Susceptibility of metallothionein-null mice to paraquat

ELSEVIER

Emironmental

Toxicology

and Pharmacology

I ( 1996)

12 I-225

Susceptibility of metallothionein-null

mice to paraquat

M. Sato ” * , M.D. Apostolova ‘, M. Hamaya ‘, J. Yamaki ‘, K.H.A. Choo ‘, A.E. Michalska ‘, N. Kodama ‘. C. Tohyama ’

Received

Ii

September

1995:

revwd

11 khruary

1996:

acceptrd

35 February

1906

Abstract Using transeenic mice in which metallothionein (MT)-1 and MT-II penes. we havpe studied a putative role of MT as a free radical scavenger against paraquat. a free radical generator. Male mice were injected S.C. with paraquat (PQ) at a single dose of 40 or 60 mg/kg of body weight (b.w.1. Two of the six MT-null mice died within 16 h at the dose of 60 mg PQ/kg. b. w. PQ administration increased hepatic MT concentration in the normal mice (C57BL/hJl, but not in the MT-null mice. The lipid peroxidation (LPI determined by thiobarbituric acid-reactive substance formation was increased by PQ in the liver of normal and MT-null mice, and the enhanced level was greater in the MT-null mice than in the C57BL/hJ mice. Administration of PQ significantly increased blood urea nitrogen only in the MT-null mice. indicating renal damage. Without paraquat administration. the hepatic concentration of non-protein sulphydryl compounds was less in the MT-null mice than in the CS7BL/hJ mice. and the basal level of LP wtas higher in the MT-null mice than in the C57BL/6J mice. The present results support the notion that MT plays an antioxidativje role against paraquat insult under physiolo@cal conditions.

1. Introduction Metallothionein (MT) is a cysteine-rich. low molecular weight (6-7 kDa) and metal-binding protein. One of the postulated roles of MT is to reduce the toxicity of heav3y metals such as cadmium and mercury (Kagi and Kojima. 1987). Recently. attention has increasingly been focused on a role of MT as a free radical scavenger (Sato and Bremner, 1993). For example. the rate constant for its reaction with hydroxyl radicals produced by the xanthinexanthine oxidase reaction is higher than that of the reduced form of glutathione (GSH) (Thomalley and Vasak. 1985). In fact, degradation of DNA by ionizing radiation is inhibited by MT in a concentration-dependent manner. Cadmium (Cd&resistant cells with enhanced levels of MT induced by Cd were more resistant to oxidative stresses causedby hydrogen peroxide or hydroxyl radicals (MelloFilho et al.. 1988). In viva studies have also shown a possible role of MT as a free radical scavenger. Mice

* Corresponding l382~668Y/Yh/Sl5.OO PI/ s1382-6689(4630001

author. Copyright

~” 1096 l-7

Elwvwr

Science

B.V.

All

right\

given inducing metalssuch as zinc (Zn), Cd, or manganese (Mn) showed increased resistance to a lethal dose of X irradiation compared with controls (Matsubara et al., 1986). Naganuma et al. (1988) reported that increased MT induced by bismuth compounds showed a protective effect against the lethal and cardiac toxicity of adriamycin, an anticancer agent that can induce free radicals. However, contrary to the above observation. Karina et al. (1990) observed that overexpression of MT in Chinese hamster ovary cells by genetic manipulation did not affect their ability to withstand oxidative challenge by bleomycin or irradiation

which

produces

hydroxyl

radicals,

but did in-

crea\e their resistanceto alkylating agents such as mitomycin C. Their data do not support the hypothesis that MT play4 a role as a scavengerof free radicals. In addition, the intracellular concentration of glutathione, which is an efficient scavenger.

is usually

much higher

than that of MT.

