Toxicology, 13 ( 1992) 179- I89 Elsevier Scientific Publishers Ireland
179 Ltd
Protective effect of diosmetin on in vitro cell membrane damage and oxidative stress in cultured rat hepatocytes Pia Villa”, Dario Covaa, Laura De Francescob, Giuseppina Palladini” and Raffaella
Amalia Peregoa
Guaitanib,
“C. N. R. (National Research Council) Center ofCytophurmucology, Depurtmvn~ of’ Phrrrmucology. Viu Vanvitelli 32, 20129 Milan and hExperimentul Liver Toxicology Unit, Isriruio di Ricrrche Furmucologichc~ Mario Negri. Viu Eritrecc 62, 20157 Milun (Italy) (Received
December
12th. 1991; accepted
April 4th. 1992)
Summary Primary cultures of rat hepatocytes were used to study the effects of the flavonoids diosmin and its main metabolite diosmetin on the cell membrane damage caused by erythromycin estolate (EE) and oxidative stress caused by ferr-butylhydroperoxide (TBHP). The damage was evaluated by the leakage of intracellular enzymes lactate dehydrogenase, aspartate-aminotransferase and the residual cell content of a lysosomal marker acid phosphatase CAP). After treating the cells for 40 h with diosmetin EE induced less enzyme leakage. The content of AP was kept higher by diosmetin pretreatment after 6 h exposure to EE. Diosmin at the same concentrations had barely any effect. Diosmetin, but not diosmin, also protected against TBHP toxicity and this was related to lower lipid peroxidation and higher glutathione content caused by pretreatment with the flavonoid. When the cells were treated simultaneously with TBHP and diosmetin after 21 h of culture, the protection by the flavonoid was even higher. In fact the antioxidant activity of diosmetin was considerably greater than that of diosmin. After 40 h exposure to both flavonoids diosmin but not diosmetin was detectable in the cell membrane fraction, suggesting that the latter’s protective effect is associated with its metabolites. Key words: Flavonoids;
Diosmin;
Diosmetin;
Cultured
hepatocytes;
Hepatotoxicity;
Glutathione
Introduction
Flavonoids are natural compounds widely found in edible plants. Some are claimed to have biological and pharmacological properties [ 1] mainly related to their effects on the production of oxygen-free radicals [2-41 and/or arachidonic acid metabolites [3,5,6]. Diosmin (Fig. 1), a semi-synthetic flavonoid utilized in the treatment of various venous diseases, inhibited the production of oxygen radicals in vivo Correspondence to; Pia Villa, Experimental l-20157 Milan, Italy.
0300-483XI92/$05.00 0 1992 Elsevier Scientific Publishers Printed and Published in Ireland
Liver Toxicology
Ireland
Ltd.
Unit, Istituto
Mario Negri, Via Eritrea
62,
180
OH \O
OIOSMIN
HO
OH
Fig. 1. Molecular
structures
of diosmin
0
and diosmetin
and in vitro [4] and reduced the inflammatory reaction in the rat by affecting the synthesis of prostaglandins (PGE*, PGFz,) and thromboxane (TX B2) [5]. Therefore flavonoids can act as antiradicals and antioxidants with a consequent stabilizing effect on cell membranes and intracellular structures. Flavonoids such as rutin and catechin influence the fragility of lysosomes of rat liver in vitro [7] and in vivo after intoxication by galactosamine and ethanol [8]. Studies with cultured rat hepatocytes have shown the protective effect of some flavonoids, catechin, silybin [9] and silymarin [lo] against toxic insult by various compounds. In order to gain insight into the effects of diosmin and its aglycon, diosmetin (Fig. l), on cell membrane damage and oxidative stress we used the model of rat hepatocytes in primary culture. Two compounds, erythromycin estolate (EE) and tert-butyl hydroperoxide (TBHP), which damage liver cells by different mechanisms, were used for the study. EE is a macrolide antibiotic which produces hepatic injury in vivo and in vitro through a hypersensitive mechanism and a membrane surfactant effect [l l- 131. TBHP is an organic hydroperoxide which causes peroxidation of membrane lipids [14,15], disturbances in intracellular Ca*+ homeostasis and the glutathione system, leakage of cytoplasmic enzymes and cell death [ 14- 171. Hepatocytes were exposed to each toxic compound together with diosmin or diosmetin after 21 h of culture or after 40 h treatment with either diosmin or diosmetin. The cytotoxic effects of EE and TBHP were evaluated by leakage of intracellular enzymes. After exposure to flavonoids their content was measured in the cells and correlated to their pharmacological effect.
181
Materials and methods Chemicals Diosmin and diosmetin were supplied by Geymonat S.p.A. (Anagni, Fr., Italy). Erythromycin estolate was a gift from Pierre1 (Milan, Italy), tert-butyl hydroperoxide was from Aldrich-Chemie (Steinheim, Germany) and butylated hydroxyanisole from Sigma Chemical Co., St. Louis, MO. All other chemicals were analytical grade. Animals Male Crl:CD(SD)BR rats weighing 180-200 g, fed an ‘open formula’ diet ad libitum and housed under controlled conditions (22 f 0.5”C, 55% relative humidity, 12/12 h light/dark cycle), were used as liver donors. Experimental procedures with animals were reviewed and approved by the institutional animal care and use committee, in compliance with institutional guidelines that adhere to the principles found in the ‘Guide for the Care and Use of Laboratory Animals’, Publication No. 85-23. NIH, Bethesda, MD (revised 1985). Isolation and primary culture of hepatocytes Hepatocytes were isolated by perfusing the liver with 0.033% w/v collagenase solution, according to Seglen’s technique [ 181, under pentobarbital sodium (50 mg/kg body weight intraperitoneally (i.p.)) anesthesia. Cell viability was tested by trypan blue exclusion and was 87 f 1%. Hepatocytes were seeded at a density of 0.85 x lo6 cells per 9-cm2 Petri dish in 1.5 ml of standard medium with 5% fetal calf serum (FCS) and cultured at 37°C in a 5% CO*-95% air humidified atmosphere. The standard medium consisted of a mixture of 75% minimum essential medium and 25% medium 199 containing 0.2% bovine serum albumin (BSA), 1 &ml bovine insulin, 50 &ml kanamycin, 50 &ml streptomycin and 50 I.U./ml penicillin [12,13]. The medium, without FCS and with 7 x 10m5M hydrocortisone hemisuccinate, was renewed 3 and 21 h after hepatocyte seeding in the absence and presence of flavonoids. Culture treatments Diosmin and diosmetin were dissolved in dimethylsulphoxide (DMSO) and 0.5 N NaOH. Two treatment schedules were used: (i) the flavonoids (12.5 and 25 PM) were added with the toxic substances 21 h after cell seeding and (ii) the flavonoids (12.5, 25 and 50 PM) were added 3 and 21 h after hepatocyte seeding for a total exposure of 40 h until treatment with the toxic compounds started. The untreated controls were exposed to the same DMSO concentration (0.05% v/v) as the treated cultures. EE (100 PM), dissolved in DMSO, and TBHP (0.5 mM) were added to the cells diluted with culture medium without BSA after either 21 h or 43 h of culture for 6 h and 50 min, respectively. After different incubation periods (2, 4 and 6 h for EE and 30, 40 and 50 min for TBHP) measured amounts of culture medium were
