Thermal inactivation of the molecular forms of acetylcholinesterase and butyrylcholinesterase

Btochtmwa et Bwphyswa Acta, 742 (1983) 509-516

509

Elsewer Bmme&cal Press BBA31489

T H E R M A L INACTIVATION OF T H E MOLECULAR F O R M S O F A C E T Y L C H O L I N E S T E R A S E AND BUTYRYLCHOLINESTERASE JA

EDWARDS and S. BRIMIJOIN *

Department of Pharmacology, Mayo Graduate School of Medw:ne, Rochester, MN 55905 (U S A )

(Received October 25th, 1982)

Key words Acetylchohnesterase, Butyrylchohnesterase, Heat stablhty, Monomerw form, (RodenO

To compare acetylcholinesterase (acetylcholine acetylhydrolase, EC 3.1.1.7) and butyrylcholinesterase (acylcholine acylhydrolase, EC 3.1.1.8), we utilized the physical parameter of thermolability. In serum or muscle extracts from mouse and rat, butyryicholinesterase was inactivated as a unimodal function of temperature. Inactivation began at 51°C and was complete at 54-570C (depending upon time of incubation). Acetylcholinesterase was inactivated in two stages. A 6(Wo decrease in activity from 42 to 48°C was followed by a plateau. The second stage of inactivation began at 51°C and was complete at 57-60°C (depending upon time of incubation). Sucrose density gradients revealed that the partial loss of acetylcholinesterase activity at 48°C was due to inactivation of the monomeric 4 S enzyme, which was the most thermolabile molecular form in each tissue examined. When heated after isolation on density gradients, most of the forms of acetylcholinesterase and butyrylcholinesterase lost activity as a single exponential function of time. The monomers of both enzymes were inactivated fastest. Inactivation of the larger forms was slower and required higher temperatures. Tetrameric 10 S acetylcholinesterase was unique in following a time course that could only be fitted by a double exponential equation (i.e., when this form was heated to 55°C, almost 60% of the activity showed a short half-life while the remainder showed a long half-life). This behavior did not reflect differences in the thermolability of soluble and membrane-derived tetramers.

Introduction

Acetylchohnesterase (acetylchohne acetylhydrolase, EC 3.1.1.7) and butyrylchohnesterase (acylchohne acylhydrolase, EC 3.1.1.8) are closely related enzymes found throughout the body. These enzymes have s~mdar actwe s~tes and mechamsms of action [1]. Both are polymorptuc, as shown by electrophoresls [2], gel chromatography [3,4] and density gradient ultracentnfugaUon [5,6]. The polymorpbasm of acetylchohnesterase is explained by the occurrence of globular monomers, &mers and tetramers, and by the assocmt~on of the latter with collagen-hke tails m larger arrays [7] A ho* To whom correspondenceshould be addressed 0167-4838/83/0000-0000/$03 00 © 1983 ElsevierBlome&calPress

mologous senes of naturally occurring forms of butyrylchohnesterase has also been identified [8] The many Slmllarlhes between the chohnesterases have led to suggestions that b u t y r y l c h o h n e s t e r a s e is a p r e c u r s o r of acetylchohnesterase [9-11]. However, the two enzymes are lmmunolog~cally non-cross-reacting [6] and are regulated by mechamsms that respond &fferently to hereditary [12] and hormonal factors [13] It t h e r e f o r e seems hkely that acetylcholinesterase and butyrylchohnesterase are products of &fferent genes. Even so, it is appropriate to ask what the genetic relatlonsh~ps are among the molecular forms of acetylcholinesterase or butyrylchohnesterase, and between the corresponding forms of these enzymes. One approach

510

to this question would be to identify genetic variants of one or more of the molecular forms of either enzyme. Such variants could then be used for a study of gene-linkages. Perhaps the most sensitive test for genetic variants that differ m primary protein structure is thermal inactivation [14]. In order to lay the groundwork for future study of enzyme variants, and for its inherent interest, it would be desirable to estabhsh the thermolablhty of the molecular forms of both chohnesterases. Some information is already available on the thermolabdlty of total acetylchohnesterase activity [15,16], of electrophoretlc isozymes of acetylchohnesterase [17], and of two molecular forms of butyrylchohnesterase [18]. However, no one has systematically compared the thermal mactlvauon of the molecular form s of acetylchohnesterase and butyrylchohnesterase This was the purpose of the present study Materials and Methods

