Cytoskeleton regulators in the control of corneal fibrosis

31 3.2 Effect of various dosing regimes of topical application of TGF-β1 on the transformation of corneal keratocytes-to-myofibroblasts in corneal fibrosis in vivo……….. 46 4.3 Co-localiz

Acknowledgements

I would like to express my sincere gratitude to Professor Roger W Beuerman for his supervision and guidance throughout this project I am grateful for his comments and advice I would like to thank all the staff of SERI especially Sia-Wey Yeo, Wing-Sam Lee, Jennifer Ng, Queenie Tan, Candice Ho, Jia Lin, Wan-e Lim and Dr Amutha Veluchamy Barathi, who assisted me in many ways My work was supported by the grant: National Research Council of Singapore grant (R739/23/2010) and Singhealth Foundation grant (R835/30/2011)

I would like to thank my family for their love and support throughout my endeavours

TABLE OF CONTENTS

CHAPTER I INTRODUCTION…….……… 1

1 Cornea structure……… ……… 1

1.1.1 Corneal stroma structureh ……… …… 1

1.1.2 Corneal stroma keratocytes……… 2

1.1.3 Cytoskeleton in biology……… … 3

1.1.4 Cytoskeleton regulators in biology……….……… 3

1.2 Corneal fibrosis as an adverse outcome……… ……… 3

1.3 Corneal stroma keratocytes in corneal fibrosis……… 4

1.4 Mediators trigger corneal fibrosis……… ……… 5

1.4.1 TGF-β1 induced fibrosis……… 5

1.4.2 TGF-β1 induced corneal fibrosis……… ……… 6

1.5 Cytoskeleton in fibrosis……….…… 7

1.5.1 Cytoskeleton regulators in fibrosis……… 8

1.6 Current treatment of corneal fibrosis……….……… 10

1.7 Answering the Experimental Questions……… 12

1.8 Hypothesis……….…… 12

1.8.1 Specific aim……… 13

CHAPTER II MATERIAL AND METHODS……… 14

2.1 Ethics……… 14

2.2 Anterior Keratectomy Procedure……… 14

2.3 Animal Experimental Groups……… …… 15

2.4 Gold-Chloride Procedure……….… 16

2.5 Cytoskeleton Regulators RT² Profiler™ PCR Array……… 16

2.6 Moesin siRNA in vivo delivery……… ……… 17

2.7 List of laboratory techniques……… …… 18

2.7.1 Preparation of tissue……… 18

2.7.2 Use of exogenous protein……… 18

2.7.3 Histological Evaluation ……….……… 19

2.7.4 Immunochemistry……… … 21

2.7.5 Imaging……… 22

2.7.6 RNA Extraction and quantification……… ……… 22

2.7.7 Reverse transcription and Polymerase chain reaction……… 24

2.7.8 Real time RT-PCR……… 25

2.7.9 Total protein extraction……… 27

2.7.10 BCA protein assay……… 28

2.7 11 Western blot analysis……… 28

2.8 Statistical analysis……… 30

Chapter III TGF-β1-induced corneal fibrosis……… 31

3.1 Phenotypic changes of corneal stroma keratocytes in response to the anterior keratectomy……… 31

3.2 Effect of various dosing regimes of topical application of TGF-β1 on the transformation of corneal keratocytes-to-myofibroblasts in corneal fibrosis in vivo……… ……… 33

3.3 TGF-β1 induced transformation of corneal stroma keratocytes-to-myofibroblasts in the corneal fibrosis……….……… 39

3.4 Up-regulation of cytoskeleton regulators in corneal fibrosis in vivo……… 41

Chapter IV Moesin and fibrosis……… 46 4.1 Moesin functions in cytoskeleton……… ……… 46

4.3 Co-localization of moesin and α-SMA within corneal stroma keratocytes in

corneal fibrosis……… 47 4.4 Characterization of moesin expression in the corneal fibrosis……… 49 4.5 Role of moesin in the corneal fibrosis in vivo……… 51

4.5.1 In vivo delivery of moesin siRNA into the cornea after an anterior

keratectomy……… 51

4.5.2 The transformation of corneal stroma keratocytes-to-myofibroblasts in the

corneal fibrosis is moesin-dependent……… 56 4.5.3 Corneal haze development in the injured cornea……… 60 Chapter V Moesin and signaling in corneal fibrosis 63 5.1 Introduction to Smad signaling in fibrosis……… 63 5.2 Effect of TGF-β1 on the activation of Smad 2 and Smad 3 in the corneal

fibrosis……… 64 5.3 Effect of moesin on TGF-β1-stimulated activation of Smad2 and Smad3 in

Vivo……… 75

6.3 Clinical application……… 79

6.4 Future studies……… 81

Chapter VII References……… 85

Chapter VIII Appendices……… 131

Appendix A: Experimental animal groups……… 132

Appendix B: Moesin siRNA sequences……… 132

Appendix C: Antibodies……… 132

Appendix D: PCR primers……… 133

Chapter X Supplementary……… 134

A Grants: ……… 134

B Patents……… 134

C Presentations at Conferences……… 134

D Publications……… 136

SUMMARY

The avascular cornea accounts for about two-thirds of the total refractive power of the eye,therefore maintenance of corneal transparency is critical for focusing light onto the retina The impact on vision is highlighted by the finding that corneal scar as a contributor to “corneal blindness” is the third leading cause of blindness world-wide Corneal scar may develop as sequel to infections and a wide spectrum of corneal stromal injuries such as refractive surgery (LASIK, etc.) Currently, corneal transplantation is the major treatment regime for corneal scar To develop more specific medical therapies or as an adjunct to surgery to prevent scar formation, the regulation of corneal scar must be understood

Fibrosis is the cellular process that leads to the formation of scar A strong association

of fibrosis with “cytoskeleton regulators” suggests this class of proteins that links the intracellular cytoskeleton with the extracellular environment may be a target for therapeutic development; however, as this is a large group of proteins, evidence for involvement of specific members of this class of proteins in the corneal fibrosis has not been developed Recently, interest in the ERM (ezrin/radixin/moesin) family members has shown that these proteins are the organizer of membrane domains and act as links to the cytoskeleton as well as signalling pathways involved in many cellular processes Cytoskeleton disruption agents have been shown to prevent fibrosis may involve this family of proteins.

