UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigating the hepatitis C virus and dengue virus interactions with host lipid pathways : from circulating… Hyrina, Anastasia 2016

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

Item Metadata

Download

Media
24-ubc_2017_february_hyrina_anastasia.pdf [ 21.1MB ]
Metadata
JSON: 24-1.0340573.json
JSON-LD: 24-1.0340573-ld.json
RDF/XML (Pretty): 24-1.0340573-rdf.xml
RDF/JSON: 24-1.0340573-rdf.json
Turtle: 24-1.0340573-turtle.txt
N-Triples: 24-1.0340573-rdf-ntriples.txt
Original Record: 24-1.0340573-source.json
Full Text
24-1.0340573-fulltext.txt
Citation
24-1.0340573.ris

Full Text

 INVESTIGATING THE HEPATITIS C VIRUS AND DENGUE VIRUS INTERACTIONS WITH HOST LIPID PATHWAYS: FROM CIRCULATING HUMAN MICRORNAS TO LIPID MODULATING AGENTS  by  Anastasia Hyrina B.A., Mount Holyoke College, 2012   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES	   (Microbiology and Immunology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   December 2016   © Anastasia Hyrina, 2016  	   ii	  Abstract  Cholesterol and lipid levels are maintained through tightly controlled and complex feedback mechanisms that involve regulation of major metabolic genes.  Dysregulation of cellular or plasma lipid levels can lead to a wide range of pathologies, including hyperlipidemia, atherosclerosis and other disorders.  A number of viruses, including important human viruses of the Flaviviridae family such as hepatitis C virus (HCV) and dengue virus (DENV), utilize and modulate host lipids to support their lifecycles, and the resulting changes in lipid metabolism may contribute to virus-associated pathologies.  The overall aim of this thesis was to determine the role of key regulators of host lipid homeostasis, including microRNAs (miR-122, miR-24 and miR-223) and proprotein convertases (SKI-1/S1P and PCSK9) during viral infection and virus-associated disease. To address this aim, we first examined the molecular interplay between three circulating microRNAs known to act as regulators of lipid homeostasis.  The data we present in Chapter 2 shows that specific signatures of the three microRNAs were associated with different treatment outcomes in patients with chronic hepatitis C (CHC), indicating that these microRNAs correlate with HCV infection.  We then tested the hypothesis that enzymatic regulators of lipid metabolism could also indicate HCV infection, and in Chapter 3 we show that PCSK9 levels were significantly upregulated in patients who achieved a treatment-based viral cure but not in relapsers.  These data indicate that changes in PCSK9 concentrations may have an important role in both HCV infection and in host lipid metabolism.  In Chapter 4, we tested whether reducing the abundance of lipid droplets via inhibition of SKI-1/S1P with a small molecule PF-429242 suppresses DENV infection.  The inhibitor blocked SKI-1/S1P-mediated accumulation of lipid droplets in hepatoma cells and reduced DENV infection, identifying SKI-1/S1P as a potential target for indirect-acting anti-DENV agents.   This study on modulators of lipid metabolism during HCV and DENV infections provides new insights into the complex host-virus interactions that associate with virally-induced disease.  We hope that our data lay the foundation for understanding disease pathogenesis and support the development of future strategies for Flaviviridae - associated diseases. 	   iii	  Preface  All of the work presented henceforth was conducted in the Life Sciences Center at the University of British Columbia, Point Grey campus.  Dr. François Jean was the supervisory author on this thesis and was involved throughout for all the projects in concept formation and manuscript composition.   All projects and associated methods involving human samples were approved by the University of British Columbia’s Research Ethics Board [certificate # H13-01770].  Biohazard Approval Certificate #B12-0106 for the Jean lab antiviral research programs.  For the work presented in Chapter 2 and 3, I was the lead investigator, responsible for experimental design, data collection and analysis.  Drs. Edward Tam and Mel Krajden provided the cohort plasma samples collected at the LAIR Centre.  Paul Steven (Qiagen) fitted several statistical models presented in Table 2.3 and Tables A1.2-4.  Dr. Andrea D. Olmstead (BCCDC) performed the experiments presented in Figure 3.3.  I wrote the first drafts of the manuscript for this project, which was revised together with Dr. François Jean, Dr. Mel Krajden and Dr. Andrea D. Olmstead.  A version of the research presented in Chapter 4 has been submitted to a journal (Hyrina, A., Meng, F., McArthur, S.J., Nabi I.R., Jean F. Human Subtilisin Kexin Isozyme-1(SKI-1)/Site-1 Protease (S1P) Regulates Cytoplasmic Lipid Droplet Abundance: A Potential Target for Indirect-Acting Anti-Dengue Virus Agents).  I was responsible for designing all the experiments described in this manuscript.  Dr. Fanrui Meng (UBC) provided technical assistance with confocal microscopy presented in Figures 4.1, 4.2 and 4.5.  Sharlene Eivemark assessed PF-429242 for cytotoxic effects in Huh-7.5.1 cells shown in Figure 4.1.  Steven McArthur performed the EC50 curve fitting shown in Figures 4.1 and 4.3.  I wrote the first drafts of the manuscript, which was revised together with Dr. François Jean and Dr. Robert Nabi (UBC).  All reagents provided by external research groups are indicated in the materials and methods.  This work was funded by a Canadian Institutes of Health Research operating grant, Qiagen and British Columbia Proteomics Network – Michael Smith Foundation for Health Research (F. Jean).  Graduate student funding and research training was provided by Canadian Network on Hepatitis C (CanHepC) under the mentorship of Dr. Mel Krajden.  	   iv	  Table of Contents Abstract	  .............................................................................................................................................	  ii	  Preface	  .............................................................................................................................................	  iii	  Table	  of	  Contents	  ...........................................................................................................................	  iv	  List	  of	  Tables	  .................................................................................................................................	  vii	  List	  of	  Figures	  ..............................................................................................................................	  viii	  List	  of	  Abbreviations	  .....................................................................................................................	  x	  Acknowledgements	  .....................................................................................................................	  xv	  Dedication	  ......................................................................................................................................	  xvi	  Chapter	  1:	  Introduction	  ...............................................................................................................	  1	  1.1	  Host	  lipid	  metabolism	  .....................................................................................................................	  1	  1.1.1	  Lipoproteins	  ...................................................................................................................................................	  1	  1.1.2	  Low-­‐density	  lipoprotein	  receptor	  ........................................................................................................	  2	  1.1.3	  Proprotein	  convertase	  subtilisin/kexin	  type	  9	  ...............................................................................	  3	  1.1.4	  The	  SREBP	  pathway	  ...................................................................................................................................	  6	  1.1.5	  Subtilisin	  kexin	  isozyme-­‐1/site-­‐1	  protease	  .....................................................................................	  6	  1.1.6.	  Viral	  dysregulation	  of	  host	  lipid	  homeostasis	  ................................................................................	  8	  1.2	  MicroRNAs	  ...........................................................................................................................................	  9	  1.2.1	  Discovery	  of	  microRNAs	  ...........................................................................................................................	  9	  1.2.2	  MicroRNA	  biogenesis	  ...............................................................................................................................	  10	  1.2.3	  Post-­‐transcriptional	  repression	  by	  microRNAs	  ...........................................................................	  10	  1.2.4	  Circulating	  microRNAs	  ............................................................................................................................	  12	  1.2.5	  Emerging	  role	  of	  microRNAs	  in	  regulating	  lipid	  metabolism	  ................................................	  15	  1.2.6	  Role	  of	  microRNAs	  during	  viral	  infection	  .......................................................................................	  16	  1.3	  Hepatitis	  C	  virus	  (HCV)	  .................................................................................................................	  18	  1.3.1	  HCV	  discovery	  and	  diversity	  .................................................................................................................	  19	  1.3.2	  HCV	  pathogenesis	  ......................................................................................................................................	  20	  1.3.3	  HCV	  proteins	  ................................................................................................................................................	  21	  1.3.4	  HCV	  lifecycle	  ................................................................................................................................................	  23	  1.3.5	  HCV	  antivirals	  ..............................................................................................................................................	  25	  1.3.6	  HCV	  and	  lipid	  metabolism	  .....................................................................................................................	  27	  1.3.7	  MicroRNAs	  during	  HCV	  infection	  .......................................................................................................	  28	  1.4	  Dengue	  virus	  (DENV)	  biology	  .....................................................................................................	  29	  1.4.1	  DENV	  epidemiology	  and	  clinical	  disease	  .........................................................................................	  30	  1.4.2	  DENV	  pathogenesis	  ...................................................................................................................................	  31	  1.4.3	  DENV	  biology	  and	  lifecycle	  ....................................................................................................................	  32	  1.4.4	  DENV	  vaccine	  and	  treatment	  ................................................................................................................	  34	  1.4.5	  DENV	  and	  lipid	  metabolism	  ..................................................................................................................	  35	  1.5	  Research	  hypothesis	  and	  rationale	  ..........................................................................................	  36	  1.5.1	  Aim	  1	  ...............................................................................................................................................................	  37	  1.5.2	  Aim	  2	  ...............................................................................................................................................................	  38	  1.5.3	  Aim	  3	  ...............................................................................................................................................................	  38	  	   v	  Chapter	  2:	  Differential	  profiles	  of	  circulating	  miR-­‐122,	  miR-­‐24	  and	  miR-­‐223	  in	  hepatitis	  C-­‐infected	  patients	  who	  achieve	  a	  treatment-­‐based	  viral	  cure	  ................	  40	  2.1	  Introduction	  .....................................................................................................................................	  40	  2.2	  Materials	  and	  methods	  .................................................................................................................	  42	  2.3	  Results	  ................................................................................................................................................	  45	  2.3.1	  Study	  design	  and	  cohort	  characteristics	  ..........................................................................................	  45	  2.3.2	  Correlation	  between	  circulating	  levels	  of	  plasma	  miR-­‐122,	  miR-­‐24,	  miR-­‐223,	  and	  clinical	  parameters	  in	  CHC	  patients	  ..............................................................................................................	  46	  2.3.3	  Circulating	  miR-­‐24	  and	  miR-­‐223	  plasma	  levels	  significantly	  increase	  in	  patients	  who	  have	  achieved	  SVR	  ................................................................................................................................................	  52	  2.4	  Discussion	  .........................................................................................................................................	  56	  2.4.1	  miR-­‐122	  in	  CHC	  infection	  .......................................................................................................................	  57	  2.4.2	  miR-­‐24	  and	  miR-­‐223	  correlate	  with	  CHC	  infection	  ....................................................................	  58	  2.4.3	  Summary	  .......................................................................................................................................................	  59	  Chapter	  3:	  Elevated	  plasma	  PCSK9	  levels	  in	  hepatitis	  C-­‐infected	  patients	  who	  achieve	  a	  treatment-­‐based	  viral	  cure	  ...................................................................................	  60	  3.1	  Introduction	  .....................................................................................................................................	  60	  3.2	  Materials	  and	  methods	  .................................................................................................................	  61	  3.3	  Results	  ................................................................................................................................................	  64	  3.3.1	  Cohort	  characteristics	  ..............................................................................................................................	  64	  3.3.2	  Plasma	  PCSK9	  levels	  increase	  after	  treatment-­‐based	  SVR	  ......................................................	  64	  3.3.3	  Extracellularly	  applied	  recombinant	  human	  wild-­‐type	  PCSK9	  and	  gain-­‐of-­‐function	  mutant	  PCSK9-­‐D374Y,	  but	  not	  loss-­‐of-­‐function	  PCSK9-­‐R194A	  variant,	  inhibited	  HCV	  infection	  in	  human	  hepatoma	  Huh-­‐7.5.1	  cells	  ..........................................................................................	  69	  3.4	  Discussion	  .........................................................................................................................................	  73	  Chapter	  4:	  Human	  subtilase	  SKI-­‐1/S1P	  regulates	  cytoplasmic	  lipid	  droplet	  abundance:	  a	  potential	  target	  for	  indirect-­‐acting	  anti-­‐dengue	  virus	  agents	  ..........	  76	  4.1	  Introduction	  .....................................................................................................................................	  76	  4.2	  Materials	  and	  methods	  .................................................................................................................	  78	  4.3	  Results	  ................................................................................................................................................	  83	  4.3.1	  Inhibition	  of	  SKI-­‐1/S1P	  enzymatic	  activity	  using	  PF-­‐429242	  impairs	  activation	  of	  the	  SREBP	  pathway	  and	  correlates	  with	  a	  dramatic	  decrease	  in	  lipid	  droplet	  abundance	  .........	  83	  4.3.2	  Huh-­‐7.5.1	  cells	  support	  DENV-­‐2	  replication	  and	  DENV	  capsid	  protein	  binding	  to	  hepatic	  lipid	  droplets	  ..........................................................................................................................................	  86	  4.3.3	  Pretreatment	  of	  Huh-­‐7.5.1	  cells	  with	  PF-­‐429242	  results	  in	  a	  dose-­‐dependent	  decrease	  in	  intracellular	  DENV-­‐2	  NS1	  expression	  and	  a	  3-­‐log	  decrease	  in	  extracellular	  viral	  titer	  ..............................................................................................................................................................................	  87	  4.3.4	  Pretreatment	  of	  Huh-­‐7.5.1	  cells	  with	  PF-­‐429242	  results	  in	  a	  robust	  decrease	  in	  intracellular	  DENV-­‐2	  RNA	  .................................................................................................................................	  90	  4.3.5	  Post-­‐treatment	  of	  DENV	  infected	  Huh-­‐7.5.1	  cells	  with	  PF-­‐429242	  does	  not	  affect	  viral	  RNA	  synthesis,	  but	  it	  does	  impair	  assembly	  and/or	  release	  of	  infectious	  virus	  particles	  ....	  93	  4.3.6	  Extracellularly	  applied	  oleic	  acid,	  an	  inducer	  of	  lipid	  droplet	  formation,	  rescues	  DENV-­‐2	  RNA	  synthesis	  in	  PF-­‐429242-­‐treated	  Huh-­‐7.5.1	  cells	  .........................................................	  96	  4.3.7	  Pretreatment	  of	  Huh-­‐7.5.1	  cells	  with	  PF-­‐429242	  results	  in	  a	  significant	  decrease	  in	  intracellular	  viral	  RNA	  for	  all	  four	  DENV	  serotypes	  ..............................................................................	  97	  4.4	  Discussion	  ......................................................................................................................................	  101	  4.4.1	  Human	  subtilase	  SKI-­‐1/S1P	  is	  a	  regulator	  of	  lipid	  droplet	  formation	  in	  Huh-­‐7.5.1	  human	  hepatoma	  cells	  .....................................................................................................................................	  101	  	   vi	  4.4.2	  Manipulation	  of	  human	  SKI-­‐1/S1P	  enzymatic	  activity	  provides	  a	  mean	  of	  effectively	  inhibiting	  viral	  infection	  of	  Huh-­‐7.5.1	  cells	  by	  DENV-­‐2	  ....................................................................	  102	  4.4.3	  Inhibition	  of	  the	  SKI-­‐1/S1P-­‐mediated	  proteolytic	  activation	  of	  the	  SREBP	  pathway	  has	  a	  pan-­‐serotypic	  inhibitory	  effect	  on	  DENV	  infection	  .................................................................	  104	  4.4.4	  SKI-­‐1/S1P	  is	  a	  potential	  target	  for	  indirect-­‐acting	  antiviral	  agents	  against	  DENV	  infection	  .................................................................................................................................................................	  105	  Chapter	  5:	  Conclusions	  and	  future	  directions	  .................................................................	  106	  5.1	  Discussion	  ......................................................................................................................................	  106	  5.1.1	  Circulating	  miRNA	  levels	  during	  HCV	  infection	  ........................................................................	  107	  5.1.2	  PCSK9	  and	  HCV	  ........................................................................................................................................	  109	  5.1.3	  SKI-­‐1/S1P	  and	  DENV	  ............................................................................................................................	  111	  5.2	  Future	  directions:	  further	  dissecting	  the	  roles	  and	  applications	  of	  regulators	  of	  lipid	  metabolism	  in	  viral	  pathogenesis	  .......................................................................................	  112	  5.2.1	  Investigate	  the	  effect	  of	  miR-­‐24	  and	  miR-­‐223	  on	  HCV	  infection	  .......................................	  112	  5.2.2	  Investigating	  the	  link	  between	  miR-­‐24	  and	  PCSK9	  .................................................................	  115	  5.2.3	  DENV	  hijacking	  lipid	  metabolic	  pathways	  ...................................................................................	  116	  5.2.4	  Dissecting	  the	  role	  of	  PF-­‐429242	  on	  ATF6	  activation	  in	  DENV	  infection	  ......................	  118	  5.2.5	  In	  vivo	  effect	  of	  SKI-­‐1/S1P	  inhibition	  on	  DENV	  infection	  ......................................................	  118	  5.3	  Conclusions	  ....................................................................................................................................	  119	  Bibliography:	  ..............................................................................................................................	  123	  Appendix	  1:	  Chapter	  2	  supplementary	  tables	  and	  figures	  ..........................................	  149	  Appendix	  2:	  Chapter	  4	  supplementary	  figures	  ...............................................................	  154	  Appendix	  3:	  Chapter	  5	  supplementary	  figures	  ...............................................................	  157	  	    	   vii	  List of Tables Table	  2.1.	  	  Patient	  demographics	  and	  baseline	  biochemical	  characteristics.	  ........	  47	  Table	  2.2.	  	  Correlations	  of	  circulating	  miR-­‐122,	  miR-­‐24,	  and	  miR-­‐223	  with	  biochemical	  parameters.	  ..........................................................................................................	  48	  Table	  2.3.	  ANOVA	  model	  analysis	  of	  the	  change	  in	  microRNA	  levels	  after	  treatment	  completion	  in	  SVR	  patients	  and	  relapsers.	  ....................................................	  55	  Table	  3.1.	  	  Patient	  demographics	  and	  baseline	  biochemical	  characteristics.	  ........	  65	  Table	  3.2.	  	  Correlations	  of	  extracellular	  PCSK9	  with	  biochemical	  parameters.	  ....	  70	  Table	  A1.1.	  Cohort	  characteristics.	  .....................................................................................	  149	  Table	  A1.2.	  Mixed	  model	  analysis	  of	  miR-­‐122	  against	  biochemical	  parameters.150	  Table	  A1.3.	  Mixed	  model	  analysis	  of	  miR-­‐24	  against	  biochemical	  parameters.	  ..	  151	  Table	  A1.4.	  Mixed	  model	  analysis	  of	  miR-­‐223	  against	  biochemical	  parameters.152	  	    	   viii	  List of Figures Figure	  1.1.	  PCSK9-­‐mediated	  post-­‐translational	  regulation	  of	  low-­‐density	  lipoprotein	  receptor	  (LDLR).	  .....................................................................................................	  4	  Figure	  1.2.	  	  The	  SREBP	  pathway	  activation	  by	  proteolytic	  cleavage.	  ...........................	  7	  Figure	  1.3.	  MicroRNA	  biogenesis	  and	  secretion.	  ..............................................................	  11	  Figure	  1.4.	  HCV	  genome	  and	  protein	  functions.	  ................................................................	  22	  Figure	  1.5.	  Development	  in	  hepatitis	  C	  therapy	  with	  respect	  to	  tolerability	  and	  efficacy.	  ...........................................................................................................................................	  26	  Figure	  1.6.	  	  DENV	  particle	  and	  genome.	  ...............................................................................	  33	  Figure	  2.1.	  Cohort	  overview	  and	  study	  design	  for	  miRNA	  analysis.	  ..........................	  43	  Figure	  2.2.	  Circulating	  miR-­‐24	  and	  miR-­‐223	  levels	  in	  plasma	  show	  a	  strong	  positive	  linear	  relationship	  in	  all	  CHC	  patients	  and	  negatively	  correlate	  with	  liver	  injury	  and	  liver	  fibrosis.	  ...........................................................................................................	  51	  Figure	  2.3.	  Circulating	  miR-­‐24	  and	  miR-­‐223	  plasma	  levels,	  not	  miR-­‐122,	  significantly	  increase	  in	  patients	  who	  have	  achieved	  SVR.	  ...........................................	  54	  Figure	  3.1.	  Plasma	  PCSK9	  levels	  in	  healthy	  individuals.	  ................................................	  66	  Figure	  3.2.	  Plasma	  PCSK9	  concentrations	  significantly	  increase	  in	  HCV-­‐infected	  patients	  who	  have	  achieved	  SVR.	  ...........................................................................................	  68	  Figure	  3.3.	  Extracellularly	  applied	  recombinant	  PCSK9	  and	  the	  gain-­‐of-­‐function	  mutant	  PCSK9-­‐D374Y	  but	  not	  the	  loss-­‐of-­‐function	  PCSK9-­‐R194A	  variant	  inhibits	  HCV	  infection	  in	  Huh-­‐7.5.1	  cells.	  ............................................................................................	  72	  Figure	  4.1.	  Inhibition	  of	  SKI-­‐1/S1P	  using	  PF-­‐429242	  prevents	  activation	  of	  the	  SREBP	  pathway	  and	  reduces	  the	  abundance	  of	  cytosolic	  lipid	  droplets.	  .................	  85	  Figure	  4.2.	  DENV-­‐infected	  Huh-­‐7.5.1	  cells	  accumulate	  the	  C	  protein	  around	  lipid	  droplets.	  .........................................................................................................................................	  89	  Figure	  4.3.	  Inhibition	  of	  SKI-­‐1/S1P	  using	  PF-­‐429242	  results	  in	  a	  dose-­‐dependent	  decrease	  in	  intracellular	  DENV-­‐2	  NS1	  expression	  and	  a	  3-­‐log	  decrease	  in	  extracellular	  viral	  titer.	  .............................................................................................................	  92	  Figure	  4.4.	  Inhibition	  of	  SKI-­‐1/S1P	  using	  PF-­‐429242	  results	  in	  a	  significant	  decrease	  of	  DENV-­‐2	  viral	  RNA	  pre-­‐	  and	  post-­‐establishment	  of	  DENV	  infection	  in	  Huh-­‐7.5.1	  cells.	  .............................................................................................................................	  95	  Figure	  4.5.	  DENV-­‐2	  intracellular	  viral	  RNA	  inhibited	  by	  PF-­‐429242	  can	  be	  rescued	  with	  exogenous	  oleic	  acid	  by	  increasing	  cytosolic	  lipid	  droplet	  abundance.	  .........	  99	  Figure	  4.6.	  Inhibition	  of	  SKI-­‐1/S1P	  using	  PF-­‐429242	  results	  in	  a	  robust	  decrease	  in	  intracellular	  viral	  RNA	  in	  Huh-­‐7.5.1	  cells	  for	  all	  four	  DENV	  serotypes.	  .............	  100	  Figure	  5.1.	  	  The	  intricate	  interplay	  between	  the	  host	  factors,	  viral	  infection	  and	  lipid	  metabolism.	  ......................................................................................................................	  122	  	   ix	  Figure	  A1.1.	  Differential	  changes	  of	  miR-­‐122,	  miR-­‐24	  and	  miR-­‐223	  before,	  during	  and	  after	  treatment	  relative	  to	  baseline	  levels	  in	  CHC-­‐infected	  patients	  based	  on	  treatment	  outcome.	  ..................................................................................................................	  153	  Figure	  A2.1.	  Oligonucleotide	  primers	  and	  fluorogenic	  probes	  used	  in	  the	  serotype-­‐specific	  DENV	  virus	  real-­‐time	  RT-­‐PCR	  assay.	  ................................................	  154	  Figure	  A2.2.	  Inhibition	  of	  SKI-­‐1/S1P	  using	  PF-­‐429242	  prevents	  activation	  of	  the	  SREBP	  pathway	  in	  DENV-­‐2	  infected	  Huh-­‐7.5.1	  cells.	  .....................................................	  155	  Figure	  A2.3.	  Characterization	  of	  AcPF-­‐429242	  as	  an	  inactive	  derivative	  of	  PF-­‐429242.	  .........................................................................................................................................	  156	  Figure	  A3.1.	  Intracellular	  miR-­‐122,	  miR-­‐24	  and	  miR-­‐223	  levels	  during	  HCV	  JC1	  infection	  of	  hepatoma	  Huh-­‐7.5	  cells.	  ..................................................................................	  157	  Figure	  A3.2.	  DENV	  hijacks	  the	  SKI-­‐1/S1P-­‐mediated	  proteolytic	  activation	  of	  the	  SREBP	  pathway	  during	  the	  late	  stages	  of	  the	  viral	  lifecycle	  in	  Huh-­‐7.5.1	  cells.	  ...	  159	  	    	   x	  List of Abbreviations ABCA1 – ATP binding cassette protein AI ABCG1 – Adenosine triphosphate-binding cassette subfamily G member 1 ADE – Antibody-dependent enhancement ADRP – Adipose differentiation related protein Ago2 – Argonaute 2 Alb/uPA – Urokinase genes under control of an albumin promoter ALT – Alanine aminotransferase ANGPTL3 – angiopoietin-like protein 3 ANOVA – Analysis of variance Apo – Apolipoprotein APRI – AST to platelet ratio index  AST – Aspartate aminotrasferase ATF-6 – Activating transcription factor 6 ATP – Adenosine triphosphate BART – Bam HI-A rightward transcript BHRF1 – Bam HI fragment H rightward open reading frame I BMP – Bis(monoacylglycero)phosphate BOC – Boceprevir BSA – Bovine serum albumin CAPRIN2 – Caprin family member 2 CBP – CREB-binding protein  CD81 – Cluster of differentiation 81 cDNA – Complementary deoxyribonucleic acid CDRD – Center for Drug Research and Development c.elegans – Caenorhabditis elegans CLDN1 – Claudin-1 CypA – Cyclophilin A CHC – Chronic hepatitis C COL1A1 – Collagen type I alpha 1  COL3A1 – Collagen type III alpha 1 	   xi	  DAA – Direct-acting antiviral DC-SIGN – Dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin  DENV – Dengue virus DF – Dengue fever DGAT1 – Diacylglycerol O-acyltransferase 1 DGCR8 – DiGeorge syndrome critical region gene 8 DHF – Dengue hemorrhagic fever DMEM – Dulbecco’s modified eagle medium DMSO – Dimethyl sulfoxide DSS – Dengue shock syndrome DSV – Dasabuvir EBV – Epstein-Barr virus ECM – Extracellular matrix EDTA – Ethylenediaminetetracetic acid eIF – Eukaryotic translation initiation factors ER – Endoplasmic reticulum FASN – Fatty acid synthase FBS – Fetal bovine serum FU – Follow-up GDP – Guanosine diphosphate GP – Glycoprotein GTP – Guanosine triphosphate HCC – Hepatocellular carcinoma HCMV – Human cytomegalovirus HCV – Hepatitis C virus HDL – High-density lipoproteins HGB – Haemoglobin HIV – Human immunodeficiency virus HMG-CoA – 3-Hydroxy-3methyl-glutaryl-coenzyme A reductase Hpi – Hours post infection HSC – Hepatic stellate cell 	   xii	  HSP90 – Heat shock protein 90 HSV – Herpes simplex virus Huh – Human hepatoma IAA – Indirect-acting antiviral IDL – Intermediate-density lipoprotein IFN – Interferon IL – Interleukin INSIG – Insulin-induced gene IRES – Internal ribosome entry site IRF-3 – Interferon regulatory factor 3 JAK/STAT – Janus kinase/signal transducer and activator of transcription JFH – Japanese fulminant hepatitis kDA – Kilo Dalton KSHV – Kaposi’s sarcoma-associated herpesvirus LD – Lipid droplet LDL – Low-density lipoprotein LDLR – Low-density lipoprotein receptor LED – Ledipasvir LPDS – Lipoprotein depleted serum LVP – Lipoviroparticles Mbtps – Membrane-bound transcription factor peptidase, site 1 miRNA – MicroRNA MV – Microvesicle MOI – Multiplicity of infection mRNA – Messenger ribonucleic acid MTP – Microsomal triglyceride transfer protein NAFLD – Non-alcoholic fatty liver disease NANB – Non-A, non-B hepatitis NASH – Non-alcoholic steatohepatitis NIAID – National Institute of Allergy and Infectious Diseases  NR – Non-responder  	   xiii	  NS – Non-structural NTP – Nucleoside triphosphate OBV – Ombitasvir OCLN – Occludin PABP1 – Polyadenylate-binding protein 1 P bodies – Processing bodies PBS – Phosphate buffered saline PC – Proprotein convertase PCAF – P300-CREB binding protein associated factor  PCR – Polymerase chain reaction PCSK9 – Proprotein convertase subtilisin/kexin type 9 PEG-IFN – Pegylated interferon PI4KA – Phosphatidylinositol 4-kinase PPAR – Peroxisome proliferator-activated receptors pre-miRNA – Precursor microRNA pri-miRNA – Primary microRNA PTV – Paritaprevir PUMA – P53 up-regulated modulator of apoptosis RXRα  – Retinoid X receptor alpha RBV – Ribavirin RIG-I – Retinoic acid-inducible gene I RISC – RNA-induced silencing complex RdRp – RNA-dependent RNA polymerase RIPA – Radioimmunoprecipitation assay RNA – Ribonucleic acid ROS – Reactive oxygen species S2P – Site-2 protease SCAP – SREBP cleavage-activating protein SCID – Severe combined immunodeficiency SDS – Sodium dodecyl sulphate SIM – Simeprevir 	   xiv	  siRNA – Small interfering RNA SKI-1/S1P – Subtilisin kexin isozyme-1/site-1 protease SOF – Sofosbuvir SRE – Sterol response element SREBP – Sterol regulatory element binding protein SR-BI – Scavenger receptor class B member 1 SVR – Sustained virological response TGF-ß – Transforming growth factor beta TRBP – Trans-activation response RNA-binding protein TVR – Telaprevir UTR – Untraslated region VLDL – Very low-density lipoprotein VLDLR – Very low-density lipoprotein receptor VP – Vesicle packets WBC – White blood cells WNV – West Nile virus WT – Wild-type Xbp-1 – X-box binding protein 1 XRN1 – 5'-3' exoribonuclease 1 YFV – Yellow fever virus    	   xv	  Acknowledgements  To my supervisor Dr. François Jean, thank you for providing great opportunities to learn and grow as a scientist and for allowing me to pursue such interesting research.  To my CanHepC mentor Dr. Mel Krajden, thank you for believing in me and giving me the opportunity to be a part of the CanHepC network.   I would like to thank all the current and previous members of the CanHepC network for their suggestions on the HCV project.  To all the current and former Jean lab members, thank you for all your help and support over the years.  In particular, thank you to Steven McArthur for all the valuable discussions.   I would like to thank committee members Drs. Georgia Perona-Wright and Wayne Vogl for your support, constructive criticism and creative suggestions throughout this thesis work.    To my Women in Science mentor Dr. Kara Carter, thank you for giving me valuable advice on career development.   I would like to thank Darrel Cook, Dr. Andrea Olmstead, Lori Lee Walston for help with clinical samples and data.  I would like to thank all the members of the Department of Microbiology and Immunology for creating a welcoming environment. Darlene Birkenhead, thank you for providing me the answers to all my questions within minutes, for caring and being so supportive.    A thank you to my parents who always encouraged me to become the independent woman I am today.  To my sister, for all the help and support since the beginning. I am so proud to have a sister like you.  To Nicolas, for being there for me through the toughest times, always finding words of encouragement and making me laugh. To my friends, including Sandra, Citlali, Kateryna and Christina, for all the fantastic memories we shared, for your friendship and love.    	   xvi	  Dedication To the best sister in the world, Genya Hyrina ♥ 	  	   1	  Chapter 1: Introduction  1.1 Host lipid metabolism Under normal physiological conditions, lipid input into the body is equal to lipid output via secretion or oxidation as an energy source.  Disruption of either the input or output pathways can result in dyslipidemia that increases the risk of cardiovascular diseases (1).  A number of enveloped viruses utilize and modulate host lipid metabolic pathways to support their lifecycles (2).  However, the regulatory mechanisms underlying the dysregulation of lipid pathways in these disease states are incompletely understood.   The liver is the body’s main metabolic organ, and it plays an essential role in regulating fatty acid and cholesterol metabolism.  Lipid metabolism involves several interdependent pathways.  Hepatic fatty acids can be derived from de novo lipogenesis, hydrolysis of triglycerides from cytoplasmic lipid droplets (LDs), direct uptake of fatty acids derived from triacylglycerols of lipoprotein remnants and uptake of non-esterified fatty acids released by adipose tissue (3).  The liver is also responsible for processing and packaging lipids into lipoproteins for transport of triglycerides and cholesterol through the blood.  Similarly, cholesterol also can be derived from de novo synthesis or absorbed from the diet, and are transported into circulation as lipoprotein particles.  Cholesterol can be stored in cells as cholesterol esters or metabolized into bile acids.  Coordinated regulation of each of these pathways is essential for the maintenance of normal lipid homeostasis.   1.1.1 Lipoproteins Lipids are stored intracellularly in hepatocytes as triglycerides and cholesterol esters, organised in specific organelles called LDs.  These LDs are dynamic lipid storage organelles made up of a core of neutral lipids surrounded by a phospholipid monolayer and specific perilipin proteins (4).  Although fat may accumulate substantially in the liver, it does not occur indefinitely.  Any excess cholesterol or fatty acids not required by the liver are 	   2	  packaged into lipid-carrying vehicles called very low-density lipoproteins (VLDLs) for secretion into the plasma and delivery to extrahepatic tissues (1).  VLDLs are assembled within the endoplasmic reticulum (ER) where triglycerides are incorporated into an apolipoprotein B (ApoB) backbone by the microsomal triglyceride transfer protein (MTP) (1).  Cholesteryl esters are incorporated into the lipoprotein core by unclear mechanisms.  Final lipidation and maturation of VLDL takes place in the Golgi apparatus and may involve fusion of LDs with assembling VLDL prior to secretion of these particles.  The mature VLDL particles consist of a core of neutral lipids, stabilized by the ApoB backbone that is surrounded by a phospholipid monolayer.  Following secretion of VLDL, their fatty acids are absorbed by tissues when the triglycerides within the core of the lipoprotein particles are hydrolyzed by lipases.  The absorption process results in particles of chylomicron, VLDL remnants, and intermediate-density lipoprotein (IDL).  The IDL can interact with hepatic lipase to become more dense and enriched with cholesteryl esters, generating low-density lipoprotein (LDL).  LDLs are the main carriers of cholesterol in human circulation and are taken up into target cells exclusively by the LDL receptor (LDLR) (1).  	   Although VLDL particles mediate the delivery of triglyceride and cholesterol molecules from the liver to muscle and fat tissue, another class of lipoproteins called high-density lipoproteins (HDL) mediate the transport of excess cholesterol from extrahepatic tissues to the liver for elimination via the bile.  HDLs are small particles, containing the least lipid and the most protein relative to other classes of lipoproteins.  HDLs are formed when newly secreted ApoA interacts with ATP binding cassette protein AI (ABCA1) on the surface of hepatocytes.  This promotes the incorporation of phospholipid and cholesterol from hepatocyte plasma membranes and the formation of spherical HDL that receive more cholesterol from cell membranes of peripheral tissues.  At the liver, HDL particles bind to scavenger receptor, class B type I (SR-BI), which promotes hepatic uptake of only the cholesteryl esters (1).   1.1.2 Low-density lipoprotein receptor  Cellular lipid and cholesterol metabolism depends on sequential cell surface binding, internalization, and intracellular degradation of plasma LDL.  LDL uptake is mediated by the 	   3	  LDLR expressed in clusters in coated pits on the cell surface (5).  The so-called ‘LDL-receptor pathway’ was characterized in detail by Michael S. Brown and Joseph L. Goldstein, who received the Nobel Prize in Physiology or Medicine in 1985 for their tremendous contribution to the field of cholesterol metabolism (6).  LDLR binds ApoB in LDL particles present in the extracellular fluid via LDLR’s extracellular domain, which induces clathrin-dependent endocytosis of the bound lipoprotein.  The complex is transported via early endosomes to the late endosomal compartment, where the acidic environment causes dissociation of the receptor–ligand complex.  The receptor is recycled to the cell surface while the LDL particle is degraded in the lysosomal compartment (7).  The physiological importance of the LDL/LDLR interaction is demonstrated in people with autosomal dominant mutation in LDLR or ApoB, which interferes with LDLR-mediated LDL uptake and degradation of LDL.  This mutation results in a genetic disorder called familial hypercholesterolemia, which is characterized by an excessive deposition of cholesterol in tissues that leads to accelerated atherosclerosis and increased risk of coronary heart disease (6).  1.1.3 Proprotein convertase subtilisin/kexin type 9  In early 2003, the proprotein convertase subtilisin/kexin type 9 (PCSK9) was identified as the third major locus contributing to autosomal dominant hypercholesterolemia after LDLR and ApoB (8).  PCSK9 was found as the major post-translation regulator of LDLR levels in the liver (9-11) (Figure 1.1).  PCSK9 is most abundant in hepatocytes and in small intestine enterocytes where it binds LDLR at the cell surface separately from the LDL binding domain.  Following receptor-mediated endocytosis, the tight binding of PCSK9 to LDLR prevents LDLR recycling back to the plasma membrane, which results in LDLR degradation in lysosomes (8).  Mutations in PCSK9 therefore affect liver LDLR cell surface expression and disrupt normal uptake of plasma cholesterol into the liver.  Gain-of-function mutations in PCSK9 have been associated with hypercholesterolemia due to lower levels of LDLR and reduced clearance of plasma LDL (12-14).  Loss-of-function PCSK9 mutations conversely are associated with abnormally low circulating cholesterol levels due to increased LDLR abundance on the surface of liver cells (15, 16).  	   4	       Figure 1.1. PCSK9-mediated post-translational regulation of low-density lipoprotein receptor (LDLR).  LDLR is a cell surface transmembrane protein, which plays a role in internalization of lipoproteins from plasma.  PCSK9, a member of the proprotein convertase family, is synthesized in the liver and secreted to the plasma through the secretory pathway.  At the cell surface PCSK9 binds LDLR, which mediates endocytosis of the complex and prevents LDLR recycling.  Instead, LDLR is degraded in endosomal and lysosomal compartments, which leads to a decreased number of LDLRs on the surface of cells.  Adapted from (17).   	   5	  PCSK9 is the ninth and the last identified member of the proprotein convertase (PC) family of secretory pathway serine proteases that are required for maturation and activation of a variety of host and pathogen precursor propeptides [reviewed in (18)].  PCSK9 is synthesized as a zymogen proprotein in the ER where it undergoes autocatalytic cleavage of its N-terminal prosegment prior to secretion and cell surface association with LDLR.  Unlike in the other PC members, the prosegment of PCSK9 remains non-covalently bound to the active site of the protein throughout secretory pathway transit and secretion, steadily inhibiting its potential catalytic activity.  Thus, PCSK9 has no other substrate than itself, and its activity in regulating LDLR levels is related to its binding and escorting the resulting complex toward the lysosomal compartment (8).  PCSK9 expression is upregulated by sterol depletion, specifically by the sterol regulatory element binding proteins (SREBPs), the master transcriptional factors in lipid biosynthetic pathways (19).  Surprisingly, both PCSK9 and LDLR are upregulated in response to low sterol levels and statin treatments.  This suggested that statins would be more effective at decreasing LDL if not for the associated rise in PCSK9 levels after statin therapy.  Indeed, this was confirmed in PCSK9-knockout mice as well as in individuals harboring the common loss-of-function R46L PCSK9 mutant (15, 20).  As high LDL-cholesterol (LDL)-C is a major contributor to atherosclerosis and coronary heart disease, many therapeutic strategies for inhibiting the expression, function or LDLR-binding activity of PCSK9 are being developed (21).  Recently, two drugs targeting PCSK9 were approved for treating patients with familial hypercholesterolemia.  Alirocumab and evolocumab are monoclonal antibodies that bind to PCSK9, lowering LDL by about 50–60%.  These drugs are approved for use in patients with cardiovascular disease or familial hypercholesterolemia whose LDL cholesterol levels are insufficiently controlled with other agents, such as statins.  