Preparat:on of ttssues Male Sprague-Dawleyrats (200-250 g) and male I C R mice (25-35 g) were used. The methods for preparation and hendhng of tissues have been described in detail previously [12]. Animals were anesthetized with ðyl ether, and the thoracic cavity was opened. Blood samples were drawn by cardiac puncture and allowed to clot for 1 h at 4°C for later isolation of serum. A catheter was introduced into the aorta while the heart was still beating, and the animal was exsangulnated by perfuslon with 0 9% NaC1 in order to remove most of the blood-borne chohnesterase. The atria or the left henudmphragms were then removed, cleaned of connective tissue, and weighed Extraction of enzyme Tissues were homogenized in cold (4°C) buffer at a ddutlon of 10 m l / g wet weight, with a Polytron homogenizer (Brinkm a n n Instruments) at setting 5 for 30 s. For extraction of acetylchohnesterase, the buffer was 0 05 M Tris-HC1, p H 7 4/1% Triton X-100/1 M NaCI/0 2 mM EDTA F o r extraction of butyrylchohnesterase, NaC1 was omitted and Tris was replaced by sodium phosphate (0 05 M, p H 7.4) in order to optimize recovery of activity. Homogenates were centrifuged at 16000 × g for 10

mln at 4°C In a Sorvall RC-5B centrifuge, and the supernatant fractions were immediately used for thermal inactivation or density gra&ent ultracentrlfugatlon (see below) In one experiment, membrane-bound acetylchohnesterase was separated from the readily sohibihzed forms of this enzyme by prehmlnary ultracentrlfugatlon as follows. Rat heml&aphragm was homogenized in 10 vol. of low-salt buffer lacking detergent (1 e , the normal homogenization buffer rmnus Triton and NaC1). The homogenate was centrifuged for 10 mln at 500 × g. The supernatant was recovered and spun for 1 h at 100000 × gay with a Type 35 rotor in a Beckman L2-65B ultracentrifuge. The high-speed supernatant (S1), contaming soluble enzyme, was retained. The highspeed pellet was rinsed by rehomogenlzation in the original volume of low-salt buffer and again centrifuged at 1 0 0 0 0 0 × g for 1 h. The supernatant was discarded. The final pellet was rehomogenlzed in an equivalent volume of the low-salt buffer to which Triton X-100 had been added in a final concentration of 1% This resuspended pellet was centrifuged at 10000 x g for 10 min in a Sorvall RC-5B. The supernatant ($2), containing 'membrane-derived' enzyme, was retained Fractions S 1 and S2 were applied to sucrose density gradients Separation of molecular forms The molecular forms of acetylchohnesterase and butyrylchollnesterase were separated according to the method of Hall [5] as previously described in detail [19] Samples (100/tl) of diaphragm or atria extract, or of serum, were layered onto 5-20% linear density gradients of sucrose made up in the homogenization buffer. Catalase (l 1.2 S) was also applied as a callbraUon standard. The gra&ents (5 ml each) were centrifuged in the L2-65B ultracentrifuge (SW 50.1 swinging bucket rotor) at 121000 ×gav for 16 h at 4°C Fractions (0.2 ml) were collected from the bottom of the tubes For study of particular enzyme forms in isolation, only the appropriate peak fraction and the two immediately adjacent fractions were pooled This procedure was intended to minimize cross-contamination. Inhtbmon of endogenous proteases In some experiments it was desired to inactivate endogenous proteases. For this purpose, tissue extracts or pooled samples of isolated molecular forms were

