry, with or without IFITM3, were then mixed with S15-

DSP8 – containing EVs, or pseudoviruses. Mixed virus and EV containing media were subsequently incubated for 4h at 37°. In the context of virus-EV fusion, media pH was then adjusted to 5.2 and incubated for 30 min at 37°C before being subsequently neutralized to pH 7.2. EVs were chilled to 4oC and centrifuged onto coverslipsat

2,000xg for 2 h, fixed with 3.7% PFA in 0.1 M PIPES buffer, mounted onto coverslips, and observed by fluorescence microscopy.

Fluorescence Microscopic Analyses of EVs

EVs on coverslips were permeabilized in PBS containing 0.2% Triton-X 100 for 5 min, exposed to PBS containing 2% FBS for 1 h, and subsequently incubated withrabbit anti-

IFITM3, mouse anti-CD9 or mouse anti-CD63 antibodies, at a 1:500 dilution, in PBS containing 2% FCS for 1 hour at room temperature. Coverslips were then rinsed, incubated with secondary antibodies, at a 1:1000 dilution, in PBS containing 2% FCA for 1 hour at room temperature, mounted onto slides using PermaFluor Mountant, and observed by fluorescence microscopy. Images were captured using a DeltaVision microscope (Applied Precision) equipped with a digital camera (CoolSNAP HQ;

Photometrics), using a 1.4-numerical aperture 60X objective lens. Images were deconvoluted with SoftWoRx deconvolution software (Applied Precision). Deconvolved imaged were analyzed for EV puncta usingthe Spots Mode of the Imaris 6.3.1 software package (Bitplane Scientific Solutions). Co-localizations between IFITM3+ and CD63+,

CD9+, or mCherry+ puncta were also analyzed using the Spots Mode of the Imaris software package. Algorithms were designed for each experiment that detected local fluorescent signals with a quality above background fluorescence. Algorithms were applied to each image in the data set, and puncta enumerated using Batch. For colocalization, the fluorescent intensity of the other signal of interest within each spot was used to assess co-localization.

Production of ifitm5 – Encoding Lentiviral Particles and Stable Cell Lines

10 cm dishes of 293T cells were transfected with 2.7 µg of ΔNRF, 2.7 µg pHEF-VSV-G, and 2.7 µg of pLVX containing CD9, the appropriate IFITM, or vector control using PEI transfection. 48 hours after transfection, virus-containing supernatants were harvested and centrifuged at 1,000 x g at 4oC for 10 minutes to clear them of cellular debris. The cleared supernatants were aliquoted and stored at -80oC until use. RAW264.7 cells were transduced with the appropriate retroviral vector, and after 4 days cells were selected with 2 μg/mL puromycin.

Immunofluoresence Microscopy of RAW Cells

RAW264.7 cells transduced with retroviral vectors encoding for IFITM5-FLAG, MALEP-

IFITM5-FLAG, or IFITM5(S40L)-FLAG were rinsed with PBS and fixed with 3.7% paraformaldehyde in 100 mM piperazine-N,N′-bis(2-ethanesulfonic acid) (PIPES) buffer

(pH 6.8). When appropriate, cells were then permeabilized in PBS containing 0.2%

Triton-X 100 for 5 minutes at RT. Cells were then rinsed and incubated in PBS containing 2% FBS for 1 hour at room temperature (RT). Antibodies were added, and cells were incubated for 1 hour at RT. Cells were rinsed with PBS, and Alexa Fluor- conjugated secondary antibodies were applied for 1 hour at RT, along with fluorescently labeled wheat germ agglutinin and Hoechst 33258 (Molecular Probes). Cells were mounted using PermaFluor Mountant (Thermo Scientific), and imaged with a

DeltaVision microscope (Applied Precision) equipped with a digital camera (CoolSNAP

HQ; Photometrics). Images were deconvolved with SoftWoRx deconvolution software

(Applied Precision). Colocalization was measured and quantified using Imaris version

6.3.1 (Bitplane Scientific Solutions).

Osteoclast Formation Assay

RAW264.7 cells expressing the indicated IFITM or vector control were plated on coverslips at a density of 5000 cells/cm2 and stimulated with 50 ng/mL sRANKL. When appropriate, cells were inoculated with 250 ng/mL amphotericin B or DMSO control 24 hours post sRANKL stimulation. 5 days after sRANKL stimulation, cells were rinsed in

PBS and fixed with 3.7% p-formaldehyde in 100 mM PIPES buffer.

Immunofluorescence microscopy was performed as indicated above. The percent of total nuclei in multinucleated cells was then quantified from 1000 cells, counted in at least 20 different fields for each experiment, from three experiments.

Osteoclast Differentiation Assay

RAW264.7 cells expressing the indicated IFITM or a vector control were plated at a density of 5000 cells/cm2 and stimulated with 50 ng/mL sRANKL. 5 days later, cells were fixed with 3.7% p-formaldehyde in 100 mM PIPES buffer and stained for tartrate – resistant acid phosphatase with the Acid Phosphatase, Leukocyte (TRAP) kit (Sigma-

Aldrich) according to manufacturer's instructions. Microscopy was performed on a Leica

DM IRB microscope (Leica Microsystems) using a MagnaFire 2.1C digital camera system (Optronics). Images were processed using ImageJ (National Institutes of

Health).

Viral Syncytia Formation Assay

RAW264.7 cells expressing the appropriate IFITM were infected with GFP – expressing

Murine Hepatitis Virus strain A59. 2 hours post – infection, inoculum was removed and cells were incubated in the presence of 250 ng/mL amphotericin B or DMSO control for

16 hours. Cells were then rinsed in PBS and fixed with 3.7% p-formaldehyde in 100 mM

PIPES buffer. Immunofluorescence microscopy was performed on a Leica DM IRB microscope (Leica Microsystems) using a MagnaFire 2.1C digital camera system

(Optronics). The percent of GFP+ nuclei in multinucleated cells was quantified from 500 cells in three experiments.

Statistical Analysis

Statistical comparisons were made by two-tailed Student’s t-test. Data are represented, unless otherwise noted, as means ± SEM.

CHAPTER THREE

RESULTS

SECTION 1: TEMs as Platforms for Virus-Cell Membrane Fusion

The Tetraspanin CD9 Facilitates MERS-CoV Entry.

Given the roles of tetraspanins in the entry of many enveloped viruses, as discussed in Chapter 1, we speculated that they would also control CoV entry, potentially at the level of membrane fusion. Since the MERS-CoV receptor DPP4 binds the tetraspanin CD9 (148), we sought to determine whether this tetraspanin played a role in MERS-CoV entry. To this end, we used CRISPR/Cas9 technologies (195) to eliminate CD9, or alternatively the tetraspanin CD81, from 293T cells. Western blot analyses of cell lysates confirmed CD9 and CD81 knockout (KO), while the expression of the tetraspanin CD63 was unaffected in both cell lines (Figure 7A).

Immunofluorescence analysis (IFA) of unpermeabilized cells confirmed that the KO cells were depleted of the appropriate tetraspanin (Figure 7B). To determine whether

293TKO cells were resistant to MERS-CoV entry, we utilized MERS-S – laden HIV- based pseudoparticle (pp) transductions, which allowed us to limit our readouts to virus- cell entry events. We rendered 293T “wild type” (WT) and KO cells permissive to

MERS-CoV transductions by transfecting them with DPP4, and subsequently transduced them with MERSpps. Notably, CD9KO restricted MERSpp transductions by over 1 log (Figure 7C). Ectopic expression of CD9 rescued MERSpp entry in these

46 cells. The requirement for CD9 was MERS-specific, as SARSpp transductions were not inhibited in CD9KO cells when its ACE2 receptor was ectopically expressed (Figure

7D). CD81KO did not significantly inhibit entry of MERS- or SARSpps, though there was a modest decrease in SARSpp entry that was eliminated when CD81 was ectopically expressed (Figure 7E-F). These results indicated that CD9, but not CD81, is a MERS-CoV entry factor. They also suggested that all CoVs do not rely on the same tetraspanin to facilitate viral entry.

Figure 7. CoV Entry into Tetraspanin KO Cells. (A) Western blot analysis of 293T and HeLa clonal cell lines. Actin and the tetraspanin CD63 are used as loading controls. (B) Immunofluorescent analysis of HeLa clonal cell lines. Unpermeabilized cells were incubated with primary antibodies against CD9, CD81 or CD63 as indicated. 293T WT, CD9KO (C-D), or CD81KO (E-F) cells were transfected with the appropriate receptors and CD9 where indicated. These cells were transduced with viruses pseudotyped with S proteins from MERS (C and E) or SARS (D and F). Pseudovirus transduction was measured by luciferase assay. Experiments performed with the aid of James Earnest.

CD9 Holds DPP4 in TEMs.

Given the ability of tetraspanins to aggregate proteins into TEMs (133), we predicted that CD9 functioned to localize DPP4 to these membrane microdomains. If so, DPP4 would localize to membranes resistant to solubilization by CHAPS detergent

(135). Thus, we ectopically expressed the MERS receptor DPP4, the SARS receptor

ACE2 (196), or the CoV-triggering protease transmembrane protease serine 2

(TMPRSS2) in WT and KO cells, and subjected the cells to surface-biotinylation. Cells were then lysed in CHAPS detergent to dissolve cell membranes without destroying

TEM protein-protein interactions. Lysates were then fractionated along a sucrose gradient. After fractionation, CHAPS-soluble proteins were in the high-density (HD) fractions at the bottom of the gradient, while CHAPS-insoluble protein-lipid complexes were in the low-density (LD) fractions at the top of the gradient. Cell surface proteins in these fractions were subsequently precipitated using streptavidin, and fractions were then probed for CoV receptors and proteases by Western blot. Notably, DPP4 was present in LD membrane fractions only when CD9 was expressed (Figure 8A), while

CD81KO had no effect on DPP4 localization. Localization of ACE2 and TMPRSS2, however, were unaffected by CD9 or CD81 KO (Figure 8B-C). These results suggested that CD9 specifically interacts with DPP4 to hold it in TEMs.

Figure 8. Association of CoV Entry Factors with CHAPS-resistant Membranes in the Presence or Absence of CD9 or CD81. 293T WT, CD9KO, or CD81KO cells were transfected with the CoV receptors DPP4 (A), ACE2 (B), or the protease TMPRSS2 (C). KO cells were also complemented with the appropriate tetraspanin. Cell-surface proteins were biotinylated before cells were lysed in cold CHAPS and cleared lysates were subjected to ultracentrifugation. Cell surface proteins were isolated by streptavidin pulldown and analyzed in high density (HD) and low density (LD) fractions by western blot. Experiment performed by James Earnest.

CD9 Brings DPP4 and TMPRSS2 into Close Proximity.

Our evaluations of CHAPS-resistant membrane fractions suggest that CD9 holds

DPP4 receptors in TEMs, which also contain the CoV-triggering protease TMPRSS2

(Figure 8). Given that MERS-S receptor binding is required for productive triggering

(24), we questioned whether CD9 held DPP4 and TMPRSS2 closely together to facilitate virus entry. We thus determined whether DPP4 and TMPRSS2 were in close proximity on cell surfaces, and whether CD9 facilitated these close interactions. We did this using proximity ligation assays (PLAs) (197), which can determine whether transmembrane proteins are within 40 nm of each other (198). Antibodies tagged with oligonucleotide probes are bound to cells, and close spacing between probes facilitates

49 oligonucleotide hybridization into templates for DNA polymerase, thereby providing substrates for fluorescent DNA synthesis.

CD9KO HeLa cells, which have relatively flat morphologies suitable for PLAs, were incubated with DPP4 and TMPRSS2 antibodies, and after incubation with secondary antibodies tagged with PLA probes and subsequent DNA amplification, fluorescent DNAs were detected by immunofluorescence assay (IFA). Notably, close interactions between DPP4 and TMPRSS2 were rarely observed in CD9KO cells

(Figure 9A), while CD9 ectopic expression increased detectable interactions by ~10 fold

(Figure 9D). When DPP4 or TMPRSS2 were overexpressed, they closely interacted even in the absence of CD9 (Figure 9B, C, E). Interactions were most prevalent when all three proteins were ectopically expressed (Figure 9F). The number of detectable

PLA foci per cell correlated with cell susceptibility to viral transduction (Figure 9G-H).

These data suggest that CD9 facilitates MERS-CoV entry by linking DPP4 and

TMPRSS2 on the cell surface. They also suggest that CD9-mediated linkages are especially important when DPP4 and TMPRSS2 levels are low.

Figure 9. Proximity Ligation Assay of DPP4 and TMPRSS2 in CD9KO Cells. (A-F) HeLa CD9KO cells were transfected with the indicated genes and a GFP reporter before being mounted on microscopy slides. Proximity ligation assay was performed using primary antibodies against hDPP4 and hTMPRSS2. Red foci indicate close proximity of the two proteins. (G) The average number of foci/cell in GFP+ cells in each group was quantified. (H) MERSpp transduction of HeLa cells overexpressing the indicated proteins. Experiment performed by James Earnest.

CD9 and TMPRSS2 Are Critical for MERS-CoV Infection in vivo.

As shown in Figure 9, CD9 links DPP4 with TMPRSS2 on the cell surface of cultured cells. These data also suggest that CD9-mediated linkages are critical for

TMPRSS2-triggered MERS-CoV early entry. However, whether these factors are required for MERS-CoV lung infection was unknown. To identify host factors critical for

MERS-CoV entry in vivo, we developed dual-expressing adenoviral serotype 5 (Ad5) vectors encoding for the MERS-CoV receptor human DPP4 (hDPP4), along with either

TMRPSS2 or shRNAs targeting CD9 or TMPRSS2 mRNAs. In this system, only Ad5- transduced cells are susceptible to MERS-CoV infection, as ectopic expression of

51 human (h)DPP4 is required for MERS-CoV entry into mouse cells (199). Therefore, all cells permissive to viral infection are also depleted of candidate entry factors.

To confirm the Ad5-expressed shRNAs depleted their target genes, mouse LET-

1 cells were transduced with Ad5 vectors, and cells were subsequently lysed and probed for DPP4, CD9, and TMPRSS2 by Western blot or RTqPCR. These assays confirmed that the Ad5 vectors overexpressed and depleted their target proteins in LET-

1 cells (Figure 10 A-B). As expected, DPP4-expressing Ad5 vectors rendered cells permissive to MERSpp entry, and TMPRSS2 overexpression enhanced entry into these cells (Figure 10C). Conversely, TMPRSS2 and CD9 depletion rendered cells less permissive to MERSpps. These data indicated that CD9 and TMPRSS2 are critical entry factors in LET-1 cells, and that Ad5-expressed gene products modulated the expression of these factors.

To determine whether CD9 and TMPRSS2 are required for MERS-CoV entry in vivo, our colleague Dr. Kun Li at the University of Iowa transduced mice intranasally with Ad5 vectors, and after 5 days challenged with MERS-CoVs. 2 days post-infection, lungs were harvested and MERS-CoV titers quantified by plaque assay on Vero cells.

Notably, depletion of either TMPRSS2 or CD9 reduced MERS-CoV titers by ~10 fold and ~20 fold, respectively (Figure 10D). TMPRSS2 overexpression, however, did not enhance MERS-CoV titers, presumably due to high endogenous expression of serine proteases in the mouse lung. These data indicate that, remarkably, MERS-CoV lung infections are largely driven by a single tetraspanin, and a single protease. These results also suggested that MERS-CoVs utilize cell surface, rather than endosomal,

52 factors to enter mouse lung cells.

Figure 10. Analysis of Adenovirus Knockdown of MERS-CoV Entry Factors in Cell Culture and in vivo. (A) LET-1 cells were transduced with an adenovirus carrying a GFP gene or adenoviruses carrying hDPP4 and either an empty vector, the TMPRSS2 gene, or a U6-driven shRNA against TMPRSS2 or CD9. After 3 days, cells were lysed and analyzed by western blot for the indicated proteins. (B) Quantitative rtPCR analysis of Tmprss2 transcripts in cells transduced with rAd5-hDPP4-EV and rAd5-hDPP4- shTmprss2. (C) LET-1 cells were transduced with the indicated Ad5 vector before transducing with VSV-MERSpp. Transduction was measured by luciferase assay. (D) The indicated Ad5-DPP4 vectors were installed intranasally in C57/Bl6 mice. 5 days later, mice were infected with MERS-CoV. Lungs were isolated at 2 dpi and viral titers were measured by plaque assay. Experiments performed with the assistance of James Earnest and Kun Li.

SECTION 2: TEMs as Platforms for Extracellular Vesicle Membrane Fusions

Membrane Fusion Drives EV Content Transfer.

EV-mediated cargo transfers facilitate critical intercellular communications, however, the molecular mechanisms of EV cargo delivery are unknown. This is, in part, due to the lack of simple quantitative assays to monitor EV-cell cytoplasmic exchange.

