Chicken Intestinal Receptor for 1,25 ... - Semantic Scholar - M.MOAM.INFO (2025)

Vol. 262, No. 3, Issue of January 25, pp. 1305-1311, 1987 Printed in U.S.A.

THE JOURNAL OF BIOLOGICAL CHEMISTRY 8 1987 by The American Society of Biological Chemists, Inc.

Chicken Intestinal Receptor for 1,25-Dihydroxyvitamin D3 IMMUNOLOGICCHARACTERIZATION AND HOMOGENEOUS ISOLATION OF A 60,000-DALTON PROTEIN* (Received for publication, March 11, 1986)

J. Wesley Pike$, Noreen M. Sleator, and Mark R. Hausslerg From the Department of Biochemistry,Arizona Health Sciences Center, University of Arizona, Tucson, Arizona 85724

The chick 1,25-dihydroxyvitamin D3 receptor has been identified via immunoblot analysis and isolated to homogeneity via positive immunoselection and SOdium dodecyl sulfate-polyacrylamide gel electrophoresis. Cytosolic extracts of intestinal mucosa, as well as purified samples highly enriched for receptor by nonimmunologic methodology were electrophoresed on denaturing gels, transferred to nitrocellulose, and probed utilizing a purified monoclonal antibody against the chick receptor. Two protein signals were detected by this approach, a major species of 60,300 daltons and a minor form at 58,600 daltons. Both immunologically identified species were present in receptor-positive tissues but were absent in receptornegative liver extracts. The two immunoreactive cytosolic proteins comigrated with two polypeptides detected via Coomassie Blue staining as well as by immunoblot analysis afterenrichment utilizing DNA-cellulose, blue dextran-Sepharose, and other chromatographic separation techniques. Increasing concentrations of the minor form during purification suggest it arises from the larger molecular weight species via proteolysis. Finally, both forms of the receptor were isolated to near homogeneity employing positive immunoselection and each individually purified to homogeneity employing sodium dodecyl sulfate-polyacrylamide gel electrophoresis. These experiments show that the chick receptor exists as a major species of 60,300 as well as a minor form of 58,600 and that both forms can be purified to homogeneity via immunoaffinity chromatography.

1,25-Dihydroxyvitamin D, (1,25-(OH)zD3)’is a eukaryotic steroid-like hormone with an essential role in the regulation of calcium and phosphorus homeostasis (1, 2). The obligate mediator of 1,25-(OH)2D3’saction is a receptor protein (3,4), which, by analogy to other steroidreceptors, is postulated to bind to promoterregions of the genome and directly alter the transcriptional rates of genes encoding hormone-dependent proteins. The macromolecule is an acidic protein that sediments at3.2-3.7 s and has apparent binding domains for both * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Recipient of National Institutes of Health Grant DK-38130 and AR38170. Present address: Dept. of Pediatrics and CellBiology, Baylor College of Medicine, One Baylor Plaza, Houston, T X 77030. Recipient of National Institutes of Health Grant AR-15781. The abbreviations used are: 1,25-(OH)zD3,1,254ihydroxyvitamin D3; cytosol, high-speed supernatants of homogenate5 of intestinal mucosa prepared in KETD-0.3; PMSF, phenylmethylsulfonylfluoride; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis.

1,25-(OH)zD3and DNA (5-8). Although molecular weight estimates via permeation chromatography have ranged from 47,000-95,000 for the avian receptor (3-6, 9-12), purification of this trace molecule followed by denaturing gel electrophoresis has generally revealed a major polypeptide of approximately 60,000-65,000 daltons (5, 6, 9, 10). Direct evidence, however, has not been obtained to show that this species is the 1,25-(OH)2D3 receptor. This fact, coupled with the extreme proteolytic sensitivity to which steroid receptors are proneduring isolation, has made it extremely tenuousto assign a specific molecular weight to partially or homogeneously purified avian 1,25-(OH)2D3 receptor.In this paper,we provide strong evidence based upon immunologic detection of receptor after resolution on SDS-polyacrylamide gel electrophoresis, that the chick 1,25-(OH)zD3receptor apparently exists incytosols as two antigenically relatedspecies of 60,300 and 58,600 daltons. The latter, however, most likely represents a proteolyticartifact of the nativemonomer. In addition, these proteins can be highly enriched by positive immunoselection and individuallyisolated to homogeneity via SDSPAGE. EXPERIMENTALPROCEDURES

Radioisotopes-1,25-Dihydro~y[26,27-~H]~itamin D, (158 Ci/ mmol) was generated from 25-hydro~y[26,27-~H]vitamin D3 as previously described (13). Crystalline 1,25-(OH)2D3was a generous gift of Dr. M. Uskokovicof Hoffmann-La Roche (Nutley, NJ). Na’”I was obtained from New England Nuclear. The 14C-protein standards bovine serum albumin (14), phosphorylase 0 ( X ) , ovalbumin (16), and lysozyme (14) were obtained from Amersham Corp. Radioiodinated protein A was obtained from ICN Biochemicals (Cleveland, OH). All isotopes were quantitated with ACS scintillation mixture (Amersham Corp.) in a Beckman LS-7500 instrument. Buffers-The compositions of the buffers used in these experiments are as follows: KETD (0.01 M Tris-HC1, pH 7.4 (at 25 “C), 1 mM EDTA, 5 mM dithiothreitol, and variable molar concentrations of KC1 as indicated, e.g. KETD-0.3); TBS (0.05 M Tris-HC1, pH 7.5 (at 25 “C),0.2 M NaCl, 0.05% Tween 20); PBS (0.02 M Na2HP0,, pH 7.2 (at 25 “C), 0.15 M NaCl). Preparation of 1,25-(0H)& Receptor-All procedures for the preparation of 1,25-(OH)2D3receptors were carried out at 4 “C or on ice. Intestinal mucosa, kidney, and liver were obtained from 4-6-week old Rhode Island Red cockerels maintained on a diet deficient in vitamin DS (17) from the day of hatch. Cytosols were prepared by homogenization of tissue in KETD-0.3, followedby ultracentrifugation a t 165,000 X g for 40 min. 1,25-(OH)& receptors were prepared from extracts of mucosal nuclei or chromatin as described in a subsection below. Preparation of Antibodies-Monoclonal antibodies 9A7y and 4A5y were obtained from serum-free cultures of SP2/0-9A7 and SP2/04A5 hybridomas and purified to homogeneity as previously described (8). They were stored at aconcentration of 1-3mg/ml in PBS containing 0.05% sodium azide. SP2/0-4A5 and SP2/0-9A7 represent hybridomas derived from separate splenic cell fusions of two different immunized rats. The 9A7y and 4A5y antibody products (selected by reactivity to the1,25-(OH)2[3H]D3.receptor complex) exhibit relative affinities for chick 1,25-(OH).& receptor of 1.8 X lo-” M and 1 X 10”’ M, respectively (8). While these antibodies display complete

