Proc. Nati. Acad. Sci. USAVol. 80, pp. 6698-6702, November 1983Neurobiology

Acetylcholinesterase of mammalian neuromuscular junctions:Presence of tailed asymmetric acetylcholinesterase insynaptic basal lamina and sarcolemma

(motor endplate/extracellular matrix components/acetylcholine receptor/16S acetylcholinesterase)

PATRICK A. DREYFUS, FRAN;OIS RIEGER, AND MARTINE PIN(ON-RAYMONDInstitut National de la Sante et de la Recherche MWdicale U153, 17, rue du Fer-A-Moulin, 75005 Paris, France

Communicated by George B. Koelle, June 22, 1983

ABSTRACT A sarcolemma-rich fraction can be isolated aftersubeellular fractionation of mouse intercostal muscles by sedi-mentation on a discontinuous sucrose gradient. The quantitativerecovery of the acetylcholine receptor in this fraction is about 50%,which indicates the presence of a high proportion of postsynapticmembranes. Acetylcholinesterase (AcChoEase; EC 3.1.1.7) is foundmainly in three different layers: the top layer, which contains sol-uble AcChoEase, the intermediate layer (fraction A), and the last,AcChoR-rich, layer (fraction C). The relative proportions of themolecular forms of AcChoEase are different in the three layers.The "16S" AcChoEase is in a higher proportion in both types ofmembrane fractions (A and C) compared to soluble AcChoEase.Both total AcChoEase and 16S AcChoEase are enriched in the Aand C fractions. In the C fraction, the sequential use of homog-enizations in the presence of detergent and high ionic strengthallows the "solubilization" of two distinct AcChoEase pools. Oneis detergent-soluble and mainly composed of slow-sedimentingforms; the other one is detergent-insoluble, high-ionic strength-soluble, and composed mainly of collagen-like, tailed, asymmetric(16S) AcChoEase. Thus, most of the asymmetric AcChoEase isspecifically localized in the synaptic extracellular matrix of themammalian muscle fiber. However, in the A fraction, most of the16S AcChoEase found is solubilized by detergent alone, suggest-ing an association with microsomal membranes. It may mean thatat least some of the basal lamina-embedded 16S AcChoEase ispreassembled intracellularly in the sarcoplasmic reticulum.

The 16S form of acetylcholinesterase (AcChoEase; EC 3.1.1.7)is highly concentrated in the regions of muscle containing end-plates (1) and is absent or barely detectable after denervation(1, 2). Membrane fractionation experiments have been widelyused to determine the cellular loci of the AcChoEase accu-mulation and localization. For example, muscle microsomal orsarcoplasmic reticulum, and also sarcolemmal fractions, havebeen prepared and analyzed for their enrichment in Ac-ChoEase activity or content of various molecular forms. In par-ticular, the sarcolemmal fraction isolated by McLaughlin et al.(3) was enriched for both 16S AcChoEase and extracellular ma-trix components (basal lamina). There is indirect evidence, es-sentially from studies on electric fish AcChoEase, that theasymmetric forms of AcChoEase, which are composed of tet-rameric protomers and a collagen-like multistranded tail (4-8),can interact with basal lamina components such as glycos-aminoglycans of unknown nature (9), fibrous material (10),or fibronectin (11). In mammalian skeletal muscle, 16S Ac-ChoEase has been found to be collagenase sensitive (12) andthus is probably homologous to the electric fish tailed enzyme.In view of its privileged localization in the region rich in motor

endplates and its possible crucial role in cholinergic neuro-transmission, it is of prime importance to determine the cel-lular structures with which it is associated and the interactionsin which this complex macromolecular component is involved,in particular at the neuromuscular junction. In this paper, wereport a muscle cell subfractionation that allows the isolation ofa neuromuscular junction-enriched fraction, characterized byits specific content in pre- and postsynaptic proteins, namelycholine acetyltransferase (ChoAcTase; acetyl-CoA:choline 0-acetyltransferase, EC 2.3.1.6) the acetylcholine receptor pro-tein (AcChoR), and 16S AcChoEase. This preparation allowsdirect biochemical studies of various neuromuscular junctionstructures or protein components that were not possible pre-viously. In particular, taking advantage of the detergent insol-ubility of synaptic basal lamina material, we have obtained di-rect proof that this material specifically contains most of thejunctional 16S AcChoEase.