To develop an approach to measure EV cargo transfer, we looked to technologies developed by Dr. Zene Matsuda and colleagues to monitor cell-cell fusion and cytoplasmic exchange (200). These technologies rely on the EV-mediated transfer of

Dual-Split Renilla luciferase (Rluc): Green Fluorescent Protein (GFP) Protein (DSP) fragments to target cells expressing the complementary DSP fragments. Upon delivery into the cell cytosol, EV-ferried fragments will complement with target cell DSP fragments to form active Rluc and GFP reporters (Figure 11). This approach provides rapid readouts of cytoplasmic exchange that are readily quantified; therefore, we sought to utilize these DSP fragments to monitor EV-cell fusion and cytoplasmic exchange.

Figure 11. Diagram of EV Cargo Transfer Assay Using the DSP Reporter System. DSP8 (green and blue) is incorporated into EVs in 293T producer cells and secreted. Released DSP8 – containing EVs bind to DSP1 expressing target cells, where subsequent EV cargo deliveries mediate DSP complementation and reporter gene expressions (yellow). To facilitate DSP incorporation into EV lumens, we noted that membrane tethered proteins are preferentially loaded into EVs (201), and thus tethered the DSPs to lipid bilayers by appending an acylated membrane anchor, S15

(MGSSKSKPKDPSQRR) (192, 202). First, to determine whether EVs carried DSP cargo, EV-containing media were collected from 293T-S15DSP8 cells and subsequently cleared of cell, cell debris, and apoptotic bodies by differential centrifugation. EVs were then pelleted by ultracentrifugation. EVs could be observed in the pelleted material by transmission electron microscopy, and displayed cup-like morphologies consistent with the literature (Figure 12A) (194). Pellets also contained DSP8 cargo, along with the tetraspanins CD63 and CD9 (Figure 12B), which are enriched in EV membranes (64).

Pre-incubating media with sodium dodecyl sulfate (SDS) prevented these proteins from being pelleted, suggesting they were contained in membranous EVs.

To determine whether the S15 tether enhanced DSP8 loading into EVs, we ectopically expressed both S15-DSP8 and S15-DSP1 in cells, thereby rendering those cells Rluc+. Rluc levels in EV-containing media harvested from DSP1/8 or

S15DSP1/S15DSP8 cells were then measured to quantify DSP cargo release. Many

EVs are under 200 nm in diameter, the limit of detection by light microscopy. We therefore relied on nanoparticle tracking analysis (NTA), which allows one to observe small particles through light-scattering (203), to enumerate vesicles in conditioned media. The results indicated that S15-tethers did enhance EV cargo loading by over 50 fold (Figure 12C, red data points), while only modestly increasing EV output (Figure

12C, blue data points). These results indicated that S15-tethering facilitated DSP loading into EVs. However, inoculation of S15-DSP8+ EVs onto S15-DSP1 target cells did not generate luciferase or fluorescent signals.

We hypothesized that EVs did not deliver S15-DSP8 cargos because they were unable to fuse with target cells. To address this concern, we ectopically expressed the fusion glycoprotein of vesicular stomatitis virus (VSVG) in EV producer cells. Notably, this fusion protein rendered EVs transfer competent (Figure 12D), with luciferase signals indicative of cargo deliveries appearing within 2 hours of incubation at 37°C.

Intracellular GFP puncta could also be observed in some inoculated cells (Figure 12E), providing additional evidence that DSPs were successfully transferred in the presence of a bona-fide virus fusion protein. These findings indicate that membrane fusion catalysts can drive EV cargo delivery, and suggest that 293T cell EVs have relatively poor membrane fusion potential.

Figure 12. Membrane Fusion Drives EV Cargo Transfers. (A) Transmission electron microscopy image of EVs harvested from 293T-S15DSP8 producer cells. Scale bar = 50 nm. (B) Conditioned media from EV-producing cells were incubated with or without 0.1% SDS. EV particles were then pelleted by ultracentrifugation and the EV- associated proteins were subjected to Western immunoblotting to detect DSP8, CD63, CD9 and β-actin. (C) EVs from 293T-DSP1/8 and 293T-S15DSP1/S15DSP8 cells were collected over a 12-h period. Rluc RLU associated with EV-containing media and producer cells were measured and plotted as the proportion of total Rluc in the media (red data points). Data bars represent means ± SEM. In parallel, EVs were collected from conditioned 293T-DSP8 cells and particle concentrations were quantified by nanoparticle tracking analysis (blue data points). Data bars represent means ± SD. (D) EV-containing media were collected from 293T-S15DSP8 cells overexpressing the fusion protein VSVG or a vector control. EVs were then inoculated onto 293T-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM. *p<0.05, compared to control DSP1-containing EV transfers onto DSP1 target cells. (E) DSP complementations in cells incubated with VSVG-laden EVs as in (D) were identified by intracellular GFP+ puncta (white arrows). Wheat-germ agglutinin (red) and Hoescht (blue) labeled cell-surface lectins and nuclei, respectively.

Tetraspanins Promote EV Release and Enable Cargo Transfer. Although cells can express virus-like fusion catalysts (12, 13), their expression does not appear universal (12, 113), and therefore many EV cargo transfers are likely enabled by other factors. Intriguingly, tetraspanin overexpression increases EV secretion (122, 123). Tetraspanins also interact with adhesive proteins (128), and CD9- positive EVs drive fusion-incompetent gametes to fuse through “fusion-from-without”

(151). Due to these known tetraspanin activities, we hypothesized that tetraspanin overexpression would enable EV cargo transfers. Indeed, ectopic expression of CD9,

CD63, or CD81 in EV producer cells permitted EV cargo transfers, with target cell complementation signals readily observed after 2 h of incubation with EV-containing media (Figure 13). Notably, tetraspanins functioned exclusively in EV producer cells to promote EV cargo transfer, as they were inert when expressed in EV target cells.

Figure 13. Tetraspanins Act in Producer Cells to Enable EV Cargo Transfers. EV- containing media were collected from 293T-DSP8 cells overexpressing the indicated tetraspanins or a vector control. EVs were then inoculated onto 293T-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM. *p<0.05, compared to control DSP1-containing EV transfers onto DSP1 target cells.

Tetraspanins could conceivably promote EV cargo transfers by enhancing EV output, EV-target cell binding, or EV-cell membrane fusion. To determine whether tetraspanins acted at the level of EV cargo loading or release, we ectopically expressed tetraspanins in Rluc+ 293T-DSP1/8 cells and quantified Rluc levels in EVs.

Nanoparticle tracking analysis was again used to enumerate vesicles in conditioned media. The results indicated that tetraspanins did enhance both EV outputs (Figure

14A, blue data points) and EV cargo release (Figure 14A, red data points). These results were confirmed by Western blot, which indicated that EVs from tetraspanin overexpressing cells were enriched in both DSP and tetraspanin cargos (Figure 14C).

EV outputs, however, did not correlate with cargo transfer readouts (Figure 14B). For

59 example, CD9 was the most potent facilitator of EV output, the least potent facilitator of

EV cargo deliveries. These results suggested that tetraspanins also act at the EV- target cell binding or fusion stages during cargo transfer.

Figure 14. Tetraspanins Promote EV Secretion and Cargo Transfer. (A) In the same experiments as Figure 12C, 293T-DSP1/8 cells were transfected with the indicated tetraspanins or a vector control. EVs were then collected from the cells over a 12-h period. Rluc RLU associated with EV-containing media and producer cells were measured and plotted as the proportion of total Rluc in the media (red data points). Data bars represent means ± SEM. In parallel, EVs were collected from conditioned 293T-DSP8 cells and particle concentrations were quantified by nanoparticle tracking analysis (blue data points). Data bars represent means ± SD. (B) EVs collected from conditioned 293T-DSP8 cells were inoculated onto 293T-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM. (C) EVs were isolated from tetraspanin- overexpressing cells by ultracentrifugation. Select proteins in the EV-producing 293T- DSP8 cells and the secreted EVs were detected by Western immunoblotting.

Tetraspanins Act at the Level of EV Cargo Delivery to Enable EV Transfers.

To determine whether tetraspanins facilitate EV-target cell binding, Rluc levels in

EV preps were used to normalize DSP cargo loads across all conditions. After normalization, EVs were then bound to target cells at equivalent input multiplicities for 2 h at 13°C, to prevent endocytosis. Unbound EVs were then removed, cells lysed, and bound EV levels quantified with a luminometer. Less than 1% of EVs bound to target cells under these conditions, with tetraspanins having no effect on EV-cell binding

(Figure 15A). To determine whether tetraspanins drove EV cargo deliveries, 293T-

DSP1 cells with bound DSP8+ EVs were incubated at 37°C to allow bound EVs to fuse and deliver their cargo into target cell cytosols. Remarkably, tetraspanin-enriched EVs delivered DSP8 into target cells, while EVs with basal tetrapanin levels did not (Figure

15B). These results indicated that tetraspanins enabled EV cargo delivery at some stage post-binding, likely at the level of membrane fusion. They also suggested that

EVs vary in their transfer competency, with tetraspanin-enriched EVs comprising a fusion-competent EV population.

Figure 15. Tetraspanins Drive EV Content Delivery into Target Cells. (A) EV-cell binding: DSP1/8 (Rluc-positive) EVs were incubated with 293T target cells at equivalent input multiplicities (150 RLUs/cell) for 2h at 13°C. Unbound EVs were removed with a PBS rinse cycle and the remaining cell-associated EVs quantified by Rluc RLU measurements. Data bars represent means ± SEM. (B) EV cargo delivery: DSP8- containing (Rluc-negative) EVs were bound to 293T-DSP1 target cells, and unbound EVs were removed, as described in (A). Cells were then replenished with EV-free media and incubated for 2h at 37oC. DSP8 cargo deliveries into target cells were quantified by Rluc RLU measurements. The percentage of cell-bound EVs that delivered their DSP8 cargoes were inferred from the binding data in (A). Data bars represent means ± SEM.

Additional Enablers of EV-mediated Cargo Transfer.

In addition to tetraspanins, several other conditions promote EV release, including ATP (204), calcium ionophores (205), ER stress agents (206), bafilomycin A1

(baf) (207), and proteases (208). However, it was unclear whether the EVs produced

63 under these conditions could deliver their cargo into target cells. We therefore treated

Rluc+ 293T-DSP1/8 cells with baf, or overexpressed TMPRSS2. As expected, these conditions facilitated the release of both EVs (Figure 16A, blue data points) and DSP1/8 cargo (Figure 16A, red data points). These results were confirmed by Western blotting of EV and cell lysates (Figure 16B). Notably, applying these conditions to 293T-DSP8 cells also endowed EVs with cargo transfer capacity (Figure 16C). Applying the same conditions to EV target cells, however, did not facilitate cargo transfer (Figure 16D).

These data indicate that a wide variety of EV producer cell modifications can endow

EVs with transfer competency, and that these factors act specifically in EV producer cells.

Figure 16. TMPRSS2 and Baf Enhance EV Outputs and Enable EV Content Transfer. (A) In the same experiments as Figure 12C, 293T-DSP1/8 cells were transfected with the protease TMPRSS2 or treated with bafilomycin. EVs were then collected from the cells over a 12-h period. Rluc RLU associated with EV-containing media and producer cells were measured and plotted as the proportion of total Rluc in the media (red data points). Data bars represent means ± SEM. In parallel, EVs were collected from conditioned 293T-DSP8 cells and particle concentrations were quantified by nanoparticle tracking analysis (blue data points). Data bars represent means ± SD. (B) EVs were isolated from modified producer cells by ultracentrifugation Select proteins in the EV-producing 293T-DSP8 cells and the secreted EVs were detected by Western immunoblotting. (C) EVs collected from modified 293T-DSP8 cells were inoculated onto 293T-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM. (D) EVs collected from 293T-DSP8 cells were inoculated onto modified 293T-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM.

Type I Interferons Block EV Cargo Transfer through IFITM3.

While VSVG and tetraspanins were sufficient to enable EV cargo transfers, it was still unclear whether membrane fusions were required for EV cargo delivery. To determine whether membrane fusions are required for EV cargo delivery, we looked for

65 similarities between EVs and enveloped viruses. We turned to type I interferons (IFNs), which suppress multiple stages of viral replication, including virus-cell fusion (161). To determine whether IFNs block EV cargo transfer, we performed transfer assays in DBT cells, which are more responsive to IFN than 293T cells. TMPRSS2-expressing DBT-

DSP8 cells produced EVs that could transfer cargo into DBT-DSP1 cells (Figure 17A), and these transfers were inhibited by IFN-β in a dose dependent manner. IFN-β did not reduce DSP expression or EV output, as determined by Western blot (Figure 17B), indicating that inhibition was not at the level of EV production.

While over 500 IFN-induced genes (ISGs) exist, only one families of ISGs, the

IFITM proteins, are known to generally block virus-cell fusion (162, 164). We thus hypothesized that IFNs suppressed cargo transfer through IFITM3, the only murine

IFITM that is interferon-induced (209). We generated IFITM3 knockout (KO) DBT cells using CRISPR-Cas – expressing lentiviral vectors, and performed EV cargo transfers in the presence of IFN-β. Notably, IFITM3 KO rescued EV cargo transfer in the presence of IFN (Figure 17C-D), indicating that IFN acted through IFITM3 to block EV cargo transfer.

Figure 17. IFN Inhibits EV-directed Cargo Transfer through IFITM3. (A) EV- producing DBT-DSP8 cells were incubated with increasing doses of IFN-β for 12 h, then EVs were harvested and inoculated onto DBT-DSP1 target cells. After 2h at 37oC, DSP8 cargo transfer was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM, (n=3). (B) Selected proteins in DBT-DSP8 producer cells and secreted EVs were detected by Western immunoblotting. IFITM3 was observed as an IFN-β induced product. (C) EV-producing WT and IFITM3 KO DBT-DSP8 cells were incubated with IFN-β for 12 h. EVs harvested from the IFN-β treated WT and IFITM3 KO cells were inoculated onto parallel WT and IFITM3 KO DBT- DSP1 target cells. After 2h at 37oC, complementation was detected by measuring Rluc RLU. Data bars represent means ± SEM, (n=3). (D) Selected proteins in WT and IFITM3 KO DBT-DSP8 producer cells were detected by Western immunoblotting.

IFITM3 Blocks EV Cargo Transfer at the Level of Membrane Fusion.

IFITMs can block virus-cell membrane fusion by lining the limiting membranes of target cells (162, 164), or by incorporating into viral envelopes (166, 167). We hypothesized that IFITMs functioned similarly to block EV-cell membrane fusion. It was therefore possible that IFITM3 acted in either EV producer or target cells to block cargo transfer. To identify the sites of EV transfer blockade, IFITM3 was expressed in either

EV producer or target cells. When expressed in producer cells, IFITM3 incorporated

67 into EVs (Figure 18B), but did not block the secretion of EVs (Figure 18A), EV- associated tetraspanins (Figure 18B), or DSP8, indicating that IFITM3 acted after EV biogenesis. IFITM3-containing EVs also bound to target cells at levels equivalent to

IFITM3-negative EVs (Figure 18C), but were unable to deliver their DSP cargos into target cell cytosols (Figure 18D). It is noteworthy that IFITM3 also blocked EV cargo transfer when expressed in target cells (Figure 18D), but IFITM3 was far more effective when present in EVs. These data indicated that IFITM3 acted to block EV cargo delivery after target cell binding.

Figure 18. IFITM3 Inhibits EV-directed Cargo Transfer. (A) EVs were harvested from IFITM3(-) or (+) 293T-DSP8 cells and particle concentrations were quantified by nanoparticle tracking analysis. Data bars represent means ± SEM. (B) EVs were harvested from IFITM3(-) or (+) 293T-DSP8 cells and pelleted by ultracentrifugation. Selected proteins in pelleted EVs and in producer cell lysates were detected by Western immunoblotting. (C) EVs were harvested from IFITM3(-) or (+) 293T-DSP1/8 producer cells and absorbed onto 293T target cells at equivalent input multiplicities (250 RLUs/cell) for 2 h at 13°C. Media were removed, cells rinsed once in PBS, and the percentage of cell-bound EVs were quantified by Rluc RLU measurements. Data represent means ± SEM. (D) EVs were harvested from IFITM3(-) or (+) 293T-DSP8 producer cells. The EVs were inoculated onto IFITM3(-) or (+) 293T-DSP1 target cells, as indicated under the graphical data. After 2 h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM.

To directly determine whether IFITM3 blocks EV-cell membrane fusions, we noted that vesicle-vesicle fusions can occur in conditioned media (210). We expected these vesicle-vesicle fusions could be registered at the single vesicle level by DSP complementations, and that IFITM3 would block these fusion events. Thus, EVs containing S15-mcherry and DSP8, with or without IFITM3, were co-incubated with HIV- based, VSVG-laden pps containing DSP1 (Figure 19A). VSVpp-EV fusions were readily observed after 30 min of incubation at pH 5.2, and these fusions were inhibited by over a log in the presence of IFITM3 (Figure 19B). Surprisingly, EV-EV fusions were also observed after 4 hours of incubation at 37°C (Figure 19C), and these fusions were also blocked by IFITM3. These data indicated that IFITM3 blocks EV-EV membrane fusions, and strongly suggest that IFITM3 acts similarly to block EV-cell fusion and cargo transfer.

