Epigenetic modifications in hyperhomocysteinemia: potential role in diabetic retinopathy and age-related macular degeneration

To study Hyperhomocysteinemia (HHcy)-induced epigenetic modifications as potential mechanisms of blood retinal barrier (BRB) dysfunction, retinas isolated from three- week-old mice with elevated level of Homocysteine (Hcy) due to lack of the enzyme cystathionine β-synthase (cbs–/–, cbs+/– and cbs+/+), human retinal endothelial cells (HRECs), and human retinal pigmented epithelial cells (ARPE-19) treated with or without Hcy were evaluated for (1) histone deacetylases (HDAC), (2) DNA methylation (DNMT), and (3) miRNA analysis. Differentially expressed miRNAs in mice with HHcy were further compared with miRNA analysis of diabetic mice retinas (STZ) and miRNAs within the exosomes released from Hcy-treated RPEs. Differentially expressed miRNAs were further evaluated for predicted target genes and associated pathways using Ingenuity Pathway Analysis. HHcy significantly increased HDAC and DNMT activity in HRECs, ARPE-19, and cbs mice retinas, whereas inhibition of HDAC and DNMT decreased Hcy-induced BRB dysfunction. MiRNA profiling detected 127 miRNAs in cbs+/– and 39 miRNAs in cbs–/– mice retinas, which were significantly differentially expressed compared to cbs+/+. MiRNA pathway analysis showed their involvement in HDAC and DNMT activation, endoplasmic reticulum (ER), and oxidative stresses, inflammation, hypoxia, and angiogenesis pathways. Hcy-induced epigenetic modifications may be involved in retinopathies associated with HHcy, such as age-related macular degeneration and diabetic retinopathy.


INTRODUCTION
Homocysteine (Hcy) is an amino acid containing a sulfur group and is a key component of the methionine recycle system. Hcy plays a fundamental role in methyldonor biosynthesis, which is biochemically linked to the principal epigenetic code found in DNA [1]. Our previous studies have reported that elevated Hcy, known as hyperhomocysteinemia (HHcy), disrupts inner [2,3] and outer [4] blood-retinal barrier (BRB) integrity. The presence of an intact BRB is critical for maintaining the structural and functional integrity of the retina. The breakdown of BRB is a hallmark of the pathogenesis and development of retinal diseases, such as diabetic retinopathy (DR) [5][6][7] or age-related macular degeneration (AMD) [8][9][10]. Over the last decade, HHcy has received special attention to its relation to DR and AMD in several clinical and epidemiological studies, suggesting an association between elevated levels of serum Hcy and the risk of DR [11][12][13][14][15][16] and AMD [17][18][19][20][21][22].
Research Paper: Gerotarget (Focus on Aging) www.impactjournals.com/oncotarget However, the underlying mechanism of HHcy-induced BRB dysfunction is not clearly defined.
Epigenetic modifications are defined as changes in phenotype and gene expression that take place without alterations of DNA sequence [35]. Recently, epigenetic modifications have been highlighted as important mechanisms for the development of number of diseases [36][37][38].There are three major epigenetic modification mechanisms:(1) DNA methylation, which is the most well-known epigenetic modification and plays an essential role in controlling global and specific gene expression. It is mediated by DNA methyl transferases (DNMTs) [39,40], (2) Histone modification, which is defined as changes occurring in histone proteins that will affect gene expression.
Chromatin is the complex of chromosomal DNA and proteins that forms chromosomes in the nucleus. DNA is packaged in chromatin by wrapping around histone proteins in units called nucleosomes. Histone proteins are highly alkaline, with Histone tails that are normally positively charged due to presence of amine groups on their basic lysine and arginine amino acids [41,42]. These positive charges on histone tails help histone proteins to interact with and bind to the DNA backbone at the negatively charged phosphate groups. During the acetylation process, amines on the histone change into amides; thereby, positive charges on the histone become neutralized and subsequently the capability of the histones to bind with DNA is reduced [43,44]. This decreased binding results in expansion of chromatin, which permits genetic transcription to occur. The Histone deacetylation is catalyzed when histone deacetylases remove those acetyl groups, which increase the positively charged amine groups of histone tails and facilitate high-affinity binding between the histones and the negatively charged phosphate groups on the DNA backbone. The increased binding of DNA to histone condenses the DNA structure, interfering with the DNA transcription process [45].
Lastly, (3) non-coding RNA regulation (miRNAs), or microRNAs, comprise a unique class of small (~22) groups of nucleotides, endogenous non-coding RNAs that play a crucial role in regulating gene expression by silencing their target mRNAs [46][47][48]. MiRNAs can represent tools for the regulation of gene expression not only within the cell in which they are produced but also in the gene expression of other remote cells [49][50][51]. MiRNAs can be released from their cells of origin and circulate to target different cells in different tissues. One of the most interesting carriers of miRNAs are extracellular vesicles.
This study explores epigenetic modifications as possible mechanisms of HHcy-induced BRB dysfunction. Targeting HHcy-induced epigenetic modifications could provide therapeutic targets to mitigate Hcy-mediated BRB dysfunction in retinal diseases associated with elevated Hcy, such as DR and AMD.

