The Role of AMPK in Viral Infection

Home , Leporipoxvirus, Toscana phlebovirus

University of Pennsylvania ScholarlyCommons

Publicly Accessible Penn Dissertations

Summer 2011

Theresa S. Moser University of Pennsylvania, [email protected]

Follow this and additional works at: https://repository.upenn.edu/edissertations

Part of the Cell Biology Commons, and the Virology Commons

Recommended Citation Moser, Theresa S., "The Role of AMPK in Viral Infection" (2011). Publicly Accessible Penn Dissertations. 975. https://repository.upenn.edu/edissertations/975

This paper is posted at ScholarlyCommons. https://repository.upenn.edu/edissertations/975 For more information, please contact [email protected]. The Role of AMPK in Viral Infection

Abstract Host factors are crucial in determining the outcome of viral infection. As obligate intracellular pathogens, viruses are highly dependent on numerous cellular proteins and processes, and also must evade multiple antiviral defense mechanisms. Since viral pathogens represent a significant threat both to human health and agriculture, a better understanding of these interactions could facilitate the treatment of viral infections and provide novel targets for therapeutics. In particular, Poxviruses include medically important human pathogens yet little is known about the specific cellular factors essential for their replication. To identify genes essential for vaccinia infection, I used high-throughput RNA interference to screen Drosophila kinases and phosphatases. I identified seven genes including the three subunits of AMPK as promoting vaccinia infection. AMPK facilitated infection both in insect and mammalian cells. I found that AMPK is required for macropinocytosis, a major endocytic entry pathway for vaccinia. Furthermore, AMPK contributes to other actin-dependent cellular processes including lamellipodia formation and wound healing, independent of the known AMPK activators LKB1 and CaMKK. Therefore, AMPK plays a highly conserved role in poxvirus infection and actin dynamics independent of its role as an energy regulator.

This led me to investigate the role of AMPK in other viral infections, including human arthropod-borne viral pathogens. Surprisingly I found that AMPK expression and activation restricts infection of the Bunyavirus Rift Valley Fever Virus (RVFV) in an LKB1-dependent manner. RVFV like other RNA viruses manipulates cellular membranes to form vesicle-like structures that support the viral replication complex in a process dependent on de novo fatty acid synthesis. I found that AMPK activation potently inhibits acetyl Co A carboxylase, crucial in fatty acid synthesis, leading to decreased levels of cellular lipids. Bypassing fatty acid synthesis by treating cells with palmitate, the first product of fatty acid synthesis rescued infection, demonstrating that AMPK restricts RVFV by inhibiting fatty acid synthesis. Additional RNA viruses that require membrane proliferations including the Togavirus Sindbis and the Alphavirus Vesicular Stomatitis Virus were also restricted by AMPK. AMPK is likely a component of the intrinsic innate immune response against RNA viruses, and may provide a target for broadly anti-viral therapeutics.

Degree Type Dissertation

Degree Name Doctor of Philosophy (PhD)

Graduate Group Cell & Molecular Biology

First Advisor Sara Cherry

Second Advisor Stuart N. Isaacs

Keywords poxvirus entry, macropinocytosis, AMPK, kinase signaling, Rift Valley Fever Virus, fatty acid metabolism Subject Categories Cell Biology | Virology

This dissertation is available at ScholarlyCommons: https://repository.upenn.edu/edissertations/975 THE ROLE OF AMPK IN VIRAL INFECTION

Theresa S. Moser

A DISSERTATION

Cell and Molecular Biology

Presented to the Faculties of the University of Pennsylvania

In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

2011

Dissertation Supervisor:

Sara Cherry, Assistant Professor of Microbiology

Graduate Group Chairperson:

Daniel S. Kessler, Associate Professor of Cell and Developmental Biology

Dissertation Committee:

Stuart N. Isaacs, Associate Professor of Medicine (Committee Chair) Morris J. Birnbaum, Professor of Medicine Frederic D. Bushman, Professor of Microbiology Amita Sehgal, Professor of Neuroscience

DEDICATION

I would like to dedicate this dissertation to my grandparents, George and Elaine Keys, who first encouraged me to go into research and offered perpetual love and support.

ACKNOWLEDGEMENTS

I would like to thank the following individuals for their contributions to this work:

Sara Cherry for guidance and support through every step of the way, always pushing me to improve, and sharing a love of science.

My thesis committee members Stuart Isaacs (chair), Morrie Birnbaum, Rick Bushman, and Amita Sehgal for their insight, mentorship, and support.

Russell Jones and Craig Thompson for providing the AMPK and LKB1 deficient cells essential for this work.

Carolyn Coyne for assistance with microscopy, macropinocytosis, and Photoshop.

Meghana Kulkarni and Norbert Perrimon for the kinase/phosphatase set of dsRNAs.

Bernard Moss, Stuart Isaacs, Robert Doms, Gary Cohen, and Roselyn Eisenberg for vaccinia virus strains and antibodies.

Fang Hua Lee and Phil Arca for assistance growing large quantities of vaccinia.

Michele Bland, Russell Miller and Morrie Birnbaum for useful reagents and technical assistance with AMPK activation studies.

Robert Rawson for useful reagents in lipid metabolism studies.

Leah Sabin for organization, kindness, and a great JoVE video.

Kaycie Hopkins for RVFV, electrical expertise, and open-mindedness.

Jie Xu for late night conversations and enthusiasm.

Sheri Hanna for support and advice especially through difficult periods.

Patrick Rose for Sindbis, plate fixing, and conversations about lab and life.

Spencer Shelly for VSV, Western expertise, and stability.

Shelly Bambina and Maggie Nakamoto for VSV and holding the lab together.

The members of the Cherry lab for suggestions, criticism, and encouragement, and making the Cherry lab an exceptional place to do research.

My friends and family for encouragement and support.

This work was funded by NIH training grant T32HG000046. iii

ABSTRACT

THE ROLE OF AMPK IN VIRAL INFECTION

Theresa S. Moser

Thesis Advisor: Sara Cherry

Host factors are crucial in determining the outcome of viral infection. As obligate intracellular pathogens, viruses are highly dependent on numerous cellular proteins and processes, and also must evade multiple antiviral defense mechanisms. Since viral pathogens represent a significant threat both to human health and agriculture, a better understanding of these interactions could facilitate the treatment of viral infections and provide novel targets for therapeutics. In particular, Poxviruses include medically important human pathogens yet little is known about the specific cellular factors essential for their replication. To identify genes essential for vaccinia infection, I used high-throughput RNA interference to screen Drosophila kinases and phosphatases. I identified seven genes including the three subunits of AMPK as promoting vaccinia infection. AMPK facilitated infection both in insect and mammalian cells. I found that

AMPK is required for macropinocytosis, a major endocytic entry pathway for vaccinia.

Furthermore, AMPK contributes to other actin-dependent cellular processes including lamellipodia formation and wound healing, independent of the known AMPK activators

LKB1 and CaMKK. Therefore, AMPK plays a highly conserved role in poxvirus infection and actin dynamics independent of its role as an energy regulator.

LKB1-dependent manner. RVFV like other RNA viruses manipulates cellular membranes to form vesicle-like structures that support the viral replication complex in a process dependent on de novo fatty acid synthesis. I found that AMPK activation potently inhibits acetyl Co A carboxylase, crucial in fatty acid synthesis, leading to decreased levels of cellular lipids. Bypassing fatty acid synthesis by treating cells with palmitate, the first product of fatty acid synthesis rescued infection, demonstrating that AMPK restricts

RVFV by inhibiting fatty acid synthesis. Additional RNA viruses that require membrane proliferations including the Togavirus Sindbis and the Alphavirus Vesicular Stomatitis

Virus were also restricted by AMPK. AMPK is likely a component of the intrinsic innate immune response against RNA viruses, and may provide a target for broadly anti-viral therapeutics.

TABLE OF CONTENTS

I. INTRODUCTION………………………………………………………………………………1

1. The Study of Virus-Host Interactions Using RNAi Technology………………….1

2. Poxvirus Infection and Tropism……………………………………………………..3

3. Vaccinia Virus………………………………………………………………………...4

a. Vaccinia Lifecycle……………………………………………………………4

b. Vaccinia Entry and Macropinocytosis……………………………………..7

4. Arboviruses………………………………………………………………………….11

a. Rift Valley Fever Virus……………………………………………………..12

b. Sindbis Virus………………………………………………………………..15

c. Vesicular Stomatits Virus………………………………………………….17

d. Lipid Metabolism in Arboviral Infection…………………………………..19

5. Cellular Energy Metabolism……………………………………………………….20

6. Downstream Functions of AMPK……………………………………………….…22

a. AMPK Regulation of Protein Synthesis………………………………….23

b. AMPK Regulation of Autophagy………………………………………….25

c. AMPK Regulation of Fatty Acid Synthesis………………………………26

d. AMPK Regulation of the Cytoskeleton…………………………………..27

7. The Role of AMPK in Metabolic Diseases……………………………………….28

8. Aims of Present Studies……………………………………………………………30

II. MATERIALS AND METHODS…………………………………………………………….34

1. Cells, Antibodies, Reagents, and Viruses………………………………………34

2. RNAi and Infections……………………………………………………………….35

3. siRNA……………………………………………………………………………….36

4. Viral Immunofluorescense………………………………………………………..36

5. Screen Analysis……………………………………………………………………36

6. Immunoblotting, Northern Blotting and RT-PCR……………………………….37

7. Plaque Assays……………………………………………………………………..37

8. Vaccinia Entry Assay……………………………………………………………...38

9. Fluid Phase Dextran Uptake Assay……………………………………………..38

10. Actin Ruffling Assay……………………………………………………………….39

11. Transferrin Uptake Assay………………………………………………………...39

12. Wound Healing Assay…………………………………………………………….39

13. Cellular Lipid Staining……………………………………………………………..40

14. Fatty Acid Synthesis Bypass Assay……………………………………………..40

III. A KINOME RNAI SCREEN IDENTIFIED AMPK AS PROMOTING POXVIRUS

ENTRY THROUGH THE CONTROL OF ACTIN DYNAMICS…………………………….41

1. Introduction…………………………………………………………………………41

2. Results……………………………………………………………………………...43

a. Vaccinia Infection in Drosophila cells…………………………………..43

b. RNAi Screen of Drosophila Kinome…………………………………….45

c. AMPK Promotes Vaccinia Infection in Mammalian Cells…………….47

d. Vaccinia Infection Activates AMPK…………………………………...... 49

e. Known Upstream Activators of AMPK Are Not Required for Vaccinia

Infection……………………………………………………………………49

f. AMPK Promotes Vaccinia Entry….…………………………………….50

vii

g. AMPK Is Required for Vaccinia Induced Macropinocytosis….………51

h. AMPK Promotes Lamellipodia Formation………………………….…..52

i. AMPK Promotes Cellular Motility……..………………………………...53

3. Discussion………………………………………………………………………….54

IV. AMP-ACTIVATED KINASE RESTRICTS RIFT VALLEY FEVER VIRUS INFECTION

BY INHIBITING FATTY ACID SYNTHESIS………………………………………………...69

1. Introduction…………………………………………………………………………69

2. Results……………………………………………………………………………...72

a. AMPK Expression Restricts RVFV Infection…………………………..72

b. LKB1 Expression Restricts RVFV Infection……………………………74

c. AMPK Activation Restricts RVFV Infection…………………………….75

d. Acetyl CoA Carboxylase Activity is Inhibited by AMPK Expression and

Activation…………………………………………………………………..76

e. Palmitate Rescues AMPK-Mediated Restriction of RVFV…………...79

f. VSV and Sindbis Are Restricted by AMPK…………………………….80

3. Discussion………………………………………………………………………….81

V. CONCLUDING REMARKS………………………………………………………………..90

1. Vaccinia Promotes Vaccinia Entry………………………………………………..91

2. AMPK Restricts RNA Viruses……………………………………………………..94

VI. APPENDIX: SUPPLEMENTARY INFORMATION……………………………………..99

VII. REFERENCES…………………………………………………………………………...115

viii

LIST OF TABLES

Table 3.1: Candidate Genes Identified and Validated…………………………………63

LIST OF ILLUSTRATIONS

Figure 1.1: AMPK regulates protein and lipid metabolism……………………………..33

Figure 3.1: Vaccinia virus undergoes infectious entry in Drosophila cells……………61

Figure 3.2: High-content RNAi screen identifies cellular factors required for vaccinia infection………………………………………………………………………...62

Figure 3.3: AMPKα promotes poxvirus infection in mammalian cells………………...64

Figure 3.4: Vaccinia infection induces AMPK activation, but is independent of LKB1 and CaMKK……………………………………………………………………65

Figure 3.5: AMPK promotes vaccinia entry……………………………………………...66

Figure 3.6: AMPK is required for vaccinia-induced macropinocytosis………………..67

Figure 3.7: AMPK is required for actin-dependent membrane ruffling and wound healing independent of LKB1………………………………………………..68

Figure 4.1: AMPK expression restricts RVFV infection………………………………...84

Figure 4.2: LKB1 expression restricts RVFV infection………………………………….85

Figure 4.3 AMPK activation restricts RVFV……………………………………………..86

Figure 4.4 Acetyl CoA Carboxylase activity is inhibited by AMPK expression and

activation……………………………………………………………………….87

Figure 4.5 Addition of palmitate restores RVFV infection in the presence of A769662………………………………………………………………………..88

Figure 4.6 VSV and Sindbis are restricted by AMPK………………………………….89

I. INTRODUCTION

1. The Study of Virus-Host Interactions Using RNAi Technology

Since viruses are obligate intracellular pathogens with limited coding capacity, they must take advantage of numerous cellular processes and pathways in order to complete their lifecycle. At the same time the host has many strategies of anti-viral defense that viruses must evade in order to be successful. Since viral pathogens represent a significant threat both to human health and agriculture, a better understanding of these interactions could facilitate the treatment of viral infections and provide novel targets for therapeutics. There are several approaches that can be used to identify these host factors. One approach is to use small molecule inhibitors to identify cellular mechanisms involved in viral infection. A drawback to this approach is that many small molecules have multiple cellular targets, and often the specific genes involved remain unknown. A second approach is to use robust hypomorphic loss of function RNA interference (RNAi) methodology. RNAi allows for systematic gene silencing of targeted mRNAs, creating a platform for unbiased loss-of-function screening in cultured cells. Because whole genomes can be screened without requiring time consuming genetic manipulation, RNAi has become a valuable methodology for reverse genetic screening [1].

Cell-based screening in a Drosophila model system is particularly amenable to this approach, as it provides an excellent model system to probe the host contribution to viral infection through RNAi screening. RNAi silencing is extremely potent in Drosophila allowing robust knockdown of the targeted gene, and the reduced redundancy displayed in the Drosophila genome increases the likelihood of identifying candidates with single

1 gene RNAi [2,3]. Moreover, many developmental and cell biological processes are conserved from Drosophila to mammals, making the fruitfly a good model for studying mammalian systems [4]. Cell-based loss of function RNAi screens have been developed in Drosophila targeting either the entire genome or a subset of the genome, such as the kinase-phosphatase screen I employed in Chapter III [2,5].

These high-throughput screens that have been generated can be used for many different cell biological applications including the study of viral infection [6]. In particular, previous studies using these approaches have dissected the host factor requirements of

Drosophila C Virus (DCV). Using both cell-based screens and fly mutations, clathrin- mediated endocytosis was identified as an important component of DCV infection and pathogenesis [7]. Additional host components were identified through RNAi screens, revealing that DCV is also exquisitely sensitive to cellular translation machinery [8].

Notably, many of the components identified through these approaches were also important for the infection of adult flies with DCV, and for the infection of human cells with the related human pathogen, poliovirus [8]. This overlap demonstrates that host factors identified in Drosophila can play conserved roles in mammalian systems.

Moreover host factors were identified that control various steps in the virus life cycle including entry, translation, and RNA replication. This approach has also successfully been used to identify factors that contribute to the host innate immune response [9]. A wide variety of viruses can infect Drosophila cells, and since many medically and agriculturally important viruses are carried or transmitted by insects, a greater understanding of infection in insects is also beneficial [10]. Thus, studies in Drosophila can increase our understanding of the host factors that contribute to viral pathogenesis

2 as well as the host response, and allow the unbiased discovery and dissection of the host contribution to viral infection.

2. Poxvirus Infection and Tropism

Poxviruses are large double stranded DNA viruses that replicate entirely in the cytoplasm of infected cells [11]. This group includes members that infect a diverse group of animals, ranging from invertebrate insects (Entomopoxviruses), to vertebrate reptiles, birds, and mammals (Chordopoxviruses), giving this group both medical and ecological importance. Chordopoxviruses have been divided into eight genera, of which the

Orthopoxviruses are the best studied. This group includes variola virus, the causative agent of smallpox, as well as other similar viruses that contribute to human disease, such as cowpox, monkeypox, and vaccinia [12]. Variola virus was declared eradicated from nature in 1980 through very successful vaccination campaigns, but frozen stocks still exist, and this virus continues to be considered a bioterrorist threat [13].

Additionally, the closely related cowpox and monkeypox viruses can be transmitted to humans zoonotically from their natural rodent hosts and remain a public health concern

[14].

The Poxvirus genome consists of a single segment of double stranded DNA

~200kb long with hairpin loops at the termini. The genome encodes approximately 200 open frames with nearly a quarter of these genes conserved across all sequenced poxviruses [15]. Nearly half of the genes are conserved across Chordopoxviruses

[15,16]. The majority of these well conserved genes have housekeeping functions required in the process of making new progeny virus. Many of the other genes have immunomodulatory roles, and are believed to be responsible for the specific species

tropism of each virus, which varies greatly [17]. Viruses such as vaccinia, cowpox, and

monkeypox are able to cause disease in a wide range of animals, while others in this

group such as Variola, and Ectromelia viruses are restricted to a single species:

humans, and mice, respectively [12]. Studies with the Leporipoxvirus Myxoma have

demonstrated that altering the host response to infection can alter viral tropism.

Myxoma virus is highly specific to infection of rabbits and has an abortive phenotype in

other species and cell types; however, knocking out the interferon receptor can allow the

virus to productively infect normally abortive cell types [18,19]. This finding demonstrates

the importance of the host contribution in determining the outcome of infection.

3. Vaccinia Virus

Vaccinia virus is the best studied member of the poxvirus family, and was

created and used successfully as the vaccine for smallpox, leading to the eradication of

this disease. Sequence analysis indicates that vaccinia is equally similar to variola and

cowpox viruses, making it impossible to determine the exact origin of this virus

[15,20,21]. While the virus is not natural, infection is still a concern in some

circumstances. Vaccinia is an infectious virus, and there are reports of recently

vaccininated individuals, particularly soldiers, transmitting the infection to close personal

contacts [22,23,24,25,26,27,28]. This is a concern since vaccinia infection can cause

severe illness in some instances [22,24].

a. Vaccinia Lifecycle

Vaccinia has a complex lifecycle that requires interactions with numerous host components. The virus must first attach, enter cells, and uncoat through complex mechanisms that will be further discussed below. Once the virus has gained entry into 4 the cytoplasm of the host cell, the virus undergoes several rounds of temporally regulated nucleic acid synthesis. The first replication step is the ATP-dependent transcription of early genes, which is mediated entirely by viral proteins [29]. The genes expressed early in infection include those involved in immune evasion, and also those required for DNA replication and intermediate protein synthesis [11]. DNA replication along with later stages in the life cycle takes place in virus factories, which are discrete partially membrane-bound cytoplasmic foci of replication in the perinuclear region of infected cells [30]. Establishment of these factories involves many changes to the cellular structure. ER-derived membrane is incorporated into the factories, and virus particles traffic along microtubules to reach the perinuclear location of the factories

[30,31,32,33]. This process may be assisted by induction of cellular motility and contractility to collect microtubules and endosomes near the factories, and thus involves both the actin and microtubule cytoskeletal networks [31,32,34]. DNA replication begins

1-2 hours after infection and produces about 10,000 genome copies per cell connected in a head-to-head, and tail-to-tail arrangement of concatamers that must be cleaved by the viral resolvase in the late stages of infection to produce single genome units for packaging [11,35]. A number of virally-encoded factors have important roles in DNA replication including the DNA polymerase, nucleoside triphosphatase, DNA glycosylase, and processivity factors [11,36]. The viral B1 kinase is also required for DNA replication, at least partially through inhibiting a cellular protein BAF that normally binds cytoplasmic

DNA and blocks replication [37]. Unknown cellular factors also play important roles in successful DNA replication, since this step in the lifecycle is frequently blocked in abortive cell lines [38,39].

Following DNA replication viral gene expression shifts from early to intermediate, and then late gene products. Several intermediate genes are required for late gene expression, maintaining the lifecycle’s temporal progression. Proteins expressed late consist of virion structural components and proteins required for early transcription [29].

In addition to virally encoded factors, host factors are required for intermediate and late synthesis [40,41,42]. The putative RNA binding protein G3BP and/or p137 are host proteins required for intermediate expression [43]. A nuclear transcription factor,

YinYang (YY1) was shown to be recruited out of the nucleus and bind to an intermediate promoter and may have a repressive role [44]. Several nuclear riboproteins A2/B1 and

RBM3 play redundant roles in late transcription, and may play an important role in targeting the replication machinery to the late promoters [45]. Finally, the TATA binding protein (TBP) has also been shown to bind intermediate and late promoters and has an essential role in transcription at these stages [46].

Once these stages of viral replication are completed, assembly of new virions begins. This process includes several stages of morphogenesis within the viral replication factories before mature virions are produced, a process that involves the trans-Golgi membrane protein golgin-97 [47,48,49]. The maturation process involves cresent formation, genome encapsidation, and membrane acquisition. Mature virus

(MV), the first infectious form of the virus, consists of the virus core, with the DNA genome and early replication machinery, surrounded by a single lipid bilayer envelope and associated proteins. These virions then have several fates. MV can remain in the cytoplasm of the infected cell until it is released by cell lysis [47]. Alternatively, MV particles can use microtubules for transport to the golgi, where they undergo wrapping and obtain additional sets of Golgi-derived lipid bilayer envelopes [50]. These

membranes are studded with a different set of vaccinia proteins that are trafficked to the

Golgi during replication. Once wrapping is complete, these virions are transported on

microtubules to the cell surface, where they fuse with the cell membrane in an actin-

dependent manner, leaving behind the outer-most Golgi-derived envelope [51,52,53].

This Extracellular Virus (EV) is a second mature and infectious form of the virus, which

can dissociate from the cell to facilitate virus spread. Alternatively, EV can remain

associated with the cell (CEV) and induce Abl family kinase dependent signaling events

that induce actin tail formation to propel the virion away from the infected cell, and aid in

its spread to adjoining cells [51,54,55,56]. In addition to facilitating spread of newly made

EV particles, generation of actin tails can accelerate infection by directing other incoming

EV particles to surrounding uninfected cells [57]. The different forms of virus are

believed to have different roles during infection, with the more stable MV particles

responsible for spread of the virus from host to host, while the EV and CEV forms are

important in dissemination within a host [51].

b. Vaccinia Entry and Macropinocytosis

Entry of enveloped viruses such as vaccinia typically occurs in four steps: virus attachment, trafficking to an appropriate compartment, activation of fusion proteins, and membrane fusion releasing the virus core into the cytoplasm [58]. Attachment may occur by several binding events. Evidence suggests that virion binding to cell surface glycosaminoglycans can facilitate attachment [59]. Certain vaccinia proteins can also bind to cellular chondroitin sulfate and heparin sulfate, which may actually inhibit infection, but it is unclear what role if any these binding events have on vaccinia entry

[59,60,61,62].

Many of the viral proteins involved in vaccinia entry have been recently characterized. The vaccinia entry fusion complex is made up of at least 8 proteins, but a number of others also play a role in entry [63,64,65,66,67,68,69,70,71,72,73,74,75,76].

These proteins are expressed late in infection, and are present on the MV membrane.

Notably, both MV and EV forms of the virus use the same entry fusion complex for virus entry, but may have different entry kinetics [77,78]. Since all forms of the virus use the same entry-fusion complex, located on the inner MV membrane, this poses an interesting problem for EV and CEV virions and necessitates the removal of the outer membrane before fusion can occur. The EV membrane is fragile and actively broken upon cellular contact. EM images suggest that the EV membrane breaks, but remains associated with the virion on the cell surface, providing a sheath that may hide and protect the virus particle during entry [79]. Studies have also suggested that EV entry requires low pH, presumably because acidic conditions are needed to disrupt the outer membrane, but studies have also indicated that MV entry requires low pH, so it is unclear if acidity plays a specific role in disruption of the EV membrane [59,80,81].

The location and cellular requirements of membrane fusion and entry are controversial. Electron microscopy studies have shown virus particles fusing at the plasma membrane, but it was unclear if this was a productive form of entry [59]. Early drug inhibitor studies indicated that actin inhibitors as well as drugs that blocked acidification of endosomes were able to block the majority of infection, which led to the belief that at least two entry mechanisms existed: fusion at the plasma membrane, or pH-dependent endosomal entry [77,78,82]. Since the virus is excluded from many routes of endosomal entry by its large size, and it is an actin dependent process, macropinocytosis was believed to be the mechanism of endosomal entry [77,83,84]. It

8 has become clear that there are multiple routes for vaccinia entry that are both virus strain and cell type specific. Vaccinia strains WR and Wyeth seem to enter predominantely using a low pH endosomal mechanism in BSC-1 cells, likely macropinocytosis, while IHDJ, Copenhagen, and Elstree are less sensitive to pH, indicating that the predominant mechanism of entry is likely fusion at the plasma membrane, or a pH-independent endosomal fusion mechanism [85]. Furthermore, macropinocytosis seems not to be the the entry mechanism in all cell types. Both WR and IHDJ entry into Chinese Hamster Ovary (CHO) cells is resistant to inhibitors of Na/H exchangers, Pak1, and PI3 kinases that are indicative of macropinocytosis [84]. Even entry of the same vaccinia strain into different cell types is variable with respect to low pH dependence, entry kinetics, and inhibition by heparin [86]. Therefore, additional mechanisms of entry exist, and there are many additional cellular factors remaining to be identified for this complicated uptake mechanism and thus, for vaccinia entry.

Recent studies in HeLa cells have demonstrated that macropinocytosis is an important endocytic route of vaccinia entry [83]. Generally, macropinocytosis is a nonselective route for bulk fluid-phase uptake and is not constitutively active, but is induced by growth factors, and also by some pathogens including vaccinia [87,88]. This active endocytic process induces extensive actin cytoskeletal rearrangement, leading to membrane ruffling, lamellipodia formation, and the internalization of extracellular fluid and membrane. Consistent with this, vaccinia entry is dependent upon modulation of the actin cytoskeleton, and initiates macropinocytosis by inducing dramatic actin-rich microvilli protrusions followed by global myosin II-dependent blebbing, thereby promoting virus uptake [78,83]. Induction triggers the activation of receptor tyrosine kinases (RTKs) such as epidermal growth factor receptor (EGFR) whose activation is

9 crucial for vaccinia entry and this kinase may serve as an entry receptor. This induction activates complex signaling cascades leading to the initiation of these actin extensions which extend the plasma membrane allowing fluid-phase capture. This process involves signaling cascades that converge on members of the Ras superfamily of GTPases.

Interestingly, multiple strains of vaccinia seem to enter HeLa cells by macropinocytosis, but the specific actin rearrangements and signaling requirements vary in different strains. While WR induces transient Rac1-dependent blebbing of the actin cytoskeleton,

IDHJ induces filipodia formation dependent on Cdc42, suggesting multiple pathways of macropinocytosis exist and are exploited by different poxviruses [88,89]. Rac1 and

Cdc42 contribute to a number of cellular processes that require extensive actin dynamics, and signaling is carefully regulated by several guanine exchange factors as well as by crosstalk with other Rho family GTPases [90,91,92]. Additional kinases such as Pak1 are then activated along with actin-associated proteins that ultimately lead to large-scale actin rearrangements, lipid modifications, and macropinosome formation

[83,87]. While some specific kinase families have been implicated in macropinocytosis for both vaccinia strains (e.g., protein kinase C (PKC), serine/threonine kinases, tyrosine kinases, and phosphatidylinositol kinases) [87], many of the specific factors have not been identified, and in some cases the specific role of factors such as PKC, is not well understood. Chapter III will discuss a novel role for the host factor AMPK in vaccinia entry by macropinocytosis.

4. Arboviruses

Arthropod-borne viruses or Arboviruses include viruses that are transmitted by arthropod vectors, particularly biting insects. Arboviruses transit through at least two

10 hosts to complete their lifecycle. One host is a vertebrate, usually a bird or small mammal, and the second host is an arthropod vector, usually a mosquito. These viruses are maintained in an enzootic cycle with little mortality. However, if other hosts are infected, such as humans or livestock epizootic outbreaks can lead to high mortality. In some cases, the mortality occurs in dead-end hosts that cannont function as a reservoir for re-infection of mosquitoes, and generally do not transmit the virus to other hosts [10].

The arboviruses infecting both humans and domestic animals have led to significant world-wide morbidity and mortality, and are both a medical and agricultural concern. Most arboviruses impacting human health fall into one of three viral families:

Flaviviridae, Togaviridae, and Bunyaviridae. Flaviviruses include medically important members such as Yellow Fever Virus, Dengue Virus, and West Nile Virus that are present in both temperate and tropical climates. West Nile Virus was originally identified in Africa, and until recently was limited to the Old World. However in 1999, there was an outbreak in New York that enabled the virus to spread across North America and into

Central America and the Caribbean Islands [93,94]. The case of West Nile Virus demonstrates the potential ability of these arboviruses to spread to new geographical areas and why they represent a public health concern even in non-endemic areas. The

Alphavirus genus within Togaviridae includes New World viruses that cause encephalitis in humans and equines, as well as Old World viruses that cause febrile and arthralgic illnesses, while Bunyaviridae includes members that cause haemorrhagic fever

[95,96,97]. The Bunyavirus Rift Valley Fever Virus (RVFV) is the focus of Chapter IV.

