The structure and organization of dopamine-β-hydroxylase in the chromaffin granule membrane

33

Biochimica et Biophysica Acta, 669 (1981) 33-381

Elsevier/North-HollandBiomedicalPress BBA 38681 THE STRUCTURE AND ORGANIZATION OF DOPAMINE-~-HYDROXYLASE IN THE CHROMAFFIN GRANULE MEMBRANE

PETER BLAKEBOROUGH*, CHARLES F. LOUIS ** and ANTHONYJ. TURNER *** Department of Biochemistry, University of Leeds, 9 Hyde Terrace, Leeds LS2 9LS (U.K.J

(Received November 3rd, 1980)

Key words: Dopamine.#-hydroxylase; Chromaffin granule,"Protein cross-linking; Catecholamine; (Bovine adrenal glandJ

Chromaffin granules have been purified from bovine adrenal medullae. The granule membranes have been crosslinked with the disulphide-bridged bifunctional imido ester, dimethyl-3,3'-dithiobissuccinimidylpropionate hydrochloride. Analysis of the cross-linked proteins by electrophuresis on agarose/acrylamide gels revealed components of M r 300 000 and 150 000. Further analysis of samples by eleetrophoresis in a second dimension containing a reducing agent revealed the monomeric species from which the cross-linked polypeptides were formed. The major component in the second dimension exhibited a molecular weight of approx. 80 000 and could be identified with dopamine-/~-hydroxylase (3,4-dihydroxyphenylethylamineascorhate:oxygen oxidoreduetase (/3-hydroxylating), EC 1.14.17.1). It is proposed that dopamine-~-hydroxylase in the intact granule membrane is arranged as a tetramer consisting of two disulphide-bridged dimers of the 80 000 subunit in close apposition. This structural arrangement of the membrane-bound form of dopamine-fl-hydroxylase is identical with that previously proposed for the soluble, intra-granular form of the enzyme.

Introduction

Dopamine-/3-hydroxylase (3,4-dihydroxyphenylethylamine,ascorbate:oxygen oxidoreductase (fl-hydroxylating), EC 1.14.17.1) catalyses the conversion of dopamine to noradrenaline in the biosynthetic pathway to adrenaline [1]. In the adrenal medulla this enzyme appears to be exclusively associated with the chromaffin granule fraction [2] where it occurs as the major membrane protein of this organelle. It also exists as a minor component of the intra-granular soluble protein [3]. Both the soluble and the membrane-bound forms of dopamine-/3-hydroxylase are * Present address: National Institute for Reseaxchin Dairying, Univeristy of Reading, Shinfield, Reading RG2 9AT, U.K. ** Department of Veteiinary Biology, University of Minnesota, St. Paul, MN 55108, U.S.A. *** To whom correspondence should be addressed. Abbreviation: DSP, dimethyl-3,3'-dithiobissuccinimidylpropionate hydrochloride.

glycoproteins and have been purified from bovine adrenal chromaffin granules by affinity chromatography on concanavalin A-Sepharose 4B [4,5]. The two forms of the enzyme appear identical with respect to electrophoretic mobilities, amino acid compositions and immunological reactivities [3,6]. The soluble form of the enzyme has been reported to exist as a tetramer of Mr 290000, containing four copper atoms [7]. The tetramer appears to be composed of two similar types of polypeptide chain that differ only in their amino-terminal regions [6,8]. Lactoperoxidase-coupled iodination of bovine chromaffin granules has revealed that the membranebound form of dopamine-/3-hydroxylase can be iodinated from both membrane surfaces, suggesting that it exists as a trans-membrane protein [9]. The subunit molecular weight of this enzyme has been estimated to be 80 000 by gel electrophoresis in the presence of SDS [9]. However, the oligomeric structure of the native enzyme in the membrane has not been ascertained. In the present study the structure and organi-

