Bioconjugate Chem. 1995, 6, 541-548
541
High-Yield Affinity Alkylation of the Atrial Natriuretic Factor Receptor Binding Site Xiaolan He,' Koji Nishio,i a n d Kunio S. Misono" Department of Molecular Cardiology, The Cleveland Clinic Foundation Research Institute, 9500 Euclid Avenue, Cleveland, Ohio 44195-5071. Received February 9, 1 9 9 P
To facilitate characterization of the atrial natriuretic factor (ANF) receptor, we have developed an affinity labeling procedure, stepwise affinity labeling, which allows specific labeling of ANF binding sites in adrenal plasma membranes a t high yields. An iodoacetyl (IAc-), bromoacetyl (BrAc-), or maleimidobenzoyl group was attached to the amino-terminal a-amino group of the ANF(4-28) peptide, and the peptide derivatives were radioiodinated at Tyr-28 to obtain affinity reagents, N4a-IA~-[1251]ANF(4-281, N4U-BrAc-[12511ANF(4-28), and N4a-(maleimidobenzoyl)-[12511ANF(4-28). Receptor labeling was carried out in a stepwise fashion as follows: (1)Membranes were treated with p-chloromercuriobenzenesulfonic acid (PCMBS) or N-ethylmaleimide to block sulfhydryl groups; (2) the affinity reagent was allowed to bind to the receptor a t 0 "C for 1 h; and (3) the membranes were washed to remove unbound reagent and were incubated a t room temperature to effect alkylation reaction. Sodium dodecylsulfate (SDSI-polyacrylamide gel electrophoresis (PAGE) followed by autoradiography revealed incorporated, the labeling specific labeling of a 130-kDa ANF receptor. On the basis of 1251-radioactivity yields were estimated to be 70%, 52%, and 21% for the reactions with I A C - [ ~ ~ ~ I I A N F BrAc(~-~~), [lz5I1ANF(4-28), and (maleimidobenzoyl)-[12511ANF(4-28), respectively. The efficiency of receptor labeling by the stepwise procedure using I A C - [ ~ ~ ~ I ] A N Fwas ( ~ -27-fold ~ ~ ) greater than that obtained by photoaffinity labeling using N ~ B Z - [ ~ ~ ~ I ] A Nand F ( 63-fold ~ - ~ ~ )greater than that by direct crosslinking using disuccinimidylsuberate and [lZ5I]ANF(4-28) under comparable conditions. Digestion of the membrane protein labeled with I A C - [ ~ ~ ~ I I A N Fby ( ~BrCN, - ~ ~ ) endoproteinase Glu-C, and endoproteinase Lys-C gave single radiolabeled bands with apparent masses of 40, 18, and 29 kDa, respectively. Reversed-phase HPLC separation of the digests also gave single major peaks. The confinement of the affinity label to one major fragment in each digest suggests that the cross-linking occurred a t a single or a limited number of sites. The stepwise affinity labeling with the high crosslinking yield and specificity may be useful for analyzing the ANF receptor binding site structure.
Atrial natriuretic factor (ANF)l is a peptide hormone secreted by the heart atrium that has potent natriuretic (de Bold et al., 1981) and vasorelaxant activities (Currie et al., 1983; Grammer et al., 1983). The actions of ANF a t its target organs, including the blood vessels, kidney, and adrenal gland, are mediated by cell membrane receptors that are directly coupled to guanylate cyclase (Cantin and Genest, 1985; Gerzer et al., 1987). The ANF receptor molecule consists of a single polypeptide chain with a molecular mass of about 130 kDa (Kuno et al., 1986; Takayanagi et al., 1987; Meloche et al., 1988), containing a n extracellular ANF-binding domain, a single transmembrane sequence, and an intracellular region containing both protein kinase-homologous domain and guanylate cyclase domain (Chinkers et al., 1989). It has been suggested that, in the basal state, the kinasehomologous domain interacts with the guanylate cyclase domain, suppressing guanylate cyclase activity. Binding of ANF to the extracellular domain causes a conforma-
* To whom correspondence should be addressed. E-mail:
[email protected]. Tel: (216) 444-2054. FAX: (21614449263.
' Present address: Department of Pharmacology, Case Western Reserve University, Cleveland, OH 44106. Present address: Department of Anatomy, Nagoya University School of Medicine, Nagoya 466, Japan. Abstract published in Advance ACS Abstracts, August 15,
*
@
1995.
Abbreviations: ANF, atrial natriuretic factor; IAc, iodoacetyl; BrAc, bromoacetyl; N3Bz-, azidobenzoyl; PCMBS, p chloromercuriobenzenesulfonic acid; SDS, sodium dodecylsulfate; PAGE, polyacrylamide gel electrophoresis. 1043-1802/95/2906-0541$09.00/0
tional change that eliminates this suppression, thus elevating guanylate cyclase activity (Chinkers and Garbers, 1989). ATP, an allosteric effector, bound to the protein kinase-homologous domain, enhances ANF stimulation of guanylate cyclase. Although such a chain of events has been postulated, the actual structure of the receptor or the mechanism of interaction between ANF and the receptor remains largely unknown. This is in part because the binding site structure of the ANF receptor is not known. Traditionally, chemical labeling of peptide receptor binding sites has been carried out either by photoaffinity labeling or by affinity cross-linking methods. However, both methods give low yields of labeled binding sites. In photoaffinity labeling, radicals are generated by photolysis. Photoradicals react indiscriminately a t multiple sites on receptor polypeptides as well as with the surrounding solute and solvent molecules. Consequently, the net yield of cross-linking is small, and the reaction yields a complex mixture of unknown derivatives, making structural analysis extremely difficult (Ruoho et al., 1984). The bifunctional reagents used in affinity cross-linking are mostly electrophiles and have some level of chemical selectivity. However, in the major reaction path, crosslinker reagents simultaneously attack the potential acceptor residues on the receptor and ligand. This simultaneous attack leaves the other end of the cross-linker unable to cross-link and, hence, is unproductive in receptor labeling. For this reason, the net yield of the cross-linking generally does not exceed 5- 10% (Pilch and Czech, 1984). Given the chemical complexity of the 0 1995 American Chemical Society
542 Bioconjugate Chem., Vol. 6, No. 5, 1995
He et al.
