Toxicology, 77 (1993) 115-121 115 Elsevier Scientific Publishers Ireland Ltd.

Effects of the antioxidant butylated hydroxyanisole on cytosolic free calcium concentration

Michae l Dav id a, G y u l a H o r v a t h b, I n g o l f Sch imke ¢, Ma th i a s M. Muel le r a a n d Ivan N a g y b

a2nd Department of Surgery, Division of Clinical Biochemistry, University of Vienna (Austria); bDivision of Clinical Chemistry, Heim Pal Pediatric Hospital, Budapest (Hungary) and CDepartment of Clinical and

Pathological Biochemistry, Humboldt-University, Berlin (Germany)

(Receive d July 2nd, 1992; accepted September 29th, 1992)

Summary

The effect of the antioxidant butylated hydroxyanisole on free intracellular calcium was investigated in rat cardiomyocytes, rat pituitary ceils, human umbilical vein endothelial cells, granulocytes and BHK- ceils. The effects of incubation time and butylated hydroxyanisole (BHA) concentrations were examined with respect to the extracellular calcium concentration. Butylated hydroxyanisole increased the free cytosolic calcium concentration in a dose- and time-dependent manner irrespective of the presence of ex- tracellular calcium. Differences between all cell types were observed in the sequel of the re-uptake of the calcium ions released by BHA. Dependence on the concentration of extraceilular calcium on the extent of sequestration was seen in rat pituitary cells and cardiomyocytes only.

Key words." Butylated hydroxyanisole (BHA); Antioxidant; Cytosolic calcium; Endothelial cells

Introduction

The phenolic compound 2,6-di-tert-butyl-4-methoxy-phenol (butylated hydroxy- anisole; BHA) is a widely used antioxidant in the food industry as well as in chemical and pharmaceutical preparations for its activity as a free radical scavenger. BHA is also very commonly used as an antioxidant in tissue culture experiments, although usually higher concentrations than in food or pharmaceutical preparations are added. In addition to its actions as an antioxidant, BHA has been shown to alter a number of biological processes including inhibition of the aggregation and ATP secretion in human platelets [1], suppression of the T-lymphocyte proliferation and IL-2 synthesis [2], blocking of GABA exocytosis in brain synaptosomes [3], inhibi- tion of the Ca2÷-Mg2÷-ATPase in skeletal muscle [4] and disturbance of membrane potential, swelling and release of calcium from isolated mitochondria [5]. To per-

Correspondence to." Michael David, National Institutes of Health, Division of Cytokine Biology, Building 29A, Room 2B-01, 8800 Roekville Pike, Bethesda, MD 20892, USA.

0300-483X/93/$06.00 © 1993 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

116

form an entire inhibition of lipid or protein oxidation, we recently used this com- pound as an 'antioxidant' in experiments to investigate the role of peroxide-induced increase of intracellular free calcium levels (unpublished data). To our surprise we found that BHA itself caused dramatic changes in free cytosolic calcium concentra- tions. As cytosolic calcium regulates many cellular processes, we investigated the effects of BHA on intracellular free calcium concentration in the presence and absence of extracellular calcium. To determine whether the actions of BHA were cell or species-specific, experiments were performed using various cell types from dif- ferent species.

Materials and methods

Reagents

Human fibronectin and hypothalamic growth factor were obtained from Biomedica. The other cell culture reagents were purchased from Gibco. All other substances were ordered from Sigma Chemicals.

Cell harvesting and culture

Endothelial cells Endothelial cells were isolated from human umbilical veins and cultured accor-

ding to the procedure of Jaffe et al. [6]. Briefly, fresh human umbilical veins were filled with 0.1% collagenase solution and incubated at 37°C for 5 min. Then the veins were perfused with Medium 199 containing 20% human serum. Cells were collected from the perfusate by centrifugation at 800 x g for 5 min and seeded onto T-75 flasks precoated with human fibronectin. Cells were cultured in M 199 medium containing 20% human serum, 100 000 units/1 penicillin and 100 000 /zg/1 strep- tomycin, 100 000 units/1 low molecular weight heparin and 30 rag/1 bovine hypothalamic growth factor. The confluent primary monolayers ( - 8 x 106 cells/flask) were trypsinized, transferred into three T-75 culture flasks and cultured for 3 days. Only cells from these first subcultures were used for the experiments described below. The cells were identified as endothelial cells by their typical cob- blestone, contact inhibited morphology [7] and by factor VIII (FVIII-vWF) staining [8].

