Reactivation of Optic Nerve Head Astrocytes by TGF-β2 and H₂O₂ Is Accompanied by Increased Hsp32 and Hsp47 Expression

Abstract

purpose. Histologic studies have previously demonstrated increased expression of small heat shock proteins (Hsps) in reactive optic nerve head (ONH) astrocytes of patients with glaucoma. Transforming growth factor (TGF)-β2 and hydrogen peroxide (H₂O₂) are known to induce ONH astrocyte reactivation. The goal of the present study was to determine whether new potentially involved Hsps, such as Hsp32, -47, -60, and -70, are expressed in the reactivation process of ONH astrocytes mediated by TGF-β2 and H₂O₂.

methods. Cultured human ONH astrocytes were treated with 1.0 ng/mL TGF-β2 for up to 48 hours. In addition, the cells were exposed to 100, 200, or 400 μM H₂O₂ for 1 hour. Expression of Hsp32, -47, -60, and -70 was examined by immunohistochemistry, real-time PCR, and Western blot analyses.

results. Treatment with TGF-β2 increased Hsp32 after 4 and 6 hours, whereas Hsp47 was upregulated after treatment with TGF-β2 for 12, 24, and 48 hours. Exposure of the cells to H₂O₂ could increase both Hsp32 and -47. No significant effects on the expression of Hsp60 and -70 were observed after treatment of the cells with TGF-β2 or H₂O₂.

conclusions. TGF-β2 increased Hsp32 after short-term treatment and Hsp47 after longer periods in cultured human ONH astrocytes. H₂O₂ increased both Hsp32 and -47 levels. No effects on Hsp60 and -70 levels were induced by TGF-β2 and H₂O₂. These results may provide further insights into the cellular stress responses of reactive human ONH astrocytes. Further extensive studies are needed to examine the potential roles of Hsps in the ONH of glaucomatous eyes.

From histologic studies it is known that glaucomatous changes of the optic nerve head (ONH) are associated with a reactivation of astrocytes, ^{1 2 3 4} which is characterized by increased expression of glial fibrillary acidic protein (GFAP) and neural cell adhesion molecule (NCAM). ^{4 5 6} Subsequent histologic experiments on glaucomatous optic neuropathy demonstrated that reactive ONH astrocytes are also associated with increased production of small heat shock proteins (Hsps) such as αB-crystallin ⁷ and Hsp27. ⁸ Hsps represent a large group of proteins, which stabilize protein folding and aggregation. ^{9 10} As molecular chaperones, they are known to be upregulated by various forms of stress. ^{11 12 13 14 15} In various neurodegenerative diseases, increased Hsp expression has been observed in reactive astrocytes. ^{11 16 17} In vitro studies with reactive human ONH astrocytes showed an induced expression of the small Hsps αB-crystallin and Hsp27 after exposure to elevated hydrostatic pressure. ⁵ Whether or not the expressions of new potentially involved Hsps such as Hsp32, -47, -60, or -70 are increased in reactive astrocytes of the ONH is still an open question that requires further investigation. 

The pathogenesis of primary open-angle glaucoma (POAG) is not only associated with an increased intraocular pressure, but is also characterized by cellular stress factors such as increased transforming growth factor-β2 (TGF-β2) ^{18 19 20 21} and oxidative stress. ^{22 23 24} Both TGF-β2 ² and oxidative stress ²³ have been suggested to play a role in the reactivation process of human ONH astrocytes of glaucomatous eyes. Results in several experimental studies have suggested that both TGF-β2 ^{25 26 27 28} and oxidative stress ^{29 30 31} are potent inducers of Hsp upregulation in various cellular systems. In the ocular tissue, TGF-β2 has been shown to augment the synthesis of the small Hsp αB-crystallin in human trabecular meshwork cells ³² and in human ONH astrocytes. ²⁹ Previous data from our laboratory have also demonstrated an upregulation of the small Hsp27 in human ONH astrocytes after TGF-β2 and hydrogen peroxide (H₂O₂) treatment. ³³ Until now, it has been unclear which of the potentially involved Hsps, such as Hsp32, -47, -60, and -70, are expressed in reactive human ONH astrocytes. Furthermore, there are no existing data about the factors responsible for the increased expression of Hsps in reactive human ONH astrocytes. 

In our study, we investigated the effects of TGF-β2 and H₂O₂ on the expression of Hsps in cultured human ONH astrocytes. Knowledge of the baseline and induced expression of Hsps may provide further insight into the cellular process of reactivation of ONH astrocytes. For Hsp47, we detected a possible involvement in the collagen synthesis and secretion in cultured human ONH astrocytes. 

