Presence of a Transcriptionally Active Glucocorticoid Receptor α in Lens Epithelial Cells

Abstract

purpose. The purpose of this study was to determine whether lens epithelial cells (LECs) contain a glucocorticoid receptor (GR) that is transcriptionally active and that is able to induce production of known glucocorticoid-inducible proteins.

methods. Protein and mRNA were obtained from human, rabbit, and bovine lens epithelia and from cultured human lens epithelial cells (B3, hLECs) and rabbit lens epithelial cells (N/N1003A, rLECs). Paraffin-embedded sections were prepared from human lenses for immunohistochemical localization of GR. RT-PCR was performed to amplify portions of GR, and the products were sequenced. Protein samples were analyzed by Western blot. hLECs and rLECs were transfected with pTAT3-luc and assayed for luciferase activity after treatment with dexamethasone (Dex) and/or RU486. Dex-treated LECs were also analyzed by quantitative real-time PCR and by Western blot for expression of specific mRNA and proteins.

results. By PCR and sequencing, products consistent with GR sequences were obtained from human, rabbit, and bovine lenses and from hLECs and rLECs. The complete GRα sequence was obtained from rLECs and was found to be 89% identical with human GR. A 1757-bp 3′ fragment of bovine GRα cDNA was also amplified from bovine lens. By Western blot, bands of approximately 94 kDa, the expected size of GR, were identified from human, rabbit, and bovine lens samples and from hLECs and rLECs, using anti-GR antibodies. Anti-GR antisera localized GR to both the cytosol of anterior and bow region LECs and to the nuclei of epithelial and early-differentiating lens fiber cells. Luciferase expression was induced in pTAT3-luc–transfected rLECs and hLECs by Dex treatment and this expression was partially (rLECs) or completely (hLECs) blocked by pretreatment with RU486. mRNA levels for type-1 glucocorticoid-induced target genes and also mRNA and protein levels for type-2 genes were upregulated after Dex exposure.

conclusions. The data confirm the existence of GR in hLECs, indicate that GR is present in rLECs, and resolve the controversy over the presence of GR in bovine lens. The GRα in hLECs and rLECs was shown to be transcriptionally active and the expression levels in hLECs of mRNAs and proteins known to be regulated by glucocorticoids were modified in these cells by glucocorticoid treatment.

As a consequence of the long-term application of steroids for treatments associated with conditions such as allergy, autoimmune disease, and transplantation, there is a high risk of development of posterior subcapsular cataract (PSC). ^{1 2 3} Several mechanisms have been proposed for the induction of steroid-induced PSC, and studies have focused principally on two areas: the binding of steroids to lens crystallins and the oxidation of lens proteins after steroid administration. ^{4 5 6 7} The evidence has not yet linked steroid treatment through either of these mechanisms to the generation of steroid-induced cataract. 

Steroids have been reported to be capable of binding to lens proteins ^{7 8} ; however, evidence against a steroid binding mechanism for PSC induction has been advanced by Dickerson et al. ⁷ who reported findings more consistent with activation of a glucocorticoid receptor (GR) and with steroid binding to lens proteins playing an incidental role. Lenses treated with glucocorticoids in vitro become opacified. Some reports have indicated that this opacification can be prevented by administration of α-tocopherol, ^{5 6} suggesting a role for free radicals. A reduction in the level of glutathione (GSH), in other studies of glucocorticoid-treated lenses ^{7 9} may support a role for oxidants in this opacification. However, only steroids with glucocorticoid activity possessed this ability and the reduction could be substantially prevented by the glucocorticoid antagonist, RU38486 (RU486), further suggesting a possible role for GR. The recent study by Lyu et al., ¹⁰ who found no GSH reduction in rat lenses coincident with the development of posterior opacities, concurs with a role for GR rather than oxidants in the development of PSC. 

Although this circumstantial evidence may suggest GR involvement in changes to lens cells that could lead or contribute to steroid-induced cataract, the existence of a functional GR in lens epithelial cells (LECs) remained to be demonstrated. Immunohistochemical studies by Stokes et al. ¹¹ and Suzuki et al. ¹² indicated reactivity to anti-GR antibodies in lens epithelium, but anti-GR antibodies can be temperamental requiring additional confirmatory evidence. Early studies ^{13 14} of glucocorticoid binding to lens proteins were consistent with the presence of GR in lens epithelium, but the Western blot analysis and binding assays of Jobling and Augusteyn ⁴ failed to indicate the presence of GR in bovine lens epithelium. Two recent studies of human and mouse epithelial cells ¹⁵ and rat lenses ¹⁰ have provided evidence for the existence of lens GR. 