The Togavirus Sindbis and the Rhabdovirus Vesicular Stomatitis virus will also briefly be mentioned.

11 a. Rift Valley Fever Virus

Bunyaviruses are arboviruses that include several disease causing members such as Sin Nombre, Hantavirus, Crimean-Congo hemorrhagic fever virus, and Rift

Valley Fever virus (RVFV). RVFV is a mosquito borne Category A agent initially endemic to sub-Saharan Africa. The virus was first identified in 1931 during an investigation of a sheep epidemic in the Rift Valley of Kenya. Since then, outbreaks have been reported in sub-Saharan and northern Africa, including a major outbreak in 1997-

98 in Kenya, Somalia, and Tanzania [98]. However, recent outbreaks of RVFV have occurred in Egypt and the Arabian Peninsula [99,100,101], indicating the potential of this virus to spread to new geographical areas, especially since a broad range of mosquito species have been linked to RVFV transmission [98].

RVFV has particular importance as an agricultural pathogen, where infection of livestock can lead to significant morbidity and mortality, particularly among young animals. Infection leads to nearly 100% abortion among pregnant livestock, and over

90% of infected lambs die from RVFV. However, mortality among adult sheep can be as low as 10% [98]. The majority of human infections result from direct or indirect contact with the blood or organs of infected animals, ingesting unpasteurized milk from an infected animal, or from mosquito transmission [98,102,103]. Most humans infected with

RVFV develop a self-limited febrile illness with symptoms similar to meningitis, although approximately 1-3% die from the disease due to hemorrhagic symptoms, and in recent outbreaks the mortality rate in humans was reported to be as high as 45%

[104,105,106,107,108]. The severity of RVFV zoonosis, along with its capacity to cause major epidemics among livestock and humans, and the lack of efficient prophylactic and

12 therapeutic measures make this pathogen a serious public health threat in endemic as well as non-endemic countries. RVFV is therefore classified as a BSL-3 or BSL3+/4 agent in Europe and the U.S. respectively, and is considered a potential bioterrorist agent [107].

Vaccines have been developed for veterinary use, including a live-attenuated vaccine MP-12, as well as inactivated virus vaccine [109,110,111]. The live vaccine provides long term immunity with only one dose, but can cause spontaneous abortion in pregnant animals, while the inactive vaccine requires multiple doses to provide protection, which may be problematic during outbreaks in endemic areas [112,113].

Moreover, vaccination during an outbreak is not recommended since it may actual intensify the outbreak. During mass animal vacciniation campaigns the use of multi-dose vials and the re-use of needles and syringes may allow further transmission of the virus throughout a herd. In humans, these vaccines can provide protective immunity; however, they are expensive to produce, and in the case of MP-12 may not be acceptable in areas where RVFV is not endemic [114,115]. No effective anti-viral therapies have yet been developed against RVFV.

RVFV has a single-stranded tripartite genome that encodes six major proteins.

The L and M segments have negative polarity, while the S segment is ambisense. The L segment encodes L, the viral RNA-dependent RNA polymerase. The M segment encodes two nonstructural proteins M segment-derived non-structural protein (NSm) of

78 kDa (NSm1) and 14 kDa (NSm2), as well as the precursor of the two structural glycoproteins: Gn (glycoprotein located at the N-terminus of the precursor molecule), and Gc (glycoprotein located at the C-terminus of the precursor molecule) [116,117].

The S segment encodes two proteins. In the positive sense orientation S encodes S segment-derived non-structural protein (NSs), while in the negative sense orientation S encodes the nucleoprotein N [117]. The viral envelope is composed of a lipid bilayer containing the Gn and Gc glycoproteins that form regularly arranged surface subunits that cover the surface in a shell of 122 glycoprotein capsomers arranged in an icosahedral lattice [118,119,120]. The viral ribonucleoproteins are composed of the three genomic segments associated with numerous copies of N and L that are packaged in the virion through a strong interaction with the cytoplasmic tail of the glycoproteins

[121,122]. The nonstructural proteins are dispensable for viral infection in cell culture, but play important roles in pathogenesis in vivo [123,124,125,126].

The lifecycle of RVFV involves attachment and binding to as yet unidentified receptor, followed by membrane fusion and entry through a low-pH dependent mechanism mediated by the envelope glycoproteins and dependent on cellular PKCε

[117,127,128]. The viral ribonucleocapsid is released into the cytoplasm and is transcribed into either mRNA for protein transcription or antigenomic cRNA for use in genome replication [117]. Replication occurs on modified perinuclear Golgi-derived tubular vesicle structures present in close association with mitochondria [129]. The cRNA present in the input virus serves as the template for synthesis of the NSs so that this virulence factor can be expressed immediately after the virus has entered the host cell [130]. Bunyaviral mRNA synthesis is initiated by a cap-snatching mechanism, where host mRNAs are cleaved through the endonuclease activity of the L protein to obtain 10-

18 nucleotide primers for transcription [131,132,133,134,135]. Synthesis of cRNA and vRNA does not require cap-snatching, and is initiated directly with 5’ nucleoside triphosphates. cRNA represents a complete copy of the viral genome, while mRNAs 14 terminate prior to the 5’ end of the template; however the mechanism of the switch between mRNA transcription and cRNA replication is unclear. The S and L segment mRNAs are transcribed on free ribosomes, while the M segment mRNA is translated on the rough ER [136,137]. The glycoproteins produced off of M are then cleaved from a common precursor molecule and migrate to the Golgi where they mature [137]. During assembly, the ribonucleocapsid, including N and L, associate with the glycoproteins and bud into the Golgi in vesicles [117]. Finally, these vesicles fuse with the plasma membrane allowing viral release.

Little is known about the host factors that contribute to RVFV infection in insects or mammals, although the interferon-inducible MxA has been shown with virus like particles to inhibit RVFV transcription, and the RVFV NSs protein has been characterized as a major virulence factor to counteract the antiviral interferon system by suppressing interferon induction and inhibiting the interferon-inducible antiviral protein

PKR [138,139,140,141]. Furthermore, mice incapable of mounting a robust type I interferon response to RVFV were more susceptible to infection, demonstrating the importance of innate immune defense in controlling this virus [142]. b. Sindbis Virus

Togaviruses are subdivided into two genera, the Alphaviruses and Rubiviruses

[143]. In Chapter IV I will briefly discuss the role of AMPK in infection of the prototypical

Alphavirus Sindbis. The Alphavirus group includes many medically and agriculturally important viruses including Eastern equine encephalitis virus, Western equine encephalitis virus, Venezuelan equine encephalitis virus, Ross River virus, and

Chikungunya virus in addition to Sindbis [144]. Sindbis was first identified in 1952 in

Cairo, Egypt, and is most common in southern and eastern Africa, Egypt, Israel,

Philippines, Australia, and northern Europe, making it the most widely distributed of the

Alphaviruses that cause arthritis in humans [145]. The primary sites of Sindbis-mediated disease in humans are in northern Europe, where Sindbis causes Ockelbo, Pogosta fever, and Karelian fever in Sweden, Finland, and Russia respectively, and in South

Africa. In other areas Sindbis causes a subclinical disease that manifests in fever, rash, or mild arthralgias. Sindbis Virus is transmitted by mosquito and maintained in nature by transmission between avian and mosquito vectors, with humans considered dead-end hosts [145].

Alphaviruses are icosahedral enveloped arboviruses with a linear single-stranded positive sense RNA genome approximately 11.5 kb in length containing a 5’ methylated nucleotide cap and a 3’ polyadenylated tail that resembles cellular mRNA [144]. Sindbis encodes four nonstructural proteins involved in virus replication and pathogenesis, and five structural proteins that compose the virion [146]. The envelope consists of a host- derived lipid bilayer embedded with 240 copies of E1 and E2 heterodimer glycoproteins further modified by palmitoylation, and smaller amounts of the K6 membrane-associated protein [146,147,148,149]. During entry, E2 is involved in receptor binding and subsequent clathrin-mediated endocytosis, while E1 is responsible for pH-dependent fusion of the viral membrane with the endosomal membrane [144]. The nucleocapsid enters the cytoplasm and dissociates allowing translation and genome replication.

Replication occurs in cytopathic vacuoles derived from lysosomal and endosomal membranes [150]. Translation of genomic RNA yields two polyproteins: P123 or the larger P1234. The larger form occurs through translational readthrough of a termination codon that occurs 10-20% of the time. Thus the predominant product is P123 [151]. The

16 resultant polyproteins are cleaved by the virally encoded protease activity in the nonstructural protein nsP2 [152]. Both fully cleaved products and cleavage intermediates have been shown to play a role in virus replication [144]. After cleavage, structural proteins are translocated into the ER for processing and post-translational modifications that include the addition of high-mannose sugars and palmitoylation [153,154,155].

Proteins are then transported through the Golgi and to the plasma membrane where interactions between the nucleocapsid and glycoproteins occur, leading to virion assembly and ultimately budding from the plasma membrane [156,157,158]. A striking feature of alphavirus infection in vertebrate cells is the ability of the virus to shut down host transcription and translation without affecting viral synthesis. These shutdowns result in a decrease in type I interferon production, attenuating the innate immune response [144]. c. Vesicular Stomatitis Virus

The Rhabodovirus family includes Rabies virus, but the prototypical and best studied member is Vesicular Stomatitis Virus (VSV), which is discussed briefly in

Chapter IV. VSV represents a widely studied prototype for the non-segmented, negative- sense RNA viruses [159]. VSV infections cycle through cytolytic infection of mammalian hosts, particularly cattle, and persistent non-cytolytic infection and transmission by insects, such as sandflies. In cattle, horses, and pigs VSV produces a disease characterized by fever, and the formation of vesicles in the oral mucosa and teat, and is primarily of concern because it is similar in appearance to the catastrophic foot and mouth disease [159].

Rhadboviruses are rod or bullet shaped negative sense single stranded RNA viruses that typically encode 5 proteins: large protein (L), glycoprotein (G), nucleoprotein

(N), phosphoprotein (P), and matrix protein (M) . The VSV-G envelope protein enables virus entry through pH-dependent clathrin-mediated endocytosis [160]. After low-pH fusion, the ribonucleocapsid is released into the cytoplasm, where it undergoes genome replication and protein synthesis. Viral P and L proteins comprise the core components of the viral RNA-dependent RNA polymerase, which is responsible for viral transcription

Ĩ 櫂 template for primary transcription by the virion RNA-dependent RNA polymerase, resulting in synthesis of leader RNA and all five monocistronic viral mRNAs that become processed to have a 5’ cap and a 3’ polyadenylated tail, and thus resemble cellular mRNAs that can be translated by ribosomes [159,161]. Viral proteins accumulate, leading to genome replication. A full-length positive strand RNA antigenome is synthesized and serves as the template for synthesis of progeny negative-strand genomes. The N protein encapsidates the RNA genome into a helical RNase-resistant core, while P acts as a chaperone to maintain the N protein in a soluble for in the cytoplasm before association with the genome [161]. The virus assembles by association of the ribonucleoprotein-matrix preassembly complexes with the G protein at the inner leaflet of the plasma membrane, and releases by budding from the plasma membrane [161]. d. Lipid Metabolism in Arboviral Infection

Lipid metabolism plays an important role in infection of many viruses that replicate in the cytoplasm including RVFV, Sindbis, and VSV. An essential feature of

18 these infections is the ability of viruses to rearrange and proliferate internal cellular membranes into distinct invaginations or vesicle-like structures to compartmentalize the viral replication complex and support genome replication [162]. These membranous structures may serve to closely coordinate the replication complexes with virion assembly and budding, act as a scaffold for the replication complex to tether viral RNA, increase the local concentration of replication components, or prevent detection by the host immune defense system [162]. Depending on the virus, these membrane modifications can be derived from distinct cellular sources, including ER, Golgi, endosomal, and mitochondrial membranes, and may have complex biogenesis derived from multiple cellular origins [150,163,164,165,166,167,168,169,170]. Bunyavirus replication occurs on modified perinuclear Golgi-derived tubular vesicle structures present in close association with mitochondria, while Alphaviruses modify lysosomal and endosomal membranes into cytopathic vacuoles to support their replication complex and

Rhabdoviruses use cytoplasmic inclusion bodies [129,150]. Despite these differences, there are many similarities in the membrane structures induced by different types of viruses suggesting commonalities in this induction across multiple virus families [171]. In some cases the viral proteins responsible for these changes are known, but host factors are also likely to contribute to this process. Importantly these modifications are highly dependent on cellular lipid biogenesis pathways and require generation of newly synthesized cellular lipids [162].

5. Cellular Energy Metabolism

Cellular energy metabolism is tightly regulated to allow organisms to balance energy requirements to maintain life. In order for growth and reproduction to occur, cells

19 and organisms must construct components such as proteins, lipids, and nucleic acids in anabolic processes that require the use of energy. In order to build these required components, energy must be harvested from the uptake and catabolic breakdown of energy-rich materials such as glucose, and stored in molecules such as ATP or GTP in a form the cell can use [172]. Energy is harvested and utilized in numerous complex cellular processes that must be carefully regulated so that energy consumption does not exceed availability. During the course of life, an organism or cell experiences many changes in energy demands and availability that can result from changes in activity level, or stresses such as glucose deprivation and starvation, hypoxia, ischemia, and oxidative or osmotic stress. Therefore, the success and survival of the cell depends on its ability to monitor cellular energy and respond to environmental stresses to maintain energy homeostasis.

AMP-activated Kinase (AMPK) is an important regulator of cellular metabolism both on the cellular and organismal level. Once activated this complex switches the cell from an anabolic to a catabolic state by activating oxidative pathways that generate energy while inhibiting synthesis and growth pathways in order to return the cell to a state of energy homeostasis [173]. AMPK is conserved in all sequenced eukaryotes, and a null mutation results in embryonic lethality possibly due to developmental roles in establishing polarity [174,175]. AMPK is a heterotrimeric complex that consists of a catalytic alpha subunit as well as beta and gamma regulatory subunits [173]. Drosophila encode only one gene for each subunit whereas more redundancy is seen in mammalian systems with multiple isoforms of each subunit encoded by several distinct genes (α1, α2, β1, β2, γ1, γ2, γ3) that can produce at least 12 possible heterotrimeric combinations [176]. In mice, no phenotype is associated with deletion of the α1 gene,

20 while deletion of the α2 is associated with metabolic phenotypes, indicating partial functional redundancy of the two catalytic subunits [177,178,179]. While expression patterns of individual subunits vary, the precise function of each subunit combination remains unclear.

AMPK is activated in response to an increase in cellular AMP or ADP compared to ATP [173,180]. Bateman domains of the regulatory gamma subunit are exchangeable competitive binding sites for AMP, ADP, or ATP. Activation is triggered through binding of AMP or ADP to these binding sites, leading to increased phosphorylation of the threonine 172 on the alpha subunit [180,181,182]. Binding of two AMP inhibits dephosphorylation of AMPK and also promotes additional allosteric activation, while tight binding of a single ADP protects AMPK from dephosphorylation without allosteric activation [180,182,183]. Evidence suggests that AMPK may be most strongly regulated by dephosphorylation [183,184]. The canonical upstream activator of AMPK that catalyzes this phosphorylation event is the constitutively active tumor suppressor LKB1

[185,186]. Other upstream activators have also been identified that may function independently of energy levels. Calmodulin-dependent protein kinase kinase beta

(CaMKKβ) has been reported to phosphorylate AMPK on the Thr172 in a calcium- dependent manner [187,188]. Additional upstream activators have been identified in yeast. In addition to Tos3p, the yeast homolog of LKB1, Sak1 and Elm1p have been shown to activate AMPK in vivo and in vitro in yeast, suggesting the possibility of additional activators in other systems as well [189]. Finally, TGFβ-activated kinase

(TAK1) has also been implicated to activate both yeast and mammalian AMPK in overexpression studies [190]. AMPK activation by TAK1 has been shown to induce autophagy, which has been implicated in innate immunity [191,192]. While unclear if this 21 activation is relevant in vivo, TAK1 is involved in innate immune signaling, and may provide a link between AMPK signaling and innate immune defense [193,194,195,196].

In addition to cellular control, AMPK is also regulated by cytokines that control whole-body energy balance, such as leptin, adiponectin, ghrelin, cannabinoids, interleukin-6, and ciliary neurotrophic factor (CNTF) [176,197,198]. In particular, AMPK is regulated by cytokines released by adipocytes including leptin, adiponectin, and resistin. These cytokines can have different effects on different tissues. In skeletal muscle, AMPK activation by leptin and adiponectin increased glucose uptake and fatty acid oxidation. In the liver, adiponectin stimulates fatty acid oxidation and inhibits glucose production, while resistin has the opposite effect [176]. AMPK activation in the brain may regulate food intake. In mice, activation of AMPK in the hypothalamus stimulates food intake, while treatment with anorexigenic agents such as insulin inhibit

AMPK and lead to a repression of food intake even in fasted mice [199]. Thus in addition to regulating cellular energy, AMPK is an important regulator of whole-body energy balance, linking it to obesity through food intake and energy expenditure [200].

6. Downstream Functions of AMPK

AMPK activation stimulates ATP-producing catabolic pathways, while inhibiting

ATP-consuming anabolic pathways rapidly through phosphorylation, and also more slowly through transcriptional regulation [173]. This activation has a number of important downstream effects. AMPK activation increases translocation of the glucose transporter

GLUT4 to the plasma membrane, thus stimulating glucose uptake in a manner somewhat redundant with insulin activation of Akt signaling [201,202,203]. At the same time, activity and expression of many biosynthetic genes and pathways are

22 downregulated by AMPK. These include protein targets involved in lipid metabolism, carbohydrate metabolism, protein metabolism, and transcription. AMPK activation also has effects on cell signaling, ion transport, and polarity given this complex control over many aspects of cell biology [173]. Several of the downstream effectors of AMPK target mechanisms important for viral infection, and are further discussed below.

a. AMPK Regulation of Protein Synthesis

AMPK activity inhibits global protein translation by multiple signaling mechanisms

(Figure 1A). AMPK inhibits the mTOR Complex 1 (mTORC1) which is an important activator of protein synthesis through two main downstream effectors [204]. mTORC1 phosphorylates p70-S6 Kinase 1 (S6K1) on multiple residues with the most critical modification occurring on the Thr-389 [205,206]. This phosphorylation event stimulates the phosphorylation and activation of S6K1. S6K1 activates the 40 S ribosomal protein

S6 and other components of the translation machinery, leading to increased translation initiation [207]. Additionally, mTORC1 has been shown to phosphorylate at least four residues of 4E-BP1 in a hierarchical manner [208]. In a non-phosphorylated state, 4E-

BP1 inhibits the translation initiation factor eIF4E by tightly binding it and preventing it from binding to 5’capped mRNAs and recruiting them to the ribosomal initiation complex.

Phosphorylation of 4E-BP1 by mTORC1 releases this binding and allows eIF4E activity, thus promoting translation initiation [208].

mTORC1 is regulated by the upstream tuberous sclerosis complex (TSC) that receives inputs from multiple upstream signaling pathways, including AMPK [204,209].

These kinase-mediated signaling pathways alter the ability of TSC1/TSC2 to function as a GTPase activating protein (GAP) towards Rheb, the major regulator of mTORC1

[210,211,212,213,214]. Rheb is active in its GTP-bound form, and is then able to activate mTORC1. TSC1/TSC2 activity therefore inhibits the activity of Rheb in unstimulated conditions [209]. Several pathways are able to alter the activity of

TSC1/TSC2. AMPK phosphorylates TSC2 at 2 sites (Ser 1270 and Ser 1388) which enhances the ability of the TSC1/TSC2 complex to act as a GAP, and blocks Rheb- dependent mTORC1 activation leading to an inhibiton of protein synthesis [215]. In contrast, Akt is another major regulator of the mTORC1 pathway that becomes activated in response to growth factors such as insulin [216,217]. Akt phosphorylates TSC2 on at least five sites distinct from those targeted by AMPK, inhibiting the activity of TSC2 and thus activating mTORC1 and protein synthesis [209]. Finally TSC1 can also be phosphorylated at distinct sites by members of the MAPK pathway to inhibit TSC1/TSC2 activity, and thus activate mTORC1 [218,219]. Therefore, regulation of protein synthesis downstream of mTORC1 is complex, and while AMPK is an important regulator, other upstream pathways also contribute to regulating translation initiation.

In addition to regulating translation initiation through mTORC1, AMPK also regulates translation elongation through Elongation Factor 2 (eEF2) [220,221]. eEF2 is the phosphoprotein that mediates the translocation step of peptide chain elongation during protein translation [222]. Phosphorylation at the Thr-56 residue inactives the protein by preventing ribosome binding [223,224]. This phosphorylation event is catalyzed by the calcium/calmodulin-dependent protein kinase eEF2 kinase, which, like mTORC1, can be regulated through multiple upstream pathways [222]. AMPK directly phosphorylates eEF2 kinase on several serine residues, with Ser-398 being its major target [221]. This phosphorylation stimulates eEF2K activity, which inhibits eEF2 and elongation. In contrast, eEF2K is inactivated by phosphorylation at Ser-366 by S6K 24 downstream of mTORC1 or by p90RSK, resulting in increased translation [225].

Likewise phosphorylation by p38 MAPK inhibits eEF2K activity [226]. Therefore, protein synthesis mediated by both mTORC1 and eEF2 are regulated by multiple pathways with considerable crosstalk, and other signaling events besides AMPK may contribute to regulation of protein synthesis during viral infection. b. AMPK Regulation of Autophagy

Autophagy is a catabolic process by which cells break down their own components by engulfing a portion of the cytoplasm in an isolation membrane, and then delivering the contents to lysosomes for degradation [192]. This process is an important repair mechanism for degrading damaged organelles, membranes, and proteins, and contributes to survival during periods of nutrient starvation by recycling non-vital components [227]. Evidence also suggests that autophagy is involved in innate immune defense through destruction of intercellular pathogens, and plays a crucial role in resistance to bacterial, viral, and protozoan infections [192].Like protein synthesis, autophagy is dually regulated by AMPK and Akt downstream of mTORC1 [227].

Activation of mTORC1 leads to inhibition of autophagy. Therefore, since AMPK activity leads to a decrease in mTORC1 activity as described above, AMPK activation is associated with a cooresponding increase in autophagy, while Akt activity leads to an increase in mTORC1 activity and decreased autophagy (Figure 1A). Thus AMPK activity could interefere with viral infection through autophagy mediated restriction.

25 c. AMPK Regulation of Fatty Acid Synthesis

AMPK is an important regulator of lipid metabolism, inhibiting both sterol and fatty acid synthesis, while promoting fatty acid oxidation (Figure 1B) [228].

Phosphorylation downstream of AMPK inactivates acetyl CoA carboxylase (ACC), a rate-limiting enzyme in fatty acid metabolism [229,230]. ACC catalyzes the irreversible conversion of acetyl-CoA to malonyl-CoA, the substrate for fatty acid synthesis, which drives de novo production of palmitate [231]. In addition to this important role in fatty acid synthesis, malonyl-CoA is a cosubstrate for the fatty acid elongation system located in the membrane of the endoplasmic reticulum [232]. This system allows chain lengthening of both endogenously synthesized and dietary-derived essential fatty acids into higher polyunsaturated fatty acids. Finally, malonyl-CoA serves a signaling function through inhibition of carnitine palmitoyltransferase I (CPT-1), an essential factor involved in the transport of fatty acids to the mitochondria for beta oxidation [231]. Thus malonyl-CoA production by ACC promotes fatty acid synthesis, while inhibiting fatty acid oxidation.

Mammalian systems encode two non-redundant ACC isoforms, ACC1 and ACC2, which are both targeted by AMPK for phosphorylation and inactivation [228]. In vitro studies revealed that AMPK could phosphorylate ACC1 on multiple serine residues, but phosphorylation of Ser79 occurs most rapidly, and is associated with inactivation

[229,230,233]. While other protein kinases have been shown to phosphorylate ACC1 in vitro, evidence of physiological relevance exists only for AMPK [229,234]. Studies suggest that malonyl-CoA produced by ACC2 is exclusively involved in fatty acid oxidation, while ACC1 contributes to fatty acid synthesis [228]. Therefore, activation of

AMPK through stress or low energy conditions induces fatty acid oxidation through

ACC2, while inhibiting fatty acid synthesis through ACC1 and could impact the ability of

viruses to subvert lipid metabolism for their replication.

d. AMPK Regulation of the Cytoskeleton

While the best understood role of AMPK is its role in metabolism, recent

evidence suggests this kinase also has a crucial role in regulating cell structure and

polarity through engagement with the actin cytoskeleton [235]. In Drosophila, loss of

AMPK leads to defects in mitotic division and epithelial cell polarity accompanied by disruption of the actin cytoskeleton [175]. In some mammalian epithelial cell lines, AMPK activation leads to polarization characterized by the formation of an actin brush-border or tight junction assembly [175,236,237,238]. Additionally, AMPK activation can induce astrocyte stellation, and actin stress fiber disassembly [239], and studies using

Compound C and AMPK activators have linked AMPK to macropinocytic uptake of albumin in murine macrophages [240]. Finally, recent evidence suggests that AMPK regulates microtubules as well as actin. AMPK phosphorylates the microtubule plus end protein CLIP-170, and knock down of AMPK impairs microtubule stabilization and directional cell migration [241]. I provide further evidence in Chapter III that AMPK is crucial in promoting actin-driven macropinocytosis and membrane ruffling as well as cell migration, and this modulation of the actin cytoskeleton is important in promoting vaccinia entry. This accumulating evidence suggests a broad and conserved role for

AMPK in a variety of cellular processes that require cytoskeletal rearrangements.

The precise signaling events that lead to these AMPK-dependent cytoskeletal changes remain unclear. Several upstream kinases have been shown to activate AMPK under different stimuli. The best studied of these is the tumor suppressor LKB1 which

27 activates AMPK in response to changes in cellular energy levels [185,186]. Additionally,

AMPK can be activated in response to changes in intracellular calcium levels by

CaMKKβ, and further evidence suggests that TGFβ-activating kinase (TAK1) may serve as a third upstream activator [187,188,190]. Previous studies demonstrated that LKB1 is an important mediator of cell polarity at least in part through signaling to AMPK, and has been shown to drive actin brush border formation, and the translocation of apical and basal markers during the establishment of polarity [175,236,237,242,243], but how this process is fine-tuned and regulated in a physiologically relevant context remains to be determined.

7. The Role of AMPK in Metabolic Diseases

Since AMPK is an important regulator of energy metabolism, it has been implicated in type II diabetes, obesity, and related metabolic diseases [244,245]. AMPK activation has pleiotrophic effects in a number of metabolically relevant tissues, such as liver, skeletal muscle, adipose, and hypothalamus [245]. During type II diabetes, liver, skeletal muscle, and adipose tissue can become insulin resistant. This state of insulin resistance leads to a dramatic decrease in glucose uptake and utilization, reduced mitochondrial capacity, and defects in fatty acid oxidation that can ultimately interfere in the ability of skeletal muscle to switch between glucose and fatty acid metabolism [246].

AMPK activation is able to counteract some of the effects of insulin resistance. Notably, many of the metabolic changes induced by AMPK would be desirable in the treatment of type II diabetes and metabolic syndrome. These changes include increased glucose uptake and mitochondrial biogenesis, increased fatty acid oxidation in muscle and liver, and decreased fatty acid synthesis in liver and adipose tissue

[247,248,249,250,251,252,253]. Indeed, drugs that activate AMPK have been shown in rats and humans to alleviate some of the metabolic symptoms associated with diabetes

[254,255,256,257,258,259,260,261,262].

Two classes of drugs used clinically to treat type II diabetes activate AMPK, and may function at least in part through this mechanism. One class of AMPK activating antidiabetic drugs is the biguanides, including Metformin and Phenformin. Metformin has been in use since 1957 and is the most widely used drug to treat type II diabetes.

However, it is a low-potency compound that is used at high doses with only moderate efficacy and can have significant side effects. It reduces hyperglycermia without stimulating insulin secretion, promoting weight gain, or causing hypoglycermia [263,264].

Decreased hepatic glucose production and increased skeletal myocyte glucose uptake are believed to be major contributors to Metformin’s glucose lowering effects, and likely results from the AMPK activating effects of drug treatment [258,263,265,266,267].

Metformin activates AMPK indirectly, and likely has additional cellular targets [262,268].

For example, studies indicate that Metformin inhibits mTORC1 independent of AMPK and reduces circulating levels of insulin and insulin-like growth factor [269].

In addition to treatment of type II diabetes, AMPK activation may have other beneficial effects on health. AMPK has been linked to longevity in model organisms such as flies and worms, and may contribute to the beneficial effects of exercise [270,271].

AMPK activation also inhibits cell proliferation and a growing body of evidence suggests that AMPK activating drugs such as Metformin can reduce the risk of cancer

[272,273,274,275]. An observational study in 2005 found that diabetics taking Metformin had a 40% lower risk of a variety of cancers than those taking other diabetic drugs [276].