Materials and Methods

Cell Culture

Primary cell cultures of human lamina cribrosa astrocytes were prepared from donor eyes from the eye bank of Ludwig-Maximilians-University (Munich, Germany). Monolayer cultures were established from eyes of five human donors between 56 and 68 years of age. These eyes were obtained 4 to 12 hours postmortem without any history of eye diseases. Methods of securing human tissue were humane, included proper consent and approval, complied with the Declaration of Helsinki, and were approved by the local ethics committee. Astrocytes of the ONH were prepared, grown, and classified as described previously. ^{4 6 29} In brief, the eyes were cut equatorially behind the ora serrata, and the ONH was isolated from the neighboring tissues. The ONH was sagittally dissected under a microscope, and the lamina cribrosa was identified. Discs of the lamina cribrosa were prepared by dissection from the pre- and postlaminar region, subsequently cut into three to four explants and placed in 60-mm Petri dishes with 2 mL Dulbecco’s modified Eagle’s medium (DMEM)/F-12 supplemented with 10% fetal bovine serum (FBS; Invitrogen-Gibco, Karlsruhe, Germany), 5 ng/mL human basic pituitary fibroblast growth factor (bFGF; Sigma-Aldrich, Deisenhofen, Germany), 5 ng/mL human platelet-derived growth factor-A chain (PDGF AA; Sigma-Aldrich), 50 U/mL penicillin and 50 μg/mL streptomycin (Invitrogen-Gibco) at 37°C in a 5% CO₂ incubator. To isolate ONH astrocytes, we first plated the primary cell cultures in serum-free astrocytes growth medium (AGM; Cambrex Bio Science, Verviers, Belgium) for 24 hours and then changed to AGM containing 5% FBS. ³⁴ Other cell populations, such as lamina cribrosa cells, failed to attach in serum-free medium and were removed by the change of medium. Subsequently, cultured ONH astrocytes were maintained in DMEM/F-12 with 10% FBS. ONH astrocytes were distinguished from adjacent cells by their morphology and immunohistochemical staining (data not shown). ^{4 29}  

Primary human ONH astrocytes were characterized by positive immunostaining for Pax2 (Abcam, Cambridge, UK), glial fibrillary acidic protein (GFAP; Sigma-Aldrich), neural cell adhesion molecule (NCAM; Serotec, Düsseldorf, Germany), S100 (Invitrogen). ^{4 6 35 36 37} Pax2 is a transcription factor labeling the nuclei and is expressed by astrocyte lineage cells in the development of adult human retina and ONH. ⁴ Positive immunostaining for GFAP depict intermediate filaments of the cytoskeleton of astrocytes. ^{4 6 35} NCAM is a cell surface adhesion molecule expressed by ONH astrocytes shown as fine granules on the cell surface. ⁶ S100 represents an astroglia-derived calcium-binding protein involved in the maintenance of homeostasis in neurons and astrocytes. ^{36 37} Furthermore, primary human ONH astrocytes are distinguished from neighboring cells by negative immunostaining for smooth muscle actin (smA; Dako, Glostrup, Denmark) and A2B5 (Chemicon International, Hampshire, UK). ^{4 6} smA is a marker of vascular smooth muscle that is absent in astrocyte growth medium. ⁴ A2B5 is a cell surface marker that appears as patches of stain and labels precursors of oligodendrocytes in the myelinated optic nerve, which were removed from the explants under the phase microscope. ^{4 6} Among 13 primary cell cultures, only 9 were at least 95% positive for Pax2, GFAP, NCAM, and S100, and negative for smA and A2B5, and therefore were used in this study. ²⁹ Morphologic characteristics distinguished ONH astrocytes, which are larger, star-shaped cells with thin and long processes, from small, flat, polygonal lamina cribrosa cells. ^{34 38}  

To investigate the effects of TGF-β2, we grew second- to fifth-passage astrocytes to confluence in 35-mm Petri dishes in DMEM/F12 supplemented with 10% FBS at 37°C and 5% CO₂. At confluence, the cells were washed and incubated overnight in serum-free DMEM/F12. After a 24-hour incubation, this medium was replaced by fresh, serum-free DMEM/F12 supplemented with active TGF-β2 (R&D Systems, Wiesbaden, Germany) to a final concentration of 1.0 ng/mL. Under these conditions, the cells were incubated for 2, 4, 6, 12, 24, or 48 hours. In control cultures, the medium was changed at the same time points, but no TGF-β2 was added. 

To test the effects of oxidative stress on astrocytes, we incubated confluent cells of passages 3 to 5 for 24 hours in serum-free DMEM/F12 at 37°C and 5% CO₂. Then, the medium was replaced by fresh serum-free DMEM/F12 medium, and the cells were exposed to 100, 200 or 400 μM hydrogen peroxide (H₂O₂) for 1 hour. After H₂O₂ treatment, the cells were placed in serum-free DMEM/F12 medium for 24 hours. In control cultures, the medium was changed at the same time points but no H₂O₂ was added. The tetrazolium dye-reduction assay (MTT; 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide; Sigma-Aldrich) was used to test cell viability before and at the end of treatment and did not reveal any signs of increased cell death in TGF-β2 or H₂O₂-treated cells (data not shown). All experiments were performed at least in triplicate in astrocyte cultures from three different donors. 