The purpose of this study was to determine the existence of a functional GR in LECs. Our results using polymerase chain reaction (PCR), Western blot analysis, and immunohistochemistry confirmed the presence of GR in human lens cells and indicate that GR is present in rabbit and bovine lens cells, thus resolving the controversy ⁴ over the presence of GR in the bovine lens. Furthermore, sequencing data indicate that the lens cells for all three species contain GRα, the active isoform of GR. Treatment of cultured LECs with Dex induced expression from the pTAT3-luc expression vector indicating that the GRα in LECs can induce transcription from positive glucocorticoid-responsive elements (GREs). In addition, transcription of mRNAs for direct type-1 glucocorticoid-induced genes (that contain positive GREs) and mRNA and protein for indirect type-2 glucocorticoid-induced genes (reported in other studies to be regulated by glucocorticoids) were induced in LECs by Dex treatment, indicating that the GRα in lens cells is transcriptionally active and functional. 

Materials and Methods

Cells and Tissues

The rabbit lens epithelial cell line (rLECs), N/N1003A (kindly provided by John Reddan, Oakland University, Rochester, MI), was maintained in culture with Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich Co., St. Louis, MO) containing 10% rabbit serum, 2.2 mg/mL sodium bicarbonate, and 50 μg/mL gentamicin (Sigma-Aldrich Co.). The human lens epithelial cell (hLEC) line, B-3, was obtained from the American Type Culture Collections (ATCC, Manassas, VA) and maintained in DMEM with 2 mM l-glutamine, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 1.5 mg/mL sodium bicarbonate, 20% fetal bovine serum, and 100 U/mL penicillin-streptomycin. rLECs and hLECs were maintained according to protocols described by Reddan et al. ¹⁶ and ATCC product sheet CRL-11421, respectively. For cell cultures exposed to Dex (Sigma-Aldrich Co.) and RU486 (Sigma-Aldrich Co.), the steroid and antagonist were added from 10- or 100-mM stock solutions prepared in dimethylsulfoxide or ethanol respectively. Control samples were exposed to solvent only. 

Bovine lenses were obtained from a local abattoir, and rabbit lenses from the Department of Comparative Medicine, Medical University of South Carolina (MUSC). The use of animals conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Human lenses, designated for research after receipt of informed consent, were obtained through the services of the South Carolina Lions Eye Bank. With a dissecting microscope, the anterior chamber of each eye was opened, the iris removed, the zonule attachments severed, and the lenses removed without adherent tissues. From a central anterior rhexis to just anterior of the bow region, the lens capsule with adherent LECs was peeled away from the fiber cells. The LECs with attached capsule were immediately processed for RNA and protein extraction and stored at −20°C. 

PCR and Sequencing

mRNA was isolated from rLECs and hLECs using extraction reagent (Trizol; Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol, and first-strand cDNA (fs cDNA) was prepared using avian myeloblastosis virus (AMV) reverse transcriptase, also according to the manufacturer’s instructions (Invitrogen). Oligonucleotide primers (Table 1) for PCR detection and amplification of rabbit GR were designed to regions of homology between published human GR (GenBank accession no. U01351) and mouse GR (accession no. X04435) sequences (http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD). These primers (human GR nucleotide numbering) were also used to amplify bovine GR from lens fs cDNA. Reaction conditions were: dissociation at 95°C for 30 seconds, annealing at 52°C for 1 minute, and extension at 72°C for 2 minutes for 35 cycles. Reaction products were detected by ethidium bromide 2% agarose gel electrophoresis. DNA excised from the gel was extracted by centrifugation of the gel slice over a bed of acid-stripped Ottawa sand in an 0.8-mL microfuge tube with an apical 30-gauge hole, and into a nested 1.5-mL tube. DNA was precipitated in 75% ethanol, centrifuged, and reconstituted in deionized (d) H₂O and submitted to the MUSC Sequencing Facility for sequencing. 

