Arch Toxicol (2004) 78: 25–33 DOI 10.1007/s00204-003-0470-y

O RG AN T OX IC ITY A N D M E CH AN I SM S

Marta Blanca Mazzetti Æ Marı´ a Cristina Taira Sandra Marcela Lelli Æ Eduardo Dascal Juan Carlos Basabe Leonor Carmen San Martı´ n de Viale

Hexachlorobenzene impairs glucose metabolism in a rat model of porphyria cutanea tarda: a mechanistic approach Received: 8 August 2002 / Accepted: 7 April 2003 / Published online: 29 July 2003
Springer-Verlag 2003

Abstract Hexachlobenzene (HCB), one of the most persistent environmental pollutants, induces porphyria cutanea tarda (PCT). The aim of this work was to analyze the effect of HCB on some aspects of glucose metabolism, particularly those related to its neosynthesis in vivo. For this purpose, a time-course study on gluconeogenic enzymes, pyruvate carboxylase (PC), phosphoenolpyruvate carboxykinase (PEPCK), glucose-6-phosphatase (G-6Pase) and on pyruvate kinase (PK), a glycolytic enzyme, was carried out. Plasma glucose and insulin levels, hepatic glycogen, tryptophan contents, and the pancreatic insulin secretion pattern stimulated by glucose were investigated. Oxidative stress and heme pathway parameters were also evaluated. HCB treatment decreased PC, PEPCK, and G-6-Pase activities. The effect was observed at an early time point and grew as the treatment progressed. Loss of 60, 56, and 37%, respectively, was noted at the end of the treatment when a considerable amount of porphyrins had accumulated in the liver as a result of drastic blockage of uroporphyrinogen decarboxylase (URO-D) (95% inhibition). The plasma glucose level was reduced (one-third loss), while storage of hepatic glucose was stimulated in a time-dependent way by HCB treatment. A decay in the normal plasma insulin level was observed as fungicide intoxication progressed (twice to four times lower). M. B. Mazzetti Æ M. C. Taira Æ S. M. Lelli L. C. S. M. de Viale (&) Departamento de Quı´ mica Biolo´gica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, C1428BGA, Ciudad Auto´noma Buenos Aires, Argentina E-mail: [email protected] Tel.: +54-11-45763342 Fax: +54-11-45763342 E. Dascal Æ J. C. Basabe Centro de Investigaciones Endocrinolo´gicas (CEDIE). Hospital de Nin˜os, Dr. Ricardo Gutierrez, C1425EDF, Ciudad Auto´noma Buenos Aires, Argentina Present address: L. C. S. M. de Viale O’Higgins 4332, CP 1429, Buenos Aires, Argentina

However, normal insulin secretion of perifused pancreatic Langerhans islets stimulated by glucose during the 3rd and 6th weeks of treatment did not prove to be significantly affected. HCB promoted a time-dependent increase in urinary chemiluminiscence (fourfold) and hepatic malondialdehide (MDA) content (fivefold), while the liver tryptophan level was only raised at the longest intoxication times. These results would suggest that HCB treatment does not cause a primary alteration in the mechanism of pancreatic insulin secretion and that the changes induced by the fungicide on insulin levels would be an adaptative response of the organism to stimulate gluconeogenesis. They showed for the first time that HCB causes impairment of the gluconeogenic pathway. Therefore, the reduced levels of glucose would thus be the consequence of decreased gluconeogenesis, enhanced glucose storage, and unaffected glycolysis. The impairment of gluconeogenesis (especially for PEPCK) and the related variation in glucose levels caused by HCB treatment could be a consequence of the oxidative stress produced by the fungicide. Tryptophan adds its effect to this decrease in the higher phases of HCB intoxication, where its levels overcome the control values possibly owing to the drastic decline of URO-D. This derangement of carbohydrates leads porphyric hepatocytes to have lower levels of free glucose. These results contribute to our understanding of the protective and modulatory effect that diets rich in carbohydrates have in hepatic porphyria disease. Keywords Gluconeogenesis Æ Glucose Æ Hexachlorobenzene Æ Insulin Æ Porphyria cutanea tarda Abbreviations ALA-S d-aminolevulinate synthase Æ G-6-Pase glucose-6-phosphatase Æ HCB hexachlorobenzene Æ MDA malondialdehyde Æ PEPCK phosphoenolpyruvate carboxykinase Æ PCT porphyria cutanea tarda Æ PC pyruvate carboxylase Æ PK pyruvate kinase Æ TCDD 2,3,7,8-tetrachlorodibenzo-p-dioxin Æ URO-D uroporphyrinogen decarboxylase