In a mouse model of lung cancer, mice given Metformin orally or by injection had a reduction in the number and size of tumors [277]. Finally, since AMPK activating drugs are already on the drug market, they may also provide useful therapeutics against certain viral infections. In Chapter IV I discuss the role of AMPK activation in control of

RNA virus infection. Metformin has already been tested in several clinical trials for the treatment of Hepatitis C Virus (HCV). Two small trials with Pioglitazone had inconsistent results, while a larger trial using metformin showed significant improvement in sustained virological response particularly among females [278,279,280]. In Chapter IV I provide evidence that AMPK activation is more broadly anti-viral among disparate RNA viruses, and therefore these AMPK activating drugs may make promising anti-viral therapeutics against multiple infections.

8. Aims of Present Studies

Host interactions play an integral role in determining the outcome of viral infection. These interactions can be required for the virus to complete certain stages of its lifecycle, or can restrict viral infection as part of an innate immune response program.

Since viral pathogens represent a significant threat both medically and agriculturally, a better understanding of both types of host interactions could facilitate the treatment of viral infections and provide novel targets for therapeutics. Host factors that are required for infection could be targeted with inhibitory molecules to block the virus lifecycle and prevent replication, while host processes that contribute to viral restriction could be enhanced to further assist the innate immune response in controlling infection. Therefore the work described in Chapters III and IV aims to investigate the role of cellular metabolic signaling in viral infection. Cellular metabolic signaling is mediated through

AMPK, the central sensor of cellular energy levels. In response to various types of cellular stress centering on low energy conditions, AMPK is activated and induces changes in numerous downstream pathways, many of which are likely to be involved in viral infection.

In Chapter III I describe an RNAi screen of Drosophila kinases and phosphatases that led to the identification of AMPK as a novel host factor that contributes to infection of the Poxvirus vaccinia. In mammalian cells AMPK contributes to vaccinia infection by facilitating entry through macropinocytosis, and importantly, AMPK contributes to other cell biological processes that depend on rearrangements of the actin cytoskeleton. This role of AMPK in cytoskeletal rearrangements has only recently been discovered, and it is likely that other novel players upstream and downstream of AMPK also contribute to this process and have yet to be identified.

Macropinocytosis is a specific actin-dependent route that a subset of viruses employ for entry. I have shown in Chapter III that the requirement for AMPK during entry is specific to poxviruses. During these studies, I found a number of RNA viruses, in particular Rift Valley Fever Virus (RVFV), that are actually restricted by AMPK, as discussed in Chapter IV. RNA viruses rearrange and proliferate internal cellular membranes into distinct invaginations or vesicle-like structures to compartmentalize the viral replication complex and support genome replication. AMPK is a potent inhibitor of fatty acid synthesis required for these membrane proliferations, and restricts RVFV through this mechanism. AMPK activation reduces levels of cellular lipids, while exogenous addition of the fatty acid palmitate restores RVFV infection by bypassing fatty acid synthesis. Since additional viruses such as VSV and Sindbis are also restricted by

AMPK, this mechanism may be broadly anti-viral. Furthermore, since AMPK-activating drugs are currently in use for treatment of type II diabetes, I postulate that they may also be useful therapeutics against a broad range of viral infections. Taken together, AMPK represents a host factor that can have either pro-viral or anti-viral effects depending on the specific life cycle requirements of each virus, and as such may represent a useful drug target.

Figure 1.1: AMPK regulates protein and lipid metabolism. A. AMPK inhibits protein synthesis by dually inhibiting activity of mTORC1 and eEF2. mTORC1 induces translation initiation through 4E‐BP1 and S6K1, and also inhibits autophagy. Phosphorylation of eEF2 by eEF2K inhibits its activity in translation elongation. B. AMPK is a key regulator of lipid metabolism, inhibiting fatty acid and sterol synthesis while promoting fatty acid oxidation.

II. MATERIALS AND METHODS

1. Cells, Antibodies, Reagents, and Viruses

Drosophila DL1 cells were grown and maintained at 25°C in Schneiders

Drosophila media supplemented with 10% FBS (JRH) as described [7]. MEFs, BSC-1,

BHK and U2OS cells were maintained at 37°C in DMEM supplemented with 10% FBS

(Sigma) and 10mM Hepes. HeLa S3 suspension cells were maintained in MEM

supplemented with 10% FBS and 0.05% Pluronic. BSC-1 cells were maintained in MEM

supplemented with 10% cosmic calf serum (Hyclone). All media were additionally

supplemented with 100 ug/ml penicillin/streptomycin and 2 mM L-glutamine. LKB1-/-

MEFs were complemented with MIGR (Vector) or FLAG-LKB1-MIGR (LKB1 cDNA) and maintained as above.

Vaccinia strains vPRA13, vSC8, and vP30CP77, were grown in HeLa S3 suspension cells supplemented with 2.5% FBS, and titered on BSC-1 cells as described

[281,282,283,284]. Cowpox Brighton Red, Vesicular Stomatitis virus (Indiana), Rift

Valley Fever Virus MP-12 and Sindbis Virus were used. Antibodies were obtained from the following sources: anti-Bgal (Promega and Cappel), anti-E3L (gift from S. Isaacs)

[285], anti-L1R (R180 gift from G. Cohen and R. Eisenberg), anti-RVFV ID8, anti-Rac1

(Millipore),and anti-P-AMPK, t-AMPK, P-ACC, t-ACC, P-eEF2, t-eEF2 (Cell Signaling

Technology). Fluorescently labeled secondary antibodies along with anti-sheep HRP were obtained from Jackson Immunochemicals or Invitrogen. All other HRP-conjugated antibodies were obtained from Amersham. AlexaFluor 488 and 594 phalloidin, FITC- conjugated dextran, and 594-conjugated Transferrin and BODIPY-TR were purchased

34 from Invitrogen. Compound C [258] and STO-609 [286] were obtained from

Calbiochem. Additional chemicals were obtained from Sigma.

2. RNAi and Infections

A mini library of dsRNAs generated against Drosophila kinase and phosphatase genes was obtained from N. Perrimon, and aliquoted onto 384 well plates at 250ng dsRNA/384 well [287]. Secondary amplicons and control dsRNA were designed using

SnapDragon and DRSC resources (www.flyrnai.org), and generated as described [288].

RNAi was performed as described [289]. Briefly, for 384 well assays, 16,000 DL1 cells were seeded onto 250ng dsRNA in 10ul serum free media. One hour later 20ul complete media was added, and cells were incubated in a humid chamber for 3 days. For other experiments, 2,000,000 cells were seeded onto 4ug of dsRNA/6 well in 1mL serum free media. One hour later 2 mL complete media was added. For viral infections, vaccinia was titered on BSC-1 cells, and MOIs added to all cell types are based on pfu/ml measured on BSC-1 cells. All other viruses were tittered on BHK cells. Media was removed and virus was added in 2% serum medium and incubated at 25°C for

Drosophila cells, and 37°C for mammalian cells. For RVFV, Sindbis, and VSV, cells were spinoculated for 1 hour at 1200 RPM for 1 hour prior to incubation. Viral innocula used was adjusted to achieve ~10% infection of Drosophila cells in the primary screen, and ~20% infection in secondary analysis. Level of infection of mammalian cells varied depending on the assay format and was optimized through viral serial dilutions. Cells were processed at the indicated time point post infection.

3. siRNA

siRNAs were obtained from ABI. U2OS cells were reverse transefected with

HiPerfect and incubated for 3 days followed by infection at MOI=10 for 8 hours, fixed, and processed for immunofluorescence.

4. Viral Immunofluorescence

Cells were fixed and processed for immunofluorescence as previously described at 48 hours post infection for Drosophila cells and 8 hours post infection in mammalian cells for vaccinia, and 10 hours post infection for RVFV, Sindbis, and VSV [8]. Briefly, cells were fixed in 4% formaldehyde/PBS, washed twice in PBS/0.1% TritonX-100

(PBST), and blocked in 2% BSA/PBST. Primary antibodies were diluted in block, added to cells, and incubated overnight at 4ºC. Vaccinia was stained with anti-E3L and anti-B- gal. RVFV was stained with anti-RVFV ID8. VSV and Sindbis expressed GFP, and did not require antibody staining. Cells were washed three times in PBST, and incubated in secondary antibody for one hour at room temperature. Cells were counterstained with

Hoescht33342 (Sigma). Plates were imaged at 20X for Drosophila cells and 10X for mammalian cells, capturing three images per well per wavelength using an automated microscope (ImageXpress Micro), and quantification was performed using MetaXpress image analysis software. Significance was determined using a Student T-test.

5. Screen Analysis

Image analysis was used to generate metrics from the captured images including the number of nuclei and the number of infected cells per site. The percent infection was calculated for each site, log-transformed, and the interquartile range (IQR) was used to

36 calculate a robust Z score for each site using the following equation: log10 [(%infection- median)/(IQR*0.74)] [290]. Candidates were identified as positive if the average robust Z score of all sites in a well was <-2 in two independent replicates.

6. Immunoblotting, Northern blotting, and RT-PCR

For protein analysis, MEFs were prechilled to 16°C for 10 minutes and then treated with vaccinia (MOI 20) for 1 hour at 16°C to synchronize infection. Cells were incubated at 37°C for 10 or 30 minutes or treated with 10uM 2DG for 30 minutes. To analyze downstream effectors of AMPK, MEFs were treated with 12 mM 2DG, 10 uM oligomycin, or 100uM A769662 for 4 hours. Cells were then washed briefly in cold PBS and lysed in NP40 lysis buffer supplemented with protease (Boehringer) and phosphatase (Sigma) inhibitor cocktails. Samples were separated by SDS-PAGE and blotted as described [170]. HRP-conjugated secondary antibodies and Western

Lightening Chemiluminescence Reagent were used for visualization.

For RNA analysis, cells were lysed in Trizol buffer, and RNA was purified and blotted as previously described with the indicated probes [8]. RT-PCR was performed using M-MLV reverse transcriptase on random primed total RNA (Invitrogen). One uL of the cDNA or a 1:10 dilution was used for PCR amplification.

7. Plaque Assays

Viruses were plaqued on MEF or BSC-1 cells as indicated. Confluent monolayers were treated with serial dilutions of virus for two hours, after which the cells were overlayed with agarose, and incubated at 37°C for 48 hours. Cells were fixed in 10% formaldehyde, followed by crystal violet staining. Plaque number was determined

37 manually, and plaque diameter was measured using MetaXpress software and used to calculate areas.

8. Vaccinia Entry Assay

MEFs plated on coverslips were chilled to 16°C for 10 minutes and then treated with vaccinia (MOI 100) at 16°C for 1 hour. Unbound virus was removed, and cells were incubated at 37°C for 1 hour, washed three times in cold PBS, and fixed. Cells were washed in ammonium chloride (50mM) and PBST and were stained with anti-L1R and then washed and incubated with secondary antibody, Hoescht 33342, and phalloidin

488. Coverslips were mounted and imaged using using a 63X objective with a Leica DMI

4000 B fluorescent microscope. Images were taken as 0.2 um z-stacks that were deconvolved using AutoQuant X2 software using Adaptive PSF with 20 iterations.

Images are displayed as max projections. To quantify, images were randomized and blindly quantified for virus entry (n>30 for each condition).

9. Fluid Phase Dextran Uptake Assay

MEFs grown on glass coverslips were chilled to 16ºC for 10 minutes and then treated with vaccinia (MOI=200) at 16º C for 1 hour. Unbound virus was removed, and

FITC-dextran (70kD, lysine fixable) was added at 0.5 mg/ml. Cells were incubated at 37º

C for 20 minutes, washed twice in PBS, and once in pH 5.5 buffer (0.1 M sodium acetate, 0.05 M NaCl) for 5 minutes. Cells were fixed and stained with Hoescht 33342 and phalloidin 594. Coverslips were mounted and imaged using a 63X objective with a

Leica DMI 4000 B fluorescent microscope. Images were randomized and blindly quantified for the percentage of cells undergoing macropinocytosis as defined by >20

38 puncate per cell. U2OS cells grown on glass coverslips were pretreated with 10uM

Compound C or vehicle for 1 hour and then assayed as above.

10. Actin Ruffling Assay

Cells were grown on glass coverslips and treated with vehicle or 1 uM PMA for 3 hours. For Rac1 localization experiments, cells were blocked in 8% BSA/PBST for 1 hour. Anti-Rac1 (Millipore) was added in 1% BSA/PBST overnight at 4ºC Cells were washed 3 times in PBST, and secondary antibodies were added for 1 hour at room temperature. For all experiments, cells were stained with Hoescht 33342 and phalloidin

488. Coverslips were mounted and imaged using a 63X objective with a Leica DMI 4000

B fluorescent microscope.

11. Transferrin Uptake Assay

Cells grown on glass coverslips were chilled to 16ºC for 10 minutes and then treated with vaccinia (MOI=100) at 16º C for 1 hour. Unbound virus was removed, and

594-transferrin was added at 20 ug/ml. Cells were incubated at 37º C for 20 minutes, washed twice in PBS, and once in pH 5.5 buffer (0.1 M sodium acetate, 0.05 M NaCl) for

5 minutes. Cells were fixed and stained with Hoescht 33342 and phalloidin 488.

Coverslips were mounted and imaged using a 63X objective with a Leica DMI 4000 B fluorescent microscope.

12. Wound Healing Assay

Cells were grown to 100% confluence overnight, then scratched with a pipet tip to wound. Several marks were made along the length of the wound, and were imaged over

39 time, using these marks as guides. Images were analyzed for wound length at the same position over time using MetaXpress software.

13. Cellular Lipid Staining

Cellular lipids were stained as previously described [291,292]. MEFs were grown to confluence overnight, and then treated with PBS vehicle or 100uM A769662 for 10 hours. Cells were fixed in 4% formaldehyde for 10 minutes and washed three times in

PBS. Staining was performed with 10 ug/ml BODIPY-TR and counterstained with

Hoescht33342 in 100mM glycine in PBS overnight. Cells were washed three times in

PBS and imaged using the ImageXpress Micro automated microscope.

14. Fatty Acid Synthesis Bypass Assay

Exogenous palmitate addition was performed as previously described [293].

Delipidated Fetal Calf Serum and Albumin-bound palmitate were prepared as described

[293] and obtained as a kind gift from Robert Rawson. U2OS cells were set up on day 0 in 96 well plates and grown over night in normal growth medium. On day 1 medium was removed and cells were washed briefly in PBS. Cells were refed with low glucose DMEM supplemented with 5% delipidated Fetal Calf Serum with or without 100uM Albumin- bound palmitate, and incubated overnight. On day 2 cells were treated with 100 uM

A769662 or PBS vehicle for 1 hour, and infected with RVFV Plates were shaken for 5 minutes, spinoculated for 1 hour at 1200 RPM, and incubated for 10 hours to allow infection. Cells were fixed in 4% formaldehyde, stained with antibodies, and imaged at

10 X using the automated microscope ImageXpress Micro, as described above.

Quantification was performed using MetaXpress image analysis software. Significance was determined using a Student T-test. 40

III. A KINOME RNAI SCREEN IDENTIFIED AMPK AS PROMOTING POXVIRUS

ENTRY THROUGH THE CONTROL OF ACTIN DYNAMICS1

1. Introduction

In order to successfully infect cells, viruses must remodel the cellular environment to allow for the reallocation of resources to viral production. Poxviruses are large dsDNA viruses that have a sophisticated lifecycle characterized by a number of temporally regulated steps. Vaccinia virus is the prototypical poxvirus, was used as the vaccine to eradicate smallpox, and has been the most thoroughly characterized [11]. To initiate infection, vaccinia first binds, enters cells, uncoats, and expresses early gene products. Next, genomic DNA replication occurs, followed by intermediate and late gene expression. Assembly, maturation, and virus release completes the cycle. Although poxviruses encode a large number of genes (>200), they remain obligate intracellular pathogens and require a multitude of activities hijacked from their host cell. While many viral factors required for various steps in the vaccinia lifecycle have been described, the specific host factor contribution is less clear.

In particular, an early step in the infection cycle involves cell penetration. This step is critical for the initial establishment of infection, and also presents a good target for anti-viral therapeutics [294]. Different families of viruses have developed diverse strategies for entering cells; some fuse at the plasma membrane, while others co-opt

1. This chapter was reprinted from Moser TS, Jones RG, Thompson CB, Coyne CB, Cherry S, 2010 A Kinome RNAi Screen Identified AMPK as Promoting Poxvirus Entry through the Control of Actin Dynamics. PLoS Pathog 6(6): e1000954. doi:10.1371/journal.ppat.1000954

41 one of the many cellular endocytic routes [58]. Studies have demonstrated that macropinocytosis is an important endocytic route of vaccinia entry [83]. Generally, macropinocytosis is a nonselective route for bulk fluid-phase uptake and is not constitutively active, but is induced by growth factors, and also by some pathogens including vaccinia [87,88]. This active endocytic process induces extensive actin cytoskeletal rearrangement, leading to membrane ruffling, lamellipodia formation, and the internalization of extracellular fluid and membrane. Consistent with this, vaccinia entry is dependent upon modulation of the actin cytoskeleton, and initiates macropinocytosis by inducing dramatic actin-rich microvilli protrusions followed by global myosin II-dependent blebbing, thereby promoting virus uptake [78,83]. Induction triggers the activation of receptor tyrosine kinases (RTKs) which activate complex signaling cascades leading to the induction of these actin extensions which extend the plasma membrane allowing fluid-phase capture. This process involves signaling cascades that converge on members of the Ras superfamily of GTPases in particular, Rab5 and Rac1

[88,89]. Rac1 contributes to a number of cellular processes that require extensive actin dynamics, and its signaling is carefully regulated by several guanine exchange factors as well as by crosstalk with other Rho family GTPases [90,91,92]. Again, as for growth factor dependent macropinocytosis, vaccinia-induced uptake is dependent upon Rac1

[78,83]. Additional kinases such as Pak1 are then activated along with actin-associated proteins that ultimately lead to large-scale actin rearrangements, lipid modifications, and ultimately macropinosome formation [83,87]. While some specific kinase families have been implicated in macropinocytosis (e.g., protein kinase C (PKC), serine/threonine kinases, tyrosine kinases, and phosphatidylinositol kinases) [87], many of the specific factors have not been identified, and in some cases the specific role of factors such as

PKC, is not well understood. Therefore, there are many additional cellular signaling factors remaining to be identified for this complicated uptake mechanism and thus, for vaccinia entry.

To take an unbiased systematic approach toward the identification of these cellular factors, we developed a system using the model organism Drosophila to perform a high-throughput RNA interference (RNAi) screen for cellular kinases and phosphatases required for early steps in vaccinia infection. The Drosophila system is particularly amenable to this approach for a number of reasons including: reduced redundancy in the genome, high conservation with mammalian systems, efficient RNAi, and previous success with this system to identify cellular factors required for viral infection [8,295]. This Drosophila system is permissive to early steps in the vaccinia infection cycle allowing us to specifically dissect the role of cellular factors involved in the infectious entry process.

Using this system, we identified seven genes that contribute to vaccinia infection, including the three subunits of the AMP-activated kinase (AMPK) complex, the master energy sensor of the cell. Importantly, the requirement for AMPK in vaccinia infection is conserved in mammalian cells, and is specifically required for vaccinia-induced macropinocytic entry. Further characterization led to the discovery that AMPK controls a variety of virus-independent actin-dependent processes including lamellipodia formation and cell migration. Altogether, we found a new role for AMPK in actin dynamics.

2. Results a. Vaccinia Infection in Drosophila Cells

Since vaccinia infection of Drosophila cells has not been reported, we first characterized the course of infection in Drosophila cells. Using a reporter virus expressing Beta-galactosidase (B-gal) under the control of an early/late promoter which is active during all stages of vaccinia infection, we found that vaccinia infection is dose- dependent with maximal expression at 48 hours post infection (hpi) (Supplementary

Figure 1). Next, we infected Drosophila cells using reporter viruses that express B-gal under the control of temporally regulated vaccinia promoters that are active during different phases of the virus replication cycle. We found that Drosophila cells were efficiently infected as measured by the production of B-gal from an early/late promoter (p

7.5) or by the production of E3L protein, a vaccinia gene product expressed early in infection while, there was very little expression of B-gal from either an intermediate promoter (G8R), or a late promoter (p11) (Figure 3.1A). Consistent with these findings, we have been unable to detect vaccinia DNA replication (data not shown) suggesting a block to infection following early protein synthesis. These findings demonstrate that while vaccinia is unable to complete all stages of the lifecycle in Drosophila cells, entry and early expression occur, providing a model system to study the host factor requirements of vaccinia entry.

Previous studies have shown that efficient vaccinia entry is dependent upon the endocytic route of macropinocytosis [83]. In order to assess whether host requirements for vaccinia entry were conserved between Drosophila and mammalian cell lines, we tested whether inhibition of macropinocytosis attenuated infection in these disparate cell

44 types. To this end, we treated cells with several known inhibitors of macropinocytosis and vaccinia entry including; an actin inhibitor Latrunculin A, PI3K inhibitor Wortmannin,

Na/H antiporter inhibitor EIPA, and PKC inhibitor Rottlerin [78,83]. We found that each of these drugs significantly inhibited vaccinia infection in both human and Drosophila cells

(Figure 3.1B-C, quantified in Supplementary Figure 2). These data show that at least early steps in the viral lifecycle are dependent upon similar pathways in insect cells allowing us to use this model to identify additional factors required for vaccinia infection in mammalian cells. b. RNAi Screen of Drosophila Kinome

In order to systematically probe the requirements for cellular signaling factors in early vaccinia infection, we developed a quantitative assay amenable to RNAi using virally encoded B-gal expression as a measure of early infection. While a non-targeting negative control dsRNA (GFP) had no effect on the percentage of infected cells, knockdown of B-gal by RNAi reduced the percentage of B-gal expressing cells 17-fold, indicating that vaccinia infection can be quantitatively assayed using this system (Figure

3.2A). Moreover, dsRNA targeting the cellular gene Rab5, a small GTPase required for many endocytic processes including macropinocytosis [89] also significantly decreased vaccinia infection, validating that we can identify cell-encoded factors required for vaccinia infection using this approach (Figure 3.2A and 3.2D).

We used this assay to perform an RNAi screen against the Drosophila kinome to identify novel signaling factors that promote vaccinia infection (Schematic diagram

Figure 3.2B). This screen consisted of approximately 440 unique genes (~200 kinases,

~90 phosphatases, and ~150 regulator factors) arrayed onto 384 well plates.

Additionally, negative control wells were included containing either no dsRNA (15 wells) or dsRNA targeting GFP (28 wells), which is not expressed in this system. Lastly, 21 positive control wells with dsRNA targeting lacZ were included (Figure 3.2B light blue).

Drosophila cells were seeded in these pre-arrayed 384 well plates, incubated for three days to allow knock down of each targeted gene, and then infected with vaccinia virus for 48 hours. For the screen, a baseline infection of 10% was within the linear range and was achieved at an MOI of 1.25 (Supplementary Figure 1C). The plates were fixed and processed for immunofluorescence using B-gal expression as a measure of infection, and counter-stained to monitor cell number. Automated microscopy and image analysis were used to quantify the percent infection (B-gal+/Total Nuclei) that was transformed into Robust Z scores for each plate, and the Robust Z scores of the 2 replicates were plotted against each other (Figure 3.2B). Positive candidates were defined as having a

Robust Z score of < -2 in duplicate screens (p<0.05). Using these metrics, we identified

8 genes (2%, orange and pink Figure 3.2B). In addition to these 8 factors, we identified

20 out of 21 (95%) of the positive control lacZ dsRNAs (Figure 3.2B light blue) and none of the 43 negative controls (non-targeting dsRNA and empty wells). We also monitored the toxicity of the dsRNA treatments and found that none of the 8 genes that inhibited infection were cytotoxic (<25% decrease in cell number, Table1). In contrast, we found that while 17 wells reduced cell number by >25% in duplicate screens, none of these genes also inhibited infection. Therefore, our screen revealed host factors required for robust infection that are not required for cell viability. Notably all 8 genes have human homologs (Table 1), including all 3 subunits of the AMP-activated kinase (AMPK) complex (SNF1A (AMPKα), SNF4Agamma (AMPKγ) and alicorn (AMPKβ)) (Figure 3.2B pink), a heterotrimeric complex involved in maintaining cellular energy homeostasis.

RNAi resulted in ~3-fold reduction in vaccinia infection when AMPK was depleted

(Figure 3.2C).

Non-overlapping secondary dsRNAs were generated for each of the 8 genes and were tested to confirm the role of each of these genes in vaccinia infection. Seven of the genes validated (88%). Importantly, among the genes that validated were those encoding the three subunits of the AMPK complex (Figure 3.2D, Supplementary Figure 3 and Table 1). Moreover, RT-PCR confirmed that AMPKα was depleted by dsRNA treatment against AMPKα (Supplementary Figure 4A). Furthermore, we found that loss of AMPKα or AMPKγ also led to a defect in early viral mRNA accumulation compared to control (Figure 3.2E), suggesting that AMPK is required upstream of viral mRNA production in Drosophila, perhaps at the stage of entry. c. AMPK Promotes Vaccinia Infection in Mammalian Cells

AMPK is an important sensor of intracellular energy that is conserved in eukaryotes ranging from yeast to humans [174]. While Drosophila encodes only one copy of each AMPK subunit, mammals have multiple isoforms of each subunit encoded by several distinct genes (α1, α2, β1, β2, γ1, γ2, γ3) which can produce at least 12 possible heterotrimeric combinations [176]. The lack of redundancy in Drosophila allowed our identification of AMPK by single gene RNAi. We used this Drosophila system as a tool to identify novel host factors that contribute to vaccinia infection, but since Drosophila is not a natural host, we were interested in determining the role of

AMPK in a more biologically relevant context.

To investigate the role of AMPK in vaccinia virus infection of mammalian cells, we took advantage of mouse embryonic fibroblasts (MEFs) that are genetically altered 47

and null for both AMPKα subunits, AMPKα1 and AMPKα2 (AMPKα1/AMPKα2 -/-) and verified the lack of these proteins by immunoblot analyses (Supplementary Figure 5)

[177,178,296]. These cell lines divide and grow indistinguishably from their sibling control AMPKα1/AMPKα2 +/+ cells (wild type) (data not shown). We challenged either the

AMPKα1/AMPKα2-/- cells or their sibling control wild type cells with vaccinia virus and

measured infection using a plaque assay. This revealed a 20-fold decrease in plaque

number and a 15-fold decrease in plaque area in AMPKα1/AMPKα2 -/- compared to wild type cells (Figure 3.3A-C). The requirement for AMPK was also observed for the closely related poxvirus, cowpox virus (Figure 3.3D, quantified in Supplementary Figure 6).

These decreases in infectivity were specific for poxviruses and not simply due to a decrease in overall cell health since several unrelated RNA viruses, including Vesicular

Stomatitis virus (VSV) grew as well in the AMPK deficient MEFs compared to wild type

(Supplementary Figure 7 and data not shown). This suggests that AMPK deficient cells are capable of supporting all stages of virus infection; including at least some forms of endocytosis and endosomal trafficking since VSV enters cells through clatherin- mediated endocytosis [297]. Therefore, we identified a specific requirement for AMPK in poxvirus infection but not for viral infection generally. In addition to plaque assays we also monitored vaccinia infection by immunofluorescence, immunoblot, and Northern blot and found that there was a significant decrease in vaccinia virus replication in

AMPKα1/AMPKα2 -/- cells in each assay (Supplementary Figure 8A-C). This suggests

that AMPK promotes early steps of the vaccinia lifecycle in mammalian cells as well as

Drosophila. Finally, to verify that the requirement for AMPK was not MEF-specific we

tested whether inhibition of AMPK in the human osteosarcoma cell line U2OS

attenutated vaccinia infection using two approaches. First, we pre-treated U2OS cells

with the AMPK inhibitor Compound C or vehicle and challenged these cells with vaccinia

[258]. Again, we found that inhibition of AMPK attenuated infection (Figure 3.4A, B).

Next, we depleted AMPKα1, AMPKα2, or both AMPKα1 and AMPKα2 using siRNAs and observed a decrease in vaccinia infection (Supplementary Figure 9A, B). We confirmed knock-down by immunoblot (Supplementary Figure 9C). Together, these data show that vaccinia infection is dependent upon AMPK for infection across disparate cell types including Drosophila, human and mouse. d. Vaccinia Infection Activates AMPK

AMPK is activated through phosphorylation of a threonine residue on the catalytic α subunit, which can be triggered by a variety of stimuli including an increase in the cellular ratio of AMP/ATP [173,184,298,299,300,301]. Since AMPK promotes vaccinia infection, we tested whether vaccinia infection activates AMPK. We used a phospho-specific antibody against AMPKα to measure AMPK activation and observed an increase in phospho-AMPKα within 10 minutes of vaccinia infection. Likewise, treatment with 2-deoxyglucose (2DG), a known activator of AMPK, led to an increase in

AMPK phosphorylation, while little phosphorylation was detected in untreated controls

(Figure 3.4C). This increase was not due to changes in total AMPK levels, and suggests that AMPK becomes activated very early in vaccinia infection. e. Known Upstream Activators of AMPK Are Not Required for Vaccinia Infection

Several upstream kinases have been implicated in AMPK activation under different conditions. The classic activator of AMPK is the tumor suppressor LKB1, which activates AMPK in response to energy deprivation [185,186]. In Drosophila, LKB1 is the only described upstream kinase required for AMPK activation and lkb1 mutants 49

phenocopy ampk mutants [175,242]. In contrast, in mammalian systems, LKB1 is the

upstream kinase in response to energy deprivation, while additional upstream kinases,

such as CaMKKβ have been implicated in AMPK activation under other conditions

[187,188]. We tested whether LKB1 was required for vaccinia infection using cells that

are null for LKB1 [302] and complemented with either vector alone (LKB1-/-; Vec), or an

LKB1 cDNA (LKB1-/-;LKB1) (Supplementary Figure 10) and found that loss of LKB1 had

no effect on vaccinia infection in mammalian cells (Figure 3.4D). Likewise, using RNAi to

deplete LKB1 in Drosophila cells, we found that it was dispensable for infection by

immunofluorescence (Figure 3.4E) and Northern blot (Figure 3.4F). RT-PCR analysis

validated that LKB1 was indeed knocked down in Drosophila cells (Supplementary

Figure 4B). We also tested whether CaMKK, the other well-established AMPK activator

in mammalian systems, was required for vaccinia infection. To this end, we pre-treated

U2OS cells with the CaMKK inhibitor STO609 prior to infection, and found no effect on

vaccinia infection with doses up to 5 ug/ml (Figure 3.4A, B). Taken together, these data

show that vaccinia infection is AMPK-dependent but LKB1 and CaMKK-independent.

f. AMPK Promotes Vaccinia Entry

Given that loss of AMPK led to a decrease in both viral mRNA and protein

production (Figure 3.2, Supplementary Figure 8), we tested whether AMPK was required

for efficient virus entry. We monitored viral entry into wild type or AMPKα1/AMPKα2 -/-

MEFs using a fluorescence-based assay. We prebound virus to the cells, and then allowed infection to proceed for one hour. Incoming virus particles were visualized using an antibody against L1R, a membrane-bound viral surface protein (Figure 3.5A).

