Carcinogenicity Study in Wistar Rats with Administration in Drinking Water

FUNDAMENTAL AND APPLIED TOXICOLOGY ARTICLE NO.

40, 75–89 (1997)

FA972374

Trimethylphosphate: A 30-Month Chronic Toxicity/Carcinogenicity Study in Wistar Rats with Administration in Drinking Water1 E. M. Bomhard,*,2 G. J. Krinke,† W. M. Rossberg,‡ and Th. Skripsky† *Department of Toxicology, Carcinogenicity, and Genotoxicity, Bayer AG, D-42096 Wuppertal, Germany; †Department of Toxicology, Novartis Crop Protection AG, CH-4002 Basel, Switzerland; and ‡Department of Drug and Safety Evaluation, Bayer Corporation, 400 Morgan Lane, West Haven, Connecticut 06516 Received April 3, 1997; accepted September 3, 1997

is 1 mg/kg in males and 10 mg/kg in females. The incidence, time of occurrence, spectrum of types, and localizations of tumors provided no indication of a tumorigenic/carcinogenic effect of the test substance. TMPO is therefore considered not to be carcinogenic in Wistar rats. q 1997 Society of Toxicology.

Trimethylphosphate: A 30-Month Chronic Toxicity/Carcinogenicity Study in Wistar Rats with Administration in Drinking Water. Bomhard, E. M., Krinke, G. J., Rossberg, W. M., and Skripsky, Th. (1997). Fundam. Appl. Toxicol. 40, 75–89. Trimethylphosphate (TMPO) was administered to 50 male and 50 female Wistar rats through their drinking water at doses of 0, 1, 10, or 100 mg/kg body weight up to 30 months. The dosage of 100 mg/kg was reduced to 50 mg/kg in week 54 for reasons of tolerance, and the animals were euthanized in week 100. Additional 10 animals per dose and sex were treated for 12 months and then euthanized for interim analysis. Weakness of the hind limbs, increased incidences of sunken flanks, distended abdomen, and poor general condition were observed in both sexes of the 100/ 50 mg/kg group beginning with week 46. Food intake was reduced in high dose males. At 10 mg/kg body weights were up to 10% (males) and at 100/50 mg/kg up to 20% (males) or 15% (females) lower than in controls. Mortality was not affected in animals receiving up to 10 mg/kg. At 100/50 mg/kg it was markedly increased, reaching about 70% at week 100. Relatively slight hematologic changes (reduced hemoglobin, hematocrit, erythrocyte counts, increased reticulocyte numbers, and thrombocyte counts as well as a shift in the differential blood count) at 100/50 mg/kg are interpreted as changes most probably secondary to the other toxic effects. Increased cholesterol concentrations in plasma, shifts in the serum protein electrophoresis (males), increased organ weights (females), and an increased incidence of necroses and lymphocytic infiltrations point to a treatment-related effect on the liver at 100/50 mg/kg. Slightly increased protein excretion, increased relative kidney weights, and an increased incidence of chronic progressive nephropathy are considered treatment-related but rather secondary effects at 100/50 mg/kg. At 100/50 mg/kg an increased incidence and severity of bilateral tubular atrophy in the testes was diagnosed. The most important toxic effect was neurotoxicity, consisting of degeneration and loss of nerve fibers in the peripheral nerves and the spinal cord, associated with myopathic changes, and occurring at 100/50 mg/kg. The no-observedadverse-effect-level, based on the suppression of body weight gain,

Trimethylphosphate (TMPO; CAS No. 512-56-1) is a potential contaminant in certain commercial phosphate esters; therefore, its toxicity is a matter of interest. Although it is not known to be a cholinesterase inhibitor, it has neurotoxic potential (Deichmann and Witherup, 1946; Schaeppi et al., 1984). TMPO is a direct acting mutagen with alkylating activity. However, its mutagenic/genotoxic potential is considered to be low in most test systems. In 1978 the U.S. Department of Health, Education and Welfare published the results of bioassays with trimethylphosphate in rats and mice. Groups of 50 Fischer 344 rats and 50 B6C3F1 mice of each sex were administered TMPO in distilled water three times per week at dose levels of either 50 or 100 mg/kg (rats) and either 250 or 500 mg/kg body weight (mice). Vehicle controls consisted of groups of 20 rats and 20 mice of each sex. The rats were dosed for 104 weeks and the mice for 103 weeks. The conclusion of these studies was that TMPO was associated with fibromas of the subcutaneous tissue in male Fischer 344 rats and was carcinogenic in female B6C3F1 mice, producing endometrial adenocarcinomas of the uterus (DHEW Publication (NIH) 78-1331, February 10, 1978). These results indicated possible tumorigenic effects of TMPO, but the studies had some shortcomings (e.g., small number of control animals, treatment only 3 days/week, etc.). In the present chronic toxicity/carcinogenicity study TMPO was administered to Wistar rats in their drinking water for up to 30 months. The objectives of the study were to identify any carcinogenic potential of TMPO as well as its organ toxicity and to establish the dose–response relationship and a no-observed-adverse-effect-level in a mammalian species.

1 Presented in part at the 35th Annual Meeting of the Society of Toxicology, Anaheim, CA, March 10–14, 1996. 2 To whom correspondence should be addressed. Fax: /49 (0)202-36 88 73. E-mail: [email protected].

MATERIAL AND METHODS Animals. Weanling Wistar rats of the strain BOR:WISW (SPF Cpb) from a specific pathogen-free colony, 4–5 weeks of age, were purchased 75

AID

FAAT 2374

/

6k23$$$$41

11-03-97 14:49:29

0272-0590/97 $25.00 Copyright q 1997 by the Society of Toxicology. All rights of reproduction in any form reserved.

ftoxa

76

BOMHARD ET AL.