1305

1306

Immunocharacterization of Chick 1,25-(0H)2D3Receptors

specificity (they do not cross-react with either estrogen or glucocorticoid receptors), they are highly reactive with all mammalian forms of the 1,25-(OHhD3 receptor thus far tested (10). Purified monoclonal antibody 9A7y and 4A5y were used to immunize New Zealand White Rabbits via standard immunization protocols. Rabbit anti-gA7y and anti-4A5y antibodies were then purified to homogeneity by fractionation of the crude antiserum on a rat IgG affinity column. One molar acetic acid-eluted fractions were neutralizedwith 1 M Tris-HCl, dialyzed into PBS, and stored a t -70 "C. The presence of anti-9A7y or anti-4A5y activity was ascertained by Ouchterlony analysis and by utilizing it, in combination with Staphylococcus aureus (Enzyme Center, Boster, MA), to precipitate 1,25-(OH)2D3 receptor-monoclonal antibody complexes (8). Anti-rat IgG was obtained from Zymed Laboratories (San Francisco, CA). Purification of Chick Intestinal 1,25-(OH),D3 Receptors-l,25(OH)*D:$receptors were purifiedfrom 1.22 kg ofchick intestinal mucosa by chromatographic methods describedpreviously(6,10). Briefly,mucosa was homogenized inKETD-0(containing 1 mM phenylmethylsulfonyl fluoride) and thereceptor prepared from chromatin subfractions asfollows: mucosal tissue was homogenized via a P-10 polytron (Brinkmann) in 5 volumes (w/v) of KETD-0, and the nuclei collected a t 4000 X g for 20 min. After an additional wash in the same buffer, the nuclei were sequentially washed in 1) 0.008 M EDTA and 0.025 M NaC1, 2) 0.01 M Tris-HCI, pH7.4, and 1% Triton X-100, and 3) 0.01 M Tris-HCI, pH 7.5. The final wash was repeated a second time. These buffers also contained 5 mM dithiothreitol and 1 mM phenylmethylsulfonylfluoride. 1,25-(OH)2D3receptors were extracted from the crude chromatin pellet for 30 min in 5 volumes of KETD-0.3 containing 1mM phenylmethylsulfonyl fluoride, and after centrifugation (17,000 X g for 20 min)to remove thechromatin material, precipitated with crystalline ammonium sulfate a t 40% of saturation. Protein pelletswere stored frozen a t -70 "C. The ammoniumsulfate-precipitatedpellets were dissolved inKETD-0and labeled overnight a t 4 "Cwith 50 nM 1,25-(OH)2[3H]D3(0.78 Ci/ mmol). Following dialysis to remove residual salt, the samples were chromatographed sequentially onDNA-cellulose (5 X 12 cm), hydroxylapatite (2.5 X 7 cm), Ultrogel AcA-44 (1.6 X 60 cm), DEAE-cellulose (2.5 X 5 cm), and blue dextran-Sepharose (1 X 3 cm). Six separate runs were necessary to fractionate the entire receptor containing sample by DNA-cellulose, whereupon the material was combined for further purification. As assessed by quantitation of radiolabeled hormone binding after DE81 filter adsorption (6-8),dissolved pellets contained 12.3nmol of receptor andthefinal purified material contained 0.55 nmol. This procedure provided an approximate yield of 5% and an estimated purityof >50% based upon both the calculation of specific activity (receptor-hormone complex/mg protein) as well as Coomassie Blue staining of the purified sample after SDSPAGE. 1,25-(OH)2D3 receptors were also isolated from salt extracts of the nuclearfractiononboth a 4A5y-Sepharoseand9A7y-Sepharose immunoaffinity columns. The resins were prepared by coupling purified 4A5y or 9A7y withCNBr-activatedSepharose(Sigma)as described by the manufacturer. Each resin contained approximately 1 mg of antibody/ml of resin. Salt extracts of the crude nuclear fraction were preparedas follows: intestinal mucosal tissue was homogenized via a glass-Teflon tissue grinder in 10volumes (w/v) of KETD-0.05, and the nuclear fraction collected by centrifugation at 3,000 X g for 15 min.The pellet was washed twice in homogenization buffer and collected each time by centrifugation as above. Receptors were extracted from the crude nuclear pellet by resuspension for 30 min. in KETD-0.3, and the soluble fraction subsequently obtainedby ultracentrifugation a t 165,000 X g for 30 min. Samples be to employed for purification by immunoaffinity selection were always prepared immediately before use. In order to assess the efficiency of adsorption of receptor to the antibody-derivatized Sepharose, preparationswere prelabeled for 2 h a t 4 "C with 8 nM 1,25-(OH)2[3H]D3 (85 Ci/mmol). Incubation of nuclear extracts with immunoaffinity beads for 16-18 h at 4 "C in ratios as high as 300-500 to 1 (v/v; 25 ml of extract and 0.05 ml of beads) resulted in the adsorption of over 85% of 1,25-(OH)2[3H]D3. receptor complexes,as estimatedby unbound receptor and subsequent elution of radioactive 1,25-(OH)2[3H]D3. The immunoaffinity resin was washed in KETD-1.0, pelleted through a cushion of 1 M sucrose in KETD-1.0, washed repeatedly in KETD-1.0 containing 0.2% Triton X-100, and then washed with KETD-0. Receptors were eluted using up to 10 resin volumes (v/v) of 3 M guanidine-HCl. Preparative purification of the1,25-(OH)2D3 receptoremployed 4-6 g eq of mucosal nuclear extracts (20-30 ml) containing 4-6 pg of receptor