Therefore, a detailed study is required to establisha role of MT as an antioxidant. Paraquat is widely used as a herbicide, and its poisoning occurs in two phases(Vuska et al., 1983). The first. early phaselasts several days after paraquat exposure and reaervrd

involves liver, kidney and other tissues. The second. late phase involves only the lung and paraquat toxicity is known to induce pulmonary fibrosis, although mechanisms of the lesion have still been unknown. Paraquat is metabolized by NADPH-cytochrome C reductase system and forms free radicals with subsequent oxidation to the dications by molecular oxygen. This metabolic pathway is essential for development of paraquat toxicity (Bus and Gibson, 1984). A tissue specific difference in the ability to metabolize paraquat has been reported and the greatest rate of paraquat radical production was found in the liver (Baldwin et al., 1975). In the present study, effect of paraquat at the first early phase on the hepatic and renal function was studied. To study the role of MT as an antioxidant. it would be advantageous to use an animal model which is defective only in MT and to evaluate the capacity of these animals to scavenge free radicals. Recently, transgenic mice deficient in MT-I and MT-II genes have been generated (Michalska and Choo, 1993: Masters et al., 1994). To examine the proposed function of MT as a radical scavenger, we studied the ability of these transgenic mice to handle oxidative stress produced by paraquat.

2. Materials

2.2. Analysis The liver was rapidly excised. minced, and homogenized in cold 1.15% KCl-3 mM Tris-HCl buffer (pH 7.4) to make a 10% homogenate under N? atmosphere. The homogenates were used for estimation of the degree of lipid peroxidation. which was measured by formation of thiobarbituric acid-reactive substances (TBA-RS) (Uchiyama and Mihara, 1978). The homogenates were centrifuged at 8000 X g for 15 min, and the supematants were used to determine MT by radioimmunoassay (Nishimura et al., 1990). Prior to analysis of Zn, aliquots of each homogenate were treated with acid mixture (HNO, : HClO,, 3 : 1, v/v). After treatment, the inorganic were dissolved in ultrapure water residues (Sybron/Bamsteard, Tokyo, Japan) and metal analysis was carried out by inductively coupled plasma emission spectrometry (ICP; model ICAP-750, Nippon-Jarrel-Ash, Tokyo. Japan). The supematants of the homogenates (1500 X g. IO min) were used for estimation of non-protein sulphydryl compounds (NP-SH) which was determined by the method of Ellman (1959) as described by Costa and Murphy (1986). For diagnosis of liver and kidney function aspartate aminotransferase CAST; EC 2.6.1.1) and blood urea nitrogen (BUN) were determined, respectively, using commercially available assay kits (Wake Pure Chemicals Ind., Osaka, Japan).

and methods

2. I. Animals

2.3. Statistics

Mice with null mutation of MT-I and -II genes were produced as described earlier (Michalska and Choo, 1993). C57BL/6J control mice were obtained from Japan Clea Co. (Tokyo. Japan). MT-null male mice (9 weeks old) were injected S.C. with paraquat (PQ, 40 or 60 mg/kg body weight). Blood was collected from vena cava inferior under pentobarbital anesthesia, and the mice were killed 16 h after PQ injection. MT-null and C57BL/6J mice injected with saline were used as control.

Table I Effect of paraquat

administration

on mortality

and tissue weight

Data were analyzed by Student’s t test. The acceptable level of significance was set at P < 0.05.

3. Results The liver weight was smaller in the MT-deficient mice than in the C57BL/6J mice, whereas the weights of lung and kidney were similar between the MT-null and

at 16 h

C57BL/6J

Mortality

h

Tissue weight Liver Kidney Lung

MT-null Saline

PQ-1 ’

O/S

o/5

5.81 * 0.71 (100) d 1.13 * 0.04 (100) 0.65 + 0.02 (100)

5.38 k 0.13 (93) I .2 1 * 0.02 (107) 0.62 i 0.01 (95)

PQ-2

,’

O/35

Saline

PC-1 d

O/5

O/S

4.62 f 0.18 (100) 1.17+0.01 (100) 0.60 + 0.01 (100)

3.94 f 0.17 ’ (85) I.12 & 0.06 (96) 0.55 * 0.03 (91)

PQ-2

a

2/6

‘

a Mice were injected with a dose of paraquat (PQ-I: 40 or PQ-2: control. ^ Significantly different from the control at P < 0.05.

5.28 * 0.08 (91) I.15 *0.01 (102) 0.61 f 0.01 (94) 60 mg/kg).

h Number

of dead mice. ’ g/lOOg

of body weight.