from breeder Winkelmann (Borchen, Germany). Animals were acclimated for 9 days in the same room in which they were housed for the study. The mean body weight (body weight range) of male rats at the start of the study was 89 g (73–104 g), while that of the females was 91 g (67–106 g). Housing and maintenance. Animals were individually housed under conventional conditions in Type II Makrolon cages on low-dust wood granules. Cages and bedding material were changed as required, at least once a week. Feed (Altromin 1324 pellets) was provided ad libitum. The cages containing the experimental animals were placed on racks, separated by groups. The racks of the same dose groups were placed separately in closed Makrolon cabins. Each cabin was connected to the air-conditioning system. This measure was intended to protect the personnel as well as to avoid cross-contamination. Animal rooms were maintained on a 12-h light/dark cycle (fluorescent light) and targeted at a temperature of 22 { 27C and a relative humidity of 50{10%. Air was changed approximately 10 times per hour. Test substance. Trimethylphosphate (TMPO, C3H9PO4, CAS No. 51256-1) is a liquid at room temperature, freely soluble in water, and stable in aqueous solution. The purity of the batch used was 99%. Treatment. The test chemical was admixed weekly to fresh tap water without any vehicle and was available ad libitum. At each mixing levels were adjusted based on the most recent body weight and water consumption determination in order to deliver a constant average intake in mg/kg body weight per day. Stability of the test chemical in the tap water was verified analytically for concentrations covering those used in the experiment. Concentration analyses of tap water formulations for each group were determined quarterly until termination. The dose levels chosen were based on the results of the NCI bioassay in conjunction with subchronic NCI studies which indicated a rather flat dose–response curve (DHEW, 1978). Thus, the dosage regimen of 0, 1, 10, and 100 mg/kg per day was selected. The high dose was expected to represent the maximum tolerated dose (MTD), while the low dose was intended to affirm a chronic no-observed-adverse-effect-level. During the course of the first year of treatment, it turned out that the 100 mg/kg dose was too high. It was therefore reduced to 50 mg/kg in week 54. Sixty rats per sex per group were utilized of which 10 rats per sex per group were sacrificed at 12 months. The remaining 50 animals per sex per group were intended for a treatment period of up to 30 months. Clinical examinations. Cage-site examination to detect moribund or dead rats and abnormal behavior and appearance of rats was conducted twice a day (once daily on weekends and on bank holidays). Detailed examination of individual animals was performed once a week. All rats were weighed once per week during the first 3 months and once every other week for the remainder of the study. The amount of food and water consumed was determined weekly throughout the study. The individual food and water intake was, in addition, measured at approximately 13-week intervals. These data were used to calculate mean individual daily food and water consumption. Ophthalmological examinations were performed in weeks 98/99 and 128 on 10 rats per group. The pupillary reflex of both eyes was first tested in a darkened room and the anterior regions of the eye were inspected. After dilating the pupils with Mydriaticum Roche drops the refractive elements of the eyes and the fundus were examined using an indirect ophthalmoscope. Clinical laboratory investigations. Clinical laboratory investigations were conducted after 6, 12, 14 (hematology and urinalyses only), 18, 23/ 24, and 30 months on 10 randomly selected animals per group. Where possible, the same 10 rats/sex/group were used for these evaluations. Rats that died prior to any of the clinical evaluations were replaced by the next animal on the randomization list in order to maintain the sample size of each group at 10/sex. At 23 months samples of the 100/50 mg/kg group were taken just before finishing treatment of this group due to poor general condition and high mortality. Blood sampling from the remaining groups

AID

FAAT 2374

/

6k23$$$$42

11-03-97 14:49:29

was 27 days later. Therefore, comparability of data at this time point is somewhat limited. The blood samples for determination of glucose concentrations in deproteinized whole blood were taken from one of the caudal veins of nonfasted, nonanesthetized animals. The blood samples used for determining the other parameters in peripheral blood were collected from the retro-orbital venous plexus of nonfasted animals anesthetized with diethylether. The hematological parameters examined at each sampling time point included erythrocytes, reticulocytes, leukocytes, differential blood, and platelet counts, hemoglobin, hematocrit, mean corpuscular hemoglobin, mean corpuscular volume, mean corpuscular hemoglobin concentration, and thromboplastin time (Hepato-Quick method; not measured at 14 months). Blood plasma or serum was evaluated for alkaline phosphatase, lactate dehydrognase, aspartate aminotransferase, and alanine aminotransferase activity, and concentrations of total bilirubin, cholesterol, creatinine, albumin, total protein, urea nitrogen, triglycerides, inorganic phosphate, calcium, potassium, sodium, and chloride. Glucose was determined in deproteinized whole blood and protein electrophoresis on serum samples was done by separation on cellulose acetate films, staining with Ponceau S and densitometric evaluation. Urine was collected at room temperature over periods of about 16 h (overnight). During the urine collection periods the animals received water ad libitum, but no feed. Urine was measured for volume, total protein, specific gravity, and pH (month 14 only) and semiquantitatively for blood, glucose, bilirubin, protein, ketone bodies, and pH (except month 14). Sediment from each urine sample was microscopically examined. Pathological evaluation. All rats that were found dead or terminated in extremis were necropsied. Ten rats/sex/group were terminated and necropsied 12 months after study initiation. At month 24 it was decided to terminate 100/50 mg/kg animals still alive because of the high mortality in this group. All other rats surviving the 30-month test period were terminated and necropsied. The adrenals, brain, heart, kidneys, liver, lungs, ovaries, spleen, and testes were weighed at scheduled necropsies. Organ weight/ final body weight ratios were calculated. The following tissues were collected from all rats: aorta, adrenal glands, bone marrow (in sternum), brain (in toto), cecum, colon, duodenum, epididymides, esophagus, eyes with optic nerve, femur with joint, Harderian glands, heart, ileum, jejunum, kidneys, liver, lungs, lymph nodes (mesenteric / cervical), mammary glands/area, muzzle, ovaries, pancreas, pituitary, prostate, rectum, salivary gland (submandibular), sciatic nerve, seminal vesicle, skeletal muscle, skin, spinal cord, spleen, sternum, stomach, testes, thymus (if present), thyroid (/ parathyroids), tongue, trachea, urinary bladder, ureter, urethra, uterus, vagina, and all gross lesions. Collected tissues were fixed in 4% neutral buffered formaldehyde. All above-named fixed tissues (except the muzzle) were embedded in paraplast, sectioned at 3–5 mm, stained with hematoxylin and eosin, and examined microscopically. Additional sections from the peripheral nerve of the rats terminated at 12 months were stained with luxol blue/cresyl violet, and from the spinal cord of all animals on study were stained with Bodian’s silver stain and luxol blue counter stain for better visualization of nerve fibers. Statistical analyses. The statistical evaluation of data related to clinical pathology, survival, and body and organ weights as well as feed and water intake was performed using SAS routines. Three types of tests were used, the choice of the tests being a function of prior knowledge obtained in former studies. Provided that the variates in question can be considered approximately normally distributed with equal variances across treatments, the Dunnett test is used; if heteroscedasticity appeared more likely a p value adjusted Welch test is applied. If the evidence based on historical experience indicates that the assumptions for a parametric analysis of variance cannot be maintained, distribution-free tests in lieu of ANOVA are carried out, i.e., the Kruskal-Wallis test followed by adjusted Mann–Whitney–Wilcoxon U tests where appropriate. Significant differencies from the control group are indicated with ‘‘*’’ for p ° 0.05 and ‘‘**’’ for p ° 0.01. Incidences of microscopically detected tumors were evaluated by the

ftoxa

CARCINOGENICITY OF TRIMETHYLPHOSPHATE Peto test (Peto et al., 1980). Where appropriate, lesions were pooled and subjected again to statistical analyses to determine a possible effect of the treatment. The test was performed one-sided (test for increase) except for the survival analysis, which was done two-sided.

