Free
Genetics  |   February 2014
Aberrant Expression of Long Noncoding RNAs in Early Diabetic Retinopathy
Author Notes
  • Eye Hospital, Nanjing Medical University, Nanjing, China 
  • Correspondence: Biao Yan, Eye Hospital, Nanjing Medical University, 138# Han-Zhong Road, Nanjing, China 210029; yanbiao1982@hotmail.com
  • Qin Jiang, Eye Hospital, Nanjing Medical University, 138# Han-Zhong Road, Nanjing, China 210029; jqin710@vip.sina.com. Jin Yao, Eye Hospital, Nanjing Medical University, 138# Han-Zhong Road, Nanjing, China 210029; dryaojin@vip.sina.com
Investigative Ophthalmology & Visual Science February 2014, Vol.55, 941-951. doi:10.1167/iovs.13-13221
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Biao Yan, Zhi-Fu Tao, Xiu-Miao Li, Hui Zhang, Jin Yao, Qin Jiang; Aberrant Expression of Long Noncoding RNAs in Early Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2014;55(2):941-951. doi: 10.1167/iovs.13-13221.

      Download citation file:


      © 2016 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Purpose.: Long noncoding RNAs (lncRNAs) are broadly classified as transcripts longer than 200 nucleotides. lncRNA-mediated biology has been implicated in a variety of cellular processes and human diseases. Diabetic retinopathy (DR) is one of the leading causes of blindness. However, little is known about the role of lncRNAs in DR The goal of this study aimed to identify lncRNAs involved in early DR and characterize their roles in DR pathogenesis.

Methods.: We established a mouse model of streptozotocin (STZ)-induced diabetes, and performed lncRNA expression profiling of retinas using microarray analysis. Based on the Pearson correlation analysis, an lncRNA/mRNA coexpression network was constructed. Gene ontology (GO) enrichment and KEGG analysis of lncRNAs–coexpressed mRNAs was conducted to identify the related biological modules and pathologic pathways. Real-time PCR was conducted to detect the expression pattern of lncRNA in the clinical samples and the RF/6A cell model of hyperglycemia.

Results.: Approximately 303 lncRNAs were aberrantly expressed in the retinas of early DR, including 214 downregulated lncRNAs and 89 upregulated lncRNAs. GO analysis indicated that these lncRNAs–coexpressed mRNAs were targeted to eye development process (ontology: biological process), integral to membrane (ontology: cellular component), and structural molecule activity (ontology: molecular function). Pathway analysis indicated that lncRNAs–coexpressed mRNAs were mostly enriched in axon guidance signaling pathway. In addition, MALAT1, a conserved lncRNA, was significantly upregulated in an RF/6A cell model of hyperglycemia, in the aqueous humor samples, and in fibrovascular membranes of diabetic patients.

Conclusions.: lncRNAs are involved in the pathogenesis of DR through the modulation of multiple pathogenetic pathways. MALAT1, a conserved lncRNA, may become a potential therapeutic target for the prognosis, diagnosis, and treatment of DR.

