UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Study of the minute virus of mice cell entry Garcin, Pierre 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_november_garcin_pierre.pdf [ 73.53MB ]
Metadata
JSON: 24-1.0135542.json
JSON-LD: 24-1.0135542-ld.json
RDF/XML (Pretty): 24-1.0135542-rdf.xml
RDF/JSON: 24-1.0135542-rdf.json
Turtle: 24-1.0135542-turtle.txt
N-Triples: 24-1.0135542-rdf-ntriples.txt
Original Record: 24-1.0135542-source.json
Full Text
24-1.0135542-fulltext.txt
Citation
24-1.0135542.ris

Full Text

 STUDY OF THE MINUTE VIRUS OF MICE CELL ENTRY  by  Pierre Garcin    M.Sc., University Bordeaux Victor Segalen, 2009    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2014  © Pierre Garcin, 2014     ii Abstract  The minute virus of mice prototype (MVMp) is a non-enveloped single stranded DNA virus of the family Parvoviridae. MVMp is one of the smallest viruses and shows intriguing abilities to preferentially infect and kill cancer cells (oncotropism/oncolytism), suggesting a potential for MVMp as an anti-cancer agent. Unfortunately, there is a lack of knowledge of the early events of MVMp infection cycle, such as binding to the cell surface and subsequent endocytosis. In an attempt to identify cellular partners of MVMp infection, our lab performed a mass spectrometry analysis of MVMp potential binding partners. Following this analysis, the galactose-binding lectin (galectin) 3 (Gal-3) was identified as binding partner for MVMp. Given the involvement of this extra-cellular matrix protein in the clustering and endocytosis of cell surface receptors, and its up-regulation in various aggressive tumor cells, I hypothesized that Gal-3 could play a role in MVMp cell entry, and potentially in its oncotropism. Using siRNA knockdown of Gal-3 in different cells followed by immunofluorescence microscopy analysis, I found that Gal-3 is necessary for an efficient MVMp cell entry and infection in different cells. Moreover, I discovered that the Golgi enzyme β1,6-acetylglucosaminyltransferase 5 (Mgat5), whose role is the addition of complex N-glycosylation to various cell surface receptors for Gal-3 binding, is required for MVMp infection. I also found that cancer cells with higher Gal-3 expression are more susceptible to MVMp infection than cells with lower Gal-3 levels. Next I used a combination of flow cytometry, immuno-fluorescence and transmission electron microscopy to characterize the early events of MVMp infection in various tissue-culture cell lines. My results show that many crucial parameters of the mesenchymal cell migration process regulate MVMp cellular entry and infection. I found that MVMp relies on cell  iii protrusions to cluster at the leading edge of migrating cells rapidly after binding to the plasma membrane, from where it is subsequently endocytosed. Moreover, transmission electron microscopy analysis revealed that MVMp uses various endocytic pathways, which was confirmed using drug inhibitors of endocytosis. Finally, I found that epithelial-mesenchymal transition, an inducer of cancer cell migration, triggers MVMp infection in highly dividing non-permissive cancer cells.                     iv Preface  The Results in Chapter 3 dealing with siRNA of Gal-3 in LA9 cells (Fig. 3.1 to 3.4) have been published as part of a peer-reviewed journal:  • Garcin, P., Cohen, S., Terpstra S., Kelly, I., Foster L.J., and Panté, N. (2012) Proteomic Analysis Identifies a Novel Function for Galectin 3 in the Cell Entry of Parvovirus. Journal of Proteomics. 79:123-32.  In this article, Dr. Sarah Cohen and Sanne Terpstra performed the immunoprecipitation experiments. The mass spectrometry analysis and statistic analysis were performed by Dr. Leonard Foster and Isabelle Kelly. I performed all the fluorescence and electron microscopy analysis, which are presented in Chapter 3.  A manuscript with the rest of the results in Chapter 3 is almost ready to be submitted to a peer-reviewed journal: • Garcin, P.O., Nabi, I.R., and Panté, N. (2014). Galectin-3 Plays a Role in the Minute Virus of Mice Infection. To be submitted to a peer-reviewed journal.  Chapter 4 has been accepted for publication in Virology: • Garcin, P.O. and Panté, N. (2014). Cell Migration is another Player of the Minute Virus of Mice Infection.  Virology. In press.   v Chapter 5 will be submitted to a peer-reviewed journal: • Garcin, P.O. and Panté, N. (2014). The Minute Virus of Mice Exploits Different Endocytic Pathways for Cellular Entry. To be submitted to a peer-reviewed journal.  With the guidance of my supervisor Dr. Nelly Panté, I designed and performed all experiments, quantified, and analyzed all data presented in this thesis. The research presented in this thesis was approved by the UBC Bio-Safety Committee (Certificate B10-0057).         vi Table of Contents 	  Abstract	  .....................................................................................................................................................	  ii	  Preface	  .....................................................................................................................................................	  iv	  Table	  of	  Contents	  ..................................................................................................................................	  vi	  List	  of	  Tables	  ..........................................................................................................................................	  ix	  List	  of	  Figures	  ..........................................................................................................................................	  x	  List	  of	  Abbreviations	  .........................................................................................................................	  xii	  Acknowledgements	  ...........................................................................................................................	  xvi	  Chapter	  1:	  Introduction	  .......................................................................................................................	  1	  1.1	  The	  MVMp	  infection	  cycle	  ......................................................................................................................	  1	  1.1.1	  Structure	  and	  composition	  of	  the	  MVMp	  ....................................................................................................	  1	  1.1.2	  MVMp	  early	  infection	  steps	  ..............................................................................................................................	  3	  1.1.3	  MVMp	  replication	  and	  cellular	  egress	  .........................................................................................................	  7	  1.2	  Overview	  of	  viral	  oncotropism	  ............................................................................................................	  9	  1.2.1	  Definition	  of	  oncotropic/oncolytic	  viruses	  ................................................................................................	  9	  1.2.2	  Oncolytic	  virotherapy	  ......................................................................................................................................	  10	  1.2.3	  Parameters	  involved	  in	  viral	  oncotropism	  .............................................................................................	  11	  1.2.4	  What	  makes	  a	  good	  oncotropic/oncolytic	  virus	  ..................................................................................	  12	  1.2.5	  Parvoviral	  oncotropism	  ..................................................................................................................................	  14	  1.2.6	  Determinants	  of	  parvoviral	  oncotropism	  ...............................................................................................	  16	  1.3	  Overview	  of	  the	  mechanism	  of	  cell	  migration	  ..............................................................................	  20	  1.3.1	  The	  basis	  of	  cell	  migration	  .............................................................................................................................	  20	  1.3.2	  Cell	  adhesion	  to	  the	  extracellular	  matrix	  ................................................................................................	  21	  1.3.3	  Galectins	  .................................................................................................................................................................	  23	  1.3.4	  Signaling	  pathways	  involved	  in	  cell	  migration	  .....................................................................................	  24	  1.3.5	  Plasma	  membrane	  and	  focal	  adhesion	  turnover	  during	  cell	  migration	  .....................................	  27	  1.4	  Cell	  migration	  during	  tumorigenesis	  ...............................................................................................	  30	  1.4.1	  Epithelial-­‐mesenchymal	  transition	  ............................................................................................................	  33	  1.4.2	  Different	  types	  of	  cancer	  cell	  migration	  ...................................................................................................	  34	  1.5	  Research	  objectives	  ...............................................................................................................................	  36	  1.5.1	  Aim	  1:	  To	  assess	  the	  involvement	  of	  Gal-­‐3	  and	  Mgat5	  in	  the	  MVMp	  early	  infection	  ...........	  37	  1.5.2	  Aim	  2:	  To	  screen	  various	  cancer	  cells	  for	  permissivity	  to	  MVMp	  infection	  .............................	  38	  1.5.3	  Aim	  3:	  To	  determine	  whether	  or	  not	  cell	  migration	  plays	  a	  role	  in	  MVMp	  infection	  ..........	  38	  1.5.4	  Aim	  4:	  To	  identify	  the	  MVMp	  endocytosis	  pathways	  .........................................................................	  39	  Chapter	  2:	  Materials	  and	  Methods	  ................................................................................................	  40	  2.1	  Antibodies,	  reagents	  and	  cell	  lines	  ...................................................................................................	  40	  2.2	  Cell	  culture	  and	  virus	  purification	  ....................................................................................................	  41	  2.2.1	  Cell	  culture	  ............................................................................................................................................................	  41	  2.2.2.	  MVMp	  purification	  ...........................................................................................................................................	  41	  2.3	  Fluorescence	  microscopy	  .....................................................................................................................	  43	   vii 2.3.1	  Cell	  preparation	  for	  fluorescence	  microscopy	  ......................................................................................	  43	  2.3.2	  Confocal	  microscopy	  and	  image	  analysis	  ................................................................................................	  43	  2.4	  MVMp	  assays	  ............................................................................................................................................	  44	  2.4.1	  MVMp	  binding	  assay	  by	  immunofluorescence	  microscopy	  ............................................................	  44	  2.4.2	  MVMp	  clustering	  assay	  by	  immunofluorescence	  microscopy	  .......................................................	  44	  2.4.3	  MVMp	  uptake	  assay	  by	  immunofluorescence	  microscopy	  ..............................................................	  45	  2.4.4	  MVMp	  infection	  assay	  ......................................................................................................................................	  46	  2.5	  Gal-­‐3	  siRNA	  transfection	  ......................................................................................................................	  46	  2.6	  Electron	  microscopy	  ..............................................................................................................................	  47	  2.7	  Flow	  cytometry	  ........................................................................................................................................	  48	  2.8	  Plaque	  assay	  .............................................................................................................................................	  49	  2.9	  Induction	  of	  epithelial-­‐mesenchymal	  transition	  .........................................................................	  50	  2.10	  Western	  blot	  ..........................................................................................................................................	  50	  2.11	  Statistical	  analysis	  ...............................................................................................................................	  51	  Chapter	  3:	  Involvement	  of	  Galectin	  3	  and	  Mgat5	  in	  MVMp	  Infection	  ...............................	  53	  3.1	  Introduction	  .............................................................................................................................................	  53	  3.2	  Results	  ........................................................................................................................................................	  54	  3.2.1	  Galectin	  3	  siRNA	  knockdown	  hampers	  MVMp	  infection	  in	  LA9	  cells	  .........................................	  54	  3.2.2	  Galectin	  3	  siRNA	  knockdown	  hampers	  MVMp	  uptake	  in	  LA9	  cells	  .............................................	  57	  3.2.3	  Galectin	  3	  siRNA	  knockdown	  does	  not	  affect	  MVMp	  binding	  to	  LA9	  cells	  ...............................	  59	  3.2.4	  Galectin	  3	  knockdown	  hampers	  MVMp	  infection	  in	  PyMT	  cells	  ...................................................	  62	  3.2.5	  Galectin	  3	  knockdown	  hampers	  MVMp	  uptake	  in	  PyMT	  cells	  .......................................................	  62	  3.2.6	  MVMp	  does	  not	  infect	  Mgat5-­‐/-­‐	  cells	  ..........................................................................................................	  65	  3.2.7	  MVMp	  uptake	  is	  reduced	  in	  Mgat5-­‐/-­‐	  cells	  ...............................................................................................	  71	  3.2.8	  MVMp	  cell	  surface	  binding	  is	  not	  affected	  in	  Mgat5-­‐/-­‐	  cells	  .............................................................	  71	  3.2.9	  Correlation	  between	  Gal-­‐3	  expression	  and	  MVMp	  infection	  ..........................................................	  74	  3.3	  Discussion	  .................................................................................................................................................	  82	  Chapter	  4:	  Cell	  Migration	  is	  another	  Player	  of	  the	  MVMp	  Infection	  ..................................	  88	  4.1	  Introduction	  .............................................................................................................................................	  88	  4.2	  Results	  ........................................................................................................................................................	  89	  4.2.1	  MVMp	  infection	  begins	  with	  the	  clustering	  of	  viral	  particles	  at	  the	  base	  of	  filopodia	  .........	  89	  4.2.2	  MVMp	  accumulates	  at	  the	  leading	  edge	  of	  migrating	  cells,	  which	  are	  more	  susceptible	  to	  viral	  uptake	  ......................................................................................................................................................................	  96	  4.2.3	  MVMp	  infection	  increases	  with	  cell	  migration	  .....................................................................................	  99	  4.2.6	  Epithelial-­‐mesenchymal	  transition	  triggers	  MVMp	  infection	  ......................................................	  107	  4.3	  Discussion	  ..............................................................................................................................................	  112	  Chapter	  5:	  MVMp	  Cell	  Entry	  Mechanisms	  .................................................................................	  117	  5.1	  Introduction	  ..........................................................................................................................................	  117	  5.2	  Results	  .....................................................................................................................................................	  118	  5.2.1	  MVMp	  cellular	  entry	  occurs	  in	  proximity	  to	  focal	  adhesions	  .......................................................	  118	  5.2.2	  MVMp	  can	  use	  a	  variety	  of	  endocytic	  pathways	  .................................................................................	  120	  5.3	  Discussion	  ..............................................................................................................................................	  126	  Chapter	  6:	  General	  Discussion	  and	  Future	  Perspectives	  ....................................................	  129	  6.1	  A	  key	  role	  for	  Gal-­‐3	  and	  Mgat5	  in	  MVMp	  infection	  ...................................................................	  130	  6.2	  Involvement	  of	  mesenchymal	  cell	  migration	  in	  MVMp	  infection	  ........................................	  134	  6.3	  MVMp	  cell	  entry	  ...................................................................................................................................	  137	  6.4	  A	  new	  model	  for	  the	  MVMp	  early	  infection	  .................................................................................	  139	  6.5	  Future	  directions	  .................................................................................................................................	  141	   viii 6.5.1	  In	  vivo	  analysis	  of	  MVMp	  oncotropism	  ...................................................................................................	  141	  6.5.2	  Investigating	  the	  Gal-­‐3	  involvement	  in	  MVMp	  infection	  ................................................................	  142	  6.5.3	  Investigating	  the	  Gal-­‐3-­‐BP	  involvement	  in	  MVMp	  infection	  .........................................................	  143	  6.5.4	  Identification	  of	  the	  MVMp	  receptor(s)	  .................................................................................................	  143	  6.6	  Concluding	  remarks	  ............................................................................................................................	  144	  References	  ..........................................................................................................................................	  145	                                      ix List of Tables  Table 1. Names and details on the cancer cells used in the screening…………………………...76    x List of Figures  Figure 1.1. Endocytic pathways used by viruses ………………………………………….......….6 Figure 1.2. Summary of MVM replication cycle …….………………………..……….….….…..8 Figure 1.3. The type 1 interferon antiviral response ………………………………..……….…..19 Figure 1.4. Integrin-mediated cell adhesion to the ECM ………………………………..........…22 Figure 1.5. Schematic diagram of RAS signaling pathways involved in cell migration……..….26 Figure 1.6. The process of FA disassembly……………………………………………………...28 Figure 1.7. Integrin turnover occurs at the leading edge of a migrating cell…………………….29 Figure 1.8. Involvement of Gal-3 and Gal-3-BP in FA stability………………………………...32 Figure 1.9. Different types of cell migration…………………………………………….………35 Figure 3.1. MVMp infectivity is reduced in Gal-3 siRNA knockdown LA9 cells………………56 Figure 3.2. MVMp cellular uptake is reduced in Gal-3 siRNA knockdown LA9 cells………....58 Figure 3.3. Binding of MVMp to the cell surface is not affected in Gal-3 siRNA knockdown LA9 cells………………………….………………………………………………………………60 Figure 3.4. Early events of MVMp binding to its receptor………………………………………61 Figure 3.5. MVMp infectivity is reduced in Gal-3 siRNA knockdown PyMT cells………….…63 Figure 3.6. MVMp cellular uptake is reduced in Gal-3 siRNA knockdown PyMT cells.............64 Figure 3.7. Detection of Mgat5 expression in different cells……………………………………67 Figure 3.8. MVMp infectivity is abolished in Mgat5-/- cells as shown by IF microscopy………68 Figure 3.9. MVMp infectivity is abolished in Mgat5-/- cells as shown by Western blot and plaque assay……………………………………………………………………………………………....69 Figure 3.10. Mgat5-/- cells proliferate more rapidly than PyMT and Mgat5-rescued cells….…..70 Figure 3.11. MVMp cellular uptake is reduced in Mgat5-/- cells………………………………...72 Figure 3.12. MVMp binding to the cell surface is not affect in Mgat5-/- cells……………….….73 Figure 3.13. Gal-3 expression profile in various cancer cells……………………………...……77 Figure 3.14. Correlation between Gal-3 expression and MVMp infection in cancer cells…...…78 Figure 3.15. Cell proliferation analysis………………………………………………………….79 Figure 3.16. No correlation between Gal-3 expression and MVMp infection in LN18 and LN229 cells……………………………………………………………………………………………….80 Figure 3.17. Cell proliferation analysis………………………………………………………….81  xi Figure 3.18. Model of the Gal-3 and Gal-3-BP involvement in MVMp early infection………...86 Figure 4.1. Electron micrographs of MVMp clustering at filopodia…………………………….91 Figure 4.2. Immunogold labeling of MVMp after the clustering assay…………………………92 Figure 4.3. Time course of MVMp clustering at the plasma membrane……………………..….94 Figure 4.4. MVMp clustering at the cell surface is dependent on actin polymerization………...95 Figure 4.5. MVMp clusters at the leading edge of migrating cells………………………....…...97 Figure 4.6. MVMp uptake is increased in migrating cells……………………………………....98 Figure 4.7. Effect of poly-K and FN substrates on LA9 and PyMT cell migration….……....100 Figure 4.8. MVMp infection increases with cell migration…………………………………….101 Figure 4.9. Effect of poly-K and FN substrates on LA9 and PyMT cell proliferation…………102 Figure 4.10. FACS analysis of MVMp cell binding and uptake on different substrates……….105 Figure 4.11. FN matrix increases MVMp accumulation at the nuclear periphery in PyMT cells…………………………………………………………………………………………...…106 Figure 4.12. EMT triggers MVMp infection in EpRas cells……………………………….…..109 Figure 4.13. EMT permits MVMp replication in EpRas cells…………………………………110 Figure 4.14. Cell proliferation analysis………………………………………………………...111 Figure 5.1. MVMp cellular entry occurs in proximity to focal adhesions……………………..119 Figure 5.2. IF analysis of MVMp endocytosis…………………………………………………122 Figure 5.3. MVMp can use various endocytic pathways as shown by EM…………………….123 Figure 5.4. MVMp can use various endocytic pathways as shown by IF microscopy…………124 Figure 5.5. MVMp can use various endocytic pathways as shown by FACS………………….125 Figure 6.1. Model for the Gal-3 and MVMp binding to an hypothetical receptor……………..133 Figure 6.2. Model for the MVMp early infection steps……………………………………...…140              xii List of Abbreviations  AAV: adeno-associated virus Ad: adenovirus bafA1: bafilomycin A1 BPV: bovine parvovirus BSA: bovine serum albumin  ˚C: degree Celsius  CD: cluster of differentiation CLICs: clathrin-independent carriers CME: clathrin-mediated endocytosis CO2: carbon dioxide  CPV: canine parvovirus CPZ: chlorpromazine Crm1: cellular chromosome maintenance protein 1 CtxB: cholera toxin b subunit CytoB: cytochalasin B DMEM: Dulbecco’s modified Eagle medium  DMSO: dimethyl sulfoxide DNA: deoxyribonucleic acid  dsDNA: double-stranded DNA E-Cad: epithelial cadherin ECM: extra-cellular matrix EDTA: Ethylenediaminetetraacetic acid EGFR: epidermal growth factor receptor EM: electron microscopy  EMT: epithelial-mesenchymal transition ER: endoplasmic reticulum  EtOH: ethanol FA: focal adhesion FACS: fluorescence-activated cell sorting  xiii FBS: fetal bovine serum  FITC: fluorescein isothiocyanate FN: fibronectin FP: filopodia FPV: feline parvovirus Gal: galectin Gal-3-BP: galectin-3-binding protein GFP: green fluorescent protein GPI-AP: glycosylphosphatidyl-inositol anchored protein GTP: guanine triphosphate H1-PV: H1 parvovirus HIV: human immunodeficiency virus HPV: human papillomavirus H-Ras: Harvey-Ras HRP: horseradish peroxidase HSV: herpes simplex virus  IF: immunofluorescence IHNV: infectious hematopoietic necrosis virus IM: infection medium IRF: interferon regulatory factor ISRE: IFN-stimulated response element kbp: kilobase pairs  KD: knock-down kDa: kilodalton  L-PHA: Phaseolus vulgaris leucoagglutinin MAPK: mitogen-activated protein kinase MCM: mesenchymal cell migration MDCK: Madin-Darby canine kidney MDa: megadalton  ME: microenvironment  Mgat5: β1,6-acetylglucosaminyltransferase 5  xiv MHC: major histocompatibility complex MMP: matrix metalloproteinase MOI: multiplicity of infection  mRNA: messenger ribonucleic acid  MV: measles virus MVMi: minute virus of mice immuno-suppressive strain MVMp: minute virus of mice prototype strain N-Cad: neural cadherin NLS: nuclear localization signal ns: not significant NS: non-structural protein nt: nucleotide OAS: 2',5'-oligoadenylate synthetase Pax: paxillin PBS: phosphate-buffered saline PE: Phyco-erythrin PFA: paraformaldehyde Pfu: plaque-forming unit pH: potential hydrogen  PKR: protein kinase R PLA2: phospholipase A2 Poly-K: poly-l-lysine PPV: porcine parvovirus PyMT: Polyoma virus middle T antigen rAAV: recombinant adeno-associated virus RGD: arginine-glycine-aspartate RNA: ribonucleic acid  RPMI: Roswell Park Memorial Institute medium RT: room temperature  SARS-Cov: SARS coronavirus SD: standard deviation SDS: sodium dodecyl sulfate   xv siRNA: small-interference RNA SOS: son of sevenless SR: serine/arginine  Src: sarcoma oncogene ssDNA: single-stranded DNA STAT: signal transducer and activator of transcription SV40: simian virus 40 TE: Tris-EDTA buffer  TEM: transmission electron microscopy  TGF-β: transforming growth factor beta VP: viral structural protein VSV: vesicular stomatitis virus wt: wild type            xvi Acknowledgements  First I would like to thank my supervisor, Dr. Nelly Panté, for all of her support and guidance throughout this thesis. Thank you for your patience, and for your contributions to the preparation of our manuscripts and this thesis. I would also like to thank my supervisory committee, Drs. Vanessa Auld, Ivan R. Nabi, and Wayne Vogl for their insights and helpful feedback about this project. I especially thank Dr. Wayne Vogl for considerable help with my electron microscopy analysis, and Dr. Ivan R. Nabi for generously sharing his cell lines and reagents, and for helpful discussions.  I greatly thank Dr. Peter Tattersall for the supply of MVMp capsid and NS1 antibodies, which played a crucial role throughout this thesis. I also thank Dr. Ivan R. Nabi (University of British Columbia) for the Mgat5-rescued, T238, TPC1, MCF7, MDA231, LNCap, and PC3 cells, as well as Dr. Christian Naus (University of British Columbia) for the LN18 and LN229 cell lines. I thank Drs. James W. Dennis (University of Toronto) and Ivan R. Nabi (University of British Columbia) for the Mgat5-wt and Mgat5-/- PyMT-transformed mouse epithelial mammary tumor cell lines, as well as Drs. Ernst Reichmann (Zurich University Children's Hospital) and Calvin Roskelley (University of British Columbia) for the EpRas cells. I thank Tak Kwong Poon (University of British Columbia) for help with my epithelial-mesenchymal transition experiments. I thank Dr. Pascal St. Pierre (University of British Columbia) for helpful advice with my confocal microscopy analysis, and for helpful discussions. I thank Dr. Cécile Boscher (University of British Columbia) for help with my galectin 3 siRNA experiments and for helpful discussions. I thank Dr. Bharat Joshi, Dr. Jay Shankar, Ray Fanrui, and Annie Aftab (University of British Columbia) for helpful discussions. I thank Jenny L. Huang for her help of proofreading this thesis.  I thank the members of the Dr. Nelly Panté laboratory for welcoming me to their lab. I thank Dr. Sarah Cohen for help with the virus preparation, Dr. Shelly Au for help with the writing of this thesis and fellowships, as well as Wei Wu for helpful discussions and Maria Acevedo for her constant entertainment. I also thank Lixin Zhou and Andy Johnson for help with my flow cytometry analysis. I am grateful for the financial support of the Zoology Department for Zoology Graduate Fellowships received during the last two years of my doctoral degree, and for operating grants to my supervisor from NSERC and CIHR.  xvii  I thank Dr. Michael Kann (Universite Victor Segalen Bordeaux2) for encouraging me to come to Vancouver for my PhD thesis, as well as all those who contributed to the great training that I benefited during my M.Sc. degree at the Universite Victor Segalen Bordeaux2. Finally, I would like to thank my mother Frederique Gueritte for teaching me the definition of DNA when I was 3 years old, and for stimulating my curiosity since the beginning. I also thank all those of my family who have supported me financially and made this accomplishment possible, as well as all my close friends (they will recognize themselves) who encouraged me all along and accepted me for who I am.                  1 Chapter 1: Introduction  The minute virus of mice (MVM) is one of the smallest viruses and shows intriguing oncotropic/oncolytic properties, suggesting a great potential for this virus as an anti-cancer agent. Unfortunately, the early events of MVM infection cycle, such as binding to the cell surface and endocytosis, are still poorly documented. Hence, I investigate these early events of MVM infection as the main objective of my thesis. A complete understanding of the infection cycle of this virus, and in particular of its underexplored early steps, could help us to assess whether these could play a role in the MVM ability to infect cancer cells preferentially. Since many aggressive tumor cells exhibit highly migrating and invasive phenotypes, which rely greatly on the endocytosis of cell surface receptors, it is reasonable to hypothesize that some oncotropic viruses may highjack these mechanisms to enter cancer cells more efficiently. Hence, I also investigated whether MVM infection is associated with host cell motility. In this chapter, I describe the MVM infection cycle, but also give a general overview of the viral oncotropism phenomenon and the mechanism of cell migration, and review various aspects of tumorigenesis with regard to the main theme of this thesis.  1.1 The MVMp infection cycle 1.1.1 Structure and composition of the MVMp The minute virus of mice prototype strain (MVMp) belongs to the family Parvoviridae, which includes many other strains such as the rat parvoviruses H1 and LuIII, the canine parvovirus (CPV), the feline parvovirus (FPV), the porcine parvovirus (PPV), the bovine parvovirus (BPV), the human parvovirus B19, the MVM immune-suppressive variant (MVMi), as well as several  2 non-autonomous parvoviruses like AAVs (reviewed in Cotmore and Tattersall, 2007). Parvoviruses are only pathogenic for young or immune-compromised hosts, and even infection with the parvovirus B19 induces little symptoms in the healthy human adults. The only exception is the MVMi, which targets lymphocytes and erythropoietic precursors, thus affecting the host immune system (reviewed in (Brownstein et al., 1991; Kontou et al., 2005; Parrish, 2010). In contrast MVMp remains asymptomatic after intranasal inoculation in newborn mice (Kimsey et al., 1986), even though it has been reported that MVMp variants generated during infection in mice can show increased virulence (Lopez-Bueno et al., 2006; Rubio et al., 2005). MVMp can infect mouse fibroblast cells and a variety of human cancer cells (Cornelis et al., 2004).  The ~5 kb single stranded DNA (ssDNA) genome of MVMp is enclosed within an icosahedral capsid of ~26 nm in diameter. It only carries 2 genes that produce a total of four proteins, which include the non-structural proteins (NS) 1 and 2 as well as the structural proteins VP1 (83 kDa) and VP2 (64 kDa). Another structural protein, VP3 (60 kDa), is generated upon proteolytic cleavage of the VP2 N-terminal region (reviewed in (Cotmore and Tattersall, 2007; Parrish, 2010). The MVMp capsid is composed of 60 copies of these structural proteins, with only ten VP1 (Tattersall et al., 1976; Tattersall et al., 1977).  Like other autonomous parvoviruses, MVMp needs to enter the nucleus and wait for the S-phase of the cell cycle to replicate its genome, as it requires the cellular factors involved in cell replication and transcription (Bashir et al., 2000). During the phase of replication of autonomous parvoviruses, the DNA-binding protein NS1 is synthesized first as it is necessary to initiate the transcription of viral DNA (Doerig et al., 1990; Doerig et al., 1988), and therefore it is widely used as a measurement of successful parvoviral infection. Nevertheless, this is not the only function of NS1. The parvoviral NS1 also induces DNA-damage response and cell cycle arrest in G2 phase to promote virus amplification (Adeyemi et al., 2010; Op De Beeck et al., 2001),  3 triggers apoptosis through accumulation of reactive oxygen species in the cytoplasm (Hristov et al., 2010), and finally allows release/egress of newly produced viruses from infected cells after gelsolin-dependent cell lysis (Bar et al., 2008). The viral NS2 protein has been associated with MVM replication (Cater and Pintel, 1992), and its interaction with the nuclear export factor cellular chromosome maintenance protein 1 (Crm1) is required for viral egress from the nucleus (Eichwald et al., 2002).   