Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of fibroblast phenotype and pericellular matrix in wound healing Mah, Wesley Brandon 2013

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

Item Metadata

Download

Media
24-ubc_2013_spring_mah_wesley.pdf [ 67.74MB ]
Metadata
JSON: 24-1.0073674.json
JSON-LD: 24-1.0073674-ld.json
RDF/XML (Pretty): 24-1.0073674-rdf.xml
RDF/JSON: 24-1.0073674-rdf.json
Turtle: 24-1.0073674-turtle.txt
N-Triples: 24-1.0073674-rdf-ntriples.txt
Original Record: 24-1.0073674-source.json
Full Text
24-1.0073674-fulltext.txt
Citation
24-1.0073674.ris

Full Text

THE ROLE OF FIBROBLAST PHENOTYPE AND PERICELLULAR MATRIX IN WOUND HEALING by Wesley Brandon Mah  B.Sc., 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 Studies (Cell and Developmental Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  March 2013 ©Wesley Brandon Mah, 2013  !  ABSTRACT Scar formation as a result of wound healing in skin is associated with increased deposition of extracellular matrix (ECM) and reduced ECM turnover by fibroblasts. Remarkably, wound healing in the human oral mucosa results in scarless healing, while wound healing in the skin can often result in scarring. Therefore comparing fibroblast phenotype and interactions with their ECM niche in the gingiva and skin may provide novel information about the factors that regulate scar formation. To this end, a novel 3D cell culture model was utilized to yield a cellular microenvironment (niche) that closely mimics the in vivo situation and primary gingival (GFBL) and skin fibroblasts (SFBL) phenotype was characterized. Furthermore, fibroblasts were reseeded on cell-free 3D ECM derived from GFBL and SFBL and the effects of the 3D ECM on cell phenotype were analyzed. Interestingly, SFBL in 3D cultures had greater expression of ECM deposition associated genes, including collagens, matricellular proteins, SLRPs, TGF-!1 and CTGF, intracellular ECM degradation and myofibroblast differentiation and function-associated genes, while GFBL had a greater expression of matrix remodeling associated genes (MMPs). We also found that the 3D cultures showed a significant difference in expression of certain genes (MMPs and myofibroblast function-associated genes) between GFBL and SFBL compared to cells reseeded on the 3D ECM or 2D control substrate. Thus, the 3D culture conditions may differentially regulate expression of a subset of genes in these cells. Interestingly, SFBL had a greater expression of matrix deposition associated genes (collagens, SLRPs, tenascins) irrespective of the culture conditions, suggesting that expression of these genes is inherently distinct between GFBL and SFBL. This was associated with greater autogenous TGF-! expression and SMAD3 phosphorylation in SFBL than GFBL, which may partly explain the innate difference in gene expression. In addition, there was greater ERK1/2 phosphorylation in fibroblasts when seeded on 3D ECM compared to 2D substrate. Greater ERK1/2 phosphorylation may have promoted greater expression of AP-1-dependent MMPs seen in SFBL and GFBL on 3D ECM. In conclusion, reduced expression of matrix deposition associated genes and greater expression of matrix remodeling genes in GFBL may contribute to scarless healing in gingiva.  !  !  ""!  TABLE OF CONTENTS ABSTRACT...................................................................................................................................ii TABLE OF CONTENTS...........................................................................................................iii LIST OF TABLES......................................................................................................................vii LIST OF FIGURES.....................................................................................................................ix LIST OF SYMBOLS AND ABBREVIATIONS................................................................xvi ACKNOWLEDGMENTS.......................................................................................................xvii 1. CHAPTER ONE – REVIEW OF THE LITERATURE......................................1 1.1. Introduction: overview of scar formation.........................................................................1 1.2. Overview of wound healing................................................................................................3 1.2.1. Hemostasis and inflammatory response...............................................................4 1.2.2. Re-epithelialization and granulation tissue formation.......................................4 1.2.3. Maturation and tissue remodeling..........................................................................6 1.3. Factors that are associated with scar formation..............................................................7 1.3.1. Prolonged inflammatory response.........................................................................7 1.3.2. Increased transforming growth factor-! (TGF-!) activity...............................8 1.3.3. Cell-ECM interaction...............................................................................................9 1.3.4. Fibroblast phenotype..............................................................................................11 1.4. Current therapies to eliminate scar formation...............................................................12 1.4.1. Surgical therapy......................................................................................................13 1.4.2. Pressure therapy......................................................................................................13 1.4.3. Silicone therapy.......................................................................................................13 1.4.4. Emerging approaches in scar-reducing therapies...........................................14 1.5. Scar-free healing in fetal wounds................................................................................... 15 1.6. Regeneration in amphibians and mammals.................................................................. 16 1.7. Scarless healing in gingiva versus scar-forming healing in skin...............................17 1.7.1. Anatomical and functional properties of the skin and gingiva......................17 1.7.2. Comparing wound healing in gingiva and skin................................................18 1.8. Fibroblasts.............................................................................................................................20 1.8.1. Properties of fibroblasts.........................................................................................20  !  !  """!  1.8.2. Fibroblast phenotype in oral mucosa and skin.................................................20 1.9. Extracellular matrix (ECM)..............................................................................................22 1.9.1. Overview of the ECM............................................................................................22 1.9.2. Pericellular matrix (PCM).................................................................................... 23 1.9.3. Fibrillar ECM protein..........................................................................................24 1.9.4. Glycoproteins and matricellular proteins...........................................................25 1.9.5. Small leucine-rich proteoglycans (SLRPs)...................................................... 26 1.9.6. Growth factors.........................................................................................................28 1.9.7. Matrix metalloproteinases and their inhibitors.................................................28 1.10. Cell-derived three-dimensional in vivo-like cell culture model.................................30 1.10.1. Generation and characterization of the cell-derived 3D cell cultures..........31 2. CHAPTER TWO – AIMS OF THE STUDY......................................................34 2.1. General hypothesis..............................................................................................................34 2.2. Aim (I)...................................................................................................................................34 2.3. Hypothesis (I).......................................................................................................................34 2.4. Aim (II)................................................................................................................................. 34 2.5. Hypothesis (II) ................................................................................................................... 34 3. CHAPTER THREE – MATERIALS AND METHODS.....................................35 3.1. Cells........................................................................................................................................35 3.2. In vivo-like three-dimensional (3D) cell culture...........................................................35 3.3. Generation of cell-free fibroblast-derived 3D ECM....................................................35 3.4. Fibroblast reseeding on fibroblast-derived 3D ECM and collagen substrates.......35 3.5. Measurement of fibroblast proliferation.........................................................................36 3.6. Quantification of total protein deposited in the cell-derived 3D ECM....................36 3.7. Immunostaining analysis...................................................................................................36 3.8. Scanning electron microscopy (SEM)........................................................................... 37 3.9. Quantitative analysis of mRNA expression using real-time RT-PCR.....................38 3.10. Quantification of collagen abundance in 3D ECM.....................................................39 3.11. Quantification of sulphated glycosaminoglycan in 3D ECM...................................40 3.12. Western blotting.................................................................................................................40 3.12.1. Preparation of ECM proteins for Western blotting..........................................40 !  !  "#!  3.12.2. Preparation of cell lysates for Western blotting...............................................41 3.12.3. Western blotting......................................................................................................41 3.13. Cell adhesion assessment..................................................................................................42 3.14. Statistical analysis..............................................................................................................42 4. CHAPTER FOUR – RESULTS..........................................................................43 4.1. Phenotypic Characterization of gingival- (GFBL) and skin-derived fibroblasts (SFBL).....................................................$$$...........................................................................43! 4.1.1. Distinct cellular morphology and ECM organization in GFBL and SFBL 3D ECM.....................................................................................................................43! 4.1.2. GFBL had greater proliferation rate than SFBL...............................................43! 4.1.3. GFBL and SFBL showed no significant difference in total protein abundance in the 3D cultures...........$$$.............................................................$$$$$43! 4.1.4. GFBL and SFBL displayed distinct gene expression profiles in 3D cultures......................................$$$$$$......................................$$$$$$...............................$44! 4.1.5. Gene expression remained significantly different between GFBL and SFBL during 3D ECM generation over time.................................................................46! 4.1.6. Characterization of ECM protein abundance....................................$$$$$$$$$$$$$$$$$46! 4.2. The role of the ECM on cell phenotype...........................................$$$$$$........................$48 4.2.1. GFBL and SFBL showed no difference in adhesion rate on either GFBLderived 3D ECM (GECM) and SFBL-derived 3D ECM (SECM)..............!48! 4.2.2. There were no significant differences between GECM and SECM in modulating gene expression..................................................................................49! 4.2.3. Cell signaling pathways........................$$$$$$..........................................................$50! 4.2.4. Phosphorylation of SMAD3 was greater in SFBL......................................$$$$$50! 4.2.5. ERK1/2 phosphorylation was greater in fibroblasts when seeded on 3D ECM compared to collagen I substrate...............................................................51! 5. CHAPTER FIVE – DISCUSSION......................................................................52 5.1. GFBL and SFBL showed an innately distinct gene expression profile...................52 5.2. Fetal-like characteristics of GFBL...................................................................................53 5.3. Characterization of the 3D ECM......................................................................................53! 5.4. Impact of the 3D culture niche on cell phenotype$$......................................................55! !  !  #!  5.5. Cell adhesion........................................................................................................................57! 5.6. Greater expression of TGF-! and SMAD3 phosphorylation in SFBL and its potential implication on matrix deposition associated gene expression..................57! 5.7. Potential association and mechanisms of differentially regulated individual genes in wound healing and scar formation.............................................................................$59! 5.7.1. Collagens..................................................................................................................$59! 5.7.2. Tenascins..................................................................................................................60! 5.7.3. SLRPs.......................................................................................................................$60! 5.7.4. MMPs........................................................................................................................60! 5.8. Putative regulation of MMP and VEGF-" expression by ERK1/2 and AP-1 pathway.................................................................................................................................$62! 6. CHAPTER SIX – CONCLUSIONS AND FUTURE DIRECTIONS...................64 6.1. Conclusions..........................................................................................................................64 6.2. Future directions..................................................................................................................65 6.2.1. Senescence phenotype............................................................................................65 6.2.2. Autogenous TGF-! signaling...............................................................................65 6.2.3. Downstream mediators of TGF-! signaling that are associated with fibrosis.......................................................................................................................65 6.2.4. Effect of Wnt/!-catenin pathway and fibroblast origin on gene expression .....................................................................................................................................66 6.2.5. Further characterization of the 3D ECM derived from GFBL and SFBL..67 6.2.6. Long-term goals......................................................................................................67 CHAPTER SEVEN – TABLES AND FIGURES......................................................68 REFERENCES.......................................................................................................121  !  !  #"!  LIST OF TABLES ! Table 1. In vitro studies comparing fibroblasts derived from skin and various regions of the oral mucosa. Modified from H#kkinen et al., 2012......................................69 Table 2. Matricellular proteins that have a role in the wound healing process and are synthesized by fibroblasts..............................................................................................................70 Table 3. Key growth factors involved in wound healing and its ability to interact with PCM molecules........................................................................................................................71 Table 4. List of target proteins and indirect activation of MMPs, cytokines and growth factors by MMPs. The target proteins are not limited to the list above. Table is a modification from McCawley and Matrisian, 2001. The MMPs listed have been shown to be expressed by fibroblasts...............................................................................................................73 Table 5. List and origin of the cells used...................................................................................74 Table 6. List of antibodies used for immunohistochemistry and Western blotting....75 Table 7. Primers used for real-time PCR analysis.................................................................76 Table 8. Summary of the gene expression analysis of the genes that were significantly different between GFBL and SFBL in 3D cultures 7 days post-seeding. Predetermined: Genes that were significantly different between GFBL and SFBL, irrespective to the culture conditions (3D cultures 7 days post-seeding, 24 h after reseeding on cell-derived 3D matrix or collagen I substrate, with or without FBS and ascorbic acid). Therefore, the expression of these genes is innately different between GFBL and SFBL, and does not depend on the environment (culture substrate and culture medium). Interestingly, the majority of these genes that showed significantly greater expression in SFBL were matrix deposition associated genes. 3D cultures 7 days postseeding: Genes that were only significantly different between GFBL and SFBL in 3D culture 7 days post-seeding. Differential expression of these genes is therefore dependent on the 3D culture environment. Overall, most matrix remodeling MMPs, ECM internalization and intracellular degradation-associated genes (Endo180 and CTSK) and various matricellular protein genes (SPARC, THBS-2, OPN) were found in this category. Serum response: Genes that showed significant differences between GFBL and SFBL only in response to serum (3D cultures 7 days post-seeding and 24 h after reseeding on collagen I substrate with the presences of FBS and ascorbic acid). Expression of CTGF was significantly affected by serum and was greater in SFBL than GFBL. Statistical analysis was performed using Student’s t-test. GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM)................................................................................................................82  !  !  #""!  Table 9. Gene expression analysis between fibroblasts seeded on 3D ECM compared to collagen type I substrate. Expression of MMP-1, -3 and -10 was significantly greater when GFBL and SFBL were seeded on 3D ECM compared to collagen I substrate. Arrow indicates that expression of target gene was greater on GECM/SECM relative to collagen substrate in GFBL and SFBL. Values indicated in the table represent p-values (* p<0.05, **p<0.01; Student’s t-test). GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) ...............................................................................................................................................83 Table 10. Summary of TGF-!-dependent genes differentially expressed by GFBL and SFBL. Greater expression of fibrillar ECM proteins and tenascin-C, tenascin-X and SLRPs in SFBL may be attributed to increased activation of endogenous TGF-! signaling and phosphorylation of SMAD3. Genes that have been previously shown to be regulated by TGF-! are shown. !": Expression of the TGF-! responsive gene was higher or lower in SFBL relative to GFBL; = : expression of the TGF-! responsive gene was the same in GFBL and SFBL.................................................................................................................................84 ! !  !  !  #"""!  LIST OF FIGURES ! Fig. 1. A schematic presentation of the role of ECM components on cell function. The fibrillar ECM provides mechanical stimuli to the cells and acts as a ligand for receptormediated signaling. Other components of the ECM including glycoproteins, proteoglycans and matricellular proteins modulate (can promote or inhibit) cell-ECM interactions through binding to ECM proteins, cell surface receptors and molecules such as proteases and growth factors.......................................................................................................85 Fig. 2. Key small leucine-rich proteoglycans (SLRPs) that regulate wound healing. Asporin (ASPN), biglycan (BGN) and decorin (DCN) belong in class I and fibromodulin (FMOD) and lumican (LUM) belong in class II of the SLRP family.....................................86 Fig. 3. Representative GFBL (A-D) and SFBL (E-H) images comparing morphology and organization in 3D cultures 7 days post-seeding. GFBL were more elongated and spindle-like compared to SFBL. This is well demonstrated from the shape of the nuclei seen in D and H. GFBL in 3D cultures were more polar and parallel to one another, while SFBL were organized in a more random fashion........................................................................87 Fig. 4. SEM images of representative GFBL (A) and SFBL (B) in 3D cultures 7 days post-seeding, and of the 3D ECM after cell removal (C-F). (A,B) GFBL were more elongated and spindle-like compared to SFBL. (C-F) The fibrillar network of GFBLderived 3D ECM was organized parallel to one another, while SFBL-derived fibrillar network was organized in a more random fashion. (E-F) Tannic acid was also incubated in the fixation step to better preserve proteoglycans and glycoproteins in the ECM. (C-D) Without tannic acid the organization of the fibrillar network was better perceived............88 Fig. 5. GFBL proliferated faster but produced equal amounts of total proteins into the 3D ECM over time as compared to SFBL. (A,B) Cell number at high density (3D) (A) and low density (B) cultures was assessed by measuring total RNA concentration ($g/$l) (A) and by MTT assay (B). (C) Total protein abundance ($g/$l) in 3D ECM of SFBL and GFBL cultures. (D) Protein abundance in 3D ECM relative to cell number (mg protein/$g RNA) between SFBL and GFBL 3D cultures. Protein assays (C,D) are representative of 3 repeated experiments. Results show mean +/- SEM. N=5 (A,C,D), N=3 (B) parallel cell lines. (*p<0.05; Student’s t-test)..............................................................89 Fig. 6. Greater expression of fibrillar ECM proteins in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, ***p<0.001; Student’s ttest)........................................................................................................................................................90 Fig. 7. Greater expression of matricellular proteins in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01,***p<0.001; Student’s t-test). THBS-1: thrombospondin-1; THBS-2: thrombospondin-2.......................91  !  !  "%!  Fig. 8. Greater expression of small leucine-rich proteoglycans in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (**p<0.01, ***p<0.001; Student’s t-test). ASPN: asporin; BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican.....................................................................................................................................92 Fig. 9. Greater expression of small leucine-rich proteoglycans in individual parallel SFBL compared to GFBL lines in 3D cultures 7 days post-seeding. Results show the relative proportion of individual SLRPs being expressed in GFBL and SFBL lines. DCN (indicated in yellow) was most highly expressed and FMOD (indicated in red) was the least expressed in all cell lines. Results were generated by 2(Ctreference-Cttarget) method. BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican.............................................93 Fig. 10. Greater expression of pro-fibrotic growth factors and Cthrc1 in SFBL compared to GFBL in 3D cultures 7 days post-seeding. (A) Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01, ***p<0.001; Student’s t-test). (B) Results show the relative proportion of TGF!1, -!2 and -!3 in individual parallel GFBL and SFBL lines. In general, expression of total TGF-! was greater in SFBL compared to parallel GFBL. On average, TGF-!3 expression made up 2% and 8% out of total TGF-! expression in GFBL and SFBL, respectively. Results were generated by 2(Ctreference-Cttarget) method. TGF-!R1: TGF-! receptor 1; TGF-!R2: TGF-! receptor 2; CXCL12: stromal-derived factor 1; Cthrc1: collagen triple helix repeat containing 1........................................................................................94 Fig. 11. Expression of MMPs and TIMP-4 was different in GFBL compared to SFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01; Student’s t-test).....................................................................................................................................................95 Fig. 12. Expression of MMPs was greater in GFBL compared to SFBL in 3D cultures 7 days post-seeding. Results show the relative proportion of individual MMPs being expressed in individual parallel GFBL and SFBL lines. Note that MMP-1 and -10 (indicated in navy and light blue) were most highly expressed out of all MMPs studied in four out of five GFBL lines and MMP-2 (indicated in red) was most abundant in all SFBL lines. Results were generated by 2(Ctreference-Cttarget) method............................................................96 Fig. 13. Expression of cell contractility and myofibroblast associated genes was greater in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, ** p<0.01, ***p<0.001; Student’s t-test). "-SMA: "-smooth muscle actin; NMMIIA: non-muscle myosin IIA; NMMIIB: non-muscle myosin IIB....................97  !  !  %!  Fig. 14. Expression of genes involved in internalization (Endo180 and LRP1) and intracellular degradation of ECM molecules (CTSK) in 3D cultures 7 days postseeding. Expression was calculated relative to one GFBL line. Results show mean +/SEM. N=5 parallel cell lines. (*p<0.05, ***p<0.001; Student’s t-test). Endo180 (CD280); LRP1: lipoprotein receptor-related protein 1; CTSK: cathepsin K.........................................98 Fig. 15. Expression of key genes by GFBL and SFBL remained significantly different in 3D cultures over time. Results showed significantly higher expression of several genes over time in SFBL compared with GFBL. Expression was calculated relative to one GFBL line. Relative mRNA expression of all of the genes remained constant over time, except that SFBL showed significantly increased expression of DCN expression over time. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01, ***p<0.001; Student’s t-test and ANOVA for multiple comparisons). DCN: decorin; BGN: biglycan; FMOD: fibromodulin; LUM: lumican; "-SMA: "-smooth muscle actin.......................................................................................................................................................99 Fig. 16. Total collagen abundance and collagen type I accumulation in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. (A,B) Using Sircol Biocolor assay, there was no significant difference in total collagen abundance relative to cell number (A) and relative to total protein (B) between GFBL- and SFBL-derived 3D ECM. Results show mean +/- SEM. N=5 parallel cell lines. (Student’s t-test). (C) Immunostaining of collagen type I before (a,b) and after (c,d) cell extraction in 3D cultures 7 days post-seeding. Collagen organization followed the distinct shape and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and displayed more parallel organization compared to SFBL. .........................................................................................................................100 Fig. 17. Total sulphated glycosaminoglycans (GAG) in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. Using Blyscan Biocolor assay, there was no significant difference in total sulphated GAG abundance, relative to cell number (A) or relative to total protein (B) between GFBL- and SFBL-derived 3D ECM. Results show mean +/SEM. N=5 parallel cell lines. (Student’s t-test).........................................................................101 Fig. 18. Fibromodulin (FMOD) abundance in the cell-derived 3D ECM after chondroitinase ABC (cABC) and keratanase enzyme pretreatment. (A) Western blot analysis of cell-free 3D ECM derived from SFBL-4-1 indicated that cABC digestion increased the available FMOD for anti-FMOD binding relative to keratanase and nonenzyme treated. Control (1) is 3D ECM sample incubated with cABC buffer alone and Control (2) is the 3D ECM sample incubated first in the buffer for cABC for 24 h then in buffer for keratanase for 24 h. (B) Coomassie blue staining was used to indicate the loading of total protein in each lane.............................................................................................102  !  !  %"!  Fig. 19. Small leucine-rich proteoglycan (SLRP) abundance in SFBL- and GFBLderived 3D ECM 7 days post-seeding. SFBL-derived 3D ECM had a significantly greater amount of biglycan (BGN) protein than GFBL-derived 3D ECM (A,E). Decorin (DCN) (B,F), fibromodulin (FMOD) (C,G) and lumican (LUM) (D,H) did not show a significant difference between 3D ECM. However, there were large differences in the amount of DCN deposited by individual cell lines. Results from four parallel GFBL- and SFBL-derived 3D ECM are shown (A-D). The results are representative of three repeated experiments. Results show mean +/- SEM. N=5 parallel cell lines (E-H). (*p<0.05; Student’s t-test).................................................................................................................................103 Fig. 20. Small leucine-rich proteoglycan (SLRP) accumulation in SFBL- and GFBLderived 3D ECM, with or without the presence of cells, 7 days post-seeding. SLRPs were found both intracellularly and extracellularly. In the 3D ECM, they were organized into a fibrillar-like orientation, suggesting that these molecules were associated with collagen or other fibrillar ECM proteins. Semi-quantitative analysis showed no difference in SLRPs abundance in GFBL- and SFBL-derived 3D ECM, except that GFBL showed greater abundance of intracellular immunostaining for DCN, FMOD and LUM as compared with SFBL. Results from one representative GFBL and SFBL line are shown. BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican..............................104 Fig. 21. EDA-Fibronectin (EDA-FN) abundance in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. There was no difference in EDA-FN abundance between SFBL- and GFBL-derived 3D ECM. However, individual cell lines showed considerable variation in abundance of EDA-FN in the 3D ECM. Results show mean +/- SEM. N=5 parallel cell lines. (Student’s t-test)..............................................................................................105 Fig. 22. Tenascin-C abundance in SFBL- and GFBL-derived 3D ECM 7 days postseeding. Analysis of 3D ECM samples by Western blotting without chondroitinase ABC (cABC) pretreatment showed that SFBL deposited significantly greater amount of tenascin-C protein than GFBL in the 3D ECM (A,C). However, when samples were analyzed after cABC treatment, there was a greater amount of tenascin-C protein in GFBL-derived 3D ECM compared to SFBL-derived 3D ECM (p=0.07). Results from four parallel GFBL and SFBL lines are shown and the results are representative of 2 repeated experiments (A-B). Results show mean +/- SEM. N=5 parallel cell lines (C-D). (*p<0.05; Student’s t-test).................................................................................................................................106 Fig. 23. Tenascin-C and EDA-FN accumulation in SFBL- and GFBL-derived 3D ECM, before and after cell extraction, 7 days post-seeding. Tenascin-C and EDA-FN abundance appear to remain unchanged after cell removal. The accumulation of tenascinC and EDA-FN followed the distinct organization and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and showed more parallel organization compared to SFBL. Results from one representative GFBL and SFBL line are shown...107  !  !  %""!  Fig. 24. SFBL appeared to have a greater adhesion rate compared to GFBL. Adhesion of GFBL and SFBL was observed on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) or collagen type I. It appeared that SFBL had a greater adhesion rate than GFBL, regardless of the substrate. Both SFBL and GFBL seemed to adhere more rapidly and adopted a more spindle-like morphology on cell-derived 3D ECM compared to collagen I substrate. There was no difference in adhesion rate within a given cell type when fibroblasts were seeded on SECM or GECM.....................................108 Fig. 25. Expression of fibrillar ECM and matricellular protein genes was significantly greater in SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Expression of type I and III collagen and elastin-1 was significantly higher in SFBL as compared with GFBL. Among the matricellular protein genes, tenascin-C and tenascin-X showed greater expression in SFBL. Expression of differentially regulated genes did not depend on the nature of the matrix. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, ***p<0.001; Student’s t-test)...109 Fig. 26. Expression of SLRP genes was significantly greater in SFBL 24 h postseeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, **p<0.01; Student’s t-test). BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican..............................110 Fig. 27. Expression of MMPs and TIMP4 in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Expression of MMPs was generally greater when fibroblasts were seeded on cellderived 3D ECM compared to collagen I substrate. Only MMP-11 expression was significantly different between SFBL and GFBL. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05; Student’s t-test)..............................................................................................................111 Fig. 28. Expression of growth factor genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No differences in expression of TGF-!1 or CCN-2/CTGF were noted between GFBL and SFBL or between different substrates. SFBL had significantly higher expression of TGF-!3 on collagen I substrate than GFBL, but when cells were seeded in 3D ECM, this difference was reduced. Vascular endothelial growth factor-" (VEGF-") expression was greater in both GFBL and SFBL when seeded on cell-derived 3D ECM compared to collagen I substrate, but this difference did not reach statistical significance. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, Student’s t-test). TGF-!R1: TGF! receptor 1; TGF-!R2: TGF-! receptor 2....................................................................................112  !  !  %"""!  Fig. 29. Expression of myofibroblast ("-SMA, P311), cell contractility (NMMIIA, NMMIIB) associated genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No significant differences in gene expression of "-smooth muscle actin ("-SMA), nonmuscle myosin IIA (NMMIIA), and non-muscle myosin IIB (NMMIIB) between GFBL and SFBL or between different substrates were noted. However, expression of P311 and "11 integrin was significantly higher in SFBL than in GFBL in all substrates. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05; Student’s t-test).....................................................................113 Fig. 30. Intracellular ECM degradation (Endo180, CTSK) associated genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No significant differences in gene expression of Endo180 (CD280) or cathepsin K (CTSK) between GFBL and SFBL or between different substrates were noted. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines..................................................114 Fig. 31. Western blot analysis of total and phosphorylated SMAD3 (p-SMAD3) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) and collagen I substrate. Total SMAD3 was greater in GFBL compared to SFBL, while p-SMAD relative to total SMAD3 was greater in SFBL in all substrates. !-actin was used as a loading control...........................................................115 Fig. 32. Quantification of total SMAD3 protein and phosphorylated SMAD3 (pSMAD3) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. Total SMAD3 was greater in GFBL (A,B), while p-SMAD3 relative to total SMAD3 was greater in SFBL (C,D). (A,C) N=3 parallel GFBL and SFBL cell lines. (B,D) Results show pooled data of GFBL and SFBL (N=9) on three different substrates. Results show mean +/- SEM. (*p<0.05; **p<0.01 Student’s t-test)..............................................................................................................116 Fig. 33. Western blot analysis of total and phosphorylated ERK1/2 (p-ERK1/2) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) and collagen I substrate. ERK1/2 phosphorylation was greater in GFBL and SFBL when seeded on cell-derived 3D ECM. No differences were found in total or phosphorylated ERK1/2 levels between SFBL and GFBL on any of the substrates. !-tubulin was used as a loading control..................................................................117  !  !  %"#!  Fig. 34. Quantification of total and phosphorylated ERK1/2 (p-ERK1/2) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. (A) No difference was found in phosphorylated ERK1/2 relative to total ERK1/2 levels between GFBL and SFBL on any of the substrates. (B,C) ERK1/2 phosphorylation was greater in GFBL (p=0.12) and SFBL (p<0.01) when seeded on cell-derived 3D ECM relative to collagen type I substrate. Results show pooled data from GECM and SECM for the given cell type. Results show mean +/- SEM. A: N=3 of three parallel cell lines. (**p<0.01; Student’s t-test)...............118 Fig. 35. Quantification of total and phosphorylated p38 (p-p38) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. No significant differences were found between GFBL and SFBL or between different substrates. !-tubulin was used as a loading control. Results show mean +/- SEM. N= 3 parallel cell lines. (Student’s t-test).............................119 Fig. 36. A schematic presentation of the signaling pathways that may be differently regulated between GFBL and SFBL (SMAD3) and by 3D ECM and 2D substrate (ERK1/2). We found that SFBL had an innately greater phosphorylation of SMAD3 compared to GFBL. Higher levels of SMAD3 phosphorylation in SFBL may underlie the increased expression of matrix deposition associated genes and decreased expression of matrix remodeling associated genes. When GFBL and SFBL were cultured on 3D ECM, there was a greater phosphorylation of ERK1/2 compared to when these cells were seeded on 2D type I collagen substrate. Higher levels of ERK1/2 phosphorylation may attribute to the greater expression of AP-1-dependent MMP and VEGF-" in GFBL and SFBL seeded on 3D ECM..........................................................................................................................120  !  !  %#!  LIST OF SYMBOLS AND ABBREVIATIONS "-SMA ASPN BGN CCN cABC CTGF Cthrc1 CTSK CXCL12 DCN ECM Endo180 EGF FAK FGF-2/bFGF FMOD GAG GECM GFBL HB-EGF HGF IFN IL KGF LAP LRP1 LRR LTBP LUM MMPs MSC NMMII PCM PDGF SECM SFBL SLRP SPARC TAK1 TGF-! THBS TIMPs VEGF-"  !  "-smooth muscle actin Asporin Biglycan CTGF cysteine-rich protein nephroblastoma overexpressed Chondroitinase ABC Connective tissue growth factor Collagen triple helix repeat containing 1 Cathepsin K Stromal-derived factor-1" Decorin Extracellular matrix CD280, uPARAP, MRC2 Epidermal growth factor Focal adhesion kinase Fibroblast growth factor 2 Fibromodulin Glycosaminoglycans Gingival fibroblast-derived 3D ECM Gingival fibroblasts Heparin-binding EGF-like growth factor Hepatocyte growth factor Interferon Interleukin Keratinocyte growth factor Latency-associated peptide Lipoprotein receptor-related protein 1 Leucine-rich repeat Latent TGF-! binding protein Lumican Matrix metalloproteinases Mesenchymal stem cells Non-muscle myosin II Pericellular matrix Platelet-derived growth factor Skin fibroblast-derived 3D ECM Skin fibroblasts Small leucine-rich proteoglycan Secreted protein acidic and rich in cysteine TGF-! activated kinase 1 Transforming growth factor-! Thrombospondin Tissue inhibitors of metalloproteinases Vascular endothelial growth factor-"  !  %#"!  ACKNOWLEDGEMENTS I would like to express my deepest gratitude to my supervisor, Dr. Lari H#kkinen, for his endless guidance and support during the course of this project. I would like to extend my sincerest appreciation to my committee members, Dr. Hannu Larjava and Dr. Calvin Roskelley, for their advice and support throughout this project. I would like to thank Dr. Guoqiao Jiang for his expertise in reverse transcriptase and real-time PCR and his help in all other aspects of work throughout my time in the lab. I also would like to thank Cristian Sperantia for his technical expertise throughout this project. In addition, I would like to acknowledge Dylan Olver and Godwin Cheung, as they helped generate important data for this project. I would like to extend my thanks to everyone else in the lab for their help and support through this project. Most of all, I wish to express my love and gratitude to my family and Diana Lam for their unconditional support throughout the duration of my studies. Grant support I would like to thank UBC CIHR Skin Research Training Centre (SRTC) Master’s Training Scholarship Award for helping fund my project for one year. The work was also supported by a grant from CIHR.  !  !  %#""!  CHAPTER ONE- REVIEW OF THE LITERATURE Scar formation as a result of wound healing in skin is associated with an increased deposition of extracellular matrix (ECM) and reduced ECM turnover by fibroblasts. Scar formation can lead to severe physical and psychosocial complications and at present, current therapies to prevent scarring are unpredictable. Interestingly, wound healing in the human oral mucosa results in scarless healing, while wound healing in the skin results in scarring. Therefore comparing fibroblast phenotype and interactions with their ECM niche in the gingiva and skin may provide novel information about the factors that regulate scar formation. This literature review concisely covers topics on the process of wound healing and scar formation, current therapies and models used to study wound healing. Moreover, the current evidence of the role of fibroblasts and ECM proteins in wound healing, as well as current literature on the differences between oral mucosa and skin fibroblasts will be reviewed. 1.1. Introduction: overview of scar formation Scar formation in skin is an unwanted condition that often forms as a result of the wound healing process. Clinically, scars range from fine lines to disfiguring hypertrophic or keloid scars that can negatively affect an individual’s quality of life (Durani et al., 2008). In general, at a cellular level, scars display excess accumulation of poorly organized collagen-rich ECM as a result of increased collagen deposition and reduced ECM turnover by fibroblasts (Cutroneo, 2003). In normal connective tissue, collagen fiber bundles are organized in a basket-weave manner (Occleston et al., 2010). However after wound healing resulting to scar formation, collagen fibers are thinner, densely packed and organized parallel to each other (Durani et al., 2008). In addition, there is an increased amount of type I collagen and other ECM components, including fibronectin, proteoglycans and matricellular proteins and growth factors present in the connective tissue (H#kkinen et al., 2012). Fine line scars typically form after two to three weeks of proliferative phase of wound healing (granulation tissue formation, angiogenesis and reepithelialization), followed by an aberrant remodeling phase that involves continued and increased accumulation of collagen in the injury site (Gurtner et al., 2008). Hypertrophic  !  &!  scars start to form four to eight weeks post-wounding, and have an extended rapid connective tissue proliferative phase for up to six months, leading to excessive accumulation of collagen-rich ECM (Gauglitz et al., 2011). In hypertrophic scars, the termination of the growth phase will eventually occur. However, in keloid scars the growth phase may persist for years with no signs of regression (Murray 1994; Gauglitz et al., 2011). Keloid scars develop slow and may take several years after wounding to develop. They can also arise spontaneously without any known injury, particularly in the mid-chest area (Gauglitz et al., 2011). This outgrowth leads to a tumor-like appearance that extends out from the margin of the original wound site and histologically the tissue shows irregular collagen organization (Gauglitz et al., 2011). This is in contrast with hypertrophic scars that, although being raised from the skin, usually do not breach the margin of the wound and histologically has parallel collagen bundle orientation (Murray, 1994; Gauglitz et al., 2011). Scar formation and its extent depend on the depth and location of the wound (Wang et al., 2008; Gauglitz et al., 2011). For instance, more severe scars are formed in deep wounds extending into the reticular layer of the dermis and in certain skin areas with greatest tension, including the chest, back and shoulders (Deitch et al., 1983). Proper wound healing may be negatively affected also by various factors, such as infection, systemic conditions, including immunocompromised state, diabetes, vascular disease, renal failure or malnourishment, and other factors, including smoking (Nauta et al., 2011). In addition, patient associated factors, such as ethnicity, sex and age, may account for the differences in healing outcomes (Gauglitz et al., 2011). The most common injuries that result in scar formation are burns, lacerations, surgeries, piercings, and even vaccinations (Gauglitz et al., 2011). These injuries often penetrate into the deeper dermal layers of the skin and as a result, stimulate a large response by fibroblasts in the wound healing process, causing excessive accumulation of ECM (Miller and Nanchahal, 2005; Gauglitz et al., 2011). Scars can lead to severe physical complications, as they reduce tensile strength of skin, affect joint mobility and cause pain and tenderness (Gauglitz et al., 2011; Penn et al.,  !  '!  2012). Even after years following injury, only 70% of tensile strength may be recovered in scar tissue compared to normal skin (Clark, 1996; Singer and Clark, 1999). Wound contraction at the end of the granulation tissue formation stage of wound healing is a beneficial process and promotes wound closure and reduces the surface area of the wound. However, persistent contraction of the wound that associates with extensive scar formation often leads into contractures, which is the shortening of scar tissue and can affect for instance joint mobility (Tredget et al., 1997). Scars can also have a psychosocial impact on individuals, as they are often present in visible areas, including the face (Occleston et al., 2010). In developed countries, approximately 100 million patients acquire scars each year as a result of surgical procedures alone (Gauglitz et al., 2011). The estimated risk of hypertrophic scars following surgery is 40% to 70% (Lewis and Sun 1990; Gauglitz et al., 2011). However, a successful anti-scarring therapy would benefit the burn patients most, as more than 70% of burns develop into hypertrophic scars (Gangemi et al., 2008). Many of these patients are children and severe scarring may affect their growth and development (Bombaro et al., 2003). It has been estimated that hypertrophic scarring treatments have a financial toll of $4 billion annually in the United States alone, while in developing countries, there is an even greater incidence of burns and other major cutaneous injuries (Aarabi et al., 2007; Occleston et al., 2010). At present, predictably efficient anti-scarring therapies are lacking. Better understanding of the biological processes that regulate wound healing leading to scar formation or scar-free tissue regeneration is needed in order to develop improved therapies to prevent scar formation and associated complications. 1.2. Overview of wound healing Wound repair is a complex process that requires systematic coordination of various cell types to promote the regeneration of damaged tissue. The wound healing process is divided into various biological events: hemostasis and inflammation, re-epithelialization, granulation tissue formation, and maturation and tissue remodeling (H#kkinen et al., 2012). These phases all overlap one another and the duration of each stage is dependent  !  (!  on the depth and size of the injury (H#kkinen et al., 2012). Based on the current understanding, biological processes involved in any stages of wound healing are important in determining the outcome of wound healing. 1.2.1. Hemostasis and inflammatory response Immediately after wounding, clot formation begins for hemostasis. Platelets and fibrin cross-link to form a fibrin clot, which serves as a temporary barrier to prevent the leakage of blood and invasion of bacteria and acts as a scaffold for wound repair, thereby mediating signaling between the ECM and the surrounding cells (Martin 1997; Gauglitz et al., 2011). In addition to its role in forming a barrier, platelets also secrete mediators that initiate wound healing, such as platelet-derived growth factor (PDGF), which attracts and activates macrophages and fibroblasts (Singer and Clark, 1999). Through chemotactic signals, neutrophils and monocytes are recruited to the site of the wound to remove foreign particles and bacteria (Martin, 1997). Neutrophils are phagocytes that engulf microbes and particles, while monocytes differentiate into macrophages at the site of the wound, and are responsible for removing contaminating bacteria through phagocytosis (Martin, 1997). Neutrophils and macrophages are critical in removing any potential pathogens, however, they are also important in initiating tissue formation phase, as these cells are a source of cytokines that activate fibroblasts, endothelial cells and keratinocytes (Martin, 1997; Singer and Clark, 1999). 1.2.2. Re-epithelialization and granulation tissue formation Re-epithelialization begins within hours after injury and is a process of keratinocytes proliferating and migrating over the damaged connective tissue of the skin or oral mucosa (Martin, 1997; Singer and Clark, 1999; Gurtner et al., 2008). The migrating keratinocytes form a border between the temporary eschar (scab) and the viable tissue (Singer and Clark, 1999). Cell proliferation and migration is induced by cytokines released during wound healing, including keratinocyte growth factor (KGF), epidermal growth factor (EGF), hepatocyte growth factor (HGF), heparin-binding EGF-like growth factor (HBEGF) and transforming growth factor (TGF)-" and -! (Werner and Grose, 2003; Miller and Nanchahal, 2005; Koivisto et al., 2012). Proliferating keratinocytes are found behind  !  )!  the leading edge of migrating cells (Singer and Clark, 1999). In addition, the keratinocytes produce and secrete collagenase to degrade the fibrin clot and the ECM of the connective tissue layer to facilitate migration (Singer and Clark, 1999). Two to four days after wounding, the fibrin and fibronectin rich provisional matrix provides a scaffold and signaling cues to direct more fibroblasts and macrophages into the injury to stimulate ECM synthesis and angiogenesis to replace the fibrin clot with new connective tissue, known as granulation tissue, by a process called fibroplasia (Singer and Clark, 1999). The newly synthesized tissue becomes highly fibrillar, being mainly composed of type III collagen (Gurtner et al., 2008). Collagen synthesis and expression is mainly regulated by pro-fibrotic TGF-! during wound repair (Singer and Clark, 1999). However, the activity of TGF-! is balanced by factors that inhibit its function. For instance, collagen expression and deposition have been shown to be reduced by an ECM protein, collagen triple helix repeat containing 1 (Cthrc1) (Pyagay et al., 2005; Leclair and Lindner, 2007; Li et al., 2011). This secreted protein is expressed by fibroblasts and its mechanism of action in reducing collagen deposition is through inhibiting SMAD2/3 activation by TGF-! (Leclair and Lindner, 2007). Furthermore, other ECM molecules, including small leucine-rich proteoglycans (SLRPs), can modulate TGF-! activity (Eckes et al., 1999). The granulation tissue is also composed of fibronectin, elastin, proteoglycans and hyaluronic acid that provide a structural scaffold for cells and modulate their functions (Kalamajski and Oldberg, 2010; Gauglitz et al., 2011). Macrophages and to a lesser extent, fibroblasts, provide an unceasing source of cytokines and growth factors to stimulate fibrous tissue formation and angiogenesis (Singer and Clark, 1999, H#kkinen et al., 2012). Newly sprouted capillaries and blood vessels are necessary to supply nutrients for cellular metabolism and support cell ingrowth at the site of the wound (Singer and Clark, 1999). The primary regulators that are essential for angiogenesis are fibroblast growth factor 2 (FGF-2/bFGF) and vascular endothelial growth factor (VEGF). FGF-2 is important for angiogenesis initiation and VEGF for granulation tissue formation phase (Tonnesen et al., 2000). Both FGF-2 and VEGF promote endothelial cell proliferation and migration to the site of the wound by  !  *!  acting as chemotactic signals or increasing vascular permeability (Gospodarowicz et al., 1987; Bao et al., 2009). Fibroblasts mediate key events in granulation tissue formation, as these cells produce and organize the ECM of the connective tissue, and modulate inflammation, reepithelialization and angiogenesis (H#kkinen et al., 2012). For instance, inflammatory cells and fibroblasts undergo cross-talk, resulting in the release of chemokines, cytokines and prostaglandins, which modulate the inflammatory response (Buckley et al., 2001; Grose and Werner, 2003; Flavell et al., 2008). Fibroblasts also undergo differentiation into myofibroblasts, which have a greater capacity in synthesizing ECM components (H#kkinen et al., 2000). Increased abundance of myofibroblasts and fibroblasts during wound healing has been associated with scarring (Cass et al., 1997; Nedelec et al., 2001). In addition, myofibroblasts contain contractile "-smooth muscle actin ("-SMA) filaments that promote the contraction of the wound (Singer and Clark, 1999). 1.2.3. Maturation and tissue remodeling During the maturation and remodeling phase, all of the processes activated during wound repair begin to cease and reorganization of the connective tissue begins (Gurtner et al., 2008). This phase begins two to three weeks following injury depending on the size and location of the wound and can last for months or even years (Gurtner et al., 2008). Cellularity of the tissue is reduced as most endothelial cells, macrophages, fibroblasts and myofibroblasts undergo apoptosis (Gurtner et al., 2008). In addition, existing type III collagen is degraded and replaced with type I collagen (Gurtner et al., 2008). The degradation of excess ECM proteins is mediated mainly by matrix metalloproteinases (MMPs), which are released by fibroblasts, keratinocytes and inflammatory cells (Gill and Parks 2008; Gurtner et al., 2008). ECM components can also be internalized by fibroblasts via endocytosis and undergo intracellular degradation by lysosomal enzymes, including cathepsin K (Holmbeck and Szabova, 2006). An important cell surface receptor, lipoprotein receptor-related protein 1 (LRP1), regulates receptor-mediated ECM endocytosis and degradation of collagens (Wienke et al., 2003; Kjoller et al., 2004; Curino et al., 2005). In addition, Endo180 (CD280, uPARAP, MRC2) is a  !  +!  transmembrane protein that fibroblasts utilize for the uptake of collagen (Wienke et al., 2003). Wound contraction that started at the end of granulation tissue formation continues during early remodeling stage. This further reduces the size of the wound and organizes the ECM molecules to withstand physical forces. Contractility is mediated by the expression of non-muscle myosin II (NMMII) and actin isoforms in fibroblasts and myofibroblasts (Bond et al., 2010). Appropriate remodeling phase is crucial to reduce excess collagen and normalize tissue structure and therefore prevent scar formation. 1.3. Factors that are associated with scar formation Wound healing requires a well-orchestrated series of biological events to promote proper tissue regeneration. However, this well-orchestrated series of events does not always result in a cosmetically favorable wound healing outcome. A variety of factors can contribute to scar formation. However, aberrations in four key biological processes in wound healing appear to associate with scar formation: (1) inflammatory response, (2) TGF-! activity, (3) cell-ECM interaction and (4) fibroblast phenotype. 1.3.1. Prolonged inflammatory response A prolonged and overly intense inflammatory phase has been associated with excess fibrosis seen in hypertrophic scars (Aarabi et al., 2009). Various growth factors and cytokines are released during inflammatory phase, which is the prerequisite for subsequent processes, such as re-epithelialization, granulation tissue and angiogenesis. Mast cells, macrophages and lymphocytes are leukocytes that play a critical role in preventing infections, but they also stimulate fibrosis at the site of the wound (Aarabi et al., 2009). These leukocytes release growth factors, including PDGF, IGF-I, TGF-! and FGF-2, which activate fibroblasts to proliferate and synthesize pro-fibrotic ECM proteins (Tredget, et al., 1997). In addition, these inflammatory cells can release and activate growth factors and cytokines, including TGF-!, VEGF and FGF to promote cell migration of fibroblasts, endothelial cells and more inflammatory cells to the site of the wound (Brown et al., 1992; Aarabi et al., 2007). Therefore, a robust and prolonged activation for granulation tissue formation by inflammatory cells, can lead to the fibrosis and scarring seen in wounds. It has been shown that oral mucosal wounds had reduced  !  ,!  number of neutrophils, macrophages and mast cells compared to skin wounds (Mak et al., 2009). As inflammation is the first stage of wound healing, its importance in releasing and activating the proper mediators is critical to pave the way for subsequent phases of wound healing. 1.3.2. Increased transforming growth factor-! (TGF-!) activity Another biological event associated with scar formation is the overproduction or prolonged activity of TGF-!. TGF-!1 is a pro-fibrotic growth factor that promotes expression of genes associated with ECM deposition, such as collagen, matricellular proteins and protease inhibitors, and suppresses expression of proteases, thus promoting a greater accumulation of ECM components (Border and Noble, 1994). In addition, TGF-! is a potent chemoattractant for fibroblasts and leukocytes, including, monocytes, macrophages and lymphocytes (O’Kane and Ferguson, 1997). TGF-! molecules are abundantly present in the ECM as a latent (inactive form) complex, which is made up of latent TGF-! binding protein (LTBP) and latency-associated peptide (LAP) (Border and Noble, 1994; Roberts, 1995). TGF-!1 is secreted by platelets, leukocytes and fibroblasts and is activated and upregulated shortly after wounding (Roberts, 1995; O’Kane and Ferguson, 1997). TGF-! can auto-induce its own expression in surrounding cells, thus increasing pro-fibrotic activity in wounds (O’Kane and Ferguson, 1997). At early gestation, fetal skin wounds heal without scar formation (Ferguson and O’Kane, 2004). This coincides with absence of inflammation and reduced expression of TGF-!1 (Krummel et al., 1988; Lin et al., 1995). However, exogenous application of TGF-!1 into the fetal wounds induced an inflammatory response and scar formation (Krummel et al., 1988; Lin et al., 1995). In humans, there are three isoforms of TGF-! that are involved in wound healing, TGF-!1, TGF-!2 and TGF-!3 (Shah et al., 1995; O’Kane and Ferguson, 1997). Of the three isoforms, TGF-!3 is known to have an anti-fibrotic role in wound healing as exogenous TGF-!3 reduces scar formation, while TGF-!1 and TGF-!2 have pro-fibrotic properties and neutralizing antibodies against TGF-!1 and TGF-!2 significantly reduced scarring in rats (Shah et al., 1994; Shah et al., 1995; Hirt-Burri et al., 2011). However, it has been suggested that the overall concentration of TGF-! may  !  -!  not entirely influence fibrogenesis, instead the ratio of pro-fibrotic TGF-!1 to antifibrotic TGF-!3 may determine whether scarring occurs (O’Kane and Ferguson, 1997; Hirt-Burri et al., 2011). In adult wounds, TGF-!3 is normally found at low levels, while TGF-!1 is found at higher levels (Ferguson and O’Kane, 2004). However, in fetal wound healing, the ratio of TGF-!1 and TGF-!3 is in the inverse relationship (Ferguson and O’Kane, 2004). In addition, TGF-! promotes angiogenesis indirectly by triggering inflammatory cells and fibroblasts to secrete pro-angiogenic factors (Wiseman et al., 1988; Berse et al., 1999; Tonnesen et al., 2000). Studies have shown that TGF-!1 expression and abundance was greater in scar tissue and fibroblasts derived from hypertrophic scars compared to normal skin tissue and fibroblasts (Ghahary et al., 1993; Zhang et al., 1995; Wang et al., 2000). Moreover, comparison of experimental wound healing in gingiva and skin of red Duroc pigs, showed that there was reduced number of cells positive for TGF-! and phosphorylated SMAD3 in scarless gingival wounds compared to scar forming skin wounds (Mak et al., 2009). Thus, it is likely that increased activity of particularly TGF-!1 during wound healing plays a role in scar formation. However, various approaches to block TGF-!1 activity and signaling during wound healing have not resulted to significant scar reduction in clinical setting, indicating that also other factors are involved. 1.3.3. Cell-ECM interaction The ECM plays an essential role in cellular behavior, such as migration, proliferation and gene expression during wound healing (Eckes et al., 1999; Tran et al., 2004; Eckes et al., 2010). In response to injury, there is an upregulation and altered distribution of ECM proteins, such as collagens, fibronectin, tenascin and proteoglycans in the wound provisional matrix (Eckes et al., 1999; H#kkinen et al., 2012). This altered and overabundance of matrix proteins would change the biochemical and mechanical environment of cells and as a result, cellular functions may be altered due to changes in cell-ECM interactions (Eckes et al., 1999). ECM modulates cellular function of fibroblasts mostly through integrin receptors. Fibroblasts recognize specific components of the ECM through the interaction of integrin  !  .!  subunits. For example, fibroblasts has been shown to bind to collagen type I through integrins "1!1, "2!1, and "11!1, (Eckes et al., 2010) whereas, fibroblasts bind to fibronectin through different integrins, "5!1 and "v!3 (Green and Yamada, 2007). Integrins do not contain enzymatic activity, instead they are linked to the actin cytoskeleton and in turn, any physical strain exerted on the integrin-ECM adhesion site will be transmitted to the cytoskeleton of the cell (Lu et al., 2012). These changes to the cytoskeleton would promote differences in gene expression and cellular behavior (Eckes et al., 1999; Lu et al., 2012). For instance, impeding "1!1 function has been shown to downregulate collagen synthesis, while blocking "2!1 had been shown to upregulate MMP-1 expression in fibroblasts (Langholz et al., 1995). Also, soluble factors, such as growth factors and cytokines, can interact with cell surface receptors and in turn, trigger a cascade of enzymatic signals to promote changes in cellular function (Werner and Grose, 2003). These signals can further interact and modulate integrin signaling and vise versa. To understand the role of the ECM on cellular function, mice with targeted deletion of various ECM proteins have been generated. For instance, knock out (KO) of matricellular proteins tenascin-C, thrombospondin and SPARC has led to irregular collagen fiber organization, altered cell-mediated re-epithelialization, angiogenesis and duration of the wound healing process (Kyriakides and Bornstein, 2003; Alford and Hankerson, 2006; Eckes et al., 2010). Also, SLRP decorin (DCN) has been shown to bind to epidermal growth factor receptor (EGFR) of fibroblasts, leading to the suppression of migration, proliferation and ECM synthesis by fibroblasts (Moscatello et al., 1998; Patel et al., 1998; Tran et al., 2004). The decreased activation in fibroblasts by DCN-EGFR interaction is believed to be beneficial during the maturation stage of wound healing (Eckes et al., 2010). In addition, the ECM primarily regulates fibroblast differentiation. Myofibroblast differentiation from fibroblasts is mediated by EDA-fibronectin, a secreted protein P311, TGF-! and platelet-derived growth factor (PDGF), as well as mechanical cues experienced by the cells from the ECM (Serini et al., 1998; Pan et al., 2002; Tomasek et al., 2002; Muro et al., 2003; Penn et al., 2012).  !  &/!  Distinct ECM proteins can also indirectly modulate cellular function by interacting with other ECM proteins, and altering their bioactivity or availability. For instance, many proteoglycans have been shown to be modulators of growth factor activity (Ruoslahi and Yamaguchi, 1991). The pro-angiogenic FGF-2 has been shown to bind to heparan sulphate side chains of proteoglycans that protects it from degradation (Ruoslahi and Yamaguchi, 1991). However, proteolysis of the proteoglycans leads to activation of FGF2 (Ruoslahi and Yamaguchi, 1991). Also, DCN and fibromodulin (FMOD), a closely related SLRPs, have been shown to interact with pro-fibrotic TGF-!1 and as a result the bioactivity of this potent pro-fibrotic growth factor was sequestered from activating cellular pathways (Yamaguchi et al., 1990; Soo et al., 2000; Stoff et al., 2007). A schematic presentation of the role of the ECM on fibroblast is found in Fig. 1. ! 1.3.4. Fibroblast phenotype In an uninjured tissue, fibroblasts are quiescent, as they are exposed to few signals from the surrounding pericellular matrix (H#kkinen et al., 2012). In this state, fibroblasts are attached to their pericellular matrix and are steadily synthesizing, degrading and organizing the ECM to promote a structurally and biochemically optimal niche for the cells (Lemons et al., 2010; De Donatis et al., 2010). However, upon tissue injury or inflammation, fibroblasts become activated for tissue repair. As the primary cell type in maintaining the connective tissue, fibroblasts deposit granulation tissue ECM and modulate inflammation, re-epithelialization, and angiogenesis (H#kkinen et al., 2012). Fibroblasts also communicate with other cells by secreting growth factors and cytokines (Eckes et al., 2010). Fibroblast phenotype may change as a result of wound repair (Clark, 1988; Sempowski et al., 1995). Isolated fibroblasts derived from both hypertrophic and keloid scars showed a distinct phenotypic difference compared to normal skin fibroblasts. For example, keloid and hypertrophic fibroblasts had an increased abundance of growth factor receptors and responded more rapidly to growth factors, such as PDGF and TGF-! (Tuan et al., 1998; Ishihara et al., 2000; Gauglitz et al., 2011). In addition, keloid and hypertrophic fibroblasts had increased expression TGF-!1 and TGF–!2 and their receptors compared  !  &&!  to normal skin fibroblasts (Babu et al., 1992; Lee et al., 1999; Wang et al., 2000; Chin et al., 2001). Also, fibroblasts derived from different anatomical areas are phenotypically heterogeneous and therefore, may have different roles in the wound healing process and have different capacity in synthesizing ECM proteins (Fries et al., 1994; Sempowski et al., 1995; H#kkinen et al., 2012). Isolated fibroblasts from the buccal mucosa and skin of the same patient, showed a different gene expression profile in the presence of growth serum (Okazaki et al., 2002; Enoch et al., 2010). Moreover, isolated fibroblasts from oropharyneal and buccal mucosa have shown to secrete different growth factors, cytokines and ECM molecules than skin fibroblasts (Knerer et al., 1999; Okazaki et al., 2002). Also, there were differences in cell proliferation, migration and adhesion, expression of extracellular receptors and response to growth factors between fibroblasts derived from the oral mucosa (oropharyneal and buccal mucosa) and skin (Knerer et al., 1999; Lee and Eun, 1999; Enoch et al., 2010; Guo et al., 2011a). Therefore, distinct fibroblast phenotypes may contribute to different wound healing outcomes. 1.4. Current therapies to eliminate scar formation Currently, there are no significant FDA-approved therapies available to prevent and eliminate scars (Nauta et al., 2011). Although there are therapies that can treat and reduce scars, these therapies are often unpredictable and unreliable. Imperfect regeneration from wound healing stems from a bidirectional feedback relationship between cell-to-cell and cell-ECM communication (Seifert et al., 2012a, Lu et al., 2012). Research in decoding the complex molecular mechanisms of wound healing and scar formation is required, which would contribute to the development of an effective anti-scarring therapy that would improve both physical and psychosocial complications that accompany scars. The leading scar therapies do not target the biochemical mechanisms of the wound healing process, but rather aim through surgical techniques to revise the scar, utilize pressure therapy or occlusive therapy, such as silicon gel sheeting, to promote a more favorable healing environment.  !  &'!  