Alterations in Ocular Surface System During the Pathogenesis of Herpes Stromal Keratitis

Wayne State University

Wayne State University Dissertations

January 2019

Alterations In Ocular Surface System During The Pathogenesis Of Herpes Stromal Keratitis

Pushpa Durgesh Rao Wayne State University, [email protected]

Follow this and additional works at: https://digitalcommons.wayne.edu/oa_dissertations

Part of the Immunology and Infectious Disease Commons

Recommended Citation Rao, Pushpa Durgesh, "Alterations In Ocular Surface System During The Pathogenesis Of Herpes Stromal Keratitis" (2019). Wayne State University Dissertations. 2181. https://digitalcommons.wayne.edu/oa_dissertations/2181

This Open Access Dissertation is brought to you for free and open access by DigitalCommons@WayneState. It has been accepted for inclusion in Wayne State University Dissertations by an authorized administrator of DigitalCommons@WayneState.

The Contents of Chapter 1 including Figure 13-23 has been Originally published in The

Journal of Immunology. Pushpa Rao, Susmit Suvas. 2019. Development of Hypoxia and the Prevalence of Glycolytic Metabolism in Progressing Herpes Stromal Keratitis

Immunologists, Inc.

DEDICATION

To my God Father, Dr. M.B Vasudevachari

My parents, Mr. Durgesh Jadkal and Mrs. Anuradha Durgesh Rao

And My Dear Friends,

All this work was made possible through your blessings and best wishes.

ACKNOWLEDGMENTS

I extend my heartiest thanks to Dr. Susmit Suvas for being an excellent mentor. I appreciate Dr. Suvas to not only steer my research project in a precise direction but also to be my cheerleader. Dr. Suvas has provided state-of-the-art training in his laboratory and I am proud to have published quality peer-reviewed journals in the

Dr.Suvas laboratory. My special thanks to the research grant received by Dr. Suvas from the National Eye Institute for supporting the research work throughout my Ph.D. career. I am also grateful to Thomas. C. Rumble graduate fellowship awarded by the

Wayne State University graduate school for financial support during my graduate education.

I am thankful to my dissertation committee members, Dr. Jena Steinle, Dr. Lalit

Pukhrambam and, Dr. Izabela Podgorski for their valuable time and to provide me with thoughtful inputs in the preparation of my thesis. I immensely appreciate my committee members for their guidance all through my Ph.D. career.

My special thanks to Dr. Linda Hazlett, the Vice Dean of Research and Graduate

Programs, Wayne State University School of Medicine to incessantly care about the upliftment of graduate students and to promote research by providing advanced facilities at the Vision core, Dept of Ophthalmology Visual and Anatomical Sciences.

I extend my gratitude to the Association of Research in Vision and

Ophthalmology (ARVO) foundation Travel Award (2017), the Gordon Research

Conference (GRC) Travel Award (2018) and, the Wayne State University Travel Award

(2015-2018). This award has provided me with a great platform to present my research work and to meet with the pioneers in the field.

iii

My sincere thanks to the members of Dr. Suvas laboratory, Subhash Gaddipati,

Ph.D., Bala Burugula, Ph.D., the former post-doctoral fellows who have helped me with the experiments. Andrew Jerome, Ph.D., as a fellow student has helped me in managing accessory laboratory work and for a good chat. Nicholas Ciavattone, the graduate student, has helped me acclimatize to a new laboratory environment when I had just joined the Dr.Suvas laboratory.

This research work would have been an ordeal without the unconditional love and emotional support provided by my parents. My mom, Mrs. Anuradha has always kept me motivated and strong to face challenging events in life. My dad, Mr. Durgesh feels proud of my achievements and my research work in the laboratory. My parent's happiness and blessings have been a driving force for all my accomplishments. My heartiest thanks to Dr. M.B. Vasudevachari, Late Dr. Vasanta Srikantan and, Ms. Prema

Shenoy to wish me success.

Last, but not least, I cannot forget to mention my special thanks to all my dear friends for their enormous help in financially supporting me when required during the hard times in life. Without their financial and emotional support, it would have been impossible to pursue my career and accomplish my dream.

TABLE OF CONTENTS

DEDICATION ...... ii

ACKNOWLEDGMENTS ...... iii

LIST OF FIGURES ………………………………………………………………………….vii

LIST OF TABLES………………………………………………………………………………..x

INTRODUCTION ...... 1

1. Herpes Simplex Virus 1 ...... 1

1.1 Epidemiology ...... 1

1.2. HSV-1 Structure and Lytic Life Cycle ...... 2

1.3. Pathogenesis of Herpes Stromal Keratitis ...... 3

2. Cornea ...... 5

2.1. Structure of normal cornea and tear film ...... 5

2.2. Corneal nerves ...... 9

2.3. Corneal hem-angiogenesis ...... 13

3. Hypoxia Inducible factor ...... 14

3.1. HIF Pathway ...... 14

4. Lacrimal gland ...... 16

4.1. Anatomy and functional role of lacrimal gland ...... 16

5. Conjunctiva...... 19

5.1. Anatomy and functional role of conjunctiva ...... 19

6. Insulin like growth factor binding protein (IGFBPs) ...... 20

6.1. Insulin growth factor (IGF) and IGFBP-3 ...... 20

6.2. IGF dependent action of IGFBPs ...... 22

6.3. IGF independent action of IGFBPs ...... 24

MATERIALS AND METHODS ...... 33

CHAPTER 1: DEVELOPMENT OF INFLAMMATORY HYPOXIA AND PREVALENCE OF GLYCOLYTIC METABOLISM IN PROGRESSING HERPES STROMAL KERATITIS LESIONS ...... 56

Abstract ...... 56

Introduction ...... 57

Results ...... 58

Discussion ...... 65

CHAPTER 2: EVALUATION OF EXTRAORBITAL LACRIMAL GLAND AND CONJUNCTIVAL INFLAMMATION DURING HSV-1 INFECTION……………………....82

Abstract ...... 82

Introduction ...... 83

Results ...... 86

Discussion ...... 94

CHAPTER 3: MORPHOLOGICAL ALTERATIONS IN THE CORNEAL EPITHELIUM DURING OCULAR HSV-1 INFECTION ...... 105

Abstract ...... 106

Introduction ...... 107

Results ...... 109

Discussion ...... 113

CHAPTER 4: IGF-1 DEPENDENT MECHANISM OF IGFBP-3 IN THE PATHOGENESIS OF HERPES STROMAL KERATITIS ...... 124

Abstract ...... 124

Introduction ...... 125

Results ...... 127

Discussion ...... 130

REFERENCES ...... 138

ABSTRACT ...... 172

AUTOBIOGRAPHICAL STATEMENT ...... 176 vi

LIST OF FIGURES

Figure 1: Key Aspects of Herpes Stromal Keratitis Development in C57BL/6 mouse model ...... 26

Figure 2: Structure of Cornea ...... 27

Figure 3: Layers of Tear Film...... 28

Figure 4: Regulation of the Hypoxia-Inducible Factor Pathway in an Oxygen-Dependent Manner...... 29

Figure 5: Histology of Lacrimal Gland from C57BL/6 mouse ...... 30

Figure 6: Histology of Conjunctiva from a C57BL/6 mouse...... 31

Figure 7: Insulin-Like Growth Factor (IGF) Dependent and Independent Action of Insulin-Like Growth Factor Binding Proteins (IGFBPs)...... 32

Figure 8: Protein Array Blot Showing No Insulin-Like Growth Factor Binding Protein-3 (IGFBP-3) Expression to Confirm the Knock-out Phenotype in IGFBP-3-/- mice...... 49

Figure 9: Phenol Red Thread Test in C57BL/6 mouse eyes...... 50

Figure 10: Dissection of Clean Cornea from a C57BL/6 mouse...... 51

Figure 11: Dissection of Extra-orbital Lacrimal Gland in C57BL/6 mouse ...... 52

Figure 12: Dissection of Adnexa from a C57BL/6 mouse...... 53

Figure 13: Development of Inflammatory Hypoxia Correlates with the Influx of Neutrophils in HSV-1 infected Corneas...... 70

Figure 14: Significant Depletion of Neutrophils Correlated with Reduced Development of Hypoxia in HSV-1 infected Corneas...... 71

Figure 15: Hypoxia Signaling Pathway RT2 Profiler PCR Array to show Elevated Expression of Glycolysis Regulating Genes in the Corneas with HSK Lesions...... 73

Figure 16: Quantitative Real-Time Polymerase Chain Reaction to show Elevated Expression of Glycolytic Genes in the Corneas...... 74

Figure 17: Prevalence of Glycolytic Metabolism in the HSV-1 infected Corneas ...... 75

Figure 18: Nuclear Localization of Hypoxia-Inducible Factor-2α in the Corneal Epithelium of HSV-1 infected Corneas During the Clinical Disease Period ...... 76

vii

Figure 19: Stabilization of Hypoxia-Inducible Factor-1α and it's Differential Expression on the Immune Cell Subsets in HSV-1 infected Corneas...... 77 Figure 20: Acriflavine Treatment Aggravates Corneal Opacity but Significantly Reduces Hem-angiogenesis in the Corneas with Herpes Stromal Keratitis...... 78

Figure 21: Acriflavine Treatment Significantly Reduces Stabilization of Hypoxia- Inducible Factor-1α and, -2α in the Immune Cell Subsets and Non-Immune Cells respectively in HSV-1 Infected Corneas ...... 79

Figure 22: Acriflavine Treatment Significantly Reduces the Influx of Non-Granulocytic Myeloid Cells and CD4 T cells in HSV-1 infected Corneas...... 80

Figure 23: A Proposed Model to Demonstrate the Development of Inflammatory Hypoxia in the HSV-1 infected Cornea During the Clinical Disease Period ...... 81

Figure 24: Deinnervation in HSV-1 Infected Corneas ...... 99

Figure 25: Significant Reduction of Tear Volume Correlating with Inflammation in Extra- orbital Lacrimal Gland in HSV-1 infected Corneas During the Clinical Disease Period...... 100

Figure 26: Detection of Infectious Virus in the Extra-orbital Lacrimal gland in the HSV-1 infected C57BL/6 mice During the Pre-Clinical Disease Period ...... 102

Figure 27: Topical Administration of Lacritin Transiently Increases the Tear Volume, and Partially the Severity of Herpes Stromal Keratitis in the HSV-1 infected Corneas...... 103

Figure 28: Loss of Goblet Cells in the Conjunctiva of HSV-1 infected eyes During the Clinical Disease Period...... 104

Figure 29: Increased Infiltration of Leukocytes in the Conjunctiva of HSV-1 Infected eyes during the Clinical Disease Period...... 105

Figure 30: Irregular Corneal Smoothness and Damage to the Ocular Surface in the HSV-1 infected eyes...... 117

Figure 31: Hyperproliferation of the Corneal Epithelium During the Clinical Onset of Herpes Stromal Keratitis ...... 118

Figure 32: Loss of K12 Expression the Corneal Epithelium During the Clinical Disease of Herpes Stromal Keratitis...... 119

Figure 33: Loss of K12 and the Gain of K10 Expression in the Corneal Epithelium During the Clinical Disease of Herpes Stromal Keratitis...... 120

Figure 34: Loss of Pax-6 Expression in the Corneal Epithelium During Clinical Disease of Herpes Stromal Keratitis ...... 121

viii

Figure 35: A Schematic Representation to Depict Alterations of Cytokeratin Expression in the Corneal Epithelium During the Clinical Disease of Herpes Stromal Keratitis. .... 122

Figure 36: A Proposed Model to Summarize the Events Involved in Ocular Surface Alterations in HSV-1 Infected Corneas During the Clinical Disease Period...... 123

Figure 37: Increase in IGFBP-3 Expression in C57BL/6 Mouse Corneas at Different Time Points Post-Ocular HSV-1 Infection...... 133

Figure 38: Intense IGF-1R Immunofluorescence Staining in HSV-1 Infected C57BL/6 Mouse Corneas During the Clinical Disease Period ...... 134

Figure 39: Significant Increase in the IGF-1R Phosphorylation Detected in Total Leukocytes in Herpes Stromal Keratitis Developing Corneas of IGFBP-3-/- in Comparison to C57BL/6 Mice ...... 135

Figure 40: Robust Development of HSK Disease in HSV-1 infected IGFBP-3-/- Mice in comparison to the C57BL/6 Mice ...... 136

Figure 41: Schematic Representation to Summarize the Aims Investigated in this Study in HSV-1 Infected Mouse Corneas…………………………………………………………137

LIST OF TABLES

Table 1: List of antibodies used for Immunofluorescence and Flow Cytometry……….54

Table 2: List of Qiagen Primers used for qRT-PCR assay………………………………55

INTRODUCTION

1. Herpes Simplex Virus 1

1.1 Epidemiology

Herpes Simplex Virus Type 1 (HSV-1) is a common existing virus that is known to cause infection in the majority of the world’s population [1,2,3]. HSV-1 can establish latency with the potential to cause recurrent disease in the host for a lifetime [4]. In the

United States, 50% of the total population has been estimated to be seropositive for

HSV-1, while in Asia and Europe reported a higher incidence of HSV-1 infection, as high as 80% [5,6,7]. The reports of HSV-1 infection across the world explicitly demonstrates its ubiquitous prevalence. Approximately, 20,000 new cases appear annually with an increased risk in children and aged individuals (45-49yr old) [5,8].

HSV-1 primarily infects the orofacial region causing skin lesions (cold sores)

[9,2]. During latency, HSV-1 harbors in the trigeminal ganglion and acquires a quiescent latency mode in the absence of infectious virions [10]. Reactivation of HSV-1 can lead to ocular infections causing blindness [11,12]. Herpes Stromal Keratitis (HSK) is the leading cause of corneal blindness in the United States [11]. Corneal epithelial defect and stromal inflammation cause corneal blindness, that obscures the clear vision during the pathogenesis of HSK [13]. Contemporary therapeutics such as anti-viral (acyclovir) and anti-inflammatory (corticosteroid) drugs have been shown to control the disease transiently but cannot subdue the HSK disease severity. Therefore, an urgent need for a promising cure is crucial to treat HSK causing corneal blindness.

1.2. HSV-1 Structure and Lytic Life Cycle

HSV-1 is a double-stranded DNA virus encapsulated within an icosahedral nucleocapsid consisting of 162 capsomeres [14,15]. The HSV-1 viral genome encodes

84 proteins that play a major role in the lytic life cycle of the virus. The first step of the lytic cycle is the entry of the HSV-1 virus into the host cell which is initiated by the attachment and fusion of the virus to the host cell surface proteins (heparan sulfates).

There are several glycoproteins namely, gB, gC, gD, gH, and gL that are involved in its interaction with the host cell surface receptors [16]. Nectin-1, herpes virus entry mediator (HVEM), and paired immunoglobulin-like type 2 receptor-alpha (PILR-α) receptor on the corneal epithelium has been reported to be the key receptors which help in the HSV-1 viral entry [17,18]. Furthermore, upon the entry of the virus into the host cell, the transcription machinery for viral DNA replication and assembly of new viral capsids will occur following the insertion of the viral genome into the host nucleus [19].

Consequently, the host cell machinery will aid in the expression of immediate early (IE), early (E), and the late (L) proteins required for the synthesis of new virions [20,21].

Finally, a newly synthesized DNA will be packaged into a nucleocapsid to form a complete virion [22].

An interesting aspect of the HSV-1 life cycle is its establishment of latency in the trigeminal ganglion. Once the active viral replication subsides, the viral genome remains in the infected host cell, thereby maintaining the virus in the latent phase [23,24]. The expression of latency associated transcripts promotes the HSV-1 latency [25]. There are ongoing studies to understand further the factors associated with the establishment and maintenance of HSV-1 latency in the trigeminal ganglion. However, reports have

3 suggested that stress can induce a shift from latent to the lytic cycle (active replication) of the HSV-1 virus which can lead to catastrophic conditions in the host. Notably, In the mouse model of HSK, the disease is not induced by the virus reactivation, since the cornea is the primary site of infection that favors being a good model to study the pathology of the disease similar as in humans.

1.3. Pathogenesis of Herpes Stromal Keratitis

1.3.1. Pre-clinical Phase

Development of HSK in the mouse models is very similar to the events that take place in the human eyes. However, in mouse model primary HSV-1 infection is conducted by directly infecting the cornea. While in human eyes the majority of HSV-1 infection occurs via reactivation of the virus, termed as the recurrent infection.

Pathogenesis of HSK can be broadly categorized into the pre-clinical and the clinical phase (Figure 1). The preclinical phase lasts through 7 days post-infection, after which the clinical phase of HSK will commence which is responsible for the development of corneal pathology (Figure 1).

In the pre-clinical phase of the HSK disease, the active replication of the virus is controlled by the innate immune response [26,27]. This response is induced by the influx of early innate immune cells such as neutrophils, macrophages and dendritic cells for the initial attempt to clear the virus [28]. The innate immune cells can secrete mediators such as tumor necrosis factor-α (TNF-α), interferon-γ (IFN-γ), nitric oxide and other reactive oxygen species that can aid in virus clearance but, can subsequently damage the corneal tissue causing detrimental effects [29]. A plethora of pro- inflammatory factors is released by the innate immune cells such as Interleukins (IL-1,

IL-6, IL-8, IL-12) and macrophage inflammatory proteins (MIP-1α) [30]. Neutrophils are the short-lived cells that survive 2-3 days post HSV-1 infection in cornea [31].

Neutrophils play an essential role in viral clearance via phagocytosis of the virus and by the release reactive oxygen species (ROS). Meanwhile, the antigen presenting cells pick up these phagocytosed viral particles to elicit the T-cell responses. Consequently, this promotes the onset of the clinical phase in the HSK pathogenesis.

1.3.1. Clinical Phase

In mouse models of HSV-1 infection, the clinical phase will commence after 8-9 days post infection. During this phase, characteristic features of HSK pathogenesis will appear in the infected cornea, such as the development of hem-angiogenesis, and corneal opacity which leads to corneal damage and blindness. T-cells have been reported to be the major cell types responsible for the HSK disease severity.

Interestingly, studies have demonstrated that the lack of T-cells will not promote the development of HSK disease [32,33]. CD4+ T cells play a predominant role in secreting pro-inflammatory factors and chemotactic signals to help the extravasation of the polymorphonuclear cells (PMNs) [30,34].

Of note, the influx of neutrophils occurs during the clinical phase, causing a biphasic response during HSK disease progression [35,36]. Apart from the T-cells, the second wave of neutrophils influx is also responsible for the HSK disease severity.

Surprisingly, the infiltrated neutrophils in the second wave are resistant to apoptosis

[31]. Eventually, the existence of neutrophils for an extended period in the infected cornea will cause increased corneal tissue damage [37,38]. To moderate this immuno-

5 inflammatory condition, there are anti-inflammatory molecules and neuropeptides produced from the corneal sensory nerve stimulation) [39,40,41]

2. Cornea

2.1. Structure of normal cornea and tear film

The cornea is a unique avascular and a transparent tissue that not only protects the anterior ocular surface but also aids in clear vision. Normally the corneal surface maintains its moist and smooth refractive surface in the presence of trilaminar tear film with a thickness of about 4-6μm. It is composed of lipid, aqueous and mucin layers [42],

(Figure 3). Tear film also plays a role in supplying the cornea with nutrients and protecting the ocular surface from invading pathogens [43,44]. Tear film (Figure 3) consists of the outer lipid (oil) layer produced from meibomian glands (modified sebaceous glands) located in the tarsal plates [45]. Meibum consists primarily of phospholipids and maintains its fluidic property, due to its low melting temperature. It acts like a surfactant that spreads over the aqueous layer of tear film [46]. Hence, the absence of a meibomian layer results in evaporation of tear film [47]. The middle aqueous layer of tear film comprises 90% of the tear film and consists of water, electrolytes, and proteins [48]. It is produced mostly from lacrimal glands that contain soluble mucins which contribute to the viscosity of the tear film. The middle aqueous layer functions in not only providing lubrication, protecting from foreign materials but also in delivering nutrients such as oxygen, glucose, and proteins to the underlying corneal tissue [49].The innermost mucus layer is secreted by the conjunctival goblet cells, in addition to corneal and conjunctival epithelium [50]. This layer consists of ocular mucus composed of primarily the mucins that protect, lubricate and anchor aqueous

6 tear film to the underlying corneal epithelium. Mucins are high molecular weight glycoproteins located on the apical surface of the epithelium [51]. There are two classes of mucins, which includes a secretory and membrane-bound form [52]. The secretary form of mucin further consists of two forms, the large gel-forming mucin and the small soluble mucin [44,53,54]. MUC5AC is one of the major large gel-forming mucin expressed in conjunctival goblet cells. The secreted forms of mucins lack a transmembrane domain which is produced by the goblet cells [55]. In contrast, the membrane-bound mucins have a short cytoplasmic tail with a single transmembrane domain and a glycosylated extracellular domain. The corneal and conjunctival epithelial cells contain membrane-bound mucins namely, MUC1, MUC4, and MUC16 that functions to form glycocalyx at the epithelial cell-tear film interface [56]. The glycocalyx protects corneal epithelium from shear forces or desiccation [57]. The corneal and conjunctival epithelium is covered by microvilli, microplicae, and glycocalyx composed of glycoproteins and glycolipids [50,58,59]. The glycocalyx extends between microplicae and microvilli through branching and associates itself to the epithelial cell membrane and also assists in binding to the mucus layer [60]. Hence, the mucus layer binds to the glycocalyx and moves freely across the cornea without firmly attaching itself to the corneal epithelium [61]. Glycocalyx helps in even spreading of the mucus across the corneal epithelium. However, deficiency in tear film production or increased tear evaporation causes dry eye conditions leads to corneal epithelial damage [52].

Structurally cornea is comprised of five morphologically distinct layers (Fig 2) that includes (a) corneal epithelium (b) Bowman’s layer, (c) stroma, (d) Descemet’s membrane and the (e) endothelium. The three major functional layers include the

7 anterior corneal epithelium, middle tough collagenous stroma and the posterior monolayer of the endothelium [62].

Corneal epithelium (Figure 2) is 50-70μm thick and is derived from the surface ectoderm during embryogenesis [63]. Corneal epithelium consists of 5-6 layers of nucleated cells, which can be further subdivided into three layers, (i) surface cell layer

(ii) wing like cell layer (iii) Basal layer. The apical (surface) layer is a 2-3 layer with stratified squamous non keratinized epithelium that exhibits surface microvilli/microplicae that contacts with the ocular mucins [63,64]. It is also a protective layer, as it continuously shed cells. The middle wing-like the cell layer is 1-2 layers with polyhedral shaped cells. The deeper basal cell layer is the germinative layer of the corneal epithelium which consists of columnar cells. The corneal epithelium is devoid of goblet cells and is continuous with the conjunctival epithelium at the limbus [65,66].

Corneal epithelial cells are actively metabolizing cells with abundant mitochondria present in the wing like cells but scarce in the basal layer, while the wing like cells and the surface cells contain rich glycogen deposits [64] Junctional complexes, such as tight junction proteins (Zonula occludens-1, occludens, claudin-1) desmosomes, and intermediate filaments play a major role in maintaining corneal epithelial cell integrity and polarity [66]. Epithelial cells express intermediate filament proteins with their diverse expression depending on the location and differentiation pattern [67]. These filaments are formed by the pairing of type-1 (acidic) and type-II molecules (basic/neutral) in a 1:1 ratio. Cytokeratins function in maintaining epithelial integrity, mechanical stability and regulating intracellular pathways such as wound healing/apoptosis [67,68]. Stability of epithelial cells is maintained by cytokeratins through their interaction with the

8 desmosomes and the hemidesmosomes between the epithelial cells and to the basement membrane. Since the expression pattern of cytokeratin sheds light on the differentiation pattern of the epithelium, it is used as a potent epithelial differentiation marker, as well as a diagnostic tumor marker [67,69]. Corneal epithelial cells express

K3 (type II)/K12 (type I) pair in all layers of the epithelium, and the suprabasal cells of the limbus region are also positive to K12. In contrast, the basal corneal stem cells in the limbus region are both K3/K12 negative [70,71]. Apart from the corneal epithelium, the conjunctiva is also known to express K15 in all layers of its epithelium with K19 expression in its basal layer only [72].

A fragile detachable epithelium characterizes an absence of cytokeratin expression in the corneal epithelial cells due to its loss of mechanical stability [73,74].

Interesting findings from several reports show that during corneal insults the expression pattern of cytokeratin changes; for ex: 1) debrided corneal epithelium showed K19/K15 expression with the appearance of goblet cells, that signifies the alteration into conjunctival type of epithelium [72]; 2) Dry eye conditions showed corneal epithelium with K1/K10 expression, signifying its alteration to skin type of epithelium (non- keratinized to keratinized type). These conditions are correlated with the development of

‘squamous metaplasia’ in the corneal epithelium that is characterized with the keratinization, loss of mucin expression, increased stratification of cells and enlargement of the surface squamous cells [75].

Corneal stroma (Figure 2) is protected by the overlying dense collagenous

Bowman’s layer. Stroma is the largest and the thickest (400-500μm) layer of the cornea that consists of specialized collagen fibers- ‘lamellae.’ A defined organization of the

9 lamellae contributes to corneal transparency. The lamella is interspersed with keratocytes and ground substance (proteoglycans and chondroitin sulfate) that plays a role in maintaining the corneal matrix. During corneal wounding, keratocytes undergo differentiation to form fibroblast cell type with a capability to migrate to the wound site in the presence of contractile actin element and subsequently to myofibroblast phenotype of cells with an increased contractile ability due to the expression of α-Smooth Muscle

Actin (α-SMA). During pathological conditions, corneal epithelial cells, are known to secrete growth factors such as transforming growth factor (TGF-β1), epidermal growth factor (EGF), fibroblast growth factor (FGF), insulin-like growth factor (IGF-1), which binds to their cognate receptors to mediate signal transduction pathways that can alter the normal phenotype of the corneal stroma [64,76].

Corneal endothelium (Figure 2) consists of a monolayer of non-dividing cells with a limited regenerative capacity. It mainly functions as a pump between the anterior cornea and the posterior aqueous humor to deliver waste out and diffuse nutrients into the cornea [63].

2.2. Corneal nerves

The cornea is an avascular tissue; however, it is also a densely innervated tissue. Corneal nerves play an essential role in tear production and secretion.

Additionally, corneal nerves help in wound healing, the blink reflex, and sensations of temperature, pain, and touch [77]. It is well known from literature studies that corneal nerve dysfunction leads to corneal opacity and loss of vision [78]. Approximately 1.57 million people worldwide have blindness due to corneal opacity [78]. Corneal nerve damage may also occur under inflammatory conditions or in the presence of an

10 infectious corneal disease leading to blindness [79]. From a recent study, it has been demonstrated that HSV-1 infection in cornea leads to corneal nerve degeneration [80].

Hence, this makes it important to understand the structure and function of corneal nerves.

The ophthalmic branch of cranial nerve V (trigeminal nerve) branches into sensory trigeminal neurons to transmit nerve fibers in the cornea and terminate into free nerve fiber endings in the corneal epithelium [81]. The nerve fibers from the trigeminal ganglion cells branch to form nerve bundles near the limbus. These nerve bundles now form the limbal plexus [82]. From the limbal plexus, stromal nerve trunks emerge to enter the peripheral corneal stroma in a radial pattern and divides into several branches

[83]. There are both unmyelinated and myelinated axons in the stromal nerve trunks

[84]. In the periphery of the mouse cornea there are myelinated nerve fibers, whereas, in the periphery and central human cornea, beaded fibers and straight fibers are present that are not myelinated and myelinated respectively [82,85].

Interestingly, corneal nerve fibers lose their myelin sheath with their outer covering ‘perineurium’ only surrounded by Schwann cells that help in conducting nerve impulses in the absence of myelination [86]. In the corneal stroma, as the nerve density increases anteriorly, the diameter of the nerve also decreases making it thin and leading to its branching into smaller fascicles eventually forming the anterior stromal plexus

[81,87,88]. Corneal stromal nerves are known to innervate keratocytes and are organized in parallel to collagen lamellae, while only a few stromal nerves are known to terminate as free nerve endings [88]. Defasciculation of stromal nerves may occur as they reach the anteriormost of the stroma and into the subbasal nerve plexus reaching

11 into the corneal epithelium. Hence, the innervation in the corneal epithelium is derived from the subbasal nerve plexus, which originated from the branches of the peripheral stromal nerves [89,90]. In the peripheral cornea, the tips of the stromal nerve enter and gives rise to the long bundles of nerves that run into the central cornea, that eventually divides into intricate branches to form a network in the corneal epithelium. The fashion in which the nerves run into the cornea gives it a form called ‘swirl- like pattern’

[83,91,92]. The beaded unmyelinated nerve fibers run through the subbasal plexus and terminate into the corneal epithelium as free nerve endings, especially with more dense nerve endings in the central cornea [87]. The nerve fibers of corneal epithelium are nociceptive fibers that respond to heat, chemical or mechanical stimuli [93]. A response to neuropathic pain may also be associated with corneal inflammation [94]. Intriguingly, the existence of neurotrophic factors in the cornea is known to maintain a healthy corneal surface by modulating inflammation and inducing wound repair [95]. Some of the corneal neurotrophic factors include nerve growth factor (NGF), Semaphorin (Sema)

3A, 3F, and 7A, Slits 1-3, Netrin 2, Ephrin (Eph) B2, Substance P (SP), calcitonin-gene related peptide (CGRP), neurotrophin (NT-3) and brain-derived neurotrophic factor

(BDNF) [96,97]. Substance P is known to be released from the nerve fibers during inflammation to promote wound healing and proliferation of corneal epithelial cells [95].

Therefore, the absence of neurotrophic factors can significantly affect the corneal nerve function.

To study corneal nerve function, there are indirect tests performed such as; a) lacrimation tests, for ex: the Schirmer and phenol red thread tests that assess the tear production & quantification of tear volumes, b) Tests with vital dyes, for ex: fluorescein,

Rose Bengal and lissamine green to observe the occurrence of corneal epithelium damage [98]. Corneal diseases can be associated with the dysfunction of the corneal nerves since the corneal sensory nerves are connected to the central nervous system

(brain) via afferent nerves while the signals from the brain are received through efferent nerves [81]. Apart from the cornea, corneal sensory nerves, and brain, this circuit system also comprises the lacrimal glands which complete the neural circuit and has an essential role in tear production. Hence indirectly, during corneal damage, malfunction of the neural circuit system can cause aberrant tear production. The absence of neuropeptides involved in signaling the neural circuit can also cause a direct effect on aberrant tear production [99]. It is evident that corneal nerves play an important role in maintaining the corneal integrity since damage caused to corneal nerves leads to its structural damage for ex: epithelial defect, perforation and reduced transparency

[100,101]. It is well known from a recent mouse model study that during HSV-1 infection in the cornea, the damage is caused to the sensory nerve fibers innervating cornea that as observed by the corneal whole mount staining for β-III tubulin (neuronal marker) [80].

Interestingly, another factor known to involve in corneal nerve formation is the

Paired box-6 (Pax-6). Pax-6 is a transcription factor that plays a major role in eye development including the corneal epithelium. Deficiency in Pax-6 expression results into various ocular anomalies, especially leading to corneal opacity, corneal epithelial defects, vascularization and discrepancy in the regular corneal nerve pattern [102]. Pax-

6 has also been reported in maintaining normal corneal morphogenesis by controlling the cellular adhesion in corneal epithelial cells via the expression of tight junction proteins [103]. Therefore Pax-6 has been identified as a role model in keeping a normal

13 healthy corneal epithelium that also functions in maintaining the corneal nerve formation.

2.3. Corneal hem-angiogenesis

Corneal hem-angiogenesis involves the sprouting of new blood vessels from capillaries and venules of pericorneal plexus of the limbal area that receives blood supply from the branches of ophthalmic arteries and ciliary arteries [104]. The process of corneal angiogenesis can be divided into several phases as follows; 1) the release of growth factors under corneal injury and their binding to the cognate receptors present on vascular endothelial cells. These cells belong to the neighboring pre-existing blood vessels that help in promoting vessel growth. 2) Vascular endothelial cells undergo a process of ‘activation’ after migrating away from the parent blood vessel by producing proteolytic enzymes to degrade the basement membrane and extracellular matrix. 3)

Activated vascular endothelial cells acquire the expression of cell surface adhesion molecules such as integrins and selectins [105].

The cornea under normal conditions, lacks both lymphatic and blood vessels, thus maintaining its transparency and visual acuity [106]. An important characteristic feature of the normal cornea is to maintain lymphatic and avascularity that also influences the “immune privileged” status of the tissue. The privileged status of cornea actively plays a role in maintaining a proper balance between angiogenic and antiangiogenic factors in corneal epithelium [104]. Several angiogenic molecules have been identified to be expressed in the cornea, most prominently being Vascular endothelial growth factor (VEGF-A also called as VEGF) which is increased during inflammation and vascularization of the cornea in both human and animal models [104]. VEGF was

14 initially named as vascular permeability factor (VPF) for its role as a stimulator of vascular permeability. VEGF plays a crucial role in angiogenic events during embryonic, physiologic, and pathological blood vessel growth in vivo104. VEGFs belong to a family of secreted polypeptides with VEGF-A as a primary member of the family [104,107].

Additionally, VEGF-B, VEGF-C, and VEGF-D are the other members of the family that differentially bind to VEGF receptors (VEGFR). VEGF-A binds to VEGFR-2 (also called

Flk-1) and VEGFR-1 (also called Flt-1), VEGF-C and VEGF-D bind to VEGFR-3 while

VEGF-B only binds to VEGFR-1 [104]. Recent studies have shown the expression of

VEGF localized to corneal epithelial and endothelial cells, vascular endothelial cells of limbal vessels and keratocytes [104,105]. Normal cornea constitutively expresses

VEGFR-1 that binds to VEGF-A a potent angiogenic stimulator and VEGFR-3 specifically by corneal epithelium that binds to angiogenic VEGF-C and VEGF-D.