Introduction
Diabetic retinopathy (DR) is a severe complication of diabetes. 1 Approximately one-third of diabetic patients have signs of DR and approximately one-tenth of patients have the vision-threatening phases of retinopathy, such as diabetic macular edema and proliferative retinopathy. 2 The pathogenesis of DR is multifactorial. Risk factors, such as poor glycemic control, longer diabetes duration, hypertension, hyperlipidemia, and albuminuria have been implicated in the initiation and progression of DR. 3,4 To date, several candidate genes, including aldose reductase, VEGF, receptor for advanced glycation end products gene, angiotensin I converting enzyme, methylenetetrahydrofolate reductase, glucose transporter, plasminogen activator inhibitor 1, α2β1 integrin, and apolipoprotein E, are found to be associated with DR susceptibility, suggesting the role of genetic factors in shaping the susceptibility to DR. 5,6  
The mammalian genome is transcribed in a complex manner, including the production of thousands of long noncoding RNAs (lncRNAs). 7 lncRNAs are defined as the transcripts of more than 200 nucleotides that structurally resemble mRNAs but do not encode proteins. lncRNAs participate in a variety of biological processes, such as chromosome imprinting, epigenetic regulation, cell-cycle control, transcription, translation, splicing, and cell differentiation. 7 Misregulation of lncRNAs is associated with the susceptibility to several human diseases, including cancers, cardiovascular diseases, and neurological diseases. 8 Genome-wide association studies (GWAS) have revealed that only 7% of diseases or trait-associated single-nucleotide polymorphisms (SNPs) reside in the protein-coding regions, whereas 43% of trait-/disease-associated SNPs are found outside of protein-coding genes, 9 suggesting that noncoding RNA alteration could affect the genetic susceptibility to DR. Thus, identifying DR-related lncRNAs contributes to better understanding the complex molecular mechanisms of DR pathogenesis. 
To reveal the potential role of lncRNAs in DR, we performed lncRNA expression profiling and compared lncRNA expression differences between diabetic retinas and nondiabetic retinas using microarray analysis. The result showed that 303 lncRNAs were aberrantly expressed in diabetic retinas. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis suggested that these differentially expressed lncRNAs may be involved in DR pathogenesis through modulating multiple pathologic signaling pathways. Metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), a highly conserved lncRNA, was abnormally expressed in an RF/6A cell model of hyperglycemia, in the fibrovascular membranes (FVMs), and in the aqueous humor samples of diabetic patients, suggesting its potential application as a biomarker of the prognosis and diagnosis of DR. To our knowledge, this is the first direct, in-depth investigation on lncRNA expression profiling of DR, providing a novel insight into the molecular mechanisms of DR pathogenesis. 
Materials and Methods
Diabetic Mice Model
All experimental animals were handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and approved by the Animal Care and Use Committee of Nanjing Medical University. Diabetes was induced chemically in 8-week-old C57BL/6 mice. Mice received an intraperitoneal injection of 50 mg/kg streptozotocin (STZ) dissolved in sodium citrate buffer (0.01 M, pH 4.5) on 5 successive days. Control animals received an injection of equal volume of citrate buffer. Blood glucose levels were measured immediately before STZ injection, or 2 days, 1 week, 1 month, or 2 months after STZ injection. Animals with blood glucose levels higher than 250 mg/dL were deemed as having diabetes. 10  
Electroretinogram
Before being euthanized for morphological and biochemical analyses, the experimental mice were subjected to analysis by electroretinogram (ERG) to evaluate the change in retinal electrical activity. Briefly, mice were dark-adapted overnight. Pupils were fully dilated using 1% tropicamide solution (Alcon, Fort Worth, TX). ERG responses were recorded from both eyes using the platinum wire corneal electrodes, forehead reference electrode, and ground electrode in the tail. ERG waveforms were recorded with a bandwidth of 0.3 to 500 Hz and samples at 2 kHz by a digital acquisition system. Statistics were shown as mean ± SEM amplitudes of A-, B-, or oscillatory potential (OP) wave of each treatment group. 11  
Microarray Profiling
Total RNAs were isolated from the retinas of diabetic mice 2 months after STZ injection (n = 9) or age-matched and sex-matched wild-type mice (n = 9) using TRIzol reagent (Invitrogen, Carlsbad, CA), respectively. Three individuals were pooled together as a biological repeat to further eliminate the individual difference. Microarray profiling was performed using Agilent Mouse Gene Expression Microarrays (Product Number G4852A; Agilent Technologies, Santa Clara, CA), including 39,430 Entrez Gene RNAs and 16,251 long intergenic noncoding RNAs (lincRNAs). Briefly, 10 μg of total RNAs were labeled using the Superscript Plus Direct cDNA labeling system (Invitrogen), and then hybridized to the chip. Microarrays were scanned using an Agilent scanner, and microarray data were extracted using Agilent Feature Extraction software FE10.5. The data quality was assessed using GeneSpring GX software (Agilent Technologies). 
Real-time PCR
Total RNAs were extracted using TRIzol reagent (Invitrogen) and then reversely transcribed using PrimeScript RT reagent Kit (TaKaRa, Dalian, China). Real-time PCR was performed using the ABI Prism 7300 sequence detection system (Applied Biosystems, Foster City, CA). The reaction mixture (20 μL) contained 10 ng cDNA template, 200 nM each of sense and antisense primers, and 10 μL 2 × SYBR-Green PCR Mix (TaKaRa). Real-time PCR was performed in duplicate for each sample, and the specificity of PCR product was estimated using the dissociation curve. β-actin was detected as the internal control. 
Bioinformatics Analysis
To investigate the potential role of the lncRNAs–coexpressed mRNAs, these mRNAs were input into the Database for Annotation, Visualization, and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov) for annotation and functional analysis, including gene set enrichment analysis and mapping gene sets to the KEGG pathway. 12 The TRANSFAC database was used to predict the transcription factor binding sites (TFBS) in the sequences of MALAT1 (http://www.gene-regulation.com). 13 The catRAPID algorithm was used to predict the potential interacting proteins of MALAT1 (http://service.tartaglialab.com). 14  
lncRNA/mRNA Coexpression Network
To construct the lncRNA/mRNA coexpression network, we calculated the Pearson correlation coefficient and R value to evaluate lncRNA-mRNA correlation. 15 The network construction procedures included the following: (1) preprocess data: the same mRNAs with different transcripts taking the median value represent the gene expression values, without special treatment of lncRNA expression value; (2) screen data: remove the subset of data according to the lists showing the differential expression of lncRNA and mRNA; (3) calculate the Pearson correlation coefficient and use R value to calculate the correlation coefficient between lncRNAs and mRNAs; and (4) screen by Pearson correlation coefficient: select the Pearson correlation coefficient greater than 0.99 as the meaningful value and draw the lncRNA/mRNA coexpression network by using the cytoscape program. 
Clinical Sample Collection
The clinical study was approved by the ethics committees of Nanjing Medical University. The surgical specimens were handled in accordance with the Declaration of Helsinki. All patients gave informed consent before inclusion in this study. The FVMs were obtained from the patients who consecutively underwent pars plana vitrectomy as treatment of proliferative DR (PDR) caused by diabetes mellitus type 2 (study group) or who underwent pars plana vitrectomy as treatment of idiopathic macular holes or preretinal membranes (control group). Samples of aqueous humor were harvested from the eyes of participating patients who had PDR or nondiabetic ocular diseases. None of the patients with nondiabetic ocular diseases had diabetes mellitus. 
Statistical Analysis
Data were presented as the mean ± SEM unless otherwise stated. Comparison between two groups was analyzed by using the two-tailed Student's t-test or two-way ANOVA. Statistical significance was defined as P less than 0.05. 
Results
Induction of Diabetes in C57Bl/6J Mice
Mice received intraperitoneal administration of STZ to induce diabetes. The control group received an injection of equal volume of citrate buffer. STZ treatment resulted in hyperglycemia and a progressive loss of body weight in STZ-treated mice (Table 1). Moreover, we compared ERG signaling in the retinas of diabetic and nondiabetic mice after 2 months of STZ injection. A-wave response was found to be virtually identical between diabetic and nondiabetic retinas (Fig. 1A). At all light intensities evaluated, the retinas of diabetic mice showed reduced B-wave and oscillatory potential amplitudes (Figs. 1B, 1C). These findings were similar to previous reports, indicating that deleterious change in ERG signals occurred in the diabetic animal model, beginning at 6 weeks after STZ injection or even earlier. 1,11 Taken together, these results suggested that STZ treatment resulted in an obvious change in retinal function in the retina of mice at the early stage of diabetes. 
Figure 1
 
ERG levels in the retinas of diabetic and nondiabetic mice. Mean ERG amplitudes for the retinas of diabetic and nondiabetic mice (n = 10 for each group) are shown. The top (A) shows the A-wave. The middle (B) represents the B-wave. The bottom (C) shows the amplitudes for the OPs.
Figure 1
 
ERG levels in the retinas of diabetic and nondiabetic mice. Mean ERG amplitudes for the retinas of diabetic and nondiabetic mice (n = 10 for each group) are shown. The top (A) shows the A-wave. The middle (B) represents the B-wave. The bottom (C) shows the amplitudes for the OPs.
Table 1
 
General Physiological Parameters in Diabetic and Nondiabetic Mice
Table 1
 
General Physiological Parameters in Diabetic and Nondiabetic Mice
Nondiabetic, n = 10 Diabetic, n = 10
2 wk after diabetic
 Body weight, g 28.1 ± 4.5 27.8 ± 3.1
 Glucose, mg/dL 115 ± 6 278 ± 36*
4 wk after diabetic
 Body weight, g 31.5 ± 5.2 29.7 ± 4.2*
 Glucose, mg/dL 110 ± 8 297 ± 44*
8 wk after diabetic
 Body weight, g 37.6 ± 5.8 31.7 ± 2.9*
 Glucose, mg/dL 105 ± 4 315 ± 39*
Overview of lncRNa Microarray Analysis
To reveal the potential role of lncRNAs in early DR, we performed a microarray analysis of the retinal tissues from STZ-induced diabetic mice and age- and sex-matched controls after 2 months of diabetes. The OD260:OD280 ratios of total RNAs were approximately 2.1, and the OD260:OD230 ratios of total RNAs were more than 1.9, suggesting that these RNAs were sufficiently pure for microarray analysis. Subsequently, these mRNAs were reversed into cDNA transcripts, and microarray hybridization was performed using Agilent Mouse Gene Expression Microarrays (product number G4852A). The box plot provided an overview of lncRNA microarray data, which displayed the differences between samples without making any assumptions of the underlying statistical distribution. After normalization, the distribution of log 2 ratios between the nondiabetic and diabetic groups is shown in Figure 2A. Scatter plot provided an overall indication of sample similarity between individual transcripts. As shown in Figure 2B, the biological replicates exhibited similar transcript levels (nondiabetic versus nondiabetic, diabetic versus diabetic), whereas there was a significant lncRNA expression difference between nondiabetic and diabetic groups. To gain a systematic comparison of lncRNA expression between diabetic and nondiabetic retinas, we also used hierarchical clustering analysis to arrange samples into groups based on their expression levels. The diabetic samples were clustered together on the same branch, whereas the nondiabetic samples were clustered on the other branch (Fig. 2C). Taken together, these results suggested that this microarray analysis was completed with high quality. From an overall perspective, there was a significant difference of lncRNA expression between nondiabetic and diabetic groups. 
Figure 2
 
