Horse kidney neutral α-d -glucosidase: purification of the detergent-solubilized enzyme; comparison with the proteinase-solubilized forms

Biochimica et Biophysica Acta

831 (1985) 59-66

59

Elsevier BBA 32285

Horse kidney neutral a-D-glucosidase: purification of the detergent-solubilized enzyme; comparison with the proteinase-solubilized forms Jean Giudicelli a, Maryse Boudouard a, Pascale Delqu6 a, Christian Vannier b and Pierre Sudaka a a Laboratoire de Biochimie, Facultb de Mbdecine, Chemin de Valombrose, and b Centre de Biochimie, Parc Valrose, 06034 Nice Cedex (France)

(ReceivedJune 3rd, 1985)

Key words: Neutral a-D-glucosidase;Amphipathicform; Proteolyticform; Brushborder enzyme;(Horse kidney) Neutral a-D-glucosidase (a-D-glucoside glucohydrolase, EC 3.2.1.20) from horse kidney brush-border membranes was solubilized using Emulphogene BC 720 and purified by an affinity chromatography technique. The enzyme preparation (390-fold purified), which was free of other known microvillus hydrolases, exhibited one precipitate line in crossed immunoelectrophoresis and migrated as a single band on sodium dodecyl sulfate polyacrylamide gel electrophoresis. Several criteria (charge-shift crossed immunoelectrophoresis and hydrophobic chromatography) revealed the purified detergent form of the enzyme to be an amphipathic molecule. The papain treatment of either brush-border membrane vesicles or the purified detergent form of neutral a-D-glucosidase released an enzymatic form devoid of these amphipathic properties. Conversely, after trypsin treatment of the 'd' form of the enzyme, two enzymatic forms were obtained: the first and major form retained these amphipathic properties; the second form exhibiting the same properties as the papain-released form. Furthermore, only a very small amount of neutral a-D-glucosidase can be released after trypsin solubilization of brush-border membrane vesicles, and the released enzyme did not exhibit amphipathic properties. These results were interpreted as meaning that the trypsin attack site on the detergent form of the enzyme had either poor affinity for, or obstructed access to, the proteinase when the enzyme was integrated in native membrane or in Triton X-100 miceiles, whereas the proteolytic site of the papain was always accessible.

Introduction

The majority of the kidney and intestinal microvillar membrane bound enzymes from various species have been purified in proteolytic and detergent forms and their molecular and kinetic properties have been described. The difference between these two enzyme forms has been explained by the existence of a hydrophobic area separable by proteolytic cleavage from the hydrophilic (proteolytic form) functional part of these hydrolases. These hydrophobic areas embedded in the lipid matrix serve as an anchor. Conforming this hy-

pothesis, these detergent-solubilized enzymes release low-molecular-weight hydrophobic fragments after proteolytic treatment [1-4]. One of these enzymes, neutral a-o-glucosidase (a-a-glucoside glucohydrolase, EC 3.2.1.20), has previously been purified as a proteolytic solubilized form from the membrane of enterocytes [5-9] and proximal renal tubules [10-12] of some species. However, this membrane-bound hydrolase has only been purified in a detergent-solubilized form from the enterocytes of pig and rat [3,13,14]. Interestingly, in this last case, the detergent-binding properties could be obtained only when enzymatic

0167-4838/85/$03.30 © 1985 Elsevier SciencePublishers B.V. (BiomedicalDivision)

60 purification was performed in the presence of proteinase inhibitors [15]. No investigation on purification of the detergent-solubilized form of kidney neutral a-D-glucosidase has been described to date. This paper reports on a purification procedure for detergent-solubilized horse kidney neutral aD-glucosidase including an affinity chromatography step. The study of the amphipathic nature of this enzyme, purified either by detergent or by proteinases, is described. The fact that the detergent-purified enzyme ex~obited strong binding with a hydrophobic gel and charge-shift properties indicated that this form of the protein presented a hydrophobic area. The differences observed between the two proteol~tic forms of neutral a-Dglucosidase are discussed. The present work was presented in part at the 15th FEBS Meeting, July 24-29, 1983, Brussels, Belgium. Materials and Methods

Materials CH-Sepharose-4B and phenyl-Sepharose CL-4B were obtained from Pharmacia; Indubiose AcA202 from Industrie Biologique Fran~aise; lyophilized trypsin from Worthington; papain from NBC; sodium deoxycholate and Triton X-100 from Merck; agarose (mr = - 0.13) was purchased from Bio-Rad Laboratories and cetyltrimethylammonium bromide from Serva; soluble starch according to Zulkowsky from Merck, the soluble starch was dialyzed against distilled water and lyophilized prior to use. All other reagents were either previously described [11] or were of the best available purity.

