Abstract
Purpose:
We investigated the extracellular matrix (ECM) of the lamina cribrosa (LC) and peripapillary sclera (PPS) and compared histone acetylation and related enzymes to identify racial differences between Korean and Caucasian donor eyes.
Methods:
Posterior segment tissues were obtained from 30 Caucasian donors and 42 age and axial length-matched Korean donors. Histone modification was assessed for histone deacetylase (HDAC) 2, HDAC3, and acetylated histone H3. The promoter regions of the major ECM in the LC and PPS including collagen type I and III, and elastic fiber components (elastin and fibrillin-1) and lysyl oxidase enzymes including lysyl oxidase-like 1 and 2 (LOXL2) were evaluated by chromatin immunoprecipitation (ChIP) assay. Protein and mRNA expression of major ECM components were assessed using real-time polymerase chain reaction analysis, western blot analysis, and immunohistochemical staining.
Results:
HDAC2 and HDAC3 expression levels were decreased and acetylated histone H3 was increased in the LC and PPS of Korean eyes than Caucasian eyes. The promoter regions of LOXL2, elastin, and fribrillin-1 genes were highly acetylated in Korean LC. Expression of LOXL2 and elastic fiber components (elastin and fibrillin-1) were significantly increased in Korean LC and PPS than Caucasians according to the real-time polymerase chain reaction, western blot analyses, and quantification of elastic fiber staining.
Conclusions:
Histone acetylation status differed in the promoter regions of the elastic fiber components and LOXL2 in the LC and PPS according to race. Further study to reveal the association with these findings to the pathogenesis of glaucoma in Korean eyes is needed.
Altered extracellular matrix (ECM) is thought to contribute to several ocular conditions.
1 Myopia involves ECM changes in the sclera, leading to eyeball elongation.
2 In pseudoexfoliation syndrome, there are genetic defects in the elastic fibers of the eye and decreased elastic components in the lamina cribrosa (LC), which seem to be related to the development of glaucoma.
3 Elevated intraocular pressure (IOP) in glaucoma results from ECM changes in the trabecular meshwork.
4 Additionally, the LC is thought to be the principle site of optic nerve damage in glaucoma, and changes in ECM components are observed in glaucomatous eyes.
5–9 ECM changes in the peripapillary sclera (PPS) have recently been implicated in glaucoma, and both the PPS and LC are thought to be important in the pathogenesis of glaucoma.
5–13
Major ECM component of the LC and the sclera, elastin, differed between Africans and Europeans, perhaps contributing to the higher prevalence of glaucoma in Africans.
14 Therefore, differences in ECM components in the LC and the sclera may determine the susceptibility to glaucoma. However, the exact mechanism by which the ECM contributes to glaucomatous damage is not well understood. Therefore, an understanding of the baseline characteristics of the LC and PPS is critical for investigating the pathogenesis of glaucoma. Normal-tension glaucoma (NTG), characterized by a relatively normal IOP leading to the same clinical presentation as glaucoma with high IOP, is the predominant type of glaucoma in Asian populations.
15,16 Clinically, the central cornea, anterior and posterior sclera, and LC were thinner in Korean study participants with NTG, and many of these NTG patients have myopia.
17–20 We therefore hypothesized that studying racial differences in ECM may reveal the contribution of the ECM to glaucoma, and the racial differences in the clinical presentation of glaucoma. Epigenetic mechanisms, including DNA methylation of cytosine-phosphate-guanine sites, posttranslational histone modifications, chromatin remodeling, and deployment of noncoding RNA, regulate gene expression, influencing gene function without altering DNA sequences.
21–23 Recent studies have shown that epigenetic regulation of ECM components may contribute to several ocular conditions, including glaucoma and myopia.
24–26 Therefore, we investigated ECM components and modulating enzymes of the LC and the sclera in human donor eyes and identified racial differences between Korean and Caucasian eyes. We compared histone acetylation and related enzymes in the PPS and LC of Korean and Caucasian individuals to determine the epigenetic differences between these races. The study targeted major fiber components of the ECM in the LC and PPS, which mechanically supports these tissues, including collagen type I and III, and elastic fiber components (elastin and fibrillin-1).
