Glucocorticoid effects on extracellular matrix proteins and integrins in ...

2 downloads 0 Views 2MB Size Report
Abstract. The trabecular meshwork (TM) is a specialized eye tissue essential for regulation of the aqueous humor outflow and control of the intraocular pressure.
INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 1: 339-346, 1998

Glucocorticoid effects on extracellular matrix proteins and integrins in bovine trabecular meshwork cells in relation to glaucoma LILI ZHOU, YUHONG LI and BEATRICE Y.J.T. YUE Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago College of Medicine, Chicago, IL 60612, USA Received November 19, 1997; Accepted December 12, 1997

Abstract. The trabecular meshwork (TM) is a specialized eye tissue essential for regulation of the aqueous humor outflow and control of the intraocular pressure. Disturbances of TM cells may lead to elevated intraocular pressure and glaucoma. This study assessed the dexamethasone effects on levels of extracellular matrix proteins and their integrin receptors in bovine TM cells. Instillation of glucocorticoids such as dexa­ methasone is known to result in ocular hypertension. The histologic changes induced resemble those seen in glaucoma. Examination of the effects of glucocorticoid therefore may provide insights into the pathogenesis of glaucoma. TM cells in either tissue culture or organ cultures were treated with 0 (control), 0.1, or 1 |lM of dexamethasone for 72 h. Immunostaining, Western, Northern and dot blot analyses showed that dexamethasone caused an increase in levels of fibronectin and collagen type IV in tissue-cultured TM cells. Increased focal contacts were also observed but the levels of laminin and collagen type I were unaffected. The dexamethasone effect was similarly demonstrated in organ cultures, with the exception that collagen type I also was enhanced. These results suggest that dexamethasone modulates extracellular matrices in the TM. Glucocorticoid may exert its effect through such a modulation in the development of steroid glaucoma. Introduction The trabecular meshwork (TM) is a specialized tissue located in the anterior chamber of the eye neighboring the cornea. It is a major site for regulation of the aqueous humor outflow and plays an important role in the control of the intraocular pressure (IOP) (1). The TM tissue is a mesh-like structure composed of sheets of trabecular beams made up of extra­ cellular matrix (ECM) elements (2). Covering the beams are TM cells that are believed to be essential for maintenance of

the outflow pathway (3). Disturbances in the vitality and functional status by genetic predisposition, aging, or other insults may result in obstruction of the aqueous outflow, leading to IOP elevation and glaucomatous conditions. Glucocorticoids, when administrated either systematically (4) or topically (5-8), can lead to the development of ocular hypertension in a subset of the population. Ocular instillation of glucocorticoids such as dexamethasone also has resulted in an IOP elevation in rabbits (9). The histologic changes induced in the TM have been shown to bear a resemblance to those seen in primary open angle glaucoma (10-12). It is therefore suggested that examination of the glucocorticoid effects may provide insights into the pathogenesis of this most common form of glaucomas. This notion is further reinforced by recent evidence that a gene identified as TMinduced glucocorticoid response (TIGR) (3,13) is linked to both juvenile and primary open angle glaucomas (14). The exact nature and the functional role of TIGR is unclear at present. One mechanism through which glucocorticoid exerts its effect or TIGR alters the TM functions may be ECM modulations. A previous study (15) showed that in perfusion culture of human eyes 0.1 |oM of dexamethasone induced IOP elevation and ECM accumulation in the TM. An abnormal accumulation of materials resembling basement membranes (16), presumably produced by TM cells, has also been observed in the eyes of steroid-induced glaucoma. ECM components, including fibronectin, laminin, and collagens, are believed to be important for the proper functioning of TM tissues. We therefore assessed the glucocorticoid effects on the levels of four major ECM components, fibronectin, laminin, and collagen types I and IV, in TM cells and on the distribution of integrin receptors. Both tissue culture and organ culture systems were used. Materials and methods

Correspondence to: Dr Beatrice Y.J.T. Yue, Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago, 1855 W. Taylor Street, Chicago, IL 60612, USA Key words: dexamethasone, extracellular matrix, glaucoma, organ culture, tissue culture, trabecular meshwork

