https://orcid.org/0000-0002-1481-1061
Figures
Abstract
Primary open angle glaucoma is a leading cause of visual impairment and blindness which is commonly treated with drugs or laser but may require surgery. Tenon’s ocular fibroblasts are involved in wound-healing after glaucoma filtration surgery and may compromise a favourable outcome of glaucoma surgery by contributing to fibrosis. To investigate changes in gene expression and key pathways contributing to the glaucomatous state we performed genome-wide RNA sequencing. Human Tenon’s ocular fibroblasts were cultured from normal and glaucomatous human donors undergoing eye surgery (n = 12). mRNA was extracted and RNA-Seq performed on the Illumina platform. Differentially expressed genes were identified using a bioinformatics pipeline consisting of FastQC, STAR, FeatureCounts and edgeR. Changes in biological functions and pathways were determined using Enrichr and clustered using Cytoscape. A total of 5817 genes were differentially expressed between Tenon’s ocular fibroblasts from normal versus glaucomatous eyes. Enrichment analysis showed 787 significantly different biological functions and pathways which were clustered into 176 clusters. Tenon’s ocular fibroblasts from glaucomatous eyes showed signs of fibrosis with fibroblast to myofibroblast transdifferentiation and associated changes in mitochondrial fission, remodeling of the extracellular matrix, proliferation, unfolded protein response, inflammation and apoptosis which may relate to the pathogenesis of glaucoma or the detrimental effects of topical glaucoma therapies. Altered gene expression in glaucomatous Tenon’s ocular fibroblasts may contribute to an unfavourable outcome of glaucoma filtration surgery. This work presents a genome-wide transcriptome of glaucomatous versus normal Tenon’s ocular fibroblasts which may identify genes or pathways of therapeutic value to improve surgical outcomes.
Citation: Roodnat AW, Callaghan B, Doyle C, Vallabh NA, Atkinson SD, Willoughby CE (2024) Genome-wide RNA sequencing of ocular fibroblasts from glaucomatous and normal eyes: Implications for glaucoma management. PLoS ONE 19(7): e0307227. https://doi.org/10.1371/journal.pone.0307227
Editor: Ted S. Acott, Oregon Health and Science University, UNITED STATES OF AMERICA
Received: March 28, 2024; Accepted: June 28, 2024; Published: July 11, 2024
Copyright: © 2024 Roodnat et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All RNA-Seq data was deposited and released in the SRA database (Study:https://www.ncbi.nlm.nih.gov/bioproject/1078594). SRA reads can be accessed separately via https://www.ncbi.nlm.nih.gov/sra/?term=PRJNA1078594 as well.
Funding: This work was supported by Glaucoma UK (formerly known as International Glaucoma Association), Fight for Sight (UK), and the Royal Liverpool University Hospital Charitable Funds. The funders had no say in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Glaucoma is a progressive irreversible optic neuropathy characterised by loss of retinal ganglion cells ultimately leading to blindness if untreated [1]. The commonest form of glaucoma is primary open angle glaucoma (POAG) which had an estimated prevalence of roughly 55 million in 2020 [2, 3]. Risk factors for POAG include age, sex, genetic background and increased intraocular pressure (IOP); which is an important modifiable risk factor [4, 5]. POAG shows signs of a fibrotic condition [6, 7] and is associated with increased levels of profibrotic cytokine TGF-β2 in the aqueous humour [8].
Initial treatment of POAG commonly consists of SLT laser treatment or drugs intended to lower the IOP such as prostaglandins, β-adrenergic inhibitors and carbonic anhydrase inhibitors [9]. These drugs are usually applied topically in the form of eyedrops and frequently contain preservatives such as benzalkonium chloride (BAC). Unfortunately, BAC is toxic to the eye and has been associated with fibrosis, cell death, mitochondrial dysfunction, the generation of reactive oxygen species and DNA damage [9, 10]. During treatment of POAG surgical intervention in the form of glaucoma filtration surgery may be required. The failure rate of this procedure can be as high as 35% when evaluated after 5 years [11] and success depends on modulation of the wound healing response and prevention of fibrosis using anti-proliferative and anti-fibrotic drugs, such as 5-fluorouracil and mitomycin-C [12].
One cell type that is of particular importance in the postoperative wound-healing response is the Tenon’s ocular fibroblast (TF; also called Tenon’s capsule fibroblast) [13] because it is a main source of pro-fibrotic myofibroblasts involved in ECM deposition and wound contraction which may lead to scarring and subsequent bleb failure [12, 14, 15].
Previously it was shown that TFs from glaucomatous patients show signs of fibrosis when compared to non-glaucomatous patients by considering differential expression of a limited predefined panel of 10 genes [16]. The main purpose of this study was to perform a hypothesis-independent genome-wide RNA sequencing analysis to obtain a more complete view of the alterations in the transcriptome of glaucomatous Tenon’s ocular fibroblast (GTF) cells when compared to normal Tenon’s ocular fibroblasts (NTFs).
Materials and methods
Tenon’s ocular fibroblasts culturing and characterisation
Primary human Tenon’s ocular fibroblasts (TFs) were obtained from patients with POAG and non-glaucomatous patients during the period 06 Oct 2015 until 28 Feb 2017 using a previously described standardised method [17, 18]. Patient information can be found in S3 Table. Cells were maintained in low glucose Dulbecco`s Modified Eagle Media (DMEM) (Sigma, Gillingham, UK) which was supplemented with 10% fetal calf serum (Biosera, Heathfield, UK) and 2mM L-glutamine (Sigma, Gillingham, UK), Penicillin-Streptomycin (Sigma, Gillingham, UK), and 2.5 μg/mL Fungizone (amphotericin B, Sigma, Gillingham, UK). The cells were incubated at 37.0 °C, 95 % humidity and 5 % CO2. Cells were sub-cultured after reaching roughly 70 % confluency; cultures were used up to passage four. Tenon’s ocular fibroblast characterisation was carried out by fluorescent staining using a mouse monoclonal antibody (Thermofisher Scientific, USA) against vimentin which is a fibroblast biomarker. Bright focus microscopy and fluorescent vimentin and DAPI staining cell images of these Tenon’s ocular fibroblasts were shown previously in another publication by our group in [19].
RNA extraction
Total RNA was obtained from the cultured Tenon’s ocular fibroblasts using the Qiagen RNeasy Mini Kit (Qiagen, Manchester UK) following the manufacturer’s instructions. RNA concentration was determined using the NanoDrop 2000 instrument (Thermofisher Scientific, Horsham, UK). RNA quality was assessed using the Bioanalyser 2100 (Agilent Technologies, Stockport, UK) combined with an RNA 6000 Nano Kit (Agilent, Santa Clara, CA, USA).
RNA sequencing
Generation of sequencing libraries was achieved using the NEBNext Ultra TM RNA library Prep Kit (NEB, Ipswich, MA, USA). Purification of mRNA was performed using magnetic poly-T oligo-attached beads. RNA sequencing was done by Genewiz (Genewiz, Bishop’s Stortford, UK) using an Illumina platform thus producing a set of FASTQ files containing paired end reads.
RNA-Seq data analysis
Quality control of the resulting paired end reads was performed using program FastQC v0.11.9 [20]. Adaptor removal and quality trimming of reads was done using Cutadapt 3.0 [21] in a linux environment thus producing high-quality trimmed reads with a Phred quality score larger than 25 which corresponds to <0.3% base calling error rate. The high-quality adapter-trimmed reads were aligned to the primary assembly GRCh38 of the human genome sequence using the STAR aligner 2.7.6a [22]. The source of the primary assembly GRCh38 was the GENCODE comprehensive gene annotation for the primary assembly [23]. To quantify reads per feature (i.e. gene) the program FeatureCounts 2.0.1 [24] was utilised. This resulted in a table containing counts per sample and feature/gene. To reduce potential mapping false positives, genes with an average count-value lower than 100 were removed. Next, the filtered counts table was used as input to determine differential gene expression which was done using the program edgeR version 3.32.1 [25] in R studio. This procedure resulted in a gene list containing false discovery rate (FDR), p-value and counts per million reads mapped (CPM). Genes with an absolute |log2FC| > 0.26 (which corresponds to at least 20%-fold change) and FDR < 0.05 were considered to be differentially expressed genes (DEGs). These DEGs were used for subsequent analysis.
Functional enrichment analysis
Using the resulting list of DEGs, a functional analysis was done using enrichment tool Enrichr [26]. Enrichr utilises a modified Fisher exact test to determine over-representation of an input gene list in a.o. gene ontologies and pathway databases. Using Enrichr, functional and pathway enrichment was determined for pathway databases WikiPathways, KEGG, Reactome, MSigDB Hallmark genesets and furthermore for Gene Ontology (GO) biological processes and molecular functions. Pathways or GO terms were assumed to be enriched if the false discovery rate FDR < 0.01. The resulting enrichment terms tend to show a large amount of overlap and redundancy which can be partially removed using clustering. To cluster and visualise the functional enrichment results, EnrichmentMap 3.3.4 [27] was used in Cytoscape (version 3.9.1) combined with AutoAnnotate [28]. To further visualise and investigate interesting enriched KEGG pathways, pathway gene annotation was performed using Pathview 1.30.1 [29] in R. In a similar fashion, visualisation of gene expression in pathways from WikiPathways was achieved using PathVisio [30].
Results
Descriptive features of RNA-Seq data
RNA sequencing of 12 samples that were obtained from 12 different donors resulted in 24.8M 150 bp reads per sample on average. Usually the initial Phred quality score was larger than 25 (which corresponds to a <0.3% base calling error rate) but some reads required modest quality trimming which was done using the program Cutadapt thus arriving at quality scores which were guaranteed to be larger than 25. These paired trimmed high-quality reads were mapped to the genome (GRCh38) with an average mapping percentage of 90%.
Differential gene expression
Using the primary comprehensive gene annotation, paired trimmed high-quality reads were aligned to the genome sequence of primary assembly GRCh38. Genes with an average low number of mapped counts < 100 were filtered out. This procedure resulted in a counts-table for 11,227 aligned genes which were subsequently used to determine differentially expressed genes using edgeR in Rstudio. Using this program, fold-change (log2FC), log2CPM, false discovery rate (FDR) and p-value per gene were determined.