26

Introduction Hexachlorobenzene (HCB) is a well-known porphyrinogenic fungicide that caused a massive outbreak of porphyria cutanea tarda (PCT) in Turkey in 1959 and is capable of inducing porphyria in several animal species (Wilson and Kueveruwa 1994). Uroporphyrinogen decarboxylase (URO-D) is the key enzyme of the heme metabolic pathway that is blocked in this porphyria, not only in human, but also in experimental HCB-induced models. This block impairs the regulation of the heme pathway (San Martin de Viale et al. 1977). PCT, the most common form of porphyria in humans (Elder 1998), is clinically characterized by severe cutaneous photosensitivity mediated by uroporphyrin and partially decarboxylated porphyrins. The source of these porphyrins is the liver where the activity of URO-D is decreased. These highly carboxylated porphyrins accumulate in the liver, circulate in plasma, and are excreted in urine (Kappas et al. 1995). Porphyria induced in rats by HCB mimics biochemically the disease in humans and is exacerbated by estrogens, iron and alcohol (San Martin de Viale et al. 1975; Smith and De Matteis 1980). The animal model showed that females are more sensitive than males in acquiring this porphyria (Wainstok de Calmanovici et al. 1991) and that the liver was the organ mainly affected (San Martı´ n de Viale et al. 1977). In spite of the fact that a great deal of information concerning the effect of HCB on the heme metabolic pathway has been reported (Wainstok de Calmanovici et al. 1984), little is known about the effect of the xenobiotic on glucose metabolism. In this respect, hepatic ultraestructural studies in HCB-treated rats showed storage products, presumed to be glycogen, replacing smooth endoplasmic reticulum (Mollenhauer et al. 1975) and an extensive cytoplasm area containing numerous glycogen particles in the hepatocytes of the intermediary zone, while there were few in the centrilobular zone (Bo¨ger et al. 1979). Moreover, studies carried out in hamsters intoxicated with HCB showed there were no positive periodic acid-Schiff structures in the cytoplasm of hepatocytes, thus suggesting that these hepatocytes are unable to synthesize glycogen (Rı´ os de Molina et al. 1996). On the other hand, a diet rich in glucose is known to have a beneficial effect on acute porphyria patients, significantly improving their clinical condition (Kappas et al. 1995). The prevention of acute experimental porphyria through high carbohydrate and/or protein intake (Rose et al. 1961) has been shown to be an example of the effect of glucose, in which carbohydrates prevent d-aminolevulinate synthase (ALA-S) induction (Tschudy et al. 1964). Other studies have demonstrated that a high-fat/high-protein diet enhances the effect of HCB. Conversely, high-carbohydrate diets suppress the effect of HCB, which is more marked in glucose diets (Ivanov et al. 1976).

Weber et al. (1991) and Fan and Rozman (1994) found that in two different rat strains, Sprague-Dawley and Long-Evans, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), one of the most powerful porphyrinogenic PCT agents, induced acute intoxication in which a dosedependent inhibition of the hepatic activity of the enzyme phosphoenolpyruvate-carboxykinase (PEPCK) was observed. The aim of the present work was to study the effect of HCB on some aspects of glucose metabolism, particularly those related to its synthesis. For this purpose, a time-course study on gluconeogenic enzymes pyruvate carboxylase (PC), phosphoenolpyruvate carboxykinase (PEPCK), and glucose-6-phosphatase (G-6-Pase) was performed. The study of pyruvate kinase (PK), a glycolytic enzyme which, depending on its activity, directs substrate flux through either the glycolytic or the gluconeogenic pathway (Pilkis et al. 1986), was also carried out. Plasma glucose and insulin levels, hepatic glycogen content, and the pancreatic insulin secretion pattern stimulated by glucose were also investigated.

Materials and methods Materials HCB (commercial grade) was a gift from Compan˜ı´ a Quı´ mica S.A., Argentina. Coenzyme A, phosophotransacetylase, citrate lyase, pyruvate, deoxyguanosine 5’-diphosphate, lactate dehydrogenase and NADH were obtained from Sigma-Aldrich Chemical Co. Uroporphyrin III was from Porphyrin Products, Logan, Utah, USA. All other chemicals were of the highest available grade. Collagenase was obtained from Serva Feinbiocem, Heidelberg, Germany. Pork monoiodine 125I-insulin was obtained from CENEXA (Facultad de Medicina, Universidad Nacional de la Plata, Argentina), and guinea pig antiporcine insulin antiserum was a generous gift from Dr. R. Gutman, Hospital Italiano, Buenos Aires, Argentina. Rat standard insulin was obtained from Novo Research Laboratories (Bagsvaerd, Denmark). Guinea pig antiporcine insulin antiserum was sufficiently nonspecific as to allow pork-labeled insulin to be displaced by rat insulin.

Animals and treatment Female Wistar rats, weighing 160–180 g at the beginning of the experiments, were used. Animals were maintained on a 12-h- light/ dark cycle and fed Purina 3 diet and water ad libitum. They were randomly assigned to two groups: group I: controls (n = 40); and group II: HCB-treated (n = 40). Two additional control groups of five animals each, with and without 24-h fasting, were tested for enzymatic activity. HCB was administered 5 days/week by stomach tube for 2, 3, 5, or 6 weeks. Each day the animals received a single dose of HCB (1 g/ kg body weight per day) suspended in water (40 mg/ml) containing Tween 20 (0.5 ml/100 ml of suspension). The dose of HCB used in this work was based on previous studies demonstrating that this amount elicits clear manifestation of hepatic porphyria in Wistar rats, without inducing appetite suppression, death, or external alterations (Billi de Catabbi et al. 1997). The low order of toxicity observed was indicative of the minimal absorption of HCB across the intestinal tract when administered in water. In this sense, it has been shown that (14C)HCB suspended in water

27 and administered intragastrically is very poorly absorbed—only 5% of the dose (Koss and Koransky 1975). Animals from the control group received the vehicle alone. All the animals were deprived of feed for 24 h prior to death by decapitation. Immediately before the animals were killed (controls, n=10; HCB-treated, n=10 each time during the different treatment weeks) blood was extracted with a capillar tube from the ophthalmic venous plexus. Blood was collected on heparin, maintained at 0C and centrifuged at 1,000 g for 15 min at 0–4ºC. Part of the plasma obtained was used for glucose determination, and the rest was frozen at )20ºC for insulin quantification.

