Nutrient Interactions and Toxicity

Transcription

1 Nutrient Interactions and Toxicity Acute Valproate Administration Impairs Methionine Metabolism in Rats 1 Natalia Úbeda, 2 Elena Alonso-Aperte and Gregorio Varela-Moreiras Sección de Nutrición, Bromatología y Dietética, Facultad de Ciencias Experimentales y de la Salud, Universidad San Pablo CEU, Madrid, Spain ABSTRACT Valproate (VPA) is a drug widely used to treat epilepsy, but it has serious adverse effects including hepatotoxicity, teratogenicity and antifolate activity. The mechanism underlying VPA toxicity is unclear although an interaction with folate and other metabolites involved in methionine metabolism has been suggested. The present study was undertaken to evaluate potential changes in the metabolic function of the methionine cycle after acute exposure to a single dose of valproate. Female Wistar rats (n 30) were treated with 400 mg/kg of VPA. Different groups of six rats were killed at 1 (t1), 3 (t3), 6 (t6), 9 (t9), and 24 (t24) hours after the injection. One group of rats was untreated (n 6) and was considered the control group. The most pronounced effects of VPA administration were observed 1 h after drug injection. VPA induced a 56% reduction in methionine adenosyltransferase activity and a 54% reduction in plasma vitamin B-6. Increases in the hepatic concentration of S-adenosylhomocysteine and oxidized glutathione, and a reduction in the S-adenosylmethionine/S-adenosylhomocysteine transmethylation ratio also occurred at 1 h. All of these alterations, however, were normalized within 24 h, parallel with a decrease in serum VPA concentration. The acute effects of VPA suggest that the alterations in the methionine cycle could be the common mechanism underlying the hepatotoxic, teratogenic and antifolate effects of the drug. J. Nutr. 132: , KEY WORDS: valproate methionine adenosyltransferase glutathione vitamin B-6 rats. Valproate (VPA, 3 2-n-propylpentanoic acid) is a widely used drug in the therapy of diverse forms of epilepsy and neurological disorders. However, it shows important potential adverse effects including hepatotoxicity, teratogenicity and antifolate activity. Hepatotoxicity is rare but can be lethal. Several studies (1,2) show two types of hepatotoxic effects, i.e., one that is dose related and reversible, and another that is dose independent and irreversible. The first effect is characterized by an elevation in liver enzyme activities, mainly transaminases, and no clinical symptoms. The second effect is much more serious and can lead to hyperammonemia, hypoglycemia, coma and death. During pregnancy, VPA therapy results in a high incidence of spina bifida aperta, 5- to 20-fold higher than in an untreated population (3 5). In animal models, VPA administration has been frequently used as a common strategy to induce neural tube defects (NTD) (6 8). VPA reduces blood folate levels and causes folate deficiency in both humans (9) and laboratory animals (10,11). The mechanisms underlying the toxic effects of VPA are unclear and have been investigated in different studies. Hepatotoxicity has been suggested to be a consequence of carnitine deficiency [reviewed in (12)] and also to be an oxidative effect. 1 Supported by San Pablo CEU University (Madrid, Spain). 2 To whom correspondence should be addressed. nubeda@ceu.es. 3 Abbreviations used: AdoHcy, S-adenosylhomocysteine; AdoMet, S-adenosylmethionine; BHMT, betaine homocysteine methyltransferase; GSH, reduced glutathione; GSSG, oxidized glutathione; MAT, methionine adenosyltransferase; MS, methionine synthase; NTD, neural tube defects; PL, pyridoxal; VPA, valproate. VPA treatment has been associated with a deficiency of selenium, which is required for glutathione peroxidase activity (13). Furthermore, depletion of glutathione has been observed in animals treated with VPA (14,15). To explain VPA teratogenesis, an interaction with folic acid metabolism has been suggested (11,16). In nonepileptic humans, folate supplementation prevented 70% of NTD cases (17,18). However, when the supplementation was assayed in VPA-induced NTD in animal models, results