27,28
Caucasian donor eyes (n = 30; male to female ratio = 11:19; mean age = 62.26 ± 17.23 years) were obtained for laboratory investigation from the Midwest Eye Bank, and all eyes were processed within 24 hours postmortem. Korean donor eyes (n = 42; male to female ratio = 22:20; mean age = 64.87 ± 21.34 years) used for corneal transplantation were obtained from the Eye Bank of Seoul St. Mary's Hospital. Immediately after enucleation, axial length was measured with a caliper. Since Korean donor eyes were provided more frequently than Caucasian donor eyes, we matched Korean samples to Caucasian samples by age and axial length. The measured mean axial length was 23.5 ± 1.3 mm for Caucasian eyes and 23.6 ± 1.5 mm for Korean eyes. For RNA and protein extraction, posterior segment tissues (retina, choroid, sclera, LC, and retrobulbar optic nerve) were prepared under a dissecting microscope and shock frozen in liquid nitrogen (n = 12 Korean and n = 12 Caucasian eyes). The scleral specimens were subdivided into the PPS, the 2-mm area surrounding the LC, and sclera from the equatorial region. For chromatin immunoprecipitation (ChIP) assay, the LC and PPS tissues were pooled and half of each sample was mixed with either acetylated histone H3 antibody or normal rabbit serum for a control (n = 6 Korean and n = 6 Caucasian eyes). Tissues for immunohistochemical analysis were cryofixed in optimal cutting temperature (OCT) compound (n = 12 Korean and n = 6 Caucasian eyes). Posterior segments from 12 Korean and six Caucasian donor eyes were fixed in 4% paraformaldehyde and embedded in paraffin for histochemical staining for elastin and collagen fibers. Informed consent for tissue donation was obtained from the patients or their relatives, and the study protocols were approved by the Institutional Review Board of the Seoul St. Mary's Hospital and adhered to the tenets of the Declaration of Helsinki for experiments involving human tissues and samples.
Total RNA was extracted from ocular tissues using the RNeasy kit (Qiagen, Hilden, Germany), including an on-column DNase I digestion step. First-strand cDNA synthesis from 0.1 μg of total RNA and quantitative RT-PCR were performed using the MyIQ thermal cycler and software (Bio-Rad Laboratories, Munich, Germany). PCRs (25 μL), run in duplicate, contained 2 μL of first-strand cDNA, 0.4 μM each of upstream and downstream primers (Macrogen, Inc., Seoul, Korea), and IQ SYBR Green Supermix (Bio-Rad Laboratories). Primers and PCR conditions are summarized in the
Table. For quantification, serially diluted standard curves of plasmid-cloned cDNA were run in parallel, and amplification specificity was checked using melt curves and sequence analyses on the CPX96 Real Time PCR Machine (Bio-Rad Laboratories). mRNA ratios relative to the housekeeping gene
GAPDH were calculated for normalizing gene expression levels. For each analyzed tissue, six samples were pooled in each race.
Indirect immunofluorescence single- or double-labeling was performed on cryosections of posterior segment tissues using primary antibodies against lysyl oxidase (LOX; 1:100; Abnova, Taipei, Taiwan), lysyl oxidase-like 1 (LOXL1; 1:50; Abnova), lysyl oxidase-like 2 (LOXL2; 1:50; Abnova), transforming growth factor (TGF)-β (1:100; Abcam, Cambridge, UK), elastin (1:100; Elastic Products Company, Inc., Owensville, MO, USA), and fibrillin-1 (1:100; Chemicon International, Inc., Temecula, CA, USA). To investigate histone modifications, antibodies against histone deacetylase (HDAC) type 2 (1:100; Abcam), HDAC type 3 (1:100; Abcam), and acetylated histone H3 (1:200; Abcam) were used. Cryostat sections (5 μm) were fixed in cold paraformaldehyde (PFA), blocked with 10% normal goat serum, and incubated in primary antibodies diluted in phosphate-buffered saline (PBS) overnight at 4°C. Antibody binding was detected by Alexa 488- and Alexa 546-conjugated secondary antibodies (Molecular Probes, Eugene, OR, USA). After further washing in 0.1 M PBS for 30 minutes, the sections were mounted using VECTASHIELD Mounting Medium with 4′,6-diamidino-2-phenylindole (DAPI) stain (Vector Laboratories, Burlingame, CA, USA). Sections were washed, coverslipped, and examined by confocal laser scanning microscopy (Carl Zeiss AG, Oberkochen, Germany). To quantify each staining, images were converted into gray scale image and then the pixel intensity was measured by threshold analysis of gray values using ImageJ (ImageJ 1.40;
http://rsb.info.nih.gov/ij/index.html; developed by Wayne Rasband, and provided in the public domain by the National Institutes of Health, Bethesda, MD, USA).
