The Pennsylvania State University the Graduate School College Of

The Pennsylvania State University

The Graduate School

College of Medicine

RETROVIRAL CAPSID MATURATION:

CONTRIBUTIONS OF THE CA C-TERMINAL DOMAIN

A Dissertation in

Microbiology and Immunology

Matthew Raymond England

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

May 2015

The dissertation of Matthew Raymond England was reviewed and approved* by the following:

Rebecca C. Craven Professor of Microbiology and Immunology Dissertation Advisor Chair of Committee

Sarah K. Bronson Associate Professor of Cellular and Molecular Physiology

David J. Spector Professor Emeritus of Microbiology and Immunology

Fang Tian Associate Professor of Biochemistry and Molecular Biology

Jianming Hu Professor of Microbiology and Immunology

Aron Lukacher Professor and Chair of Microbiology and Immunology

*Signatures are on file in the Graduate School

ABSTRACT

Maturation of virions is a critical and obligatory step toward infectivity for orthoretroviruses. Retroviruses initially bud from infected cells as immature particles. The structural proteins are synthesized as the Gag and Gag-Pol polyproteins. Concomitant with budding, the viral protease (PR) becomes active and cleaves Gag and Gag-Pol into the constituent proteins. Cleavage induces significant morphological changes to the interior of the virus, the most visually stunning aspect of maturation. Whereas the immature particle is roughly spherical, the mature virus contains a capsid of variable shape surrounding two copies of the viral positive- stranded RNA genome and associated proteins.

The mature capsid is made of the capsid protein (CA), which consists of an N-terminal domain (NTD) and C-terminal domain (CTD). The mature capsid is made of a hexameric lattice of CA punctuated by 12 pentamers variably located. The best studied CA-CA interaction is the dimerization interface between two CTD molecules. The dimerization interaction holds neighboring hexamers together and is generated by both hydrophobic and electrostatic interactions between the 310 helix and the second alpha helix of the CTD (α9). Further CTD-CTD stabilization is provided by the three-fold CTD interface which holds three hexamers together via the final alpha-helix (α11).

Much less is understood about the immature Gag particle. The size and flexibility of the domains of Gag have precluded successful x-ray crystallography and nuclear magnetic resonance imaging of Gag. The immature lattice is made exclusively of hexamers and forms spherical particles by insertion of gaps into the lattice. Recent cryo-electron microscopy (cryoEM) and cryo-electron tomography (cryoET) of assembled truncated Gag molecules has provided models for immature assembly. The interhexameric dimerization interface is also present in immature particles but is predicted to look different than that of the mature capsid.

Using Rous sarcoma virus (RSV), we tested the new structural model of the immature interhexameric dimer interface by screening residues predicted to be at the interface. The failure

iii of some of the alanine-substituted Gag proteins to assemble provides the strongest support to date that the current immature CTD dimer model is representative of the Gag lattice. Three of the mutants tested allowed Gag assembly but altered in vitro assembly of CA consistent with predictions from structural models. In total, these data support the hypothesis that the immature and mature dimers are different and regulated by distinct residues.

A novel mature CTD-CTD interaction was recently described for the three-fold axis of symmetry. Mutagenesis of RSV T214, selected for its potential influence at several stages of retrovirus production, provided support that the residue is critical for assembly of the mature capsid. Alteration of capsid stability for T214 mutants and the location of the residue in the loop between α10 and α11 of the CTD are consistent with structural models of the mature three-fold interface.

CA-SP, the capsid protein cleavage intermediate, assembled into the same types of particles as CA, albeit faster, and was able to nucleate CA assembly suggesting that CA-SP could be the first stage at which mature capsid assembly occurs. The intermediate may also serve as a nucleation point for capsid assembly. NMR analysis of the CTD-SP protein pointed to a potential close association between SP and the major homology region (MHR) and α9 of the CTD. The interaction is consistent with mutagenesis of CA in which SP mutations are able to restore infectivity to non-infectious MHR mutant viruses and the identification of maturation inhibitor escape mutants in HIV-1 that map to MHR residues.

Reported maturation effects by MHR and SP mutations pointed to role for these two regions in maturation regulation. The results presented in this dissertation support a model in which the capsid protein intermediate CA-SP establishes mature contacts in an SP-mediated fashion. In this model, residues of SP transiently interact or closely associate with the MHR and

α9, causing slight conformational changes that break immature interactions and generate mature contacts in both the dimer and three-fold interface. By this mechanism, the entire CTD and SP coordinate with each other to regulate the outcome of maturation.

TABLE OF CONTENTS

LIST OF FIGURES ...... viii LIST OF TABLES ...... ix LIST OF ABBREVIATIONS ...... x ACKNOWLEDGEMENTS ...... xiii CHAPTER I ...... 1 INTRODUCTION TO RETROVIRUS ASSEMBLY AND MATURATION ...... 1 OVERVIEW OF RETROVIRUSES ...... 2 Importance of Retroviruses ...... 2 Classification...... 2 ANATOMY OF RETROVIRUSES ...... 3 Gag Polyprotein ...... 4 MA ...... 4 CA ...... 7 NC ...... 9 PR ...... 10 Smaller Peptide Components ...... 10 THE RETROVIRUS LIFE CYCLE ...... 11 Early Events – Establishment of Infection ...... 11 Binding and Entry ...... 11 Disassembly and Reverse Transcription ...... 11 Integration ...... 13 Late Events – Virion Production ...... 14 Transcription/Translation ...... 14 Assembly and Budding of Immature Particles...... 15 CAPSID UNCOATING LEADS TO REVERSE TRANSCRIPTION/INTEGRATION ...... 17 Optimal Stability of the Capsid Leads to Reverse Transcription ...... 18 Influence of Cyclophilin A on HIV-1 Infectivity ...... 18 Uncoating and Nuclear Import ...... 19 Model for Uncoating and Nuclear Import ...... 19 THE COMPLEX CHOREOGRAPHY OF MATURATION ...... 20 Requirement for Maturation and Morphological Changes ...... 20 Immature Particle Structure ...... 22 Mature Particle Structure ...... 26 Ordered Protein Processing ...... 27 RNA Maturation ...... 28 Changes in Protein Activity ...... 30 ASSEMBLY INTERFACES ...... 31 Immature Contacts ...... 31 Intrahexameric Interactions...... 31 Interhexmeric Interactions ...... 32 Mature Contacts ...... 33 NTD-NTD ...... 33 NTD-CTD ...... 36 CTD-CTD ...... 37 Hexamers vs. Pentamers ...... 38 INFLUENCES ON CAPSID MATURATION ...... 39 The β-hairpin ...... 40 The Spacer Peptide ...... 40 The CTD 310 and α9 Helices ...... 42

The Major Homology Region ...... 43 REMAINING QUESTIONS AND AIMS OF THE DISSERTATION ...... 44 CHAPTER II ...... 47 MATERIALS AND METHODS ...... 47 CELL CULTURE AND VIRUS SPREAD ...... 48 PARTICLE RELEASE ASSAY ...... 48 DETERGENT RESISTANCE ASSAY ...... 49 PROTEIN PURIFICATION ...... 49 Full-Length Capsid Protein ...... 49 C-Terminal Domain Protein ...... 50 Gag (ΔMBDΔPR) Protein ...... 51 IN VITRO CA ASSEMBLY ...... 51 IN VITRO GAG ASSEMBLY ...... 52 ELECTRON MICROSCOPY ...... 52 Negative Stain ...... 52 Thin Section ...... 52 CIRCULAR DICHROISM AND UNFOLDING EQUILIBRIUM ...... 53 ISOTHERMAL TITRATION CALORIMETRY ...... 53 NUCLEAR MAGNETIC RESONANCE ...... 53 WESTERN BLOT ANALYSIS ...... 54 CRYO-ELECTRON MICROSCOPY AND SINGLE PARTICLE RECONSTRUCTION ...... 54 CHAPTER III ...... 56 THE 310 AND α-9 HELICES ARE A CRITICAL MATURATION SWITCH ...... 56 CHAPTER IV ...... 87 A ROLE FOR CA-SP IN NUCLEATING CAPSID MATURATION ...... 87 ABSTRACT ...... 88 IMPORTANCE ...... 88 INTRODUCTION ...... 89 RESULTS ...... 93 The Monomeric CA-SP and CA-S...... 93 SP Confers Distinctive Protein Assembly Properties ...... 93 CA-SP Nucleates Mature CA and CA-S Assembly ...... 99 CA and CA-SP Form Structurally Similar Assembly Products ...... 99 Probing the Amino Acid Sequence of SP ...... 107 Dynamics of the Spacer Peptide ...... 107 DISCUSSION ...... 108 ACKNOWLEDGEMENTS ...... 114 CHAPTER V ...... 115 IMPORTANCE OF THREONINE 214 AT THE THREE-FOLD INTERFACE ...... 115 INTRODUCTION ...... 116 RESULTS AND DISCUSSION ...... 120 Virus Spread and Particle Release ...... 120 In Vitro Assembly of Particles ...... 122 Stability of Virus Cores ...... 125 Packaging of TrpRS ...... 127 Overall Impressions of T214 ...... 129 ACKNOWLEDGEMENTS ...... 129 CHAPTER VI ...... 130 DISCUSSION: IMPORTANCE OF THE C-TERMINAL DOMAIN DURING MATURATION ...... 130 SUMMARY OF DATA CHAPTERS ...... 131

ASSEMBLY OF IMMATURE PARTICLES ...... 133 Influence of the Dimerization Interface on Immature Assembly ...... 133 Models of Immature Assembly ...... 135 INTERACTIONS LEADING TO MATURE CAPSID ASSEMBLY ...... 139 The Role of SP ...... 139 The CTD Dimer Interface ...... 141 Importance of CTD Dimerization ...... 143 The Three-Fold Interface ...... 144 Maturation of the CTD ...... 145 The Role of the NTD ...... 147 Formation of the β-hairpin ...... 147 Necessity of the Flexible Loops ...... 147 De Novo Assembly of the Mature Capsid? ...... 148 LIMITATIONS AND ALTERNATIVE APPROACHES...... 150 FINAL THOUGHTS AND POTENTIAL APPLICATIONS ...... 152 REFERENCES ...... 153 APPENDIX ...... 179 IMPORTANCE OF THE NTD FLEXIBLE LOOPS ON MATURATION ...... 179 INTRODUCTION ...... 180 RESULTS AND DISCUSSION ...... 181 Analysis of in Vitro Assembled ΔMBDΔPR Particles ...... 181 In Vitro Assembly Properties of the D87E/A134V Protein ...... 184 ACKNOWLEDEMENTS ...... 186

vii

LIST OF FIGURES

Figure 1.1. The immature retrovirus particle ...... 5 Figure 1.2. The mature retrovirus particle ...... 6 Figure 1.3. The structure of the CA protein ...... 8 Figure 1.4. Schematic of the replication cycle of retroviruses ...... 12 Figure 1.5. Uncoating of HIV-1 capsid is linked reverse transcription and nuclear import ...... 21 Figure 1.6. Assembly and Budding of prototypic retroviruses ...... 23 Figure 1.7. Global arrangement of the Gag lattice ...... 25 Figure 1.8. The interactions of CA in the mature particle ...... 34 Figure 1.9. Models for the roles of Gag subunits in the immature and mature lattices ...... 41 Figure 3.1. The immature and mature dimer interfaces ...... 60 Figure 3.2. Effects of mutations on virus spread ...... 64 Figure 3.3. Effects of mutations on particle release ...... 67 Figure 3.4. Thin sections of transfected cells ...... 69 Figure 3.5. In vitro immature particle assembly ...... 72 Figure 3.6. Stability of mutant virus particles ...... 75 Figure 3.7. In vitro mature particle assembly ...... 77 Figure 3.8. Transmission electron micrographs of assembled CA mutants ...... 78 Figure 3.9. Effects of CTD mutations on dimerization ...... 80 Figure 4.1. Proteolytic Processing During Maturation ...... 90 Figure 4.2. Purification and characterization of capsid proteins ...... 94 Figure 4.3. Effects of SP on CTD-CTD dimerization ...... 95 Figure 4.4. Kinetics of CA, CA-S, and CA-SP assembly ...... 96 Figure 4.5. Co-assembly of CA, CA-S, and CA-SP in vitro ...... 98 Figure 4.6. Negative-stain reconstruction of CA-SP ...... 100 Figure 4.7. Reconstruction of CA-SP from Cryo-EM ...... 101 Figure 4.8. Fitting of CA domains into CA-SP T=1 particle density ...... 102 Figure 4.9. In vitro assembly of a SP-localized suppressor of F167Y ...... 104 Figure 4.10. Assembly of CA-SP-8NC and proteins with the D52A substitution ...... 105 Figure 4.11. Effects of mutations on CA-SP assembly ...... 106 Figure 4.12. Dynamics of the spacer peptide ...... 109 Figure 5.1. The CTD three-fold interface ...... 117 Figure 5.2. Alignment of retrovirus CTD sequences ...... 119 Figure 5.3. Infectivity and release of virus particles ...... 121 Figure 5.4. Morphology of released virus particles ...... 123 Figure 5.5. In vitro assembly of CA protein ...... 124 Figure 5.6. Virus core stability ...... 126 Figure 5.7. Packaging of cellular trpRS ...... 128 Figure 6.1. Assembly models of the immature lattice ...... 136 Figure 6.2. Assembly model for the mature dimerization interface ...... 146 Figure A.1. Cryo-EM micrographs of ΔMBDΔPR proteins ...... 182 Figure A.2. Radial density averaging of RSV ΔMBDΔPR particles ...... 183 Figure A.3. In vitro assembly of the flexible loop mutants ...... 185

viii

LIST OF TABLES

Table 3.1. Phenotypes of WT and mutant viruses and proteins ...... 65 Table 4.1. PRE NMR Measurements for CTD-SP S241L ...... 110

LIST OF ABBREVIATIONS

Ala, A Alanine

Arg, R Arginine

ASLV Avian sarcoma/leukosis virus

Asp, D Aspartic acid

CA Capsid protein

CD Circular dichroism

CryoEM Cryo-electron microcopy

CryoET Cryo-electron tomography

CTD C-terminal domain

Cys, C Cysteine

DEAE Diethylaminoethyl

DMEM Dulbecco's modified Eagle's medium

EDTA ethylenediaminetetracetic acid

Env Envelope glycoprotein

Gag Group specific antigen

GFP Green fluorescent protein

Gln, Q Glutamine

Glu, E Glutamic acid

Gly, G Glycine

His, H Histidine

HIV-1 Human immunodeficiency virus type 1

HSQC Heteronuclear single quantum coherence

Ile, I Isoleucine

IN Integrase protein

IP Immunoprecipitation

Leu, L Leucine

LTR Long terminal repeat

Lys, K Lysine

MA Matrix protein

Met, M Methionine

MHR Major homology region

MLV Murine leukemia virus

MMTV Mouse mammary tumor virus

M-PMV Mason-Pfizer monkey virus

NC Nucleocapsid protein

NMR Nuclear magnetic resonance

NTD N-terminal domain

Phe, F Phenylalanine

PIC Preintegration complex

PR Protease protein

PRE Paramagnetic relaxation enhancement

Pro, P Proline

RSV Rous sarcoma virus

RT Reverse transcriptase protein

RTC Reverse transcription complex

SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

SEC Size exclusion chromatography

Ser, S Serine

SP Spacer peptide

SU Surface protein

Thr, T Threonine

TM Transmembrane protein

Trp, W Tryptophan

Tyr, Y Tyrosine

Val, V Valine

xii

ACKNOWLEDGEMENTS

Life of a graduate student is never easy. From always rapidly coming deadlines to frequently failed experiments, research challenges us mentally and physically in every way possible. Perseverance and determination are the keys to success. In the end, this journey is fulfilling and dare I say fun. I am fortunate enough to have experience both good and the bad during the last 6 plus years.

No one person can weather this journey alone. I have the deepest appreciation for my mentor and advisor, Dr. Rebecca Craven. Her guidance and support during my time here has made me a better scientist and person. She has challenged me in ways no other person has, and I am forever grateful for it. I would also like to thank the other members of my committee, all of whom have helped me realize the potential in my research capabilities.

Friendship is the best medicine for overcoming difficult situations. I am fortunate to have made many great friends during my time here. Although there are too many to recognize individually, I am especially grateful for the friendship of Carol Wilson, Parvez Lokhandwala

Katrina Heyrana, and Roland Myers. Being able to chat with them during the work day made it much easier to come in every day. To Dr. John Wills and the members of his lab, I owe thanks for their ability to tolerate me during the crazy times. I am indebted to Drs. John Flanagan, Maria

Bewley, and Ira Ropson for always being available to talk with me about anything. Not only have they provided great guidance and support of my research endeavors, they have also been great friends.

Finally, I would like to thank my parents and family. Their encouragement during the last few years has been invaluable. Despite not really understanding what I do, they have always been willing and eager to listen to me talk about my work. I would not be here today without them.

xiii

CHAPTER I

INTRODUCTION TO RETROVIRUS ASSEMBLY AND MATURATION

OVERVIEW OF RETROVIRUSES

Importance of Retroviruses

Retroviruses are a diverse family of RNA viruses that cause significant diseases, including cancer and immunodeficiency in a wide range of animals. Until the discovery of human immunodeficiency virus type 1 (HIV-1), avian retroviruses including the Rous sarcoma virus

(RSV) were utilized primarily to elucidate basic cellular processes as well as oncogenesis. As a result, RSV has been at the forefront of scientific discoveries. Evidence for virus-induced oncogenesis was first provided by Payton Rous in 1911 using his eponymous virus RSV (1).

Later the oncogenic property of the virus was assigned to the viral src oncogene which directly led to the discovery of cellular oncogenes. Among the more critical discoveries using RSV was the identification of reverse transcriptase, which has become an invaluable tool in molecular biology and allowed the investigation into reverse transcription events critical to human life (2,3).

Both retrotransposable elements and telomeres are generated through reverse transcription. More recently gene therapy has relied on retroviruses to efficiently deliver gene products to target cells.

Classification

The retrovirus family is broken into two distinct groups: Orthoretrovirinae and

Spumaretrovirinae (the latter will not be described further in this document). The orthoretroviruses (also referred to as retroviruses) are further classified into 6 genera by sequence relatedness with the prototype virus for each in parentheses: Alpharetrovirus (RSV),

Betaretrovirus (mouse mammary tumor virus, MMTV), Gammaretrovirus (murine leukemia virus, MLV), Deltaretrovirus (human T-lymphotropic virus type 1, HTLV-1), Epsilonretrovirus

(Walleye dermal sarcoma virus), and Lentiretrovirus (HIV-1) (4).

Two other classification systems were used for retroviruses, though advances in genetics made their use more archaic. Retroviruses were initially classified by sites of assembly of the immature particle (4,5). Type A retroviruses, such as the human endogenous retrovirus, bud into the endoplasmic reticulum but fail to leave the host cell. Type B and D viruses assemble particles

2 in the cytosol – type B viruses (MMTV) form cores with rounded morphology, whereas type D viruses (Mason-Pfizer monkey virus, MPMV) are more tubular in shape. Type C viruses (RSV and HIV-1) assemble at the plasma membrane and complete assembly as the virus buds from the cell (4,5). A broader classification system separated retroviruses into simple (like RSV and MLV) and complex (like HIV-1 and HTLV) viruses. These groups were based on the number of spliced mRNAs that the virus makes. Most simple retroviruses make no more than one, which codes for the envelope protein, and none produce multiply spliced mRNAs. In contrast, complex retroviruses encode multiply spliced mRNAs and produce a variety of accessory proteins that aid in replication and infection (6).

ANATOMY OF RETROVIRUSES

Retroviruses are a family of enveloped RNA viruses that convert their RNA genomes into DNA in a process called reverse transcription. Viruses contain two copies of the positive- sense linear, single-stranded RNA. The genome is 7-11 kb long, has a 5’ cap, and is 3’ polyadenylated, making it indistinguishable from cellular mRNAs (7-11). The genome is packaged into virions that range from about 100 to 200 nm; RSV, the focus of this dissertation is

100-120 nm in diameter. Upstream and downstream sequences termed long terminal repeats

(LTR) flank the open reading frames of the genome and serve as regulatory units for various host and viral proteins and are important for integration of the reverse transcribed DNA proviral genome (12-14). Similar to the distantly related hepadnaviruses, retroviruses reverse transcribe their RNA genome into a double stranded copy of DNA and then integrate that DNA into the host

DNA to create a provirus.

The genome of all retroviruses encodes three gene products: gag, pol, and env. The gag gene encodes the structural proteins, pol produces the replicative enzymes reverse transciptase and integrase, and env yields the envelope glycoproteins (7-11). For the orthoretroviruses, Pol is produced only as a fusion of gag and pol via a frameshift or other translation read-through of the

Gag stop codon leading to the production of Gag-Pol accounting for approximately 5 % of the

3 structural protein produced (15). Depending on the genus of retrovirus, the PR domain may be located at the C-terminus of Gag (RSV) or at the N-terminus of Pol (HIV-1). Simple retroviruses, such as RSV, contain only the structural genes and, in some cases, an oncogene (6). Complex retroviruses, like HIV-1, produce the 3 primary proteins as well as multiply spliced mRNAs coding for a variety of accessory factor proteins that assist at different stages of virus replication.

Retroviral structural proteins are initially synthesized as part of the Gag precursor protein, which forms around the viral genome. Initial assembly leads to budding and release of immature, non-infectious particles made of radially aligned Gag molecules into spherical cores

(Figure 1.1). The nucleocapsid domain (NC) is in the interior of the protein bound to the genome, the capsid domain and spacer peptide (CA and SP respectively) form the major Gag-Gag contacts, and the matrix domain (MA) attaches to the viral envelope. Surface (SU) and transmembrane (TM) proteins make up the envelope protein complex (Env) that binds to cell surface receptors and initiates fusion of the cell and viral membranes. The Pol proteins are located inside of the Gag shell as part of the Gag-Pol fusion protein (16-20).

Maturation of the immature particle generates an infectious virion. It is triggered at the time of budding and results in cleavage of Gag and Gag-Pol by virus encoded protease into structural proteins (16-20). Maturation results in widespread changes in the structure of the virion interior as well as biochemical changes in the proteins and RNA that prepare the virus to infect a new cell (Figure 1.2).

Gag Polyprotein

The matrix protein (MA) is at the amino terminus of the Gag polyprotein. Its primary functional role in assembly is to target the Gag protein to the plasma membrane either after formation of the immature particle (Type B/D viruses) or to begin the process of immature assembly (Type C viruses). For the majority of retroviruses species, membrane targeting and

Figure 1.1. The immature retrovirus particle. Retroviruses are initially produced has immature particles made of the structural protein Gag, radially aligned and attached to the viral envelope via the MA domain (red). Gag and Gag-Pol bind to two copies of the viral RNA genome (black) via the NC domain (blue). The two domains of CA (orange) provided the major intermolecular contacts holding Gag molecules together. The RSV protease (PR, yellow) forms the C-terminus of Gag and its activation initiates maturation of the particle. A varying number of the Env complexes (purple) stud the viral envelope and determine cell tropism and cause fusion of the virus with the host cell. Figure from Jared Spidel.

Figure 1.2. The mature retrovirus particle. A) After cleavage by PR, the structural proteins remodel the interior of the virion. MA (orange) remains attached to the viral envelope. CA

(orange) undergoes significant reorganization to form the mature capsid. NC (blue) and the viral genome (black) condense along with the replicative proteins (green) to form the reverse transcription complex (RTC) inside the capsid. B) Cryo-electron tomograms of RSV and three- dimensional computer modeling of the cores show the polymorphic nature of the mature capsid.

Adapted from Butan et.al. 2008.

6 binding is facilitated by myristylation of the N-terminus of MA (21). The MA protein also appears to serve other functions during production of virions. Many MA proteins contain stretches of positively charged residues postulated to be involved in an interaction with Env and acidic phospholipids, providing a possible mode of targeting viruses to specific sites of budding

(22-26). Though NC is the primary RNA binding segment of Gag, MA has both a nuclear localization signal and a detectable ability to bind RNA (27,28). These two features may be important for the transient trafficking of Gag to the nucleus, possibly to pick up newly transcribed

RNA and allow MA to accompany newly reverse transcribed DNA into the nucleus prior to integration (29,30).

The capsid protein (CA) is the largest and arguably the most important structural domain of the Gag protein and is located in the middle of the Gag molecule (20). Gag mutagenesis experiments revealed that the CA domain provides the major interaction interfaces that hold together Gag molecules and is the sole protein component of the mature capsid (31-40). CA is also critical for host virus interactions, as many cellular restriction factors bind to the mature capsid after entry into a cell.

Overall, CA is a largely globular protein made up of 11 alpha-helices (α1-α11) organized into two structural domains – the N-terminal domain (NTD) and C-terminal domain (CTD) – that are separated by a short flexible interdomain linker of 4-7 amino acids (41-46) (Figure 1.3).

Seven distinct helices (α1-α7) are present in the NTD and are preceded by a β-hairpin structure located at the amino terminus whose formation is required for maturation. The β-hairpin is established by an ionic interaction between the highly conserved amino-terminal proline and an aspartic acid residue at the top of α3 at position 52 (RSV), 51 (HIV), or 54 (MLV). Although critical for the success of the virus, the β-hairpin does not appear to directly interact with any specific interface of the mature capsid, indicating that its formation indirectly influences

Flexible loops

MHR

Figure 1.3. The structure of the CA protein. The ribbon diagram of the alpha-helical mature CA protein is shown with the NTD (red) and CTD (blue). The two domains are connected via 4-5 amino acid flexible linker depicted as dashed lines between α7 and the 310 helix. The figure was created from the PDB files: 1em9 (NTD) and 1d1d (CTD).

8 interfaces of other regions of the capsid (37,47-49). Of the 7 helices of the NTD, α1-4 and α7 form an arrowhead shaped bundle. A flexible region spans the end of α4 and includes α5 and α6 terminating with the beginning of α7 (48,50-56). This flexible loop region is the major source of structural differentiation amongst retroviral NTDs and serves as a prime target for host protein interactions with CA as these loops are surface exposed in mature capsid.

The CA CTD consists of 4 alpha-helices with a short 310 helix located at the N-terminus between the interdomain linker and α8. The 310 helix, which is made of an i+3 backbone hydrogen bonding network in contrast to the i+4 bonding of alpha-helices, along with the interdomain linker, provides both flexibility to the spatial arrangement of the NTD and CTD relative to each other and interactions in the immature and mature capsids (38,40,41,52,53,57,58).

A region composed of about 20 amino acids including the first α-helix of the CTD (α8) is named the major homology region (MHR) as it contains the region of greatest sequence homology among retroviruses as well as other reverse transcribing elements, suggesting a critical function for this region (59). In fact, mutations within the MHR profoundly affect both immature assembly and maturation by influencing the structure of the protein presumably by transmitting structural signals over long distances (60-63). The most important function of the CTD is its ability to self- associate, providing the major interactions that enable the assembly of both Gag and CA. As will be discussed in more detail, the dimerization helix (α9) contributes to the vital CTD-CTD interfaces in both immature and mature dimers, though these interfaces are quite distinct

(38,52,64). The final two helices of the CTD have recently been identified as sites of CA-CA interaction and CA-host protein interaction (65-67).

The nucleocapsid protein (NC) is a small basic protein 60-90 residues long located near the C-terminus of Gag and is the primary domain for RNA binding. The protein binds RNA through one or two Cys-His zinc finger motifs (68). Deletion of these residues abolishes packaging of the viral genome, whereas replacement of NC with another RNA-binding domain,

9 such as a leucine zipper, restores genome packaging (69-71). Additionally, NC contains both a nuclear localization signal to facilitate transport of the genome into and out of the nucleus and assembly domains that contribute to assembly and budding of immature particles (72).

Retroviruses produce their own aspartic protease (PR) that cleaves Gag and Gag-Pol into individual proteins during retrovirus maturation. In most retroviruses, the coding region for PR is part of the pol gene; however, it is located at the 3ʹ end of the RSV Gag gene. At some point during budding of the virion, homodimerization of the PR domain leads to its activation (16,73).

Failure of PR to be activated, or inhibition of PR activity, prevents maturation of retrovirions resulting in an aborted infection (74,75). Current antiretroviral cocktails typically include at least one PR inhibitor. In addition to specifically targeting the PR, a new class of PR inhibitors, called maturation inhibitors, block PR activity by binding to Gag.

Smaller Peptide Components

In the avian sarcosis/leukemia viruses MA and CA are separated by three small proteins, named p2a, p2b, and p10. Understanding of the functions of these proteins is still rudimentary; however, they appear to have a role in late stages of immature particle assembly and budding

(69,76). Additionally, the p10 domain contains a nuclear export signal that becomes exposed after

Gag binds to RNA, presumably to allow for export of the RNP complex from the nucleus.

Analogous to the p2 proteins, lentiviruses contain a 60 amino acid protein termed p6. Its role appears to be targeted to assembly of immature particles as wells as recruitment and packaging of some accessory proteins (77).

Several retrovirus genomes also encode small peptide domains between the larger structural proteins of Gag. The peptides, called spacer peptides, separate the CA and NC proteins

(SP in RSV and SP1 in HIV-1) and the NC and p6 proteins (SP2 in HIV-1). The function of

SP/SP1 will be addressed in more detail later. Briefly, it aids the tethering of six Gag molecules into a hexameric array in the immature particle (36,78). Recent evidence also suggests a role for

SP/SP1 in the maturation of retroviral capsids (62,78,79). Cleavage of CA-SP/CA-SP1 is the final step of proteolytic processing in RSV and HIV-1. Several other retroviruses also contain amino acid segments not necessarily designated as spacer peptides that likely function in ways similar to the spacer peptides of RSV and HIV-1.

THE RETROVIRUS LIFE CYCLE

Early Events – Establishment of Infection

The life cycle of retroviruses is generally separated into two distinct phases – early and late. Early stages include entry into a host cell and reverse transcription and integration of the viral genome to create a provirus. These events result in the stable persistent infection of a cell and prepare the infected cell to generate new virions (Figure 1.4).

Binding and Entry

The Env surface (SU) and transmembrane (TM) proteins form a trimer of heterodimers on the surface of the retrovirus lipid envelope and determine cell tropism by binding to specific cell surface receptor molecules (80-85). HIV-1, which infects T-cells, binds to CD4, whereas

RSV utilizes tva, tvb, or tvc molecules as its receptor (86). Entry into cells can be achieved by one of two methods – direct fusion of viral envelope with the cell membrane or receptor mediated endocytosis. Most retroviruses are thought to use only one of the entry methods; however, RSV is capable of entering through both pathways (82,87-89). Regardless of the means of cellular entry, retroviruses complete the rest of their life cycles in much the same manner.

Disassembly and Reverse Transcription

After internalization, the reverse transcription complex (RTC), which includes the capsid and RNP, traffics to the nucleus of the cell. Partial dissolution of the capsid prepares the RTC or preintegration complex (PIC) for integration (90). Further analysis of uncoating and its contribution to reverse transcription will be addressed in the next section. Reverse transcription is accomplished by the RT protein in all retroviruses. In vitro, this is the only protein needed to

Figure 1.4. Schematic of the replication cycle of retroviruses. Viruses to cellular receptors and enter cells. After a process of uncoating and trafficking intracellularly, reverse transcription of the genome occurs to generate the proviral genomic DNA. Integration of the provirus into the host genome eventually leads to transcription and translation of viral proteins by cellular machinery.

Gag, Gag-Pol, Env and the newly generated viral RNA genome traffic to the site of assembly.

Immature particles bud from the plasma membrane and undergo a complex process of maturation to generate infectious viruses. Figure adapted from Ganser et.al. 2008.

12 create DNA from an RNA template; however, it is clear that other retroviral proteins, including

NC and CA, affect the rate and efficiency of RT activity in vivo (59,63,91-95). RT generates

DNA through a complicated series of steps. The details are briefly summarized here. Synthesis is initiated from a tRNA (Lys in HIV-1 and Trp in RSV) packaged from the previously infected cell

(96-98). Negative-sense strand DNA is produced that extends to the 5’ end of the RNA genome.

This DNA segment is called strong stop and its synthesis continues at the 3’ end of the genome following strand transfer. During negative-stand DNA synthesis, RT, which also contains RNase

H activity, degrades most the RNA template. Positive-sense DNA synthesis initiates on the small segment of RNA that was not degraded by RT and is completed by a strand-transfer mechanism similar to the one that generated the negative-sense strand. Completion of reverse transcription creates the pre-integration complex (PIC), which includes the DNA, integrase (IN), and other viral and host proteins.

Retroviruses are notorious for their ability to mutate due in large part to the lack of a proof-reading mechanism in the RT enzyme (99-101). The experimental mutation rate for the

RSV RT is approximately 1.5 mutations per 104 bases (102). The mutation rate and quick replication rate produce a heterogeneous population of viruses within a single host system and even within a single cell (103-105). Further genetic mutations are introduced by the host RNA polymerase II (RNAP II) during transcription of the proviral genome (106). The full-length transcript serves two roles for the virus – it is mRNA for the Gag and Pol proteins and is also the viral genome that gets packaged into new virions. Genetic diversity of HIV-1 is a leading cause of antiviral resistance and is a serious obstacle to generation of HIV-1 vaccines and complications of drug resistance.

Integration

Infection of a host cell is not completed until the retrovirus integrates its newly synthesized DNA genome into the host genome. All retroviruses use the IN protein produced from the pol gene product and require trafficking of the PIC to the nucleus of the newly infected

13 cell. Most retrovirus PICs are not actively transported into the nucleus and therefore can only infect actively mitotic cells which break down the nuclear envelope during mitosis (107-109).

The exception is lentiviruses and spumaviruses which do actively transport PICs to the nucleus and thus can infect non-dividing cells (110). Upon gaining access to the cellular genome, the IN protein generates nicks at the ends of the viral DNA genome and the site of cellular genome integration (111-113). The protein then produces a ligation event that joins the cellular and viral genomes to create the provirus which can then by transcribed by cellular polymerases.

Late Events – Virion Production

The establishment of an infected cell during the early stage is followed by transcription and translation of retroviral proteins along with assembly, budding, and maturation of virions during the later stages of the life cycle. Late stage events ultimately lead to the synthesis of new virions that go on to infect naïve cells.

Transcription/Translation

Transcription and translation of the proviral genome are entirely dependent on host cellular factors – RNA polymerase and ribosomes. Viral transcripts are generated by the host

RNAP II beginning at the 5’ LTR (114). Simple retroviruses utilize host factors interacting with transcription enhancers in the LTR to regulate transcription, whereas, complex retroviruses also encode accessory proteins, such as the HIV-1 Tat protein, to assist in the regulation of transcription. The resulting transcript can be variably spliced to generate full-length messenger/genomic RNA, Env mRNA, or accessory protein mRNA (114).

Export of the RNA follows transcription and is mediated by a variety of cellular and viral proteins. Retroviruses contain cis-acting elements for export of their spliced RNAs in a fashion similar to host cell mRNAs (115,116). This element, termed the constitutive transport element

(CTE) located near the 3’ LTR is bound by host export factors. Unspliced RSV transcripts are similarly exported by host factors that bind to the direct repeats (DR) located at either end of the

14 src gene. Complex retroviruses contain trans-acting elements that facilitate the export of the viral unspliced and singly spliced RNA.

Recent evidence suggests a role for the Gag protein in the export of viral RNAs at least in avian retroviruses. The Gag protein of several retroviruses, including RSV and HIV-1, traffics to the nucleus where its function is under intense investigation (117-120). It is postulated that the

RSV Gag protein binds to the psi sequence at the 5’ end of the RNA and facilitates nuclear export in a CRM-1 dependent mechanism (121). Blockage of Gag trafficking to the nucleus significantly reduces packaging of the viral genome into virions suggesting that this nuclear trafficking event improves, but may not be absolutely required, for efficient virus production (72,122).

Translation of RSV and HIV-1 mRNAs is 5’ cap-dependent like that of cellular mRNA translation. Some retroviruses, like MLV and SIV, initiate translation through internal ribosome entry sites (123,124). A unique feature of retrovirus translation generates Gag-Pol, which is made by a ribosomal (-1) translation frameshift during the translation of Gag or read-through of the

Gag stop codon. This event is relatively rare, resulting in a 20:1 ratio of Gag to Gag-Pol. Most notably, this frameshift leads to differences in the production of the PR protein arising from differences in the genetic organization of retroviruses (15,125-128). The RSV PR protein is part of Gag, whereas in many other viruses it is part of Pol, thus resulting in RSV containing 20-fold more PR (125-128). The significance of this difference in abundance is not known.

Assembly and Budding of Immature Particles

The site of particle assembly and the mechanism of trafficking of Gag to the site of assembly remain much debated topics in the retrovirus field. Initially it was assumed that Gag either trafficked directly to the site of assembly or began to assemble on its way to the site of immature particle formation. Recent evidence from investigations with RSV suggests that the initial oligomerization of Gag occurs in the nucleus of the cell where Gag selects the viral genome as described above (117).

At some point, oligomers of Gag traffic to the plasma membrane to produce the immature particle. The mechanism of Gag or RNA trafficking is not entirely clear and many studies have provided conflicting evidence. In some studies, HIV-1 Gag has been shown to associate with a variety of microtubule proteins, including Kif4A and Kif3A; however, others have shown that disruption of microtubule networks has little or no effect on virus release (129-132).

Regardless of the pathway for getting to the plasma membrane, the Gag molecules of many retroviruses, including HIV-1 and M-PMV, are co-translationally myristoylated (133,134).

Upon localization to the acidic environment of the budding site, the myristoyl group of MA inserts into the plasma membrane. At the same time acyl groups of plasma membrane lipids interact with MA (135-137). The plasma membrane at the site of retrovirus budding contains many acidic phospholipids, the most important of which is proposed to be PI(4,5)P2. The N- terminal half of the MA domain is enriched with basic residues that bind to the acidic phospholipids (138,139). Insertion of the myristoyl group and binding of MA to various phospholipids facilitates attachment of Gag to the plasma membrane. Targeting of the RSV Gag, which is acetylated rather than myristoylated, is also dependent on the presence of acid phosphoinositol species and cholesterol (138-144).