and 0.05-0.1 ml of derivatizedSepharose. The guanidine/HCl/protein eluate was precipitated with4 volumes (v/v) of cold acetone (-20 "C), and after1 h a t -70°C the precipitate was collected at 17,000 x g for 20 min. The pelleted proteins were washedonce with 80% coldbufferedacetone (-20 "C) and then solubilized in electrophoresis buffer for either analysis or further purification.The 60- and 58-kDa species were isolated to homogeneity through elution ofgel slices containing either the60- or 58-kDa species with TBS. Immunologic Detection of 1,25-(OH)zD3 Receptors-Intestinal mucosa, kidney, or liver cytosols, as well as purified receptor protein were electrophoresed on 11% SDS-polyacrylamide gels as described by Laemmli (18).The gels were either stained with Coomassie Blue for direct visualization or electrophoretically transferred to nitrocellulose membranes (Bio-Rad) as described earlier (19). The elution bufferwasmodified to contain 25 mM Tris, 192 mM glycine, and 0.015% SDS. Complete transfer of all protein was achieved in 3 h at 50 V or overnighta t 20 V. Membranes were routinely incubated with PBS-3% bovine serum albumin for 2-4 h a t 22 "C and then transferred to PBS-1% bovine serum albumin containing 4 pg/ml pure 9A7y and shaken overnight at 4 "C. Washing was achieved by gently shaking the membranesfor 90 min a t 22 "C in 50 ml of TBS-0.05% Tween 20 with four changes. Following a short preincubation in PBS-3% bovine serum albumin (10 min), the sheet was developed for 2 h in PBS-1% bovine serum albumincontaining 6 X 10' cpm/ml of purified rabbitanti-9A7y previously iodinated using chloramine T (20). In later experiments, an alternative and preferablemethodology was employed; incubation of the sheets withmonoclonal antibody 4A5y or 9A7y (4 pg/ml) was followed after washing with an identical incubation containing secondary antibody (anti-9A7y or anti-4A5y antiserum; 1/10,000). Finally, the sheets were developed utilizing radioiodinated protein A (30-40 pCi/pg; 75,000 cpm/ml). After additional washing exactly as above, themembranes were driedandthenautoradiographed at -70 "C for various lengths of time (24-72 h), using Kodak X-Omat R P or AR film and a Du Pont Cronex Hiplus intensifying screen. General Techniques-Sedimentation analysis was carried out on 10-30% sucrose gradients prepared in KETD-0.3 asdescribed earlier (10). DNA-cellulose chromatography was performed also as described (6-8, 10). Protein analysis was carried out as described by Lowry et al. (21) using crystalline bovine serum albumin as standard. Silver staining was carried out as described by Merrill etal. (22). RESULTS

Immunoblot Analysis of 1,25-(OH),D3 Receptors-Utilizing 9A7y as a probe, immunoblot analysis of crude intestinal cytosol revealed two closely spaced, but distinct, radioactive signals migrating between bovine serum albumin and ovalbumin (Fig. 1). These species were also detectable when utilizing either anti-receptor antiserumor 4A5y antibody but were absent when nonspecific rat IgG was employed or when specific antibody was excluded (Fig. 1, lane 6). The immunologic signals were dependent upon cytosol concentration (Fig. 1, lanes 2-5), could be identified inhigh-saltextracts of mucosal nuclei which contain receptor (Fig. 1, lane 7), and were also detectableintotal cellular extracts prepared by homogenizing mucosal tissue directly in denaturing electrophoresis buffer (data not shown). No signals were generated when liver cytosol, atissue devoid of detectable 1,25-(OH)2D3binding, was examined (Fig. 1, lune 8). Whilethe larger molecularweightantigenicallyreactivespecies represented the major signal in cytosol, the more rapidly migrating polypeptide was consistently observed. In fact, its concentration is clearly elevated in nuclear extracts (lane7), an observation probably due to the fact that the larger form appears to be selectively solubilized during the preparation of nuclei (data not shown). The most likely explanation for the appearance of the smaller antigenically reactive species, however, is that it arises from the larger species through endogenous proteolysis. Several weaker signals a t -85 and -100-kDa are also evident in Fig. 1. The intensities of these species could not be enhanced relative to the putative receptor signals by either

Immunocharacterization of Chick l,25-(0HJ2D3 Receptors

Liver

Intestine

i

1 *)

I

x Y

5

92.5 69

53

1307

n

1

461

1

2

3

4

5

6

7

8

FIG. 1. Immunologic detection of 1,25-(OH)& receptor in extracts of chick intestinal mucosa. Intestinal mucosal or liver cytosols and mucosal “nuclear extracts” (see “Experimental Procedures”) were subjected to SDS-PAGE and then transferred to nitrocellulose for immunoblot analysis. Coomassie Blue-stain of intestinal cytosol (lane 1, 100 pg protein). Immunoblot of intestinal cytosol: lane 2, 200 pg; lane 3, 100 pg; lane 4, 50 pg; lane 5, 25 pg of cytosol protein probed with 9A7y (4 pglml). Lane 6, immunoblot of intestinal cytosol (200 pg protein) without 9A7-y. Lane 7, immunoblot of intestinal nuclear extract (200 pg) with 9A7y (4 pglml). Immunoblot of liver cytosol ( l a n e 8, 235 pg protein) with 9A7y (4 pglml). Molecular mass-stained standards are phosphorylase p, 92.5 kDa (14); bovine serum albumin, 69kDa (16); rat IgG heavy chain, 53 kDa (8); ovalbumin, 46 kDa (18); and 0-lactoglobulin, 18.4 kDa (16). Autoradiography was for 48-72 h at -70 “C.

