Discrepant Expression of MicroRNAs in Transparent and Cataractous Human Lenses

Abstract

Purpose.: Age-related cataracts are considered to be a pathological condition that arise as senescence progresses. However, little is known about the function of microRNAs (miRNAs) in the formation of age-related cataracts. The purpose of this study was to identify possible differences in miRNA expression in the central epithelium of transparent and age-related cataractous human lenses.

Methods.: Microarrays were used to determine the miRNA expression profiles of both transparent and cataractous lenses. The results were analyzed by significance analyses performed by the microarray software, and the results were confirmed by stem-loop RT-PCR. Algorithms were used to predict the target genes of the differentially expressed miRNAs.

Results.: Two hundred and six miRNAs were identified in all human lenses. The top eight miRNAs according to expression levels were miR-184, let-7b, miR-923, miR-1826, miR-125b, miR-1308, miR-26a, and miR-638 in transparent lenses. In contrast, the top eight miRNAs in cataractous lenses were miR-184, miR-1826, let-7b/c, miR-24, miR-23b, miR-923, and miR-23a. The expression levels of 20 miRNAs were increased and the levels of 12 miRNAs were decreased by more than 2-fold in transparent lenses relative to the levels in cataractous lenses. These findings were confirmed by stem-loop RT-PCR. In addition, several genes that were predicted to be targets of the identified miRNAs have been reported to be involved in lens development or cataract formation.

Conclusions.: The authors report, for the first time, the distinct expression profiles of miRNAs in the central epithelium of transparent and age-related cataractous human lenses. Significant differences in miRNA expression were identified, and the genes targeted by the relevant miRNAs were predicted. The differential expression of miRNAs suggests that these miRNAs have potential roles in lens development and/or cataract formation.

Introduction

MicroRNAs (miRNAs), a type of noncoding, small RNA in animals, plants, and fungi, regulate the expression levels of target genes by binding imperfectly to three prime untranslated region (3′ UTR) sequences in mRNAs and attenuating protein translation. ¹ The miRNAs are highly conserved and have been shown to have diverse temporal and tissue-specific expression patterns in a wide range of species. ^2,3 This evidence supports the hypothesis that miRNAs may play important roles in the regulation of diverse physiological processes (such as development and senescence) and pathological processes. For example, lin-4, a highly conserved miRNA among various species, controls the timing of larval development in Caenorhabditis elegans and regulates tissue aging in adult worms. ⁴  

Human lenses are relatively isolated tissues that contribute to the refraction of light to the retina, and are surrounded by the aqueous humor and the vitreous body. Lenses are transparent in young people, but changes occur as the body ages. These changes include the development of a hard, compact nucleus, local opacity, and, finally, the development of a pathological cataract. These characteristics allow the lenses to be used as adaptive models to study the function of miRNAs in cataracts. To date, several miRNAs have been identified in animal eyes ^5,6 ; miR-184, miR-205, miR-26a, miR-31, and miR-24 were detected in the cornea; miR-184, miR-125b, miR-31, miR-204, miR-26a, let-7b, and others were detected in the lens; and miR-181a, miR-125b, miR-124a, miR-30c, miR-182, and miR-183 have been reported to be expressed in the retina. However, the miRNA expression profiles in human lenses remain to be established, and little research has been performed to determine the miRNA expression profiles in cataracts. Cataracts are a condition caused by senescence, but the etiology has not been explained to date. 

Several methods have been used to detect miRNA expression, including microarray analysis, stem-loop RT-PCR, northern blot analysis, and in situ hybridization. ^7,8 In the current study, the authors used a microarray-based approach, which is a revolutionary approach that combines molecular biology and computer technology to allow the high-throughput, simultaneous analysis of global miRNA expression. ⁹ Prediction algorithms have been developed to predict the miRNAs' target genes through analysis of the conservation degree of the 5′ region of the miRNAs, and the binding sites in the 3′ UTR of mRNAs. ¹⁰ In this study, the authors determined the miRNA expression profiles in the central epithelium of transparent and cataractous human lenses by microarray, and, subsequently, analyzed the results to identify the miRNAs that had distinct expression levels. Finally, several commonly used and accurate miRNA prediction algorithms, TargetScan, miRanda, and PicTar, were used to predict the target genes of these miRNAs and to assess their function in human lenses. 