Synthesis and plasma membrane targeting of the Env proteins are entirely independent of the same events for Gag and Gag-Pol. After translation, Env is transported to the endoplasmic reticulum by a hydrophobic signal peptide in the N-terminus of the protein. In the ER, the signal peptide is removed by host peptidase and the Env protein is glycosylated. After transitioning to the Golgi apparatus, the protein is further cleaved by host proteases to generate the SU and TM proteins, which remain associated through disulfide bonds (145,146). More post-translational modifications occur as the proteins move to the plasma membrane. Incorporation of the Env proteins into virions is not strictly required for budding of retroviruses, but is required for infectivity (25). Association of Env with the Gag or Gag-Pol MA domain (described further in the

Maturation section) ensures proper inclusion of the glycoproteins.

The immature particle initiates budding from the infected cell. Host cell factors help pinch off of the virion, releasing it from the cell. This pinching off process appears to be directed by the L domains of viruses, found in p6 of HIV-1 and p2b of RSV, and may also be dependent on some accessory factors produced by complex retroviruses (147,148). The L domains recruit endosomal sorting complex required for transport (ESCRT) proteins leading to the completion of membrane pinching and budding of the virion (149-153). Concomitant with the later stages of budding, the viral PR becomes active via dimerization and begins cleaving Gag and Gag-Pol into individual proteins. The cleavages begin the highly coordinated process of maturation

CAPSID UNCOATING LEADS TO REVERSE TRANSCRIPTION/INTEGRATION

The pathway to infection is a dangerous journey for the viral genome. It must survive under many hostile conditions, as host cells contain many enzymes and innate immune factors to recognize and eliminate foreign DNA and RNA. Additionally, the genome must traffic to specific sites within a cell at precise time points to propagate infection. To overcome these obstacles, viruses produce capsids, simple or complex proteinaceous structures that serve many different functions during the virus life cycle. What, however, is the function of retrovirus capsids? They have envelopes to surround the ribonucleoprotein complex and NC protein to coat the viral genome, similar to negative-strand RNA viruses which do not have capsids. Therefore, one would expect the presence of a capsid to be redundant. However, a good retroviral capsid is critical to infectivity of the virus. The mature capsid plays an important role in leading to successful reverse transcription and nuclear entry of the preintegration complex (PIC) by interacting with various host cell factors.

As briefly described above, retroviruses must reverse transcribe and integrate their proviral genomes. Prior to these events the protective capsid shell must undergo a disassembly process termed uncoating. There are still many aspects of the process that are not clear, but uncoating is regulated by interactions between the capsid and several host cell proteins.

Investigating disassembly of the capsid is difficult due to a lack of good experimental approaches,

17 the heterogeneity of mature capsids, and the transient nature of the intermediates. Most of our knowledge comes from mutagenesis of the CA protein and host cell factors.

Optimal Stability of the Capsid Leads to Reverse Transcription

RTCs or PICs can be isolated from the cytoplasm of infected cells. When these complexes were fractionated through detergent and sucrose, early studies found that little CA was associated with RTCs/PICs. However, the optimal stability of HIV-1 and RSV capsids is correlated with successful reverse transcription (37,59,63,91,95,154). In addition to CA-CA interactions, host restriction factors influence the stability of retroviral capsids. Most of the restriction factors identified have been done so for HIV-1. Many of the restriction factors bind directly to the HIV-1 capsid by interacting with the flexible loop region of the CA protein. These studies highlight the importance of the capsid in during reverse transcription and integration and have aided the design of potential anti-retroviral therapeutics.

One such restriction factor, TRIM5α, binds to the HIV-1 capsid and destabilizes the interactions. This interaction leads to proteasome-mediated elimination of CA and blocks reverse transcription of the viral genome (155-158). PF74, a small molecule inhibitor of HIV-1, binds to the same region of CA as TRIM5α and leads to destabilization of the capsid in vitro. The inhibitor also blocks infectivity of the virus. HIV-1 escape mutants map to the flexible loop region of the CA protein and stabilize the capsid structure (159-162). On the opposite end of the stability spectrum, small molecule inhibitors BI-1 and BI-2 increase stability of the capsid. These molecules also decrease reverse transcription and therefore block infectivity (163,164).

Influence of Cyclophilin A on HIV-1 Infectivity

The peptidylprolyl isomerase cyclophilin A (CypA) is among the best characterized HIV-

1 capsid binding host restriction factors. CypA binds to the flexible loop region of CA between

α4 and α5 (51). Addition of cyclosporine A inhibits binding of CypA to the capsid and prevents reverse transcription of the viral genome (165,166). Mutations that render the virus CypA- independent or even cyclosporine A-dependent map to the CypA binding site or other NTD

18 flexible loop residues (167,168). These mutations stabilize the capsid, delaying the onset of uncoating in a manner similar to the CypA protein. In addition to its influence on reverse transcription, CypA also plays a role in nuclear entry of the HIV-1 PIC. Unlike most retroviruses, lentiviruses can infect non-dividing cells, indicating that there must be active import of the PIC into the nucleus. Many of the CA mutants that render the virus CypA-independent only infect dividing cells. To infect non-dividing cells, like macrophages, these mutant viruses require the

CypA-CA interaction (169,170). The binding of CypA and CA also contributes to hiding the

HIV-1 capsid from intracellular innate immune recognition proteins. WT viruses generally do not induce interferon production. However, when CypA is depleted, robust activation of interferon production commences (171,172).

Uncoating and Nuclear Import

Transportin 3 (TNPO3) a nuclear transport protein was identified as a necessary cellular factor for HIV-1 infectivity specifically at nuclear entry and integration. Mutations that alter capsid stability also show a decreased dependence on TNPO3. Mixtures of purified TNPO3 protein and HIV-1 cores increase core disassembly in vitro, which could be blocked by the inclusion of CypA. TNPO3 was originally thought to bind to CA. However, it is now known to interact with the cellular protein cleavage and polyadenylation specificity factor 6 (CPSF6).

CPSF6 is a nuclear shuttling protein that aids in nuclear import of the HIV-1 PIC. Truncations that restrict CPSF6 to the cytoplasm inhibit HIV-1 infection by blocking nuclear import.

Mutations that render the virus TNPO3-independent make the virus CSPF6-independent as well.

Therefore CPSF6 and TNPO3 are thought to work cooperatively to allow the HIV-1 PIC to bind to nucleoporins. Binding to nuclearporins enables the PIC to enter the nucleus.

Model for Uncoating and Nuclear Import

Retroviral capsid uncoating is a critical and highly regulated process that leads to the reverse transcription of the viral genome and integration of the provirus. Uncoating depends on a capsid that has enough integrity to survive attack from host restriction factors to allow trafficking

19 of the PIC to the vicinity of the nucleus but is still can undergo subsequent disassembly. From the data presented above, the active model for retrovirus uncoating includes initial stabilization of the capsid by CypA to prevent premature disassembly. This allows the PIC to bind to TNPO3 and

CPSF6 facilitating import through the nuclear pore (Figure 1.5) (173). This model proposes that reverse transcription occurs near the nucleus and enables the viral core to bypass innate immune sensors. Binding of restriction factor TRIM5α or small molecule inhibitors promote early capsid disassembly inhibiting reverse transcription. Successful maturation, leading to infection of a cell, is dependent on the production of a good, fully functional and optimally stable capsid.

THE COMPLEX CHOREOGRAPHY OF MATURATION

All retroviruses, with the exception of Spumaviruses, undergo an obligatory maturation process in which the non-infectious, immature virus particle is converted into an infectious virion.

When initially assembled, either in the cytoplasm or at the plasma membrane, the cores of retroviruses have immature morphology, and eventually become more condensed. Condensation of electron density is associated with the PR-mediated proteolytic cleavage of the Gag and Gag-

Pol polyprotein precursors and structural changes in the genomic RNA (17-19). This dramatic change in the virion morphology is clearly the most visually stunning aspect of maturation; however, it is not the only feature of the process. Several other changes must occur to produce an infectious virus. Proper maturation of the capsid may be tightly linked to these other events. The changes to the virus during maturation are further described below.

Requirement for Maturation and Morphological Changes

Maturation is an elaborate process that is absolutely required in order for orthoretroviruses infectivity. Viruses unable to properly cleave Gag and Gag-Pol through genetic manipulation or chemical inhibition are severely crippled (174-178). The most visually impressive result of retrovirus maturation is the proteolytic processing of the structural protein precursors and subsequent reorganization of the virion interior. In the absence of protein processing, viruses retain immature core morphology in which the Gag and Gag-Pol proteins are

Figure 1.5. Uncoating of HIV-1 capsid is linked reverse transcription and nuclear import. A)

After entry host CypA binds to the capsid via the flexible loops of the CA NTD stabilizing the capsid structure until the core is trafficked near the nucleus. Binding by TNPO3 and CPSF6 aid in the initiation of uncoating; this in turn leads to reverse transcription. The PIC is then able to bind to nuclear pore proteins for import into the nucleus. B) Deletion or mutation of the capsid or host factors in (A) causes premature uncoating which blocks reverse transcription. C) Binding by antiretroviral host proteins, such as TRIM5α, leads to destabilization of the capsid preventing infection.

21 radially arranged and in close association with the viral envelope (75,179). The ability to distinctly visualize genomic RNA in mature particles is often absent in immature particles. After maturation there is a significant redistribution of protein and RNA within virions. The genomic

RNA condenses and the bulk of electron density is transferred from the envelope to a more central location in MLV, RSV, and HTLV virions and more eccentric in HIV and M-PMV virions (4,6) (Figure 1.6).

Aside from the visually stunning changes that occur during maturation, biophysical transformations occur as well. When produced with defective PR domains to prevent processing, the unprocessed immature RSV cores are highly resistant to disruption by non-ionic detergents

(176). After maturation and mature capsid formation, the virus core becomes more detergent sensitive and therefore less stable (59,176,180). It is likely that this loss of stability is associated with the capacity for dissociation of the capsid after entry and reverse transcription of the genome.

Immature Particle Structure

Assembled immature particles display a radial alignment of Gag molecules in which the

MA protein is attached to the viral membrane and NC binds to the viral RNA genome near the center of the particle (16-19). The domains of CA along with SP/SP1 provide the primary protein- protein contacts that create hexamers and that occur between hexamers (36,179,181-184). Unlike mature capsids which are often distinct for each retrovirus species, the immature particles are fairly uniform – spherical in shape, with size providing the significant difference between species.

Immature RSV with its tighter curvature is slightly smaller than HIV-1 and M-PMV though all of the molecular interactions described below apply to all retroviruses that have been examined in detail (181).

The immature lattice is made of hexameric arrangement of Gag molecules connected via their CA domains. Gaps are visible when particles are reconstructed from cryo-ET, indicating that

Figure 1.6. Assembly and Budding of prototypic retroviruses. Sequential transmission electron micrographs of budding and maturing retroviruses particles are shown from left to right. HIV (A) and RSV (C) assemble as the reach the plasma membrane, whereas MMTV (B) forms complete immature particles inside the cytoplasm. All three undergo maturation post-budding (far right).

23 the particles are not completely spherical, but rather spheroid in shape (182). However the lattice of immature particles is still continuous (Figure 1.7). Based on classic fullerene structure, pentamers are needed to relieve the strain of curvature and completely close a sphere made of a hexameric lattice; however, no study to date, structural or biochemical, has been able to demonstrate the formation of pentameric units of Gag protein (36,181,182,185). Whether it is not structurally possible to occur or just very energetically unfavorable and thus difficult to observe is not clear. In any case, the gaps appearing in native immature Gag assemblies are likely caused by the strain of curvature and serve the same functions as pentamers. Accordingly, it appears that the closure of the core is only necessary after maturation and that the genome can be well protected from the environment by the envelope.

Despite the relative similarity between immature particles, in actuality no two assembled particles are exactly alike due to variation in number of Gag molecules and the locations and sizes of gaps. This heterogeneity has caused difficulties in obtaining high resolution structures. More recently, cryo-ET and subtomogram averaging of tubes created from in vitro assembly of truncated Gag proteins has been analyzed to obtain higher resolution models of assembled immature particles (186,187). Each of the individual Gag components appears to retain their protein fold even when part of the Gag polyprotein. For example, the NTD and CTD of CA have individually been purified and their structures solved by either NMR or x-ray crystallography. For the most part, each of these structures fits neatly into the cryo-density from the truncated Gag protein, and the same can be said for the NC protein. Less is known for MA, as this domain was not included in the tubes of Gag. With respect to CA, it was reported that the NTD forms rings of density associated with hexamers and the CTD forms a similar structure directly beneath the

NTDs (36,186,187). For RSV, the spacing between hexamers is 8.0 nm, slightly closer than that seen in the mature capsid described below (20,181).

Figure 1.7. Global arrangement of the Gag lattice. Purified Gag molecules were assembled in vitro in the presence of nucleic acids and analyzed by cryo-electron tomography. The centers of each hexameric unit cell are marked with hexamers, which are colored according to cross correlation on a scale from low (red) to high (green). Higher cross-correlation values indicate that the subtomogram is more similar to the average structure. Defects in the form of gaps are found throughout each particle, thought the Gag lattice is continuous. Two representative particles are shown for each virus species. Adapted from de Marco et.al. 2010.

Mature Particle Structure

The highly polymorphic nature of retrovirus capsids makes their study extremely difficult. In addition to a large spectrum of available shapes, the number of CA molecules used to make up the capsid can vary. Depending on the overall shape of the capsid, from as little as 30% to as much as 80% of the available CA protein from Gag ends up in the mature capsid, as measured by cryo-ET of native capsids (25). This observation was supported by hydrogen/deuterium exchange in which both assembled and unassembled CA protein was found in virions, indicating that not all of the enveloped CA is used to build the mature capsid (188).

The mature capsid lattices of RSV are modeled as fullerene pleomorphic polyhedrons, whereas HIV forms fullerene cones (185,189,190). In these models up to 300 hexameric units of the CA protein are arranged and closed by insertion of 12 penatmers of CA. The total number of hexamers determines capsid size whereas shape depends on the spacing of the pentamers. RSV pentamers are generally distributed randomly leading to more rounded shapes, whereas HIV-1 forms cones because 5 pentamers localize to one end (the narrow one) and 7 to the other (42-44).

As might be expected, the tubular organization common to MPMV are generated by equal distribution of the pentamers to either end of the tube, thus providing caps similar to the cones of equal curvature (189).

Due to the large variation in native capsid morphology, high resolution imaging of mature contacts requires in vitro manipulations and in some cases mutagenesis. These techniques revealed that upon maturation, the NTD forms interactions at the exterior surface of the capsid, whereas the CTD is located on the interior of the capsid closer to the reverse transcribing machinery (38,43,191-194). Additionally, the NTD pinches inward, creating a more compact pore whereas the CTD shifts outward resulting in larger hexameric spacing than seen in the immature particle. These shifts also result in a change in the relative orientation of the domains to each other. In the mature form, the CTD of one monomer nestles underneath and in between its own

NTD and the NTD of a neighboring monomer creating a unique NTD-CTD intrahexameric interface (192,194).

Ordered Protein Processing

Pulse-chase experiments with maturing virions in infected cells allowed intermediate species of Gag cleavage products to be identified. Additionally, in vitro processing of Gag demonstrated that PR prefers certain cleavage sites over others (195-199). These results lead to the proposal that most of the retroviruses proteolytically process their Gag proteins in an ordered fashion. Each retrovirus species retains similarity in the pattern though minor differences occur due to the differences in Gag and Gag-Pol constituents amongst the genera. One common feature is that cleavage of the amino-terminus of CA precedes cleavage at the carboxy-terminus. In general, the first cleavage severs CA (or p2/p10 in RSV) from MA, leaving CA-SP-NC. This is followed by freeing of NC and producing CA-SP. In retroviruses that have a true spacer peptide, like RSV and HIV-1, the final cleavage separates CA from SP (35,180,199).

At least three critical factors contribute to the orderly processing of Gag. First, the sequence of each cleavage site is a key determinant of the rate of cleavage. Exchanging cleavage site sequences alters the peptide products produced during proteolysis (200-203). Second, the context of the cleavage site is important for regulating the processing rate. For example, until the severing of the SP-NC linkage, the cleavage site between CA and SP may not be readily available to PR. There may be conformation changes after SP-NC cleavage enabling PR to access the final cleavage site (200,201). Finally, the presence of RNA has been shown to influence the rate of cleavage, suggesting a role for packaging of the genome in proteolytic processing. The presence of the viral genome likely leads to conformation changes within the Gag and Gag-Pol proteins or intermediates that allow cleavage sites to be more or less accessible to PR, regulating the rate at which the enzyme can act (204-206).

Just as there are intermediates produced during Gag cleavage, there are likely to be capsid assembly intermediates. In vitro techniques have provided structural models for these

27 particles showing the differences in assembled states, but none are predicted to represent maturation intermediates. Recent cryoET imaging of HIV-1 particles with a newer class of antiretrovirals, which only block the final cleavage between CA and SP1, revealed that these particles retain both immature and mature lattice architecture (75). However, it is impossible to say that this artificial freezing of CA-SP1 truly represents an intermediate stage in maturation.

Further development of single particle tracking and imaging, both in vivo and in vitro, will be needed to visualize the presence of any intermediate assembly structures.

Several questions remain about the variability in capsid morphology. First, what determines placement of pentamers, as there are none found in the immature particle? Second, how is continuous curvature established for the capsid when there is only a single protein to work with (194)? In many cases, in vivo imaging of native capsids shows a close association of the capsid and the lipid bylayer, hinting at the possibility that this interaction influences capsid curvature and insertion of pentamers (25). It is also possible that at some point during formation of mature contacts, strain in protein packing expels a molecule of CA from a hexamer to generate a pentamer. Continuous curvature is created due to the flexible nature of the CA protein

(42,44,191). All of the interactions described later are not rigid and slight variations in interaction angles allow for a large number of interfaces. The interdomain linker separating the NTD and

CTD also possess significant flexibility, providing additional opportunities for variation in how the domains, and ultimately CA molecules, associate with each other.

RNA Maturation

Viral genomic RNA is found in virions as a dimer linked by weak non-covalent interactions (68). Electron micrographs of dimeric retrovirus RNA show loops near the 5’ ends of the molecules suggesting that the linkage occurs somewhere near the 5ʹ end, which is highly structured and features several stem loops (207,208). An interplay between the three-dimensional structure of RNA and the NC domain of Gag leads to the selection and packaging of the viral genome as a dimer.

The RNA assumes at least two dimeric states during maturation. The initial dimer is weakly associated and must evolve into a more stable form as it condenses with the NC protein during maturation. Electrophoretic analysis of genomic RNA of MuLV revealed that at early time points post budding, the RNA migrates more slowly that RNA harvested from virions at later times (209). In either case, mild heating of the RNA led to even faster migration. Melting curves for each at each of the RNA time points showed that RNA collected earlier was less stable than that harvested 24 hours post budding. These data suggest that the RNA initially packaged in immature virions is dimeric, though the dimer is less stable than that of matured viruses (209).

Further analysis of RNA dimer maturation through mutagenesis of cleavage sites revealed that dimer maturation actually takes place during the sequential processing of Gag requiring the severing of NC from CA or SP/SP1. Mutations can be introduced into the P1-NC cleavage site of HIV-1 to prevent PR-mediated cleavage. Viruses with these mutations do not form more stable RNA dimers (210). A similar feature is seen when mutations are engineered between the MoMLV CA and NC (211). In HIV-1 severing the SP1-NC bond initially releases the NC as the NC-p1-p6 intermediate species (212,213). Gammaretroviruses contain a conserved

12 amino acid motif in the CA NTD. Deletion of these residues in MLV allows complete cleavage by PR, but the virus remains in an immature assembly state. Likewise, a deletion of the

PTAP sequence in the HIV-1 p6 domain prevents maturation of the virus despite full cleavage of the Gag polyprotein. The dimeric RNA genome of both of these mutant viruses fails to undergo maturation (214). Taken together, all of these data support a model in which maturation of the genome is tightly linked to maturation of the capsid.

RNA maturation does not just affect the structure and biophysical properties of the RNA; it also has a strong effect on reverse transcription (212). The endogenous RT activity of viruses with different cleavage site mutations were compared for early and late RT products. HIV-1 virions, locked in the immature state, had a 10-fold reduction in the amount of early RT product and over a 100-fold decrease in late RT products (212). Mutant viruses inhibited at intermediate

29 stages of Gag cleavage, which had near WT levels of RNA maturation, had WT-like endogenous

RT activity. Exogenous RT activity for these mutants was similar to WT indicating that the RT defect was due to the lack of endogenous genome maturation (212). The greater decrease in the late product formation of immature viruses is a result of failed strand transfer in the middle of reverse transcription. This process is facilitated by the mature form of RNA dimer, not the immature form.

The structural integrity of the virion is also a product, at least in part, of proper genome maturation. Preventing HIV-1 RNA dimerization and dimer maturation by a stem loop deletion or mutagenesis leads to a delay in the final cleavage between CA and SP (215). In HIV-2 deletion of the first stem loop causes accumulation of MA-CA-SP1 product and a lack of mature cores (216).

Removal of the psi-packaging sequence from MuLV RNA also perturbs maturation. In this construct, the resulting VLPs contain slightly less total RNA. To accomplish this, the VLPs increased the packaging of mRNAs that were cellular in origin (217).

The stability of the structural proteins in the packaging deficient particles was also measured. Immature particles with a full complement of viral and cellular RNA are very stable in the presence of non-ionic detergent. Addition of RNase causes the particles to become susceptible to detergent, and the majority of the protein is released to the supernatant (217). Similar release is observed for the psi-packaging mutant, suggesting that the viral and cellular RNAs play a structural role in particle assmembly (217). These data point to a function for the viral genome in proper Gag-Gag interactions and Gag processing. Whether this is due to a scaffolding function in which the genome serves as an initiation point for assembly or whether the RNA influences dynamic movements of Gag to facilitate PR cleavages is still under investigation.

Changes in Protein Activity

In addition to changing the global structure of the virion, maturation also prepares the virus for infection of a cell by altering the activity of the structural and nonstructural proteins.

The immature particle is a very stably assembled structure despite having gaps present in the Gag

30 assembly. This stability is greatly diminished after maturation (176,218,219). The CA molecule undergoes significant structural changes, including formation of the β-hairpin at the N-terminus, sequential cleavage of the C-terminus, breaking of immature interaction forces, and establishment of mature contacts (220,221). These alterations transform the virion interior into a less stable core that somehow promotes reverse transcription and ultimately integration of the proviral genome.

The first step of maturation begins with activation of the PR component of Gag through dimerization PR. The level of PR activity in intact Gag (avian retroviruses) or Pol (HIV-1 and others) is difficult to accurately measure because PR is released so rapidly by self cleavage of the polyproteins (195,222). Release of PR from Gag enhances the activity of the protein. Blocking the release of PR by mutating the processing site in Pol or Gag renders the resulting viruses noninfectious, though PR retains function in vitro (196,223). In summary, release of the retrovirus PR protein at the early stages of maturation is needed for efficient and complete cleavage of the Gag and Pol proteins.

RT must also undergo maturation in order for the virus to become infectious. In experiments with disrupted virions, the activity of RT, whereas part of the precursor Pol protein, is variable dependent upon the species of retrovirus (176,219,224-226). VLPs with unprocessed protein or processing intermediates fail to reverse transcribe RNA in permissive cell lines, though some of these same proteins purified in vitro exhibit varying levels of activity (219).

ASSEMBLY INTERFACES

Immature Contacts

Gag is the only viral protein needed for assembly of immature VLPs both in vivo and in vitro as it contains all of the intermolecular contacts needed for protein association, membrane targeting and binding, and genome binding are found within Gag. There are two main Gag-Gag interactions that occur in the immature virion – intrahexameric and interhexameric – which involve different residues of the CA NTD and CTD.

Intrahexameric Interactions

Hexameric organization of Gag is the only known arrangement of the Gag protein in the immature lattice (45). Based on higher resolution modeling of in vitro assembled tubes of M-

PMV and HIV-1, the CA helices involved in the intrahexamer interactions of Gag include only the CTD MHR and α-11 helix (186,187). In stark contrast to the 43 residues involved in intrahexameric interactions in the mature capsid, only five residues are predicted to actually make the intrahexameric contacts in the immature particle (186,187). In addition to the CTD, other elements of Gag participate in assembly of immature particles (36,69). The SP and first few residues of NC also play critical roles in assembly of the immature lattice, specifically in forming intrahexameric interactions (36).

The SP/SP1 region of the Gag polyprotein in alpharetroviruses and lentiretroviruses has become a significant area of interest in recent years. Early evidence pointed to a critical role for

SP in assembly of the immature particles (62,75,227). Genetic manipulation of this region in

HIV-1 and RSV demonstrated that SP is required for proper Gag-Gag association and is a key determinant of immature particle size (180). Deletion of this region causes abnormal maturation or prevents particle release ultimately preventing infectivity. Additionally, preventing the final cleavage event which removes the SP/SP1 residues during maturation also prevents the virus from maturing into an infectious particle (74,228). Immature VLPs of RSV, M-PMV, and HIV-1 examined by cryo-ET show a pillar of electron density sitting between the CTD and NC

(36,181,182,229,230). This density led to modeling SP as a six helix bundle that aids the CTD in building hexamers (36). Biophysical conformation of this phenomenon showed that under certain conditions, a peptide spanning the interaction domain (including the C-terminus of the CTD, all of SP, and the first few residues of NC) could form amphipathic helical bundles (78,79).

Interhexmeric Interactions

A large surface area that includes both NTD-NTD and CTD-CTD interactions links hexamers together. Residues involved in interhexameric interactions are located predominately on the NTD helices 4-7 and to a smaller extent the CTD helix 9. In the best studied protein, HIV-

1, there are only two residues in α-9 that are involved in this interaction (186,187). Although both domains participate in interhexameric interactions, there does not appear to be any interaction between the NTD and CTD in the immature particle. There appears to be major differences between the M-PMV and HIV-1 NTDs in the immature particle suggesting greater flexibility in this region (186,187). Therefore, although the RSV NTD fits within the M-PMV immature model, it is currently impossible to predict whether RSV will resemble the HIV-1 or M-PMV immature model, or if it will be an entirely different structure all to its own. Currently unpublished data from our lab indicates that the residues within the flexible loop region (the end of α-4 to the beginning of α-7) are important for maturation. These residues likely contribute to movements between the immature and mature structure or influence how cellular factors bind to the capsids.

Mature Contacts

Analysis of the intermolecular interactions between CA proteins has largely been deduced from pseudoatomic models produced by fitting x-ray crystallography or NMR structures of the NTD and CTD into cryoEM density of in vitro assembled particles. Mutagenesis and suppressor searches have refined the structural models further. These data have provided a solid understanding of the interactions between CA molecules in the mature capsid. Descriptions of these interfaces are provided below (Figure 1.8).

NTD-NTD

In contrast to immature Gag assembly, hexamers and pentamers of the mature capsid are held together through interactions between CA NTDs (44). Assembled mature hexamers have been isolated for RSV, MLV, and HIV, whereas mature CA pentamers have been identified for

RSV and HIV (44,192,194,231,232). Packing of hexamers is less compact in the mature state, the diameter of mature hexamers being about 90 Å rather than the 80 Å of immature hexamers

(25,232,233).

Figure 1.8. The interactions of CA in the mature particle. A pentameric unit of CA (white) is shown with a monomer of CA from a neighboring pentamer (gray). The mature capsid of retroviruses is held together through three distinct contacts within the CA protein. A) NTD-NTD interactions are mediated through alpha-helices 1-3 (blue) and form intrapentameric and intrahexameric interfaces. These interactions are strengthened with NTD-CTD interactions (not shown). B) CTD-CTD interactions are provided from the 310 helix and α9 (red) and are the only interpentameric and interhexameric interaction in the mature capsid. The figure was generated from the pseudoatomic model from the T=1 icosahedral particles of in vitro assembled CA presented in Cardone et.al. 2009.

The NTD-NTD interactions involve helices 1-3 of one molecule interfacing with the first three helices of the neighboring NTD, ultimately creating a bundle of 18 helices around the center of a hexamer and 15 helices for a pentamer (37,62,91,95,234-237). Hydrogen-deuterium exchange provids further support for the involvement of the first three CA helices in hexamer formation (188,238,239). Additionally, mutation of residues in these helices inhibits propagation of virus. The interface is hydrophobic with aliphatic residues in the center surrounded by polar residues at the periphery. There are no electrostatic interactions and intersubunit hydrogen bonding is mediated by water molecules at the interface.

Though it has not been demonstrated to be a strict participant in any NTD-NTD interaction, the close proximity of the N-terminal β-hairpin to the helices in question enables the structure to influence the interface indirectly (221). Experimental evidence suggests that the β- hairpin does not form until the CA protein is completely cleaved, and that formation of the β- hairpin locks in conformational states leading to mature assembly (220,240). Mutations within the

β-hairpin disrupt infectivity and the effect can be reversed by secondary mutations elsewhere in the NTD (221).

NTD-CTD

The concept of an interdomain interface was predicted initially from genetic screens in which secondary mutations in one domain could rescue lethal mutations in the other (62). Initial mapping demonstrated that the interface included two molecules of CA rather than an intramolecule interaction (44,188,239). Higher structural modeling based on cryoEM reconstructions of in vitro assembled particles confirmed the presence of this intermolecular

NTD-CTD interaction (194). Structural analysis further showed that the NTD-CTD interaction adds to the intrahexameric interaction of the NTD. The interdomain interface is made of two distinct interactions between the NTD and CTD of one monomer and the NTD of the neighboring molecule. The interaction between the N-terminal end of α4 helix in the NTD and α11 in the neighboring CTD primarily holds the interface together. Contacts between the α8/9 loop of the

CTD and α3/4 of its own NTD provide further stabilization of the interface (188,239). Formation of this interface is unique to the mature capsid as the NTD and CTD are not properly stacked to make these interactions in the immature particle.

CTD-CTD

The CTD forms homodimeric CTD-CTD interactions, linking two CA molecules. This dimerzation is at least one of the critical interactions that nucleates mature capsid assembly (241).

The HIV-1 CTD protein readily dimerizes in solution at neutral pH although the Kd of the interaction is high, whereas the RSV CTD requires low pH for dimerization (38). In either case, a variety of dimer structures have been isolated and solved to high resolution by x-ray crystallography and NMR (38,50,58,242,243). Most of the structures show parallel or near parallel packing between the second α-helix (α9) of each molecule, although some structures exhibit anti-parallel packing, whose biological function is not readily apparent. Additionally, the dimers of different retrovirus species have slightly different conformations – the HIV-1 α9 helices are parallel whereas the angle between the RSV dimer helices is approximately 45°. Given the many dimer structures that form under similar conditions, it is likely that the packing of CTDs between CA molecules provides a high degree of flexibility to the capsids.

Mutagenesis and hydrogen-deuterium exchange experiments identified α9 as the dimerization helix. Structural models agreed with this assignment identifying the first CTD-CTD interaction between α9 and the 310 helix of the CTD (the dimer interface) (38). Until cryoEM of in vitro assembled CA particles was completed, it was not known as to whether this interface represented interhexameric or intrahexameric interactions. CryoEM of both RSV full-length CA assembled in the presence of acid buffer or 500 mM sodium phosphate, as well as sodium chloride-induced assembly of HIV-1 particles, revealed that the dimer forms at the interpentameric and presumably interhexameric interface (38,194). The interactions in the RSV

CTD dimer crystallized at low pH were shown to agree with the cryoEM models, indicating that the low pH dimer likely represents mature interhexameric interfaces (232).

The dimer is held together by a hydrophobic interface formed by the 310 and α9 helices.

In HIV-1, the key contact occurs at W184 and M185 where the two dimerzation helices cross each other (58). The dimerization interaction is further strengthened by a network of electrostatic interactions involving aspartate and arginine side chains. For RSV, this network occurs between the aspartic residues D179 and D191 with R194 (38). Changing from an immature to a mature dimer involves significant rotation in orientation between the two CTD molecules

(186,187). This relationship creates a more compact dimer in the mature form and alters the binding partners for the residues involved in each interface.

More recently, a new CTD-CTD interaction seen only in the mature capsid has been described. Known as the trimer interface, the interaction occurs at the three-fold axis of in vitro assembled tubes of HIV-1 CA and involves CTDs from three hexamers (65,244). Residues within

α10 and α11 mediate the assembly of this new interface. Mutagenesis and cross-linking of residues in this region prevents infectivity, indicating that this interface plays a biological role in the virus and is not just an artifact of in vitro assembly conditions (244). The trimer interface is not seen in planar lattices of CA assembles. Therefore, in addition to the dimer interface and interdomain linker, this trimer interface likely provides a great deal of flexibility in the interactions that can occur between hexamers and pentamers, and the conformation is dependent on the rotation needed to generate specific curvatures in mature capsids (65,191).

Hexamers vs. Pentamers

Until recently, retroviral pentamers were hypothetical as there was no evidence for their existence. In vitro assembled RSV CA particles from our laboratory were analyzed by cryoEM in the laboratory of Alasdair Steven, and it was reported that the CA protein assembles into two distinct classes of spheres – T = 1, made solely of 12 pentamers and T = 3, composed of 12 pentamers and 20 hexamers. From these data, the first pseudoatomic model of a retroviral pentamer was created (194,232). Though these findings did not prove that pentamers exist in

38 vivo, the structural model fits within the context of the fullerene model of assembly, and provided the first evidence for pentamers. More recently, pentamers of HIV-1 CA protein were isolated in vitro and shown to have a structure very similar to those from RSV (245).

The overall organization of hexamers and pentamers is nearly identical; all three intermolecular interactions (NTD-NTD, NTD-CTD, and CTD-CTD) are found (194,231,245).

The primary difference between the two oligomers is a 12 degree rotation in the molecules relative to each other, altering the packing angles between each subunit resulting in a more compact pentamer. The center of this rotation is located in the middle of the three helices mediating NTD contacts, allowing the helices to maintain overall structure (37,231,245). Whether hexamers or pentamers form seems to depend, at least in part, on electrostatic interactions. For

HIV-1, R18 in α1 sits in the interface and electrostatic repulsion would be greater for the pentamer, as the arginine residues are closer there than in the hexamer. Indeed, mutations that remove this positive charge promote pentameric assembly of purified CA protein (231,245).

Ganser-Pornillos et al. have proposed that these electrostatic interactions promote hexamer assembly until placement of a hexamer is energetically impossible, in which case a pentamer is inserted (42).

INFLUENCES ON CAPSID MATURATION

Both the NTD and CTD of CA are involved in assembly of immature particles and mature capsids, although only the CTD is absolutely required for VLP formation (180,246-249).

The interhexameric and intrahexameric interfaces are distinct, and both domains switch roles after maturation (186,187). The NTD only forms intrahexameric interactions in the mature capsid, whereas the CTD in turn predominately acts at interhexameric interfaces. A critical need in the field of retrovirology is to identify the residues of CA that participate in immature or mature virus contacts. Formation of the interfaces is controlled by so-called maturation switches, structural units within CA that determine the conformations leading to either immature or mature particle assembly.

The β-hairpin

One such switch is the β-hairpin at the N-terminus of CA. As a constituent of Gag, the β- hairpin does not form because the N-terminal Pro is attached to p10 in RSV and MA in HIV-1.

After cleavage, the β-hairpin forms by electrostatic linkage between Pro1 and an aspartic acid residue at the top of α3. This interaction is thought to lock in contacts, especially in the NTD

(69,78,220,221). Formation of the β-hairpin is absolutely required to make mature capsids.

Ablation of the Pro-Asp interaction either by extending the N-terminus or by an Asp-Ala substitution has been shown to block virus infectivity and prevents proper assembly of the mature capsid in a variety of retroviruses (37,47,48,250). Hydrogen-deuterium exchange experiments revealed that formation of the β-hairpin in HIV-1 is also linked with the final cleavage of CA-

SP1, suggesting a need for long-range communication between the NTD and CTD (220,251).

The Spacer Peptide

Another proposed maturation switch in the CTD is SP (36,74,75,252-255). SP and the C- terminus of the CTD form a 6-helix bundle that stabilizes Gag hexamers (36,78,79) (Figure 1.9).

Retroviral capsid proteins are initially cleaved from Gag with a short extension at the C- terminus

(256-259). This intermediate species persists in maturing virions for up to six hours post-budding

(203,260-262). Further proteolytic processing by the RSV protease produces one of two proteins

– CA (237 amino acids) or CA-S (240 amino acids) – that compose the mature capsid; similar timed proteolytic processing occurs in other retroviruses (199). Deletion of SP renders the virus non-infectious due to a defect in Gag-Gag interactions. Failure of the final cleavage event also prevents infectivity, possibly by preventing the assembly of the mature capsid (35). So critical is this final maturation step that it is a novel target for antiretroviral drug therapy with several drugs, including Bevirimat, reaching clinical trials (263,264). However, like most HIV drugs, resistant mutants can be generated quickly and reduce the efficacy of these drugs (178,265).

A role for SP in coordinating the final stages of maturation is predicted from genetic and initial biophysical characterization of the CA-SP intermediate (62,266). Structurally, the SP

Figure 1.9. Models for the roles of Gag subunits in the immature and mature lattices. Computer generated models of the hexameric lattice of Gag and CA are depicted from cryo-electron tomograms. The NTD is shown in blue, CTD in yellow, and spacer peptide (SP) in purple. The

CTD and SP form intrahexameric interactions in the Gag immature lattice, whereas the NTD mediates interhexameric interactions. Maturation removes SP1 and causes formation of the β- hairpin at the N-terminus of the NTD, which induces hexameric ring formation in that domain.

The CTD forms interhexameric interactions that link two hexamers of CA together. Figure adapted from Wright et.al. 2007.

region along with the last 9 amino acids of the CTD is too flexible to be resolved by high resolution x-ray crystallography or NMR (57,267-269). Recent solid-state NMR analysis of the

HIV-1 CA-SP1 molecule provided evidence that the SP region may have a direct influence on the structure of the rest of CA, in particular the CTD (252). Genetic screens for spontaneous secondary site suppressors of lethal CA mutations revealed mutations in SP and SP1 that correct assembly defects in RSV and HIV-1 (62,270). Substitution of serine with leucine at position 241 in RSV suppresses the assembly defect of the MHR mutant F167Y despite the SP not being part of the mature capsid (62). Additionally, when HIV-1 was tested with the maturation inhibitor PF-

46396, which binds to SP1, escape mutants had substitutions in the MHR (270). These data suggested that SP at some point comes into contact with the MHR.

