The use of dogs as second species in regulatory testing of pesticides. Part II: Subacute, subchronic and chronic studies in the dog.

Data on 172 pesticides (fungicides, herbicides, insecticides and other pesticides) submitted for regulatory purposes during the past 40 years to the German Federal Institute for the Health Protection of Consumers and Veterinary Medicine (BgVV) were analysed to determine whether chronic studies in dogs (52/104 weeks) provide essential additional specific toxicological compared with subchronic (13 weeks) or subacute (4 weeks) studies in the same species. Comparison of the lowest observed effect levels (LOELs) in dogs revealed no significant differences between subchronic and chronic studies but a significant difference between subacute studies and subchronic or chronic studies. Moreover, there was a significant correlation between the LOELs determined in subchronic studies and those determined in chronic studies in dogs (r = 0.78-0.84). The distribution of target organ toxicity determined in chronic studies in dogs was not significantly different from the distribution determined in subchronic studies, except for effects on the spleen in studies on herbicides which were only observed in chronic studies and in combined subchronic/chronic studies, but never in subchronic studies. Organ-specific effects that were observed in chronic studies but not in subchronic studies were found in 30 of 55 studies on fungicides, in 25 of 44 on herbicides, in 17 of 38 on insecticides and in 10 of 16 on other pesticides. Compared with 26-week studies, additional organ-specific toxic effects were found in three of five, in three of four, in one of three and in one of one 52/104-week studies on fungicides, of herbicides, of insecticides and other pesticides, respectively. The organ-specific effects that were seen only in the chronic dog studies were evaluated according to their severity, e.g. significant damage to organs versus changes in enzyme activities that do not affect organ function or histology. Such effects were not considered to be specific for dogs in chronic studies if similar effects were also found in chronic studies in rodents (rat or mouse). In 15 of 141 studies in dogs serious side effects were observed in chronic studies that were not observed in subchronic studies. Furthermore, for 9 of 172 pesticides significant new effects were seen in 52/104-week studies when compared with 4- or 13-week studies and in 7 of 141 52/104-week studies when compared with 13-week studies. Analysis of the severity of organ-specific toxic effects of pesticides revealed that chronic long-term studies (52/104 weeks) in dogs do not provide specific additional information to 26-week studies in the same species.

The use of dogs as second species in regulatory testing of pesticides. I. Interspecies comparison.

The relevance of studies in dogs on regulatory testing of pesticides was examined retrospectively using data of 216 pesticides (acaricides, fungicides, growth regulators and hormones, herbicides, insecticides, molluscicides, nematicides, rodenticides, synergists for insecticides) submitted for regulatory purposes during the past 40 years to the Federal Institute of Health Protection of Consumers and Veterinary Medicine (BgVV), the competent national authority in Germany. At first the relevance of the no-observed-effect levels (NOEL) for safety assessment was evaluated for each chemical in 4-week (subacute), 13-week (subchronic) and 52/104-week (chronic) toxicity studies carried out on dogs, rats and mice. After subchronic and chronic application of fungicides the sensitivity of rats and dogs to the toxic chemicals was quite similar. However, the dog was generally a more sensitive species to toxic effects of insecticides than rat and mouse. On the other hand the NOEL was lower in the rat than the dog in chronic studies on herbicides. When the lowest-observed-effect level (LOEL) was evaluated in animal species, the dog was the most sensitive in approximately 15% of the studies. Mice were found to be the most sensitive species only in approximately 1% of the studies on 216 pesticides. Comparison of organ specific toxicity at the LOEL in subacute studies on fungicides and herbicides revealed a poor correlation of target-specific organ toxicity across species. However, in the subchronic and chronic studies (13 and 52/104 weeks) no significant differences in species-specific organ toxicity were observed in the three species rat, mouse and dog. The only exception were haematoxic effects in chronic studies on herbicides, which were more frequent in dogs (40%) than in rats and mice (20%). The results support the established concept that studies on dogs and rats are important for the safety assessment of pesticides, while studies on mice do not provide further information, except for detection of an oncogenic potential which is a further controversial issue. Further analysis of subacute, subchronic and chronic studies in dogs should reveal if all of the studies are essential for safety assessment of pesticides.

