Cellular and Molecular Neurobiology, Vol. 11, No. 1, 1991
Comparative Studies on the Primary Structure of Acetylcholinesterases from Bovine Caudate Nucleus and Bovine Erythrocytes H. Heider, 1 P. Litynski, 1 S. Stieger, ~ and U . Brodbeck ~'~ Received October 4, 1989; accepted March 2, 1990 KEY WORDS: acetylcholinesterase; amino acid sequence; amino acid analysis; bovine brain; bovine erythrocytes.
SUMMARY 1. Comparison of partial amino acid sequences of G2-acetylcholinesterase (ACHE) from bovine erythrocytes and G4-AChE from bovine caudate nucleus revealed no differences in primary structure between the two enyzmes. The first 33 residues of the N-terminal sequences were identical. 2. In addition, the amino acid sequences of four peptides generated by tryptic and cyanogen bromide cleavage were identical for bovine erthyrocyte and brain ACHE, suggesting one identical major coding exon for the adult bovine AChE forms. Comparison of these sequences with that of fetal bovine serum AChE (Doctor et al., 1988), showed differences in residues 16, 181,212, and 216. 3. Deglycosylation studies of the two adult enzyme forms revealed that the core protein of erythrocyte AChE has an approximately 4 kDa lower molecular mass than brain ACHE. This most propably reflects differences in the C-terminal sequences of the two enzymes. INTRODUCTION Cholinesterases can be divided into two major classes: pseudocholinesterases or butyrylcholinesterases (BChE; 3 EC 3.1.1.8), with a high affinity for butyrylcho11nstitut fiir Biochemie und Molekularbioiogie, Universit~it Bern, Biihlstrasse 28, CH-3012 Bern, Switzerland. 2 To whom correspondence should be addressed. 3 Abbreviations used: ACHE, acetylcholinesterase; BChE, butyrylcholinesterase; PI, phosphatidylinositol; GPI, glycosyl-phosphatidylinositol; SDS-PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis; [3H]DFP, [3H]diisopropylfluorophosphate; PVDF, polyvinylidene difluoride. 105 0272-4340/91/0200.0105506.50/0© 1991 Plenum Publishing Corporation
106
Heider, Litynski,Stieger, and Brodbeck
line as substrate, and acetylcholinesterases (ACHE; EC 3.1.1.7), which hydrolyze primarily acetylcholine--the natural transmitter substance in brain and excitable tissues. The two enzymes may be discerned by different reactivity toward a number of inhibitors (Silver, 1974) or antibodies (Brimijoin and Rakonczay, 1986). AChE occurs either bound to cell membranes and basal lamina or as soluble enzyme and exists in multiple molecular forms, which differ in the number of catalytic subunits as well as in the structure through which they are bound to membranes (for reviews see Massouli6 and Toutant, 1988; Toutant and Massouli6, 1988). Typical representatives of membrane-bound forms of AChE are the GE-enzyme from erythrocytes and G4-AChE from brain. Erythrocyte AChE is composed of two identical catalytic subunits with an apparent molecular mass of 77 kDa which are linked by disulfide bridges. Both subunits contain a hydrophobic anchor consisting of a glycophospholipid structure (Roberts et al., 1987; for review see Brodbeck, 1986). Immunological data suggest a high homology between G2-AChE from erythrocytes and the G4 form from brain. Polyclonal antisera cross-react with all molecular forms of AChE within one species, and among the numbers of monoclonal antibodies there is only one that can distinguish between bovine erythrocyte AChE and bovine brain AChE (Rakonczay and Brimijoin, 1985) and another one that discriminates between the two human forms (Rasmussen et al., 1987). In mammalian brain AChE occurs predominantly as tetrameric globular enzyme (G4 form) with small amounts of dimers and monomers. Approximately 80% of the globular forms consists of amphiphilic AChE extractable only with detergent containing buffer (for review see Rakonczay, 1986). Tetrameric AChE seems to be anchored to the membrane through a non-catalytic subunit of 20 kDa that apparently is linked to two of the four catalytic subunits by disulfide bridges (Gennaeri et al., 1987; Inestrosa et al., 1987). The assembly of four catalytic subunits together with the attached hydrophobic anchor migrates on SDS-PAGE with an apparent molecular mass of 350 kDa, whereas the monomer has an apparent molecular mass of approximately 68 kDa. According to Bon et al. (1986) the N-terminal sequence of the catalytic subunit is A s p / A l a - S e r - P r o G l u - A s p - P r o - G l u - L e u - L e u - V a l - M e t - V a l . Presently no other sequence data, neither of the catalytic subunit nor of the anchor, are available. On the other hand, complete amino acid sequences are known for human BChE (Lockridge et al., 1987; Prody et al., 1987) and for AChE from Drosophila (Hall and Spierer, 1986), Torpedo californica (Schumacher et al., 1986) and Torpedo marmorata (Sikorav et al., 1987). Among the different forms of mammalian ACHE, approximately 90% of the sequence of fetal bovine serum AChE is currently available (Smyth et al., 1988) and five tryptic peptides of human erythrocyte AChE are sequenced (Chhajlani et al., 1989). In this paper we report on the partial sequence of membrane-bound tetrameric AChE from bovine brain and dimeric AChE from bovine erythrocytes. MATERIALS A N D METHODS All chemicals used for sequencing were from Applied Biosystems, Warrington (England). Polyvinylidene diflouride (PVDF) membranes were from