Overview of lncRNAs microarray analysis. (A) The box plot displays the distributions of lncRNA expression profiling. After normalization, the distributions of log 2 ratios among different samples are shown. The box plots consist of boxes with a central line and two tails. The central line represents the median of the data, whereas the tails represent the upper and lower quartiles. (B) The scatter plot is used to compare the expression variations between nondiabetic and diabetic retinas. (C) Heatmaps were generated from the hierarchical cluster analysis to show the differential expressed lncRNAs between nondiabetic and diabetic retinas. The color scale on the top illustrates the relative expression level of lncRNAs across all samples: red denotes expression greater than 0, and green denotes expression less than 0.
Figure 2
 
Overview of lncRNAs microarray analysis. (A) The box plot displays the distributions of lncRNA expression profiling. After normalization, the distributions of log 2 ratios among different samples are shown. The box plots consist of boxes with a central line and two tails. The central line represents the median of the data, whereas the tails represent the upper and lower quartiles. (B) The scatter plot is used to compare the expression variations between nondiabetic and diabetic retinas. (C) Heatmaps were generated from the hierarchical cluster analysis to show the differential expressed lncRNAs between nondiabetic and diabetic retinas. The color scale on the top illustrates the relative expression level of lncRNAs across all samples: red denotes expression greater than 0, and green denotes expression less than 0.
Differential Expressed lncRNAs Between Nondiabetic and Diabetic Retinas
The microarray data were filtered by using the volcano plot to illustrate the differentially expressed lncRNAs between nondiabetic and diabetic retinas (Fig. 3A). We set a threshold as fold-change greater than 2.0, and identified 303 differentially expressed lncRNAs, including 214 downregulated lncRNAs and 89 upregulated lncRNAs (diabetic versus nondiabetic; Supplementary Table S1). To verify the results of microarray data, we performed real-time PCR assays to detect the expression differences of the top 10 upregulated and 10 downregulated lncRNAs between diabetic and nondiabetic retinas. We found that 9 of the 10 upregulated lncRNAs were verified to be significantly increased in diabetic retinas, whereas 8 of the 10 downregulated lncRNAs were verified to be decreased in diabetic retinas (Table 2). 
Figure 3
 
Differential expressed lncRNAs between nondiabetic and diabetic retinas. (A) The volcano plot illustrates the differential expressed lncRNAs between the diabetic and nondiabetic groups. (B) Chromosome distribution of differentially expressed lncRNAs based on microarray data is shown.
Figure 3
 
Differential expressed lncRNAs between nondiabetic and diabetic retinas. (A) The volcano plot illustrates the differential expressed lncRNAs between the diabetic and nondiabetic groups. (B) Chromosome distribution of differentially expressed lncRNAs based on microarray data is shown.
Table 2
 
Differentially Expressed lncRNAs in the Retinas of Diabetic Mice Compared With That of Nondiabetic Mice
Table 2
 
Differentially Expressed lncRNAs in the Retinas of Diabetic Mice Compared With That of Nondiabetic Mice
lncRNAs Fold Change, DR/NDR, Microarray Fold Change, DR/NDR, qRT-PCR
Downregulated
 chr4:35106125-35127700  reverse strand 13.62 10.55
 chr15:101088974-101089576  reverse strand 11.95 8.63
 chr6:126766225-126771309  reverse strand 10.85 9.44
 chr19:53624826-53625580  reverse strand 10.64 5.66
 chr13:98062189-98066613  forward strand 10.34 10.21
 chr1:45569500-45580900  reverse strand 10.18 9.86
 chr10:78969674-78981039  forward strand 10.09 6.94
 chr2:33661335-33669111 f  orward strand 9.89 7.34
 chr17:21785442-21801542  reverse strand 9.62 6.08
Upregulated
 chr10:120336296-120354471  forward strand 26.38 15.34
 chr19:23068397-23190256  forward strand 13.10 11.26
 chr8:59994102-59994797  forward strand 10.74 9.52
 chr14:44151048-44156924  reverse strand 10.68 8.77
 chr14:65878005-65878630  forward strand 8.71 7.43
 chr8:59994102-59994797  forward strand 7.42 8.10
 chr9:27148989-27155714  reverse strand 5.01 2.22
 chr18:75513062-75523889  forward strand 4.81 4.33
We found that these differentially expressed lncRNAs showed different lengths ranging from 217 bp to 33.5 kb and were transcribed from the sense and antisense directions. Further, these differentially expressed lncRNAs were distributed in nearly all of the mouse chromosomes (Fig. 3B). In general, a number of lncRNAs showed “clusters of transcription” with multiple transcripts originating from relatively short segments of the genome. 16 On mouse chromosome 1, two lncRNA clusters were found, which were respectively transcribed from a 138,499-bp (2437 bp, 10,001 bp, and 11,951 bp) or a 182,685-kb chromosomal region (3345 bp, 22,601 bp [forward (F)], and 22,601 bp [reverse (R)]); on mouse chromosome 2, three lncRNAs (18,426 bp, 634 bp, 4223 bp, and 2456 bp) were transcribed from a 176,134-kb chromosomal region; on mouse chromosome 3, three lncRNAs (580 bp, 372 bp, and 10,551 bp) were transcribed from a 846,09-kb chromosomal region; on mouse chromosome 7, five lncRNAs (7276 bp, 15,076 bp, 1708 bp, 60,751 bp, and 7166 bp) were transcribed from a 134,369-kb chromosomal region; and on mouse chromosome 8, three lncRNAs (5876 bp, 6202 bp [F], and 6202 bp [R]) were transcribed from a 108,153-kb chromosomal region. 
Construction of lncRNA/mRNA Coexpression Network
Although accumulating studies have attempted to reveal the functional significance of lncRNAs, the biological roles of most lncRNAs are still unknown. Biological processes and cellular regulation networks are very complex, involving the interactions of various molecules, such as proteins, RNAs, and DNAs. 17 The microarray data not only provided the information of lncRNA expression, but also provided mRNA expression information between nondiabetic and diabetic retinas. We thus constructed an lncRNA/mRNA coexpression network, and investigated the potential interaction between mRNAs and lncRNAs. Seventy-nine differentially expressed lncRNAs and 100 mRNAs comprised the coexpression network, which was composed of 2675 network nodes (Supplementary Table S2). We selected 15 of the most differentially expressed lncRNAs between diabetic and nondiabetic retinas, and drew the regulatory network using the cytoscape program. The coexpression network was composed of 15 lncRNAs and 74 coexpressed mRNAs (Fig. 4). The network indicated that one mRNA could correlate with a great number of target lncRNAs and so were the lncRNAs, implying the inter-regulation of lncRNAs and mRNAs occurred in the early stage of DR. 
Figure 4
 