Our clinical target is the early phase of corneal wounding, before an opaque corneal scar forms In response to corneal injury, the transformation of corneal stroma keratocytes-to-myofibroblasts is predominently responsible for the corneal fibrosis

To define the in vivo role of specific cytoskeleton regulators in mediating the corneal fibrosis, a mouse model with corneal fibrosis stimulated by the anterior keratectomy was established TGF-β1 was topically applied to accelerate the corneal fibrosis process since it activates corneal stroma keratocytes to a myofibroblast phenotype expressing α-SMA accompanied by altered expression patterns of ECM components The main results of the present study are that (1) in the corneal fibrosis, moesin was identified as the most highly induced gene among the 84 cytoskeleton regulator genes using cytoskeleton regulators RT² Profiler™ PCR array (Chapter III 3.4); (2) The up-regulation of moesin in the corneal fibrosis was confirmed by RT-PCR (Chapter III 3.4) and western blot (Chapter IV 4.4) in a time-dependent manner; (3) the appearance of myofibroblasts was analyzed by examining α-SMA expression.Dual immunofluorescent staining showed that moesin colocalized with α-SMA within corneal stroma keratocytes in the corneal fibrosis (Chapter IV 4.3); (4) Up-regulation

of α-SMA and moesin in the corneal fibrosis was reduced by moesin siRNA (Chapter

IV 4.5.1 and 4.5.2); (5) Moesin siRNA reduced corneal opacification in corneal fibrosis (Chapter IV 4.5.3); (6) Activation of Smad 2 and Smad 3 in corneal fibrosis was reduced by moesin siRNA (Chapter V 5.3)

In conclusion, moesin siRNA decreased the transformation of corneal stroma keratocytes-to-myofibroblasts in the corneal fibrosis, as defined by the expression of α-SMA, through the reduction of activated forms of Smad 2 and Smad 3 Moesin may

be a potential drug target for inhibiting corneal fibrosis and details of moesin-related signaling pathways would be critical for understanding corneal fibrosis

LIST OF TABLES

1 Experimental animal groups……… 133

2 List of moesin siRNA sequence……… 133

3 List of antibodies……… 132

4 List of PCR primers……… 132

5 Various dosing regimes for topical application of TGF-β1 on the corneal stroma after the anterior keratectomy……… 35

LIST OF FIGURES

1 Histological analysis of corneal wound healing after an anterior

keratectomy………20

2 Phenotypic changes of corneal stromal keratocytes in response to

the anterior keratectomy ……… 32

3 Effect of various dosing regimes for TGF-β1 on the transformation of

corneal keratocytes-to-myofibroblasts, as defined by the expression of α-SMA

in the corneal stroma after the anterior keratectomy 35

4 Keratocyte-to-myofibroblast transformation induced by topical application

of TGF-β1, as defined by α-SMA expression 40

5 Results for the response of cytoskeleton regulators to topical application

of TGF-ß1 after an anterior keratectomy at PO day 1 analyzed by

RT² Profiler PCR-array ……….………42

6 RT-PCR quantification of mRNA levels for cytoskeleton regulators induced

by topical application of TGF-ß1 following an anterior keratectomy 44

7 Dual immunofluorescent staining demonstrated that moesin co-localized with

α-SMA within corneal stroma keratocytes in the corneal fibrosis……… 48

8 Up-regulation of moesin in the corneal fibrosis……… ………50

9 Distribution of fluorescein–labeled moesin siRNA after in vivo delivery of moesin siRNA by iontophoresis into the cornea……….52

10 Effect of moesin siRNA on the expression of moesin in the corneal fibrosis ……….54

11 Effect of moesin siRNA on the transformation of corneal keratocytes-to-myofibroblasts in the corneal fibrosis ……….……… 58

12 In vivo slit lamp macroscopic observation of the cornea in group 2-AK and group 5-AKTSi………61

13 Effect of TGF-β1 on Smad 2 activation ……… 65

14 Effect of TGF-β1 on Smad 3 activation ……… …67

15 Effect of moesin on Smad 2 activation ……… ……….69

16 Effect of moesin on Smad 3 activation ……….……… 70

TGF- Transforming Growth Factor

ECM -extracellular matrix

siRNA- short interfering ribonucleic acid

ANOVA -Analysis of variance

PBS- phosphate buffered saline

BSA -bovine serum albumin

DEPC -Diethyl Pyro carbonate

SDS- sodium dodecyl sulphate

dNTP -deoxy nucleoside (5’-) tri phosphate

DTT 1, 4-dithiothreitol

RNA Ribonucleic acid

cDNA -complimentary deoxyribonucleic acid

PCR -Polymerase Chain Reaction

RT-PCR Reverse Transcriptase - Polymerase Chain Reaction

CHAPTER I INTRODUCTION

1.1 Cornea structure

The cornea is an avascular tissue and is comprised of five layers: The corneal surface consists of a multi-layered (5-7 cell layers) non-keratinized, stratified squamous epithelium, which is separated from the thick collagenous stroma layer

by a condensed layer of collagen-Bowman’s layer Backing the stroma and separating it from the anterior chamber of the eye is a simple squamous monolayer-endothelium, which consists of a monolayer of cells attached to the stroma via a thin acellular layer known as Descemet's membrane (Beuerman RW

et al 1998) These tissues can be separated easily from each other because each of these tissues is homogeneous in cell type and arranged in the simple layers Therefore the cellular events that occur during corneal wound healing can be documented precisely with these unique structural features (Weimar V et al 1960)

