purification of acetylcholinesterase by affinity chromatography and ... · purification of...

THE JOURNAL OF BIOLOGICAL CHEM~STRV Vol. 247, No. 5, Issue of March 10, pp. 1555-1565, 1972

Printed in U.S.A.

Purification of Acetylcholinesterase by Affinity Chromatography

and Determination of Active Site

Stoichiometry*

(Received for publication, November 9, 1971)

TERRONE L. ROSENBERRY,HAI WON CHANG,AND YUEH T. CHEN

From, th,e Department of Neurology, College of Physicians and Surgeons, Columbia University, New York, New York 1003~

SUMMARY

Specific inhibitors of acetylcholinesterase have been prepared and attached to Sepharose. When crude preparations of the enzyme are chromatographed on the aflinity gel, the enzyme is selectively retained and may be eluted by a salt gradient. Conditions have been defined under which enzyme is purified some ZO-fold in one step to a homogeneous and apparently pure species. An essential factor appears to be the distance the inhibitor is extended from the gel matrix by a chain of bridging atoms which form an attachment arm. A careful series of protein determinations was carried out to establish the ultraviolet extinction coefficient of the purified enzyme. Inherent differences among the protein values determined by various techniques appear to exist. They may in part account for discrepancies in values reported previously for the specific activity of homogeneous enzyme. With the use of an E$,~~ = 18.0, a specific activity of 610 mmoles of acetylcholine hydrolyzed per hour per mg of protein was observed for the most highly purified enzyme. Active site titration measurements revealed 3.3 active sites per enzyme molecule.

Acetylcholinesterase (EC 3. I. I. 7) was first isolated by extraction from the electric organ of Torpedo murmorata in 1938, following the discovery of an extraordinary concentration of the enzyme in this tissue (1). Purification of this enzyme from electric tissue of Electrophorus electricus by fractional precipitation of the enzyme at carefully controlled ammonium sulfate con-

centrations and pH values routinely gave a lOO- to 200.fold increase in its specific activity’ to a value of about 150, and further

* This work was supported in part by Grant NSF-GB-25362 from the National Science Foundation. Grant NS-03394 from the Sational Institutes of Health, and by’gifts from the New York Heart Association, Inc., and the Hoffmann-LaRoche Foundation.

1 Enzyme activities are given in the units mmoles of acetylcholine hydrolyzed per min per ml of enzyme solution. Enzyme specific activities are given in the units mmoles of acetylcholine hydrolyzed per hour per mg of protein, where the protein concentration is estimated using 6.j; nm = 18.0 & 0.4.

treatment by ultracentrifugation yielded a rapidly sedimenting, apparently homogeneous enzyme species which occasionally had a specific activity of 400 to 500 (2-4). Later investigations in- troduced multistep procedures using ion exchange and gel exclu- sion chromatography (5). These chromatographic procedures lead to the isolation of a homogeneous enzyme protein which has been crystallized (6), but the methods are laborious and result in an enzyme yield of less than 10%. In view of the difficulties in procuring electric tissue, such a yield is prohibitively small for use in studies aimed at elucidating the structure and properties of the enzyme. Moreover, the ultraviolet spectra and extinction coefficients reported for purified acetylcholinesterase (7, 8) are in disagreement. While this discrepancy may reflect a difference in purity of the two preparations, it seems likely that intrinsic differences in the quantitative protein determinations are also involved. The lack of a reference extinction coefficient for acetylcholinesterase has hampered the comparison of specific activities for purified enzyme with those reported in the literature (5, 6, 9). Extinction coefficients based on a variety of protein determinations will be presented in this paper, and hopefully they will facilitate the comparison of various enzyme preparations.

The technique of affinity chromatography was recently intro- duced in the purification of acetylcholinesterase (10) after it found wide application in a variety of protein purification sys- tems (11). It involves the attachment of specific, reversible acetylcholinesterase inhibitors to a Sepharose resin matrix by means of the coupling agent cyanogen bromide. When a crude preparation of acetylcholinesterase is applied to a chromatographic column containing the modified resin, the acetylcholinesterase is selectively retained. The enzyme may be eluted from the column with a salt gradient. The extent to which the target protein is retained on the affinity column has been found to de- pend upon the length of the attachment arm which links the specific inhibitor to the resin (12). A number of specific inhibitors of acetylcholinesterase with attachment arms of varying length are currently under investigation in this laboratory as candidates for the affinity column purification of the enzyme. In this study we report the purifications of Electrophorus enzyme attained with the use of the inhibitor phenyltrimethyIammonium attached to Sepharose 4B with arms of two lengths: mono-(&

1555

by guest on February 26, 2019http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

15.56 A finity Chromatography Pur$cation of Acetylcholinesterase Vol. 247, No. 5

aminocaproyl)-PTA (IV)z and di-(6.aminocaproyl)-PTA (VIII).

CH, 0

c&?& / 1 -0

-(NH-&H,),-),-fiH, 2Bre

LIT3 IV: n = 1 VIII: n = 2

The chromatographic results obtained are compared directly with those from the conventional multistep procedure (5, 6). Although both procedures result in highly purified enzyme, the purification which can be obtained with affinity chromatography is slightly higher and the yield is much higher than that with the conventional procedure. During the course of this work a report on the purification of acetylcholinesterase by aflinity chromatography appeared (9) in which a comparison of the re- tentive properties of several inhibitor-linked Agarose gels was made. The study presented here supplements that report in giving information both on somewhat different inhibitor ligands and on the elution profile and purity of the recovered enzyme. A high level of purification is achieved and a titration of enzyme- active sites is made to determine the number of active sites per molecule.

MATERIALS

Chromatographic Resins and Gels-DEAE-cellulose was obtained from Eastman; cellulose phosphate (Cellex-P) was pur- chased from Bio-Rad, Richmond, California; and Sephadex and Sepharose gels came from Pharmacia.

Titrating Agents-The synthesis of N-methyl-(7-dimethylcarbamoxy)-quinolinium iodide has been described earlier (13). 3,3’-Bis[a!- (trimethylammonium)methyl] -azobenzene bromide, whose synthesis has been described (14), and 3,3’-bis[cu-([i4C]- trimethylammonium)methyl]-azobenzene iodide, with a specific activity of 0.74 mCi per mmole, were generous gifts of Dr. N. H. Wassermann, Columbia University. For Bis-Q2 in water under conditions in which the trans-isomer predominates to greater than 90% (14), X,,, 316 nm (E, 16,000).

Synthesis oj Inhibitors

6-Carbobenzoxyaminocaproic Acid (I)-A solution of 39.4 g (0.3 mole) of 6-aminocaproic acid in 150 ml of 2 N NaOH was chilled to 0”. Carbobenzoxychloride (57.5 g (95%), 0.32 mole) and 74 ml of 4 N NaOH solution were added simultaneously with good stirring over a period of 50 min, maintaining the reaction temperature O-7”. After stirring an additional 30 min, the reaction mixture was extracted three times with about 100 ml of ether to remove excess carbobenzoxychloride, and the aqueous solution was acidified slowly to pH 3 with 6 N HCl. The product was extracted three times with 250 ml of ether, and the

* The following abbreviations are used: mono- (6-aminocaproyl) - PTA, [N-(6-aminocaproyl)-p-aminophenyl]trimethylammonium bromide hydrobromide (IV). di-(6-aminocaproyl)-PTA, [N-(6- aminocaproyl-6’-aminocaproyl) -p-aminophenyl]trimethylammonium bromide hydrobromide (VIII); Bis-Q, 3,3’-bis[ol-(trimethylammonium)methyl]-azobenzene bromide; mono-(6-aminocaproyl)- MP, methyl-[N-(6-aminocaproyl)-3-aminolpyridinium bromide hydrobromide; M7C, N-methyl-(7-dimethylcarbamoxy)-quinolinium iodide; [“C]Bis-Q, 3,3’-bis[or-([‘4C]trimethylammonium)- methyl]-azobenzene iodide; mono-(6-aminocaproyl)-BTA, [N-(6- aminocaproyl)-p-aminobenzyl]trimet.hylammonium bromide hydrobromide (XI).

combined ether layers were washed with brine solution and dried over anhydrous Na2S04. Evaporation of the ether afforded an oil which was crystallized from benzene and a small amount of hexane on chilling in a refrigerator overnight. The crude product was 74.2 g (93.5%) of white crystals, m.p. 54-56”, and the recrystallization raised the melting point to 59960”; infrared (KBr): 3320, 1725, 1695, 1535, and 1250 cm-i.

