Arch. Environ. Contam. Toxicol. 46, 528 –533 (2004) DOI: 10.1007/s00244-003-2285-5

A R C H I V E S O F

Environmental Contamination a n d Toxicology © 2004 Springer-Verlag New York LLC

Effects of Vinclozolin on Spermatogenesis and Reproductive Success in the Japanese Quail (Coturnix coturnix japonica) L. Niemann,1 B. Selzsam,1,2 W. Haider,3 C. Gericke,2 I. Chahoud2 1 2 3

Bundesinstitut fu¨r Risikobewertung, Thielallee 88-92, D-14195 Berlin, Germany Freie Universita¨t Berlin, Institut fu¨r Klinische Pharmakologie und Toxikologie, Garystrasse 5, D-14195 Berlin, Germany Institut fu¨r Tierpathologie, Scho¨nhauser Str. 22, D-13127 Berlin, Germany

Received: 6 December 2002 /Accepted: 13 October 2003

Abstract. In a one-generation reproduction study, the major agricultural fungicide vinclozolin was administered to adult Japanese quail for a period of 6 weeks at dietary levels of 125 and 500 ppm. Fertility and reproductive performance were not affected up to the highest concentration, although the examination of additional endpoints in the drakes (spermatid count, histology of the testis) provided some evidence of an inhibition of spermatogenesis at both dietary concentrations. Likewise, there were no indications of systemic toxicity in the adult birds. Plasma hormone concentrations (estradiol, testosterone, progesterone, T3, and T4) showed a large interindividual variance but treatment-related differences between the groups could not be established. There were no clear-cut indications of antiandrogenic effects in quail, although a limited transfer of the test substance into the eggs was proven.

Vinclozolin (3-[3,5-dichlorophenyl]-5-methyl-1,3-oxazolidine2,4-dione; CAS No. 50471-44-8) is the active ingredient in a number of contact fungicides that are widely used in a variety of crops including fruit, in particular, grapes, vegetables, ornamentals, grain legumes, and oil seed rape. It provides effective control of different pests like Alternaria, Monilinia and Sclerotinia spp., and Botrytis cinerea. The compound is of a very low acute toxicity to mammals and devoid of mutagenic potential. However, there is serious health concern because of certain findings suggesting endocrine disruption. In long-term studies in rodents, a dose-related increase in neoplasia of hormone-dependent or hormone-producing organs such as the testis, the adrenals, or the ovaries was observed (Anonymous 1997). Reproductive hazard became obvious in multigeneration and teratogenicity studies in rats and was mainly characterized by morphological feminization of genetically male pups. This specific alteration was evidenced by hypoplasia of the testis and the penis, a reduction in anogenital distance, and atrophy or chronic inflammation of the epididymis, the seminal

Correspondence to: L. Niemann

vesicle, the coagulation gland, and the prostate. In contrast, no such findings were noted in rabbits. These dose-related antiandrogenic effects leading to infertility of adult male rats are assumed to result from an inhibition of testosterone binding to androgen receptors by at least two metabolites exhibiting a higher affinity to the respective molecular receptor sites (Gray et al. 1994; Chapin et al. 1996). The lowest no-observed adverse effect level (NOAEL) for adverse effects of vinclozolin on rat pups was recently established by Hellwig et al. (2000) in a teratogenicity study in the Long–Evans strain at 6 mg/kg body weight/day. However, a sufficient margin of safety between dose levels causing clinically apparent endocrine disruption and the concentrations to which consumers and operators might be exposed does exist and a significant risk for humans is not anticipated (Anonymous 1997). Since exposure of birds to vinclozolin under normal environmental conditions is not unlikely, a possible impact on reproduction in this vertebrate class should be seriously considered. Reproduction studies have been conducted in the bobwhite quail (Colinus virginianus) and in the mallard duck (Anas platyrhynchos), revealing a NOAEL of 125 ppm in both species (Anonymous 1997). In a preliminary study in the Japanese quail (Coturnix coturnix japonica) in our laboratory, dietary levels of 50, 125, and 250 ppm vinclozolin were employed but no impact on any of the reproduction-related parameters has been observed up to the highest concentration (Chahoud et al. 2000, unpublished). In this paper, we report the results of a more recent one-generation reproduction study in this species that was run according to the OECD proposal for a new avian reproduction toxicity guideline (OECD 2000) with the exception that only two, instead of three, treatment groups and the control were included. The dietary levels of 125 and 500 ppm are well above the maximum residue concentrations of about 30 ppm that were expected to occur under worst-case conditions in potential avian feed such as leaves and small insects (Anonymous 1997). In excess of guideline requirements, additional toxicological endpoints (hormone concentrations, spermatid count, histology) were examined to investigate their suitability for the detection primarily of hormone-mediated effects.