Methods Assays. Neutral a-D-glucosidase and glucoamylase activities were measured as previously described in the presence of' 0.005 M of maltose or 10 m g / m l of soluble starch in sodium citrate 0.1 M buffer (pH 6.5) [16]. One unit of a-D-glucosidase activity was the amount of enzyme required to hydrolize 1 btmol of maltose/rain under the assay conditions. Aminopeptidase N (a-aminoacyl-peptide hydrolase, EC 3.4.11.1), y-glutamyltransferase ((5-glutamyl)peptide: amino acid 5glutamyltransferase, EC 2.3.2.2), alkaline phos-

phatase (orthophosphoric-monoester phosphohydrolase, EC 3.1.3.1) and trehalase (a,a-trehalose glucohydrolase, EC 3.2.1.28) were determined using i.-alanine p-nitroanilide, y-glutamyl p-nitroanilide, p-nitrophenyl phosphate and trehalose respectively as substrate, as indicated in Ref. 17. proteins were determined according to the method of Lowry et al. [181 under general conditions. In the presence of Triton X-100, protein concentrations were determined by the method of Gerristein et al. [19]. Under all conditions, bovine serum albumin was used as standard. Neutral a-D-glueosidase purifications. Fresh horse kidneys were obtained at the slaughterhouse immediately after death. All experiments were carried out at 4°C. The 'trypsin' form of neutral a-D-glucosidase ('Ti' form) was purified as previously described [11] with some variations. Thus, the affinity chromatography step was performed with paminophenyl-a-D-glucopyranoside as ligand instead of p-aminophenyl-fl-D-maltoside. The enzyme was eluted from affinity chromatography by 0.075 M p-nitrophenyl-a-D-glucopyranoside in an elution buffer consisting of 0.15 M sodium phosphate (pH 7.0). The 'papain' form of neutral a-D-glucosidase ('p' form) was purified from horse kidney brushborder preparations [16]. After membrane protein digestion for 60 min with 3% papain (w/w) relative to proteins [20], the 'p' form of the enzyme, recovered in the 105 000 × g, 60 min, supernatant, was purified by affinity chromatography as described earlier for the 'Ti' form. The 'detergent' form of neutral a-D-glucosidase ('d' form) was purified in the same way as the proteolytic form, except that proteolytic treatment with trypsin was omitted and affinity chromatography was performed in the presence of 0.1% Emulphogene BC 720 in 0.15 M sodium phosphate buffer (pH 7.4). Preparation ofantisera. Rabbits were immunized with the purified 'Ti' form of neutral a-D-glucosidase or whole vesicles brush-border membrane proteins as described by Louvard et al. [201 and Wroblewski [21], respectively. The antisera obtained were stored at - 6 0 ° C and dialyzed overnight against 0.15 M phosphate buffer saline (pH 7.4) prior to use.

61

Crossed immunoelectrophoresis. Crossed immunoelectrophoresis was performed according to Laurell [22] in 0.04 M veronal buffer (pH 8.6) on a 1 mm thick agarose gel containing 0.1% Triton X-100 (w/v). The first dimension was run at 6°C for 120 rain at 13 V / c m and the second dimension at room temperature for 16 h at 4.5 V/cm. Charge-shift crossed-immunoelectrophoresis was performed in the same conditions as above, except that the agarose and detergents used were in 0.0375 M Tris/0.1 M glycine (pH 8.6) buffer. In the first dimension, the agarose and buffer contained either 0.0125% cetyltrimethylammonium bromide/0.1% Triton X-100 or 0.2% sodium deoxycholate/0.1% Triton X-100 or 0.1% Triton X100 only [23]. In the second dimension, only 0.1% Triton X-100 was added to the agarose. Approx. 2.5 ~g purified neutral c~-D-glucosidase in 6 /~1 were applied. After the second dimension, the plates were incubated, dialyzed and stained with 0.25% Coomassie blue, as previously described [24].