Calculated: C 63.38, H 7.22, N 5.28 Found : C 63.57, H 7.27, N 5.28

N’- (6. Carbobenzoxyaminocaproyl) -p-N, N -dimefhylphenylene- &amine (II)-N,N’-Dicyclohexylcarbodiimide (5.75 g, 0.0279 mole) was added to a solution of N,N-dimethylphenylenedia- mine (3.5 g, 0.0257 mole) and 7.0 g (0.0264 mole) of Compound I in 150 ml of acetonitrile (distilled over PzO5) and stirred at room temperature overnight. The whit,e precipitate, dicyclohexylurea (4.4 g), was filtered away, and the filtrate was evaporated and replaced with boiling benzene (100 ml). It was treated with Norit, filtered, and the slightly colored crystalline product (9.8 g, 71.3 To) was obtained upon cooling. Recrystallization from benzene and hexane gave white crystals, m.p. 113-114”; infrared (KBr): 3320, 1710, 1665, 1525, and 1250 cm-i.


[N-(6-Carbobenzoxyaminocaproyl) -p-aminophenyl]trimethylammonium Iodide (III)-Lfethyl iodide (10 ml, an excess) was added to a solution of Compound II (6.5 g, 0.017 mole) in 10 ml of dimethyl formamide and heated on a steam bath for 20 min. Ethyl acetate was added dropwise with swirling to the reaction flask until slight turbidity appeared. On cooling and rubbing the side of the flask with a glass rod, 8.5 g (96%) of white crystals, Compound III, m.p. 112-114”, were obtained. Infrared (KBr): 3270, 1712, 1668, 1514, and 1232 cm-l.

Calculated: C 52.58, H G.14, N 7.99 Found: C 52.47, H 6.22, N 7.84

[N-(6-Aminocaproyl)-p-aminophenyl]trinzethylammoni~cs~ Bro- mide Hydrobromide (W-Glacial acetic acid (13 ml) which had been saturated with gaseous anhydrous hydrogen bromide II-as added to 1 g of Compound III and stirred for 10 min. When there was no further sign of decarboxylation, 25 ml of anhydrous ethyl ether were added with swirling and allowed to stand for about 5 min to ensure the separation of gummy product,. The supernatant was decanted, and the product was washed twice with 25 ml of ether. The resulting light brown, gummy material was crystallized from methanol and ethyl acetate to yield 0.78 g (97%) of white crystals, m.p. 205-207”, with decomposition. Treating the aqueous solution of the product with freshly prepared AgBr, filtering, evaporating, and following the recrystallization from methanol-ethyl acetate afforded an analytical sample, m.p. 209-210”; infrared (KBr): 3000, 1690, 1609, and 1540 cm-l; ultraviolet: X,,, 243 nm (E, 14,900).

GJL~N30Br2

Calculated: C 42.37, H 6.40, N 9.88, Br 37.58 Found : C 42.27, H 6.32, N 10.05, Br 37.56


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

[W-(6- Aminocaproyl) - p - N,N - dimethylphenylenediamineldihy- toluene (16) (3 g, 0.01668 mole), 4.4 g (0.0168 mole) of Com- drobromide (V)-To 2.2 g (0.0057 mole) of Compound II were pound I, and 3.71 g (0.018 mole) of dicyclohexylcarbodiimide added 25 ml of saturated solution of dry hydrogen bromide in were added to 100 ml of acetonitrile and stirred at room temper- glacial acetic acid and stirred for about 10 min. Upon cessa- ature overnight. The white precipitate, dicyclohexylurea, lvas tion of CO* evolution, 50 ml of anhydrous ether were added to filtered away, and the filtrate was evaporated. Sodium carbon- precipitate the amine hydrobromide formed. Since the precipi- ate solution was added to the residue and extracted three times tate was gummy rather than crystalline, the supernatant solu- with chloroform (250 ml), dried with anhydrous sodium sulfate. tion was decanted away, washed twice with 20.ml portions of filtered and evaporated. The crude product was dissolved in anhydrous ether, and crystallized from absolute ethanol-ether boiling benzene, filtered (to remove a small amount of dicyclo- to yield 2.2 g (93%) of white crystals, m.p. 181-182”; infrared hexylurea), and crystallized on cooling with a small amount of (KBr): 2950, 1678, 1542, and 1505 cm-‘. hexane. Recrystallization from 50% benzene-hexane gave 3.87

C14H2aN30B1.2 g (59%) of white crystals, m.p. 108-109”; infrared (KBr): 3300, 1700, 1670, 1530, and 1250 cnP.

Calculated: C 40.89, H 6.13, N 10.22 Found: C 40.76, H 6.20, N 10.74 CdLNzOa

A”-(6-Carbobenzoxyamirwcaproyl-6’- amirwcaproyl) - p-N,N -di- Calculated: C 69.50, H 7.86, N 10.58

methylphenylenediamine (VI)-One gram (0.00243 mole) of Com- Found : C 69.47, H 8.05, N 10.87

pound V and 0.495 g (0.0049 mole) of triethylamine were stirred [N- (6- Carbobenzoxyamirwcaproyl) -p-aminobenzyl]trimethylam- in 60 ml of acetonitrile for a few minutes, and 0.645 g (0.00243 monium Iodide (X)-Methyl iodide (0.5 ml) was added to a mole) of Compound I and 0.535 g (0.0026 mole) of dicyclohexyl- solution of Compound IX (0.265 g, 0.668 mmole) in 10 ml of carbodiimide were added. The mixture was stirred at room absolute ethanol and heated under reflux on a steam bath for 20 temperature overnight, and the precipitate (dicyclohexylurea min. Ethyl acetate was added dropwise until slight turbidity and some of triethylamine hydrobromide) was filtered. After appeared. On cooling and rubbing the side of the flask with a the filtrate was evaporated, the residue was treated with 20 ml glass rod, 0.355 g (98%) of white crystals, Compound X, m.p. of 10% Na2C03 solution and extracted three times with chloro- 128-130”, was obtained. Infrared (KBr): 3250, 1700, 1600, form (80 ml). The chloroform layers were dried over anhy- 1520, and 1253 cm-‘. drous Na$O+ evaporated, and the resulting residue was frac- tionally crystallized twice from boiling acetonitrile to afford CzdbN~OJ

0.655 g (54%) of white crystals, m.p. 126-127’; infrared (KBr): Calculated: C 53.54, H 6.37, N 7.80

3290, 2900, 1683, 1640, 1525, and 1260 cm+ Found : C 53.72, H 6.38, N 7.56

CdLoO,N~ [N-(6-Aminocaproyl)-p-aminobenzyl]trimethylammonium Bro-

Calculated: C 67.72, H 8.12, N 11.28 mide Hydrobromide (XZ)-Compound XI was obtained in 907, Found: C 67.61, H 8.50, N 11.98 yield from X as described for Compound IV, white crystals,

[a-- (B-Carbobenzoxyaminocaproyl-6’-aminocaproyl) -p -amirw- m.p. 219-220”; infrared (KBr) : 2870,1683, 1602, and 1535 cm-‘;

phenyl]trimethyZammonium Iodide (VZZ)-Compound VII was ultraviolet: X,,, 252 nm (e, 16,900).

prepared quantitatively from VI by methylation as described C1&~N30Br2 for Compound III to give white crystals, m.p. 89-92”, and recrystallization from ethanol-ethyl acetate gave m.p. 109-111”;


(KBr): 3220, 2900, 1690, 1640, 1525, and 1250 cm-l.