Although definitive clinical efficacy and long-term safety data are still needed, antibody-based PCSK9 inhibitors promise to meet much of the unmet medical need in the treatment of raised LDL cholesterol.  Also, several additional approaches to inhibiting PCSK9 are in clinical development, such anti-PCSK9 siRNA and anti-PCSK9 adnectin (engineered target binding proteins) (21).   	   6	  1.1.4 The SREBP pathway  Hepatic lipid homeostasis is controlled by numerous transcription factors and nuclear receptors.  The SREBPs are master regulators of lipid homeostasis and play a critical role in de novo lipid biosynthesis (22) (Figure 1.2).  SREBP-1 controls fatty acid and triglyceride biosynthesis, while SREBP-2 controls cholesterol metabolism, LDLR and PCSK9 expression.  SREBPs are basic-helix-loop-helix-leucine zipper transcription factors that are bound to the ER membranes as an inactive precursor (23).  Their activation is dependent on the presence of sterols and requires the cleavage of pre-SREBP to release the mature form, which then translocates to the nucleus.  When sterol levels are low, the SREBP cleavage-activating protein (SCAP) escorts pre-SREBP from the ER to the Golgi apparatus, where pre-SREBP is sequentially cleaved by two cellular proteases: subtilisin kexin isozyme-1/site-1 protease (SKI-1/S1P) and site-2 protease (S2P) (24).  These cleavages liberate the active, N-terminal fragment of SREBP (n-SREBP), which further translocates into the nucleus where it regulates the expression of various genes involved in lipid metabolism by binding to the sterol regulatory element (SRE) of its target genes, such as LDLR and PCSK9 (19, 23, 25).  When cellular sterol levels are high, the insulin-induced gene protein (INSIG) associates with SCAP, which causes the SCAP–pre-SREBP complex to be retained in the ER, thereby preventing the formation of n-SREBP and decreasing the expression of SREBP target genes (26).   1.1.5 Subtilisin kexin isozyme-1/site-1 protease   The requirement for SKI-1/S1P to mediate an essential cleavage event in SREBP activation, gives SKI-1/S1P a critical role in host lipid metabolism.  SKI-1/S1P, like PCSK9, is also a member of the proprotein convertase family (18).  SKI-1/S1P is a ubiquitously expressed transmembrane secretory pathway enzyme that cleaves after amino acid consensus motif RX(V,L)(K,F,L) (27).  In the ER, SKI-1/S1P undergoes autocatalytic cleavages to release the N-terminal propeptide domain yielding a mature ~106 kDa, membrane-anchored protein.  In some cases, an additional cleavage occurs, causing shedding of the SKI-1/S1P ectodomain.  Release of the ectodomain is required for SKI-1/S1P activity (28, 29).  Following activation, SKI-1/S1P moves to the Golgi apparatus where it can mediate   	   7	      Figure 1.2.  The SREBP pathway activation by proteolytic cleavage.   SREBP precursors are retained in the ER membranes through a tight association with SCAP and a protein of the INSIG family when sterol levels are high. When sterol levels are low, SCAP dissociates from INSIG and escorts the SREBP precursors from the ER to the Golgi apparatus. Once there, two proteases, SKI-1/S1P and S2P, sequentially cleave the precursor protein, releasing the mature N-terminal fragment of SREBPs (n-SREBP) into the cytoplasm. n-SREBP then migrates to the nucleus, where it activates sterol response element (SRE) promoter of genes involved in lipid metabolism, such as LDLR and PCSK9. Adapted from (30)    	   8	  proteolytic cleavages of its targeted substrates.  The most well studied substrates are the SREBP family members but other substrates have been identified, including activating transcription factor 6 (ATF-6), which functions as a transcription factor to promote the expression of downstream target genes involved in ER stress (31).  The importance of SKI-1/S1P in regulating plasma cholesterol and lipid metabolism is illustrated in liver-specific SKI-1/S1P knockout in mice, which show 50% reductions in circulating cholesterol and fatty acid levels (32, 33).  Furthermore, these studies suggest that targeting SKI-1/S1P activity specifically in the liver is a viable therapeutic opportunity for the treatment of hypercholesterolemia.  Interestingly, SKI-1/S1P has also been implicated in the cleavage of haemorrhagic fever virus glycoproteins (34-36) and in regulation of hepatitis C virus (HCV) infection (37), and it is possible that therapeutic manipulation of SKI-1/S1P may also be effective in these viral diseases.   1.1.6. Viral dysregulation of host lipid homeostasis  Although many host lipid metabolic pathways were first identified and characterized over 70 years ago, only recently have they been examined in the context of viral infection.  Hijacking of host lipids and their biosynthetic pathways is a common strategy utilized by many viruses and there are several known functions that host lipids serve during viral infection (38).  First, lipid membranes are structural components of all enveloped viruses.  Second, lipid bilayers often function as platforms during multiple steps of the virus lifecycles, including viral gene expression, replication of the viral genome and assembly of virus particles.  RNA viruses are known for remodelling host intracellular membranes to compartmentalize viral components, thus protecting virions from attacks by the host defense system and increasing the efficiency of viral replication by concentrating its components.  The altered lipid membranes also provide a scaffold for the assembly of virus replication complexes.  Finally, lipid-mediated signal transduction serves as a regulator of the host response to viral infection as well as viral pathogenesis (38).  Lipids have been demonstrated to be critical to the lifecycles of two highly important human viruses, HCV and dengue virus (DENV), which are the focus of this project.  The mechanisms in which these two viruses hijack host lipid metabolism to support their lifecycles will be reviewed in detail later. 	   9	  1.2 MicroRNAs  MicroRNAs (miRNAs) regulate diverse biological functions including lipid metabolism and have been recognized as potential targets in drug treatment development.  MiRNAs are small non-coding RNAs of 18-24 nucleotides in length that target mRNAs with imperfect complementarity and interfere with the process of translation (39).  Such interference disrupts production of specific proteins in a cell-intrinsic manner. MiRNAs can also regulate protein expression in distant cells, via their secretion as extracellular miRNAs.  Extracellular miRNAs enter the systemic circulation and are stable in plasma because they are protected within microvesicles, such as exosomes, or due to their association with RNA-binding proteins and lipoprotein complexes (40).  Such extracellular miRNAs correlate quantitatively with various pathogenic states (41).    1.2.1 Discovery of microRNAs   The first small non-coding RNA lin-4 with its antisense complementary sites in the 3'-UTR of the lin-14 gene was discovered by the Ambros and Ruvkun laboratories in 1993 (42, 43).  Both laboratories found that lin-4, a gene known to control the timing of Caenorhabditis elegans (C. elegans) larval development, did not code for protein but instead produced a pair of small RNAs of 22 and 61 nucleotides, where the longer one was predicted to be the precursor of the shorter one.  Both small RNAs repressed the translation of lin-14 mRNA by pairing with its 3'-UTR, which supported a part of the regulatory pathway that triggers the transition between the first and the second stage in C. elegans larval development (42, 43).   The second miRNA discovery did not happen until seven years later, when the Ruvkun laboratory identified another miRNA let-7 as a regulator of lin-41 during larval development from late larval to adult cell stage (44).  Afterwards, homologues of let-7 were identified in the human and fruit fly genomes, which led to discovery of hundreds of miRNAs in various species.  The latest release of the miRNA database has catalogued 434 miRNAs in C. elegans, 466 miRNAs in Drosophila melanogaster and 2,588 miRNAs in humans, although the functional importance of many of these miRNA annotations remains to be determined (miRBase v. 21). 	   10	  1.2.2 MicroRNA biogenesis Cellular miRNAs are transcribed in the nucleus by RNA polymerase II to primary miRNA (pri-miRNA) transcripts (45, 46) (Figure 1.3).  Pri-miRNAs are double-stranded stem loop structures of 100 - 1000 nucleotides in length and are processed to >60–70 nucleotide precursors (pre-miRNAs) by the nuclear RNAse-III type endonuclease Drosha (47).  Drosha functions together with its cofactor the DiGeorge syndrome critical region gene 8 (DGCR8) protein to process the pri-miRNAs into pre-miRNAs (48).  Pre-miRNAs are then exported to the cytoplasm via nuclear pore complexes by exportin-5 and its cofactor Ran, which converts GTP to GDP to facilitate release of the miRNA precursors into the cytoplasm (49-51).  Once there, pre-miRNAs are cleaved into ~20 base-pair miRNA/miRNA* duplexes by the RNase-III type enzyme Dicer and its cofactor trans-activation response RNA-binding protein (TRBP).  In mammals, Dicer is supported by Argonaute 2 (Ago2), a RNaseH-like endonuclease that cleaves the 3' overhanging nucleotides of pre-miRNAs, thus generating mature miRNAs (52-54).  One strand of the miRNA duplex is then incorporated into the RNA-induced silencing complex (RISC) where it directly binds to a member of the Argonaute protein family, whereas the other strand (miRNA*) in most cases is released and degraded (55, 56).   1.2.3 Post-transcriptional repression by microRNAs   Base-pairing between the miRNA and target mRNA in the RISC mediates either translational repression or mRNA degradation.  While in plants miRNAs normally base-pair to mRNAs with nearly perfect complementarity that triggers endonucleolytic mRNA cleavage (57), metazoan miRNAs in most cases pair imperfectly with their targets and result in translational repression (58-62).  Metazoan miRNAs appear to follow a set of rules regarding the targeting of mRNAs, as determined by experimental and bioinformatics analysis.  The primary requirement is a contiguous and perfect based-paring of the miRNA nucleotides 2 through 8, known as the ‘seed’ region.  Also, the majority of metazoan miRNA-binding sites are present in multiple copies in the 3'-UTR region of targeted mRNA, which is necessary for effective repression of translation (58-62).  	   11	   Figure 1.3. MicroRNA biogenesis and secretion.  Primary miRNA transcripts (pri-miRs) are originally transcribed from introns of protein-coding genes by RNA polymerase II.  The pri-miRNA is processed by the nuclear RNase Drosha and its co-factor DGCR8 to produce 70-nucleotide stem loop precursor miRNAs (pre-miRNAs).  The pre-miRNAs are exported out of the nucleus to the cytoplasm by exportin-5, where they are processed by the cytoplasmic RNase Dicer into a mature miRNA duplex.  One of the double-strand mature miRNAs is selectively loaded into the RISC complex, which contains the Argonaute (Ago) family protein as a core component.  In the cytoplasm, mature miRNAs predominantly bind to the 3'-UTR region of the target mRNA, and repress its expression through mechanisms of both translational repression and mRNA destabilization.  MiRNAs are also secreted and stably circulate in the extracellular space within vesicles, which include apoptotic bodies and exosomes, or in conjunction with RNA-binding proteins, such as Ago and lipoproteins.  Adapted from (63).  	   12	   The precise mechanism for miRNA induced repression of translation is poorly understood but extensive evidence suggests that miRNAs act at the translational initiation step (64).  Translation of mRNA can be divided into three primary steps: initiation, elongation and termination.  Initiation starts with the recognition of the mRNA 5'-terminal cap structure (m7Gppp) by the eukaryotic translation initiation factors (eIF).  Further interaction between the eIF and the polyadenylate-binding protein 1 (PABP1) circularizes mRNA, facilitates the recruitment of the 40S and 60S ribosomal subunit and initiates the elongation phase of translation (65-68).  Experimental evidence suggests that miRNAs exert their inhibitory effect on translational initiation by recruiting several factors and enzymes for mRNA cleavage and degradation including decapping enzymes, deadenylase, 3' and 5' exonucleases, and endonucleases.  Also, the Argonaute protein competes with cap binding proteins and eIF4E for binding to cap structure and inhibits translation initiation by interfering with mRNA circularization and assembly of the 40S initiation complex (69-72).  It is likely that miRNAs can inhibit translation at the post-initiation step as well.  Several proposed mechanisms include miRNA induction of ribosome drop-off, competition with elongation factors or recruitment of mRNA/protein degradation enzymes and decay factors such as exosome complexes (64).    1.2.4 Circulating microRNAs  In 2008, Chim et al. reported that placental miRNAs can be detected in maternal blood plasma and were the first to report that miRNAs are present in biological fluids (73, 74).  Since then, miRNAs have been found in all biological fluids from saliva to urine (74).  This generated much excitement in the field of extracellular miRNAs and suggested a potential novel class of disease biomarkers and possible drug targets.  Plasma miRNAs were found to be remarkably stable even under conditions as harsh as boiling, low or high pH, long-time storage at room temperature, and multiple freeze-thaw cycles (75).  Possible explanations for this unforeseen stability is miRNA association with RNA-binding proteins (Ago2) (76), lipoprotein complexes (HDL and LDL) (77) or packaging within microparticles [exosomes, microvesicles (MVs), and apoptotic bodies] (78). 	   13	   El-Hefnawy et al. were among the first to demonstrate that plasma RNA is protected from degradation by its inclusion in protein or lipid vesicles (79).  Depending on their size and mode of release from cells, these particles are known as exosomes, MVs, or apoptotic bodies.  Exosomes are small vesicles (50-100 nm) that originate from the endosome and are released from cells when multivesicular bodies fuse with the plasma membrane.  MVs are membranous vesicle that are larger (0.1-1 µm) than exosomes and are released from the cell through protrusion of the plasma membrane.  Apoptotic bodies are the largest microparticles (0.5-2 µm) and are shed from cells during apoptosis (75, 80).  Several studies reported that cells can select some miRNAs for cellular release via microparticles while others are retained (81); however, further studies are needed to dissect the underlying mechanism of miRNA selection.  There is accumulating evidence that the majority of circulating miRNAs are not found inside microvesicles but rather bound to RNA-binding proteins.  Arroyo et al. used differential centrifugation and size-exclusion chromatography and demonstrated that up to 90% of miRNAs in the circulation are present in a non-membrane–bound form primarily associated with Ago2 (76).  Interestingly, while some miRNAs (eg. let-7a) were exclusively associated with vesicles, some miRNAs (eg. miR-122) were detected only in protein-associated fractions, which may reflect different cell type-specific miRNA release mechanisms.  Since miR-122 (a liver-specific miRNA) was detected only in Ago2 complexes but not in association with vesicles, this suggests that hepatocytes release miR-122 through a protein carrier pathway only (76).  However, recently miR-122 has been identified in exosomes in complexes with replication-competent HCV RNA, Ago2 and heat shock protein 90 (HSP90), from HCV-infected hepatoma Huh-7.5 cells and HCV-infected patients (82), suggesting that miR-122 bound to the HCV genome can be recruited to exosomes and involved in receptor-independent HCV transmission (the role of miR-122 during HCV lifecycle will be discussed in detail in section 1.3.4).  It has been proposed that Ago2-miRNA complexes are passively released by death of apoptotic cells and remain in the extracellular space because of the high stability of the Ago2 protein; however, this hypothesis does not rule out the possibility of cell membrane-associated channels or receptors mediating the specific release of the Ago2-miRNA complexes (75).   The most recent findings on miRNA secretion pathways showed that miRNAs could 	   14	  also be transported via lipoproteins (HDL and LDL) (77).  HDL particles have an average size of 8 to 12 nm, which makes them substantially smaller than exosomes.  Furthermore, they contain lipids, such as phosphatidylcholine, that are known to form stable ternary complexes with nucleic acids.  They also contain ApoA as a main constituent, which has been used for the systemic delivery of small interfering RNAs in animal models.  Vickers et al. profiled miRNAs in purified HDL, LDL, and exosome pools from human plasma and revealed that the HDL-miRNA profile was distinct from the exosome-miRNA profile (77).  Also, they found that HDL from familial hypercholesterolemia subjects had a higher concentration of miRNAs and contained more individual miRNAs than HDL from healthy subjects.  One of the most abundant miRNAs in HDL and LDL is miR-223.  This is also the most regulated HDL-associated miRNA and has been shown to be increased >3,000-fold in patients with human familial hypercholesterolemia, who are at high risk of developing atherosclerosis (77).  Another group reported that HDL had >10,000 copies of miR-223 per µg of HDL total protein, while LDL had 1,500 copies per µg of LDL (83).  The presence of miRNAs in circulation has also raised the intriguing idea that they could have a function in cell-to-cell communication.  This would mean that miRNAs are selectively targeted for secretion in one cell and taken up by a distant cell, possibly to regulate gene expression.  Indeed, accumulating evidence suggests that miRNAs may function as mediators of cell-to-cell communication (84, 85).   Another area of great interest is in the use of circulating miRNAs as clinical biomarkers.  With miRNAs controlling multiple aspects of cellular protein expression, dysregulation of miRNAs is associated with numerous human pathologies, including cancers, liver disease and viral infections (40, 86, 87).  Circulating miRNAs fulfill a number of criteria of ideal biomarkers, including accessibility through non-invasive methods, a long half-life within the sample, a high degree of specificity and sensitivity, the ability to differentiate pathologies, and allowance of early detection; and numerous studies have been investigating miRNAs as circulating biomarkers for diagnosis or prognosis of various pathologies (75).   	   15	  1.2.5 Emerging role of microRNAs in regulating lipid metabolism  Emerging evidence demonstrates that miRNAs are critical regulators of lipid synthesis, fatty acid oxidation and lipoprotein formation and secretion.  A number of representative miRNAs have been shown to have a significant impact on lipid homeostasis (88).  For example, miR-122 is known to affect various genes involved in hepatic cholesterol and lipid metabolism.  Inhibition of miR-122 using antisense approaches results in reduction of plasma cholesterol levels in mice and chimpanzees (89-91).  A reduction in hepatic miR-122 expression has been reported in both human non-alcoholic steatohepatitis (NASH) and animal models of this disease.  Mice deficient in miR-122 had lower levels of serum cholesterol, LDL, HDL and serum triglyceride than wild-type mice, but still developed steatohepatitis and HCC (92, 93).    Another extensively investigated miRNA, miR-33, modulates genes involved in lipid metabolism (94).  The family of miR-33 includes two members, miR-33a and miR-33b located in intronic regions within SREBP-2 and SREBP-1 respectively (95).  Both miRNAs are co-transcribed with their host genes and regulate similar physiological processes.  Silencing of miR-33 in vivo increases plasma HDL levels by targeting ABCA1 and adenosine triphosphate-binding cassette subfamily G member 1 (ABCG1), thereby attenuating cholesterol efflux to ApoA or reducing cholesterol efflux to nascent HDL (96, 97).   Both miR-27a and miR-27b are demonstrated to target retinoid X receptor alpha (RXRα), which plays a central role in adipogenesis and regulates fat metabolism (98).  Overexpression of miR-27a accelerates adipolysis, by releasing more glycerol and free fatty acids in the adipocytes, and represses lipid storage in cells (99).  In addition, miR-27a inhibits the expression of many lipid metabolic genes, including fatty acid synthase (FASN), SREBP-1, SREBP-2, peroxisome proliferator-activated receptors PPARα and PPARγ, as well as ApoA, ApoB and ApoE (100).  Thus, miR-27 may regulate lipid metabolism by reducing lipid synthesis and increasing lipid secretion from cells (88, 100).  Another miRNA of the miR-27 gene cluster, miR-24, was shown to target INSIG1 which would have a positive effect on cellular SREBP levels (101).  The authors of this study showed that antagonizing miR-24 in diet-induced obese mice significantly reduced plasma and hepatic lipid levels through down-regulation of lipogenic gene expression (101).  Higher 	   16	  levels of miR-24 and lower levels of INSIG1 were also observed in non-alcoholic fatty liver disease (NAFLD) or NASH patients (102, 103), suggesting that crosstalk between miR-24 and INSIG1 may be important for controlling lipid homeostasis in metabolic diseases (104).  Differential miR-223 expression has also been linked to both human and murine obesity.  Zhuang et al. showed that high-fat diet-fed miR-223 knockout mice had worsened insulin resistance and that miR-223 deficient macrophages showed increased inflammatory potential compared with miR-223 containing macrophages (105).  The increased inflammatory stress in knockout animals was hypothesized to exacerbate obesity-related metabolic disease through derepression of PBX/knotted 1 homeobox 1 (Pknox 1).  Also, as previously discussed, HDL isolated from patients with familial hypercholesterolemia had 3,781-fold greater miR-223 than HDL from normal patients (77). The functional significance of differential plasma miRNA expression in metabolic disease is unknown at this time, though a repressive effect by miR-223 delivered by an exosomal fraction has been demonstrated experimentally (106).  Finally, miR-223 was shown to be robustly expressed in the liver and to regulate liver cholesterol biosynthesis and HDL uptake through repression of various targets (107), suggesting a role in obesity related cardiovascular disease (108).  1.2.6 Role of microRNAs during viral infection  Given their ability to regulate a wide variety of cellular processes, it is not surprising that miRNAs have been identified as key players in host-pathogen interactions.  During viral infection, both virally-encoded and host miRNAs may promote or limit viral replication, and may contribute to disease progression and outcome (109).  Human DNA viruses, mainly herpesviruses, encode a number of viral miRNAs that can modulate the cellular environment to allow viruses to evade the immune response and regulate viral latency (109, 110).  In 2004, the first virus-encoded miRNAs were identified in B cells latently infected with Epstein-Barr virus (EBV) (111).  To date the majority of virally encoded miRNAs have been found in EBV and Kaposi’s sarcoma-associated herpesvirus (KSHV), as well as in human cytomegalovirus (HCMV) and the herpes simplex viruses (HSV) (110, 112).   As an example, EBV is known to encode ~40 miRNAs that are located in two clusters: one cluster, Bam HI fragment H rightward open reading frame I (BHRF1), encodes 	   17	  four miRNAs and another cluster, Bam HI-A region rightward transcript (BART), encodes 36 miRNAs (113).  The miRNAs in these clusters are expressed at different levels based on the virus latency state.  EBV can exhibit one of three latency programs: latency I, latency II, or latency III. Each latency program leads to the production of a limited, distinct set of viral proteins and viral RNAs.  BHRF1 miRNAs are exclusively expressed during latency III, whereas BART miRNAs are expressed primarily during latency II (110).  BART miRNAs are also highly expressed in all of the tumour types associated with EBV, especially the carcinomas, suggesting that these miRNAs may play a role in tumour development (114, 115).  Examples include BART5 that represses the p53 up-regulated modulator of apoptosis (PUMA), thereby protecting EBV-infected cells from virally-induced apoptosis (116) and BART13* that likely plays an anti-apoptotic role by targeting the Wnt-signaling enhancer caprin family member 2 (CAPRIN2) (117). For RNA viruses, the lack of nuclear localization by most RNA viruses, the smaller genome size, and the error prone nature of RNA-dependent RNA polymerase (RdRp) may constrain the genomic incorporation of viral miRNAs.  However, there is strong evidence that host miRNAs can both enhance and restrict viral replication.  Cellular miRNAs have been shown to directly bind to human immunodeficiency virus type 1 (HIV-1) RNA and inhibit viral replication.  Huang et al. reported that a group of cellular miRNAs (miR-28, miR-125b, miR-150, miR-223 and miR-382) targets the 3'-UTR of HIV-1 mRNA in resting CD4+ T cells, restricting HIV-1 infection and possibly contributing to latency (118).  Another group has shown that miR-29 binds HIV-1, increasing viral incorporation into processing bodies (or P bodies, cytoplasmic domains that contain proteins involved in post-transcriptional processes such as mRNA degradation, translational repression or RNA-mediated gene silencing), thereby inhibiting translation of viral proteins and viral replication (119).  Many cellular miRNAs were also shown to regulate HIV-1 infection indirectly, through the modulation of the levels of HIV dependency factors such as cyclin T1, an important component of the eukaryotic RNA polymerase II elongation complex, and the HIV-1 protein Tat and its transcriptional co-activators, CREB-binding protein (CBP)/p300	  and p300-CREB binding protein associated factor (PCAF) (120).  Picornaviruses also modulate cellular miRNAs during their lifecycle.  Upon enterovirus 71 (EV71) infection, miR-141 is induced and targets eIF4E, a key element in 	   18	  cap-dependent translation (121).  Since picornaviruses contain an internal ribosome entry site (IRES) that does not require cap dependent factors such as eIF4E, up-regulation of miR-141 results in reduced cap-dependent and enhanced cap-independent translation, thus inhibiting cellular mRNA translation and favouring viral protein production.  Inhibition of miR-141 during EV71 infection was shown to dramatically reduce virus production, further suggesting that expression of this cellular miRNA is essential for the virus’ lifecycle (121).   1.3 Hepatitis C virus (HCV)  Several important human viruses have been shown to alter lipid metabolism in their host, and HCV is a canonical example.  HCV is a significant health burden in North America and worldwide.  The global prevalence of anti-HCV antibodies is estimated at 1.6% (1.3–2.1%) of the population, corresponding to 115 million (92–149 million) infections, with the majority of these infections occurring in adults (122).  Central and East Asia, North Africa, and the Middle East have the highest HCV prevalence (>3.5%), with moderate prevalence in Eastern and Western Europe (1.5–3.5%) and lowest prevalence in North America (<1.5%) (123).  HCV hijacks and manipulates host cholesterol and lipid metabolism to support every step of its lifecycle, which is highly effective since the liver is the main site of virus replication.  The successful persistence of HCV in the human liver is a principal cause of HCV-associated pathology.  HCV-associated infection causes around 350,000 deaths per year, related to complications such as cirrhosis and liver cancer (124).  For many years, treatment for chronic HCV, consisting of a combination of subcutaneous injections of pegylated interferon (PEG-IFN) and oral ribavirin (RBV), has been inadequate (success rates of ~50%, depending on genotype) and associated with significant adverse events (125).  Recent years have seen a revolution in HCV therapeutic development, with the advent of highly effective, curative, direct-acting antiviral therapies that selectively target HCV and the move toward interferon-free regimens (126).  Even with the recent tremendous advances in treatment of chronic HCV infection, individuals with late stage liver fibrosis remain at an ongoing risk of cirrhosis and hepatocellular carcinoma even after they are cured by treatment (127).  A better understanding of the key regulators of lipid metabolism that are modulated during the course of HCV infection could help to elucidate mechanisms of progressive liver 	   19	  disease and enable the discovery of therapeutic targets that are effective against progressive liver fibrosis. 	  1.3.1 HCV discovery and diversity   In 1989, the genome of the virus causing non-A, non-B (NANB) hepatitis was identified in a cDNA library derived from the plasma of a chimpanzee infected with material from a patient with NANB hepatitis, which was designated HCV (128).  Prior to the identification of the causative agent of NANB hepatitis, blood units were not screened for contamination with HCV and this omission let the virus spread around the world via contaminated blood supplies used for transfusions.  Since the discovery of the virus, screening of blood supplies for HCV became a universal requirement that significantly reduced the risk of acquiring HCV in developed countries, but not in developing countries that do not constantly screen blood products for HCV.  Also, there have been isolated outbreaks of HCV infection resulting from contaminated and inadequately sterilized syringes and needles in clinics, physician offices and tattooing salons (129).  Egypt is the country reporting the highest prevalence of HCV infections (15-20%), which is attributed to past widespread schistosomiasis (a disease caused by parasitic worms) treatment campaigns utilizing unsterilized glass syringes (130).  Injection drug use is currently the main mode of HCV transmission in developed countries including Canada (129).   Throughout the world HCV viruses show extreme genetic diversity, which is partly explained by the long evolutionary association between the virus and the host.  HCV genomes are split into six major genotypes (genetic variations of ~30%) consisting of over 80 subtypes (genetic variations of ~20%).  Genotypes 1-3 have a worldwide distribution.  Genotypes 1a and 1b are the most common, accounting for about 60% of global infections.  They predominate in Northern Europe and North America, and in Southern and Eastern Europe and Japan, respectively.  Genotype 3 is endemic in Southeast Asia and is variably distributed in different countries.  Genotype 4 is principally found in the Middle East, Egypt, and Central Africa, and genotype 5 is almost exclusively found in South Africa (122).   Additionally, HCV has a great genetic diversity within infected hosts, existing in blood as a heterogeneous mixture of circulating related genomes containing a master (most 	   20	  frequently represented) sequence and a large spectrum of mutants, referred to as quasispecies (131).  This virus diversity in vivo is a result of the error-prone viral RdRp and high levels of viraemia, that together enable rapid mutation of the virus (132).  The existence of quasispecies within an infected person has many important consequences, including rapid viral escape of neutralizing immunity in the host, which prevents vaccine development, and the rapid development of resistance to direct-acting antiviral therapeutics (133).   1.3.2 HCV pathogenesis   Hepatitis is a liver disease defined by inflammation of the liver and characterized by immune-mediated destruction of hepatocytes.  After HCV infection, acute hepatitis with jaundice occurs in about 20% of infections.  Up to 80% of individuals who become acutely infected cannot clear the virus and progress to chronic infection (134).  In most cases infection goes unnoticed for several decades before manifestations of HCV-related pathologies including cirrhosis, portal hypertension, hepatic decompensation, and/or hepatocellular carcinoma (HCC) (135).  During chronic infection, HCV actively and persistently replicates in the liver, leading to progressive fibrosis that can develop into cirrhosis in 20-30 years of infection.  Cirrhosis occurs in about 7% (in retrospective studies) to 18% (in clinical referred settings) of those chronically infected.  The risk of cirrhosis is increased in individuals using excess alcohol, in those who acquire the disease at an older age, in those with concomitant obesity, in men, in immunosuppressed HIV-positive patients, and in those with recurrent HCV after liver transplantation (136, 137).  Hepatic steatosis is also a common histological feature of chronic hepatitis C (CHC), detected in 40% to 86% of HCV-infected individuals (138).  Steatosis is characterized by hepatic fat accumulation, which occurs when the fatty acid input to the liver exceeds the amount that can be either secreted or oxidised as an energy source (output) (139).  Even though the mechanisms underlying hepatic steatosis in HCV-infected patients are still not understood, several studies have implicated a role for viral manipulation of lipid metabolism.  There is also a genetic association of HCV with steatosis, which occurs more frequently in individuals infected with HCV genotype 3 (140, 141).  Clinical significance of the 	   21	  pathological condition suggests that steatosis associates with fibrosis progression (142, 143), and has been negatively correlated with HCV response to antiviral therapy (144, 145).   A prominent mechanism linking steatosis and fibrogenesis during HCV infection is insulin resistance.  There is experimental evidence showing a direct link between HCV infection and insulin resistance, thought to occur via a virally-induced upregulation of tumour necrosis factor α (TNF‐α) that induces insulin resistance in HCV-infected individuals (146-148).  Several studies also observed the high prevalence of hypobetalipoproteinemia in HCV-infected patients compared with healthy control subjects (149, 150).  A low abundance of lipoproteins in the blood may result from HCV-regulated reduction in liver VLDL assembly and secretion and, as a consequence, a decrease of ApoB plasma level, which is likely a key contributing factor to intracellular lipid accumulation and steatosis (150).   1.3.3 HCV proteins  HCV is a positive-sense, single-stranded 9,600-kilobase RNA virus belonging to the family Flaviviridae.  A single HCV polyprotein of 3,010 amino acids is translated, and then cleaved by cellular and viral proteases into three structural proteins (core, E1, and E2) and seven non-structural proteins (p7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B) (151) (Figure 1.4).  The viral core and envelope proteins (E1 and E2) are located at the N-terminus of the polyprotein followed by the viroporin p7 that functions as an ion channel.  Each of these proteins is released from the polyprotein by host signal peptidase cleavage.  The core protein is initially synthesized in the ER but upon the second cleavage with the signal peptide peptidase in the C-terminal region of the core protein, the protein is released from the ER membrane and is free for trafficking to the surface of LDs (152).  The two envelope glycoproteins, E1 and E2, play a major role in virus entry.  They are type 1 transmembrane proteins forming non-covalent heterodimers at the surface of HCV particles and represent a major target for neutralizing antibodies (153).  Monomers of the p7 protein can assemble into hexamers or heptamers and thereby form cation-selective ion channels for pH regulation during virus assembly and release (154).  NS2 functions as a protease by participating in the cleavage at the NS2/NS3 junction of the polyprotein (155).  NS3 is a multifunctional protein consisting of serine protease domain and helicase/NTPase domain.  The protease activity of   	   22	    Figure 1.4. HCV genome and protein functions.  The HCV genome is composed of an open reading frame (ORF) flanked by 5' and 3' untranslated regions (UTRs).  Internal ribosome entry site (IRES)-mediated translation of the ORF yields a single polyprotein carrying 10 individual viral proteins. These are released by host signal peptidase, HCV non-structural NS2 and NS3/4A proteases. Host miR-122 upregulates viral RNA abundance by binding at the 5'-UTR of the HCV RNA.  Adapted from (156).    	   23	  NS3 is enhanced by the NS4A cofactor and is responsible for the polyprotein cleavage in the region downstream of NS3, which is essential for the generation of subunits of the viral RNA replication complex (155).  Also, the NS3/4A protease activity is involved in blocking the ability of the host cell to develop an innate antiviral response by interfering with double-stranded RNA signalling pathways (157).  NS3/4A inhibits the cellular RNA helicase retinoic acid-inducible gene I (RIG- I) pathway by cleaving an essential adaptor protein of interferon regulatory factor 3 (IRF-3) activation (158).  The helicase activity of NS3 performs double-stranded RNA unwinding that may be involved in multiple steps of HCV RNA replication (151).  NS4B is best known for its ability to induce membranous web formation within the cell, an important feature of HCV replication (159).  Studies on NS5A have shown the protein as a critical member of virus assembly and an essential protein in virus replication.  The phosphorylation state of NS5A has been implicated in regulation of multiple NS5A functions (160).  The final HCV protein, NS5B is the HCV RdRp, which is responsible for synthesizing viral RNA (161).   1.3.4 HCV lifecycle  HCV virions are 50-80 nm in diameter, with E1 and E2 glycoprotein heterodimers embedded in the lipid bilayer surrounding a nucleocapsid composed of core protein and the positive-sense single-stranded RNA genome (162).  HCV virions are associated with LDL and VLDL in the infected host and exist as lipoviroparticles (LVPs): the lipoproteins assist with virus entry and shield the virus from neutralization (163). Viral entry plays an important role in enforcing the liver tropism of HCV.  The LDLR and glycosaminoglycans are thought to mediate initial low-affinity cell binding with LVPs (164, 165), before E1-E2 interact with the co-receptors SR-BI (166) and CD81 (167) to mediate high-affinity cell binding.  Claudin-1 (CLDN1) and occludin (OCLN) are also required for entry (168, 169).  Virion-associated cholesterol seems to be involved at a late stage of HCV entry, at or before fusion, through interaction with the Niemann-Pick C1-Like 1 (NPC1L1) cholesterol absorption receptor (170).  Uptake occurs through clathrin-mediated endocytosis (171), and fusion requires the low pH of the endosomal compartment (172).  	   24	  These entry processes eventually lead to the release of the HCV genome into the cytoplasm, where viral genome translation first occurs. ER-associated translation is initiated by an IRES located in the HCV 5'-UTR (173).  The resulting HCV polyprotein is co- and post-translationally cleaved by cellular proteases (signalase and signal peptide peptidase) and the viral NS2 and NS3/NS4A proteases to release the ten HCV proteins (162).  RNA replication is believed to take place in association with ER-derived membranous web induced by NS4B and NS5A (159, 162).  Besides viral proteins involved in virus replication, a number of host factors influencing the viral replication cycle have been identified.  Cyclophilin A (CypA), a protein possessing peptidyl-prolyl isomerase activity, is required for HCV replication (174).  CypA, which binds NS5A, is believed to catalyze conformational changes necessary for HCV RNA replication (175). Another host factor implicated in the HCV lifecycle is the liver-specific host miR-122 that upregulates viral RNA abundance (176).  The miR-122 has two conserved seed sites at the 5'-UTR HCV RNA, while interaction with the 3' end of viral genome is also necessary to maintain HCV RNA abundance.  MiR-122 forms an oligomeric complex in which one miR-122 molecule binds to the 5' terminus of HCV RNA with 3' overhanging nucleotides masking the 5' terminal sequences of the HCV genome (177).  MiR-122-Ago2 complexes were shown to protect uncapped HCV RNA from 5'-3' degradation by exoribonuclease Xrn1 (178). It has also been suggested that miR-122-HCV RNA complexes might shield the 5' end of the HCV genome from recognition by innate immunity-inducing pattern recognition receptors, such as RIG-I (179).  Of note, miR-122 binding also has a minimal stimulatory effect on translation (180).  Virus assembly and release are tightly regulated processes connected to host cell lipid metabolism (163).  Following maturation, an HCV core protein moves with the assistance of diacylglycerol O-acyltransferase 1 (DGAT1) (181) from ER membranes to cytoplasmic LDs (152).  Host LDs surrounded by the core protein have been proposed as the main site of HCV assembly (182-184).  Disrupting HCV association with LDs blocks production of infectious virus particles (185).  Also, targeting of HCV core to LDs is associated with increased LD size and relocation of cytosolic LDs to the perinuclear region of the cell (186-188).  	   25	  During virus release, HCV hijacks the VLDL synthesis and secretion pathway, a process that transforms intracellular high-density precursors into low-density secreted particles, known as LVPs, prior to egress (189, 190).  During the process, the nucleocapsid is thought to be transferred to luminal LDs, which are precursors of VLDL particles (191).  Nucleocapsid-containing luminal LDs fuse with ApoB-containing pre-VLDL particles to form LVPs, which also acquire ApoE and ApoC (190) and exit through the Golgi (192).  The microsomal	  triglyceride	  transfer	  proteins has been shown to be responsible for transfer of triglycerides to luminal LDs and has been implicated in HCV assembly (190).   1.3.5 HCV antivirals  CHC treatment has been interferon-based for the last two decades, with the addition of ribavirin in 1998 (193), pegylated interferon (PEG-IFN) in 2001 (194), and initial protease inhibitor direct-acting antiviral (DAA) therapies (telaprevir, boceprevir) in 2011 (195, 196) providing a gradual increase in the rate of sustained virologic response (SVR, equivalent to a virological cure of infection) (Figure 1.5).    IFN treatment stimulates the host innate immune response to viral infection by signalling naïve cells to establish an antiviral state.  IFN acts through the JAK/STAT signalling pathway (a series of signalling kinases and transcriptional activators) to activate expression of IFN stimulated genes (197).  Ribavirin (RBV) is a guanosine analog that interferes with HCV infection.  The mechanism of RBV action against HCV is unclear but several mechanisms have been proposed including the shift in the T helper cell 1 (TH1/TH2 balance, termination in the premature RNA chain through misincorporation by the NS5B RdRP activity and by increasing HCV RNA mutations past an ‘error catastrophe’ threshold (198).   Despite some improvements in IFN-containing regimens, treatment uptake has remained low in most countries, ranging from <1% to a maximum of 5% of people with CHC initiating therapy each year (199).  Factors contributing to low HCV treatment rates included low efficacy, medical comorbidities and therapeutic toxicity, prolonged duration of therapy (24–48 weeks), lack of awareness of the curative potential of treatment, lack of    	   26	      Figure 1.5. Development in hepatitis C therapy with respect to tolerability and efficacy.  The timeline of hepatitis C therapy development with respect to tolerability and efficacy. Abbreviations: BOC, boceprevir; DSV, dasabuvir; IFN, interferon; LED, ledipasvir; OBV, ombitasvir; PEG-IFN, pegylated interferon; PTV, paritaprevir; RBV, ribavirin; SIM, simeprevir; SOF, sofosbuvir; TVR, telaprevir.  Adapted from (126).   	   27	  treatment infrastructure, limits on treatment reimbursement, social marginalization of many people with CHC, and low rates of HCV screening and disease assessment (126).  Fortunately, there have been drastic improvements in the treatment of CHC in the past few years.  Several oral regimens combining DAAs from different families (NS5B nucleotide inhibitors, NS5B non-nucleoside inhibitors, NS5A replication complex inhibitors and NS3/4A protease inhibitors) have been developed.  These regimens result in an increase in SVR rates to above 90% and a reduction of the duration of treatment to 12 weeks or less (200) (Figure 1.5).  A complete list of HCV DAAs currently on the market and in clinical trials has been summarized recently in Asselah et al. (200).   Despite an enhanced HCV efficacy rate, several challenges will likely appear in the future (201).  Most of these drugs have a low barrier to resistance, with the exception of NS5B nucleos(t)ide inhibitors.  Resistance-associated mutations to several DAAs have already been characterized in NS3/4A, NS5A, and NS5B as well as natural polymorphisms observed in certain genotypes and subtypes (202).  Moreover, another serious challenge is that curing HCV does not prevent patients from being reinfected nor does it address the potential for liver disease progression after achieving a virological cure.  Together, our understanding of the HCV-induced mechanisms leading to liver disease and cancer remains rather limited.  Further research on the mechanisms of progressive liver disease is urgently required to enable the development of additional preventative therapeutics (201).  