29
Acetylated histone H3 ChIP assays were performed using EZ-ChIP Chromatin Immunoprecipitation Kit (Merck Millipore, Billerica, MA, USA) as outlined by the manufacturer. In each assay, six LC and PPS were pooled in each race. In brief, the acetylated histone H3 antibody (1 μg, Abcam) or 1 μL of normal mouse IgG (Abcam) was used to precoat the assay wells. Each LC and PPS tissue was cut into little pieces and cross-linked with 1% formaldehyde. The cross-link was stopped by 1.25 M glycine solution. After tissue disaggregation and the nuclei isolation, the chromatin was sheared by sonication with five pulses of 20 seconds each separated by a 40-second rest on ice. After centrifugation, 5 μL of the diluted supernatants were used as input DNA. The other diluted supernatant (100 μL) was added to the acetylated histone H3 antibody-coated wells followed by incubation at room temperature for 60 minutes. ChIP-enriched DNA fragments were carried out by quantitative PCR (qPCR) with one cycle at 94°C for 5 minutes; 35 cycles at 94°C for 30 seconds, 58°C 30 seconds, and 72°C for 30 seconds; and one cycle at 72°C for 7 minutes. Samples were analyzed in triplicate using qPCR. Primers targeted the promoter region of interest genes including LOX, LOXL1, LOXL2, TGF-β, collagen type I, collagen type III, elastin, and fibrillin-1 were used. The concentration of the ChIP samples was calculated as a percent of the input DNA isolated from immunoprecipitated chromatin. The relative fold enrichment for acetylated histone H3 antibody versus IgG negative control was calculated. Since we are comparing the promoter acetylation for each target gene promoter between Korean and Caucasian eyes, relative promoter acetylation (fold change) was calculated as (relative fold enrichment for acetylated histone H3 to IgG negative control in Koreans)/(relative fold enrichment for acetylated histone H3 to IgG negative control in Caucasians). Fold change larger than 1.0 means that relative fold enrichment for acetylated histone H3 is greater in Korean eyes.
The LC and PPS were homogenized in radioimmunoprecipitation assay (RIPA) buffer (1% Triton X-100, 5% SDS, 5% deoxycholic acid, 0.5 M Tris–HCl pH 7.5, 10% glycerol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride [PMSF], 5 μg/mL aprotinin, 1 μg/mL leupeptin, 1 μg/mL pepstatin, 200 mM sodium orthovanadate, and 200 mM sodium fluoride). Tissue extracts were incubated for 10 minutes on ice and clarified by centrifugation at 10,000g for 25 minutes at 4°C. Total protein in tissue extracts was measured using a standard bicinchoninic acid (BCA) assay (Pierce, Rockford, IL, USA). Tissue extracts (30 μg total protein) were resuspended in 5× sample buffer (60 mM Tris–HCl pH 7.4, 25% glycerol, 2% SDS, 14.4 mM 2-mercaptoethanol, 0.1% bromophenol blue) at a 4:1 ratio, boiled for 5 minutes, and resolved by SDS-PAGE. Proteins were transferred onto a nitrocellulose membrane (Hybond-C, Amersham Pharmacia Biotech, Germany), and blots were stained with Ponceau S (Sigma-Aldrich Corp., St. Louis, MO, USA) to visualize the protein bands and ensure equal protein loading and uniform transfer. Blots were washed and blocked for 45 minutes with 5% nondried skim milk in Tris-buffered saline with Tween-20 (TBST) buffer (20 mM Tris–HCl pH 7.6, 137 mM NaCl, and 0.1% Tween 20). Blots were then probed for 24 hours using antibodies against LOX, LOXL1, LOXL2, TGF-β, elastin, fribrillin-1, HDAC2, HDAC3, acetylated histone H3, and GAPDH (Sigma-Aldrich Corp.). Blots were then probed with horseradish-peroxidase (HRP)-conjugated goat anti-rabbit secondary antibody. Bound antibodies were detected using an enhanced chemiluminescence system (ECL, Amersham, MA, USA) and X-ray film. Relative intensity was measured using an ImageMaster VDS (Pharmacia Biotech, Sunnyvale, CA, USA) and the fold changes in these protein levels compared to GAPDH are indicated below the blot. Results are representative of five independent experiments. Data are expressed as mean ± SD.
