Journal of Refractive Surgery

Translational Science Supplemental Data

Histological and microRNA Signatures of Corneal Epithelium in Keratoconus

Yu Meng Wang, MMed; Tsz Kin Ng, PhD; Kwong Wai Choy, PhD; Hoi Kin Wong, MPhil; Wai Kit Chu, DPhil; Chi Pui Pang, DPhil; Vishal Jhanji, MD, FRCOphth

Abstract

PURPOSE:

To illustrate the histopathology of keratoconic corneal epithelia and its micro-ribonucleic acid (miRNA) regulation.

METHODS:

Corneal epithelia were collected from 27 patients with keratoconus and 26 normal patients after surgery or by impression cytology. The miRNA profile was determined using miRNA microarray. The biological roles of miRNA target genes were delineated by gene ontology and pathway analyses. The expressions of significant miRNAs were validated using TaqMan polymerase chain reaction (PCR), whereas protein localization and expression of the miRNA target genes were examined by immunofluorescence and immunoblotting analyses.

RESULTS:

Histological assessment showed that corneal epithelia in patients with keratoconus were thinner with loosely packed cells compared to normal patients. Microarray analysis revealed that 12 miRNAs were significantly downregulated in keratoconic corneal epithelia. Gene ontology analysis demonstrated that the predicted miRNA target genes participated in cell junction, cell division, and motor activity, whereas pathway analysis highlighted the involvement of syndecan-mediated signaling pathway. TaqMan PCR validated the altered expression of six miRNAs in corneal epithelia from surgery (hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-28-5p, and hsa-miR-181a-2-3p) and four miRNAs in squamous corneal epithelial samples collected from impression cytology (hsa-miR-151a-3p, hsa-miR-195-5p, hsa-miR-185-5p, and hsa-miR-194-5p). In addition, higher S100A2 expression was found in the epithelial basal cell layer of keratoconic corneal epithelia.

CONCLUSIONS:

The miRNA and histological analyses in this study demonstrated structural and biological changes in keratoconic corneal epithelia, broadening the understanding of keratoconus pathology. In addition, impression cytology is useful to collect corneal epithelial tissues for gene expression analysis.

[J Refract Surg. 2018;34(3):201–211.]

Abstract

PURPOSE:

To illustrate the histopathology of keratoconic corneal epithelia and its micro-ribonucleic acid (miRNA) regulation.

METHODS:

Corneal epithelia were collected from 27 patients with keratoconus and 26 normal patients after surgery or by impression cytology. The miRNA profile was determined using miRNA microarray. The biological roles of miRNA target genes were delineated by gene ontology and pathway analyses. The expressions of significant miRNAs were validated using TaqMan polymerase chain reaction (PCR), whereas protein localization and expression of the miRNA target genes were examined by immunofluorescence and immunoblotting analyses.

RESULTS:

Histological assessment showed that corneal epithelia in patients with keratoconus were thinner with loosely packed cells compared to normal patients. Microarray analysis revealed that 12 miRNAs were significantly downregulated in keratoconic corneal epithelia. Gene ontology analysis demonstrated that the predicted miRNA target genes participated in cell junction, cell division, and motor activity, whereas pathway analysis highlighted the involvement of syndecan-mediated signaling pathway. TaqMan PCR validated the altered expression of six miRNAs in corneal epithelia from surgery (hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-28-5p, and hsa-miR-181a-2-3p) and four miRNAs in squamous corneal epithelial samples collected from impression cytology (hsa-miR-151a-3p, hsa-miR-195-5p, hsa-miR-185-5p, and hsa-miR-194-5p). In addition, higher S100A2 expression was found in the epithelial basal cell layer of keratoconic corneal epithelia.

CONCLUSIONS:

The miRNA and histological analyses in this study demonstrated structural and biological changes in keratoconic corneal epithelia, broadening the understanding of keratoconus pathology. In addition, impression cytology is useful to collect corneal epithelial tissues for gene expression analysis.

[J Refract Surg. 2018;34(3):201–211.]

Keratoconus is a corneal ectatic disorder characterized by irregular astigmatism and mild to severe visual impairment. As the disease advances, approximately 20% of the patients require keratoplasty for visual rehabilitation.1,2 The etiology of keratoconus is complex and multifactorial, involving both genetic and environmental factors. Our previous genome-wide association study identified 26 loci associated with central corneal thickness, among which FOXO1 and FNDC3B confer high risk for keratoconus.2 Another exome sequencing analysis also reported the association of a WNT10A variant with corneal thickness and increased risk for keratoconus.3 It is known that keratoconus pathology has a distinct anterior focus involving the epithelium and anterior corneal stroma with loss of cellular uniformity.4 Transcriptome analysis of keratoconus corneal epithelium identified 56 differentially expressed genes in corneal epithelia, which participate in extracellular matrix remodeling, cell–cell/cell–matrix interactions, and transmembrane signaling.5 Proteomic analysis showed that cytokeratin-3, cytokeratin-12, gelsolin, S100A4, and enolase-1 are differentially expressed in keratoconic corneal epithelia.6 These studies indicate the possible pathological changes in keratoconic cor-neal epithelia. Normal corneal epithelium accounts for approximately 10% of the corneal thickness. The keratoconus epithelial corneal thickness profile using Artemis very high-frequency digital ultrasound scanning (ArcScan, Inc., Golden, CO) demonstrated localized central thinning surrounded by an annulus of thickened epithelium.7 Fourier-domain optical coherence tomography demonstrated parameters based on corneal epithelial thickness that could differentiate keratoconic from normal eyes with high accuracy.8 In vivo confocal microscopy and histological analyses have confirmed that basal epithelial cells in keratoconic corneas exhibited irregular arrangement and a significant reduction in cell density when compared to normal corneas.4 The epithelial apoptosis could be a possible reason for the corneal epithelium thinning.9

Micro-ribonucleic acids (miRNAs) belong to a family of 18 to 24 nucleotide non-coding RNAs that negatively regulate the translation of messenger RNA. The miR-184 mutations and miR-568 polymorphism have been found to be associated with keratoconus.10 A recent deep sequencing analysis on different normal human ocular tissues identified 297 miRNAs expressed in the cornea, and 9 of them could target the genes associated with keratoconus (DOCK9, FNDC3B, FOXO1, and RAB3GAP1).11 Yet, miRNA regulation in kerato-conic corneal epithelium has not been studied. In this study, we aimed to illustrate the histophathology and miRNA regulation of corneal epithelial changes in keratoconic patients. Corneal epithelia were collected from patients with keratoconus and normal patients during surgery and their histological structures were resolved. The miRNA profiles were delineated by microarray and miRNA target genes were analyzed. In addition, the translational potential of the miRNA expression was confirmed in the corneal epithelial samples obtained by impression cytology.

Patients and Methods

Study Patients

This prospective study involved a total of 53 consecutive participants (27 keratoconic and 26 normal) recruited from The Chinese University of Hong Kong Eye Centre. The diagnosis of keratoconus was made based on retinoscopic signs (“red” reflex on direct ophthalmoscopy), biomicroscopic signs (Vogt's striae, Fleischer's ring, and corneal scarring) and corneal tomography using swept-source optical coherence tomography (Casia; Tomey Corporation, Nagoya, Japan). For the normal cohort, the study patients with no ocular abnormality except myopia and myopic astigmatism, a corrected distance visual acuity of 20/20 or better, and stable refraction for more than 1 year were included. All of them had photorefractive keratectomy. Patients with a corneal thickness of less than 500 μm, suspicion of keratoconus on corneal topography (displacement of the corneal apex, decrease in thinnest-point pachymetry, and asymmetric topographic pattern), ocular inflammation, and infection were excluded. The Ethics Committee for Human Research of The Chinese University of Hong Kong approved the study protocol. The protocol followed the tenets of the Declaration of Helsinki. An informed consent was obtained from each study patient after explanation of the nature and possible consequences of the study.

Collection of Corneal Epithelia

Central corneal epithelia (8 to 9 mm) of the study patients were collected during photorefractive keratectomy (normal patients) or corneal cross-linking (patients with keratoconus). The corneal epithelium samples were immediately washed with phosphate buffered saline (PBS) (0.01 M, pH 7.5; Amresco, Solon, OH) and stored in TRIzol (for RNA extraction; Invitrogen, Carlsbad, CA) or radioimmunoprecipitation assay (RIPA) buffer (for protein extraction; Sigma-Aldrich, St. Louis, MO) at −80°C or fixed with 10% formalin (for histological analysis; Sigma-Aldrich) at 4°C before further experiments. In addition to the surgical specimens, the squamous layer of the corneal epithelia was also collected using an impression cytology device (EYEPRIM; OPIA Technologies SAS, Paris, France). The EYEPRIM device comes with a 69-mm2, 0.2 μm-thick polysulfone membrane on one end and a plunger on the other end. Briefly, the patient was asked to look straight ahead with chin lifted slightly. A drop of preservative-free local anesthetic eye drops was instilled in the lower fornix of the eye. The plunger was pushed to touch the cornea with the membrane gently for 5 seconds. The pressure was released before removing the device. The membrane was carefully transferred from the device into a 1.5-mL tube using a pair of sterile forceps. Two EYEPRIM membranes were collected for each eye. The membranes were immediately stored in TRIzol reagent at −80°C for further miRNA analysis.