Deconvolution of was used to visualize vaccinia inside of cells (Figure 3.5A, XZ view).

Quantification revealed a ~3-fold reduction in the number of AMPK mutant cells that

internalized virus (Figure 3.5B). These data show that AMPK promotes infection at the

stage of entry, although we have not ruled out that virus binding could also be affected

by lack of AMPK.

We were also interested in whether AMPK played a role in vaccinia infection

downstream of entry. First, we monitored both early and late gene expression and found

that while the percentage of cells expressing an early protein (E3L) or a late protein

(L1R) was reduced in AMPKα1/AMPKα2 -/- MEFs compared to wild type, 100% of the

cells that expressed early genes also expressed late genes, indicating no further block to

replication (Supplementary Figure 11A). We also measured the infectivity of virus

produced from AMPKα1/AMPKα2 -/- MEFs. We found a ~3-fold decrease in virus titer produced from AMPKα1/AMPKα2 -/- MEFs compared to wild type (Supplementary Figure

11B) which is similar to the decrease in viral entry (Figure 3.5B). These data suggest that while fewer AMPK deficient cells produce virus, the vaccinia released from these cells is as infectious as virus produced from wild type cells. Therefore, while AMPK is important for entry, it is dispensable for later steps in the viral lifecycle. g. AMPK Is Required for Vaccinia Induced Macropinocytosis

Previous studies established that a major entry route for vaccinia is macropinocytosis, which is required for and induced by vaccinia infection [83]. Since we observed a defect in viral entry in the AMPK mutant cells (Figure 3.5), and that inhibitors of macropinocytosis attenuated vaccinia infection (Figure 3.1), we tested whether AMPK plays a role in vaccinia-induced macropinocytosis. Macropinocytosis can be directly monitored by fluorescently labeled dextran uptake [303]. Neither wild type nor

AMPKα1/AMPKα2 -/- MEFs efficiently endocytosed dextran under resting conditions

(Figure 3.6A). However, upon vaccinia infection, macropinocytosis was dramatically induced in wild type MEFs as measured by an increase in dextran uptake into the cell. In contrast to the large number of dextran punctae observed in the infected wild type

MEFs, AMPKα1/AMPKα2 -/- MEFs did not efficiently take up the dextran in the presence

of virus (Figure 3.6A). We quantified the level of vaccinia-induced macropinocytosis in

the wild type versus AMPK deficient cells and found an approximately five-fold reduction

in the percentage of cells undergoing macropinocytosis (Figure 3.6B).

Furthermore, we found a decrease in vaccinia-induced dextran uptake in human

U2OS cells pretreated with Compound C compared to vehicle control (Figure 3.6C).

Quantification revealed an approximately five-fold decrease in the percentage of U2OS

cells undergoing vaccinia-induced macropinocytosis when AMPK is inhibited (Figure

3.6D). Together, these data show that virus-induced macropinocytosis is dependent

upon AMPK in disparate cell types and hosts.

In contrast to this dependence of macropinocytosis on AMPK, there was no

defect in transferrin uptake in AMPKα1/AMPKα2 -/- MEFs (Supplementary Figure 12), indicating that receptor-mediated endocytosis is not controlled by AMPK. This is consistent with our observation that VSV infection is not attenuated in AMPKα1/AMPKα2

-/- MEFs as VSV enters cells by receptor-mediated endocytosis and not macropinocytosis (Supplementary Figure 7). h. AMPK Promotes Lamellipodia Formation

One of the early steps in macropinocytosis involves extensive actin remodeling characterized by membrane ruffling and lamellipodia formation. We were interested in 52 determining whether AMPK was required for this early step in the macropinocytic pathway, and whether the requirement for AMPK in macropinocytosis was vaccinia- dependent or if AMPK was required more generally for actin remodeling. Therefore, to test whether AMPK controlled actin-dependent remodeling independent of viral infection, we treated cells with phorbol myristic acid (PMA), which induces cells to undergo high levels of actin-mediated ruffling and lamellipodia formation; this is dependent on Rac1, a small Rho family GTPase that is also required for macropinocytosis and vaccinia infection [78,304,305,306]. Dramatic lamellipodia formation, seen as thick bands of actin at the cell periphery, were observed in wild type MEFs stimulated with PMA, but abrogated in AMPK-deficient cells (Figure 3.7A, arrows). We also monitored PMA- induced ruffling using live cell imaging and observed a significant defect in the AMPK mutant MEFs.

In addition, we monitored the localization of Rac1 during PMA-induced lamellipodia formation in the wild type and AMPK deficient cells and found that both

Rac1 re-localization and actin mobilization are defective in AMPK deficient cells (Figure

3.7B, arrows) suggesting that the defect is upstream of or parallel to Rac1 activation.

While our studies identified AMPK as a critical kinase required for vaccinia infection and actin dynamics, we found that LKB1 was dispensable for infection. This led us to test whether actin-dependent lamellipodia formation and ruffling was also LKB1- independent. We found that there was no defect in PMA-induced lamellipodia and ruffling in LKB1-deficient cells (Figure 3.7C, arrows). This was not unexpected since

HeLa cells, a cell type routinely used for studies on actin dynamics, are mutant for LKB1,

53 and can still undergo lamellipodia formation, and macropinocytosis [83,307,308,309].

Therefore, we found that the actin remodeling activity of AMPK is independent of LKB1. i. AMPK Promotes Cellular Motility

Since extensive actin remodeling and Rac1 membrane localization are required for lamellipodia formation, macropinocytosis, and vaccinia infection [78,304,305,306], we tested whether AMPK was required for another Rac1-dependent actin-dependent process namely in vitro wound healing [310]. For these studies, we created wounds in a confluent monolayer of either wild type or AMPKα1/AMPKα2-/- MEFs by scratching the surface, and monitored wound closure over time. Using this assay we found that there was a significant delay in the migration of AMPK-deficient MEFs into the wound compared to wild type cells (Figure 3.7D, E). While the wound was completely healed 24 hours after wounding in wild type cells, a sizable gap in the monolayer was still present in AMPK deficient cells, indicating a delay in wound healing and reduced motility. During this motile state, cells undergo a dramatic change in shape, with lamellipodia forming at the leading edge directing cell migration to close the wound. We observed the formation of lamellipodia in the WT MEFs at the edge of the wound while AMPK mutant MEFs did not polarize (Supplementary Figure 12). Consistent with our observations that the role for AMPK in actin dynamics is LKB1-independent, we found that wound healing is unaffected by the loss of LKB1 (Supplementary Figure 14).

3. Discussion

Cell penetration is a critically important step in viral infection, and one that is generally driven by cellular factors and signaling pathways. First, viruses must attach to the cell surface and bind to the viral entry receptor, which often initiates signaling within 54 the cell. Next, viruses must fuse or penetrate the cellular membrane either at the cell surface, or in many cases, within intracellular compartments by taking advantage of the endogenous endocytic machinery. Since there is an array of different endocytic mechanisms, there is great diversity in the strategies used by viruses for entry.

Therefore, studying virus entry has increased our understanding not only of viral infection, but also of the normal cellular processes of endocytosis [88].

To further dissect the cellular signaling requirements of vaccinia infectious entry, we developed an unbiased loss-of-function screening platform in Drosophila cells, and identified seven cellular factors required for vaccinia infection. Amongst the genes were all three subunits of AMPK, implicating the entire complex in vaccinia infection. Further studies showed that in addition to its role in Drosophila, AMPK is also required in murine and human cells for poxvirus infection at the stage of entry.

Studies indicate that vaccinia virus can enter cells through multiple routes via as of yet unidentified receptor(s). Studies suggest that virus particles can fuse either at the plasma membrane or from within endosomal compartments, dependent on cell type and virus strain [59,80,81,85,86]. Importantly, a major endocytic pathway for vaccinia entry has recently been described as macropinocytosis [77,83]. Our data as well as previous reports support the idea that vaccinia uses macropinocytosis for entry, but not exclusively. We and others have shown partial inhibition of vaccinia infection using a variety of drugs that are well-established inhibitors of macropinocytosis [78,83].

However, in no case was vaccinia entry completely dependent upon macropinocytosis for infection, demonstrating that the virus can use alternative routes for entry.

Nevertheless, macropinocytosis is required for efficient entry across broad cell types

55 suggesting that inhibition of this pathway may attenuate infection sufficiently to allow for immune-mediated clearance of the infection.

The process of macropinocytosis drives non-specific uptake of extracellular fluid, large portions of the plasma membrane, as well as large particles. Macropinosomes, unlike coated vesicles, are morphologically heterogenous, and can vary greatly in size from 0.2-10μm in diameter, sufficient to accommodate the large size of vaccinia virus particles (~0.3μm) [47,87]. Classic induction of macropinocytosis by growth factor receptor signaling stimulates ruffles, or sheet-like extensions of the plasma membrane, formed by assembly of actin filaments beneath the plasma membrane that form cups that contract and close to form macropinosomes [87]. This process is driven by signaling events initiated at the plasma membrane and are thought to involve a number of kinase families including PI3K, PI5K, PKC, serine/threonine kinases, and receptor tyrosine kinases. In addition, as many different inducers of macropinocytosis have been identified, there are likely multiple pathways that converge on the activation of macropinocytosis, adding to the complexity of this cell biological pathway [87]. While several specific kinases, such as Pak1, and LIM kinase have well described roles in macropinocytosis [311,312], there is much that remains unclear. Here, we have found a role for an additional kinase, AMPK, in promoting vaccinia entry through its role in macropinocytosis. We have found that AMPK deficiency attenuates vaccinia infection, concomitant with reduced entry and fluid-phase uptake, supporting an important role for

AMPK in vaccinia-induced macropinocytic entry.

The process of macropinocytosis involves several steps including extensive actin-mediated membrane ruffling, cup formation, and finally cup closure, which requires

56 the fusion of plasma membranes to close off the macropinosome, followed by fission to separate the macropinosome from the plasma membrane [87]. We discovered that

AMPK is required for the formation of lamellipodia and affects the recruitment of Rac1 to the cell periphery, suggesting that the role of AMPK in macropinocytosis lies in the initial rearrangement and reorganization of the actin cytoskeleton.

While we have shown that AMPK contributes to actin remodeling during vaccinia- induced macropinocytosis, we also found that AMPK plays a role in other virus- independent remodeling processes including cell migration. During this process, forward movement is driven by the extension of a leading edge protrusion or lamellipodium, followed by contraction at the rear. This protrusive force is generated by localized polymerization of actin mediated by Rac1 [313]. In addition to its role in controlling ruffling upstream of macropinocytosis, we found AMPK also has an essential role in cellular motility and wound healing, demonstrating a broad role in Rac1-dependent actin modulation. Previously, Rac1 has been implicated in nitric oxide production and glucose uptake downstream of AMPK [314,315]. This role for AMPK and Rac1 in glucose uptake via the translocation of the major glucose transporter GLUT-4 is quite interesting because this may provide a direct link between AMPK’s role in energy homeostasis and the cytoskeleton [316,317,318].

While the best understood role of AMPK is its role in metabolism, recent evidence suggests this kinase also has a crucial role in regulating cell structure and polarity through engagement with the actin cytoskeleton. In Drosophila, loss of AMPK leads to defects in mitotic division and epithelial cell polarity accompanied by disruption of the actin cytoskeleton [175]. In some mammalian epithelial cell lines, AMPK activation

57 leads to polarization characterized by the formation of an actin brush-border or tight junction assembly [175,236,237]. Additionally, AMPK activation can induce astrocyte stellation, and actin stress fiber disassembly [239]. Finally, studies using Compound C and AMPK activators linked AMPK to macropinocytic uptake of albumin in murine macrophages [240]. Taken together with our new data, this accumulating evidence suggests a broad and conserved role for AMPK in a variety of cellular processes that require actin cytoskeletal rearrangements.

AMPK can be activated in response to changes in intracellular calcium levels by

CaMKKβ, and further evidence suggests that TGFβ-activating kinase (TAK1) may serve as a third upstream activator [187,188,190]. Previous studies demonstrated that LKB1 is an important mediator of cell polarity at least in part through signaling to AMPK, and has been shown to drive actin brush border formation, and the translocation of apical and basal markers during the establishment of polarity [175,236,237,242,243]. We found that at least some AMPK-dependent cytoskeletal changes are independent of LKB1 and

CaMKK including lamellipodia formation, macropinocytosis and wound healing. These different actin-dependent outcomes could be controlled by the unique downstream Rho

GTPase family members that may become activated by AMPK depending on the stimulus and upstream kinase (such as Rac1, which is associated with lamellipodia formation and macropinocytosis) [91,92]. Study of the Rho family GTPases activated by

AMPK under different stimuli may resolve some of these issues.

In addition to the three subunits of AMPK discovered through screening the

kinome, we identified four other genes that promote vaccinia infection: Pi3K68D, Fab1,

Stam, and CG9311, all of which have human homologs. Pi3K68D, Fab1, and Stam are

kinases while CG9311 is the only phosphatase identified in the screen. As kinases are

druggable targets, and many known factors involved in macropinocytosis are kinases,

we are particularly interested in the role of these kinases in vaccinia infection. Both

Pi3K68D (PIK2C2A) and Fab1 (PIP5K3/PIKfyve) have roles in metabolism of

phosphatidylinositol (PtdIns), an important component of membrane trafficking,

cytoskeletal rearrangements, and macropinocytosis. In particular, Pi3K68D is a member

of the class II family of PI3Ks that produce PtdIns(3)P downstream of growth factor

stimulation, and can modulate the activity of Rho GTPases such as Rac1 and Cdc42.

Class II PI3Ks have a critical role in lamellipodia formation and in cell migration,

localizing to the leading edge of migrating cells [319,320]. Interestingly, class I PI3K

have also been implicated in vaccinia infection, during virus entry, and also later stages

of infection [83,321,322]. Fab1, the PtdIns(3)P 5-kinase that converts PtdIns(3)P into

PtdIns(3,5)P2, has been implicated in fluid-phase uptake, transport, and endosomal acidification [323,324]. The third kinase, Stam (STAM, STAM2) is activated by cytokine and growth factor stimulation, and localizes to early endosomes, where it is involved in endosomal sorting [325,326]. Since these kinases have roles either in trafficking to or from the plasma membrane, or in cytoskeletal rearrangements, and have been implicated in processes related to macropinocytosis, it is quite possible that they also play a direct role vaccinia entry. Perhaps Pi3K68D is involved in promoting cytoskeletal rearrangements that lead to macropinocytosis, while Fab1 and Stam could be involved

59 sequentially in later entry steps leading to membrane fusion once a macropinosome has been internalized.

Rearrangements in the actin cytoskeleton are crucial not only for vaccinia infection, but also for many essential cellular processes including: cell division, establishment and maintenance of polarity, cellular motility, and uptake of extracellular fluids, each of which must be carefully regulated. While these various processes have different outcomes for the cell, they share several important signaling components, with

AMPK as a central mediator. Further characterization of AMPK as well as these additional new factors is required to determine their precise role in vaccinia infection and whether they interact with AMPK, macropinocytosis, or other actin-dependent processes. How AMPK activation in response to different signals leads to these disparate changes in the actin cytoskeleton, and how these processes fit into the larger network of AMPK-dependent pathways will drive future studies. Lastly, the development of more selective AMPK inhibitors or other inhibitors of macropinocytosis may be useful against poxviruses, and other viruses that hijack this endocytic route for their entry mechanism.

Figure 3.1: Vaccinia virus undergoes infectious entry in Drosophila cells. A. Drosophila cells were infected with three different recombinant vaccinia viruses expressing B-gal driven by an early/late (p7.5), intermediate (G8R), or late (p11) vaccinia promoter. Each virus expresses E3L protein from an endogenous early promoter. After 48 hours, infection levels were assessed using B-gal or E3L-specific antibodies. B-C. Vaccinia infection is dependent upon known mediators of macropinocytosis. B. Human U2OS cells were pretreated with: Latrunculin A, Wortmannin, and Rottlerin at 5μM; EIPA at 12.5μM for 1 hour and challenged with vaccinia (MOI=10) for 8 hours. C. Drosophila DL1 cells were treated with: Latrunculin A, Wortmannin, and Rottlerin at 5μM; EIPA at 50μM for 1 hour and challenged with vaccinia (MOI=20) for 24 hours. Cells were fixed and processed for immunofluorescence using E3L expression (green) as a marker for infection, and Hoescht 33342 (blue) to visualize nuclei. Percent infection is the average of four images in a representative of three experiments.

Figure 3.2: High-content RNAi screen identifies cellular factors required for vaccinia infection. A. Cells were pre-treated with the indicated dsRNAs and infected with E/L (p7.5) B-gal vaccinia at multiplicity of infection (MOI) of 1.25 for 48 hours. Percentage of infected cells is calculated from ((B-gal+, green)/ (nuclei, blue)). B. Schematic of high-throughput RNAi screening. Cells were plated in 384 well plates arrayed with gene-specific dsRNA targeted against the Drosophila kinome. After three days, the cells were infected with E/L B-gal vaccinia at MOI=1.25 for two days and then processed for immunofluorescence. Automated image analysis was used to determine the percentage of infected cells and used to identify the positive candidates (Z<-2 in duplicate). Candidates in light blue represent the B-gal-depleted positive controls, and those in pink represent the three subunits of the heterotrimeric AMPK complex. Other candidates are indicated in orange. C. Examples of primary screen data including AMPK components that when depleted resulted in a significant decrease in the percentage of B-gal-positive cells. D. Independent dsRNAs against each of the three AMPK subunits: AMPKα (SNF1A), AMPKβ (alicorn) and AMPKγ (SNF4Agamma). RNAi was performed and cells were infected at an MOI=5. Percent infection was measured and the mean+SD is shown; * indicates p<0.05 compared to control in three independent experiments. E. Northern blot analysis was performed on RNA extracted from cell lysates pretreated with dsRNA targeting the indicated genes and probed for B-gal, and a loading control RpS6.

Table 3.1: Candidate genes identified and validated.

Robust Z score Nuclei fold change

% infection % inf ection f rom median validated?

Drosophila Human Gene Gene Screen #1 Screen #2 Screen #1 Screen #2 Screen #1 Screen #2

median 3.9 7 1 1

B-gal 0.9 1.2 -17.5 -10.8 1 1 SNF1A (AMPKα) PRKAA2 1.7 2.8 -4.5 -5.4 0.8 1 Yes SNF4Agamma (AMPKγ) PRKAG2 1.1 1.9 -3.9 -7 1 0.9 Yes Alicorn (AMPKβ) PRKAB1 0.3 4.2 -3.9 -3.4 1 1.1 Yes

Fab1 PIKfyve 3.4 4.5 -2.3 -2.2 1 1 Yes

Pi3K68D PIK3C2A 2.4 3.3 -3.8 -3.9 1.1 1.1 Yes

Stam STAM 3.3 4.5 -2.1 -2.5 1 1.1 Yes

CG9311 PTPN23 3.1 2.5 -3.8 -5.7 1 0.9 Yes

Strn-Mlck MYLK 0.6 4.8 -2.5 -2.8 1 1.1 No

Figure 3.3: AMPKα promotes poxvirus infection in mammalian cells. A. Vaccinia virus plaque assays were performed with wild type or AMPKα1/AMPKα2 -/- MEFs. Representative data from the10-5 dilution of virus is shown. B. Quantification of plaques from A. presented as the normalized mean+SD of wild type plaques from four experiments;* indicates p<0.05. C. The diameter of 30 representative plaques in each duplicate well from A. was used to calculate the average plaque area, which is displayed as the normalized mean+SD in triplicate experiments; * indicates p<0.05. D. Cowpox virus plaque assays were performed with wild type or AMPKα1/AMPKα2 -/- MEFs. Representative data from the10-5 dilution of virus is shown.

Figure 3.4: Vaccinia infection induces AMPK activation, but is independent of LKB1 and CaMKK. A. Human U2OS cells were pre-treated with vehicle, Compound C (1-5 μM), or STO609 (1-5μg/mL) and then infected with vaccinia (MOI=10) for 8 hours (E3L, green; nuclei, blue). A representative of 3 experiments is shown. B. Percent infection of A. was measured and the mean+SD is shown; * indicates p<0.05 compared to control in three independent experiments. C. Cells were pretreated with vaccinia virus (MOI=20) at 16°C for 1 hour, and then infected at 37°C or treated with 2DG, a known inducer of AMPK phosphorylation. Lysates were collected at the indicated time points post treatment and assayed by immunoblot for phospho-AMPK. A representative of 3 experiments is shown. D. LKB1 -/- MEFs were complemented with a vector control (LKB1-/-,Vec) or FLAG-LKB1 cDNA (LKB1-/-, LKB1) and subsequently infected with vaccinia at the indicated MOI for 8 hours, and processed for immunofluorescence. Percent infection was quantified using automated image analysis. A representative of two experiments is shown. E. RNAi against LKB1 in Drosophila cells had no effect on infectivity compared to RNAi against negative control luciferase as measured by immunofluorescence. F. LKB1 knock down in Drosophila has no effect on vaccinia early mRNA levels as measured by Northern blot. Cells were pretreated with dsRNAs against negative control luciferace, positive controls E3L (viral) and Rab5 (cellular), as well as AMPKα and LKB1 and probed as indicated.

Figure 3.5: AMPK promotes vaccinia entry. A. Wild type or AMPKα1/AMPKα2 -/- MEFs were either infected or mock-infected with vaccinia, and the L1R membrane protein (red) was monitored to visualize incoming virus. Nuclei (blue) and actin (green) were also stained. Images are presented as max projection along with XZ insets. Representative images of triplicate experiments are shown. B. The percentage of cells undergoing vaccinia entry was quantified (n>30 for each condition).

Figure 3.6: AMPK is required for vaccinia-induced macropinocytosis. Fluid phase dextran uptake assays were performed in the presence or absence of virus. A. Wild type and AMPKα1/AMPKα2 -/- MEFs were either infected or mock-infected and treated with FITC-dextran (red), processed for microscopy, and stained to visualize actin (phalloidin (green)) and nuclei (Hoescht 33342 (blue)). Representative images from triplicate experiments are shown. B. The percentage of MEFs with dextran punctae was quantified from three independent experiments with the mean +/- SD shown; * p<0.05. C. U2OS cells were treated or mock-treated with the AMPK inhibitor Compound C (10 μM) 1 hour prior to addition of vaccinia. Texas Red-dextran (red) was added and processed as above. D. The percentage of U2OS cells with dextran punctae was quantified from three independent experiments with the mean +/- SD shown; *p<0.05.

Figure 3.7: AMPK is required for actin-dependent membrane ruffling and wound healing independent of LKB1. A. Cells were treated with vehicle or PMA and stained with phalloidin (actin, green) and Hoescht 33342 (nuclei, blue). Arrows indicate lamellipodia. Representative images from triplicate experiments are shown. B. Cells were treated with PMA and stained with phalloidin (actin, green), anti-Rac1 (red) and Hoescht 33342 (nuclei, blue). Arrows indicate colocalization of actin and Rac1 in lamellipodia. Representative images from triplicate experiments are shown. C. MEFs null for LKB1 (LKB-/-;Vec) or rescued (LKB1-/-,LKB1) were treated with PMA or vehicle and stained as A.. Arrows indicate lamellipodia. Representative images from triplicate experiments are shown. D. A scratch was made in a confluent monolayer of wild type or AMPKα1/AMPKα2 -/- MEFs, and monitored over time for closure. Representative images from triplicate experiments are shown immediately after wounding (T=0) and after 20 hours (T=20). E. Reduction in wound width is quantified over time. Data are normalized to initial would width at T=0, and presented as means of three independent experiments with four wounds per set; * indicates p<0.05 in each experiment. Moreover, the rate of wound healing is reduced in AMPK mutants by ANOVA (p<0.001).

IV. AMP-ACTIVATED KINASE RESTRICTS RIFT VALLEY FEVER VIRUS INFECTION

BY INHIBITING FATTY ACID SYNTHESIS

1. Introduction

Arboviral pathogens infecting both humans and domestic animals have lead to significant world-wide morbidity and mortality and are both a medical and agricultural concern. Bunyaviruses are an important group of insect-borne RNA viruses that include several disease causing members such as Sin Nombre, Hantavirus, Crimean-Congo

Hemorrhagic Fever Virus, and Rift Valley Fever virus (RVFV). RVFV is a mosquito borne Category A agent initially endemic to sub-Saharan Africa. However, outbreaks of

RVFV have recently occurred in Egypt and the Arabian Peninsula, indicating the potential of this virus to spread to new geographical areas [98].

RVFV has particular importance as an agricultural pathogen, where infection of livestock can lead to significant morbidity and mortality, particularly among young animals, and cause catastrophic abortion rates [98]. Most humans infected with RVFV develop self-limited febrile illness, although approximately 1-3% die from the disease due to hemorrhagic symptoms [104,106,107,108]. A live-attenuated vaccine MP-12, as well as inactivated virus vaccine have been developed for veterinary use [109,110,111].

These vaccines can also provide protective immunity in humans; however, they are expensive to produce and in the case of MP-12 may not be acceptable in areas where

RVFV is not endemic [114,115]. Moreover, no effective anti-viral therapies have yet been developed against RVFV. Here we describe a novel mechanism of cellular restriction against RVFV mediated through energy-mediated regulation of lipid metabolism.

Lipid metabolism plays an important role in infection of many viruses that replicate in the cytoplasm including RVFV. An essential feature of these infections is the ability of viruses to rearrange and proliferate internal cellular membranes into distinct invaginations or vesicle-like structures to compartmentalize the viral replication complex and support genome replication [162]. These membranous structures may serve to closely coordinate the replication complexes with virion assembly and budding, act as a scaffold for the RC to tether viral RNA, increase the local concentration of replication components, or prevent detection by the host immune defense system [162]. Depending on the virus, these membrane modifications can be derived from distinct cellular sources, including ER, golgi, endosomal, and mitochondrial membranes, and may have complex biogenesis derived from multiple cellular origins

[150,163,164,165,166,167,168,169]. Despite these differences there are many similarities in the membrane structures induced by different types of viruses suggesting commonalities in this induction across multiple virus families [171]. In some cases the viral proteins responsible for these changes are known, but host factors are also likely to contribute to this process. Importantly these modifications are highly dependent on cellular lipid biogenesis pathways and require generation of newly synthesized cellular lipids [162].

AMP-activated Kinase (AMPK) is a heterotrimeric complex that is an essential regulator of lipid metabolism mediated through energy sensing [228]. Under conditions of energetic stress, this complex switches the cell from an anabolic to a catabolic state by activating oxidative pathways that generate energy while inhibiting synthesis and growth pathways in order to return the cell to a state of energy homeostasis [173]. Thus

AMPK activity is important for survival during periods of stress, and also has implications 70 in type II diabetes, obesity, metabolic syndrome, longevity, and cancer

[244,245,270,271,272,273,274]. AMPK is a heterotrimeric complex consisting of a catalytic alpha subunit, and regulatory beta and gamma subunits [173]. Mammals encode multiple isoforms of each subunit encoded by several distinct genes (α1, α2, β1,

β2, γ1, γ2, γ3) which can produce at least 12 possible heterotrimeric combinations [176].

Activation is triggered through binding of AMP or ADP to the Bateman domains of the gamma subunit, leading to increased phosphorylation of the threonine 172 on the alpha subunit by inhibiting dephosphorylation and inducing allosteric activation

[180,181,182,183]. The canonical upstream activator of AMPK that catalyzes the phosphorylation event is the constitutively active tumor suppressor LKB1, but additional activators such as CaMKKβ have also been identified [185,186,187,188,190].