Western Blot

Bovine, human, and rabbit lens capsules with adherent LECs were processed for protein in RIPA extraction buffer (150 mM NaCl, 50 mM Tris [pH 8.0], 2% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS) containing protease inhibitor cocktail for mammalian cells (Sigma-Aldrich Co.) with sonication (50 W, three times 10 seconds). Cultured rLECs and hLECs were washed in cold PBS, scraped off the culture flask surface, and centrifuged at 800g for 5 minutes before addition of RIPA. After extraction for 30 minutes on ice, samples were centrifuged at 12,000g for 10 minutes and the supernatant saved. Protein concentrations were determined using the bicinchoninic acid (BCA) assay (Pierce, Rockford, IL). Samples were electrophoresed on 10% to 20% Tris-glycine-polyacrylamide gels (BioWhittacker, Rockville MD) using a commercial apparatus (Mini-Protean 3; BioRad, Hercules, CA) and then blotted to nitrocellulose membrane using a transblot cell (Mini Trans-Blot; BioRad). Membranes were blocked in 5% nonfat dried milk and probed with primary and secondary antibodies by standard protocols and visualized by enhanced chemiluminescence (SuperSignal; Pierce) using a gel documentation apparatus (VersaDoc; BioRad) and digital camera. Antisera used for detection of GR were BuGR2 (Alexis Biochemicals, San Diego, CA), H-300 and P-20 (Santa Cruz Biotechnology, Santa Cruz, CA), PA1-511A (Affinity Bioreagents, Golden, CO), and NCL-GCR (Novocastra Laboratories, Newcastle-upon-Tyne, UK). 

Western blot analysis for Dex-induced protein expression were prepared from duplicate cell pellets derived from the same samples as used for quantitative (Q)-PCR (described later) and were also processed with RIPA extraction buffer. Media in all flasks were replaced with serum-free DMEM. Forty-eight hours before cells were harvested, 1 μM Dex was added to the 48-hour Dex flask, and DMSO was added to the other flasks. For the 24-, 6-, and 3-hour samples, the media were replaced with media containing Dex at 24, 42, and 45 hours, respectively. Cells in the control flask were exposed to DMSO for 48 hours. Cells from all flasks were harvested at the same time. Blots were probed using antibodies for cellular inhibitor of apoptosis protein (cIAP-2; Trevigen, Gaithersburg, MD) and mitochondrial superoxide dismutase (MnSOD; Stressgen, Victoria, British Columbia, Canada) at 1:100 dilution, followed by an anti-rabbit IgG secondary antisera (Sigma-Aldrich Co.). Blots were reprobed with anti-actin mAb with anti-mouse IgM secondary (Oncogene, La Jolla, CA), imaged as above and densitometric analyses were performed using NIH Image (ver.1.63) with values normalized to the actin signal for each sample and expressed as multiple of increase (or decrease) relative to control samples (available at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). 

Immunohistochemistry

Lenses were placed in Bouin’s fixative for 5 days, bisected anterior to posterior, embedded in paraffin, and sectioned at 5 μm. Sections were deparaffinized and brought through ethanol to H₂O. Sections were then immersed in 10 mM citrate buffer (pH 6.0) and heated at pressure in a pressure cooker for 1 minute to unmask antigens (Vector Laboratories, Burlingame, CA). Sections were blocked in 10% normal goat or rabbit serum, washed three times in PBS and exposed overnight to the primary antisera: H-300 and P-20, NCL-GCR, and PA1-511A. Tetrarhodamine isothiocyanate (TRITC)-conjugated secondary rabbit anti-mouse antisera or goat anti-rabbit antisera (Sigma-Aldrich Co.) were used to localize GR and the sections visualized by fluorescence microscopy (Axioplan-2 with Axioplan software; Carl Zeiss Meditec, Thornwood, NY). 

Gene Expression

The plasmids, pTAT3-luc, kindly supplied by Keith Yamamoto, (University of California, San Francisco), and pRL-TK (Promega, Madison, WI) were purified from JM-101 cultures by standard miniprep techniques (Qiagen, Valencia, CA). Subconfluent rLECs and hLECs were cotransfected with pTAT3-luc and pRL-TK (Renilla luciferase) according to the protocols provided with the transfection agents (GeneJammer; Stratagene, La Jolla, CA for rLECs; SuperFect, Qiagen, for hLECs), then incubated overnight in serum-free medium. After treatment with Dex or antagonist, triplicate cell samples were harvested and luciferase expression analyzed using the dual-luciferase assay system, according to the manufacturer’s instructions (Promega), with a dual-injector plate luminometer (Berthold, Oak Ridge, TN). The pTAT3-luc luminescence was normalized to the Renilla signal for each sample and expressed relative to the transfected-untreated control as multiples of increase (or decrease) in luminescence. 