Posterior segment tissues fixed in 4% buffered paraformaldehyde were processed for paraffin embedding. The 5-μm cross-sections of the LC were stained for elastin and collagen using Weigert's resorcin-fuchsin and picrosirius red histochemical stains, respectively, according to standard protocols. Sections were analyzed with a light microscope (AX70; Olympus Optical Co., Tokyo, Japan). The percentage area occupied by elastin and collagen fibers in relation to the entire area analyzed was detected with ImageJ analytical software by threshold analysis of gray values.
29 Elastin and collagen fibers were detected by standardized threshold settings, with dark fibers included and the bright background excluded.
All data are expressed as mean ± SD. Comparisons between groups were performed using the Mann-Whitney U test, with P < 0.05 set as statistical significance. Statistical analysis was performed using the SPSS statistical package (SPSS, Chicago, IL, USA).
We analyzed the degree of histone acetylation in the LC and PPS between Korean and Caucasian human donor eyes. Immunohistochemical staining of LC and PPS tissues for HDAC2, HDAC3, and acetylated histone 3 is shown in
Figures 1 and
2. The HDAC2 and HDAC3 expression levels were elevated in the nuclei of cells in the LC of Caucasian eyes (
Fig. 1, white arrowheads). As a result, acetylated histone 3 expression was lower in the LC of Caucasian eyes than Korean eyes. Increased acetylated histone 3 expression was observed in the LC of Korean eyes (
Fig. 1, white arrowheads). In the PPS, expression levels of HDAC2 and HDAC3 were elevated in the nuclei of cells in Caucasian eyes (
Fig. 2, white arrowheads). As a result, acetylated histone 3 was decreased in the PPS of Caucasian eyes compared to Korean eyes, as shown by increased nuclear staining in the PPS cells of Korean eyes (
Fig. 2, white arrowheads). These results show that there are epigenetic differences in the LC and PPS between Korean and Caucasian eyes. Acetylation of Korean LC and PPS was increased than the Caucasian LC and PPS, indicating there may be active transcription in the Korean LC and PPS.
To find out which transcriptions are active by the histone modification process in the LC and PPS, major ECM component and ECM modulating enzymes were targeted. We performed ChIP assays to find out the promoter DNA associated with acetylated histone H3, which is shown in
Figure 3. Pooled samples of LC or PPS from six eyes of each Korean and Caucasian were analyzed in one series of experiment. All data of Korean eyes have been normalized into ratios to the level of Caucasian eyes. Promoter regions of LOXL2, elastin, and fibrillin-1 showed significant increase in histone acetylation in the LC and PPS. Other ECM modulating enzymes and ECM components that are reported to be abundant in the LC and PPS, including LOX, LOXL1, collagen type I, and collagen type III, and growth factor, TGF-β, did not show significant changes.
We performed quantitative RT-PCR analysis on total RNA extracted from the retina, choroid, sclera, PPS, LC, and the optic nerve to evaluate the ECM protein composition in ocular tissues. The mRNA level of collagen type I was similar in both Korean and Caucasian eyes (
Fig. 4A). The levels of collagen type III mRNA increased significantly higher in the PPS and LC of Korean eyes than in Caucasian eyes (
Fig. 4B). Additionally, the expression levels of elastin and fibrillin-1 mRNAs were highly elevated in the PPS and the LC of Korean eyes (
Figs. 4C,
4D). Among ECM remodeling enzymes, LOX (
Fig. 4F) and LOXL2 (
Fig. 4G) mRNAs were significantly elevated in Korean eyes.