were unsatisfactory (19 21), suggesting that folate might not be the only metabolite participating in VPA teratogenesis. Folate serves as a methyl donor to homocysteine, which is converted to methionine, in a reaction catalyzed by the vitamin B-12 dependent methionine synthase (MS). This is one of the main reactions in the methionine cycle, a metabolic route that leads to the synthesis of S-adenosylmethionine (AdoMet) through the enzyme methionine adenosyltransferase (MAT). AdoMet is the main methyl group donor in many transmethylation reactions (e.g., DNA-, proteins, lipids), yielding S-adenosylhomocysteine (AdoHcy). AdoHcy is hydrolyzed to homocysteine, which can then be remethylated to methionine [via MS or betaine-homocysteine methyltransferase (BHMT), in hepatic tissue], thus completing the cycle, or be degraded via the transsulfuration pathway in which vitamin B-6 participates and leads to glutathione synthesis (22) (Fig. 1). An alteration in the methionine cycle, likely in the remethylation pathway, has been proposed to explain the occurrence of NTD in nonepileptic humans (23,24), but this hypothesis has not been sufficiently studied after exposure to /02 $ American Society for Nutritional Sciences. Manuscript received 8 November Initial review completed 3 January Revision accepted 28 May

2 2738 ÚBEDA ET AL. FIGURE 1 The methionine cycle in liver. THF: tetrahydrofolate. VPA. In mice, Wegner and Nau (16) demonstrated that VPA administration during organogenesis interfered with folate metabolism and decreased the levels of formylated folates. This might explain some aspects of VPA teratogenesis. More recently, Hishida and Nau (25) reported that a reduction in methionine levels, accompanied by inhibition of homocysteine remethylation, increases in aminothiols and decreases in glutathione, was associated with the induction of VPA-induced NTD in mice. In pregnant rats, we showed that VPA affects methionine synthesis by altering methionine synthase activity, an effect that impairs methionine availability and disrupts the methylation cycle, thereby inducing maternal and fetal DNA hypomethylation (26). All of the studies showed a relationship between VPA administration and an alteration in the methionine cycle that could explain VPA-induced teratogenesis. In the present study, we decided to examine this relationship more fully in an animal model using a single dose of VPA. A detailed time course of biochemical changes after acute exposure to the drug could be very valuable not only to define the mechanism of VPA-induced teratogenesis, but also to further understand VPA-induced hepatotoxicity. MATERIALS AND METHODS Female Wistar rats (n 36; g; Animal Service, Universidad San Pablo-CEU, Madrid, Spain) were used. Rats consumed a nonpurified laboratory diet (RMM, Harlan Interfauna Iberica, S.A, Barcelona, Spain) and water ad libitum. After overnight food deprivation, 400 mg/kg of VPA was administered subcutaneously, and different groups of six rats were killed 1 (t1), 3 (t3), 6 (t6), 9 (t9) and 24 (t24) h after injection. Food deprivation was continued from the time of injection until the time of killing. One group of rats (n 6) was similarly deprived of food overnight but was untreated (control). Sodium valproate (VPA) was obtained from Sigma-Aldrich Chemicals (Steinheim, Germany) and prepared as an aqueous solution. Rats were anesthetized with CO 2 and killed by decapitation at different time points according to the experimental groups (1, 3, 6, 9 and 24 h after injection). Whole blood was collected from all rats and the plasma and serum were separately obtained by centrifugation (1,100 g, 4 C, for 15 min) and kept at 20 C until analyzed. Livers were promptly removed, frozen in liquid nitrogen and stored at 80 C for further analyses. All animal experiments were undertaken according to the guidelines from the European Community Council (27). Amino acids and derivatives. Plasma homocysteine levels were determined by HPLC, using a Chromsystems Reagent Kit for HPLC analysis of homocysteine in plasma (Chromsystems, Munich, Germany), which uses a simple isocratic HPLC system