Fig 1 shows a volcano-plot of all genes in which thresholds were set at FDR = 0.05 and arbitrarily |log2FC| = 1.5 to show genes with highest fold-change. Of the total set of 11,227 expressed genes, 5817 were considered to be differentially expressed based on a false discovery rate FDR<0.05 and an absolute fold change |log2FC| > 0.26 (which corresponds to a fold change of at least 20%). Of these 5817 genes, 2664 genes were upregulated, and 3153 genes were downregulated. Housekeeping genes GAPDH and ACTB were not differentially expressed. Table 1 shows the top 25 upregulated and top 25 downregulated genes based on FDR and Fig 2 shows a heatmap of the normalised expression of top 50 upregulated and 50 downregulated genes with the lowest FDR. S1 Fig shows the 2 highest dimensions of a principal component analysis (PCA) which demonstrates good clustering of glaucomatous samples while normal samples show more spread, especially samples C3 and C5. A complete table containing all 11,227 genes that were expressed can be found in S1 Table.
In this figure grey and green dots represent genes that are not considered to be differentially expressed because the false discovery rate FDR>0.05, blue dots represent genes which are considered to be differentially expressed with a FDR<0.05 and |log2FC| ≤ 1.5 while red dots represent genes that are both significantly expressed and also have a high fold-change: FDR < 0.05 and |log2FC| > 1.5. In this work all genes with FDR < 0.05 and |log2FC| > 0.26 have been taken into account (all red genes and most blue genes) which amounts to 5817 genes of which 2664 genes were upregulated and 3153 genes were downregulated.
The expression pattern of each gene is shown in log2cpm has been normalised to enable comparison between genes. Hierarchical clustering of genes has been applied to group genes by normalised gene expression pattern.
Functional enrichment and pathway analysis
Using functional enrichment tool Enrichr, overexpression analysis was performed on the list of 5817 differentially expressed genes to identify significantly altered pathways in pathway databases KEGG, WikiPathways, Reactome, MSigDB Hallmark genesets and enriched molecular functions (MF) and biological processes (BP) in Gene Ontology (GO). Enrichment terms were assumed to be significantly changed between glaucomatous and normal TFs if the false discovery rate FDR<0.01. This enrichment analysis yielded a total of 787 enriched pathways and GO terms containing the majority of DEGs. These enrichment results were clustered into 176 clusters of which 89 clusters containing only one term. Those clusters containing more than one term have been visualised in Fig 3. The largest cluster contained 93 terms all related to “cell cycle” pointing to a significant overlap and redundancy in terms. A list of the top 50 enrichment results (as ranked by FDR) can be found in Table 2. A full list of clusters, terms and genes per term can be found in S2 Table.
The enrichment terms were obtained using an over-representation analysis (ORA) of a gene list of differentially expressed genes with false discovery rate FDR < 0.05 when comparing glaucomatous to normal Tenon’s capsule fibroblasts. Each node corresponds to an enrichment term in which the node intensity represents the FDR in a range 0.00 (darkest) to 0.01 (lightest) while the node size indicates the number of genes associated with a certain enrichment term. Blue lines connecting these nodes represent gene list overlap between enrichment terms above a threshold of 0.27. Clustered enrichment terms have been encircled and annotated.
Glaucomatous Tenon’s ocular fibroblasts show signs of fibroblast to myofibroblast transdifferentiation
Enrichment analysis showed significant enrichment of several fibrosis and EMT related enrichment terms such as MSigDB hallmark gene set “Epithelial Mesenchymal Transition” (FDR = 4e-15), “Positive Regulation Of Epithelial To Mesenchymal Transition” (GO:0010718). Furthermore “Epithelial to mesenchymal transition in colorectal cancer” (WP4239, p = 0.06) has a gene overlap of 56/160 and the annotated pathway shows several key genes involved in the processes of fibrosis as shown in Fig 4. Some key upregulated genes in this pathway are TGF-β1 and its receptor TGFBR2, transcription factors TWIST1, TWIST2, ZEB1, ZEB2 and transcriptional repressors SNAI1, SNAI2. Furthermore, effector genes FN1 and several collagens were upregulated which is in keeping with a profibrotic status. Not shown in Fig 4 is key fibrotic marker ACTA2 which was also upregulated (log2FC = 2.9).
Expression of genes in WikiPathway “Epithelial to mesenchymal transition in colorectal cancer” (WP4239). Differentially expressed genes in WikiPathway WP4239 have been annotated using package PathVisio. Red corresponds to upregulation in glaucomatous TF cells when compared to normal TF cells while blue corresponds to downregulation. Grey coloured genes are expressed but not differentially so. White coloured genes were not found to be expressed in our dataset or were filtered out due to low number of reads.
Increased synthesis and remodelling of the extracellular matrix in glaucomatous Tenon’s ocular fibroblasts
The enrichment analysis resulted in several terms associated with synthesis and remodelling of the extracellular matrix such as “Crosslinking Of Collagen Fibrils R-HSA-2243919” and “Regulation Of Extracellular Matrix Organization (GO:1903053)”, see S2 Fig. The expression of collagen species COL1A1, COL6, COL8, COL12 and possibly COL5A1 (FDR = 0.053) was upregulated in glaucomatous TF cells when compared to normal TF cells. Expression of laminin subunits LAMA4, LAMC1, LAMC2 was also upregulated. Furthermore, ECM components including VCAN (versican), FN1 (fibronectin), FBN1 (fibrillin-1), VTN (vitronectin) were upregulated in glaucomatous TF cells. ELN (elastin) was not differentially expressed. Latent transforming growth factor binding proteins LTBP1, LTBP2, LTBP3 were all strongly upregulated in glaucomatous TF cells and so was THBS1 (thrombospondin). With respect to collagen synthesis, ECM enzymes including LOX, LOXL1, LOXL3, LOXL4, PXDN, P4HA1, P4HA2, PLOD1, PLOD3 were all upregulated. Considering the matrix metallopeptidase family, ADAM12, ADAM10, ADAM23 were upregulated and so were MMP2 (log2FC = 2.4), MMP14 and MMP17 while MMP15 and MMP24 were downregulated. Conversely, metallopeptidase inhibitors TIMP1, TIMP2, TIMP3 were also strongly upregulated, for instance TIMP1 showed log2FC = 3.1 corresponding to an 8-fold increase with respect to normal TF cells.
Autophagy is upregulated in glaucomatous Tenon’s ocular fibroblasts
Enrichment analysis showed significant changes in autophagy in glaucomatous TFs compared to normal TFs as indicated by enriched Reactome pathways “Autophagy” (R-HSA-9612973, FDR = 1.8e-3), “Selective Autophagy” (R-HSA-9663891, FDR = 8.8e-3) and “Macroautophagy” (R-HSA-1632852) and GO biological processes “Macroautophagy” (GO:0016236) and “Regulation Of Macroautophagy” (GO:0016241). Some key genes involved in autophagy are shown in a heatmap in S3 Fig and in the annotated KEGG pathway hsa04140 in Fig 5.
The autophagy transcriptome in glaucomatous Tenon’s ocular fibroblasts by annotation of genes in KEGG pathway hsa04140 named “autophagy—animal” using R package Pathview. To enhance visibility, gene expression was normalised to a range of [–1,1] in which orange indicates upregulation and cyan indicates downregulation. Grey coloured genes are expressed but not differentially so. White coloured genes were not found to be expressed at sufficient levels in our dataset.
Several genes related to LC3 formation were upregulated such as MAP1LC3A, MAP1LC3B and GABARAPL1. Of the ATG (autophagy related) gene family, ATG2B, ATG10, ATG12, ATG14 were upregulated and ATG3, ATG5, ATG7 were moderately downregulated. Furthermore, LAMP1 and SQSTM1 (p62) were upregulated. While autophagy did appear to be upregulated, this was not clear for mitophagy. Several mitophagy-related enrichment results were identified such as “PINK1-PRKN Mediated Mitophagy” (R-HSA-5205685) but further inspection demonstrated that there was no change in expression of PINK1 and there was hardly any expression of PRKN and in fact this gene was filtered out due to low count numbers.
Mitochondrial fission is upregulated in glaucomatous Tenon’s ocular fibroblasts
Several biological processes involved in mitochondrial fission and fusion were significantly altered such as Mitochondrion Morphogenesis (GO:0070584) and Mitochondrion Organization (GO:0007005). The differentially expressed genes involved in these biological processes and other related processes are shown in Fig 6. Several key genes associated with mitochondrial fission were upregulated such as FIS1, MFF and several genes involved in mitochondrial fusion were downregulated such as MFN1 and OPA1 which suggests a functional enhancement of mitochondrial fission.
This heatmap shows genes with a false discovery rate FDR < 0.05 that are associated with key enrichment terms related to mitochondrial fission and fusion. Heatmap colours correspond to fold change (in log2FC) which has been limited to a range of [–4,4] to improve visibility.
Oxidative stress is increased in glaucomatous Tenon’s ocular fibroblasts
Oxidative stress was identified in our enrichment results, for instance “Cellular Response To Oxidative Stress (GO:0034599)” and MSigDB hallmark set “Reactive Oxygen Species Pathway”. The genes associated with some key enrichment terms are presented as a heatmap in Fig 7. Several antioxidant enzymes were upregulated including GLRX (glutaredoxin), SRXN1 (sulfiredoxin), TXNRD1 and glutathione peroxidases GPX8, GPX7, GPX4 while GPX3 was downregulated in glaucomatous TF cells. Peroxiredoxins PRDX1, PRDX4, PRDX5, PRDX6 were upregulated in glaucomatous TF cells while PRDX3 was downregulated. One component of glutathione S-transferase, namely GSTP1, was upregulated. Key regulator of antioxidant enzymes NFE2L2, also known as NRF2, was upregulated by a factor of two (log2FC = 0.94, FDR = 3e-3) in glaucomatous TF cells. There was no difference in expression of SOD (superoxide dismutase) enzymes between glaucomatous and normal TF cells. Overall, the antioxidant system and especially the glutathione subsystem was clearly upregulated in glaucomatous TF cells when compared to normal TF cells.
Heatmap showing genes with false discovery rate FDR < 0.05 that are associated with key enrichment terms related to oxidative stress and the antioxidant system. Heatmap colours correspond to fold change (in log2FC) between glaucomatous and normal Tenon’s ocular fibroblasts which has been limited to a range of [–4,4] to improve visibility.
