Debranching enzyme in fibroblasts, amniotic fluid cells and chorionic villi: pre- and postnatal diagnosis of glycogenosis type III

Clinica Chimica Acfa, 149 (1985) 129-134 Elsevier

129

CCA 03206

Debranching enzyme in fibroblasts, amniotic fluid cells and chorionic villi: pre- and postnatal diagnosis of glycogenosis type III O.P. van Diggelen

a, H.C. Janse

a and G.P.A.

Smit

b

u Department of Cell Biology and Genetics, Erasmus University, P.O. Box 1738, 3000 DR Rotterdam and ’ Department of Pediatrics,Universrty Hospital, University of Groningen, Groningen (The Netherlands) (Received

November

28th. 1984; revision March 6th. 1985)

Key words: Debranching enzyme; Glycogenosis type III; Fibroblasts; Amniotic fluid cells; Chorionic villi; Prenatal diagnosis

Summary Glycogenosis type III is characterized by a deficiency of debranching enzyme in most tissues, and it can be detected by the inability to liberate glucose from limit dextrin. However, using this assay, the deficiency is not expressed in cultured fibroblasts from patients with glycogenosis type III. We have demonstrated that the failure to detect debranching enzyme deficiency in fibroblasts is entirely due to interference of acid a-glucosidase, which can also hydrolyse limit dextrin. A method is described to remove specifically acid a-glucosidase allowing clear discrimination between fibroblasts from patients and controls, whereas heterozygotes showed intermediate values. The results with amniotic fluid cells and chorionic villi suggest the feasibility of first- and second-trimester prenatal diagnosis of glycogenosis III.

Introduction Glycogenosis type III is an autosomal recessive disorder resulting from a deficiency of glycogen debranching enzyme [l]. The clinical symptoms are usually limited to the physiological consequences of impaired glycogenolysis; sometimes increased accumulation of glycogen in muscle and heart may occur. Normally the majority of glucose units from glycogen are released by the sequential action of glycogen phosphorylase on 1,4 linkages. This enzyme halts close to 1,6 branching points, leaving the characteristic limit dextrin structure. Debranching enzyme is comprised of two components: oligo-(1,4-1,4)-glucan transferase (EC 2.4.1.25) and amylo-1,6-glucosidase (EC 3.2.1.33), which act in a two-step fashion to remove the branching points. 0009-8981/85/$03.30

0 1985 Elsevier Science Publishers

B.V. (Biomedical

Division)

