Document RjYoxrq5vb8zKvVVkRnoGvd2V

THEJOURNAL OF BIOLOGICACHLEMISTRY Vol. 2.58, No. 10. Issue of May 25, pp, 63904393. 1983 Printed in U.S.A. Inhibition of ATP-dependentMicrosomal Ca2+Sequestration during Oxidative Stress and Its Preventionby Glutathione* (Received for publication, Decembe1r0, 1982) Dean P. Jones$&Hjordis Thorn, Martyn TS. mith/\**S, arah A. Jewelln, and Sten Orreniusn From the YDepartment of Forensic Medicine, Karolinska Institutet, ,910401 Stockholm, Sweden, the +Department of Biochemistry, Emory University, School of Medicine, Atlanta, Georgia 30322, and the llToxicology Unit, Department of Pharmacology, The School of Pharmacy, 29/39 Brunswick Square, London WCZN I N , United Kingdom Downloaded from www.jbc.org by guest, on March 29, 2010 The metallochromic indicator arsenazoIII was used and extramitochondrial compartments could be quantitated to study the effect of oxidative stress on ATP-depend- by a relatively simple methodT. he sequential additionof the ent Ca2+ uptake by rat liver microsomes. Addition of protonophore FCCP' and the ionophoreA23187 in the pres- ATP caused a rapid increase in ionophore A23187-re- ence of the metallochromic indicator arsenaIzIoI allowed the leasable Ca2+ which stabilized in 2-3 manind provided spectrophotometric determination of the Ca2+concentrations arapidandvery sensitive assay forATP-dependent Ca2' sequestration.Quantitatively, this fraction was sufficient to account for virtually all of the nonmitochondrial ionophore-releasable Cao2f +rat hepatocytes. Incubation with t-butyl hydroperoxide caused a rapid loss in the ability of microsomes to sequester Ca2+ in the presence of ATP. Addition of dithiothreitol or a physiological concentration of GSH to these incuba- in these two intracellular compartments (10, 11).Using this method, we haveshownthat exposure of hepatocytesto oxidative stress,by incubation with t-butylhydroperoxide or the redox-active quinone menadione (2-methyl-1,4-naphthoquinone), causes the mobilization of Ca2+ anda decrease in both the mitochondrial and extramitochondrial Ca2+ pools (10-12). It is very likely that most of the extramitochondrial tions provided effective protection against the oxida- pool is Ca2'-sequestered by the endoplasmic reticulum, and tive damage. ATP-dependent microsomal Ca2+ seques-since this pool appears tobe a primary target for Ca2+mobi- tration is therefore sensitive to oxidative damage and lization and loss during oxidative stress (11,12), in this study may be a primary site of injury leading to disturbed we have investigated theeffects of oxidative stress onmicro- Ca2' homeostasis during the early stages of drug he- somal Ca2+ sequestration. For these studies, we developed a patotoxicity.Intracellular thiols, notably GSH, may rapidspectrophotometricmethod for the measurement of prevent these changes by protecting the microsomal ATP-dependent microsomal Ca2+uptake which measures Ca2+pump from oxidative damage. only the ionophore-releasable pool. The results clearly dem- onstrate that microsomal ATP-dependent Ca" sequestration is highly sensitiveto oxidative damagecaused by t-butyl The intracellular free Ca2+ concentration is controlled by differential Ca2+transport across mitochondriale,ndoplasmic hydroperoxide and thatGSH effectively protects against this damage. reticular, and plasma membranes. Thineflux and efflux proc- EXPERIMENTALPROCEDURES esses of the mitochondria areknown to be able to control the Microsomes were prepared from male Sprague-Dawleyrats (190ambient free Ca2' concentration (1,2), and recent studies by 220 g, fed ad libitum) as described by Ernster et al. (13).Rats were Becker et