Immunohistochemistry

Cultured human ONH astrocytes, grown in four-well plastic chamber slides, were treated with TGF-β2 or H₂O₂ as described earlier. After incubation, the cells were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde (PFA) for 15 minutes, and subsequently washed twice with PBS containing 0.1% Triton X-100. Primary incubation with all samples was performed with a rabbit anti-human Hsp32 antibody (Stressgen Bioreagents, Hamburg, Germany), mouse anti-human Hsp47 antibody (Stressgen Bioreagents), goat anti-human Hsp60 antibody (Stressgen Bioreagents), or rabbit anti-human Hsp70 antibody (Stressgen Bioreagents), diluted 1:200 in PBS containing 5% bovine serum albumin (BSA) for 4 hours at room temperature (RT). Control samples were incubated with PBS and 5% BSA but without the primary antibodies. Afterward, the cells were washed three times with PBS, and incubated with fluorescein-conjugated goat anti-rabbit Cy3 antibody, goat anti-mouse Cy3 antibody, or donkey anti-goat Cy3 antibody (diluted 1:500 in PBS; Dianova, Hamburg, Germany) for 1 hour at RT. The cells were then rinsed in PBS, mounted with Kaiser’s glycerin jelly (Merck, Darmstadt, Germany) and analyzed by fluorescence microscope (DMR; Leica Microsystems, Wetzlar, Germany). Representative areas were documented on computer (IM 1000 software; Leica Microsystems, Heerbrugg, Switzerland). All experiments were performed at least in triplicate in astrocyte cultures from three different donors. 

RNA Isolation and Real-Time PCR

Total RNA was isolated from 10-cm Petri dishes by the guanidium thiocyanate-phenol-chloroform extraction method (Stratagene, Heidelberg, Germany). The structural integrity of the RNA samples was confirmed by electrophoresis in 1% Tris-acetate-EDTA (TAE)-agarose gels. Yield and purity were determined photometrically. After RNA isolation, mRNA was transcribed to cDNA via reverse transcriptase. This cDNA was then used for specific real-time PCR. Quantification of human Hsp32, -47, -60 and -70 mRNA was performed with specific primers (Table 1)during 40 cycles with a thermocycler (LightCycler System; Roche Diagnostics, Mannheim, Germany). Primers and probes were found with a commercial computer program (ProbeFinder, ver. 2.04; Exiqon, Woburn, MA). The standard curve was obtained from probes of three different untreated human ONH astrocyte cultures. To normalize differences of the amount of total RNA added to each reaction, we simultaneously processed 18S rRNA in the same sample as an internal control. The level of Hsp32, -47, -60, and -70 mRNA was determined as the relative ratio (RR), which was calculated by dividing the levels of mRNA of those Hsps by the level of the 18S rRNA housekeeping gene in the same samples. All experiments were performed at least in triplicate in astrocyte cultures from three donors. 

Protein Extraction and Western Blot Analysis

Cells grown on 35-mm tissue culture dishes were washed twice with ice-cold PBS, collected, and lysed in RIPA cell lysis buffer. After centrifugation (19,000g for 30 minutes at 4°C) in a microfuge, the supernatants were transferred to fresh tubes and stored at −70°C for future use. The protein content was measured by the bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL). Denatured proteins (2 μg) were separated under reducing conditions by electrophoresis on a 5% SDS-polyacrylamide stacking gel and an 8% or 10% SDS-polyacrylamide separating gel. Thereafter, the proteins were transferred onto a polyvinyl difluoride membrane (Roche) with semidry blotting (Hsp32, -47, -60, and -70) or by tank blot (Col1α1) and probed with a rabbit anti-human Hsp32 antibody (Stressgen Bioreagents), mouse anti-human Hsp47 antibody (Stressgen Bioreagents), goat anti-human Hsp60 antibody (Stressgen Bioreagents), rabbit anti-human Hsp70 antibody (Stressgen Bioreagents), or rabbit anti-human Col1α1 antibody (Rockland, Gilbertsville, PA), as described before. ³² These antibodies were used at a dilution of 1:200, respectively. Chemiluminescence was detected with the imager (LAS-1000; RayTest, Pforzheim, Germany). Exposure times ranged between 1 and 20 minutes. Quantification was performed on a computer (AIDA software; RayTest). All experiments were performed at least in triplicate in astrocyte cultures from three donors. 

Transfection of siRNA

Cultured human ONH astrocytes were seeded in six-well plates at a density of 8 × 10⁵ cells per well and grown for 24 hours in DMEM/F-12 with 10% FBS. The astrocytes were transfected with an Hsp47 specific siRNA (c = 0.1 μM; On-Targetplus siRNA, cat. no. J-011230-05; Dharmacon, Lafayette, CO) using a transfection reagent (Dharmafect1; Dharmacon) according to the manufacturer’s instructions. After 24 hours, the medium was changed to serum-free medium containing 1.0 ng/mL TGF-β2 or to regular culture medium, and the astrocytes were treated for another 12 hours. Corresponding control astrocytes were treated under the same conditions without the addition of siHsp47 for the transfection procedure. After treatment, the astrocytes were harvested for RNA isolation, and the medium of the astrocytes was collected for protein extraction. Medium samples were concentrated 40× by centrifugation (Centricon columns; Millipore, Billerica, MA). Hsp47 knockdown efficiency was tested on RNA samples by semiquantitative PCR using Hsp47 primers. The effect of Hsp47 knockdown on intracellular type I collagen accumulation was evaluated by semiquantitative PCR and Western blot experiments. 

