3. UNPUBLISHED RESEARCH PPIP5K2 Mutations in Human

IN SEARCH OF GENETIC MUTATIONS FOR FAMILIAL

KERATOCONUS

Mariam Lotfy Khaled

Submitted to the Faculty of the Graduate School

of Augusta University in partial fulfillment

of the Requirements of the Degree of

(Doctor of Philosophy)

April

2019

ACKNOWLEDGEMENTS

First and foremost, I would like to express my gratitude to God Almighty for the blessing, kindness, and inspiration in lending me to accomplish my research study.

I would like to express my sincere gratitude and appreciation to my advisor Dr. Yutao Liu for providing his heartfelt support, guidance, patience, motivation, and immense knowledge. I could not have imagined having a better advisor and mentor for my Ph.D. study.

Besides my advisor, I would like to thank the rest of my thesis committee: Prof. Ruth

Caldwell, Prof. Zheng Dong, Prof. Mitchell Watsky, and Dr. Amy Estes, for their insightful comments and encouragement, and for the insightful questions and encouragement to widen my research from various perspectives. I am also grateful to Dr. Ming Zhang for his kindness for acceptance to serve as my thesis reader.

My sincere thanks also goes to the CBA Department Chair Prof. Sylvia Smith, and the Dean of the Graduate School, Prof. Mitchell Watsky, who gave access to their laboratory and research facilities. Without their precious support it would not be possible to conduct this study.

I thank my previous and current fellow lab-mates especially Jingwen Cai and Dr. Inas

Helwa for offering me help whenever I asked, for the stimulating discussions during our lab meetings, and for all the fun we have had since joining Dr. Liu’s lab.

I am also grateful to the Graduate School representatives: Prof. Mitchell Watsky, Prof.

Patricia Cameron, Prof. Mark Hamrick and Marvis Baynham, and the CBA department staff: Diane Riffe, Lynda Cotton Mitchell, Heide Andrews, Donna Tuten, and Sharon Lever, for their great support all the time. I am grateful for the NIH/NEI for funding my PhD research.

I would like to express my deepest gratitude to the people who mean a lot to me, my parents, Lotfy Khaled and Enayate Ibrahim, whose selfless sacrificial life and their great efforts with pain, tears, and unceasing prayers have enabled me to accomplish every achievement I earned since being little girl. I would never be able to pay back the love and affection showered upon by my parents. Also I am extremely grateful to my sisters Hager, and Salma for their selfless love, care, support and valuable prayers.

I would also like to extend my deepest gratitude to a very special person, my husband,

Mohammed Ibrahim for his continuous and unfailing love, support and understanding. He was always around at times I thought that it was impossible to continue. He helped me keep things in perspective. I greatly value his contribution and deeply appreciate his belief in me. I deeply appreciate my lovely daughters, Sarah, and Zaina. You have made me stronger, better and more fulfilled than I could have ever imagined. Words would never say how grateful I am to all my family. I consider myself the luckiest in the world to have such a lovely and caring family, standing beside me with their warm love and endless support. And finally, I would like to thank all my friends in USA and in Egypt for all their support, care and encouragements.

ABSTRACT

MARIAM LOTFY KHALED In search of genetic mutations for familial keratoconus (Under the direction of YUTAO LIU) Keratoconus (KC) is the most common corneal degenerative disorder and a leading cause of corneal transplantation in developed countries. KC is a multi-factorial disease with involvement of genetic, environmental, and hormonal factors. Although KC has been widely studied, the main cause of the disease and the molecular mechanism remain unknown. We aimed to study the molecular genetics of KC via utilizing next-generation sequencing technology including RNA-Seq, whole exome sequencing, and whole genome sequencing. We used RNA-Seq to study the KC-affected corneal transcriptome. We identified 436 coding RNAs and 584 lncRNAs with differential expression in the KC- affected corneas with a |fold change| ≥ 2 and a false discovery rate ≤ 0.05. Pathway analysis, using WebGestalt, indicated the enrichment of the genes involved in the extracellular matrix, protein binding, glycosaminoglycan binding, and cell migration. Co- expression analysis revealed 296 pairs of genes with significant KC-specific correlations.

The RNA-Seq data analysis highlighted the potential roles of several genes (CTGF,

SFRP1, AQP5, lnc-WNT4-2:1, and lnc-ALDH3A2-2:1) and pathways (TGF-β, WNT signaling, and PI3K/AKT pathways) in KC pathogenesis. Next, we used whole genome and exome sequencing to figure out the causal mutation(s) in a four-generation KC family with a linkage locus on Chr5q14.3-q21.1. We found a missense mutation in the phosphatase domain of PPIP5K2 (c.1255T>G, p.Ser419Ala). We found another missense

mutation in the same domain of PPIP5K2 (c.2528A>G, p.Asn843Ser) in a second KC family. PPIP5K2 is a bifunctional enzyme involved in the inositol phosphate metabolic pathway. In vitro functional assays indicated the impact of the identified mutations on the enzymatic activity of PPIP5K2. PPIP5K2 expresses at a higher level than its homolog

PPIP5K1 in both human and mouse corneas. A transgenic mouse model with the loss of phosphatase activity and elevated kinase activity of Ppip5k2 exhibited corneal structural abnormalities emphasizing the important role of PPIP5K2 in the homeostasis of corneal integrity. This study advances our knowledge of KC genetic etiology and helps in identifying a potential therapeutic target for KC.

KEY WORDS: (Keratoconus, Cornea, WGS, WES, RNA-Seq, PPIP5K2)

Table of Contents

Chapter Page

1 Introduction ...... 8

A. Statement of the problem and specific aims ...... 8

B. Significance and innovation ...... 9

C. Rational of our study ...... 10

D. Literature review ...... 11

2 Differential Expression of Coding and Long Noncoding RNAs in Keratoconus- affected Corneas (published manuscript) ...... 21

3 PPIP5K2 mutations in human familial keratoconus and mouse cornea

(unpublished research) ...... 50

4 Discussion ...... 83

5 Summary ...... 94

References ...... 96

List of Tables

Table 1: SNPs associated with keratoconus identified through genome-wide association studies ...... 16

Table 2: Clinical phenotypes of patients with KC and normal controls ...... 41

Table 3: Top coding genes with significant differential expression in KC-affected corneas

...... 42

Table 4: Twenty one differentially expressed coding genes to be validated using ddPCR.

...... 43

Table 5: Seven differentially expressed lncRNAs to be validated using ddPCR ...... 45

Table 6: Coding and long noncoding RNAs with significant differences in expression correlation (with a reasonable corneal expressions) in KC-affected corneas versus the controls...... 46

Table 7: Variant filtering pipeline for whole genome data using the SVS software...... 58

List of Figures

Figure 1: Optical coherence tomography scans of keratoconic patients with various stages of severity...... 8

Figure 2: A potential physiological model for the pathogenesis of keratoconus...... 11

Figure 3: Keratoconus genes and their involvement in other ocular diseases ...... 19

Figure 4: Differentially expressed coding genes in kc-affected human corneas ...... 47

Figure 5: Differentially expressed long non-coding RNAs (lncRNAs) in keratoconus- affected human corneas ...... 48

Figure 6: Correlation networks of the differentially expressed coding RNAs and lncRNAs displayed by Cytoscape...... 49

Figure 7: A four-generation KC family with autosomal dominant inheritance...... 53

Figure 8: The pedigree structure of a second multiplex family affected with KC...... 54

Figure 9: Variant filtering pipeline for whole exome sequencing data using the SVS software ...... 57

Figure 10: PCR-based Sanger sequencing of the identified mutations...... 67

Figure 12: (A) Inositol pyrophosphate metabolism……………………………………. 68

Figure 11: Biochemical properties of the two missense mutations ...... 70

...... 72

Figure 14: SD-OCT images of mouse anterior chamber. (A) Allele map of the

Ppip5k2K^ mice. (B)Wild-Type. (C) Ppip5k2+/K^ mouse. (D) Ppip5k2K^/K^ mouse. Arrows pointed to the corneal surface irregularities...... 74

Figure 15: Mice corneal assessment...... 75

Figure 16: H&E of the mice corneas...... 76

Figure 17: Mouse corneal curvature and thickness mapping...... 77

Figure 18: Mouse Slit lamp Examination...... 78

Table of Abbreviations:

Abbreviation Term CCT Central Cornea Thickness ddPCR Droplet Digital PCR DZ Dizygotic ECM Extracellular matrix HCEC Human corneal Epithelial Cell HCSF Human Corneal Stromal Fibroblast HD Huntington disease ICRS intrastromal corneal ring segment KC Keratoconus LncRNA Long noncoding RNA mtDNA Mitochondrial DNA MYOC Myocillin MZ Monozygotic NGS Next-Generation Sequencing PBCL Piggy Back Contact Lenses PCSK1 Proprotein Convertase Subtilisin/Kexin Type 1 PD Parkinson’s disease PPIP5K Diphosphoinositol Pentakisphosphate Kinases RGP Rigid Gas Permeable RNA-Seq RNA Sequencing ROS Reactive Oxygen Species SD-OCT Spectral Domain-Optical Coherence Tomography UV-CXL Ultra-Violet Corneal Cross-Linking WES Whole Exome Sequencing WGS Whole Genome Sequencing NONHSAT NONCODE Transcript ID for lncRNAs

1. INTRODUCTION

A. Statement of the problem and specific aims

Our main goal is to identify and characterize the genetic mutations and molecular pathways involved in the pathogenesis of keratoconus (KC). KC is the most common primary corneal ectasia, affecting 300,000 individuals in the US. It is a leading indicator of corneal transplantation in Western countries. It is a bilateral and asymmetric corneal degeneration characterized by localized corneal thinning, which leads to protrusion of the thinned corneal region. Corneal protrusion causes myopia and irregular astigmatism, leading to impairment of visual acuity. Both genetic and environmental factors contribute to the etiology of KC. Evidence of genetic contribution includes familial clustering and concordance between monozygotic twins. Genome-wide linkage studies in multiplex KC families have identified 20 chromosomal loci linked to KC. Among them, only the chr5q locus has been replicated in different studies. Chr5q14.3-21.1 locus has been identified in a four-generation Caucasian family; however, the exact causative mutation remains unknown. Moreover, many researchers have performed differential expression analyses, using RT-PCR or microarray, to identify the related genes and molecular pathways in keratoconic corneas. Next-generation sequencing has been widely used to profile DNA sequence (whole genome/exome sequencing) and RNA expression (RNA-Seq). It is necessary to apply next-generation sequencing in KC to identify DNA mutations and the underlying genomic mechanisms. In order to accomplish these goals, we propose to perform the following two specific aims:

Specific Aim 1: To determine differentially expressed genes in KC-affected human corneas We will isolate total RNA from 10 keratoconic corneas and 8 unaffected postmortem donor corneas and perform stranded Total RNA-Seq to determine the differentially expressed genes in keratoconic corneas. We will validate the differential expression of selected genes using droplet digital PCR.

Specific Aim 2: To identify and characterize genetic mutations in multiplex KC families We will identify the causative mutation in a four-generation pedigree with a pre-identified linkage region on chromosome 5 (5q14.3-21.1) using whole genome and whole exome sequencing. We will characterize the identified mutation using biochemical assays, cell culture assays, and related mouse models.

B. Significance and innovation

KC is the most common corneal ectatic disorder and a leading indicator for corneal transplantation in the developed countries 1; 2 KC usually starts at puberty and progresses until it stabilizes in the third or fourth decade 3; 4. KC is a complex multifactorial disorder.

Both genetic and environmental factors (i.e. nutrition, eye rubbing, atopy, and hormones) play important roles in the pathogenesis of KC 5; 6. The lifetime cost to treat KC-related refractive error averages $25,000 per patient in the US 7. KC patients report significant impaired vision-related quality of life which worsens with time due to disease progression

8. KC is recognized in the Strategic Plan of the National Eye Institute (NEI) at the National

Institute of Health (NIH), indicating the national priority and necessity for research in this field. The identification of causative mutations provides not only critical information regarding the pathogenesis of KC but also potential targets for novel treatment

opportunities. In this study, we utilize high-throughput whole exome and whole genome sequencing to search for genetic mutations in a four-generation multiplex KC family and characterize the mutations using biochemical assays, corneal cell culture, and mouse models. Thus, our study aligns well with the strategic goal of NIH/NEI.

Successful completion of our study significantly improves our understanding of the underlying genetic etiology of KC. Identification of novel mutations and the establishment of the first mouse model for KC can promote novel targets for KC diagnosis and treatment to reduce the disease burden for patients and society.

C. Rational of our study

Our aims are to identify the genetic basis and characterize molecular pathway involved in

KC. In this thesis, we have used whole genome sequencing and whole exome sequencing to identify the causative genetic mutations in multiplex KC families. We hypothesize that the mutations in these families cause disruption in protein-coding genes, leading to the pathological corneal phenotypes. We also use RNA-sequencing approach to identify the differentially expressed genes using keratoconic and healthy corneas. We use the differentially expressed genes in the identification of candidate genes and molecular pathways involved in the disease pathogenesis. In addition, we seek to use these candidate genes in our search for the causative mutation(s) in KC-affected families.

D. Literature review

1. Human cornea

The human cornea is a transparent, avascular, and precisely structured tissue. It accounts for 70% of the eye's refractive power and acts as a tight protective barrier 9; 10. The average thickness of the cornea is approximately 0.5 mm at the center with a gradual thickness increase towards the periphery 11; 12. The cornea receives its nutrients and anti-infectious factors anteriorly through the tear film and posteriorly through the aqueous humor 13; 14.

The cornea receives oxygen from tears which then diffuses throughout the corneal layers15. Cornea consists of five consecutive layers: epithelium, Bowman’s layer, stroma,

Descemet’s membrane, and endothelium 16; 17. The corneal epithelium is a non- keratinized stratified squamous interface with 5 to 7 arranged layers of cells (superficial cells, wing cells, basal cells, and basement membrane), and accounts for 10% of the cornea thickness 10. Bowman’s layer is an acellular fibrous meshwork between epithelium and stroma 18. Stroma is a precisely organized collagenous matrix consisting of multiple collagenous lamellae and keratocytes and accounts for approximately 80% of the corneal thickness 19. This organized architecture is crucial for the transparency of the cornea. Any disruption in its organization may lead to an opaque cornea 20 and affect visual acuity.

Stromal keratocytes are mesenchymal cells important in maintaining extracellular matrix

(ECM) environment and stromal homeostasis 21. Descemet’s membrane is an acellular collagenous layer, which acts as the endothelium basement membrane 22; 23. The endothelium is a monolayer of regularly sized polygonal cells. Apart from oxygen, the cornea obtains its nutrients from aqueous humor through the endothelium. Additionally, the endothelium regulates stromal hydration through active transport mechanisms 24. The

stroma is a highly hydrophilic collagenous layer. Without the endothelium, the cornea rapidly swells and becomes opaque 25.

2. Keratoconus

Keratoconus (KC) is the most common corneal ectasia and a leading cause of corneal transplantation in the developed countries 1; 2. KC is a multi-factorial disease with the involvement of genetic, nutritional, environmental, and hormonal factors 26. Although KC was first described in 1748, the exact cause still remains a mystery for most patients. The onset of KC is often at puberty and the disease progresses and stabilizes in the third or fourth decade 3; 4. KC patients report significantly impaired vision-related quality of life, which deteriorates with progression 8. KC manifests in all ethnicities and has no gender preference 27-29. KC prevalence varies greatly worldwide, ranging from 0.0003% in

Russia 30, 0.086% in Denmark 31, 0.249 % in Iran 32, to 2.3% in central India 33.

2. 1. Clinical signs and diagnosis

KC starts silently with a subclinical manifestation known as forme fruste (French term means incomplete form) . Its primary symptoms include photophobia, monocular diplopia, glare, and a reduction in visual acuity, which cannot be corrected to 20/20 with spectacles only 27. Clinically, KC is diagnosed by using slit lamp biomicroscopy and corneal topography/tomography 6; 27; 34. Various clinical signs of KC could be identified with slit lamp such as Fleischer’s ring (iron lines partially or completely surrounding the cone),

Vogt's striae (fine vertical lines in deep stroma and Descemet’s membrane), localized corneal thinning, and Münson’s sign (bulging of the lower lid during downgaze) 16.

Recently, the tomographic analysis is used to evaluate the whole cornea (anterior and posterior corneal surfaces and corneal thickness), iris, and lens, through the construction

of a three-dimensional image of the anterior segment 6; 35; 36. Anterior segment optical coherence tomography (OCT) is one of the imaging methods which are frequently used for

KC diagnosis 37.

2. 2. Treatment modalities

The treatment of KC varies according to the disease stage and severity including non- surgical and surgical treatments 38. Spectacles and contact lenses are the early treatment options to correct the visual acuity. These contact lenses include soft lenses, rigid gas permeable (RGP) lenses, hybrid lenses, piggy back contact lenses (PBCL), and scleral lenses 39. The most frequently used one is the RGP lens, as it allows passage of oxygen to the cornea. However, the scleral lens is usually a better option for patients who cannot tolerate other contact lenses. The scleral lens is a large diameter RGP lens, which rests on the sclera without touching the cornea and limbus.

When non-surgical options become ineffective in KC management, surgical intervention becomes necessary to restore the patient's vision. There are different types of surgeries, among which, physicians can determine based on the severity of the disease. Patients with moderate KC might need the insertion of an intrastromal corneal ring segment (ICRS), ultra-violet corneal cross-linking (UV-CXL), or Bowman’s layer transplantation. In severe cases, corneal transplantation is required, either by removal of the whole corneal tissue

(known as penetrating keratoplasty) or removal of only the upper corneal layers and keeping the endothelium (known as deep anterior lamellar keratoplasty) 40. Preserving the patient’s endothelium decreases the risk of grafting rejection 41. For ICRS, two hemispherical segments are inserted into stromal tunnels, which can correct corneal ectasia via shortening the cone length, causing corneal flattening to the periphery. However, ICRS

may not stop the disease progression completely 42. UV-CXL is a photo-oxidative procedure that involves utilizing of riboflavin and ultraviolet-A rays to increase corneal biomechanical resistance, and to stop the progression of KC 43; 44.

2. 3. KC- associated histopathological changes

KC-associated histopathological changes have been shown in different corneal layers except for the endothelium. Light and electron microscopes can be used to examine fixed corneas 45-47 while confocal microscopy and OCT are used to look at corneas in vivo 46;

48-51. Stromal thinning is the main morphological manifestation of KC. However, KC- affected corneas may exhibit other abnormalities including 1) enlargement, irregular arrangement, and a significant reduction in basal epithelial cell density 47; 52-54, 2) increased visibility of the nerve fibers on slit lamp examination 55; 56, 3) ruptured

Bowman’s layer 47; 57, 4) significantly reduced number of collagenous lamellae of stroma

58; 59, 5) significantly decreased density of anterior and posterior keratocytes 45; 54; 57;

60; 61, and 6) disrupted Descemet’s membrane 62. Based on the corneal structural changes identified from OCT, KC can be classified into five stages (Figure 1) 50. In stage 1, cornea shows a thinner epithelium and stroma at the cone. In stage 2, the Bowman’s layer exhibits hyperreflective anomalies with epithelium thickening and stromal opacity. In stage 3, the epithelium is thickened while the stroma is thinned with a disrupted Bowman’s layer. Stage

4 shows pan-stromal scarring, and ultimately, stage 5 presents a rupture in the Descemet’s membrane with the formation of hydrops (an acute form of keratoconus), leading to total corneal scarring (Figure 1, stage 5).

Figure 1: Optical coherence tomography scans of keratoconic patients with various stages of severity. All scale bars represent 250 μm. Adopted from Sandali et al.50

2. 4. KC-associated molecular mechanisms

Genetic, environmental and hormonal factors may interact with each other through different molecular pathways and ultimately lead to the KC–associated corneal pathology.

As illustrated in figure 2, there are three main suggested mechanisms to explain the structural and biochemical abnormalities in KC.

The first mechanism is the disturbed equilibrium between degradative enzymes and their inhibitors, leading to stromal thinning 63-65. Elevated expression of multiple cellular and collagenous degradative enzymes have been reported in KC-affected epithelial cells, keratocytes, and patient's tears 66; 67. These enzymes include matrix metalloproteinases

(MMP-1, -3, -7, -9 and -13), and lysosomal enzymes (acid esterases, acid phosphatases, acid lipases, and cathepsins) 63; 66; 68. At the same time, their inhibitors have shown lower expression levels, including α1-protease inhibitor, α2-macroglobulin, and tissue inhibitors of MMP 65; 69; 70. Moreover, elevated levels of various cellular adhesion molecules

(laminin, fibronectin, and desmoglein 3) and inflammatory mediators (IL-4, IL-5, IL-6, IL-

8, TNF-α, and TNF-β) have been reported in KC-affected corneas 28; 71-73.

These findings also lead to the second mechanism - the abnormal distribution in corneal wound healing cascade 74. Corneal wound healing is a tangled process involved with multiple growth factors, cytokines, and proteases produced by epithelial cells, stromal keratocytes, and inflammatory cells 72. Upon an epithelial injury, multiple cytokines are released to induce consequential actions in an attempt to repair the epithelial defect, regulate the stroma remodeling, and minimize loss of transparency and function.

The third mechanism is corneal oxidative stress in KC-affected corneas. Reduced levels of various cellular antioxidants have been identified in KC, such as glutathione, paraoxonase1, catalase, superoxide dismutase and superoxide glutathione 75; 76. The elevated oxidative stress in human corneas may damage the mitochondria and its DNA

(mtDNA) in KC-affected corneal tissues and stromal fibroblasts 77; 78. Various studies have found increased ROS production and lower ATP level in the diseased cornea 79; 80.

In addition to the aforementioned mechanisms, hormonal factors may play a role in KC development 16; 81; 82. Sex hormones receptors are widely expressed in different ocular tissues, including the cornea. Pregnancy has been postulated as a risk factor for KC because many pregnant women with KC have experienced a significant progression with severe clinical phenotypes 83; 84. Moreover, receiving an in vitro fertilization treatment have been found to worsen KC, probably due to an increase in their estrogen levels 85. All these proposed mechanisms indicate that KC is a degenerative disease and various factors may influence the disease onset and progression (Figure 2).

Figure 2: A potential physiological model for the pathogenesis of keratoconus. Adopted from Khaled et al.16.

3. Genetics of KC

Multiple studies have indicated the genetic contribution in KC pathogenesis 86. Familial inheritance was reported in 6 - 23.5% of KC patients. First-degree relatives are at risk of developing KC 15–67 times higher than the general population 87. Autosomal dominance is the most common pattern of inheritance (reviewed in 88). Higher concordance of KC severity has been found in monozygotic (MZ) twins than dizygotic twins (DZ) 86; 89; 90.

MZ twins are genetically identical, while DZ twins are developed from two different fertilized eggs, thus they are concordant in half of the alleles 91; 92. Similar concordance should be anticipated in MZ and DZ twins if environmental factors are the sole contributor to KC pathogenesis. Another piece of evidence is from the significantly elevated KC risk from the consanguineous marriage (first-cousin marriage) 93; 94.

KC could be presented as secondary to other systemic disorders in patients. KC is more prevalent in patients with chromosomal abnormality disorders including Down syndrome

(trisomy 21) and 22q11.2 deletion syndrome 16; 95-97. Additionally, multiple genetically inherited diseases have been shown to be associated with KC such as Leber congenital amaurosis 98; 99, cataract 100, retinal cone dystrophy 101, Ehlers–Danlos syndrome, and osteogenesis imperfecta 102; 103.

Despite the extensive human genetics research in KC patients, it remains unclear how these genetic factors cause/contribute to KC onset and progression. Wang et al. have postulated that KC follows Mendel's law and is inherited due to a major gene defect 104. On the other hand, other studies suggested a non-Mendelian pattern of KC due to the disease’s complexity, and the interaction of multiple genes and environmental and metabolic factors

104; 105. Genome-wide linkage and association studies are the primary tools to identify variants for Mendelian and complex diseases, respectively. Linkage analysis is to identify the genomic locus containing disease-causal mutations while genome-wide association studies (GWAS) are to identify the genetic variants associated with disease risk 106.