General procedures For the determination of PEPCK and G-6-Pase activities, liver was homogenized in three volumes of 0.25 M sucrose at 0–4ºC. The homogenate was centrifuged for 30 min at 11,000 g. The resulting supernatant was used for PEPCK determination. The pellet was resuspended in 0.25 M sucrose, 1 mM EDTA pH 7, stored at )80ºC overnight, and then thawed for determination of G-6-Pase. Supernatants for PC and PK determinations were prepared from homogenates in a single 1-h 100,000 g centrifugation step with buffer and concentrations as stated below. All assays of soluble enzymes were carried out with fresh livers immediately after death. For URO-D determination pieces of liver were homogenized with 0.154 M KCl (1:5 w/v) and centrifuged at 11,000 g for 20 min at 0–4ºC. The supernatant was used as source of enzyme.

Enzyme activities PC was measured according to Berndt et al. (1978). Liver was homogenized in five volumes of 50 mM Tris acetate, 5 mM MgSO4, 5 mM EDTA buffer, pH 6.5, and then treated with ultrasound on ice using three 20-s bursts (Perkin Elmer 450 Ultrasonic Homogenizer 4710) to release the particle-associated enzyme. Supernatant (100,000 g) was used to estimate the activity, measuring spectrophotometrically at 340 nm the NADH oxidation. PEPCK was measured according to Petrescu et al. (1979) using deoxyguanosine 5’-diphosphate as nucleotide substrate, and 100,000 g supernatant of as enzyme source. The oxaloacetate formed during the reverse enzymatic reaction was determined spectrophotometrically at 340 nm by reduction with malate dehydrogenase in the presence of NADH. G-6-Pase was measured following Baginski et al. (1974) through the inorganic phosphate released from glucose-6-phosphate. Free inorganic phosphate was determined spectrophotometrically at 700 nm by the color of the phosphomolybdate complex formed, with arsenite as the reducing agent. PK was measured according to Imamura and Tanaka (1982) with the couple NADH-LDH in the supernatant of a 1:3 (w/v) homogenate of rat liver in a 50 mM Tris-HCl buffer, pH 7.5, containing 0.1 M KCl, 5 mM MgSO4, 1 mM EDTA, 0.2 mM fructose-1,6-diphosphate, and 10 mM mercaptoethanol. URO-D was determined according to Wainstok de Calmanovici et al. (1984). Incubation were carried out in a mixture (final volume 1.5 ml) containing: 0.067 M potassium phosphate buffer, pH 6.8, 1 mM reduced GSH, 0.1 mM EDTA, using 4 lM uroporphyrinogen III as substrate, enzyme (3 mg/ml). The porphyrins formed were separated and quantified as free porphyrins by highperformance liquid chromatography (Lim et al. 1983).

Hepatic porphyrin content Porphyrin content was determined spectrophotometrically as free porphyrins in 5% (w/v) HCl. 0.1–0.3 ml of 10% (w/v) homogenates in 0.154 M KCl was used for the assay, as described by San Martı´ n de Viale et al. (1977).

Plasma glucose Fasted glucose levels were determined in plasma using a Glycemia Enzymatic Kit (Wiener Laboratory, Rosario, Santa Fe, Argentina).

Hepatic glycogen content Liver glycogen was precipitated with 2 volumes of cold ethanol from 5,000 g supernatant arising from a 1:10 (w/v) homogenate in 10% TCA. The precipitate was then dissolved in water, and glycogen content was estimated according to Dubois et al. (1956). Results are expressed in terms of glucose equivalents from glycogen/g liver.

Plasma insulin Normal insulin was measured by radioimmunoassay by the method of Herbert et al. (1965). Pork monoiodine 125I-insulin was used. Guinea pig antiporcine insulin antiserum was sufficiently nonspecific as to allow pork-labeled insulin to be displaced by rat insulin. This antiserum showed reactivity with normal insulin, but not with proinsulin or glycated insulin.

Perfusion of the Langerhans islets and insulin secretion Islets were obtained from collagenase-treated rat pancreas according to the method of Lacy and Kostianovsky (1967). The perfusion of islets was carried out according to Burr et al. (1969) with slight modifications (Basabe et al. 1986). Islets from the whole pancreas of a single rat were used in each perfusion. Samples were collected following an initial 15-min recuperation period in tubes containing 0.2 ml 0.25 M EDTA. Then they were immediately frozen at)20C. Insulin was determined in the perfusion samples by the method of Herbert et al. (1965). To evaluate insulin secretion from perfused pancreatic islets, the area under the stimulated insulin secretion curves for pancreatic islets was integrated.

Tryptophan content in hepatic tissue Tryptophan content was determined as described by Wolf and Kuhn (1983) for tissue preparation and Moran and Fitzpatrick (1999) for HPLC detection; 12,000 g supernatant was filtered in Millipore SJHU013NS filters and injected onto a Waters lBondapack C18 HPLC column (300·4.6 mm i.d.). A Waters 470 fluorescence detector with an excitation wavelength of 290 nm and an emission wavelength of 330 nm was used.

Malondialdehyde content in hepatic tissue Malondialdehyde (MDA) was assayed as described by Ennamany et al. (1995); thus the red MDA-thiobarbituric acid complex produced was extracted with butanol to avoid interference by endogenous porphyrins.

Rat urinary chemiluminescence Total urine was collected employing metabolic cages. It was homogenized and kept frozen until evaluation. The total urine chemiluminescence was measured as described by Rı´ os de Molina et al. (1998) on aliquots of 1 ml of urine centrifuged at 300 g for 10 min. A Beckman LS-3150P liquid scintillation counter, operated in the out-of-coincidence mode, was used for quantification of chemiluminescence.