182
withdrawn, centrifuged to remove cell debris and placed on ice for enzymatic analysis. The cells were washed twice with saline and immediately frozen. They were then harvested by scraping with 0.2% w/v Triton X-100 for acid phosphatase (AP) and glutathione (GSH) determinations. All experiments were performed in duplicate. Assays Lactate
dehydrogenase. Lactate dehydrogenase (LDH) was assayed according to Wroblevsky and La Due [19], aspartate-aminotransferase (AST) by standard kits (Boehringer Mannheim, Monza, Italy), both in the culture medium. AP was measured in the cell lysate by the method of Lawson et al. [20]. Hepatocyte GSH. Hepatocyte GSH was measured by the method of Sedlak and Lindsay [21] as non-protein sulphydryl groups in the supernatant fraction obtained after deproteinization of the cell lysate with 50% trichloroacetic acid (TCA). Cell lipid peroxidation. Cell lipid peroxidation was monitored as production of malondialdehyde (MDA). The cells were incubated with Krebs-Henseleit buffer for 30 min and then deproteinated with 50% TCA; MDA was assayed by treatment of the supernatant fraction with 2-thiobarbituric acid, as described by Rush et al. [ 141. Microsomal ripid pet-oxidation. Rat liver microsomes, prepared according to Cinti et al. [22], (1.2 mg protein/ml) were incubated at 37°C for 10 min with 0.25 mM TBHP in 0.15 M KCl,lO mM Tris (pH 7.4) after addition of the flavonoids or butylated hydroxy-anisole (BHA) [23]. Subsequent treatment and determination of MDA were as described above. Proteins. Proteins were measured according to Lowry et al. [24] with BSA as standard. Diosmin and diosmetin assay. After 40 h exposure to 50 PM flavonoids the cells were washed twice and then scraped with saline. The ceil suspension was homogenized and centrifuged at 4°C and 2000 x g for 15 min and two fractions, cell membrane and lysate, were obtained and frozen. The two flavonoids were assayed in both fractions by HPLC using a Perkin Elmer LC pump and LC75 detector fitted with a fixed-wavelength 345-nm filter. Chromatograms were recorded with a Shimadzu C-R3A integrator. The analytical column was a Lichrospher 100 RP-18 (5 pm). The mobile phase consisted of methanol/acetic acid/water in a ratio of 40:7:53. This solution was pumped isocratically through the column at ambient temperature with a flow rate of 1 ml/min after sonication in an ultrasonic bath. The retention times of diosmin and diosmetin were 2.2 and 2.7 min, respectively. Statistical analysis
The statistical significance of the results was established using one- and two-way analysis of variance and P < 0.05 was chosen as the minimum level of significance. Results Leakage of the intracellular enzymes LDH and AST was used as an indicator of cell injury. Enzyme release increased with the time of exposure to 100 PM EE
183
2
4
6
4
6
600 g 0
500
?I
400
.c ?A _z ii
300 200 100 0 2
time (hours) Fig. 2. LDH and AST release from hepatocyte cultures pretreated with diosmin or diosmetin for 40 h and then exposed to 100 PM EE for 6 h. Results are mean f SE of 6-8 (LDH) and 4-6 (AST) different from the untreated controls hepatocyte preparations. All the groups were significantly different (P < 0.01). 0 P < 0.05 and 00 P < 0.01vs.diosmetin 50 pM; * P < 0.05 and ** P < 0.01 vs. EE.
(Fig. 2). After treating the cultures for 40 h with diosmetin, 25 and especially 50 PM, the toxicity, evaluated by LDH and AST leakages, was significantly less than in cells treated with EE alone (about 35-40%) or after treatment with 50 PM diosmin (Fig. 2). Diosmin at the same concentrations was barely effective (Fig. 2). Only cells pretreated with diosmetin maintained a significantly higher cell content of the lysosomal enzyme, acid phosphatase, after 6 h exposure to EE (Table I). The total content of LDH, AST and AP was not affected by the pretreatment with flavonoids at the concentrations used, except for LDH which was slightly reduced by the 40-h treatment with 50 PM diosmin (1700 f 57 vs. 1984 + 71 munits/mg protein in untreated controls) (P < 0.05). The treatment with 0.5 mM TBHP caused LDH leakage proportional to the time of exposure (Fig. 3). Pretreatment with 25 and 50 PM diosmetin significantly protected the cells, reducing LDH leakage by about 50 and 70”%1respectively after 30 min exposure to TBHP (Fig. 3). At this time many ‘blebs’ appeared on the plasma membrane of the cells treated with TBHP after exposure to either the vehicle or diosmin but not in cells pretreated with 50 PM diosmetin (not shown). After 40 and 50 min exposure to TBHP, pretreatment with 50 PM diosmetin significantly protected the cells, by about 500/u (Fig. 3). The protective effect of diosmetin was related
184
TABLE
I
ACID PHOSPHATASE ESTOLATE (EE)
IN
HEPATOCYTES
AFTER
Treatment
EXPOSURE
Acid phosphatase (pg phosphorus/mg
Untreated Controls EE 100 PM”
64.56 30.21 27.02 30.69 29.49 32.86 40.18 41.37
Diosmin 12.5 pMb + EE 100 PM” Diosmin 25 pMb + EE 100 PM” Diosmin 50 pMb + EE 100 PM” Diosmetin 12.5 pMC + EE 100 pM” Diosmetin 25 pMC + EE 100 pM” Diosmetin 50 pMC + EE 100 pMa
protein/60
ERYTHROMYCIN
min)
f zt f f zt f
2.15 l.OS”*@ 1.65’ 1.93’* 2.14”§ 2.07” l 2.00°+ f 1.72”++
aAfter 40 h of culture the cells pretreated or not with flavonoids tested for the cellular content of acid phosphatase. b,CHepatocytes were pretreated with diosmin (b) or diosmetin different hepatocyte preparations. “P < 0.01 vs. untreated controls. +P < 0.05 and +‘P < 0.01 vs. diosmetin 12.5 pM.
*P < 0.01vs.diosmetin $P < 0.01 vs. diosmetin
TO
were exposed
to EE for 6 h and then
(c) for 40 h. Results are mean f SE of 4-5
25 pM. 50 pM.
2500
1 i-
q
untr control
2000
ln
F 3 f EI 2 z E -I
1500
q q
Diosmetin 25pM Diosmetln 50pM
1000
500
0 40
30
time (minutes) Fig. 3. LDH posed to 0.5 0 P < 0.05 ** P < 0.01
release from hepatocyte cultures mM TBHP for 50 min. Results and 00 P < 0.01 vs. diosmetin vs. untreated controls.
pretreated with diosmin or diosmetin for 40 h and then exare mean f SE of 4-6 different hepatocyte preparations. 25 pM; 0 P < 0.01 vs. diosmetin 50 PM; l P < 0.05 and
185
untr. control Diosmin 12.5pM Diosmin 25pM Diosmin 50pM Diosmetinl2S~M Diosmetin 25pM Diosmetin 50pM
0
18
40
time of exposure to flavonoids Fig. 4. GSH content (non-protein SH) in cells after treated with flavonoids 3 and 21 h after seeding and of treatment. Results are mean f SE of 4-6 different 25 FM; 0 P < 0.05 and 00 P < 0.01vs.diosmetin * P < 0.05 and ** P < 0.01 vs. untreated control
TABLE
(hours)
exposure to diosmin or diosmetin. The cells were harvested for GSH evaluation after 0, 18 and 40 h hepatocyte preparations. 0 P < 0.01 vs. diosmetin 50 PM; A P < 0.05 vs. untreated cells at time 0; at the same time of culture.