Quantitative Real-Time PCR

mRNA reverse transcribed to fs cDNA, as described, was prepared for use in Q-PCR analyses. Oligonucleotide primers to human glucocorticoid-induced lucine zipper protein (GILZ), human serum/glucocorticoid-induced kinase (SGK), human and rabbit cellular inhibitor of apoptosis protein (cIAP)-2 and human and rabbit mitochondrial superoxide dismutase (MnSOD) cDNAs were designed on computer (Primer Express software from Applied Biosystems, Foster City, CA). Primers were synthesized by Integrated DNA Technologies (Coralville, IA) (GILZ, SGK, cIAP-2, MnSOD) or GenBase Inc. (San Diego, CA) (actin). 

Q-PCR amplifications were performed in a 96-well plate (Applied Biosystems) in 25-μL volumes according to the manufacturer’s protocol using nucleic acid stain (SYBR Green; Applied Biosystems) as the indicator. Conditions were: dissociation at 95°C for 10 minutes followed by 45 cycles of 1 minute at 60°C and 30 seconds at 95°C (model 7000 Q-PCR Cycler; Applied Biosystems, with data analysis performed using the manufacturer’s software). Samples were run in triplicate alongside controls for actin, and the mean number of cycles to threshold (0.2) was compared with actin to determine multiples of increase or decrease in expression relative to untreated controls. 

Results

Detection of GR by RT-PCR

PCR amplifications using primer combinations designed to amplify regions of the 5′ (GR001f with GR629r and GR460f with GR995r) and 3′ (GR1349f with GR2022r and GR2038f with GR2352r) portions of GR cDNA yielded products of approximately 630, 535, 670, and 310 bp, respectively, from cDNA prepared from human and rabbit (data not shown) lens samples and also from hLECs and rLECs (Fig. 1) . These products were of the anticipated size for cDNAs within exon 1 (5′ products) and spanning exons 3 to 7 (3′ products), respectively. Sequences obtained for the hLECs derived products were identical with the corresponding regions of human GRα (accession no. U01351). Sequences for the rLECs-derived products subjected to BLAST analysis were similar to these same regions (www.ncbi.nlm.nih.gov/blast/ provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD). 

The intervening rLECs GR sequences were obtained by PCR using primers designed to regions of sequence already obtained, and the full length sequence for rabbit GR from LECs was submitted to GenBank (accession no. AY161275). The rabbit GR nucleotide sequence is 89% identical with human GR and mouse GR. The derived protein sequence for rabbit GR is shown aligned to human GR in Figure 2 and shares 91% identity. 

Primers GR1349f with GR2022r and GR2038f with GR2352r were also used in PCR amplifications of bovine fs cDNA prepared from bovine lens capsule and epithelial cells and generated two products of the expected sizes (Fig. 1) . The intervening sequence was obtained using specific primers designed to regions of these products, and a reaction with rabGR548f and rabGR1473r was conducted to obtain additional 5′ sequence. A combined 1757-nucleotide (nt) bovine GR 3′ sequence was obtained (accession no. AY238475). This bovine GR 3′ sequence is 89% and 93% identical with the comparable regions of human GRα and the rabbit GR respectively and for the derived protein is 87% identical with both human and rabbit GRαs. 

Detection of GR by Western Blot

The anti-GR antibodies, NCL-GCR, PA1-511A, H-300, and P-20, gave positive reactions to GR in Western blots of bovine, human, and rabbit lenses (samples derived from lens capsules with adherent epithelial cells) and hLECs and rLECs, at the expected molecular size for GR of approximately 94 kDa (Fig. 3) . BuGR2, which was raised to amino acids (aa) 407 to 423 of rat GR (identical with rabbit GR aa383 to 399: Fig. 2 ), as expected, reacted with the rabbit samples but did not react to those of human origin—the corresponding peptide region of human GR differs by 2 amino acids. BuGR2 also reacted to the bovine lens sample—the corresponding peptide region of bovine GR is also identical with that of rat GR. The most consistent response was observed with NCL-GCR, which was raised to the N-terminal of human GRα, and this mAb reacted with samples from all three mammalian species and both cell lines. PA1-511A reacted strongly to hLECs and weakly to bovine, human, and rabbit lens, the epitope recognized by PA1-511A is identical in human and bovine GR but rabbit GR contains two conservative substitutions in this region. H-300 (maps to aa 121–420 of human GR) and P-20 (designed to the C-terminal of GRα) both reacted with samples from all three species. 