1.3.6 HCV and lipid metabolism  Hepatic lipid metabolism is intimately linked with every step of the HCV lifecycle, and is extensively modulated by viral infection.  HCV hijacks a host lipid kinase, phosphatidylinositol 4-kinase (PI4KA), to alter intracellular membrane architecture and to drive membranous web formation to facilitate its genomic replication (203).  As mentioned above, host LDs are important for HCV as they serve as a scaffold for virus assembly	  (182-184).  HCV negatively modulates the synthesis and secretion of VLDLs (149, 150, 204).  Hijacking of the VLDL pathway is not only important for assembly but has downstream consequences.  Virus particles are secreted into plasma as a heterogeneous population with 	   28	  varying densities and are associated with various lipoprotein components that may shield HCV from the immune system and permit enhanced entry into hepatocytes (163).   The dependence of HCV for replication, morphogenesis and secretion on host lipid metabolic pathways requires that HCV modulates the lipid-rich intracellular environment to make it favourable for virus propagation.  HCV alters host lipid metabolism in three ways: enhanced lipogenesis, impaired degradation and reduced export (205).  HCV increases lipogenesis via SREBP activation (206) and reduces oxidation and lipid export (207).  HCV-associated downregulation of fatty acid breakdown and cholesterol export is thought to contribute to steatosis.  HCV infection induces generation of reactive oxygen species (ROS), which can modulate the functions of proteins and lipids via peroxidation and also suppress fatty acid oxidation and export (208).  Hypobetalipoproteinemia, another condition observed in HCV infected individuals and is a well-established cause of steatosis, likely results from viral hijacking of the VLDL pathway (150).  Both plasma cholesterol and ApoB levels of CHC patients increase upon successful antiviral treatment, whereas no such increase was observed in patients who fail to respond to treatment, further suggesting that HCV infection is responsible for these alterations in lipid metabolism (207).  1.3.7 MicroRNAs during HCV infection Many human miRNAs have been described to influence HCV pathogenesis either directly or indirectly through regulation of pathways associated with the HCV lifecycle and disease progression.  As discussed earlier, during HCV infection a liver-specific host miRNA, miR-122, interacts with the virus genome and enhances RNA abundance, translation, and virus production (209).  Other host miRNAs such as miR-448, miR-196, miR-199a and let-7b have been reported to interact directly with HCV RNA resulting in an inhibitory effect on the HCV lifecycle (210-212).  MicroRNAs affect not only HCV lifecycle but also the tissue pathology seen in CHC patients.  Hepatic fibrosis developed during CHC infection is morphologically characterized by increased deposition of extracellular matrix (ECM) proteins, including collagen types I/III, fibronectin and laminin.  Increased transforming growth factor-β (TGF-β) signalling in response to HCV-mediated damage activates hepatic stellate cells (HSCs), and strongly 	   29	  upregulates ECM protein production (213).  Previous studies suggested HCV-induced modulation in miRNA expression levels may contribute to progression of liver fibrosis (214).  HCV infection has been shown to downregulate expression of miR-29, which has an inhibitory effect on the HCV lifecycle (215).  Downregulation of miR-29 levels concomitantly increases the expression of miR-29 targeted collagen genes (collagen type I alpha 1 [COL1A1] or collagen type III alpha 1 [COL3A1]), which contribute to ECM production (214).   MicroRNAs have also been implicated in HCV-associated steatosis, particularly miR-27. Two isoforms of miR-27 exist, miR-27a and miR-27b.  They are encoded by separate gene loci and differ by one nucleotide.  Two studies have correlated miR-27 expression with the occurrence of HCV steatosis, both in cell culture and in vivo (HCV-infected SCID/Alb-uPa mice and patients) (100, 216). Shirasaki et al. demonstrated that miR-27a regulates HCV induction of lipid storage by targeting lipid transporter ABCA1 and RXRα (100).  Singaravelu et al. reported that miR-27b induces LD accumulation in HCV infected hepatocytes by targeting PPARα and angiopoietin-like protein 3 (ANGPTL3) (216).   MicroRNAs have been also implicated in host defense to HCV infection and the clinical outcomes of anti-HCV therapy.  For example, miRNA microarray profiling in liver biopsies of CHC patients prior to IFN-based anti-HCV treatment showed that specific miRNAs correlated with treatment outcome.  Expression levels of 9 miRNAs (upregulated: miR-27b, miR-122, miR-378 and miR-422b; downregulated: miR-18a, miR-34b, miR-143, miR-145 and miR-652) were significantly different between the SVR and non-responder (NR) groups, suggesting that expression patterns of these hepatic miRNAs are associated with the therapeutic outcome in CHC patients (217).    1.4 Dengue virus (DENV) biology  Dengue virus (DENV) is another member of the Flaviviridae family and, like HCV, it has been implicated in hijacking host lipid metabolism.  DENV infects humans in over 100 countries every year, with approximately 3.6 billion people at risk of infection globally (218).  DENV epidemics occur in the Americas, Asia, Africa, and Australia and affect travelers to endemic regions.  DENV epidemics have a substantial economic impact in affected countries 	   30	  with the annual economic costs of DENV in the Americas estimated at $2.1 (USD) billion including direct medical, direct non-medical, and indirect costs (eg. productivity losses) of dengue cases treated in hospital or ambulatory settings (219, 220).  Dengvaxia®, the world’s first DENV vaccine, was approved in December 2015 in Mexico. To date the vaccine has also been registered in the Philippines, Brazil and El Salvador (221).  Efficacy analysis of the two large phase III efficacy studies concluded that Dengvaxia® protects against 65.6% of symptomatic dengue disease caused by any of the four serotypes of the virus in the study population from 9 - 16 years old over the 25-month surveillance period (222). Also, there are several other vaccine candidates under development (223).  There is still no evidence of how effective these vaccines will be in preventing dengue disease in highly urban settings and in the long-term and given the absence of antiviral treatment against DENV infection, further understanding of DENV pathogenesis is necessary to help identify new therapeutic targets.   1.4.1 DENV epidemiology and clinical disease  Dengue is the most prevalent arthropod-borne viral infection in humans. DENV is spread to humans by its two principal mosquito vectors (Aedes aegypti or Aedes albopictus). Four DENV serotypes (1, 2, 3, and 4) circulate in tropical and sub-tropical areas of the world, each capable of causing severe disease.  DENV serotypes differ from each other by 25-40% at the amino acid levels.  The incidence of reported DENV infections has risen steadily over the past few decades due to increased urban populations, global travel and commerce, and inadequate mosquito control programs (224).   DENV infection is largely asymptomatic (up to 75% of infections) but may result in a range of clinical illnesses from self-limited dengue fever (DF) to severe dengue, a potentially lethal hemorrhagic and capillary leak syndrome previously called dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS).  DF is characterized by an abrupt onset febrile illness associated with headache, severe muscle and joint pain, and rash that typically lasts for one to two weeks.  Severe dengue presents the rapid onset of capillary leakage, significant thrombocytopenia and mild-to-moderate liver injury.  Hemorrhagic manifestations include bleeding in the skin and gastrointestinal tract.  Rapid fluid loss leads to hemoconcentration and hypotension that can result in mortality (224, 225).  	   31	  1.4.2 DENV pathogenesis  Although primary infection provides durable if not life-long protection against re-infection by the same DENV serotype, secondary infection by a different DENV serotype occurs frequently in endemic areas and is the primary risk factor for severe disease.  Severe dengue occurs in most cases in adults during secondary infection and in infants born to DENV-immune mothers and it is thought to be a unique feature of DENV pathogenesis, related to immune enhancement (225).  The exact mechanism by which the immune response to DENV protects against or contributes to severe disease remains unclear.  When sufficiently large numbers of antibodies bind to DENV particles, they are able to neutralize infection.  However, if antibody affinity falls below the threshold for neutralization, they can induce entry of DENV into cells via Fcγ receptors (FcγR) by a process known as “antibody-dependent enhancement” (ADE). Thus, neutralizing antibodies developed in response to primary DENV infection with one DENV serotype are protective against a secondary infection with exactly the same DENV serotype. In the case of a secondary infection with a different DENV serotype, the subtle changes in the epitope mean the antibodies can bind the virus but with low affinity and are unable to effectively neutralize the virus, and hence the antibodies stimulate ADE by inducing entry of DENV into cells via FcγR, resulting in increased viral burden and more severe disease (225). In the case of severe dengue in infants, it is believed that DENV-specific IgG antibodies transferred from mother to fetus decrease to levels that enhance cellular uptake of a newly acquired primary DENV.  Therefore, the severity of DENV infection is thought to depend on the quantity and specificity of the cross-reactive antibody response that either is pre-established in serum or quickly produced by the memory B cell response (224).  Besides ADE, during a primary infection, both serotype-specific and cross-reactive memory T cell responses are produced.  The expression of viral epitopes on infected cells during a secondary DENV infection triggers activation of serotype-cross-reactive memory T cells, with the production of pro-inflammatory cytokines ultimately resulting in plasma leakage in the vascular endothelium (220).  Also, myeloid cells or mast cells have been proposed to increase vascular permeability by producing IL-10, skewing CD4+ T cells responses or releasing vasoactive molecules through degranulation (226, 227).   	   32	  1.4.3 DENV biology and lifecycle  DENV is a spherical 50 nm virion consisting of a nucleocapsid made of capsid protein [C] surrounded by a lipid envelope.  Two other structural DENV proteins, pre-membrane [prM], and envelope [E], are embedded in the envelope and control virus entry into cells.  The DENV genome is a 10.7-kilobase positive single-stranded RNA encoding for a single polyprotein that is processed into individual 10 proteins (224) (Figure 1.6).   DENV enters target host cells via clathrin-dependent receptor-mediated endocytosis.  Numerous putative receptors have been identified on human and mosquito cells, but a definitive receptor for virus entry has not been identified and therefore the molecular basis of DENV tropism remains uncertain.  Targets for DENV infection in vivo include immune cells (monocytes, macrophages, dendritic cells, mast cells), hepatocytes and endothelial cells.  Dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) serves as a DENV attachment factor on dendritic cells (220).  During secondary infections, pre-existing antibodies bind to DENV virions and enable Fcγ receptor-mediated uptake by Fcγ receptor-bearing cells, as discussed earlier.   After endocytosis, a pH-dependent conformational change allows escape of viral RNA from the endosome.  Virus penetration into the cytoplasm is followed by translation in the ER into a single polyprotein that is subsequently cleaved into three structural (C, M and E) and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) by viral NS2B/NS3 serine protease and host proteases, including prM cleavage by furin.  Virus RNA is replicated in virus-induced invaginated membrane “vesicle packets” (VP) derived from the ER (228).  VPs are approximately 90 nm-wide and containing non-structural proteins along with double-stranded RNA. A pore appears to connect the interior of the VP with the cytosol, and presumably permits trafficking of nucleoside triphosphate (NTPs) and newly synthesized viral genomes into and out of VP, respectively (38, 228).   After association of the viral RNA with the capsid protein and budding into the ER to acquire a lipid membrane coated with membrane prM and E proteins, the virion exits through   	   33	    Figure 1.6.  DENV particle and genome.  A. Cross-section of mature dengue virion showing structural components. B. The DENV genome is composed of an ORF flanked by 5' and 3' UTRs encoding three structural proteins (the capsid (C), membrane (prM) and envelope (E) glycoproteins) and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5).    	   34	  the host secretory pathway when the virus undergoes conformational changes to achieve mature state.  Cleavage of prM into M proteins by furin proteases in the trans-Golgi network during viral egress is required for generation of mature DENV virions, which have a smooth marble-like structure, whereas immature virions have heterotrimeric spikes that project off its surface (229).  Because of differential exposure and conformation of E and prM/M proteins on the surface of mature versus immature virions, the maturation state of DENV modulates both the cell types that can be infected due to specificity of expressed receptors and the interaction of the virion with particular antibodies (229, 230). 	  1.4.4 DENV vaccine and treatment   To produce protection against four different DENV serotypes has been a significant challenge for vaccine development, as partial vaccine-induced immunity against any single serotype theoretically could result in an individual being at great risk of developing severe disease during subsequent natural infection.  Several live-attenuated tetravalent DENV vaccine candidates are under evaluation in large clinical trials.  The first one that recently has been approved in several countries is the Sanofi Pasteur CYD-TDV candidate, Dengvaxia®, which contains four chimeric viruses in which the structural genes of the yellow fever vaccine virus (YFV-17D) were replaced with those of each DENV serotype.  This vaccine showed variable protection among different DENV serotypes (35 - 78%) (231).  Other potential concerns related to the vaccine include its lack of T cell epitopes derived from the non-structural proteins because the vaccine is composed largely of sequence from YFV-17D, which could limit induction of a protective T cell response.  Also, the vaccine lacks DENV NS1, an immunogenic viral protein secreted at high levels in infected individuals, which has been hypothesized to contribute directly to disease pathogenesis (224, 232).  Two additional vaccine candidates have recently advanced to phase II trials.  DENVax developed by Takeda Pharmaceuticals and TV005 developed by the National Institute of Allergy and Infectious Diseases (NIAID) that are composed of a mixture of either a full length attenuated or modified DENV and chimeric DENV strains.  While DENVax produces a tetravalent neutralizing antibody response in 44 - 80% of recipients (233), TV005 was shown to elicit a neutralizing antibody response against all four DENV serotypes in 90% of recipients (234). 	   35	   Besides vaccine development, other strategies have also been pursued to control DENV infections.  Extensive progress has been made on reducing DENV transmission by limiting infection in the mosquito host.  For example, the infection of Aedes aegypti mosquitoes with the endosymbiotic bacteria Wolbachia resulted in invasion of mosquito populations and interference with DENV replication (235).  Wolbachia-infected Aedes aegypti mosquitoes have already been released in Australia where outbreaks of DF occur and have been stable over several years (236).  Other groups have created genetically engineered Aedes aegypti mosquitoes that inherently are resistant to DENV infection through the induction of antiviral RNA interference (237).   Also, high-throughput antiviral drug discovery screens have been performed to identify inhibitors of the fusogenic viral envelope protein (E), the NS2B/NS3 heterocomplex serine protease, integral membrane proteins required for replication (NS2A and NS4B), and the RdRp (NS5), with further pre-clinical development ongoing (238).  Nonetheless, because drugs against viral proteins could select for resistant variants, the concept of targeting host molecules required for DENV infectivity has emerged as an alternative strategy (239).  Drugs that target host processes contributing to DENV pathogenesis, and that aim to enhance innate immune responses are also being studied (239).   1.4.5 DENV and lipid metabolism  Lipid metabolism is intricately involved in multiple steps of the DENV lifecycle.  DENV entry includes fusion of the viral lipid membrane with the host endosomal membrane, and the efficiency of the fusion process is very likely to be affected by the lipid content of the entering virions as well as by the host target membrane (38, 240).  Previously, it has been shown that negatively charged lipids, such as phosphatidylserine or bis(monoacylglycero)phosphate (BMP), increase the fusion of DENV with synthetic liposomes (240, 241), and these lipids are abundant within the internal membranes of the late endosome and lysosome.  DENV RNA replication, which occurs in VPs (as described earlier), requires regions of both negative and positive membrane curvature, which may be induced by proteins embedded in the membrane and/or by the physicochemical properties of the lipids forming 	   36	  the bilayer.  Notably, biochemically isolated replication membranes for DENV are enriched in sphingolipids and sterols relative to the ER membrane from which they are derived (38).  Indeed, cholesterol may be important for assembly of the DENV RNA replication complex, perhaps explaining the virally-induced formation of cholesterol-rich microdomains within the ER.  Consistent with this hypothesis, pharmacological inhibition of squalene and HMG-CoA synthase, two enzymes necessary for cholesterol biosynthesis, inhibit DENV replication in cell culture (242).  Analysis of host factors required for DENV replication also showed that fatty acid synthetase is recruited to viral replication sites through the viral NS3 protein.  This leads to an increase in the rate of fatty acid biosynthesis in DENV-infected cells with de novo synthesized lipids preferentially colocalizing with the viral genome (243).   While LDs have been known to play a critical function in HCV particle assembly for some time (152), evidence for their role in DENV replication has only recently been uncovered (244-246).  DENV particles were shown to associate with LDs through the capsid protein, and this association was found to be necessary for infectious viral particle formation (244).  Moreover, it has been demonstrated that DENV induces the lysosomal degradation of LDs by autophagy to support its replication (247).  The increased release of triglycerides, the major component of LDs, leads to production of free fatty acids that are catabolized by mitochondrial β-oxidation generating ATP.  This latter event is essential for DENV replication (247).  1.5 Research hypothesis and rationale  Cholesterol and lipid levels are maintained through tightly controlled and complex feedback mechanisms that involve regulation of major metabolic genes.  It is well known that insufficient or excessive cellular or plasma lipid levels can lead to a wide range of pathologies, including hyperlipidemia, atherosclerosis and premature coronary heart disease.  As discussed, a number of enveloped viruses, including important human viruses of the Flaviviridae family, HCV and DENV, modulate and utilize host lipids to support their lifecycles, and the changes in lipid homeostasis that they impose may further contribute to virus-associated pathology.  However, the regulatory mechanisms underlying the dysregulation of lipid pathways in these disease states are incompletely understood.  A key 	   37	  aim of this thesis was to determine the role of key regulators of host lipid homeostasis, including miRNAs and two members of the PC family, during viral infection and virus-associated diseases. Our study provides new insights into complex pathology-associated host-virus interactions, which may be utilized as targets for antiviral development as well as biomarkers of infection and virus-associated diseases.   MicroRNAs repress expression levels of genes by binding to mRNA transcripts, resulting in translational repression and acting as master regulators of many cellular processes, including lipid metabolism (248). The presence of miRNAs in circulation also provides great potential for their use as clinical biomarkers in diagnosis or prognosis of various pathologies (41, 249).  Likewise, PCSK9 and SKI-1/S1P, members of the PC family, play essential roles in regulating lipid and cholesterol metabolism through several interdependent pathways (8, 33).  In this study, we hypothesized that miRNAs (miR-122 (90, 91), miR-223 (107) and miR-24 (101)) and proprotein convertases (PCSK9 (8) and SKI-1/S1P (33) modulate the lifecycles and pathogenesis of viruses of the Flaviviridae family through regulating host lipid metabolism.   1.5.1 Aim 1  HCV hijacks host lipid metabolic pathways as part of its lifecycle.  Chronic HCV infection is associated with altered lipid metabolism, which both contributes to disease progression and influences response to antiviral therapy.  To help understand how HCV influences important lipid metabolic pathways during chronic liver disease, we investigated the molecular interplay between three circulating miRNAs known to act as regulators of lipid homeostasis, miR-122 (90, 91), miR-223 (107) and miR-24 (101), in CHC patients. Our hypothesis was that defining a specific signature of selected circulating miRNAs that is characteristic of HCV-infected patients who achieve a treatment-based viral cure would identify important lipid metabolic pathways influenced by HCV infection during chronic liver disease.  This study is presented in Chapter 2.    Our data demonstrated differential changes in circulating miRNAs (miR-122, miR-24, miR-223) before, during and after treatment-based cures.  Our results showed a decrease in circulating miR-122 and increases in miR-24 and miR-223 after treatment compared to 	   38	  baseline in SVR.  These results reveal that miRNAs known to act as regulators of lipid metabolism are correlated with IFN-based therapeutic outcomes in patients with HCV infection.   1.5.2 Aim 2  HCV particles are termed lipoviroparticles as they associate with lipoprotein components and interact with the LDLR to promote viral uptake into hepatocytes.  PCSK9 is a host protein that controls cell surface expression of LDLR, a regulator of plasma lipoprotein uptake (8).  Increased PCSK9 expression and gain-of-function mutations in PCSK9 result in decreased LDLR levels and a reduction in clearance of plasma LDL. Thus, upregulation of extracellular PCSK9 concentrations may inhibit HCV infection and alter the host lipid metabolic pathways hijacked during viral infection (12-14).  Because of the unique function of PCSK9 and the impact of HCV infection on lipid homeostasis, in Chapter 3 we hypothesized that plasma PCSK9 concentrations would differ depending on treatment outcomes, in CHC patients undergoing antiviral treatment.   Our data from Chapter 3 demonstrated that plasma PCSK9 concentration was significantly upregulated in SVR patients but not in relapsers after treatment.  We further confirmed that increased extracellular PCSK9 limits HCV infection in human hepatoma cells.  We demonstrated that PCSK9 gain- and loss-of-function mutants allow various degrees of HCV infection inhibition.  These results reveal that increases in circulating PCSK9 levels in SVR patients could have an important biological effect on HCV by suppressing virus entry.  1.5.3 Aim 3  Lipid metabolic pathways are hijacked and utilized by a number of enveloped viruses during infection.  LDs and their components are of great importance in the lifecycles of both HCV and DENV.  SKI-1/S1P is another host protein that acts as a checkpoint in the host’s lipid metabolic system, through control of the SREBP activation (33).  Previously, the Jean laboratory reported that targeting SKI-1/S1P enzymatic activity using an active-site-directed small molecule inhibitor effectively inhibits HCV infection (37).  In Chapter 4, we 	   39	  hypothesized that inhibiting SKI-1/S1P-mediated activation of the SREBP using a small molecule inhibitor would reduce abundance of LDs and ultimately inhibit DENV infection.  Effective inhibition of DENV infection through manipulation of the host SKI-1/S1P enzymatic activity could serve as a potential target for indirect-acting anti-DENV agents.   Our data from Chapter 4 demonstrated that an active-site-directed aminopyrrolidineamide-based inhibitor of SKI-1/S1P, PF-429242 (33), effectively blocks DENV (1-4) from establishing infection in human hepatoma Huh-7.5.1 cells.  PF-429242 antiviral activity was observed both pre- and post-establishment of viral infection and was associated with a dramatic decrease in LD abundance in PF-429242-treated Huh-7.5.1 cells.  Our studies demonstrate SKI-1/S1P’s potential as a novel host-directed pan-serotypic anti-DENV target, and they reveal therapeutic opportunities associated with the use of lipid-modulating drugs for controlling DENV infection.   	   40	  Chapter 2: Differential profiles of circulating miR-122, miR-24 and miR-223 in hepatitis C-infected patients who achieve a treatment-based viral cure  2.1 Introduction  Cellular and plasma cholesterol and lipid levels are maintained through tightly controlled and complex feedback mechanisms that involve regulation of major metabolic pathways (250).  Disruption of cholesterol and lipid homeostasis can lead to a wide range of pathologies, including hyperlipidemia, atherosclerosis, and other metabolic disorders (251, 252).  There is increasing evidence of viral hijacking of host cholesterol and lipid metabolic pathways by human enveloped viruses such as hepatitis C virus (HCV), and these data are helping to uncover important and intimate connections between virus, host lipids, and pathogenesis of viral disease.  HCV establishes chronic infection in approximately 75% of infected individuals and is a major cause of cirrhosis, end-stage liver disease, and hepatocellular carcinoma (253).  Chronic HCV (CHC) infection is also associated with dysregulation of host lipid metabolism (254, 255).   It has been suggested that the dysregulation of host lipids that occurs during CHC infection may contribute to steatosis, which can further increase the risk of liver fibrosis (reviewed in (254, 255)).  Interestingly, several epidemiological studies have reported that CHC infection is associated with hypolipidemia (149, 256) that usually resolves with successful HCV treatment, returning lipid levels to baseline in those who achieve a sustained virological clearance; however, it persists in people who fail treatment (256).  Furthermore, many successfully treated patients experience low-density lipoprotein (LDL) and cholesterol rebound to such high levels that they are associated with an increased risk of developing coronary disease (256).  Taken together, these clinical and experimental studies underline the complexity of the HCV-induced metabolic changes during CHC infection and potential impact of treatment on outcome.  However, the identity of the key molecular players in lipid homeostasis remains elusive, and the impact of current treatment strategies on host lipid metabolism is unknown.  In this chapter, we tested the hypothesis that specific lipid 	   41	  metabolic indicators would be differentially associated with treatment-based viral cure versus ongoing chronic infection.  MicroRNAs (miRNAs) are small non-coding RNAs that regulate gene expression at the post-transcriptional level (248).  They influence key biological processes and can contribute to the pathogenesis of various diseases (248).  MicroRNAs also serve as stable plasma biomarkers for many disorders (41, 249) and are promising targets for novel therapeutic strategies (257, 258). MicroRNAs have recently been found to modulate cholesterol and lipid homeostasis.  In vivo inhibition of a liver-specific miRNA, miR-122, significantly lowered plasma cholesterol levels in both mice and non-human primates (90, 91).  Besides having an important role in cholesterol homeostasis, miR-122 is required for efficient HCV replication (176), and a miR-122 antagonist was shown to have strong antiviral effects in a human clinical trial (257).  In addition, miR-223 coordinates cholesterol homeostasis by targeting several key genes involved in cholesterol biosynthesis, influx and efflux (107).  Also, several miRNAs from the miR-23 cluster were reported to be involved in regulation of lipid metabolism, including miR-27b (216, 259) and miR-24 (101).  Moreover, miR-223 and miR-24 are lipoprotein- associated miRNAs (77, 83, 260) and are abundant in liver (101, 107) and platelets (261).   To further understand the impact of HCV infection on lipid metabolism, we investigated the molecular interplay between three circulating miRNAs (miR-122, miR-223, and miR-24) in a well-characterized cohort of 94 patients with CHC who had been treated with pegylated- interferon-based therapy. Our aim was to elucidate HCV-induced lipid metabolic changes during chronic HCV infection and to define their link with treatment outcome.  The CHC-infected cohort consisted of three groups of participants: patients who achieved virological cure, known as a sustained virological response (SVR); relapsers; and non-responders.  We examined potential correlations between liver-specific miR-122 and lipoprotein-associated miR-24 and miR-223 with the cohort’s clinical parameters.  We also evaluated if treatment-based viral cure of CHC infection is associated with a coordinated interplay between these three molecules to help characterize candidate biomarkers for treatment-based viral cure of CHC infection and to identify potential therapeutic targets to prevent progression of HCV-associated liver disease.  	   42	  2.2 Materials and methods Patient samples.   Stored plasma samples (-80°C) from 94 patients with CHC were retrospectively analyzed.  Patient samples were collected at an outpatient clinic (certificate # H13-01770) to monitor treatment outcomes.  Patients were treated with interferon-based antiviral therapy that included pegylated interferon-alpha (PEG-IFN) and Ribavirin (RBV) and/or boceprevir (BOC)/telaprevir (TPV) for 24-48 weeks.  Patients were classified having achieved SVR when HCV RNA was undetectable at 12 and 24 weeks after the end of therapy.  Relapsers were patients who were negative for HCV RNA at the end of therapy but were HCV RNA positive at either 12 or 24 weeks after the end of therapy.  Non-responders were patients with detectable HCV RNA at all timepoints during and at the end of therapy.  Plasma samples were collected and evaluated at baseline before therapy (week 0) as well as during therapy (weeks 4, 8, 12, 24, and 48) and after therapy (12 or 24 weeks post-treatment).  A cohort overview is summarized in Figure 2.1A.  HCV RNA was quantified using the COBAS AmpliPrep/COBAS Taqman® HCV-Test form Roche Molecular Systems. FibroScan® testing was performed at the clinic before treatment initiation to determine the degree of liver fibrosis.  Routine blood parameters were measured at sample collection timepoints.  An APRI score was calculated using the formula: APRI score = AST level (U/L)/AST upper limit of normal (U/L)/platelet count (109/L)) x 100, where 40 U/L was used as AST upper limit of normal (262, 263).  A FIB-4 score was calculated using the formula: FIB-4 score = (Age(years) x AST level (U/L))/(platelet count (109/L x√ALT(U/L) (264).  Patients co-infected with hepatitis B virus or HIV-1 were excluded.  Data from patients receiving long-lasting, lipid-modulating drugs or metformin therapy for diabetes were included.  The protocol was approved by the Ethics Committee of the University of British Columbia.   RNA isolation.   The experimental design is summarized in Figure 2.1B. Total RNA was extracted from plasma with the miRNeasy Serum/Plasma kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol.  Briefly, 250 µl plasma aliquots were thawed on ice for 1.5-2 hours, centrifuged for 10 min at 16,000 g at 4°C.  Two hundred microliters of plasma was transferred to a new tube, and 1 mL of QIAZOL was added and mixed by vortexing.  A total 	   43	   Figure 2.1. Cohort overview and study design for miRNA analysis.  (A) Plasma samples were collected from a cohort of 94 patients chronically infected with HCV.  All patients were treated with interferon-based antiviral therapy that included pegylated interferon-alpha (PEG-IFN) and ribavirin (RBV) and/or an oral HCV protease inhibitor [boceprevir (BOC)/telaprevir (TPV)] for 24-48 weeks [Treatment Week (TW) 24-TW48]. Plasma samples were collected before [Week 0 (W0)], during (TW4-TW24/TW48), and after treatment [Follow Up Week (FUW)] at the indicated timepoints (FUW12/FUW24).  (B) Circulating levels of plasma miR-122, miR-24, and miR-223 were quantified by quantitative real-time PCR after total RNA isolation.  MicroRNA levels were normalized to a non-human spike-in control [Caenorhabditis elegans miR-39  (cel-miR-339)].  Clinical relevance of circulating levels of plasma miR-122, miR-24, and miR-223 in CHC infection and their impacts on treatment outcome was evaluated using various statistical models (see Materials and Methods).   	   44	  of 5.6 x 108 copies of synthetic C. elegans miR-39 (Qiagen) were spiked in to each homogenate and later used as a reference control.  Two hundred microliters of chloroform was added and incubated at room temperature for 3 min before centrifugation for 15 min at 12,000 g at 4°C.  The aqueous layer (600 µl) was loaded into a QIAcube (Qiagen) and all further steps of RNA extraction were automated to reduce sample-to-sample variability.  Automated RNA extraction was performed using the QIAcube and according to the manufacturer’s protocol, and RNA was eluted in 14 µl of RNAse-free water.   Reverse transcription, real-time polymerase chain reaction (RT-PCR) PCR and normalization.  Complimentary DNA (cDNA) was synthesized using the miScript II RT kit (Qiagen) according to the manufacturer’s instructions with the following modifications for a 10 µL reaction; 5 µL of total plasma RNA was incubated with 2 µL 5x HiSpec Buffer, 1 µL Nucleics Mix, 1 µL RNase‐Free water, and 1 µL RT mix at 37°C for 1 hour followed by enzyme deactivation for 5 min at 95°C.  The cDNA was diluted 1:10 in RNase‐Free water and microRNAs detected using the miScript SYBR® Green PCR kit (Qiagen).  Twenty microliter reactions were run in triplicate in 96‐well plates on an Mx3005P QPCR system (Stratagene, La Jolla, CA, USA).  The reactions were incubated at 95°C for 15 min, followed by 40 cycles of 94°C for 15 sec, 55°C for 30 sec, and 70°C for 30 sec.  Dissociation curve analysis was performed, and the threshold cycle (Ct) values were determined using default settings.  MicroRNAs were detected with the commercial Ce_miR‐39_1, miR-24-3p, miR-223-3p, and miR-122a-5p miScript Primer assays (Qiagen).  MicroRNA levels were normalized using the C. elegans miR‐39 spike-in RNA and expressed as ΔCt.   Statistical analysis.  Statistical analysis was performed using SAS Version 9.2 and GraphPad Prism version 6.0 (GraphPad Software Inc., La Jolla, CA).  Data are presented as mean ± standard error of the mean (SEM).  Data were analyzed by the parametric two-tailed t-test (Table 2.1), Spearman’s correlation (Table 2.2 and Figure 2.2 (D-E), and the two-way ANOVA test (Figure 2.3).    	   45	  Statistical modelling. Several statistical models were fitted by Paul Steven (Qiagen) to investigate the possible clinical significance of the data. Model 1: Response 1 = Response 2 + patient intercept (random) + patient slope (random) 𝑦!" = 𝛼! + 𝛼!𝑥!" + 𝛽!! + 𝛽!!𝑥!" +   𝜖!" Where 𝑦!"   denotes the 𝑖-th observation of patient 𝑗 of the response (Response 1), 𝑥!" is the covariate (Response 2) and 𝜖!" is the error term.  Model 1 looks at the relationship between Response 1 and Response 2, allowing a different intercept and slope for each patient.  Response 1 is defined as miRNA ΔCt response (i.e., separate models for each miRNA investigated), and Response 2 is defined as APRI, FIB-4 score, HCV viral load, and other liver function scores (i.e., separate models to look at each relationship). Model 2: Response = covariate  𝑦! = 𝛼! + 𝛼!𝑥! +   𝜖! where 𝑦!   denotes the 𝑖-th observation, 𝑥! is an indicator variable for treatment status,  𝛼! denotes the differences in response between before and after treatment, and  𝜖! is the error term.   Model 2 is an ANOVA that looks at the difference in mean response between patients before and after treatment.  Responses assessed were ΔCt for each miRNA and APRI and FIB-4 scores for each patient outcome group (SVR and relapsers).  2.3 Results 2.3.1 Study design and cohort characteristics   This longitudinal study included 94 CHC participants, including 57 who achieved SVR, 10 relapsers, and 27 non-responders to anti-HCV therapy.  Participants were followed for 60 to 72 weeks before, during, and after treatment depending on the length of treatment (Figure 2.1 and Table A1.1).  Participants without plasma samples (n=22) at baseline (week 0) were excluded from analysis of baseline clinical parameters associated with SVR and non-	   46	  SVR patients (Table 2.1) and from paired analysis of circulating miRNAs before and after treatment in SVR and relapsers (Figure 2.3).  This yielded 39 SVR, 9 relapsers, and 24 non-responders for the two analyses.  The demographics of SVR, relapsers, and non-responders as well as baseline clinical parameters and pertinent medications are shown in Table 2.1. In agreement with previous reports, our cohort showed that older patients, patients with lower platelet counts, and patients with a higher degree of liver injury and fibrosis based on APRI, FIB-4, and Fibroscan scores were less likely to achieve SVR (265).  Importantly, baseline levels of plasma miR-122, miR-24, and miR-223 did not differ significantly between SVR and non-SVR patients (Table 2.1).  In addition, there was no significant difference in circulating miR-122, miR-24, and miR-223 levels in SVR, relapsers, or non-responders at any time during treatment compared to baseline levels (Figure A1.1).   2.3.2 Correlation between circulating levels of plasma miR-122, miR-24, miR-223, and clinical parameters in CHC patients    To determine whether there was a correlation between circulating plasma levels of miR-122, miR-24, and miR-223 and the clinical parameters measured in all CHC patients using all timepoints, we first calculated the Spearman's correlation coefficients (rhos), which assess the strength of correlation between two distinct variables (Table 2.2). For the liver-specific miR-122, we observed rhos values indicative of a moderate, positive monotonic correlation between circulating plasma levels of miR-122 and measures of liver damage, including liver aminotransferase concentration [ΔCt(miR-122): AST (rhos = -0.54); ALT rhos = -0.62)] and APRI score (rhos = -0.42).  In contrast, we observed a weak, negative monotonic correlation between the levels of circulating miR-24 and miR-223 and APRI score (ΔCt(miR-24): rhos = 0.30; ΔCt(miR-223): rhos =0.39), and levels of circulating miR-24 and miR-223 and FIB-4 score (ΔCt(miR-24): rhos = 0.35; ΔCt(miR-223): rhos =0.40).  Moreover, there was a weak to moderate, positive monotonic correlation between miR-24 and platelets (ΔCt(miR-24): rhos  = -0.39), and miR-223 and platelets (ΔCt(miR-223): rhos = -0.45), respectively.      	   47	  Table 2.1.  Patient demographics and baseline biochemical characteristics.   ALT, alanine transaminase; AST, aspartate transaminase; WBC, white blood cells; HCV,  hepatitis C virus. APRI score was calculated using the formula: AST level (U/L)/AST upper limit of normal (U/L)/platelet count (109/L) x 100, where 40 U/L was used as AST upper limit of normal. FIB-4 score was calculated using the formula: Age(years) x AST level (U/L))/(platelet count (109/L x√ALT(U/L). MicroRNAs levels are reported as ΔCts after normalizing to spiked-in RNA. The level of ΔCt is inversely related to the miRNA level in circulation.  Parametric data are shown with plus-minus values, which are the means ± SEM. P-values are for comparison between SVR and non-SVR (relapsers + non-responders). P-values were calculated using the parametric two-tailed t-test, p<0.05 were considered significant.    	   48	  Table 2.2.  Correlations of circulating miR-122, miR-24, and miR-223 with biochemical parameters.    Data are presented as Spearman’s coefficient (rhos) and P-values. ALT, alanine transaminase; AST, aspartate transaminase; HGB, haemoglobin; WBC, white blood cells. In the analysis, ΔCt (where Ct is the PCR cycle threshold; used as a measure of miRNA levels in circulation) is normalized to the reference spike-in RNA. The level of ΔCt is inversely related to the miRNA level in circulation.   	   49	   Thus, this initial Spearman’s correlation analysis showed a moderate, positive correlation between miR-122, which is predominantly expressed in the liver, and clinical parameters associated with liver functions (AST and ALT) and liver injury (APRI).  In contrast, weak, negative correlations were observed between the circulating levels of plasma miR-24 and miR-223 and liver injury scores (APRI and FIB-4).  Interestingly, the circulating levels of miR-24 and miR-223, two platelet-enriched miRNAs (261), showed a robust positive correlation with platelets.    To investigate this further, we next performed a mixed-model analysis to determine the relationship between miR-122, miR-24, and miR-223 and the clinical parameters reported in this study (see Tables A1.2-4).  A detailed description of the statistical models used is presented in the Materials and Methods section (Model 1).  Linear regressions were plotted to visualize some of the prominent relationships established using this analysis (Figure 2.2 A-C). By applying the mixed model to our clinical data sets, we demonstrated a robust, positive linear relationship between miR-122 and APRI scores (Figure 2.2A), and between miR-122 and HCV virus load (Figure 2.2C).   Consistent with the Spearman’s correlation analysis, we observed a negative linear relationship between miR-24 and miR-223 and liver injury scores [APRI (Figure 2.2A) and FIB-4 (Figure 2.2B)].  Additionally, positive, linear relationships were confirmed between circulating levels of miR-24 and miR-223 and platelet count (Tables A1.2-4).  Because we observed very similar correlations between miR-24 and miR-223 with the clinical parameters studied in our cohort, we next investigated a possible relationship between these two lipoprotein-associated miRNAs by calculating their Spearman's correlation coefficient. The rhos value observed (rhos = 0.91, p-value < 0.0001) is indicative of a very strong, positive monotonic correlation between circulating levels of miR-24 and miR-223 (Figure 2.2D). In contrast, both miR-24 or miR-223 was weakly correlated with miR-122 levels (rhos = 0.3, p-value < 0.0001) (Figure 2.2E) and (rhos = 0.3, p-value < 0.0001) (Figure 2.2F), respectively. Finally, the association between miR-24 and miR-223 was further confirmed using our linear mixed-model. The results presented in Tables A1.3-4 demonstrated a significant, positive linear relationship between circulating plasma levels of miR-24 and miR-223 in all CHC patients.   	   50	       	   51	  Figure 2.2. Circulating miR-24 and miR-223 levels in plasma show a strong positive linear relationship in all CHC patients and negatively correlate with liver injury and liver fibrosis.  (A-C) Graphs representing the negative linear relationship between miR-24 and miR-223 levels and liver injury scores [APRI (A) and FIB-4 (B)] and the robust, positive linear relationship between miR-122 and APRI scores (A), and miR-122 and HCV virus load (C).  The slopes and ordinates values (see Tables A1.2-4) of the plots presented in panels (A-C) were estimated by applying a mixed model analysis to our clinical data sets as described in the Materials and Methods section.  (D-E) Spearman's correlation plot of circulating miR-223 and miR-24 levels (D), miR-24 and miR-122 levels (E), and mir-223 and miR-122 levels in all CHC patients.  The rhos value observed (rhos = 0.91, p-value < 0.0001) is indicative of a very strong, positive monotonic correlation between circulating levels of miR-24 and miR-223 (D).  In contrast, both miR-24 or miR-223 weakly correlated with miR-122 levels (rhos = 0.3, p-value < 0.0001) (E) and (rhos = 0.3, p-value < 0.0001) (F), respectively.    	   52	  2.3.3 Circulating miR-24 and miR-223 plasma levels significantly increase in patients who have achieved SVR   Since CHC infection is associated with extensive dysregulation of host lipid metabolism, we next investigated if HCV-infected patients who achieved a treatment-based viral cure present altered levels of miR-122, miR-24, miR-223, three important regulators of lipid homeostasis.  