‘&&04NJ N- (6-Carbobenzoxyuminocaproyl) -S-aminopyridine (XZZ)-

The Compound XII was obtained in 92.8% from 3-aminopyri- Calculated: C 54.55, H 6.79, N 8.77 Found : C 54.64, H 6.74, N 9.37

dine as described for Compound II, m.p. 90-92”. Fractional recrystallization from acetonitrile gave XII, m.p. 97-97.5”; in-

[.\i-(6-Aminocaproyl-6’.aminocaproyl) -p-aminophenyl]trimeth- frared (KBr): 3280, 1695, 1670, 1540, and 1250 cm-l.

ylammonium Bromide Hydrobromide (VZZZ)-Compound VIII ChJLO~N~ was prepared in 89% yield from VII as described for Compound IV to give white crystals, m.p. 170-173”. Analytical sample,

Calculated: C 66.84, H 6.79, N 12.31

prepared by treating aqueous solution with freshly prepared Found: C 67.16, H 6.88, N 11.70

AgBr and recrystalliz:ng from boiling ethanol and ethyl acetate Methyl-[N-(6-carbobenzoxyaminocaproyl) -S-amino]pyridinium had m.p. 179-180”; infrared (KBr): 2950, 1695, 1640, and 1540 Iodide (XZZZ)-Methyl iodide (0.5 ml) was added to 0.272 g cm-‘; ultraviolet: X,,, 243 nm (E, 14,860). (0.798 mmole) of pure Compound XII in 10 ml of ethanol and a

G1H,,N402Br2 few drops of dimethylformamide and heated under reflux for 30 min.

Calculated: C 46.85, H 7.11, N 10.40, Br 29.69 Anhydrous ether was added dropwise until a slight turbid-

Found : C 47.04, H 7.19, N 10.36, Br 29.59 ity appeared, and the flask was left in a refrigerator for 2 days with occasional rubbing of the side of the flask with a glass rod.

N’-(6.Carbobenzoxynmirwcaproyl) -p-amino-N, N -dimethylben- The crystallization is slow and somewhat difficult, but finally it zylamine (IX)-The crude p-amino-cr-dimethylaminotoluene gave very pale crystals, Compound XIII, 0.347 g (90yc). Re- (15), IT-hich was prepared from tin (5 g) and concentrated hy- crystallization from ethanol and ethyl acetate gave m.p. 83-85”. drochloric acid (12 ml) reduction on p-nitro-cu-dimethylamino- Infrared (KBr): 3200, 2900, 1690 (broad), 1538, and 1250 cm’.

Issue of March 10, 1972 T. L. Rosenberry, H. W. Chang, and Y. T. Chen 1557


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1558 A&finity Chromatography Puri$cation of Acetylcholinesterase Vol. 247, Ko. 3


lVethyl-[N-(6-aminocaproyl)-S-amino]pyridinium Bromide Hy- drobromide (XIV)-Compound XIV was prepared from XIII as described for Compound IV, white crystals, m.p. 204-205”; recrystallization after treating with AgBr gave an analytical sample, m.p. 208” (decomposed). Infrared (KBr) : 2950, 1705, 1555, and 1500 cm-l; ultraviolet: X,,, 252 nm (E, 11,100) and 291 nm (E, 4,620).


METHODS

Coupling of Inhibitors to Sepharose--9 modification of the procedure given by Cuatrecasas (12) was made in order to attain a reproducible (10 to 15%) ligand concentration per ml of Sepha- rose and to conserve ligand. The following example is typical of the coupling procedure. A well washed suspension of 100 ml of Sepharose 4B on a Buchner funnel was transferred to a 200- ml beaker with 100 ml of distilled water and cooled to 4’. The ice bath was removed, a thermometer and a pH electrode were immersed in the beaker, and 1 g (9.45 mmoles) of CNBr was added at once. The pH of the solution was quickly adjusted to and maintained at 10.5 + 0.3 with 4 N NaOH with vigorous stirring. The reaction was completed in about 7 to 8 min, and the final temperature of the reaction mixture was about 15”. Ice was added to bring the temperature down, and the mixture was quickly transferred to a Buchner funnel and washed for 5 to 7 min under suction with 1 liter of cold 0.05 M carbonate-bi- carbonate buffer (pH 9.8). The activated Sepharose was rapidly suspended in 80 ml of the above buffer solution containing 250 pmoles of ligand and stirred overnight at 4”. Under these conditions the ligand is assumed to be in excess of the Sepharose activation sites. The substituted Sepharose was then washed with 1 liter of distilled water on a Buchner funnel and further washed extensively with distilled water until no ligand ultraviolet absorbance was detected in the washings. Table I sum- marizes the optimum ratios of materials used to obtain a pre- scribed level of ligand coupling for 25- to 250.ml quantities of Sepharose 4B.

TABLE I Coupling of ligand to Sepharose 4B

The preparation of cyanogen bromide-activated Sepharose and conditions for the coupling of the inhibitor ligand, in this case either mono-(6-aminocaproyl)-PTA or di-(6-aminocaproyl)-PTA, are described under “Materials.” The quantity of material coupled to the Sepharose was calculated from the amount of un- reacted ligand in the gel washings, measured by the absorbance at 243 nm.

1

Added CNBr Added ligand / Attached ligand 1 %&f$ / %%$!

?ng/ml gel p??mles/ml gel /.lmoles/ml gel % %

5 2.0 0.7 1.5 35 10 2.5 to 3.0 2.0 f 0.3 2.1 60 to 80 20 6.0 to 8.0 4.0 f 0.6 2.1 60 to 80

Benzylated DEAE-cellulose (B-DEAE-cellulose)-The preparation of B-DEAE-cellulose and the washing procedures for this and other ion exchange celluloses have been described earlier (5). The solvent for the benzylation of DEAE-cellulose was modified to 200 ml of dimethylformamide and 800 ml of methyl cellosolve per 50 g of DEAE-cellulose. Microgranular DEAE- cellulose (Whatman) was not suitable for benzylation under these conditions because it is too soluble for adequate recovery.

Crude Enzyme-Lyophilized crude enzyme preparations were obtained in collaboration with Dr. A. L. Baker of Worthington. The crude enzyme had been extracted from toluene-treated tissue and carried through an ammonium sulfate precipitation as described in Steps 1 and 2 of Reference 6, a procedure which results in a 20. to 30.fold purification relative to enzyme in a fresh tissue homogenate. These crude preparations served as the starting material for all purification procedures described in this paper. They had essentially constant specific activities’ of 25 to 32.

Enzyme Activity Assay-Enzyme activities’ were measured by the pH-stat method with the use of a Radiometer titrator. The assay medium contained 0.1 M sodium chloride, 0.02 M magne- sium chloride, 0.0057, gelatin, and the substrate, 2.7 m&r re- crystallized acetylcholine bromide (Eastman). The pH was adjusted to 7.4, and the temperature to 25”.

Protein Determination-Protein concentra,tions were routinely estimated spectrophotometrically, using either the absorbance at 280 nm or the absorbance difference between 215 and 225 nm. Extinction coefficients for purified acetylcholinesterase at these wave lengths were determined by relating the absorbance measurements to absolute protein values based on nitrogen determination, differential refractometry, and dry weight.

Nitrogen was determined by both the Dumas and the micro- Kjeldahl methods (Schwarzkopf Microanalytical Laboratories, Woodside, N.Y.) ; nitrogen determinations on a micro-scale were also performed by means of a ninhydrin complex-forming procedure (17). Nitrogen values were converted to protein with the use of the assumption that the protein contains 16.070 nitrogen.

The absolute protein contents of samples of protein carefully dialyzed against the solvent were determined by an interfero- metric technique, using a capillary type of synthetic boundary cell in the Spinco model E analytical ultracentrifuge as a differential refractometer. A value of 4.21 interference fringes per mg per ml was used and included a slight correction for the amino acid composition 08).

Dry weight measurements were carried out as described by Leuzinger et al. (8) on protein dialyzed for 3 days against 0.002 M ammonium carbonate. Several aliquots containing 50 to 300 pg of protein or appropriate blanks were dried at 80”, then re- peatedly heated at 110” for 60 min and cooled in a desiccator over phosphorus pentoxide until the weights were constant.