Semiquantitative PCR

Gene-specific PCRs were performed in a total volume of 25 μL containing 5 μL cDNA, 2.5 μL 10× PCR buffer (Mg²⁺ free), 0.5 μL 10 mM dNTP mix, 0.5 μL 10 μM primer (forward and reverse each), 0.75 μL 50 mM MgCl₂, 0.1 μL (5 U/μL) Taq polymerase (all from Invitrogen), and H₂O. The 25-μL PCR steps were 30 seconds of denaturation at 96°C, 30 seconds of annealing, and 45 seconds of extension at 72°C, followed by an end-extension step of 5 minutes at 72°C after the last cycle. The primers for Hsp47 were (forward, 5′-TTCTGCCTCCTGGAGGCG-3′; reverse, 5′-CGCTCAGCACTGCCTTGG-3′, position, 257-274; product size, 267; annealing temperature, 58.0°C), for Col1α1 were (forward, 5′-GATGGACTCAACGGTCTCC-3′; reverse 5′-CCTTGGGGTTCTTGCTGATG-3′; position, 3576-4034; product size, 458; annealing temperature, 57.0°C), and for GAPDH (forward, 5′-GAAGGTGAAGGTCGGAGTC-3′; reverse, 5′-GAAGATGGTGATGGGATTTC-3′; position, 6-231; product size, 225; annealing temperature, 57.0°C). The functionality of primers was tested on cDNAs obtained from different tissues before the experiments to exclude false-negative results (data not shown). The band intensity was measured in light units with an imaging workstation (Lumi-Imager; Roche, Mannheim, Germany). Quantification was performed with the accompanying software (Lumi-Analyst software, Roche). The final amount of PCR product was expressed as the ratio of the Hsp47 or the Col1α1 gene amplified to that of the GAPDH gene. 

Results

Characterization of Human Astrocyte Cultures

All human ONH astrocyte cultures were characterized by negative immunohistochemical staining for smA and A2B5, which excluded potential vascular smooth muscle cells and oligodendrocytes and astrocytes of the myelinated optic nerve (Fig. 1) . Furthermore, all cell cultures showed positive immunohistochemical staining for Pax2, GFAP, NCAM-1, and S100, identifiers of reactive astrocyte cultures of the ONH (Fig. 1) . 

Effects of TGF-β2 on Hsp32 Expression

Human ONH astrocytes were treated with 1.0 ng/mL TGF-β2 for 2, 4, 6, 12, 24, and 48 hours (Fig. 2) . By immunohistochemical staining, treatment with TGF-β2 for 4 hours showed a maximum increase of Hsp32 expression (Fig. 2B)compared with untreated control cells (Fig. 2A) . Since immunohistochemistry is not a reliable quantification method, we also performed real-time PCR and Western blot analysis. The signals generated by real-time PCR analysis in untreated control cells were set to 100% (Figs. 2C 2E) . Keeping in serum-free medium had no influence on Hsp expression for all investigated time periods. There was an upregulation of Hsp32 mRNA expression after TGF-β2 treatment for 4 (2.1 ± 0.7-fold) and 6 (2.6 ± 0.3-fold) hours compared with untreated control cells (Fig. 2C) . However, the Hsp32 mRNA level did not change after 12 (0.9 ± 0.1-fold), 24 (0.8 ± 0.3-fold), and 48 (0.7 ± 0.2-fold) hours of TGF-β2 treatment compared with untreated control cells (Fig. 2E) . 

At the protein level, Western blot analysis revealed an upregulation of Hsp32 expression after treatment with TGF-β2 for 4 (2.0 ± 0.3-fold) and 6 (1.4 ± 0.2-fold) hours compared with untreated control cells (Fig. 2D) . In contrast, treatment with TGF-β2 for 2 (1.1 ± 0.2-fold), 12 (0.8 ± 0.1-fold), 24 (0.7 ± 0.1-fold), and 48 (0.7 ± 0.1-fold) hours showed no marked changes of Hsp32 protein expression (Figs. 2D 2F) . 

Effects of H₂O₂ on Hsp32 Expression

Cultured human ONH astrocytes were treated with 100, 200, and 400 μM hydrogen peroxide (H₂O₂) for 1 hour (Fig. 3) . Immunohistochemical staining demonstrated maximum Hsp32 increase after exposure of cells to 200 μM H₂O₂ for 1 hour (Fig. 3B)compared with untreated control cells (Fig. 3A) . Real-time PCR analysis was performed 24 hours after stress exposure (Fig. 3C) . The signals generated in untreated control cells were set to 100% (Fig. 3C) . Exposure to 200 and 400 μM H₂O₂ for 1 hour increased the mRNA expression of Hsp32 1.9 ± 0.5- and 1.7 ± 0.7-fold compared with untreated control levels (Fig. 3C) . 