3. 1. Genome-wide linkage Analyses in KC

Family-based linkage analyses have identified a number of KC-linked chromosomal loci

107; 108. In a Northern Irish family, chr15q22.32-24.2 locus was found to be segregating with KC and congenital cataract. After ten years, the researcher group successfully identified a mutation in the MIR184 seed region (c.57C>T) 100; 109; 110. This long period of time highlights the complexity of identifying the causative genetic mutation within the linkage locus. Another linkage locus, which has been interestingly replicated in two different studies, is 5q21.2 111; 112. The first study has been achieved via screening a four- generation KC-affected Caucasian pedigree 111, while the second has included 25 different southern Italian families 112. Furthermore, KC has been reported in a clinical case with

5q31 deletion suffering from multiple musculoskeletal abnormalities 113. Thus, chromosome 5 might be the region which includes KC associated pathogenic variant(s) among different populations. Moreover, various linkage regions, in familiar KC, have been screened to identify the putative mutation. Many variants have been found to segregate with the disease such as MIR184 c.57C>T, CAST c.-40+7414T>C, DOCK9 c.2262A>C,

IL1RN c.214+242C>T and SLC4A11 c.2558+149_2558+203del54 100; 114-116. The combination of linkage and association studies is a powerful approach to identify candidate mutations in the specific linkage region. Dr. Rabinowitz and his colleagues, after identifying a chromosome 5 linkage region 5q14.3-q21.1 in a Caucasian family, have

genotyped tightly spaced SNPs in this loci utilizing both family and case-control KC cohorts. They have identified a SNP in the intronic region of CAST (rs4434401) to be significantly associated with KC in both familial and case-control datasets 114. CAST encodes for calpastatin which acts as a specific inhibitor of calpains, and is important for corneal epithelial cell turnover and wound healing in rabbits 117. It will be valuable to understand the impact of CAST (rs4434401) SNP on the cornea development and how these can contribute to the KC pathogenesis. Up-to-date, MIR184 mutation (c.57C>T) is the only functionally validated mutation using an in vitro model. The discovery of KC pathogenic mutations using linkage analysis is restrained by multiple factors including the difficulty of obtaining a multi-generation family, the limited resolution of the linkage analysis (large- sized linkage region which contains many candidate genes needed to be filtered), and large genetic complexity of KC with incomplete penetrance 106; 118.

3. 2. Genome-wide association studies (GWAS) in KC

GWAS is an approach to genotype hundreds of thousands of SNPs (Single Nucleotide

Polymorphisms) in unrelated large case-control populations to find statistically significant associations between a SNP and a trait. Thus, a SNP with a significant difference in allele frequency between cases and controls is considered to be associated with the disease 118;

119. To minimize the false discovery rate, a p-value of less than 5 × 10−8 is recommended to adjust for the multiple testing as the significance threshold in GWAS. GWAS have identified many KC-associated SNPs in different populations from Europe, USA, China, and Australia. A number of sequence variants have been identified to be associated in KC

(Table 1) and KC-related ocular phenotypes such as central corneal thickness (Figure 3).

However, it remains unclear how these specific variants contribute to KC pathogenesis.

An upstream SNP rs3735520 in the HGF gene has been found to be associated with KC

120. Interestingly, the same study found that serum HGF level was significantly elevated in the patients with the T allele of this SNP. 120. HGF encodes for hepatocyte growth factor, a protein expressed in the cornea, and showed a higher expression in response to corneal injury. In KC-affected cornea, HGF and its receptor (c-met) have shown moderate- to-strong immunoreactivity in basal epithelial cells 121. All these data indicate that HGF elevated expression could correlate to KC pathogenesis, however, the mechanism is still unclear. Another SNP (rs4954218), which has been found in two different studies, is located upstream RAB3GAP1 gene 122; 123. A regulatory role of SNP rs4954218 has been postulated toward two genes which are RAB3GAP1 (6.4 kb away), and YSK4 (21.2 kb away)). However, further functional analysis will be necessary to validate the regulatory mechanism of this SNP 122. Many KC-associated SNPs identified through GWAS are involved in collagen fibrils and ECM regulatory pathways, which have been proposed as a potential mechanism for KC such as LOX and COL5A1. LOX intronic SNPs (rs10519694 and rs2956540) have been identified in both family-based and case-control studies 124.

LOX encodes for lysyl oxidase, a copper-dependent amine oxidase, which is required for the formation of lysine-derived crosslinks in extracellular matrix proteins 125

Table 1: SNPs associated with keratoconus identified through genome-wide association studies Gene (SNP ID) Functional Category Population Reference LOX (rs10519694, rs2956540), Intronic Caucasian 124; 126 (rs2956540) Caucasian HGF (rs3735520, Upstream (Australia 120; 126 rs17501108/rs1014091) and US) RAB3GAP1 (rs4954218) Upstream Caucasian 122; 123 South FOXO1 (rs2721051) Intergenic Australian 127 and US South FNDC3B (rs4894535), Intronic Australian 127 and US South RXRA-COL5A1 Intergenic Australian 127 (rs1536482) and US RXRA-COL5A1 Intronic Latinos 128 (rs3118515) Australian, MPDZ-NF1B 127; 129; Intergenic US and (rs1324183) 130 Chinese South COL5A1 (rs7044529), Intronic Australian 127 and US

COL5A1 Intergenic/ Intronic Caucasian 131 (rs1536482/rs7044529) South BANP/ZNF469 region Intergenic Australian 127; 129 (rs9938149) and US WNT10A (rs121908120) Missense Australian 132

3. 3. KC candidate genes identified by direct screening

Several studies have searched for KC-causative mutation(s) via direct screening of specific genes using either targeted genotyping or Sanger sequencing. VSX1 (visual system homeobox 1) and SOD1 (superoxide dismutase isoenzyme 1) were the first two genes proposed as significant for KC pathogenesis. VSX1 has been extensively screened in KC- affected individuals from different populations. Various VSX1 mutations have been identified in KC and posterior polymorphous corneal dystrophy including D144E, P247R,

G160D, and L17P 133; 134. On the other hand, other studies have not found any pathogenic

VSX1 mutation(s) in their KC patients 135-144. A 7 bases deletion in SOD1 second intron, has been reported to in two different families 145; 146 143; 145; 146. However, a different study could not find any pathogenic mutation(s) in SOD1 140. SOD1 is a cytoplasmic antioxidant enzyme which catalyzes the conversion of superoxide radicals to molecular oxygen and hydrogen peroxide 147. Intriguingly, the activity of extracellular SOD1, in the central cornea of KC, has been reported with a 50% activity reduction 76. Therefore, SOD1 intronic deletion in KC patients could result in elevated corneal oxidative stress 76; 143.

KC- associated irregularities and breaks in the corneal bowman’s layer have prompted

Stabuc-Silih and his colleagues to screen COL4A3 and COLA4 in 104 unrelated KC patients versus 157 healthy individuals 148. Collagen type IV is the main constituent of 17

Bowman’s layer 149. COL4A3 and COL4A4 encode for two out of six α chains which form collagen type IV heterotrimeric molecules 150. Stabuc-Silih et al have found three coding variants, D326Y in COL4A3, M1237V and F1644F in COL4A4 to be associated with KC risk, however, they couldn’t identify any KC specific mutations 148.

3. 4. CCT-associated genetic variants

Central corneal thickness (CCT) is a normally distributed quantitative trait with an assessed heritability of up to 95% 151. Thinner CCT is a clinical demonstration, which has been considered as a risk factor for developing KC. Several GWAS have reported multiple CCT- associated genes including RXRA-COL5A1, COL8A2, FAM53B, FOXO1, and ZNF469, in various large KC cohorts 152-156. The extremely thin cornea has been found to be associated with rare congenital connective tissue disorders (such as Brittle cornea syndrome and osteogenesis imperfecta) 157; 158. CCT-associated SNPs in collagen coding genes have led researchers to examine the association between CCT and KC in large cohorts 127; 156. Intriguingly, numerous CCT associated SNPs have shown a positive association with KC in different populations, which augments the impact of CCT in KC risk 128-131; 156; 159. Pathway enrichment analysis for genes associated with both CCT and KC (including ADAMTS2, ADAMTS8, COL6A2, COL12A1, FBN1, LOXL2,

LUM/DCN/KERA, THSB2, TGFβ2, and LTBP1) reveals a significant role of collagen,

ECM, corneal thickness, and TGFβ signaling pathways in the pathogenesis of KC 127; 156.

Moreover, Ingenuity Pathway Analysis (IPA) for the KC associated genes implies the involvement of connective tissues, the visual system, and skeletal muscle disorders 156.

Corneal related phenotypes have been reported in knockout mice models for KC-associated

genes including corneal opacity (LUM) 160-162, corneal endothelium absence (TGFβ2)

163 and thin corneal stroma (FBN1, KERA, LUM, TGFβ2) 160-165.

Figure 3: Keratoconus genes and their involvement in other ocular diseases

4. Next-generation sequencing strategies

Next-generation sequencing (NGS) is a high throughput sequencing technology, during which millions of DNA fragments one sample are sequenced at once. NGS has allowed low-cost, time effective, massively parallel sequencing which revolutionized genomic and transcriptomic approaches for both basic and clinical research. Comparing to the traditional

Sanger sequencing, NGS is powerful with a lower cost, labor work, and time consumption.

NGS applications include whole genome sequencing (WGS), whole exome sequencing

(WES), targeted capture sequencing, and RNA Sequencing (RNA-Seq). During the last decade, NGS has successfully been used to identify disease-causing mutations, molecular mechanisms, and diagnostic markers 166-171. Targeted sequencing has been successful in exploring the causative genetic mutation(s) in various inherited diseases such as

neurofibromatosis type 1, Marfan syndrome, dilated cardiomyopathy, congenital disorders of glycosylation, and KC 170; 172. However, targeted sequencing is not often efficacious, because the sequenced genes might not constitute the disease-causing mutation(s), or the causal mutation(s) is not in the coding regions of the targeted genes. Thus, performing

WES and WGS may overcome these potential issues 170. WES inquires mutation(s) in the coding regions (1% of human genome) while WGS explores mutation(s) and variants in the whole genome and also any structural modifications such as copy number variance.

Researchers have used WES and WGS to identify genetic mutation(s) in complex diseases among unrelated individuals or family members 172. Using WES, a heterozygous mutation in DYNC1H1 has been found in a four-generation family affected with Charcot-

Marie-Tooth disease 173. On the other hand, by performing WGS, compound heterozygous mutations have been identified in the SH3TC2 gene in a family affected with

Charcot-Marie-Tooth disease 174. This example illuminates that complex heterogenic diseases might have different causal mutations in different families. Furthermore, complex diseases, such as cancer, attributed to various factors other than genetics including environmental and metabolic factors. Thus, researchers have used RNA-Seq to understand a disease associated molecular mechanism 175-177. RNA-Seq has been used to study the expression level of coding and noncoding regions. Non-coding RNAs include long noncoding, miRNAs, and other short RNAs (snRNA and piwiRNA), which regulate protein translation and RNA stability 178.

2. PUBLISHED MANUSCRIPTS

Differential Expression of Coding and Long Noncoding RNAs

in Keratoconus-affected Corneas

Mariam Lofty Khaled1†, Yelena Bykhovskaya2†, Sarah E.R. Yablonski1,3,

Hanzhou Li1, Michelle D. Drewry1, Inas F. Aboobakar4, Amy Estes5, Xiaoyi Gao6, W.

Daniel Stamer4, Hongyan Xu7, R. Rand Allingham4, Michael A. Hauser4,8, Yaron S.

Rabinowitz2, Yutao Liu1,9,10

1 Department of Cellular Biology and Anatomy, Augusta University, Augusta, GA

2 Regenerative Medicine Institute and Department of Surgery, Cedars-Sinai Medical

Center, Los Angeles, CA

3 STAR Program, Augusta University, Augusta, GA

4 Department of Ophthalmology, Duke University Medical Center, Durham, NC

5 Department of Ophthalmology, Augusta University, Augusta, GA

6 Department of Ophthalmology and Visual Science, University of Illinois at Chicago,

Chicago, IL

7 Department of Population Health Sciences, Augusta University, Augusta, GA

8 Department of Medicine, Duke University Medical Center, Durham, NC

9 James and Jean Culver Vision Discovery Institute, Medical College of Georgia,

Augusta University, Augusta, GA

10 Center for Biotechnology and Genomic Medicine, Augusta University, Augusta, GA

† Equal contribution from these two authors

Short Title: Differential Expression in Keratoconic Corneas

Word Count: 3,888

Number of Tables: 5

Number of Figures: 3

Section Code(s): CO

Grant Support: National Institutes of Health (NIH; Bethesda, MD, USA) grants

R01EY023242, R01EY023646, R01EY009052, and P30 EY005722.

Commercial Relationships for All Authors: None

Corresponding Author:

Yutao Liu, MD & PhD

Department of Cellular Biology and Anatomy

Medical College of Georgia, Augusta University

Augusta, GA, USA

Email: [email protected]

Conflict of Interest: None for all authors

Abstract

PURPOSE: Keratoconus (KC) is the most common corneal ectasia. We aimed to determine the differential expression of coding and long noncoding RNAs (lncRNAs) in human corneas affected with KC.

METHODS: 200ng total RNA from the corneas of 10 KC patients and 8 non-KC normal controls was used to prepare sequencing libraries with the SMARTer Stranded RNA-Seq kit after ribosomal RNA depletion, followed by paired-end 50bp sequencing with Illumina

Sequencer. Differential analysis was done using TopHat/Cufflinks with a gene file from

Ensembl and a lncRNA file from NONCODE. Pathway analysis was performed using

WebGestalt. Using the expression level of differentially expressed coding and noncoding

RNAs in each sample, we correlated their expression levels in KC and controls separately and identified significantly different correlations in KC against controls followed by visualization using Cytoscape.

RESULTS: Using |fold change| ≥ 2 and a false discovery rate ≤ 0.05, we identified 436 coding RNAs and 584 lncRNAs with differential expression in the KC-affected corneas.

Pathway analysis indicated the enrichment of genes involved in extracellular matrix, protein binding, glycosaminoglycan binding, and cell migration. Our correlation analysis identified 296 pairs of significant KC-specific correlations containing 117 coding genes enriched in functions related with cell migration/motility, extracellular space, cytokine response, and cell adhesion. Our study highlighted the potential roles of several genes

(CTGF, SFRP1, AQP5, lnc-WNT4-2:1, and lnc-ALDH3A2-2:1) and pathways (TGF-β,

WNT signaling, and PI3K/AKT pathways) in KC pathogenesis.

CONCLUSIONS: Our RNA-Seq based differential expression and correlation analyses have identified many potential KC contributing coding and noncoding RNAs.

Introduction

Keratoconus (KC) is the most common corneal ectatic disorder 1; 2. KC is characterized by bulging and thinning cone-shaped cornea, resulting in myopia, irregular astigmatism, and, eventually, visual impairment. KC usually starts at puberty and progresses until the third or fourth decade 3; 4. KC is a complex multifactorial disorder with the contribution from both genetic and environmental factors as well as biomechanical, metabolic and hormonal factors 81; 82; 179-182. KC patients report significantly impaired vision-related quality of life which worsens with time due to disease progression8.

RNA sequencing (RNA-Seq) allows the entire transcriptome of coding and/or noncoding

RNAs to be surveyed in a high-throughput and quantitative manner 183-186. Long noncoding RNAs (lncRNAs) are class of noncoding RNAs that are at least 200 nucleotides long 187; 188, and have been found to regulate gene expression transcriptionally and post- transcriptionally in both physiological and pathological conditions 188; 189.

Previous studies have attempted to understand processes that underlie KC using different approaches. Wentz-Hunter et al. have performed targeted expression analysis of the stromal layer of the cornea to identify differentially expressed transcripts in KC versus normal cornea 190. Soon afterward, microarray was used to assess RNAs isolated from either the KC-affected epithelium or keratocytes 191; 192. By 2005, differentially expressed proteins in KC patients have been investigated using the whole cornea, certain layer, or tears 193-196. In 2017, Szcześniak et al. and Kabza M et al. used RNA-Seq to determine differentially expressed RNAs between KC-affected and other disease affected corneas197; 198.

For the first time, we used Total RNA-Seq to identify differentially expressed coding/noncoding RNAs in KC-affected corneas versus non-diseased normal ones, and to determine the expression correlation. We used droplet digital PCR (ddPCR) to validate the differential expression of selected RNAs. These differentially expressed RNAs may provide a better understanding of the functions of related lncRNAs and the potential pathways involved in KC pathogenesis.

Methods

Human Cornea Samples and RNA Extraction

We followed the Tenets of Declaration of Helsinki. Unaffected human cornea samples were collected from postmortem donors from North Carolina Eye Bank, and the KC- affected corneal samples were collected during corneal transplantation surgery at the

Cedars-Sinai Medical Center (CSMC). Both studies were approved by the Institutional

Research Board offices at Duke University School of Medicine and CSMC, respectively.

Written informed consent was obtained from all CSMC patients. This study included 10

KC-affected corneas and 8 normal non-KC corneas (Table 2). The inclusion and exclusion criteria for KC diagnosis has been described previously 122; 199. The diagnosis of KC was done by a cornea fellowship trained ophthalmologist based on clinical examination and confirmed with videokeratography. Clinical examination included slit-lamp biomicroscopy, cycloplegic retinoscopy, and fundus evaluations. The slit-lamp biomicroscope was used to identify stromal thinning, Vogts’ striae, and Fleischer ring. The retinoscopy examination was performed with a fully dilated pupil (20 minutes after phenylephrine 2.5% and cyclopentolate 1% drops had been instilled in the eye) to

determine the presence or absence of retro illumination signs of KC, such as the oil droplet sign and scissoring of the red reflex. Videokeratography evaluation was performed on each eye using the Topographic Modeling System (TOMEY, Nagoya, Japan). Subjects were assigned as having KC if they had at least one clinical sign and a confirmatory videokeratography map with an asymmetric bowtie with skewed radial axis above and below the horizontal meridian (AB/SRAX) pattern 200. KC grade was assigned based on the average central keratometry readings as follows: Mild: <50D; Moderate: >50D but <

55D; Moderate-Severe: 55 -60D; Severe: >60D.

Only KC patients at the advanced stage of KC from whom all other therapeutic interventions have failed were included in this study. They were diagnosed with KC grade of above 55D (based on the average central keratometry readings). Half of the corneal button removed from surgery was used for RNA extraction while the other half was sent for pathological examination. The postmortem donors had no other ocular disorders affecting their cornea. These cornea samples were excluded from usage in corneal transplant mostly due to the lack or incomplete medical/health information or occasionally low number of corneal endothelial cells. The central button of the donor corneas was acquired using a 6-mm biopsy punch (Cat#33-38, Miltex, Inc., York, PA, USA) and the whole button was used for RNA isolation. The cornea was stored in RNALater™ solution

(Qiagen, Valencia, CA, USA) to preserve RNA quality. Total RNA was extracted using mirVana miRNA isolation kit from ThermoFisher Scientific, and with NucleoSpin®

RNA/Protein isolation kit (MACHEREY-NAGEL Inc., Bethlehem, PA), according to the manufacturer’s recommendations. RNA quality was evaluated using Agilent 2100

Bioanalyzer with Agilent’s RNA 6000 Pico Kit (Agilent, Santa Clara, CA, US). We

selected RNA samples with a minimum RIN (RNA Integrity Number) score of 6 for RNA-

Seq.

Stranded Total RNA-Seq

We used 200 ng total RNA from all subjects to prepare sequencing libraries with the

TaKaRa SMARTer Stranded RNA-Seq kit after ribosomal RNA depletion using the

RiboGone - Mammalian kit from TaKaRa (TaKaRa Bio USA, Inc., Mountain View, CA,

USA). Briefly, the RiboGone-Mammalian kit removes ribosomal RNA (rRNA) and mitochondrial RNA (mtRNA) sequences from human total RNA samples based on hybridization and RNase H digestion that specifically depletes 5S, 5.8S, 18S, and 28S nuclear rRNA sequences, as well as 12S mtRNA sequences. The rRNA-depleted total RNA was reverse-transcribed to synthesize the first-strand cDNA followed by PCR amplification using universal forward PCR primer and reverse PCR indexing primer set.

Purified RNA-Seq library was validated using the Agilent 2100 Bioanalyzer with Agilent’s

High Sensitivity DNA Kit (Agilent, Santa Clara, CA, USA). Pooled RNA-Seq libraries were sequenced with paired-end 50bp reads using an Illumina HiSeq 2500 sequencer

(Illumina, Inc., San Diego, CA, USA) at the Georgia Cancer Center Integrated Genomics

Core of Augusta University.

After quality check and control with all sequencing reads, demultiplexed reads were aligned by TopHat201 using paired-end reading with the approximation of the median library size. Counts of sequencing reads were normalized using Cufflinks in fragments per kilo bases and millions reads (FPKM) 202. After normalization, we annotated the transcripts with a gene transcript file from Ensembl database at the gene isoform level. We performed differential expression analysis using the Cuffdiff package 203. We used a fold

change of ≥ 2 and false discovery rate (FDR) ≤0.05 to identify genes with significant expression changes in KC corneas versus normal ones. The list of differentially expressed genes was uploaded to the Web-based Gene Set Analysis Toolkit (WebGestalt)

(www.webgestalt.org) 2013 version for gene ontology and KEGG pathway analysis 204.

For lncRNAs, we annotated them using a file from the NONCODE version 5 database 205.

After normalization using Cufflinks, we performed differential expression analysis using

Cuffdiff. For lncRNA data, missing expression data in specific samples was replaced with a background value of 0.001 to enable the differential analysis between cases and controls.

Without such replacement of missing data, the related lncRNAs would have been excluded from the differential expression. For lncRNA differential expression analysis, to avoid the possible impact of potential DNA contamination and RNA isolation bias, we required a minimal size of 300bp for all the lncRNAs. We used |fold change| ≥ 2 and FDR ≤ 0.05 to identify lncRNAs with significant expression changes in KC corneas versus normal ones.

Validation using droplet digital PCR technique (ddPCR)

Twenty-one differentially expressed coding RNAs were chosen for validation with ddPCR as previously described 206; 207. We selected these coding RNAs based on the following criteria: 1. reasonably high corneal expression with ≥ 10 normalized read counts; 2. relevance to corneal integrity, transparency, and cellular adhesion; 3. potential involvement in oxidative stress; and 4. related with corneal wound healing. Specific EvaGreen-based ddPCR assays were obtained from Bio-Rad Laboratories, Inc. (Hercules, CA, USA). To validate the differentially expressed lncRNAs, we selected 7 lncRNAs with reasonable corneal expression (≥10 normalized read counts) and strong correlations with various other coding and lncRNAs. Specific ddPCR expression assays with specific probes were

designed and ordered using the online ddPCR assay design tool with the specific RNA sequence of each selected lncRNA for copy number determination (https://www.bio- rad.com/digital-assays/#/). We used RNA samples from the Total RNA-Seq experiment for validation: 8 KC-affected corneas and 8 control corneas. The ddPCR assays were performed using the QX200 Droplet Digital PCR system (Bio-Rad, Hercules, CA, USA) and its associated QuantaSoft Software, as previously described 122; 199. We used 4 internal reference genes (GAPDH, PPIA, HPRT1, and RPLP0) 208 to be quantified in the samples for normalization purpose. All four reference genes were stably expressed across the KC and control cornea samples and had similar performance in expression normalization, leading to almost identical successful validation of selected target genes.

Thus, we decided to normalize with GAPDH for the figure purpose. Mann-Whitney’s test was used to analyze the differential expression data from ddPCR for the statistical analysis.