Despite these findings, the precise role SP plays in the formation of mature capsids is unclear. Recently, our laboratory reported that bacterially expressed and purified RSV CA-SP protein assembles into the same structures in vitro as the RSV CA protein (266). These data support the idea that SP promotes the assembly of mature capsids. However, cryoET analysis of

HIV-1 particles in the presence of maturation inhibitors, suggested that failure to remove SP1 prevents dissociation of the immature Gag lattice ultimately blocking the formation of mature capsids (75). Further structural and biochemical investigation is needed to elucidate the precise interactions made by SP and its function during maturation.

The CTD 310 and α9 Helices

The first half of the CTD provides numerous contacts in both immature and mature particles. The 310 and α9 helices establish CTD-CTD interactions that are unique to the mature and immature capsids (186,187). A significant degree of rotation must occur between the two

CTD molecules. The crossing angle of the two α-9 helices is predicted to change by about 140 degrees. This rotation alters the binding partners for the residues at the dimer interface. A variety of studies have investigated the residues, including the MHR and dimerzation helix, that

42 influence the formation of CTD dimers and their contribution to infectivity (37,59-63,250,271).

However, these studies were performed prior to the generation of the model of dimerization in immature particles. Therefore none have included rationally designed mutagenesis to test the accuracy of the model. Nor has there been a test of the role of these hydrophobic residues in both immature and mature capsid assembly.

The Major Homology Region

Within the CA protein, the MHR has the highest level of sequence homology amongst retrovirus species. The MHR consists of approximately 20 residues near the N-terminus of the

CTD and includes the first turn of the CTD and α-8, sitting between the two regions that make up the CTD dimerization interface (32,40,272). The MHR is buried within the CTD and helps stabilize the overall fold of the CTD with hydrophobic and aromatic residues. Structural and genetic data has implicated the MHR in assembly of both immature and mature capsids (59-

61,242,271,273-276). Three types of effects have been described for mutations in the MHR of

RSV – those that have no effect on the virus, those that block budding and are thought to effect immature assembly, and those that block mature capsid assembly by perturbing maturation

(59,63,277).

Substitutions for the three most conserved polar residues of the MHR (RSV: Gln 158,

Glu 162, and Arg 170) as well as for the hydrophobic residues in the second half of the MHR, cause severe budding defects, suggestive of a disruption in proper Gag trafficking or immature particle assembly (59,63). More conservative substitutions at some of the same sites allow normal particle release and assembly of immature virions but still exhibited reduced infectivity. The severe, though not complete, defect in infectivity is despite normal incorporation of viral proteins and RNA and normal function of the RT protein exogenously (59,63). Therefore it was postulated that these mutant virions were defective in maturation.

More recent biochemical analysis of some of these same MHR mutations revealed that they are indeed defective in normal maturation. Structurally, CTD proteins with mutations

D155Y, F167Y, or L171V fold similarly to the WT protein; however, full-length proteins with these same mutations do not assemble in vitro into capsid-like particles similarly to WT (250). To determine interaction partners of the MHR, a suppressor search was initiated and several second- site mutations that corrected the infectivity defect of the MHR mutants spontaneously arose

(62,271). Purified mutant proteins assemble in vitro into the same small spheres made by the WT protein. Many of the second-site suppressors are in the dimerization helix α-9 (62,271).

Isothermal titration calorimetry analysis to assess dimer formation revealed that the lethal MHR mutations prevent dimerization of the CTD, whereas the second-site suppressors restore the ability of the CTD to dimerize (60). Unpublished NMR data analysis of the MHR mutant CTD proteins identified several residues whose location was altered by the mutations. The altered residues included those that probably form interhexameric contacts between the 310 and α-9 helices. Taken together, these data point to a role for the MHR especially the conserved hydrophobic residues, in directing assembly of the mature virus capsid. Though not explicitly examined by the work in this dissertation, the function of the MHR is closely tied to that of the

310 and α-9 helices, and mutations in the MHR have profound effects on the function of the two helices.

REMAINING QUESTIONS AND AIMS OF THE DISSERTATION

Retroviruses undergo an obligatory step of maturation in order to generate infectious virions (17-19). Virus-encoded protease cleaves the protein products of Gag and Gag-Pol allowing an extensive reorganization of the interior of the virus, including a condensation of the reverse transcription complex and creation of a capsid distinct from that of immature particles

(16,20). There are still many remaining questions about the process. Current models suggest that maturation requires a switch between different modes of protein packing; however, how the switch between immature and mature interfaces occurs is poorly understood. Specifically what factors influence the outcome of maturation? Recent models for the immature Gag lattice have been generated, but there is little biological data that currently supports the models (186,187).

Furthermore, a new mature CTD-CTD interface has been identified by cryoEM and NMR studies but there is currently no biological support for the existence of this new interface (244,278).

Finally, although genetic data support the idea that SP is critical for maturation, analysis of its role remains to be accomplished (62,279). The experiments presented in the following data chapters will address these questions related to the influences on maturation by specific regions of the CTD and SP.

Despite high resolution structures for the mature dimer and good models for the immature dimer, few studies investigated the function of the residues participating in the dimer interfaces, including the hydrophobic residues that form abundant interactions. This is especially true for RSV. The experiments presented in Chapter III test roles of individual amino acids in assembly of immature particles and in the maturation of the capsid and provide much needed genetic and biochemical evidence in support of current structural models. Substitutions of hydrophobic residues at the mature dimer interface exerted effects on both mature particle assembly as well as immature particle formation. These data are the first biological support for the new immature dimer interface models.

Initial cleavage of Gag and Gag-Pol yield a short-lived yet important intermediate form of the capsid protein (CA-SP) which contains the 9-12 residue spacer peptide at the C-terminus

(260). Further cleavage yields two mature capsid protein species in RSV, either the 237 amino acid CA protein or the 240 amino acid CA-S protein (199). Genetic evidence suggests that mutations in the spacer peptide promote maturation, whereas structural studies show that failure to remove the spacer peptide prevents mature capsid assembly. Chapter IV describes a comparison of the in vitro assembly phenotypes of CA and CA-SP and provides a potential role for the spacer peptide during maturation. Purified CA-SP assembled with faster kinetics than CA but there was no significant difference in the types of assembled particles formed. Transient interactions of SP were identified in regions of the CTD critical for determining the outcome of maturation consistent with an active role for the SP in promoting maturation. Chapter IV is a

45 combination of the biochemical data presented in England and Purdy et.al. 2014 and structural data presented in Keller et.al. 2013 on the RSV spacer peptide.

Chapter V provides preliminary data on the function of a third site in the CTD that recently has been shown to form during maturation. The three-fold interface that forms an interdomain interface in the mature capsid was recently identified by NMR analysis of in vitro assembled HIV-1 particles (244,278). Provided in Chapter V is preliminary data from limited mutagenesis at a single site in that region. RSV Thr214 was found to be a critical residue for infectivity of the virus. In particular, the residue was important for the formation of the mature capsid, consistent with structural models of the three-fold interface.

Chapter VI summarizes the findings and presents a current model for how both the spacer peptide and CTD interfaces influence the outcome of maturation in relation to other factors that contribute to maturation, such as the MHR. Included are implications of the data for the pathway of mature capsid assembly and future directions for the study of retrovirus maturation. An appendix follows describing some preliminary data from experiments examinging the effect of some NTD mutations on the in vitro assembly of immature and mature particles as well as initial investigation of the dynamic interactions of SP with the CTD.

CHAPTER II

MATERIALS AND METHODS

CELL CULTURE AND VIRUS SPREAD

The RSV genome, propagated in Escherichia coli (E. coli) strain DH5α, was maintained on the splicing competent plasmid pRS.V8-eGFP (RS.V8) that contains the gene for GFP in place of the src oncogene (280). Mutations were generated in a pCMV-GagPol plasmid by QuikChange site-directed mutagenesis and inserted into RS.V8 via restriction digest and ligation using the SbfI and SnaBI cut sites. Plasmids were checked by sequencing (Operon).

Transformed Japanese quail fibroblast cells (QT6) were transfected with RS.V8 to produce stock virus. Virus producing cells were normalized by flow cytometry for GFP. Chicken fibroblast (DF1) cells were infected and spread of the virus through the culture was monitored by flow cytometry. The data were plotted as the ln(GFP-positive cells) vs. days post-infection, and the slope of these lines were the rate of spread (271). Infectivity was reported as the average of at least 3 independent growth curves. Due to the limit of detection of the flow cytometer, non- infectious mutant viruses were reported as < 0.05 (271,281).

PARTICLE RELEASE ASSAY

Duplicate 60 mm plates of QT6 cells were transfected with 10-20 µg of RS.V8 plasmid.

Twenty-four hr post-transfection, cells were starved for 30 min and then radio-labeled with [35S]- methionine/cysteine (100 µCi). The lysate was harvested from one plate after 15 min of labeling and the released virus was collected from the medium of the other after 4 hr of labeling (63,221).

FCRC-12, an α-CA antibody, and Staph A were used to immunoprecipitate Gag and CA from the lysate and medium. Proteins were separated by 15 % SDS-PAGE and the gel was dried prior to exposure to film or phosphor screen. For each virus, the release of CA in the medium was corrected for the level of Gag protein detected from the parallel plate. Particle release was determined by calculating the ratio of CA in the medium to Gag in the lysate and reported as the average of 3 repeats (63).

DETERGENT RESISTANCE ASSAY

For each virus construct, two plates of QT6 cells were transfected with 10 µg of plasmid

DNA. Twenty-four hr post-transfection, cells were starved for 30 min and then radio-labeled with

[35S]-methionine/cysteine (100 µCi) for 4 hr. Medium was collected and spun at 1,000 x g to pellet cells. The virus-containing supernatant was overlaid on a step gradient of 15 % sucrose (+/-

1.0 % Triton X-100) and 20 % sucrose and spun at 153,000 x g for 50 min at 4 °C (176,180).

Lysis buffer containing SDS was added to the supernatant and pellet. CA protein was immunoprecipitated with α-CA antibody and separated by 15 % SDS-PAGE. The gel was dried for autoradiography and imaged by film and phosphor screen. Stability of the capsid was analyzed by calculating the percentage of protein capable of being pelleted and comparing detergent resistance of mutant and wild type virus (176,180).

PROTEIN PURIFICATION

Full-Length Capsid Protein

RSV Prague C CA (237 residues), CA-S (240 residues), and CA-SP (249 residues) were inserted into pET-24(+) and expressed in the E. coli strain BL21(DE3) or the codon optimized strain BL21-Gold(DE3)-RIL (241,250,282,283). CA and CA-SP mutations were generated by site-directed mutagenesis (Stratagene). Expression of full-length protein was induced by autoinduction (ZYP-5052) in 2 L Fernbach flasks at either 37 °C or 32 °C. After culture saturation, the E. coli were spun down at 4,000 x g for 10 minutes at 4 °C (241,250,266,282).

Bacteria pellets were resuspended in resuspension buffer (20 mM Tris, 500 mM NaCl, 10 % glycerol, 1 mM EDTA, and 10 mM DTT, pH 7.6 with complete inhibitor and lysozyme) and sonicated. Following sonication, the sample was treated with Benzonase nuclease (Novagen) and centrifuged at 21,000 x g for 30 min at 4 °C. Soluble protein in the supernatant was precipitated with 35 % ammonium sulfate for approximately 30 min and again centrifuged at 21,000 x g for

30 min at 4 °C. The pellet was resuspended in buffer A (50 mM Tris, 1.0 mM DTT, and 0.1 mM

EDTA, pH 7.6) and dialyzed against the same buffer overnight at 4 °C. The soluble material was

49 loaded onto a column packed with 25 ml of diethylaminoethyl (DEAE) beads (EMD

Biosciences), eluted with buffer A, and collected in 5 ml fractions. Fractions containing protein were pooled and precipitated with 50 % ammonium sulfate for 30 min followed by centrifugation at 21,000 x g for 10 min. The pellet was resuspended in and dialyzed overnight at 4 °C against the same buffer. Monomeric protein was obtained by size exclusion chromatography (SEC) using

Superdex 75 resin (GE Healthcare). Fractions containing the monomeric protein were pooled, concentrated to approximately 10 mg/ml with Pierce protein concentrators, and stored in SEC buffer (20 mM Tris, 150 mM NaCl, and 0.1 mM EDTA, pH 7.6) (241,250,266).

C-Terminal Domain Protein

The CA C terminal domain (CTD) and the CTD bearing the SP were inserted into the pET18(+) expression vector and expressed as 6xHis-tagged proteins in BL21(DE3) (60). Auto- induced, saturated bacteria cultures were pelleted and resuspended in resuspension buffer. After sonication, the bacteria lysate was spun at 21,000 x g for 30 min at 4 °C and the supernatant was loaded onto a column containing 10 ml of nickel resin (GE Healthcare). Bound protein was eluted with 1 M imidizole and the 6x His-tag was removed by tobacco etch virus protease cleavage leaving 5 exogenous amino acids at the N-terminus (GHMAS). Monomeric CTD protein was isolated by SEC and stored in SEC buffer. The following extinction coefficient values were used:

24,980 M-1cm-1 (RSV CA, CA-S, and CA-SP) and 6,990 M-1cm-1 (RSV CTD and CTD-SP).

To produce 15N-labeled 6xHis-tagged CTD and CTD-SP protein, BL21(DE3) were inoculated into 10 ml of 2x YT broth and grown to an OD of 2. The starter culture was used to

15 inoculate 500 ml of M9 medium supplemented with NH4Cl in a 2 L Fernbach flask and grown to and OD of 1, at which time IPTG (final concentration1 mM) was added to induce protein production. Bacteria were lysed and proteins purified using the same nickel column purification mentioned above.

Gag (ΔMBDΔPR) Protein

The largest well-behaved, soluble form of the Gag protein lacks the first 84 residues of

MA and all of PR (ΔMBDΔPR). The pET3xc plasmid containing RSV ΔMBDΔPR (a kind gift from Volker M. Vogt) was expressed by auto-induction in BL21(DE3)-RIL E. coli as previously mentioned (39,79,284). The bacteria culture was spun down and resuspended in buffer consisting of 20 mM HEPES pH 7.5, 750 mM NaCl, 1 mM EDTA, 0.5 mM TCEP, 10 % glycerol, 0.1 %

NP-40, and lysozyme. The suspension was sonicated and nucleic acids were precipitated by addition of Polymin P (pH 7.8) to a final concentration of 0.15 % and then spun at 21,000 x g for

30 min at 4 °C. The supernatant was further precipitated with 25 % ammonium sulfate and resuspended in 20 mM HEPES pH 7.5, 100 mM NaCl, 0.1 mM EDTA, and 0.1 mM TCEP. The suspension was briefly centrifuged to remove any insoluble material and then loaded onto a cation exchange column (SP resin, GE Healthcare), equilibrated with Buffer A (20 mM HEPES pH 8.0, 100 mM NaCl, 0.1 mM EDTA, and 0.1 mM TCEP). ΔMBDΔPR protein was eluted by linear gradient from 0-75 % Buffer B (20 mM HEPES pH 8.0, 1 M NaCl, 0.1 mM EDTA, and 0.1 mM TCEP) and fractions containing protein were pooled and concentrated to approximately 2 ml.

The protein was further purified by SEC and monomeric fractions were pooled, concentrated to approximately 5 mg/ml, and stored in buffer (25 mM HEPES pH 8.0, 500 mM NaCl, 0.1 mM

EDTA, 0.1 mM TCEP, and 0.01 mM ZnSO4) at -80 °C (39,79,284).

IN VITRO CA ASSEMBLY

CA assembly was initiated in 50 µl reactions (80 µM protein, 50 mM Tris, 150 mM

NaCl, 0.1 mM EDTA) by addition of sodium phosphate (final concentration 500 mM) at pH 8.0 and 25°C in transparent 96-well half-area plates unless otherwise specified. The development of turbidity was monitored at 450 nm in at least three independent experiments using a SpectraMax

Plus384 (Molecular Devices) (241,250). To reduce settling of assembled or aggregated material, the plate was shaken between readings. The kinetic curves presented were averages of replicate

51 curves (usually 3) generated on the same day, and for clarity, only every 8-10 time points were shown.

The critical concentration for nucleating in vitro assembly, i.e. the minimum amount of protein needed to initiate assembly, was determined by assembling CA, CA-S, and CA-SP under the same standard conditions. After reaching plateau, the assembly reaction was spun at 128,000 x g in an Optima THX-1000 for 30 minutes at 25 °C. The unassembled, soluble protein in the supernatant was measured by Nanodrop-3000, and reported as the average of 4 replicates

(241,266).

IN VITRO GAG ASSEMBLY

The ΔMBDΔPR protein assembly was done in 30 µl reactions in 0.7 ml eppendorff tubes.

The protein was diluted to 1 mg/ml with 50 mM 2-(N-morpholino)ethanesulfonic acid and 100 mM NaCl, pH 6.0. Assembly was triggered by the addition of a 50-mer DNA consisting of 25 GT repeats (GT50) at a ratio of 10:1 ΔMBDΔPR:GT50. The reaction was allowed to proceed for 30 min at room temperature with occasional mixing (39,284).

ELECTRON MICROSCOPY

Negative Stain

Samples from in vitro CA or Gag assembly reactions were applied to parafilm in 10 µl droplets. Carbon and formvar-coated 200 mesh copper grids were inverted and set on the droplets for 2 min. After a series of 3 sequential water washes, the grids were stained with 2 % uranyl acetate for 5 sec and viewed on a JEOL JEM 1400 electron microscope. An Orius SC 1000 bottom-mounted CCD camera was used to capture images (241,250).

Thin Section

QT6 cells were transfected with RS.V8 plasmids and 24 hr post-transfection were fixed for 1 hr in a solution of 0.5 % glutaraldehyde and 4 % paraformaldehyde buffered with 0.1 M sodium cacodylate, pH 7.3. Following fixation, cells were washed in 0.1 M sodium cacodylate buffer and post-fixed overnight in buffered 1 % osmium and 1.5 % potassium ferrocyanide. After

52 post-fixation, cells were rinsed with buffer, dehydrated in a graded series of ethanol washes, and embedded in EMBed 812 (Electron Microscopy Sciences). A diamond knife mounted in a Porter-

Blum MT-2B ultramicrotome (Sorvall) was used to cut 90 nm thin sections (59,180). The sections were mounted on 200 mesh copper grids and stained with 2 % uranyl acetate and lead citrate. The sections were examined on a JEOL JEM 1400 electron microscope.

CIRCULAR DICHROISM AND UNFOLDING EQUILIBRIUM

Circular dichroism (CD) spectra for monomeric CA and CA-SP (0.15 mg/ml, 20 mM

Tris, 75 mM NaCl, 0.1 mM EDTA) were collected in the presence of 0 to 6 M guanidine hydrochloride (prepared using a Hamilton Microlab titrator) at 25°C using a Jasco J-710 spectropolarimeter. Data was collected between 190 and 260 nm, whereas secondary structure was monitored between 212 and 225 nm. The effects of SP on secondary structure and the midpoint of denaturant concentration were calculated using the programs CDPro and K2D3 (60).

ISOTHERMAL TITRATION CALORIMETRY

The dimerization affinity of CTD and CTD-SP proteins was measured by a VP-ITC microcalorimeter at 10 °C. Protein was dialyzed against 20 mM succinic acid, 100 mM NaCl, and

5 mM NaN3, pH 3.7 to induce dimerization and then concentrated to 0.5 M. Low pH triggers dimerization of the isolated CTD by neutralizing aspartate residues at the dimer interface and also promotes the assembly of full-length CA into capsid-like structures identical to those formed in sodium phosphate at pH 8.0 (38,232). The protein in low pH buffer was introduced into a syringe and injections of 5-10 µl were performed. Origin (MicroCal) was used to integrate the dissociation heat generated per injection, and the data was fit to a monomer-dimer equilibrium model (60). At least 2 replicates were performed for each protein.

NUCLEAR MAGNETIC RESONANCE

CTD-SP proteins bearing a C192R mutation in combination with cysteine mutations at residue 241 or 245 or a cysteine insertion at the C-terminus were incubated with 10-fold molar excess S-(1-oxyl-2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-yl)methyl methanesulfonothioate

(MTSL) at room temperature overnight. MTSL is a paramagnetic spin label that forms a disulfide linkage with cysteine residues. The presence of MTLS increases the relaxation rate of nearby nuclei and allows detection of interactions spanning up to 25 Å as opposed to 6 Å seen in traditional NMR analysis. Excess MTSL label was removed by buffer exchange spin column (GE

Healthcare) and the resulting protein was dialyzed overnight into NMR buffer (20 mM Tris, 50 mM sodium phosphate, 5 mM sodium azide, pH 7.0).

Prior to obtaining NMR spectra, deuterated H2O was added to a final percentage of 10 %.

Experiments were recorded at 25 °C on a 600 MHz Bruker Advance II spectrometer. Data were processed using the TopSpin software and assignments were made using NMRviewJ. Backbone assignments were derived from two-dimensional heteronuclear single quantum coherence

(HSQC) experiments. The intensity of peaks was compared before and after reduction of the

MTSL spin label with ascorbic acid.

WESTERN BLOT ANALYSIS

QT6 cells in 100 mm plates were transfected with 15 µg of WT, A154D, T214A, or

T214D RS.V8 plasmid. Forty-eight hr post-transfection, medium was collected and spun at 1,000 x g to pellet cells. The virus-containing supernatant was spun at 153,000 x g for 50 min at 4 °C.

Pellets were resuspended in sample buffer and separated by 15 % SDS-PAGE. Proteins were transferred to PVDF membrane and blocked. Primary antibody (α-trpRS) was added at a 1:500 ratio overnight at 4 °C. After washing with TBST, α-rabbit secondary antibody was added at a

1:10,000 dilution for 90 min. Blots were washed and incubated with ECL Plus (GE Healthcare) for 5 minutes. Bands were exposed using the Bio-Rad ChemiDoc MP system.

CRYO-ELECTRON MICROSCOPY AND SINGLE PARTICLE RECONSTRUCTION

Assembled CA-SP particles were imaged by TEM. Two distinct sizes of particles were found – 17 nm and 32 nm. Of the 17 nm particles, 200 were picked using EMAN2 and reconstructed with Robem. The same process was performed on 60 particles of the larger class.

WT, K107A, and K107R ΔMBDΔPR proteins were assembled in vitro as described above. The particles were absorbed onto holey carbon grids, flash frozen, and examined by cryoEM on a JEOL microscope. Micrographs were analyzed with EMAN and particles were picked for their proper size (~70 nm) and uniformity. The radial density from 12 of the most uniform of these particles was averaged for each mutant to produce a radial density map.

CHAPTER III

THE 310 AND α-9 HELICES ARE A CRITICAL MATURATION SWITCH

The following chapter is adapted from the following manuscript in preparation:

England MR, Heyrana KJ, Ropson IJ, Flanagan JM and Craven RC. Contributions of the 310 and α-9 helices to intermolecular C-terminal domain interfaces in Rous sarcoma virus.

ABSTRACT

Retrovirus maturation leads to major structural reorganization of the interior of the virus producing pleomorphic mature capsids. Immature particles are made of a hexameric lattice of the

Gag polyprotein, whereas mature capsids are formed by one of the Gag cleavage products, CA.

Recent pseudo-atomic models and crystal structures described dimerization interfaces between the CA C-terminal domain (CTD) in both the immature and mature particle. The interface occurs between two neighboring hexamers. The dimerization interfaces of the immature and mature particles are unique due to different packing of CTD molecules. This leads to the prediction that some CTD residues will be critical for establishing the immature dimer, whereas others will be important for mature dimerization. The goal of this study was to test this prediction biologically and biochemically using Rous sarcoma virus (RSV). We report that mutant viruses bearing

W153A in the 310 helix and L180A and V188A in α9 are not infectious, release particles poorly, and fail to assemble Gag in vitro, indicating that those residues are critical for immature assembly. Viruses bearing mutations A154D and A154V remained Gag assembly competent, but could not assemble CA in vitro suggesting that they participate primarily at the mature interface.

The Q195A virus produced particles less efficiently than WT but could assemble Gag. The mutant also assembled abnormal CA particles in vitro, suggesting that it is functional but has defects in both immature and mature particle assembly. These results provide support for the importance of particular residues in the 310 and α9 helices in assembly of immature particles and maturation and provide the strongest support to date of the relevance of current immature structural models.

IMPORTANCE

Maturation of retroviruses requires dramatic reorganization of the structural interior of the virus. Assembly of the mature capsid has been well described at a near atomic level; however, assembly structures of the immature particle are still elusive. One shared interaction in both the immature and mature particle is the dimer interface, which holds hexameric units of the structural

57 protein together. Recent models for the immature dimer predict that it is distinct from the mature dimer, though the immature model has not yet been tested. We provide here the first genetic and biochemical analysis in support of the relevance of the immature dimer model and have identified mutants that discriminate between immature assembly and mature assembly.

INTRODUCTION

The structural proteins of orthoretroviruses are initially synthesized as the polyprotein

Gag. The main structural domains of the protein are the matrix (MA) which binds to the viral envelop, capsid (CA) which forms the predominate Gag-Gag interactions, and nucleocapsid (NC) which binds to the RNA genome (45,230). To become infectious retroviruses undergo an obligatory maturation process that is initiated concomitant with budding and results in the activation of virus encoded protease (PR), which sequentially cleaves the Gag polyprotein allowing reorganization of the structural proteins (42,45). Failure of complete maturation through chemical or genetic manipulation yields non-infectious viruses (75,176,224,264). The most visually stunning consequence of maturation is the dramatic reorganization of the virion interior as the virus shifts from an immature Gag protein shell to a mature capsid made of the CA protein.

The protein has high structural homology amongst all orthoretrovirus species and consists of two domains – the N-terminal domain (NTD) and C-terminal domain (CTD) that are separated by a short, flexible linker. The NTD contains 7 alpha helices preceded by a β-hairpin at the N- terminus (40,53,56,285). The CTD is made of a short 310 helix followed by 4 alpha helices

(32,40). Both immature and mature retrovirus capsids are built based on fullerene geometric constraints, mediated through CA (69,185,286). The low resolution reconstructions of immature particles revealed a hexameric organization of Gag molecules around the hole of the six-fold axis with the CA NTD sitting directly above the CTD (20,36,179,181,182,287). Cleavage of CA from the Gag polyprotein alters the CA-CA interactions formed by breaking the immature contacts.

During these changes the CTD shifts outward whereas the NTD pinches inward, establishing new

58 intermolecular interactions that allow the CA protein to form the mature capsid with a hexameric lattice punctuated with 12 pentamers (25,44,185,193,194,231,245).

Although both domains contribute to virus infectivity and assembly in vivo, genetic studies provided evidence that the CTD and downstream spacer peptide are the only parts of Gag absolutely required for the production of particles (35,249,288,289). Recent subtomogram averaging of tubes from in vitro assembled truncated M-PMV and HIV-1 Gag protein has provided an 8 Å resolution structural model for the immature particle (186,187). In these models, the CTD is predicted to form two different homo-dimeric interactions. The interhexameric interaction involves α9 whereas the intrahexameric dimer includes α11 and a region of the CTD called the major homology region (MHR) around α8. In HIV-1 W184 and M185 of α9 are known to disrupt immature particle assembly indicating they are a point of close association between the two CTD molecules (37).

Although the fold of the CA CTD does not change much during maturation, the relative positioning of the dimers changes quite dramatically. X-ray crystallography performed on dimers of HIV-1 or RSV CTD expected to represent mature dimers show significant differences between the immature and mature dimers (32,38,242). Upon maturation, HIV-1 residues W184 and M185 remain at the interhexameric dimer interface, but there is a dramatic change in orientation of the molecules relative to each other. The shift in orientation during maturation causes the 310 helix, which is at the periphery of the dimeric interface in the Gag lattice model, to rotate toward the center of the interface. This places the RSV residue A154 near residue Q195 in α9 (Figure 3.1).

The rotation during maturation breaks the intrahexameric interaction of the MHR and α11, thereby making the interhexameric interactions the only dimer interface formed in the mature capsid.

The mature dimer interface is made largely of hydrophobic residues along with several salt bridges (32,38,242). The new structural model of the immature interhexameric dimer also predicts this interface to be held together by hydrophobic interactions; however, this hypothesis

for the mature dimer (Mature). Upper panels depict the top view and lower panels the side view

ET ET density of immature assembled MPMV dimers (Immature). Rendering of the low pH dimer of RSV

Figure Figure 3.1. The immature and mature dimer interfaces. Monomeric RSV CTD protein (PDB: 3g21) was fit into the cryo CTD is shown for each dimer. Residues selected for mutagenesis are colored as follows: W153 (black), (orange). Q195 and (blue), V188 (green), A154 (red), L180

61 has not yet been tested (186,187). Since the introduction of high resolution models for the immature particle, there have been no conformational genetic and biochemical tests of the accuracy of the model. The differences between the two dimers suggest that some CTD residues are critical for establishing the immature dimer, whereas other residues will be important for creating the mature dimer. Based on the orientation shift of the CTD, residues of the 310 and α9 helices are likely critical to regulating the maturation of the CTD dimer. As there are currently no experimentally derived models for the immature Rous sarcoma virus (RSV) Gag lattice, we decided to test the contribution of individual CTD amino acids toward production of immature particles and maturation of the RSV capsid.

Here we provide genetic and biochemical evidence for the importance of several hydrophobic residues predicted by the immature and mature structural models for stabilizing each of the interfaces. To accomplish this, we identified hydrophobic residues on the CTD 310 and α9 helices of RSV likely to contribute to dimerization and carried out a mutagenesis screen to test their effects on infectivity and maturation. We report that substitutions to hydrophobic residues at the mature dimer interface were shown to exert effects on both mature particle assembly as well as immature particle formation. All mutant viruses except A154G provided at least some perturbation to virus spread. The largest effects for mutations at A154 and Q195 occurred at mature assembly and dimerization consistent with the predicted structural rotation of the CTD.

Mutants W153A, L180A, and V188A disrupted Gag association suggesting these residues are critical to the immature dimer consistent with the immature dimer models. In summary, our mutagenesis provides much needed genetic and biochemical evidence in support of current structural models for immature and mature dimers of the CTD.

RESULTS

Analysis of Structural Models

Structural models for the immature CTD dimer interface have been generated for

HIV-1 and M-PMV from cryoEM/cryoET studies, but no such model has been developed yet for RSV. Using Chimera, monomers from the low pH dimer 3G21 for RSV CA were aligned to the HIV-1 and M-PMV immature dimer models and atomic clashes were minimized to generate a structural model for the immature RSV CTD dimer (Figure 3.1).

Analysis of each of the predictive RSV dimers by the Protein, Interfaces, Structures, and

Assemblies (PISA) tool revealed similarities and differences between the two dimers.

Each of the interfaces is created from 20 amino acids of the CTD, many of which are hydrophobic. However, the total surface area of the interface is predicted to be ~595 Å2 for the immature dimer and ~630 Å2 for the mature dimer. Residues that make up the interfaces are different as well. Both the structural models and PISA analysis predict that the A154 of the 310 helix and Q195 at the bottom of α9 do not interact in the immature dimer. Yet after maturation, there is considerable movement of CTD molecules toward each other that places these residues near each other forming a hydrogen bond in 3G21

(Figure 3.1). These data suggest that some of the residues in the CTD will be important for establishing the immature dimer rather than the mature dimer and vice versa.

Combining the results of PISA analysis along with visual inspection of the immature dimer model and 3G21, we chose 5 residues in the 310 and α9 helices for their potential contribution to the interfaces. These residues include W153 and A154 in the 310 helix, and L180, V188, and Q195 along α9. All of the residues were solvent exposed as monomers. Most residues were mutated to alanines with the exception of A154 which

Figure 3.2. Effects of mutations on virus spread. Virus spread through cultures of DF1 cells over the course of 21 days was monitored by FACS of virus-encoded GFP production.

Table 3.1. Phenotypes of WT and mutant viruses and proteins

CA Infectivitya Particle Immature Dimer Kd CA Core Protein Releaseb Assembly (mM) Assembly Stabilityc WT 0.80 ± 0.07 1.85 ± 0.24 Yes 0.11 ± 0.02 Spheres 0.47 ± 0.13 W153A < 0.05 0.26 ± 0.24 No > 1.5 No 0.01 ± 0.02 A154D < 0.05 1.44 ± 0.20 Yes > 1.5 No 0.02 ± 0.02 A154G 0.65± 0.04 1.87 ± 0.20 Yes 0.11 ± 0.02 Spheres 0.43 ± 0.02 A154V < 0.05 0.97 ± 0.55 Yes > 1.5 No 0.02 ± 0.01 L180A < 0.05 0.06 ± 0.09 No > 1.5 No 0.06 ± 0.08 V188A < 0.05 0.14 ± 0.19 No > 1.5 No 0.11 ± 0.11 Q195A 0.45 ± 0.03 0.64 ± 0.17 Yes 0.41 ± 0.03 Tubes 0.15 ± 0.03 a – calculated from the slope of (ln % GFP cells / days post infection) b – CA in medium / Gag in lysate c – fraction of CA in pellet after spinning through triton X-100

65 was mutated to conserved side chains glycine and valine and the negatively charged aspartic acid.

Virus Spread

The relative ability of mutant viruses to replicate in DF1 cells was analyzed by measuring the rate of spread of virus-encoded GFP through cell monolayers (Figure 3.2 and Table 3.1). Of the 310 helix mutants, W153A, A154D, and A154V all failed to replicate. The virus with the most conserved mutation A154G retained the ability to spread at about 80 % of WT. Both L180A atthe turn between α8 and α9 and V188A at the midpoint of α9 prevented the virus from replicating.

The rate of spread of Q195A located at the bottom of α9 was only 56 % of WT. The apparent delay in Q195A infectivity suggested a possible adaption or reversion mutation. The CA region of the viruses produced at days 21 and 28 was sequenced and found to retain only the Q195A mutation (data not shown). These data suggested that W153A, A154D, A154V, L180A, V188A, and Q195A had some type of defect in the virus replication cycle. We sought to address the stage at which each of these viruses was blocked.

Immature Particle Release

We first assessed the ability of each virus to produce and release immature particles. All mutant viruses were capable of synthesizing Gag protein in transfected QT6 cells. The CAmedium to Gaglysate ratio for the W153A (0.26 ± 0.24), V188A (0.14 ± 0.19), and L180A (0.06 ± 0.09) mutant viruses was significantly decreased compared to WT (1.85 ± 0.24), indicating very poor ability to bud particles from cells (Figure 3.3A and 3.3B). Though there was little protein in the medium for these viruses, all of the protein in the medium appeared cleaved to the mature proteins like the WT virus (Figure 3.3A).

To address if these viruses were producing any particles, transfected QT6 cells were fixed, embedded and processed by thin sectioning for examination by transmission electron microscopy (TEM). Central sections through WT virus produce particles that are 100-120 nm in

Figure 3.3. Effects of mutations on particle release. QT6 cells transfected with proviral plasmid were radio-labeled for 15 min (lysate) and 4 h (medium). A) Lysate and medium were separated by SDS-PAGE. 1-Mock, 2-WT, 3-W153A, 4-A154V, 5-A154G, 6-A154D, 7-L180A, 8-V188A,

9-WT, 10-Q195A. B) Particle release was determined by the ratio of CA in the medium to Gag in the lysate.

Figure 3.4. Thin sections of transfected cells. QT6 cells transfected with plasmids containing virus genome were fixed, sectioned and stained with uranyl acetate, and examined by TEM. Bar

= 200 nm.

70 diameter with dark centralized densities representative of the condensed ribonucleoprotein complex and fuzzy halo of glycoprotein on the exterior of the particles (Figure 3.4). The W153A virus failed to produce extracellular particles resembling the WT virus, though there was a noticeable accumulation of protein at the plasma membrane in a few fields. RSV with the V188A mutation produced rare particles that were approximately the WT size and had the characteristic fuzzy halo of glycoproteins. However there was no discernible dark centralized core in these particles. A few fields showed L180A virus particles that resembled WT, although these particles were rare and difficult to find.

RSV with the A154G substitution released particles at levels indistinguishable from the

WT virus (Figure 3.3B). The A154D virus had a distinct decrease in particle release, though the value was not statistically different than that of WT (1.44 ± 0.20 vs. 1.85 ± 0.24). Both the

A154V and Q195A viruses were intermediate for particle release (0.97 ± 0.55 and 0.64 ± 0.20, respectively), indicating a moderate defect in the ability to bud particles (Figure 3.3B). All of the

A154 mutated viruses exhibited abundant particles of WT size and morphology as determined by

TEM (Figure 3.4). The Q195A virus also produced normal looking particles, but they were rare and difficult to find (Figure 3.4).

Virus spread, particle release, and thin section EM data are all consistent with a severe

Gag assembly defect for viruses bearing the W153A, L180A, and V188A mutations. To specifically address the ability of Gag molecules to assemble independent of the virus, we used an in vitro Gag assembly system to assess the effects of mutation on the protein.

In Vitro Immature Particle Assembly

The largest construct of RSV Gag that is well behaved is a protein lacking the membrane binding domain at the N-terminus of MA and the PR domain at the C-terminus of Gag

(ΔMBDΔPR). WT ΔMBDΔPR protein assembled in vitro into large spheres of about 70 nm in diameter with a dark center when incubated with DNA oligonucleotides (Figure 3.5). The virus- like particles (VLPs) were typically double ringed due to the multiple domains of Gag and

Figure 3.5. In vitro immature particle assembly. Purified ΔMBDΔPR protein was induced to assemble in vitro by addition of a 50-mer DNA for 30 min. Particles were absorbed onto formvar and carbon coated copper grids, stained with 1 % uranyl acetate, and visualized by TEM. Bar =

100 nm.

73 occasionally incomplete spheres could be observed consistent with others observations. The

ΔMBDΔPR protein bearing the W153A, L180A, or V188A mutation was unable to assemble into structures like the WT protein consistent with a crippling defect in Gag association (Figure 3.5).