PrimaryStructureof Bovine Acetylcholinesterases
107
Millipore, Bedford (Mass.). [3H]DFP was purchased from Amersham, Buckinghamshire (England). All other chemicals w e r e either from Fluka, Buchs (Switzerland), Sigma, St. Louis (Mo.), Merck, Darmstadt (FRG), and Boehringer, Mannheim (FRG). The reagents were at least of analytical grade; chemicals for amino acid analysis were all HPLC grade and sequencer grade, respectively. Bovine blood and bovine brain were purchased from the local slaughterhouse. The detergent-soluble AChE from bovine brain was extracted from caudate nuclei and purified by affinity chromatography essentially according to the method described for human brain AChE by Sorensen et al. (1982). In order to obtain enzyme of highest purity, a..~econd:affinity chromatography step was routinely performed, yielding AChE with a specific activity of 5000 IU/mg protein at minimum. Bovine erythrocyte AChE was purified from erythrocyte membranes as detailed for the preparation of human erythrocyte AChE (Brodbeck et al., 1981). After two affinity chromatography steps, the enzyme had a specific activity of 5500 IU/mg protein.
Enzyme Assay AChE activity was measured at room temperature by the method of Ellman et al. (1961). The assay solution contained 1 mM acetylthiocholine iodide and
0.25 mM 5,5'-dithiobis(2-nitrobenzoic acid) in 100 mM phosphate buffer, pH 7.4, supplied with 0.1% Triton X-100. Enzyme activity is expressed as international units (IU) (/~mol substrate hydrolyzed/min).
Active-Site Labeling Active-site labeling was achieved by incubation with 10-SM [3H]DFP for 15 hr at room temperature. In order to remove unreacted DFP and to allow complete aging of DFP-inhibited ACHE, the enzyme was dialysed for 3 days against three changes of 10mM Tris/HCl buffer, pH 7.4, containing 144 mM NaCI.
SDS-Polyacrylamide Gel Eiectrophoresis SDS-PAGE was performed essentially according to Laemmli (1970) using 5-15% polyacrylamide gradient gels. Samples were either subjected to gel electrophoresis without previous reduction or reduced by adding 5% (v/v) mercaptoethanol. Proteins were stained with Coomassie brilliant blue R-250. For autoradiography gels containing [3H]DFP-labeled samples were immersed in Amplify, dried, and exposed to Kodak X-Omat SO 282 film at -70°C. Tricine SDS-PAGE was performed according to Schaegger and von Jagow (1988). All samples were reduced with mercaptoethanol. The slab gels (1 mm thick) consisted of three sections: a 12-cm separation gel (16% acrylamide, 0.5% bisacrylamide), a 2-cm spacer gel (9.6/0.3%), and a 2-cm stacking gel (3.8/0.12%). No urea was included in the gels. Preelectrophoresis was done at 20 mA over night. Electrophoresis was performed at 10-mA initial current. When the samples had entered the stacking gel, the constant current was adjusted to 25 mA and then the gel was run for 16-18 hr.
108
Heider, Litynski, Stieger, and Brodbeck
Eiectroblotting and Staining of the Peptides The original blotting procedure onto PVDF membranes according to Matsudeira (1987) was used. Prior to blotting the gel was immersed for 10 min in the transfer buffer (10mM Caps, pH 11.0, 20% methanol). The PVDF membrane was first wetted with 100% methanol and then equilibrated in transfer buffer. The blotting sandwich contained 2 × 6 Whatman M3 filter papers with the gel and two PVDF membranes in between. Blotting was performed in a semidry electroblotting unit (Ancos, Denmark) at 0.8 mA/cm 2 for 2 hr. Under these conditions some larger peptides may traverse the first membrane and become trapped on the second one, whereas the smaller peptides are fixed on the first membrane. After blotting the membranes were rinsed briefly with distilled water and then stained in 0.1% Coomassie brilliant blue R-250 in 50% methanol for 2 min. Destaining was performed in 50% methanol and 10% acetic acid for 10 min. Then the membranes were rinsed in distilled water, air-dried, and stored at -20°C until further processing.