1.1.2 MVMp early infection steps The earliest step in the viral infection cycle is the virus attachment to the plasma membrane of the target cell. For this process, viruses can use many different cell surface receptors, which often determine the tropism of the virus for specific tissues. In the case of the parvoviruses MVMp, CPV, FPV, BPV, as well as several AAVs, the capsid binds with high affinity to a variety sialic acids exposed at the cell surface by glycolipids or glycoprotein (Halder et al., 2014; Johnson et al., 2004; Kaludov et al., 2001; Lofling et al., 2013; Nam et al., 2006; Wu et al., 2006).  Once attached to the plasma membrane, viruses can employ many different endocytic mechanisms to enter the cell (reviewed in (Mercer et al., 2010). As shown in Figure 1.1, most viruses use the classical clathrin-mediated endocytosis (CME), this is the case of VSV, influenza A virus, and adenoviruses 2 and 5, whereas only few viruses appear to use macropinocytosis, examples of these include Vaccinia virus and adenovirus 3. The SV40 is one of the rare viruses found to use both caveolin-dependent and lipid raft-mediated endocytosis, and this model has been somewhat revised (Engel et al., 2011). The mimivirus enters the cell upon phagocytosis, in contrast to AAV2, which requires the clathrin-independent carriers (CLICs) for transduction of its genome into Hela and HEK293T cells (Nonnenmacher and Weber, 2011). Other endocytic  4 mechanisms have been identified recently, including interleukin-2-, flotillin-, and Arf6-dependent entry pathways, but no viruses have been found to use these mechanisms so far. It is sometimes complicated to decipher between all these entry routes in different cellular models. For example, CME can also depend on lipid rafts (Quattrocchi et al., 2012).   CME is the preferred entry route for several parvoviruses. For example, IF and EM analysis have revealed that the parvovirus B19 enters Ut7/Epo cells via CME in a lipid raft-dependent manner (Quattrocchi et al., 2012). Similar approaches have shown that CPV enters canine cells via CME (Parker and Parrish, 2000), and EM analysis suggests that MVMp enters LA9 mouse fibroblast cells most likely via CME (Linser et al., 1977 and 1979). In contrast to these viruses, however, PPV and AAV5 can apparently employ several endocytic mechanisms. Indeed, the use of chemical inhibitors of endocytosis has revealed that PPV enters porcine testis fibroblast cells using both CME and macropinocytosis (Boisvert et al., 2010), whereas EM analysis of AAV5 uptake in human embryo fibroblast has shown that this virus is internalized via CME and caveolar endocytosis (Bantel-Schaal et al., 2009). Finally, it was shown by IF microscopy that AAV2 requires several regulators of the CLICs formation to transduce its genome into Hela and HEK293T cells (Nonnenmacher and Weber, 2011). Hence, parvoviruses can exploit a variety of endocytic mechanisms in different cellular models.  Once in the endosome, the parvovirus capsid slowly undergoes low pH-dependent conformational changes in the capsid protein VP1 (Mani et al., 2006) that expose a phospholipase A2 (PLA2) enzymatic domain as well as several consensus nuclear localization signals (NLS) contained within the N-terminal region of the capsid protein VP1 (reviewed in (Cotmore and Tattersall, 2007). The PLA2 domain allows slow and partial digestion of the endosomal membrane (potentially because of the limited of number of VP1 in the capsid, (Tattersall et al., 1977)) which allows the virus to slowly escape from late endosomal compartments to the cytosol,  5 where it may then interact with the cytoskeleton and molecular motors for intracellular trafficking (Suikkanen et al., 2003). Once in the cytoplasm, the virus somehow requires the ubiquitin-proteasome machinery for nuclear translocation (Ros and Kempf, 2004). The parvoviral capsid is then imported intact into the nucleus where it subsequently releases its viral ssDNA genome. This nuclear import process occurs potentially via the classical nuclear import pathway as several of the NLSs contained within the capsid protein VP1 are required for onset of MVM infection (Lombardo et al., 2002), or via transient disruption of the nuclear membrane (Cohen et al., 2011; Porwal et al., 2013).    6  Figure 1.1. Endocytic pathways used by viruses. See text for details. CME: clathrin-mediated endocytosis. IL2: interleukin 2. CLICS/GEEC: clathin-independent carriers. GEEC. GPI-enriched endosomal compartment. (Adapted from Mercer et al., 2010).    7 1.1.3 MVMp replication and cellular egress Once in the nucleus, the MVMp viral ssDNA is released after capsid uncoating, and waits for the host cell to enter S-phase in order to initiate transcription of viral genes using the cellular replication machinery (Cotmore and Tattersall, 2006). The viral ssDNA genome is then converted to a double stranded DNA (dsDNA) intermediate that allows subsequent transcription of viral messenger ribonucleic acids (mRNAs) and replication of the viral genome in a complex “rolling hairpin” mechanism (Berns, 1990). After translation in the cytoplasm, the viral NS1 protein is imported into the nucleus where it acts as a driver of viral DNA replication and capsid gene transcription (reviewed in Cotmore et Tattersall, 2006). It also induces cell cycle arrest (Cotmore and Tattersall, 2006) and the DNA-damage response (Adeyemi et al., 2014; Adeyemi et al., 2010) for prolonged viral amplification.  The capsid proteins synthesized in the cytoplasm are imported into the nucleus (Lombardo et al., 2000; Lombardo et al., 2002) for assembly of progeny virions after encapsidation of the viral ssDNA. The newly synthesized virions are subsequently exported to the cytoplasm (Eichwald et al., 2002; Maroto et al., 2004; Miller and Pintel, 2002) where NS1-dependent generation of reactive oxygen species induces cell lysis after gelsolin-dependent cleavage of the actin cytoskeleton, permitting viral release from infected cells (Bar et al., 2008; Hristov et al., 2010). It has also been documented that the progeny viruses can use the endoplasmic reticulum and the Golgi apparatus for vesicular egress and release in the extracellular milieu (Bar et al., 2013). Figure 1.2 summarizes the replication cycle of MVMp described in this section, according to the current literature.   8  Figure 1.2. Summary of MVM replication cycle. (1) Binding of the virus to the cell surface. (2) Clathrin-mediated endocytosis. (3) Trafficking of the virus toward the nucleus. (4) Escape of the virus to the cytoplasm. (5) Nuclear import of the virus. (6) Uncoating of the viral ssDNA genome. (7) Conversion of viral ssDNA to dsDNA intermediate. (8) Replication of the viral genome and transcription of viral proteins. (9) Export of the viral mRNAs to the cytoplasm. (10) Translation of viral proteins. (11) Nuclear import of viral proteins. (12) Assembly of progeny virions. (13) Viral egress from the infected cell.     9 1.2 Overview of viral oncotropism 1.2.1 Definition of oncotropic/oncolytic viruses Oncotropic viruses are microorganisms that naturally evolved or were genetically engineered to infect cancer cells preferentially, while sparing non-transformed cells (Guo et al., 2008; Miest and Cattaneo, 2014; Russell et al., 2012; Zeyaullah et al., 2012). In some cases, such viral infections are followed by lysis of the tumor cells; thus, some oncotropic viruses are also called oncolytic viruses. The specific tropism of these viruses is possible because oncotropic viruses exploit the cellular abnormalities acquired by cancer cells upon the successive mutations that allow tumor growth and dissemination. In addition, the crosstalk between cell division and immune response effectors is of major importance in the viral oncotropism phenomenon. For example, the immune escape of cancer cells that often arises upon loss of function of cell cycle regulators like p53 (reviewed in (Ozaki and Nakagawara, 2011; Rivlin et al., 2011; Suzuki and Matsubara, 2011) renders them simultaneously susceptible to viral replication, in contrast to normal cells, which inhibit viral replication through establishment of an anti-viral state (reviewed in (McFadden et al., 2009; Naik and Russell, 2009; Randall and Goodbourn, 2008). The loss of growth control during cancer transformation also increases the cell division rate, and thus, promotes infection and replication of oncotropic viruses that rely on cell division effectors to replicate their genome. This is the case for parvoviruses (PVs), which require the cellular replication and transcription machinery to amplify their own genetic material (reviewed in Cotmore and Tattersall, 2006). Furthermore, some tumor cells generate and exhibit specific cell surface antigens that can be targeted by oncotropic viruses in order to promote viral binding and uptake in these cells specifically.   10 1.2.2 Oncolytic virotherapy Because oncotropic/oncolytic viruses can replicate in cancer cells specifically, they can also induce the direct destruction of cancerous tissues while causing very limited side effects to the patient (reviewed in (Hemminki, 2014; Ilkow et al., 2014). Viruses represent considerably powerful tools as they can be genetically modified to bind cancer-cell specific surface antigens, to produce cytokines that will stimulate the anti-cancer immune response, or simply have their pathogenicity attenuated. For these reasons, and because the knowledge in the field of cancer development and progression has increased significantly, oncotropic/oncolytic viruses have gained the interest of the scientific community. Thanks to this better understanding of the cellular and molecular factors involved in tumorogenesis, but also because of the clear limitations of current cancer therapies, the research field of oncolytic virotherapy has literally exploded in the past two decades (Russell et al., 2012), even though oncolytic viruses were discovered much earlier. Ten years ago, up to eleven oncolytic viruses were already tested in clinical trials, and new candidates have emerged since. Some of them naturally infect cancer cells preferentially, like the parvovirus H1 (H1-PV), reovirus, and newcastle-disease virus (Maitra et al., 2012; Nuesch et al., 2012; Phuangsab et al., 2001), whereas others such as measles (MV), adenovirus, vesicular stomatitis virus (VSV), vaccinia, and herpes simplex viruses (HSV) have been modified to target cancer cells specifically (Ammayappan et al., 2013; Bajzer et al., 2008; Cassady and Parker, 2010; Haddad et al., 2012; Hallden and Portella, 2012). The new challenge is now to select which viruses have the potential of a safe anti-cancer agent, and this is why it is of major importance to characterize in detail the infection mechanisms of oncotropic viruses such as MVMp.   11 1.2.3 Parameters involved in viral oncotropism There are three main levels of regulation of viral oncotropism that have been reported so far. The first one occurs during the viral binding to the target cell, the earliest step in a viral infection cycle. Because tumor cells sometimes express specific antigens at their surface, some oncolytic viruses have been engineered to bind such receptors specifically, which consequently increases their chances of entering and infecting these tumor cells. For example, MV, Ad, and VSV were modified to bind tumor-specific antigens only (Hasegawa et al., 2006; Kanerva et al., 2004; Miller et al., 2010; Rein et al., 2005). On the other hand, some oncotropic viruses happen to naturally prefer tumor cell surface receptors. This is the case of echovirus 1, coxsackievirus, poliovirus, and MV, which target integrin α2β1 (ovarian cancer), Decay-accelerating factor (melanoma), cluster of differentiation (CD) 155 (glioma), and CD46 (ovarian cancer) respectively (Anderson et al., 2004; Au et al., 2005; Manchester et al., 2000; Merrill et al., 2004; Shafren et al., 2005).  The second critical regulator of the oncotropic behavior of some viruses is the tumor microenvironment, as it contains higher levels of secreted enzymes such as the matrix metalloproteinases (MMPs), which allow faster degradation of the extracellular matrix (ECM) to promote tumor dissemination (reviewed in (Gialeli et al., 2011; Pytliak et al., 2012) in comparison to normal tissues. Some enveloped viruses like the SARS-coronavirus (SARS-CoV) require a proteolytic cleavage of their membrane fusion proteins in order to fuse their envelope with the endosomal membrane and escape to the cytoplasm (Matsuyama et al., 2005). Hence, scientists have inserted MMP cleavage sites in the fusion protein of MV to restrict its activation in the tumor microenvironment specifically (Springfeld et al., 2006), whereas the SARS-Cov activating proteases are expressed in human respiratory and gastrointestinal tracts (Bertram et al.,  12 2012). Similarly, the oncolytic newcastle-disease virus contains a furin cleavage site in its F protein (Collins et al., 1993) that was modified to restrict infection to prostate cancer cells (Shobana et al., 2013).  The last major regulator of viral oncotropism takes place at the replication step in the viral infection cycle, and can depend on various cellular factors. The VSV, for example, can only replicate within cells that lack the interferon (IFN) antiviral response (Paglino and van den Pol, 2011) often the case in cancer cells as they carry mutations in the p53 tumor suppressor gene (reviewed in (Ozaki and Nakagawara, 2011; Rivlin et al., 2011; Suzuki and Matsubara, 2011). Indeed, the transcription factor p53 is a major regulator of the cell cycle that is normally expressed in response to variable oncogenic aggressions such as DNA damage (reviewed in (Levine and Oren, 2009; Vousden and Prives, 2009). During viral infection in normal cells, p53 is activated as a downstream target of type 1 IFN-inducible genes in order to trigger cell cycle arrest, destruction of viral mRNAs, and apoptosis of infected cells, thus preventing viral amplification and spread (reviewed in (Randall and Goodbourn, 2008; Sadler and Williams, 2008). Another oncotropic virus, myxoma virus, requires an activated signal transducer and activator of transcription 1 (STAT1) for replication (Wang et al., 2004). In contrast, reovirus can only infect cells with activated rat sarcoma oncogene (Ras; (Shmulevitz et al., 2010)).  1.2.4 What makes a good oncotropic/oncolytic virus Not all oncolytic viruses are good candidates for oncolytic virotherapy, and those that actually work can always be further optimized. Ideally, a good oncolytic virus should gather several characteristics (reviewed in (Parato et al., 2005): it should be harmless to humans, infect tumor cells with high specificity, have a rapid infection cycle, diffuse systemically throughout the whole  13 organism, not insert into the host DNA, induce a long-term anti-cancer immunity, and allow genetic engineering. I will describe these characteristics in more detail in the following paragraph.  The first obvious requirement for oncolytic viruses is that they should cause very little side effects when administered to the patient, and thus not be pathogenic. This low pathogenicity can be ensured by using viruses that are naturally harmless to human, or by modifying those that are human pathogens in order to attenuate their natural pathogenicity. Oncolytic viruses should also infect cancer cells specifically, to prevent destruction of normal tissues and the inevitable side effects that would result from such unwanted damage. The replication cycle of an oncolytic virus should be quick, with maximum viral amplification to allow virus spread before the establishment of an anti-viral immune response. To increase the access of the virus to disseminated tumors, an oncolytic virus should have a good systemic diffusion as well. Oncolytic viruses should also behave as potent adjuvants to induce immunization of the host against cancerous antigens and prevent further relapse, and it is clear that such viruses should not integrate into the genome of the host, to avoid inappropriate recombination. Lastly, a good oncolytic virus should also be easily modifiable genetically, so that the viral genome can be inserted with transgenes that allow in vivo tracking or the expression of anti-cancer genes that will stimulate the anti-cancer immune response.  Unfortunately, it is rather complicated for a single oncotropic virus to have all these ideal properties. For example, the smaller size of some oncotropic viruses tends to promote their systemic diffusion, but in return limits their potential for genetic engineering. For several reasons, PVs would appear as very good candidates for oncolytic virotherapy. Because they are non-pathogenic for human, they could be used without risk for the patient, and the chances of therapeutic success would be increased as the subject would have never been exposed to and thus  14 immunized against these viruses. Thanks to their very small size (about 26 nm in diameter), PVs show great systemic diffusion and could reach disseminated tumors that are usually inaccessible to other (bigger) oncotropic viruses. Finally, it has been shown that the parvoviral genome offers opportunity for genetic engineering (Allaume et al., 2012; El Bakkouri et al., 2005), and the high stability of the parvoviral capsid allows long-term storage and easier handling.  1.2.5 Parvoviral oncotropism Some parvoviruses infect cancer cells preferentially, and are thus considered potential anti-cancer agents (reviewed in (Nuesch et al., 2012; Rommelaere et al., 2010). The oncotropic PVs include the rat H1-PV, the MVMp, and the rat parvovirus LuIII. In addition, the dependoviruses AAVs, which also belong the Parvovidae family, exhibit highly variable tissue tropism (Zincarelli et al., 2008) and are considered potential agents for cancer gene therapy.   Originally, oncotropic PVs were thought to be oncogenic as some of them were detected in rodent tumors (Collins and Parker, 1972). H1-PV and MVMp were also shown to protect various animal models against oncogenesis (Chen et al., 1989; Dupressoir et al., 1989; Toolan and Ledinko, 1968), first suggesting that PVs could represent potential anticancer agents. Other studies demonstrated that transformation of murine fibroblast with Harvey-ras (H-ras) or polyoma virus middle T antigen (PyMT) increases permissiveness to MVMp infection (Mousset et al., 1986), and that cell transformation with ionizing radiation, carcinogen, or SV40 increased susceptibility to H1-PV infection (Cornelis et al., 1988a; Cornelis et al., 1988b), further supporting the potential use of rodent PVs in anticancer virotherapy. Intriguingly, the presence of orthologous endogenous parvovirus sequences in the human genome suggests that these viruses coexisted with mammals for about 98 million years (Liu et al., 2011), and it is somewhat  15 fascinating to think that oncotropic PVs may have conferred a selective advantage to the infected hosts that were simultaneously exposed to oncogenic factors during evolution.   Non-autonomous PVs like AAVs are incapable of replication in the absence of a helper virus and exhibit a wide tissue tropism (Li et al., 2005a; Li et al., 2005b). Hence these viruses have been long considered relevant and safe candidates for gene delivery and gene therapy, and there are ongoing clinical trials for the use of recombinant AAVs (rAAVs) as a treatment of various genetic diseases (Daya and Berns, 2008; Ojala et al., 2014). Unlike MVMp however, AAVs do not possess any natural oncotopism and have to undergo modifications to be used as oncolytic agents. Fortunately, these AAVs can be engineered to target specific cell types such as tumor cells, and to express tumor suppressor genes. For example, the cross-linking of bifunctional antibodies to the AAV2 capsid allowed transduction of non-permissive cells (Bartlett et al., 1999), while the insertion of a L14 peptide via genetic manipulation of the capsid gene permitted transduction of cancer cell lines expressing the L14-specific integrin receptor on their surface (Girod et al., 1999).  Once AAVs are retargeted to transduce specific cancer cells, they can be further modified to express proteins with onco-suppressive properties such as the cell cycle regulator p53. Indeed, transduction of wild type (wt) p53 cDNA using a recombinant rAAV/p53 vector inhibited tumor growth after injection into immunodeficient mice implanted with H-358 (lung) tumors (Qazilbash et al., 1997). In spite of these promising advances, one major limitation for the use of AAVs as gene therapy is that they commonly infect humans (Burguete et al., 1999), so the chances are high that the patients will already be immunized against these viruses and induce their rapid destruction.    16 1.2.6 Determinants of parvoviral oncotropism It has been documented that the oncotropic behavior of PVs is determined by several main parameters including: S-phase dependence for viral replication, sensitivity to antiviral response, and Raf-1-dependent phosphorylation of viral proteins. In the following, I review these determinants.  One of the main reasons why rodent PVs are oncotropic is that they lack the elements necessary for viral replication, and thus rely, once in the nucleus, on the cellular factors involved in cell replication and transcription in order to multiply (Cotmore and Tattersall, 2006). These factors are available during the S-phase of the cell cycle, and because cancer cells often escape growth control and proliferate very rapidly, oncolytic PVs tend to replicate in these more efficiently. For example, it is known that the transcription factor E2F is required for NS1-dependent transcription of the MVMp P4 promoter that begins at early S phase (Deleu et al., 1999).  As highlighted in Figure 1.3, the establishment of an antiviral state upon activation of the IFN-stimulated response element (ISRE) prevents viral replication and spread by several means (reviewed in (Randall and Goodbourn, 2008; Sadler and Williams, 2008; Sen and Sarkar, 2007). On one hand, it inhibits the translation of viral proteins and induces degradation of viral mRNAs. On the other hand, it induces cell cycle arrest, cell death, and promotes the presentation of viral antigens by the class 1 major histocompatibility complex to initiate immunization for later destruction of infected cells by cytotoxic lymphocytes. The antiviral response is obviously a major barrier to viral infection in vivo, and viruses such as rotavirus and reovirus have acquired mechanisms that enable them to actively block the IFN response (Sherry, 2009).   17  In the case of MVMp, it has also been shown that the activation of anti-viral response upon MVMp infection in normal but not transformed mouse cells partially determines the MVMp oncotropism (Grekova et al., 2010), while the lack of type 1 IFN antiviral response upon protein kinase R disruption regulates the MVMp oncolytic behavior (Ventoso et al., 2010). On the other hand, recent studies have reported that MVMp can escape the IFN-dependent viral control in mouse embryonic fibroblast (Mattei et al., 2013), and that MVMp infection is not inhibited by the type-1 IFN antiviral response in human normal and cancer cells (Paglino et al., 2014).   In addition to IFN response, the MVMp oncotropism is associated with Raf-1 signaling, as the nuclear translocation of MVMp capsid proteins for subsequent assembly relies on Raf-1-mediated phosphorylation of the viral structural protein VP2 (Riolobos et al., 2010). Raf-1 is a downstream effector of the Ras signaling pathway, which is often found activated upon Ras mutations in cancer cells (Prior et al., 2012). Ras transformation was also indirectly involved in transcription of the MVMp P4 promotor and MVMp genome replication (Burnett et al., 2003, Paglino et al., 2007). Finally, some aspects of the viral NS1 protein function require phosphorylation by several variants of the cellular protein kinase C (Lachmann et al., 2003; Nuesch et al., 2003). Since PKC expression and activation are apparently altered in some cancers (Abu-Ghanem et al., 2007), this is considered one more potential determinant of the MVMp oncotropism (Nuesch et al., 2012).  Nevertheless, these main parameters only partially explain the intriguing parvoviral oncotropism, and further studies are necessary to completely understand why PVs are oncotropic, and before one can use these viruses against cancer. Among the many oncotropic viruses mentioned in section 1.2.5, the only one that belongs to the Parvoviridae family and was used in clinical trials is the rat parvovirus H1-PV, even though MVMp also shows promising oncotropic/oncolytic properties in vitro and in vivo (Cornelis et al., 1988a; El Bakkouri et al.,  18 2005; Guetta et al., 1990; Rommelaere and Cornelis, 1991). Indeed H1-PV proved efficient in the treatment of human glioma after combined intratumoral and intravenous injection in grafted rats (Geletneky et al., 2010), and is currently being tested in a first clinical trial (Geletneky et al., 2012).    19  Figure 1.3. The type 1 interferon antiviral response. See text for details. Type 1 IFN binding to its receptor triggers STAT1 and 2 phosphorylation and dimerization. The STAT1/2 complex is imported into the nucleus where it acts as a transcription factor after recruitment of interferon regulatory factor 9 (IRF-9). This initiates expression of 2',5'-oligoadenylate synthetase (OAS), protein kinase R (PKR), major histocompatibility complex (MHC) 1, and caspases. ISRE: IFN-stimulated response element. (Adapted from (Sadler and Williams, 2008).    20 1.3 Overview of the mechanism of cell migration  Cancer cell migration/invasion has never been associated with the viral oncotropism phenomenon, yet it often determines the outcome of the cancerous disease. Advanced stage tumors undergo enough genetic modifications to acquire migration and invasion ability, which leads to cancer metastasis. In this section, I will describe various regulators of the cell migration and invasion.  1.3.1 The basis of cell migration Whether it is during embryonic development or wound repair, some cells have the ability to migrate from one site to another within an organism (reviewed in (Lamouille et al., 2014; Savagner, 2010; Yang and Weinberg, 2008). The cell migration mechanism depends on many cellular factors ranging from transmembrane adhesion molecules that allow the connection of the intracellular cytoskeleton to the ECM, the cytoskeleton itself which maintains the integrity/shape of the cell while exerting the physical stress that actively pulls the cell forward, to the various cellular enzymes that regulate complex signaling pathways and induce rearrangement of the cytoskeleton or the turnover of adhesion receptors (reviewed in (Friedl and Wolf, 2010; Huttenlocher and Horwitz, 2011; Parsons et al., 2010). Other molecules such as the secreted galactose binding lectins (galectins) or MMPs also play crucial roles in the cell migration process by respectively promoting the clustering and activation of cell surface receptors, or degrading the ECM (Goetz et al., 2008; Wang and McNiven, 2012). Importantly, the tight regulation and coordination of these cellular factors is necessary to achieve the proper directed cell migration that permits normal embryonic development or efficient wound repair.  21 1.3.2 Cell adhesion to the extracellular matrix One obvious requirement for cell migration is the adhesion of the cell to the extracellular environment, as it differentiates migrating cells from circulating cells like leukocytes (which only migrate when activated) for example. While circulating cells simply follow the flow (unless activated), migrating cells literally make their way through the dense mesh formed by the ECM and other cell tissues. Adhesion molecules thus permit the connection between the intracellular cytoskeleton that maintains the structure of the cell, and the ECM mostly composed of a complex mixture of fibronectin (FN), collagens, laminin and growth factors (reviewed in (Frantz et al., 2010; Hubmacher and Apte, 2013; Kim et al., 2011; Lu et al., 2012).   The most common cell-ECM adhesion molecules are called integrins. Integrins are heterodimeric receptors that form upon selective pairing of 18α and 8β subunits that have variable affinities for ECM proteins (reviewed in (Barczyk et al., 2010; Campbell and Humphries, 2011). As illustrated in Figure 1.4, the extracellular domain of integrins binds ECM ligands, while their cytoplasmic tail recruits multi-molecular complexes composed of many proteins such as talin, paxillin, vinculin, integrin-linked kinase, the proto-oncogene Src kinase, and the focal adhesion kinase. This interaction allows connection of the ECM to the cytoskeleton and transmission of various cell signaling that regulate cell migration (reviewed in (Calderwood et al., 2013; Deramaudt et al., 2014; Morse et al., 2014; Moser et al., 2009). Integrin activity is regulated in part at the intracellular level through direct binding with talin, but also at the cell surface via receptor clustering that can be influenced both by the heterogeneous composition of the plasma membrane or directly by the ECM.  22  Figure 1.4. Integrin-mediated cell adhesion to the ECM. See text for details. The clustering of integrins within focal adhesions (FAs) recruits various regulators of the FA signaling to the cytoplasmic tail of integrins, and permits the interaction with actin stress fibers. FAK: focal adhesion kinase. ILK: integrin-linked kinase. ECM: extracellular matrix. (Adapted from Nagano et al, 2012).  23 1.3.3 Galectins At the extracellular level, secreted ECM proteins of the galectin family interact with glycosylated cell surface receptors, and affect the lateral movement, the clustering, and signaling activity of these receptors. Several comprehensive reviews about galectins have been published recently (Boscher and Nabi, 2013; Vasta et al., 2012a). In the next paragraphs I outline only the important aspects of these proteins that relate to this thesis.  Galectins are nearly ubiquitous proteins that share an affinity for β-galactoside. There are 15 mammalian galectins with variable affinities for glycoconjugates. The members of the galectin family are characterized by a common amino acid sequence, the carbohydrate recognition domain (CDR), which allows their binding to specific glycans. Galectins can be classified within three groups, which include the prototype, chimeric and tandem-repeat galectins. The prototype galectins only contain a CRD and can form homodimers (e.g., Gal-1). The chimeric Gal carries a N-terminal extension in addition to the CRD and can form pentamers (e.g., Gal-3). The tendem-repeat Gals are composed of two different CDRs attached by a short peptide.   Galectins have been associated with many mechanisms such as immune responses, pathogen recognition, cell survival, cell adhesion and migration, but also the progression of cancers (reviewed in (Boscher et al., 2011; Leffler et al., 2004; Newlaczyl and Yu, 2011; Takenaka et al., 2004; Vasta et al., 2012b). Studies have demonstrated that Gal-1 can affect infectious behaviors of the human immune-deficiency virus type 1 (HIV-1), influenza A and Nipah viruses (Garner et al., 2010; Mercier et al., 2008; Yang et al., 2011), but the reported involvements of galectins in viral infections remain limited so far.   The Gal-3 ability to assemble in pentamers is particularly important in the cell migration process as it allows formation of a lattice upon Mgat5-modified N-glycans cross-linking at the  24 cell surface, which restrains the movement of integrins and promotes their stabilization within FAs (Boscher et al., 2011; Goetz, 2009), and for this reason the integrin glycosylation is a crucial factor of the cell migration (Janik et al., 2010).   Another cellular protein called Gal-3-binding protein (Gal-3-BP, also called Mac2-BP or 90K) has been found to promote Gal-3 activity through its intrinsic ability to bind Gal-3 (Grassadonia et al., 2004). It is also involved in the pathogen interaction (Fornarini et al., 1999), and was found recently to induce aggregation of AAV6 (Denard et al., 2012). Similarly to Gal-3, Gal-3-BP can bind β1-integrins and form pentamers (Sasaki et al., 1998) that promote the Gal-3-lattice formation. Integrin clustering is also initiated upon direct integrin binding to ECM molecules. For example, it has been well documented that α5β1-integrins receptor bind the ECM protein fibronectin (FN) via arginine-glycine-aspartate (RGD) repeated motifs (reviewed in (Ruoslahti, 1996; Schaffner et al., 2013; Singh et al., 2010) in order to form FA sites and recruit many cytoplasmic effectors for signal transduction. For these reasons, FN is commonly used as a promoter of cell migration for in vitro studies.  1.3.4 Signaling pathways involved in cell migration Once the cell is attached to the ECM via integrins, various intracellular signaling pathways are activated to induce the changes in cell shape that are necessary for cell migration. These changes are tightly regulated by the RAS superfamily of small GTPases, which includes (among others) Ras, Rho, Rac and CDC42 (reviewed in (Heasman and Ridley, 2008; Pylayeva-Gupta et al., 2011). As illustrated in Figure 1.5, RAS GTPases can be activated in different ways once recruited to FAs. For example, recruitment by paxillin triggers CDC42 activation at nascent  25 adhesions, and Rac activation at focal complexes in a Ras-dependent manner. On the other hand, recruitment of Src and FAK to mature FAs triggers ROCK activation by RhoA.  The signaling activity of Ras GTPases mostly regulates cell migration and invasion via reorganization of the actin cytoskeleton, which allows membrane protrusions (including filopodia, lamellipodia/pseudopodia and invadopodia) and stress fibers assembly (reviewed in (Le Clainche and Carlier, 2008). At the leading edge of migrating cells, CDC42 and Rac control elongation of actin filaments and the subsequent formation of membrane protrusions to allow forward cell migration. More specifically, CDC42 drives extension of filopodia via Formins-mediated linear actin polymerization for initial cell attachment and sensing of the extracellular environment (reviewed in Huttenlocher et al., 2011), while Rac regulates membrane ruffles like lamellipodia and pseudopodia (functionally equivalent (Yamaguchi and Condeelis, 2007)) via Arp2/3 complexes-dependent branched-actin meshwork assembly. On the other hand, Rho GTPase triggers stress fibers assembly to create the traction force of the migrating cell, and it is also involved in the formation of circular dorsal ruffles (Hoon et al., 2012). When migrating cells move forward, the initial FAs that form at the cell front mature and end-up closer to the cell body, and can be converted to singular membrane protrusions named podosomes (also called invadopodia in cancer cells). The role of invadopodia is crucial during the development of cancer metastasis, for they induce MMP-dependent degradation of dense matrices such as basement membranes, and thus promote tumor cell invasion (reviewed in (Hoshino et al., 2013).  