Phenotype of Mutants in pH-Sensitive Regions of the AAV Capsid and Evidence for a pH-Induced Capsid Protease


Material Information

Phenotype of Mutants in pH-Sensitive Regions of the AAV Capsid and Evidence for a pH-Induced Capsid Protease
Physical Description:
1 online resource (104 p.)
Salganik, Maxim
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Medical Sciences, Immunology and Microbiology (IDP)
Committee Chair:
Muzyczka, Nicholas
Committee Members:
Dunn, William A
Srivastava, Arun
Condit, Richard C
Agbandje-Mckenna, Mavis


Subjects / Keywords:
aav -- protease -- virus
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Medical Sciences thesis, Ph.D.
Electronic Thesis or Dissertation
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )


Adeno-Associated Virus (AAV) is a small, non-pathogenic parvovirus that hasgained popularity as a gene therapy vector in clinical trials and a gene transfer tool inresearch. A greater understanding of the basic biology of this virus could allow for theoptimization of this vector system, and thus may have far-reaching implications in bothresearch and clinical settings. The virus is known to enter the cell by receptor-mediatedendocytosis, and traffics to the nucleus where it uncoats and transcribes its genes. Therequirement for endosomal acidification to facilitate AAV infection has been known forsome time; however, the exact role of pH in AAV biology has not been clearlydetermined. Using crystallographic data that demonstrated transitions in two pH-sensitive regions of the capsid, we made individual alanine mutants and where able toidentify Y704, E562 and E564 as critical for AAV infection. We characterized theintracellular trafficking of these mutants and were able to show evidence of a novel rolefor the AAV capsid in the nucleus. We were also able to identify a previouslyundescribed pH-sensitive protease within the AAV capsid. Our work has shown that thepH-sensitive elements of the AAV capsid are critical for AAV infection and governpreviously unknown aspects of AAV biology.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (Ph.D.)--University of Florida, 2012.
Adviser: Muzyczka, Nicholas.
Electronic Access:
Statement of Responsibility:
by Maxim Salganik.

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
lcc - LD1780 2012
System ID:

This item is only available as the following downloads:

Full Text




2 2012 Maxim Salganik


3 This work is dedicated to my family and friends


4 ACKNOWLEDGMENTS I would like to express my deepest thanks to my mentor and committee chair, Dr. Nicholas Muzyczka for his guidance and support. I would like to thank my committee members, Drs. Mavis AgbandjeMckenna, Richard Condit, Arun Srivastava and Wil liam Dunn for their insight and help throughout my graduate school tenure All members of the Muzyczka lab, past and present, including Weijun Chen, Kei Wu, Shoudong Li, Craig Myers, Kevin Nash, Fikret Aydemir, Jackie Seekamp, Julia Vaizer, Thomas Ben St ephenson and Jasbir Singh have my thanks for making the lab a pleasure to work in. Special thanks goes out to Dr. Oleg Gorbat y uk for his guidance through my dissertation. I would also like to acknowledge the contributions of Dr. Balasubramanian Venkatakris hnan, a trusted colleague and a coauthor on our protease publication (96) Many people contributed to this work with technical advice and gifts of viral stocks and reagent s and I would like to thank David Knop (ASGT), Sergei Zolotukhin, Roland Herzog, Maurice Swanson, Arun Srivastava, Mavis AgbandjeMcKenna and Jude Samulski as well as the members of their labs for their contributions Special thanks to The University of Florida vector core has played a key role in making large stocks of some of the mutants described here as well as providing control vectors, and I would like to thank Mark Potter and the rest of the staff for their help. I would finally like to acknowledge and thank my parents, Mikhail and Marina Salganik for their unwavering support and my dear friends Dr. S usan Ellor, Dr. Ch ristopher DeFraia, Dr. Eric Yang Sheela Halbur, Dr. Lisa Stow Bailly and Dr. Marc Bailly for providing me with a foundation upon which to build my life and my work.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 LIST OF A BREVIATIONS ............................................................................................. 10 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRO DUCTION .................................................................................................... 12 AdenoAssociated Virus .......................................................................................... 12 AAV Bi ology ..................................................................................................... 13 Capsid .............................................................................................................. 14 AAV Receptors, Entry and Trafficking .............................................................. 18 AAV and the Nucleus ....................................................................................... 25 Replication and Assembly ................................................................................ 29 Spec ific Aims .......................................................................................................... 30 Mutational and Trafficking Analysis of Mut22 and pH Responsive Amino Acids of the AAV2 Capsid ............................................................................. 31 Identification and Characterization of the AAV Capsid Protease ...................... 31 2 MATERIALS AND METHODS ................................................................................ 32 PCR Mutagenesis and Plasmid DNA Purification ................................................... 32 Cell Culture ............................................................................................................. 32 Virus Production ..................................................................................................... 32 AAV2 pH Mutants ............................................................................................. 32 AAV1, 2 and 9 Protease Mutants ..................................................................... 33 Wild type Serotype 110 Samples .................................................................... 33 Virus Titering ........................................................................................................... 34 Particle to Infectivity Ratios ..................................................................................... 34 Western Blotting and Dot Blotting ........................................................................... 35 Immunocytochemistry and Confocal Microscopy .................................................... 35 Subcellular Fractionation ........................................................................................ 36 Viral DNA Extraction and Real Time PCR .............................................................. 37 Uncoating Assay ..................................................................................................... 37 Capsid Autolysis Assay ........................................................................................... 38 Mass Spectrometry ................................................................................................. 38 Edman Degradation Sequencing ............................................................................ 39


6 3 AAV2 CAPSID MUTAGENESIS ............................................................................. 41 Results .................................................................................................................... 42 pH Quartet Mutant Phenotypes ........................................................................ 42 Entry and Nuclear Trafficking of pH Quartet Mutants ....................................... 42 Uncoating and Second Strand Synthesis ......................................................... 44 Characterization of Mut22 ................................................................................ 47 DNA Group ....................................................................................................... 48 Discussion .............................................................................................................. 48 pH Quartet ........................................................................................................ 48 DNA Group ....................................................................................................... 51 Summary ................................................................................................................ 53 4 PROTEASE ACTIVITY OF THE AAV CAPSID ....................................................... 66 Results .................................................................................................................... 66 Protease Activity Within the AAV Capsid .......................................................... 66 Determination of Autolytic Cleavage Sites ........................................................ 69 Identification of putative active site amino acids. ....................................... 70 The VP1 cleavage pattern is different from that of VP3. ............................ 71 Discussion and Conclusion ..................................................................................... 72 5 CONCLUSION AND FUTURE DIRECTIONS ......................................................... 87 Conclusion .............................................................................................................. 87 Future Directions .................................................................................................... 89 Y704A and the Nucleus .................................................................................... 89 AAV Protease ................................................................................................... 89 LIST OF REFERENCES ............................................................................................... 91 BIOGRAPHIC AL SKETCH .......................................................................................... 104


7 LIST OF TABLES Table page 3 1 Alignment of amino acid residues that contribute to the pH quartet and surrounding acidic clusters in AAV1 to 10 .......................................................... 55 4 1 Edman N Terminal Sequences .............................................................................. 86


8 LIST OF FIGURES Figure page 3 1 Location and Structure of the pH quartet and DNA group. ................................. 54 3 2 Mutant virus production. ..................................................................................... 55 3 3 Particle to infectivity ratios of pH quartet mutants. .............................................. 56 3 4 A20 confocal microscopy of pH quartet mutants.. .............................................. 57 3 5 Confirming nuclear and cytoplasmic fraction purity ............................................ 57 3 6 Cellular entry of WT and pH quartet mutant genomes.. ...................................... 58 3 7 Subcellular distribution of pH quartet genomes.. ................................................ 59 3 8 Percentage of uncoated nuclear genomes.. ....................................................... 60 3 9 Particle to infectivity ratios of self complimentary pH quartet and mut40. ........... 61 3 10 Wild type capsids do not compliment Y704A. ..................................................... 62 3 11 Mut22 particle to infectivity ratio. ........................................................................ 63 3 12 Mut22 A20 confocal microscopy. ........................................................................ 64 3 13 Subcellular distribution of wildtype and mut22 genomes. .................................. 65 3 14 DNA group particle to infectivity ratio.. ................................................................ 65 4 1 Position of conserved acidic cluster .................................................................... 77 4 2 AAV undergoes autolytic cleavage at pH 5.5. .................................................... 78 4 3 Mass spectrometry map of cleavage product peptides.. ..................................... 79 4 4 All of the AAV serotypes undergo autolytic cleavage. ........................................ 79 4 5 External protease activity of the AAV capsid. .................................................... 80 4 6 Protease inhibitors have no effect on autolytic pH5 cleavage.. .......................... 81 4 7 Autocatalytic activity is preserved in the mut40 acid cluster mutants. ................ 81 4 8 AAV9 specific cleavage site mutants ................................................................. 82 4 9 AAV1 specific cleavage site mutants. ................................................................ 83


9 4 10 VP1 specific cleavage. ...................................................................................... 84 4 11 Map of identified cleavage sites. ......................................................................... 85


10 LIST OF ABREVIATIONS AAV AdenoAssociated Virus Ad5 Adenovirus 5 HSPG Heparan Sulfate Proteoglycan NLS Nuclear Localization Signal NPS Nuclear Pore Complex ORF Open Reading Frame PI Post Infection SC Self Complimentary SS Single Stranded VP Viral Protein WT Wild Type


11 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 PHENOTYPE OF MUTANTS IN PHSENSITIVE REGIONS OF THE AAV CAPSID AND EVIDENCE FOR A PH INDUCED CAPSID PROTE ASE By Max Salganik December 2012 Chair: Nicholas Muzyczka Major: Medical Sciences AdenoAssociate d Virus (AAV) is a small, non pathogenic parvovirus that has gained popularity as a gene therapy vector in clinical trials and a gene transfer tool in research. A greater understanding of the basic biology of this virus could allow for the optimization of this vector system, and thus may have far reaching implications in both research and clinical settings. The virus is know n to enter the cell by receptor mediated endocytosis, and traffics to the nucleus where it uncoats and transcribes its genes. The requirement for endosomal acidification to facilitate AAV infection has been know n for some time; however, the exact role of pH in AAV biology has not been clearly determined. Using cry sta llographic data that demonstrated transition s in two pH sensitive regions of the capsid, we made individual alanine mutants and where able to identify Y704, E562 and E564 as critical for AAV infection. We characterized the intracellular trafficking of these mutants and were able to show evidence of a novel role for the AAV capsid in the nucleus. We were also able to identify a previously undescribed pH sensitive protease within the AAV capsid. Our work has shown that the pH sensitive elements of the AAV capsid are critical for AAV infection and govern previously unknown aspects of AAV biology.


12 CHAPTER 1 INTRO DUCTION Adeno Associated V irus The members of the Parvoviridae family infect a diverse range of hosts and share a common characteristic of packaging their singlestranded DNA genome into a small, nonenv eloped, icosahedral capsid. While this makes viruses of this family remarkably resilient in harsh environments it presents a unique set of challenges during infection compared to that of enveloped viruses. Unable to rely on the mechanism of membrane fusion, Parvoviruses must utilize receptor mediated endocytosis to gain access to th e complex endosomal network, then the cytoplasm and ultimately the nucleus. Throughout this journey each Parvovirus must avoid degradation and entrapment in deadend cellular compartments while priming its own tough capsid for genome release and second str an d synthesis in the nucleus AAV, a member of the Dependovirus genus must further orchestrate all of these events around a coinfection with an unrelated, autonomous helper virus; a role that can be filled by its n ame sake Adenovirus as well as human he rpes virus, cytomegalovirus and human papillomavirus (5, 17, 78, 85) The steps in intracellular trafficking, uncoating and early replication of AAV are poorly understood and yet are of critical significance since the virus is now used in numerous clinical trials as a gene delivery vector. While AAV has shown incredibly broad tropism across numerous tissue types and has proven to be highly useful as a research tool (reviewed in (4)) a deeper and more complete understanding of i t s basic biology is critical to realizing its fu ll potential in the clinic


13 AAV Biology Twelve distinct human and nonhuman AAV serotypes have thus far been identified, with hundreds of recombinants identified by PCR studies in human and nonhuman samples (35, 36, 76, 103) These serotypes can further be organized into genetic clades A F, with AAV4 and AAV5 belonging to their own unique groups. There is a 6099% sequence identity among the members of the clades, with AAV4 and 5 being the most divergent from all others. In contrast AAV1 and 6 both belong to clade A due to a high sequence homology and antigenic cross reactivity (36) While many of the AAV serotypes have been tested for gene therapy applications, the efficient infection of AAV2 in tissue culture has made it the serotype of choice for study ing the basic biology of AAV infection. For this reason most mutagenesis studies of AAV to date have focus ed on AAV2 (14, 40, 41, 71, 108, 122) The ss DNA genome of AAV is 4.7kb in length and contains two large and one small open readi ng frame [ORF] The coding sequence is flanked by two 145bp inverted terminal repeats [ITRs] which are the sole cis elements required for replication, packaging and integration (75, 97, 99) The first o f the two major ORFs encodes four overlapping replication proteins, Rep 78, Rep68, Rep52 and Rep40 The Rep68 and Rep78 proteins contain nicking activity as well as ATP dependent helicase activ ity in the 3 5 direction. The large Rep proteins (78, 68) are produced by alternative splicing of the p5 promoter transcript, while the two smaller Rep proteins are produced from the p19 transcript. During latent infection, Rep78/ 68 act to suppress AAV gene expression and maintain th e viral genome via sitespecific integration into the host genome. These same proteins serve to activate AAV DNA replication and gene expression in the


14 presence of a helper virus. Conversely, the smaller Rep proteins appear to not be crucial for DNA replic ation, but mediate genome packaging (22, 61) T he second major ORF encodes the three capsid proteins [VP1 (87kDa), VP2 (73kDa), VP3 (63kDa)]. They are generated by alternative splicing of a single mRNA transcribed from the p40 promoter. The full mRNA produces VP1, however the major splice variant remov es the VP1 start codon, leaving only one other AUG, which is at the start of the VP3 sequence. An alternate start site ( ACG ) upstream of the VP3 start codon has the proper Kozak context and thus results in a low level of VP2 expression. All three proteins share a common C terminus ( 532 amino acid VP3 sequence in AAV2 ) with VP2 and 1 having progressively more unique N terminal sequence. Recently it was discovered that a short ORF, that overlaps the CAP ORF produces a small accessory protein that is necessar y for AAV assembly. This protein has been appropriately named assembly activating Protein (AAP) (108) Capsid The AAV genome is packaged into a 26nm icosahedral capsid that has an approximate mass of 4 MDa. The capsid is comprised of 60 subunits assembled as an icosahedron with a T=1 symmetry, with the VPs 1 (87kDa), 2 (73kDa) and 3 (62kDa) present in a 1:1:810 ratio as determined by denaturing gel electrophoresis (60) The co mmon C terminal region encompassed by VP3 is responsible for forming the bulk of the AAV capsid structure and for carrying out most major functions of the capsid. The minor VP1 protein is necessary for productive infection and contains within its unique N terminal region a phos pholipaseA2 (PLA2) domain (38) as well as several basic regions that have been shown to function as nuclear localization signals (NLS) (41, 107) Recent findings indicate that the unique N terminal region also contains several


15 other trafficking signals including an SH2binding motif shown to be critical for nuclear trafficking and a PDZ binding motif with implications for nuclear entry (90) While the unique N terminal sequences of both minor proteins are localized to the interior of the capsid at room temperature, it has been show n that these domains become exposed during infection in capsids that are otherwise intact (107) This feature in combination with the fact that VP2 is nonessential has been used to redesign VP2 as a vehicle for introducing large ligands into the AAV capsids, which are incorporated into the capsid along with VP2 and displ ayed on the capsid surface (118) The capsid has several axes of symmetry, which correspond to prominent topological features. The threefold axis of symmetry is characterized by a depression which is surrounded by prominent spikes. These form the most distal points from the center of the capsid. These are separated by valleys and contain amino acids which play a prominent role in receptor binding (59, 116) and host imm une response (42, 77) In AAV2, this region contains R484, R487, K527, K532, R585, and R588, which are responsible for binding heparan sulfate proteoglycan (HSPG) AAV2s primary attachment receptor. In addition to receptor binding mutants, the AAV2 charge cluster alanine mutant m ut22 (268272 NDNHY NANAY) is also surface exposed and located along the threefold axis of symmetry. This mutant is severely defective for infection despite being able to form intact c apsids and binding HSPG In close proximity to m ut22 on the outer surface is R432 which when mutated to alanine has a null packaging phenotype (122) While R432 is far from the five fold pore, which is the site of genome packaging, it is believed that it interrupts normal interaction between the capsid and Rep, thus preventing packaging On the interior of the capsid, the three fold region