Animal studies
Animal protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at Augusta University. All experimental procedures done with animals were performed according to guidelines accepted by the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. Cystathionine beta-synthase (cbs) deficient mice were generated as described before [63]. Colonies of cbs +/+ , cbs +/− , and cbs −/− were established by breeding pairs of cbs +/− mice (B6.129P2-Cbstm1Unc/J; Jackson Laboratories, Bar Harbor, ME).

Assessment of DNA methyltransferase (DNMT) activity
Nuclear extracts were prepared from retinas of cbs +/+ , cbs +/and cbs -/mice (3 weeks old), HRECs and ARPE-19 treated with Hcy (20, 50 and 100 µM) using a nuclear extraction kit purchased from Abcam (Abcam Inc., Cambridge, MA, USA). Following extraction DNMT activity was assayed calorimetrically according to manufacturer's instructions (Cat# P-3009-48, Epigentek, Farmingdale, NY, USA). Briefly, nuclear extracts were added to a microplate coated with a universal DNMT substrate in presence of adenosyl methionine as methyl donor. The produced methylated DNA which is proportional to the DNMT activity was captured using anti-methyl cytosine antibody and then detected using detection antibody through an ELISA-like reaction. The color produced was measured at 450 nm and its density was proportional to DNMT activity.

Assessment of HDAC activity
HDAC activity was assayed by using a colorimetric assay kit (Catalog #K331-100, BioVision, CA, USA). Retinal nuclear extracts prepared from cbs +/+ , cbs +/and cbs -/mice and Hcy-treated HRECs and ARPE-19 (20, 50 and 100 µM) were incubated with a substrate containing acetylated Lysine side chain for 1 h at 37° C. Following deacetylation by sample HDAC, the reaction was terminated by adding Lysine Developer to produce a chromophore that was measured spectrophotometrically at 400 nm.

Optical coherence tomography (OCT) and fluorescein angiography (FA)
To evaluate the effect of combined Inhibition of DNA methylation and histone deacetylation on Hcyinduced retinal disruption on 24-week-old mice injected intravitreally with Hcy with and without repeated intraperitoneal injection of combined inhibitors of DNA methylation and histone deacetylation. OCT and FA were performed simultaneously after 48 and 72 hours from Hcy injection, as described in our pervious publication [2,4]. Briefly, mice were injected intravitreally with Hcy (200 μM) with and without intraperitoneal injection of combined inhibitor of HDAC, Sodium butyrate (SB, 1mg/kg, catalog # B5887-250MG sigma Aldrich) and DNMT inhibitor, 5-Azacytidine (5-AZC,2.5 mg/kg, catalog # A2385 SIGMA) [64][65][66][67]. The mice were anesthetized using 2% isoflurane and their eyes were dilated using 1% tropicamide eye drop. Each mouse was then placed on the imaging platform of the Phoenix Micron III retinal imaging microscope supplemented with OCT imaging device (Phoenix Research Laboratories, Pleasanton, CA). To keep the eye moist during imaging, lubricant gel was applied 20 μL 10% fluorescein sodium (Apollo Ophthalmics, Newport Beach, CA) were injected into the mice intraperitoneal, followed by rapid acquisition of fluorescent images ensued for ~5 minutes. Fluorescein leakage was demonstrated as indistinct vascular borders progressing to diffusely hazy fluorescence.

RNA isolation, preparation and analysis
Retinas of age matched cbs +/-, cbs -/and WT mice were enucleated and total RNA was extracted using Ambion, TRIzol Reagent (Life Technologies). RNA purity and concentration were assessed by spectrophotometry using Nano Drop ND-1000 (Thermo Fisher).

