B I O C H I M I E , 1984, 66, 291-304

Purification, characterization and kinetics of 13-N-acetylhexosaminidases A and B from the slug Arion rufus L.

Enrique VILLAR, Jos6 A. CABEZAS : and Pedro CALVO.

Department o f Biochemistry, Faculty o f Biology, University o f Salamanca and C.S.L C., Plaza de la Merced, 1, Salamanca, Spain.

(Re fu le 10-11-1983, acceptd le 8-2-1984).

R6sum6 - - Deux ~-N-acdtyihexosaminidases ont dtd purifides jusqu~ homogenditd et caractd- risdes, ~ partir de la glande digestive de la limace, A. rufus L.; les deux enzymes ont montrd une trds grande activitd spdcifique. L'hexosaminidase A (Hex A) a dtd purifide 1 300 lois avec un rendement du 12 %; l'hexosaminidase B (Hex B) a dtd purifide 1 400 fois avec un rendement du 20 %. Les hexosaminidases A et B purifides montrent des mobilitds diectrophordtiques diffdrentes, et donnent une seuie bande de protdine en diectrophor~se sur gel de polyacrvlamide. Les prdparations purifides ne possddent pas les autres activites glycosidasiques prdsentes dans l'extrait brut.

Les activitds [3- N-acdtylglucosaminidasique (Glc NA c-ase) et [~- N-acdtylgalactosaminidasique (GalNAc-ase) sont toujours assocides, et comme Ieur rapport d'activitd reste constant au cours de toutes les dtapes de purification, elles seraient catalysdes par la m6me protdine enzvmatique (Hex A ou Hex B).

Le pH optimum pour la GlcNAc-ase est de 4,5 et pour la GalNAc-ase de 4,0. L 'Her B montre 5 r u n u e alUUUlit" a iu c n u l e u r et u u u e t n e ~ t ,,t. p o u t l e l~,lua c n u n g e t n e n t n q u e L e s i n t s

isodlectriques sont de 4,5 et 5,5 pour Ies formes A et B, respectivement. La masse moldculaire est 150 000 pour l'Hex A et 320 000 pour la Hex B. La composition en acides aminds des Hex A et B montre quelques diffdrences concernant Cys, 1~r, Ser, Glu et lie, particulidrement.

Les vaieurs du rapport V, .... /K,, montrent que I'activitd GlcNAc-ase est la principaie des deux formes.

Les [3-N-acdtylglucosides et ies ~-N-acdtylgalactosides montrent un effet de compdtition complete pour &~ite actif commun dans des cxpdriences avec un mdlange de substrats. Les vaieurs de K, sont toujours concordantes pour les activitds GlcNAc-ase et GalNAc-ase avec des inhibiteurs

To whom all correspondence should be addressed.

Ahbrdviations : pNPh-glycosides : p-nitrophenyl glycosides; pNPh-GIcNAc : p-nitrophenyl 2-acetamido-2-deoxy-l~-o-gluco- pyranoside; pNPh-GalNAc : p-nitrophenyl 2-acetamido-2-deoxy-13-o-galac- topyranoside; GlcNAclac : 2-acetomido-2-deoxy-o-gluconolactone; GalNAclac : 2-acetamido-2-deoxy-o-galactonolactone: GIcNAc-ase : I~-N-acetylglucosaminidase activity;

GalNAc-ase : lt-N-acetylgalactosaminidase activity: H e x : hexosaminidase. I:l-N-acetylhexosaminidase: Con A-Sepharose : concanavalin A-Sepharose 4B: ANAG-Sepharose : 2-acetamido-N-(c-aminocaproyl)-2-deox.~ 13- o-glucosylamine-Sepharose 4B: K,, : Michaelis constant: Vmj, : maximal velocity: K, : inhibitor constant: Pb (AcO)., : lead acetate.

E n z j m e : [t-N-acetyihexosaminidase 1EC3.2.1.52): 2-ace tamido-2-deoxy-lt-hexoside acetamidodeoxyhexohydrolase.

292 E. Villar, J.A. Cabezas and P. Calvo

compdtitifs (les lactones respectives). Les rdsuitats suggdrent fortement que les deux activitds sont catalysdes par ie m~me site actif pour I'Hex A comme pour l'Hex B.

Une inhibition des activitds enzymatiques a dtd trouvde avec les lactones respectives, les N-acdthylhexosamines, le mannose, les mannosides, le CI2Hg et l'acdtate de plomb; et une activation avec le ribose, et avec ~iuelques chlorures et sulfates de cations divalents.

Mots-cl6s : [~-N-ac~tyihexosaminidase / formes enzymatiques / [~-N-ac~tylglucosaminidase / [i-N-ac~tyl- galactosaminidase / hexosaminidase.

S u m m a r y -- Two {l-N-acetylhexosaminidases have been purified to homogeneity and characteri- zed, from the digestive gland of the slug A. rufus L., showing very high specific activities. Hexosaminidase A (Hex A) was purified 1300-fold with a yield o f 12 %, and hexosaminidase B (Hex B) was purified 1400-fold with a yield o f 20 %. Purified Hex A or Hex B ~m as a single protein band in polyacrylamide gel disc electrophoresis, showing dbrferent mobilities. The purified preparations do not show any of the other glycosidase activities present in the crude extract.

{l-N-acetylglucosaminidase (GicNAc-ase) and ~-N-acetylgalactosaminidase (GalNAc-ase) activities are always associated in a single peak for each enzyme form, with constant activity ratio, in all the purification steps, since they are catalyzed by the same enzyme (Hex A or Hex B).

The optimal pH for both forms are 4.5 for GlcNAc-ase and 4.0 for GalNAc-ase activity. Hex B shows thermal and pH-stability higher than Hex A. The isoelectric points are 4.5 and 5.5 for A and B forms, respectively. The molecular weight is 150 O00 for Hex A and 320 O00 for Hex B. The amino acid composition of purified Hex A and B presents some differences concerning particularly Cys, Thr, Ser, Glu and lie.

The ratios Vm~/Km show that GlcNAc-ase is the main activity o f both enzyme forms.

fl-N-acetylglucosides and {l-N-acetylgalactosides completely compete for a common active site in mixed-substrates experiments. The K~ values are always coincident for GlcNAc-ase and GalNAc-ase activities, using competitive inhibitors (the corresponding iactones). These results strongly suggest that both activites are catalyzed by the same active site in both Hex A and B.

Inhibition of the enzyme activities was found with the corresponding lactones, N-acetyl hexosamines, mannose, mannosides, HgCI 2 and lead acetate; activation, with ribose, and with some chlorides and sulphates of divalent cations.

Key-words : [3-N-acetylhexosaminidase / enzyme forms / [~-N-acetylglucosaminidase / [~-N-acetylgalactosa- minidase / hexosaminidase.