16 appears to mediate capsid interactions with genomeassociated DNA, as well as nongenome associated nucleotides (37, 39, 68, 80, 83) While capsid assembly does not require the presence of the AAV genome or REP proteins, the resulting empty capsids have been shown to still c ontain short DNA segments A recent study examining the effects of pH on crystals of AAV8 found that one of the two regions that underwent shifts in response to lower pH appeared to be involved in capsidDNA interactions (79) This region was located on the inner surface at the three fold and at neutral pH showed a density corresponding to a dAMP, similar to whats been seen in other studies. As the pH was dropped, H631 shifted and the dAMP density was no longer observed. This region was called the DNA group and involved residues P422, P632, H631 and F629 (AAV8). Further examination of this region and its function in AAV2 will be discussed in the results section. The ordered nucleotidecapsid interactions appear to extend from the threefol d into the two fold region (37) that is defined by a depression that separates the threefold spikes on the cap sids outer surface. The residues surrounding this region have been implicated in AAV infectivity and are of particular importance to the work described here. Among the previously identified AAV2 mutants, Mut40 (561565 DEEEI AAAAI) and Mu t37 (527532 KDDEE K AAAAA) (122) were both able to produce intact, full capsids that were severely defective for infectivity. Both of these mutants are located along the sides of the t hreefold spikes, surrounding the twofold. Located nearby mut37 encompasses two out of five residues (K527, K532) responsible for heparin binding, and is in direct contact with residues from m ut40. Mut40 is in turn part of the second region identified by Nam et al, as being sensitive to pH. This second region, dubbed the


17 pH q uartet, is comprised of E566, Y707, R372 and H529 in the AAV8 sequence, which come together from three different VP monom ers. At neutral pH, Y707 and E566 appear to interact through hydrogen bonding. This interaction is broken as both residues rotate away from one another in response to acidic pH. This transition occurs fairly abruptly with approximately 50% of the capsids having their pH Quartets switched at pH6 and all being complet ely switched by pH 5.5 During this transition, the side chain of Y707 also transitions from being buried to surface exposed. Se veral of the residues in the pH q uartet have previously been implic ated in AAV2 infection. E566 (AAV8) corresponds to AAV2s E563, which is one of the residues in mut40. Likewise, Y707 (AAV8) corresponds to Y704 in AAV2 and is one of the surface tyrosines found by Zhong et al (130) to produce a mild increase in infectivity when mutated to a phenylalanine. A n additional two surface tyrosine residues from the same study (Y700, Y730), which also produced a similar phenotype, are located in close proximity to the pH quartet and come together to form part of the floor in the twofold depression. Deeper in the capsid and in close proximity to the mutants discussed above, mut47 is also defective in infectivity while forming intact capsids (122) Interestingly this mutant is also aberrant for binding the A20 monoclonal antibody which is specific for intact capsids even though the residues in question are located f ar from the antibody binding site (120) on the inner surface of the capsid, at the twofold axis of symmetry. In addition to all of the mutational data, cryoelectron microscopy has suggested that the N ter mini of the minor capsid proteins reside underneath the twofold axis in intact capsids, prior to externalization during infection (37) Clearly this region plays a crucial in AAV infection and will be further discussed in upcoming chapters.


18 The five fold axis of symmetry is characterized by a rai sed pore surrounded by a valley. The pore is the only channel between the interior and exterior environments of the intact capsid. The narrowest point is located near the inner surface of the capsid and is ~1.2nm across (14) which is large enough to accommodate DNA and peptides. The cylindrical channel of the pore is comprised of residues 322338 in AAV2 and extends above the exterior surface of the capsid, forming a raised rim around its opening. The residues at either end of this structure are hi ghly conserved, although the re is variability in t he intermediate sequence. At it s interior end, the pore meets a funnel like structure, formed by the highly conserved residues 217223 (AAV2). The primary function of the pore is believed to be for packaging the viral genome into preformed capsids, however there is evidence that the pore may also be used for externaliz ing the VP1 and 2 N termini. Mutants such as L336A, N335A and V221C in the pore have been shown to block heat mediated externalization of the VP1 and 2 in vitro as well as resulted in decreased infectivity, suggesting that this externalization is also hindered during infection (14) AAV Receptors, Entry and Trafficking Typical viral infection begins with receptor binding on the cell surface. In the case of AAV this requires the binding of both a primary glycan receptor and a secondary protein receptor. Different AAV serotypes bind different glycan receptor s. In the case of AAV2, HSPG is know n to be the primary receptor (112) HSPG is also utilized by AAV3 (112) and is at least also bound by AAV6 (123) though in that case it is not clear linked sialic acids (124) Sialic acids of different linkages are also utilized by AAVs 1, 4 and 5 (54, 124) while AAV9 has been recently shown to bind terminal galactose residues (104,


19 105) This variety of glycan receptors belies what may be an even greater number of secondary receptors. In the case of AAV2, fibroblast growth factor receptor (FGFR), 47/67kDa Laminin receptor, hepatocyte growth factor receptor integrins have been shown to act as receptors (1, 3, 55 92, 111) Although other serotypes havent been as extensively characterized, it is clear that there are both receptors that unique to individual serotypes such as platelet derived growth f actor receptor (PDGFR) and AAV5 (26) and some that are used by multiple serotypes such as 37/67kDa laminin receptor and AAVs 2, 3, 8 and 9 (1). Out of the two kinds of receptors bound by AAV during infection, only t he secondary protein receptor triggers endocytosis. It has been thought that binding to the primary glycan receptor anchored the capsid on the cell surface, which then allow ed secondary receptor binding to occur. Whatever the exact nature of the mechanism, AAV receptor binding has been believed to trigger clathrinmediated, dynamindependent endocytosis (9, 30) This original model then saw AAV, now residing within clathrin coated vesicles, traffic to early endosomes. From here, there have been varying reports on the exact path AAV would take through the endosomal system, out into the cytoplasm and ultimately into the nucleus (27 29, 56, 101, 126) Many of these reports are for AAV2 but seemed to be specific to cell type or mult iplicity of infection (MOI) used. To those that tried to tackle the question of what specific pathway AAV took through the cell, after endocytosis no clear answer emerged. In one case, infections with a higher MOI of AAV seemed to direct the virus into rec ycling endosomes, and resulted in a more efficient infection (28) However the same study found, as had others that at least a portion of the virus still trafficked through to the late endosome. Late


20 endosomes typically mature into lysosomes, but also exchange material with the trans Golgi (16, 47) It is therefor e, not surprising that AAV has also been colocalized with both of these compartments (8, 51) though in both cases it is not clear whether this was representative of virus that would later go on to the nucleus or simply a deadend. It is believed that at least the subset of AAV that ultimately gains access t o the nucleus escapes from the early or late endosome and rapidly traffics to the nucleus. Escape from the endosomal system is believed to be mediated by the PLA2 domain within the VP1unique region (38, 109) While PLAmediated endosome escape is believed to be a conserved feature among parvoviruses, findings with Canine P arvovirus (CPV) suggest that this process may be more complicated than simply punching holes in vesicle membranes. CPV infection supplemented with different sized fluorescent dextrans revealed that while CPV PLA2 is indeed capable of permeabilizing the vesicles it r esides in, the size of the holes produced were insufficient to allow the passage of a 10 kDa dextran. This would indicate that the same holes are also insufficient for mediating escape of the much large parvovirus capsid (110) While it is clear that PLA2 is necessary for endosomal escape (33, 73, 109) there are new findings which propose an alternative scenario, where small endosomal vesicles containing AAV traffic in a rapid, unidirectional manner to the nucleus along the microtubule network (125) In this scenario, rather than emerging from endosomes early, AAV permeablizes them enough to display its VP1 unique region and it s targeting domains on the exterior of the vesicle, thus directing the entire ves icle towards the nucleus. This model however, doesnt address the eventual need for the virus to escape the vesicle and enter the nucleus assuming the virus uses the nuclear pore


21 complex (NPC). It does however highlight the point that the cytoskeleton pl ays a critical role in the AAV life cycle. W hile it was believed that AAV utilized the microtubule network to travel to the nucleus via direct interaction after endosomal egress (57, 101) electron microscopy data from Xiao et al suggests that the virus is able to exploit the microtubule network from inside a vesicle (125) Given that AAV has also been colocalized with both the Golgi and the ER (50, 51) another possibility is that AAV follows a trafficking pathway similar to SV40 (102) and traffics through the Golgi into the ER where it exploits the unfolded protein response ( UPR) machinery to gain access to the cytoplasm and ultimately the nucleus. Inspite of the uncertainty about the exact pathway taken by AAV through the endosomal system, the fact that AAV uptake occurs by clathrinmediated dynamindependent endocytosis has been widely accepted and unchallenged until recently. However a recent study (84) which looked at the effects of a combination of dominant negative mutants in th e clathrin uptake pathway and a variety of specific pharmacological inhibitors of endocytosis, has cast doubt on this aspect of AAV biology and has presented evidence that AAV uptake occurs independent of clathrin, calveolin (as with SV40 (102) ) and dynamin. This new work instead implicates Cdc42, Arf1 and GRAF1, which are factors in the formation of clathrinindependet carriers, as responsible for AAV uptake into detergent resistant GPIanchoredproteinenriched endosomal compartments. The CLIC/GEEC pathway then leads to AAV trafficking into the Golgi and i nvolves compartment s that undergo acidification as seen in the early/late endosomes. While these findings do seem to correlate with the known role of pH in AAV trafficking or the reports of AAV Golgi localization, they are in direct contradiction of


22 even m ore recent findi ngs that once again implicate c l a thrin and dynamin in the internalization of AAV2 and 8 (70) The picture is further complicated by the findings that the phenotypes of insertional mutants of the AAV2 heparinbinding motif can be divided into two categories, with heparanbinding mutants entering via a clathrin and caveolinindependent pathway while nonheparanbinding mutants entering via clathrinmediated endocytosis (114) This same study also showed that the mutants entering via the clathrin and caveolinindependent pathways were unable to transduce the cells, thus implying that entry alone is not sufficient for productive infection and that the route taken is critical to the final outcome Looking closer at the three studies, the two implicating clathrin both utilized chlorpromazine as an inhibitor of the pathway. While chlorpromazine does inhibit the binding of the A2 clathrin subunit on plasma membranes, t hus preventing the formation of clathrincoated pits (117) it also has wideranging effects on a variety of receptors and signaling pathways (reviewed in (23) ). This leaves open the possibility that the inhibition of AAV transduction seen wit h chlorpromazine treatment is due to an off target effect in the cell, rather than inhibition of clathrin mediated endocytosis of the virus. That said, chlorpromazine has been widely used as a specific inhibitor of clat hrin mediated endocytosis In contras t the study implicating the CLIC/GEEC pathway used Eps15, which has been shown to be a crucial part of the A2 clathrin complex and whose truncated form can be used to abolish clathrin coated pit formation (10) This approach carries less risk for off target effects. Nonethe less the discrepancy among these studies highlights the confusion within the AAV trafficking field and underscores the need for a better understanding of the underlying cellular signaling and trafficking pathways.


23 While it is not clear which of t he above models is correct or more relevant to productive AAV infection, there are several things that are know n about AAV infection prior to arrival to the nucleus. AAV requires acidification and maturation of the endosomal compartmen ts and infection is blocked with drugs such as bafilomycin A (29) The need for endosomal acidification goes beyond the simple need to escape into the cytoplasm, since mic roinjection of AAV into the cytoplasm, and even into the nucleus fails to result in a productive infection (107) It has been believed that acidification is necessary for the externalization of the VP1 N terminal unique region (73, 107) which contains the crucial PLA2 domain, NLS signals as well as trafficking domains Acidic pH has however failed to produce VP1 externalization in vitro (107) suggesting that either other cellular factors are necessary or tha t perhaps the pH dependent process is simply reversible, and thus, not detected at the typically neutral pH during immunoblotting What is clear is that trafficking through the endosomal pathway results in processing of the AAV capsid, which is critical for infection. Some of this processing was believed to have been done by resident proteases, and indeed there is some evidence that cathepsins B and L are able to process AAV2 and 8 (2) and appear to have a modest effect on AAV infection. Given that the study also shows no such effect with AAV5, it is unclear how indicative these findings are of AAV infection in general It is also known that the proteasome plays a negative role in AAV infection. Treat ment with proteasomeblocking drugs such as MG132 results in enhanced AA V infection (31, 128) which is thought to be the result of decreased viral degradation in the cytoplasm This is in contrast to autonomous parvoviruses, which actually require


24 pr ocessing by the proteasome (95) Ubiquitation of proteins can serve to target protein to various cellular compartments, depending on the number and the lysine linkage of the additional ubiquitin molecules. Polyubiquitination at lysine 48 is a classical signal for proteasome degradation (21) One process by which proteins can be targeted for ubiquitination is phosphorylation (15, 20, 88, 89) In the case of AAV it was initially shown that inhibition of epidermal growth factor receptor protein tyrosine ki nase (EGFR PTK) resulted in increased AAV transduction (72, 91) It was later demonstrated that EGFR PTK was able to phosphorylate surface tyrosines on the AAV2 capsid, and that this phosphorylation was linked with increased ubiquitination and proteasomemediated degradation of the virus (129) Indeed mutation of four out of eight surface tyrosine residues to phenylalanine (that cant be phosphorylated) resulted in a modest to moderate increase in AAV infectivity. One such tyrosine, Y704, is located in the pH q uartet This residue will be of particular interest in our work described in Chapter 3. When discussing AAV infection, it is important to acknowledge one more aspect, which is engrained in the very nature of this genus: the helper virus. At first, it would seem natural to study AAV infection in the context of an adenovirus infection, since the virus is unable to complete it life cycle without a helper (discussed further in Replication) However the interest in AAV as a vector also complicates its study s ince Adenovirus coinfection is not practical for clinical applications While perhaps the primary utilitarian goal of studying AAV is to understand it for the sake of making a better vector, some of the inconsistencies that obviously exist when comparing different studies may exist due to the fact that helper free infection is not how the virus typically spreads in the wild and thus doesnt accurately reflect the environment in which it has


25 evolved to function. Although AAV is obviously capable of helper free transduction with robust gene expression, especially in the self complimentary vectors, the ambiguities seen in trafficking may in part be due to AAV evolving to function in the context of a cell already infected by a virus. Adenovirus has been demonstrated to enhance AAV nuclear trafficking (126) This may be due to the endosomedisrupting properties of adenovirus protein VI ( 119) which have been shown to at least mildly facilitate endosomal escape of PLA2deficient AAV (109) This of course could explain, why in Ad free infections, a large portion of the incoming AAV virions seem to scatter throughout the endosomal system in the absence of a strong endosomal disruptor like protein VI. T his suggests that there are two types of infections, with and without Adenovirus, and that AAV has developed strategies to survive both scenarios. Thus, by examining AAV in the context of Ad, rather than just focusing on Adfree infection as is often done, we increase our chances at better under standing the underlying biology of the virus and ultimately engineering it for our use. AAV and the Nucleus AAV, like most DNA viruses must deliver its genome to the nucleus for successful replication and gene expression. What has not been clear until rec ently was if the AAV capsid entered the nucleus or simply injected its genome as seen with herpesviruses and adenovirus ( review ed in (62) ). Early evidence suggested that capsid entry into the nucleus was highly inefficient (9). T his was based on confocal microscopy findings which reveal ed few Cy3 labeled capsids in the nucleus, suggesting that either capsid entry into the nucleus wasnt necessary or that the capsid disintegrated prior to nuclear entry as seen with hepadnaviruses ( reviewed in (62) ). S ubsequent efforts revealed that A20 positive capsids could indeed be observed in the nucleus (126) Further evidence


26 that the intact capsid enters the nucleus and is critical for infection, was provided when microinjection of the A20 neutralizing antibody into HeLa nuclei was shown to be inhibitory (107) Because the A20 antibody recognizes only intact AAV capsids, this meant that the entire capsid enters the nucleus priority uncoating. Having established that the intact capsid d oes enter the nucleus, the next question facing the field was how? Given that the VP1 unique region harbors four basic clusters of amino acids, which have been shown to function as nuclear localization signals (NLS)(41) it was logical to assume that AAV utilized the nuclear pore complex (NPC) to gain access to the nucleus. The typical nuclear complex can dilate up to 39nm to accommodate large cargo during active transport (87) and would thus be more than capable of allowing the 26nm AAV capsid to pass. There has however been evidence that the virus is able to enter the nucleus without the NPC. Incubation of virus with purified nuclei, revealed that AAV was able to enter even in the presence of wheat germ agglutinin (WGA), an inhibitor o f active nuclear transport (45) This study also showed that the virus entry kinetics were dependent on time, temperature and virus concentration but were not saturable, further indicating that AAV enters on its own, without the assistance of cellular machinery. Further evidence that this NPC independent entry method may be shared by all parvoviruses is seen in studies done with minute virus of mice (MVM). Initial observations of MVM microinjected into Xenopus oocytes, revealed that the virus caused disruptions within the nuclear envelope (NE) and nuclear lamina (NL), suggesting that it gained access to the nucleus directly, rather than through the NPC (24, 25) These same studies also revealed that the damage to the NE was timeand dosedependent but was not affected by inhibition


27 of the NPC; just as observed with AAV. Further examination revealed that MVM was able to achieve these disruptions without the help of its PLA2 domain, and that it did so with the help of caspase3. While caspase3 is a strongly proapoptotic enzyme when activated, in the case of MVM, the observed disruption of the NE and underlying nuclear lamina were achieved by relocating a small fraction of caspase3 that is activated in the cytoplasm at any given time, rather than activating the larger pool of caspase3 zymogen. This would of course prevent cells from prematurely entering apoptosis. While it is not clear if AAV uses the same tactic we have observed that AAV is able to cause similar disruptions in the NE and NL as judged by confocal microscopy (unpublished observations) Recent findings of a nuclear entry role for a PDZ binding domain in the N terminus of VP1 may point to underlying interactions, governing an NPCfree nuclear entry process. This leaves open the possibility that this unique aspect of parvovirus biology is conserved between the autonomous parvoviruses and dependoviruses. The fact that intact capsids must reach the nucleus suggests that AAV capsids must reach a particular nuclear location or interact with nuclear components for infection to continue. A study, looking to initially characterize the interacting partners of AAVs Rep proteins, stumbled upon the fact that a nucleolar protein, B23/nucloephosmin/NPM bound the AAV capsid with greater affinity than Rep (13) The study went on to describe how NPM appeared to facilitate Rep78 binding of the AAV ITR as well as Rep endonuclease activity. Given that NPM is known to facilitate interactions between proteins and nucleic acids in ribosomal assembly and chromatin remodeling (reviewed in (69) ), these findings are not surprising and may suggest a reason for the AAV capsid


28 to seek out and bind NPM. In another seemi ngly unrelated study, Johnson et al demonstrated that treatment with proteasome inhibitors not only resulted in enhancement of AAV transduction but also an increased translocation of A20positive capsids into the nucleolus (52) Nucleolar localization was observed to a lesser extent in untreated cell s infected with AAV in the absence of adenovir us Interestingly the same study also observed that treatment with hydroxyurea (HU), an inhibitor of DNA synthesis which has been shown to increase AAV transduction (53) resulted in release of AAV into the nucleoplasm. Both treatments resulted in increased transduction by AAV and were additive when combined suggesting they affected different aspect of AAV biology in the nucleus To further examine the link between the nucleolus and AAV, the same study also knocked down expression of NPM and nucleolin (NCL) using siRNA, and reported a modest increase in infectivity. NPM siRNA resulted in increases nucleolar accumulation and an increase in transduction, as seen with MG132. NCL siRNA treatment also resulted in increased transduction but this time accompanied by a migration of viral particles into the nucleoplasm, as seen with HU treatment. The authors thus suggested that AAV first utilizes the nucleolus as a safe haven within the nucleus where it can remain until conditions favor egress from the nucleolus and uncoating. Findings with autonomous parvoviruses show that viral infection induces marginalization of chromatin and an increase in protein mobility within the nucleus (48) suggesting a scenario in which the parvoviral capsid could use NPM to induce favorable nuclear remodeling It seems clear that the AAV capsid has a nuclear role i n productive infection, but despite tantalizing clues, the exact nature of that role is still unclear.


