Enzymatic study of GM1 gangliosidosis

283

Clinica Chimica Acta, @ Elsevier

Scientific

59 (1975) 283-290 Publishing Company,

Amsterdam

-

Printed

in The Netherlands

CCA 6857

ENZYMATIC

STUDY

OF GM1 GANGLIOSIDOSIS

TORU YUTAKA, SHINTARO HYAKUJI YABUUCHI

of Pediatrics,

Department (Received

September

OKADA,

KOJI MIMAKI,

Osaka University

TAKAHIRO

Medical School,

SUGITA

and

Osaka (Japan)

27, 1974)

Summary Two types of GM1 gangliosidosis were studied biochemically. Type 1 liver accumulated non-lipid hexosamine in addition to GM1 ganglioside, but there was no increase of hexosamine and GM, in type 2 liver. The optimum pH of liver beta-galactosidase of type 1 and type 2 was 5-6 while that of the normal control was 4.5. Type 1 brain beta-galactosidase showed a slightly acidic optimum pH, i.e. 4.0 in comparison with that of the normal control. The optimum pH of type 2 brain beta-galactosidase was 5.5, like the liver enzyme. Thermostability of liver beta-galactosidase was the same in type 1 and type 2, while that of brain was different. Beta-galactosidase of type 1 and type 2 liver is more stable at 42°C than the normal control, but a different thermostability was observed in type 1 and type 2 brain. Liver beta-galactosidase of type 1 showed one peak each at acid and neutral pH, and type 2 liver had only one peak at neutral pH.

Introduction GM, gangliosidosis is an inherited disorder characterized by the deficiency of beta-galactosidase and the storage of ganglioside GM1 in tissues [1,2]. The disease occurs in two clinical forms, that is, early onset, type 1 and late onset, type 2 [3] . These two types differ in clinical findings. The type 1 patient has Hurler-like facies, severe bony deformities, and hepatosplenomegaly. On the other hand, the type 2 patient shows a slower progress of symptoms and no abnormal facies and visceromegaly. Mucopolysacchariduria and visceral histiocytosis are observed in type 1, but not in type 2. The genetic disorder is caused by the deficiency of beta-galactosidase, however, very low activity of betagalactosidase has been observed. Several investigators reported on the nature of the residual enzyme in type 1 and type 2 [4--81. The present study was carried out in order to elucidate the difference in the nature of the residual enzyme. Liver and brain beta-galactoside from both type 1 and type 2 patients were studied.

284

Materials and Methods Tissue preparation Autopsy specimens were immediately frozen and stored at -20°C until use. Brain and liver tissues were homogenized in distilled water using a glass homogenizer. After 1 min sonication (10 kcycles/s), the homogenate was centrifuged at 15 000 X g for 20 min at 4°C and the supernatant was used for the enzyme assay. Preparation of ‘H-labeled ceramide tetrahexoside (asialo GM, ganglioside) GM, ganglioside was obtained from the cerebral gray matter of a GM1 gangliosidosis patient. Crude GM, was part,ially hydrolyzed with 0.1 N hydrochloric acid by the method of Ledeen et al. [9]. Crude ceramide tetrahexoside obtained by partial hydrolysis of GM, was oxidized by galactose oxidase and reduced by sodium [ 3H] borohydride according to the method of Radin et al. [lo] . The labeled ceramide tetrahexoside was purified on silica gel thin-layer chromatography. Procedure for analysis of lipid Total lipid was extracted from the tissue with 20 vol. of chloroform/ methanol (2 : 1, v/v) and an additional 10 vol. of same solvent. The extract was washed three times with 0.2 vol. of 0.88% KCl, 0.05 M KC1 containing Folch’s upper phase and Folch’s upper phase. Each upper phase wa9 pooled and evaporated to dryness in vacua after dialysis against water. The lower phase was’ evaporated to dryness in vacua and the total lipid was weighed. Crude gangliosides from the upper phase were separated by ascending thin-layer chromatography of silica gel H using chloroform/methanol/2.5 N ammonia (60 : 40 : 9, v/v) as solvent and each ganglioside was analyzed according to Suzuki [ll] . Sialic acid (NANA) was assayed by the resorcinol method [12] . Determination of the non-lipid hexosamine was performed according to Suzuki et al.

