Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of Hic1-expressing cells in the murine pancreas and their role in tissue regulation and regeneration Schreiner, Petra 2014

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

Item Metadata


24-ubc_2014_november_schreiner_petra.pdf [ 42.31MB ]
JSON: 24-1.0167026.json
JSON-LD: 24-1.0167026-ld.json
RDF/XML (Pretty): 24-1.0167026-rdf.xml
RDF/JSON: 24-1.0167026-rdf.json
Turtle: 24-1.0167026-turtle.txt
N-Triples: 24-1.0167026-rdf-ntriples.txt
Original Record: 24-1.0167026-source.json
Full Text

Full Text

Analysis of Hic1-expressing cells in the murine pancreas and their role in tissue regulation and regeneration  by  Petra Schreiner  B.Sc. (Hons). The University of British Columbia, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Cell and Developmental Biology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  October, 2014 © Petra Schreiner, 2014 	   ii	  ABSTRACT Mesenchymal stromal cells (MSC) play fundamental roles in tissue development, homeostasis and regeneration and are associated with many pathological conditions.  Previous studies analyzing MSCs in skeletal development identified a gene (Hypermethylated in Cancer 1, Hic1) that appears to be solely expressed within embryonic mesenchyme.  Analysis of a novel Hic1nLacZ/+ mouse line revealed that Hic1 is restricted to interstitial stromal and perivascular cells.  In pancreas, Hic1+ cells overlap extensively with Pdgfrα and Sca1, which together identify putative MSCs in various tissues.  These cells also co-express several markers of Pancreatic Stellate cells (PSC), including vimentin, desmin, nestin and neural/glial antigen 2 (NG2). PSCs are considered effector cells in many pancreatic diseases including fibrosis and cancer, and become activated under these conditions.  Using a novel fate-tracking Hic1CreERT2 knock-in mouse, we have seen that under homeostatic conditions these cells are capable of some turnover, the labeled cells doubling in number over a 3-month period following induction of the reporter gene in adult 8-week old mice.  After caerulein-induced pancreatic damage, the cells become activated, attaining smooth muscle actin α (αSMA) and Collagen-GFP expression, and incorporating EdU, with cells amassing in the damaged areas.  Consistent with a potential role for Hic1 in regulating MSC quiescence, conditional deletion of Hic1 in the adult pancreas leads to significantly increased numbers of perilipin+ adipocytes in comparison to controls, and an increased number of nLacZ+ cells.  CD45- CD31- Sca1+ cells proliferate at a higher rate after Hic1-deletion.  Mouse weight and blood glucose are unaffected.  Functional studies into the role of Hic1 	   iii	  have found that Hic1-deletion resulted in a significantly less damage after caerulein-induced damage.  This work has identified Hic1 as a novel marker of PSCs and set the stage for a greater understanding of the role of these mesenchymal stromal cells (MSCs) in pancreatic diseases.    	   iv	  PREFACE Dr. Underhill’s lab identified the gene Hypermethylated in Cancer 1 (Hic1), as playing an important role in mesenchymal stromal cells during embryonic development. A post-doctoral fellow Dr. Hwa Seon Seo and a research associate Dr. Arthur V. Sampaio completed early studies identifying these cells in various tissues. All the pancreas-specific work in this thesis, including literature research, study design, research activities, data analysis and writing, was completed by Petra Schreiner under the supervision of Dr. T. Michael Underhill. Figures 2 & 3 were completed jointly by Dr. Seo and Petra Schreiner. Wilder Scott and Heejung (Lisa) Choi performed confocal microscopy and created the data in Figure 10 and assisted with Figure 12. Some technical assistance (paraffin processing of tissues, sectioning, some IF staining) was given by the summer student Betina Ng and the technician Derek Wong. All animal work was approved by the UBC Animal Care Committee, Certificate # A09-0476, A13-0333 and A11-0187.   	   v	  TABLE OF CONTENTS Abstract	  ......................................................................................................................................	  ii	  Preface	  .......................................................................................................................................	  iv	  Table	  of	  Contents	  .....................................................................................................................	  v	  List	  of	  Tables	  ...........................................................................................................................	  vii	  List	  of	  Figures	  .......................................................................................................................	  viii	  List	  of	  Abbreviations	  .............................................................................................................	  ix	  Acknowledgements	  ...............................................................................................................	  xi	  Dedication	  ................................................................................................................................	  xii	  Chapter	  1:	  Introduction	  ........................................................................................................	  1	  HIC1	  INTRODUCTION	  .......................................................................................................................	  1	  MESENCHYMAL	  STEM	  CELLS	  –	  INTRODUCTION	  ......................................................................	  5	  REGENERATION	  VS	  REPAIR,	  IN	  GENERAL	  .................................................................................	  9	  PANCREAS	  STRUCTURE,	  FUNCTION	  &	  HISTOLOGY	  ............................................................	  14	  PANCREATIC	  DISEASES	  ................................................................................................................	  16	  PANCREATIC	  DAMAGE	  MODELS	  ................................................................................................	  19	  SUMMARY,	  OBJECTIVES	  &	  HYPOTHESES	  ................................................................................	  21	  Chapter	  2:	  Methods	  &	  Materials	  ......................................................................................	  23	  MICE	  ....................................................................................................................................................	  23	  PANCREATITIS	  DISEASE	  MODEL	  ...............................................................................................	  25	  REAGENTS	  .........................................................................................................................................	  26	  STAINING	  ...........................................................................................................................................	  26	  HIC1+	  CELL	  ISOLATION	  .................................................................................................................	  30	  MICROSCOPY	  &	  IMAGE	  COLLECTION	  .......................................................................................	  32	  Chapter	  3:	  Results	  ................................................................................................................	  33	  HIC1+	  CELL	  CHARACTERIZATION	  .............................................................................................	  33	  HIC1	  DELETION	  ...............................................................................................................................	  41	  HIC1-­‐CELL	  ACTIVATION	  ...............................................................................................................	  46	  PANCREATITIS	  MODEL	  DEVELOPMENT	  .................................................................................	  52	  TDTOMATO+	  CELLS	  PROLIFERATE	  AFTER	  DAMAGE	  ..........................................................	  55	  PANCREAS	  REGENERATION	  OF	  HIC1-­‐DELETED	  MICE	  ........................................................	  55	  Chapter	  4:	  Discussion	  .........................................................................................................	  59	  HIC1	  IS	  A	  PUTATIVE	  MARKER	  OF	  QUIESCENT	  PSCS	  ...........................................................	  59	  THE	  EFFECT	  OF	  HIC1	  DELETION	  ...............................................................................................	  61	  HIC1+	  CELLS	  IN	  HOMEOSTASIS	  AND	  IN	  PANCREATIC	  DAMAGE	  .....................................	  63	  REGENERATION	  ..............................................................................................................................	  66	  Chapter	  5:	  Conclusion	  .........................................................................................................	  69	  SUMMARY	  .........................................................................................................................................	  69	  DISCUSSION	  OF	  GOALS/OBJECTIVES	  .......................................................................................	  69	  SIGNIFICANCE	  ..................................................................................................................................	  72	  STRENGTHS	  &	  LIMITATIONS	  ......................................................................................................	  73	  	   vi	  POTENTIAL	  APPLICATIONS	  ........................................................................................................	  73	  FUTURE	  DIRECTIONS	  ....................................................................................................................	  74	  Bibliography	  ..........................................................................................................................	  76	  	  	    	   vii	  LIST OF TABLES Table 1. Characteristics of Pancreatic Mesenchymal Cells    13 Table 2. Types of Diseases Affecting the Pancreas     17 Table 3. List of Transgenic Mice Used      23 Table 4. List of Primers for Genotyping      25 Table 5. Primary Antibodies        30 Table 6. Frequency of tdTomato+ Cells in Lineage-Tracked Samples  41 Table 7. Frequency of tdTomato+ Cells in Healthy and Damaged Pancreas  52     	   viii	  LIST OF FIGURES Figure 1. Hic1 Structure              3 Figure 2. Hic1LacZ is expressed near, but not within, epithelium             6 Figure 3. Hic1-expression in comparison to PDGFRα-EGFP, CD31 and Vimentin       7 Figure 4. Roles of MSCs             10 Figure 5. Basic murine pancreas structure and histology         15 Figure 6. Various Hic1 mouse lines            24 Figure 7. Characterization of pancreatic Hic1+ cells by flow cytometry        34 Figure 8. tdTomato+ cells in cell culture                35 Figure 9. Characterization of Hic1-expressing cells in healthy tissue       37 Figure 10. No tdTomato label is present without tamoxifen administration        39 Figure 11. GFAP antibody staining.             40 Figure 12. Tracking of tdTomato+ cells over time          42 Figure 13. Effect of Hic1 deletion in adult mouse pancreas         44 Figure 14. Timeline of caerulein-induced pancreatic damage and regeneration      48 Figure 15. tdTomato-labeled cells after caerulein-induced pancreatic damage      49 Figure 16. Characterization of tdTomato+ cells in caerulein-damaged pancreas      50 Figure 17. Optimization of the pancreas damage model         54 Figure 18. EdU incorporation into tdTomato+ cells in health and caerulein-damaged pancreata               56 Figure 19. Hic1-deleted mice in pancreas regeneration         57 Figure 20. Proposed action of Hic1+ cells in pancreas         70 	   ix	  LIST OF ABBREVIATIONS ADAM12  A disintegrin and metalloprotease 12 ADRB2  β2 adrenergic receptor α-SMA, acta2  Smooth muscle actin alpha Atoh1   Atonal Homolog 1 BTB/POZ  BR-C, ttk and BAB/Pox virus and zinc finger domain CCK   Choleocystokinin CCND1  Cyclin D1 CP   Chronic pancreatitis CXCR7  C-X-C chemokine receptor 7 DBTC   Dibutyltin dichloride E   Embryonic development day ECM   Extracellular matrix EphA2   Ephrin receptor A2 FACS   Fluorescence-activated cell sorting FCM   Flow cytometry GFAP   Glial Fibrillary Acidic Protein HGPS   Hutchinson-Gilford Progeria syndrome Hic1   Hypermethylated in Cancer 1 HSC   Hepatic stellate cells IL-2   Interleukin 2 IL-10   Interleukin 10 IPF   Idiopathic Pulmonary Fibrosis 	   x	  iPSC   Induced pluripotent stem cells MRL/Mp  A mouse strain with autoimmune disease (hypoactive Il-2 allele) MSC   Mesenchymal stromal/stem cell NCAM  Neural cell adhesion molecular  NG2; Cpsg4  Neural/glial antigen 2; Chondroitin sulfate proteoglycan 4 NGF   Nerve growth factor Nkx3.2  NK3 homeobox 2, also known as Bapx1 nLacZ   nuclear β-galactosidase NuRD   Nucleosome remodeling deacetylase P57KIP2  Cyclin-dependent kinase inhibitor 1C; CDKN1C P21CIP/Waf1  Cyclin-ependent kinase inhibitor 1A; CDKN1A PDAC   Pancreatic ductal adenocarcinoma Pdgfrα   Platelet-derived growth factor receptor alpha PFBs   Pancreatic Fibroblasts Plin   Perilipin PSCs   Pancreatic stellate cells qPCR   Quantitative polymerase chain reaction SIRT1   Sirtuin 1 Sox9   SRY (sex determining region Y) box 9 SWI-SNF Switch/Sucrose non fermentable nucleosome remodeling complex TCF4   Transcription factor 4 TGF- β  Transforming growth factor beta TNBS   Trinitrobenzene sulfonic acid 	   xi	  ACKNOWLEDGEMENTS  I would like to thank my supervisor, Dr. T. Michael Underhill, for his support, mentorship, patience, and his open leadership style in allowing me to choose a project and follow where it leads. I am also grateful for the assistance and advice from other lab members, including Wilder Scott, Le Su, Dr. Arthur Sampaio, Dr. Hwa Seon Seo and Dr. Kim Wiegand and Ryan van der Werff. As well, I would like to thank the BRC Unit, specifically Krista Ranta, Les Rollins, Wei Yuan, Alexa Taylor and Weiting Hsieh, and the UBC Flow Cytometry Facility, Andy Johnson and Justin Wong. I would also like to thank my committee members, past and present, for their time and advice: Drs Fabio Rossi, Kelly McNagny, Francis Lynn and Calvin Roskelley. I also appreciate the financial support of scholarships from the National Sciences and Engineering Research Council of Canada and the UBC Faculty of Medicine, and research grants from the Canadian Institutes of Health Research and Pancreas Centre BC.    	   xii	  DEDICATION   I would like to dedicate this work to my family, who have always supported and encouraged me, and especially my father, who I know would be so proud.  	   1	  CHAPTER 1: INTRODUCTION HIC1 INTRODUCTION Mesenchymal cells play fundamental roles in the developing embryos and in the Underhill laboratory we have been studying the function of these cells in embryogenesis.   During the course of this research we discovered that a gene called Hypermethylated in cancer 1, Hic1, is restricted to embryonic mesenchyme.  The gene HIC1 was first discovered in 1995 by Baylin’s group as a putative tumor suppressor, where it resides near TP53 on human chromosome 17p131.  Their group found that the HIC1 gene is regularly hypermethylated in tumour samples, and this has been subsequently reproduced by numerous other groups who examined breast, colon, brain and lung tumours2,3.  Baylin’s group also found that mice which were heterozygous for Hic1 were more likely to develop tumours than their wild-type littermates, a finding thought to be due to later silencing of the wild-type Hic1 allele by methylation4,5.  Hic1-/- embryos die around gestational age of 10.5 in an in-bred C57Bl/6 strain, with developmental defects including decreased size and numerous abnormalities6.  Many of these results have been reproduced in our lab (unpublished data).  Hic1 structure (gene, protein) In the human genome HIC1 is on chromosome 17 in a CpG island; in the mouse Hic1 is on chromosome 111.  The nucleotide sequences between these species are 89% identical (as calculated by an NCBI BLAST search), resulting in a protein of about 80 kDa in size.  There may be as many as six different transcripts of HIC1 though there is 	   2	  only one known protein isoform7,8 (Figure 1).  