European Journal of Clinical Investigation ( 1 990) 20,470-474
Cholesterol synthesis in patients w i t h glutathione deficiency J. GUSTAFSSON*t, B. CARLSSON* & A. LARSSON*, Departments of Paediatrics* and Pharmaceutical Biochemistry?, University of Uppsala, Uppsala, Sweden
Received 21 February 1989 and in revised form 27 July 1989
Abstract. 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase catalyses the rate-limiting step in cholesterol synthesis. Glutathione (GSH) has been postulated to be an important activator of HMG-CoA reductase in uivo. HMG-CoA reductase activity was assayed in cultured fibroblasts from healthy children. Solubilized enzyme preparations were prepared by ultracentrifugation after freezing and thawing of fibroblasts. Such treatment increased the relative enzyme activity markedly. Enzymological assay conditions were established. Addition of GSH stimulated the reaction, whereas there was inhibition after addition of glutathione disulphide (GSSG). The inhibitory effect of GSSG could be reversed by the addition of excess GSH. Fibroblast preparations, deficient in GSH, were obtained from children with glutathione synthetase deficiency or from normal subjects after the growth of fibroblasts in the presence of buthionine sulphoximine. Solubilized enzyme preparations from GSH-deficient fibroblasts had HMG-CoA reductase activities lower than or comparable with those of control preparations. The results indicate only some reduction in the capacity for cholesterol synthesis in subjects with glutathione deficiency. The existence of additional activation mechanisms in vivo, alternative to GSH, for thiol-dependent modulation of HMG-CoA reductase activity seems likely. Keywords. Fibroblasts, glutathione (GSH), 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Introduction 3-hydroxy-3-methylglutaryl coenzyme A (HMGCoA) reductase catalyses the conversion of HMGCoA into mevalonate [I], a key intermediate in the biosynthesis of cholesterol [I]. The reaction is considered rate-limiting in the synthesis of cholesterol [ 13. Correspondence: J. Gustafsson MD, Ph.D, Department of Paediatrics, University Hospital, S-751 85 Uppsala, Sweden.
Several lines of evidence indicate that the activity of the HMG-CoA reductase can be regulated in different ways. Thus, regulation via liver cell membrane receptors for lipoproteins [2] has been thoroughly characterized. Enzyme phosphorylation/dephosphorylation[3] has also been suggested as a regulatory mechanism. Furthermore, potent effects of oxygenated sterols [4] and certain hormones [ 5 ] on HMG-CoA reductase activity have been described. There is evidence that maintenance of sulphydryl residues in HMG-CoA reductase is necessary for activity [6]. An active form of the enzyme can be prepared from thiol-containing rat liver microsomes [7], whereas there is an inactive form of the enzyme in thiol-depleted preparations [7]. The most likely compound for the activation of HMG-CoA reductase under in viuo conditions is glutathione (GSH) [7]. Decreased levels of GSH may occur as a consequence of inborn errors in glutathione synthesis [8] or in certain physiological and pathological situations [9]. Biochemical studies in patients with glutathione deficiency could provide information about the role of glutathione in biological processes. This paper describes studies on HMG-CoA reductase in fibroblasts from normal subjects as well as patients with glutathione synthetase deficiency. Materials and methods Materials
DL-3-(methyl-3H)-hydroxy-3-methylglutaryl coenzyme A (specific radioactivity, 10.9 Ci mmol-I) was purchased from New England Nuclear (Dreieich, FRG). DL-3-hydroxy-3-methylglutaryl coenzyme A, glutathione (GSH), L-glutamine and NADPH were obtained from Sigma Chemical Co. (St Louis, MO, USA). L-buthionine-(SR)-sulphoximinewas obtained from Chemalog (South Plainfield, NJ, USA). Preparation of jibroblasts
Skin biopsies were obtained from three patients with glutathione synthetase deficiency. The clinical courses of patients (nos 1 and 2), two sisters with glutathione
GLUTATHIONE DEFICIENCY AND CHOLESTEROL synthetase deficiency, have been described earlier [lo]. Both girls had normal serum levels of cholesterol and triglycerides as well as normal serum lipoprotein patterns. Patient no. 3 was initially described by Spielberg et al. [ I I]. His fibroblasts were generously provided by Dr J. D. Schulman. As controls, skin biopsies from three healthy children were used. Informed parental consent was obtained before the biopsies. The study was approved by an ethical committee. Fibroblasts from skin biopsies, taken from the forearm of the patients, were grown as monolayers. The cultures were kept in an humidified incubator with an atmosphere of 5% carbon dioxide at 37°C. The flasks used contained 6-25 ml of Eagle's minimum essential medium (with Earle's salts but without glutamine) supplemented with heat-inactivated foetal calf serum (10~%, v/v) and L-glutamine (2 mmol 1-l). The fibroblasts in each flask were split into three new flasks every 5-7 days. The fibroblasts were used between the fifth and tenth passages. The fibroblasts were not contaminated with mycoplasma when tested according to Chen [12]. The fibroblast preparations were frozen before use in most instances. Prior to incubation, the fibroblasts were thawed and suspended in the incubation buffer. The material was sonicated for 2 x 30 s by use of a Soniprep 150. After centrifugation at 30000 x g for 20 min, followed by centrifugation at 100000xg for 90 min, part of the resulting supernatant containing solubilized HMG-CoA reductase activity was used for incubation. Determination of protein was performed as described by Kalb & Bernlohr [I31 using a Beckman spectrophotometer, model DU 50.
