1 BIOPHYSICAL CHARACTERIZATION OF THE CELLULAR INTERACTIONS OF ADENO ASSOCIATED VIRUS SEROTYPE 2 By LAWRENCE J. TARTAGLIA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Lawrence J. Tartaglia
3 To my mom, dad, and sister
4 ACKNOWLEDGMENTS First and foremost I would like to thank my parents; Larry and Alice, and my sister, Kristen for all of their love, support, and encouragement. They have given me the strength and passion to pursue my academic dreams and have molded me into the person I am today. Beyond my immediate family, I am extremely grateful to my G randmother, A unts Sue, Mary, Sandy and Helen, and U ncles Gip, Don, and Bill for believing in me. Also to my best friend, Frank who has taught me to enjoy life to the fullest. I am genuinely honored to have studied under the tutelage of Dr. Mavis AgbandjeMcKenna who is the epitome of a mentor. She has inspired me to reach my full potential as a scientist and a person. This degree would have not been possible without her passion and commitm ent to my intellectual growth and I am forever grateful. I would also like to thank Dr. Nicholas Muzyckza for his mentorship and allowing me to complete most of my studies in his laboratory. I thank Dr. Robert McKenna for his insightf ul conversations and t eaching me to think outside of the box. My other committee members, Dr. James B. Flanegan, Dr. Arun Srivastava, and Dr. Peter Saye ski have been valuable resources and have provided critical input in the development of the research project. All member s of t he McKenna lab and Muzyczk a lab past and present have helped along the way. A special thanks to my classmates Lauren Drouin, Dr. Robert Ng, and Dr. Balasubramanian Venkatakrisnan for all of your support; Dr. Antonette Bennett, Dr. Sujata Halder, Dr. Chen Ling, and Paul Chipman for teaching me techniques used in the la b. I would also like to thank undergraduate students that I have taught over the years, especially Andrew Woodhouse, Roman Kazakov, and Alex Plattner.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES .......................................................................................................... 9 LIST OF ABBREVIATIONS ........................................................................................... 12 ABSTRACT ................................................................................................................... 14 CHAPTER 1 BACKGROUND AND INTRODUCTION ................................................................. 16 Parvoviridae ............................................................................................................ 16 AdenoAssociated Viruses (AAVs) ......................................................................... 16 Parvoviridae Capsid Structure ................................................................................ 19 AAV2 Receptor Interacti ons and Endocytosis ........................................................ 20 AAV Immune Response.......................................................................................... 23 Defensins ................................................ 25 AAV2 and antibody interactions ....................................................................... 25 Significance of Virus R eceptor and Immune Response Interactions ...................... 26 2 MATERIALS AND METHODS ................................................................................ 34 TriSystem Vector ................................................................................................. 34 ................................................................... 34 SmallScale Protein Expression/Time Course Analysis ................................... 35 Large Scale Protein Expression and Purification via Nickel Column Chromatography ........................................................................................... 36 Co ......................................................... 37 ................................... 38 .................................................................... 38 Generation of Recom binant Baculoviruses ...................................................... 39 Small............................................. 40 LargeScale Protein Expression and Purification via Nickel Column Chromatography ........................................................................................... 41 Co ........................................ 42 Purif ication of Baculovirus AAV2 VLPs ................................................................... 43 Cell Culture and Virus Infection ........................................................................ 43 Virus Purification .............................................................................................. 43 Virus Buffer Exchange and Concentration ........................................................ 44 A20 IgG and Fab Purification .................................................................................. 45
6 A20 IgG Purification ......................................................................................... 45 A20 Fab Purification ......................................................................................... 45 SDS PAGE Analysis ............................................................................................... 46 Western Blot Analysis ............................................................................................. 47 Nat ive Dot Blot ........................................................................................................ 47 Mass Spectrometry ................................................................................................. 47 Trypsin Digestion of ..................................... 47 LC MS/MS ....................................................................................................... 49 Protein Search Algorithm for Peptide Mass I dentification ................................. 49 Transduction Inhibition Assays In Vitro ................................................................... 50 Surface Plasmon Resonance ................................................................................. 51 ..................... 52 ........... 52 d AAV1, AAV2, and AAV5 ............................................................................................................. 53 ......................... 54 AAV2 A20 IgG SingleCyle Kinetics ................................................................. 54 AAV2 A20 IgG Multi Cycle Kinetics .................................................................. 55 AAV2 A20 Fab Multi Cycle Kinetics ................................................................. 56 Negative Stain Electron Microscopy ....................................................................... 56 ............................................................... 57 ............................................................................ 57 .......................................... 57 ............................................. 57 Defensins 1, 2, and 5 ...................................... 58 Cryo Electron Microscopy ....................................................................................... 58 Cryo EM Data Collection .................................................................................. 58 Three Dimensional Reconstruction .................................................................. 59 Fitting of Pseudo Atomic Models into CryoEM Density Maps ......................... 60 3 PURIFICATION, AND INTERACTION WITH AAV2 ....................................................................................................................... 76 Background ............................................................................................................. 76 Results and Discussion ........................................................................................... 78 Construct Development Rationale .................................................................... 78 ........ 78 ....................................................... 80 Baculovirus A AV2 Purification and Characterization ........................................ 81 Transduction Inhibition Assay ........................................................................... 82 Surface Plasmon Resonance ........................................................................... 82 ............................. 83 Summary ................................................................................................................ 84 4 AAV2 ....................................................................................................................... 96
7 Background ............................................................................................................. 96 Results and Discussion ........................................................................................... 9 7 ...................................................................................... 97 SmallScale Protein Expression, Purification, and LC ............................................................................... 98 LargeScale Protein Expression, Purification, and Dot Blot Characterization ... 99 Co ...................................... 100 Transduction Inhibition Assays ....................................................................... 100 Surface Plasmon Resonance ......................................................................... 101 ...................... 104 Stain EM ...................................................................................................... 105 Analysis of AAV2 EM ............................... 106 S ummary .............................................................................................................. 108 5 AAV2 AND THE IMMUNE RESPONSE ................................................................ 135 Background ........................................................................................................... 135 Results and Discussion ......................................................................................... 136 Defensins 1, 2, and 5 .............................. 136 Defensins 1, 2, and 5 and Visualized by Negative Stain EM ...................................................................................................... 137 Analysis of AAV 2 HD5 Complexes by CryoEM ............................................. 138 A20 IgG and Fab Purification ......................................................................... 140 AAV2 and A20 IgG/Fab Interactions Analyzed by SPR .................................. 141 Sum mary .............................................................................................................. 144 6 SUMMARY AND FUTURE DIRECTIONS ............................................................ 162 Summary .............................................................................................................. 162 ................. 162 .................................................. 163 AAV2 Interactions with defensins and A20 Mab and A20 Fab ........................... 165 LIST OF REFERENCES ............................................................................................. 169 BIOGRAPHICAL SKETCH .......................................................................................... 186
8 LIS T OF TABLES Table page 1 1 Adenoassociated virus receptors ...................................................................... 28 1 2 Viral receptors utilized by both AAV2 and other selected families ...................... 29 2 1 Jun genes ............................................... 62 2 2 Fos and Jun genes and a 8X histidine tag ............ 62 2 3 Ratios of AAV2 with integrins and defensins analyzed by transduction inhibition ............................................................................................................. 63 2 4 resonance ........................................................................................................... 64 2 5 Ratios of AAV2 and A20 Mab and A20 Fab analyzed by Surface Plasmon Resonance ......................................................................................................... 65 2 6 Ratios of AAV2 with integrins and defensins analyzed by negative stain EM and cryoEM ....................................................................................................... 66 2 7 Statistics for cryo EM reconstructions ................................................................ 67 5 1 scAAV2 EGFP transduction inhibition by HNP1, HNP2, and HD5. .................. 146
9 LIST OF FIGURES Figure page 1 1 Gene organization and transcription scheme for Adenoassociated virus 2 (AAV2).. .............................................................................................................. 30 1 2 AAV2 l ife cycle depicting cell surface attachment, endocytosis, intracellular trafficking, nuclear entry and the immune response. ......................................... 31 1 3 AAV2 variable regions and secondary structure assignment. ............................ 32 1 4 AAV2 VP3 trimer with highlighted heparinbinding sites and N GR motif. ........... 33 2 1 Schematic flow chart of integrin, AAV2, and A20 production and purifications.. ...................................................................................................... 68 2 2 A20 Fab purification scheme. ............................................................................. 69 2 3 Surface plasmon resonance. .............................................................................. 70 2 4 ........................... 71 2 5 ................................................ 72 2 6 A single cycle A20 IgG kinetics experiment. ....................................................... 73 2 7 A multi cycle A20 IgG kinetics experiment. ........................................................ 74 2 8 Schematic flow chart of the cryoEM reconstruction process. ............................ 75 3 1 Diagram of plasmids used for expression ................................ 85 3 2 PCR amplification of cDNAs and verification of assembled plasmids by restriction enzyme digestion. .............................................................................. 86 3 3 .......................... 87 3 4 Nickel column purification of HEK293expressed integrin. ................................. 88 3 5 Tandem MS/MS verification of expressed integrin domains. .............................. 89 3 6 Characterization of integrin V 5 using a conformational anti V 5 antibody. .. 90 3 7 A chromatogram showing peak fractions of an AAV2 purification via ion exchange chromatography. ................................................................................ 91 3 8 I on exchange purification of SF9expressed AAV2.. .......................................... 92
10 3 9 Comparative analyses of AAV2mediated transduction of HeLa cells with V 5 integrin inhibition. ...................................................................................... 93 3 10 Resonance. ........................................................................................................ 94 3 11 Negative stain EM of AAV2 and integrins. .......................................................... 95 4 1 Diagram of plasmids used .............................. 111 4 2 integrin D NA. .................................................................................................... 112 4 3 Small......................................................................................... 113 4 4 Tandem MS/MS verification of expressed integrin domains. ............................ 114 4 5 Largeintegrin. ............................................................................................................. 115 4 6 Co ........................................ 116 4 7 Comparative analyses of AAV2mediated transduction of HeLa cells with .................................................................................... 118 4 8 Resonance. ...................................................................................................... 119 4 9 AAV1, AAV2, and AAV5 analyzed by Surface Plasmon Resonance. .......................................................................... 122 4 10 M. ........................ 127 4 11 EM. ................................................................................................................... 128 4 12 linking experiments visualized by negative stain EM. .......................................................................................................... 129 4 13 Three dimensional icosahedral reconstruction of AAV2 complexed with ............................................................... 130 4 14 Three dimensional icosahedral reconstruction of AAV2 complexed with ............................................................... 132 4 15 Three dimensional asymmetric reconstruction of AAV2 complexed with ............................................................... 134
11 5 1 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 1 inhibition. ......................................................................................... 147 5 2 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 2 inhibition. ......................................................................................... 148 5 3 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 5 inhibition. ......................................................................................... 149 5 4 defensin 1, 2, and 5 complexed with AAV2 and visualized by negative stain EM. ................................................................................................................... 150 5 5 Three dimensional icosahedral reconstruction of AAV2 complexed with defensin 5. ........................................................................................................ 151 5 6 Diff erence map calculations between AAV2defensin 5 and AAV2alone density maps. ................................................................................................... 153 5 7 Cross sections of AAV2 defensin 5 and AA V2 alone density maps. ............. 155 5 8 AAV2 and defensin 5 binding and structural rearrangements. ...................... 156 5 9 A20 IgG and Fab purification. ........................................................................... 158 5 10 AAV2 A20 IgG singlecycle kinetics. ................................................................ 159 5 11 AAV2 A20 IgG multi cycle kinetics. .................................................................. 160 5 12 AAV2 A20 Fab multi cycle kinetics. .................................................................. 161
12 LIST OF ABBREVI ATIONS AAVS a denoassociated virus es AAV2 adenoassociated virus serotype 2 AAP assembly activating protein AD2 adenovirus serotype 2 CAR coxsackie adenovirus receptor CD cell differentiation CME clathrin mediated endocytosis CPV canine parvovirus CRYO EM cryo reconstruction EDTA ethylene diaminetetr acetic acid EGFR epidermal growth factor receptor EM electron microscopy FAB f ragment antibody FGFR1 fibroblast growth factor receptor 1 FPV feline panleukopenia virus FSC fourier shell correlation HADV human adenovirus HD5 defensin 5 HELA Henrietta Lacks cells HEK human embryonic k idney HGFR hepatocyte growth factor receptor HSPG heparan sulfate proteoglycan ICBR Interdisciplinary Center for Biotechnology Research
13 ITR inverted terminal repeats LAMR laminin r eceptor LC MS/MS liquid chromatography mass spectrometry mass spectrometry MAB monoclonal antibody MHC major histocompatability complex MOI multiplicity of infection MVM minute virus of mice ORF open reading frame P2 passage 2 PAMPS pathogenassociated molecular patterns PCR polymerase chain reaction PDGFR platelet derived growth factor receptor PFASTBAC1 baculovirus transfer vector PI isoelectric point PLA phospholipase PR primer PRRS pattern recognition receptors SCAAV2 EGFP self complementary AAV2 enhanced green fluorescent protein SF9 S podoptera frugiperda cells SPR surface plasmon resonance VLPS virus like particle s VP viral protein VP1U viral protein 1 u nique region
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BIOPHYSICAL CHARACTERIZATION OF THE CELLULAR INTERACTIONS OF ADENO ASSOCIATED VIRUS SEROTYPE 2 By Lawrence J. Tartaglia August 2013 Chair: Mavis Agbandje McKenna Major: Medical Sciences Biochemistry and Molecular Biology AdenoAssociated Viruses (AAVs) are currently among the most frequently used gene delivery vectors for gene therapy. Due to their nonpathogenicity, long term transgene expression, and ability to transduce non dividing and dividing cells, these viruses have attracted significant interest in the gene therapy field. T herefore, understanding the AAV life cycle determinants particularly cellular entry and the immune response, is critical for the rational design of new AAV vectors. Towards discerning these determinants, AAV2 was employed as a model for int receptors and also immune response effectors such as defensins 1,2, and 5, and A20 IgG and A20 Fab antibodies. The results show ed that interact with the AAV2 capsid in an asymmetric and transient manner as analyzed by transduction inhibition, surface plasmon resonance, and cryo electron microscopy. defens ins (1, 2, and 5) inhibit ed AAV2 by transduction inhibition assays, while cryo electron microscopy results show ed defensi n 5 b inding to the AAV2 capsid induces structural rearrangements around the twofold and fivefold axes of symmetry. Th ese preliminary data support the hypothesis defensin 5 may stop VP1u extrusion through either the twofold axis or fivefold
15 channel and thus prevent AAV2 endosomal escape during cellular trafficking. Lastly, surface plasmon resonance analyses showed A20 IgG ( Kd = 11 nM) has a 1000fold stronger binding affinity for the AAV2 capsid than A20 Fab (Kd = 2 M). These data suggest that A20 binding to the AAV2 capsid occurs in either a bivalent nature where both IgG arms bind to the capsid or a monovalent nature where both arms bind two individual capsids. The studies presented here show the biophysical interactions of AAV2 with host cellsurface r eceptors and immune response molecules
16 CHAPTER 1 BACKGROUND AND INTRODUCTION Parvoviridae The Parvoviridae family consists of small, non enveloped viruses that contain linear, single stranded DNA genomes. The family Parvoviridae contains two subfamilies: Parvovirinae and Densovirinae which infect vertebrates and invertebrates, respectively. The subfamily Parv ovirinae is further subdivided into five genera: Amdovirus, Bocavirus, Dependovirus, Erythrovirus, and Parvovirus All of the genera excluding dependoviruses are autonomous viruses that do not require a helper virus coinfection for a productive infection; however, factors expressed during cell cycle S phase are required for replication. On the contrary, dependoviruses also require helper viruses (e.g. adenovirus, herpesvirus) which provide beneficial functions that prevent concatamers and apoptos is and als o promote viral replication and gene expression (1). The most studied dependoviruses have been the Adenoassociated viruses (AAVs), particularly for their utility for gene delivery. Adeno Associated Viruses (AAV s) AAV was first discovered in 1965 as a contaminant in an Adenovirus slide preparation (2), and was later grouped under the Parvoviridae family of viruses. To date, 12 distinct human and nonhuman primate AAV serotypes (AAV112) and ov er 100 AAV variants have bee n identified and sequenced (2 11) By definition, a serotype is a viral isolate that cannot cross react with neutralizing antibodies for all of the other exis ting characterized serotypes (11) However, not all AAVs represent true serotypes since antibodies to AAV1 and A AV6 crossreact (5, 7, 12) and there is limited inform ation on the serological profiles of AAV10, AAV11 and AAV12 (3, 10) Therefore,
17 the AAVs are classified into different genetic clades (A F) and clonal isolates (e.g. AAV4 and AAV5), whose members share antigenic cross reactiv ity and have sequence similarities (7). The AAVs are not associated with any disease, infect a wide range of cell types, and have the ability to establish long term transgene expression, all traits that make these viruses attractive gene delivery vectors. A number of studies have utilized pseudotyping strategies that package the genome and ITRs of one AAV serotype into capsids of other serotypes and com pare the transduction efficiencies of serotypes of AAV vectors in different tissues (8, 10, 13 16) For example, AAV2, the most studied serotype, is known to transduce a wide range of tissue types including liver, muscle, lung, kid ney, and the central nervous system. Interestingly, AAV9 shows a similar transduction profile with a muc h higher efficiency than AAV2 (7). AAV1 and AAV7 show rapid onset and transduce skeletal muscle with up to 3 orders of magnitude higher efficiency than AAV2 (7, 8) A complete list of the hierarchy of transduction efficiency in major tissues by AAV serotype vectors is reviewed (11) AAVs have a linear, singlestranded 4.7 kb DNA genome that packages either sense or anti sense DNA strands with equal frequency. T he genome has three open reading frames (orf), rep, cap, and aap flanked by inverted terminal repeats (ITR) (Figure 1 1). The left orf encodes for four overlapping replication proteins Rep78, Rep 68, R ep 52, and R ep 4 0 that exhibit nicking and helicase activity, as well as package the AAV genome. The right orf encodes the vir al capsid proteins, VP1, VP2, and VP3 in proportions of 1:1:10, respectively (17) The entire capsid orf is encoded in VP1, while VP2 and VP3 are alternatively spliced mRNAs. VP2 is alternatively spliced from an
18 alternate upstream codon, ACG, compared to VP3 that gets spliced from a conventional downstream ATG. A total of 60 subunits of VP1:VP2:VP3 are assembled into T=1 icosahedron virion shells. Recently an additional protein, the assembly activating protein (AAP) encoded by the aap orf was discovered w ith a cryptic start site in VP2 (18) AAP has been shown to transport the unassembled VPs into the nucleolus for capsid assembly. ITRs flank all three orfs and are implicated as the only cis acting elements required for geno me replication and packaging (19) Recombinant AAV vectors are constructed with ITRs and the internal wt AAV coding sequences are removed and replaced with reporter or ther apeutic genes (19 21) These recombinant viruses infect both dividing and nondividing cells, and establish l atency in nondividing cells (22) Recent attention in AAV2 biology has focused on elucidating and understanding the virus internalization and trafficking pathways (Figure 1 2). AAV2 transduction commences with cell surface attachment to its primary receptor, heparan sulfate proteoglycan (HSPG) (23) followed by coreceptor mediated endocytosis in dynamin depend ent clathrincoated pits (23 28) However, AAV2 has also been reported to transduce CD9 cells in low HSPG conditions (29) and infect HEK293 cells in a clathrin independent manner (30) Researchers have shown that endosomal acidity is essential for viral trafficking since vacuolar H+ATPase inhibitor s (e.g. bafilomycin A1) reduce both transduction and trafficking to the nucleus (31 35) The acidic environment of the endosome is reported to externalize the N termini of both VP1 and VP2, which activates a phosophlipase A2 (PLA2) activity in VP1, and nuclear localization sig nals in both VP1 and VP2 (36 40) Following endosomal escape, AAV2 is transported along micr otubules toward the nucleus (35) however, the mechanism of genome uncoating is unclear. In
19 the presence of a helper virus, DNA replication, viral gene expression, and capsid assembly occur, which results in infectious AAV progeny. In the absence of a helper virus, viral gene expression is repressed and wt AAV2 integrates and establishes latency in chromosome 19, until a helper virus rescues the integrated genome (41, 42) Presently, the AAVs are the only known mammalian DNA virus es that have the ability to undergo sitespecific integration. Parvoviridae Capsid Structure Parvoviruses are among the smallest vir uses with a diameter of ~260 All Parvoviruses have 60 copies (in total) of viral proteins that are involved in capsid assembly; however, viral protein ratios vary among individual viruses (43) Parvovirus capsids have been implicated in a number of functional roles that maintain virus survival. These roles include host cell recognition and entry, intracellular process ing, genome delivery, uncoating, assembly and release of progeny virus. Thus, a complete understanding of the capsid structure is essential for correlation to function. A number of Parvoviridae struct ures have been solved by X ray c rystallogr aphy and cryoelectron microscope and image reconstructions (c ryo EM). Some of these studies include Minute Virus of Mice ( MVM) (44, 45) Feline Panleukopnia Virus (FPV) (46) Human P arvovirus B19 (47) Canine Parvovirus (CPV) (48) Porcine Parvovirus (49) H1 P arvovirus (50) Aleutian Mink D isease V irus (AMD V ) (51) Human B ocavirus (52) and to date, seven AAV serotype s (AAV2, AAV3b, AAV4, AAV5, AAV6, AAV8 and AAV9 ) (53 60) structures have been report e d In addition, the crystallization of several other AAV serotypes (AAV1, AAV3B AAV4, AAV5, AAV6, AAV7, AAV9) has been reported (58, 6064) Numerous com monalities exist among the capsid topology of the Parvoviridae family of viruses; however, the AAV2 capsid will serve as the model for the family for
20 further capsid explanations. The VP structure consists of a core eight stranded t o the C terminus) anti barrel a conserved helix, with surface loops in between the strands (Figure 13 A ). The loops also contain small strand structure. The loops are named after the strands between which they are inserted. F and the HI loop is inser The tops of these loops are structurally varied among other AAV serotypes, and they dictate host specific tropism, antigenic response, and transduction efficiency in viral vectors (65) Figure 13B shows the surface loops (Figure 13A) of an assembled ca psid from 60 VPs. The twofold axis of symmetry is characterized by depressions (Figure 1 3C) The threefold axis of symmetry has three se parate protrusions surrounding depressions The fivefold axis contains a cylindrical channel strands contains a central pore encircled by a canyon like depression (Figure 1 3A and C) AAV 2 Receptor I nteraction s and E ndocytosis Viruses must bind cell surface receptors in order to gain entry, deliver genome, and produce progeny virus in host cells. The first line of defense that viruses have to overcome is the glycocalyx barrier that surrounds and protects cells. This layer is composed of glycoproteins and glycolipids that viruses bind to and use as either attachment factors or true receptors. For example, as mentioned above, HSPG has been identified as a primary receptor for AAV2 (23) as well as other AAV viruses includin g AAV3 (67) and AAV6 (68) the closely related (by cellular internalization pathway) Adenovirus serotypes 2 and 5 (HAdv) (69) and other non related viruses such as Denguevirus (70) and Human Cytomegalovirus (HC MV) (71) See T able 11 for a complete list of receptors utilized by the AAVs. Upon cell surface attachment, some
21 viruses utilize coreceptors for internalization depending on t he cell type. Note when viruses use multi ple receptors for productive entry, it is convention to call the one contacted first the receptor or primary receptor and subsequent ones coreceptors (72) AAV2 has been shown to effectively bind a number of coreceptors: integrins (23, 26) hepatocyte growth factor receptor (HGFR) (24) fibrobla st growth factor receptor 1 (FGFR1) (25) and laminin receptor (27) Table 12 shows other selected viral families that utilize the same class of receptors as AAV2. Structural and mutational analyses have demonstrated that parvovirus host tropisms and transduction properties of AAV vectors have arisen from variations in surface amino acids (43) In AAV2, mutagenesis studies identified amino acids responsibl e for AAV2s HSPG phenotype (73) and more recently, a cryo EM reconstruction of AAV2 complexed with HS allowed for the mapping of th e A AV2 heparin binding site (74, 75) In the AAV2 HS complex structure reported by Levy et al (74) heparin density was shown proximal to the basic residues (Figure 1 4) at the threefold protrusion and additional structural rearrangements occurred at the twofold and fivefold axes compared to the wt AAV2 no HS reconstruction. However, these additional structural rearrangements were not seen in a si milar AAV2 HS PG cryo EM reconstruction study by ODonnell et al (75) These conformational change findings are significant since the channel at the fivefold axis is postulated as the site of externalization of the P LA2 activity in parvoviruses (76) (77, 78) and has been implicated in the insertion and release o f viral DNA during assembly (36) and infection (79) respectively. It has been shown that in vitro treatment of early endosomes with acidic pHs induces extrusion of the VP1/2 N terminus in CPV (80) and MVM (40) but
22 not AAV2. This suggests that some combination of factors including receptor /co receptor binding and pH is necessary for VP1/2 extrusion and therefore productive infection in AAV2 In addition to the fivefold structural rearrangement, the threefol d axis undergoes remodeling (74) Interestingly, the proposed integrinbinding motif (NGR) sits near the threefold axis of symmetry (26) a site that may be more exposed following HS binding. These observations of the HSPG induced structural rearrangements on the capsid surface raise a few questions regarding integrin binding. Is the interaction of AAV2 HS PG a prerequisite for integrin binding? Where is the integrin binding site on the AAV2 capsid surface? Does integrin binding further remodel the AAV2 capsid surface for productive endocytosis? So far, the integrins, and other coreceptor binding sites on the AAV2 capsid have not been structurally mapped. To date, only a sel ect number of virus integrin structural studies have been reporteds bind to and cluster at RGD motifs on the penton base, which lead to intracellular signaling events required for virus internalization (53) In Echovi s to fivefold symmetry axes, induce integrin clustering, and commit the virus rec eptor complex to cell entry via the caveolar pathway (54) Other integrinintegrin Coxsackievirus A9 show a high affinity interaction (low nanomolar), yet no structural rearrangements occur and the integrins are primarily used for cell surface attachment, not internalization the Parechovirus1 capsid with high affinity (56), but its interaction is more transient as confirmed by Surface Plasmon Resonance (SPR) These virus integrin interactions raise an additional question. Does AAV2 have a stronger affinity for one integrin
23 ntegrin) over the other? This project will focus on characterizing the interaction of integrins with AAV2, including the identification of their capsid binding site. AAV Immune R e s ponse AAV vectors are vehicles of choice for in vivo gene transfer as they t ransduce a wide variety of tissue types and mediate long term transgene expression after a single administr ation (82) A large number of animal studies establish AAV vectors as a potential therapeutic to ol for gene delivery in humans (14, 83, 84) However, translation of animal model results to clinical studi es in humans revealed limitations of these predictions (85) These differences arise from mechanisms in which the AAV capsids are recognized and inhibited by the innate and humoral immune system. Recent studies have shown that the innate immune system plays a critical role in the initial response to the AAV vector and in stimulating a deleterious humoral immune response (82) A 2006 trial using AAV2 to transfer blood coagulation factor IX to the livers of hemophilia B patients deficient for factor IX resulted i n a short term transgene expression for two weeks This was followed by a decline in transgene expression and a rise in liver enzymes, which is indicative of the destruction of hepatocytes. The hepatocyte destruction was most likely due to C D 8+ T cell responses against the AAV capsid (82) It was later discovered that input capsidderived peptides were presented by MHC class I molecules on the surface of hepatocytes following transduction with AAV2 vectors (86) Another study showed that the innate immune system recognizes nonself unique structural motifs which are referred to as pathogenassociated molecular patterns (PAMPs), via pattern recognition receptors (PRRs) (87) Other reports showed that responses to AAV vectors are mediated by activating proinflammatory cytokines that activate the NF kB pathways and inducing type I IFN
24 production (86) Additional studies show ed AAV serotypes are targeted by the innate immune response in different manners. For example, b oth AAV1 and A AV8 transduce dendritic cells. However, a report showed the AAV1 capsid was more imm un ogeni c and less tolerated when delivered to NOD mice compared to AAV8 (88) A similar study showed AAVrh32.33, an evolutionary divergent isolate from rhesus macaques, had reduced transgene expression and strong anti transgene CD8+ T cell responses compared to AAV8 when injected into skeletal muscle (89) The enhanced responses was due t o increased levels of CD4+ help er T cells, plus CD40L and CD28, which are involved in immune system activation. The humoral immune response to the AAV capsid also poses a large obstacle in gene therapy. The presence of circulating Ig G antibodies to the AAV capsid is a limitation in the efficacy of gene delivery (90) Studies have shown that these ant ibody responses to the AAV capsid prevent repeated administrations (91) and even low levels of neutralizing antibodies can abrogate AAV transductions (92) Other work demonstrated that AAV2 capsidprimed memory CD8+ T cells in humans can cross react with epitopes from AAV1 and AAV8 and result in expansion of CD8+ T cells (82) Interestingly, AAV8 has been isolated from nonhuman primates, but preexisting immune reponses in humans against it suggest s crossreactivity between them (7). Three rece nt report s have examined the poorly understood antigenic epitopes on AAV capsid by cryo EM (93 187188 ) Monoclonal fragment antibodies (Fabs ) against AAV1, AAV2, AAV6 and AAV8 were generally identified to bin d around t he top or the side of the three fold protrusions, while for AAV2 and AAV5 the binding footprint s extended from the protrusion towards the fivefold channel (93 and 188) Interestingly,
25 the antibody binding footprints for a couple of the Fabs, C37B and 4E4 overlap overlap with known receptor attachment points on the capsid surface, such as the HSPG site on AAV2 and AAV6 respectively (73, 74) thus suggesting steric inhibition of receptor attachment as possible mechanisms of neutralization. A AV2 and its I nter actions with the D efensins The defensins are 2 6 kDa peptides that are effectors of the innate immune response that are active against bacteria, fungi, and viruses (95) However, their role in antiviral immunity is poorly understood, particularly for nonenveloped viruses. A recent report showed defensins 1 and 5 block endosomal escape in HadV5, but receptor binding and virus internalization are unaffected (96) In addition to adenoviral interactions with defensins, its been shown that defensins 1 (HNP1) and 2 (HNP2) are detrimental to the use of AAV2 vectors as gene delivery vehicles for the treatment of cystic fibrosis in the conducting ai rways (97) The data showed that levels of HNP1 and HNP2 we re elevated as high as 10,000 fold in the presence of AAV2 and correlated inversely with virus transduction levels Despite these initial findings, very little information is known about how these defensins inhibit AAV2 during infection K ey questions that remain to be asked include: Do defensins directly interact with the AAV2 capsid and if so, where do they bind? At what stage in the infection do defensins act? For example, do they block AAV2 cell surface attachment internalization, endosomal escape, a nd/or uncoating ? defensins on the AAV2 capsid and analyze their ability to inhibit transduction in vitro. AAV2 and antibody interactions The presence of humoral immune responses to wildtype AAV in the hu man population is an obstacle for gene delivery. Reports suggest that neutralizing antibodies
26 have been found in up to ~70% of the population (98) In addition, na tural exposure in humans to AAV 1, AAV2, AAV 5, AAV 6 AAV 8, and AAV 9 result in production of antibodies with a predominant response to IgG1 and low responses to IgG2, IgG3, and IgG4 (98) A20 is a monoclonal antibody (Mab) which binds intact capsids only and neutralizes AAV2 infection (99, 100) This AAV2 antibody is the most widely studied to date and reports show that epitopes on the capsid surface that inhibit A20 neutralizat ion also inhibit polyclonal neutralization (101) These data suggest that the A20 epitope is likely similar to those that are recognized by the antibodies generated against the capsid in response to AAV2 infection in the cell (101) As mentioned above, the structure of AAV2 in complex with Fabs from the A20 antibody was determined (102) Structural and mutational studies show that the binding footprint extends from the canyon around the fivefold to the side of the three fold protrusion. Interestingly, the A20 epitope does not overlap the HSPG and proposed integrin coreceptor binding sites. This is consistent with earlier findings that A20 binding does not inhibit AAV2 cell surface attachment a nd that neutralization occurs post nuclear entry (99) However, t he kinetics/affinities of this interaction are unknown. Further analyses of these interactions would be beneficial to elucidate the binding specifics of AAV2A20 since kinetic rates of antibody binding typically correlate w ith neutralization sensitivity (103) This project aims to develop an optimized method to analyze the capsid affinities of neutraliz ing antibodies using the AAV2A20 interaction. This would be applicable for large screen capsid antibody tests and also for differentiating between high and low affinity epitopes. Significance of Virus R eceptor a nd Immune R esponse I nteractions Recombinant AAV vectors have risen to the forefront as gene delivery vehicles over the past 20 years (104) Most AAV gene therapy applications to date have utilized
27 AAV2, which is currently in clinical trials to combat Parkinsons disease, Lebers congenital am aurosis, muscular dystrophy and other diseases. The focus of this project is two fold: (1) A structure function analysis of AAV2 capsids to identify regions that are essential for viral entry and endocytosis and (2) characterization of the immune responses that target AAV2 We believe the combination of structural, biochemical and biophysical analyses will aid us in our effort to characterize the coreceptor and immune response binding sites on the capsid surface Not only will these studies shed light on viral entry and the immune response for nonenveloped viruses, they will serve as a model for characterizing these properties for other AAV serotypes due to the optimization of the experimental approaches present ed in this study Outcomes will provide information to support the ratio nal design of AAV vectors for evading the immune response and targeted tissue tropisms by revealing essential viral host cell interactions encountered during infection.