II
GSH CONTENT
IN CELLS
AFTER
EXPOSURE
TO TBHP
Group
Pretreatmenta
GSH (non-protein (nmol/mg protein)
Controlb Controlb
Vehicle Diosmin 50 PM Diosmetin 50 pM
39.52 f 1.11+ 35.68 f 2.60+ 67.70 f 3.57
Vehicle Diosmin 50 pM Diosmetin 50 PM
2.65 f 0.44” 2.65 f 0.44” 20.34 zt 3.76
Controlb TBHP 0.5 mMC TBHP 0.5 mMC TBHP 0.5 mMC
SH) content
(%J)d
100 90 171 6.1 6.1 51.5
aThe cells were treated with either the vehicle or flavonoids 3 and 21 h after seeding for 40 h. b,CThe cells were exposed to either the vehicle cb) or TBHP @) for 50 min and then harvested for GSH measurement. dGSH content is expressed as a percentage of the content of control cells pretreated with the vehicle. The results are mean f SE of 3-4 different hepatocyte preparations. All the groups treated with TBHP were significantly different (P < 0.01) from the corresponding controls. +P < 0.01 vs. control/diosmetin 50 PM. OP < 0.01 vs. TBHP/diosmetin 50 pM.
186
to lower lipid peroxidation; cells pretreated with 50 PM diosmetin and exposed to TBHP for 30 min produced about half the amount of MDA (0.743 + 0.123 nmol MDA/mg protein) compared to cells treated with TBHP after exposure to either the vehicle (1.535 f 0.264 nmol/mg protein) (P < 0.05) or 50 PM diosmin (1.895 + 0.280 nmol/mg protein) (P < 0.01). The high degree of protection by diosmetin was also correlated with the glutathione content. After 18 and especially 40 h treatment, diosmetin caused a significant dose-dependent increase in the content of GSH while diosmin did not affect it at 25 PM and slightly reduced it at 50 PM (Fig. 4). The effect of 50 PM diosmin on GSH may reflect some toxicity of this flavonoid as shown by the small decrease (14%) of LDH in the cells after 40 h treatment. After 50 min exposure to TBHP (Table II), GSH content declined to about 7% of the control in cells pretreated with the vehicle or diosmin; in diosmetin-treated hepatocytes GSH was reduced to about 50% of the content in control cells pretreated with the vehicle. HPLC analysis of the cell membrane and lysate of the hepatocytes after 40 h exposure to the flavonoids (50 PM) only detected diosmin in the cell membrane (7-16 pg/mg cell protein: results of two experiments). When the flavonoids were added with the toxic compounds after 21 h of culture no protection by diosmin or diosmetin was observed against EE damage (data not shown). However, diosmetin was very effective in protecting the cells when it was added with TBHP (Fig. 5); after 30 min exposure there was no damage with 12.5 and 25 PM diosmetin and after 40 and 50 min, at the higher concentration, the protection was about 90 and 70% respectively. After 50 min the cells treated with TBHP and 25 PM diosmetin maintained a significantly (P < 0.01) higher GSH content (34.9 f 7.7 nmol/mg protein) than cells treated with TBHP alone (4.7 f 1.1
2500
1 q 1n
E‘ 2000 .-v) ? 3 1500-L aa d0 lOOO.z 4
500 -
o+
untr. control TBHP 0.5 mM
q
Diosmin 12.5pM
0
Diosmin 25pM
q q
Diosmetinl2.5pM Diosmetin 25pM
L
30
40
time (minutes) Fig. 5. LDH release from hepatocyte cultures treated with 0.5 mM TBHP in the presence or not of diosmin or diosmetin after 21 h of culture. Results are mean f SE of 3-4 different hepatocyte preparations. 0 P < 0.05and 00 P < 0.01 vs. diosmetin 12.5 PM; 0 P < 0.05and 00 P < 0.01 vs. diosmetin 25 PM; * P < 0.05 and ** P < 0.01 vs. untreated controls.
187
nmol/mg protein) or TBHP and 25 PM diosmin (5.1 f 1.5 nmolimg protein). The residual GSH content in cells treated with TBHP and 25 PM diosmetin was about 80% compared with untreated controls (43.6 + 5.5 nmol/mg protein). The antioxidant activity of both flavonoids was investigated and compared with that of a classical antioxidant, BHA, using microsomal lipid peroxidation induced by TBHP [23]. The I& (concentration of inhibitor required to inhibit oxidant activity by 50%) for diosmetin was 11.3 f 1.1 PM - 3.5 times higher than that of BHA (3.2 + 0.4 PM). Diosmin was much less effective than diosmetin and its IC,, was 113.2 f 16.9 PM (P < 0.01 diosmetin vs. diosmin). Discussion Diosmetin, the main metabolite of diosmin in man, protected hepatocytes against cell injury produced by EE and TBHP. These compounds damage liver cells by different mechanisms. It has been proposed that the plasma membrane is the primary target of EE because of its surface-active properties [ 1 l- 131. EE, a combination of the propionate salt of erythromycin with a detergent, lauryl sulfate, has great surfactant activity, high lipid-water partition coefficient and a marked adsorption onto the membrane interphase [ 111. If any change occurs to the membrane the adsorption of the drug and its cytotoxic effects may be modified accordingly. The fact that diosmetin protected against EE toxicity only after treatment of the hepatocytes for 40 h and not when given with the toxic agent suggests that the flavonoid needs a period of contact with the cells in order to affect and probably stabilize the structure of the biological membranes. The smaller leakage of LDH, a soluble cytoplasmic enzyme and AST, an enzyme found in cytosol and mitochondria, indicates greater protection of the cell and mitochondrial membranes and the higher residual content of the lysosomal enzyme acid phosphatase suggests less lysosome fragility after diosmetin treatment. These data are in agreement with the effect of other flavonoids such as rutin and catechin which exert a stabilizing influence on the lysosomes [7,8]. Previous work using a fluorescent probe showed that some flavonoids interacted with the polar zone of the lipid bilayer [3]. Diosmetin may affect and stabilize the structure of biological membranes by the following mechanisms: (i) by modifying the membrane composition and eventually membrane fluidity as shown for other antioxidant chemicals; an effect on membrane fluidity has been correlated with the protection against hepatocellular toxicity in the rat [25] and (ii) by affecting the metabolism and activation of the arachidonic acid cascade which has been suggested is involved in liver cell necrosis [26]. In vivo in a model of inflammatory granuloma in the rat, 15 days of treatment with diosmin inhibited the synthesis of prostaglandins (PGE; and PGF2,) and thromboxane [5]. The fact that diosmin but not diosmetin was detectable in the cell membrane after exposure to both flavonoids suggests that the latter’s protective effect is associated with its metabolites. TBHP is an organic hydroperoxide which can be metabolized by: (i) glutathione peroxidase resulting in depletion of GSH and oxidation of NADPH with consequent disturbances of intracellular Ca2+ homeostasis which is a critical event in the for-
188