We performed western blot analysis on protein extracted from the LC and the PPS to evaluate the histone acetylation and ECM composition of the optic nerve head (ONH). As similar to the findings of immunohistochemical staining, HDAC2 and HDAC3 protein were significantly decreased and acetylated histone H3 protein was significantly increased in Korean eyes compared to Caucasian eyes (
Fig. 5). Among ECM modulating enzymes, LOXL2 was significantly increased in Korean eyes than Caucasian eyes. TGF-β and other ECM modulating enzymes, such as LOX and LOXL1, did not show any difference between the LC and PPS of Korean and Caucasian eyes. Among ECM components, elastin and fibrillin-1 were significantly increased in Korean eyes compared to Caucasian eyes. Other major ECM components, including collagen type I and III, did not show significant difference by race. Overall, the levels of protein expression of LOXL2 and components of elastic fiber (elastin and fibrillin-1) differed significantly between Korean and Caucasian eyes.
Expression of the ECM immunolabeling of the LC revealed a complex, radially oriented fiber network consisting of collagen and elastic fibers. ECM remodeling enzymes were mainly localized to the cytoplasm of cells located at the interface between axon bundles and laminar beams and within the axon bundles. LOX expression was similar between Korean and Caucasian eyes (
Fig. 6). This was also for LOXL1 expression (data not shown). However, LOXL2 was significantly increased along the laminar beams in Korean eyes (
Fig. 6). There were increased similar levels of collagen type I and type III in the laminar beams between Korean and Caucasian eyes (data not shown). The quantification of elastin was significantly increased in Korean LC than Caucasian LC (
Fig. 7), but not with fibrillin-1 (
Fig. 8) although it showed increased expression in the Korean LC.
We analyzed the collagen and elastin fiber systems qualitatively and quantitatively in cross-sections of the LC. Because there is regional variation in LC structure, in this study we focused on the inferotemporal region of the LC, which is the region most vulnerable to glaucomatous damage. The staining of the LC using picrosirius red showed no difference in the percentage of the laminar beams occupied by collagen (Korean, 77%; Caucasian, 83%;
Fig. 9). The percentage of the laminar beams occupied by elastic fibers stained with resorcin-fuchsin were significant greater in Korean LC compared to Caucasian LC (Korean, 52%; Caucasian, 33%;
P = 0.032).
We showed that the histone acetylation status differed in the LC and PPS according to race. The LC and PPS are tissues abundant with ECM, so we investigated major ECM fiber components and modulating enzymes. Mainly, the promoter regions of LOXL2 and elastic fiber component (elastin and fibrillin-1) showed different state of acetylation between Korean and Caucasian eyes. Expression analysis using western blot analysis and immnohistochemical staining revealed that LOXL2 and elastic fiber component were increased in the LC and PPS of Korean eyes than Caucasian eyes.
Elastic fibers are one of the chief components of the ECM of the LC and PPS. Notably, elastin and fibrillin assemble to form a mature elastic fiber. These elastic fibers contribute to the structure of the LC and PPS and give elasticity and ability to recoil to the tissue. Alterations in elastic content and quality have been implicated in the pathogenesis of several ECM-related diseases. Representatively, we know a type of glaucoma, which is pseudoexfoliation glaucoma, related to a defective process in elastic fiber formation. A study by Zenkel et al.
3 found dysregulated expression of LOXL1 and elastic fiber components in the LC of pseudoexfoliative syndrome human donor eyes. Similarly, we found that promotor of elastin and fibrillin-1 were acetylated in the LC and PPS of Korean eyes, and the proteins of elastin and fribriilin-1 expression were up-regulated. Quantification of elastin and fibrillin-1 assessed by immunohistochemical and Weigert's resorcin-fuchsin staining showed increased elastin in the LC of Korea eyes.