Vinclozolin and Reproductive Success in the Quail

Materials and Methods Thirteen- to 19-week-old Japanese quail were administered vinclozolin (batch No. 183; BASF AG, Ludwigshafen, Germany; purity, 99.1%) via the diet for a total of 6 weeks to investigate, in particular, adverse effects on reproduction including hatchling development but also the occurrence of systemic toxicity and possible transfer of the compound into the eggs. Young birds (breeder: Ku¨ berich, Geesdorf/ Wiesentheid, Germany) were brought into our animal quarter at 6 weeks of age, allocated to one of three groups, and adapted to the laboratory conditions for a period of 5 weeks. Male and female animals were singly housed in neighboring wire pens with sloped floors and egg catchers at a room temperature of 22 ⫾ 2°C and 55 ⫾ 10% relative humidity. Air changes per hour were kept constant (15⫻). Daily photoperiod was 16 h light:8 h dark. Light intensity in the cages was about 300 lux. Adult birds as well as chicks were fed a standard diet (Altromin Extrudat Wachtel-Haltungsfutter 0729; Altromin GmbH, Lage, Germany) and received tap water ad libitum. From week 8 onward, the quail were mated once a day (five times per week) by introducing the drake into the cage of the allocated hen for not more than 20 min. At the end of the adaptation period, the animal number was reduced to 18 pairs per group and the couples were definitely allocated to the control (Group 0) and the two treatment groups (Groups 1 and 2). Only proven breeders (i.e., with at least one egg laid before) in good health were included. Then all test parameters were monitored in a 2-week pretreatment period during which all animals were still maintained on untreated diet. From the beginning of the subsequent administration period, the test substance was admixed to the diet of the two treatment groups once a week using a premix without the need for a solvent. The intended dietary concentrations for Groups 1 and 2 were 125 and 500 ppm, respectively. Actual dietary concentrations of the test substance were subject to regular examinations. Food samples were taken from the dose groups weekly during the treatment period and once from the control group. Samples were kept frozen (⫺20°C) and sent to the manufacturer for qualitative and quantitative analysis by gas chromatography with flame ionization injection. Examination of the eggs for residues of vinclozolin or its metabolites was also performed in a laboratory of BASF AG (Agrarzentrum Limburgerhof, Germany) on pooled egg samples from the adaptation phase and from eggs laid during the first and fifth weeks of substance administration. Following complete hydrolysis, total residues are quantitatively determined as 3,5-dichloroaniline by means of gas chromatography with subsequent mass spectrometry. Adult quail as well as chicks were monitored daily for abnormal behavior, signs of toxicity, and mortality. Body weights of male and female birds were determined after arrival at our testing facility, immediately before the substance administration started, and at scheduled termination. The drakes were additionally weighed in the third week of treatment. Food consumption was only roughly estimated based on replicates of three pairs each. In the last week of treatment, about 1 ml blood per animal was collected from all adult quail by venous puncture of a wing (brachial) vein into NH4-heparinized tubes and centrifuged to separate the cellular components. Plasma was frozen for subsequent determination of the following hormone concentrations: testosterone, estradiol, progesterone (only in females), T3, and T4. ELISA kits for hormone measurements in humans were used according to the recommendations of the manufacturers (IBL, Hamburg, FRG, for testosterone and DRG, Marburg, FRG, for all other hormones). In previous experiments, these test kits proved applicable without further modification for investigations in Japanese quail, whereas a kit for LH determination was found not to be suitable (Selzsam et al., in preparation). Eggs laid during the pretreatment and the administration period