Histological Analysis

The corneal epithelia were fixed in 10% formalin, processed, and embedded in paraffin. The histological structure of the corneal epithelia in the paraffin section (5 μm) was visualized after hematoxylin–eosin staining. Briefly, the paraffin sections were dewaxed in xylene and rehydrated in 100% to 70% ethanol gradients. The sections were then stained with hematoxylin–eosin solution (Sigma-Aldrich) according to the recommended protocol. The stained sections were dehydrated and mounted. The morphological structure of corneal epithelia was visualized and imaged under a light microscope (Eclipse Ti-E; Nikon, Tokyo, Japan). The number of squamous, wing, and basal cells in normal and keratoconic corneal epithelium sections (10 sections per sample; 3 samples in each group) were counted and compared.

The total RNA, including the miRNA fraction, in TRIzol reagent was extracted according to the manufacturer's protocol (Invitrogen). The RNA concentration and quality were measured with Nanodrop 2000 (Thermo Fisher Scientific, Waltham, MA), whereas the RNA integrity was determined using the Agilent 2100 Bio-analyzer (Agilent Technologies, Inc., Santa Clara, CA). Human miRNA microarray slide (8 × 60K, release 19.0, G4872A; Agilent Technologies, Inc.) containing 2,006 human miRNAs from the Sanger miRBase (release 19.0) was used. GeneSpring GX 13.0 software (Agilent Technologies, Inc.) was used for value extraction and data analysis. A moderated t test was used for calculation of the P value for each miRNA probe. Significance was defined by a change greater than 1.5 fold and corrected P value less than .05. The Benjamin–Hochberg false discovery rate was used for adjustment of multiple tests. Principal component analysis (PCA) and hierarchical clustering were performed to provide a visual impression of how various sample groups were related. Two samples of normal corneal epithelium and two samples of keratoconic corneal epithelium were used for miRNA microarray experiment.

Downstream target genes of differentially expressed miRNAs were predicted by five databases (TargetScan, microRNA.org, TarBase, PicTar, and PITA) in the GeneSpring GX 13.0 software. Gene enrichment and gene ontology analysis were performed using the Database for Annotation, Visualization and Integrated Discovery (DAVID) Bioinformatics Resources 6.8 ( http://david.abcc.ncifcrf.gov/). An enrichment score greater than 1.3 was considered significant. Pathway analysis and gene interaction were performed using FunRich software (version 2.1.2; http://funrich.org).

Eleven differentially expressed miRNAs from microarray with greater than 10 fold change were subjected to further TaqMan validation. Six independent surgical corneal epithelial samples and eight independent impression cytology samples from each group were used in the validation experiments. Total RNA (20 ng) was reverse transcribed using the TaqMan MicroRNA Reverse Transcriptase kit (Applied Biosystems, Foster City, CA). The resultant products were quantified using the corresponding TaqMan MicroRNA Assays (Applied Biosystems) on LightCycler 480 Instrument II (Roche Diagnostics Corporation, Indianapolis, IN). Their expressions were normalized to that of U6.

Immunofluorescence Analysis

Immunofluorescence analysis was used to determine the protein location and expression of the miRNA target genes in the corneal epithelia. Briefly, the paraffin sections of corneal epithelium, after they were dewaxed and rehydrated, were incubated in freshly prepared ice-cold 50 mM ammonium chloride (Sigma-Aldrich) in PBS for 2 minutes to quench the free aldehyde and retrieve the antigen. The sections were then blocked in PBS supplemented with 1% bovine serum albumin (BSA), 0.15% saponin, and 0.01% Triton X-100 (all Sigma-Aldrich) for 2 hours, and incubated with the primary antibodies (Table A, available in the online version of this article) in PBS supplemented with 1% BSA, 0.425% saponin, 0.0015% Triton X-100, and Tween-20 (Sigma-Aldrich) at 4°C for 18 hours. After PBS washes, the sections were incubated with corresponding secondary antibodies conjugated with Alexa-488 and 4′,6′-diamidino-2-phenyindole (DAPI; for nuclear counter-stain) in PBS supplemented with 1% BSA for 1 hour at room temperature. The fluorescence signals were visualized and imaged under a fluorescence microscope (Eclipse Ti-E).

Antibodies for Protein Expression Analysis

Table A:

Antibodies for Protein Expression Analysis

Immunoblotting Analysis

The protein expression of miRNA-predicted target genes was semi-quantified by immunoblotting analysis. Briefly, the corneal epithelia were lysed in RIPA buffer supplemented with protease and phosphatase inhibitors (Roche). Total protein concentration of cell lysates was quantified using the Protein assay (Bio-Rad Laboratories, Hercules, CA). An equal amount of denatured protein samples (20 μg) was resolved with 10% or 15% SDS-PAGE gels, electro-transferred onto a nitrocellulose membrane and incubated with the primary antibodies (Table A) at 4°C for 18 hours, followed by incubation of the corresponding horseradish peroxidase (HRP)–conjugated secondary antibodies at room temperature for 1 hour. The signals were then detected using enhanced chemiluminescence (ECL) substrate (Amersham Pharmacia, Cleveland, OH) and captured by the ChemiDocTM XRS+ system (BioRad). Beta-actin was used for normalization. Relative expressions of target proteins were measured and calculated by Quantity One software (BioRad).

Statistical Analysis

Means of the keratoconic and normal groups were compared using the independent t test. All of the statistical analyses, except the microarray analysis, were performed using the commercially available software (SPSS version 22.0; SPSS, Inc., Chicago, IL). A P value of less than .05 was considered statistically significant.

Results

Clinical Assessments of Study Patients

The mean age of patients with keratoconus (n = 27; 29.75 ± 8.02 years) was comparable to that of the normal patients (n = 26; 33.39 ± 8.22 years; P = .070). Significant differences in corneal topographic parameters between the two studied groups were observed in anterior and posterior keratometry at steep axis (Ks), keratometry at flat axis (Kf), and average keratometry (Kavg) (P < .001) (Table 1). The swept-source optical coherence tomography images showed thin and conical cornea in patients with keratoconus compared to the normal patients (Figure A, available in the online version of this article).

Clinical Parameters of Keratoconic and Normal Patients

Table 1:

Clinical Parameters of Keratoconic and Normal Patients

Swept-source optical coherence tomography analysis of corneas of keratoconic and normal patients. (A) Keratometric axial power topography of normal cornea. (B) Cross-section view of normal cornea. (C) Keratometric axial power topography of keratoconic cornea. (D) Cross-section view of keratoconic cornea.

Figure A.

Swept-source optical coherence tomography analysis of corneas of keratoconic and normal patients. (A) Keratometric axial power topography of normal cornea. (B) Cross-section view of normal cornea. (C) Keratometric axial power topography of keratoconic cornea. (D) Cross-section view of keratoconic cornea.

Histological Analysis of Keratoconus Corneal Epithelium

The histological structure of surgical corneal epithelial specimens was assessed by hematoxylin–eosin staining. The corneal epithelia from normal patients were compact and tightly packed, whereas those from patients with keratoconus were thinner and loosely arranged (Figure 1A). The number of squamous cells, wing cells, and basal cells in the keratoconic corneal epithelia were 3.01 ± 0.53, 9.08 ± 1.41, and 7.42 ± 0.94 cells/100 μm, respectively, which were all significantly fewer than those in normal patients (6.42 ± 1.04, 20.48 ± 3.35, and 11.06 ± 1.00 cells/100 μm, respectively, P < .001; Table 2, Figure 1B). Immunofluorescence analysis of p53 protein on corneal epithelial sections showed an increased p53 protein expression in squamous and wing cell layers of keratoconic corneal epithelia (Figure 1C).

Histological analysis of keratoconic and normal corneal epithelia. (A) a. Corneal epithelium from normal patients was compact and tightly packed. b. Corneal epithelium was thinner and cells were loosely packed in keratoconic cornea. Scale bar: 20 μm. (B) Cell numbers in squamous cell layer, wing cell layer, and basal cell layer were significantly decreased. ***P < .001. (C) Elevated p53 protein expression in keratoconic corneal epithelium by immunofluorescence analysis, compared with normal corneal epithelium.

Figure 1.

Histological analysis of keratoconic and normal corneal epithelia. (A) a. Corneal epithelium from normal patients was compact and tightly packed. b. Corneal epithelium was thinner and cells were loosely packed in keratoconic cornea. Scale bar: 20 μm. (B) Cell numbers in squamous cell layer, wing cell layer, and basal cell layer were significantly decreased. ***P < .001. (C) Elevated p53 protein expression in keratoconic corneal epithelium by immunofluorescence analysis, compared with normal corneal epithelium.

Number of Squamous, Wing, and Basal Cells in Keratoconic and Normal Corneal Epithelia (cells/100 μm)

Table 2:

Number of Squamous, Wing, and Basal Cells in Keratoconic and Normal Corneal Epithelia (cells/100 μm)

miRNA Profiling and Validation

The global miRNA profile (miRNome) of normal and keratoconic corneal epithelia collected from surgery was determined by microarray. PCA (Figure 2A) and hierarchical clustering analysis (Figure 2B) demonstrated that the miRNA expression profiles of normal and keratoconic corneal epithelia were clustered separately. Of the 2,006 human miRNAs, 137 miRNAs showed more than 1.5-fold expression changes in the keratoconic corneal epithelia. The volcano plot identified 12 significant (Pcorr < .05 and > 1.5 fold change) differentially expressed miRNAs (hsa-miR-4713-3p, hsa-miR-151a-3p, hsa-miR-193b-5p, hsa-miR-138-5p, hsa-miR-195-5p, hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-28-5p, hsa-miR-6126, hsamiR-181a-2-3p, and hsa-miR-30e-3p; Figure 2C), all of which were downregulated in the keratoconic corneal epithelia (Table B, available in the online version of this article). In addition, we also confirmed that human corneal miRNAs were abundantly expressed in our samples (Table C, available in the online version of this article).