The signaling activity of RAS GTPases is tightly interconnected, and the crosstalk between them provides a balance in the formation of cell protrusions and stress fibers, that is necessary for directed cell migration (Sahai et al., 2001). For example, it has been shown that increasing Rho signaling simultaneously reduces Rac activity, and vice versa (reviewed in (Parri and Chiarugi, 2010).   26  Figure 1.5. Schematic diagram of RAS signaling pathways involved in cell migration. See text for details. Grb2: growth factor receptor-bound protein 2. SOS: son of sevenless. FAK: focal adhesion kinase. ROCK: Rho-associated kinase.    27 1.3.5 Plasma membrane and focal adhesion turnover during cell migration In order to achieve proper directed cell migration, it is necessary for the cell to detach from the ECM when necessary. As mentioned above, integrins are attached to the ECM and would prevent cell movement if such attachments were permanent. The best way to stop this interaction and allow the cell to move forward is to remove these attachment sites from the cell surface. This is usually achieved via endocytosis in a mechanism called FA turnover that is highly dependent on the microtubule network (reviewed in (Caswell et al., 2009; Nagano et al., 2012). As illustrated in Figure 1.6, before FA disassembly the actin stress fibers are attached to the cytoplasmic tail of integrins, and then released during the formation of the endocytic vesicle that contains the FA. The protein dynamin induces subsequent fission of this vesicle from the plasma membrane, and the vesicle can then interact with microtubules for cytoplasmic trafficking. The turnover of FAs mostly happens at the leading edge of migrating cells, and obviously involves simultaneous endocytosis of plasma membrane. Several endocytic mechanisms have been related to the process of FA disassembly, including clathrin-, caveolin-, and lipid raft-dependent endocytosis, yet only the first two are microtubule-dependent (Ezratty et al., 2005; Chao and Kunz, 2009; Ezratty et al., 2009; Nethe and Hordijk, 2011; Urra et al., 2012; Wang et al., 2013a). Importantly, this turnover can lead to both the recycling of FA for reuse at the cell surface, or the degradation of mature/aged FAs in lysosomes. As indicated in Figure 1.7, it is the FN-dependent ubiquitinylation of integrin complexes during the maturation of FAs that determines whether these will be degraded in lysosomes or recycled to the cell surface (Lobert et al., 2010).     28   Figure 1.6. The process of FA disassembly. See text for details. FAK: focal adhesion kinase. (Adapted from Nagano et al, 2012).    29   Figure 1.7. Integrin turnover occurs at the leading edge of a migrating cell. See text for details. The ubiquitinylation of integrins (double bars) at cell adhesion sites to the ECM (green) occurs during the maturation of FAs, and determines the degradation of old adhesions in lysososmes.     30 1.4 Cell migration during tumorigenesis In many ways, the migration mechanism of cancer cells is comparable to the one of normal migrating cells like fibroblasts, in the sense that the minimum cellular factors (described in section 1.3) required for migration are qualitatively the same. The main differences reside in the expression level of these factors (often promoted in cancer cells), but also in the regulation of their activity. After cancer cells located in the initial tumor multiply and start competing with each other for nutrients and oxygen, they become invasive, migrate to distant sites and form metastases (Chaffer and Weinberg, 2011; Chiang and Massague, 2008; Valastyan and Weinberg, 2011). When becoming invasive, cancer cells usually undergo epithelial-mesenchymal transition (EMT) and reacquire the ability for migration that they lost upon cell specialization (reviewed in (Lamouille et al., 2014; Savagner, 2010; Yang and Weinberg, 2008).   After this critical step of EMT, these migrating cancer cells may undergo further genetic mutations that will improve their migration and invasion ability. Because of these mutations, various cellular proteins involved in the normal cell migration process can have their expression and activity increased, while those that regulate them might be inhibited. This is for example the case of the β1,6-acetylglucosaminyltransferase 5 (Mgat5), a Golgi enzyme involved in the N-glycosylation of cell surface and secreted proteins. It is well characterized that Mgat5 overexpression in cancer cells increases the number of N-glycans exposed by cell surface receptors such as EGFR or integrins, with the consequence to promote their interaction with secreted Gal-3 (Boscher et al., 2011; Dennis et al., 2009b). In this context, the direct Gal-3 binding to Mgat5-modified N-glycans facilitates the assembly and stabilization of integrins within FAs with the assistance of Gal-3-BP, leading to greater cell migration ability (Fig. 1.8). Not surprisingly, Gal-3 and Gal-3-BP expression is also boosted in cancer cells, and this is one of  31 the reasons why these proteins are considered potent markers for aggressive tumors (Chiu et al., 2010; Shankar et al., 2012).   Other examples of potential markers for cancer cells are the ECM protein FN and the MMPs (Fernandez-Garcia et al., 2014; Roy et al., 2009; Saito et al., 2008; Sudo et al., 2013), whose roles during cancer cell migration are adhesion and ECM degradation respectively (Folgueras et al., 2004; Gialeli et al., 2011; Kessenbrock et al., 2010). Indeed, overexpression of MMPs in cancer cells induces drastic and uncontrolled destruction of the tumor microenvironment, facilitating migration, invasion, and thus dissemination of tumors from the initial niche, which mechanisms are further promoted by FN over-secretion.     32  Figure 1.8. Involvement of Gal-3 and Gal-3-BP in FA stability. See text for details. The formation of a Gal-3 lattice upon binding to Mgat5-modified N-glycans promotes the stabilization of integrins within FAs, and the subsequent FA maturation and disassembly that are necessary for cell migration. This process is further amplified after binding of Gal-3-BP pentamers to Gal-3. (Adapted from (Goetz, 2009).     33 1.4.1 Epithelial-mesenchymal transition EMT is a fundamental process that allows the conversion of polarized epithelial cells into motile mesenchymal cells. It is required for embryonic development of multicellular organisms, and during wound healing events in the adult (reviewed in Lamouille et al., 2014; Savagner, 2010; Yang and Weinberg, 2008). EMT is also a key determinant of cancer progression as it enables cancer cells to escape the initial site of tumor growth, reach blood vessels, and disseminate within the host to establish metastases. Such secondary tumor sites represent a real danger for patients as they are out of reach of conventional cancer treatments such as surgery or radiotherapy, and the development of cancer metastasis is often a sign of poor prognosis. Strikingly, up to 90% of malignant cancers originate from epithelia (carcinomas), and this process is still under intense investigation.   During cancer progression, one of the initial triggers of EMT is the loss of epithelial cadherin (E-cad), a marker of epithelial cells involved in the maintenance of cell-cell junctions and epithelial integrity, which can result from mutations in the E-cad gene, E-cad transcriptional repression, or from E-cad cell surface removal by endocytosis (Barber et al., 2008; de Beco et al., 2012; Peinado et al., 2007). The down-regulation of E-cad is dependent on complex and interconnected signaling cascades involving several effectors of the cell migration process such as Ras, Rho, Rac, and Src, in response to extracellular stimuli (reviewed in (Thiery, 2002). Growth factors such as the hepatocyte growth factor, the fibroblast growth factor, or the transforming growth factor β (TGF-β), have been shown to provide such stimuli. In cancer cells, the fibroblast growth factor can indirectly activate Src and Rac signaling, while the hepatocyte growth factor stimulates Ras and PI3K signaling upon binding to its c-MET receptor, both of which lead to EMT (reviewed in (Thiery, 2002; Trusolino and Comoglio, 2002). Similarly, the  34 binding of TGF-β1 to its receptor activates Snail for repression of E-cad expression, and Rho for actin reorganization and cell migration (reviewed in (Miyazono, 2009; Wendt et al., 2009). Nevertheless, TGF-β alone is not sufficient to induce EMT, and is has been demonstrated that mutations in H-Ras gene or transformation with H-Ras are necessary for epithelial cells to undergo EMT rather than apoptosis upon TGF-β1 treatment (Oft et al., 1996). Mutations that activate the Ras gene are found very frequently in malignant cancers (Prior et al., 2012). Finally, EMT-like malignant cell transformation can be induced using PyMT expression (Guy et al., 1992; Spandidos and Riggio, 1986), as it activates the Ras and Src signaling pathways (Raptis et al., 1991; Webster et al., 1998). For this reason, PyMT transgenic mice are used widely as a model for the study of carcinogenesis and cancer metastasis (Fantozzi and Christofori, 2006).  1.4.2 Different types of cancer cell migration There are mainly three different types of cell migration that are employed by cancer cells, which include group/collective cell migration, mesenchymal, and amoeboid cell migrations (reviewed in (Lammermann and Sixt, 2009; Theveneau and Mayor, 2013). As indicated in Figure 1.9, epithelial cells use group cell migration during wound healing. In this case, the cells maintain cell-cell junctions and migrate using integrin-dependent cell adhesion to the ECM. The loss of cell-cell junctions during EMT allows mesenchymal cell migration, which also depends on integrin-ECM adhesions. Note that both group and mesenchymal cell migrations rely on the degradation of the ECM, and that in some cases mesenchymal cells can also undergo group cell migration by partially retaining cell-cell adhesions (Theveneau and Mayor, 2011). In contrast, amoeboid cell migration involves squeezing of the cell body to cross the ECM, and is not dependent on integrin-mediated adhesion or ECM degradation.  35  Figure 1.9. Different types of cell migration. See text for details. ECM: extracellular matrix. (Adapted from (Yamazaki et al., 2005).    36 1.5 Research objectives The overall aim of this thesis was to gain insights into the cell entry mechanism of MVMp. The replication of the MVMp viral genome is very well understood at the molecular level, but the early events of the MVMp infection cycle are still poorly documented. Because the endocytosis of cell surface receptors is often promoted in cancer cells in order to achieve various cellular processes such as mesenchymal cell migration, I hypothesized that MVMp takes advantage of such increase in endocytosis to enter tumor cells.   Mass spectrometry analysis of cellular proteins immunoprecipitated by MVMp yielded significant enrichment of Gal-3 and its binding protein Gal-3-BP (Garcin et al., 2013). Both of these proteins have been associated with the clustering and activation of cell surface receptors (Inohara and Raz, 1994; Koths et al., 1993; Rosenberg et al., 1991), and it was found recently that purified human Gal-3-BP induces aggregation of AAV6 capsid (Denard et al., 2012), supporting the possibility that Gal-3-BP also interacts with MVMp.   As described in section 1.3.3, Gal-3 function depends on the Golgi enzyme Mgat5, which is responsible for the glycosylation of cell surface receptors, allowing their interaction with Gal-3 and therefore an efficient directed mesenchymal cell migration (Boscher et al., 2011; Goetz, 2009). For example, it has been shown that the retention of integrins within FAs, which allows FAs maturation and disassembly, does not occur in Mgat5-/- cells, which is the reason why their migration ability is considerably hampered in comparison to Mgat5-wt cells (Goetz et al., 2008; Granovsky et al., 2000). Importantly, over-expression of Gal-3 and Mgat5 play a pivotal role in the development and spreading of cancer cells, and Gal-3 is considered a potent marker for malignant tumors (Chiu et al., 2010; Miranda et al., 2009; Mourad-Zeidan et al., 2008; Shankar et al., 2012; Takenaka et al., 2004; Wang et al., 2013b; Wang et al., 2009). Hence, I hypothesized  37 that Gal-3 and Mgat5 may contribute to the MVMp oncotropism through their involvement in the process of mesenchymal cell migration and cancer metastasis. The questions that I asked during this thesis project were the following:  1. Is it possible that Gal-3 and Mgat5 play a role in the MVMp early infection? 2. Is there a correlation between the Gal-3 and Mgat5 expression profiles and the ability of MVMp to infect cancer cells preferentially? 3. Is it possible that the process of cell migration regulates in part the MVMp early infection? 4. How does MVMp enter its target cells?   To address these questions, I have used various complementary cell biology techniques, including fluorescence microscopy, electron microscopy, flow cytometry and chemical assays. The following are the specific objectives of my PhD thesis and a brief synopsis of my results.  1.5.1 Aim 1: To assess the involvement of Gal-3 and Mgat5 in the MVMp early infection Since our lab previously identified Gal-3 as a potential binding partner for MVMp (Garcin et al., 2013), and given the involvement of Gal-3 in the clustering and endocytosis of cell surface receptors (Furtak et al., 2001; Gao et al., 2012; Goetz et al., 2008; Lajoie et al., 2007; Partridge et al., 2004), my hypothesis is that Gal-3 plays a role in the MVMp uptake and infection. To test this hypothesis, I used siRNA knockdown (KD) of Gal-3 in two different cells lines, as well as in  38 Mgat5-/- cancer cells. I then analyzed MVMp replication, uptake, and cell surface binding by immunofluorescence (IF) microscopy and flow cytometry in the Gal-3 KD cells (Chapter 3).  1.5.2 Aim 2: To screen various cancer cells for permissivity to MVMp infection Gal-3 is considered a reliable marker for aggressive thyroid tumors (Chiu et al., 2010; Shankar et al., 2012), and is also involved in the malignant progression of various tumors (Miranda et al., 2009; Mourad-Zeidan et al., 2008; Takenaka et al., 2004; Wang et al., 2013b; Wang et al., 2009). Hence, I next hypothesized that there might be a correlation between the Gal-3 over-expression in cancer cells and the MVMp ability to infect these cells. If this is true, cancer cells with higher levels of Gal-3 should be more susceptible to MVMp infection than cells with lower Gal-3 expression. To test this hypothesis, I performed a screen of human cancer cells originating from various tissues for their susceptibility to MVMp infection, with regard to their degree of aggressiveness and Gal-3 expression levels (Chapter 3).  1.5.3 Aim 3: To determine whether or not cell migration plays a role in MVMp infection Previously it has been reported that MVMp accumulates at the base of filopodia just before endocytosis in LA9 cells as observed by EM (Linser et al., 1977). Furthermore, the cells commonly used for analysis of the MVMp replication cycle and for mesenchymal cell migration studies are fibroblast. Hence, it is possible that the passive highjack of the mesenchymal cell migration mechanism by MVMp may partially regulate its early infection. To test this, I  39 reconstituted the early steps of MVMp infection, and assessed whether these correlate with various parameters of the mesenchymal cell migration process using a combination of IF, EM and flow cytometry (Chapter 4).  1.5.4 Aim 4: To identify the MVMp endocytosis pathways Recently it is has been documented that other parvoviruses use different endocytic pathways. For example, CPV uses CME (Parker and Parrish, 2000), B19 uses CME in a lipid raft-dependent manner (Quattrocchi et al., 2012), AAV-2 enters its host cells using CLICs (Nonnenmacher and Weber, 2011), and porcine parvovirus uses several endocytic pathways (Boisvert et al., 2010). The MVMp cell entry mechanism has been poorly documented so far. It is possible that MVMp could use any or several of the various endocytic mechanisms that have been discovered recently and which are used by other parvoviruses. Hence, my last aim was to characterize the MVMp cellular entry. In Chapter 5, I used IF, EM, flow cytometry, as well as various drug inhibitors of endocytosis to determine the endocytic pathways employed by MVMp in two different cell lines.            40 Chapter 2: Materials and Methods  2.1 Antibodies, reagents and cell lines Mouse anti-MVM capsid (clone D4H1, (Kaufmann et al., 2007)) and mouse anti-NS1 (clone CE10B10, (Yeung et al., 1991)) antibodies were a generous gift from Dr. Peter Tattersall (Yale University School of Medicine). The α5-integrin-GFP (Laukaitis et al., 2001) was a generous gift from Dr. Alan F. Horwitz (University of Virginia). Rabbit fibronectin (ab2413), LAMP1 (ab24170), Arp2/3 complex (ab47654), clathrin (ab59710) and N-Cadherin (ab18203) antibodies were purchased from Abcam. Mouse E-Cadherin (610182) antibody was purchased from BD laboratories. The goat anti-mouse Phyco-erythrin (PE)-conjugated secondary antibody (20103) was purchased from ImGenex. Rabbit Galectin 3 (Gal-3, H-160) and lamin A/C (H110) antibodies were purchased from Santa Cruz biotechnology. AlexaFluor 647-conjugated phalloidin and fluorescein isothiocyanate (FITC)-conjugated transferrin (1069904) were purchased from Invitrogen. The FITC-conjugated cholera toxin B subunit (CTxB), FITC-phalloidin, Bafilomycin A1, Neuraminidase, Cytochalasin B, Dynasore hydrate, Genistein, and Chlorpromazine hydrochloride were purchased from Sigma. The human recombinant transforming growth factor β1 (TGF-β1) was purchased from Research and Development Systems. Horseradish peroxidase (HRP)-conjugated mouse and rabbit, as well as AlexaFluor 649- and 549-conjugated secondary antibodies, were purchased from Jackson ImmunoResearch laboratories.   41 2.2 Cell culture and virus purification 2.2.1 Cell culture The mouse LA9 fibroblast, mouse mammary PyMT-transformed wild type (PyMT)(Granovsky et al., 2000), Mgat5-/- (Granovsky et al., 2000) and Mgat5-rescued (Partridge et al., 2004), the human thyroid T238 and TPC1 cells, as well as the human glioblastoma LN18 and LN229 cells were maintained at 5% carbon dioxide (CO2) and 37 °C in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 1% glutamine and penicillin-streptomycin. The human mammary MCF7 and MDA231 cells, as well as the human prostate LNCap and PC3 cells were maintained at 5% CO2 and 37 °C in Roswell Park Memorial Institute medium (RPMI). EpRas (Oft et al., 1996) cells were maintained at 5% CO2 and 37 °C in DMEM F12 supplemented with 5% FBS, 1% glutamine and 5µg/ml Insulin.  2.2.2. MVMp purification  MVMp was produced and purified based on the procedures described by Tattersall et al. (Tattersall et al., 1976) and Williams et al. (Williams et al., 2004). In brief, LA9 cells were grown in suspension in a stir flask containing Suspension medium (minimal essential medium (MEM)) containing 0.1% sodium bicarbonate, 5% FBS, and 1% penicillin-streptomycin, pH 7.4) at 37 °C, 5% CO2. The cell concentration was maintained between 2x105 and 6x105 cells/ml until a sufficient volume of cell culture (about 250 ml) was reached. The cells were then infected with MVMp at a MOI of 10-3 plaque-forming units per milliliter (pfu/ml) for 30 min at 37 °C. Following infection, the cell growth was monitored each day and the cell concentration was  42 maintained constant by addition of culture medium (about 2 l of final volume), until a decrease in the growth rate (about 5 days post-infection) was observed.   MVMp was purified from the infected cells rather than the culture medium. Care was taken to perform all steps of the purification with chilled, sterile buffers, at 4°C or on ice to avoid proteolytic activity. The cells were harvested and collected in 200 ml Nalgene centrifuge flasks for subsequent centrifugation at 1,600 g. The cell pellets were then washed in TNE buffer (50 mM Tris, 150 mM NaCl, 0.5 mM EDTA, pH 8.7), re-suspended in TE (50 mM Tris, 0.5 mM EDTA, pH 8.7) by repeated pipetting, and transferred to an Oak ridge centrifuge tube. The pellets were lysed by Sonication (4 time 4 sec) and centrifuged at 17,000 g for 30 min at 4 °C using a Avanti J251 centrifuge and a JA 25.50 rotor. The supernatants were transferred to new Oak ridge tubes and the virus was precipitated with 25 mM CaCl2 on ice for 30 min. The virus was then collected after 25 min centrifugation at 3,000 g, taken up with Uptake buffer T(20)E (50 mM Tris, 20 mM EDTA, pH 8.7) and centrifuged again at 12,000 g for 10 min. The supernatant was loaded onto a continuous CsCl gradient created by loading four Beckman ultraclear centrifuge tubes (14 mm x 89 mm) with 5 ml CsCl (0.53 g/ml in TE), followed by a sucrose cushion of 0.75 ml 1 M sucrose in TE, followed by 5 ml virus-containing supernatant. Gradients were then centrifuged at 100,000 g for 20 h. The next day, the bands containing full capsids were extracted using a syringe and needle. This was followed with dialysis in TE pH 8.7 using Pierce dialysis cassettes to ensure the purity of the viral preparation. In total, four dialyses of 24 h each were performed, and the dialysis buffer was replaced each day. The virus was then aliquoted and stored in liquid nitrogen. The purified virus was characterized by immune-fluorescence (IF) microscopy (as described in section 2.3.1), electron microscopy (EM) (as described in section 2.6), and plaque assay (as described in section 2.8). The yield was assessed by plaque assay to be 2.1x109 pfu/ml.  43 2.3 Fluorescence microscopy 2.3.1 Cell preparation for fluorescence microscopy Cells grown on glass coverslips and assayed as indicated below were rinsed once with phosphate buffered saline (PBS), fixed with 4% paraformaldehyde (PFA) in H2O for 10 min at 4 °C, permeabilized with 0.2% Triton X-100 in PBS containing 2.5% bovine serum albumin (BSA) for 2 min at room temperature (RT), and blocked with PBS containing 2.5% BSA for 30 min at RT. Cells were incubated with primary antibodies in PBS containing 2.5% BSA for 1 h at RT, washed three times with PBS containing 2.5% BSA, and then incubated with fluorescently-labeled secondary antibodies (diluted 1/5000) in PBS containing 2.5% BSA for 20 min at RT. When indicated, actin filaments were labeled using 1µg/ml AlexaFluor 647-conjugated phalloidin in PBS containing 2.5% BSA during the 20 min incubation with secondary antibodies. Cells were then rinsed three times with PBS containing 2.5% BSA, once with distilled water, and mounted in Prolong Gold Antifade with DAPI.   2.3.2 Confocal microscopy and image analysis  All IF images shown here were acquired (single Z slice) using a Fluoview 1000 confocal laser-scanning microscope (Olympus). Images of single cells were obtained with a 100 X plan apochromatic objective (pinhole 185 µm), and the lower magnification images with a 60 X plan apochromatic (pinhole 150 µm). For the MVMp clustering assay, images were acquired in the plane of focal contacts. For the MVMp uptake assay, images were acquired in the plane of the nucleus. All the fluorescence images showed are representative for at least 3 independent experiments.  44 2.4 MVMp assays 2.4.1 MVMp binding assay by immunofluorescence microscopy  About 50% confluent cell monolayers were grown onto glass coverslips and incubated with MVMp at indicated MOI (see figures legend) in infection medium (IM: DMEM + 1% glutamine) for 15 min at 4 °C, washed with PBS and prepared for IF microscopy (as described in section 2.3.1). MVMp was detected with a specific anti-capsid (D4H1) antibody (dilution 1/400).  2.4.2 MVMp clustering assay by immunofluorescence microscopy  About 50% confluent cell monolayers were grown onto glass coverslips and incubated with MVMp at a MOI of 8 in IM for 15 min at 4 °C. The virus-containing medium was then replaced with fresh IM, the cells were incubated at 37 °C for 0, 5 or 10 min, washed with PBS and prepared for IF microscopy (as described in section 2.3.1). MVMp was detected with a specific anti-capsid D4H1 antibody (diluted 1/400). For the clustering assay in the presence of Cytochalasin B (CytoB), the cells were incubated with 4 µM CytoB for 30 min at 37 °C before processing for MVMp clustering assay (in the presence of 4 µM CytoB) as described above.  For analysis of MVMp co-localization with α5-integrin, the cells were transfected with 3 µg α5-integrin-GFP DNA plasmids for 24 h before infection using lipofectamin 2000 reagent (Invitrogen) according to the manufacturer’s indications. For analysis of MVMp co-localization with CtxB, 2 µg/ml FITC-CtxB was added to the IM containing MVMp during the 15 min incubation period at 4 °C. In the case of the clustering assay during wound healing, the cells were grown to 100% confluence onto glass coverslips, a wound was created using a 100 µl tip, the  45 cells were allowed to recover for about 3 h and were then incubated with MVMp for clustering assay.   2.4.3 MVMp uptake assay by immunofluorescence microscopy  About 50% confluent cell monolayers were grown onto glass coverslips and incubated with IM containing 100 nM bafilomycin A1 (bafA1, which inhibits the vacuolar H+-ATPase in the endosomal membrane that is responsible for acidification (Bayer et al., 1998), and thus arrests MVMp in early endosomes and allows for a better observation of virions that enter the cells) for 1 h at 37 °C, and then with MVMp at indicated MOI (4 and 8, see figures legend) for 15 min at 4 °C in IM. The virus-containing medium was then replaced with IM containing 100 nM bafA1, and the cells were incubated for 4 h at 37 °C and 5% CO2, before preparation for IF microscopy (as described in section 2.3.1). MVMp was detected with a specific anti-capsid D4H1 antibody (diluted 1/400).   In the case of the uptake assay during wound healing, the cells were grown to 100% confluence onto glass coverslips, a wound was created using a 100 µl tip, the cells were allowed to recover for about 3 h and were then incubated with IM containing 100 nM bafA1 for 1 h at 37 °C and 5% CO2. Cells were then incubated with MVMp at a MOI of 8 for 15 min at 4 °C in IM containing 100 nM bafA1. The virus-containing medium was then replaced with IM containing 100 nM bafA1, and the cells were incubated for 4 h at 37 °C and 5% CO2, before preparation for IF microscopy analysis (as described in section 2.3.1). MVMp was detected using a specific anti-capsid D4H1 antibody (diluted 1/400). For IF analysis of MVMp cellular uptake in presence of drug inhibitors of endocytosis, LA9 and PyMT cells were grown on glass coverslips and incubated for 1 h at 37 °C with PBS  46 containing 25 µM chlorpromazine (CPZ, an inhibitor of clathrin-mediated endocytosis), 100 µM Dynasore (an inhibitor of dynamins), 50 µg/ml Genistein (an inhibitor of caveolar endocytosis) or dimethyl sulfoxide (DMSO, diluted 1/1000 as control condition), together with 100 nM bafA1. The cells were then infected with MVMp at a MOI of 8 for 4 h at 37 °C in presence of the drugs aforementioned, and prepared for IF microscopy analysis (as described in section 2.3.1). MVMp was detected using a specific anti-capsid D4H1 antibody (diluted 1/400). Optimal concentrations of the drugs used were determined as described in section 2.7.  2.4.4 MVMp infection assay  About 50% confluent cell monolayers were grown onto plastic dishes (for Western blot analysis) or glass coverslips (for IF analysis) and incubated with MVMp at indicated MOI (see figures legend) for 15 min at 4 °C in IM. The virus-containing medium was then replaced with fresh IM, the cells were incubated for 24 h (or 48 h in the case of the cells induced for EMT) at 37 °C and prepared for IF microscopy (as described in section 2.3.1) or Western blot (as described in section 2.10) analysis. MVMp and NS1 were detected using specific anti-capsid D4H1 (diluted 1/400) and CE10B10 (diluted 1/500) antibodies respectively.  2.5 Gal-3 siRNA transfection Cells were either mock transfected with Lipofectamine RNAiMAX (Invitrogen) or transfected with 20 pmol/µl siRNA oligonucleotide sequences targeting Gal-3 (Henderson et al., 2006) accordingly to the manufacturer’s recommendations. The nucleotide (nt) sequences of the sense strands of mouse Gal-3 siRNA were as follows:  • Gal-3 sequence 1: beginning at nt 502, 5'-GAUGUUGCCUUCCACUUUAdTdT-3'  47 • Gal-3 sequence 2: beginning at nt 4, 5'-GCAGACAGCUUUUCGCUUAdTdT-3' • Gal-3 sequence 3: beginning at nt 678, 5'-GGUCAACGAUGCUCACCUAdTdT-3' • Gal-3 sequence 4: beginning at nt 190, 5'-GGACAGGCUCCUCCUAGUGdTdT-3'  A pool of four non-targeting siRNAs (ON-TARGETplus Non-Targeting Pool, Dharmacon) was used as control siRNA. Expression of Gal-3 was assessed by Western blot (as described in section 2.10) and IF (as described in section 2.3.1) using a specific Gal-3 (H-160, diluted 1/400) antibody 24, 48 and 72 hours after transfection.   2.6 Electron microscopy About 60% confluent cell monolayers grown on Aclar films (Pelco) were assayed for MVMp clustering at a MOI of 32 as described in section 2.4.2. The cells were then washed once with PBS, fixed with 4% glutharaldehyde in 0.1M sodium cacodylate (NaCac) for 1 h at 4 °C, washed again with 0.1M NaCac, post-fixed with 1% OsO4 in 0.1M NaCac for 1 h at 4 °C, washed in distilled water and stained en bloc (prior to embedding) with 1% uranyl acetate in distilled water for 1 h at RT. Cells were then dehydrated in increasing ethanol (EtOH) concentrations (50%, 70%, 90%) for 10 min each, then thrice in 100% EtOH for 10 min, and finally once in acetone for 10 min. Next, samples were infiltrated in 1/1 Epon/acetone mixture for 1 h, then 2/1 Epon/acetone mixture for 2 h, and overnight in pure Epon. The cells were finally embedded in Epon and allowed to polymerize for 2 days at 60oC. Ultrathin sections (about 60 nm thick) resulting from en-face or cross sectioning (i.e. sections parallel or perpendicular to the cell monolayer) were positively stained with 2% uranyl acetate for 7 min, washed thrice with distilled water, stained with 2% lead citrate for 3 min and washed thrice with distilled water. Images were  48 acquired using a FEI Tecnai G2 transmission electron microscope operated at an acceleration voltage of 120 kV. Micrographs were digitally recorded using an Eagle 4k CCD camera (FEI). All the EM images showed are representative for at least 3 independent experiments.  For immuno-gold labeling of MVMp during clustering assay, about 60% confluent LA9 cell monolayers were grown on Aclar films (Pelco) and assayed for MVMp clustering at a MOI of 32 as described in section 2.4.2. The cells were then fixed with 4% paraformaldehyde in PBS containing 2.5% BSA for 1 h at 4 °C, washed with PBS containing 2.5% BSA, incubated with anti-MVMp capsid antibody (diluted 1/50) for 1 h, washed with PBS containing 2.5% BSA, incubated with 10-nm gold-conjugated anti-mouse antibody (diluted 1/40) for 1 h, washed again with PBS containing 2.5% BSA, and prepared for TEM analysis after embedding in Epon as described above.  2.7 Flow cytometry To detect MVMp binding to the cell surface by fluorescence-activated cell sorting (FACS), about 60% confluent cell monolayers were grown on tissue culture dishes and incubated with MVMp at MOI of 8 for 15 min at 4 °C. The virus-containing medium was then replaced with fresh IM and the cells were incubated for 4 h either at 4 °C (for MVMp binding analysis) or 37 °C (for MVMp uptake analysis). After a wash in ice cold PBS, the cells were detached with 0.25% trypsin-EDTA for 30 min at 4 °C, fixed with 2% PFA in H2O for 15 min at 4 °C, and blocked with PBS containing 2.5% BSA for 20 min at 4 °C. Cells were then incubated with MVMp anti-capsid D4H1 antibody (dilution 1/200 in PBS containing 2.5% BSA) for 30 min at 4 °C, washed with PBS containing 2.5% BSA, and incubated with PE-conjugated secondary antibody (dilution 1/500 in PBS containing 2.5% BSA) for 30 min at 4 °C. After two washes in PBS containing  49 2.5% BSA, the cells were re-suspended in PBS containing 2% FBS and analyzed with a FACS LSRII flow cytometer (BD Biosciences).   For FACS analysis of MVMp cellular uptake in presence of drug inhibitors of endocytosis, the cells were grown on tissue culture dishes and treated for 1 h at 37 °C with PBS containing 25 µM CPZ, 100 µM Dynasore, or 50 µg/ml Genistein. As control, cells were incubated with DMSO (diluted 1/1000). The cells were then incubated with MVMp at a MOI of 8 for 15 min at 4 °C. The virus-containing medium was replaced with fresh IM containing the various drugs mentioned above, and the cells were incubated for 4 h either at 37 °C (MVMp uptake control) or at 4 °C (MVMp binding control). After a wash in ice cold PBS, the cells were detached with 0.25% trypsin-EDTA for 30 min at 4 °C and prepared for FACS analysis as described above.  Optimal concentrations of the drugs used were determined by dose-dependent control experiments testing their efficiency of inhibition during uptake experiments of FITC-conjugated transferin, or FITC-conjugated CtxB. For these control experiments, the cells were pre-treated with drugs for 1 h at 37 °C, incubated with 20 µg/ml FITC-transferin in IM for 30 min at 4 °C, or with 2 µg/ml FITC-CtxB in IM for 15 min at 4 °C, and then incubated at 37 °C for 4 h. The cells were then detached by trypsin treatment and processed for FACS analysis (as described above) immediately. The ranges of drug concentration tested were 25-100 µM for CPZ, 100-200 µM for Dynasore, and 25-100 µg/ml for Genistein.  2.8 Plaque assay About 30% confluent cell monolayers were grown in 6 cm plastic dishes, washed with PBS and incubated with 1 in 10 dilutions of MVMp (starting at a MOI of 8) in Dilution medium (DMEM  50 1% FBS, 10 mM Hepes) for 1h at 37 °C. After removal of the virus, 7 ml of Overlay medium (MEM 5% FBS, 1% tryptose phosphate, 0.75% low melting point agarose, and 0.5% gentamycin) were added to each plate and allowed to solidify at RT for 15 min. The cells were then grown for 4 days at 37 °C and 5% CO2, fixed in 10% PFA for 30 min, washed and stained with 0.3% Methylene blue for 30 min. The plaque assay showed here is representative for 2 independent experiments.  