The lncRNA/mRNA coexpression network constructed using the cytoscape program. (A) The lncRNAs and mRNAs with Pearson correlation coefficients not less than 0.99 were selected to draw the regulatory network by using the cytoscape program. (B) Real-time PCR experiments were conducted to verify the expression associations between lncRNAs and mRNAs.
Figure 4
 
The lncRNA/mRNA coexpression network constructed using the cytoscape program. (A) The lncRNAs and mRNAs with Pearson correlation coefficients not less than 0.99 were selected to draw the regulatory network by using the cytoscape program. (B) Real-time PCR experiments were conducted to verify the expression associations between lncRNAs and mRNAs.
Gene Enrichment and Pathway Analysis of lncRNAs–Coexpressed mRNAs
Gene enrichment analysis was performed to determine the gene and gene product enrichment in biological processes, cellular components, and molecular functions. 18 We found that the highest enriched GOs targeted by lncRNAs–coexpressed mRNAs were cell response to stress (ontology: biological process), integral to membrane (ontology: cellular component), and structural molecule activity (ontology: molecular function; Fig. 5A). KEGG pathway analysis indicated that the lncRNAs–coexpressed mRNAs were involved in the regulation of axon guidance, MAPK signaling pathway, complement and coagulation cascades, chemokine signaling pathway, and pyruvate metabolism. One of these pathways, the gene category “axon guidance” is shown in Figure 5B, which has been reported to be involved in the pathological process of ocular neurodegeneration. 19,20  
Figure 5
 
Gene enrichment and pathway analysis of lncRNAs–coexpressed mRNAs. (A) The GO enrichment analysis provided a controlled vocabulary to describe the differentially expressed lncRNAs–coexpressed mRNAs. The ontology covered three domains: biological process, cellular component, and molecular function (P ≤ 0.05 is recommended). (B) KEGG analysis suggested that lncRNAs–coexpressed mRNAs were mainly targeted to the “axon guidance” signaling pathway.
Figure 5
 
Gene enrichment and pathway analysis of lncRNAs–coexpressed mRNAs. (A) The GO enrichment analysis provided a controlled vocabulary to describe the differentially expressed lncRNAs–coexpressed mRNAs. The ontology covered three domains: biological process, cellular component, and molecular function (P ≤ 0.05 is recommended). (B) KEGG analysis suggested that lncRNAs–coexpressed mRNAs were mainly targeted to the “axon guidance” signaling pathway.
Bioinformatics Analysis of MALAT1
MALAT1 is one of the highly conserved lncRNAs in mammals. It is located on chromosome 11q13, and is known to be misregulated in several solid tumors and associated with cancer metastasis and recurrence. 8 In the results of our microarray data, we found that an orthology of MALAT1, linRNA chr19: 5795689-5802671, was aberrantly expressed in diabetic retinas (Fig. 6A). Transcription factors were recognized as important components of signaling cascades controlling all types of normal cellular processes as well as response to external stimulus. 13 Here, we used the TRANSFAC program to predict the TFBS in the sequences of MALAT1. The result indicated that MALAT1 could combine with nuclear factor-κB (NF-κB) motif as the cis-acting element, respectively (Fig. 6B). We also conducted catRAPID analysis to predict the potential interacting protein of MALAT1. We found a strong interaction between MALAT1 and CPNL1 (Fig. 6C). 
Figure 6
 
Bioinformatics analysis of MALAT1. (A) The chromosome location of MALAT1 is shown in the mouse genome. (B) Transcription factor binding site prediction indicated that NF-κB was the cis-acting element of MALAT1. (C) catRAPID analysis indicated a strong interaction between CPNL1 and MALAT1.
Figure 6
 
Bioinformatics analysis of MALAT1. (A) The chromosome location of MALAT1 is shown in the mouse genome. (B) Transcription factor binding site prediction indicated that NF-κB was the cis-acting element of MALAT1. (C) catRAPID analysis indicated a strong interaction between CPNL1 and MALAT1.
Clinical Relevance
To reveal the potential role of MALAT1 in the pathogenesis of ocular diseases, we first treated RF/6A cells with high levels of glucose (50 mM) to mimic the diabetic condition in vitro, and then detected the expression pattern of MALAT1. We found that high glucose resulted in approximately 40% increase in MALAT1 levels for the first 24 hours, and an approximately 2.2- and 4.1-fold increase in MALAT1 levels after 36- and 48-hour high glucose treatment. By contrast, the expression of MALAT1 was stably expressed during the experimental period in the normal glucose (5 mM) group and mannitol-treated group (Fig. 7A). 
Figure 7
 
Expression pattern of MALAT1 in RF/6A cells, the aqueous humor, and FVMs of DR patients. (A) RF/6A cells were incubated in media containing 5 mM glucose, 50 mM glucose, or 50 mM mannitol for 24 hours, 36 hours, and 48 hours. The group treated with 5 mM glucose was taken as the control group, and the group treated with mannitol was taken as the osmolar control. The levels of MALAT1 were determined by real-time PCR. (B) The FVMs were collected from the eyes of patients with PDR, or with macular holes or preretinal membranes. Real-time PCR was conducted to detect the level of MALAT1 expression. (C) Samples of aqueous humor were harvested from the eyes of patients who had PDR or nondiabetic ocular diseases. None of the patients with nondiabetic ocular diseases had diabetes mellitus. The levels of MALAT1 were determined by real-time PCR. The data of each group are expressed as the relative change compared with the control group. The asterisk indicates the significant difference compared with the control group.
Figure 7
 
Expression pattern of MALAT1 in RF/6A cells, the aqueous humor, and FVMs of DR patients. (A) RF/6A cells were incubated in media containing 5 mM glucose, 50 mM glucose, or 50 mM mannitol for 24 hours, 36 hours, and 48 hours. The group treated with 5 mM glucose was taken as the control group, and the group treated with mannitol was taken as the osmolar control. The levels of MALAT1 were determined by real-time PCR. (B) The FVMs were collected from the eyes of patients with PDR, or with macular holes or preretinal membranes. Real-time PCR was conducted to detect the level of MALAT1 expression. (C) Samples of aqueous humor were harvested from the eyes of patients who had PDR or nondiabetic ocular diseases. None of the patients with nondiabetic ocular diseases had diabetes mellitus. The levels of MALAT1 were determined by real-time PCR. The data of each group are expressed as the relative change compared with the control group. The asterisk indicates the significant difference compared with the control group.
To reveal the clinical relevance of MALAT1 misregulation, we conducted real-time PCR to examine MALAT1 levels in the aqueous humor and in the FVMs of patients with PDR. The clinical characteristics of the patients are shown in Table 3. Results of quantitative RT-PCR showed that MALAT1 expression was significantly higher in the FVMs of the nine patients with PDR than in the membranes of the six patients with idiopathic epiretinal membranes (Fig. 7B). We also found that MALAT1 level was significantly higher in the aqueous humor of PDR patients than that of nondiabetic patients (Fig. 7C). Taken together, these results support the hypothesis that MALAT1 misregulation is a potential molecular mechanism of DR pathogenesis. 
Table 3
 