Expression correlation and Network Analysis

The expression levels of differentially expressed coding RNAs and lncRNAs in all 18 corneal samples were imported into RStudio Version 0.98.1062 (RStudio, Boston, MA,

USA) 209 from CSV raw data files. The combined expression data was structured into a table where the rows were the mapped RNA names and the columns were the case or control individual de-identified with a code name. Repeated gene names were removed and genes with no expression levels in all controls or cases were also removed. The correlation between all RNA expression levels was determined for the controls data and the cases data separately using R’s built-in Pearson correlation function (cor). Using R psych package’s test of significance for correlations (r.test), the p-value comparing the respective mapped gene’s control and case correlations were calculated. The false discovery rate (FDR) was

calculated from the p-values using the R’s built-in Benjamini–Hochberg procedure function (p.adjust with “BH” parameter). We removed redundant correlations between each gene and itself as well as duplicated gene correlations (A-B is the same as B-A).

The list of expression correlations was filtered with the criteria of an FDR ≤ 0.05 and having at least a mean expression level ≥ 20 in the cases or the controls with one gene.

The entire filtered list of correlated RNAs including lncRNAs was visualized using

Cytoscape version 3.5.1 (Cytoscape Consortium, New York, NY) 210. Expression network hubs were defined as network nodes with multiple correlations to other RNAs. It was expected that this correlation expression analysis among coding RNAs and lncRNAs would help identify biologically important lncRNAs in human cornea. The list of correlated coding RNAs was loaded to WebGestalt (www.webgestalt.org) 2013 version for gene ontology and KEGG pathway analyses 204.

Results

Identification of differentially expressed coding RNAs

Each sample generated 20-44 million paired reads with 100% sample index match, > 96% bases with Q30, and mean quality score > 38. Overall, 62% - 88% of sequencing reads for all RNA samples were successfully mapped to the transcriptome file from Ensembl database, containing a total of 212,368 transcript isoforms. There were a total 95,954 and

103,440 expressed transcripts in all the normal controls and KC-affected cornea samples, respectively. The number of expressed transcripts in each sample ranged from 120,000 to

139,000. Using a cutoff of at least one count of normalized transcripts, the number of

expressed transcripts in each sample ranged from 35,000 to 45,000. There is a total of

14,111 and 19,608 expressed transcripts (≥1 count) in all the normal controls and KC- affected corneal samples, respectively. The mean expression levels of these shared transcripts (≥ 1 count) were 142 reads in control corneas and 63 read counts in the keratoconic corneas.

Using the cutoff with |fold change| ≥ 2 & FDR ≤ 0.05, we identified 436 differentially expressed coding RNAs (159 up-regulated and 277 down-regulated). Our analysis was consistent with previous expression studies by identifying differential expression of many known KC-related genes, including but not limited to TGFBR3, TIMP1, FBLN1, TIMP3,

AQP5, and SFRP1. The top 20 up- and down-regulated coding genes, showing the highest fold changes, were listed in Table 3.

Using WebGestalt with the differentially expressed coding RNAs, our gene ontology analysis revealed disruption in different biological processes (cellular proliferation, differentiation, locomotion, migration, cellular stress and wound response), molecular functions (glycosaminoglycan binding, iron ion binding, antioxidant activity, and insulin like growth factor binding), and cellular components (adherens junctions, extracellular matrix region, and basement membrane) in the keratoconic corneas.

Validation of differentially expressed coding RNAs

Out of the 436 differentially expressed coding genes, we selected 21 genes for validation by ddPCR based on the criteria described earlier (Table 4). Of these 21 genes, 19 were confirmed to have significant and similar expression trend to the RNA-Seq results (Figures

4A and 4B). The other two genes (APOD and IFI6) demonstrated similar expression trends but did not reach statistical significance.

Identification of differentially expressed lncRNAs

A total of 15,966 lncRNAs were expressed in the control and keratoconic cornea samples.

Using the cutoff with |fold change| ≥ 2 & FDR ≤ 0.05, we identified 586 differentially expressed lncRNAs (343 up-regulated and 243 down-regulated). Next, differentially expressed coding RNAs and lncRNAs were used to perform a correlation analysis.

Validation of differentially expressed lncRNAs

Based on their high cornea expression (≥10 normalized read counts), functional annotation in relation to KC-related pathways, and expression correlation with coding and lncRNAs, we selected 7 differentially expressed lncRNAs for ddPCR validation (Table 5). Specific probe-based ddPCR assays with FAM labeling were designed to validate the differential expression of these lncRNAs in KC-affected corneas versus the controls (Figure 2A). The four selected internal reference genes were also used to normalize the expression of these lncRNAs. Five lncRNA showed a similar trend to RNA-Seq results (lnc-WNT4-2:1,

MEG3, lnc-BLID-5:1, lnc-SCP2-2:1, and lnc-ALDH3A2-2:1). However,

NONHSAT174656 and lnc-DLG-1:1 didn’t show any significant differential expression in

KC-affected corneas versus controls (Figure 5A & 5B).

Correlation analysis of coding and noncoding RNAs

The expression correlation analysis identified a total of 492,135 pairs of RNAs with expression correlations in cases or controls. After comparing the correlations between cases and controls, a total of 352 significant differential correlations were identified with a

FDR ≤ 0.05. After applying a minimum mean expression of 20 for at least one of the two paired transcripts in cases or controls, we identified 296 pairs of significant differential correlations between cases and controls. Many coding and noncoding RNAs were identified in multiple correlation pairs. These included A2M, APOD, MUC16, lnc-SCP2-

2:1, lnc-DLG5-1:1, lnc-BLID-5:1, lnc-ALDH3A2-2:1, NONHSAT175650_1, and

NONHSAT217671 (Figure 6). We used Cytoscape to visualize the whole expression network of these 296 pairs. This network analysis revealed a number of genes with significant correlations in the keratoconic corneas versus the unaffected normal controls

(Table 6). We identified 117 coding RNAs in the significantly differential expression correlations. These 117 genes were enriched for their functions in negative regulation of cell migration and cell motility, extracellular space, cellular response to cytokine stimulus, cell surface receptor signaling pathway, and regulation of cell adhesion.

Discussion

KC is a complex disorder with both genetic and environmental factors. Epithelial and stromal layers of the cornea show the prominent histopathological changes during KC progression. A better understanding of the key molecular players involved in KC- associated corneal pathogenesis may provide potential therapeutic targets for disease treatment. Researchers used various techniques including RNA-Seq, microarray, RT-PCR, mass spectrometry, and western blots 191; 192; 196; 198; 211; 212. Recently, Kabza et al. have used RNA-Seq for KC-affected corneas; however, this study was complicated by the use of other diseased corneas as controls. Compared to this study, our data have shown overlapping genes with differential expression (35 up- and 135 down-regulated genes)198.

We identified decreased expression in the cytokeratin (KRT6A, KRT13, KRT15, and

KRT19) and collagen (COL8A1 and COL8A2) genes, indicating compromised integrity of KC cornea. Type VIII collagens are normally expressed in the Descemet’s membrane

213 and targeted inactivation of Col8a1/Col8a2 leads to corneal abnormalities in mice214.

Our study confirms the potential role of oxidative stress in KC-affected corneas by identifying differential expression of many genes involved in response to oxidative stress including GSTO1, SGK1, APOD, KLF9, GPX3, STK39, TXNIP, and lnc-ALDH3A2-

2:175; 215. The majority of these genes in defense mechanism for oxidative stress were down-regulated. However, lnc-ALDH3A2-2:1 is found to have more than 3-fold increased expression and correlated to other lncRNAs (Figure 3A). Lnc-ALDH3A2-2:1, according to Lncipedia database 216, is classified as an antisense transcript which overlaps with the protein coding gene ALDH3A2 on the opposite strand. However, using BLASTN, we found that lnc-ALDH3A2-2:1 sequence overlaps with the last exon of ALDH3A1 gene.

The ALDH3A1 transcript did not show significant differential expression in our study; however, lncRNAs may alter protein level without affecting mRNA 217. Another factor impacting oxidative stress is disruption of iron homeostasis, which can lead to various corneal diseases including KC 194; 195; 218; 219. Fleisher ring, a ring of iron deposition in the corneal cone, is a typical clinical sign for KC 16. Importantly, TFRC, a cellular receptor for iron-ferritin complex uptake 220, has been identified to have elevated expression in KC cornea in our study. This elevated expression in KC-affected corneas may result from iron deficiency 221. We also found down-regulation of the FTL transcript (Ferritin Light Chain) in KC corneas. Significantly, mutations in FTL have been linked to iron accumulation in the brain, causing neurodegenerative disease 222

We observed differential expression of many genes involved in corneal wound healing such as EPPK1, AQP3, AQP5, SLC2A1, and CTGF. Corneal wound healing is a complex process controlled by the integrated actions of multiple growth factors, cytokines, and proteases produced by epithelial cells, stromal keratocytes, and inflammatory cells 223.

However, unregulated healing processes, due to genetic or epigenetic factors, can lead to impairment in cellular function 224; 225. Mace et al. have shown anti-proliferative and hyper-apoptotic phenotype in KC-affected cornea which also reveals a defect in wound healing process 211. Aquaporins (AQPs) are a ubiquitous family of transmembrane water channels to maintain tear film osmolarity, stromal layer thickness, and corneal transparency226. AQP3 and AQP5, important for corneal wound healing, are highly expressed in the cornea 226. Both AQPs show reduced expressions in the KC-affected corneas (Table 3). Consistent with our results, Rabinowitz et al have reported the lack of

AQP5 expression in KC-affected corneas 227. Although this altered expression was not reported by Garfias et al., this discrepancy could be due to their heterogeneous controls

(normal corneal epithelium, sclerocorneal rims, and healthy corneal buttons) 228. EPPK1 encodes a cytoskeletal linker protein called epiplakin, which maintains the integrity of intermediate filaments networks in epithelial cells 229; 230. Migration-dependent wound healing improvement has been found in epiplakin-null epithelium 231. Up-regulation of

EPPK1 in our study suggests the potential involvement of EPPK1 in the impaired corneal wound healing in KC. Interestingly, SLC2A1 is up-regulated during corneal epithelium wound healing 232. Reduced expression of SLC2A1 in KC-affected corneas may reflect a lesser glucose uptake. This is intriguing because there is a reduced risk of developing KC

in diabetic patients) 233-236. Moreover, it has been reported that diabetic patients have thicker cornea than normal individuals 232.

CTGF is expressed in multiple ocular tissues as a downstream signaling molecule in the

TGF-β pathway 237. Importantly, loss of CTGF impairs efficient corneal wound healing

238. The decreased expression of CTGF in our study is consistent with previous studies

198; 199. Dysregulation of extracellular matrix (ECM) pathway has been reported with KC progression 16; 199. Previous studies have shown an up-regulation of ECM degradative enzymes and/or down-regulation of their inhibitors 65; 75; 198. Our data show significant down-regulation in proteinases inhibitors (TIMP-1, TIMP-3, and A2M, and SLPI), but interestingly no significant change with ECM proteinases.

WNT signaling is essential for normal corneal development. Sequence variants in WNT7B and WNT10A genes have been associated with CCT and KC risk 132; 239. A recent RNA-

Seq based study showed dysregulation in WNT signaling in KC-affected corneal epithelium 212. We found some evidence of WNT signaling disruption represented by downregulation of ANGPTL7, TGFBR3, SFRP1 and upregulation of FZD8 (frizzled class receptor 8) and EGR1 (early growth response 1). ANGPTL7 encodes angiopoietin-like 7 and is abundantly expressed in keratocytes to maintain corneal avascularity and transparency 240. SFRP1 inhibits WNT signaling by preventing the formation of the WNT-

FZD complex. Iqbal et al. have shown an increased expression of SFRP1 in the KC- affected corneal epithelium 241 while SFRP1 expression is reduced in tears from KC patients 242. TGFBR3 suppresses WNT signaling through binding to Wnt3a and β-catenin

243. We observed increased expression of Lnc-WNT4-2:1 in KC versus controls. Our

BLASTN analysis indicated that Lnc-WNT4-2:1 is a sense transcript for its overlapping

with exon 5 of WNT4 gene. Interestingly, CPAT (Coding-Potential Assessment Tool using an alignment-free logistic regression model) predicts a coding potential for this lncRNA

(https://hg19.lncipedia.org/) 216. Although no significant change in WNT4 mRNA in KC- affected corneas, this does not rule out the potential role of lnc-WNT4-2:1 in regulating

WNT signaling pathway.

Another pathway in KC pathogenesis is the PI3K/AKT pathway. This pathway is involved in cell cycle regulation and cell proliferation in the cornea 244. Our data shows down- regulation of negative regulators of PI3K/AKT pathway, including PIK3IP1, DDIT4,

IGFBP-3, -4, and -5. PIK3IP1 may suppress AKT activity by inhibiting the expression of

PI3K 245. PIK3IP1 has been reported to be down-regulated in KC-affected corneas 198.

IGFBPs belong to a family of six structurally similar proteins which antagonize the binding of insulin growth factors to their receptors, thus, controlling cell survival, mitogenesis, and differentiation 246. Our results indicate that IGFBP-3,-4, and-5 have 10-fold, 3-fold, and

8-fold reduced expression in KC corneas, respectively. Our findings are consistent with previous studies in keratocytes isolated from KC-affected corneas compared to normal corneas 191.

For the first time, we have identified pairs of RNAs showing significantly different correlations in KC versus controls (Table 6). Functions of most of the identified genes are still not clear, especially in the cornea. APOD and SFRP1 showed a significant positive correlation in KC, with both being down-regulated (Figure 3B). Both genes were reported to have tumor-suppressive functions including DNA-repair, induction of apoptosis, detoxification, differentiation, and transcriptional regulation 247. Apolipoprotein D

(APOD) is a stroma specific protein as lipophilic molecules carrier 248. SFRP1 is a

negative regulator of WNT signaling, but it is unclear whether WNT signaling disruption is a cause for the inverse correlation between the two genes.

Despite the significant findings, our study has several limitations. First is the relatively small sample size of cases and controls. Second, age, gender, and ethnicity were not matched completely between cases and controls. It will be appropriate to study how age affects the corneal expression in controls so that we can identify KC-specific but not age- related expression alterations. The source of corneal samples (i.e. surgical vs. postmortem donors) might potentially contribute to the expression profile difference between KC and controls. Third, this study used whole cornea for expression. It will be important to examine tissue-specific differential expression in the epithelium and stroma separately if possible.

Fourth, our expression study is limited by transcript quantification only. Additional proteomics or protein-based studies are necessary to confirm the differential expression in

KC-affected corneas.

We have not only confirmed many previous reported KC-related differential expressed genes, but also identified a number of novel coding and noncoding RNAs involved in KC pathogenesis. We have identified overabundance of genes with reduced expression, consistent with the destructive nature of corneal degradation associated with KC-related corneal thinning and with our previous RNA microarray study and another study from

Mace et al. 199; 211.

In summary, we have successfully used Total RNA-Seq to identify differentially expressed coding and noncoding RNAs in KC-affected whole corneal tissues. These RNAs may disturb various corneal functions including cellular adhesion, oxidative stress response,

proliferation, migration, apoptosis, and wound healing. Our expression correlation analysis between differentially expressed coding RNAs and lncRNAs has identified a number of potential expression regulators and provided potential functional annotation for lncRNAs with unknown functions.

Acknowledgments

The authors highly appreciate all the donated human corneal tissue samples from all individual donors. Without such samples, this study would not be possible.

Table 2: Clinical phenotypes of patients with KC and normal controls

Table 3: Top coding genes with significant differential expression in KC-affected corneas

Table 4: Twenty one differentially expressed coding genes to be validated using ddPCR.

Table 5: Seven differentially expressed lncRNAs to be validated using ddPCR

Table 6: Coding and long noncoding RNAs with significant differences in expression correlation (with a reasonable corneal expressions) in KC-affected corneas versus the controls.

Figure 4: Differentially expressed coding genes in keratoconus-affected human corneas identified by RNA sequencing (A) and droplet digital PCR (ddPCR). For panel A, ** FDR < 0.01, * 0.01< FDR < 0.05. For panel B, ***p- value < 0.001, ** p-value < 0.01, * 0.01< p-value < 0.05. Error bars in panels A and B represent standard error.

Figure 5: Differentially expressed long non-coding RNAs (lncRNAs) in keratoconus-affected human corneas identified using by RNA-Seq (A) and droplet digital PCR (ddPCR) (B). For panel A, ** FDR <0.01, * 0.01< FDR < 0.05. For panel B, *** p- value < 0.001, ** p-value < 0.01, * 0.01< p-value < 0.05. Error bars in panels A and B represent standard error.

Figure 6: Correlation networks of the differentially expressed coding RNAs and lncRNAs displayed by Cytoscape. The nodes represent mapped RNAs colored and sized by the degree of correlations with other RNAs. The edges are sized and colored based on the significant differences (i.e., false discovery rate, FDR) between controls and cases. Correlation networks were shown for the Lnc-ALDH3A2 -2:1 (A), APOD (B), CLIC4 (C), SLC2A1 (D), and A2M (E) genes.

3. UNPUBLISHED RESEARCH

PPIP5K2 mutations in human familial keratoconus and mouse

cornea

Introduction

Keratoconus (KC, OMIM 14830) is a multi-factorial corneal ectatic disorder1; 2. Genetic, environmental, biomechanical, and hormonal factors play important roles in KC pathogenesis26. Although the majority of KC cases are sporadic, the genetic contribution in KC pathogenesis has been validated by a large number of studies (reviewed in 86).

Familial inheritance has been reported in 6 - 23.5% of KC patients and first-degree relatives are at 15–67 times higher risk than the general population 87. Autosomal dominant inheritance has been observed in most of the multiplex KC families (reviewed in 249).

Higher concordance of KC severity has been reported in monozygotic (MZ) twins than dizygotic twins (DZ)86; 89; 90. Moreover, consanguineous marriage (first-cousin marriage) was found to be a risk factor for KC93; 94.

KC has also been reported to be associated with chromosomal abnormality disorders such as Down syndrome (trisomy 21), and 22q11.2 deletion syndrome16; 95-97. Furthermore, multiple genetically inherited disorders have been found to be associated with KC including Leber congenital amaurosis 98; 99, cataract 100, retinal cone dystrophy 101,

Ehlers–Danlos syndrome, and osteogenesis imperfecta 102; 103.

Genome-wide linkage analyses have identified many genomic loci linked with KC. These include 1p36.23-36.21, 2p24, 2q13, 3p14-q13, 5q14.3-q21.1, 5q21.2, 5q32-q33, 8q13.1- q21.11, 9q34, 13q32, 14q11.2, 14q24.3, 15q15.1, 15q22.33-24.2, 16q22.3-q23.1, and

20p13-p12.2, 20q12 (reviewed in 250). However, the exact disease-causing mutations are still unknown for most patients 106. Intriguingly, the linkage locus 5q21.2 has been replicated in two independent studies 111; 112. Follow-up studies have screened a lot of candidate genes within many of the identified linkage loci to identify the causative mutations. Genetic mutations have been reported in a limited number of genes including

MIR184 and DOCK9 100; 115. MIR184 c.57C>T has been functionally validated using an in vitro model. Thus, in vitro and/or in vivo models are necessary to elucidate the potential function of the identified mutations in KC pathogenesis. The mutation MIR184 c.57C>T has been identified in the linkage region 15q22.32-24.2 in a Northern-Irish family with KC and congenital cataract through targeted sequence-capturing followed by next-generation sequencing 100; 109; 110.

Next-generation sequencing has been successfully used to identify genetic mutations in human Mendelian disorders. Targeted sequencing has been widely used to elucidate the causative genetic mutation(s) in numerous inherited diseases such as neurofibromatosis type 1, Marfan syndrome, dilated cardiomyopathy, congenital disorders of glycosylation, and KC 170; 172. However, targeted sequencing relies on the successful identification of a significant linkage region. The advent of whole exome sequencing (WES) and whole genome sequencing (WGS) technology brings significant advantages by exploring all the coding or genome sequence in the absence of a linkage locus 170. WES is often used to identify mutation(s) in the protein-coding regions (1% of the human genome), while WGS is to explore mutations in the whole genome. We have used these revolutionized approaches to study the causal mutation(s) in a four-generation family affected with KC.

Genome-wide linkage analysis has revealed region chr5q14.3-21.1 to be specific to this family with an autosomal dominant inheritance 111; 251.

Materials and Methods

Human DNA The Institutional Review Boards at the Cedars-Sinai Medical Center and Augusta

University approved the related research protocols. After all patients and family members were informed about the study, they were consented with informed written consent. DNA was extracted from blood samples of all family members (Figure7) 111; 252 from a four- generation family, and individuals from a two-generation family (Figure 8) following a standard protocol from Puregene Gentra Systems, Inc. (Minneapolis, MN). All DNA samples were shipped at room temperature to Augusta University. The concentration of all

DNA samples was determined by DNA-specific fluorescence (PicoGreen) binding assays with a standard curve using a microplate reader Tecan M200. The quality of DNA was evaluated with the ratio of OD260/OD280, the ratio of OD260/OD230, and a 0.8% agarose gel electrophoresis with high mass DNA ladders. Samples with > 10kb in size with no or little degradation were considered to be high quality for WES and WGS. In addition, all

DNA samples were treated with RNase to remove any potential RNA contamination.

Figure 7: A four-generation KC family with autosomal dominant inheritance. A linkage peak was reported on chr5q14.3-21.1. KC-linked haplotype was identified as gray or black bar. Generation number is indicated by the bold-faced Roman numerals listed to the left of the generation. Individual identification number of each member is indicated at the right bottom side of each member. (Adapted from Tang, Y. G. et al. 2005)111. (A) Red color highlights the genotypes of individuals with PPIP5K2 missense mutation (rs35671301,

c.1255T>G, p.Ser419Ala). (B) Red color highlights the intronic variant in the PCSK1 gene (rs373951075, c.1096-10G>A).

Figure 8: The pedigree structure of a second multiplex family affected with KC. Red color highlights the PPIP5K2 missense mutation (rs781831998, c.2528A>G, p.Asn843Ser).

Whole exome sequencing (WES) and Whole genome sequencing (WGS) For WES, one µg high quality DNA per sample for 14 individuals from the two families was enriched with Roche NimbleGen SeqCap EZ Exome Library v3.0 with a 64Mb sequence capture followed by 100bp paired-end sequencing with Illumina Hi-Seq2500 sequencer at Duke University Center for Human Genome Variation (CHGV) as previously described 253. For the bioinformatics analysis, all sequencing data underwent quality control with the TrimGalore package by removing any Illumina adapter sequences or low quality base calls from the 3’-end of the reads. Sequencing reads were aligned to human reference genome NCBI build 37 with the BWA algorithm 254. Potential PCR duplicate reads were removed with Picard and variants are called with GATK, following the Broad

Institute’s Best Practices Workflow 255-257. The called variants were annotated for their

functional impact using SNPeff with annotations from Ensembl 258. All the samples were sequenced with at least 75X average coverage for all targeted regions.

Five individual DNA samples from the four-generation family were selected for WGS.

WGS was done with paired-end 150bp sequencing with Illumina HiSeq XTEN sequencer with at least 30X average coverage at New York Genome Center (NYGC, New York, NY,

USA). Bioinformatics analysis was performed using the standard pipeline at NYGC for sequence quality control, sequence alignment, sequence and structural variants calling, and annotation. All five samples were sequencing with 31-37X coverage. After we assessed the quality of WGS data along the analysis pipeline, WGS data were processed through an automated pipeline at NYGC’s high-performance computational facility. Paired-end 150bp reads were aligned to the GRCh37 human reference genome using the Burrows-Wheeler

Aligner (BWA-MEM v0.78) 254 and processed using the best-practices pipeline that includes marking of duplicate reads by using Picard, realignment around indels, and base recalibration via GATK v3.2.2 256. Variants were called using GATK HaplotypeCaller to generate a single sample Genotype VCF (GVCF) file. To improve variant call accuracy, five single-sample GVCF files were jointly genotyped using GATK GenotypeGVCFs to generate a multi-sample VCF file. Variant annotations were done using SnpEff 258,

VCFtools 259, and NYGC in-house custom software.

