Glucocorticoid receptor binding in bovine lens

Transcription

1 Glucocorticoid receptor binding in bovine lens Eugene J. Wenk, M. Rosario Hernandez,* Bernard 1. Weinstein, Gary G. Gordon, Michael W. Dunn, and A. Louis Southren The present study demonstrates glucocorticoid receptor localization in the nuclei of cells of the anterior epithelium and bow region and its absence in all other regions of the lens. Both autoradiographic and biochemical studies show that triamcinolone acetonide is inactive whereas tetrahydrocortisol (previously considered to be an inactive metabolite ofcortisol) is an active competitor of dexamethasone binding to the glucocorticoid receptor in the bovine lens. The consistency of these findings by both techniques supports the validity of this observation. In addition, it indicates that autoradiography can be used for studying glucocorticoid agonist/ antagonist relationships in the human lens, where only small amounts of tissue are available. (INVEST OPHTHALMOL VIS SCI 22: , 1982.) Key words: glucocorticoids, target cells, cataract, autoradiography, glucocorticoid competitors lucocorticoids appear to be required for the growth and differentiation of many types of cells (for review see ref. 1). The natural differentiation of the lens epithelial cell entails a cessation of cell division followed by synthesis of specific gene products (e.g., the crystallins). 2 Since these cells contain abundant glucocorticoid receptors, 3 glucocorticoids may play a physiological role in the regulation of this process. The finding of active glucocorticoids in the aqueous humor of several mammalian species, 4 ' 5 despite a sig- From the Departments of Ophthalmology, Medicine, and Anatomy, New York Medical College, Valhalla, and the New York Eye and Ear Infirmary, New York, N. Y. This investigation was supported by U.S.P.H.S. Grant EY Submitted for publication April 23, Reprint requests: A. Louis Southren, M.D., New York Medical College, MRI Building, Valhalla, N. Y *Visiting scientist from the University of Chile Medical School. nificant blood-aqueous barrier, is consistent with this hypothesis. Glucocorticoids have been associated with the development of posterior subcapsular cataract (PSCC) in man after both topical and systemic administration 6 ' 7 of pharmacological amounts of these steroids. A morphological study of steroid-associated PSCC 8 demonstrated epithelial cells in the posterior subcapsular region, suggesting that there had been an alteration in the normal differentiation of these cells into mature lens fiber cells. We have previously demonstrated the presence of a glucocorticoid receptor in cellfree homogenates of bovine lens epithelium. We now report the cellular and subcellular localization of 3 H-dexamethasone to this receptor as well as an unusual pattern of competition by a variety of steroids for these sites. Materials and methods Autoradiographic detection of 3 H-dexamethasone binding in whole lenses. Bovine (4-month /82/ $ Assoc. for Res. in Vis. and Ophthal., Inc. 599