RESULTS

Analysis of Test Substance in the Drinking Water TMPO was found to be stable in drinking water for the period of use. The results of the content checks during the study complied with the target values in all dose levels. Clinical Signs and Mortality The animals of the highest dose group revealed a weakness of hind limbs beginning with week 46. This was observed in a total of 55 males and 26 females of this group. In addition, these animals showed an increased incidence of sunken flanks (especially in males), distended abdomen (especially in females), and poor general condition. Dose reduction at week 54 had no remarkably improving effect on these clinical signs. Hind limb weakness was also noticed in a few animals of the other groups including control starting mostly around week 120, which has to be attributed to old age. Survival curves for male and female rats are shown in Figs. 1 and 2, respectively. Mortality rates were increased at 100/50 mg/kg starting within the period between weeks 39 and 52. Despite the dose reduction in week 54, mortality increased further up to 70% on average in week 100 when this group was terminated. At 1 mg/kg there was no effect while at 10 mg/kg the numbers of dead animals were slightly higher but only toward the end of the study. Body Weights, Feed and Water Intake, and Ophthalmology There was no toxicologically relevant effect on body weights at 1 mg/kg (Fig. 3). In 10 mg/kg males there was a significant retardation with up to about 10% lower than control body weights. At 100/50 mg/kg there was a marked effect (up to 20% lower body weights in males and 15% in females). Male and female rats up to 10 mg/kg consumed approximately the same amount of feed per animal per day and kg body weight as the control animals. At 100 mg/kg (during the first 54 weeks) a slightly decreased mean feed intake was noticed in males. Adjusting the feed intake to body weights revealed a slightly higher intake in both sexes receiving 100/50 mg/kg. Rats of all dose groups consumed approximately the same amount of water as the control animals (data not shown). Ophthalmological examinations did not reveal any test substance-related changes. The findings resembled those common in old-age rats.

AID

FAAT 2374

/

6k23$$$$42

11-03-97 14:49:29

77

Clinical Laboratory Investigations TMPO treatment did not affect hematology parameters in males and females at 1 and 10 mg/kg. At 100/50 mg/kg hemoglobin, hematocrit, and occasionally also erythrocyte count were significantly reduced, and the proportion of reticulocytes increased essentially in males up to month 18. There was also a trend to higher thrombocyte counts up to month 18 in this group which attains statistical significance on a few occasions. In addition, a relative increase of segmented neutrophils and a corresponding decrease of lymphocytes was observed in this dose group at various time points (data not shown). Slightly increased cholesterol concentrations were noted in 100/50 mg/kg males and females. At 12 months the relative amount of the a1-globulin fraction in 100/50 mg/kg males exceeded the historical range, in consequence the albumin and the g-globulin fractions were significantly lower than those of the controls. A similar but not statistically significant trend is apparent at months 18 and 24 (data not shown). All other values did not indicate a test substancerelated effect. Urinary protein excretion seems to be increased especially in 100/50 mg/kg males at months 18 and 23. The urinary pH value (semiquantitatively determined) was distinctly lower in high dose females at 6 months and in all treated female groups at 12 months. Additional investigations at 14 months using a quantitative method roughly confirmed the 12 months results: the pH values of all female groups were significantly but not strictly dose-dependently lower than in controls. At month 23 (shortly before termination of the highest dose group) the pH seems to be rather low in high dose males and females but comparison is difficult due to the absence of values of the remaining groups in this week (data not shown). Gross Pathology At the 12-month interim termination some male and female rats of the highest dose group showed signs of hind limb skeletal muscle wasting, and the incidences of changes of the testes (small, contents fluid) appeared to be slightly increased. In addition, three high dose males showed signs of emaciation. Necropsy of intercurrently died animals revealed higher incidences of the following findings in high dose group (taking into account the markedly shorter survival time in this group): muscle wasting, changes in the lungs (e.g., mottled, reddish, pale), changes in the heart (e.g., thick, hard, abnormal color), changes in the liver (e.g., thick, scarring,), scarring of the kidney in males and females, small seminal vesicles, changes in the testes (small, soft), and fluid contents of the thoracic cavity in males, and skin edema in females. Final necropsy of high dose animals in month 24 indicates a slight increase in the frequency of small hind

ftoxa

78

BOMHARD ET AL.

FIG. 1. Kaplan–Meier survival plot of male main group rats.

limbs, scarring of the kidneys, and small, soft testes. At study termination after 30 months no changes attributable to the treatment with TMPO were found.

to neurotoxicity and muscle wasting. No remarkable differences were observed in the remaining dose groups at the 30 month termination (data not shown).

Organ Weights

Histopathology

Assessment of organ weights at the interim termination in the high dose group is difficult due to the widely differing body weights especially of males. However, it is striking that some organs (which usually show a correlation with body weights) of high dose males and females were unaffected in terms of absolute weights or even slightly increased compared with the controls. These organs were the adrenals, heart, lungs, and kidneys in males as well as the adrenals, heart, liver, and kidneys in females. Related to body weight, the values were in some cases markedly increased, and the differences were mostly statistically significant (Table 1) when compared to control values. The increased relative organ weights in high dose group were obviously produced by decreased weight of body, to a large extent attributable

Nonneoplastic changes. At interim sacrifice after 12 months the majority of 100/50 mg/kg males and females showed a peripheral nerve and spinal cord degeneration (Figs. 4–7). In addition, a myopathy of skeletal muscles was observed in some animals. Moderate or marked interstitial edema in the testes was diagnosed in two high dosed animals while it was only minimal in three controls. Spermatozoa in both epididymides were absent in another animal (Table 2). In addition, three high dose animals of the main group, dying around the time of interim necropsy, showed a moderate (one animal) or marked (two animals) tubular atrophy of the testes. Table 3 summarizes the incidences of selected histopathological changes identified as main targets at terminal sacri-