mation of blebs on the plasma membrane and in cell death; (ii) the cytochrome P-450 system resulting in either alkoxyl (RO’) or peroxyl (ROO‘) electrophile free radicals which may lead to lipid peroxidation and cell injury [ 14- 171. The relative contribution of each of these potential mechanisms to the cytotoxicity of TBHP is still debated. Farber’s group showed that catechol reduced lipid peroxidation in parallel with the reduction of cell killing without any effect on GSH or calcium metabolism [15]. On the other hand, it has been shown that the addition of promethazine, an antioxidant molecule, inhibited lipid peroxidation without affecting cell toxicity and GSH content [14] while treatment with GSH prevented the intracellular depletion of GSH, reduced lipid peroxidation and preserved cell viability [ 171. Our results show that diosmetin pretreatment significantly protected against TBHP toxicity both by reducing lipid peroxidation and by making more GSH available; at the end of exposure to the toxic agent a lower GSH content, which is an indirect measurement of TBHP metabolism [16], was found also in the cells pretreated with diosmetin but the residual GSH level was still maintained high (about 50% of the control cells compared with 7% in cells treated with TBHP after treatment with the vehicle). Consequently this study too suggests that the depletion of GSH is an important event associated with the cell death induced by TBHP. How diosmetin increases GSH content is not known. It may be related to the antioxidant properties of flavonoids [2-51. A similar increasing effect on liver GSH content in vivo was described for the synthetic antioxidants butylated hydroxy-anisole and butylated hydroxy-toluene [27]. When the cells were treated with diosmetin and TBHP together the protection was even higher and much less GSH was consumed during exposure to the toxic agent (only 20% with 25 PM diosmetin), suggesting that in these conditions diosmetin might act as a direct antioxidant. In fact the antioxidant activity of diosmetin, three times lower than that of BHA, was in the range of the protective effect on hepatocytes (12.5-25 PM) and was significantly higher than that of diosmin. Diosmin had little effect on EE toxicity and none on TBHP although it was detectable in the cell membrane suggesting that in cultured hepatocytes it is not transformed in diosmetin. In fact the glycosidases which are capable of liberating flavonoid aglycons are present in intestinal bacterial flora but not in liver [28]. In conclusion, our findings with cultured rat hepatocytes indicate that diosmetin, but not diosmin, significantly reduces the damage caused by EE and TBHP and significantly increases the GSH content. After exposure to both flavonoids only diosmin is detectable in the cell membrane fraction. References 1 2
3
4
B. Havsteen, Flavonoids, a class of natural products of high pharmacological potency. Biochem. Pharmacol., 32 (1983) 1141. I.B. Afanas’ev, A.1. Dorozhko, A.V. Brodskii, V.A. Kostyuk and A.I. Potapovitch, Chelating and free radical scavenging mechanisms of inhibitory action of rutin and quercetin in lipid peroxidation. Biochem. Pharmacol., 38 (1989) 1763. A.K. Ratty, J. Sunamoto and N.P. Das, Interaction of flavonoids with I, I-diphenyl-2-picrylhydrazyl free radical, liposomal membranes and soybean lipoxygenase-1. Biochem. Pharmacol., 37 (1988) 989. M. Lonchampt. B. Guardiola, N. Sicot, M. Bertrand, L. Perdrix and J. Duhault, Protective effect
189
5
6
8 9 10 II 12 13 14
15 16 17 18 19 20
21 22 23 24 25
26 27 28
of a purified flavonoid fraction against reactive oxygen radicals. In vivo and in vitro study. Arzneim. Forsch., 39 (1989) 882. M. Damon, 0. Flandre, F. Michel, L. Perdrix, C. Labrid and A. Crastes de Paulet, Effect of chronic treatment with a purified flavonoid fraction on inflammatory granuloma in the rat. Study of prostaglandin E, and F,, and thromboxane B, release and histological changes. Arzneim. Forsch., 37 (1987) 1149. G.R. Reddy, N. Ueda, T. Hada, A.C. Sackeyfio, S. Yamamoto, Y. Hano, M. Aida and T. Nomura, A prenylflavone, artonin E, as arachidonate S-lipoxygenase inhibitor. Biochem. Pharmacol., 41 (1991) 115. P. Van Caneghem, Influence of some hydrosoluble substances with vitamin P activity on the fragility of lysosomes in vitro. Biochem. Pharmacol., 21 (1972) 1543. P. Niebes and G. Ponard, Stabilization of rat liver lysosomes by (+)-cyanidanol-3 in viva. Biochem. Pharmacol., 24 (1975) 905. J.C. Davila, A. Lenherr and D. Acosta, Protective effect of flavonoids on drug-induced hepatotoxicity in vitro. Toxicology, 57 (1989) 267. K.A. Mereish and R. Solow, Effect of antihepatotoxic cultured rat hepatocytes. Pharm. Res., 7 (1990) 256. CA. Dujovne, Liver cell culture toxicity and surfactant icol. Appl. Pharmacol., 32 (1975) Il.
agents
against
potency
of erythromycin
P. Villa, J.-M. Big& and A. Guillouzo, Effects of erythromycin hepatocytes. Biochem. Pharmacol., 33 (1984) 4098. P. Villa, J.-M. Begue and A. Guillouzo, Erythromycin toxicity hepatocytes. Xenobiotica, 15 (1985) 767. G.F. Rush, J.R. Gorski, M.G. Ripple, hydroperoxide-induced lipid peroxidation Pharmacol., 78 (1985) 473. N. Masaki, M.E. Kyle and J.L. Farber,
J. Sowinski, P. Bugelski and cell death in isolated ferf-Butyl
hydroperoxide
microcystin-LR
toxicity
derivatives.
in
Tox-
derivatives
on cultured
in primary
cultures
rat
of rat
and W.R. Hewitt, Organic hepatocytes. Toxicol. Appl. kills cultured
hepatocytes
by
peroxidizing membrane lipids. Arch. Biochem. Biophys., 269 (1989) 390. G.F. Rush, L.A. Yodis and D. Alberts, protection of rat hepatocytes from tert-butyl hydroperoxideinduced injury by catechol. Toxicol. Appl. Pharmacol., 84 (1986) 607. P. But-Calderon, I. Latour and M. Roberfroid, Biochemical changes in isolated hepatocytes exposed to terr-butyl hydroperoxide. Implications for its cytotoxicity. Cell Biol. Toxicol., 7 (1991) 129. P.O. Seglen, Preparation of isolated rat liver cells. Methods Cell Biol., 13 (1976) 29. F. Wroblewski and J.S. LaDue, Lactic dehydrogenase activity in blood. Proc. Sot. Exp. Biol. Med.. 90 (1955) 210. K.J. Lawson, C.J. Danpure and D.A. Fyfe, The uptake and subcellular distribution of gold in rat liver cells after in vivo administration of sodium aurothiomalate. Biochem. Pharmacol., 26 (1977) 2417. J. Sedlak and R.H. Lindsay, Estimation of total, protein-bound and nonprotein sulfhydryl groups in tissue with Ellman’s reagent. Anal. Biochem., 25 (1968) 192. D.L. Cinti, P. Moldeus and J.B. Schenkman, Kinetic parameters of drug-metabolizing enzymes in Ca*+-sedimented microsomes from rat liver. Biochem. Pharmacol., 21 (1972) 3249. SM. Borowitz and C. Montgomery, The role of phospholipase A2 in microsomal lipid peroxidation induced with f-butyl hydroperoxide. Biochem. Biophys. Res. Commun., 158 (1989) 1021. 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. H.G. Shertzer, G.L. Bannenberg, M. Rundgren and P. Moldeus, Relationship of membrane fluidity, chemoprotection and the intrinsic toxicity of butylated hydroxytoluene. Biochem. Pharmacol., 42 (1991) 1587. F. Guarner, N.K. Boughton-Smith, G.J. Blackwell and S. Moncada, Reduction by prostacyclin of acetaminophen-induced liver toxicity in the mouse. Hepatology, 8 (1988) 248. H. Jaeschke and A. Wendel, Choleresis and increased biliary emux of glutathione induced by phenolic antioxidants in rats. Toxicology, 52 (1988) 225. G. Tamura, C. Gold, A. Ferro-Luzzi and B.N. Ames, Fecalase: A model for activation of dietary glycosides to mutagens by intestinal flora. Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 4961.