511

&luted w~th an 'ant~protease soluUon' to obtain the following final concentraUons: 50 mM sodium phosphate, pH 7.4, 2 mM benzanudine hydrochloride, 5 mM N-ethylmaleinude, 1 m g / m l bacitracm, 20 # g / m l pepstatm Thermal mactwatton Samples (150/~1 of serum, muscle extract, or molecular forms from sucrose density gra&ents) were heated for varying periods of time (1-45 mln) at one of several temperatures (42-60°C) m a thermostatically regulated bath, w~th constant shaking. Control samples were kept at 4°C. All samples were assayed for enzyme act~wty at the same time, usually wltlun 1 h. Enzyme assay. Acetylchohnesterase was assayed m duphcate samples at 37°C by a modification of the rad~ometnc assay devised by Potter [20], as described previously [19]. [14C]Acetate-labeled acetylcholine was the substrate (1 mM), and ethopropazane hydrochlonde (10 -4 M, Stgma Chemical Co) was a d d e d to m h l b l t butyrylchohnesterase by over 99%. Blank samples were generated by adding BW284C51 (15-bls(4-allyldlmethylammomumphenyl)pentan-3-one dtbrormde; Burroughs Wellcome Co., Research Triangle Park, NC) at a concentration of 10 -5 M. Activity is expressed as nmol of substrate hydrolyzed per h. Butyrylchohnesterase acuwty was measured m duphcate samples at 37°C by an essentmlly smular ra&ometnc method using [14C]butyrate-labeled butyrylchohne as a substrate (0.5 mM) as prevaously described [12]. The reaction nuxtures all contained BW284C51 (10 -5 M) to lnactwate acetylchohnesterase. Blank samples contained, m ad&tlon, ethopropazlne (10 -4 M). Actwlty is expressed as nmol of substrate hydrolyzed per h. Stat,stws Stat~sucal analys~s was performed on a Hewlett-Packard Model 85 computer. Curve-fitting was carried out with the a~d of a generahzed ~teratwe non-linear regression program adapted for this computer by Peck and Barrett [21]. The calculauons yielded best estimates of the desired parameters together w~th their standard errors.

I00

10050

~,~~~

"~ ~o ==

30

60

90

N

120

~ 5e

•

~ ~o co

m

o

, 180 , 2~0 300 , 120

Q w

I

~

60,

=o ,oo

~ ~eo

P,

o

i 30

I 60

I 90

I 120

~

0

6'o 120 . . 180. 240 . 300

ASSAY DURATION, MINUTES

Fig 1 Stablhty of enzyme actavtty during assay Rat &aphragm extracts and samples of mouse serum were prepared with or without protease lrdubltors (see Materials and Methods) Control ( © ) samples were kept on ice Experimental samples were heated to 50 (A) or 55°C (W) for 10 n u n and then returned to lee untd assay Acetylchollnesterase (ACHE) actwtty (A and B) was determined only on the muscle extracts Butyrylchohnesterase (BuChE) activity (C and D) was determined only on the serum samples Rephcate samples were mvesUgated with protease mhtbltors (B and D) and without (A and C) All assays were earned out at 37°C Regression hnes were fitted by the method of least-squares

(A)

(B)

1(30

80

8O

>-

,.=,80 0 ¢,,.) tIJ

nr 40

,° 1

20

45 50 5 5 - 6 0 45 50 55 60 TEMPERATURE (°C)

Results

Enzyme stabtltty durmg assay. Since our chohnesterase assays typically involved a single

Fig 2 Acetylchohnesterase (0) or butyrylchohnesterase ( O ) actlwty rema=mng m mouse serum after incubation at varying temperatures for (A) 10 man or (B) 45 man Data are from a representatwe experiment S, rmlar patterns were obtained vath heat, ng for 5, 15, 20 and 30 man

512

duration of reaction, ~t was ~mportant to determine whether enzyme that had been partmlly mactwated by prior heating would retain its residual activity throughout the assay period. For tlus reason, extracts of rat dmphragm and mouse serum were assayed for varying intervals at 37°C after an m m a l 10 nun treatment at 50 or 55°C. The hydrolysis of substrate m these samples was found to be an absolutely hnear function of assay duration, regardless of whether protease mhlbitors were added to the mcubaUon media (Fig. 1) Ttus result l m p h e s t h a t the r e a c t i o n s c a t a l y z e d b y acetylchohnesterase and butyrylchohnesterase proceeded at constant velooty for an extended period (at least 2 and 5 h, respectwely) Inasmuch as the tested samples contained representatwe amounts of the m a j o r m o l e c u l a r f o r m s of b o t h chohnesterases, we concluded that single-point assays provided an accurate index of residual enzyme act~vlty.