The micro-ribonucleic acid (miRNA) microarray profiling of keratoconic corneal epithelia. (A) Principle component analysis of miRNA profiles. Square: normal; Triangle: keratoconus. (B) Hierarchical clustering of normal and keratoconic corneal epithelial miRNA profiles. (C) Volcano plot identified 12 differentially expression miRNAs (blue; > 1.5 fold change and corrected P < .05).

Figure 2.

The micro-ribonucleic acid (miRNA) microarray profiling of keratoconic corneal epithelia. (A) Principle component analysis of miRNA profiles. Square: normal; Triangle: keratoconus. (B) Hierarchical clustering of normal and keratoconic corneal epithelial miRNA profiles. (C) Volcano plot identified 12 differentially expression miRNAs (blue; > 1.5 fold change and corrected P < .05).

Differentially Expressed miRNAs in Keratoconic and Normal Samples by Microarray and Taqman

Table B:

Differentially Expressed miRNAs in Keratoconic and Normal Samples by Microarray and Taqman

Expression of Human Cornea microRNAs in Keratoconus Corneal EpitheliumExpression of Human Cornea microRNAs in Keratoconus Corneal EpitheliumExpression of Human Cornea microRNAs in Keratoconus Corneal EpitheliumExpression of Human Cornea microRNAs in Keratoconus Corneal Epithelium

Table C:

Expression of Human Cornea microRNAs in Keratoconus Corneal Epithelium

Eleven miRNAs with greater than 10-fold differential expression were selected for TaqMan PCR validation using 12 additional independent surgical corneal epithelial samples. Of the 11 tested miRNAs, six miRNAs showed significant (P < .05 and > 1.5 fold change) down-regulation in keratoconic corneal epithelia (hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-28-5p, and hsa-miR-181a-2-3p; Table B). This validated the differential miRNA expressions between normal and keratoconic corneal epithelia.

Gene Ontology and Pathway Analyses of miRNA Target Genes

The 12 differentially expressed miRNAs in the microarray experiment were further applied for prediction of their target genes based on the Human hg19 (UCSC) database in the GeneSpring software. A total of 1,150 significant target genes were predicted from the five databases in the GeneSpring software (P < .05). Gene ontology analysis by DAVID (1,145 DAVID identities) demonstrated that the predicted miRNA target genes are involved in cell junction (5.68%, P = 5.79 × 10−6), cell division (3.06%, P = 3.31 × 10−3), microtubule (2.53%, P = 1.22 × 10−2), and motor activity (1.05%, P = 1.32 × 10−3; Table 3).

Gene Ontology Analysis of Predicted Target Genes of Differentially Expressed miRNAs From Microarray

Table 3:

Gene Ontology Analysis of Predicted Target Genes of Differentially Expressed miRNAs From Microarray

Pathway analysis by FunRich (1,071 FunRich identities) suggested that the predicted miRNA target genes could possibly be involved in the syndecan-mediated signaling pathway (30.4%, P < .001), vascular endothelial growth factor (VEGF), and VEGF receptor (VEGFR) signaling pathway (29.3%, P < .001), endothelins (29.3%, P < .001), PDGF receptor signaling network (29.1%, P < .001), integrin family cell surface interactions (30.6%, P < .001), IFN-gamma pathway (29.1%, P < .001), and ErbB receptor signaling network (29.3%, P < .001; Figure 3). The predicted target genes are interactively connected in a vast network (Figure B, available in the online version of this article). Moreover, the FunRich analysis also identified that SLC2A10, ANTXR1, and ZNF469 from our predicted gene list were associated with keratoconus (Table D, available in the online version of this article).

Pathway analysis of predicted micro-ribonucleic acid (miRNA) target genes. Top ten significantly enriched pathways from the 1,150 predicted miRNA target genes are presented (P < .001). VEGF = vascular endothelial growth factor

Figure 3.

Pathway analysis of predicted micro-ribonucleic acid (miRNA) target genes. Top ten significantly enriched pathways from the 1,150 predicted miRNA target genes are presented (P < .001). VEGF = vascular endothelial growth factor

Interaction network of micro-ribonucleic acid (miRNA) target genes in proteoglycan syndecan-mediated signaling pathway using FunRich software (http://funrich.org).

Figure B.

Interaction network of micro-ribonucleic acid (miRNA) target genes in proteoglycan syndecan-mediated signaling pathway using FunRich software (http://funrich.org).

Predicted miRNA Target Genes Related to Previous Keratoconus Studies or Predicted to Be Linked With Keratoconus

Table D:

Predicted miRNA Target Genes Related to Previous Keratoconus Studies or Predicted to Be Linked With Keratoconus

In comparison with previous transcriptomic, genetic, and proteomic studies (Table D),3,5,12 four genes (CEACAM6, FHL2, S100A2, ARF3) from our predicted miRNA target gene list were in common with the transcriptomic study,5 four genes (LRRK1, FNDC3B, RXRA, ZNF469) with our previous GWAS analysis,3 and three genes (INSR, VEGFA, PPARGC1A) with the proteomic study,12 indicating that our miRNA profile could be coherent to the mechanisms of keratoconic corneal epithelial changes from different investigations.

Protein Expression of miRNA Target Genes

To characterize the role of miRNAs in keratoconic corneal epithelia, two miRNA target genes (S100A2 and PTK2; Table E, available in the online version of this article) were selected and further evaluated by immunofluorescence analysis. We observed a higher expression of S100A2 protein, which was predicted to be regulated by hsa-miR-138-5p and hsa-miR-185-5p (Table D), in the basal cell layer of keratoconic corneal epithelia compared to that of normal corneal epithelia (Figure 4 and Figure B). In contrast, focal adhesion kinase (FAK; PTK2) only showed slightly increased expression in the wing cell layer as compared to the normal corneal epithelia. The immunofluorescence results were further confirmed by immunoblotting analysis. S100A2 protein expression in keratoconic corneal epithelia was 2.49-fold higher than in normal corneal epithelia (P < .05; Figure 5), whereas FAK protein expression did not show a significant difference between keratoconic and normal corneal epithelia (1.12-fold, P = .591). These findings suggested that the downregulation of hsa-miR-138-5p and hsa-miR-185-5p could be correlated with the upregulation of S100A2 protein in keratoconic corneal epithelia.

Predicted miRNA Target Genes in the Syndecan-mediated Signaling PathwayPredicted miRNA Target Genes in the Syndecan-mediated Signaling PathwayPredicted miRNA Target Genes in the Syndecan-mediated Signaling Pathway

Table E:

Predicted miRNA Target Genes in the Syndecan-mediated Signaling Pathway

Immunofluorescence analysis of S100A2 and focal adhesion kinase (FAK) proteins in normal and keratoconic corneal epithelia. Higher expression of S100A2 protein was observed in the basal cell layer of keratoconic corneal epithelium compared to that of normal corneal epithelium, whereas FAK only showed slightly increased expression in the wing cell layer.

Figure 4.

Immunofluorescence analysis of S100A2 and focal adhesion kinase (FAK) proteins in normal and keratoconic corneal epithelia. Higher expression of S100A2 protein was observed in the basal cell layer of keratoconic corneal epithelium compared to that of normal corneal epithelium, whereas FAK only showed slightly increased expression in the wing cell layer.

Immunoblotting analysis of S100A2 and focal adhesion kinase (FAK) proteins in normal and keratoconic corneal epithelia. S100A2 protein expression in keratoconic corneal epithelium was 2.49 fold higher than that in normal corneal epithelium (P < .05), whereas FAK protein expression did not show a significant difference between keratoconic and normal corneal epithelia (1.12 fold, P = .591). Beta-actin was used as a housekeeping protein.

Figure 5.

Immunoblotting analysis of S100A2 and focal adhesion kinase (FAK) proteins in normal and keratoconic corneal epithelia. S100A2 protein expression in keratoconic corneal epithelium was 2.49 fold higher than that in normal corneal epithelium (P < .05), whereas FAK protein expression did not show a significant difference between keratoconic and normal corneal epithelia (1.12 fold, P = .591). Beta-actin was used as a housekeeping protein.

MiRNA Expression in Corneal Squamous Epithelia Collected From Impression Cytology

The impression cytology can obtain squamous cells of the corneal epithelia without the necessity of surgery, and the samples showed corneal epithelial marker (KRT3) expression (Figure 6A). We evaluated the expression of microarray-identified miRNAs by TaqMan PCR on 16 corneal samples collected using impression cytology. Of the 11 tested miRNAs, 4 were significantly downregulated in keratoconic corneal epithelia compared to the normal samples (hsa-miR-151a-3p, 5.01 fold, P = .036; hsa-miR-195-5p, 38.78 fold, P < .001; hsa-miR-185-5p, > 100 fold, P = .013; and hsamiR-194-5p, 46.49 fold, P = .011; Figure 6B).

Micro-ribonucleic acid (miRNA) expression in EYEPRIM (OPIA Technologies SAS, Paris, France) samples. (A) Expression of corneal and limbal specific genes in EYEPRIM samples. Membrane samples 1 and 2 expressed corneal epithelial markers KRT3. Retinoblastoma cells Y79 and membrane without cells were used as negative controls. (B) miRNA expression analysis on keratoconic and normal corneal epithelia collected using EYEPRIM by TaqMan PCR. The miRNAs hsa-151a-3p, hsa-185-5p, hsa-194-5p, and hsa-195-5p were significantly downregulated in keratoconic corneal epithelium, whereas hsa-193b-5p was significantly upregulated. *P < .05; **P < .01; ***P < .001.