Clinical Characteristics of the Patients for FVMs and Aqueous Humor Collection
Table 3
 
Clinical Characteristics of the Patients for FVMs and Aqueous Humor Collection
FVMs Aqueous Humor
Clinical Parameters DR NDR DR NDR
Number, case 9 6 10 10
Age, y 50.07 ± 2.85 55.33 ± 6.26 56.8 ± 8.8 58.4 ± 10.2
Sex, M/F 6/3 4/2 4/6 5/5
Total cholesterol 170.33 ± 30.52 180 ± 8.26 166.45 ± 10.44 175 ± 11.08
Creatinine, mg/dL 1.46 ± 0.32 2.15 ± 1.66 1.68 ± 0.76 2.00 ± 1.02
Triglyceride, mg/dL 128.33 ± 94.48 91.5 ± 23.34 134 ± 6.88 95 ± 10.82
Glycosylated hemoglobin 7.2 ± 0.85 6.8 ± 0.68
Discussion
In recent years, the concept of the functional genome has been rewritten to encompass a multitude of newly discovered noncoding RNA transcripts. The functional significance of lncRNAs has long been recognized. 7,21 However, the change in the abundance and scale of lncRNA expression in ocular diseases is just beginning to come to light. DR is a result of multiple pathogenetic processes caused by hyperglycemia and abnormalities of insulin-signaling pathways, leading to retinal microvascular defects and neuroretinal dysfunction and degeneration. 22 Although significant progress has been made, the molecular mechanisms underlying DR pathogenesis are still not fully understood. For this reason, charting the transcriptional landscape of lncRNAs is a key step in understanding the significance of lncRNAs in DR. Here, we detected lncRNA profiling in a murine model of DR by using microarray analysis. We identified that 303 lncRNAs were aberrantly expressed at the early stage of DR. Moreover, we found that MALAT1, a conserved lncRNA, was significantly upregulated in the RF/6A cell model of hyperglycemia, in the aqueous humor, and the FVMs of DR patients, implying its application as a biomarker for the prognosis and diagnosis of DR. 
To date, several lncRNAs have been implicated in eye development, including Vax2os1, RNCR2, Six3OS, and Tug1. Vax2os1 controls the cell cycle progression of photoreceptor progenitors in the mouse retina. 23 RNCR2, Six3OS, and Tug1 play a critical role in regulating retinal cell fate specification. 24 Islet-cell dysfunction is central to the pathophysiology of type 2 diabetes. Type 2 diabetes would manifest when the β-cell fails to secrete sufficient amounts of insulin to maintain normoglycemia and undergoes apoptosis. Human β cell transcriptome analysis uncovers that lncRNAs are tissue specific, dynamically regulated, and abnormally expressed in type 2 diabetes. 2527 In addition, GWAS study indicates that antisense noncoding RNA in the INK4 locus (ANRIL) is significantly associated with increased susceptibility to type 2 diabetes. 28 Thus, it is no surprise that abnormal lncRNA expression may be relevant to the molecular etiology of DR. The identification of dysregulated lncRNAs may also open a new framework to study the pathophysiology of diabetes and diabetic complications. 
Compared with the protein-coding sequences, most lncRNAs are poorly conserved throughout vertebrates. It is difficult to predict the functions of lncRNAs based on their nucleotide sequences. 29 To reveal the functional significance of lncRNAs in DR, we constructed the lncRNA/mRNA coexpression network based on the correlation analysis. lncRNAs–coexpressed mRNAs are targeted to “cellular response to stress” (ontology: biological process), “integral to membrane” (ontology: cellular component), and “structural molecule activity” (ontology: molecular function). As shown in Figure 5A, the biological processes, such as epithelium and tube development, may be involved in the regulation of retinal vascular leakage. The biological processes, including tube morphogenesis, epithelium development, and branching morphogenesis of a tube, may be involved in the process of pathological neovascularization. 19 lncRNAs–coexpressed mRNAs are also targeted to several signaling pathways, including axon guidance, MAPK signaling pathway, complement and coagulation cascades, chemokine signaling pathway, and pyruvate metabolism. These signaling pathways are tightly associated with the pathological processes, such as neurodegeneration, neovascularization, inflammation, and immunology, suggesting the lncRNA-mediated network plays a wide role in the pathogenesis of DR. Of them, “axon guidance” is essential for the establishment of proper neuronal connections during development, which gains on the top count score during KEGG analysis. 20 This result implies that neurodegeneration is an early event in the pathogenesis of DR. The finding is in accordance with previous clinical evidence from ERG, contrast sensitivity, perimetric, and color vision studies, which suggests that neuronal changes may occur before clinically detectable microvasculopathy. 30  
Compared with the protein-coding sequences, lncRNA sequences evolve very rapidly. 22 In a recent study, Mustafi et al. 31 identified only 18 conserved lncRNAs from 3133 mouse lncRNAs, which is similar to other studies that showed only a small minority of lncRNAs in the mouse or human have transcribed orthologous sequences across different species. 32 By contrast, the highly conserved lncRNAs would play a critical and conserved role across different species. 29 MALAT1 is a highly conserved lncRNA across different mammalian species. 33 MALAT1 has been reported to be significantly upregulated in several solid tumors and is linked to cancer metastasis and recurrence. In this study, we found that MALAT1 is significantly upregulated in the retinas of diabetic mice, in the RF/6A cell model of hyperglycemia, in the aqueous humor samples, and FVMs of diabetic patients. Thus, we provide the evidence from cell line, animal model, and clinical samples, which indicate that MALAT1 dysregulation may become the molecular etiology of DR occurrence. 
Regular ophthalmological examinations, timely laser therapy depending on the stage of the disease, and close interdisciplinary cooperation are essential to prevent loss of vision in DR. 33 However, the prognosis for DR patients is still poor. Aqueous humor is an important intraocular fluid responsible for supplying nutrients to and the removal of metabolic wastes from the avascular tissues of the eye. Changes in aqueous humor protein content have been associated with potentially blinding diseases, such as primary congenital glaucoma, 34 myopia, 35 and Fuchs endothelial corneal dystrophy. 36 Based on the fact that lncRNAs have been detected in other body fluids, 37 we speculated that they may also be present in aqueous humor. Importantly, we identified that MALAT1 expression is significantly upregulated in the aqueous humor of DR patients. The change in MALAT1 level may affect the physiological function of ocular tissue, which may become an indicator in the early stage of DR. MALAT1 may be developed as a biomarker for diagnosing or making a prognosis of DR based on the MALAT1 levels. 
FVMs form on the surface of the neuroretina as a sequela to PDR. FVMs are characterized by the migration and proliferation of various types of cells (e.g., retinal glial cells, fibroblasts, macrophages/monocytes, hyalocytes, laminocytes, and vascular endothelial cells). 38 The BFGF, VEGF, TNF-α, angiopoietin-2, hepatocyte growth factor, monocyte chemoattractant protein-1, IL-8, NF-κB, and activator protein-1 have been detected in FVMs collected from PDR patients. 39 These observations support the concept that a complex local milieu, rather than a single or a few growth factors, influences the generation of FVMs. Here, MALAT1 is significantly upregulated in the FVMs of patients with PDR. MALAT1 is a critical regulator of cell motility, cell cycle progression, and cell proliferation. 4042 Thus, it is no surprise that MALAT1 dysregulation could affect the migration and proliferation of FVM-related cells, which affect the DR pathogenesis. 
