Chem.-Biol. Interactions, 81 (1992) 209-218 Elsevier Scientific Publishers Ireland Ltd.
209
INHIBITION OF MICROSOMAL GLUCOSE 6-PHOSPHATASE BY UNSATURATED ALIPHATIC A L D E H Y D E S AND KETONES
BO M. JORGENSEN a, PIA AGERBO a, BENNY J E N S E N b, TORGER BORRESEN ~, and GUNHILD HOLMER c
aCenter for Food Research at DTH, bTechnological Laboratory, Danish Ministry of Fisheries and CDepartment of Biochemistry and Nutrition, Technical University Build. 221 "'~' and 224", DK-2800 Lyngby (Denmark) (Received August 5th, 1991) (Revision received November 6th, 1991) (Accepted November 8th, 1991)
SUMMARY
Aldehydes and ketones with one double bond conjugated to the carbonyl group inhibited the enzyme glucose 6-phosphatase, which is embedded in the microsomal membrane. The Michaelis constant, Km and the maximal rate of reaction, V, were affected in a way dependent on the inhibitor's chain-length: trans-2-pentenal and 1-penten-3-one increased Km linearly with concentration and had almost no effect on V, whereas trans-2-nonenal caused a large increase in V but only a small and non-linear change in Km. The effect of the short-chain aldehydes on the kinetic parameters increased with chain-length, but pentenone increased Km more than did trans-2-heptenal and conjugated dienals did not act as inhibitors. Therefore, sterical effects apparently are of importance. Washing the microsomes after incubation with hexenal or heptenal did not substantially decrease the inhibition, but with nonenal the inhibition was reduced by washing. Inhibition by the SH-group blocking reagent p-hydroxymercuribenzoate was competitive to inhibition by the alkenals. It is concluded that the a-/3 unsaturated oxo-compounds inhibit glucose 6-phosphatase by binding covalently to an important mercapto group and that perturbation of the enzyme's membrane environment also plays a part in the inhibition.
Key words: Alkenals -- Autoxidation -- Membrane perturbation -- Mercapto group -- p-Hydroxymercuribenzoate Correspondence to: Bo JCrgensen, DTH Build. 221 (FF), DK-2800 Lyngby, Denmark. Abbreviations: EDTA, ethylenediaminetetraacetic acid; f, speed of rotation; [I], inhibitor concentration; Ki(Km) and Ki(V), apparent inhibitor constant; Kin, Michaelis constant; PMB, p-hydroxymercuribenzoate; rmin and rmax, minimal and maximal rotation radii (centrifugal data); So, initial substrate concentration; v, initial rate of reaction; V, maximal rate of reaction. 0009-2797/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
210 INTRODUCTION
Glucose 6-phosphatase (EC 3.1.3.9) of mammalian liver is an enzyme located in the endoplasmic reticulum. Its properties are heavily dependent on the membrane integrity and even small perturbations of the environment, e.g. with detergent [1], may drastically alter the kinetic parameters, specificity and stability [1-6]. The enzyme is inhibited by common SH-group binding reagents [7-9] with the predominant effect being a rise in Km (glucose-6-P). But the mercapto group involved is apparently not situated at the active site as the reagent phydroxymercuribenzoate (PMB) binds without inhibition at 0°C [9]. Induced lipid autoxidation or addition of some of the secondary oxidation products also inhibit the enzyme [10- 16]. It has been shown [16 - 18; Agerbo et al., unpublished data] that among the variety of products formed, the active ones are aft-unsaturated aldehydes (e.g. 4-hydroxy-2-nonenal) and ketones, which are believed to bind to SH-groups in proteins by a conjugate nucleophilic addition to the carbon-carbon double bond [10,16-18]. Thus, reactions of that type may be suggested as the cause of inhibition of glucose 6-phosphatase. In order to further explore the mechanism, we have measured the influence of a series of unsaturated oxo-compounds on the kinetic parameters, Michaelis constant and maximal rate of reaction. Short-chain compounds increased Km linearly with concentration and had only minor effects on V, whereas longer-chain alkenals decreased V markedly and exhibited a non-linear Km relationship with inhibitor concentration. Inhibition was in both cases competitive to inhibition by PMB and enzymatic activity was only partly recovered by washing the microsomes with buffer. It is concluded that inhibition involves at least two distinct mechanisms: a binding to an 'essential' SH-group and a perturbation of the membrane lipid phase. The second effect is more pronounced with the longerchain aldehydes. MATERIALS AND METHODS
Reagents 1-Penten-3-one, 2-pentenal, 2-hexenal, 2-heptenal, 2-octenal, 2-nonenal, 2,4-hexadienal, 2,4-heptadienal and 2,4-decadienal -- all with trans-configuration around the C = C double bonds -- were purchased from Aldrich-Chemie, Germany, in the highest grade available (typically 9 3 - 99%) and used without further purification. Glucose 6-phosphate (analytical grade) and a crude preparation of partially purified glucose 6-phosphatase were obtained from Sigma Chemical Company, USA. Buffer reagents were of the highest quality commercially available.
Preparation of microsomes The liver from a laboratory rat that had been fed a pellet diet was homogenized in a medium containing 250 mM sucrose, 1 mM Tris and 0.1 mM EDTA, pH = 7.0, in a Potter-Elvehjem homogenizer with a loose-fitting pestle. The homogenate was centrifuged (4°C, 15 min, f = 11 000 min -1, rmin and
211
rmax = 5.7 and 10.7 cm) in a Sorvall RC-5B centrifuge equipped with an SS-34 rotor. The supernatant was further centrifuged (4°C, 65 rain, f = 35 000 min- 1, rmin and rmax = 3.95 and 9.19 cm) in a Beckman L8-70M ultracentrifuge (70 TI rotor) and the resultant pellet was resuspended in incubation medium (50 mM Tris, 50 mM maleate, 150 mM KC1; pH = 7.4) to a volume of 1 ml per g liver, making the protein concentration to be around 15 - 20 mg/ml. The preparation could be stored at - 3 0 ° C for more than a month without measurable loss of glucose 6-phosphatase activity.