28 Table 11. A denoassociated virus receptors Virus Glycan Receptor Co receptor Reference AAV1 N linked sialic acid Unknown (59, 105) AAV2 HSPG FGFR1, HGFR,LamR CD9, (2 3 27, 106)( AAV3 HSPG FGFR1, HGFR, LamR (16, 27, 107, 108) AAV4 O linked sialic acid Unknown (15) AAV5 N linked sialic acid PDGFR (15, 109) AAV6 N linked sialic acid EGFR (59, 105) AAV7 Unknown Unknown AAV8 Unknown LamR (27) AAV9 N linked galactose LamR (27, 110) AAV10 Unknown Unknown AAV11 Unknown Unknown AAV12 Unknown Unknown Abbreviations: Heparan sulfate proteoglycan (HSPG), hepatocyte growth factor receptor (HGFR), fibroblast growth factor receptor 1 (FGFR), laminin receptor (LamR), platelet derived growth factor receptor (PDGFR), epidermal growth factor receptor (EGFR)
29 Table 12 Viral receptors utilized by both AAV2 and other selected families Family Virus example Receptors Uptake route Reference Adenovi ridae Ad 2 and Ad 5 HSPG CME (69, 72, 111) Flaviviridae Dengue virus HSPG CME (70) Herpesviridae HCMV HSP G CME (71) Parvoviridae AAV2 HSPG CME (23 27, 112) HGFR, FGFR, Laminin Parvoviridae B19 Globoside, CME (113, 114) Papillomaviridae HPVs HSPG, CME (115, 116) Picornaviridae EV1 CME (117) Abbreviations: Heparan sulfate proteoglycan (HSPG), coxsackie receptor (CAR) hepatocyte growth factor receptor (HGFR), fibroblast growth factor receptor (FGFR), clathrinmediated endocytosis (CME)
30 Figure 11. Gene organization and transcription scheme for A denoassociated virus 2 (AAV2). The ge nome is encoded by three ORFs (rep, cap, aap) that are contained within t wo inverted terminal repeats (ITR) at the termini. The r ep gene encodes four regulatory proteins (Rep78, Rep68, Rep52, and Rep40) where Rep68 and Rep40 are alternatively spliced from Rep78 and Rep52, respectively. The number next to each Rep gene also desc ribes the molecular weight in kDa. The cap gene encodes three viral protein products (VP1, VP2, and VP3) where VP1 (87 kDa) contains a translation initiation codon, while VP2 (72 kDa) and VP3 (62 kDa) are generated by alternative splicing The as sembly act ivating protein (AAP; 23 kDa) is encoded by a nest ed, alternative ORF.
31 Figure 12. AAV2 life cycle depicting cell surface attachment, endocytosis, intracellular trafficking, nuclear entry and the immune response. The image depicts AAV2 binding to its primary receptor, HSPG, followed by interactions with coreceptors, the cytoplasm via clathrinmediated endocytosis followed by endosomal transport through multiple vesicles where the capsid eventually locates to the nucleus via microtubules. defensins) and humoral (A20 IgG) immune responses to AAV2 are also depicted. Abbreviations: Heparan sulfate proteoglycan (HSPG), fibroblast growth factor receptor 1 (FGFR1), and hepatocyte growth factor receptor (HGFR).
32 Figure 13 AAV2 variable regions and secondary structure assignment. (A) Variable loops are color coded as: VRI (purple), VRII (light blue), VRIII (red), VRIV (green), VRV (dark blue), VRVI (yellow), VRVII (orange), VRIII (pink), and VRIX (tan). Twofold, threefold, and fivefold icosahedral symmetry axes are represented as a sphere, triangle, and pentagon, respectively. Beta sheets are shown in grey. (B) The same monomer as in (A) is assembled 60 times to generate an entire capsid (43) The capsid is viewed down the twofold symmetry axis. The variable loops provide the variable topology on the surface of the capsid (C) Depth cue view of the three dimensional crystal structure of AAV2. The most interior and exterior regions of the capsid are colored blue and red, respectively. 3 5 2 5 fold 3 fold 2 fold A B C
33 Figure 14. AAV2 VP3 trimer with highlighted heparinbinding sites and NGR motif. (A) Cartoon view of surface heparin binding residues, R484, R487, K532, R585, and R588, are shown in blue and NGR motif residues (511513) are colored red. (B) Surface view. A B R588
34 CHAPTER 2 MATERIALS AND METHODS This chapter describes common proced ures and materials utilized throughout this thesis. The studies are divided into two main sections: The first involves expression and purification of AAV2 virus like particles (VLPs), and A20 IgG and Fabs. The second section involv es the biochemical, biophysical and structural defensins 1, 2, and 5. in the PQE TriSystem Vector Construct Design of rin The full Integrin (OpenBiosystems cat. # MHS1011169792), Fos (OpenBiosystems cat. # MHS101160754), and Jun (OpenBiosystems cat. # MHS101097228128) were purchased for PCR amplification of the appropriate domains. A 2 971bp fragment of Fidelity DNA polymerase; BIORAD cat. # 172 5302) with primers (see Table 2 1 ) designed with restriction endonuclease clonin g sites Nco I and Eco RI. The following thermal cycler parameters were used for all PCR amplification steps: (1) 98C 2 min, (2) 98C 30 sec, (3)58C 1 min, (4) 72C 4 min, and (5) 72C 10 min, with 35 cycles run between steps 2 4. Following amplification, the cDNA fragment was excised from an agarose gel (BIO RAD cat. # 1613102), extracted (Qiagen cat. # 28706), restriction enzyme digested with Nco I and Eco RI, and ligated into the pQE TriSystem Vector (Qiagen cat. # 33903) using the same cloning sites. A 3:1 insert to vector molar ratio was used for ligation. The ligation reaction was transformed into
35 JM109 competent cells (Stratagene cat. # 200235) and streaked on an agarose plate with ampicillin (50 g/ mL ) for overnight incubation at 37C. Colonies were selected and grown in 5 mL of LB media containing ampicillin. Clones were purified (Qiagen cat. # 27106) and identified by restriction digest and verified by DNA sequencing. A similar cloning strategy was employed for the remaining cloning procedures. A 153 bp fragment of the Fos gene involved in forming a leucine zipper with Jun was PCR amplified from the full l ength cDNA with primers (Table 21 ) designed with restriction sites Eco RI and Xho I, and a five domain and Fos was added to incorporate flexibility between the two expressed domains. The amplified Fos fragment was restriction digested, and cloned in frame utilizing the same clon ing sites in the pQE TriSystem vector. The completed construct was verified by DNA sequencing. A second construct was generated for the integr in extracellular domain. A 2 173 bp fragment was ampli fied with primers (see Table 21 ) designed with restriction sites Pci I and Hind III, and inserted into the pQE TriSytem vector using Nco I and Hind III cloning sites. A 153 bp fragment of Jun was amplified with primers (see Table 2 1 ) designed with restriction sites Hind III and Xho I a nd a glycine linker, and inserted in frame with the integrin extracellular domain utilizing the same cloning sites in the vector. The completed construct was verified by DNA sequencing. Small Scale Protein Expression/Time Course Analysis Human Embryonic Kidney 293 cells (HEK293; American Type Culture Collection) were cultured in 15 cm plates (BD Falcon cat. # 353025) and maintained in Dulbeccos modified Eagles medium (Gibco 11995) supplemented with 100 U/ mL penicillin, 100 g/ mL streptomycin, and 5% fetal bovine serum (SAFC Cat. #123030C) at 37C in a 5%
36 CO2: 95% air humidifier incubator. Cells were grown to 80% confluency and transfected precipitation (118) Briefly, 45 g of each construct were transfected per 15 cm plate (~2.5 X 107 cells) for a total of 90 g of DNA. Cells were harvested at various time points post transfection: 0 h, 12 h, 24 h, 32 h, 40 h, and 48 h. Additional time points were not taken since most of the cells lost adherence to plates 48 h post transfection. The cells were separated from the media (supernatant) by centrifugation at 1,811 x g at 4C for 20 min. The cell pellet (CP) was resuspended in 2X SDS loading dye and both the CP and supernatant were boiled at 100C for 10 min and analyzed by SDS PAGE. Large Scale Protei n Expression and Purification via Nickel Column C hromatography Figure 21. A cell factory of HEK293 cells was transfected with equimolar amounts of transfection. The cells were separated from the media after centrifugation and resuspended in 20 mL of lysis buf fer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, 0.05% Tween 20, pH 8.0) that included an EDTA free protease inhibitor mix (Roche cat. # 04693159001). The transfected cells were lysed by three/freeze thaw cycles with the addition of benzonase nuclease (Novagen cat. # 70664) after the last thaw followed by incubation at 37C for 30 min. The cell lysate was clarified by centrifugation at 1,811 xg at 4C for 20 min. The cellular debris was discarded and the supernatant was applied to an equilibrated (in lysis buffer) 1 mL NiNTA agarose (Qiagen cat. # 30210) gravity flow column (BIORAD cat. # 7311550) and the flow through sample was reloaded prior to washing and elution. A stepwise wash gradient with increasing concentrations of
37 imidazole (60 mM, 80 mM, 100 mM; three 1 mL fractions were collected for each wash) was used to remove nonspecific protein nickel interactions. Bound protein was eluted with a 250 mM imidazole and collected in 300 L fractions. The cell lysis, wash, and elution buffers were identical except for varying imidazole concentrations, and the wash and elution buffers were protease inhibitor free. The load, wash, and elution fract ions were analyzed by SDS PAGE. Based on the SDS PAGE data, fractions containing only the respective integrin dom ains were pooled and concentrated with a 9kDa cutoff concentrator (Amicon). Folding, assembly, and activation of integrin heterodimers are dependent on divalent cations, which have crucial roles in these processes (12, 119) For these proteins, Ca2+ aids heterodimer formation, while Mg2+ and Mn2+ support ligand binding. Thus samples were buffer exchanged and concentrated into buffers including 1 mM MgCl2 and 1 mM CaCl2 for further studies. In addition, samples were used for Western blots, native Dot blot, Coimmunopreciptation, negative stain EM, and SPR as described below. Co immunoprecipitation of HEK 293 cel Sepharose beads (GE Healthcare) were incubated at 4C with gentle rocking for 1 h. individual microcentrifuge tubes that included the MAb and protein A Sepharose. A fourth tube devoid of the MAb that inclu lysate was utilized as a negative control. The four tubes were incubated at 4C O/N with gentle rocking. Samples were spun at 1,811 xg f or 5 min for pelleting the protein A
38 beads followed by removal of unbound proteins in cell lysates, which were designated flow through. Beads were washed three times with 1 mL of cell lysis buffer and loading dye, boiled for 5 minutes to dissociate the immunocomplexes, collected by centrifugation, and SDS PAGE was performed with the supernatant. Construct D esign of The full MHS101161377), 58245), Fos (OpenBiosystems cat. # MHS101160754), and Jun (OpenBiosystems cat. # MHS101097228128) were purchased for PCR amplification of the appropriate domains. A 3003 integr in encoding only the extracellular domain was PCR amplified (iProof HighFidelity DNA polymerase; BIORAD ca t. # 172 5302) with primers ( Table 2 2 ) designed with restriction endonuclease cloning sites Sal I and Not I. The following thermal cycler parameters were used for all PCR amplification steps: (1) 98C 2 min, (2) 98C 30 sec, (3) 58C 1 min, (4) 72C 4 min, and (5) 72C 10 min, with 35 cycles run between steps 2 4. Following amplification, the cDNA fragment was visualized and excised from an agarose gel (BIO RAD cat. # 1613102), extracted (Qiagen cat. # 28706), restriction enzyme digested with Sal I and Not I, and ligated into the pFastBac1 vector (Invitrogen cat. # 10360014) using the same cloning sites. A 1:1 insert to vector molar ratio was used for ligation. The ligation reaction was transformed into JM109 competent cells (Stratagene cat. # 200235) and streaked on an agarose plate with mL ) for overnight incubation at 37C. Colonies were selected and grown in 5 mL of LB media containing ampicillin. Clones were purified (Qiagen cat. #
39 27106) and identified by restriction digest and verified by DNA sequencing. A similar cloning strategy was employed for the remaining cloning procedures. A 160 bp fragment of the Fos gene involved in forming a leucine zipper with Jun was PCR amplified from the full length cDNA with primers (Table 2 2 ) designed with restriction sites Not I and Xba I, and a five glycine linker. This linker region, domain and Fos was added to incorporate flexibility between the two expressed domains. The amplified Fos fragment was restriction digested, and cloned inframe utilizing the same cloning sites in the pFastBac1 vector. An 8X histidine tag was inserted inframe using complementary primers (Table 22 ) designed with restriction sites Xba I and Hind III. The incubated in a heat block for 5 min at 95C. The primers were gradually cooled to room temperature for subseq uent ligation using the same cloning sites in the pFastBac1 vector. The completed construct was verified by DNA sequencing. 2208 bp fragment was amplified with primers (see Table 2 2 ) designed with restriction sites Bam HI and Sal I, and inserted into the pFastBac1 vector using the same cloning sites. A 153 bp fragment of Jun was amplified with primers ( Table 2 2 ) designed with restriction sites Sal I and Xba I and a glycine linker integrin extracellular domain utilizing the same cloning sites in the vector. An 8X histidine tag was inserted inabove). The completed construct was verified by DNA sequencing. Generation of Recombinant B aculoviruses recombinant baculoviruses according to the manufacturers protocol (Invitrogen; Bac to -
40 Bac 5) were plaquepurified and amplified to a passage 2 (P2) stage. Each clone was analyzed for integrin DNA following agarose gel visualization by extraction and PCR amplification using primers ( Table 2 2 ) for the full length gene. Small Scale Protein E xpression ntegrin SF9 insect cells were cultured in 2800 mL Pyrex flasks and maintained in SF 900 II SFM media (Gibco cat. # 11496015) supplemented with 10% antibiotic antimycotic (Gi bco Cat. # 15240062) in a 28C incubator. All P2 stocks were titered by plaque assays and the clones with the 5, were used to infect 50 mL of 2 x106 SF9 insect cells at a multiplicity of infection (MOI) of 5. Cells were harvested 72 h post infection and separated from the media by centrifugation at 1, 811 xg at 4C for 20 min. The cell pellet was discarded and the media fraction, with the addition of an EDTA free protease inhibitor mix (Roche cat. # 04693159001) was appli ed to an equilibrated (lysis buffer; 50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, 0.05% Tween 20, pH 8.0) 1 mL Ni NTA agarose (Qiagen cat. # 30210) gravityflow column (BIORAD cat. # 7311550) and the flow through sample was reloaded prior to washing and elution. A stepwise wash gradient with increasing concentrations of imidazole (60 mM, 80 mM, 100 mM; three 1 mL fractions were collected for each wash) was used to remove nonspecific protein nickel interactions. Bound protein was eluted with a 250 mM imi dazole buffer and collected in 300 L fractions. The lysis, wash, and elution buffers were identical except for varying imidazole concentrations, and the wash and elution buffers were protease inhibitor free. The media, flow through, wash, and elution frac tions were analyzed by SDS PAGE and in addition, the elution fractions were analyzed by Western blot
41 Large Scale Protein Expression and Purification via Nickel Column C hromatography A brief diagram depicting the integrin purification scheme is shown in Figure 21. Approximately 1 L of SF9 insect cells (2 x106) in SF 900 II SFM media, were coinfected with recombinant baculovirus stocks of 5 at an MOI of 5 for each virus Cells were harvested 72 h post infection and separated from the media by centrifugation at 1, 811 g at 4C for 20 min. The cell pellet was discarded and 10% PEG 8000 (Fisher Scientific Cat. # BP2331) was added to the media to precipitate protein O/N followed by centrifugation at 17 700 g at 4C for 90 min to collect the pellet Th is pellet was resupsended in 2550 mL lysis buffer (including protease inhibitor) depending on pellet size. Following resuspension, an additional cent r ifugation step at 1 811 g at 4C for 20 min was performed to clarify the supernatant The pellet was discarded and the supernatant was applied to an equilibrated (lysis buffer) 5 mL NiNTA agaros e gravityflow column and the flow through sample was reloaded prior to washing and elution. The column was washed separately with 25 mL each of 100 mM and 150 mM imidazole. Bound protein was eluted with 400 mM imidazole and collected in 1 mL fractions. The flow through, wash, and elution fractions were analyzed by SDS PAGE. Based on the SDS PAGE data, fractions containing only the respective integrin domains were pooled and concentrated with a 9kDa cutoff concentrator (Amicon). Folding, assembly, and activation of integrin heterodimers are dependent on divalent cations, which have crucial roles in these processes (12, 119) For these proteins, Ca2+ aids heterodimer formation, while Mg2+ and Mn2+ support ligand binding. Thus samples were buffer exchanged and concentrated into buffers including 1 mM MgCl2 and 1 mM CaCl2 for further studies. Elution fractions were dialyzed into 20 mM HEPES, pH 7.4,
42 150 mM NaCl, 1 mM MgCl2, 0.1 mM CaCl2, serial diluted, analyzed by negative stain EM cryo EM and SPR as described below Co I mmunoprecipitation Integrin and AAV2 A co immunoprecipitation experiment was conducted to determine if an anti integrin antibody (Abcam cat. # P1D6) that recognizes only the integrin subunit could pull down the integrin subunit and if th e sample could pull down AAV2. Protein G Sepharose Fast Flow slurry (GE Healthcare cat. # 170618) was prepared according to the manufacturer. The slurry was centrifuged at 1,000 xg for 5 min to spin down the beads and the supernatant was dec anted. The beads were washed two more times and resuspended in binding buffer (20 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM MgCl2, 1 mM MnCl2, 0.1 mM CaCl2). Five experimental conditions were assayed: 1) To determine if integrin can pull down AAV2 50 L o f beads (100 g) + 2 L antibody (8 g) + 500 L binding buffer + 4 L (16 g) + 20 L AAV2 (24 g) ; 2) to determine if the anti integrin antibody can pull down 50 L of beads (100 g) + 2 L antibody (8 g) + 500 L binding buffer + 4 L (16 g); 3) to determine if AAV2 interacts nonspecifically with the beads 50 L of beads (100 ug) + 20 L AAV2 (24 g) + 500 L binding buffer; 4) to determine if AAV2 interacts nonspecifically with the anti tegrin antibody 50 L of beads (100 ug) + 2 L of anti a5b1 antibody (8 g) + 500 L binding buffer + 20 L AAV2 (24 ug) ; and 5) to determine if interacts nonspecifically with the beads 50 L of beads ( 100 g) + 4 L integrin (16 g). The six immunoprecipitation experiments were conducted over the course of 13 days as follows: Day 1) tubes 1 and 2 antibody, beads, and binding buffer were gently rocked at 4C for 1 h followed by the addition of and more rocking at 4C overnight. Tube 3 AAV2, beads, and binding
43 buffer were rocked overnight at 4C Tube 4 antibody, beads, and binding buffer were rocked for at 4C for 1 h followed by the addition of AAV2 and more rocking at 4C overnight. Tube 5 and binding buffer were rocked at 4C overnight. Day 2) tube 1 AAV2 was added to the mixture and rocked overnight at 4C Tubes 25 the mixture was centrifuged at 1,000 xg to pellet the beads. The supernatant was removed and saved as unbound. One mL of binding buffer was added to the beads for washing and was centrifuged at 1,000 xg for 5 min. Two more wash steps were performed and these were saved as wash 13. Fifty L of 2X SDS loading dye were added to the beads after the final wash. The mixture was centrifuged at 1,000 xg for 5 min and the supernatant was saved as elution. Day 3) Tube 1 the same procedure as day 2. All samples were run on Western blots as described below Purification of Baculovirus AAV2 VLPs Cell Culture and Virus Infection The recombinant baculovirus vector containing a modified AAV2 VP13 gene to ensure proper ratios of expressed VP1:VP2:VP3 was a generous gift from Sergei Zolotukin (Gene Therapy Center, University of Florida) (120) An appropriate volume of a titered recombinant baculovirus packaging the AAV2 VP13 gene was used to express AAV virus like particles (VLPs) by infecting SF9 insect cells maintained at 2.0 x106 cells/ mL of suspension at 27C. The cells were harvested 72 h post infection by centrifugation at 1,100 xg for 15 min and the pellet was collected. Virus Purification A brief diagram depicting the AAV2 VLP purification scheme is shown in Figure 2 1. The recombinant AAV2 was purified according to the published protocol (121) The pellet was resuspended in lysis buffer (20 mM Tris, 150 mM, pH 8.4 and freeze thawed
44 3x in liquid nitrogen and 37C water baths. One microliter of benzonase (Sigma cat. # E1014) was added to the cell lysate (50 U/ mL final concentration) and incubated at 37C for 30 min. The crude cell lysate was clarified by centrifugation at 4,000 xg for 20 min and the supernatant was then loaded onto a discontinuous 1560% iodixanol step gradient prepared from Optiprep (Nycomed cat. # 1114542). The iodi xanol gradient was then centrifuged at 350,000 xg at 18C for 1 h. The gradient was fractionated and the VLP fractions containing VP1VP3 was identified by Dot blot probed with B1 as primary antibody (data not shown) (99) The fractions from the 25% iodixanol gradient were further purified on an anion exchange 5 mL HiTrap Q column (GE Healthcare cat. # 17515901). The clarified cell lysate was diluted with 2x Buffer A (20 mM Tris, 15 mM NaCl, pH 8.5), loaded onto a column at 1 mL /min, washed with 50 mL Buffer A, and the sample was eluted with 30 mL Buffer B (20 mM Tris, 500 mM NaCl, pH 8.5). The fractions were collected and buff er exchanged and concentrated as described below. Virus Buffer Exchange and Concentration The AAV2 VLPs were buffer exchanged and concentrated in 150 kDa nominal cutoff Apollo concentrators (Orbital Biosciences cat. # AP2015010). The virus was buffer exchanged 3X by adding up to ~20 mL of the desired buffer to the sample in the retentate vial and concentrated by centfiguation at ~1,933 xg at 4C. On the last buffer exchange, the sample was concentrated by centrifugation to the desired volume. The samples were analyzed by SDS PAGE, Dot blot, and negative stain electron microscopy (EM) for monitoring VLP purity and integrity prior to use for negative stain EM, cryo EM and SPR as described below.