Incubation procedures and analyses of incubation mixtures
The incubations were performed at 37°C for 20-30 min in a total volume of 150 p1 of 50 mmol 1-' Tris-CI buffer, pH 7.4. All incubation mixtures contained 150 nmol 'H-HMG-CoA, 2 Ci mol-I, and 0.45 pmol of NADPH. Before addition of HMG-CoA, the mixture was preincubated for 10 min at 37°C. In standard incubations 100 000 x g supernatant fluid, corresponding to about 0.2 mg of protein, prepared as described above, was used. The concentration of glutathione was 20 mmol I - ' in standard incubations. The incubations were terminated by the addition of 25 pl of 10 mol I - ' HCI and then incubated for 30 min at 37°C in order to complete lactonization of the mevalonic acid formed during the incubations. Parts of the incubation mixtures were analysed for radioactivity by the mixedphase assay as described by Philipp & Shapiro [14]. This assay involves addition of an acidified, deproteinized incubation mixture to a scintillation medium based on toluene [14]. The product in the incubation, 3H-mevalonolactone, partitions into the toluene whereas the 3H-HMG-CoA remains in the water phase [14]. The radioactivity of this material will thus not be counted in liquid scintillation counting. The equipment used for radioactivity counting was an LKB 1219 Rackbeta instrument. Efficiency for 3H with the settings used was 63.5%. Determination of glutathione
Glutathione was determined as described by Tietze ~51.
80 60
40 20
0 Amount of protein ( p g )
0
0.5
1.0
HMGCOA (mmol
1-9
47 1
1.5 0
10
20
GSH (rnmo11-I)
Figure 1. Effect of time (a), amount o f protein (100000 x g supernatant fluid) (b), concentration of substrate (c) and GSH (d) on HMG-CoA reductase activity in a solubilized preparation from control fibroblasts. Standard incubation conditions were used.
J. GUSTAFSSON, B. CARLSSON & A. LARSSON
472
Results Establishment of enzymological conditions Optimal conditions for assay of human fibroblast HMG-CoA reductase activity were established. It was shown that the use of a 30000xg supernatant of a sonicated fibroblast suspension increased the activity per mg of protein compared with the sonicated fibroblast suspension. Freezing of the fibroblasts at -20 or - 196°C and thawing before sonication and centrifugation, further increased the activity. Repeated freezing and thawing up to three times gave no additional increase but rather decreased the enzyme activity. Centrifugation at 100000 x g of a 30 000 x g supernatant, prepared from fibroblasts after freezing, thawing and sonication, yielded a solubilized preparation with the highest activity per mg of protein. About 90% of the total HMG-CoA reductase activity was solubilized and found in the 100 000 x g supernatant fraction. About 10 per cent remained in the 100 000 x g pellet. Figure la, b shows that under standard conditions the rate of formation of mevalonic acid in a 100000 x g supernatant fluid from a solubilized control preparation was linear with time for 30 min and with the amount of protein up to at least 0.25 mg. Substrate saturation was obtained at a concentration of HMGCoA of 1 mmol I-' (Fig. Ic). The addition of potassium chloride up to a final concentration of 300 mmol I - ' did not influence the activity (not shown).
