αB-Crystallin in the Trabecular Meshwork Is Inducible by Transforming Growth Factor-β

Abstract

purpose. Because in glaucomatous eyes transforming growth factor-β (TGF-β) and αB-crystallin are increased in the anterior eye segment, the effect of TGF-β1 and TGF-β2 on the expression of αB-crystallin and its corresponding mRNA was studied in human trabecular meshwork (TM) cells.

methods. Monolayer cultures of “cribriform” and “corneoscleral” TM cells of 5 human donors (12–73 years of age) were treated with either 1.0 ng/ml TGF-β1, TGF-β2, or 5 × 10⁻⁷ dexamethasone (DEX) for 12 to 96 hours. Induction of αB-crystallin and the related mRNA was investigated by western and northern blot analyses. For comparison, human foreskin fibroblasts (HFF) and NIH 3T3 cells were treated in the same way as the TM cells.

results. An increase of αB-crystallin mRNA was observed after treatment of TM cells with TGF-β1 and TGF-β2, whereas DEX had no effect. In the cribriform TM cells with a high basal level, the enhancement ranged between 2 and 3 times; whereas in the corneoscleral TM cellsα B-crystallin mRNA increased between 5 and 6 times. Using western blot analysis, the increase of αB-crystallin expression in the cribriform TM cells was only small compared with the significant increase in the corneoscleral TM cells. Treatment of HFF and NIH 3T3 cells with TGF-β did not induce αB-crystallin mRNA.

conclusions. This is the first time to show that αB-crystallin is not only induced by stress factors but also by TGF-β in TM cell cultures. The difference in induction of mRNA and protein seems to be dependent onα B-crystallin concentration before treatment.

Alpha B-Crystallin, a member of the small heat shock protein family (HSP), can act as a molecular chaperone, preventing aggregation and unfolding of proteins in response to stress. ¹ Constitutively the protein is expressed not only in lens cells ² but also in a variety of nonlenticular cells. ^{3 4 5 6 7 8} Most of these cells show only minor mitotic capacity and are exposed to physiological, mechanical, osmotic, or oxidative stress over a long period. In the eye, constitutive expression of αB-crystallin has been found in the trabecular meshwork (TM). Histologic sections revealed the presence of αB-crystallin only in cells adjacent to the inner wall of Schlemm’s canal but not in the inner portions of the TM. ^{9 10 11} In the present study we showed that at the electronmicroscopic level of αB-crystallin is present only in cribriform TM cells and not in cells of the lamellated portion of the TM or in endothelial cells lining Schlemm’s canal. In many glaucomatous eyes αB-crystallin is increased in the TM and is also present in inner corneoscleral and uveal TM cells. ¹² Elevated levels of αB-crystallin have also been described in various neurodegenerative disorders such as Creutzfeld–Jacob or Alexander’s disease. ^{13 14 15} The reason for the accumulation ofα B-crystallin under these pathologic conditions is unknown. 

In vitro human TM cells accumulate αB-crystallin in response to oxidative stress and heat shock. ¹⁰ Oxidative damage is considered an important factor in the pathogenesis of glaucoma, ^{16 17 18} but direct evidence for this assumption is still lacking. On the other hand, it has been shown that the aqueous humor of many glaucomatous eyes exhibits increased amounts of transforming growth factor-β (TGF-β). ¹⁹ Moreover, it is known that glucocorticoid treatment is causative for some forms of glaucoma. It has been shown that hormone responsiveness is a characteristic of small HSP genes (e.g., human hsp27 is estrogen-responsive and is expressed in several estrogen-sensitive human tissues and breast tumors). ^{20 21 22 23} Accumulation ofα B-crystallin in response to the glucocorticoid hormone dexamethasone (DEX) has been demonstrated in murine NIH 3T3 cells. ²⁴ In the present study we investigated whether treatment with DEX or TGF-β results in the induction of αB-crystallin in TM cells. 