28 Proteins Proteins were determined according to the method of Lowry et al. (1951) using bovine serum albumin as standard.

Statistic analysis Results are expressed as means ±SEM. Data in the time course studies were submitted to one-way ANOVA and differences between means by the F-test. Statistical analysis of insulin secretion data was performed with ANOVA and Scheffe’s Test. In all cases, 0.05 was used as the level of significance.

Results URO-D activity and hepatic porphyrins content A long delay time in the manifestation of biochemical features is typical of HCB-induced porphyria. Figure 1 shows the effect of HCB treatment on the hepatic UROD activity expresed as copro’ gen III formation. As can be seen, HCB promotes URO-D decreases even from the 2nd week of HCB treatment (43% inhibition). Such decreases grow as the treatment time progresses, UROD losing dramatically its activity (90–95% of inhibition) by 5–6 weeks. Porphyrins (Fig. 1) began to accumulate in liver over control values from the week 3 (7.3±2.8 vs 3.2±1.7 lg porphyrins/g liver). Porphyrin levels became statistically significant by week 5 and reached remarkable increases at week 6 (more than 50-fold increase).

fasting period. Table 1 clearly shows that the rat strain used throughout these experiments exhibits a 40–60% increase in enzyme activity upon 24-h starvation. The specific activity of PEPCK exhibited a timedependent decrease as a consequence of HCB intoxication (Fig. 2). Results showed that the enzyme inhibition reached was about 55% at the 5th week of fungicide treatment, the time at which porphyrins accumulated remarkably in the liver (Fig. 1). HCB promoted an early decrease in specific activity of PC (Fig. 3A), which grew as the treatment progressed. A loss of 60% of this gluconeogenic enzymatic activity was observed at longer time periods. HCB showed early inhibition of G-6-Pase-specific activity (2nd week: 25% inhibition) (Fig. 3B). This effect remained constant until the 5th week. From then on, HCB provoked an even greater loss of this gluconeogenic activity (37%) when rats became porphyric. Specific PK activity was not significantly affected at any of the HCB treatment times assayed (control: 0.55±0.15 lmol NADH oxidized/min.mg protein; HCB-treated: 0.53±0.14 lmol NADH oxidized/min.mg protein).

Table 1 Response of gluconeogenic liver enzyme activities to fasting

Not fasted Fasted (24 h)

Gluconeogenic enzyme activities

PCa lmol/minÆmg

PEPCKa nmol/minÆmg

G-6-Pasea lmol/minÆmg

0.135±0.028 0.210±0.038*

8.01±0.46 11.12±0.84*

0.353±0.031 0.574±0.033*

*Significantly different from not fasted, P<0.05 Results are expressed as mean ±SEM of five animals in each group a

PEPCK, PC, and G-6-Pase enzyme activities were determined in the livers of rats with or without 24-h

Fig. 1 Time-course effect of hexachlorobenzene on hepatic uroporphyrinogen decarboxylase (URO-D) and porphyrin content. Each bar represents the mean ±SEM of ten animals. *Significantly different from control rats (P<0.05)

Fig. 2 Time-course effect of hexachlorobenzene on hepatic phosphoenol-pyruvate carboxykinase activity. Each bar represents the mean ±SEM of ten animals. The activity is expressed as lmol of NADH oxidized /min.mg protein (for the coupled reaction). *Significantly different from control rats (P<0.05)

29 Fig. 3A, B Time-course effect of hexachlorobenzene on hepatic pyruvate carboxylase activity (A) and hepatic glucose 6-phosphatase activity (B). Each bar represents the mean ±SEM of ten animals. A The activity is expressed as as lmol of NADH oxidized/min.mg protein (for the coupled system). B The activity is expressed as lmol of inorganic phosphate /min.mg protein. *Significantly different from control rats (P<0.05)

Fig. 4A, B Time-course effect of hexachlorobenzene on plasma glucose concentration (A) and hepatic glycogen content (B). Each point represents the mean ±SEM of ten animals. *Significantly different from control rats (P<0.05)

Plasma glucose and hepatic glycogen content The plasma glucose level showed a decrease as HCB treatment went on; it was statistically significant from the 3rd week and reached the lowest values (one-third loss) at the highest intoxication time (Fig. 4A). HCB significantly affected the glycogen hepatic content from the 2nd week of treatment, reaching values more than threefold higher than control in the 6th week (Fig. 4B). Plasma insulin levels and insulin secretion from perifused pancreatic islets Decay in the plasma insulin level was observed as HCB intoxication progressed (Fig. 5A). Therefore, HCB porphyric animals showed lower insulin levels than control animals (twice to four times lower). HCB does not seem to affect significantly the insulin secretion of perfused pancreatic Langerhans islets stimulated by glucose during the 3rd and 6th weeks of treatment. Perfused islets showed a similar biphasic

pattern of insulin secretion in control and treated animals during weeks 3 and 6, as can be seen in Fig. 5B, where the pattern for 6 weeks of treatment was shown. In spite of the fact that it did not reach statistical significance, data for HCB treated animals remained, in general, slightly lower than control. Lipid peroxidation and urinary chemiluminescence As can be seen in Table 2, lipid peroxidation, measured as MDA content, was augmented as a result of fungicide treatment at all times assayed both in homogenate and microsomes. These levels increased from the 2nd week, retained high values in the intermediate weeks, and showed a sharp rise at the end of HCB intoxication. Measurements of daily urinary chemiluminescence (Fig. 6) showed that this parameter grows as a function of HCB treatment time in treated animals. All the increases were statistically significant from the 2nd week where the values of the treated animals overcame twofold the control ones, reaching increases higher than fourfold at the 6th week.