order to remove endogenous GSH did not significantly change the degree of mevalonate formation when incubations were performed without the addition of GSH. Chromatography of a 30000 x g supernatant on Sepharose 4B did not yield any fraction containing HMG-CoA reductase activity that was absolutely dependent upon the addition of thiols (cf. [7]). Eflect of glutathione disulphide (GSSG) on HMGCoA reductase activity In Fig. 2 the effect on HMG-CoA reductase activity of addition of GSSG can be seen. 40 mmol I - ' GSSG inhibited the activity by about 80%. Addition of 40 mmol 1-' GSH restored the activity. Incubations offibroblast preparations from patients with glutathione synthetase deficiency Figure 3 shows the HMG-CoA reductase activity in solubilized preparations from the three patients (JG, H G and AR) with glutathione synthetase deficiency and three controls assayed in absence or presence of 20 mmol 1-' GSH. The activities of the preparations of two of the patients were about 50% of those in the controls, whereas that of the third patient was comparable with the activities of the controls. Addition of GSH stimulated the HMG-CoA reductase activities of control preparations by 30-80%, whereas those of the patients were stimulated by 100-200%. Eflect of buthionine sulphoximine ( B S O ) on HMGCoA reductase activity
Effect of GSH on HMG-CoA reductase activity Addition of 10 mmol 1-' GSH stimulated the reaction by about 60% and saturated the system (Fig. Id). The degree of stimulation varied between different preparations from 20-100%. Dialysis of preparations in
As can be seen in Fig. 4, HMG-CoA reductase activity in solubilized preparations from fibroblasts, grown up to 24 h in a medium containing I mmol I - ' BSO, was about 90% of that in corresponding control prep-
Cl
O
E
100
ti e i i P I
I
I
I
I
I
I
I
I
I I
1
I I
I
I
I I
I
I I
I I I
C
0
2
J
0
u
2
O
70
5.3
10
O
I
20
I
I
f 40
GSSG (mrno11-9
in a Figure 2. Effect of GSSG on HMG-CoA reductase activity (0) solubilized preparation (100000 x g supernatant fluid) from control fibroblasts. Except for the omission of GSH, standard incubation conditions were used. The effect of supplementation with 40 mmol I-' GSH in addition to GSSG, is also shown ( 0 ) .
E 0
200
-iz 5
100
!!
W >
5
0
0
III
III
I
Control.-i 2
-I
III 3
III +Patient I
IJI
III I-
2
3
Figure 3. HMG-CoA reductase activity in solubilized preparations (l00OOO x g supernatant fluid) from controls and patients with glutathione synthetase deficiency (patient I (JG), 2 (HG) and 3 (AR)). I, No addition of GSH; 11, addition of 20 mmol I - ' GSH. Standard incubation conditions were used.
GLUTATHIONE DEFICIENCY AND CHOLESTEROL 500
r
300
-
I
0
-E 200 C
0 in
$
loo-
0 C
0
00
12
24
Hours
Figure 4. HMG-CoA reductase activity (-) and GSH-content (- -)in solubilized preparations (100000 x g supernatant fluid) from control fibroblasts, grown up to 24 h in a medium containing I mmol I - ' BSO ( 0 ) .or in a medium without BSO ( 0 ) .
arations. In the control preparations, BSO was added just prior to incubation. The endogenous glutathione content was reduced to almost undetectable levels after 24 h of BSO treatment.
Discussion Liver cells, as well as extrahepatic cells, obtain cholesterol by synthesis de nouo or uptake via receptors for lipoproteins [ 161. There is evidence for requirement of thiol-group protecting agents in the assay of HMGCoA reductase from different sources [6]. In accordance with earlier work [17] on human fibroblasts the present values indicate dependence of HMG-CoA reductase activity on thiols. Thus, sensitivity of the enzyme activity to inhibition by GSSG followed by reactivation by GSH was shown. Similarly to what has been demonstrated with liver microsomal HMG-CoA reductase [ 181, the increase in enzymatic activity after freezing and thawing probably reflects solubilization of the whole enzyme or the active part of the enzyme. The occurrence of enzyme solubilization was clearly shown by the finding of the major part of the HMG-CoA reductase activity in the 100000 x g supernatant fluid. In the present study, there was considerable variation in the HMG-CoA reductase activity, even when optimal amounts of glutathione was present. There was no evidence indicating a relationship between the number of cell passages and the HMG-CoA reductase activity. The relatively great number of factors taking part in the regulation of the HMG-CoA reductase activity may explain the variation seen between different experiments. The present work examines the relationship between in vioo and in vitro glutathione status and in uitro enzyme activity. Although GSH had a stimulatory effect on fibroblast HMG-CoA reductase activity in oitro, there was no absolute requirement for GSH.