Variant filtering pipeline We applied a specific sequence analysis pipeline in a combination with human corneal gene expression data to filter and prioritize all the variants (Figure 9). We used SNP &

Variation Suite (SVS8.0) software from Golden Helix (Bozeman, MT, USA) for our analysis pipeline. We filtered and focused on the variants using the following criteria: 1)

good sequencing quality (≥20) and high read depth (≥20) ; 2) absence in the unaffected controls carrying no linkage risk haplotypes (indicated by the empty column); 3) minor allele frequency (MAF) ≤ 0.01 in publically available variant databases (The Exome

Aggregation Consortium (ExAC), NHLBI GO Exome Sequencing Project (ESP), Genome

Aggregation Database (gnomAD), and 1000 Genomes); 4) exonic and non- synonymous variants; 5) heterozygous in the affected patients via autosomal dominant inheritance; 6) the gene containing the variant is expressed in human cornea; and 7) located within in the linkage locus chr5q14.3-21.1. We have applied a similar filtering pipeline for the WGS data (Table 7).

Figure 9: Variant filtering pipeline for whole exome sequencing data using the SVS software

Table 7: Variant filtering pipeline for whole genome data using the SVS software

In vitro functional assay for the identified mutations

Plasmids hosting human PPIP5K2 (NM_001345875) were constructed as previously described 260. Single mutations of PPIP5K2 Ser419Ala or Asn843Ser were constructed using the Q5 site-directed mutagenesis kit (New England Biolabs, Ipswich, MA, USA).

The paired primers of human PPIP5K2 were designed as follows (mutagenic nucleotide is lower case): for S419A the sequence of the forward primer is

TGGATATAAAgCAGGGAAATTAAAAC; and the reverse is

TCACACTTTTCAAAAAGATCAAAAAATTTC. For the second mutation N843S, the sequence of the forward primer is GCCTTATGCAgTGAATCAAAG; and the reverse primer is ACCATAGCGAAGAATAGAC. All constructs were confirmed by DNA sequencing. HEK293 cells were transiently transfected with Flag-tagged PPIP5K2,

PPIP5K2 p.Ser419Ala or PPIP5K2 Asn843Ser plasmids using Lipofectamine 3000 (Life

Technologies). Cells were harvested 16- 20 hours after transfection and lysed in ice-cold buffer containing 10 mM HEPES, 130 mM NaCl, 1% Triton X-100, 10 mM NaF, 10 mM

Na2HPO4, 10 mM Na pyrophosphate and protease inhibitor cocktail (Roche, Basel,

Switzerland). The FLAG-tagged PPIP5K2 proteins were immunopurified using FLAG M2 affinity gel (MilliporeSigma, St. Louis, MO, USA). Purified FLAG-PPIP5K2 proteins were analyzed by SDS-PAGE and stained with Coomassie Blue. Densitometry analysis

(ImageJ) was used to calculate the amounts of PPIP5K2 loaded onto the gel by referencing additional lanes loaded with known amounts of the human recombinant PPIP5K2 kinase domain.

The enzymatic assays of PPIP5K2 phosphatase and kinase were performed as previously described 260; 261. Briefly, PPIP5K2 phosphatase activities were measured at 37°C after

30 minutes incubations comprising 100 µl of assay buffer containing 1 mM Na2EDTA, 50 mM KCl, 20 mM HEPES (pH 7.2), 2 mM MgCl2, 0.5 mg/ml BSA and 1µM of chemically synthesized InsP8. The activities were calculated by how much InsP8 hydrolyzed to 5-

InsP7 and was then quantified by HPLC. Kinase activities were measured at 37°C after one hour incubations comprising 100 ul of assay buffer containing 1 mM Na2EDTA, 50 mM

KCl, 20 mM HEPES pH 7.2, 7 mM MgCl2, 5 mM ATP, 0.5 mg/ml BSA, plus 0. 1 μM of

chemically synthesized 5-InsP7. The reaction was quenched and the conversion of 5-InsP7 to InsP8 was determined by HPLC.

Primary human corneal epithelial cells and stromal fibroblasts culture

Human donor eyes were received from the Georgia Eye Bank (Atlanta, GA, USA) within

24 hours after death. Primary human corneal epithelial cells (HCEC) and stromal fibroblasts (HCSF) were isolated from the donors’ eyes according to the established protocols 262; 263. Briefly, in serum-containing media, corneas were dissected under the microscope in the presence of DispaseII to detach the epithelium from the remaining tissue.

Epithelium was kept in media with the DispaseII overnight at 4°C to get a single cell suspension. HCEC were cultured in high glucose DMEM (Gibco, ThermoFisher

Scientific), 10 ng/mL human recombinant EGF (Sino Biological, Wayne, PA, USA), 10% fetal bovine serum, 1% insulin-transferrin-selenium-100X (Gibco), and 40µg/mL gentamycin (Sigma). The cells were incubated at 37°C under 95% humidity and 5% CO2, and the medium was changed every 3 to 4 days. From the remaining tissue, endothelium was removed and only the layer containing stromal cells was kept in a medium containing serum and collagenase at 37 ̊C and 5% CO2 to get HCSF. HCSF were cultured in the same medium and under the same conditions as HCEC but without human recombinant EGF.

We isolated RNA from HCEC (three different donors), and HCSF (one donor but the RNA was from three different culture plates) using Exiqon miRCURY™ RNA Isolation Kit.

PPIP5K1 and PPIP5K2 expression with droplet digital PCR (ddPCR) Specific EvaGreen-based ddPCR assays for PPIP5K1 (dHsaEG5000994) and PPIP5K2

(dHsaEG5018506) were obtained from Bio-Rad Laboratories, Inc. (Hercules, CA, USA).

We assessed the expression of PPIP5K1 and PPIP5K2 in primary HCEC and HCSF using with ddPCR as previously described 206; 207. We used two internal reference genes

(GAPDH and HPRT1) to be quantified in the samples for normalization purpose, and both showed similar stable expression among all samples. For figure purpose, we normalized with HPRT1.

Mouse husbandry We obtained breeding pairs of heterozygous B6N (Cg)-Ppip5k2tm1b (EUCOMM) Wtsi/J mice from Jackson Laboratory (Figure 14A). It is a Ppip5k2 gene trap mouse model with no phosphatase activity, but elevated kinase activity (designated as Ppip5k2K^) 264. This mouse model was generated by the Knockout Mouse Phenotyping Program (KOMP). A beta-galactosidase-containing cassette with an upstream 3’ splice site and a downstream transcriptional termination sequence with poly (A) tail was inserted as a reporter allele to replace exon 14. Mice were bred and maintained according to the guidelines described in the Association for Research in Vision and Ophthalmology Statement for the Use of

Animals in Vision and Ophthalmic Research and Augusta University animal handling guidelines. The mouse protocol was approved by the Institutional Animal Care and Use

Committee (IACUC) at Augusta University. Mice genotypes were assessed using PCR with the following primers, wild type forward 5′-GGG AGA CCT ATC CAT GGT TGT

A-3′, wild type reverse 5′-CAT TTG AGC AGA GCT TGG TG-3′, mutant forward 5′-

CGG TCG CTA CCA TTA CCA GT-3′, and mutant reverse 5′-CAG CCC TGA AAA

CAT ACA CG-3′. PCR was performed using Touchdown protocol with an annealing temperature of 57 ̊C 265. Gel electrophoresis showed different patterns of band sizes according to the genotype. The homozygous mouse Ppip5k2K^/K^ showed one PCR band at 450bp, the heterozygous mouse Ppip5k2+/K ^showed two bands at 450bp and 236bp, while the wild-type mouse Ppip5k2+/+ yielded one band at 236bp.

In vivo mouse corneal examination Mice were anesthetized using a rodent anesthesia cocktail containing ketamine 100 mg/mL, xylazine 12 mg/mL (Sigma-Aldrich Corp., St. Louis, MO, USA) for either slit lamp or spectral domain optical coherence tomography (SD-OCT) examination. GenTeal

Lubricant Eye Gel (Alcon, Ft. Worth, TX, USA) was applied after anesthesia until the examination started. Systane lubricant eye drops from Alcon were applied throughout the eye examination to avoid cornea dryness. a. Slit lamp examination The cornea and anterior chamber of the anesthetized mouse were examined with a slit lamp

(SL-D7; Topcon, Tokyo, Japan) and photo-documented with a digital camera (D100;

Nikon, Tokyo, Japan). b. SD-OCT We used SD-OCT to visualize the anterior segments in mice 266; 267. The structure of anterior chamber and corneal curvature of the anesthetized mouse were examined using the Bioptigen Spectral-Domain Ophthalmic Imaging System (Envisu R2200; Bioptigen,

Morrisville, NC, USA) 268. Briefly, we used the 12mm telecentric probe and set the reference arm at position 245. A mouse was mounted on to the cassette, and OCT laser was aligned along the optical axis of the eye until Purkinje reflections on the X Y axis were obtained. Three scans were performed for each eye which consisted of one rectangular and two radial scans. The rectangular scan parameters were: length = 4mm, width = 4mm, angle = 0 degree, A-scan/B-scan = 1000 lines, b-scan = 100 scan, frame/B-scan =1, and duration 4 seconds. The radial scan parameters were: Diameter = 3mm, A-scan/B-scan =

1000 lines, B-scan/volume = 50 scan, frame/B-scan =1, and duration 4 seconds with the first radial scan and 2 seconds for the second one. The central corneal thickness (CCT) 62

and the anterior chamber depth (ACD) were assessed using the software as shown in figure

15D.

Curvature and thickness mapping Corneal biometric parameters including corneal curvature, topography, and pachymetry were measured from the corneal OCT images using methods previously described for human corneal OCT images and then adapted for mouse corneal OCT images 269-271. In brief, after contrast enhancement, each radial corneal OCT B scan was first segmented using a semi-automatic, gradient-based algorithm to identify the corneal epithelial and endothelial surfaces. The algorithm used an initial seed segmentation that served as a pilot template for segmenting the remainder of the volume. Axial motion correction was performed by aligning segmentations at the axis of rotation of the radial volume. The data were then fitted to a sixth-order Zernike polynomial to resample the radial data into an evenly distributed Cartesian data set. Optical dewarping was implemented by path length and refraction correction with a vector form of Snell’s Law.

From the corneal surface data, pachymetry (corneal thickness) maps were created by direct z-axis subtraction of the epithelial and endothelial layers. Curvature maps were created by calculating local epithelial curvatures based on formulas outlined in the ANSI Z80.23-2007 standards for human corneal topography 272. Instantaneous curvatures were calculated along meridians across the nominal optical zone. A radius of curvature (Rc) to describe the average curvature of each specific cornea was calculated by fitting the anterior corneal surface to a sphere using least squares fitting.

Cornea sectioning and histological staining a. Mouse cornea After euthanization, the whole mouse eye was removed and fixed in Davidson’s fixative for <24hrs and was transferred to 70% EtOH. Eyes were sent to the Electron Microscopy and Histology Core laboratory at Augusta University for sectioning. After fixation, eyes were embedded in a low melting point agar following a specific protocol. Briefly, 4% agar was prepared and heated until the solution was completely dissolved and then heat was turned to medium and stirred for 20 minutes. An embedding mold was filled with 4% agar and the eye was oriented under a dissecting scope to keep the eye on the bottom of the mold. When solidified, the agar block was removed from the embedding mold and excess agar was trimmed off around the eye, then was placed in a cassette. The cassette was added to 70% ethanol and the block was ready for processing. Samples were processed on a

Sakura Tissue Tek VIP processor overnight, then, they were embedded in Polyffin embedding media (VWR#15147-839) and sectioned on a Leica RM 2155 microtome at 5

μm. Some slides were stained with H&E on a Leica Autostainer XL, while others were kept unstained for later use. Slides were then stored at room temperature until use. b. Human cornea Postmortem human corneas were obtained from Duke University Eye Center via the North

Carolina Eye Bank. The corneas were fixed in Davidson’s fixative for < 24 hours and then were transferred to 70% alcohol. Fixed corneas were sent to Georgia Esoteric and

Molecular (GEM) Laboratory at Augusta University for sectioning. Human corneas were put in agar blocks and then were processed using Excelsior AS Thermofisher. Corneas

were sectioned with a 4um thickness using a Leica Microtome. H&E staining was done by using Leica Autostainer XL. c. Immunofluorescence staining (IF) Slides with mouse or human cornea sections were rehydrated as follows: xylene (5 minutes, twice), xylene with 100% ethanol (5 minutes, once), 95% ethanol (3 minutes, twice), 75% ethanol (3 minutes, once), 50% ethanol (3 minutes, once), and finally were washed in ddH2O (3 minutes, once). Next, heat-induced epitope retrieval was performed using citrate buffer pH=6. Tissue permeabilization was done with PBS containing 0.3% Triton X-100 for 10 minutes. Non-specific antibody binding was blocked by incubation in 10% goat serum in PBS, for 1 h. Corneal tissue was incubated with rabbit anti-PPIP5K2 (ab204374,

Abcam, 1:100), or anti-PPIP5K1 (HPA039380, Sigma, 1:100) primary antibody at 4°C overnight. Both antibody dilutions were made up in 5% goat serum in PBS. Slides were washed in PBS (5 minutes, three times) and incubated with the secondary antibody (Alexa

Fluor 488 goat conjugated goat anti-rabbit, Invitrogen) for 1 h at room temperature. Slides were mounted in ProLong Diamound Antifade Mountant with DAPI (Invitrogen) and then images were taken by immunofluorescent microscopy.

Statistical analysis Data were analyzed using the Student’s t-test with GraphPad Prism 8.02. A p ≤ 0.05 was considered statistically significant.

Results Mutations within the linkage region chr5q14.3-21.1 In the four-generation multiplex family, we selected ten individuals (KC patients - II5, III4,

III5, IV4, IV5, IV10, and IV12; controls - III12, III13, and III14) for WES sequencing

(Figure 7A). Variant filtration of WES data led to identification of a non-synonymous mutation that is located within the linkage region chr5q14.3-21.1 and co-segregated with

KC in the four-generation family. We identified a missense mutation (rs35671301, c.1255T>G, p.Ser419Ala) in the PPIP5K2 gene (Diphosphoinositol Pentakisphosphate

Kinase 2) with a global MAF of 0.006 in the gnomAD database (genome aggregation database, gnomad.broadinstitute.org). Using PCR-based Sanger sequencing, we verified this mutation in all 10 selected individuals for WES (Figure 10A). We confirmed the segregation of this mutation with KC in all family members but not in two individuals III9 and IV9. The amino acid serine at position 419 was highly conserved evolutionally from zebrafish to all primates, suggesting its potential important function in the PPIP5K2 protein in biological systems including the corneas

We also selected five individuals (KC patients - III7, IV2, IV5, and IV10; controls - IV8) for WGS. Using our established variant filtering pipeline (Table 7), we not only verified the missense mutation in the PPIP5K2 gene but also identified an intronic variant in the

PCSK1 gene (proprotein convertase subtilisin/kexin type 1) (rs373951075, c.1096-10C>T) with a global MAF of 0.00026 in the gnomAD database (Figure 7B). Using PCR-based

Sanger sequencing, we verified this variant and confirmed the complete segregation of this variant in this family (Figure 10B). However, this variant is not highly conserved

evolutionally with both alleles present in different vertebrates almost equally. This lack of conservation indicates the lower possibility of its involvement in human health and disease.

Figure 10: PCR-based Sanger sequencing of the identified mutations. A and B represent mutations identified in chr5q14.3-21.1linkage loccus of the four-generation family. C represents the missense mutation in the second KC-affected family.

A second missense mutation of PPIP5K2 in another KC family Using WES data from a second KC-affected family with four selected individuals (KC –

301 and 303; controls – 201 and 202, Figure 8), we identified another missense mutation

(rs781831998, c.2528A>G, p.Asn843Ser) in the PPIP5K2 gene with a global MAF of

0.00006 in the gnomAD database. Using PCR-based Sanger sequencing, we successfully verified the mutation and screened this mutation in other family members (Figure 10C).

This finding adds more evidence that PPIP5K2 may play a role in the pathogenesis of KC.

Across 100 different vertebrates, the amino acid at this position is either aspartic acid (D) or asparagine (N). The rare mutation to Serine at this position could impact the protein function of PPIP5K2. This additional mutation provides more evidence that PPIP5K2 may play a role in the pathogenesis of KC. Thus, we focused our research on the potential role of PPIP5K2 mutations in the pathogenesis of KC.

Figure 11: (A) Inositol pyrophosphate metabolism. (B) Domain graphic for human PPIP5Ks. Adopted from Gu et al.261

Biochemical properties of the two mutations After overexpressing the PPIP5K2 protein with either Ser419Ala or Asn843Ser mutations in the HEK293 cells, the mutated PPIP5K2 proteins were purified for enzymatic assays to measure their activities of phosphatase and kinase. In vitro functional enzymatic assays

(n=4) showed that Asn843Ser point mutation caused a significant 15% reduction in the

PPIP5K2 phosphatase activity (p < 0.05) while the Ser419Ala mutation did not change the phosphatase activity significantly (p > 0.05) (Figure 11A). Although both mutations increased the kinase activity in vitro, this increased kinase activity was only significant for the N843S mutation (Figure 11B). Since these in vitro enzymatic assays do not reflect the cellular environment, it is important to examine the impact of these mutations on the kinase/phosphatase activity in cell culture in the future. It is possible that S419A mutation may affect the enzymatic activities more in the physiological condition. These in vitro biochemical assays provide direct evidence that these two mutations could be functional in cells.

Figure 12: Biochemical properties of the two missense mutations (A) PPIP5K2-Phosphatase activity level. (B) PPIP5K2-Kinase activity level. K2 represents hsa-PPIP5K2, K2S419A represents hsa-PPIP5K2 with Serine 419 to Alanine mutation, and K2N843S represents hsa-PPIP5K2 with Asparagine to Serine change. Done by Chunfang Gu, and Stephen Shears.

PPIP5K2 dominant expression over PPIP5K1 in human and mouse corneas We examined the expression of PPIP5K2 and its homolog PPIP5K1 in primary HCEC and

HCSF using ddPCR assays. PPIP5K2 showed a 5-fold higher expression in human cells than PPIP5K1 (Figure 13A). Immunofluorescence staining demonstrated a higher expression of PPIP5K2 in both human and mouse corneas (Figure 13B, D, & F), while

PPIP5K1 showed higher expression in mouse retinas (Figure 13 C & E).

Figure 13: PPIP5K2 expression level compared to PPIP5K1. (A) HCEC and HCSF show higher expression of PPIP5K2 transcripts than PPIP5K1. (B) Human cornea expresses PPIP5K2. (C) PPIP5K1 and (D) PPIP5K2 expression in mouse cornea expresses. (E) PPIP5K1 and (F) PPIP5K2 expression in mouse retina. Error bars represent standard error. Green fluorescence represents the target protein, blue fluorescence

represents DAPI. Ep: epithelium, St: stroma, End: endothelium, RPE: Retinal Pigment Epithelium; OS/IS: Outer/Inner Segment; ONL: Outer Nuclear Layer; INL: Inner Nuclear Layer; GCL: Ganglion Cell Layer.

Corneal pathological phenotypes observed in Ppip5k2+/K^ & Ppip5k2K^/K^ mouse models In vivo assessment of the mouse corneal phenotypes was performed using SD-OCT scanning and slit lamp biomicroscopy. We used the SD-OCT to visualize and examine the anterior segment including the corneas. We observed irregularities on the anterior corneal surface in the Ppip5k2+/K^ & Ppip5k2K^/K^ mouse models at three months old compared to their littermate controls (Figure 14). Approximately 40% of the

Ppip5k2+/K^ & Ppip5k2K^/K^ mice showed significantly abnormal corneal curvature of the anterior surface of mouse corneas (Figure 14A).

CCT measurements did not change significantly across three different genotypes, while the anterior chamber depth was significantly reduced in the Ppip5k2K^/K^ (Figure 15B and C).

H&E staining of mouse corneal sections revealed that these surface irregularities were mostly manifested as thickened epithelium (Figure 16). The OCT derived pachymetry mapping exhibited a significant corneal thickness reduction in the mutant mice versus their littermate controls (Figure 17). The curvature map showed slight changes in the mutant mice (Figure 17). Although, slit lamp examination revealed significant corneal opacity in mouse corneas from Ppip5k2+/K^ & Ppip5k2K^/K^ mice at 3-4 months age (Figure 18).

However we did not observe corneal cone formation in these mice.

Figure 14: SD-OCT images of mouse anterior chamber. (A) Allele map of the Ppip5k2K^ mice. (B)Wild- Type. (C) Ppip5k2+/K^ mouse. (D) Ppip5k2K^/K^ mouse. Arrows pointed to the corneal surface irregularities.

Figure 15: Mice corneal assessment. A. number of mice with sever corneal abnormalities among three groups. B. Central cornea thickness (CCT) measurements among three groups. C. Anterior chamber depth (ACD) among the three groups. D. SD-OCT image of a WT mouse showing the method of measuring the CCT and the ACD.

Figure 16: H&E of the mice corneas. A &B for Wild-Type, C for Ppip5k2+/K^ with pathological corneal phenotype, D for Ppip5k2+/K^ without pathological corneal phenotype, E for Ppip5k2K^/K^ with pathological corneal phenotype, F for Ppip5k2K^/K^ without pathological corneal phenotype. Ep: epithelium, St: stroma, End: endothelium.

Figure 17: Mouse corneal curvature and thickness mapping. The uppercase letter represents curvature map (epithelial axial map), while the lowercase letter represents the thickness map (pachymetry). Panel A for Wild-Type, B for Ppip5k2+/K^, and C for Ppip5k2K^/K^. Done by Alice Liu and Anthony Kuo

Figure 18: Mouse Slit lamp Examination. (A & D) Wild-type. (B & E) Ppip5k2+/K^ mouse. (C & F) Ppip5k2K^/K^ mouse. Upper panel shows the absence of cone formation in the mice corneas among the three groups. Lower panel shows the corneal opacity in the mutant mice compared to the wild-type

Discussion

We have successfully identified two novel pathogenic mutations of PPIP5K2 in two autosomal dominant multiplex families using WES and WGS. Both mutations are located in the PPIP5K2 phosphatase domain. PPIP5K2 (diphosphoinositol pentakisphosphate kinases) is a dual functional enzyme which encompasses both kinase and phosphatase catalytic activity. The kinase activity converts 1,5-InsP7 to InsP8 and the reversible reaction is catalyzed by the phosphatase activity (12A and B). PPIP5K2 and its products

(1,5-InsP7 and InsP8) play a vital role in the inositol phosphate signaling pathway via regulation of energy homeostasis, phosphate (Pi) sensing, and immune responses 273-275.

We assessed the mRNA and protein expression levels of PPIP5K2 relative to its homolog

PPIP5K1 in human corneal cells and tissues. PPIP5K2 shows a significantly higher expression level than PPIP5K1 in HCEC and HCSF. Furthermore, immunohistochemistry results demonstrated that human cornea expresses more PPIP5K2 than PPIP5K1. The relatively higher PPIP5K2 corneal expression suggests a potential role of this enzyme in the cornea. Interestingly, PPIP5K1 showed a markedly high expression in the mouse retina indicating the biological function of PPIP5K1 in the mouse retina.

To study the impact of the identified mutations on the PPIP5K2 catalytic activity, we introduced each point mutation in a plasmid construct. Intriguingly, the p.Asn843Ser point mutation disrupted the PPIP5K2 enzymatic activity by reducing the phosphatase and elevating the kinase activity. Although the p.Ser419Ala mutation did not affect the enzymatic activity, the effect of the mutation may via enzyme regulation under other factors including inorganic phosphate and glucose, or binding partners261. The crystal structure of the phosphatase domain has not yet been identified, which hinder our success in identifying the functional deficit related to the p.Ser419Ala mutation. Across different vertebrates, the PPIP5K2 protein sequence is highly conserved at aa419 and aa843, further supporting their potential role in KC patients.