Thermolabthty of the chohnesterases m serum and muscle. Mouse serum was heated for up to 45 m m at vanous temperatures and was subsequently assayed for acetylchohnesterase and butyrylcholinesterase. In the heated sera, butyrylchohnesterase lost actw~ty as a ummodal function of temperature. A shght actlvat~on somet~mes occurred after heating to 42°C but was not consistently observed. Loss of act~wty was first noticeable at a temperature of 51°C. All butyrylchohnesterase activity was lost after heatmg to a temperature between 54 and 57°C, depending on the durauon (Fig. 2). The acetylchohnesterase activity of mouse serum followed a more complex curve of reactivation, decreasing as a b~modal function of temperature The first stage of mactivahon began at 42°C and accounted for approxamately 60% of the actwlty The Ts0 of this stage (Le, the temperature at wluch half of the more thermolabile component was lnactwated) was 45°C. A plateau m the curve for mactwatlon of acetylchohnesterase followed from 48 to 51 °C. Higher temperatures induced a second stage of mactwat~on, w~th a Ts0 near 54°C. All acetylchohnesterase actwlty was lost at 57-60°C (Fig. 2) Slrmlar experiments were performed on the thermal mactwat~on of acetylchohnesterase and butyrylchohnesterase in extracts of muscle from

6o

.

(A)

"C J~

s

E c

40

>,p> Io <( m,m

/P',

20

J~

J

J~

i

I0

I~

I 15

i 20

15

20

',,,o.\ ' ",~ " 25

CAT

(B)

I.

°/

80 0

~

60

>

hi ~

2O

I11 5

iO

25

FRACTION

Fig 3 (A) Distribution of acetylchollnesterase (ACHE) molecular forms m mouse serum centnfuged on 5-20% sucrose denslty gra&ents O, Unheated control fractions, O, fractions heated 10 nun at 48°C before assay (B) Distribution of butyrylchohnesterase (BuChE) molecular forms m mouse serum treated as in A 0, Control, O, heated Catalase (CAT) was used for cahbratlon Data are from a representatwe expenment

mouse and rat. The results (not shown) closely resembled those obtained with sera.

Thermolabthty of the molecular forms of the chohnesterases We noticed that the proportion of thermolabde acetylchohnesterase in mouse serum (Le., the component mactwated at 48°C) was about equal to the proporUon of the 4 S or m o n o m e n c form. Hypothesizing that m o n o m e n c acetylchohnesterase was umquely sensmve to heating, we centrifuged mouse serum on sucrose density gradients and heated the separated fractions to 48°C

513

for 10 min. The profile of activity m the heated gradients showed that 4 S acetylchohnesterase was differentially inactivated (Fig. 3). An identical result was obtained when sera were heated before gradient-centrifugation. The loss of 4 S

30

(A)

CAT

io I

t~

5

I0

15

20

25

acetylchollnesterase activity in both types of experiment was at least 67%. The true inactlvatmn was probably even greater smce the shift in peakposition after heating (Fig. 3A) indicated that much of the remaining actiwty was contributed by dlmerlc 6.5 S acetylchohnesterase. Because mouse serum contmns only two of the major molecular forms of acetylchohnesterase m easily measurable amounts we examined the thermolability of other forms m another tissue. Sucrose denstty gradients of rat diaphragm were fractlonated and heated to 48°C for 10 rain, as above. In the gradient-profiles, no effect on the tetramenc (10 S) or asymmetric (16 S) forms was noticed. However, the 4 S acetylchohnesterase was again differentially inactivated, and 6.5 S acetylcholinesterase was unmasked (Fig. 4A). The observations on rat diaphragm suggested that dimeric acetylchohnesterase was much less thermolabile than the monomenc enzyme, but a richer source of dimer was needed for a careful test of this idea. Such a source ts the mouse

,B,

(B)

(A)

CAT ;~

IOO

IOO

80

80

>Iz LIJ 60

60

o o .i re

40

40

20

20

i ii 6

i 1 I0

, ,

~

,~" 'k

I

,5

//

,,,', A

~

;,

.~

.