with an attached fluorescence detector ( Ex 385 nm; Em 515 nm). During sample preparation, homocysteine is released from protein by a reduction step; after precipitation, it is subjected to a precolumn derivatization. Methionine was quantified using serum samples deproteinized by ultrafiltration and analyzed using a Beckman System 6300 High performance Amino Acid Analyzer (Beckman Instruments, Palo Alto, CA) following the modifications described by Andersson et al. (28). Hepatic AdoMet and AdoHcy levels were determined by HPLC according to the method described by Fell et al. (29), with some modifications (30). Aliquots of frozen liver ( 100 mg) were homogenized in 4 volumes of 0.4 mol/l perchloric acid, and then centrifuged at 10,000 g, 4 C, for 10 min. The clear supernatants were removed, filtered and appropriate aliquots were analyzed for AdoMet and AdoHcy. Vitamins. Serum folate levels were measured by chemiluminescence, using an Automated Chemiluminescence System (Ciba-Corning ACS, Madrid, Spain). Vitamin B-6 is the generic term for three natural pyridine derivatives and their 5-phosphate esters. Pyridoxal (PL)-5-phosphate is the active coenzyme form and is the main component of several variants in plasma. Plasma PL-5-phosphate concentration (now referred to as vitamin B-6) was determined by HPLC, using a Chromsystems Reagent Kit for HPLC analysis of vitamin B-6 (Chromsystems), which uses a simple isocratic HPLC system with an attached fluorescence detector ( Ex 300 nm; Em 400 nm). With this method, the derivatization necessary for the detection of vitamin B-6 takes place automatically when the sample is applied to the HPLC column because the derivatization reagent is present in the mobile phase. Glutathione. Hepatic reduced (GSH) and oxidized (GSSG) glutathione were measured by HPLC according to the method described by Reed et al. (31). Livers were homogenized in 4 volumes of 50 mmol/l Tris buffer (ph 8) containing 0.3 mol/l sucrose and 0.1 mmol/l EGTA. The homogenate was mixed with 0.6 mol/l trichloroacetic acid and 100 mmol/l iodoacetic acid and incubated in the dark for 30 min. Dinitrofluorobenzene (1.5%) was added and the mixture was incubated overnight at 4 C. The homogenate was then centrifuged at 4000 g for 10 min and the supernatant was again centrifuged at 13,000 g for 100 min to obtain the supernatant for GSH and GSSG assays by HPLC. Enzymes. MAT, BHMT and MS activities were measured in liver extracts using radioenzymatic assays as described by Martín- Duce et al. (32), Finkelstein and Mudd (33) and Keating et al. (34), respectively. Total protein content in liver extracts was measured by the method of Bradford (35). Valproate. Serum valproate was determined by fluorescence polarization immunoassay using a TDx System and an Abbot Reagent kit (South Pasadena, CA). Statistics. Results were expressed as means SEM, n 6. Data were subjected to one-way ANOVA and when the F-test was significant (P 0.05), multiple comparisons between means were done using Student s t test with the Bonferroni adjustment for inequality and the Student-Newman-Keuls test. Significant differences were further analyzed by the Kruskal-Wallis nonparametric test (SYSTAT Version 5.0, Systat, Chicago, IL and Excel for Windows, Microsoft Office Version 7.0). RESULTS VPA reduced hepatic MAT in all groups, except at 24 h after treatment (Fig. 2). One hour after VPA administration (group t1), there was a significant increase in hepatic AdoHcy concentration (Table 1). This was still evident 3 h after treatment (group t3), but AdoHcy levels were not different from controls in the t6, t9 and t24 groups. AdoMet concentration tended to be lower than in controls in the t1 group (P 0.11). The slight reduction in AdoMet and the significant increase in AdoHcy were reflected in a significant reduction in the AdoMet/AdoHcy concentration ratio 1 h after VPA administration. Plasma vitamin B-6 concentrations were reduced by VPA treatment 1 h after administration and then increased progressively with time to reach control values (Fig. 3). GSH concentration did not differ from the control in any of the