Introduct ion

[t-N-acetylhexosaminidase is a widely distribu- ted glycosidase having both [$-N-acetylglucosami- nidase and [i-N-acetylgalactosaminidase activities. This enzyme has been purified from different materials such as mammalian tissues [1,2], plants [3, 4] and molluscs [5-7].

Robinson and Stifling [8] were the first to separate the [I-N-acetylhexosaminidase (from human spleen) into two components, A and B. The interest in the molecular forms of this enzyme was increased by the findings of Okada and O'Brien [9], who reported a specific deficiency of the A form in patients with Tay-Sachs disease, and by those of Sandhoff [10], who described a deficiency of both A and B components in a

related disease. O'Brien et al. [11] and Dance et al. [12] reported twc, different methods to deter- mine the proportions in which both hexosamini- dases are present in humans, an important fact for the diagnosis of Tay-Sachs disease heterozygote carriers.

Those major forms of the enzyme have been studied in several mammalian tissues and sera: Human ~,lacenta [13, 14], human liver [15], beef spleen [16], bovine brain [17], horse brain [18], porcine placenta and bovine epidydimis [19], rat liver [20], colon tumours [21] and sera [22]. Recently, Ahladeff and Holzinger [23] found an atypical he tosaminidase B in human liver tu- mours, and Kind [24] studied the chromatogra- phic pattern of both hexosaminidase forms in urine from patients with renal diseases or renal

Two molecular forms o f ~-N-acetylhexosaminidase from a mollusc 293

transplantation. Changes in hexosaminidase le- vels in different diseases have been previously reported [25-28].

The digestive gland of molluscs is a very rich source of hexosaminidases and other glycosidases [29, 30]; however, the molecular forms of this enzyme have been scarcely studied in that mate- rial, specially from terrestrial molluscs. Mura- matsu [31] reported only one hexosaminidase from Turbo cornutus liver, but later, Yeung et al. [7] demonstrated the presence of two forms in this mollusc.

We describe here the purification, characteriza- tion and kinetics of two enzyme forms of [I-N- acetylhexosaminidase from the digestive gland of the slug Arion rufus L., a material not previously studied, showing very high specific activities. Since both hexosaminidases are involved in the catabolism of N-acetylglucosamine and N-acetyl- galactosamine-containing glycoconjugates, the study of their properties and kinetics may be useful for a better understanding of the physio- logical role played by different forms of the same enzyme. The catalysis of the activities of this enzyme in one or in two active sites is a contro- versial point [3, 6, 21, 32-35] that we study here.

In addition, highly purified and kinetically well-characterized [I-N-acetylhexosaminidases, such as those reported in this work, are useful as analytical tools for structural studies.

Materials and Methods

Materials

The slugs Arion rufus L. were collected in June, and stored at -- 20 °C without significant loss of hexosami- nidase activity.

pNPh-glycosides, pNPh-GIcNAc, pNPh-GaINAc, glucose, mannose, galactose, D-fucose, L-fucose, L- rhamnose, ribose, xylose, maltose, sucrose, raffinose, glucosamine (HCI), N-acetyiglucosamine, N-acetylga- lactosamine, ct-methylglucoside, a-methylmannoside, bovine serum albumin, standard proteins for molecular weight determination, acrylamide, Tris, sodium azide, DEAE-cellulose (medium mesh), and CM-cellulose (medium mesh) were obtained from Sigma Chemical Co., St. Louis, Mo, U.S.A. GIcNAclactone and gal- NAclactone were from Koch-Light Laboratories, Coin- brook, Bucks., U.K. Sephacryl S-200, Sephadex G-200, Con A-Sepharose 4B and Blue Dextran were from Pharmacia Fine Chemicals, Uppsala, Sweden. Ampholines were obtained from Serva, Heidelberg, F.R.G. Fluorescamine was from Roche, Nutley, N J, U.S.A. General laboratory chemicals were obtained

from Probus, Spain. All products were of analytical grade.

Enzyme assays

Glycosidase activities were determined in 0.1 M sodium citrate buffer pH4.5 (optimal pH for ~-N- acetylglucosaminidase activity) or pH 4.0 (optimal pH for [~-N-acetylgalactosaminidase activity), and 2 mM pNPh-GIcNAc (or other p-NPh-glycoside) or l mM pNPh-GalNAc. After incubation at 37 oC for 10 min, i ml of ! M Na2CO3 was added to the reaction mixture (! ml). The released p-nitrophenol was measured at 400 nm. All the buffer solutions contained 0.02 %NAN3 as preservative. One unit of enzyme activity (U) was defined as the amount of enzyme which hydrolyzes 1 ~mol of pNPh-glycoside per minute under the assay conditions. Specific activities were expressed as U/mg protein.

Protein determination

Proteins were determined by the methods of Lowry et al. [36] or B/ihlem et al. [37] (when assaying low protein concentration samples), using bovine serum albumin as standard.

Purification o f [i-N.acetyihexosaminidases

All procedures were carried out at 4oC unless otherwise indicated.

Step 1 : preparation of the crude extract. Digestive glands were homogenized with 50 mM sodium citrate buffer pH 6.0 (l/10, w/v) in a mortar with sand. The suspension was left to settle for 15 min and the homogenate was centrifuged at 12 000 x g for 20 min; the precipitate was discarded.

Step 2 : ammonium sulphate fractionation. Solid ammonium sulphate was added with stirring to the supernatant from the previous step, to give 30% saturation. After 20 rain, the resulting precipitate was removed by centrifugation as above, and discarded. The supernatant was adjusted to 50% saturation and after 15 rain the suspension was centrifuged at 12 000 x g for 20 min; the supernatant was discarded.

Step 3 : Gel filtration on sephacryl S-200. The precipi- tate from the last step was resuspended in 50raM sodium citrate buffer pH 6.0, and applied to a column (2.5 x 85 cm) of Sephacryl S-200 equilibrated with the same buffer; the column was eluted with that buffer at a flow rate of 13 mi/h, controlled by peristaltic pump. Fractions of 5 ml were collected and those containing hexosaminidase activity were pooled. The pooled f"actions were then dialyzed against 301 of 10raM sodium citrate buffer pH 6.0 for 24h with several changes of the buffer solution.

Step 4 : DEAE-ceilulose chromatography. The dialy- zed solution from step 3 was applied to a DEAE- cellulose column (1.2 x 37cm) equilibrated with 10 mM sodium citrate buffer pH 6.0. 1he column was

294 E. l/'iilar, J.A. Cabezas and P. Calvo

washed with this buffer to elute unadsorbed proteins. Hexosaminidase B is recovered at this stage. A NaCI linear gradient (from 0 to 0.8 M) in the same buffer was then applied. Hexosaminidase A is eluted at NaCI concentration between 0.25 and 0.50 M.