Additionally, there are also other critical anti-angiogenic molecules expressed by corneas such as thrombospondin-1 (TSP-1), endogenous IL-1 receptor antagonist (IL-

1Ra) and Tissue inhibitor of metalloproteinases (TIMPs)-1 and 2. Therefore, these factors help in maintaining corneal avascularity by inhibiting both hem-angiogenesis and lymphangiogenesis. However, under inflammatory conditions such as HSK, increased expression of VEGF has been reported, particularly by the epithelial cells of both inflammed and vascularized cornea [108].

3. Hypoxia Inducible factor

3.1.HIF Pathway

Identification of hematopoietic growth hormone EPO led to the discovery of HIF as a key regulator of hypoxia’ that can either directly or indirectly regulate gene

15 expression involved in controlling metabolic adaptations, erythropoiesis, angiogenesis, apoptosis and other cell survival processes [109]. HIF is a widely known transcription factor with three different alpha isoforms namely HIF-1α, HIF-2α, and HIF-3α all that are sensitive to oxygen tension; while the HIF-1β isoform is constitutively expressed. HIF-1β is not oxygen responsive and forms heterodimeric complexes with one of the alpha isoforms. Both HIF-1α and HIF-2α play a major role in upregulating angiogenic factors during hypoxia [110]. Under normal oxygen tension (normoxia) HIF undergoes a hydroxylation process. In contrast, under hypoxic conditions, HIF stabilizes in the nucleus forming a heterodimer with HIF-1β that triggers the activation of genes involved in both physiological and pathological conditions [111].

HIF-1β is constitutively expressed and forms a heterodimeric complex with one of the three isoforms of HIF namely, HIF-1α, HIF-2α, and HIF-3α which are upregulated during hypoxia [112]. Each of these alpha isoforms contains functional domains, with an individual specific role in dimerization (bHLH/PAS), nuclear localization signal (NLS), protein stability (ODDD-N-TAD), and transcriptional activity (N-TAD/C-TAD); bHLH – basic helix loop helix domain; PAS= Per-Arnt-Sim motif; NLS= nuclear localization signal; ODDD= oxygen-dependent degradation domain; N-TAD= N-terminal transactivation domain; C-TAD= C-terminal transactivation domain [113]. Prolyl hydroxylase enzymes (PHD1-3) play an important role in regulating the HIF pathway since it functions as the oxygen sensor in the cells [114]. During normoxia, PHD enzyme hydroxylates ODDD of HIF-α that initiates the binding of von Hippel-Lindau E3 ubiquitin ligase complex to ubiquitinate HIF-α and thereby promoting it to proteasomal degradation [115] (Figure 4). In contrast, under hypoxia, HIF-α accumulates in the cell

16 followed by its nuclear translocation to form a transcriptional complex with co-activator proteins p300 and CBP. The transcriptional complex consists of the dimers HIF-α- HIF-

1β that binds to the HRE (hypoxia response element) within the promoter of target genes. The transcribed genes mediate angiogenesis, cell motility, metabolism, apoptosis, proliferation, and other cellular functions in response to hypoxia [116]. In a nutshell, under normoxia, HIF-α degradation is regulated via hydroxylation mediated by the VHL binding, while under hypoxia HIF-α accumulation in the nucleus and its transcriptional activity is mediated by the binding of p300/CBP to HIF-α [117].

4. Lacrimal gland

4.1. Anatomy and functional role of the lacrimal gland

The lacrimal functional unit (LFU) consists of the lacrimal gland, conjunctiva, cornea and interconnecting innervations [118]. The aqueous layer of the tear film is contributed by the lacrimal gland that is essential to maintain a healthy ocular surface, especially the corneal and conjunctival tissue [99]. The prominent lacrimal gland in mice and rats is the exorbital lacrimal gland located below the ear which is connected to the ocular surface by the excretory duct [119]. The main secretory cells in lacrimal gland are the pyramid-shaped acinar cells that are linked together by tight junctions to form the racemous glands [99] (Figure 5). The intralobular and interlobular lacrimal ducts consist of duct cells that help in the modification of the primary products of the lacrimal gland with water and electrolytes from acinar cell secretions [120]. Another main cell type in the lacrimal gland is the myoepithelial cell that expresses α smooth muscle actin. These myoepithelial cells surround the acinar and ductal cells and functions via its contraction to help excretion of the lacrimal gland fluid [99,120]. Apart from these cell types, the

17 existence of immune cells has also been studied, including lymphocytes, dendritic cells, macrophages, and plasma cells. Under pathological conditions, lymphocyte populations are reported to increase their population [121]. Intriguingly, the secretion from a lacrimal gland (tubuloacinar exocrine gland) is tightly regulated by the neural reflex. The mechanism of neural response occurs via the activation of afferent sensory nerves in the cornea and conjunctiva that stimulates the efferent nerves (parasympathetic and sympathetic nerves) innervating the lacrimal gland. The efferent nerve stimulation, in turn, helps in the release of the neurotransmitters from the nerves innervating lacrimal glands and promotes secretion from the lacrimal glands onto the ocular surface. The thin myelinated sensory nerve endings in the ocular surface help in stimulating the lacrimal gland to secrete the aqueous layer of the tear film [122,123]. These nerves lose their myelin sheath as they enter the corneal stroma, where they undergo intricate branching into the subepithelial plexus. Eventually, these nerves branch out further as thin nerves into the corneal epithelium that ultimately terminates in the apical layers of the corneal epithelium forming a ‘leash type of nerve endings’ [40]. The sensory nerves that are present very close to the corneal surface will help in sensing changes in the tear film due to stress or environmental changes [81]. Therefore, the regulation of lacrimal gland secretion is controlled by the ‘neural reflex arc’ that involves the activation of sensory nerves in both cornea and conjunctiva [124]. There are three major steps involved in the neural regulation of lacrimal gland function; (i) sensory nerve activation in the cornea and conjunctiva: corneal sensitivity is the prime factor for tear secretion since sensory nerves play a major role in stimulating tear secretion from the lacrimal gland. Under conditions such as corneal lesions, microbial keratitis, or epithelial

18 defects, reduced corneal sensitivity has been reported to be associated with decreased tear production [125]. (ii) Efferent parasympathetic and sympathetic nerve stimulation: neurotransmitters such as acetylcholine and norepinephrine are released upon both parasympathetic and sympathetic nerve stimulation respectively, to control secretion from the lacrimal gland. Acetylcholine binds to muscarinic receptors, whereas norepinephrine binds to an adrenergic receptor to further initiate the signaling pathways

[126]. Despite the presence of efficient afferent neural stimulation, neurotransmitter release is critical in lacrimal gland secretion. During lacrimal inflammation, a decrease in tear production is observed due to the blockage of neurotransmitter release and disabled efferent nerve stimulation [127]. (iii) Signaling pathways in lacrimal gland: in addition to the neurotransmitter release, signaling pathways activated by its binding to cognate receptors is also essential in stimulating lacrimal gland secretion. The inhibition of the signaling pathway was observed when antagonists were bound to muscarinic receptors and thereby prevents the binding of acetylcholine leading to inefficient efferent nerve stimulation [121].

In summary, the neural reflex arc is known to involve in ‘cross-talk’ between ocular surface and lacrimal gland by activating the afferent sensory nerves in the cornea and conjunctiva that promotes the activation of efferent sympathetic and parasympathetic nerves. This cascade of events is crucial to signal tear secretion from lacrimal gland [124].

5. Conjunctiva

5.1. Anatomy and functional role of the conjunctiva

Conjunctiva is a semitransparent thin mucous membrane that covers the posterior surface of the eyelids, connects with the eyeball and eventually extends into the limbus of the cornea. It is highly innervated and vascularized tissue that functions in protecting the cornea by secreting mucins, antibacterial proteins, water, and electrolytes which forms both an aqueous and mucous layer of the tear film [128]. Importantly, conjunctiva plays an essential role in absorbing drugs (permeable to 40KDa) and actively transports via sodium-dependent, monocarboxylate or dipeptide transporters to the ocular surface [129,130]. Conjunctiva is anatomically subdivided into three subdivisions; (1) Tarsal or palpebral conjunctiva lining the eyelids (2) Forniceal conjunctiva lining the upper and lower fornices (3) Bulbar conjunctiva overlying the sclera. Conjunctiva consists of the non-keratinizing epithelium with mucin-secreting goblet cells and eventually becomes continuous with the corneal epithelium [131]

(Figure 6). From a plethora of studies, it is evident that under corneal epithelial cell damage, limbal stem cells will help in corneal re-epithelization. Interestingly, during damage to the limbus, development of corneal angiogenesis and migration of goblet cells in the newly formed corneal epithelium was observed [132]. Rich innervations in conjunctiva occur via parasympathetic and sympathetic nerves surrounding conjunctival goblet cells, whereas the sensory nerves innervate the conjunctival epithelial cells

[133,134]. Therefore, corneal sensory nerves will help to promote neural reflex to stimulate parasympathetic nerves and mucin secretion from conjunctiva [135].

Conjunctival epithelial cells synthesize membrane-bound mucins MUC- 1, -2 -4, -7 and -

16. MUC-4 is also reported to be secreted as soluble mucin by releasing itself into the tear film via ‘ectodomain shedding’[128]. Gel-forming mucin, MUC5AC is synthesized and secreted by the conjunctival goblet cells [136].

Altogether, this study broadens our knowledge on understanding first, an intricate network of lacrimal gland inflammation during HSV-1 infection in mouse cornea and its influence on tear secretion, correlating to the maintenance of corneal epithelial integrity and second, the mechanisms of hypoxia-induced angiogenic mediators in HSV-1 infected mouse cornea. More importantly, this study will also help us to identify novel strategies for maintaining corneal integrity, to increase and sustain stable tear volume to alleviate HSK disease severity and thereby control corneal blindness.

6. Insulin-like growth factor binding protein (IGFBPs)

6.1. Insulin growth factor (IGF) and IGFBP-3

Cell survival depends on a balance maintained between the signals that mediate cell survival and apoptosis. The IGF axis has a unique role to play in managing both signals. IGF has a prominent role in mediating cell survival factors by binding to IGF-1R

[137]. The IGF system constitutes peptides called Insulin-like growth factors (IGFs), which are produced in the liver and possess mitogenic activity by interacting with

Insulin-like growth factor receptors (IGFRs). IGFs are of two types that are structurally similar to insulin, namely IGF-I and IGF-II that are regulated by IGFBPs and IGFBP proteases [138]. It will be interesting to understand the role of IGFBPs in the regulation of IGF during corneal HSV-1 infection if any. A 150KDa ternary complex is formed when

IGFBP interacts with the acid -labile subunit (ALS) glycoprotein and the free IGF in serum. This interaction of IGFBPs and IGF inhibits the binding of IGF to its cognate IGF

21 receptor [139]. From the sequencing studies, six different isoforms of IGFBPs have been identified; IGFBP-1 to IGFBP-6. Among these IGFBP-3 has been reported to be highly abundant in the serum that can predominantly bind to the free IGF-I/IGF-II in the serum [140,141]. It has been shown that 1% of the circulating IGF-1 remains in the free state, while the rest is bound mainly to IGFBP-3 [142]. IGF-1 action can be modulated by IGFBP-3 either positively or negatively depending on the tissue, and it’s the pathologic state [143]. In tumor models, IGFBP-3 has been widely studied to promote different functions, either apoptosis or support cell survival signals [144]. The binding of

IGFBP-3 to IGF-1 is inhibitory, since it reduces the bioavailability of IGF-1 to promote mitogenic effects but leads to induce apoptosis. On the contrary, IGFBP-3 binding to

IGF-1 will control IGF-II function positively by slowly releasing the IGF-1 to its binding receptor IGF-1R. Therefore, IGFBP-3 will help in coordinating the exposure of IGF-1 to

IGF-1R in a well-regulated manner to promote downstream cell survival signaling [137].

The IGF-1R is a transmembrane heterotetrameric protein, which has two extracellular α- chains and two membrane-spanning β-chains in a disulfide-linked β-α-α-β configuration

[145,146].

The affinity of IGFBP-3 to IGF-1 reduces when IGFBP-3 is proteolyzed into fragments and thereby letting the IGF-1 free [147]. Apart from this, IGFBP-3 is also known to act independently of IGF-1 and functions to promote apoptosis triggered by p53, TGF-β1/TGF-β2 [148]. IGFBP-3 exists as a cleaved form in specific tissues such as the vitreous humor of diabetic patients, providing the identification of IGFBP-3 protease [149].In the cornea, IGFBP-3 has been reported to function in myofibroblast proliferation and differentiation [150]. The membrane-bound expression of IGFBP3 in

22 basal and suprabasal cells of central human corneal epithelium were observed in addition to its increased expression in the surface epithelial cells, thus signifying its role in cell surface shedding and apoptosis. Prominent expression of IGFBP3 in the basal cells of peripheral corneal epithelium also indicates its function in the ‘centripetal migration’ of epithelial cells [151]. IGFBP-3 has also been shown to have a potential role in apoptosis and anti-growth signaling in the human corneal epithelium [151].In this study, we will further determine the role of the IGF-1/IGFBP-3 axis in the development of corneal hem-angiogenesis and HSK disease severity.

6.2. IGF dependent action of IGFBPs

IGF dependent activity of IGFBPs involves sequestration of IGFs. This results in the rescue of IGF binding to its receptor (IGF-R) [152,153]. A plethora of reports has reported evidence on the interaction between IGFBPs and IGF [154,155,156]. This defines the IGF-dependent mechanism of IGFBPs which modulates the bioavailability of

IGF (Figure 7). Binding of IGF molecules to IGF-1R activates intracellular tyrosine domain of β-chain and causes autophosphorylation at several tyrosine residues in IGF-

1R [145]. Phosphorylation of tyrosine residues 1131, 1135, and 1136 plays an essential role in canonical signaling attributed to the receptor [157]. Activated IGF-1R phosphorylates insulin receptor substrates (IRS-1 and IRS-2) and initiates signaling cascade. This signaling cascade includes the activation of phosphatidylinositol-3-kinase

(PI3K)/AkT and mitogen-activated protein kinase pathways [158]. IGF-1R signaling has been well reported to have pleiotropic effects, which includes enhancing cellular proliferation, migration, neutrophil survival, and effector function, T-cell survival, promoting hem- and lymphangiogenesis and inhibition of apoptosis [159,160,161].

The IGF dependent action of IGFBPs was demonstrated by using a recombinant human mutant of IGFBP-4 that showed a reduced activity to bind IGF-2 molecules. This reduced binding resulted in the inhibition of osteoblast proliferation [162]. Additionally,

IGF dependent action of IGFBP was also evident from the release of IGF molecules either by proteolysis of IGFBPs or by alterations in the IGFBP phosphorylation. IGFBP-3 levels in serum have reported being regulated by the protease activity such as plasmin and serine proteases (Prostate-specific antigen), which degrades IGFBP-3 [154]. Matrix metalloproteinases (MMPs) are the enzymes known to have proteolytic activity and is involved in the degradation of IGFBPs. MMPs are the family of Zn2+- dependent enzymes that are involved in the remodeling of extracellular matrix components [163].

They are synthesized as proenzyme molecules and become active after removal of an amino-terminal pro-peptide. MMP-3 and MMP-9 are reported to trigger the proteolysis of

IGFBP-3 resulting in the release of IGF molecules [164,165,166]. Interestingly, increased IGFBP-3 proteolysis has been reported in pathological conditions such as cancer and arthritis [167]. Previous studies have shown that MMP-3 and MMP-9 are involved in the proteolysis of IGFBP-3 in inflamed tissue [168,169]. Interestingly, in the mouse model of HSK, the expression and activity of MMP-9 have been reported in

HSV-1 infected corneas [170]. While, the reports from our lab showed an increased level of both, MMP-9 and MMP-3 during the pre-clinical and clinical phase of HSK

(Figure 37). MMP-3 serves as an activator of pro-MMP-9 and helps in the generation of active MMP-9 [171]. Overall, these studies provide evidence on the proteolysis of

IGFBP-3 in inflamed tissue via MMPs that leads to the release of the bound IGF molecules.

6.3. IGF independent action of IGFBPs

IGFBPs can execute its function independent of IGF binding and also by not influencing IGF receptor signaling (Figure 7). Structurally IGFBP-3 consists of a conserved N-terminal domain, a mid-region that is highly variable, and a conserved C- terminal domain, with other subdomains and functional motifs which helps in the functional role of IGFBP-3 [172]. The mid-region of IGFBP-3 essentially plays a role to promote its interaction with glycosaminoglycans (cell surface proteins) [173]. Recent reports have suggested the specific binding of IGFBP-3 to cell surface receptor (not the

IGF receptor) in breast cancer cells which mediated growth inhibition in these cells.

Additionally, studies have also reported various types of receptors interacting with

IGFBPs such as type V TGFβ receptor. IGFBP-4, and -5 is known to interact with type V

TGFβ receptor in breast cancer cells [174,175]. Type V TGFβ receptor was found to be structurally similar to the low-density lipoprotein receptor-related protein-1 (LRP-1)

[176]. The report suggested that binding of IGFBP-3 to Type V TGFβ/ LRP-1 promoted growth inhibitory signaling in human lung carcinoma cells (H1299) [177]. A recent study has provided evidence on the involvement of the IGFBP-3 receptor with IGFBP-3 functions in retinal endothelial cells mediating apoptosis [176].

Several reports have demonstrated the IGF independent action of IGFBP-3, and

-5 via its localization to the nucleus [178]. Localization of IGFBP-3 into the nucleus has been reported due to the presence of a nuclear localization signal (NLS) and the nuclear transport protein importin-β. Once in the nucleus, IGFBP-3 executes transcription regulation. IGFBP-3 can also bind to nuclear receptors such as retinoid-x- receptor-α, Vitamin D receptor, peroxisome proliferator-activated receptor-γ (PPAR-γ),

Nur -77 [179]. Furthermore, the nuclear localization of IGFBPs is also known to promote the anti-apoptotic effect [180]. IGFBPs also binds to the epidermal growth factor receptor (EGFR) on the plasma membrane and then translocate into the nucleus in response to DNA damage and promotes DNA damage repair [181]. IGFBPs may likely promote cell adhesion and motility in an IGF independent manner. IGFBP-1, and -2 is known to bind integrin α5β1 (fibronectin receptor) to promote cellular growth signaling pathway [152]. Moreover, a plethora of reports have indicated the IGF independent action of IGFBP-3 in mediating cellular growth and angiogenesis in diabetes

[182,183,184]. Taken together, the reports have provided concrete evidence on the

IGF/IGF-R independent action of IGFBP-3 executing either promoting or inhibiting various cellular processes such as cellular proliferation, growth, and apoptosis.

Figure 1: Key aspects of Herpes Stromal Keratitis (HSK) development in C57BL/6 mouse model. The early response to control the virus occurs during the pre-clinical phase. The actively replicating virus ceases by day 7 post-infection. An influx of inflammatory neutrophils is detected during the pre-clinical phase. The disease pathology begins to develop during the clinical phase of the disease (after day 7 post -infection) with the development of hem- angiogenesis and corneal opacity. An increased influx of T-cells is detected. Additionally, another wave of neutrophil influx is also observed at the clinical disease phase. Therefore, there is a biphasic infiltration of neutrophils during the pathogenesis of HSK disease.

Figure 2: Structure of Cornea. Layers in cornea showing from the top (anterior) corneal epithelium layer, Bowman’s layer, corneal stroma, Descemet’s membrane, and endothelium at the bottom (posterior)

Figure 3: Layers of tear film. Tear film overlying the corneal epithelium. Membrane bound mucins on the corneal epithelium function as glycocalyx at the epithelial-tear film interface. Three layers of tear film from the top (anterior) is the lipid layer, middle aqueous layer (with soluble mucins), at the bottom (posterior) mucous layer.

Figure 4: Regulation of Hypoxia Inducible Factor (HIF) pathway in oxygen dependent manner. In presence of abundant oxygen (normoxia) HIF-1/2α (HIF isoforms) is prone to undergo proteasomal degradation. This process is initiated by the prolyl hydroxylases (PHD) followed by the binding of HIF isoforms to von-Hippel-lindau (VHL). However, under low oxygen supply (hypoxia), stabilization of HIF isoforms occurs upon its translocation to the nucleus. Heterodimerization of HIF isoforms occur in the nucleus with HIF-1β, along with the cofactor protein (p300/CBP). This heterocomplex formed in the nucleus promotes the transcription of HIF target genes that consists of hypoxia response element (HRE) sequences in their promoter region. (CBP-CREB-Binding protein) Modified concept from Franke.et.al; Erythrocytosis: the HIF pathway in control; Blood; 2013 122(7):1122-8.

Figure 5: Histology of Extra-orbital Lacrimal gland from a C57BL/6 mouse. The image is showing the histology of Extra-orbital lacrimal gland from C57BL/6 mice which was stained with Hematoxylin and Eosin (H&E). The image shows lobules separated by loose connective tissue (arrows are indicating the location). Each of these lobules consists of acini (arrows highlighting the acini within the lobules). An enlarged image of the acinus is shown in the image on the right, indicating acinar cells that are secretory in function. The lobules also contain ducts (an enlarged image displayed on the right) that helps in transporting the secreted components from the lacrimal gland.

Figure 6: Histology of Conjunctiva from a C57BL/6 mouse. The image is showing the histology of conjunctiva from C57BL/6 mice which were stained with Periodic acid-Schiff (PAS). A. The image shows the location of the conjunctiva in the cross-section of the Adnexa. B. An enlarged image is showing three different types of the conjunctiva. The one highlighted in the red is the bulbar conjunctiva or the ocular conjunctiva (covers the outer surface of the eye), highlighted in blue is the palpebral or the tarsal conjunctiva (lines the inner side of eyelids), highlighted in green is the fornical conjunctiva (present at the junction between the bulbar and palpebral conjunctiva). C. An enlarged image showing the goblet cells (PAS+) lined across the conjunctival epithelium.

Figure 7: Insulin Like Growth Factor (IGF) dependent and independent action of Insulin Like Growth Factor Binding Proteins (IGFBPs). In the left panel, showing the binding of IGF-1 to its receptor IGF-1R. This elicits the signaling cascade upon phosphorylation of the IGF-1R receptor. While, the presence of IGFBPs modulates the IGF signaling by its binding to the IGF molecules. This leads to the sequestration of the IGF molecules and inhibits its binding to the IGF receptor. On the right panel, showing IGBPs interaction with the cell surface receptors such as LRP-1/type V TGF-β receptor. Other cell surface receptors yet needs further elucidation. It is known that, IGFBPs upon its binding with the cognate receptor leads to its internalization inside the cell and gets transported into the nucleus in an importin-dependent fashion, thereby to promote the expression of specific genes involved in various cellular process (migration, proliferation, apoptosis etc). Modified concept from Wick.et.al; A Darwinian-evolutionary concept of age-related diseases; Experimental Gerontology 38 (2003) 13–25

MATERIALS AND METHODS

Mice

Eight- to twelve-week-old C57BL/6 mice were purchased from Jackson

Laboratories (Bar Harbor, ME). This mice strain (stock No 000664) was used for all experiments mentioned in this study. IGFBP-3-/- mice were provided by Dr.Jena Steinle laboratory (Wayne State University School of Medicine, Dept of Ophthalmology, Visual and Anatomical Sciences). The genotype of IGFBP-3-/- mice was confirmed (Figure 8).

All mice were housed at Wayne State University School of Medicine, Division of Animal

Laboratory Animal Resources (DLAR). All experimental animal procedures were reviewed and approved by the Investigation Animal Care and Use Committee (IACUC) of Wayne State School of Medicine and the Association for Research in Vision and

Ophthalmology.

Virus

HSV-1 McKrae was obtained from Dr.Daniel J.J. Carr laboratory at the University of Oklahoma College of Medicine (Oklahoma City, OK). In our laboratory, the virus was propagated on a monolayer of Vero cells (American Type Cell Culture, Manassas, VA;

CCL81). Infected Vero cells were then pelleted by centrifugation at 3000rpm for 30min at 4°C. The pellet was resuspended in 2ml PBS and then subjected to three cycles of flash freeze/thaw in liquid nitrogen and in 37°C water-bath, respectively. The virus was purified from the supernatant and then titrated to quantify the viral titers using viral plaque assay. Viral titers were designated as plaque forming units per milliliter (pfu/ml).

Aliquots of the virus were stored in -80°C until used [185].

Corneal HSV-1 infection

Primary HSV-1 infection in C57BL/6 mice was performed by anesthetizing the mice with an intraperitoneal injection of ketamine/xylazine (33mg/kg body weight/20mg/kg body weight respectively) in 0.2 ml of 1x phosphate buffered saline

(PBS). Mouse corneas were sacrificed using a 27 ½ gauge needle. 1x105 pfu of HSV-1

McKrae virus was topically applied only to the right cornea in 3ml of 1x PBS. All infected mice were housed in micro-isolator cages with cage tops and handled in Class II, Type

A2, biosafety cabinet to aid precaution against bacterial superinfection [186].

Clinical Scoring of Herpes Stromal Keratitis (HSK)

Handheld slit lamp biomicroscope (Kowa, Nagoya, Japan) was used to clinically score each of the HSV-1 infected eyes at day 8, day 10 and day 16 post HSV-1 infection, to determine corneal opacity and development of corneal angiogenesis. A 5- point scale is applied to determine the severity of corneal opacity as follows: 0-normal cornea; 1-mild corneal haze; 1.5- mild corneal haze and with epithelial defect; 2.0- moderate opacity with iris visible; 2.5- moderate opacity with iris visible and with epithelial defect; 3.0- severe opacity with iris not visible; 3.5- severe opacity with iris not visible and with epithelial defect; 4.0- severe opacity with corneal ulcer; 4.5- severe opacity with corneal ulcer and with epithelial defect; 5.0- corneal rupture. Corneal hem- angiogenesis was determined by measuring the centripetal growth of newly formed blood vessels in each quadrant of the cornea as follows; A 4-point scale is applied to determine the length of the blood vessels in each quadrant of the cornea between 0- with no angiogenesis and 4-the blood vessels that reach central region of the cornea.

The final score from all four quadrants was summed up to have a total range 0-16 score

35 for each eye at a designated time-point. Corneas with an angiogenesis score >4 and opacity score >2 were scored as severe HSK, whereas the corneas with an opacity score ≤2 and angiogenesis score ≤4 were scored as mild HSK disease [185,187].

Evaluation of Corneal Smoothness

Stereomicroscope imaging with the LED ring illuminator (Leica Microsystems Inc,

IL, USA) was used to capture mouse eye images on day 4, day 8, and day 10 post

HSV-1 infection to determine corneal smoothness and epithelial defects. The light from the ring illuminator projects off the corneal surface and the regularity of the circular ring can be determined depending on the smoothness of the corneal surface. The corneal smoothness is determined by dividing the projected ring into four quarters, of 3’o clock position each quarter. Based on a 5-point scale, the severity of corneal irregularity will be scored depending on the number of distorted quarters as follows: 0-No distortion; 1- distortion in one quarter of the ring; 2- distortion in two quarters of the ring; 3- distortion in three quarters of the ring; 4- distortion in all four quarters of the ring; 5-severe distortion with no ring formation observed [188,189],

Phenol red thread tear test

This test was used to measure the tear levels in each of the HSV-1 infected mouse eyes at day -4, -6, -7, -8 and -10 post HSV-1 infection. Each of the un- anesthetized mice was held on a stable surface (without touching the facial region or whiskers) to insert the phenol red (pH indicator) impregnated cotton thread tip (ZQ

1010, ZONE-QUICK, Japan) into the eyes at the lateral canthus region for the 30s

(Figure 9). An eye dressing forceps (Miltex, 18-779) was used to hold the thread.

Appropriate care is taken to avoid touch stimulation to eyelashes and whiskers. The

36 change in the thread color from yellow to red was measured in millimeters and recorded. The thread images with color change were captured immediately [190].

Rose Bengal staining

Rose Bengal staining was used to evaluate the damage to the ocular surface in

HSV-1 infected eyes.1% Rose Bengal (catalog #330000 Sigma Aldrich, MO, USA) solution was prepared in distilled water. 2µl of the solution was dropped on to the anesthetized mouse eyes by using a micropipette. After 4 secs, Rose Bengal staining was washed away from the mouse eyes by using 2µl 1x PBS from the micropipette. The eye images were acquired by using stereomicroscope. Positively stained areas with

Rose Bengal will appear as magenta/dark pink punctate staining on the ocular surface

[191].

Immunofluorescence staining and confocal microscopy

Immunofluorescence staining was performed on day 4, day 10 and day 12 post

HSV-1 infection and on uninfected (naïve) corneal frozen sections (8µm). Enucleated eye-globes from naïve, day 4 and day 10 post-infection were transferred into the blocks with OCT compound (Tissue Tek, CA, USA) that were snap-frozen in liquid nitrogen and stored at -80ºC. Corneal frozen sections of 8µm were collected on a positively charged glass slide (Thermofisher Scientific, MA, USA). The samples were air dried for 30min and later fixed using 2% buffered paraformaldehyde (PFA) for 30min at room temperature or with acetone (ice-cold) for 10min at -20ºC followed by three washes in

1X PBS for 5min each. Slides were blocked using 3% BSA in 1xPBS with 0.3% Triton

X-100 for 2 hours at room temperature. Slides were then incubated with FITC- conjugated mouse monoclonal antibody, raised against the pimonidazole adducts

37 formed during hypoxia or with cytoskeletal intermediate filament proteins (K1, K10, K12, pan-cytokeratin), paired box-6 protein (PAX-6), hypoxia-inducible factor-1α, 2α (HIF-1α,

HIF-2α), glucose transporter-1, 3 (GLUT-1, GLUT-3), monocarboxylate transporter-4

(MCT-4) and Insulin-like growth factor -1 receptor (IGF-1R). The next day, slides were washed for three times in 1x PBS for 5min each, later incubated for 2hr at RT with respective secondary antibody (Alexa 488). Slides were rewashed three times in 1X

PBS for 5min each and then mounted with DAPI containing medium (Vector

Laboratories, Burlingame, CA) [186,192]. The sources of antibody and dilution used in this study are provided in Table 2. Images were acquired in 20x and 40x objective lens by using Leica True Confocal Scanner SP8 Confocal microscope (Leica Microsystems

Inc, IL, USA).

Corneal whole mount staining and confocal microscopy

Mice were euthanized at day 4 and day 10 post HSV-1 infection to enucleate the eye globes. The eye-globes were then immediately fixed in 4% PFA for 30min at room temperature. The corneal tissue was gently dissected (under a dissecting microscope) separating it from the attaching iris tissue, ciliary body, scleral tissue, and the lens by using fine curved forceps (Miltex, York, PA). The corneal tissue dissected as a ‘cup shape’ (Figure 10) was flattened by making six- eight partial cuts from the limbus to the central cornea and the stored in 30% sucrose using 0.1% EDTA-0.01% hyalurodinase type IV-S solution at 37ºC for 90min. The tissues were blocked using 3% BSA in 1%

TritonX-100 solution for 90min at room temperature, followed by incubation with primary antibody (Tuj1) in a humidified chamber for 16hours at 4ºC. The next day, tissues were washed for four times with 0.3% TritonX-100-PBS solution for 10min each, followed by

38 overnight incubation with secondary antibody (Alexa 488) at 4ºC. The next day, tissues were washed for four times with 0.3% TritonX-100-PBS solution for 10min each and later mounted with DAPI containing mounting medium (Vector Laboratories,

Burlingame, CA, USA) [190]. The sources of antibody and dilution used in this study are provided in Table 2. Tile scan Images were acquired using a Leica True Confocal

Scanner SP8 Confocal microscope (Leica Microsystems Inc, IL, USA).

Bromodeoxyuridine (BrdU) incorporation assay

Naïve(uninfected) and HSV-1 infected mice at day 5 and day 9 post HSV-1 infection were given a single dose of intraperitoneal injection (0.2ml/mice) of BrdU

(10mg/ml) in saline solution. Mice were euthanized post 18hrs of BrdU injection, and the eye-globes were collected to snap- freeze in the OCT compound (Tissue Tek, CA,

USA). Frozen sections of the cornea were collected on a positively charged glass slide

(Thermo Scientific, MA, USA). The sections were let to air-dry overnight at room temperature and later fixed in 4% buffered para-formaldehyde solution for 15min, later washed thrice in 1xPBS. For antigen retrieval, sections were immersed in Coplin jar with

2NHCl for 30min at room temperature and subsequently neutralized with 0.1M Borate buffer (pH 9.0), twice for 5min each. The samples were blocked with 3% BSA in 1xPBS and incubated at 4ºC with primary antibody for 18hrs (overnight). The next day slides were washed three times followed by incubation with secondary antibody (Alexa 488) overnight at 4ºC. The next day, slides were washed and mounted with DAPI mounting medium (Vector Laboratories, Burlingame, CA, USA) [190]. The sources of antibody and dilution used in this study are provided in Table 2. Images of 20x and 40x magnification

39 were acquired using Leica True Confocal Scanner SP8 Confocal microscope (Leica

Microsystems Inc, IL, USA).

Hematoxylin and Eosin (H/E) and Periodic-Acid-Schiff (PAS) hematoxylin staining

Extra-orbital lacrimal glands (Figure 11) and Adnexa (Figure 12) from the naïve, day 4 and day 10 post HSV-1 infection were dissected and immediately fixed (in

‘Excalibur’s alcoholic fix solution’). The fixed samples were sent to the core facility

(Excalibur Pathology, OK, USA) to carry out paraffin embedding, sectioning, H/E staining for Extra-orbital gland tissue samples and PAS staining for conjunctival goblet cells (from Adnexa tissue sample). Images were acquired by using the Leica light microscope (Leica Microsystems Inc, IL, USA).