Amino Acid Analysis Amino acids were analyzed essentially according to Schaller et al. (1989) Areas containing the peptides of interest were excised from the PVDF membranes and hydrolyzed in gas phase with 6 M hydrochloric acid containing 0.1% (v/v) phenol and 0.05% (v/v) 2-mercaptoethanol for 22 hr at 115°C under vacuum. The samples were then dried under vacuum and the free amino acids converted to their corresponding phenylthiocarbamoyl derivatives by adding a mixture of ethanol/triethylamine/water/phenylisothiocyanate (70:10:10:1) and incubating for 20 min. The reagents were removed under vacuum and the samples redissolved in 0.14 M ammonium acetate buffer, pH 6.4. Reversed-phase HPLC analysis of the derivatives was performed on a Nova-Pak C18 column (4/~m, 3.9 x 150mm; Waters, Milford, Mass.) in a Hewlett Packard 1090 automated amino acid analyzer system essentially according to Bidlingmeyer et al. (1984).
Enzymatic and Chemical Cleavage Procedures AChE (about 400 IU in a volume of 300 ~1) together with [3H]DFP-labeled AChE (about 1/10 of the amount of unlabeled enzyme) was dialyzed against 10mM ammoniumcarbonate buffer, pH 7.4, and precipitated with ice-cold acetone to remove excess detergent. For tryptic cleavage the pellet was redissolved in 2 M urea, 50mM Tris, pH 8.0. Trypsin (5%, w/w) which was treated with N-tosyl-L-penylalanine-chloromethyl-ketone (BoehringerMannheim) was added and the samples were incubated for 2 hr at 37°C. The cleavage mixture was then immediately placed on a Tricine gel. Cyanogen bromide cleavage was performed essentially as described by Gross (1967). AChE (200 IU) was precipitated and redissolved in 70% formic acid supplied with 7% (w/v) cyanogen bromide. The reaction mixture was incubated in the dark at room temperature for 16 hr. The reaction was stopped by adding a 10-fold excess of water. The samples Were dried under vacuum and transferred to the Tricine gel.
PrimaryStructureof Bovine Acetylcholinesterases
109
N-Chlorosuccinimide cleavage was performed essentially according to the method of Lischwe and Ochs (1982) described for peptide mapping. ACHE, completely inhibited by [3H]DFP (corresponding to 100 IU), was precipitated as above and redissolved in 50#1 of a solution of urea/water/acetic acid (1 g / l m l / l m l ) supplied with 15 mM N-chlorosuccinimide. The reaction was allowed to proceed for 2 hr at room temperature and then the samples were transferred onto the Tricine gel. N-Glycosidase F Digestion Bovine erythrocyte and bovine brain AChE (30 IU each) were dried under vacuum and redissolved in 15 #1 buffer consisting of 50 mM sodium phosphate, pH 8.6, 10 mM EDTA, 1% Triton X-100, and 0.1% SDS. For the heat-denatured enzyme samples (1 min at 100°C), 0.5% octylglycoside instead of Triton X-100 and, additionally, 10 mM mercaptoethanol were used as incubation buffer. To each sample 0.4 unit of N-glycosidase F (EC 3.2.2.18; Boehringer-Mannheim, FRG) was added and incubated for 18 hr at 37°C. Then the samples were loaded on an 8-18% polyacrylamide gradient gel.
Determination of Amino Acid Sequences Bands to be sequenced were excised from PVDF membranes and cut into pieces of approximately 2 x 8 mm. Up to six pieces were placed into the reaction cartridge of the pulsed-liquid phase Applied Biosystems sequencer Model 477A. Amino acid analysis was performed with the on-line PTH-amino acid analyzer Model 120A (Applied Biosystems) using the NORMAL-1 and OLDPRO programs supplied by the manufacturer.