From this step both enzyme forms were separately purified.

Step 5 A (Hex A) : affinity chromatography on Con A-Sepharose 4B. The pooled fractions with hexosami- nidase activity from the NaC! gradien ~, elution of the last step were applied, at -'~om temperatrre (approx. 23 oC), to a column (1.2 x t3 cm) of Con ,~ ~ 4harose equilibrated in 10raM sodium citrate bvffer pH 6.0, containing 1 mM MnCI2, MgCI., and CaCI2. The co- lumn was washed with the same buffer until the absorbance at 280 nm of the effluent become zero. Hexosaminidase A was then eluted with 1 M tt-methyl- glucoside and 1 M NaCI in the citrate buffer, essen- tially as described by Akiki et al. [38], but using a very slow elution rate (1 ml/h). The fractions containing hexosaminidase activity were pooled and dialyzed against 101 of 10 mM sodium phosphate buffer pH 7.5, with several changes. The enzyme activity recovered was about 75 % at this step. A very slow elution at not too low temperature (23-37 oC) is essential to get such a high recovery under our experimental conditions.

Step 6 A (h'ex A)" affinity chromatography on ANAG-Sepharose. The dialyzed preparation from the last step was applied to a column (0.8 x 7.5 cm) of Sepharose 4B-2-acetamido- N-(e-aminocaproyl)-2- deoxy-[i-D-glucosylamine (ANAG-Sepharose). The gel was kindly provided by Dr. J.L. Stirling (Dept. Bio- chemistry, Queen Elizabeth College, London). The column was washed with the above mentioned dialysis phosphate buffer until the absorbance at 280 nm was zero. Hexosaminidase A was then eluted with 0.2 M NaCI and 60 mM N-acetyiglucosamine in the same phosphate buffer. The pooled fractions showing hexo- saminidase activity were dialyzed against 101 of 10 mM sodium phosphate buffer pH 7.5, with several changes. This dialyzed solution was used for characterization and kinetic studies.

Step 5 B (Hex 11) : CM-cellulose chromatography. The pooled fractions of hexosaminidase B from the DEAE-cellulose chromatography (step 4) were dialyzed against 241 of 10 mM sodium citrate buffer pH 4.5, with several changes. The dialyzed enzyme solution was applied to a CM-cellulose column (1.5 x 45 cm) equilibrated with this citrate buffer. The column was washed with the same buffer. Hexosamlnidase B is eluted with a NaCI linear gradient (from 0 to 0.5 M) in the citrate buffer, at a NaCI concentration between 0.2 and 0.4 M.

Step 6 B (Hex B) : affinity chromatography on Con A-Sepharose 4 B. The enzyme solution from the last step was dialyzed against 24 ! of 10 mM sodium citrate buffer pH 6.0, with several changes, and then applied to a Con A-Sepharose column, as indicated for step 5 A. Hexosaminidase B was eluted as described above

for hexosaminidase A. The fractions with hexosamini- dase activity were pooled and dialyzed against 101 of 10mM sodium citrate buffer pH 6.0. The dialyzed solution was used for characterization and kinetic studies.

Differential assay for hexosaminidases A and B

The assays to determine the proportion of each enzyme form in the crude extract were carried t,z~t as described by Dance et al. [12] using DEAE-celluiose, and by O'Brien et al. [11] following a thermal inacti- vation procedure.

Polyacrylamide gel electrophoresis

Electrophoresis of the purified samples was perfor- med, according to Davis [39] in Tris/glycine buffer pH 8.3, with 5 % acrylamide gels and 3 mA per column. Gels were stained for protein by silver staining, as described by Morrissey [40].

Molecular weight determination

The molecular weight of hexosaminidases A and B was determined by Sephadex G-200 column chromato- graphy. The column (2.5 × 100cm) was equilibrated and eluted with 50 mM sodium citrate buffer pH 6.0, at a flow rate of 5 ml/h. Standard proteins used were : Ferritin (440,000), catalase (240,000), y-globulin (160,000), bovine serum albumin (67,000), ovalbumin (45,000), and myoglobin (17,000).

Isoelectric focusing

Isoelectric focusing was carried out according to . . . . . . t,v~$ [41] in a i f0 ml LKB eiectrofocusing co- lumn. The ampholyte concentration was 1% (v/v), with a pH range from 3.5 to 10.0 in a sucrose gradient (0-50 %, w/v). The run was performed at 400 V, 4 oC for 48 h. Fractions of 2 ml were collected, and the pH of each fraction was measured at 2 oC with a digital pH meter (Radiometer pH M-62).

Amino acid analysis

The amino acid composition of both purified hexosaminidases A and B was determined with the samples from steps 6 A and 6 B, respectively. Hydroly- ses were carried out at 110 oC for 24, 48 and 72 h with 200 l.ti of 5.7 M HC! containing 0.05 % 2-mercaptoetha- nol, in evacuated and sealed tubes [42]. Cys was determined as cysteic acid after performic acid oxida- tion [43]. The analyses were performed in a Beckman 121-MB analyzer equipped with a Beckman integrator 126 data system.

Determination o f kinetic parameters

The corresponding /rNPh-glycosides were used as substrates. The reaction mixtures were incubated at

Two molecular forms of ~-N-acetylhexosaminidase from a mollusc 295

37 oC for l0 min in all the kinetic assays. Michaelis constants (Km) and maximal velocities (VmaJ were determined from the Lineweaver-Burk plots [44]. Inhi- bitor constants (K0 were determined by the Dixon plots [45]. GlcNAc lactone and GaINAc lactone were used as competitive inhibitors.

Mixed-substrates analysis

Mutual competition studies between substrates were carried out as described by Dixon and Webb [46], using equimolecular mixtures of the corresponding pNPh- glycosides as substrates (pNPh-GIcNAc/p-NPh-Gal- NAc).

The theoretical maximal velocities for competing substrates. V,+~, were calculated from the equation: Kmx,/~tK~,, = (Vx-V~+y)/(V~+y- Vy), where x is the substrate showing a higher velocity; y, the substrate showing a lower velocity; V, maximal velocity, and ct, the concentration ratio of both substrates (oc = 1 in our experiments).

Results

Separation of two enzyme forms offI-N-acetyl- hexosaminidase

Two forms, A and B, were separated on DEAE-cellulose column chromatography (Fig. l), at the 4 th step of the purification procedure. Hexosaminidase B is recovered by washing the column with buffer, whereas hexosaminidase A was bound to the ion-exchanger. Form A was obtained with a 0-0.8 M NaCl linear gradient.