3.55 i 0.19 * (77) I .07 * 0.01 (91) 0.58 + 0.02 (97) mean + SE. d % of the

223

Fig. I. Hepatic concentrations of zinc and metallothionein after paraquat admmistration, C57BL/6J and MT-null male mice (9 weeks old) were injected with a single subcutaneous dose of paraquat (PQ I: 40 or PQ 2: 60 mg/kg of body weight). MT-null and C57BL/6J mice that injected with saline were used as controls. The mice were sacrificed at I6 h after an injection. MT was determined by radioimmunoassay. * denotes significant difference from control at P < 0.05.

C57BL/6J mice (Table 1). Non-protein sulphydryl compounds which include GSH play an important role in the antioxidative system and a large amount of NP-SH exists in tissues, especially in liver. The basal level of NP-SH in the liver was less in the MT-null mice (7.50 f 0.21 pmol/g tissue, n = 5) than in the C57BL/6J mice (9.52 * 0.28 pmol/g tissue, n = 5, P < 0.05). After administration of PQ, mice became lethargic, and at the dose of 60 mg PQ/kg of body weight, two of the six MT-null mice died within 16 h (Table I). Since acute lethality is generally examined for a 7 days survival period, we failed to obtain LD,, values of PQ, but MT-null mice were found more susceptible to PQ than C57BL/6J mice. Administration of PQ significantly decreased liver weight in the MT-null mice, whereas the weights of kidney and lung were not affected. PQ administration increased hepatic concentrations of Zn and MT in the C57BL/6J mice, but not in the MT-null mice (Fig. 1). The hepatic lipid peroxidation was increased by PQ in the C57BL/6J and MT-null mice, and the enhanced level was significantly greater in the MT-null mice than in the

300

1:300

200

200

100

200

3 MT-null

C576U6J

Fig. 2. Lipid peroxidation in the liver of MT-null mice after paraquat administration. See the legend to Fig. I for animal treatment. The degree of lipid peroxidation was measured by monitoring the formation of thiobarbituric acid-reactive substances (TBA-RS). - denotes significant difference from control at P < 0.05. # denotes significant difference from a paired group of C57BL/6J mice at P < 0.05.

80

60 z 3 = It 100

100

= s E z 3 m

40

20

O conmfmz conmim

0

L

oCan ml C57Bu6J

R?2

conmim2 MT-null

Fig. 3. Plasma aspartate aminotransferase CAST) and blood urea nitrogen (BUN) in MT-null mice after paraquat animal treatment. A: AST: B: BUN. ’ denotes significant difference from control at P < 0.05.

administration.

See the legend to Fig. 1 for

C57BL/6J mice (Fig. 2). It should be pointed out that the TBA-RS level of the saline-treated MT-null mice was much higher than that of C57BL/6J mice. Thus, in the MT-null mice, the lipid peroxidation readily occurs both in basal and oxidative stress conditions. Administration of PQ resulted in the consumption of, and decrease in, hepatic NP-SH in both MT-null and C57BL/6J mice (data not shown), As shown in Fig. 3A. administration of PQ increased plasma AST levels in both C57BL/6J and MT-null mice. In the C57BL/6J mice the AST level was elevated in a dose-dependent manner. The enhanced level at the low dose of PQ (40 mg/kg b.w.) was higher in the MT-null mice than in the C57BL/6J mice. although not significant. In the MT-null mice, administration of the high dose of PQ (60 mg/kg b.w.) significantly increased plasma AST. but the enhanced level was smaller compared with the mice at the low dose of PQ (40 mg/kg b.w.). This suggests that AST reaches a maximum and declines at higher doses. since the AST level readily reaches a maximum value in PQ-treated animals (Sato, 199 I ). Paraquat did affect renal function in the MT-null mice as assessed by a dose-dependent increase in BUN level (Fig. 3B), whereas in the C57BL/6J mice BUN level was not increased by PQ.