al. (3)have shown that mitochondrialinflux-efflux treated with phenobarbital in drinking wate(r1mg/ml) for at least 7 cycling can be adjusted by energy-dependenCta2' sequestra- tion in microsomes. The Ca2+uptake systemof the endoplasmic reticulum has only recenbteleyn characterized ina variety of tissues and cell types, including the kidney (4), liver (5), pancreas (6),adipocyte (7), and submandibulargland (8).This system utilizes ATP andis thought tobe associated with the days prior to preparation of microsomes. Microsomes were washed oncebyresuspendingin0.15 M KC1 containing 5 p~ EGTAand centrifuging 30 min at 105,000 x g. The pellet was resuspended in 0.25 M sucrosecontaining 10 mM Tris-HC1,pH7.4, to give a find concentration of 10-15 m g / d and stored on ice until used. Incuba- tions were performed with approximately 1 mg of protein/ml in 125 mM KCl, 2 mM KzHP04,4m~ MgCL, and 25 m~ Hepes, pH 7.0.This microsomal Ca*+-ATPaseactivity (4, 5). Up to 50% of the total cellular calcium can be located in tehnedoplasmic reticulum of hepatocytes (9), and therefore, modulationof microsomal Caz+ may have an important regulatoryrole in maintaining intracellular Ca2+ homeostasis. In recent studies of Ca2' homeostasis in isolated hepato- was the medium used by Beckeret al. (3) to study Ca2+ regulatiobny both mitochondria and microsomes, except that the components to support mitochondrial Ca" uptake were omitted. Purified arsenazo I11wasdissolvedin the above medium, and the concentration was determined from the A s M . ~ sobtained byaddingexcessCa" to mediumwhich had been titrated withEGTA to remove CaZ+and using the Es54-M=5 2.15 X lo4 M" cm"for the Ca2+.arsenazo111 cytes, we found that the Ca2+ content of the mitochondrial complex. This AE value was determined from difference spectraof the Caz+.arsenazoI11complexin the abovemedium. Thisvalue corresponded to the A E value for these wavelengths calculatefdrom * These studies were performed in Stockholm,Sweden with support from Swedish Medical Research Council Project 03x3-2471and the Swedish Councilfor Planning and Coordinationof Research. The costs of publication of this article weredefrayedin part by the payment of pagecharges. This articlemusttherefore behereby Fig. 1 of Kendrick et al. (14), which was obtained usingsimilar conditions. Useof A for our studies was advantageous becauseof the turbidity due to the microsomes, and this approachwas usedthroughout for consistency. The measured KOwas 11.3p ~in, agreement with earlier measurements insimilar media (9).The high Mg2+concentra- marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 5 Supported by National Institutes of Health GrantGM-28176. * * Supportedby a European Science Exchange Awaordf the Royal Society of Great Britain. ' The abbreviationsused are: FCCP, carbonylcyanidep-trifluoro- methoxyphenylhydrazone; EGTA, ethylene glycol bis(B-aminoethyl ether)N,N,N',N"tetraaceticacid; arsenazo 111, 0-(1,8-dihydroxy-3,6disulfonaphthalene-2,7-bisazo)bisbenzenearsoniaccid. 6390 Microsomal ea2+Sequestration 6391 tion allowed additionof ATP and measurement of free Ca" concen- tration by absorbance changesat the wavelength pair 654 versus 685 nm without sigmficant change in the spectral characteristics or sen- sitivity of the dye due to changesin Mg2+concentration. Use of these wavelengths rather than 675 versus 685 nm was therefore preferred because they provided about 10-fold greater sensitivity. Dual wave- length and scanning spectrophotometrywas performed with either a Sigma ZWS-11 spectrophotometer or an Aminco DW2a spectropho- tometer at room temperature. Incubationswereperformedinrotatingroundbottom flasks at 37 "C. Proteinwas measured by the method of Lowry et al. (15) with bovine serum albumin as a standard. Malonaldehyde formation was measured asthe thiobarbituric acid reactive products (16). Materials-FCCP was purchasedfromBoehringer Mannheim GmbH,Mannheim,FederalRepublic of Germany. Arsenazo 111, bovineserumalbumin (crystallized), Hepes, EGTA, t-butyl hydro- peroxide, GSH, and dithiothreitolwere purchased from Sigma. The ionophore A23187 was purchased from Calbiochem-Behring. Other chemicals were at least of reagentgradeandpurchasedlocally. Deionizedwater was usedforallsolutions,exceptforFCCP and A23187, which were dissolved in dimethyl sulfoxide. RESULTS A N D DISCUSSION Measurement of exchangeable Ca2+ in the endoplasmic reticulum and the activity of the Ca2+translocase has previously been performed using isolated microsomes and either 45Ca2+as a tracer (5, 6) or direct determination of Ca2+by atomic absorption spectroscopy (9). A sensitive and rapid approach to measurement of Ca2+in biological systems involves the use of the metallochromic indicator for Ca2+,arsenazo I11 (14).We have therefore used this compound to develop a new assay for Ca2+sequestration by liver microsomes. Addition of microsomes to solutions of arsenazo I11 had no effect on the visible absorption spectrum or extinction coefficient for the Ca2+.arsenazo I11 complex. Furthermore, the dissociation constant for this complex was unaffected by addition of microsomes, and thus, the free Ca2+concentration could be obtained from the mass action equation, [Ca2+free] [arsenazo 1111 KO= [Caz+.arsenazo III] where KO is 11.3 p i , [Ca2+.arsenazo1111was determined by addition of a small excess of EGTA, and total [arsenazo 1111 was determined by adding excess Ca2+.To measure changes in Ca2+content of microsomes, it was convenient to construct a plot of L L ~as a~fun~ctio-n of~added Ca2+so that absorb- ance changes couldbe directly related to changes in Ca2+ content (see below). Freshly prepared microsomes (10-15mgof protein/ml) which had been washed with 5 IJM EGTA and maintained on ice contained the equivalent of 1-2nmol of Ca2+/mgof protein, of which 20-40% was detectable only after addition of the ionophore A23187. Dilution of microsomesto 1mg ofprotein/ ml and incubation at room temperature for 10-15 min or at 37 "C for 5-10 min resulted in the loss of most of the A23187releasable fraction. Consequently,preincubation was routinely performed to deplete this Ca2+fraction prior to measurement of ATP-dependent Ca2+uptake. Addition of 3 mM ATP to preincubated microsomes caused a rapid increase in the A23187-releasableCa2+pool (Fig. lA), which was linear for 90 s and reached a maximal value at 120s. The amount of Ca2+sequestered (Fig. 10 was calculated from the experimentally derived standard curve (Fig. 1B)and was 0.61 f 0.12 m o l of Ca2'/mg of protein/min for the f i s t minute and 1.03 f 0.20 nmol of Ca2+/mgof protein for the total uptake (n = 6). In microsomes which were not washed with EGTA, ionophore-releasable Ca2+was typically 2-3 nmol/mg of protein. Assuming 25% ofthe cellular protein is from the endoplasmicreticulum in cellsfrom phenobarbitalpretreated rat, this would indicate that 0.5-1 m o l of Ca2+/ mg ofprotein should be releasable in hepatocytes. This agrees well with the values of about 1-1.1 nmol of Ca2+/106c e b (10) for the A23187-releasable pool in hepatocytes which is not released by FCCP treatment when one considers that there are 1.6 mg of protein/106 cells (17). The characteristics of ATP-dependent Ca2+sequestration as measured by this method are similar to those obtained with other approaches (5,18). The uptake was insensitive to