Materials and Methods

Electronmicroscopic Investigation

In previous studies we have shown that localization ofα B-crystallin staining is similar in human and monkey eyes. ⁹ Sufficient morphology for ultrastructuralα B-crystallin demonstration was only obtained in primate eyes fixed immediately after enucleation. These studies were therefore only performed in monkey eyes. Cynomolgus monkey eyes were kindly provided by the primate center in Marburg (Behring Werke, Marburg, Germany). All animals were kept and treated in agreement with the Helsinki Convention on the Use of Animals in Research and conform to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The animals were killed in conjunction with other nonocular protocols. The eyes were immediately enucleated and 2-mm–wide sections of the anterior eye segment were placed in 4% paraformaldehyde and 0.1% glutaraldehyde for 4 hours. Cryoprotection was performed by washing in ice-cold 30% dimethylformamide twice for 30 minutes. The samples were then incubated in methylbutan for 30 minutes and immediately frozen in liquid nitrogen. Low-temperature embedding was performed with a EM-CPC (Leica, Germany, Bensheim) using Lowicryl HM20 (Polysciences Europe, Eppelheim, Germany) according to the instruction manual of the manufacturer. After UV-polymerization, ultrathin sagittal sections were prepared using a Reichert microtome (Ultracut OmU3). 

Sections were mounted on Ni-grids, air-dried, and preincubated with 2% bovine serum albumin in phosphate-buffered saline (PBS, pH 7.4) for 20 minutes, followed by incubation with the primary antibody (rabbit antiα B-crystallin 1:100) ⁴ for 2 hours. After washing six times for 10 minutes each, labeling was performed using goat anti-rabbit IgG 10 nm gold particles and the sections washed again. Uranyl-acetate–stained sections were viewed with a Zeiss EM 902 electronmicroscope (Zeiss, Oberkochen, Germany). 

Tissue Culture

Cell cultures derived from eyes of 5 human donors (12, 49, 57, 57, and 73 years of age, obtained 4–8 hours postmortem) were prepared, grown, and classified as described previously. ^{9 11} Cells derived from the cribriform and outer corneoscleral meshwork were termed cribriform, cells derived from the inner corneoscleral and uveal portion were termed corneoscleral trabecular cells. Confluent TM cells were distinguished from each other and from adjacent scleral spur and ciliary muscle cells by their morphology and immunohistochemical staining as described previously. ¹¹  

In brief, confluent cells of the third passage being confluent for 7 days were stained with antibodies against α-smooth muscle (sm) actin (mouse, clone No. 1A4 Ig G2a anti-bovine; Dakopatts, Hamburg, Germany) diluted 1:150, desmin (mouse, clone D33 anti-human IGG1; Dakopatts) diluted 1:10, and αB-crystallin (rabbit) ⁴ diluted 1:100. For staining with antibodies against α-sm actin and desmin, cells were fixed with ice-cold methanol for 3 minutes; for demonstration of αB-crystallin the fixation was performed with 4% paraformaldehyde for 15 minutes, followed by two washes with PBS containing 0.1% Triton X-100. All antibodies were incubated overnight at 4°C. After washing in PBS, cells were incubated with fluorescein-labeled swine anti-mouse and swine anti-rabbit immunoglobulins (Dakopatts) diluted 1:20. Slides were viewed and photographed with a Leitz Aristoplan microscope (Leitz, Wetzlar, Germany). 

Confluent cribriform and corneoscleral TM cells of passage 3 were incubated for 12, 24, 48, or 96 hours in serum-free medium supplemented with 1.0 ng/ml TGF-β1 (Boehringer Mannheim, Mannheim, Germany), 1.0ng/ml TGF-β2 (Boehringer Mannheim), 5 × 10⁻⁷ M DEX (Sigma, Deisenhofen, Germany). The treated cells were compared with cultures incubated under identical conditions but without TGF-β or DEX in the medium. 

To test whether the effect of induction of αB-crystallin mRNA is due to the HCl that is used to activate TGF-β, cribriform cells were either treated with 1 μM HCl or a combination of 1.0 ng/ml TGF-β2 and 10 mg/ml anti–TGF-β2 (RD-Systems, Wiesbaden, Germany). 

Additionally we treated primary human foreskin fibroblasts (HFF; kindly provided by the Department of Virology, Erlangen) and NIH 3T3 fibroblasts with DEX, TGF-β1, or TGF-β2, in the same way as the TM cells for 24 hours, 48 hours, and 7 days. 