1.4.1. Surgical therapy There are various surgical options in removing scars; the most common therapies are laser treatment, and skin substitution (Aarabi et al., 2007, Davari et al., 2012). Currently, the most effective laser treatment is the vascular-specific 585-nm pulsed dye laser (PDL) (Alster et al., 2007). The mechanism of scar revision through PDL is believed to reduce TGF-!1 expression, reduce fibroblast proliferation and decrease deposition of collagen type III (Alster et al., 2007). Surgical revision involves a linear surgical excision of the scar (Gauglitz et al., 2011). Following keloid excision, there is a chance of 45-100% for recurring keloid scars without adjuvant therapy (Gauglitz et al., 2011). Mainly for burn patients, skin substitution can also be an option. It can either involve (1) autograft, (2) allografts/xenografts or (3) bioengineered skin substitutes as a replacement for hypertrophic scars (Murphy and Evans, 2012). 1.4.2. Pressure therapy Pressure therapy has been used to manage hypertrophic and keloid scars for over 40 years (Mustoe et al., 2001; Zurada et al., 2006). In order to be effective, the pressure garment must be worn all day for 6 to 12 months (Mustoe et al., 2001). This treatment is believed to accelerate the wound maturation phase, thin the dermis, decrease edema and reduce blood flow and oxygen levels (Zurada et al., 2006). In addition, it is postulated that this therapy reduces collagen formation (Macintyre and Baird, 2006). Pressure therapy has limitation in the area of treatment, as it is not effective in joints and often areas of high movement (Zurada et al., 2006). Furthermore, these garments may be uncomfortable to use (Zurada et al., 2006). 1.4.3. Silicone therapy Silicone therapy has been used for over 30 years as prophylactic treatment for preventing excessive scarring and also improving already established scars (Gold et al., 2001, Mustoe et al., 2001). Presently, silicone therapy is usually present as a silicone gel to effectively manage scarring. This therapy has been especially effective in improving signs and symptoms of hypertrophic and keloid scars (Mustoe et al., 2001). The mechanism of action is not well understood, but it is believed to involve occlusion and  !  &(!  hydration of the stratum corneum epidermal layer and promote cytokine-mediated signaling from keratinocytes to dermal fibroblasts (Mustoe et al., 2001). Despite the effectiveness of silicone-based products, it is still far from fostering a scar-free healing outcome. 1.4.4. Emerging approaches in scar-reducing therapies Understanding cellular and molecular mechanisms of the wound healing process has led to therapies in manipulating biological events that may help reduce scar development (Durani et al., 2008). Currently, there are a number of pharmaceutical products under development. However, few products have advanced through clinical trials, and these include Avotermin (Juvista), Ilodecakin, interferon injection and 5-Fluorouracil (5-FU) (Occleston et al., 2010, Gauglitz et al., 2011). Avotermin (Juvista; Renovo, UK) contains human recombinant active TGF-!3 and the proposed mechanism of action is that it alters the healing environment towards a scar-free fetal healing (Occleston et al., 2010). The relative abundance of TGF-!1 and TGF-!3 in the pericellular matrix is critical for maintaining scar formation, since a high abundance of TGF-!1 promotes fibrogenesis, while TGF-!3 reduces it (Shah et al., 1995). It is postulated that Avotermin promotes better organization of newly formed ECM in the dermis, reduces excess deposition of ECM, decreases the number and persistence of inflammatory cells and accelerates the maturation phase of wound healing (Occleston et al., 2010). This product has reached phase III clinical trials (Little et al., 2012). Ilodecakin (Renovo, UK) contains recombinant human interleukin-10 (IL-10) and its mechanism of action is to suppress the inflammatory response by modulating the recruitment and differentiation of inflammatory cells, reduce ECM deposition and decrease activity of TGF-!1 (Occleston et al., 2010). This therapy has shown significant results during phase II of drug development (Occleston et al., 2010). Interferon (IFN) therapy involves injecting IFN at the site of the scar lesion (Gauglitz et al., 2011). The proposed mechanism of action is that IFN has an anti-proliferative effect  !  &)!  and improves fibrosis by directly antagonizing histamine and TGF-! activity (Gauglitz et al., 2011). It also decreases type I and III collagen synthesis and TGF-! production (Gauglitz et al., 2011; Karagoz et al., 2012). In clinical trials, administration of IFN-"2b has reduced keloid scars by 50% (Berman and Duncan, 1989; Gauglitz et al., 2011). 5-FU is a pyrimidine analog that promotes fibroblast apoptosis by inhibiting DNA synthesis through sequestering fibroblast growth factors (de Waard et al., 1998). Recently, injection of 5-FU for 12 weeks was shown to reduce the size of the scar by at least 50% in most patients with no recurrence of scar development (Nanda and Reddy, 2004; Gauglitz et al., 2011). This modality has reached phase III trials (Tziotzios et al., 2012). Despite the efficacy of these drugs, these treatments at best only reduce excessive keloid and hypertrophic scar development, but do not eliminate the cause. To prevent scar formation, further research should focus on understanding the wound healing processes that predispose or lead to scar formation. 1.5. Scar-free healing in fetal wounds In order to better understand the molecular mechanisms of wound healing and factors that contribute to scar formation, studies involving in vivo models that undergo scar-free healing have provided insight on the cellular and molecular regulators involved. During the first one-third to one-half of gestation, wounds in fetal skin of mammalians result in rapid healing with no signs of scarring (Ferguson and O’Kane, 2004). As mentioned earlier, a prolonged inflammatory phase has been shown to contribute to scarring. In embryonic wound healing, the number of inflammatory cells present is significantly reduced and the period in which these cells are present is shorter than in adult wounds (Bullard et al., 2003; Ferguson and O’Kane, 2004). In addition, a more rapid reepithelialization was observed in fetal healing than in adult healing (Whitby et al., 1991). One of the key difference in the fetus and adult is the wound healing environment. The fetus is in a moist, sterile environment, whereas in adult skin, wounds are susceptible to  !  &*!  dehydration and infection, which creates complications for wound healing (Bullard et al., 2003; Seifert et al., 2012a). In addition, the amount, type and duration of growth factors present are distinct between adult and fetal wounds (Ferguson and O’Kane, 2004). The one major difference is the abundance of TGF-! isoforms (Bullard et al., 2003; Ferguson and O’Kane 2004). As mentioned above, TGF-!1 and TGF-!2 are associated with fibrosis and scar formation, while TGF-!3 has anti-fibrotic activity (Shah et al., 1995, Hirt-Burri et al., 2011). In embryonic scar-free healing, pro-fibrotic TGF-!1 and TGF-!2 abundance is low, while anti-fibrotic TGF-!3 abundance is greater (Bullard et al., 2003; Ferguson and O’Kane 2004). In adult wounds, it is the inverse; there are high levels of TGF-!1 and TGF-!2, while there is a low abundance of TGF-!3 present (Bullard et al., 2003; Ferguson and O’Kane 2004). In addition, fetal fibroblasts had reduced expression of SMAD3 and SMAD4, which are key intracellular mediators of TGF-! signaling, compared to postnatal fibroblasts (Colwell et al., 2007). There are also differences in ECM proteins between fetal and adult wounds. In the fetus, it was reported that there are greater amount of fibronectin and tenascin-C compared to adult wounds (Whitby et al., 1991; Bullard et al., 2003). This higher abundance of adhesion proteins may contribute to more rapid re-epithelialization phase and fibroblast migration (Bullard et al., 2003). The prompt response by fibroblasts may in turn affect collagen deposition and organization. In the fetal wounds, there is also higher proportion of collagen type III compared to collagen type I than in adult wounds (Merkel et al., 1988). In scarless fetal healing, myofibroblasts, key producers of excess ECM in scars, are also absent (Cass et al., 1997). Interestingly, increased expression of MMPs and decreased expression of MMP inhibitors in scarless fetal wounds has been reported (Dang et al., 2003). 1.6. Regeneration in amphibians and mammals Studying amphibians has been an important tool in understanding wound healing, since these species have the ability to regenerate entire missing limbs (Brockes, 1997). Regeneration is different from repair, since regeneration promotes the growth of a functional tissue or organ similar to the original form, whereas repair often results in the  !  &+!  generation of a poorly functional mass of fibrotic tissue. The mechanism of regeneration stems from progenitor cells forming a blastema at the site of the wound (Sugimoto et al., 2011). In addition, cells are able to dedifferentiate into a cell type further upstream from their lineage (Sugimoto et al., 2011). Under these conditions, these mammals are able to regenerate entire limbs. Regeneration of skin in amphibians has not been studied extensively. However, a recent report showed that axolotls are capable of perfect regeneration of full thickness excisional wounds (Seifert et al., 2012a). These animals had reduced hemostasis, decreased neutrophil abundance, and delayed deposition of ECM during regeneration (Seifert et al., 2012a). Interestingly, contrary to fetal wound healing, it was shown that there were low levels of fibronectin and tenascin-C accumulation during limb regeneration in these animals (Seifert et al., 2012a). More recently, a mammalian species, African spiny mouse found in Kenya, was shown to be able to completely regenerate skin without scarring (Seifert et al., 2012b). The skin of these mice easily tears off, and during wound healing, scab formation and hemostasis was rapid relative to other mouse species skin (Seifert et al., 2012b). Most interestingly, the ECM was deposited slowly, the organization was quite porous and the wound was predominantly composed of type III collagen in the African spiny mouse skin during regeneration (Seifert et al., 2012b). The above animal models have been excellent in deciphering the molecular factors involved in scar formation, however the physiological and biological relevance may be in question when comparing wound healing in adult humans. 1.7. Scarless healing in gingival versus scar-forming healing in skin 1.7.1. Anatomical and functional properties of the skin and gingiva As mentioned above, wound healing outcome is dependent on locations of the injury. Recent findings have shown that, remarkably, wound healing in gingiva of the oral mucosa results in significantly reduced scar formation as compared to similar skin wounds (Wong et al., 2009, Mak et al., 2009). The skin and oral mucosal gingiva have similar anatomical and functional properties. The structure of skin and gingiva both  !  &,!  consists of a keratinized epithelium composed of stratified squamous cells (H#kkinen et al., 2012). Keratinocytes are the principle cell type found in the epithelial layer (Stephens and Genever, 2007). The skin and gingiva have an underlying layer of connective tissue, called dermis and lamina propria, respectively. Both the dermis and lamina propria contain a papillary and reticular layer (Stephens and Genever, 2007). These layers of connective tissue are composed of dense fibrillar and elastic matrix containing vascular and neural networks (Stephens and Genever, 2007). In addition, the function of the skin and oral mucosa mirrors one another. Both act as a physical barrier to harmful agents, to trigger an immunological response and play a role in thermoregulation, osmoregulation and electrolyte balance (Stephens and Genever, 2007). Injury to the epidermal layer of the skin usually results in complete regeneration, however if the injury penetrates deep into the dermal layer, healing tends to lead to scarring (H#kkinen et al., 2012). On the other hand, injury to the epithelium or lamina propria of the oral mucosal gingiva results in minimal scarring (H#kkinen et al., 2012). 1.7.2. Comparing wound healing in gingiva and skin Studying wound healing systematically in relation to adult humans has been limited by the lack of a good experimental model. With the use of a murine model, it has been shown that oral mucosal wounds had a reduced inflammatory response compared to skin wounds (Sciubba et al., 1978; Szpaderska et al., 2003). This includes lowered levels of macrophages, neutrophils and T-cells, as well as IL-6 and KC (IL-8) in oral mucosal wounds (Sciubba et al., 1978; Szpaderska et al., 2003). Moreover, oral mucosal wounds undergo a more rapid re-epithelialization and they showed reduced levels of TGF-!1 relative to TGF-!3 compared to skin wounds (Sciubba et al., 1978; Szpaderska et al., 2003; Schrementi et al., 2008). However, mice, rats or rabbits are loose-skinned animals, thus different from tight skinned humans and wound healing response and scar formation in these animals differs from humans. Therefore, conclusions drawn from these studies make it difficult to draw parallels to human wounds. Since studying wound healing in human skin is ethically problematic, alternative models have been more recently developed. Like humans, pigs are tight-skinned mammals (Fang and Mustoe, 2008). Oral mucosal wounds in red Duroc pigs and humans had been compared, and the outcomes of  !  &-!  wound healing were similar, both histologically and clinically in the attached palatal gingival mucosa (Wong et al., 2009). Gingival wound healing in pigs and humans showed striking similarities, such that the timing of re-epithelialization, formation of granulation tissue and molecular composition were similar (Wong et al., 2009). These findings validate that the pig model can be used to help understand wound healing in humans. When comparing similar excisional wounds in gingiva (palatal mucosa region) and skin (back region) of the pig model, there was a significant reduction in scarring in gingiva (Wong et al., 2009; Mak et al., 2009). In addition, the inflammatory phase of wound healing in gingiva was short-lived, comprising of reduced recruitment of neutrophils, mast cells and macrophages relative to skin wounds (Mak et al., 2009). Moreover, gingival wounds had a lower number of cells positive for TGF-! and phosphorylated SMAD3 (Mak et al., 2009). In addition, during early stages of wound healing, the anti-fibrotic TGF-!3 isoform was more abundant in gingival wounds than skin wounds (Eslami et al., 2009). In contrast, skin wounds contained a significantly greater amount of type I procollagen immunopositive cells and increased deposition of fibronectin (Wong et al., 2009). In gingiva, there was prolonged accumulation of the matricellular protein, tenascin-C (Wong et al., 2009). Fibroblasts play a critical role in producing ECM proteins and regulating of inflammation (H#kkinen et al., 2012). Therefore, differences in inflammatory response, ECM composition and phosphorylated SMAD3 via TGF-! activation may be associated with the cell phenotype of fibroblasts in gingiva and skin. Intrinsic environmental differences may also attribute to the differences seen in wound healing outcomes in the oral mucosa and skin. For example, the oral mucosa is covered with saliva that provides moisture and contains a unique molecular composition, including EGF, TGF-! and FGF that may promote wound healing (H#kkinen et al., 2000). In contrast, wound healing in skin occurs in a drier environment, which may be less optimal for wound healing. In addition, the microflora between the oral mucosa and skin are dissimilar, which may contribute to differences in wound healing (H#kkinen et al., 2000).  !  &.!  1.8. Fibroblasts 1.8.1. Properties of fibroblasts Fibroblasts are the most abundant cell type present in connective tissue, where they establish, maintain and remodel the tissue (Gabbiani et al., 1971). ECM provides a structural framework for tissues and fibroblasts and other cells, such as epithelial, inflammatory and vascular to firmly attach to the matrix (Sorrell and Caplan, 2009). In addition, fibroblasts release growth factors and cytokines into the cellular microenvironment to signal and communicate with fibroblasts and other cells (Sorrell and Caplan, 2009). Fibroblasts can also form cell-to-cell contacts, via gap and adherens junctions, with other cells and fibroblasts for intercellular communication (H#kkinen et al., 2012). These connections allow for a synchronized response to environmental cues, such as fibroblast activation, proliferation and formation of granulation tissue (H#kkinen et al., 2012). By definition, fibroblasts are characterized as elongated, spindle-shaped cells that have the ability to synthesize, deposit and remodel the ECM within connective tissue (Sorrell and Caplan, 2009). They can also undergo differentiation and senescence in vivo and in vitro (Sorrell and Caplan, 2009). Despite similarities in elongated shape and the ability to synthesize ECM components, fibroblasts isolated from different anatomical sites, have been shown to be functionally and morphologically distinct (Fries et al., 1994; Sempowski et al., 1995). These functional differences include proliferation rate, cell migration and gene expression profile (Fries et al., 1994; Chang et al., 2002; Sorrell and Caplan, 2009). Moreover, cytokine and ECM components produced are significantly different in fibroblasts of a different origin (Sorrell and Caplan, 2004; Sorrell and Caplan, 2009). Therefore, comparing gingival-derived (GFBL) and skin-derived fibroblasts (SFBL) may help determine phenotypic characteristics that may contribute to scarless wound healing in gingiva. 1.8.2. Fibroblast phenotype in oral mucosa and skin Fibroblasts in the oral mucosa originate from the neural crest, while most fibroblasts from the trunk dermis originate from the mesoderm (Buckley et al., 2001; Breau et al., 2008!  !  '/!  0123"4! !"# $%&'! '/&&). This may contribute to differences in cell phenotype and wound healing response. The known phenotypic differences between oral mucosal and skin fibroblasts are summarized in Table 1. Furthermore, fibroblasts from different regions of the oral mucosa are phenotypically different. This may depend on the different function of gingival tissue compared to other parts of the oral mucosa, as the gingiva surrounds and supports the teeth and alveolar bone (Ramfjord, 1979). A microarray analysis comparing GFBL and SFBL grown in 2D cultures with the presence of serum has shown that 278 out of 5284 genes were expressed significantly different (Ebisawa et al., 2011). Interestingly, the majority of the genes that were different, were ECM-associated, cytokine activity-related and growth factor-related genes (Ebisawa et al., 2011). Therefore, perhaps there are pre-programmed genes that are found to be different in GFBL and SFBL, which may explain why there are differences in wound healing outcomes in gingival tissue and skin. Expression of adhesion molecules has also been studied and it was shown that there was a distinct expression profile of integrin receptors and focal adhesion kinase (FAK) in GFBL and SFBL (Palaiologou et al., 2001; Guo et al., 2011a). In addition, it has been demonstrated that GFBL had a greater proliferation rate than SFBL (Schor et al., 1996). In terms of ECM synthesis, ECM production differs from GFBL and SFBL. For instance, there were larger dermatan sulphate proteoglycans and greater levels of hyaluronan in GFBL compared to SFBL cultures (Larjava et al., 1988; Schor et al., 1996). Thus, differences in the role of fibroblasts in gingival and skin wound healing response may be in part due to differences in interaction of the cells with ECM and in cell proliferation. It is worth mentioning that GFBL are more phenotypically similar to scarless-associated fetal fibroblasts than adult skin fibroblasts. For instance, proliferation rate was greater and similar in fetal and gingival fibroblasts compared to SFBL (Schor et al., 1996). In addition, the synthesis of hyaluronan, an ECM molecule that is associated with scarless healing in fetal wounds, was similar between gingival and fetal fibroblasts, while SFBL showed significantly lower levels of hyaluronan (Schor et al., 1996). These findings  !  '&!  suggest that differences in wound healing outcomes may be attributed to the role of fibroblasts during this complex biological process. 1.9. Extracellular matrix (ECM) 1.9.1 Overview of the ECM Fibroblasts synthesize, organize and remodel the ECM to yield an optimal niche that allows for proper cell and tissue function. Both the biochemical composition and structural rigidity regulate the functions of cells embodied in the ECM. There is immense structural and functional heterogeneity in the ECM of different tissues (Hay, 1991), therefore characterizing cell phenotype in GFBL and SFBL in terms of ECM production may help determine factors that promote differences in the wound healing outcomes. The ECM is composed of numerous proteins that play an important role in maintaining a fully functional tissue. The ECM components can be categorized into various groups: fibrillar proteins, glycoproteins, matricellular proteins, proteoglycans and growth factors. Cell-ECM interactions are crucial for the regulation of biological events. In general, fibrillar ECM proteins have a role in providing a structural scaffold and promoting cell adhesion (Widgerow, 2011). In addition, these ECM molecules can interact and regulate cell function through acting as ligands for receptor-mediated signaling and provide mechanical stimulus for mechanotransduction (Yamada et al., 2003; Ingber, 2006). Glycoproteins including fibronectin and matricellular proteins are abundant throughout the ECM and regulate cell adhesion and store growth factors (Klingberg et al., 2013). Matricellular proteins modulate cell-ECM interactions through binding of ECM proteins, cell surface receptors and molecules such as proteases and growth factors (Bornstein and Sage, 2002; Eckes et al., 2010). Many matricellular proteins are highly upregulated during response to injury (Bornstein and Sage, 2002). Various members of this group of proteins are synthesized by fibroblasts and have been implicated in having important roles during wound healing. KO of matricellular protein tenascin-C, thrombospondin, SPARC and osteopontin in mice has led to altered wound healing, characterized by  !  ''!  abnormal ECM organization, angiogenesis and immune response (Kyriakides and Bornstein, 2003; Alford and Hankerson, 2006; Eckes et al., 2010). Proteoglycans have a wide range of functions, including regulating cell proliferation, adhesion and migration (Wight et al., 1992). In addition, these ECM proteins mediate the molecular interaction of other ECM molecules (for example growth factors) with cells (Wight et al., 1992). A specific structurally related group of proteoglycans, SLRPs, regulate collagen fibril formation, which is important during wound healing and targeted deletion of certain SLRPs promotes development of certain pathological conditions, including fibrosis (Kalamajski and Oldberg, 2010). As mentioned above, growth factors play a pivotal role in regulating the wound healing process. Growth factors are often immobilized in the ECM and can be therefore considered an integral component of the ECM (Macri et al., 2007). The majority of the ECM components listed above are synthesized and deposited by fibroblasts and are essential for proper functioning of the connective tissue and have a crucial role in the wound healing process. 1.9.2. Pericellular matrix (PCM) The molecules within the ECM that are in close proximity to the cell surface, are known as the pericellular matrix (PCM) (Macri et al., 2007). The ECM components, including the ones listed above, interact with surrounding cells and can thus be considered part of the PCM. The PCM is a highly dynamic and versatile environment, as there is continuous PCM turnover and remodeling, that in turn affects cellular function (Lu et al., 2012). This feedback relationship between the cells and PCM allows for cells to rapidly adapt to their surroundings, and abnormalities to the PCM may lead to dysregulation of cellular function, such as cancer development or fibrosis (Eckes et al., 1999; Lu et al., 2012). In the case of fibrosis, there is an increased deposition of ECM, which alters the cellular microenvironment, and in turn, modifies cell-ECM interaction (Eckes et al., 1999). Thus, the role of the ECM, specifically the PCM, is to provide a structure for cells to adhere to, filter and carry proteins towards and away from cells, act as a reservoir for growth factors  !  '(!  and cytokines and modulate cellular activity (Macri et al., 2007). Consequently, the PCM is involved with regulating cell adhesion, migration, differentiation, proliferation, survival and acts as a structural scaffold for tissue and cells (Chen, 2010). Conclusions made on cellular function in a given tissue, must take into consideration the influence of the PCM that surrounds the cells. Therefore, understanding the role of the PCM components involved in wound healing and how it may modulate cellular behavior, may help contribute to the development of an effective therapy that blocks fibrosis and scarring that occur as a result of wound repair. 1.9.3. Fibrillar ECM protein The collagen family is the most abundant type of protein found in the ECM, as it accounts for 30% of all proteins in mammals and acts as the primary structural component of tissue (Nimni, 1983). The collagen family contains at least 28 different types, derived from more than 42 genes (Eckes et al., 2010). Most collagen is assembled in a triple helix formation that is stabilized through hydrogen bonds (Nimni, 1983). The hallmark of scarring as a result of wound healing is excess deposition of disorganized collagen type I (Cutroneo, 2003). Therefore, comparing collagen type I synthesis and deposition in fibroblasts derived from gingiva and skin may help uncover whether there is a distinct phenotypic difference. Collagen not only acts as a structural site for cellular attachment in tissue, but also enables ECM components to be anchored within the pericellular matrix, thus fostering their biological function in a centralized location. For example, the major cellular adhesion molecule, fibronectin, is able to bind to collagen, and as a result, this interaction allows cells to anchor to the fibrillar ECM scaffold (Gold and Pearlstein, 1980). In addition, SLRPs interact with collagen via their leucine-rich repeat (LRR) domain, which in turn allow for sequestering of soluble ECM components, including TGF-! and FGF-2 that bind to SLRPs (McCawley and Matrisian, 2001; Kalamajski and Oldberg, 2010). Elastic fibers in connective tissue are made up of elastin and fibrillin-rich microfibrils (Aumailley and Gayraud, 1998; Kielty et al., 2002;). The major components, elastin and fibrillin-1, are synthesized and secreted by fibroblasts (Kielty et al., 2002). Elastic fiber  !  ')!  formation in the ECM may be dependent on elastic fiber-binding proteins, emilins 1-3, which are also secreted by fibroblasts (Zanetti et al., 2004). Elastic fibers are found within the dermal layer of the skin, but are most abundant deep in the dermis (Amadeu et al., 2004). In gingiva, elastic fibers are only found in the deeper layers of the gingival connective tissue (Chavrier, 1990). The elastic fiber complex allows for tissue flexibility and elasticity (Kielty et al., 2002) and upon wounding, the elastic fiber system is often compromised (Bhangoo et al., 1976; Amadeu et al., 2004). It is believed that elastin production is reduced or absent during wound healing, which accounts for the reduced elasticity and breaking strength of the scar (Amadeu et al., 2004). It has been shown that fibrillin-1 protein density was lower in scars than in normal human skin (Amadeu et al., 2004). However, elastin was only lower in abundance in the superficial part of the dermis of scar tissue, while deep in the dermis, elastin was surprisingly greater than normal skin (Amadeu et al., 2004). 1.9.4. Glycoproteins and matricellular proteins Fibronectin is the main cell adhesion molecule secreted into the extracellular space of wounds (Widgerow, 2011). This glycoprotein is important during early stages of healing, as the provisional matrix is highly abundant with fibronectin, to help platelets in releasing growth factors and cytokines at the site of the wound and to regulate cell adhesion, migration and function (Sempowski et al., 1995; Widgerow, 2011). Fibronectin has an equally important role during granulation tissue formation. This glycoprotein is believed to be essential in promoting the formation of a collagen fibrillar network (Widgerow, 2011). Fibronectin also binds to growth factors such as LTBP, VEGF and FGF-2 (Bossard et al., 2004; Fontana et al., 2005; Wijelath et al., 2006; Klingberg et al., 2013). During clot formation, plasma fibronectin is found at the injured site to aid in cell migration (Martin, 1997). In addition, fibroblasts synthesize two major isoforms of fibronectin, EDA- and EDB-fibronectins during wound healing (Zardi et al., 1987). EDB-fibronectin has an extra domain (domain B), while EDA-fibronectin lacks the Bdomain but contains the A-domain (H#kkinen et al., 2012). Plasma fibronectin lacks both the A- and B-domain (Grinnell et al., 1981; Martin, 1997). EDB-fibronectin regulates fibroblast proliferation and angiogenesis, while EDA-fibronectin plays a role in  !  '*!  mediating cell adhesion to the ECM and supporting fibroblast migration and myofibroblast differentiation (Serini et al., 1998; Castellani et al., 1994; H#kkinen et al., 2012). EDA-fibronectin reaches a peak expression at day 7-14 post-wounding in skin and oral mucosa (Wong et al., 2009). At 35 and 49 day post-wounding, skin wounds had significantly greater abundance of EDA-fibronectin compared to gingival wounds in red Duroc pigs (Wong et al., 2009). EDB-fibronectin is absent in the early stages of wound healing and reaches peak abundance at day 7 post-wounding in oral mucosal wounds (H#kkinen et al., 2012). Interestingly, there was a high level of EDB-fibronectin found in scarless prenatal tissue compared to adult normal tissue (Zardi et al., 1987; Castellani et al., 1994). Matricellular proteins do not serve as a structural role, instead these proteins function as modulators of cell-ECM communication (Kyriakides and Bornstein, 2003). As a result, these proteins may significantly alter the progression of wound healing and scar formation (Kyriakides and Bornstein, 2003). However, very little is known about their expression in scar forming compared to scarless wound healing, as well as in gingival and skin fibroblasts. Table 2 summarizes of the key matricellular proteins that may have a role in the wound healing process. 1.9.5. Small leucine-rich proteoglycans (SLRPs) Many of the ECM molecules involved in wound healing can be sequestered and modulated by SLRPs. These proteoglycans are important components of the PCM, both structurally and biochemically (Kalamajski and Oldberg, 2010). SLRPs are comprised of a protein core that has O- or N-linked oligosaccharides and one or more glycosaminoglycan (GAG) side chain(s) covalently attached (Kalamajski and Oldberg, 2010). Within the protein core, there are several LRR domains, which are responsible for most of the functional activity (McEwan et al., 2006). The LRR motifs consist of tandem leucine-rich repeats and other small hydrophobic residues (McEwan et al., 2006). In addition, SLRPs help retain hydration in the PCM (Eckes et al., 2010). The GAG side chains of SLRPs are comprised of chondroitin sulphate, dermatan sulphate or keratan sulphate (McEwan et al., 2006).  !  '+!  Changes in SLRPs accumulation during different phases of wound healing, suggest that SLRPs may play a role in tissue repair, such as influencing TGF-! signaling or collagen fibril organization (Honardoust et al., 2008). Targeted deletion of SLRPs has led to abnormal collagen fibrillogenesis in several tissues (Ameye and Young, 2002). There are 15 distinct SLRP molecules and through characterization of the major proteoglycans secreted by human oral mucosal fibroblasts, the closely related SLRPs, decorin (DCN), biglycan (BGN), fibromodulin (FMOD) and lumican (LUM) were found to associate with distinct wound healing cells (Alimohamad et al., 2005; Honardoust et al., 2008) and interact with type I collagen to coordinate the regulation of fibrillogenesis and collagen fibril organization in the oral mucosa (Matheson et al., 2005). The primary SLRP species that have been studied the most in relation to wound healing and connective tissue integrity belong in either class I or class II of the SLRP family (Kalamajski and Oldberg, 2010) (Fig. 2). Studies have shown that DCN can bind to TGF-! (Yamaguchi et al., 1990; Ständer et al., 1999; Eckes et al., 1999). The binding of DCN to TGF-! results in preventing TGF-! from binding to its receptors, in turn inhibiting fibrogenesis and mitogenesis (Ständer et al., 1999). In addition, in vitro studies have shown that DCN, BGN and FMOD have the ability to bind to the active form of TGF-!1 (Hildebrand et al., 1994). Moreover, FMOD was able to minimally bind to the latent recombinant TGF-!1 precursor (Hindebrand et al., 1994). Studies involving the overexpression of DCN and FMOD reduced conjunctival and dermal scarring in rabbits, respectively (Grisanti et al., 2005; Stoff et al., 2007). Moreover, overexpression of DCN in experimental lung and peritoneal fibrosis and glomerulonephritis in rodents decreased TGF-! activity and reduced fibrosis (Kolb et al. 2001; Margetts et al., 2002; Shimizukawa et al., 2003; Huijun et al. 2005). SLRPs can also modulate cell adhesion. DCN and BGN can bind to fibronectin, inhibiting fibroblast adhesion to this molecule (Winnemöller et al., 1991; Bidanset et al., 1992). Interestingly, certain SLRPs can modulate expression of cytokines, MMPs and TIMPs. For example, upregulation of DCN in fibroblasts decreased levels of MMP-1 and MMP-3, while MMP-2 and TIMP-2 levels increased (Al Haj Zen et al., 2003; Stoff et al.,  !  ',!  2007). In addition, IL-1! was reduced and IL-4 and tumor necrosis factor-" (TNF-") was increased via DCN overexpression (Al Haj Zen et al., 2003). Therefore, these findings suggest that DCN or other closely related SLRPs could significantly affect the wound healing process. 1.9.6. Growth factors Growth factors are essential signaling proteins that regulate cellular behavior in various biological processes, including wound healing. Growth factor signaling on anchoragedependent cells, including fibroblasts, depends on cellular interactions with the ECM as integrin-mediated signals from the ECM modulate growth factor receptor signaling and vice versa (Macri et al., 2007). Growth factors are synthesized and secreted by most cell types involved in wound healing, including platelets, leukocytes, fibroblasts, endothelial cells and epithelial cells (Bennett and Schultz, 1993; Werner and Grose, 2003). As essential components in regulating cellular behavior in surrounding cells, comparing the expression and synthesis of key growth factors and cytokines in GFBL and SFBL, may help determine factors that promote differences in wound healing outcomes in skin and gingival tissue. These soluble proteins control growth, differentiation and metabolism of cells, which may lead to chemotactic signaling, proliferation, angiogenesis, matrix synthesis and enzyme production in wound healing (Steed, 1997). A list of key growth factors involved in wound healing is summarized in Table 3. 