In particular, AMPK is a potent regulator of lipid metabolism, inhibiting both sterol and fatty acid synthesis, while promoting fatty acid oxidation [228]. AMPK phosphorylation of acetyl CoA carboxylase (ACC), a rate limiting enzyme in fatty acid metabolism, inhibits its activity [229,230]. ACC catalyzes the irreversible conversion of acetyl-CoA to malonyl-CoA, a key metabolite that plays multiple roles in fatty acid metabolism. First, it is the substrate for fatty acid biogenesis, which drives de novo production of palmitate [231]. Second, malonyl-CoA is also a cosubstrate for chain lengthening of endogenously synthesized and dietary-derived essential fatty acids into higher polyunsaturated fatty acids [232]. Third, malonyl-CoA serves a signaling function through inhibition of carnitine palmitoyltransferase I (CPT-1), an essential factor in the transport of fatty acids to the mitochondria for beta oxidation [231]. Thus malonyl-CoA production by ACC promotes fatty acid synthesis, while inhibiting fatty acid oxidation.

Mammalian systems encode two nonredundant ACC isoforms, ACC1 and ACC2, which 71 are both inactivated by AMPK-mediated phosphorylation. Some studies suggest that malonyl-CoA produced by ACC2 is exclusively involved in fatty acid oxidation, while

ACC1 contributes to fatty acid biogenesis [228]. Therefore, activation of AMPK through stress or low energy conditions induces fatty acid oxidation through ACC2, while inhibiting fatty acid synthesis through ACC1.

We found that AMPK is potently antiviral against RVFV, and this restriction is dependent on the upstream activator LKB1. Furthermore, pharmacological activation of

AMPK inhibited viral infection, and correlated with decreased levels of cellular lipids.

Mechanistically, we found that AMPK is antiviral through its role in fatty acid metabolism as we could bypass the antiviral effects of AMPK by feeding cells palmitate, the metabolite produced by ACC. Palmitate treatment restored RVFV infection, suggesting that AMPK specifically restricts infection by inhibiting fatty acid synthesis. Since many viruses are dependent upon fatty acid biosynthesis for their replication, we tested whether AMPK restricted additional RNA viruses. We found that indeed, AMPK has antiviral activity against multiple RNA viruses from disparate families. Taken together, our data suggest that AMPK activation is broadly anti-viral, and may provide a novel anti-viral therapeutic target.

2. Results a. AMPK Expression Restricts RVFV Infection

We previously found that AMPK was required for efficient vaccinia infection. This finding led us to investigate the role of AMPK in other virus infections; we were particularly interested in RVFV as it is a virus that is medically important, but little is known about the mechanisms by which it establishes a productive infection. For our 72

studies we used the lab adapted strain MP12 that has only 11 amino acid differences

from the wild type strain [109]. In order to test the role of AMPK in RVFV infection, we

took advantage of mouse embryonic fibroblasts (MEF) that are genetically altered and

null for both of the catalytic α subunits, AMPKα1 and AMPKα2 (AMPKα1/AMPKα2-/-)

[177,179,296]. We challenged either the AMPKα1/AMPKα2-/- MEFs or their sibling control wild type MEFs with MP12 and measured infection by plaque assay (Figure

4.1A). We found a 6 fold increase in the number of plaques formed in

AMPKα1/AMPKα2-/- MEFs compared to wild type (Figure 4.1B) concomitant with a four- fold increase in average plaque area in AMPKα1/AMPKα2-/- MEFs (Figure 4.1C).

Moreover, RVFV infection was also increased in AMPKα1/AMPKα2-/- MEFs as measured by immunofluorescence (Figure 4.1D, quantified in Figure 4.1E), indicating that RVFV is able to infect and spread more efficiently in the absence of AMPK.

Next we examined if there were differences in infection kinetics in cells lacking

AMPK by performing a time course in wild type and AMPKα1/AMPKα2-/- MEFs using immunofluorescence to measure the percent of infected cells. In both cell types infection was first detected at 8 hours post infection (hpi)(Figure 4.1F), indicating similar infection kinetics; however, the level of infection was 2-fold higher in AMPKα1/AMPKα2-/- MEFs compared to wild type. By 12-16 hpi infection had spread to adjoining cells, as indicated by the appearance of foci of infected cells. In wild type MEFs we observed a 2-fold increase in infection by 16 hpi compared to 8 hpi, while a 3-fold increase was observed in AMPKα1/AMPKα2-/- MEFs during the same time interval (Figure 4.1F), indicating enhanced virus spread in the absence of AMPK.

This enhanced spread of the virus to surrounding cells could be due to increased

production of infectious virus or increased infectivity of virions produced. Next, we

measured the amount of infectious virus produced in wild type and AMPKα1/AMPKα2-/-

MEFs over time in a one-step growth curve. Media was collected at various times after

infection, and released virus was tittered on BHK cells. Little virus was detected at 2-4

hpi, indicating that input virus was not detected in this assay. At 12 hpi we observed a 5-

fold increase in infectious virus released from AMPKα1/AMPKα2-/- MEFs compared to wild type, indicating enhanced virus production in AMPKα1/AMPKα2-/- MEFs (Figure

4.1G). Taken together these data suggest that AMPK expression restricts RVFV both during initial infection, and through the reduced production of infectious virus. b. LKB1 Expression Restricts RVFV Infection

AMPK is activated through phosphorylation of a threonine residue on the catalytic alpha subunit. Several upstream kinases have been implicated in AMPK activation under different conditions. The classic activator of AMPK is the tumor suppressor LKB1, which phosphorylates AMPK in response to energy deprivation

[185,186]. This phosphorylation event can be triggered by a variety of stimuli that cause a reduction in cellular energy levels, such as glucose starvation or hypoxia [173]. Binding of AMP or ADP to the regulatory gamma subunit triggers increased phosphorylation of the Thr-172 of the catalytic alpha subunit by the tumor suppressor LKB1 [180,181,182].

In order to determine if LKB1 signaling was important for AMPK-mediated RVFV restriction, we tested the requirement for LKB1 during RVFV infection. We challenged

MEFs that are null for LKB1 and complemented with either vector alone (LKB1-/-; Vec),

or an LKB1 cDNA (LKB1-/-;LKB1) [327] and found increased RVFV infection in MEFs

lacking LKB1(Figure 4.2A). Quantification revealed a 2 fold increase in the number of

plaques (Figure 4.2B) and a 6 fold increase in plaque size (Figure 4.2C) in LKB1-/-; Vec

MEFs compared to the LKB1 complemented LKB1-/-;LKB1 MEFs. Moreover we

measured percent infection by immunofluoresence assay for RVFV (Figure 4.2D) and

found increased percent infection in the LKB1-/-;Vec MEFs compared to those complemented with LKB1. Finally, we also measured RVFV production in cells lacking

LKB1 using a one-step growth curve and found approximately a 4 fold increase in infectious virus produced in the LKB1-/-;LKB1 MEFs compared to those complemented

with LKB1 (Figure 4.2E). Taken together, these data indicate that AMPK signaling

mediated by LKB1 has antiviral activity against RVFV.

c. AMPK Activation Restricts RVFV

Since LKB1 is the canonical upstream activator of AMPK in response to low

energy conditions, and AMPK restricts RVFV infection through LKB1, we tested whether

RVFV was sensitive to drugs that activate AMPK by reducing levels of cellular energy.

We performed these studies both in MEFs and in the human osteosarcoma cell line

(U2OS). To this end, we pretreated U2OS cells or MEFs with the glucose analog 2-

deoxyglucose (2DG), or the ATP synthase inhibitor oligomycin for 1 hour prior to

infecting with RVFV at various MOIs. Both drugs significantly decreased infection of

RVFV compared to vehicle controls (Figure 4.3A and data not shown). These drugs

activate AMPK indirectly, and have pleiotropic effects that may also contribute to viral

infection. Additionally reduction of cellular energy levels may adversely affect viral

infection independent of AMPK. To control for these possibilities we tested the effects of

these drugs on vaccinia virus, which is not restricted by AMPK, and instead requires it

for efficient entry [327]. Vaccinia was able to enter and infect equally under all treatment

conditions (Figure 4.3B), indicating that the drug-treated cells are healthy enough to

support viral infection, and the reduced infection levels with RVFV were specific to

AMPK activation.

Furthermore, we took advantage of a recently developed thienopyridone

compound A769662 that activates AMPK independently the energy status of cells or

AMP levels [328,329]. This drug mimics both allosteric activation of AMPK and inhibition

of dephosphorylation without affecting binding of AMP to the gamma subunit [330].

Activation in the presence of this drug requires phosphorylation by an upstream

activator, but is not specific to a particular upstream pathway [330]. We tested the ability

of this drug to inhibit viral infection in U2OS cells and MEFs. RVFV infection was

significantly reduced in the presence of this compound (Figure 4.3C), indicating that

AMPK activation restricts RVFV infection independently of the pleiotropic effects of

reduced cellular energy levels. Furthermore, we tested whether A769662 has AMPK-

independent effects on infection [331], by comparing the activity of the drug against virus

in wild type and AMPK deficient MEFs. A769662 treatment significantly inhibited RVFV

infection in wild type MEFs but not in AMPKα1/AMPKα2-/- MEFs (Figure 4.3D). Taken together, these studies suggest that energy-mediated AMPK activation has antiviral activity against RVFV. d. Acetyl CoA Carboxylase Activity Is Inhibited by AMPK Expression and Activation

AMPK regulates several downstream pathways that could be important for viral infection, in particular protein translation and lipid synthesis. AMPK regulates translation through multiple pathways, inhibiting translation initiation by inactivating mTORC1, and

inhibiting translation elongation by inactivating eEF2 [204,220,221]. To investigate how

AMPK is able to restrict RVFV infection, we first examined the role of mTORC1 signaling

in RVFV infection with or without AMPK expression. Inactivation of mTORC1 by AMPK

leads to decreased translation initiation as well as increased autophagy, both of which

could be anti-viral [204]. Since AMPK activation inhibits mTORC1 activity, we

hypothesized that mTORC1, and thus protein synthesis, would be overactive in AMPK

deficient cells. We tested the role of mTORC1 signaling in RVFV infection using the

mTORC1 inhibitor Rapamycin, and found no significant difference in RVFV infection in

cells treated with Rapamycin compared to vehicle controls in either wild type or

AMPKα1/AMPKα2-/- MEFs , indicating that the mechanism of antiviral activity does not occur through mTORC1 signaling (Supplementary Figure 15). Furthermore, since autophagy has been shown to have antiviral effects in some models [192], we performed plaque assays to test whether autophagy restricted RVFV, and found no significant difference in plaque number or size in MEFs expressing a Atg5 hairpin compared to control MEFs (Supplementary Figure 16).

Next, we examined phosphorylation of several downstream targets of AMPK. eEF2 is an important regulator of translation initiation, while ACC regulates fatty acid metabolism [222,231]. Both are inactivated by phosphorylation downstream of AMPK

[173]. We tested the ability of AMPK-activating drugs to induce phosphorylation of these downstream targets in WT or AMPKα1/AMPKα2-/- MEFs. Cells were treated with AMPK activating drugs for 1 hour, lysates were collected and phosphorylation was measured by immunoblot using phosphor-specific antibodies. In WT MEFs AMPK phosphorylation was increased in response to 2DG, oligomycin, and A769662 treatment (Figure 4.4A).

We observed a concomitant increase in ACC and eEF2 phosphorylation (Figure4.4A).

As expected, neither total nor phospho-AMPK was detected in AMPKα1/AMPKα2-/-

MEFs under any treatment condition. We observed in an increase in eEF2 phosphorylation in response to treatment with 2DG and A769662 compared to vehicle treatment; although the level of eEF2 phosphorylation in AMPK deficient MEFs was reduced compared to wild type cells. In contrast, ACC phosphorylation was undetectable in the absence of AMPK (Figure 4.4A). ACC, eEF2, and the control protein tubulin were expressed equally in wild type and AMPKα1/AMPKα2-/- MEFs (Figure 4.4A). These findings suggest signaling pathways other than AMPK are important in regulating eEF2 phosphoylation during drug treatment, while ACC phosphorylation is exquisitely regulated by AMPK.

ACC is an important regulator of fatty acid metabolism, both by inhibiting fatty acid biosynthesis and activating fatty acid catabolism through beta-oxidation [231][228].

Fatty acid biosynthesis is an important component of RNA virus infection since numerous RNA viruses proliferate cellular membrane structures for proper formation of the viral replication complex [162]. A member of the Bunyavirus family related to RVFV,

Bunyamwera Virus, was reported to induce formation of a new Golgi-derived tubular structure that harbors the viral replication complex [169,332]. Disrupting formation of this structure with certain drugs or viral mutants was reportedly associated with decreased levels of virus production [332]. Since AMPK activation inhibits fatty acid synthesis by inactivating ACC, we tested whether altered levels of AMPK expression and activation cause global changes to cellular lipid levels. To this end, we examined levels of cellular lipids by staining wild type and AMPKα1/AMPKα2-/- MEFs with the lipophilic BODIPY fluorescent dye, and found increased BODIPY staining in cells lacking AMPK compared to wild type (Figure 4.4B). Furthermore, wild type MEFs treated with the AMPK activator

A769662 had reduced levels of BODIPY staining compared to untreated MEFs, while

little difference was seen in AMPKα1/AMPKα2-/- MEFs treated with A769662 compared to vehicle treated cells (Figure 4.4B). These findings suggest that the absence of AMPK leads to overproduction of cellular lipids, while AMPK activation globally inhibits cellular lipid levels.

AMPK activation leads to increased fatty acid oxidation along with decreased fatty acid synthesis. To determine if increased fatty acid oxidation inhibits RVFV infection, we pretreated U2OS cells with the fatty acid oxidation inhibitor etomoxir 1 hour prior to infection with RVFV. Treatment with etomoxir had no effect on RVFV infection

(Figure 4.4C, quantified in 4.4D), indicating that fatty acid oxidation is dispensable for infection. Therefore, AMPK activation leads to a net decrease in cellular lipids, most likely through reduced levels of fatty acid synthesis. The increased levels of cellular lipids in AMPKα1/AMPKα2-/- MEFs may be used during viral infection to allow further enhancement of the replication complex, leading to increased infection. e. Palmitate Rescues AMPK-Mediated Restriction of RVFV

If AMPK activation restricts RVFV infection by reducing levels of fatty acid synthesis, exogenous addition of fatty acids should restore infection. Therefore, we tested whether we could bypass the requirement for AMPK-regulated fatty acid synthesis by pretreating cells with palmitate, the first product of fatty acid biosynthesis.

We treated U2OS cells with palmitate overnight, and then added A769662 1 hour prior to infection with RVFV to activate AMPK. After 10 hours of infection, cells were fixed and stained for RVFV to measure percent infection. In cells treated with the AMPK activator

A769662 alone, we found a 5-fold decrease in RVFV infection, consistent with our

previous findings (Figure 4.5A, quantified in 4.5B). However, addition of palmitate prior

to treatment with A769662 was able to restore infection nearly to levels seen in

untreated cells (Figure 4.5A, quantified in 4.5B). We observed a 5-fold increase in RVFV

infection in cells treated with A769662 and palmitate compared to those treated with

A769962 alone (Figure 4.5B). Addition of palmitate alone had little effect on infection in

cells. (Figure 4.5A-B). Taken together, these data suggest that AMPK restricts RVFV

infection primarily through inhibiting acetyl CoA carboxylase, and fatty acid biosynthesis.

f. AMPK Restricts VSV and Sindbis Virus

The requirement for lipid synthesis during infection is not unique to Bunyaviruses.

Since many RNA viruses that replicate in the cytoplasm require extensive membrane

modifications and proliferations to support their replication complex [129,162], we tested

whether AMPK activation is able to restrict other viruses. To this end we tested the

ability of the Rhabdovirus Vesicular Stomatitis Virus (VSV) and the Alphavirus Sindbis to

grow in wild type and AMPKα1/AMPKα2-/- MEFs by immunofluorescense. Both VSV

(Figure 4.6A-B) and Sindbis (Figure 4.6C-D) had a 2-3 fold higher infection in

AMPKα1/AMPKα2-/- MEFs compared to wild type MEFs. Moreover, both VSV (Figure

4.6E-F) and Sindbis (Figure 4.6G-H) also had a 2-3 fold increase in percent infection in

LKB1-/-; Vec compared to MEFs expressing LKB1, indicating that both AMPK and its

canonical upstream activator LKB1 restrict additional viruses besides RVFV. To further

investigate the role of AMPK signaling in VSV infection, we performed and one-step

growth curve, and found a 2 fold increase in infectious virus production in

AMPKα1/AMPKα2-/- MEFs compared to wild type MEFs (Figure 4.6I). Finally, VSV infection was inhibited in U2OS cells pretreated with AMPK activating drugs 2DG and

80 oligomycin (Figure 4.6J). Taken together our data suggest that AMPK is broadly antiviral across disparate virus families, and may represent a novel cellular target for antiviral therapeutics.

3. Discussion

Arboviruses represent a group of emerging pathogens of both medical and agricultural importance for which there are few therapies. RVFV is a particularly important member of this group that causes disease both in humans and livestock, and is considered a Category A pathogen due to its potential for geographical spread. Here, we identified AMPK as a novel antiviral factor that restricts RVFV infection. This restriction required the upstream activator LKB1. AMPK activity regulates viral infection by reducing fatty acid biosynthesis, an essential process in RVFV infection. We extended these studies and found that additional RNA viruses that are known to require lipid biosynthesis were also restricted by this pathway. In addition, treatment with drugs that activate AMPK restricted infection, suggesting a novel therapeutic strategy for infections with RNA viruses.

AMPK is a central regulator of cellular energy, and regulates numerous cellular mechanisms. Evidence suggests that AMPK may be the only physiologically relevant regulator of ACC and fatty acid synthesis [228]. AMPK regulates translation at least in part through inhibiting eEF2; however our data suggest that there are additional regulators of this pathway since eEF2 phosphorylation was observed in the absence of

AMPK, while ACC phosphorylation was undetectable in AMPK deficient cells (Figure

4.4A). Therefore, AMPK is a very potent regulator of ACC and fatty acid synthesis, making it likely that this regulation is responsible for the anti-viral effects of AMPK on

81 viruses that require fatty acid synthesis. ACC regulates lipid metabolism through a number of different mechanisms mediated by production of malonyl CoA . Malonyl CoA is important in converting simple essential fatty acids into more complex polyunsaturated fatty acids that can be used to build triglycerides and other cellular lipids. It also inhibits fatty acid oxidation by inhibiting transport of fatty acids to mitochondria through CPT1, and finally malonyl CoA is a substrate for fatty acid synthesis, which drives de novo palmitate production. [231]. Our finding that the CPT1 inhibitor had no effect on infection, while addition of palmitate was able to rescue RVFV and overcome the restriction mediated by AMPK-activating drugs (Figure 4.5) suggests that the ability of AMPK to inhibit fatty acid biosynthesis is the most important determinant of AMPK-mediated

RVFV restriction. Therefore, tight control of fatty acid biosynthesis by AMPK is an essential component of cellular intrinsic antiviral responses.

Importantly, in addition to RVFV, we also found that the Rhabdovirus VSV and the Alphavirus Sindbis are also restricted by AMPK and LKB1 (Figure 4.6), and AMPK has been implicated to play a role in infections of Hepatitis C virus (HCV). AMPK- activating drugs inhibit the replication of RNA replicons concomitant with a decrease in cellular lipid levels, while knock down of the upstream activator LKB1 led to increased replication [291]. Moreover, addition of AMPK-activating drugs such as Metformin to current HCV treatment regimens had promising, albeit modest, effects on reducing viral loads [278,279]. Like RVFV, VSV, and Sindbis, HCV replication occurs on modified cellular membranes, in this case forming a membranous web derived from intracellular vesicles, which requires fatty acid synthesis [164,165]. Furthermore, treatment of AMPK- activating drugs was able to inhibit infectin of HCMV and HIV, consistent with an anti-

82 viral role for AMPK [333,334]. Taken together, our evidence suggests a novel and broad antiviral restrictive mechanism controlled by the energy sensor AMPK.

Cellular anti-viral defense incorporates multi-faceted mechanisms to nonspecifically protect the host cell from numerous viral infections. While each family of virus has different characteristics and strategies for infection, commonalities exist across a broad range of disparate viruses. Many essential features of virus lifecycles are conserved across different families, allowing the host immune system to restrict multiple viruses by targeting conserved required mechanisms. One such mechanism broadly conserved across RNA viruses is membrane proliferation to support the viral replication complex. AMPK is a potent inhibitor of fatty acid synthesis required for these membrane proliferations, and plays a role in restricting viruses through this mechanism. Since

AMPK activators are currently in the clinic to treat metabolic disorders including type II diabetes [261], perhaps they may provide useful therapeutics against a broad range of viral infections.

Figure 4.1: AMPK expression inhibits RVFV infection. A. Plaque assays were performed on wild type (WT) and AMPKα2 -/- MEFs. Representative data from triplicate experiments is shown. B. Quantification of plaques from A. presented as the normalized mean+SD of wild type plaques from three experiments;* indicates p<0.05. C. The diameter of 30 representative plaques in each duplicate well from A. was used to calculate the average plaque area, which is displayed as the normalized mean+SD in triplicate experiments; * indicates p<0.05. D. WT or AMPKα1/AMPKα2 -/- MEFs were infected with serial dilutions of RVFV, spinoculated 1 hour at 1200 RPM, incubated for 16 hours to allow viral growth, and processed for immunofluoresence. (RVFV, green; nuclei, blue). E. Quantification of D. presented as percent of infected cells. A representative of three experiments is shown. F.Time course of RVFV infection in WT and AMPKα1/AMPKα2 -/- MEFs. Cells were infected with RVFV, spinoculated, and fixed at indicated hours post infection. (RVFV, green; nuclei, blue) A representive of triplicate experiments is shown G. One-step growth curve of RVFV in WT or AMPKα1/AMPKα2 -/- MEFs. RVFV grown in WT or AMPKα1/AMPKα2 -/- MEFs for 4, 8, or 12 hours was tittered on BHK cells.

* *

Figure 4.2: LKB1 expression inhibits RVFV infection: A. RVFV was plaqued on LKB1-/-;LKB1 and LKB1-/-;Vec MEFs. Representative data from triplicate experiments is shown. B. Quantification of plaques from A. presented as the normalized mean+SD of wild type plaques from three experiments;* indicates p<0.05. C. The diameter of 30 representative plaques in each duplicate well from A. was used to calculate the average plaque area, which is displayed as the normalized mean+SD in triplicate experiments; * indicates p<0.05. D. RVFV percent infection in LKB1-/-;LKB1 and LKB1-/-;Vec MEFs. MEFs were infected with serial dilutions of RVFV, spinoculated 1 hour at 1200 RPM, incubated for 16 hours to allow viral growth, and processed for immunofluoresence. A representative of triplicate experiments is shown. E. Time course of RVFV infection in LKB1-/-;LKB1 and LKB1-/-;Vec MEFs. Cells were infected with RVFV, spinoculated, and fixed at indicated hours post infection. A representive of triplicate experiments is shown.

Figure 4.3: AMPK activation restricts RVFV. A-B. U2OS cells were pretreated with 10uM 2DG, 10uM oligomycin or PBS (untreated) for 1 hour and infected with serial dilutions of RVFV (A) for 10 hours after spinoculation or vaccinia (B) for 8 hours and processed for immunofluorescense. C. U2OS cells were pretreated with 12 mM 2DG, 100 uM A769662, or PBS for 1 hour and infected with RVFV (MOI 1) for 10 hours after spinoculation. D. WT and AMPKα1/AMPKα2 -/- MEFs were pretreated with 100 uM A769662 or PBS for 1 hour and infected with RVFV (MOI 1) for 10 hours after spinoculation and processed for immunofluorescence. Data are displayed as the normalized percent infection relative to the untreated control at MOI 1.25 +SD in triplicate experiments; * indicates p<0.05.

Figure 4.4: Acetyl CoA Carboxylase activity is inhibited by AMPK expression and activation. A. Phosphorylation of AMPK and downstream effectors in WT and AMPKα1/AMPKα2 -/- MEFs. Cells were treated with AMPK activators 2DG (12 mM), oligomycin (OM, 10 uM), and A769662 (100 uM) for 4 hours. Lysates were collected and assayed by immunoblot for phospho-AMPK, phospho-ACC, and phospho-eEF2 Total protein was assayed for each to ensure equal levels of expression and loading. Tubulin was measured as a loading control. B. Lipid levels in MEFs with or without AMPK activation. WT and AMPKα1/AMPKα2 -/- MEFs were treated with 100 uM A769662 for 10 hours and stained for lipids with BODIPY lipophilic fluorescent dye. (BODIPY, red; nuclei, blue). C. WT and AMPKα1/AMPKα2 -/- MEFs were pretreated with the fatty acid oxidation inhibitor Etomoxir for 1 hour, infected with RVFV by spinoculation for 10 hours, and processed for immunofluorescence. (RVFV, green; nuclei, blue) D. Quantification of percent RVFV infection from C. Representative experiments are shown.

Figure 4.5: Addition of palmitate restores RVFV infection in the presence of A769662. A. U2OS cells were pretreated with 100 uM palmitate overnight in low glucose medium supplemented with 5% delipidated fetal calf serum. 100 uM A769662 or PBS was added 1 hour prior to infection with RVFV (MOI 1). Cells were spinoculated, incubated for 10 hours, and processed for immunofluorescence. (RVFV, green; nuclei, blue) B. Quantification of A. Data are displayed as the normalized percent infection relative to the untreated control at MOI 1.25 +SD in triplicate experiments; * indicates p<0.05 compared to untreated vehicle control.

Figure 4.6: VSV and Sindbis are restricted by AMPK. A and C. WT or AMPKα1/AMPKα2 -/- MEFs were infected with serial dilutions of VSV (A) or Sindbis (C), spinoculated 1 hour at 1200 RPM, incubated for 16 hours to allow viral growth, and processed for immunofluoresence. (Virus-GFP, green; nuclei, blue). B and D are quantifications of A and C displayed as percent infection. E and G. LKB1-/-;LKB1 and LKB1-/-;Vec MEFs were infected with serial dilutions of VSV (E) or Sindbis (G), spinoculated 1 hour at 1200 RPM, incubated for 16 hours to allow viral growth, and processed for immunofluoresence. (Virus-GFP, green; nuclei, blue). F and H are quantifications of E and G.Representitives of triplicate experiments are shown. I. One- step growth curve of VSV in WT or AMPKα1/AMPKα2 -/- MEFs. VSV grown in WT or AMPKα1/AMPKα2 -/- MEFs for 4, 8, or 12 hours was tittered on BHK cells. H. U2OS cells were pretreated with 10uM 2DG, 10uM oligomycin or PBS (untreated) for 1 hour and infected with serial dilutions of VSV for 10 hours after spinoculation and processed for immunofluorescense. Data are displayed as the normalized percent infection relative to the untreated control at MOI 10 +/-SD in triplicate experiments; * indicates p<0.05 compared to untreated vehicle control.

V. CONCLUDING REMARKS

Host interactions play crucial roles in determining the outcome of viral infection.

These interactions can be required for the virus to complete certain stages of its lifecycle, or can restrict viral infection as part of an immune response program, depending on the specific stratagies a given virus uses for infection. I took an un-biased screening approach to identify host factors that impact vaccinia infection, and ultimately discovered a previously unrecognized role for AMPK in infections of multiple disparate viruses. AMPK is a central regulator of many cellular processes including protein, carbohydrate, and lipid metabolism as well as cytoskeletal regulation, polarity, growth, and proliferation, many of which are important for viral infection. While vaccinia requires

AMPK to modulate the actin cytoskeleton promoting efficient entry (Chapter III), RVFV is restricted by AMPK due to its negative regulation of fatty acid biosynthesis (Chapter IV).

Such complex interactions are not unprecedented. For example, autophagy is required for some viruses during their lifecycle, but has also been shown to have antiviral functions, specifically against VSV and Sindbis Virus [292,335,336,337,338]. Another example is Barrier to Autointegration Factor (BAF), which is required to prevent autointegration of the retroviral preintegration complex, but also plays an antiviral role in poxvirus infection by preventing DNA replication [37,339]. Therefore, the role of host factors during infection is highly dependent on the specific lifecycle requirements of each virus, and the discovery of these functions can increase our understanding of viral as well as cell biology.

1. AMPK Promotes Vaccinia Entry

AMPK is involved in multiple cellular processes that require modulation of the actin and microtubule cytoskeletal networks, including vaccinia virus induced macropinocytosis. While AMPK very quickly becomes phosphorylated upon vaccinia infection, the cellular signaling pathway leading to this activation remains unclear. Since

AMPK is expressed cytoplasmically, and is required prior to internalization of vaccinia virus, it is unlikely that vaccinia directly phosphorylates AMPK. More likely, vaccinia binds to some surface receptor, which initiates a signaling cascade ultimately leading to

AMPK phosphorylation. Epidermal Growth Factor Receptor (EGFR) has recently been identified as a receptor important for vaccinia entry through macropinocytosis [84].

EGFR activation is known to stimulate Rac1-mediated actin modifications such as filament stabilization and Arp2/3-dependent actin polymerization that contribute to macropinocytosis [87]. Since Rac1 mobilization during PMA treatment is defective in

AMPK deficient MEFs (Figure 3.7B) and Rac1 activation downstream of AMPK has been reported [314,315], it would be interesting to determine if AMPK is phosphorylated in response to EGFR stimulation, and if EGFR-induced actin dynamics are defective in

AMPK deficient cells. This would provide a link between engagement of a receptor by vaccinia, downstream activation of AMPK, and induction of macropinocytosis.

The immediate upstream activator of AMPK under these conditions remains unclear. I have shown that neither LKB1 nor CaMKKβ are required for vaccinia infection

(Figure 3.4), suggesting that AMPK phosphorylation occurs through some other route.