1.1.1 Corneal stroma structure

Corneal stroma constitutes about 90% of the total thickness of human cornea The stroma composed with corneal stroma keratocytes, extracellular matrix, and bundles of highly ordered collagenous fibers aligned in parallel layers with respect to one another (Meek KM et al 2001 and 2004; Funderburgh JL et al 1997 and 2000; Dunlevy JR et al 2000; Chakravarti S et al 2002; Kao WW et al 2002) The precise arrangement of keratocytes and extracellular matrix components in the corneal stroma are important for corneal function because they account for the major contribution towards corneal curvature, refractive power and transparency,

and allow free passage of light (Ishizaki M et al 1994; Maurice DM et al 1957 and 1962)

1.1.2 Corneal stroma keratocytes

Corneal stroma keratocytes are neural crest-derived cells and major cell type reside in the corneal stroma which synthesize extracellular matrix during development and maintain homeostasis in adults Møller-Pedersen and coworkers (1994 and 1997) showed that human cornea contains 2.0-3.5 million stroma keratocytes Light passes through up to 100 layers of light-scattering stroma keratocytes when traveling through human cornea (Møller-Pedersen et al 2004) Corneal stroma keratocytes are a principle source of light scattering (Møller-Pedersen et al 2004), and thus play an essential role in the maintenance of corneal transparency (Hay ED et al 1980)

Normal corneal stroma keratocytes are quiescent, flattened and sparsely distributed between the collagen lamellae of the corneal stroma (Nishida T et al 1988; Smelser GK et al 1965; Smith JW et al 1969); filamentous actin in normal corneal stroma keratocytes is not organized into the stress fibers (Netto MV et al 2006a) Following corneal stroma injury, corneal stroma keratocytes adjacent to the wound edge transform into the repaired cells-myofibroblasts (Martin P et al 1997; Fini ME et al 1999; Matsuda H et al 1973), which is responsible for the reduction of cornea transparency and has been identified as the predominant biological eventresponsible for corneal fibrosis (Netto MV et al 2006a; Mohan

RR et al 2003; Jester JV et al 1987, 1994, 1996, 1999b)

1.1.3 Cytoskeleton in biology

The cytoskeleton is a dynamic cell organelle which tranduces external environmental events to initiate changes in cell motility, migration and morphological changes (Brown BK et al 2001; Pollard TD et al 2003) Cytoskeleton contains three main components: microfilaments, intermediate filaments, and microtubules (Heuser JE et al 1980; Fey EG et al 1984)

micro-1.1.4 Cytoskeleton regulators in biology

Rearrangements of cytoskeleton are central to the aforementioned function of cytoskeleton Several studies have highlighted the importance of cytoskeleton regulators in the control of dynamic rearrangements of cytoskeleton (Ono S et al 2007; Silacci P et al 2004; dos Remedios CG et al 2003; Maciver SK et al 2002; Cooper JA et al 2000; Sun HQ et al 1999; Weeds AG et al 1991)

1.2 Corneal fibrosis as an adverse outcome

The precise architecture of the cornea is essential to maintain normal vision After corneal infection, or corneal stroma injury (ocular trauma including injury like large epithelial debridement or alkali burns or surgery such as refractive surgery, e.g., LASIK, etc.)(Chen JJ et al 1990 and 1991; Jester JV et al 1995 and 1997; Chew SJ et al 1995; Funderburgh JL et al 1989) that provide sufficient stimulus to disrupt basement membrane, corneal fibrosis develops as a wound-healing response, which interferes with the normal corneal structure (Saika S et al 2009) and causes visual impairment (Jester JV et al 1997) The impact on vision is

highlighted by the finding that corneal fibrosis as a contributor to “corneal blindness” is the third leading cause of blindness world-wide (Foster A et al 2005)

1.3 Corneal stroma keratocytes in corneal fibrosis

During corneal wound healing, the transformation of corneal stroma keratocyte to differentiated myofibroblast phenotype is the hallmark of corneal fibrosis (Netto

MV et al 2006a; Mohan RR et al 2003; Jester JV et al 1987, 1994, 1996, 1999b) Injury to the corneal stroma (penetrating keratoplasties, laser refractive surgery, corneal trauma, keratectomy, or photoablation by excimer laser) results in the apoptosis of the corneal stroma keratocytes in the wound region (He J et al 2008; Fini ME et al 1999; Wilson SE et al 2002) Corneal stroma keratocytes near the wound edge become activated (Netto MV et al 2006a) and differentiated into spindle shaped myofibroblasts (Mohan RR et al 2003) that proliferate and migrate towards the wound region, and repopulate the acellular wound area (Moller-Pedersen T et al 1998a, 1998b and 1998c; Zieske JD et al 2001) That myofibroblasts reduce corneal transparency (Wilson SE et al 2001) attributes to three reasons: firstly, reduced intracellular crystallin production of myofibroblasts (Jester JV et al 1999b and 1999c); secondly, myofibroblasts within the wound region cause increased scattering of light (Jester JV et al 1999b and 1999c); thirdly, the deposition of the fibrous tissue initiated by myofibroblasts contributes

to the regression of the refractive power (Jester JV et al 1999b and 1999c) The appearance of myofibroblasts temporally correlates with increased amount of rabbit corneal fibrosis (Moller-Pedersen et al 1998a and 1998b), and the

disappearance in myofibroblasts also correlates with reduced rabbit corneal fibrosis (Jester JV et al 1999b)