amino acids with o-phthalaldehyde in the presence of 3-mercaptopropionic acid and secondary amino acids with 9-fluorenylmethylchloroformate using the Amino-Quant HPLC System (Hewlett Packard, Wilmington, DE) according to the protocols provided by the manufacturer. Amino acid sequence analysis was carried out by Edman degradation in an Applied Biosystems Model 470A protein sequencer. lZ5Iradioactivity was measured in a Packard AutoGamma 500 y counter (Meriden, CT) a t a counting efficiency of 80%. To estimate the specific radioactivity of peptides, the amount of radioiodinated peptide was determined on the basis of the height of the W absorption peak in the HPLC chromatogram. Corresponding iodinated but nonradioactive peptides were prepared, quantitated by amino acid analysis, and used as the EXPERIMENTAL PROCEDURES standards in the HPLC. The molecular masses of pepMaterials. ANF(4-28) with the sequence Arg-Sertides were determined by electro-spray ionization mass Ser-Cys-Phe-Gly-Gly-Arg-Ile-Asp-Arg-Ile-Gly-Ala-Glnspectrometry carried out a t the Protein and Carbohydrate Ser-Gly-Leu-Gly-Cys-Asn-Ser-Phe-Arg-Tyr (Misono et al., Structure Facility a t the University of Michigan. 1984a) was synthesized by the solid-phase method in an Preparation of Bovine Adrenal Cortex Plasma Applied Biosystems Model 431A peptide synthesizer Membranes. Bovine adrenal cortex plasma membranes (Foster City, CAI. The initial loading of 9-fluorenylwere prepared by a modification of the method of Glossmethyloxycarbonyl(Fmoc)amino acid on a (44hydroxymann et al. (1974) as follows. Bovine adrenal glands methy1)phenoxy)methyl resin was effected by activating were obtained from a local slaughterhouse. The outerto its symmetric anhydride with dicyclohexylcarbodiimide most slices (1-2 mm thick) of the adrenal glands were in the presence of (dimethy1amino)pyridine. Subsequent collected using a Staddie-Riggs microtome. The slices coupling of Fmoc-amino acid to the growing peptide chain were homogenized in four volumes of 10 mM NaHC03 was effected by in situ activation and coupling using in a blender for 20 s. Homogenization was repeated in a 2-( lH-benzotriazol-1-y1)-1,1,3,3-tetramethyluronium hexaPolytron homogenizer (Brinkman, Westbury, NY) a t the fluorophosphate as an activating reagent in the presence setting of 7 for 40 s. The homogenate was centrifuged of N-hydroxybenzotriazole. After synthesis, the peptide a t 1500g for 15 min a t 4 "C. The supernatant was was cleaved from the resin and deprotected by treatment collected, and the centrifugation was repeated. The with a mixture of trifluoroacetic acid (10 mL), phenol supernatant then was collected and centrifuged a t 50000g (0.75 g), ethanedithiol (0.25 mL), thioanisole (0.5 mL), for 2 h. The pellet was resuspended in 50 mM Trisand dionized water (0.5 mL) a t room temperature for 2 HC1 buffer, pH 7.4, containing 0.15 M NaC1,5 mM MgC12, h. After filtration, the peptide was precipitated by 0.1% bovine serum albumin, and 0.05% bacitracin a t a methyl tert-butyl ether. The precipitate was washed membrane protein concentration of approximately 1.5 several times with methyl tert-butyl ether, dissolved in mgImL. The membrane suspension was divided into water, and lyophilized. The disulfide bond between the aliquots, quickly frozen in liquid nitrogen, and stored a t Cys-7 and Cys-23 residues was closed by oxidation with -80 "C. dropwise addition of 10 mM KJFe(CNI6I into a dilute Preparation of N4a-IA~-ANF(4-28). To ANF(4-28) solution of the peptide (approximately 0.1 mM) a t pH 7 (10 nmol) in 50 pL of 50 mM HEPES buffer, pH 8.0, was with stirring a t room temperature until a pale yellow added a 4-fold excess of IAc-N-hydroxysuccinimidein 10 color remained for 1 h (Sugiyama et al., 1984). The pL of dimethylformamide. The reaction was allowed to purification of the peptide was carried out by reversedproceed for 2 h a t room temperature. The reaction phase HPLC on a Vydac C-18 column (2.2 cm x 25 cm, mixture was acidified by adding an equal volume of 2 M The Separations Group, Hesperia, CA) using a linear acetic acid and was then chromatographed directly on a CH3CN gradient (from 2% to 60%)in 0.1% trifluoroacetic Vydac C-18 reversed-phase column (4.6 mm x 250 mm, acid in water applied over a period of 60 min a t a flow Separations Group). Elution was carried out with a rate of 15 mumin. Rat ANF(1-28) was prepared in a linear CHSCN gradient from 0% to 60% over a period of 40 similar manner. ANF(1-28) was radioiodinated and min in 0.1% trifluoroacetic acid in water a t a flow rate used in receptor binding assay. of 1 mumin. Elution was monitored by W absorption (Maleimidobenzoy1)-N-hydroxysulfosuccinimide and Ioa t 214 nm. These conditions were used as standard dogen (1,3,4,6-tetrachloro-3,6-diphenylglycoluril~ were conditions for all HPLC separations. N 4a-IAc-ANF(4obtained from Pierce (Rockford, IL). BrAc-N-hydrox28) was collected in a peak eluting a t 24.0 min, while p-chloromerysuccinimide, IAc-N-hydroxysuccinimide, authentic ANF(4-28) was eluted a t 23.0 min. Electrocuriobenzenesulfonic acid (PCMBS), and diisopropylfluspray mass spectrometric analysis of this material gave orophosphate were from Sigma (St. Louis, MO). [1251]NaI a molecular mass of 2876.4 Da, which is consistent with (2.4 mCi/nA of iodine) was obtained from Amersham the calculated mass value for N 4a-IAc-ANF(4-28). Amino (Arlington Heights, IL). Endoproteinase Glu-C (Staphyacid analysis gave a composition identical to that of ANFlococcus aureus protease V8; EC 3.4.21.19) was obtained (4-28). Edman degradation gave no phenylthiohydanfrom Worthington (Freehold, NJ) and endoproteinase toin amino acid, being consistent with iodoacetylation of Lys-C (EC 3.4.99.30) from Boehringer-Mannheim (Inthe N-terminal a-amino group. The yield of N 4a-IAcdianapolis, IN). All other chemicals were of reagent ANF(4-28) ranged from 50% to 90%, depending on the grade. preparation. Optimal yields were obtained with a 4- to Analytical Procedures. Amino acid analysis was 10-fold excess of the N-hydroxysuccinimide ester over the carried out after hydrolysis in 6 N HCl with 0.1% phenol peptide. N 4a-IAc-ANF(4-28) was aliquoted, dried in a a t 110 "C for 20 h. Analysis was performed by reversedphase HPLC after precolumn derivatization of primary Savant SpeedVac concentrator, and stored at -20 "C.
receptor protein and peptide ligand, the yield of one specific cross-linked structure is likely to be even lower. To overcome these problems, we have developed a procedure, stepwise affinity labeling, for the ANF receptor, The method allowed highly specific labeling of the ANF receptor with labeling yields as high as 70% of receptor sites in adrenal membrane preparations. Radiochemical peptide mapping of the affinity-labeled membrane after BrCN, endoproteinase Glu-C, or endoproteinase Lys-C digestion showed that the labeling was confined to a single major peptide fragment, indicating that the labeling reaction was highly specific. The potential usefulness of this method for structural characterization of the peptide receptor binding site is discussed.