Purification of hexosaminidases A and B, and criteria for homogeneity

Hexosaminidase A was purified about l 300-fold with an overall yield of 12 % from the crude extract (Table I). The specific activities were 23 U/mg for 13-N-acetylglucosaminidase agtivity and 9 U/mg for 13-N-acety:galactosamini- dase activity. The specific activity ratio I~-N-ace- tylglucosaminidase/13- N-acetylgalactosaminidase was 2.5 + 0.1 throughout the purification proce- dure.

Hexosaminidase B was purified about l 400-fold, with an overall yieid of 20 % from the crude extract (Table II). The specific activities were 32 U/mg for 13-N-acetylglucosaminidase activity and 17 U/mg for 13-N-acetylgalactosami- nidase activity. The ratio of these activities was 1.7 + 0.1 throughout the purification procedure.

The purity of the purified enzyme samples was confirmed by polyacryiamide gel disc electropho-

500

--- 400 E

E

300 ._

200

LaJ

100

10 20 30 43

0a

] 0k_-

0 10 20 30

A280 05

0.3

Fraclion number

FIG. 1. -- Elution profile o f he.x'osaminidases A and B on DEAE-eelhdose cohmm chromatography.

Experimental condition are given in Methods. Fractions of 5 ml were collected. (A,zx) Hex A, GIcNAc-ase and GalNAc-ase, respectively: ( e . o ) Hex B, GlcNAc-ase and GalNAc-ase, respectively: ( ) protein monitored by the absorbance at 280 nm; ( - - - - ) N a C ! concentration.

resis. Purified hexosaminidase A or B run as a single protein band (Fig. 2), showing different electrophoretic mobilities. The purified enzyme preparations of hexosaminidases A and B did not show any of the other glycosidase activities originally present in the crude extract.

T h a n r a n a r l t i a n ~ f a n n d f a r h a t h farm~: in t h ~

crude extract were 37 % and 63 % for A and B, respectively (Table Ill), using DEAE-cellulose in tube or in column [12]. Results from thermal inactivation [11] were not representative (see below, Discussion). All activities, purification factors and yield values were calculated on the basis of the above listed percentages, according to Sandhoff et al. [15].

Optimal pH and pH-stability

The optimal pH for both forms is 4.5 with pNPh-GIcNAc as substrate, and 4.0 with pNPh-GalNAc, in 0.1 M sodium citrate buffer.

The effect of pH on the stability of the enzyme forms was examined by preincubating them in 0.2 M citric acid/Tris buffer at various pH values. Results obtained are shown in figure 3. Each enzyme form shows the same pH-stability profile for both ~-Noacetylglucosaminidase and B-N-ace- tylgalactosaminidase activities. Nevertheless, hexosaminidases A and B have a very different

TAB

LE I

Pur

ifica

tion

of h

exos

amin

idas

e ,4

from

dig

estiv

e gla

nd o

f Ari

on r

ufus

L

E

xper

imen

tal

deta

ils

and

defi

niti

ons

of u

nits

are

ind

icat

ed i

n M

ater

ials

and

Met

hods

t,

,J

Ste

p

Tot

al

Spe

cifi

c ac

tivi

ty

Tot

al a

ctiv

ity

Pur

ific

atio

n Y

ield

prot

ein

Glc

NA

c-as

e G

alN

Ac-

ase

Glc

NA

c-as

e G

alN

Ac-

ase

Glc

NA

c-as

e G

alN

Ac-

ase

Glc

NA

c-as

e G

alN

Ac-

ase

Spe

cifi

c

acti

vity

ra

tio

GIc

NA

c-as

e/

Gal

NA

c-as

e

',D

Ox

i. C

rude

ext

ract

0.

018

2. A

tom

. S

ulph

a.

prec

ipit

atio

n 2

! 5

0. !

67

(30-

50 %

)

3. S

epha

cryl

S-2

00

14.6

!.

53

4. D

EA

E-c

ellu

lose

2.

40

5.14

5. C

on A

-Sep

haro

se

0.44

! 7

.6

6. A

NA

G-S

epha

rose

0.

23

23.3

U/m

g

U

-fol

d %

m

g 2

501

0.00

7 45

.0

17.5

1

1 10

0 10

0

0.07

6 35

.9

16.3

9

1 !

80

93

0.70

2,

2.3

10.2

85

10

0 50

58

i .98

12

.3

4.75

28

6 28

3 27

27

7.35

'7

.74

3.23

98

0 l

050

! 7

i 8

9o10

'.5

.36

2.09

l

295

l 30

0 12

12

2.6

2.4

2.5

E"

TAB

LE I

I

Pur

ifica

tion

of h

exos

amin

idas

e B

from

dig

estiv

e gl

and

of A

rion

ruf

us

L

Exp

erim

enta

l de

tail

s ar

e in

dica

ted

in M

ater

ials

and

M

etho

ds

Ste

p

Tot

al

Spe

cifi

c ac

tivi

ty

Tot

al a

ctiv

ity

Pur

ific

atio

n Y

ield

prot

ein

Glc

NA

c-as

e G

alN

Ac-

ase

Glc

NA

c-as

e G

alN

Ac-

ase

Glc

NA

c-as

e G

aiN

Ac-

ase

Glc

NA

c-as

e G

aiN

Ac-

ase

Spe

cifi

c

acti

vity

ra

tio

Glc

NA

c-as

e/

Gal

NA

c-as

e

mg

!. C

rude

ext

ract

2

501

0.03

0

2. A

mm

. S

ulph

a.

prec

ipit

atio

n 21

5 0.

28

(30-

50 %

)

3. S

epha

cryl

S-2

00

14.6

2.

61

4. D

EA

E-c

ellu

iose

7.

18

3.33

5. C

M-c

ellu

lose

0.

78

23.4

6. C

on

A-S

epha

rose

0.

41

3 ! .

6

U/m

g

U

-fol

d %

0.01

2 75

.0

3 ! .

0 1

I 1 O

0 1 O

0

O. 1

3 60

.2

27.9

9

! 0

80

90

1.20

38

. !

17.5

87

97

51

56

2.02

23

.'9

14.5

11

! 16

3 32

47

12.8

18

.13

10.0

78

0 !

032

24

32

17.1

13

.0

7.01

I

053

! 37

9 17

22

1.6

i.8

1.8

Two molecular forms of fJ-N-acetylhexosaminidase from a mollusc 297

TABLE Ill

Differential assays to determine the proportion of hexosaminidases A and B in the crude extract. Experimental conditions are indicated in the text. The total yield was about 97 % in all thest: procedures.