At the protein level, Western blot analysis showed an increase of Hsp32 protein amount after treatment with 200 μM H₂O₂ (1.8 ± 0.4-fold), and 400 μM H₂O₂ (1.6 ± 0.3-fold) compared with untreated control levels (Fig. 3D) . 

Effects of TGF-β2 on Hsp47 Expression

Immunohistochemical staining revealed maximum Hsp47 expression after 48 hours of TGF-β2 treatment (Fig. 4B)compared with untreated control levels (Fig. 4A) . Expression of Hsp47 mRNA expression was upregulated by prolonged TGF-β2 treatment (Fig. 4E) . Although the Hsp47 levels rarely changed up to 6 hours of treatment, they increased by 1.6 ± 0.2-fold after 12, 2.1 ± 0.5-fold after 24, and 2.3 ± 0.5-fold after 48 hours (Figs. 4C 4E) . 

At the protein level, no significant changes of Hsp47 protein expression was observed by Western blot analysis after short-time TGF-β2 treatment for 2 (1.0 ± 0.1-fold), 4 (0.9 ± 0.1-fold), and 6 (0.8 ± 0.1-fold) hours (Fig. 4D) . On the other hand, prolonged treatment with TGF-β2 for 12, 24, and 48 hours increased the Hsp47 protein levels by 1.6 ± 0.2-, 2.0 ± 0.3-, and 2.2 ± 0.1-fold compared with untreated control levels (Fig. 4F) . 

Effects of H₂O₂ on Hsp47 Expression

Immunohistochemical staining demonstrated maximum Hsp47 increase after exposure of cells to 200 μM H₂O₂ for 1 hour (Fig. 5B)compared with untreated control levels (Fig. 5A) . Treatment with H₂O₂ increased the Hsp47 expression both at the mRNA (Fig. 5C)and protein (Fig. 5D)level. Exposure to 100, 200, and 400 μM H₂O₂ for 1 hour upregulated the Hsp47 mRNA expression by 1.5 ± 0.2, 1.6 ± 0.2, and 1.7 ± 0.3-fold compared with untreated control levels (Fig. 5C) . At the protein level, Western blot analysis revealed increased Hsp47 protein expression by 1.5 ± 0.3-, 1.6 ± 0.5-, and 1.5 ± 0.3-fold after exposure of cells to 100, 200, and 400 μM H₂O₂ compared with untreated control levels (Fig. 5D) . 

Effects of TGF-β2 on Hsp60 Expression

Immunohistochemical staining revealed similar Hsp60 expression after 12 hours of TGF-β2 treatment (Fig. 6B)compared with untreated control levels (Fig. 6A) . There was no effects on Hsp60 mRNA expression for all investigated periods (TGF-β2 for 2 hours, 0.8 ± 0.3-fold; 4 hours, 0.9 ± 0.1-fold; 6 hours, 1.1 ± 0.3-fold; 12 hours, 1.4 ± 0.3-fold; 24 hours, 1.2 ± 0.2-fold; 48 hours, and 0.8 ± 0.4-fold; Figs. 6C 6E ). 

Western blot analysis showed no significant effects of TGF-β2 treatment on Hsp60 protein expression compared with untreated control levels (TGF-β2 for 2 hours, 1.0 ± 0.1-fold; 4 hours, 0.9 ± 0.1-fold; 6 hours, 1.0 ± 0.1-fold; 12 hours, 1.1 ± 0.01-fold; 24 hours, 0.8 ± 0.04-fold; and 48 hours, 0.8 ± 0.1-fold; Figs. 6D 6F ). 

Effects of H₂O₂ on Hsp60 Expression

Immunohistochemical staining revealed no Hsp60 increase after exposure of cells to 100 μM H₂O₂ for 1 hour (Fig. 7B)compared with untreated control levels (Fig. 7A) . An upregulation of Hsp60 mRNA was observed after treatment with 200 μM (1.5 ± 0.2-fold) and 400 μM (1.5 ± 0.2-fold) H₂O₂ for 1 hour compared with untreated control levels (Fig. 7C) . There were no significant changes of Hsp60 protein level after all investigated time periods (100 μM H₂O₂, 0.9 ± 0.1-fold; 200 μM H₂O₂, 0.8 ± 0.3-fold; and 400 μM H₂O₂, 0.8 ± 0.2-fold; Fig. 7D ). 

Effects of TGF-β2 on Hsp70 Expression

Immunohistochemical staining revealed similar Hsp70 expression after 12 hours of TGF-β2 treatment (Fig. 8B)compared with untreated control levels (Fig. 8A) . Real-time PCR analysis revealed no significant effects on Hsp70 mRNA expression after TGF-β2 treatment (TGF-β2 for 2 hours, 1.1 ± 0.2-fold; 4 hours, 1.1 ± 0.3-fold; 6 hours, 1.0 ± 0.2-fold; 12 hours, 1.2 ± 0.4-fold; 24 hours, 1.1 ± 0.5-fold; and 48 hours, 0.8 ± 0.6-fold; Figs. 8C 8E ). 