Stepwise Affinity Labeling of Peptide Binding Site
Preparation of N4a-BrA~-ANF(4-28). The synthesis and isolation of N4"-BrAc-ANF(4-28) was carried out in the same manner as above. Preparation of N4"-(Ma1eimidobenzoyl)-ANF(428). (Maleimidobenzoy1)-N-hydroxysulfosuccinimide (40 nmol) in 5 pL of water was added to ANF(4-28) (10 nmol) in 50 pL of 50 mM HEPES buffer, pH 8.0. After 2 h, the reaction mixture was acidified by addition of an equal volume of 2 M acetic acid and chromatographed on a Vydac C-18 reversed-phase column a s described above. Elution was monitored by UV absorption a t 214 nm and a t 340 nm. N4a-(Maleimidobenzoyl)-ANF(4-28)eluted a t 2-3 min after the ANF(4-28) peak and showed absorption at both 214 and 340 nm. The absorption a t 340 nm is caused by the maleimide moiety. The purified N4"-(ma1eimidobenzoyl)-ANF(4-28)was aliquoted, dried under a vacuum, and stored a t -20 "C. Radioiodination of Peptide Derivatives. N4"-IAcANF(4-28) (5-15 pg) in 60 pL of 0.2 M potassium phosphate buffer, pH 7.6, was placed in a glass tube (12 x 65 mm) coated with 50-100 pg of Iodogen. The reaction was initiated by adding 1 mCi of carrier-free [12511NaI(the specific activity a t 2.4 mCi/nmol), unless otherwise specified. After 10 min, the mixture was acidified by adding a n equal volume of 2 M acetic acid and chromatographed directly on a Vydac C-18 column as described above. Fractions were collected a t 0.5 min intervals. Pass-through fractions were collected in tubes containing a solution of NazSz05 in water to convert I2 to nonvolatile iodide. lZ5Iradioactivity was monitored using a Beckman Model 120 Radioisotope detector (Beckman, Fullerton, CA). Monoiodo and diiodo derivatives of N4"-IAc-ANF(4-28) were eluted a t about 1and 2 min respectively, after the peak of uniodinated N 4a-IAc-ANF(4-28). The combined yield of the two radioiodinated derivatives was about 33% of the [12511NaIadded to the reaction mixture. Similar results were obtained in radioiodination of N 4a-BrAc-ANF(4-28) and N 4a-(maleimidobenzoyl)-ANF(4-28). All affinity-labeling experiments were carried out using monoiodinated peptide derivatives. For some experiments, [12511NaI(1 mCi) was diluted with 1.88 nmol of unlabeled NaI to a specific activity a t 0.43 mCi/nmol before the iodination reaction. To estimate the specific activity of radioiodinated peptides, corresponding nonradioactive, mono- and diiodinated peptides were prepared, quantitated by amino acid analysis, and used as the standards in the HPLC separation. The amount of radioiodinated peptide was determined on the basis of the height of the W absorption peak in the HPLC chromatogram. The specific radioactivities of the mono- and diiodo derivatives were estimated to be 0.40 mCi/nmol and 0.78 mCi/nmol, respectively, and were consistent with calculated values. Preparation of N'-Acetyl-L-lysine Adducts from N4a-IAc-A"(4-28),N4"-BrAc-ANF(4-28),and N4"(Maleimidobenzoy1)-ANF(4-28).N4"-LAc-ANF(4-28) (8 pg) was reacted with 0.25 M Ne-acetyl-L-lysine in 40 p L of 50 mM HEPES buffer, pH 8.5, a t room temperature for 24 h. Reversed-phase HPLC of the reaction mixture under standard conditions gave a new discrete peak eluting a t about 1 min after the peak of N4"-IAc-ANF(4-28) (data not shown), which contained the N e acetyllysine adduct of N 4a-(methylenecarbonyl)-ANF(428) as determined by mass analysis. N '-Acetyllysine adducts were prepared from N4"-BrAc-ANF(4-28) and in a similar from N4a-(maleimidobenzoy1)-ANF(4-28) manner. Receptor Binding Assay. A competitive binding assay with the bovine adrenal cortex membranes was
Bioconjugate Chem., Vol. 6,No. 5, 1995 543
carried out using [12511ANF(1-28)as a radioactive ligand (Misono et al., 1985). The assay mixture consisted of 50 mM Tris-HC1 buffer, pH 7.4, 0.15 M NaCl, 1 mg/mL bovine serum albumin, 0.5 mg/mL bacitracin, [12511ANF(1-28) (300000 cpdincubation, the final concentration of [12511ANF(1-28)a t approximately 0.14 nM), and varying concentrations of unlabeled ANF or its derivatives. Binding was initiated by adding bovine adrenal membranes (20 pg membrane protein) and continued for 1h at 0 "C. The total volume was 0.5 mL. The bound 1[12511ANF(1-28)was separated from free [1251]ANF( 28) by filtration through a Whatman GF/C glass filter that was pretreated with 0.3% poly(ethy1eneimine) for 10 min. Filters were washed eight times with 4 mL of 20 mM Tris-HC1 buffer, pH 7.4, containing 0.15 M NaCl. The 1251-radioactivitytrapped on the filter was counted using the Packard Auto-Gamma 500 gamma counter. Stepwise Affinity Labeling of A" Receptor. The reaction was carried out on the bovine adrenal plasma N 4a-BrAcmembranes using N4a-IA~-[1251]ANF(4-28), [lZ5IlANF(4-28),or N 4a-(maleimidobenzoy1)-[Iz5IIANF( 428) as the affinity reagent. Prior to the reaction, the adrenal membranes were treated with PCMBS or Nethylmaleimide to block sulfhydryl groups. This pretreatment step was essential for achieving specific labeling of the ANF receptors. The adrenal membranes (150 pg) were suspended in 90 p L of ice-cold 20 mM potassium phosphate buffer, pH 7.5, containing 2 mM EDTA and 0.15 M NaCl (buffer A). Ten pL of 50 mM PCMBS was then added to the membrane suspension and incubated at room temperature for 10 min, unless otherwise specified. In some experiments, the membranes were pretreated with 5 mM N-ethylmaleimide for typically 1h. At the end of the incubation, the membrane suspension was diluted with 1mL of buffer A containing 1mM phenylmethanesulfonyl fluoride, 0.5 mM diisopropylfluorophosphate, and 2 mg/mL of carboxymethylated bovine serum albumin (Sigma). The membranes were collected by centrifugation and then resuspended in 90 pL of the same mixture. The affinity labeling experiment was carried out in a stepwise manner as follows. Step 1 . Binding of the Affinity Reagent to the Adrenal N4a-BrA~-[12511Membranes. N4a-IA~-[12511ANF(4-28), ANF(4-281, or N4a-(maleimidobenzoy1)-[12511ANF(4-28) in 10 pL of water was added to the membrane suspension to a final concentration of 1 nM, unless otherwise specified. For the control experiment, unmodified ANF(4-28) was added to a final concentration of 1pM before the addition of the affinity reagent. The mixture was incubated at 0 "C in the dark for 60 min to allow binding. After the incubation, the membrane suspension was diluted with 1 mL of cold buffer A and centrifuged a t 4 "C a t 15000g for 5 min. The supernatant containing unbound reagent was removed. Step 2. Alkylation Reaction. The membranes were resuspended in 100 p L of buffer A and incubated a t room temperature for 4 h, unless otherwise specified, to effect alkylation reactions. After the reaction, the membranes were collected by centrifugation. To remove noncovalently bound affinity reagent, the membranes were resuspended in 1 mL of 20 mM sodium acetate buffer, pH 5.0, containing 0.15 M NaCl and 2 mM EDTA and incubated a t room temperature for 30 min. The membranes were collected by centrifugation and were separated by SDS-PAGE. After the gel was dried, radiolabeled protein bands were detected by autoradiography. Photoaffinity Labeling of A" Receptor. Photoaffinity labeling was carried out using N4"-(azidobenzoyll-ANF(4-28) [N4"-N3Bz-ANF(4-28)1 as a photoaffinity reagent as described previously (Misono et al.,
He et al.