1.9.7. Matrix metalloproteinases and their inhibitors Matrix metalloproteinase (MMPs) are a group of enzymes responsible for the degradation of ECM and basement membrane components (Matrisian, 1990; Gill and Parks, 2008). MMPs are expressed by fibroblasts and have a crucial role in remodeling of the ECM. Therefore, comparing the MMP expression and synthesis in GFBL and SFBL, as well as the cell-ECM interaction in the regulation of MMPs may help uncover factors that contribute to scarring. This family of enzymes requires a zinc ion in their active site in order to have enzymatic activity (Malemud, 2006). In humans, the MMP family contains 23 members (Palavalli, 2009) and these enzymes are categorized into four distinct subsets:  !  collagenases,  gelatinases,  stromelysins,  '-!  and  membrane-type  (MT)  metalloproteinases (Armstrong and Jude, 2002). Characteristically, each member of MMPs specifically degrades at least one component of the ECM and is inhibited by tissue inhibitors of metalloproteinases (TIMPs) (Woessner, 1991). Gene expression of MMPs is regulated by growth factors, cytokines, physical stress, cellular transformation and chemical agents (e.g. hormones, drugs) (Nagase, 1999). In addition, cell-matrix and cellcell interactions regulate expression of MMPs (Nagase, 1999). These proteinases are secreted into the ECM in a pro-MMP inactive form, known as a zymogen (Woessner, 1991). Activation of latent forms of the enzyme is generally facilitated by plasminogen activator, furin convertase protein or other MMPs (McCawley and Matrisian, 2001, Malemud, 2006). MMPs play a fundamental role in connective tissue maintenance and remodeling. MMPs degrade components of the ECM to foster cell migration, deposition of new ECM and tissue development (Armstrong and Jude, 2002). In addition, MMPs play a central role in complex biological processes, such as wound repair. For example, fibroblasts utilize MT-MMPs to cleave impeding collagen and other PCM components, without disrupting the integrity of the matrix, to migrate to the wound site (H#kkinen et al., 2012). In addition, cell migration of epidermal cells is mediated by MMP-1, -2, -3 and -13, as they degrade ECM components to promote re-epithelialization (Singer and Clark, 1999). Secretion of MMPs can regulate chemokine and growth factor activity through proteolytic cleavage (Gill and Park, 2008). Remodeling of the ECM particularly occurs during the closing stages of wound healing, as excess collagen and other ECM components are degraded to allow normalization of tissue composition (H#kkinen et al., 2012). MMP expression is also regulated by growth factors, through MAP kinase pathway (Benbow and Brinckerhoff, 1997; Nagase et al., 1999; Westermarck and Kahari, 1999). The target ECM proteins, and cytokine, growth factor and MMP activation by MMPs are summarized in Table 4. The primary regulators of MMP activity are TIMPs. TIMP molecules are able to interact with the active or alternative site of MMPs and inactivate these proteases (Nagase and Woessner, 1999; Malemud, 2006). The TIMP family consists of four members (TIMP-1,  !  '.!  -2, -3, -4) (Nagase and Woessner, 1999). These MMP inhibitors have a role in cell invasion, tumorigenesis, metastasis and angiogenesis (Nagase and Woessner, 1999) and are synthesized and secreted by fibroblasts (Woessner, 1991). TIMP expression is also regulated by growth factors (Matrisian, 1990). For example, TGF-! upregulates TIMP expression, thus inhibits degradation of the ECM (Benbow and Brinckerhoff, 1997). The balance of synthesis and degradation of the ECM is largely dependent on MMP and TIMP abundance. Furthermore, the spatiotemporal abundance of MMPs and TIMPs during wound healing may play an important part on the accumulation of ECM components and whether scar formation takes place (H#kkinen et al., 2012). Interestingly, it was shown that scarless fetal skin wound healing has a greater abundance of MMPs relative to TIMPs compared to scar-forming adult skin (Dang et al., 2003; H#kkinen et al., 2012). The mechanisms of the wound healing process are not universal across all species and among all anatomical regions of the body. This is at least partly attributed to the heterogeneous populations that exist between fibroblasts, as fibroblasts are one of the key cell-types to orchestrate the wound healing process. We hypothesize that phenotypic differences between GFBL and SFBL may explain, at least partially, the scarless and scar forming healing in gingiva and skin, respectively. In addition, cellular functions are dependent on their interactions with the ECM. Therefore, the qualitative and quantitative differences in ECM would affect the biological events during wound healing. 1.10. Cell-derived three-dimensional in vivo-like cell culture model Current knowledge about GFBL and SFBL phenotype is largely based on studies using the artificial 2D cell culture environment. To study fibroblast phenotype in an in vivo-like condition, cell-derived three-dimensional (3D) cell culture models had been recently developed (Cukierman et al., 2001; Chen et al., 2007; Soucy and Romer, 2009). With the use of the 3D cell culture model, the cell-ECM interaction can be studied in a more in vivo-like situation to determine the role of the ECM on cell phenotype.  !  (/!  Cellular function in a conventional two-dimensional monolayer environment has been a useful model in deciphering biological processes, such as signaling pathways, however its in vivo physiological relevance may be in question (Green and Yamada, 2007). In addition, studies have utilized artificial 3D collagen scaffolds to mimic the in vivo-like 3D matrix. Although these 3D collagen matrices have helped understand the role of the three-dimensionality and mechanical tension from the ECM on cellular function, a 3D environment in vivo is more complex than the collagen and growth factors (from the media) present in these 3D collagen systems (Cukierman et al., 2002; Green and Yamada, 2007). Using cell-derived 3D ECM will yield a microenvironment where cells are embodied in their own ECM that may more closely mimic ECM composition and structure found in vivo (Green and Yamada, 2007). This allows a novel way to study cell phenotype and the influence of the ECM on cellular function in relation to a wound healing setting. 1.10.1. Generation and characterization of the cell-derived 3D cell cultures To generate in vivo-like 3D cultures, fibroblasts are cultured at a high density in a standard medium DMEM containing fetal bovine serum and ascorbic acid (Beacham et al., 2007). Ascorbic acid is used to promote synthesis and deposition of ECM molecules (Beacham et al., 2007). To study the composition and function of the 3D ECM, fibroblasts are extracted using an alkali detergent buffer (Beacham et al., 2007). The 3D ECM is highly fibrillar and its molecular composition resembles the connective tissue found in vivo, including many of the ECM proteins described above, for example, collagen type I, III, IV and V and various glycoproteins including fibronectin and tenascin-C and SLRPs (Cukierman et al., 2001; Chen et al., 2007; Soucy and Romer 2009; Lai et al., 2010). Further studies have compared cell adhesion, morphology, proliferation, signaling and differentiation between 2D cell culture and artificial 3D scaffolds with the cell-derived 3D ECM (Cukierman et al., 2001; Chen et al., 2007; Soucy et al., 2009; Lai et al., 2010). When cells were isolated from tissue and cultured on a 2D substrate, these cells became flattened, divide aberrantly and lost their in vivo phenotype (von der Mark et al., 1977;  !  (&!  Baker and Chen, 2012). Fibroblasts exposed to 3D ECM had a six-fold greater rate of adhesion compared to fibronectin, collagen, laminin or artificial 3D collagen gels (Cukierman et al., 2001). Not only was the adhesion rate significantly faster, but there was also a marked difference in cellular morphology. Fibroblasts on cell-derived matrix acquired the spindle-like morphology, characteristic to cells in vivo (Cukierman et al., 2001). In contrast, fibroblasts on a 2D substrate are flat and interact with the substrate only at the basal side of the cell, forming an apical-basal polarity (Cukierman et al., 2001; Baker and Chen, 2012). On the other hand, on a 3D substrate, fibroblasts form adhesion sites at all three dimensions of the cell, which is more representative to an in vivo condition (Cukierman et al., 2001; Baker and Chen, 2012). Moreover, cells seeded on a 3D ECM displayed fewer branched terminals and were more asymmetrical in structure as compared to 2D cultures (Cukierman et al., 2001). Adhesion molecule composition was also significantly different in fibroblasts cultured on 3D ECM and 2D substrates. Cell– ECM integrin-mediated adhesions facilitate physiological responses, including regulating cell proliferation, migration, differentiation, survival and ECM organization and remodeling (Cukierman et al., 2002). Surprisingly, fibroblasts on 3D ECM s depended on "5!1 integrin, and fibroblasts on 2D substrates depended on both "5!1 and "V!3 integrins (Cukierman et al., 2001). Also, fibroblasts cultured on 2D substrates had elevated levels of phosphorylation of FAK compared to fibroblasts cultured on 3D ECM (Cukierman et al., 2001). In addition, proliferation rate was twice as high when fibroblasts were seeded on a 3D ECM compared to a fibronectin coated 2D surface (Cukierman et al., 2001). Migration rate of fibroblasts on 3D ECM was enhanced by approximately 50%. In addition, the 3D ECM increased the directionality of migration (Cukierman et al., 2001). Cellular signaling between the 2D and 3D cell cultures is significantly distinct. The 3D ECM that is deposited by fibroblasts is less rigid and more pliable than the 2D cultures (Cukierman et al., 2001; Green and Yamada, 2007). This leads to differences in mechanotransduction signaling responses (Green and Yamada, 2007). For example, stiffness of 3D Hydrogel matrix regulated differentiation of mesenchymal stem cells (MSC), as intermediate stiffness promoted osteogenic differentiation and softer stiffness stimulated adipogenic lineage commitment (Huebsch et al., 2010). In addition to  !  ('!  mechanical signals, fibroblasts cultured in 2D and 3D cultures are exposed to different ECM biochemical signals that can modulate cellular behavior (Soucy et al., 2009). For instance, when normal fibroblasts were cultured on 3D ECM deposited by fibroblasts derived from tumors, the ECM induced morphological and gene expression changes relative to normal fibroblasts cultured on normal fibroblast-derived 3D ECM (Amatangelo et al., 2005). Moreover, fibroblasts and MSC retained their in vivo-like phenotype longer when seeded on 3D ECM as compared to cells cultured on standard 2D cell culture plastic (Chen et al., 2007; Quiros et al., 2008; Lai et al., 2010). For example, when MSC were seeded on bone marrow-derived 3D ECM, more cells were able to differentiate into adipocytes or osteoblasts when stimulated with differentiation medium (Chen et al., 2007; Lai et al., 2010). In addition, when endothelial cells were seeded on fibroblast-derived 3D ECM, these cells remodeled the ECM and underwent tubulogenesis, which did not occur in standard 2D cultures (Soucy and Romer, 2009). Previous study has shown that dermal and fetal fibroblast in 3D cell cultures had similiarties in fibroblast density and biochemical abundance and organization with in vivo neonatal and fetal dermis, respectively (Pouyani et al., 2012). In summary, studying the phenotypic characteristics of GFBL and SFBL in the 3D cell culture model would yield information that may more closely reflect the in vivo situation.  !  ((!  CHAPTER TWO – AIMS OF THE STUDY Scars form as a result of increased deposition of ECM and reduced ECM turnover by fibroblasts. Since wound healing in the human oral mucosa results in scarless healing, while wound healing in the skin results in scarring, comparing GFBL and SFBL phenotype and interactions with their ECM niche may provide novel information about the factors that regulate scar formation. The aim of this study is to compare GFBL and SFBL phenotype in an in vivo-like 3D cell culture that mimics ECM composition and structure found in tissues. 2.1. General hypothesis We hypothesize that differences in GFBL and SFBL phenotype may contribute to the scarless and scar forming healing in gingiva and skin, respectively. 2.2. Aim (I) To characterize GFBL and SFBL phenotype through the analyses of cell morphology and proliferation, expression of key wound healing associated genes and ECM protein abundance in in vivo-like 3D cell cultures. 2.3. Hypothesis (I) We hypothesize that GFBL and SFBL have a distinct cell phenotype and that GFBL preferentially express molecules related to matrix turnover while SFBL express genes involved in matrix deposition. 2.4. Aim (II) To determine the role of the ECM on cell phenotype, by analyzing the effects of the 3D ECM on cell adhesion, gene expression and ECM-cell signaling in GFBL and SFBL. 2.5. Hypothesis (II) We hypothesize that the 3D ECM generated by the cells determines in part the cell phenotype of GFBL and SFBL.  !  ()!  CHAPTER THREE - MATERIAL AND METHODS 3.1. Cells Five cell lines of dermal fibroblasts (SFBL) from clinically healthy human breast skin (NHDF; PromoCell, Heidelberg, Germany) and gingival fibroblasts (GFBL) from clinically healthy attached gingiva from five healthy human subjects isolated as previously described (Häkkinen et al., 1994). Cells (Table 5) were maintained at 37oC and 5% CO2 in Dulbecco’s Modified Eagle’s medium (DMEM) (Gibco Life Technologies, Inc., Grand Island, NY, USA), supplemented with 10% fetal bovine serum (FBS) (Gibco Life Technologies) and 1% antibiotic/antimycotic (Gibco Life technologies). Experiments were performed at passages 5 to 12. Cells were routinely seeded for experiments when they reached 90-100% confluency. 3.2. In vivo-like three-dimensional (3D) cell culture GFBL and SFBL were seeded at high density (5x104 cell/cm2) on plastic culture plates in DMEM, supplemented with 10% FBS, 1% antibiotic/antimycotic and 50 $g/ml of ascorbic acid (Beachem et al., 2007). Cells were allowed to grow and deposit a 3D ECM up to 14 days, with media change three times per week. 3.3. Generation of cell-free fibroblast-derived 3D ECM Fibroblasts were grown in 3D culture for up to 14 days, washed once with phosphatebuffered saline (PBS) and incubated with a cell extraction buffer composed of 0.5% Triton X-100 and 20 mM NH4OH in PBS (pH=8.0) at 37oC for 5-10 min. PBS was then added to dilute the extraction buffer 2-fold, and cultures were incubated at 4oC overnight. 3D ECM was washed with PBS and treated with 10U/ml of DNase (Roche Diagnostics; Indianapolis, IN, USA) at 37oC for 30 min. 3D ECM was then gently washed 3 times with PBS and stored at 4oC until used. 3.4. Fibroblast reseeding on fibroblast-derived 3D ECM and collagen substrates As a control substrate for 3D ECM, tissue culture surfaces were coated with 100$g/ml collagen type I (PureCol® Advanced BioMatrix; Poway, CA, USA) in 10mM HCl at  !  (*!  room temperature for 1 h, followed by PBS washing. Non-specific binding sites were blocked by incubating with 1% bovine serum albumin (BSA) in PBS at room temperature for 30 min. For the reseeding experiments, fibroblasts were seeded on the cell-free 3D ECM and collagen substrates at a density of 4x104 cell/cm2 in DMEM for 24 h and analyzed for cell adhesion, gene expression and activation of signaling pathways. 3.5. Measurement of fibroblast proliferation Proliferation of GFBL and SFBL was studied in both low (starting density 1.6x104 cell/cm2) and high (4x104 cells/cm2) density conditions. For the low density condition, cell numbers were measured using the tetrazolium-based colorimetric assay (MTT assay; Promega, Madison, WI). To this end, three parallel GFBL and SFBL lines were seeded on plastic culture plates in 6 replicates and experiment was repeated 3 times and cell numbers were recorded at day 1, 3, 6 and 8 post-seeding. To assess cell numbers at high density conditions, cells were seeded and maintained as described above for generation of 3D cell cultures for 3, 7, 10 and 14 days. Total RNA was extracted using NucleoSpin® RNA  II  kit  (Macherey-Nagel,  Bethlehem,  PA,  USA)  and  quantitated  by  spectrophotometry (GeneQuant; LKB Biochrom, Ltd, Cambridge, UK) as a measurement of cell numbers. 3.6. Quantification of total protein deposited in the cell-derived 3D ECM At day 3, 7, 10, and 14 post-seeding, cell-free 3D ECM was solubilized in SDS sample buffer (2% SDS, 0.005% bromophenol blue, 10% glycerol in 50 mM Tris-HCl; pH=6.8) and collected using a rubber policeman. Total protein content was measured using BioRad DC protein assay reagent and a spectrophotometer at 655nm (Bio-Rad; model 3550 microplate reader) according to the manufacturer's instructions. Five parallel GFBL and SFBL lines were used for the analysis. The experiment was repeated three times. 3.7. Immunostaining analysis For immunostaining, the 3D cultures and cell-free 3D ECM samples were generated on glass coverslips. Briefly, glass coverslips were coated with gelatin, before cells were seeded, to stabilize the attachment of the cells and ECM that they produce (King and  !  (+!  Parsons et al., 2011). To this end, the coverslips were incubated in 0.2% gelatin in PBS at 37oC for 1 h. After rinsing with PBS, coverslips were incubated in 1% glutaraldehyde at room temperature for 30 min, then washed with PBS, followed by incubation with DMEM at 37oC for 30 min. Coverslips were then washed with PBS and stored at 4oC or used immediately. To generate 3D cell cultures and cell-free ECM samples, three representative GFBL and SFBL lines were cultured on the coverslips as described above. At day 7 post-seeding, the cultures were fixed with 4% formaldehyde at room temperature for 20 min, with or without first removing the cells as described above. After washing with PBS, samples containing cells were treated with 0.5% Triton X-100 in PBS for 4 min. All samples were then blocked with PBS+ containing BSA (10mg/ml) and glycine (1mg/ml) at room temperature for 30 min followed by an incubation with the primary antibody (Table 6) diluted in PBS- containing BSA (1mg/ml) in a humidified chamber at 4oC overnight. The samples were then washed with PBS containing BSA (1mg/ml) and 0.01% Triton X-100 and incubated with an appropriate Alexa-conjugated secondary antibody (1:100 dilution; Alexa 488/594; Molecular Probes Inc., Eugene, OR, USA) at room temperature for 1 h. Nuclei were then stained with 300 nM DAPI in PBS for 30 min (Molecular Probes; Eugene, OR, USA). Samples were mounted with Immunomount solution (Thermo Scientific, Pittsburgh, PA) and examined using an Axioplan II Fluorescent microscope (Carl Zeiss Inc. Jena, Germany) and images were captured using Northern Eclipse software (Empix Imaging, Missisauga, ON, Canada). 3.8. Scanning electron microscopy (SEM) GFBL and SFBL were seeded for generation of 3D cultures on coverslips coated with gelatin as described above. At day 7 post-seeding, samples with or without cells (containing ECM only) were prepared as described above. Samples were then washed with 0.1M PIPES buffer (pH=7.4), fixed with 4% formaldehyde in 0.1 PIPES for 10 min and fixed with 2.5% glutaraldhyde for 10 min. In a set of experiments, 2% tannic acid was also present in the fixation step to better preserve glycoproteins and proteoglycans in the ECM. Following PIPES wash, samples were post-fixed in 1% Osmium tetroxide in 0.1 PIPES (pH=6.8) at room temperature for 1 h. The samples were dehydrated through a graded series of ethanol (from 50% to 100%) for 5 min each and then dried using critical  !  (,!  point drying (Samdri-795, Tousimis; Rockville, MD, USA). After drying, the coverslips were attached to aluminum stubs and coated with gold-palladium with a sputter coater (Hummer IV, Technics; Alexandria, VA). The samples that were not treated with tannic acid were viewed and images were captured using Cambridge 260 Stereoscan scanning electron microscope (Cambridge Instruments Ltd, Cambridge, UK).! The samples that were treated with tannic acid were viewed and images were captured using FEI Helios NanoLab 650 Focused Ion Beam scanning electron microscope (Hillsboro, Oregon, USA). 3.9. Quantitative analysis of mRNA expression using real-time RT-PCR Total RNA was extracted from cultured cells using NucleoSpin® RNA II kit and treated with rDNase according to the manufacturer's protocol (Macherey-Nagel). Briefly, cells were washed once with PBS and lysed with RA1 buffer containing 1% betamercaptoethanol at room temperature for 3-5 min. The lysate was filtrated through NucleoSpin Filter at 11,000 x g for 1 min. Supernatants were mixed with equal volume of 70% ethanol and the mixture was centrifuged in the NucleoSpin RNA II Column at 11,000 x g for 1 min. Samples were desalted with MDB buffer, followed by incubation with rDNase (10U) at room temperature for 15 min. Samples were washed with RA2 and RA3 buffer and total RNA was eluted from the column with RNase/DNase-free water. Total RNA concentration and purity was measured by spectrophotometry (Bio-Rad; model 3550 microplate reader) and samples with OD260/280 ratio of 1.8 to 2.0 were used for the study. RNA integrity was assessed by electrophoresis using an agarose gel containing formaldehyde followed by staining of RNA with 0.5$g/ml of ethidium bromide in 0.1M ammonium acetate for 30 min. Gels were assessed for integrity of 18S and 28S rRNAs bands (6333 and 2366 nucleotides, respectively). cDNA was synthesized using iScript™ Select cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions. Briefly, 1.0$g of total RNA was reverse transcribed by adding 4$l of 5x reaction buffer, 2$l of random primers and 1$l reverse transcriptase and nuclease-free water for a final volume of 20$l. The cDNA was synthesized using Mastercycler gradient 5331 Reverse-Transcriptase PCR Instrument (Eppendorf AG, Hamburg, Germany) using the following program: 1 cycle at 25oC for 5 min, 1 cycle at 42oC for 30 min and 85oC for  !  (-!  5 min to heat-inactivate the reverse transcriptase. The primers used for real-time PCR are listed in Table 7. All primers were designed on the boundaries of exons, and analyzed by BLASTn (http://blast.ncbi.nlm.nih.gov/Blast.cgi) for their specificity. The primers were designed to yield a target sequence that was 75-150 base pairs long with a GC content between 40-70%. Efficiency of target amplification was optimized (which includes annealing temperature, primer and sample concentration) up to 95% for each primer set using a 10-fold dilution series of cDNA while standard curves were made. For the reaction, cDNA from each sample was diluted to a concentration (2ng/$l) such that the Ct values were well within the range of their standard curves, and 5$l of diluted cDNA were mixed with 10$l of 2 X iQ SYBR Green I Supermix (Bio-Rad), and 5 pmoles of primers, for a final volume of 20 $l. Real-time PCR amplification was performed on the CFX96 System (Bio-Rad) using the following program: 1 cycle at 94oC for 3 min 35 cycles at 94oC for 10 s, 58oC for 20 s, and reaction completion with reading plate and a melt curve analysis from 65oC to 95oC, 5 s for each 0.5oC. Amplification reactions were conducted for target genes with ubiquitin C (UBC), glyceraldehydes-3-phosphate  dehydrogenase  (GAPDH),  hypoxanthine  phosphoribosyltransferase I (Hprt1), 18s rRNA and asparagine-linked glycosylation 9 (ALG9) as reference genes. For a given experiment, at least two reference genes with a M-value below 0.5 were chosen (Vandesompele et al., 2002). Non-transcribed RNA samples were used as a negative control. The PCR reactions were performed in triplicate for each sample. For 3D cultures, gene expression analysis was performed using five parallel GFBL and SFBL lines. For reseeding experiments, two parallel GFBL and SFBL lines were used and experiment was repeated 5-6 times. The data was analyzed based on the comparative Ct program (CFX Manager Software Version 2.1, Bio-Rad). 3.10. Quantification of collagen abundance in 3D ECM Fibroblasts were grown in 3D cultures for 7 days and cells were removed as described above. Total collagen content was determined by using Sircol Collagen Assay Kit (Biocolor Ltd., Carrickfergus, Northern Ireland, U.K.) according to the manufacturer's instructions. Briefly, collagen from the cell-free 3D ECM was solubilized in 0.1mg/ml of  !  (.!  pepsin (Sigma-Aldrich) in 0.5M acetic acid at 4oC overnight. Samples were then neutralized with an Acid Neutralising Reagent, followed by an incubation with Isolation and Concentration Reagent at 4oC overnight. Samples were centrifuged at 12,000 rpm for 10 min and Sircol Dye Reagent was then added to the pellet and incubated with shaking for 30 min. Samples were centrifuged at 12,000 rpm to allow packing of the collagen-dye complex. The supernatant was discarded and cold Acid-Salt Wash Reagent was added without disrupting the pellet. Samples were centrifuged at 12,000 rpm, the supernatant was discarded and Alkali Reagent was added to the pellet. Samples were vortexed and then measured at 595nm using a spectrophotometer (Bio-Rad). Total collagen concentration was estimated using the Collagen Standard (Biocolor Ltd). The experiment was performed in duplicate for five parallel GFBL and SFBL cell lines. 3.11. Quantification of sulphated glycosaminoglycan in 3D ECM Fibroblasts were grown in 3D cultures for 7 days and cells were removed as above. Total sulphated glycosaminoglycans (GAG) were determined using Blyscan Sulphated GAG Assay Kit (Biocolor Ltd.) according to the manufacturer's instructions. Briefly, sulphated GAG from the cell-free 3D ECM were extracted using 20mg/ml papain (Sigma-Aldrich), with occasional shaking at 65oC for 3 h. The samples were centrifuged at 12,000 rpm for 10 min and supernatants were collected for sulphated GAG quantitation. To this end, samples were incubated with Blyscan Dye Reagent for 30 min and then centrifuged at 12,000 rpm for 10 min. The pellet was incubated with the Dissociation Reagent for 10 min and color reaction was quantitated at 655nm using a spectrophotometer (Bio-Rad). Total amount of sulphated GAG were estimated relative the Glycosaminoglycan Standard (Biocolor Ltd.). The experiment was performed in duplicate for five parallel GFBL and SFBL cell lines. 3.12. Western blotting 3.12.1. Preparation of ECM proteins for Western blotting Cell-free 3D ECM from fibroblasts cultured for 7 days were prepared as above and subjected to enzymatic digestion or left untreated. For the enzyme treatment, the 3D ECM samples were incubated with 0.1U/ml of keratanase II enzyme (Seikagaku  !  )/!  Biobusiness Corp.; Tokyo, Japan) in 50 mM Tris-HCl (pH=7.4), 1U/ml of chondroitinase ABC (Sigma-Aldrich; St. Louis, MO) in 50mM acetic acid and 30mM of Tris-HCl (pH=8.0) at 4oC overnight or with both enzymes consecutively (24 h incubation with chondroitinase ABC followed by 24 h incubation with keratanase II) (Fig. 18). In a preliminary experiment, it was found that keratanase II did not increase the amount of total FMOD detected by Western blot quantification, while chondroitinase ABC treatment alone improved its detection (Fig. 18). Therefore, in the subsequent experiments, chondroitinase ABC pretreatment was used to analyze ECM proteins by Western blotting. 3.12.2. Preparation of cell lysates for Western blotting For assessing cells in reseeding experiments, cells were seeded on cell-free 3D ECM in DMEM (4.0x104 cells/cm2) and cultured for 24 h. Cells were then washed with ice-cold PBS and lysed with 20mM MOPS, 2mM EGTA, 5 mM EDTA, 1% Triton X-100 containing Complete Protease Inhibitor Cocktail (Roche Diagnostics) (pH=7.2). Lysates were collected using a rubber policeman, sonicated on ice and filtered through a NucleoSpin Filter (Macherey-Nagel) by centrifugation with 5,000g for 1 min. Three parallel GFBL (DC, OL, HN) and SFBL (4-1, 2-C, 406) were used for SMAD, ERK and p38 analysis for the reseeding experiments. 3.12.3. Western blotting Total protein concentration in samples was determined using the DC Protein Assay (BioRad). Equal amount of protein of each sample solubilized in SDS sample buffer containing 2-mercaptoethanol was separated in 7.5% SDS-polyacrylamide gel electrophoresis. After electrophoresis, the proteins were transferred onto a Hybond-ECL nitrocellulose membrane (Amersham, Little Chalfont, Buckinghamshire, UK) at 4°C overnight. The nonspecific binding sites were blocked by incubating the membranes in Odyssey Blocking Buffer (LI-COR Biosciences; Lincoln, NE, USA) at room temperature for 1 h. The membranes were then incubated with the primary antibody (Table 6) in Odyssey Blocking Buffer, containing 0.1% Tween-20 at 4oC overnight. After washing with TBS containing 0.1% Tween-20 (TBS-T), the membranes were incubated with an  !  )&!  appropriate species-specific secondary antibody conjugated with IRdye (1:10,000; LICOR Biosciences) in Odyssey Blocking Buffer, containing 0.1% Tween-20 and 0.01% SDS at room temperature in the dark for 1 h. After washing with TBS-T (4 times for 5 min each), the blots were washed in TBS for 5 min and dried. The blots were detected using the LI-COR Odyssey Infrared Imaging system (LI-COR Biosciences). The results were quantified using the Odyssey application software version 3.0. 3.13. Cell adhesion assessment Cell-derived 3D ECM and type I collagen substrates were prepared on glass coverslips as described above. GFBL and SFBL were seeded in DMEM at a density of 3x104 cell/cm2 on the substrates. At 25 min, 45 min, 75 min, 3 h and 24 h post-seeding, cells were fixed with 4% formaldehyde at 4oC overnight. The samples were washed with distilled H2O (dH2O), air-dried, and stained with 0.1% crystal violet in 200 mM boric acid at room temperature for 1 h. Cells were then rinsed with dH2O and air-dried. Cell adhesion and spreading was determined from digital images obtained using a phase contrast microscope. 3.14. Statistical analysis All data is presented as mean ± standard error of the mean of 3-5 parallel GFBL and SFBL lines. Statistical analyses were done by using Student’s t-test or two-way ANOVA, using  !  GraphPad  Prism  5.  p<0.05  was  )'!  considered  statistically  significant.  CHAPTER FOUR – RESULTS 4.1. Phenotypic characterization of gingival- (GFBL) and skin-derived fibroblasts (SFBL) 4.1.1. Distinct cellular morphology and ECM organization in GFBL and SFBL 3D ECM To assess cell morphology and organization in the 3D cultures, we used phase contrast microscopy (Fig. 3 A,E), immunostaining (Fig. 3B,C,F,G), nuclei staining (Fig. 3D,H) and SEM (Fig. 4) 7 days post-seeding. GFBL adopted a more spindle-like and elongated morphology (Fig. 4A), whereas SFBL were larger and wider in morphology (Fig. 2B). In addition, cellular organization was more polar and parallel to one another in GFBL compared to SFBL cultures (Fig. 3). This cellular arrangement appeared to affect ECM deposition and organization, as similar differences were observed in ECM organization by SEM (Fig. 4C-F). 4.1.2. GFBL had greater proliferation rate than SFBL Proliferation rate is a key phenotypic characterization that often varies from cell-type to cell-type. When seeded at low density, GFBL proliferated significantly faster than SFBL at 3 and 6 days after seeding (Fig. 5B). At day 8, cell number in GFBL cultures was also greater than in SFBL cultures, however, the difference was not statistically significant. Likewise, at high density conditions (Fig. 5A), where fibroblasts were generating and embodied in their own 3D ECM, there was a significantly greater number of GFBL than SFBL by day 7 and this difference remained statistically significant at day 10 and 14 post-seeding. 4.1.3. GFBL and SFBL showed no significant difference in total protein abundance in 3D cultures A major role for fibroblasts in wound healing is to synthesize and deposit ECM proteins to restore the connective tissue that was lost due to wounding. Therefore, GFBL and SFBL were analyzed for their capacity to deposit total protein in the 3D ECM. The amount of total protein in the GFBL-derived 3D ECM was higher at 3, 7, 10 and 14 days after cell seeding (Fig. 5C). However, when total protein abundance was calculated  !  )(!  relative to cell number, SFBL produced more protein per cell at all time-points (Fig. 5D). However, there were no statistically significant differences between GFBL-derived 3D ECM (GECM) and SFBL-derived 3D ECM (SECM) in either analyses. 4.1.4. GFBL and SFBL displayed distinct gene expression profiles in 3D cultures The gene expression profile of wound healing associated genes was characterized to determine whether SFBL are phenotypically distinct compared to GFBL. Gene expression analysis was performed at a defined time-point (day 7 post-seeding) during ECM synthesis and deposition in 3D cultures. Results demonstrated a significant difference in gene expression in several classes of genes between SFBL and GFBL. Fibrillar ECM proteins collagen type I (p<0.05), collagen type III (p<0.001) and elastin-1 (p<0.001) were more highly expressed in SFBL (Fig. 6). Apart from elastin, other components of elastic fibers, fibrillin-1, emilin-1 and -2 did not show a significant difference. Emilin-3 expression was negligible in GFBL and SFBL (Ct=31-34). Furthermore, EDA- and EDB-fibronectin isoforms, which are two ECM glycoproteins expressed abundantly during wound healing, showed no difference in gene expression between SFBL and GFBL. Matricellular proteins were more highly expressed by SFBL (Fig. 7). SFBL showed significantly greater expression of both tenascin-C (p<0.001) and tenascin-X (p<0.05) compared to GFBL. The matricellular proteins that have a role in ECM modeling and organization, SPARC-1 (p<0.01) and osteopontin (OPN) (p<0.05), were also more highly expressed in SFBL. The anti-angiogenic factor thrombospondin-2 (THBS-2) (p<0.05) also showed significantly greater expression in SFBL. Among the three CCN family members, only CCN-2/CTGF showed significantly greater expression in SFBL (p<0.05). Hevin expression was negligible in both GFBL and SFBL (Ct=30-32). Among the five SLRPs analyzed for mRNA expression, SFBL showed significantly greater expression for DCN (p<0.001), BGN (p<0.001), FMOD (p<0.01) and LUM (p<0.01) (Fig. 8). Expression of SLRPs was also assessed in proportion to one another,  !  ))!  