Figure 19. IFITM3 Inhibits in vitro EV-virus and EV-EV Fusions. (A) Diagram of in vitro fusion assay. EVs contained DSP8 and were without (-) or with (+) IFITM3. EVs also contained the red fluorescent EV marker S15-mCherry. EVs were then mixed with DSP1-containing EVs or VSVG pps (A) or not (B). GFP puncta resulting from these co-incubations reflect vesicle- vesicle fusion events. (B) EVs were mixed with DSP1-containing VSV pseudoviruses under acidic (pH 5.0) conditions for 30 min at 37oC, to trigger VSV G protein-mediated fusion catalysis. (C) EVs were mixed with DSP1-containing EVs under neutral (pH 7.0) conditions for 4 h at 37oC. After mixing, the vesicles were pelleted onto coverslips and analyzed by fluorescent microscopy. Green GFP+ punctae (products of EV-virus and EV-EV fusion) were quantified relative to the more abundant mCherry+ punctae representing unfused vesicles. Data bars represent means ± SEM. Experiment performed by Enya Qing.

IFITM3 Preferentially Incorporates into Tetraspanin-containing EVs. Given that EV-incorporated IFITM3 blocked 85-90% of DSP transfers and EV-EV fusions (Figures 18-19), we speculated that IFITM3 was present in the majority of EVs.

Immunofluorescence analysis of EVs from IFITM3 and S15-mCherry expressing producer cells, however, indicated that IFITM3 was actually present in only ~20% of all

EVs (Figure 20A). To explain this discordance, we noted that IFITMs interact with tetraspanins, and hypothesized that IFITM3 preferentially incorporated into tetraspanin- enriched EVs. Indeed, IFITM3 incorporated into ~40% of S15-mCherry EVs containing

CD9 or CD63, while only ~15% tetraspanin-negative S15-mCherry EVs contained

IFITM3 (Figure 20B-C). These data indicate that IFITM3 preferentially incorporates into tetraspanin-enriched, transfer-competent EVs.

Figure 20. IFITM3 Incorporates into a Subset of Tetraspanin – containing EVs. (A) EVs were harvested from 293T-mCherry cells that were without (-) or with (+) IFITM3. EVs were fixed, permeabilized, and stained for IFITM3 (green). The percent of mCherry (red) EVs harboring IFITM3 was quantified using Imaris Software. Data represent means ± SEM, (n=3). Scale bar = 10 µm. (B) EVs harvested from 293T-mCherry- IFITM3 cells were fixed, permeabilized, and stained for CD9 (blue) and IFITM3 (green). The levels of IFITM3 in the CD9-negative and CD9-positive EV populations were quantified using Imaris Software. Data represent means ± SEM, (n=3). (C) EVs harvested from 293T-mCherry-IFITM3 cells were fixed, permeabilized, and stained for CD63 (blue) and IFITM3 (green). The levels of IFITM3 in the CD63-negative and CD63-positive EV populations were quantified using Imaris Software. Data represent means ± SEM, (n=3). Scale bar = 10 µm.

Tetraspanins and TMPRSS2 Enable EV-Directed Transfers in Mesenchymal Stem

Cells.

By using DSP-based approaches, we identified multiple regulators of EV- mediated cargo transfers in 293T cells. We predicted that these factors would also

73 control the delivery potentials of therapeutically-relevant EVs. EVs derived from mesenchymal stem cells (MSCs) contain cargoes that inhibit apoptosis, fibrosis, and inflammation, making them attractive candidates for many EV-based therapies (58,

211). To determine whether tetraspanins and TMPRSS2 enhanced MSC EV delivery potentials, EVs from MSCs co-transfected with DSP8 and either tetraspanins or

TMPRSS2 were applied to MSCs expressing DSP1. Similar to our results in 293T cells,

EVs from MSCs overexpressing tetraspanins or TMPRSS2 delivered their DSP8 cargos into the target cells, while unstimulated MSC EVs were inert (Figure 21). These data indicated that EV cargo transfer facilitators also enhanced the delivery potential of therapeutically-relevant EVs.

Figure 21. Tetraspanins and TMPRSS2 Promote EV Cargo Transfer in MSCs. EVs collected from conditioned MSC-DSP8 cells were inoculated onto MSC-DSP1 target cells. After 2h at 37oC, DSP8 cargo delivery was measured by its complementation with DSP1 into Rluc RLU. Data bars represent means ± SEM.

SYSTEM 3: TEMs as Platforms for Monocyte Cell-Cell Fusions

Disease-causing Mutations Alter IFITM5 Subcellular Localization.

Mutations in IFITM5 lead to severe forms of osteogenesis imperfect (179, 180), indicating that this IFITM plays a critical role in skeletal homeostasis. IFITM5 is expressed in monocytic osteoclast precursors (182), and is homologous to fusion- inhibiting IFITMs (183, 184). We thus hypothesized that IFITM5 controls monocyte membrane fusion, and that disease-causing mutations dysregulate this activity.

Considering this hypothesis, we transduced ifitm5 genes into RAW264.7 cells, a monocyte/macrophage cell line, using lentiviral vectors. RAW cells have high levels of

RANK expression, and fuse into multinucleated osteoclast-like cells upon stimulation with soluble RANK ligand (sRANKL) (212). These membrane fusions occur at the termini of cellular extensions (213). Antiviral IFITMs likely act locally to impede membrane fusion (165), so we reasoned that, if IFITM5 controls pre-osteoclast membrane fusion, it would localize to pre-osteoclast fusion sites. Indeed, IFITM5-FLAG was expressed on the plasma membrane of transduced cells, and was observed at the termini of cellular extensions (Figure 22A). The distributions of IFITM5 and MALEP-

IFITM5 were identical, but notably IFITM5(S40L) was not present on the cell surface

(Figure 22A), and instead was observed in the endoplasmic reticulum (Figure 22B).

This re-localization prevented IFITM5(S40L) from interacting with CD9, a critical regulator of pre-osteoclast membrane fusion (97) (Figure 22C). These results indicated that IFITM5 is present at the sites of pre-osteoclast membrane fusion, and suggested that the S40L mutation may alter the function of IFITM5.

Figure 22. Localization of IFITM5 in RAW Macrophages. (A) Pre-osteoclasts were transduced with retroviral vectors encoding Flag-tagged IFITM5, MALEP-IFITM5, or IFITM5(S40L). 4 days post-transduction, unpermeabilized cells were incubated with anti-FLAG antibodies (red) and wheat germ agglutinin (blue) to mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. Scale bars, 10 μm. (B) Pre-osteoclasts transduced as in (A) were permeabilized 4 days post – transduction. Cells were incubated with anti- FLAG antibodies (red) and anti-KDEL antibodies (green) to mark the positions of Flag- tagged IFITMs and ER – localized proteins, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. Scale bars, 10 μm. (C) Pre-osteoclasts transduced as in (A) were permeabilized 4 days post – transduction. Cells were incubated with anti-FLAG antibodies (red) and anti-CD9 antibodies (green) to mark the positions of Flag-tagged IFITMs and tetraspanin proteins, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. Scale bars, 10 μm. (D) Colocalization of IFITM5 with CD9 was quantified from immunofluorescent images of cells stained as in C using Imaris software. Data represent means ± SD, (n=3). *, p<0.05.

IFITM5 Blocks Osteoclast Formation.

We predicted that surface-localized IFITM5 could control pre-osteoclast membrane fusion, while the intracellular IFITM5(S40L) could not. We thus stimulated ifitm5-transduced RAW cells with sRANKL, and quantified their capacity to fuse into multinucleated osteoclasts. Overexpressed IFITM5 and MALEP-IFITM5 both suppressed osteoclast formation and equally reduced the average number of nuclei per osteoclast, when compared to a vector-transduced control (Figure 23). Notably, overexpression of IFITM5(S40L) had no effect on osteoclast formation. This correlates with the absence of IFITM5(S40L) on the cell surface (Figures 22 and 23B). These data indicate that IFITM5 has the ability to restrict osteoclastogenesis, and that an OI- causing mutation destroyed this restriction. They also suggest that IFITM5 acts locally to impede osteoclast formation, similar to how other IFITMs act locally to restrict membrane fusion (Figure 19).

Figure 23. Effect of ifitm5 Overexpression on Osteoclast Formation.(A) Pre- osteoclasts stably expressing IFITM5, MALEP-IFITM5, IFITM5(S40L), or a vector control were stimulated to form osteoclasts with 50 ng/mL sRANKL. 5 days later, cells were fixed and labeled with anti-FLAG antibodies and wheat germ agglutinin to mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 marked the positions of cell nuclei. The percent of total nuclei in multinucleated cells was then quantified from 1000 cells, counted in at least 20 different fields for each experiment. Data represent means ± SD, (n=3). *, p<0.05 when compared to “NO IFITM” control. (B) Distribution of multinucleated cells from (A) was quantified from 1000 cells. Data represent means ± SD, (n=3). *, p<0.05 when compared to “NO IFITM” control for color – matched condition. (C) Representative images of osteoclast formation in the presence of the indicated IFITM. Anti-FLAG antibodies (red) and wheat germ agglutinin (blue) mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. White arrows mark large, multinucleated cells. Scale bars, 15 μm.

IFITM5 Does Not Influence Pre-Osteoclast Differentiation.

During osteoclastogenesis, monocytes first differentiate into osteoclast precursors before fusing together (214), likely because the catalysts of membrane fusion are only expressed upon differentiation. It was therefore possible that IFITM5 restricted pre-

78 osteoclast differentiation prior to membrane fusion. To determine whether IFITM5 interfered in pre-osteoclast differentiation, we assessed the expression of tartrate- resistant acid phosphatase (TRAP), a protein highly upregulated in osteoclasts and their differentiated precursors (85). After 5 days of RANKL stimulation, IFITM5-expressing

RAW cells were fixed and stained for TRAP. While natural and MALEP-IFITM5 inhibited osteoclastogenesis, TRAP staining was unaffected by the overexpression of these proteins (Figure 24). These results suggested that IFITM5 does not impede pre- osteoclast differentiation, and that IFITM5 may act at the level of membrane fusion to prevent osteoclast formation.

Figure 24. Effect of ifitm5 Overexpression on the Expression of Tartrate- Resistant Acid Phosphatase. Pre-osteoclasts stably expressing the indicated IFITM or a vector control were plated on coverslips and stimulated with 50 ng/mL sRANKL. After 5 days, cells were fixed and stained for TRAP. Blue arrows mark large, multinucleated osteoclasts.

Amphotericin B Inhibits IFITM5.

We hypothesized that IFITM5 shared the fusion-inhibiting activities of its antiviral paralogs. Antimycotic polyene compounds, like amphotericin B, interfere with the

79 fusion-inhibiting activities of IFITM3 (215), likely by intercalating into membranes and inducing fusion-facilitating membrane curvatures (216, 217). We therefore hypothesized that amphotericin B would also inhibit IFITM5. RAW cells were thus stimulated with sRANKL in the presence or absence of amphotericin B, and the average number of nuclei per cell quantified after 4 days. While amphotericin B had no effect on osteoclast formation in the absence of IFITM5 expression, it completely rescued osteoclastogenesis in the presence of natural and MALEP-IFITM5 (Figure 25). Along with previous findings (215), these data indicated that amphotericin B is a broad inhibitor of all IFITM family members, and suggest that all IFITM proteins operate similarly to suppress membrane fusion.

Figure 25. Effect of Amphotericin B on IFITM5 Activity. (A) Pre-osteoclasts stably expressing IFITM5, MALEP-IFITM5 or a vector control were stimulated with 50 ng/mL sRANKL. Beginning 24 hours post sRANKL stimulation, cells were incubated for 4 days in the presence of 250 ng/mL amphotericin B or DMSO control. Cells were then fixed and labeled with anti-FLAG antibodies and wheat germ agglutinin to mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 marked the positions of cell nuclei. The percent of total nuclei in multinucleated cells was then quantified from 1000 cells, counted in at least 20 different fields for each experiment. Data represent means ± SD, (n=3). *, p<0.05. (B) Representative images of osteoclast formation in the presence or absence of amphotericin B. Anti-FLAG antibodies (red) and wheat germ agglutinin (blue) mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. Scale bars, 15 μm.

IFITM5 Restricts Membrane Fusion. Our results suggested that IFITM5 restricts the membrane fusion-catalyzing protein(s) expressed on the surface of pre-osteoclasts. To directly confirm that IFITM5 inhibits membrane fusion catalysis, we analyzed the effect of IFITM5 on a known viral fusion protein: MERS-S. To determine whether IFITM5 blocked MERS-S – mediated membrane fusion, we transfected DBT cells with DPP4 along with natural or mutant

IFITM5. The antiviral IFITM3 was utilized as a positive control. 2 days post- transfection, cells were inoculated with VSV-based MERSpps. 18 hours later, luciferase gene expressions were quantified in target cells as a readout of viral entry. Notably, natural and MALEP-IFITM5 inhibited MERS-S – mediated membrane fusion and viral entry, while IFITM5(S40L) did not (Figure 26A). IFITM5 also restricted cell-cell fusion mediated by the Murine Hepatitis Virus (MHV) spike glycoprotein, and this restriction was obliterated by amphotericin B (Figure 26B-C). Together, these results indicate that

IFITM5 blocks membrane fusions mediated by multiple fusion catalysts, similar to its paralogs.

Figure 26. Effect of ifitm5 Overexpression on Viral Membrane Fusion. (A) MERS- CoV – permissive DBT cells were generated by cotransfection of plasmids encoding human DPP4 along with the appropriate IFITM. 48 hours post – transfection, cells were inoculated with VSV-based MERS pps for 1 hour, rinsed with serum-free MEM, and incubated for an additional 18hrs. Cells were then lysed and luminescence was measured as a readout for viral entry. Data represent means ± SD, (n=4). *, p<0.05 when compared to “NO IFITM” control. (B) Pre-osteoclasts stably expressing the appropriate IFITM were infected with GFP – expressing MHV-A59. 2 hours post – infection, inoculum was removed and cells were incubated in the presence or absence of 250 ng/mL amphotericin B. After 16 hours, cells were then fixed and labeled with Hoeschst 33258 to mark cell nuclei. The percent of GFP+ nuclei in multinucleated cells was then quantified from 1000 cells. Data represent means ± SD, (n=3). *, p<0.05. (C)Representative images of cells from (B). GFP expression marks infected cells, while Hoescht 33258 (blue) marks the positions of cell nuclei. White arrows mark multinucleated cells.

CD9 Overexpression Rescues Osteoclast Formation in the Presence of IFITM5.

IFITMs control tetraspanin localization and their interactions with other proteins. IFITM5 disturbs interactions between CD9 and CD81 (136), tetraspanins critical for osteoclastogenesis. We hypothesized that IFITM5 may suppress osteoclast membrane fusion by disrupting the activities of these critical tetraspanins. We therefore overexpressed CD9 in RAW cells, along with natural or mutant IFITM5, and assessed the ability of these cells to fuse into osteoclasts. Notably, overexpressed CD9 accumulated in large puncta on the cell surface (Figure 27A), and rescued osteoclast formation in the presence of IFITM5 (Figure 27B-C). These data indicated that CD9 overexpression overcame IFITM5 inhibitory activities, and suggested that IFITM5 acts through CD9 to suppress membrane fusion.

Figure 27. Effect of CD9 Overexpression on IFITM5 Activity. (A) Pre-osteoclasts were transduced with CD9 or a vector control. 4 days later, cells were fixed and labeled with anti-CD9 (green) and Hoeschest 33258 (grey). Scale bars, 10 μm. (B) Pre- osteoclasts stably expressing the indicated IFITM or a vector control were transduced with a vector control or CD9. 2 days post – transduction, cells were stimulated with 50 ng/mL sRANKL for 4 days. Cells were then fixed and labeled with anti-FLAG, fluorescently labeled wheat germ agglutinin and Hoeschst 33258 to mark Flag-tagged IFITMs, cell plasma membranes and nuclei, respectively. The percent of total nuclei in multinucleated cells was then quantified from 1000 cells, counted in at least 20 different fields for each experiment. Data represent means ± SD, (n=3). *, p<0.05. (C) Representative images of cells from (B). Anti-FLAG antibodies (green) and wheat germ agglutinin (red) mark the positions of Flag-tagged IFITMs and cell plasma membranes, respectively. Hoescht 33258 (grey) mark the positions of cell nuclei. Scale bars, 15 μm.

CHAPTER FOUR

DISCUSSION

Summary of Data

We sought to characterize the roles of tetraspanins in constructing virus, EV, and cell membrane fusion platforms. Our results indicate that TEMs act as platforms for a wide range of membrane fusion processes, illustrating how membrane dynamics in diverse environments are controlled by a common set of elements. These elements include the tetraspanins themselves, which can facilitate and enable membrane fusions

(Figures 1, 9-10, 13-15), as well as IFITMs, which incorporate into TEMs to inhibit membrane mergers (Figures 17-19, and 23).

First, we sought to determine whether MERS-CoV utilizes TEMs as fusion platforms in cell culture and in vivo. We determined that CD9 was critical for MERS-

CoV entry in cell culture (Figure 7), and was the tetraspanin partner of the MERS-CoV receptor DPP4 (Figure 8) (148). CD9 acted to link DPP4 with the protease TMPRSS2, thereby facilitating MERS-S proteolysis and viral entry (Figure 9). We then developed an Ad5-based approach to render mouse lung cells permissive to infection while also transiently overexpressing or depleting candidate viral entry factors. Using this approach, we demonstrated for the first time that MERS-CoVs utilize TEMs as fusion platforms in vivo (Figure 10). Thus, MERS-CoV fusion platforms include at least three factors: the receptor DPP4, the tetraspanin CD9, and the protease TMPRSS2.