ruthenium red, an inhibitor of mitochondrial Ca2+uptake, and the apparent K,,,for ATP (about0.1-0.2 mM) was somewhat lower than other values (5, 6, 18).A t the Ca2+levels present in our system, the sequestration of Ca2+was relatively insensitive to added Ca2+and was therefore consistent with an apparentK,,, of about 1 p~ total Ca2+.In contrast to these similarities, however, we observed that the microsomal ATP-dependent Ca2' sequestration was insensitive to theprotonophore FCCP (Table I). These experiments were performed both by incu- Downloaded from www.jbc.org by guest, on March 29, 2010 FIG. 1. Measurement of the ATP- dependent microsomal Ca2+seques- tration by the A23187 release method with arsenazo 111. A, micro- somes (0.8mg of protein/ml)were prein- cubated for 20 min at room temperature toreleaseresiduaslequestered Ca". (nATP (3 m ~ w)as addedpriortothe ionophoreA23187 at theintervals indicated, and arsenazo I11 was added30 s priortoaddition ofA23187.Excess EGTAwasadded as indicatedtogive zero Ca2+andthus a measure of total Ca2+.arsenazo I11 in thesystem. B, AA-W was calibrated relative to addi- tions of known amounts of Ca2+to mi- crosomes,whichhad been depleted of Ca2+by addition of EGTA.Thiscurve allowed simple calculationof the amount and rate of CaZ+sequestration. C,Ca2+ sequestration as a function of time. Data are from a single experiment represent- ative of data from six microsomal prep- arations. +ATP-ATP A 30s. 60s. 90s. 120s 210 s 300 5 12 Total Ca" added Inmol/ml) tlme Isec) 6392 Microsomal Ca2+Sequestration Downloaded from www.jbc.org by guest, on March 29, 2010 bating with FCCP before adding ATP and by adding FCCP after ATP-dependentCa2+uptake hadoccurred and provided no evidence for a sensitivity of this system to therespiratory uncoupler. Recent studies in our laboratory have shown that theme- tabolism of t-butyl hydroperoxide by isolated hepatocytes can lead to alteredCa2+homeostasis (10-12). In these studies, two intracellular Ca2+pools weredistinguished by treatment first with FCCP and then ionophore A23187 in the presence of arsenazo 111.The FCCP-releasable pool has been identified as being mitochondrial in origin. The remaining extramitochondrial Ca2+,released by subsequent addition ofA23187, was thought to be located predominantly in the endoplasmic reticulum, largely on the basis of subfractionation studies (9). g 10 C _L nn The results using the present approach show that the Ca2+ sequestering capacity of the microsomal fraction is quantita- tively sufficient to account for virtually all of the "extramitochondrial" Ca2+of liver cells. Effects of &ButylHydroperoxide on Microsomal Ca2+Sequestration-&Butyl hydroperoxide is rapidly metabolized by glutathione peroxidase in liver cells, resulting in the oxidation of GSH and pyridine nucleotides, as well as a loss in iono- FIG. 3. Effect of Mna+ on the t-butyl hydroperoxide (t-BH)dependent initiation oflipid peroxidation (A)and Cas+sequestration (B)in rat liver microsomes. Lipid peroxidation was mea- sured interms of malonaldehyde formation (16),and Ca2+sequestra- phore-releasable Ca2+and, a t high concentrations (4m ~ )c,ell tion was measured as described in the legend to Fig. 1. death (10). Incubation of microsomes with t-butyl hydroperoxide resulted in rapid loss of their ability to sequester Ca" in uptake and release processes. To test whether t-butyl hydrothe presence of ATP (Fig. 2 A ) . This lossin activity was peroxide altered the microsomal ability to retainCa", microreflected both in the rate and the maximal amount of Ca2+ somes, whichhad notbeen preincubated to deplete Ca", were sequestered (Fig. 2B). Both the net rate of