Digestion of membrane vesicles by proteinases. Solubilization of neutral a-D-glucosidase from brush-border membranes by papain was performed as described by Louvard et al. [20] with 3% papain (w/w) relative to proteins, and the membrane suspension was adjusted to 5 mg/ml. Solubilization of the enzyme from brush-border membranes by trypsin was performed as described by Maroux et al. [25] by addition of 10% trypsin (w/w) to the membrane suspension in 0.05 M Tris-HCl (pH 8.0). Aliquots (100 ~1) were removed, diluted 10-fold at 4°C with 0.1 M phosphate buffer (pH 6.2) containing 0.2 mg aprotinin/ml and spun for 30 min at 18000 × g in the angular rotor

of a Haereus Chris centrifuge. No neutral a-D-glucosidase inhibition was observed with aprotinin [11].

Polyacrylamide gel electrophoresis. The homogeneity of the purified 'detergent' enzyme was examined by 7.5% polyacrylamide gel electrophoresis [26]. The gel was stained with 0.25% Coomassie blue. Results

Purification of 'detergent' neutral c~-D-glucosidase Up until ammonium sulfate precipitation, the different steps used for purification of 'detergent' neutral a-D-glucosidase were the same as those used for purification of the 'proteolytic' enzyme [11]. After ammonium sulfate fractionation, the pellet was dissolved in 0.15 M sodium phosphate buffer (pH 6.8) containing 0.1% Emulphogene BC 720 and dialyzed overnight against the same buffer. After centrifugation, the supernatant was layered on the affinity chromatography column consisting of p-aminophenyl-a-D-glucopyranoside coupled to CH-Sepharose-4B and previously equilibrated with dialysis buffer. After elution of the non-adsorbed proteins, the 'd' enzyme was specifically desorbed by 0.075 M of p-nitrophenyl-a-D-glucopyranoside, a substrate of this enzyme [11], in the elution buffer. Fractions containing the enzyme activity, characterized by a strong yellow coloration due to p-nitrophenol formation, were pooled; the pnitrophenol was then eliminated by running the pooled fractions through a column of Indubiose AcA-202. The results of the purification procedure are

TABLE I PURIFICATION OF HORSE KIDNEY 'd' NEUTRAL Ot-D-GLUCOSIDASE

Homogenization Solubilization 30-60% ammonium sulfate fractionation Affinity chromatography (after AcA-202 step)

Total activity (U)

Total proteins (rag)

34.2 30.9

534 135.5

0.064 0.229

100 90.3

l 3.6

21.6

47

0.46

63.2

7.2

18.2

0.73

Specific activity (U/rag)

24.9

Recovery (%)

53.2

Purification factor

389

62

summarized in Table I. Purification increased specific activity by about 390-times, with an approximate yield of 53%. An adaptation of this purification method allows all three forms of the enzyme to be purified from a preparation of horse kidney brush-border membrane vesicles. This method includes solubilization of the membrane proteins with Emulphogene BC 720 [24] followed, when necessary, by proteolysis. One-step enzyme purification is then performed using affinity chromatography. These three forms of neutral c~-D-glucosidase can thus be purified from the kidney in less than 48 h.

Purity of the 'd' neutral C~-D-glucosidase Two criteria of purity were used. When the purified 'd' enzyme was subjected to crossed immunoelectrophoresis using anti-whole membrane antiserum, only one precipitate was observed (Fig. l a): this contrasts with the complex pattern obtained when detergent-solubilized horse kidney brush-border membranes were run against an anti-whole membrane antiserum (Fig. lb). The

purity of 'd' neutral C~-D-glucosidase was checked by SDS-polyacrylamide gel electrophoresis. As shown in Fig. lc, the SDS-denatured 'd' enzyme exhibited a single band. Furthermore, aminopeptidase N, y-glutamyhransferase, alkaline phosphatase and trehalase activities were not detected in the purifed 'd' form of the neutral a-D-glucosidase preparation.