Incubation with inhibitor Stock solutions (1 - 2 mM) of inhibitor in incubation medium were prepared in 100-ml measuring flasks. Four or five volumes of inhibitor, diluted with medium to the appropriate concentration, were mixed with one volume of microsomes and gently shaken at 37°C in sealed bottles.
Enzyme assay Assay buffer (500 td, 50 mM maleate, 4.6 mM EDTA, 2.3 mM NaF; pH = 6.0) were mixed with 100 ~l 200 mM glucose 6-phosphate (pH = 6.0) in a 15-ml polycarbonate tube (Sorvall). The enzymatic reaction (at 37°C) was started by adding 100 #l incubation mixture and terminated after an appropriate time (usually 20 min) by addition of 1.3 ml 10% (w/v) trichloroacetic acid. The precipitated protein was settled by centrifugation at f = 12 000 rain-1 for 10 min in a Sorvall SM-24 rotor (4°C) and 1000 ~l was taken to phosphate analysis. Blank assays were performed by addition of acid prior to enzyme.
Determination of phosphate The amount of phosphate formed in the assay was determined essentially as described by Fiske and Subbarow [19]. Sample (1000 #l) was mixed with 2000 ~l water, 1000 ~l 'molybdate reagent' (1.36 vol. conc. sulfuric acid, 3 vol. water, 5 vol. 50 g/1 hexaammonium molybdate; diluted to 10 vol with water) and 1000 ~l 'reducing reagent' (30 g/1 sodium disulfite, 10 g/1 4-(methylamino)-phenol sulfate (Photo-Rex ® ; Merck, Germany)). Absorbance at 660 nm was read after at least 40 min.
Determination of protein The protein content of the microsomal preparations was estimated by the Folin-Ciocalteu reagent [20] and with bovine serum albumin as standard. RESULTS
AND DISCUSSION
Glucose 6-phosphatase followed simple Michaelis-Menten kinetics with glucose 6-phosphate as substrate in the absence as well as in the presence of inhibitors. The kinetic constants, Km and V, were calculated from Hanes plots (So/Vversus So) [21]. 'Secondary plots', Km and 1/V versus [I], were apparently linear in the 0-1 mM range when pentenone, pentenal, hexenal or heptenal were used as inhibitor (Figs. 1 and 2). The inhibitor concentrations needed for a doubling of Km
212
20
16
,g
8
0
[
I
I
I
I
I
I
0
0.2
0.4
0.6
0.8
1.0
1.2
Inhibitor concentration
(mM)
Fig. 1. Secondary plots, K m versus [I1, %r four different inhibitors. Microsomes were incubated with inhibitor at 37°C for 15 min after which glucose 6-phosphatase was assayed at seven substrate cencentrations. Km-values were then determined from unweighted Hanes plots (see text). The inhibitors were: pentenone ( • ), pentenal (at), hexenal (m) and heptenal (0).
181 16
iF= 14 12
•
•
10
8_1"
•
•
I
I
I
I
I
I
I
0
0.2
0.4
0.6
0.8
1.0
1.2
Inhibitor concentration
{mM)
Fig. 2. Secondary plots, 1/V versus [I], for four different inhibitors. V-values were determined from the same experiments as those described in the legend to Fig. 1 and the same symbols are used.
213 TABLE I APPARENT INHIBITOR CONSTANTS AND THEIR RATIO Experimental details were as described in the legend to Fig. 1, except that a 35 rain incubation was included. Incubation (35 rain)
Incubation (15 rain)
Pentenone Pentenal Hexenal Heptenal
Ki(Km) (raM)
Ki(V) (mM)
Ki(Km)/ Ki(V)
Ki(Km) (raM)
Ki(V) (mM)
Ki(V)
Ki(Km)/
0.33 0.90 0.73 0.42
36 14 2.8 1.4
0.009 0.06 0.26 0.30
0.09 0.46 0.36 0.26
24 2.9 1.6 0.8
0.004 0.16 0.23 0.33
or a halving of V were therefore easily detected from the intersection of the secondary plots with the abscissa. They are named 'apparent inhibitor constants' with symbols Ki(Km) and Ki(V), respectively. It should be noted, however, that these Ki-values are 'nominal' as the true inhibitor concentration in the lipid phase around the enzyme is most likely different from the bulk value and furthermore may be dependent on incubation time. Table I shows values of Ki(Km) and Ki(V) after 15 and 35 min of incubation. Measurements were taken at two different incubation times as the inhibition progresses rather slowly and is not monophasic (Agerbo et al., unpublished data). It is seen from the table that pentenone showed a much greater effect on Km
0
6
0.0
I
I
I
I
I
0.3
0.6
0.9
1.2
1.5
Conc.
of nonenal
(mM)
Fig. 3. Secondary plots, K m versus [nonenal]. Microsomes were incubated with 2-nonenal at 37°C for 15 min (0) or 35 rain ( i ) after which glucose 6-phosphatase was assayed at seven substrate concentrations. Kin-values were then determined from unweighted Hanes plots.
214
than did pentenal, whereas the opposite was true regarding V. An increase in aldehyde chain-length resulted in a more efficient inhibition with the greatest change found in Ki(V). In all cases, however, K m was the most strongly affected (Table I). Such a predominant Km-effect has also been found with several SHgroup binding reagents [3,7,9] which support the hypothesis that the a,~unsaturated oxo-compounds bind to an SH-group in the enzyme. The particularly low Ki(Km) with pentenone may be explained by the lack of sterical hindrance at the/3-carbon atom, which is the primary target for a vinylogous thiol-addition. An entirely different result was obtained with the longer-chain aldehyde 2-nonenal (Figs. 3 and 4). This inhibitor exhibited a non-linear Km versus [I] relationship (Fig. 3), with a maximum which depended on incubation time both with regard to height and position. The change in Km over the nonenal concentration range used ( 0 - 1 mM) was small, whereas V was affected to a great extent (Fig. 4). 2-Octenal behaved something between nonenal and shorter-chain aldehydes: a steep rise in Km with octenal concentration was found initially, but a maximum was reached with less than I mM octenal and was also much decreased. Non-linear Km versus [I] plots of a similar type were obtained when small concentrations of detergent (Triton ® X-114) were added to microsomes [1] and may reflect a conformational change in the enzyme caused by perturbation of its lipid environment. As expected, this perturbation was smaller with the shorterchain aldehydes. The 2,4-dienals, hexadienal and heptadienal, caused hardly any alteration in the kinetic constants and only a small decrease in Km was seen with decadienal.