Next, we developed the first quantitative assay to monitor EV-cell membrane fusions and cargo deliveries (Figure 11). This assay relied on the transfer of reporter protein fragments between cell cytosols. Using this method, we were the first to determine that membrane fusion drives the delivery of luminal EV cargos into cell cytosols (Figure 12). We also determined that EV-incorporated tetraspanins enabled

EV cargo deliveries post-binding, likely at the level of membrane fusion (Figure 15).

IFITM3, in contrast, incorporated into tetraspanin-containing EVs to restrict EV fusogenicity (Figures 17-19). Together, these data indicate that TEMs act as platforms for EV membrane fusions.

We also found that EVs vary in their cargo delivery potentials, with delivery- competent EVs comprising only a small subset of vesicles released by the cell.

Transfer-competent EVs are enriched in tetraspanins (Figure 14), and are targeted by

IFITM3 (Figure 20). Deliveries by these transfer-competent EVs were only detectable upon EV producer cell modification, at least in the context of 293T cells and MSCs

(Figures 13, 16, and 21).

Finally, we found that TEMs also control monocyte membrane fusion, with the tetraspanin CD9 enhancing (Figure 27), and the bone-specific IFITM5 inhibiting (Figure

23), monocyte multinucleation and osteoclast formation. We are the first to determine that the constitutively expressed IFITM5 shares the fusion-inhibiting activities of the IFN- induced IFITMs, and that IFITM5 acts through CD9 to prevent membrane mergers

(Figure 27). CD9 overexpression rendered IFITM5 inert, as did treatment with amphotericin B (Figure 25), suggesting that IFITMs act through tetraspanins to alter membrane curvature and fluidity in ways that render them fusion-incompetent. Disease-

87 causing mutations to IFITM5 differentially effected its fusion-inhibiting activities, with one mutant, MALEP-IFITM5, retaining its fusion-inhibiting activities, while another mutant,

IFITM5(S40L) was inert (Figure 23). These results are the first to identify functional differences between mutant and natural IFITM5s, and they suggest that distinct mutations in IFITM5 cause disparate forms of OI.

Overall, these results indicate that a variety of membrane fusion reactions, across different environments, all rely on a common set of regulators. These regulators are concentrated into TEMs, likely by the tetraspanins themselves. Tetraspanins, therefore, may be central to constructing all extra-cellular platforms for membrane fusion catalysis.

Requirement of CD9 during MERS-CoV Early Entry

Our findings indicate that TEMs are platforms for MERS-CoV proteolysis and viral entry. CD9 is a central component in this fusion platform, as it links DPP4 with

TMPRSS2 (Figure 9). Close proximity between DPP4 and CoV activating proteases is likely required for early, TMPR-mediated viral entry. This is because only DPP4-bound

CoV S proteins are susceptible to productive, fusion-facilitating proteolytic cleavages

(24), and because cell surface proteases are likely concentrated TEMs (Figure 8) (28).

Additional findings by Dr. James Earnest have confirmed that CD9 is specifically used in early, but not late, CoV entry pathways (218). As discussed in Chapter 1, this could be because proteolytic triggering during late entry is catalyzed by soluble cathepsins (29), which may not require tetraspanins to encounter receptor-bound CoV-S proteins. It is therefore notable that CD9 is so critical for MERS-CoV infections in the mouse lung

(Figure 10). These results suggest that MERS-CoVs have evolved to preferentially utilize early entry routes in vivo.

It is unclear why MERS-CoVs have such a preference for early entry.

Possibilities include low levels of cathepsins in lung cells, or alternatively the preference to avoid endosome-localized antiviral factors, like the IFITMs (165). It also is unclear why CoVs maintain the ability to utilize late entry routes. Importantly, it is possible that these late entry pathways are simply byproducts of culturing CoVs in commonly-used cell lines cells. Some of these cell lines have very low levels of cell surface serine proteases (24), and CoVs preferentially use late entry routes to infect these cells.

Future investigations into CoV entry pathways would likely benefit from utilizing cells with high surface protease expression.

Selective Use of TMPRSS2 by MERS-CoVs to Infect the Lung

Our results using Ad5 vectors indicate that MERS-CoV relies heavily on a single protease, TMPRSS2, to enter mouse lung cells. While previous studies have highlighted the importance of serine proteases in CoV lung infections (33), cell culture studies suggest that CoVs can utilize a variety of proteases to trigger membrane fusion catalysis, including TMPRs (35), pro-protein convertases (219), and metalloproteases

(220). Many of these proteases also localize to TEMs. Presumably, DPP4-bound

MERS-CoVs would be able to engage several different proteases to facilitate entry.

This, however, is not the case in mouse lungs, as viral titers were significantly reduced by depleting just one protease (Figure 10). While these reasons for this specificity are unclear, it may be due to high levels of TMPRSS2 in lung epithelial cells, or due to close

89 interaction between CD9 and an unidentified tetraspanin that presumably aggregates

TMPRSS2 into TEMs (218).

Tetraspanins in the Entry of Other CoVs

In addition to MERS-CoV, we have determined that the CoV 229E also utilizes

CD9 during viral entry (218). Notably, however, not all CoVs utilize CD9 to facilitate membrane fusion, as SARS-CoV (Figure 7) and murine hepatitis virus (MHV) (218) were unaffected by CD9 depletion. One possible explanation is that SARS and MHV have evolved to rely on other tetraspanins for entry. Antibodies targeting the tetraspanins CD9, CD81, and CD63, which likely act to rigidify TEMs through tetraspanin crosslinking, inhibit the entry of all CoVs, supporting this hypothesis (28). In addition, many CoV receptors, including the SARS receptor ACE2 and MHV receptor

CEACAM, preferentially localize to TEMs (28, 218). It is intriguing to speculate why

CoVs have come to utilize TEMs as proteolytic priming platforms. Alternative proteolytic pathways do exist. For example, CoVs can be cleaved by extracellular proteases in the lung parenchyma (221). CoVs could theoretically evolve structures that allow for proteolytic “priming” before a receptor-dependent “triggering” event. It is likely, however, that proteolyzed CoV-S proteins would be unstable, and susceptible to inactivation. This may be the reason CoV-S proteins have evolved to restrict their proteolysis to receptor-bearing target cell membranes. Further investigation, however, is required to determine whether all CoVs depend on particular tetraspanins, and proteases, to enter cells.

Utilizing Recombinant Adenoviruses for Gene Knockdown

We sought to determine whether MERS-CoV utilizes TEMs as platform for viral infection in the lung. This was accomplished through the generation of recombinant adenoviruses that express both shRNAs silencing genes of interest as well as the

MERS-CoV receptor hDPP4. Using these dual-expressing Ad5 vectors, our collaborators were able to both render mouse lung cells permissive to MERS-CoV infection while also depleting candidate CoV entry factors from the same cells (Figure

10). This system overcame the time-consuming requirements of generating transgenic or knockout mice, making it an efficient was to rapidly identify viral entry factors in the mouse lung. Ad5 vectors also only acted on transduced cells, likely diminishing any off- target effects from gene knockdown or overexpression. This approach can be applied to any viral system where cells must be rendered permissive to infection, and we speculate that many will use this technology in the future to identify factors critical for viral replication and transmission.

TEMs as General Platforms for Viral Entry

Findings from our lab and others indicate that tetraspanins promote the entry of multiple viruses, including coronaviruses (Figures 7, 9-10) (28), human cytomegalovirus

(HCMV) (139), human papilloma virus (HPV) (141, 145), influenza virus (IAV) (28, 142), hepatitis C virus (HCV) (138, 146, 147), lujo virus (140), and several alphaviruses (143,

144). In many of these cases, tetraspanins promote viral entry by regulating the composition of TEMs. For example, IAV hemagglutinin is proteolyzed in TEMs, similar to CoVs (28). The tetraspanin CD81 is critical for viral entry (142), suggesting that this tetraspanin may bring IAV-activating proteases into close proximity with sialate

91 receptors. As discussed in Chapter 1, while CD81 is the primary receptor for HCV

(138), it also coalesces HCV coreceptors into a complex required for endocytosis (146,

147). CD151 operates in a similar manner to aggregate HPV coreceptors and drive endocytosis of receptor-bound HPVs (141, 145). We predict that many viruses have evolved to utilize TEM-resident entry factors due to their close proximity, and thus rely on tetraspanins to maintain these entry platforms.

Mechanisms of EV Cargo Delivery

While the phenotypic effects of EV-directed cargo transfers are readily documented (48, 51), the regulation and efficiency of this delivery process was largely unknown. Here we developed new approaches to investigate EV-mediated membrane fusions and cargo transmissions. These approaches rely on the transfer of split reporter proteins between cell and vesicle cytosols. Using these approaches, we discovered that TEMs regulate EV cargo deliveries and act as platforms for EV-cell fusions. EV cargo delivery potentials correlated with TEM composition. We found that the tetraspanins CD9, CD63, and CD81 all enabled cargo deliveries at some stage post- binding, likely at the level of membrane fusion (Figure 15). A bona-fide fusion catalyst, the viral fusion protein VSVG, also bestowed EVs with transfer-competence (Figure 12), indicating that membrane fusions can drive EV cargo transfers.

Tetraspanins may enable EV-cell fusions by generating positive membrane curvatures on both EV and cell membranes (120, 149), thereby bringing them into close proximity. Alternatively, these tetraspanins may concentrate unidentified EV fusion machinery in TEMs, thereby facilitating their interactions similar to how CD9 aggregates

MERS-CoV entry factors (Figure 9). It is notable that tetraspanins regulate the

92 localization of fusion – promoting lipids. CD81, for example, interacts directly with cholesterol (116), and also facilitates the externalization of the fusion – promoting lipid phosphatidylserine (222). In addition, individual tetraspanins interact with distinct integrins that may also bring membranes into close proximity to allow for membrane merger (130). Finally, it is possible that tetraspanins interact with true membrane fusion catalysts. Notably, some endogenous retroviral fusion proteins incorporate into EVs

(223), though whether these fusogens interact with tetraspanins is unclear. Comparing delivery-competent and –incompetent EV subsets by lipidomic and proteomic analyses may provide additional insights into the architecture of EV-cell fusion platforms.

Inhibition of EV Cargo Delivery

In contrast to the delivery-facilitating tetraspanins, IFN stimulated the ISG IFITM3 to incorporate into EVs and impede cargo deliveries (Figure 17). Our in vitro EV-virus and EV-EV fusion assays indicate that IFITM3 functions on EVs to inhibit membrane fusion and cargo delivery (Figure 19). These results, along with previous findings that

IFITM3 blocks virus-cell membrane fusions (162, 164), strongly suggest that EVs deliver hydrophilic macromolecules into the cytoplasm via membrane fusion.

Our results indicate that inhibition of cargo transfer was more potent when

IFITM3 was present in EVs, rather than in target cells (Figure 18). This could be due to high concentrations of IFITM3 on tetraspanin-containing EVs, which might be mediated by IFITM-tetraspanin interactions (136, 172). Indeed, while IFITM3 was present on only

~20% of 293T EVs, it incorporated into ~45% of tetraspanin-containing vesicles (Figure

20), supporting this hypothesis. Interactions with tetraspanins may lead to the formation of multimeric IFITM3 networks, which could be essential for membrane fusion inhibition

(158). Alternatively, EVs may not rely on TEMs on target cells to catalyze membrane fusion, thereby avoiding membrane platforms with high IFITM3 concentrations.

Tetraspanin overexpression in target cells did not permit EV cargo transfers (Figure 13), consistant with this hypothesis. Intriguingly, several enveloped viruses can also evade cellular IFITMs (163), and further investigation is required to determine whether IFITM sensitivity correlates with TEM utilization during virus or EV entry.

Potential Implications of IFITM-Mediated Inhibition of EV Fusion

IFNs and IFITMs are traditionally thought of as antiviral agents, and we speculate that these antiviral activities may extend to EV-incorporated IFITMs. EVs can transfer viral proteins and nucleic acids between cells (59, 224), and incorporated IFITMs would suppress the dissemination of these viral factors (Figurs 17-19). Similarly, EVs can also transmit pathogenic protein aggregates like prions and α-synuclein (62, 63), as well as oncogenic proteins and nucleic acids between cells (48). We speculate IFNs and

IFITMs prevent the dissemination of many harmful cargos.

Intriguingly, EVs can also ferry pro-inflammatory molecules, the very factors that induce IFNs and IFITMs, between cells. For example, stromal cell EVs can transfer pro- inflammatory RNAs to breast cancer cells, leading to RIG-I activation and the production of both IFNs and more pro-inflammatory RNAs (225). EVs can also transmit lipopolysaccharide (LPS)-induced microRNA-155, which sensitizes cells to EV target cells to subsequent LPS-induced signaling (51). These positive-feedback loops of pro- inflammatory responses may be tempered by IFITMs, which are induced by both IFN and LPS (226). This would prevent runaway inflammation and promote the return to homeostatic conditions. Notably, even in the absence of inflammation or disease, EVs

94 and IFITMs control lymphocyte activation (50, 227), skeletal homeostasis (53, 179), and tumor growth and metastasis (48, 176). It is possible that IFITMs and EVs act in concert to control these biological processes by precisely tuning intercellular communications.

While we have focused on EV luminal cargos, whose delivery requires EV-cell fusion, it is notable that fusion-incompetent vesicles could also play critical roles in intercellular communication. These roles include antigen cross-presentation (50), protein turnover (44), and signal transduction mediated by receptor binding. IFITM3 and other fusion inhibitors could be utilized to distinguish between fusion -dependent and –independent EV activities.

Limitations to EV Cargo Transfer

There were profound limitations to EV-directed cargo transfers. Indeed, transfers were undetectable unless producer cells were stimulated to hyper-secrete EVs, some of which were transfer-competent (Figures 13, 16, and 21). Still, only a small fraction

(1/10,000th) of secreted DSPs were successfully transferred and registered in target cells (Figure 15). These inefficiencies in EV-directed transfers may be due to low EV adherence to target cells, since only 0.3% EVs bound to 293T cells after 2 hours at

13°C (Figure 15). Additional transfer inefficiencies may also derive from the small fraction of EVs that are indeed transfer competent. Under conditions where IFITM3 blocked ~85% of cargo transfers (Figure 18), only ~20% of EVs contained IFITM3

(Figure 20). Given that IFITM3 likely acts locally to block membrane mergers (165)

(Figure 19), these results suggest that only a small subset of EVs released from 293T cells is transfer competent. These results extend the concept of EV heterogeneity to

95 the level of EV fusogenicity and cargo delivery potential. Further investigations in additional cell types and conditions are required to determine whether EVs also have low transfer potential in vivo, where long-distance EV-directed transmissions have been documented (78).

Membrane Fusion Inhibition as a Shared Function of IFITMs

Skeletal maintenance requires both bone-depositing mesenchymal osteoblasts and bone-resorbing hematopoietic osteoclasts (228). Dysregulation of osteoclast activity can lead to complications including osteopenia and osteoporosis (85).

Developing new therapies for these diseases will significantly benefit from elucidating the molecular mechanisms of monocyte membrane fusion. Our results indicate that monocytic pre-osteoclast membrane fusions occur in TEMs. The tetraspanin CD9 promoted pre-osteoclast multinucleation (Figure 27), while IFITM5 inhibited this process

(Figure 23). IFITM5 did not prevent osteoclast differentiation (Figure 24), and blocked virus-catalyzed membrane fusions (Figure 26), strongly suggesting that IFITM5 acts at the level of membrane fusion to suppress osteoclastogenesis. Membrane fusion inhibition is thus a shared activity across both interferon-induced and constitutively expressed IFITMs. Notably, the IFITM protein family is ancient, and IFITM5 is homologous with evolutionarily distant IFITM orthologs (183, 184). Fusion-inhibiting

IFITM-like proteins are also present in mycobateria (229), leading us to speculate that the primordial function of IFITMs was to control membrane mergers.

In addition to IFITM5, it is possible that other IFITMs also control monocyte membrane fusion. Indeed, monocytes can express IFITMs 1, 2, and 3 (230), possibly due to tonic IFN signaling. While it is unclear whether IFN-induced IFITMs control

96 osteoclastogenesis, intriguingly mice deficient in IFN-β signaling show symptoms of enhanced bone resorption and osteoclast activity (231). The role of IFN-induced

IFITMs in suppressing monocyte fusions may be even more relevant in the lung, where monocytes fuse into giant cell macrophages to defend against foreign materials and pathogens (90–92). We hypothesize that IFITMs generally act to temper monocyte membrane fusions across different tissues.