130

Liver from patients with glycogenosis type III accumulates the highly branched limit dextrin and usually lacks both debranching activities. Enzyme deficiencies have been demonstrated in muscle [l], liver [2], leucocytes [3] and erythrocytes [4]. Variants deficient in only one of the debranching activities have also been described [2] and a number of studies have shown that the separate amylo-1,6-glucosidase deficiency can be demonstrated in cultured cells [5-71. For diagnostic purposes however, it is preferable to assay the overall, two-step, debranching activity by measuring the glucose release from limit dextrin. This has not been reported for cultured fibroblasts. In this paper we describe a method for the detection of debranching enzyme deficiency in cultured fibroblasts using a limit dextrin based assay. This method is also applied to amniotic fluid cells and chorionic villi to examine the feasibility of prenatal diagnosis of glycogenosis III. Materials and methods Cell culture and preparation of cell extracts Human skin fibroblasts and amniotic fluid cells were cultured in Ham’s F10 medium supplemented with 5% fetal bovine serum plus 5% newborn bovine serum or 20% fetal bovine serum respectively [8]. The cells were harvested 7 days after the last subculture and stored at -70°C until used. Fibroblasts from heterozygotes for glycogenosis III, not related to the patients described, were kindly provided by Drs. M. Mathieu, S. Boyer and M.T. Zabot (Lyon) and Dr. M. Lluch (Barcelona). Fibroblasts of two of the patients were obtained through Dr. M. Macek (Prague) and Dr. H. Przyrembel (Rotterdam). Normal chorionic villi were obtained after elective abortion of pregnancies in the 8th-11th wk. Chorion biopsy and amniocentesis was carried out by Dr. M.G.J. Jahoda (Academic Hospital, Rotterdam). A pellet of lo6 frozen cultured cells was suspended in 150 ~1 H/N buffer (10 mmol/l histidine/NaOH, pH 7.0, 0.02% NaN,) and sonicated for 10 s. Chorionic villi homogenates (lo%, wet wt/vol) were prepared with a Potter homogenizer, followed by sonication. After centrifugation for 10 min at 7,000 X g, the supernatants were taken and used for assays. A 50% slurry (W vol packed beads) of concanavalin A (Con A)-Sepharose 4B (Pharmacia, Uppsala, Sweden) was freshly prepared in H/N buffer containing 9 mmol/l a-methylmannoside and 40 ~1 of this slurry was carefully pipetted onto the bottom of a 1.5 ml plastic reaction tube. Supernatant (80 ~1) was then pipetted with caution onto the slurry without touching the tube with the tip of the pipette. Subsequently, the reaction tube was carefully placed horizontally in a rack and gently rotated at 1 revolution per min for 45 min at room temperature. Due to the surface tension the meniscus stayed vertical and efficient mixing of beads and supernatants was achieved while all the beads remained immersed continuously. After centrifugation for 5 min at 1,000 X g the post Con A supernatant was used for the enzyme assays. Enzyme assays Limit dextrin was prepared from rabbit liver glycogen, essentially as described [9] using rabbit liver phosphorylase B (Boehringer, Mannheim, FRG). Debranching

131

enzyme was determined by mixing 10 ~1 post Con A supernatant and 20 ~1 substrate solution (9 mg limit dextrin/ml 150 mmol/l histidine/NaOH buffer, pH 6.5, containing 1 mmol/l EDTA and 0.02% NaN,). For cultured cells reaction mixtures were incubated for 24 h at 37°C under oil (80% paraffin, 20% hexadecane); chorionic villi were incubated for 5 h. The reaction was terminated by boiling the samples for 2 min after which 10 ~1 N-ethylmaleimide (100 mmol/l) was added to inactivate SH-groups which interfere with the glucose determination. Then, 200 ~1 glucose reagent was added which contained 0.5 mol/l Tris . HCl (pH 7.0) 10 mmol/l MgCl,, 2.1 mmol/l, 2,2’-azino-di-[3-ethylbenzthiazolinsulphonate(6)], 8 pg glucose oxidase and 2 pg peroxidase (grade 1 Boehringer). The tubes were placed horizontal to maximize oxygen diffusion and after 60-90 min at room temperature the samples were clarified by centrifugation and the absorbance at 420 nm was read. Acid a-glucosidase was measured with 4-methylumbelliferyl-cY-glucoside (MU-cu-glucoside) at pH 4 as described [8]. Results The commonly used assay for debranching enzyme, measuring glucose liberation from limit dextrin [9], does not allow discrimation between normal and glycogenosis III fibroblasts (Table I). The experiments in Fig. 1 demonstrate that the debranching enzyme deficiency in glycogenosis III cells is obscured by interference of acid a-glucosidase which can hydrolyze 1,4- as well as 1,6-a-glucosidic linkages [lo]. Fibroblast supernatants were mixed with increasing amounts of Con A beads in the presence of 3 mmol/l a-methylmannoside. This compound prevented binding of debranching enzyme to Con A (Fig. 1, compare closed and open squares) and had no effect on the activities of debranching enzyme or a-glucosidase. This treatment resulted in removal of 96-988 acid a-glucosidase (Fig. 1, solid circles). The concomitant effect on glucose liberation from limit dextrin is also depicted in Fig. 1