An analysis of purified protein solutions for the presence of sugar was carried out (19) with sucrose as a standard at two different phenol concentrations. Data were collected in t,erms of the phenol concentration which gave the higher percentage of sugar estimate, and a slight protein blank correction was made from control analyses of highly purified chymotrypsin (Worth- ington), lysozyme (Worthington), and bovine serum albumen (Sigma).

Puri$cation by Afinity Chromatography-Stock solutions of 0.02 M potassium phosphate buffer at pH 6.9 and 0.5 M NaCl in


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Issue of March 10, 1972 T. L. Rosenberry, H. W. Chang, and Y. T. Chen 1559

buffer were prepared and filtered with a Millipore apparatus. From this point all steps were carried out at 3-5”. A solution of 3.6 to 5.0 g of lyophilized crude enzyme dissolved in 80 ml of stock buffer was dialyzed against 10 liters of stock buffer for 24 hours at 5”. A slight turbidity was removed from the enzyme solution by rentrifugation at 18,000 X g for 20 min. Aliquots of this crude enzyme stock solution were then added to a chromatographic column of the substituted Sepharose which had been washed and packed in stock buffer. The crude enzyme aliquot was allowed to equilibrate with t,he column overnight. The aliquot addition was usually followed by further washing with stock buffer until the nonretained material had been completely eluted and the 280 nm absorbance of the washings had dropped to zero. Elution of the retained enzyme was accomplished by a buffered linear gradient of sodium chloride or potassium chloride, starting either from zero salt concentration or from a low salt concentration which had been used to further wash weakly retained impurities from the column.

The ,column eluent was continuously monitored at 254 nm with an LKB Uvicord I. Fractions were assayed both for enzyme activity on the pH-stat and for protein spectrophotometrically. Active fractions were pooled and concentrated by a vacuum dialysis procedure which uses a collodion membrane bag (Schleicher and Schuell, Inc.) for subsequent characteriza- tion studies.

Pur$cation by Standard Chromatographic Procedures-These studies were made generally as outlined in the detailed chromatographic methods given in Steps 2 through 6 of the procedure of Kremzner and Wilson (5). However, the following modifications of these procedures were applied in our work.

A concentration of 0.03 M sodium phosphate buffer was used for equilibration and elution of all columns. All steps were carried out at 3-5”. The lyophilized crude enzyme was dissolved in 2.5% ammonium sulfate (2 to 10 mg of protein per ml); in- soluble material was removed by centrifugation at 25,000 X g for 10 min; and the mixture was applied to a B-DEAE-cellulose column (3 to 12 mg of protein per ml of packed resin) in Step 2. -1 solvent higher in ionic strength than that in the previous procedure was used to minimize the presence of soluble protein ag- gregates which may interfere with purification on B-DE,4E-cellulose.3 The pooled enzyme, eluted with a stepwise sodium chloride gradient in Step 2, was concentrated (Step 3) by the vacuum dialysis procedure described above. This sample was run through a Sephadex G-200 column (1 ml of applied sample per 50 ml of packed gel) in Step 4.

The pH of the enzyme solution and of the column equilibration was adjusted from the usual 7.0 to 5.7 for cation exchange chromatography on Cellex-P (Step 5) immediately prior to application of the sample (0.2 to 0.3 mg of protein per ml of packed resin). Enzyme was eluted with a stepwise sodium chloride gradient, immediately readjusted to pH 7.0, and concentrated by vacuum dialysis.

The final purification steps were carried out with a combina- t,ion of anion exchange columns which could be used interchange- ably. The resins used in thefe steps were DEAE-cellulose, DEAE-Sephadex, and QAE-Sephadex. The applied sample protein per column volume was the same as that used in Steps 6 and 7 of Reference 5.

A-lmi~u, Acid Composition-Separate analyses were performed

3 Dr. L. Kremzner, personal communication.

on three 0.5-mg samples of purified and concentrated enzyme as described by Moore and Stein (20). To each dried sample was added 0.5 ml of fresh 6 N HCl, and the hydrolyses were carried out at 110” for 24 hours. The analyses were performetl on a Beckman amino acid analyzer.

Competitive Inhibition Con&ants-These were determined by computer analysis of reciprocal plot data (13) in which both the acetylcholine substrate and the inhibitor concentrations were varied.

Active Site Titration-Enzyme-active site normalities were dete mined either by fluorescent titration with the carbamoylating agent M7C2 as described in Reference 13 or by equilibrium dialyses using [14C]Bis-Q. The fluorescent titrations were measured at 25” in pH 6.85 buffer (0.05 af phosphate) containing 0.1 M NaCl and were monitored on an Aminco-Bowman spectro- photofluorometer with dual monochromators. Kinetic pnram- eters for the carbamoylation reaction were also determined as in the previous report (13).

Equilibrium dialyses of 0.5- to l&ml enzyme samples (0.5 to 1.0 mg per ml) in S-mm dialysis tubing were carried out for 20 hours at 1-2” with shaking in 75 ml of solvent containing 2 to 200 PM of either [14C]Bis-Q or a mixture of [%]Bis-Q and Bis-Q. The solvent for the titration was either 0.1 M NaCl in pH 6.85 buffer (0.05 M phosphate) or normal eel Ringer’s minus glucose (21) at pH 7.0. Aliquots of 100 to 500 ~1 were taken from bot,h the inside and the outside of the equilibrated dialysis solutions and added to lo-ml quantities of a slightly modified Bray’s scintillation cocktail (22) for count.ing on a Packard model 574 scintillation counter. Further additions to the counting vials were made in some cases to make the total composition of added sample in these vials identical, except for the quantity of [14C] Bis-Q, but such precautions to insure reproducible counting efficiencies had little effect on the results. -4 weighted least squares computer analysis of reriprocal plot data was used to quantitatively determine t.he binding parameters, and these values were in agreement with those estimated from Scatchard plots.

Irnmunoelectrophoresis-This procedure was carried out. in a 1.5% agar matrix (23, 24) w&h the use of sodium barbitol buffer (0.05 M, pH 8.3). The voltage was adjusted to 20 volts per cm, and the sample was run for 100 min.

Polyacrylamide Gel Electrophoresis-Polyacrylamide gel electrophoresis patterns were obtained in glass tubes (6 x 150 mm) in a Buchler apparatus thermojacketed at 10”. A procedure based on the alternate method of Davis (25) was followed, using Tris-glycine buffer at pH 8.3, 7oc polyacrylamide in the separation gel, and 1.25yo polyacrylamide in the 0.5~cm spacer gel (reagents from Canalco and Eastman). Samples were added to a top 20.~1 layer of 40% sucrose which contained bromphenol blue as a tracking dye.

Gels were protein-stained for 20 min with Xmido schwarz; staining for esterase activity was carried out in phosphate buffer (0.05 M, pH 7.0) and was based on the diazo blue B reaction with P-naphthol released during hydrolysis of @-naphthyl acetate (24). Gels were destained by diffusion in i70 acetic acid. Relative band intensities were measured on a Canalco model E microdensitometer.

Error Analysis-The analysis of variance in weighted least squares plots was carried out as outlined previously (13). The standard error (SE.) is listed with several tabulated values. This quantity is the square root of the variance of the mean


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1560 A@aity Chromatography Puri$cation of Acetylcholinesterase Vol. 247, No. 5

value and is given by the expression

S.E. = xi Wi(Zi - 3)’

(xi Wi)(n - 1)

where zi is an individual observation; z is the mean of the zi values; wi is the weight given to an individual observation; and n is the number of observations. A weighted analysis was used

TABLE II

Com.petitive inhibition dissociation constants (KI) for several inhibitors of acetylcholinesferasea

Inhibitor

Mono-(6-aminocaproyl)-PTA . Mono-(6-aminocaproyl)-PTA-Sepharoseb. Di-(6aminocaproyl)-PTA. Di-(6-aminocaproyl)-PTA-Sepharoseb. Mono-(6-aminocaproyl)-BTA. Mono-(6-aminocaproyl)-MP . Bis-Q.