,'

I 45 I10

15

20i

I 55

I 60

TEMPERATURE

25

FRACTION

I 50

Fig 5

45 50 55 60 (°C)

(A) Thermal mactlvatmn of molecular forms of

Fig 4 (A) Dlsmbuuon of acetylcho]mestcrase (ACHE) molecular forms m extracts of rat diaphragm centrifuged on 5-20~ sucrose density gradients, I , Control, O, heated l0 nun at

acetylchohnesterase from mouse diaphragm (B) Thermal macuvauon of molecular forms of butyrylchohnesterase from

48°C (B) DlstnbuUon of acetylcholmesteras¢ molecular forms m extracts of mouse atria centrffugexi on 5-20~ sucrose density gradients To facihtate separation of the lower molecular weight forms, the grachent was spun at 40000 rpm instead of the usual 35500 rpm i , Control, ©, heated as above Catalase (CAT) was used for callbratmn Data are from a representative experi-

sucrose density gradients using peak fractions plus the two adjacent fractions Samples were heated for 10 nun Data are from a representaUve expenment. A, Monomenc enzyme (4 S acetylchohnesterase or 5 S butyrylchohnesterase), O, tetram e n c e n z y m e (10 S a c e t y l c h o h n e s t e r a s e or 11 S butyrylchohnesterase), e , asymmetric enzyme (16 S acetylchohnesterase)

ment

mouse serum Molecular forms were collected from 5 - 2 0 ~

514

atrium, which has more dlmer than monomer [22]. We found that heating gradlent-fracuons of atria for l0 nun at 48°C caused no loss of activity in either the 10 S or the 6.5 S f o r m s of acetylchohnesterase (Fig. 4B). Evidently, &meric acetylchohnesterase is relatively thermostable To observe the thermal inactivation of the molecular forms of butyrylchohnesterase we studied mouse serum, winch contains appreciable quanuties of the m o n o m e n c (5 S) and tetrameric (11 S) forms of tins chohnesterase. The profiles of activity in gradient-fractions heated to 48°C for 10 mln (Fig. 3B) showed that tetramerlc serum butyrylchohnesterase was thermostable, like tetramertc acetylcholinesterase. Monomeric butyrylcholinesterase lost activity under the same conditions, like the acetylchohnesterase monomer. To exanune the thermal inactivation of the major molecular forms of acetylchohnesterase in more detail, 4, 10 and 16 S enzyme forms were isolated from mouse diaphragm on density gra&ents (see Materials and Methods) Samples of these isolated forms were then heated for up to 45 nun The behavior of the isolated acetylchohnesterase forms is shown in Fig. 5A. As expected, the monomer was most thermolablle (Ts0 = 47°C), whale the l0 and 16 S forms were stable at temperatures below 51°C (Ts0 = 56°C for both forms) Slnular results were obtained with acetylchohnesterase forms isolated from the rat dmphragm (data not shown). For the same type of analysis, the monomeric and tetrameric forms of butyrylcholinesterase were also isolated. Because the mouse diaphragm contamed too little of tins enzyme, mouse serum was used as the source. As w~th acetylchohnesterase, the m o n o m e n c butyrylcholinesterase was more thermolabde (Ts0 = 46°C) than the tetramenc form (Ts0 = 59°C). In fact, the forms of butyrylchohnesterase differed even more in their thermal reactivation curves (Fig. 5B) than did the forms of acetylchohnesterase (Fig. 5A).