RESULTS Amino Acid Analysis The amino acid composition of bovine caudate nucleus AChE was determined using a gas phase hydrolyzing system, with subsequent conversion of the amino acids to their corresponding phenylthiohydantoine derivatives (Bidlingmeyer et al., 1984). Table I shows the amino acid composition of this enzyme in comparison to previously published compositions of fetal bovine serum AChE [hydrophilic tetramer (Ralston et al., 1985)] and bovine erythrocyte AChE [GPI-anchored dimer (Grossmann and Lieflaender, 1979)]. In general the compositions of the three forms are in good accordance with the exception of the high content of isoleucine in bovine erythrocyte ACHE. This enzyme contains 3.6 mol isoleucine/100 tool, compared to 1.4 mol/100 mol and 1.5 mol/100 mol, respectively, for bovine caudate nucleus AChE and fetal bovine serum ACHE. Assuming 540 amino acids per catalytic subunit of bovine erythrocyte AChE and 580 amino acids per catalytic subunit of fetal bovine serum AChE and bovine brain ACHE, respectively, these values yield 19 isoleucines for the erythrocyte enzyme and 9 Ile residues for the other two enzyme forms. The low content of lysine residues is a common feature of all three forms. Bovine caudate nucleus AChE contains approximately 10 residues per subunit, whereas the erythrocyte as well as the serum enzyme contain even lower amounts of lysine.
U0
Heider, Litynski, Stieger, and Brodbeck Table 1. Amino Acid Composition of the Catalytic Subunits of (a) Bovine Caudate Nucleus AChE in Comparison to Published Values for the Enzyme from (b) Bovine Erythrocytes (Grossmann and Liefl~inder) and (c) Fetal Bovine Serum (Ralston et al., 1985). as Moles per 100 mol a
Asp Glu Ser Gly ~s Arg Thr Ala Do ~r Val Met I~ Leu Phe ~s ~s
a
b
c
8.4±0.3 10.4±0.5 7.8±0.5 11.1~0.2 2.2i0.3 7.8~0.6 4.0i0.3 10.0±0.2 7.5±0.5 3.6~0.3 6.7~0.3 1.1±0.5 1.4~0.1 10.7±0.1 4.8±0.1 1.7±0.1 1.2~0.1
7.6 9.2 6.2 10.3 1.8 6.4 3.7 9.8 8.7 3.1 7.6 1.4 3.6 10.0 4.4 1.5 1.4
8.2 11.0 9.1 11.9 2.6 6.2 3.6 9.9 7.4 2.1 7.6 1.5 1.5 10.1 4.6 1.4 1.3
The data in a represent the average of five amino acid determinations carded out as described under Materials and Methods. a
SDS-PAGE Patterns of the Cyanogen Bromide-Cleaved Enzymes AChEs from bovine caudate nucleus and bovine erythrocytes were treated with cyanogen bromide and the resulting peptides resolved by Tricine SDSPAGE. As shown in Fig. 1 the major peptides have molecular masses between 29 and 38 kDa. The two main peptides derived from the bovine brain enzyme had an identical N-terminal amino acid sequence, suggesting either different C-terminal
i ~a
1
2
Fig. 1. SDS-PAGE pattern (Coomassie stain) of bovine caudate nucleus (1) and bovine erythrocyte AChE (2) after cyanogen bromide clevage. EIectrophoresis was carded out according to Schaegger and von Jagow (1988) as detailed under Materials and Methods. Numbers at the fight refer to molecular weight markers (2.5-17 kDa, Sigma low molecular weight kit, 31 and 43 kDa; Bio-Rad calibration kit). Arrows pointing to bands indicate peptides from which the N-terminus was determined.
Primary Structure of Bovine Acetylcholinesterases
111
sequences of the peptides or one of the two peptides being modified. The same N-terminal sequence was also obtained with the major 38-kDa band of bovine erythrocyte ACHE. Comparison of this sequence to that of fetal bovine serum AChE (Smyth et al., 1988) revealed that it is located C-terminally from the active-site serine residue.
Cyanogen Bromide, N-Chlorosuccinimide, and Tryptic Cleavage of the [3H]DFP-Labeled AChE In order to obtain information On the sequence of the active-site serinecontaining peptide, AChE from bovine caudate nucleus and bovine erythrocytes was labeled with [3H]DFP. Cleavage was performed by treatment with cyanogen bromide, N-chlorosuccinimide, and trypsin. The resulting peptides were separated on the Tricine gel system. As shown in Fig. 2 cyanogen bromide cleavage of both enzymes resulted in two major radioactive bands at identical positions around 3.5 kDa (lanes 1 and 2). Two further bands at identical positions were obtained around 13 kDa, whereas the patterns at higher molecular masses were different. Cleavage of the two enzyme preparations with N-chlorosuccinimide resulted in a more complex band pattern (lanes 3 and 4). Peptides up to 14 kDa migrated to identical positions in the gel, whereas above this molecular mass the pattern differs. The smallest labeled peptides were found at around 3 kDa, in the -.o-- 43
ii!iiiili¸iii!iii ?iiiiii! ii! --31
.... iii!!i
Comparative Studies on the Primary Structure of Acetylcholinesterases from Bovine Caudate Nucleus and Bovine Erythrocytes H. Heider, 1 P. Litynski, 1 S. Stieger, ~ and U . Brodbeck ~'~ Received October 4, 1989; accepted March 2, 1990 KEY WORDS: acetylcholinesterase; amino acid sequence; amino acid analysis; bovine brain; bovine erythrocytes.