529

were stored in a refrigerator at 16 ⫾ 1°C and 60 –70% humidity before incubation started. Storage was limited to less than 8 days. Eggs were incubated for 2 weeks in an automatic incubator at 37.8 ⫾ 0.1°C and 60 ⫾ 5% relative humidity and turned over the egg equator several times a day. Three or four days prior to hatch, the eggs were moved to another incubator, where they were exposed to a temperature of 37.5 ⫾ 0.1°C and to 80 ⫾ 5% relative humidity. Hatched chicks were not removed from the incubator until they became completely dry. Eggshell thickness was measured weekly in one “intact” egg per pen at five sites around the equator after the eggs had been opened, washed out, and dried for at least 48 h at room temperature according to Bennett et al. (1988). The shell thickness of cracked eggs was evaluated separately. The number of eggs and their weights were assigned to the respective week of the pretreatment and treatment periods and are expressed as number of eggs and mean egg weight per hen and week. The percentage of broken/cracked eggs and of fertile eggs, the occurrence of “death in shell,” and the number and weight of hatchlings were determined for the eggs laid during the whole pretreatment and treatment period. The hatchlings from the eggs laid during the second week of pretreatment and in the second, fourth, and sixth weeks of the treatment period were allowed to grow up. All chicks from the remaining weeks were killed immediately after hatch, counted, and monitored for external malformations. At the end of the 14-day growing periods, the weight and number of survivors were registered and the sex distribution among chicks was determined. The differentiation between males and females was based on visual detection of testes or ovaries upon necropsy following sacrifice by exposure to CO2. At study termination, the adult birds were sacrificed by decapitation and examined macroscopically. The livers and the testes were removed and organ weights determined. From six animals per sex and group, the liver was fixed in 4% formaldehyde, dissected, and embedded in paraffin for subsequent histopathological examination. In the same way, the right testis from six drakes per group was prepared for histopathology. Tissue samples from both organs were stained with hematoxylin– eosin (HE) and periodic acid/Schiff’s reagent (PAS). The left testis of all males was used for spermatid counting. For this purpose, epididymis and Tunica albuginea were gently removed and the testis was minced and homogenized in 10 ml of a 0.9% NaCl solution containing 0.5% Triton X-100. After 100 ␮l of this homogenate had been diluted at a ratio of 1:9 with the same solution once more, one drop was taken and filled into a Buerker counting chamber with a field size of 0.0025 mm2 and a depth of 0.1 mm. After 5 min, the spermatides were visually counted under a Zeiss microscope equipped with a Plan-Neofluar ocular (magnification, ⫻40). Total spermatid count of the testis was calculated and is expressed as millions per testis. This method has been widely used in rats for some time (Dalsenter et al. 1996; Faqi et al. 1997) and was successfully adapted to quail in our laboratory. Body weights and reproduction parameters were subject to statistical analysis by means of the SAS standard software package (version 8). The reproduction data were collected on a weekly basis and a comparison was made between the control and the treatment groups. For testing the differences observed for statistical significance, chisquare test and ANOVA followed by Dunnett’s test or Fisher’s exact test were used. When the distributions were not clearly Gaussian, the Wilcoxon rank sum test was applied. The significance level was 5% in each test. Furthermore, 2-week intervals were considered to enable comparison of data from the initial, intermediate, and terminal phases of the treatment period and those obtained during the pretreatment phase. However, no vertical covariate testing between pretreatment and treatment phases within the same experimental group was performed.

530

L. Niemann et al.

Table 1. Residues of vinclozolin in eggs (mean values)

Table 3. Atrophy of germinating epithelium in quail testis

Week of treatment

0 ppm

125 ppm

500 ppm

1st 5th

⬍0.05a ⬍0.05a

1.99 2.31

6.21 4.69

a

Atrophy incidence

0 ppm

125 ppm

500 ppm

0/6

2/6

4/6

Limit of detection.

Table 2. Body weight (g) in male and female adult quail (mean ⫾ SD) 0 ppm

P, 2nd week T, 3rd week T, 6th week

125 ppm

500 ppm

Male

Female

Male

Female

Male

Female

237.1 ⫾ 17.7 (n ⫽ 17) 247.1 ⫾ 17.6 (n ⫽ 17) 251.2 ⫾ 22.5 (n ⫽ 16)

268.1 ⫾ 24.6 (n ⫽ 17)

242.3 ⫾ 26.1 (n ⫽ 18) 253.4 ⫾ 26.2 (n ⫽ 18) 257.9 ⫾ 26.4 (n ⫽ 16)

278.9 ⫾ 27.7 (n ⫽ 18)

239.1 ⫾ 19.0 (n ⫽ 18) 248.3 ⫾ 20.2 (n ⫽ 18) 254.4 ⫾ 23.4 (n ⫽ 18)

263.7 ⫾ 26.1 (n ⫽ 18)

291.7 ⫾ 34.3 (n ⫽ 16)

326.2 ⫾ 37.4 (n ⫽ 16)

292.5 ⫾ 35.1 (n ⫽ 18)

Note. P, pretreatment phase; T, treatment phase; n, number of animals. No statistically significant differences obtained in the ANOVA followed by Dunnett’s test.