39 f 9

30 f 8 44 f 5 14 f 2 52 f 37

3.8 f 0.5 4.6 f 0.2

Q KI values estimated from weighted least squares plots (13) (see also under “Methods”).

*Attainment of steady-state inhibition was slow with these gel-linked inhibitors. Enzyme was pre-equilibrated with the final concentration of inhibitor gel for 30 to 45 min prior to each rate measurement.

% o.5 E

= z 0.4

cu

2 0.3

5 m 0.2

E

2 0.1 a

0

I I I I

20 40 60

20 40 60

FRACTION NUMBER

0.4

0.2

0

FIG. 1. Elution of acetylcholinesterase from affinity column with continuous linear sodium chloride gradient. Crude enzyme solution with a total activity’ of 250 mmoles per min was passed through a column (1.5 X 30 cm) containing 22 ml of gel to which the affinity ligand di-(6-aminocaproyl)-PTA had been attached (for procedures see under “Methods”). The lower elution profiles show the absorbance at 280 nm, the enzyme activity, and the calculated sodium chloride concentration for each of the lO.O-ml eluted fractions. The upper profile gives the corresponding specific activity’ of each fraction. Fractions 1 to 4 contain nonretained protein which included 58 nmoles per min (23%) of the applied enzyme. Retained protein in Fractions 5 to 70 included 183 mmoles per min (73%) of the applied enzyme.

only when each weight could be determined experimentally as the reciprocal of the variance of the given observation. In un- weighted averaging, wi = 1 for all i.

RESULTS

Inhibition of Enzyme Activity-The competitive inhibition of acetylcholinesterase activity by the ligands synthesized for use in affinity chromatography is shown in Table II. The charac- teristic inhibition patterns of the ligand are altered only slightly on attachment to the gel; no significant differences are observed among the various PTA derivatives except for the di-(6-aminocaproyl)-PTA-Sepharose, which apparently binds enzyme 2 to 3 times more tightly than the free ligand. The free mono-(6- aminocaproyl)-MP2 ligand inhibits the enzyme with a competitive Kr which is about 10 times lower than the PTA derivatives and therefore may be more selective in affinity chromatography. Studies using this inhibitor with extension arms of various lengths are currently in progress.

Afinity Chromatogratphy-A preliminary affinity chromatography purification using the di-(6-aminocaproyl)-PTA-Sepha- rose is shown in Fig. 1. The enzyme is the last detectable protein component to be eluted from the column, and it is released to eluting buffer containing about 0.2 M NaCl. While the resolution of enzyme from protein contaminants appearing earlier in the elution profile may be improved by an increased period of washing with stock buffer, the preliminary chromatographic run in Fig. 1 demonstrates a sharp rise in specific activity’ of eluted enzyme to a constant value corresponding to that of essentially homogeneous enzyme (see below)

t

20 40 60

20 4c 6C 80

FRACTION NUMBER

FIG. 2. Elution of acetylcholinesterase from affinity column saturated with enzyme with combination stepwise and continuous potassium chloride gradient. Elution profiles are described in Fig. 1. Fraction size is 11.0 to 11.4 ml. Crude enzyme with a total activity of 670 mmoles per min was passed through a column (1.5 X 30 cm) containing 42 ml of affinity-liganded gel. Nonre- tained protein (up to Fraction 4) included 191 mmoles per min (29%) of the applied enzyme, and Fractions 5 to 80 include the retained enzyme which is 445 mmoles per min (67%) of the applied enzyme. The KC1 gradient emphasizes the 0.2 M region where the enzyme is eluted.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Issuo of March 10. 1972 I’. L. Rosenberry, H. W. Chang, and Y. T. CherL 1561

Slimmary of afinity chromatography purification

The chromatographic procedure is described under “Methods.” The per cent yield is expressed relative to the applied crude enzyme solution, which had a specific activity’ of 25 to 32 and an

activity defined as 100%.

Ligand Specific activity Yield

% Di-(6.aminocaproyl)-PTA”, 520-575b 35-75

Di-(6-aminocaproyl)-PTAc. 560-610 65-75d

Mono-(6-aminocaproyl)-PTAe., .I 38c-420 4&50

Q The quantity of crude enzyme added (6.7 mmoles per min per pmole of ligand) was that observed to be just sufficient to saturate the Sepharose, which contained 1.61 pmoles of di-(6-aminocap-

royl)-PTA per ml of packed gel. Under these conditions the nonretained enzyme usually did not exceed 15% of the total applied enzyme. The wide range in the yield reflects recoveries

from both linear and nonlinear salt gradient elutions. Non- linear salt gradients gave generally better resolutions, and recoveries from these column runs were always in excess of 60%.

* The corresponding specific activity at an assay pH of 7.0 is 5.0% lower.

c Only two experiments have been done using gel containing 0.71 @mole of ligand per ml of packed gel; in both, crude enzyme was added (10.6 mmoles per min per pmole of ligand) in excess of the quantity calculated as described in Footnote a to saturate

the gel. In one run the nonretained enzyme was 45yo of that applied, a result expected from the capacity observed in Foot- note a. In the second run only 5% of the enzyme was not retained. Both runs used linear salt gradient elutions.

d The yield is based on retained enzyme. e The quantity of crude enzyme added (1.3 mmoles per min per

pmole of ligand) to Sepharose containing 5.8 pmoles of mono-(6- aminocaproyl)-PTA per ml of packed gel was 20 to 30% of that required to saturate the gel. Elut,ion was carried out with a nonlinear salt gradient.

The larger preparative affinity column purification shown in

Fig. 2 again uses the di-(Gaminocaproyl)-PTA gel, but here the elution of retained protein is accomplished with a nonlinear potassium chloride gradient and gives a much better resolution of enzyme from contaminating protein. A complete purification is not, achieved in this run, however, as the specific activities of eluted enzyme are somewhat variable and correspond to about 85yo of that of enzyme purified to the highest attainable specific activity. This column had been loaded with an excess of crude enzyme solution, and the enzyme activity capacity of the column was observed to be 6.7 mmoles per min per pmole of ligand. This corresponds to about 1 enzyme molecule bound per 400 ligand sites.

A summary of the results of affinity chromatography using di- (6.aminocaproyl)-PTA-Sepharose is given in Table III. These results were obtained with gels having either 0.71 or 1.61 pmoles of ligand per ml of packed Sepharose and will be shown below to indicate the isolation of essentially pure enzyme. Purification levels nearly as high were found by means of gels with a higher concentration of attached di-(6.aminocaproyl)-PTA; but such gels retain more contaminant protein, require higher concentrations of salt to elute the enzyme, and reduce yields because of their higher capacity to retain enzyme irreversibly.

The affinity chromatography results obtained with the use of

TABLE IV

Summary of standard chromatography purijication procedures

The chromatographic procedure is described under “Methods.”

step Procedure Specific activity” Yield* -

%

1 Extraction and ammonium 31 (30-32) /lo0 (defined) sulfate precipitation I

2 B-DEAE-cellulose 140 (lW225)/ 52 (44-62) 3 Concentration and dialy- 145 1 50 (48-51)

sis 4 Sephadex G-200 250 (190310) 43 (41-46) 5 Cellex-PC 410 (340-470) 25 (23-26) 6 DEAE-cellulosed 475 (460-490) 16 (15-18) 7 DEAE-Sephadexd 500 (480-51O)l 12 (12-13) 8 DEAE-cellulosed 520 (510-540) 9 (8-10)

Q The first number is the average over-all specific activity’ for the purification step, and the numbers in parentheses are the average range of specific activities pooled.

b The per cent yield is the over-all yield for all steps leading up to a given level of purification. The first number is the average

yield for a given procedure, and the numbers in parentheses are the range of yields observed in repet,itions of the procedure.

c Adjustment of the pH to 5.7 was crucial for this step. At pH 5.9 (5) less than 20% of the enzyme activity was retained on the column. A stepwise salt gradient elution (pH 6.0) yielded enzyme at 0.1 M NaCl, and the pH was readjusted to 7.0 to maintain enzyme stability.

d Elution accomplished with a continuous linear sodium chloride gradient at pH 7.0 which yielded enzyme at about 0.2 M

NaCl.

an alternate ligand, mono-(6.aminocaproyl)-PTA, are also included in Table III. Lower levels of purification are attained with this ligand, which does not extend as far from the Sepharose polysaccharide backbone as di-@aminocaproyl)-PTA. The salt gradient elution of retained protein contaminant,< from mono-(6.aminocaproyl)-PTA-Sepharose overlaps that of the enzyme and decreases the resolution and purification from that obtained with the di-(B-aminocaproyl)-PTA gel. This difference is consistent with the somewhat higher affinity of the enzyme for di-(6.aminocaproyl)-PTA-Sepharose shown in Table

II.