To perform this analysis, circulating levels of plasma miR-122, miR-24, and miR-223 were measured in paired samples of HCV patients before (baseline) and after antiviral treatment.  Data were normalized to the miRNA baseline levels for each patient before treatment (week 0) and analyzed based on treatment outcomes (SVR and relapsers).  Non-responders were excluded from this analysis due to the lack of post treatment samples. The results revealed specific dysregulation of plasma miR-24 and miR-223 levels in patients who achieved SVR, characterized by a significant 6- and 7-fold increase in circulating miR-24 and miR-223 levels, respectively (Figure 2.3B-C) after antiviral treatment, while no differences in miR-24 and miR-223 were observed in relapsers.  In contrast, these analyses revealed that plasma miR-122 levels were specifically dysregulated in patients who relapsed after treatment, with a significant 8-fold increase in circulating miR-122 levels after antiviral treatment (Figure 2.3A).  No changes in miR-122 were observed in SVR patients.  Also, no significant changes in miR-24, miR-223 and miR-122 levels were found in SVR patients, non-responders and relapsers at any timepoint during treatment compared to baseline levels (Figure A1.1).   To explore further the possible clinical significance of the data shown in Figure 2.3, an ANOVA model was fitted to the data (Model 2).  This model was applied to analyse the difference in circulating miRNA levels and APRI and FIB-4 scores before and after treatment, specifically, the mean level at baseline compared to the mean level across the follow-up timepoints tested in SVR and relapsers (Table 2.3).  This analysis demonstrated a statistically significant difference in circulating miRNA levels after treatment compared to baseline.  As in the earlier analysis, miR-223 and miR-24 plasma levels increased after treatment in patients who achieved SVR, (ΔCt(miR-24), ΔCt change = -1.2, p-value < 0.0001; ΔCt(miR-223), ΔCt change =  -1.4, p-value < 0.0001).  In contrast, plasma levels of   	   53	     	   54	  Figure 2.3. Circulating miR-24 and miR-223 plasma levels, not miR-122, significantly increase in patients who have achieved SVR.  Plasma levels of miR-24, mir-223, and miR-122 were specifically dysregulated for the SVR patients and relapsers after antiviral treatment (FU) when normalized to individual baseline levels before treatment (WO).  The results revealed a specific dysregulation of plasma miR-24 and miR-223 levels with SVRs, characterized by a significant 6- and 7-fold increase in circulating miR-24 and miR-223 levels, respectively (A-B) after antiviral treatment.  In contrast, results of our analyses revealed that plasma miR-122 levels were specifically dysregulated for the relapsers, with a significant 8-fold increase in its circulating levels after antiviral treatment (C).  Abundance of each circulating miRNA was measured in paired samples of HCV patients before (baseline: WO) and after treatment (FU).  Data were normalized to individual baseline levels before treatment, expressed as the fold change and analyzed based on treatment outcomes.  Data are shown as means ± SEM and compared with the two-way ANOVA test. **P –value < 0.01, ***, p-value < 0.005 ns = non-significant.   	   55	  Table 2.3. ANOVA model analysis of the change in microRNA levels after treatment completion in SVR patients and relapsers. 	    	   56	  miR-122 decreased after treatment (ΔCt(miR-122), ΔCt change = 1.3, p-value < 0.0001) in this group.  The decrease in miR-122 was consistent with the previous two-way ANOVA analysis presented in Figure 2.3C, where data were normalized to individual baseline levels, although the decrease detected in the earlier analysis was not statistically significant.  No significant differences in circulating miRNA levels were found in relapsers using an ANOVA model (Model 2) (Table 2.3).   Finally, an ANOVA model analysis was also performed on the APRI and FIB-4 scores and, as previously reported (266), APRI scores significantly decreased (APRI score change = -0.6, p-value = 0.0002) in patients achieving SVR compared to their pre-treatment status (Table 2.3).  Taken together, the results of our ANOVA analysis using Model 2 showed differential miRNA levels depending on the outcomes of the antiviral therapy. Importantly, while miR-122 levels decreased, miR-24 and miR-223 levels significantly increased only in patients who achieved SVR compared to their mean baseline levels (Table 2.3).    2.4 Discussion  Cholesterol and lipid metabolism are tightly regulated by multiple cellular pathways, and their dysregulation leads to various metabolic disorders, including liver steatosis, which may be a direct cause of advanced liver pathologies (267).  HCV infection has been shown to be able to disrupt lipid homeostasis (254, 255).  Despite tremendous advances in treatment of chronic HCV infection with orally administered, direct-acting antivirals, individuals with late-stage liver fibrosis remain at an ongoing risk of cirrhosis and hepatocellular carcinoma even after they are cured by treatment (127).  A better understanding of key regulators of lipid metabolism modulated during HCV infection and involved in the development of advanced liver pathology will be important to help uncover pathways that may represent potential therapeutic targets to prevent liver disease progression.  The data presented in this chapter demonstrate differential changes in circulating miRNAs (miR-122, miR-24, miR-223) before, during and after treatment-based cures for HCV.  Our results show a decrease in circulating miR-122 and increases in miR-24 and miR-223 after treatment, compared to baseline, in SVR.  These results reveal that miRNAs known 	   57	  to act as regulators of lipid metabolism are correlated with IFN-based therapeutic outcomes in patients with HCV infection.  2.4.1 miR-122 in CHC infection  MiR-122 plays a crucial role in regulation of cholesterol and fatty acid metabolism in the adult liver (90).  MiR-122 is a liver-specific miRNA and is one of the most abundant miRNAs in the liver, accounting for about 70% of the whole hepatic miRnome in human adults (268). Antisense-mediated inhibition of hepatic miR-122 has been shown to markedly lower plasma cholesterol levels in both mice and non-human primates (89, 90, 269).  Transcriptomic analyses in mice further reveal that transient miR-122 sequestration downregulated expression of genes involved in fatty acid metabolism as well as cholesterol biosynthesis, including the rate-limiting enzyme 3-hydroxy-3-methylglutaryl-CoA-reductase (89, 90, 270).  Our results show that circulating plasma miR-122 levels decrease when patients achieve SVR compared to before treatment.  We also demonstrated a correlation between circulating miR-122 levels and those of liver transaminases that may directly reflect hepatic necroinflammation levels. Also, miR-122 correlation with HCV RNA levels has been supported by previous studies in serum (271, 272).  To our knowledge we are the first to establish that circulating miR-122 has a linear relationship with the APRI score.  Our results further extend the hypothesis that plasma/serum miR-122 can serve as a marker of therapeutic outcome and liver injury.  Contrasting results have been reported on serum miR-122 pretreatment levels in patients with different responses to IFN-based therapy (272, 273).  In agreement with Köberle et al. (272), in our study, plasma miR-122 pre-treatment levels did not differ between SVR and non-SVR patients. Similarly, in Köberle study no correlation was reported between serum miR-122 and HCV RNA levels at baseline, while the decline in HCV RNA upon beginning therapy closely correlated with the reduction of serum miR-122 (272). Furthermore, a recent study reported that serum miR-122 levels are reduced following treatment in subjects who achieve SVR when treated with IFN-free direct-acting antiviral therapy (274), which supports the hypothesis that changes in circulating miR-122 levels observed before and after treatment in SVR patients correlate with HCV infection independent of the treatment regimen. 	   58	  2.4.2 miR-24 and miR-223 correlate with CHC infection    Many studies have attempted to investigate the role of circulating miR-122 during HCV infection and liver fibrosis; however, no clinical studies have reported on circulating miR-24 and miR-223 during HCV infection, despite the involvement of both miRNAs in a number of other inflammatory disorders, including cancers and cardiovascular and metabolic diseases (275-277).  We have demonstrated more than 6- and 7-fold upregulation in miR-24 and miR-223 levels, respectively, compared to baseline levels, in patients who achieve SVR after treatment but not in those who relapse. These data suggest a direct viral effect on plasma levels of miR-24 and miR-223. The observed increase in miR-24 and miR-223 likely reflects a physiological response to viral clearance that may impact metabolic changes observed in patients who achieve SVR.  In particular, the increase in miR-223 might suggest improved liver outcomes following viral cure as a recent study showed that miR-223 levels were significantly lower in the sera and liver biopsies of patients with hepatocellular carcinoma compared to healthy volunteers (278).   We also demonstrated an excellent correlation between miR-24 and miR-223 (rhos = 0.91, p-value < 0.0001) in CHC patietns. Both miRNAs have previously been reported to be involved in metabolic diseases. Increased miR-24 levels correlate with enhanced processing of sterol regulatory element-binding proteins (SREBPs) and increased lipid accumulation in human hepatocytes, through a mechanism thought to involve targeting insulin-induced gene 1 (INSIG1).  Induction of miR-24 was also observed in livers of mice fed a high-fat diet and in human hepatocytes treated with fatty acid (101).  Studies of miR-223 have shown correlation with monocyte differentiation and the regulation of multiple inflammatory genes in monocytes and macrophages (279, 280).  Hepatic miR-223 levels were also found to significantly increase upon ischemic/reperfusion injury (281) and decrease in hepatocellular carcinoma (282).  Recently, it was found that miR-223 promoter activity and miR-223 levels correlate to intracellular cholesterol changes and that miR-223 regulates cholesterol biosynthesis, uptake, and efflux, thus establishing its critical role in the post-transcriptional regulation of cholesterol metabolism (107).  Our findings show that there is a correlation between circulating miR-24 and miR-223, therefore this could suggest a common mechanistic pathway and/or target genes between these two miRNAs.  Previous studies reported that a significant proportion of circulating miRNAs are 	   59	  associated with lipoprotein complexes (e.g., high-density lipoprotein [HDL] or low-density lipoproteins [LDL]) (77, 83, 260). MiR-223 is the most abundant miRNA associated with both HDL (>10,000 copies of miR-223 per µg of HDL total protein) and LDL (1,500 copies/µg LDL) (83).  Furthermore, both HDL-associated miR-223 and miR-24 are the most upregulated miRNAs with >3,000-fold and 65-fold increase, respectively, in human familial hypercholesterolemia (77).  It has been established previously that HCV patients who achieve SVR have increased LDL and cholesterol from baseline compared to non-SVR patients (256, 283).  This indicates that upregulated levels of circulating miR-24 and miR-223 in SVR patients could be directly associated with an upregulation in lipoproteins levels that serve as carriers of these miRNAs.  2.4.3 Summary  Limitations of this study include the small number of paired samples before and after treatment in relapsers, and the lack of after-treatment samples in non-responders for measuring changes in circulating miRNA levels.  Also, some patients, as indicated, were receiving lipid-modulating agents, which could affect plasma miRNAs levels though the number of these patients was low.  While measurement of the level of cholesterol and specific lipoproteins in plasma before and after anti-HCV treatment would have been interesting to perform, patients in this study were not instructed to fast before blood draws so quantification of these molecules would have been unreliable.   In conclusion, clearance of HCV using an IFN-based therapy results in rapid changes in levels of circulating miRNAs known to have a regulatory role in lipid metabolism, including miR-122, miR-24, and miR-223.  These data suggest that HCV replication may affect lipid homeostasis by modulating regulatory miRNAs.  The pathophysiological mechanisms that lead to the development of progressive liver disease in HCV-infected patients who achieve a treatment-based viral cure are still unclear, and new insight is needed to understand pathogenesis and suggest therapeutic interventions. Our data suggest that differential regulation of host regulators of lipid metabolism are associated with treatment outcome, and maybe associated with the progression of liver disease.  Further investigation is warranted of lipid pathways and metabolites as possible predictors of prognosis and likely treatment outcome.  	   60	  Chapter 3: Elevated plasma PCSK9 levels in hepatitis C-infected patients who achieve a treatment-based viral cure  3.1 Introduction  Hepatitis C virus (HCV) establishes infection in the liver, an organ which plays an essential role in regulating cholesterol and lipid homeostasis through production and uptake of lipoproteins (284).  HCV infection is linked to the metabolism of lipids within the hepatocytes and also dysregulation of circulating lipoprotein metabolism, which allows for efficient viral propagation and persistence (181, 182, 285, 286).  Assembly of HCV particles takes place around host lipid droplets (LDs), which serve as the source of lipids to assemble very low density lipoproteins (VLDL) (4).  HCV hijacks the VLDL synthesis and secretion pathway prior to egress from infected cell (189, 190).  As a result, secreted viral particles, also known as lipoviroparticles (LVPs), are associated with lipoprotein components that assist with virus entry and shield the virus from neutralization (163).  LVPs interact with the low-density lipoprotein receptor (LDLR) to promote viral uptake into hepatocytes (164, 165) in conjunction with other well studied HCV entry receptors including CD81 (167), scavenger receptor class B member 1(SR-BI) (166), claudin-1 and occludin (168, 169).   Proprotein convertase subtilisin/kexin type 9 (PCSK9) is the major post-translation regulator of LDLR levels in the liver (9-11).  PCSK9 is the ninth member of the proprotein convertase (PC) family of secretory pathway serine proteases that are involved in maturation and activation of a variety of host and pathogen precursor propeptides [reviewed in (18)].  PCSK9 is synthesized as a zymogen proprotein in the ER where it undergoes autocatalytic cleavage of its N-terminal prosegment prior to secretion and binding to LDLR at the cell surface separately from the LDL binding domain (8).  Following receptor-mediated endocytosis, the tight binding of PCSK9 to LDLR prevents LDLR recycling back to the plasma membrane and results in LDLR degradation in lysosomes (8).  Thus, increased PCSK9 expression and gain-of-function mutations in PCSK9 results in decreases in LDLR levels and a reduction in clearance of plasma LDL that have been associated with 	   61	  hypercholesterolemia (12-14).  Loss-of-function PCSK9 mutations are associated with low circulating cholesterol levels due to increased LDLR abundance on the surface of liver cells (15, 16).  Also, circulating concentrations of PCSK9 directly correlate with LDL and total cholesterol concentrations (8, 287, 288).  Moreover, a recent paper showed that PCSK9 levels inversely correlated with HCV LVP levels in humans (289).     The unique function of PCSK9 and the impact of HCV infection on lipid homeostasis prompted our investigation into the potential effect of viral cure on concentration of extracellular PCSK9 in a cohort of chronic HCV (CHC) patients undergoing interferon-based treatment.  We found that plasma PCSK9 concentration was significantly upregulated in SVR patients but not in relapsers after treatment.  We further confirmed that increased extracellular PCSK9 limits HCV infection in human hepatoma cells.  We also demonstrated that PCSK9 gain- and loss-of-function mutants allow various degrees of HCV infection inhibition.  Together these results suggest that increases in circulating PCSK9 levels in SVR patients could have an important biological effect on HCV by suppressing virus entry. During chronic HCV infection, the virus itself may be suppressing PCSK9, and following successful treatment, PCSK9 bounces back.  3.2 Materials and methods Patient samples.   A subset of stored plasma samples (-80°C) from the cohort of patients with CHC studied in Chapter 2 were retrospectively analyzed.  Patient samples were collected at an outpatient clinic (certificate # H13-01770) to monitor treatment outcomes.   Patients were treated with interferon-based antiviral therapy that included pegylated interferon-alpha (PEG-IFN) and Ribavirin (RBV) and/or [boceprevir (BOC)/telaprevir (TPV)] for 24-48 weeks.  Patients were classified having achieved a SVR when HCV RNA was not detected at 12 and 24 weeks after the end of therapy.  Relapsers were patients who were negative for HCV RNA at the end of therapy but were HCV RNA positive at either 12 or 24 weeks after the end of therapy.  Non-responders were patients with detectable HCV RNA at all timepoints during and at the end of therapy.  Plasma samples were evaluated at baseline before therapy (week 0) and after therapy (12 or 24 weeks post-treatment) in SVR and relapsers, while in non-responders 	   62	  plasma samples were collected at baseline only. HCV RNA was quantified using the COBAS AmpliPrep/COBAS Taqman® HCV-Test form Roche Molecular Systems.  Routine clinical blood parameters were measured at sample collection timepoints. Patients co-infected with hepatitis B virus or HIV-1 were excluded.  Data from patients receiving long-lasting, lipid-modulating drugs or metformin therapy for diabetes were included.  Plasma samples (n=8) were included as controls from healthy donors collected from volunteers at the University of British Columbia.  The protocol was approved by the Ethics Committee of the University of British Columbia (certificate # H13-01770).   EDTA plasma collection.   Collection of EDTA plasma from healthy controls was performed following the Early Detection Research Network (EDRN) Standard Operating Procedure (SOP). Briefly, whole blood was collected in EDTA Blood Collection Tubes (BD vacutainers). After collection, the blood was gently mixed by inverting the tube 8 to 10 times and centrifuged for 20 min at 11,000 g at room temperature.  Vacutainer tubes were stored upright at 4ºC until centrifugation and blood samples were centrifuged within four hours of blood collection.  After centrifugation, plasma layer formed at the top of the tube was collected, aliquoted and stored at -86ºC.   PCSK9 ELISA.   Plasma PCSK9 concentrations were determined using CircuLex human PCSK9 ELISA kits in accordance with the manufacturer's instructions (MBL, Woburn, MA, USA).  Briefly, all reagents were brought to room temperature prior to the assay and working solutions of Wash Buffer and standard curve (Std1 [0.16 ng/mL]- Std7 [10 ng/mL]) were prepared.  Plasma samples were thawed on ice and diluted 100-fold with Dilution Buffer.  Standard solutions and diluted samples were pipetted in duplicates in the CircuLex microtiter plates pre-coated with an antibody specific for PCSK9.  Following 1-hour incubation period at room temperature, shaking at 300 rpm on an orbital microplate shaker, each well of the microtiter plate was washed with Wash Buffer.  Afterwards, HRP conjugated Detection Antibody was added into each well, incubated for another hour at room temperature as before and washed with Wash Buffer.  Substrate Reagent was added to each well for 20 min and then the 	   63	  reaction was stopped with an acidic Stop Solution added in the same order as the previously added Substrate Reagent.  Absorbance of the resulting yellow product was measured using spectrophotometric microplate reader at 450 nm. PCSK9 concentration values were derived from the absorbance value in each samples based on the regression analysis of the standard curve.   Cell culture.   Human hepatoma Huh-7.5.1 cells were kindly provided by Dr. Francis Chisari (Scripps Research institute, La Jolla, CA, USA). Cultured cells were grown in Dulbecco’s modified Eagle medium (DMEM) supplemented with 1% penicillin/streptomycin, glutamine, non-essential amino acids and HEPES, and 10% fetal bovine serum (FBS) (Gibco/Invitrogen, Burlington, ON, Canada) or 10% lipoprotein-depleted serum (LPDS) (Biomedical Technologies Inc., Stoughton, MA, USA).   Effect of various recombinant extracellular PCSK9 treatment.   Huh-7.5.1 cells plated in complete media were switched to LPDS (lipoprotein-depleted serum) supplemented media for 24 hours before treatment with his-tagged, wild-type (wt) PCSK9 (wt-PCSK9), gain-of-function PCSK9-D374Y, loss-of-function PCSK9-R194A (Circulex at MBL International, Woburn, MA, USA), or bovine serum albumin (BSA) (Sigma-Aldrich Corp., St. Louis, MO, USA).  After 8 hours, the cells were infected with HCV MOI 0.5. Recombinant PCSK9 was replaced 24 hours after the first PCSK9 treatment; the cells were then fixed with 3.7% formaldehyde (Fischer Scientific, Pittsburg, PA, USA) 48 hours following HCV infection. HCV infection was detected with a mouse anti-core antibody (Abcam, Cambridge, MA, USA) and then incubated with the secondary antibody Alexa Fluor-568-conjugated donkey anti-mouse antibody (Molecular Probes/Invitrogen). Hoechst 33258 (10 µg ml-1, molecular probes) was used for detection of nuclei. The percentage of HCV infected cells was determined using Cellomics ArrayScan VTI HCS reader (Thermo Scientific, Nepean, ON, Canada).  Statistical analysis.  Statistical analysis was performed using GraphPad Prism version 6.0 (GraphPad Software 	   64	  Inc., La Jolla, CA). Data are presented as mean ± standard deviation. Data were analyzed by unpaired (Table 3.1) and paired (Figure 3.2 A-B) t-test, Spearman’s correlation (Table 3.2), and a one-way ANOVA (Figure 3.2 C).  The sigmoidal fit function in Igor Pro software (WaveMetrics, Inc., Portland, OR, USA) was used for fitting HCV and recombinant PCSK9 or PCSK9-D374Y inhibition curves and for determining EC50 values. The reported EC50 values are the average of the values calculated from at least three independent experiments plus or minus the standard deviation.  3.3 Results 3.3.1 Cohort characteristics   This study included 50 CHC participants, including 27 achieving SVR, 7 relapsers, and 16 non-responders to anti-HCV therapy. Samples of patients achieving SVR and relapsers included several timepoints: at the baseline before treatment and during the follow-ups, at weeks 12 and/or 24 after the end of treatment; while samples of non-responders were included at the baseline only. The demographics of SVR, relapsers, and non-responders as well as baseline biochemical characteristics and pertinent medications are shown in Table 3.1.   In agreement with previous reports, in our cohort, older patients, patients with lower platelet counts, and patients with higher degree of liver injury and fibrosis based on AST, FIB-4, and Fibroscan scores were less likely to achieve SVR (265).   3.3.2 Plasma PCSK9 levels increase after treatment-based SVR  To further examine the impact of HCV clearance on lipid homeostasis in HCV-infected patients who achieved SVR, we examined plasma PCSK9 concentrations by ELISA.  First, we determined that PCSK9 can be detected in plasma samples following our protocol and established that PCSK9 concentration in our cohort of healthy individuals was 367.5±161.8 ng/ml (Figure 3.1).  Next, we tested PCSK9 concentrations in CHC individuals undergoing IFN-based treatment.  Using a paired t-test (two-tailed), we compared the mean  	   65	  Table 3.1.  Patient demographics and baseline biochemical characteristics.    	   66	   Figure 3.1. Plasma PCSK9 levels in healthy individuals.  Plasma PCSK9 concentrations were determined using CircuLex human PCSK9 ELISA kit in eight healthy individuals. Results are reported as the average of two technical replicates.      	   67	   	   68	  Figure 3.2. Plasma PCSK9 concentrations significantly increase in HCV-infected patients who have achieved SVR.  A paired dependent t-test was used to determine differences associated with plasma PCSK9 concentrations before and after treatment for SVR (A) and relapsers (B).  PCSK9 levels were significantly higher (p=0.002) in plasma samples from SVR patients (A) but not from relapsers  (B). No significant differences were observed in baseline plasma PCSK9 concentrations between patients achieving SVR, non-responders (NR), or relapsers (C).  Straight line in data represents mean, and data were compared using two-tailed paired t-test (A and B) and with a one-way ANOVA (C).  Results are reported as the average of two technical replicates.   	   69	  plasma samples from SVR patients but not from relapsers (Figure 3.2A-B), suggesting a link between circulating PCSK9 and HCV.  We did not observe significant differences in baseline plasma PCSK9 concentrations between patients achieving SVR, non-responders, or relapsers (Figure 3.2C); however, there was a non-significant increase in PCSK9 concentrations in both relapsers and non-responders compared to patients achieving SVR.  Interestingly, using Spearman’s correlation, we observed a positive, significant correlation between plasma PCSK9 concentrations and circulating plasma levels of miR-24 [ΔCt(miR-24), rhos = -0.24, p-value = 0.0174] (Table 3.2).   3.3.3 Extracellularly applied recombinant human wild-type PCSK9 and gain-of-function mutant PCSK9-D374Y, but not loss-of-function PCSK9-R194A variant, inhibited HCV infection in human hepatoma Huh-7.5.1 cells  To further explore the hypothesis that circulating PCSK9 levels may be linked to HCV infection, we tested the effect of extracellularly applied human PCSK9 on HCV infection in human hepatoma Huh-7.5.1 cells.  Specifically, we evaluated the modulatory effect of increasing PCSK9 concentration on HCV infection by employing recombinant human wild-type PCSK9 (wt-PCSK9) (290, 291) and PCSK9 gain- and loss-of-function mutants (Figure 3.3A-B) in a cell culture model of HCV (JFH-1) infection. Huh-7.5.1 cells were pre-treated for 8 hours with increasing concentrations of recombinant wt-PCSK9 (from 0.15 µg/ml to 25 µg/ml) prior to HCV infection (Figure 3.3C-D).  Pretreatment of Huh-7.5.1 cells with wt-PCSK9 for 8 hours resulted in a degradation of mature LDLR (data not shown) as expected based on previous studies (11, 292).  Under these experimental conditions, 25 µg/ml of wt-PCSK9 inhibited HCV infection (98.6% or 42-fold) measured by intracellular levels of HCV core protein with a median effective concentration (EC50) value of 4.63±0.06 µg/ml (Figure 3.3D).   In contrast, when Huh-7.5.1 cells were pretreated with 25 µg/ml of the recombinant loss-of-function mutant PCSK9-R194A (Figure 3.3B), no significant reduction in HCV infection was detected compared to control cells (Figure 3.3E).  Residue R194 of PCSK9 forms a hydrogen bond with LDLR (293); thus, the R194A mutation destabilizes this  	   70	  Table 3.2.  Correlations of extracellular PCSK9 with biochemical parameters.    	   71	   	   72	  Figure 3.3. Extracellularly applied recombinant PCSK9 and the gain-of-function mutant PCSK9-D374Y but not the loss-of-function PCSK9-R194A variant inhibits HCV infection in Huh-7.5.1 cells.   (A) PCSK9 is a secreted protein highly expressed in the liver where it plays an important role as a post-translational regulator of LDLR levels. It is biosynthesized as preproprotein that contain a prodomain, catalytic domain, and cysteine-histidine-rich domain (CHRD). PCSK9 catalytic triad contains aspartate 186 (D186), histidine 226 (H226) and serine 386 (S386). (B) Gain-of-function mutations such as D374Y in PCSK9 have been associated with hypercholesterolemia due to lower levels of LDLR (10× higher affinity for LDLR) and reduced clearance of plasma LDL. Loss-of-function PCSK9 mutations (R194A) are conversely associated with abnormally low circulating cholesterol levels due to increased LDLR abundance  (lower affinity for LDLR) on the surface of liver cells. (C-D) Huh-7.5.1 cells grown in LPDS-supplemented media were treated with varying concentrations of recombinant PCSK9 or PCSK9-D374Y, BSA or untreated control or (E) 25 µg/ml PCSK9-R194A, for 8 hours. Treated cells were infected with HCV MOI 0.5 and cells were fixed 48 hours post-infection. Cells, probed with HCV core specific antibodies and stained with Hoechst dye to visualize cell nuclei, were counted using Cellomics HCS to determine the percentage of total HCV-infected cells. Representative images are shown acquired at 10X magnification objective (Cellomics HCS) (C). The EC50 values for recombinant PCSK9 and PCSK9-D374Y mutant were calculated based on the dose response (D). Values are expressed as relative HCV infection in treated cells compared to untreated cells, which are set to 1. Results (mean ± SEM) from 3 independent experiments are shown (C-E).    	   73	  infection was detected compared to control cells (Figure 3.3E).  Residue R194 of PCSK9 forms a hydrogen bond with LDLR (293); thus, the R194A mutation destabilizes this interaction (Figure 3.3B).  These results suggested that inhibition of HCV infection in Huh-7.5.1 cells by wt-PCSK9 is mediated through an LDLR-dependent mechanism.   To confirm these results, we tested the modulatory effect of the gain-of-function mutant PCSK9-D374Y (294) on HCV infection.  As shown in Figure 3.3B, an extra contact is formed between PCSK9 and LDLR at neutral pH when D374 is mutated to Y374 (295, 296).  The result of this natural mutation (D374Y) is a 10-fold increase in LDLR turnover by PCSK9-D374Y compared to wt-PCSK9 (291).  As expected, following pretreatment of Huh-7.5.1 cells with PCSK9-D37Y for 8 hours, the number of HCV core-positive cells markedly decreased (Figure 3.3C-D).  The calculated EC50 value for PCSK9-D374Y inhibition of HCV infection under these experimental conditions was 0.66±0.05 µg/ml (Figure 3.3B).  Interestingly, the calculated ratio between the two reported EC50 values in this study for the anti-HCV activities of wt-PCSK9 and PCSK9-D374Y (~ 7 fold) was closely related to the fold increase (~ 10 fold) in LDLR turnover for wt-PCSK9 and PCSK9-D374Y previously reported (291).  Hence, these results support the hypothesis that PCSK9 levels may correlate with impeded HCV infection through an LDLR-dependent mechanism.  3.4 Discussion  In this study we investigated extracellular concentrations of PCSK9 before and after anti-HCV treatment. An increase in circulating PCSK9 levels is normally associated with the reduction in LDLR in the liver, resulting in higher levels of LDL cholesterol in plasma (12).  Interestingly, in this study, circulating PCSK9 levels were upregulated in SVR patients after the end of treatment compared to baseline levels, while no changes were found in relapsers, which supports previous studies showing patients who achieved SVR had increased LDL and cholesterol from baseline compared to non-SVR patients with both IFN-based and IFN-free therapies (256, 283).  An increase in circulating PCSK9 levels in SVR patients further supports that changes to circulating miRNAs known to act as regulators of lipid metabolism identified in Chapter 2, are not coincidental and suggests dynamic modulations to the 	   74	  lipidome profile upon viral cure.  In Chapter 2, we reported that miR-24 and miR-223 levels were significantly increased in HCV-infected patients who achieve SVR (miR-24, p-value < 0.0001; miR-223, p-value < 0.0001), while miR-122 significantly decreased after HCV clearance (p-value < 0.0001) (Table 2.3). Furthermore, we have shown that upregulation in circulating PCSK9 concentration could have a biological effect on suppressing HCV infection by inhibiting viral entry via downregulation of LDLR in the liver, which we demonstrated using HCV infection in cell culture.  HCV particles are associated with lipoprotein components and are termed lipoviroparticles (LVPs) that interact with the LDLR to promote viral uptake into hepatocytes (297).  Therefore, a decreased level of PCSK9 during HCV infection could be beneficial for the virus via upregulation of the entry receptor.  Previously, it has been shown that stable overexpression of PCSK9 or a chimeric, membrane-anchored PCSK9 mutant (PCSK9-ACE2) could be used to render Huh-7 cells resistant to HCV infection and that increased extracellular PCSK9 has a suppressive role in HCV infection (292).  Also, a recent paper showed PCSK9 levels inversely correlated with HCV LVP levels in humans (289).  Here, we confirmed the suppressive effect of extracellular PCSK9 on HCV infection and demonstrated the relevance of PCSK9 mutants.  We showed that a gain-of-function PCSK9 mutant is much more efficient in inhibiting HCV (EC50 0.66±0.05 µg/ml) than wild-type PCSK9 (EC50 4.63±0.06 µg/ml) and was particularly evident at lower and more physiologically relevant concentrations previously shown to range between 0.033 µg/ml to 3 µg/ml (298).  In contrast, a loss-of-function PCSK9 variant was determined to have no effect on HCV infectivity.  These results have important implications as a variety of PCSK9 mutations found within the human population may have a significant impact on the levels of liver LDLR and may also influence host susceptibility to HCV infection and treatment outcomes.  In conclusion, clearance of HCV using an IFN-based therapy results in rapid changes of PCSK9 levels, an important regulator of lipid metabolism. This implicates an effect of HCV replication on lipid homeostasis via modulation of regulatory markers.  Mechanisms of the development of progressive liver disease in HCV-infected patients who achieve a treatment-based viral cure are unclear, but these mechanisms are important to understand for the development of appropriate monitoring and treatments.  Our data suggest that differential regulation of host regulators of lipid metabolism may be associated either with treatment 	   75	  outcome or liver disease progression and support further investigation of lipid pathways and metabolites as predictors of progressive liver disease and treatment outcome.  	   76	  Chapter 4: Human subtilase SKI-1/S1P regulates cytoplasmic lipid droplet abundance: a potential target for indirect-acting anti-dengue virus agents  4.1 Introduction Dengue virus (DENV) represents a significant threat to global public health, with approximately 390 million cases annually and about 2.5 billion people living in endemic countries (218, 299, 300).  DENV is the causative agent of dengue fever (DF) and of life-threatening severe dengue, including dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS) (301).  Although DENV was first isolated more than 70 years ago, current treatment and prevention approaches are still limited to relief of symptoms and vector control (301-304).  Currently, four DENV serotypes (DENV-1 to -4) transmitted by Aedes aegypti and Aedes albopictus mosquitoes are known to circulate in humans (218, 305).  All four DENV serotypes are considered to be hyper-endemic in most tropical and subtropical areas of the world, and they are poised to spread into new territories (218, 306).  A better understanding of host-DENV interactions and DENV pathogenesis is urgently needed to design broad-spectrum antivirals that will be effective against all four DENV serotypes.   The DENV serotypes are members of the Flavivirus genus with single-stranded positive-sense RNA genomes encoding three structural proteins (capsid [C], precursor membrane [prM], and envelope [E]) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) (307).  RNA viruses are associated with intrinsically high rates of mutation, with the DENV-4 evolution rate estimated at 6.89 × 10−4 substitutions/site/year (308, 309).  Given the importance of reliably targeting all four DENV serotypes and limiting the formation of antiviral resistance, indirect-acting antivirals (IAA) that interfere with the viral hijacking of host factors important for the viral lifecycle are an attractive therapeutic avenue (37, 310).  Cellular factors such as lipids and cholesterol are involved in the DENV lifecycle (245, 247, 311-313).  Different drugs targeting either lipid or cholesterol pathways have been tested, including an inhibitor of fatty acid synthase (C75), an inhibitor of intra-cellular 	   77	  cholesterol transport (U18666A), inhibitors of cholesterol synthesis (lovastatin, fluvastatin, and pravastatin), and the hypolipidemic agent arachidonic acid 5-lipoxygenase inhibitor (nordihydroguaiaretic acid).  All of these inhibitors achieved variable reductions in DENV virus replication or infectious particle formation (244, 314-316), underlining the importance of cellular lipids and, in particular, lipid droplets (LD) in DENV infection.  LDs are dynamic intracellular lipid storage organelles that play multiple roles during the DENV lifecycle (244, 247).  They consist of a neutral lipid core (e.g., triglycerides and cholesterol esters) surrounded by a phospholipid monolayer containing LD-associated proteins such as adipose differentiation-related protein (ADRP) (317).  In this study, we investigated the molecular functions of human subtilisin kexin isozyme-1/site-1 protease (SKI-1/S1P), a key master regulator of the lipid homeostasis/sterol regulatory element-binding protein (SREBP) pathway (33), in the formation of cellular lipid storage droplets and the DENV lifecycle.  In mammals, the biosynthesis of cholesterol, fatty acids, and triglycerides is tightly regulated by a family of transcriptional factors called SREBPs.  Two genes encode three SREBP isoforms: SREBP-1a, SREBP-1c, and SREBP-2 (23).  SREBP-2 and SREBP-1c are the predominant forms in the liver, regulating genes involved in sterol biosynthesis and fatty acid synthesis, respectively (318).  The inactive precursor of SREBP (pre-SREBP) is synthesized in the endoplasmic reticulum (ER) as a membrane-bound protein.  Its activation is dependent on the presence of sterols and requires the cleavage of pre-SREBP to release the mature form, which then translocates to the nucleus.  When sterol levels are low, the SREBP cleavage-activating protein (SCAP) escorts SREBP from the ER to the Golgi apparatus, where SREBP is sequentially cleaved by two cellular proteases: SKI-1/S1P and site-2-protease (S2P) (24).  These cleavages liberate the N-terminal fragment of SREBP (n-SREBP), which further translocates into the nucleus to activate genes involved in lipid and cholesterol metabolism, such as low density lipoprotein receptor (LDLR) and proprotein convertase subtilisin/kexin type 9 (PCSK9) (19, 23, 25).  When cellular sterol levels are high, the insulin-induced gene protein (INSIG) associates with SCAP, which causes the SCAP–pre-SREBP complex to be retained in the ER, thereby preventing the formation of n-SREBP and decreasing the expression of SREBP target genes (26).  	   78	  Given that SKI-1/S1P-mediated SREBP proteolytic activation controls the expression of genes directly involved in intracellular fatty acid and cholesterol biosynthesis (24), two important components of LDs, we hypothesized that pharmacological inhibition of the subtilase SKI-1/S1P could represent a powerful approach to developing a novel class of broad-spectrum antivirals against DENV infection.  We proposed that strategic manipulation of human SKI-1/S1P enzymatic activity would effectively inhibit DENV infection by blocking cytoplasmic LD formation and interfering with DENV hijacking of cytoplasmic LDs, a critical event in the DENV lifecycle (244). Here using an active-site-directed aminopyrrolidineamide-based inhibitor of SKI-1/S1P, PF-429242 (33), we demonstrated that inhibition of host SKI-1/S1P enzymatic activity effectively blocks DENV (1-4) from establishing infection in human hepatoma Huh-7.5.1 cells.  PF-429242 antiviral activity was observed both pre- and post-establishment of viral infection and was associated with a dramatic decrease in LD abundance in PF-429242-treated Huh-7.5.1 cells.  Our studies demonstrate SKI-1/S1P’s potential as a novel host-directed pan-serotypic anti-DENV target, and they reveal therapeutic opportunities associated with the use of lipid-modulating drugs for controlling DENV infection.  4.2 Materials and methods Cell culture and reagents.   Human hepatoma Huh-7.5.1 cells were kindly provided by Dr. Francis Chisari (Scripps Research Institute, La Jolla, CA, USA) (319).  Cultured cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 1% penicillin/streptomycin, 1% L-glutamine, 1% nonessential amino acids, 1% 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), and 10% heat-inactivated fetal bovine serum (FBS) (Gibco/ Invitrogen, Burlington, ON, Canada).  Vero E6 cells (ATCC #CC1-81) were cultured in M199 medium (Sigma-Aldrich Corp., St. Louis, MO, USA) supplemented with 1% HEPES, 1% L-glutamine, 1% sodium bicarbonate, and 5% FBS (Gibco).  Bovine serum albumin (BSA) and oleic acid were obtained from Sigma-Aldrich Corp.   	   79	  Viruses and infections.   The four serotypes of DENV were kindly provided by Dr. Mike Drebot from the National Microbiology Laboratory (Winnipeg, MB, Canada): DENV-1, Hawaiian; DENV-2, New Guinea C; DENV-3, H-87; and DENV-4, H-241. Huh-7.5.1 cells were either inoculated with DENV-1, -2, -3, or -4 (MOI = 1 or 0.01) or mock-infected at 37°C for 1 hour before the inoculum was removed and fresh complete media was added.  Cell supernatant, lysates, and/or RNA were collected at various timepoints post-infection for analysis by secondary infection/plaque assay, Western blot, and/or reverse transcription quantitative real-time PCR (RT-qPCR), respectively.   Small-molecule inhibition of SKI-1/S1P.   PF-429242 (chemical name: 4-[(Diethylamino)methyl]-N-[2-(2-methoxyphenyl)ethyl]-N-(3R)-3-pyrrolidinylbenzamide), a reversible, competitive small-molecule aminopyrrolidine-amide inhibitor of SKI-1/S1P (33, 37, 320), was synthesized by Dr. Peter Chua at the Center for Drug Research and Development (CDRD) at the University of British Columbia (Vancouver, BC, Canada) according to previously described protocols (321).  The chemical was dissolved in DMSO and stored at a concentration of 100 mM.  To serve as a negative control, an intermediate product, acetylated PF-429242 (AcPF-429242), was also synthesized at CDRD.  To investigate the antiviral activity of the small molecule, Huh-7.5.1 cells were treated with PF-429242 for 24 hours.  Media was then removed and the cells were infected with one of the DENV serotypes (DENV 1–4) for 48 hours.  Alternatively, cells were first infected with DENV-2 for 24 hours; then the media was removed and replaced with fresh media supplemented with PF-429242 for 48 hours. DMSO and AcPF-429242-treated cells were used as controls.  For lipid supplementation assays (322), Huh-7.5.1 cells were treated with PF-429242 as described above with the addition of 0.6 mM oleic acid (with BSA in molar ratio 6:1) or the equivalent concentration of oleic acid/BSA alone during DENV infection.   Cytotoxicity assay.   Cell viability was determined using CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Madison, WI, USA) following manufacturer’s instructions. Briefly, Huh-	   80	  7.5.1 cells were treated with different concentrations of PF-429242 or AcPF-429242 for 24 hours in the 96-well plate.  After 24 hours, media containing the inhibitor was removed and fresh media was added to the cells for an additional 48 hours.  Following the incubation period, 20 µl of CellTiter 96 Aqueous One Solution Reagent was added to each well containing the samples in 100 µl of culture medium and the plate was incubated for 1 hour at 37°C.  Production of formazan by cells from a tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; MTS] was detected by measuring absorbance at 490 nm.  Statistical significance was calculated using a one-way ANOVA.  Western blot analysis.   Cultured cells were washed twice with ice-cold phosphate buffered saline (PBS) and re-suspended in a cold radioimmunoprecipitation assay (RIPA) buffer [50 mM Tris–HCl (pH 8), 1% Triton X-100, 0.5% sodium deoxycholate, 150 mM NaCl, and 0.1% SDS] containing a 1X complete, EDTA-free, protease inhibitor cocktail (Roche, Laval, QC, Canada).  Whole cell extracts were vortexed and then clarified by centrifugation at 12,000×g for 15 minutes.  Soluble extracts were mixed with 2X sample buffer (62.5 mM Tris-HCl, pH 6.8, 25% glycerol, 2% SDS, 0.01% bromophenol blue, and 5% β-mercaptoethanol).  Samples stained with anti-NS1 antibody were mixed with sample buffer without β-mercaptoethanol.  Samples were electrophoresed on 10–15% SDS polyacrylamide gels and transferred to nitrocellulose membranes.  Membranes were blocked in Odyssey blocking buffer (LI-COR Biosciences, Lincoln, NE, USA), and proteins of interest were detected by probing with the appropriate primary and secondary antibodies diluted in Odyssey blocking buffer containing 0.1% Tween-20. The membrane was probed using a mouse anti-NS1 (1:50 dilution; Abcam, Cambridge, MA, USA), rabbit anti-β-tubulin (1:3,000 dilution; Abcam), and secondary antibodies IRDye 680-conjugated (red bands) or 800-conjugated (green bands) donkey anti-mouse or goat anti-rabbit antibodies (1:10,000; LI-COR Biosciences).  Protein bands were detected and quantified using the Odyssey Infrared Imaging System (LI-COR Biosciences).  