rapid preparation of mucosal cytosol, direct tissue homogenization in denaturing buffer, or by the inclusion of protease inhibitors such as sodium molybdate or phenylmethylsulfonyl fluoride. Nevertheless, in view of the general proteolytic sensitivity of steroid receptor proteins, the presence of larger molecular weight immunologically reactive species in these experiments suggested the possibility that thereceptor forms migrating between bovine serum albumin and ovalbumin might have arisen from one of these parental macromolecules through proteolysis. In order to furtherevaluate this possibility, we employed an alternative immunoblot methodology outlined under “Experimental Procedures” which makes use of a secondary antibody “bridge” and radioiodinated protein A (9A7y and 4A5y are unreactive to proteinA (8)). As observed in Fig. 2, lane 3 and 5 , several immunologic signals are generated utilizing monoclonal antibody 9A7y and the latter methodologic approach. However, only two immunoreactive species are dependent upon the monoclonal antibody, the putative receptor species and a much larger macromolecule migrating nearthe origin of the gel (compare Fig. 2, lunes 1 , 3 , and 5.) Thus, the immunologic signals evident in Figs. 1 and 2 migrating between the two 9A7y-dependent species arise exclusively because of cross-reactive elements in the secondary antisera. Equally important, however, is the observation that 4A5y similarly reacts with the putative receptor signals identified with 9A7y but is unreactive to the larger protein species migrating near the gel origin (Fig. 2, lanes 7 and 8). Moreover, preparations of 9A77, while consistently reacting with the 60-kDa species, exhibit arange of reactivity with this large species, suggestive of an unknown reactive contaminant in certain 9A7y antibody preparations. Collectively, these data strongly support the premise that theavian receptor for 1,25-(OH)2D3is a protein whose molecular mass is approximately 60,000. In order to accurately determine the molecular size of these

FIG. 2. Immunologic detection of 1,25-(OH)zD3 receptor utilizing two separate monoclonal antibodies. Aliquots of intestinal mucosal cytosol (containing 50 pg of protein and approximately 0.5 pmol receptor) were subjected to SDS-PAGE and thentransferred to nitrocellulose. Individual lanes were subjected to immunoblot analysis as follows: lane 1, anti-9A7y antiserum alone; lane 2, 9A7y alone; lane 3, 9A7y and anti-9A7y antiserum; lane 4, anti-rat IgG (Zymed) alone; lane 5,9A7y and anti-rat IgG (Zymed); lane 6, anti4A5y antiserum alone; lane 7, 4A5y and anti-4A5y antiserum; lane 8, 4A5y and anti-rat IgG (Zymed). All lanes were incubated with radioiodinated protein A as described under “Experimental Procedures.” Molecular mass standards areas in Fig. 1but include carbonic anhydrase, 30 kDa and lysozyme, 14.3kDa. Autoradiography was for 16 h a t -70 “C.

0.9

\

.A

‘B

SDSPAGEllMMUNOBLOT 0.8

& 5 8 0

0.7

-

A

E 0.6

5

50

I

200

250

300

(MrP FIG. 3. Molecular weight determination of the A and B forms of the chick 1,25-(OH)& receptor. Left, chick intestinal cytosol (200 pg protein) was subjected to SDS-PAGE and analyzed by immunoblot as described under “Experimental Procedures.” Radioactive [14C]proteinstandards are phosphorylase p, 92.5 kDa; bovine serum albumin, 69 kDa; ovalbumin, 46 kDa; carbonic anhydrase, 30 kDa; and lysozyme, 14.3kDa. The autoradiogram was exposed for uersus (RF)’~ for the immunoblot 48 h at -70 “C. Right, plot of (M,)” analysis of chick intestinal receptor presented on the left. Relative migrational standard proteins phosphorylase p, bovine serum albumin, ovalbumin, and carbonic anhydrase are indicated by e, whereas the migration of the A and B forms of receptor are indicated by 0. A plot of (M,)” versus (KJ” for the mobility of 1,25-(OH),D3receptor on Ultrogel AcA-44 as previously described (10) is also indicated where 1,25-(OH)*Dsreceptor is designated (0).Protein standards (0) are phosphorylase 0, bovine serum albumin, and carbonic anhydrase.

two receptor forms (designated A and B), we performed numerous immunoblot analyses of intestinal cytosol samples after electrophoreses on long-resolving gels. An example of this is observed in Fig. 3, which is accompanied to the right by a plot of (M,)” uersus ( I t f ) ” of the collected data. Based upon the example shown here, as well as M , estimates obtained from additional high resolution gels, the average molecular weights ( n = 6) of the two forms are 60,300 f 400 (S.E.) ( A ) and 58,600 & 700 (S.E.) ( B ) .These values compare favorably to a molecular weight of 59,000 estimated from a plot derived from gel filtration data utilizing Ultrogel AcA-