Exon 2, which is common to all transcripts, is composed of a BTB/POZ domain at the N-terminus and has five Krüppel-type C2-H2 zinc-fingers1.  BTB/POZ domains and zinc-fingers are common in DNA-binding proteins and are involved in protein-protein interaction and DNA binding 9,10.  A consensus DNA binding sequence, 5’-C/GNGC/GGGGCAC/ACC-3’, has been identified7 for Hic1, assisting in identification of target genes which are involved in epigenetic regulation, development, cell cycle control, cell adhesion and migration, and tumour development, example target genes being Sirt1, Atonal Homolog 1 (Atoh1), chemokine (C-X-C motif) receptor 7 (CXCR7), EphA2, CCND1, P57KIP2, P21CIP/Waf1, Sox9 and ADRB25,11–18,17.   Hic1 function Early research on Hic1 focused on phenotypic analysis, such as the aforementioned observations of increased tumour incidence and decreased embryo size of Hic1 mutants1–4.  Researchers also noted that overexpression of HIC1 in vitro led to decreased cell growth1.  It took several years until the research focus on Hic1 changed from phenotypic observations in cancer models to identifying interacting proteins and understanding the mechanism of action.  It has been proven that HIC1 complexes with the deacetylase Sirtuin 1(SIRT1) to repress SIRT1 transcription, thereby preventing deacetylation and inactivation of p53 by SIRT15.  HIC1 can antagonize Wnt signals by binding Transcription factor 4 (TCF-4) and sequestering β-catenin, preventing both of these transcription factors from interacting and binding to the promoters of Wnt-	   3	  	  	  	  	  	  	  	   Figure 1. HIC1 structure. Exon 2 contains the bulk of the protein, resulting in a 714 amino acid protein, with a BTB/POZ domain and 5 C2H2 zinc fingers. 	  	   4	  responsive genes19.  The Wnt/β-catenin pathway is important in regulating developing and adult cells; misregulation of this pathway often leads to increased proliferation and cancer20.  HIC1 also interacts with the nucleosome remodeling complex SWItch/Sucrose Non-Fermentable (SWI-SNF), C-terminal Binding Proteins (CtBP) and Nucleosome remodeling and histone deacetylation complex NuRD, all of which are also involved in transcriptional control and oncogenesis17,21,22.  In quiescent cells, HIC1, complexed with a subunit of NuRD, binds and inhibits transcription of the G1-phase regulator cyclin D117.  Hic1 research in the Underhill lab The field of HIC1 research is relatively small with 201 publications appearing since 1995, and the main focus of these reports has been on cancer, with a few papers examining the role of Hic1 in development and none venturing into other disease models.  However, several groups report HIC1 expression in many normal tissues, which indicates HIC1 has a role to play outside of abnormal cancer cells1,23.  Data from the Underhill laboratory (unpublished work) indicates Hic1-expressing cells are present in all tissues with the exception of cartilage, ranging from being a rare population in organs such as the spleen, to comprising 10-20% of total cell counts in lung and fat.  As there are no well validated commercial antibodies available our lab has generated our own, as well as a conditional mouse line that, following Cre-mediated Hic1 deletion, leads to expression of a nuclear LacZ reporter (see Materials & Methods).  This line was used to create a heterozygous Hic1nLacZ/+ line (using a germline cre deleter), which has proven very useful in identifying and characterizing tissue-resident Hic1+ cells.  The labeled cells are, in many tissues, immediately adjacent to blood vessels in a 	   5	  perivascular location.  Early work in our lab has found that Hic1 expression is restricted to quiescent mesenchymal stromal cells.  The nLacZ reporter does not co-express with epithelial markers such as epcam (CD326), nor do these cells appear in epithelial layers (Figure 2), nor do epithelial-specific Cre driver lines (i.e. MMTVCre) lead to induction of Hic1 expression (data not shown).  nLacZ expression overlaps with standard mesenchymal markers such as Sca1, Vimentin and Pdgfrα (Figure 3).  MESENCHYMAL STEM CELLS – INTRODUCTION Mesenchymal stem cells (MSCs), also termed multipotent stromal cells and medicinal signaling cells, are currently the topic of much research24,25.  Scientists are studying the use of MSCs derived from bone marrow, adipose tissue, umbilical cords and other sources to attempt to treat conditions such as autism, multiple sclerosis, type 1 diabetes, skin ulcers, spinal cord injuries, heart and skeletal damage, osteoarthritis and graft versus host disease (GVHD)26–33.  However, there is little standardization and different groups that purport to take MSCs from even the same tissues may not be using the same cells, due to different markers and isolation protocols34.  Though MSCs from various tissues seem to have the same potential and possibly the same roles, there is no one universal marker that can identify them, though CD146+ CD90+ CD105+ CD73+ CD45- are common in some groups for identification of human MSCs, whereas in mice, CD34+ Sca-1+ Lin- PDGFRα+ are regularly used32,35.  Numerous fundamental questions in the MSC field remain unanswered as to the mechanism, efficacy, reproducibility and safety of these cells as therapies36–42.  For example, in one clinical trial of patients with 	   6	  	  	  Figure 2. Hic1LacZ is expressed near, but not within, epithelium. (A) Pancreatic duct, (B) small intestine, (C) mammary gland, (D) lung. Counterstained with Nuclear Fast Red. 400x images. 20!µmA BC D	   7	  	  Figure 3. Hic1-expression in comparison to PDGFRα-EGFP, CD31 and Vimentin. In many tissues including pancreas, Hic1+ cells show good overlap with PDGFRα-EGFP (A-F). Hic1+ cells are near endothelial cells (G-I) and are Vimentin+(J). JSmall Intestine/CD31Mammary Gland/CD31 Heart/CD31 Pancreas/Vimentin	   8	  myocardial infarction being treated with bone marrow-derived MSCs, Hare et al found that adverse events occurred at the same rate between the MSC treated and placebo treated groups, though the MSC treated group had increased pulmonary function and ejection fraction, however, they only reported data up to a 6 month time-point post treatment43.  Another trial, similarly dealing with bone marrow derived MSCs treated to patients after myocardial infarction, found that while early after treatment there were significant differences with regards to cardiac function, by 12 and 24 months post-treatment, there were no differences44.  Ambiguity also exists as to whether therapeutic effects are due to the MSCs themselves, or to the trophic factors they produce45–47.  Though clinical trial results have been by turns encouraging and conflicting, many MSC therapies are already in the clinic without sufficient evidence for efficacy and safety36–42.   Roles of MSCs in Disease Within tissues, under homeostatic conditions MSCs are typically quiescent.  It was previously thought that their only role was to participate in the natural turnover of certain tissues; however, a variety of other functions are now beginning to emerge for MSCs31–33.  Some groups have shown that MSCs are able to dampen immune responses, inhibit T-cell activation, proliferation and cytokine production48,49.  MSCs may also support angiogenesis33,50.  Due to well-defined cell bodies with numerous long and branched dendritic processes that envelope the vasculature, it is thought that MSCs can act as early sensors of tissue damage and maintain the tissue microenvironment31,32.  MSCs reside in multiple tissue compartments35,39,51, and upon tissue injury, are activated and leave the bone marrow, or their perivascular site within the tissue, and accumulate at 	   9	  sites of need where they produce pleiotrophic factors to support regeneration and/or tumorigenesis45,52–55.  Figure 4 outlines the various roles of MSCs. What is agreed upon is that MSCs are mulitpotent progenitors, ostensibly capable of differentiating into bone, cartilage, fat, loose and dense connective tissues, and possibly muscle and tendon39,45,51,56–58.  MSCs are most likely pericytes45, and are sources of myofibroblasts cells in many fibrotic disorders and cancers: kidney, lung, liver and pancreas, for example, making them an attractive target cell for putative treatments59–66.   REGENERATION VS REPAIR, IN GENERAL Organisms require a mechanism to respond to injury in order to prevent the injury from leading to death or dysfunction.  Researchers sometimes categorize the response into two main types: repair and regeneration47,54,67.  Repair is generally understood to consist of a fibrotic response with ECM production and scar formation, resulting in incomplete healing and impaired tissue/organ function68,69.  Regeneration, on the other hand, is described as the ability to heal the injured tissue/organ to a state similar to one prior to injury, to regain near pre-injury function69,70.  Regeneration often recapitulates aspects of development, with many similar pathways being involved55,70–73.  The aim of the tissue regeneration field is to find a way to prompt injured tissues to regenerate rather than repair.  Incomplete wound healing is thought to contribute to many disease pathologies and can itself be devastating even once the initial injury has resolved.  One such example is Idiopathic Pulmonary Fibrosis (IPF) in which there is a progressive fibrosis in the lungs due to mainly unknown causes; no successful treatment has been found for this fatal disease affecting mainly older people (above 65 years old), with a 	   10	  	  MSCsCellReplacementImmuneModulationAngiogenesisRegenerativeSupportAnti- scarringMitogenic/anti-apoptoticFigure 4. Roles of MSCs. 	   11	  prevalence between 14-63 cases per 100,000 (world-wide average)74–76.  Another reason for the intense interest in MSCs and regeneration is the link to aging: as we age, the frequency and potential of MSCs decreases and we switch from regeneration to repair, which may explain why diseases seem exacerbated in the elderly45,47,77.  Groups have reported that defects in MSCs are behind the accelerated aging disease, Hutchinson-Gilford progeria syndrome (HGPS)78,79.  Scaffidi and Misteli found that progerin (the mutant form of lamin A present in HGPS) agonizes Notch signaling in MSCs, affecting their differentiation potential80.  Zhang et al reported on patient-derived induced pluripotent stem cell (iPSC) lines they created to study HGPS; progerin levels were maximal in the mesenchymal lineages, though differential potential was not affected79.  Understanding the link between aging and impaired regeneration is becoming more important as the population in many developed countries is growing older and the prevalence of fibrotic diseases is increasing53,75,81. Researchers still argue over the origin of the effector cells, fibroblasts or mesenchymal cells, in many fibrotic diseases such as IPF and kidney fibrosis, though some have been identified with various convincing evidence, such as in liver and pancreatic fibrosis59,64,82–84.  The cell of origin may not ultimately matter, however, in cases of full-blown fibrosis, where treatments must be aimed at mitigating the fibrotic factors, rather preventing fibrosis that has already begun. Though the myofibroblast/activated MSC is certainly a key player, other cells play important roles.  Immune cells produce inflammatory factors such as TGF-β which can perpetuate the fibrotic process85.  The epithelium in many tissues is often the first to be made dysfunctional and can be the initiator of fibrotic diseases - for example, a murine 	   12	  model of IPF uses the chemotherapeutic bleomycin to induce epithelial cell death, thereby causing preliminary inflammation that later leads to fibrosis86–88. The goal of the Underhill Laboratory is to understand the role of HIC1+ cells in regeneration and aging, and determine how HIC1 regulates MSC biology.  Numerous injury models have been used to demonstrate activation of Hic1+ cells following damage.  Furthermore, Hic1 deleted animals show a prominent phenotype in the pancreas which includes increased numbers of white adipocytes.  Thus, I decided to more carefully examine the distribution, nature and function of Hic1+ cells in the pancreas.    Pancreatic Stellate Cells In 1876 Karl von Kupffer noted that the liver contained cells with a star-shaped morphology that had a high concentration of lipid droplets.  These cells were named hepatic stellate cells (HSCs) due to their shape, or Ito-cells; the lipid droplets have subsequently been demonstrated to be stores of vitamin A derivatives, mostly retinyl esters89–91.  In 1982 Watari and colleagues identified similar cells in the pancreas by feeding mice vitamin A rich diets, which increased the size of the lipid droplets and allowed those cells to exhibit autofluorescence characteristic of the presence of retinoids63.  Those pancreatic stellate cells, or PSCs, are estimated to represent 4-7% of pancreatic cells and are found in periacinar, periductal and perivascular locations92,93.  In addition to being in liver and pancreas, stellate cells have also been identified in kidney and lung63,94. PSCs have common characteristics that allow for identification of these cells as opposed to pancreatic fibroblasts (PFB) and other pancreatic cell types  (see Table 1).  In 	   13	  general, under normal conditions PFBs are desmin and smooth muscle actin alpha (α-SMA; Acta2) negative and are more spindle-shaped 93.  As noted in their original   Table 1. Characteristics of Pancreatic Mesenchymal Cells* Characteristic Pancreatic Fibroblast Pancreatic Stellate Cell (quiescent) Pancreatic Stellate Cell (activated) Morphology Elongated stellate-shaped cell Prominent cell body, with dendritic processes Large, thick, with dendritic processes Proliferation -/√ - √√√ Vitamin A Droplets - √√√ - α-SMA -/√ - √ Desmin - √ √ Vimentin √ √ √ NG2 (Cpsg4) - √ √ GFAP - √ √ Nestin - √ √ Collagen/ECM √ - √√ NGF - √ √ NCAM - √ √ * With information from 64,95.  discovery, stellate cells in the healthy pancreas contain vitamin A droplets.  PSCs have been shown to exhibit variable expression of nestin, vimentin, NG2 and desmin64,92,93,95–97.  While the presence of glial fibrillary acidic protein (GFAP) has been used to identify HSCs, its use in locating PSCs has been more challenging:  once these cells are isolated and cultured (and are thereby displaying an activated phenotype) they are GFAP-positive27, however, few papers have shown convincing staining on in vivo samples.  A GFAP-LacZ transgenic mouse showed few LacZ-positive cells on poorly stained sections, though the LacZ-positive cells were in the proper periacinar location36.  As well, though it had been understood that HSCs did express GFAP, a recent fate-tracing paper 	   14	  indicated that GFAP was a poor marker for HSCs66.  The conflicting data of GFAP expression may also partially be due to differences between species – human, rat and murine stellate cells may not be identical with regards to markers.  Upon activation PSCs deplete their vitamin A droplets, up-regulate α-SMA and nestin expression and extracellular matrix protein production64,93,95.  PSCs play important roles in pancreatic cancer and pancreatic fibrosis, where, following activation, they proliferate and produce collagen and other ECM proteins93,95,97,99,100.  PANCREAS STRUCTURE, FUNCTION & HISTOLOGY The pancreas is the organ responsible for blood glucose regulation, and dysfunction of that role can lead to diabetes and obesity27,101.  Less discussed is the exocrine function, namely the production of digestive enzymes by acinar cells.  As diabetes is associated with loss of β cells within islets, much of the focus on pancreas function has been on this cell population, and less so on other important cell types27,101.     Pancreatic structure & cell types A mouse pancreas generally makes up ~1% of total body weight, weighing around 0.18 g in a young adult C57Bl/6 mouse102.  A human pancreas weighs between 85-100 g in adults, also representing ~1% of total body weight103.  Structurally, the mouse pancreas is a lobular organ which appears much more amorphous than a human pancreas.  The general structure consists of groups of acinar cells, called acini, which secrete digestive enzymes into interlobular ducts that drain into the main pancreatic duct104,105 (Figure 5).  Islets, which make up roughly ~1% of the pancreatic mass, are interspersed throughout the pancreas106.	   