using the 2(Ctreference-Cttarget) method (Fig. 9). It was found that DCN was most highly expressed among the SLRPs, while FMOD was the least expressed in both GFBL and SFBL. Expression of total SLRP mRNA was up to 10 times higher in SFBL compared to GFBL (Fig. 9). Pro-fibrotic growth factors including TGF-!1 (p<0.01) and CTGF/CCN-2 (p<0.05) were more highly expressed in SFBL, while the pro-fibrotic TGF-!2 (Ct=26-30) was more highly expressed in GFBL (Fig. 10A). In addition, the anti-fibrotic growth factor TGF-!3 was more highly expressed in SFBL (p<0.001), but the level of expression (Ct=26-30) was far less than TGF-!1 (Ct=21-24). When the proportional expression of TGF-!1, -!2 and TGF-!3 was assessed in individual cell lines by the 2(Ctreference-Cttarget) method, SFBL lines in general showed higher total TGF-! expression compared to GFBL (Fig. 10B). However, SFBL showed a higher relative proportion of TGF-!3 (8%) compared to GFBL (2%) out of the total TGF-! levels in these cell types. The secreted protein Cthrc1, which inhibits SMAD2/3 activation, also showed a significantly greater expression in SFBL (p<0.05) (Fig. 10A). The pro-angiogenic FGF-2 showed no significant difference in expression between GFBL and SFBL, but expression of pro-angiogenic VEGF-" was significantly greater in GFBL (p<0.05) (Fig. 10A). MMP expression was in general significantly greater in GFBL than in SFBL (Fig. 11 and 12). GFBL expressed significantly more MMP-1 (p<0.05), MMP-3 (p<0.05) and MMP10 (p<0.05), while SFBL expressed significantly more MMP-7 (p<0.01) and MMP-11 (p<0.05) than GFBL (Fig. 11). Analysis of overall proportion of MMPs by the 2(CtreferenceCttarget)  method showed that the total MMP expression was greater in GFBL compared to  SFBL (up to 10 times greater) (Fig. 12). In all cell lines, MMP-1 and MMP-10 were the most abundant MMPs expressed in GFBL, while MMP-2 was the most abundantly expressed in SFBL (Fig. 12). MMP-7 and MMP-11 were expressed at very low levels compared to other MMPs (Fig. 12). MMP-13 expression was negligible in GFBL and SFBL (Ct=32-34). Among the four members of the TIMP family, only TIMP-4 was more highly expressed in GFBL (p<0.05), while the other genes showed no significant difference between SFBL and GFBL (Fig. 11).  !  )*!  Genes associated with myofibroblast phenotype and cell contractility, including P311 (p<0.001), "-SMA (p<0.05), non-muscle myosin IIA (NMMIIA) (p<0.05) and nonmuscle myosin IIB (NMMIIB) (p<0.01) were significantly more expressed in SFBL (Fig. 13). In addition, genes involved with intracellular turnover of ECM, Endo180 (CD280) (p<0.05) and cathepsin K (CTSK) (p<0.001) were significantly more expressed in SFBL compared to GFBL, while there was no difference in the expression of lipoprotein receptor-related protein 1 (LRP1) (Fig. 14). 4.1.5. Gene expression remained significantly different between GFBL and SFBL during 3D ECM generation over time To determine whether gene expression remained significantly different over time between GFBL and SFBL in 3D cultures, a set of genes was analyzed at days 3, 7, 10, and 14 post-seeding (Fig. 15). The analysis was focused on genes that showed significantly different expression in GFBL compared to SFBL in the above analyses at day 7 post-seeding. In general, all the analyzed genes showed constant and significantly different level of expression in GFBL and SFBL from days 3 to 14 post-seeding. The only gene that showed significantly altered gene expression over time was DCN, as it showed significantly increased expression from days 3 to 14 in SFBL, but not in GFBL. 4.1.6. Characterization of ECM protein abundance To characterize the composition of the 3D ECM generated by the cells, the abundance of collagen, SLRPs, EDA-fibronectin and tenascin-C was analyzed. Total collagen in the 3D ECM relative to total protein or total cell number, showed no significant difference between SECM and GECM (Fig. 16A-B). Immunostaining of collagen type I showed that collagen organization followed the distinct shape and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and displayed more parallel organization compared to SFBL (Fig. 16C). To assess the abundance of sulphated proteoglycans, the amount of total sulphated GAG was analyzed relative to total protein and total cell number. No significant difference was found between the different tissue-derived 3D ECM (Fig. 17A-B).  !  )+!  To further characterize the composition of sulphated proteoglycans in the 3D ECM, we performed Western blotting of SLRPs. To improve access of antibodies directed against the core proteins of SLRPs, enzyme treatment to remove GAG chains is usually required. Therefore, we first determined the effectiveness of chondroitinase ABC (cABC) and keratanase enzymes to expose antigenic sites in SLRPs. The 3D ECM samples were either treated with cABC only, keratanase only or with both enzymes consecutively. It was found that the cABC enhanced the detection of DCN and BGN, two dermatan/chondroitin sulphate proteoglycans (data not shown). Interestingly, also the detection of FMOD was enhanced by cABC, while keratanase alone did not affect FMOD detection, even for samples that were treated with cABC first (Fig. 18). Therefore, cABC digestion was also used for detecting keratan sulphate proteoglycans FMOD and LUM. In the SLRP protein analysis using Western blotting (Fig. 19), only BGN was significantly greater in SECM than GECM (Fig. 19A,E). Abundance of BGN, FMOD and LUM was fairly uniform in GECM and SECM, but DCN showed considerably variable abundance among different individual cell lines (Fig. 19). In order to assess organization of SLRPs in the 3D ECM at day 7 post-seeding, we used immunostaining. There were no obvious differences in SLRPs abundance between GECM and SECM. However, GFBL showed a greater abundance of intracellular immunostaining for DCN, FMOD and LUM compared to SFBL (Fig. 20). In the 3D ECM, SLRPs were organized into a fibrillar-like orientation, suggesting that these molecules were associated with collagen or other fibrillar ECM proteins (Fig. 20). Abundance of cellular EDA-fibronectin (Fig. 21) and tenascin-C (Fig. 22), two molecules involved in wound healing and scar formation, was determined using Western blot analysis. There was no significant difference in abundance of EDA-fibronectin between GECM and SECM, which was consistent with the mRNA expression analysis (Fig. 21). However, there was considerable difference in EDA-fibronectin abundance in 3D ECM derived from individual GFBL and SFBL lines. The level of ECM associated tenascin-C was significantly different in the matrices. When 3D ECM samples were analyzed without cABC enzyme pretreatment, SECM contained a significantly greater abundance  !  ),!  of tenascin-C (p<0.05) (Fig. 22A). However, with cABC pretreatment, GECM contained more tenascin-C (p=0.15) (Fig. 22B). Immunostaining of EDA-fibronectin and tenascinC showed that organization of these molecules followed the distinct shape and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and showed more parallel organization compared to SFBL (Fig. 23). Pretreatment with cABC enzyme did not result in a difference in tenascin-C immunostaining in both GECM and SECM (data not shown). 4.2. The role of the ECM on cell phenotype Our findings have shown that GFBL and SFBL had a distinct gene expression profile during the generation of the 3D cultures. The composition and structural organization of the 3D ECM within these cultures can have a profound effect on cellular function and phenotype. Therefore, we assessed the effects of the 3D ECM on GFBL and SFBL phenotype by seeding these fibroblasts on GECM and SECM. Moreover, we analyzed how the 3D ECM modulated cell adhesion, gene expression and signaling pathways. 4.2.1. GFBL and SFBL showed no difference in adhesion rate on either GFBL-derived 3D ECM (GECM) and SFBL-derived 3D ECM (SECM) Cell adhesion is regulated by the molecular composition and organization of the substrate to which the cells are exposed. Therefore, adhesion rate was determined in GFBL and SFBL seeded on either GECM, SECM and on collagen type I as a control substrate for 24 h. Fibroblasts appeared to have similar adhesion rate irrespective of whether they were seeded on GECM and SECM (Fig. 24). However, relative to collagen I substrate, cellderived 3D ECM appeared to promote faster cell adhesion for both GFBL and SFBL (Fig. 24). In addition, SFBL seemed to have greater adhesion and spreading rate compared to GFBL on all three substrates. This observation was most noticeable when cells were seeded on collagen I substrate at 25 and 75 minutes. In addition, the morphology of the cells on 3D ECM adopted a more spindle-like morphology compared to cells on collagen I substrate. Lastly, the orientation of fibroblasts seemed to adopt the organization of the 3D ECM, such that cells on GECM were organized in a parallel manner, whereas cells on SECM were organized in a more random fashion.  !  )-!  4.2.2. There were no significant differences between GECM and SECM in modulating gene expression To determine whether the GECM and SECM can influence the cell phenotype, GFBL and SFBL were reseeded on these matrices and mRNA expression was analyzed for a set of genes. GECM and SECM did not modify gene expression differently in GFBL and SFBL, as gene expression within a given cell type was similar whether seeded on either 3D matrices. However, among the 30 genes that were analyzed, 16 of the genes showed a significant difference between GFBL and SFBL in all substrates (Fig. 25-30, Table 8). Similar difference for these genes was also observed in the analysis of GFBL and SFBL in 3D cultures, as shown above. Therefore, these genes were classified under the “predetermined” category, as gene expression was significantly different between GFBL and SFBL, irrespective to the cell culture conditions (Table 8). The majority of the genes in this category were matrix deposition associated genes, including types I and III collagen, elastin-1, tenascin-C, tenascin-X, SLRPs (DCN, BGN, FMOD, and LUM) and TIMP-4. The other 14 genes analyzed at 24 h after reseeding on the 3D ECM substrates showed no significant difference in gene expression between GFBL and SFBL (Fig. 25-30, Table 8). Therefore, differential expression of these genes is therefore dependent on the 3D culture environment. In this category, most MMPs, ECM internalization and intracellular degradation-associated genes (Endo180 and CTSK) and various matricellular protein genes (SPARC, THBS-2, OPN) were found significantly different between GFBL and SFBL only under 3D cultures 7 days post-seeding. To further determine whether serum (FBS and ascorbic acid) promoted differences in gene expression in GFBL and SFBL, fibroblasts were seeded on collagen I substrate with or without serum for 24 h. Among the 14 genes that were not significantly different in the reseeding experiment, expression of CTGF was significantly affected and was greater in SFBL than GFBL. Therefore, serum promotes a greater expression of pro-fibrotic CTGF expression in SFBL compared to GFBL. Collectively, this analysis allowed us to assess on whether these wound healing associated genes were distinctively regulated by different environmental cell culture conditions. Interestingly, majority of matrix deposition associated genes were innately more highly expressed in SFBL compared to  !  ).!  GFBL. Also, matrix remodeling MMPs, ECM internalization and intracellular degradation associated genes and certain matricellular proteins genes were differently regulated in GFBL and SFBL under 3D cultures 7 days post-seeding, but not when SFBL and GFBL were reseeded on various 3D ECM substrates. A detailed analysis on comparing gene expression from the 3D cultures 7 days post-seeding, reseeding experiment and serum response is summarized in Table 8. The 3D ECM did have an effect on modulating expression of certain genes, compared to collagen I substrate. Expression of MMP-1, -3 and -10 genes was significantly greater when GFBL and SFBL were seeded on 3D ECM compared to collagen I substrate (Fig. 27, Table 9). In addition, VEGF-" gene expression was greater when SFBL and GFBL were seeded on 3D ECM compared to collagen I substrate. However, this was not statistically significant (Fig. 28, Table 9). ! 4.2.3 Cell signaling pathways The extracellular environment within the 3D cultures regulated different subset of genes, particularly MMP, TGF-! and certain matricellular proteins (Table 8), in GFBL and SFBL. These genes can be regulated by TGF-! via SMAD3 intracellular signaling, or via ERK1/2 MAP kinase (MAPK) and p38 MAPK pathway. To determine whether the cellderived 3D matrices differentially regulated these signaling pathways, GFBL and SFBL were seeded on GECM, SECM and collagen type I substrates and the level of phosphorylated SMAD3, ERK1/2 and p38 were assessed by Western blotting. 4.2.4. Phosphorylation of SMAD3 was greater in SFBL TGF-! is known to be a potent growth factor in promoting the deposition of matrix protein during wound healing. The TGF-! signaling pathway involves the activation via phosphorylation of SMAD complexes, which then promotes the interaction of transcription factors in the nucleus to foster the expression of target genes. TGF-! can be present in the ECM (Roberts, 1995). Therefore, we assessed whether 3D ECM derived from GFBL and SFBL can induce differential phosphorylation of SMAD3, a downstream mediator of TGF-! signaling, in SFBL and GFBL. Phosphorylation of SMAD3, relative  !  */!  to total SMAD3 was significantly greater in SFBL, but the nature of 3D ECM did not affect that difference (Fig. 31, 32D), and GFBL contained a significantly greater amount of total SMAD3 protein compared to SFBL (Fig. 31, 32B). Greater SMAD3 phosphorylation and lower total SMAD3 in SFBL were also found in 3D cultures 7 days post-seeding (data not shown). 4.2.5. ERK1/2 phosphorylation was greater in fibroblasts when seeded on 3D ECM compared to collagen I substrate Cell-ECM interaction can also regulate the MAPK pathway (Johnson and Lapadat, 2002). In addition, TGF-! signaling interacts and activates the MAPK signaling pathway, which regulates various genes involved in wound healing (Mulder, 2000, Yu et al., 2002). ERK1/2 MAPK activation was analyzed and there was no difference in ERK phosphorylation when fibroblasts were seeded on GECM or SECM (Fig. 33, 34A). Interestingly, however, there was a significantly greater ERK1/2 phosphorylation in SFBL when seeded on 3D substrate, as compared to collagen I substrate (Fig. 33, 34C). GFBL also had a greater ERK1/2 phosphorylation when seeded on 3D ECM, but this did not reach statistical significant (p=0.12) (Fig. 34B). The p38 MAPK phosphorylation was not different when fibroblasts were seeded on GECM or SECM or on collagen I substrate (Fig. 35). In addition, there was no innate difference in p38 phosphorylation between GFBL and SFBL.  !  *&!  CHAPTER FIVE – DISCUSSION 5.1. GFBL and SFBL showed an innately distinct gene expression profile Wound healing in the human oral mucosa results in scarless healing, while wound healing in the skin results in scarring (Wong et al., 2009; Mak et al., 2009). Therefore comparing fibroblast phenotype and interactions with their ECM niche in the gingiva and skin may provide novel information about the factors that regulate scar formation. While all fibroblasts are generally similar in spindle-like shape and have the ability to synthesize ECM components, fibroblasts isolated from different anatomical sites have been shown to be functionally and phenotypically distinct (Fries et al., 1994; Sempowski et al., 1995). In our study, SFBL and GFBL had a distinct gene expression profile in 3D cultures. In order to assess whether these distinct gene expression profiles depended on the 3D culture conditions, cells were reseeded on cell-free 3D ECM derived from GFBL or SFBL or on a 2D collagen substrate and gene expression was analyzed after 24 h. Interestingly, expression of many of the ECM deposition genes remained significantly higher in SFBL, regardless of the cell growth conditions. Thus, the expression of these genes may be “pre-programmed” or innately distinct between GFBL and SFBL (Table 8). This innate difference may be partly attributed to the embryonic origin of the cells. Gingival!fibroblasts originate from the neural crest, while most fibroblasts from the trunk dermis in humans are derived from the mesodermal layer (Buckley et al., 2001; Breau et al., 2008). Previous studies have shown that GFBL and SFBL preserve their distinct HOX gene expression pattern established during embryogenesis, which may explain their distinct phenotype in adult tissues (Chang et al., 2002; Rinn et al., 2008). In addition, gene expression analyses have shown that in standard 2D cultures on a plastic cell culture substrate, GFBL and SFBL had distinct gene expression profiles (Chang et al., 2002; Guo et al., 2011a) suggesting that, indeed, developmental origin determines fibroblast phenotype and function. A more detailed analysis showed that 287 out of 5284 genes were significantly different between GFBL and facial SFBL in standard 2D cell cultures (Ebisawa et al., 2011). Specifically, cytokine and growth factor-related genes were more highly expressed in GFBL than SFBL and there were also significant differences in ECM-related gene !  *'!  expression. For example, GFBL had higher levels of collagen type IV expression and lower levels of tenascin-C and collagen type X expression compared to SFBL (Ebisawa et al., 2011). Interestingly, the only similarity found between this study and our study was that expression of tenascin-C was greater in SFBL compared to GFBL. The lack of similarities may be due to differences in origin of the cell lines used in the studies as the previous study used facial skin fibroblasts while our study utilized fibroblasts from breast tissue. Recent evidence has shown that there are considerable differences in gene expression between fibroblasts derived from different anatomical locations of skin, including the abdomen, back, thigh, foreskin and scalp (Chang et al., 2002). 5.3. Fetal-like characteristics of GFBL While fetal skin heals without scarring, post-natal skin loses this ability. This is different from adult gingival tissue that possesses scarless healing characteristics (H#kkinen et al., 2000; Ferguson and O’Kane, 2004). Thus, fetal skin and adult gingiva may possess similar characteristics that favor scarless healing. Accordingly, our findings showed similarities to previous studies assessing fetal and adult skin fibroblasts (Dang et al., 2003; Ferguson and O’Kane, 2004 Colwell et al., 2006). Like fetal skin fibroblasts, we found that GFBL had reduced TGF-!1 and CTGF expression and elevated MMP expression relative to SFBL. In addition, similar to fetal skin fibroblasts, the proliferation rate of GFBL was greater as compared to adult SFBL (Schor et al., 1996). In contrast, however, fetal skin wounds have greater relative abundance of tenascin-C and FMOD and reduced levels of DCN and BGN compared to adult skin (Whitby et al., 1991, Soo et al., 2000). An opposite relative abundance of these molecules was found between GFBL and SFBL in our study. This difference suggests that regulation of these molecules may be different in vivo and in vitro. Alternatively, regulation of scar formation by these molecules may be more complex than previously thought. 5.3. Characterization of the 3D ECM Cellular microenvironment (niche) is an important regulator of cell phenotype and function (Green and Yamada, 2007). There are quantitative and qualitative differences in ECM proteins, including collagen type III, proteoglycans and tenascin-C, between  !  *(!  gingiva and skin tissue (Narayanan et al., 1980; Bronson et al., 1988; Wong et al., 2009). In addition, the nature of this niche is determined by the phenotype of cells that produce it, but in a reciprocal manner, the niche may also impact cell function. Interestingly, GECM and SECM generally had similar effects on gene expression of the given cell type, although microscopical characterization of the 3D matrices showed differences. In particular, organization of the fibrillar 3D ECM between SFBL and GFBL cultures was distinct and followed the morphological difference of SFBL and GFBL during the generation of these matrices. Specifically, GFBL were well aligned and showed a more narrow cell morphology as compared to wider and more randomly organized SFBL. This morphological difference appears to be inherent to the cells as similar morphological difference was present regardless on whether the cells were seeded on SECM, GECM or 2D collagen substrates. In contrast to in vivo, the parallel organization of type I collagen in GECM is more similar to collagen bundle organization in scar tissue (Durani et al., 2008). Therefore, this suggests that the regulation of cell and fibrillar ECM organization may be different in vitro and in vivo. Molecular characterization of the SECM and GECM showed surprisingly little differences, as no significant differences were found in total collagen, total sulphated proteoglycans and many SLRPs. The only differences noted were in the abundance of BGN and tenascin-C between GECM and SECM. Therefore, it may not be surprising that these two matrices did not induce distinct gene expression in cells seeded on them. It is worth noting that expression of some MMPs was significantly different in fibroblasts on 3D ECM substrates compared to fibroblasts on 2D collagen substrate. Thus, the organization and composition of cell-derived 3D ECM in general had a distinct impact on gene expression when compared to standard 2D collagen substrate. This is supported by similar findings comparing function of cells on 2D substrates and 3D ECM (Cukierman et al., 2001; Soucy et al., 2009). The fact that differential gene expression of several ECM proteins by SFBL and GFBL does not translate into different composition of the cell-free 3D ECM substrates from these cell types can be explained by the fact that only a portion of molecules produced by fibroblasts in culture become incorporated into the matrix, while the majority is usually secreted into the medium. For instance, previous  !  *)!  findings using human GFBL have shown that approximately 43% of total proteoglycans were secreted into the medium, while 14% were found in ECM (Larjava et al., 1992). In addition, it remains to be shown whether the extraction protocol of cells from the 3D ECM affects its molecular composition. Interestingly, analysis of corresponding protein level of tenascin-C by Western blotting showed either higher or lower abundance in GECM, depending on whether samples were pretreated with cABC or if left untreated. Previous work has found that GAG in the PCM of fibroblast cultures derived from different anatomical sites had different susceptibility to cABC enzyme pretreatment (Bronson et al., 1988). Therefore, it is possible that assembly of the ECM was different between SECM and GECM, which would explain the increased detection of tenascin-C in GECM, but no change in SECM when cABC enzyme pretreatment was used. Previous studies have shown that chondroitin sulphate proteoglycans can interact with tenascin-C (Hoffman and Edelman, 1987; Grumet et al., 1994) and that cABC can release tenascin-C from these molecules (Hoffman and Edelman, 1987). Therefore, it is likely that cABC pretreatment may allow better exposure and/or release of tenascin-C from the proteoglycan-tenascin-C complex in the 3D ECM. To further support this idea, we also found that cABC pretreatment helped with detecting FMOD and LUM in our analysis, despite cABC not having a direct effect on keratan sulphate chains present in FMOD and LUM (McEwan et al., 2006). Non-enzyme pretreated 3D ECM samples and even keratanase enzyme pretreatment alone did not improve FMOD and LUM detection, suggesting that in a tissue-type 3D ECM, chondroitin and/or dermatan sulphate chains of proteoglycans can mask keratanase sensitive sites in keratan sulphate proteoglycans. 5.4. Impact of the 3D culture niche on cell phenotype One of the key findings in this study was that GFBL and SFBL had innately distinct expression of certain genes while other genes were differently regulated only when cells were exposed to the 3D culture environment, but not when cells were reseeded on 3D ECM. Therefore, factors present in the 3D culture can differentially regulate expression of certain genes in GFBL and SFBL. It is possible that removal of cells necessary for  !  **!  generating the cell-free 3D ECM substrates for the reseeding experiments may have affected ECM molecule composition and presence and activity of PCM-bound growth factors. For instance, alkaline pH higher than 8.0 has been shown to activate latent TGF! in the ECM (Lyons et al., 1988). Cells interact with growth factors and cytokines that are localized and bound to the PCM generated by the cells (Macri et al., 2007). Growth factors in the PCM are essential for spatiotemporal coordination of cellular function in wound healing (Macri et al., 2007). For instance, TGF-! can be bound to DCN, BGN and FMOD in the ECM and is therefore stored and sequestered from interacting with cellular receptors (Yamaguchi et al., 1990; Soo et al., 2000; Stoff et al., 2007). In addition, various other growth factors, including VEGF, FGF-1 and -2, TNF-" and PDGF can be present in the matrix (Macri et al., 2007). Therefore, the distinct gene expression profile between 3D cell culture and reseeding in fibroblasts may be partly due to differences in cell-ECM interaction. Mechanical strain from the ECM, which is determined by the composition and organization of the matrix and by forces generated by the given cells, affects cell function (Yamada et al., 2003; Schwartz, 2010). This may be different between GECM and SECM and may be affected by the cell-extraction protocol. It has been shown that mechanical strain in fibroblasts, promoted a significant change in expression in approximately 20% of the genes analyzed by microarray (Cui et al., 2004). Interestingly, mechanical strain also induced the activation of latent TGF-! and upregulation of CTGF, FGF and tenascin-C expression in fibroblast and myofibroblasts (Cui et al., 2004; Guo et al., 2011a; Klingberg et al., 2013). We found that cell contractility and myofibroblast associated genes, including "11 integrin, NMMIIA, NMMIIB, "-SMA and P311 (Singer and Clark, 1999; Pan et al., 2002; Bond et al., 2011; Talior-Volodarsky et al., 2012) were more highly expressed in SFBL compared to GFBL. Therefore, mechanisms regulated by the mechanotransduction pathway may at least partially explain the phenotypical differences between GFBL and SFBL in the 3D cultures.  !  *+!  5.5. Cell adhesion SFBL showed faster adhesion and spreading irrespective to the adhesion substrate. Thus, in addition to innately distinct gene expression and cell growth rate, SFBL and GFBL interact with ECM molecules differently. Cell adhesion and spreading depend mostly on integrin-mediated cell adhesion to ECM proteins and dynamic adaptation of the cytoskeleton to the integrin-mediated signals (Green and Yamada, 2007). Interestingly, previous studies have found that GFBL and SFBL expressed different integrin subunits in standard 2D cultures (Palaiologou et al., 2001) and that SFBL had a greater adhesion rate than GFBL on both fibronectin and collagen type I coated substrate (Guo et al., 2011a). Enhanced adhesion rate in SFBL was suggested to be due to greater expression and protein abundance of "2 and "4 integrins and FAK (Guo et al., 2011a). Thus, differential expression of integrin-type ECM receptors may underlie the distinct interaction of SFBL with the ECM proteins. Detailed integrin expression profile of cells used in the present study in the context of 3D ECM remains to be verified. 5.6. Greater expression of TGF-! and SMAD3 phosphorylation in SFBL and its potential implication on matrix deposition associated gene expression In order to get insight into the signaling pathways that may underlie differential gene expression in SFBL and GFBL, we assessed activation of three pathways, TGF-!SMAD3, ERK1/2 and p38, known to regulate expression of ECM proteins and matrix remodeling genes (Westermarck and Kahari, 1999; Laping et al., 2002; Leask et al., 2003; Mulsow et al., 2005). To this end, cells were seeded on SECM and GECM substrates or on 2D collagen and phosphorylation of the target proteins were analyzed by Western blotting at steady state 24 h after seeding. Interestingly, the results showed that regardless of the culture conditions, GFBL had a higher level of total SMAD3, whereas SFBL had a significantly greater proportion of activated (phosphorylated) SMAD3. This was associated with higher expression of total TGF-! (TGF-!1, -!2 and -!3) in the 3D cultures by SFBL, while there were no significant differences in the expression of TGF-! receptors between the cell types. Therefore, it is possible that the major fibrogenic TGF-! pathway is innately more active in SFBL, due to increased autogenous TGF-! secretion and SMAD3 phosphorylation by SFBL. This is supported by our findings showing that  !  *,!  several genes that were more highly expressed by SFBL were TGF-! signaling targets (Table 10). Interestingly, a previous study has also reported that GFBL had reduced ability to upregulate fibrogenic genes as a response to TGF-!1 as compared to SFBL (Guo et al., 2011b). We suspect that TGF-! signaling response is different between GFBL and SFBL and may in part underlie the distinct gene expression profile found in the present study. Therefore in the future, it is worth studying in greater detail the role of TGF-! mediated signaling pathways on gene expression in GFBL and SFBL. For instance, this difference may be depended on the level of TGF-! co-receptors (e.g. endoglin, betaglycan) expressed by these cells (Goumans et al., 2009), secretion of autocrine TGF-! isoforms at a protein level, differences in the level of molecules that are required for activation of TGF-! (e.g. integrins, proteases) (McCawley and Matrisian, 2001; Annes et al., 2004), natural TGF-! inhibitors (e.g. LTBP) (Border and Noble, 1994), or differential regulation of canonical (SMAD) or non-canonical (e.g. TAK1, ERK1/2) TGF-! signaling pathways (Laping et al., 2002; Attisano and Wrana, 2002). In contrast to differential activation of SMAD3, no differences in the level of total or phosphorylated ERK1/2 or p38 were found between SFBL and GFBL. However, both SFBL and GFBL showed a greater ERK1/2 phosphorylation when seeded on 3D ECM as compared to 2D collagen substrate, suggesting that organization and/or composition of the 3D ECM can induce ERK1/2 phosphorylation in these cells. Recent findings have shown that NIH-3T3 fibroblasts interacting with cell-derived 3D ECM and 2D substrates utilize a different set of adhesion molecules, including integrins, paxillin and FAK (Cukierman et al., 2001). Integrin-mediated signaling is one of the key activators of the ERK1/2 MAPK pathway (Giancotti and Ruoslahti, 1999) and therefore, the organization of the 3D ECM may affect the activation of this pathway. In addition, 3D ECM contains a distinct biochemical composition of integrin ligands, including fibronectin and collagen, and adhesion modifiers, such as matricellular proteins, which may have an effect on intracellular signaling (Yamada et al., 2003). The 3D ECM may also contain stored growth factors, including FGF-2, VEGF and TGF-!, which are known to activate MAPK pathway (Milanini et al., 1998 Mulsow et al., 2005). Finally, a previous study using the standard floating/anchored collagen gel model found that mechanosignaling  !  *-!  from the ECM regulated phosphorylation of ERK1/2 in response to serum in fibroblasts (Rosenfeldt and Grinnell, 2000). Therefore, distinct mechanical properties of the 3D and 2D culture substrates may account for differences in ERK1/2 phosphorylation on these two substrates. It is important to note that ERK1/2 activation may be distinct at different time-points, as our analysis only tested ERK1/2 phosphorylation at 24 hours post-seeding. For instance, analyzing ERK1/2 activation may be significantly different at an earlier time-point, where GFBL and SFBL had a more spread morphology on 3D ECM compared to collagen type I substrate. Interestingly, ERK1/2 signaling is crucial for cell spreading (Lai et al., 2001), however contrary to our cell spreading and adhesion analysis (Fig. 24), we found lower ERK1/2 phosphorylation in fibroblasts with larger spread morphology on collagen type I substrate than on 3D ECM, 24 hours post-seeding. Therefore, other mechanisms aside from ERK1/2 signaling are also likely involved in cell spreading on these different substrates. 5.7. Potential association and mechanisms of differentially regulated individual genes in wound healing and scar formation The following will summarize some of the potential key functions of differentially expressed genes in GFBL and SFBL in wound healing and scar formation. 5.7.1. Collagens Excess production of collagen is a hallmark of scarring (Cutroneo, 2003). Our findings showed that SFBL expressed significantly higher levels of type I and III collagen than GFBL. Previous findings have shown that there was a greater abundance of cells positive for type I procollagen in skin wounds compared to gingival wounds of red Duroc pigs (Wong et al., 2009). Therefore, innately higher expression of collagen by SFBL may at least partially underlie the increased scar formation in skin compared to gingiva.  !  *.!  5.7.2. Tenascins Expression of tenascin-C and tenascin-X, two matricellular proteins implicated in wound healing (H#kkinen et al., 2012), was significantly greater in SFBL compared to GFBL. Previous work has also shown that tenascin-C was more highly expressed in SFBL compared to GFBL in 2D cultures (Ebisawa et al., 2011). During wound healing, tenascins play a role in promoting fibronectin deposition, fibroblast migration and increasing the strength and integrity of connective tissue (Midwood et al., 2004; H#kkinen et al., 2012). However, detailed role of tenascins in scar formation remains unclear as previous studies have shown higher levels of tenascin-C in scarless fetal and gingival wounds, but lower levels during limb regeneration of amphibians (Whitby et al., 1991; Wong et al., 2009; Seifert et al., 2012a). 5.7.3. SLRPs Among SLRPs, DCN was the major molecule expressed by cells, although GFBL showed heterogeneity in its expression. In addition, FMOD was the least expressed SLRPs in both cell types. SLRPs are structurally similar and have overlapping functions, for instance in regulating collagen fibrillogenesis and TGF-! activity (Merline et al., 2009). It has been shown that overexpression of DCN and FMOD reduced conjunctival and dermal scarring in rabbits, respectively (Grisanti et al., 2005; Stoff et al., 2007). In addition, DCN expression increases with gestational age in fetal fibroblasts, where skin converts from scarless healing potential to scar-forming healing (Beanes et al., 2001). Therefore, it remains to be shown what is the biological significance of higher expression of these SLRPs by SFBL in scar formation. 5.7.4. MMPs The remodeling phase of wound healing is considered critical for appropriate wound healing and is dependent on the bioactivity of MMPs (Gill and Parks 2008; Gurtner et al., 2008). Our results showed that expression of MMP-1, -3, and -10 was greater in GFBL compared to SFBL. Previous studies have shown that overexpression of DCN or FMOD in fibroblasts resulted in decreased protein production of MMP-1 and -3, while MMP-2 increased (Al Haj Zen et al., 2003; Stoff et al., 2007). Thus, lower MMP-1 and -3  !  +/!  