Microarray analysis
A total 250 ng of RNA was labeled with biotin using the Flash Tag Biotin HSR RNA Labeling Kit (Affymetrix, Santa Clara, CA) according to the manufacturer's protocol. The labeled samples were then hybridized into the Gene Chip miRNA 3.0 array (Affymetrix). Hybridization, washing, and scanning of the arrays were carried out according to Affymetrix's recommendations. The data was imported into the Partek Genomic Suites version 6.6 (Partek, St. Louis, MO). Principal component analysis (PCA) was performed to visualize the partition among the groups. The differential expressions were calculated by using ANOVA of the Partek Package and filtered with a p-value cutoff of 0.05 and a fold-change cutoff 1.5-foldto screen highly significant miRNAs. The significant miRNA lists were used to generate hierarchical clustering plots.

Bioinformatics analysis of the data
MiRNA lists were imported into Ingenuity Pathway Analysis (Qiagen) and analyzed on MicroRNA Target Filter and Core Analysis for detection of predicted target genes of miRNAs and associated pathways.

RT-PCR
RT-PCR was performed to validate the differential expression of selected miRNAs. RNA was extracted from retinas of cbs +/mice and their age matched controls. RT-PCR was performed for miRNAs in these samples using appropriate miScript primers obtained from Qiagen. The following primers were used: Mm_miR-200c_1 miScript Primer Assay (MS00001827), Mm_miR-205_1 miScript Primer Assay (MS00001862), Mm_miR-199a-3p_1

Exosmoes isolation from cell culture media
Conditional culture media was collected, and then centrifuged for 30 minutes at 2000 × g to remove any cells or debris. Then, exosomes were isolated using Invitrogen Total Exosome Isolation Reagent (from cell culture media) (Catalog#: 4478359) according to the manufacturer's protocol. Briefly, 0.5 volumes of the exosome isolation reagent was added to the media then vortexing was done to mix the reagent with the media. The mixture was left overnight at 4° C. Next day, the mixture was centrifuged at 10,000 × g for an hour at 4° C. Supernatants were aspirated and pellets were suspended again in an appropriate PBS volume.

Zeta view nanoparticle tracking analysis (NTA)
The size and concentration of the exosomes isolated were quantified by using NTA which was done utilizing the ZetaView PMX 110 (Particle Metrix, Meerbusch, Germany) and its related software (ZetaView 8.02.28) [68]. Each sample was measured at 11 different positions, and then the size and concentration of each sample were obtained. ZetaView 8.02.28 software was used for the analysis of measurement data from the ZetaView.

Exosomal RNA isolation and measurement
RNA isolation was performed using the Total Exosome RNA & Protein Isolation Kit (catalog # 4478545; Invitrogen) according to the manufacturer's protocol. A final volume of 30 μl RNA solution was collected from each. Agilent 2100 Bioanalyzer (Santa Clara, CA) was used for measuring RNA quality and concentration at the Integrated Genomics Core of Georgia Cancer Center at Augusta University.

Statistical analysis
Different experimental groups were compared by using the student two-tailed t test or one-way analysis of variance (ANOVA), with a post hoc Tukey's test. Data considered statistically significant when P value < 0.05. The outcomes are expressed as mean ± SD. For the microarray data analysis; the fold change was calculated using least squares means of the groups.

Hcy increases DNA methylation and histone deacetylation in vivo and in vitro
Experiments were performed (in vivo and in vitro) to evaluate the effect of HHcy on the activities of DNMT enzymes ( Figure 1A) and HDAC enzymes ( Figure 1B), as key enzymes of DNA methylation and histone deacetylation respectively. In vitro experiments using HRECs and ARPE-19 treated with Hcy showed a significant increase of DNMT and HDAC activities in a dose-dependent manner. This was consistent with the in vivo effect of HHcy, where both cbs +/and cbs -/mice retina showed marked increase of the activity of DNMT enzymes as well as HDAC enzymes compared to the normal control group.

Inhibition of DNA methylation and histone deacetylation diminishes HHcy effect on mice retina
Optical coherence tomography (OCT) and fluorescein angiography (FA) examinations of 24 weeks living mice injected intravitreally with Hcy with and without intraperitoneal injection of combined inhibitors of DNA methylation (DNMT inhibitor, 5-Azacytidine) and histone deacetylation (HDAC inhibitor, Sodium butyrate) showed changes in the FA and OCT examination in the mice injected with Hcy similar to what has been reported before in our previous publication [2,4]. FA revealed diffuse hyper fluorescence, leakage and focal areas of hyper-fluorescence coming from the deeper layer (yellow arrows) suggestive of choroidal neovascularization (CNV) and the OCT examination revealed disruption of the retinal morphology with RPE changes with the development of CNV (white arrows), and intra-retinal neovascularization (arrow heads). Interestingly, combined injection of Hcy and HDAC/DNMT inhibitors protected the retina from the distributing effect of Hcy alone on mice retina as shown in both FA and OCT ( Figure 2)