Tissue culture. Freshly enucleated bovine eyes were obtained from a local slaughterhouse. Under sterile conditions, excess orbital and conjunctival tissues were trimmed off the globes. Eyes were soaked for 1 min in Providone (Baxter, Deerfield, IL), rinsed in phosphate buffered saline (PBS), and incubated in a solution of penicillin and streptomycin (Sigma Chemical Co., St. Louis, MO). The TM tissues were then dissected and cultured as previously described (16,17) in Dulbecco's modified

340

ZHOU et al: DEXAMETHASONE EFFECTS ON THE TRABECULAR MESHWORK

Eagle's minimum essential medium (MEM) supplemented with 10% fetal calf serum, 5% calf serum, nonessential and essential amino acids and antibiotics (17,18). When confluency was reached, the cells were trypsinized and passaged. First- or second-passaged cells were plated either at 20,000 cells per chamber on glass chamber slides coated with 10 jJ-g/ml of fibronectin or laminin for immunostaining, or at 400,000 cells per well on 6-well plates for Western, Northern and dot blot analysis. After exposure to 0 (control), 0.1 or 1 |J.M of dexamethasone in serum-free MEM for 3 days, cells in chamber slides were fixed for 15 min in cold methanol or paraformaldehyde-lysine-periodate fixative (pH 6.2), and permeabilized in 0.2% Triton X-100 (19) for immunostaining. The levels of fibronectin, laminin, and collagen types I and IV accumulated during the last 24 h of the dexamethasone treatment were visualized by Western blotting and measured by dot blot assays. The mRNA levels were examined by Northern blot analysis. Organ culture. After soaking in antibiotics, bovine eyes from the slaughterhouse were bisected at the equatorial meridian. The vitreous, lens and iris were removed, and the cornea and sclera were placed in an incubation assembly containing three chambers, constructed according to the description by Johnson and Tschumper (20). Each chamber unit had a tightfitting O-ring that sealed the scleral rim to create a 'closed eye'. Two 21-gauge cannulas built into the base of the chamber allowed access to the anterior chamber of the 'closed eye'. One cannula, connected to a Sage syringe pump (Orion Research, Boston, MA) via a sterilized tubing system and a Millipore filter, was perfused with serum-free MEM. Super­ ficial slits were made on the 'closed eye' 3 mm from the limbus posteriorly to allow the medium to flow through the aqueous veins (20,21). The organ culture units were assembled, sealed watertight and perfused (16.7 p:l/min) at 37°C for 3 days in a C0 2 incubator with 0, 0.1 and 1 \\M of dexamethasone in MEM. The medium would flow through the eye in a normal fashion, entering the TM, Schlemm's canal and collector channels. The IOP was measured with a pressure transducer. After the perfusion, the anterior segment was perfusionfixed in 4% formaldehyde. The tissues were dehydrated and embedded in paraffin, and 5-pim thick sections were prepared for immunostaining. Tissue extracts were also obtained for Western, dot blot and Northern blot analyses. Immunostaining. Immunostaining for fibronectin, laminin and collagen types I and IV was performed as previously described (19,22). In brief, the fixed cells or deparaffinized sections were incubated first in 10% normal goat blocking serum for 20 min and then with affinity-purified and monospecific rabbit anti-human fibronectin (1:150 dilution, Collaborative Research, Bedford, MA), anti-mouse laminin (1:50, Collaborative Research), anti-human collagen type I (1:100, HEYL, Houston, TX), anti-mouse collagen type IV (1:100, Collaborative Research), and antibodies to human a5ßl, avß3 or ßl integrin (1:50, Chemicon, Temecular, CA) for 1.5 h. After rinses with PBS, the specimens were further incubated with biotinylated goat anti-rabbit IgG (1:250, Vector Laboratories, Burlingame, CA) for 30 min, soaked in 0.3% H202-methanol for 20 min, and incubated with avidin-biotin-

