47, 133-144 (1992)
lmmunophotochemical Analysis of Mineralocortin by Polyclonal Antibodies against the Native Receptor from Rat Kidney M. MIRSHAHI,* Departments
of *Ophthalmology and TBiochemistry, des Cordeliers, 15 rue de I’Ecole
G. LAZAR,$ and tHornzone de M&decine,
AND M. K. AGARWAL$
Laboratory, Paris, France
Received October 17, 1991; and in revised form November 18, 1991 We have obtained a polyclonal antiserum by immunizing fawn Burgundy rabbits with the mineralocorticoid receptor (MCR) purified biochemically from rat kidneys. High titers of anti-MCR activity were obtained in radioimmunoassays within 3 weeks and increased with a booster shot. In Western blot analysis, the antibody revealed a major band of 94-98 kDa in renal cytosol from rat and beef kidneys. We also developed a fluorographic procedure where the MCR linked covalently to tritiated R-5020, following ultraviolet irradiation, gave imprints superimposable on the Western blot profile. The fluorographic pattern was specific since it was largely abolished in the presence of cold RU 26752 that is specific to MCR, or mineralocortin. The immune IgG precipitated rat renal MCR-[jH]RU 26752 complexes in a dose-dependent manner and also recognized MCR bound to the natural hormone aldosterone. During gel permeation chromatography on Sephacryl, the elution profile of [3H]RU 26752 shifted to high-molecular-weight regions in the presence of immune IgG. The receptor protein could be immunolocalized primarily to the principal cells of the collecting duct in rat kidney but the intercalated cells and glomeruli were not labeled, contrary to beef kidney where a uniform pattern of immunostaining was evident. These should permit large-scale purification of the MCR for detailed physicochemical studies and for screening of the MCRpositive tissues during various pathophysiological syndromes. o 19% Academic PKSS. IK.
Steroid hormones modulate target-specific gene expression in the mammal by binding to high-affinity, low-capacity receptors in the cytoplasmic and the nuclear compartments (l-3). Mineralocorticoids regulate the hydrosodic balance in the organism and a dysfunction can lead to disorders such as Conn’s syndrome and hypertension (4,5). Analysis of the mineralocorticoid receptor (MCR) by classical radioligand assays has been particularly difficult due to its low concentration in the cell, its instability, and the cross-reactivity of the natural hormone aldosterone with other groups of receptors (6-8). Spironolactone derivatives endowed with MCR specificity were synthesized only recently (9,lO). Stereospecific configurations induced by these radioprobes in the MCR protein are related to the residue in the 7-position of their B-ring, thereby permitting molecular analysis (11) and purification of the MCR from different sources (6,7). 133 08854505/92 $3.00 All
Copyright Q 1992 by Academic Press, Inc. rights of reproduction in any fomt reserved.
Polyclonal and monoclonal antibodies generated against several members of the steroid receptor superfamily have been invaluable adjuncts to understanding the distribution, localization, and domain structure of the corresponding proteins (1218). This sort of analysis has hitherto been wanting for the native MCR since the receptor was purified to a sufficient extent only recently (6,7). Receptors for androgens (16) progestins (19), and glucocorticoids (20) have been analyzed by photoaffinity ligands that introduce a covalent bond between the steroid and selected amino acids in the active site of the steroid binding domain (SBD) under the influence of ultraviolet irradiation. Such scrutiny has not been reported for the MCR due primarily to the inavailability of an appropriate, photosensitive radioprobe. The paucity of information regarding the cellular MCR was felt so acutely that doubts were raised about the very existence of this receptor (review in (21), despite the cloning of the MCR gene from human kidney (22). We present here the immunohistochemical analysis of a number of parameters in rat and bovine kidney with the aid of a polyclonal antiserum generated against the native rat renal MCR (6,7). We have succeeded, too, in the affinity labeling of the steroid binding domain of the MCR by photochemistry. Thus, ever since our original demonstration of the MCR some 15 years back (23), the progress has come full circle. A number of technical and conceptual innovations have permitted the reevaluation of age old concepts such that the MCR may now more appropriately be named Mineralocortin as a full member in the family of trunsactivating proteins with distinct structural and functional characteristics and undisputed cellular identity. MATERIALS Receptor