544 Bioconjugafe Chem., Vol. 6,No. 5, 1995
1985; Pandey et al., 1986). Bovine adrenal cortex membranes (150 pg) were incubated with N 4a-N3Bz-ANF(428) (300000 cpndincubation, approximately 0.7 nM) in 100 pL of assay buffer at 0 "C for 1 h to allow binding. Membranes were collected by centrifugation and resuspended in the same buffer. Photolysis was performed using a 2.3-W Spectroline ultraviolet lamp (Westbury, NY)a t 258 nm a t a distance of 10 cm for 5 min. Affinity Cross-Linking of ANF Receptor. Bovine adrenal membranes (150 pg) were incubated with [12511ANF(4-28) (300000 cpdincubation, 0.7 nM) in 100 pL of assay buffer a t 0 "C for 1 h to allow binding. The membranes were collected by centrifugation and resuspended in 20 mM sodium phosphate buffer, pH 7.5 containing 0.15 M NaC1. Disuccinimidyl suberate in 10 pL of dimethyl sulfoxide, or bis(sulfosuccinimidy1)suberate in 10 pL of water, was added to a final concentration of 0.1,0.5, or 1mM. Cross-linking was allowed to proceed a t room temperature for 1 h. SDS-PAGE and Autoradiography. The labeled membranes were dissolved in sample buffer consisting of 50 mM Tris-HC1 buffer, pH 6.8, 2% SDS, 10 mM dithiothreitol, and 7% glycerol, 0.001% bromophenol blue was added, and the mixture was boiled for 5 min. An aliquot of the mixture (10-20 pg membrane protein) was separated by SDS-PAGE in a 7.5% polyacrylamide gel. For the separation of protein digests, electrophoresis was carried out in a 4-20% polyacrylamide gradient gel. The gels were stained with Coomassie Blue G-250, destained, and dried. Autoradiography was carried out by exposing the dried gels to Kodak X-Omat AFt film overnight at -80 "C. Cyanogen Bromide Digestion of Affinity-Labeled ANF Receptor. Bovine adrenal membranes (200 pg) were affinity labeled with N4a-IAc-[12513ANF(4-28) by the stepwise affinity alkylation reaction described above. The labeled membranes were resuspended in 100 p L of buffer A. The membrane suspension was extracted with three volumes of a CHC13-CH30H mixture (2:l v/v) with vigorous shaking followed by centrifugation. A white pellet formed a t the interface of the aqueous and CHC13 layers and was collected. The pellet was resuspended in 100 pL of water, and the extraction with the CHC13CH30H mixture was repeated. The pellet a t the interface was collected and dried under vacuum. The dried material was dissolved in 70% formic acid a t a protein concentration of approximately 2 mg/mL. BrCN was added a t a protein-to-reagent ratio of 1:l (w/ w), and the reaction was allowed to proceed a t room temperature overnight under nitrogen in the dark. The digest was then diluted 20-fold with water and lyophilized. The dried material was resuspended in 100 pL of 10 mM NH4HC03, and the lyophilization was repeated. The lyophilized material was separated by SDS-PAGE, and radioactive peptide fragments were detected by autoradiography. Endoproteinase Digestion of Affinity-Labeled ANF Receptor. The bovine adrenal membranes (200 pg) were affinity labeled with N4a-IA~-[1251]ANF(4-28) by the stepwise affinity alkylation. The membranes were then dissolved in 50 pL of 0.5% SDS and immediately heated a t 100 "C for 10 min to inactivate endogenous proteases. After cooling, the mixture was diluted 5-fold with 0.1 M NH4HC03, pH 8. Digestion with endoproteinase Glu-C was carried out a t a protein-to-enzyme ratio of 1O:l a t 23 "C overnight. Digestion with endoproteinase Lys-C was carried out in a similar manner except that the digestion was carried out in 0.1 M sodium phosphate buffer, pH 7.5, a t 37 "C. The digestion was stopped by acidification with an equal volume of 2 M
acetic acid, and the mixture was lyophilized. The lyophilized material was separated by SDS-PAGE or reversedphase HPLC, and the affinity-labeled fragments were detected by autoradiography or by counting lZ5I-radioactivity. RESULTS AND DISCUSSION
Affinity labeling of the ANF receptor was carried out with a procedure that we have termed stepwise affinity labeling. The procedure allowed specific labeling of the ANF receptor a t yields substantially greater than those obtainable by the traditional photoaffinity labeling or affinity cross-linking methods. As the affinity reagent, we used the ANF(4-28) peptide in which a moderately reactive electrophilic moiety was incorporated synthetically. The higher labeling yield was achieved by carrying out the reaction in a stepwise fashion as follows: (1)ANF (428) with the attached electrophile was allowed to bind to the receptor in a plasma membrane suspension a t 0 "C to suppress covalent reactions; (2) the membranes were collected by centrifugation, and the supernatant containing unbound ligand was removed; and (3) the membranes were resuspended, and the cross-linking reaction was effected by incubation a t room temperature. The reaction combines the advantage of each of the photoaffinity labeling and the affinity cross-linking procedures to achieve a substantial improvement in the labeling yield. ANF is a 28-residue peptide with the sequence Ser-