Tissue Processing for Flow cytometry (FACS)

Enucleated eyes from the naïve and infected mice at day -4, -10, -12 and -13 post HSV-1 infection were collected in ice-cold RPMI 1640 medium supplemented with antibiotics and fetal bovine serum (10%). The corneal tissue was dissected (under a dissecting microscope) separating it from the attaching iris tissue, ciliary body, scleral tissue, and the lens by using fine curved forceps (Miltex, York, PA). Each of the cornea samples was collected in ice-cold RPMI 1640 media + Liberase TL (2.5mg/ml, Sigma-

Aldrich, MO, USA) and placed in a disruptor (1500rpm) at 37ºC for 45min. Each of the cornea samples was triturated using 3ml syringe plunger and passed through a 70µm cell strainer. The samples were pelleted by centrifugation at 1000rpm for 10min (4ºC) and then broken down into single cell suspension for FACS staining.

The mouse Extra-orbital lacrimal gland (EoLG) present anterior to the ear, from the naïve and infected mice at day 4 and day 10 post HSV-1 infection were collected

(Figure 11) in ice-cold RPMI 1640 media supplemented with antibiotics and fetal bovine serum (10%). Individual EoLG tissue was minced into 4-5 pieces and then were collected into RPMI 1640 media + Liberase TL (2.5mg/ml, Sigma-Aldrich, MO, USA).

These samples were placed in a disruptor (1500rpm) at 37ºC for 45min. Each of the

EoLG samples was triturated using 3ml syringe plunger and passed through a 70µm cell strainer. The samples were pelleted by centrifugation at 1000rpm for 10min (4ºC) and then broken down into single cell suspension for FACS staining.

Conjunctival tissue (pooled- right and left eyes) from the naïve and infected mice at day 4 and day 10 post HSV-1 infection were dissected from the Adnexa (Figure 12) by using a surgical blade #11. Under the dissection microscope, loosely attached conjunctival tissue (includes palpebral, whole bulbar and forniceal conjunctiva) was separated from the sclera and eyelids by corneal scissors. The pooled conjunctival tissue was collected into RPMI 1640 media + Liberase TL (2.5mg/ml, Sigma-Aldrich,

MO, USA). The samples were placed in a disruptor (1500rpm) at 37ºC for 40min, followed by trituration using 3ml syringe plunger and was passed through a 70µm cell strainer. The samples were pelleted by centrifugation at 1000rpm for 10min (4ºC) and then broken down into single cell suspension for FACS staining.

Spleen samples from the naïve and infected mice were collected in ice-cold

RPMI 1640 media supplemented with fetal bovine serum (10%) and antibiotics.

Individual spleen tissues were triturated by using 3ml syringe plunger and was passed through a 70µm cell strainer. Red Blood Cells (RBCs) were lysed by using RBC lysis buffer (Sigma-Aldrich, MO, USA). The spleen cells were counted by using

41 hemocytometer (to plate 106 cells/well in 96 well plates for cell surface flow cytometry staining).

Flow Cytometry cell surface staining

Cell surface staining was carried out on single cell suspension processed from the tissue samples. Individual samples were plated in the 96 well u-bottom plates. Cells were washed with FACS buffer (PBS+FBS+0.1%sodium azide) followed by blocking of

Fc receptors (CD16/32 antibody) on ice for 30min. Next, the individual samples were stained with fluorochrome-conjugated antibodies. The sources of antibody and dilution used in this study are provided in Table 1. Samples were fixed overnight with 1:1 2%

PFA: FACS buffer and acquired the next day on LSR Fortessa flow cytometer. The data were analyzed using FlowJo software.

Flow Cytometry intracellular staining

Intracellular staining was carried out on single cell suspension processed from the tissue samples. Individual samples were plated in 96 well u-bottom plates. Cell surface staining on individual samples was carried out after blocking with Fc block

(CD16/32 antibody) on ice for 30min. The BD Biosciences cell membrane permeabilizing staining buffer set (Cytofix/Cytoperm 554714, BD Biosciences, CA, USA) was used for intracellular pimonidazole and IGF-1R staining. The eBioscience transcription factor staining buffer set (005523-00, Thermofisher Scientific, MA, USA) was used for HIF-1α staining. The staining buffer set was used as per the manufacturer’s protocol. The sources of antibody and dilution used in this study are provided in Table 1. Samples were fixed overnight with 1:1 2% PFA: FACS buffer and

42 acquired the next day on LSR Fortessa flow cytometer. The data were analyzed using

FlowJo software.

RNA isolation, Quantitative Reverse Transcriptase Polymerase Chain Reaction (qRT-PCR), Hypoxia Signaling Pathway RT2 profiler PCR array

Individual cornea samples dissected from naïve, day 5 and day 10 post HSV-1 infection were collected into RNeasy lysis buffer (RNeasy micro kit; QIAGEN, MD, USA) and homogenized using a Qiagen Tissue Lyser (Qiagen, MD, USA). The total RNA was extracted, as per the manufacturer’s protocol (RNeasy micro kit; QIAGEN, MD, USA).

Quantification of purified RNA was carried out using NanoDrop ND-1000 spectrophotometer (Thermofisher Scientific, MA, USA). One microgram RNA was reverse transcribed using RT2 First Strand Kit (QIAGEN, MD, USA) to prepare cDNA for the qRT-PCR assay. The optimal cycling conditions required to run the qRT-PCR assay were set, starting with the hot start cycle for 10min at 95ºC, followed by the PCR amplification for 15s at 95ºC and 60s at 60ºC with cycle repeated 40 times. The 96-well

Hypoxia Signaling Pathway RT2 Profiler PCR Array (Qiagen, MD, USA) was carried out as per the manufacturer’s protocol. The qRT-PCR assay was carried out using the CFX

Connect Real-Time PCR Detection System (Bio-Rad, CA, USA). qRT-PCR Primers used in this study are provided in Table 2. The data obtained from the qRT-PCR assay was calculated as fold differences in gene expression after normalization with β-actin

(housekeeping gene). The results were represented as relative gene expression by using the 2-ΔΔCT method of calculation [ΔΔ cycle threshold (CT) = ΔCT (experiment)-

ΔCT (control)].

Lactate colorimetric assay

L (+) Lactate concentration from the naïve (uninfected) and day 10 post HSV-1 infection corneal lysates were measured by the lactate colorimetric assay kit (MAK064,

Sigma Aldrich) as per the manufacturer’s protocol. Briefly, each of the corneal samples was collected in 150μl of lactate assay buffer. The samples were homogenized using handheld homogenizers (Random # 440613) and centrifuged at 13,000 x g for 10min at

4°C to remove insoluble material. The supernatant was passed through 10KDa spin columns (ab93349, Abcam) to deproteinize and remove the remaining lactate dehydrogenase in the sample. The Elute collected from the columns (50μl) is used to prepare a master mix containing 50μl of the reaction mixture (2µl enzyme mix, 2µl lactate probe and 46µl of assay buffer). The prepared reaction mixture was then incubated at room temperature for 30min. Lactate standards were prepared by diluting

100nmole/µl lactate to 1nmole/µl standard solution. Volumes of 0,2,4,6,8 and 10µl of

1nmole/µl lactate standard were used to generate the lactate standard curve. Sample absorbance was measured using a colorimetric microplate reader (Molecular devices, spectra max plus, NH, USA) at 570nm (A570). The results were represented as nanograms per microliter (ng/µl).

Viral Titration assay

HSV-1 virus load in infected Extra-orbital lacrimal glands (EoLG) at day 5 and day 10 post HSV-1 infection were determined by collecting the tissues in 500µl serum- free DMEM media and then stored in -80ºC until used. Quantification of infectious virus was enumerated using a standard viral plaque assay. Vero cells (CCL81, American

Type Cell Culture, Manassas, VA;) were grown in a 12-well plate (0.2x106 cells/well).

The EoLG samples to be tested for viral load were thawed on ice and homogenized using the hand-held homogenizer (Random # 440613). The EoLG samples were appropriately diluted in DMEM serum free media and added to each of the wells containing confluently grown Vero cells. The 12 -well culture plates were incubated at

º 37 C/5% CO2 for 60min. Samples were removed carefully and replaced by 1ml of minimal essential media/carboxymethyl cellulose overlay media. Plates were incubated

º 37 C/5% CO2 for 48hrs to allow the plaque formation. The overlay media was later removed gently to avoid any loss of cell sheet and replaced by 1ml of glutaraldehyde fixative (50x diluted in 1xPBS) for 5min at room temperature. After removing the fixative, cells were then stained with Coomassie blue for 1hr at 37ºC. Plaques were counted under a bright light source to enumerate the number of plaques forming units (PFU) per

EoLG.

Mouse angiogenesis protein array analysis

An array of angiogenesis-related proteins was measured in HSV-1 infected corneal samples from B6 and IGFBP-3-/- mice in naïve and at day 5, day 10 and day 15 post HSV-1 infection by using Proteome Profiler Mouse Angiogenesis Array kit

(ARY015, R&D systems, MN, USA). Cornea s dissected from the naïve and infected groups from B6 and IGFBP-3-/- mice were transferred to 1XPBS with protease inhibitor cocktail (Sigma Aldrich, MO, USA) and were stored at -80ºC. The cornea samples were sonicated at 50% amplitude with a cycle of 15s-pulse, followed by resting on ice for

1min. A total of six cycles of sonication per cornea was performed. The sonicated samples were later centrifuged at 4ºC at 15,000 rpm for 10min. The supernatant was collected to estimate the protein concentration in each sample by using the BCA protein

45 assay kit (Thermo Scientific, MA, USA). Samples were pooled to obtain a minimum of

200µg of protein for angiogenesis array analysis as per the manufacturer’s instructions.

Positive signals on the membrane were quantified using Image Studio Lite Version 4.0.

Enzyme-linked immunosorbent assay (ELISA) to determine the amount of Insulin- like growth factor binding protein-3 (IGFBP-3) and Insulin-like growth factor-1 (IGF-1)

To quantitate the amount of IGFBP-3 and IGF-1 in the corneas of B6 mice, individual cornea sample was collected in 1XPBS with protease inhibitor cocktail (Sigma

Aldrich, MO, USA) and stored at -80ºC. Individual samples were sonicated at 50% amplitude with a cycle of 15s-pulse, followed by resting on ice for 1min. A total of six cycles of sonication per cornea was performed. The sonicated samples were later centrifuged at 4ºC at 15,000 rpm for 10min. The supernatant was collected for the assay using mouse IGFBP-3 Duo set (DY775, R&D systems, MN, USA) and mouse IGF-1

ELISA (22-IGF1MS-E01, ALPCO, NH, USA) and followed as per manufacturer’s instructions. Sample absorbance was measured using a colorimetric microplate reader

(Molecular devices, spectra max plus, NH, USA).

Intraperitoneal (i.p.) injection of Hypoxyprobe-1 (Pimonidazole) to detect hypoxia

Hypoxyprobe-1(HP1-100 kit; Hypoxyprobe-1 kit, Pimonidazole Hydrochloride,

Natural Pharmacia International Inc, MA, USA) i.p. Injection (1.2mg/20g body weight) was given to the naïve (uninfected) mice and at day 4, day 10 and day 15 post-infection.

The drug was dissolved in 0.9% normal saline. Mice were euthanized ninety minutes after injection, to collect the enucleated eye-globes. The Individual eye-globe was submerged immediately into the blocks containing the OCT compound (Tissue- Tek,

CA, USA) and was snap-frozen into liquid nitrogen. The OCT blocks were stored in -

80ºC until used for sectioning and immunofluorescence staining with the anti-PIMO antibody. The source of antibody and dilution used is provided in Table 1.

Quantification of Hypoxyprobe-1 (Pimonidazole) staining

Immunofluorescence images stained with anti-PIMO antibody were quantified by measuring the area of interest showing positive staining. This was acquired by using the

‘Threshold’ tool on a grayscale image by using the Image J software (1.4q version, open source software) and by setting the measurements to analyze ‘limit to threshold’ so that the ‘Area of fraction’ in the image is only limited to the threshold region that will be highlighted. Each section is set to the threshold at different areas of positively stained areas and was performed at least three times. The quantification of pimonidazole positively stained area was represented as the percentage of pimonidazole positively stained areas that depicts the percentage of pixels selected, by using the threshold tool as described previously [193,194,195].

In-vivo depletion of neutrophils by administration of anti-Ly6G (IA8) antibody

Intraperitoneal (i.p.) injection of anti-Ly6G (clone IA8; BE0075-1, Bio X Cell) monoclonal antibody (mAb) was given to the HSV-1 infected mice at the dose of

500µg/day/mouse- as the first injection dose. Subsequent injections were given at the dose of 250µg/day/mouse to the treated group of mice. The isotype-matched antibody was administered to a control group of infected mice. Anti-Ly6G mAb antibody injections were given to deplete neutrophils starting from day 7 to day 11 post HSV-1 infection for a total of five injections. On day 12 post-infection, mice from both the groups were given i.p. Injection of hypoxyprobe-1 (pimonidazole) and were euthanized ninety minutes later to collect corneas and spleen tissue. By flow cytometry, individual corneal samples were

47 analyzed to determine the extent of pimonidazole staining in the different subsets of immune cells. To confirm the in-vivo depletion of neutrophils, individual spleen samples were analyzed to enumerate the frequency of neutrophils (Gr1high).

Intraperitoneal (i.p.) injection of the Acriflavine (ACF) for inhibition of hypoxia- inducible factor-1α, 2α (HIF-1α and HIF-2α)

i.p. Injection of ACF (A8126, Sigma Aldrich, MO, USA) was given to the HSV-1 infected group of mice, and i.p. Injection of vehicle (PBS only) to the control group of

HSV-1 infected mice. ACF i.p injection of 8mg/kg body weight in PBS- 0.2ml per mice were given twice a day, from day 6 to day 9 post HSV-1 infection or from day 6 to day

11 post HSV-1 infection, for a total of 8 or 12 injections respectively. Corneal opacity and corneal hem-angiogenesis scores were determined in the infected corneas from both groups of mice by using a hand-held slit lamp microscope as described earlier in this section. Flow cytometry analysis of individual cornea samples from both the groups was performed to enumerate the frequency of immune cell subsets at day 12 post- ocular infection.

Topical administration of Lacritin to increase tear production in HSV-1 infected C57BL/6 mouse eyes

Topical administration of lacritin (product received from Dr. Robert L. Mc Kown,

Dr. Gordon W. Laurie) was given to the HSV-1 infected group of mice, four times a day at a dose of 4µM. Control group of HSV-1 infected mice received four times a day C-25

(a lacritin truncation mutant). The treatment was given from day 4 through day 10 post

HSV-1 infection. Tear levels were measured in both the treated and control group of mice at day 6, 8 and 10 post-ocular infections. Corneal opacity and corneal hem- angiogenesis scores were determined in the infected corneas from both groups of mice by using a hand-held slit lamp microscope as described earlier in this section. Flow

48 cytometry analysis of individual extra-orbital lacrimal glands from both the groups was performed to detect the influx of immune cells (CD4+ CD45+) at day 10 post-ocular infection.

Statistical Analysis

Statistical analysis was carried out by using the GraphPad Prism software (San

Diego, CA). Each of the experiments was repeated at least two times. The minimum number of mice used in all experiments were five (n=5). Significance (p values) in all experiments were calculated by using an unpaired two-tailed parametric t-test or non- parametric Mann- Whitney U test. Statistical analysis in two independent groups was performed by using One-way ANOVA.

Figure 8: Protein Array blot is showing no IGFBP-3 expression to confirm the knock out -/- phenotype in IGFBP-3 mice. A. An Image of the protein array blot showing IGFBP-3 protein in the C57BL/6 mice, while no expression of IGFBP-3 was detected in the IGFBP-3-/- mice.

Figure 9: Phenol Red Thread Test in C57BL/6 mouse eyes. Image showing inserting of the cotton thread impregnated with phenol red (ZONE-QUCIK, Japan) into the lateral canthus of C57BL/6 mouse eyes by using the forceps. The phenol red upon contact with tears will change its color from yellow to red.

Figure 10: Dissection of Clean Cornea from C57BL/6 mouse eyes. A. Showing image of dissected whole eye globe placed in ice cold sterile RPMI supplemented media with 10% FBS. B. Showing image of dissected clean corneal cup. Dissection was performed by removing out the lens, attached iris, and ciliary process. Curved corneal scissors (Miltex, York, PA) was used for this dissection.

Figure 11: Dissection of Extra-orbital lacrimal gland from C57BL/6 mouse. Image of Extra- orbital gland location in the temporal area. The circled area in the image showing the surgical area with the extra-orbital lacrimal gland (EoLG) located with the ducts (lacrimal ducts) branching off from the glands. Enlarged image of the dissected EoLG is shown in the inset.

Figure 12: Dissection of Adnexa from a C57BL/6 mouse. A. Showing dissected Adnexa with a whole eye globe and eyelid. B. Displaying an enlarged image of the Adnexa. Dissected Adnexa is used to further dissect the bulbar conjunctiva for FACS staining, or the whole adnexa is fixed for histology preparation.

Antibody Clone Fluorochrome Dilution Catalog # HIF-1 alpha Monoclonal Alexa 647 1:20 R&D systems, IC1935R HIF-2 alpha Polyclonal Unconjugated 1:100 Novus, NB100-122

GLUT-1 EPR3915 Alexa 488 1:50 Abcam, ab195359 Monoclonal GLUT-3 Polyclonal FITC 1:200 Abcam, ab136180 MCT-4 Polyclonal Un-conjugated 1:100 Novus, NBP1-81251 (SLC16A3) Hypoxyprobe-1 Monoclonal FITC 1:100 Hypoxyprobe, FITC-Mab (Pimonidazole) Donkey anti- Polyclonal Alexa 488 1:200 Molecular Probes Rabbit A21206 Isotype Control Monoclonal Alexa 647 1:20 R & D systems, IC002R

CD4 RM4-5 BV605 1:100 BD Biosciences, 563151

CD11b M1/70 PerCp-Cy5.5 1:100 BD Biosciences, 550993

CD45 30F-11 PE-Cy7 1:100 BD Biosciences, 552848

Ly6G 1A8 Alexa 700 1:100 BD Biosciences, 561236

Tuj1 1-15-79 Un-conjugated 1:100 Covance, MRB-435P-100 Monoclonal (now available from Biolegend) BrdU BU1/75 Un-conjugated 1:100 Abcam, ab6326 Monoclonal K10 EP1607-IHCY Unconjugated 1:200 Abcam, ab76318 Monoclonal

K12 SC-17101/L-15 Un-conjugated 1:100 Santa Cruz Polyclonal Biotechnologies, sc-17101 K1 EPR17870 Unconjugated 1:200 Abcam, ab185629 Monoclonal

PAX-6 Poly19013 Un-conjugated 1:150 Biolegend, PRB-278P

IGF-1R (pY1131) Alexa 647 1:10 BD Biosciences, 558588 K74-218

Table1: List of Antibodies used for Immunofluorescence and Flow cytometry

Catalog # Gene Symbol Description Ref Seq Accession #

PPM03503F Hk2 Hexokinase 2 NM_013820.3 PPM36694A Pfkp Phosphofructokinase, platelet NM_019703.4 PPM05505C Pfkfb3 6-Phosphofructo-2- NM_133232.3 kinase/fructose-2, 6- biphosphatase 3 PPM04168A Slc2a1 Solute carrier family 2 NM_011400.3 (facilitated glucose transporter), member 1 PPM04164A Slc2a3 Solute carrier family 2 NM_011401.4 (facilitated glucose transporter), member 3 PPM27328A Slc16a3 Solute carrier family 16 NM_030696.3 (monocarboxylic acid transporters), member 3 PPM28207F Pdk1 Pyruvate dehydrogenase NM_172665.5 kinase, isoenzyme 1 PPM29621B Cs Citrate synthase NM_026444.3 PPM34700A Idh2 Isocitrate dehydrogenase NM_173011.2 2(NADP)+, mitochondrial PPM05102A G6pdx Glucose-6-Phosphate NM_008062.2 dehydrogenase X-linked PPM82289A Pgls 6-Phosphogluconolactonase NM_001294269.1 PPM35183A Pgd Phosphogluconate NM_001081274.1 dehydrogenase

Table2: List of Qiagen Primers used for qRT-PCR assay

CHAPTER 1: DEVELOPMENT OF INFLAMMATORY HYPOXIA AND PREVALENCE OF GLYCOLYTIC METABOLISM IN PROGRESSING HERPES STROMAL KERATITIS LESIONS

Abstract

Herpes Stromal keratitis (HSK) is a chronic immuno-inflammatory condition in cornea caused by recurrent herpes simplex virus-1 (HSV-1) infection. One of the well- known causes of the development of hypoxia is chronic inflammation. Here, our study was focused on understanding the three key aspects of hypoxia, namely, (i) development of hypoxia in HSV-1 infected mouse cornea, (ii) identification of hypoxia- associated glycolytic gene expression in HSV-1 infected mouse cornea, and (iii) the results of blocking hypoxia-inducible factor (HIF) dimerization on the pathogenesis of

HSK. The results from this study showed the development of hypoxia associated with an increased level of glycolytic gene expression, an elevated level of lactate concentration and an increased level of neutrophils influx during clinical disease onset of HSK. However, in-vivo administration of anti-Ly6G antibody not only depleted neutrophils but also reduced the development of hypoxia that was detected by anti- pimonidazole staining in the HSV-1 infected corneas. Nuclear stabilization of HIF-1α and HIF-2α was observed in the infiltrating immune cells and corneal epithelium respectively, in the HSV-1 infected cornea. Intriguingly, in-vivo administration of acriflavine inhibited nuclear stabilization of both, HIF-1α and HIF-2α during clinical disease onset of HSK.

Additionally, acriflavine treatment also resulted in the reduced corneal hem- angiogenesis with yet, an increased corneal opacity in HSV-1 infected corneas. The

Immune cell influx upon acriflavine treatment increased neutrophil influx and in contrast,

57 a decrease in an influx of CD4 T cells and non-granulocytic myeloid cells. Overall, this study sheds light into the key factors involved in the development of inflammatory hypoxia and the expression of hypoxia-associated genes during the disease progression of HSK.

Introduction

Hypoxia is the most common phenomenon that occurs in tissues with inflammation [197]. The major event that occurs during hypoxia is the nuclear stabilization of HIF. The transcription factor, HIF remains inactive during abundant oxygen supply in the cells but gets activated in hypoxic conditions (depleted oxygen supply in the cells) [197]. The HIF isoforms, namely, HIF-1α and the HIF-2α. The isoforms get complexed as a heterodimer with the constitutively expressed isoform,

HIF-1β [197]. The heterodimer translocates into the nucleus and promotes the transcription of a battery of genes that play a role in cell survival, cell differentiation, angiogenesis, and glucose metabolism [198,199].

HSK is a chronic inflammatory condition that develops in response to HSV-1 infection in the cornea. As described earlier, HSK in a mouse model is broadly classified into pre-clinical and clinical disease phase [29]. During the pre-clinical disease phase, infiltrating innate immune cells with inflammatory monocytes and the NK cells and the type 1 interferon signaling events are involved in the process of viral replication

[200,201]. However, viral clearance occurs at 6-day post-infection. During the clinical disease phase, the influx of leukocytes occurs, especially the T-cells in the corneal stroma. The neutrophil influx has been studied as the most prominent innate immune cell type during both, the pre-clinical and the clinical disease phase of HSK [30,202].

Development of corneal hem- and lymph-angiogenesis, opacity and scarring that affects the clear vision and visual acuity are the consequences of the increased neutrophil influx in HSV-1 infected cornea [203,204]. It has been previously studied that the influx of neutrophils into mucosal tissue can re-model the development of inflammatory hypoxia [205,206]. Therefore, the focus of the current study was to investigate the development of hypoxia and hypoxia-associated gene expression during the progression of HSK disease.

Results

Development of inflammatory hypoxia in HSV-1 infected cornea

HSV-1 infected corneas with inflamed lesions are known to become hypoxic

[206,197]. During the clinical onset of HSK disease at day 7 post-infection, a massive influx of neutrophils is observed in the mouse model with HSV-1 infection in cornea

[202]. To investigate the development of hypoxia, a chemical hypoxia marker namely, pimonidazole (2-nitroimidazole) was used to detect hypoxic regions in-vivo.

Pimonidazole undergoes a cascade of chemical reactions and gets reductively activated in cells with hypoxia. Pimonidazole forms stable adducts by binding to the thiol groups in amino acids, peptides, and proteins [194,207]. These adducts can be imaged in-vivo by staining the tissue sample with FITC conjugated anti-pimo antibody. Notably, pimonidazole is not active within apoptotic and necrotic regions but specifically indicates hypoxic regions in the tissue [207]. HSV-1 infected mice were injected with pimonidazole drug intraperitoneally (i.p.) both at pre-clinical (day 4 post-infection) and at clinical disease phase of HSK (day 10 and 15 post-infection) as described in Materials and Methods. Our results showed faint staining of pimonidazole at day 4 post-infection

59 in contrast to an intense pimonidazole staining at day -10, and -15 post-infection in

HSV-1 infected cornea (Figure 13 A). The controls used for pimonidazole staining were the naïve (uninfected control), day 10 post infected cornea with no pimonidazole drug injection and, the isotype control. As expected, all the controls showed no positive staining of pimonidazole. Next, we showed a progressive increase in the influx of the neutrophils in HSV-1 infected corneas at the clinical disease onset of HSK (Figure 13

B). Interestingly, these results correlated with the magnitude of the pimonidazole staining. Additionally, quantification of pimonidazole staining was carried out that showed a significant increase in the pimonidazole stained hypoxic regions in the severe

HSK compared to the mild HSK diseased corneas (Figure 13 C).

Furthermore, to determine the contribution of neutrophils influx to the development of hypoxia in HSV-1 infected cornea, the depletion of neutrophils was carried out using the anti-Ly6G (IA8 clone) monoclonal antibody (mAb). The HSV-1 infected mice were injected i.p. with anti-Ly6G mAb from day 7 through day 11 post- infection as described in materials and methods. Our results confirmed the depletion of neutrophils by detecting the reduced frequency of Gr1 high cells (Figure 14 A) in the spleen of HSV-1 infected mice at day 12 post-infection. In correlation to the depletion of neutrophils, a significant decrease in the magnitude of the pimonidazole staining was observed in HSV-1 infected corneas (Figure 14 B). Recent reports have demonstrated the pimonidazole staining and its quantification on immune cells by using flow cytometry

[208]. In our study, HSV-1 infected corneas, from the mice treated with anti-Ly6G mAb showed a significant decrease in the pimonidazole staining on non-leukocytes and granulocytes by using flow cytometry. On the other hand, no significant decrease in

60 pimonidazole staining was noted on the CD4 T cells (Figure 14 C). Overall our results corroborate the involvement of neutrophil influx to be the cause for the development of hypoxia in HSV-1 infected corneas with progressing HSK lesions.

Increased mRNA levels of hypoxia-associated genes in HSK developing corneas

Recent reports have indicated the importance of hypoxia in altering the gene expression profile, such as regulating cellular metabolism in the inflamed tissue [199].

Therefore, we hypothesized that the development of hypoxia in HSV-1 infected cornea is not a bystander event. The genes involved in hypoxia-associated signaling was detected was using the Hypoxia Signaling Pathway RT2 Profiler PCR Array on naïve, day-5, -10 post-infection cornea samples. Increased expression of genes encoding lactate transporter, glucose transporters, and glycolytic enzymes were detected in HSV-

1 infected corneas at day-10 post-infection (Figure 15). The qRT-PCR assay was performed to corroborate the results obtained from the PCR array. Our data showed a significant increase in the mRNA levels of the enzymes involved in glycolysis namely,

Hk2 (Hexokinase-2), Pfkp (phosphofructokinase platelet), Pfkfb3 (6-phosphofructo-2- kinase/fructose-2,6-biphosphatase 3) with an average of 3.49, 6.06 and, 7.1-fold increase respectively (Figure 16 A). Additionally, an increase in mRNA levels of glucose transporters namely, Slc2a1(Glut-1; 4.0-fold increase), Slc2a3 (Glut-3; 33.25-fold increase) and, Slc16a3 (Mct-4; 22.5-fold increase) was determined in HSV-1 infected corneas at day 10 post-infection with progressing HSK lesions (Figure 16 B). Apart from glycolysis, mRNA levels of genes contributing the other oxygen-dependent metabolic pathways were explored such as the Pentose Phosphate Pathway (PPP), and

Tricarboxylic Acid Cycle (TCA) in HSV-1 infected corneas.

Interestingly, our results showed no significant increase in the genes involved in the mRNA levels of genes involved in regulating the oxidative phase of PPP namely,

G6pd (glucose-6-phosphate dehydrogenase), Pgls (6-phosphogluconolactonase), Pgd

(phosphogluconate dehydrogenase) (Figure 16 D). However, a decrease in the mRNA level of a gene involved in the TCA cycle namely, Pdk1 (pyruvate dehydrogenase kinase-1; an average of 40% decrease) was determined (Figure 16 C). PDK1 phosphorylation of PDH enzyme leads to deactivation of the enzyme activity in converting pyruvate to acetyl CoA. The acetyl CoA entering the TCA cycle is then converted to citrate by citrate synthase. Therefore, from our results, we further hypothesized that a decrease in the mRNA levels of Pdk1 would likely influence increased enzyme activity of PDH in HSV-1 infected corneas. Intriguingly, our results showed a significant increase in the mRNA levels of Idh2 (isocitrate dehydrogenase-2, mitochondrial), with an average of 25% reduction in Cs (citrate synthase) in HSK developing corneas compared to the control (naïve corneas) (Figure 16 C). Altogether, the results obtained from the gene expression profile suggests that glycolysis is not the exclusive metabolic pathway, but also TCA cycle may likely play a trivial role in HSV-1 infected corneas during the clinical onset of the HSK disease.

As aforementioned, our results showed a dramatic increase in the mRNA levels of glucose transporters (Glut-1, Glut-3) and the lactate transporter (Mct-4) in HSV-1 infected corneas (Figure 16 B). We were further interested in studying the localization pattern in the frozen sections of the naïve and HSV-1 infected corneas (day-10 post infection) by employing immunofluorescence and confocal imaging. GLUTs and MCTs are the cell membrane-associated proteins with its prime function being glucose import

62 and lactate export respectively [209,210]. Our results showed a faint membrane staining of GLUT-1 only in the basal layer of the naïve corneal epithelium, while in day 10 post- infection cornea intense staining of GLUT-1 was observed in the basal and wing cell layer of the corneal epithelium (Figure 17 A). Surprisingly, GLUT-1 staining was not observed in the corneal stroma, which was massively infiltrated with neutrophils and

CD4 T cells during HSV-1 infection.

On the other hand, GLUT-3 staining was very clearly observed in the corneal stroma (Figure 17 A), indicating its importance as an essential glucose transporter in the corneal stromal cells of HSV-1 infected cornea. Our results as shown previously (Figure

16 B), a significant increase in the mRNA levels of MCT-4 was observed (average of

33.25±9.82-fold increase) in HSV-1 infected cornea. In support with this, MCT-4 also showed an intense membrane staining in the epithelial and stromal layer of the cornea

(Figure 17 A). Additionally, flow cytometry results also showed a significant increase in the expression of MCT-4 protein on the neutrophils than the CD4 T cells in HSV-1 infected corneas (Figure 17 B). MCT-4 isoform is well known to execute lactate export and is reported to be highly expressed in hypoxic tissues with increased glycolysis.

Lactate is the metabolic end-product of glycolysis that is exported or imported by various isoforms of MCTs [210,209]. Therefore, understanding the importance of MCT-4 in exporting the intracellular lactate, we determined the lactate levels in the naïve vs.

HSV-1 infected corneas (Figure 17 C). Our data showed high levels of L-lactate in day

10- post-infection corneas compared to the uninfected corneas. Altogether, our data support the prevalence of glycolytic metabolism in HSV-1 infected corneas with progressing HSK lesions.

Stabilization of HIF-1α and HIF-2α in HSV-1 infected corneas

Nuclear localization of the HIF isoforms (HIF-1α and HIF-2α) is the hallmark of hypoxia [211]. Therefore, we ascertained the nuclear localization of HIF-1α and HIF-2α in HSV-1 infected corneas by employing confocal imaging and flow cytometry respectively. Nuclear localization of HIF-2α was distinct in the corneal epithelium of the infected corneas at day 10 post-infection (Figure 18). On the contrary, no nuclear localization of HIF-2α was observed in the naïve or the corneas at day 4 post-infection

(Figure 18).

HIF-1α is well known to regulate the immune cell function [212,213,214]. During the clinical onset of HSK disease, corneal stroma is heavily infiltrated with CD4 T cells and neutrophils. Hence, we carried out intracellular flow cytometry analysis (as described in materials and methods) to determine the HIF-1α stabilization in infiltrating immune cells in HSV-1 infected corneas. Our data showed increased HIF-1α expression in the leukocytes than the non-leukocytes in the HSV-1 infected corneas during the clinical disease period (Figure 19 B). Interestingly, differential levels of HIF-1α nuclear localization were detected in the order of myeloid cells > CD4 T cells > neutrophils

(Figure 19 B). Naïve B6 corneas only showed minimal levels of HIF-1α in the resident leukocytic cell populations (Figure 19 A). Overall, these results demonstrated the nuclear colocalization of HIF-1α/ HIF-2α showing a pattern of differential levels of its expression in specific cell types in HSV-1 infected cornea during the clinical disease period.

Acriflavine treatment aggravated corneal opacity, but reduced hem-angiogenesis in HSV-1 infected cornea As aforementioned, our results demonstrated the stabilization of HIF-1α and HIF-

2α in immune cells and corneal epithelium respectively, in HSV-1 infected corneas.