Functional Relationship between IFITMs and Tetraspanins

Our results suggest that IFITM5 acts through CD9 to block monocyte fusion, as

CD9 overexpression overcame IFITM5 inhibition (Figure 27). In addition, IFITM3 preferentially incorporated into tetraspanin-enriched EVs to suppress EV-cell fusion and cargo transfer (Figures 19 -20). It is possible that IFITMs suppress membrane fusion by disrupting critical interactions between tetraspanins and their protein partners. IFITM5 disrupts interactions between CD9 and CD81 (136), and IFITM1 disrupts tetraspanin interactions to inhibit virus-cell fusion (172), supporting this speculation. Tetraspanins induce positive membrane curvatures (120, 149), and it is possible that these curvatures require tetraspanin-tetraspanin interactions. By disrupting these interactions,

IFITMs may render membranes fusion-incompetent. Alternatively, IFITMs may rely on tetraspanins to localize to sites of membrane fusion. Overexpressed CD9 may not localize specifically to pre-osteoclast fusion platforms, and could thus dilute IFITM5 proteins away from pre-osteoclast fusion sites. Further investigation is required to determine whether disrupting protein-protein interactions in TEMs is a common mechanism by which IFITMs prevent membrane mergers, or whether tetraspanins are instead essential to concentrate IFITMs at sites of membrane fusion.

Amphotericin B and IFITM-Mediated Fusion Inhibition

In addition to CD9 overexpression, amphotericin B treatment also inhibited

IFITM5 (Figure 25). Amphotericin B and other polyene compounds integrate into the outer leaflet of lipid bilayers, increasing membrane fluidity and potentially causing outward bending of the plasma membrane away from the cytosol (216, 217, 232). By integrating into outer membrane leaflets, amphotericin B may induce positive membrane curvatures that compete against the negative-curving IFITMs (Figure 6). They also decrease the relative amount of inward-curving sterols in membranes (233, 234), which may contribute to their activities. It is notably that amphotericin B and related polyenes are used therapeutically as antifungals. Given their ability to inhibit IFITMs and thereby promote membrane fusion (215) (Figure 25), it is possible that polyenes could be applied to conditions of excess bone deposition, including patients harboring MALEP-

IFITM5 and those suffering from osteopenia.

Potential Heterogeneity in OI Type V Pathologies

Our results suggest that dysregulated osteoclastogenesis contributes to OI type

V (Figure 23). Notably, our results also suggest that distinct IFITM5 mutations may cause different forms of OI type V. While MALEP-IFITM5 blocked membrane fusion similarly to its natural counterpart, IFITM5(S40L) was inert. Patients with the S40L mutation have greater bone resorption and lower bone mineral densities than patients expressing MALEP-IFITM5 (189), suggesting that IFITM5(S40L) patients may have overabundant osteoclasts (180). Patients expressing MALEP-IFITM5, in contrast, often have excess bone deposits including hyperplastic calluses, hyperdense metaphyseal bands, and interosseous calcification (179). This correlates with MALEP-IFITM5 fusion

98 inhibition (Figure 23). MALEP-IFITM5, however, could dysregulate skeletal maintenance in other ways, since natural IFITM5 can suppress membrane fusions to a similar extent. Distinguishing between the diseases caused by IFITM5 mutants may lead to the development of distinct therapies for these forms of OI type V.

Identifying Additional Membrane Fusion Catalysts and Regulators

We are likely just beginning to identify regulators of EV-cell and cell-cell membrane fusion. Enveloped virus membrane fusions are controlled at the level of the target cell by cellular receptors, cations, proteases, temperature, pH, and membrane lipid composition (5). We have determined that tetraspanins, TMPRSS2, and endosome acidification control EV cargo transfers at the level of the EV producer cell

(Figures 13-16). However, none of these factors regulated EV-cell membrane fusions when applied to target cells. Thus target cell factors regulating EV-cell membrane fusion remain unclear. We speculate that virus, cell and EV membrane fusions share many common regulators in both producer and target cells. DSP-based approaches could potentially be used in high-throughput screens to identify EV, virus and cell fusion enhancers and suppressors. These screens could be applied to many therapeutically- relevant cell types, including MSCs and monocytic osteoclast precursors. By enhancing and suppressing EV and cell membrane fusion potentials in therapy models, we may uncover how membrane fusions control disease.

REFERENCE LIST

1. Tkach M, Théry C. 2016. Communication by Extracellular Vesicles: Where We Are and Where We Need to Go. Cell 164:1226–1232.

2. Wickner W, Schekman R. 2008. Membrane fusion. Nat Struct Mol Biol 15:658– 664.

3. Parolini I, Federici C, Raggi C, Lugini L, Palleschi S, De Milito A, Coscia C, Iessi E, Logozzi M, Molinari A, Colone M, Tatti M, Sargiacomo M, Fais S. 2009. Microenvironmental pH Is a Key Factor for Exosome Traffic in Tumor Cells. J Biol Chem 284:34211–34222.

4. Montecalvo A, Larregina AT, Shufesky WJ, Beer Stolz D, Sullivan MLG, Karlsson JM, Baty CJ, Gibson GA, Erdos G, Wang Z, Milosevic J, Tkacheva OA, Divito SJ, Jordan R, Lyons-Weiler J, Watkins SC, Morelli AE. 2012. Mechanism of transfer of functional microRNAs between mouse dendritic cells via exosomes. Blood 119:756–766.

5. Harrison SC. 2015. Viral membrane fusion. Virology.

6. Hernandez JM, Podbilewicz B. 2017. The hallmarks of cell-cell fusion. Development 144:4481–4495.

7. Ogle BM, Cascalho M, Platt JL. 2005. Biological implications of cell fusion. Nat Rev Mol Cell Biol 6:567.

8. Tsatsaronis JA, Franch-Arroyo S, Resch U, Charpentier E. 2018. Extracellular Vesicle RNA: A Universal Mediator of Microbial Communication? Trends Microbiol 26:401–410.

9. Chernomordik L V, Kozlov MM. 2008. Mechanics of membrane fusion. Nat Struct Mol Biol 15:675–683.

10. Efrat A, Chernomordik L V, Kozlov MM. 2007. Point-like protrusion as a prestalk intermediate in membrane fusion pathway. Biophys J2007/02/02. 92:L61–L63.

11. Yang L, Huang HW. 2002. Observation of a Membrane Fusion Intermediate Structure. Science (80- ) 297:1877 LP-1879.

12. Mi S, Lee X, Li X, Veldman GM, Finnerty H, Racie L, LaVallie E, Tang XY, Edouard P, Howes S, Keith Jr. JC, McCoy JM. 2000. Syncytin is a captive

retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789.

13. Fédry J, Liu Y, Péhau-Arnaudet G, Pei J, Li W, Tortorici MA, Traincard F, Meola A, Bricogne G, Grishin N V, Snell WJ, Rey FA, Krey T. 2017. The Ancient Gamete Fusogen HAP2 Is a Eukaryotic Class II Fusion Protein. Cell 168:904– 915.e10.

14. Ivanovic T, Choi JL, Whelan SP, van Oijen AM, Harrison SC. 2013. Influenza- virus membrane fusion by cooperative fold-back of stochastically induced hemagglutinin intermediates. Elife 2:e00333.

15. Plemper RK, Brindley MA, Iorio RM. 2011. Structural and Mechanistic Studies of Measles Virus Illuminate Paramyxovirus Entry. PLOS Pathog 7:e1002058.

16. Chakraborty H, Tarafdar PK, Lentz BR. 2013. A novel assay for detecting fusion pore formation: implications for the fusion mechanism. Biochemistry 52:8510– 8517.

17. Bullough PA, Hughson FM, Skehel JJ, Wiley DC. 1994. Structure of influenza haemagglutinin at the pH of membrane fusion. Nature 371:37–43.

18. Weiss SR, Navas-Martin S. 2005. Coronavirus pathogenesis and the emerging pathogen severe acute respiratory syndrome coronavirus. Microbiol Mol Biol Rev 69:635–664.

19. Assiri A, Al-Tawfiq JA, Al-Rabeeah AA, Al-Rabiah FA, Al-Hajjar S, Al-Barrak A, Flemban H, Al-Nassir WN, Balkhy HH, Al-Hakeem RF, Makhdoom HQ, Zumla AI, Memish ZA. 2013. Epidemiological, demographic, and clinical characteristics of 47 cases of Middle East respiratory syndrome coronavirus disease from Saudi Arabia: a descriptive study. Lancet Infect Dis 13:752–761.

20. Peiris JSM, Guan Y, Yuen KY. 2004. Severe acute respiratory syndrome. Nat Med 10:S88-97.

21. Bosch BJ, van der Zee R, de Haan CAM, Rottier PJM. 2003. The coronavirus spike protein is a class I virus fusion protein: structural and functional characterization of the fusion core complex. J Virol 77:8801–8811.

22. Raj VS, Mou H, Smits SL, Dekkers DHW, Muller MA, Dijkman R, Muth D, Demmers JAA, Zaki A, Fouchier RAM, Thiel V, Drosten C, Rottier PJM, Osterhaus ADME, Bosch BJ, Haagmans BL. 2013. Dipeptidyl peptidase 4 is a functional receptor for the emerging human coronavirus-EMC. Nature 495:251– 254.

23. Lu G, Hu Y, Wang Q, Qi J, Gao F, Li Y, Zhang Y, Zhang W, Yuan Y, Bao J, Zhang B, Shi Y, Yan J, Gao GF. 2013. Molecular basis of binding between novel human coronavirus MERS-CoV and its receptor CD26. Nature 500:227–231.

100

24. Park J-E, Li K, Barlan A, Fehr AR, Perlman S, McCray PB, Gallagher T. 2016. Proteolytic processing of Middle East respiratory syndrome coronavirus spikes expands virus tropism. Proc Natl Acad Sci 113:12262 LP-12267.

25. Ou X, Zheng W, Shan Y, Mu Z, Dominguez SR, Holmes K V, Qian Z. 2016. Identification of the Fusion Peptide-Containing Region in Betacoronavirus Spike Glycoproteins. J Virol 90:5586–5600.

26. Gao J, Lu G, Qi J, Li Y, Wu Y, Deng Y, Geng H, Li H, Wang Q, Xiao H, Tan W, Yan J, Gao GF. 2013. Structure of the fusion core and inhibition of fusion by a heptad repeat peptide derived from the S protein of Middle East respiratory syndrome coronavirus. J Virol 87:13134–13140.

27. Nash TC, Buchmeier MJ. 1997. Entry of mouse hepatitis virus into cells by endosomal and nonendosomal pathways. Virology 233:1–8.

28. Earnest JT, Hantak MP, Park J-E, Gallagher T. 2015. Coronavirus and influenza virus proteolytic priming takes place in tetraspanin-enriched membrane microdomains. J Virol 89.

29. Qiu Z, Hingley ST, Simmons G, Yu C, Das Sarma J, Bates P, Weiss SR. 2006. Endosomal Proteolysis by Cathepsins Is Necessary for Murine Coronavirus Mouse Hepatitis Virus Type 2 Spike-Mediated Entry. J Virol 80:5768 LP-5776.

30. Glowacka I, Bertram S, Muller MA, Allen P, Soilleux E, Pfefferle S, Steffen I, Tsegaye TS, He Y, Gnirss K, Niemeyer D, Schneider H, Drosten C, Pohlmann S. 2011. Evidence that TMPRSS2 activates the severe acute respiratory syndrome coronavirus spike protein for membrane fusion and reduces viral control by the humoral immune response. J Virol 85:4122–4134.

31. Kawase M, Shirato K, Matsuyama S, Taguchi F. 2009. Protease-mediated entry via the endosome of human coronavirus 229E. J Virol 83:712–721.

32. Kawase M, Shirato K, van der Hoek L, Taguchi F, Matsuyama S. 2012. Simultaneous treatment of human bronchial epithelial cells with serine and cysteine protease inhibitors prevents severe acute respiratory syndrome coronavirus entry. J Virol 86:6537–6545.

33. Zhou Y, Vedantham P, Lu K, Agudelo J, Carrion R, Nunneley JW, Barnard D, Pöhlmann S, McKerrow JH, Renslo AR, Simmons G. 2015. Protease inhibitors targeting coronavirus and filovirus entry. Antiviral Res 116:76–84.

34. Shulla A, Heald-Sargent T, Subramanya G, Zhao J, Perlman S, Gallagher T. 2011. A Transmembrane Serine Protease Is Linked to the Severe Acute Respiratory Syndrome Coronavirus Receptor and Activates Virus Entry. J Virol 85:873–882.

101

35. Shirato K, Kawase M, Matsuyama S. 2013. Middle East respiratory syndrome coronavirus infection mediated by the transmembrane serine protease TMPRSS2. J Virol 87:12552–12561.

36. Bohley P, Seglen PO. 1992. Proteases and proteolysis in the lysosome. Experientia 48:151–157.

37. Bertram S, Dijkman R, Habjan M, Heurich A, Gierer S, Glowacka I, Welsch K, Winkler M, Schneider H, Hofmann-Winkler H, Thiel V, Pohlmann S. 2013. TMPRSS2 activates the human coronavirus 229E for cathepsin-independent host cell entry and is expressed in viral target cells in the respiratory epithelium. J Virol 87:6150–6160.

38. Floyd DL, Ragains JR, Skehel JJ, Harrison SC, van Oijen AM. 2008. Single- particle kinetics of influenza virus membrane fusion. Proc Natl Acad Sci 105:15382 LP-15387.

39. Tellier E, Canault M, Rebsomen L, Bonardo B, Juhan-Vague I, Nalbone G, Peiretti F. 2006. The shedding activity of ADAM17 is sequestered in lipid rafts. Exp Cell Res 312:3969–3980.

40. Lucas JM, True L, Hawley S, Matsumura M, Morrissey C, Vessella R, Nelson PS. 2008. The androgen-regulated type II serine protease TMPRSS2 is differentially expressed and mislocalized in prostate adenocarcinoma. J Pathol 215:118–125.

41. Ratajczak J, Miekus K, Kucia M, Zhang J, Reca R, Dvorak P, Ratajczak MZ. 2006. Embryonic stem cell-derived microvesicles reprogram hematopoietic progenitors: evidence for horizontal transfer of mRNA and protein delivery. Leukemia 20:847–856.

42. Valadi H, Ekstrom K, Bossios A, Sjostrand M, Lee JJ, Lotvall JO. 2007. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 9:654–659.

43. Trams EG, Lauter CJ, Salem NJ, Heine U. 1981. Exfoliation of membrane ecto- enzymes in the form of micro-vesicles. Biochim Biophys Acta 645:63–70.

44. Pan BT, Teng K, Wu C, Adam M, Johnstone RM. 1985. Electron microscopic evidence for externalization of the transferrin receptor in vesicular form in sheep reticulocytes. J Cell Biol 101:942–948.

45. Li M, Zeringer E, Barta T, Schageman J, Cheng A, Vlassov A V. 2014. Analysis of the RNA content of the exosomes derived from blood serum and urine and its potential as biomarkers. Philos Trans R Soc B Biol Sci 369:20130502.

46. Harding C, Heuser J, Stahl P. 1983. Receptor-mediated endocytosis of transferrin and recycling of the transferrin receptor in rat reticulocytes. J Cell Biol 97:329–339.

102

47. Pan BT, Johnstone RM. 1983. Fate of the transferrin receptor during maturation of sheep reticulocytes in vitro: selective externalization of the receptor. Cell 33:967–978.

48. Melo SA, Sugimoto H, O’Connell JT, Kato N, Villanueva A, Vidal A, Qiu L, Vitkin E, Perelman LT, Melo CA, Lucci A, Ivan C, Calin GA, Kalluri R. 2014. Cancer exosomes perform cell-independent microRNA biogenesis and promote tumorigenesis. Cancer Cell 26:707–721.

49. Thery C, Duban L, Segura E, Veron P, Lantz O, Amigorena S. 2002. Indirect activation of naive CD4+ T cells by dendritic cell-derived exosomes. Nat Immunol 3:1156–1162.

50. Tkach M, Kowal J, Zucchetti AE, Enserink L, Jouve M, Lankar D, Saitakis M, Martin‐Jaular L, Théry C. 2017. Qualitative differences in T‐cell activation by dendritic cell‐derived extracellular vesicle subtypes. EMBO J 36:3012 LP-3028.

51. Alexander M, Hu R, Runtsch MC, Kagele DA, Mosbruger TL, Tolmachova T, Seabra MC, Round JL, Ward DM, O’Connell RM. 2015. Exosome-delivered microRNAs modulate the inflammatory response to endotoxin. Nat Commun 6:7321.

52. Ramakrishnaiah V, Thumann C, Fofana I, Habersetzer F, Pan Q, de Ruiter PE, Willemsen R, Demmers JAA, Stalin Raj V, Jenster G, Kwekkeboom J, Tilanus HW, Haagmans BL, Baumert TF, van der Laan LJW. 2013. Exosome-mediated transmission of hepatitis C virus between human hepatoma Huh7.5 cells. Proc Natl Acad Sci 110:13109–13113.

53. Zhang S, Chu WC, Lai RC, Lim SK, Hui JHP, Toh WS. 2016. Exosomes derived from human embryonic mesenchymal stem cells promote osteochondral regeneration. Osteoarthr Cartil 24:2135–2140.

54. McGough IJ, Vincent JP. 2016. Exosomes in developmental signalling. Development 143:2482–2493.

55. Wang Z, von Au A, Schnolzer M, Hackert T, Zoller M. 2016. CD44v6-competent tumor exosomes promote motility, invasion and cancer-initiating cell marker expression in pancreatic and colorectal cancer cells. Oncotarget 7:55409–55436.