120 100 80
~
40 20
0.0
I
I
I
I
I
0.3
0.6
0.9
1.2
1.5
Conc. of nonenal (mM) Fig. 4. Secondary plots, 1IV versus [nonenal]. V-values were determined from the same experiments as those described in the legend to Fig. 3 and the same symbols are used.
215
TABLE II A P P A R E N T I N H I B I T O R CONSTANTS AND T H E I R RATIO
Experimental details were as described in the legend to Fig. 1, except that the partially purified enzyme preparation was used and the incubation was for 1 h at 22°C.
Hexenal Nonenal
Ki(Km) (mM)
Ki(I0 (raM)
Ki(Km)/Ki(IO
1.01 0.41
8.1 1.34
0.12 0.31
The lack of a marked effect may be attributed to the rather rigid planar structure around the three conjugated double bonds giving rise to a severe sterical hindrance at the ~- and b-carbon atoms. A commercially available, crude preparation of partially purified glucose 6-phosphatase was inhibited differently from the microsome-bound enzyme by unsaturated aldehydes. Hexenal and nonenal behaved as classical noncompetitive inhibitors of that preparation (Table II) and the reactions were of pseudo-first order. (A similar behavior was found with N-butylmaleimide inhibition of detergent-disrupted microsomes [8]). These observations, made with an enzyme whose surrounding membrane is absent, further substantiate the assumption that two distinct mechanisms of inhibition by alkenals are operative in intact microsomes.
TABLE III E F F E C T OF PMB ON I N H I B I T I O N BY H E X E N A L
Microsomes were incubated with PMB at 0°C for 15 min and then with seven different concentrations ( 0 - 0 . 7 5 raM) of hexenal at 37°C for 35 min. The slope and intercept of plots of v - 1 versus hexenal concentration were then determined and tabled as function of the PMB concentration. [PMB] (~M) 0 20 40 60 80 100 Slope 2 b t
Slope (raM - 2 min) 10.8 9.5 11.9 14.1 16.7 12.9
± ± ± ± ± ±
5.8 a 1.6 2.6 5.7 1.7 1.1
0. 0486 c ± 0.0660 2.04
Intercept (mM - 1 rain) 11.7 13.1 13.9 14.7 15.4 16.9
± ± ± ± ± ±
2.8 0.7 1.2 2.6 0.7 0.5
0.0480 ± 0.0076 17.5
aThe '+ values' represent 95% confidence intervals. bSlope of plot of data in column 2 or 3 versus data in column 1. eNot significantly different from 0 at the 5% level.
216
The probable involvement of a mercapto group in inhibition by alkenals was further tested by addition of the well known thiol reagent PMB. This mercury compound acts as a classical inhibitor of glucose 6-phosphatase [9], i.e. the reciprocal reaction rate is linearly increasing with PMB concentration. If two inhibitors compete for a mutual site, or binding otherwise is exclusive, the slope of plots of V- 1 versus [I], obtained with one inhibitor is independent of the concentration of the other whereas the intercept is a linear function of that variable. On the other hand, if the inhibitors bind independently, both the slope and intercept are increasing with the concentration of the other inhibitor. Table III shows the effect of PMB on the slope and intercept of this type of plot obtained with hexenal as inhibitor. The intercept increased linearly as expected and the slope was virtually independent of the mercury compound. Similar results were found with nonenal or when another SH-group blocking reagent, 4,4 '-diisothiocyanostilbene 2,2'-disulfonate (DIDS), was substituted for PMB. Both of these thiol reagents have been shown by Schulze et al. [9,22,23] to bind to an 'essential' SH-group, so the present results support the hypothesis [10] that the alkenals also bind to this group. The bond resulting from a nucleophilic addition of a mercapto group to a C = C double bond is a thioether bond, which is expected to be stable. Washing the alkenal-treated microsomes, therefore, should not remove inhibition if that bonding itself is the sole cause of inhibition. Table IV shows the result of washing preinhibited microsomes by suspension in incubation medium, centrifugation and resuspension of the pellets. Non-inhibited controls were included in order to correct for any change in specific activity during the procedure. Due to the rotor geometry and the partial filling of the tubes, the sedimentation occurred with 8min = 145 S. This resulted in recovery of two-thirds of the original activity but
TABLE IV E F F E C T OF W A S H I N G T H E MICROSOMES A F T E R INCUBATION W I T H A L K E N A L S Microsomes were incubated with -- 1 mM aldehyde at 37°C for 20 rain. To 1.4 ml incubation mixture were added 2.6 ml incubation medium in 15-ml tubes, which were then centrifuged in a Sorvall SM-24 rotor (4°C, 60 min, f = 15 000 min 1, rmin and rmax = 9.7 & 11.1 cm). The pellet was gently rinsed with 1 ml medium, resuspended in 1.4 ml medium with or without aldehyde and incubated for another 20 min (37°C). Microsomes incubated without inhibitor were used as controls. Specific activity (% of control) Before washing
2-Hexenal 2-Heptenal 2-Nonenal
62 4- 1 a 47 ± 1 27 ± 2
aThe ' ± values' are calculated S.D.s.