horseradish-peroxidase complex (ABC, Vector) for 30 min. The color reaction was developed with 3,3-diaminobenzidine tetrahydrochloride (Sigma). The slides were dehydrated, mounted and photographed. The brown reaction products were examined under light microscopy. For negative controls, normal rabbit IgG was used in place of the primary antibody. For fibronectin and collagen types I and IV, some experiments were performed using an ExtrAvidin-biotinalkaline phosphatase complex method. The slides were incubated with ExtrAvidin-alkaline phosphatase conjugate (1:50, Sigma) after the secondary antibody incubation. The pink color was developed with the enzyme substrate Fast Red (Sigma). The distribution of integrins in tissue-cultured TM cells was examined by immunofluorescence methods. The cells were blocked in 10% heat-inactivated normal goat serum, incubated sequentially with primary antibody and fluorescein (FITC)-labeled goat anti-rabbit IgG (1:20, Cappel Laboratories, Burlingame, CA) and examined by fluorescence microscopy. Western blot and dot blot assays. After dexamethasone treatment, tissue culture medium was collected and the TM cells were harvested. For organ culture experiments, the entire TM tissues were dissected out. The cells or tissues were homogenized at 4°C in 0.1 M Tris buffer (pH 7.2) and 0.154 M NaCl. The homogenates were centrifuged (19,22), and the protein contents in the supernatants were measured (23) by Bradford's protein assay (Bio-Rad Laboratories, Richmond, CA) using bovine serum albumin as a standard. For Western blotting, aliquots of culture medium or tissue extracts containing 10 jig of protein along with human fibro­ nectin and mouse laminin (both from Collaborative Research) standards, were electrophoresed on 6% SDS gels under reducing conditions as previously described (24). The proteins were electroblotted onto a nitrocellulose membrane (0.2 |J,m, Schleicher and Schuell, Keine, NH). After blocking, the membrane was allowed to react with either polyclonal antifibronectin (1:10,000) or anti-laminin (1:2,500), followed by incubation with goat anti-rabbit IgG conjugated to horseradish peroxidase (1:10,000; Cappel, Durham, NC). Immunoreactive bands were visualized using an enhanced chemiluminescence detection system (Pierce, Rockford, IL). Prestained molecular weight markers (Bio-Rad) were run in parallel. For dot blot analyses (24), aliquots (50 u.1, in serial dilutions) of each TM cell or tissue sample were loaded onto nitrocellulose membranes in a 96-well dot blot apparatus. Fibronectin, laminin, bovine collagen type I (HEYL) and mouse collagen type IV (Collaborative Research) were used as standards. The membranes were blocked; they were then incubated with rabbit anti-human fibronectin, anti-laminin, or anti-collagen types I or IV (1:10,000) and allowed to react with horseradish peroxidase-labeled goat anti-rabbit IgG. The spots were visualized by ECL and were quantified by the microplate reader. Data from triplicate samples were averaged. Nonspecific binding was determined on duplicate blots incubated with only normal rabbit IgG and the secondary antibody. These were subtracted from the total binding to determine the specific binding. Resultant values were then compared with the standard curve obtained with known amounts of each protein. The dot blot values were normalized

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 1: 339-346, 1998

341

Figure 1. Immunostaining of tissue-cultured bovine trabecular meshwork cells for fibronectin. Cells in the first passage were treated for 72 h with (A), 0 (control); (B), 0.1; and (C), 1 |iM of dexamethasone. Positive staining appears as pink deposits (original magnification, x40). Note the increased staining in dexamethasone-treated specimens (B and C) compared to the control (A).

Figure 2. Immunostaining of tissue-cultured bovine trabecular meshwork cells for collagen type I. Cells in the first passage were treated for 72 h with (A), 0 (control); (B), 0.1; and (C), 1 u,M of dexamethasone. Positive staining appears as pink deposits (original magnification, x40). Note that no significant difference in staining is observed between dexamethasone-treated (B and C) and the control (A) specimens.