Therefore, we selected the Acriflavine (ACF) drug to inhibit HIF signaling via direct binding to the PAS domain of HIF-1α and HIF-2α. This results in the blocking of HIF dimerization with HIF-1β causing inhibition of HIF transcription activity [215]. Systemic administration of ACF was given to HSV-1 infected mice as described in materials and methods. The effect of the drug treatment was determined by scoring corneal opacity and corneal angiogenesis in both the control and treated group at day 10 post-infection.

Unexpectedly, our results showed a significant increase in the corneal opacity scores

(Figure 20 A), while a significant decrease in angiogenesis scores was observed in the

ACF treated group of HSV-1 infected mice (Figure 20 C). Additionally, an increased incidence of eyes with an opacity score of ≥3.0 was observed in ACF treated a group of mice (Figure 20 B). In ACF treated group of mice, a significant decrease in the mRNA levels of Pfkfb3 was observed in the infected corneas compared to the vehicle-treated group of HSV-1 infected mice (Figure 20 D). Intriguingly, our results showed a significant decrease in the mRNA levels of Pfkfb3 in correlation to a significant decrease in angiogenesis scores in the ACF treated group of infected mice. In conclusion, our data suggested that ACF treatment in HSV-1 infected mice are efficient in reducing the corneal angiogenesis alone, but not the corneal opacity.

To further determine the potency of ACF treatment in inhibiting the nuclear localization of HIF-1α and HIF-2α in a non-immune and immune cell population, we performed confocal imaging, and intracellular flow cytometry analysis on cornea

65 samples from both groups (ACF and vehicle-treated) of HSV-1 infected mice. Our results from flow cytometry analysis showed a significant decrease in the stabilization of

HIF-1α in the neutrophils and CD4 T cells in ACF treated group when compared to the vehicle-treated group of mice at day 12 post-infection (Figure 21 A). Our results from confocal images showed no stabilization of HIF-2α in the corneal epithelium of ACF treated group at day 12 post-infection (Figure 21 B)

To study the immune influx in ACF treated a group of HSV-1 infected mice, the corneas from both groups of HSV-1 infected mice were processed for flow cytometry cell surface staining. Our results showed a decreased frequency of non-granulocytic myeloid cells and CD4 T cells while an increased frequency of neutrophil influx was observed in ACF treated a group of infected mice at day 12 post-infection (Figure 22)

Overall, these results highlight the efficacy of ACF treatment in modifying the stabilization of HIF in immune and non-immune cell population and their role in modulating the leukocyte subsets in HSV-1 infected corneas at the clinical disease phase.

Discussion

Hypoxia has been well studied in association with inflammation in tissues [197].

The oxygen demand in the inflamed tissue is high due to the massive influx of metabolically active neutrophils. This is one of the likely causes for the development of hypoxia [216]. The process of neutrophil migration in the inflamed tissue requires an abundant amount of energy [216, 217]. Remarkably, in the mouse model of inflamed bowel disease, neutrophils are reported to shape the development of hypoxia.

Neutrophils in inflamed tissue carry out a mechanism of ‘respiratory burst’ executed by

NADPH oxidase (NOX) activity. During this process, oxygen consumption is high and therefore leads to the oxygen depletion in the local area.

Consequently, this causes inflammatory hypoxia [205]. Neutrophils play a crucial role in HSK development. Reports have indicated a positive correlation between the disease severity and increased influx of neutrophils in HSV-1 infected cornea [30]. In support, we proposed a model illustrating the development of inflammatory hypoxia in

HSV-1 infected cornea. The newly formed blood vessels in the HSV-1 infected cornea will supply molecular oxygen, while infiltrating neutrophils may likely cause the oxygen depletion via NADPH oxidase activity causing hypoxia (Figure 23).

Hypoxia has been well reported to influence glucose metabolism due to the increased energy demand in the inflamed tissue that harbors metabolizing active cells

[199]. A significant increase in the mRNA levels of glycolytic genes in HSV-1 infected corneas was detected. The process of glycolysis is commenced by the uptake of glucose which is carried out by the membrane-bound glucose transporters (GLUTs).

[218]. Among the various types of GLUTs, GLUT-1 has shown to be vital for CD4 function [219]. Recent reports have suggested that GLUT-1 is essential for CD4 T cells in HSV-1 infected mice since the inhibition of glucose uptake reduced the HSK disease severity [220]. Notably, our results showed membrane-bound GLUT-1 staining only in the corneal epithelial cells but not in the corneal stroma which is heavily infiltrated with

CD4 T cells and neutrophils during the clinical disease of HSK. To our surprise, prominent GLUT-3 staining was observed in the corneal stroma of HSV-1 infected mice.

Reports have indicated that GLUT-3 has a higher affinity for glucose than GLUT-1 [221].

Hence, our results discern the functionality of GLUTs in HSK cornea suggesting a

67 crucial role of GLUT-3 in mediating glycolysis. Glycolytic cycle culminates with lactate as its end product especially during hypoxia [222,223]. The lactate formed in the cytosol is transported out of the cell via lactate transporter namely, monocarboxylate transporters (MCTs). MCTs are present in different isoforms, among which MCT-4 is highly upregulated in hypoxia [210]. A significant increase in the mRNA levels of the

MCT-4 gene, in correlation to the increased levels of lactate in HSV-1, infected corneas was observed . This suggests, MCT-4 to be a designated exporter of lactate in the corneas with HSK lesions. Lactate is a metabolic waste from glycolysis that is exported out of the cell via MCT-4 and imported via MCT-1 as a metabolic fuel by cells undergoing mitochondria respiration [224]. Our data showed an increased mRNA level of MCT-1 gene in HSV-1 infected cornea (data not shown) as depicted in our proposed model (Figure 23), HSV-1 infected corneas may probably have pockets of oxygenated and hypoxic regions. In oxygenated regions mitochondrial respiration (tricarboxylic acid cycle- TCA) predominates by conversion of pyruvate into acetyl CoA. At this point, we infer from our results that lactate is a metabolic byproduct of glycolysis which serves as an auxiliary metabolic fuel to promote mitochondrial respiration and survive through hypoxia in HSK lesions. In support, our results also showed a significant increase in the mRNA levels of Pdk1 that indicates an increased generation of acetyl CoA followed by its entry into the TCA cycle in HSV-1 infected corneas. ‘Reductive carboxylation’ is another way responsible for generating acetyl CoA derived from glutamine. It is known that under a hypoxic condition or defective mitochondrial respiration, isocitrate dehydrogenase-2 promotes glutamine-derived citrate synthesis. Citrate later gets converted into acetyl CoA, thereby increasing the pool of acetyl CoA that may likely

68 enter the fatty acid metabolism or continue through the TCA cycle [225,226].

Interestingly, our results show increased mRNA levels of the Idh2 gene, that is probably executing the process of reductive carboxylation.

Stabilization (nuclear localization) of HIF-1α/ HIF-2α) is the hallmark of hypoxia.

The HIF isoform gets heterodimerized with HIF-1β and later gets translocated into the nucleus to initiate the transcription of an array of genes involved in angiogenesis, cellular metabolism, and apoptosis, etc. HIF-1α/ HIF-2α is known to have distinct functions. HIF-1α is known to regulate glycolysis and promote expression of genes encoding glucose/lactate transporters and glycolytic enzymes. While, HIF-2α may have an indirect role in initiating glycolysis via c-Myc or related factors [227]. Therefore, this indicates HIF-1α to possess a principal role in glycolytic metabolism [199,212]. Our study shows the stabilization of both, HIF-1α/ HIF-2α, which implies that both isoforms have a significant role in influencing mRNA levels of glycolytic genes in the corneas with

HSK lesions. Apart from its role in regulating cellular metabolism, HIF-1α is also known as the ‘regulator of inflammation’ [212,213]. The HIF-1α expression has been determined in the innate and adaptive cell population, whereas HIF-2α expression has been studied in various cell types including the epithelial, endothelial and immune cells

[214,228,229,230]. Intriguingly, our study showed a differential expression of HIF-1α on immune cell types including the myeloid cells, neutrophils, and CD4 T cells, while HIF-

2α was exclusively detected in the corneal epithelium of HSV-1 infected mice. Hence, we conclude from our results that HIF plays an essential role in corneal inflammation.

We were further interested in studying the consequences of blocking the HIF function by using acriflavine (ACF) in HSV-1 infected mice. ACF binds to both HIF-1α/ HIF-2α and

69 inhibits its heterodimerization with HIF-1β which leads to the blocking of HIF transcription activity [231,215]. We expected the blocking of HIF transcriptional activity by ACF to alleviate the HSK disease severity. ACF action has been widely studied in tumor models to reduce angiogenesis [215]. As expected, our results showed a significant decrease in angiogenesis in the ACF treated group compared to the control group of HSV-1 infected mice. Interestingly, these results correlated with the significant reduction in the mRNA levels of the Pfkb3 gene (encoding glycolytic enzyme). Pfkb3 is known to be HIF dependent and involved in promoting pathological angiogenesis since its inhibition reduced the extent of the angiogenesis [232].

On the other hand, our results showed that ACF treatment aggravated the corneal opacity. This was an unexpected result. Moreover, ACF treatment also showed a significant increase in the influx of neutrophils in the corneas with HSK lesions.

Studies from HIF blocking have reported increasing the ROS levels in tumor models

[233]. ACF treatment may increase ROS levels due to an increased influx of neutrophils in the corneas with HSK lesions. Hence, we concluded that ACF as a sole treatment is not efficacious in alleviating the HSK disease severity.

In summary, our research demonstrates the development of inflammatory hypoxia, correlating with the nuclear localization of HIF isoforms, and the prevalence of glycolytic metabolism in corneas with HSK lesions. Additionally, our data indicate that blocking HIF with ACF alone may not be potent in reducing HSK disease severity.

Overall, this study provides novel insights to promote the advancement of therapeutic targets in targeting both, the blocking of HIF and inhibiting corneal inflammation.

Figure 13: Development of inflammatory hypoxia corelates with influx of neutrophils in the HSV-1 infected corneas. A. Representative immunofluorescence and confocal images showing pimonidazole (hpi) staining in uninfected (naive) cornea and HSV-1 infected corneas at 4-, 10-, and 15-, post-infection. Prominent pimonidazole adduct staining (in green) is detected in infected corneas during clinical disease period (10-day through 15-day post-infection). DAPI nuclear stain is shown in blue. FITC- conjugated mouse IgG isotype control did not show hpi staining in infected cornea (Control 1). Similarly, the HSK cornea from a mouse that did not receive pimonidazole injection, does not show specific pimonidazole adduct staining in the cornea when stained with FITC conjugated anti- pimonidazole antibody (Control 2). The images shown were acquired with 40x objective, except the control slides which were acquired with a 10x objective. DAPI nuclear staining in blue and pimonidazole adduct staining (in green B. Representative FACS plots denote an increased frequency of neutrophils (Ly6G+F4/80-) in HSV-1 infected corneas during the progression of clinical disease at day 10 and day 15 post-infection. The progression of clinical disease was also associated with a decrease in the frequency of macrophages (Ly6G-F4/80+) in infected corneas. The frequency of neutrophils and macrophages was obtained from CD45+ leukocytes in infected corneas. C. Scatter plot showing the quantification of pimonidazole staining in naïve and infected corneas with mild and severe HSK at 10-day post-infection. This depicts the magnitude of hypoxic regions (n=3 naïve corneas, n=8 mild HSK corneas, n=9 severe HSK corneas). Data derived is shown from two similar experiments. p values were calculated using non-parametric Mann-Whitney U test **** p≤0.0001

Figure 14: Significant depletion of neutrophils correlated with reduced development of hypoxia in HSV-1 infected corneas. A. Representative FACS plots denote the frequency of CD11b+ Gr1+ population in the splenocytes from isotype control, and anti-Ly6G treated groups of mice on day 12 post- infection. FACS plots were derived from live-gated cells. Scatter plots (right) denotes the frequency of Gr1+ cells per spleen tissue from both groups of mice (open and closed circles) on day 12 post-infection. Each circle represents an individual spleen tissue from HSV-1 infected mice (n=6). Figure legend Continues

Figure 14: Legend continues B. Representative Eye image from the control group of mice at 12-day post ocular infection and a corresponding corneal section showing pimonidazole staining (left). Representative Eye image from Anti- Ly6G treated mice at 12-day post ocular infection and corresponding frozen corneal section showing pimonidazole staining. Bar graph (right) denoting the percentage of pimonidazole staining area (quantified by using Image J software) in both groups of mouse corneas at 12-day post ocular infection. Pimonidazole staining shown is representative of two independent experiments (n=5 corneas per group). DAPI nuclear staining in blue and pimonidazole adduct staining (in green). Magnification X400. C. Representative pseudocolor FACS plots demonstrating gating strategy to show singlets gated cell population using the forward scatter height vs. area. Furthermore, live gated the cells using side and forward scatter to distinguish three cell populations a) non-leukocytes b) Granulocytes c) CD4+ T cells. Representative histogram FACS plots (top to bottom) denoting analysis of each of the cell population to determine the mean fluorescence intensity (MFI) of pimonidazole compared to the isotype control (Black line) in the control group (Magenta line) vs anti-Ly6G (blue line) treated group of mouse corneas at 12- day post ocular infection. Scatter plot/bar graph (bottom panel) denoting pimonidazole MFI in all the three histograms analyzed cell populations at 12-day post-ocular infection. Data derived is shown from two similar experiments with n=5 per group. P values were calculated using unpaired non-parametric t-test. ** p≤0.01.

Figure 15: Hypoxia Signaling Pathway RT2 Profiler Array to show elevated expression of glycolysis regulating genes in the corneas with HSK lesions. PCR Array of hypoxia-associated signaling pathway was carried out on naive, 5-day and 10-day post-infection cornea samples showed an elevated expression of genes encoding glycolytic enzymes, glucose transporters, and the lactate transporter at 10-day post-infection.

Figure 16: Quantitative Real-Time Polymerase Chain Reaction Assay (RT-qPCR) to show elevated expression of glycolytic genes in the corneas with HSK lesions. RT-qPCR was carried out on uninfected and HSV-1 infected individual cornea samples. Bar graphs document the fold change in the level of expression of (A) glycolytic enzymes (HK2, PFKP, PFKFB3), (B) glucose transporters (Slc2a1 and Slc2a3), and lactate transporter (Slc16a3), (C) tricarboxylic acid cycle genes (Pdk1, Cs, Idh2), (D) Pentose Phosphate Pathway genes (G6pd, Pgls, Pgd) in infected individual corneas at 10-day POI in comparison to uninfected corneas. Data shown are representative of two similar experiments (n=5 corneas/group). Statistical significance was calculated using parametric two-tailed Student’s t-test.

Figure 17: Prevalence of glycolytic metabolism in the HSV-1 infected corneas. A. Showing membrane localization of GLUT-1 (Slc2a1), GLUT-3 (Slc2a3) and MCT-4 (Slc16a3) in infected corneas with HSK at 10-day post infection. Naive and HSV-1 infected frozen corneal sections were stained with anti-GLUT-1 (Slc2a1), GLUT-3 (Slc2a3) and MCT-4 (Slc16a3) antibodies followed by incubation with Alexa-488 conjugated secondary antibody. Representative images document weak GLUT1 and MCT4, but no GLUT3 staining in the corneal epithelium of naive corneas (top row). Intense GLUT-1 staining was seen only in the corneal epithelium of infected eyes, whereas GLUT3 and MCT4 staining was evident in epithelium and stroma of infected eyes at 10-day post-infection (bottom row). Image magnification is at 400X. DAPI (blue, nuclear staining), GLUT-1, GLUT-3, MCT-4 (green). B. Representative FACS plots denoting the frequency of CD4+ T cells (top panel), Ly6G + neutrophils (bottom panel) in infected corneas at 10-day post-infection. The histogram FACS plots derived from the gated population of CD4+ T cells and Ly6G+ neutrophils represent the expression level of MCT4 proteins in these immune cells. Scatter plot/bar graph (on right) demonstrate significantly increased level of MCT-4 protein in neutrophils than CD4+T cells in infected corneas. C. Scatter plot is showing significantly elevated level of L-lactate at 10-day post-infection in HSV-1 infected corneas than uninfected corneas. Results shown are representative of three independent experiments (n=6-7 corneas or mice per group). Statistical significance was calculated using unpaired parametric Student’s t-test (** p= 0.0058).

Figure 18: Nuclear localization of hypoxia inducible factor-2α (HIF-2α) in the corneal epithelium of HSV-1 infected corneas during clinical disease period. Representative immunofluorescence images of frozen corneal sections showing HIF-2α staining obtained from the naïve (top panel), during pre-clinical period (middle panel, 4-day post ocular infection) and clinical phase (bottom panel, 10-day post ocular infection) of HSK. The top and bottom panel from left to right, showing nuclear staining with DAPI (in red), HIF-2α staining (in green), and the merged images with nuclear localization of HIF-2α (in yellow). Arrowheads in the bottom row indicate the epithelial cells with nuclear localization of HIF-2α in yellow. Magnification X400. Immunofluorescence images shown are representative of five corneas at each time-point.

Figure 19: Stabilization of Hypoxia Inducible factor-1α and its differential expression on Immune cell subsets in HSV-1 infected corneas. A. Representative pseudo color FACS plots of naïve cornea denoting the gating strategy employed for CD45- leukocytes and CD45+ leukocytes. Representative histogram FACS plot (right) to show isotype control and the level of intracellular HIF-1α protein in CD45- and CD45+ populations, respectively. B. Representative pseudo color FACS plots of an infected cornea at day-13 post ocular infection, denoting the gating strategy employed for CD45- leukocytes and CD45+ leukocytes, CD4 T cells, CD11b+ Ly6G+ neutrophils and CD11b+ Ly6G- myeloid cells. Representative histogram FACS plot (top to bottom) to show isotype control and the level of intracellular HIF-1α protein in CD4 T cells, neutrophils and myeloid cells, respectively in the infected cornea. Scatter plot/bar graph in the top panel demonstrates the mean fluorescence intensity (MFI) of HIF-1α protein in CD45- and CD45+ leukocytes. Scatter plot/bar graph in the bottom panel demonstrates the MFI of HIF-1α protein in three different immune cell populations in individual infected corneas at day-13 post ocular infection. Data derived is shown from two similar experiments with n=5 per group. Statistical significance was calculated used unpaired non-parametric t-test and one-way ANOVA test. **** p≤0.0001.

Figure 20: Acriflavine treatment aggravates corneal opacity, but significantly reduces hem- angiogenesis in corneas with Herpes Stromal Keratitis. A. Scatter plot (on the left) is showing the corneal opacity score of individual corneas from control and acriflavine treated groups of HSV-1 infected mice at 10-day post-infection. B. Bar graph (right) is showing the incidence of eyes with opacity score >3.0 between both groups of infected mice. C. Scatter plot is showing hem-angiogenesis score of individual corneas from control and acriflavine treated groups of infected mice at 10-day post- infection. D. Scatter plot/bar graph is showing the expression of Pfkfb3 gene in individual corneas from control and acriflavine treated groups of infected mice at 10-day post-infection. Statistical significance was calculated using nonparametric Mann-Whitney U test. (n=5-6 HSV-1 infected mice per group in each experiment).

Figure 21: Acriflavine treatment significantly reduces stabilization of HIF isoforms (HIF-1/2α) in the immune cell subsets and nonimmune cells respectively in HSV-1 infected corneas. A. Representative pseudo color FACS plots demonstrating gating strategy to show singlets gated cell population using the forward scatter height vs area. Furthermore, live gated the cells using side and forward scatter to distinguish CD11b+ CD45+ Ly6G-, CD11b+ CD45+ Ly6G+ and CD11b- CD45+ CD4+ cell populations. Representative histogram FACS plots (top to bottom) denoting analysis of Neutrophils (CD11b+ Ly6G+) and CD4 T cells to determine the mean fluorescence intensity (MFI) of HIF-1α in the vehicle (pink peak) vs acriflavine (ACF) treated (green peak) of mouse corneas at 12-day post ocular infection. Scatter plot/bar graph (bottom panel) denoting HIF- 1α MFI in all the two histogram analyzed cell populations at 12-day post-ocular infection. Data derived is shown from two similar experiments with n=5 per group. p values were calculated using unpaired t-test. * p<0.05. B. Representative immunofluorescence images of frozen corneal sections show HIF-2α staining obtained from the vehicle (top panel), ACF treated (bottom panel), at 12-day post ocular infection) of HSK. The top and bottom panel from left to right, show nuclear staining with DAPI (in red), HIF-2α staining (in green), and the merged images with nuclear localization of HIF-2α (in yellow). Arrowheads in the top row indicate the epithelial cells with nuclear localization of HIF-2α in yellow. Magnification X400. Immunofluorescence images shown are representative of five corneas at each time-point. Data derived is shown from two similar experiments with n=5 per group. p values were calculated using unpaired t-test. * p≤0.05.

Figure 22: Acriflavine treatment significantly reduces the influx of non-granulocytic myeloid cells and CD4 T cells in HSV-1 infected corneas. Representative pseudo color FACS plots demonstrating gating strategy to show singlets gated cell population using the forward scatter height vs area. Furthermore, live gated the cells using side and forward scatter to distinguish CD11b+ CD45+ Ly6G- (non-granulocytic myeloid cells), CD11b+ CD45+ Ly6G+ (Neutrophils) and CD11b- CD45+ CD4+ cell populations in vehicle vs Acriflavine (ACF) cornea treated samples. Scatter plot/bar graph (bottom panel) denoting percentages of all three cell populations analyzed in vehicle vs ACF treated cornea samples at 12-day post-ocular infection. Data derived is shown from two similar experiments with n=5 per group. p values were calculated using unpaired t-test.* p≤0.05, **p≤0.01, ***p≤0.001.

Figure 23: A proposed model to demonstrate the development of inflammatory hypoxia in HSV-1 infected corneas during the clinical disease period. This model shows the existence of both, oxygenated and hypoxic regions in the HSV-1 infected cornea. In the prevalence of hem-angiogenesis, as one of the hallmarks of HSK disease, molecular oxygen will be likely supplied by these neo-blood vessels formed in the HSV-1 infected cornea. However, in presence of massive influx of neutrophils, the molecular oxygen may reduce into reactive oxygen species (ROS). Eventually, this process can shape the development of inflammatory hypoxia.

CHAPTER 2: EVALUATION OF EXTRA-ORBITAL LACRIMAL GLAND AND CONJUNCTIVAL INFLAMMATION DURING HSV-1 INFECTION

Abstract

Corneal innervation, main lacrimal gland, and ocular surface components including cornea, conjunctiva and meibomian gland act as a ‘functional unit’ that control tear production via the neural reflex arc. Sensory corneal nerves are part of the neural reflex arc, which regulates tear fluid production from the lacrimal gland. The loss of sensory corneal nerves is reported in both human and mouse herpes stromal keratitis

(HSK). Chronic inflammation of the functional unit or one of its components may lead to dysfunction in tear production. Our study showed a significant reduction in tear volume in HSV-1 infected corneas of C57BL/6 (B6) mice during the clinical disease period. In this study, we demonstrated that corneal HSV-1 infection in the B6 mouse model would cause reduced tear levels in correlation to extra-orbital lacrimal gland (EoLG) and conjunctival inflammation. Our study also ascertained the outcomes of decreasing tear levels and their effect on lacrimal gland inflammation. Our results showed a massive influx of CD45+ leukocytes involving CD4 and CD8 T cells and morphological changes in inflamed EOLG tissues on day 10 post-infection. Most notably, the existence of the replicating HSV-1 virus in the EoLG was detected on day 5 post-infection. Topical administration of lacritin in HSV-1 infected mice did increase the tear volume at day 8 post-infection but was not able to sustain the increase in the tear volume throughout the

HSK disease phase. However, no change was observed neither in the influx of leukocytes nor in the HSK disease severity (corneal opacity and hem-angiogenesis) in

83 the lacritin treated group of HSV-1 infected mice. The conjunctival tissue was also detected with an increased influx of leukocytes involving CD4 and CD8 T cells.

Interestingly, our results also showed a significantly reduced number of mucin contributing goblet cells in the conjunctiva at day 10 post-infection, compared to the uninfected(naïve) conjunctiva. Of note, these results correlated with the reduced tear volumes in the HSV-1 infected mice at day 10 post-infection. Taken together, our study suggested an increased leukocytes influx in EoLG and conjunctiva. Although lacritin treatment alone delayed the decrease in tear volume in infected eyes, it was not sufficient to inhibit leukocytes influx in EoLG of the HSV-1 infected mice. Additionally, conjunctival tissue likely plays an important role in reducing tear volume after ocular

HSV-1 infection.

Introduction

HSK is an immunoinflammatory condition which persists in inflamed cornea even after the clearance of infectious HSV-1 virus [234]. Studies from the mouse model have reported the role of neutrophils and T cells in the pathogenesis of HSK [30,32]. The hallmark of HSK in the mouse model involves the development of corneal opacity and angiogenesis (hem- and lymphangiogenesis) [235,236]. Our data (Figure 24) and recent studies in a mouse model of corneal HSV-1 infection have reported, the loss of sensory corneal nerves during the progression of HSK [237,80]. The afferent sensory corneal nerves are part of the neural reflex arc, which regulates the secretion of tear fluid on the ocular surface [99,238].

The tear film is a thin transparent layer coating the surface of the eye [239].

There are three distinct layers in the tear film, comprising of the top lipid layer, middle

84 aqueous and the inner mucin coat. The lipid layer is released from the meibomian glands and harderian gland (in mice) to prevent the tear evaporation from the ocular surface [240]. The aqueous layer is secreted by the lacrimal gland and helps in providing nutrients and oxygen to the corneal tissue. Lacrimal gland plays an important role in regulating the quantity and quality of tears overlying the corneal surface [99,124].

The goblet cells of the conjunctival epithelium secretes mucin contributing to the mucin layer [241].

Additionally, the differentiated squamous cells with microvilli and microplicae, in the superficial layers of the corneal epithelium, secretes glycocalyx and maintains the tear film integrity [239,242]. Most of the tear film constitutes of the aqueous and mucin layer [241]. Tear production is highly regulated by neural reflex arc, that consists of parasympathetic and sympathetic innervations in the lacrimal gland that is activated by the afferent sensory nerves in the cornea. The activation of neural reflex arc stimulates the release of neurotransmitters from nerves innervating lacrimal gland and helps in aqueous secretion onto the tear film overlying the ocular surface [243,244]. Acinar cells are the main secretory cells of the lacrimal gland and surrounding these cells are the myoepithelial cells [99]. EoLG is the prominent lacrimal gland in mice and rats [119].

Here, our study highlights on EoLG in the HSV-1 infected B6 mice. Overall, an intact tear film is required to lubricate, supply the nutrients and helps in the diffusion of atmospheric oxygen to the corneal epithelium.

It is well known that reduced tear level secretion leads to the irregularity in the corneal surface due to the decreased lubrication of the corneal surface. Consequently, it elevates the risk of damage to the corneal epithelium causing ‘dry eye’ like conditions

[99]. This condition can be correlated to the persistent inflammatory conditions, subsequently leading to the destruction of the EoLG [124]. Therefore, it is essential to understand the histopathology of the EoLG in the B6 mice with HSV-1 infection.

Histologically, a normal EoLG consists of lobules that have multiple acini lined by columnar secretory cells which play a major role in aqueous production (a prominent layer in the tear film) released on to the ocular surface [245]. Under pathological conditions EoLG is prone to undergo dysfunction with the infiltration of immune cells

(CD45+ /CD4+T cells) [124]. This causes a significant reduction in the exocrine function of the EoLG resulting in decreased tear production and ocular lubrication [243]. Herein, our study shows that chronic immuno-inflammatory disease caused by HSV-1 infection in cornea will result in inflammation of the EoLG. This leads to an alteration in tear production which will cause a direct impact on maintaining a moist ocular surface.

Topical administration of lacritin, an approach to increase tear secretion, transiently increased the tear volume in the infected eyes during the early clinical disease phase of

HSK. Also, the treatment had a moderate, but not statistically significant, the effect in reducing the severity of HSK.

Conjunctiva is a thin semitransparent mucous membrane that covers the posterior surface of the eyelids and further extends into the limbus of cornea [246,129].

Conjunctival epithelium consists of the mucin-secreting goblet cells. Parasympathetic and sympathetic innervation in conjunctiva is stimulated by corneal sensory nerves to stimulate mucin secretion from goblet cells [135]. Conjunctival goblet cells have a key role in mucin production, while its responses during HSV-1 ocular infection remain unexplored. Infiltration of innate immune cells induces inflammatory responses in the

86 conjunctiva. However, during HSV-1 infection in the cornea, there exists a paucity of knowledge with the occurrence of conjunctival inflammation and its contribution in maintaining the stable tear volume over the ocular surface. Our study suggests that

HSV-1 infection in B6 mouse cornea during clinical disease phase of HSK induces conjunctival inflammation which results in the loss of goblet cells. Interestingly, reduced tear volume correlates with the loss of goblet cells in the conjunctiva of HSV-1 infected eyes. Altogether, the purpose of this study was to evaluate inflammation in extra-orbital lacrimal gland and conjunctiva in the HSV-1 infected B6 mice.

Results

Loss of corneal nerves in HSV-1 infected eyes of C57BL/6 mice

The cornea is an avascular and highly innervated tissue. Corneal nerve damage associated with the loss of sensation has been reported previously in HSV-1 infected corneas [80]. The existence of corneal nerves plays an essential role in maintaining an intact ocular surface by interconnecting with the neural reflex arc [247]. The cross-talk between corneal nerves and the neural reflex arc is crucial in tear production [92,190].

Therefore, to elucidate the corneal innervation in HSV-1 infected eyes the cornea whole mount of the naïve and HSV-1 infected eyes were prepared for staining with the pan- neuronal marker- anti-Tuj1 antibody, (as described in materials and methods). Several reports have described corneal nerve formation, its distinct network and distribution pattern in the normal cornea [92,80]. Our results from the montage images acquired from the naïve (uninfected) cornea showed an intact network of nerves distributed from the central cornea to the peripheral (limbus) region (Figure 24). Tuji1staining (green) displayed the appearance of distinct thick nerve trunks in the limbal stromal region

(Figure 24 A). These nerves further ramified into smaller branches as they reached the paracentral region of the cornea (Figure 24 B). Prominent nerve leashes (bunches) were identified throughout the montage image of the naïve cornea. The nerve leashes further merged into the central cornea with fine branches forming a ‘central swirl’ pattern

(Figure 24 C). In contrast, the Tuj1 stained montages of HSV-1 infected cornea at day-

10 post-infection showed loss of corneal nerves, more prominently in the central cornea

(Figure 24 F) with the disappearance of distinct central swirl formation. Very faintly stained thick limbal stromal nerve trunks were observed (Figure 24 D). No nerve leashes in the paracentral region were observed (Figure 24 E) as compared to the pattern in the naïve cornea. In conclusion, HSV-1 infected cornea had no prominent staining of Tuj1 indicating the deinnervation or loss of corneal nerves that may likely pose an impact on tear production.

Reduced volume of tears in HSV-1 infected eyes during the clinical phase of HSK

In mouse model, HSK is divided into pre-clinical (day -0 to -7 post-infection) and clinical (day -7 to -18 post-infection) disease phase [31]. During the clinical phase, the development of corneal opacity and angiogenesis is detected [235,237]. A study from our lab (Figure 24) and recent studies, in a mouse model, have shown the loss of sensory corneal nerves during the clinical phase of HSK [237,80]. The sensory corneal nerves are part of the neural reflex arc, which regulates tear secretion from the lacrimal gland [238]. To measure the tear volume in HSV-1 infected eyes, phenol red thread

(PRT) test was performed on infected eyes from C57BL/6 mice at different day post- infection, as described in Materials and Methods. PRT test was also carried out on eyes with no ocular HSV1 infection (naïve). Representative phenol red thread images from

88 naïve and day 10 post-infection are shown with the color change (yellow to red) on the thread tips after the PRT (Figure 25 A). As shown in (Figure 25 A), an average of about

50% decrease in tear volume was measured in HSV-1 infected eyes on day 10, when compared with an uninfected naïve group of mice. This experiment was carried out by tracking the tear level in the same eye before and after corneal HSV-1 infection (Figure

25 A). Furthermore, the eyes with severe HSK (corneal opacity ≥3.0) were noted to have a reduced level of tears than the eyes with mild HSK (corneal opacity ≤1.0) (Figure

25 A) Together, our results showed a significantly reduced tear volume in infected eyes during the clinical phase of HSK compared to the uninfected eyes.

Histopathology of EoLG showing atrophy of acinar cells during clinical disease phase of HSK EoLG is the main contributor to the aqueous layer, which is the most vital layer of the tear film [241]. Acinar cells in EoLG play a major role in tear secretion [99].

Therefore, impaired function in the acinar cells may cause a severe deficiency in tear production. A recent study has highlighted the reduced tear volume and ‘destruction’ of acinar cells in the EoLG of Sjorgen’s syndrome mice model [248]. Hence, we hypothesized that acinar cell dysfunction might likely be the primary cause for the reduced tear volumes in the HSV-1 infected B6 mice. To address this, histopathology in

EoLG was evaluated by Hematoxylin and Eosin (H&E) staining. EoLG were excised from the naïve, at day-4 and -10 post-HSV-1 infection. The tissues were immediately fixed using Excalibur’s fixative solution (as described in Materials and Methods). The paraffin-embedded sections were later processed for H&E staining. Our results showed acinar cell atrophy (circled areas in Figure 25 C) that was prominently observed in the

EoLG that is supplying tear to the eyes with severe HSK at day 10 post-infection

89 compared to the uninfected control (naïve). Indeed, the EoLG tissue at day 10 post- infection failed to preserve the normal architecture as in control. On the other hand, the

EoLG which is supplying tears to the eyes with mild HSK showed infiltration of fewer inflammatory cells in the interstitium of the EoLG (Figure 25 C). H&E staining in the naïve showed compactly arranged acini in the lobules. A similar observation was noted in the EoLG at day 4 post-infection (Figure 25 C). Eye images depicting HSK disease severity and phenol red thread test images are shown in the (Figure 25 C) which correlates to the histopathology of EoLG. Altogether, our data suggest degeneration of acinar cells in the EoLG, which correlates with reduced tear volume in the eyes of HSV-

1 infected mice during the clinical phase of HSK.