2.9 Induction of epithelial-mesenchymal transition EpRas cells were seeded onto plastic dishes coated with 10 µg/ml FN and treated with 10 ng/ml transforming growth factor beta 1 (TGF-β1) for 72 h. The cells were then detached with trypsin and reseeded (in order to decrease the confluence and allow even spread of the cells) onto glass coverslips (for IF analysis) or plastic dishes (for Western blot analysis) coated with 10 µg/ml FN, for subsequent IF analysis of MVMp infection (as described in section 2.4.4) or Western blot analysis (as described in section 2.10) of EMT markers using E-cad (610182, diluted 1/1000) and N-cad (ab18203, diluted 1/500) antibodies.  2.10 Western blot  Cells were washed with ice-cold PBS, scrapped, centrifuged for 4 min at 1,000 g, resuspended in lysis buffer (20 mM Tris-HCl, pH 7.6, 0.5% NP-40, 250 mM NaCl, 3 mM EDTA, 3 mM EGTA containing freshly added 2 mM DTT, 0.5 mM PMSF, 1 mM sodium vanadate, 2.5 mM sodium fluoride, and 1 µM leupeptin) for 30 min at 4°C, before centrifugation for 4 min at 5,000 g. The supernatants were subsequently stored at -20oC. The protein concentrations of the cell lysates  51 were determined by Bradford assay. 10 µg total cell lysates were diluted in sample buffer (50 mM Tris-base pH 6.8, 2% SDS, 10% glycerol, 1% β-mercaptoethanol and 0.02% bromophenol blue) and run through a 10% SDS-PAGE for 1 h at 160 V, and transferred to nitrocellulose membrane by semi-dry transfer for 1 h at 20 V. Membranes were blocked with PBS containing 0.1% Tween 20 and 5% skim milk, followed by overnight incubation with either mouse NS1 (dilution 1/1000), E-cad (dilution 1/5000) or N-cad (dilution 1/1000) primary antibodies diluted in PBS containing 0.5% Tween 20 and 0.25% skim milk. After three washes in PBS containing 0.5% Tween 20, the membranes were incubated overnight with HRP-conjugated secondary antibodies. After three washes in PBS containing 0.5% Tween 20, the antibodies were detected using an enhanced chemiluminescent kit.  For detection of Mgat5-modified N-glycans, the membranes were incubated overnight with HRP-conjugated Phaseolus vulgaris leucoagglutinin (L-PHA-HRP) diluted (1/500) in PBS containing 0.5% Tween 20 and 2% BSA. The next day, the L-PHA was detected using an enhanced chemiluminescent kit after three washes in PBS containing 0.5% Tween 20. For loading control, membranes were stripped with stripping buffer (1.5% SDS, 1% β-mercaptoethanol, 0.6% Tris base in H2O) and re-probed with G-Actin (dilution 1:10000) or Lamin A/C (dilution 1:1000) primary antibodies as described above. All the Western blots showed are representative for at least 2 independent experiments.  2.11 Statistical analysis For the IF quantification of MVMp clustering, clusters bigger than 50 or 30 pixels (for LA9 or PyMT cells respectively, since I used different magnification for both) were counted using  52 ImageProPlus. For IF quantification of MVMp fluorescent signal and the percentage of NS1 positive cells, pictures were taken using a Zeiss Axioplan 2 fluorescence microscope and analyzed with ImagePro. For quantification of the fluorescence-activated cell sorting (FACS) analysis of MVMp uptake inhibition in the presence of drug inhibitors of endocytosis, the distance between the curves of each drug condition and the one of the uptake control was divided by the distance between the curves of uptake and binding controls. Values are given as mean of triplicates ± standard deviation (SD). Statistical significance was determined by unpaired Student’s t test. * p< 0.05; ** p<0.01. The unpaired t-test investigates the likelihood that the differences between two groups could have been caused by chance. P<0.05 was used as a threshold of significance.             53 Chapter 3: Involvement of Galectin 3 and Mgat5 in MVMp Infection  3.1 Introduction Previously, the extracellular matrix (ECM) protein galectin-3 (Gal-3) was identified as a potential cellular partner for the oncotropic MVMp during a mass spectrometry analysis (Garcin et al., 2013). One of the properties of Gal-3 is to bind β1,6-acetylglucosaminyltransferase 5 (Mgat5)-modified N-glycosylations exposed by cell surface glycoproteins such as integrins or EGFR in order to modulate their clustering and signaling activity (reviewed in (Boscher et al., 2011). It has been reported that Gal-1 can play a role in viral infections, but this is rather limited to HIV-1, influenza A and nipah viruses (Garner et al., 2010; Mercier et al., 2008; Yang et al., 2011), and galectins have never been shown to be associated with parvoviral infections. Nevertheless, the involvement of Gal-3 in cancer development and progression (Miranda et al., 2009; Mourad-Zeidan et al., 2008; Takenaka et al., 2004; Wang et al., 2013b; Wang et al., 2009), as well as the potential use of Gal-3 as a marker for aggressive tumor cells (Chiu et al., 2010; Shankar et al., 2012), are well characterized. Hence, because MVMp is known to infect cancer cells preferentially (reviewed in (Nuesch et al., 2012; Ponnazhagan et al., 2001; Rommelaere et al., 2010), I hypothesized that Gal-3 and Mgat5 play a role in the MVMp infection. To test this hypothesis, I assessed whether or not changes in Gal-3 and Mgat5 expression would affect MVMp infection in various cellular models. First, I used small interfering ribonucleic acid (siRNA) to silence Gal-3 expression, and followed MVMp infection, cellular uptake, and cellular binding by immunofluorescence (IF) microscopy. With this approach, I found that Gal-3 was required for efficient MVMp uptake and infection in LA9 mouse fibroblast, a model for MVMp studies, and PyMT-transformed mouse mammary epithelial tumor cells, a model for the study of  54 Gal-3 and Mgat5 involvement in cancer metastasis (Granovsky et al., 2000). In contrast, Gal-3 siRNA KD did not affect MVMp binding to the plasma membrane.   I next investigated the involvement of Mgat5 in the MVMp infection using PyMT mouse epithelial mammary tumor cells derived from Mgat5-knockout mice (Mgat5-/- cells; (Granovsky et al., 2000)). As control, I used a stable cell line resulting from Mgat5 rescue of these cells (Mgat5-/- cells stably rescued with a retroviral Mgat5 expressing vector (Partridge et al., 2004)). Strikingly, Mgat5-/- cells proved considerably resistant to MVMp infection, in comparison to wild type (wt) Mgat5 and Mgat5-rescued cells. MVMp uptake was also reduced in Mgat5-/- cells, whereas the MVMp cell surface binding was not affected. Lastly, during a screening of human cancer cells from various tissues, I discovered a correlation between the Gal-3 expression profile of these cells and their susceptibility to MVMp infection. In summary, the results presented here show that Gal-3 and Mgat5 are necessary for MVMp uptake and infection.   3.2 Results 3.2.1 Galectin 3 siRNA knockdown hampers MVMp infection in LA9 cells The soluble extracellular matrix protein Gal-3 and its binding protein Gal-3-BP were two of the best hits in the proteomic study our lab performed in collaboration with the lab of Dr. L Foster (Garcin et al., 2013). Since there is very little information concerning the roles of galectins in viral infection, I decided to characterize the potential contribution of Gal-3 to MVMp infection. For this purpose, I used siRNA-mediated KD of Gal-3, and I followed MVMp infection in the Gal-3-depleted cells. SiRNA targeting Gal-3, which has been previously shown to accomplish efficient depletion of Gal-3 (Henderson et al., 2006) was transfected into LA9 mouse fibroblast cells and the expression level of Gal-3 was followed over several days using Western blot  55 analysis of cell extracts and immunofluorescence staining of Gal-3. As control, cells were mock transfected with lipofectamine RNAiMAX without siRNA or transfected with control siRNA. As illustrated in Figure 3.1, A and B, Gal-3 expression was significantly reduced 72 h post-transfection in LA9 cells transfected with the siRNA against Gal-3, but not in the control experiments. Thus, these conditions were chosen for further experimentation throughout this study.   I then infected Gal-3-depleted cells and control-transfected cells with MVMp at MOI of 4 for 24 hours and detected the expression of the viral NS1 by immunofluorescence microscopy. NS1 is a replication initiator protein for MVMp and is used as readout of successful MVMp infection. The results show that NS1 could be detected in only approximately 13% of the Gal-3 KD infected cells, whereas 45-50% of the control cells were positive for NS1 (Fig. 3.1, C and D). Thus, Gal-3 is required for efficient MVMp infection.     56  Figure 3.1. MVMp infectivity is reduced in Gal-3 siRNA knockdown LA9 cells. (A) LA9 cells were either mock transfected with Lipofectamine, or transfected with siRNA targeting Gal-3 or control siRNA and prepared for IF microscopy 72 post-transfection. Gal-3 (green) was detected using a specific antibody and fluorescently-labeled phalloidin was used for detection of actin filaments (white). (B) Western blot analysis of LA9 cell lysates after Gal-3 siRNA KD as described in A. (C) LA9 cells were siRNA KD for Gal-3 as described in A and infected with MVMp at a MOI of 4 for 24 h at 37 °C and prepared for IF microscopy using antibodies against NS1 (magenta) and Gal-3 (green). DAPI was used for detection of the nucleus (blue). (D) Quantification of NS1 positive cells as observed by IF microscopy. (n=600, data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01). (Reproduced with permission from Garcin et al, 2013).  57 3.2.2 Galectin 3 siRNA knockdown hampers MVMp uptake in LA9 cells To further define the role of Gal-3 in the MVMp infection cycle, I asked whether depletion of Gal-3 would affect MVMp cell entry. For this purpose, Gal-3-KD cells and control-transfected cells were infected with MVMp for two hours in the presence of bafA1, which inhibits the vacuolar H+-ATPase in the endosomal membrane that is responsible for acidification (Bayer et al., 1998). MVMp is taken up via receptor-mediated endocytosis, and its escape from endosomes depends on acidification of this compartment (Mani et al., 2006). Thus, bafA1 arrests MVMp in early endosomes and allows for a better observation of virions that enter the cells. As revealed by fluorescence microscopy after labeling with MVMp antibodies, cellular uptake of MVMp decreased in the Gal-3 KD cells (Fig. 3.2). Quantification of the fluorescence intensity on these cells revealed a reduction of MVMp uptake of about 60% in the Gal-3 KD cells compared with the control cells. Thus, Gal-3 appears to significantly modulate the cellular uptake of MVMp.     58  Figure 3.2. MVMp cellular uptake is reduced in Gal-3 siRNA knockdown LA9 cells. (A) LA9 cells were either mock transfected or transfected with siRNA targeting Gal-3 or control siRNA, and infected with MVMp at a MOI of 4 for two hours at 37 °C in presence of bafA1 to inhibit endosomal acidification, and prepared for IF microscopy. MVMp (red) and Gal-3 (green) were detected using specific antibodies and DAPI was used for detection of the nucleus (blue). (B) Quantification of MVMp fluorescence intensity signal as observed by IF microscopy. (n=500, data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01). (Reproduced with permission from Garcin et al., 2013).    59 3.2.3 Galectin 3 siRNA knockdown does not affect MVMp binding to LA9 cells  Next, I evaluated the binding of MVMp to the cell surface of Gal-3-depleted cells and control-transfected cells that were incubated with MVMp at 4 °C to prevent MVMp cellular uptake. As illustrated in Figure 3.3, MVMp was able to bind to the plasma membrane of both Gal-3 KD and control-transfected cells. Quantification of the viral fluorescence intensity did not reveal statistically significant difference of the binding of MVMp to the control and Gal-3 KD cells (Fig. 3.4B). Thus, cell attachment is not impaired in Gal-3 KD cells. Close inspection of the IF signal for the MVMp cell surface binding experiment (Fig. 3.3A) revealed co-localization of Gal-3 with MVMp at the plasma membrane (Fig. 3.3A, zoom panel), indicating a possible interaction of Gal-3 with MVMp under physiological conditions. This co-localization is more obvious at regions of the plasma membrane where Gal-3 seems to be aggregated forming large immunostaining patches (Fig, 3.3A, arrows).  I also examined the MVMp binding to the plasma membrane by EM. Again, LA9 cells were incubated with MVMp at 4 °C and processed for EM analysis after embedding and ultrathin sectioning. As shown in Figure 3.4, this approach revealed distinct MVMp accumulation at the plasma membrane (Fig. 3.4).    60  Figure 3.3. Binding of MVMp to the cell surface is not affected in Gal-3 siRNA knockdown LA9 cells. (A) LA9 cells were either mock transfected or transfected with siRNA targeting Gal-3 or control siRNA, and incubated with MVMp at a MOI of 4 for two hours at 4°C and prepared for IF microscopy. MVMp (red) and Gal-3 (green) were detected using specific antibodies. The Zoom panels are close-up of cells on the Merge panel as indicated by the arrows, showing co-localization of MVMp with Gal-3. (B) Quantification of MVMp fluorescence intensity signal as observed by IF microscopy. (n=200, data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05). ns: not significant. (Reproduced with permission from Garcin et al., 2013).    61  Figure 3.4. Early event of MVMp binding to its receptor(s). Gallery of four selected examples of electron micrographs from four different LA9 cells showing viral clusters at the plasma membrane during MVMp binding to the cell surface membrane. Cells were assayed for the MVMp clustering assay at a MOI of 32 and prepared for Epon embedding and TEM analysis after ultrathin sectioning as indicated in section 2.6. Images shown are representative of three independent experiments. (Reproduced with permission from Garcin et al., 2013).    62 3.2.4 Galectin 3 knockdown hampers MVMp infection in PyMT cells  Next I decided to confirm this observation in a different cellular model derived from PyMT-induced transformation of mammary epithelial cells in mice (Granovsky et al., 2000). These cells have been used previously for the study of Gal-3. I first performed knockdown of Gal-3 using siRNA transfection of PyMT cells, and confirmed the KD efficiency by IF and Western blot analysis of Gal-3 levels, which were significantly reduced already at 48 h post-transfection (3.5, A and B), much more rapidly than in LA9 cells (Fig. 3.1A). The cells were then infected with MVMp at a MOI of 4 for 24 h at 37 °C, and prepared for IF analysis of MVMp infection. The viral protein NS1 and Gal-3 were detected using specific antibodies. As shown in Figure 3.5, B and C, the percentage of NS1-expressing cells was lower when these were treated with Gal-3 siRNA, in comparison with those treated with lipofectamin or non-targetting siRNA. Hence, Gal-3 is required for efficient MVMp infection in PyMT cells, in agreement with my previous observation in LA9 cells (Fig. 3.1).  3.2.5 Galectin 3 knockdown hampers MVMp uptake in PyMT cells I also analyzed the effect of the Gal-3 KD on MVMp cell entry by IF microscopy. PyMT cells were transfected with Gal-3 siRNA, control siRNA or lipofectamine, and subsequently infected with MVMp at a MOI of 4 for 2 h at 37 °C in the presence of bafA1. As shown in Figure 3.6, the MVMp uptake was reduced in PyMT cells treated with Gal-3 siRNA, in comparison to control conditions. This indicates that similarly to what I observed in LA9 cells (Fig. 3.2), Gal-3 is necessary for an efficient MVMp uptake in PyMT cells.    63  Figure 3.5. MVMp infectivity is reduced in Gal-3 siRNA knockdown PyMT cells. (A) PyMT cells were either mock transfected with Lipofectamine, or transfected with siRNA targeting Gal-3 or control siRNA and prepared for IF microscopy 48 h post-transfection. Gal-3 (green) was detected using a specific antibody and fluorescently-labeled phalloidin was used for detection of actin filaments (white). (B) Western blot analysis of PyMT cell lysates after Gal-3 siRNA KD as described in A. (C) PyMT cells were siRNA KD for Gal-3 as described in A and infected with MVMp at a MOI of 4 for 24 h at 37 °C and prepared for IF microscopy using antibodies against NS1 (magenta) and Gal-3 (green). DAPI was used for detection of the nucleus (blue). (D) Quantification of NS1 positive cells as observed by IF microscopy. (n=600, data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01).    64  Figure 3.6. MVMp cellular uptake is reduced in Gal-3 siRNA knockdown PyMT cells. (A) PyMT cells were either mock transfected or transfected with siRNA targeting Gal-3 or control siRNA, and infected with MVMp at a MOI of 4 for two hours at 37 °C in presence of bafA1 to inhibit endosomal acidification, and prepared for IF microscopy. MVMp (red) and Gal-3 (green) were detected using specific antibodies and DAPI was used for detection of the nucleus (blue). (B) Quantification of MVMp fluorescence intensity signal as observed by IF microscopy. (n=500, data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05).    65 3.2.6 MVMp does not infect Mgat5-/- cells It has been well characterized that Mgat5 over-expression in cancer cells promotes tumor progression and invasion in a Gal-3-dependent manner (reviewed in (Dennis et al., 2009b). Hence, I decided to test the Mgat5 involvement in MVMp infection using PyMT mouse epithelial mammary tumor cells derived from mice knockout for Mgat5 (Mgat5-/-, (Granovsky et al., 2000). LA9, PyMT, Mgat5-/-, and Mgat5-rescued (Mgat5-/- cells stably rescued with a retroviral Mgat5 expressing vector (Partridge et al., 2004)) total cell lysates were first probed by Western blot for Mga5-modified N-glycosylations using L-PHA-HRP to verify that Mgat5-/- cells lack these glycosylations. As shown in Figure 3.7, Mgat5-/- cells were indeed deficient in Mgat5-modified N-glycans, in contrast to LA9 and PyMT cells. Moreover, Mgat5-rescued cells displayed a much higher level of Mgat5 expression in comparison to LA9 and PyMT cells.   Next, LA9, PyMT, Mgat5-/- and Mgat5-rescued cells were infected with MVMp at a MOI of 8 (a higher MOI was used compared to Gal-3 siRNA experiments in Figures 3.1 and 3.5 to test the resistance of Mgat5-/- cells to MVMp infection) for 24 h at 37 °C and prepared for IF microscopy analysis to detect both the viral NS1 and the newly produced viral particles that should already be generated at 24 h post-infection with MVMp. The MVMp capsids and NS1 protein were detected using specific antibodies. As shown in Figure 3.8, about 80% of both LA9 and PyMT cells were positive for NS1 and MVMp progeny. Strikingly, Mgat5-/- cells revealed no sign of NS1 expression nor progeny virions, indicating that these cells are rather resistant to MVMp infection. Furthermore, the lack of Mgat5 clearly inhibited MVMp infection much more than the Gal-3 siRNA KD (Fig. 3.1 and 3.5). Importantly, this inhibition could be reversed in Mgat5-rescued cells, which were as permissive as the wild type PyMT cells (Fig. 3.8). I further validated these findings by Western blot analysis of NS1 expression levels in cell lysates of all  66 four cell lines after 24 h of infection with MVMp at a MOI of 8. As shown in Figure 3.8, there was no trace of NS1 in Mgat5-/- cells, whereas LA9, PyMT, and Mgat5-rescued cells showed comparable amounts of NS1.  To further test the level of resistance of Mgat5-/- cells to MVMp infection, I performed a plaque assay of MVMp in all four cell lines. In this experiment, the cells were infected for 4 days with decreasing virus concentrations, and the cell lysis induced by the virus was observed via the formation of plaques. As shown in Figure 3.9B, Mgat5-/- cells proved resistant up to 4 days post-infection with MVMp even at higher viral concentrations, while all other cell types still showed plaques at one in 104 virus dilution. To ensure that the lack of MVMp infection in Mgat5-/- cells was not due to a reduced proliferation of these cells (since MVMp requires cell entry into S-phase to establish infection), I performed cell proliferation assays with LA9, PyMT, Mgta5-/-, and Mgat5-rescued cells. This revealed that the cell division rate was much higher in Mgat5-/- cells in comparison to LA9 and PyMT cells, but also that Mgat5-rescued cells proliferated more slowly than all three other cell lines (Fig. 3.10). Hence, the inability of MVMp to infect these cells (Fig 3.8) is not due to a reduction in the cell division.     67  Figure 3.7. Detection of Mgat5 expression in different cells. (A) Total cell lysates from LA9, PyMT, Mgat5-/-, and Mgat5-rescued cells were prepared for Western blot analysis. The Mgat5-modified N-glycosylations were detected using HRP-conjugated L-PHA. Actin was detected with a specific antibody for loading control. Mgat5-Res: Mgat5-rescued. L-Pha: Phaseolus vulgaris leukoagglutinin. Image shown is representative of three independent experiments.     68  Figure 3.8. MVMp infectivity is abolished in Mgat5-/- cells as shown by IF microscopy. (A) LA9, PyMT, Mgat5-/-, and Mgat5-rescued (Mgat5-R) cells were infected with MVMp at a MOI of 8 for 24 h at 37 °C and prepared for IF microscopy analysis. MVMp (red) and the NS1 protein (pseudocolored in magenta) were detected using specific antibodies, and DAPI (blue) was used to observe the nucleus. (B) Quantification of the percentage of cells positive for NS1 expression or MVMp progeny from three experiments performed as described in A (n=1000, data are mean ± standard error of the mean measured from three independent experiments, *** p<0.005). ns: not significant.  69    Figure 3.9. MVMp infectivity is abolished in Mgat5-/- cells as shown by Western blot and plaque assay. (A) LA9, PyMT, Mgat5-/-, and Mgat5-rescued (Mgat5-R) cells were infected with MVMp at a MOI of 8 for 24 h at 37 °C and prepared for Western blot analysis. The viral protein NS1 and the cellular actin (loading control) were detected using specific antibodies. Res: Mgat5-rescued. -/-: Mgat5-/-. (B) Plaque assays of MVMp in the indicated cells. LA9, PyMT, Mgat5-/-, and Mgat5-rescued cells were mock-infected (Cont) or infected with MVMp at the dilutions indicated (starting at a MOI of 8) and processed for plaque assay as described in section 2.8. Images shown are representative of two independent experiments.       70  Figure 3.10. Mgat5-/- cells proliferate more rapidly than PyMT and Mgat5-rescued cells. LA9, PyMT, Mgat5-/-, and Mgat5-rescued (Mgat5-R) cells were grown in plastic dishes for 48 h, detached by trypsin treatment, and the viable cells were counted using a hemocytometer after trypan blue staining (data are mean ± standard error of the mean measured for three independent experiments, ** p<0.01; *** p< 0.005).    71 3.2.7 MVMp uptake is reduced in Mgat5-/- cells The previous results (Figs. 3.8-3.10) are indications that the inhibition of MVMp infection in Mgat5-/- cells occurs upstream of the replication step in the viral infection cycle. I thus hypothesized that an alteration in the MVMp uptake could be the cause of this inhibition, similarly to what I observed in Gal-3-KD cells (Figs. 3.2 and 3.6). To test this hypothesis, PyMT, Mgat5-/-, and Mgat5-rescued cells were infected with MVMp at a MOI of 8 for 4 h at 37 °C in the presence of bafA1, and the cells were prepared for IF microscopy analysis (as described in section 2.3.1). MVMp was detected using specific antibodies. As shown in Figure 3.11, MVMp uptake was reduced in Mgat5-/- cells compared to PyMT and Mgat5-rescued cells.   3.2.8 MVMp cell surface binding is not affected in Mgat5-/- cells  The previous findings indicate that a reduced MVMp uptake in Mgat5-/- cells prevents MVMp infection in these cells. In order to verify that this limited uptake was not a consequence of a reduced attachment of MVMp to the plasma membrane of Mgat5-/- cells, I analyzed the cell surface binding of MVMp in PyMT, Mgat5-/-, and Mgat5-rescued cells. MVMp cellular binding was assessed by fluorescence-activated cell sorting (FACS), which allows analysis of a considerable number of cells in a limited amount of time. MVMp was detected with a specific anti-capsid (D4H1) antibody. As shown in Figure 3.12, this approach revealed that MVMp binding to the cell surface was not reduced in Mgat5-/- cells in comparison to PyMT and Mgat5-rescued cells, as the fluorescence intensity resulting from the detection of cell-bound MVMp (Fig. 3.12, binding curve in red) was comparable in all three cell lines. Hence, the reduced MVMp uptake (Fig. 3.11) is not due to a reduced MVMp receptor binding in Mgat5-/- cells.   72   Figure 3.11. MVMp cellular uptake is reduced in Mgat5-/- cells. (A) PyMT, Mgat5-/-, and Mgat5-rescued (Mgat5-R) cells were infected with MVMp at a MOI of 8 for 4 h at 37 °C and prepared for IF microscopy analysis. MVMp (red) was detected using a specific antibody and DAPI (blue) was used to observe the nucleus. Images shown are representative of three independent experiments.   73  Figure 3.12. MVMp binding to the cell surface is not affect in Mgat5-/- cells. (A) Cells grown on plastic dishes were incubated with MVMp at a MOI of 8 for 15 min at 4 °C, washed with PBS, tripsynized and fixed for FACS analysis. MVMp was detected using a specific anti-capsid antibody. As controls of the antibodies specificity, the cells were incubated in the absence of any antibody (Auto-fluorescence) or the primary antibody was omitted (No primary ab). Graphs shown are representative of two independent experiments.    74 3.2.9 Correlation between Gal-3 expression and MVMp infection  Since Gal-3 is often associated with cancer progression (Miranda et al., 2009; Mourad-Zeidan et al., 2008; Takenaka et al., 2004; Wang et al., 2013b; Wang et al., 2009), and is also considered a reliable marker for some invasive tumors (Chiu et al., 2010; Shankar et al., 2012), I next asked whether its expression levels in human cancer cells would correlate with MVMp infection. For this purpose, I assessed the susceptibility to MVMp infection of several well-characterized human cancer cell lines originating from various tissues, with regard to their Gal-3 expression levels. For this screening experiment, I chose thyroid (T238 and TPC1), breast (MCF7 and MDA213), and prostate (LNCap and PC3) cancer cells with either low or high Gal-3 expression profiles (Table 1).   I first verified the Gal-3 cellular levels by Western blot (Fig. 3.13). In agreement with the literature, T238, MCF7, and LNCap cells displayed lower Gal-3 expression, in comparison to TPC1, MDA231, and PC3 cells respectively (Pacis et al., 2000; Sathisha et al., 2007; Shankar et al., 2012). Likely because of species-related variability (human Gal-3 ranges from 29-36 kD, Kramer et al., 2013), the mouse Gal-3 in LA9 cells had a slightly higher molecular weight than the human Gal-3 in the other cell lines (Fig. 3.13). These cells were then infected with MVMp at a MOI of 2 (a low MOI was used in order to test the sensibility of each cell line to MVMp infection) for 24 h at 37 °C and prepared for IF microscopy. The viral NS1 protein was detected using a specific antibody. As shown in Figure 3.14, the cells with higher Gal-3 expression, including TPC1, MDA231, and PC3, were infected more efficiently than the ones with lower Gal-3 levels. Strikingly, the TPC1 cells were even more susceptible to MVMp infection than LA9 cells. To ensure that the limited MVMp infection in the cells with low Gal-3 was not due to a reduced cell proliferation of these cells (since MVMp requires cell entry into S-phase to  75 establish infection), I next performed cell proliferation assays of all the cells used in the screening. This revealed that all the breast and prostate cancer cell lines used here proliferate more slowly than LA9 cells (Fig. 3.15). In contrast, the thyroid cancer cells showed greater cell proliferation rate, especially the TPC1 cells. Thus, it appears that the reduced MVMp infection in Gal-3 low cells observed in Figure 3.14 is not the result of a limited cell division, even though the greater MVMp infection observed in TPC1 cells may be partially related to the high proliferation rate of these cells.   Nonetheless, there was an exception to the trend observed during this screening, since analysis of MVMp infection (24 h at 37 °C, MOI of 2) in LN18 and LN229 brain tumor cells showed that even though these two cell lines exhibit similar Gal-3 levels during Western blot analysis (Fig. 3.16A), the LN229 cells were more permissive to MVMp infection than the LN18 cells (3.16, B and C). Again, this was not due to a hampered cell proliferation as LN18 cells were still able to proliferate at normal rates (Fig. 3.17).    76 Table 1. Names and details on the cancer cells used in the screening      77  Figure 3.13. Gal-3 expression profile in various cancer cells. Total cell lysates (names and details of the cells are indicated in Table 1) were prepared for Western blot analysis. Gal-3 and the cellular actin (loading control) were detected using specific antibodies. Image shown is representative of two independent experiments.    78  Figure 3.14. Correlation between Gal-3 expression and MVMp infection in cancer cells. (A) Cells were infected with MVMp at a MOI of 2 for 24 h at 37 °C and prepared for IF microscopy analysis. The viral NS1 protein (pseudocolored in magenta) was detected using a specific antibody, and DAPI (blue) was used to observe the nucleus, and actin filaments (pseudocolored in white) were labeled using AlexaFluor 647-conjugated phalloidin. (B) Quantification of the percentage of NS1-positive cells from three experiments performed as described in A (n=1000, data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01; *** p< 0.005).    79  Figure 3.15. Cell proliferation analysis. Cells (names and details of the cells are indicated in Table 1) were grown in plastic dishes for 48 h, detached by trypsin treatment, and the viable cells were counted using a hemocytometer after trypan blue staining (data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05; ** p<0.01).    80  Figure 3.16. No correlation between Gal-3 expression and MVMp infection in LN18 and LN229 cells. (A) Total cell lysates were prepared for Western blot analysis. Gal-3 and the cellular actin (loading control) were detected using specific antibodies. (B) Cells were infected with MVMp at a MOI of 2 for 24 h at 37 °C and prepared for IF microscopy analysis. The viral NS1 protein (pseudocolored in magenta) was detected using a specific antibody, DAPI (blue) was used to observe the nucleus, and actin filaments (pseudocolored in white) were labeled using AlexaFluor 647-conjugated phalloidin. (C) Quantification of the percentage of NS1-positive cells from two experiments performed as described in B. (n=500, data are mean ± standard error of the mean measured from two independent experiments, ** p< 0.01).    81    Figure 3.17. Cell proliferation analysis. Cells (names and details of the cells are indicated in Table 1) were grown in plastic dishes for 48 h, detached by trypsin treatment, and the viable cells were counted using a hemocytometer after trypan blue staining (data are mean ± standard error of the mean measured from three independent experiments, * p<0.05).    82 3.3 Discussion It is still unclear how MVMp is able to infect cancer cells preferentially, and the current models for the MVMp oncotropism (described in section 1.2.6) only answer part of the question. In this chapter, I report my discovery that Gal-3 and Mgat5, two major regulators of cancer progression, are crucial for MVMp infection. I show that Gal-3 is necessary for efficient MVMp uptake and infection in LA9 and PyMT cells, and that Mgat5-/- cells cannot be infected with MVMp. Moreover, I found a correlation between the level of Gal-3 expression and the susceptibility to MVMp infection in human cancer cells.  I first found that siRNA KD of the ECM protein Gal-3 hampered MVMp infection in LA9 and PyMT cells, and one could argue that this is because of the Gal-3 involvement in cell cycle progression (reviewed in Newlaczyl et al., 2011), a requirement of MVMp infection. Yet, Gal-3 has a rather anti-proliferative activity (reviewed in Dumic et al., 2006), and thus loss of Gal-3 would tend to promote MVMp infection. Moreover, MVMp cellular uptake was reduced in Gal-3 siRNA treated cells, whereas MVMp binding to the cell surface remained unaffected, indicating that Gal-3 is required in a post-binding but pre-uptake step of the MVMp infection cycle. Because Gal-3 also regulates clustering and endocytosis of cell surface receptors (Furtak et al., 2001; Gao et al., 2012; Goetz et al., 2008; Lajoie et al., 2007; Partridge et al., 2004), it is reasonable to think that Gal-3 plays a role in the MVMp cell entry, for this virus requires interaction with a cell surface receptor(s) for its early infection steps. There was little difference between LA9 and PyMT cells regarding the susceptibility to MVMp infection in control conditions, and the level of infection inhibition in Gal-3 siRNA KD PyMT cells (Fig. 3.5) was just slightly higher than the one in LA9 cells (Fig. 3.1). In contrast, the MVMp uptake inhibition in Gal-3 siRNA KD cells was slightly greater for LA9 (Fig. 3.2) compared to PyMT cells (Fig.  83 3.6). These differences regarding MVMp infection during Gal-3 siRNA experiments in LA9 and PyMT cells could be related to the Gal-3 siRNA KD efficiency. Indeed, the reduction in cellular Gal-3 after siRNA treatment observed by Western blot was already more efficient at 48 h post-treatment in PyMT cells (Fig. 3.5) than in LA9 cell (Fig. 3.1) at 72 h post-treatment.   