MATERIAL AND METHODSIn each experiment we used 20 3- to 4-week-old mice of theC3H, Cad+/Cad+ strain (INSERM U153). The animals weresacrificed by cervical dislocation and the rib cage was imme-diately dissected out and freed from the superficial parathor-acic muscles. All subsequent procedures were performed in acold room (4-6°C). The tissues (intercostal muscles and ribs)were homogenized in Krebs/phosphate medium [5% (wt/vol);Brinkmann Polytron; 1 min] and the crude homogenate was fil-tered through four stainless steel filters (1,000-, 500-, 250-, and125-,um pore size). The filtrate was centrifuged (800 X g; 15min; IEC centrifuge) and the pellet was resuspended in 48 mlof Krebs medium. Eight milliliters of this suspension (crudehomogenate) was layered on top of a preformed discontinuoussucrose gradient (layers of 8 ml of 1.8 M, 1.4 M, 1 M, and 0.6M sucrose). Sedimentation was for 150 min at 7,500 x g in aBeckman L8 ultracentrifuge with an SW 27 rotor. The mainfractions obtained were, from the meniscus to the bottom: aclear supernatant ("soluble" fraction), three bands [A in the 0.6M sucrose layer, B in the 1 M sucrose layer, and C (usually dou-ble-Cl and C2) in the 1.4 M sucrose layer], and a thick pellet.Each fraction was collected by aspiration, diluted 1:2 with Krebsmedium, and centrifuged at 22,000 rpm for 35 min (L8 ultra-centrifuge; SW 27 rotor). The supernatants were discarded andthe pellets constituted the primary fractions.

Each primary fraction was then submitted to an extractionprocedure either in a single step in a standard medium (1% Tri-ton X-100/1 M NaCl/0.001 M EGTA/0.01 M Tris'HCl, pH

Abbreviations: AcChoEase, acetylcholinesterase; BtChoEase, butyryl-cholinesterase; ChoAcTase, choline acetyltransferase; AcChoR, acetyl-choline receptor; a-BgTx, a-bungarotoxin; ECM, extracellular matrix.

6698

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertise-ment" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

7, 2

021

Proc. Natl. Acad. Sci. USA 80 (1983) 6699

7.2) or in two successive steps, in the following order: the pri-mary pellet was resuspended by vigorous shaking in 1 ml of adetergent medium, composed of 1% Triton X-100 and 0.01 MTris HCI, pH 7.2. After centrifugation the pellet was resus-pended in high salt medium (1 ml of 1 M NaCI/0.001 M EGTA/0.01 M Tris-HCI, pH 7.2). The supernatants of both centrif-ugations were kept for further analysis. In a few experiments,the procedure was performed in the inverse order.

Cytochemical Staining. A modification of the method ofKoelle and Friedenwald (13) was used to stain motor endplateAcChoEase. Each primary fraction-A, B, and C-was dis-persed into a 0.2 M Tris maleate/NaOH, pH 6.5/0.6% glu-taraldehyde fixative and kept overnight. Then, after centrifu-gation, the pellets were suspended in the standard pH 5 in-cubation medium, which has acetylthiocholine iodide as a sub-strate. After two washings by centrifugation in the acetic acid/sodium acetate, pH 5.0, buffer used in the incubation, 3% po-tassium ferrieyanide was used to resuspend the pellets and thebrown coloration was allowed to develop for 15 min. Two ad-ditional washings were performed before mounting the prep-aration in pure glycerol for light microscopy.AcChoR and Binding of '-51-Labeled a-Bungarotoxin (12MI-

a-BgTx). The intact rib cages were incubated in 10 ml of a cul-ture medium (14) containing '25I-a-BgTx (15-20 Ci/,ug; NewEngland Nuclear; 1 Ci = 3.7 X 1010 Bq) at 0.2 Ag/ml for 30min at room temperature. Then the tissues were washed twicewith the culture medium and a third time with the Krebs me-dium. Finally, all the fractions were solubilized and subjectedto sedimentation analysis (14).AcChoEase and Butyrylcholinesterase and Their Multiple