Increased influx of leukocytes in EoLG of HSV-1 infected eyes during the clinical phase of HSK EoLG play an important role in regulating the quantity and quality of aqueous component of the tear fluid in mouse [249]. In several pathological instances, EoLG is excessively infiltrated with immune cells, which may cause damage to acinar cells resulting in an insufficient tear production [250,251]. To determine whether corneal

HSV-1 infection promotes leukocyte influx in EoLG, HSV-1 infected C57BL/6 mice were euthanized on the day -4 and -10 post-infection and EoLG was collected to perform flow cytometry. As is shown in our results (Figure 25 B), an average of about 3-fold increase in the frequency and an absolute number of CD45+ cells were seen in EoLG on day-10 in comparison to day-4 post-infection. Among CD45+ cells, more than 5-fold increase in the average number of CD4 and CD8 T cells was determined in EoLG on day-10 than day-4 post-infection. Our results showed an increased influx of T cells in EoLG of HSV-

1 infected mice during the clinical phase of HSK.

Detection of Replicating virus in the EoLG tissue of HSV-1 infected B6 mice during the pre-clinical disease phase of HSK The replicating HSV-1 virus exists in cornea until day 5 to day 6 post-infection, after which clearance of the replicating virus occurs [203]. EoLG is innervated with both parasympathetic and sympathetic nerves. One of the ophthalmic branches of the trigeminal nerve (V1) sub-branches into the lacrimal nerve that enters the lacrimal gland. It is noteworthy to mention that the ophthalmic branch is interconnected with the sensory nerves of the cornea and the lacrimal nerve in the EOLG [252]. Therefore, we hypothesized that the replicating HSV-1 virus during the pre-clinical phase of HSK disease could possibly re-route towards the V1 branch and enter EoLG via the lacrimal nerve. Eventually, the residing infectious virus in the EoLG can evoke inflammation and cause tissue damage. Our results so far have indicated excessive tissue damage with the significantly increased influx of leukocytes in the EoLG of HSV-1 infected mice. To further elucidate the existence of the infectious virus, the excised EoLG from the day 5 and day 10 post- HSV-1 infected mice were minced in one/two small pieces and co- cultured with the Vero cell monolayer tissue culture flask (Figure 26 A). After the 48 hours of incubation, these Vero cell cultures were examined for phenotypic changes.

Notably, EoLG at day 5 post-infection showed predominant changes in the phenotype of

Vero cells which were the typical characteristics of virus infection. As shown in (Figure

26 B) the confluent layer of Vero cells starts getting peeled off the adherent surface and the rounding of cells occur. In contrast, EoLG at day-10 post-infection neither showed any symptoms of virus infection nor depicted any of the above-mentioned phenotypic changes in the Vero cell monolayer (Figure 26 C). The control showed the normal and healthy growth of Vero cell monolayer adherent to the surface (Figure 26 D).

To further quantitate the viral load in EoLG of HSV-1 infected mice during the pre-clinical disease phase, we carried out a viral titration assay. This assay was used to quantify the number of plaque forming units (PFU) from the individual EoLG of naïve(uninfected) and HSV-1 infected mice at day-5 post-infection. A significant increase in the PFU at day 5-post-infection was observed compared to the naïve

(uninfected control) (Figure 26 E). Of note, these results correlated with our previous result that showed phenotypic changes in the Vero cell monolayer co-cultured with day

5 post-infection EoLG. Altogether, our findings indicate the existence of the replicating virus in the EoLG of HSV-1 infected mice.

Lacritin treatment delayed but did not prevent the reduction in tear volume nor reduced the severity of HSK in HSV-1 infected C57BL/6 mice Lacritin is a tear glycoprotein, which has been shown to increase tear production in aqueous-deficient dry eye disease [253]. To determine if lacritin treatment could increase the tear level in HSV-1 infected eyes, topical administration of lacritin and its truncated control peptide (C-25) was carried out on HSV-1 infected eyes, as described in the Materials and Methods. A brief outline of the lacritin treatment conducted is shown in (Figure 27 A). HSV-1 infected eyes, receiving topical lacritin treatment, showed a significantly higher level of tears in comparison to truncated C-25 treated control group of infected mice on day 8 post-infection. However, the effect of lacritin attenuated with the progress of time, and on day 10 post-infection no significant difference in the level of tears was measured between treated and control groups of infected mice (Figure 27 A). No significant difference in the frequency and number of leukocytes infiltrating EoLG of infected mice was detected on day 10 post-infection when compared between both groups of mice (Figure 27 B). Although a reduced

92 corneal opacity was observed in lacritin treated than control C-25 treated HSV-1 infected eyes on day 10 post-infection, the data was not statistically significant.

Furthermore, the lacritin treatment alone was not effective in preventing the loss of corneal epithelium smoothness observed in HSV-1 infected eyes during the clinical disease period (Figure 27 C). Together, our results showed that Lacritin, but not C-25, treatment effectively delayed the decrease in tear volume in infected eyes. Lacritin treatment given alone to HSV-1 infected eyes had a moderate, but not statistically significant effect on reducing the severity of HSK.

Loss of goblet cells in the conjunctiva of HSV-1 infected eyes during the clinical phase of HSK Conjunctival epithelium has specialized cells secreting mucins namely, the goblet cells. Mucins are the vital component of the tear film and hence crucial in maintaining the tear film overlying the corneal epithelium [254]. Reports have implied that dry eye conditions lead to a reduced number of goblet cells in the conjunctiva of mouse models

[255]. In the agreement, our data showed a significant reduction in the tear volume in the eyes of HSV-1 infected mice (Figure 28 A) that potentially may cause dry eye conditions. Concerning this result, it was important to evaluate the mucin-secreting goblet cells in the conjunctiva of the HSV-1 infected eyes. Mucin secretion in conjunctival goblet cells was evaluated by periodic acid-Schiff (PAS) staining. Adnexa was excised from the naïve and at day-4, -10 post-HSV-1 infections. The tissues were immediately fixed using Excalibur’s fixative solution (as described in materials and methods). The paraffin-embedded sections were later processed for PAS staining. Our results showed distinct staining of PAS+ goblet cells in the naïve and at day 4 post- infection lined along the conjunctival epithelium (as indicated by the arrows, Figure 28

A) at the fornix and superior to the fornix region. On the contrary, loss of goblet cells was observed in the conjunctival epithelium of HSV-1 infected eyes at day 10 post- infection. Intriguingly, the loss of goblet cells correlated with the reduced tear volumes corresponding to the HSK disease severity (Figure 28 A).

To further enumerate the number of goblet cells, quantification of the goblet cells in the conjunctiva of naïve, at day-4, and -10 post-infection was performed using the

Image J software. The PAS+ goblet cells were counted per microscopic field

(0.01mm2). A significant reduction in the number of PAS+ goblet cells was observed in the conjunctiva at day-10 post-infection compared to the naïve or at day-4 post-infection

(Figure 28 B). Approximately 75% of the goblet cells were lost in the conjunctiva at day-

10 post-infection. Our results confirm the loss of goblet cells in the conjunctiva of HSV-

1 infected eyes that may cause dysfunction in the mucin secretion and in the maintenance of tear film integrity.

Increased influx of leukocytes in the conjunctiva of HSV-1 infected eyes during the clinical phase of HSK As mentioned earlier, conjunctiva has a prominent role in maintaining a moist and lubricated ocular surface by secretion of mucins. However, during inflammation dysfunction in the mucin secretion may occur causing dry eye conditions and damage to the ocular surface. Reports have indicated an increased infiltration of T cells in the conjunctiva during dry eye conditions [256]. Since our results confirmed the loss of goblet cells, we were further interested in characterizing the influx of leukocytes in the conjunctiva of HSV-1 infected eyes by using flow cytometry. The excised adnexa from the naïve, at day -4, and -10 post-infection were collected, and bulbar conjunctiva was dissected. The pooled conjunctiva samples were then processed into single cell

94 suspensions and stained with cell surface antibodies as mentioned in the materials and methods. As is shown in our results (Figure 29), an average of about 3-fold increase in the frequency of CD4 T cells and an average of 1-fold increase in CD8 T cells were observed in the conjunctiva on day 10 in comparison to day 4 post-infection. Our results demonstrated an increased influx of T cells in the conjunctiva of HSV-1 infected mice during the clinical phase of HSK.

Discussion

Our findings in this study showed reduced tear levels in HSV-1 infected eyes in association with inflammation in EoLG and conjunctiva. The extent of inflammation correlated with the severity of HSK, and the reduced volume of tears in HSV-1 infected eyes. We also noted the atrophy of acinar cells in the EoLG of infected mice with severe

HSK. Our approach to increasing the tear volume, while using topical lacritin treatment, transiently increased the tear fluid level in HSV-1 infected eyes. Also, lacritin treatment partially but not significantly reduced the extent of hypoxia, and the severity of HSK.

Recent reports [237,80] and our results (Figure 24) indicated an extensive loss of sensory corneal nerves in HSV-1 infected eyes during the clinical phase (day 10 post- infection) of HSK. The bulk of aqueous tear fluid production is contributed by the lacrimal gland and is tightly regulated by the neural reflex arc (NRA) [99]. The NRA includes afferent sensory corneal nerves and parasympathetic and sympathetic innervations of the lacrimal gland [257]. Inputs and outputs from the sensory corneal nerves are reported to control the tear fluid secretion from the lacrimal gland [238].

Since sensory corneal nerves are involved in regulating tear production from the lacrimal gland, their loss in HSV-1 infected eyes will likely affect reducing the tear

95 secretion from the lacrimal gland, as shown in our results. In latent stromal herpetic keratitis patients, the steady-state turnover of tears is also significantly reduced and is considered to be the outcome of impaired lacrimation [258]. In aqueous deficient dry eye condition, lacrimal gland (LG) inflammation is associated with the reduced aqueous production of tears [124]. LG inflammation may also cause the atrophy of acinar cells.

Our results showed a massive influx of CD4 and CD8 T cells, and the atrophy of acinar cells in EoLG of infected mice with severe HSK. Although the cause of increased infiltration of leukocytes in EoLG is not clear, the loss of sensory corneal nerves in HSK developing eyes may play a role in lacrimal gland inflammation. Sensory corneal nerves regulate neurotransmitter, such as acetylcholine and norepinephrine, secretion from the parasympathetic and sympathetic innervations in EoLG. It is possible that degeneration of sensory nerves in corneas during the clinical phase of HSK may cause a decrease in the release of neurotransmitters from the nerves in EoLG, and this may result in an excessive influx of T cells in lacrimal gland [99]. Pharmacological blockade of acetylcholine receptor has been shown to cause inflammation and lymphocytic infiltration in EoLG of B6 mice [259]. Interestingly, during inflammation reports have suggested the occurrence of fibrosis [260]. The process of fibrosis involves a massive deposition of extracellular cellular matrix components. In inflamed lacrimal gland, acinar cells were reported to be replaced by the ‘hyaline fibrous tissue’ [261]. In Sjogren's syndrome, the lacrimal gland was detected with ‘intralobular fibrosis’ associated with increased deposition of fat [252]. Therefore, we hypothesize that EoLG in HSV-1 infected mice with severe HSK may be undergoing the process of fibrosis in association with the degeneration of acinar cells. As our result showed (Figure 25 B), a distorted

96 architecture of the EoLG, likely indicates the replacement of acinar cells with massive deposition of fibrous tissue.

Lacritin is a tear glycoprotein which has recently been shown to promote the basal tearing in normal rabbits [262]. Besides, the topical application of lacritin effectively increased the tear secretion in aqueous-deficient dry eye disease [253]. In

HSV-1 infected eyes, lacritin treatment transiently increased the tear volume, but the effect was attenuated by day 10 post-infection. This suggests that lacritin treatment alone is not highly effective in controlling HSV-1 induced inflammation of the EoLG. In

HSK, the corneal stroma is heavily infiltrated with immune cells [263], and the latter causes stromal inflammation. In response to chronic stromal inflammation, the corneal epithelial cells may adopt an aberrant cell fate [264]. Therefore, this may have an impact on maintaining an intact tear film. The formation of an intact tear film on the corneal surface is important to preserve the corneal epithelium homeostasis. Although, lacritin is suggested to promote the corneal epithelium homeostasis [265]. lacritin treatment in association with an approach to control stromal inflammation should be more effective in reducing the severity of HSK.

Conjunctival goblet cell secretion is interconnected with the neural reflex arc. The stimulation from the reflex arc initiates goblet cells to secrete mucins which are largely the membrane-bound type of mucins. Muc16 is transmembrane mucin, while Muc5a/5b are the large secretory mucins expressed by the goblet cells in mouse conjunctiva

[266]. Mucins in the tear film play a major role in maintaining tear film integrity and a lubricated ocular surface [267]. However, in dry eye syndrome, reduced mucin secretion has been reported in correlation to the reduced number of goblet cells [242]. Our study

97 demonstrated a loss of corneal sensory nerves followed by the reduced tear volumes in the HSV-1 infected eyes. In our study, most likely due to prevalent dry eye conditions, loss of goblet cells in the conjunctiva of HSV-1 infected eyes were observed. This provides us the clues on dysfunction in mucin secretion that may directly impact the integrity of the tear film and the ocular surface. Reports have indicated a significant loss of mucin-secreting goblet cells during inflammation and in aqueous-deficient dry eyes

[268]. In mouse models, IFN-γ is known to promote loss of goblet cells [269]. Studies have indicated that dry eye conditions augment the expression of IFN-γ, which indeed causes loss of goblet cells [268]. IFN-γ is produced by CD8 T cells Th1 lymphocytes and other cell types [270]. Increased infiltration of T cells in the conjunctiva during dry eye conditions has been reported [256]. A significant increase in the influx of T cells in the conjunctiva of HSV-1 infected eyes was observed. Although the cause for loss of goblet cells is unclear in the conjunctiva of HSV-1 infected eyes, it is likely possible that the influx of T cells (CD4/CD8) are the culprits involved in causing degeneration of conjunctival goblet cells. This provokes a vicious cycle of events initiated by a dry ocular surface that leads to an increased influx of IFN-γ producing T lymphocytes, subsequently causing loss of conjunctival goblet cells. Of note, ‘Squamous metaplasia’ is a pathological phenomenon detected during the loss of goblet cells and impacts the health of ocular surface [271,272]. In the presence of dry eye conditions and existing chronic inflammation in the HSV-1 infected mice, squamous metaplasia may likely hamper the integrity of the tear film and the corneal epithelium causing loss of vision.

In summary, this study broadens our knowledge to further explore the cause for acinar cell degeneration and the loss of conjunctival goblet cells during lacrimal and conjunctival inflammation respectively, in HSV-1 infected mice.

Figure 24: Deinnervation in HSV-1 infected cornea. Representative Confocal images of Tuj-1 stained corneas. Corneal whole mount staining was performed for determining the existence of corneal nerves by using Tuj 1- neuronal marker. Montage images were acquired at 200x magnification. The enlarged images of each region from montage are shown in the insets respectively. In the Naïve cornea an intact nerve pattern was observed in the limbus showing thick limbal stromal nerve trunks (A); in the paracentral region showing nerve leashes (bunches) running in an orientation parallel to each other throughout the whole corneal mount (B). The nerve bunches ramified into fine branches and merged into the central corneal region (C) as displayed in the montages. In day 10 HSV-1 infected cornea, loss of corneal nerves was observed. There are was no intact distribution of nerves compared to the naïve cornea. Very faintly stained stromal nerve trunks were observed (D); no nerve bunches displayed in the para-central region (E); loss of corneal nerves in the central region (F). TUJ1 (green). Immunofluorescence images shown are representative of three corneas at each time point.

100

Figure 25: Significant reduction of the tear volume correlating with inflammation in Extra-orbital lacrimal gland in HSV-1 infected corneas during the clinical phase of HSK. A. Representative Images of phenol red threads used for measuring the tear volume in the eyes of naïve and at day 10 post-infection. The color change in threads from yellow to red depicts the tear volume. Scatter plots designated with colored dots represent basal tear volume in the individual naïve mouse eye followed by day 7 and day 10 POI (post-ocular infection). Each colored dot in the scatter plot corresponds to the tear volume (basal, day7 and day10 POI) of individual mouse eyes. The Bar graph on the far left is showing a significant reduction in the measurement of the tear volume in eyes with severe HSK compared to the mild HSK. Figure legend continues

101

Figure 25: Legend continues

B. Representative hematoxylin and eosin (H&E) staining in the paraffin sections of EOLG excised from the naïve, HSV-1 infected mice at day 4 post-infection and at day 10 post-infection with mild HSK (top) and severe HSK (bottom). Eye images depict the severity of corneal opacity correlating to the phenol red thread images with the measurement of reduced tear volume in severe HSK eyes. Circled areas in EOLG of infected mice with severe HSK demonstrate the atrophy of acinar cells. C. Representative FACS plots denote the frequency of CD45+ cells (top panel) and CD4/CD8 T cells (bottom panel) in EOLG of HSV-1 infected mice on 4 and 10-day post-infection. Scatter plots demonstrate a significant increase in the frequency and an absolute number of total leukocytes (top panel) and CD4/CD8 T cells (Bottom panel) in EOLG of infected mice on day10 than day 4 post-infection. Results shown are representative of two similar experiments. (n=5 to 8 eyes per group). Statistical analysis was performed by using unpaired two- tailed Student’s t-test and one-way ANOVA, (*p≤0.05, **p≤0.01, ****p≤0.0001, nsp>0.05).

102

Figure 26: Detection of Infectious virus in the Extra-orbital lacrimal gland of HSV-1 infected B6 mice at the preclinical disease phase of HSK. A. Representative image of the tissue culture flask with Vero cell monolayer, co-cultured with excised EoLG (minced into two tiny pieces) from HSV-1 infected mice at day-5 or day-10 HSV-1 post-infection. B. Representative image of the monolayer layer of Vero cells after 48hr co-culture with EoLG tissue excised from HSV-1 infected mice at day 5 post-infection. Phenotypic changes (rounding off cells and peeling off the adherent surface) in the Vero cells can be observed prominently. Image magnification at 10x taken to view prominent changes in the Vero cells in a larger area of the monolayer. C. Representative image of the monolayer layer of Vero cells after 48hr co- culture with EoLG tissue excised from HSV-1 infected mice at day 10 post-infection. No phenotypic changes were observed in the monolayer of Vero cells. D. Image is showing the normal Vero cell monolayer as a control. D. Viral titration assay was carried to quantify the number of the plaque forming units (PFU) from the individual EolG of naïve(uninfected) and HSV-1 infected mice at day-5 post- infection. A significant increase in the PFU at day 5-post-infection was observed. (n=5 for each time point post-infection). Statistical analysis was performed using unpaired two-tailed Student’s t-test and one-way ANOVA; The p values were calculated using unpaired two-tailed Student’s t-test (*p≤0.05).

103

Figure 27: Topical administration of Lacritin transiently increases the tear volume and partially reduces the severity of HSK in HSV-1 infected corneas. A. Schematic of topical lacritin/C-25 treatment regimen started at day 4 through day 10 post-infection. Images of Phenol Red Thread (PRT) test from C-25 control and Lacritin treated group of HSV-1 infected mice. The PRT test was performed to test the basal tears (before HSV-1 infection) and at day -6, -8, and -10 post-infection. Scatter plots represents the tear volume measured in C-25 (control) and lacritin treated groups of HSV-1 infected mice. Significant increase in the tear volume was observed at day 8 post-infection. B. Representative FACS plots denote the frequency of CD45+ cells (left panel) and CD4/CD8 T cells (right panel) in EOLG of C-25 (top panel) and lacritin (bottom panel) at day 10 post-infection. Scatter plots show no significant difference in the frequency and absolute number of CD45 (top panel) and CD4/CD8 T cells (bottom panel) in EOLG derived from both treated groups of mice. C. Scatter plot shows the corneal opacity score and angiogenesis score of individual mouse eye between both groups of mice at day-6, - 10 post-infection. Scatter plot shows the corneal smoothness scores of individual mouse eye between both groups of mice at 6,8 and 10 –day post-infection. Data shown are representative of two independent experiments. (n=5 to 8 eyes per group). The p values were calculated using unpaired two- tailed Student’s t test (*p≤0.05, nsp>0.05).

104

FIG 28: Loss of goblet cells in the conjunctiva of HSV-1 infected cornea during the clinical phase of HSK. A. Infected eyes with ocular adnexa were surgically excised from the naïve C57BL/6 mice and on day 10 post-infection. The paraffin sections of eyes with surrounding eyelids were stained with periodic acid Schiff's (PAS) for the identification of goblet cells. Sections were deparaffinized before staining. Arrows indicate the regions in the conjunctiva (superior to fornix) with PAS-stained -goblet cells in the naïve (uninfected), at day-4 and -10 post-infection. Eye images and the thread test corresponds to the respective adnexa. Bar graphs showing the quantification of goblet cells in 0.1mm2 performed using Image J software. B. A significant decrease was observed in the goblet cell number of the conjunctiva at day 10 post-infection. (n=4 to 6 eyes per group). The p values were calculated using one-way ANOVA ns (*** p≤0.001, p>0.05).

105

Figure 29: Increased infiltration of leukocytes in the conjunctiva of HSV-1 infected cornea during clinical phase of HSK. Representative FACS plots denote the frequency of CD45+ cells (top panel) and CD4/CD8 T cells (bottom panel) in the conjunctiva of HSV-1 infected mice on 4 and 10-day post-infection. B. Scatter plots show frequency and absolute numbers of total leukocytes (top panel) with a significant increase in the frequency and absolute numbers of CD4/CD8 T cells (bottom panel) in the conjunctiva of HSV-1 infected mice on day 10 than day 4 post-infection (bottom panel). Data shown are representative of two independent experiments. (n=5 to 8 eyes per group). The p values ns were calculated using unpaired two-tailed Student’s t test (*** p≤0.001, *p≤0.05 , p>0.05).

106

CHAPTER 3: MORPHOLOGICAL ALTERATIONS IN THE CORNEAL EPITHELIUM DURING OCULAR HSV-1 INFECTION Abstract

Corneal epithelium consists of non-keratinizing stratified squamous epithelium which plays an essential role in maintaining a transparent cornea, thereby providing a clear vision. However, previous reports have shown that during HSV-1 infection aberrant corneal epithelial differentiation occurs, resulting in the loss of corneal epithelial characteristics. Here, we hypothesized that non- keratinized corneal epithelium acquires a keratinized ‘skin-like epithelium’ (epidermis) in HSV-1 infected corneas that directly hinders clear vision and leads to corneal blindness. The previous study reported from our laboratory suggested reduced tear volume in the HSV-1 infected eyes of C57BL/6 (B6) mice during the clinical disease phase. The reduced tear volume also increases desiccation in the eyes leading to ‘dry-eye’ condition. Our study showed reduced smoothness in the HSV-1 infected cornea in association to positive

Rose Bengal staining indicative of dry eye and damaged ocular surface. It is known that intact non- keratinized corneal epithelium plays a major role in maintaining a moist and conditioned ocular surface. Corneal epithelial cytokeratin-K12 is one of the key proteins that function to maintain epithelial integrity and homeostasis. Therefore, in this study, our interest was to understand the characteristic expression pattern of cytokeratins in the corneal epithelium of HSV-1 infected eyes. Interestingly, our results showed the reduced or loss of corneal epithelial specific K12 keratin expression, while the appearance of epidermis-specific K10 expression was observed as the disease progressed during the clinical phase of HSK. To our surprise, the loss of paired box protein (Pax-6) expression indicative of abnormal differentiation of ocular surface

107 lineage was also observed in the HSV-1 infected corneal epithelium during the clinical phase of HSK. Overall, this study highlights a novel finding on the trans-differentiation of non-keratinized corneal epithelium to the keratinized epidermis type, in association with the loss of Pax-6 expression during the clinical phase of HSK disease.

Introduction

The cornea is an avascular tissue, which will receive oxygen supply and nourishment from the tear film in the anterior region and aqueous humor in the posterior region [43,44]. One of the hallmarks of HSK disease in mouse cornea includes damage of corneal nerve associated with corneal epithelial defect [237]. Corneal nerve damage is associated with the loss of corneal nerve fibers extending from the corneal stroma into the epithelium [237]. Previous results from our lab also showed a similar outcome of loss corneal nerves at day 10 post-infection in HSV-1 infected B6 mouse cornea [Figure

24]. The cornea is a densely innervated tissue. The abundant nerve supply with its density and pattern of branching is important in maintaining the corneal integrity and homeostasis of corneal epithelium [80]. Corneal nerves play a predominant role in tear production and in maintaining the corneal epithelial integrity [273]. Reduced sensory corneal nerves will result in the loss of sensation and a decrease in the density of superficial corneal epithelial cells [273]. The loss of corneal nerves will reduce the ability of efficiently providing oxygen to a corneal tissue that manifests in ‘dry eye’ conditions.

The dry eye can either be due to tearing deficiency or increased tear evaporation in the absence of intact corneal epithelium [274]. Therefore, production of the tear film with its maintenance over the intact corneal epithelium is essential to keep a moist and conditioned ocular surface. To evaluate the corneal smoothness, HSV-1 infected eyes

108 were scored as described in Materials and Methods. A significant increase in the smoothness score was observed, suggesting an increased corneal surface damage

(Figure 30 B). In clinics, Rose Bengal staining has been widely used to detect ocular surface damage associated with tear film breaks or with compromised tear film integrity

[a 275]. Our results showed purple san stained areas in HSV-1 infected eyes at day 10 post -infection but not prominent in the uninfected eyes (Figure 30 A). Rapid corneal epithelial cell turnover has been reported with the increased thickness in dry eyes [276].

The data from our study showed increased proliferation detected by BrdU

(BromodeoxyUridine) (Figure 31 A). BrdU is a proliferation marker, which incorporates into replicating DNA [263]. Further, it is detected by staining with anti-BrdU antibody and is imaged using a confocal microscope as detailed in Materials and Methods.

Damage or loss of corneal nerves has been reported to show reduced levels of tears [277], but its influence on corneal epithelial cytokeratin expression has not been studied. Cytokeratin expression on the corneal epithelial cells functions to maintain the epithelial integrity, mechanical stability through their interaction with the tight junction proteins [67,278]. Hence, it is important to study the expression pattern of cytokeratin in the corneal epithelium. In our study, we investigated the abnormalities in the corneal epithelial keratin (K12) expression during HSV-1 infection. Our results showed abnormal expression or the loss of K12 with the appearance of K10 (epidermis-specific) in the corneal epithelium of HSV-1 infected eyes at day 10 post-infection (Figure 32, 33).

Studies from other groups have also shown that reduced K12 expression is detected in correlation to reduced Pax-6 expression [279]. Pax-6 is a transcription factor which plays an essential role in the development and maintenance of ocular lineage. More

109 precisely, Pax-6 is crucial in maintaining the corneal phenotype, since Pax-6+/- mice showed reduced expression of K12 [280]. Our results showed that, in association with abnormal K12 expression, loss of Pax6 expression in HSV-1 infected cornea was observed during the clinical phase of HSK (day 10, 16 post -infection) (Figure 34).

Overall, based on our previous observation of reduced tear volume in HSV-1 infected eyes, our study sheds light on alterations in the morphology of corneal epithelium. Mainly, the changes in the corneal epithelium as observed by the loss of corneal epithelial specific cytokeratin K12 and the appearance of the epidermis-specific

K10 is a novel finding in the HSV-1 infected eyes. Intriguingly, these results corroborated with the loss of Pax6 in the corneal epithelium of HSV-1 infected eyes which is crucial in eye development. Therefore, this study emphasis on n the morphology and maintenance of corneal epithelial integrity which plays a major role in both, tear secretion and its maintenance over the corneal surface.

Results

Irregular corneal smoothness and damage to the ocular surface epithelium in the HSV-1 infected eyes A smooth and conditioned corneal surface is maintained in the presence of intact underlying corneal epithelium. Based on our previous results that showed reduced tear volumes, we were curious to understand the smoothness of the corneal surface in HSV-

1 infected eyes. Corneal smoothness scoring was employed to understand the extent of corneal surface irregularity and was evaluated as described in materials and methods.

Briefly, corneal smoothness score wan as evaluated on a five-point scale depending on the distorted quarters of the reflected ring: 0- no distortion, 5- severe distortion [281].

The corneal smoothness for each eye in the uninfected (naïve), at day- 4, -8, and -10

110 post-infection were recorded. Representative eye images of the naïve, at day 4 and day

10 with the reflecting ring is shown in (Figure 30 A). Our results showed a sign significantly increased smoothness score in the eyes during the clinical phase of HSK

(day 10 post-infection) compared to the naïve with a score of zero (Figure 30 B). An average of 2-fold increase in the corneal smoothness scores was observed at day 10- post-infection compared to the naïve (Figure 30 B).

We were further curious to elucidate the irregularity in the corneal surface. This was determined by using Rose Bengal staining in the naïve and at day 4 and 10 post- infection. Rose Bengal staining is observed in a damaged and dry corneal surface.

Positive Rose Bengal staining indicates damage to corneal epithelium and is routinely used in clinics to diagnose dry eye syndrome [275]. Our results showed a prominent magenta/ dark pink punctate staining on the corneal surface in the eyes at day 10 post- infection, while faint staining in day 4 post-infection eyes, compared to no staining in the naïve (Figure 30 A). Taken together, our data suggest the prevalence of ocular surface damage indicative of dry eye conditions in HSV-1 infected eyes.

Hyperproliferation of corneal epithelium during the clinical onset of HSK

Immunofluorescence staining for BrdU was performed as described in Materials and Methods to characterize the extent of proliferation in the corneal epithelium of HSV-

1 infected eyes. As a marker for proliferation, the nuclear staining of BrdU indicates rapidly diving cells. Based on our previous observation (Figure 31 A), we hypothesized that rapid corneal epithelial cell proliferation might likely be stimulated to compensate for the ocular surface damage. Therefore, we sought to study the magnitude of proliferation in the corneal epithelium of the naïve and at day 5 and 9 post-infection. The nuclear

111 staining with BrdU was prominent in the corneal epithelium as shown in the tile scan image (Figure 31 A) compared to the corneal epithelium at day 5 post-infection. The enlarged images acquired at high magnification shows intensely stained BrdU+ cells

(Figure 31 A). No BrdU staining was detected in the naïve (uninfected control) corneal epithelium.

Furthermore, the number of BrdU+ cells were quantified using Image J software.

BrdU+ cells were counted per microscopic field (for a total of at least three different fields/sample) at 20x magnification. A significant increase of BrdU labeling (average of

20 cells/microscopic field) in the basal and suprabasal layer of corneal epithelium at day

9 post-infection was detected, compared to the day 5 post-infection which showed few

(<10 cells/microscopic field) BrdU labeled cells at the basal layer of the corneal epithelium. Nearly a 2-fold increase in BrdU labeling was observed in the corneal epithelium at day 9 post-infection compared to day 5 post-infection. No BrdU+ cells were detected in the naïve corneal epithelium (Figure 31 B). Thus, we infer from our results that hyperproliferation is prominent due to the rapid cell turn over in the corneal epithelium during the clinical onset of HSK disease.

Trans-differentiation of corneal epithelium during the clinical phase of HSK

In support with our previous data (Figure 30) that showed ocular surface damage, we were further interested to study whether the corneal epithelial phenotype is preserved during ocular HSV-1 infection. To study the corneal epithelial integrity, we sought to investigate the cytokeratin expression pattern in the corneal epithelium of

HSV-1 infected eyes (at day-4, -10, and -16 post-infection) by using immunofluorescence staining for K12 cytokeratin. K12 is specifically expressed on the

112 corneal epithelium. Tile scan image of the corneal section at day-post infection showed the loss of K12 expression, except few regions of the paracentral cornea expressed K12

(Figure 32). In the uninfected control (Figure 32), K12 expression was observed all through the corneal epithelium in a uniform pattern. Since studies have shown that pathological conditions transform the cytokeratin expression pattern of corneal epithelium to ‘skin like epithelium’ [282] we further explored to identify epidermis like phenotype in the corneal epithelium. To do this, we carried out immunofluorescence staining for K10 cytokeratin in the HSV-1 infected eyes at day-4, -10, and -16 post- infection. Surprisingly, we detected an intense expression of K10 in the corneal epithelium on day 10 and day 16 post-infection (Figure 33 G, H). Moderately stained regions of K10 expression were visible at day 4 post-infection (Figure 33 F), while no

K10 stained regions were detected in the naïve (uninfected control) (Figure 33 E).

Notably, a dramatic loss of K12 staining was observed at day -10, and -16 post-infect tion (Figure 33 C, D), while prominent staining of K12 was observed in the naïve (Figure

33 A). Collectively, our results showed a remarkable observation of phenotypic changes in the corneal epithelium during HSV-1 infection. The occurrence of ‘transdifferentiation’ of corneal epithelium is evident from our data indicating a switch from non-keratinized to a ‘keratinized skin like’ epithelium during the clinical disease phase of HSK (Figure 35).

Loss of PAX-6 expression in the corneal epithelium during the clinical phase of HSK A characteristic feature of corneal epithelium is the expression of transcription factor Pax-6. Expression of Pax-6 in the corneal epithelium distinguishes it from the epidermis and has a key role in exclusively eye development and morphogenesis [283].