Figure 3. Western blot analysis using a polyclonal antibody to fibronectin (A) and collagen type I (B). Lanes 1-3 show extracts from tissue-cultured bovine trabecular meshwork cells treated, respectively, with 0 (control), 0.1 and 1 U.M of dexamethasone. The fibronectin intensity (A) in lane 3 is higher than that in lanes 1 and 2. The intensity of collagen type I bands (B) is not increased in dexamethasone-treated samples (lanes 2 and 3 compared to lane 1).

Collection, Rockville, MD), a5 (0.6 kb) and ßl (1.2 kb, Life Technologies, Gaithersburg, MD) and glyceraldehyde-3phosphate dehydrogenase (G3PDH, 1.27 kb, Clontech, Palo Alto, CA) cDNA probes were prepared using a random priming kit (Boehringer-Manheim, Indianapolis, IN). The nylon membranes were prehybridized and hybridized with the 32P-labeled cDNA, and were subsequently washed in a 2X standard sodium chloride-sodium phosphate-EDTA (SSPE) solution containing 0.1% SDS at room temperature, and in IX SSPE, 0.1% SDS and 0.1X SSPE, 0.1% SDS at 60°C. They were allowed to dry, and autoradiography was performed to detect the 32P-labeled cDNA-mRNA hybrids. The relative fibronectin and integrin to G3PDH mRNA levels in the samples were measured with a densitometer (Bioimage, Ann Arbor, MI). Results

to the protein content, and Student's t-test was used to determine the statistical significance of the data. Northern blot analysis. Total RNA was extracted from tissuecultured bovine TM cells after dexamethasone treatment by the acid guanidinium thiocyanate-phenol-chloroform method (18,19). The isolated RNA samples were electrophoresed (20 jig per lane) on 1.5% formaldehyde-agarose gel and transferred to nylon membranes (Amersham, Arlington Heights, IL). The membranes were baked at 80°C for 1 h and were exposed to ultraviolet light for 1.5 min to immobilize the RNA. 32P-labeled fibronectin (1.3 kb, American Type Culture

Tissue culture studies. Moderate to intense immunoreactivity for fibronectin (Fig. 1), laminin and collagen types I (Fig. 2) and IV was observed in cultures of bovine TM cells. Compared with controls, immunostaining for fibronectin (Fig. 1) and collagen type IV was stronger in cultures treated with dexamethasone. The treatment did not elicit any visible changes in either the pattern or the intensity of staining for laminin and collagen type I (Fig. 2). Using a fibronectin antibody, Western blotting (Fig. 3A) of the cell extracts showed a major protein band corresponding to fibronectin molecules (molecular weight, 220 kDa). The intensity of the fibronectin band appeared to be increased by

ZHOU et al: DEXAMETHASONE EFFECTS ON THE TRABECULAR MESHWORK

342

Figure 4. Immunofluorescence staining of tissue-cultured bovine trabecular meshwork cells for integrins avß3 (A and B) and u5ß 1 (C and D). Cells in the first passage were incubated for 72 h without (A and C) or with (B and D) 1 u,M of dexamethasone. Note the increased focal contact formation in dexamethasonetreated cells (B and D) compared to the controls (A and C). (Original magnification, x40).

Table I. Levels of fibronectin, laminin, and collagen in tissue cultures of bovine trabecular meshwork cells. Dexamethasone Level (ng/fxg protein)

0 (control)

0.1 uM

1 U.M

Fibronectin

73.4+8.9 (n=4)

120.0±8.0 (n=4)

156.9+28.1» (n=4)

Laminin

1.18+0.13 (n=5)

1.36±0.22 (n=5)

1.29+0.14 (n-5)

Collagen type I

147.4+18.5 (n=5)

146.6+10.7 (n=4)

153.9+22.7 (n=4)

Collagen type IV

111.2+10.8 (n=5)

169.3+10.8" (n=4)

177.6+12.0" (n=3)

Tissue cultured cells were treated with 0, 0.1 or 1 uM of dexamethasone for 72 h. The level of each protein was determined by dot blot assays. aP