It is now well established that Gal-3 pentamers can assemble into a lattice upon binding to various cell surface receptors (Boscher et al., 2011; Brewer et al., 2002; Lajoie et al., 2009). Gal-3 binding to plasma membrane-anchored receptors requires Mgat5-mediated N-glycosylation of these receptors; a process that takes place in the Golgi apparatus and that is directly implicated in the aggressive phenotype of various tumor cells (reviewed in (Dennis et al., 2009b). In fact, both Gal-3 and Mgat5 over-expression are associated with the aggressive phenotype of tumor cells (Guo et al., 2001; Shankar et al., 2012). Hence, given the relationship between Gal-3 and Mgat5, we could expect that preventing Mgat5 expression would also affect MVMp infection. Strikingly, Mgat5-/- cells appeared extremely resistant to MVMp infection. In comparison, Gal-3 KD only induced partial inhibition of MVMp infection at a MOI of 4 in LA9 and PyMT (Fig. 3.1 and 3.5), whereas Mgat5-/- cells did not get infected with MVMp even at a MOI of 8 and up to 4 days post-infection (Fig. 3.9). Importantly, the Mgat5-rescued cells were infected with MVMp as efficiently as the PyMT cells and even allowed synthesis of progeny virions (Fig. 3.8-3.9), which indicates that MVMp indeed requires Mgat5 expression to establish productive infection, and that the loss of permissivity in Mgat5-/- cells is not a cellular artifact. Notably, the proliferation rate of Mgat5-/- cells was much higher than the one of PyMT cells, which divided already more rapidly than LA9 and Mgat5-rescued cells (Fig. 3.10). Since MVMp requires cell entry into S-phase to establish infection (Rhode, 1973; Siegl and Gautschi, 1976), it is at a level upstream of the cell division that the MVMp infection cycle is affected in Mgat5-/- cells.   84  The binding of MVMp to the plasma membrane remained unaffected in Mgat5-/- cells (Fig. 3.12), which was the same effect I observed for Gal-3 siRNA KD in LA9 cells (Fig. 3.3). This was unexpected as my hypothesis was that Mgat5 might contribute to creating additional plasma membrane receptors for MVMp, as it does for Gal-3 (reviewed in (Dennis et al., 2009a). My results indicate that this is not the case. Therefore, I propose that it is most likely the endocytosis of MVMp receptor(s) that is affected in Mgat5-/- cells, rather than the ability of MVMp to bind this receptor(s). This scenario is somewhat plausible since previous works showed that Gal-3 molecules assemble into a pentameric structure (Dumic et al., 2006), which is a property that allows Gal-3 to crosslink Mgat5-modified glycoproteins at the cell surface, and regulate their signaling activity and endocytosis. Several studies with Gal-3 and other galectins have shown that these selectively crosslink a single species of glycoproteins to form uniform lectin–carbohydrate microdomains or lattices that influence cellular functions. For examples, Gal-3 crosslinks EGFR, transforming growth factor β (TGF-β) receptors and transmembrane neurotrophin receptors p75 and gp114 (Delacour et al., 2006; Delacour et al., 2007; Partridge et al., 2004). Similarly, Gal-1 binds to CD7, CD43, and CD45 on T cells and induces a dramatic redistribution of these glycoproteins into segregated membrane microdomains on the cell surface (Pace et al., 1999). Therefore, it is possible that Gal-3 promotes MVMp cell entry via crosslinking of the MVMp receptor(s) in an Mgat5-dependent manner, which might significantly modulate the receptor function/activity resulting in rapid endocytosis of MVMp. This would also explain why I observed Gal-3 aggregates co-localised with MVMp (Fig. 3.3).   Notably, the implication of galectins in viral endocytosis has been reported for other viruses such as HIV-1, influenza A and nipah viruses (Garner et al., 2010; St-Pierre et al., 2010; Yang et al., 2011). In the case of HIV-1, it has been shown that soluble Gal-1 promotes viral binding to its target cells (CD4+ lymphocytes) via direct interaction of Gal-1 with both the viral  85 gp120 and the host CD4 receptor. In contrast, Gal-1 displayed anti-viral activities against influenza A and nipah viruses via direct Gal-1 binding to the viral enveloped glycoproteins involved in the cell entry mechanism of these viruses. Based on my results, Gal-3 is necessary for efficient MVMp infection (Fig. 3.1 and 3.5), but the binding of MVMp to the plasma membrane of LA9 cells was not affected during Gal-3 siRNA (Fig. 3.3). Hence, it seems likely that the mechanism by which Gal-3 regulates MVMp early infection is different than the one of Gal-1 in HIV-1 infection. This might be related to the fact that unlike HIV-1, MVMp is a non-enveloped virus, and thus lacks the envelope glycoproteins required for attachment of HIV-1 to its target cells.   During our proteomic analysis of MVMp cellular partners (Garcin et al., 2013), we also detected Gal-3-BP, one of the native ligands of Gal-3. Gal-3-BP is a secreted protein heavily glycosylated containing a scavenger receptor cysteine-rich domain, and plays a role in the regulation of immune responses (reviewed in (Grassadonia et al., 2004). It also binds to integrins and multiple proteins of the ECM. Consistent with our proteomic findings, Gal-3-BP has recently been identified as binding partner of human parvovirus AAV6 (Denard et al., 2012). In fact, Gal-3-BP literally induces aggregation of AAV6 in a concentration-dependent way, and it is rather intriguing that the small clusters of membrane-bound MVMp I observed during EM experiments (Fig. 3.4) resemble these AAV6/Gal-3-BP aggregates.   Based on my results presented in this chapter, I propose a role for Gal-3 in the regulation of MVMp early infection (Fig. 3.18). In this model, the direct MVMp binding to its receptor allows lateral movement of the viral particle at the plasma membrane, until it encounters the Gal-3 lattice, which retains receptor-bound MVMp in membrane micro-domains and promotes viral endocytosis.   86  Figure 3.18. Model of the Gal-3 involvement in MVMp early infection. MVMp particles first bind to glycosylations exposed by cell surface receptor(s) and diffuse with these latter at the cell surface, until they encounter a Gal-3-rich membrane micro-domain (A). The Gal-3 lattice then retains MVMp receptors within these cellular domains, and activates the intracellular signaling pathways required for subsequent MVMp endocytosis (B).    87  Because of its various properties described in section 1.3.3, Gal-3 is considered a potential marker for aggressive tumors (Chiu et al., 2010; Shankar et al., 2012). This is particularly true in thyroid cancers, where Gal-3 appears to be “the single most accurate marker for the diagnosis of differentiated thyroid cancer” (Shankar et al., 2012). Since I found that both Gal-3 and Mgat5 play a crucial role in the MVMp infection, I expected that there could be a correlation between Gal-3/Mgat5 expression and the ability of MVMp to infect cancer cells preferentially. Interestingly, this is what I observed during my small screening for MVMp permissivity in cancer cells originating from various tissues, with the exception of the brain cells LN18 and LN229 (Fig. 3.16). Indeed, as shown in Figure 3.14, MVMp was barely able to infect any of the poorly invasive cancer cells with low Gal-3 levels, while the cells with higher Gal-3 expression could always be infected significantly. This somewhat indicates that similarly to Gal-3 (Chiu et al., 2010; Shankar et al., 2012), MVMp could be used as a marker for aggressive tumors. Even more important, the targeting of Gal-3/Magt5 positive tumors cells by MVMp would enable tumor destruction, as the virus replication can lead to cell lysis, with considerable and localized amplification of the oncolytic material. One could also measure the tumor progression (or remission) after injection (either intratumoral or intravenous) of variable doses of MVMp.   In conclusion, the results presented in this chapter demonstrate that the ECM protein Gal-3 and the Golgi enzyme Mgat5 are newly identified determinants of MVMp infection. Moreover, the correlation between Gal-3 expression and MVMp infection in various human cancer cells suggests that Gal-3 could be one more player of the MVMp oncotropism.    88 Chapter 4: Cell Migration is another Player of the MVMp Infection  4.1 Introduction As described in Introduction (Section 1.2.6), various intracellular properties of cancer cells have been identified as factors that regulate MVMp oncotropism/oncolytism. Yet, these factors act once the virus enters the cell, and it is still unclear whether MVMp oncotropism might be regulated at earlier time of infection, as for example during cell entry of the virus. Another characteristic of many types of aggressive tumor cells is their increased migration/invasion ability, which rely on the uptake and degradation of cell surface receptors (Nagano et al., 2012). Therefore, I hypothesized that MVMp and potentially other oncotropic viruses take advantage of some aspects of the cellular migration machinery in order to infect highly migrating/invasive cancer cells preferentially. Supporting this hypothesis, most of the cellular models for MVMp studies are mouse fibroblast cells, a migrating cell type widely used for analysis of all aspects of the mesenchymal cell migration (MCM) process, such as the structure of FA complexes, or the turnover of cell membrane/surface receptors. Furthermore, Linser et al. (1977) showed by EM that MVMp particles regroup around cellular filopodia just before endocytosis in LA9 cells (Linser et al., 1977), but also that MVMp can enter LA9 cells from cell ECM contact sites (Linser et al., 1979). These are indications that MVMp cell entry could occur at the leading edge of migrating fibroblast. Lastly, my data presented in Chapter 3 showing that Gal-3 and Mgat5, two crucial regulators of the cancer cell migration and metastasis, are involved in the MVMp infection and oncotropism also support this hypothesis.  In this chapter, I report my studies addressing the question of whether the cell migration process could play a role in the MVMp infection. First, I used EM to confirm the previous results  89 reported by Linser et al. that MVMp virions accumulate at the base of filopodia (Linser et al., 1977), which I observed in both LA9 mouse fibroblast and PyMT mouse epithelial mammary tumor cells. Next, using IF microscopy, I found that MVMp requires cell protrusions (filopodia and lamellipodia/pseudopodia) to cluster at the leading edge of migrating cells, in a temperature-dependent manner. Furthermore, cells that migrate during wound healing assay showed higher levels of MVMp uptake. Similarly, promoting cell migration on a FN matrix increased MVMp infection. Lastly, induction of EMT triggered MVMp infection in highly dividing non-permissive epithelial cells. In summary, the results presented in this chapter show that cell migration promotes MVMp entry and infection of its targets cells.  4.2 Results 4.2.1 MVMp infection begins with the clustering of viral particles at the base of filopodia Previous studies indicate that some parvoviruses rely on filopodia formation and integrin activity to infect their target cells (Linser et al., 1977; Weigel-Kelley et al., 2003), two cellular factors that are also involved in the process of cell migration. As many aggressive cancer cells migrate very rapidly, this could be a reason why some parvoviruses can infect certain tumor cells preferentially. Previously EM analysis showed clusters of MVMp particles accumulated at filopodia just before viral entry into LA9 cells (Linser et al., 1977).  I decided to confirm these observations in LA9 and PyMT cells using a similar approach. Cells were incubated with MVMp at 4 °C for 2 h, and switched to 37 °C for 5-10 min (Clustering assay). After sample preparation for EM, ultrathin en-face sections of the cells (i.e. sections parallel to the cell monolayer) were positively stained and observed using a TEM (as described in section 2.6). As shown in Figure  90 4.1, this analysis revealed small groups of MVMp particles associated with filopodia and accumulating at their base in both cell types. Pre-embedding immunogold labeling using an anti-MVMp capsid antibody confirmed that the observed clusters were indeed MVMp particles (Fig. 4.2).     91  Figure 4.1. Electron micrographs of MVMp clustering at filopodia. Cells were assayed for MVMp clustering at a MOI of 30 and prepared for EM. Images were acquired using a transmission electron microscope (TEM) after en-face ultrathin sectioning (i.e. sectioning parallel to the cell monolayer), and positive staining. Panel A and B show electron micrographs of MVMp particles accumulating at the base of filopodia from LA9 and PyMT cells respectively. The arrowheads point to areas that are enlarged in panels Z1 and Z2. FP indicates filopodia. Images shown are representative for three independent experiments.  92   Figure 4.2. Immunogold labeling of MVMp after clustering assay. LA9 cells were grown on Aclar films and assayed for MVMp clustering at a MOI of 30. As control (Mock) cells were incubated in the absence of virus and processed for EM analysis as described in section 2.6. Arrows point to MVMp particles. Arrowheads point to a gold particle clearly associated with virions. Images shown are representative for two independent experiments.    93  Next I repeated this clustering assay in LA9 and PyMT cells, but this time the cells were prepared for IF microscopy analysis (as described in section 2.3.1). As shown in Figure 4.3, clusters of MVMp began to form at the base of filopodia in both LA9 and PyMT cells after 5 min incubation at 37 °C. However, those clusters were not observed when cells were not incubated at 37 °C after the initial 4 °C incubation period (Fig. 4.3, panel labeled 0 min). Moreover, the number of clusters in LA9 cells still increased until 10 min incubation at 37 °C, in contrast to PyMT cells in which the maximum MVMp clustering was already reached after 5 min of incubation at 37 °C. This indicates that the dynamic of MVMp clustering is faster in PyMT cells compared to LA9 cells, and this is why I chose to use 10 and 5 min incubations at 37 °C for LA9 and PyMT cells respectively during the following clustering assays reported below. It also shows that MVMp particles do not bind filopodia directly, but rather diffuse with their receptors toward the edge of the cells.  Based on the results presented in Figures 4.1 to 4.3, I hypothesized that the formation of cell protrusions (including filopodia and lamellipodia/pseudopodia) was necessary for the MVMp clustering event. Hence, I next tested the effect of a drug inhibitor of actin polymerization on the MVMp clustering ability in LA9 and PyMT cells. For this experiment, the MVMp clustering assay was performed in the presence Cytochalasin B (CytoB, an inhibitor of actin polymerization, (Flanagan and Lin, 1980)). Short treatment (less than 1 h) with CytoB at low concentration (less than 10 µM) prevents formation of cell protrusions without inducing drastic cell detachment from the coverslips or rounding, and it has been shown that this drug prevents motile processes in fibroblast cells (Bliokh et al., 1980; Lin et al., 1978). As shown in Figure 4.4, CytoB hampered the clustering of MVMp particles in both LA9 and PyMT cells, indicating that the MVMp clustering mechanism is indeed dependent on actin polymerization and formation of cell protrusions at the leading edge of migrating cells.  94   Figure 4.3. Time course of MVMp clustering at the plasma membrane. (A) Cells were grown on glass coverslips, incubated with MVMp at a MOI of 8 for 2 h at 4 °C, and shifted to 37 °C for 0, 5 or 10 min before preparation for IF microscopy. MVMp (red) was detected using a specific anti-capsid antibody, and actin filaments (pseudocolored in white) were labeled using AlexaFluor 647-conjugated phalloidin. Arrowheads point to MVMp clusters. (B) Quantification of the number of MVMp clusters (bigger than 50 pixels) per cell for all the conditions from three experiments performed as described in A. (n=100, data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05; ** p<0.01). 0 min: 0 minute; 5 min: 5 minutes; 10 min: 10 minutes.  95  Figure 4.4. MVMp clustering at the cell surface is dependent on actin polymerization. (A) LA9 and PyMT cells were grown onto glass coverslips, assayed for MVMp clustering in the absence (Cont) or presence of CytoB, and prepared for IF microscopy analysis. MVMp (red) was detected with a specific antibody, and actin filaments (white) using AlexaFluor 647-conjugated phalloidin. (B) Quantification of the number of viral clusters (as described in section 2.11) per cells for all the conditions from three experiments performed as described in A. (n=100, data are mean ± standard error of the mean measured from three independent experiments, * p<0.05; ** p<0.01).    96 4.2.2 MVMp accumulates at the leading edge of migrating cells, which are more susceptible to viral uptake In addition to the clustering of MVMp at the base of filopodia observed by EM (Fig. 4.1), when performing my MVMp clustering assay with both LA9 and PyMT cells followed by IF microscopy analysis, I also observed localization of MVMp clusters at the leading edge of both cell types, along lamellipodial/pseudopodial structures (Fig. 4.5A). To confirm this observation, I repeated these experiments and immunolabeled both MVMp and Arp 2/3 complex, a marker for the leading edge of migrating cells (Suraneni et al., 2012). As expected, these experiments revealed a close proximity of MVMp clusters with Arp 2/3 complexes in both cell types (Fig. 4.5B), which may be an indication of the role of the cell migration process in the early steps of MVMp infection.   To further investigate the involvement of cell migration in MVMp infection, I tested whether cells engaged in migration are more susceptible to infection. For this experiment, LA9 and PyMT cells were seeded at 100% confluence on glass coverslips, a wound was created, and the cells were processed for MVMp clustering assay (as described in section 2.4.2) but also tested for viral uptake (as described in section 2.4.3). As illustrated in Figure 4.6, the cells moving into the wound (allowed to migrate) displayed higher levels of MVMp clustering and greater MVMp uptake at 4 hours post-infection in both LA9 and PyMT cells. These results thus indicate that cell migration is important for efficient MVMp uptake, which could occur from the leading edge of the cell.    97  Figure 4.5. MVMp clusters at the leading edge of migrating cells. (A) Cells were assayed for MVMp clustering (as described in section 2.4.2) at a MOI of 8 and prepared for IF microscopy. MVMp (red) was detected using a specific anti-capsid antibody, and actin filaments (pseudocolored in white) were labeled with AlexaFluor 647-conjugated phalloidin. Arrows point to lamellipodia at the leading edge of the cells. (B) Cells were assayed for MVMp clustering and processed as described in A. MVMp particles (red) and Arp 2/3 complex (green) were detected using specific antibodies. Zoom panel shows high magnification images of areas marked with “Z”. Images shown are representative of three independent experiments.   98  Figure 4.6. MVMp uptake is increased in migrating cells. Cells were grown to 100% confluence, a wound was created and the cells were assayed for MVMp clustering (as described in section 2.4.2) or uptake (as described in section 2.4.3) before processing for IF microscopy. MVMp (red) was detected using a specific antibody, AlexaFluor 647-conjugated phalloidin (pseudocolored in white) was used to detect actin filaments, and DAPI (blue) was used to observe the nucleus. Images show half of a wound. The doted lines indicate the beginning of the wound. After the wound was created, the cells were allowed to re-attach for 3 h before infection with MVMp. Images shown are representative of three independent experiments.     99 4.2.3 MVMp infection increases with cell migration Next I tested whether promoting cell migration would increase MVMp infection. LA9 and PyMT cells were seeded onto culture dishes coated with fibronectin (FN, a promoter of cell migration) or with poly-l-lysine (poly-K, whose positive charge causes the cells to adhere tightly to the coverslips without stimulating cell motility) as a control, and assayed for IF and Western blot analysis of MVMp infection (24 h infection at 37 °C, as described in section 2.4.4) at a MOI of 4. In this experiment, the production of NS1 was again used as a measurement of MVMp infection. Wound healing assays first confirmed that the cells grown onto FN-coated surfaces migrate faster than those grown on poly-K-coated surfaces, but also showed that PyMT cells migrate much faster than LA9 cells (Fig. 4.7A and B). As illustrated in Figure 4.8, LA9 cells growing on FN-coated surfaces showed a higher percentage of NS1-positive cells by IF (Fig. 4.8, A and B), and greater NS1 expression by Western blot (Fig. 4.8C). In contrast, the percentage of NS1-positive PyMT cells observed by IF was about twice as high for cells grown on FN matrix as the one of cells grown on poly-K (Fig. 4.8, A and B), and the NS1 expression levels detected by Western blot were much higher for PyMT cells grown on FN-coated surface (Fig. 4.8C). Hence, the ECM protein FN promotes both cell migration and MVMp infection, particularly in PyMT cells. To ensure that the reduced MVMp infection in LA9 and PyMT cells grown on poly-K matrix was not due to a reduced cell proliferation (since MVMp requires cell entry into S-phase to establish infection, (reviewed in Cotmore et Tattersall, 2006), I performed cell proliferation assays with LA9 and PyMT cells grown on poly-K- or FN-coated plastic dishes (Fig. 4.9). This experiment revealed that LA9 cells grow slightly more rapidly on poly-K matrix compared to FN, whereas the proliferation rate of PyMT cells was similar in both conditions. Hence the increased MVMp infection on FN matrix is not the result of an increased cell proliferation.  100  Figure 4.7. Effect of poly-K and FN substrates on LA9 and PyMT cell migration. (A) Cells were grown onto plastic dishes coated with 10 µg/ml FN or 0.01% poly-K until 100% confluence, a wound was created and the healing was monitored at 12 h and 24 h after the creation of the wound using an inverted light microscope (as described in (Granovsky et al., 2000)). Pictures are representative for at least 3 independent experiments. (B) Quantification of the distance traveled by LA9 (white) and PyMT cells (grey) 12 h post-wound. Shown are the mean and standard error measured from 3 independent experiments. (n=60, data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05; ** p<0.01).    101  Figure 4.8. MVMp infection increases with cell migration. (A) LA9 and PyMT cells were grown onto glass coverslips coated with 10 µg/ml FN or 0.01% poly-l-lysine, assayed for MVMp infection (24 h at 37 °C) at a MOI of 4 and prepared for IF microscopy analysis. The viral protein NS1 was detected with a specific antibody and DAPI (blue) was used to observe the nucleus. (B) Quantification of the percentage of NS1 positive cells from three experiments performed as described in A. (n=1000, data are mean ± standard error of the mean measured from three independent experiments, * p< 0.05; ** p<0.01). (C) LA9 and PyMT cells were grown onto glass coverslips coated with 10 µg/ml FN or 0.01% poly-K, assayed for MVMp infection (24 h at 37 °C) at a MOI of 4 and prepared for Western blot analysis. The viral protein NS1 and the cellular lamin A/C (loading control) were detected using specific antibodies. LA9 and PyMT cells were also MOCK infected as a control.     102  Figure 4.9. Effect of poly-K and FN substrates on LA9 and PyMT cells proliferation. Cells were grown in plastic dishes for 48 h, detached by trypsin treatment, and the viable cells were counted using a hemocytometer after trypan blue staining. (Data are mean ± standard error of the mean measured from three independent experiments). ns: not significant.    103  I next hypothesized that FN could promote the cell surface binding and uptake of MVMp, as it increased MVMp infection in LA9 and PyMT cells (Fig. 4.8). I thus analyzed by FACS (as described in section 2.7) the MVMp binding and uptake in these cells. In this experiment, LA9 and PyMT cells were grown onto plastic dishes coated with FN or poly-K, infected with MVMp at a MOI of 8 for 4 h at 4 °C (binding control) or at 37 °C (uptake control), and prepared for FACS analysis (as described in section 2.7). As shown in Figure 4.10A, the binding of MVMp to the plasma membrane was not affected by the poly-K or FN matrix, neither in LA9 nor in PyMT cells. In contrast, the decrease in MVMp fluorescence intensity (at 37 °C) observed for LA9 and PyMT cells grown on poly-K was higher than the one observed in cells grown on FN-coated surfaces (Fig. 4.10B), indicating that MVMp uptake is slightly greater on poly-K than on FN matrix. Note that in this experiment, the cells were not permeabilized, and the decrease in viral fluorescence reflects the removal of the virus from the cells surface by endocytosis (since I could not find a way to completely remove the virus from the cell surface and thus measure accurately the virus that actually entered the cell).  My results indicate that promoting cell migration on FN matrix increases MVMp infection, and this was not the result of MVMp cellular uptake because cellular uptake was higher for cells grown on poly-K matrix than those grown on FN matrix. Hence I hypothesized that the post-entry trafficking of MVMp toward the nucleus (and thus MVMp ability to infect its target cells) could be affected when cells are grown on different matrices. To answer this question, LA9 and PyMT cells grown on glass coverslips coated with poly-K or FN were assayed for IF analysis of MVMp uptake (as described in section 2.4.3). MVMp was detected with a specific antibody, and the nucleus of the cells with DAPI. As shown in Figure 4.11, there was barely any difference in the MVMp accumulation at the nuclear periphery between LA9 cells grown on poly-K or FN. In PyMT cells, however, the FN matrix promoted MVMp accumulation at the nuclear periphery  104 in comparison to poly-K. Moreover, the poly-K substrate appeared to delay the traffic of MVMp-containing vesicles towards the nucleus in PyMT cells, inducing accumulation of MVMp at the periphery of the cells. These observations indicate that the FN matrix promotes MVMp infection by increasing the proportion of MVMp particles that reach the nucleus of target cells.     105  Figure 4.10. FACS analysis of MVMp cell binding and uptake on different substrates. (A) Cells were grown on plastic dishes coated with 10 µg/ml FN or 0.01% poly-K, incubated with MVMp at a MOI of 8, and prepared for FACS analysis of MVMp binding (as described in section 2.7). MVMp was detected with a specific antibody. As controls, the cells were incubated with no antibody (Auto-fluorescence), or the primary antibody was omitted (No primary ab). (B) Cells were grown as described above, infected with MVMp at MOI of 8, and prepared for FACS analysis of MVMp uptake (as described in section 2.7). The level of uptake is reflected by the decrease in MVMp fluorescence intensity, as the cells were not permeabilized and only the virus still present at the cell surface was detected. Graphs shown are representative of three independent experiments.   106  Figure 4.11. FN matrix increases MVMp accumulation at the nuclear periphery in PyMT cells. Cells were grown onto glass coverslips coated with 10 µg/ml FN or 0.01% poly-K, assayed for MVMp uptake at a MOI of 8 and prepared for IF microscopy (as described in section 2.3.1). MVMp (red) was detected using a specific antibody and DAPI (blue) was used to observe the nucleus. Images shown are representative of three independent experiments.    107 4.2.6 Epithelial-mesenchymal transition triggers MVMp infection EMT is one of the most common inducers of cancer cell migration (reviewed in Lamouille et al., 2014; Savagner, 2010; Thiery, 2002; Yang and Weinberg, 2008). EMT triggers a switch from a non-migrating epithelial to a highly migrating mesenchymal cellular phenotype, accompanied with a loss of cell-cell junctions via down-regulation of E-cad and an increase of FN and N-Cad expression to achieve efficient cell migration. I found previously that cell migration plays a role in the MVMp infection (sections 4.2.1 to 4.2.3), and hypothesized that EMT could represent another regulator of MVMp infection and oncotropism. Supporting this possibility, the two main cellular models used in this thesis (LA9 and PyMT) clearly exhibit mesenchymal properties. To determine whether EMT plays a role in the MVMp replication cycle, I used a model derived from Ras-transformation of EpH4 mouse mammary epithelial cells (EpRas, a model cell line that easily undergoes EMT, (Oft et al., 1996)). EpRas cells were induced for EMT through a 96 h TGF-β1 treatment, and the success of EMT was assessed by Western blot analysis of E-cad and N-Cad expression. As shown in Figure 4.12A, this treatment induced a partial reduction of E-cad and an increase of N-Cad expression in EpRas cells, even though a proportion of the cells remained epithelial.  Next I tested whether TGF-β1-induced EMT would affect the MVMp infection efficacy in EpRas cells by detecting both the viral NS1 and the newly produced viral particles. For this experiment, EpRas cells were first treated with TGF-β1 as described above, and then infected with MVMp (MOI of 8) at 37 °C for 48 h before preparation for IF microscopy analysis (as described in section 2.3.1). As shown in Figure 4.12B, EpRas cells showed no sign of NS1 expression even at 48 h post-infection in the absence of TGF-β1 treatment, suggesting that MVMp cannot infect these epithelial cancer cells, even when Ras-transformed. In contrast,  108 inducing EMT allowed subsequent MVMp infection (Fig. 4.12B and C) and production of progeny virions (Fig. 4.13). However, only a limited number of cells could be infected (less than 20%; Figs. 4.12B and 4.13). This is most likely due to a limited portion of EpRas cells undergoing EMT (Fig. 4.12A) and reduced cell proliferation (data not shown). These findings further highlight the requirement for EMT and MCM in the MVMp early infection. Indeed, the lack of NS1 expression in EpRas cells without TGF-β1 treatment indicates that the limiting factors for MVMp infection in these conditions are located upstream of the viral replication step, potentially at virus cell entry or nuclear delivery stages. Moreover, since MVMp requires cell entry into S-phase to establish infection (Cotmore et Tattersall, 2006), I performed a cell proliferation analysis of EpRas cells. This experiment revealed that the division rate is considerably higher in EpRas compared to LA9 cells (Fig. 4.14), and thus the resistance to MVMp infection observed in EpRas cells is not related to a reduced cell replication. These are indications that the TGF-β1-induced permissivity for MVMp in EpRas cells is most likely related to the effect of TGF-β1 on cell differentiation rather than on cell proliferation.    109  Figure 4.12. EMT triggers MVMp infection in EpRas cells. (A) Western blot analysis of EMT markers in EpRas cells after a 96 h incubation in the presence/absence of TGF- β1. (B) EpRas cells were induced for EMT and assayed for MVMp infection (48 h at 37 °C) at a MOI of 8 before preparation for IF microscopy analysis (as described in section 2.3.1). NS1 (magenta) was detected using a specific antibody, AlexaFluor 647-conjugated phalloidin (pseudocolored in white) was used to detect actin filaments, and DAPI (blue) was used to observe the nucleus. (C) Quantification of the percentage of NS1 positive cells from three experiments performed as described in A. (n=1000, data are mean ± standard error of the mean measured from three independent experiments, *** p<0.005).   110  Figure 4.13. EMT permits MVMp replication in EpRas cells. EpRas cells were induced for EMT and assayed for MVMp infection (48 h at 37 °C) before preparation for IF microscopy analysis (as described in section 2.3.1). MVMp (red) was detected using a specific antibody, AlexaFluor 647-conjugated phalloidin (pseudocolored in white) was used to detect actin filaments, and DAPI (blue) was used to observe the nucleus. Arrowheads point at cells with MVMp progeny. Images shown are representative of three independent experiments.    111    Figure 4.14. Cell proliferation analysis. Cells were grown in plastic dishes for 48 h, detached by trypsin treatment, and the viable cells were counted using a hemocytometer after trypan blue staining (data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01).    112 4.3 Discussion Many parameters may contribute to the MVMp infection, and in Chapter 3, I identified some of them, Gal-3 and Mgat5. Here I show that MVMp relies on many aspects of the MCM to infect its target cells. My combined EM, IF microscopy, and FACS analysis revealed that MVMp requires cell protrusions (filopodia and lamellipodia/pseudopodia) to cluster at the leading edge of migrating LA9 and PyMT cells. I also demonstrated that cells engaged in migration are more susceptible to MVMp uptake, and that promoting cell migration on a FN matrix increases MVMp infection. Moreover, I found that EMT triggers MVMp infection in non-permissive EpRas epithelial cells.   I demonstrated that the first steps of MVMp infection include the binding and subsequent clustering of the virus at the leading edge of migrating cells. It had been shown previously by Linser et al. (1977) that MVMp accumulates at the base of filopodia just before endocytosis in LA9 cells (Linser et al., 1977). I observed the same phenomenon during EM analysis in both LA9 and PyMT cells (Fig. 4.1), suggesting that it is not limited to one cell line. Furthermore, my IF analysis showed that the clustering of MVMp at the base of filopodia, and thus at the leading edge of migrating cells, is temperature-dependent. This indicates that the membrane diffusion of the MVMp receptor(s) toward the leading edge is an active mechanism that is not controlled by the virus itself, or by ECM proteins such as Gal-3. This concept is supported by the discovery that another parvovirus, the CPV, is only able to bind and crosslink a limited number of cell surface receptors, and thus diffuses randomly at the cell surface until it encounters a clathrin-coated pit for subsequent endocytosis (Cureton et al., 2012). Intriguingly, the dynamics of MVMp clustering, faster in PyMT compared to LA9 cells (Fig. 4.3), suggests that this virus might use different cell surface receptors in these different cells. Indeed, it has been shown for  113 example that lateral diffusion of integrins within the plasma membrane is hampered by the mesh of cortical actin (Lepzelter and Zaman, 2010), and is thus slower than the highly dynamic diffusion of GPI-anchored proteins which already occurs at room temperature (Owen et al., 2009). Note that both integrins and GPI-AP can be glycosylated and carry sialic acids (the MVMp receptor, (Halder et al., 2014; Nam et al., 2006)), and thus represent potential cell surface receptors for MVMp.   