3 VALPROATE IMPAIRS METHIONINE METABOLISM IN RATS 2739 FIGURE 2 Effect of valproate administration on hepatic methionine adenosyltransferase (MAT) activity in rats at 1, 3, 6, 9 and 24 h after treatment and in the untreated control group. Values are means SEM, n 6. Means without a common letter differ, P VPA-treated groups but the GSH concentration was lower in groups t3 and t9 than in group t1 (Table 2). The GSSG concentration was greater in group t1 than in all other groups. Serum methionine concentrations were significantly lower in groups t3 and t6 compared with control and t1 groups (Table 3). Plasma homocysteine concentrations were significantly reduced 3 and 24 h after VPA injection compared with group t1 but none of the VPA-treated groups differed. Serum folate concentrations and hepatic activities of MS and BHMT were not affected by VPA treatment (Table 3). One hour after treatment, serum VPA concentrations were close to 2.5 mmol/l (Fig. 4). The concentration of the drug decreased by 51% 3 h after the injection, and continued to decrease until 24 h after treatment when it had almost disappeared from serum. Significant correlations between serum VPA levels and several variables were found: plasma vitamin B-6, r 0.60, P 0.001; hepatic AdoHcy, r 0.50, P 0.01; and the AdoMet/AdoHcy ratio, r 0.39, P DISCUSSION The most pronounced effect of valproate administration was observed after 1 h, i.e., treatment with 400 mg/kg of valproate induced a 56% reduction in MAT activity. This is the first time such an effect has been described although MAT inhibition has been studied under different circumstances. TABLE 1 Hepatic concentrations of S-adenosylmethionine (AdoMet), S-adenosylhomocysteine (AdoHcy) and AdoMet/AdoHcy ratio in rats at 1, 3, 6, 9 and 24 h after valproate administration and in the untreated control rats1 Group AdoMet AdoHcy nmol/g AdoMet/ AdoHcy Control b a t a b t a a t b a t b a t b a 1 Values are means SEM, n 6; means in a column with different superscript letters differ, P FIGURE 3 Effect of valproate administration on plasma vitamin B-6 levels in rats at 1, 3, 6, 9 and 24 h after treatment and in the untreated control group. Values are means SEM, n 6. Means without a common letter differ, P Vitamin B-6 was measured as pyridoxal-5-phosphate. Direct measure of MAT activity in liver biopsies from cirrhotic patients showed a marked reduction in its activity (32). In animal models, hepatotoxic agents such as carbon tetrachloride and buthionine sulfoximine have markedly reduced MAT activity and decreased hepatic glutathione and AdoMet levels (36,37). Recently, Lu et al. (38) described a predisposition to liver injury and an increase in liver proliferation in a MAT knockout mouse model. MAT is the enzyme responsible for the synthesis of AdoMet using methionine and ATP as substrates (Fig. 1). Its importance is due to the fact that AdoMet is the main methyl donor for most of the transmethylation reactions in the cell, and therefore is involved in many critical biological functions. The reduction of MAT activity in injured liver has not been clearly explained, but it could be related to oxidative stress. In vitro, the presence of GSSG can inhibit the enzyme activity and this effect can be modulated by GSH, although GSH alone has no effect on the enzyme. Therefore, MAT activity in vitro is regulated by the GSH/GSSG ratio (39). In vivo, the reduction in MAT activity is concomitant with reduced cellular levels of glutathione, both in human cirrhosis and in animal models intoxicated with buthionine sulfoximine and carbon tetrachloride (36,37). Glutathione depletion seems to be a direct consequence of VPA treatment in rats (14,15), and depression of glutathione levels has been proposed recently as a possible mechanism underlying VPA ter- TABLE 2 Hepatic concentrations of reduced (GSH) and oxidized glutathione (GSSG) in rats at 1, 3, 6, 9 and 24 h after valproate administration and in the untreated control rats1 Group GSH GSSG nmol/g Control ab b t a a t b b t ab b t b b t ab b 1 Values are means SEM, n 6; means in a column with different superscript letters differ, P 0.05.

