Eur. J. Biochcm. 37, 469-480 (1975)

Multiple Molecular Forms of Purified Human Erythrocyte Acetylcholinesterase Peter OTT, Bruno JENNY, and Urs BRODBECK Medizinisch-chemisches Institut der Universitiit. Bern (Received March 17iMay 30,1975)

1. Human erythrocyte acetylcholinesterase was solubilized by Triton X-100 and purified by affinity chromatography to a specific activity of 3800 IU/mg of protein. The yield of the purified enzyme was 25 - 45 2. Gel filtration on Sepharose 4-B in the presence of Triton X-100 revealed one peak of enzyme activity with a Stokes’ radius of 8.7 nm. Density gradient centrifugation in 0.1 % Triton X-100 showed one peak of enzyme activity with an S4 value of 6.3s. 3. Isoelectric focusing in Triton X-100 resolved the enzyme into five molecular forms with isoelectric points of 4.55, 4.68, 4.81, 4.98 and 5.18. Upon incubation with neuraminidase the enzyme activity in the first four forms was decreased with a concommitant increase in activity in the form with the higher isoelectric point. 4. After removal of excess Triton X-100 on Bio-Gel HTP, polyacrylamide gel electrophoresis showed seven bands of protein and corresponding bands of enzyme activity. Density gradient centrifugation of the detergent-depleted enzyme at high ionic strength revealed five multiple molecular forms with S4 values of 6.3 S, 10.2 S, 12.2 S, 14.2 S and 16.3 S. At low ionic strength, higher aggregates were observed in addition to the other forms. Dodecylsulfate-polyacrylamide gel electrophoresis gave one subunit only with an apparent molecular weight of 80000. 5. These results suggest that human erythrocyte acetylcholinesterase, solubilized by Triton X-100, exists in various forms differing in net charge but of apparently similar molecular dimensions. After removal of the detergent, forms with different molecular sizes are observed.

x.

Previously, evidence for the existence of multiple forms of acetylcholinesterase from human erythrocytes has accumulated. Shafai and Cortner resolved the enzyme by DEAE-Sephadex column chromatography into two components of apparently similar molecular weight [I] and postulated that Triton X-100 solubilized acetylcholinesterase is a hybrid dimer composed of two unlike subunits ( M and p ) of equal size [2]. Wright and Plummer solubilized the enzyme by the action of detergent (Triton X-100) and that of high ionic strength (1.2 M KCI) [3]. They found evidence by polyacrylamide gel electrophoresis for the existence of up to six multiple molecular forms with apparent molecular weights ranging from 66000- 562000 [4]. According to the suggestion of Whittaker and Charlier ~~

[5], human erythrocyte acetylcholinesterase exists in various states of aggregation with a monomer unit of 75000 molecular weight, in close agreement to an earlier reported figure [6]. Several workers have described the partial purification of this enzyme [1,4,5,7-121. Except for the bovine erythrocyte enzyme that was obtained by affinity chromatography in a highly purified form [13,14], no other red cell preparation of similar purity is described. This paper reports on the purification of human erythrocyte acetylcholinesterase and on some of the properties of its multiple molecular forms. Preliminary accounts of this work have been presented [15,16].

~

Ahbret1iarion.c. Dip-F, diisopropyl-phosphorofluoridate; dodecylsulfate, sodium dodecylsulfate; Nbs,, 5,5-dithio-bi@nitrobenzoic acid). Enzymes. Acetylcholinesterase (EC 3.1.1.7) ; cholinesterase (EC 3.1.I .8) ; neurdminidase (EC 3.2.1.18) ; catdlase (EC 1.11.1.6).

Eur. J. Biochem. 57 (1975)

MATERIALS AND METHODS Enzymes Human erythrocyte sediments from one-day-old collections of blood were obtained from the central

410

Molecular Forms of Human Erythrocyte Acctylcholinesterase

blood bank of the Swiss Red Cross and were used as starting material for the purification of acetylcholinesterase. After removing the buffy coat, the red cells were routinely washed twice by the addition of equal amounts of a solution of 0.9% NaCI. Erythrocyte ghosts were prepared according to the procedure of Dodge et al. [17]). Acetylcholinesterase was solubilized in a solution of 1.0 7; Triton X-100. Neuraminidase, type VI, purified from Clostridium perfringens and catalase were purchased from Sigma Chemical Company (St. Louis, Mo., U.S.A.). Neuraminidase, purified from Vibrio cholrrar, was a product of Serva, Heidelberg. Bovine serum albumin was obtained from Poviet Producten (Amsterdam). ‘251-labelled thyroglobulin was a gift of Dr H. Kohler, Inselspital (Bern). Triton X-100 was from Rohm & Haas (Philadelphia) and Ampholine from LKB Producter AB (Bromma). All other chemicals were standard commercial products obtained either from Fluka AG (Buchs) or from Merck AG (Darmstadt). BioGel P-10 and BioGel HTP were obtained from Bio Rad Laboratories (Richmond), Sepharose 4 B and polyacrylamide gradient gels PAA 4/30 from Pharinacia AB (Stockholm).

Enzyme Assaj

Acetylcholinesterase activity was estimated according to the method of Ellman et a/. [18]. Unless otherwise specified the assay mixture contained 1 mM acetylthiocholine, 0.125 mM Nbs, and 0.05 Triton X-100 in a total volume of 3.0 ml 100 mM sodium phosphate buffer pH 7.4. The reaction was followed spectrophotoinetrically by the increase in absorbance at 41 2 nm on a Beckman DB-G spectrophotometer equipped with a W W recorder 3002. Catalase was determined spectrophotometrically according to Aebi ~91.