This is in contrast with previous polarity studies that have demonstrated that LKB1 is an important mediator of cell polarity at least in part through AMPK signaling

[175,236,237,242,243]. Recent studies with the AMPK activator A769662 have indicated that expression of LKB1 or CaMKKβ is required for A769662-induced phosphorylation of

AMPK, but the specific upstream activator was unimportant and may vary in different cell types depending on which is highly expressed [330]. LKB1 activity is believed to be constitutive, with dephosphorylation by protein phosphatase 2Calpha being most strongly responsible for its regulation of AMPK [183,340,341]. It is possible that LKB1 and CaMKKβ play redundant roles in vaccinia-mediated AMPK phosphorylation so that interfering with one of the upstream activators still allowed phosphorylation by the other.

Dually inhibiting both LKB1 and CaMKKβ during vaccinia infection could clear up this issue. Alternatively it is possible that AMPK is phosphorylated by TAK1 or a novel upstream activator in response to vaccinia infection.

Rho-family GTPases regulate actin cytoskeletal structures in response to external stimuli. Specifically, Rac1 is involved in membrane ruffling, wound healing and macropinocytosis discussed in Chapter III. I found a defect in Rac1 mobilization to the plasma membrane in response to PMA treatment in AMPK deficient MEFs (Figure 3.7B), indicating that AMPK is important in signaling this event. Studies have shown that Rac1 is activated in response to vaccinia infection, and this activation is required for vaccinia entry through macropinocytosis [83]. However the virus strain used in these studies was

WR, as I used in Chapter III. A different strain of vaccinia IHDJ employs a different type of macropinocytosis in which stable filopodia are formed during infection, rather than the transient actin blebs induced by WR [84]. Moreover, while both virus strains induced activation of both Rac1 and Cdc42, WR entry was dependent on Rac1, while IHDJ entry was dependent on Cdc42 [84]. Since the form of macropinocytosis employed by IHDJ is clearly different from WR and likely involves different upstream signaling events, it would

92 be interesting to determine if IHDJ entry is also dependent on AMPK, as I would expect if AMPK is specifically involved in Rac1-mediated actin modulation, rather than Cdc42. I found that like vaccinia WR, cowpox virus requires AMPK for efficient infection (Figure

3.3D). The mechanism of cowpox entry has not been well characterized, so it is unclear if this virus enters by macropinocytosis, and if its entry is dependent on Rac1 or Cdc42.

These studies may help to further elucidate the role of AMPK in macropinocytosis and poxvirus infection.

During infection several mature forms of vaccinia are produced. Mature virus

(MV) consists of the virus core surrounded by one lipid bilayer envelope that contains associated membrane proteins including the vaccinia entry-fusion complex [77]. MV remains in the cytoplasm of an infected cell, and is released only upon cell lysis. This is the more stable form of the virus believed to be responsible for host to host transmission. Extracellular virus (EV) consists of an MV particle wrapped in an additional

Golgi-derived membrane studded with a different set of viral proteins. EV is released from the cell by fusion at the plasma membrane, and can either disseminate or reassociate with the cell (CEV) through interactions with the actin cytosketon to propel itself on actin tails [11]. Studies have suggested that MV and EV may use different strategies for entry since requirements and kinetics of entry were different [78]. CEV entry is not well understood [77]. It would be interesting to investigate the role of AMPK in entry of the EV form of the virus in order to compare the importance of this host factor in both mature forms of the virus.

Since EV particles interact with the actin cytoskeleton upon egress to form CEV, it is possible that AMPK could play a role in this process as well. This idea is further

93 supported by Figure 3.3A-C in which I show both a decrease in plaque number and plaque size in cells lacking AMPK. The decrease in plaque number is likely due to decreased infectivity resulting from inefficient entry in the absence of AMPK. The decrease in plaque size could be due to decreased virus production or inefficient spread.

Although fewer cells become infected in cells lacking AMPK, those that do become infected are able to complete all replicative stages of vaccinia infection (Supplementary

Figure 11A), making decreased virus production less likely. Furthermore, in a one step growth curve, I found approximately a 2-fold reduction in vaccinia produced in AMPK deficient cells, consistent with our observed 2-fold reduction in vaccinia entry

(Supplementary Figure 11B). Since there was no further decrease in vaccinia production, I believe virus production to be comparable in WT and AMPK deficient cells.

Therefore, it is likely that the small plaque size observed in these cells is due to decreased spread. This decrease in spread could result from an entry block in subsequent rounds of infection, or it could result from an inability of CEV to form actin tails. Since actin modulation is required for tail formation and vaccinia spread, AMPK could play a role in this process as well as in entry, and it would be interesting to investigate this possibility.

2. AMPK Restricts RNA Viruses

While vaccinia virus requires AMPK for efficient infection, several RNA viruses examined in this study were restricted by AMPK expression and activation. RVFV was very dramatically inhibited, while VSV and Sindbis virus were moderately restricted

(Figure 4.1 and 4.6). I found that this restriction was dependent on the upstream activator LKB1 (Figure 4.2 and 4.6), and signaling mediated through low energy

94 conditions was able to control infection (Figure 4.3). This is in contrast to our studies with

Vaccinia virus, in which LKB1 deficiency or treatment with energy-lowering AMPK- activating compounds had no effect on infection (Figure 3.4). Furthermore, while AMPK is required for vaccinia entry through macropinocytosis, VSV and Sindbis enter cells through clathrin-mediated endocytosis and RVFV enters through a Rab5 and Rab7 pH- dependent endocytic route that has not been well characterized [342]. I tested the ability of cells lacking AMPK take up transferrin through clathrin-mediated endocytosis, and found that this endocytic route is intact in AMPK deficient cells and has similar efficiency to WT MEFs (Supplementary Figure 12). Therefore, it is clear that the role of AMPK in

RNA virus infection occurs through a very different mechanism than I observed for

Vaccinia virus. The upstream activation pattern is distinct, as well as the downstream mechanism of viral action.

I have shown that vaccinia virus induces AMPK phosphorylation (Figure 3.4C). It is unclear if infection with these RNA viruses induces a similar increase in phosphorylation. Studies with the RNA virus HCV have indicated that expression of HCV proteins from a stably-expressed replicon inhibits AMPK activity through an Akt- mediated pathway in hepatocytes [291]. Since HCV infection is also inhibited by AMPK activation, it is likely that the virus encodes some mechanism to counteract the negative effects of AMPK by promoting dephosphorylation, but it is unclear which viral proteins are responsible for this effect. I have some data to suggest that VSV is also able to inhibit phosphorylation of AMPK or its downstream target ACC (data not shown), but this idea must be further explored.

It is unclear if AMPK activity provides a constitutive anti-viral presence since there is some level of basal AMPK phosphorylation, or if AMPK is specifically induced upon viral infection. Distinguishing between these possibilities could be difficult if these viruses do indeed encode mechanisms to inhibit AMPK phosphorylation. Testing individual non-infectious components, such as UV-inactivated virus particles, virus like particles, the ribonucleoprotein, or the naked RNA genome may help to resolve this issue. Infection is likely to introduce new stresses and energy demands on the cell, which may cause a local, if not global increase in AMPK phosphorylation, leading to viral restriction. This could be determined by measuring cellular energy levels during infection. Alternatively, it is possible that AMPK becomes activated in response to some viral sensing program. TAK1 has been implicated as an upstream activator of AMPK

[190], and also participates in activation of innate immune signaling downstream of toll- like receptors and the cytoplasmic NOD sensors [193,194,195,196]. It is therefore possible that recognition by some of these sensors could lead to AMPK activation. While treatment with LPS has shown a decrease in AMPK phosphorylation in certain cell types, agonists to other sensors have not been tested. It would be interesting to treat cells with a variety of sensor agonists, and examine the effect of treatment on AMPK phosphorylation. This may help to determine if innate immune signaling pathways contribute to AMPK activation and viral restriction.

Taken together with published reports, evidence suggests that members of at least four different RNA virus families may be restricted by AMPK. I have shown that restriction of RVFV is mediated through AMPK’s inhibition of fatty acid synthesis, and correlative studies indicate the same may be true for HCV [291]. It will be important to further investigate the role of AMPK in VSV and Sindbis infection to determine if

96 restriction occurs by the same mechanism for these viruses. Furthermore, it will be important to test additional disparate viruses to examine the full range of AMPK mediated restriction. RVFV, VSV, SINV, and HCV are all enveloped single-stranded

RNA viruses that replicate in the cytoplasm. These viruses require manipulation of cellular lipids at two distinct stages of their lifecycle. They rearrange and proliferate internal cellular membranes into vesicle-like structures to support their replication complex, and also require cellular lipids during assembly when they are encapsulated in a cellularly-derived lipid envelope. It is unclear whether AMPK activation interferes with the replication complex or encapsulation steps or both. While non-enveloped viruses such as Polio Virus do not require this membrane encapsulation step, many still require membrane rearrangements to support their replication complex. Determining if non- enveloped viruses are also restricted by AMPK could help to provide an answer.

Additional studies with viruses that replicate in the nucleus such as Influenza virus could further elucidate the spectrum of viruses that are restricted by AMPK.

Several classes of AMPK-activating pharmaceuticals are already in use for treatment of type II diabetes. These drugs may also have uses as antiviral therapeutics, but first it will be important to establish if they have antiviral effects in vivo. Metformin and TZDs are believed to activate AMPK indirectly and have other targets during treatment. Since they are associated with negative side effects, development of more specific AMPK activators to treat viral infection may be beneficial. Therefore it is important to investigate if AMPK activation itself is antiviral in vivo. To perform this study, mice should be treated with the specific AMPK activator A769662 along with clinically approved anti-diabetic drugs that activate AMPK and infected with a panel of viruses known to be restricted by AMPK. Finally, since groups of diabetic patients have been

97 closely monitored, it may be possible to perform epidemiological studies to determine if patients receiving Metformin or TZD treatment show decreased incidence of viral infection. Similar studies have been already been used to link Metformin treatment with decreased risk of cancer, and could also be beneficial in investigating infectious disease.

These studies could be very helpful in developing broadly anti-viral therapeutics that target AMPK.

VI. APPENDIX: SUPPLEMENTARY FIGURES

Supplementary Figure 1: Vaccinia infection in Drosophila cells. A. Drosophila DL1 cells were infected with vaccinia virus expressing B-gal driven by an early/late promoter (p7.5) for indicated time, and stained for X-gal production. A representative of 2 experiments is shown. B. Titration of vaccinia infection in Drosophila cells seeded in 384 well plates. Cells were fixed and processed 48 hpi and stained for early B-gal expression (green) and nuclei (blue). C. Quantification of B. Percent infection is the average of 6 wells, with 3 images per well in duplicate experiments. Bars represent average percent infection for each experiment.

Supplementary Figure 2: Inhibitors of macropinocytosis inhibit vaccinia infection in mammalian and Drosophila cells. A. Human U2OS cells were pretreated with: Latrunculin A (Lat A, 5μM), Wortmannin (Wort, 5μM), Rottlerin (10μM), or EIPA (12.5μM) for 1 hour, challenged with vaccinia (MOI=10) for 8 hours, and quantified for percent infection. B. Drosophila DL1 cells were treated with: Latrunculin A (Lat A, 5μM), Wortmannin (Wort, 5μM), and Rottlerin (5μM), or EIPA (50μM) for 1 hour and challenged with vaccinia (MOI=20) for 24 hours. Cells were fixed and processed for immunofluorescence using E3L expression as a marker for infection, and Hoescht 33342 to visualize nuclei. Mean percent infection + SD in triplicate experiments is shown; * indicates p<0.05 compared to control in three independent experiments.

100

Supplementary Figure 3: Validation of eight candidates that promote vaccinia infection identified in RNAi screen of Drosophila kinases and phosphatases. Independent dsRNA targeting different sequences of each candidate gene were tested, and percent infection was determined by immunofluorescence measuring B-gal expressing cells. Luciferase was used as a nontargeting negative control. B-gal and Rab5 were added as positive controls for decreased infection. A representative of duplicate experiments is shown. Error bars represent standard deviation of 12 different wells with 3 images taken per well. * indicate p-value of <0.001 in both experiments.

101

Supplementary Figure 4: dsRNA against AMPKα or LKB1 leads to depletion of the cognate mRNA in Drosophila cells. RNAi was performed against luciferase (luc) or AMPKα (A) or LKB1 (B) in Drosophila cells. RNA was collected from lysates and RT- PCR was performed to measure mRNA levels. A 1:10 dilution of control cDNA (luc) was included to demonstrate that the depletion was greater than 10-fold. Clathrin heavy chain (chc) was used as a loading control.

102

Supplementary Figure 5: AMPKα1/AMPKα2 -/- MEFs do not express AMPKα. Wild type or AMPKα1/AMPKα2 -/- MEF protein lysates were collected and probed by immunoblot for total-AMPKα and tubulin.

103

Supplementary Figure 6: AMPK promotes efficient cowpox virus infection. Plaque assays were performed on wild type or AMPKα1/AMPKα2 -/- MEFs and quantified in duplicate experiments. Error bars show the individual values; * p<0.05 in each replicate.

104

Supplementary Figure 7: AMPK is not required for Vesicular Stomatitis Virus infection. Plaque assays were performed on wild type or AMPKα1/AMPKα2 -/- MEFs. There was no decrease in plaque number observed in the mutant cells. A representative experiment of three is shown.

105

Supplementary Figure 8: AMPK promotes early vaccinia infection in mammalian cells. A. Lack of AMPKα leads to decreased vaccinia infectivity in MEFs. Wild type or AMPKα1/AMPKα2 -/- MEFs were infected with the indicated MOI for 8 hours and processed for immunofluorescence. Data is displayed as average percentage of infected cells for a representative experiment. B. Loss of AMPKα leads to a decrease in viral mRNA production in AMPKα1/AMPKα2 -/- MEFs. Northern blot of viral mRNA levels in WT or AMPKα1/AMPKα2 -/- MEFs at indicated times post infection (MOI=10). Blots were probed for virally encoded E3L or a ribosomal RNA loading control. C. Wild type or AMPKα1/AMPKα2 -/- cells were infected (MOI=10) for the indicated times and probed for E3L by immunoblot

106

Supplementary Figure 9: siRNA targeting AMPK inhibits vaccinia infection in mammalian cells. A. U2OS cells were treated with non-targeting siRNA (siCON) or siRNA targeting AMPKα1 or AMPKα2 and infected with vaccinia virus (MOI 10), and stained for E3L expression after 8 hours. B. Quantification of percent infection from A. The average of duplicate experiments; error bars represent mean of the percent infection for each experiment. C. Western blot probing total AMPKα after siRNA treatment.

107

Supplementary Figure 10: LKB1 cDNA rescues the LKB1 null cells. LKB1 -/- MEFs were complemented with a vector control (Vec) or FLAG-LKB1 (LKB1) cDNA and were mock treated, or treated with 2-deoxyglucose (2DG) which leads to LKB1-dependent AMPK phosphorylation for 30 min. Protein lysates were collected and probed by immunoblot for FLAG, phospho- or total-AMPKα expression.

108

Supplementary Figure 11: Vaccinia produced in AMPK deficient cells is infectious. A. AMPK is not required for late vaccina protein expression. WT and AMPKα1/AMPKα2 - /- MEFs infected with vaccinia for 8 hours were stained for early (E3L, green) and late (L1R, red) vaccinia protein expression. B. Infectious virus is produced in AMPK deficient cells. Vaccinia grown for 12 hours in WT and AMPKα1/AMPKα2 -/- MEFs was titered in BSC-1 cells. Rifampicin (Rif) was added as a control for detecting incoming virus. The relative pfu/ml in BSC-1 cells was graphed as the mean + standard deviation of triplicate experiments. The decrease in virus produced was similar to decrease in virus entry.

109

Supplementary Figure 12: AMPK deficient cells undergo efficient receptor- mediated endocytosis. Transferrin uptake assays were performed in the presence or absence of virus. Wild type and AMPKα1/AMPKα2 -/- MEFs were either infected or mock-infected and treated with 594-transferrin (red), processed for microscopy, and stained to visualize actin (phalloidin (green)) and nuclei (Hoescht 33342 (blue)). Representative images from triplicate experiments are shown.

110

Supplementary Figure 13: AMPK deficient cells are defective in lamellipodia formation during wound healing. A scratch was made in a confluent monolayer of wild type or AMPKα1/AMPKα2 -/- MEFs, and monitored over time. Images were taken using a 20X and 63X objective immediately after wounding (T=0), and again after 3 and 6 hours to determine the morphology of the cells at the wound front. Polarized cells with lamellipodia are visible at the wound front of wild type MEFs (arrows). Representative images from triplicate experiments are shown.

111

Supplementary Figure 14: Cellular motility is LKB1-independent. Scratches were introduced into a confluent monolayer of LKB1 -/- , Vec or LKB1 -/-; LKB1 cDNA MEFs, and monitored over time for closure. Representative images from triplicate experiments are shown immediately after wounding (T=0) and after 12 or 24 hours. The reduction in wound width is quantified over time. Data are normalized to initial would width at T=0, and presented as means of three independent experiments with four wounds per set.

112

90 80 70 60 50 untreated 40 Rapamycin 30 20 10 0 WT AMPK ko

Supplementary Figure 15: mTORC1 is not required for AMPK-mediated restriction of RVFV. WT and AMPKα1/AMPKα2 -/- MEFs were pretreated with 10 nM Rapamycin or PBS for 1 hour and infected with RVFV (MOI 1) for 10 hours after spinoculation and processed for immunofluorescence. A representative of duplicate experiments is shown.

113

Supplementary Figure 16: RVFV is not restricted by autophagy. Plaque assays were performed with RVFV on MEFs stably expressing a hairpin RNA targeting Atg5 or a control hairpin.

114

VII. REFERENCES

1. Mohr S, Bakal C, Perrimon N Genomic screening with RNAi: results and challenges. Annu Rev Biochem 79: 37‐64. 2. Echeverri CJ, Perrimon N (2006) High‐throughput RNAi screening in cultured cells: a user's guide. Nat Rev Genet 7: 373‐384. 3. Perrimon N, Friedman A, Mathey‐Prevot B, Eggert US (2007) Drug‐target identification in Drosophila cells: combining high‐throughout RNAi and small‐molecule screens. Drug Discov Today 12: 28‐33. 4. Veraksa A, Del Campo M, McGinnis W (2000) Developmental patterning genes and their conserved functions: from model organisms to humans. Mol Genet Metab 69: 85‐100. 5. Boutros M, Kiger AA, Armknecht S, Kerr K, Hild M, et al. (2004) Genome‐wide RNAi analysis of growth and viability in Drosophila cells. Science 303: 832‐835. 6. Cherry S (2008) Genomic RNAi screening in Drosophila S2 cells: what have we learned about host‐pathogen interactions? Curr Opin Microbiol 11: 262‐270. 7. Cherry S, Perrimon N (2004) Entry is a rate‐limiting step for viral infection in a Drosophila melanogaster model of pathogenesis. Nat Immunol 5: 81‐87. 8. Cherry S, Doukas T, Armknecht S, Whelan S, Wang H, et al. (2005) Genome‐wide RNAi screen reveals a specific sensitivity of IRES‐containing RNA viruses to host translation inhibition. Genes Dev 19: 445‐452. 9. Sabin LR, Zhou R, Gruber JJ, Lukinova N, Bambina S, et al. (2009) Ars2 regulates both miRNA‐ and siRNA‐ dependent silencing and suppresses RNA virus infection in Drosophila. Cell 138: 340‐351. 10. Weaver SC, Barrett AD (2004) Transmission cycles, host range, evolution and emergence of arboviral disease. Nat Rev Microbiol 2: 789‐801. 11. Moss B (2001) Poxviridae: The viruses and their replication. In: Howley DMKaPM, editor. Fields Virology. Philadelphia: Lippincott, Williams, and Wilkins. pp. 2849‐2884. 12. McFadden G (2005) Poxvirus tropism. Nat Rev Microbiol 3: 201‐213. 13. McFadden G Killing a killer: what next for smallpox? PLoS Pathog 6: e1000727. 14. Essbauer S, Pfeffer M, Meyer H Zoonotic poxviruses. Vet Microbiol 140: 229‐236. 15. Gubser C, Hue S, Kellam P, Smith GL (2004) Poxvirus genomes: a phylogenetic analysis. J Gen Virol 85: 105‐117. 16. Upton C, Slack S, Hunter AL, Ehlers A, Roper RL (2003) Poxvirus orthologous clusters: toward defining the minimum essential poxvirus genome. J Virol 77: 7590‐7600. 17. Werden SJ, Rahman MM, McFadden G (2008) Poxvirus host range genes. Adv Virus Res 71: 135‐171. 18. Wang F, Ma Y, Barrett JW, Gao X, Loh J, et al. (2004) Disruption of Erk‐dependent type I interferon induction breaks the myxoma virus species barrier. Nat Immunol 5: 1266‐ 1274. 19. Johnston JB, Nazarian SH, Natale R, McFadden G (2005) Myxoma virus infection of primary human fibroblasts varies with cellular age and is regulated by host interferon responses. Virology 332: 235‐248. 20. Lefkowitz EJ, Wang C, Upton C (2006) Poxviruses: past, present and future. Virus Res 117: 105‐118.

115

21. Xing K, Deng R, Wang J, Feng J, Huang M, et al. (2006) Genome‐based phylogeny of poxvirus. Intervirology 49: 207‐214. 22. Vora S, Damon I, Fulginiti V, Weber SG, Kahana M, et al. (2008) Severe eczema vaccinatum in a household contact of a smallpox vaccinee. Clin Infect Dis 46: 1555‐1561. 23. Silva DC, Moreira‐Silva EA, Gomes Jde A, Fonseca FG, Correa‐Oliveira R Clinical signs, diagnosis, and case reports of Vaccinia virus infections. Braz J Infect Dis 14: 129‐134. 24. Lederman E, Miramontes R, Openshaw J, Olson VA, Karem KL, et al. (2009) Eczema vaccinatum resulting from the transmission of vaccinia virus from a smallpox vaccinee: an investigation of potential fomites in the home environment. Vaccine 27: 375‐377. 25. Hughes CM, Blythe D, Li Y, Reddy R, Jordan C, et al. Vaccinia Virus Infections in Martial Arts Gym, Maryland, USA, 2008. Emerg Infect Dis 17: 730‐733. 26. (2009) Progressive vaccinia in a military smallpox vaccinee ‐ United States, 2009. MMWR Morb Mortal Wkly Rep 58: 532‐536. 27. (2009) Human vaccinia infection after contact with a raccoon rabies vaccine bait ‐ Pennsylvania, 2009. MMWR Morb Mortal Wkly Rep 58: 1204‐1207. 28. Vaccinia virus infection after sexual contact with a military smallpox vaccinee ‐ Washington, 2010. MMWR Morb Mortal Wkly Rep 59: 773‐775. 29. Broyles SS (2003) Vaccinia virus transcription. J Gen Virol 84: 2293‐2303. 30. Tolonen N, Doglio L, Schleich S, Krijnse Locker J (2001) Vaccinia virus DNA replication occurs in endoplasmic reticulum‐enclosed cytoplasmic mini‐nuclei. Mol Biol Cell 12: 2031‐2046. 31. Carter GC, Rodger G, Murphy BJ, Law M, Krauss O, et al. (2003) Vaccinia virus cores are transported on microtubules. J Gen Virol 84: 2443‐2458. 32. Schramm B, de Haan CA, Young J, Doglio L, Schleich S, et al. (2006) Vaccinia‐virus‐induced cellular contractility facilitates the subcellular localization of the viral replication sites. Traffic 7: 1352‐1367. 33. Katsafanas GC, Moss B (2007) Colocalization of transcription and translation within cytoplasmic poxvirus factories coordinates viral expression and subjugates host functions. Cell Host Microbe 2: 221‐228. 34. Sanderson CM, Way M, Smith GL (1998) Virus‐induced cell motility. J Virol 72: 1235‐1243. 35. Culyba MJ, Harrison JE, Hwang Y, Bushman FD (2006) DNA cleavage by the A22R resolvase of vaccinia virus. Virology 352: 466‐476. 36. Druck Shudofsky AM, Silverman JE, Chattopadhyay D, Ricciardi RP Vaccinia virus D4 mutants defective in processive DNA synthesis retain binding to A20 and DNA. J Virol 84: 12325‐12335. 37. Wiebe MS, Traktman P (2007) Poxviral B1 kinase overcomes barrier to autointegration factor, a host defense against virus replication. Cell Host Microbe 1: 187‐197. 38. Li Y, Hall RL, Moyer RW (1997) Transient, nonlethal expression of genes in vertebrate cells by recombinant entomopoxviruses. J Virol 71: 9557‐9562. 39. Bengali Z, Satheshkumar PS, Yang Z, Weisberg AS, Paran N, et al. Drosophila S2 cells are non‐permissive for vaccinia virus DNA replication following entry via low pH‐ dependent endocytosis and early transcription. PLoS One 6: e17248. 40. Rosales R, Sutter G, Moss B (1994) A cellular factor is required for transcription of vaccinia viral intermediate‐stage genes. Proc Natl Acad Sci U S A 91: 3794‐3798.

116

41. Gunasinghe SK, Hubbs AE, Wright CF (1998) A vaccinia virus late transcription factor with biochemical and molecular identity to a human cellular protein. J Biol Chem 273: 27524‐27530. 42. Wright CF, Hubbs AE, Gunasinghe SK, Oswald BW (1998) A vaccinia virus late transcription factor copurifies with a factor that binds to a viral late promoter and is complemented by extracts from uninfected HeLa cells. J Virol 72: 1446‐1451. 43. Katsafanas GC, Moss B (2004) Vaccinia virus intermediate stage transcription is complemented by Ras‐GTPase‐activating protein SH3 domain‐binding protein (G3BP) and cytoplasmic activation/proliferation‐associated protein (p137) individually or as a heterodimer. J Biol Chem 279: 52210‐52217. 44. Broyles SS, Liu X, Zhu M, Kremer M (1999) Transcription factor YY1 is a vaccinia virus late promoter activator. J Biol Chem 274: 35662‐35667. 45. Wright CF, Oswald BW, Dellis S (2001) Vaccinia virus late transcription is activated in vitro by cellular heterogeneous nuclear ribonucleoproteins. J Biol Chem 276: 40680‐40686. 46. Knutson BA, Liu X, Oh J, Broyles SS (2006) Vaccinia virus intermediate and late promoter elements are targeted by the TATA‐binding protein. J Virol 80: 6784‐6793. 47. Condit RC, Moussatche N, Traktman P (2006) In a nutshell: structure and assembly of the vaccinia virion. Adv Virus Res 66: 31‐124. 48. Alzhanova D, Hruby DE (2007) A host cell membrane protein, golgin‐97, is essential for poxvirus morphogenesis. Virology 362: 421‐427. 49. Alzhanova D, Hruby DE (2006) A trans‐Golgi network resident protein, golgin‐97, accumulates in viral factories and incorporates into virions during poxvirus infection. J Virol 80: 11520‐11527. 50. Herrero‐Martinez E, Roberts KL, Hollinshead M, Smith GL (2005) Vaccinia virus intracellular enveloped virions move to the cell periphery on microtubules in the absence of the A36R protein. J Gen Virol 86: 2961‐2968. 51. Smith GL, Vanderplasschen A, Law M (2002) The formation and function of extracellular enveloped vaccinia virus. J Gen Virol 83: 2915‐2931. 52. Ward BM, Moss B (2001) Vaccinia virus intracellular movement is associated with microtubules and independent of actin tails. J Virol 75: 11651‐11663. 53. Ward BM, Moss B (2001) Visualization of intracellular movement of vaccinia virus virions containing a green fluorescent protein‐B5R membrane protein chimera. J Virol 75: 4802‐4813. 54. Newsome TP, Weisswange I, Frischknecht F, Way M (2006) Abl collaborates with Src family kinases to stimulate actin‐based motility of vaccinia virus. Cell Microbiol 8: 233‐241. 55. Reeves PM, Bommarius B, Lebeis S, McNulty S, Christensen J, et al. (2005) Disabling poxvirus pathogenesis by inhibition of Abl‐family tyrosine kinases. Nat Med 11: 731‐ 739. 56. Newsome TP, Scaplehorn N, Way M (2004) SRC mediates a switch from microtubule‐ to actin‐based motility of vaccinia virus. Science 306: 124‐129. 57. Doceul V, Hollinshead M, van der Linden L, Smith GL Repulsion of superinfecting virions: a mechanism for rapid virus spread. Science 327: 873‐876. 58. Marsh M, Helenius A (2006) Virus entry: open sesame. Cell 124: 729‐740. 59. Carter GC, Law M, Hollinshead M, Smith GL (2005) Entry of the vaccinia virus intracellular mature virion and its interactions with glycosaminoglycans. J Gen Virol 86: 1279‐1290.