2 Invest. Ophthalmol. Vis. Set. May Wenk et al 1B Fig. 1. Autoradiographs of the bow region of lenses incubated with 10~7M 3H-dexamethasone and exposed to the emulsion for 6 months. (H&E; X2000.) A, Lens incubated with 3 H-dexamethasone only. Note the concentration of grains over nuclei. B, Lens incubated with 3H-dexamethasone and a 200-fold excess of nonlabeled dexamethasone. Note the marked decrease in the number of grains over nuclei. old calves) eyes were obtained immediately after slaughter, placed on ice, and transported to the laboratory where dissection was carried out within 4 hr. After the cornea and iris had been removed and all zonule fibers had been severed, lenses were carefully removed and transferred to glass test tubes containing various concentrations of steroids* in 5 ml aliquots of HEPES-buffered RPMI *Generic names of steroids: dexamethasone = 1,4pregnadien-9-fluoro-16-methyMl 3,17a,21-triol-3-20dione; tetraliydrocortisol = 5/3-pregnan-3a<,ll/J, 17a,21tetrol-20-one; triamcinolone acetonide = 1,4-pregnadien-9-fliioro-ll/3,16a, 17a,21-tetrol-3,20-dione 16,17acetonide tissue culture medium. All incubations contained 10"8M to 10~7M 3H-dexamethasone (sp. act. 30 Ci/mmol; Amersham Searle Corp.). Where indicated, a 200-fold molar excess of various nonlabeled steroids (triamcinolone acetonide from Sigma, the others from Steraloids, Inc.) was included. The purity of 3H-dexamethasone was found to be greater than 95% by thin-layer chromatography. The identification and purity of the nonlabeled tetrahydrocortisol was verified by a mixed melting point determination. The lenses were incubated at 31 for 30 min, cooled, and washed nine times with 10 ml aliquots of heatinactivated rabbit serum at 0 over a 90 min period. The lenses were then frozen in Freon precooled with liquid nitrogen and stored in liquid nitrogen. The samples were processed by dry autoradiography using a method suitable for diffusible substances.9" 10 After 3 to 6 months of exposure, the autoradiographs were developed for 45 sec in D-19 (Kodak) at 20 and stained with hematoxylin and eosin (H&E). In one experiment lenses were incubated with 3H-dexamethasone and washed as described above. The epithelium was removed and then extracted with solvent, and the radioactive compounds were analyzed by thin-layer chromatography. In this manner the bound labeled steroid was found to be more than 90% dexamethasone. Silver grains over nuclei and equivalent areas of cytoplasm were counted independently by two of the authors from coded photomicrographs magnified 2000 times. At least 100 nuclei and equivalent cytoplasmic areas were counted for each determination. Biochemical detection of 3 H-dexamethasone binding in lens cell-free homogenates. The bovine lenses were divided into anterior capsule with adhering epithelial cells, posterior capsule, cortex, and nuclear regions as previously described 3 and frozen at 80. Extracts were prepared from dissected lens tissues by homogenizing in Tricine buffer (2 x 10~2M Tricine, 2 x 10"3M CaCl2, 10":iM MgCl2, and 5 X 10"4M dithiothreitol, ph 7.9) with a Brinkmann Polytron PT 10-ST (2 X 15 sec, low speed) and centrifuged for 50 min at 140,000 X g to obtain particle-free supernatants. These supernatants were incubated with 50 nm 3 H-dexamethasone, a concentration previously shown to be sufficient to saturate binding to the glucocorticoid receptor in bovine lens epithelium. 3 Incubations were carried out with and without a 200-fold molar excess of nonlabeled steroids at 0, and at the indicated times aliquots were filtered through a small Sephadex G-50 col-

3 Volume 22 Number 5 Glucocorticoid receptor binding in bovine lens 601 NUCLEUS CYTOPLASM NUCLEUS CYTOPLASM NUCLEUS CYTOPLASM Fig. 2. Distribution of silver grains in the bow region after incubation at three concentrations of 3 H-dexamethasone, with and without nonlabeled steroid. Solid bars, 3 H-dexamethasone alone; hatched bars, 3 H-dexamethasone with a 200-fold excess of nonlabeled steroid; 2 /x.ci/ml corresponds to 10" 7 M dexamethasone in this experiment. Standard deviations are indicated. *p < 0.025; **p < limn to separate the bound and free hormone. Specifically bound steroid was calculated by subtracting the nonspecifically bound from the total as previously described." Results Autoradiographs of the bow region of lenses incubated with 3 H-dexamethasone demonstrated silver grains distributed throughout the cells, but predominantly over nuclei (Fig. 1, A). In parallel incubations using an excess of nonlabeled dexamethasone, nuclear localization of the grains was markedly reduced (Fig. 1, B). Fig. 2 shows the distribution of silver grains over nuclei and cytoplasm at various concentrations of 3 H-dexamethasone with and without an excess of nonlabeled steroid. At each concentration of dexamethasone, the number of grains over nuclei was significantly reduced by the presence of nonlabeled hormone. Thus the binding to the nuclei is to saturable sites, characteristic of steroid hormone receptors. By contrast, the localization of the grains over the cytoplasm was not significantly decreased by an excess of nonlabeled hormone. When incubated with 3 H-dexamethasone, all regions of the anterior epithelium contained large number of grains in both the cytoplasm and the nucleus (Fig. 3, A). As in the bow region, the addition of an excess of nonlabeled hormone to the incubation medium resulted in a large decrease in the number of silver grains over the nuclei but not over the cytoplasm (Fig. 3, B). In addition, there were relatively few grains found over the anterior (Fig. 3, A and B) and posterior lens capsule. Table I shows the amount of dexa-