+

was termined by linking trimethyl(m-aminopheny1)ammonium chloride hydrochloride to the extended side chain.

Enzyme Purification Step I : Ajfinity Chromatography. Triton X-100 solubilized acetylcholinesterase was purified essentially as described by Berman [14]. The batch procedure was used to adsorb the enzyme onto the affinity gel. When 60-70% of the total activity was adsorbed onto the gel, it was packed into a column and protein, unspecifically bound to the resin, was eluted with a buffer of 20 mM Tris-HC1 pH 7.4, containing 1 M NaCl and 1 ”/, Triton X-100. Acetylcholinesterase was then eluted by the addition of 20 mM decamethonium to the same buffer. Fractions, 5 ml each, were collected. The enzyme was dialyzed for three days against a buffer of 20 mM Tris-HCI pH 7.4, containing 1 M NaCl and 0.1 % Triton X-100. Step II: Removal of Triton X-100. The enzyme was passed through a column (2 x 25 cm) of BioGel P-10 equilibrated and eluted with a buffer of 10 mM sodium phosphate pH 6.0. After pooling, the enzyme was adsorbed onto a column (1 x 10 cm) of BioGel HTP. The column was washed with the same buffer and the decrease in Triton X-100 in the eluant fractions was monitored at 280 nm. When the absorbance was below 0.01, the enzyme was desorbed with 200 mM sodium phosphate buffer pH 7.4.

Concen trat ion of Pro tr in Solution Protein solutions were concentrated on a stirred Amicon ultrafiltration cell using a Diaflo PM-10 ultrafilter. Isoelectric. Focusing

Protein Concentration Protein was estimated according to the method of Lowry et al. [20] using bovine serum albumin as standard. Yreparatiotz qf’A,f;fi’nityGel Sepharose 4 B was treated essentially as described by Cuatrecasas and Anfinsen [30] and by Berman and Young [13] except that hexamethylene diamine was used instead of 3,3’-diaminodipropylamine[31]. The succinylation steps were carried out at pH 8.0. After three cycles of coupling (hexamethylenediamine followed by succinic anhydride), the chain elongation

lsoelectric focusing was carried out according to Vesterberg and Svenson [21] in a Uniphor column electrophoresis system (LKB Prpducter AB, Bromma) with pH gradients of pH 3- 10 and 4-6. Solutions of 1 yi Ampholine were used and as support a density gradient was prepared from 0 to 45% sucrose. The enzyme was dialyzed against a ininium of 100 volumes of 1 ”.’, glycine buffer solution pH 7.0 and applied to the column as 2-ml samples. To prevent the formation of Joule’s heat the voltage was increased gradually over a period of 12- 18 h, keeping the total current to 2.5 mA or lower. After reaching 1000 V, the electrofocusing was continued for 2 days. Thereafter the column was emptied at a flow rate of 2 mlhnin. L ~ i r J. .

Biochcm 57 (1975)

47 1

P. Ott, B. Jenny, and U . Brodbeck

Treatment qf Acet~lclzolinesterasewith Neuraminidase

Staining Procedures

Sialic acid residues were removed from acetylcholinesterase essentially according to the procedure of Carlson and Svensmark [22]. Acetylcholinesterase was diluted with an equal volume of 0.1 M sodium acetate buffer pH 5.5, containing 0.2 M NaCl, 20 mM CaCI,, either in presence or in absence of 0.1 Triton X-100. To 200 units of enzyme activity 1 mg of purified neuraminidase was added. The solution was incubated at 30 “C for 24 to 72 h. Before isoelectric focusing the mixture was dialyzed as described above.

In dodecylsulfate-polyacrylamide gel electrophoresis, protein was stained using Coomassie brilliant blue R-250. The gels were destained in a solution of acetic acid/methanol/water ( 3 5 : 10 :75, by vol.). In polyacrylamide gradient gel electrophoresis protein was stained with 0.7% amidoblack 10B in 7 % acetic acid. The gels were destained in 7 % acetic acid. Carbohydrates were visualized with the periodateSchiff stain according to the method of Zacharius et al. [26] and enzyme activity according to Uriel [27] and Karnovsky and Roots [28].

Density Gradient Centrijugation

Density gradient centrifugation on 5 to 30 ”/, (wp) linear sucrose gradients were performed according to Martin and [23i. The gradients were prepared with a Beckman syringe gradient mixer. The enzyme samples were dialyzed for 14 h against the same buffer that was used in preparing the sucrose gradients. The marker proteins of known sedimentation values were then added to the enzyme solution and 0.1 ml of this mixture was layered on top of the gradients. Centrifugation was performed at 4 ’C and 40000 rev./min for 15 h in a MSE 6 x 14 ml titanium swing-out rotor on a MSE superspeed 65 ultracentrifuge. The tubes were emptied with a peristaltic pump at a flow rate of 0.5 ml/min and fractions of 0.25 ml were collected. ElectrophoreJ is