The MCR, or Mineralocortin, was purified from rat kidney by our original procedure consisting of cytosol saturation with MCR-specific ligand RU 26752, chromatography on phosphocellulose, activation at room temperature, batch extraction from DNA-cellulose, and finally lyophilization, in that order (6,7). Immunization
The lyophilized MCR (150 pg protein) was reconstituted in Freund’s complete adjuvant for 30 min and injected intradermally at four different foci into adult fawn, Burgundy rabbits, maintained on pellet food and water in a temperatureand light-controlled animal house. A booster shot was given 3 weeks later and blood samples from the ear vein were analyzed by ELISA for anti-MCR activity. Enzyme-linked
Dynatech (UK) polystyrene microplates (96 wells) were coated with 100 ~1 purified MCR (3 pg/ml) in phosphate-buffered saline (PBS), containing (per liter) 0.25 g potassium dihydrogen phosphate, 1.38 g disodium phosphate, 0.25 g KCl, 9 g NaCl, and 0.01% sodium azide, overnight at 37°C. After three washes with 0.2 ml PBS-0.1% Tween 20, they were incubated for 2 h at 37°C with 100 ~1
serial dilutions of rabbit MCR antiserum. Following three washes with PBSTween, the antigen-antibody complexes were saturated with 100 ~1 sheep antirabbit biotinylated antibody (1: 1000 in PBS-Tween-0.1% bovine serum albumin). Three washes with PBS-Tween followed, after which 100 ~1 of 1: 1000 streptavidin biotinylated horseradish peroxidase complex was added to the wells and the peroxidase activity was finally quantitated by further addition of 100 ~1 of orthophenylene diamine (0.04%)-hydrogen peroxide (0.01%). The reaction was stopped after 3 min with 50 ~1 of 2.5 N sulfuric acid and the color intensity was measured at 492 nm. A series of controls in parallel included MCR or rabbit serum or MCR + nonimmune serum, alone (24). Immunoblots
Organ cytosols were denatured in 220 mM Tris-HCl, ph 6.8, 2% SDS, 15% glycerol and electrophoresed for 3-4 h on 15% acrylamide gels in Tris-glycineSDS (13 g + 14.4 g + 1 g/liter water). The gels were colored with the Coomassie reagent for 10 min and finally destained with 10% each of acetic acid and methanol for the desired period of time. A series of standards (Bio-Rad) of known molecular weight were elctrophoresed with each run to assess the molecular mass of the MCR (6,7,24). The proteins separated by SDS-PAGE as above were electrotransferred to Millipore (USA) nitrocellulose membranes for 1 h at 5 mA in Tris-glycinemethanol (1.5 g + 7.2 g + 100 ml/liter). The membranes were blocked for 1 h with 15% nonfat dry milk at 37°C in PBS and incubated with rabbit anti-MCR (1: 250 in PBS-albumin) for 2 h at 4°C and the antigen-antibody complexes were saturated with sheep anti-rabbit biotinylated antibody (1: 500) for 90 min, followed by 90 min in the presence of streptavidin biotinylated horseradish peroxidase complex (1:500), all diluted in PBS-albumin. The blots were washed 3 x 7-10 min between each of the successive steps mentioned above. Finally, the nitrocellulose papers were developed in the dark for lo-20 min at room temperature with a mixture of 4-chloro-1-naphthol(O.O5%), hydrogen peroxide (O.Ol%), methanol (lo%), dried, and stored at -20°C as described before (24). Auto&orography
The organ cytosol in buffer A was incubated (2 h 4°C) with 20 nM [3H]R 5020 alone or in the presence of loo-fold excess of RU 26752 that is specific to MCR and then irradiated at a distance of l: 10000) of MCR-specific antibody could be detected by the enzyme-linked immunoabsorbent assay (ELISA) just 3 weeks after immunization with 150 hg of protein purified in the presence of [3H]RU 26752 and an excess of RU 38486 from rat kidney (not shown). Three weeks after a booster with 140 pg MCR, the titers increased even further such that ELISA was still positive at >1:32000 dilution of the antiserum. The antiserum obtained in this manner was analyzed further by several techniques to assess MCR specificity and the nature of the cellular receptor. Western