RNA Isolation and Northern Blot Analysis

Total RNA was isolated from 35-mm petri dishes by the guanidinium thiocyanate–phenol–chloroform extraction method (RNA isolation kit, Stratagene, Heidelberg, Germany). RNA (15 μg) was denatured and size-fractionated by gel electrophoresis in 1% agarose gels containing 2.2 M formaldehyde. The RNA was then vacuum-blotted onto a nylon membrane (Boehringer Mannheim) and cross-linked (1600μ J, Stratalinker; Stratagene). To assess the amount and quality of the RNA, the membrane was stained with methylene blue. Images were taken with the Lumi-Imager (Boehringer Mannheim). Prehybridizations were performed at 68°C for 1 hour in Dig Easy Hyb (Boehringer Mannheim). Hybridizations were at 68°C overnight in prehybridization solution containing 50 ng/ml αB-crystallin–specific 450-bp antisense riboprobe. Riboprobes were synthesized by using a combined polymerase chain reaction (PCR). Briefly, a DNA fragment was amplified from cDNA of TM cells using the reaction conditions and primers described previously, ¹¹ except that the downstream primer (MWG-Biotech, Ebersberg, Germany) contained the sequence for the T7-promoter. After purification with a Quiagen (Hilden, Germany) PCR Purification Kit, 1 μg DNA was used as a template for in vitro transcription using the digoxigenin labeling RNA Kit from Boehringer Mannheim. Labeling efficiency was checked by direct detection of the labeled RNA probe with anti–digoxigenin–alkaline phosphatase. After hybridization, the membrane was washed twice with 2× SSC, 0.1% sodium dodecyl sulfate (SDS) at room temperature, followed by two washes in 0.1× SSC, 0.1% SDS, for 15 minutes at 68°C. After hybridization and posthybridization washes, the membrane was washed for 5 minutes in washing buffer (100 mM maleic acid, 150 mM NaCl; pH 7.5; 0.3% Tween-20) and incubated for 60 minutes in blocking solution (100 mM maleic acid, 150 mM NaCl, pH 7.5, 1% blocking reagent, Boehringer Mannheim). Anti–digoxigenin–alkaline phosphatase (Boehringer Mannheim) was diluted 1:10,000 in blocking solution, and the membrane incubated for 30 minutes. The membrane was washed four times, 15 minutes per wash, in washing buffer. The membrane was equilibrated in detection buffer (100 mM Tris–HCl, 100 mM NaCl, pH 9.5) for 10 minutes. For chemiluminescence detection CDP-star (Boehringer Mannheim) was diluted 1:100 in detection buffer, and the filter incubated for 5 minutes at room temperature. After air-drying, the semidry membrane was sealed in a plastic bag. Chemiluminescence was detected with the Lumi-Imager workstation. Exposure times ranged between 10 minutes and 1 hour. The quantification was performed with Lumi Analyst (Boehringer Mannheim). 

Western Blot Analysis

Cells grown on tissue culture dishes were washed twice with PBS (pH 7.2), collected, and lysed in SDS sample buffer for gel analysis. ²⁵ The samples for gel analysis were boiled for 5 minutes, and protein content was measured using BCA protein assay reagent (Pierce, Rockford, IL). For analysis of the proteins (2 μg), a 5% SDS–polyacrylamide gel electrophoresis (PAGE) for stacking gel and 12% SDS–PAGE for separating gel were used. After electrophoresis the proteins were transferred with semidry blotting onto a polyvinyl difluoride membrane (Boehringer Mannheim). The membrane was incubated with PBS containing 0.1% Tween-20 (PBST, pH 7.2) and 5% bovine serum albumin for 1 hour. The primary antibody (αB-crystallin diluted 1:4000) ⁴ was then added and allowed to react overnight at room temperature. After washing three times in PBST, an alkaline phosphatase–conjugated swine–anti-rabbit antibody (diluted 1:20,000; Dianova, Hamburg, Germany) was added for 30 minutes. Visualization of the alkaline phosphatase was achieved using chemiluminescence. CDP-star was diluted 1:100 in detection buffer, and the filter was incubated for 5 minutes at room temperature. After air-drying the semidry membrane was sealed in a plastic bag. Chemiluminescence was detected with the Lumi-Imager workstation. Exposure times ranged between 1 and 5 minutes. The quantification was performed with Lumi Analyst. 