After washing No extra aldehyde
With extra aldehyde
70 ± 3 50 ± 2 55 ± 2
37 ± 1 20 ± 1 6 ± 1
217
only about one-half of the protein content. As seen from the table, microsomes treated with hexenal or heptenal virtually retained their degree of inhibition shown by the small change in relative specific activity. But inhibition by nonenal was much reduced. A new incubation of the washed microsomes with aldehyde resulted in further inhibition in all cases, because the reactions were not complete during the first incubation period. This extra inhibition shows that the rise in specific activity of the nonenal-treated microsomes by washing was not due to a loss of capability of being inhibited. Instead, it is another example of the differences between aldehydes of different chain-length and is in accordance with the hypothesis that not only formation of the thioether bond but also a distortion of the lipid phase play a part in the inhibition of microsomal glucose 6-phosphatase by the longer-chain alkenals. ACKNOWLEDGEMENT
The excellent technical assistance of Ms. Lis Berner is gratefully acknowledged. This work was carried out at Marine Biotechnology Center under The Biotechnological Research and Development Programme 1987-1990. REFERENCES 1
2 3
4 5 6 7 8 9
10
11
12
H.-U. Schulze, R. Kannler and B. Junker, Latency studies on rat liver microsomal glucose 6-phosphatase. Correlation of membrane modification and solubilization by Triton X-114 with the enzymatic activity, Biochim. Biophys. Acta, 814 (1985) 85-95. R.C. Nordlie, Glucose 6-phosphatase, hydrolytic and synthetic activities, in: P.D. Boyer, (Ed.), The Enzymes, Vol. IV, Academic Press, New York, 1971, 543-610. B.K. Wallin and W.J. Arion, The requirement for membrane integrity in the inhibition of hepatic glucose 6-phosphatase by sulfhydry] reagents and taurocholate, Biochem. Biophys. Res. Commun., 48 (1972) 694-699. W.J. Arion and H.E. Walls, The importance of membrane integrity in kinetic characterizations of the microsomal glucose 6-phosphate system, J. Biol. Chem., 257 (1982) 11217-11220. D. Zakim and D.E. Edmonson, The role of the membrane in the regulation of activity of microsomal glucose 6-phosphatase, J. Biol. Chem., 257 (1982) 1145-1148. K.A. Sukalski and R.C. Nordlie, Glucose 6-phosphatase: two concepts of membrane-function relationship, Adv. Enzymol., 62 (1988) 93- 117. W. Colilla and R.C. Nordlie, Effects of sulfhydryl reagents on synthetic and hydrolytic activities of multifunctional glucose 6-phosphatase, Biochim. Biophys. Acta, 309 (1973) 328-338. B. Vakili and M. Banner, The effect of N-alkylmaleimides on the activity of rat liver glucose 6-phosphatase, Biochem. J., 194 (1981) 319- 325. H.-U. Schulze, B. Nolte and R. Kannler, Evidence for changes in the conformational status of rat liver microsomal glucose 6-phosphate phosphohydrolase during detergent-dependent membrane modification. Effect of p-mercuribenzoate and organomercurial agarose gel on glucose 6-phosphatase of native and detergent-modified microsomes, J. Biol. Chem., 261 (1986) 16571 - 16578. G. Bertone and M.U. Dianzani, Inhibition by aldehydes as a possible further mechanism for glucose 6-phosphatase inactivation during CCl4-poisoning, Chem.-Biol. Interact., 19 (1977) 91 - 100. A. Benedetti, A.F. Casini, M. Ferrali and M. Comporti, Extraction and partial characterization of dialysable products originating from the peroxidation of liver microsomal lipids and inhibiting microsomal glucose 6-phosphatase activity, Biochem. Pharmacol., 28 (1979) 2909-2918. H. de Groot, T. Noll and B. Rymsa, Alterations of the microsomal glucose 6-phosphatase system
218
13
14
15
16 17
18
19 20 21 22
23
evoked by ferrous iron- and haloalkane free-radical-mediated lipid peroxidation, Biochim. Biophys. Acta, 881 (1986) 350-355. H. de Groot, T. Noll and T. TSlle, Loss of latent activity of liver microsomal membrane enzymes evoked by lipid peroxidation. Studies of nucleoside diphosphatase, glucose 6-phosphatase and UDP glucuronyltransferase, Biochim. Biophys. Acta, 815 (1985) 91- 96. G. Inselmann, J. Hannemann and K. Baumann, Cyclosporine A induced lipid peroxidation and influence on glucose 6-phosphatase in rat hepatic and renal microsomes, Res. Commun. Chem. Pathol. Pharmacol., 68 (1990) 189-203. R. Fulceri, A. Pompella, A. Benedetti and M. Comporti, On the role of lipid peroxidation and protein-bound aldehydes in the haloalkane-induced inactivation of microsomal glucose 6-phosphatase, Res. Commun. Chem. Pathol. Pharmacol., 68 (1990) 73-88. M. Comporti, Lipid peroxidation and cellular damage in toxic liver injury, Lab. Invest., 53 (1985) 599-623. A. Benedetti, H. Esterbauer, M. Ferrali, R. Fulceri and M. Comporti, Evidence for aldehydes bound to liver microsomal protein following CC14or BrCCl:t poisoning, Biochim. Biophys. Acta, 711 (1982) 345-356. H. Esterbauer, H. Zollner and R.J. Schaur, Aldehydes formed by lipid peroxidation: Mechanisms of formation, occurrence and determination, in: C. Vigo-Pelfrey (Ed.), Membrane Lipid Oxidation, Vol. I, CRC Press, Boca Raton, FL, 1990, 239-268. C.H. Fiske and Y. Subbarow, The colorimetric determination of phosphorus, J. Biol. Chem., 66 (1925) 375 - 400. O.H. Lowry, N.J. Rosebrough, A.L. Farr and R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275, C.S. Hanes, Studies on plant amylases. I. The effect of starch concentration upon the velocity of hydrolysis by the amylase of germinated barley, Biochem. J., 26 (1932) 1406- 1421. M. Speth and H.-U. Schulze, On the nature of the interaction between 4,4'-diisothiocyanostilbene 2,2'-disulfonic acid and microsomal glucose 6-phosphatase. Evidence for the involvement of sulfhydryl groups of the phosphohydrolase, Eur. J. Biochem., 174 (1988) 111- 117. M. Speth, N. Baake and K.-U. Schulze, Topographical localization and characterization of microsomal glucose 6-phosphatase binding sites accessible to 4,4'diazidostilbene 2,2'disulfonic acid, Arch. Biochem. Biophys., 275 (1989) 202-214.