blot analysis of rat kidney cytosol revealed a major band of 94-98 kDa (Fig. 1, lane C), in good agreement with the theoretical value of 107 kDa estimated from the MCR cloned from human kidney (22) and the antigen purified biochemically (677). Since a specific probe for MCR endowed with photoactivable configuration is currently wanting, we used promegestone or R 5020, which exhibits some affinity for the MCR (8,27) and which has been successfully employed for the affinity labeling of PR by photochemical techniques (19). The data in Fig. 1 show the fluorographic pattern of rat renal MCR irradiated in the presence of 20 nM [3H]R 5020 under various conditions. The labeling was confined to the 94- to 98-kDa band (lane F), confirming the results with immunoblots. The labeling was greatly reduced in the presence of loo-fold excess of cold RU 26752 that is specific to MCR (lane E). [3H]R 5020 alone gave a negative fluorograph (lane D), similar to nonirradiated [3H]R 5020-MCR complexes (not shown). Pretreatment of kidney cytosol with the anti-MCR IgG also reduced the extent of photolabeling but did not eliminate it since the immune IgG is unable to precipitate all of the MCR (see below). Thus, a new tool for the chemical analysis of Mineralocortin is now available. Immunoprecipitation of radiolabeled ligand-receptor complexes by the antibody has been used to assess the specificity of various antisera (12-14,18). Dose-dependent precipitation of rat renal MCR- [3H]RU 26752 complex, where the affinity of the steroid for Mineralocortin excludes association with receptors for other classes of steroid hormones (9,11), was obtained with our polyclonal antiserum (not shown). Immunoprecipitation was limited to no more than 30-35% of the total MCR in renal cytosol, irrespective of the quantity of the specific IgG used, as with other receptor classes (12-14,18). It should be kept in mind that immunoprecipitation was carried out in the presence of high salt and that the antibody may have only limited access to the untransformed heterooligomer (18). The data in Fig. 2 show that incubation of the MCR with immune IgG increased the molecular mass of the receptor. One major peak of about 40 kDa was observed with [3H]RU 26752 on the Sephacryl column in the presence of nonimmune IgG, with two humps in 230- and 660-kDa regions. These suggest degradation and aggregation of the MCR during the elution process, in agreement with our earlier studies (6,7,11). In the presence of immune IgG, the position of the major peak was calculated to be 50 kDa with one minor component in the 305-kDa region and an even larger one whose macromolecular mass could not be estimated accurately. The appearance of new, high-molecular-weight entities in the presence of specific antiserum has also been reported for other receptor classes but the molecular mass in these studies was not provided (12-14,18), making comparisons difficult. In order to maximize the antigen-antibody interaction, and to minimize any nonspecific association, high salt was used throughout these experiments, so the behavior of untransformed, cytosolic MCR remains unknown. Finally, we used this specific antiserum to localize the MCR in paraffin-embedded sections of the kidney. The data in Fig. 3 show that immunostaining was limited to cells lining the distal tubule, particularly the principal cells of the collecting duct, but all of the cells could not be stained (A,B). The glomerular
FRACTION NUMBER 2. Displacement of rat renal Mineralocortin by immune IgG. Renal cytosol labeled with [3H]RU 26752 was passed through the Sephacryl S-300-HR column in the presence of preimmune (0) or immune (0) IgG and the fractions were monitored for absorbance and radioactivity. The protein profile was superimposable and is therefore not shown. Arrows indicate approximate molecular mass in kilodaltons. FIG.