1.4 Mediators trigger corneal fibrosis

Since the transformation of keratocytes into myofibroblastshas been identified as the predominant biological process responsible for corneal fibrosis (Jester et al 1999a; Jester et al 1999b; Folger PA et al 2001; Netto MV et al 2006a; Imanishi J

et al 2000), an important issue is to characterize the mediators involved in the regulation of this phenotype transition The transformation of keratocytes to myofibroblasts occurs as a consequence of multiple stimuli, but focus is put on TGF-β1 and cytoskeleton regulators in this thesis since the cytoskeletal tension and TGF-β1 play a main role in regulating the acquisition and maintenance of the myofibroblast phenotype (Hinz B et al 2001b; Arora PD et al 1999a; Blaauboer

ME et al 2011; Hinz B et al 2003)

1.4.1 TGF-β1 induced fibrosis

TGF-β1 is a multifunctional regulatory cytokine and has impact on diverse cellular activities (Sporn MB et al 1992) Under ideal circumstances, TGF-β1 leads to the normal tissue architecture remodeling; however, excessive TGF-β1 prevents myofibroblast apoptosis and contributes to a pathologic tissue repair process- fibrosis (Zhang HY et al 1999) TGF-β1 has been linked to a variety of the human fibrotic disease (Leask A et al 2004) For example, persistent elevation

of TGF-β1 in the kidney leads to tubulointerstitial fibrosis (Border WA et al

1994); increased TGF-β1 expression in chronic liver disease is a hallmark of hepatic fibrosis (Fausto N et al 1991); TGF-β1 contributes to systemic sclerosis which is an autoimmune disease characterized by excessive fibrosis (Liu X et al 2005) Increased amounts of TGF-β protein was observed in the lung biopsies from patients with pulmonary fibrosis (Broekelmann TJ et al 1991) Active TGF-

β in astrocytes induces fibrosis formation (Asher RA et al, 2000) In the experimental animal model, transgenic mice with aberrant TGF-β1 expression developed the fibrosis of liver and kidney (Clouthier DE et al 1997)

1.4.2 TGF-β1 induced corneal fibrosis

TGF-β1 is a central mediator of corneal fibrosis (Saika S et al 2009) The transformation of corneal stroma keratocyte to myofibroblast phenotype is a predominant biological event responsible for the corneal fibrosis (Netto MV et al 2006a; Mohan RR et al 2003; Jester JV et al 1987, 1994, 1996, 1999b) TGF-β1 induces the transformation of corneal stroma keratocyte to a myofibroblastic state

as seen in the corneal fibrosis (Tandon A et al 2010; Jester JV et al 1999a; Kaur H

et al 2009; Karamichos D et al 2010; Carrington LM et al 2006) The expression

of α-SMA has been used extensively in research and pathology to monitor this transformation (Carrington LM et al 2006; Darby I et al 1990; Tomasek JJ et al 2002) The TGF-β1-induced corneal fibrosis model has been extensively used to investigate signaling pathways involved in fibrosis and to explore anti-fibrotic therapeutic targets In vitro, TGF-β1 induced keratocytes to develop myofibroblastic characteristics in the cultured human (Ohji M et al 1993) and

rabbit corneal keratocytes (Jester JV et al 1996 and 1999a; Masur SK et al 1996) TGF-β1 stimulated cultured human corneal keratocytes to generate a model that resembles in vivo human corneal fibrotic processes (Karamichos D et al 2010) In

an organ culture study, when bovine corneas were injuried with a trephine, myofibroblasts differentiation was induced in the corneal stroma after topical application of 100ng/ml of TGF-β1 (Carrington LM et al 2006) Jester JV and his colleagues (1997) showed that neutralizing antibody to inactivate TGF-β1 reduced the rabbit corneal fibrosis after lamellar keratectomy In the rabbit eye, fibrosis developed after photorefractive keratectomy followed by topical application of TGF-β1 (1ug/ml) (Myers JS et al 1997)

1.5 Cytoskeleton in fibrosis

The defining criterion for myofibroblasts is the neo-expression of α-SMA (a contractile cytoskeletal element that imparts contractile properties to the cell) (Jester JV et al 1994) and the incorporation of α-SMA into stress fibers (Dugina V

et al 2001; Katoh K et al 2001) The transformation of keratocytes to myofibroblasts is characterized by a change in the morphology from stellate to spindle-shape associated with rearrangement of cytoskeleton to form stress fibers (Jester JV et al 1994) Chaurasia SS et al (2009) suggested that rearrangement of cytoskeleton architecture in the corneal stroma keratocytes results in increased expression of α-SMA after rabbit corneal injury

Myofibroblast differentiation in mesangial cells is modulated by structure and composition of the cytoskeleton (Patel K et al 2003) Fibroblasts-to-myofibroblasts differentiation in the lung fibrosis is controlled by cytoskeleton

tension (Blaauboer ME et al 2011) Cytoskeleton disruption agents prevented fibrosis (Hordichok AJ et al 2007) Fibrosis after spinal cord injury (Hellal F et al 2011) and fibrosis associated with systemic sclerosis (Liu X et al 2005) are decreased by microtubule stabilization