uptake and the incubated with and without the hydroperoxide, and the rate amount sequestered are a function of the balance of the of loss of the A23187-releasable pool was measured. These results indicate that the hydroperoxide had no effect on the TABLEI Effects of inhibitors on microsomal Ca" sequestration Liver microsomes were incubated for 10 min as described under "ExDerimental Procedures" with additions as indicated. Additions Ca2+sequestered nmol/mgprotein None FCCP (10p ) Ruthenium red (20p ) Metyrapone (100 p ~ ) MnC12 (250 PM) &Butylhydroperoxide (200 p ) &Butylhydroperoxide + Metyrapone (100p ) t-Butyl hydroperoxide + MnC12 (250pM) 0.98 f 0.05 0.85 f 0.07 0.82 f 0.04 0.84 f 0.03 *0.81f 0.08 0.13 0.09 *0.61 f 0.07 0.17 0.05 net rate of Ca2+loss from microsomes in short term incubations in the absence of ATP, but that atlonger time points an increased release was observed, especially in microsomes preincubated with ATP(datanot shown). Therefore, we presently cannot distinguish between a possible inhibition of uptake as opposed to a stimulation of release. Indeed, both processes may occur simultaneously. t-Butyl hydroperoxide is largely metabolized by glutathione peroxidase but is also a substrate for cytochrome P-450 (19). Investigation of this activity of cytochrome P-450 indicates that a free radical is generated (19)which couldaccount for the observed effectof the hydroperoxide on Ca2+sequestration. Addition of metyrapone, a well known inhibitor of cytochrome P-450, prevents the inactivation of Ca2' sequestration by t-butyl hydroperox- ide (Table I).To distinguish between this process and radical- mediated lipid peroxidation,experiments were performedwith added Mn2+,an inhibitor of free radical-mediated lipid per- oxidation? The results (Fig. 3) show that Mn" inhibits mal- onaldehyde production following t-butyl hydroperoxide treat- ment but has little effect on the hydroperoxide-induced loss ofCa2' sequestration. These results suggest that the inacti- vation of Ca2+sequestration may be a direct effect of a free radical product on the Ca2+pump and not mediated through a generalized peroxidation of membrane lipids. The loss of the A23187-releasableCa2+pool in hepatocytes following t-butyl hydroperoxide treatment occurs only after GSH depletion and can be prevented by dithiothreitol (10). 10 llrne ( m m l 2b tlrne(sec1 FIG. 2. Inhibition of microsomal Cas+ sequestration by tbutyl hydroperoxide. A, microsomeswere incubated with 200 p tbutyl hydroperoxide for 1,5,10,and 18 minprior to addition of ATP. 90s after addition of ATP, arsenazo I11 was added, and after 30 s, the "4.A due to addition of A23187 was recorded. Time points are, therefore, expressed as the totalincubation time after addition of the We therefore examined the effect of GSH and dithiothreitol on ATP-dependent Ca2+sequestration by microsomes. Incubation of microsomes with GSH or dithiothreitol did not change the ability of the microsomes to sequester Ca2+in the presence of ATP(data not shown). However, addition of either of these thiols to microsomes incubated with 200 pM tbutyl hydroperoxide provided a complete protection against loss of ATP-dependent Ca2' sequestration for up to 30 min hydroperoxide. B, time course of Ca2+sequestration following addi- (Fig. 4). This protection is probably not due to glutathione tion of ATP to microsomes incubated 5 min either without (0)or with (02)00 p~ t-butyl hydroperoxide. S . Orrenius and L. Emster, unpublished results. Microsomal Ca2+Sequestration 6393 AB REFERENCES iIk L tt "l oIW "I tI st -t 2 O4 L sotw t " f L t 2ow FIG. 4. Prevention ofthe t-butylhydroperoxide inhibition of microsomal Ca2+sequestration by GSH and dithiothreitol. Microsomes (1mg of protein/ml) were incubated at 37 "C for 15 min with the treatmentsindicated prior to addition of A T P and measure- ment of the A23187-releasable Ca*'. A , control; B , 200 ,UMt-butyl hydroperoxide; C, 200 PM t-butyl hydroperoxide and 2 mM dithiothre- itol; D, 200 ,UMt-butyl hydroperoxide and 1 mM GSH. Ionophore A23187 (0and EGTA were added as indicated by the arrows. 