Results

Immunoelectronmicroscopy

Immunogold labeling for αB-crystallin was seen only in the 3 to 5 layers of cribriform cells (Fig. 1) . The endothelial cells lining Schlemm‘s canal as well as the TM cells covering the corneoscleral and uveal trabecular lamellae were completely unstained. 

Cell Culture

In phase-contrast microscopy, cells derived from the cribriform and corneoscleral TM showed slight differences in their morphology similar to what has been described previously. ^{9 11} The cribriform cells appeared somewhat larger than the corneoscleral cells. Cribriform cells stained intensively for αB-crystallin, whereas most of the corneoscleral cells remained unstained (Figs. 2 A, 2B). These staining differences were seen in all cultures derived from all five donors. None of the cultures stained for desmin, only single cells in both TM cultures stained for α-sm actin. 

Effects of TGF–β and DEX on αB-Crystallin mRNA

Northern blot analysis of untreated corneoscleral TM cells showed two faint bands, which were approximately 0.8 and 1.1 kb in length. Untreated cribriform TM cells showed a much higher amount ofα B-crystallin mRNA in the 0.8-kb band (3 times; Fig. 3 ), whereas the 1.1-kb band appeared only slightly increased. Additionally, a band at approximately 2.6 to 2.9 kb hybridized with theα B-crystallin probe. 

Treatment of both corneoscleral and cribriform TM cells with 1.0 ng/ml TGF-β1 or TGF-β2 significantly increased the amount ofα B-crystallin mRNA. There were, however, differences in the responsiveness of the two cell types. In the cribriform TM cells the increase ranged between 2 and 3 times (Fig. 3) , whereas the corneoscleral TM cells, which had a 3 times lower basal level than the cribriform TM cells, showed a much higher increase (6–8 times) ofα B-crystallin mRNA after 12-hour treatment with TGF-β1 or TGF-β2. In both cell types the increase was mainly in the 0.8-kb band, whereas the 1.1-kb band nearly disappeared. Additionally, in cribriform and corneoscleral TM cell northern blot analysis, the αB-crystallin probe hybridized with the 2.6- to 2.9-kb band after treatment with TGF-β. Treatment of the corneoscleral or cribriform TM cells for 24, 48, or even 96 hours with TGF-β1 and TGF-β2 showed the same results (Fig. 3 , Table 1 ). 

Withdrawal of TGF-β1 and TGF-β2 after 24-hour treatment of corneoscleral TM cells led to a normalization of the elevatedα B-crystallin mRNA levels after 24 hours (Fig. 4) . 

When corneoscleral TM cells were treated with HCl or a combination of TGF-β2 and a neutralizing antibody, only TGF-β2 alone inducedα B-crystallin mRNA (data not shown). 

Treatment with DEX for 12, 24, 48, and 96 hours had no effect onα B-crystallin mRNA expression, either in the cribriform or in the corneoscleral cells (Fig. 3 , Table 1 ). 

Effect of TGF-β and DEX on αB-Crystallin Expression

Using western blot analysis, the cribriform TM cells showed a 6 times higher amount of αB-crystallin than the corneoscleral TM cells (Fig. 5) . 

After treatment with TGF-β1 and TGF-β2 for 12 or 96 hours there was a significant increase in the amount of αB-crystallin in the corneoscleral cells, whereas the increase in the cribriform TM cells was only small (Fig. 5) . In the corneoscleral TM cells αB-crystallin increased after 12 hours of treatment with TGF-β1 and TGF-β2 between 4.5 and 6.8 times, whereas in the cribriform TM cells the increase ranged between 2 and 3 times. After 96 hours of treatment, the increase in the corneoscleral and cribriform TM cells was in the same range. 

Treatment with DEX for 12 to 96 hours had no effect on the amount ofα B-crystallin, either in cribriform or in corneoscleral TM cells. 

Treatment of NIH 3T3 cells and HFFs

NIH 3T3 and HFFs showed a clear difference in their response to DEX and TGF-β when compared with TM cells. In NIH 3T3 cells treatment with TGF-β1 and TGF-β2 for 24 hours, 48 hours, and 7 days did not induce αB-crystallin mRNA. On the other hand, αB-crystallin mRNA was induced by treatment with DEX. This induction was seen after 24 hours of DEX treatment and increased after 48 hours and 7 days of treatment (Fig. 6) . 