There is a growing body of evidence to support the theory that the baseline characteristics of the ECM in the PPS and LC play a role in the pathogenesis and susceptibility to glaucoma.
30–32 We propose that racial differences in the baseline ECM properties that are observed in the present study may contribute to glaucoma susceptibility and to various racially disparate clinical characteristics of glaucoma. In Korean glaucoma patients, almost 80% of total glaucoma cases are NTG, and 70% of these patients have myopia.
15,33 A similar glaucoma prevalence is reported in Japan.
16 Both countries have a high prevalence of myopia.
34,35 This is in contrast with Western countries, where NTG is reported to comprise approximately 20% to 50% of the total glaucoma cases, and approximately 10% to 30% is associated with myopia.
36–38
Studies of Korean glaucoma patients revealed that the posterior sclera and the LC were thinner in NTG eyes compared to glaucomatous eyes with high IOP.
39,40 Greater elastic LC and PPS prone to deformation under normal range IOP and contribute to changes within the ONH, such as disc tilt or peripapillary atrophy, leading to increased susceptibility of glaucoma. Our previous clinical investigations showed that glaucoma with myopia or NTG had greater changes in the optic disc morphology, which was related to the glaucomatous damage location, and this seems to be related to the features of the LC and the posterior sclera.
19,41–46 Further investigations are needed to reveal the exact association and clinical meanings of the present findings.
In contrast to the study by Zenkel et al.,
3 our study found increased LOXL2 expression in the LC and PPS of Koreans and no change in LOX or LOXL1 expression. LOX family have many isotypes and they share a catalytic domain, which makes them share similar functions.
47,48 LOX family have a central role in the morphogenesis and repair of ECM, initiates cross-linking of collagen and elastin, and stabilize collagen and elastic fibers.
49 Among them, LOX and LOXL1 are the major forms and knockout of these genes result in death and devastating conditions.
49 The LOXL1 is below normal level in pseudoexfoliation syndrome and glaucoma and this is thought to contribute to the decreased elastic fiber formation and cross-linking in the LC. However, there are also reports showing that LOXL1 is upregulated in early stages of pseudoexfoliation process leading to defective process of elastic fiber synthesis and formation of the pseudoexfoliative materials.
50,51 So, upregulated LOXL2 in the present study may also contribute to defective process at the point of glaucoma development. On the other hand, LOXL2 is thought to be a modulating enzyme rather than direct effect on the synthesis of collagen and elastin. Recently, a new role of LOXL2 to remove the amino group located in lysine of histone H3 when it is methylated has been found.
52,53 LOXL2 is located to the perinuclear region and is able to interact with transcriptional factors. Although this cannot be proven from the present study, elevated LOXL2 may contribute to the epigenetic modification in the LC and PPS.
In summary, histone acetylation status differed in the PPS and LC of the ONH between Caucasian and Korean donor eyes. Especially, the acetylation of the promoter region of LOXL2 and elastic fiber component (elastin and fibrillin-1) showed significant difference in the Caucasian and Korean ONH. This was related to different ECM protein expression in the LC and PPS by race. Further study to reveal the association with these findings to the pathogenesis of glaucoma in Korean eyes is needed.
The authors thank the human donors and their relatives who agreed and participated in this study. We thank Jee Won Mok (Catholic Institute for Visual Science, The Catholic University of Korea) for experiment advices. We thank Suk-Woo Yang (Department of Ophthalmology, The Catholic University of Korea) and Choun-Ki Joo (Catholic Institute for Visual Science, Department of Ophthalmology, The Catholic University of Korea) for providing human donor eyes. We thank Man Soo Kim (Department of Ophthalmology, Chair of Eyebank at Seoul St. Mary's Hospital, The Catholic University of Korea) for assistance in taking care of the donor eyes. We thank Midwest Eye-Banks, who provided Caucasian donor eyes.
Supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. NRF-2014R1A1A3049403). This study was supported by the Busan Sungmo Eye Hospital Sodam Scholarship Committee.
Disclosure: H.-Y.L. Park, None; J.H. Kim, None; Y. Jung, None; C.K. Park, None