Kmetws of macttvatton of the chohnesterase forms. The rates of inactivation of the molecular forms of acetylchohnesterase were determined on enzyme from the rat diaphragm. Samples of each enzymatic form were heated for 1-40 nun at temperatures capable of inducing near-total inactivation. It was found that the time-dependent loss of

(A)

(C)

(B)

ioo

~ oo,4 ~ooo,

go

uJ ,)o

g ~

(±ooor

20

' k ~ 2r2 t o ~ (

,o

,o

20

~o

,o TIME

~0 ~0

.

.

.

.

.

.

MINUTES

Fig 6 Thermal reactivation of molecular forms of rat diaphragm acetylchohnesterase collected from 5-20% sucrose gradients using peak fracuon plus the adjacent fractions A, Actwlty of 4 S acetylchohnesterase on a senuloganthnuc scale as a funcUon of time of incubation at 48°C Regression line fitted by computer (r E= 0 969) to a smgle exponentml equauon, y = A e - k, B, Activity of l0 S acetylchohnesterase heated at 55°C Regression lines fitted by computer (r 2 = 0 896) to a double exponentml equation, y = A e - k V + ( l - A ) e -k2t C, Heated at 55°C Regression line fined by computer (r 2 = 0 650) to a single exponentml equaaon, y = A e - kt The rate constants, ' k ' , have umts of n u n - 1

acetylchohnesterase activity in the 4 S form (at 48°C) and in the 16 S form (at 55°C) could each be described by a single exponential equation (Fig 6A and C). On the other hand, the inactivation of l0 S acetylchohnesterase (at 55°C) could only be described by a double exponential equaUon (Fig 6B) Tins curve had a rapid phase that accounted for 58 _ 7% of the total 10 S actlwty and a slow phase that accounted for the remainder. An apparently biphaslc inactivation of l0 S acetylchohnesterase nught have reflected effects of heat on endogenous proteases ( 1 . e , a c t i v a t i o n at moderate temperature and inactivation at Ingh temperature). The hkehhood of such an artifact was reduced by results of a second experiment in winch protease ininbitors were added to the pooled 10 S acetylchohnesterase before heating (see Materials and Methods) Subsequent assays showed that the curve of inactivation at 55°C again had a rapid phase (rate constant, 0.26 _ 0 08 rmn - I ) accounting for 5 6 + 13% of the 10 S acetylchohnesterase activity, and a slow phase (rate constant, 0 033 + 0.012 n u n - l ) accounting for the remainder. We also considered whether the biphasic time course of the inactivation of 10 S acetylchohnesterase could have reflected dif-

515 ferences in the thermolabihty of membrane-derived as compared w~th soluble or secreted enzyme. The readily solublhzed acetylchohnesterase and the membrane-associated forms of thts enzyme were prepared as described in Materials and Methods. Each sample was subjected to density gradient centnfugaUon, and the isolated 10 S acetylchohnesterase was heated for varying penods of time at 55°C. Both samples followed a blphaslc lnactxvatlon-curve The rapid phase of macuvatlon accounted for 37 + 8% of the soluble form, and 47 + 4% of the membrane-derived form, with rate constants of 0.521 + 0.288 mln - l and 1.110+ 0.401 mln -1, respectwely. The rate constants for reactivation of the remaining actlvlty were 0.021 +0.006 mln -~ for the soluble and 0.026 + 0.005 min -1 for the membrane-derived 10 S form. None of the differences between the properUes of soluble and membrane-derived enzyme were statistically significant We concluded that the btphasic heat-Inactivation kmetlcs of 10 S acetylchohnesterase did not reflect the presence of enzyme subclasses that can be separated according to their ease of solubfllZatlon. Finally, the kinetics of inactivation of the molecular forms of mouse serum butyrylchohnesterase were examined. The m o n o m e n c 5 S form of this enzyme exlublted a single exponential loss of activity (rate constant, 0 067 + 0.015 n u n - l at 48°C). Tlus rate of inactwatlon is Slnular to, although somewhat slower than, that of monomerlc acetylchohnesterase from the same tissue The tetramenc 11 S butyrylchohnesterase contrasted with the corresponding form of acetylchohnesterase m showing no lnactlvaUon at 55°C and a single, slow, exponential dechne in activity at 60°C (rate constant, 0.090 + 0 005 m l n - 1)

Discussion Our observations of differences between the thermolabdltles of acetylchohnesterase and butyrylcholinesterase are consistent with the results of an earher study on the chohnesterases of rat superior cerxacal ganglia [16]. Given the lugh probabdlty that the thermolabdlty of a protein will be altered by a single random amino acid substitution [14], such differences are hardly surprising.