0005-2795/81/0000-000/$02.50 © Elsevier/North-HollandBiomedicalPress

34 zation of dopamine-/~-hydroxylase within the chromaffm granule membrane has been investigated using the cleavable, disulphide-bridged cross-linking compound dimethyl-3,3'-dithiobissuccinimidylpropionate hydrochloride (DSP). This bifunctional reagent reacts with free amino groups of proteins at neutral pH and is able to cross-link membrane polypeptides the reactive groups of which are within 1.1 nm of each other [10]. DSP has previously been used to cross-link proteins present in the erythrocyte plasma membrane [10,11] and sarcoplasmic reticulum membranes [12], where specific protein interactions could be demonstrated. Dopamine-~-hydroxylase is here shown to be cross-linked by DSP to a tetrameric structure of approx. Mr 300 000, which appeared to be composed of similar subunits. It is suggested that the native enzyme may exist as a tetramer within the granule membrane.

Experimental Materials Both acrylamide and methylene bis-acrylamide were recrystallised from the minimum volume of boiling acetone. N,N,N',N'-Tetramethylene-ethylenediamine (TEMED) was redistilled under reduced pressure. SDS was specially pure from B.D.H. Chemicals Ltd., Poole, Dorset, U.K. Agarose (Seakem Brand) was obtained from Marine Colloids Inc., Springfield, NJ, U.S.A. and DSP was obtained from Pierce Chemical Co., Rockford, IL, U.S.A. Water used in all solutions was double-distilled and all other chemicals, where possible, were of analytical grade. Methods Preparation of chromaffin granule membranes. Crude chromaffm granules were prepared from bovine adrenal glands and then further purified by centrifugation through a discontinuous sucrose gradient as described previously [9]. A membrane fraction was prepared from these purified granules by repeated hypotonic lysis and centrifugation as described by Phillips [13]. Cross-linking of the proteins of granule membranes. Granule membranes were suspended in 20 mM Hepes buffer, pH 7.2, at a protein concentration of 4 mg/ml. DSP was dissolved in dimethyl sulphoxide to a concentration of 10 mg/ml. Aliquots of

the membrane fraction (1 mg protein) were incubated with DSP (20 /A) for 20 min at room temperature. Control incubations were conducted using dimethyl sulphoxide (20 gl) in the absence of DSP. The reaction was terminated by the addition of 50/A 1 M Tris buffer, pH 8.5, containing 1.6 M iodoacetamide and 1% (w/v) SDS. In experiments where the reduction of the disulphide bonds present in the cross-linked proteins was required, dithiothreitol (5 mg) was also added at this stage. The samples were dialyzed overnight against 0.1% (w/v) SDS at 4°C. The samples were then analysed within 24 h by agarose/acrylamide gel electrophoresis or by a two-dimensional gel system as described below. SDS acrylamide gel electrophoresis. This was carried out in a conventional gel electrophoresis unit supplied by Hoefer Scientific Instruments, San Francisco, CA, U.S.A. Analysis of proteins on 10% polyacrylamide gels was carried out according to the method of Ugel et al. [14]. For analysis of proteins in two dimensions after reaction with DSP, the procedure described by Louis et al. [12] was adopted. Electrophoresis in the first dimension was performed on an agarose/acrylarnide disc gel containing 0.5% agarose (w/v), 2.2% acrylamide (w/v) and 0.1 M sodium phosphate buffer, pH 7. The gels were inverted in their glass tubes after polymerisation in order to obtain a fiat meniscus and held in their tubes during electrophoresis by fine nylon gauze. The electrophoresis buffer was 0.1 M sodium phosphate, pH 7, containing 0.1% SDS (w/v). Bromphenol blue was used as tracking dye. The second dimension was constructed in a slab gel electrophoresis apparatus (Model A-C 4-10, E.C. Aparatus Corp., PA, U.S.A.). The system used was similar to that described by Ugei et al. [14], which had previously allowed the identification of dopamine~-hydroxylase from the granule membrane [9]. The resolving gel, which was composed of 10% acrylamide (w/v) and 11 mM sodium phosphate buffer (pH 7), was supported on a 1% agarose (w/v) plug in the base of the apparatus. The stacking gel consisted of 1% acrylamide (w/v), 0.04% potassium persulphate (w/v), 16% sucrose (w/v), 0.025% TEMED, 5 mM sodium phosphate adjusted to pH 7. This was overlaid with 1% agarose (w/v), 10% fl-mercaptoethanol (v/v), 0.1% SDS (w/v), 5.5 mM sodium phosphate (pH 7). The /3-mercaptoethanol served to