Antigenic identity In order to verify that the 'p' and 'd' neutral C~-D-glucosidases presented the same antigenic characteristics, both purified forms of this protein were subjected to a parallel crossed immunoelectrophoresis [27]. As shown in Fig. 2, both forms of the enzyme were identical. The presence of another precipitate line with the 'd' form of enzyme could be explained by the loss of antigenic determinant(s) during papain treatment. Interestingly, the mobility of the 'd' form of the enzyme was lower than that of the 'p' form; such differences have previously been reported for other integral proteins and complexes [28-30].

Demonstration of the amphipathic properties of neutral C~-D-glucosidase The charge of detergent added to the medium affects protein mobility only if the detergent molecules interact with the whole protein or a part of it. The amphipathic properties of the 'd' enzyme were thus determined by studying mobility vari-

c G

Fig. 1. Purity of detergent neutral a-D-glucosidase. Crossed immunoelectrophoresis of purified 'd' neutral a-D-glucosidase (2.5 /~g) and microvillar proteins (20 ~g). (a), purified 'd' form of neutral a-t~-glucosidase. (b), microvillar membrane Triton Xq00-solubilized proteins; the arrow indicates the neutral c~D-glucosidase peak. The gel for the second dimension had an immunoglobulin content of 14 ~ g / c m z. (c), polyacrylamide gel electrophoresis of neutral a-D-glucosidase, Electrophoresis was carried out at 6 m A / t u b e with a gel of 8x0.5 cm (30 ~tg of purified 'd' neutral a-D-glucosidase); the arrow on the left indicates the origin of the gel. For details see Materials and Methods.

Fig. 2. Parallel crossed immunoelectrophoresis of ' p ' (~,) and 'd' (*) neutral a-o-glucosidase (2.5 ~g). These enzymatic forms were subjected to immunoelectrophoresis under the same conditions as those described in Fig. 1.

63 TABLE II CROSSED CHARGE-SHIFT IMMUNOELECTROPHORESIS OF THE THREE FORMS OF NEUTRAL a-I)-GLUCOSIDASE The mobilities, in the first dimension, are expressed relative to the mobility of human haemoglobin, a hydrophilic marker protein. For details see Material and Methods section. Abbreviations used: CTAB, cetyltrimethylammonium bromide: DOC. sodium deoxycholate; TX-100, Triton X-100. Neutral a-D-glucosidase form 'd' form 'Ti' form 'p' form or 'Ti' form after papain treatment

Detergent used in first dimension TX-100

CTAB/TX-100

DOC/TX-100

1.04 1.04 1.45 (70%) (30%) 1.45

0.794 0.794 - 1.45 (70%) (30%) 1.43

1.42 1.44

a t i o n as a function of the charge of detergent a d d e d to the e l e c t r o p h o r e t i c m e d i u m [23]. In this e x p e r i m e n t , a h y d r o p h i l i c protein, h u m a n hemoglobin, was a d d e d as an internal s t a n d a r d (to e l i m i n a t e effects due to variations in electrop h o r e t i c conditions). W h e n subjected to chargeshift crossed i m m u n o e l e c t r o p h o r e s i s and comp a r e d with i m m u n o e l e c t r o p h o r e s i s p e r f o r m e d in the presence of T r i t o n X-100 alone, the a m p h i p a t h i c nature of the purified ' d ' enzyme was d e m o n s t r a t e d u n a m b i g u o u s l y . Indeed, the relative mobility was increased when s o d i u m d e o x y c h o l a t e was a d d e d to the gel a n d buffer. Conversely, the a d d i t i o n of c e t y l t r i m e t h y l a m m o n i u m b r o m i d e resulted in a c a t h o d i c shift of the enzyme (Table II). T h e p r o t e o l y t i c forms of neutral a-D-glucosidase (purified from the p a p a i n - or t r y p s i n - t r e a t e d ' d ' form of the enzyme) were subjected to crossed i m m u n o e l e c t r o p h o r e s i s u n d e r the same c o n d i t i o n s as described for the ' d ' enzyme: the ' p ' form always p r e s e n t e d the same relative m o b i l i t y (1.44), regardless of the detergent used. W h e n the ' T i ' f o r m was run into c e t y l t r i m e t h y l a m m o n i u m b r o m i d e / T r i t o n X-100 or T r i t o n X-100 alone, two p e a k s were o b t a i n e d . In each case, the first p e a k ( a b o u t 70%) p r e s e n t e d the same relative m o b i l i t y as that o b s e r v e d for the ' d ' neutral a-D-glucosidase when run in the same conditions. The seco n d p e a k ( a b o u t 30%) p r e s e n t e d the same m o b i l i t y as that o b s e r v e d when ' p ' neutral a-D-glucosidase was run into a n y detergent. W h e n the ' T i ' form was run into s o d i u m deoxycholate, o n l y one prec i p i t a t e line, with the same m o b i l i t y as that obt a i n e d with ' p ' form, was o b s e r v e d (Table II). A n alternative criterion of h y d r o p h o b i c i t y for