Standard Chromatography-The results obtained by standard chromatographic procedures are outlined in Table IV. Puri- fication levels higher than those reported previously (5, 6) were

observed in the first five steps, but the specific activities obtained in the last three steps are somewhat lower. The enzyme from Step 8 appears very close to being chromatographically pure; no protein is eluted in fractions separate from the active peak, and

the specific activities’ of fractions on the tails of this peak re- main above 450. The chromatography on B-DEAE-cellulose

(Step 2) revealed that a major component of the crude enzyme preparation was retained with greater affinity than acetylcholin- e&erase. Although at low levels of column loading (1.2 mmoles per mini per ml of packed resin) purification was not hindered, higher loading levels decreased both the purification and the yield. At, high loading levels (5.0 mmoles per min per ml of packed resin), this contaminant displaced up to 75c/, of t.he applied enzyme prior to the initiation of the salt gradient. The displaced enzyme could be rechromatographed on B-DEAE-cel-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1562 Aqinity Chromatography PurL’Jication of Acetylcholinesterase Vol. 247, No. 5

FIG. 3 (top left). Polyacrylamide gel electrophoresis of crude and purified preparations of acetylcholinesterase (E). The enzyme had been purified by affinity chromatography on di-(6 aminocaproyl)-PTA-Sepharose to a specific activity’ of 490. Gels were run at 7 volts per cm until the tracking dye reached the bottom of the gel, or about 8 hours. The applied samples, reading from the left, were (1) 8.6 ~g of purified E stained for esterase activity; (2) 8.6 rg of purified E stained for protein; (3) 86 pg of purified E stained for protein; (4) 50 pg of crude E stained for ester&se activity; (5) 50 pg of crude E stained for protein.

FIG. 4. (bottom l:fl). Polyacrylamide gel electrophoresis of purified acetylcholiiresterase at increasing sample loads. The enzyme had been purified as in Fig. 3 to a specific activity’ of 520. Gels were run at 2.5 ma per tube for 8 hours; the spacer gel was omitted. The applied samples, reading from the left, were (1) 8.6 fig; (2) 43 a; (3) 86 PR.

FIG. 5 (right’. Immunoelectrophoresis of acetylcholinesterase at various stages of purification. Antisera both to a crude en-

ulose to give specific activities up to 260 with only slight contamination from the tight bindirq component.

The elution volume correspondng to the peak acetylcholinesterase activity during chromatography of the partially purified enzyme on Sephadex G-200 was used to estimate the molecular weight of the enzyme (26). The observed value, based on three chromatographic runs, was 245,000 f 10,000.

Polyacrylamide Gel Electrophoresis-The homogeneity of enzyme purified on di-(6-aminocaproyl)-PTA-Sepharose (Table III) was determined both by polyacrylamide gel electrophoresis and by immunoelectrophoresis procedures. In Fig. 3 the gel electrophoresis patterns of enzyme purified to a specific activity of 490 are compared with those of the crude enzyme applied to

zyme extract (specific activity’ 2 to 3) and to partially purified enzyme (specific activity 192) which had been prepared previously (27) were applied to 40- to lOO-pg protein samples which had been subjected to electrophoresis as described under “Methods.” a, upper trough, antisera to the crude extract (Anti-C) ; well, enzyme purified on di-(6-aminocaproyl)-PTA-Sepharose to a specific activity of 520; lower trough, antisera to the partially purified enzyme (Anti-P). b, upper well, enzyme partially purified by standard chromatographic procedures with a specific activity of 340; trough, Anti-P; lower well, enzyme partially purified by standard chromatographic procedures with a specific activity of 330. (En- zyme in b was obtained from a fractionation on DEAE-Sephadex and was prepared by Dr. G. Lovinger.) c, upper frough, anti-C; upper well, crude enzyme extract of specific activity 2 to 3; lower trough, mixture of equal parts of Anti-C and Anti-P; lower well, enzyme purified first by standard chromatographic procedures and further run on mono-(6-aminocaproyl)-PTA-Sepharose with a specific activity of 550.

affinity chromatography. This enzyme had lost activity on storage from an original specific activity of 520 to the value of 490 obtained on the day of the electrophoresis run. No bands other than an active esterase band are observed when an 8.6q.q protein sample is run; however, at the higher 86qg protein load a faint diffuse band of inactive impurity becomes apparent which corresponds to 10% of the total applied protein, according to a microdensitometer trace analysis. Essentially all the protein in the purified sample passes through the space gel and enters the stacking gel. Gel electrophoresis patterns of purXied enzyme with a higher specific activity of 520 in Fig. 4 demon- strate that an impurity, here corresponding to 10% of the total applied protein, may only be observed with high protein load.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Issue of March 10, 1972 T. L. Rosenberr~y, H. W. Chary, and Y. T. Chen 1563

TABLE v

Erliirdion co~ficients hased on .several mclhods oj &sol&e prolein determ.ination

‘l’hc protein content of various samples of purified acctylcholin-

esterase having specific activities’ of 520 to 560 was determined by the following techniques, described in detail under ‘LMethods.” With the use of t,he corresponding absorbance at 280 nm of the

sample, the extinction coefficient was obtained. No differences in ext.inction coefficient were found among samples whose specific activity varied from 520 to 560 by any of the techniques.

.._~ ____

Protein determination Number of

determinations

/ -.--- 1. Analysis of nitrogen content by the

numas or micro-Kjeldahl method”. 2. Analysis of nit,rogen content by rrin -

hydrin micro-method*.

3. Analysis of nitrogen content, from the amino acid compositionb.

4. Differential refractometry.. 5. Dry weight

4

4

2 2

6

21.4 f 0.3

21.8 + 0.7

18.8 * 0.5 17.6 f 0.3

18.2 f 1.1

C The extinction coefi?cient at 280 nm is defined as the absorbance of a solution containing 10 mg of protein per ml and is listed

as the mean and standard error for t,he number of determinations sho\vn. The difference absorbance A(e:z ,,m - ez’$ “,J may also be used to estimate the protein concentration. The ratio of this difference extinction coefficient to & “,,, was very reproducible

for sa.mples highly purified either by affinity or by standard chromatographic procedures and was observed lo be 3.52 & 0.05. The corresponding ratio (E:,$ “,J&$ .,) was 2.1 + 0.1.

b Conversion of nitrogen to protein was tna.de assuming t,hat.

the protein contained 16.0% nitrogen.

Immunoelectrophoresis-Patterns obtained at various stages of enzyme purificalion are shown in Fig. 5. While antisera to crude enzyme measurably precipitated only protein impurities, antisera to partially purified enzyme precipitated both enzyme and certain impurities remaining at relatively high levels of purification. Only a single band is observed for enzyme purified to specific activities of 520 and 550, consistent with essentially homogeneous enzyme.

Profein Determination-A careful determination of the specific actirit,y of the homogeneous protein demonstrated above required that the spectrophotometric protein determination be directly related to an absolute protein analysis. Three different techniques for tneasuring absolute protein content were em- ployed, and the caorresponding spectrophotometric extinction coefficients obtained from each technique are listed in Table V. A significant difference in protein estimates is obtained between techniques measuring micro quantities of nitrogen as NH3 or St and those monitoring dry weight or refractive index. The nitrogen analysis from the amino acid composition data (Table VI) tends to support the lower extinction coefficient obtained from the dry weight and refractive index. It appears that micro determinat,ions of NH3 or NI are underestimating the protein content by 10 to 150/b. The difference could be real if the protein contained significantly less than the 16.0% nitrogen assumed or if contaminants containing little nitrogen were present. (An analysis of t,he purified protein sarnple for polysaccharides and their derivatives with free or potentially free reducing groups

TABLE VI

Amino acid composition of acetylcholinesterase

Analysis procedures are described under “Methods.”