During my IF analysis, I showed that inhibition of actin polymerization with CytoB prevents MVMp clustering in both LA9 and PyMT cells (Fig. 4.4), suggesting a requirement for cell protrusions (filopodia and lamellipodia/pseudopodia) in this viral clustering event, in agreement with my previous observations that MVMp accumulates at the base of filopodia (Fig. 4.1) and along lamellipodia (Fig. 4.5A). Another parvovirus that shows a similar phenomenon is the canine parvovirus, which binds to filopodia of canine cells as clusters of several virions that then accumulate at the base of filopodia already at 5 min post-infection (Harbison et al., 2009). Other viruses have also been observed associated with cell protrusions. For example, Semliki Forest virus binds to microvilli before entering cells by endocytosis (Helenius et al., 1980). More recently, using single particle imaging and fluorescently labeled viruses it has been documented that viruses such as, Murine leukemia virus (Lehmann et al., 2005), human papillomavirus type 16 (Schelhaas et al., 2008), vaccinia virus (Mercer and Helenius, 2008), adenovirus type 2 (Burckhardt and Greber, 2009), and herpes simplex virus type-1 (Oh et al., 2010), interact with filopodia and move (or “surf”) down these protrusions to reach the cell body for entry into their target cells. While some of these viruses surf down filopodia as single particles, MVMp might move down filopodia as clusters of few virions, because we observed large clusters at the tip of filopodia (Fig. 1).   114  While some viruses rely on filopodia formation to establish infection, other viruses induce filopodia formation. For example, cells respond to herpes simplex virus type-1 infection by promoting filopodia formation (Oh et al., 2010), while high titers of adeno-associated virus type 2 trigger filopodia formation to allow viral endocytosis (Nonnenmacher and Weber, 2011). But unlike these viruses, MVMp does not seem to induce filopodia because we never noticed any difference in the number/appearance of filopodial extensions in the absence/presence of MVMp, even at higher virus concentrations. Instead, it appears that MVMp infection depends on these structures for the virus to cluster before entering the cell.  I next showed that LA9 and PyMT cells engaged in migration during wound healing are more susceptible to MVMp uptake (Fig. 4.6). This further supports the requirement of directed cell migration and cell protrusions in the MVMp early infection, since fully confluent cells (in which MVMp uptake is very limited) obviously have no leading edge and are not able to form filopodia or lamellipodia/pseudopodia. A comparable mechanism has been reported for the human papillomavirus 16 (HPV-16), which uses surface transport along filopodia to infect epithelial cells (Schelhaas et al., 2008). In contrast to MVMp, however, HPV-16 is not oncotropic, but rather induces oncogenic transformation of infected tissues (reviewed in (Narisawa-Saito and Kiyono, 2007).  Consistent with my findings that migrating cells are more susceptible to MVMp uptake (Fig. 4.6), promoting LA9 and PyMT cell migration on a FN matrix (Fig. 4.7) increased permissivity to MVMp infection in comparison to poly-K (Fig. 4.8). This was rather limited in LA9 cells compared to PyMT cells, potentially because LA9 cells are fibroblasts and secrete FN constitutively. Furthermore, this result was somewhat unexpected as I found that MVMp uptake is more efficient when cells are grown on poly-K matrix (Fig. 4.10B). Nevertheless, viral uptake does not always lead to productive infection. It is now well accepted that the ability of MVMp to  115 reach the nucleus is directly dependent on both endosomal acidification (to escape late endosomes (Mani et al., 2006)) and the ubiquitin/proteasome system (Ros and Kempf, 2004). Hence, it is possible that FN induces ubiquitinylation and lysosomal degradation of cell surface receptors used by MVMp (Lobert et al., 2010), and thus promotes MVMp infection. In this model, the FN matrix likely increases the proportion of virus that enters the acidifying endocytic compartments, whereas poly-K simply induces direct recycling of the virus at the plasma membrane. This hypothesis is supported by my observation during IF microscopy analysis that MVMp accumulation at the nuclear periphery is greater in PyMT cells grown on FN-coated coverslips (Fig. 4.11). Hence, it appears that cell migration on a FN matrix is required for efficient MVMp trafficking toward the nucleus and subsequent infection.   I also found that induction of EMT with TGF-β1 treatment in EpRas cells rendered these cells susceptible to MVMp infection (Fig. 4.12). TGF-β1 is not just involved in cellular differentiation (EMT), but also affects cell replication via the cell cycle regulator p21 (Seoane, 2006). However, TGF-β1 has anti-proliferative properties on epithelial cells (Seoane, 2006), thus one would expect it to hamper rather than promote MVMp infection; because MVMp requires cell division to replicate its genome. Hence, as I showed that cell migration is required for early events of the MVMp replication cycle, it is probable that TGF-β1 triggered permissivity to MVMp infection in EpRas cells (Fig. 4.12) through EMT induction.  In this context, it would appear like the process of MCM is a “minimum required” for MVMp infection, and it might be no coincidence that LA9 and PyMT cells (both of which are highly permissive to MVMp infection) exhibit a mesenchymal phenotype. In fact, to my knowledge the vast majority of cell lines used for MVMp studies are fibroblastoid cells, a typical mesenchymal-like type of cell. In addition, since EpRas cells proved extremely resistant to MVMp infection (Fig. 4.13) even though the division rate of these cells was considerably high  116 (Fig. 4.14), one could speculate that MVMp is not able to infect cancer cells with an epithelial phenotype, no matter how fast these replicate. Although this has to be probed by testing a large number of cells, my preliminary results with Mgat5-/-, MDCK, and EpH4 cells, which proliferate very rapidly and have an epithelial phenotype, indicate that these cells were also resistant to MVMp infection (data not shown). An explanation for these results could be that MVMp requires the signaling activity that takes place at the leading edge of migrating mesenchymal cells, and that involves endocytosis of plasma membrane receptors for degradation in lysosomes in a FN-dependent manner (as described in section 1.3.5). This is also supported by my previous findings that Gal-3 and Mgat5 (Chapter 3) are required for efficient MVMp infection, for both of these proteins were directly implicated in the MCM process by permitting the FA signaling at the leading edge of migrating cells (reviewed in (Goetz, 2009). Yet the resistance of these epithelial cells to MVMp infection could also reside in their lack of Gal-3 or FN expression, since we found that both Gal-3 (Chapter 1) and FN (Fig. 4.8) play a role in MVMp infection.  In summary, the results presented in this chapter demonstrate that cell migration is another determinant of MVMp infection. Cell migration being a key factor in many cases of aggressive and metastatic cancers, it is possible that the highjack of the cell migration machinery by MVMp would also partially explain its intriguing oncotropic properties. More importantly, this model would not oppose, but rather complete the ones already established for the MVMp oncotropism (described in section 1.2.6), in which the loss of function of the cell cycle regulator p53 increases cell proliferation while inhibiting the antiviral response. Intriguingly, p53 was also recently associated with the cell migration process, since loss of p53 during tumorigenesis is apparently one more determinant of EMT (Muller et al., 2011). Finally, the combined p53, Gal-3, Mgat5 and MCM regulation of MVMp oncotropism would somewhat confer to this virus a considerably high degree of specificity for aggressive/invasive tumor cells.  117 Chapter 5: MVMp Cell Entry Mechanisms  5.1 Introduction In contrast to other parvoviruses, the MVMp cellular entry mechanism(s) remains elusive. Nevertheless, my results presented in Chapter 3 indicate that Gal-3 is necessary for efficient MVMp cell entry, and Gal-3 has been several times associated with endocytosis of cell surface receptors (Furtak et al., 2001; Gao et al., 2012; Goetz et al., 2008; Lajoie et al., 2007; Partridge et al., 2004). Moreover, I showed in Chapter 4 that MVMp clusters at the leading edge of migrating LA9 and PyMT cells, and that cell migration promotes MVMp infection. Because various endocytic mechanisms take place at the leading edge of migrating cells to allow the turnover and degradation of FAs, I next hypothesized that MVMp cell entry could occur in this area.  In this chapter, I aimed to elucidate the endocytosis mechanism used by MVMp to enter LA9 and PyMT cells. Using a combination of EM, IF microscopy and FACS, I found that MVMp is endocytosed from the base of filopodia and from cell-ECM contact sites at the leading edge of migrating LA9 and PyMT cells, via at least three different entry pathways. Intriguingly, each of these pathways, which include clathrin-, caveolin- as well as clathrin-independent carriers (CLICs)-mediated endocytosis, have been associated with the process of FA disassembly in migrating cells. Hence, I propose that MVMp highjacks the turnover of plasma membrane and FAs during cell migration to establish infection, and that it can enter cells via a variety of endocytosis mechanisms.   118 5.2 Results 5.2.1 MVMp cellular entry occurs in proximity to focal adhesions The results described in Chapter 4 show that many crucial parameters of the MCM process regulate MVMp early infection. Moreover, the observation that MVMp clusters at the leading edge of migrating LA9 and PyMT cells (Fig. 4.5) indicate that this virus might be endocytosed in proximity to FA sites, which play a pivotal role during cell migration. This hypothesis is further supported by my discovery that MVMp infectivity is abolished in Mgat5-/- cells (Fig. 3.8), since it was reported that the cell migration defect in these cells results from their inability to form proper FAs (Goetz et al., 2008). Various cellular proteins are recruited to FAs during cell migration (as described in section 1.3.2), and I next analyzed the proximity of MVMp clusters with two of these FA markers, paxillin (Pax) and α5-integrin. For these experiments, LA9 and PyMT cells were assayed for MVMp clustering (as described in section 2.4.2) and prepared for IF microscopy using antibodies for MVMp and Pax, or the MVMp clustering assay was performed in cells that were first transfected with α5-integrin-GFP. As shown in Figure 5.1A, there was a clear proximity between MVMp clusters and both markers of FAs, indicating that MVMp cell entry could take place at cell-ECM contact sites, as already reported by Linser et al., in 1979. This observation was confirmed by EM visualization of cells assayed for MVMp clustering and sectioned vertically (i.e perpendicular to the cell monolayer) to allow observation of FA sites and ECM. Using this approach, I visualized vesicles containing MVMp particles forming at cell-ECM contact sites (Fig. 5.1B, right panel), confirming my IF studies (Fig. 5.1A). This analysis also revealed vesicles containing MVMp particles at the base of filopodia (Fig. 5.1B, left panel). Thus, in migrating cells MVMp appears to be endocytosed during the turnover of plasma membrane or FAs, either at the base of filopodia, or from FA sites.   119  Figure 5.1. MVMp cellular entry occurs in proximity to focal adhesions. (A) Cells were transfected (α5-GFP panel) or not (Pax panel) for 24 h with a α5-integrin-GFP (green) construct, assayed for MVMp clustering (as described in section 2.4.2), and prepared for IF microscopy. MVMp (red) and paxillin (pseudocolored in green) were detected using specific antibodies. Zoom panel shows high magnification images of areas marked with “Z”. (B) Electron micrographs of MVMp endocytosis at the leading edge of migrating cells and in proximity to FAs. Cells were assayed for MVMp clustering and prepared for EM. Images were acquired using a TEM after ultrathin cross sectioning (i.e section perpendicular to the cell monolayer) and positive staining. Arrowheads point to endocytic vesicles containing MVMp. FP: filopodia. ECM: extracellular matrix. Images shown are representative of three independent experiments.  120 5.2.2 MVMp can use a variety of endocytic pathways My EM analysis of MVMp-infected cells revealed virion internalized in vesicles that have the hallmark of clathrin-coated pits (Fig. 5.1B). To verify this observation, LA9 and PyMT cells were again assayed for MVMp clustering (as described in section 2.4.2), and prepared for IF microscopy analysis (as described in section 2.3.1) using a clathrin-specific antibody. Because some clathrin-independent entry pathways are lipid raft-mediated, I also performed IF experiments detecting lipid raft using FITC-conjugated cholera toxin B subunit (CtxB) in cells assayed for MVMp clustering. As illustrated in Figure 5.2, there was a clear co-localization of MVMp with clathrin as expected, but also with CtxB, indicating that in LA9 and PyMT cells MVMp uses both CME and lipid-raft mediated endocytosis.  En-face EM sections of both LA9 and PyMT cells after the viral clustering assay also revealed multiple MVMp particles endocytosed in clathrin-coated vesicles (Fig. 5.3, left panels). Nonetheless, single MVMp particles were also observed in flask-shaped vesicles (Fig. 5.3, middle panels), but also in elongated tubular compartments with the appearance of CLICs (Fig. 5.3, right panels). These results clearly indicate that MVMp can use various endocytic mechanisms.   To verify that MVMp uses several endocytic pathways, I next studied MVMp uptake in the presence of various drug inhibitors of endocytosis using IF and FACS analysis. For IF, LA9 and PyMT cells grown on glass coverslips were infected with MVMp at a MOI of 8 for 4 h at 37 °C in the presence bafA1 in addition to chlorpromazine (CPZ, an inhibitor of clathrin-mediated endocytosis; (Wang et al., 1993)), genistein (an inhibitor of caveolar endocytosis; (Pelkmans et al., 2002)) Dynasore (an inhibitor of dynamins; (Macia et al., 2006)), or DMSO (control), and prepared for IF microscopy analysis (as described in section 2.3.1). As shown in Figure 5.4, CPZ  121 and genistein partially reduced MVMp uptake (and accumulation at the nuclear periphery) in both LA9 and PyMT cells. In contrast, the presence of Dynasore during infection completely prevented MVMp uptake in LA9 cells: instead of the typical accumulation at the nuclear periphery (Fig. 5,4, Control panel), the virus accumulated at the cell periphery (Fig. 5.4, zoom panel).  To better quantify MVMp cellular uptake in the presence of drugs that inhibit several endocytic pathways, LA9 and PyMT cells grown onto plastic dishes were infected with MVMp at a MOI of 8 for 4 h at 4 °C (binding control) or at 37 °C in the presence of CPZ, genistein, Dynasore or DMSO (uptake control), and prepared for FACS analysis (as described in section 2.7). In this experiment, the cells were not permeabilized, and the decrease in viral fluorescence reflects the removal of the virus from the cells surface by endocytosis (since I could not find a way to completely remove the virus from the cell surface and thus measure accurately the virus that actually entered the cell). In agreement with my IF and EM studies (Fig. 5.2 to 5.4), I found that CPZ and genistein partially reduced the cellular uptake of MVMp in both LA9 and PyMT cells, but not completely (Fig. 5.5). In contrast to CPZ and genistein, Dynasore completely blocked MVMp cell entry in LA9 cells (Fig. 5.5). Yet, Dynasore did not completely prevent MVMp entry in PyMT cells, although it was more efficient in inhibiting MVMp uptake than CPZ or genistein in these cells (Fig. 5.5). Hence, these results indicate that MVMp is endocytosed in a dynamin-dependent manner using both clathrin-dependent and clathrin-independent endocytosis in LA9 cells, but in PyMT cells MVMp might use in addition a dynamin-independent endocytic pathway. Altogether, my EM, IF, and FACS analysis demonstrate that MVMp uses various endocytic pathways to enter its target cells.  122  Figure 5.2. IF analysis of MVMp endocytosis. (A) Cells were assayed for MVMp clustering (as described in section 2.4.2) at a MOI of 8 in the presence of FITC-conjugated CtxB (CtxB panel, green) or not (Clath panel), and prepared for IF microscopy. MVMp (red) and Clathrin (green) were detected using specific antibodies. Zoom panel shows high magnification images of areas marked with “Z”. Images shown are representative of three independent experiments.   123  Figure 5.3. MVMp can use various endocytic pathways as shown by EM. Cells were assayed for MVMp clustering (as described in section 2.4.2) at a MOI of 32 and prepared for EM analysis. Images were acquired using a TEM after en-face (i.e. section parallel to the cell monolayer) ultrathin sectioning and positive staining. Arrowheads point to MVMp particles. Shown are electron micrographs illustrating the presence of MVMp particles in clathrin-coated vesicles (left panels), in flask-shaped vesicles and in invaginations of the plasma membrane (middle panels), and in elongated tubular compartments (right panels). Images shown are representative of three independent experiments.   124  Figure 5.4. MVMp can use various endocytic pathways as shown by IF microscopy. LA9 and PyMT Cells were grown onto glass coverslips, assayed for MVMp uptake at a MOI of 8 in the presence of drug inhibitors of endocytosis (as described in section 2.4.3), and prepared for IF microscopy (as described in section 2.3.1). MVMp (red) was detected using a specific antibody, and DAPI was used to observe the nucleus. Zoom panel shows high magnification images of areas marked with “Z”. CPZ: chlorpromazine; Genist: genistein; Dyna: Dynasore. Images shown are representative of three independent experiments.    125  Figure 5.5. MVMp can use various endocytic pathways as shown by FACS. (A) LA9 and PyMT Cells were grown onto plastic dishes, assayed for MVMp uptake at a MOI of 8 in the presence of drug inhibitors of endocytosis (as described in section 2.4.3), and prepared for FACS analysis (as described in section 2.7). MVMp was detected using a specific antibody. The cells were not permeabilized, and the decrease in viral fluorescence reflects the removal of the virus from the cells surface by endocytosis. (B) Quantification of the percentage of MVMp uptake inhibition from three experiments performed as described in A (n=30000, data are mean ± standard error of the mean measured from three independent experiments, ** p<0.01). CPZ: chlorpromazine; Genist: genistein; Dyna: dynasore.    126 5.3 Discussion Many studies have characterized the DNA replication mechanism of MVMp and other parvoviruses. In contrast, less is known about the molecular details of earlier steps of MVMp infection. Understanding the mechanism of MVMp entry into target cells might help shed light on the oncotropic properties of this virus. In this chapter, I report my studies on the characterization of the mechanism of MVMp cell entry in LA9 mouse fibroblast and PyMT mouse epithelial mammary tumor cells.  I found that the clustering of MVMp particles at the leading edge of migrating LA9 and PyMT cells (Fig. 4.5) is followed by virus endocytosis in this areas. Moreover, MVMp clusters showed some co-localisation with markers of FAs (Fig. 5.1A), indicating a potential involvement of these structures in MVMp cell entry. This was further supported with my EM analysis of vertically sectioned cells assayed for MVMp clustering, in which I observed formation of MVMp-containing endocytic vesicles from the base of filopodia, but also from cell-ECM contact sites (Fig. 5.1B). These are indications that MVMp cell entry occurs in part during the process of FA disassembly. This concept is further supported by my previous finding that Mgat5-modified N-glycosylations are required for MVMp infection (Chapter 3), for Mgat-5 null cells were shown deficient in cell migration because of their inability to form and mature FAs (Goetz et al., 2008). Moreover, the disruption of microtubules with nocodazole, a common inhibitor of FA disassembly and thus cell migration (Stehbens and Wittmann, 2012), also hampers MVMp infection (Ros et al., 2002). Focal adhesions allow cell attachment to the ECM, and the coordination of FA turnover or degradation is necessary to achieve directed cell migration. Hence, since I found previously (Chapter 4) that MCM promotes MVMp infection, it is rather plausible that this virus takes advantage of FA disassembly to enter its target cells.  127  Early EM studies have visualized MVMp in clathrin-coated vesicles (Linser et al., 1977), and this entry pathway was thought to be the uptake mechanism for MVMp and others parvoviruses. Nevertheless, recent publications have documented that in addition to clathrin-mediated endocytosis, some parvoviruses can use internalization by CLICs, caveolae-dependent internalization and macropinocytosis (Bantel-Schaal et al., 2009; Boisvert et al., 2010; Nonnenmacher and Weber, 2011; Quattrocchi et al., 2012). Thus, MVMp could use any or several of the internalization mechanisms described for other parvoviruses. To address this in more detail I first confirmed that MVMp co-localizes with clathrin by IF microscopy, and that MVMp is found in clathrin-coated vesicles by EM. However, my EM analysis also revealed MVMp particles in flask-shaped vesicles and in elongated tubular compartments with the appearance of CLICs. These results clearly indicate that MVMp can use various endocytic mechanisms. This conclusion is also supported by my IF co-localization of MVMp with CtxB, indicating that MVMp uses both clathrin- and lipid-raft mediated endocytosis. To further investigate the involvement of clathrin, I performed IF and FACS analyses of MVMp uptake with CPZ (an inhibitor of clathrin-mediated endocytosis), and found that MVMp uptake was partially reduced in the presence of this inhibitor (Figs. 5.4 and 5.5). Thus, documenting that MVMp could enter its target cells not only by clathrin-dependent endocytosis, but also by additional routes.  To further define the endocytic pathways used by MVMp, I studied the effect of genistein, an inhibitor of caveolar endocytosis, and found that MVMp cellular uptake was also reduced, but not completely inhibited by genistein (Figs. 5.4 and 5.5), indicating that in addition to clathrin, caveolin could drive endocytosis of MVMp. To narrow down these mechanisms I tried Dynasore, an inhibitor of dynamins, and found slightly different results for the two cell lines I used. While Dynasore completely blocked the MVMp uptake in LA9 cells, its inhibitory effect was not complete for PyMT cells. These results indicate that both clathrin and caveolin drive the MVMp  128 endocytosis in a dynamin-dependent manner in LA9 cells, but in PyMT cells MVMp can use an extra dynamin-independent entry route.  It is well accepted that the leading edge of migrating cells hosts high levels of endocytosis to allow the recycling of plasma membrane or the degradation of mature FA complexes (reviewed in (Caswell et al., 2009; Nagano et al., 2012; Webb et al., 2002); these mechanisms can occur via clathrin-, caveolin-, as well as CLICs-mediated endocytosis (Ezratty et al., 2009; Howes et al., 2010; Shi and Sottile, 2008). Taken together, my findings that MVMp can use all of these internalization mechanisms, that MVMp clusters at the leading edge of migrating cells, and that MVMp infection is abolished in Mgat5-/- cells, suggest that in order to enter its target cells MVMp might take advantage of the various endocytic pathways available for the turnover of plasma membrane or the disassembly of FAs during cell migration.            129 Chapter 6: General Discussion and Future Perspectives  Little was known about the early infection mechanism of the MVMp, and its oncotropic behavior was also poorly documented, even though oncotropic parvoviruses represent new potential agents in the battle against cancer. In this thesis, I have shown for the first time how and where this virus enters its target cells, and I have also identified key players of its infection mechanism in different tissue-culture cells. Indeed, I have demonstrated that MVMp requires Gal-3 and Mgat5 expression to establish infection, and I have found a correlation between Gal-3 expression and MVMp infection in various human cancer cells. Moreover, I have shown that MVMp particles cluster at the leading edge of migrating LA9 and PyMT cells for subsequent endocytosis in proximity to FAs, and that the process of MCM promotes MVMp infection. Finally, I have shown that MVMp can use various endocytic mechanisms to enter LA9 and PyMT cells. Protein glycosylation and secreted ECM proteins such as Gal-1 have been connected to viral infections in the past, but only HPV-16 was shown to require the cell migration process for infection (Schelhaas et al., 2008). Below I will first discuss my proposed model of the complex interplay between Mgat5 and Gal-3 that regulates the MVMp cell entry, and then I will describe the involvement of the cell migration process in MVMp infection. Finally I propose a new model for the MVMp early infection mechanism, and discuss future directions that should be considered to further elucidate the molecular mechanisms of the early steps of MVMp infection that might improve our understanding of the MVMp oncotropism.   130 6.1 A key role for Gal-3 and Mgat5 in MVMp infection The results presented in Chapter 3 show that both Gal-3 and Mgat5 are necessary for an efficient MVMp cell entry and infection in LA9 and PyMT cells. The lattice formed by Gal-3 pentamers upon binding to Mgat5-modified N-glycosylations exposed by cell surface receptors modulates their membrane diffusion, clustering, activation, and subsequent recycling or degradation by endocytosis. This is not an isolated example as the variety of galectins and their affinity for different glycosylation patterns play pivotal roles in host/pathogen interactions, but also in various functions of the host immune system (reviewed in Leffler et al., 2004; Vasta, 2012).  Based on previous studies, Gal-3 has a higher affinity for β-galactoside chains whereas MVMp binds sialylated glycans preferentially (Halder et al., 2014; Nam et al., 2006). This leads to a model (Figure 6.1A) where both Gal-3 and MVMp bind to the same receptor(s) without competition, in agreement with my finding that Gal-3 is required for efficient MVMp uptake (Fig. 3.2 and 3.6). This would explain why MVMp binding to the cell surface was not affected during Gal-3 siRNA KD or in Mgat5-/- cells (Figs. 3.3 and 3.12). In this context, it is most likely the signaling activity of the MVMp receptor that is affected upon loss of Gal-3 and Mgat5 expression, rather than the ability of MVMp to bind this receptor. This possibility is further supported by previous reports showing that the lack of Mgat5-modified N-glycosylations prevents Gal-3 lattice formation and affects the signaling ability of cell surface receptors such as integrins and EGFR (reviewed in (Boscher et al., 2011). Hence, the limited MVMp uptake that I observed in Gal-3 KD (Fig. 3.2 and 3.6) and Mgat5-/- cells (Fig. 3.12) may be the result of reduced signaling activity and internalization of the MVMp receptor resulting from a limited stabilization of this receptor within signaling platforms at the leading edge of migrating cells.  131  Nonetheless, it is also possible that the MVMp receptor has no specific activity, and simply depends on the signaling of a co-receptor (Fig. 6.1B). In this case, the MVMp receptor would simply allow the virus to reach domains of the plasma membrane where the signaling activity (Gal-3 and Mgat5-dependent) of its co-receptor promotes endocytosis (at the leading edge in migrating cells). This is in fact the case of the human parvovirus B19, which requires activation of its co-receptor α5β1-integrin by extracellular FN to infect erythroleukemia cells (Weigel-Kelley et al., 2003). During IF experiments, I observed some co-localization of MVMp with α5-integrin-GFP at FAs just before viral endocytosis (Fig. 5.1), and FN-coating promoted MVMp infection in different cells (Fig. 4.8), indicating that α5β1-integrins could also be a co-receptor for MVMp. Moreover, the Gal-3 binding to Mgat5-modified N-glycans exposed by α5β1-integrins promotes their stabilization within FAs (Goetz, 2009), which is necessary for the subsequent FN-dependent integrin ubiquitinylation that occurs upon maturation of these FAs (Lobert et al., 2010). Hence, because MVMp also requires both endosomal acidification (Mani et al., 2006) and the ubiquitin-proteasome complex (Ros and Kempf, 2004) for nuclear delivery, the lack of infectivity in Mgat5-/- cells could be the result of insufficient ubiquitinylation of the MVMp receptor(s) (or co-receptors).   I found a correlation between Gal-3 expression and MVMp infection in several human cancer cells, suggesting a possible role for Gal-3 in the MVMp oncotropism. There was, however, one exception with the brain tumor cells LN18 and LN229, which displayed similar Gal-3 expression levels but different susceptibility to MVMp infection (Fig. 3.16). This indicates that the involvement of Gal-3 in the MVMp infection is cell type-dependent, which we could expect since Gal-3 plays different roles in different cells. For example, Gal-3 promotes cell migration and invasion of breast tumor cells (Boscher and Nabi, 2013), but regulates the transport of apical proteins in epithelial cells (Delacour et al., 2007), even though it does contribute to the  132 maintenance of cell polarity in both of these cases. Nevertheless, it is also possible that other limiting factors involved in later steps of MVMp infection are lacking in the cells with low Gal-3 and MVMp-permissivity.  In summary, I have identified Gal-3 and Mgat5 as key players of the MVMp infection cycle in several tissue culture cells, and found a correlation between Gal-3 expression and MVMp infection in human cancer cells. The complexity in the Gal-3 and Mgat5 function described in Figure 6.1 supports the necessity of identifying potential candidates of the MVMp receptor(s), as this will allow better understanding of the involvement of Gal-3 and its receptors (which might simply vary in different cells) in the MVMp infection. Animal experimentation of the Mgat5 and Gal-3 involvement in MVMp infection will be required to evaluate their potential role in the oncotropism of MVMp.   133  Figure 6.1. Models for the Gal-3 and MVMp binding to their receptors. MVMp particles and Gal-3 pentamers may bind different glycosylations on the same receptor (A), or different glycosylations on different receptors (B).   134 6.2 Involvement of mesenchymal cell migration in MVMp infection The results presented in Chapter 4 show that many parameters of the cellular migration process regulate the MVMp cell entry and early infection. MVMp particles rapidly cluster at the leading edge of migrating LA9 and PyMT cells (Fig. 4.5), and the MVMp cell entry occurs in proximity to FAs (Fig. 5.1). Furthermore, MVMp uptake was greater in cells allowed to migrate during wound healing assay (Fig. 4.6), and promoting cell migration on a FN matrix increased MVMp infection (Fig. 4.8). Lastly, induction of EMT in EpRas cells allowed subsequent MVMp infection and amplification (Fig. 4.12 to 4.13).  The mechanism employed by MVMp to cluster at the leading edge of the migrating cells (Fig. 4.5) remains unclear. Based on my data, this process is temperature-dependent (Fig. 4.3) and requires cell protrusions (filopodia and lamellipodia/pseudopodia) upon actin polymerization at the cell front (Fig. 4.4). Moreover, it is unlikely that the trafficking of MVMp particles (together with their receptors) on the plasma membrane all the way to the leading edge of the cell is driven by ECM proteins such as Gal-3 (which would technically induce receptor clustering at 4 °C), since I never observed significant MVMp clustering after 2 h incubation at 4 °C (Fig. 4.3). Hence, it is possible that similarly to CPV (Cureton et al., 2012), MVMp randomly follows the membrane diffusion of its receptor(s) until it reaches sites of endocytosis (at the leading edge of migrating cells in the case of MVMp). The Gal-3 lattice would then maintain the MVMp receptor(s) within such plasma membrane domains to optimize viral uptake. Nevertheless, the fast dynamic and the directionality of the MVMp clustering indicate that some intra-cellular partners (likely involved in cell migration) might contribute to this process. Identifying the MVMp receptor(s) would provide further insight into the MVMp clustering mechanism.  135  Surprisingly, this is the first time that an oncotropic virus is shown to rely on the cell migration for infection. Nonetheless, it was proposed previously that the human papilloma virus 16 (HPV16) requires cell migration during wound healing for infection of epithelial cells, even though this virus is more oncogenic than oncotropic (Raff et al., 2013). HPV16 is a small non-enveloped virus that was also observed associated with filopodia by electron microscopy (Schelhaas et al., 2008), similarly to what I observed with MVMp (Fig. 4.1).   Unlike HVP16, however, MVMp seems unable to infect epithelial cells, since none of the cell lines with an epithelial phenotype (Mgat5-/-, MCF7, MDCK, EpH4, and EpRas cells) that I tested for their susceptibility to MVMp infection were ever permissive for viral replication. Even Ras-transformed epithelial cells resisted MVMp infection, whereas TGF-β1-induction of EMT in these cells allowed subsequent MVMp replication (Fig. 4.13). Hence, it appears possible that the MCM process represents a minimum requirement of the MVMp infection and oncotropism. Indeed, MVMp requires the S-phase of the cell cycle to establish infection (Rhode, 1973; Siegl and Gautschi, 1976), but the cells with the faster division rate (Mgat5-/-, MDCK, EpH4 and EpRas cells) were also the most MVMp-resistant during my experiments. This indicates that the MCM process is required upstream of the MVMp replication step, potentially for efficient trafficking of incoming virus toward the nucleus. My observation that MVMp accumulation at the nuclear periphery is reduced in Mgat5-/- compared to the highly mesenchymal PyMT and Mgat5-rescued cells (Fig. 3.11) supports this hypothesis. Similarly, the FN matrix increased MVMp accumulation at the nuclear periphery in PyMT cells (Fig. 4.11).  