Differential expression profile of miRNAs in cbs -/and cbs +/mice retina
In order to identify the impact of severely elevated Hcy levels on the retinal miRNA profile, a microarray analysis was done for retinas of cbs -/mice compared to control cbs +/+ retinas at 3 weeks old ( Figure 3A-3C). The data was filtered by using p value less than 0.05 which reveals 217 miRNAs significantly differentially expressed versus cbs +/+ (p value < 0.05) of which 124 were downregulated and 93 were upregulated. From these statistically significant miRNAs, 39 miRNAs were more than 1.5-fold change, 20 miRNAs of them were downregulated (Table 1A) and 19 were upregulated (Table 1B). In order to identify the impact of mild to moderate elevation of Hcy levels on the retinal miRNA profile, miRNA microarray analysis was done for the cbs +/mice retina compared to the cbs +/+ control ( Figure 3D-3F). The differential miRNA expression profiling of the cbs +/retinas showed 307 statistically significant and differentially expressed miRNAs compared to the cbs +/+ control (157 miRNAs were downregulated and 150 were upregulated). Further filtration of these miRNAs based on the fold change cutoff more than 1.5, revealed that 127 miRNAs were significant with fold change more than 1.5; of which 49 were downregulated (Table 2A) and 78 were upregulated (Table 2B).

Comparison of the common differentially expressed miRNAs in cbs -/-& cbs +/and WT& diabetic (STZ) mouse retina
In order to explore miRNAs that are commonly changed with either complete or partial deletion of the cbs enzyme, data was loaded into the IPA (Ingenuity Pathway Analysis) for comparison analysis. The comparison of the 2 data sets showed the presence of 25 miRNAs that were commonly changed in both cbs -/or cbs +/mouse retina. Out of these 25 miRNAs, 18 miRNAs were differentially expressed in a consistent manner between the 2 groups ( Figure 4A, highlighted); 8 miRNAs were downregulated in both groups (miR-16, miR-200, miR-205, miR-3064, miR-379, miR-431, miR-485 and miR-491) and 10 miRNAs were upregulated in both groups (miR-194, miR-1894, miR-211, miR-3072, miR-3077, miR-4436, miR-5128, miR-669a, miR-669c and miR-6967). Microarray data results were validated by subjecting retinal RNA samples from cbs +/and cbs +/+ mice to RT-PCR analysis ( Figure 4B). Consistently with the microarray results, miR-205 (p value = 0.001), miR-206 (p value = 0.01) and miR-27 (p value = 0.04) were significantly downregulated in cbs +/compared to control cbs +/+ (p value < 0.05). MiR-29a showed a tendency of downregulation but it did not reach the statistical significance. However, miR-199 (p value = 0.01) and miR-200 (p value = 0.004) in contrary to the microarray results were significantly upregulated in cbs +/in comparison with the control cbs +/+ (p value < 0.05). In contrast, miR-31 was upregulated in the microarray data but was downregulated with RT-PCR validation (p value = 0.04). U6 was used as an internal control.

Differential expression profile of retinal miRNAs during diabetes
We aimed to detect the differential expression of retinal miRNAs under the influence of diabetes ( Figure 4C-4E).

Figure 2: Effect of inhibition of DNA methylation and histone deacetylation on Hcy-induced BRB dysfunction.
OCT and FA examinations of 24-week-old mice injected intravitreally with Hcy with and without intraperitoneal injection of combined inhibitors of DNA methylation (DNMT inhibitor, 5-Azacytidine) and histone deacetylation (HDAC inhibitor, Sodium butyrate). The FA and OCT examination in the mice injected with Hcy revealed diffuse hyper fluorescence, leakage and focal areas of hyper-fluorescence coming from the deeper layer (yellow arrows) suggesting choroidal neovascularization (CNV). OCT examination revealed disruption of the retinal morphology with RPE changes with the development of CNV (white arrows) and intra-retinal neovascularization (arrow heads). Interestingly, combined injection of Hcy and HDAC&DNMT inhibitors protected the retina from the distributing effect of Hcy alone on mice retina as shown in both FA and OCT (A)The leakage was confirmed by measuring the albumin leakage in the retinas by western blotting (B), which was significantly increased in the Hcy-injected mice eyes compared to non-injected eyes and the eyes injected with Hcy along with intraperitoneal injection of combined DNMT&HDAC inhibitors, * p < 0.05.
WT animals were treated with streptozotocin (STZ) 50 mg/kg for five consecutive days. Diabetes was confirmed by measuring blood glucose level (mice with blood glucose level > 250 mg/dl were considered diabetic). Microarray analysis was performed for diabetic retinal RNA in comparison with the non-diabetic controls. 312 miRNAs were found to be changed after 1.5-fold change cutoff and p value < 0.05 (129 miRNAs were downregulated and 183 miRNAs were upregulated).