35 reduce disulphide bridges in the cross-linked proteins before stacking and resolution. The first dimension gel (agarose/acrylamide) was placed on top of the reducing layer and sealed with the agarose mixture containing #-mercaptoethanol and 5.5 mM phosphate buffer (pH 7). The cathode buffer solution was composed of 50 mM Tris/50 mM glycine/0.1% (w/v) SDS, pH 9.45. The anode buffer was 60 mM Tris/50 mM HCI, pH 7. Electrophoresis was conducted at 150 V for 4 - 5 h. All gels were stained for protein using Coomassie Blue as described by Fairbanks et al. [15], and were photographed with an orange filter. Molecular weight calibration curves. Molecular

weights of polypeptide species resolved by the agarose/acrylamide gel system were determined from the calibration curve shown in Fig. 1. A sample of bovine serum albumin, treated as described above for cross-linking the proteins of granule membranes, was used to provide a series of markers. The migration distances of the markers on SDS polyacrylamide gels were linearly related to log M r in the range 65 0 0 0 325 000. For the gel system of Ugel et al. [14], the molecular weights of different polypeptide components were determined using calibration proteins of known molecular weight [16]. Protein determination. Protein concentrations were determined by the method of Lowry et al. [17] using bovine serum albumin as standard. Results and D i s c u s s i o n

1

0

I

I

0-5

1"0

Fig. 1. Calibration of SDS agarose/acrylamide gels for the determination of molecular weights of proteins. Plot of the logarithm of the molecular weights of cross-linked oligomcrs of bovine serum albumin versus their relative mobility in agarose/acrylamide gels. Bovine serum albumin was crosslinked with DSP as described in Methods, solubilised in 1% (w]v) SDS and then carboxyamidomethylated. Samples (50 xtg protein) were analysed on agarose/acrylamide gels as descried by Louis et al. [12]. The gels were stained with Coomassie Blue [15] and scanned photometrically at 540 nm using a Unicam SP1800 ultraviolet spectrophotometer adapted for scanning polyacrylamide gels. Results are the mean of three experiments.

The conditions used in this study for the crosslinking of membrane proteins were similar to those reported by Louis et al. [12] for the reaction of DSP with the sarcoplasmic reticulum membrane isolated from skeletal muscle. No significant differences in cross-linking were obtained by increasing either the concentration of DSP or the reaction time in the conditions described above. Fig. 2 shows the analysis of protein components of the granule membrane on agarose/acrylamide gels both in the presence and absence of DSP. When both DSP and reducing agents were absent from the sample preparations (Fig. 2a), relatively few components appeared on the gels. The molecular weight of the major band was approx. 150 000 and there was a relatively minor species of Mr 300 000. In addition there was a third component of apparently low molecular weight that migrated close to the region of the dyefront. When samples of granule membranes were treated with a reducing agent (dithiothreitol) before electrophoresis (Fig. 2c), a different pattern of polypeptide species emerged. The component of M r 150 000 had diminished greatly in staining intensity and that of 300000 was now undetectable. The major polypeptide component migrated with an apparent molecular weight of approx. 80 000. When the granule membrane preparation was treated with the cross-linking reagent (Fig. 2b), the pattern of polypeptide components was similar to that seen in Fig. 2a. The major difference was that the component

36

1I

~ ~i~~iii~!ii~ilI ~!!i~i~~

300 150 80

L

"11"