1.5.1 Cytoskeleton regulators in fibrosis

Differentiated myofibroblasts show a significant increase in cytoskeleton regulators content compared with fibroblasts (Dugina V et al 2001). Malmström

et al (2004) showed that TGF-β1 up-regulated the expression of a variety of cytoskeleton regulators (cofilin, profilin etc.) The details about several cytoskeleton regulators which are implicated to be involved in the fibrosis are shown below

Mechanical stimulation is transduced from extracellular environment into intracellular cytoskeleton of myofibroblasts through transmembrane receptors-integrins that link to intracellular stress fibers through a variety of cytoskeleton regulators (vinculin, talin, paxillin, gelsolin, FAK, ezrin–radixin–moesin proteins) (Geiger B et al 2001) Phenotypic transition of human lung fibroblasts-to-myofibroblasts induced by TGF-β1 depends on integrin signaling via focal adhesion kinase (FAK) (Thannickal VJ et al 2003) FAK, a nonreceptor protein tyrosine kinase, is central for the mechanoresponses, and cytoskeletal organization (Schober M 2007; Albinsson S et al 2007; Mitra SK et al 2005) FAK activation regulated myofibroblast differentiation by down-regulating α-SMA, thereby preventing fibrosis (Greenberg RS et al 2006) FAK silencing was

demonstrated to attenuate the rise in the fibrosis in overloaded left ventricles (Clemente CF et al 2009) Increased multifocal interstitial fibrosis was found in mice with inactivation of FAK in cardiomyocytes (Peng X et al 2006)

Paxillin has a pivotal role in transducing signals from extracellular environment into intracellular cytoskeleton (Turner CE et al 2000) and mediates the rearrangement of actin cytoskeleton into contractile elements (Cowin AJ et al 2003; O’Kane S et al 1997) The ability of paxillin to modulate fibronectin activities in wound healing may be important in the mechanism of fibrosis-free fetal wound repair (Cowin AJ et al 2003) The expression of paxillin was up-regulated in the transition of cardiac fibroblast to myofibroblastic phenotype (Santiago JJ et al 2010) Fibroblasts cultured from fibrotic lesionsof patients with systemic sclerosis (fibrotic disease) showed elevated expression of paxillin (Shi-Wen X et al 2004)

Gelsolin and cofilin both have a critical role in the cytoskeleton rearrangement (Sun H et al 1995; Burtnick LD etal 1997; Kwiatkowski DJ 1986; Yin HL et al 1987) Stress fiber formation is one of the myofibroblast features (Hinz B et al 2001) Cofilin regulates stress fiber formation (Bosselut N et al 2010) In porcine aortic cardiac valve interstitial cells, cofilin is a marker of myofibroblast transformation and is required for SMA incorporation into stress fibers (Pho M et

al 2008) The formation of stress fibers was delayed in gelsolin-depleted cells (Arora PD et al 1999) Mice deficient in gelsolin are protected from bleomycin-induced fibrosis and gelsolin expression is crucial for the pulmonary fibrosis (Oikonomou N et al 2009)

Profilin participates in the reorganization of the cytoskeleton (Hinz B et al 2001b) Profilin mRNA was observed to be up-regulated by using in situ hybridization in hypertrophic fibrosis (Wu J et al 2004)

Vinculin, moesin, and ezrin, were elevated in pulmonary fibrosis (Fukuda Y et al 1995; LeRoy EC et al 1974; Shi-Wen X et al 2004; Bogatkevich GS et al 2008)

1.6 Current treatment of corneal fibrosis

Mitomycin C (MMC) is used intraoperatively to inhibit keratocyte proliferation and activation forthe prevention of the corneal fibrosis (Thornton I et al 2008; Bedei A et al 2006; Camellin M et al 2004). Although MMC reduces the sub-epithelial fibrosis after refractive corneal surgery, several complications reported with its topical use, such as corneal abnormal wound healing, limbal/scleral necrosis, endothelial damage, and loss of keratocytes (Dougherty PJ et al 1996; Safianik B et al 2002; Rubinfeld RS et al 1992; Wu KY et al 1999)

TGF-β1 is another major target of developing anti-fibrotic strategies Many approaches focusing on the blockade of TGF-β1 signaling, involving interfering the ligand or its receptor (Netto MV et al 2006a; Chang JH et al 1998; Bilgihan K

et al 2000; Nassaralla BA et al 1995; Cellini M et al 2006; Arora R et al 2005; Sharma A 2009; Arora R et al 2005) by using antibodies (Shah M et al 1995; Cordeiro MF et al 1999), antagonists (Grisanti S et al 2005), and nucleotide-based methods (Cordeiro MF et al 2003), which ameliorate the fibrotic process in the animal model of fibrosis (Ueno H et al 2000; Giri SN et al 1993; Wang Q et al 1999) For example, the development of dermal or corneal fibrosis was blocked

by the application of neutralizing antibodies to inactivate TGF-β (Jester et al 1997; Shah M et al 1994; Shah M et al 1995; Moller–Pedersen T et al 1998b) However, because TGF-β plays many essential roles in a variety of developmental and pathological processes (Massague J et al 1998), including immune regulation, cancer surveillance, wound healing, and TGF-β is an archetype of pleiotropy; therefore, as a focus of efforts to reverse fibrosis, other complications may arise (Massague J et al 1998; Mohan RR et al 2005 and 2010)