It is often convenient to use mouse models to study human genetic disorders due to their phylogenetic relation and physiological similarity to humans, the ease of breeding and maintaining them in the laboratory, and the availability of multiple inbred strains 276.

Accordingly, we studied the role of PPIP5K2 in the cornea using a mouse model with elevated kinase activity. We used a Ppip5k2 gene trapped mouse model with beta- galactosidase-containing cassette to replace exon 14. We examined the cornea anterior

curvature from three groups of mice: Ppip5k2+/K^, Ppip5k2K^/K^, and their littermate wild type at 3-4 months of age. This age was chosen because both murine epithelium and stroma thickness are stabilized after 2 months of age 267, and because this age is equivalent to the teenage of onset of KC in human patients 6. We used complementary approaches to assess mouse cornea structural changes, including SD-OCT, slit lamp, and corneal histology. The slit lamp examination did not show the existence of corneal ectasia resembling that seen in

KC patients. However, slit lamp biomicroscopy did reveal a various degree of corneal opacities and shallow anterior chamber in the Ppip5k2+/K^ and Ppip5k2K^/K^ mice versus their littermate controls. Additionally, SD-OCT examination showed abnormal cornea structure with surface irregularities in these mice with hyperkinase activity. Among the three different groups (wild type, Ppip5k2+/K^, and Ppip5k2K^/K^), approximately 40% of the mutant mice exhibited pathological corneas with surface irregularities, indicating the incomplete penetrance. SD-OCT confirmed the significantly shallower anterior chamber depth in the Ppip5k2K^/K^ mice observed in the slit lamp exams. Central corneal thickness

(CCT) of mice did not show a significant difference among the three different groups. The

CCT of the mutant mice were highly variable, but their average was similar to the wild- type. Thus, the lack of significant differences in CCT should not mislead us on the potential impact of PPIP5K2 on corneal structure and function. Our OCT-based anterior curvature and thickness mapping clearly indicate the abnormally thinned localized corneal regions in the Ppip5k2+/K^and Ppip5k2K^/K^ mice. Additional transgenic mouse models containing the two identified human mutations in the PPIP5K2 gene are needed to determine the specific functions of these mutations and their relationship to KC pathogenesis.

WGS data showed an intronic mutation within PCSK1 (rs373951075, c.1096-10G>A) in the four-generation family. This mutation might contribute to KC pathogenesis as well.

The mutation is located in a consensus sequence for binding with JunD which is AP-1 transcription factor subunit. The consensus sequence includes cAMP-responsive element

(CRE) motifs to which JunD binds and forms heterodimers with the ATF family members

(activating transcription factor) 277; 278. JunD acts as a transcriptional activator or repressor of specific target genes to regulate diverse vital processes including cellular proliferation, differentiation and oxidative stress279. Further investigation is necessary to understand the impact of this intronic mutation on JunD binding and its functional impact on the corneal physiological function as well in KC pathogenesis.

Finally, our study succeeded to identify the potential causative mutation of KC in two different families; however, our study had limitations. First, the disease haplotype exists in family members without any KC manifestation, which is likely due to the incomplete penetrance of KC. Second, the MAF of p.Ser419Ala varies greatly across different ethnic populations in public databases such as gnomAD. As aforementioned, p.Ser419Ala is found in the four–generation US family with European ancestry. MAF of p.Ser419Ala is

0.01084 in the European population and is 0.0028 in the North American population while it is 0.0068 in all populations combined together. Third, both PPIP5K2 (p.Ser419Ala) and

PCSK1 intronic mutations exist in all the four-generation family members carrying the disease haplotype; however, two members in this family do not have p.Ser419Ala mutation. These two members are in the same branch within the pedigree, one with KC and the other without KC. This finding highlights the importance of the intronic mutation on the PCSK1 but should not discount the functional impact of p.Ser419Ala on KC

pathogenesis. It is remotely possible that these two members could be chimeric and the p.Ser419Ala mutation exists in the ocular tissues but not in their lymphoblasts (from which we isolate DNA). Forth, although the mouse cornea showed structural abnormalities, the mouse model in our study did not exactly show the same KC-associated clinical phenotypes that are present in human patients. This could be due to the contribution or interaction of other non-genetic factors including hormonal, environmental, eye rubbing, and biomechanical factors.

In conclusion, we have identified two novel mutations in the phosphatase domain of

PPIP5K2 gene in two US families. Our in vitro functional assays and in vivo mouse model strongly suggest the potential impact of PPIP5K2 in the pathogenesis of KC.

4. DISCUSSION

KC is a complex multifactorial disease. KC is associated with histopathological changes in all the corneal layers except the corneal endothelium. The effective treatment for KC patients often involves surgical procedures, yet a pharmacological treatment will be beneficial especially with younger aged patients. Although KC has been widely studied, the main cause of the disease and the molecular mechanism remain mysterious. In my thesis, I aimed to study the molecular genetics of KC via the application of next-generation sequencing (NGS) technology. The advent of NGS leads to a revolution in molecular human genetics as it reproduces enormous sequence data in a short time at an affordable cost. The successful application of NGS in identifying genetic mutations in a large number of human genetic disorders motivates our group to study KC using NGS. It is necessary to identify KC-associated transcriptional alterations due to the pathological process in human patients. Thus, it is important to search the genetic mutations in KC families and to profile the expression of both coding and noncoding genes in KC-affected corneas versus normal controls.

In the first aim, I have used stranded Total RNA-Seq to characterize the KC-affected corneal transcriptome. The derived corneal expression data was utilized to prioritize variants in cornea-expressed genes aiming to identify potential causative mutation(s) in KC families. I also used RNA-Seq data to identify novel KC-associated pathways and gene networks containing both coding and noncoding RNAs. For the first time I have identified many long noncoding RNAs that could play important roles in KC pathogenesis. The second aim was to identify the causal mutation(s) in a four-generation family affected with

KC. Integrating both aims can definitely advance our knowledge about the disease pathogenesis and help identify potential therapeutic targets for future therapy.

RNA-Seq has been used to measure the expression of both coding and noncoding RNAs and to study the role of these RNAs in different human disorders280; 281. Approximately

70% of the human genome is transcribed as noncoding RNAs. These noncoding RNAs

(ncRNA) include small ncRNAs (miRNAs, siRNAs, snRNA, and piRNA, < 200 nucleotides) and long ncRNAs (lncRNAs, > 200 nucleotides). While the role of miRNAs has been studied extensively, the biological relevance of most lncRNAs to human disorders is still unclear. Recently, many studies have revealed the regulatory role of lncRNAs in different diseases. In accordance, lncRNAs may regulate gene expression through epigenetic, transcriptional, post-transcriptional, or translational regulation. For example, lncRNA HOTAIR recruits polycomb repressive complex 2 to certain target genes, resulting in an epigenetic re-programming. Elevated expression of HOTAIR has been linked to cancer metastasis282. Another lncRNA, MEG3 regulates PTEN/PI3K/AKT signaling cascade in neurons throughout the synaptic plasticity process 283. Moreover, MEG3 has been significantly down-regulated in the retinas of diabetic mice and knockdown of MEG3 leads to retinal vessel dysfunction exaggeration284. It is necessary to study the role of lncRNAs in KC pathogenesis.

Identifying KC-associated molecular pathways has been investigated using different techniques including microarray, RT-PCR, mass spectrometry, and western blots 191; 192;

196; 198; 211; 212. Recently Kabza et al. have published a RNA-Seq analysis for keratoconic corneas; however, they compared the RNAs expression versus non-KC diseased affected corneas (including: bullous keratopathy, corneal scarring, ulcers, and

perforations), which probably affects their data analysis 198. In our analysis, we avoided this drawback by using unaffected normal corneas as controls. We identified 436 coding

RNAs and 586 lncRNAs to be differentially expressed in KC-affected corneas. Analysis of our data showed that various corneal functions have been compromised. Genes involved in regulation of corneal integrity, transparency and cellular adhesion have differential expressions in the keratoconic corneas. These genes include KRT6A, KRT13, KRT15,

KRT19, COL8A1 and COL8A2. Corneal wound healing linked genes including AQP5 showed disrupted expression. Consistent with another study, expression of AQP5 was reduced by 57 folds in KC, which highlights its main role in KC pathogenesis227. Our data have shown differential expression of various coding RNAs and lncRNAs to be attributed to oxidative stress, which is one of postulated mechanisms in KC progression.

For example, GPX3 (Glutathione peroxidase 3, catalyzes H2O2 reduction) was 16-fold lower than the normal control. GPX has shown significantly depleted expression in rabbit corneas after exposing to ultraviolet-B (UV-B) rays 285. Intriguingly, UV rays are a contributing factor in KC progression, which can be linked to the lessened expression of antioxidant enzymes and accumulation of reactive oxygen species 286.

It is well known that in both physiological and pathological state genes exert their roles via interacting with each other. Complimentary to the traditional differential expression analysis, it is necessary to perform a differential co-expression analysis in cases and controls by grouping the genes into different expression patterns and associating them with similar biological processes287; 288. Additionally, co-expression or correlated analysis is a valuable approach to annotate many lncRNAs through their correlation with coding genes and derive their biological functions. We found 296 pairs of coding and lncRNAs to have

significant differential correlations between cases and controls. Using Cytoscape, we were able to visualize the networking between those correlated genes.

Accordingly, we established a specific filtering pipeline to select the candidate mutation(s) in a four-generation family affected with KC. Through WES we identified a missense mutation in PPIP5K2 (rs35671301, c.1255T>G, p.Ser419Ala). Another genetic mutation was identified on PCSK1 intronic region (rs373951075, c.1096-10C>T) after filtering the

WGS variants. We could have missed the intronic mutation with WES only. We searched for additional PPIP5K2 mutations in other KC-affected families. Interestingly, we found another missense mutation in the PPIP5K2 (rs781831998, c.2528A>G, p.Asn843Ser) in a two-generation family. Across different vertebrates, PPIP5K2 protein sequence is highly conserved at aa419 and aa843, while the PCSK1 intronic variant is not conserved with both alleles present in different vertebrates almost equally. These data supports the potential role of the missense PPIP5K2 mutations in KC pathogenesis, which prompt us to study the impact of the PPIP5K2 missense mutations.

PPIP5K2 encodes the diphosphoinositol pentakisphosphate kinases with both phosphatase and kinase activities. It belongs to the inositol metabolic cascade, which starts with the phosphorylation of the secondary messenger, inositol trisphosphate (InsP3), to inositol hexakisphosphate (InsP6). Subsequently, inositol hexakisphosphate kinases (IP6Ks) and

PPIP5Ks raise the inositol pyrophosphates (PP-InsP) messengers, a unique class of signaling molecules that have one or two high-energy phosphoanhydride bonds289. PP-

InsP includes InsP7 and InsP8, which regulate multiple cellular activities including apoptosis, vesicle trafficking, cytoskeletal dynamics, insulin signaling, cell migration, exocytosis, telomere maintenance, and adaptations to environmental stress273; 274.

PPIP5Ks have a triple domain structure composed of an N-terminal kinase domain, a phosphatase domain, and a C-terminal intrinsically disordered region, which facilitates their binding, regulatory, and signaling functions. The kinase domain phosphorylates 5-

InsP7 to InsP8, while the phosphatase domain dephosphorylates InsP8 back to 5-InsP7.

The phosphatase domain includes a Pleckstrin Homology (PH) region which is contained in various signaling proteins and can bind to phospholipids290. PPIP5Ks have two homologs PPIP5K1 and PPIP5K2.

For the first time we have identified two novel mutations in the PPIP5K2 gene in two different families affected with KC. We decided to figure out the expression pattern of

PPIP5K2 in the corneas. PPIP5K2 has a much higher expression than PPIP5K1 in both epithelial and stromal cells indicating the potential role of PPIP5K2 in the human cornea.

Although the expression of PPIP5K2 transcript was high in both the HCEC and HCSF, immunofluorescence staining indicated that only the epithelium have shown high expression of PPIP5K2 protein. The possible explanation is either the PPIP5K2 transcript in the stromal cells does not translated to the protein, or the change in the phenotype of stromal keratocytes to fibroblasts led to the increased expression of PPIP5K2. As previously mentioned, both epithelium and stroma show histopathological changes in KC.

Low or even no expression of PPIP5K2 in the human keratocytes does not rule out the possible influence of PPIP5K2 on the corneal keratocytes. This is because many reasons including: 1) there are a dual direction communication between stroma and epithelium during normal development, homeostasis, and wound healing291, 2) the epithelium layer directly interacts with the external insults and several environmental factors which affect

the epithelium renewal 292, 3) up to date the site of KC initiation remains unknown, with hypotheses implicating the epithelium and the stroma293; 294.

Next, we studied how the two mutations influence the catalytic activities of PPIP5K2. Both mutations are located in the phosphatase domain of PPIP5K2. We collaborated with Dr.

Stephen B. Shears, an expert in the biological function of PPIP5K2, and his colleagues to overexpress PPIP5K2 with either of these mutations and check the catalytic activities of mutated PPIP5K2 in vitro. The p.Asn843Ser causes a significant reduction in the phosphatase activity (Figure 11A), which was aggravated in the presence of inorganic phosphate (Pi) (Data not shown). Moreover, PPIP5K2 kinase showed higher activity with both mutations (p.Ser419Ala and p.Asn843Ser); however, it was only significant with p.Asn843Ser. These results imply the impact of p.Asn843Ser mutation on the enzymatic activity, but do not discount the functional impact of p.Ser419Ala mutation. The conservation of p.Ser419 amino acid across different species reflects the importance of this amino acid in the PPIP5K2 biological function. The lack of crystal structure of the phosphatase domain prevents us from computationally predicting the functional deficit due to the p.Ser419Ala mutation.

Dr. Shears’ group has previously shown that Pi directly regulates the catalytic activities of

PPIP5K2 by inhibiting phosphatase and stimulating kinase activities 261. Moreover, following cellular starvation, InsP8 levels decreased due to restoration in the InsP8 phosphatase activity. Later, the same group has also shown that a hyperkinase low phosphatase mutant (R399A) in PPIP5K1 leads to decrease in the cellular ATP level 260.

These studies provide a proof of concept about the bioenergetics influence of the PPIP5Ks.

In accordance, we postulate that our novel PPIP5K2 mutations are likely to impact the

mitochondrial function, based on two justifications. First is the reported association between KC pathogenesis and mitochondrial dysfunction. Hao et al. have reported KC corneas with significant mtDNA damage, lower mitochondrial membrane potential, higher

ROS levels, and lower ATP levels80. In addition, Pathak et al. have found 18 non- synonymous variants in the mitochondrial complex I of the KC cases 79. Second, the loss of 1,5-InsP8, after PPIP5K knockout in HCT116 cells, has led to significant elevation of mitochondrial bioenergetic capacity and ATP levels. However, the previously mentioned point mutation in the phosphatase domain (R399A) has led to increased 1,5-InsP8 levels and decreased cellular ATP levels 295. It will be valuable to test our hypothesis by assessing the mitochondrial function in the immortalized lymphoblast cells derived from

KC-affected family members carrying these mutations.

PPIP5K2 is a conserved enzyme and the affected amino acids by the two mutations are well conserved across different species from zebrafish to all primates. Thus, we aimed to understand the role of PPIP5K2 by examining the cornea phenotypes using a mouse model lacking the phosphatase domain of PPIP5K2. This helped advance our knowledge about

PPIP5K2 phosphatase domain function to support our genetic findings in KC families.

Comparing against the littermate controls, our OCT analysis showed pathological corneal phenotypes in both Ppip5k2+/K^ and Ppip5k2K^/K^ mice after 3 months of age. As aforementioned, curvature and thickness mapping are used to diagnose KC patients in the clinic. Accordingly, we collaborated with Dr. Anthony Kao to analyze our OCT data among the three groups. Interestingly, both Ppip5k2+/K^ and Ppip5k2K^/K^ mice showed significant localized thinning and curvature changes. Those phenotype in the mutant mice

may be due to the high levels of InsP8 which has been reported to be elevated in the cell under stress, aging, or high temperature296.

The mouse corneal surface irregularities were only manifested in 40% of both

Ppip5k2+/K^ and Ppip5k2K^/K^ mice, which indicates incomplete penetrance. Incomplete penetrance has been reported in KC linkage analyses including the four generation family that we have studied111. Interestingly, we found that same litters would manifest same phenotype. This observation may be due to the effect of different environmental factors such as the age of the mother, the proximity of the cage to light, and other epigenetic modifiers. We have expected that 100% of Ppip5k2K^/K^ would have the corneal surface irregularities, which did not occur. Probably not only the elevated kinase is the main inducer for the corneal phenotype, but also other environmental contributors. The best way to understand the inconsistent phenotype in Ppip5k2K^/K^ is to measure the corneal levels of

InsP7 and InsP8 in our mouse model. However, the levels of these pyrophosphates are too low to be detected with HPLC. Thus, we planned to isolate mouse corneal epithelial and stromal cells to measure those pyrophosphates, however, the survival of these cells were low. The loss of the PPIP5K2 phosphatase domain may affect the survival of isolated mouse cornea cells in culture conditions.

The mouse slit lamp examination did not show the corneal bulging phenotype that is considered as one of the main KC manifestations in patients. However, in many cases mouse models may not manifest the identical phenotypes of the matched human disorders including Huntington disease (HD) and Parkinson’s disease (PD). HD is a late-onset neurodegenerative disorder and causes dyskinesia (involuntary movements) in patients.

Significant phenotypic variations have been reported in mouse models, which express

either the truncated or full-length human HTT gene, or the full-length mouse, mutant Htt gene 297. Regarding PD, LRRK2 mutations account for 5–6 % of patients with familial

PD and 1–3 % of sporadic cases. However, most Lrrk2 transgenic mouse models expressed minimal or no neurodegeneration 298. Another example is cardiomyopathies, which encompass a diverse array of phenotypic expression including heart failure and sudden cardiac death. Many groups have reported either partial resemblance or dissociation of the animal models for the human cardiomyopathy phenotype299-301. These examples suggest that mouse model may express a minimal or no human-like phenotype, which is the case in KC, especially being a multifactorial disease. Another possibility for not gaining KC- associated cornea phenotypes in our mouse model is the lack of other environmental, hormonal, or biomechanical factors in mice.

The PCSK1 intronic mutation (rs373951075, c.1096-10G>A) in the four-generation family might also contribute to KC. The mutation located in a consensus sequence for JUND binding, which is the AP-1 transcription factor subunit. JunD binds to 8-mer cAMP- responsive element (CRE) motifs TGACGTCA where it forms hetero dimers with members of the ATF (activating transcription factor) family 277; 278. We found that the fifth nucleotide in this binding region was changed from G to A, which probably affects the binding capability of JUND to ATF members.

Finally, under the supervision of my mentor, I successfully identified the potential mutations involved in familial KC using comprehensive approaches. The analysis of the

RNA-Seq data has advanced out knowledge of KC-related coding/noncoding genes. For the first time, we successfully performed a correlation analysis between the differentially expressed coding RNAs and lncRNAs in KC versus normal corneas. This network analysis

identified several genes that could co-regulate and interact with each other to impact KC pathogenesis. Another strength of my thesis is the identification of two missense mutations in two different KC-affected families within the phosphatase domain of PPIP5K2.

Moreover, the mouse model without the phosphatase domain has demonstrated corneal structural irregularities indicating the role of PPIP5K2 in KC-associated corneal pathogenesis.

Our study also had a few limitations. First, the sample size of the RNA-Seq study is relatively small with only 18 total samples. We plan to expand the sample size to a total of

80 corneal samples. Second, the differentially expressed genes in KC-affected corneas could not be determined to be causal to KC or caused by KC in human patients. Third, the segregation of the identified mutations does not follow autosomal dominant inheritance completely probably due to environmental, hormonal, and biomechanical factors as well as incomplete penetrance of the mutations. Fourth, the MAF of p.Ser419Ala is relatively high in populations with European ancestry. The detailed biochemical functional impact of this mutation remains to be determined.

Overall, we have successfully characterized the differential expression of coding and long noncoding genes in KC-affected human corneas and identified two novel mutations in the

PPIP5K2 gene in two multiplex KC families. Our mouse model with hyperkinase activity of PPIP5K2 manifests localized abnormal anterior corneal curvature and localized regional corneal thinning in both Ppip5k2+/K^ and Ppip5k2K^/K^ mice. These mice could represent an important novel animal model for further KC research in identifying novel targets and developing novel therapies for KC.

5. SUMMARY

• Keratoconus is the most common primary corneal ectasia in the US, affecting

300,000 patients, and one of the most frequent causes of corneal transplantation

• We aimed to identify and characterize the genetic mutations and molecular

pathways involved in the pathogenesis of KC

• Using RNA-Seq we identified 436 coding RNAs and 586 lncRNAs to be

differentially expressed in KC-affected corneas

• Gene ontology analysis revealed disturbances in various corneal functions

including cellular adhesion, oxidative stress response, proliferation, migration,

apoptosis, and wound healing

• Co-expression analysis revealed significant correlations between coding and

lncRNAs in cases versus controls, indicating many potential networks may play

potential roles in KC pathogenesis

• Using whole exome and genome sequencing, we have identified two missense

mutations in PPIP5K2 in two different families affected with KC

• PPIP5K2 is a bi-functional enzyme regulating cellular signaling and homeostasis

• PPIP5K2 is expressed in the cornea higher than its homolog PPIP5K1

• Loss of Ppip5k2 phosphatase domain may cause corneal defects in mice

• Integrating the revolutionized sequencing technology with both in vitro and in

vivo models has significantly promoted our genetic research in keratoconus and

established potential animal models for keratoconus

REFERENCES

Uncategorized References 1. Kang, P.C., Klintworth, G.K., Kim, T., Carlson, A.N., Adelman, R., Stinnett, S., and Afshari, N.A. (2005). Trends in the indications for penetrating keratoplasty, 1980- 2001. Cornea 24, 801-803. 2. Ghosheh, F.R., Cremona, F.A., Rapuano, C.J., Cohen, E.J., Ayres, B.D., Hammersmith, K.M., Raber, I.M., and Laibson, P.R. (2008). Trends in penetrating keratoplasty in the United States 1980-2005. Int Ophthalmol 28, 147- 153. 3. Tuft, S.J., Moodaley, L.C., Gregory, W.M., Davison, C.R., and Buckley, R.J. (1994). Prognostic factors for the progression of keratoconus. Ophthalmology 101, 439- 447. 4. Ertan, A., and Muftuoglu, O. (2008). Keratoconus clinical findings according to different age and gender groups. Cornea 27, 1109-1113. 5. McMillan, J. (2018). Spectrum of Darkness, Agent of Light: Myopia, Keratoconus, Ocular Surface Disease, and Evidence for a Profoundly Vitamin D-dependent Eye. Cureus 10, e2744. 6. Sharif, R., Bak-Nielsen, S., Hjortdal, J., and Karamichos, D. (2018). Pathogenesis of Keratoconus: The intriguing therapeutic potential of Prolactin-inducible protein. Prog Retin Eye Res 67, 150-167. 7. Rebenitsch, R.L., Kymes, S.M., Walline, J.J., and Gordon, M.O. (2011). The lifetime economic burden of keratoconus: a decision analysis using a markov model. American journal of ophthalmology 151, 768-773 e762. 8. Kymes, S.M., Walline, J.J., Zadnik, K., Gordon, M.O., and Collaborative Longitudinal Evaluation of Keratoconus study, g. (2004). Quality of life in keratoconus. American journal of ophthalmology 138, 527-535. 9. Hanlon, S.D., Behzad, A.R., Sakai, L.Y., and Burns, A.R. (2015). Corneal stroma microfibrils. Exp Eye Res 132, 198-207. 10. DelMonte, D.W., and Kim, T. (2011). Anatomy and physiology of the cornea. J Cataract Refr Surg 37, 588-598. 11. Feizi, S., Jafarinasab, M.R., Karimian, F., Hasanpour, H., and Masudi, A. (2014). Central and peripheral corneal thickness measurement in normal and keratoconic eyes using three corneal pachymeters. J Ophthalmic Vis Res 9, 296-304. 12. Martola, E.L., and Baum, J.L. (1968). Central and peripheral corneal thickness. A clinical study. Arch Ophthalmol 79, 28-30. 13. Cwiklik, L. (2016). Tear film lipid layer: A molecular level view. Biochim Biophys Acta 1858, 2421-2430. 14. Goel, M., Picciani, R.G., Lee, R.K., and Bhattacharya, S.K. (2010). Aqueous humor dynamics: a review. Open Ophthalmol J 4, 52-59.