Similar results were achieved at the protein levels. Treatment with TGF-β2 for 2 (1.2 ± 0.1-fold), 4 (0.8 ± 0.1-fold), 6 (0.9 ± 0.1-fold), 12 (1.2 ± 0.1-fold), 24 (0.8 ± 0.2-fold) and 48 (0.9 ± 0.1-fold) hours showed no significant effects on Hsp70 protein expression (Figs. 8D 8F) . 

Effects of H₂O₂ on Hsp70 Expression

Immunohistochemical staining demonstrated no Hsp70 increase after exposure of cells to 200 μM H₂O₂ for 1 hour (Fig. 9B)compared with untreated control levels (Fig. 9A) . Exposure of cells to 100, 200, and 400 μM H₂O₂ demonstrated minor changes in Hsp70 mRNA expression compared with untreated control levels (100 μM H₂O₂, 1.0 ± 0.3-fold; 200 μM H₂O₂, 1.2 ± 0.1-fold; and 400 μM H₂O₂, 0.8 ± 0.2-fold; Fig. 9C ). Western blot analysis also revealed no significant effects of H₂O₂ on Hsp70 protein levels (100 μM H₂O₂, 1.1 ± 0.1-fold; 200 μM H₂O₂, 1.2 ± 0.1-fold; and 400 μM H₂O₂, 1.0 ± 0.1-fold) compared with untreated control levels (Fig. 9D) . 

Involvement of Hsp47 Knockdown on TGF-β2–Induced Col1α1 Expression

The effect of siHsp47 on the TGF-β2–mediated upregulation of Col1α1 was analyzed by semiquantitative PCR and Western blot analysis. By semiquantitative PCR analysis, it can be seen that Hsp47 is effectively knocked down by the use of Hsp47 siRNA (Figs. 10A 10B) . Treatment with 1.0 ng/mL TGF-β2 for 12 hours increased Col1α1 mRNA expression by 1.5 ± 0.2-fold compared with untreated control levels (Figs. 10A 10B) . However, when cells were transfected with Hsp47 siRNA before TGF-β2 treatment, there was no marked change of Col1α1 mRNA expression (1.1 ± 0.1-fold) compared with that in untreated control cells (Figs. 10A 10B) . 

To examine the effect of siHsp47 on the Col1α1 secretion in TGF-β2-treated astrocytes, we conducted Western blot analyses on the media of treated cells (Figs. 10C 10D) . Treatment with TGF-β2 for 12 hours could induce an increased extracellular Col1α1 protein secretion into the media by 1.8 ± 0.2-fold compared with untreated control levels, whereas the media of siHsp47-transfected astrocytes treated with TGF-β2 showed only a 1.2 ± 0.1-fold expression of Col1α1 protein compared with untreated control levels (Figs. 10C 10D) . 

Discussion

Under neuropathologic conditions, increased synthesis of heat shock proteins (Hsps) in reactive astrocytes may provide a protective cellular response to stressful events. ^{11 16 17} Until now, little is known about the expression of Hsps in reactive human ONH astrocytes, which may play a role in ocular diseases such as in glaucomatous neurodegeneration of the optic nerve. TGF-β2 ^{2 29} and H₂O₂ ²³ have been used to induce reactivation of astrocytes. In this study, we examined the effects of these two reactivation inducing factors on the expression of Hsp32, -47, -60, and -70 in cultured human ONH astrocytes. 

Hsp32 is known as the stress-inducible form of the enzyme heme oxygenase (HO)-1, which catabolizes the pro-oxidant heme to biliverdin, free iron, and carbon monoxide. ¹¹ The Hsp32 gene is explicitly sensitive to stress stimuli including TGF-β2 and H₂O₂. ^{39 40} In our experiments, both TGF-β2 and H₂O₂ were able to upregulate Hsp32 mRNA and protein expression. Hsp32 expression was increased after short-time treatment with TGF-β2 for 4 and 6 hours and after exposure of cells to H₂O₂ for 1 hour. In contrast, prolonged treatment with TGF-β2 for 12, 24, and 48 hours demonstrated no significant effects on Hsp32 mRNA and protein expression. These results are consistent with in vitro studies on cerebral astrocytes, which showed maximally increased mRNA and protein Hsp32 levels within 2 to 4 hours after stress exposure and reached basal levels within 8 hours after stress exposure. ^{41 42} This early increase in Hsp32 expression, which subsided after prolonged TGF-ß2 exposure up to 48 hours, may indicate that Hsp32 is explicitly a marker of early stress response. Increased production of Hsp32 has been implied in the cellular protection against oxidative stress and may play an important role in antioxidant defense responses. ^{42 43 44} It is overexpressed in several neurodegenerative conditions including Alzheimer and Parkinson disease. ^{39 45 46} Whether or not Hsp32 is increasingly expressed in reactive ONH astrocytes in glaucomatous optic neuropathy awaits further investigation. 