15	  	  	  Figure 5. Basic murine pancreas structure and histology. A wholemount image of a murine pancreas is shown, H&E stained 100X and zoomed in areas of duct, acini and islet. DuctAciniIslet	   16	  The endocrine function is accomplished by the islets, which are composed of 5 cell types, named α, β, γ (or PP), δ and ε-cells104,107.  They are responsible for glucagon, insulin, pancreatic polypeptide, somatostatin and ghrelin production, respectively104,107.  There are differences in pancreas structure as well as islet composition and organization between species, and even between different mouse strains; however, as this thesis is only minimally concerned with islets, those differences will not be delved into here102.  The remainder of the pancreas is composed of neuronal cells, ductal cells, acinar cells and fibroblast/ mesenchymal cells.  The former cell types can be marked, for example, by Tuj1, cytokeratin 19 and amylase, respectively108.  The fibroblast/ mesenchymal compartment, however, requires more complex identification.  Fibroblasts are generally identified by their location (outside the acini and ducts) and morphology, as well as the fact that they reside within the interstitial space and are not typically closely associated with blood vessels, and are often α-SMA positive when cultured (Table 1).  The ducts are lined with cuboidal epithelial cells, with loose connective tissue separating most of the ducts from the acini and islets.  The acini produce bicarbonate and digestive enzymes which are contained in organelles called zymogen granules, often in a pro-enzyme form109–111.  Upon stimulation the enzymes are released into the intralobular and interlobular ducts and activated.  PANCREATIC DISEASES All of the aforementioned pancreatic cell types play important roles in organ and organismal health; understandably, there are many types of diseases resulting from malfunction of these cells.  The most common and well-known pancreatic disease is 	   17	  diabetes mellitus, with a resultant dysregulation of glucose metabolism which can negatively affect many other organs and tissues101.  Several cancers can also arise from the exocrine pancreas, the most common being adenocarcinoma, though some neuroendocrine cancers may also arise from islet cells101 (Table 2).  Many pancreatic diseases are relatively rare and are often asymptomatic until later stages of disease.  Complicating these diagnoses is the mostly inaccessible organ location.  One such disease for which this is true is chronic pancreatitis (CP).  Table 2. Types of Diseases Affecting the Pancreas Disease Cell Type Responsible Prevalence / Incidence Notes Risk Factors Adenocarcinoma (Pancreatic Ductal Adenocarcinoma) Ductal cells Varies among countries 8-40 per 100K incidence112,113  § Lowest 5-year survival of any major cancer (5%)114 § 90% of pancreatic cancers are adenocarcinoma114,115 § Smoking, diet, genetics, age112 Diabetes mellitus β-cells, or all body cells Over 380 million adults world-wide116 § Results in many secondary diseases (obesity, organ failure)  § Obesity, diet, genetics, pregnancy, chronic pancreatitis Pancreatitis Acinar Cells 50 per 100K prevalence113 5-12 per 100K incidence113,117,118 § Chronic pancreatitis can often lead to pancreatic cancers111,119,120 and diabetes mellitus § Alcohol, smoking, genetics, diet, idiopathic, hypercalcemia117,120,121 	  Chronic pancreatitis & fibrosis Chronic pancreatitis is defined as “an inflammatory process that leads to the progressive and irreversible destruction of exocrine and endocrine glandular pancreatic parenchyma which is replaced by fibrotic tissue”118,122.  	   18	  CP can take several years to fully develop, mainly being diagnosed past middle age.  Though a large percentage (over 70%) of cases can be found to have links to alcoholism119, many cases are familial in nature or idiopathic118,122.  It is challenging to diagnose the early stages of the disease as it can sometimes be mistaken for an episode of acute pancreatitis, and patients often may have negative test results121.  An additional complexity may be that patients who have recurrent episodes of acute pancreatitis will develop chronic pancreatitis later on.  Though the incidence and prevalence of CP are relatively low, they result in repeated hospital admissions and loss of productivity or employment, and have a negative impact on patients’ quality of life, as the disease is very painful and can affect metabolism, leading to exocrine and endocrine insufficiency113,117,118,121. Due to lack of effective treatments physicians often attempt administration of pain medication, antioxidants and/or enzyme supplements, or recommend surgeries including pancreatectomy121.  One common hypothesis is that alcohol, trauma, an obstruction (calcification or gallstones) or other disturbance interferes with normal acinar and ductal functions such that digestive enzymes are released or activated prematurely117,121.  Other hypotheses include the idea that oxidative stress or intrapancreatic lipid deposition are responsible, or some other event that causes necrosis, leading to pancreatitis and fibrosis117.  The latter hypothesis may gain traction, as an animal model found that necrosis caused by an infection could worsen pancreatitis123.  A unifying feature of all these hypotheses is that, no matter the initial mechanism, the pancreas undergoes self-digestion, resulting in inflammation, tissue atrophy, and with prolonged bouts of injury, eventually fibrosis. 	   19	  Many researchers have determined that PSCs are the effector cells in many pancreatic diseases, contributing to an increased need to better understand these cells and their roles in order to find therapeutic targets64,124–127.  PANCREATIC DAMAGE MODELS There are several murine in vivo pancreatic damage models that typically involve induction of pancreatic self-digestion, and the models exhibit varying degrees of severity and time to presentation.  These models fall within a few categories, namely: surgically-induced; diet-induced; immune-modulated and secretagogue-induced.  Generally, some damage or abnormal signaling is required, which causes irregular zymogen activation or breakdown of paracellular membranes.  Those freed, activated pancreatic enzymes in turn cause cell death and inflammation, as well as impaired blood flow and edema111.  Pancreatic duct ligations or obstructions are surgical methods of inducing pancreatitis.  These models follow a component of the human disease, as gallstone occlusion of the human duct has been known to be a precursor of pancreatitis; autopsies of some patients have found a similar phenomenon111,113,122,128.  Ductal flow is blocked, causing a backup of secretory outflow111.  The resultant release of Ca2+ and activation of zymogens leads to localized damage and pancreatitis111.  Another surgical model relies on the injection of sodium taurocholate or trinitrobenzene sulfonic acid (TNBS) into the pancreatic duct, leading to similar results111.  Some groups report that models with only bile acid infusion do not result in as consistent and severe a pancreatitis as those that are combined with ductal ligation or obstruction64. 	   20	  Diet-induced models include feeding mice alcohol or a choline-deficient diet, enhanced with ethionine (CDE diet).  The ethanol produces reactive oxygen species (ROS) which then damage the acinar cells and activate PSCs119.  As ~70% of CP cases may be linked to alcohol, this model is particularly apt119. The CDE diet unfortunately results in near 100% mortality, and in male mice requires estrogen administration109,111. A third category is that of immune-modulated pancreatitis models.  These models mimic human autoimmune pancreatitis, which is quite rare (prevalence ~ 1/100,000 people) with mainly elderly people being affected129.  These models treat transgenic mouse lines, for example IL-10-/- or MRL/Mp mice, with polyinosinic: polycytidylic acid to speed up the disease progression129–131.  Another model in the immune-modulated category uses infectious agents to induce pancreatitis111.  Infection with Coxsackie virus group B variants results in acute pancreatitis that can develop into chronic pancreatitis111. Lastly is secretagogue-induced pancreatitis109,111,119.  These models have the benefit of speed, with pancreatitis resulting within 1-2 days after injection, as well as not requiring complex surgical techniques.  L-arginine injection results in acute necrotizing pancreatitis, though a lower dosage can cause chronic pancreatitis.  Serial injection of the choleocystokinin (CCK) analog caerulein with a supramaximal dose can cause a milder pancreatitis.  CCK binds to cell receptors on acinar cells64,132 and potentially on PSCs; both human and rat PSCs express CCK2 receptor, but rat PSCs do not express CCK1 receptor93,133.  At maximal doses, caerulein induces pancreatic secretion and increased protein synthesis, depleting enzyme stores109,111,119.  At supramaximal doses, however, secretion is reduced leading to increased concentrations of activated pancreatic enzymes within the pancreas, resulting in pancreatic damage109,111,119.  Caerulein can be 	   21	  administered by continuous intravenous infusion or serial intraperitoneal injection.  Caerulein is the most common model of pancreatitis, as the animals display little pain or weight loss, the damage is mild, the inflammation is homogeneous and no special expertise is required.  Normally, treatment is hourly over 7-10 hours for one day only, with inflammation peaking within 24 hours109,111.  Longer-term treatment of caerulein (2-8 wks) is also the most common model for chronic pancreatitis.  The caerulein model is a good mimic of the human disease (both acute and chronic), as a strong hypothesis is that intra-pancreatic zymogen activation and tryptic digestion causes pancreatitis.  Another model for chronic pancreatitis utilizes a single injection of dibutyltin dichloride (DBTC) but requires two months for the disease to peak109,111,119.  SUMMARY, OBJECTIVES & HYPOTHESES To summarize, fibrosis is a common factor of chronic inflammatory diseases and is thought to reflect over-exuberant repair or incomplete regeneration68,69.  MSCs are thought to participate in homeostasis as well as pathogenesis in fibrotic disease but many questions remain as to their identity and specific roles53–55,81,134.  In ongoing studies in the Underhill lab we have found that Hic1 expression marks quiescent MSCs in different tissue compartments.  In this thesis I have taken advantage of our unique mouse models to study Hic1-expressing cells in the pancreas and their potential role in pancreas homeostasis and regeneration.  The overarching hypotheses are: 1) Hic1 expression marks PSCs, 2) Hic1 expression is required in pancreatic tissue homeostasis and regeneration, and  	   22	  3) Loss of Hic1 will negatively affect tissue homeostasis and regeneration. 	  The objectives of this work were therefore three-fold: 1) Characterize Hic1+ cells in the pancreas. Hic1-expressing cells were characterized by immunofluorescence and flow cytometry with regard to expression of PSC markers.  2) Determine the fate of Hic1+ cells during pancreatic damage and regeneration. Hic1-expressing cells were genetically labeled with a fluorescent reporter and tracked over time and through pancreatic damage and regeneration.  3) Determine the effect of loss of Hic1 expression in pancreas homeostasis and regeneration. Mice that were Hic1-deleted were monitored and pancreas histology analyzed for phenotypic changes. Mice were also challenged with caerulein-induced pancreatic damage, and the extent of regeneration studied.   	   23	  CHAPTER 2: METHODS & MATERIALS MICE All mice were housed in groups, given laboratory chow and water ad libidum, with a 12 h light/dark cycle. For characterization work, mice were generally 8-12 weeks old.  For caerulein-induced pancreatic damage, mice were treated between 15-20 weeks old.  Mouse strains used are listed in Table 3.  A schematic of the Hic1 strains is shown in Figure 6.  Table 3. List of Transgenic Mice Mouse Strain Background Source Wild-type C57Bl/6 From Jackson Laboratory, maintained in-house UBCCreERT2/+; Hic1f/f C57Bl/6 UBCCreERT2/+ from Jackson Laboratory, Hic1f/f were generated under contract for the Underhill lab by genOway Inc. Hic1citrine/+ Mixed 129/SvJ & C57Bl/6 – backcrossed onto C57Bl/6 Dr. V Korinek23 Hic1LacZ/+ C57Bl/6 Derived from Hic1f/f mice using a germline Cre-deleter mouse. Hic1CreERT2/+; Rosa-tdTomatof/+ C57Bl/6 Hic1CreERT2/+ were generated under contract for the Underhill lab by genOway Inc, and then crossed to Rosa-tdTomato mice, which were obtained from Jackson Laboratory PDGFRaH2B-EGFP/+ C57Bl/6 Originally from Dr. P. Soriano,	  ordered	  from	  Jax	  Laboratories pOBCol3.6GFP C57Bl/6, CD-1 mixed From Dr. David Rowe135 Hic1CreERT2/+; Rosa-DTAf/+ C57Bl/6 Hic1CreERT2/+ were generated under contract for the Underhill lab by genOway Inc, and then crossed to Rosa-DTA mice, which were obtained from Jackson Laboratory  	   24	  	  	  	  	  Figure 6. Various Hic1 mouse lines. 	   25	  Genotyping: Ear punches were taken from mice between 3-8 weeks of age, DNA isolated and genotyping was performed by PCR amplicon analysis by Taka Murakami of the UBC Genotyping Facility.  Primer sequences are listed in Table 4.  Table 4. List of Primers for Genotyping Gene Forward Primers (5’-3’) Reverse Primers (5’-3’) Cre GCG GTC TGG CAG TAA AAA CTA TC GTG AAA CAG CAT TGC TGT CAC TT GFP AAG TTC ATC TGC ACC ACC G TCC TTG AAG AAG ATG GTG CG Citrine GAT GGA GGC GCC TAT GGT GA GGC GAA GCA CAT CAG GCC GT Tomato WT AAG GGA GCT GCA GTG GAG TA CCG AAA ATC TGT GGG AAG TC Tomato mutant GGC ATT AAA GCA GCG TAT CC CTG TTC CTG TAC GGC ATG G Hic1 TCC GAA AGG GTC ATT TCG CCC TAA TGT CAT GGT CCA GGT TTA GCA GGT AND TGA AGA GTT TGT CCT CAA CCG CGA   Glucose measurements: To monitor any potential impact on pancreas endocrine function, blood glucose levels were monitored bi-weekly by collecting a small drop (< 10 µl) from the tail using a tail poke SOP.  Glucose was measured with a OneTouch® UltraMini® Blood Glucose Meter from LifeScan Inc.  PANCREATITIS DISEASE MODEL Pancreatic Damage: Pancreatic damage was induced by intraperitoneal injections of the CCK analog caerulein.  Though this model is normally used to induce pancreatitis and thereby study inflammation, injections at high enough concentrations, or repeated, can cause pancreatic fibrosis.  Injection repeated over two weeks or more can lead, over time, 	   26	  to a pancreatic cancer model.  For our purposes, 8 hourly injections in one day (at 50 µg/kg) were enough to create focal lesions in the pancreas at day 3-5, which were completely resolved by day 8.  To get more robust widespread damage, three days of such injections over 5 alternating days was performed.  Whichever model was used, the mice tolerated it very well, with minimal weight loss and few visible signs of discomfort.  The date of the last IP injection was termed day 0.  REAGENTS Caerulein was ordered from Sigma-Aldrich (cat # C9026) and aliquots were made at 500 µg/ml and stored at -20o C.  Tamoxifen (TAM) was obtained from Sigma-Aldrich (cat # T5648), stored at 4o C, with fresh stock made weekly at 20 mg/ml in corn oil.  Stocks were stored at 4o C, and re-warmed at 37o C in a rotating chamber prior to use.  Animals were injected IP at ~2 mg/ 20 g mouse.  Three days of injection were performed to label cells (i.e. Rosa-tdTomato), five days of serial injections were completed for conditional deletion of Hic1.  Experiments were started, or animals were collected, 5 days after the end of labeling.  STAINING Sample processing: Tissue samples were processed in various ways, depending on whether tissues needed to have LacZ enzyme or endogenous fluorescent protein preserved, and the type of embedding into either paraffin or frozen blocks.  In all cases, after the animal was euthanized the heart was perfused using a 27-gauge needle with 10 	   27	  mL of PBS (with 2 mM EDTA), followed by perfusion with 10 mL of the relevant fixative.   Non-reporter tissues: the fixative used was 10% neutral buffered formalin.  Tissues were fixed at least overnight before washing and preparing for paraffin embedding.  For LacZ-containing samples: a ‘LacZ fixative’ was used.  