Figure 6.

Micro-ribonucleic acid (miRNA) expression in EYEPRIM (OPIA Technologies SAS, Paris, France) samples. (A) Expression of corneal and limbal specific genes in EYEPRIM samples. Membrane samples 1 and 2 expressed corneal epithelial markers KRT3. Retinoblastoma cells Y79 and membrane without cells were used as negative controls. (B) miRNA expression analysis on keratoconic and normal corneal epithelia collected using EYEPRIM by TaqMan PCR. The miRNAs hsa-151a-3p, hsa-185-5p, hsa-194-5p, and hsa-195-5p were significantly downregulated in keratoconic corneal epithelium, whereas hsa-193b-5p was significantly upregulated. *P < .05; **P < .01; ***P < .001.

Discussion

Keratoconus has long been described as a corneal stromal disease. However, in vivo confocal images showed abnormal features in every layer of the cornea, including abnormal epithelial and stromal keratocytes.8 Microscopic features of keratoconic corneas indicated irregular arrangement in basal epithelial cells,4,8 reduction in basal epithelial cell and keratocyte densities,4,13 and epithelial apoptosis along with corneal epithelium thinning.9 Likewise in our study, the keratoconic corneal epithelial cells were loosely arranged (Figure 1A) and significantly fewer in number compared to normal corneal epithelia (Figure 1B). The apoptosis-related marker, p53 protein, was upregulated in keratoconic corneal epithelia (Figure 1C), suggesting degeneration of this layer. Previous proteomic studies also identified a marked increase in apoptosis-related proteins, including p53 protein, in corneal epithelia.6,12 In addition, because the cathepsins are involved in the apoptotic pathway,14 the elevated corneal expression of cathepsin B and G in keratoconus indicates that cell apoptosis occurs in keratoconic corneal epithelia.15 The upregulation of interleukin-6, tumor necrosis factor-α, and matrix metalloproteinase-9 in tears and corneal epithelia of patients with keratoconus further suggests the presence of chronic inflammation, cell apoptosis, and destruction of cell integrity in keratoconic corneal epithelia.16 Lysyl oxidase, which is essential for stabilization, integrity, and elasticity of collagen fibrils, is decreased in keratoconic corneal epithelia.17 A previous transcriptome study on keratoconic corneal epithelia also suggested changes in cytoskeleton, reduced extracellular matrix remodeling, and modified cell-to-cell and cell-to-matrix interactions.5 Together with the gene ontology analysis on cell adhesion and cell division predicted from the miRNA target genes in our study (Table 3), these findings imply that cell integrity alteration and cell apoptosis are the possible pathological changes in the degeneration of corneal epithelium in keratoconus.

The miRNA regulation is one of the possible pathological mechanisms in various corneal diseases, including diabetic corneal nerve regeneration,18Pseudomonas aeruginosa–induced keratitis,19 and keratoconus.10 With reference to the recent deep sequencing analysis on normal human cornea,20,21 297 miRNAs were identified and expressed in human cornea, with which 172 miRNAs have an average expression signal greater than 5 (Table C). Our microarray analysis identified 211 miRNAs in our corneal epithelial samples and 94 miRNAs (55%) were commonly expressed when compared to the deep sequencing study, suggesting the corneal specificity of our samples as indicated by our miRNA profile. Moreover, the deep sequencing analysis on normal human cornea also suggested that miR-143-3p, miR-153-3p, miR-182-5p, miR-183-5p, miR-186-5p, miR-223-3p, miR-27a-3p, miR-9-5p, and miR-96-5p could target the genes associated with keratoconus (DOCK9, FNDC3B, FOXO1, and RAB3GAP1).11 In our study, 12 human miRNAs were downregulated in keratoconic corneal epithelia by microarray analysis (Table B). Using TaqMan PCR, we confirmed the downregulation of six miRNAs in corneal epithelia from surgery and four miRNAs in squamous corneal epithelial samples collected from impression cytology, suggesting that they could potentially be applied for keratoconus prediction.

The miR-151a has been found to be expressed differentially in atopic dermatitis (an inflammatory disease)21 and Alzheimer's disease (a neuronal degenerative disease),22 and inhibition of miR-151a is related to suppression of tumor growth.23 The miR-194 plays a crucial role in tumor progression and inhibits proliferation and metastasis in cancer cells.24 These findings suggest that downregulation of these miRNAs in keratoconic corneal epithelium could contribute to its degenerative processes, including promotion of cell apoptosis and inhibition of cell migration, as indicated by our gene ontology analysis of miRNA target genes (Table 3).

Previous transcriptome analysis identified 56 candidate genes differentially expressed in keratoconus samples,5 and four of them (CEACAM6, FHL2, S100A2, and ARF3) are in common with our predicted miRNA target genes (Table D). We selected S1002A (predicted to be regulated by hsa-miR-138-5p and hsa-miR-185-5p) and PTK2 (predicted to be regulated by hsa-miR-138-5p) for further analysis, especially the former being reported in the transcriptome study (Table D) and involved in the proteoglycan syndecan-mediated signaling pathway (Figure 3, Figure C, available in the online version of this article, and Table E), whereas PTK2 is involved in cell adhesion and motor activity.25 We confirmed that S1002A protein expression is up-regulated in keratoconic corneal epithelia (Figure 5), with a specific location in the epithelial basal cell layer (Figure 4). Our findings are concordant with a previous immunohistochemistry study.26

Immunofluorescence analysis with no primary antibody in normal and keratoconic corneal epithelia.

Figure C.

Immunofluorescence analysis with no primary antibody in normal and keratoconic corneal epithelia.

The pathway analysis of our miRNA target genes identified the proteoglycan syndecan-mediated signaling pathway, integrin family cell surface interactions, and VEGF and VEGFR signaling pathway (Figure 3). A study on syndecan-1 knockout mice identified the role of syndecan-1 in mediating cell proliferation and regulation of integrin expression in normal and wounded corneal epithelial tissues.27 Loss of syndecan-1 in corneal stromal cells could influence cell migration rates, fibrillar adhesions, and integrin activation.28 Therefore, syndecan-mediated signaling pathway-related cell migration and cell adhesion closely match our gene ontology analysis of miRNA target genes (Table 3). The S100A2 involved in the proteoglycan syndecan-mediated signaling pathway (Table E) has been shown to be upregulated in keratoconic corneal epithelia in previous immunohistochemistry study26 and in our protein expression analyses (Figures 45), suggesting that the proteoglycan syndecan-mediated signaling pathway dysregulation could be a potential mechanism for keratoconic corneal epithelia degeneration. How the proteoglycan syndecan-mediated signaling pathway is involved in keratoconus and whether other mechanisms, such as VEGF pathway and integrin family cell surface interactions, could be involved warrants further investigation. Nevertheless, a previous proteomic study identified VEGFA as a predicted upstream regulator in keratoconic corneal epithelia,12 and this also matches with miRNA target genes in our study (Table D).

To translate our findings of miRNA expression into clinical application, we evaluated the EYEPRIM conjunctival impression cytology device, which can procure cells from squamous corneal epithelia for biological testing without the necessity of surgery. The device allows quick sampling of conjunctival cells, which can then be analyzed for dry eye bio-markers.29 We used it for collection of corneal epithelia for the first time. In our study, we confirmed that the samples collected by impression cytology are cornea-specific because they expressed the cor-neal epithelial marker (Figure 6A). We propose that the EYEPRIM or an equivalent device can be used as a molecular diagnostic tool for corneal diseases. The miRNA expression of the EYEPRIM-collected squamous corneal epithelial cells could be a convenient and feasible method for keratoconus prediction (Figure 6B).

This study resolved the histological structure of keratoconic corneal epithelia and identified cell apoptosis, cell integrity alteration, and the downregulation of miRNAs as the potential mechanisms for keratoconic corneal epithelial degeneration. In addition, impression cytology is a useful modality to collect and analyze the miRNA expression of corneal epithelia.