1. Bygrave, F. L. (1978) Biol. Rev. Camb. Philos. SOC.53,43-79 2. Carafoli, E., and Crompton, M. (1978) Curr. Top.Memb. Transp. 10,151-216 3. Becker. G. L.. Fiskum. G.. and Lehnine"er,. A. L. (.1980.) J. Biol. Cheh. 255,'9009-9012 ' 4. Moore, L., Fitzpatrick, D. F., Chen, T. S., and Landon, E. J. (1974) Biochim. Biophys. Acta 345,405-418 5. Moore, L., Chen, T., Knapp, H. R., Jr., and Landon, E. J . (1975) J. Biol. Chem. 260,4562-4568 6. Ponnappa, B. C., Dormer, R.L., and Williams, J. A. (1981) Am. J. Physiol. 240, G122-G129 7. Bruns, D. E., McDonald, J. M., and Jarett, L. (1976) J. Biol. Chem. 251,7191-7197 8. Watson, E. L., and Siegel, I. A. (1977) Biochem. Pharmacol. 26, 125-127 9. Murphy, E., Coll, K., Rich, T. L., and Williamson, J. R. (1980) J. Biol. Chem. 255,6600-6608 10. Bellomo, G., Jewell, S. A., Thor, H., and Orrenius, S . (1982)Proc. Natl. Acad. Sci. U. S. A . 79,6842-6846 peroxidase activity associated with the microsomes because 11. Jewell, S. A., Bellomo, G., Thor, H., Orrenius, S., and Smith, M. dithiothreitol is not a substrate for glutathione peroxidase T. (1982) Science (Wash.D. C . )217, 1257-1259 (20). Previous studies have shown that the microsomal Ca2+ pump is sensitive to sulfhydryl reagents (5), and we also found that 1 m~ diamide inhibited its activity (data not shown). It has further been shown by Moore and co-workers (5, 21-23) that the microsomal Ca2+pump is inhibited during carbon 12. Thor, H., Smith, M. T., Hartzell, P., Bellomo, G., Jewell, S. A., and Orrenius, S. (1982)J . Biol. Chem. 257, 12419-12425 13. Emster, L.,Siekevitz, P., and Palade, G. E. (1962) J. Cell Biol. 15,541-562 14. Kendrick, N. C., Ratzlaff, R. W., and Blaustein, M. P. (1977) Anal. Biochem. 83,433-450 tetrachloride andcarbon disulfide metabolism in liver micro- 15. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. somes. It therefore appears that thiols, notably GSH in liver (1951)J.Biol. Chem. 193,265-275 cells, may protect the Ca2+sequestering system of the endo- 16. Smith, M. T., Thor, H., Hartzell, P., and Orrenius, S. (1982) plasmic reticulum by preventing the oxidation of thiol group(s) critical for Ca2+-ATPaseactivity. Biochem. Pharmacol. 31, 19-26 17. Jones, D. P., and Mason, H. S. (1978)J.Biol. Chem. 253, 4874- 4880 These resultsare consistent with our recent finding that the 18. Bygrave, F.L. (1978) Biochem. J. 170,87-91 extramitochondrial Ca2+pool is regulated in hepatocytes by 19. O'Brien, P. J., and Rahimtula, A. D. (1975)J. Agric. Food Chem. the glutathione redox state and is susceptible to alteration by 23, 154-158 oxidative stress (10-12). The Ca2+sequestering system of the liver endoplasmic reticulum may be a primary target in both hepatotoxic injury and in the physiological response of liver cellsto drugs and hormones. Its furthecrharacterization therefore seems an area worthy of future investigation. 20. Wendel, A. (1981) in Enzymatic Basis ofDetorication (Jakoby, W. B., ed) pp. 333-353, Academic Press, New York 21. Moore, L., Davenport, G. R., and Landon, E. J. (1976) J. Biol. Chem. 251, 1197-1201 22. Moore, L. (1980)Biochem. Pharrnacol. 29, 2505-2511 23. Moore, L. (1982)Biochem. Pharmacol. 31, 1465-1467 Downloaded from www.jbc.org by guest, on March 29, 2010