Glaucomatous Tenon’s ocular fibroblasts show cell cycle arrest and senescence
Enrichment cluster “cell cycle” was the largest cluster in our enrichment results and contained 93 enrichment terms all related to the cell cycle such as KEGG “cell cycle” hsa04110 (FDR = 4e-14) and “Cell Cycle R-HSA-1640170” in Reactome, see S2 Table, indicating that there are differences in proliferation and cell cycle between glaucomatous and normal TFs. To illustrate key differences in cell cycle gene expression, KEGG diagram hsa04110 was annotated using Pathview (see Fig 8), demonstrating clear differences in gene expression between glaucomatous and normal TFs. Further inspection of Cyclins and Cdk inhibitor genes in glaucomatous versus normal TFs showed that CCND1 was upregulated and CCNB2, CCNE2, CCNB1, CCNA2 were all highly downregulated (log2FC around -3.5) in glaucomatous TFs, indicating that these cells were most likely in the G1 phase. Furthermore, CCNG1 was upregulated and Cdk inhibitor CDKN1A (p21, Cip1) was upregulated while all other Cdk inhibitors were either downregulated or not differentially expressed. Cyclin expression is in part regulated by transcription regulators of the E2F family. Within this family, E2F1, E2F3, E2F4 were all downregulated and so were their inhibitors RB1, RBL1 and RBL2. In addition to cell cycle arrest, glaucomatous TFs also showed signs of a senescence-associated secretory phenotype including upregulation of TGF-β1 (TGFB1, log2FC = 1.5, FDR = 0.006), IL6 (log2FC = 1.5, FDR = 0.045), IGFBP2, IGFBP3, IGFBP4, IGFBP6. SERPINE1 (PAI-1) was also upregulated but with FDR = 0.06. All VEGF species VEGFA, VEGFB, VEGFC were significantly upregulated. In summary, the glaucomatous TF cells show cell cycle arrest and a senescence-associated secretory phenotype. With respect to the TGF-β family it is interesting to note that while TGF-β1 (TGFB1) and TGF-β3 (TGFB3) were both upregulated, expression of TGF-β2 (TGFB2) was strongly downregulated (log2FC = -3.5, FDR = 2e-5) in glaucomatous TF cells when compared to normal cells.
Alterations in cell cycle have been shown by annotation of genes in KEGG pathway hsa04110 named “cell cycle” using R package Pathview. To enhance visibility, gene expression has been normalised to a range of [–1,1] in which orange indicates upregulation and cyan indicates downregulation. Grey coloured genes are expressed but not differentially so. White coloured genes were not found to be expressed at sufficient levels in our dataset.
Upregulation of the unfolded protein response in glaucomatous Tenon’s ocular fibroblasts
The unfolded protein response (UPR) appeared in several enrichment terms within cluster “UPR; cellular stress; senescence” such as “XBP1(S) Activates Chaperone Genes” (R-HSA-381038), “Unfolded Protein Response (UPR)” (R-HSA-381119), “ERAD Pathway” (GO:0036503) and “ATF4 Activates Genes In Response To Endoplasmic Reticulum Stress R-HSA-380994”. Further inspection and pathway analysis showed increased expression of DDIT3 (CHOP), see Fig 9 and the gene-enrichment heatmap in S4 Fig. There are three UPR branches that may be activated and that start by dissociation of HSPA5 (BiP) from EIF2AK3 (PERK), ATF6 or ERN1 (IRE1α). HSPA5 itself was not differentially expressed. EIF2AK3 (PERK) was also not differentially expressed but a paralog of this gene, EIF2AK4, was upregulated possibly indicating activation of this UPR branch. ATF6B was moderately upregulated. ERN1 (IRE1α) was not differentially expressed but its associated gene XBP1 was upregulated albeit with an FDR = 0.06. In summary the UPR is upregulated or at least altered in glaucomatous TFs when compared to normal TF cells.
Changes in the unfolded protein response and ER stress in Tenon’s ocular fibroblasts have been shown by annotation of genes in KEGG pathway hsa04141 “protein processing in endoplasmic reticulum” which also contains the Unfolded Protein Response (UPR). Annotation has been achieved using R package Pathview. To enhance visibility, gene expression has been normalised to a range of [–1,1] in which orange indicates upregulation and cyan indicates downregulation. Grey coloured genes are expressed but not differentially so. White coloured genes were not found to be expressed at sufficient levels in our dataset.
Glaucomatous Tenon’s ocular fibroblasts show altered apoptotic pathways
Enrichment analysis resulted in two clusters involved in apoptosis, “regulation of apoptosis” and “intrinsic apoptotic signalling pathway” and several unclustered terms. The intrinsic apoptotic pathway showed both pro-apoptotic and anti-apoptotic changes in glaucomatous TF cells when compared to normal cells, see the annotated KEGG pathway hsa04210 in Fig 10. BH3-only sensitizers PMAIP1 (also known as NOXA, log2FC = -2.7, FDR = 2e-4) and BMF were downregulated while BAD was upregulated. Pro-survival genes BCL2 (log2FC = -0.94) and MCL1 were downregulated while expression of BCL2L1 (BCL-XL) was unchanged and expression of pro-survival gene BCL2L2 (also known as BCL-W) was upregulated. BH3-only activators BCL2L11 (BIM) and BID were downregulated while BBC3 (PUMA) was upregulated. Effector BAX was upregulated while BAK1 was downregulated. As mentioned before, DDIT3 (CHOP) was upregulated in glaucomatous tissue, linking the UPR to apoptosis. The extrinsic apoptotic pathway seemed upregulated in glaucomatous TF cells and showed upregulation of several genes coding for death receptors such as FAS and TNFRSF1A (TNF-R1). Transcription factor TP53 (P53) was downregulated by a factor 4 (log2FC = -2.1) and so was the expression of some of its transcriptional targets such as PMAIP1 but on the other hand expression of other TP53-regulated genes BBC3 (PUMA), BAX and FAS was upregulated. Executioner caspases CASP3 and CASP7 were downregulated while expression of Caspase inhibitor XIAP was unchanged.
The pathway analysis is performed by annotation of genes in KEGG pathway hsa04210 “apoptosis” using R package Pathview. To enhance visibility, gene expression has been normalised to a range of [–1,1] in which orange indicates upregulation and cyan corresponds to downregulation. Grey coloured genes are expressed but not differentially so. White coloured genes were not found to be expressed at sufficient levels in our dataset.
Changes in inflammation in glaucomatous Tenon’s fibroblasts
Several enrichment results pointed to differences in inflammation between GTF and NTF cells such as Reactome pathways “MyD88-independent TLR4 Cascade” (R-HSA-166166), “TRAF6 Mediated Induction Of NFkB And MAP Kinases Upon TLR7/8 Or 9 Activation” (R-HSA-975138),”Interleukin-17 Signaling” (R-HSA-448424), “Interleukin-6 Signaling” (R-HSA-1059683) and “Activation Of NF-kappaB In B Cells” (R-HSA-1169091). Further inspection on gene level identified upregulation of cytokine IL6 (log2FC = 1.5, FDR = 0.04) and downregulation of IL1B (IL-1β) (log2FC = -4.5, FDR = 0.001). Of the cytokine receptors, IL31RA was strongly downregulated (log2FC = -4.1), IL17RA was upregulated and so was IL6 receptor associated gene IL6ST (log2FC = 1.4). Toll-like receptor 4 (TLR4) was upregulated (log2FC = 2.1, FDR = 0.03). Several members of the TNF receptor superfamily were upregulated, notably TNFRSF1A and TNFRSF1B. As noted earlier, expression of TGF-β1 was upregulated (TGFB1, log2FC = 1.5, FDR = 0.006) while expression of TGF-β2 was downregulated (log2FC = -3.5, FDR = 2e-5) in glaucomatous TF cells when compared to normal cells.
Discussion
In this study we performed a genome-wide hypothesis-independent RNA sequencing analysis to compare Tenon’s ocular fibroblasts from glaucoma (POAG) patients to those originating from non-glaucomatous patients. The large number of differentially expressed genes indicates that Tenon’s ocular fibroblasts from glaucomatous patients have undergone dramatic changes when compared to those from non-glaucomatous patients. Enrichment analysis revealed fibroblast to myofibroblast transdifferentiation and changes in mitochondrial organisation, autophagy, oxidative stress, ECM remodelling, inflammatory state, cell cycle arrest and senescence, unfolded protein response and apoptosis. These themes will be discussed in more detail below.
When comparing TF cells originating from glaucoma patients with those from normal patients, we found that glaucomatous TF cells showed upregulation of genes associated with a myofibroblast cell type such as ACTA2 (α-SMA), ZEB1, ZEB2, SNAI1 (SNAIL) and SNAI2 (SLUG) suggesting fibroblast to myofibroblast transdifferentiation. Myofibroblasts are contractile, ECM-producing cells that normally appear transiently during normal wound healing but become persistent in pathological fibrotic conditions such as systemic sclerosis [31], idiopathic pulmonary fibrosis [32] and liver fibrosis [33]. Myofibroblasts may originate from epithelial cells (via a process called EMT), endothelial cells (via EndMT), regular fibroblasts and other cells [34, 35]. Myofibroblasts and corresponding biological processes such as EMT have been reported to play a role in the context of glaucoma, for instance in retinal pigment epithelial cells [36], lamina cribrosa cells [37] and trabecular meshwork cells [38, 39]. In keeping with our results, Tenon’s ocular fibroblasts from glaucomatous patients were reported to show signs of a myofibroblast phenotype by determining differential expression of a predefined panel of 10 genes [16].
One mechanism by which fibroblasts may transdifferentiate into myofibroblasts is via exposure to POAG-related extracellular signal molecules such as TGF-β2 which is upregulated in the aqueous humour (AH) of POAG patients [8, 40]. It is plausible that AH containing POAG-associated ROS and cytokines can reach Tenon’s capsule and therefore TFs by drainage via the subconjunctival space and/or lymphatic vessels of the conjunctiva [41]. Also, TGF-β1 can cause fibroblast to myofibroblast transdifferentiation [42, 43] and may play an autocrine role in our experiment via a positive feedback mechanism because TGF-β1 (TGFB1) was upregulated in glaucomatous TF cells when compared to normal cells in our results. Another mechanism by which fibroblast to myofibroblast transdifferentiation may occur is in response to altered mechanical forces exerted onto the cell [44, 45]. It is not clear if such a mechanism has played a role in GTFs when compared to NTFs. The use of preservative benzalkonium chloride which is often present in topical glaucoma drugs has been demonstrated to cause TF fibroblast to myofibroblast transdifferentiation at lower concentrations than used in actual eyedrops [46] suggesting this could be a factor contributing to the fibroblast to myofibroblast transdifferentiation we observed in our results. Furthermore, exposure to reactive oxygen species (ROS) may also promote fibroblast to myofibroblast transdifferentiation [47, 48].