TGF-β2 ^{26 47 48 49} and oxidative stress ^{50 51} are known stress stimuli for Hsp47 induction as observed in various cellular systems. Our experiments revealed a major induction of Hsp47 expression after prolonged treatment with TGF-β2 for 12, 24, and 48 hours, whereas treatment with TGF-β2 for a shorter time did not lead to significant changes of Hsp47 protein expression. Similarly, previous studies on cortical reactive astrocytes of the rat brain demonstrated induced Hsp47 expression after prolonged stress exposure from 10 hours to 14 days. ¹¹ Our experiments also showed increased Hsp47 levels after H₂O₂ exposure. The reasons for Hsp47 induction are as yet unknown. There are assumptions based on various studies that induced Hsp47 expression may participate in glial cell protection and adaptation to damage. ¹¹ Another aspect of the role of Hsp47 in reactive ONH astrocytes may provide its function as an important collagen-specific molecular chaperone and its role in collagen biosynthesis as observed in different diseases such as in arteriosclerosis, pulmonary fibrosis, and keloid formation. ^{52 53 54 55 56 57} Treatment of astrocytes with TGF-β2 induced an increased Col1α1 expression, which is consistent with results of our previous studies. ⁵⁸ By transfection of ONH astrocytes with Hsp47-specific siRNA before TGF-β2 treatment, we could observe a less marked increase of Col1α1 synthesis and secretion. These results confirmed that Hsp47 may play a critical role in collagen accumulation by enhancing synthesis and secretion of collagen in vitro. Whether Hsp47 is also involved in collagen synthesis in the ONH in vivo is still unknown. 

Our results on Hsp60 induction revealed no significant effects of TGF-β2 and H₂O₂ on both Hsp60 mRNA and protein expression. In agreement with our observations, no enhanced expression of Hsp60 was found after TGF-β2 treatment of human cerebral astrocytes. ²⁸ Similarly, previous investigations in vitro have shown no induction of Hsp60 after H₂O₂ exposure. ^{59 60} In accordance with our in vitro results, Tezel et al. ⁸ have observed no prominent immunohistochemical staining for Hsp60 in reactive ONH astrocytes of glaucomatous eyes. 

Comparable results were obtained for Hsp70 induction in cultured human ONH astrocytes. Our experiments revealed no significant effects of both TGF-β2 and H₂O₂ treatment on Hsp70 levels. These results are in contrast to observations with cerebral astrocytes, which respond to numerous stress stimuli by rapid upregulation of Hsp70. ^{61 62 63} While extensive studies exist on stress-induced Hsp70 expression in neuronal cells, ^{11 64} little is known about stress-induced Hsp70 expression in astrocytes. ⁶⁴ Therefore, whether Hsp70 is increased in human glaucomatous ONH astrocytes awaits further investigations. 

In summary, we were able to show that both TGF-β2- and H₂O₂-induced reactivation of ONH astrocytes is accompanied by increased Hsp32 and -47 expression in vitro. These results may provide further insights into the cellular stress responses of reactive human ONH astrocytes. Further extensive in vitro and in vivo studies are needed to examine the potential roles of Hsps in the ONH of glaucomatous eyes. 

 Supported by FoeFoLe (ALY) and the German Ophthalmological Society Research (UW-L)

 Submitted for publication March 1, 2008; revised July 30, 2008; accepted January 26, 2009.

 Disclosure: A.L. Yu, None; J. Moriniere, None; M. Birke, None; C. Neumann, None; R. Fuchshofer, None; A. Kampik, None; H. Bloemendal, None; U. Welge-Lussen, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Primers Used for Real-Time PCR

Gene Target	Gene Sequence	Gene Position
Hsp32	5′-gggtgatagaagaggccaaga-3′	673–693
	5′-agctcctgcaactcctcaaa-3′	720–739
Hsp47	5′-gcgggctaagagtagaatcg-3′	122–141
	5′-atggccaggaagtggtttg-3′	213–231
Hsp60	5′-tcagtgtgccttgaactctatga-3′	1364–1386
	5′-ttatctaaatcctggagtacaacctg-3′	1430–1455
Hsp70	5′-cagcagacaccagcagaaaa-3′	1739–1758
	5′-cttggatccagcttgagagg-3′	1785–1804
18S rRNA	5′-ctcaacacgggaaacctcac-3′	1348–1367
	5′-cgctccaccaactaagaacg-3′	1438–1457

Figure 1.

 

Immunofluorescence staining of the cultured ONH astrocytes. Cells were characterized by negative staining for smA and A2B5 (shown with DAPI nuclear counterstaining) and positive staining for Pax2, GFAP, NCAM, and S100.

Figure 2.

 

TGF-β2 increased the expression of Hsp32. (A) By immunohistochemistry, basal levels of Hsp32 staining were observed in untreated control ONH astrocytes. (B) Treatment with 1.0 ng/mL TGF-β2 for 4 hours increased the expression of Hsp32 compared with untreated control levels. (C, E) Real-time PCR analysis of TGF-β2-induced Hsp32 mRNA expression. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp32 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D, F) For Western blot analysis of Hsp32 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp32 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors (*P < 0.05). Co, control.