4 Invest. Ophthalmol. Vis. Sci. May Wenk et al. Table I. Specific binding of H-dexamethasone and its heat lability in extracts from various regions of the bovine lens 3 Tissue Lens epithelium Lens cortex Lens nucleus Lens posterior capsule Femtomoles dexamethasone specifically bound per ing protein* Percent of specific binding that is heat labile t *Extracts were incubated with 3 H-dexamethasone at 0 overnight. t Extracts were heated to 50 for 30 min and then cooled to 0 Viefore addition of dexamethasone. Ca Fig. 3. Autoradiographs of the anterior epithelium of lenses after 3 months of exposure. (H&E; X1750.) Capsule (Ca) and epithelial cell nuclei (N). A, Lens incubated with 3H-dexamethasone only. Silver grains are concentrated in a band overlying the entire epithelial cell layer. Note the paucity of grains over capsule. B, Lens incubated with 3H-dexamethasone and a 200-fold excess of nonlabeled dexamethasone. The number of grains over the region of epithelial cell nuclei is greatly diminished. methasone specifically bound in cell-free homogenates prepared from the various lenticular regions and the percentage of the specific binding that is heat labile (a characteristic of steroid receptors). The amount of specific binding of 3 H-dexamethasone in homogenates of lens epithelium and its heat lability are similar to those found in other glucocorticoid target tissues such as the liver. I2 ' 13 This is in contrast to homogenates from lens cortex, nucleus, and posterior capsule where the specific binding is barely detectable and is heat stable. Fig. 4 shows the autoradiographic localization of 3 H-dexamethasone to cell nuclei in the presence and absence of a 200-fold molar excess of various nonlabeled steroids. As can be seen, nonlabeled dexamethasone and tetrahydrocortisol both suppress nuclear localization of 3H-dexamethasone (Fig. 4, B and D). Triamcinolone acetonide (Fig. 4, C) is inactive as a competitor. Statistical analysis of these binding data (Fig. 5), in which between 100 and 300 nuclei were counted from autoradiographs, demonstrates the significance (p < 0.005) of this competition with dexamethasone and tetrahydrocortisol. The decrease seen with triamcinolone acetonide was not statistically significant. The pattern of competition obtained with varying concentrations of these same nonlabeled steroids in cell-free homogenates of lens epithelium is shown in Fig. 6. Dexamethasone significantly inhibited the binding at 20- and 200-fold molar excess, indicating that the conditions of the assay were sufficient to saturate the glucocorticoid receptor. Tetrahydrocortisol competed partially at the lower concentrations and almost as much as dexamethasone itself at the higher concentration. Triam-

5 rvolume 22 Number 5 Glucocorticoid receptor binding in bovine lens 603 V:- -Mr' ; 48,1 4A- 4D Fig. 4, Autoradiographs of the bow region of lenses incubated with 3H-dexamethasone with or without an excess of a nonlabeled competitor. Time of exposure, 3 months. (H&E; X1750.) A, No competitor added. Note concentration of silver grains over nuclei. B, Nonlabeled dexamethasone added. Grains over nuclei clearly reduced in number compared to A. C, Nonlabeled triamcinolone acetonide added. Nuclear concentration of grains approximately the same as in A. D, Nonlabeled tetrahydrocortisol added. Note the reduction in number of grains over the nuclei compared to A. cinolone acetonide did not significantly compete at either 20- or 200-fold molar excess. Thus the pattern of competition in cell-free homogenates at 0 is virtually identical to that found autoradiographically in whole lenses at 31 Discussion The present study demonstrates glucocorticoid receptor localization in the nuclei of cells of the anterior epithelium and bow region and its absence in all other regions of the lens. This confirms and extends our previous findings of a high-affinity binding protein with many of the properties of a glucocorticoid receptor in cell-free homogenates of bovine lens epithelium. The saturable nuclear binding of 3H-dexamethasone most likely represents cytoplasmic receptor that has been translocated to the cell nucleus during the in vitro incubation {V% hr at 31 ). Both autoradiographic and biochemical studies reported here indicate that triamcinolone acetonide is inactive whereas tetrahydrocortisol (considered to be an inactive metabo-