All immunoblots were scanned at a wavelength of 700 nm for detecting IRDye 680-labeled antibodies and at a wavelength of 800 nm for IRDye 800-conjugated antibodies.  Signal intensities were quantified by means of the Odyssey software version 3.0. β-tubulin was 	   81	  consistently used as a loading control and for normalizing protein expression.  A one-way ANOVA was used for data sets with more than two groups to calculate statistical significance, which is represented in the figures by the following notation: * denotes p >0.05, ** denotes p >0.01, and *** denotes p >0.005.  Curve-fitting, half-maximal effective concentration (EC50) determination.   A custom hyperbolic fit function [y=a+bx/(1+cx)] in Igor Pro software (WaveMetrics, Inc., Portland, OR, USA) was used for fitting DENV-2 NS1 expression and PF-429242 inhibition curves and for determining EC50 value.  The reported EC50 values are the average of the values calculated from three independent experiments (± SEM).   Plaque assay.   DENV titers were determined by performing plaque assay as previously described (323).  Briefly, Vero E6 cells monolayers were seeded in 12-well plates (Falcon; Becton Dickinson, Lincoln Park, NJ, USA) and incubated at 37°C in a CO2 incubator.  Supernatants of DENV or mock-infected Huh-7.5.1 cells were tested using tenfold dilutions starting at 1:10.  Plaques were visualized on day 5 by staining with 4% neutral red solution (Sigma-Aldrich Corp.) in PBS.  Statistical significance was calculated by a student’s t-test (paired) based on three independent experiments.  RNA isolation.   Total RNA was isolated using a miniRNeasy kit (QIAGEN, Mississauga, ON, Canada) according to the manufacturer’s instructions.  The concentration and purity of RNA were determined by a NanoDrop ND-1000 Spectrophotometer (Thermo Scientific, Nepean, ON, Canada).  Quantitative real-time PCR (qRT-PCR).   A total of 500 ng of purified total RNA was reverse-transcribed to cDNA using TaqMan reverse transcription reagents (random hexamers; Applied Biosystems, Foster City, CA, USA).  Quantitative RT-PCR was carried out on an Mx3005P real-time PCR system (Stratagene, La Jolla, CA, USA) using Brilliant III Fast QPCR reagents (Stratagene) 	   82	  according to the manufacturer’s instructions.  DENV RNA was analyzed using a previously reported serotype-specific DENV primer probe set (324) (Figure A2.1).  DENV RNA levels were quantified across samples and normalized to β-actin RNA levels using 500 nM primers (forward: 5′-GCC CTG AGG CAC TCT TCC-3′ and reverse: 5′-GGA TGT CCA CGT CAC ACT TC-3′) and 250 nM probe (5′-AC TCC ATG CCC AGG AAG GAA GGC-3′ with a 5′ Cy5 fluorophore and 3′ black hole quencher).  The expression levels of seven cellular mRNAs were quantified by qRT-PCR using TaqMan gene expression assays [Applied Biosystems, assays ID (Hs00965485_g1:FURIN; Hs00545399_m1:PCSK9; Hs00921626_m1:SKI-1/S1P; Hs01092524_m1:LDLR; Hs01081784_m1: SREBP-2; Hs01088691_m1:SREBP-1c;  Hs00210639_m1: S2P].  For data analysis, the 2-ΔΔCt method was used, and mean fold changes in expression are shown relative to mock or control treated samples.  The data were analyzed with one-way or two-way ANOVAs.  Confocal microscopy.   Huh-7.5.1 cells seeded in µ-Slide 8 Well IbiTreat (Ibidi, Madison, WI, USA) were fixed in 3% v/v paraformaldehyde in PBS, then permeabilized in PBS containing 0.01% digitonin or 0.1% Triton X-100.  Cells were probed with primary rabbit anti-capsid (C) antibodies (244) (1:1000, kindly provided by Dr. Andrea V. Gamarnik (Fundación Instituto Leloir-CONICET, Argentina)), then incubated with secondary antibodies (Alexa 647 conjugated goat anti-rabbit), Hoechst dye, and Nile red dye diluted in PBS.  Nile red, 9-diethylamino-5H-benzo[alpha]phenoxazine-5-one, is a selective fluorescent  stain for detecting intracellular lipid droplets  (325).  The wells were then imaged using Leica SP8 confocal microscope (Leica Microsystems, Wetzlar, Germany) with a 100x objective (HC PL APO 100x/1.40 OIL).  All quantified images were acquired using the same laser intensity and gain settings, and LDs were enumerated by applying the same threshold setting to each image. LD abundance was quantified by LD–positive areas (µm2), shown as the means per cell ± SEM (>50 cells analyzed), and LD numbers, shown as the means per cell ± SEM (>50 cells analyzed), based on Nile red fluorescence.   	   83	  4.3 Results 4.3.1 Inhibition of SKI-1/S1P enzymatic activity using PF-429242 impairs activation of the SREBP pathway and correlates with a dramatic decrease in lipid droplet abundance   To test our hypothesis that strategic manipulation of SKI-1/S1P dependent activation of the SREBP pathway would effectively inhibit DENV infection by blocking cytoplasmic lipid droplet (LD) formation, we first investigated the effect of the small-molecule SKI-1/S1P inhibitor PF-429242 on the SREBP pathway in mock-infected and DENV-2-infected Huh-7.5.1 cells. The amino-pyrrolidine amide compound PF-429242 (Figure 4.1A) is a potent and selective catalytic site-directed inhibitor of SKI-1/S1P endoproteolytic activity (33, 37, 321, 326); and under our experimental conditions, the 50% cytotoxic concentration (CC50) of PF-429242 in Huh-7.5.1 cells was > 81	  µM (Figure 4.1B and (37)).  To evaluate the effects of PF-429242 on the SREBP pathway, Huh-7.5.1 cells were pre-treated with 10 or 20 µM PF-429242 for 24 hours.  The inhibitor was removed and the cells were mock-infected or infected with DENV-2 for 48 hours before total RNA was harvested and analyzed by qRT-PCR.  We observed a strongly reduced expression of SREBP-1c and SREBP-2 mRNAs in both the mock-infected and DENV-2-infected Huh-7.5.1 cells (Figure 4.1C and Figure A4.2).  This result is consistent with previous findings showing that transcription of both SREBP-1c and SREBP-2 is stimulated by SREBPs in a feed-forward mechanism that requires sterol regulatory element (SRE) sequences in the promoters of these genes (30, 32).  As expected, the expression of two of the SREBP target genes, PCSK9 and LDLR, was robustly decreased in both mock-infected and DENV-2-infected Huh-7.5.1 cells treated with PF-429242 compared to control (Figure 4.1C and Figure A2.2).  In contrast, expression of two SREBP-independent genes, SKI-1/S1P and furin, was approximately the same in these samples as in controls (Figure 4.1C and Figure A2.2). These results demonstrate that inhibition of SKI-1/S1P enzymatic activity using PF-429242 impairs activation of the SREBP pathway in both mock-infected and DENV-2 infected Huh-7.5.1 cells. Given that SKI-1/S1P-mediated SREBP proteolytic activation controls the expression of genes directly involved in intracellular fatty acid and cholesterol biosynthesis (24), two  	   84	     	   85	  Figure 4.1. Inhibition of SKI-1/S1P using PF-429242 prevents activation of the SREBP pathway and reduces the abundance of cytosolic lipid droplets.  (A) Chemical structure of PF-429242 is shown. (B-E) Huh-7.5.1 cells were treated with various concentrations of DMSO or PF-429242 for 24 hours before the inhibitor was removed and fresh complete media was added to the cells for an additional 48 hours. (B) The relative cytotoxicity of the compound was then determined using an MTS-based cell viability assay. The absorbance measured at 490 nm is proportional to the number of living cultured cells (C) Total RNA was extracted, and the mRNA levels of SREBP-1c, SREBP-2, PCSK9, LDLR, FURIN, and SKI-1/S1P were quantified by qRT-PCR. Results were normalized against β-actin mRNA levels and expressed as fold change. Statistical significance was calculated for PF-429242-treated cells (20 µM) compared to 0.02% DMSO-treated cells (control) with a two-way ANOVA for each mRNA presented.  (D) Representative images of the effect of PF-429242 (20 µM) and 0.02% DMSO on lipid droplets are shown. Fixed cells were permeabilized with Triton X-100 and stained for cell nuclei using Hoechst dye and for lipid droplets using Nile red. Images were obtained using a Leica SP8 confocal microscope with a 100x objective. (E) Abundance of lipid droplets was quantified measuring the lipid droplet-positive area (µm2)/cell and the average number of lipid droplets/cell based on Nile red fluorescence in untreated, 0.02% DMSO-treated, and PF-429242-treated cells (20 µM) using Fiji software in three independent experiments.  Statistical significance was calculated with a two-way ANOVA with a Bonferroni’s post-test. Values represent average ± SEM of three independent experiments. ns = not significant. ****, p < 0.001.  	   86	  important components of LDs, we next tested the hypothesis that pharmacological inhibition of the subtilase SKI-1/S1P by PF-429242 could represent a powerful approach to regulating LD abundance in Huh-7.5.1 cells. To investigate the efficacy of PF-429242 to reduce cytosolic LD abundance, we used Nile red, a selective fluorescent stain for intracellular LDs (325). DMSO-treated (vehicle control), untreated, and 20 µM PF-429242-treated Huh-7.5.1 stained cells were examined by confocal microscopy (Figure 4.1D).  The quantification of the LD-positive area and count using fluorescence intensity of stained cells with Nile red demonstrated a specific effect of PF-429242 in Huh-7.5.1-treated cells, with an overall 50% reduction in LD area and count (Figure 4.1E). In contrast, LD-positive area and count were approximately the same in the DMSO (vehicle control) and untreated cells (Figure 4.1E).  Taken together, these results demonstrate that inhibition of SKI-1/S1P enzymatic activity in Huh-7.5.1 cells using the potent and selective inhibitor PF-429242 impairs activation of the SREBP pathway and results in a decrease in LD formation.  4.3.2 Huh-7.5.1 cells support DENV-2 replication and DENV capsid protein binding to hepatic lipid droplets  Since the SKI-1/S1P-dependent activation of the SREBP pathway and LD formation can be inhibited with PF-429242, we hypothesized that PF-429242 could act as an antiviral agent against DENV infection in Huh-7.5.1 cells.  To test the antiviral properties of PF-429242 in these cells, we first needed to show that Huh-7.5.1 cells support DENV-2 infection.  Therefore, Huh-7.5.1 cells were mock-infected or infected with DENV-2 at a multiplicity of infection (MOI) of 1.  This approach allowed the monitoring of DENV-2 infection in Huh-7.5.1 cells over prolonged periods, whereas at higher MOI a rapid virus-induced cell death is observed thus precluding gene expression profiling of infected cells at the late stage of infection.  Total RNA was extracted at 0, 4, 8, 24, 48, and 72 hpi, and viral replication efficiency was determined by quantifying the expression level of DENV-2 RNA by qRT-PCR.  Alternatively, viral infection was demonstrated by visualizing DENV-2 Capsid (C) protein by indirect immunofluorescence. The temporal expression of DENV-2 RNA is presented in Figure 4.2A.  Newly 	   87	  biosynthesized viral RNA is detected at 24 hpi and continued to increase steadily thereafter reaching the highest level at 72 hpi.  Similarly, confocal microscopic analysis of DENV-2 infected cells at 24 and 48 hpi revealed a robust biosynthesis of DENV-2 C protein in most cells that were analyzed using indirect immunofluorescence (Figure 4.2B and C).  When DENV-infected cells were fixed with paraformaldehyde and permeabilized with Triton X-100, the C protein was found in both the nucleus and cytoplasm at 48 hpi (Figure 4.2B).  When Huh-7.5.1 cells were permeabilized with digitonin, which, unlike Triton X-100, preferentially permeabilizes the plasma membrane leaving the nuclear envelope intact (327), the C protein was primarily distributed to the cytoplasm (Figure 4.2C), forming a ring-like shape on the surface of the LDs at 48 hpi (Figure 4.2D).  These results are consistent with previous studies demonstrating that under specific experimental conditions, the subcellular localization of C protein can be detected either in the cytoplasm or the nucleus of DENV-infected cells (244, 328).  Reduction of the abundance of LDs in cells infected with DENV-2 at 24 and 48 hpi were quantified using fluorescence intensity of stained cells with Nile red (Figure 4.2E). These results are consistent with earlier study by Heaton et al., which demonstrated that DENV infection leads to an autophagy-dependent reduction of LDs (247).  Taken together, our findings demonstrate that Huh-7.5.1 cells support DENV-2 infection, which is consistent with our observation that the DENV capsid protein re-localized to LD in Huh-7.5.1 cells, an important step in the DENV lifecycle (244).   4.3.3 Pretreatment of Huh-7.5.1 cells with PF-429242 results in a dose-dependent decrease in intracellular DENV-2 NS1 expression and a 3-log decrease in extracellular viral titer    In order to examine the effectiveness of PF-429242 as an anti-DENV agent, Huh-7.5.1 cells were pretreated with increasing concentrations of PF-429242 (0.1 µM to 20 µM) prior to infection with DENV-2.  The inhibitor was removed after 24 hours, and the cells were then infected with DENV-2 for 48 hours.  Viral protein synthesis in Huh-7.5.1 cells was monitored by determining the level of DENV-2 NS1 protein abundance in total cell lysates using Western blot analysis.    	   88	    	   89	  Figure 4.2. DENV-infected Huh-7.5.1 cells accumulate the C protein around lipid droplets.  (A-D) Huh-7.5.1 cells were infected with DENV-2 (MOI 1) or mock-infected. (A) Total RNA was extracted at various timepoints (0, 4, 8, 24, 48, and 72h) post-infection, and mRNA expression of DENV-2 RNA was quantified by qRT-PCR. Results were normalized against control β-actin mRNA levels and expressed as fold change relative to the time-matched mock-infected controls. (B-D) Cells were harvested at 24 and 48 hours post-infection. Cells were fixed with paraformaldehyde and permeabilized with Triton X-100 (B) or digitonin (C).  (D) DENV C accumulation around lipid droplets in Huh-7.5.1 cells infected with DENV (MOI 1) at 48 hpi. Fixed cells were stained for cell nuclei using Hoechst dye (blue), for lipid droplets using Nile red (red), and DENV anti-capsid (green).  (E) Abundance of lipid droplets was quantified by measuring the lipid droplet-positive area (µm2)/cell and the average number of lipid droplets/cell based on Nile red fluorescence using Fiji software Images were obtained using a Leica SP8 confocal microscope with a 100x objective. Results (mean ± SEM) from at least three independent experiments are shown. Statistical significance was calculated for DENV-2-infected cells compared to mock-infected cells at timepoints corresponding to DENV-2-infected cells with a two-way ANOVA with a Bonferroni’s post-test. *, p < 0.05; ****, p < 0.001.   	   90	  The host cell pretreatment with PF-429242 resulted in a dose-dependent decrease in the intracellular level of DENV-2 NS1 protein (Figure 4.3A).  A near-complete block in DENV infection was observed following treatment with 20 µM of inhibitor (Figure 4.3B).  Under these experimental conditions, a 19.3-fold reduction in intracellular DENV NS1 was observed when treated with 20 µM PF-429242 (Figure 4.3B).  It was shown that PF-429242 inhibits DENV NS1 synthesis with an EC50 concentration of 1.15 ± 0.23 µM. Next, the effect of inhibiting host SKI-1/S1P on the formation of infectious DENV virions was investigated by plaque assay.  Supernatants from PF-429242 (20 and 30 µM) or DMSO-treated Huh-7.5.1 cells infected with DENV-2 were collected 48 hours post-infection and titrated on naïve Vero E6 cells; plaques were counted five days post-infection.  Results showed a ~3-log decrease in DENV-2 titer in cells pretreated with 20 or 30 µM of PF-429242 compared to the DMSO-treated control (Figure 4.3C). These findings clearly demonstrate the antiviral activity of PF-429242 against DENV-2 in Huh-7.5.1 cells: pretreatment of Huh-7.5.1 cells with PF-429242 results in a dose-dependent inhibition of DENV infection with an EC50 of 1.15 µM and a ~3-log decrease in DENV-2 titer with 20 mM of PF-429242.  4.3.4 Pretreatment of Huh-7.5.1 cells with PF-429242 results in a robust decrease in intracellular DENV-2 RNA  To further dissect the steps in the DENV lifecycle impacted by PF-429242 inhibition of host cell SKI-1/S1P, we examined the relative levels of intracellular DENV-2 RNA in primary and secondary infected cells.  First, Huh-7.5.1 cells were pretreated with different concentrations of PF-429242, 20 µM AcPF-429242 (inactive analog; Figure A2.3A), or DMSO (control) for 24 hours before being infected with DENV-2.  Under our experimental conditions, neither DMSO nor AcPF-429242 was toxic to Huh-7.5.1 cells (Figure A2.3B).  At 48 hpi, DENV genomic RNA was isolated and relatively quantified in cell extracts using qRT-PCR.  DENV-2 RNA levels were normalized to β-actin transcript levels.  In agreement with our previous findings on the relative intracellular abundance of DENV NS1 protein (Figure 4.3) in PF-429242-treated cells, we found that intracellular levels of DENV-2 RNA were also markedly decreased by an average of 74% in the 20 µM PF-429242-treated cells  	   91	     	   92	  Figure 4.3. Inhibition of SKI-1/S1P using PF-429242 results in a dose-dependent decrease in intracellular DENV-2 NS1 expression and a 3-log decrease in extracellular viral titer.  (A-B) Huh-7.5.1 cells were either non-treated (untreated) or treated with increasing concentrations of PF-429242 for 24 hours. The inhibitor was removed and the cells were then infected with DENV-2 (MOI 0.01) for 48 hours. (A) Cell lysates were probed for DENV NS1 (green) and normalized to β-tubulin (red). Representative Western blot for the effect of PF-429242 on DENV NS1 protein level is shown. (B) Dose response curve of normalized, averaged NS1 signal quantified from Western blots of three independent experiments (EC50 = 1.15 ± 0.23 µM). (C) Huh-7.5.1 cells were treated with increasing concentrations of PF-429242 or with 0.03% DMSO (control) for 24 hours. The inhibitor was removed and the cells were then infected with DENV-2 (MOI 0.01). At 48 hours post-infection, the supernatant was collected and viral titer was determined by infecting naïve Vero E6 cells and counting plaques 5 days post-infection. Results (mean ± SEM) from three independent experiments are shown. Statistical significance was calculated for PF-429242-treated cells compared to control with a ratio paired student’s t-test (C) (**, p < 0.01).     	   93	   compared to the DMSO-treated controls and 20 µM AcPF-429242-treated cells (Figure 4.4A).  Next, to evaluate the effect of PF-429242 on DENV-2 infectious virus particle production and its spread to naïve cells, we performed an assay involving DENV-2 secondary infection.  Huh-7.5.1 cells were pretreated with PF-429242 for 24 hours prior to DENV-2 infection.  At 48 hpi, supernatants were collected and incubated with naïve Huh-7.5.1 cells for one hour. Following incubation, the supernatants were removed and the cells were supplemented with fresh complete media for 48 hours before DENV-2 RNA was quantified by qRT-PCR.  Consistent with our results observed during primary infection of Huh-7.5.1 cells with DENV-2, we measured a statistically significant reduction in DENV-2 RNA during secondary infection following initial PF-429242 treatment, compared to DMSO or AcPF-429242 (Figure 4.4B).  These results suggest that pretreatment of Huh-7.5.1 cells with PF-429242 may impair viral replication (Figure 4.4), and it may also compromise viral protein biosynthesis (Figure 4.3A-B) and production of infectious DENV-2 virus particles (Figure 4.3C) in Huh-7.5.1 cells with low cytosolic LD abundance.  4.3.5 Post-treatment of DENV infected Huh-7.5.1 cells with PF-429242 does not affect viral RNA synthesis, but it does impair assembly and/or release of infectious virus particles   We next examined whether PF-429242 can impair DENV-2 RNA synthesis when added to Huh-7.5.1 cells with already established infection.  To achieve this goal, Huh-7.5.1 cells were first infected with DENV-2 for 24 hours to allow uninterrupted DENV replication and establishment of infection.  At the end of the infection period, the Huh-7.5.1 cells were treated with different concentrations of PF-429242, AcPF-429242, or DMSO for 48 hours before DENV-2 RNA levels were quantified by qRT-PCR.  Interestingly, in contrast to our finding with the pretreatment of Huh-7.5.1 cells with PF-429242 (Figure 4.4A-B), post-treatment of DENV-2- infected Huh-7.5.1 cells with PF-429242 had no effect on the level of intracellular DENV-2 RNA (Figure 4.4C).  	   94	     	   95	  Figure 4.4. Inhibition of SKI-1/S1P using PF-429242 results in a significant decrease of DENV-2 viral RNA pre- and post-establishment of DENV infection in Huh-7.5.1 cells.  (A-B) Huh-7.5.1 cells were treated either with 20 µM AcPF-429242, with various concentrations of PF-429242 (10-30 µM) or DMSO (0.01-0.03%) (control) for 24 hours. The inhibitor was removed and the cells were then infected with DENV-2 (MOI 0.01) for 48 hours. (A) Total RNA was harvested and DENV-2 RNA levels, normalized to β-actin transcript levels, were relatively quantified in cell extracts using qRT-PCR. (B) Collected supernatant was cultured with naïve Huh-7.5.1 cells for 48 hours, and DENV-2 RNA levels were quantified. (C-E) Huh-7.5.1 cells were infected with DENV-2 (MOI 0.01). At 24 hours post-infection, cells were treated either with 0.03% DMSO (control), 20 µM AcPF-429242, or 20/30 µM PF-429242 for 48 hours. Total intracellular RNA during primary infection (C) and secondary infection (D), and extracellular RNA during primary infection (E) were harvested and analyzed for DENV RNA levels.  Intracellular DENV-2 RNA levels (C, D) were normalized to β-actin transcript levels, while extracellular DENV-2 RNA levels (E) were normalized by volume and then relatively quantified using qRT-PCR.  Values represent mean ± SEM of three independent experiments. Statistical significance was calculated compared to control with a one-way ANOVA with a Bonferroni’s post-test  (**, p < 0.01; ***, p < 0.005).     	   96	   To investigate the potential effects of PF-429242 on the production of infectious DENV-2 virus particles when applied after DENV-2 infection, we examined the intracellular levels of DENV-2 during secondary infection of Huh-7.5.1 naïve cells.  We found that DENV-2 RNA was reduced approximately 50% during secondary infection, when naïve Huh-7.5.1 cells were treated with the supernatants from cells that had first been infected for 24 hours, then treated with PF-429242 for 48 hours, compared to supernatants from cells that had first been infected for 24 hours, and then treated with AcPF-429242 or DMSO (Figure 4.4D).   To examine further the potential effects of PF-429242 on the production of extracellular infectious virus particles under these experimental conditions, the supernatants from primary infected cells were analyzed for the presence of extracellular viral RNA.  This analysis revealed that extracellular DENV-2 RNA is decreased by more than 50% from cells treated with PF-429242 after established infection compared to cells treated with DMSO and AcPF-429242 (Figure 4.4E).  While this analysis does not rule out possible effects of PF-429242 on DENV virus assembly, the correlation observed between the impaired secondary infection (Figure 4.4D) and reduction of extracellular DENV RNA from the primary infection (Figure 4.4E) suggests that production of extracellular infectious virus particles may be compromised in PF-429242-treated cells.  Importantly, pharmacological treatment of already infected DENV-2 cells using PF-429242 resulted in a 50% reduction of DENV extracellular RNA.  Taken together, these results indicate that inhibiting SKI-1/S1P can interrupt the DENV lifecycle at multiple stages of viral infection, both preventing naïve cells from becoming infected and preventing the assembly and/or release of infectious virus particles from already infected cell populations.  4.3.6 Extracellularly applied oleic acid, an inducer of lipid droplet formation, rescues DENV-2 RNA synthesis in PF-429242-treated Huh-7.5.1 cells    Inhibition of DENV infectivity by PF-429242 suggested that active lipid metabolism in the host cell is important for the viral lifecycle.  To determine whether the availability of 	   97	  intracellular fatty acids, and specifically their accumulation in cytosolic LDs, was a limiting factor for DENV-2 infection, Huh-7.5.1 cells treated with PF-429242 were supplemented with an exogenously added fatty acid (oleic acid) to induce LD formation during DENV-2 infection (322). DENV-2 RNA levels were quantified by qRT-PCR at 24 hpi.  As expected, DENV-2 RNA was almost fully rescued in cells treated with 10 µM PF-429242 and oleic acid/BSA, compared to cells treated with either oleic acid/BSA alone or PF-429242 alone (Figure 4.5A).  Reduction of the abundance of LDs in cells treated with PF-429242 and the dramatic increase in LD abundance in PF-429242-treated cells with supplemented oleic acid was quantified using fluorescence intensity of stained cells with Nile red (Figure 4.5B and C).   These results demonstrate that the inhibitory effect of PF-429242 on DENV-2 infection can be attributed to its intrinsic capacity to significantly reduce total intracellular lipid levels, specifically triglycerides and cholesterol esters (37), two major constituents of cellular LDs (4).  These findings were confirmed by using Nile red, a dye selective for neutral lipids such as triglycerides and cholesterol esters that make up the core of LDs (317, 325).  Importantly, our results reveal human SKI-1/S1P as a regulator of LD abundance further confirming that LDs are necessary for DENV infection.  4.3.7 Pretreatment of Huh-7.5.1 cells with PF-429242 results in a significant decrease in intracellular viral RNA for all four DENV serotypes   Finally, to establish whether replication with other DENV serotypes can be inhibited by PF-429242 in the same manner as DENV-2, Huh-7.5.1 cells were pretreated with 20 µM of PF-429242, DMSO (control) or AcPF-429242 for 24 hours.  The inhibitor was removed and the cells were then infected with DENV-1, -2, -3, or -4 (MOI 0.01) for 48 hours.  Total cellular RNA was harvested, and DENV-1, -2, -3, and -4 RNA levels, normalized to β-actin transcript levels, were relatively quantified in cell extracts using qRT- PCR.  Results showed that DENV-1 RNA was reduced by 84%, DENV-2 by 74%, DENV-3 by 95%, and DENV-4 by 95% compared to control-treated cells (Figure 4.6).  These results demonstrate that    	   98	     	   99	  Figure 4.5. DENV-2 intracellular viral RNA inhibited by PF-429242 can be rescued with exogenous oleic acid by increasing cytosolic lipid droplet abundance.  (A-C) Huh-7.5.1 cells were treated either with 0.01% DMSO (control) or 10 µM PF-429242 for 24 hours. The inhibitor was removed and the cells were then infected with DENV-2 (A) or mock-infected (B-C) with or without addition of 0.6 mM oleic acid (with BSA in molar ratio 6:1) for 24 hours. (A) Total RNA was harvested and DENV-2 RNA levels, normalized to β-actin transcript levels, were relatively quantified in cell extracts using qRT-PCR. (B) Representative images of the effect of PF-429242 and oleic acid on lipid droplets are shown. Fixed cells were permeabilized with Triton X-100 and stained for cell nuclei using Hoechst dye and for lipid droplets using Nile red. Images were obtained using a Leica SP8 confocal microscope with a 100x objective. (C) Abundance of lipid droplets was quantified by measuring the lipid droplet-positive area (µm2)/cell and the average number of lipid droplets/cell based on Nile red fluorescence using Fiji software.  Statistical significance was calculated with a two-way ANOVA with a Bonferroni’s post-test (*, p < 0.05; **, p < 0.01; ***, p < 0.005; ****, p < 0.001). Values represent mean ± SEM of three independent experiments.    	   100	     Figure 4.6. Inhibition of SKI-1/S1P using PF-429242 results in a robust decrease in intracellular viral RNA in Huh-7.5.1 cells for all four DENV serotypes.  Huh-7.5.1 cells were treated either with 0.02% DMSO (control), 20 µM AcPF-429242, or 20 µM PF-429242 for 24 hours. The inhibitor was removed and the cells were then infected with one of the four DENV serotypes (MOI 0.01) for 48 hours. Total RNA was harvested and DENV RNA levels of four serotypes, normalized to β-actin transcript levels, were relatively quantified in cell extracts using qRT-PCR. Values represent mean ± SEM of three independent experiments. Statistical analysis was performed for PF-429242 and AcPF-429242 treated cells compared to control with a two-way ANOVA with a Bonferroni’s post-test (**, p < 0.01; ***, p < 0.005).    	   101	  inhibiting the enzymatic activity of human subtilase SKI-1/S1P using PF-429242 dramatically reduces viral replication of all four DENV serotypes in Huh-7.5.1 cells.  4.4 Discussion  In this study, using PF-429242, an active-site-directed inhibitor of SKI-1/S1P, we demonstrated that manipulation of human SKI-1/S1P enzymatic activity in Huh-7.5.1 human hepatoma cells provides a means of interfering with the SKI-1/S1P-mediated proteolytic activation of the SREBP pathway and results in PF-429242-treated cells with a robust reduction of the abundance of cytosolic LDs, an important organelle hijacked during the DENV lifecycle (244). Collectively, our results reveal human SKI-1/S1P as a potential target for indirect-acting pan-serotypic anti-dengue virus agents.  4.4.1 Human subtilase SKI-1/S1P is a regulator of lipid droplet formation in Huh-7.5.1 human hepatoma cells   As hypothesized, inhibiting the SKI-1/S1P dependent activation of the SREBP pathway and expression of SREBP-activated genes resulted in a robust decrease in cytoplasmic LD formation in PF-429242-treated cells.  Our results demonstrated that pharmacological inhibition of the subtilase SKI-1/S1P by PF-429242 represents a powerful approach to regulating the LD abundance and size in Huh-7.5.1 cells.  Interestingly, identification of human SKI-1/S1P as a regulator of LD formation is consistent with the identification of the SREBP and SREBP cleavage activating protein (SCAP) as key genes involved in LD formation using a genome-wide RNA interference screen in Drosophila S2 cells (329).  Collectively, these observations suggest SKI-1/S1P inhibitors could serve as LD modulating agents.   	   102	  4.4.2 Manipulation of human SKI-1/S1P enzymatic activity provides a mean of effectively inhibiting viral infection of Huh-7.5.1 cells by DENV-2  Lipids and cholesterol have previously been reported as playing important roles in the DENV lifecycle (243, 244, 247, 312, 314).  The viral NS3 serine protease was implicated in recruiting fatty acid synthase to the replicase complex, leading to increased cellular fatty acid synthesis (243).  Additionally, LDs were found to be necessary during the viral assembly steps, as the capsid protein of DENV is recruited to LDs and this step is crucial for forming infectious viral particles (244).  Furthermore, degradation of LDs by autophagy for ATP generation is necessary to support viral replication (247).  Cholesterol in the viral envelope was proposed as being important in the fusion process for DENV entry, while intracellular cholesterol transport is necessary both for viral entry and replication (312, 314).  However, although the collective significance of lipids and cholesterol in most steps of the DENV lifecycle has been recognized, it is still not clear whether the targeting of major metabolic lipid and cholesterol pathways, such as the SREBP pathway, is a viable option for developing host-directed therapeutics for DENV infection. In this study, we demonstrated that inhibition of the SKI-1/S1P-dependent activation of the SREBP pathway and SREBP-activated genes using PF-429242 resulted in a dose-dependent decrease in DENV-2 infection in Huh-7.5.1 cells (Figure 4.3). Importantly, the antiviral effect of PF-429242 was associated with a decrease in cytosolic LD abundance.  The anti-DENV activity of PF-429242 was characterized by strong reductions in DENV NS1 protein abundance, reduction in intracellular and extracellular DENV RNA, and infectious virions (Figures 4.3 and 4.4).  Of note, the Jean lab reported previously that PF-429242 treatment of HCV-infected Huh-7.5.1 cells inhibited HCV infection with an EC50 of 6.4 µM (37).  Here, the same treatment led to a reduction in DENV-2 infection with an EC50 of 1.15 µM, showing that inhibition of SKI-1/S1P has a more potent antiviral effect against DENV-2 than HCV. Using a lipid complementation assay, we demonstrated that the antiviral effects of PF-429242 can be almost fully blocked by adding exogenous oleic acid to DENV-infected cells (Figure 4.5A).  We demonstrated that cytosolic LD abundance is increased in response to oleic acid in untreated and PF-429242-treated Huh-7.5.1 cells (Figure 4.5B and C). These findings are consistent with the prior work of Rohwedder et al. using lipidomic analysis of 	   103	  oleic acid-treated Huh-7 cells that showed an increase in triglyceride and cholesterol ester content in response to oleate stimulation (330).  In addition, the Jean lab previously reported that PF-429242 treatment of Huh-7.5.1 cells reduced cholesterol ester levels by 63% and total intracellular triglycerides by 51%, two major constituents of cellular LDs (37).  Collectively, these observations indicate that the anti-dengue activity of PF-429242 is associated with the inhibition of SKI-1/S1P-mediated activation of the SREBP transcriptional network, a novel and important host pathway for LD formation and the DENV lifecycle. In addition, since the SREBP pathway regulates cholesterol biosynthesis and cholesterol plays a significant role in the DENV lifecycle (26, 311, 312), inhibition of SKI-1/S1P by PF-429242 could interfere with viral infection by depleting intracellular cholesterol levels.  This is best exemplified by the recent reports of Petersen et al. (331) and Kleinfelter et al. (332) that demonstrated the critical roles of the SREBP cholesterol regulatory pathway in the viral lifecycle of hantaviruses using the SKI-1/S1P inhibitor, PF-429242. Finally, our observation that post-treatment of DENV-2-infected Huh-7.5.1 cells with PF-429242 does not affect intracellular viral RNA synthesis indicated that DENV replication was not compromised; however, extracellular viral RNA and viral RNA levels during secondary infection were still reduced (Figure 4.4E and D, respectively), suggesting that the production and/or release of infectious virus particles had been compromised in the virally infected cells. This also suggests that existing intracellular fatty acid and cholesterol levels were sufficient to support the establishment of the primary infection but that the post-replication steps (e.g., virion assembly, maturation, release, or re-entry into naïve cells) were compromised.  Since reduction in extracellular viral RNA and intracellular viral RNA during secondary infection was comparable relative to control treated cells, re-entry into naïve cells was not the factor that was compromised by PF-429242 treatment. These findings are in agreement with a study by Samsa et al. reporting that DENV assembles capsids on LDs (244).  Accordingly, SKI-1/S1P inhibition by PF-429242 in Huh-7.5.1 cells, which leads to smaller/fewer LDs, would impair assembly and release of infectious virus particles.   Our results are in partial agreement with those of Uchida et al., who recently reported suppressive effects of PF-429242 on DENV propagation in HeLa cells, which are derived from cervical cancer cells (333).  Whereas the first part of the work by Uchida and collaborators agrees with my findings on the pan-serotypic inhibitory effect of PF-429242 on 	   104	  DENV infection (Figure 4.6), we reached an opposite conclusion on the mechanism of action of PF-429242. Uchida et al. concluded that the PF-429242-associated depletion of LD and cholesterol in HeLa cells are not direct causes of the virus inhibition. In our results, however, SKI-1/S1P inhibition by PF-429242 in human hepatoma Huh-7.5.1 cells was reversed by exogenous oleic acid, which acts as an inducer of LD formation in PF-429242-treated cells.  These discrepancies may be related to differences in the cell line studied.  We selected human hepatoma Huh-7.5.1 cells as the model system to examine the molecular functions of SKI-1/S1P, a key regulator of the lipid homeostasis/SREBP pathway, in the formation of cellular lipid storage droplets and the DENV lifecycle.  Perhaps cell-intrinsic differences in LD biology between cervical cancer cells and human hepatoma cells could account for the differing results (322).  In addition, the concentration of supplemented oleic acid, the addition of BSA during the oleic acid treatment, and the time of addition of oleic acid during the rescue experiment are other key differences between this study and that of Uchida et al.  Of note, a study by Ricchi et al. demonstrated concentration- and time-dependent effect of extracellularly applied oleic acid to hepatic cell cultures on the accumulation of triglycerides and SREBP-1 activation (334)).   4.4.3 Inhibition of the SKI-1/S1P-mediated proteolytic activation of the SREBP pathway has a pan-serotypic inhibitory effect on DENV infection  Here, we show that infection of all DENV serotypes depends on human SKI-1/S1P enzymatic activity and that its inhibition has a pan-serotypic inhibitory effect on the DENV lifecycle (Figure 4.6).  Infection with any of the DENV serotypes may develop into non-severe dengue fever (DF) or into life-threatening dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS).  It has been observed that DHF and DSS more often result from secondary infection with a heterologous DENV serotype (335).  After infection with one serotype, an individual develops immunity to all four serotypes but only for a short period of time.  The immunity to other serotypes usually wanes over two to three months, leaving life-long immunity established only against the primary infecting serotype (336).  Subsequent exposure to heterologous serotypes increases an individual’s risk of developing DHF/DSS, which is believed to arise from a phenomenon known as antibody-dependent 	   105	  enhancement (ADE) (337, 338).  Since all four DENV serotypes are now circulating in most tropical and subtropical areas of the world and are endemic in over 100 countries, candidates for DENV therapeutic agents need to be able to target all DENV serotypes (218).  4.4.4 SKI-1/S1P is a potential target for indirect-acting antiviral agents against DENV infection  One of the advantages of targeting a cellular enzyme for antiviral therapy is that it dramatically reduces the likelihood of inducing antiviral resistance.  Our study demonstrated that, PF-429242 targets all DENV serotypes with similar efficacy and therefore may be useful as a broad-spectrum antiviral against other viruses that depend on lipid homeostasis/SREBP pathways for their lifecycles.  Although cellular SKI-1/S1P activity plays an important role in preventing failure of lysosomal functions (339), the ER stress response (340), and bone mineralization (341, 342), its short-term inhibition in the case of acute infections by DENV and other LD-dependent microbes may represent a viable therapeutic strategy in DENV-associated disease (343).    	   106	  Chapter 5: Conclusions and future directions  5.1 Discussion  The aim of this work was to identify novel host factors associated with HCV and DENV pathogenesis focusing on the regulators of lipid metabolism.  Viruses modulate host cellular metabolism to establish the physical environment and produce energy needed for their lifecycles.  Several viruses remodel specific lipid microenvironments to facilitate different stages of their lifecycles, including entry, replication and assembly (2).  During chronic HCV infection, retention of lipids in the liver is dramatically increased leading to steatosis (344), while the levels of circulating lipoproteins are reduced inducing a pathology known as hypolipoproteinemia (150).  Steatosis and hypolipoproteinemia have both been linked to increased severity of chronic HCV-induced liver disease.  However, major gaps in our knowledge exist on how viruses, host metabolic pathways and regulators of lipid metabolism contribute to virulence and pathogenesis.  A better understanding of host-pathogen interactions during HCV and DENV infections could reveal cellular targets or pathways that are associated with viral pathogenesis.  We chose to investigate several important regulators of lipid homeostasis, including miRNAs (miR-122, miR-24 and miR-223) and two host proprotein convertases, PCSK9 and SKI-1/S1P, during HCV and DENV infection.  MicroRNAs have emerged as essential post-transcriptional regulators of gene expression.  They are predicted to control over 60% of transcripts (345) and to influence diverse biological processes, including metabolism (346).  We explored the molecular interplay between circulating miRNAs (miR-122, miR-24, and miR-223), known to regulate lipid homeostasis (90, 91, 101, 107) in CHC patients undergoing anti-HCV treatment.  The studies presented herein provide evidence that the amount of human circulating miRNAs (miR-122, miR-24, and miR-223) is affected by CHC infection and achieving a treatment-based cure.   We also investigated the role of circulating PCSK9 during HCV infection. PCSK9 is an important regulator of cholesterol and lipid homeostasis and acts by degrading the HCV co-receptor, LDLR (8).  Additionally, we investigated the role of another member of the PC 	   107	  family during DENV infection, SKI-1/S1P, which regulates processing of transcriptional factors in one of the major lipid metabolic pathway, SREBP (33). Overall, the research presented here has revealed the complex role that lipid regulators play during HCV and DENV infection, as well as potential new targets for development of therapeutics against important viral diseases.   5.1.1 Circulating miRNA levels during HCV infection  In Chapter 2, we investigated circulating levels of three important human miRNAs in CHC-infected patients and how these miRNAs are modulated based on different aspects of HCV infection: levels of HCV RNA, liver transaminases, liver fibrosis score, blood cell count and antiviral treatment outcome.    Our results showed a decrease in circulating miR-122 and increases in miR-24 and miR-223 after treatment compared to baseline in SVR (Figure 2.3, Table 2.3).  While miR-122 is a liver-specific miRNA and is one of the most abundant miRNAs in the liver, accounting for about 70% of the whole hepatic miRnome in the adult human (268), miR-24 and miR-223 are expressed in other cell types besides hepatocytes, including T cells and cells of the myeloid cell lineage, respectively (347).  It is not surprising therefore that the observed decrease in circulating miR-122 in patients who achieved SVR also correlates with the levels of liver transaminases (Table 2.2, Table A1.2), which confirms previous studies (348-351).  Therefore, the secretion of miR-122 in circulation during chronic HCV infection may occur passively, as by-products of dead cells, and may directly reflect hepatic necroinflammation levels rather than result from active secretion via microvesicles.  Thus, miR-122 may serve as a good marker of liver inflammation and potentially fibrosis.    MiR-122 has been previously investigated as a potential marker of liver inflammation and fibrosis (348).  Recently, Matsuura et al. studied 130 CHC patients prospectively followed over several decades and identified miR-122 as a candidate predictor of liver disease progression.  Cross-sectional analyses at the time of initial liver biopsy showed that miR-122 levels in plasma were positively correlated with inflammatory activity, but not fibrosis.  However, a limitation of the study by Matsuura et al. is the low number of patients, especially those with significant hepatic fibrosis (348).  	   108	   The mode of secretion of circulating miRNAs in plasma/serum is controversial. Gallo et al. indicated that circulating miRNAs in plasma/serum are predominantly concentrated in exosomes (352).  Arroyo et al. however reported that up to 90% of circulating miRNAs are associated with proteins and showed that some miRNAs are more likely to be enriched within specific extracellular complexes (eg. circulating miR-122 was detected in the protein fractions rather than in the vesicles in healthy individuals) (76).  Furthermore, another study has suggested that the distributions of miR-122 and miR-155 between exosomes or protein complexes may differ depending on the underlying pathological condition as demonstrated in mice with alcoholic liver disease, inflammatory liver damage or drug induced liver disease (353). Overall, these studies indicate that the appropriate extracellular complex for measuring miRNA levels depends on the underlying pathology as well as the specific miRNA being measured (348).    Unlike miR-122, circulating miR-24 and miR-223 were shown to be the most abundant miRNAs within lipoprotein complexes (77, 83).  Also, it has been previously established that HCV patients who achieve SVR have increased LDL and cholesterol from baseline compared to non-SVR patients (256, 283).  This indicates that upregulated levels of circulating miR-24 and miR-223 in SVR patients observed herein could be directly associated with an upregulation in lipoproteins levels that serve as carriers of these miRNAs.  Yet, this interpretation should be considered with caution as the study on the extent HDL-bound miRNAs contribute to the total pool of circulating miRNAs showed that HDL-bound miR-223 contributes to ~8% of the total plasma pool (83). Also, in the same study, miR-223 level varied only in the plasma fraction independently of the concentrations in HDL-bound fraction (83).  