4 2740 ÚBEDA ET AL. TABLE 3 Serum methionine, plasma homocysteine and serum folate concentrations and hepatic enzyme activities involved in the remethylation pathway in rats at 1, 3, 6, 9 and 24 h after valproate administration and in the untreated control rats1 Group Methionine Homocysteine Folate MS 2 BHMT 2 mol/l nmol/l nmol/(h mg protein) Control a ab t a a t b b t b ab t ab ab t ab b Values are means SEM, n 6; means in a column with different superscript letters differ, P MS, methionine synthase; BHMT, betaine homocysteine methyltransferase. atogenesis in mice (25). These studies, however, did not discriminate between the oxidized and reduced forms of glutathione. It remains unclear whether the first insult is a depression of glutathione that leads to MAT inactivation or an inhibition of MAT activity that leads to glutathione depletion. In the transsulfuration pathway, homocysteine is converted to cystathionine and cysteine, which is further metabolized and excreted as inorganic sulfate, or it is used for the synthesis of glutathione (Fig. 1). The depletion of liver glutathione levels as a result of liver damage could lead to inactivation of MAT, but this in turn could result in a further decrease in glutathione levels and thus worsen the deficiency of MAT (37). In our study, we found significant reductions in MAT activity 1, 3, 6 and 9 h after VPA administration. Enzyme activity had recovered by 24 h, when VPA had almost disappeared from the serum. We did not observe a depletion in total glutathione due to VPA but a significant increase in the oxidized form (GSSG) 1 h after injection. On the basis of these findings, we propose that VPA inhibits MAT in vivo. The mechanism could be related to the increase in GSSG because this oxidized form inhibits MAT in vitro (39). Because MAT inactivation is strongly related to liver injury (see above), these metabolic conditions could both be related to the hepatotoxicity associated with VPA. Aliphatic compounds may inhibit MAT in vitro (40) and it is possible that VPA has that effect. We used crude liver extracts for our assays and did not test for the presence of a soluble inhibitor. The effect of VPA on plasma vitamin B-6 was also marked. There was a negative correlation between serum VPA and vitamin B-6 levels (r 0.60, P 0.001), with a 54% FIGURE 4 Valproate serum levels in rats at 1, 3, 6, 9 and 24 h after drug administration. Values are means SEM, n 6. decrease in plasma B-6 concentration 1 h after VPA administration. This effect has not been reported before. We know that chronic treatment with anticonvulsant drugs in humans may induce vitamin B-6 deficiency (41), and that vitamin B-6 deficiency impairs homocysteine/methionine metabolism (42 44), but these observations are due to chronic drug treatment. In our study, the effect of VPA on vitamin B-6 was acute. The explanation we propose is that the effects result from a disregulation of the methionine cycle. One of the major regulatory sites is the distribution of homocysteine between remethylation and cystathionine synthesis (22), and an increased need for glutathione synthesis might affect this distribution (45). In addition, the activity of cystathionine- synthase is enhanced under oxidative conditions (46). Changes in the ambient redox potential could affect the flux of homocysteine between the two competing pathways, increasing this flux two- to threefold through the transsulfuration pathway to replenish the glutathione pool, which was diminished in response to an oxidant insult (47). In turn, homocysteine levels tended to be higher (P 0.06) 1 h after VPA administration. These results suggest that VPA treatment and the impairment of MAT activity could be inducing increases in transsulfuration enzymes to prevent homocysteine elevation. The induction of the transsulfuration pathway could explain the reduction in vitamin B-6 levels because this vitamin acts as a cofactor for two enzymes involved in the pathway, cystathionine- -synthase and -cystathionase. Serum folate concentration did not change after VPA administration, as we showed previously after drug treatment on d 8, 9 and 10 of pregnancy (26). Conversely, VPA has antifolate activity, in both humans (9) and laboratory animals (10,11). These apparently contradictory observations suggest that VPA might indeed alter folate metabolism without reducing total folate concentration. VPA also induced a marked and significant increase in hepatic AdoHcy concentration. The reaction catalyzed by S-adenosylhomocysteine hydrolase is reversible, thermodynamically favoring AdoHcy synthesis. However, the reaction normally functions in the opposite direction because the products, homocysteine and adenosine, are removed enzymatically (see Fig. 1) (22). Nevertheless, an alteration in methionine/ homocysteine metabolism could lead to accumulation and conversion into AdoHcy, with concomitant elevation in AdoHcy levels. We previously showed this elevation in AdoHcy due to impaired homocysteine metabolism, in both VPA-treated rats (26) and carbon tetrachloride induced hepatic injury in rats (48).