29 Replication and Assembly Productive AAV replication requires coinfection with a helper virus such as Adenovirus, Herpesvirus Cytomegalovirus or Papilloma virus (5, 17, 78, 85) In the case of Adenovirus it has been established that wt AAV infection requires at the minimum the E1a, E1b, E2a, E4Orf6, and the virus associa ted [VA] RNA (18, 19, 43, 67, 93, 94) E1A is a transactivator that induces cells into S phase and is responsible for upregulating transcription of Adenoviral and AAV genes. E1B works in concert with E4 to facilitate transport of AAV mRNA, with E4 also playing a role in AA V DNA replication and secondstrand synthesis E2A and VA RNA proteins are likewise responsible for enhanced accumulation of AAV mRNA and proteins (34, 78, 81, 98) In the absence of helper virus infection, wildtype AAV has been shown to undergo sitespecific integration on human chromosome 19 ( 19q13.3 qter ), at a site called AAVS1 (63 65, 100) In the absence of Rep proteins, as is the case with recombinant AAV (rAAV) vectors, the genome has been shown to be maintaine d long term in the form of an episome (127) A AV genomic DNA is replicated by a strandd i splacement mechanism, with the 3 ITR serving as a primer for complimentary strand synthesis This model, typical of parvoviruses, requires the nicking activity of Rep78/68 to facilitate terminal resolution(49, 106) A mutation within the 3 ITR terminal resolution site ( TRS ), prevents terminal resolution and results in a duplex genome which can still be packaged (74) The use of duplex genomes in rAAV vectors circumvents the secondstrand synthesis bottleneck and leads to rapid gene expression in the absence of helper virus. This however comes at the cost of reduced packaging capacity (121) With wild type and


30 single stranded rAAV, genomes of both polarities are replicated and packaged equally in the nucleus (11, 12) AAV capsid formation occurs independent of the Rep proteins or the viral genome but requires the expression of AAP (108) AAP facilitates nucleolar localization of VPs and the formation of intact, A20 positive capsids The monomer interfaces forming the fivefold and twofold axis, are relatively simple and asse mbly along them is believed to occur spontaneously In contrast, the three fold axis of symmetry is formed by complex hook in loop interactions, which have been hypothesized to require chaperones for proper formation. Specific Aims As AAV continues to be developed for therapeutic use and deployed extensively in clinical trials, the gaps in our knowledge of the molecular events which govern AAV infection are made all the more apparent. Developments, such as the phenylalanine/tyrosine mutants and self compl imentary vectors, underscore the practical importance of having a thorough and complete understanding of the basic biology of AAV. Despite numerous studies implicating acidic pH as critical to AAV infection, and identifying regions within the capsid that are responsiv e to changes in pH, the exact role of these regions in AAV infection is still unknown. Additionally, a number of mutants have been identified as being critical to AAV infection post entry, however the exact role of these residues as it pertains to AAV trafficking remains unknown. To this end, the project described here has developed into two interconnected directions.


31 Mutational and Traf ficking Analysis of M ut22 and p H R esponsive Amino Acid s of the AAV2 C apsid We examined the infectivity and characterized the trafficking of the previously identified mut22, in an effort to gain insight into the nature of its defect. Similarly we mut ated amino acids within the pH q uartet and the DNA group to examine the role of these pH sensitive elements in AAV infection. We also mutated individual residues within mut40 as well as Y700 and Y730 due to t heir close proximity to the pH q uartet and their previous im plication in AAV infection (122, 130) All studies were conducted in the presence of Adenovirus 5 coinfection. Identification and Characterization of the AAV Capsid Protease Structural analysis of the pH q uartet region revealed similarities with acid protease active sites. Our aim was to determine if the capsid indeed had protease activity and to characterize it. We identified a pH dependent protease that is capable of autolysis of the AAV VPs in response to acidic pH. We demonstrated that this activity is conserved across serot ypes and appears to involve multiple cleavage sites within the viral coat protein. We were able to identify, through both biochemical and mutational analysis, unique cleavage sites in AAV1 and 9. Our close collaborator additionally showed that the capsid also possessed protease activity against external subst rates and that mutating the pH q uartet residue, E563 (AAV2), could abolish this activity (96)


32 CHAPTER 2 MATERIALS AND METHOD S PCR Mutagenesis and Plasmid DNA Purification All AAV2 mutants were generated in a pIM45 background using the Quikchange II XL site directed mutagenesis kit from Agilent (Santa Clara, CA). Similarly AAV1 and AAV9 mutants where generated using the same method in a pXR1 and Rep2Cap9 background respectively. Afte r PCR, products were digested with DpnI, transformed into XL10Gold ultracompetent cells and plated on ampicillincontaining agar plates. Single colonies were picked and grown up overnight in liquid cultures with ampicillin. A small sample from each cultur e was frozen in 15% glycerol at 80 C for future use while the rest of the culture was extracted for plasmid DNA using a Mini Prep kit (Qiagen, Valencia, CA ). Mutations and a proper background were verified by sequencing the full cap ORF Cell Culture HEK293 and HeLa cells were maintained at 37 C and 5% CO2 in Dulbeco Modified Essential Media (DMEM), supplemented with 10% heat inactivated fetal bovine serum, 100 U/m L penicillin, and 100 mg/m L streptomycin. Virus Production AAV2 pH Mutants Mutant viral stocks were produc ed by a threeplasmid transfection of HEK293 cells using polyethylenimine (linear molecular mass ~ 25000) with either wildtype or mutant pIM45, pXX6 and either pTR UF11 (single stranded eGFP genome) or pds eGFP (self complimentary eGFP g enome). At 68 hrs post transfection lysates were collected and a small fraction was saved for dot blots while the remainder was purified by iodixanol step


33 gradient centrifugation and heparin column chromatography as previously described (131) Typical preparation size consisted of 20 x 15 cm2 plates of HEK293 cells. For large scale preps, mutant viruses were produced by the UF Vector Core. AAV1 2 and 9 Protease Mutants Mutant viruses were prepared by polyethylenimine transfection o f 5 x 15 cm2 plates of HEK293 cells and purified by iodi xanol step gradient centrifugation as previously described (23). All mutant viruses were packaged with pTR UF11 containing a GFP expression cassette. After iodixanol c entrif ugation, the material from the 40%/60% interface was washed in a 150 kDa cutoff concentr ator (Apollo; Orbital Bioscience, Topsfield, MA) with 50 m L of 1 x TD (1 x PBS plus 1 mM MgCl2 and 2.5 mM KCl) plus 1 M NaCl. The virus was then buffer exchanged into 1x PBS (pH 7.4) and concentrated to 500 L Wild type Serotype 1 10 Samples Wild type capsid samples were obtained for serotypes 110 for various experiments from the following sources and prepared as follows. AAV1 to 3, AAV5 to 8, and AAV10 virus particles containing a green fluorescent protein (GFP) expression cassette (pTR UF11) were prepared by DNA transfection in HEK293 cells and purified by iodixanol step gradients and Mono Q FPLC chromatography as previously described (132) AAV9 virus particles (VPs) containing a GFP cassette were purified by iodixanol gradient, washed with 1 M NaCl, and buffer exchanged into 1x phosphatebuffered saline (PBS). In addition, AAV1 containing an 1 antitrypsin (AAT) cass ette (a kind gift from David Knop at Applied Genetic Technologies Corporation [AGTC], Gainesville, FL) was made by an alternative protocol using herpesvirus infected BHK cells and purified as previously described (113) Briefly, BHK cells growing in suspension were coinfected


34 with two recombinant herpes simplex virus (rHSV) vectors and then simultaneously lysed and digested (1% [vol/vol] Triton X 100, 25 U/mL Benzonase, 2 mM MgCl2) 24 h later. The lysate was filter clarified and then column purified by CIM Q (BIA Separations) and affinity (AVB; GE Healthcare) chromatography. The GFP containing virus particles were used for autoproteolysis assays as described below. AAV1 containing an AAT cassette was used for AAV1 mass spectroscopy and Edman sequencin g. Virus stocks for autoproteolysis made by DNA transfection were obtained from the University of Florida Vector Core Laboratory (AAV1, 2, 5, 8, and 9), Sergei Zolotukhin (rh10), Mavis AgbandjeMcKenna (AAV3 and 6), and Roland Herzog (AAV7) Virus Tit ering All viral stocks were titered by real time PCR using BioRad MyIQ cycler and iQ SYBRGreen supermix (Bio Rad, Hercules, CA) with primers against the chicken beta actin (CBA) promoter common to both ss DNA and ds DNA genomes (FWD: TCCCATAGTAACGCCAATAGG REV: CATCAAGTGTATCATATGCCAAG). P article to Infectivity Ratios HeLa cells were seeded at 1x104 cells per well in 96 well plates 12 hrs prior to infection. Cells were infected with serial dilutions of mutant and wild type AAV ranging from 0. 1 to 1x106 viral genomes per cell and all infections were supplemented with MOI =10 of Adenovirus 5 and done in triplicate. At 48 hrs post infection ( pi ) plates were imaged with a Zeiss Axiom inverted fluorescent microscope using GFP and/or RFP filters. Cells posit ive for the proper fluorescent protein were counted in the imaged area and the total number of positive cells in the well was back calculated. Cells infected with only Ad5 were used as negative controls for autofluorescence. Particle to infectivity ratio


35 w as calculated by dividing the known virus genome input by the number of GFP/RFP positive cells. Western Blotting and Dot Blotting For dot blots of transfected lysates, equal amounts of protein (60 re loaded on either Whatman Protran nitrocellulose blotting m embranes [ Figure 3 2 and 42] (GE Healthcare), or Immun Blot polyvinylidene difluoride (PVDF) membranes [all other blots] (Bio Rad, Hercules, CA) in their native state, to assess intact capsids, using a vac uum blotting apparatus For assessment of total VP present samples were boiled, separated on 420% gradient Tris HCl polyacrylamide gels, and then transferred to nitrocellulose membranes. Blots were then blocked for 2 hrs with 5% nonfat milk (BioRad Hercules, CA ) in 1x PBS/ 0.5% Tween20 and pr obed with either B1 monoclonal antibody (1:3000) to measure total capsid protein or A20 (1:20) to measure intact capsids for 1hr at room temperature. Membranes were then washed 3 x 5 min in 1x PBS/ 0.5% Tween and incubated with anti mouse (1:5000) or anti rabbit secondary HRP conjugated antibodies (company) for 1 hr at RT. After a final 3 x 5 min washes, membranes were developed using Millipore ECL reagent (Millipore, Billerica, MA) and imaged using HyBlot CL Autoradiography Film (Denville Scientific I nc, Metuchen, NJ) Immunocytochemistry and Confocal Microscopy For confocal microscopy 16well chambered slides (BD) were seeded with 7x103 HeLa cells 24hrs prior to infection, skipping wells to ensure that no cross contamination would occur during infec tion. Cells were infected with 2x104 vg/c ell of AAV with and without the addition of MOI =10 Ad5. At each time point, cells were washed three times with 37C 1x PBS and then fixed and stained as previously described (52) with the following additions. Nuclei were labeled using Lamin B1 rabbit polyclonal antibody


36 (Abcam; 1:600) and all samples were processed using imageIT FX s ignal enhancer (Life Technologies ) prior to blocking with immunofluorescence buffer (Johnson). All images were acquired with a Leica SP5 scanning laser confocal microscope and processed using Volocity (Perkin Elmer) and Photoshop (Adobe). We also found the A20 epitope to be particularly prone to destruction by over fixation either with higher concentrations of paraformaldehyde (PFA; 4% or greater) or when using pure, methanol free formaldehyde (2%). We also found that the manner in which PFA was prepared to be critical for success. Briefly, PFA was made fresh from powder prior to each experimen t in 10 mL batches. 62.5 N NaOH was added to 500 ddH2O in a small glass test tube. PFA powder (0.2 g) was then added to the test tube and mixed by swirling in an 80 C water bath until dissolved (approximate 2 minutes). The sample was then promptly brought up to a final working volume of 10 mL with 1x PBS. Subc ellular Fractionation HeLa cells were seed at 1x105 cells per well in 6 well plates, 12 hrs prior to infection. Cells were infected with 1x104 viral genomes per cell of wildtype or mutant rAAV and co infected Ad5 at an MOI of 10 Cells were infected by incubating with virus at 4 C for 30 min in 1mL of serum containing media. After 30 min cells were shifted to 37 C for 15 min to allow entry of bound virus and then virus containing media was removed and replaced with fresh, prewarmed serum containing media. At the appropriate time, cells were collected by trypsinization and pelleted in a refrigerated centrifuge a t 250 x G Cells were then washed 3x with icecold 1x PBS and fractionated into cytoplasmic and nuclear fractions with the NE PER kit (Pierce; Rockford, IL) as per the manufacturer s instructions. The fractionation process resulted in cytoplasmic and


37 nuclear fractions as well as a post nuclear pellet that contained the nuclear membrane and cellular DNA. Each fraction was adjusted to 200 resuspended in 200 2O. All infections were done in triplicate. Viral DNA Extraction and Real Time PCR Subcellular fractions were supplemented with 20 mM Tris HCl pH 8, 100 mM EDTA, 5% SDS) a nd 2 U of proteinase K ( Fermentas, Thermo Scientific, Glen Burnie, MD ). Samples were allowed to digest for 1 hr at 50 C and then heated to 95 C for 20 min as a final denaturing step. Proteins were then extracted 3x with phenol:chloroform:isoamyl alcohol ( 25:24:1 Life Technologies ) and DNA was precipitated with the addition of 1 lue ( Ambion, Life Technolgies ), 1/10th sample volume of sodium acetate (3 M, pH 5.2) and 2.5x volumes of 95% ethanol. Samples were allowed to precipitate overnight at 20 C and were then spun down at 20,000 RCF at 4 C for 30 min. DNA pellets were washed with 70% ethanol and then allo wed to very briefly air dry prior to resuspension in 100 2O. 1/10th dilutions were then made of all samples and viral genomes were measure by real time PCR with primers against the CBA promoter (see virus titering ). All samples were measured in tripl icate. Uncoating Assay When assessing the relative portion of uncoated virions in the nucleus, cells were prepared, infected and fractionated as described above. Once nuclear fractions (soluble nuclear fraction and post nuclear pellet) were obtained using the NE PER kit (Thermo Fisher, Rockford, IL), fractions were adjusted to 200 L with benzonase buffer (10 mM Tris pH 7.5, 5 mM MgCl2) and then split in half. One half was labeled as total and processed as described in the subcellular fractionation section while the other half was