Methods

Inclusion Criteria, Classification, and Tissue Grouping

Forty lens samples were collected from postmortem eyes (20 donors, donor age range was 58–65 years, free of ocular diseases) and age-related cataract patients (20 patients, patient age range was 60–68 years, free of other ocular diseases). The degree of lenticular opacification was determined by the Lens Opacities Classification System III (LOCSIII). ^11,12 Lenses from postmortem eyes were obtained from the Eye Bank of Zhongshan Ophthalmic Center in Guangzhou, China, 8 to 24 hours after death. The lenticular opacification of these lenses was graded 1 to approximately 2, and these samples were used as the transparent lens samples. The cataractous lens samples had lenticular opacification of grade 4 to approximately 6. These lenses were acquired during surgeries performed on age-related cataract patients after informed consent of each patient was obtained. In each cataractous lens, a nuclear cataract was obvious, and the nuclear cataract type was identified (see Supplementary Material and Supplementary Table S1). There were no statistically significant differences between the transparent and cataractous human lenses with respect to the age of the donor (P > 0.05, Independent Sample t-test). This study was performed in accordance with the tenets of the Declaration of Helsinki for Research Involving Human Tissue. The authors also received consent for research use with the approval of the Sun Yat-sen University Institutional Human Experimentation Committee. 

Tissue Preparation and RNA Extraction

All lens sample specimens were centered anterior capsules with a 5- to approximately 6-mm diameter obtained using anterior continuous curvilinear capsulorhexis. First, RNA samples sourced from the central epithelium of one lens do not contain sufficient RNA for a microarray assay, therefore, the authors randomly pooled the central epithelia of five lenses (transparent or cataractous) to obtain enough RNA as one sample to run in the microarray; thus, creating three sets of pooled samples for both transparent and cataractous lenses. No statistically significant differences existed among the three pooled samples in terms of donor age (P > 0.05, Independent Sample t-test), and one microarray was used for each set of pooled samples (see Supplementary Material and Supplementary Fig. S1A). After the microarray analysis was performed, the central epithelia from five transparent lenses and five age-related cataract patients were acquired for stem-loop RT-PCR. These epithelia were randomly divided into five groups, with each group consisting of one pair of individual transparent and cataractous lenses. The results of the microarray analysis were confirmed by stem-loop RT-PCR. Five independent experiments were performed on each pair (see Supplementary Material and Supplementary Fig. S1B). 

RNA was extracted from continuous curvilinear capsulorhexis specimens. All tissue samples were cooled with liquid nitrogen and homogenized in TRIzol reagent (Invitrogen, Carlsbad, CA). Total RNA from the samples was then isolated according to the manufacturer's instructions. RNA quality was confirmed by calculating the OD260/280 ratio using absorbances measured using a spectrophotometer (Bio-Rad, West Berkeley, CA), and the integrity of the RNA was verified by agarose electrophoresis. The minimum amount of RNA for a microarray is 1 μg as determined by measuring the absorbance at 260 nm. 

Microarray Analysis

The miRNA expression profiling was performed three separate times for each RNA sample using a Gene Chip-Human Genome Array (HG-U133 Plus 2.0, Affymetrix, Santa Clara, CA) at CapitalBio Corporation in Beijing, China; the Gene Chip microarray service of CapitalBio Corporation is certified by Affymetrix. The microarray processing was performed according to the Gene Chip Expression Analysis Technical Manual (Affymetrix). Briefly, 1 μg of total RNA was used to synthesize double-stranded cDNA, and then biotin-tagged complementary RNA (cRNA) was produced using the MessageAmp II aRNA Amplification Kit (Ambion, Austin, TX). The resulting biotin-tagged cRNA was fragmented to strands of 35 to 200 bases in length according to protocols supplied by Affymetrix. The fragmented cRNA was then hybridized to the microarray. Hybridization was performed at 45°C with rotation for 16 hours in the Affymetrix GeneChip Hybridization Oven 640 (Affymetrix). The GeneChip arrays were washed and then stained with streptavidin-phycoerythrin on an Affymetrix Fluidics Station 450 (Affymetrix) followed by scanning on a GeneChip Scanner 3000 (Affymetrix). 