Dodecylsulfate-polyacrylamide slab gel electrophoresis was performed using the Ortec equipment. Gels and enzyme samples were prepared according to the method of Weber and Osborn [24]. After application of the samples, a layer of 5 % polyacrylamide gel in 10 mM sodium phosphate buffer pH 7.1 was added on top. Electrophoresis was carried out at 8 0 m A per slab gel using an Ortec 4100 pulsed constant power supply. The front dye was bromphenol blue. Bovine serum albumin and egg albumin served as marker proteins for molecular weight determinations. Despite the presence of mercaptoethanol, small amounts of aggregates of serum albumin [25] and egg albumin are observed in dodecylsulfate gel electrophoresis, the molecular weights of which can be used for calibration. Polyacrylamide gradient gel electrophoresis was performed using the Pharmacia equipment. Electrophoresis was carried out at 125 V for 15 h in 80 mM Tris-borate buffer, pH 8.35, containing 0.01 EDTA. The buffer temperature was kept constant at 12 ‘C. Eur. J . Biochem. 57 (1975)

Gel Filtration

Gel filtration was carried out with a 2x95-cm column on Sepharose 4B. The Stokes’ radii were determined as described by Siegel and Monty [29],

RESULTS Enzyme Purification

Human erythrocyte acetylcholinesterase was solubilized by 1 % Triton X-100 and purified by affinity chromatography either from isolated red cell ghosts or directly from a hemolysate containing the same amount of detergent. Starting from the isolated ghosts, the yield in purified enzyme exceeds the one obtained from the hemolysate. At this stage of purification, neither the specific activity of the enzyme nor its relative purity could accurately be assessed as Triton X-100 strongly interfered with the Lowry protein determination as well as with polyacrylamide gel electrophoresis. Removal of Triton X-100 was attempted in several ways. As this detergent forms stable micelles with an apparent molecular weight of approximately 200000, neither gel filtration on Sepharose 4 B nor prolonged dialysis proved to be successful. The use of the hydrophobic amberlite XAD-2 resin to remove this detergent from aqueous protein-containing solutions was suggested by Holloway [32] but proved to be unsuccessful in the case of human erythrocyte acetylcholinesterase. The enzyme together with Triton X-100 was quantitatively adsorbed onto the resin from which it could not be eluted in an active form. By chromatography of the enzyme on BioGel HTP, however, excess Triton X-100 could be removed. The enzyme preparation thus obtained was suitable for protein determination as well as for polyacrylamide gel electrophoresis. A summary of the results from the enzyme purification is listed in Table 1.

Molecular Forms of Human Erythrocyte Acetylcholinesterase

412 Table 1. Purijicution of lzunzan erythrocj,te ucer~~lcholinesteruse Fraction

Volume -

Hemolysate After affinity column After BioGel HTP column

Total protein

Total activity

ml

"g

5 900

53 1 000

~

~

-

Recovery of activity

Specific activity

Puritication

pmol/inin

70

pmolmin-' mg-'

-fold

11 200

100

0 021 (4350)

(207 000)

3 808

181000

-

~~

13

(1.09)

4 74s

42

16 X

1.019

3881

34.6

"

5

15

10

Stokes radius (nm) t

; *, ,

4 II

I

, ,L

Elution volume ( m l )

Fig. 1, Gel filtration on Sepharose 4 B ofpurijieed human erjthrocjte acetjlcholinesterase. Enzyme (2 ml) containing 50 IU/ml was applied onto a column (2 x 95 cm) equilibrated and developed with20 mM Tris-HC1 buffer, p H 7.4 containing0.l yi Triton X-100. The flow rate was 10ml:h Blue Dextran 2000 (----)was used to deterand fractions, 2.5 ml each, were collected and assayed for acetylcholinesterase activity (0-0). mine the void volume (V,) and catalase, ferritin, fibrinogen and hemocyanin served as markers to calibrate the column. Inset: Relation of the Stokes' radius of acetylcholinesterase in 0.1 Triton X-100 (indicated by the arrow) to the Stokes' radii of the marker proteins

u',

Properties of Puvijkd Acetylcholinesterasr in the Presence of Triton X-100 Gel filtration through Sepharose 4 B gave one peak of enzyme activity (Fig. 1). The tailing of enzyme activity in the elution profile indicated the possible presence of lower-molecular-weight forms of acetylcholinesterase. The tailing was more pronounced when either a sample containing Triton X-100 was chromatographed through a detergent-free column, or when the Triton X-100 content of the sample applied was higher than the detergent concentration of the eluting buffer. Such behaviour indicated a possible artefact. Consequently samples from the tail end of the elution profile were rechromatographed through Sepharose 4 B in presence of 0.1 % Triton X-100 without prior concentration of the enzyme solution. The elution volume after rechromatography always decreased to the elution volume shown in Fig. 1, regardless of the

position of the sample in the tail end portion of the first chromatography. Upon density gradient centrifugation one form of enzyme with a sedimentation coefficient of 6.3 S (Fig. 2) was obtained. Isoelectric focusing, however, resolved the enzyme into five molecular forms with isoelectric points of 4.55, 4.68, 4.81, 4.98 and 5.18 (Fig. 3 A). Reelectrofocusing under the same condition of the peak fractions showed that each form could be obtained individually (Fig. 3 B - F). Acetylthiocholine was the preferred substrate for all of them and butyrylthiocholine was hydrolyzed at a rate of about 1.5 % of that of acetylthiocholine. The K , values for acetylthiocholine and the constant Ki for the inhibition of the enzyme by excess Dip-F, determined according to the method of Aldridge [35] are listed in Table 2. The decrease in activity upon inhibition by excess Dip-F follows as expected pseudofirst-order kinetics for the individual enzyme forms. The rates of inhibition of the unresolved enzyme, Eur. J. Biochem. 57 (1975)