DEAE-cellulose (in tube)

GlcNAc-ase Proportion

DEAE-cellulose (in column) Thermal inactivation (50°C)

GicNAc-ase Proportion GlcNAc-ase Proportion

U/ml % U/ml % U/ml %

Hex A 0.62 36.9 0.62 37.3 1.65* 97.1 Hex B i.06 63.1 1.04 62.7 0.05 2.9

* This activity was calculated by subtracting the GicNAc-ase activity, founa after partial thermal inactivation, irom the total activity before that treatment.

FIG. 2. -- Polyacrylamide-gel disc electrophoresis of purified h~rosaminidases A and B.

Experimental details are indicated in the text. Gels were stained for protein by silver staining. ~,, hexosaminidase A; B, hexosaminidase B.

FiG. 3. -- pH-Stability of hexosaminidases A and B.

Samples of the enzyme forms were preincubated at 37°C for 2 h in 0.2 M citric acid/Tris buffer at pH values ranging from 3.0 to 8.5. The enzyme activities were subsequently determined as described in the Methods section, a, GlcNAc- ase; b, GalNAc-ase.

pH-stabil i ty pattern. Hexosaminidase A is com- pletely stable only in the narrow pH range between 6.0 and 7.0, whereas hexosaminidase B was found to be stable in a very wide pH range, between 4.5 and 8.5.

Thermal stability

Both enzyme forms showed s imilar profiles for [3-N-acetylglucosaminidase and [t-N-acetylgalac- tosaminidase activities (Fig. 4). However, hexo- saminidase A has different temperature depen-

!00

i i i , i

3 /, 5 6

:. - Hex B

Hex A

pH

100 / Hex H xB A

,, i , i i i i i i

3 ~ 5 6 7 6 9 pH

298 E. Villar, J.A. Cabezas and P. Calvo

100[ a

ex B

M i i ! i , ~0 50 60 70

Temperature( *C )

100

a 50 ==

E

ex B

I I I I

/.0 50 60 70 Temperature (°C)

FtG. 4. -- Thermal stabifity of hexosaminidases A and B.

Samples of the enzyme forms were heated in 50 mM citric acid/Tris buffer pH 7.0 for 5 min at temperatures ranging between 37 and 70°C. Remaining activities were determined as indicated above, a) GlcNAc-ase; b) GalNAc-ase.

dence than hexosaminidase B, showing the typi- cal differences o f hexosaminidases A and B in their thermal ~tahillty n s = t t = = r n c ~ e nr,=t,;,~.~eh, indicated in other materials [22]. At 50 oC, hexo- saminidase B is completely stable for more than 2 h, whereas hexosaminidase A loses 50 % of its enzyme activities after 20 rain at that temperature.

Molecular Weight

The molecular weight was about 150,000 for hexosaminidase A and 320,000 for hexosamini- dase B, as determined by gel filtration on Sepha- dex G-200 (Fig. 5).

Isoelectric point

Both hexosaminidases A and B run as a single peak (Fig. 6) showing both J3-N-acetylglucosami- nidase and J3-N-acetylgalactosaminidase activities. The corresponding isoelectric points are 4.5 and 5.5 for A and B forms, respectively.

Amino acid composition

The amino acid composit ions o f purified hexosaminidases A and B are shown in Table IV.

501

o m

3l; 101

_=

rrJtin

HEX B X',,pxCat alase

"~'- Globulin

~ S e r u m albumin

Myoglobin

Vo , I , i i , I | i i . i , ,

200 300 ~00

Elution volume(ml)

FIG. 5. - - Molecular weight determination of hexosanzinida- ses A and B by Sephadex G-200 column chromatography.

Details are described in Methods.

A280

A | 2.0

E 50 1 pH

--_>" 1.5 12 .>_ o 10

1.0 8 ~" 2s " ' 6

0.5

2

0 0

Fraction number

A280 .B I I 20

30 pH - it.

:~ 1.5 12

20. to

~, to e

m I 0 -

05 /, /

2

n 0 10 20 30 /,0 50 60

Fraction number

FIG. 6. - - lsoelectric focusing of hexosaminidases A and B.

Experimental conditions are indicated in Methods. A, hexosaminidase A; B, hexosaminidase B. (A,o) GIcNAc-ase; (A,o) GaiNAc-ase; (m) protein monitored by the absorbance at 280 nm; ( ) pH.

Two molecular forms of ~-N-acetylhexosaminidase from a mollusc ,299

TABLE IV

Amino acid composition of hexosaminidases A and B.

Figures are average of values from 24 h, 48 h and 72 h hydrolyzates, except for Val (72 h hydrolysis value only). Cys was determined as cysteic acid after perfor- mic acid oxidation. Values for Thr and Ser were obtained by extrapolation to zero hours hydrolysis. Trp was not determined. Experimental details are given in Materials and Methods.

Amino acid Hex A Hex B

mol/100 mol

Lys 4.2 3.5 His 3. l 2.2 Arg 3.8 3.4 Cys 4.3 3.3 Asp 12.5 12.6 Thr 6.2 7.6 Ser 5.9 7.0 Glu 10.1 8.8 Pro 5.0 5.3 Gly 7.3 7.4 Ala 7.4 6.5 Val 6.9 7.5 Met 0.7 1.0 lie 5.2 6.3 Leu 8. l 8.0 Tyr 4.7 4.7 Phe 4.6 4.9

These composit ions present differences concer- ning particularly cysteine, threonine, serine, glu- tamic acid and isoleucine. In contrast, the propor- tions o f aspartic acid, glycine, leucine and tyro- sine are almost coincident in both enzyme forms.

Kinetic studies

Both purified hexosaminidase A and B follow typical Michaelis-Menten kinetics in the rate of hydrolysis of pNPh-GicNAc and pNPh-GalNAc. The apparent Michaelis constants (Kin) and maximal velocities (Vmax) for each substrate and enzyme form are given in Table V. Both 13-N- acetyihexosaminidases A and B show the highest Vmax (27 U / m g and 44 U/mg, respectively) with pNPh-GlcNAc as substrate. The Vmax values determined with pNPh-GaINAc as substrate are about one ,half lower (12 U / m g and 25 U/mg). The Km values are similar for both substrates : 0.33 mM and 0.31 mM for hexosaminidase A, and 0.25 mM and 0.20 mM for hexosaminidase B. The ratios Vmax/Km, that indicate the efficacies of several substrates used by a single enzyme [47], are 8 i.8 ml.mg-~.min-~ and 38.7 for hexosamini- dase A, with pNPh-GlcNAc and pNPh-GalNAc, respectively; and 176 ml .mg-Lmin -t and 125 for hexosaminidase B, with pNPh-GlcNAc and pNPh-GalNAc, respectiveG. These results show that ~-N-acetylglucosaminidase is the main acti-

TABLE V

Kinetic parameters of hexosaminidases A and B Enzyme activities of Hex A and Hex B were assayed under the conditions described in Materials and Methods.