Myocytes Rat cardiomyocytes were harvested by trypsin treatment of hearts obtained from

1- to 3-day-old rats. The collected cells were cultured in a selection medium in order to eliminate fibroblasts according to Wollenberger et al. [9]. The cells were used 3 days after preparation when they were identified as myocytes by synchronous beating in the culture. Adherent cells were resuspended for measurement as describ- ed below.

Granulocytes The cells were collected by multiple density centrifugation of human blood as

described by Mueller et al. [10] and experiments were carried out immediately.

117

BHK-cells Baby hamster kidney (BHK) cells were derived from the established cell line M

300 from the Institute of Immunology, Humboldt-University, Berlin and used 3 days after inoculation of the culture flask. The adherent, confluent cells were trypsinized prior to the calcium measurement.

Rat pituitary cells Rat anterior hypophysis were taken from adult female rats of - 200 g body weight

and trypsinized [11]. Experiments with the cell suspension were performed within 6h.

Determination of intracellular free Ca 2+

After trypsinization, cell suspensions were centrifuged for 10 min at 800 x g and the pellets were resuspended in HBSS (Hepes-buffered salt solution: 140 mM NaC1, 3.9 mM KCI, 0.5 mM Na2HPO4, 0.7 mM KH2PO4, 11 mM glucose, 0.5 mM MgCI2, 2.2 mM CaCI2 and 20 mM Hepes (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid)). FURA-2/AM was added to a final concentration of 10 #M. Cell suspensions were incubated for 30 min at 37°C with occasional agitation. After rinsing twice with HBSS the cells were resuspended to a concentration of 5-6 x 10 6 cells/ml. After addition of 20 #1 100-fold concentrated BHA-solution (in 70% ethanol, concentra- tions see Fig. 1) fluorimetric measurements were performed with 200 #1 of cell suspension. The fluorescence of the FURA 2 loaded cells was measured with a Perkin Elmer 3000 fluorescence spectrophotometer (excitation: 342 nm; emission: 492 nm) equipped with a stirring device. For each sample, calibration was performed by the addition of 50 #1 EGTA solution (0.4 M EGTA in 1.5 M Tris), 20 #1 Triton (10o/o Triton X-100) and 100 #1 CaCl2-solution (1 M). The Ca2+-concentrations

[Ca]i (nmol/I)

1200-

1000-

800"

600"

400"

200"

0 0.0 015 110 115 210

BHA-Concentration (mmol/I)

Fig. 1. Concentration-dependent increase of cytosolic free calcium level after addition of BHA to endo- thelial cells (peak values, mean ± STD).

i18

were determined and calculated from the equation:

[Ca2+]i = K d x ( F - Fmin)/(Fmax - F)

as described previously [12]. Experiments performed in Ca2+-depleted buffer were done in the presence of 2 mM EGTA in Ca2+-free HBSS.

Statistical analysis

Data are given as mean ± SD of three experiments unless otherwise indicated. For determination of significance the results were subjected to a matched-pairs t-test.

R ~

The addition of BHA to endothelial cells resulted in a concentration-dependent increase of cytosolic calcium (Fig. 1). Other cell types showed similar effects (data not shown). Maximal increases in cytosolic calcium concentrations were approx- imately the same in each cell type (Table I) and occurred - 2 0 s after addition of BHA (Fig. 2).

However, various cell types showed large differences in the rate of decay in max- imal cytosolic calcium concentration (Fig. 2/1-5). Free cytosolic calcium levels in endothelial cells and granulocytes returned nearly to the basal concentrations within 10 min in the continued presence of BHA. Under the same experimental conditions, elevated calcium concentrations in BHK cells declined below resting levels after l0 min whereas in rat cardiomyocytes and pituitary cells intracellular free calcium con- centration remained approximately at peak values with a slight tendency to a further increase.

The influence of extracellular calcium on the actions of BHA was examined by the addition of EGTA before the addition of BHA (Fig. 2/6-10). Peak cytosolic calcium concentrations were reached after the same incubation time with BHA and were in

TABLE I

PEAK VALUE OF CYTOSOLIC CALCIUM CONCENTRATION (nM) AFTER ADMINISTRA- TION OF 1 mM BHA IN THE PRESENCE AND IN THE ABSENCE OF EXTRACELLULAR (ext) CALCIUM

Mean ± STD, n in parentheses, *p < 0.0001 vs. basal value.