473

P. Ott, B. Jenny, and U. Brodbeck

10

0

10

30

20

40

Fraction number

Fig. 2. Sucrose density gradient centrijugation of ucet~lcholine.sterusein presence of’O.1 Triton X-100. Centrifugation was performed as described in Materials and Methods in 20 mM Tris-HCI buffer, pH 7.4, containing 0.1 ”/: Triton X-100. Enzyme, 100 pl containing 15 IU. were applied together with 0.2 ing of catalase. The collected fractions were assayed for acetylcholinesterase (AChE, 0-0) and catalase activity (A-A)

A

E

1 2 3 4 5

h

pH 4’98

Fraction numbei

Fig. 3. Isoelectric ,focusing of acet~lcholi~zesteruse in 0.1 ”/, 7riton X-100. Isoelectric focusing was carried out as described in Materials and Methods. Enzyme activity (0-0) and p H value (A--A) were determined in each fraction (A). The peak fractions were pooled and from each pool equal amounts of enzyme activity were subjected to reelectrofocusing under the same conditions. (B - F) Individual patterns after reelectrofocusing of pools 1- 5 respectively. Enzyme units shown in Fig. B - F represent a ten-fold magnification of measured activity Eur. J. Biochem. 57 (1975)

474

Molecular Forms o f Human Erythrocyte Acetylcholinesterase

however, was faster during the first 6 min than thereafter, yielding a biphasic semilogarithmic plot. Since the enzyme routinely was purified from a mixture of red cells with blood groups 0,A, B and AB, the possibility existed, that the occurrence of multiple forms arose from a combination of genetic variant enzymes. Consequently acetylcholinesterase was isolated from red cells of two individual donors each, of the blood groups 0, A, B and AB. The activity profile of the 8 samples did not differ significantly with respect to the individual isoelectric points or to the relative amounts of each enzyme form present.

Table 2. C'ornparisori of kinetic parurnric'rs qf' acctjlc.lioli~iester.ase hrfbw and afier separution h j isoelec.tr.icfocwsi~g Enzyme

lsoelectric point

Bcforc separation

Pool 1 2 3 4 5 ~

4.55 4.68 4.x1 4.98 5.18

K,,, for acetyl thiocholine

Ratio of enzyme activity"

lo4 x K , for Dip-F

25

58

3.62b 2.18"

22 15 23 16 n. d."

59 56 58 61 n. d.d

2.33 2.27 2.45 3.00 3.56

Acetylcholinesterase purified from a mixture of human red cells was incubated with neuraminidases from CI. per-Jiingens and from I/. clzolerae for times ranging from 24 to 72 h. This treatment reduced the amount of enzyme activity present in the molecular forms with isoelectric points of 4.55 to 4.98 and increased concommitantly the enzyme activity in the form with an isoelectric point of 5.18. In absence of 0.1 < ; Triton X-100 the conversion to the species with the highest isoelectric point seemed to proceed at a faster rate than in presence of detergent. Losscs of enzyme activity up to 90 and more were, however, encountered during prolonged incubation of acetylcholinesterase at pH 5.5 in the absence of the detergent. A solution of 0.1 Triton X-100 effectively protccted the enzyme from denaturation at pH 5.5. As reviewed by Herz and Kaplan [33], acetylcholinesterase activity is consistently reduced in red cells of patients with paroxysmal nocturnal hemoglobinuria and a change in the properties of this enzyme has been reported [34]. Consequently blood from one patient carrying this rare disease was processed similarly to normal blood. Although acetylcholinesterase activity was reduced, no change in the distribution among the five different forms could be detected. Proper-t ies of the A cetj.lcho1i11es t erase Depleted of Triton X-I00

~~

Activity towards acetylthiocholiiie,activity towards butyrylthiocholine. " For the first 6 inin. After 6 inin. n. d. = not deterinincd. "

Upon sucrose density gradient centrifugation in a high ionic strength medium, the enzyme depleted of Triton X-100 w'as resolved into several active forms differing in their sedimentation coefficients (Fig. 4).

I 1

u)

u)

N

m.

N

0 d

t 10

22

30

40

50

0

Fraction number

Fig. 4. Drrr.\ity grotlirtir ( , ( ' i i t l . i f u ~ ~ ~of/ i ~c licr. ~ , ~ ~ , / ~ ~ / ~ o l i n c ~ .tkpkircl r / c ~ r t r . \ cof , Ij.ilo/i X-100. Centrifugation \\:I\ pcrl'ormed as described in Materials and Methods in 20 m M Tris-HC'I buKer, pH 7.4 i n absence of Trilon X-300. Enzymc. 100 wl containing 30 I U , was applied together %it11 0.2 ing of catalase. Thc collected fractions mcrc assayed for acetylclio1inester;ise (.---o)

and catalasc activity (A

A)

Eur. J . Biochem. 57 (1975)

415

P. Ott. B. Jenny. and U. Brodbeck A

0.4 0.3 0.2 0.1

0 Sq

-E

2

14.85

E

0.4

.

50.3

-- 0.2 2.

.->

0 0

10

20

30

40

Fraction number

m

5.