SUMMARY 1. Comparison of partial amino acid sequences of G2-acetylcholinesterase (ACHE) from bovine erythrocytes and G4-AChE from bovine caudate nucleus revealed no differences in primary structure between the two enyzmes. The first 33 residues of the N-terminal sequences were identical. 2. In addition, the amino acid sequences of four peptides generated by tryptic and cyanogen bromide cleavage were identical for bovine erthyrocyte and brain ACHE, suggesting one identical major coding exon for the adult bovine AChE forms. Comparison of these sequences with that of fetal bovine serum AChE (Doctor et al., 1988), showed differences in residues 16, 181,212, and 216. 3. Deglycosylation studies of the two adult enzyme forms revealed that the core protein of erythrocyte AChE has an approximately 4 kDa lower molecular mass than brain ACHE. This most propably reflects differences in the C-terminal sequences of the two enzymes. INTRODUCTION Cholinesterases can be divided into two major classes: pseudocholinesterases or butyrylcholinesterases (BChE; 3 EC 3.1.1.8), with a high affinity for butyrylcho11nstitut fiir Biochemie und Molekularbioiogie, Universit~it Bern, Biihlstrasse 28, CH-3012 Bern, Switzerland. 2 To whom correspondence should be addressed. 3 Abbreviations used: ACHE, acetylcholinesterase; BChE, butyrylcholinesterase; PI, phosphatidylinositol; GPI, glycosyl-phosphatidylinositol; SDS-PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis; [3H]DFP, [3H]diisopropylfluorophosphate; PVDF, polyvinylidene difluoride. 105 0272-4340/91/0200.0105506.50/0© 1991 Plenum Publishing Corporation
106
Heider, Litynski,Stieger, and Brodbeck
line as substrate, and acetylcholinesterases (ACHE; EC 3.1.1.7), which hydrolyze primarily acetylcholine--the natural transmitter substance in brain and excitable tissues. The two enzymes may be discerned by different reactivity toward a number of inhibitors (Silver, 1974) or antibodies (Brimijoin and Rakonczay, 1986). AChE occurs either bound to cell membranes and basal lamina or as soluble enzyme and exists in multiple molecular forms, which differ in the number of catalytic subunits as well as in the structure through which they are bound to membranes (for reviews see Massouli6 and Toutant, 1988; Toutant and Massouli6, 1988). Typical representatives of membrane-bound forms of AChE are the GE-enzyme from erythrocytes and G4-AChE from brain. Erythrocyte AChE is composed of two identical catalytic subunits with an apparent molecular mass of 77 kDa which are linked by disulfide bridges. Both subunits contain a hydrophobic anchor consisting of a glycophospholipid structure (Roberts et al., 1987; for review see Brodbeck, 1986). Immunological data suggest a high homology between G2-AChE from erythrocytes and the G4 form from brain. Polyclonal antisera cross-react with all molecular forms of AChE within one species, and among the numbers of monoclonal antibodies there is only one that can distinguish between bovine erythrocyte AChE and bovine brain AChE (Rakonczay and Brimijoin, 1985) and another one that discriminates between the two human forms (Rasmussen et al., 1987). In mammalian brain AChE occurs predominantly as tetrameric globular enzyme (G4 form) with small amounts of dimers and monomers. Approximately 80% of the globular forms consists of amphiphilic AChE extractable only with detergent containing buffer (for review see Rakonczay, 1986). Tetrameric AChE seems to be anchored to the membrane through a non-catalytic subunit of 20 kDa that apparently is linked to two of the four catalytic subunits by disulfide bridges (Gennaeri et al., 1987; Inestrosa et al., 1987). The assembly of four catalytic subunits together with the attached hydrophobic anchor migrates on SDS-PAGE with an apparent molecular mass of 350 kDa, whereas the monomer has an apparent molecular mass of approximately 68 kDa. According to Bon et al. (1986) the N-terminal sequence of the catalytic subunit is A s p / A l a - S e r - P r o G l u - A s p - P r o - G l u - L e u - L e u - V a l - M e t - V a l . Presently no other sequence data, neither of the catalytic subunit nor of the anchor, are available. On the other hand, complete amino acid sequences are known for human BChE (Lockridge et al., 1987; Prody et al., 1987) and for AChE from Drosophila (Hall and Spierer, 1986), Torpedo californica (Schumacher et al., 1986) and Torpedo marmorata (Sikorav et al., 1987). Among the different forms of mammalian ACHE, approximately 90% of the sequence of fetal bovine serum AChE is currently available (Smyth et al., 1988) and five tryptic peptides of human erythrocyte AChE are sequenced (Chhajlani et al., 1989). In this paper we report on the partial sequence of membrane-bound tetrameric AChE from bovine brain and dimeric AChE from bovine erythrocytes. MATERIALS A N D METHODS All chemicals used for sequencing were from Applied Biosystems, Warrington (England). Polyvinylidene diflouride (PVDF) membranes were from