Immunochuracterization of Chick 1,25-(0H)2D3 Receptors

1308

44. These immunoblot results reveal not only the precise SDS-PAGE-derived molecular weight of the receptor in the absence of purification but also show that the receptor may be subject to rapid in vitro proteolysis. Immunologic Characterization of Receptor after DNA-Cellulose Chromatography-We next chromatographed 1,25(OH)2D3 receptors on DNA-cellulose and then examined selected fractions by immunoblot methodology in order to establish the DNA-binding characteristics of both the A and B receptor forms. DNA-binding capabilities of receptor are particularly sensitive to proteolysis (23), and it was conceivable that one form might be unreactive with the affinity medium. As illustrated in Fig.4, left, a typical receptor profile was achieved on DNA-cellulose in which a portion of the tritium (-25%) was unreactive (fall through) while the majority was retained and eluted as a typical receptor-hormone complex at 0.22 M KCl. As assessed by DEAE filter adsorption or antibody sedimentation displacement analysis, only a small fraction of the radioactivity not retained by the columnwas receptor-bound, and that level could be decreased by more rapid preparation of cytosols as well as by shorter labeling periods (data notshown). Both the A and B forms of the 1,25(OH)2D3 receptor bound to DNA and eluted with authentic 1,25-(OH)2[3H]D3.receptor complexes (see immunoblot inset, Fig. 4, left); neither couldbe detected in fractions which contain non-DNA-binding species (Fig. 4, right, lanes 1 and 2). These results provide additional evidence that the immunoreactive species we have identified in cytosol are 1,25(OH)2D3receptor forms. Interestingly, close inspection of the elution pattern of A and B suggests that the B form may exhibit slightly increased affinity for DNA, thus explaining the relatively greater resistance of this form to extraction from the nuclear fraction by KC1. It is also evident that while the 60-kDa species is most abundant in cytosol, after chromatography the 58-kDa form shows a major increase in signal intensity relative to the A form. This is neither a hormonenor time-dependent phenomenon since direct immunoblot of cytosol to which hormone has been added shows no conversion to or enhancement of the B form and no obvious relative decrease in the concentration of the 60-kDa species for as long as 24 h at 4 "C (data not shown). Thus, the nature of

this increase in the B form after chromatography remains unknown, although it likely represents a chromatographyfacilitated proteolytic cleavage. This could arise either through the selective enrichment of specific proteases capable of cleaving receptor or because of the elimination of bulk protein during the chromatography itself. In any event, these data provide the best evidence that theB form is a truncated version ofA. These findings also show that both A and B forms of 1,25-(OH)2D3receptor bind similarly to DNA and are thus notanalogous in properties to theA and Bpolypeptides comprising the avian progesterone receptor (24). Purification of 1,25-(oH)2D3 Receptor Via Group-selective Affinity Chromatography-In order to obtain additional supportive evidence that the cytosolic species we detected by immunoblot methodologywere 1,25-(OH)2D3receptors, we purified this protein from chick intestine so that comigration studies could be performed. 1,25-(OH)2D3receptors were isolated by sequential chromatography of high-salt chromatin extracts on DNA-cellulose, hydroxylapatite, Ultrogel AcA-44 gel filtration, DEAE-cellulose, and blue dextran-Sepharose as described earlier (6, 10). Fig. 5 illustrates the final chromatographic step of the 1,25-(OH)2D3receptor on blue dextranSepharose in which a typical symmetrical peak of radioactivity was eluted at 0.45 M KC1. The inset represents the silverstaining pattern of polypeptides after SDS-gel resolution of the proteins present in several indicated fractions. While a number of species are present inthe 45 to 60-kDa range, two especially abundant proteins appear to be particularly good candidates for the A and Bforms of the receptor. Immunologic Detection of Purified 1,25-(0H)2D3Receptor and Comigration with Cytosolic Forms-Prior to performing immunoblot analysis of the purified receptor sample, we assessed whether the enriched native 1,25-(OH)2[3H]D3.receptor complex retained immunoreactivity. This was important because all of our previously reported observations focused

--

"

n I

12

3 4

10

5

J. 0

10

20

30

40

50

Fraction Number

FIG. 4. Immunoblot analysis of DNA-binding forms of the chick intestinal receptor. Left,intestinal cytosol (10% homogenate (w/v), 5 ml) was labeled with 1,25-(OH)$H]D3 (4 nM) for 1.5 h a t 4 "C. Followingre-equilibration in KETD-0.05 by exclusion chromatography on Sephadex G-25, cytosol was then chromatographed on a 2.5 X 3 cm DNA-cellulose column and eluted during a 0.05-0.5 M KC1 gradient. Inset reveals immunoblot analysis of the radioactivity (fractions 38-48) eluting between 0.2-0.4 M KCl. Right, immunoblot analysis of the DNA-cellulose chromatographic fractions 4 (lane I ) and 5 (lane2) (representing non-DNA-binding proteins), fractions40 (lane 3) and (lane 4 ) 41 (peak DNA-binding fractions), and crude intestinal cytosol (lane5) (50 pg protein). Molecular weight standards are as in Fig. 1. Autoradiography was for 72 h a t -70 "C.

25

30

35

40

Fraction Number FIG. 5. Chromatography of highly enriched chick 1,25(OH)2Dsreceptor on blue dextran-Sepharose. 1,25-(OH),[3H] D3. receptor complexes (4 pmol) previously fractionated as described under "Experimental Procedures" were chromatographed on a 1X 2cm column of blue dextran-Sepharose and eluted as a single peak of radioactivity at 0.45 M KCl. Inset, aliquots (0.05 ml) of the indicated fractions (32-34) were analyzed by SDS-PAGE, and thesubsequently resolvedproteins were stained with silver. Molecular weight standards are as indicated in Figs. 1 and 2.

nes

Immunochuracterization of Chick 1,25-(0HjZD3 Receptors upon immunologicreactivity with crude receptors and because proteolysis during isolation could potentially releaselarge hormone-bound receptor fragments incapable of reactivity to monoclonal antibody. Fig. 6A illustrates complete reactivity of the 1,25-(OH)zD3.receptorcomplex to the 9A7y antibody after sedimentation displacement analysis was carried out in 10-30% linear sucrose gradients. As a result, immunoblot analysis was then performed on 5 and 20 ng of purified receptor sample and compared to the Coomassie Blue stain of approximately 2 pg. In addition, intestinal cytosol wasalso electrophoresed and immunoblotted to evaluate the migrational characteristics of the putative receptor species compared to that generated after enrichment. Clearly, as seen in Fig. 6B,strong immunologically generated signals are evident (Fig. 6B, lanes 2 and 3) which comigrate with two stainable species at 58,600 and 60,300 daltons (lane 1).Most important, these species exactly comigrated with the A and B forms identifiable in cytosol extracts. These observations, coupled with that of sedimentation displacement (Fig. 6A), suggest that the protein species identified by Coomassie Blue stain but unreactive to antibody are unlikely to have contained 1,25-(OH)2D3prior to denaturation, although they could be receptor fragments which no longer contain the epitope for monoclonal antibody. Since both A and B forms, as well as a weakly reactive species at 50 kDa were identified by immunoblot analysis, we cannot ascertain which of these species bound 1,25-(OH)zD3 prior to denaturation. It is alsoapparent that the B form of the chick receptor has clearly becomethe dominant species after purification, both by stain visualization as well as by immunoreactivity. This observation is consistent with results identified in Fig. 4 and suggests a progressive proteolytic conversion to form B. Positive Immunoselection and Homogeneous Isolation of A and B Forms of the Avian 1,25-(0H,,D3 Receptor-Having ascertained the identity of the 1,25-(OH)zD3receptor, we then determined the ability of the antibodies to immunoselect and enrich receptors for 1,25-(OH)2D3.Nuclear extracts were incubated batchwise overnight with 9A7y or 4A5y derivatized