15. Beebe, D.C. (2008). Maintaining transparency: a review of the developmental physiology and pathophysiology of two avascular tissues. Semin Cell Dev Biol 19, 125-133. 16. Khaled, M.L., Helwa, I., Drewry, M., Seremwe, M., Estes, A., and Liu, Y. (2017). Molecular and Histopathological Changes Associated with Keratoconus. Biomed Res Int 2017, 7803029. 17. Farjo, A., McDermott, M., and Soong, H.K. (2009). Corneal anatomy, physiology, and wound healing. Ophthalmology 44, 203-208. 18. Meek, K.M., and Knupp, C. (2015). Corneal structure and transparency. Prog Retin Eye Res 49, 1-16. 19. Boote, C., Dennis, S., Newton, R.H., Puri, H., and Meek, K.M. (2003). Collagen fibrils appear more closely packed in the prepupillary cornea: optical and biomechanical implications. Invest Ophthalmol Vis Sci 44, 2941-2948. 20. Giugliani, R., Federhen, A., Rojas, M.V., Vieira, T., Artigalas, O., Pinto, L.L., Azevedo, A.C., Acosta, A., Bonfim, C., Lourenco, C.M., et al. (2010). Mucopolysaccharidosis I, II, and VI: Brief review and guidelines for treatment. Genet Mol Biol 33, 589-604. 21. Hovakimyan, M., Falke, K., Stahnke, T., Guthoff, R., Witt, M., Wree, A., and Stachs, O. (2014). Morphological analysis of quiescent and activated keratocytes: a review of ex vivo and in vivo findings. Curr Eye Res 39, 1129-1144. 22. McTigue, J.W. (1967). The human cornea: a light and electron microscopic study of the normal cornea and its alterations in various dystrophies. Transact American Ophthalmol Society 65, 591-660. 23. Efron, N., Perez-Gomez, I., Mutalib, H.A., and Hollingsworth, J. (2001). Confocal microscopy of the normal human cornea. Cont Lens Anterior Eye 24, 16-24. 24. Bonanno, J.A. (2012). Molecular mechanisms underlying the corneal endothelial pump. Exp Eye Res 95, 2-7. 25. He, Z., Forest, F., Gain, P., Rageade, D., Bernard, A., Acquart, S., Peoc'h, M., Defoe, D.M., and Thuret, G. (2016). 3D map of the human corneal endothelial cell. Sci Rep 6, 29047. 26. Karolak, J.A., and Gajecka, M. (2017). Genomic strategies to understand causes of keratoconus. Mol Genet Genomics 292, 251-269. 27. Rabinowitz, Y.S. (1998). Keratoconus. Survey of ophthalmology 42, 297-319. 28. Galvis, V., Sherwin, T., Tello, A., Merayo, J., Barrera, R., and Acera, A. (2015). Keratoconus: an inflammatory disorder? Eye (Lond) 29, 843-859. 29. Ionescu, C., Corbu, C.G., Tanase, C., Jonescu-Cuypers, C., Nicula, C., Dascalescu, D., Cristea, M., and Voinea, L.M. (2016). Inflammatory Biomarkers Profile as Microenvironmental Expression in Keratoconus. Dis Markers 2016, 1243819. 30. Gorskova, E.N., and Sevost'ianov, E.N. (1998). [Epidemiology of keratoconus in the Urals]. Vestn Oftalmol 114, 38-40. 31. Nielsen, K., Hjortdal, J., Aagaard Nohr, E., and Ehlers, N. (2007). Incidence and prevalence of keratoconus in Denmark. Acta ophthalmologica Scandinavica 85, 890-892. 32. Ziaei, H., Jafarinasab, M.R., Javadi, M.A., Karimian, F., Poorsalman, H., Mahdavi, M., Shoja, M.R., and Katibeh, M. (2012). Epidemiology of keratoconus in an Iranian population. Cornea 31, 1044-1047.

33. Jonas, J.B., Nangia, V., Matin, A., Kulkarni, M., and Bhojwani, K. (2009). Prevalence and associations of keratoconus in rural maharashtra in central India: the central India eye and medical study. American journal of ophthalmology 148, 760-765. 34. Romero-Jimenez, M., Santodomingo-Rubido, J., and Wolffsohn, J.S. (2010). Keratoconus: a review. Cont Lens Anterior Eye 33, 157-166; quiz 205. 35. Fan, R., Chan, T.C., Prakash, G., and Jhanji, V. (2018). Applications of corneal topography and tomography: a review. Clin Exp Ophthalmol 46, 133-146. 36. Wilson, S.E., Lin, D.T., and Klyce, S.D. (1991). Corneal topography of keratoconus. Cornea 10, 2-8. 37. Belin, M.W., and Khachikian, S.S. (2006). New devices and clinical implications for measuring corneal thickness. Clin Exp Ophthalmol 34, 729-731. 38. McGhee, C.N., Kim, B.Z., and Wilson, P.J. (2015). Contemporary Treatment Paradigms in Keratoconus. Cornea 34 Suppl 10, S16-23. 39. Rathi, V.M., Mandathara, P.S., and Dumpati, S. (2013). Contact lens in keratoconus. Indian J Ophthalmol 61, 410-415. 40. Parker, J.S., van Dijk, K., and Melles, G.R. (2015). Treatment options for advanced keratoconus: A review. Survey of ophthalmology 60, 459-480. 41. Watson, S.L., Ramsay, A., Dart, J.K., Bunce, C., and Craig, E. (2004). Comparison of deep lamellar keratoplasty and penetrating keratoplasty in patients with keratoconus. Ophthalmology 111, 1676-1682. 42. Andreanos, K.D., Hashemi, K., Petrelli, M., Droutsas, K., Georgalas, I., and Kymionis, G.D. (2017). Keratoconus Treatment Algorithm. Ophthalmol Ther 6, 245-262. 43. Wollensak, G., Spoerl, E., and Seiler, T. (2003). Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. American journal of ophthalmology 135, 620-627. 44. Raiskup-Wolf, F., Hoyer, A., Spoerl, E., and Pillunat, L.E. (2008). Collagen crosslinking with riboflavin and ultraviolet-A light in keratoconus: long-term results. J Cataract Refract Surg 34, 796-801. 45. Mathew, J.H., Goosey, J.D., and Bergmanson, J.P. (2011). Quantified histopathology of the keratoconic cornea. Optom Vis Sci 88, 988-997. 46. Brewitt, H. (1979). Light-Microscope and Scanning Electron-Microscope Findings in Acute Keratoconus. Klin Monatsbl Augenh 174, 605-613. 47. Scroggs, M.W., and Proia, A.D. (1992). Histopathological variation in keratoconus. Cornea 11, 553-559. 48. Hollingsworth, J.G., Efron, N., and Tullo, A.B. (2005). In vivo corneal confocal microscopy in keratoconus. Ophthalmic Physiol Opt 25, 254-260. 49. Efron, N., and Hollingsworth, J.G. (2008). New perspectives on keratoconus as revealed by corneal confocal microscopy. Clin Exp Optom 91, 34-55. 50. Sandali, O., El Sanharawi, M., Temstet, C., Hamiche, T., Galan, A., Ghouali, W., Goemaere, I., Basli, E., Borderie, V., and Laroche, L. (2013). Fourier-domain optical coherence tomography imaging in keratoconus: a corneal structural classification. Ophthalmology 120, 2403-2412.

51. Brautaset, R.L., Nilsson, M., Miller, W.L., Leach, N.E., Tukler, J.H., and Bergmanson, J.P. (2013). Central and peripheral corneal thinning in keratoconus. Cornea 32, 257-261. 52. Bitirgen, G., Ozkagnici, A., Bozkurt, B., and Malik, R.A. (2015). In vivo corneal confocal microscopic analysis in patients with keratoconus. Int J Ophthalmol 8, 534-539. 53. Weed, K.H., MacEwen, C.J., Cox, A., and McGhee, C.N. (2007). Quantitative analysis of corneal microstructure in keratoconus utilising in vivo confocal microscopy. Eye (Lond) 21, 614-623. 54. Mocan, M.C., Yilmaz, P.T., Irkec, M., and Orhan, M. (2008). In vivo confocal microscopy for the evaluation of corneal microstructure in keratoconus. Curr Eye Res 33, 933-939. 55. Barr, J.T., Schechtman, K.B., Fink, B.A., Pierce, G.E., Pensyl, C.D., Zadnik, K., and Gordon, M.O. (1999). Corneal scarring in the Collaborative Longitudinal Evaluation of Keratoconus (CLEK) Study: baseline prevalence and repeatability of detection. Cornea 18, 34-46. 56. Sherwin, T., and Brookes, N.H. (2004). Morphological changes in keratoconus: pathology or pathogenesis. Clin Exp Ophthalmol 32, 211-217. 57. Sykakis, E., Carley, F., Irion, L., Denton, J., and Hillarby, M.C. (2012). An in depth analysis of histopathological characteristics found in keratoconus. Pathology 44, 234-239. 58. Morishige, N., Wahlert, A.J., Kenney, M.C., Brown, D.J., Kawamoto, K., Chikama, T., Nishida, T., and Jester, J.V. (2007). Second-harmonic imaging microscopy of normal human and keratoconus cornea. Invest Ophthalmol Vis Sci 48, 1087-1094. 59. Takahashi, A., Nakayasu, K., Okisaka, S., and Kanai, A. (1990). [Quantitative analysis of collagen fiber in keratoconus]. Nippon Ganka Gakkai Zasshi 94, 1068- 1073. 60. Erie, J.C., Patel, S.V., McLaren, J.W., Nau, C.B., Hodge, D.O., and Bourne, W.M. (2002). Keratocyte density in keratoconus. A confocal microscopy studyaaInternetAdvance publication at ajo.com Sept 17, 2002. American journal of ophthalmology 134, 689-695. 61. Niederer, R.L., Perumal, D., Sherwin, T., and McGhee, C.N. (2008). Laser scanning in vivo confocal microscopy reveals reduced innervation and reduction in cell density in all layers of the keratoconic cornea. Invest Ophthalmol Vis Sci 49, 2964-2970. 62. Wolter, J.R., Henderson, J.W., and Clahassey, E.G. (1967). Ruptures of descemet's membrane in keratoconus causing acute hydrops and posterior keratoconus. American journal of ophthalmology 63, 1689-1692. 63. Zhou, L., Sawaguchi, S., Twining, S.S., Sugar, J., Feder, R.S., and Yue, B.Y. (1998). Expression of degradative enzymes and protease inhibitors in corneas with keratoconus. Invest Ophthalmol Vis Sci 39, 1117-1124. 64. Cristina Kenney, M., and Brown, D.J. (2003). The Cascade Hypothesis of Keratoconus. Cont Lens and Anterior Eye 26, 139-146. 65. Balasubramanian, S.A., Pye, D.C., and Willcox, M.D. (2010). Are proteinases the reason for keratoconus? Curr Eye Res 35, 185-191.

66. Sawaguchi, S., Yue, B.Y., Sugar, J., and Gilboy, J.E. (1989). Lysosomal enzyme abnormalities in keratoconus. Arch Ophthalmol 107, 1507-1510. 67. Nelidova, D., and Sherwin, T. (2012). Advances in Ophthalmology.(InTech). 68. Sherwin, T., Brookes, N.H., Loh, I.P., Poole, C.A., and Clover, G.M. (2002). Cellular incursion into Bowman's membrane in the peripheral cone of the keratoconic cornea. Exp Eye Res 74, 473-482. 69. Whitelock, R.B., Fukuchi, T., Zhou, L., Twining, S.S., Sugar, J., Feder, R.S., and Yue, B.Y. (1997). Cathepsin G, acid phosphatase, and alpha 1-proteinase inhibitor messenger RNA levels in keratoconus corneas. Invest Ophthalmol Vis Sci 38, 529-534. 70. Sawaguchi, S., Twining, S.S., Yue, B.Y., Wilson, P.M., Sugar, J., and Chan, S.K. (1990). Alpha-1 proteinase inhibitor levels in keratoconus. Exp Eye Res 50, 549- 554. 71. Wisse, R.P., Kuiper, J.J., Gans, R., Imhof, S., Radstake, T.R., and Van der Lelij, A. (2015). Cytokine Expression in Keratoconus and its Corneal Microenvironment: A Systematic Review. Ocul Surf 13, 272-283. 72. Ashby, B.D., Garrett, Q., and Willcox, M.D.P. (2014). Corneal Injuries and Wound Healing – Review of Processes and Therapies. Austin J Clin Ophthalmol 1, 1-25. 73. Nielsen, K., Heegaard, S., Vorum, H., Birkenkamp-Demtroder, K., Ehlers, N., and Orntoft, T.F. (2005). Altered expression of CLC, DSG3, EMP3, S100A2, and SLPI in corneal epithelium from keratoconus patients. Cornea 24, 661-668. 74. Deng, Y., Liu, S., and Cai, R. (2001). Immunohistochemistry research on keratoconus. Yan Ke Xue Bao 17, 65-67, 75. 75. Wojcik, K.A., Blasiak, J., Szaflik, J., and Szaflik, J.P. (2014). Role of biochemical factors in the pathogenesis of keratoconus. Acta Biochim Pol 61, 55-62. 76. Behndig, A., Karlsson, K., Johansson, B.O., Brannstrom, T., and Marklund, S.L. (2001). Superoxide dismutase isoenzymes in the normal and diseased human cornea. Invest Ophthalmol Vis Sci 42, 2293-2296. 77. Atilano, S.R., Coskun, P., Chwa, M., Jordan, N., Reddy, V., Le, K., Wallace, D.C., and Kenney, M.C. (2005). Accumulation of mitochondrial DNA damage in keratoconus corneas. Invest Ophthalmol Vis Sci 46, 1256-1263. 78. Chwa, M., Atilano, S.R., Hertzog, D., Zheng, H., Langberg, J., Kim, D.W., and Kenney, M.C. (2008). Hypersensitive response to oxidative stress in keratoconus corneal fibroblasts. Invest Ophthalmol Vis Sci 49, 4361-4369. 79. Pathak, D., Nayak, B., Singh, M., Sharma, N., Tandon, R., Sinha, R., Titiyal, J.S., and Dada, R. (2011). Mitochondrial complex 1 gene analysis in keratoconus. Mol Vis 17, 1514-1525. 80. Hao, X.D., Chen, Z.L., Qu, M.L., Zhao, X.W., Li, S.X., and Chen, P. (2016). Decreased Integrity, Content, and Increased Transcript Level of Mitochondrial DNA Are Associated with Keratoconus. PLoS One 11, e0165580. 81. McKay, T.B., Hjortdal, J., Sejersen, H., Asara, J.M., Wu, J., and Karamichos, D. (2016). Endocrine and Metabolic Pathways Linked to Keratoconus: Implications for the Role of Hormones in the Stromal Microenvironment. Sci Rep 6, 25534. 82. McKay, T.B., Hjortdal, J., Sejersen, H., and Karamichos, D. (2017). Differential Effects of Hormones on Cellular Metabolism in Keratoconus In Vitro. Sci Rep 7, 42896.

100

83. Hoogewoud, F., Gatzioufas, Z., and Hafezi, F. (2013). Transitory topographical variations in keratoconus during pregnancy. J Refract Surg 29, 144-146. 84. Pizzarello, L.D. (2003). Refractive changes in pregnancy. Graefes Arch Clin Exp Ophthalmol 241, 484-488. 85. Yuksel, E., Yalinbas, D., Aydin, B., and Bilgihan, K. (2016). Keratoconus Progression Induced by In Vitro Fertilization Treatment. J Refract Surg 32, 60-63. 86. Tuft, S.J., Hassan, H., George, S., Frazer, D.G., Willoughby, C.E., and Liskova, P. (2012). Keratoconus in 18 pairs of twins. Acta Ophthalmol 90, e482-486. 87. Rabinowitz, Y.S. (2003). The genetics of keratoconus. Ophthalmol Clin North Am 16, 607-620, vii. 88. Wheeler, J., Hauser, M.A., Afshari, N.A., Allingham, R.R., and Liu, Y. (2012). The Genetics of Keratoconus: A Review. Reprod Syst Sex Disord, 1-19. 89. Liskova, P., Willoughby, C.E., George, S., and Tuft, S.J. (2011). Phenotype and Hereditability of Keratoconus in Twins. Invest Ophthalmol Vis Sci 52, 1094- 1094. 90. Bechara, S.J., Waring, G.O., 3rd, and Insler, M.S. (1996). Keratoconus in two pairs of identical twins. Cornea 15, 90-93. 91. Machin, G. (2009). Non-identical monozygotic twins, intermediate twin types, zygosity testing, and the non-random nature of monozygotic twinning: a review. Am J Med Genet C Semin Med Genet 151C, 110-127. 92. Marceau, K., McMaster, M.T., Smith, T.F., Daams, J.G., van Beijsterveldt, C.E., Boomsma, D.I., and Knopik, V.S. (2016). The Prenatal Environment in Twin Studies: A Review on Chorionicity. Behav Genet 46, 286-303. 93. Gordon-Shaag, A., Millodot, M., Essa, M., Garth, J., Ghara, M., and Shneor, E. (2013). Is consanguinity a risk factor for keratoconus? Optom Vis Sci 90, 448- 454. 94. Jamali, H., Beigi, V., and Sadeghi-Sarvestani, A. (2018). Consanguineous Marriage as a Risk Factor for Developing Keratoconus. Medical Hypothesis, Discovery and Innovation in Ophthalmology 7, 17-21. 95. Burdon, K.P., and Vincent, A.L. (2013). Insights into keratoconus from a genetic perspective. Clin Exp Optom 96, 146-154. 96. Saffra, N., and Reinherz, B. (2015). Keratoconus in an adult with 22q11.2 deletion syndrome. BMJ Case Rep 2015, bcr2014203737. 97. Bassett, A.S., Hodgkinson, K., Chow, E.W., Correia, S., Scutt, L.E., and Weksberg, R. (1998). 22q11 deletion syndrome in adults with schizophrenia. Am J Med Genet 81, 328-337. 98. Elder, M.J. (1994). Leber congenital amaurosis and its association with keratoconus and keratoglobus. Journal of pediatric ophthalmology and strabismus 31, 38-40. 99. Damji, K.F., Sohocki, M.M., Khan, R., Gupta, S.K., Rahim, M., Loyer, M., Hussein, N., Karim, N., Ladak, S.S., Jamal, A., et al. (2001). Leber's congenital amaurosis with anterior keratoconus in Pakistani families is caused by the Trp278X mutation in the AIPL1 gene on 17p. Can J Ophthalmol 36, 252-259. 100. Hughes, A.E., Bradley, D.T., Campbell, M., Lechner, J., Dash, D.P., Simpson, D.A., and Willoughby, C.E. (2011). Mutation altering the miR-184 seed region causes familial keratoconus with cataract. American journal of human genetics 89, 628- 633.

101

101. Wilhelmus, K.R. (1995). Keratoconus and progressive cone dystrophy. Ophthalmologica 209, 278-279. 102. Robertson, I. (1975). Keratoconus and the Ehlers-Danlos syndrome: a new aspect of keratoconus. The Medical journal of Australia 1, 571-573. 103. Beckh, U., Schonherr, U., and Naumann, G.O. (1995). [Autosomal dominant keratoconus as the chief ocular symptom in Lobstein osteogenesis imperfecta tarda]. Klinische Monatsblatter fur Augenheilkunde 206, 268-272. 104. Wang, Y., Rabinowitz, Y.S., Rotter, J.I., and Yang, H. (2000). Genetic epidemiological study of keratoconus: evidence for major gene determination. Am J Med Genet 93, 403-409. 105. Kriszt, A., Losonczy, G., Berta, A., Vereb, G., and Takacs, L. (2014). Segregation analysis suggests that keratoconus is a complex non-mendelian disease. Acta Ophthalmol 92, e562-568. 106. Londin, E., Yadav, P., Surrey, S., Kricka, L.J., and Fortina, P. (2013). Use of Linkage Analysis, Genome-Wide Association Studies, and Next-Generation Sequencing in the Identification of Disease-Causing Mutations. In Pharmacogenomics: Methods and Protocols, F. Innocenti and R.H.N. van Schaik, eds. (Totowa, NJ, Humana Press), pp 127-146. 107. Nowak, D.M., and Gajecka, M. (2011). The genetics of keratoconus. Middle East Afr J Ophthalmol 18, 2-6. 108. Nielsen, K., Hjortdal, J., Pihlmann, M., and Corydon, T.J. (2013). Update on the keratoconus genetics. Acta Ophthalmol 91, 106-113. 109. Dash, D.P., Silvestri, G., and Hughes, A.E. (2006). Fine mapping of the keratoconus with cataract locus on chromosome 15q and candidate gene analysis. Mol Vis 12, 499-505. 110. Hughes, A.E., Dash, D.P., Jackson, A.J., Frazer, D.G., and Silvestri, G. (2003). Familial keratoconus with cataract: linkage to the long arm of chromosome 15 and exclusion of candidate genes. Invest Ophthalmol Vis Sci 44, 5063-5066. 111. Tang, Y.G., Rabinowitz, Y.S., Taylor, K.D., Li, X., Hu, M., Picornell, Y., and Yang, H. (2005). Genomewide linkage scan in a multigeneration Caucasian pedigree identifies a novel locus for keratoconus on chromosome 5q14.3-q21.1. Genet Med 7, 397-405. 112. Bisceglia, L., De Bonis, P., Pizzicoli, C., Fischetti, L., Laborante, A., Di Perna, M., Giuliani, F., Delle Noci, N., Buzzonetti, L., and Zelante, L. (2009). Linkage analysis in keratoconus: replication of locus 5q21.2 and identification of other suggestive Loci. Invest Ophthalmol Vis Sci 50, 1081-1086. 113. Rosenfeld, J.A., Drautz, J.M., Clericuzio, C.L., Cushing, T., Raskin, S., Martin, J., Tervo, R.C., Pitarque, J.A., Nowak, D.M., Karolak, J.A., et al. (2011). Deletions and duplications of developmental pathway genes in 5q31 contribute to abnormal phenotypes. Am J Med Genet A 155A, 1906-1916. 114. Li, X., Bykhovskaya, Y., Tang, Y.G., Picornell, Y., Haritunians, T., Aldave, A.J., Szczotka-Flynn, L., Iyengar, S.K., Rotter, J.I., Taylor, K.D., et al. (2013). An association between the calpastatin (CAST) gene and keratoconus. Cornea 32, 696-701. 115. Czugala, M., Karolak, J.A., Nowak, D.M., Polakowski, P., Pitarque, J., Molinari, A., Rydzanicz, M., Bejjani, B.A., Yue, B.Y., Szaflik, J.P., et al. (2012). Novel