1.44

the ' d ' form of the enzyme was p r o v i d e d by the b i n d i n g of this e n z y m a t i c form to p h e n y l - S e p h arose CL-4B. The results o b t a i n e d are shown in T a b l e III. Both the ' T i ' a n d ' d ' enzymes exhibited c o n s i d e r a b l e b i n d i n g on this h y d r o p h o b i c chrom a t o g r a p h y , although the ' d ' form presented higher b i n d i n g than the ' T i ' form. In the same conditions, o n l y a very low percentage of the ' p ' neutral a-D-glucosidase ( a b o u t 2.5%) was a d s o r b e d on the gel. Indeed, after p a p a i n treatment, only a b o u t 5% of the ' T i ' form was able to interact with the p h e n y l - S e p h a r o s e CL-4B gel.

TABLE III HYDROPHOBIC CHROMATOGRAPHY OF THREE FORMS OF NEUTRAL a-D-GLUCOSIDASE ON COLUMNS OF PHENYL-SEPHAROSE CL-4B About 20 mU of each form were applied to the column (4x 1.5 cm). 'p' and 'Ti' forms were purified after proteolytic treatment of 'd' neutral a-D-glucosidase. The samples were applied in a buffer containing 10 mM sodium phosphate, 150 mM NaC1 and 0.l% (w/v) Triton X-100, (pH 7.4). After wash, the columns were eluted by a medium containing 1 mM sodium phosphate and 2.5% (w/v) Triton X-100, (pH 7.4). Enzymatic forms

'd' 'p' ' Ti' 'Ti' + papain

neutral a-D-glucosidase activity Unbound (%)

Adsorbed and eluted (%)

Total recovery (%)

13.3 98 27.6 94.7

86.3 2.3 72.2 5.2

99.6 100.3 99.8 99.9

64 ]'ABLE IV KINETIC PARAMETERS OF THREE FORMS OF N E U T R A L a-o-GLUCOSIDASE Neutral a-D-glucosidase

Kinetic constants of substrate hydrolysis

Nature of maltose hydrolysis inhibition

forms

maltose

Tris

phlorizine

Km

V

Km *

V

"d' 'Ti' 'p'

0.46 0.45 0.48

58.4 66 68.1

0.4 0.4 0.43

19.2 22.5 26.3

M M M

NC NC NC

starch

V is expressed as/~mol linkage hydrolyzed/min per mg protein, K m * as mg/mt. K m as mM. NC is non-competitive inhibition: M is mixed inhibition.

Solubilization of neutral a-o-glucosidase brush-border membranes by proteinases

from

Approx. 95% of the neutral C~-D-glucosidase activity was solubilized after 10 min of papain treatment. All enzyme activity was found in the supernatant obtained after incubation for 60 min with this proteinase. When brush-border membrane vesicles were incubated with trypsin at 10% (w/w), only 5% of the neutral u-D-glucosidase activity was solubilized after 10 rain and about 35% of it was recovered in the supernatant obtained after 60 min trypsin incubation. Interestingly, the neutral a-D-glucosidase activity released after brush-border membrane vesicle incubation for 60 rain with each proteinase was never adsorbed on hydrophobic chromatography.