However, HDL-bound miR-223 contribution to the total plasma pool was measured in healthy individuals only and may not reflect the level of HDL-bound miR-223 in patients with different pathologies, including the CHC-patients who achieved SVR.   We also observed herein an excellent correlation between miR-24 and miR-223 levels (rhos = 0.91, p-value < 0.0001) measured in all CHC-patients (Figure 2.2).  This correlation could be accidental or due to a common mechanistic pathway and/or target genes between the two miRNAs, such as co-regulation of lipid metabolism.  Recently, both miRNAs were reported to be involved in metabolic diseases.  Vickers et al. identified that miR-223 and miR-24 are the most regulated miRNAs in atherosclerosis with >3,000-fold and 65-fold 	   109	  increase, respectively, in human familial hypercholesterolemia (77).  Also, miR-24 was shown to enhance processing of SREBPs by targeting INSIG1 and accumulating in livers of high-fat diet-treated mice and in human hepatocytes treated with fatty acids (101).  Alongside, miR-223 promoter activity and mature levels were found to be sensitive to intracellular cholesterol changes and that miR-223 regulates cholesterol biosynthesis, uptake, and efflux, thus establishing its critical role in post-transcriptional regulation of cholesterol metabolism (107).  Whether upregulation in miR-223 and miR-24 levels in SVR patients after treatment and their excellent correlation with each other is due to their role in regulation of lipid metabolism and what is the exact biological link with the HCV infection will require further study.  Overall, clearance of HCV with use of an IFN-based therapy results in rapid changes in the levels of circulating miRNAs known as regulators of lipid metabolism including miR-122, miR-24, and miR-223.  This implicates an effect of HCV infection on lipid homeostasis via modulation of regulatory markers and supports further investigation of lipid pathways and metabolites as predictors of HCV-associated progressive liver disease and treatment outcome.   5.1.2 PCSK9 and HCV  In Chapter 3, the role of circulating PCSK9 in HCV infection was explored. PCSK9 is the last identified member of the PC family and regulates the complex regulatory pathway of host cholesterol metabolism.  The extracellular functionality of this protein and its control of HCV entry receptors CD81 and LDLR highlight its importance for HCV infection by reducing its ability to penetrate liver cells (292, 354, 355).  The hypothesis that PCSK9 plays a protective role in HCV infection is also supported by evidence that VLDLR, an independent receptor of HCV, is also targeted by PCSK9 (356).  In this study, we observed an upregulation of circulating PCSK9 level in SVR patients after the end of treatment compared to baseline levels (Figure 3.2), which further implicates the link between PCSK9 and HCV.  In agreement, a recent study has shown that higher PCSK9 and LDL-C plasma concentrations correlate with reduced HCV titers in patients chronically infected with HCV genotype 3 (289).  There are many reasons why 	   110	  PCSK9 could be dysregulated in SVR patients, one being that the virus itself is suppressing PCSK9, and following successful treatment, PCSK9 bounces back.  The observation though may also be a result of other metabolic changes that occur/return to normal following SVR.  Dysregulation of circulating PCSK9 in SVR patients combined with the knowledge that PCSK9 controls expression of multiple HCV receptors prompted us to investigate the hypothesis that PCSK9 may have a direct impact on HCV infection.  Pretreatment of Huh-7.5.1 cells with 25 µg/ml PCSK9 showed complete inhibition of HCV infection (98.6% or 42-fold) (Figure 3.3), which confirms results from a previous study (292). Additionally, a gain-of-function PCSK9 mutant was observed to be much more efficient at inhibiting HCV than wild-type PCSK9.  In contrast, a loss-of-function PCSK9 variant was determined to have no effect on HCV infectivity (Figure 3.3).  These results have important implications as a variety of PCSK9 mutations found within the human population may have significant impact on the levels of liver LDLR and may influence host susceptibility to HCV infection and treatment outcomes.  This observation is particularly important as monoclonal antibodies (mAbs) targeting PCSK9 are now administered in patients with cardiovascular diseases (357, 358).  It is possible that the use of PCSK9 inhibitors may be hazardous in HCV-infected patients as they could increase LDLR mediated LVP uptake and enhance the severity of infection in HCV (289).  Therefore, the effect of mAbs targeting PCSK9 on HCV infection and HCV-associated liver disease progression in patients should be further investigated.  Another interesting observation in this study was the correlation between circulating PCSK9 and miR-24 in CHC-patients (Table 3.2).  This correlation may suggest a first biological link between PCSK9 and miR-24.  It is predicted that PCSK9 is one of the miR-24 targets though only with one binding site in its 3'-UTR (359). However, no studies previously validated miR-24 binding to PCSK9, other than a small pilot study on miR-24 targeting PCSK9 investigated by a former Jean laboratory member, Emma-Kate Loveday.  In this study, it was shown that PCSK9 mRNA expression levels were significantly reduced at the highest miR-24 concentration (45nM) in transfected HeLa cells at 48 and at 72 hours post-transfection (359).  It is therefore hypothesized that miR-24 regulation of PCSK9 may play an important role in lipid metabolism and the lifecycles of viruses that depend on lipids as well as LDLR and/or LDLR related receptors for infection.  Examples include HCV, DENV, rhinovirus, noroviruses and vesicular stomatitis virus (356). 	   111	  5.1.3 SKI-1/S1P and DENV  In Chapter 4, SKI-1/S1P was explored as a potential target for inhibiting DENV infection.  A suppressive effect of the drug PF-429242 was shown here in four serotypes of DENV1-4 propagation in cultured Huh-7.5.1 cells (Figure 4.3, Figure 4.4 and Figure 4.6). Importantly, the antiviral effect of PF-429242 was associated with a decrease in cytosolic LD abundance (Figure 4.1D), while using a lipid complementation assay, the antiviral effects of PF-429242 was almost fully blocked (Figure 4.5).   The anti-viral effects of PF-429242 have been previously reported for other viruses.  As in the case of DENV infection, PF-429242 targeted inhibition of SKI-1/S1P blocks HCV propagation via reduction of intracellular LD abundance (37).  Reduction of LD abundance showed mislocalization of the HCV core protein to cell nuclei, thus impeding HCV assembly and secretion (37).  On the other hand, in arenavirus infection, PF-429242 by targeting SKI-1/S1P-mediated cleavage of arenavirus glycoprotein directly inhibits virus propagation (320, 360).   Another interesting observation is that intracellular LD levels were decreased in DENV-infected cells compared with mock-infected cells already at 24 and 48 hpi (Figure 4.2).  A LD is an organelle mainly consisted of triacylglycerol and cholesterol ester, which serves as an energy source when its lipid content is depleted (361).  The relationships between LDs and viral replication have been reported for HCV (37, 182, 326), DENV (244), rotavirus (362), and orthoreovirus (363).  These studies indicated that LDs are crucial in multiple steps of viral infection.  In Vogt et al. because of the observed LD degradation in HCV replicon study suggested that LDs could be consumed during HCV replication as a potential source for membranes or energy (364).  Thus, LD levels might be decreased in DENV-infected cells also as a source of energy, which is required for virus replication.  Treatment of Huh-7.5.1 cells with PF-429242 also leads to reduction in LD abundance (Figure 4.1).  Hence, pretreatment of cells with PF-429242 prior to DENV infection may be detrimental for the virus as cells become depleted of the needed lipids in LDs prior to establishment of viral infection that otherwise would serve as an energy source for DENV propagation.  Targeting SKI-1/S1P during DENV infection may not be the safest option as it potentially could lead to host-side effects as SKI-1/S1P is necessary in preventing aberration 	   112	  of lysosomal function (339), the ER stress response (340), and bone mineralization (341, 342).  However, short-term impact of SKI-1/S1P inhibition is still not clear.  Additionally, since inhibition of DENV infection with PF-429242 most likely occurs by reduction of LDs contents via inhibition of the SREBP pathway, inhibiting other targets involved in LD biogenesis, fatty acid biosynthesis and the SREBP pathway activation should also be considered for inhibition of DENV infection.  Several studies previously reported that targeting host lipid metabolism could inhibit viral infection.  Samsa et al. previously showed that a small molecule C75, inhibitor of fatty acid synthase, is able to reduce DENV infection (244).  Also, several studies have demonstrated that statins were able to inhibit HCV replication in vitro (365, 366).  A specific inhibitor of DGAT1, an enzyme required for LD formation, was also reported to block HCV virion assembly and release (181).   5.2 Future directions: further dissecting the roles and applications of regulators of lipid metabolism in viral pathogenesis  The studies presented in Chapter 2, 3 and 4 provide novel findings on different regulators of lipid metabolism in HCV and DENV infections, including potential disease biomarkers as well as potential host targets for blocking viral infection.  These studies however open a wide range of questions that need to be explored to determine further the interplay between host factors mediating lipid metabolism and viral infections.  5.2.1 Investigate the effect of miR-24 and miR-223 on HCV infection  In the data presented in Chapter 2, we observed a significant upregulation in circulating miR-24 and miR-223 levels after treatment compared to baseline levels in CHC-infected patients who achieved SVR but not in patients who relapsed (Figure 2.3, Table 2.3).  Whether these upregulations are a direct or indirect effect of viral clearance or related to effects on host cellular pathways remains to be determined.   To investigate the role of miR-24 and miR-223 in HCV infection, we first propose using an HCV infectious clone to infect human hepatoma Huh-7 cells and measure changes 	   113	  in intracellular and extracellular miR-24 and miR-223 levels at different timepoints of infection.  This experiment may reveal whether HCV infection has a direct impact on dysregulation of the two miRNA levels.  Preliminary experiments (Figure A3.1) performed in collaboration with the Houghton lab from the University of Alberta showed no changes in intracellular miR-24 or miR-223 expression at 72 hours post-infection with HCV JC1 infectious clone (a genotype 2a/2a	  chimera between J6 and JFH-1 infectious clones (367)).  Previously, a comprehensive microarray analysis was performed using human hepatoma Huh-7.5.1 cells to measure dynamic host miRNAs expression alteration during in vitro acute HCV infection (368).  In the study, miR-24 was down-regulated post-HCV infection at day 3 and 4 post-infection.  Further analysis on the effect of transfected miR-24 mimic in HCV replicon cell line (genotype 1b) showed about two-fold decrease in the intracellular HCV RNA abundance.  These results suggest that systematic administration of miR-24 might not only decrease the virus replication level but also inhibit hepatic fibrosis, since TGF-β is one of the potential targets of miR-24 (368).  Therefore, upregulated circulating levels of miR-24 upon HCV cure in patients who achieved SVR may reflect the decrease in HCV RNA abundance and indicate biological processes happening to a host after viral cure, such as inhibition of hepatic fibrosis and/or changes to metabolic pathways that are disrupted by viral infection.   However, drawbacks of the existing HCV cell culture systems are that they may not reflect natural HCV infection.  In vitro culture involves a cancer-derived cell line, requires the uncommon HCV genotype 2a and does not create a chronic viral infection that is normally established by HCV in humans.  For example, in the study described above (368), there was a lack of concordance of the effects of miRNAs between the replicon system and the JFH-1-derived cell culture-grown HCV (HCVcc), as no changes in RNA abundance were observed after transfection of miR-24 mimic into HCVcc system.  Ideally, to determine the effect of chronic HCV infection on intracellular levels of miR-24 and miR-223, samples of liver biopsies from patients at different stages of CHC infection should be analyzed before and after SVR was achieved.  To further reduce inter-patients variability, miRNA analysis should be performed on liver biopsies collected from the same patient at different timepoints of HCV infection.   	   114	   Furthermore, unlike with miR-24, no studies previously attempted to investigate the role of miR-223 in HCV cell culture system.  Therefore, overexpression of miR-223 mimic in Huh-7 cells following HCV infection potentially could also inhibit HCV infection and could help uncover the role of this miRNA in the HCV lifecycle.  Utilizing similar techniques described in Chapter 3 and 4 to measure HCV RNA and proteins levels could be further studied the roles of this miRNA during different stages of the virus lifecycle.   In the data presented in Chapter 2, circulating miR-24 and miR-223 levels were measured from total RNA isolated from patients’ plasma samples.  It is still unknown what the mode of secretion is for these miRNAs.  Knowing this information could elucidate the mechanism of miRNA modulation during HCV infection.  To investigate the mode of secretion of these miRNAs in plasma, the same samples as the ones used in Chapter 2 could be processed for isolation of lipoproteins, exosomes and Ago-bound complexes independently by sucrose gradient ultracentrifugation followed by immunoprecipitation with antibodies targeting different vehicles of miRNA transmission.  The amount of miR-24 and miR-223 bound to lipoproteins and Ago complexes, or found inside exosomes, could be determined by qRT-PCR method as previously described.   Successful anti-HCV treatment is normally associated with changes to cholesterol and beta-lipoprotein levels (207).  To determine whether upregulation in circulating miR-24 and miR-223 correlates with metabolic changes, profiling of metabolites could be measured in plasma samples collected before and after anti-HCV treatment.  Changes in lipid metabolites in CHC-patients who achieved SVR could be further mapped to lipid pathways using online bioinformatics tools such as Lipid Metabolites and Pathways Strategy (LIPID Maps® http://www.lipidmaps.org/).  Common pathways between the changed metabolites and targets of miR-24 and miR-223 could be then investigated to determine the link between the miRNAs and lipid metabolism.  This proposed study could reveal the mechanisms of HCV-associated metabolic disorders and identify predictors of progressive liver disease and treatment outcome.  A better understanding of key regulators of lipid metabolism modulated during HCV infection and involved in the development of advanced liver pathologies could also help uncover new therapeutic targets for the drug development against liver disease progression and extra-hepatic metabolic manifestations.   	   115	  5.2.2 Investigating the link between miR-24 and PCSK9  In Chapter 3, our correlation data for the first time demonstrated a link between miR-24 and PCSK9 in vivo.  Correlation analysis using Spearman’s correlation test between PCSK9 and multiple biochemical parameters showed that PCSK9 levels correlated only with normalized circulating miR-24 levels [Δct(miR-24), rhos = -0.24, p-value = 0.0174] (Table 3.2).  Further validation of this association is necessary to understand the interplay between miR-24 and PCSK9.  An unpublished preliminary report by a former Jean lab member showed that PCSK9 may be a target of miR-24 in cell culture because miR-24 overexpression led to reduction of PCSK9 mRNA expression levels (359).  Even though the experiment showed that there is a link between miR-24 and PCSK9 levels, additional experiments are needed to validate PCSK9 as the direct target of miR-24.  One experiment would be to test miR-24 regulation of PCSK9 mRNA expression in a human hepatoma cell line (eg. Huh-7 cells and/or its clonal descendants) as this cell line is derived from hepatocytes, cells normally expressing high levels of both PCSK9 and miR-24 (8, 101).  A luciferase reporter assay containing the PCSK9 3'-UTR can further confirm the ability of miR-24 to specifically target PCSK9.  Also, miR-24 targeting of PCSK9 can be investigated by co-immunoprecipitation of miRNA:mRNA complex by the pull-down of the RISC components or Ago (369).  Target mRNAs, in this case PCSK9, undergoing direct regulation are co-immunoprecipitated along with RISC (or Ago) can be validated by qRT-PCR or deep sequencing.  Furthermore, overexpression/inhibition of miR-24 should also be investigated on the protein levels of PCSK9 as well as its known targets LDLR, VLDLR and CD81 (356).   In addition to validating PCSK9 as a target of miR-24, the interplay between the two should be further investigated during HCV infection.  Increased levels of PCSK9 result in turnover of the LDLR and reduced CD81 on the surface of hepatocytes, both of which block HCV entry into the cells (292). Therefore, inhibition of cellular miR-24 activity or expression hypothetically should lead to upregulation of PCSK9 followed by reduction in the LDLR and CD81 and thus decrease in HCV infection.  Inhibition of miR-24 activity can be achieved either with one of the commercially available inhibitors (eg. miRCURY LNA™ microRNA inhibitor) or reduction of miR-24 expression can be done with CRISPR/cas9 genome editing tool (370).  Surprisingly, in our current study we found that circulating miR-24 directly correlates 	   116	  with PCSK9, which is the opposite relationship that should be detected if PCSK9 is the direct target of miR-24.  However, in our study we measured circulating levels of both miR-24 and PCSK9, which may not be indicative of the expression levels of miR-24 and PCSK9 inside the cells.  Therefore, the circulating plasma miR-24 and PCSK9 levels should be measured in parallel with the intracellular miR-24 and PCSK9 from liver biopsies in patients at different stages of CHC infection.   5.2.3 DENV hijacking lipid metabolic pathways  In Chapter 4, we reported that DENV depends on the SKI-1/S1P-mediated activation of the SREBP pathway and LDs during infection. It is still unclear what is the mechanism of DENV modulation of lipid metabolism.   To investigate the mechanism of DENV modulation of lipid metabolism, we propose to investigate the effect of DENV infection on the SREBP pathway and associated SREBP transcriptional regulatory network. In the preliminary experiments on the SKI-1/S1P-mediated proteolytic activation of the SREBP pathway during the DENV-2 lifecycle, we profiled expression of seven SREBP-associated cellular genes during all stages of DENV-2 infection in human hepatoma (Huh-7.5.1) cells.   We initially focused on the two key host cell proteases involved in the proteolytic activation of the SREBP pathways, S1P and S2P (Figure A3.2A).  We examined the mRNA expression level of S1P and S2P in DENV-2-infected and mock-infected cells at various timepoints post-infection.  Our quantitative RT-PCR data showed that the SKI-1/S1P mRNA expression level in DENV-2-infected cells at 48 and 72 hpi was significantly increased over time-matched mock-infected controls (Figure A3.2A). In contrast, the expression level of S2P mRNA showed no increase between DENV-infected cells and their time-matched mock-infected controls (Figure A3.2A).   We next examined the mRNA expression levels of two of the SREBP transcription factor substrates, SREBP-1c and SREBP-2, during DENV-2 infection. As previously observed with SKI-1/S1P mRNA, the expression level of each SREBP-1c and SREBP-2 mRNA was significantly increased at 48 and 72 hpi in DENV-2-infected cells compared to time-matched mock-infected controls (Figure A3.2B).  	   117	   We subsequently determined the mRNA expression levels of two SREBP target genes, LDLR and PCSK9. LDLR expression was significantly increased at 48 and 72 hpi in DENV-2-infected cells compared to time-matched mock-infected controls (Figure A3.2C), which is consistent with our observation that both the protease (SKI-1/S1P) and substrate (SREBP-2) mRNA levels are increased during the late stages of the virus lifecycle. Interestingly, PCSK9 mRNA level was significantly decreased at 72 hpi in DENV-2-infected cells (Figure A3.2C). Since furin regulates PCSK9 mRNA levels in hepatocytes (371), we examined the level of furin mRNA during DENV-2 infection. Consistent with these data, we observed a robust up-regulation of furin mRNA during DENV-2 infection at 48 and 72 hpi compared to time-matched mock-infected controls (Figure A3.2C).   Collectively, these results demonstrate a stage-specific modulation of the SREBP transcriptional network [e.g., SREBP protease (SKI-1/S1P), SREBP substrates (SREBP-1c and SREBP-2), and SREBP-2 target gene (LDLR)] at the late stages of DENV infection in Huh-7.5.1 cells. Surprisingly, we observed a virus-induced transcriptional down-regulation of PCSK9 gene expression concomitant with a transcriptional up-regulation of furin, the key in vivo-inactivating protease of circulating PCSK9 (371). Further studies are however required to investigate the role of PCSK9 and its interaction with furin during DENV infection (372).  To further confirm at the protein level that DENV induces a temporal activation of the SREBP-2 transcriptional network in human hepatoma cells, we examined the proteolytic maturation of the SREBP-2 precursor proteins during the viral lifecycle.  We determined the amount of SREBP-2 precursor (SREBP-2 P) and the mature (cleaved) form of SREBP-2 (SREBP-2 C) by quantitative Western blot analysis.  Cellular lysates from mock-infected and DENV-2-infected Huh-7.5.1 cells at 4, 8, 24, and 48 hpi were separated by SDS-PAGE and immunoblotted with the indicated antibodies (Figure A3.2D). As previously reported by Peña and Harris (2012), no changes in protein abundance were detected by Western blotting analysis of SREBP-2 P and SREBP-2 C molecular forms in virus-infected cells over time-matched mock-infected controls during early DENV-2 infection [(Figure A3.2D): 4, 8 and 24 hpi] (373). In contrast, a significant reduction at 48 hpi in SREBP-2 P protein abundance was observed over time-matched mock-infected controls, suggesting a possible increase in the SREBP-2 proteolytic processing (Figure A3.2D), consistent with our transcription data. This 	   118	  observation was confirmed by quantifying the band intensities associated with SREBP-2 P and SREBP-2 C molecular forms using the Odyssey infrared imaging system, normalized to β-tubulin, and expressing the results as a percentage of SREBP-2 cleavage (Figure A3.2E).  To confirm productive infection of Huh-7.5.1 cells during the time course of these experiments, the cellular lysates were probed for DENV NS1 (Figure A3.2D).   A significant reduction at 48 hpi in SREBP-2 P protein abundance observed over time-matched mock-infected controls could also be due to a decrease in translation of the precursor protein instead of the suggested increase in the SREBP-2 proteolytic processing. One possible experiment to clarify this observation would be to measure active SREBP-2 in cell nuclei during DENV infection. Also, protein levels of the SREBP-2 target genes, LDLR and PCSK9, could be measured during DENV infection. Increased protein levels of the SREBP-2 target genes could validate upregulation the SKI-1/S1P-mediated proteolytic activation of the SREBP pathway during DENV infection.   5.2.4 Dissecting the role of PF-429242 on ATF6 activation in DENV infection  While it is well established that SKI-1/S1P is involved in regulating activation of SREBP and lipid biogenesis, SKI-1/S1P also activates the ER resident protein and activator of the unfolded protein response (UPR) ATF6 through the same mechanism (340).  Studies with DENV and West Nile virus have demonstrated activation of ATF6 and the modulation of the UPR (374, 375).  The data presented in Chapter 4 showed that PF-429242-mediated inhibition of DENV infection by the block of SKI-1/S1P-mediated activation of the SREBP pathway and LD formation.  To dissect another potential mechanism of PF-429242-mediated inhibition of DENV infection, the protein levels of ATF6 and its target genes such as X-box binding protein 1 (Xbp-1) could be measured after treatment with the small molecule in naïve and DENV-infected cells.   5.2.5 In vivo effect of SKI-1/S1P inhibition on DENV infection  The development of a credible murine model of DENV infection has been challenging, because DENV clinical isolates do not easily replicate or cause pathology in 	   119	  immunocompetent mice.  There are several immunocompromised mouse models as well as mouse-adapted viruses that enable research to address specific questions otherwise impossible to answer in human studies.  Specifically for testing anti-viral drug studies in vivo, AG129 mice have increasingly become the standard mouse model.  AG129 are mice deficient in IFN-α/β and -γ receptors (376).  In humans, DENV inhibits IFN signalling to establish infection, which does not happen in mouse cells.  Thus, AG129 mice deficient in IFN receptors can support DENV replication and show signs of severe dengue disease following intravenous or intraperitoneal infection with some DENV-strains.  Wild-type mice however develop severe dengue only after intracranial challenge, which is not a natural route of DENV infection.  Mortality, viral load, and signs of disease can all be used to assess the drug efficacy in AG129 mice (376).  Previously, an adenosine nucleoside was shown to limit viremia and reduce mortality in DENV-infected AG129 mice, likely by blocking viral RNA synthesis (377). Other DENV inhibitors targeting the NS3 helicase (378) or the capsid protein (379) have also successfully reduced viremia and organ viral titers in AG129 mice.  Using this mouse model the application of PF-429242 or other lipid-lowering candidates could be tested in reducing DENV infection.  Alternatively, immunocompetent mice expressing the mutant Mtbs1 (SKI-1/S1P) allele could be intracranially challenged with DENV to determine the in vivo importance of SKI-1/S1P in DENV infection.  5.3 Conclusions  Lipids and cholesterol play profound roles during the lifecycles of many human viruses, including HCV and DENV.  These viruses not only regulate lipid microenvironments during infection but also depend on specific lipids for entry, replication and assembly (2).  A better understanding of key regulators of lipid metabolism modulated during HCV and DENV infections will help uncover pathways that may represent potential therapeutic targets to inhibit viral infection and/or prevent liver disease progression.  Our studies focused on several important regulators of lipid homeostasis, including miRNAs (miR-122, miR-24 and miR-223) and two host proprotein convertases, PCSK9 and SKI-1/S1P, during HCV and DENV infection. 	   120	   The ability of miRNAs to target host proprotein convertases and mediate host lipid metabolism and viral lifecycle adds complexity to understanding the host-pathogen interactions during viral infections (Figure 5.1).  We attempted to understand the role of miRNAs, which are known to regulate lipid metabolism, during HCV infection.  This was done by assessing circulating levels of miR-122, miR-24 and miR-223 during different stages of HCV infection and anti-HCV therapy.  By investigating miRNA levels in relation to different parameters associated with HCV infection, we identified differential changes in these circulating miRNAs before, during and with treatment-based cures.  The observed changes likely reflect a physiological response to viral clearance that may impact metabolic changes normally occurring when CHC-patients achieve SVR, including changes of cholesterol and beta-lipoprotein levels to baseline levels (207).  Within our data sets presented in Chapter 3, circulating miR-24 levels were correlated with extracellular PCSK9 concentrations in CHC-infected patients indicating for the first time an in vivo link between the two.  Previously, miR-24 has been predicted to target PCSK9 (359), an important protein of the PC family, associated with LDLR turnover (356), which is required for HCV cell entry.  We also demonstrated that extracellular PCSK9 levels are increased in CHC-infected patients, who achieved SVR, while treatment of Huh-7.5.1 cells with extracellular PCSK9 resulted in inhibition of HCV infection.  These findings describe an intricate interplay between circulating miR-24 and PCSK9 regulating lipid metabolism and HCV infection (Figure 5.1).   In addition to dissecting the importance of extracellular PCSK9 in HCV infection, we identified that SKI-1S1P, another member of the PC family, is involved in DENV infection.  SKI-1/S1P regulates processing of transcription factors in one of the major host lipid metabolic pathways, SREBP (33).  We found that SKI-1/S1P inhibition by downregulating the abundance of host LDs blocked DENV infection at multiple steps of the virus lifecycle.  This study again reiterates the theme of a well-coordinated regulation between the host factors, lipid metabolism and viral infection (Figure 5.1).  Overall, these data demonstrate the important role of the regulators of lipid metabolism during viral infections.  This study provides insights into complex pathology-associated host-virus interactions, which lays the foundation for future drug discovery and diagnostic strategies for Flaviviridae-associated diseases.  	   121	     	   122	  Figure 5.1.  The intricate interplay between the host factors, viral infection and lipid metabolism.   Interactions among the host factors and lipid metabolism are modulated during viral infections, which disrupt stable internal conditions in order to support their lifecycles.  Viral infections, including HCV and DENV, modulate host lipid metabolism.  Also, expression of miRNAs and proprotein converatases are modulated upon viral infection.  MiRNAs are able to regulate expression of multiple proprotein convertases.  Finally, miRNAs (miR-122, miR-24 and miR-223) and proprotein convertase (SKI-1/S1P and PCSK9) may also regulate host lipid metabolic pathways as well as viral infection (HCV and DENV) that were investigated in this study.  Thus, this study demonstrates an intricate interplay between viral infection, host factors and lipid metabolism.    	   123	  Bibliography:  1.	   Cohen	  DE,	  Fisher	  EA.	  2013.	  Lipoprotein	  metabolism,	  dyslipidemia,	  and	  nonalcoholic	  fatty	  liver	  disease.	  Semin	  Liver	  Dis	  33:380-­‐388.	  2.	   Heaton	  NS,	  Randall	  G.	  2011.	  Multifaceted	  roles	  for	  lipids	  in	  viral	  infection.	  Trends	  Microbiol	  19:368-­‐375.	  3.	   Nguyen	  P,	  Leray	  V,	  Diez	  M,	  Serisier	  S,	  Le	  Bloc'h	  J,	  Siliart	  B,	  Dumon	  H.	  2008.	  Liver	  lipid	  metabolism.	  J	  Anim	  Physiol	  Anim	  Nutr	  (Berl)	  92:272-­‐283.	  4.	   Olofsson	  SO,	  Bostrom	  P,	  Andersson	  L,	  Rutberg	  M,	  Perman	  J,	  Boren	  J.	  2009.	  Lipid	  droplets	  as	  dynamic	  organelles	  connecting	  storage	  and	  efflux	  of	  lipids.	  Biochim	  Biophys	  Acta	  1791:448-­‐458.	  5.	   Goldstein	  JL,	  Brown	  MS,	  Anderson	  RG,	  Russell	  DW,	  Schneider	  WJ.	  1985.	  Receptor-­‐mediated	  endocytosis:	  concepts	  emerging	  from	  the	  LDL	  receptor	  system.	  Annu	  Rev	  Cell	  Biol	  1:1-­‐39.	  6.	   Goldstein	  JL,	  Brown	  MS.	  2009.	  The	  LDL	  receptor.	  Arterioscler	  Thromb	  Vasc	  Biol	  29:431-­‐438.	  7.	   Soutar	  AK,	  Naoumova	  RP.	  2007.	  Mechanisms	  of	  disease:	  genetic	  causes	  of	  familial	  hypercholesterolemia.	  Nat	  Clin	  Pract	  Cardiovasc	  Med	  4:214-­‐225.	  8.	   Seidah	  NG,	  Awan	  Z,	  Chretien	  M,	  Mbikay	  M.	  2014.	  PCSK9:	  a	  key	  modulator	  of	  cardiovascular	  health.	  Circ	  Res	  114:1022-­‐1036.	  9.	   Park	  SW,	  Moon	  YA,	  Horton	  JD.	  2004.	  Post-­‐transcriptional	  regulation	  of	  low	  density	  lipoprotein	  receptor	  protein	  by	  proprotein	  convertase	  subtilisin/kexin	  type	  9a	  in	  mouse	  liver.	  J	  Biol	  Chem	  279:50630-­‐50638.	  10.	   Maxwell	  KN,	  Fisher	  EA,	  Breslow	  JL.	  2005.	  Overexpression	  of	  PCSK9	  accelerates	  the	  degradation	  of	  the	  LDLR	  in	  a	  post-­‐endoplasmic	  reticulum	  compartment.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  102:2069-­‐2074.	  11.	   Maxwell	  KN,	  Breslow	  JL.	  2004.	  Adenoviral-­‐mediated	  expression	  of	  Pcsk9	  in	  mice	  results	  in	  a	  low-­‐density	  lipoprotein	  receptor	  knockout	  phenotype.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  101:7100-­‐7105.	  12.	   Abifadel	  M,	  Varret	  M,	  Rabes	  JP,	  Allard	  D,	  Ouguerram	  K,	  Devillers	  M,	  Cruaud	  C,	  Benjannet	  S,	  Wickham	  L,	  Erlich	  D,	  Derre	  A,	  Villeger	  L,	  Farnier	  M,	  Beucler	  I,	  Bruckert	  E,	  Chambaz	  J,	  Chanu	  B,	  Lecerf	  JM,	  Luc	  G,	  Moulin	  P,	  Weissenbach	  J,	  Prat	  A,	  Krempf	  M,	  Junien	  C,	  Seidah	  NG,	  Boileau	  C.	  2003.	  Mutations	  in	  PCSK9	  cause	  autosomal	  dominant	  hypercholesterolemia.	  Nat	  Genet	  34:154-­‐156.	  13.	   Leren	  TP.	  2004.	  Mutations	  in	  the	  PCSK9	  gene	  in	  Norwegian	  subjects	  with	  autosomal	  dominant	  hypercholesterolemia.	  Clin	  Genet	  65:419-­‐422.	  14.	   Timms	  KM,	  Wagner	  S,	  Samuels	  ME,	  Forbey	  K,	  Goldfine	  H,	  Jammulapati	  S,	  Skolnick	  MH,	  Hopkins	  PN,	  Hunt	  SC,	  Shattuck	  DM.	  2004.	  A	  mutation	  in	  PCSK9	  causing	  autosomal-­‐dominant	  hypercholesterolemia	  in	  a	  Utah	  pedigree.	  Hum	  Genet	  114:349-­‐353.	  15.	   Berge	  KE,	  Ose	  L,	  Leren	  TP.	  2006.	  Missense	  mutations	  in	  the	  PCSK9	  gene	  are	  associated	  with	  hypocholesterolemia	  and	  possibly	  increased	  response	  to	  statin	  therapy.	  Arterioscler	  Thromb	  Vasc	  Biol	  26:1094-­‐1100.	  	   124	  16.	   Yue	  P,	  Averna	  M,	  Lin	  X,	  Schonfeld	  G.	  2006.	  The	  c.43_44insCTG	  variation	  in	  PCSK9	  is	  associated	  with	  low	  plasma	  LDL-­‐cholesterol	  in	  a	  Caucasian	  population.	  Hum	  Mutat	  27:460-­‐466.	  17.	   Lambert	  G,	  Sjouke	  B,	  Choque	  B,	  Kastelein	  JJ,	  Hovingh	  GK.	  2012.	  The	  PCSK9	  decade.	  J	  Lipid	  Res	  53:2515-­‐2524.	  18.	   Seidah	  NG.	  2011.	  The	  proprotein	  convertases,	  20	  years	  later.	  Methods	  Mol	  Biol	  768:23-­‐57.	  19.	   Costet	  P,	  Cariou	  B,	  Lambert	  G,	  Lalanne	  F,	  Lardeux	  B,	  Jarnoux	  AL,	  Grefhorst	  A,	  Staels	  B,	  Krempf	  M.	  2006.	  Hepatic	  PCSK9	  expression	  is	  regulated	  by	  nutritional	  status	  via	  insulin	  and	  sterol	  regulatory	  element-­‐binding	  protein	  1c.	  J	  Biol	  Chem	  281:6211-­‐6218.	  20.	   Rashid	  S,	  Curtis	  DE,	  Garuti	  R,	  Anderson	  NN,	  Bashmakov	  Y,	  Ho	  YK,	  Hammer	  RE,	  Moon	  YA,	  Horton	  JD.	  2005.	  Decreased	  plasma	  cholesterol	  and	  hypersensitivity	  to	  statins	  in	  mice	  lacking	  Pcsk9.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  102:5374-­‐5379.	  21.	   Ajufo	  E,	  Rader	  DJ.	  2016.	  Recent	  advances	  in	  the	  pharmacological	  management	  of	  hypercholesterolaemia.	  Lancet	  Diabetes	  Endocrinol	  doi:10.1016/S2213-­‐8587(16)00074-­‐7.	  22.	   Eberle	  D,	  Hegarty	  B,	  Bossard	  P,	  Ferre	  P,	  Foufelle	  F.	  2004.	  SREBP	  transcription	  factors:	  master	  regulators	  of	  lipid	  homeostasis.	  Biochimie	  86:839-­‐848.	  23.	   Brown	  MS,	  Goldstein	  JL.	  1997.	  The	  SREBP	  pathway:	  regulation	  of	  cholesterol	  metabolism	  by	  proteolysis	  of	  a	  membrane-­‐bound	  transcription	  factor.	  Cell	  89:331-­‐340.	  24.	   Brown	  MS,	  Goldstein	  JL.	  1999.	  A	  proteolytic	  pathway	  that	  controls	  the	  cholesterol	  content	  of	  membranes,	  cells,	  and	  blood.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  96:11041-­‐11048.	  25.	   Jeong	  HJ,	  Lee	  HS,	  Kim	  KS,	  Kim	  YK,	  Yoon	  D,	  Park	  SW.	  2008.	  Sterol-­‐dependent	  regulation	  of	  proprotein	  convertase	  subtilisin/kexin	  type	  9	  expression	  by	  sterol-­‐regulatory	  element	  binding	  protein-­‐2.	  J	  Lipid	  Res	  49:399-­‐409.	  26.	   Espenshade	  PJ.	  2006.	  SREBPs:	  sterol-­‐regulated	  transcription	  factors.	  J	  Cell	  Sci	  119:973-­‐976.	  27.	   Seidah	  NG,	  Mowla	  SJ,	  Hamelin	  J,	  Mamarbachi	  AM,	  Benjannet	  S,	  Toure	  BB,	  Basak	  A,	  Munzer	  JS,	  Marcinkiewicz	  J,	  Zhong	  M,	  Barale	  JC,	  Lazure	  C,	  Murphy	  RA,	  Chretien	  M,	  Marcinkiewicz	  M.	  1999.	  Mammalian	  subtilisin/kexin	  isozyme	  SKI-­‐1:	  A	  widely	  expressed	  proprotein	  convertase	  with	  a	  unique	  cleavage	  specificity	  and	  cellular	  localization.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  96:1321-­‐1326.	  28.	   Toure	  BB,	  Munzer	  JS,	  Basak	  A,	  Benjannet	  S,	  Rochemont	  J,	  Lazure	  C,	  Chretien	  M,	  Seidah	  NG.	  2000.	  Biosynthesis	  and	  enzymatic	  characterization	  of	  human	  SKI-­‐1/S1P	  and	  the	  processing	  of	  its	  inhibitory	  prosegment.	  J	  Biol	  Chem	  275:2349-­‐2358.	  29.	   Elagoz	  A,	  Benjannet	  S,	  Mammarbassi	  A,	  Wickham	  L,	  Seidah	  NG.	  2002.	  Biosynthesis	  and	  cellular	  trafficking	  of	  the	  convertase	  SKI-­‐1/S1P:	  ectodomain	  shedding	  requires	  SKI-­‐1	  activity.	  J	  Biol	  Chem	  277:11265-­‐11275.	  30.	   Horton	  JD,	  Goldstein	  JL,	  Brown	  MS.	  2002.	  SREBPs:	  activators	  of	  the	  complete	  program	  of	  cholesterol	  and	  fatty	  acid	  synthesis	  in	  the	  liver.	  J	  Clin	  Invest	  109:1125-­‐1131.	  	   125	  31.	   Haze	  K,	  Yoshida	  H,	  Yanagi	  H,	  Yura	  T,	  Mori	  K.	  1999.	  Mammalian	  transcription	  factor	  ATF6	  is	  synthesized	  as	  a	  transmembrane	  protein	  and	  activated	  by	  proteolysis	  in	  response	  to	  endoplasmic	  reticulum	  stress.	  Mol	  Biol	  Cell	  10:3787-­‐3799.	  32.	   Yang	  J,	  Goldstein	  JL,	  Hammer	  RE,	  Moon	  YA,	  Brown	  MS,	  Horton	  JD.	  2001.	  Decreased	  lipid	  synthesis	  in	  livers	  of	  mice	  with	  disrupted	  Site-­‐1	  protease	  gene.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  98:13607-­‐13612.	  33.	   Hawkins	  JL,	  Robbins	  MD,	  Warren	  LC,	  Xia	  D,	  Petras	  SF,	  Valentine	  JJ,	  Varghese	  AH,	  Wang	  IK,	  Subashi	  TA,	  Shelly	  LD,	  Hay	  BA,	  Landschulz	  KT,	  Geoghegan	  KF,	  Harwood	  HJ,	  Jr.	  2008.	  Pharmacologic	  inhibition	  of	  site	  1	  protease	  activity	  inhibits	  sterol	  regulatory	  element-­‐binding	  protein	  processing	  and	  reduces	  lipogenic	  enzyme	  gene	  expression	  and	  lipid	  synthesis	  in	  cultured	  cells	  and	  experimental	  animals.	  J	  Pharmacol	  Exp	  Ther	  326:801-­‐808.	  34.	   Lenz	  O,	  ter	  Meulen	  J,	  Klenk	  HD,	  Seidah	  NG,	  Garten	  W.	  2001.	  The	  Lassa	  virus	  glycoprotein	  precursor	  GP-­‐C	  is	  proteolytically	  processed	  by	  subtilase	  SKI-­‐1/S1P.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  98:12701-­‐12705.	  35.	   Beyer	  WR,	  Popplau	  D,	  Garten	  W,	  von	  Laer	  D,	  Lenz	  O.	  2003.	  Endoproteolytic	  processing	  of	  the	  lymphocytic	  choriomeningitis	  virus	  glycoprotein	  by	  the	  subtilase	  SKI-­‐1/S1P.	  J	  Virol	  77:2866-­‐2872.	  36.	   Vincent	  MJ,	  Sanchez	  AJ,	  Erickson	  BR,	  Basak	  A,	  Chretien	  M,	  Seidah	  NG,	  Nichol	  ST.	  2003.	  Crimean-­‐Congo	  hemorrhagic	  fever	  virus	  glycoprotein	  proteolytic	  processing	  by	  subtilase	  SKI-­‐1.	  J	  Virol	  77:8640-­‐8649.	  37.	   Olmstead	  AD,	  Knecht	  W,	  Lazarov	  I,	  Dixit	  SB,	  Jean	  F.	  2012.	  Human	  subtilase	  SKI-­‐1/S1P	  is	  a	  master	  regulator	  of	  the	  HCV	  Lifecycle	  and	  a	  potential	  host	  cell	  target	  for	  developing	  indirect-­‐acting	  antiviral	  agents.	  PLoS	  Pathog	  8:e1002468.	  38.	   Villareal	  VA,	  Rodgers	  MA,	  Costello	  DA,	  Yang	  PL.	  2015.	  Targeting	  host	  lipid	  synthesis	  and	  metabolism	  to	  inhibit	  dengue	  and	  hepatitis	  C	  viruses.	  Antiviral	  Res	  124:110-­‐121.	  39.	   Ambros	  V.	  2004.	  The	  functions	  of	  animal	  microRNAs.	  Nature	  431:350-­‐355.	  40.	   Szabo	  G,	  Bala	  S.	  2013.	  MicroRNAs	  in	  liver	  disease.	  Nat	  Rev	  Gastroenterol	  Hepatol	  doi:10.1038/nrgastro.2013.87.	  41.	   Cortez	  MA,	  Calin	  GA.	  2009.	  MicroRNA	  identification	  in	  plasma	  and	  serum:	  a	  new	  tool	  to	  diagnose	  and	  monitor	  diseases.	  Expert	  Opin	  Biol	  Ther	  9:703-­‐711.	  42.	   Lee	  RC,	  Feinbaum	  RL,	  Ambros	  V.	  1993.	  The	  C.	  elegans	  heterochronic	  gene	  lin-­‐4	  encodes	  small	  RNAs	  with	  antisense	  complementarity	  to	  lin-­‐14.	  Cell	  75:843-­‐854.	  43.	   Wightman	  B,	  Ha	  I,	  Ruvkun	  G.	  1993.	  Posttranscriptional	  regulation	  of	  the	  heterochronic	  gene	  lin-­‐14	  by	  lin-­‐4	  mediates	  temporal	  pattern	  formation	  in	  C.	  elegans.	  Cell	  75:855-­‐862.	  44.	   Slack	  FJ,	  Basson	  M,	  Liu	  Z,	  Ambros	  V,	  Horvitz	  HR,	  Ruvkun	  G.	  2000.	  The	  lin-­‐41	  RBCC	  gene	  acts	  in	  the	  C.	  elegans	  heterochronic	  pathway	  between	  the	  let-­‐7	  regulatory	  RNA	  and	  the	  LIN-­‐29	  transcription	  factor.	  Mol	  Cell	  5:659-­‐669.	  45.	   Cai	  X,	  Hagedorn	  CH,	  Cullen	  BR.	  2004.	  Human	  microRNAs	  are	  processed	  from	  capped,	  polyadenylated	  transcripts	  that	  can	  also	  function	  as	  mRNAs.	  RNA	  10:1957-­‐1966.	  46.	   Lee	  Y,	  Kim	  M,	  Han	  J,	  Yeom	  KH,	  Lee	  S,	  Baek	  SH,	  Kim	  VN.	  2004.	  MicroRNA	  genes	  are	  transcribed	  by	  RNA	  polymerase	  II.	  EMBO	  J	  23:4051-­‐4060.	  	   126	  47.	   Lee	  Y,	  Ahn	  C,	  Han	  J,	  Choi	  H,	  Kim	  J,	  Yim	  J,	  Lee	  J,	  Provost	  P,	  Radmark	  O,	  Kim	  S,	  Kim	  VN.	  2003.	  The	  nuclear	  RNase	  III	  Drosha	  initiates	  microRNA	  processing.	  Nature	  425:415-­‐419.	  48.	   Han	  J,	  Lee	  Y,	  Yeom	  KH,	  Kim	  YK,	  Jin	  H,	  Kim	  VN.	  2004.	  The	  Drosha-­‐DGCR8	  complex	  in	  primary	  microRNA	  processing.	  Genes	  Dev	  18:3016-­‐3027.	  49.	   Bohnsack	  MT,	  Czaplinski	  K,	  Gorlich	  D.	  2004.	  Exportin	  5	  is	  a	  RanGTP-­‐dependent	  dsRNA-­‐binding	  protein	  that	  mediates	  nuclear	  export	  of	  pre-­‐miRNAs.	  RNA	  10:185-­‐191.	  50.	   Kim	  VN.	  2004.	  MicroRNA	  precursors	  in	  motion:	  exportin-­‐5	  mediates	  their	  nuclear	  export.	  Trends	  Cell	  Biol	  14:156-­‐159.	  51.	   Lund	  E,	  Guttinger	  S,	  Calado	  A,	  Dahlberg	  JE,	  Kutay	  U.	  2004.	  Nuclear	  export	  of	  microRNA	  precursors.	  Science	  303:95-­‐98.	  52.	   Hutvagner	  G,	  McLachlan	  J,	  Pasquinelli	  AE,	  Balint	  E,	  Tuschl	  T,	  Zamore	  PD.	  2001.	  A	  cellular	  function	  for	  the	  RNA-­‐interference	  enzyme	  Dicer	  in	  the	  maturation	  of	  the	  let-­‐7	  small	  temporal	  RNA.	  Science	  293:834-­‐838.	  53.	   Ketting	  RF,	  Fischer	  SE,	  Bernstein	  E,	  Sijen	  T,	  Hannon	  GJ,	  Plasterk	  RH.	  2001.	  Dicer	  functions	  in	  RNA	  interference	  and	  in	  synthesis	  of	  small	  RNA	  involved	  in	  developmental	  timing	  in	  C.	  elegans.	  Genes	  Dev	  15:2654-­‐2659.	  54.	   Knight	  SW,	  Bass	  BL.	  2001.	  A	  role	  for	  the	  RNase	  III	  enzyme	  DCR-­‐1	  in	  RNA	  interference	  and	  germ	  line	  development	  in	  Caenorhabditis	  elegans.	  Science	  293:2269-­‐2271.	  55.	   Schwarz	  DS,	  Hutvagner	  G,	  Du	  T,	  Xu	  Z,	  Aronin	  N,	  Zamore	  PD.	  2003.	  Asymmetry	  in	  the	  assembly	  of	  the	  RNAi	  enzyme	  complex.	  Cell	  115:199-­‐208.	  56.	   Khvorova	  A,	  Reynolds	  A,	  Jayasena	  SD.	  2003.	  Functional	  siRNAs	  and	  miRNAs	  exhibit	  strand	  bias.	  Cell	  115:209-­‐216.	  57.	   Jones-­‐Rhoades	  MW,	  Bartel	  DP,	  Bartel	  B.	  2006.	  MicroRNAS	  and	  their	  regulatory	  roles	  in	  plants.	  Annu	  Rev	  Plant	  Biol	  57:19-­‐53.	  58.	   Doench	  JG,	  Sharp	  PA.	  2004.	  Specificity	  of	  microRNA	  target	  selection	  in	  translational	  repression.	  Genes	  Dev	  18:504-­‐511.	  59.	   Brennecke	  J,	  Stark	  A,	  Russell	  RB,	  Cohen	  SM.	  2005.	  Principles	  of	  microRNA-­‐target	  recognition.	  PLoS	  Biol	  3:e85.	  60.	   Lewis	  BP,	  Burge	  CB,	  Bartel	  DP.	  2005.	  Conserved	  seed	  pairing,	  often	  flanked	  by	  adenosines,	  indicates	  that	  thousands	  of	  human	  genes	  are	  microRNA	  targets.	  Cell	  120:15-­‐20.	  61.	   Grimson	  A,	  Farh	  KK,	  Johnston	  WK,	  Garrett-­‐Engele	  P,	  Lim	  LP,	  Bartel	  DP.	  2007.	  MicroRNA	  targeting	  specificity	  in	  mammals:	  determinants	  beyond	  seed	  pairing.	  Mol	  Cell	  27:91-­‐105.	  62.	   Nielsen	  CB,	  Shomron	  N,	  Sandberg	  R,	  Hornstein	  E,	  Kitzman	  J,	  Burge	  CB.	  2007.	  Determinants	  of	  targeting	  by	  endogenous	  and	  exogenous	  microRNAs	  and	  siRNAs.	  RNA	  13:1894-­‐1910.	  63.	   Cortez	  MA,	  Bueso-­‐Ramos	  C,	  Ferdin	  J,	  Lopez-­‐Berestein	  G,	  Sood	  AK,	  Calin	  GA.	  2011.	  