179 Ltd
Protective effect of diosmetin on in vitro cell membrane damage and oxidative stress in cultured rat hepatocytes Pia Villa”, Dario Covaa, Laura De Francescob, Giuseppina Palladini” and Raffaella
Amalia Peregoa
Guaitanib,
“C. N. R. (National Research Council) Center ofCytophurmucology, Depurtmvn~ of’ Phrrrmucology. Viu Vanvitelli 32, 20129 Milan and hExperimentul Liver Toxicology Unit, Isriruio di Ricrrche Furmucologichc~ Mario Negri. Viu Eritrecc 62, 20157 Milun (Italy) (Received
December
12th. 1991; accepted
April 4th. 1992)
Summary Primary cultures of rat hepatocytes were used to study the effects of the flavonoids diosmin and its main metabolite diosmetin on the cell membrane damage caused by erythromycin estolate (EE) and oxidative stress caused by ferr-butylhydroperoxide (TBHP). The damage was evaluated by the leakage of intracellular enzymes lactate dehydrogenase, aspartate-aminotransferase and the residual cell content of a lysosomal marker acid phosphatase CAP). After treating the cells for 40 h with diosmetin EE induced less enzyme leakage. The content of AP was kept higher by diosmetin pretreatment after 6 h exposure to EE. Diosmin at the same concentrations had barely any effect. Diosmetin, but not diosmin, also protected against TBHP toxicity and this was related to lower lipid peroxidation and higher glutathione content caused by pretreatment with the flavonoid. When the cells were treated simultaneously with TBHP and diosmetin after 21 h of culture, the protection by the flavonoid was even higher. In fact the antioxidant activity of diosmetin was considerably greater than that of diosmin. After 40 h exposure to both flavonoids diosmin but not diosmetin was detectable in the cell membrane fraction, suggesting that the latter’s protective effect is associated with its metabolites. Key words: Flavonoids;
Diosmin;
Diosmetin;
Cultured
hepatocytes;
Hepatotoxicity;
Glutathione
Introduction
Flavonoids are natural compounds widely found in edible plants. Some are claimed to have biological and pharmacological properties [ 1] mainly related to their effects on the production of oxygen-free radicals [2-41 and/or arachidonic acid metabolites [3,5,6]. Diosmin (Fig. 1), a semi-synthetic flavonoid utilized in the treatment of various venous diseases, inhibited the production of oxygen radicals in vivo Correspondence to; Pia Villa, Experimental l-20157 Milan, Italy.
0300-483XI92/$05.00 0 1992 Elsevier Scientific Publishers Printed and Published in Ireland
Liver Toxicology
Ireland
Ltd.
Unit, Istituto
Mario Negri, Via Eritrea
62,
180
OH \O
OIOSMIN
HO
OH
Fig. 1. Molecular
structures
of diosmin
0
and diosmetin
and in vitro [4] and reduced the inflammatory reaction in the rat by affecting the synthesis of prostaglandins (PGE*, PGFz,) and thromboxane (TX B2) [5]. Therefore flavonoids can act as antiradicals and antioxidants with a consequent stabilizing effect on cell membranes and intracellular structures. Flavonoids such as rutin and catechin influence the fragility of lysosomes of rat liver in vitro [7] and in vivo after intoxication by galactosamine and ethanol [8]. Studies with cultured rat hepatocytes have shown the protective effect of some flavonoids, catechin, silybin [9] and silymarin [lo] against toxic insult by various compounds. In order to gain insight into the effects of diosmin and its aglycon, diosmetin (Fig. l), on cell membrane damage and oxidative stress we used the model of rat hepatocytes in primary culture. Two compounds, erythromycin estolate (EE) and tert-butyl hydroperoxide (TBHP), which damage liver cells by different mechanisms, were used for the study. EE is a macrolide antibiotic which produces hepatic injury in vivo and in vitro through a hypersensitive mechanism and a membrane surfactant effect [l l- 131. TBHP is an organic hydroperoxide which causes peroxidation of membrane lipids [14,15], disturbances in intracellular Ca*+ homeostasis and the glutathione system, leakage of cytoplasmic enzymes and cell death [ 14- 171. Hepatocytes were exposed to each toxic compound together with diosmin or diosmetin after 21 h of culture or after 40 h treatment with either diosmin or diosmetin. The cytotoxic effects of EE and TBHP were evaluated by leakage of intracellular enzymes. After exposure to flavonoids their content was measured in the cells and correlated to their pharmacological effect.
181
Materials and methods Chemicals Diosmin and diosmetin were supplied by Geymonat S.p.A. (Anagni, Fr., Italy). Erythromycin estolate was a gift from Pierre1 (Milan, Italy), tert-butyl hydroperoxide was from Aldrich-Chemie (Steinheim, Germany) and butylated hydroxyanisole from Sigma Chemical Co., St. Louis, MO. All other chemicals were analytical grade. Animals Male Crl:CD(SD)BR rats weighing 180-200 g, fed an ‘open formula’ diet ad libitum and housed under controlled conditions (22 f 0.5”C, 55% relative humidity, 12/12 h light/dark cycle), were used as liver donors. Experimental procedures with animals were reviewed and approved by the institutional animal care and use committee, in compliance with institutional guidelines that adhere to the principles found in the ‘Guide for the Care and Use of Laboratory Animals’, Publication No. 85-23. NIH, Bethesda, MD (revised 1985). Isolation and primary culture of hepatocytes Hepatocytes were isolated by perfusing the liver with 0.033% w/v collagenase solution, according to Seglen’s technique [ 181, under pentobarbital sodium (50 mg/kg body weight intraperitoneally (i.p.)) anesthesia. Cell viability was tested by trypan blue exclusion and was 87 f 1%. Hepatocytes were seeded at a density of 0.85 x lo6 cells per 9-cm2 Petri dish in 1.5 ml of standard medium with 5% fetal calf serum (FCS) and cultured at 37°C in a 5% CO*-95% air humidified atmosphere. The standard medium consisted of a mixture of 75% minimum essential medium and 25% medium 199 containing 0.2% bovine serum albumin (BSA), 1 &ml bovine insulin, 50 &ml kanamycin, 50 &ml streptomycin and 50 I.U./ml penicillin [12,13]. The medium, without FCS and with 7 x 10m5M hydrocortisone hemisuccinate, was renewed 3 and 21 h after hepatocyte seeding in the absence and presence of flavonoids. Culture treatments Diosmin and diosmetin were dissolved in dimethylsulphoxide (DMSO) and 0.5 N NaOH. Two treatment schedules were used: (i) the flavonoids (12.5 and 25 PM) were added with the toxic substances 21 h after cell seeding and (ii) the flavonoids (12.5, 25 and 50 PM) were added 3 and 21 h after hepatocyte seeding for a total exposure of 40 h until treatment with the toxic compounds started. The untreated controls were exposed to the same DMSO concentration (0.05% v/v) as the treated cultures. EE (100 PM), dissolved in DMSO, and TBHP (0.5 mM) were added to the cells diluted with culture medium without BSA after either 21 h or 43 h of culture for 6 h and 50 min, respectively. After different incubation periods (2, 4 and 6 h for EE and 30, 40 and 50 min for TBHP) measured amounts of culture medium were