Shared miRNAs among cbs -/-, cbs +/and diabetic mouse retina
To investigate the similarity in alteration of retinal miRNAs profile between HHcy and diabetes, IPA software was used to compare differential retinal miRNA profiles in cbs +/vs the diabetic retinas ( Figure 5A). The analysis showed that 37 miRNAs were commonly changed in both conditions. In the microarray data, 12 miRNAs were      miR-5128 and miR-669c) that were commonly changed among cbs -/-, cbs +/and diabetes; 2 of these miRNAs were consistently changed among the three groups (miR-194 was upregulated and miR-16 was downregulated).

HHcy induces alteration of miRNAs similar to AMD
We have reported that HHcy induces AMD-like changes in mouse retina [4]. Therefore, the current work aimed to check if HHcy induces changes in miRNAs similar to what have been reported to be implicated in the development of AMD [69]. Interestingly, miRNAs have also been shown to play critical roles in different pathways in a laser-induced choroidal neovascularization mouse model [70]. miR-205, miR-27, miR-29 and miR-31 were significantly changed in our cbs +/retina microarray and were also reported to be involved in AMD. The study revealed downregulation of miR-205, miR-27, miR-31, and miR-29 in the cbs +/retinas. MiR-31 was reported to be significantly decreased in an ischemia-induced mouse model of retinal neovascularization and in a laser-induced mouse model of choroidal neovascularization in the absence of ischemia [71]. While intraocular injection with pre-miR-31 significantly reduced the size of the choroidal neovascular lesions. A recent study also demonstrated that knockdown of miR-27, which downregulates the antiangiogenic factors Sprouty2 and semaphorin 6A (Sema6A), is protective against laser-induced choroidal neovascularization [70].

Differentially expressed miRNAs in exosomes isolated from Hcy-treated ARPE
Exosomes released from RPE represent a very important and innovative way of retinal intercellular communications. The outer blood-retinal barrier integrity is formed by tight junctions among retinal pigment epithelial cells [72,73]. The outer BRB is mainly affected in AMD. To investigate if HHcy induces changes in the miRNA cargo of RPE-released exosomes as a possible early epigenetic modification tool that can impact the retinal barrier function, conditioned media of ARPE-19 cells treated with Hcy 100 µM for 24 hr was collected and exosomes were isolated using the appropriate protocol. The concentration and size of the isolated exosomes were determined using NTA as indicated in the methods section. Total extracted RNA from the isolated exosomes and purity of the extracted RNA was validated using the Bioanalyzer chip (Supplementary Figure 1). Following exosomal RNA isolation, miRNA microarray analysis was performed. The analysis revealed that Hcy altered RPEexosomal-miRNA content. 88 miRNAs were identified (39 miRNAs were downregulated and 49 miRNAs were upregulated) with1.4-fold change cutoff and p < 0.05 ( Figure 5B-5D and Table 3A and 3B)