[131. Standard assay of beta-galactosidase Activity of beta-galactosidase was assayed using 4-methylumbelliferyl betaD-galactopyranoside (Pierce Co.) as substrate. The reaction mixture (400 ~1) contained 200 ~1 of 0.1 M citrate/phosphate buffer (pH 3.6 or pH 6.5), 0.5 mM of 4-methylumbelliferyl beta-D -ga.lactopyranoside, 0.1 M of sodium chloride and 100 1.11of enzyme preparation. Following incubation for 30 min at 37”C, the reaction was stopped by the addition of 2 ml of 0.2 M glycine/carbonate buffer, pH 10.6. The 4-methylumbelliferone released was assayed on a spectrofluorimeter at an excitation wavelength of 364 nm and an emission wavelength of 447 nm. Specific glycosphingolipid beta-galactosidases were assayed using tritiumlabeled lactosylceramide and ceramide tetrahexoside according to Gatt and Rapport [14] . Liver was homogenized in six volumes of Tris/magnesium/mercaptoethanol/buffer and was sonicated (10 kcycles/s) for 1 min. The reaction mixture containing 200 nM of ceramide tetrahexoside (2 - lo6 cpm/nM), 0.5 mg of Triton X-100, 2 mg of sodium taurocholate, liver homogenate (3 mg of

285

tissue) and 0.04 M of sodium acetate buffer pH 5.0, was incubated at 37°C for 2 h. The reaction was terminated by the addition of chloroform/methanol (2 : 1, v/v) and unlabeled galactose. Galactose was separated by partition and the radioactivity of the total upper phase was assayed on a liquid scintillation counter. Results Case study Case 1 (GM1 gangliosidosis type 1) was a male. Bony deformity was first noticed at 6 months of age. His face was abnormal with a Hurler-like appearance. Vacuolated lymphocytes in the peripheral blood and foam cells in the marrow were observed, but hepatosplenomegaly was not recognized. He died at 25 months of age. Case 2 (GM1 gangliosidosis type 2) was a female, whose mental and motor development was nearly normal for the first year, but she never walked and at the age of 17 months generalized seizures occurred. Since then her mental and motor development began to deteriorate. It progressed rapidly and she died at 45 months of age, The total lipid content of cerebral gray matter and its constituents were in the normal range. Ganglioside of the cerebral gray matter was markedly increased and was 2,229 pg NANA/g of wet tissue weight for case 1 and 2,139 pg NANA for case 2. GM1 ganglioside amounted to 77.9% of the total ganglioside in the gray matter of both patients (Table I). On the other hand the non-lipid hexosamine of gray matter from both patients was in the normal range. Ganglioside analysis of the liver by thin-layer chromatography revealed an obvious increase of GM, ganglioside in case 1 and no increase in case 2. The sialic acid content of liver ganglioside fraction was 144 pg/g of wet tissue weight in case 1 and this is 2 times more than case 2 and control. Non-lipid hexosamine was also increased 7 times more in case 1 than in the control, but that of case 2 was in normal range (Table II). TABLE

I

GANGLIOSIDE

ANALYSIS

OF

CEREBRAL

Case 1 (type Age:

Non-lipid Total

residue*

lipid*

water**

32

GRAY

MATTER

Case 1)

months

2

(type 45

Control (6 cases)

2)

months

2-8

year

62.8

56.1

58.6

37.2

43.9

41.4

90.2

86.6

87.1

2.28

1.59

0.58

GM3

3.3

3.3

1.8

GM2

3.0

3.0

2.0

GMI

77.9

77.9

23.2

GDla

5.0

7.7

37.7

GDIb

5.0

1.9

16.3

GTI

5.8

6.1

19.0

Ganglioside

* **

NANA*

Percent

of dry

weight.

Percent

of wet

weight.

Each

ganglioside

was expressed

as percent

distribution

of NANA.

TABLE II CHEMICAL

ANALYSIS

OF LIVER

Case

Non-lipid residue* Total lipid* Non-lipid hexosamine** Ganglioside NANA* * *

1

Case 2

Control

(type 1)

(type 2)

70.1 29.9 5326 144

62.1 37.3 1030 70

64.8 35.2 744 73

* Percent of dry weight. ** pg/lOO mg of non-lipid residue. *** fig/g of wet weight.

Activity of beta-galactosidase of liver and brain Liver and brain beta-galactosidase was examined at various pH values. Liver beta-galactosidase of both cases was hardly detectable at lower pH and very slight activity was observed at neutral pH. Type 2 liver was more active at neutral pH than type 1, but it was significantly lower than the normal control. The optimum pH of beta-galactosidase of both type 1 and type 2 liver was 5-6 whereas that of the normal control was 4.5 (Fig. 1). Brain beta-galactosidase showed maximum activity at different pH values, i.e. pH 4.5 in the normal control, pH 4.0 in type 1, and pH 5.5 in type 2 (Fig. 2). Thermostability of beta-glactosidase of liver and brain Liver and brain homogenate was heated at 42°C in citrate/phosphate buffer, pH 6.5. Normal liver beta-galactosidase lost 40% of its activity in 60 min

3

4

5

6

7

8

P’+

Fig. 1. pH-activity curves of liver beta-galactosidase. 4-Methylumbelliferyl beta-D-galactopyranoside are described in the text.

type 2.