Sustrate Km '~ Vmax (a} V~ + V: 'h' V~+:exp 'c' V'+:compt'd' GlcNAclac GaiNAclac

mM U/mg U/mg U/mg U/mg pM pM

pNPh-GlcNAc (x) 0.33 27 5.0 3.1 Hex A 39 16 19

pNPh-GaINAc (y) 0.31 12 4.8 3.1 m

pNPh-GlcNAc (x) 0.25 44 5.0 2.4 Hex B 69 37 34

pNPh-GalNAc (y) 0.20 25 5.2 2.4

'~ Km and Vm~, values were determined from Lineweaver-Burk plots. The subslrates were used at :oncentrations ranging from 0.04 to 3.4 mM for pNPh-GicNAc, and from 0.04 to 1.7 mM for pNPh-GalNAc.

~h) Vm~,X + Vine, y, when the enzyme was separately acting on the substrates x and y. ~c) Experimental Vma, value obtained from Lineweaver-Burk plots when using equimolecular mixtures of the substrates x

and y (at concentrations ranging from 0.04 to 1.0 mM), (dr Theoretical Vr~, calculated from the equation applicable for competing substrates, as indicated in Materials and Methods. (" K, were determined by Dixon plots, at substrate concentrations from 0.1 to 0.8 mM. GlcNAclac and GaINAclac were

used as competitive inhibitors at concentrations ranging from 0 to 40 pM.

300 E. Villar, J.A. Cabezas and P. Calvo

vity of both enzyme forms, specially for hexosa- minidase A.

Association o f activities in a common active site

[~-N-acetylglucosaminidase and [3-N-acetylga- lactosaminidase activities of hexosaminidases A or B were always associated in a single peak in all the chromatographic steps of the purification procedure and in isoelectric focusing. Further- more, the activity ratio was constant for both enzyme forms throughout the purification, as indicated. Both hexosaminidases A and B puri- fied to homogeneity still show [~-N-acetylgluco- saminidase and [~-N-acetylgalactosaminidase acti- vities since these activities are catalyzed by the same enzyme protein (A or B).

The kinetics of competition with mixed subs- trates were investigated (Fig. 7) to determine whether the enzyme forms have one or two active sites for the catalysis of [~-N-acetylglucosamini- dase and [I-N-acetylgalactosaminidase activities.

When two substrates are transformed in the same active site the experimental maximal velo- city determined with the mixture of both subs- trates must coincide with the theoretical maximal velocity calculated from the equation applicable for competing substrates, and the total rate of the reaction must be less than the sum of the rate of the reactions measured with separated substrates [46]. From the results obtained (Table 5) it may be concluded that [3-N-acetylglucosides and [3-N-ace- t v l e r r z l a e t n ~ i d J = e l ' n m n l P t o l v c o m p s , t o f ' ,~r ~ z ~ a m m ~ n

active site in both hexosaminidases A and B : the Vma, obtained with the mixture of both substrates (16 U/mg and 37 U/mg) are almost coincident with the Vmax corresponding to competition (19 U/mg and 34 U/rag), and they are less than the Vma~ expected if both activities were catalyzed in separate sites (39 U/rag and 69 U/rag). The occurrence of a single active site is confirmed by the inhibition kinetics.

The inhibition type is competitive for both activities and for both A and B forms, with GlcNAc lactone and with GalNAc lactone as inhibitors. The inhibitor constants (K0 were determined by Dixon plots.

The Ki values (Table 5) show that GalNAc lactone is a stronger inhibitor than G!cNAc lactone for both hexosaminidases A and B. When an enzyme catalyzes the hydrolysis of two subs- trates at the same active site, a competitive inhibi- tor must have the same Ki when assayed with either substrate [46]. The K~ values are always coincident for [3-N-acetylglucosaminidase and

o.~

,>

O2 i

I I I I

5 io 20 i

[s]-t( mM )-I

0.2

"7

~ 0.1

S -5

o

Hex B

, ~ I 2i0 5 10 [$]-t(~N)-I

FIG. 7. -- Lineweaver-Burk plots for hexosaminidases A and

The enzyme activities were assayed as indicated in Me- thods, pNPh-GlcNAc ( - , e ) , pNPh-GalNAc (~,o)and mixtu- res of both substrates (o,m) were used. Substrate in mixtures always had the same concentration.

[3-N-acetylgalactosaminidase activities with both lactones, in both A and B forms. From all these results it can be concluded that [I-N-acetylgluco- saminidase and [i-N-acetylgalactosaminidase acti- vities are catalyzed in the same active site in both [3-N-acetylhexosaminidases A and B purified from A. rufus.

Effect o f some compounds on the enzyme activities

As indicated in Table VI, inhibition of A and B forms was found with GalNAc lactone, GlcNAc lactone, N-acetylgalactosamine, N-ace- tylglucosamine, glucosamine (H Cl), a-methyl-

Two molecular forms of ~-N-acetylhexosaminidase from a mollusc 301

TABLE VI

Effect of some carbohydrates The enzyme samples of Hex A and Hex B were incubated in the presence of the indicated compounds at 100 mM concentration. GlcNAc-ase and GalNAc-ase activities were determined as indicated above. Results are ~iven as the average value of 4 experiments (standard deviations < 12 %).

Relative activity (%)

100 mM Hex A Hex B

GlcNAc-ase GalNAc-ase GlcNAc-ase GalNAc-ase

i 00 ! 00 100 100 Ribose 113 126 I ! 8 115 Mannose 42 59 35 43 tx-methylmannoside 34 40 30 34 Glucosamine (HCI) 68 76 70 80 N-acetylglucosamine 2 ! 23 24 30 N-acetylgalactosamine 5 6 ! 0 9 N-acetylgluconolactone 0 0 0 0 N-acer:,lgalactonolactone 0 0 0 0

mannoside, and mannose ; furthermore, activation of A and B forms was unexpectedly found with ribose, for both activities. The rest of the assayed carbohydrates (glucose, galactose, D-fucose, L-fucose, L-rhamnose, xylose, maltose, sucrose, raffinose, and tx-methylglucoside) showed no effect on the enzyme activities.

Both A and B forms were activated (Table VII)

by MnCl2, CaCI2, BaCI2 and MgSO4. Activation or inhibition of A and B forms, respectively, was found with CuSO4 and NiSO4. Both A and B forms were inhibited by HgCI2 and Pb(AcO),. The rest o f the assayed compounds (Zn SO4, Na2SO4, (NH4)2SO4, MgCI2, NaCI, KCI, LiBr, KCN, EDTA and NAN3) showed no effect on the enzyme activities.