This was made in-house and consisted of 2% paraformaldehyde, 2 mM MgCl2, 1.25 mM EGTA, 0.2% glutaraldehyde in PBS, at pH 7.4.  Tissues were fixed for 2-3 h and then washed and either whole-mount stained in LacZ staining solution (see below) followed by further fixation and paraffin embedding, or prepared for cryoembedding (see below), and LacZ stained on the slides.   For fluorescent protein containing samples, the fixative was 2% PFA.  Tissues were fixed for 48 h, and then washed and prepared for cryoembedding as below.  Formalin-fixed paraffin embedded (FFPE) samples: after fixing, tissues were washed minimum 3 x 30 min in PBS, and then placed into 70% ethanol minimum overnight.  Tissues were dehydrated in 95% ethanol, two washes of 100% ethanol, xylene (Fisher Scientific, cat # X3S4), 50:50 xylene:paraffin (in 60o C oven), paraffin (Fisher Scientific, cat # 23-021-401) for 1 h each step, and then kept in fresh paraffin overnight prior to embedding the next day.  Tissues were typically sectioned at 6 µm with a Leica RM2255 microtome and subsequently stored at room temperature.  	   28	  Cryoprocessing: after fixing, tissues were washed minimum 3 x 30 min in PBS, and then placed into 10% sucrose in PBS for 1 h, followed by 20% sucrose and 30% sucrose overnight prior to cryoembedding in OCT (Sakura Tech).  Tissues were sectioned at typically 6 µm with a Leica CM3050 S cryostat at -18o C, and stored at -80o C until use.  Staining: Whether the slides were used for basic histological stains or immunofluorescence (IF), the following procedures were used.  For cryoslides, slides were thawed for 30 min at room temperature (in the dark for IF), and washed 3 x 10 min with PBS, after which slides were stained with X-gal solution if required.  For FFPE sections, slides were dewaxed and rehydrated through 2 x 7 min xylene, 2 x 5 min 100% ethanol, followed by 5 min of 95% and 70% ethanol solutions, and 2 x 2 min distilled water.  LacZ staining: For in situ enzymatic staining on cryoslides, after thawing and PBS washes, the slides were placed into a 37o C chamber in a humidified slide box, and 500-750 µl LacZ staining solution pipetted onto each slide.  The solution was made up the same way regardless of whether it was for whole-mount or cryoslide staining.  Stocks at 10X were used to prepare a final solution that contained 20 mM MgCL2-6H2O, 0.02% Nonidet P-40, 0.01% sodium deoxycholate, 10 mM potassium ferrocyanide, and 10 mM potassium ferricyanide.  Slides were stained either for 4 h (at which time successful staining would be verified) or overnight.  Slides were then washed in PBS 3 x 10 min, counterstained with Nuclear Fast Red (Vector Labs, H-3403) for 5 min, and dehydrated, brought into xylene and coverslipped. 	   29	   H&E: Once in distilled water slides were transferred to hematoxylin (Vector Labs, cat # H-3401) for 2 min, rinsed in tap water, and counterstained with eosin (Sigma-Aldrich, cat # E4382) for 5 min, dehydrated in ethanol, placed into xylene and mounted with Cytoseal 60 (Richard-Allan Scientific, cat # 8310-4).   Immunofluorescence: In brief, to minimize autofluorescence in the various tissue samples, histological sections (frozen or FFPE) were treated for 1h with 10 mg/ml sodium borohydride, followed by 2 x 2 min PBS washes.  Slides were blocked with 2.5% serum (rabbit or goat depending on the species of the primary antibody) and 2.5 % BSA in PBS for 1 h or overnight.  Primary antibodies were used as at the dilutions indicated in Table 5.  Secondary antibodies were typically used at 1:1000, and included goat-anti rat or goat-anti rabbit conjugated to Alexa 488, Alexa 594 or Alexa 647 (Invitrogen/Life Technologies).  Slides were counterstained with DAPI (1:500) and mounted with Aqua Poly/Mount media (Polysciences Inc, cat # 18606-5).   	   30	  Table 5. Primary Antibodies Antigen Manufacturer Catalogue # Concentration Notes αSMA Abcam Ab5694 1:200 Rabbit CD31 BD Biosciences 550274 1:50 Rat Collagen I Abcam Ab34710 1:250 Rabbit Collagen IV Abcam Ab6586 1:250 Rabbit Desmin Abcam Ab8592 1:80 Rabbit GFAP Dako Z02334 1:300 Rabbit Laminin Abcam Ab11575 1:50 Rabbit Nestin Stem Cell Technologies 60051.1 1:50 Rat NG2 (Cpsg4) Millipore Ab5320 1:200 Rabbit Perilipin Cell Signaling 9349 1:200 Rabbit Sca1 Abcam Ab51317 1:100 Rat, 2h room temp S100A4 Abcam Ab27957 1:400 Rabbit Vimentin Cell Signaling 5741 1:100 Rabbit  HIC1+ CELL ISOLATION The pancreas was mechanically and enzymatically digested to allow for isolation of live Hic1+ cells.  The protocol was based on Apte’s protocol, with modifications92.  An enzyme cocktail for tissue digestion was made up with 11 mL per mouse pancreas containing 1 mg/ml protease E (Sigma Aldrich, cat # P5147), 1 mg/ml Collagenase P (Roche, cat # 11249002001) and 10 µg/ml DNase1 (Promega, cat # Z358A) in HBSS.  After euthanization by fatal avertin overdose, followed by perfusion through the left ventricle with sterile saline, the pancreas was gently excised and placed in a petri dish on ice.  Using tweezers to gently hold the pancreas, 3 mL of the digestion solution was injected through a 3 mL syringe with a 27 G needle, and several injections were performed throughout the tissue, to expand and fill the lobes.  Next, the pancreas was placed in the remaining 8 mL of digestion solution in a 15 mL falcon tube and shaken at 37o C, 200 rpm for 2 min; 120 rpm for 3 min (Forma Scientific Orbital Shaker), and then minced by hand with scissors for 3 min.  The tissues was placed back into the same 	   31	  digestion solution and shaken at 120 rpm for 5 further minutes.  Then the digested tissue was disrupted sequentially with a 5 ml pipet followed by a 2 ml pipet, and strained through a 40 µm cell strainer.  For flow cytometry (FCM) or Fluorescence-activated cell sorting (FACS), strained cells were added to 30 ml FACS buffer (2 mM EDTA, 5% FBS in PBS).  The tube was topped up to 50 ml volume, and spun at 450 g for 5min at 4o C, re-suspended in FACS buffer and re-filtered into a FACS tube before spinning again.  At this point, cells were ready for plating, antibody staining, or FCM analysis.  Antibody staining for FCM/FACS: For both FCM and FACS, cells were stained with a combination of the following antibodies for 30 min: 1:400 PerCP-conjugated anti-mouse CD45 (BD Pharmingen, cat # 557235), 1:30 APC-conjugated anti-mouse Podocalyxin (R&D Systems, cat # FAB1556A), 1:300 PeCy7-conjugated CD31 (eBiosciences, cat # 25-0311-82), 1:100 efluor450-conjugated epcam (eBiosciences, cat # 48-5791-82), and 1:100 PeCy7-conjugated Sca-1 (eBiosciences, cat # 25-5981-82).  For cell proliferation analysis, mice were IP injected with 1 mg EdU for 4 days prior to analysis.  Life Technologies EdU Click-iT kits (cat # C10420) were used.  In brief, cells were fixed for 15 min with 4% PFA, incubated in saponin permeabilization & wash reagent for 25 min, and then incubated with the EdU Click-iT reaction buffer for 30 min.  Flow cytometry and FACS: Samples were run on an LSRII (BD).  Flow cytometry data was analyzed and plotted with FlowJo v 10.0.6 (Tree Star Inc).  All FACS samples were processed through UBCFlow Cytometry Facility, using a BD Influx sorter.  	   32	  MICROSCOPY & IMAGE COLLECTION All imaging (other than live cells) was performed on an Olympus BX63 microscope using either an Olympus DP72 camera (brightfield and some fluorescent images) or an Olympus XM10 camera (fluorescent images).  Olympus’ CellSens program was used for image collection and editing, along with Adobe Acrobat’s Photoshop.  Live cells were imaged on a Zeiss Axiovert S100 microscope with a QImaging Retiga 1300 camera.  Image J software was used for nuclei/cell counts and tomato area quantification by thresholding.  CellSens software (Olympus) was used for analysis of fibrotic areas.  In brief, sections 654 µm apart were H&E stained, imaged at 40X magnification, and the fibrotic area measured, as well as the total pancreatic area.  Percent fibrosis was calculated by dividing the fibrotic area by the total area.  Total pancreatic volume was calculated by multiplying the total pancreatic area of each slide by the distance between the imaged sections.  Statistical analysis: T-tests and ANOVA were performed using Prism software.  For ANOVA, Boneferroni post-hoc analyses were used.      	   33	  CHAPTER 3: RESULTS HIC1+ CELL CHARACTERIZATION MSCs are thought to play a variety of important functions in tissue development, growth, homeostasis, regeneration and diseases31,33,53,54,77.  Preliminary studies in the Underhill lab aimed to determine whether Hic1-expressing cells may be MSCs.  Hic1 expression was found to overlap with MSC markers in various tissues and herein this was more rigorously evaluated in the pancreas, in addition to defining the relationship of Hic1+ cells to PSCs.  Initial examination of the Hic1nLacZ/+ (Figure 6) reporter in histological sections showed no X-gal staining in cells that were overtly epithelial (i.e. cuboidal epithelium in lung airway or small intestine) (Figure 2).  Immunohistochemistry on X-gal stained slides found that nLacZ+ cells were negative for epithelial markers, such as Muc1 and SPC in the lung (data not shown).  Further preliminary work showed that Hic1+ cells, in various tissues, appear to co-express commonly accepted markers of MSCs such as PDGFRα, Sca-1 and vimentin, suggesting that MSCs and Hic1-expressing cells are a common cell population (Figures 2, 3 & 7).  The cells exhibit mesenchymal/fibroblastic morphology, form CFU-F colonies in vitro and are capable of extensive expansion (Figure 8).  HIC1+ CELL CHARACTERIZATION – PANCREAS SPECIFIC An important mesenchymal cell type in the pancreas is the PSC and we sought to address if Hic1+ cells represent PSCs.  We first examined the distribution and frequency of Hic1+ cells and in comparison to reported data, there was good concordance between the distribution and frequency of both cell populations.  The literature records that PSCs 	   34	  	  Figure 7. Characterization of pancreatic Hic1+ cells by flow cytometry. Pancreatic cells from Hic1ci/t+ mice were isolated and stained for various markers. The citrine-positive cells did not express the endothelial markers CD31 nor podocalyxin. They did not co-express the epithelial marker CD326 (Epcam), nor were they hematopoietic cells (CD45). 93 ± 3% of the citrine+ cells were positive for Sca-1. CD31CitrinePodocalyxinCitrineCD326CitrineCD45Citrine CitrineSca1	   35	  	  Figure 8. tdTomato+ cells in cell culture. (A) tdTomato+ cells isolated from pancreas one day after plating. (B) The same well after 1 month, showing these cells can enter the cell cycle and proliferate many times. (C) The morphology of the tdTomato+ cells is mesenchymal/ fibroblastic. (D) Lipid droplets are present in tdTomato+ cells.	  100x	  images.	  A BC D	   36	  make up 4-7% of cells within the pancreas92,93, and our data showed 7.8 ± 2.1% by histology, and 3.8 ± 0.3% of all cells by flow cytometry, or 4.8 ± 0.4% of CD45- cells. Furthermore, PSCs are described as being in periacinar and perivascular locations, and Hic1+ cells were predominantly located at these sites (Figure 9 A, B).  Thirdly, Hic1+ cells expressed many of the markers indicative of PSCs.  Hic1-expressing cells, as identified by tamoxifen treated Hic1creERT2/+; tdTomatoflox/flox, were immunostained for commonly accepted PSC markers including NG2, nestin and desmin (Figure 9D-F). Though there was generally not complete overlap seen with any of these markers and tdTomato, a large degree of co-expression was observed (Figure 9H).  Oil-injected controls showed no tdTomato+ signal – see Figure 10.  To date, no single marker has been identified as being PSC specific; instead it is the analysis of a constellation of markers that defines PSCs.  Unfortunately, this does not adequately take into account the variable and non-uniform expression of PSC markers, which may be related to cellular state (i.e. activation).  For instance, various studies indicate that similar to HSCs, GFAP is expressed in quiescent and activated PSCs.  However, a GFAPLacZ transgenic reporter mouse showed limited LacZ expression136, inconsistent with reports of PSC cell frequency92,93, indicating GFAP may not be a reliable marker for PSCs.  In addition, our own analysis showed no appreciable GFAP reactivity in the pancreas, whereas control sections of both brain and liver showed expected GFAP expression (Figure 11).  TdTomato-labeled Hic1+ cells (from TAM treated Hic1creERT2/+; tdTomatoflox/flox mice) did exhibit perinuclear lipid droplets both in vitro (Figure 8), and somewhat more rarely, in vivo, consistent with published reports of this cell type.  Together, these findings indicate that Hic1+ cells overlap extensively with PSCs.   	   37	  	  	  	  	   	  TdTomatomerge CD31Desminmerge TdTomatoDLamininmerge TdTomatoBmerge TdTomatoPDGFRα-EGFPCA	   38	  	  Figure 9. Characterization of Hic1-expressing cells in healthy tissue. tdTomato+ pancreas cryosections were stained for various markers. The tdTomato+ cells are near the endothelial cells, but are not endothelial cells (A). They were outside the basal lamina (B). They co-expressed the markers PDGFRα-EGFP (C), Desmin (D), Nestin (E) and NG2 (F). They did not express collagen-GFP (G). 400X images from several samples were quantified for the number of double positive cells as a % of either all marker+ cells or of all tdTomato+ cells (H). A similar analysis was done with flow cytometry data for Sca-1 stained cells on pancreata from Hic1ci/t+ mice (n = 3). NG2merge TdTomatoCollagen-GFPmerge TdTomato20+m20+mNestinmerge TdTomatoEFGH0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Desmin PDGFR_ NG2 Nestin Sca1 % of marker+ cells % of Tomato+ cells % of marker+ cells % of Citrine+ cells 	   39	  	  	  Figure 10. No tdTomato label is present without tamoxifen administration. Hic1CreERT2/+; Rosa-tdTomatoflox/+ mice were injected with oil (left panels) or TAM. Without TAM, the tdTomato reporter is not activated. Top panels are wholemount images on a dissection microscope. Bottom panels are from 200X confocal images. OILToto-3, confocalTAMWhole-mount	   40	  	  tdTomato GFAP100!µmLiver Brain PancreasFigure 11. GFAP antibody staining. Cryosections of liver, brain and pancreas from Hic1CreERT2/+; Rosa-tdTomatoflox/+ mice were stained with the GFAP antibody from DAKO (green). No signal was seen in the pancreas sections, but the positive control tissues of liver and brain had signals. 	   41	  Another feature of PSCs is that they are normally quiescent in healthy tissue.  TdTomato expression was induced in Hic1creERT2/+; tdTomatoflox/flox mice with TAM injection at 8 wk of age, and the frequency and distribution of tdTomato+ cells was analyzed for up to 180 days afterward.  Pancreatic tdTomato+ cells exhibited some expansion, with the d90 and d180 samples showing a ~2-3 fold increase of tdTomato+ cells compared to the d0 sample (Table 6).  There was a negligible difference between the d90 and d180 samples (Figure 12).  This finding is congruent with reports in the literature indicating that PSCs are normally quiescent and exhibit limited turnover in healthy tissue.   Table 6. Frequency of tdTomato+ Cells in Lineage-Tracked Samples Statistic D0 D90 D180 Fold Change Area Fraction (%) 0.4 ± 0.1 1.2 ± 0.4 0.9 ± 0.2 2-3x tdTomato+ cells (% of DAPI cells) 4.1 ± 1.5 11.7 ± 4.1*** 13.3 ± 3.0*** 2-3x ±SE, *** p < 0.0001, when comparing d0 to d90, or d0 to d180   HIC1 DELETION Effect of Hic1 deletion on homeostasis Previous reports have shown that complete loss of Hic1 leads to embryonic lethality, whereas heterozygous Hic1 mice exhibited an increased incidence of tumours of various origins4,137–139.  A conditional knock-out mouse line was generated (Hic1flox/flox) and bred to UBCCreERT2 mice to produce UBC-CreERT2+/-; Hic1flox/flox mice.  We assessed the effect of Hic1 deletion on adult mice.  Once mice reached 8 weeks of age they were given five daily IP doses of 2 mg TAM to induce Hic1 deletion. Control 	   42	  	  d0d180 CD31TdTomato20+mCD31TdTomatoα-SMA TdTomatoα-SMA 20+mmergemergemergemerge TdTomatod0d180 Figure 12. Tracking of tdTomato+ cells over time. Hic1-expressing cells, as labeled by tdTomato, are normally closely associated with vasculature. Labeling and tracking these cells in the pancreas for 180 days shows no change in that regard. Normally these cells are α-SMA negative; however, by d180, some of the tdTomato+ cells have begun expressing α-SMA. Left-most panels: 400X magnification images, blown up in the middle and right panels. Top two panels: CD31. Bottom two panels: α-SMA. 	   43	  mice were either UBC-CreERT2-null, or were given oil in place of tamoxifen.  