region in rat kidney did not stain at all, contrary to bovine kidney where the immunostaining was uniform (CD). The pattern of staining was almost identical in kidney sections from the normotensive WKY and the hypertensive SHR strains (tail arterial tension 150 and 20.5 mm, respectively). DISCUSSION Having characterized (11) and purified (6,7) the MCR with the aid of synthetic derivatives endowed with receptor specificity, we now proceeded to analyze the Mineralocortin with the aid of a polyclonal antiserum directed against the native receptor and to phtotolabel the cytosolic MCR. From the outset we decided against the idiotype approach because of the specificity of such antibodies for the ligand used for immunization (28), whereas flexibility of the active site is a hallmark of the native protein and forms the very rationale for the development of new receptor probes (3,8). We also decided against the use of synthetic sequences of cloned receptors as immunogens for a number of reasons. Thus, such fragments supposedly engender receptor-specific antibodies (18), despite the fact that the conserved domain structure shares close sequence homology among various members of the steroid receptor superfamily (1,3,22). Monoclonal antibodies have been raised against a number of steroid receptors (13,29) and offered an attractive possibility. But our efforts to immunize mice with rat renal MCR were uniformly negative due possibly to close structural
FIG. 3. Mineralocortin-specific immunostaining in the kidney. The immunoreactive proteins were limited to the principal cells of the collecting duct in rat kidney cortex (dark spots), whereas the glomeruli and the intercalated cells (arrow) were not stained (B); bovine kidney showed staining (dark patches against light background) in both cortical (C) and glomerular regions (D); control serum did not stain any area at all in rat kidney (A) similar to bovine kidney (not shown). All sections x288.
homology of Mineralocortin between these two species. We are currently applying this technique to bovine MCR purified by immunoaffinity. We therefore opted for the time-tested procedure of a polyclonal antiserum in the rabbit generated in response to the activated MCR purified from rat kidney by biochemical procedures developed in our laboratory (6,7). The specificity of this antiserum was first tested by a number of criteria. The antibody titers in the rabbit increased with time over several weeks, and a booster shot led to further rise in the antimineralocortin titer in radioimmunoassays. This antiserum recognized a major band of about 98 kDa, in good agreement with both the cloned (22) and the biochemically purified (6,7) protein. It is therefore clear that the antiserum against the activated receptor was able to recognize the native, untransformed MCR in kidney cytosol. Further proof for the specificity of the antiserum was forthcoming from autofluorography of the MCR linked covalently to [3H]R 5020 by ultraviolet irradiation. Promegestone effectively labeled the 94- to 98-kDa band that diminished greatly when MCR-specific RU 26752 was allowed to compete with [3H]R 5020. Furthermore, no labeling at all could be observed with the free steroid alone or in the absence of irradiation that is required for the formation of a covalent bond between the ligand and selected amino acids of the SBD. The anti-MCR antiserum,
too, interfered with the photolabeling but did not eliminate it because access to native MCR in the cytosol appears limited. Since R 5020 is the “ideal” ligand for the PR, including the photoaffinity labeling of the receptor (19), its association with the MCR is further proof that progestins antagonized mineralocorticoid action (30)) possibly through a hydrophobic pocket (31) shared by MCR, PR, and the glucocorticoid receptor (GCR). Thus, a new tool has been found to dissect the steroid binding domain of the MCR, and particularly to identify the amino acids that maintain the tridimensional configuration of the active site in Mineralocortin. Precipitation of the radioactive ligand-receptor complex by the antibody has been used for the GCR (17) the ER (12), the PR (13), and the AR (14). Although the varying experimental conditions employed by different groups do not permit precise comparisons, a number of tenets are common to them all. Thus PBS has been the buffer of choice in the presence of high salt to limit the precipitation of nonspecific proteins and to permit maximal interaction between the receptor and its antibody. Our data are entirely in keeping with those of earlier studies using other receptor classes. As in other studies, we were unable to immunoprecipitate the native, heterooligomeric form of