56. Genschmer KR, Russell DW, Lal C, Szul T, Bratcher PE, Noerager BD, Abdul Roda M, Xu X, Rezonzew G, Viera L, Dobosh BS, Margaroli C, Abdalla TH, King RW, McNicholas CM, Wells JM, Dransfield MT, Tirouvanziam R, Gaggar A, Blalock JE. 2019. Activated PMN Exosomes: Pathogenic Entities Causing Matrix Destruction and Disease in the Lung. Cell 176:113–126.e15.

57. Yim N, Ryu SW, Choi K, Lee KR, Lee S, Choi H, Kim J, Shaker MR, Sun W, Park JH, Kim D, Do Heo W, Choi C. 2016. Exosome engineering for efficient

103

intracellular delivery of soluble proteins using optically reversible protein-protein interaction module. Nat Commun 7:12277.

58. Luther KM, Haar L, McGuinness M, Wang Y, Lynch Iv TL, Phan A, Song Y, Shen Z, Gardner G, Kuffel G, Ren X, Zilliox MJ, Jones WK. 2018. Exosomal miR-21a- 5p mediates cardioprotection by mesenchymal stem cells. J Mol Cell Cardiol 119:125–137.

59. Bukong TN, Momen-Heravi F, Kodys K, Bala S, Szabo G. 2014. Exosomes from Hepatitis C Infected Patients Transmit HCV Infection and Contain Replication Competent Viral RNA in Complex with Ago2-miR122-HSP90. PLoS Pathog 10:e1004424.

60. McKnight KL, Xie L, González-López O, Rivera-Serrano EE, Chen X, Lemon SM. 2017. Protein composition of the hepatitis A virus quasi-envelope. Proc Natl Acad Sci .

61. Robinson SM, Tsueng G, Sin J, Mangale V, Rahawi S, McIntyre LL, Williams W, Kha N, Cruz C, Hancock BM, Nguyen DP, Sayen MR, Hilton BJ, Doran KS, Segall AM, Wolkowicz R, Cornell CT, Whitton JL, Gottlieb RA, Feuer R. 2014. Coxsackievirus B Exits the Host Cell in Shed Microvesicles Displaying Autophagosomal Markers. PLOS Pathog 10:e1004045.

62. Danzer KM, Kranich LR, Ruf WP, Cagsal-Getkin O, Winslow AR, Zhu L, Vanderburg CR, McLean PJ. 2012. Exosomal cell-to-cell transmission of alpha synuclein oligomers. Mol Neurodegener 7:42.

63. Fevrier B, Vilette D, Archer F, Loew D, Faigle W, Vidal M, Laude H, Raposo G. 2004. Cells release prions in association with exosomes. Proc Natl Acad Sci U S A 101:9683–9688.

64. Kowal J, Arras G, Colombo M, Jouve M, Morath JP, Primdal-Bengtson B, Dingli F, Loew D, Tkach M, Théry C. 2016. Proteomic comparison defines novel markers to characterize heterogeneous populations of extracellular vesicle subtypes. Proc Natl Acad Sci 113:E968–E977.

65. Baietti MF, Zhang Z, Mortier E, Melchior A, Degeest G, Geeraerts A, Ivarsson Y, Depoortere F, Coomans C, Vermeiren E, Zimmermann P, David G. 2012. Syndecan-syntenin-ALIX regulates the biogenesis of exosomes. Nat Cell Biol 14:677–685.

66. Wollert T, Hurley JH. 2010. Molecular mechanism of multivesicular body biogenesis by ESCRT complexes. Nature 464:864–869.

67. Perez-Hernandez D, Gutiérrez-Vázquez C, Jorge I, López-Martín S, Ursa A, Sánchez-Madrid F, Vázquez J, Yáñez-Mó M. 2013. The Intracellular Interactome of Tetraspanin-enriched Microdomains Reveals Their Function as Sorting Machineries toward Exosomes . J Biol Chem 288:11649–11661.

104

68. Chairoungdua A, Smith DL, Pochard P, Hull M, Caplan MJ. 2010. Exosome release of β-catenin: a novel mechanism that antagonizes Wnt signaling. J Cell Biol 190:1079–1091.

69. Trajkovic K, Hsu C, Chiantia S, Rajendran L, Wenzel D, Wieland F, Schwille P, Brügger B, Simons M. 2008. Ceramide Triggers Budding of Exosome Vesicles into Multivesicular Endosomes. Science (80- ) 319:1244 LP-1247.

70. Kanada M, Bachmann MH, Hardy JW, Frimannson DO, Bronsart L, Wang A, Sylvester MD, Schmidt TL, Kaspar RL, Butte MJ, Matin AC, Contag CH. 2015. Differential fates of biomolecules delivered to target cells via extracellular vesicles. Proc Natl Acad Sci 112:E1433–E1442.

71. Sedlik C, Vigneron J, Torrieri-Dramard L, Pitoiset F, Denizeau J, Chesneau C, de la Rochere P, Lantz O, Thery C, Bellier B. 2014. Different immunogenicity but similar antitumor efficacy of two DNA vaccines coding for an antigen secreted in different membrane vesicle-associated forms. J Extracell Vesicles 3:24646.

72. Minakaki G, Menges S, Kittel A, Emmanouilidou E, Schaeffner I, Barkovits K, Bergmann A, Rockenstein E, Adame A, Marxreiter F, Mollenhauer B, Galasko D, Buzas EI, Schlotzer-Schrehardt U, Marcus K, Xiang W, Lie DC, Vekrellis K, Masliah E, Winkler J, Klucken J. 2017. Autophagy inhibition promotes SNCA/alpha-synuclein release and transfer via extracellular vesicles with a hybrid autophagosome-exosome-like phenotype. Autophagy 1–61.

73. Stein JM, Luzio JP. 1991. Ectocytosis caused by sublytic autologous complement attack on human neutrophils. The sorting of endogenous plasma- membrane proteins and lipids into shed vesicles. Biochem J 274:381–386.

74. Johnstone RM, Adam M, Hammond JR, Orr L, Turbide C. 1987. Vesicle formation during reticulocyte maturation. Association of plasma membrane activities with released vesicles (exosomes). J Biol Chem 262:9412–9420.

75. Hoshino A, Costa-Silva B, Shen T-L, Rodrigues G, Hashimoto A, Tesic Mark M, Molina H, Kohsaka S, Di Giannatale A, Ceder S, Singh S, Williams C, Soplop N, Uryu K, Pharmer L, King T, Bojmar L, Davies AE, Ararso Y, Zhang T, Zhang H, Hernandez J, Weiss JM, Dumont-Cole VD, Kramer K, Wexler LH, Narendran A, Schwartz GK, Healey JH, Sandstrom P, Jørgen Labori K, Kure EH, Grandgenett PM, Hollingsworth MA, de Sousa M, Kaur S, Jain M, Mallya K, Batra SK, Jarnagin WR, Brady MS, Fodstad O, Muller V, Pantel K, Minn AJ, Bissell MJ, Garcia BA, Kang Y, Rajasekhar VK, Ghajar CM, Matei I, Peinado H, Bromberg J, Lyden D. 2015. Tumour exosome integrins determine organotropic metastasis. Nature 527:329–335.

76. Clayton A, Turkes A, Dewitt S, Steadman R, Mason MD, Hallett MB. 2004. Adhesion and signaling by B cell-derived exosomes: the role of integrins. FASEB J .

105

77. Mulcahy LA, Pink RC, Carter DRF. 2014. Routes and mechanisms of extracellular vesicle uptake. J Extracell Vesicles 3:10.3402/jev.v3.24641.

78. Ridder K, Keller S, Dams M, Rupp A-K, Schlaudraff J, Del Turco D, Starmann J, Macas J, Karpova D, Devraj K, Depboylu C, Landfried B, Arnold B, Plate KH, Höglinger G, Sültmann H, Altevogt P, Momma S. 2014. Extracellular Vesicle- Mediated Transfer of Genetic Information between the Hematopoietic System and the Brain in Response to Inflammation. PLOS Biol 12:e1001874.

79. Nguyen TA, Smith BRC, Tate MD, Belz GT, Barrios MH, Elgass KD, Weisman AS, Baker PJ, Preston SP, Whitehead L, Garnham A, Lundie RJ, Smyth GK, Pellegrini M, O’Keeffe M, Wicks IP, Masters SL, Hunter CP, Pang KC. 2017. SIDT2 Transports Extracellular dsRNA into the Cytoplasm for Innate Immune Recognition. Immunity 47:498–509.e6.

80. Okabe M. 2013. The cell biology of mammalian fertilization. Development 140:4471–4479.

81. Gerbaud P, Pidoux G. 2015. Review: An overview of molecular events occurring in human trophoblast fusion. Placenta 36 Suppl 1:S35-42.

82. Kim JH, Jin P, Duan R, Chen EH. 2015. Mechanisms of myoblast fusion during muscle development. Curr Opin Genet Dev 32:162–170.

83. Vignery A. 2005. Macrophage fusion: the making of osteoclasts and giant cells. J Exp Med 202:337–340.

84. McNally AK, Anderson JM. 2011. Macrophage fusion and multinucleated giant cells of inflammation. Adv Exp Med Biol 713:97–111.

85. Shevde NK, Bendixen AC, Dienger KM, Pike JW. 2000. Estrogens suppress RANK ligand-induced osteoclast differentiation via a stromal cell independent mechanism involving c-Jun repression. Proc Natl Acad Sci U S A 97:7829–7834.

86. Roodman GD, Windle JJ. 2005. Paget disease of bone. J Clin Invest 115:200– 208.

87. Demonbreun AR, Biersmith BH, McNally EM. 2015. Membrane fusion in muscle development and repair. Semin Cell Dev Biol2015/10/30. 45:48–56.

88. Gauster M, Moser G, Orendi K, Huppertz B. 2009. Factors involved in regulating trophoblast fusion: potential role in the development of preeclampsia. Placenta 30 Suppl A:S49-54.

89. Bastida-Ruiz D, Van Hoesen K, Cohen M. 2016. The Dark Side of Cell Fusion. Int J Mol Sci 17:638.

106

90. Davison AG, Haslam PL, Corrin B, Coutts II, Dewar A, Riding WD, Studdy PR, Newman-Taylor AJ. 1983. Interstitial lung disease and asthma in hard-metal workers: bronchoalveolar lavage, ultrastructural, and analytical findings and results of bronchial provocation tests. Thorax 38:119–128.

91. Byrd TF. 1998. Multinucleated giant cell formation induced by IFN-gamma/IL-3 is associated with restriction of virulent Mycobacterium tuberculosis cell to cell invasion in human monocyte monolayers. Cell Immunol 188:89–96.

92. Lay G, Poquet Y, Salek-Peyron P, Puissegur MP, Botanch C, Bon H, Levillain F, Duteyrat JL, Emile JF, Altare F. 2007. Langhans giant cells from M. tuberculosis- induced human granulomas cannot mediate mycobacterial uptake. J Pathol 211:76–85.

93. Milde R, Ritter J, Tennent GA, Loesch A, Martinez FO, Gordon S, Pepys MB, Verschoor A, Helming L. 2015. Multinucleated Giant Cells Are Specialized for Complement-Mediated Phagocytosis and Large Target Destruction. Cell Rep 13:1937–1948.

94. Wright LM, Maloney W, Yu X, Kindle L, Collin-Osdoby P, Osdoby P. 2005. Stromal cell-derived factor-1 binding to its chemokine receptor CXCR4 on precursor cells promotes the chemotactic recruitment, development and survival of human osteoclasts. Bone 36:840–853.

95. Ishii M, Egen JG, Klauschen F, Meier-Schellersheim M, Saeki Y, Vacher J, Proia RL, Germain RN. 2009. Sphingosine-1-phosphate mobilizes osteoclast precursors and regulates bone homeostasis. Nature 458:524–528.

96. Matsuzaki K, Udagawa N, Takahashi N, Yamaguchi K, Yasuda H, Shima N, Morinaga T, Toyama Y, Yabe Y, Higashio K, Suda T. 1998. Osteoclast differentiation factor (ODF) induces osteoclast-like cell formation in human peripheral blood mononuclear cell cultures. Biochem Biophys Res Commun 246:199–204.

97. Ishii M, Iwai K, Koike M, Ohshima S, Kudo-Tanaka E, Ishii T, Mima T, Katada Y, Miyatake K, Uchiyama Y, Saeki Y. 2006. RANKL-Induced Expression of Tetraspanin CD9 in Lipid Raft Membrane Microdomain Is Essential for Cell Fusion During Osteoclastogenesis. J Bone Miner Res 21:965–976.

98. McNally AK, Anderson JM. 2002. Beta1 and beta2 integrins mediate adhesion during macrophage fusion and multinucleated foreign body giant cell formation. Am J Pathol 160:621–630.

99. Saginario C, Sterling H, Beckers C, Kobayashi R, Solimena M, Ullu E, Vignery A. 1998. MFR, a Putative Receptor Mediating the Fusion of Macrophages. Mol Cell Biol 18:6213–6223.

107

100. Han X, Sterling H, Chen Y, Saginario C, Brown EJ, Frazier WA, Lindberg FP, Vignery A. 2000. CD47, a Ligand for the Macrophage Fusion Receptor, Participates in Macrophage Multinucleation. J Biol Chem 275:37984–37992.

101. Kukita T, Wada N, Kukita A, Kakimoto T, Sandra F, Toh K, Nagata K, Iijima T, Horiuchi M, Matsusaki H, Hieshima K, Yoshie O, Nomiyama H. 2004. RANKL- induced DC-STAMP is essential for osteoclastogenesis. J Exp Med 200:941– 946.

102. Helming L, Gordon S. 2009. Molecular mediators of macrophage fusion. Trends Cell Biol 19:514–522.

103. Oursler MJ. 2010. Recent advances in understanding the mechanisms of osteoclast precursor fusion. J Cell Biochem 110:1058–1062.

104. Blond J-L, Lavillette D, Cheynet V, Bouton O, Oriol G, Chapel-Fernandes S, Mandrand B, Mallet F, Cosset F-L. 2000. An Envelope Glycoprotein of the Human Endogenous Retrovirus HERV-W Is Expressed in the Human Placenta and Fuses Cells Expressing the Type D Mammalian Retrovirus Receptor. J Virol 74:3321–3329.

105. Blaise S, de Parseval N, BÃ©nit L, Heidmann T. 2003. Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci U S A 100:13013– 13018.

106. Lavialle C, Cornelis G, Dupressoir A, Esnault C, Heidmann O, Vernochet C, Heidmann T. Paleovirology of “syncytins”, retroviral env genes exapted for a role in placentation. Philos Trans R Soc Lond B Biol Sci 368:20120507.

107. Dupressoir A, Vernochet C, Harper F, Guegan J, Dessen P, Pierron G, Heidmann T. 2011. A pair of co-opted retroviral envelope syncytin genes is required for formation of the two-layered murine placental syncytiotrophoblast. Proc Natl Acad Sci U S A 108:E1164-73.

108. Søe K, Andersen TL, Hobolt-Pedersen A-S, Bjerregaard B, Larsson L-I, Delaissé J-M. 2011. Involvement of human endogenous retroviral syncytin-1 in human osteoclast fusion. Bone 48:837–846.

109. Redelsperger F, Raddi N, Bacquin A, Vernochet C, Mariot V, Gache V, Blanchard-Gutton N, Charrin S, Tiret L, Dumonceaux J, Dupressoir A, Heidmann T. 2016. Genetic Evidence That Captured Retroviral Envelope syncytins Contribute to Myoblast Fusion and Muscle Sexual Dimorphism in Mice. PLoS Genet 12:e1006289.

110. Johnson MA, von Besser K, Zhou Q, Smith E, Aux G, Patton D, Levin JZ, Preuss D. 2004. Arabidopsis hapless mutations define essential gametophytic functions. Genetics 168:971–982.

108

111. Liu Y, Tewari R, Ning J, Blagborough AM, Garbom S, Pei J, Grishin N V, Steele RE, Sinden RE, Snell WJ, Billker O. 2008. The conserved plant sterility gene HAP2 functions after attachment of fusogenic membranes in Chlamydomonas and Plasmodium gametes. Genes Dev 22:1051–1068.

112. Cole ES, Cassidy-Hanley D, Fricke Pinello J, Zeng H, Hsueh M, Kolbin D, Ozzello C, Giddings TJ, Winey M, Clark TG. 2014. Function of the male-gamete- specific fusion protein HAP2 in a seven-sexed ciliate. Curr Biol 24:2168–2173.

113. Fedry J, Forcina J, Legrand P, Péhau-Arnaudet G, Haouz A, Johnson M, Rey FA, Krey T. 2018. Evolutionary diversification of the HAP2 membrane insertion motifs to drive gamete fusion across eukaryotes. PLOS Biol 16:e2006357.

114. Zaitseva E, Zaitsev E, Melikov K, Arakelyan A, Marin M, Villasmil R, Margolis LB, Melikyan GB, Chernomordik L V. 2017. Fusion Stage of HIV-1 Entry Depends on Virus-Induced Cell Surface Exposure of Phosphatidylserine. Cell Host Microbe 22:99–110.e7.