Cell type Basal Ca(ext ) = 2.2 mM Ca(ext ) = 0.0 mM

Endothelial 123.9 ± 14.1 (13) 1137.4 ± 210.1 (9)* 1065.8 ± 87.3 (4)* Cardiomyocytes 125.0 ± 6.0 (6) 885.2 ± 18.8 (3)* 903.5 ± 74.8 (2)* Pituitary 133.8 ± 4.5 (14) 902.3 ± 14.8 (8)* 866.4 ± 12.6 (6)* BHK 130.3 ± 4.8 (4) 993.7 ± 14.4 (2)* 961.5 ± 27.8 (2)* Granuiocytes 112.3 ± 6.8 (4) 1173.6 ± 31.5 (2)* 1092.4 ± 63.2 (2)*

Ca*~= = 2.2 mM ÷ +

Caext = 0 mM

119

Pituitary cells

Cardiomyocytes

o f

Endothelial cells

Granulocytes

BHK-CelIs

1000

-130

70

nM 5 min

Fig. 2. Different behaviour of cytosolic calcium concentration after addition of 1000 ~M BHA to several cell types in the presence (1-5) and the absence (6-10) of extraceilular calcium.

the same range as in those experiments performed in the presence of 2.2 mM ex- tracellular calcium. However, the effects of BHA were cell-specific in so far as that free cytosolic calcium decreased below basal concentrations in the continuous presence of BHA in pituitary and BHK cells; in endothelial cells and granulocytes it returned nearly to resting levels, whereas in myocytes it remained elevated.

120

Furthermore, peak values obtained within these experiments were not altered by the presence of 0.2 mM LaCI 3 or nifedipine. Addition of CoC12 and BHA to the measurement buffer resulted in the same effects as the addition of EGTA and BHA (data not shown).

Discussion

Alexandre et al. [1] demonstrated that the lipophilic antioxidant BHA in the range 20-200/~M inhibits the thrombin-evoked increases of free intracellular calcium level in a concentration-dependent manner. We found that BHA on its own was able to induce an enhancement of cytosolic calcium in a concentration-dependent manner within the range 250 #M to 2 mM. Furthermore, this effect of BHA is not cell- or species-specific. In an additional series of experiments we evaluated the influence of extracellular calcium on peak values and the steady-state of free intracellular calcium. The maximal value reached after stimulation with 1 mM BHA was not altered when extracellular calcium was complexed by EGTA whereas significant dif- ferences in the rate of calcium sequestration were observed in four of the five cell types used. The presence of the membrane-located calcium channel blockers La 3÷- ions or nifedipine was without effect on this phenomenon (data not shown). These results indicate that the increased cytosolic calcium concentrations evoked by BHA are caused only by a release from intracellular stores. This fact confirms and extends the findings~of Thompson et al. [5] who found isolated mitochondria release calcium after BHA treatment; although the similar compound BHT has been reported to have a protective rather than an uncoupling effect on mitochondria by preventing the increase in permeability induced by swelling agents [13]. The rate of the calcium enhancement caused by BHA is very rapid and consistent so that we suppose an ionophore.-like mechanism of action. Stimulation of protein kinase C, which also causes a dramatic increase in cytosolic calcium concentrations [14], could also be another possible mechanism of BHA action [15]. An increase of cytosolic calcium mediated via ATP depletion of the mitochondria or by inhibition of membrane ATPases [4] seems unlikely as it would (i) take a longer time and (ii) be irrevers- ible. Nevertheless, the blocking of the ATPases may be involved in the subsequent sequestration of BHA-released calcium. BHA is known to be a general inhibitor of Ca2+-Mg2+-ATPase activity [4]. Therefore, different membrane ATPase activities in different cell types may account for the subsequent cell-specific rates of calcium se- questration after BHA treatment. In Fig. 2 we rank the cell types we used according to their ability to respond to physiological stimuli. We believe that the cells are able to re-uptake the intracellular calcium ions released from intracellular stores after BHA treatment but cannot overcome additional calcium influxes from the ex- tracellular medium by membrane leakage - - which is part of the membrane calcium flux - - since their membrane ATPases are blocked. In myocytes and pituitary cells, known to have voltage-dependent channels and a higher membrane circulation of Ca 2÷, the slight decrease after the peak value is followed by a further increase (Fig. 2/1-2). This rise may be due to the inhibition of the ATPase-dependent Ca2+-pump leading to an impossibility to counteract against the high membrane leakage. This proposal is confirmed by the fact that the second increase was not observed in the

121

absence of extracellular calcium or in the presence of Co2+-ions which are known to be potent blockers of membrane calcium channels. The delay of the enhanced calcium levels in cardiomyocytes also in the absence of extracellular calcium may be due to the high calcium storage capacity of this cell type. The other cells used were able to sequester enhanced free intracellular calcium concentrations with varying kinetics and to different extents.