182
withdrawn, centrifuged to remove cell debris and placed on ice for enzymatic analysis. The cells were washed twice with saline and immediately frozen. They were then harvested by scraping with 0.2% w/v Triton X-100 for acid phosphatase (AP) and glutathione (GSH) determinations. All experiments were performed in duplicate. Assays Lactate
dehydrogenase. Lactate dehydrogenase (LDH) was assayed according to Wroblevsky and La Due [19], aspartate-aminotransferase (AST) by standard kits (Boehringer Mannheim, Monza, Italy), both in the culture medium. AP was measured in the cell lysate by the method of Lawson et al. [20]. Hepatocyte GSH. Hepatocyte GSH was measured by the method of Sedlak and Lindsay [21] as non-protein sulphydryl groups in the supernatant fraction obtained after deproteinization of the cell lysate with 50% trichloroacetic acid (TCA). Cell lipid peroxidation. Cell lipid peroxidation was monitored as production of malondialdehyde (MDA). The cells were incubated with Krebs-Henseleit buffer for 30 min and then deproteinated with 50% TCA; MDA was assayed by treatment of the supernatant fraction with 2-thiobarbituric acid, as described by Rush et al. [ 141. Microsomal ripid pet-oxidation. Rat liver microsomes, prepared according to Cinti et al. [22], (1.2 mg protein/ml) were incubated at 37°C for 10 min with 0.25 mM TBHP in 0.15 M KCl,lO mM Tris (pH 7.4) after addition of the flavonoids or butylated hydroxy-anisole (BHA) [23]. Subsequent treatment and determination of MDA were as described above. Proteins. Proteins were measured according to Lowry et al. [24] with BSA as standard. Diosmin and diosmetin assay. After 40 h exposure to 50 PM flavonoids the cells were washed twice and then scraped with saline. The ceil suspension was homogenized and centrifuged at 4°C and 2000 x g for 15 min and two fractions, cell membrane and lysate, were obtained and frozen. The two flavonoids were assayed in both fractions by HPLC using a Perkin Elmer LC pump and LC75 detector fitted with a fixed-wavelength 345-nm filter. Chromatograms were recorded with a Shimadzu C-R3A integrator. The analytical column was a Lichrospher 100 RP-18 (5 pm). The mobile phase consisted of methanol/acetic acid/water in a ratio of 40:7:53. This solution was pumped isocratically through the column at ambient temperature with a flow rate of 1 ml/min after sonication in an ultrasonic bath. The retention times of diosmin and diosmetin were 2.2 and 2.7 min, respectively. Statistical analysis
The statistical significance of the results was established using one- and two-way analysis of variance and P < 0.05 was chosen as the minimum level of significance. Results Leakage of the intracellular enzymes LDH and AST was used as an indicator of cell injury. Enzyme release increased with the time of exposure to 100 PM EE
183
2
4
6
4
6
600 g 0
500
?I
400
.c ?A _z ii
300 200 100 0 2
time (hours) Fig. 2. LDH and AST release from hepatocyte cultures pretreated with diosmin or diosmetin for 40 h and then exposed to 100 PM EE for 6 h. Results are mean f SE of 6-8 (LDH) and 4-6 (AST) different from the untreated controls hepatocyte preparations. All the groups were significantly different (P < 0.01). 0 P < 0.05 and 00 P < 0.01vs.diosmetin 50 pM; * P < 0.05 and ** P < 0.01 vs. EE.
(Fig. 2). After treating the cultures for 40 h with diosmetin, 25 and especially 50 PM, the toxicity, evaluated by LDH and AST leakages, was significantly less than in cells treated with EE alone (about 35-40%) or after treatment with 50 PM diosmin (Fig. 2). Diosmin at the same concentrations was barely effective (Fig. 2). Only cells pretreated with diosmetin maintained a significantly higher cell content of the lysosomal enzyme, acid phosphatase, after 6 h exposure to EE (Table I). The total content of LDH, AST and AP was not affected by the pretreatment with flavonoids at the concentrations used, except for LDH which was slightly reduced by the 40-h treatment with 50 PM diosmin (1700 f 57 vs. 1984 + 71 munits/mg protein in untreated controls) (P < 0.05). The treatment with 0.5 mM TBHP caused LDH leakage proportional to the time of exposure (Fig. 3). Pretreatment with 25 and 50 PM diosmetin significantly protected the cells, reducing LDH leakage by about 50 and 70”%1respectively after 30 min exposure to TBHP (Fig. 3). At this time many ‘blebs’ appeared on the plasma membrane of the cells treated with TBHP after exposure to either the vehicle or diosmin but not in cells pretreated with 50 PM diosmetin (not shown). After 40 and 50 min exposure to TBHP, pretreatment with 50 PM diosmetin significantly protected the cells, by about 500/u (Fig. 3). The protective effect of diosmetin was related
184
TABLE
I
ACID PHOSPHATASE ESTOLATE (EE)
IN
HEPATOCYTES
AFTER
Treatment
EXPOSURE
Acid phosphatase (pg phosphorus/mg
Untreated Controls EE 100 PM”
64.56 30.21 27.02 30.69 29.49 32.86 40.18 41.37
Diosmin 12.5 pMb + EE 100 PM” Diosmin 25 pMb + EE 100 PM” Diosmin 50 pMb + EE 100 PM” Diosmetin 12.5 pMC + EE 100 pM” Diosmetin 25 pMC + EE 100 pM” Diosmetin 50 pMC + EE 100 pMa
protein/60
ERYTHROMYCIN
min)
f zt f f zt f
2.15 l.OS”*@ 1.65’ 1.93’* 2.14”§ 2.07” l 2.00°+ f 1.72”++
aAfter 40 h of culture the cells pretreated or not with flavonoids tested for the cellular content of acid phosphatase. b,CHepatocytes were pretreated with diosmin (b) or diosmetin different hepatocyte preparations. “P < 0.01 vs. untreated controls. +P < 0.05 and +‘P < 0.01 vs. diosmetin 12.5 pM.
*P < 0.01vs.diosmetin $P < 0.01 vs. diosmetin
TO
were exposed
to EE for 6 h and then
(c) for 40 h. Results are mean f SE of 4-5
25 pM. 50 pM.
2500
1 i-
q
untr control
2000
ln
F 3 f EI 2 z E -I
1500
q q
Diosmetin 25pM Diosmetln 50pM
1000
500
0 40
30
time (minutes) Fig. 3. LDH posed to 0.5 0 P < 0.05 ** P < 0.01
release from hepatocyte cultures mM TBHP for 50 min. Results and 00 P < 0.01 vs. diosmetin vs. untreated controls.
pretreated with diosmin or diosmetin for 40 h and then exare mean f SE of 4-6 different hepatocyte preparations. 25 pM; 0 P < 0.01 vs. diosmetin 50 PM; l P < 0.05 and
185
untr. control Diosmin 12.5pM Diosmin 25pM Diosmin 50pM Diosmetinl2S~M Diosmetin 25pM Diosmetin 50pM
0
18
40
time of exposure to flavonoids Fig. 4. GSH content (non-protein SH) in cells after treated with flavonoids 3 and 21 h after seeding and of treatment. Results are mean f SE of 4-6 different 25 FM; 0 P < 0.05 and 00 P < 0.01vs.diosmetin * P < 0.05 and ** P < 0.01 vs. untreated control
TABLE
(hours)
exposure to diosmin or diosmetin. The cells were harvested for GSH evaluation after 0, 18 and 40 h hepatocyte preparations. 0 P < 0.01 vs. diosmetin 50 PM; A P < 0.05 vs. untreated cells at time 0; at the same time of culture.