115. Thorp EB, Gallagher TM. 2004. Requirements for CEACAMs and cholesterol during murine coronavirus cell entry. J Virol 78:2682–2692.

116. Zimmerman B, Kelly B, McMillan BJ, Seegar TCM, Dror RO, Kruse AC, Blacklow SC. 2016. Crystal Structure of a Full-Length Human Tetraspanin Reveals a Cholesterol-Binding Pocket. Cell 167:1041–1051.e11.

117. Chen MS, Tung KS, Coonrod SA, Takahashi Y, Bigler D, Chang A, Yamashita Y, Kincade PW, Herr JC, White JM. 1999. Role of the integrin-associated protein CD9 in binding between sperm ADAM 2 and the egg integrin alpha6beta1: implications for murine fertilization. Proc Natl Acad Sci U S A 96:11830–11835.

118. Shilagardi K, Li S, Luo F, Marikar F, Duan R, Jin P, Kim JH, Murnen K, Chen EH. 2013. Actin-propelled invasive membrane protrusions promote fusogenic protein engagement during cell-cell fusion. Science 340:359–363.

119. Huang S, Yuan S, Dong M, Su J, Yu C, Shen Y, Xie X, Yu Y, Yu X, Chen S, Zhang S, Pontarotti P, Xu A. 2005. The phylogenetic analysis of tetraspanins projects the evolution of cell–cell interactions from unicellular to multicellular organisms. Genomics 86:674–684.

120. Bari R, Guo Q, Xia B, Zhang YH, Giesert EE, Levy S, Zheng JJ, Zhang XA. 2011. Tetraspanins regulate the protrusive activities of cell membrane. Biochem Biophys Res Commun2011/11/02. 415:619–626.

121. Yanez-Mo M, Siljander PRM, Andreu Z, Zavec AB, BorrÃ s FE, Buzas EI, Buzas K, Casal E, Cappello F, Carvalho J, ColÃ¡s E, Silva AC, Fais S, Falcon-Perez JM, Ghobrial IM, Giebel B, Gimona M, Graner M, Gursel I, Gursel M, Heegaard NHH, Hendrix A, Kierulf P, Kokubun K, Kosanovic M, Kralj-Iglic V, KrÃ¤mer- Albers E-M, Laitinen S, LÃ¤sser C, Lener T, Ligeti E, LinÄ“ A, Lipps G, Llorente

109

A, LÃ¶tvall J, ManÄ•ek-Keber M, Marcilla A, Mittelbrunn M, Nazarenko I, Hoen ENMN-â€˜t, Nyman TA, O’Driscoll L, Olivan M, Oliveira C, PÃ¡llinger Ã, del Portillo HA, ReventÃ3s J, Rigau M, Rohde E, Sammar M, SÃ¡nchez-Madrid F, SantarÃ©m N, Schallmoser K, Ostenfeld MS, Stoorvogel W, Stukelj R, Van der Grein SG, Vasconcelos MH, Wauben MHM, De Wever O. 2015. Biological properties of extracellular vesicles and their physiological functions. J Extracell Vesicles 4:10.3402/jev.v4.27066.

122. Boker KO, Lemus-Diaz N, Rinaldi Ferreira R, Schiller L, Schneider S, Gruber J. 2018. The Impact of the CD9 Tetraspanin on Lentivirus Infectivity and Exosome Secretion. Mol Ther 26:634–647.

123. Hurwitz SN, Cheerathodi MR, Nkosi D, York SB, Meckes DGJ. 2018. Tetraspanin CD63 Bridges Autophagic and Endosomal Processes To Regulate Exosomal Secretion and Intracellular Signaling of Epstein-Barr Virus LMP1. J Virol 92.

124. Yang X, Claas C, Kraeft S-K, Chen LB, Wang Z, Kreidberg JA, Hemler ME. 2002. Palmitoylation of tetraspanin proteins: modulation of CD151 lateral interactions, subcellular distribution, and integrin-dependent cell morphology. Mol Biol Cell 13:767–781.

125. Charrin S, Manie S, Oualid M, Billard M, Boucheix C, Rubinstein E. 2002. Differential stability of tetraspanin/tetraspanin interactions: role of palmitoylation. FEBS Lett 516:139–144.

126. Berditchevski F, Gilbert E, Griffiths MR, Fitter S, Ashman L, Jenner SJ. 2001. Analysis of the CD151-alpha3beta1 integrin and CD151-tetraspanin interactions by mutagenesis. J Biol Chem 276:41165–41174.

127. Charrin S, le Naour F, Silvie O, Milhiet P-E, Boucheix C, Rubinstein E. 2009. Lateral organization of membrane proteins: tetraspanins spin their web. Biochem J 420:133–154.

128. Barreiro O, Zamai M, Yanez-Mo M, Tejera E, Lopez-Romero P, Monk PN, Gratton E, Caiolfa VR, Sanchez-Madrid F. 2008. Endothelial adhesion receptors are recruited to adherent leukocytes by inclusion in preformed tetraspanin nanoplatforms. J Cell Biol 183:527–542.

129. Slupsky JR, Seehafer JG, Tang SC, Masellis-Smith A, Shaw AR. 1989. Evidence that monoclonal antibodies against CD9 antigen induce specific association between CD9 and the platelet glycoprotein IIb-IIIa complex. J Biol Chem 264:12289–12293.

130. Serru V, Le Naour F, Billard M, Azorsa DO, Lanza F, Boucheix C, Rubinstein E. 1999. Selective tetraspan-integrin complexes (CD81/alpha4beta1, CD151/alpha3beta1, CD151/alpha6beta1) under conditions disrupting tetraspan interactions. Biochem J 340 ( Pt 1:103–111.

110

131. Saint-Pol J, Eschenbrenner E, Dornier E, Boucheix C, Charrin S, Rubinstein E. 2017. Regulation of the trafficking and the function of the metalloprotease ADAM10 by tetraspanins. Biochem Soc Trans 45:937–944.

132. Boucheix C, Rubinstein E. 2001. Tetraspanins. Cell Mol Life Sci C 58:1189– 1205.

133. Hemler ME. 2005. Tetraspanin functions and associated microdomains. Nat Rev Mol Cell Biol 6:801–811.

134. Claas C, Stipp CS, Hemler ME. 2001. Evaluation of prototype transmembrane 4 superfamily protein complexes and their relation to lipid rafts. J Biol Chem 276:7974–7984.

135. Garner AE, Smith DA, Hooper NM. 2008. Visualization of detergent solubilization of membranes: implications for the isolation of rafts. Biophys J 94:1326–1340.

136. Hanagata N, Li X. 2011. Osteoblast-enriched membrane protein IFITM5 regulates the association of CD9 with an FKBP11–CD81–FPRP complex and stimulates expression of interferon-induced genes. Biochem Biophys Res Commun 409:378–384.

137. Zuidscherwoude M, Göttfert F, Dunlock VME, Figdor CG, van den Bogaart G, Spriel AB van. 2015. The tetraspanin web revisited by super-resolution microscopy. Sci Rep 5:12201.

138. Pileri P, Uematsu Y, Campagnoli S, Galli G, Falugi F, Petracca R, Weiner AJ, Houghton M, Rosa D, Grandi G, Abrignani S. 1998. Binding of Hepatitis C Virus to CD81. Science (80- ) 282:938–941.

139. Hochdorfer D, Florin L, Sinzger C, Lieber D. 2016. Tetraspanin CD151 Promotes Initial Events in Human Cytomegalovirus Infection. J Virol 90:6430–6442.

140. Raaben M, Jae LT, Herbert AS, Kuehne AI, Stubbs SH, Chou Y-Y, Blomen VA, Kirchhausen T, Dye JM, Brummelkamp TR, Whelan SP. 2017. NRP2 and CD63 Are Host Factors for Lujo Virus Cell Entry. Cell Host Microbe 22:688–696.e5.

141. Scheffer KD, Gawlitza A, Spoden GA, Zhang XA, Lambert C, Berditchevski F, Florin L. 2013. Tetraspanin CD151 mediates papillomavirus type 16 endocytosis. J Virol 87:3435–3446.

142. He J, Sun E, Bujny M V, Kim D, Davidson MW, Zhuang X. 2013. Dual Function of CD81 in Influenza Virus Uncoating and Budding. PLoS Pathog 9:e1003701.

143. Ooi YS, Stiles KM, Liu CY, Taylor GM, Kielian M. 2013. Genome-Wide RNAi Screen Identifies Novel Host Proteins Required for Alphavirus Entry. PLOS Pathog 9:e1003835.

111

144. Stiles KM, Kielian M. 2016. Role of TSPAN9 in Alphavirus Entry and Early Endosomes. J Virol 90:4289–4297.

145. Spoden G, Freitag K, Husmann M, Boller K, Sapp M, Lambert C, Florin L. 2008. Clathrin- and Caveolin-Independent Entry of Human Papillomavirus Type 16â€”Involvement of Tetraspanin-Enriched Microdomains (TEMs). PLoS One 3:e3313.

146. Bartosch B, Vitelli A, Granier C, Goujon C, Dubuisson J, Pascale S, Scarselli E, Cortese R, Nicosia A, Cosset F-L. 2003. Cell entry of hepatitis C virus requires a set of co-receptors that include the CD81 tetraspanin and the SR-B1 scavenger receptor. J Biol Chem 278:41624–41630.

147. Grigorov B, Reungoat E, Gentil Dit Maurin A, Varbanov M, Blaising J, Michelet M, Manuel R, Parent R, Bartosch B, Zoulim F, Ruggiero F, Pecheur E-I. 2017. Hepatitis C virus infection propagates through interactions between Syndecan-1 and CD81 and impacts the hepatocyte glycocalyx. Cell Microbiol 19.

148. Okamoto T, Iwata S, Yamazaki H, Hatano R, Komiya E, Dang NH, Ohnuma K, Morimoto C. 2014. CD9 negatively regulates CD26 expression and inhibits CD26-mediated enhancement of invasive potential of malignant mesothelioma cells. PLoS One 9:e86671–e86671.

149. Runge KE, Evans JE, He Z-Y, Gupta S, McDonald KL, Stahlberg H, Primakoff P, Myles DG. 2007. Oocyte CD9 is enriched on the microvillar membrane and required for normal microvillar shape and distribution. Dev Biol 304:317–325.

150. Kaji K, Oda S, Shikano T, Ohnuki T, Uematsu Y, Sakagami J, Tada N, Miyazaki S, Kudo A. 2000. The gamete fusion process is defective in eggs of Cd9-deficient mice. Nat Genet 24:279–282.

151. Miyado K, Yoshida K, Yamagata K, Sakakibara K, Okabe M, Wang X, Miyamoto K, Akutsu H, Kondo T, Takahashi Y, Ban T, Ito C, Toshimori K, Nakamura A, Ito M, Miyado M, Mekada E, Umezawa A. 2008. The fusing ability of sperm is bestowed by CD9-containing vesicles released from eggs in mice. Proc Natl Acad Sci U S A 105:12921–12926.

152. Tachibana I, Hemler ME. 1999. Role of transmembrane 4 superfamily (TM4SF) proteins CD9 and CD81 in muscle cell fusion and myotube maintenance. J Cell Biol 146:893–904.

153. Burton DR, Poignard P, Stanfield RL, Wilson IA. 2012. Broadly neutralizing antibodies present new prospects to counter highly antigenically diverse viruses. Science 337:183–186.

154. Bailey CC, Kondur HR, Huang IC, Farzan M. 2013. Interferon-induced transmembrane protein 3 is a type II transmembrane protein. J Biol Chem 288:32184–32193.

112

155. Sallman Almen M, Bringeland N, Fredriksson R, Schioth HB. 2012. The dispanins: a novel gene family of ancient origin that contains 14 human members. PLoS One 7:e31961.

156. Chesarino NM, Compton AA, McMichael TM, Kenney AD, Zhang L, Soewarna V, Davis M, Schwartz O, Yount JS. 2017. IFITM3 requires an amphipathic helix for antiviral activity. EMBO Rep 18:1740–1751.

157. Ling S, Zhang C, Wang W, Cai X, Yu L, Wu F, Zhang L, Tian C. 2016. Combined approaches of EPR and NMR illustrate only one transmembrane helix in the human IFITM3. Sci Rep 6:24029.

158. John SP, Chin CR, Perreira JM, Feeley EM, Aker AM, Savidis G, Smith SE, Elia AEH, Everitt AR, Vora M, Pertel T, Elledge SJ, Kellam P, Brass AL. 2013. The CD225 Domain of IFITM3 Is Required for both IFITM Protein Association and Inhibition of Influenza A Virus and Dengue Virus Replication. J Virol 87:7837– 7852.

159. Friedman RL, Manly SP, McMahon M, Kerr IM, Stark GR. 1984. Transcriptional and posttranscriptional regulation of interferon-induced gene expression in human cells. Cell 38:745–755.

160. Reid LE, Brasnett AH, Gilbert CS, Porter AC, Gewert DR, Stark GR, Kerr IM. 1989. A single DNA response element can confer inducibility by both alpha- and gamma-interferons. Proc Natl Acad Sci U S A 86:840–844.

161. Schneider WM, Chevillotte MD, Rice CM. 2014. Interferon-Stimulated Genes: A Complex Web of Host Defenses. Annu Rev Immunol 32:513–545.

162. Desai TM, Marin M, Chin CR, Savidis G, Brass AL, Melikyan GB. 2014. IFITM3 restricts influenza A virus entry by blocking the formation of fusion pores following virus-endosome hemifusion. PLoS Pathog 10:e1004048.

163. Brass AL, Huang I-C, Benita Y, John SP, Krishnan MN, Feeley EM, Ryan BJ, Weyer JL, van der Weyden L, Fikrig E, Adams DJ, Xavier RJ, Farzan M, Elledge SJ. 2009. The IFITM Proteins Mediate Cellular Resistance to Influenza A H1N1 Virus, West Nile Virus, and Dengue Virus. Cell 139:1243–1254.

164. Feeley EM, Sims JS, John SP, Chin CR, Pertel T, Chen L-M, Gaiha GD, Ryan BJ, Donis RO, Elledge SJ, Brass AL. 2011. IFITM3 Inhibits Influenza A Virus Infection by Preventing Cytosolic Entry. PLoS Pathog 7:e1002337.

165. Huang I-C, Bailey CC, Weyer JL, Radoshitzky SR, Becker MM, Chiang JJ, Brass AL, Ahmed AA, Chi X, Dong L, Longobardi LE, Boltz D, Kuhn JH, Elledge SJ, Bavari S, Denison MR, Choe H, Farzan M. 2011. Distinct Patterns of IFITM- Mediated Restriction of Filoviruses, SARS Coronavirus, and Influenza A Virus. PLOS Pathog 7:e1001258.

113

166. Tartour K, Nguyen X-N, Appourchaux R, Assil S, Barateau V, Bloyet L-M, Burlaud Gaillard J, Confort M-P, Escudero-Perez B, Gruffat H, Hong SS, Moroso M, Reynard O, Reynard S, Decembre E, Ftaich N, Rossi A, Wu N, Arnaud F, Baize S, Dreux M, Gerlier D, Paranhos-Baccala G, Volchkov V, Roingeard P, Cimarelli A. 2017. Interference with the production of infectious viral particles and bimodal inhibition of replication are broadly conserved antiviral properties of IFITMs. PLOS Pathog 13:e1006610.

167. Yu J, Li M, Wilkins J, Ding S, Swartz TH, Esposito AM, Zheng YM, Freed EO, Liang C, Chen BK, Liu SL. 2015. IFITM Proteins Restrict HIV-1 Infection by Antagonizing the Envelope Glycoprotein. Cell Rep 13:145–156.

168. Zhu X, He Z, Yuan J, Wen W, Huang X, Hu Y, Lin C, Pan J, Li R, Deng H, Liao S, Zhou R, Wu J, Li J, Li M. 2015. IFITM3-containing exosome as a novel mediator for anti-viral response in dengue virus infection. Cell Microbiol 17:105– 118.

169. Li K, Markosyan RM, Zheng Y-M, Golfetto O, Bungart B, Li M, Ding S, He Y, Liang C, Lee JC, Gratton E, Cohen FS, Liu S-L. 2013. IFITM Proteins Restrict Viral Membrane Hemifusion. PLoS Pathog 9:e1003124.

170. Yount JS, Moltedo B, Yang Y-Y, Charron G, Moran TM, LÃ3pez B. C, Hang HC. 2010. Palmitoylome profiling reveals S-palmitoylationâ€“dependent antiviral activity of IFITM3. Nat Chem Biol 6:610–614.

171. Yount JS, Karssemeijer RA, Hang HC. 2012. S-palmitoylation and ubiquitination differentially regulate interferon-induced transmembrane protein 3 (IFITM3)- mediated resistance to influenza virus. J Biol Chem 287:19631–19641.

172. Wilkins C, Woodward J, Lau DT, Barnes A, Joyce M, McFarlane N, McKeating JA, Tyrrell DL, Gale Jr. M. 2013. IFITM1 is a tight junction protein that inhibits hepatitis C virus entry. Hepatology 57:461–469.

173. Chen YX, Welte K, Gebhard DH, Evans RL. 1984. Induction of T cell aggregation by antibody to a 16kd human leukocyte surface antigen. J Immunol 133:2496– 2501.