AID

FAAT 2374

/

6k23$$$$43

11-03-97 14:49:29

ftoxa

CARCINOGENICITY OF TRIMETHYLPHOSPHATE

79

FIG. 2. Kaplan–Meier survival plot of female main group rats.

fice. This table includes all main group animals available for histopathological diagnosis irrespective of the cause and timepoint of death. At 100/50 mg/kg an increased incidence of degeneration (males) and loss of nerve fibers (males and females) in the spinal cord occurred. In the peripheral nerve, fiber damage was associated with reactive cell proliferation, mainly of Schwann cells, resulting in hypercellularity. Changes of the nervous system, including sporadic hypercellularity, in animals dosed up to 10 mg/kg were those normally observed in aging laboratory rats. Therefore, they are considered spontaneous and unrelated to treatment. In the testes the incidence of tubular atrophy is increased in high dose males. The changes are more severe in degree and mostly bilateral, whereas in the remaining groups unilateral changes are predominant. The spectrum of findings in several high dose animals observed in the heart, lungs, liver, and kidneys indicates a dysfunction of the cardiovascular and respiratory system, which may be at least partly related to disturbed innervation, especially if the nerve fibers regulating the breathing or those

AID

FAAT 2374

/

6k23$$$$43

11-03-97 14:49:29

in the autonomic nervous system were affected. This, however, could not be demonstrated using the routine histopathologic evaluation. Such findings were, e.g., chronic congestion of the lungs and kidneys, formation of thrombi in the heart atria in males and females, necrosis, and lymphocytic infiltration of the liver in males. A few female animals of this group also showed further changes which possibly were associated with circulatory and respiratory dysfunction, such as edema of subcutaneous tissue and increased hematopoietic activity in the adrenal glands and bone marrow (observed as hypercellularity). Some normally occurring, age-related changes appeared to be enhanced in the high-dose group. They included atrophy of the retina and chronic progressive nephropathy in both sexes and hemosiderosis of spleen and adrenals in males. Neoplastic changes. Table 4 lists all primary neoplasms observed in animals scheduled for terminal necropsy with respect to incidence, type, malignancy, and localization. The 12-month interim examination revealed no neoplastic changes. Once more, the shorter survival time of the animals

ftoxa

80

BOMHARD ET AL.

FIG. 3. Body weight curves of male (/) and female (o) rats.

of the high dose group must be taken into account in the analysis of neoplasms. This is especially relevant for a number of tumors increasing sharply with older age, such as the adenoma of the pituitary in females, the medullary tumors in the adrenals of males, or the fibroadenoma of the mammary gland in females. In contrast to this, other intergroup differences such as tumors in the uterus and testes are not so pronounced. This could possibly be explained by the fact that, in this case, there is less pronounced dependency on the animals’ age and/or that, although these tumors do not essentially reduce the life expectancy, they nevertheless occur in earlier stages of life. Table 5 provides the numbers of animals with tumors according to whether the animals died spontaneously or were sacrificed in moribund states, at scheduled necropsy, and for all animals. These numbers indicate neither a dose-dependent difference in the total number of animals with tumors nor in tumor-bearing animals euthanized or dying during treatment or euthanized on schedule. The number of animals with more than one tumor was likewise not affected by treatment. DISCUSSION AND CONCLUSIONS

No treatment-related increase in the incidence of clinical signs was observed in dose levels up to and including 10 mg/kg. A weakness of the hind legs was observed in animals of both sexes of the high dose group beginning with week 46. In addition, an increased incidence of sunken flanks (es-

AID

FAAT 2374

/

6k23$$$$43

11-03-97 14:49:29

pecially in males), distended abdomen (especially in females), and poor general condition was noticed in this group. The dose reduction at week 54 had no remarkable improving effect on these clinical signs. Within the time period between weeks 39 and 52 mortality started to increase in high dose animals and rose to 70% in week 100, when the surviving animals were euthanized. Mortality in the dose group 10 mg/kg corresponded to that in the control group during the first 2 years of study. In the last 6 months mortality, especially of males, was slightly higher than in the control group. A substance-related effect, however, seems to be questionable as the rate was still within the range of historical studies (Bomhard, 1992) and no indications of substance-specific causes of death were given. The survival rates of the 1 mg/ kg group corresponded to those of the controls. There was no toxicologically relevant effect on body weights in males at 1 mg/kg and in females up to and including 10 mg/kg. In 10 mg/kg males body weights were up to 10% lower during the whole study period. Growth retardation was marked in high dose males (up to 20%) and females (up to 15% lower body weight compared with controls). Absolute feed intake was not affected in animals up to 10 mg/kg, while high dose male animals consumed less food. Water intake was not affected to a relevant extent. Ophthalmological and histopathological examination did not indicate an oculotoxic potential of TMPO. Hematological investigations did not reveal effects attributable to the treatment at dose levels up to 10 mg/kg. At the

ftoxa

81

CARCINOGENICITY OF TRIMETHYLPHOSPHATE

TABLE 1 Results of Organ Weighing (Interim Termination) Dose (mg/kg)

Adrenals

Heart

Lungs

Liver

Kidneys

Absolute organ weightsa Males 0 1 10 100/50

43 48 49 45

{ { { {

12.2 12.3 15.5 8.3

1071 1110 1031 1086

{ { { {

98.0 114.6 85.1 159.1

1417 1430 1449 1457

{ { { {

121.1 128.2 149.3 125.2

16089 16728 14850 14160

{ { { {

1738.7 1750.1 1499.1 1528.1*

2785 2761 2702 2689

{ { { {

269.1 215.2 346.1 299.2

{ { { {

102.0 45.6 112.6** 75.3

9676 9912 9260 10076

{ { { {

1510.3 991.5 1114.8 766.5

1821 1867 1828 1999

{ { { {

226.5 191.0 96.1 111.3

Females 0 1 10 100/50

70 66 67 72

{ { { {

20.6 12.6 19.0 13.4

789 787 779 829

{ { { {

78.7 75.5 75.0 82.0

1091 1005 963 1057

Relative organ weightsb Males 0 1 10 100/50

10 11 12 13

{ { { {

2.9 2.4 3.9 2.5

253 255 251 325

{ { { {

11.7 21.3 17.8 43.4**

336 329 352 438

{ { { {

15.8 28.7 32.5 44.8**

3801 3850 3609 4241

{ { { {

152.1 426.9 331.1 360.9*

659 635 656 807

{ { { {

30.9 47.9 73.8 88.7**

{ { { {

55.6 26.7 57.9 52.1

3668 4034 3746 4300

{ { { {

164.8 269.1* 231.0 449.0**

695 760 745 851

{ { { {

52.4 65.1* 63.6 43.6**

Females 0 1 10 100/50

27 27 27 31

{ { { {

8.1 4.6 8.6 8.1

301 321 316 352

{ { { {

17.2 29.6 24.2 21.0**

420 410 393 452

a

Data in mg. Data in mg/100 g b.w. * Difference against control for p ° 0.05 significant. ** Difference against control for p ° 0.01 significant. b