117

60. Hsiao JC, Chung CS, Chang W (1999) Vaccinia virus envelope D8L protein binds to cell surface chondroitin sulfate and mediates the adsorption of intracellular mature virions to cells. J Virol 73: 8750‐8761. 61. Chung CS, Hsiao JC, Chang YS, Chang W (1998) A27L protein mediates vaccinia virus interaction with cell surface heparan sulfate. J Virol 72: 1577‐1585. 62. Lin CL, Chung CS, Heine HG, Chang W (2000) Vaccinia virus envelope H3L protein binds to cell surface heparan sulfate and is important for intracellular mature virion morphogenesis and virus infection in vitro and in vivo. J Virol 74: 3353‐3365. 63. Senkevich TG, Ward BM, Moss B (2004) Vaccinia virus entry into cells is dependent on a virion surface protein encoded by the A28L gene. J Virol 78: 2357‐2366. 64. Senkevich TG, Ward BM, Moss B (2004) Vaccinia virus A28L gene encodes an essential protein component of the virion membrane with intramolecular disulfide bonds formed by the viral cytoplasmic redox pathway. J Virol 78: 2348‐2356. 65. Senkevich TG, Ojeda S, Townsley A, Nelson GE, Moss B (2005) Poxvirus multiprotein entry‐ fusion complex. Proc Natl Acad Sci U S A 102: 18572‐18577. 66. Townsley AC, Senkevich TG, Moss B (2005) The product of the vaccinia virus L5R gene is a fourth membrane protein encoded by all poxviruses that is required for cell entry and cell‐cell fusion. J Virol 79: 10988‐10998. 67. Townsley AC, Senkevich TG, Moss B (2005) Vaccinia virus A21 virion membrane protein is required for cell entry and fusion. J Virol 79: 9458‐9469. 68. Ojeda S, Domi A, Moss B (2006) Vaccinia virus G9 protein is an essential component of the poxvirus entry‐fusion complex. J Virol 80: 9822‐9830. 69. Ojeda S, Senkevich TG, Moss B (2006) Entry of vaccinia virus and cell‐cell fusion require a highly conserved cysteine‐rich membrane protein encoded by the A16L gene. J Virol 80: 51‐61. 70. Brown E, Senkevich TG, Moss B (2006) Vaccinia virus F9 virion membrane protein is required for entry but not virus assembly, in contrast to the related L1 protein. J Virol 80: 9455‐9464. 71. Bisht H, Weisberg AS, Moss B (2008) Vaccinia virus l1 protein is required for cell entry and membrane fusion. J Virol 82: 8687‐8694. 72. Wagenaar TR, Moss B (2007) Association of vaccinia virus fusion regulatory proteins with the multicomponent entry/fusion complex. J Virol 81: 6286‐6293. 73. Nelson GE, Sisler JR, Chandran D, Moss B (2008) Vaccinia virus entry/fusion complex subunit A28 is a target of neutralizing and protective antibodies. Virology 380: 394‐ 401. 74. Nelson GE, Wagenaar TR, Moss B (2008) A conserved sequence within the H2 subunit of the vaccinia virus entry/fusion complex is important for interaction with the A28 subunit and infectivity. J Virol 82: 6244‐6250. 75. Satheshkumar PS, Moss B (2009) Characterization of a newly identified 35‐amino‐acid component of the vaccinia virus entry/fusion complex conserved in all chordopoxviruses. J Virol 83: 12822‐12832. 76. Wolfe CL, Moss B Interaction between the G3 and L5 proteins of the vaccinia virus entry‐ fusion complex. Virology 412: 278‐283. 77. Moss B (2006) Poxvirus entry and membrane fusion. Virology 344: 48‐54.

118

78. Locker JK, Kuehn A, Schleich S, Rutter G, Hohenberg H, et al. (2000) Entry of the two infectious forms of vaccinia virus at the plasma membane is signaling‐dependent for the IMV but not the EEV. Mol Biol Cell 11: 2497‐2511. 79. Law M, Carter GC, Roberts KL, Hollinshead M, Smith GL (2006) Ligand‐induced and nonfusogenic dissolution of a viral membrane. Proc Natl Acad Sci U S A 103: 5989‐ 5994. 80. Townsley AC, Weisberg AS, Wagenaar TR, Moss B (2006) Vaccinia virus entry into cells via a low‐pH‐dependent endosomal pathway. J Virol 80: 8899‐8908. 81. Vanderplasschen A, Hollinshead M, Smith GL (1998) Intracellular and extracellular vaccinia virions enter cells by different mechanisms. J Gen Virol 79 ( Pt 4): 877‐887. 82. Sieczkarski SB, Whittaker GR (2002) Dissecting virus entry via endocytosis. J Gen Virol 83: 1535‐1545. 83. Mercer J, Helenius A (2008) Vaccinia virus uses macropinocytosis and apoptotic mimicry to enter host cells. Science 320: 531‐535. 84. Mercer J, Knebel S, Schmidt FI, Crouse J, Burkard C, et al. Vaccinia virus strains use distinct forms of macropinocytosis for host‐cell entry. Proc Natl Acad Sci U S A 107: 9346‐9351. 85. Bengali Z, Townsley AC, Moss B (2009) Vaccinia virus strain differences in cell attachment and entry. Virology 389: 132‐140. 86. Whitbeck JC, Foo CH, Ponce de Leon M, Eisenberg RJ, Cohen GH (2009) Vaccinia virus exhibits cell‐type‐dependent entry characteristics. Virology 385: 383‐391. 87. Swanson JA (2008) Shaping cups into phagosomes and macropinosomes. Nat Rev Mol Cell Biol 9: 639‐649. 88. Mercer J, Helenius A (2009) Virus entry by macropinocytosis. Nat Cell Biol 11: 510‐520. 89. Lanzetti L, Palamidessi A, Areces L, Scita G, Di Fiore PP (2004) Rab5 is a signalling GTPase involved in actin remodelling by receptor tyrosine kinases. Nature 429: 309‐314. 90. Bos JL, Rehmann H, Wittinghofer A (2007) GEFs and GAPs: critical elements in the control of small G proteins. Cell 129: 865‐877. 91. Heasman SJ, Ridley AJ (2008) Mammalian Rho GTPases: new insights into their functions from in vivo studies. Nat Rev Mol Cell Biol 9: 690‐701. 92. Jaffe AB, Hall A (2005) Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol 21: 247‐269. 93. Lanciotti RS, Roehrig JT, Deubel V, Smith J, Parker M, et al. (1999) Origin of the West Nile virus responsible for an outbreak of encephalitis in the northeastern United States. Science 286: 2333‐2337. 94. O'Leary DR, Marfin AA, Montgomery SP, Kipp AM, Lehman JA, et al. (2004) The epidemic of West Nile virus in the United States, 2002. Vector Borne Zoonotic Dis 4: 61‐70. 95. Zacks MA, Paessler S Encephalitic alphaviruses. Vet Microbiol 140: 281‐286. 96. Hollidge BS, Gonzalez‐Scarano F, Soldan SS Arboviral encephalitides: transmission, emergence, and pathogenesis. J Neuroimmune Pharmacol 5: 428‐442. 97. Weaver SC, Reisen WK Present and future arboviral threats. Antiviral Res 85: 328‐345. 98. Flick R, Bouloy M (2005) Rift Valley fever virus. Curr Mol Med 5: 827‐834. 99. (2000) Outbreak of Rift Valley fever‐‐Yemen, August‐October 2000. MMWR Morb Mortal Wkly Rep 49: 1065‐1066. 100. (2000) Update: outbreak of Rift Valley Fever‐‐Saudi Arabia, August‐November 2000. MMWR Morb Mortal Wkly Rep 49: 982‐985.

119

101. (2000) Outbreak of Rift Valley fever‐‐Saudi Arabia, August‐October, 2000. MMWR Morb Mortal Wkly Rep 49: 905‐908. 102. Wilson ML (1994) Rift Valley fever virus ecology and the epidemiology of disease emergence. Ann N Y Acad Sci 740: 169‐180. 103. Wilson ML, Chapman LE, Hall DB, Dykstra EA, Ba K, et al. (1994) Rift Valley fever in rural northern Senegal: human risk factors and potential vectors. Am J Trop Med Hyg 50: 663‐675. 104. LaBeaud AD, Muchiri EM, Ndzovu M, Mwanje MT, Muiruri S, et al. (2008) Interepidemic Rift Valley fever virus seropositivity, northeastern Kenya. Emerg Infect Dis 14: 1240‐ 1246. 105. (2007) Rift Valley fever outbreak‐‐Kenya, November 2006‐January 2007. MMWR Morb Mortal Wkly Rep 56: 73‐76. 106. Al‐Hazmi M, Ayoola EA, Abdurahman M, Banzal S, Ashraf J, et al. (2003) Epidemic Rift Valley fever in Saudi Arabia: a clinical study of severe illness in humans. Clin Infect Dis 36: 245‐252. 107. Borio L, Inglesby T, Peters CJ, Schmaljohn AL, Hughes JM, et al. (2002) Hemorrhagic fever viruses as biological weapons: medical and public health management. JAMA 287: 2391‐2405. 108. Madani TA, Al‐Mazrou YY, Al‐Jeffri MH, Mishkhas AA, Al‐Rabeah AM, et al. (2003) Rift Valley fever epidemic in Saudi Arabia: epidemiological, clinical, and laboratory characteristics. Clin Infect Dis 37: 1084‐1092. 109. Caplen H, Peters CJ, Bishop DH (1985) Mutagen‐directed attenuation of Rift Valley fever virus as a method for vaccine development. J Gen Virol 66 ( Pt 10): 2271‐2277. 110. Rossi CA, Turell MJ (1988) Characterization of attenuated strains of Rift Valley fever virus. J Gen Virol 69 ( Pt 4): 817‐823. 111. Vialat P, Muller R, Vu TH, Prehaud C, Bouloy M (1997) Mapping of the mutations present in the genome of the Rift Valley fever virus attenuated MP12 strain and their putative role in attenuation. Virus Res 52: 43‐50. 112. Morrill JC, Mebus CA, Peters CJ (1997) Safety of a mutagen‐attenuated Rift Valley fever virus vaccine in fetal and neonatal bovids. Am J Vet Res 58: 1110‐1114. 113. Morrill JC, Mebus CA, Peters CJ (1997) Safety and efficacy of a mutagen‐attenuated Rift Valley fever virus vaccine in cattle. Am J Vet Res 58: 1104‐1109. 114. Frank‐Peterside N (2000) Response of laboratory staff to vaccination with an inactivated Rift Valley fever vaccine‐‐TSI‐GSD 200. Afr J Med Med Sci 29: 89‐92. 115. Pittman PR, Liu CT, Cannon TL, Makuch RS, Mangiafico JA, et al. (1999) Immunogenicity of an inactivated Rift Valley fever vaccine in humans: a 12‐year experience. Vaccine 18: 181‐189. 116. Gerrard SR, Nichol ST (2007) Synthesis, proteolytic processing and complex formation of N‐terminally nested precursor proteins of the Rift Valley fever virus glycoproteins. Virology 357: 124‐133. 117. Bouloy M, Weber F Molecular biology of rift valley Fever virus. Open Virol J 4: 8‐14. 118. Sherman MB, Freiberg AN, Holbrook MR, Watowich SJ (2009) Single‐particle cryo‐ electron microscopy of Rift Valley fever virus. Virology 387: 11‐15.

120

119. Huiskonen JT, Overby AK, Weber F, Grunewald K (2009) Electron cryo‐microscopy and single‐particle averaging of Rift Valley fever virus: evidence for GN‐GC glycoprotein heterodimers. J Virol 83: 3762‐3769. 120. Freiberg AN, Sherman MB, Morais MC, Holbrook MR, Watowich SJ (2008) Three‐ dimensional organization of Rift Valley fever virus revealed by cryoelectron tomography. J Virol 82: 10341‐10348. 121. Overby AK, Pettersson RF, Neve EP (2007) The glycoprotein cytoplasmic tail of Uukuniemi virus (Bunyaviridae) interacts with ribonucleoproteins and is critical for genome packaging. J Virol 81: 3198‐3205. 122. Overby AK, Popov VL, Pettersson RF, Neve EP (2007) The cytoplasmic tails of Uukuniemi Virus (Bunyaviridae) G(N) and G(C) glycoproteins are important for intracellular targeting and the budding of virus‐like particles. J Virol 81: 11381‐11391. 123. Won S, Ikegami T, Peters CJ, Makino S (2006) NSm and 78‐kilodalton proteins of Rift Valley fever virus are nonessential for viral replication in cell culture. J Virol 80: 8274‐ 8278. 124. Vialat P, Billecocq A, Kohl A, Bouloy M (2000) The S segment of rift valley fever phlebovirus (Bunyaviridae) carries determinants for attenuation and virulence in mice. J Virol 74: 1538‐1543. 125. Gerrard SR, Bird BH, Albarino CG, Nichol ST (2007) The NSm proteins of Rift Valley fever virus are dispensable for maturation, replication and infection. Virology 359: 459‐465. 126. Bird BH, Albarino CG, Nichol ST (2007) Rift Valley fever virus lacking NSm proteins retains high virulence in vivo and may provide a model of human delayed onset neurologic disease. Virology 362: 10‐15. 127. Filone CM, Heise M, Doms RW, Bertolotti‐Ciarlet A (2006) Development and characterization of a Rift Valley fever virus cell‐cell fusion assay using alphavirus replicon vectors. Virology 356: 155‐164. 128. Filone CM, Hanna SL, Caino MC, Bambina S, Doms RW, et al. Rift valley fever virus infection of human cells and insect hosts is promoted by protein kinase C epsilon. PLoS One 5: e15483. 129. Novoa RR, Calderita G, Arranz R, Fontana J, Granzow H, et al. (2005) Virus factories: associations of cell organelles for viral replication and morphogenesis. Biol Cell 97: 147‐172. 130. Ikegami T, Won S, Peters CJ, Makino S (2005) Rift Valley fever virus NSs mRNA is transcribed from an incoming anti‐viral‐sense S RNA segment. J Virol 79: 12106‐12111. 131. Patterson JL, Holloway B, Kolakofsky D (1984) La Crosse virions contain a primer‐ stimulated RNA polymerase and a methylated cap‐dependent endonuclease. J Virol 52: 215‐222. 132. Bouloy M, Pardigon N, Vialat P, Gerbaud S, Girard M (1990) Characterization of the 5' and 3' ends of viral messenger RNAs isolated from BHK21 cells infected with Germiston virus (Bunyavirus). Virology 175: 50‐58. 133. Gro MC, Di Bonito P, Accardi L, Giorgi C (1992) Analysis of 3' and 5' ends of N and NSs messenger RNAs of Toscana Phlebovirus. Virology 191: 435‐438. 134. Jin H, Elliott RM (1993) Characterization of Bunyamwera virus S RNA that is transcribed and replicated by the L protein expressed from recombinant vaccinia virus. J Virol 67: 1396‐1404.

121

135. Simons JF, Pettersson RF (1991) Host‐derived 5' ends and overlapping complementary 3' ends of the two mRNAs transcribed from the ambisense S segment of Uukuniemi virus. J Virol 65: 4741‐4748. 136. Gerrard SR, Nichol ST (2002) Characterization of the Golgi retention motif of Rift Valley fever virus G(N) glycoprotein. J Virol 76: 12200‐12210. 137. Wasmoen TL, Kakach LT, Collett MS (1988) Rift Valley fever virus M segment: cellular localization of M segment‐encoded proteins. Virology 166: 275‐280. 138. Habjan M, Penski N, Wagner V, Spiegel M, Overby AK, et al. (2009) Efficient production of Rift Valley fever virus‐like particles: The antiviral protein MxA can inhibit primary transcription of bunyaviruses. Virology 385: 400‐408. 139. Habjan M, Pichlmair A, Elliott RM, Overby AK, Glatter T, et al. (2009) NSs protein of rift valley fever virus induces the specific degradation of the double‐stranded RNA‐ dependent protein kinase. J Virol 83: 4365‐4375. 140. Billecocq A, Spiegel M, Vialat P, Kohl A, Weber F, et al. (2004) NSs protein of Rift Valley fever virus blocks interferon production by inhibiting host gene transcription. J Virol 78: 9798‐9806. 141. Bouloy M, Janzen C, Vialat P, Khun H, Pavlovic J, et al. (2001) Genetic evidence for an interferon‐antagonistic function of rift valley fever virus nonstructural protein NSs. J Virol 75: 1371‐1377. 142. do Valle TZ, Billecocq A, Guillemot L, Alberts R, Gommet C, et al. A new mouse model reveals a critical role for host innate immunity in resistance to Rift Valley fever. J Immunol 185: 6146‐6156. 143. Kuhn RJ (2007) Togaviridae: The Viruses and Their Replication. In: David M KaPMH, editor. Fields Virology Philadelphia: Lippincott Williams & Wilkins. pp. 1001‐1022. 144. Jose J, Snyder JE, Kuhn RJ (2009) A structural and functional perspective of alphavirus replication and assembly. Future Microbiol 4: 837‐856. 145. Griffin DE (2007) Alphaviruses. In: David M KaPMH, editor. Fields Virology Philadelphia: Lippincott Williams & Wilkins. pp. 1023‐1067. 146. Strauss JH, Strauss EG (1994) The alphaviruses: gene expression, replication, and evolution. Microbiol Rev 58: 491‐562. 147. Gaedigk‐Nitschko K, Schlesinger MJ (1990) The Sindbis virus 6K protein can be detected in virions and is acylated with fatty acids. Virology 175: 274‐281. 148. Lusa S, Garoff H, Liljestrom P (1991) Fate of the 6K membrane protein of Semliki Forest virus during virus assembly. Virology 185: 843‐846. 149. Schmidt MF, Bracha M, Schlesinger MJ (1979) Evidence for covalent attachment of fatty acids to Sindbis virus glycoproteins. Proc Natl Acad Sci U S A 76: 1687‐1691. 150. Froshauer S, Kartenbeck J, Helenius A (1988) Alphavirus RNA replicase is located on the cytoplasmic surface of endosomes and lysosomes. J Cell Biol 107: 2075‐2086. 151. Li G, Rice CM (1993) The signal for translational readthrough of a UGA codon in Sindbis virus RNA involves a single cytidine residue immediately downstream of the termination codon. J Virol 67: 5062‐5067. 152. Strauss EG, De Groot RJ, Levinson R, Strauss JH (1992) Identification of the active site residues in the nsP2 proteinase of Sindbis virus. Virology 191: 932‐940.

122

153. Liljestrom P, Garoff H (1991) Internally located cleavable signal sequences direct the formation of Semliki Forest virus membrane proteins from a polyprotein precursor. J Virol 65: 147‐154. 154. Gaedigk‐Nitschko K, Schlesinger MJ (1991) Site‐directed mutations in Sindbis virus E2 glycoprotein's cytoplasmic domain and the 6K protein lead to similar defects in virus assembly and budding. Virology 183: 206‐214. 155. Ivanova L, Schlesinger MJ (1993) Site‐directed mutations in the Sindbis virus E2 glycoprotein identify palmitoylation sites and affect virus budding. J Virol 67: 2546‐ 2551. 156. Zhao H, Garoff H (1992) Role of cell surface spikes in alphavirus budding. J Virol 66: 7089‐ 7095. 157. Strauss JH, Strauss EG, Kuhn RJ (1995) Budding of alphaviruses. Trends Microbiol 3: 346‐ 350. 158. Suomalainen M, Liljestrom P, Garoff H (1992) Spike protein‐nucleocapsid interactions drive the budding of alphaviruses. J Virol 66: 4737‐4747. 159. Lyles DS (2007) Rhabdoviridae. In: David M KaPMH, editor. Fields Virology Philadelphia: Lippincott Williams & Wilkins. pp. 1363‐1408. 160. Matlin KS, Reggio H, Helenius A, Simons K (1982) Pathway of vesicular stomatitis virus entry leading to infection. J Mol Biol 156: 609‐631. 161. Jayakar HR, Jeetendra E, Whitt MA (2004) Rhabdovirus assembly and budding. Virus Res 106: 117‐132. 162. Miller S, Krijnse‐Locker J (2008) Modification of intracellular membrane structures for virus replication. Nat Rev Microbiol 6: 363‐374. 163. Schlegel A, Giddings TH, Jr., Ladinsky MS, Kirkegaard K (1996) Cellular origin and ultrastructure of membranes induced during poliovirus infection. J Virol 70: 6576‐ 6588. 164. Egger D, Wolk B, Gosert R, Bianchi L, Blum HE, et al. (2002) Expression of hepatitis C virus proteins induces distinct membrane alterations including a candidate viral replication complex. J Virol 76: 5974‐5984. 165. Gosert R, Egger D, Lohmann V, Bartenschlager R, Blum HE, et al. (2003) Identification of the hepatitis C virus RNA replication complex in Huh‐7 cells harboring subgenomic replicons. J Virol 77: 5487‐5492. 166. Kujala P, Ikaheimonen A, Ehsani N, Vihinen H, Auvinen P, et al. (2001) Biogenesis of the Semliki Forest virus RNA replication complex. J Virol 75: 3873‐3884. 167. Magliano D, Marshall JA, Bowden DS, Vardaxis N, Meanger J, et al. (1998) Rubella virus replication complexes are virus‐modified lysosomes. Virology 240: 57‐63. 168. Miller DJ, Schwartz MD, Ahlquist P (2001) Flock house virus RNA replicates on outer mitochondrial membranes in Drosophila cells. J Virol 75: 11664‐11676. 169. Salanueva IJ, Novoa RR, Cabezas P, Lopez‐Iglesias C, Carrascosa JL, et al. (2003) Polymorphism and structural maturation of bunyamwera virus in Golgi and post‐Golgi compartments. J Virol 77: 1368‐1381. 170. Cherry S, Kunte A, Wang H, Coyne C, Rawson RB, et al. (2006) COPI activity coupled with fatty acid biosynthesis is required for viral replication. PLoS Pathog 2: e102. 171. den Boon JA, Diaz A, Ahlquist P Cytoplasmic viral replication complexes. Cell Host Microbe 8: 77‐85.

123

172. Bruce Alberts AJ, Julian Lewis, Martin Raff, Keith Roberts, and Peter Walter (2002) Molecular Biology of the Cell. New York: Garland Science. 173. Hardie DG (2007) AMP‐activated/SNF1 protein kinases: conserved guardians of cellular energy. Nat Rev Mol Cell Biol 8: 774‐785. 174. Hardie DG, Scott JW, Pan DA, Hudson ER (2003) Management of cellular energy by the AMP‐activated protein kinase system. FEBS Lett 546: 113‐120. 175. Lee JH, Koh H, Kim M, Kim Y, Lee SY, et al. (2007) Energy‐dependent regulation of cell structure by AMP‐activated protein kinase. Nature 447: 1017‐1020. 176. Kahn BB, Alquier T, Carling D, Hardie DG (2005) AMP‐activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab 1: 15‐25. 177. Jorgensen SB, Viollet B, Andreelli F, Frosig C, Birk JB, et al. (2004) Knockout of the alpha2 but not alpha1 5'‐AMP‐activated protein kinase isoform abolishes 5‐aminoimidazole‐ 4‐carboxamide‐1‐beta‐4‐ribofuranosidebut not contraction‐induced glucose uptake in skeletal muscle. J Biol Chem 279: 1070‐1079. 178. Viollet B, Andreelli F, Jorgensen SB, Perrin C, Geloen A, et al. (2003) The AMP‐activated protein kinase alpha2 catalytic subunit controls whole‐body insulin sensitivity. J Clin Invest 111: 91‐98. 179. Viollet B, Andreelli F, Jorgensen SB, Perrin C, Flamez D, et al. (2003) Physiological role of AMP‐activated protein kinase (AMPK): insights from knockout mouse models. Biochem Soc Trans 31: 216‐219. 180. Xiao B, Sanders MJ, Underwood E, Heath R, Mayer FV, et al. Structure of mammalian AMPK and its regulation by ADP. Nature 472: 230‐233. 181. Scott JW, Hawley SA, Green KA, Anis M, Stewart G, et al. (2004) CBS domains form energy‐sensing modules whose binding of adenosine ligands is disrupted by disease mutations. J Clin Invest 113: 274‐284. 182. Suter M, Riek U, Tuerk R, Schlattner U, Wallimann T, et al. (2006) Dissecting the role of 5'‐ AMP for allosteric stimulation, activation, and deactivation of AMP‐activated protein kinase. J Biol Chem 281: 32207‐32216. 183. Sanders MJ, Grondin PO, Hegarty BD, Snowden MA, Carling D (2007) Investigating the mechanism for AMP activation of the AMP‐activated protein kinase cascade. Biochem J 403: 139‐148. 184. Davies SP, Helps NR, Cohen PT, Hardie DG (1995) 5'‐AMP inhibits dephosphorylation, as well as promoting phosphorylation, of the AMP‐activated protein kinase. Studies using bacterially expressed human protein phosphatase‐2C alpha and native bovine protein phosphatase‐2AC. FEBS Lett 377: 421‐425. 185. Alessi DR, Sakamoto K, Bayascas JR (2006) LKB1‐dependent signaling pathways. Annu Rev Biochem 75: 137‐163. 186. Shaw RJ, Kosmatka M, Bardeesy N, Hurley RL, Witters LA, et al. (2004) The tumor suppressor LKB1 kinase directly activates AMP‐activated kinase and regulates apoptosis in response to energy stress. Proc Natl Acad Sci U S A 101: 3329‐3335. 187. Hawley SA, Pan DA, Mustard KJ, Ross L, Bain J, et al. (2005) Calmodulin‐dependent protein kinase kinase‐beta is an alternative upstream kinase for AMP‐activated protein kinase. Cell Metab 2: 9‐19.

124

188. Hurley RL, Anderson KA, Franzone JM, Kemp BE, Means AR, et al. (2005) The Ca2+/calmodulin‐dependent protein kinase kinases are AMP‐activated protein kinase kinases. J Biol Chem 280: 29060‐29066. 189. Hong SP, Leiper FC, Woods A, Carling D, Carlson M (2003) Activation of yeast Snf1 and mammalian AMP‐activated protein kinase by upstream kinases. Proc Natl Acad Sci U S A 100: 8839‐8843. 190. Momcilovic M, Hong SP, Carlson M (2006) Mammalian TAK1 activates Snf1 protein kinase in yeast and phosphorylates AMP‐activated protein kinase in vitro. J Biol Chem 281: 25336‐25343. 191. Herrero‐Martin G, Hoyer‐Hansen M, Garcia‐Garcia C, Fumarola C, Farkas T, et al. (2009) TAK1 activates AMPK‐dependent cytoprotective autophagy in TRAIL‐treated epithelial cells. EMBO J 28: 677‐685. 192. Levine B, Mizushima N, Virgin HW Autophagy in immunity and inflammation. Nature 469: 323‐335. 193. Ninomiya‐Tsuji J, Kishimoto K, Hiyama A, Inoue J, Cao Z, et al. (1999) The kinase TAK1 can activate the NIK‐I kappaB as well as the MAP kinase cascade in the IL‐1 signalling pathway. Nature 398: 252‐256. 194. Sato S, Sanjo H, Takeda K, Ninomiya‐Tsuji J, Yamamoto M, et al. (2005) Essential function for the kinase TAK1 in innate and adaptive immune responses. Nat Immunol 6: 1087‐ 1095. 195. Shim JH, Xiao C, Paschal AE, Bailey ST, Rao P, et al. (2005) TAK1, but not TAB1 or TAB2, plays an essential role in multiple signaling pathways in vivo. Genes Dev 19: 2668‐ 2681. 196. Takaesu G, Surabhi RM, Park KJ, Ninomiya‐Tsuji J, Matsumoto K, et al. (2003) TAK1 is critical for IkappaB kinase‐mediated activation of the NF‐kappaB pathway. J Mol Biol 326: 105‐115. 197. Ruderman NB, Keller C, Richard AM, Saha AK, Luo Z, et al. (2006) Interleukin‐6 regulation of AMP‐activated protein kinase. Potential role in the systemic response to exercise and prevention of the metabolic syndrome. Diabetes 55 Suppl 2: S48‐54. 198. Watt MJ, Dzamko N, Thomas WG, Rose‐John S, Ernst M, et al. (2006) CNTF reverses obesity‐induced insulin resistance by activating skeletal muscle AMPK. Nat Med 12: 541‐548. 199. Kola B, Hubina E, Tucci SA, Kirkham TC, Garcia EA, et al. (2005) Cannabinoids and ghrelin have both central and peripheral metabolic and cardiac effects via AMP‐activated protein kinase. J Biol Chem 280: 25196‐25201. 200. Hardie DG (2008) AMPK: a key regulator of energy balance in the single cell and the whole organism. Int J Obes (Lond) 32 Suppl 4: S7‐12. 201. Holmes BF, Kurth‐Kraczek EJ, Winder WW (1999) Chronic activation of 5'‐AMP‐activated protein kinase increases GLUT‐4, hexokinase, and glycogen in muscle. J Appl Physiol 87: 1990‐1995. 202. Kurth‐Kraczek EJ, Hirshman MF, Goodyear LJ, Winder WW (1999) 5' AMP‐activated protein kinase activation causes GLUT4 translocation in skeletal muscle. Diabetes 48: 1667‐1671.