The A154G, A154D, A154V, and Q195A substitutions in ΔMBDΔPR did not prohibit assembly of the protein. All assembled into the same large double ringed spheres as the WT ΔMBDΔPR protein, suggesting Gag-Gag interactions were preserved in these mutant proteins (Figure 3.5).

Viruses with the W153A, L180A, and V188A mutations are severely crippled in Gag assembly consistent with a failure to bud and assemble Gag in vitro. The in vitro Gag assembly indicates that the Gag molecules are unable to associate, but we cannot rule out an additional defect at an earlier step of immature particle assembly, such as Gag trafficking. In contrast the

A154G and A154D viruses are competent for Gag assembly. Viruses with the A154V and Q195A mutations have an intermediate phenotype but are capable of forming Gag-Gag associations.

These data suggest that the defects in spread of viruses with mutations at A154D, A154V, and

Q195A may be due in part to a failure to form mature capsids during maturation.

Capsid Stability

The quality of mature capsids can be assessed in vivo by centrifuging virus-containing medium through non-ionic detergent and measuring the fraction of CA that exists in stable interactions. RSV produces highly heterogeneous mixtures of particles, only a fraction of which contain properly formed and functional capsids. As a result, only about 40 % of the CA protein remains resistant to detergent. Similar to previously published results, 47 % of the CA protein for

WT virus was resistant to non-ionic detergent (Figure 3.6 and Table 3.1) (176,180). A154G substituted virions were nearly equally resistant as WT indicating that the capsids of those viruses were likely fully matured consistent with virus spread (Figure 3.6 and Table 3.1). Viruses with

A154D and A154V mutations yielded particles that were almost entirely susceptible to detergent.

Taken with infectivity, Gag assembly, and thin sections, these two mutant viruses likely have a failure of maturation. Viruses bearing the Q195A substitution were susceptible to detergent,

Figure 3.6. Stability of mutant virus particles. QT6 cells 1 d post-tranfection were radio-labeled for 4 h and the medium was harvested. The fraction of CA from remaining in the pellet after centrifugation in the absence (hatched bars) or presence (white bars) of Triton x-100 was reported. Numbers above the bars represents the fraction of CA resistant to detergent for the Gag assembly-competent viruses.

75 though slightly more resistant than the defective A154 mutants (15 % versus 2 %) (Figure 3.6 and

Table 3.1). Additionally, the Gag assembly defective viruses (W153A, L180A, or V188A) failed to produce resistant capsids (Figure 3.6).

The failure of A154D, A154V, and Q195A viruses to form stable CA interactions is consistent with the viruses having a defect in maturation. To determine if there was a capsid assembly defect we used an in vitro CA assembly system.

In Vitro CA Assembly

We have previously established an efficient in vitro system for monitoring the ability of

CA protein to assemble capsid-like structures. Addition of sodium phosphate to purified CA protein induced nucleation-driven self-association of the protein into spherical structures of 17 nm and 32 nm in diameter visualized by TEM (241). The in vitro system faithfully recapitulated the results of maturation-defective viruses with mutations in the major homology region

(250,271).

All of the dimer interface mutations were tested for their ability to form capsid-like structures by the in vitro CA assembly assay. As expected from the capsid stability assay, the

A154G CA protein assembled with kinetics similar to the WT protein and also produced small spheres visualized by TEM (Figure 3.7A and 3.8). CA protein containing the mutations A154D and A154V failed to assemble organized products although both reactions became turbid after 4-5 hours. These data indicated that the proteins were capable of self-association, but unable to create capsid-like particles, confirming that these substitutions inhibit mature capsid assembly. Q195A

CA developed turbidity with slower kinetics than WT and primarily yielded tubes (Figure 3.7A and 3.8). The in vitro results suggest Q195A is reluctant to assemble and forms altered CA interactions different from that of the WT protein. All of the Gag assembly defective mutants consequently were also unable to assemble CA in vitro (Figure 3.7A and 3.8).

Figure 3.7. In vitro mature particle assembly. A and B) Purified CA protein was assembled in vitro with 500 mM sodium phosphate. The kinetics of assembly was monitored by turbidity at

450 nm.

Figure 3.8. Transmission electron micrographs of assembled CA mutants. Particles from assembly reactions in Figure 3.7 were absorbed onto formvar and carbon coated copper grids, stained with 1 % uranyl acetate, and visualized by TEM. Bar = 100 nm.

PISA analysis of the mature CTD dimer (3G21) points to a potential hydrogen bond between A154 and Q195 suggesting a possible reason for the perturbation of in vitro CA assembly with the A154D and Q195A proteins. In an attempt to mimic this potential interaction we engineered an electrostatic mimic of the hydrogen bond with A154D and Q195K. The single mutant CA protein Q195K was able to assemble into tubes similar to Q195A, although the assembly rate was even slower (Figure 3.7B and 3.8). However, the double mutant

A154D/Q195K was unable to assemble at all indicating that the double mutant was not compatible with mature CA contacts (Figure 3.7B and 3.8). Whereas, we cannot rule out the possibility of the hydrogen bonding network from these results, A154 and Q195 may have evolved to establish a hydrogen bonding network with other residues in the mature dimer interface. Addition of charge to this region, as was done with the A154D/Q195K double mutant, may cause significant disruption of the interaction network thus we are unable to accurately interpret a negative result.

In vitro assembly data confirm the prediction that the A154D and A154V fail to assemble

CA in vitro. Assembly of the Q195A CA protein is delayed and forms structures with abnormal morphology suggestive of altered CA protein packing. These results are consistent with a defect in maturation for each of the viruses. The Gag defective mutants are also defective in CA assembly, indicating that W153A, L180A, and V188A mutations block assembly of both immature and mature particles. All of these residues are found at the mature dimer interface based on the low pH dimer (3G21), so it was likely that the in vitro CA assembly defects could be a result of failed CTD dimerization, at least for some of the mutants.

In Vitro Dimerization

The RSV CTD dimerizes in acidic buffer (pH 3.7) by neutralizing the aspartic residues and can promote assembly of full-length CA. Isothermal titration calorimetry has been used to show that the defect of maturation-incompetent MHR mutant proteins is at the stage of dimerization, consistent with in vitro assembly data. We directly tested the influence of all the

Figure 3.9. Effects of CTD mutations on dimerization. The dissociation constants (Kd) of the purified CTD and CTD-SP dimers were determined at low pH (3.7) by isothermal titration calorimetry. NDH, normalized ΔH (cal/mol per injection).

80 mutations on CTD dimerization by ITC. Purified isolated WT RSV CTD assembles into dimers under acid conditions with a weak but reproducible dissociation constant of 0.11 mM, similar to previously described results measured by ITC (Figure 3.9 and Table 1). A154G CTD assembled dimers with the same dissociation constant as WT. Q195A CTD dimerization was detectable but

4-fold weaker, consistent with its abnormal assembly behavior (Figure 3.9 and Table 1).

Dimerization was inhibited with A154D and A154V indicating that these mutants fail to assemble mature capsids as result of blocked dimerization. These results were consistent with well described maturation-defective CA mutations in which failure of pH-induced dimerization of

CTD mutants correlated with defects in virus spread. Likewise, the Gag assembly defective mutants were also unable to dimerize at low pH.

Summary of Mutants

All together these data allowed us to segregate the 7 mutants tested into 4 distinct groups.

The first included only the A154G mutant which was nearly identical to the WT virus for all assays. The second group consists of A154D and A154V which were generally competent for

Gag assembly but was non-infectious and had a maturation defect that prevented dimerization of the mature CTD-CTD interface and thus disrupted mature capsid assembly. The third group includes W153A, L180A, and V188A all of which were non-infectious were defective in Gag associations. These mutants were also unable to form mature capsids or dimerize. The sole constituent of the final group is Q195A which was infectious though it spread more slowly than the WT virus. It had deficiencies in particle release but was capable of assembling Gag.

Additionally Q195A produced abnormal in vitro assembled CA particles and had a 4-fold weaker dimer affinity than WT. Together this likely means that the Q195A mutant, though functional as a virus, is poorly efficient at both immature particle assembly and maturation leading to slower spread of the virus.

DISCUSSION

Retrovirus maturation is a choreography of organized structural rearrangements that requires the severing of immature lattice contacts and synthesis of new ones that build mature capsids. Structurally, the CA domain of Gag most prominently facilitates the interactions that create the immature hexameric lattice. Once freed from Gag the CA protein must then generate the fullerene structural lattice of the mature capsid through new intermolecular interactions

(25,185,194). The NTD and CTD of CA have been well characterized at the atomic level by x-ray crystallography and NMR. Hexameric and pentameric lattices of mature assembled CA have been mapped by cryoEM for several retroviruses, including HIV and RSV (44,194,231,245). However, high-resolution imaging of Gag and the immature lattice have been more complicated due to the inherently flexible nature of the molecule. Recent reconstruction of tubular assemblies of truncated Gag proteins to 8 Å resolution provided insight into immature interactions (186,187).

The structural models for both the immature and mature lattices highlight a critical CTD-

CTD interface between the 310 and α9 helices. The interface for both lattices creates interhexameric dimer linkages that are quite distinct despite both being composed of hydrophobic residues. These differences lead to the prediction that some of the residues at the interface are critical for forming the immature dimer whereas others are important for creating the mature dimer. We sought to test the prediction that hydrophobic residues are critical for establishing the immature and mature dimer interfaces though genetic and biochemical characterization of select mutant viruses.

The data presented here provide some of the strongest support for the models of maturation developed from the structural work on immature assembled particles. We have identified three critical mutants (W153A, L180A, and V188A) that ablate early stages of virus production, specifically the budding of immature particles and Gag assembly. From the immature dimer model presented here, these are three of the residues with the strongest predicted interaction. These same residues also have a role in making mature CTD dimer contacts as

82 predicted by the low pH RSV dimer structure. Furthermore, A154D, A154V, and Q195A provided moderate to no effect on Gag assembly, but are important for proper mature capsid assembly consistent with the predictions that the dimer interface is different between the immature and mature particles. The importance of all the residues on CA assembly is consistent with structural models that predict the low pH dimer of RSV (3G21) represents the CTD-CTD interaction occurring in mature capsids. In this model, the 310 helix of one molecule interacts with the α9 helix of the other molecule to form an interface that is linked via intermolecular and intramolecular salt bridges between aspartic and arginine residues (38). Our mutagenesis has provided support that non-electrostatic interactions are also important for establishing this interface.

Previous work in HIV established the critical importance of the residues W184 and

M185, located on the α9 helix, to the establishment of interhexameric and interpentameric interactions as mutation of both these residues to alanine resulted in production of isolated hexamers and pentamers (32,37,231,245). The recent cryo-ET structural models of HIV demonstrated that these same residues are likely important for formation of the immature dimer interface as well and act as a pivot point between the two interfaces, requiring a rotation of approximately 140 ° in order to establish mature contacts (186,187). Based on the low pH RSV

CTD dimer and modeling of this region into the existing MPMV immature electron density, it is likely that an intermolecular interaction between two V188A residues serves as the pivot point between the dimer conformations in RSV (38). The results presented here confirm the importance of V188 as a possible RSV CTD pivot point as substitution to alanine ablates assembly of both immature and mature particles.

W153 and L180 play important roles in both Gag and CA. The rotation in angles between the two interhexameric CTD molecules leads to large-scale changes in their side chain interactions. Fitting the RSV CTD into the density of the immature dimer model predicts W153 to be nearly equally distant from both intra- and intermolecular L180 residues (~4.1 Å and ~4.7 Å

83 respectively). Upon maturation to the well established mature CTD dimer, the W153 and L180 interactions change such that W153 moves closer to its intramolecular L180 (~4A) and away from L180 of the other molecule (~8A). Shifting of the CTD during maturation likely allows these residues to provide further stability to the electrostatic interactions as well as the intermolecular V188 association. Characterization of the W153 and L180 residues provides support for the models that predict both residues contribute to immature and mature interhexameric interfaces, as assembly of CA and ΔMBDΔPR is crippled when either residue is mutated to alanine. The hydrophobic interplay among the W153, L180, and V188 residues may be a critical feature for organizing the dimers and regulating maturation (38).

Rotation of the CTD molecules during maturation also alters the environment of residues

A154 and Q195. The immature model predicts that these residues do not provide any considerable contribution to interhexameric or intrahexameric interactions (186,187). The successful assembly of ΔMBDΔPR proteins bearing mutations at these residues and production of normal looking particles visualized by thin section supports the prediction. The mature dimer structure brings these side chains within 3 Å of each other with a possible interaction between

A154 and Q195 after CA maturation (38). Whereas we cannot confirm the presence of such an interaction, mutations at both residues altered assembly of mature CA protein in vitro. With the exception of glycine substitution, mutagenesis at residue A154 resulted in significant maturation defects pertaining to capsid assembly. Mutation of Q195 was more tolerable; however, the resulting products differ from the wild type protein and appear to be far less stable when exposed to non-ionic detergent. This could be due to a distortion in hydrogen bonding to A154 which leads to a capsid that is of lower quality and therefore cannot contribute adequately to successful reverse transcription (59,91,176).

Regulation of dimerization requires far-reaching intermolecular and intramolecular interactions, including from residues not expected to be at the interfaces. The maturation defects of A154D and A154V are reminiscent of the phenotypes of several MHR and MHR boundary

84 mutations previously described. In particular the D155Y substitution, which borders both the 310 helix and the MHR, allows for proper budding and assembly of immature particles but has a profound defect in maturation and creation of mature capsids (59,60,250,271). Recent published and unpublished data from our lab points to the contribution of the MHR to dimerization through a coordinated association with the 310 and α9 helices, despite the MHR helix not participating directly at the interface (60,250). However, the MHR sits directly between the 310 helix and α9, putting it in an excellent position to regulate how each of the helices coordinates at both dimer interfaces.

Both the NTD and CTD must somehow coordinate their interactions to work together to regulate assembly of both immature and mature particles. In spontaneous second-site suppressor screens, mutations arising in the CTD have been capable of restoring infectivity to lethal NTD mutations and vice versa (221,271). The vast surface area of the interhexameric (dimer) interface, along with the long-range interaction of NTD and CTD mutations implicates that initiation of assembly begins with the formation of the dimeric interface consistent with previous data

(60,241). This would imply that creation of a complete immature particle or capsid is not achieved by sticking hexamers and pentamers together, but rather, a smaller oligomer, such as a dimer or trimer, would function as the primary assembly unit. Proteolytic activation of PR would then lead to disruption of the immature lattice enabling rotational changes that establish the mature capsid. The data provided are consistent with the importance of formation of dimeric interactions in both the immature particle and mature capsid.

ACKNOWLEDGEMENTS

We would like to thank Anita Wade for performing the initial mutagenesis and in vitro

CA assembly characterization. We are grateful to Roland Myers for processing the thin sections presented here and the imaging core for their support with EM. We are also appreciative of Boon

Goh, Juan Perilla, and Klaus Schulten for their helpful insights into modeling the RSV immature

85 dimer interface. This work was supported by NIH grants R01 CA100322 (R.C.C.) and T32

CA060395 (M.R.E.) and by Tobacco CURE funds from the Pennsylvania Department of Health.

CHAPTER IV

A ROLE FOR CA-SP IN NUCLEATING CAPSID MATURATION

The following chapter is adapted from the following published manuscripts:

England MR*, Purdy JG*, Ropson IJ, Dalessio PM, and Craven RC. Potential role for CA-SP in nucleating retroviral capsid maturation. 2014. J. Virol. 88:7170-7177.

Keller PW, Huang R, England MR, Waki K, Cheng N, J. Heymann B, Craven RC, Freed EO, Steven AC. Retroviral capsid can mature by a disassembly-reassembly pathway to produce viable virions or by a displacive pathway producing aberrant virions. 2013. J Virol. 87:13655-13664.

ABSTRACT

During virion maturation the Rous sarcoma virus (RSV) capsid protein is cleaved from the Gag protein as the proteolytic intermediate CA-SP. Further trimming at two C-terminal sites removes the spacer peptide (SP) producing the mature capsid proteins CA and CA-S. Abundant genetic and structural evidence shows that the SP plays a critical role in stabilizing hexameric

Gag interactions that form and stabilize immature particles. Freeing CA-SP from Gag breaks immature interfaces and initiates formation of mature capsids. The transient persistence of CA-SP in maturing virions and identification of second-site mutations in SP that restore infectivity to maturation-defective mutant viruses led us to hypothesize that SP may play an important role in promoting assembly of mature capsids. This study presents a biophysical and biochemical characterization of CA-SP and its assembly behavior. Our results confirm previously published cryo-EM structures by Keller et al (2013) showing monomeric CA-SP is fully capable of assembling into capsid-like structures identical to those formed by CA. Further, SP confers aggressive assembly kinetics, suggestive of higher affinity CA-SP interactions than observed with either of the mature capsid proteins. The aggressive assembly observed with CA-SP is largely independent of the SP amino acid sequence, but formation of well-ordered particles is sensitive to the presence of the N-terminal β-hairpin. Additionally, CA-SP can nucleate assembly of CA and

CA-S. These results suggest a model in which CA-SP, once separated from the Gag lattice, can actively promote the interactions that form mature capsids and provide a nucleation point for mature capsid assembly.

IMPORTANCE

The spacer peptide is a documented target for antiretroviral therapy. This study examines the biochemical and biophysical properties of CA-SP, an intermediate form of the retrovirus capsid protein. The results demonstrate a previously unrecognized activity of SP in promoting capsid assembly during maturation.

INTRODUCTION

The CA protein is the major structural protein in retroviruses, forming the capsid structure in mature virions (42,45,230). CA is processed from the polyprotein Gag by the viral protease (PR) whose activity is initiated coincident with or immediately upon virus budding from the cell surface. The highly complex rearrangement process that ensues leads to condensation of an ribonucleoprotein (RNP) structure comprised of the nucleocapsid (NC) protein and viral RNA with the replicative enzymes reverse transcriptase (RT) and integrase (IN) and the building of the mature capsid (a hexameric lattice punctuated by irregularly spaced CA pentamers) around the genome complex (25,44,181,185,194,200,233,287,290,291). Integrity of the capsid shell is critical for proper function of the genome complex upon entry to a new cell

(49,59,63,91,271,292). However, the mechanistic or structural factors that control the molecular rearrangements of maturation are poorly understood.

The release of the mature capsid subunit CA from Gag is a multistep process that in viruses of the alpha- and lentiretrovirus groups requires cleavage at a unique upstream site at the

N-terminus of CA and 2 -3 cleavages between CA and the downstream NC domain (Figure 4.1)

(45,74,75,199,200,293). The initial cleavages at the N-termini of CA and NC free NC to begin formation of the RNP and also yield immature capsid proteins that bear a short spacer peptide

(SP) at the C-terminus. These intermediates, known as CA-SP in Rous sarcoma virus (RSV, an alpharetrovirus) and CA-SP1 in the human immunodeficiency virus (HIV-1, a lentivirus), persist for several hours (180,200,260). In RSV, PR gradually removes either 9 or 12 residues from the

C-terminus of CA-SP producing a mixture of two mature species – CA (237 amino acids)

Figure 4.1. Proteolytic Processing During Maturation. A multistep sequence of cleavages of immature Gag molecules first produces the capsid intermediate CA-SP. Over 3-6 hours 9 or 12 spacer peptide residues are removed, leaving CA and CA-S to form the mature capsid. The gray arrow shows the initial cleavage generating CA-SP (249 aa) whereas black arrows represent the terminal SP cleavages resulting in the mature capsid proteins CA (237 aa) and CA-S (240 aa).

90 and CA-S (240 residues) (199). HIV-1 generally contains only a single species of CA (CA/p24) in wild-type virus although cryptic cleavage site(s) have been identified in the SP region that can yield alternative CA form(s) (200,291). The fully matured RSV and HIV proteins are highly similar in structure and consist of two alpha-helical domains, N-terminal (NTD) and C-terminal

(CTD), separated by a short linker. A short unstructured region follows the last helix in the CTD

(32,40,48,50,51,57,294).

Abundant evidence supports a critical role for the SP sequence both in the earliest events of Gag assembly and in forming stabilizing interactions in the immature Gag lattice. In immature

Gag shells of HIV, RSV, and M-PMV (the Mason-Pfizer monkey virus) cryo-electron tomographic studies identified intrahexameric SP interactions that stabilize rings of Gag molecules, whereas the CA CTD and NTD domains make inter-hexameric contacts to build the larger lattice (36,79,181,182,254,287). Biophysical studies of monomeric Gag fragments and synthetic SP peptides suggest that the region is more extended but capable of undergoing a coil- helix equilibrium transition (78,79,267-269). Furthermore, RNA binding to the NC domain of

HIV and RSV Gag triggers structural changes within SP (267) and promotes Gag dimerization via the upstream CA CTD domain (78,284). This suggests a model wherein the concentration of

Gag on RNA produces structural changes transmitted via the SP to the CA CTD allowing dimerization and the subsequent growth of the lattice (267).

This structural role for SP is supported by extensive genetic studies that demonstrate the extreme sensitivity of immature particle assembly to SP mutations. In HIV, highly aberrant structures or even total disruption of particle release is seen upon alteration or ablation of SP1 cleavage (75,263,291,295). Deletions and sequence alterations within the RSV Gag SP and flanking residues of CA and NC crippled infectivity and caused disordered Gag assembly and production of particles with altered size and morphology (247,254). Although precise deletion of the 9 residues of RSV SP allowed production of some particles of normal size and morphology, considerable size heterogeneity was noted and the cores of these particles were abnormal and

91 infectivity was destroyed (180,247). These studies emphasize the role of SP in the immature lattice but do not address any role during the maturation process.

Cryo-electron tomography studies of HIV-1 particles in which the sequence of proteolysis has been interrupted by cleavage site mutation or by treatment with a maturation inhibitor compound have provided some insights into activities of SP in the maturing particle.

When cleavage at the SP-NC junction occurs, interactions that helped to stabilize the intrahexameric contacts are destroyed and the condensation of NC with genomic RNA to form the nucleocapsid proceeds normally. However, if CA-SP1 is fully cleaved from Gag, but the subsequent trimming of the SP sequence from the CA C-terminus is prevented either by inhibitor treatment or by mutagenesis of the CA/CA-SP1 junction, the Gag lattice remains partially intact and only limited protein patches that resemble the mature capsid lattice are seen. The formation of complete capsids and acquisition of infectivity are blocked (74,75,287,291). Thus, trimming of the SP from CA during maturation is essential for successful completion of the structural protein rearrangements in virions.

Genetic studies in RSV provide a strong argument that the transient presence of SP at the end of CA influences the outcome of maturation. A second site mutation (S241L) in SP is capable of partially suppressing the capsid assembly defect caused by a primary mutation (F167Y) in the first α-helix of the CTD that compromises the ability of CA to initiate capsid formation (62).

Since little is known about the self-interaction properties of the CA-SP processing intermediate, we undertook a detailed analysis of purified RSV CA-SP to compare its biophysical and assembly properties to those of the well-characterized CA protein (241,250). These findings document a clear difference in the assembly behavior of CA and CA-SP which suggests that the transient presence of CA-SP in maturing virions may actually promote the formation of capsids by nucleating mature CA interactions and provide a biological rationale for the conservation of staggered cleavages.

RESULTS

The Monomeric CA-SP and CA-S

Recombinant CA-SP, the 249 residue processing intermediate, and the 240 residue mature CA-S proteins were expressed in E. coli and purified for comparison with the mature CA

(237 residue) that has been extensively characterized in vitro (241,250,266,282). The purified

CA-S and CA-SP, like CA, were almost exclusively monomeric (Figure 4.2A). The CA-SP protein eluted slightly faster from size exclusion chromatography column than either CA or CA-S as expected from its higher mass. In SDS-page electrophoresis, the purified CA-SP exhibited the same anomalously fast migration that was previously described in virus-infected cells (Figure

4.2B), suggesting that CA-SP adopts a more compact structure in SDS than do the two shorter capsid proteins (199,260).

The biophysical properties of the monomeric CA-S and CA-SP were indistinguishable from CA. The circular dichroism spectra for CA, CA-S, and CA-SP were superimposable, suggesting that all three proteins are properly folded and have similar secondary structures

(Figure 4.2C). The midpoint of unfolding in the presence of the denaturant guanidine hydrochloride was the same (~2.2 M) for CA and CA-SP indicating that SP has no effect on the stability of the monomeric protein (Figure 4.2D). In addition, the SP extension had no significant effect on dimerization of the CTD as measured by isothermal titration calorimetry (Figure 4.3).

SP Confers Distinctive Protein Assembly Properties

CA-S represents nearly half of the capsid protein in mature RSV virions (199).

Considering that as much as 80% of the total available capsid protein is incorporated into the mature virion core during maturation, it is expected that both CA and CA-S contribute to the capsid shell (25). The self-interaction ability of CA-S was tested in vitro using a sodium phosphate-triggered assembly reaction previously developed for RSV CA (241). Under standard conditions of 80 µM protein and 500 mM sodium phosphate at pH 8.0, the monomeric CA-S protein indeed was capable of assembly (Figure 4.4A). The kinetics of assembly, followed

Figure 4.2. Purification and characterization of capsid proteins. A) CA, CA-S, and CA-SP expressed and purified from E. coli were analyzed by SEC using a Superdex S75 column. B)

Molecular weights were identified by 15% SDS-PAGE. CA-SP migrates faster as has been described for CA-SP isolated from virions. C) Secondary structure of full-length CA, CA-S, and

CA-SP was compared by circular dichroism from 195-250 nm. D) The stability of each protein in 0-5.5 M guanidine hydrochloride was determined by monitoring α-helical content at 222 nm.

CA-S was not shown for clarity

Figure 4.3. Effects of SP on CTD-CTD dimerization. The dissociation constants of the purified

CTD and CTD-SP dimers were determined at low pH (3.7) by isothermal titration calorimetry.

Curves of the integrated heat are shown and the dissociation constant for each is reported.

Figure 4.4. Kinetics of CA, CA-S, and CA-SP assembly. A) In vitro assembly of CA, CA-S, and

CA-SP at 40 µM and 80 µM protein concentration was triggered with 500 mM sodium phosphate and turbidity at 450 nm was monitored. Resulting assembly products of B) CA, C) CA-S, and D)

CA-SP were examined by negative stain TEM. Bar = 100 nm. E) In vitro assembled protein (80

µM) was collected after the curves reached plateau and the samples were spun at 128,000 x g for

30 minutes. Unassembled protein was determined by Nanodrop-3000. Error bar represents 4 replicates. *P < 0.01, **P < 0.005.

96 turbidimetrically, were very slightly faster than observed with CA alone, although this effect was not uniformly reproducible. In a co-assembly reaction containing CA and CA-S, the kinetics of assembly and the final turbidity achieved were indistinguishable from that observed in single protein reactions (Figure 4.5A). The turbidity profiles in all cases resembled typical nucleation- driven reactions with a distinct lag followed by rapid rise.

Both mature capsid proteins (CA and CA-S) form abundant spheres of 17 nm diameter in sodium phosphate (Figure 4.4B and 4.4C). Spheres of 32 nm diameter made up 1.7 ± 1.1 (CA, n=504) and 1.5 ± 2.1 (CA-S, n=210) percent of the population. Tubes and larger angular and multi-lamellar structures that resemble the capsid-related structures observed in authentic virions were seen rarely (250). The co-assembled CA and CA-S samples contained a similar distribution of structures (Figure 4.5C). Thus, the three amino acid extension confers a slight increase to assembly properties to the protein, but does not alter the types of particles that can form.

The behavior of the cleavage intermediate CA-SP was strikingly different than that of either mature protein CA and CA-S. The presence of the extra 9 spacer peptide residues conferred very fast assembly behavior (Figure 4.4A). Under standard conditions CA-SP assembled rapidly with no discernible lag phase. As with CA and CA-S, CA-SP self-assembly was protein concentration-dependent, with the lag and growth phases becoming clear with reduction of protein concentration (Figure 4.4A). The dramatic difference between CA-SP and the mature proteins CA and CA-S was most evident at 40 µM protein concentration, where the mature proteins failed to produce any detectable turbidity (Figure 4.4A). Consistent with this, the concentration of protein remaining unassembled (resistant to pelleting at 128,000 x g) was ~ 6

µM for CA-SP, about 3-fold and 2-fold lower than measured for CA and CA-S respectively

(Figure 4.4E), suggesting a lower critical concentration for nucleation of assembly.

Figure 4.5. Co-assembly of CA, CA-S, and CA-SP in vitro. Various ratios of A) CA, CA-S, and

CA-SP and B) CA and CA-SP were mixed and assembled as in Figure 4.4. TEM images of C)

CA:CA-S:CA-SP (60:40:0, x in A) and D) CA:CA-SP (95:5, Δ in B) are shown. Bar = 100 nm.

CA-SP Nucleates Mature CA and CA-S Assembly

To test whether CA-SP could initiate the assembly of the mature CA and CA-S in vitro, immature and mature proteins were mixed together in varying ratios, keeping the total protein concentration constant at 80 µM, prior to addition of sodium phosphate. In all reactions containing CA-SP, assembly was robust and proceeded with a shorter lag than either CA or CA-S

(Figure 4.5A). In reactions containing CA and CA-SP, a proportion of immature protein, as low as 5% of the total, was able to stimulate the assembly of CA (Figure 4.5B). The co-assembly reactions yielded small spheres that resembled those formed by CA alone (Figure 4.5D). Thus, it appears that the presence of CA-SP can nucleate the assembly of the shorter mature proteins.

CA and CA-SP Form Structurally Similar Assembly Products

EM examination of the CA-SP assembly products showed structures resembling those of

CA and CA-S and included abundant spherical particles of 17 nm in diameter and multilayered particles 32 nm in diameter (Figure 4.4D). In a typical experiment, the larger particles represented

7.8 ± 1.7 (n=857) percent of the population, slightly higher than frequency observed with CA or

CA-S (above). In addition, CA-SP formed occasional tubular and capsid-like structures. Three- dimensional single-particle reconstructions from 200 negatively-stained 17 nm particles and from

60 of the 32 nm particles revealed that they are consistent with T = 1 and T = 3 icosahedral symmetry respectively, similar to those reported previously for CA (Figure 4.6A and 4.6B)

(194,232).

These results were further extended in a collaborative study by high resolution cryo- electron microscopy. T = 1 particles at 8.5 Å resolution showed that CA-CA interfaces in in vitro assembled CA-SP are nearly indistinguishable from those in CA particles and that the spacer peptide itself did not contribute regular interactions within the particles (Figure 4.7A and Figure

4.8A and 4.8B) (74). T = 3 particles were at 18 Å resolution appeared very similar to those seen with CA protein (Figure 4.7B). T = 1 capsids were found nested within each of the T = 3 particles for each RSV protein (Figure 4.7C). Unlike the previous finding of CA, in which the two nested

Figure 4.6. Negative-stain reconstruction of CA-SP. Purified CA-SP protein was assembled in vitro with 500 mM sodium phosphate and the resulting products were stained and visualized by

TEM. A) Small spheres (17 nm) were reconstructed by EMAN and determined to have T = 1 icosahedral symmetry. B) Larger spheres (32 nm) were reconstructed with T = 3 symmetry.

100

Figure 4.7. Reconstruction of CA-SP from Cryo-EM. CA-SP was induced to assemble in vitro by addition of sodium phosphate. Particles were applied to holey carbon grids, frozen, and examined by cryo-EM. A) Reconstructions of the smaller particles showed T = 1 icosahedral symmetry.

The larger spheres were made of two layers of CA-SP protein. B) The outer layer was reconstructed with T = 3 icosahedral symmetry whereas C) the inner shell had T = 1 symmetry

(74).

101

Figure 4.8. Fitting of CA domains into CA-SP T=1 particle density. (A) RSV CA-NTD and CA-

CTD atomic models were fitted into the CA-SP T=1 reconstruction independently, as rigid bodies. (B) The modelled CA-SP pentamer (magenta) is compared to the mature CA pentamer structure (cyan). The two structures are shown superimposed both from the top and inside view

(74).

102 capsids are in register with their vertices aligned, reconstructing the CA-SP assembly product with respect to the T = 3, resulted in the T = 1 particle appearing as a smooth sphere and vice versa. It was likely that the presence of SP in the larger particle disrupted the ability of the smaller particle to align with the larger particle. Although SP could not be resolved due to its flexible nature, density associate only with CA-SP (i.e. density that was likely attributed to SP) in the larger particle came in close contact with the MHR region of the smaller sphere.

The similarity of CA and CA-SP protein assembly behavior was confirmed by testing the influence of mutations known to specifically cripple the ability of CA protein to build capsid structures. The conservative substitution F167Y in the first α-helix of the CTD is a well characterized allele that cripples the nucleation of capsid formation in vitro and in maturing virions whereas allowing apparently normal Gag function (59,61,62). This mutation destroyed the ability of CA-SP to assemble in vitro (Figure 4.9), as reported previously for CA. The previously identified second-site suppressor for F167Y (S241L) was unable to rescue assembly of the protein, indicating that the influence the SP has toward fixing the F167Y defect is complex and cannot be replicated by our in vitro system. (Figure 4.9) Next the importance of the N-terminal β- hairpin was tested. Formation of this structure, which forms upon cleavage of CA from Gag and is a hallmark of mature capsid assembly, can be prevented by substitution of an alanine at the aspartate at position 52 in helix 3 which serves as a docking site for the N-terminal proline in mature capsid proteins (49,95,229,250). Monomeric D52A CA develops turbidity with rapid kinetics but is unable to form organized particles (250) (also Figure 4.10A and 4.10B). The same substitution in CA-SP likewise caused rapid precipitation without organized particle formation

(Figure 4.10A and 4.10C). The assembly-destroying activity of D52A was prevented, however, by an extension of CA-SP by eight residues of the NC protein so that the entire length of the known Gag interaction domain, spanning the SP sequence and flanking regions, was present.

Both the WT and D52A forms of the CA-SP-8NC assembled rapidly under standard conditions

(Figure 4.10A) and produced spheres that appear by negative stain EM to resemble the 17 and 32

103

Figure 4.9. In vitro assembly of a SP-localized suppressor of F167Y. In vitro assembly of purified

CA-SP proteins was completed as in Figure 4.4 and monitored turbidimetrically. EM images of

CA-SP F167Y, S241L, and F167Y/S241L are shown. Bar = 100 nm.

104

Figure 4.10. Assembly of CA-SP-8NC and proteins with the D52A substitution. A) In vitro assembly of proteins was completed as in Figure 4.4. EM images of B) D52A CA, C) D52A CA-

SP D) CA-SP-8NC, and E) D52A CA-SP-8NC are shown as in Figure 4. Black arrows show new class of larger, angular particles seen in CA-SP-8NC constructs. Bar = 100 nm.

105

Figure 4.11. Effects of mutations on CA-SP assembly. A-C) The ability of CA-SP to assemble with single and triple alanine substitutions or triple proline substitutions was analyzed as in

Figure 4.4. Representative TEM images of assembly-competent mutant proteins are depicted by

D) M240A and E) CA-SP (A1). F) CA-SP (P1) only formed aggregated complexes visualized by

TEM. Bar = 100 nm.

106 nm structures formed by CA-SP (Figure 4.10D and 4.10E). In addition, both proteins formed a new subclass of larger, less regular particles formed that made up 4.6 ± 0.9 (CA-SP-8NC, n=2083) and 3.8 ± 1.1 (D52 CA-SP-8NC, n=1464) percent of the total population, suggesting some tendency of the longer proteins to form interactions not seen with CA-SP protein (Figure

4.10D and 4.10E, arrows). The slight increase in assembly kinetics of the D52A protein suggests that an unfolded N-terminus promotes association. Taken together these findings indicate that

CA-SP assembles with properties characteristic of mature capsid protein, rather than its Gag precursor.

Probing the Amino Acid Sequence of SP

A series of mutations in the SP sequence were created to assess the importance of specific residues to the unique assembly characteristics of CA-SP (Figure 4.11). None of the single or triple alanine substitutions tested disrupted CA-SP assembly. Each displayed the rapid assembly kinetics that is characteristic of CA-SP and produced similar populations of small icosahedral particles. Compared to WT CA-SP, each of the triple alanine substitution mutants assembled slightly more slowly although the final products were similar (Figure 4.11B-4.11E).

To create potentially more disruptive alterations, clusters of three residues were replaced with prolines. Each of the three triple-proline mutants developed turbidity with the rapid kinetics of the wild-type CA-SP (Figure 4.11A and 4.11C). In contrast to the single and triple alanine mutants, however, the P1 triple proline mutant completely failed to produce well ordered particles (Figure

7F), whereas P2 and P3 produced some small spherical particles along with abundant amounts of precipitated protein. These observations suggest that the influence of SP on protein assembly is largely independent of the amino acid sequence but may be influenced by unknown structural properties of the spacer peptide.

Dynamics of the Spacer Peptide

NMR and crystallography have failed to determine the structure of SP due to its inherently dynamic nature. The cryoEM study from Keller et al. identified electron density

107 exclusively associated with the CA-SP protein, but was unable to specifically locate amino acids of SP. The larger particles of CA-SP contained nested T = 1 and T = 3 particles that were not in register with each other. However, it did appear as though the MHR of one came in close association with the SP of the other. In a study of the effects of maturation inhibitors on HIV-1 replication, escape mutants were found within the MHR. These two studies point to a potential interaction between the MHR and SP during maturation.

NMR HSQC spectra for purified CTD and CTD-SP proteins were nearly indistinguishable consistent with assembly and cryoEM data and previously reported NMR structures (40). To address the dynamics of the SP, we performed PRE NMR analysis on three different mutant CTD-SP proteins (S241C, Q245C, and 250C insertion) which were labeled with

MTSL. PRE NMR allows detection of transient and long range associations (~ 25 Å). HSQC spectra were taken before and after quenching the spin label with ascorbic acid. Residues that have the greatest reduction in intensity when the spin label is active come in closest contact with the labeled residue. Several residues were identified as likely candidates for close SP association, including MHR residues L171 and E176 and α9 residues R185, A186, and D191 (Figure 4.12 and

Table 4.1). Importantly many of these residues have been reported for perturbing maturation or correcting a maturation defect. These PRE NMR data, though preliminary, indicate the SP may come close to the MHR and/or α9 helix influencing how the CTD matures after initial Gag cleavages consistent with cryoEM and mutagenesis of CA-SP particles.