PrimaryStructureof Bovine Acetylcholinesterases
107
Millipore, Bedford (Mass.). [3H]DFP was purchased from Amersham, Buckinghamshire (England). All other chemicals w e r e either from Fluka, Buchs (Switzerland), Sigma, St. Louis (Mo.), Merck, Darmstadt (FRG), and Boehringer, Mannheim (FRG). The reagents were at least of analytical grade; chemicals for amino acid analysis were all HPLC grade and sequencer grade, respectively. Bovine blood and bovine brain were purchased from the local slaughterhouse. The detergent-soluble AChE from bovine brain was extracted from caudate nuclei and purified by affinity chromatography essentially according to the method described for human brain AChE by Sorensen et al. (1982). In order to obtain enzyme of highest purity, a..~econd:affinity chromatography step was routinely performed, yielding AChE with a specific activity of 5000 IU/mg protein at minimum. Bovine erythrocyte AChE was purified from erythrocyte membranes as detailed for the preparation of human erythrocyte AChE (Brodbeck et al., 1981). After two affinity chromatography steps, the enzyme had a specific activity of 5500 IU/mg protein.
Enzyme Assay AChE activity was measured at room temperature by the method of Ellman et al. (1961). The assay solution contained 1 mM acetylthiocholine iodide and
0.25 mM 5,5'-dithiobis(2-nitrobenzoic acid) in 100 mM phosphate buffer, pH 7.4, supplied with 0.1% Triton X-100. Enzyme activity is expressed as international units (IU) (/~mol substrate hydrolyzed/min).
Active-Site Labeling Active-site labeling was achieved by incubation with 10-SM [3H]DFP for 15 hr at room temperature. In order to remove unreacted DFP and to allow complete aging of DFP-inhibited ACHE, the enzyme was dialysed for 3 days against three changes of 10mM Tris/HCl buffer, pH 7.4, containing 144 mM NaCI.
SDS-Polyacrylamide Gel Eiectrophoresis SDS-PAGE was performed essentially according to Laemmli (1970) using 5-15% polyacrylamide gradient gels. Samples were either subjected to gel electrophoresis without previous reduction or reduced by adding 5% (v/v) mercaptoethanol. Proteins were stained with Coomassie brilliant blue R-250. For autoradiography gels containing [3H]DFP-labeled samples were immersed in Amplify, dried, and exposed to Kodak X-Omat SO 282 film at -70°C. Tricine SDS-PAGE was performed according to Schaegger and von Jagow (1988). All samples were reduced with mercaptoethanol. The slab gels (1 mm thick) consisted of three sections: a 12-cm separation gel (16% acrylamide, 0.5% bisacrylamide), a 2-cm spacer gel (9.6/0.3%), and a 2-cm stacking gel (3.8/0.12%). No urea was included in the gels. Preelectrophoresis was done at 20 mA over night. Electrophoresis was performed at 10-mA initial current. When the samples had entered the stacking gel, the constant current was adjusted to 25 mA and then the gel was run for 16-18 hr.
108
Heider, Litynski, Stieger, and Brodbeck
Eiectroblotting and Staining of the Peptides The original blotting procedure onto PVDF membranes according to Matsudeira (1987) was used. Prior to blotting the gel was immersed for 10 min in the transfer buffer (10mM Caps, pH 11.0, 20% methanol). The PVDF membrane was first wetted with 100% methanol and then equilibrated in transfer buffer. The blotting sandwich contained 2 × 6 Whatman M3 filter papers with the gel and two PVDF membranes in between. Blotting was performed in a semidry electroblotting unit (Ancos, Denmark) at 0.8 mA/cm 2 for 2 hr. Under these conditions some larger peptides may traverse the first membrane and become trapped on the second one, whereas the smaller peptides are fixed on the first membrane. After blotting the membranes were rinsed briefly with distilled water and then stained in 0.1% Coomassie brilliant blue R-250 in 50% methanol for 2 min. Destaining was performed in 50% methanol and 10% acetic acid for 10 min. Then the membranes were rinsed in distilled water, air-dried, and stored at -20°C until further processing.