expression in SFBL may be due to higher SLRPs expression seen in SFBL compared to GFBL in the present study. Moreover, MMP-1, -3 and -10 degrade major ECM proteins, including types I and III collagen, tenascin and proteoglycans (Matrisian et al., 1990; Lijnen and Collen, 1999; McCawley and Matrisian, 2001; Saunders et al., 2005). Interestingly, however, expression of MMP-7 and -11, was found to be significantly greater in SFBL. These MMPs do not predominantly degrade collagen, such as types I and III in the ECM. Instead, MMP-7 effectively degrades other ECM components, including elastin, fibronectin, gelatins, laminin-1, proteoglycans and type IV collagen (Miyazaki et al., 1990; Murphy et al., 1991; McCawley and Matrisian, 2001) and MMP11 degrades laminin, fibronectin and aggrecan (McCawley and Matrisian, 2001). Therefore, GFBL appear to have a higher capacity to remodel fibrillar ECM than SFBL. In addition, IL-6 stimulation failed to upregulate MMP-1 or -3 expression and protein levels in hypertrophic scar fibroblasts, compared to normal SFBL (Dasu et al., 2004). Therefore, scarring may be due to a lack of response to IL-6 or other MMP activators in fibroblasts to secrete MMPs. MMP expression and abundance in established scars has yet to be investigated. However, the presence of MMPs may not be essential for proper wound healing, since broad-spectrum MMP inhibitors or MMP-3, -9 and -13 KO mice demonstrated no significant effect on the remodeling phase and scar formation, while other events, including inflammation, re-epithelialization, myofibroblast differentiation were affected (Bullard et al., 1999; Mirastschijski et al., 2004; Hattori et al., 2009; Kyriakides et al., 2009). However, these studies were largely performed in loose skinned animals (rodents) where scar formation, unlike to tight skinned animals and humans, does not usually occur. While the role of MMPs in matrix remodeling and scar formation still remains unclear, MMPs may be more important in mediating other biological events in wound healing, through regulating the bioavailability and activity of cytokines and growth factors embedded in the ECM. For instance, MMPs can directly activate growth factors and cytokines, including TNF-", hepatoma-derived growth factor, HB-EGF, IL-8 (Gearing et al, 1994; Rodríguez et al., 2010). In addition, MMPs can indirectly activate cytokines and  !  +&!  growth factors. For instance, MMP-1 and -3 can cleave the proteoglycan perlecan, which ultimately leads to the release of FGF-2 bound to the proteoglycan (Whitelock et al., 1996; Mongiat et al., 2001). In addition, MMP-2, -3, and -7 can release TGF-! through DCN cleavage and MMP-3 can activate TGF-! by LTBP cleavage (Imai et al., 1997; Maeda et al., 2002; Mott and Werb, 2004). Also, MMP species can cleave and inactivate chemokines, including CXCL12, IL-8 and monocyte-specific chemokine 3 (Rodríguez et al., 2010). Our analysis showed that there was considerable difference in expression of MMP species between GFBL and SFBL, thus these fibroblasts may have a different capacity to modulate cytokine and growth factor activity. As discussed above, during wound healing and scar formation, ECM remodeling may be mediated by proteolytic degradation outside of cells by MMPs. In addition, intracellular degradation of ECM proteins by Endo180-mediated endocytosis and lysosomal CTSK degradation may also play a role (Wienke et al., 2003; Holmbeck and Szabova, 2006). For instance, it was shown that expression of Endo180 was spatiotemporally upregulated during the remodeling stage of gingival wound healing (Honardoust et al., 2006). While MMPs were more highly expressed in GFBL, there was a greater expression of Endo180 and CTSK in SFBL. Interestingly, despite the remodeling properties of CTSK, high expression of CTSK is associated with young scars and keloids (Rünger et al., 2006). Therefore, the mechanism of ECM degradation and remodeling may be significantly different in the skin and gingival tissue. 5.8. Putative regulation of MMP and VEGF-" expression by ERK1/2 and AP-1 pathway Interestingly, the promoter regions of MMP-1, -3, -7, and -10 show conserved cis-acting regulatory elements, specifically the activator protein-1 (AP-1) binding site, whereas MMP-11 do not contain such elements (Benbow and Brinckerhoff, 1997; Westermarck and Kahari, 1999). Therefore, greater expression of MMP-11 in SFBL may be due to an increased activation of a signaling pathway that is different from the MMPs that were more expressed in GFBL (with the exception of MMP-7). Our results from the reseeding experiment support the notion that MMP-11 is differently regulated than the other MMPs. We found a significantly greater expression of MMP-1, -3, and -10 genes in both  !  +'!  GFBL and SFBL when these cells were seeded on 3D ECM compared to collagen type I substrate. Interestingly, ERK1/2 MAPK has been shown to activate c-Fos and in turn, cFos can stimulate AP-1 transcription factor activation (Deng and Karin 1994; González et al., 2008). We also found that relative to fibroblasts on collagen I substrate, there was a greater ERK1/2 phosphorylation in both GFBL and SFBL when seeded on 3D ECM. This may explain the greater expression of AP-1-dependent MMP-1, -3, -7, and -10 in GFBL and SFBL. On the other hand, MMP-11 that does not appear to contain an AP-1 binding site (Benbow and Brinckerhoff, 1997; Westermarck and Kahari, 1999), showed no difference in gene expression between different substrates. Thus, our findings suggest that the 3D ECM promoted the upregulation of AP-1-dependent MMPs. In addition, despite not being statistically significant, gene expression for VEGF-" was greater in both GFBL and SFBL when seeded on 3D ECM compared to fibroblasts seeded on collagen I substrate. Interestingly, it has been shown that VEGF expression can be mediated by ERK1/2 signaling in fibroblasts (Wu et al., 2004) and expression may be induced through AP-1 binding site in the promoter region (Cho et al., 2006). Thus, the 3D ECM may have promoted greater AP-1 activation by the ERK1/2 pathway compared to collagen I substrate, which stimulated the greater expression of AP-1-dependent MMP and VEGF genes seen in this present study.  !  +(!  CHAPTER SIX – CONCLUSIONS AND FUTURE DIRECTIONS 6.1. Conclusions In summary, our findings demonstrated a distinct gene expression profile between GFBL and SFBL in 3D cultures. Specifically, pro-fibrotic growth factors, cell contractility and myofibroblast associated genes and ECM deposition proteins were more highly expressed in SFBL, while GFBL showed higher expression of immunomodulatory and matrix remodeling MMPs. In addition, SFBL had greater autogenous TGF-! expression and SMAD3 phosphorylation compared to GFBL. These factors may partly explain why expression of the majority of matrix deposition associated genes was greater in SFBL compared to GFBL, regardless of the culture growth conditions. Therefore, SFBL are “pre-programmed” to express these ECM deposition proteins in greater levels than GFBL. We also showed that the 3D cell cultures promoted a significant difference in gene expression profile between GFBL and SFBL compared to reseeding of cells on the cellfree 3D ECM or 2D control substrate. Therefore, GFBL and SFBL had a distinct response to the cellular microenvironment in 3D culture that includes ECM, ECMassociated growth factors and culture medium supplements (serum and ascorbic acid). This difference may be stemmed from the cells interaction with the surrounding ECM niche, including, cell-ECM integrin-mediated signaling, growth factor and cytokine signaling, distinct matricellular protein composition in the ECM, and mechanical strain. We also found that ERK1/2 phosphorylation was greater in both GFBL and SFBL when seeded on 3D ECM compared to 2D collagen substrate. This may have led to an upregulation of AP-1-dependent gene transcription, particularly MMPs and VEGF-". These findings, specifically reduced expression of matrix deposition associated genes and greater expression of matrix remodeling genes in GFBL may contribute to scarless healing in gingiva as compared to scar formation in skin.  !  +)!  6.2. Future directions We will investigate the underlying cause for the distinct gene expression in GFBL and SFBL. First, we will determine whether differences in gene expression profiles are due to a greater senescence-like phenotype in GFBL than in SFBL. In addition, we will analyze the role of TGF-! signaling pathways on gene expression in GFBL and SFBL. We will also determine if the Wnt/!-catenin pathway and/or embryonic origin accounts for these differences between GFBL and SFBL. Lastly, we will further characterize the composition of the GECM and SECM. 6.2.1. Senescence phenotype •  Fibroblasts in senescence phase are characterized by greater expression of MMPs genes and lower expression of ECM deposition associated genes (Jun and Lau, 2010; Pitiyage et al., 2011). Therefore, we will determine whether differences in gene expression profiles are a result of a greater level of senescence-like phenotype in GFBL than in SFBL. To analyze the level of senescence, senescence-associated !-galactosidase staining will be performed on passage-matched GFBL and SFBL parallel cell lines. 6.2.2. Autogenous TGF-! signaling  •  Determine whether there is a greater synthesis and secretion of TGF-! in SFBL compared to GFBL in 3D cell cultures, by analyzing total TGF-! abundance using biochemical techniques, such as ELISA.  •  Determine the effects of blocking TGF-! signaling using SB-431542 on gene expression in GFBL and SFBL. Analyze whether blocking TGF-! mediated signaling promotes downregulation of ECM deposition associated genes and upregulation of matrix remodeling associated genes in SFBL, similar to GFBL expression levels. 6.2.3. Downstream mediators of TGF-! signaling that are associated with fibrosis If TGF-! signaling is involved in promoting these differences in gene expression, we will explore pathways further downstream such as SMAD, TAK1 and ERK.  •  Inhibit SMAD activation by siRNA or SIS3 and analyze the effects on gene expression in GFBL and SFBL. !  +*!  •  Determine if there is a difference in phosphorylation of TAK1, an intracellular mediator of TGF-! signaling that specifically regulates fibrosis associated genes, including TGF!1, collagen and CTGF (Guo et al., 2013) in GFBL and SFBL. Investigate the effects of inhibiting TAK1 with 5Z-7-oxozeaenol on gene expression in GFBL and SFBL.  •  Determine if TGF-! signaling affects ERK1/2 pathway by blocking TGF-! signaling via SB-431542 and analyze phosphorylation of ERK1/2. Furthermore, if ERK1/2 is related to TGF-! signaling in GFBL and SFBL, analyze the effects of blocking ERK1/2 via PD98059/PD184161 on gene expression profiles of GFBL and SFBL.  •  If the pathways mentioned above affect differential gene expression in GFBL and SFBL, we will further investigate early growth receptor factor 1/2 (EGR1/2) which is a downstream mediator of ERK1/2 signaling and regulates expression of profibrotic genes (Bhattacharyya et al., 2008) and p38, JNK, c-Fos and c-Jun activation, which are intracellular mediators downstream of TAK1 (Moriguchi et al., 1996; Shirakabe et al., 1997).  6.2.4. Effect of Wnt/!-catenin pathway and fibroblast origin on gene expression •  Wnt-!-catenin overactivity has been linked to scar formation in vitro and in vivo (Bielefeld et al., 2012; Hamburg and Atit, 2012; Profyris et al., 2012). This pathway is also important during development and interacts with the TGF-! signaling pathway (Lam and Gottardi, 2011). The importance of this pathway in determining differential gene expression in SFBL and GFBL will be studied by assessing the expression and/or phosphorylation of key components of this pathway (e.g. !-catenin, Wnts, frizzled, LRP 5/6, GSK3) and by using specific antagonists to modulate this pathway as above.  •  We will analyze phenotypic differences between fibroblasts from similar embryonic origins. Previous work has compared the global gene expression patterns of fibroblasts derived from different tissues and sorted these fibroblasts by hierarchical clustering (Chang et al., 2002). Based on this analysis, we will compare expression in gingival and toe fibroblasts, as they were the most similar in the hierarchical clustering analysis. Comparing dermal papilla cells (which are derived from the neural crest) to GFBL may be an option, but dermal papilla cells are not readily available.  !  ++!  6.2.5. Further characterization of the 3D ECM derived from GFBL and SFBL •  Determine growth factor and cytokine abundance in 3D cultures using biochemical techniques, such as Western blotting and ELISA.  •  Analyze total molecular composition of the 3D ECM using mass spectrometry. 6.2.6. Long-term goals  •  Analyze the biochemical and cellular composition of gingiva and skin in vivo, by using the pig wound healing and scar formation model available in the laboratory (Mak et al., 2009; Wong et al., 2009), specifically investigating the components that were found to be different  in  the  present  study  through  biochemical  techniques,  such  as  immunohistochemistry and Western blotting, gene expression profiling using microarrays, or with the use of mass spectrometry and proteomics analysis. •  Contribute to the understanding of the molecular mechanisms of scar formation and contribute to the development of novel anti-scarring therapies that would improve both the physical and psychosocial complications that are associated with scars.  !  +,!  7.1. TABLES AND FIGURES  !  +-!  Table 1. In vitro studies comparing fibroblasts derived from skin and various regions of the oral mucosa. Modified from H#kkinen et al., 2012. Adult oral mucosal fibroblasts Adult skin fibroblasts Refs Longer proliferative lifespan.  Shorter proliferative lifespan.  Greater proliferation rate.  Lower proliferation rate.  Higher expression of MMP-2, -3.  Lower expression of MMP-2, -3.  Stephens et al., 2001; McKeown et al., 2007  Faster fibroblast migration.  Slower fibroblast migration.  Enoch et al., 2010  Higher expression of HGF and KGF.  Lower expression of HGF and KGF.  Okazaki et al., 2002; Shannon et al., 2006  TGF-! suppresses proliferation.  TGF-! promotes proliferation.  Meran et al., 2011  Differences in sulphated glycosaminoglycan abundance and size of dermatan sulphate proteoglycans.  !  +.!  Enoch et al., 2009 Schor et al., 1996; Lee and Eun, 1999  Bronson et al., 1988; Larjava et al., 1988  Table 2. Matricellular proteins that have a role in the wound healing process and are synthesized by fibroblasts. Matricellular protein  Role in wound healing  Refs  Anti-adhesive protein that promotes fibroblast migration within fibrillin-fibronectin matrices. Affects fibronectin deposition. Promotes wound contraction.  When it appears during wound healing Upregulated at the wound margins during granulation tissue formation.  Tenascin-C  Tenascin-X  Increases strength and integrity of connective tissue. KO showed reduced breaking stress of wounds without any evident effect on collagen fibrils and wound closure rate.  Upregulated during granulation tissue formation.  H!kkinen et al., 2012  Periostin  Regulates collagen type I fibrillogenesis and maturation during remodeling phase of wound healing. Influences cell-ECM interaction, as well as cell adhesion, proliferation and differentiation.  Upregulated during ECM remodeling and wound healing.  Kudo et al., 2007; Hamilton, 2008; Jackson-Boeters et al., 2009; Zhou et al., 2010; H!kkinen et al., 2012  Thrombospondin-1 (THBS-1)  Anti-angiogenic factor. Reduces MMP levels in the ECM. Activates TGF-"1.  Upregulated during granulation tissue formation.  Kyriakides and Bornstein 2003; H!kkinen et al., 2012  Thrombospondin-2 (THBS-2)  Anti-angiogenic factor. Regulates collagen fibrillogenesis and platelet formation. Reduces MMP levels in the ECM. Regulates fibroblast-ECM interactions to promote ECM remodeling.  Upregulated during granulation tissue formation.  Schiro et al., 1991; Kyriakides and Bornstein 2003; H!kkinen et al., 2012  SPARC  Regulates ECM deposition and organization and angiogenesis. Anti-adhesive protein, promotes cell migration. Inhibits cell cycle progression.  Upregulated upon injury.  Kyriakides and Bornstein, 2003; H!kkinen et al., 2012  CCN Family  Mediates cell-ECM interactions and regulates cell migration, adhesion, differentiation and survival, as well as matrix deposition.  Upregulated during granulation tissue formation.  Jun and Lau 2010; H!kkinen et al., 2012  Hevin  Regulates ECM deposition and organization. Anti-adhesive protein, suppresses cell migration.  Osteopontin  Regulates ECM reorganization during wound healing. KO mice display more disorganized ECM and altered collagen fibrillogenesis, during wound repair.  !  Bornstein and Sage 2002; Wong et al., 2009; Midwood et al., 2004; H!kkinen et al., 2012  H!kkinen et al., 2012  "#!  Upregulated during inflammatory phase.  Liaw et al., 1998; Kyriakide and Bornstein, 2003  Table 3. Key growth factors involved in wound healing and its ability to interact with PCM molecules. Growth Factors  Cell type released Platelets Macrophages Fibroblasts Keratinocytes Endothelial cells Platelets Lymphocytes Fibroblasts Macrophages Keratinocytes  Interaction with PCM Collagen IV, V Fibronectin SPARC Hyaluronan  Role in wound healing  Refs  Acts as a chemokine for cell migration of neutrophils, monocytes, and fibroblasts during inflammatory phase. Increases proliferation and ECM synthesis in fibroblasts for granulation tissue formation. Stimulates fibroblasts to contract collagen matrices and induces myofibroblast differentiation during granulation tissue formation. Potent stimulators of expression of ECM proteins and cell proliferation in fibroblasts. Acts as a chemokine for neutrophils, monocytes and fibroblasts during inflammatory phase. Stimulates angiogenesis, myofibroblast cell differentiation. In addition, stimulates re-epithelialization by promoting keratinocyte migration, but inhibits their proliferation.  Macri et al., 2007; H!kkinen et al., 2012  TGF-"3  Fibroblasts Keratinocytes  Biglycan Decorin Collagen IV Fibronectin SPARC  Decreases collagen and other ECM protein deposition.  Ferguson and O’Kane, 2004; Macri et al., 2007  CTGF  Fibroblasts  Heparan sulphatecontaining proteoglycans  Promotes proliferation, ECM production and migration in fibroblasts. Promotes endothelial cell proliferation, migration, survival, adhesion and angiogenesis.  Shi-Wen et al., 2008; H!kkinen et al., 2012  KGF (FGF-7)  Fibroblasts  Regulates keratinocyte proliferation and migration, therefore has a role in re-epithelialization.  EGF  Platelets Fibroblasts Mast cells Plasma  Regulates keratinocyte proliferation and migration for reepithelialization. Acts as a chemokine for endothelial cells for angiogenesis. Promotes proliferation, collagenase expression, migration and modulation of ECM deposition.  Werner and Grose, 2003; Koivisto et al., 2006; Werner and Grose 2003; Koivisto 2006; H!kkinen et al., 2012  HB-EGF  Macrophages Keratinocytes  Stimulates proliferation in fibroblasts and keratinocytes, therefore plays a role in re-epithelialization and granulation tissue formation. In addition, stimulates MMP-1 expression to help with keratinocyte migration.  Werner and Grose 2003; Koivisto et al., 2006  HGF  Keratinocytes  Promotes keratinocyte migration, proliferation, and matrix metalloproteinase production. Stimulates angiogenesis.  Werner and Grose 2003; Macri et al., 2007  PDGF  TGF-"1, -"2  !  Biglycan Decorin Collagen IV Fibronectin SPARC  Fibronectin Heparan sulphate/heparin Vitronectin  "#!  Werner and Grose, 2003; Macri et al., 2007; H!kkinen et al., 2012  Growth Factors  Cell type released Fibroblasts  Interaction with PCM  VEGF-#  Fibroblasts Keratinocytes Macrophages  FGF-2  Fibroblasts Macrophages Mast Cells Keratinocytes Endothelial cells  Stromal-derived factor (SDF-#) (CXCL12)  !  Role in wound healing  Refs  Promotes recruitment of hematopoietic and mesenchymal stem cells to the site of the wound. Expression is elevated during later stages of the wound healing and is more prominent at the wound margins.  Fedyk et al., 2001; Toksoy et al., 2007  Fibronectin Heparan sulphate/ heparin SPARC  Promotes angiogenesis during granulation tissue formation.  Werner and Grose, 2003; Macri et al., 2007; H!kkinen et al., 2012  Fibronectin Heparan sulphate/heparin SPARC  Promotes angiogenesis during early stages of wound healing by stimulating migration and proliferation of endothelial cells. Promotes proliferation, migration, ECM deposition and collagenase synthesis in fibroblasts.  Werner and Grose, 2003; Macri et al., 2007; H!kkinen et al., 2012  "$!  Table 4. List of target proteins and indirect activation of MMPs, cytokines and growth factors by MMPs. The target proteins are not limited to the list above. Table is a modification from McCawley and Matrisian, 2001. The MMPs listed have been shown to be expressed by fibroblasts.1 MMP  Target substrate  Non-traditional targets  MMP-1 (Collagenase-1)  Collagen I,II,III,VII, X, gelatin, entactin, aggrecan, tenascin  Perlecan Pro-MMP-1 Pro-MMP-2 Pro-TNF-#  MMP-2 (Gelatinase A)  Collagen I,IV,V,VII,X,XI, elastin, fibronectin, laminin, aggrecan vitronectin  Decorin Pro-TGF-B2 Pro-MMP-1 Pro-MMP-2 Pro-MMP-13  MMP-3 (Stromelysin-1)  Collagen I,III,IV,V, IX,X,XI, entactin, tenascin, Decorin vitronectin, fibrin/fibrinogen Perlecan Plasminogen Pro-MMP-1,3,7,8,9,13 Pro-TNF-#  TGF-" FGF Angiostatin MMP-1,3,7,8,9,13 TNF-#  MMP-7 (Matrilysin)  Collagen III,IV,V,IX,X,XI, proteoglycans, laminin, fibronectin, gelatin, tenascin, vitronectin  Decorin Pro-MMP-2 Pro-MMP-7 Pro-TNF-# Plasminogen  TGF-" MMP-2 MMP-7 TNF-# Angiostatin  MMP-10 (Stromelysin-2)  Collagen III,IV,V, IX,X,XI, entactin, tenascin, vitronectin, fibrin/fibrinogen  Pro-MMP-1 Pro-MMP-8 Pro-MMP-10  MMP-1 MMP-8 MMP-10  MMP-11 (Stromelysin-3)  Aggrecan, laminin, fibronectin  MMP-12 (Metalloelastase) Elastin, fibronectin, laminin, proteoglycan, fibrin/fibrinogen  Plasminogen  Angiostatin  MMP-13 (Collagenase-3)  Pro-MMP-9 Pro-MMP-13  MMP-9 MMP-13 IL-8  Collagen I,II,III,VII,X, gelatin, entactin, aggrecan, tenascin  1  Cytokine/ growth factor/MMP activation FGF MMP-1 MMP-2 TNF-# IL-8 TGF-"1 TGF-"2 MMP-1 MMP-2 MMP-13  Armstrong and Jude, 2002; Inoue et al., 1995; Nagase 1999; Mäkelä et al., 1994; Vaalamo et al., 1999; Ravanti et al., 2001; Selvey et al., 2004; Tervahartiala et al., 2000; Saunders et al., 2005; Test et al., 2007; Gill and Parks 2008.  !  "%!  Table 5. List and origin of the cells used. Cell name GFBL-DC GFBL-OL GFBL-HN GFBL-DW GFBL-IE SFBL-2-C SFBL-1-2 SFBL-4-1 SFBL-302 SFBL-406 !  !  Origin Attached gingiva Attached gingiva Attached gingiva Attached gingiva Attached gingiva Caucasian breast dermis Caucasian breast dermis Caucasian breast dermis Caucasian breast dermis Caucasian breast dermis  "#!  Sex Male Male Female Female Male Female Female Female Female Female  Age of donor 41 year old 30 year old 18 year old 30 year old 26 year old 40 year old 44 year old 41 year old 38 year old 35 year old  Table 6. List of antibodies used for immunohistochemistry and Western blotting. Antibody Product Source Dilution Immunostaining Western blotting Anti-human BGN Abnova Corp.; Mouse 1:100 1:500 Taipei, Taiwan Anti-human DCN R&D systems Inc.; Mouse 1:500 1:500 Minneapolis, MN Anti-human FMOD (H-50) Santa Cruz; Rabbit 1:100 1:1000 Santa Cruz, CA Anti-human LUM (H-90) Santa Cruz; Rabbit 1:100 1:1000 Santa Cruz, CA Anti-bovine collagen Biodesign; Saco, ME Rabbit 1:300 type I Anti-human Sigma; St. Louis, Rabbit 1:1000 fibronectin (plasma & MO cytoplasmic) Anti-human Abcam Inc.; Mouse 1:100 1:400 fibronectin-EDA Cambrige, MA (cytoplasmic) Anti-human tenascin- Sigma; St. Louis, Mouse 1:4000 1:4000 C MO Anti-human !-SMA Sigma; St. Louis, Mouse 1:100 MO Anti-" tubulin Abcam Inc.; Goat 1:5000 (ab21057) Cambrige, MA Anti-human actin Abcam Inc.; Rabbit 1:5000 (ab8227) Cambrige, MA Anti-ACTIVE® p38 Promega; Madison, Rabbit 1:2000 (pTGpY) WI, USA Anti-human p38 Cell Signaling Tech.; Mouse 1:1000 MAP kinase (L53F8) Danvers, MA Anti-human ERK1 Abcam Inc.; Rabbit 1:500 (ab7947) Cambrige, MA Anti-active MAPK Abcam Inc.; Rabbit 1:2000 (ERK1) (pTEpY) Cambrige, MA Anti-human SMAD3 Abcam Inc.; Rabbit 1:2000 (ab28379) Cambrige, MA Anti-human phospho- Abcam Inc.; Rabbit 1:2000 SMAD3 Cambrige, MA (ab52903)  !  "#!  !"#$%&'(&)*+,%*-&.-%/&01*&*%"$23+,%&)45&"6"$7-+-(& Structural ECM GenBank Target Gene Primer sequence Number  Amplicon  BC036531  Collagen type I (alpha 1)  AACCAAGGCTGCAACCTGGA GGCTGAGTAGGGTACACGCAGG  Forward Reverse  3951-3970 4030-4009  80  NM_000090  Collagen type III (alpha 1)  CTCCTGGGATTAATGGTAGT CCAGGAGCTCCAGGAAT  Forward Reverse  1271-1290 1340-1324  70  NM_000501  Elastin Fibrillin-1  NM_007046  Emilin-1  NM_032048  Emilin-2  NM_052846  Emilin-3  NM_212482  EDA-fibronectin  NM_212482  EDB-fibronectin  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  174-191 236-219 8577-8595 8679-8660 3162-3183 3226-3208 3513-3532 3600-3580 962-984 1033-1013 5633-5652 5700-5680 4168-4188 4232-4206  62  NM_000138  GGGTCCCTGGGGCCATTC CAGGCCCCAGCGCTGGAT AGGAAACGGAGAAGCACAA CTGTCTTCTCAACATCCCAA CCTTCGACAGAGTCCTGCTCAA CGCTGTGAACACGCCTGTC GTGAACATGGCCACTGACTT GCCCTCAGAGTGTAGATACAG TCCCTTAGACGAGATCCTAAGCA TCTAGAAGCTGCACCTTGGTC CACAGTCAGTGTGGTTGCCT CTGTGGACTGGGTTCCAATCA CAGTAGTTGCGGCAGGAGAA GTATCCTACTGAGGAGTCCACAAAATC  Matricellular protein GenBank Number NM_003246  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  Thrombospondin1  CATCTTGTTCTGTGACATGTGG TTCACAGGGTTTCCCGTTC  Forward Reverse  1513-1534 1604-1586  92  Thrombospondin2  TTGTGTTCAACCCAGACCA CAGGACACACATCATCAATATC "#!  Forward Reverse  2989-3007 3096-3075  107  NM_003247 !  Orientation Location  103 65 87 72 68 65  NM_003118  SPARC-1  NM_002160  Tenascin-C  NM_019105  Tenascin-X  NM_000582  Osteopontin  NM_006475  Periostin  NM_001901  CTGF/CCN-2  NM_001554  CCN-1  NM_002514  CCN-3  NM_001128310  Hevin-1  NM_001128310  Hevin-2  Small LeucineRich Proteoglycan GenBank Number NM_001711 BT019800  !  AGCAAGAAGCCCTGCCTGA TCCTACTTCCACCTGGACAGGATT CAACCTGATGGGGAGATATGGGGA GAGTGTTCGTGGCCCTTCCAG CCTGGAGCCAGACCATAAATACAAG CGCTGGCCACCGTGGAA CAGTTGCAGCCTTCTCAG CTAGGAGGCAAAAGCAAATC ACACGAGAAGAACGAATCAT GTAACAATTTCTTCAGAGTTTCTTC ATGATGTTCATCAAGACCTGTGCCTG CTTCCTGTAGTACAGGGATTCAAAGAT GTC CATGATGATCCAGTCCTGCAA TGTCATTGAACAGCCTGTAGAAG CTCAGATCTGGAGCCATGCGA CGTGCAGATGCCAGTCTGGTTG CTGCAGCTGCAATCCCGA TCTTCAGCTTCAGCCCTTAAACT CCAGCATCCTCCTCTGTTC TGGTGCATAGGTTGTTGTCA  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  164-182 264-241 6769-6792 6846-6826 6334-6358 6390-6374 118-136 199-180 199-181 199-182 199-183 199-184  101  Forward Reverse Forward Reverse Forward Reverse Forward Reverse  199-185 199-186 199-187 199-188 199-189 199-190 199-191 199-192  86  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  Biglycan (BGN)  CTCAAGCTCCTCCAGGTGGTC CCGAAGCCCATGGGACAGAAGTC CTGACACAACTCTGCTAGAC GACAAGAATCAATGCGTGAAG  Forward Reverse Forward Reverse  93  Decorin (DCN)  ""!  1067-1087 1151-1127 242-261 339-319  75 57 82 97 80  85 115 132  97  NM_002023  Fibromodulin (FMOD)  CACAATGAGATCCAGGAAG TCCGAAGGTGGTTATAACTC  Forward Reverse  761-779 845-826  85  BT006707  Lumican (LUM) Asporin (ASPN)  Forward Reverse Forward Reverse  635-658 720-700 1183-1213 1244-1264  85  NM_017680  TAGACAACAATAAGATCAGCAACA TTCGTTGTGAGATAAACGCAG CCTCCAGATAATCTTCCTTCATTCTA CTTCATCTTTGGCACTGTTGG  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  TGF-!1  CAACGAAATCTATGACAAGTTCAAGC AG CTTCTCGGAGCTCTGATGTG TGGTGAAAGCAGAGTTCAGAG CACAACTTTGCTGTCGATGTAG ACACCAATTACTGCTTCCGCAA GCCTAGATCCTGTCGGAAGTC GTGTATAGCTGAAATTGACTTAA TGATTGCAGCAATATGTTGTA  Forward  1218-1245  76  Reverse Forward Reverse Forward Reverse Forward Reverse  1294-1275 1883-1903 2022-2001 1161-1182 1242-1220 286-308 384-364  CTGGTGAGACTTTCTTCATGTG CTGATGCCTGTCACTTGAAA AGTGTGTGCCCACTGAGGA GTGCTGTAGGAAGCTCATCTC AGTTGGTATGTGGCACTGAA GTATAGCTTTCTGCCCAGGTC TACAGATGCCCATGCCGA CTGAAGGGCACAGTTTGGAG TGGAAGGACTTTGTGAAGGA AATTCCATCCAGTAGAAGCATC  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  768-789 894-875 1316-1334 1413-1393 831-850 886-906 174-191 266~247 689-708 793-772  Growth Factors GenBank Number NM_000660  !  NM_003238  TGF-!2  NM_003239  TGF-!3  NM_001130916  TGF-! receptor 1 (TGF-!R1)  NM_003242 NM_001171630  TGF-! receptor 2 (TGF-!R2) VEGF-"  NM_002006  FGF-2  NM_199168  CXCL12/SDF-1"  NM_138455  Cthrc-1  "#!  81  140 81 99  127 97 75 93 104  Myofibroblastassociated and cell contractilityassociated genes GenBank Number NM_001613  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  "-smooth muscle actin (" –SMA)  AGCGTGGCTATTCCTTCGT CTCATTTTCAAAGTCCAGAGCTACA  Forward Reverse  637-655 733-707  97  NM_001142483  P311 "11 integrin  NM_002473  NM_001101  Non-muscle Myosin IIA Non-muscle Myosin IIB !-actin  NM_001614  !-Actin  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  321-447 508-489 1144-1163 1204-1186 722-741 803-783 395-418 493-472 254-270 320-303 134-155 218-199  78  NM_001004439  CCTGGACTGAAGAGAGG CAGACAAAGAGTTCTGGGTA GAAGGCACCAACAAGAACGA AGGAAAAGCCCGTCTGTGA ACCGAGAAGATCAATCCATC AGATACTGGATGACCTTCTTG CCGTTTTACATAATCTGAAGGATC TTGGAAGATTCTTGTAAGGGTT AGGCCCAGAGCAAGAGA CAGTTGGTGACGATGCCG GTCGCAATGGAAGAAGAGATCG AGCGTCGTCCCCAGCAAAA  Intracellular degradation GenBank Number NM_000396  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  Cathepsin K (CTSK)  TCGACTATCGAAAGAAAGGATA AAAGCCCAACAGGAACCA  Forward Reverse  585-606 655-637  70  Endo180 (CD280)  AAGAGGCCCAGCTGGTCA GCATGGAGGCCAATCCAAAG  Forward Reverse  3144-3161 3243-3224  100  NM_005964  AF134838  !  "$!  60 81 98 72 85  NM_002332  !  LRP1 (lipoprotein ACGCCTCTGACGTGGTCCT receptor-related protein 1) CATTTCTTGCGGTCACATGGGTTG  Forward  MMPs and their Inhibitors GenBank Number NM_002421  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  MMP-1 MMP-2  NM_001166308  MMP-3  NM_214207  MMP-7  NM_002425  MMP-10  NM_005940  MMP-11  NM_002426  MMP-12  NM_002427  MMP-13  NM_003254  TIMP-1  NM_003255  TIMP-2  NM_000362  TIMP-3  NM_003256  TIMP-4  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  100  NM_001127891  GCTAACAAATACTGGAGGTATGATG GTCATGTGCTATCATTTTGGGA AGACATACATCTTTGCTGGAG ATCTGCGATGAGCTTGG ATGATGAACAATGGACAAAGGA GAGTGAAAGAGACCCAGGGA TATGCTGCAACTCATGAAC CGTAGGTTGGATACATCAC TTATACACCAGATTTGCCAAGA TTCAGAGCTTTCTCAATGG CTATCCTCCAAAGCCATTGTAA CAACTGTGTTTAATGACAATCCTC GTATGATGAAAGGAGACAGATGAT TACGTTGGAGTAGGAAGTCAT CAGGAATTGGTGATAAAGTAGAT CTGTATTCAAACTGTATGGGTC CTGTGTCCCACCCCACC GAACTTGGCCCTGATGACGA ACATTTATGGCAACCCTATCAA TCAGGCCCTTTGAACATCTTTA AGGACGCCTTCTGCAAC CTCCTTTACCAGCTTCTTCC ACCTGTCCTTGGTGCAGA TGTAGCAGGTGGTGATTTGG #%!  Reverse  1286112879 1293312910  1250-1275 1304-1325 1900-1920 1988-1972 661-682 751-732 722-740 804-785 394-415 450-432 2258-2179 2254-2231 1303-1326 1469-1449 1302-1324 1365-1386 267-283 330-311 481-502 550-529 1281-1297 1348-1329 927-944 1004-985  73  88 91 82 56 106 166 85 64 70 68 80  Housekeeping genes GenBank Number NM_002046  Target Gene  Primer/Probe sequence  Orientation Location  Amplicon  GAPDH HPRT1  NM_021009  UBC  NM_004048  B2M  BC009255  ALG9  Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse  70  M31642  CTTTGTCAAGCTCATTTCCTGGTA GGCCATGAGGTCCACCA TGTTGGATTTGAAATTCCAGACAAG CTTTTCCAGTTTCACTAATGACACAA GTGGCACAGCTAGTTCCGT CTTCACGAAGATCTGCATTGTCA TGTCTTTCAGCAAGGACTGGTCTTTC ATGGTTCACACGGCAGGCATA GAATGACCAGAATCTAGAAGAGCCA TCTCATGGTGTCCAAATCCACTAAA  !  !  #&!  1020-1043 1089-1073 619-643 727-700 371-389 444-467 281-306 351-372 1668-1692 1749-1725  107 96 92 82  Table 8. Summary of the gene expression analysis of the genes that were significantly different between GFBL and SFBL in 3D cultures 7 days post-seeding. Predetermined: Genes that were significantly different between GFBL and SFBL, irrespective to the culture conditions (3D cultures 7 days post-seeding, 24 h after reseeding on cell-derived 3D matrix or collagen I substrate, with or without FBS and ascorbic acid). Therefore, the expression of these genes is innately different between GFBL and SFBL, and does not depend on the environment (culture substrate and culture medium). Interestingly, the majority of these genes that showed significantly greater expression in SFBL were matrix deposition associated genes. 3D cultures 7 days postseeding: Genes that were only significantly different between GFBL and SFBL in 3D culture 7 days post-seeding. Differential expression of these genes is therefore dependent on the 3D culture environment. Overall, most matrix remodeling MMPs, ECM internalization and intracellular degradation-associated genes (Endo180 and CTSK) and various matricellular protein genes (SPARC, THBS-2, OPN) were found in this category. Serum response: Genes that showed significant differences between GFBL and SFBL only in response to serum (3D cultures 7 days post-seeding and 24 h after reseeding on collagen I substrate with the presences of FBS and ascorbic acid). Expression of CTGF was significantly affected by serum and was greater in SFBL than GFBL. Statistical analysis was performed using Student’s t-test. GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM). Predetermined Genes (Significantly different in 3D culture 7 days post-seeding and reseeding experiment 24 h post-seeding)  Target Gene  Collagen I Collagen III Elastin-1 Tenascin-C Tenascin-X DCN FMOD BGN LUM MMP-11 TIMP-4 TGF-"3 TGF-"2 VEGF-! P311 !11 integrin  GFBL relative to SFBL ! ! ! ! ! ! ! ! ! ! " ! " " ! !  Collagen  GECM  SECM  Target Gene  P<0.05 P<0.05 0.06 P<0.05 P<0.001 P<0.01 P<0.05 P<0.05 P<0.05 P<0.05 P<0.05 P<0.05 P<0.001 P<0.05 P<0.05 P<0.05  P<0.05 P<0.05 P<0.05 0.13 P<0.05 0.08 P<0.05 P<0.05 0.08 P<0.01 P<0.05 0.14 0.08 0.08 P<0.05 P<0.05  0.05 P<0.05 P<0.05 0.09 P<0.05 0.08 P<0.01 P<0.05 0.06 P<0.01 0.06 0.14 P<0.05 0.08 P<0.05 P<0.05  !-SMA MMP-1 MMP-3 MMP-7 MMP-10 TGF-"1 SPARC THBS-2 OPN Endo180 CTSK NMMIIA NMMIIB  !  !  3D Cultures 7 Days Post-seeding  "#!  GFBL relative to SFBL ! " " ! " ! ! ! ! ! ! ! !  Serum Response (3D culture 7 days postseeding and 24h after seeded on collagen I in the presence of FBS)  Target Gene CTGF  GFBL relative to SFBL !  Table 9. Gene expression analysis between fibroblasts seeded on 3D ECM compared to collagen type I substrate. Expression of MMP-1, -3 and -10 was significantly greater when GFBL and SFBL were seeded on 3D ECM compared to collagen I substrate. Arrow indicates that expression of target gene was greater on GECM/SECM relative to collagen substrate in GFBL and SFBL. Values indicated in the table represent p-values (* p<0.05, **p<0.01; Student’s t-test). GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM). GFBL Target gene  SFBL  GECM vs collagen substrate  Relative to collagen substrate  SECM vs collagen substrate  Relative to collagen substrate  GECM vs collagen substrate  Relative to collagen substrate  SECM vs collagen substrate  Relative to collagen substrate  MMP-1  0.18  !!  0.29  !!  0.01*  !!  0.06  !!  MMP-3  0.21  !!  0.16  !!  0.04*  !!  0.09  !!  MMP-10  0.04*  !!  0.15  !!  0.004**  !!  0.008**  !!  VEGF-!  0.10  !!  0.10  !!  0.06  !!  0.15  !!  !  "#!  Table 10. Summary of TGF-!-dependent genes differentially expressed by GFBL and SFBL. Greater expression of fibrillar ECM proteins and tenascin-C, tenascin-X and SLRPs in SFBL may be attributed to increased activation of endogenous TGF-" signaling and phosphorylation of SMAD3. Genes that have been previously shown to be regulated by TGF-" are shown. !": Expression of the TGF-" responsive gene was higher or lower in SFBL relative to GFBL; = : expression of the TGF-" responsive gene was the same in GFBL and SFBL. Greater SMAD3 phosphorylation in Genes that may not be triggered by SFBL may have resulted in  increased SMAD phosphorylation1  Collagen I  !!  