6 604 Wenk et al. Invest. Ophthalmol. Vis. Sci. May 1982 V) UJ o UJ a. 1 l0 cc cc Ul m 0 NONE TRIAMCINOLONE TETRAHYDRO- DEXAMETHASONE ACETONIDE CORTISOL NONLABELED STEROID Fig. 5. Concentration of silver grains per nucleus in the bow region after incubation with and without a 200-fold molar excess of various nonlabeled steroids. Standard deviations are indicated. 0 I0" 6 I0" 5 [NONLABELED STEROID],Molar Fig. 6. Competition of nonlabeled steroids for 3 H-dexamethasone binding in a cell-free homogenate of lens epithelium. Aliquots were incubated at 0 with 3 H-dexamethasone (50 nm) with and without a 20- and 200-fold molar excess of the indicated nonlabeled steroids for 1 to 3 hr. The bound dexamethasone was separated from the free by gel filtration as described in the text., Dexamethasone;, tetrahydrocortisol; A, triaminolone acetonide. lite of cortisol) is an active competitor of dexamethasone binding to the glucocorticoid receptor in the bovine lens. The consistency of these findings by both techniques supports the validity of this observation. A previous biochemical study of dexamethasone binding to bovine liver cytosol showed that triamcinolone was inactive as a competitor of dexamethasone binding at low concentrations (10- fold molar excess) but did compete for 65% of the binding when present at a higher concentration (100-fold excess). 14 Our present study of bovine lens epithelium shows that triamcinolone acetonide did not compete even at a 200-fold molar excess of nonlabeled steroid. This is similar to our findings with homogenates of bovine iris-ciliary body. 15 Thus the glucocorticoid receptor in these bovine eye tissues seems to be different from that reported for the bovine liver. 14 The consistency of the findings by both biochemical and autoradiographic techniques indicates that autoradiography can be used for studying glucocorticoid agonist/antagonist relationships in the human lens, where only small amounts of tissue are available.

7 Volume 22 Number 5 Glucocorticoid receptor binding in bovine lens 605 REFERENCES 1. Bottenstein J, Hayashi I, Hutchings S, et al: The growth of cells in serum-free hormone-supplemented media. Methods Enzymol 58:94, Papaconstantinou J: Molecular aspects of lens cell differentiation. Science 156:338, Southren AL, Gordon GG, Yeh HS, Dunn MW, and Weinstein BI: Receptors for glucocorticoids in the lens epithelium of the calf. Science 200:1177, Obenberger J, Starka L, and Hampl R: Quantitative determination of endogenous corticosteroids in the rabbit plasma and aqueous humor. Albrecht von Graefes Arch Klin Exp Ophthalmol 183:203, Unpublished observation. 6. Black RL, Oglesby RB, van Sallmann L, and Bunim JJ: Posterior subcapsular cataracts induced by corticosteroids in patients with rheumatoid arthritis. JAMA 174:166, Becker B: Cataracts and topical corticosteroids. Am J Ophthalmol 58:872, Greiner JV and Chylack LT: Steroid-associated posterior subcapsular cataracts: a morphologic study. INVEST OPHTHALMOL VIS SCI 16(ARVO Suppl.):143, Stumpf WE and Roth LJ; High resolution autoradiography with dry mounted, freeze-dried frozen sections. Comparative study of six methods using two diffusible compounds, 3 H-estradiol and 3 H-mesobilirubinogen. J Histochem Cytochem 14:274, Tchernitchin A, VVenk EJ, Hernandez MR, Weinstein BI, Dunn MW, Gordon GG, and Southren AL: Glucocorticoid localization by radioautography in the rabbit eye following systemic administration of 3 H-dexamethasone. INVEST OPHTHALMOL VIS SCI 19:1231, Weinstein BI, Altman K, Gordon GG, Dunn MW, and Southren AL: Specific glucocorticoid receptor in the iris-ciliary body of the rabbit. INVEST OPHTHALMOL VIS SCI 16:973, Ballard P, Baxter JD, Higgins SJ, Rousseau GG, and Tomkins GM: General presence of glucocorticoid receptors in mammalian tissues. Endocrinology 94:998, Baxter JD and Tomkins GM: Specific cytoplasmic glucocorticoid hormone receptors in hepatoma tissue culture cells. Proc Natl Acad Sci USA 68:932, Funder JM and Barlow JW: Heterogeneity of glucocorticoid receptors. Circ Res 46(Suppl. I):l-83, Wenk EJ, Weinstein BI, Hernandez MR, Gordon GG, Dunn MW, and Southren AL: Species specificity of competitors of dexamethasone binding to the glucocorticoid receptor in ocular tissues. INVEST OPHTHALMOL VIS SCI 20(ARVO Suppl.):132, 1981.