0.1

ti

Fig. 5. Density padient centrifugutioiz of ncrt~lc/io~inf~.stc.rcr.ve depleted of Triton X-100 mid it1 presrrlce of I ‘$ cc-methyl D-tilurlnosidf’. Centrifugation was performed as described in Materials and Methods. The sucrose gradients were made in 20 inM Tris-HCI buffer, pH 7.4, containing 1 2)a-methyl D-mannoside. Acetylcholincsterase ,).-.( 100 pl containing 200 IU, were applied together with catabase (A A) and 12’ I-labelled thyroglobulin (U---O). Purified human erythrocyte acctylcholinesterasc is shown under (A). The peak fraction were pooled (6- 1) and froin each pool equal ainoinits of enzyme activity were taken for recentrifugation under identical conditions. (B - C,) Individual patterns after recentrifugation of pools 6- 1 respectively. Numbers refer to s, values oTacetylcholir1esterasc

0 0.4

sq

= 10.9s

F

G

0.3

0.2

~~~~~

When the ionic strength was decreased to 0.02, heavier aggregates were also observed, moving to the bottom of the sucrose gradient. When the depleted enzyme was kept at 4 :‘C for periods up to 4 months, no changes in the distribution pattern of the 10- 16-S forms were observed. However, a significant decrease in the 6.3-S form with a concommitant formation of a 7.3-S form was obtained. Reincubation of the detergent-depleted acetylcholinesterase in 0.1 Triton X-100 apparently disaggregated the enzyme, yielding an s,-value of 6.3 S equal to the one obtained before removing Triton X-100. When a micellar solution of TritonX-100 was subjected to the same treatment, the micelles, as expected, floated on top of the sucrose gradient. The possibility thus existed that acetylcholinesterase in the presence of Triton X-100 yields a complex of artificially low sedimentation coefficient. To check this hypothesis, the elongated oligomeric forms of eel acetylcholinesterase with known sedimentation coefficients of 14 S and 18 S were isolated in the presence and absence of Triton X-100 and subjected to sucrose density gradient centrifugation in media with and without Triton X-100. Regardless of the mode of Eur. J. Biochein. 57 (1975)

0.1

0 Fraction number

isolation, the s, values obtained were the same in presence and absence of TritonX-100 and corresponded to the published values. These data suggested that the presence of Triton X-100 did not change the molecular properties of eel acetylcholinesterase. Attempts to filter the detergent-depleted human red cell enzyme through Sepharose 4 B failed when the column was not equilibrated with the detergent. The majority of the enzyme (more than 80% of the activity applied) was retained on the column. The remaining activity eluted as a broad band. ending after the inner volume of the gel. This behavior indicated a strong retention of acetylcholinesterase by Sepharose 4B. When gel filtration was performed in presence of 1 ”/, a-methyl D-mannoside, the yield in enzyme activity in the eluate fractions exceeded 85%. Thus the presence of this sugar prevented the enzyme from interacting with the carbohydrate moiety of the gel matrix. When the sucrose density gradient centrifugation experiment were carried out in presence of 1 a-methyl D-mannoside, the distribution pattern of the multiple forms of the detergent-depleted enzyme differed

416

Molecular Forms of Human Erythrocyte Acetylcholinesterdse

Table 3. Comparison of molecular parameters of acetylcholine.sterase ufier separation by .sucro.si’ density-gradient centrifugation Values of s, were obtained after recentrifugation of pools 1 - 6. Pool 1-6 refer to pools shown in Fig. 5 A Enzyme

Pool 3 2 3 4 5 6

In 0.1 7; Triton X-100

Sedimentation coefficient, .s4

Stokes’ radius

S

nm

7.3 10.9 12.9 14.8 16.8 18.1

6.8 10.1 10.4 10.7 11.1 12.9

6.3

8.7

8

Start

-

A

6

C

0 Fig. 7. Dodi~c~lsulfh.te-polyacryla~~ide gel electrophoresis of reduced human erythrocyte ucetylcholinesterasr. Preparation o f protein samples and electrophoresis were carried out as dcscribed in Materials and Methods. The samples applied contained 50 pg of protein. After the run, the gel was cut and stained for protein (A) and carbohydrate residues (B)

I 50

0 Fig. 6. Polyucrjlaniidc grudiivii gcl electropliorr.si.s o/ deiergentElectrophoresis was carried out as depleted acet~~1~holine.sterasc. described in Materials and Methods. Each enzyme sample applied (20 pl) contained approximately I0 pg of protein. Three samples were applied onto one gel slab. After the run, the slab was cut into three slices and stained for enzyme activity (A), for protein (B) and for carbohydrate residues (C)

markedly from the one obtained in the absence of the mannoside (Fig. 5A). Pools 6 to 1 were recentrifuged under identical conditions and gave the distribution patterns shown in Fig.5B to G. The Stokes radii of each multiple molecular form were estimated by gel filtration on Sepharose 4 B and the results are summarized in Table 3. Polyacrylamide gradient gel electrophoresis gave seven enzymically active bands corresponding to the protein bands. Six of these bands could also be detected after staining for carbohydrate residues (Fig. 6). On the other hand, dodecylsulfate-polyacrylamide gel electrophoresis of reduced acetylcholinesterase revealed the presence of one heavily stained band (Fig. 7) with an apparent molecular weight of 80000 k 1000

1

.-c

-

20 .

s ‘f‘

0

10

.

5 .

I

0.1

0.2

0.3