TABLE VII

Effect of some ionic compounds The enzyme samples of Hex A and Hex B were incubated in the presence of the indicated compounds at 10 mM concentration. GlcNAc-ase and GalNAc-ase activities were determined as indicated above. Results are gi,,'en as the average value of 4 experiments (standard deviations < 10 %).

Relative activity (%)

10 mM Hex A Hex B

GIcNAc-ase GalNAc-ase GIcNAc-ase GalNAc-ase

!00 100 100 100 CuSO4 147 120 93 91 NiSO4 ! 05 199 92 73 MgSO4 ! 03 104 109 116 MnCI2 113 116 ! 18 129 CaCi2 105 ! 16 ! 10 122 BaCI2 109 109 ! 1 ! 113 HgCI2 8 9 9 9 Pb(AcO)2 50 27 56 30

302 E. Villar, J.A. Cabezas and P. Calvo

Discussion

The specific activities of hexosaminidases A and B purified from the digestive gland of the slug Arion rufus L. are higher than most of the hexosaminidase activities reported different sour- ces. Since high specific activities are always an advantage for applied studies such as those in- dicated in the Introduction, and for other pur- poses, that material is very suitable as a source for this enzyme.

Thermal inactivation procedure [11] did not give representative results to establish the propor- tion of each enzyme form in the crude extract.

Saifer and Perle [48] found by thermal inactiva- tion an overestimation of thermolabile hexosami- nidase, like in our results (Table III) that show an overestimated percentage (97 %) for hexosamini- dase A. This percentage strongly disagrees with the chromatographic pattern through DEAE-cel- lulose (Fig. 1), so that it has not been considered. The proportions found for A and B forms (37 and 63 %, respectively) by the DEAE-cellulose me- thod [12] in column or in tube are similar to those reported by Sandhoff et al. [15] for human liver, and they are in agreement with the DEAE-cellu- lose profile (Fig. 1); therefore, those percentages have been considered for calculations.

The high stability of hexosaminidase B over a very wide pH-range was specially useful in the purification of this form by CM-cellulose chroma- tography, that resulted in a highly efficient step.

In the affinity chromatographies on Con A-Sepharose we found that the elution with ¢t- methylg!ucoside :,raproved both purification fac- tor and yield with respect to the elution with tt-methylmannoside. Furthermore, mannose and manno-_.ides are strong inhibitors of hexosamini- dase activities [19] (see Table VI). Both hexosami- nidast:z A and B are glycoproteins, as is usual for glyco: .~lases, because of their binding to conca- navaJin A.

The different molecular weights found for hexosam nidases A and B (150,000 and 320,000, respectively) in an approximate proportion 1 "2 suggest that hexosaminidase B presents, in its quaternary structure, an aggregation order higher than hexosaminidase A, from monomeric subu- nits of similar molecular weight in both enzyme forms, as is usual for glycosidases. Munn et al. [49] also presume that the different molecular weight forms of hexosaminidase from intestinal mucosa of newborn pigs represent a series of aggregates or associations of a basic unit. We

always found the indicated molecular weight values, using samples from different purification steps, and using different ionic strengths (from 50 to 200 mM) in the elution citrate buffer. On the other hand, Hasilik et al. [50, 51] reported that metabolically active hexosaminidases are synthe- sized as higher molecular weight precursors. Those precursors are also active. This fact may be related to the occurrence of different sized molecules.

The differences in the amino acid composition of our purified hexosaminidases A and B concern in most cases the same amino acids than in other hexosaminidases [13]. Those differences are consistent with the reported isoelectric points of both enzyme forms.

The Km values of our purified forms A and B are similar to those reported for other molluscs by P6rez and Cabezas [5], Calvo et al. [6] and Santoro and Dain [52]; but much lower than those reported by Yeung [7].

All the kinetic data confirm that [i-N-acetylglu- cosaminidase and I~-N-acetylgalactosaminidase activities are catalyzed in the same active site in the hexosaminidases A and B reported here. These results agree with those reported for other hexosaminidases from different sources such as jack bean meal [3], beef spleen [16], fenugreek seeds [33], the terrestrial snail Helicella ericetorum [6], and sheep liver [53]. However, the occurrence of one or two active sites in hexosaminidases, for the catalysis of 13-N-acetylglucosaminidase and [3-N-acetylgalactosaminidase activities, is a controversial point.

Two active sites have been suggested for the catalysis of the activities of this enzyme from other sources: the fungus Sclerotinia fructigena [32], rat colon [21, 34], bovine kidney [35] and a human colonic carcinoma cell line [54]. Neverthe- less, some of those reported results were obtained with only partially purified enzymes, and no definitive conclusion can be inferred from impure preparations. All the results that we describe here have been obtained with completely pure prepa- rations of both hexosaminidases A and B.

The inhibition found with some of the carbo- hydrates tested (Table VI) corresponds to their structural analogies with respect to the substrates; therefore, those sugars would compete with them in the active site. Similar inhibition results were found in horse brain [18]. The inhibition of the enzyme activities produced by mannose and mannosides may be explained by taking into consideration the critical location of mannose

Two molecular forms o f [~-N-acetylhexosaminidase f rom a mollusc 303

residues in N-acctylhexosamines-containing gly- coconjugates [19].

The stimulation of both activities found with ribose in both enzyme forms might be explained by a mechanism of transglycosylation currently assigned to glycosidases, as previously reported [ss].

The activation by chlorides (Table VII) was also described for hexosaminidases from other molluscs [7]. The strong inhil:ition found with Hg 2+ suggests that this enzyme may have essential thiol groups involved in the catalysis of all the activities, as is usual in acid-catalyzed hydrolysis of glycosidases [56].

Acknowledgments :

This work, supported by grants of the Ministerio de Educaci6n y Ciencia, Comisi6n Asesora de lnvestigaci6n Cientifica y Tdcnica and lnstituto Nacional de Ayuda y Promoci6n del Estudiante (Spain), fulfils, in part. requio rements for a Ph. D. for E.V. We thank Dr. J.L. Stirling (Queen Elizabeth College. London)for his kind gift of the ANAG-Sepharose gel, and Dr. E. Mdndez (Centro Ram6n y Cajal. Madrid)for the amino acid analysis. The secretarial work by Miss Beldn Bi6zquez is also ackno- wledged.

REFERENCES

I ! ~,~ I i= q ~ V r h c h l d a , A 1 1 0 7 6 ~ R i a r h o m _! I~0_

535-539. 2. Overdijk, B., van Steijn, G., Wolf, J.M. & Lisman,

J.J.W. (1982) Int. J. Biochem., 14, 25-31. 3. Li, S.C. & Li, Y.T. (1970) J. Biol. Chem., 245,

5153-5160. 4. Bouquelet, S. & Spik, G. (1978) Eur. J. Biochem..