1309

to Sepharose-4Bthrough a CNBr-activated intermediate. Nuclear extracts were utilized becauseof the slight (severalfold) enrichment which can be achievedby this maneuver, although cytosols were equally useful.After overnight incubation, the resins were extensively washed and the receptor eluted with 3 M guanidine-HC1as described under "Experimental Procedures." As illustrated in Fig. 7, several relatively abundant protein species wereapparent after isolating receptors via this protocol, two of which appeared to be similar concentrations of the A and B forms of the avian 1,25-(OH)zD3receptor. This was consistent with the observation in Fig. 1that nuclear extracts favor a more equalabundance of each of these forms, 1,25-(OH)z[3H]D3 receptors could not be adsorbed when unsubstituted Sepharose was utilized, and no protein species analogous to A and B in electrophoretic mobility could likewise beidentified after SDS-PAGEand Coomassie Bluestaining. We immunoblotted several percent of the samples isolated by immunoaffinity to see if the two protein species we suspected to be receptor retained immunoreactivity after denaturation. This analysis represents a more stringent testfor 1,25-(OH)zD3 receptor as itfavors predominantly high affinity interactions. In both cases, a very strong duplex signal was generated by this maneuver, providingcorroborative evidence that these species did represent the 1,25-(OH)2D3receptor. The guanidine-HC1-eluted receptor proteins were capable of

I

c3

X

""1

-A 'B

A

B

B

A

-B

14.3 D

j

1 2 3 4 5 6 FIG.7. Positive immunoselection and homogeneous isolation of the chick 1,25-(OH)2Ds receptor from nuclear highsalt extracts. Nuclear extract from 6-8 g of intestinal mucosa was shaken batchwise overnight a t 4 "C with either 4A5y-Sepharose or 9A7y-Sepharose (0.05-0.1ml). The beads were collected by brief centrifugation, washed extensively (see "Experimental Procedures"), I DJ and receptor-eluted with 3 M guanidine-HC1. The receptor-containing 1 2 3 4 0 10 20 30 40 samples were then subjected to SDS-PAGE and either Coomassie Number Fraction Blue-stained or analyzed via immunoblot. Lane I , Coomassie stain of FIG. 6. Immunologic identification of purified chick 1,25- the protein sample resulting from purification of mucosal receptor on (OH)2Ds receptor A and B forms and their comigration with 4A5y-Sepharose. Lune 2, immunoblot of 2% of sample. Lane 3, species detected in cytosol. A , purified 1,25-(OH)2[3H]D3.receptor Coomassie stain of protein sample resulting from purification of complexes (5.2 pmol) were incubated for 2 h a t 4 "C with (a)or receptor on 9A7y-Sepharose. Lane 4, immunoblot of 2% of the without (0)a 30-fold molar excess of 9A7y and then sedimented on sample. In both cases, approximately 4-5 pg of receptor in either the 10-30% sucrose gradients prepared in KETD-0.3. Fractionation was A or B forms (arrows)was isolated. Lanes 5 and 6 reveal SDS-PAGE from the top, and the arrows represent external protein standards 1) analysis of receptor isolated from 6-8 pg of intestinal mucosal nuclei ovalbumin (3.7 s); 2) bovine serum albumin (4.4 s); and ratIgG (7 s). with 9A7y-Sepharose as above. The immunoaffinity isolates were B, purified chick ~ , ~ E J - ( O Hreceptor ) ~ D ~ was subjected to SDS-PAGE subjected to SDS-PAGE, and theunstained protein excised from the and then analyzed by Coomassie Blue stain ( l a n e 1, 1.6 pg receptor gel individually based upon comigration with a Coomassie Blueprotein) or by immunoblots ( l a n e 2, 5 ng and lane 3, 20 ng receptor stained parallel gel. The A ( l a n e 5) and B ( l a n e 6 ) forms each were protein). Crude chick cytosol was evaluated by immunoblot ( l a n e 4, eluted in 1ml of 0.05 M Tris-HC1, pH 7.5,0.2 M NaCl, 0.1% SDS for 200 pg total protein). Protein standards are as in Fig. 2. Autoradiog- 1h a t 4 "C, acetone precipitated, and thesamples reanalyzed by SDSPAGE. raphy was for 48 h a t -70 "C.