Effect of S-adenosyl-L-methionine on the development of preneoplastic foci and the activity of some carbohydrate metabolizing enzymes in the liver, during experimental hepatocarcinogenesis.

gamma-Glutamyltranspeptidase (GGT)-positive foci and glutathione-S-transferase, placental (GST-P)-positive lesions occupied 36% and 54% of liver parenchyma, respectively, in Wistar rats 8 weeks after initiation with diethylnitrosamine, followed by selection. The administration of S-adenosyl-L-methionine (SAM, 384 mumol/kg/day) caused 77% and 42% falls in the percentage of GGT-positive and GST-P-positive lesions, respectively. There also occurred a 46% decrease in labeling index of GGT-positive foci, in SAM-treated rats. These changes were associated with decrease in liver pyruvate kinase (PK), lactate dehydrogenase and glycerol-3-phosphate dehydrogenase. SAM did not affect these enzymatic activities in normal and uninitiated controls, but it caused a consistent increase in initiated rats. Enolase, fructose-biphosphatase and malic enzyme (ME) activities increased in the liver of initiated rats. SAM did not modify significantly these enzymatic activities, either in control or in initiated rats. Glucose-6-phosphate dehydrogenase (G6PDH) was 113% higher in the liver of initiated rats than in uninitiated controls. SAM treatment did not significantly affect this enzymatic activity in uninitiated rats, but caused a great decrease in initiated ones. As expected, there occurred a marked rise in GGT activity in the liver of initiated rats, with respect to controls. SAM caused an increase in GGT activity in normal and uninitiated controls, but it caused a 77% fall in GGT activity in initiated rats, coupled with a 380% rise in remodeling of GGT-positive lesions. Histochemical determination of G6PDH and ME activities showed that in the absence of SAM many preneoplastic lesions expressed higher G6PDH and ME activities than surrounding liver. SAM did not affect ME-positive lesions, while it caused a decrease in the number of G6PDH-positive lesions. Immunohistochemical determination of PK activity, isoenzyme L, showed a decrease in GST-P-positive lesions. Many of these lesions were no longer recognizable as lesions expressing a low PK activity, in SAM-treated rats. However, a relatively small number of GST-P-positive lesions expressing a low PK activity were still present in these rats. These data suggest that glucose channelled into triacylglycerol and pyruvate synthesis decreases in rat liver, during the development of preneoplastic foci, while the production of reducing equivalents and pentose phosphates increases, thus favoring DNA synthesis and detoxification reactions. Decrease in DNA synthesis, in SAM-treated rats, is paralleled by a partial reversion of carbohydrate metabolic features to those present in normal liver.

Hormone-sensitive carbohydrate metabolism in rat hepatocyte-hepatoma hybrid cells.

Hepatocyte-hepatoma hybrid cells were obtained by fusion of hepatocytes from adult rats and Fao hepatoma cells in the presence of polyethylene glycol. These hybrids were called hepatocytoma cells. The preservation of liver-specific enzyme activities and metabolic functions was studied in the hybrid clone 1E3. 1) The proliferating hepatocytoma cells formed monolayers presenting morphological similarity to primary cultures of hepatocytes. 2) In contrast to Fao hepatoma cells, activities of all gluconeogenic key enzymes were preserved at normal or reduced levels. 3) Lactate-dependent glucose formation was maintained at a state reduced to 36% of the gluconeogenesis in hepatocytes; no glucose formation was detected in Fao hepatoma cells. 4) The activity of the liver-specific glucokinase was reduced in hepatocytoma cells, but it was still present in contrast to Fao cells. The liver-specific isoenzyme pyruvate kinase type L was replaced by the isoenzyme type M2. 5) Gluconeogenic and glycolytic enzyme activities were regulated in hepatocytoma cells by glucagon (0.1 microM) and by insulin (0.1 microM). 6) The genome of hepatocytoma cells and its expression were stable for at least one year, when spontaneously dedifferentiating cells were removed by recloning in hypoxanthine-aminopterine-thymidine (HAT) medium.

Di(2-ethylhexyl)phthalate alters carbohydrate enzyme activities and foci incidence in rat liver.