5 VALPROATE IMPAIRS METHIONINE METABOLISM IN RATS 2741 Another consequence of MAT inactivation could be a reduction in the availability of AdoMet. We did not find a significant reduction in hepatic AdoMet concentration but VPA significantly reduced the AdoMet/AdoHcy ratio, the so-called transmethylation ratio, 1 h after exposure to the drug. In the same sense, studies in humans (49) and rats (37) have shown that marked reductions in MAT activity do not necessarily lead to reduced AdoMet levels, possibly because the rate of AdoMet utilization is adjusted to the rate of synthesis. However, transmethylation reactions could be inhibited because of a lower availability of AdoMet and/or increased AdoHcy concentration. Moreover, a reduced AdoMet/AdoHcy ratio also inhibits transmethylation reactions (50). In fact, we previously showed genomic DNA hypomethylation in rats exposed to VPA (26). The time-dependent effects of VPA on the methionine cycle deserve further comment. We observed significant correlations between serum VPA levels and the plasma vitamin B-6 decrease, the hepatic AdoHcy increase and the AdoMet/ AdoHcy ratio decrease. At 24 h after VPA administration, when the drug has disappeared from serum, MAT activity and vitamin B-6 plasma levels as well as all of the other variables measured did not differ from the untreated control rats. We do not know whether the effects of VPA on MAT activity and vitamin B-6 would become irreversible due to cyclical drug effects or would even have different consequences due to the chronic administration of the drug. In a previous study, we showed that VPA administered on gestation days 8, 9 and 10 in rats also disrupted the methionine cycle, and in this case, the manifestations were decreased methionine synthase activity and methionine levels on gestation day 21 (26). We have reported for the first time significant reductions in hepatic MAT activity and plasma vitamin B-6 levels in rats exposed to a single dose of VPA (400 mg/kg). We propose that VPA alters methionine metabolism by inhibiting MAT activity in vivo. The mechanism could be related to the increase in GSSG because this form of glutathione inhibits MAT in vitro (39). An alteration of methionine metabolism could be the single mechanism underlying the pathophysiology of the different potentially toxic effects of VPA, i.e., hepatotoxicity, teratogenicity and the antifolate effect, for the following reasons: 1) MAT enzyme activity is impaired in several forms of liver disease; 2) an alteration in the methionine cycle can lead to glutathione depletion, and glutathione is the major detoxificant metabolite in the cell; 3) the methionine cycle is closely related to folate metabolism; and 4) transmethylation reactions are involved in many aspects of cellular function. However, the relationship of these metabolic changes to hepatic injury and teratogenicity remains to be elucidated. ACKNOWLEDGMENTS We thank J. M. Mato and F. Corrales (Universidad de Navarra, Spain) for their help and assistance in the methionine and glutathione determinations; and M. Caamaño (Universidad San Pablo CEU, Madrid, Spain) for VPA determination. LITERATURE CITED 1. Dreifus, F. E., Santilli, R. N., Langer, D. H., Sweeney, K. P., Moline, K. A. & Menander, K. B. (1987) Valproic acid hepatic fatalities: a retrospective review. Neurology 37: Nurge, M. E., Anderson, C. R. & Bates, E. B. (1991) Metabolic and nutritional implications of valproic acid. Nutr. Res. 11: Bjerkedal, T., Czeizel, A., Goujard, J., Kallen, B., Mastroiacova, P., Nevin, N., Oakley, G. & Robert, E. (1982) Valproic acid and spina bifida [letter]. Lancet 2: Robert, E. & Rosa, F. (1983) Valproate and birth defects [letter]. Lancet 2: Lindhout, D. & Meinardi, H. (1984) Spina bifida and in utero exposure to valproate [letter]. Lancet 2: Voorhes, C. V. (1987) Teratogenicity and developmental toxicity of valproic acid in rats. Teratology 35: Ehlers, K., Stürje, H., Merker, H. J. & Nau, H. (1992) Valproic acid induced spina bifida: a mouse model. Teratology 45: Menegola, E., Broccia, M. L., Nau, H., Prati, M., Ricolfi, R. & Giavini E. (1996) Teratogenic effects of sodium valproate in mice and rats at midgestation and at term. Teratog. Carcinog. Mutagen. 