Molecular Forms. Total AcChoEase or butyrylcholinesterase(BtChoEase; cholinesterase; EC 3.1.1.8) activity was estimatedby the method of Ellman et al., using acetylthiocholine or bu-tyrylthiocholine iodide, respectively, as substrate (15). Analysisof the multiple forms of AcChoEase was performed on a 5-20%continuous sucrose gradient (16). The specific reversible Ac-ChoEase inhibitor BW 284 C 51 (Burroughs Wellcome, Re-search Triangle Park, NC) or the irreversible BtChoEase in-hibitor isooctamethyl pyrophosphoramide (Sigma) were some-times used to assess enzyme specificities.

ChoAcTase. ChoAcTase activity was determined by themethod of Fonnum (17), using [1-'4C]acetyl-coenzymeA (Amer-sham; 50 mCi/,umol), and is expressed as ,umol of ['4C]ace-tylcholine synthesized per min. Protein concentrations wereestimated by the method of Lowry et al. (18).

Collagenase Treatment. Pure collagenase (240 internationalunits; Advance Biofactures, Lynbrook, NY) was added to 100-,ul aliquots of high-salt extracts of the various subcellular frac-tions. Ca2+ concentrations were adjusted to 0.01 M. The col-lagenolytic action was allowed to occur at 37°C for 1 hr, andsedimentation analysis was started immediately.

RESULTS

Analysis of Primary Subeellular Fractions. Distribution ofAcChoR. '25I-a-BgTx binding sites are highly concentrated inthe C fractions. Table 1 shows that about 50% of the specificbinding of '"I-a-BgTx is found in the C1 fraction. Only 6% isfound in the supernatant fraction, and the rest of the specificbinding is found in the C2 fraction (18%) and in the pellet (36%).It is in the C1 fraction that the specifically bound radioactivityper milligram of protein is at its highest value: more than 10times higher than in the crude homogenate before fractionation(Table 1).

Evidence for isolated motor endplates in the AcChoR-richfraction. Fig. 1 shows several aspects of the numerous sub-cellular structures that stain heavily after AcChoEase cyto-

Table 1. Distribution of a-BgTx binding sites (AcChoR) aftersubcellular fractionation of mouse intercostal muscles:Isolation of a junctional membrane-enriched fraction

Total Specific125I-a-BgTx,* activity,cpm/g fresh cpm/mg Recovery,

Fraction tissue protein %

Crude homogenate 18,235 ± 551 151 ± 5 100Supernatant (top

of gradient) 1,096 ± 94 356 ± 31 6Fraction A NS - -

Fraction B NS - -

Fraction C1 9,027 ± 466 1,514 ± 79 49.5Fraction C2 3,243 ± 302 702 ± 67 17.8Pellet 6,588 ± 230 - 36.1

Results are the mean ± SEM of three independent experiments. Thefresh tissue is the bulk of the costal grids, weighed immediately afterdissection, ribs included. NS, not significant.* Specific binding (after gradient analysis).

chemistry. This staining is mainly due to AcChoEase, becauseit is not obtained in the presence of 10 ,uM BW 284 C 51, thespecific inhibitor of AcChoEase. The morphological appear-ance of these subcellular structures (size and internal organi-zation, with apparent postsynaptic folds in many of them) sug-gests the presence of numerous well-preserved motor endplatesin fraction C1.