Several other treatments have also been proposed to treat corneal fibrosis These include topical application of steroids (O’Brart DP et al 1994) , human amniotic membrane (Muller M et al 2009), urokinase plasminogen activator (Csutak A et

al, 2004), and interferon α2b (Gillies MC et al 1996) To date, the results have been quite disappointing and the tested compounds are either ineffective or accompanied by various side effects Consequently, corneal transplantation is the major treatment regimens for corneal fibrosis (Karamichos D et al 2010), despite its limitations, such as postoperative complications and the lack of high-quality donor corneas (Galiacy SD et al 2011)

To develop more specific medical therapies or as an adjunct to surgery, a more detailed understanding of the mechanisms and regulation of corneal fibrosis should be useful to develop better targets

1.7 Answering the Experimental Questions

The goal of this work has been to provide new information toward answering the question “are cytoskeletal regulators involved in the control of fibrosis in the cornea”

1.8 Hypothesis

Based on the strong associations of fibrosis with “cytoskeleton regulators”, my hypothesis is that this class of proteins which links the intracellular cytoskeleton with the extracellular environment may be a target for therapeutic development for corneal scar; however, as this is a large group of proteins, evidence for involvement of specific members of this class of proteins in the development of corneal fibrosis has not been developed

1.8.1 Specific aim

The specific aims in this study were:

1 To develop a mouse model with corneal fibrosis

2 To dissect specific cytoskeleton regulators active in corneal fibrosis

3 To evaluate the in vivo role of specific cytoskeleton regulators in the regulation of corneal fibrosis

4 To investigate the role of specific cytoskeleton regulators in the signalling pathways relevant to corneal fibrosis

‘;

CHAPTER II MATERIAL AND METHODS

2.1 Ethics

All experiments were conducted in compliance with the Association for Research

in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and with approval of the SingHealth Institutional Animal Care and Use Committee (IACUC)

2.2 Anterior Keratectomy Procedure

After pre-operative ocular slit-lamp examination for exclusion criteria such as ocular disease, wound, or infection mice were anesthetized by intraperitoneal injections of xylazine (10 mg/kg body weight, Troy Laboratories, Smith-field, Australia) and ketamine (80 mg/kg body weight, Ketamine, Parnell Laboratories, Alexandria, Australia) One drop of 0.5% proparacaine hydrochloride ophthalmic solution was applied to the eye prior to surgery An anterior keratectomy was performed in a similar fashion as in previous work (Reidy JJ et al 1996) Anesthetized mice were placed under a dissecting microscope (Zeiss, Oberkochen, Germany) and light source, and cornea wound was made centrally by light application of a 2 mm trephine, to cut partially through the cornea which permitted a disc of corneal epithelial-stroma tissue within the wound to be removed with a #5 Dumont forceps Fucithalmic ointment (Leo Pharmaceutical Products, Ballerup, Denmark) was applied after surgery and daily Animals were observed during the recovery period until awakening in a heated chamber In different experiments the post-operative (PO) period was 1, 3 or 5 days At the

conclusion of the experiment mice were killed by an overdose of sodium pentobarbital, cornea stroma were removed and frozen for RNA and protein analyses

2.3 Animal Experimental Groups

C57BL6 mice (8–10 weeks old) weighing 20–25 grams were obtained from the National University of Singapore Prior to and during the experiments, the animals were housed in appropriate cages with free access to food and water in the Singapore Experimental Medicine Center (AAALAC approved)

Mice were divided into the following groups as shown in appendix A: Group NC: no procedure (n=279 eyes); Group 2-AK: anterior keratectomy (n=396 eyes); Group 3-AKT: myofibroblast induction (n=435 eyes) by anterior keratectomy and topical application of TGF-β1 (1µg/ml) TGF-β1 was applied 3 times at the day of anterior keratectomy and post-operative (PO) day 3; however, for those mice sacrificed at PO day 1 and 3, TGF-β1 was applied 3 times only on the day of anterior keratectomy; Group 4-AKSi: anterior keratectomy with moesin siRNA (n=186 eyes) SiRNA was delivered into the cornea by iontophoresis everyday from the day of anterior keratectomy until mice were sacrificed An identical application of control siRNA to the contralateral eye was made using the same conditions; Group 5-AKTSi: Anterior keratectomy plus TGF-β1 and moesin siRNA (n=360 eyes) as in group 4 with TGF-β1 applied using the same conditions as in group 3

1-2.4 Gold-Chloride Procedure

The gold-chloride technique, as previously described (Beuerman RW et al 1996), was used to visualize the morphology of corneal keratocytes Micrographs were made with microscope (Zeiss, Oberkochen, Germany)

2.5 Cytoskeleton Regulators RT² Profiler™ PCR Array

Cytoskeleton regulators RT² Profiler™ PCR Array (Sabioscience, Federick, MD) was used to analyze the expression of 84 cytoskeleton regulatory genes plus five housekeeping genes Controls were included for RNA quality, genomic DNA contamination, and general PCR performance

To complete the PCR array procedure, RNA (500ng) extracted from corneal stroma (corneal stroma after removing epithelium and endothelium) from the group 1-NC (no procedure) and group 3-AKT (anterior keratectomy with topical application of TGF-β1), was converted into first strand cDNA, performed as previously described (Zhu HY et al 2010) Equal volumes of the mixture were aliquoted into each well of the same PCR array plate which contained pre-dispensed gene specific primer sets The real-time PCR cycling program was run Fold-change calculations based on the ∆∆Ct method with normalization of the data using the corneal stroma from the group 3-AKT to the data of the normal corneal stroma from the group 1-NC was automatically performed by the integrated web-based software