45 A20 IgG and Fab P urification A20 IgG P urification Approximatel y 4 mL of concentrated hybridoma supernatant stock of A20 (1A2 HL2185) were supplied from the hybridoma core at the Interdisciplinary Center for Biotechnology Research ( ICBR) at the University of Florida. Approximately 4 mL of Protein G Sepharose Fast Flow 50% ethanol slurry with ( GE Healthcare) were added to a 15 mL falcon tube and spun at 2,500 xg for 3 min. Ethanol was decanted and the Protein G beads were resupsended in 10 column volumes of binding buffer ( 10 mL 20 mM N aH2Po4). This wash step was repeated twice. After the third wash, the 4 mL of A20 supernatant was applied to the beads and the total volume was brought up to 12 mL using binding buffer. The sample and beads were rotated in an endover end mixer at 4 C over night. The next morning, the mixture was added to a gravity flow column (BIORAD cat. # 731 1550) and washed with a total of 40 mL of binding buffer. The bound sample was eluted (0.1 M Glycine HCl, pH 3.0) and collected in 1 mL fractions in microcentrifuge tubes containg150 L of neutral ization buffer (1 M Tris, pH 8.8). The samples were analyzed by SDS PAGE, Western blot, SPR, and digested to produce Fabs as described below A20 Fab P urification A brief diagram depicting the A20 Fab purification scheme is s hown in Figures 2 1 and 22 The A20 Fab purification protocol was conducted according to the Fab preparation kit (Pierce cat. # 44985). Briefly, 500 L (4 mg/ mL ) of concentrated A20 IgG was centrifuged in the provided desalting column to bind low molecular weight contaminant s, while the A20 IgG was in the flow through. The flow through was added to an immobilized papain column and incubated in an endover end mixer at 37C for
46 exactly 3 h. The column was then centrifuged and the flow through was collected. An additional wash step using 500 L of 1X PBS was performed and these two fractions were pooled for a total volume of 1 mL The contents were mixed end over end in a prepackaged Protein A column for 10 min at room temperature. The Protein A colu mn was centrifuged at 1, 000 xg and a 1 mL fraction was collected. This should contain the Fab fragments. Two additional spins with 1 mL of 1X PBS were performed to recover the maximum amount of Fab fragments. Three more centrifugation steps using 1 mL of e lution buffer per step was performed to remove undigested Fc and IgG fragments. Approximately 100 L of a 1 M Tris, pH 8.8 solution were added to each tube for pH neutralization. The six fractions, plus the A20 IgG control (starting material) were analyzed by SDS PAGE. Samples were also utilized for SPR as described below. SDS PAGE Analysis All purified protein samples [AAV2 ( VP1 87 kDa, VP2 72 kDa, VP3 62 kDa), 85 kDa) 85 kDa) integrins, A20 IgG (h eavy chain 50 kDa, light chain 25 kDa) and Fabs (25 kDa) ] were analyzed. For both integrin purifications, protein bands were cut from the gel and the slices were stored in destain until the samples were prepared for protein verification by mass spectro metry (see below). All fraction samples were boiled for 10 min, centrifuged briefly, and loaded onto a 420% polyacrylamide gel (BioRad cat. # 345003). SDS PAGE was performed at 125 V for 90 min (122) Gels were stained in a solution consisting of 12.5% coomassie blue stain stock R 250, 10% acetic acid, and 50% methanol for 1 h. Gels were destaine d in solution consisting of 10% acetic acid and 50% methanol overnight.
47 Western Blot A nalysis Fraction samples were prepared as above for SDS PAGE analysis prior to Western blot analysis. Briefly, gels were transferred to nitrocellulose membranes (Whatman cat. # 10401197) at 1 amp for 1 h 30 min. Transferred membranes were blocked in 5% dry milk/1X PBS, 0.05% Tween for 1 h, incubated with a mouse anti H is primary antibody for integrins (Abcam ca t. # 18184) in a 1% dry milk solution, and incubated with a secondary anti mouse IgG antibody (1:5,000; GE Healthcare cat. # NA931V). Three consecutive washes (5min; 1X PBS, 0.05% Tween) were conducted between each step. Integrins were detected using chemi luminescence (Millipore cat. WBKLS0500). Native Dot B lot U n boiled elution fractions (total 5 g) that have not been denatured, previously indicated by SDS PAGE and/or Western blot to contain AAV2 VLPs or integrin domains were loaded onto nitrocellulose membranes and treated under the same conditions as a Western blot. A20 Mab and C37 Mab for AAV2 (99) anti inte and anti all of which recognize conformational epitopes were utilized for detection. Additional primary Mabs, anti integrin which recognizes a linear conformational epitope for t egrin subunit, and B1 (negative control) (99) which recognizes a linear epitope for AAV2 were used. Mass Spectrometry Trypsin Digestion of A wash solution of 200 L of 50 mM NH4HCO3/50% acetonitrile (pH 8.4) was added to the excised SDS PAGE gel bands containing proteins believed to be the integrin subunits. The contents of each tube were vortexed f or 45 minutes and the
48 solution was discarded. The wash step was repeated until the gel slices were clear and then they were dried (Speedvac) to remove liquid. A freshly prepared 45 mM dithiothreitol solution ( DTT ) was added to cover the gel slices for 45 m DTT solution was replaced with approximately the same volume of freshly prepared 100 mM iodoacetamide followed by incubation at RT for 45 min. The iodoacetamide solution was removed and the gel slices were washed with 200 L of 50 mM NH4HCO3/50% acetonitrile (pH 8.4) for 30 min with vortexing. The wash solution was removed and this procedure was repeated 3 times. The gel slices were dried in a Speedvac. Approximately 510 L of 12.5 ng/ L Trypsin (Promega) in 25 mM NH4HCO3 (pH 8.4) was added to each gel slice and incubated at 4C for 30 min. Excess enzyme solution was removed, the required volume of 25 mM NH4HCO3 (pH 8.4) to cover the gel slices topped by adding aqueous 5% acetic acid to a final concentration of 0.5%. The tubes containing the gel slices were agitated for 15 min and centrifuged at 10,000 g for 1 min on a benchtop microcentrifuge. The supernatant (containing tryptic peptides) was tr ansferred to a clean 0.5mL tube (BioRad). Approximately 50 L of extraction solution (80 % acetonitrile, 1 % TFA) was added to the gel slices followed by sonication in an ultrasonic water bath for 5 min. The supernatant (containing additional tryptic pe ptides) was transferred to the same 0.5mL tube. The gel slices were extracted again with an additional 50 L of acetonitrile and agitation by gentle vortexing for 10 min and centrifuged at x 10,000 g for 1 min. The supernatant was combined with the previous two extractions in the 0.5 mL tube. The pooled extracted peptides were dried by Speedvac. Approximately 15 L of Loading Buffer (3% ACN, 1% acetic acid, and 0.1
49 % TFA) was added to each tube and gently agitated by vortexing. The supernatants were transf erred to a vial for Liquid Chromatography Tandem Mass Spectrometry (LC MS/MS) analysis. LC MS/MS The enzymatically digested integrin samples were injected onto a capillary trap (LC Packings PepMap) and desalted for 5 min with a flow rate 10 L /min of 0.1% v/v acetic acid. The samples were loaded onto a column (LC Packing C18 Pep Map) on the Tempo nanoflow HPLC system. The elution gradient of the HPLC column started at 97% solvent A (0.1% v/v acetic acid, 3% v/v ACN, and 96.9% v/v H2O), 3% solvent B (0.1% v/v acetic acid, 96.9% v/v ACN), and 3% v/v H2Oand finished at 60% solvent B, 40% solvent A and ran for 20 min. For protein identification LC MS/MS analysis was conducted using a hybrid quadrupolelinear ion trap mass spectrometer (4000QTRAP, Applied Bios ystems). The focusing potential and ion spray voltage was set to 275 V and 2600 V, respectively. The informationdependent acquisition (IDA) mode of operation was employed in which a survey scan from m/z 400 1500 was acquired followed by collision induced dissociation (CID) of the two most intense ions. Survey and MS/MS spectra for each IDA cycle were accumulated for 1 and 3 s, respectively. Protein Search A lgorithm for Peptide Mass I dentification Tandem mass spectra for the integrin samples were extracted by ABI Analyst version 1.4.1. All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK; version 2.0.01). Mascot was set up to search the IPI Human database (67,524 entries) assuming the enzyme used for protein digestion was trypsin. Mascot was searched with a fragment ion mass tolerance of 0.60 Da and a parent ion tolerance of 1.0 Da. Iodoacetamide derivative of Cys, deamidation of Asn and Gln, oxidation of
5 0 Met were specified in Mascot as variable modifications. Scaffold (version Scaffold0 2 0301, Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 95.0% probability as specified by th e Peptide Prophet al gorithm (123) Protein identifications were accepted if established at greater than 99.0% probability and contain at least 2 identified unique peptides. Protein probabilities were assigned by th e Protein Prophet algorithm (124) Transduction Inhibition Assays In V itro To determine the capacity of defensins 1,2, and 5 to inhibit AAV2 transduction, various experimental conditions were assayed. HeLa cells at a concentration of 1 X 106 were employed for all experiments. Highly purified stocks of self complementary r AAV2 GFP viruses (5 X 107 viral genomes per microliter) packaging the enhanced green fluorescence gene (scAAV2EGFP) was a kind gift from Drs. Arun Srivastava and Chen Ling (Univer sity of Florida, Gainesville, FL). The following experiments were conducted: 1) 100 g/ mL and 10 g/ mL 100 g/ mL 3) defensin 1 100 g/ mL 10 g/ mL and 1 g/ mL 4) defensin 2 100 g/ mL 10 g/ mL and 1 g/ mL 5) defensin 5 200 g/ mL 20 g/ mL and 200 ng/ mL Based on the premise that 1 virus contains 60 VP binding sites, complexing ratios were established where a ratio of 1 VP:1 integrin is equivalent to 60 integrin attachments on one capsid. All complexing ratios are shown in Table 23. The virus and integrin or defensins Eagle Medium (DMEM; Life Technologies) lacking FBS and antibiotic/antimycotic and incubated at room temperature for 30 min prior to infection of Human cervical cancer (HeLa) cells. The cells were washed once with 1 X PBS and infected at a multiplicity of
51 infection (MOI) of 1,25 0 viral genomes per cell in various experiments. After 2 h, an Technologies) were added to each well. Seventy two h after infection, transgene expression was detected by fluorescence microscopy using an Axiovert 25 fluorescence microscope (Carl Zeiss, Inc., Thornwood, NY). Images from three independent visual fields were analyzed quantitatively by ImageJ analysis software (National Institutes of Health, Bethesda, MD). Transgene expression was as sessed as total area of green fluorescence (pixel2) per visual field (mean SD). Analysis of variance (ANOVA) was used to compare test results with the control, and they were determined to be statistically significant when p <0.05. Surface Plasmon Resonan ce A sensorgram depicting the multiple stages (e.g. association, dissociation, regeneration) and types of kinetic experiments conducted by SPR are depicted in Figure 23 Mul t i ple protein samples were used for difference reactions (AAV2integrin, AAV2A20 IgG, and AAV2A20 Fab). Real time measurements of analyteligand associations were carried out with an automated biosensor on BIAcore 2000 or BIAcore X 100 instruments (GE Healthcare). Ligands are defined as protein coupled to chip and the analyte is protein flowed over ligand. Carboxymethyl dextran coated biosensor chips (CM5 or CM3; GE Healthcare) or carboxymethyl 1 chips (CM1; GE Healthcare) were used for amine coupling experiments. Ligands were diluted in 10 mM sodium phosphate buffers with a low pH (pH range 4.0 5.0) for coupling, typically 1 pH unit under the isoelectric point (pI) to impart a net positive charge on the ligand. The biosensor chips were activated with N hydroxysuccimide and N ethyl N(dimethylaminopropyl) carbodiimide
52 followed by ligand immobilization. Immobilization buffers without Tris such as 1 XPBS or 20 mM HEPES, 150 mM NaCl, pH 7.4 were used. Activated sites that remained uncoupled were quenched (i.e. blocked) by a ddition of 1 M ethanolamineHCL, pH 8.5. An additional flow cell was activated and quenched on each chip for reference subtraction. Response values that depict changes in refractive indices on sensorgrams are measured in response units (RU). One RU unit is equivalent to 1 picogram of protein per mm2. Additional information is provided for each experiment below. This experiment was performed on the BIAcore 2000 instrument. A brief diagram depicting the immobilization, binding, and regeneration steps are shown in Figure 24 AAV2 was diluted in 10 mM sodium acetate buffer, pH 4.5 to a final concentration of 50 g/ mL and immobilized on flow cell 2 at 1296 RU on a C1 chip. Flow cell 1 was blocked and served as a reference for subtraction. The following amounts of sample were injected over both flow cells at at 5 L/min for 2 min: 1) A20 IgG (2.2 M), 2) 50 mm diethylam ine regeneration at 50 L /min for 24 sec, 3) integrin (4.1 M) in the presence of 1 mM CaCl2 and 1 mM MgCl2, 4) integrin (4.1 M) in the presence 20 mM MnCl2 and the same concentration of divalent cations as in injection 3, 5) integrin (8.2 M) in the presence of 1 mM CaCl2 and 1 mM MgCl2. Ligand and analyte ratios are depicted in Table 24 Manual R uns of I ntegrin Injected o ver I mmobilized AAV2 This experiment was performed on the BIAcore 2000 instrument. AAV2 was diluted in 10 m M sodium acetate buffer, pH 4.5 to a final concentration of 25 g/ mL and immobilized on flow cell 2 at 187 RU on a CM5 chip. Flow cell 1 was blocked and served as a reference for subtraction. The following samples were flowed over all four
53 flow cells at 5 L ( 2 M) in the the presence of 1 mM CaCl2, 1 mM MnCl2, and 1 mM MgCl2, 3) the D11 antibody (10 M) against AAV of 1 mM CaCl2 and 1 mM MgCl2 5) 20 mM EDTA regeneration at 50 L /min for 24 sec, 6) 50 mM diethylamine at 50 L 1 mM CaCl2, 1 mM MnCl2, and 1 mM MgCl2, 8) 20 mM EDTA as a regeneration buffer at 50 L 2, 1 mM MnCl2, and 1 mM MgCl2 + 20 mM EDTA. Ligand and analyte ratios are depicted in Table 24 I ntegrin Injected o ver I mmobilized AAV1, AAV2, and AAV5 This experiment was performed on the BIAcore 2000 instrument. AAV1, AAV2, and AAV5 were diluted in 10 mM sodium acetate buffer, pH 4.5 to a final concentration of 50 g/ mL and immobilized on flow cells, three, two, and four respectively on a CM3 chip Flow cell one was blocked and used for reference subtraction for all experiments. The immobilized levels of sample were 3709 RU (AAV1), 4250 RU (AAV2), and 3271 RU (AAV5). The following amounts of sampl e were flowed over all four flow cells at 5 L/min for 2 min unless otherwise noted mM MgCl2, and 1 mM MnCl2, 3) A20 antibody (2. 2 M) control, 4) 50 mM diethylamine as a regeneration buffer at 50 L /min for 24 sec. The following amounts of sample were flowed over flow cells 2 and 1 at 5 L /min for 2 min unless otherwise noted integrin (5 M) in the presence of 1 mM CaCl2 and 1 mM MgCl2, 6) 20 mM EDTA regeneration at 50 L/min for 24 sec, 7) 50 mM diethylamine regeneration at 50 L /min
54 2,1 mM MgCl2, and 1 mM MnCl2 CaCl2,1 mM MgCl2, and 1 mM MnCl2, 10) 50 mM diethylamine regeneration as a regeneration buffer at 50 L/min divalent cations at a 1 mM concentration, 12) 50 mM diethylamine re generation at 50 L three divalent cations at a 1 mM concentration, 14) 50 mM diethylamine regeneration as a regeneration buffer at 50 L M) + 20 mM EDTA in the presence of all three divalent cations at a 1 mM concentration. Ligand and analyte ratios are depicted in Table 24 Manual Run of AAV2 Injected ove r I This experiment was performed on the BIAcore 2000 instrument. A brief diagram depicting the immobilization, binding, and regeneration steps are shown in Figure 25 was diluted in 10 mM sodium acetate buffer, pH 4.0 to a final concentration of 25 g/ mL and immobilized on flow cell 2 at 3104 RU on a CM3 chip. Flow cell 1 was blocked for reference subtraction. The following amounts of sample were flowed over both flow cells at 5 L /min for 2 min unless otherwise noted: 1) AAV2 (150 nM) in the presence of 1 mM CaCl2,1 mM MgCl2, and 1 mM MnCl2, 2) anti integrin antibody (1 M). Ligand and analyte ratios are depicted in Table 24 AAV2 A20 IgG Single Cyle Kinetics This experiment was performed on the BIAcore X100 instrument. A brief diagram depicting the immobilization, capture, and binding experiments is depicted in Figure 26 A20 IgG was diluted in 10 mM sodium acetate, pH 4.5 buffer to a final concentration of 25 g/ mL and immobilized on a CM3 chip at 100 RU. A singlecycle capture kinetics
55 experiment was conducted by capturing AAV2 (10 g/ mL ) at ~100 RU on the immobilized A20 IgG Following capture, a concentration series of A20 IgG was flowed over captured AAV2 as follows: 1) Buffer only 3 replicates, 2) 1.6 nM, 3) 8 nM, 4) 40 nM, 5) 200 nM, 6) 1000 nM. Ligand and analyte ratios are depicted in Table 25. Other experimental details are as follows: AAV2 capture time = 60 sec, stabilization period, 10 sec, A20 IgG injection time = 3 min, flow rate = 30 L /min, dissociation time = 30 min. Ligand and analyte ratios are depicted in Table 26. The surface was regenerated with 50 mM diethylamine only once after the 1000 nM A20 IgG dissociation. Kinetic binding data, including association ( ka) and dissociation ( kd) rate constants, were determined with BIAevaluation software for BIAcore X100. AA V2 A20 IgG MultiCycle Kinetics This experiment was performed on the BIAcore X100 instrument. A brief diagram depicting the immobilization and direct binding experiments is depicted in Figure 27 AAV2 was diluted in 10 mM sodium acetate, pH 4.5 buffer to a final concentration of 17 g/ mL and immobilized on a CM3 chip at 100 RU. A multi cycle direct binding kinetics experiment was conducted using a concentration series of A20 IgG as follows: 1) Buffer only 3 replicates, 2) 0.5 nM, 3) 2 nM, 4) 5 nM, 6) 10 nM, 7) 50 nM 2 replicates, 8) 100 nM, 9) 500 nM, 10) 1000 nM. Ligand and analyte ratios are depicted in Table 25 Other experimental det ails are as follows: Flow rate = 30 L /min, injection time = 3 min, stabilization time = 0 min, dissociation time = 30 min. The surface was regenerated after each binding interval using 15 L of 50 mM diethylamine at a flow rate of 30 L /min for 30 sec. Kinetic binding data, including associat ion ( ka) and dissociation ( kd) rate constants, were determined with BIAevaluation software for BIAcore X100.
56 AAV2 A20 Fab Multi Cycle Kinetics This experiment was performed on the BIAcore 2000 instrument. AAV2 was diluted in 10 mM sodium acetate, pH 4.5 b uffer to a final concentration of 25 g/ mL and immobilized on a CM3 chip at 1508 RU. A multi cycle direct binding kinetics experiment was conducted using a concentration series of A20 Fab as follows: 1) Buffer only 3 replicates, 2) 20 nM, 3) 100 nM, 4) 250 nM, 5) 500 nM, 5) 1000 nM 2 replicates, 6) 2500 nM, 7) 5000 nM, 8) 7500 nM, 9) 10000 nM. Ligand and analyte ratios are depicted in Table 25 Other experimental details are as follows: Flow rate = 30 L /min, injection time = 3 min, stabilization time = 0 min, dissociation time = 30 min. The surface was regenerated after each binding interval by using 15 L of 50 mM diethylamine at a flow rate of 30 L /min for 30 sec. Kinetic binding data, including association ( ka) and dissociation ( kd) rate constants were determined with BIAevaluation software version 4.0. Negative Stain Electron M icroscopy Five microliter s of samples were pipetted onto glow discharged C flat Holey Carbon Grids (Electron Microscopy Sciences) and wa shed three times by inversion on 15 L drops of diH2O. Proteins were then stained for 30 sec by inversion on two 15 L drops of either 2% uranyl acetate or 2% uranyl formate (Polyscience, Inc. cat. # 24762). The grids were blotted using filter paper after the last wash and staining steps The samples were visualized by using a Tecnai G2 Spirit Transmission Electron Microscope (FEI) at 50K magnification. Complexing ratios are shown in Table 26
57 ntegrin AAV2 (1 .0 mg/ mL ) was complexed with (0.1 3 mg/ mL ) in a ratio of 8 VP:1 integrin (Table 26 ) The complexing was performed in 20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM MgCl2, 1 mM CaCl2. The samples were stained with 2% uranly acetate prior to viewing with the el ectron microscope I ntegrin Dilution S eries A dilution series was conducted to visualize integrin molecules by analyzing stock (0.6 mg/ mL ) and 2X, 10X, 20X, 100X, and 200X dilutions. The complexing was performed in 20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM MgCl2, 0.1 mM CaCl2 1 mM MnCl2 and stained with 2% uranlyl formate prior to viewing with the electron microscope. AAV2 C omplexed with ntegrins AAV2 (0.82 mg/ mL mL mg/ mL ) in a ratio of 1 VP:1 integrin (Table 26 ) The AAV2 and both integrin samples were dialyzed into 20 mm HEPES, pH 7 .4, 150 mM NaCl, and 20 mM HEPES pH 7.4, 150 mM NaCl, 1mM MgCl2, 1 mM CaCl2, respectively prior to complexing The complexes were sta ined by 2% uranyl acetate prior to visualization by the electron microscope Protein C rosslinking of AAV2 and ntegrins integrin were cross linked via BS3 crosslinkers (Pierce cat. # 21580). The conjugation and quenching buffers were 20 mm HEPES, 150 mM NaCl, 1 mM MgCl2, 0.1 mm CaCl2, 1 mM MnCl2 and 1 M Tris HCl, pH 7.5, respectively. The BS3 sample was prepared in water according to the manufacturers protocol. Two separate reactions were conducted: 1) AAV2 (0.82 mg/ mL ) and integrin (3.2
58 mg/ mL ) were complexed in a ratio of 1 VP :1 integrin (Table 26 ) 2) AAV2 (0.82 mg/ mL ) and integrin (0.006 mg/ mL ) complexed in a ratio of 1 VP :0.002 integrin (Table 26 ) Following complexing, a final concentration of 0.42 mM BS3 was added to the mixed proteins and incubated for 30 min at room temperature. The reactions were quenched for 15 min at room temperature by adding quenching buffer to a final concentration of 30 mM. AAV2 VLPs D efensins 1, 2, and 5 AAV2 (0.9 mg/ mL ) was defensins 1 (Peptide Institute, Inc cat. #4271s) 2 (Peptide Institute Inc cat. #4428s) and 5 (Peptide Institute Inc cat. #4415 s) in ratios of 1 VP:2.5 defensins (Table 26 ) The complexing buffer was 20 mM HEPES, 150 mM NaCl. The samples were stained with 2% uranyl acetate prior to visualization by the electron microscope. Cryo Electron M icroscopy The following procedures describe the methods employed for analyzing complexes of AAV2defensin 5 by cryo EM The complex ratios are depicted in Table 26 and a general overview of the procedure is shown in Figure 28. The statistics for the individual reconstructions are collated in Table 27. Cryo EM Data Collection Small aliquots (~3.0 L ) of complexed samples wer e incubated on ice for 1 h and then vitrified using a Mark 4 Vitrobot (FEI) on holey carbon grids The samples were examined using a Tecnai G2 F20 (FEI) microscope at an accelerating voltage of 200 kV at a nominal magnification of 67,000 X. D igital microgr aphs were collected on an Ultrascan 4000 CCD camera (Gatan Inc.) using a step size of 2.24 /pixel The CCD images were saved in the Gatan DM3 file forma t
59 Three Dimensional Reconstruction Individual images of complexes were manually selected from the digitized micrographs and extracted within box sizes 1.5X to 2X the length of the longest axis of the particle using the interactive graphics program, EMAN2 (125) Preprocessing of the selected images and estimation of the defocus levels of each micrograph were determined use EMAN2. Particles were manually selected, screened, and separ a ted into data sets for initial model building and refinements. Particles were removed from data sets if they were too close to one another or there was heterogeneity in the sample. Table 27 shows the starting and final number of particles. Initial models for all data sets were generated using 10% 15% of the total number of boxed particles selected from the furthest from focus micrographs. The init ial models served as starting models for full orientation and origin determination of the entire data set. Both icosahedrally averaged and asymmetric reconstructions started with icosahedrally av eraged initial models, however, the asymmetric refinements em ployed the C1 option (no symmetry) in EMAN2 Refinements were conducted in two to three steps using decreasing angular search values (delta) and the previous cycle s last refinement model The resolution of the f inal 3D maps, reconstructed from the select e d particle images were estimated using the 0.143 threshold of the Fourier Shell Correlation criterion (FSC) (data not shown). The handedness of the 3D reconstructed maps was determined by comparison of the surface features of the AAV2 crystal structure (PD B accession number 1LP3) (76) using the fit command in Chimera. If the handedness of the cryo EM density map was not accurate the vop zflip #(map) command (vop = volume editor) in Chimera ( http://www.cgl.ucsf.edu/chimera ) corrected the handed ness of the cryo EM density map to mat ch the AAV2 crystal structure.