(o.) ( b ) ) Fig. 2. Analysis of cross-linked proteins of the chromaffin granule membrane by electrophoresis on agarose/acrylamide gels. Membrane ghosts were prepared from purified chromaff'm granules. Samples (1 mg protein) were reacted with various concentrations of DSP as described in the Methods section; (a), untreated membranes; (b), 0.2 mz DSP; (c), untreated membranes reduced with dithiothreitol. Protein samples (50 /~g) were analysed on agarose/acrylamide gels [11], stained with Coomassie Blue [15] and photographed. The numbers on the figure refer to the molecular weights (×10 -a) of the resolved protein components.

of Mr 300 000 had markedly increased in staining intensity and that of 150000 had, correspondingly, decreased. No oligomers of molecular weight greater than 300 000 were observed. In order to investigate further the relationship between these various species, they were analysed by gel electrophoresis in two dimensions. When electrophoresis of granule membrane proteins was carried out in the absence of cross-linking reagent (Fig. 3), the component that had migrated near the dye-front in the agarose/acrylamide electrophoresis system (Fig. 2a) was resolved into several components by the second-dimension gel containing/3-mercaptoethanol.

Fig. 3. Analysis of proteins of chromaffin granule membranes by two-dimensional SDS gel electrophoresis. Granule membranes were prepared from purified chromaffin granules. Samples (1 mg protein) were solubilised in 1% (w/v) SDS and carboxyamidomethylated. Protein samples (75 /lg) were analysed using the two-dimensional SDS acrylamide gel system described by Louis et al. [12]. The gels were stained with Coomassie Blue [15] and photographed. I, direction of first dimension (agarose/acrylamide) gel; II, direction of second dimension (polyacrylamide) gel. The accompanying photographs of disc gels show separate electrophoresis of samples in either the first dimension (horizontal) or second dimension (vertical). The numbers on the figure refer to the molecular weights (×10 -3) of the resolved protein components.

However, the component with an apparent molecular weight of 150 000 was resolved into a polypeptide of apparent Mr 80 000. This protein exhibited the same mobility in the second dimension gel as the major polypeptide component of the granule membrane previously identified as dopamine-/~-hydroxylase [9]. We cannot exclude the possibility that the 80 000 dalton spot is composed of more than one polypeptide using the present gel system. However, Skotland et al. [8] and Aunis et al. [6] have demonstrated that the 8 0 0 0 0 dalton polypeptides of dopamine-/3hydroxylase exhibit some differences in their structure. The membrane-bound enzyme may also exhibit differences in the size of its 80 000 dalton polypeptide monomers which could account for the heterogeneity in Fig. 3. These results suggest that the species o f Mr 80 000 is present in the native membrane as a disulphide-bridged dimer which was cleaved by fl-mercaptoethanol when electrophoresis was performed in the second dimension. After reaction of granule membranes with the cross-linking reagent the components o f Mr 3 0 0 0 0 0

37

I

300 |50

1I

Fig. 4. Analysis of Cross-linked proteins of chromaffin granule membranes by two-dimensional SDS gel electrophoresis. Granule membranes were prepared from purified chromaffin granules. Samples (1 mg protein) were incubated for 20 min at room temperature with 0.2 mg of the cross-linking reagent DSP as described in Methods. The samples were then solubilised in 1% (w/v) SDS and carboxyamidomethylated. Protein samples were analysed using the two-dimensional SDS acrylamide gel system decsr~ed in the legend to Fig. 3. and 150 000 observed on the agarose/acrylamide gels appeared as a single polypeptide of Mr 80 000 in the second dimension after disulphide cleavage (Fig. 4). They therefore seem to form a tetramer and dimer, respectively, of the 80 000 dalton dopamine-fl-hydroxylase subunit. Because DSP is able to react with proteins within 1.i nm o f each other, it is suggested that two disulphide-linked dimers of dopamine-fl-hydroxylase are in close apposition in the granule membrane. The association of dimers to form tetramers apparently does not involve disulphide linkages and is to a large extent destroyed when SDS is used to solubilise the membrane components. The structure postulated here for the membranebound form of dopamine-fl-hydroxylase is identical to that previously proposed for the soluble form o f the enzyme [7]. This enzyme, therefore, appears to exist in the same oligomeric state whether soluble or membrane-associated. The labelling of chromaffin granules by lactoperoxidase-catalysed iodination has revealed that dopamine-fl-hydroxylase may exist as a trans-membrane protein, although relatively little of its structure is exposed at the external surface [9,18]. The occurrence of both amphiphilic and hydrophilic forms of dopamine-fl-hydroxylase [19] suggests that the membrane-bound form of this enzyme may have a structure similar to that seen with the proteases of