30

Fig. 5A, B Changes in plasmatic insulin levels vs hexachlorobenzene treatment time (A) and effect of HCB on pancreatic insulin secretion curve from rat islets of Langerhans perifused and stimulated with glucose (B) . A Each point represents the mean ±SEM of ten animals after the different times of hexachlorobenzene treatment. B Each point represents the mean of insulin determinations ±SEM of ten animals at each time point after the glucose stimulus. Samples from minutes 1 and 2 were used for baseline determination. A stimulus of 16.5 mM glucose was added to the perifusion buffer from minute 3 to 40. The first secretory peak was integrated between minutes 3 and 7, and the second peak between minutes 8 and 40 of perifusion. This experiment was carried out with animals treated during 6 weeks with HCB. *Significantly different from control rats (P<0.05)

Table 2 Time-course of hexachlorobenzene (HCB) effect on malondialdehyde (MDA) content Treatment MDA in homogenate MDA in microsomal fraction time (weeks) (nmol MDA/g liver)a (nmol MDA/mg protein)a Control 2 3 5 6

220.6±13.8* 249.4±15.3* 254.5±12.1* 278.8±12.5* 363.5±41.2*

22.4±3.4* 30.0±5.6* 35.5±2.6* 37.3±1.0* 111.9±24.6*

*Significantly different from control, p< 0.05 a Results are expressed as mean ± SEM of ten animals in each group

Hepatic tryptophan content Control rats exhibited a tryptophan content in the liver of: 7.42±0.61 lg/g liver. Rats treated with HCB during weeks 2, 3, and 5 showed this indolic compound in the liver (6.46±0.43 lg/g liver), similar to the control values. Longer HCB treatment (6–8 weeks) promoted a moderate and statistically significant increase in the tryptophan levels (10.55±0.37 lg/g liver), which overcame the control values in about 20%.

Discussion As the present results show, 24-h fasting strongly induces gluconeogenic enzyme activities in the female Wistar rats strain, as reported in another rat strain (Weber et al. 1991). This study shows that gluconeogenesis was remarkably reduced in this experimental PCT porphyria. In fact, the three enzymes studied in this metabolic pathway decreased their hepatic activities. PEPCK and PC, the regulatory enzymes of this pathway, were the most affected in a time-dependent way as a consequence of HCB treatment. The decrease in PEPCK, here observed, at the 6th week, doubled that found at 3rd weeks. The half-life of HCB has been estimated as 57 days after a cumulative dose of 840 mg/kg body wt. and 150 days after a

Fig. 6 Time-course effect of hexachlorobenzene on total daily urine chemiluminescence. Each bar represents the mean ±SEM of ten animals. *Significantly different from control rats (P<0.05)

cumulative dose of 5,250 mg/kg body wt. (KuiperGoodman et al.1977; Koss et al. 1978). Administration of HCB (5·daily) repeated each week (as in the study reported here) represents cumulative HCB doses of 900 mg/kg body wt. at 3 weeks and 1,800 mg/kg body wt. at 6 weeks; (Billi de Catabbi et al. 2000); this will lead to a linear accumulation. Therefore, measurements

31

taken at 6 weeks represent doubling of the body burden compared to 3 weeks and are far away from the steady state (Rozman et al. 1977). Times would then represent doses. It is worth mentioning that this decline in PEPCK promoted by HCB at 6 weeks covers the full-dose response for inhibition of PEPCK by TCDD (Fan and Rozman 1995). The present results clearly demonstrate that HCBinduced reduction of gluconeogenesis was not related to a preponderance of glycolysis, since the activity of the key and regulatory glycolytic enzyme PK was not affected in intoxicated animals. It is interesting that G-6-Pase is the only membranebound enzyme of the gluconeogenic pathway whose activity does not depend on allosteric or hormonal regulation (Hers and Hue 1983). Furthermore, the data from spin label studies suggest that the HCB molecule is incorporated between the fatty-acid chains of membrane lipids (Ko´szo´ et al. 1982). Thus, it is possible that HCB, by modifying the lipid environment of G-6-Pase, affects and reduces its activity. Moreover, the amount of HCB that the membrane is able to uptake was rapidly completed (perhaps at the 2nd week). This would explain why the G-6-Pase activity is reduced early, and this reduction remains constant during HCB treatment. Our results show that both the rate of gluconeogenesis and plasma insulin levels are low in HCB treatment. These results disagree with the known regulatory effect of this hormone on gluconeogenesis. High insulin levels inhibit gluconeogenesis (Granner and Andreone 1985). However, the similar response of these parameters to TCDD, a porphyrinogenic drug mechanistically related to HCB, has been reported (Gorski and Rozman 1987). To solve this apparent paradox, Weber et al. (1991) proposed that changes in gluconeogenic enzyme activities precede hormonal changes. This would suggest that HCB-induced changes in hormonal homeostasis, at least with respect to insulin, are a compensatory response of the organism to stimulate gluconeogenesis, thus agreeing with the proposal for TCDD (Weber et al. 1991). The fact that the normal plasma insulin level was significantly lower in HCB-treated animals and the pancreatic insulin secretion pattern was similar to controls would suggest that there is no primary alteration in the insulinsecretion mechanism. Therefore, the insulin levels are adaptive and follow the variations in glucose levels that, as shown here, are depressed by HCB treatment. This decline should probably occur through an increased hormone degradation in the liver, rather than through the formation of the glycated insulin. In fact, hyperglycemia has been reported to be responsible for insulin glycation (Abdel-Wahab et al. 1996), but this was not the case in the present study where a time-course decrease in glucose levels was observed. The increases in lipid peroxidation and urinary chemiluminescence observed mirror oxidative stress and free radical production caused by the slow metabolization of HCB (Ferioli et al. 1984). These results agree