‘)--------O, control; O--d , type1; x-x, was used as substrate. Details of the assay method

Fig. 2. pH-activity curves of brain beta-galactosidase. O----O, control; n -------0, type 2. The assay method was the same as for liver beta-galactosidase.

type 1; x----x,

287

30

min

Fig. 3. Thermostability of liver (A) and brain (B) beta_galactosidase. o----+J, control; n-----o. type 1; X---X. type 2. The liver and brain homogenate were heated to 42OC at PH 6.5 assayed at the same PH using 4-methylumbelliferyl beta D-galactopyranoside as substrate. Residual activities were expressed as a percentage of the initial activity.

but the decrease of activity was less in type 1 and type 2 liver. On the other hand, the brain beta-galactosidase of type 1 patient was like that of the normal control and beta-galactosidase of type 2 brain showed a minimum decrease of activity (Fig. 3). Specific beta-galactosidase activity Ceramide tetrahexoside labeled with tritium was used as the substrate of beta-galaetosidase. The activity of type 1 and type 2 liver homogenate was not detectable. Tritium-labeled ceramide lactoside was also used as substrate, and ceramide lactosidase of liver from both types showed normal activity (Table III). TABLE

III

BETA-GALACTOSIDASE

ACTIVITY

OF LIVER

AND BRAIN

Lactosylceramide and asialo GM1 beta-galactosidase activities were expressed as dpm/mg of tissue weight per h. 4-Methylumbelliferyl (4-MU) beta-galactosidase activities were estimated at pH 4.5 and expressed as nmoles of released C-methylumbelliferon/mg of protein per h. Liver Substrate

Case 1 (type 1) Case 2 (type 2) Control (n = 2)

Bl%iIl

Lactosylceramide

Asialo GMI

4-MU-betagdlactoside

4-MU betagalactoside

5270 5553 4374-4515

nil nil 15658-81155

5.1 44.5 204-426

9.2 3.5 45-68

288

30

10 Fraction

50 No.

4. Sephadex G-ZOO column chromatography of liver beta-galactosidase from control (A), type 1 (B), and type 2 (C). Sephadex G-200 was packed in a 60 cm X 0.9 cm column. Supernatant of liver homogenate was eluted with 0.01 M sodium phosphate buffer pH 7.5 containing 0.1 M NaCl and 1 mM EDTA. and 0.8sml fractions were collected. The beta-galactosidase activity was assayed at pH 3.6 and 6.5 in each fraction. l ------0, activity at pH 3.6; (l------c), activity at pH 6.5. Beta-galactosidase activity was arbitrary scaled.

Fig.

Chromatographic

separation

of liver beta-galactosidase

Liver beta-galactosidase was separated into two parts by Sephadex G-200 column chromatography when the activity was assayed using 4-methylumbelliferyl beta-D -galactopyranoside as substrate. Both peaks showed a pH optimum of 4-5. The apparent Km value is 0.36 . 10V3 M for both peaks. Sephadex G-200 column chromatography of type 1 liver beta-galactosidase gave a small peak which coincided with the first peak of the normal control when assayed at pH 3.6, but type 2 liver did not show activity at pH 3.6. The main enzyme activities at pH 6.5 in the liver of both type 1 and type 2 were eluted at the same fraction of the second peak of normal beta-galactosidase (Fig. 4). Apparent Km values of type 1 and type 2 liver homogenate were 0.18 - 10e3 M at both 3.6 and 6.5. Discussion Studies on the residual beta-galactosidase of two clinical types of GM, gangliosidosis have been carried out by many authors. The optimum pH of liver beta-galactosidase of GM, gangliosidosis type 1 and type 2 showed an obvious shift toward neutral pH from the normal control when assayed with whole homogenate using 4-methylumbelliferyl beta-D -galactoside as substrate. This is