473
Thus GSH-depleted preparations, regardless if prepared from subjects with inborn errors in glutathione synthesis or by addition of BSO to fibroblast cultures from normal subjects [19], did not have enzyme activities below 50% of the activities of control preparations. In addition, similar results were obtained with control preparations subjected to dialysis or gel filtration in order to achieve a state of depletion of thiols. The results thus indicate only some reduction in the capacity for cholesterol synthesis in subjects with congenital glutathione deficiency. Recently, a regulatory role for glutathione in liver microsomal cholesterol 7a!-hydroxylation, rate-limiting reaction in bile acid formation from cholesterol, has been suggested [20]. Thus as a whole, some decrease in cholesterol synthesis under conditions of glutathione deficiency could be encountered by a decreased rate of cholesterol metabolism. There is evidence that changes in content of reduced glutathione may in fact be a link in feeding-related regulation of cholesterol synthesis and metabolism [20]. In light of the results in this paper, the existence of additional mechanisms for thiol-dependent modulation of HMG-CoA reductase activity seems likely. Such mechanisms could involve the thioredoxin system [21] or a protein corresponding to the rat liver HMG-CoA reductase activating protein (RAP), recently described by Dotan & Shechter [22].
Acknowledgments Supported by the Swedish Medical Research Council (projects 4792 and 7514).
References I Sabine JR. General distribution and importance of HMGCoA reductase. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:3-10. 2 Edwards PA, Fogelman AM, Tanaka RD. Physiological control of HMGCoA reductase activity. In: Sabine JR, ed. 3-Hydroxy-3Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:93-105. 3 Ingebritsen TS. Molecular control of HMGCoA reductase: regulation by phosphorylation/dephosphorylation. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:129-52. 4 Gibbons GF. Molecular control of 3-hydroxy-3-methyl-glutaryl coenzyme A reductase: the role of oxygenated sterols. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:153-68. 5 Dugan RE. Regulation of HMG-CoA reductase. In: Porter JW & Spurgeon SL,eds. Biosynthesis of Isoprenoid Compounds, vol. 1. New York: John Wiley & Sons. 1981:95-159. 6 Roitelman J, Schechter I. Regulation of rat liver 3-hydroxy-3methylglutaryl coenzyme A reductase. Evidence for thiol-dependent allosteric modulation of enzyme activity. J Biol Chem 1984;259:870-7. 7 Dotan I, Schechter 1. Thiol-disulfide-dependentinterconversion of active and latent forms of rat hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase. Biochim Biophys Acta 1982;713~427-34. 8 Meister A, Anderson ME. Glutathione. Ann Rev Biochem 1983;52:711-60.
474
J. GUSTAFSSON, B. CARLSSON & A. LARSSON
9 Meister A. New aspects of glutathione biochemistry and transport: selective alteration of glutathione metabolism. Fed Proc 1984;43:3031-42, 10 Larsson A, Wachtmeister L, von Wendt L, Andersson R, Hagenfeldt L. Ophthalmological, psychometric and therapeutic investigation in two sisters with hereditary glutathione synthetase deficiency (5-oxoprolinuria). Neuropediatrics 1985;16: 131-6. 11 Spielberg SP, Kramer LI, Goodman SI, Butler J, Tietze F, Quinn P, Schulman JD. 5-Oxoprolinuria: biochemical observations and case report. J Pediatrics 1977;91:23741. 12 Chen TR. In siru detection of mycoplasma contamination in cell cultures by fluorescent Hoechst 33258 stain. Exptl Cell Res 1971;104:255-62. 13 Kalb V F Jr, Bernlohr RW. A new spectrophotometric assay for protein in cell extracts. Anal Biochem 1977;82:362-7 I . 14 Philipp BW, Shapiro DJ. Improved methods for the assay and activation of 3-hydroxy-3-methylglutarylcoenzyme A reductase. J Lipid Res 1979;20:588-93. 15 Tietze F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione. Application to mammalian blood and other tissues. Analyt Biochem I969;27:502-22.
16 Brown MS, Goldstein JL. Lipoprotein receptors in the liver: control signals for plasma cholesterol traffic. J Clin Invest 1983;72:743-7. 17 Georgopapadakou NH, Dix BA. A modified radiometric assay for 3-hydroxy-3-methylglutarylcoenzyme A reductase. Lipids 1984;I9:966-70. 18 Edwards PA, Lemongello D, Fogelman AM. Improved methods for the solubilization and assay of hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase. J Lipid Res 1979;20:40-6. 19 Griffith OW, Meister A. Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). J Biol Chem 1979;254:7558-60. 20 Hassan A H . Feeding-induced regulation of cholesterol metabolism: A unified proposal. Proc SOCExp Biol Med 1986;182:14350.
21 Holmgren A. Thioredoxin: structure and functions. Trends in Biochemical Sciences 1981;6:26-9. 22 Dotan I, Shechter I. Isolation and purification of a rat liver 3hydroxy-3-methylglutarylcoenzymeA reductase activating protein (RAP). J Biol Chem 1987;262:17058-64.