DR is one of the most important diabetic complications. Genetic factors, environmental factors, and the complex gene/environment interactions may be implicated in the pathogenesis of DR. Here, we present the detailed data designed to serve as a resource for elucidating lncRNA-mediated DR pathogenesis. Further, we found that MALAT1, a conserved lncRNA, may become a potential biomarker for the prognosis and diagnosis of DR. More studies are required to investigate the correlations between lncRNA change and DR development at different stages. Moreover, in vivo and in vitro studies should be conducted to elucidate the molecular mechanisms of lncRNA-mediated DR occurrence and estimate their potential for the prognosis, diagnosis, and treatment of DR. 
Supplementary Materials
Acknowledgments
Supported by the National Natural Science Foundation Grants 81300241 (BY), 81271028 (JY), and 81371055 (QJ); the National Clinical Key Construction Project Grant (2012) 649 (QJ); and the Medical Science and Technology Development Project Fund of Nanjing Grants ZKX12047 (QJ), YKK13227 (BY), and YKK12208 (JY). 
Disclosure: B. Yan, None; Z.-F. Tao, None; X.-M. Li, None; H. Zhang, None; J. Yao, None; Q. Jiang, None 
References
McArthur K Feng B Wu Y Chen S Chakrabarti S. MicroRNA-200b regulates vascular endothelial growth factor-mediated alterations in diabetic retinopathy. Diabetes . 2011; 60: 1314–1323. [CrossRef] [PubMed]
Yau JW Rogers SL Kawasaki R Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care . 2012; 35: 556–564. [CrossRef] [PubMed]
Fu Y-P Hallman DM Gonzalez VH Identification of diabetic retinopathy genes through a genome-wide association study among Mexican-Americans from Starr County, Texas. J Ophthalmol . 2010; 2010: 861291. [PubMed]
Doria A. Genetics of diabetes complications. Curr Diabetes Rep . 2010; 10: 467–475. [CrossRef]
Liew G Klein R Wong TY. The role of genetics in susceptibility to diabetic retinopathy. Int Ophthalmol Clin . 2009; 49: 35–52. [CrossRef] [PubMed]
Sobrin L Green T Sim X Candidate gene association study for diabetic retinopathy in persons with type 2 diabetes: the Candidate gene Association Resource (CARe). Invest Ophthalmol Vis Sci . 2011; 52: 7593–7602. [CrossRef] [PubMed]
Mercer TR Dinger ME Mattick JS. Long non-coding RNAs: insights into functions. Nat Rev Genet . 2009; 10: 155–159. [CrossRef] [PubMed]
Wapinski O Chang HY. Long noncoding RNAs and human disease. Trends Cell Biol . 2011; 21: 354–361. [CrossRef] [PubMed]
Hindorff LA Sethupathy P Junkins HA Potential etiologic and functional implications of genome-wide association loci for human diseases and traits. Proc Natl Acad Sci U S A . 2009; 106: 9362–9367. [CrossRef] [PubMed]
Sasaki M Ozawa Y Kurihara T Neurodegenerative influence of oxidative stress in the retina of a murine model of diabetes. Diabetologia . 2010; 53: 971–979. [CrossRef] [PubMed]
Jiang Y Zhang Q Liu L Tang J Kern TS Steinle JJ. β2-adrenergic receptor knockout mice exhibit a diabetic retinopathy phenotype. PLoS One . 2013; 8: e70555. [CrossRef] [PubMed]
Dennis Jr G Sherman BT Hosack DA . DAVID: database for annotation, visualization, and integrated discovery. Genome Biol . 2003; 4: P3. [CrossRef] [PubMed]
Wingender E Chen X Hehl R TRANSFAC: an integrated system for gene expression regulation. Nucleic Acids Res . 2000; 28: 316–319. [CrossRef] [PubMed]
Agostini F Zanzoni A Klus P Marchese D Cirillo D Tartaglia GG. catRAPID omics: a web server for large-scale prediction of protein–RNA interactions. Bioinformatics . 2013; 29: 2928–2930. [CrossRef] [PubMed]
Liao Q Liu C Yuan X Large-scale prediction of long non-coding RNA functions in a coding–non-coding gene co-expression network. Nucleic Acids Res . 2011; 39: 3864–3878. [CrossRef] [PubMed]
Ponting CP Oliver PL Reik W. Evolution and functions of long noncoding RNAs. Cell . 2009; 136: 629–641. [CrossRef] [PubMed]
Qureshi IA Mehler MF. Emerging roles of non-coding RNAs in brain evolution, development, plasticity and disease. Nat Rev Neurosci . 2012; 13: 528–541. [CrossRef] [PubMed]
Subramanian A Tamayo P Mootha VK Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A . 2005; 102: 15545–15550. [CrossRef] [PubMed]
Campochiaro PA. Ocular neovascularization. J Mol Med . 2013; 91: 311–321. [CrossRef] [PubMed]
Lange J Yafai Y Noack A The axon guidance molecule Netrin-4 is expressed by Müller cells and contributes to angiogenesis in the retina. Glia . 2012; 60: 1567–1578. [CrossRef] [PubMed]
Rinn JL Chang HY. Genome regulation by long noncoding RNAs. Annu Rev Biochem . 2012; 81: 145–166. [CrossRef] [PubMed]
Kutter C Watt S Stefflova K Rapid turnover of long noncoding RNAs and the evolution of gene expression. PLoS Genet . 2012; 8: e1002841. [CrossRef] [PubMed]
Meola N Pizzo M Alfano G Surace EM Banfi S. The long noncoding RNA Vax2os1 controls the cell cycle progression of photoreceptor progenitors in the mouse retina. RNA . 2012; 18: 111–123. [CrossRef] [PubMed]
Rapicavoli NA Poth EM Blackshaw S. The long noncoding RNA RNCR2 directs mouse retinal cell specification. BMC Dev Biol . 2010; 10: 49. [CrossRef] [PubMed]
Morán I Akerman İ van de Bunt M Human β cell transcriptome analysis uncovers lncRNAs that are tissue-specific, dynamically regulated, and abnormally expressed in type 2 diabetes. Cell Metab . 2012; 16: 435–448. [CrossRef] [PubMed]
Rapicavoli NA Poth EM Zhu H Blackshaw S. The long noncoding RNA Six3OS acts in trans to regulate retinal development by modulating Six3 activity. Neural Dev . 2011; 6: 1–15. [CrossRef] [PubMed]
Young T Matsuda T Cepko C. The noncoding RNA Taurine Upregulated Gene 1 is required for differentiation of the murine retina. Curr Biol . 2005; 15: 501–512. [CrossRef] [PubMed]
Tsai F-J Yang C-F Chen C-C A genome-wide association study identifies susceptibility variants for type 2 diabetes in Han Chinese. PLoS Genet . 2010; 6: e1000847. [CrossRef] [PubMed]
Chodroff RA Goodstadt L Sirey TM Long noncoding RNA genes: conservation of sequence and brain expression among diverse amniotes. Genome Biol . 2010; 11: R72. [CrossRef] [PubMed]
Zhang X Wang N Bao S Gillies M. Diabetic retinopathy: neuron protection as a therapeutic target. Int J Biochem Cell Biol . 2013; 45: 1525–1529. [CrossRef] [PubMed]
Mustafi D Kevany BM Bai X Evolutionarily conserved long intergenic non-coding RNAs in the eye. Hum Mol Genet . 2013; 22: 2992–3002. [CrossRef] [PubMed]
Cabili MN Trapnell C Goff L Integrative annotation of human large intergenic noncoding RNAs reveals global properties and specific subclasses. Gene Dev . 2011; 25: 1915–1927. [CrossRef] [PubMed]
Kollias AN Ulbig MW. Diabetic retinopathy: early diagnosis and effective treatment. Dtsch Arztebl Int . 2010; 107: 75–83. [PubMed]
Bouhenni RA Al Shahwan S Morales J Identification of differentially expressed proteins in the aqueous humor of primary congenital glaucoma. Exp Eye Res . 2011; 92: 67–75. [CrossRef] [PubMed]
Duan X Lu Q Xue P Proteomic analysis of aqueous humor from patients with myopia. Mol Vis . 2008; 14: 370–377. [PubMed]
Richardson MR Segu ZM Price MO Alterations in the aqueous humor proteome in patients with Fuchs endothelial corneal dystrophy. Mol Vis . 2010; 16: 2376–2783. [PubMed]
Ayers D. Long non-coding RNAs: novel emergent biomarkers for cancer diagnostics. Nature . 2013; 1: 31–35.
Yoshida S Ogura A Ishikawa K Gene expression profile of fibrovascular membranes from patients with proliferative diabetic retinopathy. Br J Ophthalmol . 2010; 94: 795–801. [CrossRef] [PubMed]
Chung AS Ferrara N. Developmental and pathological angiogenesis. Annu Rev Cell Dev Biol . 2011; 27: 563–584. [CrossRef] [PubMed]
Tano K Mizuno R Okada T MALAT-1 enhances cell motility of lung adenocarcinoma cells by influencing the expression of motility-related genes. FEBS Lett . 2010; 584: 4575–4580. [CrossRef] [PubMed]
Tripathi V Shen Z Chakraborty A Long noncoding RNA MALAT1 controls cell cycle progression by regulating the expression of oncogenic transcription factor B-MYB. PLoS Genet . 2013; 9: e1003368. [CrossRef] [PubMed]
Tripathi V Ellis JD Shen Z The nuclear-retained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Mol Cell . 2010; 39: 925–938. [CrossRef] [PubMed]
Footnotes
 BY and Z-FT contributed equally to the work presented here and should therefore be regarded as equivalent authors.
Figure 1
 