0.4 0.5 0.6 Relative migration

0.7

0.8

Fig, 8. Determinu t ion of the subunit ni olecular nviglif o/ fk.1e,gen tdepleted Iiumun erythrocyte acetylcholinrsterasi~.The subunit molecular weight was determined by dodecylsulfate-polyacrylamide gel electrophoresis in the presence and absence of mercaptoethanol as described in Materials and Methods. Ovalbumin and bovine serum albumin served as marker proteins

(Fig. 8). Two lightly stained faster-moving bands with apparent molecular weights of 65 000 and 41 000 could also be detected. It was estimated, that the heavily stained band accounted for at least 90% of the total protein applied to the gel. When stained for carbohyEur. J . Biochem. 57 (1975)

411

P. Ott, B. Jenny, and U. Brodbeck

drate residues the same patterns were observed as with the protein stain. Incorporation of [3H]Dip-F into acetylcholinesterase showed after dodecylsulfatepolyacrylamide gel electrophoresis, that the radioactive label was associated with the 80 000 molecular weight band. Small amounts of radioactivity could also be found in the 65000 and 41 000 molecular weight bands. Dodecylsulfate-polyacrylamide gel electrophoresis of nonreduced acetylcholinesterase yielded one heavily stained band at 159000 molecular weight.

DISCUSSION

Enzyme Purification Several investigators have described the partial purification of erythrocyte acetylcholinesterase. Using affinity chromatography Berman [13,14] obtained a pure bovine red cell acetylcholinesterase. Ciliv and Ozand [9] presented evidence by disc gel electrophoresis and electron microscopy for a relatively homogenous human erythrocyte enzyme with a specific activity of 30 IU per mg of protein. More recently Shiotang [12] obtained an apparently pure human enzyme with a specific activity of 120 IU per mg of protein and an average yield of 13 %. When the specific activity described in this paper (3800 IU per mg of protein) is compared to the values listed above it is obvious that isolating the enzyme by affinity chromatography gives a preparation with at least a 30-fold increase in specific activity as compared to the former values. Moreover affinity chromatography produced purified acetylcholinesterase with recoveries of 30-40%. It should be noted that the length of the spacer attached to the Sepharose matrix is essential for this procedure. Extending the spacer only for two instead of three cycles of coupling (hexamethylene diamine followed by succinylation) yielded, after attaching the high-affinity ligand, a resin that bound human erythrocyte acetylcholinesterase only weakly. Such a resin was not suitable for the purification of this enzyme. Contrary to the bovine red cell enzyme that can be solubilized from ghosts by hypertonic saline treatment [36], the one from human erythrocytes required the use of a detergent, confirming the observation made by Miller [37]. Of various agents tested (Tween 80, Lubrol WX, sodium deoxycholate) only Triton X-100 gave satisfactory results throughout the purification procedure. Besides the high absorption coefficient (A:& n m = 21), this detergent strongly interferes with the Lowry protein determination as well as with polyacrylamide gel electrophoresis. Although the standard curve of the Lowry assay was established in the presence of Triton X-100 the values obtained Eur. J. Biochem. 57 (1975)

for the purified enzyme possibly are subject to error and therefore are listed in parenthesis (Table 1). After removal of the excess of Triton X-100 no such interference was observed.

Properties of the Enzyme in the Presence of Triton X-100 With respect to molecular dimensions, human erythrocyte acetylcholinesterase in the presence of Triton X-100 appears as a seemingly homogenous protein preparation. As detailed by Tanford and coworkers [38] membrane proteins solubilized by the use of mild detergents most probably retain their native configuration. Helenius and Simons [39]showed that the lipophilic proteins of human erythrocyte apostroma bind large amounts of detergents in a micellar solution of Triton X-100 whereas hydrophilic proteins bind little if any of this detergent. The studies of Umbreit and Strominger [40] on a membrane-bound carboxypeptidase revealed that this Triton X-100 solubilized enzyme appeared as an artefactually large protein-detergent complex with a very low sedimentation coefficient. From the Stokes’ radius (8.7 nm) and the low s, value (6.3 S) it is concluded that the human erythrocyte acetylcholinesterase solubilized by Triton X-100 is not a spherical protein but assumes an elongated shape. Consequently estimates of molecular weight by gel filtration alone are subject to error and previously published values should be viewed with caution [I, 4,5,91. These results also suggest that the protein was combined with detergent molecules which “floated” the enzyme in the sucrose density gradient. Thus neither the estimates of molecular weights from s4 values alone according to the method of Martin and Ames [23] nor the calculation from s4 values and Stokes’ radii as described by Siege1 and Monty [29] can give reliable results for the molecular weight of human erythrocyte acetylcholinesterase. As suggested by Tanford [38], Triton X-100 most probably binds to lipophilic regions of the enzyme thus increasing the partial specific volume of the protein. The trailing of enzyme activity encountered upon gel filtration in the presence of TritonX-100 led several authors to conclude that enzymic species with apparent molecular weights as low as 93000 existed [4,5]. When in the present study samples from various positions on the trailing end of the elution profile were rechromatographed without prior concentration of the enzyme solution, the elution volume always decreased to the one shown in Fig. 1. This indicated that the presence of Triton X-100 induced an artefactual behaviour. As noted by Cann [41],