Cholesterol synthesis in patients w i t h glutathione deficiency J. GUSTAFSSON*t, B. CARLSSON* & A. LARSSON*, Departments of Paediatrics* and Pharmaceutical Biochemistry?, University of Uppsala, Uppsala, Sweden
Received 21 February 1989 and in revised form 27 July 1989
Abstract. 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase catalyses the rate-limiting step in cholesterol synthesis. Glutathione (GSH) has been postulated to be an important activator of HMG-CoA reductase in uivo. HMG-CoA reductase activity was assayed in cultured fibroblasts from healthy children. Solubilized enzyme preparations were prepared by ultracentrifugation after freezing and thawing of fibroblasts. Such treatment increased the relative enzyme activity markedly. Enzymological assay conditions were established. Addition of GSH stimulated the reaction, whereas there was inhibition after addition of glutathione disulphide (GSSG). The inhibitory effect of GSSG could be reversed by the addition of excess GSH. Fibroblast preparations, deficient in GSH, were obtained from children with glutathione synthetase deficiency or from normal subjects after the growth of fibroblasts in the presence of buthionine sulphoximine. Solubilized enzyme preparations from GSH-deficient fibroblasts had HMG-CoA reductase activities lower than or comparable with those of control preparations. The results indicate only some reduction in the capacity for cholesterol synthesis in subjects with glutathione deficiency. The existence of additional activation mechanisms in vivo, alternative to GSH, for thiol-dependent modulation of HMG-CoA reductase activity seems likely. Keywords. Fibroblasts, glutathione (GSH), 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Introduction 3-hydroxy-3-methylglutaryl coenzyme A (HMGCoA) reductase catalyses the conversion of HMGCoA into mevalonate [I], a key intermediate in the biosynthesis of cholesterol [I]. The reaction is considered rate-limiting in the synthesis of cholesterol [ 13. Correspondence: J. Gustafsson MD, Ph.D, Department of Paediatrics, University Hospital, S-751 85 Uppsala, Sweden.
Several lines of evidence indicate that the activity of the HMG-CoA reductase can be regulated in different ways. Thus, regulation via liver cell membrane receptors for lipoproteins [2] has been thoroughly characterized. Enzyme phosphorylation/dephosphorylation[3] has also been suggested as a regulatory mechanism. Furthermore, potent effects of oxygenated sterols [4] and certain hormones [ 5 ] on HMG-CoA reductase activity have been described. There is evidence that maintenance of sulphydryl residues in HMG-CoA reductase is necessary for activity [6]. An active form of the enzyme can be prepared from thiol-containing rat liver microsomes [7], whereas there is an inactive form of the enzyme in thiol-depleted preparations [7]. The most likely compound for the activation of HMG-CoA reductase under in viuo conditions is glutathione (GSH) [7]. Decreased levels of GSH may occur as a consequence of inborn errors in glutathione synthesis [8] or in certain physiological and pathological situations [9]. Biochemical studies in patients with glutathione deficiency could provide information about the role of glutathione in biological processes. This paper describes studies on HMG-CoA reductase in fibroblasts from normal subjects as well as patients with glutathione synthetase deficiency. Materials and methods Materials
DL-3-(methyl-3H)-hydroxy-3-methylglutaryl coenzyme A (specific radioactivity, 10.9 Ci mmol-I) was purchased from New England Nuclear (Dreieich, FRG). DL-3-hydroxy-3-methylglutaryl coenzyme A, glutathione (GSH), L-glutamine and NADPH were obtained from Sigma Chemical Co. (St Louis, MO, USA). L-buthionine-(SR)-sulphoximinewas obtained from Chemalog (South Plainfield, NJ, USA). Preparation of jibroblasts
Skin biopsies were obtained from three patients with glutathione synthetase deficiency. The clinical courses of patients (nos 1 and 2), two sisters with glutathione
GLUTATHIONE DEFICIENCY AND CHOLESTEROL synthetase deficiency, have been described earlier [lo]. Both girls had normal serum levels of cholesterol and triglycerides as well as normal serum lipoprotein patterns. Patient no. 3 was initially described by Spielberg et al. [ I I]. His fibroblasts were generously provided by Dr J. D. Schulman. As controls, skin biopsies from three healthy children were used. Informed parental consent was obtained before the biopsies. The study was approved by an ethical committee. Fibroblasts from skin biopsies, taken from the forearm of the patients, were grown as monolayers. The cultures were kept in an humidified incubator with an atmosphere of 5% carbon dioxide at 37°C. The flasks used contained 6-25 ml of Eagle's minimum essential medium (with Earle's salts but without glutamine) supplemented with heat-inactivated foetal calf serum (10~%, v/v) and L-glutamine (2 mmol 1-l). The fibroblasts in each flask were split into three new flasks every 5-7 days. The fibroblasts were used between the fifth and tenth passages. The fibroblasts were not contaminated with mycoplasma when tested according to Chen [12]. The fibroblast preparations were frozen before use in most instances. Prior to incubation, the fibroblasts were thawed and suspended in the incubation buffer. The material was sonicated for 2 x 30 s by use of a Soniprep 150. After centrifugation at 30000 x g for 20 min, followed by centrifugation at 100000xg for 90 min, part of the resulting supernatant containing solubilized HMG-CoA reductase activity was used for incubation. Determination of protein was performed as described by Kalb & Bernlohr [I31 using a Beckman spectrophotometer, model DU 50.