Figure 3.

 

H₂O₂ increased the expression of Hsp32. (A) By immunohistochemistry, basal levels of Hsp32 staining were observed in untreated control ONH astrocytes. (B) Treatment with 200 μM H₂O₂ for 1 hour increased the expression of Hsp32 compared with untreated control levels. (C) Real-time PCR analysis of H₂O₂-induced Hsp32 mRNA expression. Results were normalized to 18S rRNA as a reference. The steady state mRNA level of Hsp32 in untreated control cells was considered to be100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D) For Western blot analysis of Hsp32 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp32 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors (*P < 0.05). Co, control.

Figure 4.

 

TGF-β2 increased the expression of Hsp47. (A) By immunohistochemistry, basal levels of Hsp47 staining were observed in untreated control ONH astrocytes. (B) Treatment with 1.0 ng/mL TGF-β2 for 48 hours increased the expression of Hsp47 compared with untreated control levels. (C, E) Real-time PCR analysis of TGF-β2-induced Hsp47 mRNA expression. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp47 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D, F) For Western blot analysis of Hsp47 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp47 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors (*P < 0.05). Co, control.

Figure 5.

 

H₂O₂ increased the expression of Hsp47. (A) By immunohistochemistry, basal levels of Hsp47 staining were observed in untreated control ONH astrocytes. (B) Treatment with 200 μM H₂O₂ for 1 hour increased the expression of Hsp47 compared with untreated control levels. (C) Real-time PCR analysis of H₂O₂-induced Hsp47 mRNA expression. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp47 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D) For Western blot analysis of Hsp47 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp47 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors (*P < 0.05). Co, control.

Figure 6.

 

TGF-β2 had no effect on Hsp60 expression. (A) By immunohistochemistry, basal levels of Hsp32 staining were observed in untreated control ONH astrocytes. (B) TGF-β2 treatment did not induce the expression of Hsp60 compared with untreated control levels. (C, E) Real-time PCR analysis of Hsp60 mRNA expression after TGF-β2 treatment. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp60 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D, F) For Western blot analysis of Hsp60 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp60 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors. Co, control.

Figure 7.

 

H₂O₂ had no influence on Hsp60 expression. (A) By immunohistochemistry, basal levels of Hsp60 staining were observed in untreated control ONH astrocytes. (B) H₂O₂ exposure did not induce Hsp60 expression compared with untreated control levels. (C) Real-time PCR analysis Hsp60 mRNA expression after H₂O₂ exposure. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp60 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D) For Western blot analysis of Hsp60 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp60 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors (*P < 0.05). Co, control.

Figure 8.

 

TGF-β2 had no effect on Hsp70 expression. (A) By immunohistochemistry, basal levels of Hsp70 staining were observed in untreated control ONH astrocytes. (B) TGF-β2 treatment did not induce the expression of Hsp70 compared with untreated control levels. (C, E) Real-time PCR analysis of Hsp70 mRNA expression after TGF-β2 treatment. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp70 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D, F) For Western blot analysis of Hsp70 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp70 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors. Co, control.

Figure 9.

 

H₂O₂ had no influence on Hsp70 expression. (A) By immunohistochemistry, basal levels of Hsp70 staining were observed in untreated control ONH astrocytes. (B) H₂O₂ exposure did not induce Hsp70 expression compared with untreated control levels. (C) Real-time PCR analysis Hsp70 mRNA expression after H₂O₂ exposure. Results were normalized to 18S rRNA as the reference. The steady state mRNA level of Hsp70 in untreated control cells was considered to be 100%. Results are given as the mean ± SD of nine experiments with three different cell cultures from different donors. (D) For Western blot analysis of Hsp70 protein expression, lysates containing approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection of Hsp70 content. Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of nine experiments with three different cell cultures from different donors. Co, control.

Figure 10.

 

(A) Semiquantitative PCR analysis of Hsp47 and Col1α1 in untreated control cells, in TGF-β2-treated cells, and in cells transfected with siHsp47 before TGF-β2 treatment. Application of equal cDNA amounts in the PCR was controlled by GAPDH PCR. (B) Data represent the mean ratio of the optical density of the Hsp47 or Col1α1 PCR products normalized to the GAPDH amplicon of the same cDNA and are expressed as x-fold changes compared with untreated control levels. Results are given as the mean ± SD of six experiments with three different cell cultures from different donors (**P < 0.05). Co, control. (C) Representative Western blot analysis on concentrated medium of astrocytes treated under the same conditions as described in (A). Coomassie staining of gels demonstrates equal loading and protein contents of media probes. (D) Data are expressed as x-fold changes compared with untreated control levels and represent the mean ± SD of results of six experiments with three different cell cultures from different donors (**P < 0.05). Co, control.

The authors thank Katja Obholzer for excellent technical assistance.