MicroRNAs	  in	  body	  fluids-­‐-­‐the	  mix	  of	  hormones	  and	  biomarkers.	  Nat	  Rev	  Clin	  Oncol	  8:467-­‐477.	  64.	   Filipowicz	  W,	  Bhattacharyya	  SN,	  Sonenberg	  N.	  2008.	  Mechanisms	  of	  post-­‐transcriptional	  regulation	  by	  microRNAs:	  are	  the	  answers	  in	  sight?	  Nat	  Rev	  Genet	  9:102-­‐114.	  	   127	  65.	   Merrick	  WC.	  2004.	  Cap-­‐dependent	  and	  cap-­‐independent	  translation	  in	  eukaryotic	  systems.	  Gene	  332:1-­‐11.	  66.	   Kapp	  LD,	  Lorsch	  JR.	  2004.	  The	  molecular	  mechanics	  of	  eukaryotic	  translation.	  Annu	  Rev	  Biochem	  73:657-­‐704.	  67.	   Wells	  SE,	  Hillner	  PE,	  Vale	  RD,	  Sachs	  AB.	  1998.	  Circularization	  of	  mRNA	  by	  eukaryotic	  translation	  initiation	  factors.	  Mol	  Cell	  2:135-­‐140.	  68.	   Derry	  MC,	  Yanagiya	  A,	  Martineau	  Y,	  Sonenberg	  N.	  2006.	  Regulation	  of	  poly(A)-­‐binding	  protein	  through	  PABP-­‐interacting	  proteins.	  Cold	  Spring	  Harb	  Symp	  Quant	  Biol	  71:537-­‐543.	  69.	   Chekulaeva	  M,	  Hentze	  MW,	  Ephrussi	  A.	  2006.	  Bruno	  acts	  as	  a	  dual	  repressor	  of	  oskar	  translation,	  promoting	  mRNA	  oligomerization	  and	  formation	  of	  silencing	  particles.	  Cell	  124:521-­‐533.	  70.	   Richter	  JD,	  Sonenberg	  N.	  2005.	  Regulation	  of	  cap-­‐dependent	  translation	  by	  eIF4E	  inhibitory	  proteins.	  Nature	  433:477-­‐480.	  71.	   Cho	  PF,	  Poulin	  F,	  Cho-­‐Park	  YA,	  Cho-­‐Park	  IB,	  Chicoine	  JD,	  Lasko	  P,	  Sonenberg	  N.	  2005.	  A	  new	  paradigm	  for	  translational	  control:	  inhibition	  via	  5'-­‐3'	  mRNA	  tethering	  by	  Bicoid	  and	  the	  eIF4E	  cognate	  4EHP.	  Cell	  121:411-­‐423.	  72.	   Valinezhad	  Orang	  A,	  Safaralizadeh	  R,	  Kazemzadeh-­‐Bavili	  M.	  2014.	  Mechanisms	  of	  miRNA-­‐Mediated	  Gene	  Regulation	  from	  Common	  Downregulation	  to	  mRNA-­‐Specific	  Upregulation.	  Int	  J	  Genomics	  2014:970607.	  73.	   Chim	  SS,	  Shing	  TK,	  Hung	  EC,	  Leung	  TY,	  Lau	  TK,	  Chiu	  RW,	  Lo	  YM.	  2008.	  Detection	  and	  characterization	  of	  placental	  microRNAs	  in	  maternal	  plasma.	  Clin	  Chem	  54:482-­‐490.	  74.	   Turchinovich	  A,	  Weiz	  L,	  Burwinkel	  B.	  2012.	  Extracellular	  miRNAs:	  the	  mystery	  of	  their	  origin	  and	  function.	  Trends	  Biochem	  Sci	  37:460-­‐465.	  75.	   Creemers	  EE,	  Tijsen	  AJ,	  Pinto	  YM.	  2012.	  Circulating	  microRNAs:	  novel	  biomarkers	  and	  extracellular	  communicators	  in	  cardiovascular	  disease?	  Circ	  Res	  110:483-­‐495.	  76.	   Arroyo	  JD,	  Chevillet	  JR,	  Kroh	  EM,	  Ruf	  IK,	  Pritchard	  CC,	  Gibson	  DF,	  Mitchell	  PS,	  Bennett	  CF,	  Pogosova-­‐Agadjanyan	  EL,	  Stirewalt	  DL,	  Tait	  JF,	  Tewari	  M.	  2011.	  Argonaute2	  complexes	  carry	  a	  population	  of	  circulating	  microRNAs	  independent	  of	  vesicles	  in	  human	  plasma.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  108:5003-­‐5008.	  77.	   Vickers	  KC,	  Palmisano	  BT,	  Shoucri	  BM,	  Shamburek	  RD,	  Remaley	  AT.	  2011.	  MicroRNAs	  are	  transported	  in	  plasma	  and	  delivered	  to	  recipient	  cells	  by	  high-­‐density	  lipoproteins.	  Nat	  Cell	  Biol	  13:423-­‐433.	  78.	   Valadi	  H,	  Ekstrom	  K,	  Bossios	  A,	  Sjostrand	  M,	  Lee	  JJ,	  Lotvall	  JO.	  2007.	  Exosome-­‐mediated	  transfer	  of	  mRNAs	  and	  microRNAs	  is	  a	  novel	  mechanism	  of	  genetic	  exchange	  between	  cells.	  Nat	  Cell	  Biol	  9:654-­‐659.	  79.	   El-­‐Hefnawy	  T,	  Raja	  S,	  Kelly	  L,	  Bigbee	  WL,	  Kirkwood	  JM,	  Luketich	  JD,	  Godfrey	  TE.	  2004.	  Characterization	  of	  amplifiable,	  circulating	  RNA	  in	  plasma	  and	  its	  potential	  as	  a	  tool	  for	  cancer	  diagnostics.	  Clin	  Chem	  50:564-­‐573.	  80.	   Kosaka	  N,	  Iguchi	  H,	  Yoshioka	  Y,	  Takeshita	  F,	  Matsuki	  Y,	  Ochiya	  T.	  2010.	  Secretory	  mechanisms	  and	  intercellular	  transfer	  of	  microRNAs	  in	  living	  cells.	  J	  Biol	  Chem	  285:17442-­‐17452.	  	   128	  81.	   Pigati	  L,	  Yaddanapudi	  SC,	  Iyengar	  R,	  Kim	  DJ,	  Hearn	  SA,	  Danforth	  D,	  Hastings	  ML,	  Duelli	  DM.	  2010.	  Selective	  release	  of	  microRNA	  species	  from	  normal	  and	  malignant	  mammary	  epithelial	  cells.	  PLoS	  One	  5:e13515.	  82.	   Bukong	  TN,	  Momen-­‐Heravi	  F,	  Kodys	  K,	  Bala	  S,	  Szabo	  G.	  2014.	  Exosomes	  from	  hepatitis	  C	  infected	  patients	  transmit	  HCV	  infection	  and	  contain	  replication	  competent	  viral	  RNA	  in	  complex	  with	  Ago2-­‐miR122-­‐HSP90.	  PLoS	  Pathog	  10:e1004424.	  83.	   Wagner	  J,	  Riwanto	  M,	  Besler	  C,	  Knau	  A,	  Fichtlscherer	  S,	  Roxe	  T,	  Zeiher	  AM,	  Landmesser	  U,	  Dimmeler	  S.	  2013.	  Characterization	  of	  levels	  and	  cellular	  transfer	  of	  circulating	  lipoprotein-­‐bound	  microRNAs.	  Arterioscler	  Thromb	  Vasc	  Biol	  33:1392-­‐1400.	  84.	   Zernecke	  A,	  Bidzhekov	  K,	  Noels	  H,	  Shagdarsuren	  E,	  Gan	  L,	  Denecke	  B,	  Hristov	  M,	  Koppel	  T,	  Jahantigh	  MN,	  Lutgens	  E,	  Wang	  S,	  Olson	  EN,	  Schober	  A,	  Weber	  C.	  2009.	  Delivery	  of	  microRNA-­‐126	  by	  apoptotic	  bodies	  induces	  CXCL12-­‐dependent	  vascular	  protection.	  Sci	  Signal	  2:ra81.	  85.	   Zhang	  Y,	  Liu	  D,	  Chen	  X,	  Li	  J,	  Li	  L,	  Bian	  Z,	  Sun	  F,	  Lu	  J,	  Yin	  Y,	  Cai	  X,	  Sun	  Q,	  Wang	  K,	  Ba	  Y,	  Wang	  Q,	  Wang	  D,	  Yang	  J,	  Liu	  P,	  Xu	  T,	  Yan	  Q,	  Zhang	  J,	  Zen	  K,	  Zhang	  CY.	  2010.	  Secreted	  monocytic	  miR-­‐150	  enhances	  targeted	  endothelial	  cell	  migration.	  Mol	  Cell	  39:133-­‐144.	  86.	   Braconi	  C,	  Henry	  JC,	  Kogure	  T,	  Schmittgen	  T,	  Patel	  T.	  2011.	  The	  role	  of	  microRNAs	  in	  human	  liver	  cancers.	  Semin	  Oncol	  38:752-­‐763.	  87.	   Shwetha	  S,	  Gouthamchandra	  K,	  Chandra	  M,	  Ravishankar	  B,	  Khaja	  MN,	  Das	  S.	  2013.	  Circulating	  miRNA	  profile	  in	  HCV	  infected	  serum:	  novel	  insight	  into	  pathogenesis.	  Sci	  Rep	  3:1555.	  88.	   Yang	  Z,	  Cappello	  T,	  Wang	  L.	  2015.	  Emerging	  role	  of	  microRNAs	  in	  lipid	  metabolism.	  Acta	  Pharm	  Sin	  B	  5:145-­‐150.	  89.	   Krutzfeldt	  J,	  Rajewsky	  N,	  Braich	  R,	  Rajeev	  KG,	  Tuschl	  T,	  Manoharan	  M,	  Stoffel	  M.	  2005.	  Silencing	  of	  microRNAs	  in	  vivo	  with	  'antagomirs'.	  Nature	  438:685-­‐689.	  90.	   Esau	  C,	  Davis	  S,	  Murray	  SF,	  Yu	  XX,	  Pandey	  SK,	  Pear	  M,	  Watts	  L,	  Booten	  SL,	  Graham	  M,	  McKay	  R,	  Subramaniam	  A,	  Propp	  S,	  Lollo	  BA,	  Freier	  S,	  Bennett	  CF,	  Bhanot	  S,	  Monia	  BP.	  2006.	  miR-­‐122	  regulation	  of	  lipid	  metabolism	  revealed	  by	  in	  vivo	  antisense	  targeting.	  Cell	  Metab	  3:87-­‐98.	  91.	   Lanford	  RE,	  Hildebrandt-­‐Eriksen	  ES,	  Petri	  A,	  Persson	  R,	  Lindow	  M,	  Munk	  ME,	  Kauppinen	  S,	  Orum	  H.	  2010.	  Therapeutic	  silencing	  of	  microRNA-­‐122	  in	  primates	  with	  chronic	  hepatitis	  C	  virus	  infection.	  Science	  327:198-­‐201.	  92.	   Tsai	  WC,	  Hsu	  SD,	  Hsu	  CS,	  Lai	  TC,	  Chen	  SJ,	  Shen	  R,	  Huang	  Y,	  Chen	  HC,	  Lee	  CH,	  Tsai	  TF,	  Hsu	  MT,	  Wu	  JC,	  Huang	  HD,	  Shiao	  MS,	  Hsiao	  M,	  Tsou	  AP.	  2012.	  MicroRNA-­‐122	  plays	  a	  critical	  role	  in	  liver	  homeostasis	  and	  hepatocarcinogenesis.	  J	  Clin	  Invest	  122:2884-­‐2897.	  93.	   Hsu	  SH,	  Wang	  B,	  Kota	  J,	  Yu	  J,	  Costinean	  S,	  Kutay	  H,	  Yu	  L,	  Bai	  S,	  La	  Perle	  K,	  Chivukula	  RR,	  Mao	  H,	  Wei	  M,	  Clark	  KR,	  Mendell	  JR,	  Caligiuri	  MA,	  Jacob	  ST,	  Mendell	  JT,	  Ghoshal	  K.	  2012.	  Essential	  metabolic,	  anti-­‐inflammatory,	  and	  anti-­‐tumorigenic	  functions	  of	  miR-­‐122	  in	  liver.	  J	  Clin	  Invest	  122:2871-­‐2883.	  94.	   Sun	  X,	  Feinberg	  MW.	  2014.	  MicroRNA-­‐management	  of	  lipoprotein	  homeostasis.	  Circ	  Res	  115:2-­‐6.	  	   129	  95.	   Najafi-­‐Shoushtari	  SH,	  Kristo	  F,	  Li	  Y,	  Shioda	  T,	  Cohen	  DE,	  Gerszten	  RE,	  Naar	  AM.	  2010.	  MicroRNA-­‐33	  and	  the	  SREBP	  host	  genes	  cooperate	  to	  control	  cholesterol	  homeostasis.	  Science	  328:1566-­‐1569.	  96.	   Rayner	  KJ,	  Suarez	  Y,	  Davalos	  A,	  Parathath	  S,	  Fitzgerald	  ML,	  Tamehiro	  N,	  Fisher	  EA,	  Moore	  KJ,	  Fernandez-­‐Hernando	  C.	  2010.	  MiR-­‐33	  contributes	  to	  the	  regulation	  of	  cholesterol	  homeostasis.	  Science	  328:1570-­‐1573.	  97.	   Marquart	  TJ,	  Allen	  RM,	  Ory	  DS,	  Baldan	  A.	  2010.	  miR-­‐33	  links	  SREBP-­‐2	  induction	  to	  repression	  of	  sterol	  transporters.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  107:12228-­‐12232.	  98.	   Ji	  J,	  Zhang	  J,	  Huang	  G,	  Qian	  J,	  Wang	  X,	  Mei	  S.	  2009.	  Over-­‐expressed	  microRNA-­‐27a	  and	  27b	  influence	  fat	  accumulation	  and	  cell	  proliferation	  during	  rat	  hepatic	  stellate	  cell	  activation.	  FEBS	  Lett	  583:759-­‐766.	  99.	   Wang	  T,	  Li	  M,	  Guan	  J,	  Li	  P,	  Wang	  H,	  Guo	  Y,	  Shuai	  S,	  Li	  X.	  2011.	  MicroRNAs	  miR-­‐27a	  and	  miR-­‐143	  Regulate	  Porcine	  Adipocyte	  Lipid	  Metabolism.	  Int	  J	  Mol	  Sci	  12:7950-­‐7959.	  100.	   Shirasaki	  T,	  Honda	  M,	  Shimakami	  T,	  Horii	  R,	  Yamashita	  T,	  Sakai	  Y,	  Sakai	  A,	  Okada	  H,	  Watanabe	  R,	  Murakami	  S,	  Yi	  M,	  Lemon	  SM,	  Kaneko	  S.	  2013.	  MicroRNA-­‐27a	  regulates	  lipid	  metabolism	  and	  inhibits	  hepatitis	  C	  virus	  replication	  in	  human	  hepatoma	  cells.	  J	  Virol	  87:5270-­‐5286.	  101.	   Ng	  R,	  Wu	  H,	  Xiao	  H,	  Chen	  X,	  Willenbring	  H,	  Steer	  CJ,	  Song	  G.	  2014.	  Inhibition	  of	  miR-­‐24	  expression	  in	  liver	  prevents	  hepatic	  lipid	  accumulation	  and	  hyperlipidemia.	  Hepatology	  doi:10.1002/hep.27153.	  102.	   Cheung	  O,	  Puri	  P,	  Eicken	  C,	  Contos	  MJ,	  Mirshahi	  F,	  Maher	  JW,	  Kellum	  JM,	  Min	  H,	  Luketic	  VA,	  Sanyal	  AJ.	  2008.	  Nonalcoholic	  steatohepatitis	  is	  associated	  with	  altered	  hepatic	  MicroRNA	  expression.	  Hepatology	  48:1810-­‐1820.	  103.	   Smith	  EM,	  Zhang	  Y,	  Baye	  TM,	  Gawrieh	  S,	  Cole	  R,	  Blangero	  J,	  Carless	  MA,	  Curran	  JE,	  Dyer	  TD,	  Abraham	  LJ,	  Moses	  EK,	  Kissebah	  AH,	  Martin	  LJ,	  Olivier	  M.	  2010.	  INSIG1	  influences	  obesity-­‐related	  hypertriglyceridemia	  in	  humans.	  J	  Lipid	  Res	  51:701-­‐708.	  104.	   Jeon	  TI,	  Osborne	  TF.	  2016.	  miRNA	  and	  cholesterol	  homeostasis.	  Biochim	  Biophys	  Acta	  doi:10.1016/j.bbalip.2016.01.005.	  105.	   Zhuang	  G,	  Meng	  C,	  Guo	  X,	  Cheruku	  PS,	  Shi	  L,	  Xu	  H,	  Li	  H,	  Wang	  G,	  Evans	  AR,	  Safe	  S,	  Wu	  C,	  Zhou	  B.	  2012.	  A	  novel	  regulator	  of	  macrophage	  activation:	  miR-­‐223	  in	  obesity-­‐associated	  adipose	  tissue	  inflammation.	  Circulation	  125:2892-­‐2903.	  106.	   Ismail	  N,	  Wang	  Y,	  Dakhlallah	  D,	  Moldovan	  L,	  Agarwal	  K,	  Batte	  K,	  Shah	  P,	  Wisler	  J,	  Eubank	  TD,	  Tridandapani	  S,	  Paulaitis	  ME,	  Piper	  MG,	  Marsh	  CB.	  2013.	  Macrophage	  microvesicles	  induce	  macrophage	  differentiation	  and	  miR-­‐223	  transfer.	  Blood	  121:984-­‐995.	  107.	   Vickers	  KC,	  Landstreet	  SR,	  Levin	  MG,	  Shoucri	  BM,	  Toth	  CL,	  Taylor	  RC,	  Palmisano	  BT,	  Tabet	  F,	  Cui	  HL,	  Rye	  KA,	  Sethupathy	  P,	  Remaley	  AT.	  2014.	  MicroRNA-­‐223	  coordinates	  cholesterol	  homeostasis.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  111:14518-­‐14523.	  108.	   Deiuliis	  JA.	  2016.	  MicroRNAs	  as	  regulators	  of	  metabolic	  disease:	  pathophysiologic	  significance	  and	  emerging	  role	  as	  biomarkers	  and	  therapeutics.	  Int	  J	  Obes	  (Lond)	  40:88-­‐101.	  	   130	  109.	   Gottwein	  E,	  Cullen	  BR.	  2008.	  Viral	  and	  cellular	  microRNAs	  as	  determinants	  of	  viral	  pathogenesis	  and	  immunity.	  Cell	  Host	  Microbe	  3:375-­‐387.	  110.	   Skalsky	  RL,	  Cullen	  BR.	  2010.	  Viruses,	  microRNAs,	  and	  host	  interactions.	  Annu	  Rev	  Microbiol	  64:123-­‐141.	  111.	   Pfeffer	  S,	  Zavolan	  M,	  Grasser	  FA,	  Chien	  M,	  Russo	  JJ,	  Ju	  J,	  John	  B,	  Enright	  AJ,	  Marks	  D,	  Sander	  C,	  Tuschl	  T.	  2004.	  Identification	  of	  virus-­‐encoded	  microRNAs.	  Science	  304:734-­‐736.	  112.	   Cullen	  BR.	  2013.	  MicroRNAs	  as	  mediators	  of	  viral	  evasion	  of	  the	  immune	  system.	  Nat	  Immunol	  14:205-­‐210.	  113.	   Qiu	  J,	  Smith	  P,	  Leahy	  L,	  Thorley-­‐Lawson	  DA.	  2015.	  The	  Epstein-­‐Barr	  virus	  encoded	  BART	  miRNAs	  potentiate	  tumor	  growth	  in	  vivo.	  PLoS	  Pathog	  11:e1004561.	  114.	   Qiu	  J,	  Cosmopoulos	  K,	  Pegtel	  M,	  Hopmans	  E,	  Murray	  P,	  Middeldorp	  J,	  Shapiro	  M,	  Thorley-­‐Lawson	  DA.	  2011.	  A	  novel	  persistence	  associated	  EBV	  miRNA	  expression	  profile	  is	  disrupted	  in	  neoplasia.	  PLoS	  Pathog	  7:e1002193.	  115.	   Cosmopoulos	  K,	  Pegtel	  M,	  Hawkins	  J,	  Moffett	  H,	  Novina	  C,	  Middeldorp	  J,	  Thorley-­‐Lawson	  DA.	  2009.	  Comprehensive	  profiling	  of	  Epstein-­‐Barr	  virus	  microRNAs	  in	  nasopharyngeal	  carcinoma.	  J	  Virol	  83:2357-­‐2367.	  116.	   Choy	  EY,	  Siu	  KL,	  Kok	  KH,	  Lung	  RW,	  Tsang	  CM,	  To	  KF,	  Kwong	  DL,	  Tsao	  SW,	  Jin	  DY.	  2008.	  An	  Epstein-­‐Barr	  virus-­‐encoded	  microRNA	  targets	  PUMA	  to	  promote	  host	  cell	  survival.	  J	  Exp	  Med	  205:2551-­‐2560.	  117.	   Riley	  KJ,	  Rabinowitz	  GS,	  Yario	  TA,	  Luna	  JM,	  Darnell	  RB,	  Steitz	  JA.	  2012.	  EBV	  and	  human	  microRNAs	  co-­‐target	  oncogenic	  and	  apoptotic	  viral	  and	  human	  genes	  during	  latency.	  EMBO	  J	  31:2207-­‐2221.	  118.	   Huang	  J,	  Wang	  F,	  Argyris	  E,	  Chen	  K,	  Liang	  Z,	  Tian	  H,	  Huang	  W,	  Squires	  K,	  Verlinghieri	  G,	  Zhang	  H.	  2007.	  Cellular	  microRNAs	  contribute	  to	  HIV-­‐1	  latency	  in	  resting	  primary	  CD4+	  T	  lymphocytes.	  Nat	  Med	  13:1241-­‐1247.	  119.	   Nathans	  R,	  Chu	  CY,	  Serquina	  AK,	  Lu	  CC,	  Cao	  H,	  Rana	  TM.	  2009.	  Cellular	  microRNA	  and	  P	  bodies	  modulate	  host-­‐HIV-­‐1	  interactions.	  Mol	  Cell	  34:696-­‐709.	  120.	   Swaminathan	  G,	  Navas-­‐Martin	  S,	  Martin-­‐Garcia	  J.	  2014.	  MicroRNAs	  and	  HIV-­‐1	  infection:	  antiviral	  activities	  and	  beyond.	  J	  Mol	  Biol	  426:1178-­‐1197.	  121.	   Ho	  BC,	  Yu	  SL,	  Chen	  JJ,	  Chang	  SY,	  Yan	  BS,	  Hong	  QS,	  Singh	  S,	  Kao	  CL,	  Chen	  HY,	  Su	  KY,	  Li	  KC,	  Cheng	  CL,	  Cheng	  HW,	  Lee	  JY,	  Lee	  CN,	  Yang	  PC.	  2011.	  Enterovirus-­‐induced	  miR-­‐141	  contributes	  to	  shutoff	  of	  host	  protein	  translation	  by	  targeting	  the	  translation	  initiation	  factor	  eIF4E.	  Cell	  Host	  Microbe	  9:58-­‐69.	  122.	   Gower	  E,	  Estes	  C,	  Blach	  S,	  Razavi-­‐Shearer	  K,	  Razavi	  H.	  2014.	  Global	  epidemiology	  and	  genotype	  distribution	  of	  the	  hepatitis	  C	  virus	  infection.	  J	  Hepatol	  61:S45-­‐57.	  123.	   Mohd	  Hanafiah	  K,	  Groeger	  J,	  Flaxman	  AD,	  Wiersma	  ST.	  2013.	  Global	  epidemiology	  of	  hepatitis	  C	  virus	  infection:	  new	  estimates	  of	  age-­‐specific	  antibody	  to	  HCV	  seroprevalence.	  Hepatology	  57:1333-­‐1342.	  124.	   Zaltron	  S,	  Spinetti	  A,	  Biasi	  L,	  Baiguera	  C,	  Castelli	  F.	  2012.	  Chronic	  HCV	  infection:	  epidemiological	  and	  clinical	  relevance.	  BMC	  Infect	  Dis	  12	  Suppl	  2:S2.	  125.	   Fried	  MW,	  Shiffman	  ML,	  Reddy	  KR,	  Smith	  C,	  Marinos	  G,	  Goncales	  FL,	  Jr.,	  Haussinger	  D,	  Diago	  M,	  Carosi	  G,	  Dhumeaux	  D,	  Craxi	  A,	  Lin	  A,	  Hoffman	  J,	  Yu	  	   131	  J.	  2002.	  Peginterferon	  alfa-­‐2a	  plus	  ribavirin	  for	  chronic	  hepatitis	  C	  virus	  infection.	  N	  Engl	  J	  Med	  347:975-­‐982.	  126.	   Dore	  GJ,	  Feld	  JJ.	  2015.	  Hepatitis	  C	  virus	  therapeutic	  development:	  in	  pursuit	  of	  "perfectovir".	  Clin	  Infect	  Dis	  60:1829-­‐1836.	  127.	   Chung	  RT,	  Baumert	  TF.	  2014.	  Curing	  chronic	  hepatitis	  C-­‐-­‐the	  arc	  of	  a	  medical	  triumph.	  N	  Engl	  J	  Med	  370:1576-­‐1578.	  128.	   Choo	  QL,	  Kuo	  G,	  Weiner	  AJ,	  Overby	  LR,	  Bradley	  DW,	  Houghton	  M.	  1989.	  Isolation	  of	  a	  cDNA	  clone	  derived	  from	  a	  blood-­‐borne	  non-­‐A,	  non-­‐B	  viral	  hepatitis	  genome.	  Science	  244:359-­‐362.	  129.	   Cornberg	  M,	  Razavi	  HA,	  Alberti	  A,	  Bernasconi	  E,	  Buti	  M,	  Cooper	  C,	  Dalgard	  O,	  Dillion	  JF,	  Flisiak	  R,	  Forns	  X,	  Frankova	  S,	  Goldis	  A,	  Goulis	  I,	  Halota	  W,	  Hunyady	  B,	  Lagging	  M,	  Largen	  A,	  Makara	  M,	  Manolakopoulos	  S,	  Marcellin	  P,	  Marinho	  RT,	  Pol	  S,	  Poynard	  T,	  Puoti	  M,	  Sagalova	  O,	  Sibbel	  S,	  Simon	  K,	  Wallace	  C,	  Young	  K,	  Yurdaydin	  C,	  Zuckerman	  E,	  Negro	  F,	  Zeuzem	  S.	  2011.	  A	  systematic	  review	  of	  hepatitis	  C	  virus	  epidemiology	  in	  Europe,	  Canada	  and	  Israel.	  Liver	  Int	  31	  Suppl	  2:30-­‐60.	  130.	   Frank	  C,	  Mohamed	  MK,	  Strickland	  GT,	  Lavanchy	  D,	  Arthur	  RR,	  Magder	  LS,	  El	  Khoby	  T,	  Abdel-­‐Wahab	  Y,	  Aly	  Ohn	  ES,	  Anwar	  W,	  Sallam	  I.	  2000.	  The	  role	  of	  parenteral	  antischistosomal	  therapy	  in	  the	  spread	  of	  hepatitis	  C	  virus	  in	  Egypt.	  Lancet	  355:887-­‐891.	  131.	   Martell	  M,	  Esteban	  JI,	  Quer	  J,	  Genesca	  J,	  Weiner	  A,	  Esteban	  R,	  Guardia	  J,	  Gomez	  J.	  1992.	  Hepatitis	  C	  virus	  (HCV)	  circulates	  as	  a	  population	  of	  different	  but	  closely	  related	  genomes:	  quasispecies	  nature	  of	  HCV	  genome	  distribution.	  J	  Virol	  66:3225-­‐3229.	  132.	   Fishman	  SL,	  Branch	  AD.	  2009.	  The	  quasispecies	  nature	  and	  biological	  implications	  of	  the	  hepatitis	  C	  virus.	  Infect	  Genet	  Evol	  9:1158-­‐1167.	  133.	   Lauring	  AS,	  Andino	  R.	  2010.	  Quasispecies	  theory	  and	  the	  behavior	  of	  RNA	  viruses.	  PLoS	  Pathog	  6:e1001005.	  134.	   Alter	  HJ,	  Seeff	  LB.	  2000.	  Recovery,	  persistence,	  and	  sequelae	  in	  hepatitis	  C	  virus	  infection:	  a	  perspective	  on	  long-­‐term	  outcome.	  Semin	  Liver	  Dis	  20:17-­‐35.	  135.	   Ascione	  A,	  Tartaglione	  T,	  Di	  Costanzo	  GG.	  2007.	  Natural	  history	  of	  chronic	  hepatitis	  C	  virus	  infection.	  Dig	  Liver	  Dis	  39	  Suppl	  1:S4-­‐7.	  136.	   Thein	  HH,	  Yi	  Q,	  Dore	  GJ,	  Krahn	  MD.	  2008.	  Estimation	  of	  stage-­‐specific	  fibrosis	  progression	  rates	  in	  chronic	  hepatitis	  C	  virus	  infection:	  a	  meta-­‐analysis	  and	  meta-­‐regression.	  Hepatology	  48:418-­‐431.	  137.	   Thein	  HH,	  Yi	  Q,	  Dore	  GJ,	  Krahn	  MD.	  2008.	  Natural	  history	  of	  hepatitis	  C	  virus	  infection	  in	  HIV-­‐infected	  individuals	  and	  the	  impact	  of	  HIV	  in	  the	  era	  of	  highly	  active	  antiretroviral	  therapy:	  a	  meta-­‐analysis.	  AIDS	  22:1979-­‐1991.	  138.	   Asselah	  T,	  Rubbia-­‐Brandt	  L,	  Marcellin	  P,	  Negro	  F.	  2006.	  Steatosis	  in	  chronic	  hepatitis	  C:	  why	  does	  it	  really	  matter?	  Gut	  55:123-­‐130.	  139.	   Kwiterovich	  PO,	  Jr.	  2000.	  The	  metabolic	  pathways	  of	  high-­‐density	  lipoprotein,	  low-­‐density	  lipoprotein,	  and	  triglycerides:	  a	  current	  review.	  Am	  J	  Cardiol	  86:5L-­‐10L.	  140.	   Hourioux	  C,	  Patient	  R,	  Morin	  A,	  Blanchard	  E,	  Moreau	  A,	  Trassard	  S,	  Giraudeau	  B,	  Roingeard	  P.	  2007.	  The	  genotype	  3-­‐specific	  hepatitis	  C	  virus	  core	  	   132	  protein	  residue	  phenylalanine	  164	  increases	  steatosis	  in	  an	  in	  vitro	  cellular	  model.	  Gut	  56:1302-­‐1308.	  141.	   Kumar	  D,	  Farrell	  GC,	  Fung	  C,	  George	  J.	  2002.	  Hepatitis	  C	  virus	  genotype	  3	  is	  cytopathic	  to	  hepatocytes:	  Reversal	  of	  hepatic	  steatosis	  after	  sustained	  therapeutic	  response.	  Hepatology	  36:1266-­‐1272.	  142.	   Adinolfi	  LE,	  Utili	  R,	  Ruggiero	  G.	  1999.	  Body	  composition	  and	  hepatic	  steatosis	  as	  precursors	  of	  fibrosis	  in	  chronic	  hepatitis	  C	  patients.	  Hepatology	  30:1530-­‐1531.	  143.	   Hourigan	  LF,	  Macdonald	  GA,	  Purdie	  D,	  Whitehall	  VH,	  Shorthouse	  C,	  Clouston	  A,	  Powell	  EE.	  1999.	  Fibrosis	  in	  chronic	  hepatitis	  C	  correlates	  significantly	  with	  body	  mass	  index	  and	  steatosis.	  Hepatology	  29:1215-­‐1219.	  144.	   Antunez	  I,	  Aponte	  N,	  Fernandez-­‐Carbia	  A,	  Rodriguez-­‐Perez	  F,	  Toro	  DH.	  2004.	  Steatosis	  as	  a	  predictive	  factor	  for	  treatment	  response	  in	  patients	  with	  chronic	  hepatitis	  C.	  P	  R	  Health	  Sci	  J	  23:57-­‐60.	  145.	   Soresi	  M,	  Tripi	  S,	  Franco	  V,	  Giannitrapani	  L,	  Alessandri	  A,	  Rappa	  F,	  Vuturo	  O,	  Montalto	  G.	  2006.	  Impact	  of	  liver	  steatosis	  on	  the	  antiviral	  response	  in	  the	  hepatitis	  C	  virus-­‐associated	  chronic	  hepatitis.	  Liver	  Int	  26:1119-­‐1125.	  146.	   Neuman	  MG,	  Benhamou	  JP,	  Malkiewicz	  IM,	  Ibrahim	  A,	  Valla	  DC,	  Martinot-­‐Peignoux	  M,	  Asselah	  T,	  Bourliere	  M,	  Katz	  GG,	  Shear	  NH,	  Marcellin	  P.	  2002.	  Kinetics	  of	  serum	  cytokines	  reflect	  changes	  in	  the	  severity	  of	  chronic	  hepatitis	  C	  presenting	  minimal	  fibrosis.	  J	  Viral	  Hepat	  9:134-­‐140.	  147.	   Halse	  R,	  Pearson	  SL,	  McCormack	  JG,	  Yeaman	  SJ,	  Taylor	  R.	  2001.	  Effects	  of	  tumor	  necrosis	  factor-­‐alpha	  on	  insulin	  action	  in	  cultured	  human	  muscle	  cells.	  Diabetes	  50:1102-­‐1109.	  148.	   Uysal	  KT,	  Wiesbrock	  SM,	  Marino	  MW,	  Hotamisligil	  GS.	  1997.	  Protection	  from	  obesity-­‐induced	  insulin	  resistance	  in	  mice	  lacking	  TNF-­‐alpha	  function.	  Nature	  389:610-­‐614.	  149.	   Serfaty	  L,	  Andreani	  T,	  Giral	  P,	  Carbonell	  N,	  Chazouilleres	  O,	  Poupon	  R.	  2001.	  Hepatitis	  C	  virus	  induced	  hypobetalipoproteinemia:	  a	  possible	  mechanism	  for	  steatosis	  in	  chronic	  hepatitis	  C.	  J	  Hepatol	  34:428-­‐434.	  150.	   Petit	  JM,	  Benichou	  M,	  Duvillard	  L,	  Jooste	  V,	  Bour	  JB,	  Minello	  A,	  Verges	  B,	  Brun	  JM,	  Gambert	  P,	  Hillon	  P.	  2003.	  Hepatitis	  C	  virus-­‐associated	  hypobetalipoproteinemia	  is	  correlated	  with	  plasma	  viral	  load,	  steatosis,	  and	  liver	  fibrosis.	  Am	  J	  Gastroenterol	  98:1150-­‐1154.	  151.	   Dubuisson	  J.	  2007.	  Hepatitis	  C	  virus	  proteins.	  World	  J	  Gastroenterol	  13:2406-­‐2415.	  152.	   McLauchlan	  J,	  Lemberg	  MK,	  Hope	  G,	  Martoglio	  B.	  2002.	  Intramembrane	  proteolysis	  promotes	  trafficking	  of	  hepatitis	  C	  virus	  core	  protein	  to	  lipid	  droplets.	  EMBO	  J	  21:3980-­‐3988.	  153.	   Bartosch	  B,	  Cosset	  FL.	  2006.	  Cell	  entry	  of	  hepatitis	  C	  virus.	  Virology	  348:1-­‐12.	  154.	   Gentzsch	  J,	  Brohm	  C,	  Steinmann	  E,	  Friesland	  M,	  Menzel	  N,	  Vieyres	  G,	  Perin	  PM,	  Frentzen	  A,	  Kaderali	  L,	  Pietschmann	  T.	  2013.	  hepatitis	  c	  Virus	  p7	  is	  critical	  for	  capsid	  assembly	  and	  envelopment.	  PLoS	  Pathog	  9:e1003355.	  155.	   Lindenbach	  BD,	  Rice	  CM.	  2005.	  Unravelling	  hepatitis	  C	  virus	  replication	  from	  genome	  to	  function.	  Nature	  436:933-­‐938.	  	   133	  156.	   Abdel-­‐Hakeem	  MS,	  Shoukry	  NH.	  2014.	  Protective	  immunity	  against	  hepatitis	  C:	  many	  shades	  of	  gray.	  Front	  Immunol	  5:274.	  157.	   Foy	  E,	  Li	  K,	  Wang	  C,	  Sumpter	  R,	  Jr.,	  Ikeda	  M,	  Lemon	  SM,	  Gale	  M,	  Jr.	  2003.	  Regulation	  of	  interferon	  regulatory	  factor-­‐3	  by	  the	  hepatitis	  C	  virus	  serine	  protease.	  Science	  300:1145-­‐1148.	  158.	   Meylan	  E,	  Curran	  J,	  Hofmann	  K,	  Moradpour	  D,	  Binder	  M,	  Bartenschlager	  R,	  Tschopp	  J.	  2005.	  Cardif	  is	  an	  adaptor	  protein	  in	  the	  RIG-­‐I	  antiviral	  pathway	  and	  is	  targeted	  by	  hepatitis	  C	  virus.	  Nature	  437:1167-­‐1172.	  159.	   Egger	  D,	  Wolk	  B,	  Gosert	  R,	  Bianchi	  L,	  Blum	  HE,	  Moradpour	  D,	  Bienz	  K.	  2002.	  Expression	  of	  hepatitis	  C	  virus	  proteins	  induces	  distinct	  membrane	  alterations	  including	  a	  candidate	  viral	  replication	  complex.	  J	  Virol	  76:5974-­‐5984.	  160.	   Fridell	  RA,	  Qiu	  D,	  Valera	  L,	  Wang	  C,	  Rose	  RE,	  Gao	  M.	  2011.	  Distinct	  functions	  of	  NS5A	  in	  hepatitis	  C	  virus	  RNA	  replication	  uncovered	  by	  studies	  with	  the	  NS5A	  inhibitor	  BMS-­‐790052.	  J	  Virol	  85:7312-­‐7320.	  161.	   Bartenschlager	  R,	  Frese	  M,	  Pietschmann	  T.	  2004.	  Novel	  insights	  into	  hepatitis	  C	  virus	  replication	  and	  persistence.	  Adv	  Virus	  Res	  63:71-­‐180.	  162.	   Scheel	  TK,	  Rice	  CM.	  2013.	  Understanding	  the	  hepatitis	  C	  virus	  life	  cycle	  paves	  the	  way	  for	  highly	  effective	  therapies.	  Nat	  Med	  19:837-­‐849.	  163.	   Bartenschlager	  R,	  Penin	  F,	  Lohmann	  V,	  Andre	  P.	  2011.	  Assembly	  of	  infectious	  hepatitis	  C	  virus	  particles.	  Trends	  Microbiol	  19:95-­‐103.	  164.	   Barth	  H,	  Schafer	  C,	  Adah	  MI,	  Zhang	  F,	  Linhardt	  RJ,	  Toyoda	  H,	  Kinoshita-­‐Toyoda	  A,	  Toida	  T,	  Van	  Kuppevelt	  TH,	  Depla	  E,	  Von	  Weizsacker	  F,	  Blum	  HE,	  Baumert	  TF.	  2003.	  Cellular	  binding	  of	  hepatitis	  C	  virus	  envelope	  glycoprotein	  E2	  requires	  cell	  surface	  heparan	  sulfate.	  J	  Biol	  Chem	  278:41003-­‐41012.	  165.	   Agnello	  V,	  Abel	  G,	  Elfahal	  M,	  Knight	  GB,	  Zhang	  QX.	  1999.	  Hepatitis	  C	  virus	  and	  other	  flaviviridae	  viruses	  enter	  cells	  via	  low	  density	  lipoprotein	  receptor.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  96:12766-­‐12771.	  166.	   Scarselli	  E,	  Ansuini	  H,	  Cerino	  R,	  Roccasecca	  RM,	  Acali	  S,	  Filocamo	  G,	  Traboni	  C,	  Nicosia	  A,	  Cortese	  R,	  Vitelli	  A.	  2002.	  The	  human	  scavenger	  receptor	  class	  B	  type	  I	  is	  a	  novel	  candidate	  receptor	  for	  the	  hepatitis	  C	  virus.	  EMBO	  J	  21:5017-­‐5025.	  167.	   Pileri	  P,	  Uematsu	  Y,	  Campagnoli	  S,	  Galli	  G,	  Falugi	  F,	  Petracca	  R,	  Weiner	  AJ,	  Houghton	  M,	  Rosa	  D,	  Grandi	  G,	  Abrignani	  S.	  1998.	  Binding	  of	  hepatitis	  C	  virus	  to	  CD81.	  Science	  282:938-­‐941.	  168.	   Evans	  MJ,	  von	  Hahn	  T,	  Tscherne	  DM,	  Syder	  AJ,	  Panis	  M,	  Wolk	  B,	  Hatziioannou	  T,	  McKeating	  JA,	  Bieniasz	  PD,	  Rice	  CM.	  2007.	  Claudin-­‐1	  is	  a	  hepatitis	  C	  virus	  co-­‐receptor	  required	  for	  a	  late	  step	  in	  entry.	  Nature	  446:801-­‐805.	  169.	   Ploss	  A,	  Evans	  MJ,	  Gaysinskaya	  VA,	  Panis	  M,	  You	  H,	  de	  Jong	  YP,	  Rice	  CM.	  2009.	  Human	  occludin	  is	  a	  hepatitis	  C	  virus	  entry	  factor	  required	  for	  infection	  of	  mouse	  cells.	  Nature	  457:882-­‐886.	  170.	   Sainz	  B,	  Jr.,	  Barretto	  N,	  Martin	  DN,	  Hiraga	  N,	  Imamura	  M,	  Hussain	  S,	  Marsh	  KA,	  Yu	  X,	  Chayama	  K,	  Alrefai	  WA,	  Uprichard	  SL.	  2012.	  Identification	  of	  the	  Niemann-­‐Pick	  C1-­‐like	  1	  cholesterol	  absorption	  receptor	  as	  a	  new	  hepatitis	  C	  virus	  entry	  factor.	  Nat	  Med	  18:281-­‐285.	  	   134	  171.	   Blanchard	  E,	  Belouzard	  S,	  Goueslain	  L,	  Wakita	  T,	  Dubuisson	  J,	  Wychowski	  C,	  Rouille	  Y.	  2006.	  Hepatitis	  C	  virus	  entry	  depends	  on	  clathrin-­‐mediated	  endocytosis.	  J	  Virol	  80:6964-­‐6972.	  172.	   Tscherne	  DM,	  Jones	  CT,	  Evans	  MJ,	  Lindenbach	  BD,	  McKeating	  JA,	  Rice	  CM.	  2006.	  Time-­‐	  and	  temperature-­‐dependent	  activation	  of	  hepatitis	  C	  virus	  for	  low-­‐pH-­‐triggered	  entry.	  J	  Virol	  80:1734-­‐1741.	  173.	   Hoffman	  B,	  Liu	  Q.	  2011.	  Hepatitis	  C	  viral	  protein	  translation:	  mechanisms	  and	  implications	  in	  developing	  antivirals.	  Liver	  Int	  31:1449-­‐1467.	  174.	   Watashi	  K,	  Hijikata	  M,	  Hosaka	  M,	  Yamaji	  M,	  Shimotohno	  K.	  2003.	  Cyclosporin	  A	  suppresses	  replication	  of	  hepatitis	  C	  virus	  genome	  in	  cultured	  hepatocytes.	  Hepatology	  38:1282-­‐1288.	  175.	   Yang	  F,	  Robotham	  JM,	  Nelson	  HB,	  Irsigler	  A,	  Kenworthy	  R,	  Tang	  H.	  2008.	  Cyclophilin	  A	  is	  an	  essential	  cofactor	  for	  hepatitis	  C	  virus	  infection	  and	  the	  principal	  mediator	  of	  cyclosporine	  resistance	  in	  vitro.	  J	  Virol	  82:5269-­‐5278.	  176.	   Jopling	  CL,	  Yi	  M,	  Lancaster	  AM,	  Lemon	  SM,	  Sarnow	  P.	  2005.	  Modulation	  of	  hepatitis	  C	  virus	  RNA	  abundance	  by	  a	  liver-­‐specific	  MicroRNA.	  Science	  309:1577-­‐1581.	  177.	   Machlin	  ES,	  Sarnow	  P,	  Sagan	  SM.	  2011.	  Masking	  the	  5'	  terminal	  nucleotides	  of	  the	  hepatitis	  C	  virus	  genome	  by	  an	  unconventional	  microRNA-­‐target	  RNA	  complex.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  108:3193-­‐3198.	  178.	   Li	  Y,	  Masaki	  T,	  Yamane	  D,	  McGivern	  DR,	  Lemon	  SM.	  2013.	  Competing	  and	  noncompeting	  activities	  of	  miR-­‐122	  and	  the	  5'	  exonuclease	  Xrn1	  in	  regulation	  of	  hepatitis	  C	  virus	  replication.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  110:1881-­‐1886.	  179.	   Shimakami	  T,	  Yamane	  D,	  Jangra	  RK,	  Kempf	  BJ,	  Spaniel	  C,	  Barton	  DJ,	  Lemon	  SM.	  2012.	  Stabilization	  of	  hepatitis	  C	  virus	  RNA	  by	  an	  Ago2-­‐miR-­‐122	  complex.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  109:941-­‐946.	  180.	   Jangra	  RK,	  Yi	  M,	  Lemon	  SM.	  2010.	  Regulation	  of	  hepatitis	  C	  virus	  translation	  and	  infectious	  virus	  production	  by	  the	  microRNA	  miR-­‐122.	  J	  Virol	  84:6615-­‐6625.	  181.	   Herker	  E,	  Harris	  C,	  Hernandez	  C,	  Carpentier	  A,	  Kaehlcke	  K,	  Rosenberg	  AR,	  Farese	  RV,	  Jr.,	  Ott	  M.	  2010.	  Efficient	  hepatitis	  C	  virus	  particle	  formation	  requires	  diacylglycerol	  acyltransferase-­‐1.	  Nat	  Med	  16:1295-­‐1298.	  182.	   Miyanari	  Y,	  Atsuzawa	  K,	  Usuda	  N,	  Watashi	  K,	  Hishiki	  T,	  Zayas	  M,	  Bartenschlager	  R,	  Wakita	  T,	  Hijikata	  M,	  Shimotohno	  K.	  2007.	  The	  lipid	  droplet	  is	  an	  important	  organelle	  for	  hepatitis	  C	  virus	  production.	  Nat	  Cell	  Biol	  9:1089-­‐1097.	  183.	   Hope	  RG,	  McLauchlan	  J.	  2000.	  Sequence	  motifs	  required	  for	  lipid	  droplet	  association	  and	  protein	  stability	  are	  unique	  to	  the	  hepatitis	  C	  virus	  core	  protein.	  J	  Gen	  Virol	  81:1913-­‐1925.	  184.	   Shavinskaya	  A,	  Boulant	  S,	  Penin	  F,	  McLauchlan	  J,	  Bartenschlager	  R.	  2007.	  The	  lipid	  droplet	  binding	  domain	  of	  hepatitis	  C	  virus	  core	  protein	  is	  a	  major	  determinant	  for	  efficient	  virus	  assembly.	  J	  Biol	  Chem	  282:37158-­‐37169.	  185.	   Boulant	  S,	  Targett-­‐Adams	  P,	  McLauchlan	  J.	  2007.	  Disrupting	  the	  association	  of	  hepatitis	  C	  virus	  core	  protein	  with	  lipid	  droplets	  correlates	  with	  a	  loss	  in	  production	  of	  infectious	  virus.	  J	  Gen	  Virol	  88:2204-­‐2213.	  	   135	  186.	   Fukasawa	  M,	  Tanaka	  Y,	  Sato	  S,	  Ono	  Y,	  Nitahara-­‐Kasahara	  Y,	  Suzuki	  T,	  Miyamura	  T,	  Hanada	  K,	  Nishijima	  M.	  2006.	  Enhancement	  of	  de	  novo	  fatty	  acid	  biosynthesis	  in	  hepatic	  cell	  line	  Huh7	  expressing	  hepatitis	  C	  virus	  core	  protein.	  Biol	  Pharm	  Bull	  29:1958-­‐1961.	  187.	   Moriya	  K,	  Yotsuyanagi	  H,	  Shintani	  Y,	  Fujie	  H,	  Ishibashi	  K,	  Matsuura	  Y,	  Miyamura	  T,	  Koike	  K.	  1997.	  Hepatitis	  C	  virus	  core	  protein	  induces	  hepatic	  steatosis	  in	  transgenic	  mice.	  J	  Gen	  Virol	  78	  (	  Pt	  7):1527-­‐1531.	  188.	   Boulant	  S,	  Douglas	  MW,	  Moody	  L,	  Budkowska	  A,	  Targett-­‐Adams	  P,	  McLauchlan	  J.	  2008.	  Hepatitis	  C	  virus	  core	  protein	  induces	  lipid	  droplet	  redistribution	  in	  a	  microtubule-­‐	  and	  dynein-­‐dependent	  manner.	  Traffic	  9:1268-­‐1282.	  189.	   Huang	  H,	  Sun	  F,	  Owen	  DM,	  Li	  W,	  Chen	  Y,	  Gale	  M,	  Jr.,	  Ye	  J.	  2007.	  Hepatitis	  C	  virus	  production	  by	  human	  hepatocytes	  dependent	  on	  assembly	  and	  secretion	  of	  very	  low-­‐density	  lipoproteins.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  104:5848-­‐5853.	  190.	   Gastaminza	  P,	  Cheng	  G,	  Wieland	  S,	  Zhong	  J,	  Liao	  W,	  Chisari	  FV.	  2008.	  Cellular	  determinants	  of	  hepatitis	  C	  virus	  assembly,	  maturation,	  degradation,	  and	  secretion.	  J	  Virol	  82:2120-­‐2129.	  191.	   Counihan	  NA,	  Rawlinson	  SM,	  Lindenbach	  BD.	  2011.	  Trafficking	  of	  hepatitis	  C	  virus	  core	  protein	  during	  virus	  particle	  assembly.	  PLoS	  Pathog	  7:e1002302.	  192.	   Coller	  KE,	  Heaton	  NS,	  Berger	  KL,	  Cooper	  JD,	  Saunders	  JL,	  Randall	  G.	  2012.	  Molecular	  determinants	  and	  dynamics	  of	  hepatitis	  C	  virus	  secretion.	  PLoS	  Pathog	  8:e1002466.	  193.	   McHutchison	  JG,	  Gordon	  SC,	  Schiff	  ER,	  Shiffman	  ML,	  Lee	  WM,	  Rustgi	  VK,	  Goodman	  ZD,	  Ling	  MH,	  Cort	  S,	  Albrecht	  JK.	  1998.	  Interferon	  alfa-­‐2b	  alone	  or	  in	  combination	  with	  ribavirin	  as	  initial	  treatment	  for	  chronic	  hepatitis	  C.	  Hepatitis	  Interventional	  Therapy	  Group.	  N	  Engl	  J	  Med	  339:1485-­‐1492.	  194.	   Manns	  MP,	  McHutchison	  JG,	  Gordon	  SC,	  Rustgi	  VK,	  Shiffman	  M,	  Reindollar	  R,	  Goodman	  ZD,	  Koury	  K,	  Ling	  M,	  Albrecht	  JK.	  2001.	  Peginterferon	  alfa-­‐2b	  plus	  ribavirin	  compared	  with	  interferon	  alfa-­‐2b	  plus	  ribavirin	  for	  initial	  treatment	  of	  chronic	  hepatitis	  C:	  a	  randomised	  trial.	  Lancet	  358:958-­‐965.	  195.	   Jacobson	  IM,	  McHutchison	  JG,	  Dusheiko	  G,	  Di	  Bisceglie	  AM,	  Reddy	  KR,	  Bzowej	  NH,	  Marcellin	  P,	  Muir	  AJ,	  Ferenci	  P,	  Flisiak	  R,	  George	  J,	  Rizzetto	  M,	  Shouval	  D,	  Sola	  R,	  Terg	  RA,	  Yoshida	  EM,	  Adda	  N,	  Bengtsson	  L,	  Sankoh	  AJ,	  Kieffer	  TL,	  George	  S,	  Kauffman	  RS,	  Zeuzem	  S,	  Team	  AS.	  2011.	  Telaprevir	  for	  previously	  untreated	  chronic	  hepatitis	  C	  virus	  infection.	  N	  Engl	  J	  Med	  364:2405-­‐2416.	  196.	   Poordad	  F,	  McCone	  J,	  Jr.,	  Bacon	  BR,	  Bruno	  S,	  Manns	  MP,	  Sulkowski	  MS,	  Jacobson	  IM,	  Reddy	  KR,	  Goodman	  ZD,	  Boparai	  N,	  DiNubile	  MJ,	  Sniukiene	  V,	  Brass	  CA,	  Albrecht	  JK,	  Bronowicki	  JP,	  Investigators	  S-­‐.	  2011.	  Boceprevir	  for	  untreated	  chronic	  HCV	  genotype	  1	  infection.	  N	  Engl	  J	  Med	  364:1195-­‐1206.	  197.	   Sen	  GC.	  2001.	  Viruses	  and	  interferons.	  Annu	  Rev	  Microbiol	  55:255-­‐281.	  198.	   Feld	  JJ,	  Hoofnagle	  JH.	  2005.	  Mechanism	  of	  action	  of	  interferon	  and	  ribavirin	  in	  treatment	  of	  hepatitis	  C.	  Nature	  436:967-­‐972.	  199.	   Dore	  GJ,	  Ward	  J,	  Thursz	  M.	  2014.	  Hepatitis	  C	  disease	  burden	  and	  strategies	  to	  manage	  the	  burden	  (Guest	  Editors	  Mark	  Thursz,	  Gregory	  Dore	  and	  John	  Ward).	  J	  Viral	  Hepat	  21	  Suppl	  1:1-­‐4.	  	   136	  200.	   Asselah	  T,	  Boyer	  N,	  Saadoun	  D,	  Martinot-­‐Peignoux	  M,	  Marcellin	  P.	  2016.	  Direct-­‐acting	  antivirals	  for	  the	  treatment	  of	  hepatitis	  C	  virus	  infection:	  optimizing	  current	  IFN-­‐free	  treatment	  and	  future	  perspectives.	  Liver	  Int	  36	  Suppl	  1:47-­‐57.	  201.	   Douam	  F,	  Ding	  Q,	  Ploss	  A.	  2016.	  Recent	  advances	  in	  understanding	  hepatitis	  C.	  F1000Res	  5.	  202.	   Poveda	  E,	  Wyles	  DL,	  Mena	  A,	  Pedreira	  JD,	  Castro-­‐Iglesias	  A,	  Cachay	  E.	  2014.	  Update	  on	  hepatitis	  C	  virus	  resistance	  to	  direct-­‐acting	  antiviral	  agents.	  Antiviral	  Res	  108:181-­‐191.	  203.	   Berger	  KL,	  Cooper	  JD,	  Heaton	  NS,	  Yoon	  R,	  Oakland	  TE,	  Jordan	  TX,	  Mateu	  G,	  Grakoui	  A,	  Randall	  G.	  2009.	  Roles	  for	  endocytic	  trafficking	  and	  phosphatidylinositol	  4-­‐kinase	  III	  alpha	  in	  hepatitis	  C	  virus	  replication.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  106:7577-­‐7582.	  204.	   Perlemuter	  G,	  Sabile	  A,	  Letteron	  P,	  Vona	  G,	  Topilco	  A,	  Chretien	  Y,	  Koike	  K,	  Pessayre	  D,	  Chapman	  J,	  Barba	  G,	  Brechot	  C.	  2002.	  Hepatitis	  C	  virus	  core	  protein	  inhibits	  microsomal	  triglyceride	  transfer	  protein	  activity	  and	  very	  low	  density	  lipoprotein	  secretion:	  a	  model	  of	  viral-­‐related	  steatosis.	  FASEB	  J	  16:185-­‐194.	  205.	   Syed	  GH,	  Amako	  Y,	  Siddiqui	  A.	  2010.	  Hepatitis	  C	  virus	  hijacks	  host	  lipid	  metabolism.	  Trends	  Endocrinol	  Metab	  21:33-­‐40.	  206.	   Waris	  G,	  Felmlee	  DJ,	  Negro	  F,	  Siddiqui	  A.	  2007.	  Hepatitis	  C	  virus	  induces	  proteolytic	  cleavage	  of	  sterol	  regulatory	  element	  binding	  proteins	  and	  stimulates	  their	  phosphorylation	  via	  oxidative	  stress.	  J	  Virol	  81:8122-­‐8130.	  207.	   Negro	  F,	  Sanyal	  AJ.	  2009.	  Hepatitis	  C	  virus,	  steatosis	  and	  lipid	  abnormalities:	  clinical	  and	  pathogenic	  data.	  Liver	  Int	  29	  Suppl	  2:26-­‐37.	  208.	   Tardif	  KD,	  Waris	  G,	  Siddiqui	  A.	  2005.	  Hepatitis	  C	  virus,	  ER	  stress,	  and	  oxidative	  stress.	  Trends	  Microbiol	  13:159-­‐163.	  209.	   Wilson	  JA,	  Sagan	  SM.	  2014.	  Hepatitis	  C	  virus	  and	  human	  miR-­‐122:	  insights	  from	  the	  bench	  to	  the	  clinic.	  Curr	  Opin	  Virol	  7C:11-­‐18.	  210.	   Pedersen	  IM,	  Cheng	  G,	  Wieland	  S,	  Volinia	  S,	  Croce	  CM,	  Chisari	  FV,	  David	  M.	  2007.	  Interferon	  modulation	  of	  cellular	  microRNAs	  as	  an	  antiviral	  mechanism.	  Nature	  449:919-­‐922.	  211.	   Cheng	  JC,	  Yeh	  YJ,	  Tseng	  CP,	  Hsu	  SD,	  Chang	  YL,	  Sakamoto	  N,	  Huang	  HD.	  2012.	  Let-­‐7b	  is	  a	  novel	  regulator	  of	  hepatitis	  C	  virus	  replication.	  Cell	  Mol	  Life	  Sci	  69:2621-­‐2633.	  212.	   Murakami	  Y,	  Aly	  HH,	  Tajima	  A,	  Inoue	  I,	  Shimotohno	  K.	  2009.	  Regulation	  of	  the	  hepatitis	  C	  virus	  genome	  replication	  by	  miR-­‐199a.	  J	  Hepatol	  50:453-­‐460.	  213.	   Singaravelu	  R,	  Russell	  RS,	  Tyrrell	  DL,	  Pezacki	  JP.	  2014.	  Hepatitis	  C	  virus	  and	  microRNAs:	  miRed	  in	  a	  host	  of	  possibilities.	  Curr	  Opin	  Virol	  7:1-­‐10.	  214.	   Murakami	  Y,	  Toyoda	  H,	  Tanaka	  M,	  Kuroda	  M,	  Harada	  Y,	  Matsuda	  F,	  Tajima	  A,	  Kosaka	  N,	  Ochiya	  T,	  Shimotohno	  K.	  2011.	  The	  progression	  of	  liver	  fibrosis	  is	  related	  with	  overexpression	  of	  the	  miR-­‐199	  and	  200	  families.	  PLoS	  One	  6:e16081.	  215.	   Bandyopadhyay	  S,	  Friedman	  RC,	  Marquez	  RT,	  Keck	  K,	  Kong	  B,	  Icardi	  MS,	  Brown	  KE,	  Burge	  CB,	  Schmidt	  WN,	  Wang	  Y,	  McCaffrey	  AP.	  2011.	  Hepatitis	  C	  virus	  infection	  and	  hepatic	  stellate	  cell	  activation	  downregulate	  miR-­‐29:	  miR-­‐29	  	   137	  overexpression	  reduces	  hepatitis	  C	  viral	  abundance	  in	  culture.	  J	  Infect	  Dis	  203:1753-­‐1762.	  216.	   Singaravelu	  R,	  Chen	  R,	  Lyn	  RK,	  Jones	  DM,	  O'Hara	  S,	  Rouleau	  Y,	  Cheng	  J,	  Srinivasan	  P,	  Nasheri	  N,	  Russell	  RS,	  Tyrrell	  DL,	  Pezacki	  JP.	  2014.	  Hepatitis	  C	  virus	  induced	  up-­‐regulation	  of	  microRNA-­‐27:	  a	  novel	  mechanism	  for	  hepatic	  steatosis.	  Hepatology	  59:98-­‐108.	  217.	   Murakami	  Y,	  Tanaka	  M,	  Toyoda	  H,	  Hayashi	  K,	  Kuroda	  M,	  Tajima	  A,	  Shimotohno	  K.	  2010.	  Hepatic	  microRNA	  expression	  is	  associated	  with	  the	  response	  to	  interferon	  treatment	  of	  chronic	  hepatitis	  C.	  BMC	  Med	  Genomics	  3:48.	  218.	   Bhatt	  S,	  Gething	  PW,	  Brady	  OJ,	  Messina	  JP,	  Farlow	  AW,	  Moyes	  CL,	  Drake	  JM,	  Brownstein	  JS,	  Hoen	  AG,	  Sankoh	  O,	  Myers	  MF,	  George	  DB,	  Jaenisch	  T,	  Wint	  GR,	  Simmons	  CP,	  Scott	  TW,	  Farrar	  JJ,	  Hay	  SI.	  2013.	  The	  global	  distribution	  and	  burden	  of	  dengue.	  Nature	  496:504-­‐507.	  219.	   Shepard	  DS,	  Coudeville	  L,	  Halasa	  YA,	  Zambrano	  B,	  Dayan	  GH.	  2011.	  Economic	  impact	  of	  dengue	  illness	  in	  the	  Americas.	  Am	  J	  Trop	  Med	  Hyg	  84:200-­‐207.	  220.	   Guzman	  MG,	  Harris	  E.	  2015.	  Dengue.	  Lancet	  385:453-­‐465.	  221.	   Bernal	  A.	  2016.	  	  Dengvaxia®	  Approved	  in	  El	  Salvador,	  on	  Sanofi	  Pasteur.	  http://www.dengue.info/#overlay=content/dengvaxia-­‐approved-­‐el-­‐salvador.	  Accessed	  March,	  26.	  222.	   Hadinegoro	  SR,	  Arredondo-­‐Garcia	  JL,	  Capeding	  MR,	  Deseda	  C,	  Chotpitayasunondh	  T,	  Dietze	  R,	  Muhammad	  Ismail	  HI,	  Reynales	  H,	  Limkittikul	  K,	  Rivera-­‐Medina	  DM,	  Tran	  HN,	  Bouckenooghe	  A,	  Chansinghakul	  D,	  Cortes	  M,	  Fanouillere	  K,	  Forrat	  R,	  Frago	  C,	  Gailhardou	  S,	  Jackson	  N,	  Noriega	  F,	  Plennevaux	  E,	  Wartel	  TA,	  Zambrano	  B,	  Saville	  M,	  Group	  C-­‐TDVW.	  2015.	  Efficacy	  and	  Long-­‐Term	  Safety	  of	  a	  Dengue	  Vaccine	  in	  Regions	  of	  Endemic	  Disease.	  N	  Engl	  J	  Med	  373:1195-­‐1206.	  223.	   Guy	  B,	  Jackson	  N.	  2016.	  Dengue	  vaccine:	  hypotheses	  to	  understand	  CYD-­‐TDV-­‐induced	  protection.	  Nat	  Rev	  Microbiol	  14:45-­‐54.	  224.	   Diamond	  MS,	  Pierson	  TC.	  2015.	  Molecular	  Insight	  into	  Dengue	  Virus	  Pathogenesis	  and	  Its	  Implications	  for	  Disease	  Control.	  Cell	  162:488-­‐492.	  225.	   Halstead	  SB.	  2007.	  Dengue.	  Lancet	  370:1644-­‐1652.	  226.	   Halstead	  SB,	  Mahalingam	  S,	  Marovich	  MA,	  Ubol	  S,	  Mosser	  DM.	  2010.	  Intrinsic	  antibody-­‐dependent	  enhancement	  of	  microbial	  infection	  in	  macrophages:	  disease	  regulation	  by	  immune	  complexes.	  Lancet	  Infect	  Dis	  10:712-­‐722.	  227.	   Syenina	  A,	  Jagaraj	  CJ,	  Aman	  SA,	  Sridharan	  A,	  St	  John	  AL.	  2015.	  Dengue	  vascular	  leakage	  is	  augmented	  by	  mast	  cell	  degranulation	  mediated	  by	  immunoglobulin	  Fcgamma	  receptors.	  Elife	  4.	  228.	   Welsch	  S,	  Miller	  S,	  Romero-­‐Brey	  I,	  Merz	  A,	  Bleck	  CK,	  Walther	  P,	  Fuller	  SD,	  Antony	  C,	  Krijnse-­‐Locker	  J,	  Bartenschlager	  R.	  2009.	  Composition	  and	  three-­‐dimensional	  architecture	  of	  the	  dengue	  virus	  replication	  and	  assembly	  sites.	  Cell	  Host	  Microbe	  5:365-­‐375.	  229.	   Pierson	  TC,	  Diamond	  MS.	  2012.	  Degrees	  of	  maturity:	  the	  complex	  structure	  and	  biology	  of	  flaviviruses.	  Curr	  Opin	  Virol	  2:168-­‐175.	  	   138	  230.	   Austin	  SK,	  Dowd	  KA,	  Shrestha	  B,	  Nelson	  CA,	  Edeling	  MA,	  Johnson	  S,	  Pierson	  TC,	  Diamond	  MS,	  Fremont	  DH.	  2012.	  Structural	  basis	  of	  differential	  neutralization	  of	  DENV-­‐1	  genotypes	  by	  an	  antibody	  that	  recognizes	  a	  cryptic	  epitope.	  PLoS	  Pathog	  8:e1002930.	  231.	   Villar	  L,	  Dayan	  GH,	  Arredondo-­‐Garcia	  JL,	  Rivera	  DM,	  Cunha	  R,	  Deseda	  C,	  Reynales	  H,	  Costa	  MS,	  Morales-­‐Ramirez	  JO,	  Carrasquilla	  G,	  Rey	  LC,	  Dietze	  R,	  Luz	  K,	  Rivas	  E,	  Miranda	  Montoya	  MC,	  Cortes	  Supelano	  M,	  Zambrano	  B,	  Langevin	  E,	  Boaz	  M,	  Tornieporth	  N,	  Saville	  M,	  Noriega	  F,	  Group	  CYDS.	  2015.	  Efficacy	  of	  a	  tetravalent	  dengue	  vaccine	  in	  children	  in	  Latin	  America.	  N	  Engl	  J	  Med	  372:113-­‐123.	  232.	   Muller	  DA,	  Young	  PR.	  2013.	  The	  flavivirus	  NS1	  protein:	  molecular	  and	  structural	  biology,	  immunology,	  role	  in	  pathogenesis	  and	  application	  as	  a	  diagnostic	  biomarker.	  Antiviral	  Res	  98:192-­‐208.	  233.	   George	  SL,	  Wong	  MA,	  Dube	  TJ,	  Boroughs	  KL,	  Stovall	  JL,	  Luy	  BE,	  Haller	  AA,	  Osorio	  JE,	  Eggemeyer	  LM,	  Irby-­‐Moore	  S,	  Frey	  SE,	  Huang	  CY,	  Stinchcomb	  DT.	  2015.	  Safety	  and	  Immunogenicity	  of	  a	  Live	  Attenuated	  Tetravalent	  Dengue	  Vaccine	  Candidate	  in	  Flavivirus-­‐Naive	  Adults:	  A	  Randomized,	  Double-­‐Blinded	  Phase	  1	  Clinical	  Trial.	  J	  Infect	  Dis	  212:1032-­‐1041.	  234.	   Kirkpatrick	  BD,	  Durbin	  AP,	  Pierce	  KK,	  Carmolli	  MP,	  Tibery	  CM,	  Grier	  PL,	  Hynes	  N,	  Diehl	  SA,	  Elwood	  D,	  Jarvis	  AP,	  Sabundayo	  BP,	  Lyon	  CE,	  Larsson	  CJ,	  Jo	  M,	  Lovchik	  JM,	  Luke	  CJ,	  Walsh	  MC,	  Fraser	  EA,	  Subbarao	  K,	  Whitehead	  SS.	  2015.	  Robust	  and	  Balanced	  Immune	  Responses	  to	  All	  4	  Dengue	  Virus	  Serotypes	  Following	  Administration	  of	  a	  Single	  Dose	  of	  a	  Live	  Attenuated	  Tetravalent	  Dengue	  Vaccine	  to	  Healthy,	  Flavivirus-­‐Naive	  Adults.	  