In HFF neither treatment with TGF-β1 or TGF-β2 for the different time spans nor treatment with DEX had any effect on the induction ofα B-crystallin mRNA (Fig. 7) . Untreated and treated HFFs showed a faint hybridization of the antisense αB-crystallin probe at 0.8 kb. 

Discussion

The induction of αB-crystallin has so far only been shown in various cell cultures after heat shock, oxidative damage, and osmotic or mechanical stress. ^{26 27 28} The present study demonstrates for the first time that TGF-β, a growth factor known to be increased in the aqueous humor of a number of glaucomatous eyes, stimulates expression of αB-crystallin in TM cells. In both cultures of trabecular cells, αB-crystallin mRNA and protein expression were not induced by treatment with DEX. On the other hand, αB-crystallin mRNA and protein expression were induced by TGF-β treatment, which had no effect on NIH 3T3 fibroblasts or on HFF cell lines. 

Before treatment, western blot analysis showed clear differences in the amount of αB-crystallin between corneoscleral and cribriform TM cells. Cribriform TM cells constitutively express the proteinα B-crystallin, whereas in corneoscleral TM cells the protein is virtually absent. These results confirmed the immunohistochemical differences between these two cell types. By northern blot analysis we have demonstrated that parallel to the protein expression, the amount of αB-crystallin mRNA is high in cribriform TM cells but very low or even undetectable in corneoscleral TM cells. 

Stimulation at the mRNA level is paralleled by a similar effect at the protein level. As can be concluded from Table 1 , treatment of corneoscleral TM cells with growth factors TGF-β1 and -β2 had a pronounced effect on αB-crystallin mRNA and protein levels, whereas in the cribriform TM cells the increase was only small. These findings indicate that high constitutive levels of αB-crystallin result in lowered induction of mRNA and protein. 

Our findings lead to the assumption that TGF-β inducesα B-crystallin mRNA expression in TM cells at the transcriptional level. For HSP70 and HSP90 it has previously been shown that TGF-β acts in this way. ²⁹ Induction of αB-crystallin during heat shock, hypertonic stress, cadmium exposure, and treatment with TNF-α is also thought to be mediated at the level of transcription. ³⁰ In the 5′flanking region of the humanα B-crystallin gene a considerable number of cis-regulatory sequences have been described, including binding sites for the heat shock transcription factor 1 (HSF1) ^{31 32} and an alkaline phosphatase-1–like consensus sequence. ³⁰ In rat astrocytes two different transcriptional regulation mechanisms ofα B-crystallin mRNA were shown. ³⁰ After cadmium exposure, an increase in αB-crystallin mRNA level and activation of HSF1 were demonstrated. With hypertonic stress αB-crystallin mRNA was induced, but no activation of HSF1 took place. The elucidation of the exact mechanism of αB-crystallin induction by TGF-β awaits further experiments. 

It is not known why TGF-β1 and TGF-β2 stimulate αB-crystallin expression in TM cells but not in fibroblasts, and it is unknown why treatment with DEX induces αB-crystallin in NIH 3T3 cells but not in TM cells or HFFs. In a number of glaucomatous eyes nearly the entire TM but not the adjacent fibroblasts are stained with antibodies againstα B-crystallin, reflecting differences in stimulation of protein expression between the two cell types also in vivo. ¹² Because TGF-β has been shown to be increased in the aqueous humor of glaucomatous eyes, ¹⁹ this increase might be involved in the increase of αB-crystallin levels in glaucomatous TM. 

 Supported by Grant BIOMED PL 961593 of the European Commission to HB and ELD, grant SFB 539 (Glaukome) der Deutschen Forschungsgemeinschaft Bonn, and the Academy of Science, Mainz, Germany.

 Submitted for publication February 3, 1999; accepted March 29, 1999.

 Proprietary interest category: N.

 Corresponding author: Elke Lütjen–Drecoll, Department of Anatomy II, University of Erlangen–Nürnberg, Universitätsstr. 19, D-91054 Erlangen, Germany. E-mail: [e-mail]

 

Figure 1.