brush-border membranes [20,21]. The major part of the enzyme (the hydrophilic portion) may be located externally to the membrane but anchored to it by a hydrophobic 'foot' which traverses the membrane (see e.g., Refs. 20, 21). Such an organization for dopamine-fl-hydroxylase might explain how the two forms of the enzyme Which are apparently identical can exist either as soluble or membrane-bound components. We suggest, therefore, that dopamine-flhydroxylase exists in the chromaff'm granule membrane as a tetrameric structure, the main bulk of which is located at the inner surface o f the membrane. The portion of the enzyme which directly interacts with the membrane is presumed to be relatively small. Acknowledgements P.B. was supported by a studentship from the Science Research Council.

References 1 Levin, E.Y., Levenberg, B. and Kaufman, S. (1960) J. Biol. Chem. 235, 2080-2086 2 Laduron, P. and Belpaire, F. (1968) Biochem. Pharmacol. 17, 1127-1140 3 H6rtnagl, H., Winkler, H. and Lochs, H. (1972) Biochem. J. 129,187-195 4 Aunis, D., Boucher, M., Pescheloche, M. and Mandel, P. (1977) J. Neurochem. 29,439-447 5 Ljones, T., Skotland, T. and Flatmark, T. (1976) Eur. J. Biochem. 61,525-533 6 Aunis, D., Miras-Portugal, M.-T. and Mandel, P. (1974) Biochim. Biophys. Acta 365,259-273 7 Wallace, E.F., Krantz, M.J. and Lovenberg, W. (1973) Proc. Natl. Acad. Sci. U.S.A. 70, 2253-2255 8 Skotland, T., Ljones, T., Flatmark, T. and Sletten, K. (1977) Biochem. Biophys. Res. Commun. 74, 1483-1489 9 Blakeborough, P. and Louis, C.F. (1979) J. Neurochem. 33, 1285-1293 10 Wang, K. and Richards, F.M. (1974) J. Biol. Chem. 249, 8005-8018 11 Wang, K. and Richards, F.M. (1975) J. Biol. Chem. 250, 6622-6626 12 Louis, C.F., Saunders, M.J. and Holroyd, J.A. (1977) Biochim. Biophys. Acta 493, 78-92 13 Phillips, J.H. (1973) Biochem. J. 136,579-587 14 Ugel, A.R., Chrambach, A. and Rodbard, D. (1971) Anal. Biochem. 43,410-426 15 Fairbanks, G., Steck, T.L. and WaUach, D.F.H. (1971) Biochemistry 10, 2606-2617

38 16 Louis, C.F. and Shooter, E.M. (1972) Arch. Biochem. Biophys. 153, 641-655 17 Lowry, O.H., Rosebrough, N.J., Fazz, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193,265-275 18 Abbs, M.T. and Phillips, J.H. (1980) Biochim. Biophys. Acta 595,200-221

19 Skotland, T. and Flatmark, T. (1979) J. Neurochem. 32, 1861-1863 20 Macnair, R.D.C. and Kenny, A.J. (1979) Biochem. J. 179,379-395 21 Booth, A.G. and Kenny, AJ. (1980) Biochem. J. 187, 31-44