As aforementioned our results showed trans-differentiation of corneal epithelium that

113 showed loss of K12 expression in the HSV-1 infected cornea epithelium (Figure 32).

Expression of K12 is specific to the corneal epithelium. Therefore, to further investigate whether the corneal epithelial phenotype is maintained in HSV-1 infected corneas, we carried out immunofluorescence staining with the anti-Pax-6 antibody. The frozen corneal sections from the naïve (uninfected control), at day -4, and -10 post-infection were used for staining with an anti-Pax-6 antibody, followed by confocal image acquisition. Remarkably, our results showed a dramatic loss of Pax-6 in the corneal epithelium during the clinical disease phase of HSK (at day -10, -16 post-infection)

(Figure 34 G-L). In contrast, a predominant distribution of positively stained Pax-6 in the nucleus was observed throughout the corneal epithelium in naïve and at day 4 post- infection (Figure 34 A-F). Here in, our results confirm the loss of corneal epithelial identity and the loss of differentiation in the ocular surface lineage.

Discussion

In this study, we investigated a novel phenomenon on alterations of the corneal epithelium and its trans-differentiation into ‘skin-like epithelium’ in HSV-1 infected eyes, which is havoc to clear vision. The results from our study show the existence of dry eye associated with corneal epithelial damage, K12-/K10+ expression and the loss of Pax6 expression in the corneal epithelium of the HSV-1 infected eyes.

The existence of a healthy, smooth and lubricated corneal surface is critical for the maintenance of transparent cornea to aid clear vision. An intact cornea epithelium not only functions as a protective barrier but also aids in tear production and its maintenance. Consequently, this helps in keeping the ocular surface moist and healthy.

However, during dry eye condition, ocular surface damage has been reported [276]. Our

114 data showed the existence of corneal epithelial damage via positive staining of Rose bengal with reduced corneal smoothness in HSV-1 infected eyes. Rose Bengal (RB) is widely known to diagnose dry eye conditions in the clinics. It is known that binding of RB is prominent in the regions with altered mucin layer on the ocular surface or with ocular surface damage [284]. Moreover, our results previously (Figure 25 A) showed reduced tear volumes in HSV-1 infected eyes that may also influence ‘desiccating’ ocular surface. In the mouse model of ‘desiccating dry eye,’ increased epithelial cell proliferation was observed [276]. Our results showed a similar outcome with increased proliferation in the corneal epithelium during HSV-1 infection determined by a positive

BrdU staining. Of note, a phenomenon of ‘hyperplasia’ may be occurring due to the rapid turn-over of the cells and to compensate the damage in the corneal epithelium of

HSV-1 infected eyes. Altogether, these results indicate the occurrence of ocular surface damage causing ‘dry eye’ conditions. Importantly, the increase in tear osmolarity has been detected in dry eye diseases which eventually induces corneal epithelial damage

[285,286]. Thus, may pose a severe impact on preserving the corneal epithelial integrity.

The expression of cytokeratin is crucial in maintaining the integrity of the corneal epithelium [287] Hence, cytokeratin proteins have been widely used in elucidating epithelial phenotype and its differentiation. Corneal epithelium comprises non- keratinized stratified squamous epithelium. Cumulative studies have reported the specific expression of K3 and K12 cytokeratin in the normal corneal epithelium

[288,71,289,290]. Nevertheless, under pathological conditions non-keratinized epithelium transforms into a keratinized epithelium that closely matches the features of

115 skin epithelium (epidermis). The keratin expression profile determines this switch in keratinization in the cornea. Typically, as previously mentioned K3/K12 are the keratins that are specific to the corneal epithelium, which is replaced by epidermis-specific

K1/K10 expression under pathological conditions [282]. Intriguingly, our results showed the loss of K12 expression concomitant to the gain of K10 expression the corneal epithelium at day 10 and day 16 post-HSV-1 ocular infection. This suggests an occurrence of transformation or trans-differentiation in the corneal epithelium during the clinical stage of HSK disease progression (Figure 35).

Additionally, Pax-6 is a prominent marker of ocular surface lineage and has a crucial role in the maintenance, regulation, and differentiation of corneal epithelium

[291]. Nonetheless, mouse model studies have showed that Pax6+/- mice are prone to have brittle corneal epithelium [292]. Cumulative evidence from the reports have indicated that the corneal epithelium in Pax6+/- mice displayed an improper differentiation and maturation of the corneal epithelium. [293,279] Intriguingly, our data showed a dramatic loss of Pax-6 in the corneal epithelium during the clinical disease phase of HSK (at day 10 and day 16 post HSV-1 ocular infection). The loss of Pax-6 is known to induce upregulation of epidermis-specific expression of cytokeratin (K1/K10) in the corneal epithelium via a ‘default differentiation pathway’ [283] Several ocular disease disorders such as Stevens-Johnson syndrome, Sjogren's syndrome, and dry eye have reported a reduced expression of Pax-6 in association to the loss of K12, increased K10 expression and increased proliferation in the corneal epithelium [294].

Moreover, Pax-6 expression is known to be interconnected with the K12 expression, since Pax-6 stimulated the K12 promoter action [295]. In support, our results showed

116 the loss of K12 with the emergence of K10, increased proliferation, and the loss of Pax-

6 in the corneal epithelium during the clinical disease phase of HSK. Altogether, it can be concluded that Pax-6 is a key player in driving the ocular surface phenotype and in the development of corneal epithelium. The members of Wnt pathway play an essential role in signaling Pax-6 and K12 expression in the normal cornea [282] Henceforth, it will be interesting to further explore the factors related to the loss of Pax-6 in the corneal epithelium of HSV-1 infected eyes.

In conclusion, our results confirm the transformation in corneal epithelium leading to the loss of corneal epithelial identity. ‘Squamous metaplasia’ is a phenomenon characterized by an abnormal differentiation of corneal epithelium into the skin like epithelium [296]. It is considered as a hallmark in Sjogren’s syndrome and Stevens-

Johnson syndrome [296,297] Notably, one of the characteristic features of squamous metaplasia include the loss of K3/K12 and the appearance of K1/K10 in the corneal epithelium [271,272]. A recent study has reported the occurrence of epithelial to mesenchymal transition (EMT) in HSV-1 infected corneas [298]. Overall, this evidence supports our study to hypothesize that, the process of squamous metaplasia may likely be occurring in the corneal epithelium of HSV-1 infected corneas.

Taken together, this study provides a comprehensive guide to indicate corneal epithelial damage and transdifferentiation in the corneal epithelium of HSV-1 infected eyes.

117

Figure 30: Irregular corneal smoothness and damage to the ocular surface in the HSV-1 infected eyes. A. The top panel shows eye images with corneal smoothness scores. From left, naïve eyes showing a score of 0; at day 4 post-infection showing a score of 1; at day 10 post-infection showing a score of 5. Corneal smoothness scores were recorded depending on the reflected ring structure on the corneal surface. The intact round ring structure (naïve) signifies a corneal surface, in contrast to a distorted ring structure that signifies a defective/irregular corneal surface (day 10 post-infection). The bottom panel shows eye images stained with Rose Bengal (RB) staining. From left, naïve eyes showing no positive staining of RB; at day 4 post-infection with faint staining of RB indicated by arrows; at day 10 post-infection with a prominent magenta/dark pink punctate stained area indicated by arrows. B. Bar graphs show the corneal smoothness in the naïve, at day-4, -8, and -10 HSV-1 infected eyes. The data were pooled from three independent experiments with n=4-6 eyes per group/experiment. The p values ns were calculated using one-way ANOVA (** p≤0.01, p>0.05).

118

Figure 31: Hyperproliferation of the corneal epithelium during the clinical onset of HSK. A. Representative immunofluorescence tile scan images of the frozen corneal sections stained with anti- BrdU antibody in the naïve, at day -5, and -9 post-infection. Areas showing prominent BrdU staining (colocalized with nuclear staining) is shown in the insets at a higher magnification (x400). Tile scan images were acquired at x200 magnification. DAPI (red, nuclear staining), BrdU (green). B. Bar graphs showing the quantification of BrdU+ cells in the naïve, at day -5, and -9 post-infection. Image J software was used for quantification. Statistical analysis was performed by using unpaired two-tailed Student’s t- test (** p≤0.01, *p≤0.05).

119

Figure 32: Loss of K12 expression the corneal epithelium during the clinical disease of HSK. Representative immunofluorescence tile scan images shown with anti-K12 staining in the naïve cornea (top panel) and at day 10 post-infection (bottom panel). The naïve corneal tile scan image shows an intense membrane staining of K12 throughout the entire corneal epithelium. The insets (A,B) shows magnified images (x400 magnification) of the K12 staining. The day 10 post- infection corneal tile scan image shows loss of K12. K12 staining is not observed throughout the corneal epithelium. The insets (C,D) shows magnified images showing the areas with loss of K12 staining. Tile scan images were acquired at x200 magnification. DAPI (blue, nuclear staining), K12 (green). Immunofluorescence images shown are representative of three corneas at each time-point.

120

Figure 33: Loss of K12 expression and the gain of K10 expression in the corneal epithelium during the clinical disease of HSK. Representative immunofluorescence images of frozen corneal sections stained with K12 (A-D); K10 (E-H). K12 is a corneal epithelial specific cytokeratin that shows intense staining in the naïve corneal epithelium (A) and progressively disappears at day 4 (B) showing patchy K12 staining; at day 10 (C) shows loss of K12, only small regions in the corneal epithelium show K12 staining; at day 16 (D) No staining or the loss of K12) detected. K10 is an epidermis (skin) specific cytokeratin that shows prominent K10 staining at day 10 (G) and at day 16 (H); while a patchy staining pattern of K10 is observed at day 4(F); No K10 staining is observed in the naïve (control, E). Magnification x200. DAPI blue- nucleus staining, K12- green {A-D}. DAPI red- nucleus staining, K12- green {E-H}. Immunofluorescence images shown are representative of three corneas at each time-point.

121

Figure 34: Loss of Pax-6 expression in the corneal epithelium during the clinical disease of HSK. Representative immunofluorescence images of frozen corneal sections showing Pax-6 (green) DAPI (in red), and the merged images with nuclear localization of Pax-6 (in yellow). Pax-6 staining in the naïve (A- C), prominent staining in the nucleus observed as shown in the insets with the zoomed image. At day 4 post-infection (D-F), prominent staining in the nucleus observed as shown in the insets with zoomed images. At day 10 post-infection (G-I), loss of Pax-6 expression observed. At day -16 post-infection (J-L), loss of Pax-6 observed. Magnification X400. Immunofluorescence images shown are representative of three corneas at each time-point.

122

Figure 35: A schematic representation to depict alterations of cytokeratin expression in the corneal epithelium during the clinical disease period. This diagram shows the process of trans differentiation in the corneal epithelium of HSV-1 infected eyes. A loss of K12 expression (corneal epithelial specific) to the gain of the epidermis (skin) specific K10 expression is detected in the corneal epithelium of HSV-1 infected eyes during the clinical disease phase of HSK.

123

Figure 36: A proposed model to summarize the events of ocular surface alterations in HSV-1 infected cornea during the clinical disease period. This model emphasis on the impact of reduced tear levels observed in the HSV-1 infected cornea.

124

CHAPTER 4: IGF-1 DEPENDENT MECHANISM OF IGFBP-3 IN THE PATHOGENESIS OF HERPES STROMAL KERATITIS Abstract

Herpes Stromal Keratitis (HSK) is a chronic immunoinflammatory condition that develops in response to recurrent corneal infection with herpes simplex virus-1 (HSV-1) and is considered a major cause of infection-induced blindness in the United States.

Clinical manifestations of HSK involve the development of angiogenesis and opacity into the avascular cornea. A better understanding of immunoinflammatory events that promote angiogenesis and opacity in infected cornea could provide novel therapeutic targets to treat HSK. IGFBP-3 binds to 75% of serum IGF proteins and regulates their bioavailability and IGF-1R signaling. The latter is known to enhance angiogenesis and immune cell survival and effector function. An elevated level of IGFBP-3 protein was detected in HSV-1 infected corneas while carrying out angiogenic protein array in infected corneal samples from C57BL/6 (B6) mice. To determine the role of IGFBP-3 in the pathogenesis of HSK, B6 and IGFBP-3-/- mice received corneal HSV-1 infection.

Our results showed a significant increase in hem-angiogenesis and opacity scores in infected corneas of IGFBP-3-/- than control B6 mice. IGFBP-3 may exert its protective effect by sequestering IGF molecules and reducing IGF-1R mediated signaling. In support, our results showed increased phosphorylation of IGF-1R expressing leukocytes in infected corneas of IGFBP-3-/- mice. This was associated with an increased number of granulocytes in HSK lesions of IGFBP-3-/- than B6 mice. IGFBP-3 is known to undergo proteolysis by matrix metalloproteinase-3 (MMP-3), releasing bioactive IGF molecules. Previous reports from our laboratory detected an enzymatically active MMP-3 (by Zymography) in B6 corneas with HSK. Together, our

125 results suggest that IGFBP-3 is likely exerting a protective effect since their absence causes an increase in opacity and robust hem-angiogenesis in infected corneas.

IGFBP-3, in IGF-dependent mode, regulates the pathogenesis of HSK.

Introduction

IGFBP-3 is one of the six highly conserved Insulin Growth Factor (IGF)-binding proteins that bind to 75% of the serum IGF proteins to form complexes and regulates

IGF-1R signaling by sequestering IGF molecule. Indeed, IGFBP-3 helps in regulating the bioavailability of IGF proteins and IGF-1R mediated signaling [299,300,154] IGF binding to IGF-1R results in phosphorylation of tyrosine residues in IGF-1Rβ and thereby initiates IGF-1R signaling [301,157]. IGF-1R signaling is reported to initiate hem- and lymphangiogenesis and promote survival of immune cells [302,303,304,305]

IGFBP-3 also acts in IGF-independent mode, which involves cell surface binding, cytosolic accumulation and nuclear localization of IGFBP-3 [306]. IGF-dependent and independent action of IGFBP-3 is known to regulate hem-angiogenesis [307,184]. In this study, we investigated the contribution of IGF dependent action IGFBP-3 and its role in immune survival and pathogenesis of HSK in a mouse model.

Development of hem-angiogenesis is the hallmark of HSK in HSV-1 infected cornea [308]. Our interest was to ascertain the levels of protein expression involved in regulating hem-angiogenesis in HSV-1 infected cornea. An elevated level of IGFBP-3 protein was detected in HSV-1 infected corneas by using a membrane-based mouse angiogenesis array in HSV-1 infected mouse cornea (Figure 37 A). Previous studies have shown that IGFBP-3 can either promote or inhibit hem-angiogenesis

[309,141,310]. Therefore, we next evaluated the role of IGFBP-3 in the pathogenesis of

126

HSK by infecting both, the B6 and IGFBP-3-/- mice with corneal HSV-1 infection. Our results showed a significant increase in hem-angiogenesis and opacity scores in infected corneas of IGFBP-3-/- than control B6 mice (Figure 40 B). As aforementioned, binding of IGFBP-3 to IGF likely exerts its protective effect by sequestering IGF molecules and by reducing IGF-1R mediated signaling. Notably, these results associated with an increased influx and enhanced phosphorylation of IGF-1R expressing leukocytes in infected corneas of IGFBP-3-/- mice (Figure 39). Apart from this, IGFBP-3 is also well known to undergo proteolysis during pathological conditions such as diabetes, arthritis, and cancer by matrix metalloproteinases (MMPs) and may release the bound IGF molecules. MMP-3 and MMP-9 have been reported to involve in the proteolysis of IGFBP-3 in inflamed tissue [167]. Intriguingly, previous data from our lab showed a significant increase in the enzymatically active MMP-3 and MMP-9, which was detected by Zymography. From this, we conclude that the active mechanism of

IGFBP-3 might have been compromised in the presence of proteases which cleaves the

IGFBP-3 molecule. Therefore, this results in the increased hem-angiogenesis and opacity scores in the corneas of HSV-1 infected IGFBP-3-/- mice. Overall, our results suggest that IGFBP-3, in IGF-dependent mode, regulates the pathogenesis of HSK.

In summary, IGFBP-3 plays a defensive role by binding to IGF-1 and thereby inhibits IGF-1R phosphorylation that signals to promote hem-angiogenesis in HSV-1 infected cornea. In support to this, our results have also shown a possible protective role of IGFBP-3 in inhibiting IGF-1R signaling (phosphorylation) and thereby, the development of hem-angiogenesis in HSV-1 infected cornea during the clinical onset of

HSK disease.

127

Results

An Increased level of IGFBP-3 protein and mRNA expression in HSV-1 infected B6 corneas Development of corneal opacity and neovascularization in an avascular cornea is the hallmark of HSK [308]. Primarily, the objective of this study was to understand the underlying mechanisms involved in the pathogenesis of HSK. Hence, a membrane- based angiogenesis array was performed (as described in materials and methods) to detect the level of angiogenesis-regulating proteins in the corneal lysates of uninfected

(naïve) and at day -5, -10, and -15 post-HSV-1 infection. Interestingly, our results showed an increased pixel density of IGFBP-3 in HSV-1 infected corneas (Figure 37 A).

The pixel density was determined by using the Image studio lite software. We further sought to confirm this result by performing ELISA and qRT-PCR assay to estimate the

IGFBP-3 protein and mRNA levels of IGFBP-3 in HSV-1 infected corneas, respectively.

A Significant increase in, mRNA levels of IGFBP-3 gene (Figure 37 B) and the IGFBP-3 protein expression (Figure 37 C) and was detected in the HSV-1 infected corneas during the clinical disease phase of the HSK disease (day -10 and -15 post infection, respectively) when compared to the corneas during pre-clinical phase of HSK disease

(day 5 post-infection).

Furthermore, our angiogenesis protein array data (Figure 37 A) showed an increased level of matrix metalloproteinase-3, and -9 (MMP-3, -9) in infected corneas with HSK. Recently, the zymography data from our lab also confirmed the enzymatic activity of MMP-3 in HSK corneas (data not shown). Collectively, these results provoked us to focus on the role of IGFBP-3 protein in the pathogenesis of HSK and to further study the impact of MMP-3, -9 on IGFBP-3.

128

A significant increase in IGF-1R phosphorylation detected in total leukocytes present in HSK developing corneas of IGFBP3-/- than the control C57BL/6 (B6) mice IGFBP-3 under specific conditions has been reported to either promote or inhibit hem-angiogenesis [309,307,184] Furthermore, IGFBP-3 is known to exert their function via insulin-like growth factor (IGF)-dependent and -independent mechanisms (Figure 7).

IGF-dependent activity includes sequestration of IGF-I, and -II proteins and thereby regulates their bioavailability to the cell surface IGF-1 receptor (IGF-1R) [153]. IGF molecules bind to the IGF-receptor to mediate its autophosphorylation. Subsequently, downstream signaling events occur to initiate the process of hem-angiogenesis, immune cell survival and effector function [304,305,311] (Figure 38A). However, proteolysis of IGFBP-3 can induce enhanced IGF signaling, due to an increased

IGF/IGF-R interaction [166,312]. A study from our laboratory has shown an increased expression of MMP-3 and its enzymatic activity detected in HSV-1 cornea. In support, our protein array data (Figure 37 A) also showed higher levels of MMP-3 and MMP-9 in

HSV-1 infected corneas. Several reports have suggested that in inflamed tissue, MMPs are involved in the proteolysis of IGFBP-3 [168,167] Taken together, we hypothesized that active IGF-1R signaling might be induced due to increased expression of MMPs, that leads to the IGFBP-3 proteolysis in HSV-1 infected corneas.

Concerning this, our foremost interest was to study the phosphorylated

(activated) IGF-1R expression in the HSV-1 infected corneas of B6 mice.

Immunofluorescence staining of the frozen corneal section showed positive staining for phosphorylated IGF-1R (pIGF-1R) in the HSV-1 infected B6 cornea at the clinical disease phase of HSK (day -10 and -17 post-infection) compared to the naïve (Figure

38 B). IGF-1R staining was carried out in the frozen corneal section by using the anti-

129

IGF-IR antibody (Table 1) that distinguishes the phosphorylated tyrosine 1161 (pY1161) in the IGF-1 receptor. Intense staining was detected explicitly in the corneal stroma of

HSV-1 infected eyes at day 17 post-infection (Figure 38 B). This result suggests active

IGF-1R signaling in the corneas of the HSV-1 infected eyes of B6 mice.

To further quantitate the IGF-1R phosphorylation and to study the IGF dependent action of IGFBP-3, we opted to use IGFBP-3 -/- mice in this study. The IGFBP-3 -/- mice on the B6 background [313] were provided by Dr. Steinle laboratory as mentioned in

Materials and Methods. Quantification of IGF-1R phosphorylation was determined by employing flow cytometry analysis in IGFBP-3 -/- and the control B6 HSV-1 infected mice. Our data showed a significant increase in the frequency of CD45+ cells

(leukocytes) and an absolute number of CD4+ T cells in IGFBP-3 -/- than the B6 mice on day 10 post-infection (Figure 39 A, B). Intriguingly, a significant increase in the IGF-1R phosphorylation at tyrosine residue 1131 (pY1131) was detected in leukocytes (CD45+ cells) infiltrating HSK corneas in IGFBP-3 -/- than the B6 mice on day 10 post-infection

(Figure 39 C).

Notably, these results suggest an increased immune cell influx and active IGF-1R signaling in leukocytes infiltrating IGFBP-3 -/- mice in HSK corneas. Moreover, this study provides us the cues on IGF dependent action of IGFBP-3 in HSV-1 infected corneas.

Development of robust HSK disease with a significant increase in hem- angiogenesis and corneal opacity in HSV-1 infected eyes of IGFBP3-/- than the control B6 mice Upon understanding the existence of active IGF-1R signaling in leukocytes infiltrating HSV-1 infected IGFBP-3 -/- mouse corneas, we were next inclined to determine the role of IGFBP-3 in regulating the severity of HSK. Active IGF-1R signaling is known to promote hem-angiogenesis and enhance inflammation [304,311,314].

130

Therefore, we were prompted to hypothesize that, HSV-1 infected IGFBP-3 -/- mouse corneas may display a severe HSK disease. To test this, B6 and IGFBP-3 -/- HSV-1 infected mouse corneal opacity, and hem-angiogenesis scores were assessed by using a hand-held slit lamp as described in Materials and Methods. As expected, our results showed a robust HSK disease in the IGFBP-3 -/- than the B6 HSV-1 infected mouse corneas (Figure 40 A). A significant increase in the corneal opacity and hem- angiogenesis scores were evaluated in the HSV-1 infected corneas of IGFBP-3 -/- mouse corneas during the clinical disease phase of HSK (at day-10, and -12 post- infection) (Figure 40 B). Together, these results suggest that IGFBP-3 plays an essential role in governing the severity of HSK.

Discussion

Corneal hem-angiogenesis and associated immunoinflammatory events in HSV-

1 infected cornea are the prominent features of HSK [234,108]. Hence, a better understanding of the molecules involved in the regulating angiogenesis, immune cell survival and effector function in HSK corneas is essential. Our results from the angiogenesis array data showed an increased expression of IGFBP-3 in HSV-1 infected corneas (Figure 37 A). IGFBP-3 under specific conditions has been reported to either promote or inhibit hem-angiogenesis [309,307,184,310]. Notably, an increased hem- angiogenesis and massive influx of leukocytes were detected in the HSV-1 infected corneas of IGFBP-3 -/- than the B6 infected mice. This indicates an essential role of

IGFBP-3 in the pathogenesis of HSK.

Interestingly, our angiogenesis array data showed an increased expression of

MMP-3 and -9 in HSV-1 infected corneas in B6 mice (Figure 37 A). MMP-3 is known to

131 cause proteolysis of IGFBP-3 [167,165,166]. MMP-3 and -9 are reported to trigger proteolysis of IGFBP-3 resulting in the in the enhancement of the IGF- mediated actions

[166,165,167]. In the mouse model of HSK, the expression and activity of MMP-9 have been reported in HSV-1 infected corneas [170]. MMP-3 serves as an activator of pro-

MMP-9 and is involved in the generation of an active form of MMP-9 [171]. Thus, in

HSK corneas, MMP-3 may regulate the effect of MMP-9 and more importantly in the proteolysis of IGFBP-3.

The mechanism of action of IGFBP-3 can either be in an IGF-dependent or independent manner. Briefly, IGF- dependent action is executed by sequestration of

IGF-I and II proteins and thereby modulating their bioavailability to cell-surface IGF-1 receptor (IGF-1R). IGF-independent action involves cell surface binding of IGFBP-3 and its cytosolic or nuclear localization [179,153]. The aim of this study was to determine the

IGF dependent action of IGFBP-3. Since IGF molecules get complexed with IGFBP-3, the proteolysis of IGFBP-3 in HSV-1 infected cornea may liberate IGF molecules.

Subsequently, the liberated IGF molecules can bind to IGF-1R which indeed can lead to enhanced activation of IGF-1R signaling (autophosphorylation of IGF-1R). Our results showed an intense staining for phosphorylated IGF-1R (IGF-1R, Tyr1161) in HSV-1 infected B6 corneas during the clinical disease period (day-10, -17 post -infection)

(Figure 38). This suggests active IGF-1R signaling, which could be due to the proteolysis of IGFBP-3 by enzymatically active MMP-3.

IGF-1R signaling has been widely reported to have pleiotropic effects, which include enhancing cell proliferation and migration, promoting hem-angiogenesis, immune cell survival and its effector function [314,160,304,305]. In support, our results

132 showed enhanced IGF-1R phosphorylation in infiltrating leukocytes of HSV-1 infected

IGFBP-3 -/- mice than the B6 mice (Figure 39). A robust development of HSK disease was observed with increased corneal opacity and hem-angiogenesis in the HSV-1 infected IGFBP-3 -/- mice than the B6 mice (Figure 40). These results elucidate the involvement of MMP-3 in the proteolysis of IGFBP-3 which consequently leads to activated signaling of the IGF receptor and its adverse effect on HSK disease severity.

To further ensure the role of IGFBP-3 in HSK pathogenesis and its IGF dependent mechanism, an approach to block IGF-1R phosphorylation was carried out in

HSV-1 infected mice. PPP inhibits IGF-1R phosphorylation in a non-ATP competitive manner [316].

In this study, we deciphered the role of IGFBP-3 in HSK developing corneas that may provide cues for novel therapeutic targets and reduce HSK disease severity.

133

Figure 37: A. Increase in IGFBP3 protein expression in C57BL/6 mouse corneas at different time points post-ocular HSV-1 infection. A. Images showing angiogenesis array blots of corneal lysates prepared from the naïve and HSV-1 infected cornea at day 5, 10 and 15 post-ocular infections. Bar diagram on the right showing the quantification of dots represented as signal intensity of IGFBP3, MMP-3, MMP-9 and TIMP-1 using image studio lite software. MMPs -Matrix metalloproteinases, and TIMPs- Tissue inhibitor of Matrix Metalloproteinases. B. Scatter plot showing the relative mRNA expression of IGFBP3 in HSV-1 infected B6 mouse corneas at day 5 and 10 post infection (DPI). C. Bar graph showing the protein levels of IGFBP-3 using ELISA in HSV-1 infected B6 mouse corneas at day 5 and day 15 post-infection.

134

FIGURE 38: Intense IGF-1R staining in HSV-1 infected C57BL/6 (B6) mouse cornea during the clinical disease period. A. schematic of proposed mechanism hypothesizing active IGF-1R signaling by binding to IGF molecule to initiate Tyr-kinase phosphorylation (pY1161). Increased IGF-1R phosphorylation may promote hem-lymph angiogenesis and Immune cell survival. B. From the left showing frozen naïve corneal tissue section of B6 mice with no prominent IGF-1R staining, on day 10 post-infection showing prominently in the corneal epithelium and the stroma, on day 17 post-infection showing intense IGF-1R staining in the corneal stroma. DAPI staining nucleus in blue; pIGF-1R detecting phosphorylated IGF-1R (pY1161) in green. Magnification x400.

135

Figure 39: Significant increase in IGF-1R phosphorylation detected in total leukocytes in HSK developing corneas of IGFBP-3-/- in comparison to the C57BL/6 mice. A. Representative FACS plots are denoting the frequency of leukocytes in HSV-1 infected cornea from the B6 and IGFBP-3 KO mice at day10 post-ocular infection (POI). B. Scatter plot is showing a significant increase in the frequency of CD45+ cells in IGFBP-3KO mice than the B6 on day 10 POI. Bar graph is showing a significant increase in the absolute number of CD4+ T cells in IGFBP-3KO mice than the B6 on day 10 POI. C. Overlay depicts the higher levels of phosphorylated IGF-1R in infected corneas of IGFBP-3 KO compared to the B6 mice. The scatter plot on the right showing a significant increase in Mean fluorescence intensity (MFI) of the phosphorylated IGF-1R in the total leukocytes present in the infected corneas of IGFBP3-KO mice than the B6 on day10 POI. Results shown are representative of two similar experiments. (n=5 to 8 eyes per group) p values were calculated using two-tailed t-test. **p≤0.01, ***p≤0.001, ***p≤0.0001

136

Figure 40: Robust development of HSK disease in HSV-1 infected IGFBP-3-/- mice in comparison to the C57BL/6 mice. A. Representative eye images of B6 and IGFBP-3-/- mice showing more severe disease in the IGFBP-3KO mice with robust development of angiogenesis than in the B6 mice at day 10 and 12 post-ocular HSV-1 infections (POI) during the clinical disease of HSK. B. Scatter plots showing corneal opacity scores and corneal angiogenesis scores at day 7, day 10 and day 12 post-HSV-1 infection, measured using a slit lamp. Both groups of mice were infected with a low dose of virus (10 to 4 p.f.u). Results shown are representative of two similar experiments. (n=5 to 8 eyes per group). p values were calculated using two-tailed t-test. **p≤0.01, ***p≤0.001.

137

Figure 41: Schematic Representation to summarize the Aims investigated in this study in HSV-1 infected mouse corneas.

138

REFERENCES

1. Pepose, J. S., Keadle, T. L. & Morrison, L. A. Ocular Herpes Simplex: Changing

Epidemiology, Emerging Disease Patterns, and the Potential of Vaccine

Prevention and Therapy. Am. J. Ophthalmol. 141, 547–557.e2 (2006).

2. Liesegang, T. J. Herpes simplex virus epidemiology and ocular importance.

Cornea 20, 1–13 (2001).

3. Banks, T. A. & Rouse, B. T. Herpesviruses--Immune Escape Artists? Clin. Infect.

Dis. 14, 933–941 (1992).

4. Smith, J. S. et al. Type-specific seroprevalence of HSV-1 and HSV-2 in four

geographical regions of Poland. Sex. Transm. Infect. 82, 159–63 (2006).

5. Xu, F. et al. Trends in Herpes Simplex Virus Type 1 and Type 2 Seroprevalence

in the United States. JAMA 296, 964 (2006).

6. Smith, J. S. & Robinson, N. J. Age‐Specific Prevalence of Infection with Herpes

Simplex Virus Types 2 and 1: A Global Review. J. Infect. Dis. 186, S3–S28

(2002).

7. Young, R. C., Hodge, D. O., Liesegang, T. J. & Baratz, K. H. Incidence,

recurrence, and outcomes of herpes simplex virus eye disease in Olmsted

County, Minnesota, 1976-2007: the effect of oral antiviral prophylaxis. Arch.

Ophthalmol. (Chicago, Ill. 1960) 128, 1178–83 (2010).

8. Chong, E.-M. et al. Herpes simplex virus keratitis in children. Am. J. Ophthalmol.

138, 474–475 (2004).

9. Nguyen, K., Gallagher, J. & Hitchmoth, D. Herpes Zoster Ophthalmicus: A

Teaching Case Report. Optometric Education 65, (2014).

139

10. Decman, V., Freeman, M. L., Kinchington, P. R. & Hendricks, R. L. Immune

Control of HSV-1 Latency. Viral Immunol. 18, 466–473 (2005).

11. Guess, S., Stone, D. U. & Chodosh, J. Evidence-based treatment of herpes

simplex virus keratitis: a systematic review. Ocul. Surf. 5, 240–50 (2007).

12. Stuart, P. M. & Keadle, T. L. Recurrent herpetic stromal keratitis in mice: a model

for studying human HSK. Clin. Dev. Immunol. 2012, 728480 (2012).

13. Knickelbein, J. E., Hendricks, R. L. & Charukamnoetkanok, P. Management of

Herpes Simplex Virus Stromal Keratitis: An Evidence-based Review. Surv.

Ophthalmol. 54, 226–234 (2009).

14. Kieff, E. D., Bachenheimer, S. L. & Roizman, B. Size, composition, and structure

of the deoxyribonucleic acid of herpes simplex virus subtypes 1 and 2. J. Virol. 8,

125–32 (1971).

15. Furlong, D., Swift, H. & Roizman, B. Arrangement of herpesvirus deoxyribonucleic

acid in the core. J. Virol. 10, 1071–4 (1972).

16. Spear, P. G. Herpes simplex virus: receptors and ligands for cell entry. Cell.

Microbiol. 6, 401–410 (2004).

17. Shah, A., Farooq, A. V, Tiwari, V., Kim, M.-J. & Shukla, D. HSV-1 infection of

human corneal epithelial cells: receptor-mediated entry and trends of re-infection.

Mol. Vis. 16, 2476–86 (2010).

18. Karaba, A. H., Kopp, S. J. & Longnecker, R. Herpesvirus entry mediator and

nectin-1 mediate herpes simplex virus 1 infection of the murine cornea. J. Virol.

85, 10041–7 (2011).

19. Ojala, P. M., Sodeik, B., Ebersold, M. W., Kutay, U. & Helenius, A. Herpes

140

simplex virus type 1 entry into host cells: reconstitution of capsid binding and

uncoating at the nuclear pore complex in vitro. Mol. Cell. Biol. 20, 4922–31

(2000).