(http://pcrdataanalysis.sabiosciences.com/pcr/arrayanalysis.php)

2.6 Moesin siRNA in vivo delivery

Mouse moesin siRNA sequences (Qiagen, California, USA) are seen in appendix

B Total RNA and protein were isolated from corneal stroma, and the assessment

of moesin suppression was done by westernblot The uptake of labeled moesin siRNA in the cornea was verified by fluorescence microscopy Moesin siRNA powder was constituted at 80µM in ribonuclease (RNase)-free water provided by the manufacturer Immediately following the anterior keratectomy, the moesin siRNA solution (80µM/10ul) was applied on the mouse cornea according to the iontophoresis protocol as described by Hao J et al (2009) Briefly, the cathodal (cathode electrode in the delivering chamber) iontophoresis was performed with a 0.5 mA current for 1 min and 2 mA for 10 s Moesin siRNA was administered using the same parametersas the first dose every day till mice were sacrificed Control siRNA was administered using the same parameters

fluorescein-as for moesin siRNA

2.7 List of laboratory techniques

2.7.1 Preparation of tissue

Preparation of tissue for staining (frozen sections)

Details of tissue preparation have been described (Zhu HY et al 2010) Briefly, the whole mouse eyeball were cut and embedded in a mixture of OCT and frozen

in liquid nitrogen Frozen sections (10 µm thick) were used for haematoxylin and eosin (HE) staining and immuno-fluorescent staining

2.7.2 Use of exogenous protein

TGF-β1

Constituted porcine TGF-β1 (101-B1, R and D systems) at 1µg/mL in sterile 4

mM HCl was used for topical application to the mouse cornea For long term storage as a dilute solution, a carrier protein (1% bovine serum albumin) was added to the vial and stored at -20° C; however, constituted batch of TGF-β1 was used for only two weeks at a time

2.7.3 Histological Evaluation

The intact eye globes were embedded in OCT Histological sections of these tissues were stained with hematoxylin and eosin

Steps:

1 Air dry the sections, and then rehydrate in 95% ethanol for 5 minutes

2 Wash the slides in distilled water for 2 minutes

3 Stain the sections in hematoxylin for 2-3 minutes (staining time can be modified depending on desired staining density)

4 Wash the slides in running tap water for 2 minutes

5 Identify the nucleus staining with Scott’s tap water for 2 minutes

6 Wash the slides in tap water for 2 minutes

7 Counterstain in eosin for 3 minutes (staining time can be modified depending on desired staining intensity

8 Rinse in 95% ethanol for 3-5 dips, then dehydrate the sections in 95% ethanol for 2 minutes

9 Follow by rinsing in 100% ethanol for 3-5 dips, then transfer the slides in 100% ethanol for 2 minutes

10 Clear in two changes of xylene, 2 minutes each

11 Mount with xylene based mounting medium

2.7.4 Immunochemistry

Immunofluorescent Staining

Immunofluorescent staining was performed as described previously (Zhu HY et al 2010) Briefly, the intact globe was embedded in OCT compound Prepared tissue blocks were sectioned at 10µm

Fixation was performed with 4% paraformaldehyde (Sigma, Singapore) for 20 minutes, followed by washing with 0.2% Triton X-100 (Sigma, Singapore) in 1x phosphate-buffered saline (PBS, 1st Base, Singapore) for 3 times, 5 minutes per time at room temperature

Blocking was performed by addition of 5% normal goat serum (Invitrogen, Singapore) in 1x PBS for at least 30 minutes at room temperature

Primary antibodies (appendix C) were then applied at the specified dilution in 5% goat serum and the section was incubated overnight at 4 °C

After washing with 1x PBS, the appropriate AlexaFluor 488 or 594 anti-mouse, anti-rat and anti-rabbit secondary antibodies (1:500, Invitrogen, Carlsbad, California) were applied respectively in 5% goat serum for 1 hour in a dark incubation chamber After washing with 1xPBS, UltraCruz Mounting Medium containing DAPI (4, 6-diamidino-2-phenylindole, Santa Cruz Biotechnology, CA) was applied

2.7.5 Imaging

Fluorescence microscopy (direct)

The fluorescence microscope (Zeiss, Oberkochen, Germany) was used

2.7.6 RNA Extraction and quantification

RNA was extracted using TRIZOL® method UV spectrophotometry with the Spectrophotometer (Beckman Coulter Inc, Fullerton, CA) was used to measure the quantification of RNA

Steps:

1 HOMOGENIZATION

The tissues wrapped in the aluminium foil are snap frozen in a liquid nitrogen, the homogenization of the tissue was done by mortar and pestle Tissue is solubilized Using TRIZOL® Reagent (15596-018, Invitrogen-Gibco, Grand Island, NY, USA)

to dissolve the tissue powder and transfer the solubilized contents into the Eppendorf” 1.5 ml capped polypropylene tubes (Treff# 96.7246.9.01) Homogenize tissue samples in 1 ml of TRIZOL Reagent., Chloroform addition and phase separation followed homogenization

4 RNA WASH

The supernatant was removed and the RNA pellet was washed twice with 80% ethanol, adding 1 ml of 80% ethanol per 1 ml of TRIZOL The sample was mixed

by vortexing and centrifuing at no more than 12,000 × g for 25 minutes at 4°C

5 REDISSOLVING THE RNA

RNA pellet was briefly air-dried for 5 minutes Incubating RNA pellet for 10 minutes at 60°C RNA was stored at -80°C