16: Hendel, J., Dam, M., Gram, L., Winkel, P. & Jorgensen H. (1984) The effects of carbamazepine and valproate on folate metabolism in man. Acta Neurol. Scand. 69: Smith, D. B. & Carl, J. F. (1982) Interactions between folates and carbamazepine or valproate in the rat. Neurology 32: Carl, G. F. (1986) Effect of chronic valproate treatment on folate dependent methyl biosynthesis in the rat. Neurochem. Res. 11: Raskind, J. Y. & El-Chaar, G. M. (2000) The role of carnitine supplementation during valproic acid therapy. Ann. Pharmacother. 34: Hurd, R. W., Vanrisnvelt, H. A., Wilder, B. J., Maenhaut, W. & Dereu, L. (1984) Selenium, zinc and copper changes with valproic acid: possible relation to drug side effects. Neurology 34: Raza, M., Al-Bekairi, A. M., Ageel, A. M. & Qureshi, S. (1997) Biochemical basis of sodium valproate hepatotoxicity and renal tubular disorder: time dependence of peroxidative injury. Pharmacol. Res. 35: Seçkin, S., Basaran-Küçükgergin, C. & Müjdat, U. (1999) Effect of acute and chronic administration of sodium valproate on lipid peroxidation and antioxidant system in rat liver. Pharmacol. Toxicol. 85: Wegner, C. & Nau, H. (1992) Alteration of embryonic folate metabolism by valproic acid during organogenesis: implications for mechanism of teratogenesis. Neurology 42 (Suppl.): MRC Vitamin Study Research Group. (1991) Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet 338: Czeizel, A. & Dudás, I. (1992) Prevention of the first occurrence of neural tube defects by periconceptional vitamin supplementation. N. Engl. J. Med. 327: Hansen, D. K. & Grafton, T. F. (1991) Lack of attenuation of valproic acid induced effects by folinic acid in rat embryos in vitro. Teratology 43: Elmazar, M. A., Thiel, R. & Nau, H. (1992) Effect of supplementation with folinic acid, vitamin B6, and vitamin B12 on valproic acid induced teratogenesis in mice. Fund. Appl. Toxicol. 18: Hansen, D. K., Grafton, T. F., Dial, S. L., Gehring, T. A. & Siitonen, P. H. (1995) Effect of supplemental folic acid on valproic acid induced embryotoxicity and tissue zinc levels in vivo. Teratology 52: Finkelstein, J. D. (1990) Methionine metabolism in mammals. J. Nutr. Biochem. 1: Steegers-Theunissen, R.P.M., Boers, G.H.J., Trijbels, F.J.M., Finkelstein, J. D., Blom, J. G., Thomas, C.M.G., Borm, G. F., Wouters, M.G.A.J. & Eskes, T.K.A.B. (1994) Maternal hyperhomocysteinemia: a risk factor for neural tube defects? Metab. Clin. Exp. 43: Mills, J. L., McPartlin, J. M., Kirke, P. N., Lee, Y. J., Conley, M. R., Weir, D. G. & Scott, J. M. (1995) Homocysteine metabolism in pregnancies complicated by neural-tube defects. Lancet 345: Hishida, R. & Nau, H. (1998) VPA-induced neural tube defects in mice. I. Altered metabolism of sulfur amino acids and glutathione. Teratog. Carcinog. Mutagen. 18: Alonso-Aperte, E., Úbeda, N., Achón, M., Pérez-Miguelsanz, J. & Varela- Moreiras, G. (1999) Impaired methionine synthesis and hypomethylation in rats exposed to valproate during gestation. Neurology 52: European Community Council (1986) Directional guides related to animal housing and care N L358/1 to N L358/28. Barcelona, Spain. 28. Anderson, A., Brattstrom, L., Isaksson, A., Israelson, B. & Hultberg, B. (1989) Determination of homocysteine in plasma by ion-exchange chromatography. Scand. J. Clin. Lab. Invest. 49: Fell, D., Benjamin, L. E. & Steele, R. D. (1985) Determination of adenosine and adenosyl derivatives of sulfur amino acids in rat liver by high performance liquid chromatography. J. Chromatogr. 345: Miller, J. W., Nadeau, M. R., Smith, J., Smith, D. & Selhub, J. (1994) Folate deficiency-induced homocysteinemia in rats: disruption of S-adenosylmethionine s coordinate regulation of homocysteine metabolism. Biochem. J. 298: Reed, D. J., Babson, J. R., Beatty, P. W., Brodie, A. E., Ellis, W. W. & Potter, D. W. (1980) High-performance liquid chromatography analysis of nanomole levels of glutathione, glutathione disulfide, and related thiols and disulfides. Anal. Biochem. 106: Martín-Duce, A., Ortiz, P., Cabrero, C. & Mato, J. M. (1988) S-Adenosyl-L-methionine synthetase and phospholipid methyltransferase are inhibited in human cirrhosis. Hepatology 8: Finkelstein, J. D. & Mudd, S. H. (1967) Trans-sulfuration in mammals: the methionine-sparing effect of cystine. J. Biol. Chem. 242: Keating, J. N., Weir, D. G. & Scott, J. M. (1985) Demonstration of methionine synthetase activity in the intestinal mucosal cells of the rat. Clin. Sci. (Lond.) 3: Bradford, M. M. (1976) A rapid and sensitive method for the determi-