102

mutation and three other sequence variants segregating with phenotype at keratoconus 13q32 susceptibility locus. European journal of human genetics : EJHG 20, 389-397. 116. Nowak, D.M., Karolak, J.A., Kubiak, J., Gut, M., Pitarque, J.A., Molinari, A., Bejjani, B.A., and Gajecka, M. (2013). Substitution at IL1RN and deletion at SLC4A11 segregating with phenotype in familial keratoconus. Invest Ophthalmol Vis Sci 54, 2207-2215. 117. Shearer, T.R., Azuma, M., David, L.L., Yamagata, Y., and Murachi, T. (1990). Calpain and calpastatin in rabbit corneal epithelium. Curr Eye Res 9, 39-44. 118. Ott, J., Kamatani, Y., and Lathrop, M. (2011). Family-based designs for genome- wide association studies. Nat Rev Genet 12, 465-474. 119. Pe'er, I., Yelensky, R., Altshuler, D., and Daly, M.J. (2008). Estimation of the multiple testing burden for genomewide association studies of nearly all common variants. Genet Epidemiol 32, 381-385. 120. Burdon, K.P., Macgregor, S., Bykhovskaya, Y., Javadiyan, S., Li, X., Laurie, K.J., Muszynska, D., Lindsay, R., Lechner, J., Haritunians, T., et al. (2011). Association of polymorphisms in the hepatocyte growth factor gene promoter with keratoconus. Invest Ophthalmol Vis Sci 52, 8514-8519. 121. You, J., Wen, L., Roufas, A., Hodge, C., Sutton, G., and Madigan, M.C. (2015). Expression of HGF and c-Met Proteins in Human Keratoconus Corneas. J Ophthalmol 2015, 852986. 122. Li, X., Bykhovskaya, Y., Haritunians, T., Siscovick, D., Aldave, A., Szczotka- Flynn, L., Iyengar, S.K., Rotter, J.I., Taylor, K.D., and Rabinowitz, Y.S. (2012). A genome-wide association study identifies a potential novel gene locus for keratoconus, one of the commonest causes for corneal transplantation in developed countries. Hum Mol Genet 21, 421-429. 123. Bae, H.A., Mills, R.A., Lindsay, R.G., Phillips, T., Coster, D.J., Mitchell, P., Wang, J.J., Craig, J.E., and Burdon, K.P. (2013). Replication and meta-analysis of candidate loci identified variation at RAB3GAP1 associated with keratoconus. Invest Ophthalmol Vis Sci 54, 5132-5135. 124. Bykhovskaya, Y., Li, X., Epifantseva, I., Haritunians, T., Siscovick, D., Aldave, A., Szczotka-Flynn, L., Iyengar, S.K., Taylor, K.D., Rotter, J.I., et al. (2012). Variation in the lysyl oxidase (LOX) gene is associated with keratoconus in family-based and case-control studies. Invest Ophthalmol Vis Sci 53, 4152-4157. 125. Kagan, H.M., and Trackman, P.C. (1991). Properties and function of lysyl oxidase. Am J Respir Cell Mol Biol 5, 206-210. 126. Liskova, P., Dudakova, L., Krepelova, A., Klema, J., and Hysi, P.G. (2017). Replication of SNP associations with keratoconus in a Czech cohort. PLoS One 12, e0172365. 127. Lu, Y., Vitart, V., Burdon, K.P., Khor, C.C., Bykhovskaya, Y., Mirshahi, A., Hewitt, A.W., Koehn, D., Hysi, P.G., Ramdas, W.D., et al. (2013). Genome-wide association analyses identify multiple loci associated with central corneal thickness and keratoconus. Nature genetics 45, 155-163. 128. Gao, X., Gauderman, W.J., Liu, Y., Marjoram, P., Torres, M., Haritunians, T., Kuo, J.Z., Chen, Y.D., Allingham, R.R., Hauser, M.A., et al. (2013). A genome-wide

103

association study of central corneal thickness in Latinos. Invest Ophthalmol Vis Sci 54, 2435-2443. 129. Sahebjada, S., Schache, M., Richardson, A.J., Snibson, G., MacGregor, S., Daniell, M., and Baird, P.N. (2013). Evaluating the association between keratoconus and the corneal thickness genes in an independent Australian population. Invest Ophthalmol Vis Sci 54, 8224-8228. 130. Hao, X.D., Chen, P., Chen, Z.L., Li, S.X., and Wang, Y. (2015). Evaluating the Association between Keratoconus and Reported Genetic Loci in a Han Chinese Population. Ophthalmic Genet 36, 132-136. 131. Li, X., Bykhovskaya, Y., Canedo, A.L., Haritunians, T., Siscovick, D., Aldave, A.J., Szczotka-Flynn, L., Iyengar, S.K., Rotter, J.I., Taylor, K.D., et al. (2013). Genetic association of COL5A1 variants in keratoconus patients suggests a complex connection between corneal thinning and keratoconus. Invest Ophthalmol Vis Sci 54, 2696-2704. 132. Cuellar-Partida, G., Springelkamp, H., Lucas, S.E., Yazar, S., Hewitt, A.W., Iglesias, A.I., Montgomery, G.W., Martin, N.G., Pennell, C.E., van Leeuwen, E.M., et al. (2015). WNT10A exonic variant increases the risk of keratoconus by decreasing corneal thickness. Hum Mol Genet 24, 5060-5068. 133. Heon, E., Greenberg, A., Kopp, K.K., Rootman, D., Vincent, A.L., Billingsley, G., Priston, M., Dorval, K.M., Chow, R.L., McInnes, R.R., et al. (2002). VSX1: a gene for posterior polymorphous dystrophy and keratoconus. Hum Mol Genet 11, 1029-1036. 134. Bisceglia, L., Ciaschetti, M., De Bonis, P., Campo, P.A., Pizzicoli, C., Scala, C., Grifa, M., Ciavarella, P., Delle Noci, N., Vaira, F., et al. (2005). VSX1 mutational analysis in a series of Italian patients affected by keratoconus: detection of a novel mutation. Invest Ophthalmol Vis Sci 46, 39-45. 135. Liskova, P., Ebenezer, N.D., Hysi, P.G., Gwilliam, R., El-Ashry, M.F., Moodaley, L.C., Hau, S., Twa, M., Tuft, S.J., and Bhatacharya, S.S. (2007). Molecular analysis of the VSX1 gene in familial keratoconus. Mol Vis 13, 1887-1891. 136. Tang, Y.G., Picornell, Y., Su, X., Li, X., Yang, H., and Rabinowitz, Y.S. (2008). Three VSX1 gene mutations, L159M, R166W, and H244R, are not associated with keratoconus. Cornea 27, 189-192. 137. Aldave, A.J., Yellore, V.S., Salem, A.K., Yoo, G.L., Rayner, S.A., Yang, H., Tang, G.Y., Piconell, Y., and Rabinowitz, Y.S. (2006). No VSX1 gene mutations associated with keratoconus. Invest Ophthalmol Vis Sci 47, 2820-2822. 138. Tanwar, M., Kumar, M., Nayak, B., Pathak, D., Sharma, N., Titiyal, J.S., and Dada, R. (2010). VSX1 gene analysis in keratoconus. Mol Vis 16, 2395-2401. 139. Dash, D.P., George, S., O'Prey, D., Burns, D., Nabili, S., Donnelly, U., Hughes, A.E., Silvestri, G., Jackson, J., Frazer, D., et al. (2010). Mutational screening of VSX1 in keratoconus patients from the European population. Eye (Lond) 24, 1085-1092. 140. Stabuc-Silih, M., Strazisar, M., Hawlina, M., and Glavac, D. (2010). Absence of pathogenic mutations in VSX1 and SOD1 genes in patients with keratoconus. Cornea 29, 172-176.

104

141. Verma, A., Das, M., Srinivasan, M., Prajna, N.V., and Sundaresan, P. (2013). Investigation of VSX1 sequence variants in South Indian patients with sporadic cases of keratoconus. BMC Res Notes 6, 103. 142. Dehkordi, F.A., Rashki, A., Bagheri, N., Chaleshtori, M.H., Memarzadeh, E., Salehi, A., Ghatreh, H., Zandi, F., Yazdanpanahi, N., Tabatabaiefar, M.A., et al. (2013). Study of VSX1 mutations in patients with keratoconus in southwest Iran using PCR-single-strand conformation polymorphism/heteroduplex analysis and sequencing method. Acta Cytol 57, 646-651. 143. Moschos, M.M., Kokolakis, N., Gazouli, M., Chatziralli, I.P., Droutsas, D., Anagnou, N.P., and Ladas, I.D. (2015). Polymorphism Analysis of VSX1 and SOD1 Genes in Greek Patients with Keratoconus. Ophthalmic Genet 36, 213-217. 144. Karolak, J.A., Polakowski, P., Szaflik, J., Szaflik, J.P., and Gajecka, M. (2016). Molecular Screening of Keratoconus Susceptibility Sequence Variants in VSX1, TGFBI, DOCK9, STK24, and IPO5 Genes in Polish Patients and Novel TGFBI Variant Identification. Ophthalmic Genet 37, 37-43. 145. Udar, N., Atilano, S.R., Brown, D.J., Holguin, B., Small, K., Nesburn, A.B., and Kenney, M.C. (2006). SOD1: a candidate gene for keratoconus. Invest Ophthalmol Vis Sci 47, 3345-3351. 146. Udar, N., Atilano, S.R., Small, K., Nesburn, A.B., and Kenney, M.C. (2009). SOD1 haplotypes in familial keratoconus. Cornea 28, 902-907. 147. Fridovich, I. (1995). Superoxide radical and superoxide dismutases. Annu Rev Biochem 64, 97-112. 148. Stabuc-Silih, M., Ravnik-Glavac, M., Glavac, D., Hawlina, M., and Strazisar, M. (2009). Polymorphisms in COL4A3 and COL4A4 genes associated with keratoconus. Mol Vis 15, 2848-2860. 149. Kenney, M.C., Nesburn, A.B., Burgeson, R.E., Butkowski, R.J., and Ljubimov, A.V. (1997). Abnormalities of the extracellular matrix in keratoconus corneas. Cornea 16, 345-351. 150. Hudson, B.G., Reeders, S.T., and Tryggvason, K. (1993). Type IV collagen: structure, gene organization, and role in human diseases. Molecular basis of Goodpasture and Alport syndromes and diffuse leiomyomatosis. J Biol Chem 268, 26033-26036. 151. Dimasi, D.P., Burdon, K.P., and Craig, J.E. (2010). The genetics of central corneal thickness. Br J Ophthalmol 94, 971-976. 152. Hoehn, R., Zeller, T., Verhoeven, V.J., Grus, F., Adler, M., Wolfs, R.C., Uitterlinden, A.G., Castagne, R., Schillert, A., Klaver, C.C., et al. (2012). Population-based meta-analysis in Caucasians confirms association with COL5A1 and ZNF469 but not COL8A2 with central corneal thickness. Hum Genet 131, 1783-1793. 153. Vitart, V., Bencic, G., Hayward, C., Skunca Herman, J., Huffman, J., Campbell, S., Bucan, K., Navarro, P., Gunjaca, G., Marin, J., et al. (2010). New loci associated with central cornea thickness include COL5A1, AKAP13 and AVGR8. Hum Mol Genet 19, 4304-4311. 154. Vithana, E.N., Aung, T., Khor, C.C., Cornes, B.K., Tay, W.T., Sim, X., Lavanya, R., Wu, R., Zheng, Y., Hibberd, M.L., et al. (2011). Collagen-related genes influence the glaucoma risk factor, central corneal thickness. Hum Mol Genet 20, 649-658.

105

155. Lu, Y., Dimasi, D.P., Hysi, P.G., Hewitt, A.W., Burdon, K.P., Toh, T., Ruddle, J.B., Li, Y.J., Mitchell, P., Healey, P.R., et al. (2010). Common genetic variants near the Brittle Cornea Syndrome locus ZNF469 influence the blinding disease risk factor central corneal thickness. PLoS Genet 6, e1000947. 156. Iglesias, A.I., Mishra, A., Vitart, V., Bykhovskaya, Y., Hohn, R., Springelkamp, H., Cuellar-Partida, G., Gharahkhani, P., Bailey, J.N.C., Willoughby, C.E., et al. (2018). Cross-ancestry genome-wide association analysis of corneal thickness strengthens link between complex and Mendelian eye diseases. Nature communications 9, 1864. 157. Pedersen, U., and Bramsen, T. (1984). Central corneal thickness in osteogenesis imperfecta and otosclerosis. ORL J Otorhinolaryngol Relat Spec 46, 38-41. 158. Evereklioglu, C., Madenci, E., Bayazit, Y.A., Yilmaz, K., Balat, A., and Bekir, N.A. (2002). Central corneal thickness is lower in osteogenesis imperfecta and negatively correlates with the presence of blue sclera. Ophthalmic Physiol Opt 22, 511-515. 159. Abu-Amero, K.K., Helwa, I., Al-Muammar, A., Strickland, S., Hauser, M.A., Allingham, R.R., and Liu, Y. (2015). Case-control association between CCT- associated variants and keratoconus in a Saudi Arabian population. J Negat Results Biomed 14, 10. 160. Yeh, L.K., Liu, C.Y., Kao, W.W., Huang, C.J., Hu, F.R., Chien, C.L., and Wang, I.J. (2010). Knockdown of zebrafish lumican gene (zlum) causes scleral thinning and increased size of scleral coats. J Biol Chem 285, 28141-28155. 161. Chakravarti, S., Magnuson, T., Lass, J.H., Jepsen, K.J., LaMantia, C., and Carroll, H. (1998). Lumican regulates collagen fibril assembly: skin fragility and corneal opacity in the absence of lumican. J Cell Biol 141, 1277-1286. 162. Saika, S., Shiraishi, A., Liu, C.Y., Funderburgh, J.L., Kao, C.W., Converse, R.L., and Kao, W.W. (2000). Role of lumican in the corneal epithelium during wound healing. J Biol Chem 275, 2607-2612. 163. Saika, S., Saika, S., Liu, C.Y., Azhar, M., Sanford, L.P., Doetschman, T., Gendron, R.L., Kao, C.W., and Kao, W.W. (2001). TGFbeta2 in corneal morphogenesis during mouse embryonic development. Dev Biol 240, 419-432. 164. Liu, C.Y., Birk, D.E., Hassell, J.R., Kane, B., and Kao, W.W. (2003). Keratocan- deficient mice display alterations in corneal structure. J Biol Chem 278, 21672- 21677. 165. White, T.L., Lewis, P., Hayes, S., Fergusson, J., Bell, J., Farinha, L., White, N.S., Pereira, L.V., and Meek, K.M. (2017). The Structural Role of Elastic Fibers in the Cornea Investigated Using a Mouse Model for Marfan Syndrome. Invest Ophthalmol Vis Sci 58, 2106-2116. 166. Wang, J.L., Yang, X., Xia, K., Hu, Z.M., Weng, L., Jin, X., Jiang, H., Zhang, P., Shen, L., Guo, J.F., et al. (2010). TGM6 identified as a novel causative gene of spinocerebellar ataxias using exome sequencing. Brain 133, 3510-3518. 167. Ng, S.B., Bigham, A.W., Buckingham, K.J., Hannibal, M.C., McMillin, M.J., Gildersleeve, H.I., Beck, A.E., Tabor, H.K., Cooper, G.M., Mefford, H.C., et al. (2010). Exome sequencing identifies MLL2 mutations as a cause of Kabuki syndrome. Nature genetics 42, 790-793.

106

168. Ma, A.S., Grigg, J.R., Ho, G., Prokudin, I., Farnsworth, E., Holman, K., Cheng, A., Billson, F.A., Martin, F., Fraser, C., et al. (2016). Sporadic and Familial Congenital Cataracts: Mutational Spectrum and New Diagnoses Using Next- Generation Sequencing. Hum Mutat 37, 371-384. 169. Bilguvar, K., Ozturk, A.K., Louvi, A., Kwan, K.Y., Choi, M., Tatli, B., Yalnizoglu, D., Tuysuz, B., Caglayan, A.O., Gokben, S., et al. (2010). Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Nature 467, 207-210. 170. Zhang, W., Cui, H., and Wong, L.-J.C. (2014). Application of Next Generation Sequencing to Molecular Diagnosis of Inherited Diseases. In Chemical Diagnostics: From Bench to Bedside, N. L.S. Tang and T. Poon, eds. (Berlin, Heidelberg, Springer Berlin Heidelberg), pp 19-45. 171. Davey, J.W., Hohenlohe, P.A., Etter, P.D., Boone, J.Q., Catchen, J.M., and Blaxter, M.L. (2011). Genome-wide genetic marker discovery and genotyping using next- generation sequencing. Nat Rev Genet 12, 499-510. 172. Koboldt, D.C., Steinberg, K.M., Larson, D.E., Wilson, R.K., and Mardis, E.R. (2013). The next-generation sequencing revolution and its impact on genomics. Cell 155, 27-38. 173. Weedon, M.N., Hastings, R., Caswell, R., Xie, W., Paszkiewicz, K., Antoniadi, T., Williams, M., King, C., Greenhalgh, L., Newbury-Ecob, R., et al. (2011). Exome sequencing identifies a DYNC1H1 mutation in a large pedigree with dominant axonal Charcot-Marie-Tooth disease. American journal of human genetics 89, 308-312. 174. Lupski, J.R., Reid, J.G., Gonzaga-Jauregui, C., Rio Deiros, D., Chen, D.C., Nazareth, L., Bainbridge, M., Dinh, H., Jing, C., Wheeler, D.A., et al. (2010). Whole-genome sequencing in a patient with Charcot-Marie-Tooth neuropathy. N Engl J Med 362, 1181-1191. 175. Costa, V., Aprile, M., Esposito, R., and Ciccodicola, A. (2013). RNA-Seq and human complex diseases: recent accomplishments and future perspectives. European journal of human genetics : EJHG 21, 134-142. 176. Li, Y., Liu, X., Tang, H., Yang, H., and Meng, X. (2017). RNA Sequencing Uncovers Molecular Mechanisms Underlying Pathological Complete Response to Chemotherapy in Patients with Operable Breast Cancer. Med Sci Monit 23, 4321- 4327. 177. Wang, H., Li, Y., Ryder, J.W., Hole, J.T., Ebert, P.J., Airey, D.C., Qian, H.R., Logsdon, B., Fisher, A., Ahmed, Z., et al. (2018). Genome-wide RNAseq study of the molecular mechanisms underlying microglia activation in response to pathological tau perturbation in the rTg4510 tau transgenic animal model. Mol Neurodegener 13, 65. 178. Han, Y., Gao, S., Muegge, K., Zhang, W., and Zhou, B. (2015). Advanced Applications of RNA Sequencing and Challenges. Bioinform Biol Insights 9, 29- 46. 179. Gordon-Shaag, A., Millodot, M., Shneor, E., and Liu, Y. (2015). The genetic and environmental factors for keratoconus. Biomed Res Int 2015, 795738.

107

180. Karamichos, D., Zieske, J.D., Sejersen, H., Sarker-Nag, A., Asara, J.M., and Hjortdal, J. (2015). Tear metabolite changes in keratoconus. Exp Eye Res 132, 1- 8. 181. McKay, T.B., Hjortdal, J., Priyadarsini, S., and Karamichos, D. (2017). Acute hypoxia influences collagen and matrix metalloproteinase expression by human keratoconus cells in vitro. PLoS One 12, e0176017. 182. Karamichos, D., Funderburgh, M.L., Hutcheon, A.E., Zieske, J.D., Du, Y., Wu, J., and Funderburgh, J.L. (2014). A role for topographic cues in the organization of collagenous matrix by corneal fibroblasts and stem cells. PLoS One 9, e86260. 183. Wang, Z., Gerstein, M., and Snyder, M. (2009). RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10, 57-63. 184. Bradford, J.R., Hey, Y., Yates, T., Li, Y., Pepper, S.D., and Miller, C.J. (2010). A comparison of massively parallel nucleotide sequencing with oligonucleotide microarrays for global transcription profiling. BMC Genomics 11, 282. 185. Kukurba, K.R., and Montgomery, S.B. (2015). RNA Sequencing and Analysis. Cold Spring Harb Protoc 2015, 951-969. 186. Esteller, M. (2011). Non-coding RNAs in human disease. Nat Rev Genet 12, 861- 874. 187. Eddy, S.R. (2001). Non-coding RNA genes and the modern RNA world. Nat Rev Genet 2, 919-929. 188. Rogoyski, O.M., Pueyo, J.I., Couso, J.P., and Newbury, S.F. (2017). Functions of long non-coding RNAs in human disease and their conservation in Drosophila development. Biochem Soc Trans 45, 895-904. 189. Harries, L.W. (2012). Long non-coding RNAs and human disease. Biochem Soc Trans 40, 902-906. 190. Wentz-Hunter, K., Cheng, E.L., Ueda, J., Sugar, J., and Yue, B.Y. (2001). Keratocan expression is increased in the stroma of keratoconus corneas. Mol Med 7, 470-477. 191. Ha, N.T., Nakayasu, K., Murakami, A., Ishidoh, K., and Kanai, A. (2004). Microarray analysis identified differentially expressed genes in keratocytes from keratoconus patients. Curr Eye Res 28, 373-379. 192. Nielsen, K., Birkenkamp-Demtroder, K., Ehlers, N., and Orntoft, T.F. (2003). Identification of differentially expressed genes in keratoconus epithelium analyzed on microarrays. Invest Ophthalmol Vis Sci 44, 2466-2476. 193. Balasubramanian, S.A., Wasinger, V.C., Pye, D.C., and Willcox, M.D. (2013). Preliminary identification of differentially expressed tear proteins in keratoconus. Mol Vis 19, 2124-2134. 194. Chaerkady, R., Shao, H., Scott, S.G., Pandey, A., Jun, A.S., and Chakravarti, S. (2013). The keratoconus corneal proteome: loss of epithelial integrity and stromal degeneration. J Proteomics 87, 122-131. 195. Joseph, R., Srivastava, O.P., and Pfister, R.R. (2011). Differential epithelial and stromal protein profiles in keratoconus and normal human corneas. Exp Eye Res 92, 282-298. 196. Nielsen, K., Vorum, H., Fagerholm, P., Birkenkamp-Demtroder, K., Honore, B., Ehlers, N., and Orntoft, T.F. (2006). Proteome profiling of corneal epithelium and

108

identification of marker proteins for keratoconus, a pilot study. Exp Eye Res 82, 201-209. 197. Szczesniak, M.W., Kabza, M., Karolak, J.A., Rydzanicz, M., Nowak, D.M., Ginter- Matuszewska, B., Polakowski, P., Ploski, R., Szaflik, J.P., and Gajecka, M. (2017). KTCNlncDB-a first platform to investigate lncRNAs expressed in human keratoconus and non-keratoconus corneas. Database (Oxford) 2017, baw168. 198. Kabza, M., Karolak, J.A., Rydzanicz, M., Szczesniak, M.W., Nowak, D.M., Ginter- Matuszewska, B., Polakowski, P., Ploski, R., Szaflik, J.P., and Gajecka, M. (2017). Collagen synthesis disruption and downregulation of core elements of TGF-beta, Hippo, and Wnt pathways in keratoconus corneas. European journal of human genetics : EJHG 25, 582-590. 199. Bykhovskaya, Y., Gromova, A., Makarenkova, H.P., and Rabinowitz, Y.S. (2016). Abnormal regulation of extracellular matrix and adhesion molecules in corneas of patients with keratoconus. Int J Keratoconus Ectatic Corneal Dis 5, 63-70. 200. Rabinowitz, Y.S. (1995). Videokeratographic indices to aid in screening for keratoconus. J Refract Surg 11, 371-379. 201. Trapnell, C., Pachter, L., and Salzberg, S.L. (2009). TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105-1111. 202. Trapnell, C., Williams, B.A., Pertea, G., Mortazavi, A., Kwan, G., van Baren, M.J., Salzberg, S.L., Wold, B.J., and Pachter, L. (2010). Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28, 511-515. 203. Trapnell, C., Roberts, A., Goff, L., Pertea, G., Kim, D., Kelley, D.R., Pimentel, H., Salzberg, S.L., Rinn, J.L., and Pachter, L. (2012). Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc 7, 562-578. 204. Wang, J., Duncan, D., Shi, Z., and Zhang, B. (2013). WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013. Nucleic Acids Res 41, W77-83. 205. Fang, S., Zhang, L., Guo, J., Niu, Y., Wu, Y., Li, H., Zhao, L., Li, X., Teng, X., Sun, X., et al. (2018). NONCODEV5: a comprehensive annotation database for long non-coding RNAs. Nucleic Acids Res 46, D308-D314. 206. Cai, J., Perkumas, K.M., Qin, X., Hauser, M.A., Stamer, W.D., and Liu, Y. (2015). Expression Profiling of Human Schlemm's Canal Endothelial Cells From Eyes With and Without Glaucoma. Invest Ophthalmol Vis Sci 56, 6747-6753. 207. Liu, Y., Bailey, J.C., Helwa, I., Dismuke, W.M., Cai, J., Drewry, M., Brilliant, M.H., Budenz, D.L., Christen, W.G., Chasman, D.I., et al. (2016). A Common Variant in MIR182 Is Associated With Primary Open-Angle Glaucoma in the NEIGHBORHOOD Consortium. Invest Ophthalmol Vis Sci 57, 4528-4535. 208. Kulkarni, B., Mohammed, I., Hopkinson, A., and Dua, H.S. (2011). Validation of endogenous control genes for gene expression studies on human ocular surface epithelium. PLoS One 6, e22301. 209. Team, R. (2015). RStudio: Integrated Development Environment for R. In. (Boston, MA, RStudio, Inc. 210. Su, G., Morris, J.H., Demchak, B., and Bader, G.D. (2014). Biological network exploration with Cytoscape 3. Curr Protoc Bioinformatics 47, 8 13 11-24.