The effect of di(2-ethylhexyl)phthalate (DEHP) on diethylnitrosamine (DEN)-initiated preneoplastic liver lesions with expression of gamma-glutamyltranspeptidase (GGTase) and loss of adenosine triphosphatase (ATPase) as well as alterations of hepatic carbohydrate metabolism in male and female Sprague-Dawley rats have been investigated. Two treatment schedules have been compared with respect to their sensitivity by the histochemical demonstration of preneoplastic islands and by the biochemical determination of alterations in enzyme activities of liver homogenates and of serum, the last indicating hepatotoxicity. For initiation, a single dose of DEN was given, followed by treatment with various doses of DEHP given three times weekly by gavage for 7 or 11 consecutive weeks. As histochemical enzyme markers, the expression of positive GGTase as well as the deficiency in ATPase were used for identification of liver foci. The weanling female rats (protocol A) were found to be more sensitive to the carcinogenic effect of DEN in view of foci incidence than the mature male rats which underwent partial hepatectomy prior to DEN application. The administration of 200 mg DEHP/kg body wt increased the incidence of ATPase-deficient foci in both male and female rats; however, concentrations of 1000 and 2000 mg DEHP/kg decreased the incidence of liver foci. The number of foci with expression of GGTase was only slightly increased in female rats following a DEHP concentration of 50 mg/kg, and 200 mg/kg body wt. DEHP alone did not induce preneoplastic lesions that could be identified by these two markers. Biochemical investigations indicate that DEHP alters the metabolic pattern in liver. An increase of the NADP-linked enzymes glucose-6-phosphate dehydrogenase (G6PDH), malic enzyme, extra-mitochondrial ICDH as well as an enhancement of NAD-dependent alpha-G3PDH and lactate dehydrogenase were found following DEHP administration. On the other hand the glycolytic enzymes pyruvate kinase (PK) and enolase as well as the gluconeogenetic enzyme fructose-1,6-bisphosphatase (FBPase) were significantly reduced. In protocol B (male rats) the reactions of PK, FBPase and malic enzyme were more altered after DEHP exposure than in protocol A, while the activity of G6PDH was more increased in protocol A. Most enzymes being involved in the carbohydrate metabolism are influenced by DEHP in a dose-dependent manner. There was no increase in serum FBPase activity in both male and female rats after DEHP treatment but a reduction of glutamate-oxalate-transaminase and glutamate-pyruvate-transaminase activities was observed.(ABSTRACT TRUNCATED AT 400 WORDS)

Tumor promotion by the peroxisome proliferator nafenopin involving a specific subtype of altered foci in rat liver.

The involvement of tumor promotion in the hepatocarcinogenic action of peroxisome proliferators has not been generally accepted. We studied the effect of nafenopin (NAF) as a model compound in a two-stage initiation-promotion protocol. Carcinogenesis was initiated by a single dose of aflatoxin B1 (AFB1) in female (AFB1, 5 mg/kg) and male (AFB1, 2 mg/kg) Wistar rats. After recovery NAF was fed via the diet, providing a daily dose of 100 mg/kg body weight. Phenobarbital (PB) (50 mg/kg body weight) was fed to female rats as a positive control. The following results were obtained. (a) At weeks 40, 55, 59, and 70, significantly more and larger liver tumors were present in AFB1-NAF-treated rats than in rats receiving either compound alone, and the effect of the combined treatment was clearly more than additive, in three independent experiments including both sexes. This suggests tumor promotion by NAF. Male rats responded more strongly than females. Similarly, PB enhanced the yield of liver tumors. Histologically, tumors were hepatocellular adenoma or carcinoma. In group AFB1-PB the majority consisted of eosinophilic and glycogenstoring cells. However, adenoma and carcinoma of groups AFB1-NAF and O-NAF consisted of weakly basophilic cells. (b) Phenotypically altered foci were evaluated in hematoxylin- and eosin-stained liver sections from the female rats. NAF treatment after AFB1 had little effect on number and size of eosinophilic-clear cell foci and decreased the number of trigroid foci. However, it led to a dramatic increase (20-fold after 70 weeks of NAF treatment) in number and size of foci of a special phenotype that was extremely rare after AFB1 alone and virtually absent in group AFB1-PB. Hepatocytes in these foci are characterized by weak diffuse basophilia and some eosinophilia, similar to the phenotype in adenoma and carcinoma, and by absence of gamma-glutamyltranspeptidase (GGT) expression. Based on these findings, we propose the hypothesis that NAF promotes the development of liver tumors via a mechanism involving amplification of a specific subtype of altered hepatic foci.