TABLE

I

Debranching enzyme activity before and after Con A treatment amniotic fluid cells and chorionic villi glc Liberation

of supernatants

from limit dextrin (nmol/h

Untreated Mean

of cultured

fibroblasts,

per mg protein)

Con A treated Range

Mean

Range

Fibroblasts Glycogenosis III patients (n = 8) Glycogenosis III heterozygous (n = 6) Normal (n = 25) Amniotic fluid cells

72 82 105

46-139 55-96 38-208

1.4 14 24

0.4-3 13-17 13-34

Normal Chorionic

(n = 18) villi

61

32-146

22

12-36

Normal

(n = 10)

232

124-372

101

74-138

Normal

Glycogenosic

Glycogenosls

tyPen

tYPem

’ Ii

I

k 0

I1 10

20 3( I( 1 10 20 Con A- Sepharose

30 0 added

10 20 (~1)

30

Fig. 1. Debranching enzyme activity in various fibroblast strains after removal of acid a-glucosidase with Con A-Sepharose. Reaction mixtures of 40 pl contained 40-50 pg supernatant protein, amounts of 50% Con A-Sepharose slurry as indicated in the figure and 3 mmol/l a-methylmannoside (0 and n) or without a-methylmannoside (0 and q). Acid a-glucosidase (MU-a-ghtcosidase), 0 and 0. Limit dextrin hydrolysis, 0 and H. The enzymes were determined in post Con A supernatants as described in the ‘Materials and Methods’ section,

(solid squares) showing a reduction down to approximately 25%. These results suggest that a large fraction of limit dextrin hydrolysis by normal fibroblast supernatants is caused by acid a-glucosidase. This was proven by analogous experiments with glycogenosis type II fibroblasts, which have a genetic deficiency of acid a-glucosidase. With these cells, Con A treatment had no effect on limit dextrin hydolysis (Fig. 1, middle panel). In contrast, Con A treatment of glycogenosis III fibroblast supernatants resulted in a virtually complete loss of limit dextrin hydrolysis which parallelled the loss of acid a-glucosidase (Fig. 1, right panel). This indicates that the entire limit dextrin hydrolysis by glycogenosis III fibroblast supernatants is caused by acid a-glucosidase. Using post Con A supernatant from normal fibroblasts debranching enzyme activity was measured as a function of incubation time (Fig. 2). After an initial lag period of 45-90 min, debranching enzyme activity was linear with time up to 24 h and also linear with protein amounts ranging from 2-20 pg. Debranching enzyme was then assayed under standard conditions (see ‘Materials and Methods’) in various cell types (Table I). Fibroblasts‘ from patients with glycogenosis III were clearly deficient, whereas those from heterozygotes had activities intermediate between normal and mutant cells. Amniotic fluid cells were found

133

0

5

10 lncubatlon

15 time

20

25

(h)

Fig. 2. Dependence of debranching enzyme activity on incubation time and amount of protein. Debranching enzyme was determined under standard conditions after incubation periods as indicated in the figure. The various incubation mixtures contained 20 pg protein (v), 10 pg (Cl), 5 pg (A) or 2.5 pg (0).

to have activities comparable to fibroblasts. On the other hand in chorionic villi debranching enzyme activity is much higher (Table I) and an incubation time of 5 h suffices to measure the activity. Prenatal diagnosis was requested in one case (index patient included in Table I). The chorionic villi, obtained in the 9th week of this pregnancy, had low-normal debranching enzyme activity (77 nmol/h per mg), indicating a normal fetus. The diagnosis was confirmed with amniotic fluid cells obtained after amniocentesis in the 16th week of pregnancy. In these cells, low normal debranching enzyme activity was found too (12 nmol/h per mg). The pregnancy had not yet reached term. Discussion Conditions described for the standard, limit dextrin based assay of debranching enzyme in muscle, liver or leukocytes appeared not to be adequate for the assay in cultured fibroblasts. We showed that only 15-45% of total glucose liberation is accounted for by debranching enzyme. This obviously precludes detection of glycogenosis III in cultured fibroblasts by this assay and is probably the reason why limit dextrin based assays in these cells have not been reported. We have modified this assay by precipitation of interfering acid a-glucosidase with Con A-Sepharose. With this modification a marked deficiency of debranching enzyme was demonstrated in the cell strains of the 8 glycogenosis III patients studied, which were derived from seven unrelated families. This suggests that, as a rule, the overall debranching