4. Discussion In the MT-null mice. marked susceptibility of tissues to oxidative stresses due to an inability of MT gene expression was expected. Paraquat, a well-known active oxygen generator, causes damage in the liver in normal animals (Bus and Gibson, 1984; Sato, 1991). Paraquat can induce lipid peroxidation in the liver and kidney, though how lipid peroxidation is involved in the mechanism of its toxicity in vivo has not been established. Especially, in the lung there are conflicting results on participation of lipid peroxidation in paraquat toxicity in vivo and in vitro (Steffen and Netter, 1979; Komburst and Mais, 1980). Whether the enhancement of lipid peroxidation reflects a consequence of or a cause of the cell death has not been answered. But. increased lipid peroxidation, at least, reflects formation of paraquat radicals and the superoxide radical production in the liver and kidney. In fact, administration of PQ did not induce MT synthesis. and considerably increased lipid peroxidation and plasma BUN in the MT-null mice conpared with the C57BL/6J mice. Furthermore, a decrease in liver weight as well as lethal effects was more conspicuous in the MT-null mice than in the C57BL/6J mice. The present results suggest that PQ induced liver and renal damage in the MT-null mice. and that this was caused by a deficiency of MT. The evidence obtained by the in viva study is consistent with the in vitro findings that embryonic cells derived from MT-null mice are susceptible to active oxygen produced by paraquat (Lazo et al., 1995). A very interesting finding is that the MT-null mice had

a lower basal level of NP-SH and conversely a higher basal level of LP under normal physiological conditions. The NP-SH seems to be consumed by exposure to oxidative stresses in the MT-null mice because of MT deficiency. The data indicate that NP-SH, including GSH. could be the first defensive process for oxidative damage, because a large amount of GSH exists in the cells, and GSH can non-specifically react with any species of free radicals. The data also suggest that the total antioxidative capacity in the liver is lower in the MT-null mice than in the C57BL/6J mice. We have reported that exposure to cytokines such as tumor necrosis factor and interleukin-6 induces expression of not only MT mRNA but also Mn-superoxide dismutase (SOD) mRNA. and have suggested that both constituents cooperatively play antioxidative roles in the liver (Sato et al.. 1995). However, to what extent the increased MT contributes to antioxidative action by cytokines is not known. The present study confirmed that MT is able to scavenge active oxygen species. Since activities of the antioxidant factors other than MT. viz. GSH peroxidases, SOD, catalase, or ceruloplasmin, may change in the MTnull mice in order to cope with MT deficiency, studies on this aspect are under investigation in our laboratories. A putative role of MT as an antioxidant has so far been reported in cells and tissues in which a large amount of MT is accumulated by exposure to heavy metals such as Zn (Sato and Bremner. 1993). However, there are possibilities that an antioxidative action of Zn is caused not only by the metal induced-MT but also by action of stabilizing membranes by the Zn ions, thereby preventing lipid peroxidation. The present study showed that a constitutively expressed, low basal level of MT is able to protect tissues from oxidative damage. Using MT-null mice, the most widely postulated function of the MTs (i.e. to protect against toxicity of heavy metals: Kagi and Kojima. 1987) has been confirmed, since MT-null mice are more susceptible to hepatic poisoning by Cd than wild type animals (Michalska and Choo, 1993; Masters et al., 1994). In addition. the present study demonstrated that MT can play a role as an antioxidant in tissues after a single administration of paraquat. Chronic administration of paraquat may cause remarkable differences in the response of normal and MT-null mice, because the constitutive level of MT in the liver of normal mice is rather low and the half-life of oxygen radicals formed from paraquat is very short. The ability of radical scavenging by repeated exposure to oxidative stress in MT-null mice is under investigation in our laboratories. These mice are a useful in vivo model to allow study of proposed function for MTs, for example, a role of MT and essential metals such as Zn and Cu during early developmental stages. To establish the biological roles of MT, various kinds of detailed studies using different oxygen radical generators, different strains and high concentrations of oxygen are required.

Acknowledgements The authors are grateful to Mr. T. Oda for his kind cooperation in the breeding of MT-null mice at NIES. and to Dr. J Beattie, Rowett Research Institute for carefully reading the manuscript. One of us (M.P.A) received a fellowship from Science and Technology Agency, Japan and took a leave of absence from the Molecular Biology Institute, Sofia, Bulgaria.

References Baldwin, R.C., A. Pasi. J.T. MacGregor and C.H. Hine. 1975. The rate> of radical formation from the dipyridium herbicides paraquat, diquat. and marfamaquat in homogenates of rat lung, kidney. and liver: An inhibitory effect of carbon monooxide. Toxicol. Appl. Pharmacol. 32. 298.