Data Processing and Analysis

The hybridization data was analyzed using GeneChip Operating software (GCOS 1.4; Affymetrix). The scanned images were first assessed by visual inspection and then analyzed to generate raw data files (saved as CEL files using the default setting of GCOS 1.4). An invariant set normalization procedure was performed to normalize the different arrays using a DNA-chip analyzer (dChip). Two hundred and six miRNAs were detected in both the transparent and cataractous samples. 

To perform a comparison analysis, the authors applied a two-class unpaired method using Significant Analysis of Microarray software (SAM) (CapitalBio Corporation, Beijing, China) to identify miRNAs that exhibited significant differences in expression between transparent and cataractous samples. The microarray assays were repeated three times, and the miRNA expression appeared to vary greatly among the assays. The authors first screened the results for miRNAs that had false discovery rate (FDR)-corrected P values less than 0.05 among the three samples. The expression levels of the miRNAs, which were identified using this method, varied only slightly among the assays. The authors next screened for miRNAs that exhibited a 2-fold change in expression levels. Thirty-two miRNAs fit the selection criteria, and these miRNAs were validated by an Independent Samples t-test to confirm the statistical significance. 

Real-Time Quantification of MicroRNAs by Stem-Loop RT-PCR

The results of the microarray analysis were subsequently confirmed by stem-loop RT-PCR. Stem-loop RT-PCR is a highly sensitive RT-PCR method for the detection and quantification of miRNAs; this method was previously described by Erika Varkonyi-Gasic et al., ¹³ and has also been performed extensively in the authors' laboratory. The authors used three primers: a stem-loop specific RT primer, a miRNA specific forward primer, and a universal reverse primer (Table 1). This method is a two-step process. First, total RNA was extracted from the transparent and cataractous lens samples using TRIzol reagent according to the manufacturer's instructions. Total RNA (1000 ng) hybridized with the stem-loop RT primer was reverse transcribed for 45 minutes at 37°C, and for 5 seconds at 85°C using the PrimeScript RT reagent Kit (TaKaRa, Dalian, China) using the SYBR Green I assay (TaKaRa, Dalian, China). The results were obtained using the comparative threshold cycle method and were normalized to β-actin, which served as an internal control. The amplification conditions used were as follows: 3 minutes at 95°C; 39 cycles of 15 seconds at 95°C; 30 seconds at 60°C; and ramp up from 65°C to 95°C at 0.5°C intervals for 5 seconds each. All PCRs were performed in triplicate. The stem-loop RT-PCR was run five times for each sample (pooled central epithelium of five individual transparent and cataractous human lenses) and calculated the mean values for all samples. 

Table 1. 

 

Primer Sequences Used for Stem-Loop RT-PCR

miRNA Name	miRNA Sequence (5′–3′)	Stem-loop RT Primer (5′–3′)	Forward Primer (5′–3′)
miR-105	UCAAAUGCUCAGACUCCUGUGGU	*ACCACA	†TCAAATGCTCAGACT
miR-1207-5P	UGGCAGGGAGGCUGGGAGGGG	*CCCCTC	†TGGCAGGGAGGCT
miR-933	UGUGCGCAGGGAGACCUCUCCC	*GGGAGA	†TGTGCGCAGGGAGA
miR-145	GUCCAGUUUUCCCAGGAAUCCCU	*AGGGAT	†GTCCAGTTTTCCCAG
miR-22	AAGCUGCCAGUUGAAGAACUGU	*ACAGTT	†AAGCTGCCAGTTGA
miR-143	UGAGAUGAAGCACUGUAGCUC	*GAGCTA	†TGAGATGAAGCAC
miR-339-5p	UCCCUGUCCUCCAGGAGCUCACG	*CGTGAG	†TCCCTGTCCTCCAGG
miR-34a	UGGCAGUGUCUUAGCUGGUUGU	*ACAACC	†TGGCAGTGTCTTAG
miR-421	AUCAACAGACAUUAAUUGGGCGC	*GCGCCC	†ATCAACAGACATTAA
miR-133a	UUUGGUCCCCUUCAACCAGCUG	*CAGCTG	†TTTGGTCCCCTTCA

*  Additional sequence (GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGAC) before the stem-loop RT primer.

†  Additional sequence (CGGCG) before the forward primer.

 Universal reverse primer GTGCAGGGTCCGAGGT (5′–3′).