38 subjected to digestion with 250 U of Benzonase (Sigma Aldrich, St. Louis, MO) for 1 hr at 37 C. Samples were then processed with proteinase K as detailed before. Both the total and benzonased samples were subjected to real time PCR with CBA primers as described and the relative fraction of DNAse resi stant genomes was determined. Capsid Autolysis Assay Viruses containing a GFP expression cassette, which were made by the DNA transfect ion method, were used for autolysis, mass spectroscopy (MS), and Edman sequencing. Approxi mately 1 x 1010 viral genomeco ntaining particles were incubated in ei ther 1x PBS (pH 7.4) or 0.1 M citrate buffer (pH 5.5) for 1 h at 37 C. Samples were incubated at 95 C for 5 min in the presence of Laemmli buffer, separated by SDS PAGE using 4 t o 20% gradient gels, and trans ferred onto Immun Blot polyvinylidene difluoride (PVDF) membranes (0.2m pore size; BioRad, Hercules, CA). Membranes were blocked (5% milk) and probed with either B1 (1:3,000) or A1 (1:20) monoclonal antibody for 1 h at room temperature, followed by incubation with anti mouse horseradish peroxidase (HRP) conjugated (1:5,000) secondary antibodies (GE Healthcare Bioscience, Pittsburgh, PA) and enhanced chemiluminescence (ECL) visualization. Estimation of fragment sizes was done with ImageQuant TL software (G E Hea lthcare Bioscience, Pitts burgh, PA) analysis of scanned immunoblots using Precision Plus prestained protein size standards (Bio Rad, Hercules, CA). Mass Spectrometry Mass spectroscopy (MS) analysis was conducted on AAV1AAT cleavage fragments with sizes of ~ 17 kDa and ~ 50 kDa. Prior to analysis, 1 x 1011 genome containing partic les were subjected to autolysis at acidic pH as described above. Fragments were then separated by SDS PAGE as described above, and cleavage


39 bands were visualized by staining the gel with biosafe Coomassie blue (BioRad; Hercules, CA). The desired cleavage bands were excised with a scalpel and submitted to the Protein Core of the Univer sity of Florida Interdisciplin ary Center for Biotechnology Research, where they were treated with 10 0 mM iodoacetamide (45 min at room temperature) and try psin in preparation for MS analysis. Peptides were extracted from the gel slices with 200 L 80:20 acetonitrilewater containing 0.1% formic acid and vacuumed to dryness. Samples were dissolved in 3% acentonitrile, 0.1% acetic acid, and 0.01% trifluoroacetic acid (TFA). Liquid chromatography tandem MS (LC MS/MS) analysis was carried out on a LTQ Orbitrap XL mass spec trometer (ThermoFisher Scientific, West Palm Beach, FL). All MS/MS samples were analyzed with Mascot vers ion 2.2.2 (Matrix Science, Lon don, United Kingdom). Scaffold version (Proteome Software, Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. Peptide identifications wer e accepted if they could be es tablished at greater than 95.0% probability, as specified by t he Peptide Prophet algorithm (58) Protein identifications were accepted if they could be established at greater than 99.0% probability and contain at least 2 identified unique peptides. Protein probabil ities were assigned by the Protein Prophet algorithm (82) Edman Degradation Sequencing For Edman sequencing, 2 x 1012 genome containing AAV1AAT or AAV9 GFP particles wer e subjected to acidic autolysis. The s am ples were then boiled in Laemml i buffer and concentrated to 30 L by vacuum centrifugation. The concentrated samples were buffer exchanged into 10 mM Tris HCl using Micro Bio spin 6 size exclusion columns per the manufacturers instructions (BioRad, Hercules, CA). Samples were then separated by gradient SDS PAGE as previously described and trans ferred to


40 Immobilon PSQ PVDF membranes (Millipore, Biller ica, MA). Membranes were washed with doubledistilled H2O (ddH2O) and then stained with fresh Coomassie R 250 brilliant blue for 30 s followed by destaining with 40% methanol 5% acetic acid. Membranes were washed 3 times with ddH2O and allowed to air dry prior to cutting of bands. Individual bands were submitted for N terminal sequencing (Edman deg radation) to Alphalyse (Palo Alto, CA).


41 CHAPTER 3 AAV2 CAPSID MUTAGENESIS Endosomal processing is required for productive AAV infection (9, 46, 107) Part of this processing entails a low pH environment, and drugs blocking endosomal acidification have b een shown to inhibit AAV infection (9). Recent crystallographic evidence points to two regions undergoing pH dependent shift (79) Both areas are comprised of highly conserved residues (Table 3 1 ). The first, located on the inner surface, surrounding the three fold axis of symmet ry (Figure 3 1C) appears to be involved in pH dependent interactions between the capsid and DNA. This group is comprised of four amino acids, (P419, P630, H629 and F628) and will be referred to as the DNA group (Figure 3 1D) The second cluster of amino acids, is located near the two fold axis of symmetry (Figure 3 1A) and is also made up of four amino acids (R389, E563, Y704 and H526 Figure 3 1B ). This group includes residues previously i mplicated in AAV infection (122, 130) and will be referred to as the pH q uartet. In this section we detail our mutational analysis of these two regions in AAV2 and describe the trafficking implications for some of the mutants utilizing immunocytochemistry, and subcellular fractionation. Additionally, we examine the trafficking of two of the most severe capsid infectivity mutants previously identified ( mut40, mut22, (122) ) one of which is p artially encompassed by the pH q u artet. We also describe the alanine mutants of two tyrosine residues (Y700, Y730) which are adjacent to the pH quartet, and that have been previously i mplicated in AAV infection (130)


42 Results pH Quartet Mutant Phenotypes Within the pH quartet we found that all mutants produced normal levels of intact (Figure 3 2 A and B ), packaged virus (data not shown). The E563A mutant demonstrated a modest effect on infectivity (22 fold) as seen in Figure 33, while Y704A produced a defect so severe that it was below the level of detection even at high MOI (> 8 logs). It is important to note that this same tyrosine residue, when m utated to a phenylalanine by Zho ng et al (130) produced a mild increase in infectivity. Despite having displayed similar degrees of shift in the pH crystal structures of AAV8 the severity of the defects displayed by the E563A and Y 704A mutants in AAV2 differed greatly and we decide d to further examine the defect displayed by these two mutants. Entr y and Nuclear Trafficking of pH Quartet Mutants To learn more about the nature of the defect in the Y704A and E563A mutants, we examined capsid cell entry and nuclear trafficking by confocal microscopy as a function of time post infection. Intact capsids were visualized with A20 monoclonal antibody, an antibody that recognizes only intact AAV2 capsids, at different times post infection. Lamin B1 antibody was used to delineate the nuclear membrane. As previously reported, wildtype virus was observed entering the nucleus as early as 8 hrs post infection (52) and by 16hrs, appeared in the nucleus i n significant quantities (Figure 3 4 ). We observed some concentrated clusters of wildtype virus in the nucleus which resembled the nucleolar localization previously seen by others (52) but this phenomenon w as observed in a small number of cells. In contrast E563A showed very little nuclear entry even at 24hrs post infection (data not shown). In the case of Y704A we observed an almost complete lack of viral signal in the nucleus regardless of time


43 post infect ion or the presence of Adenovirus. Despite the decrease in nuclear staining, both E563A and Y704A showed strong perinuclear A20 signals, suggesting that neither mutant was defective for cell entry. Instead, it appeared that both mutants were defective in nuclear entry Although this could explain the low infectivity of Y704A, the confocal data for E563A was not consistent with the fact that E563A was only 20 fold less infectious than wild type virus. To confirm the confocal findings, cells infected with w ild type, E563A, or Y704A were fractionated as described in Methods to separate the cytoplasmic and nuclear fractions. Virus was bound to cells at 4oC for 30 min at a MOI of 10,000 vector genomes per cell, and then the cells where incubated at 37oC to init iate cell entry. The cells were also coinfected with Ad5 at an MOI of 10 to simulate a normal productive infection. Quantitative real time PCR measurements of viral DNA isolated from the two fractions were used to track virus progression through the cytoplasm and into the nucleus as a function of time post infection. The nuclear and cytoplasmic markers GAPDH and Lamin B1 were used to confirm the purity of the nuclear and cytoplasmic fractions ( Figure 3 5 ). Wild type virus continued to enter the cell for the first 8 hours of infection at which point nearly all of the (10,000/cell) wild type input genomes had entered the cells (Figure 3 6) After 8hrs post infection, the total wild type viral genome content in the cell began to drop, in all likelihood as a result of losing some of the input virus in various degradation pathways. Both mutants also continued to enter the cell after the first h our post infection. However, the peak genome levels found inside the cell were 38% (16 hrs) and 15% (4 hrs) of input f or Y704 A and E563A, respectively (Figure 3 6 ) It was not clear whether this reflected a defect in entry for the two mutants or an increased rate of degradation for the mutant genomes after entry into


44 the cell. The 8 fold lower level of E563A genomes found in the cell could in part account for the 22 fold lower infectivity of this mutant; but clearly, the 3 fold lower entry of Y704A could not account for its 8 log drop in infectivity. Surprisingly, when the distribution of viral genomes between the nuc leus and the cytoplasm was determined, we found that contrary to the confocal findings, both mutant viruses delivered their DNA to the nucleus in significant quantities. These were however lower than WT due to the previously mentioned defects. By 8hrs pi, 62% of the E563A mutant genomes that entered the cell could be detected in the nuclear fraction, which was comparable to wild type (61%), and by 16 hrs the fraction of E563A in the nucleus was significantly higher than wt (79% vs 66%) (Figure 3 7 A). In ad dition, the rate of entry into the nucleus during the first 4 hrs pi was similar for wt and E563A compare Figure 3 7B with C). In contrast, Y704A displayed a lag in nuclear entry during the first 4 8 hrs and delivered a significantly lower fraction of genome copies to the nucleus compared to wt at 8 hrs (Figure 3 7 A, D). However, by 16 hrs, there was no significant difference in the fraction of genomes in the nucl eus between wt and Y704A (Figure 3 7 A, 68% and 66% for wt and mutant, respectively). We concl uded that although there were modest differences in the rate of genome entry into the nucleus, they were too small to contribute to the phenotype of either mutant in a significant way. Both Y704A and E563A delivered approximately the same fraction of intracellular viral genomes to the nucleus. Uncoating and SecondStrand Synthesis Since the defect in cell entry and nuclear trafficking seen with Y704A could not account for the >8 log deficit in transduction, we hypothesized that this mutant might be defectiv e for uncoating after entering the nucleus. To test this nuclear fractions were


45 obtained at 16 hrs as in the previous experiment and split in half. One half was treated with benzonase to measure genomes that were still encapsidated, while the other half was used to determine the number of total genomes present. To our surprise we found that the fraction of DNAse sensitive genomes (presumably uncoated) was not significantly different between wildtype and mutant viruses (Figure 3 8 ). In addition, when we calculated the average number of genomes per nucleus that were uncoated (DNase sensitive) there was no difference between wt and Y704A, approximately 100 genom es per cell in each case (Figure 3 8 ). Thus Y704A was approximately 2 times better than wt at unc oating once it entered the nucleus ( Figure 3 8 ), and this compensated for the fact that it delivered only half the number of total genomes to the nucleus as wt (at 16 hrs) This clearly showed that this mutants severe defect in infectivity was not simply due to uncoating. The only step left in the known infectious process is second strand synthesis, which must occur prior to transcription. To test if Y704A had a defect in second strand synthesis we compared the particle to infectivity ratio of Y704A containing self complementary (SC) GFP genomes with those containing single stranded (SS) genomes. We also tested the infectivity of the remaining residues (D561,E562, E564) in the previously identified mut40, which include the two perfectly conserved glut amic acids that surround E563. Both SS and SC versions of these mutants were tested and the resulting particle to inf ectivity ratios are shown in Figure 3 9 Infection with SS Y704A was beyond our detection limits (>8 log defect), however SC Y704A was able 5 vg/cell), This however still resulted in a particle to infectivity ratio that was 7 logs lower than wildtype. Mut40 infectivity was below our detection limits (>8 log defect) in both SS and SC versions of


46 the v irus. Mutation of D561, E562 and E564 revealed that the majority of the defect was the result of E562A and E564A mutations which with an SS genome displayed a defect of 7 logs and >8 logs, respectively. SC genome packaging produced a 3 log increase in inf ectivity for E564A, but as with Y704A, the resulting virus was still ~5 logs less infective than wildtype The same was not observed for E562A, which had a nearly identical particle to infectivity ratio, regardless of genome complementarity. While the inc rease in infectivity seen with SC564 and SC704 mutants is novel, the remaining defect in these viruses suggests that the self complimentary genomes are not relieving an inherent defect in secondstrand synthesis. What is more likely, is that at high MOI (w hen positive cells are observed for the SC viruses) the nucleus is flooded with double stranded DNA, some of which we have shown to be uncoated, and thus even Y704A and E564A are able to produce a limited number of positive cells The results with Y704A suggest that this single mutation disrupts a previously unidentified role of the capsid in gene expression that functions after nuclear entry, uncoating and secondstrand synthesis. To see if we could compliment this defect, we coinf ected wildtype capsids that were either empty or packaged with a CBA driven red fluorescent protein (RFP) cassette. We tested complementation of various concentrations of Y704A with 1x104 vg/cell of RFP wild type virus or an equivalent number of empty wil d type particles and examined cells for both GFP (Y704A) and RFP (WT) fluorescence. Unfortunately we observed some bleedthrough of the RFP signal in the GFP channel (Figure 3 10). While this prevented us from making a definitive conclusion, we believe that WT virus failed to compliment the mutant. This conclusion is based on the fact, that we observed a loss of signal in both channels as the ratio of wild-


47 t ype:mutant virus was shifted from 100:1 to 1:10. This implies that the mutant virus began to out compete the wildtype virus at the higher ratios thu s preventing RFP expres sion, and suppressing signal in the RFP channel ( and bleed through in the GFP ) Had there been complementation, we would have expect ed to see some loss of RFP at higher rati os but with GFP expression remaining relatively intact Alternately, our result could also be interpreted as Y704A capsids having a dominant negative effect on wildtype infection The empty capsid complementation produced no effect on Y704A infection (Figu re 3 10). Characterization of Mut22 Mut22 ( 268 272 NDNHY NANAY ) was originally characterized as a virus that produced a full, noninfectious particle. Subsequent analysis, by Wu Xiao (dissertation) determined that mut22 was able to bind heparin, enter cel ls efficiently, and traffic to the nucleus at levels comparable to wildtype, yet it remained severely defective for infection. In our analysis, we found that mut22 exhibited a very mild defect in assembly, which was estimate type (Fi gure 3 3 B ). We also confirm ed the defect in infectivity as being more than 8 logs (Figure 3 11), as compared to wildtype and beyond the detection limit of our assay. This phenotype was the same regardless of the the genome packaged (SS vs SC) Our initial characterization of mut22 trafficking by A20 immunocytochemistry over a 1 24hr time course revealed an apparent inability to enter the nucleus (Figure 3 12) This defect was in sharp contrast to wildtype virus and was present even at 24hr pi with and without Adenovirus 5 coinfection. As with our other mutants we attempted to confirm this lack of nuclear trafficking by a time course infection and measurement of viral genomes in cytoplasmic and nuclear fractions. Real time PCR measure of vir al genomes revealed that mut22 displayed a greater than 1 log


48 defect in cell entry (Figure 3 13). Despite this, the mutant displayed nuclear translocation similar to wild type (as percentage of total virus in cell). Mut22 also reached a peak intracellular viral load at 8hrs pi, just like wild type. Although our findings differ somewhat from those of Wu, in that we have identified a modest entry defect and a larger fraction of virus entering the nucleus, the conclusion thus far is the same. It appears that m ut22 is defective at steps post nuclear entry. Given our findings with Y704A, we have to assume that the defect may involve either uncoating, as originally predicted, or the new capsid role in the nucleus which functions after uncoating and secondstrand synthesis. DNA Group In the DNA region, all single mutants produced intact particles ( Figure 3 2 A and B ) at levels comparable to wt rAAV. Despite the interaction with DNA suggested by the crystallographic data, all mutants pack aged to the same extent as rAAV and only P419A demonstrated a significant (13 fold) defect in infectivity. The remaining three mutants did not significantly differ in infectivi ty from wild type rAAV (Figure 314). A quadruple alanine substitution of all four DNA group residues resul ted in an assembly defect in which no detectable virus was produced (data not shown) Discussion pH Quartet Our mutagenesis of the two pH sensitive regions in the AAV2 capsid has shown both regions to be important to AAV biology and has yielded several un expected results. In the case of the four residues in the pH quartet we identified Y704A as a severe (>8 log) mutant and E563A as a moderate mutant (22 fold), in comparison to Y704A We were able to show that E563A had corresponding defects in cell entry (1