Overall, the increased expression of myofibroblast markers in glaucomatous TF tissue may be explained by exposure to POAG-related cytokines, growth factors and reactive oxygen species (ROS) in combination with ingredients of glaucoma drugs. Several aspects of fibrosis and myofibroblasts such as oxidative stress, autophagy, mitochondrial changes, extracellular matrix remodelling, inflammation, apoptosis and senescence will be discussed in more detail.
Our results showed increased expression of several ECM components such as collagen species, FN1, FBN1, FBN2 and others together with upregulation of ECM-modifying enzymes such as lysyl oxidases, prolyl hydroxylases, peroxidasin in glaucomatous TF cells when compared to normal TF cells. Upregulation of collagens is a key feature of many fibrotic conditions and especially the expression of COL1, COL3 and COL5 is increased in fibrosis [45]. Similar to our results, upregulation of MMP2 and MMP14 together with upregulation of TIMP1 has been shown in liver fibrosis [49]. Upregulation of TIMP2 and other TIMP species has been found in several forms of glaucoma as well [50–52] and will promote ECM accumulation. In keeping with our results, metallopeptidase ADAM12 has also been reported to be upregulated in glaucomatous lamina cibrosa cells [53] and is also associated with liver-fibrosis [54]. The expression of ADAM12 is regulated by TGF-β1 [55] and expression was shown to be upregulated in vitro in human trabecular meshwork cells after treatment with TGF-β1 [56]. Several lysyl oxidases were upregulated in our results which has also been shown in other myofibroblasts [34]. Upregulation of LOXL1 was also observed in vitro in Tenon’s ocular fibroblasts treated with TGF-β1 [57]. Since TGF-β1 was also upregulated in our data, LOXL1 expression may have been increased as a result of autocrine signalling via TGF-β1. Overall, glaucomatous TF cells show higher expression of structural ECM components while restraining the activity of matrix metalloproteases by increased expression of TIMPs which is expected to result in increased deposition of extracellular matrix proteins which highly correlates with fibrosis and fibroblast to myofibroblast transdifferentiation [45].
We observed differences in several autophagy related genes when comparing glaucomatous to normal TF cells. Several LC3 components of the MAP1LC3 family were upregulated but on the other hand ATG5 and ATG7 were moderately downregulated with a log2FC around -0.5. To assess differences in autophagy it would also have been interesting to determine the phosphorylation state of certain autophagy-related proteins such as LC3 which was beyond the scope of this study. Autophagy has been reported to be associated with several fibrotic conditions and can either promote or reduce fibrosis depending on cell type and cell state [58–60]. Furthermore, autophagy also plays a role in glaucoma. In POAG, autophagy in the trabecular meshwork has been shown to be dysregulated via the mTOR pathway [61] and autophagy may contribute to TM cell death [62] while autophagy deficiency protects against increased IOP and neurodegeneration in a mouse model [63, 64]. In pseudoexfoliation glaucoma dysfunctional autophagy may also play a causative role [65]. Overall, it appears that the observed differences in autophagy between glaucomatous and normal TF cells may be part of a fibrotic response.
Our results showed increased mitochondrial fission in glaucomatous TFs when compared to normal TFs. In several cell types mitochondrial fission has been reported to be associated with fibrosis. In corneal myofibroblast differentiation, mitochondrial fission was observed and in fact proved to be necessary to transdifferentiate corneal fibroblasts into myofibroblasts [66]. Mitochondrial fission and fragmentation were also reported in interstitial fibroblasts originating from fibrotic kidneys [67]. In myofibroblasts obtained from fibrotic lung tissue, mitochondrial fission was also found to be upregulated, presumably to meet higher energy demands [68]. With respect to glaucoma, increased mitochondrial fission has been observed after elevation of intraocular pressure in a mouse model [69]. Mitochondrial dysfunction and mitochondrial changes are strongly associated with glaucoma [70, 71] and with fibrotic conditions in general [72]. Mitochondrial changes have also been associated with exposure to glaucoma drug preservatives notably benzalkonium chloride (BAC). BAC has been reported to be causing mitochondrial dysfunction resulting in reduced ATP synthesis, production of reactive oxygen species [9, 10, 73] and mitochondrial fragmentation, although it should be noted that the latter observation was demonstrated in yeast [74].
The observed mitochondrial fission in our results is probably part of a fibrotic state which itself may have been caused by a combination of POAG pathology and glaucoma drug ingredients. The fact that mitochondrial changes have also been observed in animal models without BAC [69] suggests that at least POAG itself plays a role here.
There was upregulation of several key antioxidant enzymes and their primary regulator NFE2L2 (NRF2) in glaucomatous TF cells when compared to normal cells indicating that glaucomatous TF cells may have been subjected to increased oxidative stress levels compared to normal cells. In previous work our group demonstrated increased oxidative stress levels in glaucomatous Tenon’s ocular fibroblasts using the CellROX green assay [19]. Oxidative stress has also been reported to be associated with several fibrotic conditions such as pulmonary fibrosis [75]. One possible source of oxidative stress is BAC which has been shown to cause an increase in reactive oxygen species (ROS) and subsequent stress-induced premature senescence [10, 76] but oxidative stress is also a key feature of POAG and other forms of glaucoma [77, 78] and can be induced in a BAC-free mouse model using microbead occlusion [79]. Several transcriptional targets of NFE2L2 were upregulated in our results such as GPX4, PRDX6 and TXNRD1 [80] suggesting that NFE2L2 has not only been expressed at a higher level but also has a higher level of activation (i.e. phosphorylation), possibly via the PI3K/Akt pathway as described in [79]. NFE2L2 (NRF2) has been linked to glaucoma [81] and TGF-β2 [82] and may be a useful drug target to improve the outcome of glaucoma filtration surgery by reducing bleb fibrosis [83]. Reactive oxygen species have been shown to liberate latent TGF-β1 from the Large Latent Complex (consisting of TGF-β1, LAP and LTBP) by conformational change of LAP-beta1 [84] and possibly by other mechanisms as well [85].
When comparing glaucomatous TF cells to normal TF cells our results showed cell cycle arrest and upregulation of Cdk inhibitor protein CDKN1A (p21). CDKN1A binds to G1-Cdk thus inhibiting cell cycle progression at G1 [86]. It is interesting to note that CDKN1A is associated with POAG [87] and also with certain fibrotic conditions such as idiopathic pulmonary fibrosis [88]. Furthermore, our results showed signs of a senescence-associated secretory phenotype (SASP) as described in [89]. Senescence and cell cycle arrest are associated with glaucoma [90] as demonstrated by increased senescence in the outflow facility of POAG patients [91]. Upregulation of CDKN1A has also been linked to exposure to benzalkonium chloride, at least in human lung epithelial cells [92]. Cell cycle arrest can be induced by TGF-β2 (TGFB2) and is in this case associated with upregulation of CDKN2B [82] but in our results CDKN2B was not differentially expressed. In a model of pulmonary fibrosis, upregulation of CDKN1A and cell cycle arrest was also reported together with myofibroblast activation, telomere shortening and release of inflammatory cytokines pointing to a both senescent and fibrotic condition in which DNA damage causes cell cycle arrest via TP53 (p53) [93]. A similar mechanism may explain our results as DNA damage was enriched in our dataset. Interestingly expression of TP53 (p53) was strongly reduced in glaucomatous TFs which agrees with a study of pulmonary fibroblasts in which expression of TP53 was reported to be reduced in IPF patients [94]. Cyclin expression is in part regulated by transcription regulators of the E2F family [95, 96] and E2F1, E2F3, E2F4 were all downregulated. Apart from regulating several Cyclins, members of the E2F family of transcription regulators are also known to regulate their own transcription resulting in a positive feedback loop so it seems these regulators are at a stable low state in glaucomatous TFs.
The unfolded protein response (UPR) is associated with both POAG and fibrosis and was also found to be altered in our results. In POAG, activation of the UPR has been demonstrated in trabecular meshwork tissue in which expression of HSPA5, GRP94, ATF-4, ERO-1α and CHOP were all increased [97] and can be induced in trabecular meshwork cells when treated with TGF-β1 [56]. Activation of the UPR is also strongly associated with several fibrotic conditions [98, 99]. Our results show upregulation of DDIT3 (CHOP) which is associated with prolonged ER stress [99]. Expression of DDIT3 (CHOP) can be regulated by all three branches of UPR namely via EIF2AK3 (PERK), ATF6 and ERN1 (IRE1α) [100]. In our case ATF6B and it’s associated protease MBTPS1S1P (S1P) were upregulated suggesting that ATF6 may have caused upregulation of DDIT3. Further upstream, oxidative stress can activate several MAPKs such as JNK leading to upregulated expression of DDIT3 and induction of the UPR. Interestingly, increased expression of DDIT3 (CHOP) was also reported in vitro in human lung epithial cells after treatment with BAC [92]. Overall, our results suggest that glaucomatous Tenon’s ocular fibroblasts have been subjected to prolonged ER stress and activation of the UPR which could have been caused by exposure to cytokines, growth factors or ROS associated with POAG pathology and may be further exacerbated by exposure to glaucoma drugs and glaucoma drugs preservatives.