60 Fitting of PseudoAtomic Models into Cryo EM Density Maps The atomic coordinates for the AAV2 crystal structure solved at 3 (1LP3; 53) w as used for capsid docking and scaling of cryo EM density maps. For structural comparisons, t he AAV2 crystal structure Protein Data Bank (PDB) file was converted from PDB to MRC (electron density) format to match the /pixel, resolution, and box size values of the cryo EM density map using the e2pdb2mrc.py A (/pixels) R ( resolution) B (box size) AAV2.pdb AAV2.mrc command in the linux shell using EMAN2 (125) The structure factor s (126) of the AAV2.mrc map were matched to the cryo EM.mrc density map using the e2proc3d.py cryoEM_ map.mrc x tmp.mrc -calcsf= cryo EM .sf and e2proc3d.py AAV2.mrc AAV2filtere d.mrc --setsf= cryo EM .sf commands and also to filter artifacts from the PDB density map (AAV2filtered.mrc) to more accurately match the filtration of the cryo EM density map. For example, the AAV2 PDB artifacts are a result of the lower resolution filter ed map still retaining higher resolution data. For visual comparisons between the AAV2 filtered .mrc and cryo EM.mrc density maps, the contour levels (density) of the map with the lower RMS (root mean squared) value was scaled in C himera using the vop scale #(map) factor ( hig her RMS value / lower RMS value) command. For docking, the AAV2 crystal structure was docked into the AAV2 electron density map using the Fit command in Chimera. The correlation coefficient between the AAV2 crystal struc ture and electr on density was >95%. Difference maps were generated to subtract differences between volumes to identify or isolate masses in the cryo EM maps The AAV2 filtered.mrc map was subtracted from the AAV2ity changes (gray) that could account for integrin binding. Negative differences (blue) were generated by subtracting the maps in the reverse order. Similarly, an AAV2 filtered.mrc map was
61 generated to the same parameters as the AAV2HD5 density map and th en subtracted from the AAV2 HD5 map to i dentify positive density differences (orange) that could account for HD5 binding. Negative density differences (blue) were calculated by subtracting the maps in the reverse order. The difference maps were generated i n Chimera using the following command vop #(map1) subtract (map2). The changes on were compared to the AAV2 crystal structure utilizing the Real Space Refine command in COOT (187)
62 Table 21 Primer s eq uence Jun genes Primer Name Sequences aPr REb 5 gag ccatgg cttttccgccgcggcgac 3 1 RE 5 gag gaattc tggctgaatgccccaggtgac 3 2 Jun_F1RE 5 gag aagctt ggcggcggcggcggc agaatcgcccggctggagg 3 3 Jun_R1RE 5 gag ctcgag gtggttcatgactttctgtttaagc 3' 4 RE 5 gag acatgt tgccgcgggccccggcgc 3 5 RE 5 gag aagctt gttgggggtgtttccacac 3 6 Fos_F2 RE 5 gag gaattc ggcggcggcggcggc ctgactgatacactccaagc 3 7 Fos_R2RE 5 gag ctcgag gtgagctgccaggatgaactctag 3 8 aIndicates the primers used as shown in Figure 31; brestriction sites are italicized and glycine linkers are underlined. Table 2Jun genes and a 8X histidine tag Primer Name Sequences aPr REb 5 gag gtcgac gccaccatggggagccggacgccag 3 1 RE 5 gag gcggccgca tttctgccttggtccattgc 3 2 Fos_F4 RE 5 gag gcggccgc t ggtggtggtggtggt ctgactgatacactccaagcg3 3 Fos_R3RE 5 gag tctagagtgagctgccaggatgaactc 3 4 His_F1 RE 5 ctagacatcatcatcatcatcatcattaaa 3 5 His_R1 RE 5 agctt ttaatgatgatgatgatgatgatgt 3 6 RE 5 gag ggatcc gccaccatgggtaatttacaaccaattttc 3 7 RE 5 gag gtcgac gtctggaccagtgggacac 3 8 Jun_F2RE 5 gag gtcgac ggtggtggtggtggt agaatcgccggctggagg 3 9 Jun_R2RE 5 gag tctaga gtggttcatgactttctg 3' 10 aIn dicates the primers used as shown in Figure 41; brestriction sites are italicized and glycine linkers are underlined.
63 Table 23 Ratios of AAV2 with integrins and defensins analyzed by transduction inhibition Virus VPa Protein Moleculesb Ratioc AAV2 5.0X109 3.01X1014 1:6.0X104 AAV2 5.0X109 3.01X1013 1:6.0X103 AAV2 5.0X109 3.01X1014 1:6.0X104 AAV2 5.0X109 HNP1 1.72X1016 1:3.4X106 AAV2 5.0X109 HNP1 1.72X1015 1:3.4X105 AAV2 5.0X109 HNP1 1.72X1014 1:3.4X104 AAV2 5.0X109 HNP2 1.72X1016 1:3.4X106 AAV2 5.0X109 HNP2 1.72X1015 1:3.4X105 AAV2 5.0X109 HNP2 1.72X1014 1:3.4X104 AAV2 5.0X109 HD5 3.44X1016 1:6.9X106 AAV2 5.0X109 HD5 3.44X1015 1:6.9X105 AAV2 5.0X109 HD5 3.44X1013 1:6.9X103 aI ndicates the number of viral proteins,bindicates the number of protein molecules, cindicates the inhibition ratio (viral proteins/protein molecules). Abbreviations: Viral proteins ( VP), adenoassociated virus defensin 5 (HD5).
64 Table 24 R atios of AAV2 and integrins analyzed by surface plasmon resonance Chip Liganda VP /Moleculesb Analytec Molarityd Moleculese Ratiof C1 AAV2 1.17X 1010 A20 Mab 2.2 M 1.33X1018 1:1.1X108 AAV2 1.17X1010 integrin 4.1 M 2.47X1018 1:2.1X108 AAV2 1.17X1010 integrin 8.2 M 4.94X1018 1:4.2X108 CM5 AAV2 1.69X109 2.0 M 1.21X1018 1:7.1X108 AAV2 1.69X109 6.0 M 3.61X1018 1:2.1X109 AAV2 1.69X109 D11 Mab 10 M 6.02X1018 1:3.6X109 AAV2 1.69X109 2.0 M 1.20X1018 1:7.1X108 CM3 AAV1 3.35X 1010 integrin 2.0 M 1.20X1018 1:3.5X107 AAV2 3.84X 1010 integrin 2.0 M 1.20X1018 1:3.1X107 AAV5 2.96X 1010 integrin 2.0 M 1.20X1018 1:4.1X107 AAV1 3.35X 1010 integrin 5.0 M 3.01X1018 1:9.0X107 AAV2 3.84X 1010 integrin 5.0 M 3.01X1018 1:7.8X107 AAV5 2.96X 1010 integrin 5.0 M 3.01X1018 1:1.0X108 AAV1 3.35X 1010 A20 Mab 2.2 M 1.33X1018 1:4.5X107 AAV2 3.84X 1010 A20 Mab 2.2 M 1.33X1018 1:3.5X107 AAV5 2.96X 1010 A20 Mab 2.2 M 1.33X1018 1:4.5X107 AAV2 3.84X 1010 10 M 6.02X1018 1:1.6X108 AAV2 3.84X 1010 integrin 3.0 M 1.81X1018 1:4.7X107 AAV2 3.84X 1010 1.5 M 3.03X1017 1:2.4X107 AAV2 3.84X 1010 20 M 1.24X1019 1:3.1X108 CM3 3.35X1012 AAV2 150 nM 5.42X101 8 1:1.6X106 3.35X1012 Mab 1.0 M 6.02X1017 1:1.8X105 aI ndicates the ligand immobilized to the chip, bindicates the number of viral proteins or molecules c indicates the analyate that is flowed over immobilized ligand, dindicates the analyte molarity, eindicates the number of analyte molecules, findicates the SPR ratio (ligand VPs or molecules/analyte molecules). Abbreviations: Viral proteins (VP), adenoassociated virus serotype 1 (AAV1), adenoassociat ed virus serotype 2 (AAV2), adenoassoc iated virus serotype 5 (AAV5), monoclonal antibody (Mab), and surface plasmon resonance (SPR).
65 Table 25 Ratios of AAV2 and A20 Mab and A20 Fab analyzed by Surface Plasmon Resonance Chip Liganda VP/Moleculesb Analytec Molarityd Moleculese Ratiof CM3 AAV2 9.03X108 A20 Mab 1.6 nM 9.63X101 4 1:1.1X106 AAV2 9.03X108 A20 Mab 8.0 nM 4.82X101 5 1:5.3X106 AAV2 9.03X108 A20 Mab 40 nM 2.41X1016 1:2.7X107 AAV2 9.03X108 A20 Mab 200 nM 1.20X101 7 1:1.3X108 AAV2 9.03X108 A20 Mab 1000 nM 6.02X1017 1:6.7X108 CM3 AAV2 9.03X108 A20 Mab 0.5 nm 3.02X1014 1:3.3X105 AAV2 9.03X108 A20 Mab 2 nM 1.21X1015 1:1.3X106 AAV2 9.03X108 A20 Mab 5 nM 3.02X101 5 1:3.3X106 AAV2 9.03X108 A20 Mab 10 nM 6.04X1015 1:6.7X106 AAV2 9.03X108 A20 Mab 5 0 nm 3.02X101 6 1:3.3X107 AAV2 9.03X108 A20 Mab 100 nm 6.04X1016 1:6.7X107 AAV2 9.03X108 A20 Mab 5 00 nm 3.02X101 7 1:3.3X108 AAV2 9.03X108 A20 Mab 1000 nm 6.04X1017 1:6.7X108 CM3 AAV2 1.36X1010 A20 F ab 20 nm 1.21X1015 1:8.9X104 AAV2 1.36X1010 A20 F ab 100 nm 6.04X1016 1:4.4X106 AAV2 1.36X1010 A20 F ab 250 nm 1.51X1017 1:1.1X107 AAV2 1.36X1010 A20 F ab 5 00 nm 3.02X101 7 1: 2.2 X107 AAV2 1.36X1010 A20 F ab 1000 nm 6.04X1017 1:4.4X107 AAV2 1.36X1010 A20 F ab 2500 nm 1.51X1018 1:1.1X108 AAV2 1.36X1010 A20 F ab 5 000 nm 3.02X101 8 1: 2.2 X108 AAV2 1.36X1010 A20 F ab 7500 nm 4.52X101 8 1: 3.3 X108 AAV2 1.36X1010 A2 0 F ab 10000 nm 6.04X1018 1:4.4X108 aindicates the ligand immobilized to the chip, bindicates the number of viral proteins c indicates the analyate that is flowed over immobilized ligand, dindicates the analyte molarity, eindicates the number of analyte molecules, findicates the SPR ratio (ligand Vps/analyte molecules). Abbreviations: Viral proteins (VP), adeno associated virus serotype 1 (AAV1), adenoassociated virus serotype 2 (AAV2), adeno associated virus serotype 5 (AAV5), monoclonal antibody (Mab), fragmented antibody (Fab), and surface plasmon resonance (SPR).
66 Table 26 R atios of AAV2 with integrins and defensins analyzed by negative stain EM and cryoEM Experiment Virus VPa Protein Moleculesb Ratioc Negative stain EM AAV2 1.81X1016 2.34X1015 8:1 AAV2 1.50X1016 1.47X1016 1:1 AAV2 1.50X1016 1.62X1016 1:1 AAV2 8.13X1015 HNP1 2.04X1016 1:2.5 AAV2 8. 13X1015 HNP2 2.04X1016 1:2.5 AAV2 8.13X1015 HD5 2.04X1016 1:2.5 Negative stain EM AAV2 1.86X1016 2.41X1016 1:2 with crossilnkers AAV2 1.86X1016 3.72X1013 1:0.002 Cryo EM AAV2 1.71X1016 1.77X1016 1:1 AAV2 1.62X1016 HD5 4.07X1016 1:2.5 aI ndicates the number of viral proteins,bindicates the number of protein molecules, cindicates the complexing ratio (viral proteins/protein molecules). Abbreviations: Viral proteins (VP), electron microscopy (EM), cryo E M (cryo reconstructions), adenoassociated virus serotype 2 (AAV2), defensin 1 (HNP1), defensin 2 (HNP2), defensin 5 (HD5).
67 Table 27 Statistics for cryo EM reconstructions AAV2 AAV2 AAV2 AAV2 HD5 Parameters icosahedral asymmetric Icosahedral icosahedral No. of micrographs 44 44 44 40 Start No. of part. 1795 1795 1070 2119 Final No. of part. 1734 1734 981 2105 Defocus range (m) 1.2 5.5 1.2 5.5 1.2 5.5 1.5 3.0 Box size 196 196 256 196 Delta values () 5, 4, and 3 7 and 4 7, 4, and 2 7, 4, and 2 Resolution () 13.8 N/A 17.2 9.7
68 Figure 21. Schematic flow chart of integrin AAV2, and A20 production and purifications. (A) AAV2, (B) Integrins, and (C) A20 Mab and A20 Fab. A B C
69 Figure 22 A20 Fab purification scheme. (A) A20 IgG is treated with (B) papain resin to cleave the Fc portion from the two Fabs. (C)The mixture is flowed over Protein A resin in which the Fc portion binds Protein A and the Fabs are purified in the flow through. A B C
70 Figure 23 Surface plasmon resonance. (A) A typical sensorgram depicting association, dissociation, and regeneration phases. (B) Singlecycle kinetics experiment with increasing concentrations of analyte. A regeneration step is performed after the last analyte injection. (C) Multi cycle kinetics experiment with various concentrations of analyte. A regeneration step is performed after each analyte injection. The diagram was provided by GE Healthcare. A B C
71 Figure 24 surface regeneration using 50 mM diethylamine. A B C
72 Figure 25 immobilization, 2) AAV2 injection, and 3) direct binding measurements are closed conformation. 1 2 3
73 Figure 26 A single cycle A20 IgG kinetics experiment. Steps: 1) A20 IgG (red) is immobilized on a CM3 chip, 2) AAV2 is injected and captured, 3) A20 IgG (yellow) is injected and direct binding measurements are taken between AAV2 and A20 IgG (yellow).
74 Figure 27 A multi cycle A20 IgG kinetics experiment. Steps: 1) AAV2 immobilization, 2) A20 IgG (red) injection, 3) direct binding measurements are taken between AAV2 and A20 IgG.
75 Figure 28. Schematic flow chart of the cryo EM reconstruction process. The red box encompasses the steps used for refinement.
76 CHAPTER 3 INTEGRIN EXPRESSION PURIFICATION, AND INTERACTION WITH AAV2 B ackground The goal of this study was to produce highly purified, soluble recombinant integrin as well as AAV2 VLPs for stru cture function studies A previous study identified receptor for AAV2 and its role in mediating viral entry into the cell (112) Understanding viral entry can have important implications for elucidating the mechanisms that dictate tissue tropism for the rational design of gene delivery vectors. Integrins are cell adhesion receptors that transmit information via inside out and outsidein signaling mechanisms that link both intracellular and extracellular environments (127, 128) subunits each containing extracellular, transmembrane, and cytoplasmic domains. In combinations that have a multitude of biological functions including fertilization, embryonic development, immune responses, platelet aggregation, and receptor attachment (129, 130) Furthermore, integrin defects have been implicated in various cancers and numerous human diseases, including genetic and autoimmune disorders (131) most widely studied and serves as a receptor for extracellular matrix protein vitronectin, implicated in neural crest cell migration, angiogenesis, and tumor progression (132) that lead to embryonic lethality and neuroepithelial defects, resp ectively (127) This receptor is thus widely studied in the field of receptor biology in efforts to understand its
77 function. The identificat ion of the k structural, cell biology, and signaling standpoint s will elucidate the underlying mechanisms involved in integrin biology and provide information for the development of therapeutics. This effort was undertaken bec ause currently available commercial sources provide limited amounts of samples, which are often lyophilized leading to decreased protein activity, and are reconstituted with additional ingredients, for example, Octyl D glucopyranoside, TritonTM X 100, an d sodium azide that are inhibitory to many experimental procedures. Furthermore, there are only a few example studies in the literature that focus solely on expression of the extracellular domains of integrins for ligand receptor interaction characterizati on where the insoluble transmembrane domains are removed (133) The approach utilized here aimed to create an expression system that would ensure post suitable quantities of the dimerized domains for biophysical and structural experiments. ing the pQE TriSystem vector, an 8X histidinetagged plasmid that possesses the necessary molecular components to facilitate expression in bacteria, insect, and mammalian cells. The protein was successfully purified in a one step method employing nickel column chromatography with yields up to 2 mg per cell factory (equivalent to forty, 15 cm plates). The sequence of the protein and its functional state were verified by mass spectrometry and inhibition assays, respectively. In addition to VLPs were characterized by negative stain EM for intact capsids and probed with conformational antibodies for capsid integrity. The AAV2 VLPs were analyzed for
78 Results and Discussion Construct Development R ationale integrins, with the primer code to locate the primers used (Table 2 1) for amplification of the vari ous domains is shown in Figure 31 Since the pQE TriSystem vector system incorporates a eukaryotic Kozak sequence for translation initiation at the Nco I restriction site, primers were designed for inframe insertion in this region. The size of the DNA products obtained by restriction digests of the vectors following PCR amplification of the extracelluar, Fos, and Jun domains, and ligation into the vector we re o f the expected sizes (Figure 32 observed at ~150 bp (Figure 32 A). The size of the fragments from the rest riction 3 2 B and C). The Fos and Jun digestion product bands were less intense in the agarose gels compared to the integrin domains possibly due to reduced incorporation of ethi dium bromide in the small DNA products. In these constructs the Fos and Jun dimerization domains were inserted in place of the transmembrane domains. Integrin Expression in HEK 293 Mammalian Cells and P urification HEK 293 cells were selected for the e physiologically relevant system in which the proteins will undergo post translational modifications, such as glycosylation, as reported to be important for integrin receptors (134136) An expression time course analysis determined that optimal protein expression was at 48 h p ost transfection (Figure 33 lane 11). His tagged integrins
79 were identified in cell lysate fractions obtained from the cell pellet but were not secreted into the media. In the one step purification of the expresse d integrins using a nickel column, cellular proteins were removed by the 60100 mM imidazole concentrations (Fig ure 34 lanes 48), while the bound integrin, migrating at apparent molecular weights of ~150 50 mM elution fractions (Figure 34 A, lanes 9PAGE, probably due to post translational modifications seen w ith other integrin heterodimers (137) Th that it is post translationally modified. The yields are 23 mg per cell factory based on the pool ed elution fractions (Figure 34 A lanes 1017and Figure 34 B & C). These are co nsistent with those estimated for recombinant protein expression in mammalian cells that range from 0.01% 1% of the total protein content given our starting material of ~500 mg. The onestep purification method reduces recombinant protein losses that are typically seen when multiple rounds of purification are necessary. However, we do MS/MS analysis o f th e two integrin bands (Figure 34 B) confirmed the presence of the multiple unique peptides for in sequences (Figure 35 ) and combined with SDS PAGE analysis we estimate purity greater than 90% 95% (Figure 34 B). In approximately 20% of the purification runs a contaminating ~50 kDa protein was eluted within the first few fractio ns (data not shown). This protein was sequenced by LC MS/MS and identified as the mammalian transcription factor, Non POU domain-
80 containing octamer binding protein. This protein contained numerous patches of basic amino acid residues predicted to have mediated the interactions with the nickel resin. Concentrators with a 150kDa cutoff were employed for removal of the contaminating protein. ntegrin The conformational integrity of the expressed integrins was investigated wi th a studies including endocytosis of crocidolite fibers in human epithelial cells (138) integrin binding in photoreceptor rods in human retinal pigment epithelium (139) adenoviral gene therapy in bladder carcinoma cells (140) Co immunoprecipitation experiments integrin heterodimer (Figure 36 A, lane 8) in comparison to the noantibody control (Figure 36 lane 4). Nonspecific binding, albeit he protein A beads was also observed, however, the s greatly enhanced (Figure 36 A, compare lanes 4 (Figure 3 6 A, lanes 9 16) did not interact with the P1F6 MAb with the majority of the sample in the unbound fraction (Figure 3 6 A, lanes 10 and 14) and the lack of positive signal in the elution fractions (Figure 3 6 further substantiated by a Native dot blot in which the nondenatured (unboiled) sample was recognized while the denatured (boiled) sample was not ( Figure 36 B). These heterodimer exists in a native folded state. This verifies that the use of the Fos and Jun dimerization domains as a molecular zipper was effective in compensating for the function of the transmembrane domain of this heterodimer. The Fos and Jun
81 dimerization domains fold into two parallel coiledcoil heterodimers and have been used complex (MHC) HLA DR2 proteins in yeast, and T cell receptor (141, 142) As observed in these other studies, this utility decreased protein aggregation, which is essential for functional studies. Baculovirus AAV2 Purification and C haracterization To purify baculovirus expressed AAV2, cell lysates were run on iodixanol gradients. Typically, the 25% iodixanol fraction contains the highest concentration of empty (i.e. no DNA containing particles) AA V2 particles compared to 40% 25%, 40%, and 4060% fractions (data not shown). Therefore, the 25% fraction was further purified on a Q sepharose column and the peak fractions were analyzed (Figure 3 7) The FPLC chromatogram shows three peaks, however, AAV2 empty particles are routinely eluted from the first peak (fractions 1017) only and thus this peak was collected and analyzed by SDS PAGE (Figure 3 8 A), while the other two peaks contained cellular contaminants and iodixanol (data not shown), respectively Fractions 1116 contained bands corresponding to the expected VP1 (87 kDa), VP2 (73 kDa), and VP3 (62 kDa) sizes. Since the OD (optical density) 280 readings of these purified samples ranged from 0.7 mg/ mL 1.5 mg/ mL no concentration steps were perform ed. Due to its high concentration (~1.5 mg/ mL ), fraction 12 was selected and analyzed by native Dot blot (Figure 3 8 B). The sample was probed with conformational antibodies, A20 and C37 (99) in which the native (unboiled) sample showed a positive signal compared to the denatured (boiled) sample. In addition, B1, an antibody that reacts with a linear epitope recognized only the d enatured sample. In addition, the sample contained intact capsids as visualized by negative stain EM (Figure 38 C).
82 Transduction Inhibition A ssay Transduction inhibition studies were completed for two reasons: 1) Determine if can compete with cell surface receptors for AAV2 binding, and 2) t o active. For these reasons we tested the transduction efficiency of scAAV2EGFP with and without preincubating the vi int egrin in HeLa cells (Figure 39 ) From the results shown in Figure 39 A it is evident that whereas the mock infected HeLa cells showed no green fluorescent protein signal, the transduction efficiency of scAAV2EGFP alone was the highest c ompared to the preintegrin. Specifically, the transduction efficiency of scAAV2EGFP was inhibited by 75% and 23% when premL (1 virus: 6.0X104 integrin; Table 23) and (1 virus: 6.0X103 integrin; Table 23) to scAAV2 EGFP alone (Figure 3 9 B). In summary, this study clearly demonstrates that receptor interacti on in a biological assay. Surface Plasmon R esonance In order to determine binding interactions of soluble with intact AAV2 particles, we performed binding studies via surface plasmon resonance. AAV2 particles were amine coupled to a biosensor chip and various amounts of soluble integrin were passed over the im mobilized particles (Figure 310). The first injection of the positive control, A20 Mab, resulted in an increase of 640 RU. Based on the amount of AAV2 coupled to the chip (Table 24) and the RU i ncrease obtained after the A20 M ab injection, there are roughly five A20 Mabs binding per capsid (60 binding sites). Although, the stoichiometry of this interaction may be considerably higher since the A20
83 Mab injection served soley as a pos itive control and nonsaturating conditions of virus were not tested. Following surface regeneration with 50 mM diethylamine to remove the A20 Mab, three injections of integrin was conducted with concentrations ranging from 4.1 M to 8.2 M in the presence of different divalent cations. These injections of integr in resulted in no change in RU, despite using higher concentrations concentrations compared to the A20 Mab positive control. These data show that interactions with AAV2 may be transient and nondetectable by SPR. These results are not completely unexpected since coreceptors generally serve as a means to concentrate the virus on the cell surface via transient interactions (143) and in some cases, as with the integrins, mediate endocytosis. However, despite these findings, other viruses such as Adenovirus serotype 2 (Ad 2) bind in a relatively tight manner (KD = 73 nM). A possible explanation for this is Ad2 is ~8X larger than AAV2, which based on surface area alone, may accommodate more integrin binding sites. Negati ntegrin Negative stain transmission E M was performed on AAV2 complexed with integrin to visualize an interaction. This preliminary study was performed when integrin was first expressed and purifi ed in the lab. The goal was to capture an AAV2 binding event, despite no n saturating concentrations of Field of views from the integrin only (Figure 311A), AAV2 only (Figure 311B), and AAV2 complexed with integrin (Figure 311 C) are shown. Interestingly, it appears a single integrin heterodimer is binding t o the capsid surface (Figure 311C), however, this is the only binding event captured to date. The lack of integrin binding on the capsid surface may be due to nonsaturating quantities of integrins that
84 were added to virus (8 VP:1) for this complexing study. In addition, this interaction may be asymmetric as discussed in multiple reports including integrin and cytosekeletal interactions in migrating fibroblasts (144) position depdendent linkages of fibronectinintegrin and the cytoskeleton (145) and (81) Summary In this study, the extracelluar domain of integrin was expressed, purified in an efficient oneste p scheme, and the conformational and functional state were characterized. The interactions between AAV2 and integrin were studied by transduction inhibition assays, SPR, and negative stain EM. This interaction was only detectable in the cell based transduction inhibit ion assay where a ratio of 1 VP: 103 to 104 integrin molecules (Table 23) were needed for inhibition. In comparison, the SPR ratio was 1 VP:108 integrin molecules (Table 2 4) yet the interaction was not detectable These results suggest that the AAV2integrin interaction is weaker than AAV2 A20 Mab since roughly the same amounts of both of these proteins were flowed over immobilized AAV2. In addition, there may be a cellular factor that promoted the AAV2 integrin in teraction in the transduction inhibition experiment that was not present in the SPR study. The results of the negative stain EM experiment showed that there may be one potential integrin binding to a capsid, however, these concentrations were not saturating and the experiment will be repeated with saturating amounts of integrin.
85 Figure 31 (B) extracellular domains were inserted in frame with a glycine linker, Fos and Jun dimerization domains, respectively, and an 8X histidinetag. The positions of the primers (Pr, see Table 21) with their unique enzyme restriction sites used for plasmid assembly are show n.
86 Figure 32 P CR amplification of cDNAs and verification of assembled plasmids by (2.97 kb), Fos (0.153 kb), and Jun (0.153 kb). (B and C) Restriction enzyme Jun, re spectively. DNA bands verify that the correct plasmid sizes were present.