highest dose level hemoglobin, hematocrit, occasionally also erythrocyte count were significantly reduced, and the numbers of reticulocytes increased (males only). In addition, there was a trend to a higher thrombocyte count and a shift in the differential blood count (increased percentage of segmented neutrophils, decreased percentage of lymphocytes) in this dose group. The slight intensity of these effects suggests that TMPO has no primary effect on the hematopoietic system and the observed results are presumably a secondary effect caused by various dysfunctions in other organs, associated with the poor general condition of the animals. The results from clinical chemistry, gross pathology, and histopathology do not indicate treatment-related effects on the liver function and morphology at dose levels up to and including 10 mg/kg. In the dose group 100/50 mg/kg substance-related damage to the liver is indicated by occasionally significantly increased cholesterol concentrations in plasma, shifts in the serum-protein fractions in males, an

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

increased incidence of macroscopic changes (thick/scarring), increased liver weights in females as well as, histopathologically, a slightly increased incidence of necroses. Clinical laboratory investigations, organ weights, and gross pathological and histological evaluations do not point to substance-related damage to the kidneys in dose groups up to and including 10 mg/kg. At 100/50 mg/kg the protein excretion, especially of males, was slightly elevated at times, and the urinary pH value of males and females diminished. At necropsy of the main group males an increased incidence of scarring and chronic progressive nephropathy was observed with increased frequency in males and females. The relative kidney weights in this dose group were significantly increased at interim necropsy after 1 year. By and large, the results do not indicate a relevant nephrotoxic potential of TMPO. The effects observed could very well be a secondary result, e.g., of neurotoxicity. Clinical observations and gross pathological and histopathological investigations gave no indications of damage

ftoxa

82

BOMHARD ET AL.

repeated cutaneous application (2 ml/kg body weight, exposure period 2 h/day on a total of 20 days within a period of 28 days) and after repeated oral administration (0.3 ml/kg body weight daily on 6 consecutive days) and found unequivocal clinical evidence of neurotoxic effects such as flaccid paralysis followed by spasticity, unsteadiness, weakness of extremities, and fine tremors. In addition to this, information is available from an abstract by Jones and Jackson (1969) describing hind leg paresis in rats which had been treated weekly at 5 1 100 mg/kg orally for over 1 year. However, no results of gross and histopathological investigations are reported. On the other hand, no evidence of neurotoxic effects was found in adult chickens after subcutaneous treatment (Hollingshaus et al., 1981; Thyssen and Kaliner, 1981). In consideration of the reduced survival times and the findings made at interim necropsy, the incidence and degree of myopathies of the skeletal muscle in male rats of the high dose group is also to be seen as increased in a dose-related manner. The early myopathic changes consisted of atrophy

FIG. 4. Degeneration and loss of nerve fibers in the sciatic nerve of a high-dose group animal treated for 12 months (luxol blue and cresyl violet, 1360).

to the central or peripheral nervous system in dose groups up to and including 10 mg/kg. The peripheral nervous damage already clinically apparent (see above) in males and females of the high dose group manifested itself essentially as muscle wasting of the hind limbs as seen during gross pathological investigations. Histopathologically, males and females showed a peripheral nerve degeneration. In addition, a remarkable spinal cord degeneration and a myopathy of skeletal muscles were observed in several animals. The neurotoxic effect of TMPO has already been described in previous studies for other species (dog, rabbit) (Schaeppi et al., 1984; Deichmann and Witherup, 1946). Schaeppi et al. (1984) treated Beagle dogs with a daily dose of 1 to 2 ml TMPO per animal in gelatin capsules for 1 to 4 months. Neurological investigations revealed neurotoxicity already after 4-week treatment. Special neuropathological techniques have characterized the neuropathy as a distally accentuated, primary axonal lesion. Deichmann and Witherup (1946) investigated the effects on rabbits both after

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

FIG. 5. Well preserved nerve fibers in the sciatic nerve of a control animal examined at 12 months (luxol blue and cresyl violet, 1360).

ftoxa

CARCINOGENICITY OF TRIMETHYLPHOSPHATE

83

studies. The abstract by Jones and Jackson (1968) states that ‘‘500 mg/kg administered orally render male rats sterile for the ensuing 3 weeks, 5 times this amount, although tolerable, completely disorganizes spermatogenesis without damaging tubular architecture.’’ In addition to this, they report on rats ‘‘treated weekly at 5 1 100 mg/kg orally for over one year that have remained sterile but recover fertility 3–5 weeks from terminating treatment.’’ Furthermore, Wyrobek and Bruce (1975) reported a slight increase in sperm abnormalities in mice after intraperitoneal treatment at doses of 750 or 1000 mg/kg over 5 days. On the basis of these data reported in literature, special attention was paid to potential changes in the testes and findings in the epididymides and accessory reproductive glands. The results of the gross and histopathological investigations allow the conclusion that TMPO in the high dose leads to an increase in bilateral tubular atrophy in the testes. The toxicological relevance of this finding is, however, limited by the fact that some affected animals were debilitated by neuropathy and associated

FIG. 6. Loss of nerve fibers in the dorsal spinal columns of the cervical spinal cord of the same treated animal as in Fig. 4. The damage area is dark and triangular in shape, corresponding to the gracile fascicle (bodian and luxol blue, 1240).

and breakdown of groups of muscle fibers, indicating the possible neurogenic origin of the lesion. More advanced myopathy, however, was very similar to more diffuse muscle damage occurring in aging rats. Degenerating fibers in the peripheral nerve showed fragmentation of axons and myelin sheaths and in the spinal cord the lesion was most prominent in the dorsal columns at the cervical level. These features are compatible with distally accentuated axonopathy. A number of studies deal with the question of sterility described as temporary in the majority of cases, which occurred after oral or parenteral doses of 100 mg/kg or higher in mice, rats, and rabbits (Cho and Park, 1994; Dean and Thorpe, 1972; Gandy et al., 1987, 1990; Harbison et al., 1976; Hollingshaus et al., 1981, 1982; Jackson and Jones, 1968; Toth et al., 1992). These sterility effects were observed mostly after treatment periods of a few days and up to 4 weeks. However, histological investigations of the testes and accessory reproductive glands were not performed in these

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

FIG. 7. Cervical dorsal spinal columns of the same control animal as shown in Fig. 5 are well preserved (bodian and luxol blue, 1240).

ftoxa

84

BOMHARD ET AL.