with previous reports from our laboratory (Billi de Catabbi et al. 1997; Rı´ os de Molina et al. 1998). Considering that the decrease in PEPCK and the generation of free radicals caused by the porphyrinogenic fungicide (HCB) occur in a more or less synchronized way and that they become evident in the second week, it could be that PEPCK reduction is due to the action of these free radicals. This effect may be exerted at the DNA level, affecting the expression of the PEPCK gene, and thus reducing the mRNA levels of PEPCK (as has been reported in the case of TCDD; Stahl et al. 1993) or acting directly on the enzymatic protein, as has been reported for other proteins (Wolff and Dean 1986). These factors should be taken into account: 1. The numerous analogies between the action of TCDD and dioxin-type-HCB. a. Decrease in PEPCK and PC activities and plasma insulin levels (Weber et al. 1991; this work). b. Increase in glucocorticoids, as reported by Vaena et al. (1998) for HCB, and Stahl et al. (1993) for dioxin. c. Lack of PEPCK response to physiological stimuli (glucocorticoids, insulin), whose alterations should increase enzyme levels instead of decreasing them (Stahl et al. 1993; this work). 2. That TCDD would reduce PEPCK gene expression through the glycocorticoid receptor system and protein phosphorylation (Stahl et al. 1993). 3. That oxidative stress plays an important role in cell regulation, modulating through free radicals kinase and phosphatase activities, which act on transcription factors (Guyton et al. 1996). Moreover, nuclear factor NF-1, involved in PEPCK transcription (Hanson 1997), seems particularly sensitive to oxidative stress (Morel and Barouki 1998). 4. That the oxidative stress produced by HCB may modify DNA and typically produces 8-hydroxy-deoxyguanosine (Horva´th et al. 2001), which provokes deleterious effects (Cheng et al. 1992), affecting the transcription process. It could be postulated that free radicals produced by the oxidative stress from HCB treatment might reduce PEPCK, perhaps altering its genetic expression, either directly on the DNA, or indirectly on transcription factors through kinasedependent phosphorylations. Since it has already been demonstrated that tryptophan administration resulted in hypoglycemia and that tryptophan and its metabolites alter hepatic gluconeogenesis (Smith and Pogson 1977), another possible agent responsible for gluconeogenic enzyme decrease might be tryptophan. However, the kinetics of the activity alteration of the gluconeogenic enzymes (PEPCK, PC and G-6-Pase) and tryptophan do not match. The increase of tryptophan observed only in the last weeks of HCB treatment might be due to the marked deregulation in the heme metabolic pathway observed at these times as a

32

consequence of a drastic and sustained decrease in URO-D. This fact would cause a decrease in heme levels. This decrease would produce a reduction in tryptophan pyrrolase since heme is its prosthetic group. Since tryptophan pyrrolase is the rate-limiting step that regulates the overall physiologic availability of tryptophan, its decline would lead to an increase of tryptophan. In this respect, Weber et al. (1992) found a sharp decrease in tryptophan pyrrolase in the liver of TCDD-treated rats with a concomitant increase of plasma tryptophan levels and a decrease of PEPCK activity. Therefore, if tryptophan contributes to the decrease of PEPCK, it would somehow be involved in the higher phases of the fungicide intoxication. Thus, tryptophan would add its effect to the action exerted by the free radicals already mentioned. As the present results show, the reduced levels of glucose would be the result of a decreased gluconeogenesis, together with an enhanced glucose storage, without affecting glycolysis. This derangement of carbohydrates leads porphyric hepatocytes to have a lower level of free glucose. Since glucose represses ALA-S (Tschudy et al. 1964), low glucose would lead to depression of this enzyme, a fact that contributes to the induction of this regulatory enzyme promoted by HCB (Wainstok de Calmanovici et al. 1984). This condition of depressed endogenous glucose also helps us to understand why high doses of exogenous glucose (in diets) suppress the porphyrinogenic effect of HCB (Ivanov et al. 1976). These results contribute to our understanding of the protective and modulatory effect that diets rich in carbohydrates have in hepatic porphyria disease (Kappas et al. 1995). It is known that HCB and TCDD (both polychlorinated xenobiotics) affect heme pathway. The present results add to this common feature another metabolic characteristic, since both chemicals affect the carbohydrate pathway and its regulation in a similar way. In summary, the present study provides evidence of gluconeogenesis impairment in experimental PCT modeled by HCB. This impairment adds another negative biochemical feature to the multiple coexisting events that control this pathology. Acknowledgements This work was supported by grants from the University of Buenos Aires and CONICET. L.C. San Martı´ n de Viale and J.C. Basabe are Scientific Research Career members of the CONICET (Consejo Nacional de Investigaciones Cientı´ ficas y Te´cnicas). Procedures involving animals (care and use) were conducted according to international guidelines (Guide for Care and Use of Laboratory Animals, National Research Council, USA, 1996, and the Council of the European Communities Directive, 86/ 609/ECC).