DISCUSSION
Recently, we reported deleterious effects of elevated Hcy on BRB integrity [2,4,76]. Elevated Hcy level was reported as an epigenetic modulator in various diseases      such as chronic kidney disease [77], cardiovascular diseases [78] and atherosclerosis [79]. In the present study, epigenetic modifications were proposed as possible mechanisms of HHcy-induced BRB dysfunction. Currently, it is believed that this is the first study to screen for the epigenetic changes that occur in retina and retinal endothelial cells under HHcy. The major findings of the current study are 1) Hcy induced a significant increase of both HDACs and DNMTs activity in mouse retina, cultured HRECs and RPE; 2) Inhibition of DNMT and HDAC ameliorates Hcy-induced BRB dysfunction.
3) miRNA profiling of cbs +/and cbs -/detected 127 miRNAs in cbs +/that were statistically significant and differentially expressed in comparison to cbs +/+ .4) We detected similarity between some miRNAs altered under HHcy and miRNAs altered in DR or AMD. 4) miRNA pathway analysis showed HHcy-dysregulated miRNAs involvement in ER stress, oxidative stress, inflammation, angiogenesis, and hypoxia pathways. 5) DNMTs and HDACs are direct targets of miRNAs dysregulated in HHcy. Among different epigenetic pathways, DNA methylation is the major epigenetic modification associated with HHcy. Methylation of DNA is mediated by DNMTs and requires the presence of S-adenosylmethionine (SAM), a methyl donating compound that delivers methyl groups to maintain other metabolic reactions in human body [80]. Hcy is an intermediate product released during methionine cycle. Methionine combines with adenosine to produce SAM. After releasing its methyl group, SAM is converted to SAH which is further hydrolyzed to Hcy and adenosine [81]. Therefore, the Hcy-methionine cycle is very critical for maintenance of a balanced amount of SAM, and therefore normal DNMTs activity. Indeed, differential levels of DNA methylation, from hyper-to hypo-methylation, have been reported to be induced by HHcy [77]. A likely explanation for such phenomena is the fact that other mechanisms are involved in regulating DNA methylation and histone modification. In the current study, the Hcy increased DNMTs activity both in vivo and in vitro, suggesting HHcy-induced global increase in retinal DNA methylation.
Interestingly, epigenetic modifications were reported to play a crucial role in HHcy induced-blood brain barrier dysfunction, with profound involvement of miRNAs in the pathogenesis of the leaky cerebral vasculature in cbs +/mice [90].
To narrow the focus, the differentially expressed miRNAs in both cbs +/and cbs -/mice were compared to identify commonly expressed miRNAs in HHcy. The analysis revealed 25 miRNAs were similarly affected by elevated Hcy in both models. Furthermore, miRNAs were reported to be linked to the pathogenesis of diabetic complications [91]. Therefore, we evaluated that the differentially expressed miRNAs in mouse models of elevated Hcy and in STZ-mice as a model of Type 1 diabetes. The analysis revealed 6 miRNAs commonly differentially expressed in both STZ and HHcy mice. Among those miRNAs, 2 miRNAs (miR-16-5p and miR-194) were consistently changing among the three different groups. Interestingly, miR-16-5p has been reported to be tissue protective and to be decreased in a diabetic rat's kidney [92].
In addition to the reported role of HHcy in DR [15,93,94], our previous work reported that increased Hcy was associated with features of AMD in mice including RPE dysfunction [4]. Therefore, the current study also aimed to evaluate the similarity of miRNAs profile in mouse models of HHcy and what has been reported in patient with AMD, especially various miRNAs were linked to AMD pathogenesis [95]. The study revealed downregulation of miR-205, miR-27, miR-31, and miR-29 in the cbs +/retinas, these miRNAs were also reported to be downregulated in vitreous [68] and plasma of AMD patients [69]. Furthermore, miR-31 which was altered in cbs +/retinas was suggested as a regulator of choroidal neovascularization [70].The analysis revealed similarity in miRNAs differentially expressed in both AMD patients and retina of cbs +/mice.
Recently, studies uncovered evidence that miRNAs travel around the body are contained within exosomes [96]. Increasing evidence points to exosomes and micro vesicles as a unique alternative mechanism for protein transfer and intercellular communication, allowing exchange of proteins, lipids and nucleic acids between cells [97]. Accumulating evidence supported the role of exosomal miRNAs in eye diseases [98,99].
Related to the fact that there is no definite animal model of AMD and numerous models mimic some of the significant pathological features seen in AMD, however none recreated all of its characteristics. That is why, having an animal model that mimics both the early and late features of AMD has been challenging [100]. Our previous work reported that, HHcy induces features of AMD both in vivo (cbs mice) and in vitro (Hcy-treated RPE) [4]. Consequently, we evaluated miRNA differential profile in exosomes released from Hcy-treated RPE. Two miRNAs (MiR-17-5p and let-7a-3p) were found to be commonly differentially expressed in retinas of cbs -/mice, diabetic mice as well as RPE exosomes. Interestingly, those two miRNAs were also reported to be altered in DR and AMD [91,101,102]. Moreover, miR-17-5p has been reported to be involved in both pro-and anti-proliferative signals via dysregulation of normal cell cycle [103].
In conclusion, understanding HHcy-induced epigenetic modifications lead to possible prospects in evaluating miRNAs as molecular biomarkers, prognostic tools, diagnostic tools and therapeutic agents for eye diseases associated with elevated Hcy such as DR and AMD.