Both the intrinsic and extrinsic apoptotic pathways were altered in glaucomatous TF cells when compared to normal cells. Apoptosis in fibrotic conditions has been reported to be altered in myofibroblasts with respect to fibroblasts in the sense that myofibroblasts are primed for apoptosis but manage to escape this fate, for instance by upregulating pro-survival proteins such as BCL2L1 (BCL-XL) [34]. Evasion of apoptosis by myofibroblasts has been addressed by a class of drugs called BH3 mimetics which promote myofibroblast apoptosis thus reducing fibrosis [101]. Changes in pro-survival gene expression were also observed in our results: BCL2L2 (also known as BCL-W) was upregulated which may point to a similar apoptosis-evading mechanism because BCL2L2 suppresses effector BAX. BCL2L1 (BCL-XL) on the other hand was unchanged in our results. Apoptotic resistance is also affected by extracellular signal molecules TGF-β1 (TGFB1) and EDN1 (endothelin, ET-1). As noted before, TGF-β1 is upregulated in our results while expression of EDN1 was unchanged. TGF-β1 can be both pro-apoptotic or anti-apoptotic depending on cell type and context. For myofibroblasts TGF-β1 has been reported to act as a pro-survival cytokine and can inhibit sensitiser BAD via the FAK-AKT pathway [102, 103]. TGF-β1 also influences the extrinsic apoptotic pathway by inhibiting FAS-FASL signalling [34]. Biomechanical signalling is yet another survival signal affecting the myofibroblast apoptotic pathway and relaxation of the ECM can trigger apoptosis in fibroblasts [104]. This type of signalling is mediated via integrins and especially integrin subunits β1 (ITGB1) and αν (ITGAV), both of which were upregulated in our results. Integrins can also mediate the force-dependent release of TGF-β1 from the ECM which may be another pro-survival mechanism [34]. A Tenon’s ocular fibroblast cultured in vitro in a regular tissue culture flask will likely not experience the same mechanical forces compared to its normal in vivo environment so it could be that apoptosis is affected by the in vitro culturing conditions in our experiment. Although there is a phenomenon called biomechanical memory lasting for several weeks [105] which may also play a role here. Apoptosis is highly associated with the pathogenesis of glaucoma, primarily via cell death of retinal ganglion cells but also in other ocular tissues such as the trabecular meshwork [106]. Several studies have reported changes in apoptosis in ocular tissues due to BAC treatment [10, 92, 107, 108] indicating that BAC treatment may also contribute to the observed changes in apoptotic genes in glaucomatous Tenon’s ocular fibroblasts when compared to normal fibroblasts. One study compared apoptosis in untreated versus treated POAG patients and demonstrated increased apoptosis in the trabecular meshwork and iris from treated patients suggesting that the treatment itself may increase apoptosis [109].
Overall glaucomatous TF cells show altered expression of genes associated with apoptosis which may be due to a combination of cytokines and ROS originating from POAG itself and topical glaucoma drug ingredients. Glaucomatous TF cells may resist apoptosis by both upregulation of pro-survival genes such as BCL2L2 and by means of autocrine TGF-β1 signalling similar to other myofibroblasts.
Chronic low-level inflammation plays a key role in glaucoma [110–112] and also in the pathogenesis of fibrotic disorders [113, 114]. Our results demonstrated changes in inflammatory state when comparing glaucomatous to normal TF cells. In keeping with our findings expression of the pro-inflammatory cytokine IL6 (IL-6) has been shown to be increased in retinal tissue, trabecular meshwork and aqueous humour of glaucoma patients when compared to normal controls [111] and expression of this cytokine is also increased in several fibrotic conditions [115, 116]. Furthermore, increased preoperative levels of IL6 in the aqueous humour are correlated with poor glaucoma surgery outcome [117]. IL6 upregulation may be explained by activation of the NF-κB pathway via toll-like receptors [118]. With respect to the TLR pathway our results showed upregulation of TLR4 which has also been shown to play a role in fibroblast activation [119] and in glaucoma [111, 120, 121]. Another important inflammatory pathway is the TNF pathway which can also lead to NF-κB activation [122]. Our results indicated upregulation of TNF-α receptor TNFRSF1A and furthermore it has been reported that the concentration of TNF-α in the aqueous humor of glaucoma patients is elevated [123] so it is conceivable that this pathway was activated in glaucomatous TF cells.
Overall, we observed increased fibroblast to myofibroblast transdifferentiation in glaucomatous TFs when compared to normal TFs. This transdifferentiation may have been induced by POAG-associated ROS, cytokines and growth factors such as TGF-β2 originating from the aqueous humour and may be further exacerbated by ingredients of topical glaucoma drugs such as BAC and polyquat.
It is not clear from the results if POAG pathology itself or topical glaucoma drug ingredients are the dominant factor causing transdifferentiation in glaucomatous TFs, most likely it is a combination of both. The resulting myofibroblast phenotype shows increased ECM synthesis and remodelling, altered autophagy, increased mitochondrial fission, changes in the antioxidant system, cell cycle arrest, inflammation, upregulated unfolded protein response and altered apoptotic pathways. Furthermore, the myofibroblast phenotype seems to remain in its fibrotic state via a positive feedback loop due to autocrine TGF-β1 and IL6 signalling in addition to upregulation of transcription factors such as ZEB1 and ZEB2.
Our results indicate that glaucomatous TFs are already in a fibrotic myofibroblast state before glaucoma filtration surgery which may compromise a favourable outcome of this procedure [15]. This may also add to the variability of the surgical outcome because the preoperative Tenon’s ocular fibroblast state may likely range from fairly normal to fibrotic. This has implications for clinical practice in that surgical outcomes are likely to be impacted by the duration of the glaucomatous state, topical medication (number, duration and preserved or unpreserved) and epigenetic and genetic factors. Earlier surgical interventions with glaucoma filtration surgery or minimally invasive glaucoma surgery (MIGS) devices may result in improved outcomes but further research is required in this area [124–126]. Concomitant tissue assessment at the time of surgery may also predict the degree of fibrotic activation and epigenetic/genetic risks of fibrosis to direct more personalised therapeutic interventions [127].
Several of the identified pathways and associated genes in this work could be utilised to find suitable drug targets to reduce both preoperative and postoperative fibrosis and indeed several genes are already being investigated to improve the outcome of glaucoma filtration surgery such as the TGF-β pathway, VEGFs, PDGF, MMPs, SPARC and p53 [12, 128, 129]. Furthermore, it may also be attempted to reverse myofibroblast transdifferentiation prior to surgery e.g. by metformin treatment [130, 131]. In addition, given the fact that glaucomatous TFs are already in a profibrotic myofibroblastic state is clinically relevant to animal studies of glaucoma filtration surgery investigating novel therapeutic approaches. In these animal models the Tenon’s ocular fibroblasts have not been exposed to the profibrotic environment of the glaucomatous state. This may be one explanation why promising therapeutic candidates from animal studies fail in clinical trials in glaucoma patients [132]. The use of Tenon’s ocular fibroblast/bulbar conjunctiva interface biomimetic models seeded with GTFs may provide a better model system to investigate new anti-fibrotic therapies [133].
Conclusions
To our knowledge this is the first study reporting a genome-wide transcriptome comparing glaucomatous to non-glaucomatous Tenon’s ocular fibroblasts. Our results indicate that glaucomatous Tenon’s ocular fibroblasts are transdifferentiated into myofibroblasts with associated changes in ECM remodelling, autophagy, mitochondrial morphology, antioxidant system, cell cycle, inflammation, unfolded protein response and apoptosis. These results may provide insight into wound-healing and fibrosis associated with glaucoma filtration surgery and support the development of novel therapeutic interventions to improve the outcome of glaucoma filtration surgery.
Supporting information
S1 Fig. Principal component analysis of RNA-Seq result.
Two-dimensional Principal Component Analysis (PCA) plot showing grouping of normal samples C1..C6 in blue and glaucomatous samples C7..C12 in orange. All glaucomatous samples cluster together while the set of normal samples shows more spread, especially samples C3 and C5, but is still well separated from the glaucomatous set of samples.
https://doi.org/10.1371/journal.pone.0307227.s001
(TIF)
S2 Fig. Extracellular matrix gene-enrichment heatmap.
Heatmap showing genes (FDR<0.05) associated with key enrichment terms related to remodelling of the extracellular matrix. Heatmap colours correspond to fold change expressed as log2FC which has been limited to a range of [–4,4] for clarity.
https://doi.org/10.1371/journal.pone.0307227.s002
(TIF)
S3 Fig. Autophagy gene-enrichment heatmap.
Heatmap showing genes (FDR<0.05) associated with key enrichment terms related to autophagy. Heatmap colours correspond to fold change ex-pressed as log2FC which has been limited to a range of [–4,4] for clarity.
https://doi.org/10.1371/journal.pone.0307227.s003
(TIF)
S4 Fig. Unfolded protein response gene-enrichment heatmap.
Heatmap showing genes (FDR<0.05) associated with key enrichment terms related to the unfolded protein response. Heatmap colours correspond to fold change ex-pressed as log2FC which has been limited to a range of [–4,4] for clarity.
https://doi.org/10.1371/journal.pone.0307227.s004
(TIF)
S1 Table. Differential expression of genes.
This table contains fold-change (log2FC), false discovery rate (FDR), p-value and counts per million mapped reads (CPM) for all genes expressed above a threshold of 100 reads averaged over samples.
https://doi.org/10.1371/journal.pone.0307227.s005
(XLSX)
S2 Table. Enrichment results.
This table contains all enrichment results using a list of differentially expressed genes (FDR<0.05) as input for databases WikiPathways, KEGG, Reactome, MSigDB Hallmark genesets, Gene Ontology (GO) biological processes and molecular functions. Each entry contains a metric of significance, a genelist relevant for the enrichment result and a cluster to which the enrichment term has been assigned, if applicable.
https://doi.org/10.1371/journal.pone.0307227.s006
(XLSX)
S3 Table. Donor information.
Table containing donor information for the 12 Tenon’s capsule tissue samples.
https://doi.org/10.1371/journal.pone.0307227.s007
(DOCX)
References
- 1. Braunger BM, Fuchshofer R, Tamm ER. The aqueous humor outflow pathways in glaucoma: A unifying concept of disease mechanisms and causative treatment. European Journal of Pharmaceutics and Biopharmaceutics. Elsevier B.V.; 2015. pp. 173–181. pmid:25957840
- 2. Allison K, Patel D, Alabi O. Epidemiology of Glaucoma: The Past, Present, and Predictions for the Future. Cureus. 2020;12. pmid:33391921
- 3. Tham YC, Li X, Wong TY, Quigley HA, Aung T, Cheng CY. Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology. 2014;121: 2081–2090. pmid:24974815
- 4. Weinreb RN, Aung T, Medeiros FA. The pathophysiology and treatment of glaucoma: a review. JAMA. 2014;311: 1901–1911. pmid:24825645
- 5. Alqawlaq S, Flanagan JG, Sivak JM. All roads lead to glaucoma: Induced retinal injury cascades contribute to a common neurodegenerative outcome. Exp Eye Res. 2019;183: 88–97. pmid:30447198
- 6. Zhavoronkov A, Kanherkar RR, Izumchenko E, Teka M, Cantor C, Manaye K, et al. Pro-fibrotic pathway activation in trabecular meshwork and lamina cribrosa is the main driving force of glaucoma. Cell Cycle. 2016;15: 1643. pmid:27229292
- 7. Wallace DM O’Brien CJ. The role of lamina cribrosa cells in optic nerve head fibrosis in glaucoma. Exp Eye Res. 2016;142: 102–109.