 

Ultrathin sagittal section through the inner wall of Schlemm’s canal (SC) and the adjacent cribriform layer of the trabecular meshwork (TM) in a cynomolgus monkey, stained for αB-crystallin (×34,000). Note that all layers of cribriform TM cells show immunogold labeling forα B-crystallin, whereas the endothelium (arrowhead) of SC is completely unstained.

Figure 2.

 

Immunofluorescence staining for αB-crystallin in monolayer cultures of cribriform TM (A) and of corneoscleral TM (B) cells (donor male, 57 years of age, ×200). (A) Cribriform TM cells stained intensively for αB-crystallin. (B) In monolayers of corneoscleral TM cells only single cells were stained.

Figure 3.

 

Northern blot analysis of αB-crystallin mRNA in confluent corneoscleral and cribriform TM cells (TMC) 12 and 96 hours after treatment with either 1.0 ng/ml TGF-β1 or TGF-β2 or 5 × 10⁻⁷ M DEX. Note the higher basal level in the untreated controls (Co.) of cribriform cells, whereas in the corneoscleral cells nearly no αB-crystallin mRNA is detectable. Methylene blue staining of the 28S and 18S rRNA bands is also shown, demonstrating relative integrity and even loading of the RNA. MW, molecular weight; RDI, relative densitometric intensity (normalized to 28S rRNA).

Table 1.

 

Relative Increase in αB-Crystallin mRNA and Protein after 12 and 96 Hours of Treatment of TM Monolayers with either TGF-β1, TGF-β2, or DEX

	Control	DEX	TGF-β1	TGF-β2	Time, h
Corneoscleral TM (mRNA)	1	1	6.5	8.5	12
Corneoscleral TM (Protein)	1	0.8	4.8	6.4	12
Cribriform TM (mRNA)	1	1	2.5	3.2	12
Cribriform TM (protein)	1	1	1.8	2.4	12
Corneoscleral TM (mRNA)	1	1	5.2	5.5	96
Corneoscleral TM (protein)	1	1.2	7.8	9.6	96
Cribriform TM (mRNA)	1	0.9	2.3	2.6	96
Cribriform TM (protein)	1	1.1	1.4	1.6	96

Figure 4.

 

Northern blot analysis of αB-crystallin mRNA in confluent corneoscleral TM cells 24 hours after treatment with either HCl solution, 1.0 ng/ml TGF-β1 or TGF-β2, and replacement of the TGF-β–containing medium with serum-free medium for 12 to 24 hours. Methylene blue staining of the 28S and 18S rRNA bands is also shown, demonstrating relative integrity and even loading of the RNA. MW, molecular weight; Co., control; RDI, relative densitometric intensity (normalized to 28S rRNA).

Figure 5.

 

Western blot analysis of αB-crystallin in corneoscleral and cribriform TM cell (TMC) monolayers 12 hours after treatment with either 1.0 ng/ml TGF-β1 or TGF-β2, or 5 × 10⁻⁷ M DEX. Lysates from approximately equal amounts of protein (2 μg) were separated by SDS-PAGE and blotted for immunochemical detection ofα B-crystallin content as described in the Materials and Methods section. The number below each band shows the chemiluminescence measurement. MW, molecular weight; Co., control.

Figure 6.

 

Northern blot analysis of αB-crystallin mRNA in NIH 3T3 cells 24 hours, 48 hours, and 7 days after treatment with either HCl solution, 1.0 ng/ml TGF-β1 or TGF-β2, or 5 × 10⁻⁷ M DEX. Methylene blue staining of the 28S and 18S rRNA bands is also shown, demonstrating relative integrity and even loading of the RNA. MW, molecular weight; Co., control.

Figure 7.

 

Northern blot analysis of αB-crystallin mRNA in HFFs 96 hours after treatment with either HCl solution, 1.0 ng/ml TGF-β1 or TGF-β2, or 5 × 10⁻⁷ M DEX. As a positive control for hybridization conditions, cribriform TM cells (TMC) are shown after treatment with TGF-β. Methylene blue staining of the 28S and 18S rRNA bands is also shown, demonstrating relative integrity and even loading of the RNA. MW, molecular weight; Co., control.

We thank Sandra Hartmann, Angelika Pach, Gerti Link, and Marco Gößwein for expert technical assistance.