87 Figure 33 CP, (2) 0 h S, (3) 12 h CP, (4) 12 h S, (5) 24 h CP, (6) 24 h S, (7) 32 h CP, (8) 32 h S, (9) 40 h CP, (10) 40 h S, (11) 48 h CP, (12), 48 h S. Cell pellet and supernatant are abbreviated CP and S, respectively. The Western blot was probed with an anti His antibody.
88 Figure 3 4 Nickel column purification of HEK293expressed integrin. (A) SDS PAGE, Lanes: (1) Molecular weight marker (M), (2) cell lysate (CL), (3) flow through (FT), (4) 60 mM imidazole wash 1 (W1), (5) 80 mM imidazole wash 1 (W4), (6) 100 mM imidazole wash 1 (W7), (7) 100 mM imidazole wash 2 (W8), (8) 100 mM imidazole wash 3 (W9), (917) elutions 19. Three wash steps were performed for each imidazole concentration; however, only representative washes are shown. (B) SDS PAGE of pooled and concentrated elution fract ions 29 (shown in lanes 1017 in (A) ) (C) Western blot of sa mple shown in B probed with an anti H is antibody A B C
89 Figure 35 showing the amino acid sequence for the proteins with the unique peptides identified highlighted in yellow. Cysteines and methionines are shown in green. A B
90 Figure 36 Characterization of integrin V 5 using a conformational anti V 5 antibody. (A) Western blot of fractions from a coimmunoprecipitation of cell lysate (CL) with anti V 5 integrin antibody (P1F6) showing cell lysate (CL), unbound (U), wash (W), and elut ion (E) for different transfections as labeled at the bottom of the figure. Asterisks indicate the heavy and light chains of the antibody of the P1F6. The W ester n blot was probed with an anti H is antibody. (B) Native dot blot of unboiled and boiled (denatured) integrin V 5 probed with the anti V 5 antibody. B A *
91 F igure 37 A chromatogram showing peak fractions of an AAV2 purification via ion exchange chromatography Peak 1 pure AAV2 VLPs, Peak 2 cellular contaminants, Peak 3 iodixanol. Peak 1 Peak 2 Peak 3
92 Figure 38 Ion exchange purification of SF9expressed AAV2. (A) SDS PAGE, Lanes: (1) Molecular weight marker (M), (2) flow through, (3) wash, (411) fractions 1017. (B) Dot blots of fraction 12 probed with Mabs A20, C37, and B1. (C) Fraction 12 analyzed by negative stain EM Scale bar is at 100 nm. B C 100 nm A20 Unboiled Boiled C37 B1 A 1 2 3 4 5 6 7 8 9 10 11
93 Figure 3 9 Comparative analyses of AAV2mediated transduction of HeLa cells with V 5 integrin inhibition. (A) Cells were mock infected, infected with scAAV2 EGFP alone, and infected with scAAV2mL mL of V 5 integrin at an MOI of 1,250 viral genomes/cell. Transgene expression was detected by fluorescence microscopy 72 h post infection. A representative visual field view is shown from each triplicate experiment. Original magnification X100. (B) Quantitative analyses of AAV2 transduction efficiency in HeLa cells. Images from three visual fields were analyzed quantitatively by ImageJ analysis software. Transgene expression was assessed as total area of green fluorescence (pixel2) per visual field (mean SD). ANOVA was used to compare test results with the control, and they were determined to be statistically significant. p <0.05 or ** p <0. 01 vs scAAV2 EGFP alone A B Mock Virus alone
94 Figure 3 10 used to regenerate the AAV2 surface. Injection 1: A20 IgG (2.2 M) Response: + ~640 RU response M) Response: 0 RU M) Response: 0 RU M) Response: 0 RU Injection 2: 50 mM diethylamine
95 Figure 311. Negative stain EM of AAV2 and integrins. (A) Integrins alone. (B) AAV2 alone. (C) AAV2 complexed with integrins. EM images were taken at 50K magnification. B A C
96 CHAPTER 4 AND INTERACTION WITH AAV2 B ackground The goal of this study was to produce highly purified, s oluble recombinant integrin and AAV2 VLPs for structurefunction studies (similar to Chapter 3) A previous integrin as a coreceptor for A AV2 and its role in mediating viral entry into the cell (26) studied integrin known for its direct interactions with the extracellular matrix protein, fibronectin. These interactions are directly involved in embryogenesis, angiogenesis, and wound healing (146) integrin expression between normal and tumor cells (~10 fold increase) support involvement in tumor progression and aggressiveness (147) It has been shown that a integrin in mice results in defects in neural tube development and blood cell leakage during embryogenesis (148) exhibit vascular remodeling defects caused by adhesion and migration alteration and reduced survival of endothelial cells (149) integrin is also studied for its roles in cellsurface virus attachment and internalization (26, 114, 150) leukocyte rolling (151) and cytoskeletal activation and rearrangement (1 52) ively studied receptor since it plays a role in many facets of biology. Moreover, the ability to express and purify this receptor in the quantities needed for numerous assays including structural and cell based is cr ucial to the integrin receptor field.
97 The purpose of this study was to produce a highly soluble and functional integrin for studies with AAV2 We utilized the pFastBac vector with an 8X histidine tag as the vector backbone for production of recombinant bac uloviruses. Each baculovirus was frame with Fos and Jun dimerization domains, respectively. Coinfections were performed in Spodoptera frugiperda 9 (SF9) cells and an efficient onestep nickel column purification was employed that yielded purified protein amounts ~4 mg per liter. The purified integrin sample was analyzed for interactions with AAV2 via co immunoprecipi tation, negative stain EM, cryoEM transduction inhibition, and s urface plasmon resonance. Results and Discussion Cons Integrin S ubunits integrins, with the primer code to locat e the primer s (Table 22 ) used for amplification of t he various domains, are shown in Figure 41 A K ozak sequence was inserted into the original pFastBac plasmid vector for translation initiation at the Sal I and Bam HI sites, directly upstream of the integrin extracellular domains, which are followed by Fos and Jun domains, respectively. This was done due to numerous reports in the literature suggesting that a Kozak sequence enhance protein expression in insect cells (153155) An 8X histidine tag, inserted in addition to a stop codon for translation termination inframe with the Fos and Jun domains, enable a onestep nickel column purification. The Fos and Jun domains, known to self dimerize, were inserted in transmembrane domains to prevent prot ein aggregation and precipitation which is often observed when expressing and purifying membrane
98 proteins. A restriction enzyme domains from the rest of the and ~2.2 kb expected (Figure 4 2A). Five plaquepurified baculovirus Fos and jun DNA (Figure 42B and C) Small Scale Protein Expression, Purif ication, and LC ubunits The use of SF9 insect cells for protein expres sion ensures high yields and correct post translational processing (156) A small volume (50 mL ) expression in SF9 cells followed by nickel column purification was used to validate the integrin constructs. Infection of SF9 cells with the highest titered baculovirus recombinant for each integrin 5, following amplification to a P3 stage, produced soluble integrins in the media fraction that were purified via nickel column chromatography for 5 integrins (Figure 41 (Figure 43B and C). Eluted fractions contain when visualized by SDS PAGE. The bands are higher than the theoretical molecular ~ 85 kDa) integrin. A possible explanation for this observed higher gel migration is glycosylation of the integrin heterodimer, which is required for functional activity (133). The eluted bands ( elution 3, Figure 4 3A and B) MS/MS (Figure 4 4A and B). Based on previous cell pellet purifications (data not shown) the ratio of secreted to cytosolic integrins are typically 75% to 25% despite the lack secretion signals in th e integrin constructs. In addition, integrin purification from the PEG precipi t ated media samples were more efficient at 90% 95% purity compared to the samples from cell
99 lysates isolated from freezethawed cell pellets at 60% 70% pure. This is most likely due to higher levels of contaminating host cell proteins in cell lysates compared to the PEG precipitated media. Large Scale Protein E x pression, Purification, and Dot Blot C haracterization To produce large (up to 4 mg) 1 L volumes of SF9 insect c ells were routinely coinfected with produce the integrin heterodimer followed by a onestep nickel column purification C ellular proteins were removed by 100 mM and 150 mM imidazole concent rations (Figure 45A, lanes 34), while the bound integrins were detected in the 400 mM imidazole elution fractions (Figure 45A, lanes 517). Washing with 60100 mM imidazole failed to remove cellular contaminants, while stringent washing steps of 100150 mM imidazole was effective. A yield of ~4 mg of purified per 1 L of infected SF9 cells was obtainable from pooled (Figure 4 5A, lanes 5 17) and concentrated (Figure 4 5B) elution fractions. A loss of ~ 10% 20% was estimated to occur due to protein adhesion to the surface membrane of the concentrator used at the last purification step. determined by Dot blot analysis (Figure 45C) with function blocking monoclonal antib odies, anti (157) and anti (158) The nondenatured (unboiled) sample was recognized by both Mabs, while the nonnative (boiled sample) reacted to the anti only. Both antibodies have been reported to recognize conformational epitopes, however, the anti also reported to recognize a linear epitope consistent with the observed positive signal for the denatured sample. These
100 observations are consistent wi th the adoption of a native state by the expressed and purified integrin heterodimer. Co I mmunoprecipitation of 1 I ntegrin and AAV2 A co immunoprecipitation experiment was conducted to: 1) Determine if an anti s only the integrin subunit could pull down the u nit and 2) if the 4 6). In the first experiment, the anti integrins (Figure 46A, l ane 9), however, when compared to the beads and integrin control (Figure 46A, lane18), the integrins appear to bind nonspecifically to the beads. In the second experiment (Figure 46B, lanes 46), AAV2 appears to interact with more AAV2 is found in the elution fraction (Figure 46B, lane 6) compared to the wash fraction (Figure 46B, lane 5), which suggests AAV2 is being eluted from the beads. On the contrary, a control experiment showed that AAV2 interacts nonspecifically with the beads (Figure 46B, lanes 1012). The two results described are inconclusive due to nonspec ific binding events. To date, a and AAV2 interaction has not been empirically determined by immunoprecipitation. Transduction Inhibition A ssays P revious studies have shown that is a co receptor for AAV2 and serves to promote virus internalization into the cell (24). Thus to further demonstrate that the expressed and purified nativ e functional state, the ability of the purified heterodimer to compete for receptor attachment by AAV2 was tested in a transduction assay. HeLa cell transduction efficiency by scAAV2 EGFP with and without preshowed that the protein was capable of inhibiting cellular internalization (Fig ure 4 7).
101 Presumably by competing with endogenous integrin binding. M ockinfected HeLa cells showed no GFP transgene expression, the virus alone sample showed efficient GFP expression, while preincubation with the integrin heterodimer resulted in a 68% decrease in transduction as previously reported (24). This capacity of the expressed ability to mimic a reported virus receptor interaction. Surface Plasmon R esonance In order to determine binding interactions of soluble and integrin with intact AAV2 VLPs we performed binding studies via SPR AAV2 particles were amine coupled to a biosensor chip at a low surface density (187 RU) and 2 M to 6 M amounts of soluble and were flowed over the im mobilized particles (Figure 48 ). A low surface density of AAV2 was employed to decrease mass transport effects of individual integrin molecules from binding to many AAV2 particles which have the potential to skew kinetic analyses. The first injection of integrin resulted in a n increase of 80 RU (Figure 48 A) and a 7 VP: 1 integrin ratio. For integrin, multiple injections of 2 M to 6 M concentrations resul ted in RU increases ( highest of 30 RU with a 8 VP:1 integrin ratio; Figure 48 B and C), which suggest a binding interaction for the heterodimer and individual subunit. Previous SPR studies with Adenovirus 2 and at divalent cations are necessary for binding and that the addition of 20 mM EDTA mixed with the integrin sample or used as a regeneration buffer abolishes these divalent cationdependent interactions (159) Based on the EDTA data with Adenovirus integrins, 20 mM EDTA was utilized as a regeneration buffer (Figure 48 C, injection 8) and also directly mixed in with integrin sample that included 1 mM CaCl2, 1 mM MgCl2, and 1 mM MnCl2 (Figure 4 8 C, injection 9), respectively. Both of
102 these results show that the addition of EDTA removed bound integrin from AAV2 and when added directly to the integrin sample, no binding occurred as can be seen by lack of increase in RU following injection. These results demonstrate the essential roles played by divalent cations to facilitate, in this case, integrin activation for virus receptor interactions. In addition, kinetic tests were performed, but these analyses (data not shown) resulted in no detectable interactions. This is most likely due to the higher flow rates (30 L /min) required for kinetic studies compared to the slower flow r ate (5 L/min) used in this analysis. Manual runs with slow flow rates are typically used first in SPR studies to confirm ligandanalyte interactions, followed by kinetic analysis if the manual runs confirm binding. A second set of experiments using the same AAV2 sample as the first experiment was designed to repeat the AAV2integrin binding (Figure 49 A) and also test integrin binding to AAV1 and AAV5. All three AAV serotypes were immobilized at high surface densities (3709 RU AAV1, 4250 RU AAV2, and 3 271 RU AAV5) to ensure adequate virus coverage on the sensor chip for integrin binding. The first injection of integrin resulted in positive RU responses for all three AAV serotypes, while the second injection showed gains in only AAV1 and AAV2 (Figure 4 9 A). The minimal RU increases in injection 2 may be a consequence of not regenerating the surface after injection 1 to remove bound integrin. Some of these initial runs were soley conducted to identify a binding interaction, which resulted in multiple sample injections without regeneration. The idea was to preserve the coupled ligand on the chip without disrupting its conformational integrity w ith harsh regeneration buffers so additional kinetic analyses could be performed. Next, the rationale for testing AAV1 and
103 AAV5 was based on amino acid sequence alignments, both AAV1 and AAV2 contain an NGR motif, but AAV5 contains an EGA motif its place. The NGR motif has been proposed as the integrin binding site on the AAV2 capsid surface (26) Despite this report, AAV5 does show minimal binding to integrin (Figure 49 A, injection 1), but at the expense of the typical response curve seen with A AV2. To date, t he EGA motif has not been documented as a binding site for integrin This result suggests a number of possible explanations: 1) the EGA site is an undocumented recognition sequence for integrin. This could be further investigated by site directed mutagenesis of the EGA sequence and/or a structural approach to characterize these interactions. 2) Another possibility is that the level of integr in binding to AAV5 may indicate a level of nonspecific binding. Based on the RU responses and sensorgram curves, AAV2 appears to have the strongest binding of the three AAV serotypes, followed by AAV1 and then AAV5. The A20 Mab, which specifically recognizes AAV2, was flowed over all three AAVs as a control (Figure 49 B). As expected, a signif icant RU response (2400 RU) which represents an interaction with the coupled AAV2 (4 VP:1 A20 Mab) was determined, while there was no detectable interaction for AAV1 and AAV5. Next, using the same biosensor chip as the previous experiment (Figure 49 A and B), the flow direction was diverted over flow cells 1 (blocked for reference subtraction) and 2 (AAV2) to measure AAV2 interactions with integrin in more detail. A series of injections were conducted with varying concentrations of integrin and diff erent divalent cations. Injections 5 (5 M ; 1 mM CaCl2 and 1 mM MgCl2), 8 (10 M ; 1 mM CaCl2, 1 mM MgCl2 and 1 mM MnCl2), and 9 (3 M ; 1 mM CaCl2, 1 mM MgCl2 and 1 mM MnCl2) resulted in RU increases of 70, 38, and 65 in a non
104 concentrationdependent manner (Figure 49 C). Moreover, the addition or removal of 1 mM MnCl2 did not enhance or inhibit integrin binding. To recapitulate the inhibitory EDTA chelation results in Figure 410C, 20 mM EDTA was added to integrin samples in injections 11 (1.5 M ) 13 (1.5 M ), and 15 (20 M ), all of which contained 1 mM CaCl2, 1 mM MgCl2 and 1 mM MnCl2. Contradictory to the previous results, the 20 mM EDTA did not inhibit AAV2integrin binding despite utilizing the same AAV2 sample for coupling In an effor t to test the AAV2integrin interaction in the reverse manner, integrin was immobilized on a biosensor chip at a high surface density (3104 RU) and A AV2 was flowed over (Figure 49D ). The first injection of AAV2 (150 nM) resulted in no change in R U, while the second injection of the anti integrin antibody, which served as a positive control, showed an RU increase of 557 (1502 :1 integrin Mab ntegrin Dilution Series Visualized by Negative Stain EM Negative stain transmission EM was used to visualize the integrity and conformational state of the purified integrin heterodimer Figure 410 shows images of n egativestain only control (Figure 4 10A) and (Figure 4 10B G ) A dilution series enabled the visualization of individual heterodimers in a number of different conformations at a concentration of 0.003 mg/ mL ( Figure 410B D). The conformation of the integrin molecules in the inset (Figure 410G and H) were observed to be in the V shape or bent conformation, similar to those previously reported for integrins not bound to ligands (160) Thus the purified heterodi mer appears to be in a native state.
105 AAV2 C omplexed with EM Negative stain transmission EM was performed on AAV2 VLP complexed with to visualize binding (Figure 4 11). Complexing of both samples was performed in a 1 VP :1 integrin ratio. The appearance 4 11C) show AAV2 capsids with possible integ rin attachments ( black boxes; Figure 411A). A high integrin concentration has been shown to create a masking effect that makes deciphering binding events difficult (Figure 4 10B D ). Furthermore, as stated in Chapter 3, virus receptor interactions are tr ansient; therefore protein cross linkers were employed in an attempt to capture these binding events (Figure 412). Bis, (sulfosyccinimidyl) substrate (BS3) cross linkers are commonly used to crosslink cell surface proteins via amineamine bonds and have b een shown to link other virus receptor interactions, particularly for Gibbon Ape Leukemia Virus and Amphotropic Murine Leukemia V irus (16 1) Compa risons of AAV2 alone (Figure 412B) with AAV2 complexed with in a 1 VP:1 integrin ratio (Figure 4 12C) and 1:0.002 ratio (Figure 412D) were conducted. Interestingly, the higher the integrin concentration (while keeping capsid concentration constant) the less capsids were seen in a field of view. This suggests that the capsids are potentially aggregating due to AAV2integrin interactions. Additionally, a few capsids with the higher integrin concentration may have a few integrin at tachments decorating the capsid surface (Figure 412 C) co mpared to AAV2 alone (Figure 412 B), however, this is difficult to determine since an uncomplexed integrin could be laying next to the capsid. Therefore, a much lower in tegrin concentration (Fig ure 410G, 0.003 mg/ mL ) was
106 complexed with AAV2 (Figure 412D) to decrease the masking effect and better resolve individual integrin molecules. These results show that most integrin molecules are not binding to capsids and are only seen in the background. Analysis of AAV2Cryo EM of AAV2, integrin was complexed with AAV2 VLPs in a 1:1 ratio (Table 26). Figure 413 shows an icosahedrally averaged AAV213.8 (Figure 4 13A and B) using a box size of 196. The AAV2 PDB crystal structure (1LP3) was converted to density and then scaled and filtered (Figure 413C) to match the AAV2 cryo EM structure factors and then direct comparisons with the AAV2 cryo EM complex (Figure 414D) were conducted. The AAV2 PDB density and AAV2AAV2 PDB density displayed more defined features at the threefold protrusions compared to the AAV2 13D). This phenomenon is a result of scaling the AAV2 PDB density from 3 (AAV2 crystal structure) to 13.8 resolution, which sometimes results in lower resol ution maps retaining higher resolution data. The two maps were further analyzed by calculating difference maps to identify 13E shows the AAV2 AAV2 PDB density map (positive density shown in gray ) and also the AAV2 PDB density map subtracted by the AAV2changes (gray; Figure 413E) when superimposed on the AAV2 PDB density are not visible on the surface of the capsid (Figure 413F). These results suggest the positive density changes are not due to integrin binding. The negative density differences are
107 most likely a result of the AAV2 PDB density map still retaining higher resolution data as can be seen around the threefold axis of symmetry (Figure 413D). The negative density differences suggest the cryo reconstruction may have more ordered density in the core of the structure compared to the original AAV2 crystal st ructure that was converted to PDB density. This phenomenon is not unexpected since the AAV2 crystal structure lacks ordered density for VP1 and VP2, which could account for these diffe rences when compared to the AAV2EM map. Add itional efforts were taken to re examine the data by using a larger box size for particle selection in case integrin attachments were occluded by box size (196) Figure 14A and B shows an icosahedrally averaged AAV2final resolution of 17.8 using a box size of 256 (Table 27) Similar to the approach demonstrated in the previous figure (Figure 4 13), t he AAV2 crystal structure was converted to a density map and scaled appropriately (Figure 4 14C) to match the structure factors of the AAV2cryo EM map. One notic eable difference between maps in Figure s 4 13 and 4 14 is the latter have more defined and resolved features along twofold, threefold, and fivefold axes of symmetry (compare radial images) The most reasonable explanation for this is the increase in box size (196 256) since EMAN2 favors larger box sizes for noise filtering and weighting factors (102). Further analyses of the AAV2cryo EM map and AAV2 PDB density map (Figure 4 14C E) s how no real positive density changes on the surface of the AAV2 capsid. These results are similar to those presented in Figure 413. not detectable (Figure 413 and 414), a different approach was taken to address the
108 asymmetric reconstructions were calculated with a box size of 196 (Table 27; Figure 415). Attempts at determinin g resolution were unsuccessful most likely due to EMAN2 software issues. The AAV2 changes within the same map. This result may be due to a number of reasons: integrin is binding to AAV2 in an asymmetric man ner as shown by an increase in density at the twofold and threefold symmetry axes (compare Figure 415 A with B) and at a lesser extent at the fivefold axis of symmetry (compare Figure 4 15 A with B) These changes may be the result of a ing site footprint. The full length integrin may not be visible since they have been reported to be highly flexible and assume multiple conformations (104) and thus are not easily resolvable. 2) The second hypothesis is there are too few particles to resolve all surface features of the capsid. The particle set contained only 1734 particles in comparison to a CyanBacteriophage asymmetric reconstruction where 15,000 particles were utilized to achieve just 9 resolution (162) Furthermore, attempts at performing an as ymmetric reconstruction with the larger box size of 256 and particle count of 981 failed. The reconstructed map contained gaps of missing information (data not shown), most likely due to the lack o f particles to fulfill all orientations of the AAV2 capsid. Difference map calculations of the AAV2 was not determined. This will be further investig ated. Summary integrin was expressed, purified in an efficient oneste p scheme, and the conformational and functional state were characterized. The interactions between AAV2 and integrin were studied by
109 t ransduction inhibition assays, co immunoprecipiation, SPR, negative stain EM, and cryo EM The AAV2 based transduction inhibition assay where a ratio of 1 VP: 104 integrin molecules (Table 23) were needed for inh ibition. This result further substantiated a prior study that used multiple techniques including solid phase integrin binding, cell adhesion, and viral transduction inhibition es were inconclusive since not all of the experiments were repeatable. However, these results do suggest the AAV2interactions measured by SPR show viruses that have varying levels of affinity for their respective receptors For example, Adenovirus 2 while both CA9 enterovirus (163) and Parechovirus 1 (56) bind to multiple integrin receptor s with low and high affinities For AAV2, multiple cor eceptors (Table 11) have been identified to aid in cellular internalization. Based on this, one of the other nonintegrin receptors may have a stronger binding affinity. integrin was also analyzed from a structural stand point by negative stain EM and cryo EM Both techniques showed that integrin binding to AAV2 may occur in an asymmetric fashion where a few integrin molecules bind to the capsid surface. The AAV2reconstruction map (Figure 413 and 414) did not show integrin binding site footprints or larger integrin molecules when subtracted from AAV2 PDB density maps. In addition, a few of the negative stain EM micrographs (Figure 41 1 B and C; Figure 412C) depict potential integrin attachments on the capsid surface. Other viruses such as Coxsackievirus A9 (55) and Canine parvovir us (164) show asymmetric binding with their
110 respective coreceptors. The results presented here suggest integrin binding does occur in a transient and asymmetric f ashion.
111 Figure 4 (B) extracellular domains were inserted in frame with a glycine linker, Fos and Jun dimerization domains, respectively, and an 8X histidinetag. The positions of the primers (Pr see Table 22 ) with their unique enzyme restriction sites used for plasmid assembly are shown.
112 Figure 42 DNA. (A) Uncut (U) and restriction enzymedigested plasmid DNA for 5 grin are shown. (B and C). PCR amplified bands verify that all five clones for each respective recombinant baculovirus packaged the appropriate DNA. A B C
113 Figure 4 3. Small scale PAGE Lanes: (1) Molecular weight marker (M), (2) Media (M), (3) flow through (FT), (4) 100 mM imidazole wash 3 (W3), (5 9) elutions 1 5. (C) Western blot of elution samples shown in A and B probed with an anti H is antibody. A B C
114 Figure 44 unique peptides identified highlighted in yellow. Cysteines and methionines are shown in green. A B
115 Figure 45. Largeintegrin. (A) SDS PAGE Lanes: (1) Molecular weight marker (M), (2) flow through (FT), (3) 100 mM imidazole wash 1 (W1), (4) 150 mM imidazole wash 2 (W2), (5 18) elutions 114. (B) Lanes: (1) SDS PAGE of pooled and concentrated elution fractions 114 (shown in lanes 518 in (A)), (2) Western blot of same sample shown in lane 1 probed with an anti H is antibody. (C) anti A B C
116 Figure 46. Co 1) Marker, 2) integrin + AAV2 flow through, 5) beads + anti wash 3, 6) beads + anti e lution, 7) beads + anti flow through, 8) beads + anti integrin wash 3, 9) beads + anti elution, 10) beads + AAV2 flow through, 11) beads + AAV2 wash 3, 12) beads + AAV2 elution, 13) beads + anti a5b1 integrin antibody + AAV2 flow through, 14) beads + anti a5b1integrin antibody + AAV2 wash 3, 15) beads + anti a5b1 integrin antibody + AAV2 elution, 16) beads + flow th wash 3, 18) beads + elution. The Western was probed with an anti his antibody. (B) Lanes are the same as (A). The Western was probed with B1 antibody.