References

  1. Jhanji V, Sharma N, Vajpayee RB. Management of keratoconus: current scenario. Br J Ophthalmol. 2011;95:1044–1050. doi:10.1136/bjo.2010.185868 [CrossRef]
  2. Lu Y, Vitart V, Burdon KP, et al. Genome-wide association analyses identify multiple loci associated with central corneal thickness and keratoconus. Nat Genet. 2013;45:155–163. doi:10.1038/ng.2506 [CrossRef]
  3. Cuellar-Partida G, Springelkamp H, Lucas SE, et al. WNT10A exonic variant increases the risk of keratoconus by decreasing corneal thickness. Hum Mol Genet. 2015;24:5060–5068. doi:10.1093/hmg/ddv211 [CrossRef]
  4. Mathew JH, Goosey JD, Bergmanson JP. Quantified histopathology of the keratoconic cornea. Optom Vis Sci. 2011;88:988–997. doi:10.1097/OPX.0b013e31821ffbd4 [CrossRef]
  5. Nielsen K, Birkenkamp-Demtröder K, Ehlers N, Orntoft TF. Identification of differentially expressed genes in keratoconus epithelium analyzed on microarrays. Invest Ophthalmol Vis Sci. 2003;44:2466–2476. doi:10.1167/iovs.02-0671 [CrossRef]
  6. Nielsen K, Vorum H, Fagerholm P, et al. Proteome profiling of corneal epithelium and identification of marker proteins for keratoconus: a pilot study. Exp Eye Res. 2006;82:201–209. doi:10.1016/j.exer.2005.06.009 [CrossRef]
  7. Reinstein DZ, Gobbe M, Archer TJ, Silverman RH, Coleman DJ. Epithelial, stromal, and total corneal thickness in keratoconus: three-dimensional display with Artemis very-high frequency digital ultrasound. J Refract Surg. 2010;26:259–271. doi:10.3928/1081597X-20100218-01 [CrossRef]
  8. Maeda N, Nakagawa T, Higashiura R, Fuchihata M, Koh S, Nishida K. Evaluation of corneal epithelial and stromal thickness in keratoconus using spectral-domain optical coherence tomography. Jpn J Ophthalmol. 2014;58:389–395. doi:10.1007/s10384-014-0338-0 [CrossRef]
  9. Kim WJ, Rabinowitz YS, Meisler DM, Wilson SE. Keratocyte apoptosis associated with keratoconus. Exp Eye Res. 1999;69:475–481. doi:10.1006/exer.1999.0719 [CrossRef]
  10. Hughes AE, Bradley DT, Campbell M, et al. Mutation altering the miR-184 seed region causes familial keratoconus with cataract. Am J Hum Genet. 2011;89:628–633. doi:10.1016/j.ajhg.2011.09.014 [CrossRef]
  11. Drewry M, Helwa I, Allingham RR, Hauser MA, Liu Y. miRNA profile in three different normal human ocular tissues by miRNA-seq. Invest Ophthalmol Vis Sci. 2016;57:3731–3739. doi:10.1167/iovs.16-19155 [CrossRef]
  12. Chaerkady R, Shao H, Scott SG, Pandey A, Jun AS, Chakravarti S. The keratoconus corneal proteome: loss of epithelial integrity and stromal degeneration. J Proteomics. 2013;87:122–131. doi:10.1016/j.jprot.2013.05.023 [CrossRef]
  13. Grieve K, Georgeon C, Andreiuolo F, et al. Imaging microscopic features of keratoconic corneal morphology. Cornea. 2016;35:1621–1630. doi:10.1097/ICO.0000000000000979 [CrossRef]
  14. Szentmáry N, Szende B, Süveges I. P53, CD95, cathepsin and surviving pathways in Fuchs' dystrophy and pseudophakic bullous keratopathy corneas. Histol Histopathol. 2008;23:911–916.
  15. Zhou L, Sawaguchi S, Twining SS, Sugar J, Feder RS, Yue BY. Expression of degradative enzymes and protease inhibitors in corneas with keratoconus. Invest Ophthalmol Vis Sci. 1998;39:1117–1124.
  16. Lema I, Durán JA. Inflammatory molecules in the tears of patients with keratoconus. Ophthalmology. 2005;112:654–659. doi:10.1016/j.ophtha.2004.11.050 [CrossRef]
  17. Shetty R, Sathyanarayanamoorthy A, Ramachandra RA, et al. Attenuation of lysyl oxidase and collagen gene expression in keratoconus patient corneal epithelium corresponds to disease severity. Mol Vis. 2015;21:12–25.
  18. Wang Y, Zhao X, Wu X, Dai Y, Chen P, Xie L. microRNA-182 mediates sirt1-induced diabetic corneal nerve regeneration. Diabetes. 2016;65:2020–2031. doi:10.2337/db15-1283 [CrossRef]
  19. Muraleedharan CK, McClellan SA, Barrett RP, et al. Inactivation of the miR-183/96/182 cluster decreases the severity of pseudomonas aeruginosa-induced keratitis. Invest Ophthalmol Vis Sci. 2016;57:1506–1517. doi:10.1167/iovs.16-19134 [CrossRef]
  20. Karali M, Peluso I, Gennarino VA, et al. miRNeye: a microRNA expression atlas of the mouse eye. BMC Genomics. 2010;11:715. doi:10.1186/1471-2164-11-715 [CrossRef]
  21. Chen XF, Zhang LJ, Zhang J, et al. MiR-151a is involved in the pathogenesis of atopic dermatitis by regulating interleukin-12 receptor beta2. Exp Dermatol. 2016.
  22. Satoh J, Kino Y, Niida S. MicroRNA-Seq data analysis pipeline to identify blood biomarkers for Alzheimer's disease from public data. Biomark Insights. 2015;10:21–31. doi:10.4137/BMI.S25132 [CrossRef]
  23. Chiyomaru T, Yamamura S, Zaman MS, et al. Genistein suppresses prostate cancer growth through inhibition of oncogenic microRNA-151. PLoS One. 2012;7:e43812. doi:10.1371/journal.pone.0043812 [CrossRef]
  24. Zhu X, Li D, Yu F, et al. miR-194 inhibits the proliferation, invasion, migration, and enhances the chemosensitivity of non-small cell lung cancer cells by targeting forkhead box A1 protein. Oncotarget. 2016;7:13139–13152.
  25. Michael KE, Dumbauld DW, Burns KL, Hanks SK, Garcia AJ. Focal adhesion kinase modulates cell adhesion strengthening via integrin activation. Mol Biol Cell. 2009;20:2508–2519. doi:10.1091/mbc.E08-01-0076 [CrossRef]
  26. Nielsen K, Heegaard S, Vorum H, Birkenkamp-Demtröder K, Ehlers N, Orntoft TF. Altered expression of CLC, DSG3, EMP3, S100A2, and SLPI in corneal epithelium from keratoconus patients. Cornea. 2005;24:661–668. doi:10.1097/01.ico.0000153556.59407.69 [CrossRef]
  27. Stepp MA, Gibson HE, Gala PH, et al. Defects in keratinocyte activation during wound healing in the syndecan-1-deficient mouse. J Cell Sci. 2002;115:4517–4531. doi:10.1242/jcs.00128 [CrossRef]
  28. Stepp MA, Daley WP, Bernstein AM, et al. Syndecan-1 regulates cell migration and fibronectin fibril assembly. Exp Cell Res. 2010;316:2322–2339. doi:10.1016/j.yexcr.2010.05.020 [CrossRef]
  29. López-Miguel A, Gutiérrez-Gutierrez S, García-Vázquez C, Enríquez-de-Salamanca A. RNA collection from human conjunctival epithelial cells obtained with a new device for impression cytology. Cornea. 2017;36:59–63. doi:10.1097/ICO.0000000000000977 [CrossRef]

Clinical Parameters of Keratoconic and Normal Patients

ParameterKeratoconus Mean ± SD (Range)Normal Mean ± SD (Range)P
Age (y)29.75 ± 8.02 (18.00 to 47.00)33.39 ± 8.22 (21.00 to 49.00).070
Gender (M/F)11/1515/12.335
Anterior Ks (D)52.35 ± 3.998 (45.20 to 61.60)44.70 ± 1.52 (41.60 to 46.70)8.650 × 10−11
Anterior Kf (D)48.10 ± 3.46 (43.50 to 60.50)43.57 ± 1.36 (40.80 to 45.90)8.699 × 10−7
Anterior K avg (D)50.20 ± 3.55 (44.30 to 61.10)44.13 ± 1.41 (41.40 to 46.30)2.093 × 10−9
Anterior cylinder (D)4.10 ± 2.34 (0.50 to 10.30)1.12 ± 0.62 (0.10 to 2.80)2.000 × 10−6
Posterior K s (D)−8.03 ± 0.85 (−10.40 to −6.60)−6.40 ± 0.27 (−7.00 to −5.90)9.141 × 10−10
Posterior Kf (D)−7.21 ± 0.84 (−9.80 to −6.10)−5.54 ± 2.53 (−6.40 to −6.30).004
Posterior Kavg (D)−7.61 ± 0.83 (−10.10 to −6.40)−6.23 ± 0.22 (−6.50 to −5.80)1.580 × 10−8
Posterior cylinder (D)0.84 ± 0.39 (0.30 to 1.60)0.35 ± 0.19 (0.10 to 1.00)9.000 × 10−6
Pachymetry apex (μm)478.30 ± 30.39 (389.00 to 528.00)520.42 ± 28.64 (478.00 to 576.00)9.000 × 10−6
Pachymetry thinnest (μm)462.67 ± 39.81 (342.00 to 550.00)515.29 ± 29.75 (475.00 to 569.00)7.000 × 10−6
Anterior BFS (mm)7.14 ± 0.43 (5.88 to 7.89)7.71 ± 0.26 (7.35 to 8.22)3.032 × 10−7
Posterior BFS (mm)5.83 ± 0.39 (4.65 to 6.52)6.49 ± 0.23 (6.04 to 6.93)3.364 × 10−9

Number of Squamous, Wing, and Basal Cells in Keratoconic and Normal Corneal Epithelia (cells/100 μm)

ParameterKeratoconus Mean ± SD (Range)Normal Mean ± SD (Range)P
Squamous cells3.01 ± 0.53 (2.82 to 3.20)6.42 ± 1.04 (5.62 to 7.48)< .001
Wing cells9.08 ± 1.41 (8.44 to 9.72)20.48 ± 3.35 (18.39 to 24.55)< .001
Basal cells7.42 ± 0.94 (7.02 to 7.82)11.06 ± 1.00 (10.29 to 11.9)< .001

Gene Ontology Analysis of Predicted Target Genes of Differentially Expressed miRNAs From Microarray