J	  Infect	  Dis	  212:702-­‐710.	  235.	   Moreira	  LA,	  Iturbe-­‐Ormaetxe	  I,	  Jeffery	  JA,	  Lu	  G,	  Pyke	  AT,	  Hedges	  LM,	  Rocha	  BC,	  Hall-­‐Mendelin	  S,	  Day	  A,	  Riegler	  M,	  Hugo	  LE,	  Johnson	  KN,	  Kay	  BH,	  McGraw	  EA,	  van	  den	  Hurk	  AF,	  Ryan	  PA,	  O'Neill	  SL.	  2009.	  A	  Wolbachia	  symbiont	  in	  Aedes	  aegypti	  limits	  infection	  with	  dengue,	  Chikungunya,	  and	  Plasmodium.	  Cell	  139:1268-­‐1278.	  236.	   Hoffmann	  AA,	  Iturbe-­‐Ormaetxe	  I,	  Callahan	  AG,	  Phillips	  BL,	  Billington	  K,	  Axford	  JK,	  Montgomery	  B,	  Turley	  AP,	  O'Neill	  SL.	  2014.	  Stability	  of	  the	  wMel	  Wolbachia	  Infection	  following	  invasion	  into	  Aedes	  aegypti	  populations.	  PLoS	  Negl	  Trop	  Dis	  8:e3115.	  237.	   Franz	  AW,	  Sanchez-­‐Vargas	  I,	  Raban	  RR,	  Black	  WCt,	  James	  AA,	  Olson	  KE.	  2014.	  Fitness	  impact	  and	  stability	  of	  a	  transgene	  conferring	  resistance	  to	  dengue-­‐2	  virus	  following	  introgression	  into	  a	  genetically	  diverse	  Aedes	  aegypti	  strain.	  PLoS	  Negl	  Trop	  Dis	  8:e2833.	  238.	   Whitehorn	  J,	  Yacoub	  S,	  Anders	  KL,	  Macareo	  LR,	  Cassetti	  MC,	  Nguyen	  Van	  VC,	  Shi	  PY,	  Wills	  B,	  Simmons	  CP.	  2014.	  Dengue	  therapeutics,	  chemoprophylaxis,	  and	  allied	  tools:	  state	  of	  the	  art	  and	  future	  directions.	  PLoS	  Negl	  Trop	  Dis	  8:e3025.	  239.	   Krishnan	  MN,	  Garcia-­‐Blanco	  MA.	  2014.	  Targeting	  host	  factors	  to	  treat	  West	  Nile	  and	  dengue	  viral	  infections.	  Viruses	  6:683-­‐708.	  240.	   Zaitseva	  E,	  Yang	  ST,	  Melikov	  K,	  Pourmal	  S,	  Chernomordik	  LV.	  2010.	  Dengue	  virus	  ensures	  its	  fusion	  in	  late	  endosomes	  using	  compartment-­‐specific	  lipids.	  PLoS	  Pathog	  6:e1001131.	  	   139	  241.	   Nour	  AM,	  Li	  Y,	  Wolenski	  J,	  Modis	  Y.	  2013.	  Viral	  membrane	  fusion	  and	  nucleocapsid	  delivery	  into	  the	  cytoplasm	  are	  distinct	  events	  in	  some	  flaviviruses.	  PLoS	  Pathog	  9:e1003585.	  242.	   Rothwell	  C,	  Lebreton	  A,	  Young	  Ng	  C,	  Lim	  JY,	  Liu	  W,	  Vasudevan	  S,	  Labow	  M,	  Gu	  F,	  Gaither	  LA.	  2009.	  Cholesterol	  biosynthesis	  modulation	  regulates	  dengue	  viral	  replication.	  Virology	  389:8-­‐19.	  243.	   Heaton	  NS,	  Perera	  R,	  Berger	  KL,	  Khadka	  S,	  Lacount	  DJ,	  Kuhn	  RJ,	  Randall	  G.	  2010.	  Dengue	  virus	  nonstructural	  protein	  3	  redistributes	  fatty	  acid	  synthase	  to	  sites	  of	  viral	  replication	  and	  increases	  cellular	  fatty	  acid	  synthesis.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  107:17345-­‐17350.	  244.	   Samsa	  MM,	  Mondotte	  JA,	  Iglesias	  NG,	  Assuncao-­‐Miranda	  I,	  Barbosa-­‐Lima	  G,	  Da	  Poian	  AT,	  Bozza	  PT,	  Gamarnik	  AV.	  2009.	  Dengue	  virus	  capsid	  protein	  usurps	  lipid	  droplets	  for	  viral	  particle	  formation.	  PLoS	  Pathog	  5:e1000632.	  245.	   Carvalho	  FA,	  Carneiro	  FA,	  Martins	  IC,	  Assuncao-­‐Miranda	  I,	  Faustino	  AF,	  Pereira	  RM,	  Bozza	  PT,	  Castanho	  MA,	  Mohana-­‐Borges	  R,	  Da	  Poian	  AT,	  Santos	  NC.	  2012.	  Dengue	  virus	  capsid	  protein	  binding	  to	  hepatic	  lipid	  droplets	  (LD)	  is	  potassium	  ion	  dependent	  and	  is	  mediated	  by	  LD	  surface	  proteins.	  J	  Virol	  86:2096-­‐2108.	  246.	   Martins	  IC,	  Gomes-­‐Neto	  F,	  Faustino	  AF,	  Carvalho	  FA,	  Carneiro	  FA,	  Bozza	  PT,	  Mohana-­‐Borges	  R,	  Castanho	  MA,	  Almeida	  FC,	  Santos	  NC,	  Da	  Poian	  AT.	  2012.	  The	  disordered	  N-­‐terminal	  region	  of	  dengue	  virus	  capsid	  protein	  contains	  a	  lipid-­‐droplet-­‐binding	  motif.	  Biochem	  J	  444:405-­‐415.	  247.	   Heaton	  NS,	  Randall	  G.	  2010.	  Dengue	  virus-­‐induced	  autophagy	  regulates	  lipid	  metabolism.	  Cell	  Host	  Microbe	  8:422-­‐432.	  248.	   He	  L,	  Hannon	  GJ.	  2004.	  MicroRNAs:	  small	  RNAs	  with	  a	  big	  role	  in	  gene	  regulation.	  Nat	  Rev	  Genet	  5:522-­‐531.	  249.	   Wittmann	  J,	  Jack	  HM.	  2010.	  Serum	  microRNAs	  as	  powerful	  cancer	  biomarkers.	  Biochim	  Biophys	  Acta	  1806:200-­‐207.	  250.	   Goldstein	  JL,	  DeBose-­‐Boyd	  RA,	  Brown	  MS.	  2006.	  Protein	  sensors	  for	  membrane	  sterols.	  Cell	  124:35-­‐46.	  251.	   LaRosa	  JC,	  Hunninghake	  D,	  Bush	  D,	  Criqui	  MH,	  Getz	  GS,	  Gotto	  AM,	  Jr.,	  Grundy	  SM,	  Rakita	  L,	  Robertson	  RM,	  Weisfeldt	  ML,	  et	  al.	  1990.	  The	  cholesterol	  facts.	  A	  summary	  of	  the	  evidence	  relating	  dietary	  fats,	  serum	  cholesterol,	  and	  coronary	  heart	  disease.	  A	  joint	  statement	  by	  the	  American	  Heart	  Association	  and	  the	  National	  Heart,	  Lung,	  and	  Blood	  Institute.	  The	  Task	  Force	  on	  Cholesterol	  Issues,	  American	  Heart	  Association.	  Circulation	  81:1721-­‐1733.	  252.	   Maxfield	  FR,	  Tabas	  I.	  2005.	  Role	  of	  cholesterol	  and	  lipid	  organization	  in	  disease.	  Nature	  438:612-­‐621.	  253.	   Chen	  SL	  MT.	  2006.	  The	  Natural	  History	  of	  Hepatitis	  C	  Virus	  (HCV)	  Infection.	  Int	  J	  Med	  Sci	  3:47-­‐52.	  254.	   Bassendine	  MF,	  Sheridan	  DA,	  Bridge	  SH,	  Felmlee	  DJ,	  Neely	  RD.	  2013.	  Lipids	  and	  HCV.	  Semin	  Immunopathol	  35:87-­‐100.	  255.	   Felmlee	  DJ,	  Hafirassou	  ML,	  Lefevre	  M,	  Baumert	  TF,	  Schuster	  C.	  2013.	  Hepatitis	  C	  virus,	  cholesterol	  and	  lipoproteins-­‐-­‐impact	  for	  the	  viral	  life	  cycle	  and	  pathogenesis	  of	  liver	  disease.	  Viruses	  5:1292-­‐1324.	  	   140	  256.	   Corey	  KE,	  Kane	  E,	  Munroe	  C,	  Barlow	  LL,	  Zheng	  H,	  Chung	  RT.	  2009.	  Hepatitis	  C	  virus	  infection	  and	  its	  clearance	  alter	  circulating	  lipids:	  implications	  for	  long-­‐term	  follow-­‐up.	  Hepatology	  50:1030-­‐1037.	  257.	   Janssen	  HL,	  Reesink	  HW,	  Lawitz	  EJ,	  Zeuzem	  S,	  Rodriguez-­‐Torres	  M,	  Patel	  K,	  van	  der	  Meer	  AJ,	  Patick	  AK,	  Chen	  A,	  Zhou	  Y,	  Persson	  R,	  King	  BD,	  Kauppinen	  S,	  Levin	  AA,	  Hodges	  MR.	  2013.	  Treatment	  of	  HCV	  infection	  by	  targeting	  microRNA.	  N	  Engl	  J	  Med	  368:1685-­‐1694.	  258.	   Lindow	  M,	  Kauppinen	  S.	  2012.	  Discovering	  the	  first	  microRNA-­‐targeted	  drug.	  J	  Cell	  Biol	  199:407-­‐412.	  259.	   Vickers	  KC,	  Shoucri	  BM,	  Levin	  MG,	  Wu	  H,	  Pearson	  DS,	  Osei-­‐Hwedieh	  D,	  Collins	  FS,	  Remaley	  AT,	  Sethupathy	  P.	  2013.	  MicroRNA-­‐27b	  is	  a	  regulatory	  hub	  in	  lipid	  metabolism	  and	  is	  altered	  in	  dyslipidemia.	  Hepatology	  57:533-­‐542.	  260.	   Michell	  DL,	  Vickers	  KC.	  2016.	  Lipoprotein	  carriers	  of	  microRNAs.	  Biochim	  Biophys	  Acta	  doi:10.1016/j.bbalip.2016.01.011.	  261.	   Willeit	  P,	  Zampetaki	  A,	  Dudek	  K,	  Kaudewitz	  D,	  King	  A,	  Kirkby	  NS,	  Crosby-­‐Nwaobi	  R,	  Prokopi	  M,	  Drozdov	  I,	  Langley	  SR,	  Sivaprasad	  S,	  Markus	  HS,	  Mitchell	  JA,	  Warner	  TD,	  Kiechl	  S,	  Mayr	  M.	  2013.	  Circulating	  microRNAs	  as	  novel	  biomarkers	  for	  platelet	  activation.	  Circ	  Res	  112:595-­‐600.	  262.	   Lin	  ZH,	  Xin	  YN,	  Dong	  QJ,	  Wang	  Q,	  Jiang	  XJ,	  Zhan	  SH,	  Sun	  Y,	  Xuan	  SY.	  2011.	  Performance	  of	  the	  aspartate	  aminotransferase-­‐to-­‐platelet	  ratio	  index	  for	  the	  staging	  of	  hepatitis	  C-­‐related	  fibrosis:	  an	  updated	  meta-­‐analysis.	  Hepatology	  53:726-­‐736.	  263.	   Chou	  R,	  Wasson	  N.	  2013.	  Blood	  tests	  to	  diagnose	  fibrosis	  or	  cirrhosis	  in	  patients	  with	  chronic	  hepatitis	  C	  virus	  infection:	  a	  systematic	  review.	  Ann	  Intern	  Med	  158:807-­‐820.	  264.	   Sterling	  RK,	  Lissen	  E,	  Clumeck	  N,	  Sola	  R,	  Correa	  MC,	  Montaner	  J,	  M	  SS,	  Torriani	  FJ,	  Dieterich	  DT,	  Thomas	  DL,	  Messinger	  D,	  Nelson	  M,	  Investigators	  AC.	  2006.	  Development	  of	  a	  simple	  noninvasive	  index	  to	  predict	  significant	  fibrosis	  in	  patients	  with	  HIV/HCV	  coinfection.	  Hepatology	  43:1317-­‐1325.	  265.	   Shirakawa	  H,	  Matsumoto	  A,	  Joshita	  S,	  Komatsu	  M,	  Tanaka	  N,	  Umemura	  T,	  Ichijo	  T,	  Yoshizawa	  K,	  Kiyosawa	  K,	  Tanaka	  E,	  Nagano	  Interferon	  Treatment	  Research	  G.	  2008.	  Pretreatment	  prediction	  of	  virological	  response	  to	  peginterferon	  plus	  ribavirin	  therapy	  in	  chronic	  hepatitis	  C	  patients	  using	  viral	  and	  host	  factors.	  Hepatology	  48:1753-­‐1760.	  266.	   Martinez	  SM,	  Fernandez-­‐Varo	  G,	  Gonzalez	  P,	  Sampson	  E,	  Bruguera	  M,	  Navasa	  M,	  Jimenez	  W,	  Sanchez-­‐Tapias	  JM,	  Forns	  X.	  2011.	  Assessment	  of	  liver	  fibrosis	  before	  and	  after	  antiviral	  therapy	  by	  different	  serum	  marker	  panels	  in	  patients	  with	  chronic	  hepatitis	  C.	  Aliment	  Pharmacol	  Ther	  33:138-­‐148.	  267.	   Day	  CP,	  James	  OF.	  1998.	  Hepatic	  steatosis:	  innocent	  bystander	  or	  guilty	  party?	  Hepatology	  27:1463-­‐1466.	  268.	   Chang	  J,	  Nicolas	  E,	  Marks	  D,	  Sander	  C,	  Lerro	  A,	  Buendia	  MA,	  Xu	  C,	  Mason	  WS,	  Moloshok	  T,	  Bort	  R,	  Zaret	  KS,	  Taylor	  JM.	  2004.	  miR-­‐122,	  a	  mammalian	  liver-­‐specific	  microRNA,	  is	  processed	  from	  hcr	  mRNA	  and	  may	  downregulate	  the	  high	  affinity	  cationic	  amino	  acid	  transporter	  CAT-­‐1.	  RNA	  Biol	  1:106-­‐113.	  269.	   Elmen	  J,	  Lindow	  M,	  Schutz	  S,	  Lawrence	  M,	  Petri	  A,	  Obad	  S,	  Lindholm	  M,	  Hedtjarn	  M,	  Hansen	  HF,	  Berger	  U,	  Gullans	  S,	  Kearney	  P,	  Sarnow	  P,	  Straarup	  	   141	  EM,	  Kauppinen	  S.	  2008.	  LNA-­‐mediated	  microRNA	  silencing	  in	  non-­‐human	  primates.	  Nature	  452:896-­‐899.	  270.	   Bandiera	  S,	  Pfeffer	  S,	  Baumert	  TF,	  Zeisel	  MB.	  2015.	  miR-­‐122-­‐-­‐a	  key	  factor	  and	  therapeutic	  target	  in	  liver	  disease.	  J	  Hepatol	  62:448-­‐457.	  271.	   Cermelli	  S,	  Ruggieri	  A,	  Marrero	  JA,	  Ioannou	  GN,	  Beretta	  L.	  2011.	  Circulating	  microRNAs	  in	  patients	  with	  chronic	  hepatitis	  C	  and	  non-­‐alcoholic	  fatty	  liver	  disease.	  PLoS	  One	  6:e23937.	  272.	   Koberle	  V,	  Waidmann	  O,	  Kronenberger	  B,	  Andrei	  A,	  Susser	  S,	  Fuller	  C,	  Perner	  D,	  Zeuzem	  S,	  Sarrazin	  C,	  Piiper	  A.	  2013.	  Serum	  microRNA-­‐122	  kinetics	  in	  patients	  with	  chronic	  hepatitis	  C	  virus	  infection	  during	  antiviral	  therapy.	  J	  Viral	  Hepat	  20:530-­‐535.	  273.	   Su	  TH,	  Liu	  CH,	  Liu	  CJ,	  Chen	  CL,	  Ting	  TT,	  Tseng	  TC,	  Chen	  PJ,	  Kao	  JH,	  Chen	  DS.	  2013.	  Serum	  microRNA-­‐122	  level	  correlates	  with	  virologic	  responses	  to	  pegylated	  interferon	  therapy	  in	  chronic	  hepatitis	  C.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  110:7844-­‐7849.	  274.	   Waring	  JF,	  Dumas	  EO,	  Abel	  S,	  Coakley	  E,	  Cohen	  DE,	  Wade	  Davis	  J,	  Podsadecki	  T,	  Dutta	  S.	  2015.	  Serum	  miR-­‐122	  may	  serve	  as	  a	  biomarker	  for	  response	  to	  direct	  acting	  antivirals:	  effect	  of	  paritaprevir/R	  with	  dasabuvir	  or	  ombitasvir	  on	  miR-­‐122	  in	  HCV-­‐infected	  subjects.	  J	  Viral	  Hepat	  doi:10.1111/jvh.12470.	  275.	   Katoh	  M.	  2014.	  Cardio-­‐miRNAs	  and	  onco-­‐miRNAs:	  circulating	  miRNA-­‐based	  diagnostics	  for	  non-­‐cancerous	  and	  cancerous	  diseases.	  Front	  Cell	  Dev	  Biol	  2:61.	  276.	   Haneklaus	  M,	  Gerlic	  M,	  O'Neill	  LA,	  Masters	  SL.	  2013.	  miR-­‐223:	  infection,	  inflammation	  and	  cancer.	  J	  Intern	  Med	  274:215-­‐226.	  277.	   Taibi	  F,	  Metzinger-­‐Le	  Meuth	  V,	  Massy	  ZA,	  Metzinger	  L.	  2014.	  miR-­‐223:	  An	  inflammatory	  oncomiR	  enters	  the	  cardiovascular	  field.	  Biochim	  Biophys	  Acta	  1842:1001-­‐1009.	  278.	   Bhattacharya	  S,	  Steele	  R,	  Shrivastava	  S,	  Chakraborty	  S,	  Di	  Bisceglie	  AM,	  Ray	  RB.	  2016.	  Serum	  miR-­‐30e	  and	  miR-­‐223	  as	  Novel	  Noninvasive	  Biomarkers	  for	  Hepatocellular	  Carcinoma.	  Am	  J	  Pathol	  186:242-­‐247.	  279.	   Li	  T,	  Morgan	  MJ,	  Choksi	  S,	  Zhang	  Y,	  Kim	  YS,	  Liu	  ZG.	  2010.	  MicroRNAs	  modulate	  the	  noncanonical	  transcription	  factor	  NF-­‐kappaB	  pathway	  by	  regulating	  expression	  of	  the	  kinase	  IKKalpha	  during	  macrophage	  differentiation.	  Nat	  Immunol	  11:799-­‐805.	  280.	   Johnnidis	  JB,	  Harris	  MH,	  Wheeler	  RT,	  Stehling-­‐Sun	  S,	  Lam	  MH,	  Kirak	  O,	  Brummelkamp	  TR,	  Fleming	  MD,	  Camargo	  FD.	  2008.	  Regulation	  of	  progenitor	  cell	  proliferation	  and	  granulocyte	  function	  by	  microRNA-­‐223.	  Nature	  451:1125-­‐1129.	  281.	   Yu	  CH,	  Xu	  CF,	  Li	  YM.	  2009.	  Association	  of	  MicroRNA-­‐223	  expression	  with	  hepatic	  ischemia/reperfusion	  injury	  in	  mice.	  Dig	  Dis	  Sci	  54:2362-­‐2366.	  282.	   Wong	  QW,	  Lung	  RW,	  Law	  PT,	  Lai	  PB,	  Chan	  KY,	  To	  KF,	  Wong	  N.	  2008.	  MicroRNA-­‐223	  is	  commonly	  repressed	  in	  hepatocellular	  carcinoma	  and	  potentiates	  expression	  of	  Stathmin1.	  Gastroenterology	  135:257-­‐269.	  283.	   Meissner	  EG,	  Lee	  YJ,	  Osinusi	  A,	  Sims	  Z,	  Qin	  J,	  Sturdevant	  D,	  McHutchison	  J,	  Subramanian	  M,	  Sampson	  M,	  Naggie	  S,	  Patel	  K,	  Remaley	  AT,	  Masur	  H,	  Kottilil	  S.	  2015.	  Effect	  of	  sofosbuvir	  and	  ribavirin	  treatment	  on	  peripheral	  and	  	   142	  hepatic	  lipid	  metabolism	  in	  chronic	  hepatitis	  C	  virus,	  genotype	  1-­‐infected	  patients.	  Hepatology	  61:790-­‐801.	  284.	   Ory	  DS.	  2004.	  Nuclear	  receptor	  signaling	  in	  the	  control	  of	  cholesterol	  homeostasis:	  have	  the	  orphans	  found	  a	  home?	  Circ	  Res	  95:660-­‐670.	  285.	   Kim	  KH,	  Hong	  SP,	  Kim	  K,	  Park	  MJ,	  Kim	  KJ,	  Cheong	  J.	  2007.	  HCV	  core	  protein	  induces	  hepatic	  lipid	  accumulation	  by	  activating	  SREBP1	  and	  PPARgamma.	  Biochem	  Biophys	  Res	  Commun	  355:883-­‐888.	  286.	   Kapadia	  SB,	  Chisari	  FV.	  2005.	  Hepatitis	  C	  virus	  RNA	  replication	  is	  regulated	  by	  host	  geranylgeranylation	  and	  fatty	  acids.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  102:2561-­‐2566.	  287.	   Alborn	  WE,	  Cao	  G,	  Careskey	  HE,	  Qian	  YW,	  Subramaniam	  DR,	  Davies	  J,	  Conner	  EM,	  Konrad	  RJ.	  2007.	  Serum	  proprotein	  convertase	  subtilisin	  kexin	  type	  9	  is	  correlated	  directly	  with	  serum	  LDL	  cholesterol.	  Clin	  Chem	  53:1814-­‐1819.	  288.	   Dubuc	  G,	  Tremblay	  M,	  Pare	  G,	  Jacques	  H,	  Hamelin	  J,	  Benjannet	  S,	  Boulet	  L,	  Genest	  J,	  Bernier	  L,	  Seidah	  NG,	  Davignon	  J.	  2010.	  A	  new	  method	  for	  measurement	  of	  total	  plasma	  PCSK9:	  clinical	  applications.	  J	  Lipid	  Res	  51:140-­‐149.	  289.	   Bridge	  SH,	  Sheridan	  DA,	  Felmlee	  DJ,	  Crossey	  MM,	  Fenwick	  FI,	  Lanyon	  CV,	  Dubuc	  G,	  Seidah	  NG,	  Davignon	  J,	  Thomas	  HC,	  Taylor-­‐Robinson	  SD,	  Toms	  GL,	  Neely	  RD,	  Bassendine	  MF.	  2015.	  PCSK9,	  apolipoprotein	  E	  and	  lipoviral	  particles	  in	  chronic	  hepatitis	  C	  genotype	  3:	  evidence	  for	  genotype-­‐specific	  regulation	  of	  lipoprotein	  metabolism.	  J	  Hepatol	  62:763-­‐770.	  290.	   Qian	  YW,	  Schmidt	  RJ,	  Zhang	  Y,	  Chu	  S,	  Lin	  A,	  Wang	  H,	  Wang	  X,	  Beyer	  TP,	  Bensch	  WR,	  Li	  W,	  Ehsani	  ME,	  Lu	  D,	  Konrad	  RJ,	  Eacho	  PI,	  Moller	  DE,	  Karathanasis	  SK,	  Cao	  G.	  2007.	  Secreted	  PCSK9	  downregulates	  low	  density	  lipoprotein	  receptor	  through	  receptor-­‐mediated	  endocytosis.	  J	  Lipid	  Res	  48:1488-­‐1498.	  291.	   Lagace	  TA,	  Curtis	  DE,	  Garuti	  R,	  McNutt	  MC,	  Park	  SW,	  Prather	  HB,	  Anderson	  NN,	  Ho	  YK,	  Hammer	  RE,	  Horton	  JD.	  2006.	  Secreted	  PCSK9	  decreases	  the	  number	  of	  LDL	  receptors	  in	  hepatocytes	  and	  in	  livers	  of	  parabiotic	  mice.	  J	  Clin	  Invest	  116:2995-­‐3005.	  292.	   Labonte	  P,	  Begley	  S,	  Guevin	  C,	  Asselin	  MC,	  Nassoury	  N,	  Mayer	  G,	  Prat	  A,	  Seidah	  NG.	  2009.	  PCSK9	  impedes	  hepatitis	  C	  virus	  infection	  in	  vitro	  and	  modulates	  liver	  CD81	  expression.	  Hepatology	  50:17-­‐24.	  293.	   Lambert	  G,	  Charlton	  F,	  Rye	  KA,	  Piper	  DE.	  2009.	  Molecular	  basis	  of	  PCSK9	  function.	  Atherosclerosis	  203:1-­‐7.	  294.	   Naoumova	  RP,	  Tosi	  I,	  Patel	  D,	  Neuwirth	  C,	  Horswell	  SD,	  Marais	  AD,	  van	  Heyningen	  C,	  Soutar	  AK.	  2005.	  Severe	  hypercholesterolemia	  in	  four	  British	  families	  with	  the	  D374Y	  mutation	  in	  the	  PCSK9	  gene:	  long-­‐term	  follow-­‐up	  and	  treatment	  response.	  Arterioscler	  Thromb	  Vasc	  Biol	  25:2654-­‐2660.	  295.	   Cunningham	  D,	  Danley	  DE,	  Geoghegan	  KF,	  Griffor	  MC,	  Hawkins	  JL,	  Subashi	  TA,	  Varghese	  AH,	  Ammirati	  MJ,	  Culp	  JS,	  Hoth	  LR,	  Mansour	  MN,	  McGrath	  KM,	  Seddon	  AP,	  Shenolikar	  S,	  Stutzman-­‐Engwall	  KJ,	  Warren	  LC,	  Xia	  D,	  Qiu	  X.	  2007.	  Structural	  and	  biophysical	  studies	  of	  PCSK9	  and	  its	  mutants	  linked	  to	  familial	  hypercholesterolemia.	  Nat	  Struct	  Mol	  Biol	  14:413-­‐419.	  	   143	  296.	   Fisher	  TS,	  Lo	  Surdo	  P,	  Pandit	  S,	  Mattu	  M,	  Santoro	  JC,	  Wisniewski	  D,	  Cummings	  RT,	  Calzetta	  A,	  Cubbon	  RM,	  Fischer	  PA,	  Tarachandani	  A,	  De	  Francesco	  R,	  Wright	  SD,	  Sparrow	  CP,	  Carfi	  A,	  Sitlani	  A.	  2007.	  Effects	  of	  pH	  and	  low	  density	  lipoprotein	  (LDL)	  on	  PCSK9-­‐dependent	  LDL	  receptor	  regulation.	  J	  Biol	  Chem	  282:20502-­‐20512.	  297.	   Owen	  DM,	  Huang	  H,	  Ye	  J,	  Gale	  M,	  Jr.	  2009.	  Apolipoprotein	  E	  on	  hepatitis	  C	  virion	  facilitates	  infection	  through	  interaction	  with	  low-­‐density	  lipoprotein	  receptor.	  Virology	  394:99-­‐108.	  298.	   Lakoski	  SG,	  Lagace	  TA,	  Cohen	  JC,	  Horton	  JD,	  Hobbs	  HH.	  2009.	  Genetic	  and	  metabolic	  determinants	  of	  plasma	  PCSK9	  levels.	  J	  Clin	  Endocrinol	  Metab	  94:2537-­‐2543.	  299.	   Gibbons	  RV,	  Vaughn	  DW.	  2002.	  Dengue:	  an	  escalating	  problem.	  BMJ	  324:1563-­‐1566.	  300.	   Guzman	  MG,	  Halstead	  SB,	  Artsob	  H,	  Buchy	  P,	  Farrar	  J,	  Gubler	  DJ,	  Hunsperger	  E,	  Kroeger	  A,	  Margolis	  HS,	  Martinez	  E,	  Nathan	  MB,	  Pelegrino	  JL,	  Simmons	  C,	  Yoksan	  S,	  Peeling	  RW.	  2010.	  Dengue:	  a	  continuing	  global	  threat.	  Nat	  Rev	  Microbiol	  8:S7-­‐16.	  301.	   WHO.	  2015.	  Dengue	  and	  severe	  dengue.	  Fact	  sheet	  N°117.	  302.	   Laughlin	  CA,	  Morens	  DM,	  Cassetti	  MC,	  Costero-­‐Saint	  Denis	  A,	  San	  Martin	  JL,	  Whitehead	  SS,	  Fauci	  AS.	  2012.	  Dengue	  research	  opportunities	  in	  the	  Americas.	  J	  Infect	  Dis	  206:1121-­‐1127.	  303.	   Lim	  SP,	  Wang	  QY,	  Noble	  CG,	  Chen	  YL,	  Dong	  H,	  Zou	  B,	  Yokokawa	  F,	  Nilar	  S,	  Smith	  P,	  Beer	  D,	  Lescar	  J,	  Shi	  PY.	  2013.	  Ten	  years	  of	  dengue	  drug	  discovery:	  progress	  and	  prospects.	  Antiviral	  Res	  100:500-­‐519.	  304.	   Hotta	  S.	  1952.	  Experimental	  studies	  on	  dengue.	  I.	  Isolation,	  identification	  and	  modification	  of	  the	  virus.	  J	  Infect	  Dis	  90:1-­‐9.	  305.	   Simmons	  CP,	  Farrar	  JJ,	  Nguyen	  v	  V,	  Wills	  B.	  2012.	  Dengue.	  N	  Engl	  J	  Med	  366:1423-­‐1432.	  306.	   Messina	  JP,	  Brady	  OJ,	  Scott	  TW,	  Zou	  C,	  Pigott	  DM,	  Duda	  KA,	  Bhatt	  S,	  Katzelnick	  L,	  Howes	  RE,	  Battle	  KE,	  Simmons	  CP,	  Hay	  SI.	  2014.	  Global	  spread	  of	  dengue	  virus	  types:	  mapping	  the	  70	  year	  history.	  Trends	  Microbiol	  22:138-­‐146.	  307.	   Murray	  CL,	  Jones	  CT,	  Rice	  CM.	  2008.	  Architects	  of	  assembly:	  roles	  of	  Flaviviridae	  non-­‐structural	  proteins	  in	  virion	  morphogenesis.	  Nat	  Rev	  Microbiol	  6:699-­‐708.	  308.	   Twiddy	  SS,	  Holmes	  EC,	  Rambaut	  A.	  2003.	  Inferring	  the	  rate	  and	  time-­‐scale	  of	  dengue	  virus	  evolution.	  Mol	  Biol	  Evol	  20:122-­‐129.	  309.	   Bennett	  SN,	  Holmes	  EC,	  Chirivella	  M,	  Rodriguez	  DM,	  Beltran	  M,	  Vorndam	  V,	  Gubler	  DJ,	  McMillan	  WO.	  2003.	  Selection-­‐driven	  evolution	  of	  emergent	  dengue	  virus.	  Mol	  Biol	  Evol	  20:1650-­‐1658.	  310.	   Loveday	  EK,	  Diederich	  S,	  Pasick	  J,	  Jean	  F.	  2015.	  Human	  microRNA-­‐24	  modulates	  highly	  pathogenic	  avian-­‐origin	  H5N1	  influenza	  A	  virus	  infection	  in	  A549	  cells	  by	  targeting	  secretory	  pathway	  furin.	  J	  Gen	  Virol	  96:30-­‐39.	  311.	   Soto-­‐Acosta	  R,	  Mosso	  C,	  Cervantes-­‐Salazar	  M,	  Puerta-­‐Guardo	  H,	  Medina	  F,	  Favari	  L,	  Ludert	  JE,	  del	  Angel	  RM.	  2013.	  The	  increase	  in	  cholesterol	  levels	  at	  	   144	  early	  stages	  after	  dengue	  virus	  infection	  correlates	  with	  an	  augment	  in	  LDL	  particle	  uptake	  and	  HMG-­‐CoA	  reductase	  activity.	  Virology	  442:132-­‐147.	  312.	   Carro	  AC,	  Damonte	  EB.	  2013.	  Requirement	  of	  cholesterol	  in	  the	  viral	  envelope	  for	  dengue	  virus	  infection.	  Virus	  Res	  174:78-­‐87.	  313.	   Li	  Y,	  Kakinami	  C,	  Li	  Q,	  Yang	  B,	  Li	  H.	  2013.	  Human	  apolipoprotein	  A-­‐I	  is	  associated	  with	  dengue	  virus	  and	  enhances	  virus	  infection	  through	  SR-­‐BI.	  PLoS	  One	  8:e70390.	  314.	   Poh	  MK,	  Shui	  G,	  Xie	  X,	  Shi	  PY,	  Wenk	  MR,	  Gu	  F.	  2012.	  U18666A,	  an	  intra-­‐cellular	  cholesterol	  transport	  inhibitor,	  inhibits	  dengue	  virus	  entry	  and	  replication.	  Antiviral	  Res	  93:191-­‐198.	  315.	   Martinez-­‐Gutierrez	  M,	  Castellanos	  JE,	  Gallego-­‐Gomez	  JC.	  2011.	  Statins	  reduce	  dengue	  virus	  production	  via	  decreased	  virion	  assembly.	  Intervirology	  54:202-­‐216.	  316.	   Soto-­‐Acosta	  R,	  Bautista-­‐Carbajal	  P,	  Syed	  GH,	  Siddiqui	  A,	  Del	  Angel	  RM.	  2014.	  Nordihydroguaiaretic	  acid	  (NDGA)	  inhibits	  replication	  and	  viral	  morphogenesis	  of	  dengue	  virus.	  Antiviral	  Res	  109C:132-­‐140.	  317.	   Brasaemle	  DL,	  Barber	  T,	  Wolins	  NE,	  Serrero	  G,	  Blanchette-­‐Mackie	  EJ,	  Londos	  C.	  1997.	  Adipose	  differentiation-­‐related	  protein	  is	  an	  ubiquitously	  expressed	  lipid	  storage	  droplet-­‐associated	  protein.	  J	  Lipid	  Res	  38:2249-­‐2263.	  318.	   Horton	  JD,	  Shah	  NA,	  Warrington	  JA,	  Anderson	  NN,	  Park	  SW,	  Brown	  MS,	  Goldstein	  JL.	  2003.	  Combined	  analysis	  of	  oligonucleotide	  microarray	  data	  from	  transgenic	  and	  knockout	  mice	  identifies	  direct	  SREBP	  target	  genes.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  100:12027-­‐12032.	  319.	   Zhong	  J,	  Gastaminza	  P,	  Cheng	  G,	  Kapadia	  S,	  Kato	  T,	  Burton	  DR,	  Wieland	  SF,	  Uprichard	  SL,	  Wakita	  T,	  Chisari	  FV.	  2005.	  Robust	  hepatitis	  C	  virus	  infection	  in	  vitro.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  102:9294-­‐9299.	  320.	   Urata	  S,	  Yun	  N,	  Pasquato	  A,	  Paessler	  S,	  Kunz	  S,	  de	  la	  Torre	  JC.	  2011.	  Antiviral	  activity	  of	  a	  small-­‐molecule	  inhibitor	  of	  arenavirus	  glycoprotein	  processing	  by	  the	  cellular	  site	  1	  protease.	  J	  Virol	  85:795-­‐803.	  321.	   Hay	  BA,	  Abrams	  B,	  Zumbrunn	  AY,	  Valentine	  JJ,	  Warren	  LC,	  Petras	  SF,	  Shelly	  LD,	  Xia	  A,	  Varghese	  AH,	  Hawkins	  JL,	  Van	  Camp	  JA,	  Robbins	  MD,	  Landschulz	  K,	  Harwood	  HJ,	  Jr.	  2007.	  Aminopyrrolidineamide	  inhibitors	  of	  site-­‐1	  protease.	  Bioorg	  Med	  Chem	  Lett	  17:4411-­‐4414.	  322.	   Fujimoto	  Y,	  Onoduka	  J,	  Homma	  KJ,	  Yamaguchi	  S,	  Mori	  M,	  Higashi	  Y,	  Makita	  M,	  Kinoshita	  T,	  Noda	  J,	  Itabe	  H,	  Takanoa	  T.	  2006.	  Long-­‐chain	  fatty	  acids	  induce	  lipid	  droplet	  formation	  in	  a	  cultured	  human	  hepatocyte	  in	  a	  manner	  dependent	  of	  Acyl-­‐CoA	  synthetase.	  Biol	  Pharm	  Bull	  29:2174-­‐2180.	  323.	   Medina	  F,	  Medina	  JF,	  Colon	  C,	  Vergne	  E,	  Santiago	  GA,	  Munoz-­‐Jordan	  JL.	  2012.	  Dengue	  virus:	  isolation,	  propagation,	  quantification,	  and	  storage.	  Curr	  Protoc	  Microbiol	  Chapter	  15:Unit	  15D	  12.	  324.	   Johnson	  BW,	  Russell	  BJ,	  Lanciotti	  RS.	  2005.	  Serotype-­‐specific	  detection	  of	  dengue	  viruses	  in	  a	  fourplex	  real-­‐time	  reverse	  transcriptase	  PCR	  assay.	  J	  Clin	  Microbiol	  43:4977-­‐4983.	  325.	   Greenspan	  P,	  Mayer	  EP,	  Fowler	  SD.	  1985.	  Nile	  red:	  a	  selective	  fluorescent	  stain	  for	  intracellular	  lipid	  droplets.	  J	  Cell	  Biol	  100:965-­‐973.	  	   145	  326.	   Blanchet	  M,	  Seidah	  NG,	  Labonte	  P.	  2012.	  SKI-­‐1/S1P	  inhibition:	  a	  promising	  surrogate	  to	  statins	  to	  block	  hepatitis	  C	  virus	  replication.	  Antiviral	  Res	  95:159-­‐166.	  327.	   Griffis	  ER,	  Xu	  S,	  Powers	  MA.	  2003.	  Nup98	  localizes	  to	  both	  nuclear	  and	  cytoplasmic	  sides	  of	  the	  nuclear	  pore	  and	  binds	  to	  two	  distinct	  nucleoporin	  subcomplexes.	  Mol	  Biol	  Cell	  14:600-­‐610.	  328.	   Sangiambut	  S,	  Keelapang	  P,	  Aaskov	  J,	  Puttikhunt	  C,	  Kasinrerk	  W,	  Malasit	  P,	  Sittisombut	  N.	  2008.	  Multiple	  regions	  in	  dengue	  virus	  capsid	  protein	  contribute	  to	  nuclear	  localization	  during	  virus	  infection.	  J	  Gen	  Virol	  89:1254-­‐1264.	  329.	   Guo	  Y,	  Walther	  TC,	  Rao	  M,	  Stuurman	  N,	  Goshima	  G,	  Terayama	  K,	  Wong	  JS,	  Vale	  RD,	  Walter	  P,	  Farese	  RV.	  2008.	  Functional	  genomic	  screen	  reveals	  genes	  involved	  in	  lipid-­‐droplet	  formation	  and	  utilization.	  Nature	  453:657-­‐661.	  330.	   Rohwedder	  A,	  Zhang	  Q,	  Rudge	  SA,	  Wakelam	  MJ.	  2014.	  Lipid	  droplet	  formation	  in	  response	  to	  oleic	  acid	  in	  Huh-­‐7	  cells	  is	  mediated	  by	  the	  fatty	  acid	  receptor	  FFAR4.	  J	  Cell	  Sci	  127:3104-­‐3115.	  331.	   Petersen	  J,	  Drake	  MJ,	  Bruce	  EA,	  Riblett	  AM,	  Didigu	  CA,	  Wilen	  CB,	  Malani	  N,	  Male	  F,	  Lee	  FH,	  Bushman	  FD,	  Cherry	  S,	  Doms	  RW,	  Bates	  P,	  Briley	  K,	  Jr.	  2014.	  The	  major	  cellular	  sterol	  regulatory	  pathway	  is	  required	  for	  Andes	  virus	  infection.	  PLoS	  Pathog	  10:e1003911.	  332.	   Kleinfelter	  LM,	  Jangra	  RK,	  Jae	  LT,	  Herbert	  AS,	  Mittler	  E,	  Stiles	  KM,	  Wirchnianski	  AS,	  Kielian	  M,	  Brummelkamp	  TR,	  Dye	  JM,	  Chandran	  K.	  2015.	  Haploid	  Genetic	  Screen	  Reveals	  a	  Profound	  and	  Direct	  Dependence	  on	  Cholesterol	  for	  Hantavirus	  Membrane	  Fusion.	  MBio	  6:e00801-­‐00815.	  333.	   Uchida	  L,	  Urata	  S,	  Ulanday	  GE,	  Takamatsu	  Y,	  Yasuda	  J,	  Morita	  K,	  Hayasaka	  D.	  2016.	  Suppressive	  Effects	  of	  the	  Site	  1	  Protease	  (S1P)	  Inhibitor,	  PF-­‐429242,	  on	  Dengue	  Virus	  Propagation.	  Viruses	  8.	  334.	   Ricchi	  M,	  Odoardi	  MR,	  Carulli	  L,	  Anzivino	  C,	  Ballestri	  S,	  Pinetti	  A,	  Fantoni	  LI,	  Marra	  F,	  Bertolotti	  M,	  Banni	  S,	  Lonardo	  A,	  Carulli	  N,	  Loria	  P.	  2009.	  Differential	  effect	  of	  oleic	  and	  palmitic	  acid	  on	  lipid	  accumulation	  and	  apoptosis	  in	  cultured	  hepatocytes.	  J	  Gastroenterol	  Hepatol	  24:830-­‐840.	  335.	   Mangold	  KA,	  Reynolds	  SL.	  2013.	  A	  review	  of	  dengue	  fever:	  a	  resurging	  tropical	  disease.	  Pediatr	  Emerg	  Care	  29:665-­‐669.	  336.	   Halstead	  SB.	  1974.	  Etiologies	  of	  the	  experimental	  dengues	  of	  Siler	  and	  Simmons.	  Am	  J	  Trop	  Med	  Hyg	  23:974-­‐982.	  337.	   Halstead	  SB,	  Nimmannitya	  S,	  Cohen	  SN.	  1970.	  Observations	  related	  to	  pathogenesis	  of	  dengue	  hemorrhagic	  fever.	  IV.	  Relation	  of	  disease	  severity	  to	  antibody	  response	  and	  virus	  recovered.	  Yale	  J	  Biol	  Med	  42:311-­‐328.	  338.	   Halstead	  SB,	  Heinz	  FX,	  Barrett	  AD,	  Roehrig	  JT.	  2005.	  Dengue	  virus:	  molecular	  basis	  of	  cell	  entry	  and	  pathogenesis,	  25-­‐27	  June	  2003,	  Vienna,	  Austria.	  Vaccine	  23:849-­‐856.	  339.	   Marschner	  K,	  Kollmann	  K,	  Schweizer	  M,	  Braulke	  T,	  Pohl	  S.	  2011.	  A	  key	  enzyme	  in	  the	  biogenesis	  of	  lysosomes	  is	  a	  protease	  that	  regulates	  cholesterol	  metabolism.	  Science	  333:87-­‐90.	  340.	   Ye	  J,	  Rawson	  RB,	  Komuro	  R,	  Chen	  X,	  Dave	  UP,	  Prywes	  R,	  Brown	  MS,	  Goldstein	  JL.	  2000.	  ER	  stress	  induces	  cleavage	  of	  membrane-­‐bound	  ATF6	  by	  the	  same	  proteases	  that	  process	  SREBPs.	  Mol	  Cell	  6:1355-­‐1364.	  	   146	  341.	   Patra	  D,	  Xing	  X,	  Davies	  S,	  Bryan	  J,	  Franz	  C,	  Hunziker	  EB,	  Sandell	  LJ.	  2007.	  Site-­‐1	  protease	  is	  essential	  for	  endochondral	  bone	  formation	  in	  mice.	  J	  Cell	  Biol	  179:687-­‐700.	  342.	   Gorski	  JP,	  Huffman	  NT,	  Chittur	  S,	  Midura	  RJ,	  Black	  C,	  Oxford	  J,	  Seidah	  NG.	  2011.	  Inhibition	  of	  proprotein	  convertase	  SKI-­‐1	  blocks	  transcription	  of	  key	  extracellular	  matrix	  genes	  regulating	  osteoblastic	  mineralization.	  J	  Biol	  Chem	  286:1836-­‐1849.	  343.	   Seidah	  NG,	  Prat	  A.	  2012.	  The	  biology	  and	  therapeutic	  targeting	  of	  the	  proprotein	  convertases.	  Nat	  Rev	  Drug	  Discov	  11:367-­‐383.	  344.	   Roingeard	  P.	  2013.	  Hepatitis	  C	  virus	  diversity	  and	  hepatic	  steatosis.	  J	  Viral	  Hepat	  20:77-­‐84.	  345.	   Friedman	  RC,	  Farh	  KK,	  Burge	  CB,	  Bartel	  DP.	  2009.	  Most	  mammalian	  mRNAs	  are	  conserved	  targets	  of	  microRNAs.	  Genome	  Res	  19:92-­‐105.	  346.	   Rottiers	  V,	  Naar	  AM.	  2012.	  MicroRNAs	  in	  metabolism	  and	  metabolic	  disorders.	  Nat	  Rev	  Mol	  Cell	  Biol	  13:239-­‐250.	  347.	   Pedersen	  I,	  David	  M.	  2008.	  MicroRNAs	  in	  the	  immune	  response.	  Cytokine	  43:391-­‐394.	  348.	   Matsuura	  K,	  De	  Giorgi	  V,	  Schechterly	  C,	  Wang	  RY,	  Farci	  P,	  Tanaka	  Y,	  Alter	  HJ.	  2016.	  Circulating	  let-­‐7	  levels	  in	  plasma	  and	  extracellular	  vesicles	  correlate	  with	  hepatic	  fibrosis	  progression	  in	  chronic	  hepatitis	  C.	  Hepatology	  doi:10.1002/hep.28660.	  349.	   Bihrer	  V,	  Friedrich-­‐Rust	  M,	  Kronenberger	  B,	  Forestier	  N,	  Haupenthal	  J,	  Shi	  Y,	  Peveling-­‐Oberhag	  J,	  Radeke	  HH,	  Sarrazin	  C,	  Herrmann	  E,	  Zeuzem	  S,	  Waidmann	  O,	  Piiper	  A.	  2011.	  Serum	  miR-­‐122	  as	  a	  biomarker	  of	  necroinflammation	  in	  patients	  with	  chronic	  hepatitis	  C	  virus	  infection.	  Am	  J	  Gastroenterol	  106:1663-­‐1669.	  350.	   Trebicka	  J,	  Anadol	  E,	  Elfimova	  N,	  Strack	  I,	  Roggendorf	  M,	  Viazov	  S,	  Wedemeyer	  I,	  Drebber	  U,	  Rockstroh	  J,	  Sauerbruch	  T,	  Dienes	  HP,	  Odenthal	  M.	  2013.	  Hepatic	  and	  serum	  levels	  of	  miR-­‐122	  after	  chronic	  HCV-­‐induced	  fibrosis.	  J	  Hepatol	  58:234-­‐239.	  351.	   van	  der	  Meer	  AJ,	  Farid	  WR,	  Sonneveld	  MJ,	  de	  Ruiter	  PE,	  Boonstra	  A,	  van	  Vuuren	  AJ,	  Verheij	  J,	  Hansen	  BE,	  de	  Knegt	  RJ,	  van	  der	  Laan	  LJ,	  Janssen	  HL.	  2013.	  Sensitive	  detection	  of	  hepatocellular	  injury	  in	  chronic	  hepatitis	  C	  patients	  with	  circulating	  hepatocyte-­‐derived	  microRNA-­‐122.	  J	  Viral	  Hepat	  20:158-­‐166.	  352.	   Gallo	  A,	  Tandon	  M,	  Alevizos	  I,	  Illei	  GG.	  2012.	  The	  majority	  of	  microRNAs	  detectable	  in	  serum	  and	  saliva	  is	  concentrated	  in	  exosomes.	  PLoS	  One	  7:e30679.	  353.	   Bala	  S,	  Petrasek	  J,	  Mundkur	  S,	  Catalano	  D,	  Levin	  I,	  Ward	  J,	  Alao	  H,	  Kodys	  K,	  Szabo	  G.	  2012.	  Circulating	  microRNAs	  in	  exosomes	  indicate	  hepatocyte	  injury	  and	  inflammation	  in	  alcoholic,	  drug-­‐induced,	  and	  inflammatory	  liver	  diseases.	  Hepatology	  56:1946-­‐1957.	  354.	   Le	  QT,	  Blanchet	  M,	  Seidah	  NG,	  Labonte	  P.	  2015.	  Plasma	  Membrane	  Tetraspanin	  CD81	  Complexes	  with	  Proprotein	  Convertase	  Subtilisin/Kexin	  Type	  9	  (PCSK9)	  and	  Low	  Density	  Lipoprotein	  Receptor	  (LDLR),	  and	  Its	  Levels	  Are	  Reduced	  by	  PCSK9.	  J	  Biol	  Chem	  290:23385-­‐23400.	  	   147	  355.	   Syed	  GH,	  Tang	  H,	  Khan	  M,	  Hassanein	  T,	  Liu	  J,	  Siddiqui	  A.	  2014.	  Hepatitis	  C	  virus	  stimulates	  low-­‐density	  lipoprotein	  receptor	  expression	  to	  facilitate	  viral	  propagation.	  J	  Virol	  88:2519-­‐2529.	  356.	   Seidah	  NG.	  2016.	  New	  developments	  in	  proprotein	  convertase	  subtilisin-­‐kexin	  9's	  biology	  and	  clinical	  implications.	  Curr	  Opin	  Lipidol	  27:274-­‐281.	  357.	   Sabatine	  MS,	  Giugliano	  RP,	  Wiviott	  SD,	  Raal	  FJ,	  Blom	  DJ,	  Robinson	  J,	  Ballantyne	  CM,	  Somaratne	  R,	  Legg	  J,	  Wasserman	  SM,	  Scott	  R,	  Koren	  MJ,	  Stein	  EA,	  Open-­‐Label	  Study	  of	  Long-­‐Term	  Evaluation	  against	  LDLCI.	  2015.	  Efficacy	  and	  safety	  of	  evolocumab	  in	  reducing	  lipids	  and	  cardiovascular	  events.	  N	  Engl	  J	  Med	  372:1500-­‐1509.	  358.	   Robinson	  JG,	  Farnier	  M,	  Krempf	  M,	  Bergeron	  J,	  Luc	  G,	  Averna	  M,	  Stroes	  ES,	  Langslet	  G,	  Raal	  FJ,	  El	  Shahawy	  M,	  Koren	  MJ,	  Lepor	  NE,	  Lorenzato	  C,	  Pordy	  R,	  Chaudhari	  U,	  Kastelein	  JJ,	  Investigators	  OLT.	  2015.	  Efficacy	  and	  safety	  of	  alirocumab	  in	  reducing	  lipids	  and	  cardiovascular	  events.	  N	  Engl	  J	  Med	  372:1489-­‐1499.	  359.	   Loveday	  EK.	  2014.	  Regulating	  the	  regulators:	  emerging	  roles	  of	  cellular	  microRNAs	  during	  influenza	  A	  virus	  infection.	  Doctor	  of	  Philosophy.	  The	  University	  of	  British	  Columbia.	  360.	   Pasquato	  A,	  Rochat	  C,	  Burri	  DJ,	  Pasqual	  G,	  de	  la	  Torre	  JC,	  Kunz	  S.	  2012.	  Evaluation	  of	  the	  anti-­‐arenaviral	  activity	  of	  the	  subtilisin	  kexin	  isozyme-­‐1/site-­‐1	  protease	  inhibitor	  PF-­‐429242.	  Virology	  423:14-­‐22.	  361.	   Walther	  TC,	  Farese	  RV,	  Jr.	  2012.	  Lipid	  droplets	  and	  cellular	  lipid	  metabolism.	  Annu	  Rev	  Biochem	  81:687-­‐714.	  362.	   Cheung	  W,	  Gill	  M,	  Esposito	  A,	  Kaminski	  CF,	  Courousse	  N,	  Chwetzoff	  S,	  Trugnan	  G,	  Keshavan	  N,	  Lever	  A,	  Desselberger	  U.	  2010.	  Rotaviruses	  associate	  with	  cellular	  lipid	  droplet	  components	  to	  replicate	  in	  viroplasms,	  and	  compounds	  disrupting	  or	  blocking	  lipid	  droplets	  inhibit	  viroplasm	  formation	  and	  viral	  replication.	  J	  Virol	  84:6782-­‐6798.	  363.	   Coffey	  CM,	  Sheh	  A,	  Kim	  IS,	  Chandran	  K,	  Nibert	  ML,	  Parker	  JS.	  2006.	  Reovirus	  outer	  capsid	  protein	  micro1	  induces	  apoptosis	  and	  associates	  with	  lipid	  droplets,	  endoplasmic	  reticulum,	  and	  mitochondria.	  J	  Virol	  80:8422-­‐8438.	  364.	   Vogt	  DA,	  Camus	  G,	  Herker	  E,	  Webster	  BR,	  Tsou	  CL,	  Greene	  WC,	  Yen	  TS,	  Ott	  M.	  2013.	  Lipid	  droplet-­‐binding	  protein	  TIP47	  regulates	  hepatitis	  C	  Virus	  RNA	  replication	  through	  interaction	  with	  the	  viral	  NS5A	  protein.	  PLoS	  Pathog	  9:e1003302.	  365.	   Ikeda	  M,	  Abe	  K,	  Yamada	  M,	  Dansako	  H,	  Naka	  K,	  Kato	  N.	  2006.	  Different	  anti-­‐HCV	  profiles	  of	  statins	  and	  their	  potential	  for	  combination	  therapy	  with	  interferon.	  Hepatology	  44:117-­‐125.	  366.	   Kim	  SS,	  Peng	  LF,	  Lin	  W,	  Choe	  WH,	  Sakamoto	  N,	  Kato	  N,	  Ikeda	  M,	  Schreiber	  SL,	  Chung	  RT.	  2007.	  A	  cell-­‐based,	  high-­‐throughput	  screen	  for	  small	  molecule	  regulators	  of	  hepatitis	  C	  virus	  replication.	  Gastroenterology	  132:311-­‐320.	  367.	   Pietschmann	  T,	  Kaul	  A,	  Koutsoudakis	  G,	  Shavinskaya	  A,	  Kallis	  S,	  Steinmann	  E,	  Abid	  K,	  Negro	  F,	  Dreux	  M,	  Cosset	  FL,	  Bartenschlager	  R.	  2006.	  Construction	  and	  characterization	  of	  infectious	  intragenotypic	  and	  intergenotypic	  hepatitis	  C	  virus	  chimeras.	  Proc	  Natl	  Acad	  Sci	  U	  S	  A	  103:7408-­‐7413.	  	   148	  368.	   Liu	  X,	  Wang	  T,	  Wakita	  T,	  Yang	  W.	  2010.	  Systematic	  identification	  of	  microRNA	  and	  messenger	  RNA	  profiles	  in	  hepatitis	  C	  virus-­‐infected	  human	  hepatoma	  cells.	  Virology	  398:57-­‐67.	  369.	   Thomson	  DW,	  Bracken	  CP,	  Goodall	  GJ.	  2011.	  Experimental	  strategies	  for	  microRNA	  target	  identification.	  Nucleic	  Acids	  Res	  39:6845-­‐6853.	  370.	   Chang	  H,	  Yi	  B,	  Ma	  R,	  Zhang	  X,	  Zhao	  H,	  Xi	  Y.	  2016.	  CRISPR/cas9,	  a	  novel	  genomic	  tool	  to	  knock	  down	  microRNA	  in	  vitro	  and	  in	  vivo.	  Sci	  Rep	  6:22312.	  371.	   Essalmani	  R,	  Susan-­‐Resiga	  D,	  Chamberland	  A,	  Abifadel	  M,	  Creemers	  JW,	  Boileau	  C,	  Seidah	  NG,	  Prat	  A.	  2011.	  In	  Vivo	  Evidence	  That	  Furin	  from	  Hepatocytes	  Inactivates	  PCSK9.	  J	  Biol	  Chem	  286:4257-­‐4263.	  372.	   Lipari	  MT,	  Li	  W,	  Moran	  P,	  Kong-­‐Beltran	  M,	  Sai	  T,	  Lai	  J,	  Lin	  SJ,	  Kolumam	  G,	  Zavala-­‐Solorio	  J,	  Izrael-­‐Tomasevic	  A,	  Arnott	  D,	  Wang	  J,	  Peterson	  AS,	  Kirchhofer	  D.	  2012.	  Furin-­‐cleaved	  proprotein	  convertase	  subtilisin/kexin	  type	  9	  (PCSK9)	  is	  active	  and	  modulates	  low	  density	  lipoprotein	  receptor	  and	  serum	  cholesterol	  levels.	  J	  Biol	  Chem	  287:43482-­‐43491.	  373.	   Pena	  J,	  Harris	  E.	  2012.	  Early	  dengue	  virus	  protein	  synthesis	  induces	  extensive	  rearrangement	  of	  the	  endoplasmic	  reticulum	  independent	  of	  the	  UPR	  and	  SREBP-­‐2	  pathway.	  PLoS	  One	  7:e38202.	  374.	   Pena	  J,	  Harris	  E.	  2011.	  Dengue	  virus	  modulates	  the	  unfolded	  protein	  response	  in	  a	  time-­‐dependent	  manner.	  J	  Biol	  Chem	  286:14226-­‐14236.	  375.	   Ambrose	  RL,	  Mackenzie	  JM.	  2011.	  West	  Nile	  virus	  differentially	  modulates	  the	  unfolded	  protein	  response	  to	  facilitate	  replication	  and	  immune	  evasion.	  J	  Virol	  85:2723-­‐2732.	  376.	   Zellweger	  RM,	  Shresta	  S.	  2014.	  Mouse	  models	  to	  study	  dengue	  virus	  immunology	  and	  pathogenesis.	  Front	  Immunol	  5:151.	  377.	   Chen	  YL,	  Yin	  Z,	  Duraiswamy	  J,	  Schul	  W,	  Lim	  CC,	  Liu	  B,	  Xu	  HY,	  Qing	  M,	  Yip	  A,	  Wang	  G,	  Chan	  WL,	  Tan	  HP,	  Lo	  M,	  Liung	  S,	  Kondreddi	  RR,	  Rao	  R,	  Gu	  H,	  He	  H,	  Keller	  TH,	  Shi	  PY.	  2010.	  Inhibition	  of	  dengue	  virus	  RNA	  synthesis	  by	  an	  adenosine	  nucleoside.	  Antimicrob	  Agents	  Chemother	  54:2932-­‐2939.	  378.	   Byrd	  CM,	  Grosenbach	  DW,	  Berhanu	  A,	  Dai	  D,	  Jones	  KF,	  Cardwell	  KB,	  Schneider	  C,	  Yang	  G,	  Tyavanagimatt	  S,	  Harver	  C,	  Wineinger	  KA,	  Page	  J,	  Stavale	  E,	  Stone	  MA,	  Fuller	  KP,	  Lovejoy	  C,	  Leeds	  JM,	  Hruby	  DE,	  Jordan	  R.	  2013.	  Novel	  benzoxazole	  inhibitor	  of	  dengue	  virus	  replication	  that	  targets	  the	  NS3	  helicase.	  Antimicrob	  Agents	  Chemother	  57:1902-­‐1912.	  379.	   Byrd	  CM,	  Dai	  D,	  Grosenbach	  DW,	  Berhanu	  A,	  Jones	  KF,	  Cardwell	  KB,	  Schneider	  C,	  Wineinger	  KA,	  Page	  JM,	  Harver	  C,	  Stavale	  E,	  Tyavanagimatt	  S,	  Stone	  MA,	  Bartenschlager	  R,	  Scaturro	  P,	  Hruby	  DE,	  Jordan	  R.	  2013.	  A	  novel	  inhibitor	  of	  dengue	  virus	  replication	  that	  targets	  the	  capsid	  protein.	  Antimicrob	  Agents	  Chemother	  57:15-­‐25.	     	   149	  Appendix 1: Chapter 2 supplementary tables and figures  Table A1.1. Cohort characteristics.     	   150	  Table A1.2. Mixed model analysis of miR-122 against biochemical parameters.     	   151	  Table A1.3. Mixed model analysis of miR-24 against biochemical parameters.     	   152	  Table A1.4. Mixed model analysis of miR-223 against biochemical parameters.      	   153	    Figure A1.1. Differential changes of miR-122, miR-24 and miR-223 before, during and after treatment relative to baseline levels in CHC-infected patients based on treatment outcome.  Abundance of miR-122, miR-24 and miR-223 were measured at different timepoints before, during and after treatment in individuals achieving SVR (A), relapsers (B), and non-responders (C).  Relative microRNAs levels are represented as a fold change of microRNAs normalized to spike‐in control and relative to baseline (week 0). Results are shown as mean ± SEM and statistical significance was calculated with one‐way ANOVA. FU stands for follow-up (week 12 or 24 after treatment).  *P ≤ 0.05, ****P ≤ 0.0001"  	   154	  Appendix 2: Chapter 4 supplementary figures   Figure A2.1. Oligonucleotide primers and fluorogenic probes used in the serotype-specific DENV virus real-time RT-PCR assay.   Probes were hybridized with the HEX fluorophore, and the black hole quencher-1 (BHQ-1) was used as the fluorophore-quencher.    	   155	   Figure A2.2. Inhibition of SKI-1/S1P using PF-429242 prevents activation of the SREBP pathway in DENV-2 infected Huh-7.5.1 cells.  Huh-7.5.1 cells were treated either with 0.02% DMSO (control) or 10/20 µM PF-429242 for 24 hours. The inhibitor was removed and the cells were then infected with DENV-2 (MOI 0.01) for 48 hours. Total RNA was extracted and the mRNA levels of SREBP-1c, SREBP-2, PCSK9, LDLR, FURIN, and SKI-1/S1P were quantified by qRT-PCR in DENV-2-infected cells. Statistical significance was calculated with a two-way ANOVA with Bonferroni’s post-test. Results were normalized against β-actin mRNA levels and expressed as fold change. Values represent average ± SEM of three independent experiments. **, p < 0.01;; ****, p < 0.001.    	   156	   Figure A2.3. Characterization of AcPF-429242 as an inactive derivative of PF-429242.  (A) Chemical structures are shown of PF-429242 and its acetyl derivative (AcPF-429242). (B) AcPF-429242 was evaluated on cytotoxicity on Huh-7.5.1 cells.  Huh-7.5.1 cells were treated with DMSO  (0.01% and 0.02%) or AcPF-429242 (10 µM and 20 µM) for 24 hours before the inhibitor was removed and fresh complete media was added to the cells for an additional 48 hours. The relative cytotoxicity of the compounds was then determined using an MTS-based cell viability assay. The absorbance measured at 490 nm is proportional to the number of living cultured cells. Results (mean ± SEM) from three independent experiments are shown. Statistical significance was calculated with a one-way ANOVA with Bonferroni’s post-test.    	   157	  Appendix 3: Chapter 5 supplementary figures   Figure A3.1. Intracellular miR-122, miR-24 and miR-223 levels during HCV JC1 infection of hepatoma Huh-7.5 cells.  Huh-7.5 cells were infected with HCV JC1 infectious clone. Total RNA was isolated with Trizol at 72 hpi and miRNA expression levels were measured by qRT-PCR as described in Chapter 2.  Results (mean ± SEM) from two independent experiments are shown.   miRNAs in HCV-infected cellsmiR-122miR-24miR-223SNORD68spike-in010203040Ct valueUninfectedJC1-infected	   158	    	   159	  Figure A3.2. DENV hijacks the SKI-1/S1P-mediated proteolytic activation of the SREBP pathway during the late stages of the viral lifecycle in Huh-7.5.1 cells.   (A-C) Huh-7.5.1 cells were infected with DENV-2 (MOI 1) or mock-infected. Total RNA was extracted at various timepoints (0, 4, 8, 24, 48, and 72h) post-infection, and mRNA expression of SKI-1/S1P and S2P (A), SREBP-1c and SREBP-2 (B), LDLR, PCSK9, and FURIN (C) were quantified by qRT-PCR. Results were normalized against control β-actin mRNA levels and expressed as fold change relative to the time-matched mock-infected controls. (D) Huh-7.5.1 cells were infected with DENV-2 (MOI 1) or mock-infected. Cells were harvested at multiple intervals (4, 8, 24, and 48h) post-infection. Cell lysates were probed for SREBP-2 and DENV NS1. Expression of the proteins was quantified in DENV-2-infected cells relative to the time-matched mock-infected controls and normalized to β-tubulin. Representative Western blots of three independent experiments are shown. (E) % of SREBP-2 processing was calculated from the ratio of the cleaved SREBP-2 form to total protein expression. Protein bands were quantified using the Odyssey infrared imaging system. Results (mean ± SEM) from at least three independent experiments are shown. Statistical significance was calculated for DENV-2-infected cells compared to mock-infected cells at timepoints corresponding to DENV-2-infected cells with a two-way ANOVA with Bonferroni’s post-test. *, p < 0.05; **, p < 0.01; ***, p < 0.005; ****, p < 0.001.  

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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0340573/manifest

Comment

Related Items