Bioinformatics Analysis to Identify Possible MicroRNA Target Genes

Three miRNA prediction algorithms were used to predict the miRNAs' target genes. TargetScan predicts the biological targets of miRNAs by searching for the presence of conserved 8mer and 7mer sites that match the seed region of each miRNA. The miRanda algorithm incorporates current biological knowledge of target rules and uses an up-to-date compendium of mammalian miRNAs. The target sites predicted by miRanda are scored for the likelihood of mRNA-induced down-regulation using mirSVR, a regression model that is based on sequence and contextual features of the predicted miRNA to mRNA duplex. Finally, PicTar is an algorithm that identifies miRNA targets and provided details regarding (1) miRNA target predictions in vertebrates, (2) miRNA target predictions in seven Drosophila species, (3) miRNA targets in three nematode species, and (4) human miRNA targets that are not conserved but are co-expressed. These prediction algorithms have been previously cited in the literature. ^14–16  

Statistical Analysis

Differences between the two groups in microarray data were assessed using the FDR-corrected P values less than 0.05 provided by SAM. Differences between the two groups were also estimated with an Independent Samples t-test and a statistical computer program (SPSS 16.0; SPSS, Chicago, IL). A P value less than 0.05 was considered indicative of statistical significance. 

Results

MicroRNA Expression Profiles in the Central Epithelium of Human Lenses

The miRNA expression in transparent and cataractous lenses was initially determined by three separate microarray assays, which identified 206 miRNAs. All miRNAs were ranked according to average expression levels, and the authors found several differences between the expression profiles of transparent and cataractous lenses. The top eight differentially expressed miRNAs included miR-184, let-7b, miR-923, miR-1826, miR-125b, miR-1308, miR-26a, and miR-638 in transparent samples, and miR-184, miR-1826, let-7b/c, miR-24, miR-23b, miR-923, and miR-23a in cataractous lenses (Table 2). The expression levels of miR-184, let-7b, miR-923, and miR-1826 were high in both types of lenses, but the expression levels of these four miRNAs differed between transparent and cataractous samples. In particular, the expression levels of let-7b and miR-923 were significantly higher in transparent samples than in cataractous samples (P < 0.05) (Fig. 1). The level of miR-184, the most abundant miRNA in both transparent and cataractous samples, was not altered by the status of the lens, and exhibited an elevated level of expression relative to the levels of other miRNAs in the samples tested. 

Table 2. 

 

Average Expression Levels and SD of the Top Eight miRNAs in Transparent and Cataractous Samples

miRNA	Transparent Samples (Expression level/SD)	miRNA	Cataractous Samples (Expression level/SD)
miR-184	11,565/774	miR-184	11,572/1619
let-7b	4864/227	miR-1826	4673/222
miR-923	4635/217	let-7b	3101/261
miR-1826	4476/446	let-7c	2922/241
miR-125b	3388/208	miR-24	2741/280
miR-1308	2908/236	miR-23b	2556/131
miR-26a	2796/59	miR-923	2454/88
miR-638	2753/93	miR-23a	1579/162

Figure 1. 

 

Relative expression levels of miR-184, let-7b, miR-923, and miR-1826, which were highly expressed in both transparent and cataractous samples. The values (relative expression level) represent the mean ± SD of three independent microarray assays. The miRNAs examined are shown on the x-axis. In cataractous samples, the expression levels of let-7b and miR-923 were significantly higher than in cataractous samples (#P < 0.05 versus cataractous samples). MiR-184 exhibited elevated expression levels in both samples relative to the levels of the other miRNAs.

Differential Expression of MicroRNAs between Transparent and Cataractous Samples

The authors identified 206 miRNAs in the microarray analysis that exhibited distinct and varying expression levels between the two profiles (Fig. 2). They subsequently analyzed three separate microarrays and used an Independent Samples t-test. A 2-fold expression difference (FDR-corrected P values < 0.05) was used as a cutoff level. The results of these analyses demonstrated that 32 miRNAs (15.53%) were significantly differentially expressed between the transparent and cataractous samples. It was found that 20 (62.5%) of these 32 miRNAs were up regulated in transparent samples. The greatest change in expression was exhibited by miR-933, the level of which was approximately 19 times greater in transparent samples than in cataractous samples (Fig. 3). Twelve miRNAs that were down regulated in transparent samples were identified, miR-34a exhibited the greatest decrease in expression; its expression in transparent samples was approximately 4.65 times lower than in cataractous samples. 