Distribution of ChoAcTase. The major part of muscle Cho-AcTase activity is solubilized in the procedure and is found inthe supernatant of the discontinuous gradient after fractiona-tion (about 60% of total tissue ChoAcTase activity). Only 5% oftotal ChoAcTase is found in the isolated motor endplate-richfraction C1. If we hypothesize that ChoAcTase is essentiallylocalized in the nerve terminals, then only a small proportionof the presynaptic cytoplasm is recovered in the C1 fraction.

a

f.,..;# ..:: ,-Rae

b 9olI

P

rI

c

1oIr,",

d

FIG. 1. Isolated motor endplates in a synaptic membrane-en-riched subcellular fraction (Cl fraction). Cytochemical staining of Ac-ChoEase (19). (a) Intact muscle fibers: region rich in motor endplates.(b) Suspension of isolated stained motor endplates (large field view).(c and d) Selected examples of isolated motor endplates, with theirsynaptic gutters.

Neurobiology: Dre*s et A

Ah.&L.

P

. 4

N.. -4,

..:t

.%.,

Dow

nloa

ded

by g

uest

on

Mar

ch 2

7, 2

021

6700 Neurobiology: Dreyfus et al.

4.' S14

C.) co

4a~ ~ ~

5 10 15 20 25 30 35Fraction

FIG. 2. Sedimentation analysis of AcChoEase and its molecularforms: Relative enrichment of 16S AcChoEase in the C1 fraction. Com-plete solubilization medium was used. . , Crude homogenate be-fore subcellular fractionation; *-*, C1 fraction. The two curves can-not be directly quantitatively compared, because their incubation timesare different. Here and in Figs. 3 and 4: Men., meniscus; size markersare 3galactosidase (I3GZ), catalase (Cat), and alcohol dehydrogenase(ADH).

Distribution of AcChoEase and its multiple molecularformsin the primary fractions. AcChoEase activity is found in mostof the fractions studied, but in variable amounts, and is also

inequally distributed in its multiple molecular forms. The fivemain molecular forms of AcChoEase are identified by theirsedimentation coefficients: 16 S, 12 S, 10 S, 6.5 S, and 4 S. Theirsedimentation profiles are composed of three main peaks ofactivity (Fig. 2): "16S" (16S and minor accompanying forms),"10S" (1OS and a small 12S contribution), and "4S" (4S and a

minor 6.5S contribution). The specific activity of total Ac-ChoEase is about 10-fold higher in the Cl fraction than in thecrude homogenate (Table 2), which represents a purificationfactor of the same order as for AcChoR in this fraction. The rel-ative proportion of 16S AcChoEase is significantly higher in theCl fraction than in the crude homogenate before fractionation(Fig. 2) and is about 33% compared to 20% in the crude ho-mogenate (Table 2). The AcChoEase activity found as 16SAcChoEase in the C1 fraction represents 10% of 16S Ac-ChoEase activity in the crude homogenate. We found ratherhigh total and specific AcChoEase activities in the A fraction,higher than in the crude homogenate or any other fraction. Itis in the same fraction that 10S AcChoEase is predominant (45%

Table 3. Distribution of total AcChoEase and tailed symmetric16S AcChoEase after solubilization of the subcellular fractionsin the detergent medium

Specific activity, 16-19SRecovery, AA412/hr per AcChoEase,

Fraction % mg protein % of totalCrude homogenate 100 12.0 ± 1.04 6.5 ± 1.4Fraction A 42.2 25.5 ± 0.64 21.3 ± 4.1Fraction B 1.5 3.3 ± 0.14 13.9 ± 5.3Fraction C1 22.5 2.4 ± 0.33 13.9 ± 5.9Fraction C2 24.2 6.5 ± 0.09 7.6 ± 2.3

Results are the mean ± SEM of three independent experiments.

of total AcChoEase). The A fraction does not contain any sig-nificant amount of AcChoR or any cytochemically identifiableisolated motor endplates. The C2 fraction contains a higher rel-ative proportion of 16S AcChoEase than the C1 fraction (39%)but less AcChoR and less ChoAcTase activity.