Results Analytical determination of vinclozolin in the diet revealed concentrations that were sufficiently close to the nominal values, although in some weeks rather large deviations in both directions were observed. In the individual weeks, deviations from the intended dietary levels ranged from 0.5 to 10.5% for Group 1 and from 0.1 to 9.7% for Group 2. Average food concentrations for the whole 6-week administration period accounted for 124.5 and 482.65 ppm, respectively. A limited, dose-related transfer of the test substance into the eggs was noted (Table 1). A longer duration of exposure did not result in an increase in egg residues. Administration of vinclozolin did not cause any clinical signs in adult birds. The few animal losses during the course of the study were confined to the control and low-dose groups. These unscheduled mortalities were due to pericarditis or the occurrence of bumble foot requiring euthanasia for humane reasons and, thus, clearly not compound-related. At study termination, all 18 pairs were still alive in Group 2, whereas only 16 pairs were available for evaluation in the control and low-concentration groups. Body weight and body weight gain did not differ in either sex among the three groups (Table 2). Likewise, food consumption was obviously not altered. At necropsy, no gross pathological lesions were seen that could be attributed to substance administration or were dose-related. With regard to the absolute and relative organ weights of the liver and the testes, no differences between the control and the treated groups became apparent. Histopathological examination of the liver provided some rather equivocal evidence of a weak hepatotoxic potential of vinclozolin since the frequently seen degenerative (fat deposition, vacuolization, occurrence of signet-ring cells) and inflammatory changes in both sexes appeared more pronounced in the treated birds. However, the total number of affected animals was similar (four to six per sex and dietary level) among the three groups, and thus, a final conclusion cannot be drawn. In contrast, the dose-related atrophy of the germinating

epithelium in quail testis might suggest an effect of the test substance on this organ in a more convincing manner (Table 3), although these alterations were evaluated by the pathologist as minor changes only. Microscopically, the atrophy was characterized by a reduced epithelial height and the occurrence of gaps in the epithelium. The histology-based assumption of a possible weak impact of the test compound on spermatogenesis was confirmed by a significant reduction in mean spermatid count at the highest treatment level (Table 4). A comparison of the plasma hormone concentrations measured during the last week of the administration period (Table 5) did not reveal statistically significant or dose-related differences between the control and the treated groups in either male or female birds. Reproductive performance was not affected by treatment. Neither the absolute number of eggs per week nor the number of eggs laid per hen showed significant differences among the three groups on study. Mean egg weight was slightly higher in the groups receiving vinclozolin. Since this increase was not dose-related and was already apparent during the pretreatment period (Table 6), it is not considered to result from compound administration but, rather, from random distribution. It could be shown that there were three hens in the control group with particularly light eggs and one hen in each of the two dose groups laying outstandingly heavy eggs throughout the study. Absolute number and percentage of cracked eggs were slightly increased in the group receiving 125 ppm but were very similar to the control at the highest dietary level again (Table 6). Eggshell thickness was not different between the groups when only “intact” eggs were considered. Separate assessment of the cracked eggs revealed a significant deviation between the control and both treated groups but this was clearly due to the particularly thin eggshell in the control animals compared to the “intact” eggs from the same group. Although the percentage of fertile eggs among all eggs incubated was rather low throughout the study, with weekly

Vinclozolin and Reproductive Success in the Quail

531

Table 4. Spermatid count 0 ppm

Spermatids ⫻ 10 per testis 6

125 ppm

500 ppm

n

Mean ⫾ SD

n

Mean ⫾ SD

n

16

339 ⫾ 55

16

309 ⫾ 33

18

Mean ⫾ SD 250 ⫾ 58*

Note. n, number of drakes examined. * p ⬍ 0.05, ANOVA followed by Dunnett’s test. Table 5. Mean ⫾ SD plasma concentrations of hormones (sixth week of treatment) 0 ppm

Estradiol (pg/ml) Testosterone (ng/ml) T3 (ng/ml) T4 (ng/ml) Progesterone (ng/ml)

125 ppm

500 ppm

Male

Female

Male

Female

Male

Female

10.1 ⫾ 3.7 (6) 7.3 ⫾ 1.8 (16) 1.0 ⫾ 0.3 (16) 33.0 ⫾ 8.0 (16)

98.8 ⫾ 44.1 (16) 0.5 ⫾ 0.3 (16) 1.1 ⫾ 0.5 (16) 21.2 ⫾ 7.7 (16) 0.5 ⫾ 0.7 (15)

11.0 ⫾ 3.9 (6) 7.4 ⫾ 2.5 (16) 1.0 ⫾ 0.3 (16) 38.4 ⫾ 26.5 (15)