Kinetic studies As shown in Table IV, study of the catalytic activities of both forms of neutral a-D-glucosidase did not reveal any significant variation in the kinetic constants of substrate hydrolysis. Discussion

The 'detergent' forms of neutral a-D-glucosidase have been purified from pig and rat intestines [3,13,14]. However, this 'detergent' enzyme had never been purified before from kidney. The purpose of this work was purification of horse kidney neutral Ct-D-glucosidase by affinity chromatography and characterization of its amphipathic nature. For purification of the different enzymatic

forms, p-aminophenyl-a-D-glucopyranoside was preferred to p-aminophenyl-fl-D-maltoside as the ligand for affinity chromatography [11]. This choice is the result of several observations: this new ligand is obtained by catalytic reduction of a pure commercialized product ( p-nitrophenyl-a-D-glucopyranoside) whereas the other ligand requires a long and difficult organic synthesis. Furthermore, nonspecific adsorption is reduced to only 10-15% of the adsorbed proteins, and use of this new ligand increases the number of times the affinity chromatography gel can be used, Moreover, this technique, including an affinity chromatography step, seems to be particularly effective for the purification of all three forms of neutral C~-D-glucosidase, as reflected by the short time required for purification (48 h from brush border vesicles) and the yield (higher than 50% from the crude kidney homogenate). The enzyme was always desorbed from the affinity chromatography by p-nitrophenyl-c~-D-glucopyranoside, a substrate of the enzyme [11]. The neutral t~-D-glucosidase bound to the affinity gel was never desorbed by glucose and maltose and no significant hydrolysis of the ligand or maltose (when tested as a desorbent) was observed. These results imply that the ligand-enzyme interactions are localized in an area that overlaps both a portion of the active site of maltose hydrolysis and a nearby but distinct area. The interactions between this last area and the ligand could be hydrophobic: this would explain why, contrary to p-nitrophenyl~x-D-glucopyranoside, maltose does not desorb the enzyme and is not hydrolyzed. Such enzyme-ligand

65 interactions outside of the active site of substrate hydrolysis have been described for other glycosidases [31,32]. The different behavior observed for the 'd' and 'p' forms of neutral ~-D-glucosidase during hydrophobic chromatography and charge shift crossed immunoelectrophoresis could be explained by the presence of a hydrophobic area on the 'd' enzyme. This hydrophobic area would be responsible for the interactions with phenyl-Sepharose CL-4B gel and for the enzyme mobility modifications obtained by substituting ionic detergents for the Triton X-100 during crossed immunoelectrophoresis [1,23]. By contrast, the 'p' form, which lacks this hydrophobic area, does not exhibit binding on hydrophobic chromatography or electrophoretic mobility modification when the charge of detergent is added to the agarose gel and buffer. The detergent form of the neutral a-D-glucosidase of horse kidney would thus be similar to the other stalked 'detergent'-solubilized microvillar enzymes: they are amphipathic proteins which, by proteolytic treatment, would be converted into hydrophilic forms with elimination of hydrophobic area [1]. While these results agree with those described for maltase-glucoamylase of pig intestine brush borders [3,13], they differ from those obtained with the purified enzyme from rat intestine [14], which does not display any detergent property; recently, Lee et al. [15] were able to obtain the amphiphatic form of this enzyme when purification was performed in the presence of proteinase inhibitors. When the proteolytic 'p' and 'Ti' forms of neutral a-D-glucosidase are subjected to chargeshift crossed immunoelectrophoresis, only one precipitate line (1.42) is observed with the 'p' form whatever the detergent used. However, when the 'Ti' form is run into cetyltrimethylammonium bromide/Triton X-100 or Triton X-100 alone, two peaks are obtained. By contrast, when the 'Ti' form was run into deoxycholate/Tri.ton X-100, only one precipitate line was obtained, and had the same relative mobility (1,44) observed for 'p' form. Several points can be deduced from these results: the fact that the 'p' and 'd' forms of neutral a-D-glucosidase, run into sodium deoxycholate/Triton X-100, present the same relative mobility (1.44) explains the single peak oh-