References Abdel-Wahab YHA, O’Harte FPM, Barnett CR, Flatt PR (1996) Glycation of insulin in the islet of Langerhans of normal and diabetic animals. Diabetes 45:1489–1496

Baginski ES, Foa PP, Zak B (1974) Glucose-6-phosphatase. In: Bermeyer HU (ed) Method of enzymatic analysis, 2nd edn, vol 1, Verlag Chemie/Academic Press, Weinheim, New York, pp 876–880 Basabe JC, Karabatas LM, Arata M, Pivetta OH, Cresto JC (1986) Secretion and effect of somatostatin in early stages of the diabetic syndrome in C57BL/KsJ-mdb mice. Diabetologia. 29:485– 488 Berndt J, Messner B, Still, J (1978) Optical assay of pyruvate carboxylase in crude liver homogenate. Anal Biochem 86:154–158 Billi de Catabbi S, Sterin-Speziale N, Fernandez MC, Minutolo C, Aldonatti C, San Martı´ n de Viale L (1997) Time course hexachlorobenzene-induced alterations of lipid metabolism and their relation to porphyria. Int J Biochem Cell Biol 29:335–344 Billi de Catabbi SC, Aldonatti C, San Martı´ n de Viale LC (2000) Heme metabolism after discontinued hexachlorobenzene administration in rats: possible irreversible changes and biomarker for hexachlorobenzene persistence. Comp Biochem Physiol Part C 127:165–175 Bo¨ger A, Koss G, Koransky W, Naummann R, Frenzel H (1979) Rat liver alteration after chronic treatment with hexachlorobenzene. Virchows Arch 382:127–137 Burr IM, Balant L, Stauffacher W, Renold AE, Grodsky G (1969) Dynamic aspects of proinsulin release from perifused rat pancreas. Lancet 2:882–883 Cheng KC, Cahill DS, Kasai H, Nishimura S, Loeb LA (1992) 8-hydroxy-guanine, an abundant form of oxidative DNA damage caused G fi T and A fi C substitutions. J Biol Chem 267:166–172 Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956) Colorimetric method for determination of sugar and related compound. Anal Chem 28:350–356 Elder GH (1998) Porphyria cutanea tarda. Semin Liver Dis 18:67– 75 Ennamany R, Marzetto S, Saboreaud, D, Creppy EE (1995) Lipid peroxidation induced by bolesatine, a toxin of Boletus Satanas: implications in m5dC variation in Vero cells related to inhibition of cell growth. Cell. Biol. Toxicol. 11:347–354 Fan F, Rozman KK (1994) Relationship between acute toxicity of 2,3,7,8,-tetrachlorodibenzo-p-dioxin (TCDD) and disturbance of intermediary metabolism in the Long-Evans rat. Arch Toxicol 69:73–78 Fan F, Rozman KK (1995) Short- and long-term biochemical effects of 2,3,7,8-tetrachlorodibenzo-p- dioxin in female LongEvans rats. Toxicol Lett 75:209–216 Ferioli A, Harvey C, De Matteis F (1984) Drug induced accumulation of uroporphyrin in chicken hepatocytes cultures. Biochem J 224:769–777. Gorski JR, Rozman K (1987) Dose-response and time course of hypothyroxinemia and hypoinsulinemia and characterization of insulin hypersensitivity in 2,3,7,8-tetrachlorobenzo-p-dioxin (TCDD) treated rats. Toxicology 44:297–307 Granner DK, Andreone, TL (1985) Insulin modulation of gene expression. Diabet Metab Rev 1:139–170 Guyton KZ, Lui Y, Gorospe M, Xu Q, Holbrook NJ (1996) Activation of mitogen- activated protein kinase by H2O2. Role in cell survival following oxidant injury. J Biol Chem 271:4138– 4142 Hanson RW (1997) Regulation of phosphoenolpyruvate carboxykinase (GTP) gene expression .Annu Rev Biochem 66:581–611 Herbert V, Lau K, Gottlieb CW (1965) Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375–1384 Hers HG, Hue L (1983) Gluconeogenesis and related aspects of glycolysis. Annu Rev Biochem 52:617–653 Horva´th ME, Faux SP, Smith AG, Blazovics A, van der Looij, Fehe´r J, Cheeseman KH (2001) Vitamin E protects against iron-hexachlorobenzene induced porphyria and formation of 8-hydroxydeoxyguanosine in liver of C57BL/10 Sc5n mice. Toxicol Lett 122:97–102 Imamura K, Tanaka T (1982) Pyruvate kinase isozymes from rat. In: Wood WA (ed) Methods in enzymology, vol 90: Carbohydrate metabolism, part E. Academic Press, New York, pp 150–165