The FN matrix also promoted MVMp infection in PyMT cells (Fig. 4.8), and this was not a result of FN-stimulated cell proliferation since FN did not affect the cell growth (Fig. 4.9). It has been shown that FN mediates the attachment and cell entry of Gammaretrovirus and the infectious hematopoietic necrosis virus (IHNV) by a mechanism involving direct binding of the  136 virus to FN, which results in cellular uptake of FN-bound viral particles (Beer and Pedersen, 2007; Liu and Collodi, 2002). In contrast, the FN matrix did not affect the binding of MVMp to the plasma membrane in LA9 nor PyMT cells, and the MVMp uptake was slightly greater when cells were grown on poly-K compared to FN matrix (Fig. 4.10). This indicates that the involvement of FN in the MVMp replication cycle takes place after the binding step, and before viral replication. As mentioned in section 1.1.2, MVMp requires both endosomal acidification (Mani et al., 2006) and the ubiquitin-proteasome complex (Ros and Kempf, 2004) to establish infection, and it has been shown that FN matrix promotes ubiquitinylation of integrins within FAs for proper cell migration (Lobert et al., 2010). Hence, it is possible that the greater MVMp infectivity in PyMT cells grown on FN matrix results from increased ubiquitinylation of the MVMp receptor(s) (or co-receptors). The limited difference regarding MVMp infection observed between LA9 (mouse fibroblast) cells grown on poly-K and FN matrix could be explained by the fact that fibroblast cells constitutively secrete FN (Grinnell and Feld, 1979; Mao and Schwarzbauer, 2005).  As shown in Figure 1.9, there are different types of migration for cancer cells, including group cell migration, mesenchymal and amoeboid cell migration (reviewed in Friedl and Wolf, 2003; Lammermann and Sixt, 2009; Theveneau and Mayor, 2013; Yamazaki et al., 2005).  Based on my findings presented in Chapter 4, it is already clear that MCM is required for an efficient MVMp cellular uptake and infection. In contrast, the group cell migration used by epithelial cells during wound healing does not seem to allow MVMp infection as I could never infect cells with an epithelial phenotype. In additional experiments, it would be interesting to test whether cancer cells with amoeboid cell migration are susceptible to MVMp infection. Indeed, this would further confirm (or not) the specific MCM implication in the MVMp infection cycle, and also give us an estimation of the potential use of MVMp as virotherapy against amoeboid-like cancers.  137 Intriguingly, the findings that parvovirus B19 cell entry and nuclear delivery in erythroleukemia cells only occur when these cells are allowed to attach to FN matrix (Weigel-Kelley et al., 2003) indicate that this virus may rely on amoeboid cell migration to establish infection. In contrast, the lack of FAs in the amoeboid migration model suggests that MVMp uptake and infection will be limited in amoeboid cells, since MVMp particles cluster at FAs and seem to be endocytosed in part during the process of FA disassembly (Fig. 5.1).  In summary, I have discovered that the MVMp early infection mechanism is regulated by various parameters of the MCM process in different tissue culture cells. MVMp requires filopodia formation to cluster at the leading edge of LA9 and PyMT cells; cell migration during wound healing is necessary for efficient MVMp uptake; promoting cell migration on a FN matrix increases MVMp replication; and EMT rendered the non-permissive epithelial EpRas cells susceptible to MVMp infection. Given the role of MCM in the establishment of cancer metastasis, I propose that this could be another determinant of the MVMp oncotropism. Nonetheless, animal experimentation will be necessary to validate this model.  6.3 MVMp cell entry  As described in section 1.1.2, viruses employ many different cellular entry mechanisms to infect their target cells. Similar to other viruses, I have demonstrated that MVMp can use various cell entry pathways in LA9 and PyMT cells. My IF analysis revealed clear co-localisation of MVMp with clathrin and CtxB during early viral entry (Fig. 5.2), and my EM analysis allowed visualization of MVMp particles within distinct endocytic vesicles that exhibit the hallmark of clathrin-, caveolin-, and CLICs- mediated endocytosis (Fig. 5.3). Furthermore, IF and FACS analysis of the MVMp cellular uptake in the presence of drug inhibitors of endocytosis showed  138 that MVMp cell entry is clathrin-, caveolin- and dynamin-dependent in LA9 cells, but also partially dynamin-independent in PyMT cells (Figs. 5.4 and 5.5). Hence, the entry mechanism of this virus (and likely other viruses) is cell-type dependent, and MVMp is so far one of the rare viruses reported to use both caveolin- and CLICs-mediated endocytosis. Notably, it has been documented recently that Gal-3 plays a role in the formation of CLICs (Lakshminarayan et al., 2014), and I showed in Chapter 3 that Gal-3 is required for efficient cell entry of MVMp. Only few viruses have been reported to use caveolar endocytosis (reviewed in (Mercer et al., 2010). Recently another parvovirus, the AAV2, was shown to use CLICs mediated endocytosis for viral transduction (Nonnenmacher and Weber, 2011), in contrast to the CPV and human parvovirus B19, which enter cells via clathrin-mediated endocytosis (Parker and Parrish, 2000; Quattrocchi et al., 2012).  Unlike some other viruses however, it appears that MVMp is not able to actively control its mechanism of endocytosis, otherwise the MVMp binding to its plasma membrane receptor(s) would be sufficient for efficient viral uptake. Instead, I found that the cell migration was required for significant MVMp uptake in LA9 and PyMT cells (Fig. 4.6). The reason for this could reside in the small size of the MVMp capsid (~26 nm in diameter), which is unlikely to induce major receptor clustering and activation at the plasma membrane, in contrast to bigger viruses such as SV40 or Influenza A, which can apparently crosslink their cell-surface receptors to actively trigger viral endocytosis (reviewed in Grove and Marsch, 2011). This concept is also supported by the report that one CPV capsid can only crosslink a maximum of five cell surface receptors (Cureton et al., 2012), and that this virus just randomly follows the diffusion of its receptors within the plasma membrane until it encounters clathrin-coated pits for endocytosis. In a similar way, it seems like MVMp follows its receptor(s) all the way to the leading edge of migrating cells for endocytosis in proximity to FA sites (Fig. 5.1). In the absence of leading edge, the viral  139 uptake is limited, in agreement with my finding that LA9 and PyMT cells allowed for migration during wound healing exhibit greater MVMp uptake (Fig. 4.6). These are also indications that MVMp requires the disassembly of FAs to enter the cell, and intriguingly it has been shown that this process is dependent on the same endocytic mechanisms (CME, caveolar endocytosis, and CLICs) as I observed for MVMp (Chao and Kunz, 2009; Ezratty et al., 2009; Howes et al., 2010; Nethe and Hordijk, 2011; Urra et al., 2012; Wang et al., 2013a). Finally, MVMp showed drastic co-localization with CtxB (a marker of lipid rafts, Fig. 5.2), and lipid rafts apparently control the cell migration by permitting FA disassembly in human melanoma cells (reviewed in (Wang et al., 2013a). In conclusion, my data demonstrate that MVMp highjacks the endocytic machinery available at the leading edge of migrating cells to enter these late using a variety of entry routes.  6.4 A new model for the MVMp early infection From the experiments presented in this thesis I would like to propose a new model for the MVMp early infection mechanism. As illustrated in Figure 6.2, in this model MVMp particles first bind a still unidentified glycosylated receptor(s) (carrying sialic acids, (Halder et al., 2014; Nam et al., 2006)) at the cell surface, and rapidly cluster at the leading edge of the migrating cell (Fig. 4.3) during protrusion of the plasma membrane (Figs. 4.1 and 4.4), where they are retained by the Gal-3 lattice. The virus is then endocytosed immediately at the base of filopodia or from FA sites (Fig. 5.1), where it takes advantage of the turnover of plasma membrane or the disassembly of FAs for cell entry. Finally, the internalized vesicles containing MVMp migrate along microtubules toward the nucleus of the cell, and escape to the cytoplasm upon endosomal maturation.   140  Figure 6.2. Model for the MVMp early infection steps. Four steps are illustrated labeled 1-5. (1) In the first step, MVMp binds to its receptor at the plasma membrane; (2) in a second step, MVMp particles use cell protrusions to accumulate at the leading edge of the cell, where they are retained by the Gal-3 lattice; this is followed by (3) MVMp endocytosis during FA disassembly and plasma membrane turnover; (4) trafficking of MVMp-containing endosomes along microtubules, and (5) MVMp escape from late endosomes to the cytoplasm.    141 6.5 Future directions There are many more experiments that would be interesting to perform with regard to the MVMp infection and oncotropism mechanisms. For example, the study of the Gal-3, Mgat5, and MCM roles in the MVMp oncotropism should be tested in animals to fully validate my model, and to improve the use of oncotropic parvoviruses in virotherapies. Moreover, the molecular function of Gal-3 and Gal-3-BP in the MVMp early infection could be further tested in tissue culture cells. Identifying the cell surface receptor(s) used by MVMp to attach, enter and infect its target cells would also provide even more insight into the MVMp early infection mechanism, and allow better understanding of the complex interplay that I identified between the cell migration process, MVMp, and Gal-3. In this section, I will discuss in more detail some experimental approaches that should be considered to address these issues.  6.5.1 In vivo analysis of MVMp oncotropism I found a correlation between Gal-3 expression and MVMp infection in human cancer cells, as well as an involvement of the MCM process in MVMp early infection. For in vivo analysis of MVMp oncotropism, mice could be grafted with either highly Gal-3 positive human tumors, or with highly mesenchymal and invasive cancers (carcinomas). This could be followed with injections of MVMp preparations (either intravenously or intratumoraly) and monitoring of the tumor growth. Moreover, radioactive-labeling of MVMp would allow in vivo tracking of the virus using single-photon emission computed tomography, and I could determine whether MVMp accumulates in the grafted tumors preferentially.  In the hypothetical situation where MVMp would show in-vivo oncolytic activities against Gal-3 positive tumors or invasive carcinoma, genetic engineering of the virus could be attempted.  142 For example, the MVMp genome could be modified to express cytokines (or GFP to detect MVMp-infected cells directly) in infected cancer cells and thus promote the anti-cancer immune response, similarly to OncoVex-GM-CSF, an oncolytic herpes virus that has been engineered to express granulocyte-macrophage colony-stimulating factor (Senzer et al., 2009).  6.5.2 Investigating the Gal-3 involvement in MVMp infection I found that Gal-3 is required for efficient MVMp uptake in different cells, but the molecular details of this mechanism remain unclear. One way to shed light on this mechanism would be to test whether adding purified Gal-3 to the medium during infection would promote MVMp cell entry. This could be easily measured by IF microscopy or FACS analysis of MVMp cellular uptake (in LA9 and PyMT cells) in the presence of increasing Gal-3 concentrations. Further experiments could be conducted to assess whether Gal-3 and MVMp are endocytosed together, since Gal-3 has been involved in the clustering and endocytosis of cell surface receptors (Furtak et al., 2001; Gao et al., 2012; Goetz et al., 2008; Lajoie et al., 2007; Partridge et al., 2004). In addition, it would be interesting to determine whether or not Gal-3 expression is increased during MVMp infection, as it is the case for Gal-9 during Influenza A and cytomegalovirus infections (McSharry et al., 2014; Katoh et al., 2014). Furthermore, since Gal-3 is associated with endocytosis mechanisms, it would be worth testing whether or not Gal-3 siRNA (or addition of extracellular Gal-3) affects the entry pathways used by MVMp using IF and EM analysis of MVMp uptake. The MVMp entry routes that I identified in Chapter 5 should also be further studied using siRNA KD of proteins specific for each of these entry pathways, followed by IF and EM analysis of MVMp uptake.    143 6.5.3 Investigating the Gal-3-BP involvement in MVMp infection For this analysis, I could test the ability of Gal-3-BP to induce aggregation of MVMp, as it was observed for the parvovirus AAV6 (Denard et al., 2012). Increasing Gal-3-BP concentrations could be added to MVMp in solution, and the MVMp aggregation could be assessed by EM after negative staining. Moreover, Gal-3-BP could be added together with MVMp during early infection of cultured cells, and the MVMp clustering could be assessed by IF microscopy. In addition, the effect of Gal-3-BP siRNA KD on MVMp binding, uptake and replication, could be tested by IF microscopy analysis in different cells. Finally, co-injection of MVMp with increasing Gal-3-BP concentrations in animal models could be attempted to assess the effect of Gal-3-BP on MVMp systemic diffusion and recognition by the immune system. In these experiments, radiolabeled MVMp particles could be tracked in situ using single-photon emission computed tomography to monitor systemic spread. MVMp recognition and destruction by the immune system could be followed after animal injection by analysis of anti-MVMp antibodies and virus concentrations in the serum, either by enzyme-linked immune-sorbent assay or Western blot.  6.5.4 Identification of the MVMp receptor(s) As stated above, although MVMp binds to sialic acids, the specific membrane protein(s) to which these residues are attached, and that MVMp binds to, is still unknown. In order to identify the MVMp receptor(s), the best experimental approach would be to incubate MVMp with LA9 and PyMT cells grown onto plastic dishes, crosslink MVMp particles to their receptor(s) using 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide, immunoprecipitate the virus using the D4H1 antibody after cell lysis, and analyze the resulting precipitates by mass spectrometry. The  144 potential candidates could then be studied during infection with MVMp by IF (for co-localization and co-trafficking analysis) and FACS (for binding analysis) experiments. One could also knock them down using siRNA to confirm their involvement in MVMp binding, uptake, and infection steps.  6.6 Concluding remarks In conclusion, this work provides evidence that MVMp early infection is regulated by several cellular factors including the ECM protein Gal-3, the Golgi enzyme Mgat5, and the MCM process, in addition to those already identified by other research groups. I have shown that MVMp depends on Gal-3 and Mgat5 expression to enter and infect LA9 and PyMT cells efficiently, and that the Gal-3 expression profile in various human cancer cells correlates with their susceptibility to MVMp infection. I have also shown that MVMp highjacks the MCM process to infect its target cells, as MVMp particles clustered at the leading edge of migrating cells for subsequent endocytosis in proximity to FAs via different entry pathways, all involved in the process of FA disassembly. To my knowledge, this is the first time the migrating and invasive behavior of cancer cells is associated to a viral infection, and this might not be the last.         145 References  Abu-Ghanem, S., G. Oberkovitz, D. Benharroch, J. Gopas, and E. Livneh. 2007. PKCeta expression contributes to the resistance of Hodgkin's lymphoma cell lines to apoptosis. Cancer Biol. 6:1375-1380. Adeyemi, R.O., M.S. Fuller, and D.J. Pintel. 2014. Efficient Parvovirus Replication Requires CRL4Cdt2-Targeted Depletion of p21 to Prevent Its Inhibitory Interaction with PCNA. PLoS Pathog. 10:e1004055. Adeyemi, R.O., S. Landry, M.E. Davis, M.D. Weitzman, and D.J. Pintel. 2010. Parvovirus minute virus of mice induces a DNA damage response that facilitates viral replication. PLoS Pathog. 6:e1001141. Allaume, X., N. El-Andaloussi, B. Leuchs, S. Bonifati, A. Kulkarni, T. Marttila, J.K. Kaufmann, D.M. Nettelbeck, J. Kleinschmidt, J. Rommelaere, and A. Marchini. 2012. Retargeting of rat parvovirus H-1PV to cancer cells through genetic engineering of the viral capsid. J Virol. 86:3452-3465. Ammayappan, A., K.W. Peng, and S.J. Russell. 2013. Characteristics of oncolytic vesicular stomatitis virus displaying tumor-targeting ligands. J Virol. 87:13543-13555. Anderson, B.D., T. Nakamura, S.J. Russell, and K.W. Peng. 2004. High CD46 receptor density determines preferential killing of tumor cells by oncolytic measles virus. Cancer Res. 64:4919-4926. Au, G.G., A.M. Lindberg, R.D. Barry, and D.R. Shafren. 2005. Oncolysis of vascular malignant human melanoma tumors by Coxsackievirus A21. Int J Oncol. 26:1471-1476. Bajzer, Z., T. Carr, K. Josic, S.J. Russell, and D. Dingli. 2008. Modeling of cancer virotherapy with recombinant measles viruses. J Theor Biol. 252:109-122. Bantel-Schaal, U., I. Braspenning-Wesch, and J. Kartenbeck. 2009. Adeno-associated virus type 5 exploits two different entry pathways in human embryo fibroblasts. J Gen Virol. 90:317-322. Bar, S., L. Daeffler, J. Rommelaere, and J.P. Nuesch. 2008. Vesicular egress of non-enveloped lytic parvoviruses depends on gelsolin functioning. PLoS Pathog. 4:e1000126. Bar, S., J. Rommelaere, and J.P. Nuesch. 2013. Vesicular transport of progeny parvovirus particles through ER and Golgi regulates maturation and cytolysis. PLoS Pathog. 9:e1003605. Barber, M., A. Murrell, Y. Ito, A.T. Maia, S. Hyland, C. Oliveira, V. Save, F. Carneiro, A.L. Paterson, N. Grehan, S. Dwerryhouse, P. Lao-Sirieix, C. Caldas, and R.C. Fitzgerald. 2008. Mechanisms and sequelae of E-cadherin silencing in hereditary diffuse gastric cancer. J Pathol. 216:295-306. Barczyk, M., S. Carracedo, and D. Gullberg. 2010. Integrins. Cell Tissue Res. 339:269-280. Bartlett, J.S., J. Kleinschmidt, R.C. Boucher, and R.J. Samulski. 1999. Targeted adeno-associated virus vector transduction of nonpermissive cells mediated by a bispecific F(ab'gamma)2 antibody. Nat Biotechnol. 17:181-186. Bashir, T., R. Horlein, J. Rommelaere, and K. Willwand. 2000. Cyclin A activates the DNA polymerase deltaµ-dependent elongation machinery in vitro: A parvovirus DNA replication model. Proc Natl Acad Sci. 97:5522-5527. Bayer, N., D. Schober, E. Prchla, R.F. Murphy, D. Blaas, and R. Fuchs. 1998. Effect of bafilomycin A1 and nocodazole on endocytic transport in HeLa cells: implications for viral uncoating and infection. J Virol. 72:9645-9655.  146 Beer, C., and L. Pedersen. 2007. Matrix fibronectin binds gammaretrovirus and assists in entry: new light on viral infections. J Virol. 81:8247-8257. Berns, K.I. 1990. Parvovirus replication. Microbiol Rev. 54:316-329. Bertram, S., A. Heurich, H. Lavender, S. Gierer, S. Danisch, P. Perin, J.M. Lucas, P.S. Nelson, S. Pohlmann, and E.J. Soilleux. 2012. Influenza and SARS-coronavirus activating proteases TMPRSS2 and HAT are expressed at multiple sites in human respiratory and gastrointestinal tracts. PloS One. 7:e35876. Bliokh, Z.L., L.V. Domnina, O.Y. Ivanova, O.Y. Pletjushkina, T.M. Svitkina, V.A. Smolyaninov, J.M. Vasiliev, and I.M. Gelfand. 1980. Spreading of fibroblasts in medium containing cytochalasin B: formation of lamellar cytoplasm as a combination of several functional different processes. Proc Natl Acad Sci. 77:5919-5922. Boisvert, M., S. Fernandes, and P. Tijssen. 2010. Multiple pathways involved in porcine parvovirus cellular entry and trafficking toward the nucleus. J Virol. 84:7782-7792. Boscher, C., J.W. Dennis, and I.R. Nabi. 2011. Glycosylation, galectins and cellular signaling. Curr Opin Cell Biol. 23:383-392. Boscher, C., and I.R. Nabi. 2013. Galectin-3- and phospho-caveolin-1-dependent outside-in integrin signaling mediates the EGF motogenic response in mammary cancer cells. Mol Biol Cell. 24:2134-2145. Brewer, C.F., M.C. Miceli, and L.G. Baum. 2002. Clusters, bundles, arrays and lattices: novel mechanisms for lectin-saccharide-mediated cellular interactions. Curr Opin Struct Biol. 12:616-623. Brownstein, D.G., A.L. Smith, R.O. Jacoby, E.A. Johnson, G. Hansen, and P. Tattersall. 1991. Pathogenesis of infection with a virulent allotropic variant of minute virus of mice and regulation by host genotype. Lab Invest. 65:357-364. Burckhardt, C.J., and U.F. Greber. 2009. Virus movements on the plasma membrane support infection and transmission between cells. PLoS Pathog. 5:e1000621. Burguete, T., M. Rabreau, M. Fontanges-Darriet, E. Roset, H.D. Hager, A. Koppel, P. Bischof, and J.R. Schlehofer. 1999. Evidence for infection of the human embryo with adeno-associated virus in pregnancy. Hum Reprod. 14:2396-2401. Burnett, E., and P. Tattersall. 2003. Reverse genetic system for the analysis of parvovirus telomeres reveals interactions between transcription factor binding sites in the hairpin stem. J Virol. 77:8650-8660. Calderwood, D.A., I.D. Campbell, and D.R. Critchley. 2013. Talins and kindlins: partners in integrin-mediated adhesion. Nat Rev Mol Cell Biol. 14:503-517. Campbell, I.D., and M.J. Humphries. 2011. Integrin structure, activation, and interactions. Cold Spring Harb Perspect Biol. 3. 10:1101. Cassady, K.A., and J.N. Parker. 2010. Herpesvirus vectors for therapy of brain tumors. Open Virol J. 4:103-108. Caswell, P.T., S. Vadrevu, and J.C. Norman. 2009. Integrins: masters and slaves of endocytic transport. Nat Rev Mol Cell Biol. 10:843-853. Cater, J.E., and D.J. Pintel. 1992. The small non-structural protein NS2 of the autonomous parvovirus minute virus of mice is required for virus growth in murine cells. J Gen Virol. 73 ( Pt 7):1839-1843. Chaffer, C.L., and R.A. Weinberg. 2011. A perspective on cancer cell metastasis. Science. 331:1559-1564. Chao, W.T., and J. Kunz. 2009. Focal adhesion disassembly requires clathrin-dependent endocytosis of integrins. Fed Eur Biochem Soc J. 583:1337-1343.  147 Chen, Y.Q., M.C. Tuynder, J.J. Cornelis, P. Boukamp, N.E. Fusenig, and J. Rommelaere. 1989. Sensitization of human keratinocytes to killing by parvovirus H-1 takes place during their malignant transformation but does not require them to be tumorigenic. Carcinogenesis. 10:163-167. Chiang, A.C., and J. Massague. 2008. Molecular basis of metastasis. N Engl J Med. 359:2814-2823. Chiu, C.G., S.S. Strugnell, O.L. Griffith, S.J. Jones, A.M. Gown, B. Walker, I.R. Nabi, and S.M. Wiseman. 2010. Diagnostic utility of galectin-3 in thyroid cancer. Am J Pathol. 176:2067-2081. Cohen, S., A.K. Marr, P. Garcin, and N. Pante. 2011. Nuclear envelope disruption involving host caspases plays a role in the parvovirus replication cycle. J Virol. 85:4863-4874. Collins, M.J. Jr., and J.C. Parker. 1972. Murine virus contaminants of leukemia viruses and transplantable tumors. J Natl Cancer Inst. 49:1139-1143. Collins, M.S., J.B. Bashiruddin, and D.J. Alexander. 1993. Deduced amino acid sequences at the fusion protein cleavage site of Newcastle disease viruses showing variation in antigenicity and pathogenicity. Arch Virol. 128:363-370. Cornelis, J.J., P. Becquart, N. Duponchel, N. Salome, B.L. Avalosse, M. Namba, and J. Rommelaere. 1988a. Transformation of human fibroblasts by ionizing radiation, a chemical carcinogen, or simian virus 40 correlates with an increase in susceptibility to the autonomous parvoviruses H-1 virus and minute virus of mice. J Virol. 62:1679-1686. Cornelis, J.J., N. Salome, C. Dinsart, and J. Rommelaere. 2004. Vectors based on autonomous parvoviruses: novel tools to treat cancer? J Gene Med. 6 Suppl 1:S193-202. Cornelis, J.J., N. Spruyt, P. Spegelaere, E. Guetta, T. Darawshi, S.F. Cotmore, J. Tal, and J. Rommelaere. 1988b. Sensitization of transformed rat fibroblasts to killing by parvovirus minute virus of mice correlates with an increase in viral gene expression. J Virol. 62:3438-3444. Cotmore, S.F., and P. Tattersall. 2006. A rolling-hairpin strategy: basic mechanism of DNA replication in the parvoviruses. In: Kerr JR, Cotmore SF, Bloom ME, Linden RM, Parrish CR, editors. Parvoviruses. London: Hodder Arnold. p. 171-88. Cotmore, S.F., and P. Tattersall. 2007. Parvoviral host range and cell entry mechanisms. Adv Virus Res. 70:183-232. Cureton, D.K., C.E. Harbison, E. Cocucci, C.R. Parrish, and T. Kirchhausen. 2012. Limited transferrin receptor clustering allows rapid diffusion of canine parvovirus into clathrin endocytic structures. J Virol. 86:5330-5340. Daya, S., and K.I. Berns. 2008. Gene therapy using adeno-associated virus vectors. Clin Microbiol Rev. 21:583-593. de Beco, S., F. Amblard, and S. Coscoy. 2012. New insights into the regulation of E-cadherin distribution by endocytosis. Int Rev Cell Mol Biol. 295:63-108. Delacour, D., C.I. Cramm-Behrens, H. Drobecq, A. Le Bivic, H.Y. Naim, and R. Jacob. 2006. Requirement for galectin-3 in apical protein sorting. Curr Biol. 16:408-414. Delacour, D., C. Greb, A. Koch, E. Salomonsson, H. Leffler, A. Le Bivic, and R. Jacob. 2007. Apical sorting by galectin-3-dependent glycoprotein clustering. Traffic. 8:379-388. Deleu, L., A. Pujol, S. Faisst, and J. Rommelaere. 1999. Activation of promoter P4 of the autonomous parvovirus minute virus of mice at early S phase is required for productive infection. J Virol. 73:3877-3885.  148 Denard, J., C. Beley, R. Kotin, R. Lai-Kuen, S. Blot, H. Leh, A. Asokan, R.J. Samulski, P. Moullier, T. Voit, L. Garcia, and F. Svinartchouk. 2012. Human galectin 3 binding protein interacts with recombinant adeno-associated virus type 6. J Virol. 86:6620-6631. Dennis, J.W., K.S. Lau, M. Demetriou, and I.R. Nabi. 2009a. Adaptive regulation at the cell surface by N-glycosylation. Traffic. 10:1569-1578. Dennis, J.W., I.R. Nabi, and M. Demetriou. 2009b. Metabolism, cell surface organization, and disease. Cell. 139:1229-1241. Deramaudt, T.B., D. Dujardin, F. Noulet, S. Martin, R. Vauchelles, K. Takeda, and P. Ronde. 2014. Altering FAK-Paxillin interactions reduces adhesion, migration and invasion processes. PloS One. 9:e92059. Doerig, C., B. Hirt, J.P. Antonietti, and P. Beard. 1990. Nonstructural protein of parvoviruses B19 and minute virus of mice controls transcription. J Virol. 64:387-396. Doerig, C., B. Hirt, P. Beard, and J.P. Antonietti. 1988. Minute virus of mice non-structural protein NS-1 is necessary and sufficient for trans-activation of the viral P39 promoter. J Gen Virol. 69 ( Pt 10):2563-2573. Dumic, J., S. Dabelic, and M. Flogel. 2006. Galectin-3: an open-ended story. Biochim Biophys Acta. 1760:616-635. Dupressoir, T., J.M. Vanacker, J.J. Cornelis, N. Duponchel, and J. Rommelaere. 1989. Inhibition by parvovirus H-1 of the formation of tumors in nude mice and colonies in vitro by transformed human mammary epithelial cells. Cancer Res. 49:3203-3208. Eichwald, V., L. Daeffler, M. Klein, J. Rommelaere, and N. Salome. 2002. The NS2 proteins of parvovirus minute virus of mice are required for efficient nuclear egress of progeny virions in mouse cells. J Virol. 76:10307-10319. El Bakkouri, K., C. Servais, N. Clement, S.C. Cheong, J.D. Franssen, T. Velu, and A. Brandenburger. 2005. In vivo anti-tumour activity of recombinant MVM parvoviral vectors carrying the human interleukin-2 cDNA. J Gene Med. 7:189-197. Engel, S., T. Heger, R. Mancini, F. Herzog, J. Kartenbeck, A. Hayer, and A. Helenius. 2011. Role of endosomes in simian virus 40 entry and infection. J Virol. 85:4198-4211. Ezratty, E.J., M.A. Partridge, and G.G. Gundersen. 2005. Microtubule-induced focal adhesion disassembly is mediated by dynamin and focal adhesion kinase. Nat Cell Biol. 7:581-590. Ezratty, E.J., C. Bertaux, E.E. Marcantonio, and G.G. Gundersen. 2009. Clathrin mediates integrin endocytosis for focal adhesion disassembly in migrating cells. J Cell Biol. 187:733-747. Fantozzi, A., and G. Christofori. 2006. Mouse models of breast cancer metastasis. Breast Cancer Res. 8:212. Fernandez-Garcia, B., N. Eiro, L. Marin, S. Gonzalez-Reyes, L.O. Gonzalez, M.L. Lamelas, and F.J. Vizoso. 2014. Expression and prognostic significance of fibronectin and matrix metalloproteases in breast cancer metastasis. Histopathology. 64:512-522. Flanagan, M.D., and S. Lin. 1980. Cytochalasins block actin filament elongation by binding to high affinity sites associated with F-actin. J Biol Chem. 255:835-838. Folgueras, A.R., A.M. Pendas, L.M. Sanchez, and C. Lopez-Otin. 2004. Matrix metalloproteinases in cancer: from new functions to improved inhibition strategies. Int J Dev Biol. 48:411-424. Fornarini, B., S. Iacobelli, N. Tinari, C. Natoli, M. De Martino, and G. Sabatino. 1999. Human milk 90K (Mac-2 BP): possible protective effects against acute respiratory infections. Clin Exp Immunol. 115:91-94.  149 Frantz, C., K.M. Stewart, and V.M. Weaver. 2010. The extracellular matrix at a glance. J Cell Sci. 123:4195-4200. Friedl, P., and K. Wolf. 2010. Plasticity of cell migration: a multiscale tuning model. J Cell Biol. 188:11-19. Furtak, V., F. Hatcher, and J. Ochieng. 2001. Galectin-3 mediates the endocytosis of beta-1 integrins by breast carcinoma cells. Biochem Biophys Res Commun. 289:845-850. Gao, X., D. Liu, Y. Fan, X. Li, H. Xue, Y. Ma, Y. Zhou, and G. Tai. 2012. The two endocytic pathways mediated by the carbohydrate recognition domain and regulated by the collagen-like domain of galectin-3 in vascular endothelial cells. PloS One. 7:e52430. Garcin, P., S. Cohen, S. Terpstra, I. Kelly, L.J. Foster, and N. Pante. 2013. Proteomic analysis identifies a novel function for galectin-3 in the cell entry of parvovirus. J Proteomics. 79:123-132. Garner, O.B., H.C. Aguilar, J.A. Fulcher, E.L. Levroney, R. Harrison, L. Wright, L.R. Robinson, V. Aspericueta, M. Panico, S.M. Haslam, H.R. Morris, A. Dell, B. Lee, and L.G. Baum. 2010. Endothelial galectin-1 binds to specific glycans on nipah virus fusion protein and inhibits maturation, mobility, and function to block syncytia formation. PLoS Pathog. 6:e1000993. Geletneky, K., I. Kiprianova, A. Ayache, R. Koch, Y.C.M. Herrero, L. Deleu, C. Sommer, N. Thomas, J. Rommelaere, and J.R. Schlehofer. 2010. Regression of advanced rat and human gliomas by local or systemic treatment with oncolytic parvovirus H-1 in rat models. Neuro Oncol. 12:804-814. Geletneky, K., J. Huesing, J. Rommelaere, J.R. Schlehofer, B. Leuchs, M. Dahm, O. Krebs, M. von Knebel Doeberitz, B. Huber, and J. Hajda. 2012. Phase I/IIa study of intratumoral/intracerebral or intravenous/intracerebral administration of Parvovirus H-1 (ParvOryx) in patients with progressive primary or recurrent glioblastoma multiforme: ParvOryx01 protocol. BMC Cancer. 12:99. Gerdes, H.H., and R.N. Carvalho. 2008. Intercellular transfer mediated by tunneling nanotubes. Curr Opin Cell Biol. 20:470-475. Gialeli, C., A.D. Theocharis, and N.K. Karamanos. 2011. Roles of matrix metalloproteinases in cancer progression and their pharmacological targeting. Fed Eur Biochem Soc J. 278:16-27. Girod, A., M. Ried, C. Wobus, H. Lahm, K. Leike, J. Kleinschmidt, G. Deleage, and M. Hallek. 1999. Genetic capsid modifications allow efficient re-targeting of adeno-associated virus type 2. Nat Med. 5:1052-1056. Goetz, J.G. 2009. Bidirectional control of the inner dynamics of focal adhesions promotes cell migration. Cell Adh Migr. 3:185-190. Goetz, J.G., B. Joshi, P. Lajoie, S.S. Strugnell, T. Scudamore, L.D. Kojic, and I.R. Nabi. 2008. Concerted regulation of focal adhesion dynamics by galectin-3 and tyrosine-phosphorylated caveolin-1. J Cell Biol. 180:1261-1275. Granovsky, M., J. Fata, J. Pawling, W.J. Muller, R. Khokha, and J.W. Dennis. 2000. Suppression of tumor growth and metastasis in Mgat5-deficient mice. Nat Med. 6:306-312. Grassadonia, A., N. Tinari, I. Iurisci, E. Piccolo, A. Cumashi, P. Innominato, M. D'Egidio, C. Natoli, M. Piantelli, and S. Iacobelli. 2004. 90K (Mac-2 BP) and galectins in tumor progression and metastasis. Glycoconj J. 19:551-556. Grekova, S., R. Zawatzky, R. Horlein, C. Cziepluch, M. Mincberg, C. Davis, J. Rommelaere, and L. Daeffler. 2010. Activation of an antiviral response in normal but not transformed mouse cells: a new determinant of minute virus of mice oncotropism. J Virol. 84:516-531.  150 Grinnell, F., and M.K. Feld. 1979. Initial adhesion of human fibroblasts in serum-free medium: possible role of secreted fibronectin. Cell. 17:117-129. Grove, J., and M. Marsh. 2011. The cell biology of receptor-mediated virus entry. J Cell Biol. 195:1071-1082. Guetta, E., M. Mincberg, S. Mousset, C. Bertinchamps, J. Rommelaere, and J. Tal. 1990. Selective killing of transformed rat cells by minute virus of mice does not require infectious virus production. J Virol. 64:458-462. Guo, H.B., Y. Zhang, and H.L. Chen. 2001. Relationship between metastasis-associated phenotypes and N-glycan structure of surface glycoproteins in human hepatocarcinoma cells. J Can Res Clin Onc. 127:231-236. Guo, Z.S., S.H. Thorne, and D.L. Bartlett. 2008. Oncolytic virotherapy: molecular targets in tumor-selective replication and carrier cell-mediated delivery of oncolytic viruses. Biochim et Biophys Acta. 1785:217-231. Guy, C.T., R.D. Cardiff, and W.J. Muller. 1992. Induction of mammary tumors by expression of polyomavirus middle T oncogene: a transgenic mouse model for metastatic disease. Mol Cell Biol. 12:954-961. Haddad, D., N. Chen, Q. Zhang, C.H. Chen, Y.A. Yu, L. Gonzalez, J. Aguilar, P. Li, J. Wong, A.A. Szalay, and Y. Fong. 2012. A novel genetically modified oncolytic vaccinia virus in experimental models is effective against a wide range of human cancers. An Surg Onc. 19 Suppl 3:S665-674. Halder, S., S. Cotmore, J. Heimburg-Molinaro, D.F. Smith, R.D. Cummings, X. Chen, A.J. Trollope, S.J. North, S.M. Haslam, A. Dell, P. Tattersall, R. McKenna, and M. Agbandje-McKenna. 2014. Profiling of glycan receptors for minute virus of mice in permissive cell lines towards understanding the mechanism of cell recognition. PloS One. 9:e86909. Hallden, G., and G. Portella. 2012. Oncolytic virotherapy with modified adenoviruses and novel therapeutic targets. Exp Op Ther Tar. 16:945-958. Harbison, C.E., S.M. Lyi, W.S. Weichert, and C.R. Parrish. 