Animals were sacrificed at various time-points post-TAM and processed for histology.  qPCR analysis showed 83 ± 6 % reduction in Hic1 mRNA expression in Hic1-deleted pancreas compared to controls.  In general, other examined tissues (lung, skeletal muscle) showed similar levels of deletion.  Over a one-year period, no appreciable differences in weight between the various genotypes was observed (Figure 13H).  Blood glucose was also tested regularly (two sets, n ≤ 3 per group).  Prior to tamoxifen administration, the mice had the same levels of blood glucose (~8 mmol/L).  Over 20 weeks after tamoxifen administration the groups showed mainly similar blood glucose levels within a healthy range (Figure 13G).  These results indicate that deletion of Hic1 does not appear to affect mouse health. Histological analysis of Hic1-deleted samples did show a prominent phenotype.  As early as two weeks after tamoxifen-induced Hic1 deletion, the pancreas (as well as other tissues, including skeletal muscle and heart) showed increased fat deposition compared to wild-type and heterozygous controls.  As the mice aged, the differences between the extent of fat deposition between control and deleted mice became more apparent (Figure 13A-D).  The adipocytes were first noted by hematoxylin and eosin staining, and confirmed by perilipin immunofluorescence.  Older control mice were found to have a few random adipocytes, but the older Hic1-deleted mice had numerous tracks of perilipin-positive cells distributed throughout the pancreas, but not within the islets. Hic1 deletion therefore causes increased aberrant adipogenesis in the pancreas.   	   44	  	  	  	  	   	  Hic1f/-Cre+ or -TAM~ 5 m8 wkHic1f/f;+/+UbcCreERT2TAM9 wk8 wk100 +m200 +mC DA B0.000.501.001.502.002.50Hic1+/+ Hic1+/- Hic1 -/-% of CD45- CD31- cellsEdU positive, Sca1 positiveFp < 0.05E% LacZ+ nucleiHic1f/-Hic1f/-Cre+Cre-p < 0.0050 2 4 6 8 10 12 14 G HHic1+/+Hic1-/-% Weight Months post Hic1 deletion 80% 100% 120% 140% 160% 180% 200% 220% 240% 260% 0 2 4 6 8 10 12 14 WT Hic1 KO 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0 5 10 15 20 25 30 Blood Gluclose (mmol/L) Weeks Post Hic1 deletion WT Hic1 KO 	   45	  	  Figure 13. Effect of Hic1 deletion in adult mouse pancreas. Induced conditional deletion of floxed Hic1 alleles results in ectopic fat deposition in various tissues and in increased numbers of nLacZ+ cells. Mice were given five doses of tamoxifen at 8 weeks of age, and samples collected after 9 weeks or 5 months. Top panel: after 9 weeks, the Hic1-/- mice (B) displayed an increased number of adipocytes (as stained by perilipin, red) compared to wild-type controls (A). Bottom panel: by 5 months after tamoxifen administration to induce Emc deletion, the pancreas of knock-out mice (D) had even more amounts of fat, compared to heterozygous controls (C) (Dapi, blue; perilipin, red. 100X magnification). (E) After Hic1 deletion, there is a ~40% increase in nLacZ+ nuclei in most tissues including the pancreas. (F) An increased proportion of CD31- Sca1+ cells (as a partial marker of these cells) are proliferating as monitored by EdU incorporation in flow cytometry. Female Hic1-deleted mice were monitored for blood glucose (G) and weight (H). 	   46	  Frequency of HIC1-expressing cells As loss of Hic1 has been shown, in vitro, to lead to cell cycle re-entry, we analyzed numbers of LacZ+ cells to determine whether the LacZ+ cells were increased in number after Hic1 deletion.  The Hic1nLacZ mice are functionally heterozygous, expressing a nuclear LacZ in place of one of the Hic1 alleles.  In Hic1-deleted cells, both alleles express LacZ.  In healthy heterozygous Hic1nLacZ mice, nLacZ+ cells comprised 7.8 ± 2.1% of all pancreatic cells, falling near the established range for PSCs.  In Hic1-deleted samples, however, the nLacZ+ cells increased to 12.3 ± 1.8% of all pancreatic cells, an increase of over 50% (p value < 0.0001, n > 9) (Figure 13E).  Similar increases were seen in all tissues examined (data not shown).  To further examine the impact of Hic1 deletion on cell proliferation, an EdU incorporation experiment was performed on Hic1-deleted animals.  Sca-1 staining was used as a proxy marker to detect the cell population of interest, with CD45+ and CD31+ cells gated out.  There was a trend of increasing percentage of EdU+ Sca1+ cells (of CD45- CD31- cells) in the Hic1-deleted samples compared to WT (p > 0.05, n = 3) (Figure 13F), which indicates that non-endothelial Sca1+ cells increase proliferation after Hic1-deletion.  HIC1-CELL ACTIVATION HIC1-cells are activated upon caerulein-induced pancreatic damage It is well documented that PSCs are activated in response to a number of stimuli, including tumor growth, injury and this in part involves inflammation or necrosis95,96,124,140.  To examine the behavior of Hic1+ cells under these conditions, we challenged the Hic1creERT2/+; tdTomatoflox/flox mice with caerulein-induced pancreatic 	   47	  damage.  The early stages after caerulein administration (9-48 h) consist of tissue destruction that is accompanied by pronounced inflammatory infiltrate109,111.  During this time no difference was seen in tdTomato+ cell location, morphology, nor numbers (Figure 14).  The tdTomato+ cells appeared to become activated at d2 after damage. Maximal focal damage was observed on d3-4 and this was associated with numerous tdTomato+ cells distributed throughout the overtly damaged regions (Figure 14). The tdTomato+ cells in the damaged areas exhibited characteristics akin to those of activated PSCs.  Namely, they increased in number, became α-SMA-positive, and produced Collagen Type 1, as indicated by a Col1-GFP reporter (Figures 15D, G & 16E).  In undamaged pancreas, tdTomato+ cells did not express the collagen-GFP reporter (Figure 9G).  The activated cells remained positive for NG2, PDGFRα-EGFP and desmin, and some tdTomato+ cells gained GFAP expression (Figure 16).   In the healthy pancreas, the tdTomato label comprised 1.4 ± 0.6% of total area, compared to 12 ± 2% in damaged areas (n ≥ 2 mice, minimum two 200X images/mouse, p < 0.0005).  Counts on cryosections showed that in normal pancreas (tdTomato/DAPI), the tdTomato+ cells accounted for 4 ± 2% of all cells, compared to 13 ± 3% of cells in damaged areas (see Table 7).  By FCM quantification, the tdTomato+ cells increased almost 5-fold compared to healthy pancreas (p < 0.05, n = 3).   	   48	  	  	   Figure 14. Timeline of caerulein-induced pancreatic damage and regeneration. Pancreata from tdTomato-labeled mice were collected at different time-points after a single day of caerulein injections. Serial sections were stained with H&E or with DAPI. At d2 after damage, the pancreas displays signs of inflammation and edema, but the tdTomato+ cells do not yet show signs of activation. By d4, the pancreas contains foci with massive acute fibrosis, in which the tdTomato+ cells proliferate and amass. By d6 after caerulein injections, regeneration is almost complete, with only small foci of fibrosis left. By d9, regeneration is complete, and the tdTomato+ cells are in similar frequency and location as the saline control pancreas. 100X images. Cer d2Cer d4Cer d6Cer d9100+mSaline	   49	  	  Figure 15.  tdTomato-labeled cells after caerulein-induced pancreatic damage. Hic1+ cells become activated after caerulein-induced damage, proliferating and becoming αSMA positive. (B) Control pancreas. (A, C, D) Damaged pancreas at d4 after cerulein-induced damage. tdTomato+ cells are still closely associated with vasculature (A, C, green CD31) and have become αSMA positive (D, green). The increase of tdTomato+ cells is restricted to damaged areas (E, F parallel sections. (E), hematoxylin & eosin staining; (F), DAPI). Such damaged areas are positive for Collagen I (G) and Collagen IV (not shown). I = islets. Yellow outlined area = damaged area. 	   50	  	  TdTomato LamininMergeTdTomatoMerge DesminTdTomato NG2MergeBACMerge TdTomato Col1a-EGFPMergeTdTomato PDGFRα-EGFP20+m 20+mTdTomatoMerge GFAPDEF	   51	  	  Figure 16. Characterization of tdTomato+ cells in caerulein-damaged pancreas. tdTomato+ pancreata were collected 3 days after one day of caerulein injections and cryosectioned. Damaged areas were immunostained for various markers (A-D), shown in green in the right-most panels. Transgenic mice harboring a knock-in GFP reporter for Collagen 1 (E) or PDGFRα (F) were also crossed to Hic1CreERT2; RosatdTomatoflox/+ mice and challenged with caerulein as well. 	   52	  Table 7. Frequency of tdTomato+ Cells in Healthy and Damaged Pancreas  Statistic Control pancreas Damaged Pancreas Fold Increase Histology Area Fraction (%) 1.4 ± 0.6 12 ± 2*** 8.6 tdTomato+ cells (% of DAPI cells) 4 ± 2 13 ± 3 ** 3.3 Flow Cytometry tdTomato+ podocalyxin- cells (% of total) 1.2 ± 1 5.2 ± 1* 4.4 ±SE, * p < 0.05, ** p < 0.001, *** p < 0.0001,   PANCREATITIS MODEL DEVELOPMENT The caerulein-induced pancreatitis model is quite well established, however, it is mainly used to study the induction of inflammation, with most endpoints being one or two days after damage.  Using this established model in conjunction with the Hic1CreERT2; tdTomato mice, activation was observed to only occur in small patches by day two, with maximal numbers of tdTomato+ cells at day 3 and day 4 (Figure 14, 15).  Though the initial cell destruction and inflammation was quite homogeneous and widespread in the early stages, by days 2 - 4 there were only a few focal regions of activated cells.  Therefore, in order to produce enough damage for functional analysis, and to create a more homogeneous sample of activated cells for future FACS sorting and RNA-seq, the model needed to be altered to produce more widespread and consistent damage. First, an experiment was performed to compare the effect of fasting the mice prior to caerulein-injections, vs not fasting.  Most published caerulein protocols mentioned fasting the mice overnight, reasoning that lack of food would increase the amount of digestive enzymes available in the pancreas.  However, discussions with Dr. Pin from the University of Western Ontario, who studies the early stages of pancreatitis using this model, indicated better results are obtained without fasting the mice.  A comparison was 	   53	  performed on C57Bl/6 mice that were IP injected with caerulein 8 times over 7 hours, and half of the mice were fasted the night before (Figure 17A).  The entire pancreas was isolated and the damage analyzed by H&E staining representative sections throughout the entire pancreas, as described in the Methods section.  The damaged area containing activated tdTomato cells can easily be identified by H&E staining (refer to Figure 14  & Figure 15F & G).  The fasted mice had a damaged area comprising 1.5  ± 0.9% of the pancreas.  The non-fasted mice seemed to have more damage, at 4.6 ± 1.3% (p = ns). To try to maximize the amount of damaged area further, non-fasted C57Bl/6 mice were IP injected with caerulein 8 times over 7 hours, either for one day, or for two days, and analyzed as before.  The pancreas from the one-day injected mice (20 wk) contained damaged region comprising of 4.2% ± 3% of the total pancreas area, whereas the two-day injected mice had 6 ± 2% damaged area (p = ns) (Figure 17B).  To increase the extent of damage, additional injections were performed.  In this regard, a third experiment comparing two-day injected mice with mice that were injected three times in one week (alternating days) was performed.  The latter method was inspired by the model for inducing chronic pancreatitis (which requires a minimum of 2 weeks of 3x weekly injections), leading to pancreatic cancer over many months111,119.  Samples were collected 3 days after the last injection, and the damage analyzed as before.  The two-day injected mice (15 wk) contained 2.3  ± 0.4% damage, the 3x injected mice had 11.7  ± 4.7% damage, which was substantially more (p = ns) (Figure 17C).  Therefore, for experiments for which the goal was to analyze the effect of Hic1 deletion on regeneration, this protocol of 3x weekly injections was used.  	   54	  	  	  Figure 17. Optimization of the pancreas damage model. C57Bl/6 mice were dosed with caerulein at various times according to the graphics above, and the damaged area calculated from the collected pancreata. Two-tailed t-tests were done on the results. n ≥ 2. (A) Fasted vs not fasted mice with 1 day of 8 doses of caerulein injections. (B) Non-fasted mice comparing 1 and 2 days of caerulein injections. (C) Non-fasted mice comparing 2 subsequent days of 8 injections with 3 alternating days of 6 caerulein injections. D -1 D0 D1 D2CollectD -2IP CerX6D -3D -4 D3IP CerX6IP CerX6D -1 D0 D1 D2CollectD3IP CerX8IP CerX8D -1 D0 D1 D2CollectD3IP CerX8IP CerX8D0 D1 D2CollectD3IP CerX8D0 D1 D2CollectIP CerX8Fast18hD0 D1 D2CollectIP CerX820wkC57Bl/620wkC57Bl/615wkC57Bl/6p = ns2 days3x weekly051015201 day2 days02468FastedNon-Fasted0246p = nsp = nsABC% Damage % Damage % Damage 	   55	  TDTOMATO+ CELLS PROLIFERATE AFTER DAMAGE To produce maximal damage, with maximum opportunity for labeling proliferating cells with EdU, caerulein was injected for 6 h on three alternating days (d-4, d-2, d0), as described in the third experiment above.  EdU was injected for four days total (d-1, d0, d1, d2).  Results are seen in Figure 18.  The healthy control mice had 3.5 ± 0.9% of tdTomato+ cells incorporating EdU, whereas in damaged pancreas there were 7.6 ± 1.5% of tdTomato+ cells cycling (n ≥ 2, p < 0.05). This indicates that tdTomato+ cells in the pancreas do display some proliferation even in healthy tissue, their EdU incorporation increases after pancreatic damage.  PANCREAS REGENERATION OF HIC1-DELETED MICE To study the role of Hic1 in regeneration mice were challenged with cerulein, and the magnitude of damage/impaired regeneration measured by % fibrotic area and total pancreas volume, as covered in the Methods & Materials section.  Mice were collected at d3 after the 3x weekly caerulein injections, as described above.  Weight loss was comparable between the groups (Figure 19A), except two mice from the Hic1-deleted group were euthanized at d-2 (during the second day of caerulein injections) as they reached humane endpoint.  No difference in total pancreas volume was seen (Figure 19B).  However, by histological analysis the heterozygous (UBCCreERT2-/-; Hic1f/-)and Hic1-deleted mice showed less % fibrotic area than the control group, with the difference between control and Hic1-deleted being significant (Figure 19C; 9.2 ± 0.3% vs 1.9 ± 1.7%, p < 0.05, n ≥ 3).  The control group consisted of mice that were genetically 	   56	  	  Figure 18. EdU incorporation into tdTomato+ cells in healthy and caerulein-damaged pancreata. HealthyDamaged0246810% EdU+ Tom+ of all Tom+ cells p < 0.05Figure 18. EdU incorporation into tdTomato+ cells in healthy and caerulein-damaged pancreata. 	   57	  	  	  	  	  Figure	  19.	  Hic1-­‐deleted	  mice	  in	  pancreas	  regeneration.	  (A)	  Weight	  loss	  of	  caerulein-­‐challenged	  mice.	  Mice	  were	  administered	  6	  hourly	  doses	  of	  caerulein	  on	  days	  -­‐4,	  -­‐2	  and	  0,	  and	  were	  collected	  at	  d3.	  Two	  Hic1	  KO	  mice	  reached	  humane	  endpoint	  on	  d-­‐2	  and	  were	  euthanized.	  (B)	  total	  pancreas	  volume.	  (C)	  %	  damaged	  area.	  Body Weight Pancreas Volume (mm3 )ABCControlHet KO0246810% Damage p < 0.05Experimental Day60% 70% 80% 90% 100% 110% -4 -3 -2 -1 0 1 2 3 4 Saline KO Control cer Het cer KO cer KO SalineControl cerHet cerKO cer020406080100	   58	  Hic1CreERT2/+; RosaDTAf/+, however they were not given TAM and were therefore phenotypically and physiologically WT C57Bl/6.   	   59	  CHAPTER 4: DISCUSSION PSCs are known to play important roles in disease progression and regeneration64,95,97.  However, there is no single specific marker to identify these important cells.  It is universally agreed that PSCs become activated and are effector cells in diseases such as chronic pancreatitis and fibrosis and are involved in pancreatic cancer, comprising a great deal of the tumour stroma, being the prime source of ECM production and supporting tumour growth, metastasis and fibrosis63,64,141,142.  