49 log) and uncoating (2 fold) that could explain the extent of its defect in infectivity. In contrast, while Y704A showed a mild defect in cell entry (~3 fold) an d delayed nuclear trafficking, we could not identify the reason for its complete lack of transgene expression. Indeed, this particular mutant appeared to be able to enter the cell (Figure 3 6) traffic to the nucleus (Figure 3 7) and uncoat (Figure 3 8) but would then fail to result in a productive infection. Similarly to E564A, packaging Y704A wi th a self complimentary genome (Figure 3 9) did result in a modest (1 log) relief of its >8 log mutant phenotype. The resulting virus was however still ~7 log defective compared to wild type, suggesting that rather than alleviating a major fault in secondstrand synthesis the self complimentary genome allowed for some bleedthrough gene expression when the cell was flooded with double stranded viral DNA in high MOI infections. The fact that E562A did not ex hibit the same SC phenotype (Figure 3 9), suggest s that it may be defective for nuclear entry or uncoating. The fact that with only a single amino acid difference, the capsid was able to nearly completely shut down gene expression despite successful trafficking to the nucleus and uncoating suggests that the AAV capsid plays a new and unidentified role in AAV infection. Our preliminary analysis demonstrated that this function could not be complimented in trans with either full or empty wild type capsids (Figure 3 10) further suggesting that the capsid and its genome may still be intimately tied and functioning as a unit even after uncoating. Indeed our assay for uncoating only measured protection against a nuclease and did not provide any data as to the proximity of an uncoated genome to its parent capsid. These findings add to a growing list of observations that the capsid has an unknown but critical role in the nucleus. In support of this, the findings of Sonntag et al (107) showed that micro-


50 injection of the A20 monoclonal antibody into the nuclei of infected cells was able to neutralize infection, suggesting that the intact capsid was required in the nucleus for infection. Similarly with our mutants it is clear that while both are able to deliver their genomes to the nucleus and both retain at least a portion of those genomes in a DNAse protected state, the capsids are structurally different from w ild type as seen by the almost complete lack of A20 signal in the nucleus (confocal) as compared to wildtype (Figure 3 4) The findings with Y704A are reminiscent of the observations with the autonomous parvovirus Minute Virus of Mice (MVM). As previousl y described, only a two amino acid change within the MVM capsid could switch the virus between lymphotropic (I) or fibrotropic (P) phenotypes (6, 7) Further more this cell type restricti on occurred post entry While it was not directly shown that the particular MVM subtypes could traffic to nucleus and uncoat in nonpermissive cell typ es as seen with our Y704A mutant, it was implied since transcription was observed for both virus subtypes, regardless of cell type. Interestingly when examining the levels of the R1 and R2 transcripts, which encode the NS1 and NS2 proteins, the two subtypes displayed different ratios R1:R2 when infecting a hybrid cell line that was permissive to both viruses (7). This suggested that despite the minute difference in aminoacid sequence of the capsid protein, the two viruses differed in their interactions with the cellular splicing machinery, thus pointing us to a potential function that may be disrupted in the Y704A mutant. It is not clear if transcription occurs at all for our mutant however both the single stranded pTR UF11 and self complimentary pDS CB eGFP genomes that were used contain introns that must be spliced out for proper GFP expression. While the apparent A20 conformational


51 requirement suggest s that the capsid plays a direct role, such as a platform for recruiting cellular enzymes, the fact that this function could not be complimented in trans with wild type capsids implies that either the AAV genome must continue to function as a unit with its parent capsid or that a unique localization ( such as to the nucleolus or nuclear speckles) could be necessary and may be aberrant in our mutant. Further work examining the subnuclear localization of this mutant, the nature of its transcript profile and any of interactions it may have with nuclear ma chinery will be necessary to understand this unique new function for the AAV capsid. DNA Group In the case of the DNA group (628, 629, 630, 419) one mutant displayed a 13 fold infectivity defect (P419A), suggesting that despite the lack of any obvious defects in packaging the region may play a role in priming the genome for uncoating in response to an acidic environment. Even more interesting however is the fact that the simultaneous mutation of all four residues led to a complete assembly defect. This is r emarkable since none of the mutated residues are located on the interfaces that define monomer interactions between AAV VPs, and individually none result in an assembly defect. The assembly defect of the quadruple mutant does suggest that the DNA interacti ons observed in the crystal structures may actually be necessary for assembly. In addition to the findings published by Nam et al (79) there have also been observations that densities corresponding to ordered n ucleotides (either single or short chains) are found in virus like particles (VLPs) expressed without Rep prot eins and in the absence of AAV DNA (39, 68, 80, 83) thus suggesting that some form of DNA must be present for capsid assembly to occur even in the absence of the viral genome These findings were also based on structures of virus like particles (VLPs), which


52 lacked Rep and AAV genomes, further suggesting that DNA is required for AAV assembly. T he fa ct that the dodecahedral DNA cage observed in this study was also within the vicinity of previously identified assembly mutants (122) further supports the role of DNA in AAV assembly and lends merit to the authors proposal that DNA serves as a nucleating and binding agent to bring together VP proteins at the dodecahedral interfaces. A more recent study (37) has suggested that the threefold axis of symmetry may play host to two kinds of DNA capsid interactions. The above mentioned nongenome nucleotide interactions appeared to occur deeper within the depression on the inner surface of the capsid around the threefold axis, while the interactions with the genome appeared to take pl ace further away from the three f old but still within the same region. This would suggest that the r egion directly around the three fold may be crucial for assembly while the surrounding area may play a role in genome organization and perhaps uncoating. This is reflected with our own mutan ts since the only mutant to show a defect in infectivity (P419A) i s located distal from the three fold relative to the remaining three residues. It is unclear if the dAMP observed by Nam et al was a residue belonging to the assembly or genomespecific DNA content of the capsid and this it is not clear if the pHdependent mechanism was one that would result in the disruption of capsidgenome interactions or one that would affect the stability and conformation of the capsid, or both. As previously noted (37) the inner surface of the capsid appear s to show the most discrepancies between crystal and solutionbased cryo electron structures suggesting that further mutagenic and functional analysis of the region is necessary to dissect these two mechanisms and their relationship to one another.


53 Summary In summary, our findings demonstrate that the pH quartet is critical for infection. Mutants in this region have exhibited defects in cell entry, nuclear trafficking and uncoating. Even more importantly our findings point to a new role for the capsid in t he nucleus, which functions after uncoating and secondstrand synthesis and cannot be complimented in trans This data implies a previous ly undescribed role for the AAV capsid in transcription, splicing or RNA transport. We have also demonstrated that mut 22 exhibits a >1 log defect in cell entry, but that the majority of this mutants >8 log infectivity defect is accounted for by events occurring after nuclear entry. Finally our data also shows that P419, within the DNA group, plays a role in AAV infection and that the cluster as a whole is critical for virus assembly This data provides further evidence that nonspecific DNA facilitates the assembly of the AAV capsid.


54 Figure 3 1 Location and Structure of the pH quartet and DNA group. A) Two neighbori ng pH quartets surround the twofold axis of symmetry. B) The pH quartet is comprised of R389, Y704, E563 and H526. In addition to E563, the remaining residues from mut40 (D561, E562, E564) are located in close proximity. The four residues of the pH quartet are contributed by three different VP monomers (shown in yellow, green and cyan) C) Three DNA groups are seen located around the threefold axis of symmetry on the inner surface of the AAV2 capsid. D) At neutral pH, P419 and P630 equivalents in AAV8 are se en base stacking with an adenine residue (not shown) while H629 interacts with the DNA backbone.


55 Table 31 Alignment of amino acid residues that contribute to the pH quartet and surrounding acidic clusters in AAV1 to 10 AAV 389 526 528531 561564 704 1 R H DDED D E E E Y 2 R H DDEE D E E E Y 3 R H DDEE D E E E Y 4 R A PADS S E E E Y 5 R N LQGS S E S E Y 6 R H DDKD D E E E Y 7 R H DDED N E E E F 8 R H DDEE S E E E Y 9 R H EGED N E E E Y 10 R H DDEE S E E E Y F igure 32 Mutant virus production. A) B1 western blots show total viral protein from clarified cell lysates. 60g total protein was loaded into each well. B) A20 dot blot for intact capsids. 60g total protein from clarified lysates was loaded into each well. Top row compares cell only and known amounts of pure wt virus. + indicates a positive control with 1x1010 particles of pure AAV2 capsid in both panels. Cell only used to indicate antibody background reactivity to VP free cell lysates.


56 Figure 3 3 Particle to infectivity ratios of pH quartet mutants. All infections were done in HeLa cells at rAAV inputs ranging from 11x106 viral genomes per cell, in the presence of Ad5 (MOI =10). GFP positive cells were counted 48 hrs pi, and particle to infectivity ratios were calculated by div iding the input number of particles by the total number of GFP positive cells. All infections were done in triplicate. Y704A produced a defect that was beyond the detection limit of our assay ( > 8 logs worst than WT).


57 Figure 34 A20 confocal microscopy of pH quartet mutants. Intact capsids were stained A20 monoclonal antibody and detected with Alexa 488 anti mouse antibody (green). The nucleus was defined by Lamin B1 staining (Alexa 647; red). Panels display single Z slices through the midsect ion of the nucleus. Ad+ samples were cinfected with Ad5 (10 MOI) while Adwere infected with rAAV only. Wt virus displayed robust nuclear entry. Both mutants shows a marked lack of nuclear A20 signal, despite A20 staining in the cytoplasm. Figure 35 Confirming nuclear and cytoplasmic fraction purity. Fractions were checked for purity by western blotting against GAPDH (cytoplasmic marker) and Lamin B1 (nuclear marker). Scans of western blot films indicated >90% fraction purity. Each mutant was done in triplicate. N = nuclear fraction; C: cytoplasmic fraction


58 Figure 36 Cellular entry of WT and pH quartet mutant genomes. Pulse infections were done in 6well dishes with wt, E563A and Y704A in presence of Ad5 (MOI=10). Cells were collected at each time point, fractionated into cytoplasmic and nuclear fractions. Viral DNA was then isolated and quantitated by real time PCR. The total measure of viral DNA from all fractions from each sample is shown above. AAV input was 1x104 viral genomes per cell. W e were able to detect nearly 100% of input wt virus at 8hrs pi. Y704A display a 3fold decrease in entry while E563A had a 10fold defect in cell entry as compared to wt.


59 Figure 37 Subcellular distribution of pH quartet genomes. Time course infecti ons were done with wt, E563A and Y704A in presence of Ad5 (MOI=10). Cells were collected at each time point, fractionated into cytoplasmic and nuclear fractions. Viral DNA was then isolated and quantitated by real time PCR. AAV input was 1x104 viral genomes per cell. A) Shows percent of virus in the cell that localized to the nucleus at 8 and 16 hrs pi. B),C), D) show viral genomes in cytoplasm and nucleus at all time points for wt, E563A and Y704A respectively. All infections at all time points were done i n triplicate.


60 Figure 38 Percentage of uncoated nuclear genomes. Cell were infected with wt or mutant rAAV at 1x104 viral genomes per cell in 6well plates and conifected with Ad5 (MOI = 10). Cell were collected 16hrs pi and fractionated into nuc lear and cytoplasmic fractions. The percentage of DNAsesensitive (uncoated) genomes was determined by splitting nuclear fractions in half, and treating one half with benzonase prior to viral DNA isolation in order to remove any uncoated (DNAsesensitive) genomes. The other half was used as a total DNA control. We were able to back calculate the average number of uncoated genomes per nucleus for each virus and this is indicated above each bar. Each infection was done in triplicate.


61 Figure 39 Particle to infectivity ratios of self complimentary pH quartet and mut40 mutants. All infections were done in HeLa cells at rAAV inputs ranging from 11x106 viral genomes per cell, in the presence of Ad5 (MOI =10). GFP positive cells were counted 48 hrs pi, and particle to infectivity ratios were calculated by dividing the input number of particles by the total number of GFP positive cells. All infections were done in triplicate. Viruses packaged with self complimentary genomes are indicated with SC. Al l other viruses were packaged with singlestranded pTR UF11. The following viruses produced defects that were beyond our detection limits (> 8 log worst than WT): Y704A (single stranded), E564 (singlestranded), mut40 (with either genome).


62 Figure 3 10. Wild type capsids do not compliment Y704A. We attempted in trans complementation of the Y704A mutant packaged with a singlestranded (ss) GFP expression cassette (pTR UF11). Complementation was attempted with wt virus packaged with an ss DNA RFP gen ome as well as WT VLPs. WT* = wild type only at 1X104 vg/cell. For wt GFP and RFP virus, infections were done separately. Complementation was attempted with either an excess of WT virus (1x104 WT vg/cell : 1x102 Y704A vg/cell) or an excess of the mutant vi rus (1x104 WT vg/cell : 1x105 Y704A vg/cell). With an excess of WT virus we observed bleed through of the RFP signal in the GFP channel, but no actual GFP signal from the mutant virus. When cells were infected with an excess of mutant virus, WT RFP express ion was suppressed either by the mutant virus out competing the WT virus at the cell surface or by a potential dominant negative phenotype of the Y704A mutant. This resulted in the loss of fluorescent signal in both the RFP and GFP channels.


63 Figure 311. Mut22 particle to infectivity ratio. All infections were done in HeLa cells at rAAV inputs ranging from 11x106 viral genomes per cell, in the presence of Ad5 (MOI =10). GFP positive cells were counted 48 hrs pi, and particle to infectivity ratios we re calculated by dividing the input number of particles by the total number of GFP positive cells. All infections were done in triplicate. Virus packaged with self complimentary genomes is indicated with SC while the other virus was packaged with single stranded pTR UF11. In both cases, the mutant produced a defect that was beyond our detection limit (> 8 logs worst than WT).


64 Figure 312. Mut22 A20 confocal microscopy. Intact capsids were stained A20 monoclonal antibody and detected with Alexa 488 anti mouse antibody (green). The nucleus was defined by Lamin B1 staining (Alexa 647; red). Panels display single Z slices through the midsection of the nucleus. Ad+ samples were c o infected with Ad5 (10 MOI) while Adwere infected with rAAV only. Wt virus displayed robust nuclear entry and GFP expression in the presence of Ad5. Mut22 shows a total lack of nuclear A20 signal, despite A20 staining in the cytoplasm.


65 Figure 313. Subcellular distribution of wildtype and mut22 genomes. Time course infections were done with wt, and mut22 in presence of Ad5 (MOI=10) as previous described in FIG36. Cells were collected at each time point, fractionated into cytoplasmic and nuclear fractions. Viral DNA was then isolated and quantitated by r eal time PCR. A) and B) show the distribution of viral genomes in the cell at different time points for wt and mut22 respectively. Figure 314. DNA group particle to infectivity ratio. All infections were done in HeLa cells at rAAV inputs ranging from 1 1x106 viral genomes per cell, in the presence of Ad5 (MOI =10). GFP positive cells were counted 48 hrs pi, and particle to infectivity ratios were calculated by dividing the input number of particles by the total number of GFP positive cells. All infec tions were done in triplicate.


66 CHAPTER 4 PROTEASE ACTIVITY OF THE AAV CAPSID Previous studies have suggested that AAV undergoes proteolytic processing within the endosomal compartments (2), which might allow for structural transitions that are necessary for endosomal escape, nuclear trafficking and uncoating The region surrounding the AAV pH q uartet encompasses mutants previously identified as critical for infection (71, 122, 1 30) which contain several highl y conserved (Figure 4 1 and Table 3 1) acidic clusters including 562EEE564 and 528DDEE531. Structural analysis of this region reveals similarities to an ac id protease active site prompting the question of whether or not the AAV capsid exhibits protease activity Detailed here, is our examination of AAVs ability to undergo pH dependent autolysis as well as its ability to cleave external protein substrates. This activity is conserv ed across the major AAV serotypes, with cleavage patterns that segregate by serotype sequence similarity. Through the use of a variety of biochemical tools, we have identified and characterized some of the auto lytic site s in AAVs 1 and 9. The putative acti ve sites for these respective cleavages were confirmed by mutational analysis. While it is not clear if AAV does also require cellular proteases, we now show that the capsid has intrinsic protease activity unlike that seen in other viruses. Results Protease Activity Within the AAV Capsid To see if AAV undergoes pH induced protease cleavage, a purified preparation of AAV2 was incubated for 60 min at pH 7.4 or pH 5.5, and the molecular mass of the three caps id proteins was examined by i m munoblotting with B1 antibody, a capsid antibody that recognizes an epitope near the C termi nal end of the capsid protein (120)


67 Virus that was treated at pH 7.4 showed no change in the integrity of the three capsid proteins c ompared to untreated virus (Figure 4 2; AAV2). However, virus that was treated at pH 5.5 showed a decrease of the two larger capsid proteins, VP1 and VP2, as well as the appearance of several new lower molecular mass viral bands or an increase in their intensity. Treatment of AAV1 at low pH produced similar changes in t he B1 immunoblot pattern (Figure 4 2 ; AAV1), namely, a decrease in VP1 and 2 and an increase in sev eral lower molecular mass bands. Two of the cleavage bands that wer e produced when AAV1 was incubated at pH 5.5, ~ 17 kDa and ~ 55 kDa, were isolated by gel electrophoresis and subjected to MS analysis to identify the approximate positions of the cleavage sites (see Figure 4 3 ). The 17kDa ba nd generated peptides that clus t ered near the C terminus of the capsid protein, as expected, but also generated peptides located in the middle of the capsid and at the N terminus. The same was seen with the 55kDa band. This suggested that both the 17 kDa and 55 kDa bands contained mul ti ple capsid cleavage products, and this was consistent with at least two and probably more proteolytic cleavage sites in the capsid. To check for the possibility that a cellular protease might be contaminating the virus prepara tions, the untreated AAV1 preparations were also analyzed by MS. No proteins other than the three AAV capsid proteins were consistently found in the virus stocks. Trypsin, which had been added to cleave the capsid prot eins during mass spectroscopy, was the only protease found, but prev ious mapping of the trypsin sites on AAV capsids (115) was not consistent with the cleavage patterns seen following pH 5.5 autolysis The effect of pH 5.5 on the other AAV serotypes was similar ( Figure 4 4 ). In