TABLE 2 Incidences of Selected Nonneoplastic Changes in Animals Scheduled for Interim Sacrifice Dose (mg/kg): Sex:

0

No. of animals examined:

10

Peripheral nerve Degeneration of nerve fiber Grade 1 Grade 2 Grade 3 Spinal cord Degeneration of nerve fiber Grade 1 Grade 2 Grade 3 Skeletal muscle Myopathy Grade 1 Grade 3 Testes Tubular atrophy Grade 2 Grade 3 Edema Grade 1 Grade 3 Epididymides Absence of spermatozoa Grade 3

1

10

100/50

0

1

10 Females

10

10

10

10

10

Males 10

1 3 4

2 2

1 2

1 1

FAAT 2374

/

10

1 6 3

3 1

1

1 1

3 2

1

pathology in life-sustaining organ systems, as well as by the fact that a number of other age-associated changes in various organs appeared to be enhanced in high dose animals. The investigations produced no evidence of substance-related damage to any organs in the dose groups up to and including 10 mg/kg. At 100/50 mg/kg there were some indications in various organs of dysfunctions of the cardiovascular and respiratory system, which are most likely secondary effects of the neurotoxicity. Altogether, the no-effect-level is thus 1 mg/kg for male (based on a 10% lower than control body weight at 10 mg/kg) and 10 mg/kg for female Wistar rats. The tumor incidence, time of occurrence, spectrum of types, and localizations provided no indication of any relationship to treatment in the dose range investigated. Owing to the earlier termination of the high-dose group, a definitive statement about a potential carcinogenic effect of TMPO at the high-dose level appears somewhat limited. However, the survival-adjusted statistical analyses revealed no evidence of carcinogenicity in the 100/50 mg/kg group. Moreover, when the data from this group are compared to historical data from 2-year studies (Bomhard and Rinke, 1994), no evidence of a carcinogenic effect is provided. The obvious choice for more strict evaluation is to compare the doses 1 and 10 mg/kg with the respective control groups. The tumor

AID

100/50

6k23$$$$44

11-03-97 14:49:29

findings made in these two groups were neither dose-related nor exceeding the range of spontaneous incidences (Bomhard, 1992). Some single conspicuous incidences (e.g., four hemangiomas in the mesenteric lymph nodes of males at 10 mg/kg; 11 animals with benign medullary tumors in the adrenals at 1 mg/kg in males compared with 5 in control males, which, however, additionally showed two malignant tumors of that type) can, on the basis of historical data, be easily classified as incidental findings. On the other hand, there are also conspicuously higher frequencies in the control group (e.g., four hepatomas in males, none in the other male groups; three pituitary adenomas in males, also none in the other male groups). The overall conclusion of this study is, therefore, that TMPO proved not to be carcinogenic in Wistar rats. Particularly, the incidence of fibromas/fibrosarcomas in the subcutaneous tissue of male rats was not increased in either a statistically or a biologically significant manner. The few tumors of this type that occurred were within the range of historical controls known from 2-year and 30-month studies (Bomhard, 1992; Bomhard and Rinke, 1994). Therefore, the observations of the previous study with F344 rats (DHEW, 1978) could not be reproduced. The distribution of fibromas in the subcutaneous tissue of male F344 rats as diagnosed in the NCI study cited above

ftoxa

85

CARCINOGENICITY OF TRIMETHYLPHOSPHATE

TABLE 3 Incidences of Selected Nonneoplastic Changes in Animals Scheduled for Terminal Sacrifice Dose (mg/kg): Sex:

0

No. of animals examined:

50

12 15 2

Peripheral nerve Degeneration of nerve fiber Grade 1 Grade 2 Grade 3 Hypercellularity Grade 1 Grade 2 Grade 3 Spinal cord Degeneration of nerve fiber Grade 1 Loss of nerve fiber Grade 1 Grade 2 Grade 3 Skeletal muscle Myopathy Grade 1 Grade 2 Grade 3 Testes Tubular atrophy Unilateral Grade 1 Bilateral Grade 1 Unilateral Grade 2 Bilateral Grade 2 Unilateral Grade 3 Bilateral Grade 3

1

10

100/50

0

1

10 Females

49

48

47

49

49

50

50

10 9 2

12 10 5

4 4 3

10 17 3

10 16 1

14 9 2

5 6

1

4 4 3

2

1

1

6

Males

2

7 7 1

18 11 3

7 14 3

6 17 9

6 16 12

2 2 2 1 6 4

1 1 1 6 1

2 7 1

1 6 1 11

was as follows: 0/20 in the control group, 2/50 at 50 mg/ kg, 9/49 at 100 mg/kg. No historical control data were presented in the NCI technical report, but data published 1 year later by Goodman et al. (1979) indicate that the background rate of this tumor during that time period was only 2.6% (48/1794). A considerable study-to-study variation, however, can be assumed, since already a spot-check of some NTP reports from that time revealed at least two studies (NTP Reports 92 and 107) with a control incidence of 7/47 Å 14.9% subcutaneous fibromas. Statistical evaluation by the one-tailed Fisher exact test revealed a p value of 0.036 for high-dose males and a p Å 0.006 value in the Cochran– Armitage test for linear trend. This led to the conclusion that ‘‘trimethylphosphate was associated with the induction of benign fibromas of the subcutaneous tissue in male Fischer344 rats’’ (DHEW, 1978). In the female F344 rat and the male B6C3F1 mouse there was no indication of a carcinogenic effect. The increase in endometrial adenocarcinomas in the uterus of female B6C3F1 mice (0/16 in controls; 7/40 Å 18% at 250 mg/kg;

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

100/50

4 2

6 4

8 7 5

12 6 2

13 10 2

6 3 2

13/37 Å 35% at 500 mg/kg), however, appears to be distinct and dose-related and a treatment-related effect cannot be doubted in view of the historical data. The spontaneous incidence of this tumor type seems to be very low in B6C3F1 mice, i.e., ° 0.5% (Chandra and Frith, 1992; Haseman et al., 1985; Sheldon et al., 1995; Tamano et al., 1988; Ward et al., 1979). However, for hormone-dependent tissues such as the endometrium of the uterus, nongenotoxic mechanisms of tumor formation are more the rule than the exception (Neumann, 1991). Thus, indications of a carcinogenic effect are limited to one tissue in one species and one sex only. This seems to contradict the known genotoxic potential of this substance, which has alkylating effects in vivo and in vitro. Among the huge number of tests conducted with a variety of in vitro and in vivo experimental models there are only a few in which TMPO had no effect. In numerous mutagenicity studies TMPO was even used as a positive control substance. In view of the clear mutagenic and genotoxic potential the question arises as to why no carcinogenic effects were

ftoxa

86

BOMHARD ET AL.