Herein I have shown that Hic1 expression faithfully marks potential PSCs and that Hic1+ cells become activated in response to pancreatic damage, proliferate, produce collagen and play a role in tissue regeneration.  HIC1 IS A PUTATIVE MARKER OF QUIESCENT PSCS Currently researchers can only identify PSCs through marker analysis, and/or by an isolation method that relies on their lighter gravity/density due to their perinuclear lipid droplets92.  However, once these cells are activated their retinyl-ester stores are depleted, changing their density, and such isolation is much more difficult95,97.  Here we have shown that Hic1 expression, as marked by Hic1LacZ/+, or by Hic1+/CreERT2; tdTomatoflox/flox, has good overlap with known PSC markers such as vimentin, NG2, nestin and desmin, as well as the general MSC markers PDGFRα and Sca-1 (Figures 3, 7, 9)35,95.  As well, tdTomato+ cells contain perinuclear lipid droplets, which are known to be present in PSCs (Figure 8).  That there is incomplete overlap with markers is expected.  Firstly, PSC identification requires a mix of these markers, and no published report has quantified the variable expression of these markers. Secondly, it may be that 	   60	  Hic1-expression marks only a subset of PSCs or is slightly more upstream.  That can be examined with future studies, including clonal analysis and differentiation potential, and single-cell characterization to determine whether each Hic1-expressing cell bears the hallmarks of PSCs.  As well, we have found in other tissues (such as muscle) that Hic1-expressing cells are able to differentiate into fat and bone lineages. Since we noted that Hic1-deleted pancreas contains increased numbers of adipocytes, which potentially come from the Hic1-deleted cells, it is possible that Hic1-expressing cells are upstream of PSCs.  No published study has described the lineage potential of PSCs other than noting they differentiate into αSMA+ myofibroblasts upon activation95,97,143,144.  These results are important because most reports examining PSCs seem to not consider them MSCs, and this work shows that Hic1-expressing cells display markers of both cell types. One marker which could not be compared to our cells of interest was GFAP (Figure 11).  Many papers refer to GFAP as one of the most important markers of PSCs and HSCs, though it is not sufficient to unequivocally identify PSCs63,64,92,98.  Immunostaining on cryoslides failed to detect GFAP in pancreas (not in tdTomato+ cells, nor in other cells), though staining was successful in the positive controls of brain and liver.  However, there is no compelling evidence showing GFAP expression in murine tissue sections: most reports only assess cultured activated PSCs64,92,95 or examine other species92,97.  A group which made a GFAP-LacZ98 transgenic mouse has presented only a single X-gal stained image with limited resolution, and though the location of the GFAP-LacZ+ cells was correct for PSCs, few cells were LacZ positive.  Another group reported that GFAP is in actuality a poor marker for HSCs66.  Therefore, the fact that we cannot conclusively state whether Hic1+-cells express GFAP is not very meaningful.  	   61	  One point that must be made is that many published studies on ‘activated’ PSCs are based on cells in vitro, the idea being that once these cells are in culture and proliferating, they are ‘activated’, which is obviously different from the in vivo situation64,95. Groups also employ different methods for isolating PSCs. Though the density gradient method may be the most rigorous92,143, resulting in a high purity of PSCs, many groups use outgrowth/explant methods145,146, from which they may get PFBs and macrophages as well as PSCs.  Furthermore, most groups use media (i.e. DMEM, Ham’s F-12) that does not select for mesenchymal cells96,143,145,147 (as opposed to StemCell Technologies’ Mesencult + Mesenpure).  In addition, few groups use a low-oxygen incubator; therefore, their cells are cultured in, in effect, hyperoxic conditions148, though it is well accepted that mesenchymal cells exhibit greater growth in low oxygen conditions26,79,149.  All these points make it difficult to rely on published in vitro data to understand PSCs.  Therefore, this work is not only the first to characterize pancreatic Hic1-expressing cells, validating many previously published reports on PSC markers, but is potentially more accurate as the vast majority of this work was based on in vivo samples.  THE EFFECT OF HIC1 DELETION qPCR analysis on pancreata from Hic1-deleted mice showed 83 ± 6 % reduction of Hic1 mRNA, suggesting that Hic1 deletion was efficient.  Histological examination identified ectopic adipocytes in several tissues but most prominently and consistently in the pancreas (Figure 13).  The fat depots were found surrounding the acini rather than the ducts or islets.  Such a finding may provide an important link, as groups have reported 	   62	  fatty degeneration is a risk factor for PDAC150, and it is known that loss of Hic1 correlates to an increased incidence of tumours3,4,12,151.  The adipocytes likely arose from the Hic1-deleted cells, as we have seen in other tissues that the tdTomato-labeled cells have the ability to differentiate into adipocytes.  That has also been seen by single-cell FACS-sorted lineage potential experiments on Hic1+ cells from other tissues (data not shown).  Another possibility is that Hic1 deletion causes a change in paracrine signaling, inducing another cell to differentiate into an adipocyte.  For example, a previous report by Bonal et al found that inactivation of c-myc resulted in the transdifferentiation of acinar cells into adipocytes.  The best way to address this would be to generate a mouse model where Hic1-expressing cells can be labeled, and then Hic1 deleted, and track only the labeled cells.  However, it is most likely that the Hic1-deleted cells differentiated into adipocytes, perhaps in response to the increased cycling of these cells due to Hic1 deletion (Figure 13E & F).  This phenotype has important links to aging: as organisms age regulatory networks are negatively affected, and aberrant fibrosis and adipogenesis increase, interfering with organ function152,153. The Hic1-deleted mice were monitored for up to a year and collected at various time-points.  Weight data and blood glucose analysis found no differences between groups.  That there was no effect seen with regards to blood glucose can easily be explained by two factors.  Firstly, islets make up only ~1% of the pancreas106, so most of the ectopic adipocytes would displace acinar cells, not islets.  Secondly, few Hic1+ cells are seen in the islets (Figure 15E & F), so islets are less likely to be affected.  More subtle differences may be teased out with a glucose tolerance test, but in summary, loss of Hic1 expression did not affect murine body weight nor glucose regulation. 	   63	   HIC1+ CELLS IN HOMEOSTASIS AND IN PANCREATIC DAMAGE tdTomato-labeled cells were tracked over time to determine whether these cells are mainly quiescent or are more dynamic.  We had expected there to be little noticeable difference between time-points, but between the d0 and d90 samples there was a greater than two-fold increase in the number of tdTomato+ cells.  However, between d90 and d180 there was no difference (Figure 12, Table 6).  The pancreas has a constant low level of cell turnover due to incessant small injuries, which could explain these results.  The cells may exhaust themselves after a time, however, and are replaced by newer cells from a more upstream stem cell, explaining why there is no difference between d90 and d180.  Caerulein-induced pancreatic damage first causes an inflammatory response followed by acute fibrosis if the damage is severe enough (Figure 15).  During the inflammatory phase (d0-d2) tdTomato+ cells did not appear to be activated (Figure 14). These cells may be slow to respond so as to prevent over-reaction.  Similarly, in a skeletal muscle damage model, activation of tissue-resident MSCs following damage occurred ~ 36-48 h post-injury35.  Three days post damage the tdTomato+ cells are found distributed throughout the damaged regions, where they can be observed circumscribing the acini.  Those tdTomato+ cells also became positive for Collagen-GFP production (d2-d6), suggesting they produce extracellular matrix (Figure 16).  The tdTomato+ cells greatly increased in numbers, however, only directly in the fibrotic areas (Figure 15, Table 7).  The morphology of the tdTomato+ cells also changed, with the cells growing larger and less spindly, again, only in the affected areas.  As the tdTomato+ cells increased EdU incorporation during caerulein-induced damage, the cells must have 	   64	  entered the cell cycle (Figure 18).  Along with the co-expression of α-SMA we can conclude that Hic1-expressing cells are activated as a result of the damage. The exact size of the increase in tdTomato+ labeled cells is difficult to quantify, due to several factors.  One reason is that the tdTomato labeling is not 100% efficient. For example, with the Hic1citrine/+ mice, by flow cytometry the citrine reporter marks 3.8 ± 0.3% of all cells. With the tdTomato+ label the marked cells only mark 1.2 ± 1% of all cells.  Between mice there is also variation in the response to caerulein and the magnitude of the damage, resulting in high variation for the numbers of tdTomato+ cells. Flow cytometry experiments measure the entire pancreas, and  as increases are concentrated in the damaged areas, the overall effect may be diminished.  As well, the flow cytometry experiments show all cells of the pancreas (Table 7).  In damaged pancreas, CD45+ cells can make up ½ of the total cells (compared to ~10-20% of healthy pancreas) (data not shown).  Lastly, it appears that even in a resting state, tdTomato+ cells still display some proliferation (3.5 ± 0.9%), which may be due to the fact that the pancreas is a tissue with high turnover, but it cannot be ruled out to be due to the tamoxifen administration.  An experiment comparing the effect of tamoxifen and oil on pancreas cell proliferation should be carried out, to find the accurate level of proliferation in healthy pancreas. The histological analysis of tdTomato+ cells in damaged pancreas may be more applicable (Table 7).  However, as in the damaged areas there is also an increase in non-tdTomato+ cells (i.e. inflammatory cells, immune cells, PFBs), the increase in tdTomato+ cells is diluted out.  That is why the measurements of tdTomato+ area showed the greatest increase, at 8.6-fold.  However, this measurement may overestimate the amount of the increase, as the activated tdTomato+ cells are larger in size and less spindly than the 	   65	  quiescent cells.  Conversely, this method may also underestimate the increase, as the cells may overlap with each other, diminishing the overall area measured. Using both the FCM and the histology data together provide the clearest picture, and shows that tdTomato+ cells are increased after severe pancreatic damage.  PROLIFERATION OF TDTOMATO-LABELED CELLS IN RESPONSE TO PANCREATIC DAMAGE The tdTomato-labeled cells showed increased EdU incorporation in response to caerulein-induced damage (Figure 18).  The amount of proliferation, at a 2.6-fold increase from the resting state, is likely underestimated.  One reason is again the tdTomato labeling efficiency, as discussed above.  Probably more of the EdU+ cells would also be tdTomato-positive if the labeling was more efficient.  Another reason is the experimental outline.  As we had seen with the single-day of caerulein injections, the cells did not increase in number until d2, with maximal damage and numbers of tdTomato+ cells at d3 and d4 (Figure 14).  With the 3x weekly model EdU was only injected starting d-1 (after the second day of damage); it is likely that some of the cells entered the cell cycle earlier, and any proliferation between d-4 and d-2 would have been missed.  Thirdly, there may be differing amounts of damage between mice, and again, the flow cytometry only shows an average of the entire pancreas (see discussion above).  The addition of other markers including CD45 and α-SMA may clarify these points.  Of course, these data do not exclude the possibility that tdTomato+ cells proliferate elsewhere and then migrate to the pancreas; however, it is likely that the Hic1-expressing cells in the pancreas, nearest the damaged areas, are the proliferating cells. These various 	   66	  data indicate that upon caerulein-induced damage tdTomato+ cells cycle at an increased rate, and thus at least some of the increase in tdTomato+ numbers is due to proliferation.   REGENERATION Effect of Hic1-deletion on Pancreatic Regeneration Regarding the effect of Hic1-deletion on pancreas regeneration after caerulein-induced damage, no difference was noted in total pancreas volume at d3 (Figure 19). However, the Hic1-deleted mice displayed significantly less damaged area than the controls (Figure 19C; 1.9 ± 1.7% vs 9.2 ± 0.3%, p < 0.05).  The fact that the pancreas volume was no different than in the other caerulein treated groups suggests lack of Hic1 may not affect regeneration of lost tissue.  However, the data showing minimal % damaged area in the Hic1-deleted mice suggests that loss of Hic1 expression either impedes fibrosis in some way or allows for faster clearing of the collagen-producing cells.  As Hic1 inhibits cell proliferation1,14,154, we hypothesize that this cell population downregulates Hic1 in order to activate; therefore, deleting Hic1 may ‘prime’ these cells to respond more quickly.  The tdTomato+ labeled model shows that after regeneration is complete the activated cells disappear (Figure 16); therefore, it is also possible that the Hic1-deleted, ‘primed’ cells were cleared by normal mechanisms, before they could contribute to the fibrosis.   An important caveat is the lack of proper TAM wild-type control.  Though the control caerulein group is functionally wild-type, genetically they were not (though Figure 10 proves Hic1CreERT2 is not activated without TAM).  The lack of TAM administration to this group means we cannot deduce the effect of TAM on the pancreas.  	   67	  Though the experiment was performed several days after TAM administration had finished, effects on cell populations may be longer-lasting.  For example, Sheheta et al assessed the effect on TAM on mammary gland in puberty and adult mice155. Within a few days of TAM administration (0.2 mg to 5 mg) there were changes in various cell populations, though the stromal compartment was only affected at the highest dose and changes were normalized after 8 weeks.  A repeat of this experiment with proper controls is necessary. Other effects may be noticed later in the regenerative process: since d3 is the peak of the damage perhaps d5-8 may be more pertinent.  Analysis of Hic1-deleted samples at various time-points will identify the effect of Hic1-deletion on damage severity and spread (early time-points), as well as on regeneration (later time-points). This work does not directly address what the Hic1+ cells do in the damaged areas.  It is possible that they act as a physical barrier, isolating the damage from spreading further, as well as supporting regeneration with the secretion of trophic factors. It is known that PSCs can ‘go rogue’ and contribute overly much to fibrosis and impair tissue regeneration, such as what happens in PDAC, for example64,100,156, as do other mesenchymal cells in fibrotic diseases59,64,82–84.  An important experiment will be to test the effect of Hic1+-cell ablation on regeneration.  A paper that examined the role of MSCs in pancreas development, using a pancreatic mesenchymal-specific cre, Nkx3.2, found that ablation of pancreatic mesenchyme led to reduced pancreas size, development and β-cell and acinar cell proliferation157.  Another group saw that ablation of injury-induced, collagen-producing ADAM12+ cells (which also express PDGFRα and Sca1) in an injury model resulted in decreased fibrosis, as defined by fewer aSMA+ cells, less 	   68	  interstitial collagen and decreased leucocyte infiltration158.  As the tdTomato+ cells, which represent similar cell populations as in those papers, are present and activated in damaged areas, it is likely that their loss would have a profound affect.    	   