Amino Acid Analysis Amino acids were analyzed essentially according to Schaller et al. (1989) Areas containing the peptides of interest were excised from the PVDF membranes and hydrolyzed in gas phase with 6 M hydrochloric acid containing 0.1% (v/v) phenol and 0.05% (v/v) 2-mercaptoethanol for 22 hr at 115°C under vacuum. The samples were then dried under vacuum and the free amino acids converted to their corresponding phenylthiocarbamoyl derivatives by adding a mixture of ethanol/triethylamine/water/phenylisothiocyanate (70:10:10:1) and incubating for 20 min. The reagents were removed under vacuum and the samples redissolved in 0.14 M ammonium acetate buffer, pH 6.4. Reversed-phase HPLC analysis of the derivatives was performed on a Nova-Pak C18 column (4/~m, 3.9 x 150mm; Waters, Milford, Mass.) in a Hewlett Packard 1090 automated amino acid analyzer system essentially according to Bidlingmeyer et al. (1984).
Enzymatic and Chemical Cleavage Procedures AChE (about 400 IU in a volume of 300 ~1) together with [3H]DFP-labeled AChE (about 1/10 of the amount of unlabeled enzyme) was dialyzed against 10mM ammoniumcarbonate buffer, pH 7.4, and precipitated with ice-cold acetone to remove excess detergent. For tryptic cleavage the pellet was redissolved in 2 M urea, 50mM Tris, pH 8.0. Trypsin (5%, w/w) which was treated with N-tosyl-L-penylalanine-chloromethyl-ketone (BoehringerMannheim) was added and the samples were incubated for 2 hr at 37°C. The cleavage mixture was then immediately placed on a Tricine gel. Cyanogen bromide cleavage was performed essentially as described by Gross (1967). AChE (200 IU) was precipitated and redissolved in 70% formic acid supplied with 7% (w/v) cyanogen bromide. The reaction mixture was incubated in the dark at room temperature for 16 hr. The reaction was stopped by adding a 10-fold excess of water. The samples Were dried under vacuum and transferred to the Tricine gel.
PrimaryStructureof Bovine Acetylcholinesterases
109
N-Chlorosuccinimide cleavage was performed essentially according to the method of Lischwe and Ochs (1982) described for peptide mapping. ACHE, completely inhibited by [3H]DFP (corresponding to 100 IU), was precipitated as above and redissolved in 50#1 of a solution of urea/water/acetic acid (1 g / l m l / l m l ) supplied with 15 mM N-chlorosuccinimide. The reaction was allowed to proceed for 2 hr at room temperature and then the samples were transferred onto the Tricine gel. N-Glycosidase F Digestion Bovine erythrocyte and bovine brain AChE (30 IU each) were dried under vacuum and redissolved in 15 #1 buffer consisting of 50 mM sodium phosphate, pH 8.6, 10 mM EDTA, 1% Triton X-100, and 0.1% SDS. For the heat-denatured enzyme samples (1 min at 100°C), 0.5% octylglycoside instead of Triton X-100 and, additionally, 10 mM mercaptoethanol were used as incubation buffer. To each sample 0.4 unit of N-glycosidase F (EC 3.2.2.18; Boehringer-Mannheim, FRG) was added and incubated for 18 hr at 37°C. Then the samples were loaded on an 8-18% polyacrylamide gradient gel.
Determination of Amino Acid Sequences Bands to be sequenced were excised from PVDF membranes and cut into pieces of approximately 2 x 8 mm. Up to six pieces were placed into the reaction cartridge of the pulsed-liquid phase Applied Biosystems sequencer Model 477A. Amino acid analysis was performed with the on-line PTH-amino acid analyzer Model 120A (Applied Biosystems) using the NORMAL-1 and OLDPRO programs supplied by the manufacturer.