1310

Immunocharacterization of Chick 1,25-(OH)2D3 Receptors

rebinding 1,25-(OH)2D3when the denaturant was reduced to Moreover, these reactive bands comigrate on SDS-PAGE with 0.3 M (data not shown). Due to the presence of other protein the same receptor bands detected immunologically in crude species and the low recovery of hormone binding, however, cytosol. These resultsclearly show that while proteolysis leads no conclusions could be drawn. DNA-binding activity could to a significant depletion of the intact A monomer (the 58not, in contrast, be recovered (data not shown). The58- and kDaspecies is predominant), the same 60-kDa protein is 60-kDa forms of the 1,25-(OH),D3 receptor were then homo- presentinboth crude and purifiedreceptor preparations. geneously isolated by running gel lanes fromwhich slices Equally important is the observation that immobilized antia parallel body is capable of selecting these receptor species directly containing eitherform were cut after alignment with Coomassie Blue-stained gel. The buffer-eluted proteins were from nuclear extracts and, in tandem with SDS-PAGE, can essentially homogeneous as analyzed by an identical second- be employed to purify easily the 1,25-(OH),D3 receptor to ary electrophoresis and provide for the first time the unequiv-apparent homogeneity. ocal purification of the monomeric form of the avian 1,25The significance andinterrelationship between the two (OH),D3 receptor as well as what may be an important pro- antigenically related A and B proteins which represent the teolytic cleavage product of this protein. 1,25-(OH),D3 receptor remaintobe unequivocally determined. The most likely explanation consistent with the data DISCUSSION presented here is that theB form (58 kDa) arises artifactually as a result of degradation. Certainly, increases in the apparent The results described here represent compelling evidence concentration of thissmallerpeptideareevident in vitro that we haveidentified the chicken 1,25-(OH),D3 receptor immunologically after SDS-polyacrylamide gel electrophore- during both DNA-cellulose chromatography and the sequensis. The protein is present in homogenates of tissues such as tial chromatographic steps essential for purification. Moreover, as described by Allegretto et al. (26), the 1,25-(OH)2D3 intestineand kidney,which contain1,25-(OH),D3-binding activity but is absent in liver which is low or devoid of such a receptor is clearly susceptible to additional forms of endogeproperty. The immunoreactivespecies bind toDNA-cellulose nous proteolytic degradation. Alternatively, one might specunder low ionic strength conditions and coelute during a linear ulate that the A and B peptides represent synthetic products complex. of differentially processedmRNA transcripts as recently idensalt gradient with the 1,25-(OH)2[3H]D3.receptor The receptor displays an M , of 60,300 which is consistent tified forglucocorticoid receptors (27)or represent post-transwith previous estimates of its molecular weight by gel filtra- lationally modified species. This possibility has been shown receptor as a result of tion (4-11). Of major importance in the determination of this to occur for mammalian 1,25-(OH)2D3 receptor size estimate is the fact that was it made independent phosphorylation (28). Until specific proteolytic enzymes are of any purification steps andis thus less likely to be compro- identified which permit a systematic evaluation of the cleavmised by proteolytic artifacts often associated with the puri- age of the 60-kDa species to 58 kDa, these alternativepossification of steroid receptors. While it is conceivable that the bilities cannot be entirelyruled out. Dameet al. (29)recentlypartially purified the porcine 60-kDa receptor arises during tissue preparation via rapid and complete proteolysis of a larger parental form, we obtained receptor for l,25-(OH)*D3 and, via immunologic as well as no evidence for this possibility even when solubilizing and hormone-binding renaturation studies, have identified it as a boiling tissue directly in SDS-denaturing buffer. Thus, it is polypeptide of approximately 55 kDa. While a definitive coniikely that the chicken 1,25-(OH)2D3 receptor is a protein of clusion of the size of'the porcine receptor cannot be drawn in possibility of proteolytic degradation 60 kDa although these data do not define the composition of these studies due to the the functional intracellular form (25). This immunologic char- within the purified material, we have likewise identified this acterization, coupled with a previous report (23),provides the receptor as a species of 54 kDa by immunoblot analysis of basis for the work ina companionpaper(26) where we solubilized porcine kidney cells (30) as well as by cell-free describe the use of proteases to structurally map the chicken translation of poly(A)+ RNA isolated from these same cells (31). Taken together, these data, coupled withsimilar analyses 1,25-(OH),D,,receptor and characterize its important funcof mouse, rat, monkey, and human 1,25-(OH),D3 receptors as tional domains. Additional support for the successful immunoblot detection proteins of 54.5, 54, 52, and 52 kDa, respectively(30-32), of the 1,25-(0H),D3 receptorderives from the purification of support the premise that the chicken 1,25-(OH),D3receptor the protein by nonimmunologic means. We originally devised represents a larger molecular mass version of this protein. procedures under which the receptor could be purified to near These results are consistent with the finding that mammalian homogeneity by conventional techniquescomprised of group- receptors sediment more rapidly during sucrose density graselective affinity and gel filtration chromatography (6, 10). dient analysis (33). This difference in molecular weight is of of the current strucThesesequentialchromatographic procedures resulted in considerable interest, particularly in light emergingfrom the highly enriched receptor preparations as determined by cal- turalandfunctionalinsightpresently molecular cloning of other steroid receptor genes (27, 34-37). culation of specific activity(1,25-(OH)2D3-bound/mgprotein), which were successfully utilized as immunogen toraise These datasuggest a similarity of gene organization such that the putative DNA-binding and hormone-binding domains are specific antisera to the 1,25-(OH)2D3 receptor in rats (10). However, despite the appearance of protein bands at 58, 60, positionally conserved from the C terminus, and differences in the size of the steroid receptors arise from additional Nand 65 kDa after SDS-PAGE, several of which comigrated terminal amino acids. If this overall organization similarly with the radiolabeled hormone-receptor complex on DNAcellulose, we were unable todefinitely ascertain which of these describes the 1,25-(OH)2D:j receptor, and the results of our species represented the 1,25-(OH)2D:3 receptor. This failure to companion paper (26) are consistentwith this interpretation, conclusively identify the receptor was also true of other at- then the increase in size of the chicken receptor may be due tempts toisolate the same protein(5,9). In contrast, the data to additional unconserved amino-terminal sequence. Clearly, obtained here employing similar isolation procedures as above the molecular cloning of the chicken lr25-(OH)2D3receptor gene will enable a more precise structural analysis of this indicate that two major Coomassie Blue-stainable bands at protein to be carried out. 58 kDa and 60 kDa are evident, which also exhibit strong In conclusion, the chicken receptor for 1,25-(OH)2D3 has immunoreactivity when examined by immunoblot techniques.