Acknowledgments

We wish to thank Drs. Haberland, Heidecke, Mueller, Grunow and Mrs. Barabas for help in cell preparation; Drs. Makara and Gevai for discussions and Dr. Lamer for reviewing this manuscript.

References

1 A. Aiexandre, M.G. Doni, E. Padoin and R. Deana, Inhibition by antioxidants of agonist evoked cytosolie Ca 2+ increase, ATP- secretion and aggregation of aspirinated human platelets. Biochem. Biophys. Res. Comm., 139 (1986) 509.

2 J. Dornand and M. Gerber, Inhibition of murine T-ceU response by antioxidants: the targets of lipo- oxygenas¢ pathway inhibitors. Immunology, 68 (1989) 384.

3 F. Zoccarato, M. Pandolfo, R. Deana and A. Alexandre, Inhibition by some phenolic antioxidants of Ca 2+ uptake and neurotransmitter release from brain synaptosomes. Biochem. Biophys. Res. Comm. 146 (1987) 603.

4 P.M. Sokolove, E.X. Albuquerque, F.C. Kauffman, T.F. Spande and J.W. Daly, Phenolic antioxi- dants: potent inhibitors of the (Ca 2+ + Mg2+)-ATPase of sarcoplasmatic reticulum. FEBS Lett., 203 (1986) 121.

5 D. Thompson and P. Moldeus, Cytotoxicity of butylated hydroxyanisole and butylated hydrox- ytoluenein isolated rat bepatocytes. Biochem. Pharmacol., 37 (1988) 2201.

6 E.A. Jaffe, R.L. Nachman, C.G. Becket and C.R. Minck, Culture of human endothelial cell derived from umbilical veins. J. Clin. Invest., 52 (1973) 2745.

7 C. Haudenschild, R.S. Cotran, M.A. Gimbrone and J. Folkman, Fine structure of vascular¢ endo- thelium in culture. J. Ultrastruct. Res., 50 (1975) 22.

8 E.A. Jail'e, L.W. Hoyer and R.L. Nachman, Synthesis of antihaemophilic factor antigen by cultured endothelial cells. J. Clin. Invest., 52 (1973) 2757.

9 A. Wollenberger, The cultured myocardial cell as a model in heart research, in H. Abe, M. Tada and L.H. Opie (Eds.), Regulation of cardiac function; molecular, cellular and patho-physioiogical aspects, Japan Scientific Societus Press, Tokyo, 1984, p. 269.

10 G.M. Mueller, R. Baehr, E. Seidler and H. Tanzmann, Mikrotechnik zur Bestimmung der stimulierten Superoxid-Bildung dutch Phagozyten. Z. Med. Lab. Diagn., 28 (1987) 185.

11 Gy. Rappay, A. Gyevai, L. Kondics and E. Stark, Growth and fine structure of monolayers derived from adult rat adenohypophysal cell suspensions. In Vitro, 8 (1973) 301.

12 G. Grynkiewicz, M. Poenie and R.Y. Tsien, A new generation of Ca 2+ indicators with greatly im- proved fluorescence properties. J. Biol. Chem., 260 (1985) 3440.

13 D. Carbonera and G.F. Azzone, Permeability of inner mitochondrial membrane and oxidative stress. Biochim. Biophys. Acta, 943 (1988) 245.

14 A. Griesmacher, G. Weigel, M. David, Gy. Horvath and M.M. Mueller, Functional implications of cAMP and Ca 2+ on prostaglandin 12 and thromboxane A 2 synthesis by human endothelial cells, Arterioscler. Thromb., 12 (1992) 512.

15 M. Ruzzen¢, A. Donella-Deana, A. Alexandre, M.A. Francesoni and R. Dcana, The antioxidant butylated hydroxytoluene stimulates platelet protein kinase C and inhibits subsequent protein phosphorylation induced by thrombin. Biochim. Biophys. Acta 1094, (1991) 121.