478

the trailing peaks observed in gel filtration experiments in the presence of detergents are due to complex equilibria existing between free enzyme, the detergent and aggregates thereof. Such trailing peaks should not be taken as evidence for the existence of enzymic& ly active forms of lower molecular weights. On the other hand heterogeneity with respect to overall charge could be demonstrated by isoelectric focusing experiments (Fig. 3). The resolution of human erythrocyte acetylcholinesterase into several enzymic species differing in their isoelectric points parallels earlier observations made for a number of acetylcholinesterases of different origins and of butyrylcholinesterases. Similar to the plaice enzyme and i n contrast to the plasma enzymes [42-441, no differences in catalytic properties have been found so far for the five forms of human red cell acetylcholinesterase. Incubation with neuraminidase changed the distribution of thesc multiple molecular forms in favor of the one with the highest isoelectric point. This observation contrasts the one of Wright and Plummcr, who found no change in electrophoretic mobility of acetylcholinesterase upon incubation with neuraminidase although this treatment yielded an increase in free sialic acid residues [4]. The change in isoelectric points induced by neuraminidase is similar to that observed by Svenmark and coworkers for human serum [45], horse serum [46] and human brain acetylcholinesterase [22].

Properties of the D i z ~ ~ n Depleted zc qf’ Tritm X-100 After removal of excess Triton X-100 a more complex sedimentation pattern is obtained. Besides the “floating“ 6.3-S form, a number of clearly distinguishable heavier aggregates are observed. The occurrence of multiple molecular forms differing in their sedimentation coefficients parallels earlier observations made on acetylcholinesterases from the electric eel [47-501. Our early attempts to estimate the Stokes’ radii of the multiple molecular forms of the human red cell enzyme failed, as the enzyme could not be properly filtered through a column of Sepharose 4 B. The strong retention of the enzyme was attributed to interactions between sugar residues in the enzyme and the carbohydrates in the gel matrix. As noted earlier the cnzyme depleted of Triton X-100 was quantitatively adsorbed on a column of concanavalinA-Sepharose from which it could be eluted by a solution of %-methyl D-mannoside [52]. Interestingly the presence of this sugar during gel filtration prevented the enzyme from being retained on Sepharose 4B. This in turn allowed the Stokes’ radii to be estimated. As listed in Table 3 the increase i n sedimentation coefficient is accompanied by an increase in the

Molecular Forms of Huinan Frythrocyte Acctylcholinestcrasc

molecular diameter. Similar properties were found by Massoulie and coworkers for the eel enzyme [50]. Both eel and human erythrocyte acetylcholinesterase thus assume an elongated structure. In general, human erythrocyte acetylcholinesterase can be regarded as an integral membrane protein [52]. It is of interest to note that the elongated structures of this enzyme seem to represent an exception to the general rule that molecules of integral membrane proteins are more or less globular [53,54]. The presence ofcc-methyl D-mannoside during sucrose density gradient centrifugation almost completely abolished the presence of the “floating” 6.3-S form suggesting that this sugar might be able to displace excess Triton X-100 bound to acetylcholinesterase. Although some molecular parameters were established during the course of this investigation, reliable estimates on the molecular weight of each oligomer can not be given at present. On the other hand the subunit molecular weight of the reduced enzyme was determined as 80000 & 1 000. This value agrees closely with that obtained from dodecylsulfate-polyacrylamide gel electrophoresis of [3H]Dip-F-labelled mercaptoethanol-treated human erythrocyte membranes [5,6,55] but contrasts the results of Berman [14] who found in the bovine enzyme two subunits of 126000 and 75000 molecular weight. In the human enzyme two faster-moving bands of molecular weight 65 000 and 41000 could also be detected. The intensity increased primarily with increasing time of autolysis or with increasing age of the preparation. This situation appears to be similar to the one described for 1I-S eel acetylcholinesterase [56]. The molecular weight of non-reduced human erythrocyte acetylcholinesterase was 159 000 again in good agreement with the number given by Bellhorn et a/. [55]. The positive periodate- Schiff stain together with the observation that the human red cell enzyme binds to concanavalin A [51] verifies an earlier observation on an impure preparation [9] that this enzyme together with acetylcholinesterases from eel [57,58] and torpedo [59] is a glycoprotein. From the low sedimentation coefficient of 6.3 S obtained in the presence of Triton X-100 and from a subunit molecular weight of 80000, it might be concluded that in presence of this detergent the enzyme exists in a catalytically active monomeric state. This hypothesis however can be refuted on the basis that single subunits are obtained only after dodecylsulfate denaturation followed by disulfide reduction. It is likely however that the 7.343 species obtained after removal of Triton X-100 represents the protomer of erythrocyte acetylcholinesterase. Its sedimentation coefficient and its Stokes’ radius of 6.8 nm compare favorably to the ones of form Gi of the eel enzyme Eur. J. Biochem. 57 (1975)