OPN  %!  Collagen III  !  SPARC  %!  Tenascin-C  !!  THBS-2  %!  Tenascin-X  !!  TGF-"1  %!  Elastin-1  !  TGF-"2  "!!  DCN  !  TIMP-4  "!!  BGN  !  VEGF-!  "!!  FMOD  !  !-SMA  %  LUM  !!  CTSK  %!  TGF-"3  !!  Endo180  %!  !  MMP-1, -3, -10, -7  %!  !  MMP-11  !!!  ! 1  Gene expression difference is opposite to published TGF-" response or there is no difference in the expression of TGF-"-response genes between GFBL and SFBL.  !  "$!  !!!!!!!  ! Fig. 1. A schematic presentation of the role of ECM components on cell function. The fibrillar ECM provides mechanical stimuli to the cells and acts as a ligand for receptor-mediated signaling. Other components of the ECM including glycoproteins, proteoglycans and matricellular proteins modulate (can promote or inhibit) cell-ECM interactions through binding to ECM proteins, cell surface receptors and molecules such as proteases and growth factors.  !  "&!  !!!!!!!  Fig. 2. Key small leucine-rich proteoglycans (SLRPs) that regulate wound healing. Asporin (ASPN), biglycan (BGN) and decorin (DCN) belong in class I and fibromodulin (FMOD) and lumican (LUM) belong in class II of the SLRP family.  !  "'!  !!!!!!!  Fig. 3. Representative GFBL (A-D) and SFBL (E-H) images comparing morphology and organization in 3D cultures 7 days post-seeding. GFBL were more elongated and spindle-like compared to SFBL. This is well demonstrated from the shape of the nuclei seen in D and H. GFBL in 3D cultures were more polar and parallel to one another, while SFBL were organized in a more random fashion.  !  "#!  !!!!!!!  Fig. 4. SEM images of representative GFBL (A) and SFBL (B) in 3D cultures 7 days post-seeding, and of the 3D ECM after cell removal (C-F). (A,B) GFBL were more elongated and spindle-like compared to SFBL. (C-F) The fibrillar network of GFBLderived 3D ECM was organized parallel to one another, while SFBL-derived fibrillar network was organized in a more random fashion. (E-F) Tannic acid was also incubated in the fixation step to better preserve proteoglycans and glycoproteins in the ECM. (C-D) Without tannic acid the organization of the fibrillar network was better perceived.  !  ""!  !!!!!!!  Fig. 5. GFBL proliferated faster but produced equal amounts of total proteins into the 3D ECM over time as compared to SFBL. (A,B) Cell number at high density (3D) (A) and low density (B) cultures was assessed by measuring total RNA concentration (!g/!l) (A) and by MTT assay (B). (C) Total protein abundance (!g/!l) in 3D ECM of SFBL and GFBL cultures. (D) Protein abundance in 3D ECM relative to cell number (mg protein/!g RNA) between SFBL and GFBL 3D cultures. Protein assays (C,D) are representative of 3 repeated experiments. Results show mean +/- SEM. N=5 (A,C,D), N=3 (B) parallel cell lines. (*p<0.05; Student’s t-test).  !  "#!  !!!!!!!  Fig. 6. Greater expression of fibrillar ECM proteins in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, ***p<0.001; Student’s ttest).  !  #$!  Fig. 7. Greater expression of matricellular proteins in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01,***p<0.001; Student’s t-test). THBS-1: thrombospondin-1; THBS-2: thrombospondin-2. !  !  "#!  Fig. 8. Greater expression of small leucine-rich proteoglycans in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (**p<0.01, ***p<0.001; Student’s t-test). ASPN: asporin; BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican. !  !  "#!  !!!!!!!  Fig. 9. Greater expression of small leucine-rich proteoglycans in individual parallel SFBL compared to GFBL lines in 3D cultures 7 days post-seeding. Results show the relative proportion of individual SLRPs being expressed in GFBL and SFBL lines. DCN (indicated in yellow) was most highly expressed and FMOD (indicated in red) was the least expressed in all cell lines. Results were generated by 2(Ctreference-Cttarget) method. BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican.  !  "#!  !!!!!!!  Fig. 10. Greater expression of pro-fibrotic growth factors and Cthrc1 in SFBL compared to GFBL in 3D cultures 7 days post-seeding. (A) Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01, ***p<0.001; Student’s t-test). (B) Results show the relative proportion of TGF-!1, -!2 and -!3 in individual parallel GFBL and SFBL lines. In general, expression of total TGF-! was greater in SFBL compared to parallel GFBL. On average, TGF-!3 expression made up 2% and 8% out of total TGF-! expression in GFBL and SFBL, respectively. Results were generated by 2(Ctreference-Cttarget) method. TGF-!R1: TGF-! receptor 1; TGF-!R2: TGF-! receptor 2; CXCL12: stromal-derived factor 1; Cthrc1: collagen triple helix repeat containing 1. !  "$!  Fig. 11. Expression of MMPs and TIMP-4 was different in GFBL compared to SFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01; Student’s t-test). !  !  "#!  !!!!!!!  Fig. 12. Expression of MMPs was greater in GFBL compared to SFBL in 3D cultures 7 days post-seeding. Results show the relative proportion of individual MMPs being expressed in individual parallel GFBL and SFBL lines. Note that MMP-1 and -10 (indicated in navy and light blue) were most highly expressed out of all MMPs studied in four out of five GFBL lines and MMP-2 (indicated in red) was most abundant in all SFBL lines. Results were generated by 2(Ctreference-Cttarget) method.  !  "#!  !!!!!!!  Fig. 13. Expression of cell contractility and myofibroblast associated genes was greater in SFBL compared to GFBL in 3D cultures 7 days post-seeding. Expression was calculated relative to one GFBL line. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, ** p<0.01, ***p<0.001; Student’s t-test). !-SMA: !-smooth muscle actin; NMMIIA: non-muscle myosin IIA; NMMIIB: non-muscle myosin IIB.  !  "$!  !!!!!!!  Fig. 14. Expression of genes involved in internalization (Endo180 and LRP1) and intracellular degradation of ECM molecules (CTSK) in 3D cultures 7 days postseeding. Expression was calculated relative to one GFBL line. Results show mean +/SEM. N=5 parallel cell lines. (*p<0.05, ***p<0.001; Student’s t-test). Endo180 (CD280); LRP1: lipoprotein receptor-related protein 1; CTSK: cathepsin K.  !  "%!  !!!!!!!  Fig. 15. Expression of key genes by GFBL and SFBL remained significantly different in 3D cultures over time. Results showed significantly higher expression of several genes over time in SFBL compared with GFBL. Expression was calculated relative to one GFBL line. Relative mRNA expression of all of the genes remained constant over time, except that SFBL showed significantly increased expression of DCN expression over time. Results show mean +/- SEM. N=5 parallel cell lines. (*p<0.05, **p<0.01, ***p<0.001; Student’s t-test and ANOVA for multiple comparisons). DCN: decorin; BGN: biglycan; FMOD: fibromodulin; LUM: lumican; !-SMA: !-smooth muscle actin.  !  ""!  !!!!!!!  Fig. 16. Total collagen abundance and collagen type I accumulation in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. (A,B) Using Sircol Biocolor assay, there was no significant difference in total collagen abundance relative to cell number (A) and relative to total protein (B) between GFBL- and SFBL-derived 3D ECM. Results show mean +/- SEM. N=5 parallel cell lines. (Student’s t-test). (C) Immunostaining of collagen type I before (a,b) and after (c,d) cell extraction in 3D cultures 7 days post-seeding. Collagen organization followed the distinct shape and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and displayed more parallel organization compared to SFBL.  !  &''!  !!!!!!!  Fig. 17. Total sulphated glycosaminoglycans (GAG) in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. Using Blyscan Biocolor assay, there was no significant difference in total sulphated GAG abundance, relative to cell number (A) or relative to total protein (B) between GFBL- and SFBL-derived 3D ECM. Results show mean +/SEM. N=5 parallel cell lines. (Student’s t-test).  !  &'&!  !!!!!!!  Fig. 18. Fibromodulin (FMOD) abundance in the cell-derived 3D ECM after chondroitinase ABC (cABC) and keratanase enzyme pretreatment. (A) Western blot analysis of cell-free 3D ECM derived from SFBL-4-1 indicated that cABC digestion increased the available FMOD for anti-FMOD binding relative to keratanase and nonenzyme treated. Control (1) is 3D ECM sample incubated with cABC buffer alone and Control (2) is the 3D ECM sample incubated first in the buffer for cABC for 24 h then in buffer for keratanase for 24 h. (B) Coomassie blue staining was used to indicate the loading of total protein in each lane.  !  &'(!  !!!!!!!  Fig. 19. Small leucine-rich proteoglycan (SLRP) abundance in SFBL- and GFBLderived 3D ECM 7 days post-seeding. SFBL-derived 3D ECM had a significantly greater amount of biglycan (BGN) protein than GFBL-derived 3D ECM (A,E). Decorin (DCN) (B,F), fibromodulin (FMOD) (C,G) and lumican (LUM) (D,H) did not show a significant difference between 3D ECM. However, there were large differences in the amount of DCN deposited by individual cell lines. Results from four parallel GFBL- and SFBL-derived 3D ECM are shown (A-D). The results are representative of three repeated experiments. Results show mean +/- SEM. N=5 parallel cell lines (E-H). (*p<0.05; Student’s t-test).  !  &')!  !!!!!!!  Fig. 20. Small leucine-rich proteoglycan (SLRP) accumulation in SFBL- and GFBLderived 3D ECM, with or without the presence of cells, 7 days post-seeding. SLRPs were found both intracellularly and extracellularly. In the 3D ECM, they were organized into a fibrillar-like orientation, suggesting that these molecules were associated with collagen or other fibrillar ECM proteins. Semi-quantitative analysis showed no difference in SLRPs abundance in GFBL- and SFBL-derived 3D ECM, except that GFBL showed greater abundance of intracellular immunostaining for DCN, FMOD and LUM as compared with SFBL. Results from one representative GFBL and SFBL line are shown. BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican.  !  &'*!  !!!!!!!  Fig. 21. EDA-fibronectin (EDA-FN) abundance in SFBL- and GFBL-derived 3D ECM 7 days post-seeding. There was no difference in EDA-FN abundance between SFBL- and GFBL-derived 3D ECM. However, individual cell lines showed considerable variation in abundance of EDA-FN in the 3D ECM. Results show mean +/- SEM. N=5 parallel cell lines. (Student’s t-test).  !  &'+!  !!!!!!!  Fig. 22. Tenascin-C abundance in SFBL- and GFBL-derived 3D ECM 7 days postseeding. Analysis of 3D ECM samples by Western blotting without chondroitinase ABC (cABC) pretreatment showed that SFBL deposited significantly greater amount of tenascin-C protein than GFBL in the 3D ECM (A,C). However, when samples were analyzed after cABC treatment, there was a greater amount of tenascin-C protein in GFBL-derived 3D ECM compared to SFBL-derived 3D ECM (p=0.07). Results from four parallel GFBL and SFBL lines are shown and the results are representative of 2 repeated experiments (A-B). Results show mean +/- SEM. N=5 parallel cell lines (C-D). (*p<0.05; Student’s t-test). !  &'#!  !!!!!!!  Fig. 23. Tenascin-C and EDA-FN accumulation in SFBL- and GFBL-derived 3D ECM, before and after cell extraction, 7 days post-seeding. Tenascin-C and EDA-FN abundance appear to remain unchanged after cell removal. The accumulation of tenascinC and EDA-FN followed the distinct organization and orientation of GFBL and SFBL, as GFBL were more narrow and spindle-like and showed more parallel organization compared to SFBL. Results from one representative GFBL and SFBL line are shown.  !  &'$!  !!!!!!!  Fig. 24. SFBL appeared to have a greater adhesion rate compared to GFBL. Adhesion of GFBL and SFBL was observed on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) or collagen type I. It appeared that SFBL had a greater adhesion rate than GFBL, regardless of the substrate. Both SFBL and GFBL seemed to adhere more rapidly and adopted a more spindle-like morphology on cell-derived 3D ECM compared to collagen I substrate. There was no difference in adhesion rate within a given cell type when fibroblasts were seeded on SECM or GECM.  !  &'%!  Fig. 25. Expression of fibrillar ECM and matricellular protein genes was significantly greater in SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Expression of type I and III collagen and elastin-1 was significantly higher in SFBL as compared with GFBL. Among the matricellular protein genes, tenascin-C and tenascin-X showed greater expression in SFBL. Expression of differentially regulated genes did not depend on the nature of the matrix. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, ***p<0.001; Student’s t-test). !  !  "#$!  !!!!!!!  Fig. 26. Expression of SLRP genes was significantly greater in SFBL 24 h postseeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, **p<0.01; Student’s t-test). BGN: biglycan; DCN: decorin; FMOD: fibromodulin; LUM: lumican.  !  ""#!  !!!!!!!  Fig. 27. Expression of MMPs and TIMP4 in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. Expression of MMPs was generally greater when fibroblasts were seeded on cell-derived 3D ECM compared to collagen I substrate. Only MMP-11 expression was significantly different between SFBL and GFBL. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05; Student’s t-test).  !  """!  !!!!!!!  Fig. 28. Expression of growth factor genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No differences in expression of TGF-!1 or CCN2/CTGF were noted between GFBL and SFBL or between different substrates. SFBL had significantly higher expression of TGF-!3 on collagen I substrate than GFBL, but when cells were seeded in 3D ECM, this difference was reduced. Vascular endothelial growth factor-" (VEGF-") expression was greater in both GFBL and SFBL when seeded on cellderived 3D ECM compared to collagen I substrate, but this difference did not reach statistical significance. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05, Student’s t-test). TGF-!R1: TGF! receptor 1; TGF-!R2: TGF-! receptor 2.  !  ""#!  !!!!!!!  Fig. 29. Expression of myofibroblast (!-SMA, P311), cell contractility (NMMIIA, NMMIIB) associated genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No significant differences in gene expression of "-smooth muscle actin ("-SMA), non-muscle myosin IIA (NMMIIA), and non-muscle myosin IIB (NMMIIB) between GFBL and SFBL or between different substrates were noted. However, expression of P311 and "11 integrin was significantly higher in SFBL than in GFBL in all substrates. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines. (*p<0.05; Student’s t-test).  !  ""$!  !!!!!!!  Fig. 30. Intracellular ECM degradation (Endo180, CTSK) associated genes in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen type I. No significant differences in gene expression of Endo180 (CD280) or cathepsin K (CTSK) between GFBL and SFBL or between different substrates were noted. Results show mean +/- SEM. N=5-6 repeated experiments of two parallel GFBL (-DC, -OL) and SFBL (-2-C, -4-1) cell lines.  !  ""#!  !!!!!!!  Fig. 31. Western blot analysis of total and phosphorylated SMAD3 (p-SMAD3) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) and collagen I substrate. Total SMAD3 was greater in GFBL compared to SFBL, while p-SMAD relative to total SMAD3 was greater in SFBL in all substrates. !-actin was used as a loading control.  !  ""$!  !!!!!!!  Fig. 32. Quantification of total SMAD3 protein and phosphorylated SMAD3 (pSMAD3) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. Total SMAD3 was greater in GFBL (A,B), while p-SMAD3 relative to total SMAD3 was greater in SFBL (C,D). (A,C) N=3 parallel GFBL and SFBL cell lines. (B,D) Results show pooled data of GFBL and SFBL (N=9) on three different substrates. Results show mean +/SEM. (*p<0.05; **p<0.01 Student’s t-test).  !  ""%!  !!!!!!!  Fig. 33. Western blot analysis of total and phosphorylated ERK1/2 (p-ERK1/2) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBLderived 3D ECM (SECM) and collagen I substrate. ERK1/2 phosphorylation was greater in GFBL and SFBL when seeded on cell-derived 3D ECM. No differences were found in total or phosphorylated ERK1/2 levels between SFBL and GFBL on any of the substrates. !-tubulin was used as a loading control.  !  ""&!  !!!!!!!  Fig. 34. Quantification of total and phosphorylated ERK1/2 (p-ERK1/2) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. (A) No difference was found in phosphorylated ERK1/2 relative to total ERK1/2 levels between GFBL and SFBL on any of the substrates. (B,C) ERK1/2 phosphorylation was greater in GFBL (p=0.12) and SFBL (p<0.01) when seeded on cell-derived 3D ECM relative to collagen type I substrate. Results show pooled data from GECM and SECM for the given cell type. Results show mean +/- SEM. A: N=3 of three parallel cell lines. (**p<0.01; Student’s ttest).  !  ""'!  !!!!!!!  Fig. 35. Quantification of total and phosphorylated p38 (p-p38) in GFBL and SFBL 24 h post-seeding on GFBL-derived 3D ECM (GECM), SFBL-derived 3D ECM (SECM) and collagen I substrate. No significant differences were found between GFBL and SFBL or between different substrates. !-tubulin was used as a loading control. Results show mean +/- SEM. N= 3 parallel cell lines. (Student’s t-test).  !  ""(!  !!!!!!!  ! ! Fig. 36. A schematic presentation of the signaling pathways that may be differently regulated between GFBL and SFBL (SMAD3) and by 3D ECM and 2D substrate (ERK1/2). We found that SFBL had an innately greater phosphorylation of SMAD3 compared to GFBL. Higher levels of SMAD3 phosphorylation in SFBL may underlie the increased expression of matrix deposition associated genes and decreased expression of matrix remodeling associated genes. When GFBL and SFBL were cultured on 3D ECM, there was a greater phosphorylation of ERK1/2 compared to when these cells were seeded on 2D type I collagen substrate. Higher levels of ERK1/2 phosphorylation may attribute to the greater expression of AP-1-dependent MMP and VEGF-" in GFBL and SFBL seeded on 3D ECM.  !  ")*!  !!!!!!!  REFERENCES 1.  Aarabi S, Longaker MT, Gurtner GC. Hypertrophic scar formation following burns and trauma: new approaches to treatment. PLoS Med. 2007 Sep;4(9) e234.  2.  Adzick NS, Longaker MT. Scarless fetal healing. Therapeutic implications. Ann Surg. 1992 Jan;215(1):3-7.  3.  Al Haj Zen A, Lafont A, Durand E, Brasselet C, Lemarchand P, Godeau G, Gogly B. Effect of adenovirus-mediated overexpression of decorin on metalloproteinases, tissue inhibitors of metalloproteinases and cytokines secretion by human gingival fibroblasts. Matrix Biol. 2003 May;22(3):251-8.  4.  Alford AI, Hankenson KD. Matricellular proteins: Extracellular modulators of bone development, remodeling, and regeneration. Bone. 2006 Jun;38(6):749-57.  5.  Alimohamad H, Habijanac T, Larjava H, Häkkinen L. Colocalization of the collagenbinding proteoglycans decorin, biglycan, fibromodulin and lumican with different cells in human gingiva. J Periodontal Res. 2005 Feb;40(1):73-86.  6.  Alster T, Zaulyanov L. Laser scar revision: a review. Dermatol Surg. 2007 Feb;33(2):131-40. Review. Erratum in: Dermatol Surg. 2007 Jun;33(6):770.  7.  Amadeu TP, Braune AS, Porto LC, Desmoulière A, Costa AM. Fibrillin-1 and elastin are differentially expressed in hypertrophic scars and keloids. Wound Repair Regen. 2004 Mar-Apr;12(2):169-74.  8.  Amatangelo MD, Bassi DE, Klein-Szanto AJ, Cukierman E. Stroma-derived threedimensional matrices are necessary and sufficient to promote desmoplastic differentiation of normal fibroblasts. Am J Pathol. 2005 Aug;167(2):475-88.  9.  Ameye L, Young MF. Mice deficient in small leucine-rich proteoglycans: novel in vivo models for osteoporosis, osteoarthritis, Ehlers-Danlos syndrome, muscular dystrophy, and corneal diseases. Glycobiology. 2002 Sep;12(9):107R-16R.  10. Annes JP, Chen Y, Munger JS, Rifkin DB. Integrin alphaVbeta6-mediated activation of latent TGF-beta requires the latent TGF-beta binding protein-1. J Cell Biol. 2004 Jun 7;165(5):723-34. 11. Armstrong DG, Jude EB. The role of matrix metalloproteinases in wound healing. J Am Podiatr Med Assoc. 2002 Jan;92(1):12-8. 12. Attisano L, Wrana JL. Signal transduction by the TGF-beta superfamily. Science. 2002 May 31;296(5573):1646-7.  !  ")"!  !!!!!!! 13. Aumailley M, Gayraud B. Structure and biological activity of the extracellular matrix. J Mol Med (Berl). 1998 Mar;76(3-4):253-65. 14. Babu M, Diegelmann R, Oliver N. Keloid fibroblasts exhibit an altered response to TGF-beta. J Invest Dermatol. 1992 Nov;99(5):650-5. 15. Baker BM, Chen CS. Deconstructing the third dimension: how 3D culture microenvironments alter cellular cues. J Cell Sci. 2012 Jul 1;125(Pt 13):3015-h. 16. Bao P, Kodra A, Tomic-Canic M, Golinko MS, Ehrlich HP, Brem H. The role of vascular endothelial growth factor in wound healing. J Surg Res. 2009 May 15;153(2):347-58. 17. Beacham DA, Amatangelo MD, Cukierman E. Preparation of extracellular matrices produced by cultured and primary fibroblasts. Curr Protoc Cell Biol. 2007 Jan;Chapter 10:Unit 10.9. 18. Beanes SR, Dang C, Soo C, Wang Y, Urata M, Ting K, Fonkalsrud EW, Benhaim P, Hedrick MH, Atkinson JB, Lorenz HP. Down-regulation of decorin, a transforming growth factor-beta modulator, is associated with scarless fetal wound healing. J Pediatr Surg. 2001 Nov;36(11):1666-71. 19. Benbow U, Brinckerhoff CE. The AP-1 site and MMP gene regulation: what is all the fuss about? Matrix Biol. 1997 Mar;15(8-9):519-26. 20. Bennett NT, Schultz GS. Growth factors and wound healing: biochemical properties of growth factors and their receptors. Am J Surg. 1993 Jun;165(6):728-37. 21. Berman B, Duncan MR. Short-term keloid treatment in vivo with human interferon alfa-2b results in a selective and persistent normalization of keloidal fibroblast collagen, glycosaminoglycan, and collagenase production in vitro. J Am Acad Dermatol. 1989 Oct;21(4 Pt 1):694-702. 22. Berse B, Hunt JA, Diegel RJ, Morganelli P, Yeo K, Brown F, Fava RA. Hypoxia augments cytokine (transforming growth factor-beta (TGF-beta) and IL-1)-induced vascular endothelial growth factor secretion by human synovial fibroblasts. Clin Exp Immunol. 1999 Jan;115(1):176-82. 23. Bhangoo KS, Quinlivan JK, Connelly JR. Elastin fibers in scar tissue. Plast Reconstr Surg. 1976 Mar;57(3):308-13. 24. Bhattacharyya S, Chen SJ, Wu M, Warner-Blankenship M, Ning H, Lakos G, Mori Y, Chang E, Nihijima C, Takehara K, Feghali-Bostwick C, Varga J. Smadindependent transforming growth factor-beta regulation of early growth response-1 and sustained expression in fibrosis: implications for scleroderma. Am J Pathol. 2008 Oct;173(4):1085-99.  !  "))!  !!!!!!! 25. Bidanset DJ, LeBaron R, Rosenberg L, Murphy-Ullrich JE, Hook M. Regulation of cell substrate adhesion: effects of small galactosaminoglycan-containing proteoglycans. J Cell Biol. 1992 Sep;118(6):1523-31. 26. Bielefeld KA, Amini-Nik S, Alman BA. Cutaneous wound healing: recruiting developmental pathways for regeneration. Cell Mol Life Sci. 2012 Oct 4. 27. Bombaro KM, Engrav LH, Carrougher GJ, Wiechman SA, Faucher L, Costa BA, Heimbach DM, Rivara FP, Honari S. What is the prevalence of hypertrophic scarring following burns? Burns. 2003 Jun;29(4):299-302. 28. Bond JE, Ho TQ, Selim MA, Hunter CL, Bowers EV, Levinson H. Temporal spatial expression and function of non-muscle myosin II isoforms IIA and IIB in scar remodeling. Lab Invest. 2011 Apr;91(4):499-508. 29. Border WA, Noble NA. Transforming growth factor beta in tissue fibrosis. N Engl J Med. 1994 Nov 10;331(19):1286-92. 30. Bornstein P, Sage EH. Matricellular proteins: extracellular modulators of cell function. Curr Opin Cell Biol. 2002 Oct;14(5):608-16. 31. Bossard C, Van den Berghe L, Laurell H, Castano C, Cerutti M, Prats AC, Prats H. Antiangiogenic properties of fibstatin, an extracellular FGF-2-binding polypeptide. Cancer Res. 2004 Oct 15;64(20):7507-12. 32. Breau MA, Pietri T, Stemmler MP, Thiery JP, Weston JA. A nonneural epithelial domain of embryonic cranial neural folds gives rise to ectomesenchyme. Proc Natl Acad Sci U S A. 2008 Jun 3;105(22):7750-5. 33. Brockes JP. Amphibian limb regeneration: rebuilding a complex structure. Science. 1997 Apr 4;276(5309):81-7. 34. Bronson RE, Argenta JG, Siebert EP, Bertolami CN. Distinctive fibroblastic subpopulations in skin and oral mucosa demonstrated by differences in glycosaminoglycan content. In Vitro Cell Dev Biol. 1988 Nov;24(11):1121-6. 35. Brown LF, Yeo KT, Berse B, Yeo TK, Senger DR, Dvorak HF, van de Water L. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med. 1992 Nov 1;176(5):13759. 36. Buckley CD, Pilling D, Lord JM, Akbar AN, Scheel-Toellner D, Salmon M. Fibroblasts regulate the switch from acute resolving to chronic persistent inflammation. Trends Immunol. 2001 Apr;22(4):199-204.  !  ")+!  !!!!!!! 37. Bullard KM, Lund L, Mudgett JS, Mellin TN, Hunt TK, Murphy B, Ronan J, Werb Z, Banda MJ. Impaired wound contraction in stromelysin-1-deficient mice. Ann Surg. 1999 Aug;230(2):260-5. 38. Bullard KM, Longaker MT, Lorenz HP. Fetal wound healing: current biology. World J Surg. 2003 Jan;27(1):54-61. 39. Cass DL, Sylvester KG, Yang EY, Crombleholme TM, Adzick NS. Myofibroblast persistence in fetal sheep wounds is associated with scar formation. J Pediatr Surg. 1997 Jul;32(7):1017-21. 40. Castellani P, Viale G, Dorcaratto A, Nicolo G, Kaczmarek J, Querze G, Zardi L. The fibronectin isoform containing the ED-B oncofetal domain: a marker of angiogenesis. Int J Cancer. 1994 Dec 1;59(5):612-8. 41. Chang HY, JT Chi, S Dudoit, C Bondre, M van de Rijn, D Botstein and PO Brown. Diversity, topographic differentiation, and positional memory in human fibroblasts. Natl Acad Sci. 2002 Oct;99(20):12877-12882. 42. Chavrier C. The elastic system fibres in healthy human gingiva. Arch Oral Biol. 1990;35 Suppl:223S-225S. 43. Chen XD, Dusevich V, Feng JQ, Manolagas SC, Jilka RL. Extracellular matrix made by bone marrow cells facilitates expansion of marrow-derived mesenchymal progenitor cells and prevents their differentiation into osteoblasts. J Bone Miner Res. 2007 Dec;22(12):1943-56. 44. Chen XD. Extracellular matrix provides an optimal niche for the maintenance and propagation of mesenchymal stem cells. Birth Defects Res C Embryo Today. 2010 Mar;90(1):45-54. 45. Chin GS, Liu W, Peled Z, Lee TY, Steinbrech DS, Hsu M, Longaker MT. Differential expression of transforming growth factor-beta receptors I and II and activation of Smad 3 in keloid fibroblasts. Plast Reconstr Surg. 2001 Aug;108(2):4239. 46. Cho ML, Jung YO, Moon YM, Min SY, Yoon CH, Lee SH, Park SH, Cho CS, Jue DM, Kim HY. Interleukin-18 induces the production of vascular endothelial growth factor (VEGF) in rheumatoid arthritis synovial fibroblasts via AP-1-dependent pathways. Immunol Lett. 2006 Mar 15;103(2):159-66. 47. Clark RAF. (1996). The molecular and cellular biology of wound repair. New York, London: Plenum Press.  !  ")#!  !!!!!!! 48. Colwell AS, Yun R, Krummel TM, Longaker MT, Lorenz HP. Keratinocytes modulate fetal and postnatal fibroblast transforming growth factor-beta and Smad expression in co-culture. Plast Reconstr Surg. 2007 Apr 15;119(5):1440-5. 49. Cui W, Bryant MR, Sweet PM, McDonnell PJ. Changes in gene expression in response to mechanical strain in human scleral fibroblasts. Exp Eye Res. 2004 Feb;78(2):275-84. 50. Cukierman E, Pankov R, Stevens DR, Yamada KM. Taking cell-matrix adhesions to the third dimension. Science. 2001 Nov 23;294(5547):1708-12. 51. Cukierman E, Pankov R, Yamada KM. Cell interactions with three-dimensional matrices. Curr Opin Cell Biol. 2002 Oct;14(5):633-9. 52. Curino AC, Engelholm LH, Yamada SS, Holmbeck K, Lund LR, Molinolo AA, Behrendt N, Nielsen BS, Bugge TH. Intracellular collagen degradation mediated by uPARAP/Endo180 is a major pathway of extracellular matrix turnover during malignancy. J Cell Biol. 2005 Jun 20;169(6):977-85. 53. Cutroneo KR. How is Type I procollagen synthesis regulated at the gene level during tissue fibrosis. J Cell Biochem. 2003 Sep 1;90(1):1-5. 54. Dang CM, Beanes SR, Lee H, Zhang X, Soo C, Ting K. Scarless fetal wounds are associated with an increased matrix metalloproteinase-to-tissue-derived inhibitor of metalloproteinase ratio. Plast Reconstr Surg. 2003 Jun;111(7):2273-85. 55. Dasu MR, Hawkins HK, Barrow RE, Xue H, Herndon DN. Gene expression profiles from hypertrophic scar fibroblasts before and after IL-6 stimulation. J Pathol. 2004 Apr;202(4):476-85. 56. Davari P, Gorouhi F, Hashemi P, Behnia F, Ghassemi A, Nasiri-Kashani M, Firooz A. Pulsed dye laser treatment with different onset times for new surgical scars: a single-blind randomized controlled trial. Lasers Med Sci. 2012 Sep;27(5):1095-8. 57. De Donatis A, Ranaldi F, Cirri P. Reciprocal control of cell proliferation and migration. Cell Commun Signal. 2010 Sep 7;8:20. 58. de Waard JW, de Man BM, Wobbes T, van der Linden CJ, Hendriks T. Inhibition of fibroblast collagen synthesis and proliferation by levamisole and 5-fluorouracil. Eur J Cancer. 1998 Jan;34(1):162-7. 59. Deitch EA, Wheelahan TM, Rose MP, Clothier J, Cotter J. Hypertrophic burn scars: analysis of variables. J Trauma. 1983 Oct;23(10):895-8. 60. Deng T, Karin M. c-Fos transcriptional activity stimulated by H-Ras-activated protein kinase distinct from JNK and ERK. Nature. 1994 Sep 8;371(6493):171-5.  !  ")$!  !!!!!!!  61. Durani P, Occleston N, O'Kane S, Ferguson MW. Avotermin: a novel antiscarring agent. Int J Low Extrem Wounds. 2008 Sep;7(3):160-8. 62. Ebisawa K, Kato R, Okada M, Sugimura T, Latif MA, Hori Y, Narita Y, Ueda M, Honda H, Kagami H. Gingival and dermal fibroblasts: their similarities and differences revealed from gene expression. J Biosci Bioeng. 2011 Mar;111(3):255-8. 63. Eckes B, Kessler D, Aumailley M, Krieg T. Interactions of fibroblasts with the extracellular matrix: implications for the understanding of fibrosis. Springer Semin Immunopathol. 1999;21(4):415-29. 64. Eckes B, Nischt R, Krieg T. Cell-matrix interactions in dermal repair and scarring. Fibrogenesis Tissue Repair. 2010 Mar 11;3:4. doi: 10.1186/1755-1536-3-4. 65. Enoch S, Wall I, Peake M, Davies L, Farrier J, Giles P, Baird D, Kipling D, Price P, Moseley R, Thomas D, Stephens P. Increased oral fibroblast lifespan is telomeraseindependent. J Dent Res. 2009 Oct;88(10):916-21. 66. Enoch S, Peake MA, Wall I, Davies L, Farrier J, Giles P, Kipling D, Price P, Moseley R, Thomas D, Stephens P. 'Young' oral fibroblasts are geno/phenotypically distinct. J Dent Res. 2010 Dec;89(12):1407-13. 67. Eslami A, Gallant-Behm CL, Hart DA, Wiebe C, Honardoust D, Gardner H, Häkkinen L, Larjava HS. Expression of integrin alphavbeta6 and TGF-beta in scarless vs scar-forming wound healing. J Histochem Cytochem. 2009 Jun;57(6):54357. 68. Fang RC, Mustoe TA. Animal models of wound healing: utility in transgenic mice. J Biomater Sci Polym Ed. 2008;19(8):989-1005. 69. Fedyk ER, Jones D, Critchley HO, Phipps RP, Blieden TM, Springer TA. Expression of stromal-derived factor-1 is decreased by IL-1 and TNF and in dermal wound healing. J Immunol. 2001 May 1;166(9):5749-54. 70. Ferguson MW, O'Kane S. Scar-free healing: from embryonic mechanisms to adult therapeutic intervention. Philos Trans R Soc Lond B Biol Sci. 2004 May 29;359(1445):839-50. 71. Flavell SJ, Hou TZ, Lax S, Filer AD, Salmon M, Buckley CD. Fibroblasts as novel therapeutic targets in chronic inflammation. Br J Pharmacol. 2008 Mar;153 Suppl 1:S241-6. 72. Fontana L, Chen Y, Prijatelj P, Sakai T, Fässler R, Sakai LY, Rifkin DB. Fibronectin is required for integrin alphavbeta6-mediated activation of latent TGF-beta complexes containing LTBP-1. FASEB J. 2005 Nov;19(13):1798-808.  !  ")%!  !!!!!!! 73. Fries KM, Blieden T, Looney RJ, Sempowski GD, Silvera MR, Willis RA, Phipps RP. Evidence of fibroblast heterogeneity and the role of fibroblast subpopulations in fibrosis. Clin Immunol Immunopathol. 1994 Sep;72(3):283-92. 74. Gabbiani G, Ryan GB, Majne G. Presence of modified fibroblasts in granulation tissue and their possible role in wound contraction. Experientia. 1971 May 15;27(5):549-50. 75. Gangemi EN, Gregori D, Berchialla P, Zingarelli E, Cairo M, Bollero D, Ganem J, Capocelli R, Cuccuru F, Cassano P, Risso D, Stella M. Epidemiology and risk factors for pathologic scarring after burn wounds. Arch Facial Plast Surg. 2008 MarApr;10(2):93-102. 76. Gauglitz GG, Korting HC, Pavicic T, Ruzicka T, Jeschke MG. Hypertrophic scarring and keloids: pathomechanisms and current and emerging treatment strategies. Mol Med. 2011 Jan-Feb;17(1-2):113-25. 77. Ghahary A, Shen YJ, Scott PG, Gong Y, Tredget EE. Enhanced expression of mRNA for transforming growth factor-beta, type I and type III procollagen in human postburn hypertrophic scar tissues. J Lab Clin Med. 1993 Oct;122(4):465-73. 78. Giancotti FG, Ruoslahti E. Integrin signaling. Science. 1999 Aug 13;285(5430):102832. 79. Gill SE, Parks WC. Metalloproteinases and their inhibitors: regulators of wound healing. Int J Biochem Cell Biol. 2008;40(6-7):1334-47. Epub 2007 Oct 26. 80. Gold LI, Pearlstein E. Fibronectin-collagen binding and requirement during cellular adhesion. Biochem J. 1980 Feb 15;186(2):551-9. 81. Gold MH, Foster TD, Adair MA, Burlison K, Lewis T. Prevention of hypertrophic scars and keloids by the prophylactic use of topical silicone gel sheets following a surgical procedure in an office setting. Dermatol Surg. 2001 Jul;27(7):641-4. 82. González JM, Navarro-Puche A, Casar B, Crespo P, Andrés V. Fast regulation of AP1 activity through interaction of lamin A/C, ERK1/2, and c-Fos at the nuclear envelope. J Cell Biol. 2008 Nov 17;183(4):653-66. 83. Gospodarowicz D, Neufeld G, Schweigerer L. Fibroblast growth factor: structural and biological properties. J Cell Physiol Suppl. 1987;Suppl 5:15-26. 84. Goumans MJ, Liu Z, ten Dijke P. TGF-beta signaling in vascular biology and dysfunction. Cell Res. 2009 Jan;19(1):116-27. 85. Green JA, Yamada KM. Three-dimensional microenvironments modulate fibroblast signaling responses. Adv Drug Deliv Rev. 2007 Nov 10;59(13):1293-8.  !  ")&!  !!!!!!! 86. Grinnell F, Billingham RE, Burgess L. Distribution of fibronectin during wound healing in vivo. J Invest Dermatol. 1981 Mar;76(3):181-9. 87. Grisanti S, Szurman P, Warga M, Kaczmarek R, Ziemssen F, Tatar O, Bartz-Schmidt KU. Decorin modulates wound healing in experimental glaucoma filtration surgery: a pilot study. Invest Ophthalmol Vis Sci. 2005 Jan;46(1):191-6. 88. Grose R, Werner S. Wound healing studies in transgenic and knockout mice. A review. Methods Mol Med. 2003;78:191-216. 89. Grumet M, Milev P, Sakurai T, Karthikeyan L, Bourdon M, Margolis RK, Margolis RU. Interactions with tenascin and differential effects on cell adhesion of neurocan and phosphacan, two major chondroitin sulfate proteoglycans of nervous tissue. J Biol Chem. 1994 Apr 22;269(16):12142-6. 90. Guo F, Carter DE, Mukhopadhyay A, Leask A. Gingival fibroblasts display reduced adhesion and spreading on extracellular matrix: a possible basis for scarless tissue repair? PLoS One. 2011a;6(11):e27097. 91. Guo F, Carter DE, Leask A. Mechanical tension increases CCN2/CTGF expression and proliferation in gingival fibroblasts via a TGF!-dependent mechanism. PLoS One. 2011b;6(5):e19756. 