II
GSH CONTENT
IN CELLS
AFTER
EXPOSURE
TO TBHP
Group
Pretreatmenta
GSH (non-protein (nmol/mg protein)
Controlb Controlb
Vehicle Diosmin 50 PM Diosmetin 50 pM
39.52 f 1.11+ 35.68 f 2.60+ 67.70 f 3.57
Vehicle Diosmin 50 pM Diosmetin 50 PM
2.65 f 0.44” 2.65 f 0.44” 20.34 zt 3.76
Controlb TBHP 0.5 mMC TBHP 0.5 mMC TBHP 0.5 mMC
SH) content
(%J)d
100 90 171 6.1 6.1 51.5
aThe cells were treated with either the vehicle or flavonoids 3 and 21 h after seeding for 40 h. b,CThe cells were exposed to either the vehicle cb) or TBHP @) for 50 min and then harvested for GSH measurement. dGSH content is expressed as a percentage of the content of control cells pretreated with the vehicle. The results are mean f SE of 3-4 different hepatocyte preparations. All the groups treated with TBHP were significantly different (P < 0.01) from the corresponding controls. +P < 0.01 vs. control/diosmetin 50 PM. OP < 0.01 vs. TBHP/diosmetin 50 pM.
186
to lower lipid peroxidation; cells pretreated with 50 PM diosmetin and exposed to TBHP for 30 min produced about half the amount of MDA (0.743 + 0.123 nmol MDA/mg protein) compared to cells treated with TBHP after exposure to either the vehicle (1.535 f 0.264 nmol/mg protein) (P < 0.05) or 50 PM diosmin (1.895 + 0.280 nmol/mg protein) (P < 0.01). The high degree of protection by diosmetin was also correlated with the glutathione content. After 18 and especially 40 h treatment, diosmetin caused a significant dose-dependent increase in the content of GSH while diosmin did not affect it at 25 PM and slightly reduced it at 50 PM (Fig. 4). The effect of 50 PM diosmin on GSH may reflect some toxicity of this flavonoid as shown by the small decrease (14%) of LDH in the cells after 40 h treatment. After 50 min exposure to TBHP (Table II), GSH content declined to about 7% of the control in cells pretreated with the vehicle or diosmin; in diosmetin-treated hepatocytes GSH was reduced to about 50% of the content in control cells pretreated with the vehicle. HPLC analysis of the cell membrane and lysate of the hepatocytes after 40 h exposure to the flavonoids (50 PM) only detected diosmin in the cell membrane (7-16 pg/mg cell protein: results of two experiments). When the flavonoids were added with the toxic compounds after 21 h of culture no protection by diosmin or diosmetin was observed against EE damage (data not shown). However, diosmetin was very effective in protecting the cells when it was added with TBHP (Fig. 5); after 30 min exposure there was no damage with 12.5 and 25 PM diosmetin and after 40 and 50 min, at the higher concentration, the protection was about 90 and 70% respectively. After 50 min the cells treated with TBHP and 25 PM diosmetin maintained a significantly (P < 0.01) higher GSH content (34.9 f 7.7 nmol/mg protein) than cells treated with TBHP alone (4.7 f 1.1
2500
1 q 1n
E‘ 2000 .-v) ? 3 1500-L aa d0 lOOO.z 4
500 -
o+
untr. control TBHP 0.5 mM
q
Diosmin 12.5pM
0
Diosmin 25pM
q q
Diosmetinl2.5pM Diosmetin 25pM
L
30
40
time (minutes) Fig. 5. LDH release from hepatocyte cultures treated with 0.5 mM TBHP in the presence or not of diosmin or diosmetin after 21 h of culture. Results are mean f SE of 3-4 different hepatocyte preparations. 0 P < 0.05and 00 P < 0.01 vs. diosmetin 12.5 PM; 0 P < 0.05and 00 P < 0.01 vs. diosmetin 25 PM; * P < 0.05 and ** P < 0.01 vs. untreated controls.
187
nmol/mg protein) or TBHP and 25 PM diosmin (5.1 f 1.5 nmolimg protein). The residual GSH content in cells treated with TBHP and 25 PM diosmetin was about 80% compared with untreated controls (43.6 + 5.5 nmol/mg protein). The antioxidant activity of both flavonoids was investigated and compared with that of a classical antioxidant, BHA, using microsomal lipid peroxidation induced by TBHP [23]. The I& (concentration of inhibitor required to inhibit oxidant activity by 50%) for diosmetin was 11.3 f 1.1 PM - 3.5 times higher than that of BHA (3.2 + 0.4 PM). Diosmin was much less effective than diosmetin and its IC,, was 113.2 f 16.9 PM (P < 0.01 diosmetin vs. diosmin). Discussion Diosmetin, the main metabolite of diosmin in man, protected hepatocytes against cell injury produced by EE and TBHP. These compounds damage liver cells by different mechanisms. It has been proposed that the plasma membrane is the primary target of EE because of its surface-active properties [ 1 l- 131. EE, a combination of the propionate salt of erythromycin with a detergent, lauryl sulfate, has great surfactant activity, high lipid-water partition coefficient and a marked adsorption onto the membrane interphase [ 111. If any change occurs to the membrane the adsorption of the drug and its cytotoxic effects may be modified accordingly. The fact that diosmetin protected against EE toxicity only after treatment of the hepatocytes for 40 h and not when given with the toxic agent suggests that the flavonoid needs a period of contact with the cells in order to affect and probably stabilize the structure of the biological membranes. The smaller leakage of LDH, a soluble cytoplasmic enzyme and AST, an enzyme found in cytosol and mitochondria, indicates greater protection of the cell and mitochondrial membranes and the higher residual content of the lysosomal enzyme acid phosphatase suggests less lysosome fragility after diosmetin treatment. These data are in agreement with the effect of other flavonoids such as rutin and catechin which exert a stabilizing influence on the lysosomes [7,8]. Previous work using a fluorescent probe showed that some flavonoids interacted with the polar zone of the lipid bilayer [3]. Diosmetin may affect and stabilize the structure of biological membranes by the following mechanisms: (i) by modifying the membrane composition and eventually membrane fluidity as shown for other antioxidant chemicals; an effect on membrane fluidity has been correlated with the protection against hepatocellular toxicity in the rat [25] and (ii) by affecting the metabolism and activation of the arachidonic acid cascade which has been suggested is involved in liver cell necrosis [26]. In vivo in a model of inflammatory granuloma in the rat, 15 days of treatment with diosmin inhibited the synthesis of prostaglandins (PGE; and PGF2,) and thromboxane [5]. The fact that diosmin but not diosmetin was detectable in the cell membrane after exposure to both flavonoids suggests that the latter’s protective effect is associated with its metabolites. TBHP is an organic hydroperoxide which can be metabolized by: (i) glutathione peroxidase resulting in depletion of GSH and oxidation of NADPH with consequent disturbances of intracellular Ca2+ homeostasis which is a critical event in the for-
188