the MCR from rat tissues. This is somewhat surprising since the antibody does indeed form a complex with the native, albeit denatured, MCR immobilized in polyacrylamide gels for Western blot analysis. Immunoprecipitation also appeared steroid specific to some extent since ZK 91587-MCR could not be precipitated at all, whereas aldosterone-MCR complexes were precipitated to a limited extent. The stereospecificity of the SBD therefore is particularly complex such that all epitopes are not equally accessible to the antibody. This is also evident from the fact that prior exposure of renal cytosol to immune IgG did not alter subsequent association of RU 26752 although the antibody effectively precipitated such complexes. Recent observations also suggest several binding sites on the receptor, contrary to the classical notion that all grades of agonists and antagonists bind the same site with varying degrees of affinity, in keeping with the predicted dynamic flow model of steroid hormone action (3). Collectively, these data confirm the specificity of the polyclonal antiserum for the MCR or Mineralocortin. The interaction of the immune IgG with transformed renal MCR led to the formation of high-molecular-weight complex during gel permation chromatography, in the presence of high salt,.similar to other receptor classes (12-14) where precise molecular weights were not provided, furthermore confirming specificity of our antiserum. The immunohistochemical studies here support the physiological action of mineralocorticoids on the principal cells of the collecting duct in the distal tubule (32,33). Surprisingly, bovine glomeruli showed MCR-specific staining, whereas the hypertensive disease apparently did not alter the MCR staining pattern in the rat. Collectively, we present here the only available evidence for a polyclonal antibody against native Mineralocortin. We have also succeeded in the photoaffinity labeling of the MCR, thereby providing a specific tool for the identification of
this receptor. A combination of these cross-reactivity, should permit closer Mineralocortin by immunopurification the role of the MCR in various states
methods, and particularly the interspecies scrutiny of the functional domains of the of large quantities of this protein and of of health and disease.
ACKNOWLEDGMENTS Some of the steroids used here were kindly provided by Drs. D. Philibert, Romainville, France, and W. Losert, Schering AG, Berlin, Germany. This work was aided in part by financial assistance from UFR Broussais Hotel Dieu.
REFERENCES 1. Beato M. Transcriptional control by nuclear receptors. FASEB J 5:204&2051, 1991. 2. Wahli W., Martinez E. Superfamily of steroid nuclear receptors: Positive and negative regulators of gene expression. FASEB J 5:2243-2249, 1991. 3. Agarwal MK. Evolving trends in steroid hormone receptor research. Die Natur 77: 170-175, 1990. 4. Conn JW. Aldosterone and hypertension. J Am Med Assoc. 183~775-781. 1955. 5. Sharp GWG, Leaf A. Mechanism of action of aldosterone. Physiol Rev 46:593-633, 1966. 6. Lazar G, Pagan0 M, Agarwal MK. Purification and characterization of the activated mineralocorticoid receptor from rat myocardium. Biochim Siophys Actn 1033:41-48, 1990. 7. Lazar G, Pagan0 M, Agarwal MK. Purification and characterization of the activated mineralocorticoid receptor from rat kidney. fnt J. Biochem 22621-630. 1990. 8. Agarwal MK, Lazar G. Antimineralocorticoids. Renal Physiol Biochem 14~217-223, 1991. 9. Perroteau I, Netchitailo P, Delarue C, Leboulanger F, Philibert D, Dereadt R., Vaudry H. The effect of the antimineralocorticoid RU 28318 on aldosterone biosynthesis in vitro. J Steroid Biochem 20:853-856, 1984. 10. Losert W, Bittler D, Buse M, Stenzel JC, Haberey M, Laurent H, Nickish K, Schillinger E, Wiechert R. Mespirenone and other 15,16-methylene-17-spirolactones, a new type of steroidal aldosterone antagonists. Drug Res 36:1583-1600, 1986. 11. Agarwal MK, Kalimi M. Paradoxical differences in the receptor binding of two new antimineralocorticoids. Biochim Biophys Actu 964:105-l 12, 1988. 12. Greene GL, Gloss LE, Fleming H. DeSombre ER. Jensen EV. Antibodies to estrogen receptor: Immunochemical similarity of estrophilin from various mammalian species. Proc Nat1 Acad Sci USA 74:3681-3685, 1977. 13. Sullivan WP, Beito TG, Proper J, Krco CJ, Toft DO. Preparation of monoclonal antibodies to the avian progesterone receptor. Endocrinology 119:1549-1557. 1986. 14. Young CYF. Murthy LR, Prescott JL, Johnson MP, Rowley Dr, Cunningham GR. Killian CS, Scardino PT. VonEschenbach A, Tindall DJ. Monoclonal antibodies against the androgen receptor: recognition of human and other mammalian androgen receptors. Endocrinology 123:60-610, 1988.