Incubation procedures and analyses of incubation mixtures
The incubations were performed at 37°C for 20-30 min in a total volume of 150 p1 of 50 mmol 1-' Tris-CI buffer, pH 7.4. All incubation mixtures contained 150 nmol 'H-HMG-CoA, 2 Ci mol-I, and 0.45 pmol of NADPH. Before addition of HMG-CoA, the mixture was preincubated for 10 min at 37°C. In standard incubations 100 000 x g supernatant fluid, corresponding to about 0.2 mg of protein, prepared as described above, was used. The concentration of glutathione was 20 mmol I - ' in standard incubations. The incubations were terminated by the addition of 25 pl of 10 mol I - ' HCI and then incubated for 30 min at 37°C in order to complete lactonization of the mevalonic acid formed during the incubations. Parts of the incubation mixtures were analysed for radioactivity by the mixedphase assay as described by Philipp & Shapiro [14]. This assay involves addition of an acidified, deproteinized incubation mixture to a scintillation medium based on toluene [14]. The product in the incubation, 3H-mevalonolactone, partitions into the toluene whereas the 3H-HMG-CoA remains in the water phase [14]. The radioactivity of this material will thus not be counted in liquid scintillation counting. The equipment used for radioactivity counting was an LKB 1219 Rackbeta instrument. Efficiency for 3H with the settings used was 63.5%. Determination of glutathione
Glutathione was determined as described by Tietze ~51.
80 60
40 20
0 Amount of protein ( p g )
0
0.5
1.0
HMGCOA (mmol
1-9
47 1
1.5 0
10
20
GSH (rnmo11-I)
Figure 1. Effect of time (a), amount o f protein (100000 x g supernatant fluid) (b), concentration of substrate (c) and GSH (d) on HMG-CoA reductase activity in a solubilized preparation from control fibroblasts. Standard incubation conditions were used.
J. GUSTAFSSON, B. CARLSSON & A. LARSSON
472
Results Establishment of enzymological conditions Optimal conditions for assay of human fibroblast HMG-CoA reductase activity were established. It was shown that the use of a 30000xg supernatant of a sonicated fibroblast suspension increased the activity per mg of protein compared with the sonicated fibroblast suspension. Freezing of the fibroblasts at -20 or - 196°C and thawing before sonication and centrifugation, further increased the activity. Repeated freezing and thawing up to three times gave no additional increase but rather decreased the enzyme activity. Centrifugation at 100000 x g of a 30 000 x g supernatant, prepared from fibroblasts after freezing, thawing and sonication, yielded a solubilized preparation with the highest activity per mg of protein. About 90% of the total HMG-CoA reductase activity was solubilized and found in the 100 000 x g supernatant fraction. About 10 per cent remained in the 100 000 x g pellet. Figure la, b shows that under standard conditions the rate of formation of mevalonic acid in a 100000 x g supernatant fluid from a solubilized control preparation was linear with time for 30 min and with the amount of protein up to at least 0.25 mg. Substrate saturation was obtained at a concentration of HMGCoA of 1 mmol I-' (Fig. Ic). The addition of potassium chloride up to a final concentration of 300 mmol I - ' did not influence the activity (not shown).