Sequential Solubilization of the Primary Subeellular Frac-tions by Detergent and High-Salt Media. Extracellular MatrixIsolation and Association with Asymmetric AcChoEase. De-tergent treatment. The treatment of the primary fractions bythe detergent-containing (1% Triton X-100) and salt-free me-dium results in the total "solubilization" of specifically bound'"I-a-BgTx in all fractions where it has been found, becauseit is quantitatively recovered in the supernatant of the low-speedcentrifugation. Fig. 3A shows the "solubilization" obtained fromthe AcChoR-rich C1 fraction by the detergent treatment, andTable 3 shows the recovery of total AcChoEase and the low per-centages of 16S AcChoEase recovered in the supernatants fromeach fraction (about 14% in the C1 fraction).

High-salt treatment of detergent-insoluble material. The de-tergent-insoluble material can be "solubilized" by high-salt (1M NaCl) media and is most probably composed of the extra-cellular matrix material of the muscle fibers. In the C1 fraction,which is highly enriched in AcChoR and 16S AcChoEase, weare then dealing with extracellular matrix material mostly ofsynaptic origin. The high-salt treatment "solubilizes" a very highproportion of 16S AcChoEase (Fig. 3B). No AcChoR is solu-bilized from the pellet of the detergent-treated C1 fraction byhigh concentrations of salt. Table 4 shows the relative propor-tions of the 16S peak of AcChoEase, after extraction of the de-tergent-treated subcellular fractions by the high-salt medium;in the C1 fraction, 16S AcChoEase represents 61.2 ± 2.1% oftotal AcChoEase recovered on the sucrose gradient. After a shortexposure to pure collagenase at 37°C the high-salt-extracted 16SAcChoEase is readily converted into a major 10S form, withoutloss of AcChoEase activity (Fig. 3B), demonstrating that thisspecies is a collagen-like tailed form of AcChoEase.

BtChoEase, and particularly its 16S molecular form, is barely

Table 2. Distribution of total AcChoEase and its multiple molecular forms after subcellularfractionation of mouse intercostal muscles

Specific Relative proportions,activity Recoveportons,

Recovery, AA412/hr per % of total recovered AcChoEaseFraction % mg protein* 4-6.5S 10-12.5S 16-19S

Crude homogenate 100 0.28 ± 0.001 57.0 ± 3.5 22.2 ± 2.7 19.7 ± 1.2Supernatant 12.9 6.55 ± 0.37 54.6 ± 0.6 25.6 ± 2.4 13.1 ± 2.6Fraction A 40.2 17.49 ± 0.86 26.2 ± 3.0 45.5 ± 4.6 24.1 ± 2.7Fraction B 10.5 1.99 ± 0.07 27.2 ± 3.9 37.5 ± 3.8 32.7 ± 3.2Fraction C1 16.0 2.82 ± 0.06 28.5 ± 1.8 32.9 ± 2.3 33.2 ± 1.1Fraction C2 11.4 4.31 ± 0.06 22.0 ± 2.3 32.1 ± 2.0 39.1 ± 2.0

Results are the mean ± SEM of three independent experiments.* One A412 unit corresponds to 73.5 ,umol of hydrolyzed acetylthiocholine.

Proc. Natl. Acad. Sci. USA 80 (1983)

Dow

nloa

ded

by g

uest

on

Mar

ch 2

7, 2

021

Neurobiology:Dreyfusetal.~~~~Proc.Nati. Acad. Sci. USA 80 (1983) 6701

.~.

Cd tC.).-d<eqS.