69	  CHAPTER 5: CONCLUSION SUMMARY In conclusion, this work has shown reliable evidence that Hic1+cells in the pancreas represent pancreatic stellate cells (PSCs) (Figure 20).  It is clear that Hic1 expression is a robust marker for quiescent PSCs, which prior to this work did not have a singular marker.  Upon Hic1-deletion, the pancreas presents with increased amounts of perilipin+ adipocytes.  We have shown that after light pancreatic damage (inflammation only) these cells do not appear to be activated; however, in more extensive damage (acute fibrosis, necrosis) these cells become activated, entering the cell cycle, changing their morphology, being distributed throughout the damaged areas.  Once regeneration is complete a subset of the cells is hypothesized to become quiescent again and return to their original frequency and location.  In the Hic1-deleted mice, the damage comprised as significantly smaller portion of the pancreas compared to controls, though the total pancreas volume was unchanged.  The fate of the expanded tdTomato+ cells following activation is being investigated.    DISCUSSION OF GOALS/OBJECTIVES Two of the three goals were fully met. 1) Characterize Hic1+ cells in the pancreas.  	   70	  	  	  Figure 20. Proposed action of Hic1+ cells in pancreas. “Quiescent” Hic1+ cells co-express markers of both MSCs and PSCs. Upon “activation”, potentially due to loss of Hic1 expression after injury or mutation, these cells enter the cell cycle, and can putatively differentiate into aSMA+ myofibroblasts or Plin1+ adipocytes. Upon the completion of regeneration, most cells probably undergo apoptosis or become quiescent again. Grey = hypothesized. PancreaticStellateCellHic1+PDGFRASca1activationdifferentiationdeathmesenchymalcell++senescence(quiescence)Hic1Hic1+Desmin+NG2+Nestin+PDGFRASca1++Desmin+NG2+Nestin+_SMA+_SMA+Plin1Col1+	   71	  Hic1-expressing cells were characterized using the Hic1CreERT2; RosatdTomato transgenic mouse line.  tdTomato+ cells did not appear to be epithelial (CD326, flow cytometry), endothelial (CD31 & Podocalyxin, immunofluorescence and flow cytometry) endocrine (histology – location, morphology) nor neuroendocrine in nature.  tdTomato+ cells did co-express markers relating to PSCs, namely desmin, NG2, nestin, as well as having lipid droplets when quiescent.  2) Determine what happens to Hic1+ cells during pancreatic damage and regeneration.  After pancreatic damage and during regeneration, Hic1+ cells became activated.  The cells increased in number, surrounded and infiltrated the damaged areas.  They gained expression of α-SMA, synthesized collagen and entered the cell cycle.  Upon completion of regeneration, tdTomato+ cells returned to their original frequency and location.  They did not appear to contribute directly to regeneration of the exocrine and endocrine compartments of the pancreas, as these cells were not tdTomato+.  3) Determine the effect of loss of Hic1 expression in pancreas homeostasis and regeneration.  It was found that loss of Hic1 expression led to increased numbers of nLacZ+ cells, increased EdU incorporation into CD45- CD31- Sca1+ cells, as well as resulting in increased amounts of perilipin+ adipocytes in the pancreas.  The Hic1-deleted samples 	   72	  had significantly less damage but the entire pancreas had not recovered differently in terms of volume, suggesting fibrosis was impeded but total organ re-growth was not affected by Hic1 deletion.  However, this experiment should be repeated to rule out the influence of TAM administration.  Also, a more detailed ‘timeline’ should be generated, to provide more insight into what time-points should be examined.  As well, no group has done the model with this treatment set before, looking 3-5 days after damage.  Most groups look immediately after caerulein-treatment, in the first 9-48 h, or researchers look for chronic pancreatitis or cancer, weeks later.  SIGNIFICANCE  This work contains several important firsts.  This is the first work that examines the role of HIC1 and Hic1-expressing cells in homeostasis and tissue regeneration; previous work has focused on cancer with a few reports on the role of HIC1 in development.  This work is also the first to identify a single marker for quiescent PSCs that can be used to track/modify these cells throughout damage/regeneration. This transgenic model can be used to allow for more pure isolation of PSCs, or better study of molecular mechanisms and pathways involved in PSC regulation.  Thirdly, this is the first in vivo work with an in-depth examination and characterization of Hic1-expressing cells, and the first to hint at a functional role for Hic1 outside of development and cancer.  The phenotype of increased adipogenesis in the Hic1-deleted mice suggests that loss of Hic1, whether by mutation, hypermethylation or mis-regulation, may have a role in aging.  Finally, this thesis supports previous reports suggesting PSCs are an important clinical target.  	   73	   STRENGTHS & LIMITATIONS The strengths of this study lie mainly in the unique mouse models that were employed, with which the Hic1-expressing cells could be securely identified, characterized and tracked.  The animal models also gave more trustworthy insight than most previous PSC research, which often rely on in vitro cultures, whose purity is not confirmed and whose similarity to in vivo PSCs is uncertain. An obvious limitation of this work is that it is underpowered.  A greater ‘n’ would have led to a higher signal-to-noise ratio, perhaps allowing significance to be reached. The work would also benefit from more appropriate control groups.  As well, the current method of analyzing damage and regeneration is too high-level.  Including other readouts, such as a serum amylase kit, may give a fuller picture.  POTENTIAL APPLICATIONS  One important application of this research is the validation of the Hic1CreERT2/+ mouse line to label and follow PSCs.  The Hic1CreERT2/+ mouse line so far represents the most accurate way of labeling and tracking PSCs; other mouse lines rely on collagen or αSMA expression to label these cells, which are much less specific.  As PSCs are important not only in pancreatic fibrosis, but also in cancers, the role of PSCs in the formation and metastasis of pancreatic cancer can be studied, along with a host of other diseases.  Along with the shortened pancreatic fibrosis model, which was optimized here, therapeutics can be tested and the effect on PSC activation directly identified, either by histology or by FACS analyses of the labeled cells.  As well, this knowledge can be 	   74	  extended to assist in further understanding the behavior of the mesenchymal component in aging, as well as fibrotic diseases affecting other tissues.  FUTURE DIRECTIONS The results from this work can instigate and support many other projects.  Some future directions include: 1) Further dissecting the role of HIC1 and Hic1+cells in tissue regeneration.  Repeating and refining the regeneration experiments on Hic1-deleted mice will prove useful.  Use of iDTR mice to conditionally ablate Hic1-expressing cells will provide insight into the role of these cells in tissue regeneration. Further defining and optimizing the model (i.e. making a timeline of damage and regeneration to better identify the stage to examine where the effect of Hic1-deletion or Hic1+-cell ablation will be maximal), as well as identifying other ways to quantify the damage (i.e. use a pancreatic amylase kit to test serum samples to test for damage at several time-points, while mice are still alive) will be beneficial. 2) Examining the potential of these cells by in vitro methods, for example performing single-cell sorting to quantify how many of these cells are able to enter the cell cycle, and whether those cells have multi-lineage potential and can contribute to various mesenchymal lineages. 3) Starting to analyze the downstream targets underlying Hic1 function by doing RNA-seq on WT and ko cells (using the Hic1citrine/+ and Hic1citrine/flox with FACS), and on tdTomato+ cells to compare healthy/quiescent and activated states. 	   75	  4) Lastly, it will be interesting to look at what happens to the activated cells once regeneration is complete.  Most likely, the cells go through programmed cell death, though some may become quiescent again.  Understanding that pathway may lead to therapeutic targets for clinical applications, though it remains to be discovered whether enhancing or inhibiting apoptosis of activated cells would help or hinder regeneration. 	    	   76	  BIBLIOGRAPHY 1. Wales, M. M. et al. p53 activates expression of HIC-1, a new candidate tumour suppressor gene on 17p13.3. Nature 1, 570–577 (1995). 2. Fujii, H. et al. Methylation of the HIC-1 candidate tumor suppressor gene in human breast cancer. Oncogene 16, 2159–64 (1998). 3. Fleuriel, C. et al. HIC1 (Hypermethylated in Cancer 1) epigenetic silencing in tumors. Int. J. Biochem. Cell Biol. 41, 26–33 (2009). 4. Chen, W. Y. et al. Heterozygous disruption of Hic1 predisposes mice to a gender-dependent spectrum of malignant tumors. Nat. Genet. 33, 197–202 (2003). 5. Chen, W. Y. et al. Tumor suppressor HIC1 directly regulates SIRT1 to modulate p53-dependent DNA-damage responses. Cell 123, 437–48 (2005). 6. Carter, M. G. et al. Mice deficient in the candidate tumor suppressor gene Hic1 exhibit developmental defects of structures affected in the Miller-Dieker syndrome. Hum. Mol. Genet. 9, 413–9 (2000). 7. Pinte, S. et al. The tumor suppressor gene HIC1 (hypermethylated in cancer 1) is a sequence-specific transcriptional repressor: definition of its consensus binding sequence and analysis of its DNA binding and repressive properties. J. Biol. Chem. 279, 38313–24 (2004). 8. Mondal, A. M. et al. Identification and functional characterization of a novel unspliced transcript variant of HIC-1 in human cancer cells exposed to adverse growth conditions. Cancer Res. 66, 10466–77 (2006). 9. Lim, J.-H. Zinc finger and BTB domain-containing protein 3 is essential for the growth of cancer cells. BMB Rep. 47, 405–10 (2014). 10. Lee, S.-U. & Maeda, T. POK/ZBTB proteins: an emerging family of proteins that regulate lymphoid development and function. Immunol Rev 247, 107–119 (2013). 11. Rood, B. R. & Leprince, D. Deciphering HIC1 control pathways to reveal new avenues in cancer therapeutics. Expert Opin. Ther. Targets 17, 811–27 (2013). 12. Briggs, K. J. et al. Cooperation between the Hic1 and Ptch1 tumor suppressors in medulloblastoma. Genes Dev. 22, 770–85 (2008). 13. Van Rechem, C. et al. Scavenger chemokine (CXC motif) receptor 7 (CXCR7) is a direct target gene of HIC1 (hypermethylated in cancer 1). J. Biol. Chem. 284, 20927–35 (2009). 	   77	  14. Boulay, G. et al. Loss of Hypermethylated in Cancer 1 (HIC1) in breast cancer cells contributes to stress-induced migration and invasion through β-2 adrenergic receptor (ADRB2) misregulation. J. Biol. Chem. 287, 5379–89 (2012). 15. Foveau, B. et al. The receptor tyrosine kinase EphA2 is a direct target gene of hypermethylated in cancer 1 (HIC1). J. Biol. Chem. 287, 5366–78 (2012). 16. Mohammad, H. P. et al. Loss of a single Hic1 allele accelerates polyp formation in Apc(Δ716) mice. Oncogene 30, 2659–69 (2011). 17. Van Rechem, C. et al. Differential regulation of HIC1 target genes by CtBP and NuRD, via an acetylation/SUMOylation switch, in quiescent versus proliferating cells. Mol. Cell. Biol. 30, 4045–59 (2010). 18. Dehennaut, V., Loison, I., Boulay, G., Van Rechem, C. & Leprince, D. Identification of p21 (CIP1/WAF1) as a direct target gene of HIC1 (hypermethylated in cancer 1). Biochem. Biophys. Res. Commun. 430, 49–53 (2013). 19. Valenta, T., Lukas, J., Doubravska, L., Fafilek, B. & Korinek, V. HIC1 attenuates Wnt signaling by recruitment of TCF-4 and beta-catenin to the nuclear bodies. EMBO J. 25, 2326–37 (2006). 20. Zhang, J. et al. High β-catenin/Tcf-4 activity confers glioma progression via direct regulation of AKT2 gene expression. Neuro. Oncol. 13, 600–9 (2011). 21. Van Rechem, C., Boulay, G. & Leprince, D. HIC1 interacts with a specific subunit of SWI/SNF complexes, ARID1A/BAF250A. Biochem. Biophys. Res. Commun. 385, 586–90 (2009). 22. Deltour, S., Pinte, S., Guerardel, C., Wasylyk, B. & Leprince, D. The Human Candidate Tumor Suppressor Gene HIC1 Recruits CtBP through a Degenerate GLDLSKK Motif. 22, 4890–4901 (2002). 23. Pospichalova, V. et al. Generation of two modified mouse alleles of the Hic1 tumor suppressor gene. Genesis 49, 142–51 (2011). 24. Caplan, A. I. & Correa, D. The MSC: An Injury Drugstore. Cell Stem Cell 9, 11–15 (2011). 25. Bianco, P. “Mesenchymal” Stem Cells. Annu. Rev. Cell Dev. Biol. 1–28 (2014). doi:10.1146/annurev-cellbio-100913-013132 26. Dasari, V. R., Veeravalli, K. K. & Dinh, D. H. Mesenchymal stem cells in the treatment of spinal cord injuries: A review. World J. Stem Cells 6, 120–133 (2014). 	   78	  27. Chhabra, P. & Brayman, K. L. Stem Cell Therapy to Cure Type 1 Diabetes  : From Hype to Hope. Stem Cells Transl. Med. 2, 328–336 (2013). 28. Figliuzzi, M., Bonandrini, B., Silvani, S. & Remuzzi, A. Mesenchymal stem cells help pancreatic islet transplantation to control type 1 diabetes. World J. Stem Cells 6, 163–172 (2014). 29. Siniscalco, D., Bradstreet, J. J., Sych, N. & Antonucci, N. Mesenchymal stem cells in treating autism: Novel insights. World J. Stem Cells 6, 173–178 (2014). 30. Siniscalco, D. et al. Autism spectrum disorders: is mesenchymal stem cell personalized therapy the future? J. Biomed. Biotechnol. 2012, 480289 (2012). 31. Dimarino, A. M., Caplan, A. I. & Bonfield, T. L. Mesenchymal stem cells in tissue repair. Front. Immunol. 4, 201 (2013). 32. Murphy, M. B., Moncivais, K. & Caplan, A. I. Mesenchymal stem cells: environmentally responsive therapeutics for regenerative medicine. Exp. Mol. Med. 45, e54 (2013). 33. Singer, N. G. & Caplan, A. I. Mesenchymal stem cells: mechanisms of inflammation. Annu. Rev. Pathol. 6, 457–78 (2011). 34. Shen, H. Stricter standards sought to curb stem-cell confusion. Nature 499, 389 (2013). 35. Joe, A. W. B. et al. Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nat. Cell Biol. 12, 153–63 (2010). 36. Bianco, P. Don’t market stem-cell products ahead of proof. Nature 499, 255 (2013). 37. Eisenstein, M. Stem cells: Don’t believe the hype. Nature 484, S5–S5 (2012). 38. Trounson, A., Thakar, R. G., Lomax, G. & Gibbons, D. Clinical trials for stem cell therapies MSC Clinical Trials by Disease. BMC Med. 9, (2011). 39. Barry, F. & Murphy, M. Mesenchymal stem cells in joint disease and repair. Nat. Rev. Rheumatol. 9, 584–94 (2013). 40. Ikegaya, H. Tighten up Japan’s stem-cell practices. Nature 489, 33 (2012). 41. Cyranoski, D. Texas prepares to fight for stem cells. Nature 477, 377–8 (2011). 42. Sipp, D. & Turner, L. U . S . Regulation of Stem Cells as Medical Products. Science (80-. ). 338, 1296–1297 (2012). 	   79	  43. Hare, J. et al. A Randomized, Double-Blind, Placebo-Controlled, Dose-Escalation Study of Itravenous Adult Human Mesenchymal Stem Cells (Prochymal) After Acute Myocardial Infarction. J Am Coll Cardiol 54, 2277–2286 (2009). 44. Gao, L. R. et al. A critical challenge: dosage-related efficacy and acute complication intracoronary injection of autologous bone marrow mesenchymal stem cells in acute myocardial infarction. Int. J. Cardiol. 168, 3191–9 (2013). 45. Caplan, A. Why are MSCs therapeutic? New data: new insight. J. Pathol. 217, 318–24 (2009). 46. Wagner, J., Kean, T., Young, R., Dennis, J. E. & Caplan, A. I. Optimizing mesenchymal stem cell-based therapeutics. Curr. Opin. Biotechnol. 20, 531–6 (2009). 47. Caplan, A. I. Adult mesenchymal stem cells for tissue engineering versus regenerative medicine. J. Cell. Physiol. 213, 341–7 (2007). 48. Ding, Y. et al. Mesenchymal stem cells prevent the rejection of fully allogenic islet grafts by the immunosuppressive activity of matrix metalloproteinase-2 and -9. Diabetes 58, 1797–806 (2009). 49. Di Nicola, M. et al. Human bone marrow stromal cells suppress T-lymphocyte proliferation induced by cellular or nonspecific mitogenic stimuli. Blood 99, 3838–43 (2002). 50. Keating, A. Mesenchymal stromal cells: new directions. Cell Stem Cell 10, 709–16 (2012). 51. Cristancho, A. G. & Lazar, M. a. Forming functional fat: a growing understanding of adipocyte differentiation. Nat. Rev. Mol. Cell Biol. 12, 722–34 (2011). 52. Kota, D. J. et al. Differential MSC activation leads to distinct mononuclear leukocyte binding mechanisms. Sci. Rep. 4, 4565 (2014). 