tained when the 'Ti' form of the enzyme is run into these detergents. This peak is actually due to superposition of the two enzymatic forms (hydrophilic and amphipathic forms) obtained after trypsin treatment. The presence of two precipitate lines when the 'Ti' form of the enzyme is run into cetyltrimethylammonium-bromide/Triton X-100 or Triton X-100 alone could be explained by the presence on the enzyme, as already described for other hydrolases (for review, see Refs. 1, 33), of proteolytic attack sites near the hydrophilic area, most probably a the 'junctional segment' described by Hussain et al. [34]. The trypsin attack site on neutral a-D-glucosidase would thus have obstructed access to this proteinase when the enzyme is integrated in the native membrane or in Triton X-100 micelles. Conversely, the papain activity site would display free access to the proteinase. The modification of the molecular properties (hydrophobic chromatography and charge-shift immunoelectrophoresis) of the 'Ti' neutral c~-Dglucosidase form, after papain treatment and the different kinetics of solubilization, of this hydrolase from horse kidney brush-border membranes by the two proteinases, seem to support this hypothesis. Fulcher and Kenny [35] have recently reported that the hydrophobic domain of the purified detergent form of pig kidney microvillar endopeptidase could not be removed by trypsin treatment alone, and these authors suggested that the length of the junctional peptide must be greater than 3 nm to allow access by proteinases. Interestingly, the length of the junctional peptide of pig kidney microvillar endopeptidase, on which trypsin is uneffective, is 2 nm [35,36] whereas for pig intestinal microvillus aminopeptidase, separated from the membrane leaflet by a 5 nm gap [34], proteinase action succeeded. No information is yet available on the junctional part of neutral c~-Dglucosidase. No variation was observed in the kinetic parameters of substrate hydrolysis between the 'proteolytic' and 'detergent' forms of neutral c~-Dglucosidase; these results imply that the hydrophobic area of the 'd' enzyme is located in a protein area independent of the active site of substrate hydrolysis, and suggest that proteolytic treatment induces no or very few modifications in the hydrophilic part of the enzyme. These results

66

support the notion that the hydrophobic area is located near one end of the polypeptide chain, as previously described for other microvillar enzymes [I]. This assumption is confirmed by the immunological identity observed between these two enzymatic forms by crossed immunoelectrophoresis. Further experiments on the ability of the 'Ti' and 'd' forms of neutral c~-D-glucosidase to be incorporated in artificial phospholipidic vesicles, the characterization of purified hydrophobic peptides and the determination of C- and N-extremities of these forms of the enzyme are under way to determine the exact nature of the intermediate section, between the hydrophilic and hydrophobic domains, on which the proteolytic sites are located.

Acknowledgements We gratefully acknowledge the expert technical assistance of Miss Nicole Eimare and Miss Margareth Chischpoertich for the revision of this manuscript. This work was supported by grants from the 'Federation Nationale des Centres de Lutte Contre le Cancer' and the I.N.S.E.R.M. (contrat de recherche externe No 84-1011).

References 1 Kenny, A.J. and Maroux, S. (1982) Physiol. Rev. 62, 91 128 2 Fulcher, I.S. and Kenny, A.J. (1983) Biochem. J. 211, 743 753 3 Maroux, S. and Louvard. D. (1976) Biochim. Biophys. Acta 419, 189-195 4 Brunner, J., Hauser, H., Braun, H., Wilson, K.J.. Waker, H., O'Neill. B. and Semenza, G. (1979) J. Biol. Chem. 254, 1821 1828 5 SchlegeI-Haueter, S., Hore, P., Kerry, K.R. and Semenza, G. (1972) Biochim. Biophys. Acta 258, 506 519 6 Sivakami, S. and Radhakrishnan, A.N. (1973) Ind. J. Biochem. Biophys. 10. 283-284 7 Flanagan, P.R. and Forstner, G.G. (1978) Biochem. J. 173, 553 563 8 Tsuboi, K.K., Kwong, L.K., Burill, P.H. and Sunshine, P. (1979) J. Membrane Biol. 50, 101 122 9 KEN\,. J. and Alpers, D.M. (1973) Biochim. Biophys. Acta 315. 113-122