- 8. Agarwal P, Daher AM, Agarwal R. Aqueous humor TGF-β2 levels in patients with open-angle glaucoma: A meta-analysis. Mol Vis. 2015;21: 612. Available: /pmc/articles/PMC4445076/
- 9. Hedengran A, Kolko M. The molecular aspect of anti-glaucomatous eye drops—are we harming our patients? Mol Aspects Med. 2023;93. pmid:37459821
- 10. Rasmussen CA, Kaufman PL, Kiland JA. Benzalkonium Chloride and Glaucoma. Journal of Ocular Pharmacology and Therapeutics. 2014;30: 163. pmid:24205938
- 11. Gedde SJ, Feuer WJ, Lim KS, Barton K, Goyal S, Ahmed II, et al. Treatment Outcomes in the Primary Tube Versus Trabeculectomy Study after 5 Years of Follow-up. Ophthalmology. 2022;129: 1344–1356. pmid:35835337
- 12. Shao CG, Sinha NR, Mohan RR, Webel AD. Novel Therapies for the Prevention of Fibrosis in Glaucoma Filtration Surgery. Biomedicines 2023, Vol 11, Page 657. 2023;11: 657. pmid:36979636
- 13. Kakizaki H, Takahashi Y, Nakano T, Asamoto K, Ikeda H, Ichinose A, et al. Anatomy of Tenons capsule. Clin Exp Ophthalmol. 2012;40: 611–616. pmid:22172019
- 14. Yamanaka O, Kitano-Izutani A, Tomoyose K, Reinach PS. Pathobiology of wound healing after glaucoma filtration surgery. BMC Ophthalmol. 2015;15: 19–27. pmid:26818010
- 15. Broadway DC, Chang LP. Trabeculectomy, risk factors for failure and the preoperative state of the conjunctiva. J Glaucoma. 2001;10: 237–249. pmid:11442190
- 16. Trelford CB, Denstedt JT, Armstrong JJ, Hutnik CML. The Pro-Fibrotic Behavior of Human Tenon’s Capsule Fibroblasts in Medically Treated Glaucoma Patients. Clin Ophthalmol. 2020;14: 1391. pmid:32546947
- 17. De Falco E, Scafetta G, Napoletano C, Puca R, Vingolo EM, Ragona G, et al. A standardized laboratory and surgical method for in vitro culture isolation and expansion of primary human Tenon’s fibroblasts. Cell Tissue Bank. 2013;14: 277–287. pmid:22820760
- 18. Vallabh NA, Armstrong J, Czanner G, McDonagh B, Choudhary A, Criddle DN, et al. Evidence of impaired mitochondrial cellular bioenergetics in ocular fibroblasts derived from glaucoma patients. Free Radic Biol Med. 2022;189: 102–110. pmid:35872337
- 19. Callaghan B, Vallabh NA, Willoughby CE. Deuterated polyunsaturated fatty acids provided protection against oxidative stress in ocular fibroblasts derived from glaucoma patients. Mech Ageing Dev. 2023;211. pmid:36716826
- 20.
Babraham Bioinformatics. Fastqc. 2019 [cited 6 Sep 2022]. Available: https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
- 21. Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J. 2011;17: 10.
- 22. Dobin A, Gingeras TR. Mapping RNA-seq Reads with STAR. Current protocols in bioinformatics / editoral board, Andreas D Baxevanis. [et al]. 2015;51: 11.14.1. pmid:26334920
- 23.
GENCODE. GRCh38.p13. 2020 [cited 6 Sep 2022]. Available: https://www.gencodegenes.org/human/
- 24. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2013;30: 923–930. pmid:24227677
- 25. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2009/11/11. 2010;26: 139–140. pmid:19910308
- 26. Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics. 2013;14: 128. pmid:23586463
- 27. Merico D, Isserlin R, Stueker O, Emili A, Bader GD. Enrichment map: a network-based method for gene-set enrichment visualization and interpretation. PLoS One. 2010;5: e13984–e13984. pmid:21085593
- 28. Kucera M, Isserlin R, Arkhangorodsky A, Bader GD. AutoAnnotate: A Cytoscape app for summarizing networks with semantic annotations. F1000Res. 2016;5: 1717. pmid:27830058
- 29. Luo W, Brouwer C. Pathview: an R/Bioconductor package for pathway-based data integration and visualization. Bioinformatics. 2013/06/04. 2013;29: 1830–1831. pmid:23740750
- 30. van Iersel MP, Kelder T, Pico AR, Hanspers K, Coort S, Conklin BR, et al. Presenting and exploring biological pathways with PathVisio. BMC Bioinformatics. 2008;9: 399. pmid:18817533
- 31. Van Caam A, Vonk M, Van Den Hoogen F, Van Lent P, Van Der Kraan P. Unraveling SSc Pathophysiology; The Myofibroblast. Front Immunol. 2018;9. pmid:30483246
- 32. Moore MW, Herzog EL. Regulation and Relevance of Myofibroblast Responses in Idiopathic Pulmonary Fibrosis. Curr Pathobiol Rep. 2013;1: 199. pmid:25705577
- 33. Song G, Pacher M, Balakrishnan A, Yuan Q, Tsay HC, Yang D, et al. Direct Reprogramming of Hepatic Myofibroblasts into Hepatocytes In Vivo Attenuates Liver Fibrosis. Cell Stem Cell. 2016;18: 797–808. pmid:26923201
- 34. Hinz B, Lagares D. Evasion of apoptosis by myofibroblasts: a hallmark of fibrotic diseases. Nat Rev Rheumatol. 2020;16: 11–31. pmid:31792399
- 35. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. J Clin Invest. 2009;119: 1420. pmid:19487818
- 36. O’Driscoll E, Hughes E, Irnaten M, Kuehn M, Wallace D, O’Brien C. Role of Epithelial-to-Mesenchymal Transition of Retinal Pigment Epithelial Cells in Glaucoma Cupping. J Clin Med. 2023;12. pmid:37048820
- 37. Irnaten M, O’Brien CJ. Calcium-Signalling in Human Glaucoma Lamina Cribrosa Myofibroblasts. Int J Mol Sci. 2023;24. pmid:36674805
- 38. Pattabiraman PP, Maddala R, Rao PV. Regulation of Plasticity and Fibrogenic Activity of Trabecular Meshwork Cells by Rho GTPase Signaling. J Cell Physiol. 2014;229: 927. pmid:24318513
- 39. Tamm ER, Fuchshofer R. What increases outflow resistance in primary open-angle glaucoma? Surv Ophthalmol. 2007;52 Suppl 2. pmid:17998032
- 40. Tripathi RC, Li J, Chan WFA, Tripathi BJ. Aqueous Humor in Glaucomatous Eyes Contains an Increased Level of TGF-β2. Exp Eye Res. 1994;59: 723–728. pmid:7698265
- 41. Subileau M, Vittet D. Lymphatics in Eye Fluid Homeostasis: Minor Contributors or Significant Actors? Biology (Basel). 2021;10. pmid:34201989
- 42. Hong S, Lee JB, Iizuka Y, Song YK, Seong GJ, Han SH. The role of focal adhesion kinase in the TGF-β-induced myofibroblast transdifferentiation of human Tenon’s fibroblasts. Korean J Ophthalmol. 2012;26: 45–48. pmid:22323885
- 43. Lin HL, Qin YJ, Zhang YL, Zhang YQ, Chen YL, Niu YY, et al. Epigallocatechin-3-gallate (EGCG) inhibits myofibroblast transformation of human Tenon’s fibroblasts. Exp Eye Res. 2020;197. pmid:32603658
- 44. D’Urso M, Kurniawan NA. Mechanical and Physical Regulation of Fibroblast–Myofibroblast Transition: From Cellular Mechanoresponse to Tissue Pathology. Front Bioeng Biotechnol. 2020;8: 609653. pmid:33425874
- 45. Klingberg F, Hinz B, White ES. The myofibroblast matrix: implications for tissue repair and fibrosis. J Pathol. 2013;229: 298. pmid:22996908
- 46. Yamashiro C, Tokuda K, Kobayashi Y, Higashijima F, Yoshimoto T, Ota M, et al. Benzalkonium chloride-induced myofibroblastic transdifferentiation of Tenon’s capsule fibroblasts is inhibited by coculture with corneal epithelial cells or by interleukin-10. Sci Rep. 2021;11: 16096. pmid:34373467
- 47. Richter K, Kietzmann T. Reactive oxygen species and fibrosis: further evidence of a significant liaison. Cell and Tissue Research 2016 365:3. 2016;365: 591–605. pmid:27345301
- 48. Bocchino M, Agnese S, Fagone E, Svegliati S, Grieco D, Vancheri C, et al. Reactive Oxygen Species Are Required for Maintenance and Differentiation of Primary Lung Fibroblasts in Idiopathic Pulmonary Fibrosis. PLoS One. 2010;5: e14003. pmid:21103368
- 49. Hemmann S, Graf J, Roderfeld M, Roeb E. Expression of MMPs and TIMPs in liver fibrosis—a systematic review with special emphasis on anti-fibrotic strategies. J Hepatol. 2007;46: 955–975. pmid:17383048
- 50. Määttä M, Tervahartiala T, Harju M, Airaksinen J, Autio-Harmainen H, Sorsa T. Matrix metalloproteinases and their tissue inhibitors in aqueous humor of patients with primary open-angle glaucoma, exfoliation syndrome, and exfoliation glaucoma. J Glaucoma. 2005;14: 64–69. pmid:15650607
- 51. Weinreb RN, Robinson MR, Dibas M, Stamer WD. Matrix Metalloproteinases and Glaucoma Treatment. J Ocul Pharmacol Ther. 2020;36: 208–228. pmid:32233938
- 52. Schlötzer-Schrehardt U, Lommatzsch J, Küchle M, Konstas AGP, Naumann GOH. Matrix Metalloproteinases and Their Inhibitors in Aqueous Humor of Patients with Pseudoexfoliation Syndrome/Glaucoma and Primary Open-Angle Glaucoma. Invest Ophthalmol Vis Sci. 2003;44: 1117–1125. pmid:12601038
- 53. Smyth A, Irnaten M, Powell SK, O’Brien CJ. Elevated Expression of ADAM 12 and 19 in Human Glaucoma Lamina Cribrosa Cells. Invest Ophthalmol Vis Sci. 2021;62: 2770. Available: https://iovs.arvojournals.org/article.aspx?articleid=2775759