84, 551-559. 5. P6rez, N. & Cabezas, J.A. (1977) Biochimie. 59,

729-733. 6. Calvo, P., Reglero, A. & Cabezas, J.A. (1978) Bio-

chem. J., 175, 743-750. 7. Yeung, K.K., Owen, A.J. & Dain, J.A. (1979) Comp.

Biochem. Physiol., 6311, 329-334. 8. Robinson, D. & Stifling, J.L. (1968) Biochem. J..

107, 321-327. 9. Okada, S. & O'Brien, J. (1969) Science. 165,

698-700. 10. Sandhoff, K. (1969) FEBS Lett., 4, 351-354. !1. O'Brien, JS., Oka6a, S., Chen, M.D. & Fillerup,

D.L. (19'70) New Engl. J. Med., 283, 15-20. 12. Dance, N., Pierce, R.G. & Robinson, D. (1970)

Biochim. Biophys. Acta. 222, 662-664. 13. Geiger, B. & Arnon, R. (1976) Biochemistry, 15,

3484-3493.

14. Beutler, E., Yoshida, A., Kuhl, W. & Lee, J.E.S. (1976) Biochem. J., 159, 541-543.

15. Sandhoff, K., Conzelmann, E. & Nehrkorn, H. (1977) Hoppe-Seyler's Z. Physiol. Chem.. 358, 779-787.

16. Verpoorte, J.A. (1972) J. Biol. Chem., 247, 4787-4793.

17. Lisman, J.W. & Overdijk, B. (1978) Hoppe Seyler's Z. Physiol. Chem., 359, I 0 ! 9-1022.

18. Reglero, A., Esteban, M. & Cabezas, J.A. (1981) Int. J. Biochem., 13, 837-842.

19. Reglero, A. (1979) Int. J. Biochem.. 10, 285-288. 20. Oberkotter, L.V., Kern, R. & Koldowsky, O. (1981)

Biochim. Biophys. dcta, 630, 279-286. 21. Mian, N., Herries, D.G., Cowen, D.M. & Batte,

E.A. (1977) Biochem., 3.. i77, 319-330. 22. Calvo, P., Reviila, M.G. & Cabezas, J.A. (1978)

Comp. Biochem. Physiol.. 61B, 581-585. 23. Ahladeff, J.A. & Holzinger, R.T. (1982) Biochem. J.,

201, 95-99. 24. Kind, P.R.N. (1982) Clin. Chim. Acta. 119, 89-97. 25. Reglero, A., Carretero, M.I. & Cabezas, J.A. (1980)

Clin. Chim. ,4cta, 103, ! 55- ! 58. 26. Calvo, P., Barba, J.L. & Cabezas, J.A. (1982) Clin.

Chim. Acta. 119, 15-19. 27. Miralles, J.M., Corrales, J.J., Garcia-D[ez, L.C.,

Cabezas, J.A. & Reglero, A. (1982) Ciin. Chim. Acta, 121, 373-378.

28. Cabezas-Delamare, M.J., Reglero, A. & Cabezas, J.A. (1983) Clin. Chim. Acta, 128, 53-59.

29. Cabezas., J.A., Calvo, P., Dnez, T., Melgar, M.J., OrtiT: M.V., de Pedro, A., P6rez, N., Reglero, A., Santamaria, M.G. & Villar, E. (1982) Rev. Esp. Fisiol., 38 supl., 73-80.

30. Cabezas, J.A., Reglero, A. & Calvo, P. (1983) Int. J. Biochem.. 15, 243-259.

31. Muramatsu, T. (1968) J. owcnem., . . . . . . . . . . o,o, 521-531. 32. Reyes, F. & Byrde, R.J.W. (1973) Biochem. J.. 131,

38 ! -388. 33. Bouquelet, S. & Spik, G. (1976) FEBS Lett.. 63,

95-101. 34. Mian, N., Herries, D.G. & Batte, E.A. (1978) Bio.

chim. Biophys. Acta. 523, 454-468. 35. Pope, A.J., Mian, N. & Herries, D.G. (1978) FEBS

Lett.. 93, ! 74- ! 76. 36. Lowry, O.H. Rosebrough, N.J., Farr, A.L. &

Randall, R.J. (1951) J. Biol. Chem.. 193, 265-275. 37. B61em, P., Stein, S., Dairman, N. & Undenfriend,

S. (1973) Arch. Biochem. Bio[,,~vs., 115, 213-220. 38. Akiki, C., Bourbouze, R., Luporsi, C. & Percheron,

F. (1980) J. Chromatogr.. 188, 435-438. 39. Davis, B.J. (1964) Ann. N.Y. Acad. Sci.. 121,

404-427. 40. Morrissey, J.H. ( ! 981 ) Anal, Biochem.. i ! 7, 307-310. 41. Vesterberg, O. (1971) Methods Enzymol., 22,

389-412. 42. Moore, S. & Stein, W.H. (i963) Methods Enzymol..

6, 819-831. 43. Moore, S. (1963) J. Biol. Chem., 238, 235-237. 44. Lineweaver, H. & Burk, D. (1934) J. Am. Chem.

Soc.. 56, 658-666.

304 E. Viilar, J.A. Cabezas and P. Calvo

45. Dixon, M. (1953) Biochem. J., 55, 170-171. 46. Dixon, M. & Webb, E.C. (1964) Enzymes, 2nd edn,

pp. 54-90, 201-209, Longmans Green & Co., London.

47. Sols, A. & Crane, R.K. (1954) J. Biol. Chem., 210, 581-595.

48. Saifer, A. & Perle, G. (1973) Am. J. Obstet. Gynecol., !15, 553-555.

49. Munn, E.A. Greenwood, C.A., Orlacchio, A. & Porcellati, G. (1979) Int. J. Biochem., 10, 1045-1052.

50. Hasilik, A. & Neufeld, E. (1980) J. Biol. Chem., 255, 4937-4945.

51. Hasilik, A., Von Figura, K., Conzelmann, E.,

Nehrkorn, H. & Sandhoff, K. (1982) Eur. J. Biochem., 125, 317-321.

52. Santoro, P.F. & Dain, J.A. (1981) Comp. Biochem. Physiol., 69B, 337-344.

53. Donoso, L.A. & Spikes, J.D. (1980) Enzyme, 25, Ill-117.

54. Kimball, P.M., Brattain, M.G. & White, W.E. (1981) Biochem. J. 193, I09-113.

55. Calvo, P., Santamaria, M.G. Meigar, M.J. & Cabezas, J.A. (1983). Int. J. Biochem., 15, 685-693.

56. Wallenfels, K. & Weil, R. (1972) In The Enzymes (Edited by Boyer P.D.), 3rd edn, vol. 7, pp. 617-663, Academic Press, New York.