109

211. Mace, M., Galiacy, S.D., Erraud, A., Mejia, J.E., Etchevers, H., Allouche, M., Desjardins, L., Calvas, P., and Malecaze, F. (2011). Comparative transcriptome and network biology analyses demonstrate antiproliferative and hyperapoptotic phenotypes in human keratoconus corneas. Invest Ophthalmol Vis Sci 52, 6181- 6191. 212. You, J., Corley, S.M., Wen, L., Hodge, C., Hollhumer, R., Madigan, M.C., Wilkins, M.R., and Sutton, G. (2018). RNA-Seq analysis and comparison of corneal epithelium in keratoconus and myopia patients. Sci Rep 8, 389. 213. Tamura, Y., Konomi, H., Sawada, H., Takashima, S., and Nakajima, A. (1991). Tissue distribution of type VIII collagen in human adult and fetal eyes. Invest Ophthalmol Vis Sci 32, 2636-2644. 214. Hopfer, U., Fukai, N., Hopfer, H., Wolf, G., Joyce, N., Li, E., and Olsen, B.R. (2005). Targeted disruption of Col8a1 and Col8a2 genes in mice leads to anterior segment abnormalities in the eye. FASEB J 19, 1232-1244. 215. Toprak, I., Kucukatay, V., Yildirim, C., Kilic-Toprak, E., and Kilic-Erkek, O. (2014). Increased systemic oxidative stress in patients with keratoconus. Eye (Lond) 28, 285-289. 216. Volders, P.J., Verheggen, K., Menschaert, G., Vandepoele, K., Martens, L., Vandesompele, J., and Mestdagh, P. (2015). An update on LNCipedia: a database for annotated human lncRNA sequences. Nucleic Acids Res 43, D174-180. 217. Wang, K.C., and Chang, H.Y. (2011). Molecular mechanisms of long noncoding RNAs. Mol Cell 43, 904-914. 218. Loh, A., Hadziahmetovic, M., and Dunaief, J.L. (2009). Iron homeostasis and eye disease. Biochim Biophys Acta 1790, 637-649. 219. Huelsken, J., and Birchmeier, W. (2001). New aspects of Wnt signaling pathways in higher vertebrates. Curr Opin Genet Dev 11, 547-553. 220. Goralska, M., Ferrell, J., Harned, J., Lall, M., Nagar, S., Fleisher, L.N., and McGahan, M.C. (2009). Iron metabolism in the eye: a review. Exp Eye Res 88, 204-215. 221. Tong, X.J., Kawabata, H., and Koeffler, H.P. (2002). Iron deficiency can upregulate expression of transferrin receptor at both the mRNA and protein level. Brit J Haematol 116, 458-464. 222. Curtis, A.R., Fey, C., Morris, C.M., Bindoff, L.A., Ince, P.G., Chinnery, P.F., Coulthard, A., Jackson, M.J., Jackson, A.P., McHale, D.P., et al. (2001). Mutation in the gene encoding ferritin light polypeptide causes dominant adult-onset basal ganglia disease. Nature genetics 28, 350-354. 223. Benjamin D Ashby, Q.G.a.M.D.W. (2014). Corneal Injuries and Wound Healing – Review of Processes and Therapies. Austin J Clin Ophthalmol 1, 1017. 224. Guo, S., and Dipietro, L.A. (2010). Factors affecting wound healing. J Dent Res 89, 219-229. 225. Elsharkawi-Welt, K., Hepp, J., and Scharffetter-Kochanek, K. (2008). [Genetic causes of impaired wound healing. Rare differential diagnosis of the non-healing wound]. Hautarzt 59, 893-903. 226. Schey, K.L., Wang, Z., J, L.W., and Qi, Y. (2014). Aquaporins in the eye: expression, function, and roles in ocular disease. Biochim Biophys Acta 1840, 1513-1523.

110

227. Rabinowitz, Y.S., Dong, L., and Wistow, G. (2005). Gene expression profile studies of human keratoconus cornea for NEIBank: a novel cornea-expressed gene and the absence of transcripts for aquaporin 5. Invest Ophthalmol Vis Sci 46, 1239- 1246. 228. Garfias, Y., Navas, A., Perez-Cano, H.J., Quevedo, J., Villalvazo, L., and Zenteno, J.C. (2008). Comparative expression analysis of aquaporin-5 (AQP5) in keratoconic and healthy corneas. Mol Vis 14, 756-761. 229. Wang, W., Sumiyoshi, H., Yoshioka, H., and Fujiwara, S. (2006). Interactions between epiplakin and intermediate filaments. J Dermatol 33, 518-527. 230. Jang, S.I., Kalinin, A., Takahashi, K., Marekov, L.N., and Steinert, P.M. (2005). Characterization of human epiplakin: RNAi-mediated epiplakin depletion leads to the disruption of keratin and vimentin IF networks. J Cell Sci 118, 781-793. 231. Kokado, M., Okada, Y., Goto, M., Ishikawa, K., Miyamoto, T., Yamanaka, O., Fujiwara, S., and Saika, S. (2013). Increased Fragility, Impaired Differentiation, and Acceleration of Migration of Corneal Epithelium of Epiplakin-Null Mice. Invest Ophthalmol Vis Sci 54, 3780-3789. 232. McNamara, N.A., Brand, R.J., Polse, K.A., and Bourne, W.M. (1998). Corneal function during normal and high serum glucose levels in diabetes. Invest Ophthalmol Vis Sci 39, 3-17. 233. Woodward, M.A., Blachley, T.S., and Stein, J.D. (2016). The Association Between Sociodemographic Factors, Common Systemic Diseases, and Keratoconus: An Analysis of a Nationwide Heath Care Claims Database. Ophthalmology 123, 457- 465 e452. 234. Seiler, T., Huhle, S., Spoerl, E., and Kunath, H. (2000). Manifest diabetes and keratoconus: a retrospective case-control study. Graefes Arch Clin Exp Ophthalmol 238, 822-825. 235. Kuo, I.C., Broman, A., Pirouzmanesh, A., and Melia, M. (2006). Is there an association between diabetes and keratoconus? Ophthalmology 113, 184-190. 236. Goldich, Y., Barkana, Y., Gerber, Y., Rasko, A., Morad, Y., Harstein, M., Avni, I., and Zadok, D. (2009). Effect of diabetes mellitus on biomechanical parameters of the cornea. J Cataract Refract Surg 35, 715-719. 237. Sriram, S., Robinson, P., Pi, L., Lewin, A.S., and Schultz, G. (2013). Triple combination of siRNAs targeting TGFbeta1, TGFbetaR2, and CTGF enhances reduction of collagen I and smooth muscle actin in corneal fibroblasts. Invest Ophthalmol Vis Sci 54, 8214-8223. 238. Gibson, D.J., Pi, L., Sriram, S., Mao, C., Petersen, B.E., Scott, E.W., Leask, A., and Schultz, G.S. (2014). Conditional knockout of CTGF affects corneal wound healing. Invest Ophthalmol Vis Sci 55, 2062-2070. 239. Gao, X., Nannini, D.R., Corrao, K., Torres, M., Chen, Y.I., Fan, B.J., Wiggs, J.L., International Glaucoma Genetics, C., Taylor, K.D., Gauderman, W.J., et al. (2016). Genome-wide association study identifies WNT7B as a novel locus for central corneal thickness in Latinos. Hum Mol Genet 25, 5035-5045. 240. Toyono, T., Usui, T., Yokoo, S., Taketani, Y., Nakagawa, S., Kuroda, M., Yamagami, S., and Amano, S. (2015). Angiopoietin-like 7 is an anti-angiogenic protein required to prevent vascularization of the cornea. PLoS One 10, e0116838.

111

241. Iqbal, O., Fisher, G., Vira, S., Syed, D., Sadeghi, N., Freeman, D., Campbell, E., Sugar, J., Feder, R., Fareed, J., et al. (2013). Increased expression of secreted frizzled-related protein-1 and microtubule-associated protein light chain 3 in keratoconus. Cornea 32, 702-707. 242. You, J., Hodge, C., Wen, L., McAvoy, J.W., Madigan, M.C., and Sutton, G. (2013). Tear levels of SFRP1 are significantly reduced in keratoconus patients. Mol Vis 19, 509-xxx. 243. Jenkins, L.M., Singh, P., Varadaraj, A., Lee, N.Y., Shah, S., Flores, H.V., O'Connell, K., and Mythreye, K. (2016). Altering the Proteoglycan State of Transforming Growth Factor beta Type III Receptor (TbetaRIII)/Betaglycan Modulates Canonical Wnt/beta-Catenin Signaling. J Biol Chem 291, 25716- 25728. 244. Yao, B., Wang, S., Xiao, P., Wang, Q., Hea, Y., and Zhang, Y. (2017). MAPK signaling pathways in eye wounds: Multifunction and cooperation. Exp Cell Res 359, 10-16. 245. Zhu, Z., He, X., Johnson, C., Stoops, J., Eaker, A.E., Stoffer, D.S., Bell, A., Zarnegar, R., and DeFrances, M.C. (2007). PI3K is negatively regulated by PIK3IP1, a novel p110 interacting protein. Biochem Biophys Res Commun 358, 66-72. 246. Clay Bunn, R., and Fowlkes, J.L. (2003). Insulin-like growth factor binding protein proteolysis. Trends in Endocrin & Metabolism 14, 176-181. 247. Lodygin, D., Epanchintsev, A., Menssen, A., Diebold, J., and Hermeking, H. (2005). Functional epigenomics identifies genes frequently silenced in prostate cancer. Cancer Res 65, 4218-4227. 248. Xuan, M., Wang, S., Liu, X., He, Y., Li, Y., and Zhang, Y. (2016). Proteins of the corneal stroma: importance in visual function. Cell Tissue Res 364, 9-16. 249. Wheeler, J., Hauser, M.A., Afshari, N.A., Allingham, R.R., and Liu, Y. (2012). The Genetics of Keratoconus: A Review. Reproductive system & sexual disorders : current research. 250. Bykhovskaya, Y., Margines, B., and Rabinowitz, Y.S. (2016). Genetics in Keratoconus: where are we? Eye Vis (Lond) 3, 16. 251. Bykhovskaya, Y., Li, X., Taylor, K.D., Haritunians, T., Rotter, J.I., and Rabinowitz, Y.S. (2016). Linkage Analysis of High-density SNPs Confirms Keratoconus Locus at 5q Chromosomal Region. Ophthalmic genetics 37, 109-110. 252. Bykhovskaya, Y., Li, X., Taylor, K.D., Haritunians, T., Rotter, J.I., and Rabinowitz, Y.S. (2016). Linkage Analysis of High-density SNPs Confirms Keratoconus Locus at 5q Chromosomal Region. Ophthalmic genetics 37, 109-110. 253. Bykhovskaya, Y., Fardaei, M., Khaled, M.L., Nejabat, M., Salouti, R., Dastsooz, H., Liu, Y., Inaloo, S., and Rabinowitz, Y.S. (2017). TSC1 Mutations in Keratoconus Patients With or Without Tuberous Sclerosis. Invest Ophthalmol Vis Sci 58, 6462-6469. 254. Li, H., and Durbin, R. (2009). Fast and accurate short read alignment with Burrows- Wheeler transform. Bioinformatics 25, 1754-1760. 255. McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky, A., Garimella, K., Altshuler, D., Gabriel, S., Daly, M., et al. (2010). The Genome

112

Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res 20, 1297-1303. 256. DePristo, M.A., Banks, E., Poplin, R., Garimella, K.V., Maguire, J.R., Hartl, C., Philippakis, A.A., del Angel, G., Rivas, M.A., Hanna, M., et al. (2011). A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nature genetics 43, 491-498. 257. Van der Auwera, G.A., Carneiro, M.O., Hartl, C., Poplin, R., Del Angel, G., Levy- Moonshine, A., Jordan, T., Shakir, K., Roazen, D., Thibault, J., et al. (2013). From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr Protoc Bioinformatics 43, 11 10 11-33. 258. Cingolani, P., Platts, A., Wang le, L., Coon, M., Nguyen, T., Wang, L., Land, S.J., Lu, X., and Ruden, D.M. (2012). A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin) 6, 80-92. 259. Danecek, P., Auton, A., Abecasis, G., Albers, C.A., Banks, E., DePristo, M.A., Handsaker, R.E., Lunter, G., Marth, G.T., Sherry, S.T., et al. (2011). The variant call format and VCFtools. Bioinformatics 27, 2156-2158. 260. Gu, C., Nguyen, H.N., Ganini, D., Chen, Z., Jessen, H.J., Gu, Z., Wang, H., and Shears, S.B. (2017). KO of 5-InsP7 kinase activity transforms the HCT116 colon cancer cell line into a hypermetabolic, growth-inhibited phenotype. Proc Natl Acad Sci U S A 114, 11968-11973. 261. Gu, C., Nguyen, H.N., Hofer, A., Jessen, H.J., Dai, X., Wang, H., and Shears, S.B. (2017). The Significance of the Bifunctional Kinase/Phosphatase Activities of Diphosphoinositol Pentakisphosphate Kinases (PPIP5Ks) for Coupling Inositol Pyrophosphate Cell Signaling to Cellular Phosphate Homeostasis. J Biol Chem 292, 4544-4555. 262. Kobayashi, T., Yoshioka, R., Shiraishi, A., and Ohashi, Y. (2009). New technique for culturing corneal epithelial cells of normal mice. Mol Vis 15, 1589-1593. 263. Gipson, I.K., and Grill, S.M. (1982). A technique for obtaining sheets of intact rabbit corneal epithelium. Invest Ophthalmol Vis Sci 23, 269-273. 264. Yousaf, R., Gu, C., Ahmed, Z.M., Khan, S.N., Friedman, T.B., Riazuddin, S., Shears, S.B., and Riazuddin, S. (2018). Mutations in Diphosphoinositol- Pentakisphosphate Kinase PPIP5K2 are associated with hearing loss in human and mouse. PLoS Genet 14, e1007297. 265. Green, M.R., and Sambrook, J. (2018). Touchdown Polymerase Chain Reaction (PCR). Cold Spring Harb Protoc 2018, pdb prot095133. 266. Han, S.B., Liu, Y.C., Noriega, K.M., and Mehta, J.S. (2016). Applications of Anterior Segment Optical Coherence Tomography in Cornea and Ocular Surface Diseases. J Ophthalmol 2016, 4971572. 267. Hanlon, S.D., Patel, N.B., and Burns, A.R. (2011). Assessment of postnatal corneal development in the C57BL/6 mouse using spectral domain optical coherence tomography and microwave-assisted histology. Exp Eye Res 93, 363-370. 268. Wang, J., Saul, A., Cui, X., Roon, P., and Smith, S.B. (2017). Absence of Sigma 1 Receptor Accelerates Photoreceptor Cell Death in a Murine Model of Retinitis Pigmentosa. Invest Ophthalmol Vis Sci 58, 4545-4558.

113

269. Zhao, M., Kuo, A.N., and Izatt, J.A. (2010). 3D refraction correction and extraction of clinical parameters from spectral domain optical coherence tomography of the cornea. Opt Express 18, 8923-8936. 270. Alice Sicong Liu, D.B., Rachel Elizabeth Conn, Ryan P. McNabb, Machelle T. Pardue, Anthony N. Kuo. (2019). Mouse corneal topography, pachymetry, and keratometry utilizing OCT with keratometer validation. In ARVO Annual Meeting Abstract (VANCOUVER,BC. 271. McNabb, R.P., Kuo, A.N., and Izatt, J.A. (2013). Quantitative single and multi- surface clinical corneal topography utilizing optical coherence tomography. Opt Lett 38, 1212-1214. 272. American National Standards Institute, Corneal Topography System: Standard Terminology, Requirements.Washington, DC. ANSI Z80.23–2007

In. ( . 273. Shears, S.B. (2009). Diphosphoinositol polyphosphates: metabolic messengers? Mol Pharmacol 76, 236-252. 274. Barker, C.J., Illies, C., Gaboardi, G.C., and Berggren, P.O. (2009). Inositol pyrophosphates: structure, enzymology and function. Cell Mol Life Sci 66, 3851- 3871. 275. Williams, S.P., Gillaspy, G.E., and Perera, I.Y. (2015). Biosynthesis and possible functions of inositol pyrophosphates in plants. Front Plant Sci 6, 67. 276. Morse, H.C. (2007). Chapter 1 - Building a Better Mouse: One Hundred Years of Genetics and Biology. In The Mouse in Biomedical Research (Second Edition), J.G. Fox, M.T. Davisson, F.W. Quimby, S.W. Barthold, C.E. Newcomer, andA.L. Smith, eds. (Burlington, Academic Press), pp 1-11. 277. Karin, M., Liu, Z., and Zandi, E. (1997). AP-1 function and regulation. Curr Opin Cell Biol 9, 240-246. 278. Srivastava, P.K., Hull, R.P., Behmoaras, J., Petretto, E., and Aitman, T.J. (2013). JunD/AP1 regulatory network analysis during macrophage activation in a rat model of crescentic glomerulonephritis. BMC Syst Biol 7, 93. 279. Hernandez, J.M., Floyd, D.H., Weilbaecher, K.N., Green, P.L., and Boris-Lawrie, K. (2008). Multiple facets of junD gene expression are atypical among AP-1 family members. Oncogene 27, 4757-4767. 280. Shi, X., Sun, M., Liu, H., Yao, Y., and Song, Y. (2013). Long non-coding RNAs: a new frontier in the study of human diseases. Cancer Lett 339, 159-166. 281. Wilusz, J.E., Sunwoo, H., and Spector, D.L. (2009). Long noncoding RNAs: functional surprises from the RNA world. Genes Dev 23, 1494-1504. 282. Gupta, R.A., Shah, N., Wang, K.C., Kim, J., Horlings, H.M., Wong, D.J., Tsai, M.C., Hung, T., Argani, P., Rinn, J.L., et al. (2010). Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature 464, 1071-1076. 283. Tan, M.C., Widagdo, J., Chau, Y.Q., Zhu, T., Wong, J.J., Cheung, A., and Anggono, V. (2017). The Activity-Induced Long Non-Coding RNA Meg3 Modulates AMPA Receptor Surface Expression in Primary Cortical Neurons. Front Cell Neurosci 11, 124.

114

284. Qiu, G.Z., Tian, W., Fu, H.T., Li, C.P., and Liu, B. (2016). Long noncoding RNA- MEG3 is involved in diabetes mellitus-related microvascular dysfunction. Biochem Biophys Res Commun 471, 135-141. 285. Cejkova, J., Stipek, S., Crkovska, J., and Ardan, T. (2000). Changes of superoxide dismutase, catalase and glutathione peroxidase in the corneal epithelium after UVB rays. Histochemical and biochemical study. Histol Histopathol 15, 1043- 1050. 286. Cejkova, J., Stipek, S., Crkovska, J., Ardan, T., Platenik, J., Cejka, C., and Midelfart, A. (2004). UV Rays, the prooxidant/antioxidant imbalance in the cornea and oxidative eye damage. Physiol Res 53, 1-10. 287. Michalopoulos, I., Pavlopoulos, G.A., Malatras, A., Karelas, A., Kostadima, M.A., Schneider, R., and Kossida, S. (2012). Human gene correlation analysis (HGCA): a tool for the identification of transcriptionally co-expressed genes. BMC Res Notes 5, 265. 288. Gov, E., and Arga, K.Y. (2017). Differential co-expression analysis reveals a novel prognostic gene module in ovarian cancer. Sci Rep 7, 4996. 289. Wu, M., Chong, L.S., Perlman, D.H., Resnick, A.C., and Fiedler, D. (2016). Inositol polyphosphates intersect with signaling and metabolic networks via two distinct mechanisms. Proc Natl Acad Sci U S A 113, E6757-E6765. 290. Gokhale, N.A., Zaremba, A., and Shears, S.B. (2011). Receptor-dependent compartmentalization of PPIP5K1, a kinase with a cryptic polyphosphoinositide binding domain. Biochem J 434, 415-426. 291. Wilson, S.E., Liu, J.J., and Mohan, R.R. (1999). Stromal-epithelial interactions in the cornea. Prog Retin Eye Res 18, 293-309. 292. Yoon, J.J., Ismail, S., and Sherwin, T. (2014). Limbal stem cells: Central concepts of corneal epithelial homeostasis. World J Stem Cells 6, 391-403. 293. Shetty, R., Vunnava, K.P., Dhamodaran, K., Matalia, H., Murali, S., Jayadev, C., Murugeswari, P., Ghosh, A., and Das, D. (2019). Characterization of Corneal Epithelial Cells in Keratoconus. Transl Vis Sci Technol 8, 2. 294. Song, P., Wang, S., Zhang, P., Sui, W., Zhang, Y., Liu, T., and Gao, H. (2016). The Superficial Stromal Scar Formation Mechanism in Keratoconus: A Study Using Laser Scanning In Vivo Confocal Microscopy. BioMed Research International 2016, 10. 295. Gu, C., Nguyen, H.-N., Ganini, D., Chen, Z., Jessen, H.J., Gu, Z., Wang, H., and Shears, S.B. (2017). KO of 5-InsP7 kinase activity transforms the HCT116 colon cancer cell line into a hypermetabolic, growth-inhibited phenotype. Proceedings of the National Academy of Sciences 114, 11968-11973. 296. Shah, A., Ganguli, S., Sen, J., and Bhandari, R. (2017). Inositol Pyrophosphates: Energetic, Omnipresent and Versatile Signalling Molecules. J Indian Instit Sci 97, 23-40. 297. Ehrnhoefer, D.E., Butland, S.L., Pouladi, M.A., and Hayden, M.R. (2009). Mouse models of Huntington disease: variations on a theme. Dis Model Mech 2, 123- 129. 298. Xu, Q., Shenoy, S., and Li, C. (2012). Mouse models for LRRK2 Parkinson's disease. Parkinsonism Relat Disord 18 Suppl 1, S186-189.

115

299. Lombardi, R., Bell, A., Senthil, V., Sidhu, J., Noseda, M., Roberts, R., and Marian, A.J. (2008). Differential interactions of thin filament proteins in two cardiac troponin T mouse models of hypertrophic and dilated cardiomyopathies. Cardiovasc Res 79, 109-117. 300. Garcia-Gras, E., Lombardi, R., Giocondo, M.J., Willerson, J.T., Schneider, M.D., Khoury, D.S., and Marian, A.J. (2006). Suppression of canonical Wnt/beta- catenin signaling by nuclear plakoglobin recapitulates phenotype of arrhythmogenic right ventricular cardiomyopathy. J Clin Invest 116, 2012-2021. 301. Marian, A.J. (2011). Modeling human disease phenotype in model organisms: "It's only a model!". Circ Res 109, 356-359.

116

APPENDIX

117

118