68 all cases, there was a significant decrease of VP1 and VP2 and the appearance of additional lower molecular mass bands (or an increase in their intensity) when the viruses were incubated at pH 5.5. Comparison of the cleavage patterns seen in the different serotypes suggested that there were several different cleavage patterns as well as multiple cleavage sites in each serotype. For example, AAV1, 2, and 3 produced an approximately 17kDa Cterminal peptide, while AAV8, 9, and 10 produced an ~ 10 kDa Ct erminal band. Taken together, it appeared that all AAV serotypes were capable of some autolytic activity and that this activity w as activated by low pH. To see if AAVs can also cleave an external substrate, the PDQ colorimetric protease assay was us ed. I ncubation of AAV1, 2, 5, and 8 at a 30 nM capsid concentration with the protease substrate showed robust protease activity for each. The protease activity in the capsids was comparable (within 2fold) to that seen with 30 nM trypsin (Figure 4 5 A). Howev er, unlike autolytic cleavage, which was activated by low pH, cleavage of the external substrate was only observed at pH 7.5 and was largely abolished at pH 5.5. In addition, cleavage of the PDQ substrate by all serotypes was abolished by the addit ion of protease inhibitors (Figure 4 5 C). In contrast, protease inhibitors did not seem to have an effect on autolytic cleavage at pH5 (Figure 46). The possibility that the capsids might be contaminated by a cellular protease that copurified wi th capsid was also tested. Con trol cell extracts that had been purified the same way as capsid preparations were negative for protease activity (Figure 4 5 B; mock infected). In addition, a portion of the AAV8 stock was filtered through a concentrator (Amicon) that would retain the 3x 106kDa capsid but permit a 100kDa soluble protease to flow through. When both the material that passed through the filter and the capsids were


69 tested for prot ease activity, the capsid frac tion retained al l of the protease activity (Figure 4 5 D) To see if E563, which had been identified as a pH sensitive amino acid by X ray crystallography in AAV8, was essential for external protease activity, it was mutated to an alanine in AAV2. The E563A mutation completely abolished protease activit y on an external substrate (Figure 4 5 B). However, when we tested the ability of this mutant to carry out autolytic cleavage of the capsid, there appeared to be no change in its ability t o cleave capsid proteins (Figure 4 7 ). In addition, mutants with mutations in amino acids that flanked the E563 position and were completely conserved in all AAV serotypes, E562A and E564A, also appeared to be capable of autolytic cleavage. A fourth mutant, E561A, as well as a previously isolated mutant (122) that contained all four of these aci dic resi dues mutated to alanine (Mut40; E561 to 564A), al so shared this phenotype (Figure 4 7 ). Taken together, these findings suggested a possible model in which the pH quartet region might not itself be a protease active site but instead might be acting li ke a pH sensi tive switch that induces a structural change in the capsid. This in turn activates protease sites throughout the capsid or represses them. Determination of Autolytic Cleavage Sites To identify specific autolytic cleavage sites, two serotypes w ere chosen for Edman sequencing; AAV1 an d 9. These seroty pes were chosen because both were available in large quantity and produced pH induced cleavage fragments that could be separated from con taminating fragments. Table 4 2 shows that the apparent 10kDa band produced by AAV9 was the result of cleavage between amino acids (aa) 657 and 658 and has a calculated molecular mass of 9.3 kDa. Similarly, the apparent 17kDa band from AAV1 came from a cleavage between aa 590 and 591 to produce a fragment that


70 has a calculated molecular mass of 16.3 kDa. Finally, a 40kDa band from AAV1 was cleaved between aa 219 and 220. Its molecular mass was ap parently the result of cleavage at both the 219 and 590 positions, thus producing the apparent 40kDa band isolated from the gel. Identification of putative active site amino acids. The posi tion 657 cleavage site in AAV9 is part of the socalled HI loop, a loop that surrounds the 5fold pore and extends from one asym metric unit over a neighbor ing unit in the pentamer (Fig ure 4 8 A). The site is surrounded by several amino acids that could participate in acidinduced cleavage, including D657, S669, Y674, and T676 (Figure 4 8 B). Each of these amino acids was individually mutated to alanine and tested for its ability to cleave at position 657 to pro duce the diagnostic 10kDa Cterminal fragment. Mutation o f D657 completely abolished production of the 10kDa fra gment (Figure 4 8 C). Mutation of S669 and T676 did not affect cleavage at 657; however, these mutants appeared to be unable to produce a fragment of approximately 30 kDa that was present in wild type AAV9. Finally, mutation of amino acid 674 produced capsi d proteins that could no longer assemble, consistent with its location at a 5fold interface. Such viruses typically degrade their monomer capsid protein, and, thus, no capsid protein was detected. The ability of D657A to transduce HEK293 cells when the vi rus con tained a GFP expression cassette was also tested. Although AAV9 does not efficiently infect cells in culture, no major difference between D657A and wildtype virus transduction efficiencies was detected when equal amounts of wildtype virus and mutant virus were compared (data not shown). A similar set of experiments was carried out to identify amino acids involved in the AAV1 cleavage at position 590, which is located on the capsid surface at the top of the


71 three protrusions surrounding the icosahedral 3 fold axes (Figure 4 9 A). Four amino acids in the vicinity of position 590 (aa 590, 593, 504, and 583) were changed to alanine resi dues (Figure 4 9 B). Of these, only the E590A mutation eliminated cleavage at the 590 position when the products of the c leavage reaction were immunoblotted with B1 antibody (Figure 4 9 C). The VP1 cleavage pattern is different from that of VP3. The cleavage products from wildtype AAV1 were also immune blotted with A1 antibody. This monoclonal antibody recognizes an epitope that is present only at the N terminus of VP1 (120) and would visualize only VP1related cleavage products. When this was done, approximately 60% of AAV1 VP1 was cleaved in 30 min, and three new A1 posit ive bands were produced or increased in intensity (Figure 4 10). Thus, in AAV1, there was a minimum of three pH induced cleavage sites within VP1. The prominent band at ~ 67 kDa was expecte d and is consis tent with cleavage at aa 590 (Table 4 1). (Note that fulllength VP1 is 81.4 kDa, so 81.4 16.3 kDa = 65.1 kDa.) However, a band at 23 kDa was also expected due to cleavage at aa 219, and this band was not detected from VP1 (Figure 4 10). Thus, cleavage at position 219 occurs only in VP3, and possibly VP2, but not in VP1. Finally, it should be noted that one of the bands present at pH 7.4 was not significantly increased in intensity at pH 5. 5 (Figure 4 10; ~ 50 kDa). This band, therefore, may be due to cleavage by a cellular protein, either d uring capsid asse mbly or during virus purification. Bands that did not respond to pH induction were also seen in B1 antibody immunoblots (Figure 4 2 and 44 )


72 Discussion and Conclusion We have presented evidence that the AAV capsid contains protease activity. Several lines of evidence support this conclusion. First, we published that purified VLPs of serotypes 1, 2, 5, and 8 were capable of cleaving an external protease substrate, and the activity was sensitive to protease inhibitors. Furthermore, a single substitution mu tation in AAV2, E563A, complet ely abolished AAV2 protease ac tivity on an external substrate. In addition, size fractionation of the AAV8 preparation to remove any potential low mole cular mass protease contaminants ( ~ 100 kDa) showed that protease activity copurified with the AAV capsid. We also demonstrated that all of the AAV serotypes tested (AAV1 to 3 and AAV5 to 10) were capable of autolytic cleavage at multiple sites within th e capsid. This activity was induced by incubation at pH 5.5, the approximate pH of late endosomes. When we mapped two of the cleavage sites in AAV1 ( Figure 4 11), one of them was at amino acid (aa) 219, a position that is located in the interior of the AAV 1 capsid (unpublished observation). Since it was unlikely that a contaminating cellular protease could enter the capsid through the 5fold pore, cleavage at aa 219 is likely to be autocatalytic. In addition, MS analysis of the AAV1 preparation failed to reveal a contaminating protease other than the one added to perform MS. It is also worth noting that the virus preparations used in this study were isolated using three different production and purification methods (baculov irus infection, herpesvirus in fecti on, and DNA transfection) and contained both full and empty capsids. Finally, as in most autocleavage reactions, one of the amino acids at the site of cleavage was essential for autolytic cleav age. Mutation of both AAV9 D657 and AAV1 D590 to alanine aboli shed cleavage at these posi tions. Taken together, our evi dence supports the conclusion that


73 the AAV capsid has intrinsic protease activity A curious aspect of our data was that protease activity on an external substrate and autolytic protease activity had different pH sensitivities. Autolytic protease activity was induced by pH 5.5 and was not active at pH 7.4. In contrast, protease activity on an external substrate was seen only at pH 7.5, not at pH 5.5. Moreover, external protease activity was completely eliminated by the E563A mutation in AAV2, an amino acid structurally equivalent to one that had previously been seen to undergo a structural change when AAV8 crystals were examined at pH 5.5 (14). In contrast, autolytic cleavage appeared not to be affecte d by this mutation. Autolytic cleavage was eliminated only by mutation of acidic residues at the sites of cleavage (amino acids 590 and 657). In addition, we note that some of the autolytic cleavage sites were already cleaved in some viral samples that had been prepared and purified at neutral pH. We interpret this to mean that cleavages can occur at neutral pH at a reduced rate and accumulate over time. The positions of these pH sensitive sites are illustrated in Figure 4 11 It is immediately apparent that all of the cleavages that were mapped (amino acids 219, 590, and 657) are well separated from the pH quartet region. Both D657 and D219 are near the 5fold pore, outside and inside the capsid, respectively, while D590 is located at the protrusions surrounding the 3fold axis. This begs the question is there one protease active site that participates in all of these cleavages or multiple active sites? Typically, acid protease active sites have two acidic amino acids that coordinate a water molecule that is the nucleophile. An intriguing possibility is that the N terminal sequences of VP1 or VP2, which are the only regions in the capsid that are not str ucturally constrained, may supply the second acidic residue that participates at


74 multiple sites. In this re gard, we note that when we examined the cleavage of VP1 using an antibody specific for this capsid protein, we did not see cleavage at residue 219. This would have produced a 23kDa Nterminal fragment from VP1; thus, cleavage at 219 apparently oc curs onl y in VP3, and perhaps VP2. As mentioned earlier, the pH quartet region contains amino acids from three symmetry related VP3s that are near the inter section of the 2 3 and 5fold axes ( Figure 4 11). It was identified as one of two sites in the AAV8 capsid that undergoes an ordered structural shift when viral cryst als are incubated at pH 5.5 (79) Mutations in the equivalent region of AAV2 are capable of assem bly and DNA packaging but have significant defects in their ability to transduce cells (59, 71, 86, 122) The nature of the defect has not been identified but appears to occur at a post entry step, in viral trafficking, uncoating, or gene expression (as shown in previous chapter) It has also been shown by several groups that exposure to the acidic environment of cellular endosomes is an essential step for viral infection (9, 29, 44) Taken together, these facts suggest that the pH quartet region and the induction of protease activity perform an essential function in the viral life cycle. One consequence of passage through the acidic endosomal compartment is the extrusion of the N terminus of VP1, which contains both nuclear localization sites and a phospholipase activ ity that are essential for infection (38, 40, 66) Although treatment of virus with acidic pH in vitro does not in itself induce exposure of the VP1 N terminus, passage through acidic endosomes appears to be essential (107) Thus, acidification (plus some other process that occurs during entry and has not yet been identified) is essential for VP1 extrusion and infectivity. W e note that the autonomous par vovirus


75 minute virus of mice (MVM) undergoes cleavage of its major capsid protein VP2 approximately 22 amino acids from its N terminus, either sometime after assembly or during cell entry. This cleavage is believed to be due to one or more cel lular proteases and is important for the subsequent extrusion of the VP1 N terminus (32) It is therefore tempting to speculate that the pH quartet region is a pH sensitive molecular switch that induces a structural change in the capsid, thereby exposing protease active sites within the capsid at several locations ( Figure 4 11; amino acids 219, 590, and 657). This in turn may facilitate extrusion of VP1 to the capsid surface. The identification of cleavage sites both external to the 5fold pore and immediately internal to the pore (Table 4 2 and Figure 4 11; amino acids 657 and 219) supports the possibility that the role of capsid cleavage may be to prepare the capsid for VP1 ex trusion. Alternatively, cleavage of the capsid at several locations may be essential to promote viral uncoating once the virus enters the nucleus. However, mutation of E563, at least i n the context of the AAV2 serotype, did not noticeably change the pH induction of autolytic protease activity ( Figure 4 7 ). Thus, if the pH quartet is indeed a molecular switch, additional cofactors provided by the cell may be required. An alternative model is that the pH quartet is itself a protease that is active only on an external substrate and the other cleavages that we have mapped (amino aci ds 219, 590, and 657) are inde pendent pH sensitive active sites. This possibility suggests that the role of the pH quartet region, which includes E563, is to cleave only an external substrate during entry (i.e., some cellular protein that must be cleaved for efficient infection). However, this model would then suggest that AAV particles contain multiple


76 protease ac tive sites that are independently controlled by pH. We are not aware of a precedent for this in other viruses.


77 Figure 4 1 A) A portion of the AAV2 crystal structure showing a reference capsid protein (blue) that is contained in a pentamer (gray), a trimer (orange), and a dimer (cyan). The pH quartet amino acids are shown in the red spacefilling form. The complete cluster is shown flanking the 2fold axis of symmetry (oval), where all three polypeptide chains contri buting to the cluster are pres ent. Smaller clusters are shown at other locations where only one or two of the participating capsid polypeptides are present. The oval, triangle, and pentamer indicate the 2, 3 and 5fold axes of symmetry, respectively. B) The positions of the pH quartet am ino acids in ball and stick form (Y704, E563, R389, and H526) in AAV2. These amino acids are potenti ally interacting through hydrogen bonding. In AAV8, the equivalent amino acids to Y704 and E563 chang e their positions at acidic pH (79) Also present are two highly conserved glutamates (E562 and E563) and a second charged cluster that contributes E531, all in stick form. Mutations in the two acidic clusters make normal amounts of packaged virus but are extremely defective for infectivity (59, 71, 86, 122)


78 Figure 4 2 AAV undergoes autolytic cleavage at pH 5.5. (Top) Approximately 100 fmol (1010 particles) of AAV2 was either untreated (UN) or incubated at pH 7.4 or pH 5.5 for 60 min at 37C and then electrophoresed on an SDS acryl amide gel and immunoblotted with B1 antibody, which recognizes a common epitope present at the C terminal end of all t hree capsid proteins. The 0.5 UN lane contains half the amount of virus shown in the UN lane and allows estimation of the loss of VP1 and 2 in the pH 5.5 lane. The positions of the VP1, 2, and 3 bands and several cleavage bands are also indicated to the left and right, respectively. (Bottom) AAV1 was treated at pH 7.4 or pH 5.5 for 60 min at 37C, electrophoresed on an SDS acrylamide gel, and immunoblotted with B1 antibody.


79 Figure 4 3 Mass spectrometry map of cleavage product peptides. Combined try psin and chymotrypsin peptides generated from the B1 positive ~17 (green) and ~55kDa (red) cleavage fragments mapped against the sequence of VP1. Pink bars show predicted cleavage sites for 17kDa and 55kDa. Figure 4 4 All of the AAV serotypes undergo autolytic cleavage. Serotypes 1 to 3 and 5 to 10 were incubated at pH 7.4 and 5.5 as described in the legend to Fig. 2 and immunoblotted with B1 antibody.


80 *Figure courtesy of Balasubramanian Venkatakrishnan, from (96) Figure 4 5 External protease activity of the AAV capsid. The AAV capsid can cleave an external protein substrate. A ) The PDQ colorimetric protease assay was used to compare the protease activity of AAV serotypes at pH 7.5 (open symbols) and pH 5.5 (closed symbols) to those of trypsin at pH 7.5 (red diamonds) and the noenzyme control (buffer; gray diamonds). AAV1, orange squares; AAV2, blue circles; AAV5, purple triangles; AAV8, green hexagons. B) The E563A mutant abolishes protease activity. The protease activities of wildtyp e (wt) AAV2 VLPs (circles) were compared to those of the E563A mutant (squares) and to a mock virus extract (gray triangles with dotted line). C ) Virus was incubated at pH 7.4 with PDQ protease substrate for 1 to 5 h in the presence (closed symbols) or abs ence (open symbols) of protease inhibitor (Halt protease inhibitor cocktail; see Materials and Methods) and compared to the noenzyme control (gray diamonds with dotted line). AAV1, orange squares; AAV2, blue circles; AAV5, purple tr iangles; AAV8, green he xagons. D) AAV8 was fractionated through a 100kDa filter (Amicon), and the flow through (low molecular mass [MM]) fraction and retained (virus) fractions were tested for protease activity.