Figure 2. 

 

Microarray assays were used to determine the genome-wide miRNA expression in the central epithelium of transparent and cataractous human lenses. The authors performed three separate microarray assays for each group. Colors indicate the expression relative to the mean for each miRNA. The blue color in the transparent samples indicates a lower expression level relative to the level in cataractous samples, and the yellow color indicates a higher expression level in the transparent samples than in the cataractous samples. Fold differences correspond to shading of the color. The expression levels appeared to vary greatly among the individual assays, and, therefore, the FDR-corrected P values (< 0.05) for the three assays were used to avoid error.

Figure 3. 

 

Fold change in miRNA expression between the central epithelium of transparent and cataractous lenses. The expression levels of 32 miRNAs exhibited a change greater than 2-fold. The fold change, which is shown on the x-axis, ranged from −4.65 to −2.00 and from 2.00 to 19.85, the miRNAs screened are shown on the y-axis. A negative number (–) indicates that the expression levels in the transparent samples were lower than those in cataractous samples, and a positive number indicates that the expression levels in transparent samples were higher than those in cataractous samples. MiR-933 exhibited the greatest increase in expression, whereas miR-34a exhibited the greatest decrease in expression.

Verification of Altered MicroRNA Expression Levels by Stem-Loop RT-PCR

The miRNA expression profiles were first identified using microarray assays. As mentioned above, 32 miRNAs exhibited statistically significant differential expression between transparent and cataractous lenses. The authors then chose 10 of these 32 miRNAs, including those with the greatest differential expression (miR-933 and miR-145), and confirmed the results of the microarray using stem-loop RT-PCR. The primers were designed according to the methods described previously (Table 1). Five independent stem-loop RT-PCR experiments for each pair of individual transparent and cataractous samples were repeated to obtain mean expression levels. The results were normalized using β-actin as an internal control. The expression levels of several miRNAs (e.g., miR-933, miR-1207-5p, and miR-133a) were significantly higher in transparent samples than in cataractous samples (P < 0.05). Other miRNAs, such as miR-34a and 339-5p, exhibited significantly lower expression levels in transparent samples than in cataractous samples (P < 0.05). The miR-933 miRNA also exhibited the greatest fold change in the stem-loop RT-PCR assay (the four miRNAs that exhibited the greatest change are shown in Fig. 4; for the other six miRNAs see Supplementary Material and Supplementary Fig. S2). PCR data for all detected miRNAs exhibited high concordance with the microarray data. 

Figure 4. 

 

The relative expression levels of miRNAs were confirmed by stem-loop RT-PCR. The values shown represent the mean ± SD of five independent experiments. These four miRNAs exhibited significantly higher expression levels in transparent samples than in cataractous samples (#P < 0.05 versus cataractous samples). The PCR data exhibited high concordance with the microarray data.

Prediction of Target Genes

Candidate target genes for several up regulated or down regulated miRNAs were identified using three commonly used prediction algorithms, TargetScan, miRanda, and PicTar, as described above in the methods section. Each prediction algorithm was established according to the principle that the 5′ region of the miRNA would pair with the 3′ UTR of target mRNAs of genes. The heteroduplexes formed between several miRNAs and their target mRNAs are shown in Figure 5 as examples. The authors found that one miRNA may regulate hundreds of genes based on the results of the prediction programs, and one gene may be regulated by several miRNAs. They then individually screened the entire set of putative target genes using the National Center for Biotechnology Information (NCBI) databases. The analyses showed that several screened genes appeared to correlate with lens development, aging, or disorders (e.g., cataracts), and may be involved in signal pathways for cell growth, apoptosis, cell proliferation, or the regulation of the cell cycle. The authors identified one possible target sequence for miR-933 in brain-derived neurotrophic factor (BDNF), and possible targets sequences for miR-145 in EYA3/4, MEIS1 and E2F3 genes, and for miR-204 in the MEIS1 gene (Fig. 5). The miRNAs and their putative target genes that correlated with lenses are presented in Table 3. The position of the heteroduplex and references detailing these genes' functions are also shown.  

Figure 5. 

 

Heteroduplexes formed between miRNAs and their target mRNAs. The 5′ region of the miRNA pairs with the 3′ UTR of mRNAs through 7- to 8bp complementary sequences starting 1 or 2 bps from the 5′ end of miR-933, miR-145, and miR-204. Complementary regions were identified in the 3′ UTRs of BDNF, EYA4, and MEIS1.