20. Triezenberg, S. J., Kingsbury, R. C. & McKnight, S. L. Functional dissection of

VP16, the trans-activator of herpes simplex virus immediate early gene

expression. Genes Dev. 2, 718–29 (1988).

21. Sadowski, I., Ma, J., Triezenberg, S. & Ptashne, M. GAL4-VP16 is an unusually

potent transcriptional activator. Nature 335, 563–564 (1988).

22. Skepper, J. N., Whiteley, A., Browne, H. & Minson, A. Herpes Simplex Virus

Nucleocapsids Mature to Progeny Virions by an

Envelopment Deenvelopment Reenvelopment Pathway. J. Virol. 75, 5697–

5702 (2001).

23. Toma, H. S. et al. Ocular HSV-1 Latency, Reactivation and Recurrent Disease.

Semin. Ophthalmol. 23, 249–273 (2008).

24. Baringer, J. R. & Swoveland, P. Recovery of Herpes-Simplex Virus from Human

Trigeminal Ganglions. N. Engl. J. Med. 288, 648–650 (1973).

25. Bloom, D. C. HSV LAT and neuronal survival. Int. Rev. Immunol. 23, 187–98

26. Bauer, D., Mrzyk, S., van Rooijen, N., Steuhl, K. P. & Heiligenhaus, A.

Macrophage-depletion influences the course of murine HSV-1 keratitis. Curr. Eye

Res. 20, 45–53 (2000).

27. Tumpey, T. M., Chen, S. H., Oakes, J. E. & Lausch, R. N. Neutrophil-mediated

suppression of virus replication after herpes simplex virus type 1 infection of the

murine cornea. J. Virol. 70, 898–904 (1996).

141

28. Rowe, A. M. et al. Herpes keratitis. Prog. Retin. Eye Res. 32, 88–101 (2013).

29. Biswas, P. S. & Rouse, B. T. Early events in HSV keratitis—setting the stage for a

blinding disease. Microbes Infect. 7, 799–810 (2005).

30. Thomas, J., Gangappa, S., Kanangat, S. & Rouse, B. T. On the essential

involvement of neutrophils in the immunopathologic disease: herpetic stromal

keratitis. J. Immunol. 158, 1383–91 (1997).

31. Biswas, P. S. & Rouse, B. T. Early events in HSV keratitis--setting the stage for a

blinding disease. Microbes Infect. 7, 799–810 (2005).

32. Niemialtowski, M. G. & Rouse, B. T. Predominance of Th1 cells in ocular tissues

during herpetic stromal keratitis. J. Immunol. 149, 3035–9 (1992).

33. Niemialtowski, M. G. & Rouse, B. T. Phenotypic and functional studies on ocular

T cells during herpetic infections of the eye. J. Immunol. 148, 1864–70 (1992).

34. Doymaz, M. Z. & Rouse, B. T. Herpetic stromal keratitis: an immunopathologic

disease mediated by CD4+ T lymphocytes. Invest. Ophthalmol. Vis. Sci. 33,

2165–73 (1992).

35. Liu, T., Khanna, K. M., Carriere, B. N. & Hendricks, R. L. Gamma Interferon Can

Prevent Herpes Simplex Virus Type 1 Reactivation from Latency in Sensory

Neurons. J. Virol. 75, 11178–11184 (2001).

36. Ohara, P. T., Chin, M. S. & LaVail, J. H. The spread of herpes simplex virus type

1 from trigeminal neurons to the murine cornea: an immunoelectron microscopy

study. J. Virol. 74, 4776–86 (2000).

37. Mistry, S. K., Zheng, M., Rouse, B. T. & Morris, S. M. Induction of arginases I and

II in cornea during herpes simplex virus infection. Virus Res. 73, 177–82 (2001).

142

38. Scapini, P., Calzetti, F. & Cassatella, M. A. On the detection of neutrophil-derived

vascular endothelial growth factor (VEGF). J. Immunol. Methods 232, 121–9

(1999).

39. Brogden, K. A., Guthmiller, J. M., Salzet, M. & Zasloff, M. The nervous system

and innate immunity: the neuropeptide connection. Nat. Immunol. 558, (2005).

40. Belmonte, C., Acosta, M. C. & Gallar, J. Neural basis of sensation in intact and

injured corneas. Exp. Eye Res. 78, 513–25 (2004).

41. Sternberg, E. M. Neural regulation of innate immunity: a coordinated nonspecific

host response to pathogens. Nat. Rev. Immunol. 6, 318–28 (2006).

42. Gipson, I. K. Age-Related Changes and Diseases of the Ocular Surface and

Cornea. Investig. Ophthalmology Vis. Sci. 54, ORSF48 (2013).

43. Holly, F. J. Tear film physiology. Am. J. Optom. Physiol. Opt. 57, 252–7 (1980).

44. Prydal JL, Artal P, Woon H, C. F. Study of precorneal tear film thickness and

structure by interferometry and confocal microscopy. IOVS 33, 2006–2011 (1992).

45. Driver, P. J. & Lemp, M. A. Meibomian gland dysfunction. Surv. Ophthalmol. 40,

343–67 (1996).

46. Greiner, J. V., Glonek, T., Korb, D. R., Booth, R. & Leahy, C. D. Phospholipids in

Meibomian Gland Secretion. Ophthalmic Res. 28, 44–49 (1996).

47. Gilbard JP, Rossi SR, H. K. Tear film and ocular surface changes after closure of

the meibomian gland orifices in the rabbit. Ophthalmology 96, 1180–1186 (1989).

48. CLAO, J. L. M. Clinical trials in Dry eye disease. Rep. Natl. EYE institute/ Ind.

Work. 21, 221–32 (1995).

49. Iwata. S. Chemical composition of the aqueous phase. Int. Ophthalmol. Clin. 13,

143

29–46 (1983).

50. Dilly PN. Conjunctival cells, subsurface vesicles, and tear film mucus. Precorneal

Tear Film Heal. Dis. Contact Lens Wear (ed.Holly FJ)Dry Eye Inst. , Inc, Lubbock,

Texas 677–687 (1986).

51. Mantelli F, A. P. Functions of Ocular surface mucins in health and disease. Curr

Opin Allergy Clin Immunol 8, 477–83 (2008).

52. Davidson, H. J. & Kuonen, V. J. The tear film and ocular mucins. Vet. Ophthalmol.

7, 71–77 (2004).

53. Pflugfelder SC, Liu Z, Monroy D, et al. Detection of sialomucin complex (MUC4)

in human ocular surface epithelium and tear fluid. IOVS 41, (2000).

54. Argueso P, Spurr-Michaud S, Russo CL, et al. MUC16 mucin is expressed by the

human ocular surface epithelia and carries the H185 carbohydrate epitope. IOVS

44, (2003).

55. Inatomi T, Spurr-Michaud S, Tisdale AS, Zhan Q, Feldman ST, G. I. Expression of

secretory mucin genes by human conjunctival epithelia. IOVS 37, 1684–92

(1996).

56. Argueso P, Balaram M, S.-M. S. Decreased levels of the goblet cell mucin

MUC5AC in tears of patients with Sjogren syndrome. IOVS 43, 1004–11 (2002).

57. Mc Kenzie RW, Jumblatt JE, J. M. Quantification of MUC2 and MUC5AC

transcripts in human conjunctiva. IOVS 41, 703–708 (2000).

58. Nicholas B, Dawson CR, T. B. Surface features of the conjunctiva and cornea.

IOVS 24, 570–576 (1983).

59. Chen M, Wang Y, Begley CG, W. J. Synthesis of rabbit corneal epithelial

144

glycocalyx in vitro. Exp. Eye Res. 58, 267–276 (1994).

60. Gipson IK, Yankauckas M, S.-M. S. et al. Characteristics of a glycoprotein in the

ocular surface glycocalyx. IOVS 33, 218–227 (1992).

61. Ashutosh S. Energetics of corneal epithelial cell -ocular mucus-tear film

interactions: some surface -chemical pathways of corneal defense. Biophys.

Chem. 47, (1993).

62. Lens, A., Nemeth, S. C. & Ledford, J. K. Ocular anatomy and physiology.

(SLACK, 2008).

63. Lens A, Nemeth S, L. J. Ocular Anatomy and Physiology. SLACK Inc. Thorofare,

NJ (2007).

64. AJ, B. The architecture of the corneal stroma. Br J Ophthalmol. 85, 379–381

(2001).

65. Dartt D, Bex P, D’amore P, Dana P Mcloon L, N. J. Ocular Periphery and

disorders. Acad. Press. San Diego, CA (2011).

66. Pflugfelder, S. C. Tear Dysfunction and the Cornea: LXVII Edward Jackson

Memorial Lecture. Am J Ophthalmol 152, 900–909 (2012).

67. Moll, R., Divo, M. & Langbein, L. The human keratins: biology and pathology.

Histochem. Cell Biol. 129, 705–33 (2008).

68. Schweizer J, Langbein L, R. M. W. H. Hair follicle specific keratins and their

diseases. Exp Cell Res 313, 2010–2020 (2007).

69. RG, O. Intermediate filaments: a historical perspective. Exp Cell Res 313, 1981–

1994 (2007).

70. Schermer A, Galvin S, S. T. Differentiation-related expression of a major 64 K

145

corneal epithelial stem cells. J Cell Biol 103, 341–354 (1986).

71. Pitz S, M. R. Intermediate-filament expression in ocular tissue. Prog Retin Eye

Res 21, 241–262 (2002).

72. Yoshida, S., Shimmura, S., Kawakita, T. & Miyashita, H. Cytokeratin 15 Can Be

Used to Identify the Limbal Phenotype in Normal and Diseased Ocular Surfaces

AND. 47, 4780–4786 (2016).

73. Mc Lean WH, I. A. Disorders of keratinisation: from rare to common genetic

diseases of skin and other epithelial tissues. Ulster Med J 76, 72–82 (2007).

74. Irvine AD, M. L. W. Mutations in cornea-specific keratin K3 or K12 genes case

Meesman’s corneal dystrophy. Nat Genet 16, 184–187 (1997).

75. Li, W. et al. Down-regulation ofPax6 is associated with abnormal differentiation of

corneal epithelial cells in severe ocular surface diseases. J. Pathol. 214, 114–122

(2008).

76. Kathern E. Myrna, DVM, Simon A. Pot, DVM, and Christopher J. Murphy, DVM,

PhD, D. Meet the Corneal Myofibroblast: the role of myofibroblast transformation

in corneal wound healing and pathology. Vet Ophthalmol. 12, 25–27 (2012).

77. Shaheen BS, Bakir M, J. S. Corneal nerves in health and disease. Surv.

Ophthalmol. 59, 263–85 (2014).

78. Pascolini D, M. S. Global estimates of visual impairment:2010. Br J Ophthalmol

96, 614–8 (2012).

79. Cruzat A, Witkin D, Baniasadi N, Zheng L, Ciolino JB, Jurkunas UV, Chodosh J,

Pavan-Langston D, Dana R, H. P. Inflammation and the nervous system: The

connection in the cornea in patients with infectious keratitis. IOVS 52, 5136–43

146

(2011).

80. Chucair-Elliott, A. J., Zheng, M. & Carr, D. J. J. Degeneration and regeneration of

corneal nerves in response to HSV-1 infection. Invest. Ophthalmol. Vis. Sci. 56,

1097–107 (2015).

81. Müller LJ, Marfurt CF, Kruse F, T. T. Corneal nerves: structure, contents and

function. Exp Eye Research 76, 521–542 (2003).

82. Al-Aqaba, M. A., Fares, U., Suleman, H., Lowe, J. & Dua, H. S. Architecture and

distribution of human corneal nerves. Br. J. Ophthalmol. 94, 784–789 (2010).

83. Marfurt CF, Cox J, Deek S, D. L. Anatomy of the human corneal innervation. EXP

Eye Research 90, 478–92 (2010).

84. Murphy PJ, Patel S, Kong N, Ryder RE, M. Noninvasive assessment of corneal

sensitivity in young and elderly diabetic and nondiabetic subjects. IOVS 45, 1737–

42 (2004).

85. Namavari A, Chaudhary S, Sarkar J, Yco L, Patel K, Han KY, Yue BY, Chang JH,

J. S. In vivo serial imaging of regenerating corneal nerves after surgical

transection in transgenic thy1-YFP mice. IOVS 52, 8025–32 (2011).

86. Corfas G, Velardez MO, Ko CP, Ratner N, P. E. Mechanisms and roles of axon-

Schwann cell interactions. J Neurosci. 24, 9250–60 (2004).

87. He, J., Bazan, N. G. & Bazan, H. E. P. Mapping the entire human corneal nerve

architecture. Exp. Eye Res. 91, 513–523 (2010).

88. Müller LJ, Pels L, V. G. Ultrastructural organization of human corneal nerves.

IOVS 37, 476–88 (1996).

89. Hiura A, N. H. Innervation of TRPV1-, PGP-, and CGRP-immunoreactive fibers in

147

the subepithelial layer of a whole mount preparation of the rat cornea. Okajimas

Folia Anat Jpn. 89, 47–50 (2012).

90. Sakimoto T, Kim TI, Ellenberg D, Fukai N, Jain S, Azar DT, C. J. Collagen XVIII

and corneal reinnervation following keratectomy. FEBS Lett 582, 3674–80 (2008).

91. Patel DV, M. C. Mapping of the normal human corneal subbasal nerve plexus by

in vivo laser scanning confocal microscopy. IOVS 46, 4485–8 (2005).

92. Shaheen, B. S., Bakir, M. & Jain, S. Corneal nerves in health and disease. Surv.

Ophthalmol. 59, 263–285 (2014).

93. Belmonte C, Garcia-Hirschfeld J, G. J. Neurobiology of ocular pain. Prog Retin

Eye Res 16, (1997).

94. Murata Y, M. S. Peripheral and central distribution of TRPV1, substance P and

CGRP of rat corneal neurons. Brain Res. 1085, (2006).

95. Nakamura M, Nishida T, Ofuji K, Reid TW, Mannis MJ, M. C. Synergistic effect of

substance P with epidermal growth factor on epithelial migration in rabbit cornea.

Exp Eye Research 65, 321–9 (1997).

96. Namavari A, Chaudhary S, Ozturk O, Chang JH, Yco L, Sonawane S, Katam N,

Khanolkar V, Hallak J, Sarkar J, J. S. Semaphorin 7a links nerve regeneration

and inflammation in the cornea. Cornea 53, 4575–85 (2012).

97. Kubilus JK, L. T. Developmental guidance of embryonic corneal innervation: roles

of Semaphorin 3A and Slit2. Dev Biol 344, 172–84 (2010).

98. Gipson, I. K., Spurr-Michaud, S. J., Senchyna, M., Ritter, R. & Schaumberg, D.

Comparison of Mucin Levels at the Ocular Surface of Postmenopausal Women

With and Without a History of Dry Eye. Cornea 30, 1346–1352 (2011).

148

99. Dartt, D. A. Neural regulation of lacrimal gland secretory processes: Relevance in

dry eye diseases. Prog. Retin. Eye Res. 28, 155–177 (2009).

100. Bonini, S., Rama, P., Olzi, D. & Lambiase, A. Neurotrophic keratitis. Eye 17, 989–

995 (2003).

101. Ueno H, Ferrari G, Hattori T, Saban DR, Katikireddy KR, Chauhan SK, D. R.

Dependence of corneal stem/progenitor cells on ocular surface innervation. IOVS

53, 867–72 (2012).

102. Leiper, L. J., Ou, J., Walczysko, P., Kucerova, R. & Lavery, D. N. Europe PMC

Funders Group Control of Patterns of Corneal Innervation by Pax6. 50, 1122–

1128 (2009).

103. Davis, J., Duncan, M. K., Robison, W. G. & Piatigorsky, J. Requirement for Pax6

in corneal morphogenesis: a role in adhesion. J. Cell Sci. 116, 2157–67 (2003).

104. Azar, D. T. Corneal Angiogenic privilege: Angiogenic and Antiangiogenic Factors

In Corneal Avascularity, Vasculogenesis, and Wound healing (an American

Ophthalmological Society Thesis). Trans Am Ophthalmol Soc 104, 264–302

(2006).

105. PJ, P. The pathophysiology of angiogenesis. Crit. Rev Oral Biol Med 6, 230–47

(1995).

106. Stefano Barabino, Yihe Chen, Sunil Chauhan, and R. D. Ocular Surface

Immunity: Homeostatic Mechanisms and Their Disruption in Dry Eye Disease.

Prog Retin Eye Res 31, 271–285 (2013).

107. Author, D. I. H. and I. Z. The vascular endothelial growth factor (VEGF) family:

angiogenic factors in health and disease. Genome Biol 6, (2005).

149

108. Fernanda Glimenez, Amol suryawanshi, B. T. R. Pathogenesis of Herpes stromal

keratitis- a focus on corneal neovascularization. Prog Retin Eye Res 33, 1–9

(2013).

109. Johnson, A. W. and R. S. Biology of HIF-1α. Cell Death Differ. 5, 621–627

(20081).

110. Arjamaa O, N. M. Oxygen-dependent diseases in the retina: role of hypoxia-

inducible factors. Exp Eye Research 83, 473–483 (2006).

111. Chua, Y. L. et al. Stabilization of Hypoxia-inducible Factor-1α Protein in Hypoxia

Occurs Independently of Mitochondrial Reactive Oxygen Species Production. J.

Biol. Chem. 285, 31277–31284 (2010).

112. Ziello, J. E., Jovin, I. S. & Huang, Y. Hypoxia-Inducible Factor (HIF)-1 regulatory

pathway and its potential for therapeutic intervention in malignancy and ischemia.

Yale J. Biol. Med. 80, 51–60 (2007).

113. Krock BL, Skuli N, S. M. Hypoxia-induced angiogenesis: good and evil. Genes

Cancer 2, 1117–33 (2011).

114. Aragones J, Fraisl P, Baes M, C. P. Oxygen sensors at the crossroad of

metabolism. Cell Metab 9, 11–22 (2009).

115. Huang LE, Gu J, Schau M, B. H. Regulation of hypoxia-inducible factor 1α is

mediated by an O2-dependent degradation domain via the ubiquitin-proteasome

pathway. PNAS 95, 7987–7992 (1998).

116. Wenger RH, Stiehl DP, C. G. Integration of oxygen signaling at the consensus

HRE. Sci STKE. 2005

117. Keith, B., Johnson, R. S. & Simon, M. C. HIF1α and HIF2α: sibling rivalry in

150

hypoxic tumour growth and progression. Nat. Rev. Cancer 12, 9–22 (2012).

118. Manfred Zierhuta, , M.Reza Danab, Michael E Sternc, D. A. S. Immunology of

lacrimal gland and ocular tear film. Trends Immunol. 23, 333–335 (2002).

119. Venable, J. H. & Grafflin, A. L. Gross Anatomy of the Orbital Glands in the Albino

Rat. J. Mammal. 21, 66–71 (1940).

120. AK, M. Lacrimal fluid and electrolyte secretion: a review. Curr. Eye Res. 8, 607–

617 (1989).

121. Rios JD, Horikawa Y, Chen LL, Kublin CL, Hodges RR, Dartt DA, Z. D. Age-

dependent alterations in mouse exorbital lacrimal gland structure, innervation and

secretory response. Exp Eye Research 80, 477–491 (2005).

122. GL., R. Changes in nerve terminals and acini of the lacrimal gland and changes in

secretion induced by autonomic denervation. Z Zellforsch Mikrosk Anat. 94, 261–

281 (1969).

123. Mutch JR. The Lacrimation Reflex. Br J Ophthalmol. 28, 317–336 (1944).

124. Zoukhri, D. Effect of inflammation on lacrimal gland function. Exp. Eye Res. 82,

885–98 (2006).

125. Cousen P, Cackett P, Bennett H, Swa K, D. B. Tear production and corneal

sensitivity in diabetes. J Diabetes Complicat. 21, 371–373 (2007).

126. Hodges, R. R. & Dartt, D. A. Regulatory pathways in lacrimal gland epithelium.

Int. Rev. Cytol. 231, 129–96 (2003).

127. Zoukhri D, K. C. Impaired neurotransmitter release from lacrimal and salivary

gland nerves of a murine model of Sjogren’s syndrome. IOVS 42, 925–932

(2001).

151

128. Hodges, R. R. & Dartt, D. A. Tear film mucins: Front line defenders of the ocular

surface; comparison with airway and gastrointestinal tract mucins. Exp. Eye Res.

117, 62–78 (2013).

129. Johnny J Yanga, Hideo Uedaa, Kwang-Jin Kimb, V. H. . L. Meeting future

challenges in topical ocular drug delivery:: Development of an air-interfaced

primary culture of rabbit conjunctival epithelial cells on a permeable support for

drug transport studies. J. Control. Release 65, 1–11 (2000).

130. Horibe, Y., Hosoya, K., Kim, K. J. & Lee, V. H. Carrier-mediated transport of

monocarboxylate drugs in the pigmented rabbit conjunctiva. Invest. Ophthalmol.

Vis. Sci. 39, 1436–43 (1998).

131. Greiner JV, A. M. Effect of contact lens wear on the conjunctival mucous system.

Ophthalmology 88, 821–32 (1981).

132. Kruse, F. E., Chen, J. J., Tsai, R. J. & Tseng, S. C. Conjunctival

transdifferentiation is due to the incomplete removal of limbal basal epithelium.

Invest. Ophthalmol. Vis. Sci. 31, 1903–13 (1990).

133. Macintosh, S. The innervation of the conjunctiva in monkeys. Albr. von Graefes

Arch. für Klin. und Exp. Ophthalmol. 192, 105–116 (1974).

134. Elsås T, Edvinsson L, Sundler F, U. R. Neuronal pathways to the rat conjunctiva

revealed by retrograde tracing and immunocytochemistry. Exp Eye Research 58,

117–26 (1994).

135. Kessler, T. L., Mercer, H. J., Zieske, J. D., McCarthy, D. M. & Dartt, D. A.

Stimulation of goblet cell mucous secretion by activation of nerves in rat

conjunctiva. Curr. Eye Res. 14, 985–92 (1995).

152

136. Argüeso, P. & Gipson, I. K. Epithelial Mucins of the Ocular Surface: Structure,

Biosynthesis and Function. Exp. Eye Res. 73, 281–289 (2001).

137. Grimberg, A. p53 and IGFBP-3: Apoptosis and Cancer Protection. Mol. Genet.

Metab. 70, 85–98 (2000).

138. Dung V. Nguyen, Sergio Li Calzi, Lynn C. Shaw, Jennifer L. Kielczewski, Hannah

E. Korah, and M. B. G. An ocular view of the IGF-IGFBP system. Growth Horm

IGF Res 23, (2013).

139. Baxter RC, M. J. Binding proteins for the insulin-like growth factors: structure,

regulation and function. Prog. Growth Factor Res. 1, 49–68 (1989).

140. Ferry RJ, Cerri RW, C. P. Insulin-like growth factor binding proteins: new proteins,

new functions. Horm. Res. 51, 53–67 (1999).

141. Chang KH, Chan-Ling T, McFarland EL, Afzal A, Pan H, Baxter LC, et al. IGF

binding protein-3 regulates hematopoietic stem cell and endothelial precursor cell

function during vascular development. Proc Natl Acad Sci 104, 10595–600

(2007).

142. Pollak M. Insulin and insulin-like growth factor signalling in neoplasia. Nat Rev

Cancer 8, 915–28 (2008).

143. Izumi K, Kurosaka D, Iwata T, Oguchi Y, Tanaka Y, Mashima Y, T. K.

Involvement of insulin-like growth factor-I and insulin-like growth factor binding

protein-3 in corneal fibroblasts during corneal wound healing. IOVS 47, 591–8

(2006).

144. Grkovic S, O’Reilly VC, Han S, Hong M, Baxter RC, F. S. IGFBP-3 binds GRP78,

stimulates autophagy and promotes the survival of breast cancer cells exposed to

153

adverse microenvironments. Oncogene 1, 1–9 (2012).

145. Kato, H., Faria, T. N., Stannard, B., Roberts, C. T. & LeRoith, D. Role of tyrosine

kinase activity in signal transduction by the insulin-like growth factor-I (IGF-I)

receptor. Characterization of kinase-deficient IGF-I receptors and the action of an

IGF-I-mimetic antibody (alpha IR-3). J. Biol. Chem. 268, 2655–61 (1993).

146. Leroith, D., Werner, H., Beitner-Johnson, D. & Roberts, A. T. Molecular and

Cellular Aspects of the Insulin-Like Growth Factor I Receptor. Endocr. Rev. 16,

143–163 (1995).

147. BlatC, VillaudyJ, B. In vivo proteolysis of serum insulin-like growth factor (IGF)

binding protein-3 results in increased availability of IGF to target cells. J. Clin.

Investigation 93, 2286–2290 (1994).

148. Mohan S, B. IGF-binding proteins are multifunctional and act via IGF-dependent

and -independent mechanisms. J. Endocrinol. 1751, 19–31 (2002).

149. Schoen, T. J. et al. Identification and partial characterization of a proteinase

specific for insulin-like growth factor binding protein-3 in aqueous and vitreous

humors. Curr. Eye Res. 14, 127–35 (1995).

150. Izumi, K. et al. Involvement of Insulin-like Growth Factor-I and Insulin-like Growth

Factor Binding Protein-3 in Corneal Fibroblasts during Corneal Wound Healing.

Investig. Ophthalmology Vis. Sci. 47, 591 (2006).

151. Robertson DM, Ho SI, Hansen BS, Petroll WM, C. H. Insulin-like Growth Factor

Binding Protein-3 expression in the human corneal epithelium. Exp Eye Research

85, 492–501 (2007).

152. Firth, S. M. & Baxter, R. C. Cellular Actions of the Insulin-Like Growth Factor

154

Binding Proteins. Endocr. Rev. 23, 824–854 (2002).

153. Clemmons, D. R. et al. Role of insulin-like growth factor binding proteins in the

control of IGF actions. Prog. Growth Factor Res. 6, 357–66 (1995).

154. JONES, J. I. & CLEMMONS, D. R. Insulin-Like Growth Factors and Their Binding

Proteins: Biological Actions*. Endocr. Rev. 16, 3–34 (1995).

155. Kelley, K. M. et al. Insulin-like growth factor-binding proteins (IGFBPs) and their

regulatory dynamics. Int. J. Biochem. Cell Biol. 28, 619–37 (1996).

156. Rajaram, S., Baylink, D. J. & Mohan, S. Insulin-Like Growth Factor-Binding

Proteins in Serum and Other Biological Fluids: Regulation and Functions 1.

Endocr. Rev. 18, 801–831 (1997).

157. Kato, H., Faria, T. N., Stannard, B., Roberts, C. T. & LeRoith, D. Essential role of

tyrosine residues 1131, 1135, and 1136 of the insulin-like growth factor-I (IGF-I)

receptor in IGF-I action. Mol. Endocrinol. 8, 40–50 (1994).

158. Dufourny, B. et al. Mitogenic signaling of insulin-like growth factor I in MCF-7

human breast cancer cells requires phosphatidylinositol 3-kinase and is

independent of mitogen-activated protein kinase. J. Biol. Chem. 272, 31163–71

(1997).

159. Walsh, P. T., Smith, L. M. & O’Connor, R. Insulin-like growth factor-1 activates Akt

and Jun N-terminal kinases (JNKs) in promoting the survival of T lymphocytes.

Immunology 107, 461–71 (2002).

160. Dupont, J., Pierre, A., Froment, P. & Moreau, C. The Insulin-like Growth Factor

Axis in Cell Cycle Progression. Horm. Metab. Res. 35, 740–750 (2003).

161. Walsh, P. T. & O’Connor, R. The insulin-like growth factor-I receptor is regulated

155

by CD28 and protects activated T cells from apoptosis. Eur. J. Immunol. 30,

1010–1018 (2000).

162. Qin, X., Strong, D. D., Baylink, D. J. & Mohan, S. Structure-function analysis of

the human insulin-like growth factor binding protein-4. J. Biol. Chem. 273, 23509–

16 (1998).

163. Mott, J. D. & Werb, Z. Regulation of matrix biology by matrix metalloproteinases.

Curr. Opin. Cell Biol. 16, 558–64 (2004).

164. Mañes, S. et al. The matrix metalloproteinase-9 regulates the insulin-like growth

factor-triggered autocrine response in DU-145 carcinoma cells. J. Biol. Chem.

274, 6935–45 (1999).

165. Fowlkes, J. L. et al. Regulation of Insulin-Like Growth Factor (IGF)-I Action by

Matrix Metalloproteinase-3 Involves Selective Disruption of IGF-I/IGF-Binding

Protein-3 Complexes. Endocrinology 145, 620–626 (2004).

166. Fowlkes, J. L., Enghild, J. J., Suzuki, K. & Nagase, H. Matrix metalloproteinases

degrade insulin-like growth factor-binding protein-3 in dermal fibroblast cultures. J.

Biol. Chem. 269, 25742–6 (1994).

167. Bunn, R. C. & Fowlkes, J. L. Insulin-like growth factor binding protein proteolysis.

Trends Endocrinol. Metab. 14, 176–81

168. Coussens, L. M. & Werb, Z. Matrix metalloproteinases and the development of

cancer. Chem. Biol. 3, 895–904 (1996).

169. Sternlicht, M. D. & Werb, Z. How Matrix Metalloproteinases Regulate Cell

Behavior. Annu. Rev. Cell Dev. Biol. 17, 463–516 (2001).

170. Lee, S., Zheng, M., Kim, B. & Rouse, B. T. Role of matrix metalloproteinase-9 in

156

angiogenesis caused by ocular infection with herpes simplex virus. J. Clin. Invest.

110, 1105–11 (2002).

171. Ogata, Y., Enghild, J. J. & Nagase, H. Matrix metalloproteinase 3 (stromelysin)

activates the precursor for the human matrix metalloproteinase 9. J. Biol. Chem.

267, 3581–4 (1992).

172. Yan, X., Forbes, B. E., McNeil, K. A., Baxter, R. C. & Firth, S. M. Role of N- and

C-terminal Residues of Insulin-like Growth Factor (IGF)-binding Protein-3 in

Regulating IGF Complex Formation and Receptor Activation. J. Biol. Chem. 279,

53232–53240 (2004).

173. Grkovic, S. et al. IGFBP-3 binds GRP78, stimulates autophagy and promotes the

survival of breast cancer cells exposed to adverse microenvironments. Oncogene

32, 2412–2420 (2013).

174. Leal, S. M., Liu, Q., Huang, S. S. & Huang, J. S. The type V transforming growth

factor beta receptor is the putative insulin-like growth factor-binding protein 3

receptor. J. Biol. Chem. 272, 20572–6 (1997).

175. Leal, S. M., Huang, S. S. & Huang, J. S. Interactions of high affinity insulin-like

growth factor-binding proteins with the type V transforming growth factor-beta

receptor in mink lung epithelial cells. J. Biol. Chem. 274, 6711–7 (1999).

176. Zhang, Q., Soderland, C. & Steinle, J. J. Regulation of retinal endothelial cell

apoptosis through activation of the IGFBP-3 receptor. Apoptosis 18, 361–368

(2013).

177. HUANG, S. S. et al. Cellular growth inhibition by IGFBP-3 and TGF-β 1 requires

LRP-1. FASEB J. 17, 2068–2081 (2003).

157

178. Janssen, J. A. M. J. L. et al. Free and Total Insulin-Like Growth Factor I (IGF-I),

IGF-Binding Protein-1 (IGFBP-1), and IGFBP-3 and Their Relationships to the

Presence of Diabetic Retinopathy and Glomerular Hyperfiltration in Insulin-

Dependent Diabetes Mellitus 1. J. Clin. Endocrinol. Metab. 82, 2809–2815 (1997).

179. Baxter, R. C. Nuclear actions of insulin-like growth factor binding protein-3. Gene

569, 7–13 (2015).

180. Liu, B. et al. Direct Functional Interactions between Insulin-like Growth Factor-

binding Protein-3 and Retinoid X Receptor-α Regulate Transcriptional Signaling

and Apoptosis. J. Biol. Chem. 275, 33607–33613 (2000).

181. Lin, M. Z., Marzec, K. A., Martin, J. L. & Baxter, R. C. The role of insulin-like

growth factor binding protein-3 in the breast cancer cell response to DNA-

damaging agents. Oncogene 33, 85–96 (2014).

182. Zhang, Q. et al. Compound 49b Prevents Diabetes-Induced Apoptosis through

Increased IGFBP-3 Levels. Investig. Ophthalmology Vis. Sci. 53, 3004 (2012).

183. Gribben, L., Baxter, R. C. & Marsh, D. J. Insulin-like growth factor binding protein-

3 inhibits migration of endometrial cancer cells. Cancer Lett. 317, 41–48 (2012).

184. Kim, J.-H. et al. Antiangiogenic antitumor activities of IGFBP-3 are mediated by

IGF-independent suppression of Erk1/2 activation and Egr-1-mediated

transcriptional events. Blood 118, 2622–2631 (2011).

185. Gaddipati, S., Estrada, K., Rao, P., Jerome, A. D. & Suvas, S. IL-2/anti-IL-2

antibody complex treatment inhibits the development but not the progression of

herpetic stromal keratitis. J. Immunol. 194, 273–82 (2015).

186. Rao, P. & Suvas, S. Development of Inflammatory Hypoxia and Prevalence of

158

Glycolytic Metabolism in Progressing Herpes Stromal Keratitis Lesions. J.

Immunol. ji1800422 (2018). doi:10.4049/jimmunol.1800422

187. Dana, M. R., Zhu, S. N. & Yamada, J. Topical modulation of interleukin-1 activity

in corneal neovascularization. Cornea 17, 403–9 (1998).