Detection of GR by Immunohistochemistry

Sections of human lenses were evaluated for the immunohistochemical localization of GR using four anti-GR antibodies, the mAb NCL-GCR and the rabbit polyclonals H-300, P20, and PA1-511A. The results are shown in Figure 4 . PA1-511A reacted predominantly with the nuclei of epithelial cells in the anterolateral regions of the lens, the bow region and the differentiating lens fiber cells. NCL-GCR stained the epithelial cells poorly (data not shown). 

Although reactivity of cell nuclei was only marginally increased over controls for H-300 and P-20, these antisera also indicated reactivity in the cytoplasm of the LECs that was absent from the controls. This cytoplasmic reactivity was seen in LECs from the anterior region to the bow region and appeared as a granular fluorescence. 

Detection of GR Activity with pTAT3-luc

rLECs and hLECs preincubated in serum-free media overnight were exposed to 1 μM Dex for 0, 2, or 4 hours or to 100 μM RU486 for 2 hours or to RU486 for 2 hours followed by Dex for 2 or 4 hours. The data are shown in Figure 5 . Addition of Dex induced luciferase expression, which increased between 2 and 4 hours, and RU486 completely (hLECs) or partially (rLECs) blocked this expression. 

Expression of GILZ, SGK, cIAP-2, and MnSOD mRNA in Response to Dex Treatment

Real-time Q-PCR analysis of Dex-treated hLECs and rLECs are shown in Figure 6 . These data indicate that GILZ message increased by approximately 11-fold with 3 hours of Dex exposure and that mRNA levels remained elevated through the 48 hours of treatment. SGK mRNA levels were elevated by 3.8-fold at 3 hours of Dex treatment and declined to 2.4-fold by 48 hours. cIAP-2 message was increased by approximately twofold in hLECs at 48 hours and threefold in rLEC. Levels of MnSOD mRNA were only marginally increased in hLECs but were approximately twofold increased by 3 hours in rLECs and declined subsequently. 

cIAP-2 and MnSOD Protein in Response to Dex Treatment

Western blot analyses are shown in Figure 7 for hLECs and rLECs together with the densitometric analyses of these blots. The expression of cIAP-2 protein was elevated approximately 2.1-fold in hLECs and over 6-fold in rLEC after 48 hours of Dex exposure. MnSOD protein expression was elevated approximately 4.3-fold in hLECs and approximately 11-fold in rLECs. 

Discussion

Through PCR, sequencing, Western blot analysis, and immunohistochemistry, our data confirm the existence ¹⁵ of GR in human lens cells, indicate the presence of GR in rabbit and bovine lens cells—resolving the controversy ⁴ over the existence of GR in bovine lens—and indicate that lens cells of all three species contain the active GRα isoform. 

Products of the expected sizes were obtained by PCR amplification of cDNA obtained from human, rabbit, and bovine lens and from hLECs and rLECs, with oligonucleotide primers designed to known GR sequences. Sequences from the amplified products of fs cDNA from these cells and also from bovine lens cells, obtained with the 3′ primers, matched with a region of cDNA spanning exons 3 to 7 from human GR, indicating the products represented mRNA and not genomic DNA. The complete sequence for rabbit GR was obtained from rLECs. This sequence was compatible with the predominant GRα isoform and is the transcriptionally active form. ¹⁷ The partial bovine GR sequence was also compatible with the GRα isoform. 

Immunohistochemical staining with anti-GR antibodies localized the GR both to the nuclei and cytosol of LECs in the anterior and lateral capsule regions, bow region, and nuclei of cells in the early differentiating fiber cells. The four antisera used reacted at different intensities in the immunolocalization studies, and their reactivities in Western blot analysis were also different. In the case of BuGR2, this mAb was raised to the 407- to 423-aa region of rat GR and failed to react to samples of human origin, but did react in Western blot analysis with samples of cow and rabbit origin. The absence of antibody reactivity in Western blot analysis of bovine lens material in the earlier study of Jobling and Augusteyn ⁴ may have resulted from their method of sample preparation or from their choice of antibody. PA1-511A (also referred to as clone 57 by the suppliers, and used by Jobling and Augusteyn ⁴ ) reacted only weakly with the bovine lens sample in Western blot analysis (Fig. 3) , even though the epitope recognized by this antibody is conserved in bovine GR. 