Functional AnnotationEnrichment ScoreaCount%P
Cell junction4.07655.685.79 × 10−6
Src homology-3 domain3.00252.189.60 × 10−4
Cell division2.43353.063.31 × 10−3
Zinc-finger2.2913411.708.86 × 10−5
FERM domain1.97100.871.24 × 10−3
Serine/threonine-protein kinase1.69332.881.52 × 10−2
Synapse1.60383.329.81 × 10−5
Transcription regulation1.4615913.891.54 × 10−3
Microtubule1.42292.531.22 × 10−2
Pleckstrin homology-like domain1.39393.412.63 × 10−3
Motor activity1.33121.051.32 × 10−3

Antibodies for Protein Expression Analysis

ProteinCompanyCatalog No.SourceDilution Factor
S100A2Abcamab109494rabbit1:500
FAKUpstate06-543rabbit1:500
P53Abcamab1101mouse1:500
ß-ACTIN-HRPSigmaA3854mouse1:2,500

Differentially Expressed miRNAs in Keratoconic and Normal Samples by Microarray and Taqman

miRNAsMicroarrayTaqMan


Fold (abs)RegulationPcorraFold (abs)RegulationPcorra
hsa-miR-4713-3p84.96down.0091.47down.246
hsa-miR-151a-3p73.95down.0021.50down.002
hsa-miR-193b-5p57.29down.0001.19down.166
hsa-miR-138-5p49.34down.0101.64down.003
hsa-miR-195-5p48.58down.0091.32down.115
hsa-miR-185-5p40.30down.0121.32down.055
hsa-miR-146b-5p39.65down.0491.67down.013
hsa-miR-194-5p37.04down.0001.59down.023
hsa-miR-28-5p36.49down.0031.55down.003
hsa-miR-612635.49down.0021.03up.915
hsa-miR-181a-2-3p18.86down.0021.65down.029
hsa-miR-30e-3p1.79down.013

Expression of Human Cornea microRNAs in Keratoconus Corneal Epithelium

microRNAsNormalKeratoconusFoldPcorrAvg Seq Signala
hsa-let-7a-5p13.9113.02−1.851.0008,418.14
hsa-let-7b-3p0.000.001.001.00034.29
hsa-let-7b-5p13.4312.53−1.86.385621.00
hsa-let-7c10.849.90−1.92.385941.71
hsa-let-7d-3p0.000.001.001.00026.86
hsa-let-7d-5p9.919.02−1.86.683132.43
hsa-let-7e-5p11.4710.68−1.73.288304.71
hsa-let-7f-2-3p0.000.001.001.0008.14
hsa-let-7f-5p13.0112.00−2.021.0004,693.86
hsa-let-7g-5p12.0311.08−1.931.0001,033.71
hsa-let-7i-5p10.419.37−2.06.926463.57
hsa-miR-100-5p7.465.91−2.94.882244.29
hsa-miR-101-3p5.862.82−8.201.000202.86
hsa-miR-103a-3p11.1610.28−1.841.000380.14
hsa-miR-106b-5p8.867.99−1.831.00010.71
hsa-miR-10a-5p0.000.001.001.00030.29
hsa-miR-10b-5p0.000.001.001.000687.57
hsa-miR-125a-5p9.368.61−1.68.978467.71
hsa-miR-125b-1-3p0.000.001.001.00022.71
hsa-miR-125b-5p9.198.31−1.841.000596.43
hsa-miR-126-3p0.000.001.001.00054.86
hsa-miR-126-5p0.000.001.001.0001,239.71
hsa-miR-127-3p0.000.001.001.000185.00
hsa-miR-1280.000.001.001.00021.00
hsa-miR-1307-5p0.000.001.001.00064.14
hsa-miR-130a-3p7.355.74−3.06.93553.43
hsa-miR-132-3p0.000.001.001.0008.00
hsa-miR-135a-5p0.000.001.001.0009.71
hsa-miR-135b-5p5.902.66−9.481.0008.71
hsa-miR-136-3p0.000.001.001.00038.29
hsa-miR-138-5p5.620.00−49.349.545 × 10−342.14
hsa-miR-140-5p0.000.001.001.0005.57
hsa-miR-141-3p9.889.52−1.281.000108.00
hsa-miR-141-5p0.000.001.001.00040.43
hsa-miR-142-5p0.000.001.001.00070.57
hsa-miR-143-3p0.000.001.001.0001,399.43
hsa-miR-145-5p0.000.001.001.00018.71
hsa-miR-146a-5p0.000.001.001.00071.43
hsa-miR-146b-5p5.310.00−39.65.04924.71
hsa-miR-148a-3p8.357.35−1.991.0006,566.00
hsa-miR-148a-5p0.000.001.001.000107.86
hsa-miR-148b-3p0.000.001.001.00058.57
hsa-miR-149-5p7.136.22−1.871.00052.86
hsa-miR-150-5p0.000.001.001.00023.43
hsa-miR-151a-3p6.210.00−73.951.840 × 10−3348.00
hsa-miR-151a-5p7.486.50−1.97.074210.57
hsa-miR-155-5p0.000.001.001.0007.43
hsa-miR-15a-5p7.877.06−1.751.0006.14
hsa-miR-15b-5p10.549.60−1.911.00019.29
hsa-miR-16-5p11.6610.79−1.821.000835.86
hsa-miR-17-5p8.107.11−1.991.00014.43
hsa-miR-181a-2-3p4.240.00−18.861.840 × 10−3128.57
hsa-miR-181a-3p0.000.001.001.00077.71
hsa-miR-181a-5p10.249.33−1.881.0006,724.00
hsa-miR-181b-5p8.867.80−2.07.683283.43
hsa-miR-181c-5p0.000.001.001.00095.57
hsa-miR-181d0.000.001.001.00011.86
hsa-miR-182-5p2.060.00−4.181.0002,063.00
hsa-miR-183-5p8.177.20−1.96.329138.00
hsa-miR-18418.2817.40−1.85.50462,657.86
hsa-miR-186-5p0.000.001.001.000602.14
hsa-miR-191-5p0.000.001.001.0004,670.29
hsa-miR-192-5p2.500.00−5.671.000585.86
hsa-miR-193a-3p6.235.65−1.501.0006.86
hsa-miR-193a-5p0.000.001.001.0007.86
hsa-miR-193b-3p10.469.71−1.681.000137.14
hsa-miR-194-5p5.210.00−37.044.315 × 10−419.43
hsa-miR-195-5p5.600.00−48.588.830 × 10−320.43
hsa-miR-197-3p5.102.73−5.191.00017.00
hsa-miR-199a-3p0.000.001.001.00058.57
hsa-miR-199a-5p0.000.001.001.00057.71
hsa-miR-199b-3p0.000.001.001.00061.57
hsa-miR-199b-5p0.000.001.001.000146.29
hsa-miR-19b-3p9.678.95−1.641.00021.00
hsa-miR-200a-3p8.667.58−2.101.00052.43
hsa-miR-200a-5p2.440.00−5.411.00015.00
hsa-miR-200b-3p12.0411.15−1.85.88232.29
hsa-miR-200c-3p12.2711.58−1.611.000386.57
hsa-miR-203a10.559.74−1.751.000368.43
hsa-miR-204-5p12.0911.61−1.401.0004,619.00
hsa-miR-205-5p14.7613.76−2.011.0004,434.71
hsa-miR-20a-5p9.709.00−1.621.00022.29
hsa-miR-2108.357.57−1.721.00061.00
hsa-miR-211-3p7.163.30−14.531.000351.14
hsa-miR-21-3p0.000.001.001.00060.29
hsa-miR-21-5p11.2610.98−1.221.0002,595.14
hsa-miR-221-3p6.815.69−2.191.000120.00
hsa-miR-222-3p7.506.81−1.611.00021.00
hsa-miR-22-3p11.8611.08−1.72.92613,686.00
hsa-miR-23a-3p12.5511.60−1.931.000294.71
hsa-miR-23b-3p11.1810.15−2.041.000306.71
hsa-miR-23c0.000.001.001.00043.29
hsa-miR-24-3p12.6311.73−1.86.73121.71
hsa-miR-25-3p8.317.23−2.13.807338.29
hsa-miR-26a-5p12.1811.16−2.031.00022,348.71
hsa-miR-26b-5p10.819.79−2.031.000272.29
hsa-miR-27a-3p13.1012.25−1.801.000235.86
hsa-miR-27a-5p0.000.001.001.0005.14
hsa-miR-27b-3p10.329.37−1.921.0003,342.00
hsa-miR-27b-5p0.000.001.001.00033.29
hsa-miR-28-3p0.000.001.001.000327.57
hsa-miR-28-5p5.190.00−36.493.190 × 10−39.57
hsa-miR-29a-3p11.0310.23−1.741.000152.71
hsa-miR-29b-3p6.836.68−1.101.00020.29
hsa-miR-29c-3p10.329.77−1.461.000100.29
hsa-miR-301a-3p0.000.001.001.00038.00
hsa-miR-30a-3p2.760.00−6.781.00067.71
hsa-miR-30a-5p7.456.37−2.12.760177.86
hsa-miR-30b-5p9.899.11−1.721.000379.57
hsa-miR-30c-1-3p0.000.001.001.0007.71
hsa-miR-30c-2-3p0.000.001.001.0005.29
hsa-miR-30c-5p9.368.42−1.921.000103.29
hsa-miR-30d-3p0.000.001.001.00011.14
hsa-miR-30d-5p8.607.60−1.99.646159.14
hsa-miR-30e-3p6.000.00−64.15.055210.29
hsa-miR-30e-5p7.756.72−2.041.00038.29
hsa-miR-320a9.949.18−1.70.452153.43
hsa-miR-3280.000.001.001.00014.57
hsa-miR-331-3p7.906.81−2.141.0008.71
hsa-miR-335-3p0.000.001.001.0006.86
hsa-miR-335-5p0.000.001.001.0006.86
hsa-miR-338-3p0.000.001.001.0005.29
hsa-miR-339-3p0.000.001.001.0005.71
hsa-miR-340-5p0.000.001.001.00045.43
hsa-miR-342-3p8.427.80−1.531.00025.29
hsa-miR-345-5p0.000.001.001.00011.57
hsa-miR-34a-5p7.536.42−2.161.00022.57
hsa-miR-34c-5p0.000.001.001.00016.57
hsa-miR-361-5p8.267.15−2.16.18331.43
hsa-miR-365a-3p8.227.24−1.981.00016.43
hsa-miR-365b-3p0.000.001.001.00012.86
hsa-miR-374a-3p0.000.001.001.00041.71
hsa-miR-374a-5p7.115.77−2.531.0009.14
hsa-miR-3750.000.001.001.00037.71
hsa-miR-378a-3p9.428.56−1.81.838524.71
hsa-miR-378a-5p0.000.001.001.0009.71
hsa-miR-381-3p0.000.001.001.00015.86
hsa-miR-409-3p0.000.001.001.0009.71
hsa-miR-4100.000.001.001.00013.57
hsa-miR-411-5p0.000.001.001.00029.14
hsa-miR-4210.000.001.001.00017.57
hsa-miR-423-3p0.000.001.001.000201.14
hsa-miR-423-5p6.345.34−2.00.31043.86
hsa-miR-425-5p8.026.79−2.34.3008.86
hsa-miR-4298.938.16−1.711.00069.43
hsa-miR-45100.000.001.001.00027.57
hsa-miR-454-3p0.000.001.001.00010.00
hsa-miR-4662a-5p0.000.001.001.0007.57
hsa-miR-4840.000.001.001.00016.43
hsa-miR-486-5p0.000.001.001.00088.29
hsa-miR-497-5p0.000.001.001.00018.71
hsa-miR-508-3p0.000.001.001.0005.14
hsa-miR-532-5p2.590.00−6.031.00041.86
hsa-miR-574-3p7.927.59−1.251.00074.43
hsa-miR-5982.650.00−6.271.0005.86
hsa-miR-6510-3p0.000.001.001.0009.29
hsa-miR-660-5p5.630.00−49.69.07125.00
hsa-miR-664a-3p0.000.001.001.00024.43
hsa-miR-671-3p0.000.001.001.00015.57
hsa-miR-708-3p0.000.001.001.0005.57
hsa-miR-708-5p0.000.001.001.00019.14
hsa-miR-744-5p0.000.001.001.00027.29
hsa-miR-769-5p0.000.001.001.000136.14
hsa-miR-77060.000.001.001.0006.43
hsa-miR-92a-3p7.656.38−2.42.5441,586.43
hsa-miR-92b-3p0.000.001.001.000210.14
hsa-miR-93-5p8.677.61−2.091.00071.00
hsa-miR-9410.000.001.001.00028.57
hsa-miR-96-5p8.487.48−1.991.00010.86
hsa-miR-98-5p6.125.26−1.821.000278.00
hsa-miR-99a-5p8.197.32−1.831.000233.14
hsa-miR-99b-5p7.656.34−2.48.413470.00