209
INHIBITION OF MICROSOMAL GLUCOSE 6-PHOSPHATASE BY UNSATURATED ALIPHATIC A L D E H Y D E S AND KETONES
BO M. JORGENSEN a, PIA AGERBO a, BENNY J E N S E N b, TORGER BORRESEN ~, and GUNHILD HOLMER c
aCenter for Food Research at DTH, bTechnological Laboratory, Danish Ministry of Fisheries and CDepartment of Biochemistry and Nutrition, Technical University Build. 221 "'~' and 224", DK-2800 Lyngby (Denmark) (Received August 5th, 1991) (Revision received November 6th, 1991) (Accepted November 8th, 1991)
SUMMARY
Aldehydes and ketones with one double bond conjugated to the carbonyl group inhibited the enzyme glucose 6-phosphatase, which is embedded in the microsomal membrane. The Michaelis constant, Km and the maximal rate of reaction, V, were affected in a way dependent on the inhibitor's chain-length: trans-2-pentenal and 1-penten-3-one increased Km linearly with concentration and had almost no effect on V, whereas trans-2-nonenal caused a large increase in V but only a small and non-linear change in Km. The effect of the short-chain aldehydes on the kinetic parameters increased with chain-length, but pentenone increased Km more than did trans-2-heptenal and conjugated dienals did not act as inhibitors. Therefore, sterical effects apparently are of importance. Washing the microsomes after incubation with hexenal or heptenal did not substantially decrease the inhibition, but with nonenal the inhibition was reduced by washing. Inhibition by the SH-group blocking reagent p-hydroxymercuribenzoate was competitive to inhibition by the alkenals. It is concluded that the a-/3 unsaturated oxo-compounds inhibit glucose 6-phosphatase by binding covalently to an important mercapto group and that perturbation of the enzyme's membrane environment also plays a part in the inhibition.
Key words: Alkenals -- Autoxidation -- Membrane perturbation -- Mercapto group -- p-Hydroxymercuribenzoate Correspondence to: Bo JCrgensen, DTH Build. 221 (FF), DK-2800 Lyngby, Denmark. Abbreviations: EDTA, ethylenediaminetetraacetic acid; f, speed of rotation; [I], inhibitor concentration; Ki(Km) and Ki(V), apparent inhibitor constant; Kin, Michaelis constant; PMB, p-hydroxymercuribenzoate; rmin and rmax, minimal and maximal rotation radii (centrifugal data); So, initial substrate concentration; v, initial rate of reaction; V, maximal rate of reaction. 0009-2797/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
210 INTRODUCTION
Glucose 6-phosphatase (EC 3.1.3.9) of mammalian liver is an enzyme located in the endoplasmic reticulum. Its properties are heavily dependent on the membrane integrity and even small perturbations of the environment, e.g. with detergent [1], may drastically alter the kinetic parameters, specificity and stability [1-6]. The enzyme is inhibited by common SH-group binding reagents [7-9] with the predominant effect being a rise in Km (glucose-6-P). But the mercapto group involved is apparently not situated at the active site as the reagent phydroxymercuribenzoate (PMB) binds without inhibition at 0°C [9]. Induced lipid autoxidation or addition of some of the secondary oxidation products also inhibit the enzyme [10- 16]. It has been shown [16 - 18; Agerbo et al., unpublished data] that among the variety of products formed, the active ones are aft-unsaturated aldehydes (e.g. 4-hydroxy-2-nonenal) and ketones, which are believed to bind to SH-groups in proteins by a conjugate nucleophilic addition to the carbon-carbon double bond [10,16-18]. Thus, reactions of that type may be suggested as the cause of inhibition of glucose 6-phosphatase. In order to further explore the mechanism, we have measured the influence of a series of unsaturated oxo-compounds on the kinetic parameters, Michaelis constant and maximal rate of reaction. Short-chain compounds increased Km linearly with concentration and had only minor effects on V, whereas longer-chain alkenals decreased V markedly and exhibited a non-linear Km relationship with inhibitor concentration. Inhibition was in both cases competitive to inhibition by PMB and enzymatic activity was only partly recovered by washing the microsomes with buffer. It is concluded that inhibition involves at least two distinct mechanisms: a binding to an 'essential' SH-group and a perturbation of the membrane lipid phase. The second effect is more pronounced with the longerchain aldehydes. MATERIALS AND METHODS
Reagents 1-Penten-3-one, 2-pentenal, 2-hexenal, 2-heptenal, 2-octenal, 2-nonenal, 2,4-hexadienal, 2,4-heptadienal and 2,4-decadienal -- all with trans-configuration around the C = C double bonds -- were purchased from Aldrich-Chemie, Germany, in the highest grade available (typically 9 3 - 99%) and used without further purification. Glucose 6-phosphate (analytical grade) and a crude preparation of partially purified glucose 6-phosphatase were obtained from Sigma Chemical Company, USA. Buffer reagents were of the highest quality commercially available.
Preparation of microsomes The liver from a laboratory rat that had been fed a pellet diet was homogenized in a medium containing 250 mM sucrose, 1 mM Tris and 0.1 mM EDTA, pH = 7.0, in a Potter-Elvehjem homogenizer with a loose-fitting pestle. The homogenate was centrifuged (4°C, 15 min, f = 11 000 min -1, rmin and
211
rmax = 5.7 and 10.7 cm) in a Sorvall RC-5B centrifuge equipped with an SS-34 rotor. The supernatant was further centrifuged (4°C, 65 rain, f = 35 000 min- 1, rmin and rmax = 3.95 and 9.19 cm) in a Beckman L8-70M ultracentrifuge (70 TI rotor) and the resultant pellet was resuspended in incubation medium (50 mM Tris, 50 mM maleate, 150 mM KC1; pH = 7.4) to a volume of 1 ml per g liver, making the protein concentration to be around 15 - 20 mg/ml. The preparation could be stored at - 3 0 ° C for more than a month without measurable loss of glucose 6-phosphatase activity.