DISCUSSION

Retroviral maturation in RSV begins in a protein environment that is in flux. Due to the staggered timing of cleavages that separate the CA and NC domains in Gag, the declining pools of the longer CA-SP intermediate co-exist for some hours with the accumulating mature CA and

CA-S proteins. This persistence of CA-SP in the maturing virion, as well as the conservation of such staggered cleavages across retrovirus subfamilies, are consistent with the idea that the SP plays a role (or even multiple roles) in controlling the outcome of the maturation process.

108

Figure 4.12. Dynamics of the spacer peptide. 15N-labeled C192R CTD-SP with secondary mutations S241C (shown), Q245C, and 250C insertion were covalently bound to MTSL spin label and analyzed by PRE. NMR HSQC spectra were taken before and after addition of ascorbic acid. Residues which are affected most by the presence of the spin label are depicted in red on a

RSV CTD monomer from 3g21.

109

Table 4.1. PRE NMR Measurements for CTD-SP S241L - MTSL + MTSL Volume ID Intensity Volume Intensity Volume Change 191 3.7897 34.39609 0 0 -100% 227 3.0331 27.75041 0 0 -100% 234 3.3305 19.99543 0 0 -100% 235 2.9177 13.80626 0 0 -100% 185 5.0548 30.69065 2.2236 4.406998 -86% 211 2.3074 8.392413 2.158 2.158001 -74% 219 4.5237 30.38291 2.614 9.346554 -69% 229 4.371 36.08601 2.7521 11.68162 -68% 186 5.1501 32.35598 3.4504 13.58406 -58% 221 6.6887 45.5775 3.4935 19.9249 -56% 231 5.5367 52.15033 3.3142 24.11923 -54% 200 4.6587 34.51778 3.6381 16.93276 -51% 176 5.0668 37.65905 3.6448 22.49485 -40% 222 3.6482 36.22907 3.0039 22.13513 -39% 203 4.9027 43.60436 4.3856 27.02626 -38% 224 6.9038 48.4248 4.8257 31.68197 -35% 166 4.6698 43.40822 4.1327 28.86874 -33% 165 5.3674 39.58218 3.6317 26.44517 -33% 223 5.1462 36.83215 4.0518 24.6172 -33% 172 3.8401 30.8773 3.3091 23.40412 -24% 154 5.0672 42.11558 5.1744 33.90561 -19% 201 3.8238 30.10338 3.3677 24.26904 -19% 194 2.1411 13.70237 3.0446 11.29095 -18% 217 6.804 101.2978 5.5962 83.51812 -18% 177 4.3527 29.89469 4.1725 25.87416 -13% 148 3.7779 25.65397 3.5243 22.38774 -13% 158 2.7366 18.47458 2.488 16.20083 -12% 212 4.3826 44.24949 4.3278 39.92572 -10% 225 4.7059 34.52879 4.1791 31.41439 -9% 233 4.121 44.328 4.3645 40.46738 -9% 163 3.9284 31.08287 3.9297 28.60841 -8% 168 4.1624 18.00728 3.4915 16.88232 -6% 173 4.3528 28.95991 3.3207 27.54115 -5% 162 5.002 49.38964 4.5052 47.30817 -4% 228 6.6679 58.62051 6.4132 61.24981 4% 167 2.912 22.66497 3.269 24.17095 7%

110

The underlying basis of the slow trimming of spacer peptide from CA-SP is incompletely understood. In the case of the HIV-1 PR, slow cleavage at the CA/SP1 site due in part to its amino acid sequence but also to unknown “context” features of the protein (293). The biological consequences of staggered cleavages, however, have been examined in detail by electron microscopy with HIV-1 (74,75,287,291). Cleavage at the SP1/NC junction is sufficient to allow condensation of the nucleocapsid to proceed. If cleavage of SP1 from CA is prevented, much of the CA-SP1 remains entangled in remnants of Gag lattice, and the building of a proper capsid is aborted. Thus, it is clear that one consequence of the staggered cleavages in the spacer region is the temporal control of the formation of the different layers of the viral core.

In addition to controlling the timing of the maturation process, studies in RSV and HIV

(62,74,75,78,251,296) suggest that the transient presence of CA-SP in the maturing virion also exerts a direct influence on capsid protein structure that promotes the formation of mature CA-

CA contacts. An F167Y substitution in the CTD hydrophobic core allows apparently normal Gag function but cripples capsid maturation by a potent effect on the formation of the mature CTD-

CTD dimer interface. The subtle perturbation of CTD structure caused by F167Y can be accommodated by the cooperative action of any one of several interface-stabilizing mutations elsewhere in CA that can promote nucleation of capsid assembly and restore infectivity

(60,61,250,271). Relevant to this study, F167Y lethality can also be partially suppressed by a compensatory substitution in the spacer peptide (S241L) suggesting that the unique protein interactions that build the mature capsid may be initiated prior to removal of the SP peptide (62).

In a similar observation with HIV, a replication-crippling NTD mutation H62F at a position contributing to the mature inter-domain interface is suppressed by a compensating change in the

SP1 (296).

The transient presence of SP at the end of the capsid protein could influence the outcome of capsid formation in at least two different ways. The SP sequence may exert a negative regulatory effect on CA-SP to prevent mature-type interfaces from forming until the final

111 cleavage removes SP at the appropriate stage of maturation. Alternatively, the SP may actively promote the formation of a CA-SP intermolecular interaction that is a critical intermediate step of capsid maturation. The genetic studies alone cannot distinguish between these possibilities.

Therefore, in this current study we have utilized an in vitro capsid assembly assay developed and characterized by our laboratory to test directly the influence of SP on formation of mature CA-

CA interfaces.

The results presented here demonstrate that E. coli-expressed CA-SP protein, identical in amino acid sequence to the maturation intermediate detected in maturing virions, is capable of robust assembly in vitro. The CA-SP protein forms structures that resemble closely the mature capsid-like structures formed under similar conditions by the RSV CA protein, including the unique NTD-CTD interface that distinguishes the mature capsid lattice from the CA interactions formed by Gag (74). The powerful inhibitory effects of D52A and F167Y mutations, which are known to prevent normal CA capsid assembly, further support the interpretation that CA-SP possesses assembly properties that closely mirror those of CA.

Thus, we conclude that the extra C-terminal residues that comprise the spacer peptide do not confer an inherent inhibitory effect on CA-SP. Rather, the assembly competence of CA-SP in vitro predicts that CA-SP generated in maturing virions early upon activation of the proteolytic processing is also competent to begin capsid formation, at least once it is freed from entanglement in the remnants of the Gag lattice. Consistent with this is the observation of partially complete capsid-like structures in HIV virions in which the trimming of CA-SP1 to CA is interrupted by mutation or drug treatment (74,75).

No convincing difference in the assembly ability of CA and CA-S, the three residue longer mature form of capsid protein, were uncovered in this work. In fact, there exists no evidence to date supporting the possibility that the two species make unique contributions to capsid formation (297). Instead, the presence of the nine further residues of the SP confers an aggressive assembly capability compared to either CA or CA-S. CA-SP particles form under

112 standard conditions with little lag. The presence of CA-SP is also able to nucleate the aggressive assembly of mature capsid protein in trans, indicating that CA-SP, after release from Gag, can quickly begin building mature protein contacts and may serve as the initiation point of capsid assembly. Such an activity is consistent with the identification of spacer peptide mutations that promote capsid maturation in HIV and RSV.

What properties of CA-SP confer its unique assembly behavior in vitro? We considered the possibility that the SP sequence may either form a direct inter-subunit bridge to help promote protein assembly or, alternatively, that the peptide may make an intra-subunit interaction with the

CTD that allosterically influences the affinity of the CTD interactions. Either scenario would be consistent with observations in HIV suggesting that in the Gag lattice the SP1 sequence lies in close proximity to the CTD, specifically the major homology region or MHR (74,227,298).

However, the relative insensitivity of CA-SP assembly to alanine or proline substitutions in the spacer peptide seems to argue against such scenarios that require a direct docking of SP against the CTD. This is in contrast to in vitro assembly of Gag protein which was largely intolerant of

SP mutations (79).

It appears likely that the mere presence of extra amino acids appended to the end of the capsid protein promotes CA-CA interactions by influencing the structure of a critical upstream element in the CTD, possibly the last alpha-helix which is known to contribute to both CTD-CTD interactions and CTD-NTD inter-domain interface. Although an NMR analysis of the RSV CA,

CA-S and CA-SP proteins (232) did not detect any obvious effect of the SP on the structure of the alpha-helical bundle, preliminary data on CTD-SP presented here suggest that the SP can come close to the MHR and α9 helix. A transient interaction between these regions could influence how the CTD assembles mature interfaces thus regulating the outcome of maturation. Further analysis of protein dynamics will be needed to confirm this structural effect of SP. Retroviral capsid proteins are characterized by an unusual degree of internal dynamics in regions that are important to their assembly/disassembly capabilities, e.g., the CTD and the flexible loop region on the

113 outside surface of the NTD (65,299,300). In particular, the recent solid-state NMR comparison of assembled HIV-1 CA and CA-SP1 proteins has provided detailed evidence that the presence of

SP1 causes long-distance structural effects within these regions of the CTD and the NTD and provides strong support for this interpretation of the RSV CA-SP results presented here (252).

The maturation process that produces a functional capsid is a complex event that requires a finely tuned balance of many molecular events – the temporal pattern of PR cleavage kinetics at staggered sites, the severing of assembly domains that stabilized the Gag lattice of immature virions and the establishment of novel contacts to build the mature capsid. The results presented here demonstrate that the intermediate CA-SP, once freed from the Gag lattice, may promote the nucleation of the mature capsid and suggest that the transient persistence of CA-SP in maturing virions may be evolutionarily conserved in part for its importance to the outcome of maturation.

ACKNOWLEDGEMENTS

We are grateful to Roland Myers and the imaging core for their support with EM, Susan

Hafenstein and Lindsey Organtini for their assistance in negative stain reconstructions, and John

Flanagan for his advice and assistance on protein purification. This work was supported by NIH grants R01 CA100322 (R.C.C.) and T32 CA060395 (M.R.E.) and by Tobacco CURE funds from the Pennsylvania Department of Health.

114

CHAPTER V

IMPORTANCE OF THREONINE 214 AT THE THREE-FOLD INTERFACE

115

INTRODUCTION

Biochemical and structural analysis of the mature CTD-CTD dimer indicate that the domains are quite flexible and are capable of orienting themselves to each other in slightly different ways (32,38,44,58,244,299,301,302). The seemingly trivial variation of the dimer conformation is absolutely critical in order to establish the curvature necessary to make mature capsids (191,245,303,304). The plasticity of the dimer is caused by slight conformational variations in two ways – at the tertiary level as the fold of the CTD has some flexibility to it in regards to the way the helices pack against each other, and the quaternary level due to variations in the crossing angle between two CTD molecules (194,231,245). In vitro assembly CA products of various retroviruses have not accurately portrayed the multitude of dimer interfaces within a single particle (231,244). However, computer modeling algorithms of HIV-1 have demonstrated that all of the different dimers found in vitro can be used to make a complete, mature HIV-1 capsid (245,304).

One such recently identified assembly CTD-CTD interaction occurs at the three-fold axis of symmetry. Cryo-electron microscopy of HIV-1 tubes that form a lattice of hexamers revealed that the CTD residues at the three-fold axis form an interaction involving polar residues (Figure

5.1) (244). Previous HIV-1 dimers were identified in purified CTD alone or in lattices from sheets of CA, and therefore the three-fold interface had not been identified in prior studies.

Formation of the three-fold axis only occurs after the final maturation cleavage that separates CA from SP1 (278). Thus, it appears that the three-fold interface is critical to the formation of the curvature in mature capsids.

The structural proteins of retroviruses are noted for their generally low level of primary sequence identity compared to the significant structural homology amongst retrovirus species.

There have been two previously identified regions of the CA protein that are relatively conserved amongst the retroviruses – the MHR and the flexible loop region of the NTD. Both regions are

116

Figure 5.1. The CTD three-fold interface. NMR and cryo-EM of HIV-1 CA molecules assembled into tubes revealed a new mature CTD-CTD interface. The three-fold interface includes α11 of the CTD from three neighboring hexamers. The pentameric version of this same interface is shown from the pseudo-atomic model of T=1 icosahedral RSV CA protein. The three-fold axis of symmetry is depicted by the white triangle.

117 important to virus infectivity. The MHR serves as a regulator of assembly and maturation; the flexible loops provide contact points for a variety of cellular factors. With the discovery of the three-fold interface, a third region of conservation was revealed (244,278). Within in the interface, at the top of α11, there is a Threonine residue conserved among several retroviruses, including RSV (T214), HIV-1 (T210), M-PMV (T204), and EIAV (T195) (Figure 5.2).

This conserved threonine is of particular interest because it is also a putative phosphorylation site as identified by the NetPhos 2.0 Server (Figure 5.2). A variety of viruses utilize post-translational modifications, including phosphorylation, in order to regulate events within their life cycles, including assembly and genomic packaging. The presence and absence of phosphorylation of specific residues in the core protein of HBV are thought to act as a determinant for genome packaging and reverse transcription during hepadnavirus maturation

(305-307). Gag can be phosphorylated, specifically the MA and p6 domains of the HIV-1 Gag protein, though not all the functions of phosphorylated Gag have been elucidated (308-310).

However, there has been little direct evidence that the CA domain in phosphorylated at any stage of Gag production and assembly. Despite a variety of potential phosphorylation sites within the

CA protein, only the HIV-1 S16 residue has been studied further. This residue is phosphorylated by the Erk2 kinase after entry into a new cell, and is therefore not a maturation switch, but rather thought to be a determinant of capsid uncoating (311,312). Currently no CA phosphorylation site that regulates maturation of the capsid has been discovered.

The intrigue of the conserved threonine residue is furthered by its localization at the beginning of the final CTD α-helix, as α11 has been identified as the site of interaction between the HIV-1 Gag molecule and the lysine tRNA synthetase (lysRS) (67,313). This interaction allows incorporation of tRNALys utilized for priming reverse transcription (314). Alanine substitutions for residues in α11, including T210A, demonstrated a two to three-fold decrease in binding between the HIV-1 CTD and lysRS thought the effect on virus infectivity and genome packaging was not analyzed (67). In RSV, trpRS is packaged in order to bring in the tRNATrp

118

Figure 5.2. Alignment of retrovirus CTD sequences. The CTD sequences of four different retroviruses are aligned from the beginning of the MHR. The putative phosphorylation sequence is underlined and the potentially phosphorylated threonine is denoted by the black arrow.

119 used in priming RT; however, it is not clear if the same CTD-trpRS binding partnership occurs in

RSV (98).

Based on these data, we predicted that the conserved threonine T214 is a phosphorylation site in RSV that acts as a critical switch for maturation of the retrovirus capsid and has a role in packaging trpRS. Considering all the evidence, it is likely that the residue is phosphorylated whereas part of Gag and becomes dephosphorylated to trigger maturation. To test this hypothesis, we made alanine and aspartate mutations at position T214 in RSV to mimic the constitutively dephosphorylated and phosphorylated threonine respectively. Although both mutant viruses were rendered non-infectious, there did not appear to be any significant defects on the ability of CA to assemble products in vitro similar to WT. There were also no detectable defects in budding, production of virions with normal looking cores, or packaging the trpRS protein. In summary, whereas the T214 residue is critical for viral infectivity, we are currently unable to confirm at what step the mutants inhibit replication. Preliminary data suggests that there was a defect in the stability in the mature core which could perturb normal reverse transcription and/or integration.

RESULTS AND DISCUSSION

Virus Spread and Particle Release

Viruses containing the T214A or T214D mutation in CA were transfected into QT6 cells.

Twenty-four hours later medium from these cells were placed onto uninfected DF1 cells. Spread of RSV throughout the cell culture was monitored by spread of virus-encoded GFP (271). WT virus infects the entire monolayer within 5-7 days, whereas DF1 cells showed no signs of infection of either the T214A or T214D virus after 21 days in culture (Figure 5.3A). QT6 cells transfected with either mutant virus genome produced Gag indistinguishably from WT. The Gag proteins of T214A and T214D were properly cleaved by the virus PR and found in the medium with equal efficiency as WT, suggestive of properly budded virus (Figure 5.3B).

However, the particle release assay did not address the quality of the virus particles released into the medium. Thin section electron microscopy of each of the mutant viruses was

120

Figure 5.3. Infectivity and release of virus particles. A) Plasmids containing the WT RS.V8 genome or mutations at CA residue T214 were transfected into QT6 cells. Medium was transferred to DF1 cells 24 hours later. Virus infectivity was monitored by spread of virus- encoded GFP through the culture. B) Virus release was measured by labeling virus protein with

[35]S-Met/Cys for 15 min (lysate) or 4 hours (medium). Gag and CA protein were immunoprecipitated and analyzed by autoradiography. The ratio of CA / Gag was compared to determine the efficiency of particle release.

121 used to assess whether or each could form a normal looking core. Each of the viral genomes was transfected into QT6 cells. Transfection efficiency was confirmed by visualization of GFP fluorescence and the resulting cells were fixed, sectioned, and stained with uranyl acetate (Figure

5.4). WT virus produces virions that range from 100 to 120 nm in diameter and have a dark usually centralized density representative of the genomic RNA condensed with the reverse transcription complex (180). Upon inspection, cells transfected with both T214A and T214D produced extracellular particles that were indistinguishable from the WT virus.

Although we did not directly test the ability of each mutant to form immature particles through in vitro assembly of Gag, it is likely that both could produce normal immature structures.

In general mutations that hamper assembly of the immature particle will have a defect in protein production, particle release, or show signs of abnormal virions by thin section microscopy. Those mutations that only inhibit maturation of the capsid often look like normal WT particles in thin section images (59). Since all of these functions appear similar to WT, these data suggested that the defect in infectivity was not due to a failure of budding, protein synthesis, or immature particle assembly.

In Vitro Assembly of Particles

Assembly of a fully functional mature capsid is critical to the infectivity of retroviruses.

Since the T214 residue falls within the three-fold interface created as a result of maturation, it remained possible that the mutations were affecting the formation of the capsid. Purified RSV

WT CA protein assembles into small spheres of about 17 nm in diameter in the presence of divalent salts, such as sodium phosphate (250,266). The reaction can be monitored by measuring the optical density of the solution at a wavelength of 450 nm (Figure 5.5A). There is a delay in

WT turbidity of up to an hour prior to a sigmoidal increase in the optical density, suggestive of nucleation-driven assembly (266). When sodium phosphate was added to T214A or T214D CA protein, no detectible turbidity was seen. However, when the products were stained and examined

122

Figure 5.4. Morphology of released virus particles. QT6 cells transfected with plasmids containing virus genome were fixed, sectioned, stained with uranyl acetate, and examined by transmission electron microscopy.

123

A1.0

0.8 WT CA

0.6 T214A CA 450 T214D CA OD 0.4

0.2

0.0 0 100 200 300 Time (Min)

B T214A T214D

Figure 5.5. In vitro assembly of CA protein. A) Purified WT, T214A, and T214D protein was assembled in vitro with 500 mM sodium phosphate. Turbidity of the solution was monitored at a wavelength of 450 nm. B) Resulting particles were absorbed onto formvar and carbon-coated copper grids, stained with 1 % uranyl acetate, and visualized by electron microscopy. Bar = 100 nm.

124 by transmission electron microscopy, small spheres indistinguishable from the WT protein were seen (Figure 5.5B). Consistently fewer spheres per grid were seen with the T214D protein than either the WT or T214A protein, indicating that the T214D protein may be more reluctant to assemble. It should be noted, though, that meaningful quantitation of the number of assembled particles is difficult to interpret due to the many variables that affect how well the particles stick to the grids and are stained. Regardless, neither mutation at T214 prevents mature capsid assembly in vitro suggesting that these mutations do not affect intermolecular interactions within the CA protein that lead to assembly of the either the immature or mature capsid.

Stability of Virus Cores

Matured retrovirus cores are noticeably less stable than the related immature Gag particle. Optimal stability of the capsid is necessary to allow the RTC access to nucleic acids for reverse transcription and the reverse transcribed provirus access the host genome for integration. Reduction in stability during maturation likely enables the capsid to begin disassembling after entry into a new cell prior to reverse transcription

(59,91,176,315). The sensitivity of the viruses containing the T214 mutations was tested by pelleting the virus through sucrose in the presence or absence of the non-ionic detergent triton X-100. Cores resistant to detergent will have CA in the pellet fraction whereas detergent-sensitive cores will break up during the spin and the CA protein will appear in the supernatant.

In the absence of triton X-100, WT cores are almost exclusively found in the pellet fraction (Figure 5.6). Similar to previously published results, 46 % of the CA protein for WT virus was resistant to detergent (180). Preliminary data collected on the

T214A virus, indicated that the core was less stable than the WT core. There was a marked increase of CA protein in the supernatant fraction even without detergent.

125

Figure 5.6. Virus core stability. QT6 cells were transfected with plasmids containing virus genome. Virus-encoded protein was labeled for 24 hours with [35]S-Met/Cys. Medium was spun through a sucrose cushion in the presence (+) or absence (-) of triton X-100. Pellet (P) and supernatant (S) fractions were run on a 15 % SDS-PAGE gel and autoradiography was analyzed.

126

Additionally, there was very little CA in the pellet fraction when spun through triton X-

100. Cores from the T214D virus appeared to be slightly more stable than those of WT.

When pelleted through triton X-100, there was a greater percentage of CA protein in the pellet fraction than was seen with the WT virus.

These data suggest that there is a potential defect in mature capsid stability. The less stable T214A core could cause the RTC to dissolve too soon, exposing the viral genome to cellular pathogen receptors, such as toll-like receptor-7 (TLR-7). A more stable core as seen in the T214D virus could result in a block of reverse transcription or integration if nucleic acids or the RTC are unable to penetrate the capsid. Further analysis of core stability is needed to determine if this is a reproducible defect in the virus.

Packaging of TrpRS

Viruses require packaging of host cellular tRNA to initiate reverse transcription of the genome. Packaging of tRNA is mediated by the tRNA-RS, trpRS in the case of RSV and lysRS for HIV-1 (314). LysRS is packaged with Gag by binding to α11 of the CTD (67,313). Mutation of HIV-1 T210A reduced binding of the CTD and lysRS three-fold, but the in vivo effect is currently unknown (67). Considering T214 mutations ablate RSV infectivity, we examined if

T214A and T214D mutations would prevent packaging of the trpRS in virions. Medium from viral genome-transfected QT6 cells was pelleted on a sucrose cushion and western blot was performed for the chicken trpRS protein (Figure 5.7). Both viruses containing the T214 mutations packaged trpRS like the WT virus. Virus with the A154D mutation was also tested as an additional control. This virus also prevented infectivity but is found in the CTD 310 helix and not involved in trpRS binding. As expected, A154D had no effect on the packaging of trpRS. These data suggest that the defect in infectivity was not due to a failure in packaging the trpRS protein.

Although there was no difference seen in protein packaging, it is still possible that the T214 mutations disrupt the normal binding of the CTD and trpRS in a way that prevents or diminishes

127

Figure 5.7. Packaging of cellular trpRS. QT6 cells were transfected with WT, T214A, T214D, or

A154D proviral genome. After 24 hours of virus production, medium was collected and was spun through a sucrose cushion. Pellets were resuspended in sample buffer and separated by 15 %

SDS-PAGE. Western blotting was performed for the chicken trpRS protein.

128 packaging of the viral genome. Successful genome packaging and reverse transcription have not yet been tested.

Overall Impressions of T214

The experiments undertaken in this study sought to identify the potential of the T214 residue as a phosphorylation site. Based on the results provided, it is unlikely that this residue is phosphorylated at any stage of the virus life cycle, though the possibility cannot be completely ruled out. Viruses with either the T214A or T214D mutation had nearly identical phenotypes – neither was infectious, yet produced normal looking virus particles and could assemble mature capsids in vitro. The only appreciable difference between the two viruses is the stability of the mature cores. The T214A mutant was less stable than WT whereas the T214D virus was more stable. The ultimate consequence of these two opposing results is the same, but the reason behind each is yet to be elucidated. It is clear that the T214 residue is critical to the virus, likely by effecting the molecular interactions of the mature capsid. Further analysis of this residue is needed to identify its exact function.

ACKNOWLEDGEMENTS

We would like to thank Roland Myers for processing the thin sections presented here and the imaging core for their support with EM. This work was supported by NIH grants R01 CA100322 (R.C.C.) and T32 CA060395 (M.R.E.) and by Tobacco CURE funds from the Pennsylvania Department of Health.

129

CHAPTER VI

DISCUSSION: IMPORTANCE OF THE C-TERMINAL DOMAIN DURING

MATURATION

130

Retroviruses undergo an obligate step of maturation to produce infectious virions. This process is initiated during or shortly after budding begins. Maturation produces major changes in the virus. Among them are the condensation of the dimeric viral RNA genome into a more stable state, activation of the RT and IN proteins, and activation of the Env proteins for fusion. The hallmark of maturation is the dramatic reorganization of the internal structure of the virion

(176,316,317). Initial contacts that bind Gag molecules in the immature particles are lost during sequential proteolytic cleavage of Gag. The newly released CA protein then generates new contacts establishing the mature capsid around the RTC. A critical need in understanding retrovirus maturation is identifying the underlying mechanisms. The experiments presented in the previous chapters probe regions of the CA CTD that are key contributors to assembling the immature and mature particles. The results identify residues in the CTD that participate in interactions required for assembly and/or maturation.

SUMMARY OF DATA CHAPTERS

Most of our knowledge of the assembly and structure of retroviral particles stems from in vitro assembly assays and mutagenesis. These studies identified a number of interfaces critical to establishing complexes (32,37,38,58,62,64,65,69,241,271,299). Lower resolution imaging of native mature capsids provides evidence that many of the contacts seen in vitro are representative of the in vivo interactions (25,36,185,233). However, much less was known about interactions within the immature particle. Recent subtomogram averaging of structures that mimic those in immature particles provided the first higher resolution look at the intermolecular interactions

(186,187). Based on these models, we tested a variety of residues for their effect on the assembly of the immature particle and subsequent maturation (Figure 3.1). In doing so, we provided the first direct biological support for the interpretations of the Bharat immature CTD dimer model.

These results also provide support for the role of hydrophobic residues in establishing the mature

CTD dimer structure.

131

All of the hydrophobic residues chosen within the 310 helix and α9 showed some level of perturbation in mature capsid assembly when mutations were introduced into the CA protein

(Figure 3.6). Because these residues were expected to be in the mature dimer interface, this result was not surprising. However, we also found that many of these same residues also disrupted the ability of the immature particle to assemble (Figure 3.4). In particular mutations in W153, L180, or V188 interfered with assembly of both particles. In contrast A154 and Q195A interfered with assembly of mature capsids probably by disrupting hydrogen bonds.

A more recently described CTD-CTD mature interface maps to the three-fold axis and is largely influenced by residues at the top of α11. The threonine located at that spot is conserved among a variety of retroviruses pointing to a potential critical role for the virus (Figure 5.1 and

5.2). Also, the amino acid sequence context of this threonine is suggestive of a possible phophorylation site. The work presented in chapter V suggests that this residue in RSV (T214) is important for virus infectivity, but phosphorylation or dephosphorylation mimetic mutations did not have a discernible effect on assembly of either immature or mature particles. The only key difference between the T214 mutants and WT comes from preliminary experiments on core stability in which T214A is less stable than WT whereas T214D is more stable (Figure 5.6).

These results support a role for this loop in formation of the three-fold mature interface. Further studies are needed to confirm and extend these results.

The SP region of the Gag molecule is a critical determinant of immature particle size.

Initial cleavages of Gag produce an intermediate species of the capsid protein that retains SP

(180,199). The CA-SP species is a product, at least in part, of slower cleavage kinetics due to steric hindrance of PR activity at the CA and SP boundary until after the SP-NC cleavage event.

The CA-SP protein can be detected up to several hours after budding, leading to the possibility that SP has a function in successful maturation of the capsid (260). A question remained as to whether CA-SP could form the same types of interactions as CA or if CA-SP would assemble more Gag-like interactions. We addressed this through genetic and biochemical characterization

132 of purified CA-SP. The CA-SP assembled with faster kinetics than CA and could cause an excess of CA to assemble more rapidly (Figure 4.4 and 4.5). Despite the altered kinetics CA-SP formed the same classes of particles as CA. Higher resolution cryoEM imaging demonstrated that the assembly of CA and CA-SP yielded nearly indistinguishable structures with the exception of slight changes in the flexible loop region of the NTD and the very C-terminus of the CA protein proximal to the SP (Figure 4.7 and 4.8). CA-SP tolerated most alanine and proline mutations

(Figure 4.11). Addition of the first 8 amino acids from NC (CA-SP-8NC) resulted in assembly of a new class of particles that were larger and more angular (Figure 4.10). These results indicated that SP did not provide an inherent block to mature CA assembly, and that the CA-SP intermediate may initiate interactions of mature capsids.

NMR analysis of the dynamics of the SP point to a potential interaction with regions of the CTD critical for the formation of the dimerization interface (Figure 4.12). From all of these results, we conclude that the CA-SP intermediate is not merely a bystander of maturation and byproduct of slow cleavage kinetics. Rather SP is dynamic and is likely to form transient interactions with the MHR and α9 helix that could dictate how these regions participate in the dimer interface.

ASSEMBLY OF IMMATURE PARTICLES

Influence of the Dimerization Interface on Immature Assembly

The results described in this dissertation provide critical support for the current model of immature retrovirus particles as it pertains to the formation of the interhexameric CTD dimer.

This model predicts that the α9 helices of the two CTD molecules cross each other at angles different from those in the mature dimer. During maturation, the two CTD molecules rotate approximately 120-140 ° relative to each other along the plane of the dimer interface (186,187), so that the interhexameric dimer interface is quite distinct from that of the mature dimer.

Comparison of the Bharat immature dimers to crystal structures of the mature dimer reveals residues that likely retain the same intermolecular interaction partner in both (186,187). The so

133 called pivot point of rotation was identified in HIV-1 to be W184 and M185. The W184 and

M185 residues of one monomer appear to form an interaction with W184 and M185 from the other molecule in both dimers, although the angle of interaction is different for each dimer.

Mutation of each of these residues to alanine disrupts assembly of Gag and CA in vitro and allows for the isolation of pentamers and hexamers of HIV-1 CA dependent on mutations in the

NTD (189,231). Additionally these mutations cause a decrease in Gag assembly as measured by western blot (37). Superimposing the RSV CTD (PDB: 3G21) on the models of immature HIV-1

(4D1K) or M-PMV (4ARD) and comparing the structure to that of 3G21 allowed us to predict that V188 serves as the pivot point in RSV (38). Substitution of alanine at V188 renders the virus non-infectious. Gag and CA-containing particles are released into the medium at a significantly lower rate than WT virus and the Gag protein fails to assemble into immature particles in vitro.

All of these results are similar to those seen with the W184A/M185A mutations of HIV-1 indicating that V188 is essential to assembly of both dimer interfaces. The potential remains that

V188 is the pivot point in the RSV dimer interface (i.e. two V188 residues remain in contact with each other in both dimers with just a change in angle between them) similar to W184 and M185 in HIV-1.

Based on the dimer models we can predict that further stability of the interaction made by

V188 is provided by L180 and W153. Alanine substitutions at either L180 or W153 caused widespread reduction in particle production and completely abolished virus infectivity (Figure 3.2 and

3.3). The immature protein shell also did not assemble in vitro (Figure 4.4). From modeling the

RSV CTD into the Bharat immature dimer (4D1K/4ARD), we predict that W153 helps influence the intermolecular interface by through hydrophobic interactions with L180 of the second CTD molecule. In contrast, in the mature dimer W153 helps to establish the hydrophobic core of a single CTD molecule. Therefore the relationship between W153 and L180 changes during maturation and is a signature of the different states of the CTD dimer of the immature and mature particles.

134

The distinction between the mature and immature dimer interfaces is clearest with respect to the location of residues A154 and Q195. We provided evidence for a previously undiscovered interaction between the A154 and Q195 residues in the mature dimer interface (see below and

Chapter III). Based on the Bharat dimer model, this interaction is absent in the immature dimer because of the large rotation of the molecules relative to each other (Figure 3.1). The rotation separates A154 and Q195 from any predicted interface. Alanine substitutions at both residues did not prevent the assembly of the immature Gag lattice (Chapter III). Thus, the association of A154 and Q195 appears to be a hallmark of mature dimerization leading to assembly of the mature capsid.

Models of Immature Assembly

Initial multimerization of Gag is thought to occur early after synthesis of the protein, although the location where these interactions initiate is still under debate. Gag dimers associated with RNA are found in the cytoplasm; however, multimerization beyond dimers does not occur to any significant level until localizing to the plasma membrane (318,319). Step-wise growth of the particle than occurs until the immature virion buds from the cell. The mechanism behind this process is not entirely clear but three of the most likely pathways are described.

In the first, Gag molecules dimerize and three dimers come together to create a hexamer.

Hexamers would then serve as the basic assembly unit of the immature particle (Figure 6.1A).

Based on this model, the first dimer interaction to occur would be the intrahexameric (creating hexamers) CTD-CTD interaction. After completion of the hexamer, interhexameric dimerization

(between neighboring hexamers) would occur, building the particle. In the second model, a trimer of dimers also creates a 6-mer of Gag molecules; however, this interaction does not directly create a hexamer. Instead the 6-mer generates the three-fold axis between three hexamers (Figure

6.1B) which initiates particle assembly. Completion of the immature particle can be achieved by addition of more 6-mers. This model is particularly attractive because it quickly generates both interhexameric and intrahexameric dimers as well as all three interfaces – two-fold, three-fold,

135

Figure 6.1. Assembly models of the immature lattice. A) Monomeric Gag molecules dimerize in the cytoplasm shortly after synthesis. A trimer of dimer then associates in dimers to form hexamers of Gag protein. The hexamers serve as the units that create the immature lattice. Green- green and blue-blue interactions represent interhexameric dimers. Blue-green interactions represent intrahexameric interactions. B) A trimer of dimers forms to create a 6-mer of Gag proteins in which the common point of contact forms the three-fold axis between NTD molecules.

Red-red and blue-blue interactions represent interhexameric dimers. Blue-red interactions represent intrahexameric interactions. C) Gag monomers dimerize quickly in the cytoplasm. From there, the dimers are able to form tetramers of Gag made of both inter- and intrahexameric interactions. The tetramers serve as the nucleation point for immature lattice formation to which dimers and tetramers can be added. Red-red and blue-blue interactions represent interhexameric dimers. Blue-red interactions represent intrahexameric interactions.

136

137

and five-fold – an arrangement that could lead quickly to very stable intermediates. In the third model, there is dimer/tetramer equilibrium of Gag molecules that forms early in the assembly pathway. The tetramer serves as the nucleation substrate for particle assembly and dimers and tetramers of Gag are added to complete the immature particle. This scheme also generates both interhexameric and intrahexameric dimers early in the assembly process (Figure 6.1C).

The first pathway is common to many viruses that produce icosahedral capsids, including the picornaviruses. In these viruses, 12 individual pentamers are created and then combined to create the provirus capsid. If this same process were true for retroviruses, then Gag molecules would need to first form hexamers and then assemble them at the plasma membrane. Several lines of evidence suggest that this pathway is not the mechanism for immature capsid assembly. First, the geometric architecture makes the hexameric assembly unlikely. To generate a sphere out of hexamers there must be gaps (179,181,186,320). Simply aligning hexamers next to each other would likely produce a planar Gag lattice without some sort of outside component to generate curvature, such as the host cell factors at the plasma membrane. In the absence of any other outside component, spherical VLPs consisting of Gag and oligonucleotides can be generated in vitro (39,79,284,321). This would favor models that do not rely on the formation of hexamers as building blocks.

Second, the recent cryoEM/cryoET models generated for CA-NC assemblies predict a large interhexameric interface relative to the intrahexameric interface (186,187). The interhexameric interface is made of residues from the NTD as well as the dimerization interaction in the CTD (186,187). In RSV, mutations to residues that are predicted to make up the immature dimer interface prevent in vitro assembly of the Gag molecules into VLPs (Chapter III). These same mutations also cause the reduction of particle release in vivo consistent with a function early in Gag-Gag associations.

138

Finally, results with HIV Gag also argue against the first model. Discrete steps in HIV-1

Gag assembly have been identified by gradient centrifugation (322). Mutants of Gag, previously shown to disrupt Gag assembly in vitro or I vivo, were selected based on their location in the intra- or interhexameric interface as predicted by the Bharat model for immature dimers.

Complexes early in Gag assembly are isolated when mutations that interfere with interhexameric interactions are introduced into Gag whereas mutations affecting the intrahexameric interface allow formation of complexes consistent with later stages of assembly (323,324). Therefore the second or third pathway is the more likely mechanism for immature particle assembly as these rely on early interhexameric interactions between neighboring hexamers.

Further investigation into the mechanism of immature particle assembly is currently underway using anisotropy measurements of RSV and HIV-1 Gag molecules in the presence of

RNA or DNA oligonucleotides (J. Flanagan, personal communication). When low concentrations of Gag and short, fluorescently labeled oligonucleotides are mixed together, an accumulation of

Gag dimers and tetramers, and possibly hexamers accumulate before the larger complexes form.

Unfortunately the anisotropy signals for Gag tetramers, hexamers, and larger complexes all are near the detection limit for the system, making interpretation of the results difficult. The mutant proteins described in this dissertation are currently being examined by anisotropy. Ideally, one or more of these mutants will be able to discriminate between the three models of assembly by preferentially creating small oligomers. These complexes will then be analyzed by cryoEM and small angle x-ray scattering to determine their structure and interactions.