6 2742 ÚBEDA ET AL. nation of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: Corrales, F., Ochoa, P., Rivas, C., Martín Lomas, M., Mato, J. M. & Pajares, M. A. (1991) Inhibition of gluthatione synthesis in the liver leads to S-adenosyl-L-methionine synthetase reduction. Hepatology 14: Corrales, F., Giménez, A., Alvarez, L., Caballería, J., Pajares, M. A., Andreu, H., Parés, A., Mato, J. M. & Rodés, J. (1992) S-Adenosylmethionine treatment prevents carbon tetrachloride-induced S-adenosylmethionine synthase inactivation and attenuates liver injury. Hepatology 16: Lu, S. C., Alvarez, L., Huang, Z. Z., Chen, L., An, W., Corrales, F. J., Avila, M. A., Kanel, G. & Mato, J. M. (2001) Methionine adenosyltransferase 1A knockout mice are predisposed to liver injury and exhibit increased expression of genes involved in proliferation. Proc. Natl. Acad. Sci. USA 98: Pajares, M. A., Durán, C., Corrales, F., Pliego, M. M. & Mato, J. M. (1992) Modulation of rat liver S-adenosylmethionine synthetase activity by glutathione. J. Biol. Chem. 267: Coulter, A. W., Lombardini, J. B. & Talalay, P. (1974) Structural and conformational analogues of L-methionine as inhibitors of the enzymatic synthesis of S-adenosyl-L-methionine. I. Saturated and unsaturated aliphatic amino acids. Mol. Pharmacol. 10: Schwaninger, M., Ringleb, P., Winter, R., Kohl, B., Fiehn, W., Rieser, P. A. & Walter Sack, I. (1999) Elevated plasma concentrations of homocysteine in antiepileptic drug treatment. Epilepsia 40: She, Q. B., Nagao, I., Hayakawa, T. & Tsuge, H. (1994) A simple HPLC method for the determination of S-adenosylmethionine and S-adenosylhomocysteine in rat tissues: the effect of vitamin B6 deficiency on these concentrations in rat liver. Biochem. Biophys. Res. Commun. 205: She, Q. B., Hayakawa, T. & Tsuge, H. (1995) Alteration in the phosphatidylcholine biosynthesis of rat liver microsomes caused by vitamin B6 deficiency. Biosci. Biotechnol. Biochem. 59: Miller, J. W., Nadeau, M. R., Smith, D. & Selhub, J. (1994) Vitamin B6 deficiency vs folate deficiency: comparison of responses to methionine loading in rats. Am. J. Clin. Nutr. 59: Finkelstein, J. D. (1998) Methionine metabolism: the next 10 years. In: Methionine Metabolism: Molecular Mechanism and Clinical Implications (Mato, J. M. & Caballero, A., eds.), pp CSIC Press, Madrid, Spain. 46. Taoka, S., Ohja, S., Shan, X., Kruger, W. D. & Banerjee, R. (1998) Evidence of heme-mediated redox regulation of human cystathionine- -synthase activity. J. Biol. Chem. 273: Mosharov, E., Cranford, M. R. & Banerjee, R. (2000) The quantitatively important relationship between homocysteine metabolism and glutathione synthesis by the transsulfuration pathway and its regulation by redox changes. Biochemistry 39: Varela-Moreiras, G., Alonso-Aperte, E., Rubio, M., Gassó, M., Deulofeu, R., Alvarez, L., Caballería, J., Rodés, J. & Mato, J. M. (1995) Carbon tetrachloride induced hepatic injury is associated with global DNA hypomethylation and homocysteinemia: effects of S-adenosylmethionine treatment. Hepatology 22: Cabrero, C., Martín-Duce, A., Ortiz, P., Alemany, S. & Mato, J. M. (1988) Specific loss of high-molecular-weight form of S-adenosylmethionine synthetase in human liver cirrhosis. Hepatology 8: Cantoni, G. L., Richards, H. H. & Chiang, P. K. (1979) Inhibitors of S-adenosylhomocysteine hydrolase and their role in the regulation of biological methylation. In: Transmethylation (Usdin, E., Borchardt, R. T. & Creveling, D. E., eds.), pp Elsevier/North-Holland, New York, NY.