order to remove endogenous GSH did not significantly change the degree of mevalonate formation when incubations were performed without the addition of GSH. Chromatography of a 30000 x g supernatant on Sepharose 4B did not yield any fraction containing HMG-CoA reductase activity that was absolutely dependent upon the addition of thiols (cf. [7]). Eflect of glutathione disulphide (GSSG) on HMGCoA reductase activity In Fig. 2 the effect on HMG-CoA reductase activity of addition of GSSG can be seen. 40 mmol I - ' GSSG inhibited the activity by about 80%. Addition of 40 mmol 1-' GSH restored the activity. Incubations offibroblast preparations from patients with glutathione synthetase deficiency Figure 3 shows the HMG-CoA reductase activity in solubilized preparations from the three patients (JG, H G and AR) with glutathione synthetase deficiency and three controls assayed in absence or presence of 20 mmol 1-' GSH. The activities of the preparations of two of the patients were about 50% of those in the controls, whereas that of the third patient was comparable with the activities of the controls. Addition of GSH stimulated the HMG-CoA reductase activities of control preparations by 30-80%, whereas those of the patients were stimulated by 100-200%. Eflect of buthionine sulphoximine ( B S O ) on HMGCoA reductase activity
Effect of GSH on HMG-CoA reductase activity Addition of 10 mmol 1-' GSH stimulated the reaction by about 60% and saturated the system (Fig. Id). The degree of stimulation varied between different preparations from 20-100%. Dialysis of preparations in
As can be seen in Fig. 4, HMG-CoA reductase activity in solubilized preparations from fibroblasts, grown up to 24 h in a medium containing I mmol I - ' BSO, was about 90% of that in corresponding control prep-
Cl
O
E
100
ti e i i P I
I
I
I
I
I
I
I
I
I I
1
I I
I
I
I I
I
I I
I I I
C
0
2
J
0
u
2
O
70
5.3
10
O
I
20
I
I
f 40
GSSG (mrno11-9
in a Figure 2. Effect of GSSG on HMG-CoA reductase activity (0) solubilized preparation (100000 x g supernatant fluid) from control fibroblasts. Except for the omission of GSH, standard incubation conditions were used. The effect of supplementation with 40 mmol I-' GSH in addition to GSSG, is also shown ( 0 ) .
E 0
200
-iz 5
100
!!
W >
5
0
0
III
III
I
Control.-i 2
-I
III 3
III +Patient I
IJI
III I-
2
3
Figure 3. HMG-CoA reductase activity in solubilized preparations (l00OOO x g supernatant fluid) from controls and patients with glutathione synthetase deficiency (patient I (JG), 2 (HG) and 3 (AR)). I, No addition of GSH; 11, addition of 20 mmol I - ' GSH. Standard incubation conditions were used.
GLUTATHIONE DEFICIENCY AND CHOLESTEROL 500
r
300
-
I
0
-E 200 C
0 in
$
loo-
0 C
0
00
12
24
Hours
Figure 4. HMG-CoA reductase activity (-) and GSH-content (- -)in solubilized preparations (100000 x g supernatant fluid) from control fibroblasts, grown up to 24 h in a medium containing I mmol I - ' BSO ( 0 ) .or in a medium without BSO ( 0 ) .
arations. In the control preparations, BSO was added just prior to incubation. The endogenous glutathione content was reduced to almost undetectable levels after 24 h of BSO treatment.
Discussion Liver cells, as well as extrahepatic cells, obtain cholesterol by synthesis de nouo or uptake via receptors for lipoproteins [ 161. There is evidence for requirement of thiol-group protecting agents in the assay of HMGCoA reductase from different sources [6]. In accordance with earlier work [17] on human fibroblasts the present values indicate dependence of HMG-CoA reductase activity on thiols. Thus, sensitivity of the enzyme activity to inhibition by GSSG followed by reactivation by GSH was shown. Similarly to what has been demonstrated with liver microsomal HMG-CoA reductase [ 181, the increase in enzymatic activity after freezing and thawing probably reflects solubilization of the whole enzyme or the active part of the enzyme. The occurrence of enzyme solubilization was clearly shown by the finding of the major part of the HMG-CoA reductase activity in the 100000 x g supernatant fluid. In the present study, there was considerable variation in the HMG-CoA reductase activity, even when optimal amounts of glutathione was present. There was no evidence indicating a relationship between the number of cell passages and the HMG-CoA reductase activity. The relatively great number of factors taking part in the regulation of the HMG-CoA reductase activity may explain the variation seen between different experiments. The present work examines the relationship between in vioo and in vitro glutathione status and in uitro enzyme activity. Although GSH had a stimulatory effect on fibroblast HMG-CoA reductase activity in oitro, there was no absolute requirement for GSH.