Effects of hypolipidemic drugs nafenopin and clofibrate on phenotypic expression and cell death (apoptosis) in altered foci of rat liver.

Phenotypically altered liver foci were produced in female Wistar rats by a single dose of N-nitrosomorpholine followed by promotion with phenobarbital (PB) for 20 or 28 weeks. Then treatment was changed to either hexachlorocyclohexane (HCH), or cyproterone acetate (CPA), or nafenopin (Naf) or clofibrate (Clof), two hypolipidemic drugs. Foci were identified by a positive reaction for gamma-glutamyl-transpeptidase (GGT) and other cytological markers. HCH and CPA could substitute for PB as foci promoters; in contrast, Naf and Clof decreased expression of GGT in foci resulting in a decline of number and area of detectable foci, effects particularly pronounced with Naf. Immunohistochemical investigations of serial sections revealed that Naf also reduced expression of the altered phenotype when cytochrome P450-PB and pyruvate kinase (type L) were used as foci markers, but not when glutathione-S-transferase B (GST-B) was used. Thus, the number of foci with enhanced GST-B did not decline significantly after the change from PB to Naf treatment. Furthermore, the reduction of GGT and the decrease of foci number during Naf treatment were not associated with increased evidence of cell death by apoptosis in foci, in contrast to the situation after PB withdrawal. These findings strongly suggest that the disappearance of GGT-positive foci after Naf is due to a phenotypic change resulting in a suppression of GGT expression rather than to physical elimination of foci.

Comparison of three rat liver foci bioassays--incidence of preneoplastic foci initiated by diethylnitrosamine.

Three rat liver foci bioassays have been compared with respect to their sensitivity by the histochemical demonstration of preneoplastic foci, and by the biochemical determination of alterations in enzyme activities of serum indicating hepatotoxicity. We studied the initiation/promotion schedules according to Oesterle and Deml (A), and according to Pereira (B, Broad Spectrum Protocol), and the initiation/selection protocol according to Tatematsu et al. (C), with diethylnitrosamine (DEN), given as a single initiating dose of 10 and 30 mg/kg body wt respectively. With all schedules Sprague-Dawley rats, either females, 3 weeks old (A), or males, 6 weeks old (B, C) were used. For promotion polychlorinated biphenyls (A) or phenobarbital (B) were administered. Selection was performed with 2-acetylaminofluorene (C). The rats in schemes (B) and (C) underwent partial hepatectomy one day prior to initiation. The number and total area of foci deficient in adenosine-5'-triphosphatase (ATPase) and positive in gamma-glutamyltranspeptidase (GGTase) was evaluated. In the complete schedule with 30 mg of DEN in system (A) foci incidence exceeded that of the other systems by about 7-fold (ATPase) and 2-fold (GGTase) respectively. The lower dose of DEN and all control experiments resulted in a respective lower foci yield. With scheme (C), but not with schemes (A) and (B), e.g. serum fructose-1.6-bisphosphatase and alkaline phosphatase were increased, suggesting liver cell damage. Thus tested with DEN, scheme (A) is most sensitive and causes a low impairment of animals' welfare.

Effects of non-genotoxic hepatocarcinogens phenobarbital and nafenopin on phenotype and growth of different populations of altered foci in rat liver.