Table 3. 

 

miRNAs and Their Putative Target Genes Predicted by TargetScan, miRanda, and PicTar

miRNA Name	Putative Gene	Accession	Position of Heteroduplex	References Detailing the Functions of These Genes in Lenses
hsa-miR-933	BDNF	NM_001709	193–199	Cao et al. 17
hsa-miR-339-5P	PITX3	NM_005029	101–108, 157–163, 247–253	Dutta et al. 18
hsa-miR-143	FGF1	NM_000800	1544–1550	Del Rio-Tsonis et al. 19
hsa-miR-34a	E2F3	NM_001949	2714–2720	Chen et al. 21
	NOTCH1	NM_017617	1146–1152, 1232–1238, 1818–1826	Rowan et al. 23
hsa-miR-22	MEIS2	NM_002399	775–781	Heine et al. 24
	BDNF	NM_170735	1019–1025, 2807–2814	Cao et al. 17
hsa-miR-421	MEIS2	NM_002399	323–329	Heine et al. 24
hsa-miR-133a	CRYGS	NM_017541	34–40, 154–160	Liu et al. 25
	MEIS1	NM_170675	226–232, 646–652	Erickson et al. 24
	PITX3	NM_005029	250–256, 265–271	Dutta et al. 18
hsa-miR-105	PAX6	NM_000280	267–273	Hanson et al. 26
hsa-miR-145	MEIS1	NM_002398	1238–1244	Erickson et al. 24
	E2F3	NM_001949	2013–2019	Chen et al. 21

Discussion

The miRNAs have recently emerged as a prominent class of gene regulators. Although miRNAs have been identified in many types of mammalian tissues and organs, including eyes, their transcriptome of human lenses remains unknown. This article is the first description of the identification of miRNAs in the central epithelium of both transparent and cataractous human lenses. In the current study, the miRNA expression profiles in human lenses were identified through the use of microarrays, and detected 206 miRNAs in RNA samples sourced from the central epithelium of transparent and cataractous lenses. When ranking expression levels of the top eight miRNAs in each profile, the authors found that four of the eight miRNAs were differentially expressed in transparent versus cataractous lenses. The four miRNAs, which relative expression levels were high in both of the two profiles, were miR-184, let-7b, miR-923, and miR-1826 (Fig. 1). In particular, the relative expression level of miR-184 was greatly increased over the relative expression levels of the other miRNAs (Table 2). According to previously published research, miR-184 was first reported in mammalian cells in 2006, where high expression was observed in the corneal epithelium in a mouse model. ⁶ Subsequent work also indicated that miR-184 was related to avascularity of the cornea. ²⁷ Lenses play important roles in transmitting light to the retina, which is required for transparency. The authors hypothesize that high expression of miR-184 in lens samples contributes to the maintenance of avascularity and transparency in human lenses. Recently, Hughes et al. ²⁸ reported that a mutation in the seed region of miR-184 is also responsible for severe familial keratoconus combined with early-onset anterior polar cataracts. The authors' research effectively confirmed that miR-184 is essential for the maintenance of transparent lenses. 

The fact that miRNA expression profiles differ between the central epithelium of transparent and cataractous lens samples suggests that the miRNAs that are differentially expressed may play roles in lenticular development and cataractogenesis. The authors identified 32 miRNAs with a statistically significant fold change (> 2-fold changes) in expression. The miRNA expression levels of 20 (62.5%) of these 32 mRNAs were greater in transparent lenses than in cataractous lenses, and 12 miRNAs (37.5%) exhibited decreased expression in transparent samples (Fig. 3). These results suggest that the global expression of miRNAs in transparent lenses is increased relative to the expression levels in cataractous lenses. These findings were additionally confirmed by stem-loop RT-PCR. Conte et al. ²⁹ reported that miR-204 controlled lens cell differentiation by modulating the expression of lens placode differentiation genes in early embryogenesis, thereby revealing the function of miRNAs in lenticular development. The authors' data revealed that the expression level of miR-204 in transparent lenses was 2.31 times greater than those in cataractous lenses; this altered expression provides avenues for further research into the function of miRNAs in lenticular senescence. 