INTERACTIONS LEADING TO MATURE CAPSID ASSEMBLY

The Role of SP

Up to now, little was known about the role of SP during maturation. In vitro assembly of bacterially expressed and purified CA-SP demonstrated the protein is intrinsically assembly- competent. The rate of phosphate-induced assembly of CA-SP was much greater than CA previously described (Figure 4.4). Like CA, CA-SP assembles into spheres with T = 1 and T = 3

139 icosahedral symmetry (Figure 4.4 and 4.6). Reconstruction of CA-SP particles by cryoEM confirms that the two proteins are nearly indistinguishable save for a few amino acids in the NTD flexible loops and the C-terminus of the protein (Figure 4.7 and 4.8). NMR spectra of the CTD and CTD-SP were nearly identical confirming the similarity between CA and CA-SP consistent with previous findings (57). Furthermore, in mixed protein species reactions, CA could be induced to assemble by small amounts of CA-SP, indicating that CA-SP can nucleate CA assembly in vitro.

Our observation that CA-SP is competent for assembly is at apparent odds with in vivo structural analyses in which proteolytically blocked Gag shells are unable to complete maturation.

When HIV-1 infected cells are treated with maturation inhibitors that block the final cleavage between CA and SP1, the resulting capsid has a hybrid mature and immature structure (74,75). In

Gag, SP forms an alpha-helix along with the C-terminus of the CA CTD and the N-terminus of

NC. This region assembles into a 6-helix bundle that helps generate Gag hexamers in the immature particle. PR initially cleaves in the middle of the helix so SP and last few residues of

CA may still be involved in contacts that are normally rapidly dissociated. The retention of some of these Gag interactions in chemically treated particles could explain the failure of the particles to complete maturation. In the absence of chemical treatment, this region of Gag disassembles during the PR cleavage events, allowing intermediate stages of assembly to occur.

Several CA mutants have defects in assembly of mature capsids. The best characterized of these is the MHR mutant F167Y (59,271). The defect can be relieved by spontaneous secondary mutations arising elsewhere in CA. These suppressor mutations are found primarily in the CTD but have also been discovered in the NTD and SP (62,250). Whereas the F167Y mutant prevents dimerization of the CTD, the suppressor mutations reestablish the dimer contacts, at least for those residues tested (60). However, suppression of the lethal MHR mutation F167Y by the SP suppressor S241L could not be replicated in the in vitro assembly assay (Figure 4.8).

There are two possible explanations for this result. First, suppression of F167Y by S241L in vivo

140 could depend on the SP to retain some of the helical arrangement found shortly after initial Gag cleavage as described above. The substrate in the in vitro assembly assay does not contain this structure and therefore does not necessarily replicate the relationship between S241L and F167Y.

The second explanation considers the kinetics of SP cleavage. Cleavage of the S241L mutant protein to produce the mature CA and CA-S proteins in vivo is very rapid (62). It is possible that the suppression of F167Y by S241L is due to the increased kinetics of final CA cleavage conferred by the S241L mutation. Though CA (F167Y) cannot assemble mature contacts in vitro, it is possible that faster removal of SP in vivo, allows the CA protein to adopt a slightly different conformation than is seen in the WT. This small change may be amenable to mature capsid formation and thus restore some of the infectivity to the double mutant virus which is easily replicated in vitro. Again, the substrate for the in vitro assembly system does not undergo processing. Support for processing-dependent maturation comes from isolation of a suppressor for a different MHR mutant. Infectivity of CA mutant E162D is restored by a PR mutant (R28H) that has a slightly slower rate of final cleavage of SP (325). Thus, the dynamics of protein processing and maturation appear to be tightly linked.

The CTD Dimer Interface

The purified RSV CTD is monomeric at neutral pH conditions but dimerization is induced by low pH. X-ray crystallography of the acid-dependent dimer at sub-angstrom level resolution provides an explanation for the low pH triggered dimerzation. D179 of one monomer forms a hydrogen bond with the protonated D191 of the other monomer (38). This interaction enables the domains to dimerize. The dimer interface of RSV and HIV-1 is studded with hydrophobic residues in the 310 and α9 helices that likely further stabilize the proton-linked dimerization of the CTD (38).

In RSV, residue A184 of α9 in the dimer interface forms the closest contact between molecules. A cysteine substitution for this residue triggers spontaneous dimerization in the absence of reducing agent (38). However, further anaylsis of the contribution of specific

141 hydrophobic residues in the 310 and α9 helices at the mature dimer interface had not been done previously. Using PISA analysis, we identified several candidate residues in the low pH dimer interface (Figure 3.1). Like A184, L180 and V188 are predicted to associate closely with their matching residues in the second molecule. The hydrophobic pocket created by L180, A184, and

V188 at the top of the dimer is likely further stabilized by the presence of the W153 residue.

Alanine substitutions for W153, L180, and V188 disrupted dimerization of the CTD and blocked assembly of the CA protein in vitro, confirming the contribution of these residues to the formation of the mature dimer (Figure 3.5).

Another critical feature of the location of these hydrophobic residues, L180 in particular, is their relationship to the tripartite interface, an intrahexameric, three-domain interaction between the NTD and CTD of one CA molecule and the NTD of its neighbor. Association of the tripartite interface is governed by positively charged residues at the top of the CTD and bottom of the NTD (194,241,266). Sodium phosphate is thought to induce assembly of the RSV CA protein in vitro by shielding these positive charges. The position of the moderately bulky hydrophobic residue L180 near the tripartite interface likely contributes to the formation of the interface by influencing the packing of neighboring residues. In this way the protein is able to couple formation of the two interfaces, both of which are implicated in nucleation of capsid assembly.

A154 and Q195 are also contributing residues to the formation of the dimer toward the bottom of the interface. Substitution of the A154 to aspartate destroyed dimerization and assembly. A more conserved mutation to valine also failed to dimerize and assemble, whereas substitution to glycine was fully functional (Figure 3.5). All three mutant viruses produced virion in the cytoplasm with normal cores indicating that there were likely no issues with the virus until maturation (Figure 3.2). CTD and CA with the Q195A substitutions were able to dimerize and assemble, respectively. However, the Kd of the dimer was 4-fold weaker than that of WT and the full-length protein assembled primarily into tubes (Figure 3.4 and 3.5). Additionally, mature cores produced from the Q195A virus were far less stable than those from the WT virus (Figure 3.8).

142

Proper assembly of the bottom of the dimer requires specific spacing between the 310 helix and

α9. The spacing of A154 and Q195 in the dimer structure coupled with the results of our mutagenesis suggests that the two residues are hydrogen bonded to each other. Although we could not mimic this bond via electrostatic interactions, it is impossible to determine what adding charges at both residues would do to the overall structure of the CTD (Figure 3.7).

Importance of CTD Dimerization

In vitro assembly of RSV CA is a nucleation-driven process that is stimulated by divalent anions. Addition of dimer or other small oligomeric complexes of RSV CA stimulates rapid assembly of a monomeric fraction of the protein eliminating the lag phase of assembly typically observed with monomers alone (241). Based on mutagenesis, these data suggest that a dimeric unit or some other small oligomer of CA is an important step toward nucleating capsid assembly.

The dimer interfaces involved in this nucleation event are most likely to be intrahexameric NTD-

CTD and interhexameric CTD-CTD (60,241,250,271). The data presented in this dissertation provide further support for the importance of the CTD-CTD interaction as part of the nucleating complex. The Q195A mutant primarily forms tubes instead of spheres in vitro indicative of altered packing of CA molecules. Whereas larger structural perturbations cannot be excluded, the most likely explanation for this phenotype is altered CTD dimerzation. The Q195A CTD dimer interaction was 4-fold weaker than that of the WT protein. The Q195A dimer may be less stable than WT, have an altered structure, or both. None of these possibilities can currently be ruled out.

Either way, the dimer interface is at the very least, a contributor to the nucleating complex for the in vitro assembly of CA.

Although the CTD-CTD dimer interface in mature particles appears to be restricted to the

310 helix and α9, regulation of dimerization itself depends on other CTD elements. A154 mutants have phenotypes similar to several MHR mutants. Mutants D155Y, neighboring A154, and

F167Y produce normal looking particles in thin sections but prevent mature CA assembly by disrupting the ability of the CTD to dimerize (60,63,250,271). Defects caused by D155Y and

143

F167Y are relieved by a secondary substitution in α9, R185W. The suppressor mutation restores low pH dimerization of the CTD and promotes assembly of full-length CA in vitro

(60,61,250,271). Given the similarity between the phenotypes of the mutants studied in this dissertation and those of D155Y and F167Y, we predict that R185W can correct the assembly and dimerization defect of the A154 mutants. If so, it would provide strong evidence for a cooperative role of the 310 helix, MHR, and the dimerization helix in dimerization of the CTD.

Participation of the MHR in formation of the dimer interface may also involve SP. High resolution imaging of CA-SP(SP1) and other truncated Gag proteins are unable to resolve the SP region due to its dynamic nature (40). Preliminary PRE NMR studies reveal a potential transient interaction between the SP and residues in the hydrophobic pocket created by residues of the dimerization helix and MHR (Figure 4.12). These data provide support for structural studies of

CA-SP which was noted for a potential interaction in nested capsids. The T = 1 and T = 3 nested particles were not oriented in register, but the SP region of one appeared near the MHR of the other. The interaction between SP, α9, and MHR may cause a conformational change in the dimerization interface that serves as the starting point for mature CTD-CTD interactions and ultimately assembly of the mature capsid. Support for this model comes from the discovery of escape mutations in the MHR that provide resistance to drugs that act as maturation inhibitors. An interaction between SP and the MHR could help orient the 310 helix, MHR, and α9 with each other in a way that promotes dimerization of the CTD.

The Three-Fold Interface

Early structural studies on HIV-1 assemblies were performed on planar structures. When curved particles were analyzed, a new CTD-CTD interaction was discovered. This three-fold interface is made by association of the CTD from three different hexamers and is one of several interfaces that form upon maturation (244,278). Residues in α11 mediate the interaction. The last helix is also important for packaging of the tRNA synthetase that helps incorporate the tRNA that primes reverse transcription (67,313). Also found in this region is a putative phosphorylation site

144 at a conserved threonine in the loop between α10 and α11 (Figure 5.2). As post-translational modifications can be sources of assembly and maturation regulation, we investigated whether

T214 in RSV could serve as a phosphorylation site. Viruses with the T214A or T214D mutation were non-infectious but produced normal looking virus particles (Figure 5.3 and 5.4). Therefore early stages of virus production probably do not depend on T214 as suggest by the structural models. CA protein with either T214 mutation assembles mature capsids in vitro. However, in preliminary experiments, cores made from the T214A mutant are less stable than WT whereas the

T214D cores are more stable suggesting an influence on CA-CA association (Figure 5.5 and 5.6).

It is clear that the T214 residue is critical to the virus though its exact function cannot be determined from the experiments described in this dissertation. The residue probably effects the molecular interactions of the mature capsid at the three-fold interface. The ability of these mutants to form functional reverse transcription initiation complexes should also be examined.

Maturation of the CTD

Assembly of CA Q195A demonstrates that mutations in α9 result in alterated assembly products probably by perturbing the dimerization interface. Extended to the WT protein, the dynamic interaction between the MHR and SP help to regulate the formation of the dimerization interface. Small perturbations in the conformation of the 310 helix and α9, lead to changes in how the dimer forms affecting the assembly of the mature capsid (Figure 6.2). The dimer interaction is further stabilized and the conformation controlled by the assembly of the three-fold axis. The precision of the all these interactions likely leads to regulating whether pentamer or hexamer form. This regulation may extend to determining where in the capsid a particular hexamer or pentamer is located (i.e. in the wider end or narrower end).

145

Figure 6.2. Assembly model for the mature dimerization interface. Immature particles are made of the polyprotein Gag. Interhexameric interactions are mediated by the CTD of one hexamer

(purple) with the CTD of the neighboring hexamer (black) via the α9 helix (center of diagram).

Hexamers are created by CTD associations (not depicted) and the SP (red) 6-helix bundle (not depicted). Cleavage upstream of CA and between the SP and NC (green) releases the CA-SP protein which can begin to develop mature CTD contacts due to an interaction of the SP with the

MHR and α9. Further cleavage of SP leaves the mature capsid protein to build the capsid.

146

The Role of the NTD

Formation of the β-hairpin

Formation of the β- hairpin is critical for maturation of several retroviruses, and assembly of RSV CA into capsid-like structures is dependent on the presence of the β-hairpin, as substitution of D52A ablates normal particle assembly (37,250). Though the rate of protein association is accelerated, the interactions fail to create organized particles. Hydrogen/deuterium exchange on assembled HIV-1 protein complexes thought to represent various stages of maturation led to the prediction that the β-hairpin does not form until after the SP is removed

(220,251). Whereas we can neither support nor refute this prediction as it pertains to viruses made in cells, CA-SP assembles in vitro with the β-hairpin present (Figure 4.4). Like CA, assembly of the CA-SP in vitro is dependent on the presence of the β-hairpin at the N-terminus as D52A ablates assembly of both proteins in vitro (250). Addition of the first 8 residues of NC relieves this dependence as D52A CA-SP-8NC assembles (Figure 4.10). Therefore it is likely, at least for

RSV, that cleavage upstream and downstream of CA-SP induces the formation of the β-hairpin.

The formation of the β-hairpin causes transmission of conformational signals for mature assembly down the length of α1. This signal can then be transmitted to the CTD through an interdomain interaction involving α1 and α11, which is proximal to SP, promoting mature assembly (296).

Necessity of the Flexible Loops

The flexible loop region of the NTD is known to be a binding partner of several host restriction factors (158,168,169,312,326-328). In terms of structural influence of the capsid itself, mutations in the flexible loop are able to rescue infectivity-defective viruses with mutations in the

β-hairpin (221). Further analysis of residues in the flexible loop revealed that viruses with D87E mutations had reduced infectivity capacity that could be overcome by the second-site mutation

A134V. The D87E substitution does not appear to cause the inhibition of immature particle production. It also does not appear to inhibit assembly of the CA protein in vitro (Figure A.3).

147

Therefore the defect seen must be a result of an interaction that is not replicated in vitro. There are two explanations for the in vivo effect. First, whereas intrinsically the protein is able to assemble both types of particles, the subtle effect of D87E may prevent the dynamic reorganization of the particle in some way that perturbs maturation in vivo. Second, a currently unidentified avian protein needed for virus infectivity, similar to CypA for HIV-1, is unable to bind to the capsid with the D87E mutation. Neither of these possibilities can be favored or refuted currently.

De Novo Assembly of the Mature Capsid?

Questions about the sequence of events in maturation remain. How does the mature capsid form? There are two prevailing thoughts on the subject. First, based on studies of other viruses, mature capsid assembly could result from condensation, or slight conformational changes, of the immature lattice. However, there is little experimental evidence that supports this model of capsid maturation.

The second mechanism involves disassembly of the immature lattice and de novo assembly of the mature capsid. Several lines of evidence support this model for capsid assembly.

First, the monomeric CA protein is able to assemble in vitro in a nucleation-driven reaction. Self- association of purified CA does not require any other viral component

(44,193,241,250,266,329,330). Second, the mature capsid is made of a hexameric lattice punctuated by 12 pentamers, whereas no pentamers are found in the immature lattice

(25,179,181,182,184,185,320,331). Assembly of pentamers is most easily explained by the breakdown of the Gag lattice and reassembly of subunits into the mature capsid. Third, not all of the CA protein available from Gag is used to make the mature capsid. Estimates of up to half of

CA found within the virion fails to assemble into the capsid (25,179,233,332). Finally, nested capsids and multilayered shells are observed in native virions. This feature does not seem compatible with condensation of the immature lattice. Instead de novo assembly could result in

148 more than one nucleation event potentially producing multiple capsids within a single viral envelope or capsids forming around one another (25,233,290,333).

Although the results presented here do not directly address the question of de novo assembly of the mature capsid, the findings are consistent with this process. The fact that the CA protein is able to assemble into particles with interactions resembling the WT virus in the absence of other domains of Gag, suggest that de novo assembly is at least a possibility (241,266). The biggest drawback of the de novo assembly model is the energetic waste associated with disassembling a lattice only to rebuild another. However, the assumption has generally been that the immature lattice disassociates to monomers. This is not necessarily, and almost certainly not, true. Instead, it is likely that the substrate for de novo assembly is oligomers of CA-SP or CA.

The CTD-CTD dimerization interface is shared between the lattices of immature and mature particles, though the orientation of the proteins is different. Therefore it is likely that the cleavage of CA-SP from Gag triggers the conformational rotation of the two CTDs. This organization breaks the interhexameric association of the NTD and destabilizes the intrahexameric CTD and

SP interactions. The dimers of CA or CA-SP are then left to nucleate assembly of the mature capsid.

In vitro assembly of the CA protein provides further support for need of a dimer to nucleate assembly of the mature capsid. Mutations in the MHR that disrupt dimerization of the

CTD, prevent monomers of the protein from forming organized particles in vitro

(60,241,250,266). Reestablishing dimerization by a second-site suppressor mutation restores assembly of the previously assembly-incompetent mutant protein (60,61,241,250,271). The critical need for CA dimers for de novo assembly of the mature capsid is further supported from some of the CTD dimer interface mutants (Chapter III). Gag protein with the Q195A mutations was capable of assembling particles in vitro, whereas virus with this same mutation could bud from cells. However, the Q195A mutation prohibited the CA protein from assembling into the same structures in vitro as the WT protein, and the Q195A mutant virus does not produce stable

149 mature cores in vivo. These results are more in line with a maturation pathway that includes the disassembly of the immature particle and de novo assembly of the mature capsid.

The nature of the immature particle also lends itself to de novo assembly of the mature capsid. In vivo, blocking cleavage of Gag at the last step allows only part of the mature capsid to form from CA-SP. Our in vitro assembly data indicates that CA-SP is capable of assembling from a monomeric state. These two seemingly contradictory observations, although obtained with two different viruses (HIV and RSV, respectively), can be reconciled if capsid assembly originates at

“defects” in the immature particle. Gaps found in the Gag hexameric lattice offer a spot where released CA-SP from Gag could dissociate into smaller oligomers of the protein and nucleate reassembly. As more and more Gag protein is cleaved and SP is removed from CA, the contacts that make up the mature capsid could be established.

LIMITATIONS AND ALTERNATIVE APPROACHES

Determination of the structure of the immature lattice has been difficult due to the high level of heterogeneity of authentic retrovirions isolated from infected cells. The inherent flexibility of the Gag protein and its ability to participate in different kinds of interactions precludes crystallography and the protein is too large for NMR. Therefore, most of our models are based on phenotypes resulting from mutagenesis and low resolution imaging of particles. The lack of good structures has made selecting specific residues for genetic analysis largely a guessing game. We chose hydrophobic residues at the dimer interface primarily from their location in the mature RSV CTD dimer. Comparison of this structure with that of the high pH dimer, which is currently the best model for the RSV immature dimer interface, allowed us to predict that some of the residues would also affect Gag assembly. However, without high resolution structures of the immature RSV particle, it is possible that we have missed residues that are important for formation of the interaction interfaces. Combining cryoEM and cryoET of

RSV ΔMBDΔPR proteins similar to what has been done in M-PMV and HIV-1 will eventually

150 provide better structural models for RSV immature assembly. These models can then be used to direct investigations of influence of specific residues on maturation.

Several of the dimer residues tested failed to produce assembled immature particles in vitro (Figure 3.4). Together with decreased budding efficiency of the virus, these results indicate a defect in Gag-Gag interactions. However, from these experiments we cannot conclusively rule out that the Gag molecules are unable to interact with each other. It is possible that small oligomers of the protein can assemble into structures such as hexamers which are not easily visible by negative stain TEM. Recent work measuring anisotropy of Gag mixed with oligonucleotides has made possible the detection of small oligomeric assembly of Gag. Using this technique it should be possible to determine whether the assembly-incompetent mutant

ΔMBDΔPR proteins make any complexes at all and, if so, what their nature is. Furthermore, if one or more of these mutants accumulates only in hexamers, structures of these complexes might be obtained by cryoEM.

One of the aims of studying the CA-SP protein was to identify the structure of SP in an assembled particle. Unfortunately, the SP region is still disordered in CA-SP icosahedra (Figure

4.8). Obtaining a structure of the SP will make identifying its function significantly easier, as we will be able to determine interaction partners. From preliminary PRE-NMR studies we can infer that the SP region associates with the MHR and α9 at least transiently (Figure 4.12). Further enhancement of PRE-NMR studies will determine if these interactions are real and important to the function of the SP. Addition of 8 residues from NC to the CA-SP protein created a new class of assembled particles (Figure 4.10). Though imaging of these particles has not been completed, it is possible that the 8-residue extension will provide stability to the SP that is sufficient to enable high resolution structural analysis of the region, as has been obtained for HIV-1 and M-

PMV. Molecular dynamics modeling of the SP-NC linkage predicts that the 6-helix bundle is stabilized by electrostatic interactions (79). Mutation of some of these residues to cysteine may

151 allow disulfide crosslinking that stabilizes the bundle, again allowing analysis by cryoEM or crystallography.

FINAL THOUGHTS AND POTENTIAL APPLICATIONS

The CTD of retroviruses is a multifunctional domain that must accomplish a lot with only

80 amino acids and a relatively conserved structure during the replication cycle of the virus.

Initially the CTD provides primary contacts for hexamers of Gag and tie those hexamers together via interhexameric interactions. Additionally, the domain facilitates packaging of the reverse transcription tRNA primer by binding to tRNA synthetase. All of these interactions save for the interhexameric interface dissolve completely at the onset of PR mediated maturation. After proteolytic cleavage, the orientation of the CTD changes to lock in at the dimerization interface characteristic of the mature particle. The formation of different CTD-CTD dimers depends on contacts between different residues in the 310 helix and α9. We provide here the best genetic and biochemical support to date for the recently described model of immature particle structure as well as evidence for maturation switches in the CTD and SP region. However, higher resolution imaging of these contacts is needed to further advance our knowledge of the interfaces of immature particles and the mechanisms that regulate maturation of the capsid.

The results provided in this dissertation point to several critical residues in the CTD required for assembly of both immature and mature particles. Based on the interactions the CTD must make and the fact that residues in the 310 helix and α9 are part of interfaces in both the immature and mature dimer, this region is a potential therapeutic target. Disrupting contacts required for assembly of both particles could inhibit retrovirus replication at two stages, making the development of resistance more difficult. Further structural analysis of this region should facilitate discovery and testing of therapeutics for their capacity to disrupt the dimer interface.

152

REFERENCES

1. Rous, P. 1911. A SARCOMA OF THE FOWL TRANSMISSIBLE BY AN AGENT SEPARABLE FROM THE TUMOR CELLS. J.Exp.Med. 13:397-411

2. Baltimore, D. 1970. RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature 226:1209-1211

3. Temin, H. M. and S. Mizutani. 1970. RNA-dependent DNA polymerase in virions of Rous sarcoma virus. Nature 226:1211-1213

4. de Harven, E. 1974. Remarks on the ultrastructure of type A, B, and C virus particles. Adv.Virus Res. 19:221-264

5. BERNHARD, W. 1960. The detection and study of tumor viruses with the electron microscope. Cancer Res. 20:712-727

6. Vogt, V. M. 1997. Retroviral Virions and Genomes, p. 27-69. In: J. M. Coffin, S. H. Hughes, and H. Varmus (eds.), Retroviruses. Cold Spring Harbor Laboratory Press, Woodbury, NY.

7. Wang, L., D. Galehouse, P. Mellon, P. Duesberg, W. S. Mason, and P. K. Vogt. 1976. Mapping oligonucleotides of Rous sarcoma virus RNA that segregate with polymerase and group-specific antigen markers in recombinants. Proc.Natl.Acad.Sci.U.S.A 73:3952-3956

8. Beemon, K. L., A. J. Faras, A. T. Hasse, P. H. Duesberg, and J. E. Maisel. 1976. Genomic complexities of murine leukemia and sarcoma, reticuloendotheliosis, and visna viruses. J.Virol. 17:525-537

9. Beemon, K., P. Duesberg, and P. Vogt. 1974. Evidence for crossing-over between avian tumor viruses based on analysis of viral RNAs. Proc.Natl.Acad.Sci.U.S.A 71:4254-4258

10. Joho, R. H., M. A. Billeter, and C. Weissmann. 1975. Mapping of biological functions on RNA of avian tumor viruses: location of regions required for transformation and determination of host range. Proc.Natl.Acad.Sci.U.S.A 72:4772-4776

11. Toyoshima, K. and P. K. Vogt. 1969. Temperature sensitive mutants of an avian sarcoma virus. Virology 39:930-931

12. Chow, S. A., K. A. Vincent, V. Ellison, and P. O. Brown. 1992. Reversal of integration and DNA splicing mediated by integrase of human immunodeficiency virus. Science 255:723-726

13. LaFemina, R. L., P. L. Callahan, and M. G. Cordingley. 1991. Substrate specificity of recombinant human immunodeficiency virus integrase protein. J.Virol. 65:5624-5630

14. Masuda, T., M. J. Kuroda, and S. Harada. 1998. Specific and independent recognition of U3 and U5 att sites by human immunodeficiency virus type 1 integrase in vivo. J.Virol. 72:8396-8402

153

15. Jacks, T., M. D. Power, F. R. Masiarz, P. A. Luciw, P. J. Barr, and H. E. Varmus. 1988. Characterization of ribosomal frameshifting in HIV-1 gag-pol expression. Nature 331:280-283

16. Yeager, M., E. M. Wilson-Kubalek, S. G. Weiner, P. O. Brown, and A. Rein. 1998. Supramolecular organization of immature and mature murine leukemia virus revealed by electron cryo-microscopy: Implications for retroviral assembly mechanisms. PNAS 95:7299-7304

17. Bolognesi, D. P., R. C. Montelaro, H. Frank, and W. Schafer. 1978. Assembly of type C oncornaviruses: a model. Science 199:183-186

18. Montelaro, R. C., S. J. Sullivan, and D. P. Bolognesi. 1978. An analysis of type-C retrovirus polypeptides and their associations in the virion. Virology 84:19-31

19. Montelaro, R. C. and D. P. Bolognesi. 1978. Structure and morphogenesis of type-C retroviruses. Adv.Cancer Res. 28:63-89

20. Briggs, J. A., M. N. Simon, I. Gross, H. G. Krausslich, S. D. Fuller, V. M. Vogt, and M. C. Johnson. 2004. The stoichiometry of Gag protein in HIV-1. Nat.Struct.Mol.Biol. 11:672-675

21. Murray, P. S., Z. Li, J. Wang, C. L. Tang, B. Honig, and D. Murray. 2005. Retroviral matrix domains share electrostatic homology: models for membrane binding function throughout the viral life cycle. Structure. 13:1521-1531

22. Dalton, A. K., D. Ako-Adjei, P. S. Murray, D. Murray, and V. M. Vogt. 2007. Electrostatic interactions drive membrane association of the human immunodeficiency virus type 1 Gag MA domain. J.Virol. 81:6434-6445

23. Dalton, A. K., P. S. Murray, D. Murray, and V. M. Vogt. 2005. Biochemical characterization of rous sarcoma virus MA protein interaction with membranes. J.Virol. 79:6227-6238

24. Tang, C., E. Loeliger, P. Luncsford, I. Kinde, D. Beckett, and M. F. Summers. 2004. Entropic switch regulates myristate exposure in the HIV-1 matrix protein. Proc.Natl.Acad.Sci U.S.A 101:517-522

25. Butan, C., D. C. Winkler, J. B. Heymann, R. C. Craven, and A. C. Steven. 2008. RSV Capsid Polymorphism Correlates with Polymerization Efficiency and Envelope Glycoprotein Content: Implications that Nucleation Controls Morphogenesis. J.Mol.Biol. 376:1168-1181

26. Freed, E. O. and M. A. Martin. 1996. Domains of the human immunodeficiency virus type 1 matrix and gp41 cytoplasmic tail required for envelope incorporation into virions. J.Virol. 70:341-351

27. Ott, D. E., L. V. Coren, and T. D. Gagliardi. 2005. Redundant roles for nucleocapsid and matrix RNA-binding sequences in human immunodeficiency virus type 1 assembly. J.Virol. 79:13839-13847

154

28. Sun, M., I. F. Grigsby, R. J. Gorelick, L. M. Mansky, and K. Musier-Forsyth. 2014. Retrovirus-specific differences in matrix and nucleocapsid protein-nucleic acid interactions: implications for genomic RNA packaging. J.Virol. 88:1271-1280

29. Stansell, E., R. Apkarian, S. Haubova, W. E. Diehl, E. M. Tytler, and E. Hunter. 2007. Basic residues in the Mason-Pfizer monkey virus gag matrix domain regulate intracellular trafficking and capsid-membrane interactions. J.Virol. 81:8977-8988

30. Dvorin, J. D. and M. H. Malim. 2003. Intracellular trafficking of HIV-1 cores: journey to the center of the cell. Curr.Top.Microbiol.Immunol. 281:179-208

31. Dorfman, T., A. Bukovsky, A. Ohagen, S. Hoglund, and H. G. Gottlinger. 1994. Functional domains of the capsid protein of human immunodeficiency virus type 1. J.Virol. 68:8180-8187

32. Gamble, T. R., S. Yoo, F. F. Vajdos, U. K. von Schwedler, D. K. Worthylake, H. Wang, J. P. McCutcheon, W. I. Sundquist, and C. P. Hill. 1997. Structure of the carboxyl-terminal dimerization domain of the HIV-1 capsid protein. Science 278:849-853

33. Alfadhli, A., T. C. Dhenub, A. Still, and E. Barklis. 2005. Analysis of human immunodeficiency virus type 1 Gag dimerization-induced assembly. J.Virol. 79:14498- 14506

34. Datta, S. A. K., Z. Zhao, P. K. Clark, S. Tarasov, J. N. Alexandratos, S. J. Campbell, M. Kvaratskhelia, J. Lebowitz, and A. Rein. 2007. Interactions between HIV-1 Gag Molecules in Solution: An Inositol Phosphate-mediated Switch. J.Mol.Biol. 365:799-811

35. Krausslich, H. G., M. Facke, A. M. Heuser, J. Konvalinka, and H. Zentgraf. 1995. The spacer peptide between human immunodeficiency virus capsid and nucleocapsid proteins is essential for ordered assembly and viral infectivity. J.Virol. 69:3407-3419

36. Wright, E. R., J. B. Schooler, H. J. Ding, C. Kieffer, C. Fillmore, W. I. Sundquist, and G. J. Jensen. 2007. Electron cryotomography of immature HIV-1 virions reveals the structure of the CA and SP1 Gag shells. EMBO J 26:2218-2226

37. von Schwedler, U. K., K. M. Stray, J. E. Garrus, and W. I. Sundquist. 2003. Functional surfaces of the human immunodeficiency virus type 1 capsid protein. J Virol. 77:5439-5450

38. Bailey, G. D., J. K. Hyun, A. K. Mitra, and R. L. Kingston. 2009. Proton-linked dimerization of a retroviral capsid protein initiates capsid assembly. Structure 17:737-748

39. Yu, F., S. M. Joshi, Y. M. Ma, R. L. Kingston, M. N. Simon, and V. M. Vogt. 2001. Characterization of Rous sarcoma virus Gag particles assembled in vitro. J Virol. 75:2753- 2764

40. Kingston, R. L., T. Fitzon-Ostendorp, E. Z. Eisenmesser, G. W. Schatz, V. M. Vogt, C. B. Post, and M. G. Rossmann. 2000. Structure and self-association of the Rous sarcoma virus capsid protein. Structure 8:617-628

41. Berthet-Colominas, C., S. Monaco, A. Novelli, G. Sibai, F. Mallet, and S. Cusack. 1999. Head-to-tail dimers and interdomain flexibility revealed by the crystal structure of

155

HIV-1 capsid protein (p24) complexed with a monoclonal antibody Fab. EMBO J. 18:1124-1136

42. Ganser-Pornillos, B. K., M. Yeager, and O. Pornillos. 2012. Assembly and architecture of HIV. Adv.Exp.Med.Biol. 726:441-465

43. Ganser-Pornillos, B. K., M. Yeager, and W. I. Sundquist. 2008. The structural biology of HIV assembly. Curr.Opin.Struct.Biol. 18:203-217

44. Ganser-Pornillos, B. K., A. Cheng, and M. Yeager. 2007. Structure of Full-Length HIV- 1 CA: A Model for the Mature Capsid Lattice. Cell 131:70-79

45. Sundquist, W. I. and H. G. Krausslich. 2012. HIV-1 assembly, budding, and maturation. Cold Spring Harb.Perspect.Med. 2:a006924

46. Arvidson, B., J. Seeds, M. Webb, L. Finlay, and E. Barklis. 2003. Analysis of the retrovirus capsid interdomain linker region. Virology 308:166-177

47. Bouamr, F., C. C. Cornilescu, S. P. Goff, N. Tjandra, and C. A. Carter. 2005. Structural and dynamics studies of the D54A mutant of human T cell leukemia virus-1 capsid protein. J.Biol.Chem. 280:6792-6801

48. Gitti, R. K., B. M. Lee, J. Walker, M. F. Summers, S. Yoo, and W. I. Sundquist. 1996. Structure of the amino-terminal core domain of the HIV-1 capsid protein. Science 273:231- 235

49. von Schwedler, U. K., T. L. Stemmler, V. Y. Klishko, S. Li, K. H. Albertine, D. R. Davis, and W. I. Sundquist. 1998. Proteolytic refolding of the HIV-1 capsid protein amino-terminus facilitates viral core assembly. EMBO J 17:1555-1568

50. Momany, C., L. C. Kovari, A. J. Prongay, W. Keller, R. K. Gitti, B. M. Lee, A. E. Gorbalenya, L. Tong, J. McClure, L. S. Ehrlich, M. F. Summers, C. Carter, and M. G. Rossmann. 1996. Crystal structure of dimeric HIV-1 capsid protein. Nat.Struct.Biol. 3:763-770

51. Gamble, T. R., F. F. Vajdos, S. Yoo, D. K. Worthylake, M. Houseweart, W. I. Sundquist, and C. P. Hill. 1996. Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV-1 capsid. Cell 87:1285-1294

52. Jin, Z., L. Jin, D. L. Peterson, and C. L. Lawson. 1999. Model for lentivirus capsid core assembly based on crystal dimers of EIAV p26. J.Mol.Biol. 286:83-93

53. Khorasanizadeh, S., R. Campos-Olivas, and M. F. Summers. 1999. Solution structure of the capsid protein from the human T-cell leukemia virus type-I. J.Mol.Biol. 291:491-505

54. Mortuza, G. B., D. C. Goldstone, C. Pashley, L. F. Haire, M. Palmarini, W. R. Taylor, J. P. Stoye, and I. A. Taylor. 2008. Structure of the Capsid Amino-Terminal Domain from the Betaretrovirus, Jaagsiekte Sheep Retrovirus. J.Mol.Biol.

156

55. Mortuza, G. B., M. P. Dodding, D. C. Goldstone, L. F. Haire, J. P. Stoye, and I. A. Taylor. 2008. Structure of B-MLV capsid amino-terminal domain reveals key features of viral tropism, gag assembly and core formation. J.Mol.Biol. 376:1493-1508

56. Mortuza, G. B., L. F. Haire, A. Stevens, S. J. Smerdon, J. P. Stoye, and I. A. Taylor. 2004. High-resolution structure of a retroviral capsid hexameric amino-terminal domain. Nature 431:481-485

57. Campos-Olivas, R., J. L. Newman, and M. F. Summers. 2000. Solution structure and dynamics of the Rous sarcoma virus capsid protein and comparison with capsid proteins of other retroviruses. J.Mol.Biol. 296:633-649

58. Worthylake, D. K., H. Wang, S. Yoo, W. I. Sundquist, and C. P. Hill. 1999. Structures of the HIV-1 capsid protein dimerization domain at 2.6 A resolution. Acta Crystallogr.D Biol.Crystallogr. 55:85-92

59. Craven, R. C., A. E. Leuredupree, R. A. Weldon, and J. W. Wills. 1995. Genetic- Analysis of the Major Homology Region of the Rous-Sarcoma Virus Gag Protein. J.Virol. 69:4213-4227

60. Dalessio, P. M., R. C. Craven, P. M. Lokhandwala, and I. J. Ropson. 2013. Lethal mutations in the major homology region and their suppressors act by modulating the dimerization of the rous sarcoma virus capsid protein C-terminal domain. Proteins 81:316- 325

61. Butan, C., P. M. Lokhandwala, J. G. Purdy, G. Cardone, R. C. Craven, and A. C. Steven. 2010. Suppression of a morphogenic mutant in Rous sarcoma virus capsid protein by a second-site mutation: a cryoelectron tomography study. J.Virol. 84:6377-6386

62. Bowzard, J. B., J. W. Wills, and R. C. Craven. 2001. Second-site suppressors of Rous sarcoma virus CA mutations: Evidence for interdomain interactions. J.Virol. 75:6850-6856

63. Cairns, T. M. and R. C. Craven. 2001. Viral DNA synthesis defects in assembly- competent Rous sarcoma virus CA mutants. J.Virol. 75:242-250

64. Mayo, K., D. Huseby, J. McDermott, B. Arvidson, L. Finlay, and E. Barklis. 2003. Retrovirus capsid protein assembly arrangements. J.Mol.Biol. 325:225-237

65. Byeon, I. J., G. Hou, Y. Han, C. L. Suiter, J. Ahn, J. Jung, C. H. Byeon, A. M. Gronenborn, and T. Polenova. 2012. Motions on the millisecond time scale and multiple conformations of HIV-1 capsid protein: implications for structural polymorphism of CA assemblies. J.Am.Chem.Soc. 134:6455-6466

66. Dewan, V., T. Liu, K. M. Chen, Z. Qian, Y. Xiao, L. Kleiman, K. V. Mahasenan, C. Li, H. Matsuo, D. Pei, and K. Musier-Forsyth. 2012. Cyclic peptide inhibitors of HIV-1 capsid-human lysyl-tRNA synthetase interaction. ACS Chem.Biol. 7:761-769

67. Kovaleski, B. J., R. Kennedy, A. Khorchid, L. Kleiman, H. Matsuo, and K. Musier- Forsyth. 2007. Critical role of helix 4 of HIV-1 capsid C-terminal domain in interactions with human lysyl-tRNA synthetase. J.Biol.Chem. 282:32274-32279

157

68. D'Souza, V. and M. F. Summers. 2005. How retroviruses select their genomes. Nat.Rev.Microbiol. 3:643-655

69. Phillips, J. M., P. S. Murray, D. Murray, and V. M. Vogt. 2008. A molecular switch required for retrovirus assembly participates in the hexagonal immature lattice. EMBO J. 27:1411-1420

70. Lee, E. G., A. Alidina, C. May, and M. L. Linial. 2003. Importance of basic residues in binding of rous sarcoma virus nucleocapsid to the RNA packaging signal. J.Virol. 77:2010- 2020

71. Dannull, J., A. Surovoy, G. Jung, and K. Moelling. 1994. Specific binding of HIV-1 nucleocapsid protein to PSI RNA in vitro requires N-terminal zinc finger and flanking basic amino acid residues. EMBO J. 13:1525-1533

72. Garbitt-Hirst, R., S. P. Kenney, and L. J. Parent. 2009. Genetic evidence for a connection between Rous sarcoma virus gag nuclear trafficking and genomic RNA packaging. J.Virol. 83:6790-6797

73. Wills, J. W. and R. C. Craven. 1991. Form, function, and use of retroviral gag proteins. AIDS 5:639-654

74. Keller, P. W., R. Huang, M. England, K. Waki, N. Cheng, J. B. Heymann, R. C. Craven, E. O. Freed, and A. C. Steven. 2013. A two-pronged structural analysis of retroviral maturation indicates that core formation proceeds by a disassembly-reassembly pathway, rather than a displacive transition. J.Virol.