Immunocharacterization of Chick 1,25-(OHj2D3 Receptors been identified under denaturing conditions as a polypeptide of 60,300 daltons. The unequivocal identification and isolation of this protein should facilitate further biochemical characterization and functional analysis.

-

Acknowledements-We eratefullv acknowledee the technical assistance of S. L. Marion inpreparing antiserumto monoclonal antibody 9A7y and 4A5y. We also thank Karen Olson and Paula Leece for UreDarine -~ - the manuscript. Y

I

REFERENCES 1. DeLuca, H. F., and Schnoes* H. K. (1976)Annu.

Biochem. 45,631-666 2. Haussler, M. R., and McCain, T.A. (1977) End. J. Med. 2979 974-978,1041-1050 3. Brumbaugh, p. F.7 and Haussler~ M. R. (1975)J. BioL. 250,1588-1594 4. Wecksler, W. R., and Norman, A. W. (1980)J. Steroid Biochem. 13,977-989 5. Simpson, R. U., and DeLuca, H. F. (1982)Proc. Natl. Acad. Sci. U. S. A. 79,16-20 6. Pike, J. W., and Haussler, M. R. (1979)Proc. Natl. Acud. Sci. U. S. A. 76, 5485-5489 7. Pike, J. W. (1982)J. Biol. Chem. 257,6766-6775 8. Pike, J. W. (1984)J. Biol. Chem. 259, 1167-1173 9. Simpson, R. u., Hamstra, A.3 Kendrick, N. c., and DeLuca, H. F. (1983)Biochemistry 22,2586-2594 10. Pike, J. W., Marion, s. L., Donaldson, c. A., and Haussler, M. R. (1983)J . Biol. Chem. 258, 1289-1296 11. Franceschi, R. T., and DeLuca, H. F. (1979)J. Biol. Chem. 254, 11629-11635 12. Bishop, J. E., Hunziker, w., and Norman,A. w . (1982)Biochem. Biophys. Res. Commun. 108, 140-145 13. Haussler, M. R., Chandler, J. s.,Hagan, L. A., and Pike, J. w. (1980)Methods Enzymol. 67, 529-542 14. Tanford, C., Kawahara, K., and Lapanje, s. (1967)J . Am. Chem. SOC.89, 729-742 15. Titani, K., Koides, A., Hermann, J., Ericsson, L. H., Kumar, S., Wade, R. D., Walsh, K.A., Neurath, H., and Fiscber, E. H. (1977)Proc. Natl. Acad. Sci. U. S. A. 74,4762-4766 16. Castellino, F. J., and Barker, R. (1968)Biochemistry 7, 22072217 17. McNutt, K. W., and Haussler, M. R. (1973)J. Nutr. 103, 681689

1311

18. Laemmli, U. K. (1970)Nature 227,680-685 19. Burnette, w. N. (lg81) Biochem. 112$ 195-203 20. Greenwood, F. C., Hunter, W. M., and Glover, J. S. (1963) Biochem. J. 89,114-123 21. Loww, 0. H.. Rosebrouph. N. J., Farr. A. L.. and Randall. R. J. (19ii)J. B~OZ. Chem. i93,265-275 22. Merrill. C. R.. Goldman. D.. Sedman. S. A,. and Ebert. M. H. (1980)Science 21 1, 1437-1438 23. Allegretto, E. A., and Pike, J . W. (1985)J. Biol.Chem. 260, 10139-10145 24. Schrader, W. T., Birnbaumer, M. E., Hughes, M. R., Weigel, N. L., Grady, W. W., and O’Malley, B. W. (1981)Rec. Prog. Horm. Res. 37, 583-633 25. Sherman, M. R., and Stevens, J . (1984)Annu. Reu. Physiol. 46, 83-105 26. Allegretto, E. A., Pike, J . W., and Haussler, M. R. (1987)J. Biol. Chem. 1312-1319 27. Hollenburg, S. M., Weinberger, C., Ong, E. S., Cerelli, G., Oro, A,, Lebo, R., Thompson, E. B., Rosenfeld, M. G., and Evans, R. M. (1985)Nature 318,635-641 28. Pike, J. W., and Sleator, N. M. (1985)Biochem.Biophys.Res. Commun. 131,378-385 29. Dame, M. C., Pierce, E. A., and DeLuca, H. F. (1985)Proc. Natl. Acud. Sci. U. S. A. 82,7825-7829 30. Sleator, N.J., and Pike, J. W. (1985)Am. SOC.Bone Mineral Res. Seventh Annual Scientific Meeting, Washington, D.C., June 15-18 (abst. 135) 31. Mangelsdorf, D. J., Pike, J. W., and Haussler, M. R. (1987)Proc. Natl. Acad. Sci. U. S. A., in press 32. Pike, J. W. (1985)in Vitamin D, Chemical,Biochemical, and Clinical Update (Norman, A. W., Schaefer, K., Grigoleit, H.G., and D. V. Herrath, eds) pp. 97-104, Walter de Gruyter, Berlin 33. Kream, B. E., Yamada, S., Schnoes, H. K., and DeLuca, H. F. (1977)J. Biol. Chem. 252, 4501-4505 34. Greene, G. L., Gilna, P., Waterfield, M., Baker, A., Hort, Y., and Shine, J. (1986)Science 231, 1150-1154 35. Krust, A., Green, S., Argos, P., Kuman, V., Walter, P., Bornert, and J., Chambon, P. (1986)EMBO J. 5,891-897 36. Conneely, 0.M., Sullivan, W. P., Toft, D. O., Birnbaumer, M., Cook, R. G., Maxwell, B. L., Zarucki-Schulz, T., Greene, G. L., Schrader, W. T., and O’Malley, B. W. (1986)Science 233, 767-770 37. Meisfeld, R., Rusconi, S., Godowski, P. J., Maler, B. A., Okret, S., Wikstrom, A.C., Gustafsson, J.-A., and Yamamoto, K. R. (1986)Cell 46,389-399