81 Figure 46 Protease inhibitors have no effect on autolytic pH5 cleavage. WT AAV2 was incubated at 37 C for 1hr in pH7.4 (Lane 1) or pH5 (Lane 2). Virus was also incubated at pH5 in the presence of a protease inhibitor cocktail (Lane 3) to determine if autolytic cleavage could be inhibited. The addition of these inhibitors did not appear to influence pH dependent autolytic cleavage. Figure 47 Autocatalytic activity is preserved in the mut40 acid cluster mutants. The E563A mutation and mutations in flanking acidic amino acids do not abolish autolytic cleavage of AAV2. Wildtype (WT) and mutant virus preparations were incubated at pH 7.4 and 5.5 as described in the legend to Figure 4 2 and immunoblotted with B1 antibody. Mut40 contains all four of the other aci dic mutations.


82 Figure 48 AAV9 specific cleavage site mutants. A) AAV9 amino acids 657 and 658 (blue and red spacefilling molecules) are illustrated on the AAV9 pentamer (gray and red). The reference molecule (red subunit) illustrates that the cleav age site is located on the HI loop, which extends over the nearest neighbor capsid protein in the pent amer. B) The amino acids near the site of the 657/658 peptide bond (arrow) that were potentially involved in the position 657 cleavage reaction were subst ituted for with alanine residues and tested for autolytic cleavage activity at position 657. C) Wild type (WT) AAV9 and mutant AAV9 were incubated at pH 7.4 or 5.5 at 37C for 60 min and immunoblotted with B1 antibody.


83 Figure 49 AAV1 specific cleavage site mutants. A) AAV1 amino acids 590 and 591 (gold spacefilling molecules) are illustrated on the AA V1 trimer (gray). B) The amino acids near the site of the 590591 peptide bond (arrow) that were potentially involved in the position 590 cleavage were s ubstituted for with alanine residues and tested for autolytic clea vage activity at position 590. C) Wild type (WT) AAV1 and mutant AAV1 were incubated at pH 7.4 or 5.5 at 37C for 60 min and immunoblotted with B1 antibody.


84 Figure 410. VP1 specific cleavage. AAV1 was incubated at pH 7.4 or 5.5 for 60 min at 37C as described in the legend to Fig. 4 2 and then immunoblotted with A1 antibody, which recog nizes an epitope that is present only in VP1.


85 Figure 411. Map of identified cleavage sites. T he position of the pH quartet region (orange) is shown in spacefilling models on the gray wire background of the AAV9 crystal structure. The AAV9 position 657 cleavage site (blue), the AAV1 219 site (red), and the AAV1 590 (green) cleavage site are also s hown as space f illing models. The oval, pentamer, and triangle indicate the positions of the 2, 5 and 3fold axes, respectively.


86 Table 41 Edman N Terminal Sequences Cleavage site Confidence Cleavage sequence* C term fragment (kDa) Gel fragment (kDa) AAV1 590 100% TD/PATGDV 16.3 17 219 90% AD/GV G NAs 58.1 40 AAV9 657 90% AD/ PP TAFN 9.3 13 Lowercase indicates a mismatch with the published sequence, and underlining indicates that more than one amino acid was detected at the position.


87 CHAPTER 5 CONCLUSION AND FUTURE DIRECTIONS Conclusion Our work has demonstrated that the pH responsive region known as the pH q uartet is critical for AAV infection. Within this region we have shown, that the Y704A mutant is severely defective in infectivity despite being able to enter the cell, traffic to the nucleus and uncoat. B y a process of elimination, we have concluded that the phenotype of this mutant points to a new critical role for the AAV capsid in the nucleus, which functions after un coating and secondstrand synthesis. We were also able to demonstrate that the second pH sensitive region within the capsid, the DNA group, appears to play a role in viral assembly and may be involved in viral infectivity. Finally, we were able to co nfirm the severe nature of the m ut22 defect, and have shown that despite a modest decrease in cell entry, the mutant is block ed at a step after nuclear entry. The residues of the pH q uartet region have also led us to identify and characterize a novel protease activity within the AAV capsid. We were able to demonstrate capsid autolytic events in response to acidic pH. The autolytic activity produced serotype specific cleavage patterns, and we were able to identify three unique cleavage sites in AAVs 1 and 9. Usin g mutagenesis we re able to show that in two of the cases (AAV1 590, AAV9 657) t he aspartic acid residues adjacent to the cleavage sites were either involved in cleavage or constituted a part of a recognition site The work of our collaborator, Dr. Balasubramanian Venkatakrishnan has also indicated that the capsid has protease activity against external substrates at neutral pH and that E563A, one of the pH q uartet amino acids, abolished this activity. The same mutation, as well as other


88 mutants in the region, failed to abolish autolytic activity, and the two activities seem to have different pH optima, with neutral pH producing only external cleavage, while acidic pH produces only autolytic cleavage. This suggests that either the pH q uartet is an independent protease that functions separately from the one governing autolysis or that it is in fact a pH sensitive switch that is able to activate proteolytic sites elsewhere. During the course of this dissertation work we have demonstrated several new roles for t he AAV capsid. As a protease, the capsid is able to process itself in response to acidic pH, which maybe a prerequisite to uncoating or other structural transitions like the externalization of the N termini. This role was previously thought to be carried out by cellular enzymes but now appears to be encoded for in the capsid protein. Likewise the external protease function can be envis ioned playing a number of roles. These could include the cleavage of receptors post entry, to liberate the capsid, or the generation of signaling peptides through cleavage of cellular proteins. AAV could also use its external protease activity to enter the nucleus by degrading the nuclear lamina, as seen with MVM. The novel observation that the capsid is necessary in the nucleus after uncoating suggest that the capsid could be used to recruit cellular enzymes that are necessary for viral transcription, or even splicing. Given the protease activity of the capsid, it is not unreasonable to speculate that in addition to recruitment, the capsid could augment cellular proteins through proteolysis thus either selectively degrading certain members of multi protein complexes, such as the splice o some or by modifying individual proteins. Such modifications could allow for the preferential activity of these cellular enzymes on the viral genome instead of their usual cellular targets. A greater understanding of these


89 functions could lead to an enhancements of future generations o f AAV vectors and to a better understanding of the post entry barriers to infection that have been observed for some cell type and AAV serotype combinations. Future Directions Y704A and the Nucleus Further study of the Y704A mutant is needed, in order to isolate and characterize the new role for the AAV capsid that our current data has suggested. While it is clear that the mutant enters and uncoats in the nucleus, we must now determine if the DNA is transcribed and the resulting mRNAs spliced. Likewise, we have shown residues surrounding Y704 (namely E562, E564) to be critical for infection, and these too must be characterized to determine if the y play the same role as Y704 or if these highly conserved residues affect cell entry, nuclear trafficking or uncoating. We must also determine if mut22 is able to uncoat in the nucleus. This work is currently underway and should help determine if mut22 is an uncoating mutant as previously thought or if its defect is after uncoating as seen with Y704A AAV Protea se Although we have identified three cleavage sites, and shown two to have aspartic acids that are necessary for infection, we have yet to identify the active site(s) within the AAV capsid protease or the mechanism by which they function. In the future, we need to determine the exact nature of the protease domains responsible for both autolysis and external cleavage as well as if and how these domains are interrelated. This could be achieved both through mutagenesis and the use of specific protease inhibitors. We must also examine the possibility that the VP1 unique region may participate in autolysis thus linking several distant cleavage sites across the capsid. We must also


90 determine if other mutants within the pH q uartet are defective for external protease activity It also remains to be determined if the defect in E563A (and potentially other residues) is the result of disrupting a protease active site or the inactivation of a pH sensitive switch Finally, we need to determine the role of this newly discovered protease activity in AAV infection. Preliminary infections with the D657A AAV9 mutant revealed no apparent phenotype, suggesting that we may need to first identify and inactivate all protease site s, internal and external, prior to being able to determine how this new feature fits in the global picture of AAV biology.


91 LIST OF REFERENCES 1. Akache B Grimm D Pandey K Yant SR Xu H Kay MA 2006. The 37/67kilodalto n laminin receptor is a receptor for adenoassociated virus serotypes 8, 2, 3, and 9. Journal of Virology 80:9831 9836. 2. Akache B Grimm D Shen X Fuess S Yant SR Glazer DS Park J Kay MA 2007. A twohybrid screen identifies cathepsins B and L as uncoating factors for adenoassociated virus 2 and 8. Mol Ther 15 :330 339. 3. Asokan A Hamra JB Govindasamy L, Agbandje McKenna M Samulski RJ 2006. Adenoassociated virus type 2 contains an integrin alpha5beta1 binding domain essential for viral cell entry. Journal of Virology 80:8961 8969. 4. Asokan A Schaffer DV Samulski RJ 2012. The AAV vector toolkit: poised at the clinical crossroads. Mol Ther 20 :699 708. 5. ATCHISON RW CASTO BC HAMMON WM 1965. ADENOVIRUS ASSOCIATED DEFECTIVE VIRUS PARTICLES. Science 149 :754 756. 6. Ball Goodrich LJ Moir RD Tattersall P 1991. Parvoviral target cell specificity: acquisition of fibrotropism by a mutant of the lymphotropic strain of minute virus of mice involves multiple amino acid substitutions within the capsid. Virology 184:175 186. 7. Ball Goodrich LJ Tattersall P 1992. Two amino acid substitutions within the capsid are coordinately required for acquisition of fibrotropism by the lymphotropic strain of minute virus of mice. Journal of Virology 66:3415 3423. 8. Bantel Schaal U Hub B Kartenbeck J 2002. Endocytos is of adeno associated virus type 5 leads to accumulation of virus particles in the Golgi compartment. Journal of Virology 76:2340 2349. 9. Bartlett JS Wilcher R Samulski RJ. 2000. Infectious entry pathway of adenoassociated virus and adenoassociated virus vectors. Journal of Virology 74:2777 2785. 10. Benmerah A Bayrou M Cerf Bensussan N Dautry Varsat A 1999. Inhibition of clathrincoated pit assembly by an Eps15 mutant. Journal of Cell Science 112 ( Pt 9) :1303 1311. 11. Berns KI Bohenzky RA 1 987. Adenoassociated viruses: an update. Adv. Virus Res. 32 :243 306. 12. Berns KI Kotin RM Labow MA 1988. Regulation of adenoassociated virus DNA replication. Biochim. Biophys. Acta 951 :425 429.


92 13. Bevington JM Needham PG Verrill KC Collaco RF Basrur V Trempe JP 2007. Adenoassociated virus interactions with B23/Nucleophosmin: identification of subnucleolar virion regions. Virology 357:102 113. 14. Bleker 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. Journal of Virology 79:2528 2540. 15. Brown K Gerstber ger S Carlson L, Franzoso G Siebenlist U 1995. Control of I kappa B alpha proteolysis by sitespecific, signal induced phosphorylation. Science 267 :1485 1488. 16. Brown WJ Goodhouse J Farquhar MG 1986. Mannose6 phosphate receptors for lysosomal enz ymes cycle between the Golgi complex and endosomes. J Cell Biol 103 :1235 1247. 17. Buller RM Janik JE Sebring ED Rose JA 1981. Herpes simplex virus types 1 and 2 completely help adenovirus associated virus replication. Journal of Virology 40:241 247. 18. Buller RM Rose JA 1978. Characterization of adenovirus associated virus induced polypeptides in KB cells. Journal of Virology 25:331 338. 19. Carter BJ Antoni BA Klessig DF 1992. Adenovirus containing a deletion of the early region 2A gene allow s growth of adenoassociated virus with decreased efficiency. Virology 191:473 476. 20. Cenciarelli C Wilhelm KG Guo A Weissman AM 1996. T cell antigen receptor ubiquitination is a consequence of receptor mediated tyrosine kinase activation. J Biol Chem 271 :8709 8713. 21. Chau V Tobias JW Bachmair A Marriott D Ecker DJ, Gonda DK Varshavsky A 1989. A multiubiquitin chain is confined to specific lysine in a targeted short lived protein. Science 243 :1576 1583. 22. Chejanovsky N Carter BJ 1989. Mutagenesis of an AUG codon in the adenoassociated virus rep gene: effects on viral DNA replication. Virology 173:120 128. 23. Choi SY Kim YH Lee YK Kim KT 2001. Chlorpromazine inhibits storeoperated calcium entry and subsequent noradrenaline secretion in PC12 cells. Br. J. Pharmacol. 132 :411 418. 24. Cohen S Behzad AR Carroll JB Pant N 200 6. Parvoviral nuclear import: bypassing the host nuclear transport machinery. J Gen Virol 87 :3209 3213.


93 25. Cohen S Pant N 2005. Pushing the envelope: microinjection of Minute virus of mice into Xenopus oocytes causes damage to the nuclear envelope. J G en Virol 86:3243 3252. 26. Di Pasquale G Davidson BL, Stein CS Martins I Scudiero D Monks A Chiorini JA 2003. Identification of PDGFR as a receptor for AAV 5 transduction. Nat Med 9 :1306 1312. 27. Ding W Zhang L, Yan Z Engelhardt JF. 2005. Intrac ellular trafficking of adenoassociated viral vectors. Gene Ther 12 :873 880. 28. Ding W Zhang LN Yeaman C Engelhardt JF. 2006. rAAV2 traffics through both the late and the recycling endosomes in a dosedependent fashion. Mol Ther 13 :671 682. 29. Douar AM Poulard K Stockholm D Danos O 2001. Intracellular trafficking of adenoassociated virus vectors: routing to the late endosomal compartment and proteasome degradation. Journal of Virology 75:1824 1833. 30. Duan D Li Q Kao AW Yue Y, Pessin JE Engelhardt JF. 1999. Dynamin is required for recombinant adenoassociated virus type 2 infection. Journal of Virology 73:10371 10376. 31. Duan D Yue Y, Yan Z Yang J Engelhardt JF. 2000. Endosomal processing limits gene transfer to polarized airway epit helia by adenoassociated virus. J. Clin. Invest. 105 :1573 1587. 32. Farr GA Cotmore SF Tattersall P 2006. VP2 cleavage and the leucine 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. Journal of Virology 80:161 171. 33. Farr GA Zhang LG Tattersall P 2005. Parvoviral virions deploy a capsidtethered lipolytic enzyme to breach the endosomal membrane during cell entry. Proc Natl Acad Sci USA 102 :17148 171 53. 34. Ferrari FK Samulski T Shenk T Samulski RJ 1996. Secondstrand synthesis is a ratelimiting step for efficient transduction by recombinant adenoassociated virus vectors. Journal of Virology 70:3227 3234. 35. 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 USA 99:11854 11859.


94 36. Gao G Vandenberghe LH Alvira MR Lu Y Calcedo R Zhou X Wilson JM 2004. Clades of Adeno associated viruses are widely disseminated in human tissues. Journal of Virology 78:6381 6388. 37. Gerlach B Kleinschmidt JA Bttcher B 2011. Conformational changes in adenoassociated virus type 1 induced by genome packaging. J Mol Biol 409:427 438. 38. Girod A Wobus CE Zdori Z, Ried M Leike K Tijssen P Kleinschmidt JA Hallek 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. 39. Govindasamy L, Padron E McKenna R Muzyczka N Kaludov N Chiorini JA Agbandje McKenna M 2006. Structurally mapping the diverse phenotype of adeno associated virus serotype 4. Journal of Virology 80:11556 11570. 40. Grieger JC Johnson JS Gurda Whitaker B Agbandje McKenna M Samulski RJ 2007. Surfaceexposed adeno associated virus Vp1NLS capsid fusion protein rescues infectivity of noninfectious wildtype Vp2/Vp3 and Vp3only capsids but not that of fivefold pore mutant virions. Journal of Virology 81:7833 7843. 41. Grieger JC Snowdy S Samulski RJ 2006. Separate basic region motifs within the adenoassociated virus capsid proteins are essential for infectivity and assembly. Journal of Virology 80:5199 5210. 42. Gurda BL, Raupp C Popa Wagner R Naumer M Olson NH Ng R, McKenna R Baker TS Kleinschmidt JA Agbandje McKenna M 2012. Mapping a neutralizing epitope onto the capsid of adenoassociated virus serotype 8. Journal of Virology 86:7739 7751. 43. 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. 44. Hansen J Qing K Srivastava A 2001. Adenoassociated viru s type 2 mediated gene transfer: altered endocytic processing enhances transduction efficiency in murine fibroblasts. Journal of Virology 75:4080 4090. 45. Hansen J Qing K Srivastava A 2001. Infection of purified nuclei by adenoassociated virus 2. Mol Ther 4 :289 296.