Amino acid

Lysine. ................ 9.2 6.9 7.3 7.8 Histidine ..............

:

4.1 3.6 4.0 3.9 Ammonia. ............. 37.3 17.8 22.2 26 Arginine. .............. 8.1 9.3 9.1 8.8 Half-cystine (as cystei C

acid) ................ 3.1 2% Aspartic acid. ......... 21.8 21.4 23.6 22.3 Threonine ............. Serine. ................ :/

8.3 7.3 7.5 7.7 11.4 11.8 11.6 11.6

Glutamic acid. ........ .i 17.6 17.5 17.7 17.6 Proline. ............... .i 8.5 10.9 10.8 10.1 Glycine. ............... 15.3 14.4 14.6 14.8

Alanine. ............... 12.2 9.6 10.0 10.6 Valine. ................ 12 Methionine. ...........

:I 4.3 4.8 4.6

Isoleucine ...... ...... 6.0 6.6 6.7 6.4 Leucine ............... 13.6 15.1 15.2 14.6 Tyrosine .............. 5.7 6.7 6.2 Phenylalanine ......... 8.6 9.2 9.3 9.0 Tryptophan ........... 3.4d -

a Data are tabulated in terms of valine, which showed a constant mole per cent based on total amino acid content of 7.05 f

0.03. Twelve valine residues were used as a reference number to facilitate comparison with a previous report (6). The total amino acid residues reported thus correspond to a polypeptide

of molecular weight 20,100. Data appear as observed values uncorrected for any decomposition occurring during the 24-hour hydrolysis.

*The sealed hydrolysis tube contained oxygen allowing a cysteic acid determination.

c Samples subjected to performic acid oxidation according to the method of Hirs (28).

d Estimated spectrophotometrically (29).

- Residues per 1’2 valine residues’

Sample ATWage

I II III*

(19) did give a positive reaction equivalent to about 4 to 5% sugar. This could arise from carbohydrate contamination from the Sepharose itself.) However, the fact that the nitrogen analysis from the amino acid composition data is in reasonable agreement with the protein estimated from the dry weight or the refractive index does not support this explanation and seems to indicate that relatively less importance should be attached to the micro determinations of nitrogen as estimates of protein content. A value of 18.0 i 0.4 was assumed for the extinction coefficient at 280 nm.

Amino Acid Composition-Table VI shows the amino acid composition analysis obtained for enzyme purified on di-(6- aminocaproyl)-PTA-Sepharose. The average number of residues for each amino acid may be compared with values reported previously (6). Agreement within 1 residue is observed in all cases except proline, where the current analyses show 4 fewer residues per an arbitrary 20,100 molecular weight unit.

Active Site Titrations-The normality of enzyme catalytic sites was determined bot,h by direct spectrofluorometric titra-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1564 Afinity Chromatography Puri$catim of Acetylcholinesterase Vol. 247, x0. 5

TABLE VII

Stoichiometric titration of acetylcholinesterase active sites and calculation of number of active sites per molecule

The titration conditions are described under “Methods,” and the calculations are explained under “Results.” In each case enzyme was purified on di-(G-aminocaproyl)-PTA-Sepharose.

Ligand Turnover number Apparent sites Corrected sites per molecule” per molecule*

m&z-’

M7C”. :.:

7.66 f 0.25 X lo5 2.85 zt 0.13 3.44 f 0.21 M7CY.. 8.03 f 0.26 X 105 3.15 + 0.15 3.28 f 0.20 Bis-Q8. 7.63 z!c 0.59 X 106 2.92 f 0.24 3.46 f 0.31

Bis-Qt.... 8.80 f 0.52 X lo5 2.81 f 0.21 2.98 f 0.25

a Assumes the protein solution is 100% active acetylcholinesterase and that the enzyme molecular weight is 260,000 f

10,000. * Corrects for contaminant or inact,ive protein by normalizing

the apparent value at the observed specific activity to the value corresponding to a pure enzyme solution with a specific activity’

of 610 f 15. c Enzyme specific activity was 505 f 15; the protein concentra-

tion was 14.4 f 0.19 I.rg per ml; and the observed normality was

1.58 zt 0.03 x lo-’ N.

d Enzyme specific activity was 585 f 10; the protein concentration was 25.9 f 0.55 fig per ml; and the observed normality

was 3.14 X!Z 0.05 x lo-’ N.

6 Enzyme specific activity was 515 + 15; the protein concentra-

tion was 917 f 27 pg per ml; and the observed normality was 103.1 f 6.9 X 1O-7 N.

f Enzyme specific activity was 575 f 15; the protein concerl-

tration was 950 f 19 pg per ml; and the observed normality was

102.6 f 6.1 X lo+ N.

tion with the carbamoylating agent M7C and by equilibrium dialysis with l&-Q as described under “Methods.” From the

normality one may calculate both the turnover member’ for

enzyme-catalyzed acetylcholine hydrolysis and, if the molarit

of the enzyme solution can be established, the number of enzyme catalytic sites per molecule. To determine the enzyme mo- larity one must know both the mg of active enzyme per ml and the enzyme molecular weight. In this study the molecular weight was assumed to be 260,000, consistent with the gel filtra- tion value reported above and with values measured by sedi-

mentation equilibrium (30, 31). The total protein concentra-

tion was determined spectrophotometrically; and the active

enzyme concentration was calculated from the specific activity of the solution, assuming pure enzyme to have a specific activity of 610. This sequence of calculations is shown in Table VII and leads to an average value of 3.29 f 0.11 active sites per

molecule of enzyme. An attractive feature of the use of Lf7C: as a titrating agent is

the fact that the entire course of the carbamoylation reaction may be observed directly. Hence the following kinetic param-

eters, rigorously defined in a previous paper (13), were observed

for the titration reaction: the specific carbamoylation rate constant, kz, was 5.1 + 0.8 min-‘; the specific decarbamoylation

4 The enzyme turnover number per site, obtained directly from the enzyme activity and normality, is given in the units mmoles of acetylcholine hydrolyzed per min per meq of enzyme. Note that this is turnover under the assay conditions, in which the acetylcholine concentration is 2.7 mM. Turnover at the extrap- olated enzyme maximum velocity (13) is about 4% higher.

fl

0 IO ?O 3.0

B FIG. 6. Equilibrium dialysis titration of acetylcholinesterase

with Bis-Q displayed on a Scatchard plot. The procedure is out,- lined under “Methods.” I,, pmolar concentration of ligand; B, moles of ligand bound per mole of active enzyme. The enzyme had a specific activity’ of 575 (see Table VII, Footnote f),

rate constant, kt, was 0.033 & 0.010 milr’; and the pre-carbamoylation (reversible) titrant dissociation constant, h’,., was 6.7 + 1.8 PM. These values are identical, within experimental error, with those reported previously for enzyme of specific activity 70 (13) and thus support the conclusion that the specific catalytic rate constants for acetylcholinesterase are not altered by affinity chromatography.

A typical result from the equilibrium t,itration of acetylcholinesterase with Bis-Q is shown in Fig. 6. The average dissociation constant, KD, observed for the complex of Bis-Q with enzyme was 5.4 f 0.5 FM, and complex formation was completely reversible. This K. agrees with the competitive inhibition C’OII- stant Kr obtained with Es-Q (Table II) and supports the conclusion that the Bis-Q binding observed occurs at the enzyme

active site.