mation of blebs on the plasma membrane and in cell death; (ii) the cytochrome P-450 system resulting in either alkoxyl (RO’) or peroxyl (ROO‘) electrophile free radicals which may lead to lipid peroxidation and cell injury [ 14- 171. The relative contribution of each of these potential mechanisms to the cytotoxicity of TBHP is still debated. Farber’s group showed that catechol reduced lipid peroxidation in parallel with the reduction of cell killing without any effect on GSH or calcium metabolism [15]. On the other hand, it has been shown that the addition of promethazine, an antioxidant molecule, inhibited lipid peroxidation without affecting cell toxicity and GSH content [14] while treatment with GSH prevented the intracellular depletion of GSH, reduced lipid peroxidation and preserved cell viability [ 171. Our results show that diosmetin pretreatment significantly protected against TBHP toxicity both by reducing lipid peroxidation and by making more GSH available; at the end of exposure to the toxic agent a lower GSH content, which is an indirect measurement of TBHP metabolism [16], was found also in the cells pretreated with diosmetin but the residual GSH level was still maintained high (about 50% of the control cells compared with 7% in cells treated with TBHP after treatment with the vehicle). Consequently this study too suggests that the depletion of GSH is an important event associated with the cell death induced by TBHP. How diosmetin increases GSH content is not known. It may be related to the antioxidant properties of flavonoids [2-51. A similar increasing effect on liver GSH content in vivo was described for the synthetic antioxidants butylated hydroxy-anisole and butylated hydroxy-toluene [27]. When the cells were treated with diosmetin and TBHP together the protection was even higher and much less GSH was consumed during exposure to the toxic agent (only 20% with 25 PM diosmetin), suggesting that in these conditions diosmetin might act as a direct antioxidant. In fact the antioxidant activity of diosmetin, three times lower than that of BHA, was in the range of the protective effect on hepatocytes (12.5-25 PM) and was significantly higher than that of diosmin. Diosmin had little effect on EE toxicity and none on TBHP although it was detectable in the cell membrane suggesting that in cultured hepatocytes it is not transformed in diosmetin. In fact the glycosidases which are capable of liberating flavonoid aglycons are present in intestinal bacterial flora but not in liver [28]. In conclusion, our findings with cultured rat hepatocytes indicate that diosmetin, but not diosmin, significantly reduces the damage caused by EE and TBHP and significantly increases the GSH content. After exposure to both flavonoids only diosmin is detectable in the cell membrane fraction. References 1 2
3
4
B. Havsteen, Flavonoids, a class of natural products of high pharmacological potency. Biochem. Pharmacol., 32 (1983) 1141. I.B. Afanas’ev, A.1. Dorozhko, A.V. Brodskii, V.A. Kostyuk and A.I. Potapovitch, Chelating and free radical scavenging mechanisms of inhibitory action of rutin and quercetin in lipid peroxidation. Biochem. Pharmacol., 38 (1989) 1763. A.K. Ratty, J. Sunamoto and N.P. Das, Interaction of flavonoids with I, I-diphenyl-2-picrylhydrazyl free radical, liposomal membranes and soybean lipoxygenase-1. Biochem. Pharmacol., 37 (1988) 989. M. Lonchampt. B. Guardiola, N. Sicot, M. Bertrand, L. Perdrix and J. Duhault, Protective effect
189
5
6
8 9 10 II 12 13 14
15 16 17 18 19 20
21 22 23 24 25
26 27 28
of a purified flavonoid fraction against reactive oxygen radicals. In vivo and in vitro study. Arzneim. Forsch., 39 (1989) 882. M. Damon, 0. Flandre, F. Michel, L. Perdrix, C. Labrid and A. Crastes de Paulet, Effect of chronic treatment with a purified flavonoid fraction on inflammatory granuloma in the rat. Study of prostaglandin E, and F,, and thromboxane B, release and histological changes. Arzneim. Forsch., 37 (1987) 1149. G.R. Reddy, N. Ueda, T. Hada, A.C. Sackeyfio, S. Yamamoto, Y. Hano, M. Aida and T. Nomura, A prenylflavone, artonin E, as arachidonate S-lipoxygenase inhibitor. Biochem. Pharmacol., 41 (1991) 115. P. Van Caneghem, Influence of some hydrosoluble substances with vitamin P activity on the fragility of lysosomes in vitro. Biochem. Pharmacol., 21 (1972) 1543. P. Niebes and G. Ponard, Stabilization of rat liver lysosomes by (+)-cyanidanol-3 in viva. Biochem. Pharmacol., 24 (1975) 905. J.C. Davila, A. Lenherr and D. Acosta, Protective effect of flavonoids on drug-induced hepatotoxicity in vitro. Toxicology, 57 (1989) 267. K.A. Mereish and R. Solow, Effect of antihepatotoxic cultured rat hepatocytes. Pharm. Res., 7 (1990) 256. CA. Dujovne, Liver cell culture toxicity and surfactant icol. Appl. Pharmacol., 32 (1975) Il.
agents
against
potency
of erythromycin
P. Villa, J.-M. Big& and A. Guillouzo, Effects of erythromycin hepatocytes. Biochem. Pharmacol., 33 (1984) 4098. P. Villa, J.-M. Begue and A. Guillouzo, Erythromycin toxicity hepatocytes. Xenobiotica, 15 (1985) 767. G.F. Rush, J.R. Gorski, M.G. Ripple, hydroperoxide-induced lipid peroxidation Pharmacol., 78 (1985) 473. N. Masaki, M.E. Kyle and J.L. Farber,
J. Sowinski, P. Bugelski and cell death in isolated ferf-Butyl
hydroperoxide
microcystin-LR
toxicity
derivatives.
in
Tox-
derivatives
on cultured
in primary
cultures
rat
of rat
and W.R. Hewitt, Organic hepatocytes. Toxicol. Appl. kills cultured
hepatocytes
by
peroxidizing membrane lipids. Arch. Biochem. Biophys., 269 (1989) 390. G.F. Rush, L.A. Yodis and D. Alberts, protection of rat hepatocytes from tert-butyl hydroperoxideinduced injury by catechol. Toxicol. Appl. Pharmacol., 84 (1986) 607. P. But-Calderon, I. Latour and M. Roberfroid, Biochemical changes in isolated hepatocytes exposed to terr-butyl hydroperoxide. Implications for its cytotoxicity. Cell Biol. Toxicol., 7 (1991) 129. P.O. Seglen, Preparation of isolated rat liver cells. Methods Cell Biol., 13 (1976) 29. F. Wroblewski and J.S. LaDue, Lactic dehydrogenase activity in blood. Proc. Sot. Exp. Biol. Med.. 90 (1955) 210. K.J. Lawson, C.J. Danpure and D.A. Fyfe, The uptake and subcellular distribution of gold in rat liver cells after in vivo administration of sodium aurothiomalate. Biochem. Pharmacol., 26 (1977) 2417. J. Sedlak and R.H. Lindsay, Estimation of total, protein-bound and nonprotein sulfhydryl groups in tissue with Ellman’s reagent. Anal. Biochem., 25 (1968) 192. D.L. Cinti, P. Moldeus and J.B. Schenkman, Kinetic parameters of drug-metabolizing enzymes in Ca*+-sedimented microsomes from rat liver. Biochem. Pharmacol., 21 (1972) 3249. SM. Borowitz and C. Montgomery, The role of phospholipase A2 in microsomal lipid peroxidation induced with f-butyl hydroperoxide. Biochem. Biophys. Res. Commun., 158 (1989) 1021. 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. H.G. Shertzer, G.L. Bannenberg, M. Rundgren and P. Moldeus, Relationship of membrane fluidity, chemoprotection and the intrinsic toxicity of butylated hydroxytoluene. Biochem. Pharmacol., 42 (1991) 1587. F. Guarner, N.K. Boughton-Smith, G.J. Blackwell and S. Moncada, Reduction by prostacyclin of acetaminophen-induced liver toxicity in the mouse. Hepatology, 8 (1988) 248. H. Jaeschke and A. Wendel, Choleresis and increased biliary emux of glutathione induced by phenolic antioxidants in rats. Toxicology, 52 (1988) 225. G. Tamura, C. Gold, A. Ferro-Luzzi and B.N. Ames, Fecalase: A model for activation of dietary glycosides to mutagens by intestinal flora. Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 4961.