188. Yoon, K.-C. et al. Tear Production and Ocular Surface Changes in Experimental

Dry Eye after Elimination of Desiccating Stress. Investig. Ophthalmology Vis. Sci.

52, 7267 (2011).

189. De Paiva, C. S. et al. Disruption of TGF-β Signaling Improves Ocular Surface

Epithelial Disease in Experimental Autoimmune Keratoconjunctivitis Sicca. PLoS

One 6, e29017 (2011).

190. Gaddipati, S. et al. Loss of Neurokinin-1 Receptor Alters Ocular Surface

Homeostasis and Promotes an Early Development of Herpes Stromal Keratitis. J.

Immunol. 197, 4021–4033 (2016).

191. Kojima, T. et al. The effects of 3% diquafosol sodium application on the tear

functions and ocular surface of the Cu,Zn-superoxide dismutase-1 (Sod1)-

knockout mice. Mol. Vis. 20, 929–38 (2014).

192. Twardy, B. S., Channappanavar, R. & Suvas, S. Substance P in the Corneal

Stroma Regulates the Severity of Herpetic Stromal Keratitis Lesions. Investig.

Ophthalmology Vis. Sci. 52, 8604 (2011).

193. Jankovic, B. et al. Comparison between pimonidazole binding, oxygen electrode

measurements, and expression of endogenous hypoxia markers in cancer of the

uterine cervix. Cytom. Part B Clin. Cytom. 70B, 45–55 (2006).

194. Aguilera, K. Y. & Brekken, R. A. Hypoxia Studies with Pimonidazole in vivo. Bio-

159

protocol 4, (2014).

195. Coutelle, O. et al. Dual targeting of Angiopoietin-2 and VEGF potentiates effective

vascular normalisation without inducing empty basement membrane sleeves in

xenograft tumours. Br. J. Cancer 112, 495–503 (2015).

196. Girnita, A. et al. Cyclolignans as inhibitors of the insulin-like growth factor-1

receptor and malignant cell growth. Cancer Res. 64, 236–42 (2004).

197. Eltzschig, H. K. & Carmeliet, P. Hypoxia and Inflammation. N. Engl. J. Med. 364,

656–665 (2011).

198. Majmundar, A. J., Wong, W. J. & Simon, M. C. Hypoxia-Inducible Factors and the

Response to Hypoxic Stress. Mol. Cell 40, 294–309 (2010).

199. Corcoran, S. E. & O’Neill, L. A. J. HIF1α and metabolic reprogramming in

inflammation. J. Clin. Invest. 126, 3699–3707 (2016).

200. Royer, D. J. & Carr, D. J. J. A STING-dependent innate-sensing pathway

mediates resistance to corneal HSV-1 infection via upregulation of the antiviral

effector tetherin. Mucosal Immunol. 9, 1065–1075 (2016).

201. Frank, G. M., Buela, K.-A. G., Maker, D. M., Harvey, S. A. K. & Hendricks, R. L.

Early Responding Dendritic Cells Direct the Local NK Response To Control

Herpes Simplex Virus 1 Infection within the Cornea. J. Immunol. 188, 1350–1359

(2012).

202. Divito, S. J. & Hendricks, R. L. Activated Inflammatory Infiltrate in HSV-1-Infected

Corneas without Herpes Stromal Keratitis. Investig. Ophthalmology Vis. Sci. 49,

1488 (2008).

203. Giménez, F., Suryawanshi, A. & Rouse, B. T. Pathogenesis of herpes stromal

160

keratitis – A focus on corneal neovascularization. Prog. Retin. Eye Res. 33, 1–9

(2013).

204. Bryant-Hudson, K., Conrady, C. D. & Carr, D. J. J. Type I interferon and

lymphangiogenesis in the HSV-1 infected cornea – Are they beneficial to the

host? Prog. Retin. Eye Res. 36, 281–291 (2013).

205. Campbell, E. L. et al. Transmigrating neutrophils shape the mucosal

microenvironment through localized oxygen depletion to influence resolution of

inflammation. Immunity 40, 66–77 (2014).

206. Colgan, S. P. & Taylor, C. T. Hypoxia: an alarm signal during intestinal

inflammation. Nat. Rev. Gastroenterol. Hepatol. 7, 281–287 (2010).

207. Kizaka-Kondoh, S. & Konse-Nagasawa, H. Significance of nitroimidazole

compounds and hypoxia-inducible factor-1 for imaging tumor hypoxia. Cancer Sci.

100, 1366–73 (2009).

208. Ohta, A. et al. In vivo T cell activation in lymphoid tissues is inhibited in the

oxygen-poor microenvironment. Front. Immunol. 2, 27 (2011).

209. Halestrap, A. P. & Wilson, M. C. The monocarboxylate transporter family-Role

and regulation. IUBMB Life 64, 109–119 (2012).

210. Ullah, M. S., Davies, A. J. & Halestrap, A. P. The Plasma Membrane Lactate

Transporter MCT4, but Not MCT1, Is Up-regulated by Hypoxia through a HIF-1α-

dependent Mechanism. J. Biol. Chem. 281, 9030–9037 (2006).

211. Semenza, G. L. Hypoxia-Inducible Factors in Physiology and Medicine. Cell 148,

399–408 (2012).

212. Imtiyaz, H. Z. & Simon, M. C. in Current topics in microbiology and immunology

161

345, 105–120 (2010).

213. Lin, N. & Simon, M. C. Hypoxia-inducible factors: key regulators of myeloid cells

during inflammation. J. Clin. Invest. 126, 3661–3671 (2016).

214. Cramer, T. et al. HIF-1alpha is essential for myeloid cell-mediated inflammation.

Cell 112, 645–57 (2003).

215. Lee, K. et al. Acriflavine inhibits HIF-1 dimerization, tumor growth, and

vascularization. Proc. Natl. Acad. Sci. U. S. A. 106, 17910–5 (2009).

216. Borregaard, N. & Herlin, T. Energy metabolism of human neutrophils during

phagocytosis. J. Clin. Invest. 70, 550–7 (1982).

217. Pollard, T. D. & Borisy, G. G. Cellular motility driven by assembly and

disassembly of actin filaments. Cell 112, 453–65 (2003).

218. Thorens, B. & Mueckler, M. Glucose transporters in the 21st Century. Am. J.

Physiol. Metab. 298, E141–E145 (2010).

219. Macintyre, A. N. et al. The Glucose Transporter Glut1 Is Selectively Essential for

CD4 T Cell Activation and Effector Function. Cell Metab. 20, 61–72 (2014).

220. Varanasi, S. K., Donohoe, D., Jaggi, U. & Rouse, B. T. Manipulating Glucose

Metabolism during Different Stages of Viral Pathogenesis Can Have either

Detrimental or Beneficial Effects. J. Immunol. 199, 1748–1761 (2017).

221. Simpson, I. A. et al. The facilitative glucose transporter GLUT3: 20 years of

distinction. Am. J. Physiol. Metab. 295, E242–E253 (2008).

222. Pucino, V., Bombardieri, M., Pitzalis, C. & Mauro, C. Lactate at the crossroads of

metabolism, inflammation, and autoimmunity. Eur. J. Immunol. 47, 14–21 (2017).

223. Haas, R. et al. Intermediates of Metabolism: From Bystanders to Signalling

162

Molecules. Trends Biochem. Sci. 41, 460–471 (2016).

224. Benton, C. R. et al. PGC-1α increases skeletal muscle lactate uptake by

increasing the expression of MCT1 but not MCT2 or MCT4. Physiol. Genomics

35, 45–54 (2008).

225. Wise, D. R. et al. Hypoxia promotes isocitrate dehydrogenase-dependent

carboxylation of α-ketoglutarate to citrate to support cell growth and viability. Proc.

Natl. Acad. Sci. 108, 19611–19616 (2011).

226. Mullen, A. R. et al. Reductive carboxylation supports growth in tumour cells with

defective mitochondria. Nature 481, 385–388 (2012).

227. Gordan, J. D., Thompson, C. B. & Simon, M. C. HIF and c-Myc: sibling rivals for

control of cancer cell metabolism and proliferation. Cancer Cell 12, 108–13

(2007).

228. Walmsley, S. R. et al. Hypoxia-induced neutrophil survival is mediated by HIF-

1alpha-dependent NF-kappaB activity. J. Exp. Med. 201, 105–15 (2005).

229. McNamee, E. N., Korns Johnson, D., Homann, D. & Clambey, E. T. Hypoxia and

hypoxia-inducible factors as regulators of T cell development, differentiation, and

function. Immunol. Res. 55, 58–70 (2013).

230. Doedens, A. L. et al. Hypoxia-inducible factors enhance the effector responses of

CD8(+) T cells to persistent antigen. Nat. Immunol. 14, 1173–82 (2013).

231. Wu, D., Potluri, N., Lu, J., Kim, Y. & Rastinejad, F. Structural integration in

hypoxia-inducible factors. Nature 524, 303–8 (2015).

232. Schoors, S. et al. Partial and Transient Reduction of Glycolysis by PFKFB3

Blockade Reduces Pathological Angiogenesis. Cell Metab. 19, 37–48 (2014).

163

233. Zhao, T. et al. HIF-1-mediated metabolic reprogramming reduces ROS levels and

facilitates the metastatic colonization of cancers in lungs. Sci. Rep. 4, 3793

(2014).

234. Rowe, A. M. et al. Herpes keratitis. Prog. Retin. Eye Res. 32, 88–101 (2013).

235. Zheng, M., Schwarz, M. A., Lee, S., Kumaraguru, U. & Rouse, B. T. Control of

Stromal Keratitis by Inhibition of Neovascularization. Am. J. Pathol. 159, 1021–

1029 (2001).

236. Wuest, T. R. & Carr, D. J. J. VEGF-A expression by HSV-1-infected cells drives

corneal lymphangiogenesis. J. Exp. Med. 207, 101–15 (2010).

237. Yun, H., Rowe, A. M., Lathrop, K. L., Harvey, S. A. K. & Hendricks, R. L.

Reversible nerve damage and corneal pathology in murine herpes simplex

stromal keratitis. J. Virol. 88, 7870–80 (2014).

238. Meng, I. D. & Kurose, M. The role of corneal afferent neurons in regulating tears

under normal and dry eye conditions. Exp. Eye Res. 117, 79–87 (2013).

239. Dartt, D. A. & Willcox, M. D. P. Complexity of the tear film: importance in

homeostasis and dysfunction during disease. Exp. Eye Res. 117, 1–3 (2013).

240. Bron, A. J. & Tiffany, J. M. The meibomian glands and tear film lipids. Structure,

function, and control. Adv. Exp. Med. Biol. 438, 281–95 (1998).

241. Conrady, C. D., Joos, Z. P. & Patel, B. C. K. Review: The Lacrimal Gland and Its

Role in Dry Eye. J. Ophthalmol. 2016, 7542929 (2016).

242. Rolando, M. & Zierhut, M. The ocular surface and tear film and their dysfunction in

dry eye disease. Surv. Ophthalmol. 45 Suppl 2, S203-10 (2001).

243. Ruskell, G. L. Changes in nerve terminals and acini of the lacrimal gland and

164

changes in secretion induced by autonomic denervation. Z. Zellforsch. Mikrosk.

Anat. 94, 261–81 (1969).

244. Mutch, J. R. THE LACRIMATION REFLEX. Br. J. Ophthalmol. 28, 317–36 (1944).

245. Walcott, B. The Lacrimal Gland and Its Veil of Tears. News Physiol. Sci. 13, 97–

103 (1998).

246. Horibe, Y., Hosoya, K., Kim, K. J. & Lee, V. H. Carrier-mediated transport of

monocarboxylate drugs in the pigmented rabbit conjunctiva. Invest. Ophthalmol.

Vis. Sci. 39, 1436–43 (1998).

247. Cruzat, A., Pavan-Langston, D. & Hamrah, P. In Vivo Confocal Microscopy of

Corneal Nerves: Analysis and Clinical Correlation. (2010).

doi:10.3109/08820538.2010.518133

248. Ushio, A. et al. Pathological analysis of ocular lesions in a murine model of

Sjögren’s syndrome. Int. J. Mol. Sci. 18, (2017).

249. Stevenson, W. et al. Extraorbital Lacrimal Gland Excision. Cornea 33, 1336–1341

(2014).

250. Bian, F. et al. Altered balance of interleukin-13/interferon-gamma contributes to

lacrimal gland destruction and secretory dysfunction in CD25 knockout model of

Sjögren’s syndrome. Arthritis Res. Ther. 17, 53 (2015).

251. Shah, M. et al. Rapamycin Eye Drops Suppress Lacrimal Gland Inflammation In a

Murine Model of Sjögren’s Syndrome. Investig. Ophthalmology Vis. Sci. 58, 372

(2017).

252. Conrady, C. D., Joos, Z. P. & Patel, B. C. K. Review: The Lacrimal Gland and Its

Role in Dry Eye. J. Ophthalmol. 2016, 1–11 (2016).

165

253. Vijmasi, T. et al. Topical Administration of Lacritin Is a Novel Therapy for

Aqueous-Deficient Dry Eye Disease. Investig. Ophthalmology Vis. Sci. 55, 5401

(2014).

254. Hodges, R. R. & Dartt, D. A. Tear film mucins: Front line defenders of the ocular

surface; comparison with airway and gastrointestinal tract mucins. Exp. Eye Res.

117, 62–78 (2013).

255. Barbosa, F. L. et al. Goblet Cells Contribute to Ocular Surface Immune

Tolerance-Implications for Dry Eye Disease. Int. J. Mol. Sci. 18, (2017).

256. Dartt, D. A. et al. Conjunctival Goblet Cell Secretion Stimulated by Leukotrienes Is

Reduced by Resolvins D1 and E1 To Promote Resolution of Inflammation. J.

Immunol. 186, 4455–4466 (2011).

257. Stern, M. E., Gao, J., Siemasko, K. F., Beuerman, R. W. & Pflugfelder, S. C. The

role of the lacrimal functional unit in the pathophysiology of dry eye. Exp. Eye

Res. 78, 409–16 (2004).

258. Association for Research in Vision and Ophthalmology., S., Best, J. A. van, Lelij,

A. Van der & Jager, M. J. Investigative Ophthalmology & visual science.

Investigative Ophthalmology & Visual Science 43, ([Association for Research in

Vision and Ophthalmology, etc.], 2002).

259. Pitcher, J. D. et al. Pharmacological Cholinergic Blockade Stimulates

Inflammatory Cytokine Production and Lymphocytic Infiltration in the Mouse

Lacrimal Gland. Investig. Ophthalmology Vis. Sci. 52, 3221 (2011).

260. Wynn, T. A. Cellular and molecular mechanisms of fibrosis. J. Pathol. 214, 199–

210 (2008).

166

261. Sclaret, G. & Luxton, R. W. FIBROSIS OF THE THYROID AND LACRIMAL

GLANDS*. Brit. J. Ophthal 51, 173 (1967).

262. Samudre, S. et al. Lacritin, a Novel Human Tear Glycoprotein, Promotes

Sustained Basal Tearing and Is Well Tolerated. Investig. Opthalmology Vis. Sci.

52, 6265 (2011).

263. Gaddipati, S. et al. Loss of Neurokinin-1 Receptor Alters Ocular Surface

Homeostasis and Promotes an Early Development of Herpes Stromal Keratitis. J.

Immunol. 197, 4021–4033 (2016).

264. Nowell, C. S. et al. Chronic inflammation imposes aberrant cell fate in

regenerating epithelia through mechanotransduction. Nat. Cell Biol. 18, 168–180

(2016).

265. Wang, N. et al. Lacritin rescues stressed epithelia via rapid forkhead box O3

(FOXO3)-associated autophagy that restores metabolism. J. Biol. Chem. 288,

18146–61 (2013).

266. Gipson, I. K. Goblet cells of the conjunctiva: A review of recent findings. Prog.

Retin. Eye Res. 54, 49–63 (2016).

267. Gipson, I. K. & Argüeso, P. Role of mucins in the function of the corneal and

conjunctival epithelia. Int. Rev. Cytol. 231, 1–49 (2003).

268. Pflugfelder, S. C. et al. Aqueous Tear Deficiency Increases Conjunctival

Interferon-γ (IFN-γ) Expression and Goblet Cell Loss. Investig. Opthalmology Vis.

Sci. 56, 7545 (2015).

269. De Paiva, C. S. et al. Dry Eye–Induced Conjunctival Epithelial Squamous

Metaplasia Is Modulated by Interferon-γ. Investig. Opthalmology Vis. Sci. 48,

167

2553 (2007).

270. Pollard, K. M., Cauvi, D. M., Toomey, C. B., Morris, K. V & Kono, D. H. Interferon-

γ and systemic autoimmunity. Discov. Med. 16, 123–31 (2013).

271. Tseng, S. C. Staging of conjunctival squamous metaplasia by impression

cytology. Ophthalmology 92, 728–33 (1985).

272. Tseng, S. C. et al. Possible mechanisms for the loss of goblet cells in mucin-

deficient disorders. Ophthalmology 91, 545–52 (1984).

273. Hamrah, P. et al. Cellular changes of the corneal epithelium and stroma in herpes

simplex keratitis: an in vivo confocal microscopy study. Ophthalmology 119,

1791–7 (2012).

274. Lemp, M. A. Report of the National Eye Institute/Industry workshop on Clinical

Trials in Dry Eyes. CLAO J. 21, 221–32 (1995).

275. Machado, L. M., Castro, R. S. & Fontes, B. M. Staining Patterns in Dry Eye

Syndrome: Rose Bengal Versus Lissamine Green. Cornea 28, 732–734 (2009).

276. Fabiani, C., Barabino, S., Rashid, S. & Dana, M. R. Corneal epithelial proliferation

and thickness in a mouse model of dry eye. Exp. Eye Res. 89, 166–71 (2009).

277. Battat, L., Macri, A., Dursun, D. & Pflugfelder, S. C. Effects of laser in situ

keratomileusis on tear production, clearance, and the ocular surface.

Ophthalmology 108, 1230–5 (2001).

278. Schweizer, J., Langbein, L., Rogers, M. A. & Winter, H. Hair follicle-specific

keratins and their diseases. Exp. Cell Res. 313, 2010–2020 (2007).

279. Ramaesh, T. et al. Developmental and cellular factors underlying corneal

epithelial dysgenesis in the Pax6+/− mouse model of aniridia. Exp. Eye Res. 81,

168

224–235 (2005).

280. Hill, R. E. et al. Mouse Small eye results from mutations in a paired-like

homeobox-containing gene. Nature 354, 522–525 (1991).

281. Yoon, K. C. et al. Tear production and ocular surface changes in experimental dry

eye after elimination of desiccating stress. Investig. Ophthalmol. Vis. Sci. 52,

7267–7273 (2011).

282. Ouyang, H. et al. WNT7A and PAX6 define corneal epithelium homeostasis and

pathogenesis. Nature 511, 358–361 (2014).

283. Koroma, B. M., Yang, J. M. & Sundin, O. H. The Pax-6 homeobox gene is

expressed throughout the corneal and conjunctival epithelia. Invest. Ophthalmol.

Vis. Sci. 38, 108–20 (1997).

284. Maggs, D. J., Miller, P. E., Ofri, R. & Slatter, D. H. Slatter’s fundamentals of

veterinary ophthalmology. (Elsevier Saunders, 2008).

285. Potvin, R., Makari, S. & Rapuano, C. J. Tear film osmolarity and dry eye disease:

a review of the literature. Clin. Ophthalmol. 9, 2039–47 (2015).

286. McGinnigle, S., Naroo, S. A. & Eperjesi, F. Evaluation of Dry Eye. Surv.

Ophthalmol. 57, 293–316 (2012).

287. Porter, R. M. & Lane, E. B. Phenotypes, genotypes and their contribution to

understanding keratin function. Trends Genet. 19, 278–85 (2003).

288. Yoshida, S. et al. Cytokeratin 15 Can Be Used to Identify the Limbal Phenotype in

Normal and Diseased Ocular Surfaces. Investig. Ophthalmology Vis. Sci. 47,

4780 (2006).

289. Fuchs, E. Keratins as biochemical markers of epithelial differentiation. Trends

169

Genet. 4, 277–81 (1988).

290. Sun, T. T. et al. Keratin classes: molecular markers for different types of epithelial

differentiation. J. Invest. Dermatol. 81, 109s–15s (1983).

291. Kitazawa, K. et al. PAX6 regulates human corneal epithelium cell identity. Exp.

Eye Res. 154, 30–38 (2017).

292. Ou, J. et al. Chronic wound state exacerbated by oxidative stress in Pax6+/−

aniridia-related keratopathy. J. Pathol. 215, 421–430 (2008).

293. Ramaesh, T. et al. Corneal Abnormalities in Pax6 +/− Small Eye Mice Mimic

Human Aniridia-Related Keratopathy. Investig. Ophthalmology Vis. Sci. 44, 1871

(2003).

294. Matsuo, T. et al. A mutation in the Pax-6 gene in rat small eye is associated with

impaired migration of midbrain crest cells. Nat. Genet. 3, 299–304 (1993).

295. LIU, J. J., KAO, W. W.-Y. & WILSON, S. E. Corneal Epithelium-specific Mouse

Keratin K12 Promoter. Exp. Eye Res. 68, 295–301 (1999).

296. Dong, N. et al. Abnormal Epithelial Differentiation and Tear Film Alteration in

Pinguecula. Investig. Ophthalmology Vis. Sci. 50, 2710 (2009).

297. Kobayashi, A., Wajima, R., Sugiyama, K., Nonomura, A. & Huang, A. J. W.

Idiopathic Limbal Squamous Metaplasia. Arch. Ophthalmol. 121, 1473 (2003).

298. Royer, D. J., Elliott, M. H., Le, Y. Z. & Carr, D. J. J. Corneal Epithelial Cells Exhibit

Myeloid Characteristics and Present Antigen via MHC Class II. Invest.

Ophthalmol. Vis. Sci. 59, 1512–1522 (2018).

299. Baxter, R. C. IGF binding proteins in cancer: mechanistic and clinical insights.

Nat. Rev. Cancer 14, 329–341 (2014).

170

300. Baxter, R. C. Circulating binding proteins for the insulin-like growth factors.

Trends Endocrinol. Metab. 4, 91–6 (1993).

301. Butler, A. A. et al. Insulin-like growth factor-I receptor signal transduction: at the

interface between physiology and cell biology. Comp. Biochem. Physiol. B.

Biochem. Mol. Biol. 121, 19–26 (1998).

302. Lee, O. H. et al. Identification of angiogenic properties of insulin-like growth factor

II in in vitro angiogenesis models. Br. J. Cancer 82, 385–391 (2000).

303. Bjorndahl, M. et al. Insulin-like growth factors 1 and 2 induce lymphangiogenesis

in vivo. Proc. Natl. Acad. Sci. 102, 15593–15598 (2005).

304. Smith, L. E. H. et al. Regulation of vascular endothelial growth factor-dependent

retinal neovascularization by insulin-like growth factor-1 receptor. Nat. Med. 5,

1390–1395 (1999).

305. Kooijman, R., Coppens, A. & Hooghe-Peters, E. IGF-I Inhibits Spontaneous

Apoptosis in Human Granulocytes. Endocrinology 143, 1206–1212 (2002).

306. Baxter, R. C. Insulin-like growth factor binding protein-3 (IGFBP-3): Novel ligands

mediate unexpected functions. J. Cell Commun. Signal. 7, 179–89 (2013).

307. Chang, K.-H. et al. IGF binding protein-3 regulates hematopoietic stem cell and

endothelial precursor cell function during vascular development. Proc. Natl. Acad.

Sci. 104, 10595–10600 (2007).

308. Wilhelmus, K. R. Diagnosis and management of herpes simplex stromal keratitis.

Cornea 6, 286–91 (1987).

309. GRANATA, R. et al. Insulin-like growth factor binding protein-3 induces

angiogenesis through IGF-I- and SphK1-dependent mechanisms. J. Thromb.

171

Haemost. 5, 835–845 (2007).

310. Oh, S.-H. et al. Insulin-like growth factor binding protein-3 suppresses vascular

endothelial growth factor expression and tumor angiogenesis in head and neck

squamous cell carcinoma. Cancer Sci. 103, 1259–66 (2012).

311. Kondo, T. et al. Knockout of insulin and IGF-1 receptors on vascular endothelial

cells protects against retinal neovascularization. J. Clin. Invest. 111, 1835–1842

(2003).

312. Lee, C. Y. & Rechler, M. M. Proteolysis of insulin-like growth factor (IGF)-binding

protein-3 (IGFBP-3) in 150-kilodalton IGFBP complexes by a cation-dependent

protease activity in adult rat serum promotes the release of bound IGF-I.

Endocrinology 137, 2051–2058 (1996).

313. Ning, Y. et al. Diminished Growth and Enhanced Glucose Metabolism in Triple

Knockout Mice Containing Mutations of Insulin-Like Growth Factor Binding

Protein-3, -4, and -5. Mol. Endocrinol. 20, 2173–2186 (2006).

314. Piñeiro-Hermida, S. et al. IGF1R deficiency attenuates acute inflammatory

response in a bleomycin-induced lung injury mouse model. Sci. Rep. 7, 4290

(2017).

315. Duan, Z. et al. Insulin-like growth factor-I receptor tyrosine kinase inhibitor

cyclolignan picropodophyllin inhibits proliferation and induces apoptosis in

multidrug resistant osteosarcoma cell lines. Mol. Cancer Ther. 8, 2122–2130

(2009).

316. Girnita, A. et al. Cancer Research. Cancer Res. 60, 4152–4160 (2004).

172

ABSTRACT

ALTERATIONS IN OCULAR SURFACE SYSTEM DURING THE PATHOGENESIS OF HERPES STROMAL KERATITIS

PUSHPA DURGESH RAO

May 2019

Advisor: Dr. Susmit Suvas

Major: Anatomy and Cell Biology

Degree: Doctor of Philosophy

Herpes stromal keratitis (HSK) is a chronic immuno-inflammatory ocular disease caused by Herpes simplex virus-1 (HSV-1) infection in the cornea. HSK is characterized by the development of corneal opacity and angiogenesis accompanied by the loss of corneal transparency. Despite extensive studies on inhibiting corneal angiogenesis and strategies to reduce HSK disease development, a key underlying mechanism to maintain an intact and a transparent corneal surface during HSK disease progression, remain obscure. Our study has addressed possible causes for the HSK disease severity by identifying the key factors such as the development of inflammatory hypoxia, reduced tear volume, Inflammation of extra-orbital lacrimal gland and the conjunctiva and the transdifferentiation of the corneal epithelium in the HSK developing eyes.

Intriguingly, our study also emphasizes the protective role of IGFBP-3 in inhibiting the severe disease progression of HSK.

During the pre-clinical and the clinical disease phase of HSK, the influx of neutrophils have been studied as the most prominent innate immune cell type.

173

Development of corneal hem- and lymph-angiogenesis, opacity and epithelial defects hinders the clear vision and is known to be the cause of the increased neutrophil influx in HSV-1 infected cornea. Previous studies have indicated that the influx of neutrophils into mucosal tissue can re-model the development of inflammatory hypoxia. Therefore, the aim of this study was to investigate the development of hypoxia and hypoxia- associated gene expression during the progression of HSK disease. Our results showed an increased influx of neutrophils during the clinical disease of HSK that leads to the development of hypoxia, and the upregulation of hypoxia-associated genes. This study provided us novel insights on the prevalence of the glycolytic metabolism in the HSK developing eyes. Additionally, our data demonstrated that the blocking of Hypoxia-

Inducible Factor (HIF) with Acriflavine alone might not be an efficacious drug in reducing

HSK disease severity. Overall, these results pave way for using novel therapeutic targets that play a dual role to block HIF and control corneal inflammation during HSK.

Maintaining a healthy corneal epithelium, retaining a steady tear volume and its secretion could be the other factors that may aid in alleviating HSK disease severity.

Mounting evidence corroborates corneal nerves as a critical factor in the maintenance of corneal integrity and homeostasis of the ocular surface by mediating the reflex actions of blinking and tearing. In support, in a mouse model during HSV-1 ocular infection, corneal nerve damage has been demonstrated. Hence, corneal transparency and clear vision are critically dependent upon the integrity of corneal tissue maintained by the intricate network of corneal nerves, the presence of intact corneal epithelium and tear film that is controlled by a cross-talk between the components of the lacrimal functional unit. Our study showed a reduced measurement of tear volume in the HSK

174 developing eyes correlating to the inflammation in the extra-orbital lacrimal gland

(EoLG) and the conjunctiva. Atrophy of tear secreting acinar cells in the EoLG, the significant loss of goblet cells in the conjunctiva and the influx of leukocytes and CD4 T cells were the major events occurring during the HSV-1 infection. These events caused dysfunction to the lacrimal functional unit, thereby inducing severe HSK and ‘dry eye’ conditions. Nevertheless, topical treatment of lacritin (tear glycoprotein) delayed but did not prevent the reduction in tear volume nor reduced the severity of HSK.

Existence of a healthy, smooth and lubricated corneal surface is critical for the maintenance of transparent cornea to aid clear vision. Corneal epithelium comprises of non-keratinized stratified squamous epithelium. Nevertheless, under pathological conditions non-keratinized epithelium transforms into a keratinized epithelium that closely matches the features of skin epithelium (epidermis). The keratin expression profile determines this switch in keratinization in the cornea. Cytokeratin-12 (K12) is one of the keratins specific to the corneal epithelium, which is replaced by epidermis-specific cytokeratin-10 (K10) expression under pathological conditions. Intriguingly, our data showed the loss of K12 expression with the emergence of K10 expression during the clinical stage of HSK disease. Our data suggested the occurrence of corneal epithelial transformation. Another prominent marker, PAX-6 (paired box -6 transcription factor) has a crucial role in the maintenance of corneal epithelial identity by regulation of its differentiation. Our data showed a dramatic loss of PAX-6 expression in the HSK developing eyes. ‘Squamous metaplasia’ is the pathological phenomenon characterized by abnormal differentiation of corneal epithelium into the skin like epithelium. Overall,

175 our results signified the ‘trans-differentiation of corneal epithelial cells to the skin like epithelium, in HSV-1 infected mouse cornea.

Corneal hem-angiogenesis is one of the hallmarks of HSK. Hence, it is essential to understand better the molecules involved in regulating hem-angiogenesis, immune cell survival, and its effector function. Intriguingly, angiogenesis array data showed an increased expression of IGFBP-3 in the HSV-1 infected corneas of C57BL/6 (B6) mice.

Therefore, we were inclined to study the role of IGFBP-3 molecule in HSV-1 infected B6 corneas. IGFBP-3 is one of the six highly conserved IGF-binding proteins. IGFBP-3 binds to 75% of the serum IGF-I and IGF-II in complexes and regulates IGF-1R signaling by sequestering IGF molecule. IGF binding to IGF-1R results in phosphorylation of tyrosine residues in IGF-1Rβ chain. IGF-1R signaling that has been reported to promote angiogenesis (hem- and lymph-angiogenesis). Our data indicated severe HSK disease with robust angiogenesis in IGFBP-3-/- mice associated with a significant increase in IGF-1R phosphorylation in the infiltrated leukocytes. These results suggest the protective role of IGFBP-3 in inhibiting IGF-1R phosphorylation that leads to the development of severe HSK disease progression.

Altogether, this study broadens our knowledge on understanding several factors involved in the alterations of the ocular surface system during the pathogenesis of HSK. In summary, this study highlights on five crucial factors that may play an essential role in reducing HSK disease pathogenesis in the cornea; 1) regulating the development of inflammatory hypoxia 2) maintaining of the stable tear volume 3) reducing the lacrimal and conjunctival inflammation 4) maintaining the corneal epithelial integrity 3) alleviating corneal hem-angiogenesis (Figure 42).

176

AUTOBIOGRAPHICAL STATEMENT Education

2014-2019 Wayne State University School of Medicine, Detroit, MI, USA Doctor of Philosophy, Anatomy and Cell Biology

2008-2011 University of Connecticut Health, CT, USA Master of Science, Biomedical Sciences

2000-2002 Bengaluru University, Karnataka, India Bachelor of Science, Microbiology

Honors and Awards

2018 Short Talk selected from poster presentation and Travel Grant Awarded, Gordon Research Conference, Cornea and Ocular Surface Biology and Pathology 2017 Thomas C. Rumble Fellowship, Wayne State University School of Medicine 2017 Travel Grant, ARVO Foundation for Eye Research 2016 Third Best Poster Presentation with an Award Certificate, 3rd Annual Kresge Eye Institute Vision Research Workshop, Wayne State University School of Medicine 2015 Second Best Oral Presentation Award, 2nd Annual Kresge Eye Institute Vision Research Workshop, Wayne State University School of Medicine

Publications

Rao P, Suvas S. Development of Inflammatory Hypoxia and Prevalence of Glycolytic Metabolism in Progressing Herpes Stromal Keratitis Lesions. J Immunol. 2019 Jan 15;202(2):514-526.

Rao P, Suvas S. Development of Extra-orbital Lacrimal Gland Inflammation in the Mouse Model of Herpes Stromal Keratitis. Exp Eye Res. Manuscript under review.

Rao P, Suvas S. Role of Insulin-like Growth Factor Binding Protein-3 in the Pathogenesis of HSK. Manuscript in preparation. Gaddipati S, Rao P, Jerome AD, Burugula BB, Gerard NP, Suvas S. Loss of Neurokinin-1 Receptor Alters Ocular Surface Homeostasis and Promotes an Early Development of Herpes Stromal Keratitis. J Immunol. 2016 Nov;197(10):4021-33.

Gaddipati S, Estrada K, Rao P, Jerome AD, Suvas S. IL-2/anti-IL-2 antibody complex treatment inhibits the development but not the progression of herpetic stromal keratitis. J Immunol. 2015 Jan 1;194(1):273-82.