In the immunohistochemical study of human lens sections, PA1-511A reacted predominantly with the epithelial cell nuclei, whereas NCL-GCR, which responded most strongly by Western blot to samples of human origin, gave a poor response. These differences probably reflect the differential reactivities of the epitopes to which the antibodies were raised and the extent to which these epitopes were available on blots or within the cells in the lens sections. 

Various studies have reported that unliganded GR is localized to both to the cytosol and the nucleus, ^{18 19 20 21} although in many reports the predominant location appears to be cytosolic. From either location, liganded GR translocates and binds to nuclear chromatin. The cytosolic versus nuclear immunolocalization observed in the human lens sections in Figure 4 with the different antisera may reflect epitope availability in GR to specific antibodies; conformational changes related to the state of GR activation; binding to other proteins, particularly Hsp90; or its degree of phosphorylation. ¹⁷  

After ligand binding and translocation, GR dimerizes at target GREs when acting as a transcription factor in direct type-1 interactions. GR can also act through indirect type-2 interactions with other transcription factors, including NFκB and AP-1. The GR in LECs was shown to be capable of inducing direct gene transcription by using the expression vector pTAT3-luc. Both hLECs and rLECs, transfected with pTAT3-luc, which consists of the tyrosine aminotransferase gene containing three upstream GREs, coupled to the firefly luciferase gene, indicated that native GR in these lens cells is able to function as a transcription activator. These data substantiate similar results obtained recently in human and mouse LECs. ¹⁵ In addition, competent direct glucocorticoid-induced type-1 gene transcription by GR was also suggested by the increase in transcription of GILZ and SGK mRNAs from hLECs after exposure to Dex. 

The GILZ gene contains upstream GR recognition motifs and is directly induced by glucocorticoids. GILZ also indirectly regulates the expression of other proteins through interaction with the transcription factors NFκB and AP-1. ^{22 23} SGK is a member of the immediate early-response genes ²⁴ and is also directly regulated by GR ²⁵ through a GRE. ²⁶ Among the upstream transcription factor recognition sites for both cIAP-2 and MnSOD are sites for GR, AP-1, and NFκB. Upregulation of both these proteins has been reported after glucocorticoid administration, ^{27 28 29 30 31} and there are indications, for both cIAP-2 and MnSOD that glucocorticoid induction may be mediated through NFκB. ^{28 31 32} If LEC GR affects the activity of NFκB (and AP-1), then the levels of many lens cell proteins could be affected by steroid treatment. 

The purpose of this study was to help resolve the controversy concerning the presence or absence of GR in the lens. To this end, we believe the data support the presence of GR in human lens and cultured hLECs, indicate GR is present in rabbit lens and LECs, and resolve the presence of GR in the bovine lens. Furthermore, our data suggest that the GR present in LECs is transcriptionally active. The targets we selected for analysis of mRNA and protein expression levels clearly represent a very small sample of those proteins that are likely to have expression modified by glucocorticoid treatment. Reports link the four proteins included in this study to a range of different biochemical pathways, and thus glucocorticoid treatment of LECs is likely to perturb the dynamics of a number of cellular activities. 

Conclusions

Our data indicate that the active GRα isoform is present in LECs and that the expression levels of proteins known to be regulated by glucocorticoids are modified in these cells by glucocorticoid treatment, indicating that this lens GRα is functional. 

 Supported by National Eye Institute Grant EY13786 and by Research to Prevent Blindness, Inc.

 Submitted for publication April 22, 2003; revised August 11, 2003; accepted August 12, 2003.

 Disclosure: E.R. James, None; L. Robertson, None; E. Ehlert, None; P. Fitzgerald, None; N. Droin, None; D.R. Green, 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.

 

Oligonucleotide Primers Used to Obtain Initial PCR Products and Sequences of Bovine, Human, and Rabbit GR and in Q-PCR Amplifications of Target fs cDNAs