Bus, J.S. and J.E. Gibson, 1984, Paraquat: model for oxidant-initiated toxicity, Environ. Health Perspect. 55. 37. Costa, L.C. and S.D. Murphy, 1986, Effect of diethyl maleate and other glutathione depletors on protein synthesis, Biochem. Pharmacol. 19, 3383.

Ellman,

G.L.,

1959. Tissue sulfhydryl

groups.

Arch.

Biochem.

Biophyh.

82. 70.

Kagi, J.H.R. and Y. Kojima, 1987. Chemistry and biochemistry ot metallothionein, Experientia 52 (suppI.). 25. Karina. B., H. Lohrer. M. Karin and P. Herlich, 1990, Overexpressed metallothionein IIA gene protects Chinese Hamster ovary cells from killing by alkylating agents. Proc. Natl. Acad. Sci. USA 87, 2710. Komburqt, D.J. and R.D. Mais. 1980. The effect of paraquat on microsomal lipid peroxidation in vitro and in viva. Toxicol. Appl. Pharmacol. 523,

Lwo.

323.

J.S.. Y. Kondo, D. Dellapiazza, A.E. Michalska, K.H.A. Choo and B.R. Pitt, 1995, Enhanced sensitivity to oxidative stress in cultured embryonic cells from transgenic mice deficient in metallothlonein I and II genes, J. Biol. Chem. 270. 5506.

Masters. B.A., E.D. Kelly. C.I. Quaife. R.L. Mrinster and R.D. Palmiter, 1994, Targeted disruption of metallothionein I and II genes increases sensitivity to cadmium, Proc. Natl. Acad. Sci. USA 91. 584. Matsubara, J., T. Shida. K. Ishioka, S. Egawa. T. Inada and K. Machida, 1986. Protective effect of zinc against lethality in irradiated mice, Environ. Res. II. 558. Mello-Filho, A.C.. L.S. Chub&u and R. Meneghini, 1988, V79 Chinese hamster cells rendered to high cadmium concentrations also become resistant to oxidative stress. Biochem. J. 256, 475. Michalska. A.E. and K.H.A. Choo, 1993, Targeting and germ-line transmission of a null mutation at the metallothionein I and II loci in mouse, Proc. Natl. Acad. Sci. USA. 90, 8088. Naganuma. A., M. Satoh and N. Imura. 1988, Specific reduction of toxic side effects of adriamycin by induction of metallothionein in mice, Jpn. J. Cancer Res. (Gann) 79, 406. Nishimura. H.. N. Nishimura and C. Tohyama, 1990, Localization of metallothionein in the genital organs of the male rat, J. Histochem. Cytochem. 38. 927. Sato. M.. 1991, Dose-dependent increases in metdllothionein synthesis in the lung and liver of paraquat-treated rats, Toxicol. Appl. Pharmacol. 107. 9x. Sato. M. and I. Bremner, 1993, Metallothionein and oxygen radicals. Free Rad. Biol. Med. IS. 325. Sato. M., M. Sasaki and H. Hojo, 1995, Antioxidative roles of metallothionein and manganese superoxide-dismutase induced by tumor necrosi\ factor-alpha and interleukin-6, Arch. Biochem. Biophys. 316. 738.

Steffen, C. and K.J. Netter, 1979, On the mechanism of paraquat action on microsomal oxygen reduction and its relation to lipid peroxidation. Toxicol. Appl. Pharmacol. 37. 593. Thomalley, P.J. and M. Vasak, 1985, Possible role for metallothionein in protection against radiation-induced oxidative stress: Kinetics and mechanism of its reaction with superoxide and hydroxy radicals. Biochim. Biophys Acta 827. 36. Uchiyama, M. and A. Mihara. 1978, Determination of malonaldehyde precursor in tissues by thiobarbituric acid test, Anal. Biochem. 86, 271. Vuska. M., N. Neskovic, S. Vitrovic and V. bran, 1983, Subacute toxicity of paraquat in rats - Biochemical effects. Ecotoxicol. Environ. Safety 7. 475.