134

enzyme activity is deficient in glycogenosis III fibroblasts; in contrast, separate amylo-3,6-glucosidase activity is not always deficient [6]. For diagnostic purposes the limit dextrin based assay has obvious advantages since it monitors defects in both activities: the transferase as well as the amylo-1,6-glucosidase. In untreated homogenates of muscle, liver, or leucocytes debranching enzyme deficiency can clearly be detected with a limit dextrin based assay [l-3]; apparently acid Lu-glucosidase does not interfere severely in these tissues. This is explained by the relatively low ratios of acid glycogen-a-glucosidase/debranching enzyme in these tissues: muscle 0.3, liver 4 and leukocytes 1.5. In contrast this ratio was found to be much higher in fibroblasts, amniotic fluid cells and chorionic villi: 50, 2.5 and 15, respectively. Consequently, substantial hydrolysis of limit dextrin is caused by acid cY-glucosidase in these tissues. The debranching enzyme activity found in normal amniotic fluid cells, chorionic villi and in material from the prenatal diagnosis, suggest that glycogenosis type III can be detected prenatally by using the modified limit dextrin based assay. Acknowledgements

We thank Prof. II. Galjaard, Prof. J.F. Koster and Dr. W.J. Kfeijer for their valuable suggestions during the experimental work and preparation of the manuscript. References I Illingwor~ G, Cori GT, Cori CF. Amylo-1,6-~ucosidase in muscle tissue in generalized storage disease. J Biol Chem 1956; 218: 123-129. 2 Van Hoof F, Hers HG. The subgroups of type III glycogenosis. Eur J Biochem 1967; 2: 265-270. 3 Huijing F. Amylo-1,6glucosidase activity in normal leucocytes and in leucocytes of patients with glycogen-storage disease. Clin Cltim Acta 1964; 9: 269-272. 4 Van Hoof F. Amylo-L6glucosidase activity and glycogen content of erythrocytes of normal subjects, patients with glycogen storage diseases and heterozygotes. Eur J Biochem 1967; 2: 271-274. 5 Justice P, Ryan C, Hsia Dy-Y. Amylo-I,4-~u~sid~e in human fibroblasts: studies in type III glycogen storage disease. Biochem Biophys Res Commun 1970; 39: 301-306. 6 DiMauro S, Rowland LP, Mellman WJ. Glycogen metabolism of human diploid fibroblast cells in culture. I, Studies of cells from patients with glycogenosis types I, III and V. Pediatr Res 1973; 7: 739-744, 7 Besley GTN. Cohen PTW, Fead MJW, Wolstenbolme J. Amylo-1,6-glucosidase activity in cultured cells: a deficiency in type III glycogenosis with prenatal studies. Prenat Diag 1983; 3: 13-19. 8 Galjaard H. Genetic metabolic diseases Amniotic fluid cell cultivation. Amsterdam: Elsevier/North Holland, 1980: 542-546; Appendix II: a-glucosidase p 809-810. 9 Hers HG, Verhue W, Van Hoof F. The determination of amylo-1,6glucosidase. Eur J Biochem 1967; 2: 257-264. 10 Illingworth Brown B, Brown DH, Jeffrey PL. Simultaneous absence of a-1,4glucosidase and a-1,6-glucosidase activities (pH4) in tissues of children with type II glycogen storage disease. Biochem 1970: 9: 1423-1428.