53. Lepperdinger, G. Inflammation and mesenchymal stem cell aging. Curr. Opin. Immunol. 23, 518–24 (2011). 54. Caplan, A. I. & Dennis, J. E. Mesenchymal stem cells as trophic mediators. J. Cell. Biochem. 98, 1076–84 (2006). 55. Muñoz-Espín, D. & Serrano, M. Cellular senescence: from physiology to pathology. Nat. Rev. Mol. Cell Biol. 15, 482–96 (2014). 	   80	  56. Wang, Y.-K. & Chen, C. S. Cell Adhesion and Mechanical Stimulation in the Regulation of Mesenchymal Stem Cell Differentiation. J Cell Mol Med 17, 823–832 (2013). 57. Luu, Y. K., Pessin, J. E., Judex, S., Rubin, J. & Rubin, C. T. Mechanical Signals As a Non-Invasive Means to Influence Mesenchymal Stem Cell Fate, Promoting Bone and Suppressing the Fat Phenotype. Bonekey Osteovision 6, 132–149 (2009). 58. Chen, X. et al. Force and scleraxis synergistically promote the commitment of human ES cells derived MSCs to tenocytes. Sci. Rep. 2, 977 (2012). 59. Humphreys, B. D. et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am. J. Pathol. 176, 85–97 (2010). 60. Rock, J. R. et al. Multiple stromal populations contribute to pulmonary fibrosis without evidence for epithelial to mesenchymal transition. Proc. Natl. Acad. Sci. U. S. A. 108, E1475–83 (2011). 61. Brody, A. R., Salazar, K. D. & Lankford, S. M. Mesenchymal stem cells modulate lung injury. Proc. Am. Thorac. Soc. 7, 130–3 (2010). 62. Kramann, R., Dirocco, D. P. & Humphreys, B. D. Understanding the origin, activation and regulation of matrix-producing myofibroblasts for treatment of fibrotic disease. J. Pathol. 231, 273–89 (2013). 63. Omary, M. B., Lugea, A., Lowe, A. W. & Pandol, S. J. The pancreatic stellate cell  : a star on the rise in pancreatic diseases. J. Clin. Invest. 117, 50–59 (2007). 64. Wilson, J. S., Pirola, R. C. & Apte, M. V. Stars and stripes in pancreatic cancer: role of stellate cells and stroma in cancer progression. Front. Physiol. 5, 52 (2014). 65. Lua, I., James, D., Wang, J., Wang, K. S. & Asahina, K. Mesodermal mesenchymal cells give rise to myofibroblasts, but not epithelial cells, in mouse liver injury. Hepatology (2014). doi:10.1002/hep.27035 66. Mederacke, I. et al. Fate tracing reveals hepatic stellate cells as dominant contributors to liver fibrosis independent of its aetiology. Nat. Commun. 4, 2823 (2013). 67. Zhang, H., van Olden, C., Sweeney, D. & Martin-Rendon, E. Blood vessel repair and regeneration in the ischaemic heart. Open Hear. 1, e000016–e000016 (2014). 68. Gurtner, G. C., Werner, S., Barrandon, Y. & Longaker, M. T. Wound repair and regeneration. Nature 453, 314–21 (2008). 	   81	  69. Atala, A., Irvine, D. J., Moses, M. & Shaunak, S. Wound Healing Versus Regeneration: Role of the Tissue Environment in Regenerative Medicine. MRS Bull. 35, (2010). 70. Stappenbeck, T. S. & Miyoshi, H. The role of stromal stem cells in tissue regeneration and wound repair. Science 324, 1666–9 (2009). 71. Özpolat, B. et al. REGENERATION OF THE ELBOW JOINT IN THE DEVELOPING CHICK EMBRYO RECAPITULATES DEVELOPMENT. Dev. Biol. 372, 229–238 (2013). 72. Doppler, S. a, Deutsch, M.-A., Lange, R. & Krane, M. Cardiac regeneration: current therapies-future concepts. J. Thorac. Dis. 5, 683–97 (2013). 73. Razzell, W., Wood, W. & Martin, P. Recapitulation of morphogenetic cell shape changes enables wound re-epithelialisation. Development 141, 1814–20 (2014). 74. Hecker, L. & Thannickal, V. J. Nonresolving fibrotic disorders: idiopathic pulmonary fibrosis as a paradigm of impaired tissue regeneration. Am. J. Med. Sci. 341, 431–4 (2011). 75. Thannickal, V. J. Mechanistic links between aging and lung fibrosis. Biogerontology (2013). doi:10.1007/s10522-013-9451-6 76. Nalysnyk, L., Cid-Ruzafa, J., Rotella, P. & Esser, D. Incidence and prevalence of idiopathic pulmonary fibrosis: review of the literature. Eur. Respir. Rev. 21, 355–61 (2012). 77. Fehrer, C. & Lepperdinger, G. Mesenchymal stem cell aging. Exp. Gerontol. 40, 926–30 (2005). 78. Burtner, C. R. & Kennedy, B. K. Progeria syndromes and ageing: what is the connection? Nat. Rev. Mol. Cell Biol. 11, 567–78 (2010). 79. Zhang, J. et al. A human iPSC model of Hutchinson Gilford Progeria reveals vascular smooth muscle and mesenchymal stem cell defects. Cell Stem Cell 8, 31–45 (2011). 80. Scaffidi, P. & Misteli, T. Lamin A-dependent misregulation of adult stem cells associated with accelerated ageing. Nat. Cell Biol. 10, 452–459 (2008). 81. Kapetanaki, M. G., Mora, A. L. & Rojas, M. Influence of age on wound healing and fibrosis. J. Pathol. 229, 310–22 (2013). 82. Hinz, B. et al. The myofibroblast: one function, multiple origins. Am. J. Pathol. 170, 1807–16 (2007). 	   82	  83. Crisan, M. et al. A perivascular origin for mesenchymal stem cells in multiple human organs. Cell Stem Cell 3, 301–13 (2008). 84. Lama, V. N. & Phan, S. H. The extrapulmonary origin of fibroblasts: stem/progenitor cells and beyond. Proc. Am. Thorac. Soc. 3, 373–6 (2006). 85. Tatler, A. L. & Jenkins, G. TGF-β Activation and Lung Fibrosis. Proc. Am. Thorac. Soc. 9, 130–6 (2012). 86. Mouratis, M. a & Aidinis, V. Modeling pulmonary fibrosis with bleomycin. Curr. Opin. Pulm. Med. 17, 355–61 (2011). 87. Chung, M. P. et al. Role of repeated lung injury and genetic background in bleomycin-induced fibrosis. Am. J. Respir. Cell Mol. Biol. 29, 375–80 (2003). 88. Degryse, A. L. & Lawson, W. E. Progress toward improving animal models for idiopathic pulmonary fibrosis. Am. J. Med. Sci. 341, 444–9 (2011). 89. Bataller, R. & Brenner, D. A. Liver fibrosis. J. Clin. Invest. 115, 209–218 (2005). 90. Blaner, W. S. et al. HEPATIC STELLATE CELL LIPID DROPLETS: A SPECIALIZED LIPID DROPLET FOR RETINOID STORAGE. Biochim. Biophys. Acta 1791, 467–473 (2010). 91. Friedman, S. L. Hepatic Stellate Cells: Protean, Multifunctional, and Enigmatic Cells of the Liver. Physiol. Rev. 88, 125–172 (2010). 92. Apte, M. V et al. Periacinar stellate shaped cells in rat pancreas: identification, isolation, and culture. Gut 43, 128–33 (1998). 93. Erkan, M. et al. StellaTUM: current consensus and discussion on pancreatic stellate cell research. Gut 61, 172–8 (2012). 94. Rehan, V. K. et al. Evidence for the presence of lipofibroblasts in human lung. Exp. Lung Res. 32, 379–93 (2006). 95. Omary, M. B., Lugea, A., Lowe, A. W. & Pandol, S. J. The pancreatic stellate cell  : a star on the rise in pancreatic diseases. J. Clin. Invest. 117, 50–59 (2007). 96. Hu, Y. et al. Imbalance of wnt/dkk negative feedback promotes persistent activation of pancreatic stellate cells in chronic pancreatitis. PLoS One 9, e95145 (2014). 97. Haber, P. S. et al. Activation of pancreatic stellate cells in human and experimental pancreatic fibrosis. Am. J. Pathol. 155, 1087–95 (1999). 	   83	  98. Ding, Z., Maubach, G., Masamune, a & Zhuo, L. Glial fibrillary acidic protein promoter targets pancreatic stellate cells. Dig. Liver Dis. 41, 229–36 (2009). 99. Sparmann, G. et al. Generation and characterization of immortalized rat pancreatic stellate cells. Am. J. Physiol. Gastrointest. Liver Physiol. 287, G211–9 (2004). 100. Rucki, A. A. & Zheng, L. Pancreatic cancer stroma: Understanding biology leads to new therapeutic strategies. World J. Gastroenterol. 20, 2237–2246 (2014). 101. Mastracci, T. L. & Sussel, L. The endocrine pancreas: insights into development, differentiation, and diabetes. Wiley Interdiscip. Rev. Dev. Biol. 1, 609–28 (2012). 102. Bock, T., Pakkenberg, B. & Buschard, K. Genetic Background Determines the Size and Structure of the Endocrine Pancreas. Diabetes 54, 133–137 (2005). 103. Bockman, D. E. The Pancreas: Biology, Pathobiology and Disease. 1–8 (Raven Press, 1993). 104. Jørgensen, M. C. et al. An illustrated review of early pancreas development in the mouse. Endocr. Rev. 28, 685–705 (2007). 105. Delporte, C. Aquaporins in salivary glands and pancreas. Biochim. Biophys. Acta 1840, 1524–1532 (2014). 106. Naya, F. J. et al. Diabetes, defective pancreatic morphogenesis, and abnormal enteroendocrine differentiation in BETA2/NeuroD-deficient mice. Genes Dev. 11, 2323–2334 (1997). 107. Gittes, G. K. Developmental biology of the pancreas: a comprehensive review. Dev. Biol. 326, 4–35 (2009). 108. Dorrell, C. et al. Isolation of mouse pancreatic alpha, beta, duct and acinar populations with cell surface markers. Mol. Cell. Endocrinol. 339, 144–50 (2011). 109. Hyun, J. J. & Lee, H. S. Experimental Models of Pancreatitis. Clin. Endosc. 47, 212–216 (2014). 110. Arda, H. E., Benitez, C. M. & Kim, S. K. Gene regulatory networks governing pancreas development. Dev. Cell 25, 5–13 (2013). 111. Lerch, M. M. & Gorelick, F. S. Models of acute and chronic pancreatitis. Gastroenterology 144, 1180–93 (2013). 112. Lowenfels, A. B. & Maisonneuve, P. Epidemiology and prevention of pancreatic cancer. Jpn. J. Clin. Oncol. 34, 238–44 (2004). 	   84	  113. Yadav, D. & Lowenfels, A. B. The epidemiology of pancreatitis and pancreatic cancer. Gastroenterology 144, 1252–61 (2013). 114. Reznik, R., Hendifar, A. E. & Tuli, R. Genetic determinants and potential therapeutic targets for pancreatic adenocarcinoma. Front. Physiol. 5, 87 (2014). 115. Raijman, I. & Levin, B. The Pancreas: Biology, Pathobiology and Diseases. 899–912 (Lippincott Williams & Wilkins, 1993). 116. Ley, S. H., Hamdy, O., Mohan, V. & Hu, F. B. Prevention and management of type 2 diabetes: dietary components and nutritional strategies. Lancet 383, 1999–2007 (2014). 117. DiMagno, E. P., Layer, P. & Clain, J. E. The Pancreas: Biology, Pathobiology and Disease. 665–706 (Raven Press, 1993). 118. Jupp, J., Fine, D. & Johnson, C. D. The epidemiology and socioeconomic impact of chronic pancreatitis. Best Pract. Res. Clin. Gastroenterol. 24, 219–31 (2010). 119. Aghdassi, A. a et al. Animal models for investigating chronic pancreatitis. Fibrogenesis Tissue Repair 4, 26 (2011). 120. Rickels, M. R. et al. Detection, evaluation and treatment of diabetes mellitus in chronic pancreatitis: recommendations from PancreasFest 2012. Pancreatology 13, 336–42 (2013). 121. Forsmark, C. E. Management of chronic pancreatitis. Gastroenterology 144, 1282–91.e3 (2013). 122. Spanier, B. W. M., Dijkgraaf, M. G. W. & Bruno, M. J. Epidemiology, aetiology and outcome of acute and chronic pancreatitis: An update. Best Pract. Res. Clin. Gastroenterol. 22, 45–63 (2008). 123. Poxleitner, P. J., Seifert, G., Richter, S. C., Hopt, U. T. & Wittel, U. a. Infected pancreatic necrosis increases the severity of experimental necrotizing pancreatitis in mice. Pancreas 42, 1150–6 (2013). 124. Mews, P. et al. Pancreatic stellate cells respond to inflammatory cytokines: potential role in chronic pancreatitis. Gut 50, 535–41 (2002). 125. Shimizu, K. Mechanisms of pancreatic fibrosis and applications to the treatment of chronic pancreatitis. J. Gastroenterol. 43, 823–32 (2008). 126. Talukdar, R. & Tandon, R. K. Pancreatic stellate cells: new target in the treatment of chronic pancreatitis. J. Gastroenterol. Hepatol. 23, 34–41 (2008). 	   85	  127. Masamune, A., Watanabe, T., Kikuta, K. & Shimosegawa, T. Roles of pancreatic stellate cells in pancreatic inflammation and fibrosis. Clin. Gastroenterol. Hepatol. 7, S48–54 (2009). 128. Schepers, N. J., Besselink, M. G. H., van Santvoort, H. C., Bakker, O. J. & Bruno, M. J. Early management of acute pancreatitis. Best Pract. Res. Clin. Gastroenterol. 27, 727–43 (2013). 129. Ketwaroo, G. a & Sheth, S. Autoimmune pancreatitis. Gastroenterol. Rep. 1, 27–32 (2013). 130. Schwaiger, T. et al. Autoimmune pancreatitis in MRL/Mp mice is a T cell-mediated disease responsive to cyclosporine A and rapamycin treatment. Gut 63, 494–505 (2014). 131. Demols, A. et al. Endogenous interleukin-10 modulates fibrosis and regeneration in experimental chronic pancreatitis. Am. J. Physiol. Gastrointest. Liver Physiol. 282, G1105–12 (2002). 132. Holst, J. J., Jensen, S. L. & Morley, J. S. Neural regulation of pancreatic hormone secretion by the C-terminal tetrapptide of CCK. Nature 284, 33–38 (1980). 133. Phillips, P. a et al. Pancreatic stellate cells produce acetylcholine and may play a role in pancreatic exocrine secretion. Proc. Natl. Acad. Sci. U. S. A. 107, 17397–402 (2010). 134. Meirelles, L. D. S., Fontes, A. M., Covas, D. T. & Caplan, A. I. Mechanisms involved in the therapeutic properties of mesenchymal stem cells. Cytokine Growth Factor Rev. 20, 419–27 (2009). 135. Kalajzic, I. et al. Use of Type I Collagen Green Fluorescent Protein Transgenes to Identify Subpopulations of Cells at Different Stages of the Osteoblast Lineage. J Bone Min. Res 17, 15–25 (2002). 136. Tennakoon, A. H. et al. Characterization of glial fibrillary acidic protein (GFAP)-expressing hepatic stellate cells and myofibroblasts in thioacetamide (TAA)-induced rat liver injury. Exp. Toxicol. Pathol. 65, 1159–71 (2013). 137. Zhang, Q. et al. Metabolic regulation of SIRT1 transcription via a HIC1:CtBP corepressor complex. Proc. Natl. Acad. Sci. U. S. A. 104, 829–33 (2007). 138. Dehennaut, V., Loison, I., Pinte, S. & Leprince, D. Molecular dissection of the interaction between HIC1 and SIRT1. Biochem. Biophys. Res. Commun. 421, 384–8 (2012). 	   86	  139. Cheng, G. et al. HIC1 silencing in triple-negative breast cancer drives progression through misregulation of LCN2. Cancer Res. (2013). doi:10.1158/0008-5472.CAN-13-2420 140. Gao, X. et al. Bone Morphogenetic Protein Signaling Protects against Cerulein-Induced Pancreatic Fibrosis. PLoS One 9, e89114 (2014). 141. Apte, M. V, Wilson, J. S., Lugea, A. & Pandol, S. J. A Starring Role for Stellate Cells in the Pancreatic Cancer Microenvironment. Gastroenterology 144, 1210–1219 (2013). 142. Bachem, M. G. et al. Pancreatic carcinoma cells induce fibrosis by stimulating proliferation and matrix synthesis of stellate cells. Gastroenterology 128, 907–921 (2005). 143. McCarroll, J. a et al. Vitamin A inhibits pancreatic stellate cell activation: implications for treatment of pancreatic fibrosis. Gut 55, 79–89 (2006). 144. Apte, M. V. et al. Desmoplastic Reaction in Pancreatic Cancer. Pancreas 29, 179–187 (2004). 145. Gao, X. et al. BMP2 inhibits TGF-β-induced pancreatic stellate cell activation and extracellular matrix formation. Am. J. Physiol. Gastrointest. Liver Physiol. 304, G804–13 (2013). 146. Ishiwatari, H. et al. Treatment of pancreatic fibrosis with siRNA against a collagen-specific chaperone in vitamin A-coupled liposomes. Gut 62, 1328–39 (2013). 147. Masamune, A. et al. Connexins regulate cell functions in pancreatic stellate cells. Pancreas 42, 308–16 (2013). 148. Dimmeler, S., Ding, S., Rando, T. a & Trounson, A. Translational strategies and challenges in regenerative medicine. Nat. Med. 20, 814–821 (2014). 149. Binder, B. Y. K., Sagun, J. E. & Leach, J. K. Reduced Serum and Hypoxic Culture Conditions Enhance the Osteogenic Potential of Human Mesenchymal Stem Cells. Stem Cell Rev. (2014). doi:10.1007/s12015-014-9555-7 150. Tomita, Y. et al. Pancreatic Fatty Degeneration and Fibrosis as Predisposing Factors for the Development of Pancreatic Ductal Adenocarcinoma. Pancreas 00, 1–10 (2014). 151. Zheng, J. et al. Signification of Hypermethylated in Cancer 1 (HIC1) as Tumor Suppressor Gene in Tumor Progression. Cancer Microenviron. 1, (2012). 	   87	  152. Bethel, M., Chitteti, B., Srour, E. & Kacena, M. The Changing Balance Between Osteoblastogenesis and Adipogenesis in Aging and its Impact on Hematopoiesis. Curr. Osteoporos. Rep. 11, 99–106 (2013). 153. Hecker, L. et al. Reversal of Persistent Fibrosis in Aging by Targeting Nox4-Nrf2 Redox Imbalance. Sci. Transl. Med. 6, 231ra47–231ra47 (2014). 154. Teng, I.-W. et al. Targeted methylation of two tumor suppressor genes is sufficient to transform mesenchymal stem cells into cancer stem/initiating cells. Cancer Res. 71, 4653–63 (2011). 155. Shehata, M., van Amerongen, R., Zeeman, A. L., Giraddi, R. R. & Stingl, J. The influence of tamoxifen on normal mouse mammary gland homeostasis. Breast Cancer Res. 16, 411 (2014). 156. Marzoq, A. J., Giese, N., Hoheisel, J. D. & Alhamdani, M. S. S. Proteome Variations in Pancreatic Stellate Cells upon Stimulation with Proinflammatory Factors. J. Biol. Chem. 288, 32517–27 (2013). 157. Landsman, L. et al. Pancreatic mesenchyme regulates epithelial organogenesis throughout development. PLoS Biol. 9, e1001143 (2011). 158. Dulauroy, S., Di Carlo, S. E., Langa, F., Eberl, G. & Peduto, L. Lineage tracing and genetic ablation of ADAM12(+) perivascular cells identify a major source of profibrotic cells during acute tissue injury. Nat. Med. 18, (2012).     


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items