ERG levels in the retinas of diabetic and nondiabetic mice. Mean ERG amplitudes for the retinas of diabetic and nondiabetic mice (n = 10 for each group) are shown. The top (A) shows the A-wave. The middle (B) represents the B-wave. The bottom (C) shows the amplitudes for the OPs.
Figure 1
 
ERG levels in the retinas of diabetic and nondiabetic mice. Mean ERG amplitudes for the retinas of diabetic and nondiabetic mice (n = 10 for each group) are shown. The top (A) shows the A-wave. The middle (B) represents the B-wave. The bottom (C) shows the amplitudes for the OPs.
Figure 2
 
Overview of lncRNAs microarray analysis. (A) The box plot displays the distributions of lncRNA expression profiling. After normalization, the distributions of log 2 ratios among different samples are shown. The box plots consist of boxes with a central line and two tails. The central line represents the median of the data, whereas the tails represent the upper and lower quartiles. (B) The scatter plot is used to compare the expression variations between nondiabetic and diabetic retinas. (C) Heatmaps were generated from the hierarchical cluster analysis to show the differential expressed lncRNAs between nondiabetic and diabetic retinas. The color scale on the top illustrates the relative expression level of lncRNAs across all samples: red denotes expression greater than 0, and green denotes expression less than 0.
Figure 2
 
Overview of lncRNAs microarray analysis. (A) The box plot displays the distributions of lncRNA expression profiling. After normalization, the distributions of log 2 ratios among different samples are shown. The box plots consist of boxes with a central line and two tails. The central line represents the median of the data, whereas the tails represent the upper and lower quartiles. (B) The scatter plot is used to compare the expression variations between nondiabetic and diabetic retinas. (C) Heatmaps were generated from the hierarchical cluster analysis to show the differential expressed lncRNAs between nondiabetic and diabetic retinas. The color scale on the top illustrates the relative expression level of lncRNAs across all samples: red denotes expression greater than 0, and green denotes expression less than 0.
Figure 3
 
Differential expressed lncRNAs between nondiabetic and diabetic retinas. (A) The volcano plot illustrates the differential expressed lncRNAs between the diabetic and nondiabetic groups. (B) Chromosome distribution of differentially expressed lncRNAs based on microarray data is shown.
Figure 3
 
Differential expressed lncRNAs between nondiabetic and diabetic retinas. (A) The volcano plot illustrates the differential expressed lncRNAs between the diabetic and nondiabetic groups. (B) Chromosome distribution of differentially expressed lncRNAs based on microarray data is shown.
Figure 4
 
The lncRNA/mRNA coexpression network constructed using the cytoscape program. (A) The lncRNAs and mRNAs with Pearson correlation coefficients not less than 0.99 were selected to draw the regulatory network by using the cytoscape program. (B) Real-time PCR experiments were conducted to verify the expression associations between lncRNAs and mRNAs.
Figure 4
 