It is more interesting to consider the differences in thermolabihty among the molecular forms of the two cholinesterases. The most marked differences were between the monomeric forms and the dlmerlc or mult~menc forms of each enzyme. In tlus respect, our results agree fully with those of Lockrldge et al. [18], who found that monomerlc butyrylchollnesterase from human serum was much more sensmve to heating than was dlmenc butyrylchohnesterase. To evaluate these findings one must consider the subunlt organization of the chohnesterases. As presently understood [7,23,24], the monomer is the basic budding block of the acetylchohnesterase forms An lntersubumt disulfide-bridge is thought to connect two monomers to form a dimer. Two dlmers then associate by means of non-covalent forces into a tetramer. The asymmetric forms are composed of three-standed collagen-hke hehcal 'tails' covalently hnked by disulfide bonds to a dlmer that is associated through quaternary interactions to another dlmer. A parallel scheme has been proposed to account for the molecular forms of butyrylchohnesterase [8,18]. Three features rmght cause the molecular forms of the chohnesterases to differ m their thermolablllty: (1) mtersubunlt disulfide bonds (in dlmers and larger forms), (2) hydrophoblc or lomc interactions between dlmers (m tetramers and larger forms), (3) covalent hnkages with collagen (m asymmetric forms) All of these features are lacking m monomers, but there is some evadence that the absence of mtersubunit &sulfide bonds is the mmn reason for the comparatwe thermolablhty of these forms. Thus, it is known that reduction of dimerlc butyrylchohnesterase by dlthloerythntol increases thermolabfllty without altenng the state of aggregation (as judged by sedimentation rate) and therefore presumably without ehminatlng non-covalent binding [18]. Unfortunately, Vlgny et al [6] found that effectwe reduction of tetramerlc acetylchohnesterase reqmres prehmlnary proteolysis by trypsin. So far, m experiments on rat serum acetylchohnesterase, we have been unable to conSlstently dissocmte tetramerlc acetylchohnesterase into monomer or to convert it to a thermolabde form by treatment with dlthloerythritol m the presence or absence of trypsin (unpublished data) Therefore, the precise role of mtersubumt disulfide

516 b o n d s in the t h e r m o s t a b l h t y of this enzyme rem a m s speculative. It m a y b e noticed that the t h e r m o l a b d i t y of the isolated forms of acetylchohnesterase and b u t y r y l c h o l i n e s t e r a s e was not entirely In a c c o r d wath the behavaor of the enzyme in crude extracts. F o r example, the t e m p e r a t u r e at wluch half of the m o u s e serum b u t y r y l c h o l i n e s t e r a s e was i n a c t i v a t e d was between 51 a n d 5 4 ° C ( F i g 2) O n the o t h e r h a n d , the Isolated t e t r a m e r i c e n z y m e was half ma c t w a t e d at 59°C. I n general, the Isolated forms of b o t h enzymes were slightly m o r e resistant to heatm g than were the crude extracts. W h e t h e r the differences in t h e r m o s t a b i l i t y reflect differences in p r o t e i n - p r o t e i n interactions, s t a b l h z l n g effects of the sucrose used in the d e n s i t y g r a d i e n t s or other factors c a n n o t be stated at present. It ~s interesting that the t h e r m a l i n a c t i v a t i o n of t e t r a m e r l c a c e t y l c h o h n e s t e r a s e followed a b l p h a s l c time course Willie the m a c t l v a U o n of every o t h e r f o r m of tlus e n z y m e a n d of b o t h forms of b u t y r y l c h o h n e s t e r a s e was m o n o p h a s i c . O n e possible e x p l a n a t i o n is that the different phases of i n a c t i v a t i o n represent different subclasses of 10 S a c e t y l c h o h n e s t e r a s e [25], even though subclasses with distinct thermolabflltles could n o t be &st m g u l s h e d b y their ease of solubfllzauon. A n o t h e r possibility IS that each phase o f lnacUvatlon represents a distinct c o n f o r m a t l o n a l change in the enzyme: (1) a r a p i d change that l m p m r s b u t does n o t e h n u n a t e enzyme activity a n d (2) a slow c h a n g e that causes c o m p l e t e reactivation. E l u c i d a t i o n of this m a t t e r wall require further study. O n e clear i m p h c a t l o n of o u r results Is that a n y a t t e m p t to identify t h e r m o l a b l h t y variants of a c e t y l c h o h n e s t e r a s e a n d b u t y r y l c h o h n e s t e r a s e for genetic studies will have to t a k e account of the r a d i c a l l y different b e h a v i o r of the separate molecular forms. A second, practical l m p h c a t l o n is that selective t h e r m a l l n a c t w a t l o n m a y be able to rep l a c e density gradients as a m e a n s to identify a n d quantltate the monomerlc forms of the chohnesterases. This simple m e t h o d nught facilitate study of the synthesis a n d regulation of these enzymes.