75. Keller, P. W., C. S. Adamson, J. B. Heymann, E. O. Freed, and A. C. Steven. 2011. HIV-1 maturation inhibitor bevirimat stabilizes the immature Gag lattice. J.Virol. 85:1420- 1428

76. Bieniasz, P. D. 2006. Late budding domains and host proteins in enveloped virus release. Virology 344:55-63

77. Strack, B., A. Calistri, S. Craig, E. Popova, and H. G. Gottlinger. 2003. AIP1/ALIX is a binding partner for HIV-1 p6 and EIAV p9 functioning in virus budding. Cell 114:689-699

78. Datta, S. A., L. G. Temeselew, R. M. Crist, F. Soheilian, A. Kamata, J. Mirro, D. Harvin, K. Nagashima, R. E. Cachau, and A. Rein. 2011. On the role of the SP1 domain in HIV-1 particle assembly: a molecular switch? J.Virol. 85:4111-4121

79. Bush, D. L., E. B. Monroe, G. J. Bedwell, P. E. Prevelige, Jr., J. M. Phillips, and V. M. Vogt. 2014. Higher Order Structure of the Rous Sarcoma Virus SP Assembly Domain. J.Virol.

80. Stein, B. S., S. D. Gowda, J. D. Lifson, R. C. Penhallow, K. G. Bensch, and E. G. Engleman. 1987. pH-independent HIV entry into CD4-positive T cells via virus envelope fusion to the plasma membrane. Cell 49:659-668

158

81. Earp, L. J., S. E. Delos, R. C. Netter, P. Bates, and J. M. White. 2003. The avian retrovirus avian sarcoma/leukosis virus subtype A reaches the lipid mixing stage of fusion at neutral pH. J.Virol. 77:3058-3066

82. Barnard, R. J. O., S. Narayan, G. Dornadula, M. D. Miller, and J. A. T. Young. 2004. Low pH Is Required for Avian Sarcoma and Leukosis Virus Env-Dependent Viral Penetration into the Cytosol and Not for Viral Uncoating. J.Virol. 78:10433-10441

83. McClure, M. O., M. Marsh, and R. A. Weiss. 1988. Human immunodeficiency virus infection of CD4-bearing cells occurs by a pH-independent mechanism. EMBO J. 7:513- 518

84. Pauza, C. D. and T. M. Price. 1988. Human immunodeficiency virus infection of T cells and monocytes proceeds via receptor-mediated endocytosis. J.Cell Biol. 107:959-968

85. Zhu, P., J. Liu, J. Bess, E. Chertova, J. D. Lifson, H. Gris+¬, G. A. Ofek, K. A. Taylor, and K. H. Roux. 2006. Distribution and three-dimensional structure of AIDS virus envelope spikes. Nature advanced online publication:

86. Bates, P., L. Rong, H. E. Varmus, J. A. Young, and L. B. Crittenden. 1998. Genetic mapping of the cloned subgroup A avian sarcoma and leukosis virus receptor gene to the TVA locus. J.Virol. 72:2505-2508

87. Fackler, O. T. and B. M. Peterlin. 2000. Endocytic entry of HIV-1. Curr.Biol. 10:1005- 1008

88. Schaeffer, E., R. Geleziunas, and W. C. Greene. 2001. Human immunodeficiency virus type 1 Nef functions at the level of virus entry by enhancing cytoplasmic delivery of virions. J Virol 75:2993-3000

89. Colman, P. M. and M. C. Lawrence. 2003. The structural biology of type I viral membrane fusion. Nat.Rev.Mol.Cell Biol. 4:309-319

90. Arhel, N. 2010. Revisiting HIV-1 uncoating. Retrovirology. 7:96

91. Forshey, B. M., U. von Schwedler, W. I. Sundquist, and C. Aiken. 2002. Formation of a human immunodeficiency virus type 1 core of optimal stability is crucial for viral replication. J.Virol. 76:5667-5677

92. Borroto-Esoda, K. and L. R. Boone. 1994. Development of a human immunodeficiency virus-1 in vitro DNA synthesis system to study reverse transcriptase inhibitors. Antiviral Res. 23:235-249

93. Borroto-Esoda, K. and L. R. Boone. 1991. Equine infectious anemia virus and human immunodeficiency virus DNA synthesis in vitro: characterization of the endogenous reverse transcriptase reaction. J.Virol. 65:1952-1959

94. Alin, K. and S. P. Goff. 1996. Amino acid substitutions in the CA protein of Moloney murine leukemia virus that block early events in infection. Virology 222:339-351

159

95. Tang, S., T. Murakami, B. E. Agresta, S. Campbell, E. O. Freed, and J. G. Levin. 2001. Human immunodeficiency virus type 1 N-terminal capsid mutants that exhibit aberrant core morphology and are blocked in initiation of reverse transcription in infected cells. J Virol 75:9357-9366

96. Gilboa, E., S. W. Mitra, S. Goff, and D. Baltimore. 1979. A detailed model of reverse transcription and tests of crucial aspects. Cell 18:93-100

97. Haseltine, W. A., D. G. Kleid, A. Panet, E. Rothenberg, and D. Baltimore. 1976. Ordered transcription of RNA tumor virus genomes. J Mol.Biol. 106:109-131

98. Cen, S., H. Javanbakht, S. Kim, K. Shiba, R. Craven, A. Rein, K. Ewalt, P. Schimmel, K. Musier-Forsyth, and L. Kleiman. 2002. Retrovirus-specific packaging of aminoacyl- tRNA synthetases with cognate primer tRNAs. J.Virol. 76:13111-13115

99. Bebenek, K., J. Abbotts, S. H. Wilson, and T. A. Kunkel. 1993. Error-prone polymerization by HIV-1 reverse transcriptase. Contribution of template-primer misalignment, miscoding, and termination probability to mutational hot spots. J.Biol.Chem. 268:10324-10334

100. Battula, N. and L. A. Loeb. 1976. On the fidelity of DNA replication. Lack of exodeoxyribonuclease activity and error-correcting function in avian myeloblastosis virus DNA polymerase. J.Biol.Chem. 251:982-986

101. Williams, K. J. and L. A. Loeb. 1992. Retroviral reverse transcriptases: error frequencies and mutagenesis. Curr.Top.Microbiol.Immunol. 176:165-180

102. Leider, J. M., P. Palese, and F. I. Smith. 1988. Determination of the mutation rate of a retrovirus. J.Virol. 62:3084-3091

103. Cheng-Mayer, C., C. Weiss, D. Seto, and J. A. Levy. 1989. Isolates of human immunodeficiency virus type 1 from the brain may constitute a special group of the AIDS virus. Proc.Natl.Acad.Sci.USA 86:8575-8579

104. Cheng-Mayer, C., D. Seto, M. Tateno, and J. A. Levy. 1988. Biologic features of HIV-1 that correlate with virulence in the host. Science 240:80-82

105. Coffin, J. M. 1995. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 267:483-489

106. O'Neil, P. K., G. Sun, H. Yu, Y. Ron, J. P. Dougherty, and B. D. Preston. 2002. Mutational analysis of HIV-1 long terminal repeats to explore the relative contribution of reverse transcriptase and RNA polymerase II to viral mutagenesis. J.Biol.Chem. 277:38053-38061

107. Lewis, P. F. and M. Emerman. 1994. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J.Virol. 68:510-516

108. Humphries, E. H. and H. M. Temin. 1972. Cell cycle-dependent activation of rous sarcoma virus-infected stationary chicken cells: avian leukosis virus group-specific antigens and ribonucleic acid. J.Virol. 10:82-87

160

109. Roe, T., T. C. Reynolds, G. Yu, and P. O. Brown. 1993. Integration of murine leukemia virus DNA depends on mitosis. EMBO J. 12:2099-2108

110. Fassati, A., D. Gorlich, I. Harrison, L. Zaytseva, and J. M. Mingot. 2003. Nuclear import of HIV-1 intracellular reverse transcription complexes is mediated by importin 7. EMBO J. 22:3675-3685

111. Ciuffi, A. and F. D. Bushman. 2006. Retroviral DNA integration: HIV and the role of LEDGF/p75. Trends Genet. 22:388-395

112. Ciuffi, A., M. Llano, E. Poeschla, C. Hoffmann, J. Leipzig, P. Shinn, J. R. Ecker, and F. Bushman. 2005. A role for LEDGF/p75 in targeting HIV DNA integration. Nat.Med. 11:1287-1289

113. Katzman, M. and M. Sudol. 1995. Mapping domains of retroviral integrase responsible for viral DNA specificity and target site selection by analysis of chimeras between human immunodeficiency virus type 1 and visna virus integrases. J.Virol. 69:5687-5696

114. Rabson, A. B. and B. J. Graves. 1997. Synthesis and Processing of Viral RNA.

115. Fogel, B. L., L. M. McNally, and M. T. McNally. 2002. Efficient polyadenylation of Rous sarcoma virus RNA requires the negative regulator of splicing element. Nucleic Acids Res. 30:810-817

116. McNally, L. M. and M. T. McNally. 1999. U1 small nuclear ribonucleoprotein and splicing inhibition by the rous sarcoma virus negative regulator of splicing element. J.Virol. 73:2385-2393

117. Stake, M. S., D. V. Bann, R. J. Kaddis, and L. J. Parent. 2013. Nuclear trafficking of retroviral RNAs and Gag proteins during late steps of replication. Viruses. 5:2767-2795

118. Beyer, A. R., D. V. Bann, B. Rice, I. S. Pultz, M. Kane, S. P. Goff, T. V. Golovkina, and L. J. Parent. 2013. Nucleolar trafficking of the mouse mammary tumor virus gag protein induced by interaction with ribosomal protein L9. J.Virol. 87:1069-1082

119. Parent, L. J. 2011. New insights into the nuclear localization of retroviral Gag proteins. Nucleus. 2:92-97

120. Scheifele, L. Z., E. P. Ryan, and L. J. Parent. 2005. Detailed mapping of the nuclear export signal in the Rous sarcoma virus Gag protein. J Virol 79:8732-8741

121. Gudleski, N., J. M. Flanagan, E. P. Ryan, M. C. Bewley, and L. J. Parent. 2010. Directionality of nucleocytoplasmic transport of the retroviral gag protein depends on sequential binding of karyopherins and viral RNA. Proc.Natl.Acad.Sci.U.S.A 107:9358- 9363

122. Scheifele, L. Z., R. A. Garbitt, J. D. Rhoads, and L. J. Parent. 2002. Nuclear entry and CRM1-dependent nuclear export of the Rous sarcoma virus Gag polyprotein. Proc.Natl.Acad.Sci U.S.A 99:3944-3949

161

123. Chiu, Y. L., C. K. Ho, N. Saha, B. Schwer, S. Shuman, and T. M. Rana. 2002. Tat stimulates cotranscriptional capping of HIV mRNA. Mol.Cell 10:585-597

124. Zhou, M., L. Deng, F. Kashanchi, J. N. Brady, A. J. Shatkin, and A. Kumar. 2003. The Tat/TAR-dependent phosphorylation of RNA polymerase II C-terminal domain stimulates cotranscriptional capping of HIV-1 mRNA. Proc.Natl.Acad.Sci.U.S.A 100:12666-12671

125. Harger, J. W., A. Meskauskas, J. Nielsen, M. C. Justice, and J. D. Dinman. 2001. Ty1 retrotransposition and programmed +1 ribosomal frameshifting require the integrity of the protein synthetic translocation step. Virology 286:216-224

126. Wang, Y., N. M. Wills, Z. Du, A. Rangan, J. F. Atkins, R. F. Gesteland, and D. W. Hoffman. 2002. Comparative studies of frameshifting and nonframeshifting RNA pseudoknots: a mutational and NMR investigation of pseudoknots derived from the bacteriophage T2 gene 32 mRNA and the retroviral gag-pro frameshift site. RNA. 8:981- 996

127. Marcheschi, R. J., M. Tonelli, A. Kumar, and S. E. Butcher. 2011. Structure of the HIV-1 frameshift site RNA bound to a small molecule inhibitor of viral replication. ACS Chem.Biol. 6:857-864

128. Marcheschi, R. J., D. W. Staple, and S. E. Butcher. 2007. Programmed ribosomal frameshifting in SIV is induced by a highly structured RNA stem-loop. J.Mol.Biol. 373:652-663

129. Tang, Y., U. Winkler, E. O. Freed, T. A. Torrey, W. Kim, H. Li, S. P. Goff, and H. C. Morse, III. 1999. Cellular motor protein KIF-4 associates with retroviral Gag. J.Virol. 73:10508-10513

130. Martinez, N. W., X. Xue, R. G. Berro, G. Kreitzer, and M. D. Resh. 2008. Kinesin KIF4 regulates intracellular trafficking and stability of the human immunodeficiency virus type 1 Gag polyprotein. J.Virol. 82:9937-9950

131. Gaudin, R., B. C. de Alencar, M. Jouve, S. Berre, B. E. Le, M. Schindler, A. Varthaman, F. X. Gobert, and P. Benaroch. 2012. Critical role for the kinesin KIF3A in the HIV life cycle in primary human macrophages. J.Cell Biol. 199:467-479

132. Jouvenet, N., S. J. Neil, C. Bess, M. C. Johnson, C. A. Virgen, S. M. Simon, and P. D. Bieniasz. 2006. Plasma membrane is the site of productive HIV-1 particle assembly. PLoS.Biol. 4:e435

133. Zhou, W., L. J. Parent, J. W. Wills, and M. D. Resh. 1994. Identification of a membrane-binding domain within the amino-terminal region of human immunodeficiency virus type 1 Gag protein which interacts with acidic phospholipids. J.Virol. 68:2556-2569

134. Yuan, X., X. Yu, T. H. Lee, and M. Essex. 1993. Mutations in the N-terminal region of human immunodeficiency virus type 1 matrix protein block intracellular transport of the Gag precursor. J.Virol. 67:6387-6394

162

135. Vlach, J. and J. S. Saad. 2013. Trio engagement via plasma membrane phospholipids and the myristoyl moiety governs HIV-1 matrix binding to bilayers. Proc.Natl.Acad.Sci.U.S.A 110:3525-3530

136. Fledderman, E. L., K. Fujii, R. H. Ghanam, K. Waki, P. E. Prevelige, E. O. Freed, and J. S. Saad. 2010. Myristate exposure in the human immunodeficiency virus type 1 matrix protein is modulated by pH. Biochemistry 49:9551-9562

137. Saad, J. S., S. D. Ablan, R. H. Ghanam, A. Kim, K. Andrews, K. Nagashima, F. Soheilian, E. O. Freed, and M. F. Summers. 2008. Structure of the myristylated human immunodeficiency virus type 2 matrix protein and the role of phosphatidylinositol-(4,5)- bisphosphate in membrane targeting. J.Mol.Biol. 382:434-447

138. Dick, R. A., S. L. Goh, G. W. Feigenson, and V. M. Vogt. 2012. HIV-1 Gag protein can sense the cholesterol and acyl chain environment in model membranes. Proc.Natl.Acad.Sci.U.S.A 109:18761-18766

139. Chan, J., R. A. Dick, and V. M. Vogt. 2011. Rous sarcoma virus gag has no specific requirement for phosphatidylinositol-(4,5)-bisphosphate for plasma membrane association in vivo or for liposome interaction in vitro. J.Virol. 85:10851-10860

140. Ono, A., S. D. Ablan, S. J. Lockett, K. Nagashima, and E. O. Freed. 2004. Phosphatidylinositol (4,5) bisphosphate regulates HIV-1 Gag targeting to the plasma membrane. Proc.Natl.Acad.Sci.U.S.A 101:14889-14894

141. Nadaraia-Hoke, S., D. V. Bann, T. L. Lochmann, N. Gudleski-O'Regan, and L. J. Parent. 2013. Alterations in the MA and NC domains modulate phosphoinositide- dependent plasma membrane localization of the Rous sarcoma virus Gag protein. J.Virol. 87:3609-3615

142. Inlora, J., V. Chukkapalli, D. Derse, and A. Ono. 2011. Gag localization and virus-like particle release mediated by the matrix domain of human T-lymphotropic virus type 1 Gag are less dependent on phosphatidylinositol-(4,5)-bisphosphate than those mediated by the matrix domain of HIV-1 Gag. J.Virol. 85:3802-3810

143. Dick, R. A., E. Kamynina, and V. M. Vogt. 2013. Effect of multimerization on membrane association of Rous sarcoma virus and HIV-1 matrix domain proteins. J.Virol. 87:13598-13608

144. Callahan, E. M. and J. W. Wills. 2000. Repositioning basic residues in the M domain of the Rous sarcoma virus gag protein. J.Virol. 74:11222-11229

145. Leung, K., J. O. Kim, L. Ganesh, J. Kabat, O. Schwartz, and G. J. Nabel. 2008. HIV-1 assembly: viral glycoproteins segregate quantally to lipid rafts that associate individually with HIV-1 capsids and virions. Cell Host.Microbe 3:285-292

146. Ono, A. and E. O. Freed. 2001. Plasma membrane rafts play a critical role in HIV-1 assembly and release. Proc.Natl.Acad.Sci.USA 98:13925-13930

147. Garnier, L., L. J. Parent, B. Rovinski, S. X. Cao, and J. W. Wills. 1999. Identification of retroviral late domains as determinants of particle size. J.Virol. 73:2309-2320

163

148. Garnier, L., L. Ratner, B. Rovinski, S. X. Cao, and J. W. Wills. 1998. Particle size determinants in the human immunodeficiency virus type 1 Gag protein. J.Virol. 72:4667- 4677

149. Morita, E., V. Sandrin, S. L. Alam, D. M. Eckert, S. P. Gygi, and W. I. Sundquist. 2007. Identification of human MVB12 proteins as ESCRT-I subunits that function in HIV budding. Cell Host.Microbe 2:41-53

150. Langelier, C., U. K. von Schwedler, R. D. Fisher, D. De, I, P. L. White, C. P. Hill, J. Kaplan, D. Ward, and W. I. Sundquist. 2006. Human ESCRT-II complex and its role in human immunodeficiency virus type 1 release. J.Virol. 80:9465-9480

151. McCullough, J., R. D. Fisher, F. G. Whitby, W. I. Sundquist, and C. P. Hill. 2008. ALIX-CHMP4 interactions in the human ESCRT pathway. Proc.Natl.Acad.Sci.USA 105:7687-7691

152. Scott, A., J. Gaspar, M. D. Stuchell-Brereton, S. L. Alam, J. J. Skalicky, and W. I. Sundquist. 2005. Structure and ESCRT-III protein interactions of the MIT domain of human VPS4A. Proc.Natl.Acad.Sci U.S.A 102:13813-13818

153. Martin-Serrano, J. and S. J. Neil. 2011. Host factors involved in retroviral budding and release. Nat.Rev.Microbiol. 9:519-531

154. Reicin, A. S., A. Ohagen, L. Yin, S. Hoglund, and S. P. Goff. 1996. The role of Gag in human immunodeficiency virus type 1 virion morphogenesis and early steps of the viral life cycle. J Virol 70:8645-8652

155. Stremlau, M., M. Perron, M. Lee, Y. Li, B. Song, H. Javanbakht, F. Diaz-Griffero, D. J. Anderson, W. I. Sundquist, and J. Sodroski. 2006. Specific recognition and accelerated uncoating of retroviral capsids by the TRIM5alpha restriction factor. Proc.Natl.Acad.Sci U.S.A 103:5514-5519

156. Anderson, J. L., E. M. Campbell, X. Wu, N. Vandegraaff, A. Engelman, and T. J. Hope. 2006. Proteasome inhibition reveals that a functional preintegration complex intermediate can be generated during restriction by diverse TRIM5 proteins. J.Virol. 80:9754-9760

157. Wu, X., J. L. Anderson, E. M. Campbell, A. M. Joseph, and T. J. Hope. 2006. Proteasome inhibitors uncouple rhesus TRIM5alpha restriction of HIV-1 reverse transcription and infection. Proc.Natl.Acad.Sci.U.S.A 103:7465-7470

158. Kutluay, S. B., D. Perez-Caballero, and P. D. Bieniasz. 2013. Fates of retroviral core components during unrestricted and TRIM5-restricted infection. PLoS.Pathog. 9:e1003214

159. Shi, J., J. Zhou, U. D. Halambage, V. B. Shah, M. J. Burse, H. Wu, W. S. Blair, S. L. Butler, and C. Aiken. 2014. Compensatory Substitutions in the HIV-1 Capsid Reduce the Fitness Cost Associated with Resistance to a Capsid-Targeting Small-Molecule Inhibitor. J.Virol.

160. Blair, W. S., C. Pickford, S. L. Irving, D. G. Brown, M. Anderson, R. Bazin, J. Cao, G. Ciaramella, J. Isaacson, L. Jackson, R. Hunt, A. Kjerrstrom, J. A. Nieman, A. K.

164

Patick, M. Perros, A. D. Scott, K. Whitby, H. Wu, and S. L. Butler. 2010. HIV capsid is a tractable target for small molecule therapeutic intervention. PLoS.Pathog. 6:e1001220

161. Blair, W. S., J. Cao, J. Fok-Seang, P. Griffin, J. Isaacson, R. L. Jackson, E. Murray, A. K. Patick, Q. Peng, M. Perros, C. Pickford, H. Wu, and S. L. Butler. 2009. New small-molecule inhibitor class targeting human immunodeficiency virus type 1 virion maturation. Antimicrob.Agents Chemother. 53:5080-5087

162. Xu, H., T. Franks, G. Gibson, K. Huber, N. Rahm, C. S. De Castillia, J. Luban, C. Aiken, S. Watkins, N. Sluis-Cremer, and Z. Ambrose. 2013. Evidence for biphasic uncoating during HIV-1 infection from a novel imaging assay. Retrovirology. 10:70

163. Lamorte, L., S. Titolo, C. T. Lemke, N. Goudreau, J. F. Mercier, E. Wardrop, V. B. Shah, U. K. von Schwedler, C. Langelier, S. S. Banik, C. Aiken, W. I. Sundquist, and S. W. Mason. 2013. Discovery of novel small-molecule HIV-1 replication inhibitors that stabilize capsid complexes. Antimicrob.Agents Chemother. 57:4622-4631

164. Kortagere, S., N. Madani, M. K. Mankowski, A. Schon, I. Zentner, G. Swaminathan, A. Princiotto, K. Anthony, A. Oza, L. J. Sierra, S. R. Passic, X. Wang, D. M. Jones, E. Stavale, F. C. Krebs, J. Martin-Garcia, E. Freire, R. G. Ptak, J. Sodroski, S. Cocklin, and A. B. Smith, III. 2012. Inhibiting early-stage events in HIV-1 replication by small- molecule targeting of the HIV-1 capsid. J.Virol. 86:8472-8481

165. Braaten, D., E. K. Franke, and J. Luban. 1996. Cyclophilin A is required for an early step in the life cycle of human immunodeficiency virus type 1 before the initiation of reverse transcription. J.Virol. 70:3551-3560

166. Ptak, R. G., P. A. Gallay, D. Jochmans, A. P. Halestrap, U. T. Ruegg, L. A. Pallansch, M. D. Bobardt, M. P. de Bethune, J. Neyts, C. E. De, J. M. Dumont, P. Scalfaro, K. Besseghir, R. M. Wenger, and B. Rosenwirth. 2008. Inhibition of human immunodeficiency virus type 1 replication in human cells by Debio-025, a novel cyclophilin binding agent. Antimicrob.Agents Chemother. 52:1302-1317

167. Yang, R. and C. Aiken. 2007. A mutation in alpha helix 3 of CA renders human immunodeficiency virus type 1 cyclosporin A resistant and dependent: rescue by a second- site substitution in a distal region of CA. J Virol 81:3749-3756

168. Takemura, T., M. Kawamata, M. Urabe, and T. Murakami. 2013. Cyclophilin A- dependent restriction to capsid N121K mutant human immunodeficiency virus type 1 in a broad range of cell lines. J.Virol. 87:4086-4090

169. Qi, M., R. Yang, and C. Aiken. 2008. Cyclophilin A-dependent restriction of human immunodeficiency virus type 1 capsid mutants for infection of nondividing cells. J.Virol. 82:12001-12008

170. Yamashita, M. and M. Emerman. 2009. Cellular restriction targeting viral capsids perturbs human immunodeficiency virus type 1 infection of nondividing cells. J.Virol. 83:9835-9843

171. Rasaiyaah, J., C. P. Tan, A. J. Fletcher, A. J. Price, C. Blondeau, L. Hilditch, D. A. Jacques, D. L. Selwood, L. C. James, M. Noursadeghi, and G. J. Towers. 2013. HIV-1

165

evades innate immune recognition through specific cofactor recruitment. Nature 503:402- 405

172. Lahaye, X., T. Satoh, M. Gentili, S. Cerboni, C. Conrad, I. Hurbain, M. A. El, C. Lacabaratz, J. D. Lelievre, and N. Manel. 2013. The capsids of HIV-1 and HIV-2 determine immune detection of the viral cDNA by the innate sensor cGAS in dendritic cells. Immunity. 39:1132-1142

173. Ambrose, Z. and C. Aiken. 2014. HIV-1 uncoating: connection to nuclear entry and regulation by host proteins. Virology 454-455:371-379

174. Crawford, S. and S. P. Goff. 1985. A deletion mutation in the 5' part of the pol gene of Moloney murine leukemia virus blocks proteolytic processing of the gag and pol polyproteins. J.Virol. 53:899-907

175. Kohl, N. E., E. A. Emini, W. A. Schleif, L. J. Davis, J. C. Heimbach, R. A. Dixon, E. M. Scolnick, and I. S. Sigal. 1988. Active human immunodeficiency virus protease is required for viral infectivity. Proc.Natl.Acad.Sci.U.S.A 85:4686-4690

176. Stewart, L., G. Schatz, and V. M. Vogt. 1990. Properties of avian retrovirus particles defective in viral protease. J Virol 64:5076-5092

177. Sommerfelt, M. A., S. R. Petteway, Jr., G. B. Dreyer, and E. Hunter. 1992. Effect of retroviral proteinase inhibitors on Mason-Pfizer monkey virus maturation and transmembrane glycoprotein cleavage. J.Virol. 66:4220-4227

178. Adamson, C. S., S. D. Ablan, I. Boeras, R. Goila-Gaur, F. Soheilian, K. Nagashima, F. Li, K. Salzwedel, M. Sakalian, C. T. Wild, and E. O. Freed. 2006. In Vitro Resistance to the Human Immunodeficiency Virus Type 1 Maturation Inhibitor PA-457 (Bevirimat). J.Virol. 80:10957-10971

179. Briggs, J. A. G., M. C. Johnson, M. N. Simon, S. D. Fuller, and V. M. Vogt. 2006. Cryo-electron Microscopy Reveals Conserved and Divergent Features of Gag Packing in Immature Particles of Rous Sarcoma Virus and Human Immunodeficiency Virus. J.Mol.Biol. 355:157-168

180. Craven, R. C., A. E. Leure-duPree, C. R. Erdie, C. B. Wilson, and J. W. Wills. 1993. Necessity of the spacer peptide between CA and NC in the Rous sarcoma virus gag protein. J.Virol. 67:6246-6252

181. de Marco A., N. E. Davey, P. Ulbrich, J. M. Phillips, V. Lux, J. D. Riches, T. Fuzik, T. Ruml, H. G. Krausslich, V. M. Vogt, and J. A. Briggs. 2010. Conserved and variable features of Gag structure and arrangement in immature retrovirus particles. J.Virol. 84:11729-11736

182. Briggs, J. A., J. D. Riches, B. Glass, V. Bartonova, G. Zanetti, and H. G. Krausslich. 2009. Structure and assembly of immature HIV. Proc.Natl.Acad.Sci.USA

183. Carlson, L. A., J. A. Briggs, B. Glass, J. D. Riches, M. N. Simon, M. C. Johnson, B. Muller, K. Grunewald, and H. G. Krausslich. 2008. Three-dimensional analysis of

166

budding sites and released virus suggests a revised model for HIV-1 morphogenesis. Cell Host Microbe 4:592-599

184. Nermut, M. V., D. J. Hockley, J. B. Jowett, I. M. Jones, M. Garreau, and D. Thomas. 1994. Fullerene-like organization of HIV gag-protein shell in virus-like particles produced by recombinant baculovirus. Virology 198:288-296

185. Heymann, J. B., C. Butan, D. C. Winkler, R. C. Craven, and A. C. Steven. 2008. Irregular and semi-regular polyhedral models for Rous sarcoma virus cores. Comp.Math.Meth.Med. 9:197-210

186. Bharat, T. A., L. R. Castillo Menendez, W. J. Hagen, V. Lux, S. Igonet, M. Schorb, F. K. Schur, H. G. Krausslich, and J. A. Briggs. 2014. Cryo-electron microscopy of tubular arrays of HIV-1 Gag resolves structures essential for immature virus assembly. Proc.Natl.Acad.Sci.U.S.A 111:8233-8238

187. Bharat, T. A., N. E. Davey, P. Ulbrich, J. D. Riches, M. A. de, M. Rumlova, C. Sachse, T. Ruml, and J. A. Briggs. 2012. Structure of the immature retroviral capsid at 8 A resolution by cryo-electron microscopy. Nature 487:385-389

188. Lanman, J., T. T. Lam, M. R. Emmett, A. G. Marshall, M. Sakalian, and P. E. Prevelige, Jr. 2004. Key interactions in HIV-1 maturation identified by hydrogen- deuterium exchange. Nat.Struct.Mol.Biol. 11:676-677

189. Ganser-Pornillos, B. K., U. K. von Schwedler, K. M. Stray, C. Aiken, and W. I. Sundquist. 2004. Assembly Properties of the Human Immunodeficiency Virus Type 1 CA Protein. J.Virol. 78:2545-2552

190. Fowler, P. W. and D. E. Manolopoulas. 1995. An Atlas of Fullerenes. Dover Publications, Mineola, NY.

191. Bailey, G. D., J. K. Hyun, A. K. Mitra, and R. L. Kingston. 2012. A structural model for the generation of continuous curvature on the surface of a retroviral capsid. J.Mol.Biol. 417:212-223

192. Kingston, R. L., N. H. Olson, and V. M. Vogt. 2001. The organization of mature Rous sarcoma virus as studied by cryoelectron microscopy. J.Struct.Biol. 136:67-80

193. Ganser, B. K., A. Cheng, W. I. Sundquist, and M. Yeager. 2003. Three-dimensional structure of the M-MuLV CA protein on a lipid monolayer: a general model for retroviral capsid assembly. EMBO J. 22:2886-2892

194. Cardone, G., J. G. Purdy, N. Cheng, R. C. Craven, and A. C. Steven. 2009. Visualization of a missing link in retrovirus capsid assembly. Nature 457:694-698

195. Krausslich, H. G., H. Schneider, G. Zybarth, C. A. Carter, and E. Wimmer. 1988. Processing of in vitro-synthesized gag precursor proteins of human immunodeficiency virus (HIV) type 1 by HIV proteinase generated in Escherichia coli. J.Virol. 62:4393-4397

167

196. Burstein, H., D. Bizub, M. Kotler, G. Schatz, V. M. Vogt, and A. M. Skalka. 1992. Processing of avian retroviral gag polyprotein precursors is blocked by a mutation at the NC-PR cleavage site. J.Virol. 66:1781-1785

197. Hizi, A., L. E. Henderson, T. D. Copeland, R. C. Sowder, C. V. Hixson, and S. Oroszlan. 1987. Characterization of mouse mammary tumor virus gag-pro gene products and the ribosomal frameshift site by protein sequencing. Proc.Natl.Acad.Sci.U.S.A 84:7041-7045

198. Hatanaka, M. and S. H. Nam. 1989. Identification of HTLV-I gag protease and its sequential processing of the gag gene product. J.Cell Biochem. 40:15-30

199. Pepinsky, R. B., I. A. Papayannopoulos, E. P. Chow, N. K. Krishna, R. C. Craven, and V. M. Vogt. 1995. Differential proteolytic processing leads to multiple forms of the CA protein in avian sarcoma and leukemia viruses. J.Virol. 69:6430-6438

200. Pettit, S. C., M. D. Moody, R. S. Wehbie, A. H. Kaplan, P. V. Nantermet, C. A. Klein, and R. Swanstrom. 1994. The p2 domain of human immunodeficiency virus type 1 Gag regulates sequential proteolytic processing and is required to produce fully infectious virions. J.Virol. 68:8017-8027

201. Tritch, R. J., Y. E. Cheng, F. H. Yin, and S. Erickson-Viitanen. 1991. Mutagenesis of protease cleavage sites in the human immunodeficiency virus type 1 gag polyprotein. J.Virol. 65:922-930

202. Schock, H. B., V. M. Garsky, and L. C. Kuo. 1996. Mutational anatomy of an HIV-1 protease variant conferring cross-resistance to protease inhibitors in clinical trials. Compensatory modulations of binding and activity. J.Biol.Chem. 271:31957-31963

203. Krausslich, H. G., R. H. Ingraham, M. T. Skoog, E. Wimmer, P. V. Pallai, and C. A. Carter. 1989. Activity of purified biosynthetic proteinase of human immunodeficiency virus on natural substrates and synthetic peptides. Proc.Natl.Acad.Sci.USA 86:807-811

204. Jamjoom, G. A., R. B. Naso, and R. B. Arlinghaus. 1976. Selective decrease in the rate of cleavage of an intracellular precursor to Rauscher leukemia virus p30 by treatment of infected cells with actinomycin D. J.Virol. 19:1054-1072

205. Sheng, N., S. C. Pettit, R. J. Tritch, D. H. Ozturk, M. M. Rayner, R. Swanstrom, and S. Erickson-Viitanen. 1997. Determinants of the human immunodeficiency virus type 1 p15NC-RNA interaction that affect enhanced cleavage by the viral protease. J.Virol. 71:5723-5732

206. Sheng, N. and S. Erickson-Viitanen. 1994. Cleavage of p15 protein in vitro by human immunodeficiency virus type 1 protease is RNA dependent. J.Virol. 68:6207-6214

207. Murti, K. G., M. Bondurant, and A. Tereba. 1981. Secondary structural features in the 70S RNAs of Moloney murine leukemia and Rous sarcoma viruses as observed by electron microscopy. J.Virol. 37:411-419

208. Hoglund, S., A. Ohagen, J. Goncalves, A. T. Panganiban, and D. Gabuzda. 1997. Ultrastructure of HIV-1 genomic RNA. Virology 233:271-279

168

209. Fu, W. and A. Rein. 1993. Maturation of dimeric viral RNA of Moloney murine leukemia virus. J.Virol. 67:5443-5449

210. Shehu-Xhilaga, M., H. G. Kraeusslich, S. Pettit, R. Swanstrom, J. Y. Lee, J. A. Marshall, S. M. Crowe, and J. Mak. 2001. Proteolytic processing of the p2/nucleocapsid cleavage site is critical for human immunodeficiency virus type 1 RNA dimer maturation. J.Virol. 75:9156-9164

211. Housset, V. and J. L. Darlix. 1996. Mutations at the capsid-nucleocapsid cleavage site of gag polyprotein of Moloney murine leukemia virus abolish virus infectivity. C.R.Acad.Sci.III 319:81-89

212. Ohishi, M., T. Nakano, S. Sakuragi, T. Shioda, K. Sano, and J. Sakuragi. 2011. The relationship between HIV-1 genome RNA dimerization, virion maturation and infectivity. Nucleic Acids Res. 39:3404-3417

213. Jalalirad, M. and M. Laughrea. 2010. Formation of immature and mature genomic RNA dimers in wild-type and protease-inactive HIV-1: differential roles of the Gag polyprotein, nucleocapsid proteins NCp15, NCp9, NCp7, and the dimerization initiation site. Virology 407:225-236

214. Fu, W., Q. Dang, K. Nagashima, E. O. Freed, V. K. Pathak, and W. S. Hu. 2006. Effects of Gag mutation and processing on retroviral dimeric RNA maturation. J.Virol. 80:1242-1249

215. Liang, C., L. Rong, E. Cherry, L. Kleiman, M. Laughrea, and M. A. Wainberg. 1999. Deletion mutagenesis within the dimerization initiation site of human immunodeficiency virus type 1 results in delayed processing of the p2 peptide from precursor proteins. J.Virol. 73:6147-6151

216. L'Hernault, A., E. U. Weiss, J. S. Greatorex, and A. M. Lever. 2012. HIV-2 genome dimerization is required for the correct processing of Gag: a second-site reversion in matrix can restore both processes in dimerization-impaired mutant viruses. J.Virol. 86:5867-5876

217. Muriaux, D., J. Mirro, D. Harvin, and A. Rein. 2001. RNA is a structural element in retrovirus particles. Proc.Natl.Acad.Sci.USA 98:5246-5251

218. Park, J. and C. D. Morrow. 1993. Mutations in the protease gene of human immunodeficiency virus type 1 affect release and stability of virus particles. Virology 194:843-850