Leu-Arg-Arg-Ser-Ser-Cys-Phe-Gly-Gly-Arg-Ile-Asp-ArgIle-Gly-Ala-Gln-Ser-Gly-Leu-Gly-Cys-Asn-Ser-Phe-ArgTyr, in which Cys-7 and Cys-23 are disulfide-linked (Misono et al.,1984a,b). The disulfide bond, the sequence in the ring structure, and the carboxyl-terminal sequence, Asn-24 through Arg-27, are critical for biological activity. On the other hand, the amino-terminal region, Ser-1 through Ser-6, and the carboxyl-terminal Tyr-28 residue are not essential for biological activity (for review, see Bovy (1990)). Certain substitutions can be introduced in these nonessential regions without significantly affecting the biological activity or the ability of the peptide to bind to the receptor. In the present study, the electrophilic reactive group was incorporated a t the amino-terminal a-amino group of the ANF(4-28) peptide. The affinity reagents, N 4a-IAc-ANF(4-28), N 4a-BrAcANF(4-28), and N4a-(maleimidobenzoy1)-ANF(4-28), were prepared by reacting the ANF(4-28) peptide with IAc-N-hydroxysuccinimide,BrAc-N-hydroxysuccinimide, or (maleimidobenzoy1)-N-hydroxysuccinimide,respectively, a t pH 8. Because ANF(4-28) contains no amino acid side chains that readily react with N-hydroxysuccinimide esters, only the amino-terminal a-amino group was expected to be derivatized. The reaction products were purified by reversed-phase HPLC and then radioiodinated a t the Tyr-residue to obtain N 4a-IA~-[12511ANFor N4"-(maleimido(4-28), N 4a-BrA~-[1251]ANF(4-28), benzoyl)-[lZ5I]ANF(4-28). Because high affinity binding is the prerequisite for efficient receptor labeling, the effect of the aminoterminal modifications on the ANF binding was first tested as follows. N 4a-IAc-ANF(4-28), N 4a-BrAc-ANF(4-28), and N 4a-(maleimidobenzoyl)-ANF(4-28)were reacted with the a-amino group of N'-acetyl-L-lysine to obtain unreactive adducts, which in turn were used as competing ligands in competitive binding assays against [1251]ANF( 1-28) (Figure 1). The N'-acetyl-L-lysine adducts derived from N 4a-IAc-ANF(4-28),N 4a-BrAc-ANF(4-28), and N 4a-(maleimidobenzoyl)-ANF(4-28)were able to compete against [ 1251]ANF(1-28)with efficiencies nearly equal to that of unmodified ANF(4-28). Scat-
Bioconjugafe Chem., Vol. 6, No. 5, 1995 545
Stepwise Affinity Labeling of Peptide Binding Site
n
1
I
a
b
c
d
a
b
c
d
e
f
g
-
205 K
116K97K66K-
45 K -
1
04
-12
-
7
-11
.
-
-10
-
-
-9
I
.
-8
-7
-6
29K-
LOG CONCENTRATION, M Figure 1. Competitive binding of amino-terminal-modified ANF derivatives against [1251]ANF(1-28). Adrenal membranes (20 pg) were incubated with 0.14 nM [12511ANF(1-28)in the absence or the presence of varying concentrations of ANF(1or the W-acetyllysine adduct prepared from N4=28) (0) (maleimidobenzoyl)-ANF(4-28) (O), N 4a-BrAc-ANF(4-28) (W), or N4"-IAc-ANF(4-28) (A)at 0 "C for 1 h.
chard analysis gave Kd values of 1.8,2.0, and 2.0 nM for the adducts obtained from N &-IAc-ANF(4-28), N 4aBrAc-ANF(4-28), and N 4a-( maleimidobenzoy1)-ANF(428), respectively. The same analysis yielded a & of 0.7 and 1.1nM for ANF(1-28) and ANF(4-28), respectively. The calculated B,, values were unchanged at about 2 pmol/mg membrane protein. These results indicated that the amino-terminal modifications did not significantly affect the binding. An initial application of the stepwise affinity labeling maleprocedure to the adrenal membranes using N 4Q-( imidobenzoy1)-[12511ANF(4-28)(29 nM) as a n affinity reagent produced a substantial amount of labeled 130kDa ANF receptor band but also produced a large amount of background labeling (Figure 2 (top and bottom), lanes f and g). This result suggested that the reaction of the maleimidobenzoyl moiety was occurring much faster than the binding of the affinity reagent to the ANF receptor. The strongest nucleophile in protein side chains is the cystein sulfhydryl group. It seems likely that the large background labeling was the result of a rapid reaction between sulfhydryl groups of the membrane proteins and the maleimidobenzoyl moiety of the affinity reagent. Consistent with this speculation, the nonspecific labeling was nearly completely eliminated when sulfhydryl groups were blocked by pretreating the membranes with 5 mM PCMBS (Figure 2 (top), lanes a-e) or 5 mM N-ethylmaleimide (Figure 2 (bottom), lanes a-e). Moreover, the labeling of the 130-kDa ANF receptor was greater than the labeling obtained with untreated membranes, presumably because more of the reagent remained available to bind to the receptor. The labeling of the 130-kDa protein was completely abolished by inclusion of 1pM ANF(4-28) (Figure 2 (top and bottom), lane e), demonstrating the labeling specificity. In the pretreatment of the adrenal membranes with PCMBS, a 10-min incubation was sufficient to obtain maximum specific labeling of the ANF receptor. With N-ethylmaleimide, the maximum effect was observed after a 60-min pretreatment. In either case, blocking the sulfhydryl group before addition of the affinity reagent effectively reduced nonspecific labeling reactions. Hence, all the subsequent affinity labeling experiments were carried out with membranes pretreated with 5 mM PCMBS for 10 min. These results also indicate that sulfhydryl groups are not involved in ANF binding to the receptor.
e
f
g
w 205 K
a
w
116 K 97K66K-
Figure 2. Effect of sulfhydryl group blockade with PCMBS or N-ethylmaleimide on the specificity of stepwise affinity labeling of the ANF receptor in bovine adrenal cortex plasma membranes by N 4a-(maleimidobenzoyl)-[12511ANF(4-28).(Top) the adrenal membranes (150 pg) were incubated with 5 mM PCMBS for 0, 30,60, or 90 min (lanes a, b, c, and d, respectively) before the initial binding step in the stepwise affinity labeling procedure. Lane e shows the control experiment where the labeling reaction was carried out in the presence of 1pM ANF(4-28) with the membranes treated with 5 mM PCMBS for 90 min. Lanes f and g show results of the labeling reaction performed with untreated membranes in the absence (lane f) or the presence of 1pM ANF(4-28) (lane g). (Bottom) The same set of experiments as above were carried out except using N-ethylmaleimide.