CZ 1-

0

'--4

x

in.3 C

1

Fraction

FIG. 3. Sequential solubilization of tailed asymmetric AcChoEase and ofAc~hoR from the Cl fraction. (A) Detergent solubilization (NaCl-free).(B) High-salt solubilization (Triton-free) of the centrifugation pellet obtained inA. O-* , AcChoEase; m....U, AcChoR (1251-a-BgTx binding, ex-pressed as total cpm per fraction after sedimentation analysis); A.A.. , AcChoEase after collagenase treatment.

detectable in the high-salt-soluble synaptic extracellular matrix(ECM) material (Fig. 4). About 42 ± 5% of the initial Bt-ChoEase activity present in the crude homogenate is recoveredin the Cl fraction. Only 11% of the BtChoEase of the Cl frac-tion can be "solubilized" by high salt, which results in aBtChoEase recovery of 4.5% of the total activity present in thecrude homogenate. Thus, BtChoEase, and especially its veryminor 168 component, is not accumulated in the synaptic ECMmaterial.

Increased extraction yield of total and 16S AcChoEase in thesequential double-step procedure, compared to the single-stepprocedure. We consistently found more total AcChoEase ex-tracted with detergent followed by high salt than with deter-gent and high salt added together. Table 5 shows that it is inthe Cl fraction that both total AcChoEase and 168 AcChoEaseare recovered in highly increased amounts in the double-stepprocedure. The two-step versus one-step ratios for 168 AcCho-Ease were 1.93, 1.34, and 4.95 for the crude homogenate, frac-tion A, and fraction Cl, respectively.

Hydrophilic and hydrophobic 16S AcChoEase. It is of in-terest to estimate the relative amounts of 168 AcChoEase foundin each "solubilization" step (detergent followed by high-salttreatments) compared to the sum of 168 AcChoEase solubilizedin both -steps. Table 6 shows that 168 AcChoEase may have twodifferent solubility behaviors: in the C1 fraction, where 58% ofthe original crude homogenate "168" AcChoEase is found, mostof it is "solubilized" in the high-salt step (88%). Moreover, whenthe primary fraction Cl is first treated by high salt, then bydetergent (inversion of the two steps of "solubilization"), the

Table 4. Distribution of total AcChoEase and tailed asymmetric16S AcChoEase after solubilization of the detergent-treatedsubcellular fractions in the high-salt medium

Specific activity, 16-19SRecovery, AA412/hr per AcChoEase,

Fraction %mg protein % of totalCrude homogenate 100 6.5 ± 0.08 45.3 ± 2.5Fraction A 22.2 23.03 ± 0.03 27.0 ± 8.9Fraction B 7.4 2.88 ± 0.07 47.3 ± 4.7Fraction C1 54.2 4.13 ± 0.26 61.2 ± 2.1Fraction C2 6.8 1.19 ± 0.09 55.2 ± 2.8

Results are the mean -± SEM of five independent experiments.

previous result holds: the percentage of 168 AcChoEase re-covered in the high-salt extraction, when this step is performedfirst, is found to be 94 + 10% (n = 2) for the fraction C1.

In contrast to the Cl fraction, the detergent step extractsmost of the 168 -AcChoEase contained in the A fraction. Thus,168 AcChoEase must have a different subcellular location inthe fractions Cl and A: the synaptic ECM of the isolated motorendplate in fraction Cl and the lipid-rich membranes fromplasmalemma or sarcoplasmic reticulum in fraction A.

DISCUSSIONOur results constitute direct evidence for the association of tailedasymmetric AcChoEase with ECM material of mammalianmuscle fiber. Moreover, we find tailed asymmetric AcCho-Ease in synaptic membrane-enriched subcellular fractions,

/3GZ Cat ADH Men.

a)-4

i~~~~~~~~~~~~~~~i

5 10 15 20 25 30 35Fraction

FIG. 4. BtChoEase and its molecular forms in the motor endplate-enriched fraction Cl: Detergent and high-salt extractions. The sub-strate was butyrylthiocholine iodide. e...., Extraction by detergentand high salt (standard medium); .-*, extraction by high salt. Thetwo curves are quantitatively directly comparable.