- 54. Schmidt-Arras D, Rose-John S. Regulation of Fibrotic Processes in the Liver by ADAM Proteases. Cells 2019, Vol 8, Page 1226. 2019;8: 1226. pmid:31601007
- 55. Le Pabic H, Bonnier D, Wewer UM, Coutand A, Musso O, Baffet G, et al. ADAM12 in human liver cancers: TGF-β-regulated expression in stellate cells is associated with matrix remodeling. Hepatology. 2003;37: 1056–1066. pmid:12717386
- 56. Roodnat AW, Callaghan B, Doyle C, Henry M, Goljanek-Whysall K, Simpson DA, et al. Genome-Wide RNA Sequencing of Human Trabecular Meshwork Cells Treated with TGF-β1: Relevance to Pseudoexfoliation Glaucoma. Biomolecules. 2022;12: 1693. pmid:36421707
- 57. Zenkel M, Krysta A, Pasutto F, Juenemann A, Kruse FE, Schlötzer-Schrehardt U. Regulation of Lysyl Oxidase-like 1 (LOXL1) and Elastin-Related Genes by Pathogenic Factors Associated with Pseudoexfoliation Syndrome. Invest Ophthalmol Vis Sci. 2011;52: 8488–8495. pmid:21948647
- 58. Hill C, Wang Y. Autophagy in pulmonary fibrosis: friend or foe? Genes Dis. 2022;9: 1594–1607. pmid:36119644
- 59. Bernard M, Dieudé M, Yang B, Hamelin K, Underwood K, Hébert MJ. Autophagy fosters myofibroblast differentiation through MTORC2 activation and downstream upregulation of CTGF. Autophagy. 2014;10: 2193–2207. pmid:25495560
- 60. Migneault F, Hébert MJ. Autophagy, tissue repair, and fibrosis: a delicate balance. Matrix Biol. 2021;100–101: 182–196. pmid:33454422
- 61. Adornetto A, Parisi V, Morrone LA, Corasaniti MT, Bagetta G, Tonin P, et al. The Role of Autophagy in Glaucomatous Optic Neuropathy. Front Cell Dev Biol. 2020;8. pmid:32211404
- 62. Rao A, Sahay P, Chakraborty M, Prusty BK, Srinivasan S, Jhingan GD, et al. Switch to Autophagy the Key Mechanism for Trabecular Meshwork Death in Severe Glaucoma. Clin Ophthalmol. 2021;15: 3027. pmid:34285469
- 63. Dixon A, Shim MS, Nettesheim A, Coyne A, Su CC, Gong H, et al. Autophagy deficiency protects against ocular hypertension and neurodegeneration in experimental and spontaneous glaucoma mouse models. Cell Death & Disease 2023 14:8. 2023;14: 1–11. pmid:37620383
- 64. Nettesheim A, Dixon A, Sup Shim M, Coyne A, Walsh M, Liton PB. Autophagy in the Aging and Experimental Ocular Hypertensive Mouse Model. Invest Ophthalmol Vis Sci. 2020;61: 31–31. pmid:32797200
- 65. Want A, Gillespie SR, Wang Z, Gordon R, Iomini C, Ritch R, et al. Autophagy and Mitochondrial Dysfunction in Tenon Fibroblasts from Exfoliation Glaucoma Patients. PLoS One. 2016;11. pmid:27391778
- 66. Jeon KI, Kumar A, Wozniak KT, Nehrke K, Huxlin KR. Defining the Role of Mitochondrial Fission in Corneal Myofibroblast Differentiation. Invest Ophthalmol Vis Sci. 2022;63. pmid:35377925
- 67. Wang Y, Lu M, Xiong L, Fan J, Zhou Y, Li H, et al. Drp1-mediated mitochondrial fission promotes renal fibroblast activation and fibrogenesis. Cell Death & Disease 2020 11:1. 2020;11: 1–14. pmid:31949126
- 68. Guo T, Jiang CS, Yang SZ, Zhu Y, He C, Carter AB, et al. Mitochondrial fission and bioenergetics mediate human lung fibroblast durotaxis. JCI Insight. 2023;8. pmid:36422990
- 69. Ju WK, Kim KY, Lindsey JD, Angert M, Duong-Polk KX, Scott RT, et al. Intraocular pressure elevation induces mitochondrial fission and triggers OPA1 release in glaucomatous optic nerve. Invest Ophthalmol Vis Sci. 2008;49: 4903–4911. pmid:18469184
- 70. Zhang ZQ, Xie Z, Chen SY, Zhang X. Mitochondrial dysfunction in glaucomatous degeneration. Int J Ophthalmol. 2023;16: 811. pmid:37206187
- 71. Vallabh NA, Lane B, Simpson DA, Fuchs M, Choudhary A, Criddle D, et al. Evidence of somatic mitochondrial DNA mutations in primary open angle glaucoma. Invest Ophthalmol Vis Sci. 2019;60: 1613–1613.
- 72. Li X, Zhang W, Cao Q, Wang Z, Zhao M, Xu L, et al. Mitochondrial dysfunction in fibrotic diseases. Cell Death Discovery 2020 6:1. 2020;6: 1–14. pmid:32963808
- 73. Datta S, Baudouin C, Brignole-Baudouin F, Denoyer A, Cortopassi GA. The Eye Drop Preservative Benzalkonium Chloride Potently Induces Mitochondrial Dysfunction and Preferentially Affects LHON Mutant Cells. Invest Ophthalmol Vis Sci. 2017;58: 2406–2412. pmid:28444329
- 74. Rogov AG, Goleva TN, Sukhanova EI, Epremyan KK, Trendeleva TA, Ovchenkova AP, et al. Mitochondrial Dysfunctions May Be One of the Major Causative Factors Underlying Detrimental Effects of Benzalkonium Chloride. Oxid Med Cell Longev. 2020;2020. pmid:32104543
- 75. Cheresh P, Kim SJ, Tulasiram S, Kamp DW. Oxidative Stress and Pulmonary Fibrosis. Biochim Biophys Acta. 2013;1832: 1028. pmid:23219955
- 76. Sedlak L, Wojnar W, Zych M, Wyględowska-Promieńska D. Influence of timolol, benzalkonium-preserved timolol, and benzalkonium-preserved brimonidine on oxidative stress biomarkers in the tear film. Cutan Ocul Toxicol. 2020;39: 260–268. pmid:32594785
- 77. Tanito M, Kaidzu S, Takai Y, Ohira A. Status of systemic oxidative stresses in patients with primary open-angle glaucoma and pseudoexfoliation syndrome. PLoS One. 2012/11/26. 2012;7: e49680–e49680. pmid:23189153
- 78. Takayanagi Y, Takai Y, Kaidzu S, Tanito M. Evaluation of Redox Profiles of the Serum and Aqueous Humor in Patients with Primary Open-Angle Glaucoma and Exfoliation Glaucoma. Antioxidants (Basel). 2020;9: 1305. pmid:33352680
- 79. Naguib S, Backstrom JR, Gil M, Calkins DJ, Rex TS. Retinal oxidative stress activates the NRF2/ARE pathway: An early endogenous protective response to ocular hypertension. Redox Biol. 2021;42: 101883. pmid:33579667
- 80. Dodson M, Castro-Portuguez R, Zhang DD. NRF2 plays a critical role in mitigating lipid peroxidation and ferroptosis. Redox Biol. 2019;23: 101107. pmid:30692038
- 81. Wang M, Li J, Zheng Y. The Potential Role of Nuclear Factor Erythroid 2-Related Factor 2 (Nrf2) in Glaucoma: A Review. Med Sci Monit. 2020;26. pmid:31949124
- 82. Callaghan B, Lester K, Lane B, Fan X, Goljanek-Whysall K, Simpson DA, et al. Genome-wide transcriptome profiling of human trabecular meshwork cells treated with TGF-β2. Scientific Reports 2022 12:1. 2022;12: 1–12. pmid:35689009
- 83. Li X, Leng Y, Jiang Q, Wang Z, Luo P, Zhang C, et al. Eye Drops of Metformin Prevents Fibrosis After Glaucoma Filtration Surgery in Rats via Activating AMPK/Nrf2 Signaling Pathway. Front Pharmacol. 2020;11. pmid:32903813
- 84. Jobling MF, Mott JD, Finnegan MT, Jurukovski V, Erickson AC, Walian PJ, et al. Isoform-Specific Activation of Latent Transforming Growth Factor β (LTGF-β) by Reactive Oxygen Species. Radiat Res. 2006;166: 839–848. pmid:17149983
- 85. Liu RM, Desai LP. Reciprocal regulation of TGF-β and reactive oxygen species: A perverse cycle for fibrosis. Redox Biology. Elsevier B.V.; 2015. pp. 565–577. pmid:26496488
- 86. Dutto I, Tillhon M, Cazzalini O, Stivala LA, Prosperi E. Biology of the cell cycle inhibitor p21(CDKN1A): molecular mechanisms and relevance in chemical toxicology. Arch Toxicol. 2015;89: 155–178. pmid:25514883
- 87. Springelkamp H, Iglesias AI, Mishra A, Höhn R, Wojciechowski R, Khawaja AP, et al. New insights into the genetics of primary open-angle glaucoma based on meta-analyses of intraocular pressure and optic disc characteristics. Hum Mol Genet. 2017;26: 438–453. pmid:28073927
- 88. Korthagen NM, van Moorsel CHM, Barlo NP, Kazemier KM, Ruven HJT, Grutters JC. Association between variations in cell cycle genes and idiopathic pulmonary fibrosis. PLoS One. 2012;7. pmid:22291954
- 89. Kumari R, Jat P. Mechanisms of Cellular Senescence: Cell Cycle Arrest and Senescence Associated Secretory Phenotype. Front Cell Dev Biol. 2021;9. pmid:33855023
- 90. Danford ID, Verkuil LD, Choi DJ, Collins DW, Gudiseva H V., Uyhazi KE, et al. Characterizing the “POAGome”: A bioinformatics-driven approach to primary open-angle glaucoma. Prog Retin Eye Res. 2017;58: 89–114. pmid:28223208
- 91. Liton PB, Challa P, Stinnett S, Luna C, Epstein DL, Gonzalez P. Cellular senescence in the glaucomatous outflow pathway. Exp Gerontol. 2005;40: 745. pmid:16051457
- 92. Kim SH, Kwon D, Lee S, Son SW, Kwon JT, Kim PJ, et al. Concentration- and Time-Dependent Effects of Benzalkonium Chloride in Human Lung Epithelial Cells: Necrosis, Apoptosis, or Epithelial Mesenchymal Transition. Toxics. 2020;8. pmid:32121658