RESULTS Amino Acid Analysis The amino acid composition of bovine caudate nucleus AChE was determined using a gas phase hydrolyzing system, with subsequent conversion of the amino acids to their corresponding phenylthiohydantoine derivatives (Bidlingmeyer et al., 1984). Table I shows the amino acid composition of this enzyme in comparison to previously published compositions of fetal bovine serum AChE [hydrophilic tetramer (Ralston et al., 1985)] and bovine erythrocyte AChE [GPI-anchored dimer (Grossmann and Lieflaender, 1979)]. In general the compositions of the three forms are in good accordance with the exception of the high content of isoleucine in bovine erythrocyte ACHE. This enzyme contains 3.6 mol isoleucine/100 tool, compared to 1.4 mol/100 mol and 1.5 mol/100 mol, respectively, for bovine caudate nucleus AChE and fetal bovine serum ACHE. Assuming 540 amino acids per catalytic subunit of bovine erythrocyte AChE and 580 amino acids per catalytic subunit of fetal bovine serum AChE and bovine brain ACHE, respectively, these values yield 19 isoleucines for the erythrocyte enzyme and 9 Ile residues for the other two enzyme forms. The low content of lysine residues is a common feature of all three forms. Bovine caudate nucleus AChE contains approximately 10 residues per subunit, whereas the erythrocyte as well as the serum enzyme contain even lower amounts of lysine.
U0
Heider, Litynski, Stieger, and Brodbeck Table 1. Amino Acid Composition of the Catalytic Subunits of (a) Bovine Caudate Nucleus AChE in Comparison to Published Values for the Enzyme from (b) Bovine Erythrocytes (Grossmann and Liefl~inder) and (c) Fetal Bovine Serum (Ralston et al., 1985). as Moles per 100 mol a
Asp Glu Ser Gly ~s Arg Thr Ala Do ~r Val Met I~ Leu Phe ~s ~s
a
b
c
8.4±0.3 10.4±0.5 7.8±0.5 11.1~0.2 2.2i0.3 7.8~0.6 4.0i0.3 10.0±0.2 7.5±0.5 3.6~0.3 6.7~0.3 1.1±0.5 1.4~0.1 10.7±0.1 4.8±0.1 1.7±0.1 1.2~0.1
7.6 9.2 6.2 10.3 1.8 6.4 3.7 9.8 8.7 3.1 7.6 1.4 3.6 10.0 4.4 1.5 1.4
8.2 11.0 9.1 11.9 2.6 6.2 3.6 9.9 7.4 2.1 7.6 1.5 1.5 10.1 4.6 1.4 1.3
The data in a represent the average of five amino acid determinations carded out as described under Materials and Methods. a
SDS-PAGE Patterns of the Cyanogen Bromide-Cleaved Enzymes AChEs from bovine caudate nucleus and bovine erythrocytes were treated with cyanogen bromide and the resulting peptides resolved by Tricine SDSPAGE. As shown in Fig. 1 the major peptides have molecular masses between 29 and 38 kDa. The two main peptides derived from the bovine brain enzyme had an identical N-terminal amino acid sequence, suggesting either different C-terminal
i ~a
1
2
Fig. 1. SDS-PAGE pattern (Coomassie stain) of bovine caudate nucleus (1) and bovine erythrocyte AChE (2) after cyanogen bromide clevage. EIectrophoresis was carded out according to Schaegger and von Jagow (1988) as detailed under Materials and Methods. Numbers at the fight refer to molecular weight markers (2.5-17 kDa, Sigma low molecular weight kit, 31 and 43 kDa; Bio-Rad calibration kit). Arrows pointing to bands indicate peptides from which the N-terminus was determined.
Primary Structure of Bovine Acetylcholinesterases
111
sequences of the peptides or one of the two peptides being modified. The same N-terminal sequence was also obtained with the major 38-kDa band of bovine erythrocyte ACHE. Comparison of this sequence to that of fetal bovine serum AChE (Smyth et al., 1988) revealed that it is located C-terminally from the active-site serine residue.
Cyanogen Bromide, N-Chlorosuccinimide, and Tryptic Cleavage of the [3H]DFP-Labeled AChE In order to obtain information On the sequence of the active-site serinecontaining peptide, AChE from bovine caudate nucleus and bovine erythrocytes was labeled with [3H]DFP. Cleavage was performed by treatment with cyanogen bromide, N-chlorosuccinimide, and trypsin. The resulting peptides were separated on the Tricine gel system. As shown in Fig. 2 cyanogen bromide cleavage of both enzymes resulted in two major radioactive bands at identical positions around 3.5 kDa (lanes 1 and 2). Two further bands at identical positions were obtained around 13 kDa, whereas the patterns at higher molecular masses were different. Cleavage of the two enzyme preparations with N-chlorosuccinimide resulted in a more complex band pattern (lanes 3 and 4). Peptides up to 14 kDa migrated to identical positions in the gel, whereas above this molecular mass the pattern differs. The smallest labeled peptides were found at around 3 kDa, in the -.o-- 43
ii!iiiili¸iii!iii ?iiiiii! ii! --31
.... iii!!i