Neurobiology: Dreyfus et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

7, 2

021

6702 Neurobiology: Dreyfus et al.

Table 5. Increased yield of total AcChoEase in the sequentialextraction procedure

Total AcChoEase activity,AA412/hr per g tissue

Two-stepFraction One-step* First stept Second step*

Crude homogenate 154.3 ± 22.9 147.2 ± 39 108.5 ± 35.0Fraction A 62.0 ± 6.5 62.1 ± 39 25.1 ± 0.9Fraction C1 24.7 ± 0.3 33.1 ± 0.5 58.8 ± 8.9

Results are mean ± SEM.* Detergent plus high salt (n = 3).t Detergent (n = 3).t High salt (n = 5).

which suggests that it is probably concentrated in the musclejunctional basal lamina sheaths. However, we find that ECMis not the unique localization of tailed asymmetric AcChoEase.A number of studies had previously shown that asymmetric formsof AcChoEase are highly concentrated in the motor endplate-rich regions of rat or mouse muscle (1, 2). The possibility, orlikelihood, of the association of junctional AcChoEase withmuscle basal lamina arose from the observation that collagenasesolubilized most of AcChoEase activity found in the synapticcleft (19). However, it was discovered that asymmetric Ac-ChoEase was collagenase sensitive (5, 8, 9, 12) and its tail partwas collagen-like (20). The solubilization from tissues may thushave occurred as a consequence of a direct collagenolytic actionon AcChoEase itself. Only a direct study of the muscle ECMmaterial could settle this point. The direct isolation of ECM ismade simply by its solubility properties. It is not solubilized bynonionic detergent only, but it is soluble in the presence of highlyconcentrated salts (21, 22). We did not use the intact muscle forisolation, but subcellular membranous fractions obtained afterdiscontinuous sucrose gradient centrifugation. One of thesefractions, the C fraction, composed of two distinct bands, con-tains a very high proportion of both AcChoR and asymmetricAcChoEase, which are found concentrated in intact muscle, inthe region rich in motor endplates. The degree of contami-nation of the synaptic membrane-enriched C fraction by ex-trajunctional material is difficult to assess, but we may inferfrom the high density of equilibration of this fraction that otherplasmic or microsomal membranes are not major contaminants.We thus studied a neuromuscular junction-enriched prepara-tion. This preparation was submitted to the ECM isolation pro-cedure. In the first step of the procedure, a detergent solu-bilization, we obtained total solubilization of the AcChoR.Therefore, most of the postsynaptic plasmic membrane ma-terial must have been solubilized. The resulting detergent-in-soluble material is probably mainly composed of the synapticECM. However, it may still be associated with an unknownamount of the cytoskeleton and other nonextracted membraneproteins (23, 24). In the second step of the procedure, a high-salt extraction, a major part of the detergent-insoluble materialis solubilized. It is at this stage that most of asymmetric AcCho-Ease is solubilized. Asymmetric AcChoEase, which is com-

Table 6. Two different solubility behaviors of 16S AcChoEaseHigh-salt vs. total

Fraction 16S AcChoEase, %Homogenate 74.6 ± 13.7Fraction A 20.4 ± 8.5Fraction C1 85.6 ± 3.9

Total 16S AcChoEase is that extracted by high salt plus detergent.Results are mean ± SEM for three to five independent experiments.

posed mainly of 16S AcChoEase, but also of other minor peaks(18.5S and 12S), is thus embedded in the ECM material or inthe portion of it that may be solubilized by high-salt media.We found that both total AcChoEase and (especially) 16S