125

203. Sano H, Kane S, Sano E, Miinea CP, Asara JM, et al. (2003) Insulin‐stimulated phosphorylation of a Rab GTPase‐activating protein regulates GLUT4 translocation. J Biol Chem 278: 14599‐14602. 204. Shaw RJ (2009) LKB1 and AMP‐activated protein kinase control of mTOR signalling and growth. Acta Physiol (Oxf) 196: 65‐80. 205. Burnett PE, Barrow RK, Cohen NA, Snyder SH, Sabatini DM (1998) RAFT1 phosphorylation of the translational regulators p70 S6 kinase and 4E‐BP1. Proc Natl Acad Sci U S A 95: 1432‐1437. 206. Pearson RB, Dennis PB, Han JW, Williamson NA, Kozma SC, et al. (1995) The principal target of rapamycin‐induced p70s6k inactivation is a novel phosphorylation site within a conserved hydrophobic domain. EMBO J 14: 5279‐5287. 207. Wullschleger S, Loewith R, Hall MN (2006) TOR signaling in growth and metabolism. Cell 124: 471‐484. 208. Gingras AC, Raught B, Gygi SP, Niedzwiecka A, Miron M, et al. (2001) Hierarchical phosphorylation of the translation inhibitor 4E‐BP1. Genes Dev 15: 2852‐2864. 209. Huang J, Manning BD (2008) The TSC1‐TSC2 complex: a molecular switchboard controlling cell growth. Biochem J 412: 179‐190. 210. Zhang Y, Gao X, Saucedo LJ, Ru B, Edgar BA, et al. (2003) Rheb is a direct target of the tuberous sclerosis tumour suppressor proteins. Nat Cell Biol 5: 578‐581. 211. Inoki K, Li Y, Xu T, Guan KL (2003) Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling. Genes Dev 17: 1829‐1834. 212. Tee AR, Manning BD, Roux PP, Cantley LC, Blenis J (2003) Tuberous sclerosis complex gene products, Tuberin and Hamartin, control mTOR signaling by acting as a GTPase‐ activating protein complex toward Rheb. Curr Biol 13: 1259‐1268. 213. Castro AF, Rebhun JF, Clark GJ, Quilliam LA (2003) Rheb binds tuberous sclerosis complex 2 (TSC2) and promotes S6 kinase activation in a rapamycin‐ and farnesylation‐ dependent manner. J Biol Chem 278: 32493‐32496. 214. Garami A, Zwartkruis FJ, Nobukuni T, Joaquin M, Roccio M, et al. (2003) Insulin activation of Rheb, a mediator of mTOR/S6K/4E‐BP signaling, is inhibited by TSC1 and 2. Mol Cell 11: 1457‐1466. 215. Shaw RJ, Bardeesy N, Manning BD, Lopez L, Kosmatka M, et al. (2004) The LKB1 tumor suppressor negatively regulates mTOR signaling. Cancer Cell 6: 91‐99. 216. Manning BD, Tee AR, Logsdon MN, Blenis J, Cantley LC (2002) Identification of the tuberous sclerosis complex‐2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3‐kinase/akt pathway. Mol Cell 10: 151‐162. 217. Inoki K, Li Y, Zhu T, Wu J, Guan KL (2002) TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling. Nat Cell Biol 4: 648‐657. 218. Tee AR, Anjum R, Blenis J (2003) Inactivation of the tuberous sclerosis complex‐1 and ‐2 gene products occurs by phosphoinositide 3‐kinase/Akt‐dependent and ‐independent phosphorylation of tuberin. J Biol Chem 278: 37288‐37296. 219. Roux PP, Ballif BA, Anjum R, Gygi SP, Blenis J (2004) Tumor‐promoting phorbol esters and activated Ras inactivate the tuberous sclerosis tumor suppressor complex via p90 ribosomal S6 kinase. Proc Natl Acad Sci U S A 101: 13489‐13494.

126

220. Horman S, Browne G, Krause U, Patel J, Vertommen D, et al. (2002) Activation of AMP‐ activated protein kinase leads to the phosphorylation of elongation factor 2 and an inhibition of protein synthesis. Curr Biol 12: 1419‐1423. 221. Browne GJ, Finn SG, Proud CG (2004) Stimulation of the AMP‐activated protein kinase leads to activation of eukaryotic elongation factor 2 kinase and to its phosphorylation at a novel site, serine 398. J Biol Chem 279: 12220‐12231. 222. Browne GJ, Proud CG (2002) Regulation of peptide‐chain elongation in mammalian cells. Eur J Biochem 269: 5360‐5368. 223. Ryazanov AG, Shestakova EA, Natapov PG (1988) Phosphorylation of elongation factor 2 by EF‐2 kinase affects rate of translation. Nature 334: 170‐173. 224. Carlberg U, Nilsson A, Nygard O (1990) Functional properties of phosphorylated elongation factor 2. Eur J Biochem 191: 639‐645. 225. Wang X, Li W, Williams M, Terada N, Alessi DR, et al. (2001) Regulation of elongation factor 2 kinase by p90(RSK1) and p70 S6 kinase. EMBO J 20: 4370‐4379. 226. Knebel A, Haydon CE, Morrice N, Cohen P (2002) Stress‐induced regulation of eukaryotic elongation factor 2 kinase by SB 203580‐sensitive and ‐insensitive pathways. Biochem J 367: 525‐532. 227. Rabinowitz JD, White E Autophagy and metabolism. Science 330: 1344‐1348. 228. Hardie DG, Pan DA (2002) Regulation of fatty acid synthesis and oxidation by the AMP‐ activated protein kinase. Biochem Soc Trans 30: 1064‐1070. 229. Munday MR, Campbell DG, Carling D, Hardie DG (1988) Identification by amino acid sequencing of three major regulatory phosphorylation sites on rat acetyl‐CoA carboxylase. Eur J Biochem 175: 331‐338. 230. Davies SP, Sim AT, Hardie DG (1990) Location and function of three sites phosphorylated on rat acetyl‐CoA carboxylase by the AMP‐activated protein kinase. Eur J Biochem 187: 183‐190. 231. Saggerson D (2008) Malonyl‐CoA, a key signaling molecule in mammalian cells. Annu Rev Nutr 28: 253‐272. 232. Cinti DL, Cook L, Nagi MN, Suneja SK (1992) The fatty acid chain elongation system of mammalian endoplasmic reticulum. Prog Lipid Res 31: 1‐51. 233. Ha J, Daniel S, Kong IS, Park CK, Tae HJ, et al. (1994) Cloning of human acetyl‐CoA carboxylase cDNA. Eur J Biochem 219: 297‐306. 234. Davies SP, Carling D, Munday MR, Hardie DG (1992) Diurnal rhythm of phosphorylation of rat liver acetyl‐CoA carboxylase by the AMP‐activated protein kinase, demonstrated using freeze‐clamping. Effects of high fat diets. Eur J Biochem 203: 615‐623. 235. Koh H, Chung J (2007) AMPK links energy status to cell structure and mitosis. Biochem Biophys Res Commun 362: 789‐792. 236. Zhang L, Li J, Young LH, Caplan MJ (2006) AMP‐activated protein kinase regulates the assembly of epithelial tight junctions. Proc Natl Acad Sci U S A 103: 17272‐17277. 237. Zheng B, Cantley LC (2007) Regulation of epithelial tight junction assembly and disassembly by AMP‐activated protein kinase. Proc Natl Acad Sci U S A 104: 819‐822. 238. Miranda L, Carpentier S, Platek A, Hussain N, Gueuning MA, et al. AMP‐activated protein kinase induces actin cytoskeleton reorganization in epithelial cells. Biochem Biophys Res Commun 396: 656‐661.

127

239. Favero CB, Mandell JW (2007) A pharmacological activator of AMP‐activated protein kinase (AMPK) induces astrocyte stellation. Brain Res 1168: 1‐10. 240. Guest CB, Chakour KS, Freund GG (2008) Macropinocytosis is decreased in diabetic mouse macrophages and is regulated by AMPK. BMC Immunol 9: 42. 241. Nakano A, Kato H, Watanabe T, Min KD, Yamazaki S, et al. AMPK controls the speed of microtubule polymerization and directional cell migration through CLIP‐170 phosphorylation. Nat Cell Biol 12: 583‐590. 242. Mirouse V, Swick LL, Kazgan N, St Johnston D, Brenman JE (2007) LKB1 and AMPK maintain epithelial cell polarity under energetic stress. J Cell Biol 177: 387‐392. 243. Baas AF, Kuipers J, van der Wel NN, Batlle E, Koerten HK, et al. (2004) Complete polarization of single intestinal epithelial cells upon activation of LKB1 by STRAD. Cell 116: 457‐466. 244. Long YC, Zierath JR (2006) AMP‐activated protein kinase signaling in metabolic regulation. J Clin Invest 116: 1776‐1783. 245. Zhang BB, Zhou G, Li C (2009) AMPK: an emerging drug target for diabetes and the metabolic syndrome. Cell Metab 9: 407‐416. 246. Kelley DE, Goodpaster B, Wing RR, Simoneau JA (1999) Skeletal muscle fatty acid metabolism in association with insulin resistance, obesity, and weight loss. Am J Physiol 277: E1130‐1141. 247. Merrill GF, Kurth EJ, Hardie DG, Winder WW (1997) AICA riboside increases AMP‐ activated protein kinase, fatty acid oxidation, and glucose uptake in rat muscle. Am J Physiol 273: E1107‐1112. 248. Lochhead PA, Salt IP, Walker KS, Hardie DG, Sutherland C (2000) 5‐aminoimidazole‐4‐ carboxamide riboside mimics the effects of insulin on the expression of the 2 key gluconeogenic genes PEPCK and glucose‐6‐phosphatase. Diabetes 49: 896‐903. 249. Koo SH, Flechner L, Qi L, Zhang X, Screaton RA, et al. (2005) The CREB coactivator TORC2 is a key regulator of fasting glucose metabolism. Nature 437: 1109‐1111. 250. Velasco G, Geelen MJ, Guzman M (1997) Control of hepatic fatty acid oxidation by 5'‐ AMP‐activated protein kinase involves a malonyl‐CoA‐dependent and a malonyl‐CoA‐ independent mechanism. Arch Biochem Biophys 337: 169‐175. 251. Sullivan JE, Brocklehurst KJ, Marley AE, Carey F, Carling D, et al. (1994) Inhibition of lipolysis and lipogenesis in isolated rat adipocytes with AICAR, a cell‐permeable activator of AMP‐activated protein kinase. FEBS Lett 353: 33‐36. 252. Corton JM, Gillespie JG, Hawley SA, Hardie DG (1995) 5‐aminoimidazole‐4‐carboxamide ribonucleoside. A specific method for activating AMP‐activated protein kinase in intact cells? Eur J Biochem 229: 558‐565. 253. Zong H, Ren JM, Young LH, Pypaert M, Mu J, et al. (2002) AMP kinase is required for mitochondrial biogenesis in skeletal muscle in response to chronic energy deprivation. Proc Natl Acad Sci U S A 99: 15983‐15987. 254. Song XM, Fiedler M, Galuska D, Ryder JW, Fernstrom M, et al. (2002) 5‐Aminoimidazole‐ 4‐carboxamide ribonucleoside treatment improves glucose homeostasis in insulin‐ resistant diabetic (ob/ob) mice. Diabetologia 45: 56‐65. 255. Bergeron R, Previs SF, Cline GW, Perret P, Russell RR, 3rd, et al. (2001) Effect of 5‐ aminoimidazole‐4‐carboxamide‐1‐beta‐D‐ribofuranoside infusion on in vivo glucose and lipid metabolism in lean and obese Zucker rats. Diabetes 50: 1076‐1082.

128

256. Iglesias MA, Ye JM, Frangioudakis G, Saha AK, Tomas E, et al. (2002) AICAR administration causes an apparent enhancement of muscle and liver insulin action in insulin‐resistant high‐fat‐fed rats. Diabetes 51: 2886‐2894. 257. Buhl ES, Jessen N, Pold R, Ledet T, Flyvbjerg A, et al. (2002) Long‐term AICAR administration reduces metabolic disturbances and lowers blood pressure in rats displaying features of the insulin resistance syndrome. Diabetes 51: 2199‐2206. 258. Zhou G, Myers R, Li Y, Chen Y, Shen X, et al. (2001) Role of AMP‐activated protein kinase in mechanism of metformin action. J Clin Invest 108: 1167‐1174. 259. Shu Y, Sheardown SA, Brown C, Owen RP, Zhang S, et al. (2007) Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action. J Clin Invest 117: 1422‐1431. 260. Shaw RJ, Lamia KA, Vasquez D, Koo SH, Bardeesy N, et al. (2005) The kinase LKB1 mediates glucose homeostasis in liver and therapeutic effects of metformin. Science 310: 1642‐1646. 261. Hawley SA, Gadalla AE, Olsen GS, Hardie DG (2002) The antidiabetic drug metformin activates the AMP‐activated protein kinase cascade via an adenine nucleotide‐ independent mechanism. Diabetes 51: 2420‐2425. 262. Owen MR, Doran E, Halestrap AP (2000) Evidence that metformin exerts its anti‐diabetic effects through inhibition of complex 1 of the mitochondrial respiratory chain. Biochem J 348 Pt 3: 607‐614. 263. Stumvoll M, Nurjhan N, Perriello G, Dailey G, Gerich JE (1995) Metabolic effects of metformin in non‐insulin‐dependent diabetes mellitus. N Engl J Med 333: 550‐554. 264. Wiernsperger NF, Bailey CJ (1999) The antihyperglycaemic effect of metformin: therapeutic and cellular mechanisms. Drugs 58 Suppl 1: 31‐39; discussion 75‐82. 265. Hundal RS, Krssak M, Dufour S, Laurent D, Lebon V, et al. (2000) Mechanism by which metformin reduces glucose production in type 2 diabetes. Diabetes 49: 2063‐2069. 266. Hundal HS, Ramlal T, Reyes R, Leiter LA, Klip A (1992) Cellular mechanism of metformin action involves glucose transporter translocation from an intracellular pool to the plasma membrane in L6 muscle cells. Endocrinology 131: 1165‐1173. 267. Galuska D, Nolte LA, Zierath JR, Wallberg‐Henriksson H (1994) Effect of metformin on insulin‐stimulated glucose transport in isolated skeletal muscle obtained from patients with NIDDM. Diabetologia 37: 826‐832. 268. El‐Mir MY, Nogueira V, Fontaine E, Averet N, Rigoulet M, et al. (2000) Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I. J Biol Chem 275: 223‐228. 269. Kalender A, Selvaraj A, Kim SY, Gulati P, Brule S, et al. Metformin, independent of AMPK, inhibits mTORC1 in a rag GTPase‐dependent manner. Cell Metab 11: 390‐401. 270. Tohyama D, Yamaguchi A A critical role of SNF1A/dAMPKalpha (Drosophila AMP‐ activated protein kinase alpha) in muscle on longevity and stress resistance in Drosophila melanogaster. Biochem Biophys Res Commun 394: 112‐118. 271. Mair W, Morantte I, Rodrigues AP, Manning G, Montminy M, et al. Lifespan extension induced by AMPK and calcineurin is mediated by CRTC‐1 and CREB. Nature 470: 404‐ 408.

129

272. Hirsch HA, Iliopoulos D, Tsichlis PN, Struhl K (2009) Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission. Cancer Res 69: 7507‐7511. 273. Zakikhani M, Dowling R, Fantus IG, Sonenberg N, Pollak M (2006) Metformin is an AMP kinase‐dependent growth inhibitor for breast cancer cells. Cancer Res 66: 10269‐ 10273. 274. Vazquez‐Martin A, Oliveras‐Ferraros C, del Barco S, Martin‐Castillo B, Menendez JA (2009) The antidiabetic drug metformin: a pharmaceutical AMPK activator to overcome breast cancer resistance to HER2 inhibitors while decreasing risk of cardiomyopathy. Ann Oncol 20: 592‐595. 275. Vazquez‐Martin A, Oliveras‐Ferraros C, Menendez JA (2009) The antidiabetic drug metformin suppresses HER2 (erbB‐2) oncoprotein overexpression via inhibition of the mTOR effector p70S6K1 in human breast carcinoma cells. Cell Cycle 8: 88‐96. 276. Evans JM, Donnelly LA, Emslie‐Smith AM, Alessi DR, Morris AD (2005) Metformin and reduced risk of cancer in diabetic patients. BMJ 330: 1304‐1305. 277. Memmott RM, Mercado JR, Maier CR, Kawabata S, Fox SD, et al. Metformin prevents tobacco carcinogen‐‐induced lung tumorigenesis. Cancer Prev Res (Phila) 3: 1066‐1076. 278. Romero‐Gomez M, Diago M, Andrade RJ, Calleja JL, Salmeron J, et al. (2009) Treatment of insulin resistance with metformin in naive genotype 1 chronic hepatitis C patients receiving peginterferon alfa‐2a plus ribavirin. Hepatology 50: 1702‐1708. 279. Serfaty L, Fartoux L, Poupon R (2009) Pioglitazone as adjuvant therapy in chronic hepatitis C: sequential rather than concomitant administration with pegylated interferon and ribavirin? J Hepatol 50: 1269‐1271. 280. Overbeck K, Genne D, Golay A, Negro F (2008) Pioglitazone in chronic hepatitis C not responding to pegylated interferon‐alpha and ribavirin. J Hepatol 49: 295‐298. 281. Alexander WA, Moss B, Fuerst TR (1992) Regulated expression of foreign genes in vaccinia virus under the control of bacteriophage T7 RNA polymerase and the Escherichia coli lac repressor. J Virol 66: 2934‐2942. 282. Cochran MA, Puckett C, Moss B (1985) In vitro mutagenesis of the promoter region for a vaccinia virus gene: evidence for tandem early and late regulatory signals. J Virol 54: 30‐37. 283. Chakrabarti S, Brechling K, Moss B (1985) Vaccinia virus expression vector: coexpression of beta‐galactosidase provides visual screening of recombinant virus plaques. Mol Cell Biol 5: 3403‐3409. 284. Ramsey‐Ewing A, Moss B (1995) Restriction of vaccinia virus replication in CHO cells occurs at the stage of viral intermediate protein synthesis. Virology 206: 984‐993. 285. Weaver JR, Shamim M, Alexander E, Davies DH, Felgner PL, et al. (2007) The identification and characterization of a monoclonal antibody to the vaccinia virus E3 protein. Virus Res 130: 269‐274. 286. Tokumitsu H, Inuzuka H, Ishikawa Y, Ikeda M, Saji I, et al. (2002) STO‐609, a specific inhibitor of the Ca(2+)/calmodulin‐dependent protein kinase kinase. J Biol Chem 277: 15813‐15818. 287. Morrison DK, Murakami MS, Cleghon V (2000) Protein kinases and phosphatases in the Drosophila genome. J Cell Biol 150: F57‐62.

130

288. Armknecht S, Boutros M, Kiger A, Nybakken K, Mathey‐Prevot B, et al. (2005) High‐ throughput RNA interference screens in Drosophila tissue culture cells. Methods Enzymol 392: 55‐73. 289. Moser TS, Sabin LR, Cherry S RNAi screening for host factors involved in Vaccinia virus infection using Drosophila cells. J Vis Exp. 290. Zhang XD, Yang XC, Chung N, Gates A, Stec E, et al. (2006) Robust statistical methods for hit selection in RNA interference high‐throughput screening experiments. Pharmacogenomics 7: 299‐309. 291. Mankouri J, Tedbury PR, Gretton S, Hughes ME, Griffin SD, et al. Enhanced hepatitis C virus genome replication and lipid accumulation mediated by inhibition of AMP‐ activated protein kinase. Proc Natl Acad Sci U S A 107: 11549‐11554. 292. Heaton NS, Randall G Dengue virus‐induced autophagy regulates lipid metabolism. Cell Host Microbe 8: 422‐432. 293. Hannah VC, Ou J, Luong A, Goldstein JL, Brown MS (2001) Unsaturated fatty acids down‐ regulate srebp isoforms 1a and 1c by two mechanisms in HEK‐293 cells. J Biol Chem 276: 4365‐4372. 294. Qian K, Morris‐Natschke SL, Lee KH (2009) HIV entry inhibitors and their potential in HIV therapy. Med Res Rev 29: 369‐393. 295. Hao L, Sakurai A, Watanabe T, Sorensen E, Nidom CA, et al. (2008) Drosophila RNAi screen identifies host genes important for influenza virus replication. Nature 454: 890‐ 893. 296. Laderoute KR, Amin K, Calaoagan JM, Knapp M, Le T, et al. (2006) 5'‐AMP‐activated protein kinase (AMPK) is induced by low‐oxygen and glucose deprivation conditions found in solid‐tumor microenvironments. Mol Cell Biol 26: 5336‐5347. 297. Sun X, Yau VK, Briggs BJ, Whittaker GR (2005) Role of clathrin‐mediated endocytosis during vesicular stomatitis virus entry into host cells. Virology 338: 53‐60. 298. Stein SC, Woods A, Jones NA, Davison MD, Carling D (2000) The regulation of AMP‐ activated protein kinase by phosphorylation. Biochem J 345 Pt 3: 437‐443. 299. Hawley SA, Selbert MA, Goldstein EG, Edelman AM, Carling D, et al. (1995) 5'‐AMP activates the AMP‐activated protein kinase cascade, and Ca2+/calmodulin activates the calmodulin‐dependent protein kinase I cascade, via three independent mechanisms. J Biol Chem 270: 27186‐27191. 300. Hawley SA, Davison M, Woods A, Davies SP, Beri RK, et al. (1996) Characterization of the AMP‐activated protein kinase kinase from rat liver and identification of threonine 172 as the major site at which it phosphorylates AMP‐activated protein kinase. J Biol Chem 271: 27879‐27887. 301. Pan DA, Hardie DG (2002) A homologue of AMP‐activated protein kinase in Drosophila melanogaster is sensitive to AMP and is activated by ATP depletion. Biochem J 367: 179‐186. 302. Bardeesy N, Sinha M, Hezel AF, Signoretti S, Hathaway NA, et al. (2002) Loss of the Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 419: 162‐167. 303. Jones AT (2007) Macropinocytosis: searching for an endocytic identity and role in the uptake of cell penetrating peptides. J Cell Mol Med 11: 670‐684.

131

304. Keller HU (1990) Diacylglycerols and PMA are particularly effective stimulators of fluid pinocytosis in human neutrophils. J Cell Physiol 145: 465‐471. 305. Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A (1992) The small GTP‐binding protein rac regulates growth factor‐induced membrane ruffling. Cell 70: 401‐410. 306. Swanson JA (1989) Phorbol esters stimulate macropinocytosis and solute flow through macrophages. J Cell Sci 94 ( Pt 1): 135‐142. 307. Sapkota GP, Deak M, Kieloch A, Morrice N, Goodarzi AA, et al. (2002) Ionizing radiation induces ataxia telangiectasia mutated kinase (ATM)‐mediated phosphorylation of LKB1/STK11 at Thr‐366. Biochem J 368: 507‐516. 308. Kurokawa K, Matsuda M (2005) Localized RhoA activation as a requirement for the induction of membrane ruffling. Mol Biol Cell 16: 4294‐4303. 309. Wei Q, Adelstein RS (2002) Pitx2a expression alters actin‐myosin cytoskeleton and migration of HeLa cells through Rho GTPase signaling. Mol Biol Cell 13: 683‐697. 310. Pollard TD, Borisy GG (2003) Cellular motility driven by assembly and disassembly of actin filaments. Cell 112: 453‐465. 311. Edwards DC, Sanders LC, Bokoch GM, Gill GN (1999) Activation of LIM‐kinase by Pak1 couples Rac/Cdc42 GTPase signalling to actin cytoskeletal dynamics. Nat Cell Biol 1: 253‐259. 312. Liberali P, Kakkonen E, Turacchio G, Valente C, Spaar A, et al. (2008) The closure of Pak1‐ dependent macropinosomes requires the phosphorylation of CtBP1/BARS. EMBO J 27: 970‐981. 313. Raftopoulou M, Hall A (2004) Cell migration: Rho GTPases lead the way. Dev Biol 265: 23‐ 32. 314. Levine YC, Li GK, Michel T (2007) Agonist‐modulated regulation of AMP‐activated protein kinase (AMPK) in endothelial cells. Evidence for an AMPK ‐> Rac1 ‐> Akt ‐> endothelial nitric‐oxide synthase pathway. J Biol Chem 282: 20351‐20364. 315. Lee YM, Lee JO, Jung JH, Kim JH, Park SH, et al. (2008) Retinoic acid leads to cytoskeletal rearrangement through AMPK‐Rac1 and stimulates glucose uptake through AMPK‐p38 MAPK in skeletal muscle cells. J Biol Chem 283: 33969‐33974. 316. Yamaguchi S, Katahira H, Ozawa S, Nakamichi Y, Tanaka T, et al. (2005) Activators of AMP‐activated protein kinase enhance GLUT4 translocation and its glucose transport activity in 3T3‐L1 adipocytes. Am J Physiol Endocrinol Metab 289: E643‐649. 317. Horie T, Ono K, Nagao K, Nishi H, Kinoshita M, et al. (2008) Oxidative stress induces GLUT4 translocation by activation of PI3‐K/Akt and dual AMPK kinase in cardiac myocytes. J Cell Physiol 215: 733‐742. 318. Ueda S, Kataoka T, Satoh T (2008) Activation of the small GTPase Rac1 by a specific guanine‐nucleotide‐exchange factor suffices to induce glucose uptake into skeletal‐ muscle cells. Biol Cell 100: 645‐657. 319. Domin J, Harper L, Aubyn D, Wheeler M, Florey O, et al. (2005) The class II phosphoinositide 3‐kinase PI3K‐C2beta regulates cell migration by a PtdIns3P dependent mechanism. J Cell Physiol 205: 452‐462. 320. Maffucci T, Cooke FT, Foster FM, Traer CJ, Fry MJ, et al. (2005) Class II phosphoinositide 3‐ kinase defines a novel signaling pathway in cell migration. J Cell Biol 169: 789‐799.

132

321. Soares JA, Leite FG, Andrade LG, Torres AA, De Sousa LP, et al. (2009) Activation of the PI3K/Akt pathway early during vaccinia and cowpox virus infections is required for both host survival and viral replication. J Virol 83: 6883‐6899. 322. Zaborowska I, Walsh D (2009) PI3K signaling regulates rapamycin‐insensitive translation initiation complex formation in vaccinia virus‐infected cells. J Virol 83: 3988‐3992. 323. Ikonomov OC, Sbrissa D, Foti M, Carpentier JL, Shisheva A (2003) PIKfyve controls fluid phase endocytosis but not recycling/degradation of endocytosed receptors or sorting of procathepsin D by regulating multivesicular body morphogenesis. Mol Biol Cell 14: 4581‐4591. 324. Rusten TE, Rodahl LM, Pattni K, Englund C, Samakovlis C, et al. (2006) Fab1 phosphatidylinositol 3‐phosphate 5‐kinase controls trafficking but not silencing of endocytosed receptors. Mol Biol Cell 17: 3989‐4001. 325. Komada M, Kitamura N (2005) The Hrs/STAM complex in the downregulation of receptor tyrosine kinases. J Biochem 137: 1‐8. 326. Mizuno E, Kawahata K, Okamoto A, Kitamura N, Komada M (2004) Association with Hrs is required for the early endosomal localization, stability, and function of STAM. J Biochem 135: 385‐396. 327. Moser TS, Jones RG, Thompson CB, Coyne CB, Cherry S A kinome RNAi screen identified AMPK as promoting poxvirus entry through the control of actin dynamics. PLoS Pathog 6: e1000954. 328. Cool B, Zinker B, Chiou W, Kifle L, Cao N, et al. (2006) Identification and characterization of a small molecule AMPK activator that treats key components of type 2 diabetes and the metabolic syndrome. Cell Metab 3: 403‐416. 329. Hawley SA, Ross FA, Chevtzoff C, Green KA, Evans A, et al. Use of cells expressing gamma subunit variants to identify diverse mechanisms of AMPK activation. Cell Metab 11: 554‐565. 330. Goransson O, McBride A, Hawley SA, Ross FA, Shpiro N, et al. (2007) Mechanism of action of A‐769662, a valuable tool for activation of AMP‐activated protein kinase. J Biol Chem 282: 32549‐32560. 331. Moreno D, Knecht E, Viollet B, Sanz P (2008) A769662, a novel activator of AMP‐activated protein kinase, inhibits non‐proteolytic components of the 26S proteasome by an AMPK‐independent mechanism. FEBS Lett 582: 2650‐2654. 332. Fontana J, Lopez‐Montero N, Elliott RM, Fernandez JJ, Risco C (2008) The unique architecture of Bunyamwera virus factories around the Golgi complex. Cell Microbiol 10: 2012‐2028. 333. Kudchodkar SB, Del Prete GQ, Maguire TG, Alwine JC (2007) AMPK‐mediated inhibition of mTOR kinase is circumvented during immediate‐early times of human cytomegalovirus infection. J Virol 81: 3649‐3651. 334. Zhang HS, Wu MR (2009) SIRT1 regulates Tat‐induced HIV‐1 transactivation through activating AMP‐activated protein kinase. Virus Res 146: 51‐57. 335. Dreux M, Gastaminza P, Wieland SF, Chisari FV (2009) The autophagy machinery is required to initiate hepatitis C virus replication. Proc Natl Acad Sci U S A 106: 14046‐ 14051.

133

336. Shelly S, Lukinova N, Bambina S, Berman A, Cherry S (2009) Autophagy is an essential component of Drosophila immunity against vesicular stomatitis virus. Immunity 30: 588‐598. 337. Zhou Z, Jiang X, Liu D, Fan Z, Hu X, et al. (2009) Autophagy is involved in influenza A virus replication. Autophagy 5: 321‐328. 338. Orvedahl A, MacPherson S, Sumpter R, Jr., Talloczy Z, Zou Z, et al. Autophagy protects against Sindbis virus infection of the central nervous system. Cell Host Microbe 7: 115‐ 127. 339. Lee MS, Craigie R (1998) A previously unidentified host protein protects retroviral DNA from autointegration. Proc Natl Acad Sci U S A 95: 1528‐1533. 340. Lizcano JM, Goransson O, Toth R, Deak M, Morrice NA, et al. (2004) LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR‐1. EMBO J 23: 833‐843. 341. Sakamoto K, Goransson O, Hardie DG, Alessi DR (2004) Activity of LKB1 and AMPK‐ related kinases in skeletal muscle: effects of contraction, phenformin, and AICAR. Am J Physiol Endocrinol Metab 287: E310‐317. 342. Lozach PY, Mancini R, Bitto D, Meier R, Oestereich L, et al. Entry of bunyaviruses into mammalian cells. Cell Host Microbe 7: 488‐499.

134