15. Sprangers SA, Brenner RM, Bethea CL. Estrogen and progesterone receptor immunocytochemistry in lactotropes versus gonadotropes of monkey pituitary cell cultures. Endocrinology 124:14621470, 1989. 16. Demyan WF, Sarkar FH, Murty CVR, Roy AK. Purification and immunochemical characterization of the cytoplasmic androgen-binding protein of rat liver. Biochemistry 2&1732-1736, 1989. 17. Antakly T, Raquidan D, O’Donnell D, Katnick L. Regulation of glucocorticoid expression. I. Use of a specific radioimmunoassay and antiserum to a synthetic peptide of the N-terminal domain. Endocrinology 126:1821-1828, 1990. 18. Traish A. Ettinger R, Kim N, Rothstein AM, Wotiz HH. Development and characterization of monoclonal antibodies to a specific domain of human estrogen receptor. Steroids 55:196-208, 1990. 19. Clarke CL, Satyaswaroop PG. Photoaffinity labelling of the progesterone receptor from human endometrial carcinoma. Can Res 45~5417-5420, 1985.
20. Hermann T, Schramm K, Ghraf R. Photoaffinity labelling with 3H-RU 28362: A powerful tool for the study of rat brain glucocorticoid receptors. J Steroid Biochem 26~417-423, 1987. 21. Agarwal MK, Raynaud JP. Steroid antagonists. FEBS Lett 245:1-4, 1989. 22. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Houseman DE, Evans RM. Cloning of human mineralocorticoid receptor complementary DNA: Structural and functional kinship with the glucocorticoid receptor. Science 237: 268-275, 1987. 23. Agarwal MK. Chromotographic demonstration of mineralocorticoid specific receptors in rat kidney. Nature 254,623-625, 1975. 24. Mirshahi M, Pagan0 M, Mirshahi A, Agarwal MK. Generation of polyclonal antibodies against the mineralocorticoid receptor and analysis of mineralocortin in rat myocardium by immunophotochemistry. Biochim Biophsy Acta, in press. 25. Keppler D, Fonandeche MC, Fumeron VD, Pagan0 M, Burtin P. Immunohistochemical and biochemical study of a cathepsin B-like proteinase in human colonic cancers. Can Res 48:68556862, 1988. 26. Bradford D. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 72~248-254, 1976. 27. Wambach G. The mineralocorticoid receptor and the activity of aldosterone antagonists. In Receptor Mediated Antisteroid Action (Agarwal MK, Ed.). Berlin: de Gruyter, 1987, pp. 169-196. 28. Agarwal MK, Cayanis E. Evidence for differences in the steroid binding domains of the glucocorticoid receptor versus the idiotype antibody. Biochem Biophys Res Commun l36:470-475, 1986. 29. King WJ, Greene GL. Monoclonal antibodies localize oestrogen receptor in the nuclei of target cells. Nature 307:745-747, 1984. 30. Landau RL. Progesterone versus aldosterone. In Antihormones (Agarwal MK, Ed.). Amsterdam: Elsevier/North-Holland, 1979, pp. 153-166. 31. Philibert D, Costerousse G, Moguilewsky GM, Nedelec L, Nique F, Tournemine C, Teutsch, G. From RU 38486 towards dissociated antiglucocorticoid and antiprogesterone. In Antihormones in Health and Disease (Agarwal MK, Ed.). Basel: Karger, 1991, pp. l-17. 32. Stanton BA, Biemsderfer D, Wade JB, Giebisch G. Structural and functional study of the rat distal nephron: Effects of potassium adaptation and depletion. Kidney Int 19~36-41, 1981. 33. Madsen KM, Tisher CC. Structural-functional relationships along the distal nephron. Am J Physiol 2JO:Fl-F6. 1986.