473
Thus GSH-depleted preparations, regardless if prepared from subjects with inborn errors in glutathione synthesis or by addition of BSO to fibroblast cultures from normal subjects [19], did not have enzyme activities below 50% of the activities of control preparations. In addition, similar results were obtained with control preparations subjected to dialysis or gel filtration in order to achieve a state of depletion of thiols. The results thus indicate only some reduction in the capacity for cholesterol synthesis in subjects with congenital glutathione deficiency. Recently, a regulatory role for glutathione in liver microsomal cholesterol 7a!-hydroxylation, rate-limiting reaction in bile acid formation from cholesterol, has been suggested [20]. Thus as a whole, some decrease in cholesterol synthesis under conditions of glutathione deficiency could be encountered by a decreased rate of cholesterol metabolism. There is evidence that changes in content of reduced glutathione may in fact be a link in feeding-related regulation of cholesterol synthesis and metabolism [20]. In light of the results in this paper, the existence of additional mechanisms for thiol-dependent modulation of HMG-CoA reductase activity seems likely. Such mechanisms could involve the thioredoxin system [21] or a protein corresponding to the rat liver HMG-CoA reductase activating protein (RAP), recently described by Dotan & Shechter [22].
Acknowledgments Supported by the Swedish Medical Research Council (projects 4792 and 7514).
References I Sabine JR. General distribution and importance of HMGCoA reductase. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:3-10. 2 Edwards PA, Fogelman AM, Tanaka RD. Physiological control of HMGCoA reductase activity. In: Sabine JR, ed. 3-Hydroxy-3Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:93-105. 3 Ingebritsen TS. Molecular control of HMGCoA reductase: regulation by phosphorylation/dephosphorylation. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:129-52. 4 Gibbons GF. Molecular control of 3-hydroxy-3-methyl-glutaryl coenzyme A reductase: the role of oxygenated sterols. In: Sabine JR, ed. 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase. Boca Raton, Florida: CRC Press Inc. 1983:153-68. 5 Dugan RE. Regulation of HMG-CoA reductase. In: Porter JW & Spurgeon SL,eds. Biosynthesis of Isoprenoid Compounds, vol. 1. New York: John Wiley & Sons. 1981:95-159. 6 Roitelman J, Schechter I. Regulation of rat liver 3-hydroxy-3methylglutaryl coenzyme A reductase. Evidence for thiol-dependent allosteric modulation of enzyme activity. J Biol Chem 1984;259:870-7. 7 Dotan I, Schechter 1. Thiol-disulfide-dependentinterconversion of active and latent forms of rat hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase. Biochim Biophys Acta 1982;713~427-34. 8 Meister A, Anderson ME. Glutathione. Ann Rev Biochem 1983;52:711-60.
474
J. GUSTAFSSON, B. CARLSSON & A. LARSSON
9 Meister A. New aspects of glutathione biochemistry and transport: selective alteration of glutathione metabolism. Fed Proc 1984;43:3031-42, 10 Larsson A, Wachtmeister L, von Wendt L, Andersson R, Hagenfeldt L. Ophthalmological, psychometric and therapeutic investigation in two sisters with hereditary glutathione synthetase deficiency (5-oxoprolinuria). Neuropediatrics 1985;16: 131-6. 11 Spielberg SP, Kramer LI, Goodman SI, Butler J, Tietze F, Quinn P, Schulman JD. 5-Oxoprolinuria: biochemical observations and case report. J Pediatrics 1977;91:23741. 12 Chen TR. In siru detection of mycoplasma contamination in cell cultures by fluorescent Hoechst 33258 stain. Exptl Cell Res 1971;104:255-62. 13 Kalb V F Jr, Bernlohr RW. A new spectrophotometric assay for protein in cell extracts. Anal Biochem 1977;82:362-7 I . 14 Philipp BW, Shapiro DJ. Improved methods for the assay and activation of 3-hydroxy-3-methylglutarylcoenzyme A reductase. J Lipid Res 1979;20:588-93. 15 Tietze F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione. Application to mammalian blood and other tissues. Analyt Biochem I969;27:502-22.
16 Brown MS, Goldstein JL. Lipoprotein receptors in the liver: control signals for plasma cholesterol traffic. J Clin Invest 1983;72:743-7. 17 Georgopapadakou NH, Dix BA. A modified radiometric assay for 3-hydroxy-3-methylglutarylcoenzyme A reductase. Lipids 1984;I9:966-70. 18 Edwards PA, Lemongello D, Fogelman AM. Improved methods for the solubilization and assay of hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase. J Lipid Res 1979;20:40-6. 19 Griffith OW, Meister A. Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). J Biol Chem 1979;254:7558-60. 20 Hassan A H . Feeding-induced regulation of cholesterol metabolism: A unified proposal. Proc SOCExp Biol Med 1986;182:14350.
21 Holmgren A. Thioredoxin: structure and functions. Trends in Biochemical Sciences 1981;6:26-9. 22 Dotan I, Shechter I. Isolation and purification of a rat liver 3hydroxy-3-methylglutarylcoenzymeA reductase activating protein (RAP). J Biol Chem 1987;262:17058-64.