33 Ivanov DE, Chernev K, Adjarov D, Krustev L, Georgiev G (1976) The effect of different diets on porphyrin metabolism and microsomal liver enzyme induction in experimental hexachlorobenzene porphyria. In: Porphyrins in human diseases. 1st Int. Porphyrin Meet. Freiburg/Br 1975. Karger, Basle, pp 438– 444 Kappas A, Sassa S, Galbraith RA, Nordmann Y (1995) The porphyrias. In: Scriver CR, Beaudet AL, Sly WS, Valle D, Stanbury JB, Wibgaarden JB, Fredickson DS (eds) Metabolic and molecular basis of inherited disease. McGraw-Hill, New York, pp 2103–2159 Koss G, Koransky W (1975). Studies on the toxicology of hexachlorobenzene. I. Pharmacokinetics. Arch Toxicol 34:203–212 Koss G, Seubert S, Seubert A, Koransky W, Ippen H (1978). Studies on the toxicology of hexachlorobenzene. III. Observation in a long term experiment. Arch Toxicol 40:285–294 Ko´szo´ F, Horvath L, Simon N, Siklosi C, Kiss, M (1982) The role of possible membrane damage in porphyria cutanea tarda: a spin label study of rat liver cell membranes. Biochem Pharmacol 31:11–17 Kuiper-Goodman T, Grant DL, Moodie CA, Korsrud GO, Munro IC (1977) Subacute toxicity of hexachlorobenzene in the rat. Toxicol Appl Pharmacol 40:529–549 Lacy PE, Kostianovsky M (1967) Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes 16:35–39 Lim CK, Rideout JM, Wright DJ (1983) Separation of porphyrins isomers by high-performance liquid chromatography. Biochem J 211:435–438 Lowry OH, Rosebrough NJ, Randall RJ (1951) Protein measurement with the folin phenol reagement. J Biol Chem 193:265–275 Mollenhauer HH, Jhonson JH, Younger RL, Clark DE (1975) Ultraestructural changes in the liver fed rat hexachlorobenzene. Am J Vet Res 37:1777–1781 Moran GR, Fitzpatrick PF (1999) A continuous fluorescence assay for trypthophan hydrolase. Anal Biochem 266:148–152 Morel Y, Barouki R (1998) Down-regulation of cytochrome P450 1A1 gene promoter by oxidative stress. Critical contribution of nuclear factor 1. J Biol Chem 273:26969–26976 Petrescu I, Bojan O, Saied M, Barzu O, Schmidt F, Ku¨hnle HF (1979) Determination of phosphoenolpyruvate carboxykinase activity with deoxyguanosine 5’-diphosphate as nucleotide substrate. Anal Biochem 96:279–281 Pilkis SJ, Fox E, Wolfe L, Rothbarth L, Colosia A, Stewart HB, El-Maghrabi MR (1986) Hormonal modulation of key hepatic regulatory enzymes in the gluconeogenic/glycolytic pathway. Ann NY Acad Sci 478:1-19 Rı´ os de Molina MC, Iglesias S, Laurı´ a L, San Martı´ n de Viale LC (1996) Precancerous pathology evoked by hexachlorobenzene treatment. Acta Physiol Pharmacol Latinoam 46:71–87 Rı´ os de Molina MC, Suarez Lissi MA, Armesto A, Lafourcade C, Lissi E (1998) Rat urinary chemoluminiscence. Biolumin Chemilumin 13:63–68 Rose JA, Hellman ES, Tschudy DP (1961) Effect on diet on induction of experimental porphyria. Metabolism 10:514–521

Rozman K, Mueller W, Coulson, F, Korte F (1977) Long-term feeding study of hexachlorobenzene in Rhesus monkeys. Chemosphere 3:81–84 San Martı´ n de Viale LC, Tomio JM, Ferramola AM, Sancovich AH, Tigier HA (1975) Experimental porphyria induced in rats by hexachlorobenzene. Studies on enzyme associated with heme pathway. Effect of 17-b-estradiol. In: Doss M (ed) First Int Porphyrin Meet: Porphyrins in human diseases. Karger, Basle, pp 453–458 San Martı´ n de Viale LC, Rı´ os de Molina MC, Wainstok de Calmanovici R, Tomio, JM (1977) Porphyrins and porphyrinogen carboxylyase in hexachlorobenzene induced porphyria. Biochem J 168:393–400 Smith AG, De Matteis F (1980) Drugs and hepatic porphyrias. Clin Haematol 9:399 )425 Smith SA, Pogson CI (1977) Tryptophan and the control of plasma glucose concentrations in the rat. Biochem J 168:495–506 Stahl BH, Beer DG, Weber LWD, Rozman K (1993) Reduction of hepatic phosphoenolpyruvate carboxykinase (PEPCK) activity by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is due to decreased mRNA levels. Toxicology 79:81–95 Tschudy DP, Welland F, Collins A, Hunter G (1964) The effect of carbohydrate feeding of d-aminolevulinic acid synthetase. Metabolism 13:396–406 Vaena de Avalos S, Martini CN, San Martin de Viale LC (1998) Effect of hexachlorobenzene on steroids synthesis in adrenal cortex of rats. Proceedings of the XXXIV Annual Meeting of the Argentinian Society of Biochemistry and Molecular Biology (Mendoza) M 175:34 Wainstok de Calmanovici R, Rios de Molina MC, Tomio JM, San Martı´ n de Viale, LC (1984) Mechanism of hexachlorobenzene induced porphyria in rats. effect of phenobarbitone pretreatment. Biochem J 218: 753–763 Wainstok de Calmanovici R, Cocho´n A, Aldonatti C, Cabral JR, San Martı´ n de Viale LC (1991) Sex comparasion of heme pathway in rays bearing hepatic tumors. Tumori 77:379–384 Weber LWD, Lebofsky M, Greim H, Rozman K (1991) Key enzymes of gluconeogenesis are dose-dependently reduced in 2,3,7,8-tetrachlodibenzo-p-dioxin (TCDD)-treated rats. Arch Toxicol 65:119–123 Weber LWD, Stahl BU, Rozman K (1992) Are serotonergic mechanisms involved in the acute toxicity of chlorinated dibenzo-p-dioxins (CDDs). Chemosphere 25:161–164 Wilson J, Kueveruwa SS (1994) Toxicological profile for hexachlorobenzene. Update. US Department of Health and Human Services. Public Health Service Agency for Toxic Substances Disease Registry Wolf WA, Kuhn DM (1983) Simultaneous determination of 5-H TRP, its amino acid precursors and acid metabolite in discrete brain regions by HPLC with fluorescence detection. J Chromatog 275:1-9 Wolff SP, Dean RT (1986) Fragmentation of proteins by free radicals and its effect on their susceptibility to enzyme hydrolysis. Biochem J 234:299–403