289

consistent with the results obtained by Singer et al. [6] . The optimum pH of brain beta-galactosidase of type 1 was about pH 4 and was slightly more acidic than the normal control, while that of type 2 was like the liver enzyme. In contrast, Chou et al. [7] have reported that the optimum pH of type 1 and type 2 brain was pH 3. The reason for this discrepancy is not known, but their specimen of type 2 was fetal brain and our specimen was autopsied material. The thermostability of beta-galactosidase has been studied in the liver, skin fibroblast and brain, and the difference has been observed between normal and patient samples [4,7,15]. Our results confirmed their findings. Beta-galactosidase of both type 1 and type 2 liver showed nearly identical stabilities, but that of the normal control was more labile at pH 6.5. On the other hand, the stabilities of brain beta-galactosidase were clearly different. Type 2 was the most stable, type 1 was intermediate, and the normal control was the least stable. Norden and O’Brien [16] reported that the GM1 beta-galactosidase activities in normal liver correspond with that of 4-methylumbelliferyl beta-galactosidase. In contrast they could not find GM1 beta-galactosidase activity in GM1 gangliosidosis type 1 and type 2, although the residual 4-methylumbelliferyl beta-glactosidase was detectable in both types. We examined ceramide tetrahexoside beta-galactosidase and could not find activity either in type 1 or type 2 liver, whereas liver and brain of both types showed a normal activity of lactosylceramide beta-galactosidase. Therefore, the enzyme activities toward the artificial substrates do not reflect the condition of enzymes toward natural substrates. Suzuki and Suzuki [B] observed an abnormal pattern of lactosylceramide beta-galactosidase in type 1 and very low activity in type 2 liver. Thus more cases have to be studied in order to elucidate the discrepancy of the activity of lactosylceramide beta-galactosidase. In spite of the deficiency of GM, and asialo GM, beta-galactosidase both in type 1 and type 2, the isozyme patterns of 4-methylumbelliferyl beta-galactosidase of type 1 and type 2 liver are different. Moreover, thermostability and optimum pH of brain beta-galactosidase also differ in type 1 and type 2. GM1 , asialo GM, and galactose containing glycoprotein accumulate in type 1, and GM1 and asialo GM1 in type 2. This difference in accumulated substances may be due to the nature of the residual enzyme. The data on the nature of the residual enzyme reported up to this time indicate a difference in type 1 and type 2. These results suggest that the two types of GM1 gangliosidosis are different biochemical mutants. The precise nature of the enzyme should be elucidated with the appropriate natural substrates. Acknowledgements The authors thank Dr Tadao Orii for tissue specimens from GM, gangliosidosis type 1 patient and Dr Shizuo Handa for a sample of tritiated lactosylceramide. This investigation was supported in part by Research Grants from the Ministry of Education and from the Ministry of Public Welfare.

290

References 1

J.S.

O’Brien,

M.B.

Stern,

B.H.

Landing,

J.K.

O’Brien

and

G.N.

Donnell,

Am.

J. Dis.

Child.,

109

(1965)

338 2

S. Okada

3

D.M.

and

4

L. Pinsky,

E. Powell

5

Y.

A.C.

6

H.S.

Derry.

Suzuki. Singer

7

L. Chou.

8

Y.

Suzuki

R.

Ledeen,

9 10

N.S.

J.S. J.S.

Suzuki.

K.

L. Svennerholm,

13

K.

14

S. Gatt

Suzuki,

15

M.W. A.G.W.

and

Salsman,

L. Hof.

11

16

and Suzuki.

Life K.

Sci.,

M.M.

Ho and Norden

and

and

O’Brien,

Nature.

Nadler.

and

Wolfe,

Neural.,

(1972)

8 (1974)

(1974)

A.

Taghavy,

R.O.

Brady,

18

(1968)

340

1093 24 (1971)

24

Res.,

249

Neurology,

(1970)

Genet.,

Pediat.

and

1002 L.S. 228

Arch.

Chem.,

Bradley

3 (1964)

and

J. Hum.

J. Gonatas

R.M.

and J.S.

(1968)

Suzuki,

J. Biol.

Rapport.

J.S.

K.

Am.

H.L.

Biochim. Suzuki

160

J. Callahan.

Schafer.

K.

K.

12

and

Kaye

and

Radin,

I.A.

Science,

F. Andermann

Crocker

and

C.I.

O’Brien, Fawcett,

58

454 120

2113 J. Neuropathol. Brain

Exp.

Res..

14

Exp.

Neural..

Neural.,

(1969)

24

497

1227

Biophys.

Acta,

S. Kamoshita, Biochem. Clin. O’Brien,

24

(1957)

604

J. Neuropathol. J.. 101

Chim. Arch.

Acta,

(1966) 32

Biochem.

28

(1969)

680

(1971)

443

Biophys.,

159

(1973)

383

25

(1965)

341