P. Ott. I3 Jenny. and U . Brodbcck

(7.7 S and 6.4 nm respectively). This forin consists of two subunits [49] interlinked by disulfide bonds [56,61]. As reasoned elsewhere [62] the oligomeric forms of the huinan enzyme might be built up in multiples of the proposed dimeric protomer ( 7 . 3 3 species). If this hypothesis is correct, the human enzyme would be dif-ferent with respect to its subunit composition from the eel enzyme of which forms A, C and D are composed of4.8 and 12 subunits attached to a tail-like structure [58]. C~I~~~~ZIS~OMS

Human erythrocyte acetylcholinesterasc is more firmly bound to the red cell membrane than the bovine enzyme or the one located in the electric organs of the electric fishes. The human enzyme solubilized by Triton X-100, though apparcntly homogenous with respect to molecular dimensions most probably occurs as a detergent-induced artefact. After removal of the detergent the enzyme assumes different states of aggregation. The occurrence of multiple molecular forins closely parallels similar observations made on the membrane-bound enzyme from the electric organs of eel and torpedo [60]. 'Tlic authors iirc indebted to Drs M .Rottenberg and W. Hopff For helpful discussions and valuable suggestions regarding this \vork. to Prof. H. Achi for providing financial assistance and for his continuous interest in thcse studies and to M r M. Schlechten for skilful technical assistance. The authors thank the central blood hank of the Swiss Red Cross for its gencrous supply ol' Ii~imanred cell sediments. Dr H. Kohler for gifts of' '251-labelled thyroglobulin and Prol'. B. Fulpius for making available to ub the Pharmacia electrophoresis equipment.

REFERENCElS I . ShaPai. T. & Cortner. J. A. (1971) Bioc./iin?. Biophj.s. Ac/a. 236. 6 1 2 - 6 18. 2. Shafai. T. & Cortncr, .I.A . (1971) Biochi/ri. Bio1dij.s. Acru, 2.50. 1 17 120. 3. Wright. D. L. & Plummer, D. T. (1972) Bioc,/iini. B i o p h j ~ . Ac~/a,261. 398--401. 4. Wright. I>. L. & Plummer. D. T. (1973) Bioclwni. J . 133. 521 ~-577. 5. Whittaker, M . & Charlier. R. A. (1973) Ah.rrr. Comnzun. Ytli 1/11, C'ongr. Bioclwni. Abs. 2, o, 34. 6. Gentinetta, R. & Brodbeck, LJ. (1972) E.\-i'c~rioifin/Base/) 28. 735. 7. Mentha. J.. Sprinj.. H. & Barnard. K.(1947) J . Riol. C'hrnz. 167, 623. 8 Zittle, C. A,, Della Monica, E. S. & Custer, J . H. (1954) A r d i . Rioclierl?. Bi0pllJ.S. 48, 43 ~49. 9. Ciliv, G . & &and, P. T . (1972) Biodiini. Bin1ih.v.s. Acrrr, 2x4, 136-- 156. 10. Paniker, N . l'., Arnold, A. B. & Hartmann, R. C. (1973) Proc. Soc. E.vj1. Sin/. Mid. 144. 492-497. 11. Wright. 11. L. & Plummer, D. 1.. (1973) Bioc~hcm.Soc. Trcnw. 1. 1123.- 1125. 12. Sihotang, K . (1974) Bioc~hin?.Biophvs. Actcr. 370, 468 -476. -

t u r . J. Biochem. 57 (1975)

479 13. Bcrman, J . 11. & Young. M. (1971) Proc. N u t / Actrtl. hX. 3 9 s 398. 14. Bcrman. J . 0. (1973) ~ i 0 C h ~ w i W ' l ;/ 2 , 1710- 1715. 15. Jciinq. B.. Oit. P. & Brodbeck. U . (1974) ;I.K l n ~ .C'hcwr. K/in, Broc~/wii.12. 242. 16. Ott, P. & Brodbeck, U . (1974) A h s / r . Coninnrn. Y t / i :Wee'/. /.>el. h r . Bioc,hcrn.Soc. Abs. f I.g 4. 17. Dodge. J . T., Mitchell. C. & Hanahan. D. J . (1963) A r c h . Bioc./wrv. Biophj.s. 100. 119 - 130. 18. Lllnian, Cr. L.. Courtney. D . K., Andres. V. & Feailierstonc. K.M. (1961) Biocl7em. Pliarniucol. 7, 88-95. 19. Acbi. H. ( 3 974) 12ildiotlen t k r ~ ' / / ~ ~ / l ~ a r / , \Anu/j..sc c l / [ , / ? (Bergmeyer. 1 1 . U., ed.) 3rd edn. vol. 1. pp. 713-724, Vcrlag Chemie. Wcinheiin, Bcrgstr. 20. Loury, H. O., Kosehrough, N. J.. Farr, A. & Randall, R . .I. (1951) .I. Biol. Clrcw. 193, 265-275. 21. Vcstcrbcrg, 0 . & Svensson, 13. (1966) Ac/ti C'hc~ni. .Sctrnc/. 20. 820~k834. 22. Carlsen, J. B. & Svensmark, 0. (1970) Riochini. Bioplij,~.,4c,/tr, 207. 477- 484. 23. Martin. K . G. & Amcs, B. N . (1961)J. Biol. C'/icw?. 236, 1372-1379. 24. Webcr. K . & Osborn, M. (1969)J. Bid. (%eni. 244.4406--4412. 25. Dunker, A. K . & Rueckcrt, R. R. (1969) J . Bioi. Clicwi. 244. 5074- 5080. 26. Zacliarius. R. M., Zell. T. E.. Morrison, J . H. & Woodlock, J. J. (1969) ilnrd. Biochem. 30, 148~-152. 27. Grid, J. (1961) Ann. I m t . Pcrsrewr., 101, 104-1 19. 28. Karnovsky, M . J. & Roots. L. (1964) J . Hi.sro~hcn?.Cjrochcni. 12,219-221, 29. Siegel, L. M. & Monty, K . J. (1966) Biodiirn. Biop/ij..c. Acta. I12. 346--362. 30. Cuatrecasas, P. & Anfinsen. C. B. (1971) M d x ~ u ' :G, i ~ ~ ~ n 22. iul. 345 - 378. 31. Hopl'f. W. H., Riggio, G. & Waser, P. G. (1973) E%BS Lttr. 35,220 - 222. 32. Holloway, P. W. (1973) Anal. Biuc,/ieni.53. 304G 308. 33. klerz, F. & Kaplan, E. (1973) Pediar. Res. 7. 204-214. 34. Jackson, P. & Whittaker. M. (1972)