Incubation with inhibitor Stock solutions (1 - 2 mM) of inhibitor in incubation medium were prepared in 100-ml measuring flasks. Four or five volumes of inhibitor, diluted with medium to the appropriate concentration, were mixed with one volume of microsomes and gently shaken at 37°C in sealed bottles.
Enzyme assay Assay buffer (500 td, 50 mM maleate, 4.6 mM EDTA, 2.3 mM NaF; pH = 6.0) were mixed with 100 ~l 200 mM glucose 6-phosphate (pH = 6.0) in a 15-ml polycarbonate tube (Sorvall). The enzymatic reaction (at 37°C) was started by adding 100 #l incubation mixture and terminated after an appropriate time (usually 20 min) by addition of 1.3 ml 10% (w/v) trichloroacetic acid. The precipitated protein was settled by centrifugation at f = 12 000 rain-1 for 10 min in a Sorvall SM-24 rotor (4°C) and 1000 ~l was taken to phosphate analysis. Blank assays were performed by addition of acid prior to enzyme.
Determination of phosphate The amount of phosphate formed in the assay was determined essentially as described by Fiske and Subbarow [19]. Sample (1000 #l) was mixed with 2000 ~l water, 1000 ~l 'molybdate reagent' (1.36 vol. conc. sulfuric acid, 3 vol. water, 5 vol. 50 g/1 hexaammonium molybdate; diluted to 10 vol with water) and 1000 ~l 'reducing reagent' (30 g/1 sodium disulfite, 10 g/1 4-(methylamino)-phenol sulfate (Photo-Rex ® ; Merck, Germany)). Absorbance at 660 nm was read after at least 40 min.
Determination of protein The protein content of the microsomal preparations was estimated by the Folin-Ciocalteu reagent [20] and with bovine serum albumin as standard. RESULTS
AND DISCUSSION
Glucose 6-phosphatase followed simple Michaelis-Menten kinetics with glucose 6-phosphate as substrate in the absence as well as in the presence of inhibitors. The kinetic constants, Km and V, were calculated from Hanes plots (So/Vversus So) [21]. 'Secondary plots', Km and 1/V versus [I], were apparently linear in the 0-1 mM range when pentenone, pentenal, hexenal or heptenal were used as inhibitor (Figs. 1 and 2). The inhibitor concentrations needed for a doubling of Km
212
20
16
,g
8
0
[
I
I
I
I
I
I
0
0.2
0.4
0.6
0.8
1.0
1.2
Inhibitor concentration
(mM)
Fig. 1. Secondary plots, K m versus [I1, %r four different inhibitors. Microsomes were incubated with inhibitor at 37°C for 15 min after which glucose 6-phosphatase was assayed at seven substrate cencentrations. Km-values were then determined from unweighted Hanes plots (see text). The inhibitors were: pentenone ( • ), pentenal (at), hexenal (m) and heptenal (0).
181 16
iF= 14 12
•
•
10
8_1"
•
•
I
I
I
I
I
I
I
0
0.2
0.4
0.6
0.8
1.0
1.2
Inhibitor concentration
{mM)
Fig. 2. Secondary plots, 1/V versus [I], for four different inhibitors. V-values were determined from the same experiments as those described in the legend to Fig. 1 and the same symbols are used.
213 TABLE I APPARENT INHIBITOR CONSTANTS AND THEIR RATIO Experimental details were as described in the legend to Fig. 1, except that a 35 rain incubation was included. Incubation (35 rain)
Incubation (15 rain)
Pentenone Pentenal Hexenal Heptenal
Ki(Km) (raM)
Ki(V) (mM)
Ki(Km)/ Ki(V)
Ki(Km) (raM)
Ki(V) (mM)
Ki(V)
Ki(Km)/
0.33 0.90 0.73 0.42
36 14 2.8 1.4
0.009 0.06 0.26 0.30
0.09 0.46 0.36 0.26
24 2.9 1.6 0.8
0.004 0.16 0.23 0.33
or a halving of V were therefore easily detected from the intersection of the secondary plots with the abscissa. They are named 'apparent inhibitor constants' with symbols Ki(Km) and Ki(V), respectively. It should be noted, however, that these Ki-values are 'nominal' as the true inhibitor concentration in the lipid phase around the enzyme is most likely different from the bulk value and furthermore may be dependent on incubation time. Table I shows values of Ki(Km) and Ki(V) after 15 and 35 min of incubation. Measurements were taken at two different incubation times as the inhibition progresses rather slowly and is not monophasic (Agerbo et al., unpublished data). It is seen from the table that pentenone showed a much greater effect on Km
0
6
0.0
I
I
I
I
I
0.3
0.6
0.9
1.2
1.5
Conc.
of nonenal
(mM)
Fig. 3. Secondary plots, K m versus [nonenal]. Microsomes were incubated with 2-nonenal at 37°C for 15 min (0) or 35 rain ( i ) after which glucose 6-phosphatase was assayed at seven substrate concentrations. Kin-values were then determined from unweighted Hanes plots.
214
than did pentenal, whereas the opposite was true regarding V. An increase in aldehyde chain-length resulted in a more efficient inhibition with the greatest change found in Ki(V). In all cases, however, K m was the most strongly affected (Table I). Such a predominant Km-effect has also been found with several SHgroup binding reagents [3,7,9] which support the hypothesis that the a,~unsaturated oxo-compounds bind to an SH-group in the enzyme. The particularly low Ki(Km) with pentenone may be explained by the lack of sterical hindrance at the/3-carbon atom, which is the primary target for a vinylogous thiol-addition. An entirely different result was obtained with the longer-chain aldehyde 2-nonenal (Figs. 3 and 4). This inhibitor exhibited a non-linear Km versus [I] relationship (Fig. 3), with a maximum which depended on incubation time both with regard to height and position. The change in Km over the nonenal concentration range used ( 0 - 1 mM) was small, whereas V was affected to a great extent (Fig. 4). 2-Octenal behaved something between nonenal and shorter-chain aldehydes: a steep rise in Km with octenal concentration was found initially, but a maximum was reached with less than I mM octenal and was also much decreased. Non-linear Km versus [I] plots of a similar type were obtained when small concentrations of detergent (Triton ® X-114) were added to microsomes [1] and may reflect a conformational change in the enzyme caused by perturbation of its lipid environment. As expected, this perturbation was smaller with the shorterchain aldehydes. The 2,4-dienals, hexadienal and heptadienal, caused hardly any alteration in the kinetic constants and only a small decrease in Km was seen with decadienal.