TABLE 4 Incidences of Neoplasms of Animals Scheduled for Terminal Sacrifice Dose (mg/kg): Sex:

0

1

10 Males

No. of animals examined:

50

49

b b m

1 5 2

m

1

m

1

Adrenal glands Cortex lipoma Medullary tumor Medullary tumor Bone Fibrosarcoma Brain Astrocytic glioma Cerebral meninges Meningioma Granular cell tumor Harderian gland Squamous cell carcinoma Intestine, large Sarcoma, nos Liver Cholangiocarcinoma Hepatoma Histiocytic sarcoma Lungs Squamous cell carcinoma Lymphoreticular tissue Lymphoma Mammary gland Carcinoma, initial Carcinoma Adenoma Fibroma Fibroadenoma Mesenteric lymph node Hemangioma Ovaries Tubular adenoma Granulosa/theca cell tumor Pancreatic islets Adenoma Pancreas Exocrine adenoma Periocular tissue Sarcoma, nos Peripheral nerve Neuroma Peritoneum Fibroma Pituitary gland Adenoma Preputial/clitoral gland Squamous cell papilloma Squamous cell carcinoma, initial Squamous cell carcinoma Renal pelvis Transitional cell papilloma Retroperitoneum Leiomyosarcoma

AID

FAAT 2374

/

100/50

0

1

10 Females

48

47

49

49

50

11

9

2

b b

1 1

1

1

m

1

m m b m

1

1

1

1 4

1 1 1

m

2

1

m m b b b

5 2 8

b

1

4

1 4 3 1 6

5 2 10

1 3

3

1

b b

1 1

b

1

b

1

m

1

1

1

b

1

2

1

b

1 3

b m m

10

13

12

1 1 1

b

1

m

6k23$$2374

50

1

m

b

100/50

1

11-03-97 14:49:29

ftoxa

3

87

CARCINOGENICITY OF TRIMETHYLPHOSPHATE

TABLE 4—Continued Dose (mg/kg): Sex:

0

1

10 Males

No. of animals examined:

50

49

48

Skin Squamous cell papilloma Squamous cell carcinoma Basal cell carcinoma Basosquamous carcinoma Sebaceous adenoma Leiomyosarcoma Spinal cord Astrocytic glioma Spleen Lipoma Subcutaneous tissue Undifferentiated sarcoma Sarcoma, nos Fibroma Fibrosarcoma Fibrous histiocytoma Lipoma Leiomyosarcoma Angiosarcoma Neurinoma Testes Interstitial cell tumor Interstitial cell tumor Thyroid gland Follicular carcinoma C-cell adenoma Follicular adenoma Thymus Squamous cell carcinoma Thymoma Thymoma Urinary bladder Transitional cell carcinoma Uterus Carcinoma Adenoma Undifferentiated sarcoma Fibroma Leiomyosarcoma Hemangioma Vagina Sarcoma, nos Primary site uncertain Unclassified tumor Histiocytic sarcoma

b m m m b m

100/50

0

1

10 Females

47

49

49

50

1

1 1

50

1 2 1 1

1 1 1

b

1

b

1

m m b m b b m m m

1 1 1

b m

5

1 1

1

2

1

2 1 1 1

1 1 1

m b b

2

m b m

1

3 1

3

3

1 3

1

1

1 2

1

1 1

1

1 1

m

1

m b m b m b

4

1

4

4 1 1

1

1 1

1 2

m m m

100/50

1 1 1

1 1

1 1

1

Note. b, benign; m, malignant; nos, not otherwise specified.

observed in male and female Wistar rats, female Fischer344 rats, and male B6C3F1 mice and there was not even a single malignant skin tumor in the top dose in male Fischer344 rats. Could it be, for example, that in the range of the mostly very high dosages at which TMPO was tested for genotoxic

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

effects, a predictive statement on carcinogenic effects can no longer be made? Of great interest in this respect could be the clarification of the mechanisms of the uterine tumor formation in the B6C3F1 mouse. The following possibilities exist:

ftoxa

88

BOMHARD ET AL.

TABLE 5 Number of Tumor Bearing Animals Dose (mg/kg): Sex:

0

1

10

100/50

0

1

10 Females

24 12 8 1 3 0

31 25 18 7 0 0

34 13 12 1 0 0

23 10 7 3 0

25 17 11 6 0

19 10 8 2 0

16 5 3 2 0

49 29 19 9 0 1

49 29 19 7 3 0

50 35 26 9 0 0

50 18 15 3 0 0

Males

100/50

Unscheduled deaths Number of Animals examined Animals with neoplasms Animals with 1 neoplasm Animals with 2 neoplasms Animals with 3 neoplasms Animals with 4 neoplasms

23 13 9 3 1 0

21 14 12 2 0 0

30 17 12 5 0 0

33 11 10 1 0 0

26 19 12 6 0 1

Scheduled necropsies Number of Animals examined Animals with neoplasms Animals with 1 neoplasm Animals with 2 neoplasms Animals with 3 neoplasms

27 11 6 2 3

28 11 9 2 0

18 9 7 1 1

14 3 3 0 0 All animals

Number of Animals examined Animals with neoplasms Animals with 1 neoplasm Animals with 2 neoplasms Animals with 3 neoplasms Animals with 4 neoplasms

50 24 15 5 4 0

49 25 21 4 0 0

48 26 19 6 1 0

• the tumors that occurred with increased frequency only at high dosages and only in the female B6C3F1 mouse (possibly also the skin tumors observed in the male Fischer 344 rat in the high dose group) are caused by a primarily genotoxic mechanism; however, this occurs only at very high dosages irrelevant for the exposure conditions of humans; • the higher tumor incidences are not causally related to the genotoxic/mutagenic effects of TMPO, but are due to epigenetic mechanisms. The data reported here for the Wistar rat and the presented reevaluation of the available carcinogenicity data base cast further doubts on the relevance of the observed mutagenic potential of TMPO for humans. A closer examination of the metabolism and fate of TMPO may be more meaningful than further carcinogenicity studies. Our results also demonstrate that carcinogenicity studies with potentially neurotoxic agents should be carried out using reasonably low doselevels to avoid premature loss of high-dose animals owing to neurotoxicity. REFERENCES Bomhard, E. (1992). Frequency of spontaneous tumors in Wistar rats in 30-months studies. Exp. Toxicol. Pathol. 44, 381–392.