The prediction of target genes is a key step toward understanding the function of specific miRNAs. Several prediction algorithms have been developed based on the principle that the 5′ region of the miRNA pairs with the 3′ UTR of targeted mRNAs to produce a marked effect on mRNA translation. Previous research into prediction algorithms revealed many potential miRNA/mRNA binding pairs, which were subsequently confirmed by additional experimental data. ^14–16,30 In the current study, three different algorithms were used to predict the target genes of miRNAs that exhibited a statistically significant change in expression between transparent and cataractous lenses. The results of each of the algorithms were then pooled together. The target genes that were involved in lens development, aging, or disorders (e.g., cataracts) were selected for additional study based on information obtained from the NCBI databases (Table 3). Previously reported data demonstrates that age-related nuclear cataracts may result from changes in a single major gene, ^31,32 and candidate genes have been identified to have altered expression in nuclear cataracts. ^33,34 The authors hypothesize that the genes found to be altered in their screen may become a single major gene because these genes are associated with changes in lens development, lens cell proliferation, differentiation, or apoptosis. For example, PAX6 is considered to be one of the most important genes in lens development and is a putative target of miR-105 (Table 3). Additionally, previous research indicates that ectopic PAX6 expression disturbed lens fiber cell differentiation and induced cataract formation. ^35,36 The gamma-S crystallin gene (CRYGS), has also been predicted to bind with miR-133a (Table 3). CRYGS is generally accepted as an essential gene for the maintenance of lens transparency, as a decrease in CRYGS expression leads to cataract formation. ³³ Another gene, E2F3, which is targeted by miR-145 and miR-34a, is known to be critical regulators of the cell cycle. E2F3 can induce quiescent lens cells to re-enter the cell cycle, and inappropriate cell cycle re-entry can be accompanied by programmed cell death. However, E2F3 can also induce lens epithelium cells to exit the cell cycle and to differentiate into fibrous cells, promoting cataractogenesis. ^20–22  

The authors' study established the relationship between miRNAs and their target genes in transparent and age-related nuclear cataract lens samples. They also identified putative target genes for the distinct expression of miRNAs. Additional target genes were detected in transparent and cataractous samples by real-time RT-PCR (see Supplementary Material and Supplementary Fig. S3) and it was found that the expression levels of certain miRNAs were higher in cataractous samples with a corresponding decrease in expression of the target mRNAs. However, whether or not the decreased expression of target mRNAs is a result of the increased expression of miRNAs remains to be seen. The authors also failed to establish whether the altered miRNA expression profile is a cause or effect of cataract formation. The changes observed in this study could be the result of a biological compensatory mechanism in response to the disease pathology in the central epithelium. The samples studied are a limited model, as they would only exhibit obvious pathology in the form of a nuclear cataract. 

In conclusion, this study was conducted to investigate miRNA expression profiles in human lens samples. Of the 206 miRNAs identified with microarray assays, 32 (15.53%) exhibited greater than 2-fold changes in relative expression between the two types of samples. The expression of most miRNAs was found to be similar between transparent and cataractous human lenses. This study is the first to date that different expression profiles of miRNAs in the central epithelium of transparent and age-related cataractous human lenses. The miRNAs that had significantly different expression levels between the two lens types were selected to focus the research on miRNAs which were specifically related to lens diseases. Additionally, special algorithms were adopted to analyze miRNAs and to predict their target genes. The miRNAs and their target genes that were predicted in this study have yet to be studied in vivo or in vitro to confirm the presence of any regulatory relationships. 

Supplementary Materials

doc S1 

Footnotes

 Supported by grants from the Guangdong Recruitment Program of Creative Research Groups, the Natural Science Foundation of China (30972620, 81101255, 81000389), the National Basic Research Program of China (973 Program) (2010CB912200), the Key Projects for Hospital Clinical Disciplines of Ministry of Health of China (2010-2012-175), the Guangdong Natural Science Foundation (9251008901000022, 10451008901004204), the Guangdong Technology Programs (2009B080701017, 2008B060600026, 2009B060700089), the Open Research Program (303060202400309), and the Fundamental Research Funds of the State Key Laboratory of Ophthalmology.

Footnotes

 Disclosure: C. Wu, None; H. Lin, None; Q. Wang, None; W. Chen, None; H. Luo, None; W. Chen, None; H. Zhang, None

Footnotes

³  These authors contributed equally to this work and should therefore should be considered equivalent authors.