10 De Burlet, G. and Sudaka, P. (1976) Biochimie 58, 621 623 11 Giudicelli, J., Emiliozzi, R., Vannier, C., De BurleL G. and Sudaka, P. (1980) Biochim, Biophys. Acta 612, 85 96 12 Reiss, U. and Sacktor, B. (1981) Arch. Biochem. Biophys. 209. 342 348 13 Sorensen, S.H., Nor6n, O.. Sj6strOm, M. and Danielsen, E.M. (1982) Eur. J. Biochem. 126, 559-568 14 Lee, L.M.Y., Salvatore, A.K., Flanagan, R.P. and Forstner, G.G. (1980) Biochem. J. 187, 437-446 15 Lee, L and Forstner, G. (1984) Can. J. Biochem. Cell. Biol. 62, 36-43 16 De Burlet, G, and Sudaka, P. (1977) Biochimie 59, 1-6 17 Poiree, J.C., Vannier, C., Sudaka, P. and Fehlmann. M. (1978) Biochimie 60, 645-651 18 Lowry, O.H., Rosebrough, N.J., Farr, A . L and Randall, R.J. (1951) J. Biol. Chem. 193, 265 275 19 Gerristein, W.J., Verkley, A.J.Z., Zwaal, R.F.A. and Van Deenen, LL.M. (1978) Eur. J. Biochem. 85. 255 261 20 Louvard, D., Maroux, S., Vannier, C. and Desnuelle, P. (1975) Biochim. Biophys. Acta 375, 236-248 21 Wroblenski, H. (1975) Biochimie 57, 1095-1098 22 Laurell, C.B. (1965) Anal. Biochem 10, 358 361 23 Bhakdi, S., Bhakdi-kehnen, B. and Bjerrum, O.J. (1977) Biochim. Biophys. Acta 470, 35 44 24 Vannier, C., Boudouard, M., Giudicelli, J., Starita-Geribaldi, M. and Sudaka, P. (1981) Biochimie 63, 375 387 25 Maroux, S., Louvard, D. and Baratti, J. (1973) Biochim, Biophys. Acta 321, 282 295 26 Weber, K., Pringle, J.R. and Osborn, M. (1972) Methods Enzymol. 26, 3-27 27 Axelsen, N.H., Kroll, J. and Weeke, B. (1973) in A Manual of Quantitative Immunoelectrophoresis: Methods and Applications, (Axelsem N.H., Kroll. J. and Weeke, B. eds.), U niversitetsforlaget, Oslo. 28 Owem P. and Smyth, C.J. (1977) in Immunochemistry of Enzymes and their Antibodies (Salton, M.R.J., ed.), pp. 147 202, John Wiley & Sons, New York 29 Sj6str6m, M., Nordn, O., Jeppesen, L., Staun, M., Svensson, B. and Christiansen, L. (1978) Eur. J. Biochem. 88, 503 511 30 Svensson, B.. Danielsen, M., Staun, M., Jeppsesem L., Nordn, O. and Sj6str/Sm, M. (1978) Eur. J. Biochem. 90, 489-498 31 Megm T. and Matsushima, Y. (1977) J. Biochem. 81, 571 578 32 Steers, E., Jr., Cuatrecasas, P. and Pollar, H.B. (1971) J. Biol. Chem. 246, 196 200 33 Hauser, H. and Semenza, G. (1983) CRC Crit. Rev. Biochem. 14(4), 319 345 34 Hussain, M.M., Tranum-Jansen, J., Nordn, O., Sj6strom, M. and Christiansen, K. (1981) Biochem. J. 199, 179 186 35 Kenny, A.J., Fulcher, I.S., McGill, K.A. and Kershaw, D. (1983) Biochem. J. 211,755-762 36 Kenny, A.J. and Fulcher, I.S. (1983) in Ciba Foundation Symposium: Vol. 95, pp. 12-33, Pitman, London.