The time course of the labeling reaction with N&-BrAc[1251]ANF(4-28)(specific radioactivity at 0.43 mCi/nmol)
is shown in Figure 3. As the first step, the adrenal membranes were incubated with N4"-BrAc-[1251JANF(428) at 0 "C for 1h and then washed to remove unbound reagent. The covalent labeling reaction was initiated by raising the temperature of the membrane suspension to 23 "C. At time intervals, aliquots were taken, and the reaction was stopped by addition of 10 mM dithiothreitol. The incorporation of the 1251-labeledligand into the 130kDa protein increased with time, reaching the maximal labeling in 2-3 h. The course of the reaction of N 4a-IA~[ 1251]ANF(4-28)was similar to that of N4a-BrAc-[12511ANF(4-28). The reaction of N 4a-(maleimidoben~oyl)[12513ANF(4-28)was slightly slower, reaching a maximum in 4-6 h (data not shown). The first step of the incubation of the membranes with the affinity reagent at 0 "C for 1h followed by washing to remove unbound reagent was essential in minimizing nonspecific labeling. Direct reactions with the affinity reagents at 23 "C resulted in increased levels of nonspecific labeling. Figure 4 shows the results of labeling reactions carried out using different concentrations of N 4a-IAc-[12511ANF-
He et al.
546 Bioconjugate Chem., Vol. 6, No. 5, 1995
1
14
2
3
4
5
6
130 K -
0
2
I
I
I
I
4
6
8
10
REACTIONTIME (h)
Figure 3. Time-course of ANF receptor labeling during the stepwise affinity labeling using N4a-BrAc-[12513ANF(4-28). Adrenal membranes (150 pg) pretreated with 5 mM PCMBS for 10 min were incubated with 28 nM N4a-BrAc-[12511ANF(428) a t 0 "C for 1h to allow binding. After unbound reagent was removed, the covalent labeling reaction was allowed to proceed a t 23 "C. The 1251-radioactivity incorporated into the 130-kDa band in SDS-PAGE was plotted against the incubation time. A a
b
B c
d
a
b
C c
d
a
b
c
d
130K-
Figure 5. Comparison of the efficiency of ANF receptor labeling by the stepwise affinity labeling, photoaffinity labeling, and direct affinity cross-linking methods. Top: arena1 membranes (150pg) pretreated with 5 mM PCMBS for 10 min were affinity-labeled using the same concentrations (0.7 nM) of the affinity reagent, N4a-IAc-[12511ANF(4-28)(lanes 1and 2), N 4nNsB~-[~~~1lANF(4-28) (lanes 3 and 41, and [12511ANF(4-28) (lanes 5 and 6) at 0 "C for 1 h in the presence (lanes 1 , 3 , and 5) or absence of 1 pM ANF(4-28) (lanes 2, 4, and 6). The reaction conditions are described in the text. Bottom: a n autoradiogram of the same dried gel developed after longer period of exposure to a n X-ray film.
Figure 4. Stepwise affinity labeling of the ANF receptor in the bovine adrenal membranes using varying concentrations of N 4a-IAc-[1251 lANF(4-281, N 4a-BrAc-[lZ5I]ANF(4-28), and Nk(maleimidobenzoyl)-[12sIlANF(4-28).Adrenal membranes (150 pg) pretreated with 5 mM PCMBS for 10 min were used. (A) The stepwise labeling reactions were carried out with 10, 30, 45 nM N4a-IAc-[12513ANF(4-28) (lanes a, b, and c, respectively). The control experiment (lane d) was carried out with 45 nM N 4a-IAc-[1251]ANF(4-28)in the presence of 1pM ANF(4-28). (B) The labeling reactions with 10, 30, and 45 nM N&-BrAc[1251JANF(4-28)(lanes a, b and c, respectively). The control experiment (lane d) was carried out with 45 nM Nk-BrA~-[*2511ANF(4-28) in the presence of 1pM ANF(4-28). (C) The labeling reactions with 10,30, and 45 nM N4a-(maleimidobenzoy1)-[12511ANF(4-28) (lanes a, b, and c, respectively). The control experiment (lane d) was carried out with 45 nM N4"-(maleimidobenzoyl)-[12511ANF(4-28) in the presence of 1pM ANF(4-28).
(4-28), N4Q-BrAc-[12511ANF(4-28), or Nk-(maleimidobenzoyl)-[1251]ANF(4-28). The reagents with specific radioactivities at 0.43 mCi/nmol were used. The intensity of the labeled bands increased slightly with the concentrations of the reagents. The extent of labeling appears to depend mostly on the initial binding of the reagent to the receptor site at 0 "C.To estimate the yield of receptor labeling, sections of polyacrylamide gel containing the labeled 130-kDa band were excised, and the 1251-radioactivitywas counted. On the basis of the theoretical specific activity of the reagents (0.43 mCi/ nmol at the reference time), the amounts of reagent incorporated into the 130-kDa receptor band were estimated to be 1.4,1.04, and 0.42 fmob'pg of protein for the membranes that were labeled using 45 nM concentrations of N&-IAc-[12511ANF(4-28),N4"-BrAc-[1251JANF(4-
28), and N 4a-(maleimidobenzoyl)-[1251]ANF(4-28), respectively. On the basis of the B,, of approximately 2 pmob'mg membrane protein as determined by the binding assay, the yields of receptor labeling were estimated to be 70%,51%, and 22%, respectively. The yields may be improved further by using saturating concentrations of the affinity reagent during the initial binding step. The efficiency of labeling by the stepwise method using N4a-IAc-[12513ANF(4-28) was compared against those efficiencies obtained by photoaffinity labeling using NJBz-[1251JANF(4-28)and by direct affinity cross-linking using [12513ANF(4-28)and a cross-linker, disuccinimidyl suberate (Figure 5). To permit direct comparison, all the affinity ligands were radioiodinated using the same lot of carrier-free [1251]NaI,and monoiodinated derivatives were used at the same concentrations (300000 cpml incubation, at the final concentration of 0.7 nM) during the initial binding step. The stepwise affinity labeling gave a substantially greater labeling yield (lanes 1and 2) than that obtained by photoaffinity labeling (lanes 3 and 4) or direct affinity cross-linking (lanes 5 and 6). On the basis of the amount of 1251-radioactivityincorporated into the 130-kDa bands, the ratio of the labeling yields by the stepwise method, photoaffinity labeling, and direct cross-linking was 63:2.3: 1.0. In the direct cross-linking reaction, varying the cross-linker concentration (0.1,0.5, and 1.0 mM) or the use of a water-soluble cross-linker, bis(sulfosuccinimidyl)suberate, did not improve the extent of cross-linking.