Name	Sequence (5′–3′)
GR001f	GGATGGACTCCAAAGAATC
GR460f	AAACCTCARTAGGTCGA
rabGR578f	AGCACCTTTGACATTTTGC
GR629r	GCTCCTGTCTTTAACTTGGG
GR995r	ACGCCATGAACAGAAATG
GR1349f	AGAGCAGTGGAAGGACA
rabGR1473r	TGCTTCCAAGTTCATTCCAG
GR2022r	AGAGAGAAGCAGTAAGGTT
GR2038f	GTTTGAAGAGCCAAGAGC
GR2352r	TCATTTCTGATGAAACAGAAG
Actin forward	CCCCCAGCACCATGAAGAT
Actin reverse	GCCGATCCACACGGAGTACT
hcIAP-2 forward	TGCCTGTGGTGGGAAGCT
hcIAP-2 reverse	GGAAAATGCCTCCGGTGTT
rcIAP-2 forward	AAGGTCAAATGCTTCTGCTGTG
rcIAP-2 reverse	AGAAAAACTACAGCTGGGATATAACTGTT
GILZ forward	GCGTGAGAACACCCTGTTGA
GILZ reverse	TCAGACAGGACTGGAACTTCTCC
hMnSOD forward	TTCTGGACAAACCTCAGCCC
hMnSOD reverse	CAGTTTGATGGCTTCCAGCA
rMnSOD forward	TGCGTGTGCGAATCAGGA
rMnSOD reverse	TCAATCCCCAGCAGTGGAAT
SGK forward	TCCCCAACTCCATTGGCAA
SGK reverse	AGCTTCCTTGACGCTGGCT

Figure 1.

 

Ethidium bromide–stained 2% agarose gel of products from PCR reactions with GR primers. Lane 1: molecular size standards; lanes 2 to 8 contained products from the following reactions: primers GR1349f and GR2022r with fs cDNA prepared from bovine lens capsule/epithelial cells as template (lane 2); GR2038f and GR2352r with bovine lens fs cDNA (lane 3); GR1349f and GR2022r with fs cDNA prepared from hLECs (lane 4); GR460f and GR995r with hLECs fs cDNA (lane 5); GR001f and GR629r with fs cDNA prepared from rLECs (lane 6); GR1349f and GR2022r with rLECs fs cDNA (lane 7); and GR2038f and GR2352r with rLECs fs cDNA (lane 8).

Figure 2.

 

Alignment of derived amino acid sequence of rabbit GRα (accession no. AY161275) with human GRα (accession no. U01351). The glucocorticoid binding domain (aa 564–748) and DNA binding domain (aa 414–486) are in boxes. The peptide regions recognized by antibodies PA1-511A (human aa 346–347) and BuGR2 (rabbit aa 383–399) are underscored.

Figure 3.

 

Western blot of protein samples (100 μg/lane) prepared from the lens capsule and LEC layer of bovine (lane 1), human (lane 2), and rabbit (lane 3) lenses and from hLECs (B-3; lane 4) and rLECs (N/N1003A; lane 5). The blots were probed with the anti-GR monoclonal antibodies, NCL-GCR and BuGR-2, and the rabbit polyclonal antisera, PA1-511A, H-300, and P-20. The approximate size of GR is 94 kDa.

Figure 4.

 

Immunolocalization of GR in human lens. (A) Section from the anterior capsule and (B) bow regions probed with the polyclonal Ab, PAI-511A. (C) Section from the anterior region probed with PA1-511A and the inhibitory peptide PEP-001. (D) Bow region control from which the primary antibody was omitted. (E) Section from the anterior and (F) bow regions probed with the rabbit antiserum H-300. (G) Section from the anterior and (H) the bow regions probed with P-20.

Figure 5.

 

Luciferase expression from hLECs and rLECs transfected with pTAT3-luc exposed to 1 μM Dex for 0, 2, or 4 hours; to 100 μM RU486 for 2 hours; or to RU486 for 2 hours followed by RU486 + Dex for 2 or 4 hours; or to RU486 for 6 hours and Dex for 4 hours.

Figure 6.

 

Change in expression by hLECs of GILZ, SGK, cIAP-2, and MnSOD mRNA and by rLECs of cIAP-2 and MnSOD mRNA, analyzed by Q-PCR normalized to actin and expressed as multiples of increase or decrease relative to control (0) after Dex treatment for 0 to 48 hours.

Figure 7.

 

Change in expression by hLECs of cIAP-2 and MnSOD proteins and by rLECs of cIAP-2 and MnSOD proteins (Western blot analysis with densitometric scans of blots normalized to actin and expressed as multiples of increase or decrease relative to control [0]) after Dex treatment for 0 to 48 hours.

The authors thank Keith Yamamoto (University of California San Francisco) for providing pTAT3-luc, John R. Reddan for providing the N/N1003A cell line, and Gian Re (Medical University of South Carolina) for providing access to the Q-PCR sequencer.