Predicted miRNA Target Genes Related to Previous Keratoconus Studies or Predicted to Be Linked With Keratoconus

SourcesGenesDescriptionmiRNAsPa
TranscriptomicsCEACAM6carcinoembryonic antigen related cell adhesion molecule 6hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-181a-2-3p, hsa-miR-151a-3p.039
TranscriptomicsFHL2four and a half LIM domains 2hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-181a-2-3p, hsa-miR-146b-5p.046
TranscriptomicsS100A2S100 calcium binding protein A2hsa-miR-185-5p, hsa-miR-138-5p.037
TranscriptomicsARF3ADP ribosylation factor 3hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p.022
ProteomicsINSRinsulin receptorhsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p, hsa-miR-138-5p.000
ProteomicsVEGFAvascular endothelial growth factor Ahsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-138-5p.015
ProteomicsPPARGC1APPARG coactivator 1 alphahsa-miR-138-5p, hsa-miR-194-5p.002
GeneticsLRRK1leucine rich repeat kinase 1hsa-miR-185-5p, hsa-miR-28-5p.013
GeneticsFNDC3Bfibronectin type III domain containing 3Bhsa-miR-181a-2-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-193b-5p, hsa-miR-146b-5p.025
GeneticsRXRAretinoid X receptor alphahsa-miR-138-5p.031
GeneticsZNF469zinc finger protein 469hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-138-5p.038
FunRichSLC2A10solute carrier family 2 member 10hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-194-5p.027
FunRichANTXR1anthrax toxin receptor 1hsa-miR-185-5p, hsa-miR-151a-3p, hsa-miR-138-5p.017
FunRichZNF469zinc finger protein 469hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-138-5p.038