A specific activity’ of 660 for acetylcholinesterase highly purified by the standard chromatographic procedures (Table IV) was observed by Kremzner and Wilson (5). However, relative ultraviolet absorption value ,s given for t,his preparation (e&b ,,rn = 22.9; A (E;; ,,m - &J/@, ,111, = 2.78; &,,,m/&, “1’1

= 1.67 (7)) do not agree with the values found here (Table V). I f the assay conditions and est’inction coefficient assumed here are used to recalculate the reported specific activity of 660, a value of 550 is obtained, in good agreement with the maximum value of 540 observed for the standard purification procedures here, but slightly less than that obtained with affinity chromatography. Leuzinger and Baker (6) report a specific activity of 730 for enzyme purified by the standard chromatographic procedures; but a more recent report by Leu- zinger (32) finds the same specific activity by means of a higher assay pH and a higher protein extinction coefficient, conditions under which one would expect the specific activity to differ by 25 to 30%.

The difficulties inherent to obtaining the specific activity of purified homogeneous enzyme may not be trivial, apart from the


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ISSUES of March 10, 1972 T. L. Rosenberm, H. W.

fat-t that there may be esperimental bias toward reporting higher specific activities. Highly purified enzyme has been reported t.o be somewhat unstable (5), and we have observed that the stability of enzyme above a specific activity of 500 varies greatly from preparation to preparation. In general, the stability is best preserved in a frozen solution of high enzyme concentration. A second inherent difficulty, which was illustrated urider “Results,” is the method of protein determination used (Ta- ble V) .

These difficulties are reflected in attempts to determine the stoichiometry of enzyme-active sites. A simultaneous determination of the turnover number per site4 is highly recorn- mended, because this number is independent both of the protein concentration and the enzyme specific activity. Turnover numbers reported in Table VII agree with a series of turnover numbers for acetylcholinesterase obtained with a wide variety of titrants tabulated previously (13), a condition which we feel should be met for all active site stoichiometry determinations. Furthermore, it was shown previously that if a site turnover number, an enzyme molec*ular weight, and a specific activity for pure enzyme are assumed, it necessarily follows that a unique solution for the number of active sites per molecule may be found (13). This solution for the data in Table VII is 3.3 & 0.1 active sites per molecule. Similar logic applied to a recent report of two active sites per molecule (32) would require a site turnover number about twice that indicated in Table VII, in disagreement with all previous turnover number determinations. Previous reports of four active sites per molecule (7, 13) were based on higher reported specific activities for pure enzyme of 660 and 730, values which have not been confirmed in this study.

A value of three active sites per molecule is somewhat dis- comforting in view of t’he symmetry considerations that seem to dictate an even number of active sites for oligomeric: enzymes (33). The most uncaertain data entering into this calculation is the specific activity of pure enzyme. If specific activities higher than 610 can be demonstrated, the number of active sites per molecule would increase. Since we have not been able to demon- strate such higher activities, the possibility of three active sites per molecule must be seriously considered. In this regard, recent reports of subunit molecular weights corresponding to sis subunits per molecule (31), rather than the four previously reported (30), are of interest, because three active sites per hexamer would indicate higher symmetry than three active sites per tetramer.

Acknowledgments-We express sincere appreciation to Pro- fessor David Nachmansohn for his advice and encouragement during the course of this study. We also gratefully acknowl- edge the considerable contributions made by Dr. Valee Haris- dangkul to the immunoelectrophoresisand nitrogen determination

Chang, and Y. T. Chen 1565

results and by Dr. William Poillon to the differential refractometry measurement.

REFERENCES

1.

2.

3. 4. 5.

6.

7.

8.

9.

10.

11.

12. 13.

14.

15. 16. 17.

18. 19.

NACHMANSOHN, II., AND LEDERER, E. (1939) Bull. Sot. Chim. Biol. 21, 797

ROTHERRERG, M. A., .IND NACHMANSOHN, I>. (1947) J. Biol. Chem. 168, 223

L~WLER, H. C. (1959) J. Biol. Chem. 234, 799 LAWLER, H. C. (1963) J. Riol. Chem. 238, 132 KREMZNER, L. T., AND WILSON, I. B. (1963) J. Biol. Chem.

238. 1714 LEUZ~NGER, W., .%ND BAKER, A. L. (1967) Proc. Nat. Acad.

Sci. U. S. A. 67, 446 KREMZNER, L. T., AND WILSON, I. B. (1964) Biochemistry 3,

1902 LEUZINGER, W., BAKER, A. L., AND CAUVIN, E. (1968 Proc.

Nat. Acad. Sci. U. S. A. 69,620 BERMAN, J. Il., AND YOUNG, M. (1971) Proc. Nat. Acad. Sci.

U. S. A. 68, 395 KALDERON, N., SILMAN, I., BLUMBERG, S., AND DUDAI, Y.

(1970) Biochim. Biophys. Acta 207, 560 CUATRECASAS, P., WILCHFX, M., AND ANFINSIBN, C. B. (1968)

Proc. Nat. Acad. Sci. c’. S. A. 61, 636 CUATRECAS~S, P. (1970) J. Biol. Chem. 246, 3059 ROSENBERRY, T. L., END BERNHARD, S. A. (1971) Biochemis-

try 10, 4114 BARTELS, E., W.ISSERM.INN, N. H., AND ERLANGER, B. F.

(1971) Proc. Nat. Acad. Sci. U. S. A. 68. 1820 BENNETT, G. M., AND WILLIS, G. H. (1929) J. Chem. Sot. 264 STEDMAN, E. (1927) J. Chem. Sot. 1905 SCHIFFM~N, G., K~B~T, E. A., AND THOMPSON, W. (1964) Bio-

chemistry 3, 113

20.

BABUL, J., AND STBLL~BGEN, E. (1969) Anal. Riochem. 28,216 DUBOIS, M., GILLISS, K. A., HAMILTON, J. K., R~;.:DEHs, P. A.,

AND SMITH, F. (1956) Anal. Chem. 28, 350 MOORE, S., AND STEIN, W. H. (1957) in 8. P. COLOWICK AND

N. 0. KAPLAN (Editors), Methods in enzymology, Vol. 6, p. 819, Academic Press, New York

21. HIGMAN, H. B., PODLESKI, T. R., AND BARTELS, E. (1964) Biochim. Biophus. Acta 79. 138

22. 23.

BRAY, G. A. (1‘960) Anal. Biochem. 1, 279 GRABAR, P., AND WILLIAMS, C. A. (1953) Biochim. Biophys.

Acta 10, 193 24. URIEL, J. (1963) Ann. N. Y. Acad. Sci. 103, 956 25. DAVIS, B. J. (1964) Ann. N. Y. Acad. Sci. 121, 404 26. ANDREWS, P. (1965) Biochem. J. 96.595 27. HARISDA~GKUL, V., LOVINGER, G.,~AND KAI~AT, E. A. (1970)

Fed. Proc. 29, 376 28. HIRS, C. H. W. (1956) J. Biol. Chem. 219, 611 29. BEAVEN, G. H., AND HOLIDAY, E. R. (1952) in M. L. ANSON.

K. BAILEY, AND J. T. EDSAL; (Editors), Advances in protein chemistru, Vol. 7. v. 319. Academic Press. New York

30.

31.

LIOUZINGE;; W., GOLDBERG, M., AND CALVIN, E. (1969) J. Mol. Biol. 40, 217

MILLAR, D. B., .&ND (+R.%FIUS, M. A. (1970) Fed. Eur. Biochem. Sot. Lett. 12, 61

32. 33.

LEUZINGER, W. (1971) Biochem. J. 123, 139 MONOD, J., WYMIN, J., .IND CH~NGEUX, J.-P. (1965) J. Mol.

Biol. 12, 88


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Terrone L. Rosenberry, Hai Won Chang and Yueh T. ChenDetermination of Active Site Stoichiometry

Purification of Acetylcholinesterase by Affinity Chromatography and

1972, 247:1555-1565.J. Biol. Chem.

http://www.jbc.org/content/247/5/1555Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/247/5/1555.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/247/5/1555

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;247/5/1555&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/247/5/1555

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=247/5/1555&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/247/5/1555

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/247/5/1555.full.html#ref-list-1

http://www.jbc.org/

purification of acetylcholinesterase by affinity chromatography and ... · purification of...

Documents