The lncRNA/mRNA coexpression network constructed using the cytoscape program. (A) The lncRNAs and mRNAs with Pearson correlation coefficients not less than 0.99 were selected to draw the regulatory network by using the cytoscape program. (B) Real-time PCR experiments were conducted to verify the expression associations between lncRNAs and mRNAs.
Figure 5
 
Gene enrichment and pathway analysis of lncRNAs–coexpressed mRNAs. (A) The GO enrichment analysis provided a controlled vocabulary to describe the differentially expressed lncRNAs–coexpressed mRNAs. The ontology covered three domains: biological process, cellular component, and molecular function (P ≤ 0.05 is recommended). (B) KEGG analysis suggested that lncRNAs–coexpressed mRNAs were mainly targeted to the “axon guidance” signaling pathway.
Figure 5
 
Gene enrichment and pathway analysis of lncRNAs–coexpressed mRNAs. (A) The GO enrichment analysis provided a controlled vocabulary to describe the differentially expressed lncRNAs–coexpressed mRNAs. The ontology covered three domains: biological process, cellular component, and molecular function (P ≤ 0.05 is recommended). (B) KEGG analysis suggested that lncRNAs–coexpressed mRNAs were mainly targeted to the “axon guidance” signaling pathway.
Figure 6
 
Bioinformatics analysis of MALAT1. (A) The chromosome location of MALAT1 is shown in the mouse genome. (B) Transcription factor binding site prediction indicated that NF-κB was the cis-acting element of MALAT1. (C) catRAPID analysis indicated a strong interaction between CPNL1 and MALAT1.
Figure 6
 
Bioinformatics analysis of MALAT1. (A) The chromosome location of MALAT1 is shown in the mouse genome. (B) Transcription factor binding site prediction indicated that NF-κB was the cis-acting element of MALAT1. (C) catRAPID analysis indicated a strong interaction between CPNL1 and MALAT1.
Figure 7
 
Expression pattern of MALAT1 in RF/6A cells, the aqueous humor, and FVMs of DR patients. (A) RF/6A cells were incubated in media containing 5 mM glucose, 50 mM glucose, or 50 mM mannitol for 24 hours, 36 hours, and 48 hours. The group treated with 5 mM glucose was taken as the control group, and the group treated with mannitol was taken as the osmolar control. The levels of MALAT1 were determined by real-time PCR. (B) The FVMs were collected from the eyes of patients with PDR, or with macular holes or preretinal membranes. Real-time PCR was conducted to detect the level of MALAT1 expression. (C) Samples of aqueous humor were harvested from the eyes of patients who had PDR or nondiabetic ocular diseases. None of the patients with nondiabetic ocular diseases had diabetes mellitus. The levels of MALAT1 were determined by real-time PCR. The data of each group are expressed as the relative change compared with the control group. The asterisk indicates the significant difference compared with the control group.
Figure 7
 
Expression pattern of MALAT1 in RF/6A cells, the aqueous humor, and FVMs of DR patients. (A) RF/6A cells were incubated in media containing 5 mM glucose, 50 mM glucose, or 50 mM mannitol for 24 hours, 36 hours, and 48 hours. The group treated with 5 mM glucose was taken as the control group, and the group treated with mannitol was taken as the osmolar control. The levels of MALAT1 were determined by real-time PCR. (B) The FVMs were collected from the eyes of patients with PDR, or with macular holes or preretinal membranes. Real-time PCR was conducted to detect the level of MALAT1 expression. (C) Samples of aqueous humor were harvested from the eyes of patients who had PDR or nondiabetic ocular diseases. None of the patients with nondiabetic ocular diseases had diabetes mellitus. The levels of MALAT1 were determined by real-time PCR. The data of each group are expressed as the relative change compared with the control group. The asterisk indicates the significant difference compared with the control group.
Table 1
 
General Physiological Parameters in Diabetic and Nondiabetic Mice
Table 1
 
General Physiological Parameters in Diabetic and Nondiabetic Mice
Nondiabetic, n = 10 Diabetic, n = 10
2 wk after diabetic
 Body weight, g 28.1 ± 4.5 27.8 ± 3.1
 Glucose, mg/dL 115 ± 6 278 ± 36*
4 wk after diabetic
 Body weight, g 31.5 ± 5.2 29.7 ± 4.2*
 Glucose, mg/dL 110 ± 8 297 ± 44*
8 wk after diabetic
 Body weight, g 37.6 ± 5.8 31.7 ± 2.9*
 Glucose, mg/dL 105 ± 4 315 ± 39*
Table 2
 
Differentially Expressed lncRNAs in the Retinas of Diabetic Mice Compared With That of Nondiabetic Mice
Table 2
 
Differentially Expressed lncRNAs in the Retinas of Diabetic Mice Compared With That of Nondiabetic Mice
lncRNAs Fold Change, DR/NDR, Microarray Fold Change, DR/NDR, qRT-PCR
Downregulated
 chr4:35106125-35127700  reverse strand 13.62 10.55
 chr15:101088974-101089576  reverse strand 11.95 8.63
 chr6:126766225-126771309  reverse strand 10.85 9.44
 chr19:53624826-53625580  reverse strand 10.64 5.66
 chr13:98062189-98066613  forward strand 10.34 10.21
 chr1:45569500-45580900  reverse strand 10.18 9.86
 chr10:78969674-78981039  forward strand 10.09 6.94
 chr2:33661335-33669111 f  orward strand 9.89 7.34
 chr17:21785442-21801542  reverse strand 9.62 6.08
Upregulated
 chr10:120336296-120354471  forward strand 26.38 15.34
 chr19:23068397-23190256  forward strand 13.10 11.26
 chr8:59994102-59994797  forward strand 10.74 9.52
 chr14:44151048-44156924  reverse strand 10.68 8.77
 chr14:65878005-65878630  forward strand 8.71 7.43
 chr8:59994102-59994797  forward strand 7.42 8.10
 chr9:27148989-27155714  reverse strand 5.01 2.22
 chr18:75513062-75523889  forward strand 4.81 4.33
Table 3
 
Clinical Characteristics of the Patients for FVMs and Aqueous Humor Collection
Table 3
 
Clinical Characteristics of the Patients for FVMs and Aqueous Humor Collection
FVMs Aqueous Humor
Clinical Parameters DR NDR DR NDR
Number, case 9 6 10 10
Age, y 50.07 ± 2.85 55.33 ± 6.26 56.8 ± 8.8 58.4 ± 10.2
Sex, M/F 6/3 4/2 4/6 5/5
Total cholesterol 170.33 ± 30.52 180 ± 8.26 166.45 ± 10.44 175 ± 11.08
Creatinine, mg/dL 1.46 ± 0.32 2.15 ± 1.66 1.68 ± 0.76 2.00 ± 1.02
Triglyceride, mg/dL 128.33 ± 94.48 91.5 ± 23.34 134 ± 6.88 95 ± 10.82
Glycosylated hemoglobin 7.2 ± 0.85 6.8 ± 0.68
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×