Predicted miRNA Target Genes in the Syndecan-mediated Signaling Pathway

GenesDescriptionmiRNAsPa
ACHEacetylcholinesterase (Cartwright blood group)hsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-194-5p.033
ACTA2actin, alpha 2, smooth muscle, aortahsa-miR-146b-5p.022
ACTN3actinin alpha 3 (gene/pseudogene)hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-181a-2-3p, hsa-miR-138-5p.041
AGTR1angiotensin II receptor type 1hsa-miR-185-5p.022
ARandrogen receptorhsa-miR-28-5p, hsa-miR-151a-3p.023
ARF4ADP ribosylation factor 4hsa-miR-185-5p, hsa-miR-138-5p.023
ASAP1ArfGAP with SH3 domain, ankyrin repeat and PH domain 1hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p.026
ATF1activating transcription factor 1hsa-miR-138-5p, hsa-miR-151a-3p.013
ATF3activating transcription factor 3hsa-miR-181a-2-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-193b-5p, hsa-miR-146b-5p.020
BDNFbrain-derived neurotrophic factorhsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p, hsa-miR-138-5p.004
BMPR1Abone morphogenetic protein receptor type 1Ahsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-146b-5p.021
BRK1BRICK1, SCAR/WAVE actin nucleating complex subunithsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-146b-5p.032
BTRCbeta-transducin repeat containing E3 ubiquitin protein ligasehsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-193b-5p, hsa-miR-181a-2-3p, hsa-miR-151a-3p.012
CCNE1cyclin E1hsa-miR-138-5p, hsa-miR-195-5p.013
CCNKcyclin Khsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-181a-2-3p, hsa-miR-146b-5p.022
CD8ACD8a moleculehsa-miR-185-5p.043
CDC25Bcell division cycle 25Bhsa-miR-185-5p, hsa-miR-28-5p.003
CLIP1CAP-Gly domain containing linker protein 1hsa-miR-28-5p, hsa-miR-138-5p.048
CSE1Lchromosome segregation 1 likehsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-138-5p.037
CXCR4C-X-C motif chemokine receptor 4hsa-miR-185-5p, hsa-miR-146b-5p.019
DDX18DEAD-box helicase 18hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-28-5p, hsa-miR-4713-3p.022
DOCK1dedicator of cytokinesis 1hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-28-5p, hsa-miR-146b-5p.045
DUSP10dual specificity phosphatase 10hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p.040
DVL2dishevelled segment polarity protein 2hsa-miR-185-5p, hsa-miR-28-5p, hsa-miR-193b-5p, hsa-miR-138-5p.034
ENAHenabled homolog (Drosophila)hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-146b-5p.008
EPB41erythrocyte membrane protein band 4.1hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-138-5p.040
FESFES proto-oncogene, tyrosine kinasehsa-miR-28-5p, hsa-miR-138-5p.025
FGF23fibroblast growth factor 23hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-4713-3p.017
FHL2four and a half LIM domains 2hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-181a-2-3p, hsa-miR-146b-5p.046
FOXP3forkhead box P3hsa-miR-138-5p.022
FZD6frizzled class receptor 6hsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-193b-5p.030
GFRA1GDNF family receptor alpha 1hsa-miR-185-5p, hsa-miR-28-5p.006
GIPC1GIPC PDZ domain containing family member 1hsa-miR-194-5p, hsa-miR-138-5p.040
GIT2GIT ArfGAP 2hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p.038
GNAI3G protein subunit alpha i3hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-194-5p, hsa-miR-195-5p.001
GPX1glutathione peroxidase 1hsa-miR-185-5p, hsa-miR-4713-3p, hsa-miR-195-5p.006
GRB10growth factor receptor bound protein 10hsa-miR-181a-2-3p, hsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-193b-5p.005
GRB2growth factor receptor bound protein 2hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-193b-5p, hsa-miR-181a-2-3p.040
HDAC3histone deacetylase 3hsa-miR-185-5p, hsa-miR-28-5p, hsa-miR-146b-5p.011
HK1hexokinase 1hsa-miR-185-5p, hsa-miR-138-5p.029
HK2hexokinase 2hsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-28-5p, hsa-miR-146b-5p.019
ICAM1intercellular adhesion molecule 1hsa-miR-138-5p.043
IL6Rinterleukin 6 receptorhsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-193b-5p, hsa-miR-181a-2-3p, hsa-miR-146b- 5p, hsa-miR-138-5p.000
INSRinsulin receptorhsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p, hsa-miR-138-5p.000
IQGAP1IQ motif containing GTPase activating protein 1hsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-146b-5p.016
IRF7interferon regulatory factor 7hsa-miR-151a-3p.034
ITGA2integrin subunit alpha 2hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p.030
ITKIL2 inducible T-cell kinasehsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p.006
JUPjunction plakoglobinhsa-miR-193b-5p, hsa-miR-185-5p, hsa-miR-195-5p.037
KALRNkalirin, RhoGEF kinasehsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-151a-3p.028
KAT8lysine acetyltransferase 8hsa-miR-185-5p, hsa-miR-146b-5p.021
KDM1Alysine demethylase 1Ahsa-miR-28-5p, hsa-miR-4713-3p.046
KIF13Bkinesin family member 13Bhsa-miR-185-5p, hsa-miR-28-5p.016
KIR3DL1killer cell immunoglobulin like receptor, three Ig domains and long cytoplasmic tail 1hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-138-5p.021
KLC1kinesin light chain 1hsa-miR-185-5p, hsa-miR-195-5p.023
KPNA1karyopherin subunit alpha 1hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-151a-3p, hsa-miR-146b-5p.014
KRT1keratin 1hsa-miR-28-5p, hsa-miR-181a-2-3p, hsa-miR-146b-5p, hsa-miR-195-5p.001
MAPK12mitogen-activated protein kinase 12hsa-miR-193b-5p, hsa-miR-185-5p, hsa-miR-195-5p.027
MAPK3mitogen-activated protein kinase 3hsa-miR-195-5p, hsa-miR-185-5p.038
MAPKAPK2mitogen-activated protein kinase-activated protein kinase 2hsa-miR-195-5p, hsa-miR-185-5p.031
MLLT4myeloid/lymphoid or mixed-lineage leukemia; translocated to, 4hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-185-5p.006
NR3C1nuclear receptor subfamily 3 group C member 1hsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-138-5p.039
PA2G4proliferation-associated 2G4hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-193b-5p, hsa-miR-138-5p.046
PAK2p21 (RAC1) activated kinase 2hsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-138-5p.030
PATZ1POZ/BTB and AT hook containing zinc finger 1hsa-miR-195-5p, hsa-miR-185-5p, hsa-miR-146b-5p.010
PCBP4poly(rC) binding protein 4hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p, hsa-miR-138-5p.005
PEG10paternally expressed 10hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-138-5p.025
PLCB2phospholipase C beta 2hsa-miR-185-5p.022
PLD2phospholipase D2hsa-miR-138-5p.032
PLEKHA7pleckstrin homology domain containing A7hsa-miR-185-5p, hsa-miR-28-5p, hsa-miR-195-5p.001
POU1F1POU class 1 homeobox 1hsa-miR-185-5p.022
PPARGC1APPARG coactivator 1 alphahsa-miR-138-5p, hsa-miR-194-5p.002
PPM1Jprotein phosphatase, Mg2+/Mn2+ dependent 1Jhsa-miR-185-5p.043
PPP2CAprotein phosphatase 2 catalytic subunit alphahsa-miR-151a-3p, hsa-miR-195-5p.018
PRKACGprotein kinase cAMP-activated catalytic subunit gammahsa-miR-181a-2-3p, hsa-miR-138-5p, hsa-miR-151a-3p.004
PRKCGprotein kinase C gammahsa-miR-185-5p, hsa-miR-195-5p.007
PRKG1protein kinase, cGMP-dependent, type Ihsa-miR-181a-2-3p, hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-195-5p, hsa-miR-28-5p.047
PTK2protein tyrosine kinase 2hsa-miR-138-5p.031
RANRAN, member RAS oncogene familyhsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-181a-2-3p, hsa-miR-151a-3p.018
RAP1BRAP1B, member of RAS oncogene familyhsa-miR-28-5p, hsa-miR-194-5p.027
RELARELA proto-oncogene, NF-kB subunithsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-28-5p, hsa-miR-181a-2-3p, hsa-miR-138-5p.009
RFWD2ring finger and WD repeat domain 2hsa-miR-185-5p, hsa-miR-195-5p.046
ROCK2Rho associated coiled-coil containing protein kinase 2hsa-miR-138-5p.031
RORARAR related orphan receptor Ahsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p.033
RPS6KA1ribosomal protein S6 kinase A1hsa-miR-28-5p, hsa-miR-4713-3p, hsa-miR-138-5p.047
RRAGARas related GTP binding Ahsa-miR-146b-5p, hsa-miR-195-5p.037
RUNX1runt related transcription factor 1hsa-miR-185-5p, hsa-miR-146b-5p, hsa-miR-195-5p.046
S100A2S100 calcium binding protein A2hsa-miR-185-5p, hsa-miR-138-5p.037
SCN3Bsodium voltage-gated channel beta subunit 3hsa-miR-185-5p, hsa-miR-138-5p, hsa-miR-146b-5p.009
SERPINI1serpin family I member 1hsa-miR-195-5p.043
SETDB1SET domain bifurcated 1hsa-miR-28-5p, hsa-miR-4713-3p.050
SFXN3sideroflexin 3hsa-miR-194-5p, hsa-miR-193b-5p, hsa-miR-181a-2-3p, hsa-miR-138-5p.009
SH3GL2SH3 domain containing GRB2 like 2, endophilin A1hsa-miR-181a-2-3p, hsa-miR-138-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-146b-5p.002
SIN3ASIN3 transcription regulator family member Ahsa-miR-138-5p.031
SLC11A1solute carrier family 11 member 1hsa-miR-181a-2-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-146b-5p.029
SMURF1SMAD specific E3 ubiquitin protein ligase 1hsa-miR-195-5p, hsa-miR-185-5p, hsa-miR-138-5p.008
SNTA1syntrophin alpha 1hsa-miR-185-5p, hsa-miR-138-5p.016
STAMBPSTAM binding proteinhsa-miR-185-5p, hsa-miR-193b-5p, hsa-miR-181a-2-3p, hsa-miR-146b-5p.035
SYNJ1synaptojanin 1hsa-miR-181a-2-3p, hsa-miR-151a-3p, hsa-miR-138-5p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p, hsa-miR-146b-5p.003
TAF10TATA-box binding protein associated factor 10hsa-miR-138-5p.043
TCF3transcription factor 3hsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-28-5p, hsa-miR-138-5p.013
TERTtelomerase reverse transcriptasehsa-miR-138-5p.016
TFDP1transcription factor Dp-1hsa-miR-181a-2-3p, hsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-28-5p, hsa-miR-193b-5p, hsa-miR-146b-5p.001
TFRCtransferrin receptorhsa-miR-194-5p, hsa-miR-185-5p, hsa-miR-146b-5p.004
TLE4transducin like enhancer of split 4hsa-miR-28-5p, hsa-miR-195-5p, hsa-miR-185-5p.009
TNCtenascin Chsa-miR-185-5p, hsa-miR-138-5p.010
TNFRSF1Atumor necrosis factor receptor superfamily member 1Ahsa-miR-194-5p, hsa-miR-138-5p.048
TRAF6TNF receptor associated factor 6hsa-miR-194-5p, hsa-miR-193b-5p, hsa-miR-151a-3p, hsa-miR-146b-5p.011
UBE2Iubiquitin conjugating enzyme E2 Ihsa-miR-28-5p, hsa-miR-195-5p.029
USP6NLUSP6 N-terminal likehsa-miR-151a-3p, hsa-miR-185-5p, hsa-miR-194-5p, hsa-miR-195-5p.026
VDRvitamin D (1,25- dihydroxyvitamin D3) receptorhsa-miR-185-5p, hsa-miR-194-5p.021
VEGFAvascular endothelial growth factor Ahsa-miR-185-5p, hsa-miR-195-5p, hsa-miR-138-5p.015
WWP1WW domain containing E3 ubiquitin protein ligase 1hsa-miR-195-5p, hsa-miR-185-5p.031
YAP1Yes associated protein 1hsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-138-5p.049
YWHABtyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein betahsa-miR-28-5p, hsa-miR-185-5p, hsa-miR-146b-5p.003
ZCCHC12zinc finger CCHC-type containing 12hsa-miR-185-5p, hsa-miR-138-5p.049
Authors

From the Departments of Ophthalmology and Visual Sciences (YMW, TKN, WKC, CPP, VJ) and Obstetrics and Gynaecology (KWC, HKW), The Chinese University of Hong Kong, Hong Kong, China; the Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (VJ); and Centre for Eye Research Australia, University of Melbourne, Melbourne, Victoria, Australia (VJ).

The authors have no financial or proprietary interest in the materials presented herein.

Drs. Wang and Ng contributed equally to this work and should be considered as equal first authors.

Supported by a direct grant from the Medicine Panel of The Chinese University of Hong Kong (grant no. 4054239), Hong Kong, China.

AUTHOR CONTRIBUTIONS

Study concept and design (TKN, WKC, CPP, VJ); data collection (YMW, HKW); analysis and interpretation of data (TKN, KWC, HKW, WKC); writing the manuscript (YMW, TKN, VJ); critical revision of the manuscript (YMW, TKN, KWC, HKW, WKC, CPP, VJ); administrative, technical, or material support (VJ); supervision (TKN, WKC, VJ)

Correspondence: Vishal Jhanji, MD, FRCOphth, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, China. E-mail: jhanjiv@upmc.edu

Received: August 16, 2017
Accepted: December 08, 2017

10.3928/1081597X-20171215-02

Sign up to receive

Journal E-contents