2.7.7 Reverse transcription and Polymerase chain reaction

Steps:

1

50ng random primers 1µL

10 mM dNTP 1 µL

Total RNA appropriate volume for 1 µg

Water (12 µL - volume of total RNA)

2 Heat mixture at 65 °C for 5 min and quick chill on ice for 5 min Briefly centrifuge and add the following master mix:

5X first strand buffer 4 µL

0.1M DTT 2 µL

RNase Out 1 µL

3 Mix contents gently Incubate contents at 25 °C for 2 minutes

4 Add 1 µL superscript III Reverse transcriptase (Invitrogen Life Tech, Carlsbad, CA) and mix by pipetting up and down

5 Incubate contents at 25 °C for 10 minutes

6 Incubate contents at 42 °C for 50 minutes

7 Inactive the reaction at 70 °C for 15 minutes

2.7.8 Real time RT-PCR

The real-time RT-PCR was performed to measure the relative quantification of mRNA levels The web-based software (www.universalprobelibrary.com) was used to design the primers of specific genes First Base Pte Ltd (Singapore) synthesized the primers Real time PCR was performed in conjunction with Mouse Universal Probe Library set (#04 683 641 001, Roche Applied Science, Mannheim, Germany)

Each reaction (a total volume of 10 µL, triplicates) within a single well of well plate was used The single reaction mix consisted of

384-
1 µL of forward primer

1 µL of reverse primer (primer stock was diluted 1:10)

0.1 µL of the probe from the library

5 µL of a 2X Master Mix (#04 707 494 001, Roche Applied Science)

0.5 µL of cDNA

2.6 µL of PCR-grade water

Plate was centrifuged at 2000 g for 3 minutes at 4 °C

The Lightcycler 480 System (Roche Diagnostics, Basel, Switzerland) and the Lightcycler 480 software release 1.2.0.0625 (Idaho Technology, Inc) was used to perform the real time PCR

Setting of cycling conditions:

Preincubation: 95°C at 10 minutes (ramp rate 4.8 °C/sec)

Amplication (45 cycles): in each cycle, denaturation at 95°C for 10 s with ramp rate of 4.8 °C/sec, annealing at 56°C for 10 s with ramp rate of 2.5 °C/sec and extension at 72°C for 30 s with ramp rate of 4.8 °C/sec

Incubation: 40°C with ramp rate 2.5 °C/sec

DNA contamination was detected by the non-template control The endogenous reference was GAPDH In order to detect specific gene, the primers in conjunction with probe were listed in appendix D: PCR primers

2.7.9 Total protein extraction

Freeze tissue (corneal stroma after removing epithelium and endothelium) in liquid nitrogen after excision and store at -80 °C until needed Powder frozen tissue with mortar and pestle Using RIPA lysis buffer (Santa Cruz, Singapore) with protease inhibitor and phosphotases inhibitor (Pierce Biotechnology, Rockford, IL, USA) to dissolve the tissue powder and transfer the solubilized contents into the Eppendorf” 1.5 ml capped polypropylene tubes (Treff# 96.7246.9.01) Samples were sonicated Put sample on ice for 45 minutes and followed by centrifugation at 16,100xg (Eppendorf Centrifuge 5415 D or equivalent) for 10 minutes at 4 degree C Decant the supernatant to a fresh Eppendorf tube The samples are stored at -80oC until needed

2.7.10 BCA protein assay

Protein solutions were diluted 2X in millipore water Albumin standards provided with the BCA kit were diluted with millipore water into concentrations from 2 mg/ml to 0.25 mg/ml After addition of 1 µL of diluted standard or protein solution of unknown concentration into 200 µL of BCA reagent, the reaction was incubated for 30 minutes at 37 degrees C in the dark The absorbance at 570 nm was read

2.7 11 Western blot analysis

For Western blot analysis, protein was obtained from three independent experiments The corneal stroma was dissected and harvested in RIPA buffer, and the lysates analyzed by Western blot Nitrocellulose membrane was from Millipore, Billerica, MA, USA The gel electrophoresis buffer contains Tris/glycine buffer (Bio-Rad, Hercules, CA, USA) and 20% methanol TBS was purchased from Bio-Rad, Hercules, CA, USA, Equal loading of total protein was verified by immunoblot for Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Santa Cruz Biotechnology, CA, USA)

Gel electrophoresis

The same amount of protein (30 µg) was loaded per gel lane for electrophoresis Protein was mixed with beta-mercaptoethanol and boiled for 5 minutes at 100 degree C before loading Protein standards were loaded in one lane (precision,

Singapore) Electrophoresis was performed at a voltage of 100 volts for stacking gel and 150 volts for separating gel

Transfer

Electrophorectic transfer to nitrocellulose paper was performed with 120 V over

90 minutes Blots were washed in 1xTBST Blocking was performed for one hour with 5% skimmed milk in 1xTBST

Blotting and Visualisation

Primary antibody incubation was overnight at 4 degree C After washing, the appropriate secondary antibody (1:2000 for anti-rabbit antibody sc-2030, 1:2000 for anti-mouse antibody sc-2005, Santa Cruz, CA) was applied for one hour Blots were subsequently reprobed with GAPDH (Biotechnology, Santa Cruz, CA, USA)

to confirm equal loading The membrane was developed with SuperSignal West Femto or Dual chemiluminescent substrates from Pierce Biotechnology (Rockford, IL) The x-ray films (Pierce Biotechnology) were scanned and the band intensity was quantified by densitometry using Kodak Image Station 4000R (Carestream Molecular Imaging., New Haven, Connecticut, USA)