117 Figure 46. Continued. 250 150 100 75 50 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 250 150 100 75 50 37 A B
118 Figure 47 Comparative analyses of AAV2mediated transduction of HeLa cells with mockinfected, infected with scAAV2EGFP alone, and infected with scAAV2EGFP and 100 g/ mL integrin at an MOI of 1,250 viral genomes/cell. Transgene expression was detected by fluorescence microscopy 72 h post infection. A representative visual f ield view is shown from each triplicate experiment. Original magnification X100. (B) Quantitative analyses of AAV2 transduction efficiency in HeLa cells. Images from three visual fields were analyzed quantitatively by ImageJ analysis software. Transgene ex pression was assessed as total area of green fluorescence (pixel2) per visual field (mean SD). ANOVA was used to compare test results with the control, and they were determined to be statistically significant. **p<0.01 vs scAAV2 EGFP alone. A B Virus alone V integrin Mock
119 Figure 4 8 ( 2 M) in the presence of MgCl2, CaCl2, MnCl2 the presence of all three divalent c ations at 1 mM, 3) a negative c ontrol 1 mM CaCl2 and 1 mM MgCl2 5) 20 mM EDTA regeneration at 50 L /min for 24 sec, 6) 50 mM diethylamine at 50 L in the presence of all three divalent cations at a 1 mM concentration, 8) 20 mM EDTA regeneration at 50 L ) in the presence of all three divalent cations at a 1 mM concentration + 20 mM EDTA. Inj integrin (2 M) Response: + 80 RU A
120 Figure 48 Continued. integrin (6 M) Response: + 30 RU Injection 3: D11 IgG (10 M) Response: 0 RU B
121 Figure 48 Continued. C integrin (2 M) Response: + 11 RU Injection 8: 20 mM EDTA Injection 6: 50 mM diethylamine Injection 5: 20 mM EDTA integrin (2 M) Response: + 11 RU integrin (2 M) + 20 mM EDTA Response: 0 RU
122 Figure 49 integrin flowed over immobilized AAV1, AAV2, and AAV5 analyzed by integrin (2 M) in the presence of 1 mM CaCl2 and 1 mM MgCl2integrin (5 M) in the presence of 1 mM CaCl2, 1 mM MgCl2 and 1 mM MnCl2. (B ) Injections: 3) A20 antibody (2.2 M) integrin (5 M) in the presence of 1 mM CaCl2 and 1 mM MgCl2, (6) 20 mM integrin (10 M)integrin (10 M ) in the presence of all three divalent cations at a 1 mM concentration, 9) integrin (3 M) in the presence of all three divalent cations at a 1 mM concentrations, 10) 50 mM diethy l a integrin (1.5 M ) + 20 mM EDTA in the presence of all three divalent cations at a 1 mM concentration, integrin (1.5 M ) + 20 mM EDTA in the presence of all three divalent cations at a 1 mM concentration, integrin (20 M ) + 20 mM E DTA in the presence of all three divalent cations at a 1 mM concentration.
123 Figure 49 Continued. A AAV2 AAV1 AAV5 integrin (2 M) AAV2: + 69 RU AAV1: + 32 RU AAV5: + 15 RU integrin (5 M) AAV2: + 30 RU AAV1: + 9 RU AAV5: 0 RU
124 Figure 49 Continued. AAV2 AAV1 AAV5 Injection 3: A20 IgG (2.2 M) AAV2: + 2400 RU AAV1: 0 RU AAV5: 0 RU Injection 4: 50 mM diethylamine B
125 Figure 49 Continued. integrin (5 M) Re sponse: + 70 RU Injection 6: 20 mM EDTA Injection7): 50 mM diethylamine integrin (10 M) Response: + 38 RU integrin (3 M) Response: + 65 RU Injection 10: 50 mM diethylamine integrin (1.5 M ) + 20 mM EDTA Response: + 50 RU Injection 12 & 14: 50 mM diethylamine integrin (20 M ) + 20 mM EDTA Response: + 70 RU integrin (1.5 M ) + 20 mM EDTA Response: + 70 RU C
126 Figure 49 Continued. Injection 1: AAV2 (150 nM) Response: 0 RU Injection 2: anti integrin IgG (1 M ) Response: + 557 RU D
127 Figure 4 10. A dilution series integrin visualized by negative stain EM (A) Control grid with uranyl formate stain only. stock (0 .6 mg / mL ) (C) 2X dilution. (D) 10X dilution. (E) 20X dilution. (F) 100X dilution. (G) 200X dilution. (H ) Close up view of ( G ). Samples were stained with uranyl formate and images were taken at 50 K magnification. Scale bar is 100 nm. A B C D E F G H
128 Figure 411 integrin integrin in a 1 VP :1 integrin ratio. Black boxes contain AAV2 particles with possible integrin attachments. Samples were stained with uranyl formate and images were taken at 50 K magnification. Scale bar is 100 nm. A B C
129 Figure 412integrin and AAV2 cross linking experiments visualized by negative stain EM. (A) Control grid with integrin in a 1 VP :1 integrin ratio. (D) AAV2 integrin in a 1:0.002 ratio. Black boxes contain AAV2 particles with possible integrin attachments. Images were taken at 50 K magnification. Scale bar is 100 nm A C D B
130 Figure 413. Three dimensional icosahedral reco nstruction of AAV2 complexed with integrin using a box size of 196. (A) Radial coloring of the final density map of the AAV2 integrin reconstruction at 13.8 (B) Same as (A) except colored in gray. (C ) AAV2 PDB (1LP3) scaled and filtered density map to match /pixel, resolution, box size, and structure factors of AAV2 integrin complex density map. (D) An overlay of (B) and (C) using the Fit function in Chimera to match surface features between both maps with >95% correlation. A box around a threefold axis depicts density differences between both maps. (E) Difference map calculations between the (B) AAV2integrin reconstruction map subtracted by the (C) AAV2 density map. The positive density differences are shown in gray. Negative density differences were calculated by subtracting (B) from (C) and are shown in blue. (F) An overlay of the (C) AAV2 PDB density map with the positive and negative density differences shown in (E). The positive density differences are not visible on the surface of the AAV2 PDB density map. All views are down the twofold axis of symmetry. The cryo EM parameters are shown in Table 27
131 Figure 413. Continued. 1 1 4 1 32 A B C D E F 3 fold 5 fold 2 fold
132 Figure 414. Threedimensional icosahedral reconstruction of AAV2 complexed with integrin using a box size of 256. (A) Radial coloring of the final density map of the AAV2 integrin reconstruction at 13.8 (B) Same as (A) except colored in gray. (C) AAV2 PDB (1LP3) scaled and filtered density map to match /pixel, resolution, box size, and structure factors of AAV2 integrin complex density map.(D) An overlay of (B) and (C) using the Fit function in Chimera to match surface features between both maps with >95% correlation. A box around a threefold axis depicts density differences between both maps. (E) Difference map calculations between the (B) AAV2integrin reconstruction map subtracted by the (C) AAV2 density map. The positive density differences are shown in gray. Negative density differences were calculated by subtracting (B) from (C) and are shown in blue. (F) An overlay of the (C) AAV2 PDB density map with the positive and negative density differences from (E). The positive density differences are not visible on the surface of the AAV2 PDB density map. All views are down the twofold axis of symmetry. The cryo EM parameters are shown in Table 27.
133 Figure 414. Continued A B C D E F 1 1 4 136 3 fold 5 fold 2 fold
134 Figure 415. Three dimensional asymmetric reconstruction of AAV2 complexed with integrin using a box size of 196. (A) The reconstructed map visualized down a twofold ax is of symmetry (B) The same map as in (A) rotated 180 horizontally. Both views (compare A with B) show surface heterogeneity on the twofold threefold, and fivefold axes of symmetry 1 1 4 1 32 3 fold 5 fold 2 fold 1 1 4 1 32
135 CHAPTER 5 AAV2 AND THE IMMUNE RESPONSE Background The goal of this study was to evaluate the interaction of AAV2 with innate immune system effector molec defensins) and a known humoral antibody response (A20 IgG) by structural and biophysical techniques respectively Antimic robial peptides (AMPs) are polypeptides that are an evolutionarily conversed component of the innate immune response. One class of AMPs is the defens ins. The defensins function as host defense peptides and are active against bacteria (165) fungi (166) plants (167) and viruses (168) In h umans, defensins are defensins consist of a trip le sheet structure, have a molecular weight between 3 kDa and 6 kDa, defensins have been isolated from granules of neutrophils and were named human neutrophil peptides 14 (HNP1 4 defensins were identified from intestinal Peneth cells and were named HD5 and HD6. The typical mechanism of action defensins is to disrupt lipid membranes as seen in both bacteria (169) and enveloped viruses (170) defensins, specifically HD5 have been shown to neutralize nonenveloped viruses, despite the lack of a lipid membrane (171 176) Furthermore, a previous report showed that HNP1 and HNP2 act to neutralize AAV2 in cystic fibrosis patients (97) However, very little is known about how AAV2 interacts with HNP1and HNP2, particularly from a structural perspective. Moreover, the question remains as to whether HD5 binds to AAV2 like the aforementioned viruses.
136 defensins, the humoral immune response also plays a role. When vertebrates are infected by viruses, antibodies are produced and bind to various epitopes on the virus surface. Some antibodies are neutralizing in that they prevent the natural progression of infection by inhibiting virus receptor binding, endocytosis, and/or virus uncoating and genome release. Other antibodies are nonneutralizing and may increase virus uptake into the cell. The immune responses to the A AVs pose a significant barrier in the delivery of therapeutic genes since neutraliz ing antibodies have been found in up to 60% of the population (98) These antibodies specifically target the capsid, while immune r esponses to the transgene product are uncommon (177) A20 is a monoclonal antibody that neutralizes AAV2 by targeting fully assembled capsids and functions at a step post cel l and nuclear entry (65). It is the most widely studied monoclonal antibody against AAV2. The A20 binding footprints on the AAV2 capsi d surface have been identified have span across are large portion on the capsid surface. However, the stoichiometries and valency of the wt AAV2A20 binding interactions are unknown. Here the interactions of AAV2 with HNP1, 2, and HD5 were characterized by negative stain EM and transduction inhibition studies as well as cryo EM for AAV2 HD5. In addition, AAV2 A20 IgG and AAV 2 A20 Fab interactions were analyzed by SPR. Results and Discussion D efensins 1, 2, and 5 AAV2 transduction inhibition studies were conducted by preincubating scAAV2EGFP with and without HNP1 (Figure 51), HNP2 (Fi gur e 52 ), and HD5 (Figure 53) in HeLa cells. All three defensins inhibited scAAV2EGFP transduction, with HD5 and HNP1 inhibiting at the highest and the lowest levels, respectively (Table 5 1). These
137 defensin study in human papillomaviruses and the BKV pol yomavirus where HD5 inhibits infection to a greater degree than HNP1 (174, 175) Only one study to date has shown that HNP1 and HNP2 from bronchial secretions from cystic fibrosis patients inhibits AAV2 infection (97) However, this analysis only monitored a 1:1 mixture of HNP1 and HNP2, not the individual defensins. The results presented in Table 51 clearly show HNP2 inhibits AAV2 more than HNP1 since the same concentrations of the two defensins were analyzed. The step in the viral life cycle where the defensins are inhibiting AA V2 infection is currently unknown. In other nonenveloped virus studies, the defensins have varying inhibitory roles. For example, in papillomavirus, HD5treated viral particles bind to and enter host cells, but become retained within the cellular endos ome (174) However in BKV pol yomavirus, HD5 induces aggregation of viral particles that inhibits viral attachment to host cells by reducing the viral surface area and sequestering binding epitopes ( 175). defensininduced AAV2 aggregati on has not been seen (Figure 54 C D). Future experiments are thus warranted to elucidate the exact mechanisms of HNP1, HNP2, HD5 inhibition of AAV2. These are beyond the scope of the current study. Our goal was to visualize the interaction between these molecules and the AAV2 capsid. D efensins 1, 2, and 5 and Visualized by Negative S tain EM Negative stain transmission EM was performed on AAV2 complexed with HNP1, HNP2, and HD5 to visualize binding. Field of views from an AAV2 only control (Figure 5 4 A), HD5 only ( Figure 54 B), and AAV2 complexed with HNP1 (Figure 54 C), HNP2 (Figure 5 4 D), and HD5 (Figure 54 E) are shown. The AAV2 capsids with the
138 defensins are ~3.5 kDa, only subtle changes to the capsid surface are expected when complexed with AAV2. In a related cryo EM study wi th multiple Ad species complexed with HD5, the defensin bound all of the exposed major capsid proteins. Specifically, the potential sites for defensin binding are located between the penton base and fiber which are critical neutralization sites that prevent fiber dissociation and block subsequent steps (176) On the contrary, scanning EM results of Trypanosoma cruzi complexed with HNP1 show distinct membrane pore for mation (178) In this case, a lipid membrane served as the substrate for HNP1. Regarding AAV2 the capsid does not contain a lipid membrane and the AAV2 particles appe ar to be intact (Figure 54 C D). Therefore, a more detailed analysis of these interactions by cryo EM specifically AAV2 HD5, was undertaken to identify the HD5 binding site on the capsid surface. Analysis of AAV 2 HD5 Complexes by Cryo EM As mentioned above, AAV2 VLPs and HD5 were complexed in a 1 VP1:2.5 HD5 ratio (Table 26) and analyzed by cryoEM. The final map was solved at 9.7 resolution (Table 27) (Figure 5 5A) and structurally compared to an AAV2alone PDB density map filtered to the same resolution (Figure 55B).The twofold, threefold, and fivefold axes of symmetry from both maps were visually analyzed in Figure 55C E. Noticeably the AAV2 HD5 cryo EM map contains additional density spanning the twofold and fivefold axes of symmetry compared to the AAV2alone PDB density map. Thus, a difference map was calculated to further analyze these density differences in more detail (Figure 5 6). The positive (orange) and negative (blue) density differences (Figure 5 6C) were projected on the AAV2 HD5 cryo EM map. Minor density differences were observed on the surface of the capsid around the threefold axis of symmetry on the complex
139 compared to the AAV2 alone structure (Figure 56D) and more pronounced differences were observed on the inner surface of the capsid around the fivefold axis of symmetry (Figure 5 6F). Similarly, projection of the density differences onto the AAV2alone density map highlighted the additional positive density on the surface of the capsid a round the twofold and fivefold axes of symmetry (Figure 56E) and also on the inner surface of the capsid around the twofold and threefold axes of symmetry (Figure 56G). Both maps were then visualized in a cross section view (Figure 57A, B) and overlaid on each other (Figure 57C). The zoomedin views on the overlay show additional density on the outer portion of the fivefold channel and twofold axis (outer and inner), and a lack of density on the inner surface of the fivefold channel of the AAV2HD5 cryo EM map compared to the AAV2 alone PDB density map. In an effort to identify the VP regions at which differences occur on the AAV2 capsid surface involved in HD5 binding and/or conformational changes associated with HD5 binding, the AAV2 crystal structure (PDB ID #1LP3) was docked, by rigid body rotation and translation, into the AAV2HD5 cryo EM map and loop regions were fitted by real 5 8A F) had undergone a conformational c hange to reduce the diameter of the top of the channel at the fivefold axis compared to the AAV2 crystal structure (blue). The additional density at the twofold axis (Figure 58G) is the most likely due to HD5 binding on the capsid surface. Further analyses will be conducted to determine the proper orientation of the HD5 crystal structure in this density map. The VP amino acids in closest proximity to this density are shown in Figure (58G).
140 The data presented here has potential significance in AAV biology since the fivefold channel and twofold axis have been implicated in a number of viral functions. Mutagenesis of AAV2 capsid residues surrounding the fivefold channel resulted in significant (~3 to 9 times) reduction in genome packaging ( 33, 34). In additio n, opening of the fivefold channel has been proposed to prime the capsid for externalization of the VP1 unique (VP1u) region (34). On the contrary, another study implicated the twofold axis of symmetry as the most probable site for VP1u extrusion (188). Furthermore, the same tyrosine residues (700 and 704) implicated in that study were in close proximity to the additional HD5 density observed in this study. It was found that HD5 interactions with multiple Adenovirus serotypes inhibits protein VI externalization and prevents endosomal escape, but does not affect receptor binding or clathrinmediated endocytosis (62). Based on this report and the preliminary AAV2HD5 cryo EM analysis, HD5 may inhibit VP1u extrusion at either the fivefold or twofold symmetry ax es and thus prevent endosomal escape. This hypothesis is beyond the scope of this study and will be analyzed in future experiments. A20 IgG and Fab P urification A20 IgG was purified from hybridoma supernatants for SPR studies with AAV2 and also for the pro duction of Fabs. A20 IgG was purified on a protein G column and verified by W estern blo t analysis (Figure 59 A). By Western blot analysis, on ly elution fractions (Figure 59 A, lanes 514) were and contained proteins that hav e migrated to the appropriate sized band for heavy (50 kDa) and light (25 kDa) chains. The elutions were pooled, concentrated, and analyzed to be ~ 90% pure by SDS PAGE (Figure 5 9 B, lane 8) and densitometry The concentrated A20 IgG sample was cleaved with papain and the fractions were analyzed by SDS PAGE (Figure 5 9 B). The majority of the Fabs
141 are in the flow through (Fi gure 59 B, lanes 2 and 3), while the papain cleaved Fc portions are in both the flow through (Figure 59 B, lanes 2 and 3) and elution (Figure 59 B, lanes 5 and 6) frac tions. According to the kit manufacturer (Thermo Scientific), the Fab and Fc components should migrate to the 25 kDa and 2830 kDa marks, respectively. Based on this assumption, the lowest molecular weight band in fractions FT 1 and 2 is most likely the Fab portion. It does appear that the lowest band in FT1 is lower than 25 kDa, however, the gel s lopes in this region (Figure 59 B, compare lanes 2 and 3). The Fc portion binds to the Protein G and should only be in the elution fractions. Interestingly, both the FT and elution fractions contain Fc bands at ~30 kDa. The additional bands may be Fc that were incompletely digested or Fc components that failed to bind the Protein G. These observations are common according to the kit manufacturer. AAV2 and A20 IgG/ Fab Interactions A nalyzed by SPR The AAV2 and A20 IgG interactions were analyzed by SPR u sing a single cycle (Figure 5 10) and multi cycle (Figure 511) kinetics approach. Multiple concentrations of purified A20 IgG (Figure 59 B, lane 8) were manually flow ed over immobilized AAV2 in a dosedependent manner (data not shown). Based on these observations, a concentration series of A20 IgG ranging from 1.6 nM to 1000 nM (VP1:106 to 108 ratio) w ere flowed over AAV2 (Figure 510A). These concentrations fulfilled the three fundamental rules of singlecycle BIAcore kinetics: 1) Inject five increasing concentrations analyte (A20 IgG) over immobilized protein (AAV2), 2) include a concentration where saturation is reached, and 3) include at least one concentration bel o w the dissociation constant (Kd) value. The 1000 nM analyte concentration appears to have reached saturation based on the fast downwar d slope of the curve (Figure 5-
142 10A, 1000 nM injection) after the injection of sample. This result is further substantiated when directly comparing the 40 nM injection of sample that does not dissociate after injection (as seen by a horizontal line with no slope) to the 200 nM and 1000 nM injections. The experimental curves were fitted according to a 1:1 binding model (180) that result ed in a Kd value of 2 nM (Figure 510B). To recapitulate these observations from a slightly different approach, multi cycle kinetics were employed. The principles are essentially the same; however, this technique relies on a surface regeneration to remove analyte after each injection. In this case, a concentration series ranging from 0.5 nM to 1000 nM (VP1:105 to 108 ratio) A20 IgG were flowed over immobilized AAV2 (Figure 5 11A). It is apparent that saturation was reached since both the 500 nM and 1000 nM injections resulted in almost identical curves. The data was fitted according to a 1:1 model (F igure 511B) that resulted in a Kd value of 1 nM which is within the accepted standard deviation of this technique according to GE Healthcare. Both of these dat a sets suggest a very strong affinity between AAV2 and A20 IgG. Other viruses such as Adenovirus 2 (159) and Ebolavirus (181) have shown comparable low nM KD values with their respective antibodies. Monoclonal antibodies have the ability to bind ligands in a bivalent or monovalent fashion. The valency of AAV2 and A20 IgG interactions are unknown. In an effort to unravel this question, AAV2 and A20 Fabs were analyzed in a mutli cycle kineti cs approach (Figure 5 12) to determine Kd values for direct comparison to A20 IgG Kd values After initial tests with manual injections of A20 Fab over AAV2, it became apparent that much higher concentrations of A20 Fab were needed to have similar binding responses to the data shown in Figure 511 A concentration series of A20 Fab
143 ranging from 20 nM to 10,000 nM w ere flowed over AAV2 (Figure 512A). The data was fitted to a 1:1 model that resulted i n a Kd value of 2 M (Figure 512B). It is apparent AAV2 b i nding to the A20 IgG (Figure 512 A) is much tighter than the Fab, which suggests bivalent binding A recent cryo reconstruction of the AAV2A20 Fab complex suggests that the A20 binding footprints spans multiple peptides contributed by three different sy mmetry related subunits and at least 14 potential amino acids involved in binding (102) In addition, from a structural standpoint, the researchers believe only a monovalent binding model of A20 to AAV2 is possible due to the high levels of distortion of the antibody elbow angle that are needed for bivalent binding. On the contrary, the Kd values for the A20 IgG (12 nM) compared to A20 Fab (2 M) are 1000fold greater. At first glance, it is expected that the A20 IgG will provide a higher RU signal compared to the A20 Fab bas ed on mass (IgG 150 kDa compared to Fab 50 kDa). However, the mass differences only provide a theoretical threefold change, while the Kd values between both experiments are 1000fold different. This suggests that A20 Fabs dissociate quicker and bind l ess tightly than the A20 IgGs. Further analyses are clearly necessary to not only elucidate the valency of these interactions, but also determine the critical amino acids and/or epitopes (101) that contribute the most to this interaction. In addition to AAV2A20 antibody binding, this technique can also be utilized for screening other virus ant ibody interactions. Since AAV vectors are detrimentally affected by the humoral immune response, recent attention has focused on developing AAV escape mutants. The SPR technique developed here could provide an efficient and rapid method for screening AAV variant antibody interactions.
144 Summary In this study, HNP1, HNP2, and HD5 were shown to inhibit AAV2 as confirmed by transduction inhibition assays (Figures 51, 5 2, and 53). It was observed that HD5 inhibits scAAV2 EGFP transduction the most followe d by HNP2 and then HNP1 the least The amounts of defensins used for transduction inhibition wer e comp a rable to a study where HNP1 and HD5 were utilized for Adenovirus inhibition (62). These interactions were then visualized by negative stain EM where binding events were not easily detectable. In an effort to analyze these interactions in more detail HD5 was chosen to complex with AAV2 for a cryo EM analysis (Figure 5 5 5 6, 5 7, 5 8 ), which resulted in positive density changes around the fivefold axis of symmetry on the AAV2 capsid. HD5 molecules were fit ted into the positive density and the binding sites on the AAV2 capsid were identified at positions 327G and 328D at the peak of the DE loop. The proposed mechanism of interaction is HD5 binding inhibits conformational changes at the fivefold axis of symmetry and thus prevents VP1u externalization and endosomal escape. The hypothesis is preliminary and will be tested in future studies. SPR analyses were conducted on AAV2A20 IgG (KD =11 nM) and AAV2A20 Fab (Kd=2 M) molecules (Figures 510, 511, and 512 ). The binding appeared to occur in a bivalent nature, despite a report suggesting AAV2A20 interactions are monovalent (68). A possible explanation for this discrepancy is that the A20 IgGs do not i nteract with all potential binding sites in a bivalent manner. The idea is that the binding could be heterogeneous where some IgGs bind to the AAV2 capsid, but occlude other IgGs from binding due to steric hindrance. Another possibility is the A20 IgG is binding to two different AAV2 capsids monovalently. This could account for the increase in
145 binding affinity from the IgG compared to the Fabs. Clearly a cryo EM analysis of AAV2 with A20 IgGs is necessary to reconcile these different results .
146 Table 51. scAAV2 EGFP transduction inhibition by HNP1, HNP2, and HD5. defensin n ame Concentration Percent inhibition of AAV2 HNP1 100 g/ mL 99 HNP1 10 g/ mL 39 HNP1 1 g/ mL 33 HNP2 100 g/ mL 93 HNP2 10 g/ mL 88 HNP2 1 g/ mL 69 HD5 200 g/ mL 100 HD5 20 g/ mL 99 HD5 200 ng/ mL 90 defensin5 (HD5)
147 Figure 51 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 1 inhibition. (A) Cells were mock infected, infected with scAAV2 EGFP alone, and infected wi th scAAV2 mL mL and 1 mL of defensin 1 at a MOI of 1,250 viral genomes/cell. Transgene expression was detected by fluorescenc e microscopy 72 h post infection. A representative visual field view is shown from each triplicate experiment. Original magnification X100. (B) Quantitative analyses of AAV2 transduction efficiency in HeLa cells. Images from three visual fields were analyz ed quantitatively by ImageJ analysis software. Transgene expression was assessed as total area of green fluorescence (pixel2) per visual field (mean SD). ANOVA was used to compare test results with the control, and they were determined to be statisticall y significant. p <0.05 or ** p <0.01 vs scAAV2 EGFP alone. * *
148 Figure 52 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 2 inhibition. (A) Cells were mock infected, infected with scAAV2 EGFP alone, and infected wi th scAAV2 mL mL and 1 mL of defensin 1 at a MOI of 1,250 viral genomes/cell. Transgene expression was detected by fluorescence microscopy 72 h post infection. A representative visual field view is shown from each triplicate experiment. Original magnification X100. (B) Quantitative analyses of AAV2 transduction efficiency in HeLa cells. Images from three visual fields were analyzed quantitatively by ImageJ analysis software. Transgene expression was assessed as total area of green fluorescence (pixel2) per visual field (mean SD). ANOVA was used to compare test results with the control, and they were determined to be statistically significant. p <0.05 or ** p <0.01 vs scAAV2 EGFP alone. ** ** **
149 Figure 53 Comparative analyses of AAV2mediated transduction of HeLa cells with defensin 5 i nhibition. (A) Cells were mock infected, infected with scAAV2 EGFP alone, and infected wi th scAAV2 mL 20 mL and 200 ng / mL of defensin 5 at a MOI of 1,250 viral genomes/cell. Transgene expression was detected by fluorescence microscopy 72 h post infection. A representative visual field view is shown from each triplicate experiment. Original magnification X100. (B) Quantitative analyses of AAV2 transduction efficiency in HeLa cells. Images from three visual fields were analyzed quantitatively by ImageJ analysis software. Transgene expression was assessed as total area of green fluorescence (pixel2) per visual field (mean SD). ANOVA was used to compare test results with the control, and they were determined to be statistically significant. p <0.05 or ** p <0.01 vs scAAV2 EGFP alone. ** * *
150 Figure 54 defensin 1, 2, and 5 complexed with AAV2 and visualized by negative defensin 5. Samples were complexed in a 1:1 ratio and stained with uranly acetate. Images were taken at 50 K magnification. Scale bar is 100
151 Figure 55. Threedimensional icosahedral reconstruction of AAV2 complexed with defensin 5. (A) Radial coloring of the fin al density map of the AAV2defensin 5 reconstruction at 9.7 (B) Radial coloring of the AAV2alone density map at 9.7 Both maps were viewed along the (C, D) twofold, (E, F) threefold, and (G, H) fivefold axes of symmetry. The cryoEm parameters are shown in Table 27.
1 52 Figure 55. Continued. AAV2 HD5 (9.7 ) AAV2 alone (9.7 ) A B C D E F G H Twofold Threefold Fivefold 114 132
153 Figure 56 Difference map calculations between AAV2defensin 5 and AAV2alone density maps. (A) AAV2defensin 5 density map colored in orange. (B) AAV2 alone density map colored in blue. (C) Difference map calculations between maps (A) and (B). Positive and negative density is colored in orange and blue, respectively. (D) Negative density changes superimposed on the AAV2 defensin 5 density map. (E) Positive density changes superimposed on t he AAV2 alone density map. (F) Cross section view of the map in (D). (G) Cross section view of the map in (E).