AID

FAAT 2374

/

6k23$$$$44

11-03-97 14:49:29

47 14 13 1 0 0

Bomhard, E., and Rinke, M. (1994). Frequency of spontaneous tumours in Wistar rats in 2-year studies. Exp. Toxic. Pathol. 46, 17–29. Chandra, M., and Frith, C. H. (1992). Spontaneous neoplasms in B6C3F1 mice. Toxicol. Lett. 60, 91–98. Cho, N. H., and Park, C. (1994). Effects of dimethyl methylphosphonate (DMMP) and trimethylphosphate (TMP) on spermatogenesis of rat testis. Yonsei Med. J. 35, 198–208. Dean, B. J., and Thorpe, E. (1972). Studies with dichlorovos vapour in dominant lethal mutation tests on mice. Arch. Toxikol. 30, 51–59. Deichmann, W. B., and Witherup, S. (1946). Observations on the effects of trimethyl phosphate upon experimental animals. J. Pharmacol. Exp. Ther. 88, 338–342. DHEW (1978). Carcinog. Testing Program: Bioassay of Trimethylphosphate for Possible Carcinogenicity. CAS No. 512-56-1. DHEW/PUB/ NIH-78-1331, NCI-CG-TR-81; Order PB-285851. Epstein, S. S. (1970). Mutagenicity of trimethylphosphate in mice. Science 168, 584–586. Gandy, J., Teaf, C. M., Adatsi, F. A., James, R. C., and Harbison, R. D. (1987). The dependency of male reproductive toxicity on germ cell stage. In Functional Teratogenesis (T. Fujii and P. M. Adams, Eds.), pp. 133– 145. Teikyo University Press, Japan. Gandy, J., Millner, G. C., Bates, H. K., Casciano, D. A., and Harbison, R. D. (1990). Effects of selected chemicals on the glutathione status in the male reproductive system of rats. J. Toxicol. Environ. Health. 29, 45–57. Goodman, D. G., Ward, J. M., Squire, R. A., Chu, K. C., and Linhart, M. S.

ftoxa

CARCINOGENICITY OF TRIMETHYLPHOSPHATE (1979). Neoplastic and nonneoplastic lesions in aging F344 rats. Toxicol. Appl. Pharmacol. 48, 237–248. Harbison, R. D., Dwivedi, C., and Evans, M. A. (1976). A proposed mechanism for trimethylphosphate-induced sterility. Toxicol. Appl. Pharmacol. 35, 481–490. Haseman, J. K., Huff, J. E., Rao, G. N., Arnold, J. E., Boorman, G. A., and McConnell, E. E. (1985). Neoplasms observed in untreated and corn oil gavage control groups of F344/N rats and (C57BL/6N 1 C3H/HeN)F1 (B6C3F1) mice. J. Natl. Cancer Inst. 75, 975–984. Hollingshaus, J. G., Armstrong, D., Toia, R. F., McCloud, L., and Fukuto, R. T. (1981). Delayed toxicity and delayed neurotoxicity of phosphorothioate and phosphorothioate esters. J. Toxicol. Environ. Health 8, 619– 627. Hollingshaus, J. G., and Fukuto, R. F. (1982). Effect of chronic exposure to pesticides on delayed neurotoxicity. In Effects of Chronic Exposures to Pesticides on Animal Systems (J. E. Chambers and J. D. Yarbrough, Eds.), pp. 85–120. Raven Press, New York. Jackson, H., and Jones, A. R. (1968). Antifertility action and metabolism of trimethylphosphate in rodents. Nature 220, 591–592. Jones, A. R., and Jackson, H. (1969). Chemosterilant action of trimethylphosphate in rodents. Br. J. Pharmacol. 37, 531. [Abstract] Neumann, F. (1991). Early indicators for carcinogenesis in sex-hormonesensitive organs. Mutat. Res. 248, 341–356.

AID

FAAT 2374

/

6k23$$$$45

11-03-97 14:49:29

89

Peto, R., Pike, M. C., Day, N. E., Gray, R. G., Lea, P. N., Parish, S., Peto, J., Richards, S., and Wahrendorf, J. (1980). Guidelines for simple, sensitive significance tests for carcinogenic effects in long-term animal experiments. In IARC Monographs, Supplement 2, Long-Term and Short-Term Screening Assays for Carcinogens: A Critical Appraisal, IARC, Lyon. Schaeppi, U., Krinke, G., and Kobel, W. (1984). Prolonged exposure to trimethylphosphate induces sensory motor neuropathy in the dog. Neurobehav. Toxicol. Teratol. 6, 39–50. Sheldon, W. G., Bucci, T. J., Hart, R. W., and Turturro, A. (1995). Agerelated neoplasia in a lifetime study of ad libitum-fed and food-restricted B6C3F1 mice. Toxicol. Pathol. 23, 458–476. Tamano, S., Hagiwara, A., Shibata, M. A., Kurata, Y., Fukushima, S., and Ito, N. (1988). Spontaneous tumors in aging (C57BL/6N 1 C3H/HeN)F1 (B6C3F1) mice. Toxicol. Pathol. 16, 321–326. Thyssen, J., and Kaliner, G. (1981). Unpublished data of Bayer AG. Toth, G. P., Wang, S.-R., McCarthy, H., Tocco, D. R., and Smith, M. K. (1992). Effects of three male reproductive toxicants on rat cauda epididymal sperm motion. Reprod. Toxicol. 6, 507–515. Ward, J. M., Goodman, D. G., Squire, R. A., Chu, K. C., and Linhart, M. S. (1979). Neoplastic and nonneoplastic lesions in aging (C57BL/6N 1 C3H/HeN)F1 (B6C3F1) mice. J. Natl. Cancer Inst. 63, 849–854. Wyrobek, A. J., and Bruce, W. R. (1975). Chemical induction of sperm abnormalities in mice. Proc. Natl. Acad. Sci. USA 72, 4425–4429.

ftoxa