92. Guo F, Hutchenreuther J, Carter DE, Leask A. TAK1 is required for dermal wound healing and homeostasis. Journal of Investigative Dermatology (2013). doi: 10.1038/jid.2013.28 93. Gurtner GC, Werner S, Barrandon Y, Longaker MT. Wound repair and regeneration. Nature. 2008 May 15;453(7193):314-21. 94. Hay ED: Cell Biology of Extracellular Matrix, 2nd edition. New York: Plenum Press; 1991. 95. Häkkinen L, Heino J, Koivisto L, Larjava H. Altered interaction of human granulation-tissue fibroblasts with fibronectin is regulated by alpha 5 beta 1 integrin. Biochim Biophys Acta. 1994 Oct 20;1224(1):33-42. 96. Häkkinen L, Uitto VJ, Larjava H. Cell biology of gingival wound healing. Periodontol 2000. 2000 Oct;24:127-52. 97. Häkkinen L, Larjava H, Koivisto L. (2012). Granulation Tissue Formation and Remodeling. In Oral Wound Healing- Cell Biology and Clinical Management, ed Larjava H, Wiley-Blackwell, Ames, IA, pp.125-173. 98. Hamburg EJ, Atit RP. Sustained !-catenin activity in dermal fibroblasts is sufficient for skin fibrosis. J Invest Dermatol. 2012 Oct;132(10):2469-72.  !  ")'!  !!!!!!! 99. Hamilton DW. Functional role of periostin in development and wound repair: implications for connective tissue disease. J Cell Commun Signal. 2008 Jun;2(1-2):917. 100. Hattori N, Mochizuki S, Kishi K, Nakajima T, Takaishi H, D'Armiento J, Okada Y. MMP-13 plays a role in keratinocyte migration, angiogenesis, and contraction in mouse skin wound healing. Am J Pathol. 2009 Aug;175(2):533-46. 101. Hildebrand A, Romarís M, Rasmussen LM, Heinegård D, Twardzik DR, Border WA, Ruoslahti E. Interaction of the small interstitial proteoglycans biglycan, decorin and fibromodulin with transforming growth factor beta. Biochem J. 1994 Sep 1;302 ( Pt 2):527-34. 102. Hirt-Burri N, Ramelet AA, Raffoul W, de Buys Roessingh A, Scaletta C, Pioletti D, Applegate LA. Biologicals and fetal cell therapy for wound and scar management. ISRN Dermatol. 2011;2011:549870. 103. Hoffman S, Edelman GM. A proteoglycan with HNK-1 antigenic determinants is a neuron-associated ligand for cytotactin. Proc Natl Acad Sci U S A. 1987 Apr;84(8):2523-7. 104. Holmbeck K, Szabova L. Aspects of extracellular matrix remodeling in development and disease. Birth Defects Res C Embryo Today. 2006 Mar;78(1):11-23. 105. Honardoust HA, Jiang G, Koivisto L, Wienke D, Isacke CM, Larjava H, Häkkinen L. Expression of Endo180 is spatially and temporally regulated during wound healing. Histopathology. 2006 Dec;49(6):634-48. 106. Honardoust D, Eslami A, Larjava H, Häkkinen L. Localization of small leucine-rich proteoglycans and transforming growth factor-beta in human oral mucosal wound healing. Wound Repair Regen. 2008 Nov-Dec;16(6):814-23. 107. Huebsch N, Arany PR, Mao AS, Shvartsman D, Ali OA, Bencherif SA, RiveraFeliciano J, Mooney DJ. Harnessing traction-mediated manipulation of the cell/matrix interface to control stem-cell fate. Nat Mater. 2010 Jun;9(6):518-26. 108. Huijun W, Long C, Zhigang Z, Feng J, Muyi G. Ex vivo transfer of the decorin gene into rat glomerulus via a mesangial cell vector suppressed extracellular matrix accumulation in experimental glomerulonephritis. Exp Mol Pathol. 2005 Feb;78(1):17-24. 109. Karagoz H, Sever C, Bayram Y, Sahin C, Kulahci Y, Ulkur E. A Review of the Prevention and Treatment of Hypertrophic Scars: Part I Clinical Aspects Arch Clin Exp Surg. 2012; 1(4).  !  ")(!  !!!!!!! 110. Ingber DE. Cellular mechanotransduction: putting all the pieces together again. FASEB J. 2006 May;20(7):811-27. 111. Inoue M, Kratz G, Haegerstrand A, Ståhle-Bäckdahl M. Collagenase expression is rapidly induced in wound-edge keratinocytes after acute injury in human skin, persists during healing, and stops at re-epithelialization. J Invest Dermatol. 1995 Apr;104(4):479-83. 112. Ishihara H, Yoshimoto H, Fujioka M, Murakami R, Hirano A, Fujii T, Ohtsuru A, Namba H, Yamashita S. Keloid fibroblasts resist ceramide-induced apoptosis by overexpression of insulin-like growth factor I receptor. J Invest Dermatol. 2000 Dec;115(6):1065-71. 113. Jackson-Boeters L, Wen W, Hamilton DW. Periostin localizes to cells in normal skin, but is associated with the extracellular matrix during wound repair. J Cell Commun Signal. 2009 Jun;3(2):125-33. 114. Johnson GL, Lapadat R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science. 2002 Dec 6;298(5600):1911-2. 115. Jun JI, Lau LF. The matricellular protein CCN1 induces fibroblast senescence and restricts fibrosis in cutaneous wound healing. Nat Cell Biol. 2010 Jul;12(7):676-85. 116. Jun JI, Lau LF. Cellular senescence controls fibrosis in wound healing. Aging (Albany NY). 2010 Sep;2(9):627-31. 117. Kalamajski S, Oldberg A. The role of small leucine-rich proteoglycans in collagen fibrillogenesis. Matrix Biol. 2010 May;29(4):248-53. 118. Kielty CM, Sherratt MJ, Shuttleworth CA. Elastic fibres. J Cell Sci. 2002 Jul 15;115(Pt 14):2817-28. 119. King SJ, Parsons M. Imaging cells within 3D cell-derived matrix. Methods Mol Biol. 2011;769:53-64. 120. Kjøller L, Engelholm LH, Høyer-Hansen M, Danø K, Bugge TH, Behrendt N. uPARAP/endo180 directs lysosomal delivery and degradation of collagen IV. Exp Cell Res. 2004 Feb 1;293(1):106-16. 121. Klingberg F, Hinz B, White ES. The myofibroblast matrix: implications for tissue repair and fibrosis. J Pathol. 2013 Jan;229(2):298-309. 122. Knerer B, Formanek M, Temmel A, Martinek H, Schickinger B, Kornfehl J. The role of fibroblasts from oropharyngeal mucosa in producing proinflammatory and mitogenic cytokines without prior stimulation. Eur Arch Otorhinolaryngol. 1999;256(5):266-70.  !  "+*!  !!!!!!! 123. Koivisto L, Jiang G, Häkkinen L, Chan B, Larjava H. HaCaT keratinocyte migration is dependent on epidermal growth factor receptor signaling and glycogen synthase kinase-3alpha. Exp Cell Res. 2006 Sep 10;312(15):2791-805. 124. Koivisto L, Häkkinen, Larjava H. Re-epithelialization of Wounds. In Oral Wound Healing- Cell Biology and Clinical Management, ed Larjava H, Wiley-Blackwell, Ames, IA, pp.81-123. 125. Kolb M, Margetts PJ, Sime PJ, Gauldie J. Proteoglycans decorin and biglycan differentially modulate TGF-beta-mediated fibrotic responses in the lung. Am J Physiol Lung Cell Mol Physiol. 2001 Jun;280(6):L1327-34. 126. Krummel TM, Michna BA, Thomas BL, Sporn MB, Nelson JM, Salzberg AM, Cohen IK, Diegelmann RF. Transforming growth factor beta (TGF-beta) induces fibrosis in a fetal wound model. J Pediatr Surg. 1988 Jul;23(7):647-52. 127. Kudo Y, Siriwardena BS, Hatano H, Ogawa I, Takata T. Periostin: novel diagnostic and therapeutic target for cancer. Histol Histopathol. 2007 Oct;22(10):1167-74. 128. Kyriakides TR, Bornstein P. Matricellular proteins as modulators of wound healing and the foreign body response. Thromb Haemost. 2003 Dec;90(6):986-92. 129. Kyriakides TR, Wulsin D, Skokos EA, Fleckman P, Pirrone A, Shipley JM, Senior RM, Bornstein P. Mice that lack matrix metalloproteinase-9 display delayed wound healing associated with delayed reepithelization and disordered collagen fibrillogenesis. Matrix Biol. 2009 Mar;28(2):65-73. 130. Lai CF, Chaudhary L, Fausto A, Halstead LR, Ory DS, Avioli LV, Cheng SL. Erk is essential for growth, differentiation, integrin expression, and cell function in human osteoblastic cells. J Biol Chem. 2001 Apr 27;276(17):14443-50. 131. Lai Y, Sun Y, Skinner CM, Son EL, Lu Z, Tuan RS, Jilka RL, Ling J, Chen XD. Reconstitution of marrow-derived extracellular matrix ex vivo: a robust culture system for expanding large-scale highly functional human mesenchymal stem cells. Stem Cells Dev. 2010 Jul;19(7):1095-107. 132. Lam AP, Gottardi CJ. !-catenin signaling: a novel mediator of fibrosis and potential therapeutic target. Curr Opin Rheumatol. 2011 Nov;23(6):562-7. 133. Langholz O, Röckel D, Mauch C, Kozlowska E, Bank I, Krieg T, Eckes B. Collagen and collagenase gene expression in three-dimensional collagen lattices are differentially regulated by alpha 1 beta 1 and alpha 2 beta 1 integrins. J Cell Biol. 1995 Dec;131(6 Pt 2):1903-15.  !  "+"!  !!!!!!! 134. Laping NJ, Grygielko E, Mathur A, Butter S, Bomberger J, Tweed C, Martin W, Fornwald J, Lehr R, Harling J, Gaster L, Callahan JF, Olson BA. Inhibition of transforming growth factor (TGF)-beta1-induced extracellular matrix with a novel inhibitor of the TGF-beta type I receptor kinase activity: SB-431542. Mol Pharmacol. 2002 Jul;62(1):58-64. 135. Larjava H, Heino J, Krusius T, Vuorio E, Tammi M. The small dermatan sulphate proteoglycans synthesized by fibroblasts derived from skin, synovium and gingiva show tissue-related heterogeneity. Biochem J. 1988 Nov 15;256(1):35-40. 136. Larjava H, Häkkinen L, Rahemtulla F. A biochemical analysis of human periodontal tissue proteoglycans. Biochem J. 1992 May 15;284 ( Pt 1):267-74. 137. Leask A, Holmes A, Black CM, Abraham DJ. Connective tissue growth factor gene regulation. Requirements for its induction by transforming growth factor-beta 2 in fibroblasts. J Biol Chem. 2003 Apr 11;278(15):13008-15. 138. Lee HG, Eun HC. Differences between fibroblasts cultured from oral mucosa and normal skin: implication to wound healing. J Dermatol Sci. 1999 Nov;21(3):176-82. 139. Lee TY, Chin GS, Kim WJ, Chau D, Gittes GK, Longaker MT. Expression of transforming growth factor beta 1, 2, and 3 proteins in keloids. Ann Plast Surg. 1999 Aug;43(2):179-84. 140. Lemons JM, Feng XJ, Bennett BD, Legesse-Miller A, Johnson EL, Raitman I, Pollina EA, Rabitz HA, Rabinowitz JD, Coller HA. Quiescent fibroblasts exhibit high metabolic activity. PLoS Biol. 2010 Oct 19;8(10):e1000514. 141. Lewis WH, Sun KK. Hypertrophic scar: a genetic hypothesis. Burns. 1990 Jun;16(3):176-8. 142. Li J, Cao J, Li M, Yu Y, Yang Y, Xiao X, Wu Z, Wang L, Tu Y, Chen H. Collagen triple helix repeat containing-1 inhibits transforming growth factor-b1-induced collagen type I expression in keloid. Br J Dermatol. 2011 May;164(5):1030-6. 143. Liaw L, Birk DE, Ballas CB, Whitsitt JS, Davidson JM, Hogan BL. Altered wound healing in mice lacking a functional osteopontin gene (spp1). J Clin Invest. 1998 Apr 1;101(7):1468-78. 144. Lijnen HR, Collen D. Matrix metalloproteinase system deficiencies and matrix degradation. Thromb Haemost. 1999 Aug;82(2):837-45. 145. Lin RY, Sullivan KM, Argenta PA, Meuli M, Lorenz HP, Adzick NS. Exogenous transforming growth factor-beta amplifies its own expression and induces scar formation in a model of human fetal skin repair. Ann Surg. 1995 Aug;222(2):146-54.  !  "+)!  !!!!!!! 146. Lin RY, Adzick NS. The role of the fetal fibroblast and transforming growth factorbeta in a model of human fetal wound repair. Semin Pediatr Surg. 1996 Aug;5(3):165-74. 147. Little JA, Murdy R, Cossar N, Getliffe KM, Hanak J, Ferguson MW. TGF ! 3 immunoassay standardization: comparison of NIBSC reference preparation code 98/608 with avotermin lot 205-0505-005. J Immunoassay Immunochem. 2012 Jan;33(1):66-81. 148. Lu P, Weaver VM, Werb Z. The extracellular matrix: a dynamic niche in cancer progression. J Cell Biol. 2012 Feb 20;196(4):395-406. 149. Lyons RM, Keski-Oja J, Moses HL. Proteolytic activation of latent transforming growth factor-beta from fibroblast-conditioned medium. J Cell Biol. 1988 May;106(5):1659-65. 150. Macintyre L, Baird M. Pressure garments for use in the treatment of hypertrophic scars--a review of the problems associated with their use. Burns. 2006 Feb;32(1):105. 151. Macri L, Silverstein D, Clark RA. Growth factor binding to the pericellular matrix and its importance in tissue engineering. Adv Drug Deliv Rev. 2007 Nov 10;59(13):1366-81. 152. Maeda S, Dean DD, Gomez R, Schwartz Z, Boyan BD. The first stage of transforming growth factor beta1 activation is release of the large latent complex from the extracellular matrix of growth plate chondrocytes by matrix vesicle stromelysin-1 (MMP-3). Calcif Tissue Int. 2002 Jan;70(1):54-65. 153. Mak K, Manji A, Gallant-Behm C, Wiebe C, Hart DA, Larjava H, Häkkinen L. Scarless healing of oral mucosa is characterized by faster resolution of inflammation and control of myofibroblast action compared to skin wounds in the red Duroc pig model. J Dermatol Sci. 2009 Dec;56(3):168-80. 154. Mäkelä M, Salo T, Uitto VJ, Larjava H. Matrix metalloproteinases (MMP-2 and MMP-9) of the oral cavity: cellular origin and relationship to periodontal status. J Dent Res. 1994 Aug;73(8):1397-406. 155. Malemud CJ. Matrix metalloproteinases (MMPs) in health and disease: an overview. Front Biosci. 2006 May 1;11:1696-701. 156. Margetts PJ, Gyorffy S, Kolb M, Yu L, Hoff CM, Holmes CJ, Gauldie J. Antiangiogenic and antifibrotic gene therapy in a chronic infusion model of peritoneal dialysis in rats. J Am Soc Nephrol. 2002 Mar;13(3):721-8.  !  "++!  !!!!!!! 157. Martin P. Wound healing--aiming for perfect skin regeneration. Science. 1997 Apr 4;276(5309):75-81. 158. Matheson S, Larjava H, Häkkinen L. Distinctive localization and function for lumican, fibromodulin and decorin to regulate collagen fibril organization in periodontal tissues. J Periodontal Res. 2005 Aug;40(4):312-24. 159. Matrisian LM. Metalloproteinases and their inhibitors in matrix remodeling. Trends Genet. 1990 Apr;6(4):121-5. 160. McCawley LJ, Matrisian LM. Matrix metalloproteinases: they're not just for matrix anymore! Curr Opin Cell Biol. 2001 Oct;13(5):534-40. 161. McEwan PA, Scott PG, Bishop PN, Bella J. Structural correlations in the family of small leucine-rich repeat proteins and proteoglycans. J Struct Biol. 2006 Aug;155(2):294-305. 162. McKeown ST, Barnes JJ, Hyland PL, Lundy FT, Fray MJ, Irwin CR. Matrix metalloproteinase-3 differences in oral and skin fibroblasts. J Dent Res. 2007 May;86(5):457-62. 163. Meran S, Luo DD, Simpson R, Martin J, Wells A, Steadman R, Phillips AO. Hyaluronan facilitates transforming growth factor-!1-dependent proliferation via CD44 and epidermal growth factor receptor interaction. J Biol Chem. 2011 May 20;286(20):17618-30. 164. Merkel JR, DiPaolo BR, Hallock GG, Rice DC. Type I and type III collagen content of healing wounds in fetal and adult rats. Proc Soc Exp Biol Med. 1988 Apr;187(4):493-7. 165. Merline R, Schaefer RM, Schaefer L. The matricellular functions of small leucinerich proteoglycans (SLRPs). J Cell Commun Signal. 2009 Dec;3(3-4):323-35. 166. Milanini J, Viñals F, Pouysségur J, Pagès G. p42/p44 MAP kinase module plays a key role in the transcriptional regulation of the vascular endothelial growth factor gene in fibroblasts. J Biol Chem. 1998 Jul 17;273(29):18165-72. 167. Miller MC, Nanchahal J. Advances in the modulation of cutaneous wound healing and scarring. BioDrugs. 2005;19(6):363-81. 168. Mirastschijski U, Zhou Z, Rollman O, Tryggvason K, Agren MS. Wound healing in membrane-type-1 matrix metalloproteinase-deficient mice. J Invest Dermatol. 2004 Sep;123(3):600-2.  !  "+#!  !!!!!!! 169. Miyazaki K, Hattori Y, Umenishi F, Yasumitsu H, Umeda M. Purification and characterization of extracellular matrix-degrading metalloproteinase, matrin (pump1), secreted from human rectal carcinoma cell line. Cancer Res. 1990 Dec 15;50(24):7758-64. 170. Mongiat M, Otto J, Oldershaw R, Ferrer F, Sato JD, Iozzo RV. Fibroblast growth factor-binding protein is a novel partner for perlecan protein core. J Biol Chem. 2001 Mar 30;276(13):10263-71. 171. Moriguchi T, Kuroyanagi N, Yamaguchi K, Gotoh Y, Irie K, Kano T, Shirakabe K, Muro Y, Shibuya H, Matsumoto K, Nishida E, Hagiwara M. A novel kinase cascade mediated by mitogen-activated protein kinase kinase 6 and MKK3. J Biol Chem. 1996 Jun 7;271(23):13675-9. 172. Moscatello DK, Santra M, Mann DM, McQuillan DJ, Wong AJ, Iozzo RV. Decorin suppresses tumor cell growth by activating the epidermal growth factor receptor. J Clin Invest. 1998 Jan 15;101(2):406-12. 173. Mott JD, Werb Z. Regulation of matrix biology by matrix metalloproteinases. Curr Opin Cell Biol. 2004 Oct;16(5):558-64. 174. Mulder KM. Role of Ras and Mapks in TGFbeta signaling. Cytokine Growth Factor Rev. 2000 Mar-Jun;11(1-2):23-35. 175. Mulsow JJ, Watson RW, Fitzpatrick JM, O'Connell PR. Transforming growth factorbeta promotes pro-fibrotic behavior by serosal fibroblasts via PKC and ERK1/2 mitogen activated protein kinase cell signaling. Ann Surg. 2005 Dec;242(6):880-7, discussion 887-9. 176. Muro AF, Chauhan AK, Gajovic S, Iaconcig A, Porro F, Stanta G, Baralle FE. Regulated splicing of the fibronectin EDA exon is essential for proper skin wound healing and normal lifespan. J Cell Biol. 2003 Jul 7;162(1):149-60. 177. Murphy G, Cockett MI, Ward RV, Docherty AJ. Matrix metalloproteinase degradation of elastin, type IV collagen and proteoglycan. A quantitative comparison of the activities of 95 kDa and 72 kDa gelatinases, stromelysins-1 and -2 and punctuated metalloproteinase (PUMP). Biochem J. 1991 Jul 1;277 ( Pt 1):277-9. 178. Murphy PS, Evans GR. Advances in wound healing: a review of current wound healing products. Plast Surg Int. 2012;2012:190436. 179. Murray JC. Keloids and hypertrophic scars. Clin Dermatol. 1994 Jan-Mar;12(1):2737.  !  "+$!  !!!!!!! 180. Mustoe TA, Cooter RD, Gold MH, Hobbs FD, Ramelet AA, Shakespeare PG, Stella M, Téot L, Wood FM, Ziegler UE; International Advisory Panel on Scar Management. International clinical recommendations on scar management. Plast Reconstr Surg. 2002 Aug;110(2):560-71. 181. Nagase H, Woessner JF Jr. Matrix metalloproteinases. J Biol Chem. 1999 Jul 30;274(31):21491-4. 182. Nanda S, Reddy BS. Intralesional 5-fluorouracil as a treatment modality of keloids. Dermatol Surg. 2004 Jan;30(1):54-6; discussion 56-7. 183. Narayanan AS, Page RC, Meyers DF. Characterization of collagens of diseased human gingiva. Biochemistry. 1980 Oct 28;19(22):5037-43. 184. Nauta A, Gurtner G, Longaker MT. Wound healing and regenerative strategies. Oral Dis. 2011 Sep;17(6):541-9 185. Nedelec B, Shankowsky H, Scott PG, Ghahary A, Tredget EE. Myofibroblasts and apoptosis in human hypertrophic scars: the effect of interferon-alpha2b. Surgery. 2001 Nov;130(5):798-808. 186. Nimni ME. Collagen: structure, function, and metabolism in normal and fibrotic tissues. Semin Arthritis Rheum. 1983 Aug;13(1):1-86. 187. O'Kane S, Ferguson MW. Transforming growth factor beta s and wound healing. Int J Biochem Cell Biol. 1997 Jan;29(1):63-78. 188. Occleston NL, Metcalfe AD, Boanas A, Burgoyne NJ, Nield K, O'Kane S, Ferguson MW. Therapeutic improvement of scarring: mechanisms of scarless and scar-forming healing and approaches to the discovery of new treatments. Dermatol Res Pract. 2010;2010. doi:pii: 405262. 10.1155/2010/405262. 189. Okazaki M, Yoshimura K, Uchida G, Harii K. Elevated expression of hepatocyte and keratinocyte growth factor in cultured buccal-mucosa-derived fibroblasts compared with normal-skin-derived fibroblasts. J Dermatol Sci. 2002 Nov;30(2):108-15. 190. Palaiologou AA, Yukna RA, Moses R, Lallier TE. Gingival, dermal, and periodontal ligament fibroblasts express different extracellular matrix receptors. J Periodontol. 2001 Jun;72(6):798-807. 191. Palavalli LH, Prickett TD, Wunderlich JR, Wei X, Burrell AS, Porter-Gill P, Davis S, Wang C, Cronin JC, Agrawal NS, Lin JC, Westbroek W, Hoogstraten-Miller S, Molinolo AA, Fetsch P, Filie AC, O'Connell MP, Banister CE, Howard JD, Buckhaults P, Weeraratna AT, Brody LC, Rosenberg SA, Samuels Y. Analysis of the matrix metalloproteinase family reveals that MMP8 is often mutated in melanoma. Nat Genet. 2009 May;41(5):518-20.  !  "+%!  !!!!!!! 192. Pan D, Zhe X, Jakkaraju S, Taylor GA, Schuger L. P311 induces a TGF-beta1independent, nonfibrogenic myofibroblast phenotype. J Clin Invest. 2002 Nov;110(9):1349-58. 193. Patel S, Santra M, McQuillan DJ, Iozzo RV, Thomas AP. Decorin activates the epidermal growth factor receptor and elevates cytosolic Ca2+ in A431 carcinoma cells. J Biol Chem. 1998 Feb 6;273(6):3121-4. 194. Penn JW, Grobbelaar AO, Rolfe KJ. The role of the TGF-! family in wound healing, burns and scarring: a review. Int J Burns Trauma. 2012;2(1):18-28. 195. Pitiyage GN, Slijepcevic P, Gabrani A, Chianea YG, Lim KP, Prime SS, Tilakaratne WM, Fortune F, Parkinson EK. Senescent mesenchymal cells accumulate in human fibrosis by a telomere-independent mechanism and ameliorate fibrosis through matrix metalloproteinases. J Pathol. 2011 Apr;223(5):604-17. 196. Pouyani T, Papp S, Schaffer L. Tissue-engineered fetal dermal matrices. In Vitro Cell Dev Biol Anim. 2012 Sep;48(8):493-506. 197. Profyris C, Tziotzios C, Do Vale I. Cutaneous scarring: Pathophysiology, molecular mechanisms, and scar reduction therapeutics Part I. The molecular basis of scar formation. J Am Acad Dermatol. 2012 Jan;66(1):1-10; quiz 11-2. 198. Pyagay P, Heroult M, Wang Q, Lehnert W, Belden J, Liaw L, Friesel RE, Lindner V. Collagen triple helix repeat containing 1, a novel secreted protein in injured and diseased arteries, inhibits collagen expression and promotes cell migration. Circ Res. 2005 Feb 4;96(2):261-8. 199. Quiros RM, Valianou M, Kwon Y, Brown KM, Godwin AK, Cukierman E. Ovarian normal and tumor-associated fibroblasts retain in vivo stromal characteristics in a 3-D matrix-dependent manner. Gynecol Oncol. 2008 Jul;110(1):99-109. 200. Ramfjord SP. Periodontology and Periodontics. St. Louis: Ishiyaku Euroamerica, Inc; 1979. 201. Ravanti L, Toriseva M, Penttinen R, Crombleholme T, Foschi M, Han J, Kähäri VM. Expression of human collagenase-3 (MMP-13) by fetal skin fibroblasts is induced by transforming growth factor beta via p38 mitogen-activated protein kinase. FASEB J. 2001 Apr;15(6):1098-100. 202. Rinn JL, Wang JK, Liu H, Montgomery K, van de Rijn M, Chang HY. A systems biology approach to anatomic diversity of skin. J Invest Dermatol. 2008 Apr;128(4):776-82. 203. Roberts AB. Transforming growth factor-beta: activity and efficacy in animal models of wound healing. Wound Repair Regen. 1995 Oct-Dec;3(4):408-18.  !  "+&!  !!!!!!!  204. Rodríguez D, Morrison CJ, Overall CM. Matrix metalloproteinases: what do they not do? New substrates and biological roles identified by murine models and proteomics. Biochim Biophys Acta. 2010 Jan;1803(1):39-54. 205. Rosenfeldt H, Grinnell F. Fibroblast quiescence and the disruption of ERK signaling in mechanically unloaded collagen matrices. J Biol Chem. 2000 Feb 4;275(5):308892. 206. Ruoslahti E, Yamaguchi Y. Proteoglycans as modulators of growth factor activities. Cell. 1991 Mar 8;64(5):867-9. 207. Saunders WB, Bayless KJ, Davis GE. MMP-1 activation by serine proteases and MMP-10 induces human capillary tubular network collapse and regression in 3D collagen matrices. J Cell Sci. 2005 May 15;118(Pt 10):2325-40. 208. Schor SL, Ellis I, Irwin CR, Banyard J, Seneviratne K, Dolman C, Gilbert AD, Chisholm DM. Subpopulations of fetal-like gingival fibroblasts: characterisation and potential significance for wound healing and the progression of periodontal disease. Oral Dis. 1996 Jun;2(2):155-66. 209. Schrementi ME, Ferreira AM, Zender C, DiPietro LA. Site-specific production of TGF-beta in oral mucosal and cutaneous wounds. Wound Repair Regen. 2008 JanFeb;16(1):80-6. 210. Schwartz MA. Integrins and extracellular matrix in mechanotransduction. Cold Spring Harb Perspect Biol. 2010 Dec;2(12):a005066. doi: 10.1101/cshperspect.a005066. 211. Sciubba JJ, Waterhouse JP, Meyer J. A fine structural comparison of the healing of incisional wounds of mucosa and skin. J Oral Pathol. 1978 Aug;7(4):214-27. 212. Seifert AW, Monaghan JR, Voss SR, Maden M. Skin regeneration in adult axolotls: a blueprint for scar-free healing in vertebrates. PLoS One. 2012;7(4):e32875. 213. Seifert AW, Kiama SG, Seifert MG, Goheen JR, Palmer TM, Maden M. Skin shedding and tissue regeneration in African spiny mice (Acomys). Nature. 2012 Sep 27;489(7417):561-5. 214. Selvey S, Haupt LM, Thompson EW, Matthaei KI, Irving MG, Griffiths LR. Stimulation of MMP-11 (stromelysin-3) expression in mouse fibroblasts by cytokines, collagen and co-culture with human breast cancer cell lines. BMC Cancer. 2004 Jul 25;4:40.  !  "+'!  !!!!!!! 215. Sempowski GD, Borrello MA, Blieden TM, Barth RK, Phipps RP. Fibroblast heterogeneity in the healing wound. Wound Repair Regen. 1995 Apr-Jun;3(2):12031. 216. Serini G, Bochaton-Piallat ML, Ropraz P, Geinoz A, Borsi L, Zardi L, Gabbiani G. The fibronectin domain ED-A is crucial for myofibroblastic phenotype induction by transforming growth factor-beta1. J Cell Biol. 1998 Aug 10;142(3):873-81. 217. Shah M, Foreman DM, Ferguson MW. Neutralising antibody to TGF-beta 1,2 reduces cutaneous scarring in adult rodents. J Cell Sci. 1994 May;107 ( Pt 5):113757. 218. Shah M, Foreman DM, Ferguson MW. Neutralisation of TGF-beta 1 and TGF-beta 2 or exogenous addition of TGF-beta 3 to cutaneous rat wounds reduces scarring. J Cell Sci. 1995 Mar;108 ( Pt 3):985-1002. 219. Shamis Y, Hewitt KJ, Carlson MW, Margvelashvilli M, Dong S, Kuo CK, Daheron L, Egles C, Garlick JA. Fibroblasts derived from human embryonic stem cells direct development and repair of 3D human skin equivalents. Stem Cell Res Ther. 2011 Feb 21;2(1):10. 220. Shannon DB, McKeown ST, Lundy FT, Irwin CR. Phenotypic differences between oral and skin fibroblasts in wound contraction and growth factor expression. Wound Repair Regen. 2006 Mar-Apr;14(2):172-8. 221. Shi-Wen X, Leask A, Abraham D. Regulation and function of connective tissue growth factor/CCN2 in tissue repair, scarring and fibrosis. Cytokine Growth Factor Rev. 2008 Apr;19(2):133-44. 222. Shimizukawa M, Ebina M, Narumi K, Kikuchi T, Munakata H, Nukiwa T. Intratracheal gene transfer of decorin reduces subpleural fibroproliferation induced by bleomycin. Am J Physiol Lung Cell Mol Physiol. 2003 Mar;284(3):L526-32. 223. Shirakabe K, Yamaguchi K, Shibuya H, Irie K, Matsuda S, Moriguchi T, Gotoh Y, Matsumoto K, Nishida E. TAK1 mediates the ceramide signaling to stress-activated protein kinase/c-Jun N-terminal kinase. J Biol Chem. 1997 Mar 28;272(13):8141-4. 224. Singer AJ, Clark RA. Cutaneous wound healing. N Engl J Med. 1999 Sep 2;341(10):738-46. 225. Soo C, Hu FY, Zhang X, Wang Y, Beanes SR, Lorenz HP, Hedrick MH, Mackool RJ, Plaas A, Kim SJ, Longaker MT, Freymiller E, Ting K. Differential expression of fibromodulin, a transforming growth factor-beta modulator, in fetal skin development and scarless repair. Am J Pathol. 2000 Aug;157(2):423-33.  !  "+(!  !!!!!!! 226. Sorrell JM, Caplan AI. Fibroblast heterogeneity: more than skin deep. J Cell Sci. 2004 Feb 15;117(Pt 5):667-75. 227. Sorrell JM, Caplan AI. Fibroblasts-a diverse population at the center of it all. Int Rev Cell Mol Biol. 2009;276:161-214. 228. Soucy PA, Romer LH. Endothelial cell adhesion, signaling, and morphogenesis in fibroblast-derived matrix. Matrix Biol. 2009 Jun;28(5):273-83. 229. Ständer M, Naumann U, Wick W, Weller M. Transforming growth factor-beta and p21: multiple molecular targets of decorin-mediated suppression of neoplastic growth. Cell Tissue Res. 1999 May;296(2):221-7. 230. Steed DL. The role of growth factors in wound healing. Surg Clin North Am. 1997 Jun;77(3):575-86. 231. Stephens P, Genever P. Non-epithelial oral mucosal progenitor cell populations. Oral Dis. 2007 Jan;13(1):1-10. 232. Stephens P, Davies KJ, Occleston N, Pleass RD, Kon C, Daniels J, Khaw PT, Thomas DW. Skin and oral fibroblasts exhibit phenotypic differences in extracellular matrix reorganization and matrix metalloproteinase activity. Br J Dermatol. 2001 Feb;144(2):229-37. 233. Stoff A, Rivera AA, Mathis JM, Moore ST, Banerjee NS, Everts M, Espinosa-de-losMonteros A, Novak Z, Vasconez LO, Broker TR, Richter DF, Feldman D, Siegal GP, Stoff-Khalili MA, Curiel DT. Effect of adenoviral mediated overexpression of fibromodulin on human dermal fibroblasts and scar formation in full-thickness incisional wounds. J Mol Med (Berl). 2007 May;85(5):481-96. 234. Stratton R, Rajkumar V, Ponticos M, Nichols B, Shiwen X, Black CM, Abraham DJ, Leask A. Prostacyclin derivatives prevent the fibrotic response to TGF-beta by inhibiting the Ras/MEK/ERK pathway. FASEB J. 2002 Dec;16(14):1949-51. 235. Sugimoto K, Gordon SP, Meyerowitz EM. Regeneration in plants and animals: dedifferentiation, transdifferentiation, or just differentiation? Trends Cell Biol. 2011 Apr;21(4):212-8. 236. Szpaderska AM, Zuckerman JD, DiPietro LA. Differential injury responses in oral mucosal and cutaneous wounds. J Dent Res. 2003 Aug;82(8):621-6. 237. Talior-Volodarsky I, Connelly KA, Arora PD, Gullberg D, McCulloch CA. "11 integrin stimulates myofibroblast differentiation in diabetic cardiomyopathy. Cardiovasc Res. 2012 Nov 1;96(2):265-75.  !  "#*!  !!!!!!! 238. Tervahartiala T, Pirilä E, Ceponis A, Maisi P, Salo T, Tuter G, Kallio P, Törnwall J, Srinivas R, Konttinen YT, Sorsa T. The in vivo expression of the collagenolytic matrix metalloproteinases (MMP-2, -8, -13, and -14) and matrilysin (MMP-7) in adult and localized juvenile periodontitis. J Dent Res. 2000 Dec;79(12):1969-77. 239. Toksoy A, Müller V, Gillitzer R, Goebeler M. Biphasic expression of stromal cellderived factor-1 during human wound healing. Br J Dermatol. 2007 Dec;157(6):114854. 240. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA. Myofibroblasts and mechano-regulation of connective tissue remodelling. Nat Rev Mol Cell Biol. 2002 May;3(5):349-63. 241. Tonnesen MG, Feng X, Clark RA. Angiogenesis in wound healing. J Investig Dermatol Symp Proc. 2000 Dec;5(1):40-6. 242. Tran KT, Lamb P, Deng JS. Matrikines and matricryptins: Implications for cutaneous cancers and skin repair. J Dermatol Sci. 2005 Oct;40(1):11-20. 243. Tran KT, Griffith L, Wells A. Extracellular matrix signaling through growth factor receptors during wound healing. Wound Repair Regen. 2004 May-Jun;12(3):262-8. 244. Tredget EE, Nedelec B, Scott PG, Ghahary A. Hypertrophic scars, keloids, and contractures. The cellular and molecular basis for therapy. Surg Clin North Am. 1997 Jun;77(3):701-30. 245. Tuan TL, Nichter LS. The molecular basis of keloid and hypertrophic scar formation. Mol Med Today. 1998 Jan;4(1):19-24. Review. 246. Tziotzios C, Profyris C, Sterling J. Cutaneous scarring: Pathophysiology, molecular mechanisms, and scar reduction therapeutics Part II. Strategies to reduce scar formation after dermatologic procedures. J Am Acad Dermatol. 2012 Jan;66(1):1324; quiz 25-6. 247. Vaalamo M, Kariniemi AL, Shapiro SD, Saarialho-Kere U. Enhanced expression of human metalloelastase (MMP-12) in cutaneous granulomas and macrophage migration. J Invest Dermatol. 1999 Apr;112(4):499-505. 248. von der Mark K, Gauss V, von der Mark H, Müller P. Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture. Nature. 1977 Jun 9;267(5611):531-2. 249. Wang J, Dodd C, Shankowsky HA, Scott PG, Tredget EE; Wound Healing Research Group. Deep dermal fibroblasts contribute to hypertrophic scarring. Lab Invest. 2008 Dec;88(12):1278-90.  !  "#"!  !!!!!!! 250. Wang R, Ghahary A, Shen Q, Scott PG, Roy K, Tredget EE. Hypertrophic scar tissues and fibroblasts produce more transforming growth factor-beta1 mRNA and protein than normal skin and cells. Wound Repair Regen. 2000 Mar-Apr;8(2):128-37. 251. Werner S, Grose R. Regulation of wound healing by growth factors and cytokines. Physiol Rev. 2003 Jul;83(3):835-70. 252. Westermarck J, Kähäri VM. Regulation of matrix metalloproteinase expression in tumor invasion. FASEB J. 1999 May;13(8):781-92. 253. Whitby DJ, Longaker MT, Harrison MR, Adzick NS, Ferguson MW. Rapid epithelialisation of fetal wounds is associated with the early deposition of tenascin. J Cell Sci. 1991 Jul;99 ( Pt 3):583-6. 254. Whitelock JM, Murdoch AD, Iozzo RV, Underwood PA. The degradation of human endothelial cell-derived perlecan and release of bound basic fibroblast growth factor by stromelysin, collagenase, plasmin, and heparanases. J Biol Chem. 1996 Apr 26;271(17):10079-86. 255. Widgerow AD. Cellular/extracellular matrix cross-talk in scar evolution and control. Wound Repair Regen. 2011 Mar-Apr;19(2):117-33. 256. Wienke D, MacFadyen JR, Isacke CM. Identification and characterization of the endocytic transmembrane glycoprotein Endo180 as a novel collagen receptor. Mol Biol Cell. 2003 Sep;14(9):3592-604. 257. Wight TN, Kinsella MG, Qwarnström EE. The role of proteoglycans in cell adhesion, migration and proliferation. Curr Opin Cell Biol. 1992 Oct;4(5):793-801. 258. Wijelath ES, Rahman S, Namekata M, Murray J, Nishimura T, Mostafavi-Pour Z, Patel Y, Suda Y, Humphries MJ, Sobel M. Heparin-II domain of fibronectin is a vascular endothelial growth factor-binding domain: enhancement of VEGF biological activity by a singular growth factor/matrix protein synergism. Circ Res. 2006 Oct 13;99(8):853-60. Epub 2006 Sep 28. 259. Winnemöller M, Schmidt G, Kresse H. Influence of decorin on fibroblast adhesion to fibronectin. Eur J Cell Biol. 1991 Feb;54(1):10-7. 260. Wiseman DM, Polverini PJ, Kamp DW, Leibovich SJ. Transforming growth factorbeta (TGF beta) is chemotactic for human monocytes and induces their expression of angiogenic activity. Biochem Biophys Res Commun. 1988 Dec 15;157(2):793-800. 261. Woessner JF Jr. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 1991 May;5(8):2145-54.  !  "#)!  !!!!!!! 262. Wong JW, Gallant-Behm C, Wiebe C, Mak K, Hart DA, Larjava H, Häkkinen L. Wound healing in oral mucosa results in reduced scar formation as compared with skin: evidence from the red Duroc pig model and humans. Wound Repair Regen. 2009 Sep-Oct;17(5):717-29. 263. Wu Y, Zhang Q, Ann DK, Akhondzadeh A, Duong HS, Messadi DV, Le AD. Increased vascular endothelial growth factor may account for elevated level of plasminogen activator inhibitor-1 via activating ERK1/2 in keloid fibroblasts. Am J Physiol Cell Physiol. 2004 Apr;286(4):C905-12. 264. Yamada KM, Pankov R, Cukierman E. Dimensions and dynamics in integrin function. Braz J Med Biol Res. 2003 Aug;36(8):959-66. 265. Yamaguchi K, Shirakabe K, Shibuya H, Irie K, Oishi I, Ueno N, Taniguchi T, Nishida E, Matsumoto K. Identification of a member of the MAPKKK family as a potential mediator of TGF-beta signal transduction. Science. 1995 Dec 22;270(5244):2008-11. 266. Yamaguchi Y, Mann DM, Ruoslahti E. Negative regulation of transforming growth factor-beta by the proteoglycan decorin. Nature. 1990 Jul 19;346(6281):281-4. 267. Yang Z, Kyriakides TR, Bornstein P. Matricellular proteins as modulators of cellmatrix interactions: adhesive defect in thrombospondin 2-null fibroblasts is a consequence of increased levels of matrix metalloproteinase-2. Mol Biol Cell. 2000 Oct;11(10):3353-64. 268. Yu L, Hébert MC, Zhang YE. TGF-beta receptor-activated p38 MAP kinase mediates Smad-independent TGF-beta responses. EMBO J. 2002 Jul 15;21(14):3749-59. 269. Zanetti M, Braghetta P, Sabatelli P, Mura I, Doliana R, Colombatti A, Volpin D, Bonaldo P, Bressan GM. EMILIN-1 deficiency induces elastogenesis and vascular cell defects. Mol Cell Biol. 2004 Jan;24(2):638-50. 270. Zardi L, Carnemolla B, Siri A, Petersen TE, Paolella G, Sebastio G, Baralle FE. Transformed human cells produce a new fibronectin isoform by preferential alternative splicing of a previously unobserved exon. EMBO J. 1987 Aug;6(8):233742. 271. Zhang K, Garner W, Cohen L, Rodriguez J, Phan S. Increased types I and III collagen and transforming growth factor-beta 1 mRNA and protein in hypertrophic burn scar. J Invest Dermatol. 1995 May;104(5):750-4. 272. Zhou HM, Wang J, Elliott C, Wen W, Hamilton DW, Conway SJ. Spatiotemporal expression of periostin during skin development and incisional wound healing: lessons for human fibrotic scar formation. J Cell Commun Signal. 2010 Jun;4(2):99107.  !  "#+!  !!!!!!!  273. Zurada JM, Kriegel D, Davis IC. Topical treatments for hypertrophic scars. J Am Acad Dermatol. 2006 Dec;55(6):1024-31.  !  "##!  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073674/manifest

Comment

Related Items