2009. Early steps in cell infection by parvoviruses: host-specific differences in cell receptor binding but similar endosomal trafficking. J Virol. 83:10504-10514. Hasegawa, K., T. Nakamura, M. Harvey, Y. Ikeda, A. Oberg, M. Figini, S. Canevari, L.C. Hartmann, and K.W. Peng. 2006. The use of a tropism-modified measles virus in folate receptor-targeted virotherapy of ovarian cancer. Clin Cancer Res. 12:6170-6178. Heasman, S.J., and A.J. Ridley. 2008. Mammalian Rho GTPases: new insights into their functions from in vivo studies. Nat Rev Mol Cell Biol. 9:690-701. Helenius, A., J. Kartenbeck, K. Simons, and E. Fries. 1980. On the entry of Semliki forest virus into BHK-21 cells. J Cell Biol. 84:404-420. Hemminki, A. 2014. Oncolytic Immunotherapy: Where Are We Clinically? Scientifica (Cairo). 2014:862925. Henderson, N.C., A.C. Mackinnon, S.L. Farnworth, F. Poirier, F.P. Russo, J.P. Iredale, C. Haslett, K.J. Simpson, and T. Sethi. 2006. Galectin-3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci. 103:5060-5065. Hoon, J.L., W.K. Wong, and C.G. Koh. 2012. Functions and regulation of circular dorsal ruffles. Mol Cell Biol. 32:4246-4257. Hoshino, D., K.M. Branch, and A.M. Weaver. 2013. Signaling inputs to invadopodia and podosomes. J Cell Sci. 126:2979-2989. Howes, M.T., M. Kirkham, J. Riches, K. Cortese, P.J. Walser, F. Simpson, M.M. Hill, A. Jones, R. Lundmark, M.R. Lindsay, D.J. Hernandez-Deviez, G. Hadzic, A. McCluskey, R.  151 Bashir, L. Liu, P. Pilch, H. McMahon, P.J. Robinson, J.F. Hancock, S. Mayor, and R.G. Parton. 2010. Clathrin-independent carriers form a high capacity endocytic sorting system at the leading edge of migrating cells. J Cell Biol. 190:675-691. Hristov, G., M. Kramer, J. Li, N. El-Andaloussi, R. Mora, L. Daeffler, H. Zentgraf, J. Rommelaere, and A. Marchini. 2010. Through its nonstructural protein NS1, parvovirus H-1 induces apoptosis via accumulation of reactive oxygen species. J Virol. 84:5909-5922. Hubmacher, D., and S.S. Apte. 2013. The biology of the extracellular matrix: novel insights. Curr Opin Rheumatol. 25:65-70. Huttenlocher, A., and A.R. Horwitz. 2011. Integrins in cell migration. Cold Spring Harb Perspect Biol. 3:a005074. Ilkow, C.S., S.L. Swift, J.C. Bell, and J.S. Diallo. 2014. From scourge to cure: tumour-selective viral pathogenesis as a new strategy against cancer. PLoS Pathog. 10:e1003836. Inohara, H., and A. Raz. 1994. Identification of human melanoma cellular and secreted ligands for galectin-3. Biochem Biophys Res Com. 201:1366-1375. Janik, M.E., A. Litynska, and P. Vereecken. 2010. Cell migration-the role of integrin glycosylation. Biochim Biophys Acta. 1800:545-555. Johnson, F.B., L.B. Fenn, T.J. Owens, L.J. Faucheux, and S.D. Blackburn. 2004. Attachment of bovine parvovirus to sialic acids on bovine cell membranes. J Gen Virol. 85:2199-2207. Kaludov, N., K.E. Brown, R.W. Walters, J. Zabner, and J.A. Chiorini. 2001. Adeno-associated virus serotype 4 (AAV4) and AAV5 both require sialic acid binding for hemagglutination and efficient transduction but differ in sialic acid linkage specificity. J Virol. 75:6884-6893. Kanerva, A., G.J. Bauerschmitz, M. Yamamoto, J.T. Lam, R.D. Alvarez, G.P. Siegal, D.T. Curiel, and A. Hemminki. 2004. A cyclooxygenase-2 promoter-based conditionally replicating adenovirus with enhanced infectivity for treatment of ovarian adenocarcinoma. Gene Ther. 11:552-559. Katoh, S., M. Ikeda, H. Shimizu, K. Mouri, Y. Obase, Y. Kobashi, K. Fukushima, M. Hirashima, and M. Oka. 2014. Increased levels of plasma galectin-9 in patients with influenza virus infection. Tohoku J. Exp Med. 232:263-267. Kaufmann, B., A. Lopez-Bueno, M.G. Mateu, P.R. Chipman, C.D. Nelson, C.R. Parrish, J.M. Almendral, and M.G. Rossmann. 2007. Minute virus of mice, a parvovirus, in complex with the Fab fragment of a neutralizing monoclonal antibody. J Virol. 81:9851-9858. Kessenbrock, K., V. Plaks, and Z. Werb. 2010. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 141:52-67. Kim, S.H., J. Turnbull, and S. Guimond. 2011. Extracellular matrix and cell signalling: the dynamic cooperation of integrin, proteoglycan and growth factor receptor. J Endocrinol. 209:139-151. Kimsey, P.B., H.D. Engers, B. Hirt, and C.V. Jongeneel. 1986. Pathogenicity of fibroblast- and lymphocyte-specific variants of minute virus of mice. J Virol. 59:8-13. Kontou, M., L. Govindasamy, H.J. Nam, N. Bryant, A.L. Llamas-Saiz, C. Foces-Foces, E. Hernando, M.P. Rubio, R. McKenna, J.M. Almendral, and M. Agbandje-McKenna. 2005. Structural determinants of tissue tropism and in vivo pathogenicity for the parvovirus minute virus of mice. J Virol. 79:10931-10943. Koths, K., E. Taylor, R. Halenbeck, C. Casipit, and A. Wang. 1993. Cloning and characterization of a human Mac-2-binding protein, a new member of the superfamily defined by the macrophage scavenger receptor cysteine-rich domain. J Biol Chem. 268:14245-14249.  152 Kramer, F. 2013. Galetin-3: clinical utility and prognostic value in patients with heart failure. Res Rep Clin Card. 2013-4 13-22. Lachmann, S., J. Rommeleare, and J.P. Nuesch. 2003. Novel PKCeta is required to activate replicative functions of the major nonstructural protein NS1 of minute virus of mice. J Virol. 77:8048-8060.  Lajoie, P., J.G. Goetz, J.W. Dennis, and I.R. Nabi. 2009. Lattices, rafts, and scaffolds: domain regulation of receptor signaling at the plasma membrane. J Cell Biol. 185:381-385. Lajoie, P., E.A. Partridge, G. Guay, J.G. Goetz, J. Pawling, A. Lagana, B. Joshi, J.W. Dennis, and I.R. Nabi. 2007. Plasma membrane domain organization regulates EGFR signaling in tumor cells. J Cell Biol. 179:341-356. Lakshminarayan, R., C. Wunder, U. Becken, M.T. Howes, C. Benzing, S. Arumugam, S. Sales, N. Ariotti, V. Chambon, C. Lamaze, D. Loew, A. Shevchenko, K. Gaus, R.G. Parton, and L. Johannes. 2014. Galectin-3 drives glycosphingolipid-dependent biogenesis of clathrin-independent carriers. Nat Cell Biol. Lammermann, T., and M. Sixt. 2009. Mechanical modes of 'amoeboid' cell migration. Cur Op Cell Biol. 21:636-644. Lamouille, S., J. Xu, and R. Derynck. 2014. Molecular mechanisms of epithelial-mesenchymal transition. Nat Rev Mol Cell Biol. 15:178-196. Laukaitis, C.M., D.J. Webb, K. Donais, and A.F. Horwitz. 2001. Differential dynamics of alpha 5 integrin, paxillin, and alpha-actinin during formation and disassembly of adhesions in migrating cells. J Cell Biol. 153:1427-1440. Le Clainche, C., and M.F. Carlier. 2008. Regulation of actin assembly associated with protrusion and adhesion in cell migration. Physio Rev. 88:489-513. Leffler, H., S. Carlsson, M. Hedlund, Y. Qian, and F. Poirier. 2004. Introduction to galectins. Glycoconj J. 19:433-440. Lehmann, M.J., N.M. Sherer, C.B. Marks, M. Pypaert, and W. Mothes. 2005. Actin- and myosin-driven movement of viruses along filopodia precedes their entry into cells. J Cell Biol. 170:317-325. Lepzelter, D., and M.H. Zaman. 2010. Clustered diffusion of integrins. Biophys J. 99:L106-108. Levine, A.J., and M. Oren. 2009. The first 30 years of p53: growing ever more complex. Nat Rev. Can. 9:749-758. Li, X., J. Zhang, H. Gao, E. Vieth, K.H. Bae, Y.P. Zhang, S.J. Lee, S. Raikwar, T.A. Gardner, G.D. Hutchins, D. VanderPutten, C. Kao, and M.H. Jeng. 2005a. Transcriptional targeting modalities in breast cancer gene therapy using adenovirus vectors controlled by alpha-lactalbumin promoter. Mol Cancer Ther. 4:1850-1859. Li, X., Y.P. Zhang, H.S. Kim, K.H. Bae, K.M. Stantz, S.J. Lee, C. Jung, J.A. Jimenez, T.A. Gardner, M.H. Jeng, and C. Kao. 2005b. Gene therapy for prostate cancer by controlling adenovirus E1a and E4 gene expression with PSES enhancer. Cancer Res. 65:1941-1951. Lin, S., D.C. Lin, and M.D. Flanagan. 1978. Specificity of the effects of cytochalasin B on transport and motile processes. Proc Natl Acad Sci. 75:329-333. Linser, P., H. Bruning, and R.W. Armentrout. 1979. Uptake of minute virus of mice into cultured rodent cells. J Virol. 31:537-545. Linser, P., H. Bruning, and R.W. Armentrout. 1977. Specific binding sites for a parvovirus, minute virus of mice, on cultured mouse cells. J Virol. 24:211-221. Liu, H., Y. Fu, J. Xie, J. Cheng, S.A. Ghabrial, G. Li, Y. Peng, X. Yi, and D. Jiang. 2011. Widespread endogenization of densoviruses and parvoviruses in animal and human genomes. J Virol. 85:9863-9876.  153 Liu, X., and P. Collodi. 2002. Novel form of fibronectin from zebrafish mediates infectious hematopoietic necrosis virus infection. J Virol. 76:492-498. Lobert, V.H., A. Brech, N.M. Pedersen, J. Wesche, A. Oppelt, L. Malerod, and H. Stenmark. 2010. Ubiquitination of alpha 5 beta 1 integrin controls fibroblast migration through lysosomal degradation of fibronectin-integrin complexes. Dev Cell. 19:148-159. Lofling, J., S. Michael Lyi, C.R. Parrish, and A. Varki. 2013. Canine and feline parvoviruses preferentially recognize the non-human cell surface sialic acid N-glycolylneuraminic acid. Virology. 440:89-96. Lombardo, E., J.C. Ramirez, M. Agbandje-McKenna, and J.M. Almendral. 2000. A beta-stranded motif drives capsid protein oligomers of the parvovirus minute virus of mice into the nucleus for viral assembly. J Virol. 74:3804-3814. Lombardo, E., J.C. Ramirez, J. Garcia, and J.M. Almendral. 2002. Complementary roles of multiple nuclear targeting signals in the capsid proteins of the parvovirus minute virus of mice during assembly and onset of infection. J Virol. 76:7049-7059. Lopez-Bueno, A., M.P. Rubio, N. Bryant, R. McKenna, M. Agbandje-McKenna, and J.M. Almendral. 2006. Host-selected amino acid changes at the sialic acid binding pocket of the parvovirus capsid modulate cell binding affinity and determine virulence. J Virol. 80:1563-1573. Lu, P., V.M. Weaver, and Z. Werb. 2012. The extracellular matrix: a dynamic niche in cancer progression. J Cell Biol. 196:395-406. Macia, E., M. Ehrlich, R. Massol, E. Boucrot, C. Brunner, and T. Kirchhausen. 2006. Dynasore, a cell-permeable inhibitor of dynamin. Dev Cell. 10:839-850. Maitra, R., M.H. Ghalib, and S. Goel. 2012. Reovirus: a targeted therapeutic--progress and potential. Mol Cancer Res. 10:1514-1525. Manchester, M., D. Naniche, and T. Stehle. 2000. CD46 as a measles receptor: form follows function. Virology. 274:5-10. Mani, B., C. Baltzer, N. Valle, J.M. Almendral, C. Kempf, and C. Ros. 2006. Low pH-dependent endosomal processing of the incoming parvovirus minute virus of mice virion leads to externalization of the VP1 N-terminal sequence (N-VP1), N-VP2 cleavage, and uncoating of the full-length genome. J Virol. 80:1015-1024. Mao, Y., and J.E. Schwarzbauer. 2005. Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Mat Biol: J Int Soc Mat Biol. 24:389-399. Maroto, B., N. Valle, R. Saffrich, and J.M. Almendral. 2004. Nuclear export of the nonenveloped parvovirus virion is directed by an unordered protein signal exposed on the capsid surface. J Virol. 78:10685-10694. Matsuyama, S., M. Ujike, S. Morikawa, M. Tashiro, and F. Taguchi. 2005. Protease-mediated enhancement of severe acute respiratory syndrome coronavirus infection. Proc Natl Acad Sci. 102:12543-12547. Mattei, L.M., S.F. Cotmore, P. Tattersall, and A. Iwasaki. 2013. Parvovirus evades interferon-dependent viral control in primary mouse embryonic fibroblasts. Virology. 442:20-27. McFadden, G., M.R. Mohamed, M.M. Rahman, and E. Bartee. 2009. Cytokine determinants of viral tropism. Nat Rev Immunol. 9:645-655. McSharry, B.P., S.K. Forbes, J.Z. Cao, S. Avdic, E.A. Machala, D.J. Gottlieb, A. Abendroth, and B. Slobedman. 2014. Human cytomegalovirus upregulates expression of the lectin, galectin-9, via induction of interferon-beta. J Virol. 01259-14 Mercer, J., and A. Helenius. 2008. Vaccinia virus uses macropinocytosis and apoptotic mimicry to enter host cells. Science. 320:531-535.  154 Mercer, J., M. Schelhaas, and A. Helenius. 2010. Virus entry by endocytosis. Annu Rev Biochem. 79:803-833. Mercier, S., C. St-Pierre, I. Pelletier, M. Ouellet, M.J. Tremblay, and S. Sato. 2008. Galectin-1 promotes HIV-1 infectivity in macrophages through stabilization of viral adsorption. Virology. 371:121-129. Merrill, M.K., G. Bernhardt, J.H. Sampson, C.J. Wikstrand, D.D. Bigner, and M. Gromeier. 2004. Poliovirus receptor CD155-targeted oncolysis of glioma. Neuro Oncol. 6:208-217. Miest, T.S., and R. Cattaneo. 2014. New viruses for cancer therapy: meeting clinical needs. Nat Rev Microbiol. 12:23-34. Miller, C.L., and D.J. Pintel. 2002. Interaction between parvovirus NS2 protein and nuclear export factor Crm1 is important for viral egress from the nucleus of murine cells. J Virol. 76:3257-3266. Miller, J.M., S.M. Bidula, T.M. Jensen, and C.S. Reiss. 2010. Vesicular stomatitis virus modified with single chain IL-23 exhibits oncolytic activity against tumor cells in vitro and in vivo. Int J Infereron Cytokine Mediator Res. 2010:63-72. Miranda, F.A., M.K. Hassumi, M.C. Guimaraes, R.T. Simoes, T.G. Silva, R.C. Lira, A.M. Rocha, C.T. Jr. Mendes, E.A. Donadi, C.P. Soares, and E.G. Soares. 2009. Galectin-3 overexpression in invasive laryngeal carcinoma, assessed by computer-assisted analysis. J Histochem Cytochem. 57:665-673. Miyazono, K. 2009. Transforming growth factor-beta signaling and cancer: the 28th Sapporo Cancer Seminar, 25-27 June 2008. Cancer Sci. 100:363-365. Morse, E.M., N.N. Brahme, and D.A. Calderwood. 2014. Integrin cytoplasmic tail interactions. Biochemistry. 53:810-820. Moser, M., K.R. Legate, R. Zent, and R. Fassler. 2009. The tail of integrins, talin, and kindlins. Science. 324:895-899. Mourad-Zeidan, A.A., V.O. Melnikova, H. Wang, A. Raz, and M. Bar-Eli. 2008. Expression profiling of Galectin-3-depleted melanoma cells reveals its major role in melanoma cell plasticity and vasculogenic mimicry. Am J Pathol. 173:1839-1852. Mousset, S., J. Cornelis, N. Spruyt, and J. Rommelaere. 1986. Transformation of established murine fibroblasts with an activated cellular Harvey-ras oncogene or the polyoma virus middle T gene increases cell permissiveness to parvovirus minute-virus-of-mice. Biochimie. 68:951-955. Muller, P.A., K.H. Vousden, and J.C. Norman. 2011. p53 and its mutants in tumor cell migration and invasion. J Cell Biol. 192:209-218. Nagano, M., D. Hoshino, N. Koshikawa, T. Akizawa, and M. Seiki. 2012. Turnover of focal adhesions and cancer cell migration. Int J Cell Biol. 2012:310616. Naik, S., and S.J. Russell. 2009. Engineering oncolytic viruses to exploit tumor specific defects in innate immune signaling pathways. Expert Opin Biol Ther. 9:1163-1176. Nam, H.J., B. Gurda-Whitaker, W.Y. Gan, S. Ilaria, R. McKenna, P. Mehta, R.A. Alvarez, and M. Agbandje-McKenna. 2006. Identification of the sialic acid structures recognized by minute virus of mice and the role of binding affinity in virulence adaptation. J Biol Chem. 281:25670-25677. Narisawa-Saito, M., and T. Kiyono. 2007. Basic mechanisms of high-risk human papillomavirus-induced carcinogenesis: roles of E6 and E7 proteins. Can Sci. 98:1505-1511. Nethe, M., and P.L. Hordijk. 2011. A model for phospho-caveolin-1-driven turnover of focal adhesions. Cell Adh Migr. 5:59-64.  155 Newlaczyl, A.U., and L.G. Yu. 2011. Galectin-3--a jack-of-all-trades in cancer. Cancer Lett. 313:123-128. Nonnenmacher, M., and T. Weber. 2011. Adeno-associated virus 2 infection requires endocytosis through the CLIC/GEEC pathway. Cell Host Mic. 10:563-576. Nuesch, J.P., S. Lachmann, R. Corbau, and J. Rommelaere. 2003. Regulation of minute virus of mice NS1 replicative functions by atypical PKClambda in vivo. J Virol. 77:433-442. Nuesch, J.P., J. Lacroix, A. Marchini, and J. Rommelaere. 2012. Molecular pathways: rodent parvoviruses--mechanisms of oncolysis and prospects for clinical cancer treatment. Clin Cancer Res. 18:3516-3523. Oft, M., J. Peli, C. Rudaz, H. Schwarz, H. Beug, and E. Reichmann. 1996. TGF-beta1 and Ha-Ras collaborate in modulating the phenotypic plasticity and invasiveness of epithelial tumor cells. Genes Dev. 10:2462-2477. Oh, M.J., J. Akhtar, P. Desai, and D. Shukla. 2010. A role for heparan sulfate in viral surfing. Biochem Biophy Res Com. 391:176-181. Ojala, D.S., D.P. Amara, and D.V. Schaffer. 2014. Adeno-associated virus vectors and neurological gene therapy. Neuroscientist. Op De Beeck, A., J. Sobczak-Thepot, H. Sirma, F. Bourgain, C. Brechot, and P. Caillet-Fauquet. 2001. NS1- and minute virus of mice-induced cell cycle arrest: involvement of p53 and p21(cip1). J Virol. 75:11071-11078. Owen, D.M., D. Williamson, C. Rentero, and K. Gaus. 2009. Quantitative microscopy: protein dynamics and membrane organisation. Traffic. 10:962-971. Ozaki, T., and A. Nakagawara. 2011. p53: the attractive tumor suppressor in the cancer research field. J Biomed Biotechnol. 2011:603925. Pace, K.E., C. Lee, P.L. Stewart, and L.G. Baum. 1999. Restricted receptor segregation into membrane microdomains occurs on human T cells during apoptosis induced by galectin-1. J Immunol. 163:3801-3811. Pacis, R.A., M.J. Pilat, K.J. Pienta, K. Wojno, A. Raz, V. Hogan, and C.R. Cooper. 2000. Decreased galectin-3 expression in prostate cancer. Prostate. 44:118-123. Paglino, J., E. Burnett, and P. Tattersall. 2007. Exploring the contribution of distal P4 promoter elements to the oncoselectivity of Minute Virus of Mice. Virology. 361:174-184. Paglino, J.C., W. Andres, and A.N. van den Pol. 2014. Autonomous parvoviruses neither stimulate nor are inhibited by type I interferon response in human normal or cancer cells. J Virol. 88(9):4932-42 Paglino, J.C., and A.N. van den Pol. 2011. Vesicular stomatitis virus has extensive oncolytic activity against human sarcomas: rare resistance is overcome by blocking interferon pathways. J Virol. 85:9346-9358. Parato, K.A., D. Senger, P.A. Forsyth, and J.C. Bell. 2005. Recent progress in the battle between oncolytic viruses and tumours. Nat Rev. Can. 5:965-976. Parker, J.S., and C.R. Parrish. 2000. Cellular uptake and infection by canine parvovirus involves rapid dynamin-regulated clathrin-mediated endocytosis, followed by slower intracellular trafficking. J Virol. 74:1919-1930. Parri, M., and P. Chiarugi. 2010. Rac and Rho GTPases in cancer cell motility control. Cell Commun Signal. 8:23. Parrish, C.R. 2010. Structures and functions of parvovirus capsids and the process of cell infection. Curr Top Microbiol Immunol. 343:149-176. Parsons, J.T., A.R. Horwitz, and M.A. Schwartz. 2010. Cell adhesion: integrating cytoskeletal dynamics and cellular tension. Nat Rev Mol Cell Biol. 11:633-643.  156 Partridge, E.A., C. Le Roy, G.M. Di Guglielmo, J. Pawling, P. Cheung, M. Granovsky, I.R. Nabi, J.L. Wrana, and J.W. Dennis. 2004. Regulation of cytokine receptors by Golgi N-glycan processing and endocytosis. Science. 306:120-124. Peinado, H., D. Olmeda, and A. Cano. 2007. Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev. Can. 7:415-428. Pelkmans, L., D. Puntener, and A. Helenius. 2002. Local actin polymerization and dynamin recruitment in SV40-induced internalization of caveolae. Science. 296:535-539. Phuangsab, A., R.M. Lorence, K.W. Reichard, M.E. Peeples, and R.J. Walter. 2001. Newcastle disease virus therapy of human tumor xenografts: antitumor effects of local or systemic administration. Cancer Lett. 172:27-36. Pierre, S., A.S. Bats, and X. Coumoul. 2011. Understanding SOS (Son of Sevenless). Biochem Pharmacol. 82:1049-1056. Ponnazhagan, S., D.T. Curiel, D.R. Shaw, R.D. Alvarez, and G.P. Siegal. 2001. Adeno-associated virus for cancer gene therapy. Cancer Res. 61:6313-6321. Porwal, M., S. Cohen, K. Snoussi, R. Popa-Wagner, F. Anderson, N. Dugot-Senant, H. Wodrich, C. Dinsart, J.A. Kleinschmidt, N. Pante, and M. Kann. 2013. Parvoviruses cause nuclear envelope breakdown by activating key enzymes of mitosis. PLoS Pathog. 9:e1003671. Prior, I.A., P.D. Lewis, and C. Mattos. 2012. A comprehensive survey of Ras mutations in cancer. Cancer Res. 72:2457-2467. Pylayeva-Gupta, Y., E. Grabocka, and D. Bar-Sagi. 2011. RAS oncogenes: weaving a tumorigenic web. Nat Rev Cancer. 11:761-774. Pytliak, M., V. Vargova, and V. Mechirova. 2012. Matrix metalloproteinases and their role in oncogenesis: a review. Onkologie. 35:49-53. Qazilbash, M.H., X. Xiao, P. Seth, K.H. Cowan, and C.E. Walsh. 1997. Cancer gene therapy using a novel adeno-associated virus vector expressing human wild-type p53. Gene Ther. 4:675-682. Quattrocchi, S., N. Ruprecht, C. Bonsch, S. Bieli, C. Zurcher, K. Boller, C. Kempf, and C. Ros. 2012. Characterization of the early steps of human parvovirus B19 infection. J Virol. 86:9274-9284. Raff, A.B., A.W. Woodham, L.M. Raff, J.G. Skeate, L. Yan, D.M. Da Silva, M. Schelhaas, and W.M. Kast. 2013. The evolving field of human papillomavirus receptor research: a review of binding and entry. J Virol. 87:6062-6072. Randall, R.E., and S. Goodbourn. 2008. Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures. J Gen Virol. 89:1-47. Raptis, L., R. Marcellus, M.J. Corbley, A. Krook, J. Whitfield, S.K. Anderson, and T. Haliotis. 1991. Cellular ras gene activity is required for full neoplastic transformation by polyomavirus. J Virol. 65:5203-5210. Rein, D.T., M. Breidenbach, T.O. Kirby, T. Han, G.P. Siegal, G.J. Bauerschmitz, M. Wang, D.M. Nettelbeck, Y. Tsuruta, M. Yamamoto, P. Dall, A. Hemminki, and D.T. Curiel. 2005. A fiber-modified, secretory leukoprotease inhibitor promoter-based conditionally replicating adenovirus for treatment of ovarian cancer. Clin Cancer Res. 11:1327-1335. Rhode, S.L., 3rd. 1973. Replication process of the parvovirus H-1. I. Kinetics in a parasynchronous cell system. J Virol. 11:856-861. Riolobos, L., N. Valle, E. Hernando, B. Maroto, M. Kann, and J.M. Almendral. 2010. Viral oncolysis that targets Raf-1 signaling control of nuclear transport. J Virol. 84:2090-2099. Rivlin, N., R. Brosh, M. Oren, and V. Rotter. 2011. Mutations in the p53 Tumor Suppressor Gene: Important Milestones at the Various Steps of Tumorigenesis. Gen Can. 2:466-474.  157 Rommelaere, J., and J.J. Cornelis. 1991. Antineoplastic activity of parvoviruses. J Virol Methods. 33:233-251. Rommelaere, J., K. Geletneky, A.L. Angelova, L. Daeffler, C. Dinsart, I. Kiprianova, J.R. Schlehofer, and Z. Raykov. 2010. Oncolytic parvoviruses as cancer therapeutics. Cytokine Growth Factor Rev. 21:185-195. Ros, C., C.J. Burckhardt, and C. Kempf. 2002. Cytoplasmic trafficking of minute virus of mice: low-pH requirement, routing to late endosomes, and proteasome interaction. J Virol. 76:12634-12645. Ros, C., and C. Kempf. 2004. The ubiquitin-proteasome machinery is essential for nuclear translocation of incoming minute virus of mice. Virology. 324:350-360. Rosenberg, I., B.J. Cherayil, K.J. Isselbacher, and S. Pillai. 1991. Mac-2-binding glycoproteins. Putative ligands for a cytosolic beta-galactoside lectin. J Biol Chem. 266:18731-18736. Roy, R., J. Yang, and M.A. Moses. 2009. Matrix metalloproteinases as novel biomarkers and potential therapeutic targets in human cancer. J Clin Onc. 27:5287-5297. Rubio, M.P., A. Lopez-Bueno, and J.M. Almendral. 2005. Virulent variants emerging in mice infected with the apathogenic prototype strain of the parvovirus minute virus of mice exhibit a capsid with low avidity for a primary receptor. J Virol. 79:11280-11290. Ruoslahti, E. 1996. RGD and other recognition sequences for integrins. Annu Rev Cell Dev Biol. 12:697-715. Russell, S.J., K.W. Peng, and J.C. Bell. 2012. Oncolytic virotherapy. Nat Biotechnol. 30:658-670. Sadler, A.J., and B.R. Williams. 2008. Interferon-inducible antiviral effectors. Nat Rev Immunol. 8:559-568. Sahai, E., M.F. Olson, and C.J. Marshall. 2001. Cross-talk between Ras and Rho signalling pathways in transformation favours proliferation and increased motility. Eur Mol Biol Organ J. 20:755-766. Saito, S., Y. Hosoya, K. Togashi, K. Kurashina, H. Haruta, M. Hyodo, K. Koinuma, H. Horie, Y. Yasuda, and H. Nagai. 2008. Prevalence of synchronous colorectal neoplasms detected by colonoscopy in patients with gastric cancer. Surg Today. 38:20-25. Sasaki, T., C. Brakebusch, J. Engel, and R. Timpl. 1998. Mac-2 binding protein is a cell-adhesive protein of the extracellular matrix which self-assembles into ring-like structures and binds beta1 integrins, collagens and fibronectin. Eur Mol Biol Organ J. 17:1606-1613. Sathisha, U.V., S. Jayaram, M.A. Harish Nayaka, and S.M. Dharmesh. 2007. Inhibition of galectin-3 mediated cellular interactions by pectic polysaccharides from dietary sources. Glycoconj J. 24:497-507. Savagner, P. 2010. The epithelial-mesenchymal transition (EMT) phenomenon. Ann Oncol. 21 Suppl 7:vii89-92. Schaffner, F., A.M. Ray, and M. Dontenwill. 2013. Integrin alpha5beta1, the Fibronectin Receptor, as a Pertinent Therapeutic Target in Solid Tumors. Cancers (Basel). 5:27-47. Schelhaas, M., H. Ewers, M.L. Rajamaki, P.M. Day, J.T. Schiller, and A. Helenius. 2008. Human papillomavirus type 16 entry: retrograde cell surface transport along actin-rich protrusions. PLoS Pathog. 4:e1000148. Sen, G.C., and S.N. Sarkar. 2007. The interferon-stimulated genes: targets of direct signaling by interferons, double-stranded RNA, and viruses. Curr Top Microbiol Immunol. 316:233-250. Senzer, N.N., H.L. Kaufman, T. Amatruda, M. Nemunaitis, T. Reid, G. Daniels, R. Gonzalez, J. Glaspy, E. Whitman, K. Harrington, H. Goldsweig, T. Marshall, C. Love, R. Coffin, and J.J. Nemunaitis. 2009. Phase II clinical trial of a granulocyte-macrophage colony- 158 stimulating factor-encoding, second-generation oncolytic herpesvirus in patients with unresectable metastatic melanoma. J Clin Onc. 27:5763-5771. Shafren, D.R., D. Sylvester, E.S. Johansson, I.G. Campbell, and R.D. Barry. 2005. Oncolysis of human ovarian cancers by echovirus type 1. Int J Cancer. 115:320-328. Shankar, J., S.M. Wiseman, F. Meng, K. Kasaian, S. Strugnell, A. Mofid, A. Gown, S.J. Jones, and I.R. Nabi. 2012. Coordinated expression of galectin-3 and caveolin-1 in thyroid cancer. J pathol. 228:56-66. Sherry, B. 2009. Rotavirus and reovirus modulation of the interferon response. J Interferon Cytokine Res. 29:559-567. Shi, F., and J. Sottile. 2008. Caveolin-1-dependent beta1 integrin endocytosis is a critical regulator of fibronectin turnover. J Cell Sci. 121:2360-2371. Shmulevitz, M., L.Z. Pan, K. Garant, D. Pan, and P.W. Lee. 2010. Oncogenic Ras promotes reovirus spread by suppressing IFN-beta production through negative regulation of RIG-I signaling. Cancer Res. 70:4912-4921. Shobana, R., S.K. Samal, and S. Elankumaran. 2013. Prostate-specific antigen-retargeted recombinant newcastle disease virus for prostate cancer virotherapy. J Virol. 87:3792-3800. Siegl, G., and M. Gautschi. 1976. Multiplication of parvovirus LuIII in a synchronized culture system. III. Replication of viral DNA. J Virol. 17:841-853. Singh, P., C. Carraher, and J.E. Schwarzbauer. 2010. Assembly of fibronectin extracellular matrix. Annu Rev Cell Dev Biol. 26:397-419. Sowinski, S., C. Jolly, O. Berninghausen, M.A. Purbhoo, A. Chauveau, K. Kohler, S. Oddos, P. Eissmann, F.M. Brodsky, C. Hopkins, B. Onfelt, Q. Sattentau, and D.M. Davis. 2008. Membrane nanotubes physically connect T cells over long distances presenting a novel route for HIV-1 transmission. Nat Cell Biol. 10:211-219. Spandidos, D.A., and M. Riggio. 1986. Polyoma virus middle T gene can trigger malignant transformation of early passage rodent cells. J Gen Virol. 67 ( Pt 4):793-799. Springfeld, C., V. von Messling, M. Frenzke, G. Ungerechts, C.J. Buchholz, and R. Cattaneo. 2006. Oncolytic efficacy and enhanced safety of measles virus activated by tumor-secreted matrix metalloproteinases. Cancer Res. 66:7694-7700. St-Pierre, C., M. Ouellet, M.J. Tremblay, and S. Sato. 2010. Galectin-1 and HIV-1 Infection. Meth Enz. 480:267-294. Stehbens, S., and T. Wittmann. 2012. Targeting and transport: how microtubules control focal adhesion dynamics. J Cell Biol. 198:481-489. Sudo, T., T. Iwaya, N. Nishida, G. Sawada, Y. Takahashi, M. Ishibashi, K. Shibata, H. Fujita, K. Shirouzu, M. Mori, and K. Mimori. 2013. Expression of mesenchymal markers vimentin and fibronectin: the clinical significance in esophageal squamous cell carcinoma. Ann Surg Oncol. 20 Suppl 3:S324-335. Suikkanen, S., T. Aaltonen, M. Nevalainen, O. Valilehto, L. Lindholm, M. Vuento, and M. Vihinen-Ranta. 2003. Exploitation of microtubule cytoskeleton and dynein during parvoviral traffic toward the nucleus. J Virol. 77:10270-10279. Suraneni, P., B. Rubinstein, J.R. Unruh, M. Durnin, D. Hanein, and R. Li. 2012. The Arp2/3 complex is required for lamellipodia extension and directional fibroblast cell migration. J Cell Biol. 197:239-251. Suzuki, K., and H. Matsubara. 2011. Recent advances in p53 research and cancer treatment. J Biomed Biotechnol. 2011:978312.  159 Takenaka, Y., T. Fukumori, and A. Raz. 2004. Galectin-3 and metastasis. Glycoconj J. 19:543-549. Tattersall, P., P.J. Cawte, A.J. Shatkin, and D.C. Ward. 1976. Three structural polypeptides coded for by minite virus of mice, a parvovirus. J Virol. 20:273-289. Tattersall, P., A.J. Shatkin, and D.C. Ward. 1977. Sequence homology between the structural polypeptides of minute virus of mice. J Mol Biol. 111:375-394. Theveneau, E., and R. Mayor. 2011. Can mesenchymal cells undergo collective cell migration? The case of the neural crest. Cell Adh Migr. 5:490-498. Theveneau, E., and R. Mayor. 2013. Collective cell migration of epithelial and mesenchymal cells. Cell Mol Life Sci. 70:3481-3492. Thiery, J.P. 2002. Epithelial-mesenchymal transitions in tumour progression. Nat Rev. Can. 2:442-454. Toolan, H.W., and N. Ledinko. 1968. Inhibition by H-1 virus of the incidence of tumors produced by adenovirus 12 in hamsters. Virology. 35:475-478. Trusolino, L., and P.M. Comoglio. 2002. Scatter-factor and semaphorin receptors: cell signalling for invasive growth. Nat Rev. Can. 2:289-300. Urra, H., V.A. Torres, R.J. Ortiz, L. Lobos, M.I. Diaz, N. Diaz, S. Hartel, L. Leyton, and A.F. Quest. 2012. Caveolin-1-enhanced motility and focal adhesion turnover require tyrosine-14 but not accumulation to the rear in metastatic cancer cells. PloS One. 7:e33085. Vacchelli, E., A. Eggermont, C. Sautes-Fridman, J. Galon, L. Zitvogel, G. Kroemer, and L. Galluzzi. 2013. Trial watch: Oncolytic viruses for cancer therapy. Oncoimmunology. 2:e24612. Valastyan, S., and R.A. Weinberg. 2011. Tumor metastasis: molecular insights and evolving paradigms. Cell. 147:275-292. Vasta, G.R., H. Ahmed, M.A. Bianchet, J.A. Fernandez-Robledo, and L.M. Amzel. 2012a. Diversity in recognition of glycans by F-type lectins and galectins: molecular, structural, and biophysical aspects. Ann N Y Acad Sci. 1253:E14-26. Vasta, G.R., H. Ahmed, M. Nita-Lazar, A. Banerjee, M. Pasek, S. Shridhar, P. Guha, and J.A. Fernandez-Robledo. 2012b. Galectins as self/non-self recognition receptors in innate and adaptive immunity: an unresolved paradox. Front Immunol. 3:199. Ventoso, I., J.J. Berlanga, and J.M. Almendral. 2010. Translation control by protein kinase R restricts minute virus of mice infection: role in parvovirus oncolysis. J Virol. 84:5043-5051. Vousden, K.H., and C. Prives. 2009. Blinded by the Light: The Growing Complexity of p53. Cell. 137:413-431. Wang, F., Y. Ma, J.W. Barrett, X. Gao, J. Loh, E. Barton, H.W. Virgin, and G. McFadden. 2004. Disruption of Erk-dependent type I interferon induction breaks the myxoma virus species barrier. Nat Immunol. 5:1266-1274. Wang, L.H., K.G. Rothberg, and R.G. Anderson. 1993. Mis-assembly of clathrin lattices on endosomes reveals a regulatory switch for coated pit formation. J Cell Biol. 123:1107-1117. Wang, R., J. Bi, K.K. Ampah, X. Ba, W. Liu, and X. Zeng. 2013a. Lipid rafts control human melanoma cell migration by regulating focal adhesion disassembly. Biochim Biophys Acta. 1833:3195-3205. Wang, Y., V. Balan, X. Gao, P.G. Reddy, D. Kho, L. Tait, and A. Raz. 2013b. The significance of galectin-3 as a new basal cell marker in prostate cancer. Cell Death Dis. 4:e753.  160 Wang, Y., and M.A. McNiven. 2012. Invasive matrix degradation at focal adhesions occurs via protease recruitment by a FAK-p130Cas complex. J Cell Biol. 196:375-385. Wang, Y., P. Nangia-Makker, L. Tait, V. Balan, V. Hogan, K.J. Pienta, and A. Raz. 2009. Regulation of prostate cancer progression by galectin-3. Am J Pathol. 174:1515-1523. Webb, D.J., J.T. Parsons, and A.F. Horwitz. 2002. Adhesion assembly, disassembly and turnover in migrating cells -- over and over and over again. Nat Cell Biol. 4:E97-100. Webster, M.A., J.N. Hutchinson, M.J. Rauh, S.K. Muthuswamy, M. Anton, C.G. Tortorice, R.D. Cardiff, F.L. Graham, J.A. Hassell, and W.J. Muller. 1998. Requirement for both Shc and phosphatidylinositol 3' kinase signaling pathways in polyomavirus middle T-mediated mammary tumorigenesis. Mol Cell Biol. 18:2344-2359. Weigel-Kelley, K.A., M.C. Yoder, and A. Srivastava. 2003. Alpha5beta1 integrin as a cellular coreceptor for human parvovirus B19: requirement of functional activation of beta1 integrin for viral entry. Blood. 102:3927-3933. Wendt, M.K., T.M. Allington, and W.P. Schiemann. 2009. Mechanisms of the epithelial-mesenchymal transition by TGF-beta. Future Oncol. 5:1145-1168. Williams, W.P., L. Tamburic, and C.R. Astell. 2004. Increased levels of B1 and B2 SINE transcripts in mouse fibroblast cells due to minute virus of mice infection. Virology. 327:233-241. Wu, Z., E. Miller, M. Agbandje-McKenna, and R.J. Samulski. 2006. Alpha2,3 and alpha2,6 N-linked sialic acids facilitate efficient binding and transduction by adeno-associated virus types 1 and 6. J Virol. 80:9093-9103. Yamaguchi, H., and J. Condeelis. 2007. Regulation of the actin cytoskeleton in cancer cell migration and invasion. Biochim Biophys Acta. 1773:642-652. Yamazaki, D., S. Kurisu, and T. Takenawa. 2005. Regulation of cancer cell motility through actin reorganization. Cancer Sci. 96:379-386. Yang, J., and R.A. Weinberg. 2008. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. 14:818-829. Yang, M.L., Y.H. Chen, S.W. Wang, Y.J. Huang, C.H. Leu, N.C. Yeh, C.Y. Chu, C.C. Lin, G.S. Shieh, Y.L. Chen, J.R. Wang, C.H. Wang, C.L. Wu, and A.L. Shiau. 2011. Galectin-1 binds to influenza virus and ameliorates influenza virus pathogenesis. J Virol. 85:10010-10020. Yeung, D.E., G.W. Brown, P. Tam, R.H. Russnak, G. Wilson, I. Clark-Lewis, and C.R. Astell. 1991. Monoclonal antibodies to the major nonstructural nuclear protein of minute virus of mice. Virology. 181:35-45. Zeyaullah, M., M. Patro, I. Ahmad, K. Ibraheem, P. Sultan, M. Nehal, and A. Ali. 2012. Oncolytic viruses in the treatment of cancer: a review of current strategies. Pathol Oncol Res. 18:771-781. Zincarelli, C., S. Soltys, G. Rengo, and J.E. Rabinowitz. 2008. Analysis of AAV serotypes 1-9 mediated gene expression and tropism in mice after systemic injection. J Mol Ther. 16:1073-1080.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0135542/manifest

Comment

Related Items