154 Figure 56. Continued. F AAV2 HD5 (9.7 ) AAV2 alone (9.7 ) A B C D E G
155 Figure 57. Cross sections of AAV2 defensin 5 and AAV2alone density maps. (A) Cross section of AAV2 defensin 5 density map colored in orange. (B) Cross section of AAV2alone density map colored in blue. (C) An overaly of the maps shown in (A, B). The insets show zoomedin views of the fivefold and threefold symmetry axes. AAV2 HD5 (9.7 ) AAV2 alone (9.7 ) 2f 3f 5f 5fold axes 2fold axes A B C
156 Figure 58. AAV2 and defensin 5 binding and structural rearrangements. (A) Top view of the s tructural rearrangements of the DE loops in t he AAV2 defensin 5 PDB (orange) map superimposed on the AAV2alone PDB (blue; 1LP3). (B C ) The DE loops from both PDBs fit in the AAV2defensin 5 density map (orange) and AAV2alone density map (blue). (D F) Side views of (A C), respectively. (G) Twofold view of positive density on the AAV2alone density map. The amino acids in closest proximity to the positive density are labeled.
157 Figure 58. Continued. A B C D E F G
158 Figure 59 A20 IgG and Fab purification. (A) Western blot of A20 IgG purification using Protein G Lanes: (1) Molecular weight marker (M), (2) flow through (FT), (3) open, (4) wash (W), (5 14) elutions 110. The blot was probed with a mouse secondary antibody. (B) SDS PAGE of A20 Fab purification. Lanes: (1) Molecular weight marker, (24) flow through 1 3, (5 7) elutions 13, (8) A20 IgG contro l. B
159 Figure 5 10. AAV2 A20 IgG single cycle kinetics. (A) Experimental curves (shown in purple) for the following A20 IgG injections: 1) 1.6 nM, 2) 8 nM, 3) 40 nM, 4) 200 nM, and 5) 1000 nM. (B) Experimental (shown in purple) and fitted curves (shown in black).
160 Figure 511. AAV2 A20 IgG multi cycle kinetics. (A) Experimental curves for the following A20 IgG injections: 1) Buffer only 3 replicates, 2) 0.5 nM, 3) 2 nM, 4) 5 nM, 6) 10 nM, 7) 50 nM 2 replicates (50 nM, dark and li ght blue), 8) 100 nM (purple), 9) 500 nM (green), 10) 1000 nM (red). Only visible curves are denoted with a color. (B) Experimental and fitted (black) curves 1000 nM 1000 nM 1 000 nM
161 Figure 512. AAV2 A20 Fab multi cycle kinetics. (A) Experimental curves for the following A20 Fab injections: 1) Buffer only 3 replicates, 2) 20 nM (red), 3) 100 nM, 4) 250 nM (cyan), 5) 500 nM (blue), 5) 1000 nM 2 replicates (brown), 6) 2500 nM (purple), 7) 5000 nM (green), 8) 7500 nM (blue), 9) 10000 nM (red). (B) Experim ent al and fitted (black) curve 1000 nM
162 CHAPTER 6 SUMMARY AND FUTURE DIRECTIONS Summary The focus of this study was to structurally characterize AAV2 and its interactions with its receptors In addition, HNP1, HNP2, and HD5 were identified as binding partners and inhibitors of AAV2 during infection in HeLa cells, while the HD5 binding footprint was structurally identified on the capsid surface. Lastly, both AAV2 A20 IgG and AAV2A20 Fab interactions were characterized by SPR analyses. in tegrin Protein production in the typical quantities required for biophysical and structural biology purposes is often a bottleneck for structurefunction correlation efforts. With the advent of new DNA vectors that facilitate multi system expression, numerous proteins can now be expressed with the specific research interests of the lab in mind. For example, heterologous protein expression in E.coli yield higher quantities than insect and mammalian cells, however, at the expense of posttranslational modifications. The first part of the study involved designing an efficient integrin expression and purification system for the quantities necessary to perform structural studies with AAV2. The pQE TriSystem vec tor used in this study provides the flexibility of protein expression in all three systems, which the researcher can troubleshoot, optimize, and determine the most appropriate for the protein of interest. sub clon ed into this system for cotransfections in HEK293 cells. Due to sub cloning problems in the pQE vector was employed to develop recombinant baculoviruses that package these genes for co infect ions in SF9 insect cells. Both systems contained an 8X histidine tag for an
163 efficient onestep purification using nickel column chromatography. The amounts purified here ranged from 24 mgs per cell factory of HEK293 cell and 1 L of SF9 insect cells. The e fforts to produce the recombinant integrins were undertaken because currently available commercial resources provide limited amounts of samples, which are often lyophilized leading to decreased protein activity, and are reconstituted with additional ingredients, for example, Octyl b D glucopyranoside, TritonTM X 100, and sodium azide that are inhibitory to many experimental procedures. In addition, the expression systems presented here are cost effective compared to purchasing these integrins from commercial sources. For example, just 25 g of integrin from an unnamed vendor costs ~$270, and to purchase up to 4 mg of these p roteins costs ~ $43,000, compared to a ~$1,000 cost of producing the same amount of integrins in 1 L of insect cells in the lab. Lastly, there are only a few example studies in the literature that focus solely on expression of the extracellular domains for integrins for ligandreceptor interaction characterization, where the insoluble transmembrane domains are removed. T h e large scale protein production strategy presented here may be applied to other proteins with a specific interest on heterodimers. AAV2 In The transient nature of these virus receptor interactions is not unexpected since viru ses typically use these receptors as attachments points and/or and a means to enter the cell followed by receptor release. Table 11 shows the multiple receptors used by AAV2 and other AAV serotypes. It is no surprise that these viruses have a broad tissue tropism since many virus receptor interactions determine host range and tissue tropism (6). With respect to the AAVs, the surface loops (Figure 1 3A) have been shown to provide the variability that aids in determining receptor attachment. Mutagenic,
164 biochemical, and structural studies have shown that residues in these variable regions are crucial in virus receptor binding (182) The data presented in this study showed that t as shown by the SPR analyses. The integrin since a signal for the latter was not attainable. However, t he data was not fully reproducible for the AAV2 interactions. For example, the addition of EDTA typically inhibits integrin binding to Adenovirus (159) however the SPR studies conducted here showed EDTA inhibit AAV28C), but also have no effect on inhibition (Figure 49C). Other studies have shown that cell surface proteins typically need t o be embedded in their natural environment, within living cell or tissues, to exhibit their characteristic binding properties (183) This assumption supports the transduction inhibition studies presented in Figures 39 and 47 EGFP infection (F igures 39 and 47) This poses the question if there is an additional cellular factor that aids the AAV2integrin interaction? Additional experiments such as expressing both AAV2 and integrins in the cell, followed by imm unoprecipitation with A20 Mab (f or AAV2) or an anti his antibody for integrin may pull down other cellular factors that potential form a tertiary complex. The initial hypothesis was that the addition of heparin may be the third factor in this interaction, however, SPR (data not shown) and negative stain EM studies (data not shown) showed no evidence of this interaction aiding in integrin binding The cryo EM analysis of the icosahedrally average AAV2showed no changes in density when the AAV2 PDB cryo EM maps were used for
165 difference map calculations (Figures 413 and 414). These observations in combination with the SPR data where the AAV2dependent suggest integrin binding may occur in an asymmetric manner. This account has been seen with other viruses such as Canine Parvovirus where the transferrin receptor was shown to bind to one or a few of the 60 icosahedrally equivalent sites on the capsid surface (164) Another recent study, Coxsackievirus A9 (CVA9) and its cryo EM asymmetric reconstruction (81) Based on these reports and the data presented in t his study, an asymmetric reconstruction of the AAV215) was conducted to identify the binding footprint on the capsid surface. The observations supports that the capsid surface was not symmetrical and there was additional density, albeit little, at and around the twofold, threefold, and fivefold axes. However, these observations may be due to the lack of particles in the final reconstruction. This result suggest s that more data should be collected to perform an asymmetric reconstruction to obtain higher resolution Other methods such as cryo tomography of individual complexed viruses could be performed to determine integrin oc cupancy on the capsid surface. From a purification standpoint, an integrin masking effect have been potent ially detrimental effect to the structural studies by causing noisier reconstructions the refore the AAV2 integrin complex should be purified by gel filtration with and without protein cross linkers to removed excess integrin. Lastly, efforts should be made to use wt AAV2 with genome instead of AAV2 VLPs since the former is a more physiological interaction. AAV2 Interactions with defensins and A20 Mab and A20 Fab The AAVs have emerged as attractive gene delivery vectors due to the ability to establish long term transgene expression, the ability to establish long term transgene
166 expression, and the lack of pathogeni city from wt virus to name a few. However, the host immune responses to the viral capsids have become a serious obstacle in gene delivery. In this study HNP1 and HNP2 were characterized for their interactions with the AAV2 capsid (Figures 51 and 52), and for the first time, HD5 was identified as a binding partner to AAV2 (Figure 53, 5 5 5 6, 5 7, and 58 ) To date, this is the second stud defensins have been analyzed. The first study showed that both HNP1 and HNP2 inhibited recombinant AAV2 transduction in bronchial secretions from cystic fibrosis patients (63). However, the data presented here analyzed the individual HNP1 and HNP2 molecules, where the previous study (63) used defensins. Furthermore, the HD5 binding site on the capsid surface was identified by cryo EM to potentially bind at the twofold axis of symmetry In addition, conformational changes due to HD5 binding occurred at the fivefold channel. Based on these preliminary results, HD5 may potentially stop VP1u extrusion and therefore the phospholipase activity of the capsid which has been shown to be critical for endosom al escape (184) Th is hypothesis may parallel similar viral life cycle events seen in the human Adenoviruses (62) and also the Human Papillomaviruses (147) and their interactions with the defensins (62). Th e hyp o thesis could be tested by incubating the AAV2 capsids at a te mperature (65C) shown to externalize VP1u (184) defensins and analyze by Dot blot by probing with the A1 antibody that recognizes an epitope in VP1u. In addition, immunofluorescent confocal defensins fol lowed by inoculation with tagged AAV2 could be done. The cells would then be stained with an antibody to defensins are
167 inhibiting AAV2 transduction. These tests and the transduction inhibition could also be conducted for other AAV serotypes Other research has shown that the antibacterial properties of defensins are not unifor ml y conformation dependent. In some cases, incorrectly folded analogs are more potent antibacterial agents than the corr ectly folded defensin molecule (185) De sp ite these findings in bacteria, the inhibitory properties of defensins in human Adenoviruses are dependent upon defensin conformation (62). This could b e tested by AAV2 transduction inhibition with scAAV2EGFP and either heat defensin derivatives that have lost a native tertia ry structure. The amino acids at the twofold axis of symmetry that are proposed to interact with HD5 (Figure 58G) could be tested by sitedirected mutagenesis followed by AAV2GFP transduction inhibition experiments to test the mutant phenotypes. Lastly, the AAV2 and A20 IgG and A20 Fab interactions were analyzed by SPR (Figures 5 7, 5 8, and 59). The findings support that AAV2 binds to A20 IgG (11 nM) with greater affinity than the A20 Fabs (2 M). A recent cryo EM analysis of AAV2 complexed with A20 F abs showed that this interaction is monovalent. Based on these two studies, an additional cryo EM analysis of AAV2 complexed with A20 IgG should be conducted to elucidate the true binding interaction. Since the AAVs are prominently being utilized for gene delivery in vivo, understanding the host immune system responses to these serotypes is critical for designing better gene delivery vectors. Therefore, the SPR methodology developed in this study could be utilized for screening antibody interactions with AA V variants. For example, two recent studies showed that AAV neutralization by antibodies primarily occurs at the threefold axis of symmetry, the
168 same site for receptor binding (93 186 ) The goal would be to develop AAV variants that escape the immune response, while maintaining receptor binding. These variants could be coupled to BIAcore chips and antibodies and/or receptors could be flowed over to determine binding capabilities. In addition, other innate system effectors that have been implicated in anti viral interactions such as lactoferrin, liver expressed antimicrobial peptides, and secretory leukocyte protease inhibitor molecules all of which could be tested by BIAcore analyses with all of the AAV serotype.
169 LIST OF REFERENCES 1. Hueffer K Parrish CR 2003. Parvovirus host range, cell tropism and evolution. Curr. Opin Microbiol 6 :392 398. 2. Atchison RW Casto BC, Hammon W 1965. Adenovirus associated def ective virus particles. Science 149 :754 756. 3. Schmidt M Voutetakis A Afione S Zheng C Mandikian D Chiorini JA 2008. Adenoassociated virus type 12 (AAV12): a novel AAV serotype with sialic acid and heparan sulfate proteoglycanindependent transduction activity. J Virol 82:1399 1406. 4. Hoggan MD Blacklow NR Rowe WP 1966. Studies of small DNA viruses found in various adenovirus preparations: physical, biological, and immunological characteristics. Proc. Natl. A cad Sci U S A 55:1467 1474. 5. Parks WP Green M Pia M Melnick JL 1967. Physicochemical characterization of adenoassociated satellite virus type 4 and its nucleic acid. J Virol 1 :980 987. 6. Bantel Schaal U Hausen zur H 1984. Characterization of the DNA of a defective human parvovirus isolated from a genital site. Virology 134 :52 63. 7. Gao G Vandenberghe LH Alvira MR Lu Y Calcedo R Zhou X Wilson JM 2004. Clades of Adeno associated viruses are widely disseminate d in human tissues. J. Virol 78 :6381 6388. 8. Gao G P Alvira MR Wang L, Calcedo R Johnston J Wilson JM 2002. Novel adenoassociated viruses from rhesus monkeys as vectors for human gene therapy. Proc Natl Acad Sci U S A 99 :11854 11859. 9. Schmi dt M, Grot E Cervenka P Wainer S Buck C Chiorini JA 2006. Identification and characterization of novel adenoassociated virus isolates in ATCC virus stocks. J. Virol 80:5082 5085. 10. Mori S Wang L, Takeuchi T, Kanda T. 2004. Two novel adenoassociated viruses from cynomolgus monkey: pseudotyping characterization of capsid protein. Virology 330:375 383. 11. Wu Z, Asokan A Samulski RJ. 2006. Adenoassociated virus serotypes: vector toolkit for human gene therapy. Mol Ther 14 :316 327. 12. Gailit J Ruoslahti E 1988. Regulation of the fibronectin receptor affinity by divalent cations. J Biol Chem 263 :12927 12932.
170 13. Burger C Gorbatyuk OS Velardo MJ Peden CS Williams P Zolotukhin S Reier PJ, Mandel RJ Muzy czka N 2004. Recombinant AAV Viral Vectors Pseudotyped with Viral Capsids from Serotypes 1, 2, and 5 Display Differential Efficiency and Cell Tropism after Delivery to Different Regions of the Central Nervous System. Mol Ther 10 :16 16. 14. Davidson BL, Stein CS Heth JA Martins I Kotin RM Derksen TA Zabner J Ghodsi A Chiorini JA 2000. Recombinant adenoassociated virus type 2, 4, and 5 vectors: transduction of variant cell types and regions in the mammalian central nervous system. Proc Natl. Acad Sci U S A 97:3428 3432. 15. Kaludov N Brown KE Walters RW Zabner J Chiorini JA 2001. Adenoassociated virus serotype 4 (AAV4) and AAV5 both require sialic acid binding for hemagglutination and efficient transduction but differ in sialic acid linkage specificity. J. Virol 75 :6884 6893. 16. Rabinowitz JE Rolling F Li C Conrath H Xiao W Xiao X, Samulski RJ 2002. Cross packaging of a single adenoassociated virus (AAV) type 2 vector genome into multiple AAV serotypes enables transduction w ith broad specificity. J. Virol 76 :791 801. 17. Muralidhar S Becerra SP Rose JA 1994. Sitedirected mutagenesis of adenoassociated virus type 2 structural protein initiation codons: effects on regulation of synthesis a nd biological activity. J. Virol. 68:170176. 18. Sonntag F, Schmidt K Kleinschmidt JA 2010. A viral assembly factor promotes AAV2 capsid formation in the nucleolus. Proc Natl. Acad Sci USA 107:10220 10225. 19. McLaughlin SK Collis P Hermonat PL, Muzyczka N 1988. Adeno associated virus general transduction vectors: analysis of proviral structures. J Virol 62:1963 1973. 20. Hermonat PL, Muzyczka N 1984. Use of adenoassociated virus as a mammalian DNA cloning vector: transduction of neomycin resistance into mamm alian tissue culture cells. Proc Natl. Acad Sci U S A 81 :6466 6470. 21. Rossmann MG Morais MC Leiman PG Zhang W 2005. Combining X ray crystallography and electron microscopy. Structure 13:355362. 22. Carter PJ Samulski RJ 2000. Adenoassociated viral vectors as gene delivery vehicles. Int J Mol Med 6 :17 27. 23. Summerford C Samulski RJ. 1998. Membrane associated heparan sulfate proteoglycan is a receptor for adenoassociated virus type 2 virions. J Virol 72:1438 1445.
171 24. Kashiwakura Y Tama yose K Iwabuchi K Hirai Y Shimada T Matsumoto K Nakamura T Watanabe M Oshimi K Daida H 2005. Hepatocyte growth factor receptor is a coreceptor for adenoassociated virus type 2 infection. J Virol 79:609 614. 25. Qing K Mah C Hansen J, Zhou S Dwarki V Srivastava A 1999. Human fibroblast growth factor receptor 1 is a coreceptor for infection by adenoassociated virus 2. Nat. Med. 5 :71 77. 26. Asokan A Hamra JB Govindasamy L, Agbandje McKenna M Samulski RJ 2006. Adenoassociated virus t ype 2 contains an integrin alpha5beta1 binding domain essential for viral cell entry. J Virol 80 :8961 8969. 27. Akache B Grimm D Pandey K Yant SR Xu H 2006. The 37/67kilodalton laminin receptor is a receptor for adenoassociated virus serotypes 8, 2, 3, and 9. J. Virol. 19 :98319836. 28. Duan D Li Q Kao AW Yue Y, Pessin JE Engelhardt JF. 1999. Dynamin is required for recombinant adenoassociated virus type 2 infection. J Virol 73:10371 10376. 29. Kurzeder C Koppold B Sauer G Pabst S 2007. CD9 promotes adenoassociated virus type 2 infection of mammary carcinoma cells with low cell surface expression of heparan sulphate proteoglycans. Int. J. Mol. Med. 19 :325 323. 30. Nonnenmacher M Weber T 2011. Adenoassociated virus 2 infection requires endocytosis through the CLIC/GEEC pathway. Cell Host Microbe. 10 :563 576. 31. Bartlett JS Wilcher R Samulski RJ. 2000. Infectious entry pathway of adenoassociated virus and adenoassociated virus vectors. J Virol 74 :2777 2785. 32. Douar AM Poulard K Stockholm D Danos O 2001. Intracellular trafficking of adenoassociated virus vectors: routing to the late endosomal compartment and proteasome degradation. J Virol 75 :1824 1833. 33. Duan D Yue Y, Yan Z Yang J Engelhardt JF. 2000. En dosomal processing limits gene transfer to polarized airway epithelia by adenoassociated virus. J Clin Invest 105 :1573 1587. 34. Hansen J Qing K Srivastava A 2001. Adenoassociated virus type 2mediated gene transfer: altered endocytic processing enhances transduction efficiency in murine fibroblasts. J Virol 75:4080 4090.
172 35. Sanlioglu S Benson PK Yang J Atkinson EM Reynolds T, Engelhardt JF. 2000. Endocytosis and nuclear trafficking of adenoassociated virus type 2 are controlled by rac1 and phosphatidylinositol 3 kinase activation. J Virol 74 :9184 9196. 36. Bleker S Pawlita M Kleinschmidt JA 2006. Impact of capsid conformation and Repcapsid interactions on adenoassociated virus type 2 genome packaging. J Virol 80:810 820. 37. B leker S Sonntag F Kleinschmidt JA 2005. Mutational analysis of narrow pores at the fivefold symmetry axes of adenoassociated virus type 2 capsids reveals a dual role in genome packaging and activation of phospholipase A2 activity. J Virol 79 :2528 254 0. 38. Farr GA Tattersall P 2004. A conserved leucine that constricts the pore through the capsid fivefold cylinder plays a central role in parvoviral infection. Virology 323:14 14. 39. Farr GA Cotmore SF Tattersall P 2006. VP2 cleavage and the leuc ine ring at the base of the fivefold cylinder control pH dependent externalization of both the VP1 N terminus and the genome of minute virus of mice. J Virol 80:161 171. 40. Mani B Baltzer C Valle N Almendral JM Kempf C Ros C. 2006. Low pH dependent endosomal processing of the incoming parvovirus minute virus of mice virion leads to externalization of the VP1 N terminal sequence (N VP1), N VP2 cleavage, and uncoating of the full length genome. J Virol 80:1015 1024. 41. Giraud C Winocour E Berns KI 1994. Site specific integration by adenoassociated virus is directed by a cellular DNA sequence. Proc Natl. Acad Sci USA 21 :1003910043. 42. Wang XS Ponnazhagan S Srivastava A 1995. Rescue and Replication Signals of the Adenoassociated Virus 2 Genome. J Mol Biol. 250 :8 8. 43. Agbandje McKenna M, Chapman MS 2006. Correlating structure with function in the viral capsid. Parvoviruses. Edward Arnold, Ltd., New York 125139. 44. Agbandje McKenna M LlamasSaiz AL Wang F, Tatt ersall P Rossmann MG 1998. Functional implications of the structure of the murine parvovirus, minute virus of mice. Structure 6 :1369 1381. 45. Kontou M Govindasamy L Nam HJ Bryant N Llamas Saiz AL Foces Foces C Hernando E Rubio M P McKenna R Almendral JM Agbandje McKenna M 2005. Structural determinants of tissue tropism and in vivo pathogenicity for the parvovirus minute virus of mice. J Virol 79 :10931 10943.
173 46. Agbandje M McKenna R Rossmann MG Strassheim ML Parrish CR 1993. Structu re determination of feline panleukopenia virus empty particles. Proteins 16:155 171. 47. Kaufmann B Simpson AA Rossmann MG 2004. The str ucture of human parvovirus B19. Proc. Natl. Acad. Sci. USA 101 :11628 11633. 48. Tsao J Chapman MS Agbandje M Kel ler W Smith K, Wu H, Luo M, Smith TJ, Rossmann MG, Compans RW The ThreeDimensional Structure of Canine Parvovirus and Its Functional Implications. Science 251:14561464. 49. Simpson AA Hbert B Sullivan GM Parrish CR Zdori Z, Tijssen P Rossmann MG 2002. The structure of porcine parvovirus: comparison with related viruses. J Mol Biol. 315 :1189 1198. 50. Halder S Nam HJ Govindasamy L Vogel M Dinsart C Salom N McKenna R Agbandje McKenna M 2013. Structural characterization of H 1 Parvov irus: comparison of infectious virions to empty capsids. J Virol 87 :51285140. 51. McKenna R Olson NH Chipman PR Baker TS Booth TF, Christensen J Aasted B Fox JM Bloom ME Wolfinbarger JB Agbandje McKenna M 1999. Threedimensional structure of Aleutian mink disease parvovirus: implications for disease pathogenicity. J Virol 73:6882 6891. 52. Gurda BL, Parent KN Bladek H Sinkovits RS DiMattia MA Rence C Castro A McKenna R Olson N Brown K Baker TS Agbandje McKenna M 2010. Human bocavirus capsid structure: insights into the structural repertoire of the parvoviridae. J Virol 84 :5880 5889. 53. Xie Q Bu W Bhatia S Hare J, Somasundaram T, Azzi A Chapman MS 2002. The atomic structure of adenoassociated virus (AAV 2), a vector for human gene therapy. Proc Natl. Acad Sci U S A 99 :10405 10410. 54. Walters RW Agbandje McKenna M Bowman VD, Moninger TO, Olson NH, Seiler M, Chiorini JA, Baker TS, Zabner J 2004. Structure of adenoassociated virus serotype 5. J Virol. 78 :33613371. 55. Padron E Bowman V Kaludov N Govindasamy L, Levy H, Nick P, McKenna R, Muzyczka N, Chiorini JA, Baker TS, Agbandje McKenna M 2005. Structure of adenoassociated virus type 4. J Virol. 8 :50475058. 56. Govindasamy L, Padron E McKenna R Muzyczka N Kaludov N Chiorini JA Agbandje McKenna M 2006. Structurally mapping the diverse phenotype of adenoassociated virus serotype 4. J Virol 80 :11556 11570.
174 57. Nam HJ Lane MD Padron E Gurda B McKenna R Kohlbrenner E Aslanidi G, Byrn e B Muzyczka N Zolotukhin S Agbandje McKenna M 2007. Structure of adenoassociated virus serotype 8, a gene therapy vector. J Virol 81:12260 12271. 58. Lerch TF, Xie Q Ongley HM Hare J, Chapman MS 2009. Twinned crystals of adenoassociated virus serotype 3b prove suitable for structural studies. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 65:177 183. 59. Ng R, Govindasamy L, Gurda BL, McKenna R Kozyreva OG Samulski RJ Parent KN Baker TS Agbandje McKenna M 2010. Structural charact erization of the dual glycan binding adenoassociated virus serotype 6. J Virol 84:12945 12957. 60. DiMattia M, Govindasamy L, Levy HC Gurda Whitaker B Kalina A Kohlbrenner E Chiorini JA McKenna R Muzyczka N Zolotukhin S Agbandje McKenna M 2005. Production, purification, crystallization and preliminary X ray structural studies of adenoassociated virus serotype 5. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 61:917 921. 61. Miller EB Gurda Whitaker B Govindasamy L, McKenna R Zolotukhin S Muzyczka N Agbandje McKenna M 2006. Production, purification and preliminary X ray crystallographic studies of adenoassociated virus serotype 1. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 62 :1271 1274. 62. Quesada O Gurda B Govindasamy L, McKenna R Kohlbrenner E Aslanidi G Zolotukhin S Muzyczka N Agbandje McKenna M 2007. Production, purification and preliminary X ray crystallographic studies of adenoassociated virus serotype 7. Acta Crystallogr. Sect. F Struct. Biol. C ryst. Commun. 63:1073 1076. 63. Xie Q Ongley HM Hare J, Chapman MS 2008. Crystallization and preliminary X ray structural studies of adenoassociated virus serotype 6. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 64:1074 1078. 64. Mitchell M Nam HJ Carter A McCall A Rence C Bennett A Gurda B McKenna R Porter M Sakai Y Byrne BJ Muzyczka N Aslanidi G, Zolotukhin S Agbandje McKenna M 2009. Production, purification and preliminary X ray crystallographic studies of adenoassociated virus serotype 9. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 65 :715 718. 65. Halder S Ng R, Agbandje McKenna M 2012. Parvoviruses: structure and infection. Future Virology 7 :253278. 66. Chapman MS, Agbandje McKenna M 2006. Atomic structure of viral particles Parvoviruses. Edward Arnold Lt., New York, NY. 107123.