120 100 80
~
40 20
0.0
I
I
I
I
I
0.3
0.6
0.9
1.2
1.5
Conc. of nonenal (mM) Fig. 4. Secondary plots, 1IV versus [nonenal]. V-values were determined from the same experiments as those described in the legend to Fig. 3 and the same symbols are used.
215
TABLE II A P P A R E N T I N H I B I T O R CONSTANTS AND T H E I R RATIO
Experimental details were as described in the legend to Fig. 1, except that the partially purified enzyme preparation was used and the incubation was for 1 h at 22°C.
Hexenal Nonenal
Ki(Km) (mM)
Ki(I0 (raM)
Ki(Km)/Ki(IO
1.01 0.41
8.1 1.34
0.12 0.31
The lack of a marked effect may be attributed to the rather rigid planar structure around the three conjugated double bonds giving rise to a severe sterical hindrance at the ~- and b-carbon atoms. A commercially available, crude preparation of partially purified glucose 6-phosphatase was inhibited differently from the microsome-bound enzyme by unsaturated aldehydes. Hexenal and nonenal behaved as classical noncompetitive inhibitors of that preparation (Table II) and the reactions were of pseudo-first order. (A similar behavior was found with N-butylmaleimide inhibition of detergent-disrupted microsomes [8]). These observations, made with an enzyme whose surrounding membrane is absent, further substantiate the assumption that two distinct mechanisms of inhibition by alkenals are operative in intact microsomes.
TABLE III E F F E C T OF PMB ON I N H I B I T I O N BY H E X E N A L
Microsomes were incubated with PMB at 0°C for 15 min and then with seven different concentrations ( 0 - 0 . 7 5 raM) of hexenal at 37°C for 35 min. The slope and intercept of plots of v - 1 versus hexenal concentration were then determined and tabled as function of the PMB concentration. [PMB] (~M) 0 20 40 60 80 100 Slope 2 b t
Slope (raM - 2 min) 10.8 9.5 11.9 14.1 16.7 12.9
± ± ± ± ± ±
5.8 a 1.6 2.6 5.7 1.7 1.1
0. 0486 c ± 0.0660 2.04
Intercept (mM - 1 rain) 11.7 13.1 13.9 14.7 15.4 16.9
± ± ± ± ± ±
2.8 0.7 1.2 2.6 0.7 0.5
0.0480 ± 0.0076 17.5
aThe '+ values' represent 95% confidence intervals. bSlope of plot of data in column 2 or 3 versus data in column 1. eNot significantly different from 0 at the 5% level.
216
The probable involvement of a mercapto group in inhibition by alkenals was further tested by addition of the well known thiol reagent PMB. This mercury compound acts as a classical inhibitor of glucose 6-phosphatase [9], i.e. the reciprocal reaction rate is linearly increasing with PMB concentration. If two inhibitors compete for a mutual site, or binding otherwise is exclusive, the slope of plots of V- 1 versus [I], obtained with one inhibitor is independent of the concentration of the other whereas the intercept is a linear function of that variable. On the other hand, if the inhibitors bind independently, both the slope and intercept are increasing with the concentration of the other inhibitor. Table III shows the effect of PMB on the slope and intercept of this type of plot obtained with hexenal as inhibitor. The intercept increased linearly as expected and the slope was virtually independent of the mercury compound. Similar results were found with nonenal or when another SH-group blocking reagent, 4,4 '-diisothiocyanostilbene 2,2'-disulfonate (DIDS), was substituted for PMB. Both of these thiol reagents have been shown by Schulze et al. [9,22,23] to bind to an 'essential' SH-group, so the present results support the hypothesis [10] that the alkenals also bind to this group. The bond resulting from a nucleophilic addition of a mercapto group to a C = C double bond is a thioether bond, which is expected to be stable. Washing the alkenal-treated microsomes, therefore, should not remove inhibition if that bonding itself is the sole cause of inhibition. Table IV shows the result of washing preinhibited microsomes by suspension in incubation medium, centrifugation and resuspension of the pellets. Non-inhibited controls were included in order to correct for any change in specific activity during the procedure. Due to the rotor geometry and the partial filling of the tubes, the sedimentation occurred with 8min = 145 S. This resulted in recovery of two-thirds of the original activity but
TABLE IV E F F E C T OF W A S H I N G T H E MICROSOMES A F T E R INCUBATION W I T H A L K E N A L S Microsomes were incubated with -- 1 mM aldehyde at 37°C for 20 rain. To 1.4 ml incubation mixture were added 2.6 ml incubation medium in 15-ml tubes, which were then centrifuged in a Sorvall SM-24 rotor (4°C, 60 min, f = 15 000 min 1, rmin and rmax = 9.7 & 11.1 cm). The pellet was gently rinsed with 1 ml medium, resuspended in 1.4 ml medium with or without aldehyde and incubated for another 20 min (37°C). Microsomes incubated without inhibitor were used as controls. Specific activity (% of control) Before washing
2-Hexenal 2-Heptenal 2-Nonenal
62 4- 1 a 47 ± 1 27 ± 2
aThe ' ± values' are calculated S.D.s.