174. Tanaka SS, Yamaguchi YL, Tsoi B, Lickert H, Tam PPL. 2005. IFITM/Mil/Fragilis Family Proteins IFITM1 and IFITM3 Play Distinct Roles in Mouse Primordial Germ Cell Homing and Repulsion. Dev Cell 9:745–756.

175. Li D, Peng Z, Tang H, Wei P, Kong X, Yan D, Huang F, Li Q, Le X, Li Q, Xie K. 2011. KLF4-Mediated Negative Regulation of IFITM3 Expression Plays a Critical Role in Colon Cancer Pathogenesis. Clin Cancer Res 17:3558–3568.

176. Rosati A, Basile A, D/’Auria R, d/’Avenia M, De Marco M, Falco A, Festa M, Guerriero L, Iorio V, Parente R, Pascale M, Marzullo L, Franco R, Arra C, Barbieri A, Rea D, Menichini G, Hahne M, Bijlsma M, Barcaroli D, Sala G, di

114

Mola FF, di Sebastiano P, Todoric J, Antonucci L, Corvest V, Jawhari A, Firpo MA, Tuveson DA, Capunzo M, Karin M, De Laurenzi V, Turco MC. 2015. BAG3 promotes pancreatic ductal adenocarcinoma growth by activating stromal macrophages. Nat Commun 6.

177. Moffatt P, Gaumond M-H, Salois P, Sellin K, Bessette M-C, Godin É, de Oliveira PT, Atkins GJ, Nanci A, Thomas G. 2008. Bril: A Novel Bone-Specific Modulator of Mineralization. J Bone Miner Res 23:1497–1508.

178. Cho T-J, Lee K-E, Lee S-K, Song SJ, Kim KJ, Jeon D, Lee G, Kim H-N, Lee HR, Eom H-H, Lee ZH, Kim O-H, Park W-Y, Park SS, Ikegawa S, Yoo WJ, Choi IH, Kim J-W. 2012. A Single Recurrent Mutation in the 5′-UTR of IFITM5 Causes Osteogenesis Imperfecta Type V. Am J Hum Genet 91:343–348.

179. Semler O, Garbes L, Keupp K, Swan D, Zimmermann K, Becker J, Iden S, Wirth B, Eysel P, Koerber F, Schoenau E, Bohlander SK, Wollnik B, Netzer C. 2012. A Mutation in the 5′-UTR of IFITM5 Creates an In-Frame Start Codon and Causes Autosomal-Dominant Osteogenesis Imperfecta Type V with Hyperplastic Callus. Am J Hum Genet 91:349–357.

180. Hoyer-Kuhn H, Semler O, Garbes L, Zimmermann K, Becker J, Wollnik B, Schoenau E, Netzer C. 2014. A Nonclassical IFITM5 Mutation Located in the Coding Region Causes Severe Osteogenesis Imperfecta With Prenatal Onset. J Bone Miner Res 29:1387–1391.

181. Glorieux FH, Rauch F, Plotkin H, Ward L, Travers R, Roughley P, Lalic L, Glorieux DF, Fassier F, Bishop NJ. 2000. Type V Osteogenesis Imperfecta: A New Form of Brittle Bone Disease. J Bone Miner Res 15:1650–1658.

182. Smith RA, Young J, Weis JJ, Weis JH. 2006. Expression of the mouse fragilis gene products in immune cells and association with receptor signaling complexes. Genes Immun 7:113–121.

183. Hickford D, Frankenberg S, Shaw G, Renfree M. 2012. Evolution of vertebrate interferon inducible transmembrane proteins. BMC Genomics 13:155.

184. Zhang Z, Liu J, Li M, Yang H, Zhang C. 2012. Evolutionary Dynamics of the Interferon-Induced Transmembrane Gene Family in Vertebrates. PLoS One 7:e49265.

185. Ranganath P, Stephen J, Iyengar R, Phadke SR. 2016. Worsening of Callus Hyperplasia after Bisphosphonate Treatment in Type V Osteogenesis Imperfecta. Indian Pediatr 53:250–252.

186. Lietman CD, Marom R, Munivez E, Bertin TK, Jiang M-M, Chen Y, Dawson B, Weis M, Eyre D, Lee B. 2014. A transgenic mouse model of OI type V supports a neomorphic mechanism of the IFITM5 mutation. J Bone Miner Res n/a-n/a.

115

187. Farber CR, Reich A, Barnes AM, Becerra P, Rauch F, Cabral WA, Bae A, Quinlan A, Glorieux FH, Clemens TL, Marini JC. 2014. A novel IFITM5 mutation in severe atypical osteogenesis imperfecta type VI impairs osteoblast production of pigment epithelium-derived factor. J Bone Min Res 29:1402–1411.

188. Reich A, Bae AS, Barnes AM, Cabral WA, Hinek A, Stimec J, Hill SC, Chitayat D, Marini JC. 2015. Type V OI primary osteoblasts display increased mineralization despite decreased COL1A1 expression. J Clin Endocrinol Metab 100:E325-32.

189. Guillen-Navarro E, Ballesta-Martinez MJ, Valencia M, Bueno AM, Martinez-Glez V, Lopez-Gonzalez V, Burnyte B, Utkus A, Lapunzina P, Ruiz-Perez VL. 2014. Two mutations in IFITM5 causing distinct forms of osteogenesis imperfecta. Am J Med Genet A 164a:1136–1142.

190. Sanjana NE, Shalem O, Zhang F. 2014. Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods. United States.

191. Hitoshi N, Ken-ichi Y, Jun-ichi M. 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108:193–199.

192. Campbell EM, Perez O, Melar M, Hope TJ. 2007. Labeling HIV-1 virions with two fluorescent proteins allows identification of virions that have productively entered the target cell. Virology 360:286–293.

193. Boscarino JA, Logan HL, Lacny JJ, Gallagher TM. 2008. Envelope Protein Palmitoylations Are Crucial for Murine Coronavirus Assembly. J Virol 82:2989– 2999.

194. Thery C, Amigorena S, Raposo G, Clayton A. 2006. Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr Protoc cell Biol Chapter 3: Unit 3.22.

195. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. 2013. Genome engineering using the CRISPR-Cas9 system. Nat Protoc 8:2281–2308.

196. Li W, Moore MJ, Vasilieva N, Sui J, Wong SK, Berne MA, Somasundaran M, Sullivan JL, Luzuriaga K, Greenough TC, Choe H, Farzan M. 2003. Angiotensin- converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature 426:450–454.

197. Soderberg O, Gullberg M, Jarvius M, Ridderstrale K, Leuchowius K-J, Jarvius J, Wester K, Hydbring P, Bahram F, Larsson L-G, Landegren U. 2006. Direct observation of individual endogenous protein complexes in situ by proximity ligation. Nat Meth 3:995–1000.

198. Zhou H, Bouwman K, Schotanus M, Verweij C, Marrero JA, Dillon D, Costa J, Lizardi P, Haab BB. 2004. Two-color, rolling-circle amplification on antibody

116

microarrays for sensitive, multiplexed serum-protein measurements. Genome Biol 5:R28.

199. Zhao J, Li K, Wohlford-Lenane C, Agnihothram SS, Fett C, Zhao J, Gale MJ, Baric RS, Enjuanes L, Gallagher T, McCray PB, Perlman S. 2014. Rapid generation of a mouse model for Middle East respiratory syndrome. Proc Natl Acad Sci 111:4970 LP-4975.

200. Kondo N, Miyauchi K, Meng F, Iwamoto A, Matsuda Z. 2010. Conformational Changes of the HIV-1 Envelope Protein during Membrane Fusion Are Inhibited by the Replacement of Its Membrane-spanning Domain. J Biol Chem 285:14681–14688.

201. Fang Y, Wu N, Gan X, Yan W, Morrell JC, Gould SJ. 2007. Higher-Order Oligomerization Targets Plasma Membrane Proteins and HIV Gag to Exosomes. PLOS Biol 5:e158.

202. Rodgers W. 2002. Making membranes green: construction and characterization of GFP-fusion proteins targeted to discrete plasma membrane domains.BioTechniques. England.

203. Dragovic RA, Gardiner C, Brooks AS, Tannetta DS, Ferguson DJP, Hole P, Carr B, Redman CWG, Harris AL, Dobson PJ, Harrison P, Sargent IL. 2011. Sizing and phenotyping of cellular vesicles using Nanoparticle Tracking Analysis. Nanomedicine 7:780–788.

204. MacKenzie A, Wilson HL, Kiss-Toth E, Dower SK, North RA, Surprenant A. 2001. Rapid Secretion of Interleukin-1β by Microvesicle Shedding. Immunity 15:825– 835.

205. Savina A, Furlán M, Vidal M, Colombo MI. 2003. Exosome Release Is Regulated by a Calcium-dependent Mechanism in K562 Cells. J Biol Chem 278:20083– 20090.

206. Kanemoto S, Nitani R, Murakami T, Kaneko M, Asada R, Matsuhisa K, Saito A, Imaizumi K. 2016. Multivesicular body formation enhancement and exosome release during endoplasmic reticulum stress. Biochem Biophys Res Commun 480:166–172.

207. Edgar JR, Manna PT, Nishimura S, Banting G, Robinson MS. 2016. Tetherin is an exosomal tether. Elife 5.

208. Aatonen MT, Öhman T, Nyman TA, Laitinen S, Grönholm M, Siljander PR-M. 2014. Isolation and characterization of platelet-derived extracellular vesicles. J Extracell Vesicles 3:10.3402/jev.v3.24692.

209. Bailey CC, Huang I-C, Kam C, Farzan M. 2012. Ifitm3 Limits the Severity of Acute Influenza in Mice. PLOS Pathog 8:e1002909.

117

210. Zhou J, Aiken C. 2001. Nef enhances human immunodeficiency virus type 1 infectivity resulting from intervirion fusion: evidence supporting a role for Nef at the virion envelope. J Virol 75:5851–5859.

211. Borger V, Bremer M, Ferrer-Tur R, Gockeln L, Stambouli O, Becic A, Giebel B. 2017. Mesenchymal Stem/Stromal Cell-Derived Extracellular Vesicles and Their Potential as Novel Immunomodulatory Therapeutic Agents. Int J Mol Sci 18.

212. Hsu H, Lacey DL, Dunstan CR, Solovyev I, Colombero A, Timms E, Tan H-L, Elliott G, Kelley MJ, Sarosi I, Wang L, Xia X-Z, Elliott R, Chiu L, Black T, Scully S, Capparelli C, Morony S, Shimamoto G, Bass MB, Boyle WJ. 1999. Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci 96:3540– 3545.

213. Jansen IDC, Vermeer JAF, Bloemen V, Stap J, Everts V. 2012. Osteoclast Fusion and Fission. Calcif Tissue Int 90:515–522.

214. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ. 1998. Osteoprotegerin Ligand Is a Cytokine that Regulates Osteoclast Differentiation and Activation. Cell 93:165–176.

215. Lin T-Y, Chin CR, Everitt AR, Clare S, Perreira JM, Savidis G, Aker AM, John SP, Sarlah D, Carreira EM, Elledge SJ, Kellam P, Brass AL. 2013. Amphotericin B Increases Influenza A Virus Infection by Preventing IFITM3-Mediated Restriction. Cell Rep 5:895–908.

216. Henry-Toulme N, Sarthou P, Seman M, Bolard J. 1989. Membrane effects of the polyene antibiotic amphotericin B and of some of its derivatives on lymphocytes. Mol Cell Biochem 91:39–44.

217. Younsi M, Ramanandraibe E, Bonaly R, Donner M, Coulon J. 2000. Amphotericin B resistance and membrane fluidity in Kluyveromyces lactis strains. Antimicrob Agents Chemother 44:1911–1916.

218. Earnest JT, Hantak MP, Li K, McCray PB, Perlman S, Gallagher T. 2017. The tetraspanin CD9 facilitates MERS-coronavirus entry by scaffolding host cell receptors and proteases. PLoS Pathog 13.

219. Millet JK, Whittaker GR. 2014. Host cell entry of Middle East respiratory syndrome coronavirus after two-step, furin-mediated activation of the spike protein. Proc Natl Acad Sci U S A 111:15214–15219.

220. Phillips JM, Gallagher T, Weiss SR. 2017. Neurovirulent Murine Coronavirus JHM.SD Uses Cellular Zinc Metalloproteases for Virus Entry and Cell-Cell Fusion. J Virol 91.

118

221. Bonnin A, Danneels A, Dubuisson J, Goffard A, Belouzard S. 2018. HCoV-229E spike protein fusion activation by trypsin-like serine proteases is mediated by proteolytic processing in the S2’ region. J Gen Virol 99:908–912.

222. Shoham T, Rajapaksa R, Kuo C-C, Haimovich J, Levy S. 2006. Building of the tetraspanin web: distinct structural domains of CD81 function in different cellular compartments. Mol Cell Biol 26:1373–1385.

223. Vargas A, Zhou S, Ethier-Chiasson M, Flipo D, Lafond J, Gilbert C, Barbeau B. 2014. Syncytin proteins incorporated in placenta exosomes are important for cell uptake and show variation in abundance in serum exosomes from patients with preeclampsia. FASEB J Off Publ Fed Am Soc Exp Biol 28:3703–3719.

224. Masciopinto F, Giovani C, Campagnoli S, Galli-Stampino L, Colombatto P, Brunetto M, Yen TSB, Houghton M, Pileri P, Abrignani S. 2004. Association of hepatitis C virus envelope proteins with exosomes. Eur J Immunol 34:2834– 2842.

225. Nabet BY, Qiu Y, Shabason JE, Wu TJ, Yoon T, Kim BC, Benci JL, DeMichele AM, Tchou J, Marcotrigiano J, Minn AJ. 2017. Exosome RNA Unshielding Couples Stromal Activation to Pattern Recognition Receptor Signaling in Cancer. Cell 170:352–366.e13.

226. Ranjbar S, Haridas V, Jasenosky LD, Falvo J V, Goldfeld AE. 2015. A Role for IFITM Proteins in Restriction of Mycobacterium tuberculosis Infection. Cell Rep 13:874–883.

227. Evans SS, Lee DB, Han T, Tomasi TB, Evans RL. 1990. Monoclonal antibody to the interferon-inducible protein Leu-13 triggers aggregation and inhibits proliferation of leukemic B cells. Blood 76:2583–2593.

228. Raggatt LJ, Partridge NC. 2010. Cellular and Molecular Mechanisms of Bone Remodeling. J Biol Chem 285:25103–25108.

229. Melvin WJ, McMichael TM, Chesarino NM, Hach JC, Yount JS. 2015. IFITMs from Mycobacteria Confer Resistance to Influenza Virus When Expressed in Human Cells. Viruses 7:3035–3052.

230. Wrensch F, Karsten CB, Gnirss K, Hoffmann M, Lu K, Takada A, Winkler M, Simmons G, Pohlmann S. 2015. Interferon-Induced Transmembrane Protein- Mediated Inhibition of Host Cell Entry of Ebolaviruses. J Infect Dis 212 Suppl:S210-8.

231. Takayanagi H, Kim S, Matsuo K, Suzuki H, Suzuki T, Sato K, Yokochi T, Oda H, Nakamura K, Ida N, Wagner EF, Taniguchi T. 2002. RANKL maintains bone homeostasis through c-Fos-dependent induction of interferon-beta. Nature 416:744–749.

119

232. Cohen BE. 1998. Amphotericin B toxicity and lethality: a tale of two channels. Int J Pharm 162:95–106.

233. Coorssen JR, Rand RP. 1990. Effects of cholesterol on the structural transitions induced by diacylglycerol in phosphatidylcholine and phosphatidylethanolamine bilayer systems. Biochem Cell Biol 68:65–69.

234. Chen Z, Rand RP. 1997. The influence of cholesterol on phospholipid membrane curvature and bending elasticity. Biophys J 73:267–276.

VITA

Michael Hantak was born in Crete, Illinois to Phil and Kim Hantak. In 2007, he received his first laboratory experience at the University of Chicago, where he worked with Devkumar Mustafi, Ph.D., to develop novel MRI contrast agents. He attended

Illinois State University in Normal, IL, where he earned a Bachelor’s of Science in

Biochemistry/Molecular Biology and Biological Sciences in May 2012. During his undergraduate studies, Michael worked in the lab of Brian Wilkinson, Ph.D., where he studied mechanisms of cold tolerance in Listeria monocytogenes. In August 2012,

Michael entered the Interdisciplinary Program in Biomedical Sciences at Loyola

University Chicago, and joined the Department of Microbiology and Immunology shortly after. He completed his doctoral work in the laboratory of Tom Gallagher, Ph.D., where he focused on characterizing platforms of membrane fusion and cellular communication.

This work was supported by the Arthur J. Schmitt Dissertation Fellowship and a T32

Immunology Training Grant awarded to Dr. Katherine Knight. This work was also supported by the FALK Institute and the Doyle Fund. After completion of his graduate studies, Michael will continue his postdoctoral research in the laboratory of Jason

Shepherd, Ph.D., at the University of Utah in Salt Lake City. There Michael will study the role of endogenous retroviruses in the maintenance of long-term memory.

121