AcChoEase are recovered in much higher yields when deter-gent and high-salt extractions are sequentially used rather thanthe single-step procedure, when detergent and high salt areboth present in the extraction medium. We have no clear ex-planation for this phenomenon, which also occurs when highsalt (with EGTA) is utilized first in the sequential procedure.It is interesting to note that both steps give nearly as much eachas the single-step procedure, at least for the C1 fraction. Someprotection or activation may occur in the sequential procedurethat does not act with the simultaneous use of these extractingagents. Some of muscle asymmetric AcChoEase is found in alighter, low density fraction (A) and is solubilized by detergent.It clearly corresponds to a very distinct pool of 16S AcChoEase,which is associated not with the basal lamina but with some li-pidic membranes. These lipidic membranes may be either ex-trajunctional plasma membranes or, more probably, sarco-plasmic reticulum membranes, as recently suggested by electronmicroscopy. This lipid membrane-associated 16S AcChoEase isprobably physicochemically homologous to the chick retina (20S)asymmetric AcChoEase, which has been proposed to be an-chored into the cell plasma membrane (25). In contrast, Bt-ChoEase is nearly absent from ECM material. Our results callfor a detailed biochemical study of the muscle ECM material,to explore further the fundamental aspects of the cellular bio-synthesis, transport, and localization of AcChoEase.

1. Hall, Z. W. (1973) J. Neurobiol. 4, 343-362.2. Vigny, M., Koenig, J. & Rieger, F. (1976)J. Neurochem. 27, 1347-

1353.3. McLaughlin, J., Engel, K. W. & Boggi, R. N. (1978) J. Neuro-

chem. 31, 783-788.4. Rieger, F., Bon, S., Massoulie, J. & Cartaud, J. (1973) Eur. J.

Biochem. 34, 439-457.5. Rieger, F., Bon, S. & Massoulie, J. (1973) FEBS Lett. 36, 12-36.6. Dudai, Y., Herzberg, M. & Silman, I. (1973) Proc. Natl. Acad. Sci.

USA 70, 2473-2476.7. Cartaud, J., Rieger, F., Bon, S. & Massoulie, J. (1975) Brain Res.

88, 127-130.8. Lwebuga-Mukasa, J. S., Lappi, S. & Taylor, P. (1976) Biochem-

istry 15, 1425-1434.9. Bon, S. & Massoulie, J. (1978) Eur. J. Biochem. 89, 89-94.

10. Viratelle, D. M. & Bernhard, S. A. (1980) Biochemistry 19, 4999-5007.

11. Emmerling, M. R., Johnson, C. D., Mosher, D. F., Lipton, B.H. & Lillien, J. E. (1981) Biochemistry 20, 3242-3247.

12. Rieger, F., Ruberg, M. & Shelanski, M. L. (1979) Brain Res. 170,568-571.

13. Koelle, G. B. & Friedenwald, J. S. (1949) Proc. Soc. Exp. Biol. Med.70, 617-622.

14. Rieger, F. & Pinqon-Raymond, M. (1981) Dev. Biol. 87, 85-101.15. Ellman, G. L., Courtney, K. D., Andres, V. & Featherstone, R.

M. (1961) Biochem. Pharmacol. 7, 88-95.16. Rieger, F., Koenig, J. & Vigny, M. (1980) Dev. Biol. 76, 358-365.17. Fonnum, F. (1975) J. Neurochem. 24, 407-409.18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.

(1951)1. Biol. Chem. 193, 265-275.19. Hall, Z. W. & Kelly, R. B. (1971) Nature (London) New Biol. 232,

62-63.20. Anglister, L. & Silman, I. (1978)J. Mol. Biol. 125, 293-311.21. Kefalides, N. A., Alper, R. & Clark, C. C. (1979) Int. Rev. Cytol.

61, 167-228.22. Fulton, A. B., Prives, J., Farmer, S. R. & Penman, S. (1981)J.

Cell Biol. 91, 103-112.23. Schliwa, M. (1982)J. Cell Biol. 92, 231-236.24. Prives, J., Fulton, A. B., Penman, S., Daniels, M. P. & Chris-

tian, C. N. (1982)1. Cell Biol. 92, 231-236.25. Barat, A., Escudero, E., Gomez-Barriocanal, J. & Ramirez, G.

(1981) Neurosci. Lett. 20, 205-210.

Proc. Natl. Acad. Sci. USA 80 (1983)

Dow

nloa

ded

by g

uest

on

Mar

ch 2

7, 2

021