175 67. Handa A Muramatsu S Qiu J Mizukami H Brown KE 2000. Adeno associated virus (AAV) 3 based vectors transduce haematopoietic cells not susceptible to transduction with AAV 2 based vectors. J Gen Virol 81:2077 2084. 68. Halbert CL Allen JM Miller AD 2001. Adenoassociated virus type 6 (AAV6) vectors mediate efficient transduction of airway epithelial cells in mouse lungs compared to that of AAV2 vectors. J Virol 75:6615 6624. 69. Dechecchi MC Tamanini A Bonizzato A Cabrini G 2000. Heparan sulfate glycosaminoglycans are involved in adenovirus type 5 and 2host cell interactions. Virology 268:382 390. 70. Chen Y Maguire T, Hileman RE Fromm JR 1997. Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nature 8 :866 871. 71. Compton T Nowlin DM Cooper NR 1993. Initiation of human cytomegalovirus infection requires initial interaction with cell surface heparan sulfate. Virology 193:834 841. 72. Medina Kauwe LK 2003. Endocytosis of adenovirus and adenovirus capsid proteins. Adv. Drug Deliv. Rev. 55:1485 1496. 73. Opie SR Warrington KH Agbandje McKenna M Zolotukhin S Muzyczka N 2003. Identification of amino acid residues in the capsid proteins of adenoassociated virus type 2 that contribute to heparan sulfate proteoglycan binding. J Virol 77:6995 7006. 74. Levy HC Bowman VD Govindasamy L McKen na R Nash K Warrington K Chen W Muzyczka N Yan X Baker TS Agbandje McKenna M 2009. Heparin binding induces conformational changes in Adenoassociated virus serotype 2. J Struct Biol 165 :146 156. 75. O'Donnell J Taylor KA Chapman MS 2009. Adenoassociated virus 2 and its primary cellular receptor Cryo EM structure of a heparin complex. Virology 385:10 10. 76. Zdori Z, Szelei J, Lacoste MC Li Y Garipy S Raymond P Allaire M Nabi IR Tijssen P 2001. A viral phospholipase A2 is required for parvovirus infectivity. Dev. Cell 1 :291 302. 77. Girod A Wobus CE Zdori Z, Ried M Leike K Tijssen P Kleinschmidt JA Hal lek M 2002. The VP1 capsid protein of adenoassociated virus type 2 is carrying a phospholipase A2 domain required for virus infectivity. J Gen Virol 83:973 978.
176 78. Stahnke S Lux K Uhrig S Kreppel F Hsel M Coutelle O Ogris M Hallek M Bning H 2011. Intrinsic phospholipase A2 activity of adenoassociated virus is involved in endosomal escape of incoming particles. Virology 409:77 83. 79. Bottaro DP Rubin JS Faletto DL Chan AM Kmiecik TE Vande Woude GF, Aaronson SA 1991. Identification o f the hepatocyte growth factor receptor as the c met protooncogene product. Science 251 :802 804. 80. Suikkanen S Antila M Jaatinen A VihinenRanta M Vuento M 2003. Release of canine parvovirus from endocytic vesicles. Virology 316:267 280. 81. Shakeel S Seitsonen JJT Kajander T Laurinmaki P Hyypia T Susi P Butcher SJ 2013. Structural and Functional Analysis of Coxsackievirus A9 6 Binding and Uncoating. J Virol 87:3943 3951. 82. Mingozzi F, High KA 2007. Immune responses to AAV in clinical trials. Current gene therapy 7 :316 324. 83. Nathwani AC Gray JT Ng CY Zhou J Spence Y Waddington S N Tuddenham EG Kemball Cook G McIntosh J BoonSpijker M Mertens K Davidoff AM. 2006. Self complementary adenoassociated virus vectors containing a novel liver specific human factor IX expression cassette enable highly efficient transduction of murine and nonhuman primate liver. Blood 107:2653 2661. 84. Niemeyer GP Herzog RW Mou nt J Arruda VR Tillson DM Hathcock J, van Ginkel FW High KA Lothrop CD 2009. Long term correction of inhibitor prone hemophilia B dogs treated with liver directed AAV2mediated factor IX gene therapy. Blood 113 :797 806. 85. Gardner K Corry DB Lewi s DE Puellmann K 2006. Cytokine storm and an anti CD28 monoclonal antibody. N Engl J. Med 350 :351 356. 86. Rogers GL, Martino AT Aslanidi GV Jayandharan GR, Srivastava A, Herzog RW 2011. Innate immune responses to AAV vectors. Frontiers in Microbi ology 2 :1 10. 87. Akira S Uematsu S Takeuchi O 2006. Pathogen recognition and innate immunity. Cell. 124 :783 801. 88. Lu Y Song S 2009. Distinct immune responses to transgene products from rAAV1 and rAAV8 vectors. Proc Natl Acad Sci U S A 106 :17158 17162. 89. Mays LE Wilson JM 2011. The complex and evolving story of T cell activation to AAV vector encoded transgene products. Mol Ther 19 :16 27.
177 90. Calcedo R Vandenberghe LH Gao G Lin J Wilson JM 2009. Worldwide epidemiology of neutra lizing antibodies to adenoassociated viruses. J. Infect. Dis 199:381 390. 91. Manno CS Pierce GF Arruda VR Glader B Ragni M Rasko J Rasko J, Ozelo MC Hoots K Blatt P Konkle B Dake M Kaye R Razavi M Zajko A Zehnder J Rustagi PK Nak ai H Chew A Leonard D Wright JF Lessard RR, Sommer JM Tigges M Sabatino D Luk A Jiang H Mingozzi F Couto L Ertl HC High KA Kay MA 2006. Successful transduction of liver in hemophilia by AAV Factor IX and limitations imposed by the host immune response. Nat. Med. 12 :342 347. 92. Scallan CD Jiang H Liu T, Patarroyo White S Sommer JM Zhou S Couto LB Pierce GF 2006. Human immunoglobulin inhibits liver transduction by AAV vectors at low AAV2 neutralizing titers in SCID mice. Blood 107 :1810 1817. 93. Gurda BL, DiMattia MA Miller EB Bennett A McKenna R Weichert WS Nelson CD, Chen W J Muzyczka N Olson NH Sinkovits RS Chiorini JA Zolotutkhin S Kozyreva OG Samulski RJ Baker TS Parrish CR Agbandje McKenna M 2013. Capsid Antibodies to Different AdenoAssociated Virus Serotypes Bind Common Regions. J Virol. doi:10.1128/JVI.0062213 94. Asuri P Bartel MA Vazin T, Jang JH Wong TB Schaffer DV 2011. Directed evolution of adenoassociated virus for enhanced gene delivery and gene targeting in human pluripotent stem cells. Mol Ther. 20 :329 338. 95. White SH Wi mL ey WC Selsted ME 1995. Structure, function, and membrane integration of defensins. Curr Opin Struct Biol 5 :521 527. 96. Smith JG Silvestry M Lindert S Lu W Nemerow GR Stewart PL 2010. Insight into the mechanisms of adenovirus capsid disassembly from studies of defensin neutralization. PLoS Pathog 6 : e 1000959. 97. Virella Lowell I Poirier A Chesnut KA Brantly M Flotte TR 2000. Inhibition of recombinant adeno associated virus (rAAV) transduction by bronchial secretions from cystic fibrosis patients. Gene Ther 7 :1783 1789. 98. Boutin S Monteilhet V Veron P Leborgne C Benveniste O Montus MF Masurier C 2010. Prevalence of serum IgG and neutralizing factors against adenoassociated virus (AAV) types 1, 2, 5, 6, 8, and 9 in the healthy population: implications for gene therapy using AAV vectors. Audio, Transactions of the IRE Professional Group on 21:704 712.
178 99. Wobus CE Hgle Drr B Girod A Pe tersen G Hallek M Kleinschmidt JA 2000. Monoclonal antibodies against the adenoassociated virus type 2 (AAV 2) capsid: epitope mapping and identification of capsid domains involved in AAV 2 cell interaction and neutralization of AAV 2 infection. J Vir ol 74 :9281 9293. 100. Moskalenko M Chen L, van Roey M Donahue BA Snyder RO McArthur JG Patel SD 2000. Epitope mapping of human anti adenoassociated virus type 2 neutralizing antibodies: implications for gene therapy and virus structure. J Virol 74:1761 1766. 101. Lochrie MA Tatsuno GP Christie B McDonnell JW Zhou S Surosky R Pierce GF Colosi P 2006. Mutations on the external surfaces of adenoassociated virus type 2 capsids that affect transduction and neutralization. J Virol 80:821 834. 102. McCraw DM O'Donnell JK Taylor KA Stagg SM Chapman MS 2012. Structure of adenoassociated virus 2 in complex with neutralizing monoclonal antibody A20. Virology 431 :40 49. 103. Steckbeck JD Orlov I, Chow A Grieser H Miller K Bruno J Robinson JE Montelaro RC Cole KS 2005. Kinetic rates of antibody binding correlate with neutralization sensitivity of variant simian immunodeficiency virus strains. J Virol 79:12311 12320. 104. Daya S Berns KI 2008. Gene therapy using adenoassoci ated virus vectors. Clin Microbiol Rev 21:583 593. 105. Wu Z, Miller E Agbandje McKenna M Samulski RJ 2006. Alpha2,3 and alpha2,6 N linked sialic acids facilitate efficient binding and transduction by adenoassociated virus types 1 and 6. J Virol 80:9093 9103. 106. Kurzeder C Koppold B Sauer G Pabst S Kreienberg R Deissler H 2007. CD9 promotes adenoassociated virus type 2 infection of mammary carcinoma cells with low cell surface expression of heparan sulphate proteoglycans. Int J Mol Med 19:325 333. 107. Blackburn SD Steadman RA Johnson FB 2006. Attachment of adenoassociated virus type 3H to fibroblast growth factor receptor 1. Arch. Virol. 151:617 623. 108. Glushakova LG Lisankie MJ Eruslanov EB Ojano Dirain C Zolotukhin I Liu C Srivastava A Stacpoole PW 2009. AAV3 mediated transfer and expression of the pyruvate dehydrogenase E1 alpha subunit gene causes metabolic remodeling and apoptosis of human liver cancer cells. Mol. Genet. Metab. 98:289 299.
179 109. Walters RW Yi SM Keshavjee S Brown KE Welsh MJ, Chiorini JA Zabner J 2001. Binding of adenoassociated virus type 5 to 2,3linked sialic acid is required for gene transfer. J Biol Chem 276 :20610 20616. 110. Shen S Bryant KD Brown SM Randell SH Asokan A 2011. Terminal N linked galactose is the primary receptor for adenoassociated virus 9. J Biol. Chem 286 :13532 13540. 111. Wickham TJ Filardo EJ Cheresh DA 1994. Integrin alpha v beta 5 selectively promotes adenovirus mediated cell membrane permeabilization. J Cell. Biol. 1 :257 264. 112. Samulski RJ Summerford C Bartlett JS 1999. alphaVbeta5 integrin: a coreceptor for adenoassociated virus type 2 infection. Nat. Med. 5 :78 82. 113. Brown KE Anderson SM Young NS 1993. Erythrocyte P antigen: cellular receptor for B19 parvovirus. Science 262 :114 117. 114. Weigel Kelley KA Yoder MC Chen L, Srivastava A 2006. Role of integrin crossregulation in parvovirus B19 targeting. Hum Gene Ther 17 :909 920. 115. Drobni P Mistry N McMillan N Evander M 2003. Carboxy fluorescein diacetate, succinimidyl ester labeled papillomavirus virus like particles fluoresce after internalization and interact with heparan sulfate for binding and entry. Virology 310:163 172. 116. Evander M Frazer IH Pay ne E Qi YM Hengst K McMillan NA 1997. Identification of the alpha6 integrin as a candidate receptor for papillomaviruses. J. Virol 71 :2449 2456. 117. Marjomki VV Pietiinen VV Matilainen HH Upla PP Ivaska JJ Nissinen LL, Reunanen HH Huttunen PP, Hyypi TT, Heino JJ 2002. Internalization of echovirus 1 in caveolae. J Virol 76 :1856 1865. 118. Jordan M Schallhorn A Wurm FM 1996. Transfecting mammalian cells: optimization of critical parameters affecting calcium phosphate precipitate formation. Nucleic Acids Res. 24 :596 601. 119. Tiwari S Askari JA Humphries MJ Bulleid NJ 2011. Divalent cations regulate the folding and activation status of integrins during their intracellular trafficking. J. Cell. Sci. 124:1672 1680. 120. Kohlbrenner E Aslanidi G, Nash K Shklyaev S Campbell Thompson M Byrne BJ Snyder RO Muzyczka N Warrington KH Zolotukhin S 2005. Successful pr oduction of pseudotyped rAAV vectors using a modified baculovirus expression system. Mol Ther 12 :1217 1225.
180 121. Zolotukhin S Potter M, Zolotukhin I Sakai Y Loiler S Fraites TJ, Chiodo VA Phillipsberg T, Muzyczka N Hauswirth W W Flotte TR Byrne BJ Snyder RO 2002. Production and purification of serotype 1, 2, and 5 recombinant adenoassociated viral vectors. Methods 28 :158 167. 122. Laem mL i UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 :680 685. 123. Keller A Nesvizhskii AI Kolker E Aebersold R 2002. Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Anal. Chem. 74 :5383 5392. 124. Nesvizhskii AI Keller A Kolker E Ae bersold R 2003. A statistical model for identifying proteins by tandem mass spectrometry. Anal. Chem. 75 :4646 4658. 125. Tang G Peng L, Baldwin PR Mann DS Jiang W Rees I Ludtke SJ 2007. EMAN2: an extensible image processing suite for electron microscopy. J. Struct Biol. 157 :38 46. 126. Estrozi LF, Navaza J. 2008. Fast projection matching for cryoelectron microscopy image reconstruction. J Struct Biol. 162 :324 334. 127. Hynes RO 2002. Integrins: bidirectional, allosteric signaling machines. Cell 110:673 687. 128. Nagae M RS Mihara E Nogi T, Sugita Y Takagi J. 2012. Crystal structure of 1 :13 1 140. 129. Wegener KL Campbell ID 2008. Transmembrane and cytoplasmic domains in integrin activation and proteinprotein interactions. Mol Membr Biol. 25:376 387. 130. Millard M Odde S Neamati N 2011. Integrin targeted therapeutics. Theranostics 1 :154 188. 131. Cox D, Brennan M Moran N 2010. Integrins as therapeutic targets: lessons and opportunities. Nat Rev Drug Discov 9 :804 820. 132. Memmo LM McKeown Longo P 1998. The alphavbeta5 integrin functions as an endocytic receptor for vitronectin. J. Cell. Sci. 111 :425 433. 133. Mathias P Galleno M Nemerow GR 1998. Interactions of soluble recombinant integrin alphav beta5 with human adenoviruses. J Virol 72 :86 69 8675.
181 134. Lehmann M Battari El A Abadie B Martin JM, Marvaldi J 1996. Role of alphavbeta5 and alphavbeta6 integrin glycosylation in the adhesion of a colonic adenocarcinoma cell line (HT29D4). J Cell. Biochem 61:266 277. 135. Janik M A 2010. Cell migrationThe role of integrin glycosylation. Biochimic Biophysic Acta 6 :545 555. 136. Gu J Taniguchi N 2004. Regulation of integrin functions by N glycans. Glycoconj. J. 21 :9 15. 137. Veiga S Elias MC Gremski W Porcionatto MA da Sil va R Nader HB, Brentani R 1997. Post translational modifications of alpha5beta1 integrin by glycosaminoglycan chains. The alpha5beta1 integrin is a facultative proteoglycan. J Biol Chem 272 :12529 12535. 138. Pande P Mosleh TA Aust AE 2006. Role of alphavbeta5 integrin receptor in endocytosis of crocidolite and its effect on intracellular glutathione levels in human lung epithelial (A549) cells. Toxicol. Appl. Pharmacol. 210 :70 77. 139. Lin H Clegg DO 1998. Integrin alphavbeta5 par ticipates in the binding of photoreceptor rod outer segments during phagocytosis by cultured human retinal pigment epithelium. Invest. Ophthalmol. Vis. Sci. 39:1703 1712. 140. Buscarini M Quek ML Gilliam Hegarich S Kasahara N Bochner B 2007. Adenovir al receptor expression of normal bladder and transitional cell carcinoma of the bladder. Urol. Int. 78:160 166. 141. Glover JN Harrison SC 1995. Crystal structure of the heterodimeric bZIP transcription factor c Fos cJun bound to DNA. Nature 373 :257 26 1. 142. Engel J Kammerer RA 2000. What are oligomerization domains good for? Matrix Biol. 19:283 288. 143. Marsh M Helenius A 2006. Virus entry: open sesame. Cell 124 :729 740. 144. Schmidt CE Horwitz AF Lauffenburger DA Sheetz MP 1993. Integrincytoskeletal interactions in migrating fibroblasts are dynamic, asymmetric, and regulated. J Cell. Biol. 123 :977 991. 145. Nishizaka T Shi Q Sheetz MP 2000. Positiondependent linkages of fibronectinintegrin cytoskeleton. Proc Natl. Acad Sci U S A 97 :692 697. 146. Albelda SM Buck CA 1990. Integrins and other cell adhesion molecules. Faseb J 11:286880.
182 147. Schaffner F Ray A Dontenwill M Receptor, as a Pertinent Therapeutic Target in Solid Tum ors. Cancers 5 :27 47. 148. George EL GeorgesLabouesse EN Patel King RS Rayburn H Hynes RO 1993. Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development 119 :1079 1091. 149. Carlson TR Hu H, Braren R Kim YH Wang RA 2008. Cell autonomous requirement for beta1 integrin in endothelial cell adhesion, migration and survival during angiogenesis in mice. Development 135 :2193 2202. 150. Schornberg KL, Shoemaker CJ, Dube D Abshire MY Delos SE Bouton A H White JM 2009. Alpha5beta1integrin controls ebolavirus entry by regulating endosomal cathepsins. Proc Natl. Acad Sci USA 106 :8003 8008. 151. Sun WY Pitson SM Bonder CS 2010. Tumor necrosis factor induced neutrophil adhesion occurs via sphingosine kinase1 depen dent activation of endothelial alpha5beta1 i ntegrin. Am. J. Pathol. 177 :436 446. 152. Milliano MT Luxon BA 2003. Initial signaling of the fibronectin receptor alpha5beta1 integrin in hepatic stellate cells is independent of tyr osine phosphorylation. J. Hepatol. 39 :32 37. 153. Sano K I Maeda K Oki M Mada Y 2002. Enhancement of protein expression in insect cells by a lobster tropomyosin cDNA leader sequence. FEBS Lett. 532:143 146. 154. Alam MN Haque A Sreedhar M Pangbur n MK 2004. A novel vector for the expression of SCR domains in insect cells. J. Immunol. Methods 293 :107 113. 155. Zhang F, Dong L, Cai M Shen J Wang Y 2006. Heterologous expression of lipoproteinassociated phospholipase A2 in different expression sy stems. Protein Expr. Purif. 48 :300 306. 156. Belzhelarskaia SN 2011. Baculovirus expression systems for recombinant protein production in insect and mammalian cells Mol. Biol. 45:142 159. 157. Rowland TJ Miller LM Blaschke AJ, Doss EL Bonham AJ Hiki ta ST Johnson LV Clegg DO 2010. Roles of integrins in human induced pluripotent stem cell growth on Matrigel and vitronectin. Stem Cells Dev. 19:1231 1240. 158. Rajshankar D Downey GP McCulloch CA 2012. ILadhesion to degraded fibronectin. FASEB J. 26 :4429 4444. 159. Chiu CY Mathias P Nemerow GR Stewart PL 1999. Structure of adenovirus complexed with its internalization receptor, alphavbeta5 integrin. J Virol 73:6759 6768.
183 160. Adair BD Yeager M 2007. Electron microscopy of integrins. Meth Enzymol 426:337373. 161. Liu M Eiden MV 2011. The receptors for gibbon ape leukemia virus and amphotropic murine leukemia virus are not downregulated in productively infected cells. Retrovirology 8 :53 67. 162. Liu X Zhang Q Murata K Baker ML Sullivan MB Fu C Dougherty MT, Schmid MF, Osburne MS Chisholm SW Chiu W 2010. Structural changes in a marine podovirus associated with release of its genome into Prochlorococcus. Audio, Transactions of the IRE Professional Group on 17 :830 836. 163. Heikkil O Susi P Stanway G Hyypi T 2009. Integrin alphaVbeta6 is a highaffinity receptor for coxsackievirus A9. J Gen Virol 90 :197 204. 164. Hafenstein S Palermo LM Kostyuchenko VA Xiao C Morais MC Nelson CDS Bowman VD Battisti AJ Chipman PR Parrish CR Rossmann MG 2007. Asymmetric binding of transferrin receptor to parvovirus capsids. Proc Natl. Acad Sci U S A 104:6585 6589. 165. Yount NY Yeaman MR 2013. Peptide ant imicrobials: cell wall as a bacterial target. Ann. N. Y. Acad. Sci. 1277 :127 138. 166. Zhu S 2008. Discovery of six families of fungal defensinlike peptides provides insights into origin and evolution of the CSalphabeta defensins. Mol. Immunol. 45:828 8 38. 167. van der Weerden NL Anderson MA 2012. Plant defensins: common fold, multiple functions. Fungal Biology Reviews 26:121 131. 168. Ding J Chou YY Chang TL. 2009. Defensins in viral infections.J Innate Immun 1 :413 420. 169. Brogden KA 2005. Ant imicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nature Reviews Microbiology 3 :238250. 170. Klotman ME Chang TL. 2006. Defensins in innate antiviral immunity. Nature Reviews Immunology 6 :447 456. 171. Gropp R Frye M Wagner TO Bargon J 1999. Epithelial defensins impair adenoviral infection: implication for adenovirus mediated gene therapy. Hum Gene Ther 10 :957 964. 172. Bastian A Schfer H 2001. Human alphadefensin 1 (HNP 1) inhibits adenoviral infection in vitro. Regul. Pept. 101 :157 161.
184 173. Harvey SAK Romanowski EG Yates KA Gordon YJ 2005. Adenovirus directed ocular innate immunity: the role of conjunctival defensinlike chemoki nes (IP 10, I TAC) and phagocytic human defensinalpha. Invest. Ophthalmol. Vis. Sci. 46:3657 3665. 174. Buck CB, Day PM Thompson CD Lubkowski J Lu W Lowy DR Schiller JT 2006. Human alphadefensins block papillomavirus infection. Proc Natl. Acad S ci U S A 103 :1516 1521. 175. Dugan AS Maginnis MS Jordan JA Gasparovic ML Manley K Page R Williams G, Porter E O'Hara BA Atwood WJ 2008. Human alphadefensins inhibit BK virus infection by aggregating virions and blocking binding to host cells. J. Biol. Chem 283 :31125 31132. 176. Smith JG Nemerow GR 2008. Mechanism of adenovirus neutralization by Human alphadefensins. Cell Host Microbe 3 :11 19. 177. Chirmule N Xiao W Truneh A Schnell MA Hughes JV Zoltick P Wilson JM 2000. Humoral imm unity to adenoassociated virus type 2 vectors following administration to murine and nonhuman primate muscle. J Virol 74:2420 2425. 178. Madison MN Kleshchenko YY Nde PN, Simmons KJ Lima MF Villalta F 2007. Human defensin alpha 1 causes Trypanosoma cruzi membrane pore formation and induces DNA fragmentation, which leads to trypanosome destruction. Infect. Immun. 75 :4780 4791. 179. Szy k A Wu Z, Tucker K De D Yang Lu W Lubkowski J 2006. Crystal structures of human alphadefensins HNP4, HD5, and H D6. Protein Sci. 15:2749 2760. 180. O'Shannessy DJ Winzor DJ 1996. Interpretation of deviations from pseudofirst order kinetic behavior in the characterization of ligand binding by biosensor technology. Anal. Biochem. 236 :275 283. 181. Yu J S Liao H X Gerdon AE Huffman B Scearce RM McAdams M Alam SM Popernack PM Sullivan NJ Wright D Cliffel DE Nabel GJ Haynes BF 2006. Detection of Ebola virus envelope using monoclonal and polyclonal antibodies in ELISA, surface plasmon resonance and a quartz crystal microbalance immunosensor. J Virol Methods 137 :219 228. 182. Agbandje McKenna M Kleinschmidt J 2011. AAV Capsid Structure and Cell Interactions. Methods Mol Biol. 807 :47 92.
185 183. Frei AP Jeon O Y Kilcher S Moest H Henning LM Jost C Plckthun A Mercer J, Aebersold R Carreira EM Wollscheid B 2012. Direct identification of ligand receptor interactions on living cells and tissues. Nat. Biotechnol. 30:997 1001. 184. Popa Wagner R Porwal M Kann M Reuss M Weimer M Florin L Kleinschmidt JA 2012. Impact of VP1specific protein sequence motifs on adenoassociated virus type 2 intracellular trafficking and nuclear entry. J Virol 86:9163 9174. 185. Wei GG, Pazgier M de Leeuw E Rajabi M Li J Zou G Jung G Yuan WW Lu W YW Lehrer RIR Lu WW 2010. Trp 26 imparts functional versatility to human alphadefensin HNP1. J Biol Chem 285:16275 16285. 186. Gurda BL, Raupp C, PopWagner R, Naumer M, Olson NH, Ng R, McKenna R, Baker TS, Kleinschmidt JA, AgbandjeMcKenna M 2012. Mapping a neutralizing epitope onto the capsid of AdenoAssociated Virus Serotype 8. J. Virol. 86 :77397751. 187. Emsley P, Cowtan K 2004. Coot: model building tools for molecular graphics. Crystallographica Section D Biological Crystallography 60:21262132. 188. Horowitz ED, Finn MG, Asokan A 2012. Tyrosine cross linking reveals interfacial dynamics in adenoassociated viral capsids during infection. ACS Chem. Biol. 6 :10591066.
186 BIOGRAPHICAL SKETCH Lawrence Tartaglia grew up in Somerdale, New Jersey. He had a blessed childhood, growing up with loving parents, Larry and Alice. His one and only sibling, Kristen, was his playmate, whether she liked it or not. As a child he always found himself playing sports and avoiding homework at all costs. It wasnt until high school where he developed a true passion for the biological sciences. He graduated from Trition Regional High school in the spring of 1998 and enrolled at Rutgers U niversity that fall. During that time, he began his research career by studying the psychrophilic ice worm, Mesenchytraeus solifugus under the supervision of Dr. Daniel H. Shain. He used molecular and biochemical techniques to determine how ice worm microtubules remain functional at low physiological temperatures. He graduated with a B.A. in biology in 2003 and decided to parlay his undergraduate research into a Masters degree. In the final months of his M asters he went to Alaska and trekked across glaciers to retrieve these worms. In the spring 2008 he joined the lab of Dr. M avis AgbandjeMcKennas lab at the University of Florida. Specifically the Department of Biochemistry and Molecular Biology for his Ph.D. dissertation research project that focused on the characterization of the interaction of Adenoassociated virus seroty pe 2 (a promising gene delivery vector) and its integrin coreceptors. His studies involved the development of constructs expression systems. The interactions were charac terized using multiple techniques including surface plasmon resonance and cryoelectron microscopy. He received his Ph.D. in Biochemistry and Molecular Biology in the summer of 2013.