219. Kaplan, A. H., P. Krogstad, D. J. Kempf, D. W. Norbeck, and R. Swanstrom. 1994. Human immunodeficiency virus type 1 virions composed of unprocessed Gag and Gag-Pol precursors are capable of reverse transcribing viral genomic RNA. Antimicrob.Agents Chemother. 38:2929-2933

220. Cortines, J. R., E. B. Monroe, S. Kang, and P. E. Prevelige, Jr. 2011. A retroviral chimeric capsid protein reveals the role of the N-terminal beta-hairpin in mature core assembly. J.Mol.Biol. 410:641-652

169

221. Spidel, J. L., C. B. Wilson, R. C. Craven, and J. W. Wills. 2007. Genetic Studies of the beta-hairpin loop of Rous sarcoma virus capsid protein. J.Virol. 81:1288-1296

222. Debouck, C., J. G. Gorniak, J. E. Strickler, T. D. Meek, B. W. Metcalf, and M. Rosenberg. 1987. Human immunodeficiency virus protease expressed in Escherichia coli exhibits autoprocessing and specific maturation of the gag precursor. Proc.Natl.Acad.Sci.U.S.A 84:8903-8906

223. Kotler, M., G. Arad, and S. H. Hughes. 1992. Human immunodeficiency virus type 1 gag-protease fusion proteins are enzymatically active. J.Virol. 66:6781-6783

224. Craven, R. C., R. P. Bennett, and J. W. Wills. 1991. Role of the avian retroviral protease in the activation of reverse transcriptase during virion assembly. J.Virol. 65:6205-6217

225. Panet, A. and D. Baltimore. 1987. Characterization of endonuclease activities in Moloney murine leukemia virus and its replication-defective mutants. J.Virol. 61:1756-1760

226. Witte, O. N. and D. Baltimore. 1978. Relationship of retrovirus polyprotein cleavages to virion maturation studied with temperature-sensitive murine leukemia virus mutants. J.Virol. 26:750-761

227. Waki, K., S. R. Durell, F. Soheilian, K. Nagashima, S. L. Butler, and E. O. Freed. 2012. Structural and functional insights into the HIV-1 maturation inhibitor binding pocket. PLoS.Pathog. 8:e1002997

228. England, M. R., J. G. Purdy, I. J. Ropson, P. M. Dalessio, and R. C. Craven. 2014. Potential Role for CA-SP in Nucleating Retroviral Capsid Maturation. J.Virol. 88:7170- 7177

229. Hadravova, R., M. A. de, P. Ulbrich, J. Stokrova, M. Dolezal, I. Pichova, T. Ruml, J. A. Briggs, and M. Rumlova. 2012. In vitro assembly of virus-like particles of a gammaretrovirus, the murine leukemia virus XMRV. J.Virol. 86:1297-1306

230. Briggs, J. A. and H. G. Krausslich. 2011. The molecular architecture of HIV. J.Mol.Biol. 410:491-500

231. Pornillos, O., B. K. Ganser-Pornillos, B. N. Kelly, Y. Hua, F. G. Whitby, C. D. Stout, W. I. Sundquist, C. P. Hill, and M. Yeager. 2009. X-ray structures of the hexameric building block of the HIV capsid. Cell 137:1282-1292

232. Hyun, J. K., M. Radjainia, R. L. Kingston, and A. K. Mitra. 2010. Proton-driven assembly of the Rous Sarcoma virus capsid protein results in the formation of icosahedral particles. J.Biol.Chem. 285:15056-15064

233. Briggs, J. A., T. Wilk, R. Welker, H. G. Krausslich, and S. D. Fuller. 2003. Structural organization of authentic, mature HIV-1 virions and cores. EMBO J. 22:1707-1715

234. Auerbach, M. R., K. R. Brown, and I. R. Singh. 2007. Mutational analysis of the N- terminal domain of Moloney murine leukemia virus capsid protein. J.Virol. 81:12337- 12347

170

235. Auerbach, M. R., K. R. Brown, A. Kaplan, D. Las Nueces, and I. R. Singh. 2006. A Small Loop in the Capsid Protein of Moloney Murine Leukemia Virus Controls Assembly of Spherical Cores. J.Virol. 80:2884-2893

236. Tang, C., E. Loeliger, I. Kinde, S. Kyere, K. Mayo, E. Barklis, Y. Sun, M. Huang, and M. F. Summers. 2003. Antiviral inhibition of the HIV-1 capsid protein. J Mol.Biol. 327:1013-1020

237. Wildova, M., R. Hadravova, J. Stokrova, I. Krizova, T. Ruml, E. Hunter, I. Pichova, and M. Rumlova. 2008. The effect of point mutations within the N-terminal domain of Mason-Pfizer monkey virus capsid protein on virus core assembly and infectivity. Virology 380:157-163

238. Lanman, J. and P. E. Prevelige, Jr. 2005. Kinetic and mass spectrometry-based investigation of human immunodeficiency virus type 1 assembly and maturation. Adv.Virus Res. 64:285-309

239. Lanman, J., T. T. Lam, S. Barnes, M. Sakalian, M. R. Emmett, A. G. Marshall, and P. E. Prevelige. 2003. Identification of novel interactions in HIV-1 capsid protein assembly by high-resolution mass spectrometry. J.Mol.Biol. 325:759-772

240. Rumlova-Klikova, M., E. Hunter, M. V. Nermut, I. Pichova, and T. Ruml. 2000. Analysis of Mason-Pfizer monkey virus Gag domains required for capsid assembly in bacteria: role of the N-terminal proline residue of CA in directing particle shape. J Virol 74:8452-8459

241. Purdy, J. G., J. M. Flanagan, I. J. Ropson, and R. C. Craven. 2009. Retroviral capsid assembly: a role for the CA dimer in initiation. J.Mol.Biol. 389:438-451

242. Yu, X., Q. Wang, J. C. Yang, I. Buch, C. J. Tsai, B. Ma, S. Z. Cheng, R. Nussinov, and J. Zheng. 2009. Mutational analysis and allosteric effects in the HIV-1 capsid protein carboxyl-terminal dimerization domain. Biomacromolecules. 10:390-399

243. Mateu, M. G. 2002. Conformational stability of dimeric and monomeric forms of the C- terminal domain of human immunodeficiency virus-1 capsid protein. J.Mol.Biol. 318:519- 531

244. Byeon, I. J., X. Meng, J. Jung, G. Zhao, R. Yang, J. Ahn, J. Shi, J. Concel, C. Aiken, P. Zhang, and A. M. Gronenborn. 2009. Structural convergence between Cryo-EM and NMR reveals intersubunit interactions critical for HIV-1 capsid function. Cell 139:780-790

245. Pornillos, O., B. K. Ganser-Pornillos, and M. Yeager. 2011. Atomic-level modelling of the HIV capsid. Nature 469:424-427

246. Weldon, R. A., Jr. and J. W. Wills. 1993. Characterization of a small (25-kilodalton) derivative of the Rous sarcoma virus Gag protein competent for particle release. J Virol 67:5550-5561

247. Krishna, N. K., S. Campbell, V. M. Vogt, and J. W. Wills. 1998. Genetic determinants of Rous sarcoma virus particle size. J.Virol. 72:564-577

171

248. Wang, M. Q. and S. P. Goff. 2003. Defects in virion production caused by mutations affecting the C-terminal portion of the Moloney murine leukemia virus capsid protein. J Virol 77:3339-3344

249. Borsetti, A., A. Ohagen, and H. G. Gottlinger. 1998. The C-terminal half of the human immunodeficiency virus type 1 Gag precursor is sufficient for efficient particle assembly. J.Virol. 72:9313-9317

250. Purdy, J. G., J. M. Flanagan, I. J. Ropson, K. E. Rennoll-Bankert, and R. C. Craven. 2008. Critical role of conserved hydrophobic residues within the major homology region in mature retroviral capsid assembly. J.Virol. 82:5951-5961

251. Monroe, E. B., S. Kang, S. K. Kyere, R. Li, and P. E. Prevelige, Jr. 2010. Hydrogen/deuterium exchange analysis of HIV-1 capsid assembly and maturation. Structure. 18:1483-1491

252. Han, Y., G. Hou, C. L. Suiter, J. Ahn, I. J. Byeon, A. S. Lipton, S. D. Burton, I. Hung, P. L. Gor'kov, Z. Gan, W. W. Brey, D. Rice, A. M. Gronenborn, and T. E. Polenova. 2013. Magic Angle Spinning NMR Reveals Sequence-Dependent Structural Plasticity, Dynamics, and the Spacer Peptide 1 Conformation in HIV-1 Capsid Protein Assemblies. J.Am.Chem.Soc.

253. Lee, S. K., M. Potempa, and R. Swanstrom. 2012. The choreography of HIV-1 proteolytic processing and virion assembly. J.Biol.Chem. 287:40867-40874

254. Keller, P. W., M. C. Johnson, and V. M. Vogt. 2008. Mutations in the spacer peptide and adjoining sequences in Rous sarcoma virus Gag lead to tubular budding. J.Virol. 82:6788- 6797

255. Guo, X., J. Hu, J. B. Whitney, R. S. Russell, and C. Liang. 2004. Important role for the CA-NC spacer region in the assembly of bovine immunodeficiency virus Gag protein. J.Virol. 78:551-560

256. Elder, J. H., M. Schnolzer, C. S. Hasselkus-Light, M. Henson, D. A. Lerner, T. R. Phillips, P. C. Wagaman, and S. B. Kent. 1993. Identification of proteolytic processing sites within the Gag and Pol polyproteins of feline immunodeficiency virus. J.Virol. 67:1869-1876

257. Henderson, L. E., R. E. Benveniste, R. Sowder, T. D. Copeland, A. M. Schultz, and S. Oroszlan. 1988. Molecular characterization of gag proteins from simian immunodeficiency virus (SIVMne). J.Virol. 62:2587-2595

258. Henderson, L. E., R. C. Sowder, G. W. Smythers, and S. Oroszlan. 1987. Chemical and immunological characterizations of equine infectious anemia virus gag-encoded proteins. J.Virol. 61:1116-1124

259. Mervis, R. J., N. Ahmad, E. P. Lillehoj, M. G. Raum, F. H. Salazar, H. W. Chan, and S. Venkatesan. 1988. The gag gene products of human immunodeficiency virus type 1: alignment within the gag open reading frame, identification of posttranslational modifications, and evidence for alternative gag precursors. J.Virol. 62:3993-4002

172

260. Bennett, R. P., S. Rhee, R. C. Craven, E. Hunter, and J. W. Wills. 1991. Amino acids encoded downstream of gag are not required by Rous sarcoma virus protease during gag- mediated assembly. J.Virol. 65:272-280

261. Konvalinka, J., A. M. Heuser, O. Hruskova-Heidingsfeldova, V. M. Vogt, J. Sedlacek, P. Strop, and H. G. Krausslich. 1995. Proteolytic processing of particle-associated retroviral polyproteins by homologous and heterologous viral proteinases. Eur.J.Biochem. 228:191-198

262. Tobin, G. J., R. C. Sowder, D. Fabris, M. Y. Hu, J. K. Battles, C. Fenselau, L. E. Henderson, and M. A. Gonda. 1994. Amino acid sequence analysis of the proteolytic cleavage products of the bovine immunodeficiency virus Gag precursor polypeptide. J Virol 68:7620-7627

263. Li, F., R. Goila-Gaur, K. Salzwedel, N. R. Kilgore, M. Reddick, C. Matallana, A. Castillo, D. Zoumplis, D. E. Martin, J. M. Orenstein, G. P. Allaway, E. O. Freed, and C. T. Wild. 2003. PA-457: A potent HIV inhibitor that disrupts core condensation by targeting a late step in Gag processing. Proc.Natl.Acad.Sci.USA 100:13555-13560

264. Sakalian, M., C. P. McMurtrey, F. J. Deeg, C. W. Maloy, F. Li, C. T. Wild, and K. Salzwedel. 2006. 3-O-(3',3'-Dimethysuccinyl) Betulinic Acid Inhibits Maturation of the Human Immunodeficiency Virus Type 1 Gag Precursor Assembled In Vitro. J.Virol. 80:5716-5722

265. Zhou, J., C. H. Chen, and C. Aiken. 2006. Human Immunodeficiency Virus Type 1 Resistance to the Small Molecule Maturation Inhibitor 3-O-(3',3'-Dimethylsuccinyl)- Betulinic Acid Is Conferred by a Variety of Single Amino Acid Substitutions at the CA- SP1 Cleavage Site in Gag. J.Virol. 80:12095-12101

266. Purdy, J. G. 2009. Ph.D. thesis. Penn State University, College of Medicine, Hershey, PA.

267. Taylor, G. M., L. Ma, V. M. Vogt, and C. B. Post. 2010. NMR relaxation studies of an RNA-binding segment of the rous sarcoma virus gag polyprotein in free and bound states: a model for autoinhibition of assembly. Biochemistry 49:4006-4017

268. Morellet, N., S. Druillennec, C. Lenoir, S. Bouaziz, and B. P. Roques. 2005. Helical structure determined by NMR of the HIV-1 (345-392)Gag sequence, surrounding p2: implications for particle assembly and RNA packaging. Protein Sci. 14:375-386

269. Newman, J. L., E. W. Butcher, D. T. Patel, Y. Mikhaylenko, and M. F. Summers. 2004. Flexibility in the P2 domain of the HIV-1 Gag polyprotein. Protein Sci. 13:2101- 2107

270. Tang, S., S. Ablan, M. Dueck, W. Ayala-Lopez, B. Soto, M. Caplan, K. Nagashima, I. K. Hewlett, E. O. Freed, and J. G. Levin. 2007. A second-site suppressor significantly improves the defective phenotype imposed by mutation of an aromatic residue in the N- terminal domain of the HIV-1 capsid protein. Virology 359:105-115

271. Lokhandwala, P. M., T. L. Nguyen, J. B. Bowzard, and R. C. Craven. 2008. Cooperative role of the MHR and the CA dimerization helix in the maturation of the functional retrovirus capsid. Virology 376:191-198

173

272. McDermott, J., S. Karanjia, Z. Love, and E. Barklis. 2000. Crosslink analysis of N- terminal, C-terminal, and N/B determining regions of the Moloney murine leukemia virus capsid protein. Virology 269:190-200

273. Chien, A. I., W. H. Liao, D. M. Yang, and C. T. Wang. 2006. A domain directly C- terminal to the major homology region of human immunodeficiency type 1 capsid protein plays a crucial role in directing both virus assembly and incorporation of Gag-Pol. Virology 348:84-95

274. Larsen, L. S., M. Zhang, N. Beliakova-Bethell, V. Bilanchone, A. Lamsa, K. Nagashima, R. Najdi, K. Kosaka, V. Kovacevic, J. Cheng, P. Baldi, G. W. Hatfield, and S. Sandmeyer. 2007. Ty3 capsid mutations reveal early and late functions of the amino-terminal domain. J.Virol. 81:6957-6972

275. Orlinsky, K. J., J. Gu, M. Hoyt, S. Sandmeyer, and T. M. Menees. 1996. Mutations in the Ty3 major homology region affect multiple steps in Ty3 retrotransposition. J.Virol. 70:3440-3448

276. Mammano, F., A. Ohagen, S. Hoglund, and H. G. Gottlinger. 1994. Role of the major homology region of human immunodeficiency virus type 1 in virion morphogenesis. J.Virol. 68:4927-4936

277. Chang, Y. F., S. M. Wang, K. J. Huang, and C. T. Wang. 2007. Mutations in capsid major homology region affect assembly and membrane affinity of HIV-1 Gag. J.Mol.Biol. 370:585-597

278. Meng, X., G. Zhao, E. Yufenyuy, D. Ke, J. Ning, M. Delucia, J. Ahn, A. M. Gronenborn, C. Aiken, and P. Zhang. 2012. Protease cleavage leads to formation of mature trimer interface in HIV-1 capsid. PLoS.Pathog. 8:e1002886

279. Waki, K., S. R. Durell, F. Soheilian, K. Nagashima, S. L. Butler, and E. O. Freed. 2012. Structural and functional insights into the HIV-1 maturation inhibitor binding pocket. PLoS.Pathog. 8:e1002997

280. Callahan, E. M. and J. W. Wills. 2003. Link between genome packaging and rate of budding for Rous sarcoma virus. J.Virol. 77:9388-9398

281. Wang, G. P. and F. D. Bushman. 2006. A statistical method for comparing viral growth curves. J.Virol.Methods 135:118-123

282. Kovari, L. C., C. A. Momany, F. Miyagi, S. Lee, S. Campbell, B. Vuong, V. M. Vogt, and M. G. Rossmann. 1997. Crystals of Rous sarcoma virus capsid protein show a helical arrangement of protein subunits. Virology 238:79-84

283. Studier, F. W. 2005. Protein production by auto-induction in high-density shaking cultures. Protein Expression and Purification 41:207-234

284. Ma, Y. M. and V. M. Vogt. 2004. Nucleic acid binding-induced Gag dimerization in the assembly of Rous sarcoma virus particles in vitro. J.Virol. 78:52-60

174

285. Tang, C., Y. Ndassa, and M. F. Summers. 2002. Structure of the N-terminal 283-residue fragment of the immature HIV-1 Gag polyprotein. Nat.Struct.Biol. 9:537-543

286. Wilk, T., I. Gross, B. E. Gowen, T. Rutten, F. de Haas, R. Welker, H. G. Krausslich, P. Boulanger, and S. D. Fuller. 2001. Organization of immature human immunodeficiency virus type 1. J Virol 75:759-771

287. de Marco A., B. Muller, B. Glass, J. D. Riches, H. G. Krausslich, and J. A. Briggs. 2010. Structural analysis of HIV-1 maturation using cryo-electron tomography. PLoS.Pathog. 6:e1001215

288. Wills, J. W., C. E. Cameron, C. B. Wilson, Y. Xiang, R. P. Bennett, and J. Leis. 1994. An assembly domain of the Rous sarcoma virus Gag protein required late in budding. J Virol 68:6605-6618

289. Sakalian, M., J. W. Wills, and V. M. Vogt. 1994. Efficiency and selectivity of RNA packaging by Rous sarcoma virus Gag deletion mutants. J.Virol. 68:5969-5981

290. Benjamin, J., B. K. Ganser-Pornillos, W. F. Tivol, W. I. Sundquist, and G. J. Jensen. 2005. Three-dimensional structure of HIV-1 virus-like particles by electron cryotomography. J.Mol.Biol. 346:577-588

291. Wiegers, K., G. Rutter, H. Kottler, U. Tessmer, H. Hohenberg, and H. G. Krausslich. 1998. Sequential steps in human immunodeficiency virus particle maturation revealed by alterations of individual Gag polyprotein cleavage sites. J Virol 72:2846-2854

292. Strambio-de-Castillia, C. and E. Hunter. 1992. Mutational analysis of the major homology region of Mason-Pfizer monkey virus by use of saturation mutagenesis. J Virol 66:7021-7032

293. Lee, S. K., M. Potempa, M. Kolli, A. Ozen, C. A. Schiffer, and R. Swanstrom. 2012. Context surrounding processing sites is crucial in determining cleavage rate of a subset of processing sites in HIV-1 Gag and Gag-Pro-Pol polyprotein precursors by viral protease. J.Biol.Chem. 287:13279-13290

294. Campos-Olivas, R., J. L. Newman, Y. Ndassa, and M. F. Summers. 1999. 1H, 13C and 15N chemical shift assignments of the capsid protein from Rous sarcoma virus. J.Biomol.NMR 15:267-268

295. Gross, I., H. Hohenberg, T. Wilk, K. Wiegers, M. Grattinger, B. Muller, S. Fuller, and H. G. Krausslich. 2000. A conformational switch controlling HIV-1 morphogenesis. EMBO J. 19:103-113

296. Noviello, C. M., C. S. Lopez, B. Kukull, H. McNett, A. Still, J. Eccles, R. Sloan, and E. Barklis. 2011. Second-site compensatory mutations of HIV-1 capsid mutations. J.Virol. 85:4730-4738

297. Mayo, K., M. L. Vana, J. McDermott, D. Huseby, J. Leis, and E. Barklis. 2002. Analysis of Rous sarcoma virus capsid protein variants assembled on lipid monolayers. J.Mol.Biol. 316:667-678

175

298. Nguyen, A. T., C. L. Feasley, K. W. Jackson, T. J. Nitz, K. Salzwedel, G. M. Air, and M. Sakalian. 2011. The prototype HIV-1 maturation inhibitor, bevirimat, binds to the CA- SP1 cleavage site in immature Gag particles. Retrovirology. 8:101

299. Alcaraz, L. A., M. del Alamo, F. N. Barrera, M. G. Mateu, and J. L. Neira. 2007. Flexibility in HIV-1 assembly subunits: solution structure of the monomeric C-terminal domain of the capsid protein. Biophys.J. 93:1264-1276

300. Alcaraz, L. A., M. del Alamo, M. G. Mateu, and J. L. Neira. 2008. Structural mobility of the monomeric C-terminal domain of the HIV-1 capsid protein. FEBS J. 275:3299-3311

301. Ternois, F., J. Sticht, S. Duquerroy, H. G. Krausslich, and F. A. Rey. 2005. The HIV-1 capsid protein C-terminal domain in complex with a virus assembly inhibitor. Nat.Struct.Mol.Biol. 12:678-682

302. Ivanov, D., O. V. Tsodikov, J. Kasanov, T. Ellenberger, G. Wagner, and T. Collins. 2007. Domain-swapped dimerization of the HIV-1 capsid C-terminal domain. Proc.Natl.Acad.Sci.USA 104:4353-4358

303. Bayro, M. J., B. Chen, W. M. Yau, and R. Tycko. 2014. Site-specific structural variations accompanying tubular assembly of the HIV-1 capsid protein. J.Mol.Biol. 426:1109-1127

304. Zhao, G., J. R. Perilla, E. L. Yufenyuy, X. Meng, B. Chen, J. Ning, J. Ahn, A. M. Gronenborn, K. Schulten, C. Aiken, and P. Zhang. 2013. Mature HIV-1 capsid structure by cryo-electron microscopy and all-atom molecular dynamics. Nature 497:643-646

305. Deroubaix, A., Q. Osseman, A. Cassany, D. Begu, J. Ragues, S. Kassab, S. Laine, and M. Kann. 2014. Expression of viral polymerase and phosphorylation of core protein determine core and capsid localization of the human hepatitis B virus. J.Gen.Virol.

306. Ludgate, L., X. Ning, D. H. Nguyen, C. Adams, L. Mentzer, and J. Hu. 2012. Cyclin- dependent kinase 2 phosphorylates s/t-p sites in the hepadnavirus core protein C-terminal domain and is incorporated into viral capsids. J.Virol. 86:12237-12250

307. Ludgate, L., C. Adams, and J. Hu. 2011. Phosphorylation state-dependent interactions of hepadnavirus core protein with host factors. PLoS.One. 6:e29566

308. Saad, J. S., A. Kim, R. H. Ghanam, A. K. Dalton, V. M. Vogt, Z. Wu, W. Lu, and M. F. Summers. 2007. Mutations that mimic phosphorylation of the HIV-1 matrix protein do not perturb the myristyl switch. Protein Sci 16:1793-1797

309. Radestock, B., I. Morales, S. A. Rahman, S. Radau, B. Glass, R. P. Zahedi, B. Muller, and H. G. Krausslich. 2013. Comprehensive mutational analysis reveals p6Gag phosphorylation to be dispensable for HIV-1 morphogenesis and replication. J.Virol. 87:724-734

310. Kaushik, R. and L. Ratner. 2004. Role of human immunodeficiency virus type 1 matrix phosphorylation in an early postentry step of virus replication. J.Virol. 78:2319-2326

176

311. Dochi, T., T. Nakano, M. Inoue, N. Takamune, S. Shoji, K. Sano, and S. Misumi. 2014. Phosphorylation of human immunodeficiency virus type 1 capsid protein at serine 16, required for peptidyl-prolyl isomerase-dependent uncoating, is mediated by virion- incorporated extracellular signal-regulated kinase 2. J.Gen.Virol. 95:1156-1166

312. Misumi, S., M. Inoue, T. Dochi, N. Kishimoto, N. Hasegawa, N. Takamune, and S. Shoji. 2010. Uncoating of human immunodeficiency virus type 1 requires prolyl isomerase Pin1. J.Biol.Chem. 285:25185-25195

313. Javanbakht, H., R. Halwani, S. Cen, J. Saadatmand, K. Musier-Forsyth, H. Gottlinger, and L. Kleiman. 2003. The interaction between HIV-1 Gag and human lysyl- tRNA synthetase during viral assembly. J.Biol.Chem. 278:27644-27651

314. Kleiman, L., C. P. Jones, and K. Musier-Forsyth. 2010. Formation of the tRNALys packaging complex in HIV-1. FEBS Lett. 584:359-365

315. Dismuke, D. J. and C. Aiken. 2006. Evidence for a Functional Link between Uncoating of the Human Immunodeficiency Virus Type 1 Core and Nuclear Import of the Viral Preintegration Complex. J.Virol. 80:3712-3720

316. Peng, C., B. K. Ho, T. W. Chang, and N. T. Chang. 1989. Role of human immunodeficiency virus type 1-specific protease in core protein maturation and viral infectivity. J.Virol. 63:2550-2556

317. Katoh, I., Y. Yoshinaka, A. Rein, M. Shibuya, T. Odaka, and S. Oroszlan. 1985. Murine leukemia virus maturation: protease region required for conversion from "immature" to "mature" core form and for virus infectivity. Virology 145:280-292

318. Kutluay, S. B. and P. D. Bieniasz. 2010. Analysis of the initiating events in HIV-1 particle assembly and genome packaging. PLoS.Pathog. 6:e1001200

319. Jouvenet, N., S. M. Simon, and P. D. Bieniasz. 2009. Imaging the interaction of HIV-1 genomes and Gag during assembly of individual viral particles. Proc.Natl.Acad.Sci.U.S.A 106:19114-19119

320. Carlson, L. A., M. A. de, H. Oberwinkler, A. Habermann, J. A. Briggs, H. G. Krausslich, and K. Grunewald. 2010. Cryo electron tomography of native HIV-1 budding sites. PLoS.Pathog. 6:e1001173

321. Campbell, S. and A. Rein. 1999. In vitro assembly properties of human immunodeficiency virus type 1 Gag protein lacking the p6 domain. J Virol 73:2270-2279

322. Lingappa, J. R., R. L. Hill, M. L. Wong, and R. S. Hegde. 1997. A multistep, ATP- dependent pathway for assembly of human immunodeficiency virus capsids in a cell-free system. J.Cell Biol. 136:567-581

323. Robinson, B. A., J. C. Reed, C. D. Geary, J. V. Swain, and J. R. Lingappa. 2014. A temporospatial map that defines specific steps at which critical surfaces in the Gag MA and CA domains act during immature HIV-1 capsid assembly in cells. J.Virol. 88:5718-5741

177

324. Singh, A. R., R. L. Hill, and J. R. Lingappa. 2001. Effect of mutations in Gag on assembly of immature human immunodeficiency virus type 1 capsids in a cell-free system. Virology 279:257-270

325. Cairns, T. M. 2000. Ph.D. thesis. Penn State University, College of Medicine, Hershey, PA.

326. Henning, M. S., B. N. Dubose, M. J. Burse, C. Aiken, and M. Yamashita. 2014. In vivo functions of CPSF6 for HIV-1 as revealed by HIV-1 capsid evolution in HLA-B27-positive subjects. PLoS.Pathog. 10:e1003868

327. Takeuchi, H. and T. Matano. 2008. Host factors involved in resistance to retroviral infection. Microbiol.Immunol. 52:318-325

328. Perron, M. J., M. Stremlau, M. Lee, H. Javanbakht, B. Song, and J. Sodroski. 2007. The human TRIM5alpha restriction factor mediates accelerated uncoating of the N-tropic murine leukemia virus capsid. J.Virol. 81:2138-2148

329. Gross, I., H. Hohenberg, and H. G. Krausslich. 1997. In vitro assembly properties of purified bacterially expressed capsid proteins of human immunodeficiency virus. Eur.J.Biochem. 249:592-600

330. Ehrlich, L. S., B. E. Agresta, and C. A. Carter. 1992. Assembly of recombinant human immunodeficiency virus type 1 capsid protein in vitro. J.Virol. 66:4874-4883

331. Nermut, M. V., P. Bron, D. Thomas, M. Rumlova, T. Ruml, and E. Hunter. 2002. Molecular Organization of Mason-Pfizer Monkey Virus Capsids Assembled from Gag Polyprotein in Escherichia coli. J.Virol. 76:4321-4330

332. Vogt, V. M. and M. N. Simon. 1999. Mass determination of rous sarcoma virus virions by scanning transmission electron microscopy. J Virol 73:7050-7055

333. Briggs, J. A., K. Grunewald, B. Glass, F. Forster, H. G. Krausslich, and S. D. Fuller. 2006. The mechanism of HIV-1 core assembly: insights from three-dimensional reconstructions of authentic virions. Structure 14:15-20

334. Scholz, I., B. Arvidson, D. Huseby, and E. Barklis. 2005. Virus particle core defects caused by mutations in the human immunodeficiency virus capsid N-terminal domain. J Virol. 79:1470-1479

335. Auerbach, M. R., C. Shu, A. Kaplan, and I. R. Singh. 2003. Functional characterization of a portion of the Moloney murine leukemia virus gag gene by genetic footprinting. Proc.Natl.Acad.Sci.USA 100:11678-11683

178

APPENDIX

IMPORTANCE OF THE NTD FLEXIBLE LOOPS ON MATURATION

179

INTRODUCTION

The CA NTD is dispensable for production of VLPs, but is a critical component of native virions (37,246,334). Biological evidence suggests that the flexible loop region of the NTD spanning the end of α4 through the beginning of α7 serves as a regulatory role in particle assembly and maturation (221). Deletion of segments of the flexible loops results in misregulation of particle size (247). Several point mutations in the flexible loop arose when viruses with β-hairpin mutations were passaged in cell culture (221). Flexible loop mutations result in inefficient budding and production of virions with heterogeneous size (37,234,335). In addition, a variety of host cell proteins are known to bind to the NTD and provided both beneficial and inhibitory functions to the virus.

Recent higher resolution structural modeling provided support for the biological evidence of importance of the flexible loop region. Subtomogram averaging of tubes made of the CA-NC region of M-PMV provided insight into immature interactions between the CA domains of Gag molecules (187). Modeling of HIV-1, RSV, and HTLV-1 into the same density showed similar interhexameric interactions between the flexible loops. The interactions seen in the immature particle are distinct from those in the mature capsid and the contribution of individual residues is still not entirely clear (186,187).

A systematic examination of the contributions of flexible loop residues was undertaken by our lab. Specific residues within the RSV flexible loop were chosen for their conservation among retroviruses and ability to correct an infectivity defect as a result of β-hairpin mutation.

Nearly all of the mutant viruses tested had some level of infectivity defect. In particular D87A,

E99A, and K107A had a significant decrease in viral infectivity despite near WT levels of particle release and presence of normal looking virus particles as seen by EM. Mutation D87E caused a significant inhibition of virus infectivity despite being a conserved mutation. After several serial passages on DF1 cells, the suppressor mutation A134V was discovered. A134 is located on α7 quite distant from D87. Based on the Bharat model of immature assembly interfaces the A134

180 position contributes to the same interhexameric interface that D87 does, providing some insight into the mechanism by which A134V can correct the D87E defect (187). In the present study, we furthered the previous observations by analyzing the quality of in vitro assembled Gag particles as well as assessing the production of capsid-like particles in vitro.

RESULTS AND DISCUSSION

Analysis of in Vitro Assembled ΔMBDΔPR Particles

We previously demonstrated that the flexible loop mutations had little to no effect on in vitro assembly of the truncated ΔMBDΔPR Gag protein. K107R, D87E, and E99A all looked indistinguishable from the WT protein – double ringed structures about 70 nm in diameter

(39,284). K107A assembled particles of the same size and shape as WT but the outer ring of the assembled complex appeared fuzzier than the other proteins suggestive of some disorder in the outer shell. To further investigate the organization of the assembled ΔMBDΔPR particles we employed radial density averaging from cryo-EM images. Assembled complexes of WT, K107A, and K107R ΔMBDΔPR protein were absorbed onto holey carbon grids, flash-frozen, and imaged by cryo-EM (Figure A.1). Particles were assessed for their completeness and detectable rings of protein. From each subset of data, the twelve most uniform particles were chosen and their radial density was averaged and plotted (Figure A.2).

All three proteins assembled into structures that allowed identification of density associated with the NC/oligo, CTD, and NTD segments of the ΔMBDΔPR protein. The NTD peak for all was broader than that of the CTD due to the larger size of the NTD. This feature has been noted previously for the ΔMBDΔPR protein and native immature virus particles. The

K107R density was shifted to the right by about 1 nm suggesting that it forms slightly larger particles than the WT protein. Both WT and K107R produced particles with NTD density peaks taller than the CTD. This was different that what has been reported for in vitro assembled

ΔMBDΔPR particles, which have NTD and CTD peaks of equal height, but similar to that of in

181

K107A

K107R

Figure A.1. Cryo-EM micrographs of ΔMBDΔPR proteins. RSV ΔMBDΔPR Gag proteins were assembled with a DNA oligonucleotide for 30 min. The resulting particles were absorbed onto holey grids and frozen. Micrographs were analyzed by EMAN and selected for their quality based on size, completeness, and distinct striation of protein. Representative particles are shown.

182

Figure A.2. Radial density averaging of RSV ΔMBDΔPR particles. Twelve particles of WT and each mutant ΔMBDΔPR proteins were chosen based on their uniformity. The average radial density was plotted as a function of the particle radius. Three distinct peaks were identified representing the NC domain, the CTD, and NTD.

183 vivo immature particles. The K107A protein resembled that of previously published WT protein with NTD and CTD peaks of roughly equal height. The significance of these findings is not entirely clear. However, these data do suggest that there are differences between the WT protein and K107A, possibly due to an inherent disorder in the NTD brought out by the K107A mutation.

In Vitro Assembly Properties of the D87E/A134V Protein

Assembly of the D87E and D87E/A134V ΔMBDΔPR proteins was indistinguishable from the WT protein, suggesting that the defect of D87E was not in making immature particles.

To investigate if the D87E mutation perturbed mature capsid assembly, we engineered D87E,

A134V, and D87E/A134V into the CA expression construct. Sodium phosphate was used to induce assembly of the purified proteins (Figure A.3). All three proteins assembled with near WT kinetics, with the A134V protein having a slight delay. D87E assembled small (~17 nm) and large (~32 nm) spheres like the WT protein (228,241,250). Both A134V and D87E/A134V assembled the small spheres, albeit more rarely than the WT protein. They also had a tendency to form tubular assemblies of CA as has been occasionally seen for the WT protein. These results indicate that the D87E mutation does not provide an inherent defect in assembly of the mature capsid. Therefore the defect seen must be a result of an interaction that cannot be replicated by our in vitro models. Whether the interaction occurs within the capsid or between the capsid and a host protein, we cannot conclude.

Given ability of the ΔMBDΔPR and CA proteins with the D87E and K107A substitutions to assemble there does not appear to be a problem with the proteins themselves. However, the flexible loop region of the NTD is a particularly dynamic region of the protein has the potential for forming a variety of Gag-Gag, CA-CA, and CA-host protein interactions. All of these contacts must occur in a coordinated fashion to allow the virus to mature properly. Based on previous work and the data presented here, it is likely that the mutant proteins are able to form immature particles, but are unable to go through the final stages of maturation leading to a fully functional capsid. This could be due to a failure to bind particular host factors, such as CypA-like molecule,

184

Figure A.3. In vitro assembly of the flexible loop mutants. Purified CA proteins bearing the

D87E, A134V, and D87E/A134V mutations were assembled with sodium phosphate. Turbidity of the solutions was monitored at 450 nm (upper panel). The resulting particles were stained with 1

% uranyl acetate and examined by TEM (lower panels).

185 that would help the virus mature properly. It is also possible that the conformational changes that must occur between the immature and mature capsid are disrupted by the NTD mutations. This possibility is not replicated in our in vitro system. Our current work supports models that point to the NTD flexible loop as a critical region for capsid maturation. Further investigation is needed to fully elucidate the mechanism(s) by which the flexible loop region acts.

ACKNOWLEDEMENTS

We would like to thank Tam-Linh Nguyen and Dr. Katrina Heyrana for initiating studies on the flexible loop regions of the NTD. We are also appreciative of Dr. Maria Bewley for her assistance with mutagenesis and helpful discussions of the results. Bob Ashley, Hyunwook Lee, and Dr. Susan Hafenstein are acknowledged for their assistance with cryoEM data collection and processing. We are grateful to Roland Meyers for his help with electron microscopy.

186

VITA

Matthew Raymond England

Education

Ph.D. in Microbiology and Immunology 2008-2015 The Pennsylvania State University, College of Medicine Hershey, Pennsylvania

B.S. in Biochemistry, magna cum laude 2004-2008 University of Delaware, College of Arts and Sciences Newark, Delaware

Honors and Awards

NIH Training Grant: T32 CA060395 2011-2013 The Pennsylvania State University, College of Medicine

ASV Travel Grant for Oral Presentation July 2011 American Society for Virology, University of Minnesota

Karl H. Beyer Jr. MD/PhD Award for Highest GPA October 2009 The Pennsylvania State University, College of Medicine

Penn State Graduate Fellowship 2008-2011 The Pennsylvania State University, College of Medicine

Phi Beta Kappa Honor Society May 2008 University of Delaware

Publications

England MR*, Purdy JG*, Ropson IJ, Dalessio PM, and Craven RC. Potential role for CA-SP in nucleating retroviral capsid maturation. 2014. J. Virol. 88:7170-7177.

Colella R, Lu G, Glazewski L, Korant B, Matlapudi A, England MR, Craft C, Frantz CN, Mason RW. Induction of Cell Death in Neuroblastoma by Inhibition of Cathepsins B and L. 2010. Cancer Lett. 294:195-203.

Professional Membership

American Society for Microbiology, student member since 2013

American Society for Virology, student member since 2010