- 93. Lv X, Liu C, Liu S, Li Y, Wang W, Li K, et al. The cell cycle inhibitor P21 promotes the development of pulmonary fibrosis by suppressing lung alveolar regeneration. Acta Pharm Sin B. 2022;12: 735–746. pmid:35256943
- 94. McElhinney K, Irnaten M, O’Brien C. p53 and Myofibroblast Apoptosis in Organ Fibrosis. International Journal of Molecular Sciences 2023, Vol 24, Page 6737. 2023;24: 6737. pmid:37047710
- 95. Dynlacht BD, Moberg K, Lees JA, Harlow E, Zhu L. Specific regulation of E2F family members by cyclin-dependent kinases. Mol Cell Biol. 1997;17: 3867–3875. pmid:9199321
- 96. Akimoto M, Hangai M, Okazaki K, Kogishi J, Honda Y, Kaneda Y. Growth inhibition of cultured human Tenon’s fibroblastic cells by targeting the E2F transcription factor. Exp Eye Res. 1998;67: 395–401. pmid:9820786
- 97. Peters JC, Bhattacharya S, Clark AF, Zode GS. Increased Endoplasmic Reticulum Stress in Human Glaucomatous Trabecular Meshwork Cells and Tissues. Invest Ophthalmol Vis Sci. 2015;56: 3860–3868. pmid:26066753
- 98. Kropski JA, Blackwell TS. Endoplasmic reticulum stress in the pathogenesis of fibrotic disease. J Clin Invest. 2018/01/02. 2018;128: 64–73. pmid:29293089
- 99. Lenna S, Trojanowska M. The role of endoplasmic reticulum stress and the unfolded protein response in fibrosis. Curr Opin Rheumatol. 2012;24: 663–668. pmid:22918530
- 100. Yang Y, Liu L, Naik I, Braunstein Z, Zhong J, Ren B. Transcription Factor C/EBP Homologous Protein in Health and Diseases. Front Immunol. 2017;8: 1612. pmid:29230213
- 101. Kuehl T, Lagares D. BH3 mimetics as anti-fibrotic therapy: Unleashing the mitochondrial pathway of apoptosis in myofibroblasts. Matrix Biology. 2018;68–69: 94–105. pmid:29408011
- 102. Jun JB, Kuechle M, Min J, Seung CS, Kim G, Montenegro V, et al. Scleroderma fibroblasts demonstrate enhanced activation of Akt (protein kinase B) in situ. J Invest Dermatol. 2005;124: 298–303. pmid:15675946
- 103. Zhang HY, Phan SH. Inhibition of myofibroblast apoptosis by transforming growth factor beta(1). Am J Respir Cell Mol Biol. 1999;21: 658–665. pmid:10572062
- 104. Grinnell F, Zhu M, Carlson MA, Abrams JM. Release of mechanical tension triggers apoptosis of human fibroblasts in a model of regressing granulation tissue. Exp Cell Res. 1999;248: 608–619. pmid:10222153
- 105. Li CX, Talele NP, Boo S, Koehler A, Knee-Walden E, Balestrini JL, et al. MicroRNA-21 preserves the fibrotic mechanical memory of mesenchymal stem cells. Nat Mater. 2017;16: 379–389. pmid:27798620
- 106. Saccà SC, Izzotti A. Focus on molecular events in the anterior chamber leading to glaucoma. Cell Mol Life Sci. 2014;71: 2197–2218. pmid:24142347
- 107. Brignole-Baudouin F, Riancho L, Liang H, Nakib Z, Baudouin C. In vitro comparative toxicology of polyquad-preserved and benzalkonium chloride-preserved travoprost/timolol fixed combination and latanoprost/timolol fixed combination. J Ocul Pharmacol Ther. 2011;27: 273–280. pmid:21410334
- 108. Pisella PJ, Debbasch C, Hamard P, Creuzot-Garcher C, Rat P, Brignole F, et al. Conjunctival proinflammatory and proapoptotic effects of latanoprost and preserved and unpreserved timolol: an ex vivo and in vitro study. Invest Ophthalmol Vis Sci. 2004;45: 1360–1368. pmid:15111589
- 109. Aktas Z, Karaca EE, Gonul II, Hasanreisoglu M, Onol M. Apoptosis in the iris and trabecular meshwork of medically treated and untreated primary open angle glaucoma patients. Int J Ophthalmol. 2013;6: 827–830. pmid:24392332
- 110. Liesenborghs I, Eijssen LMT, Kutmon M, Gorgels TGMF, Evelo CT, Beckers HJM, et al. Comprehensive bioinformatics analysis of trabecular meshwork gene expression data to unravel the molecular pathogenesis of primary open-angle glaucoma. Acta Ophthalmol. 2020;98: 48–57. pmid:31197946
- 111. Baudouin C, Kolko M, Melik-Parsadaniantz S, Messmer EM. Inflammation in Glaucoma: From the back to the front of the eye, and beyond. Prog Retin Eye Res. 2021;83. pmid:33075485
- 112. Coyle S, Khan MN, Chemaly M, Callaghan B, Doyle C, Willoughby CE, et al. Targeting the NLRP3 Inflammasome in Glaucoma. Biomolecules. 2021;11. pmid:34439904
- 113. Wynn TA, Ramalingam TR. Mechanisms of fibrosis: therapeutic translation for fibrotic disease. Nature Medicine 2012 18:7. 2012;18: 1028–1040. pmid:22772564
- 114. Henderson NC, Rieder F, Wynn TA. Fibrosis: from mechanisms to medicines. Nature 2020 587:7835. 2020;587: 555–566. pmid:33239795
- 115. Li Y, Zhao J, Yin Y, Li K, Zhang C, Zheng Y. The Role of IL-6 in Fibrotic Diseases: Molecular and Cellular Mechanisms. Int J Biol Sci. 2022;18: 5405. pmid:36147459
- 116. Johnson BZ, Stevenson AW, Prêle CM, Fear MW, Wood FM. The Role of IL-6 in Skin Fibrosis and Cutaneous Wound Healing. Biomedicines. 2020;8. pmid:32365896
- 117. Cvenkel B, Kopitar AN, Ihan A. Inflammatory molecules in aqueous humour and on ocular surface and glaucoma surgery outcome. Mediators Inflamm. 2010;2010. pmid:20467456
- 118. Nyati KK, Masuda K, Mahabub-Uz Zaman M, Dubey PK, Millrine D, Chalise JP, et al. TLR4-induced NF-κB and MAPK signaling regulate the IL-6 mRNA stabilizing protein Arid5a. Nucleic Acids Res. 2017;45: 2687. pmid:28168301
- 119. Bhattacharyya S, Wang W, Qin W, Cheng K, Coulup S, Chavez S, et al. TLR4-dependent fibroblast activation drives persistent organ fibrosis in skin and lung. JCI Insight. 2018;3. pmid:29997297
- 120. Mzyk P, Hernandez H, Le T, Ramirez JR, McDowell CM. Toll-Like Receptor 4 Signaling in the Trabecular Meshwork. Front Cell Dev Biol. 2022;10. pmid:35912101
- 121. Shibuya E, Meguro A, Ota M, Kashiwagi K, Mabuchi F, Iijima H, et al. Association of Toll-like Receptor 4 Gene Polymorphisms with Normal Tension Glaucoma. Invest Ophthalmol Vis Sci. 2008;49: 4453–4457. pmid:18586872
- 122. Holbrook J, Lara-Reyna S, Jarosz-Griffiths H, McDermott M. Tumour necrosis factor signalling in health and disease. F1000Res. 2019;8. pmid:30755793
- 123. Balaiya S, Edwards J, Tillis T, Khetpal V, Chalam K V. Tumor necrosis factor-alpha (TNF-α) levels in aqueous humor of primary open angle glaucoma. Clin Ophthalmol. 2011;5: 553–556. pmid:21607023
- 124. Bicket AK, Le JT, Azuara-Blanco A, Gazzard G, Wormald R, Bunce C, et al. Minimally Invasive Glaucoma Surgical Techniques for Open-Angle Glaucoma: An Overview of Cochrane Systematic Reviews and Network Meta-analysis. JAMA Ophthalmol. 2021;139: 983–989. pmid:34264292
- 125. King AJ, Fernie G, Hudson J, Kernohan A, Azuara-Blanco A, Burr J, et al. Primary trabeculectomy versus primary glaucoma eye drops for newly diagnosed advanced glaucoma: TAGS RCT. Health Technol Assess. 2021;25: i–158. pmid:34854808
- 126. Burr J, Azuara-Blanco A, Avenell A, Tuulonen A. Medical versus surgical interventions for open angle glaucoma. Cochrane Database Syst Rev. 2012 [cited 17 Feb 2024]. pmid:22972069
- 127. Yu-Wai-Man C, Khaw PT. Personalized Medicine in Ocular Fibrosis: Myth or Future Biomarkers. Adv Wound Care (New Rochelle). 2016;5: 390–402. pmid:27679750
- 128. Yu-Wai-Man C, Khaw PT. Developing novel anti-fibrotic therapeutics to modulate post-surgical wound healing in glaucoma: big potential for small molecules. Expert Rev Ophthalmol. 2015;10: 65–76. pmid:25983855
- 129. Khaw PT, Bouremel Y, Brocchini S, Henein C. The control of conjunctival fibrosis as a paradigm for the prevention of ocular fibrosis-related blindness. “Fibrosis has many friends.” Eye (Lond). 2020;34: 2163–2174. pmid:32587389
- 130. Hurley DJ, Irnaten M, O’brien C. Metformin and Glaucoma—Review of Anti-Fibrotic Processes and Bioenergetics. Cells. 2021;10. pmid:34440899
- 131. Kheirollahi V, Wasnick RM, Biasin V, Vazquez-Armendariz AI, Chu X, Moiseenko A, et al. Metformin induces lipogenic differentiation in myofibroblasts to reverse lung fibrosis. Nat Commun. 2019;10. pmid:31278260
- 132. Khaw P, Grehn F, Holló G, Overton B, Wilson R, Vogel R, et al. A phase III study of subconjunctival human anti-transforming growth factor beta(2) monoclonal antibody (CAT-152) to prevent scarring after first-time trabeculectomy. Ophthalmology. 2007;114. pmid:17908591
- 133. Kozdon K, Caridi B, Duru I, Ezra DG, Phillips JB, Bailly M. A Tenon’s capsule/bulbar conjunctiva interface biomimetic to model fibrosis and local drug delivery. PLoS One. 2020;15. pmid:33141875