95 46. Hauck B Zhao W High K Xiao W 2004. Intracellular viral processing, not single stranded DNA accumulation, is crucial for recombinant adeno associated virus transduction. Journal of Virology 78:13678 13686. 47. Hirst J Futter CE Hopkins CR 1998. The kinetics of mannose 6phosphate receptor trafficking in the endocytic pathway in HEp2 cells: the receptor enters and rapidly leaves multivesicular endosomes without accumulating in a prelysosomal compartment. Mol Biol Cell 9 :809 816. 48. Ihalainen TO Niskanen EA Jylhv J Paloheimo O Dross N Smolander H Langowski J Timonen J VihinenRanta M 2009. Parvovirus induced alterations in nuclear architecture and dynamics. PLoS ONE 4 :e5948. 49. Im DS Muzyczka N 1990. The AAV origin binding protein Rep68 is an ATP dependent sitespecific endonuclease with DNA helicase activity. Cell 61:447 457. 50. Johnson JS Gentzsch M Zhang L, Ribeiro CMP Kantor B Kafri T Pickles RJ Samulski RJ. 2011. AAV Exploits Subcellular Stress Associated with Inflammation, Endoplasmic Reticulum Expansion, and Misfolded Proteins in Models of Cystic Fibrosis. PLoS Pathog 7 :e1002053. 51. Johnson JS Li C DiPrimio N Weinberg MS McCown TJ Samulski RJ 2010. Mutagenesis of adenoassociated virus type 2 capsid protein VP1 uncovers new roles for basic amino acids in trafficking and cell specific transduction. Journal of Virology 84:8888 8902. 52. Johnson JS Samulski RJ 2009. Enhancement of adenoassociated virus infection by mobiliz ing capsids into and out of the nucleolus. Journal of Virology 83:2632 2644. 53. Ju X D Lou S Q Wang W G Peng J Q Tian H 2004. Effect of hydroxyurea and etoposide on transduction of human bone marrow mesenchymal stem and progenitor cell by adenoasso ciated virus vectors. Acta Pharmacol Sin 25 :196 202. 54. 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. Journal of Virology 75:6884 6893. 55. Kashiwakura Y Tamayose 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. Journal of Virology 79:609 614.


96 56. Keiser NW Yan Z Zhang Y Lei Butters DCM Engelhardt JF. 2011. Unique characteristics of AAV1, 2, and 5 viral entr y, intracellular trafficking, and nuclear import define transduction efficiency in HeLa cells. Human Gene Therapy 22:1433 1444. 57. Kelkar S De BP Gao G Wilson JM Crystal RG Leopold PL. 2006. A common mechanism for cytoplasmic dyneindependent microt ubule binding shared among adenoassociated virus and adenovirus serotypes. Journal of Virology 80:7781 7785. 58. 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. 59. Kern A Schmidt K Leder C Mller OJ Wobus CE Bettinger K Lieth Von der CW King JA Kleinschmidt JA 2003. Identification of a heparinbinding motif on adenoassociated virus type 2 capsids. Journal of Virology 77:11072 11081. 60. Kerr JR 2006. Parvoviruses. A Hodder Arnold Publication. 61. King JA Dubielzig R Grimm D Kleinschmidt JA 2001. DNA helicase mediated packaging of adenoassociated virus type 2 genomes into preformed capsids. EMBO J 20:3282 3291. 62. Kobiler O Drayman N Butin Israeli V Oppenheim A 2012. Virus strategies for passing the nuclear envelope barrier. Nucleus 3 63. Kotin RM Linden RM Berns KI 1992. Characterization of a preferred site on human chromosome 19q for integration of adenoassociated virus DNA by nonhomologous recombination. EMBO J 11:5071 5078. 64. Kotin RM Menninger JC Ward DC Berns KI 1991. Mapping and direct visualization of a regionspecific viral DNA integration site on chromosome 19q13qter. Genomics 10:831 834. 65. Kotin RM Siniscalco M Samulski RJ Zhu XD Hunter L, Laughlin CA McLaughlin S Muzyczka N Rocchi M Berns KI 1990. Site specific integration by adenoassociated virus. Proc Natl Acad Sci USA 87 :2211 2215. 66. Kronenberg S Bttcher B Lieth von der CW Bleker S Kleinschmidt JA 2005. A conformational change in the adenoassociated virus type 2 capsid leads to the exposure of hidden VP1 N termini. Journal of Virology 79:5296 5303.


97 67. Laughlin CA Jones N Carter BJ 1982. Effect of deletions in adenovirus early region 1 genes upon replication of adenoassociated virus. Journal of Virology 41:868 876. 68. Lerch TF, Xie Q Chapma n MS 2010. The structure of adenoassociated virus serotype 3B (AAV 3B): insights into receptor binding and immune evasion. Virology 403:26 36. 69. Lindstrm MS 2011. NPM1/B23: A Multifunctional Chaperone in Ribosome Biogenesis and Chromatin Remodeling. Biochem Res Int 2011 :195209. 70. Liu Y Joo K I Wang P 2012. Endocytic processing of adenoassociated virus type 8 vectors for transduction of target cells. G ene Ther. 71. 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. Journal of Virology 80:8 21 834. 72. Mah C Qing K Khuntirat B Ponnazhagan S Wang XS Kube DM Yoder MC Srivastava A 1998. Adenoassociated virus type 2mediated gene transfer: role of epidermal growth factor receptor protein tyrosine kinase in transgene expression. Journal of Virology 72:9835 9843. 73. Mani B Baltzer C Valle N Almendral JM Kempf C Ros C. 2006. L ow 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. Journal of Virology 80:1015 1024. 74. Mcc arty DM Fu H Monahan PE Toulson CE Naik P Samulski RJ 2003. Adenoassociated virus terminal repeat (TR) mutant generates self complementary vectors to overcome the rate limiting step to transduction in vivo. Gene Ther 10:2112 2118. 75. McLaughlin SK Collis P Hermonat PL, Muzyczka N 1988. Adenoassociated virus general transduction vectors: analysis of proviral structures. Journal of Virology 62:1963 1973. 76. 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.


98 77. 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. Journal of Virology 74:1761 1766. 78. Muzyczka N 1992. Use of adenoassociated virus as a general transduction vector for mammalian cells. Cur r. Top. Microbiol. Immunol. 158 :97 129. 79. Nam HJ Gurda BL, McKenna R Potter M Byrne B Salganik M Muzyczka N Agbandje McKenna M 2011. Structural studies of adenoassociated virus serotype 8 capsid transitions associated with endosomal trafficking Journal of Virology 85:11791 11799. 80. Nam HJ Lane MD Padron E Gurda B McKenna R Kohlbrenner E Aslanidi G, Byrne B Muzyczka N Zolotukhin S Agbandje McKenna M 2007. Structure of adenoassociated virus serotype 8, a gene therapy vector. Journal of Virology 81:12260 12271. 81. Nayak R Pintel DJ 2007. Positive and negative effects of adenovirus type 5 helper functions on adenoassociated virus type 5 (AAV5) protein accumulation govern AAV5 virus production. Journal of Virology 81:2205 2212. 82. Nesvizhskii AI Keller A Kolker E Aebersold R 2003. A statistical model for identifying proteins by tandem mass spectrometry. Anal Chem 75:4646 4658. 83. Ng R, Govindasamy L, Gurda BL, McKenna R Kozyreva OG Samulski RJ Parent KN Baker TS Agbandje McKenna M 2010. Structural characterization of the dual glycan binding adenoassociated virus serotype 6. Journal of Virology 84:12945 12957. 84. Nonnenmacher M Weber T 2011. Adenoassociated virus 2 infection requires endocytosis through the CLIC/GEEC pathway. Cell Host Microbe 10:563 576. 85. Ogston P Raj K Beard P 2000. Productive replication of adenoassociated virus can occur in human papillomavirus type 16 (HP V 16) episomecontaining keratinocytes and is augmented by the HPV 16 E2 protein. Journal of Virology 74:3494 3504. 86. 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. Journal of Virology 77:6995 7006. 87. Pant N Kann M 2002. Nuclear pore complex is able to transport macromolecules with diameters of about 39 nm. Mol Biol Cell 13 :425 434.


99 88. Paolini R Serra A Molfetta R Piccoli M Frati L Santoni A 1999. Tyrosine kinase dependent ubiquitination of CD16 zeta subunit in human NK cells following receptor engagement. Eur. J. Immunol. 29 :3179 3187. 89. Penrose KJ Garcia Alai M de Prat Gay G McBride AA 2004. Casein Kinase II phosphorylationinduced conformational switch triggers degradation of the papillomavirus E2 protein. J Biol Chem 279 :22430 22439. 90. Popa Wagner R Porwal M Kann M Reuss M Weimer M Florin L, Klein schmidt JA 2012. Impact of VP1Specific Protein Sequence Motifs on AdenoAssociated Virus Type 2 Intracellular Trafficking and Nuclear Entry. Journal of Virology 86:9163 9174. 91. Qing K Hansen J, Weigel Kelley KA Tan M Zhou S Srivastava A 2001. Ade no associated virus type 2mediated gene transfer: role of cellular FKBP52 protein in transgene expression. Journal of Virology 75:8968 8976. 92. Qing K Mah C Hansen J, Zhou S Dwarki V Srivastava A 1999. Human fibroblast growth factor receptor 1 is a co receptor for infection by adenoassociated virus 2. Nat Med 5 :71 77. 93. Quinn CO Kitchingman GR 1986. Functional analysis of the adenovirus type 5 DNA binding protein: sitedirected mutants which are defective for adenoassociated virus helper acti vity. Journal of Virology 60:653 661. 94. Richardson WD Westphal H 1984. Requirement for either early region 1a or early region 1b adenovirus gene products in the helper effect for adenoassociated virus. Journal of Virology 51:404 410. 95. Ros C, Kemp f C 2004. The ubiquitinproteasome machinery is essential for nuclear translocation of incoming minute virus of mice. Virology 324:350 360. 96. Salganik M Venkatakrishnan B Bennett A Lins B Yarbrough J Muzyczka N Agbandje McKenna M McKenna R 2012. Evidence for pH Dependent Protease Activity in the Adenoassociated Virus Capsid. Journal of Virology. 97. Samulski RJ Chang LS Shenk T 1987. A recombinant plasmid from which an infectious adenoassociated virus genome can be excised in vitro and its use to study viral replication. Journal of Virology 61:3096 3101. 98. Samulski RJ Shenk T 1988. Adenovirus E1B 55Mr polypeptide facilitates timely cytoplasmic accumulation of adenoassociated virus mRNAs. Journal of Virology 62:206 210.

PAGE 100

100 99. Samulski RJ Srivastava A Berns KI Muzyczka N 1983. Rescue of adenoassociated virus from recombinant plasmids: gene correction within the terminal repeats of AAV. Cell 33 :135 143. 100. Samulski RJ Zhu X Xiao X Brook JD Housman DE Epstein N Hunte r LA 1991. Targeted integration of adenoassociated virus (AAV) into human chromosome 19. EMBO J 10 :3941 3950. 101. 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. Journal of Virology 74:9184 9196. 102. Schelhaas M Malmstrm J Pelkmans L Haugstetter J, Ellgaard L Grnewald K Helenius A 2007. Simian Virus 40 depends on ER protein folding and quality control factors for entry into host cells. Cell 131 :516 529. 103. 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. Journal of Virology 82:1399 1406. 104. 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. 105. Shen S Bryant KD Sun J Brown SM Troupes A Pulicherla N Asokan A 2012. Glycan binding avidity determines the systemic fate of adenoassociated virus type 9. Journal of Virology 86:10408 10417. 106. Snyder RO Im DS Muzyczka N 1990. Evidence for covalent attachment of the adenoassociated virus (AAV) rep protein to the ends of the AAV genome. Journal of Virology 64:6204 6213. 107. Sonntag F, Bleker S Leuchs B Fischer R Kleinschmidt JA 20 06. Adenoassociated virus type 2 capsids with externalized VP1/VP2 trafficking domains are generated prior to passage through the cytoplasm and are maintained until uncoating occurs in the nucleus. Journal of Virology 80:11040 11054. 108. Sonntag F, Schm idt K Kleinschmidt JA 2010. A viral assembly factor promotes AAV2 capsid formation in the nucleolus. Proceedings of the National Academy of Sciences 107:10220 10225. 109. 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.

PAGE 101

101 110. Suikkanen S Antila M Jaatinen A VihinenRanta M Vuento M 2003. Release of canine parvovirus f rom endocytic vesicles. Virology 316:267 280. 111. Summerford C Bartlett JS Samulski RJ 1999. AlphaVbeta5 integrin: a coreceptor for adenoassociated virus type 2 infection. Nat Med 5 :78 82. 112. Summerford C Samulski RJ. 1998. Membrane associated heparan sulfate proteoglycan is a receptor for adenoassociated virus type 2 virions. Journal of Virology 72:1438 1445. 113. Thomas DL, Wang L, Niamke J, Liu J Kang W Scotti MM, Ye G J Veres G Knop DR 2009. Scalable recombinant adeno associated virus production using recombinant herpes simplex virus type 1 coinfection of suspensionadapted mammalian cells. Human Gene Therapy 20:861 870. 114. Uhrig S Coutelle O Wiehe T Perabo L, Hallek M Bning H 2012. Successful target cell transduction of capsidengineered rAAV vectors requires clathrin dependent endocytosis. Gene Ther 19 :210 218. 115. Van Vliet K Blouin V Agbandje McKenna M Snyder RO 2006. Proteolytic mapping of the adenoassociated virus capsid. Mol Ther 14:809 821. 116. 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. 117. Wang LH Rothberg KG Ande rson RG 1993. Mis assembly of clathrin lattices on endosomes reveals a regulatory switch for coated pit formation. J Cell Biol 123:1107 1117. 118. Warrington KH Gorbatyuk OS Harrison JK Opie SR Zolotukhin S Muzyczka N 2004. Adenoassociated virus t ype 2 VP2 capsid protein is nonessential and can tolerate large peptide insertions at its N terminus. Journal of Virology 78:6595 6609. 119. Wiethoff CM, Wodrich H Gerace L Nemerow GR 2005. Adenovirus protein VI mediates membrane disruption following c apsid disassembly. Journal of Virology 79:1992 2000. 120. Wobus CE Hgle Drr B Girod A Petersen G Hallek M Kleinschmidt JA 2000. Monoclonal antibodies against the adenoassociated virus type 2 (AAV 2) capsid: epitope mapping and identification of c apsid domains involved in AAV 2 cell interaction and neutralization of AAV 2 infection. Journal of Virology 74:9281 9293.

PAGE 102

102 121. Wu J Zhao W Zhong L, Han Z Li B Ma W Weigel Kelley KA Warrington KH, Srivastava A 2007. Self complementary recombinant adenoassociated viral vectors: packaging capacity and the role of rep proteins in vector purity. Human Gene Therapy 18:171 182. 122. Wu P Xiao W Conlon T, Hughes J Agbandje Mckenna M Ferkol T, Flotte T Muzyczka N 2000. Mutational analysis of the adenoassociated virus type 2 (AAV2) capsid gene and construction of AAV2 vectors with altered tropism. Journal of Virology 74:8635 8647. 123. Wu Z, Asokan A Grieger JC Govindasamy L, Agbandje McKenna M Samulski RJ 2006. Single amino acid changes can influence titer, heparin binding, and tissue tropism in different adeno associated virus serotypes. Journal of Virology 80:11393 11397. 124. 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. Journal of Virology 80:9093 9103. 125. Xiao P J Samulski RJ 2012. Cytoplasmic trafficking, endosomal escape, and p erinuclear accumulation of adenoassociated virus type 2 particles are facilitated by microtubule network. Journal of Virology 86:10462 10473. 126. Xiao W Warrington KH Hearing P Hughes J Muzyczka N 2002. Adenovirus facilitated nuclear translocation of adenoassociated virus type 2. Journal of Virology 76:11505 11517. 127. Xiao X Li J Samulski RJ 1996. Efficient long term gene transfer into muscle tissue of immunocompetent mice by adenoassociated virus vector. Journal of Virology 70:8098 8108. 1 28. Yan Z Zak R Luxton GWG Ritchie TC Bantel Schaal U Engelhardt JF. 2002. Ubiquitination of both adenoassociated virus type 2 and 5 capsid proteins affects the transduction efficiency of recombinant vectors. Journal of Virology 76:2043 2053. 129. Zhong L, Li B Jayandharan G Mah CS Govindasamy L Agbandje McKenna M Herzog RW Weigel Van Aken KA Hobbs JA Zolotukhin S Muzyczka N Srivastava A 2008. Tyrosinephosphorylation of AAV2 vectors and its consequences on viral intracellular trafficking and transgene expression. Virology 381:194 202.

PAGE 103

103 130. Zhong L, Li B Mah CS Govindasamy L, Agbandje McKenna M Cooper M Herzog RW Zolotukhin I Warrington KH Weigel Van Aken KA Hobbs JA Zolotukhin S Muzyczka N Srivastava A 2008. Next generati on of adenoassociated virus 2 vectors: point mutations in tyrosines lead to highefficiency transduction at lower doses. Proceedings of the National Academy of Sciences 105:7827 7832. 131. Zolotukhin S Byrne BJ Mason E Zolotukhin I Potter M, Chesnut K Summerford C Samulski RJ, Muzyczka N 1999. Recombinant adenoassociated virus purification using novel methods improves infectious titer and yield. Gene Ther 6 :973 985. 132. Zolotukhin S Potter M, Zolotukhin I Sakai Y Loiler S Fraites TJ, Chiodo VA Phillipsberg T, Muzyczka N Hauswirth WW Flotte TR Byrne BJ Snyder RO 2002. Production and purification of serotype 1, 2, and 5 recombinant adenoassociated viral vectors. Methods 28 :158 167.

PAGE 104

104 BIOGRAPHICAL SKETCH Max Salganik was born July of 1982, in Kiev, Ukraine. He graduated with Bachelor of Science in biotechnology from Cook College at Rutgers, The State University of New Jersey in 2005. After completing his first year in the Interdisciplinary Program for Biomedical Sciences, Max joined the lab of Dr. Nicholas Muzyczka in May of 2006. He received his Ph.D. from the University of Florida in the fall of 2012.