Stepwise Affinity Labeling of Peptide Binding Site 1 2
40 K
-
3 4 5 6
6 W
0
- 29.4 K - 18.0 K
Figure 6. Separation of the affinity-labeledpeptide fragment by SDS-PAGE using a 4- 20%gradient gel. The bovine adrenal membranes were affinity-labeled using N&-IAc-[ 12513ANF(428) in the presence (lanes 1,3, and 5 ) or absence of 1 pM ANF(4-28) (lanes 2,4, and 6). The membrane protein was extracted and digested with BrCN (lanes 1 and 21, endoproteinase Glu-C (lanes 3 and 41, or endoproteinase Lys-C (lanes 5 and 6). The labeled peptides were detected by autoradiography.
To examine the chemical selectivity of the reaction for the amino acid side chains, the protein fraction extracted from the affinity-labeled adrenal membranes was analyzed by radiochemical peptide-mapping (Figure 6). The adrenal membranes were affinity labeled using N &-IAc[1251JA"(4-28) by the stepwise method. The protein extracted from the labeled membranes was digested with BrCN, endoproteinase Glu-C, or endoproteinase Lys-C and separated by SDS-PAGE, and peptide fragments labeled with 1251were detected by autoradiography. BrCN-digestion of the labeled membrane protein yielded a single labeled fragment with an apparent molecular mass of 40 kDa (lane 2). This band was absent in the control experiment, in which the stepwise affinity labeling was carried out in the presence of 1pM unmodified ANF(4-28) as a competing ligand (lane l), demonstrating the specificity of labeling. The results also indicate that the linkage between the affinity label and the receptor polypeptide was stable under the conditions of the BrCN cleavage reaction. The digestion with endoproteinase Glu-C was carried out in 0.1 M NH4HC03, in which proteolysis is expected to occur only at glutamyl bonds (Houmard and Drapeau, 1972). Because ANF(4-28) does not contain the Glu residue, the digestion does not remove 1251-labelthat is incorporated a t Tyr-28 in the affinity reagent. SDSPAGE of the digest gave a single labeled band with an apparent mass of 18 kDa (Figure 6, lane 4). This band was again absent in the control experiment (lane 3). Similarly, digestion with endoproteinase Lys-C gave a specific band with an apparent mass of 29 kDa (lanes 5 and 6). Reversed-phase HPLC separation of the endoproteinase Lys-C digest also gave a single major 1251-radioactivity peak (data not shown). This peak was eluted with a retention time (37 min) substantially longer than that of unreacted Nk-IAc-[ 12513ANF(4-28)(32 min), suggesting that the peak may contain a peptide fragment crosslinked with the affinity ligand. Similar results were obtained with the endoproteinase Glu-C digest, in which a single 1251-radioactivitypeak appeared a t 38 min. The generation of only one major labeled fragment in each digestion mixture suggests that the covalent labeling of the ANF receptor by the affinity reagent occurs a t a single site or at a limited number of sites. These results clearly indicate considerable chemical selectivity in the reaction with the receptor side chains. Evidently, the
Bimnjugaate Chem., Vol. 6, No. 5, 1995 547
specific labeling of the ANF receptor depends on the specific binding of the ANF peptide to the receptor binding site. The reactive moiety attached to the ANF peptide would then react with the receptor protein residues that are within its geometric reach. Residues that may react with the haloacetyl and maleimide reagents include Cys, His, Met, Lys, Tyr, Ser, Thr, Asp, and Glu (Wilchek and Givol, 1977). Among these residues, the Cys-SH residue is the most reactive. The bifunctional reagents utilized in this study, IAc-Whydroxysuccinimide, BrAc-N-hydroxysuccinimide,or (maleimidobenzoy1)-N-hydroxysuccinimide, have previously been used to cross-link proteins and peptides through a sulfhydryl group and an amino group (for a review, see Wong (1991)). The cross-linking reactions have been utilized, for example, to couple antigen peptides to carrier proteins (Kitagawa and Aikawa, 1976) or toxins to antibodies (Thorpe et al., 1984). However, because sulfhydryl groups react too rapidly with the IAc, BrAc, or maleimido group of the affinity reagents, it was necessary to block all accessible sulfhydryl groups in the membranes prior to the affinity labeling steps. Among the side chain residues that are within the span of the reagent arm, a residue with the strongest nucleophilicity and with the optimal orientation would react most preferentially, thereby conferring a significant degree of chemical specificity to the reaction. The reaction of the electrophile, either the haloacetate or the maleimide, with any of those amino acid side chains is expected to yield a cross-linked structure of a form that has been well characterized. Identification of the affinitylabeled residue can also be facilitated by incorporation of a radioisotope, such as 14C and 3H, in the iodoacetyl or maleimide moiety. It is necessary to note that the a-hydrogen of the carboxylmethylene group is an acidic hydrogen and undergoes slow exchange with solvent proton. Photoaffinity labeling of the ANF receptor using a p-benzoylphenylalanine-containing ANF peptide analog has been reported to provide a high yield of receptor labeling (McNicoll et al., 1992). However, the radical nature of the photoactivated reactant species may lead to cross-linking at multiple sites and yield a mixture of cross-linked structures that are of unknown forms (Dorman and Prestwich, 1994). The increase in the yield of affinity labeling of the ANF receptor obtained here by the stepwise affinity labeling method represents a major improvement over the commonly used photoaffinity labeling and affinity crosslinking procedures. The high labeling yield, chemical selectivity of the reaction, and stability of the linkage are the essential requirements for a chemical probe that can be used to determine peptide receptor binding site structures. The stepwise procedure described here appears to satisfy all these requirements. Determination of the binding site sequence would be greatly facilitated by the availability of purified receptor protein. However, purification of cell membrane receptors, including the ANF receptor, generally requires a lengthy procedure and provides limited amounts of purified protein. Yet, certain membrane receptors cannot be solubilized in an active form and, hence, cannot be purified. These difficulties have often precluded structural characterization of the peptide receptor binding site. The present approach that provides high-yield specific labeling of the ANF receptor may allow a binding site sequence to be determined directly from a plasma membrane preparation after labeling and appropriate digestion of the membrane protein. Purification of an affinity-labeled binding site peptide from a complex mixture of protein digest may
He et al.