Non-genotoxic hepatocarcinogens share the ability to induce liver growth in rodents. Phenobarbital (PB), as one prototype compound, promotes the development of liver tumors; altered cell foci of the clear-eosinophilic phenotype, also identified by gamma-glutamyltransferase expression, appear to be precursor lesions. These foci seem to over-respond to the growth-inducing effect of PB. In contrast, the question as to whether peroxisome inducers are also tumor promoters is still unsettled. We will present evidence which strongly suggests that the peroxisome inducer, nafenopin (Naf), promotes tumor development in rat livers by stimulating selective growth of a hitherto undescribed subtype of altered foci. This subtype is characterized by weak diffuse cytoplasmic basophilia of its hepatocytes. Initiation in rats by aflatoxin B1 followed by promotion with Naf produced numerous adenomas and carcinomas; their morphology resembled that of the weakly basophilic foci. Both clear-eosinophilic and weakly basophilic foci appear "spontaneously" in the liver of aging rats. Promotion of such lesions by PB-type compounds or peroxisome inducers may explain cancer formation by these non-genotoxic agents.

Pyruvate kinase isoenzymes in altered foci and carcinoma of rat liver.

Pyruvate kinase (PK) isoenzymes, rate limiting for the last steps of glycolysis, were studied in normal rat liver, putative preneoplastic foci, neoplastic nodules and hepatocellular carcinoma. These lesions were produced by an initiation-promotion protocol: treatment with a single dose of N-nitrosomorpholine (NNM) was followed by feeding diets containing phenobarbital (PB) or alpha-hexachlorocyclohexane (alpha-HCH), or basal diet. PK was demonstrated (i) by immunocytochemistry on histological sections with antibodies specifically directed against the L and M2 isoenzymes, (ii) by electrophoretic separation of isoenzymes in homogenates from liver and larger tumors, and (iii) by electrophoretic separation of isoenzymes in parenchymal and stromal cells isolated from liver and tumors. Immunocytochemistry showed decreases of L-PK (L-PK-) in hepatocytes of most of the foci, nodules and carcinomas. Most L-PK- foci showed increases in gamma-glutamyltransferase (gamma-GT) and epoxide hydrolase (EH). PB or alpha-HCH treatment further decreased expression of L-PK in foci, but not in normal liver. Cells and foci with enhanced L-PK (L-PK+) were also found after carcinogen treatment. These did not show increases of gamma-GT or EH or any distinct morphological alterations with the exception of some which were basophilic ('tigroid') in H and E stained sections. No L-PK+ tumors were found. We could not demonstrate the M2-type PK in parenchymal cells of liver or any of the lesions described above. This isoenzyme was restricted to stromal cells in normal rat liver and in all stages of carcinogenesis as shown by immunohistology and by electrophoresis of preparations from isolated cell populations. However, stromal cells from hepatocellular carcinomas exhibited a 3-fold increase of M2-PK compared with stromal cells from normal liver. These results do not support an isoenzyme shift from L to M2-PK in the course of malignant transformation of hepatocytes as suggested previously.

Binding of cationized and native ferritin to cellular structures of the electric organ of Torpedo marmorata.

The distribution of polycationic and polyanionic binding sites in the electric organ of Torpedo marmorata was investigated by incubation of tissue with native (NF) ferritin. 1) Collagen fibrils from the electric organ carry rosettes of polyanionic sites on their surface with a periodicity of 60 nm, corresponding to the pattern of crossbanding in collagen fibrils. The CF-binding sites are abut 30 nm in size and project 20 nm beyond the surface of the fibril. 2) As revealed by incubation of tissue homogenates, CF heavily stains the intraperiod line of the axonal myelin and also tubular structures in the axonal cytoplasm. 3) Neither the extracellular aspects of the pre- nor the postsynaptic membrane became labeled with either NF or CF. After incubation of tissue homogenates. labeling of the electron-dense material of the cytoplasmic aspect of the postsynaptic membrane was observed with NF and, in particular, with CF. The ventral basal lamina of the electroplaque cell revealed uniform labeling with NF. In contrast, CF-binding sites were distributed in the lamina densa of the basal lamina as a lattice of discrete binding sites, approximately 45 nm in diameter. The presence of polyanionic sites in the basal lamina, which also proceeds through the synaptic cleft, suggests the existence of a diffusion barrier for the released neurotransmitter acetylcholine. It is proposed that this facilitates hydrolysis of acetylcholine in the synaptic cleft and recirculation of the products of hydrolysis to the axon terminal.