References 1 Main, A R (1976) in Biology of Chohnergm Function (Goldberg, A M and Hamn, I eds ), pp 269-353, Raven Press, New York 2 Davis, C A and Agranoff, B W (1968) Nature 220, 277-280 3 Hollunger, E G and NLklasson, B H (1973) J Neurochem 20, 821-836 4 Adamson, R D, Ayers, S E, Deussen, Z,A and Graham, C F (1975) Blochem J 147, 205-214 5 Hall, Z W (1973)J Neuroblol 4, 343-361 6 Vlgny, M, Bon, S, Massoull~, J and Gmlger, V (1979) J Neurochem 33, 559-565 7 Massouh6, J and Bon, S (1982) Annu Rev Neuroscl 5, 57-106 8 Lyles, J M , Stlman, I and Barnard, EA (1979) J Neurochem 33, 727-738 9 Koelle, G B, Koelle, W A and Smyrl, E G (1977) J Neurochem 28, 313-319 10 Koelle, W A , Koelle, GB and Smyrl, E G (1976) Proc Natl Acad Scl U S A 73, 2936-2938 11 Koelle, W A, Smyrl, E G, Ruch, G A, Siddons, V E and Koelle, G B (1977) J Neurochem 28, 307-311 12 Edwards, J A and Bnnujom, S (1982) J Neurochem 38, 1393-1403 13 Edwards, J A and Bnmtjom, S (1983) Blochem Pharmacol, in the press 14 Pmgen, K (1971) in Enzyme Synthesis and Degradation in Mammalian Systems (Rechclgl, M Jr, ed ), pp 1-46, Umverslty Park Press, Baltimore 15 Coleman, M H and Eley, D D (1963) Bloclum Blophys Acta 67, 646-657 16 Vlgny, M Glslger, V and Massouh~, J (1978) Proc Natl Acad Sci U.SA 75, 2588-2592 17 Bajgar, J (1979)J Neurochem 32, 1875-1879 18 Lockndge, O, Eckerson, H W and La Du, B N (1979) J Blol Chem 254, 8324-8330 19 Bnrmjom, S (1979) Mol Pharmacol 15, 641-648 20 Potter, LT (1967) J Pharmacol Exp Ther 156, 500-506 21 Peck, C C and Barrett, B B (1979) J Pharmacokm Blopharmacol 5, 537-541 22 Bnrmjom, S and Schrelber, PA (1982) Muscle Nerve 5, 405-410 23 McCann, W F X and Rosenberry, T L (1977) Arch Blochem Blophys 183, 347-352 24 Rosenberry, T L and Richardson, J M (1977) Biochemistry 16, 3550-3558 25 Grass1, J, Vlgny, M and Massouhe, J (1982) J Neurochem 38, 457-468