After washing No extra aldehyde
With extra aldehyde
70 ± 3 50 ± 2 55 ± 2
37 ± 1 20 ± 1 6 ± 1
217
only about one-half of the protein content. As seen from the table, microsomes treated with hexenal or heptenal virtually retained their degree of inhibition shown by the small change in relative specific activity. But inhibition by nonenal was much reduced. A new incubation of the washed microsomes with aldehyde resulted in further inhibition in all cases, because the reactions were not complete during the first incubation period. This extra inhibition shows that the rise in specific activity of the nonenal-treated microsomes by washing was not due to a loss of capability of being inhibited. Instead, it is another example of the differences between aldehydes of different chain-length and is in accordance with the hypothesis that not only formation of the thioether bond but also a distortion of the lipid phase play a part in the inhibition of microsomal glucose 6-phosphatase by the longer-chain alkenals. ACKNOWLEDGEMENT
The excellent technical assistance of Ms. Lis Berner is gratefully acknowledged. This work was carried out at Marine Biotechnology Center under The Biotechnological Research and Development Programme 1987-1990. REFERENCES 1
2 3
4 5 6 7 8 9
10
11
12
H.-U. Schulze, R. Kannler and B. Junker, Latency studies on rat liver microsomal glucose 6-phosphatase. Correlation of membrane modification and solubilization by Triton X-114 with the enzymatic activity, Biochim. Biophys. Acta, 814 (1985) 85-95. R.C. Nordlie, Glucose 6-phosphatase, hydrolytic and synthetic activities, in: P.D. Boyer, (Ed.), The Enzymes, Vol. IV, Academic Press, New York, 1971, 543-610. B.K. Wallin and W.J. Arion, The requirement for membrane integrity in the inhibition of hepatic glucose 6-phosphatase by sulfhydry] reagents and taurocholate, Biochem. Biophys. Res. Commun., 48 (1972) 694-699. W.J. Arion and H.E. Walls, The importance of membrane integrity in kinetic characterizations of the microsomal glucose 6-phosphate system, J. Biol. Chem., 257 (1982) 11217-11220. D. Zakim and D.E. Edmonson, The role of the membrane in the regulation of activity of microsomal glucose 6-phosphatase, J. Biol. Chem., 257 (1982) 1145-1148. K.A. Sukalski and R.C. Nordlie, Glucose 6-phosphatase: two concepts of membrane-function relationship, Adv. Enzymol., 62 (1988) 93- 117. W. Colilla and R.C. Nordlie, Effects of sulfhydryl reagents on synthetic and hydrolytic activities of multifunctional glucose 6-phosphatase, Biochim. Biophys. Acta, 309 (1973) 328-338. B. Vakili and M. Banner, The effect of N-alkylmaleimides on the activity of rat liver glucose 6-phosphatase, Biochem. J., 194 (1981) 319- 325. H.-U. Schulze, B. Nolte and R. Kannler, Evidence for changes in the conformational status of rat liver microsomal glucose 6-phosphate phosphohydrolase during detergent-dependent membrane modification. Effect of p-mercuribenzoate and organomercurial agarose gel on glucose 6-phosphatase of native and detergent-modified microsomes, J. Biol. Chem., 261 (1986) 16571 - 16578. G. Bertone and M.U. Dianzani, Inhibition by aldehydes as a possible further mechanism for glucose 6-phosphatase inactivation during CCl4-poisoning, Chem.-Biol. Interact., 19 (1977) 91 - 100. A. Benedetti, A.F. Casini, M. Ferrali and M. Comporti, Extraction and partial characterization of dialysable products originating from the peroxidation of liver microsomal lipids and inhibiting microsomal glucose 6-phosphatase activity, Biochem. Pharmacol., 28 (1979) 2909-2918. H. de Groot, T. Noll and B. Rymsa, Alterations of the microsomal glucose 6-phosphatase system
218
13
14
15
16 17
18
19 20 21 22
23
evoked by ferrous iron- and haloalkane free-radical-mediated lipid peroxidation, Biochim. Biophys. Acta, 881 (1986) 350-355. H. de Groot, T. Noll and T. TSlle, Loss of latent activity of liver microsomal membrane enzymes evoked by lipid peroxidation. Studies of nucleoside diphosphatase, glucose 6-phosphatase and UDP glucuronyltransferase, Biochim. Biophys. Acta, 815 (1985) 91- 96. G. Inselmann, J. Hannemann and K. Baumann, Cyclosporine A induced lipid peroxidation and influence on glucose 6-phosphatase in rat hepatic and renal microsomes, Res. Commun. Chem. Pathol. Pharmacol., 68 (1990) 189-203. R. Fulceri, A. Pompella, A. Benedetti and M. Comporti, On the role of lipid peroxidation and protein-bound aldehydes in the haloalkane-induced inactivation of microsomal glucose 6-phosphatase, Res. Commun. Chem. Pathol. Pharmacol., 68 (1990) 73-88. M. Comporti, Lipid peroxidation and cellular damage in toxic liver injury, Lab. Invest., 53 (1985) 599-623. A. Benedetti, H. Esterbauer, M. Ferrali, R. Fulceri and M. Comporti, Evidence for aldehydes bound to liver microsomal protein following CC14or BrCCl:t poisoning, Biochim. Biophys. Acta, 711 (1982) 345-356. H. Esterbauer, H. Zollner and R.J. Schaur, Aldehydes formed by lipid peroxidation: Mechanisms of formation, occurrence and determination, in: C. Vigo-Pelfrey (Ed.), Membrane Lipid Oxidation, Vol. I, CRC Press, Boca Raton, FL, 1990, 239-268. C.H. Fiske and Y. Subbarow, The colorimetric determination of phosphorus, J. Biol. Chem., 66 (1925) 375 - 400. O.H. Lowry, N.J. Rosebrough, A.L. Farr and R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275, C.S. Hanes, Studies on plant amylases. I. The effect of starch concentration upon the velocity of hydrolysis by the amylase of germinated barley, Biochem. J., 26 (1932) 1406- 1421. M. Speth and H.-U. Schulze, On the nature of the interaction between 4,4'-diisothiocyanostilbene 2,2'-disulfonic acid and microsomal glucose 6-phosphatase. Evidence for the involvement of sulfhydryl groups of the phosphohydrolase, Eur. J. Biochem., 174 (1988) 111- 117. M. Speth, N. Baake and K.-U. Schulze, Topographical localization and characterization of microsomal glucose 6-phosphatase binding sites accessible to 4,4'diazidostilbene 2,2'disulfonic acid, Arch. Biochem. Biophys., 275 (1989) 202-214.