548 Bioconjugafe Chem., Vol. 6,No. 5, 1995
be greatly facilitated by anti-ANF antibody affinity chromatography. The stepwise affinity labeling method for ANF receptors described in the present report may also be adaptable to certain other peptide hormone receptors. ACKNOWLEDGMENT
Houmard J., and Drapeau G. R. (1972) Staphylococcal protease: a proteolytic enzyme specific for glutamoyl bonds. Proc. Nut. Acad. Sci. U.S.A. 69, 3506-3509. Kitagawa, T., and Aikawa, T. (1976) Enzyme coupled immunoassay of insulin using J . Biochem. - a novel coupling - reagent. 79, 233-236. Kuno, T., Andresen, J. W., Kamisaki, Y., Waldmen, S. A., Chang, L. Y.. Saheki. S.. Leitman. D. C.. Nakane. M.. and Murad. F. (1986) Co-purification of a n atrial natriuretic' factor receptor and particulate guanylate cyclase from rat lung. J . Biol. Chem. 261, 5817-5823. McNicoll, M., Escher, E., Wilkes, B. C., Schiller, P. W., Ong, H., and De Lean, A. (1992) Highly efficient photoaffinity labeling of the hormone binding domain of atrial natriuretic factor receptor. Biochemistry 31, 4487-4493. Meloche, S., McNicoll, N., Liu, B., Ong, H., and De Lean, A. (1988) Atrial natriuretic factor R1 receptor from bovine adrenal zona glomerulosa: Purification characterization and modulation by amiloride. Biochemistry 27, 8151-8158. Misono, K. S., Grammer, R. T., Fukumi, H., and Inagami, T. (1984a) Rat atrial natriuretic factor: Isolation, structure and biological activities of four major peptides. Biochem. Biophys. Res. Commun. 123, 444-451. Misono, K. S., Fukumi, H., Grammer, R. T., and Inagami, T. (1984b) Rat atrial natriuretic factor: Complete amino acid sequence and disulfide linkage essential for biological activity. Biochem. Biophys. Res. Commun. 219, 524-529. Misono, K. S., Grammer, R. T., Rigby, J. W., and Inagami, T. (1985) Photoafinity labeling of atrial natriuretic factor receptor in bovine and rat adrenal cortical membranes. Biochem. Biophys. Res. Commun. 130, 994-1001. Pandey, K. N. Inagami, T., and Misono, K. S. (1986) Atrial natriuretic factor receptor on cultured Leydig tumor cells: ligand binding and photoaffinity labeling. Biochemistry 25, 8467-8472. Pilch P. F., and Czech M. P. (1983) Affinity cross-linking of peptide hormones and their receptors. Recept. Biochem. Methodol. 1 , 161-175. Ruoho, A. E., Rashidbaigi, A., and Roeder, P. E. (1983) Approaches to the identification of receptors utilizing photoaffinity labeling. Recept. Biochem. Methodol. 1 , 119-160. Sugiyama, M., Fukumi, H., Grammer, R. T., Misono, K. S., Yabe, Y., Morisawa, Y., and Inagami, T. (1984) Synthesis of atrial natriuretic peptides and studies of structural factors in tissue specificity. Biochem. Biophys. Res. Commun. 123, 338-344. Takayanagi, R., Inagami, T., Snajdar, R. M., Imada, T., Tamura, M., and Misono, K. S. (1987) Two distinct forms of receptors for atrial natriuretic factor in bovine adrenocortical cells: purification ligand binding and peptide mapping. J . Biol. Chem. 262, 12104-12113. Thorpe, P. E., Ross, W. C. J., Brown, A. N. F., Myers, C. D., Cumber, A. J., Foxwell, B.M. J., and Forrester, J. T. (1984) Blockade of the galactose-binding sites of ricin by its linkage to antibody specific cytotoxic effects of the conjugates. Eur. J . Biochem. 140, 63-67. Wilchek M., and Givol D. (1977) Haloacetyl derivatives. Methods Enzymol. 46, 153-157. Wong S. S. (1991) Chemistry ofprotein conjugation and crosslinking, CRC Press, Boca Raton, FL. I
This work was supported by grants HL37399 and HL33713 from the National Institute of Health. We thank Cassandra Talerico for editorial assistance and Robin Lewis for assistance in preparation of the manuscript. Supporting InformationAvailable: HPLC chromatograms showing preparation and purification of N 4a-IAcANF(4-281, N 4a-BrAc-ANF(4-28), N 4a-(maleimidobenzoyl)-ANF(4-28), N 4a-IAc-[12511ANF(4-28),N 4a-BrAc[12511ANF(4-28),and N4U-(maleimidobenzoyl)-[12511ANF(428) and HPLC peptide-mapping of the endoproteinase Lys-C digest of adrenal membranes that were affinitylabeled with I A C - [ ~ ~ ~ I ] A N Fby( ~the - ~stepwise ~) method (9 pages). Ordering information is given on any current masthead page. LITERATURE CITED Bow, P. R. (1990) Structure Activity in the atrial natriuretic peptide family. Med. Res. Rev. 10, 115-142. Cantin, M., and Genest, J. (1986) The heart as a n endocrine organ. Clin. Invest. Med. 9, 319-327. Chinkers M., and Garbers, D. L. (1989) The protein kinase domain of the ANP receptor is required for signaling. Science 245, 1392-1394. Chinkers, M., Garbers, D. L., Chang, M. S., Lowe, D. G., Chin, H., Goeddel, D. V., and Schulz, S. (1989) A membrane form of guanylate cyclase is a n atrial natriuretic peptide receptor. Nature 338, 78-83. Currie, M. G, Geller, D. M., Cole, B. R., Boylan, J. G., Yu Sheng, W., Holmberg, S. W., and Needleman, P. (1983) Bioactive cardiac substances: Potent vasorelaxant activity in mammalian atria. Science 221, 71-73. de Bold, A. J.,Borenstein, H. B., Veress, A. T., and Sonnenberg, H. (1981) A rapid and potent natriuretic response to intravenous injection of atrial myocardial extract in rats. Life Sci. 28, 89-94. Dorman, G., and Prestwich, G. D. (1994) Benzophenone photophores in biochemistry. Biochemistry 33, 5661-5673. Gerzer, R, Heim, J.-M., Schutte, B., and Weil, J. (1987) Cellular mechanism of action of atrial natriuretic factor. Klin. Wochenschr. 65, 109-114. Glossmann, H., Baukal, A. J., and Catt, K. J. (1974) Properties of angiotensin I1 receptors in the bovine and rat adrenal cortex. J . Biol. Chem. 825-834. Grammer, R. T., Fukumi, H., Inagami, T., and Misono, K. S. (1983) Rat atrial natriuretic factor: purification and vasorelaxant activity. Biochem. Biophys. Res. Commun. 116, 696703.
BC9500453
/