99.9 ⫾ 37.0 (16) 0.5 ⫾ 0.3 (16) 1.2 ⫾ 0.3 (16) 21.1 ⫾ 7.9 (8) 0.6 ⫾ 0.8 (15)

10.0 ⫾ 2.2 (4) 7.0 ⫾ 2.3 (18) 1.1 ⫾ 0.4 (18) 37.9 ⫾ 18.0 (17)

83.2 ⫾ 20.7 (18) 0.4 ⫾ 0.2 (18) 1.0 ⫾ 0.3 (17) 22.5 ⫾ 8.3 (10) 0.5 ⫾ 0.6 (17)

Note. No statistically significant differences obtained in the ANOVA followed by Dunnett’s test. Table 6. Egg parameters (mean ⫾ SD) during the pretreatment (2nd week; 13 weeks old) and treatment (14th to 19th weeks) phases

Pretreatment 0 ppm 125 ppm 500 ppm Treatment 0 ppm 125 ppm 500 ppm

Egg weight (g)

Number and percentage of cracked eggs

Eggshell thickness (mm)

Total number of eggs laid

Intact eggs

Cracked eggs

104 108 104

11.8 ⫾ 0.9 12.4 ⫾ 0.9* 12.2 ⫾ 1.2*

7 ⫽ 7% 7 ⫽ 6% 13 ⫽ 12%

0.150 ⫾ 0.014 0.154 ⫾ 0.013 0.158 ⫾ 0.015*

0.131 ⫾ 0.010 0.144 ⫾ 0.010 0.130 ⫾ 0.017

621 616 615

11.9 ⫾ 1.0 12.4 ⫾ 0.9* 12.2 ⫾ 1.2*

44 ⫽ 7% 67 ⫽ 11%† 55 ⫽ 9%

0.151 ⫾ 0.017 0.151 ⫾ 0.017 0.152 ⫾ 0.016

0.135 ⫾ 0.013 0.146 ⫾ 0.015* 0.146 ⫾ 0.021*

Note. * p ⬍ 0.05, ANOVA followed by Dunnett’s test. † p ⬍ 0.05, chi-square test.

rates of only about 70%, no statistically significant or concentration-related differences between the control and the treated groups were recorded. Likewise, administration of vinclozolin had no impact on the hatchling rate, viability of hatchlings, or frequency of external malformations. With regard to the body weights of the hatchlings and of the chicks after the 14-day growing period, large deviations between the groups and study weeks became apparent. Sometimes these differences gained statistical significance but a clear dose response was lacking. Taking into consideration that similar deviations occurred during the pretreatment period and were frequently even more pronounced then, a treatment-related effect is not assumed. The 14-day survival rate was characterized by a high variability within the groups and individual weeks, ranging from only 48 up to 76%. However, there was no evidence of an adverse effect of vinclozolin since a better survival rate was observed at the highest dietary level compared to the control group throughout the study. Because of the specific antiandrogenic activity of vinclozolin in male rat pups, it was of particular interest to determine the quantitative sex ratio among the quail chicks surviving until day 14 (Table 7). On one hand, it must be acknowledged that, during the

pretreatment period, the number of male chicks in all groups markedly exceeded that of the females. When vinclozolin was administered, this pattern shifted toward a more equal sex distribution in both treated groups but not in the control. For the high-concentration group, this difference achieved statistical significance (52% female chicks, compared to only 34% in the control group), suggesting a treatment-related effect. On the other hand, this apparent predominance of the male sex in chicks from the untreated group had not been noted in our laboratory before, although the origin of the Japanese quail was always the same. However, the historical database is rather limited, and to our knowledge there is no published data on sex distribution among quail chicks in reproduction studies. Up to now, not more than five studies have been run in our laboratory during which sexing of the chicks was performed (Table 8).

Discussion Summarizing our results, it can be concluded that the dietary administration of the fungicide vinclozolin over 6 weeks to Japanese quail did not cause general toxicity (mortality, clini-

532

L. Niemann et al.

Table 7. Distribution between male and female quail chicks during the pretreatment and the administration period 0 ppm

P, 2nd week T, 2nd week T, 4th week T, 6th week Total

125 ppm

500 ppm

Male

Female

Male

Female

Male

Female

16 11 17 17 45

11 8 10 6 24

18 11 14 17 42

7 14 11 12 37

20 14 13 12 39*

10 15 16 12 43*

Note. P, pretreatment phase; T, treatment phase. * p ⬍ 0.05, chi-square test. Table 8. Sex distribution among 14-day-old chicks from untreated hens in five reproduction studies during pretreatment (all groups included) and administration period (control groups only) Pretreatment period

Administration period

Males

Males

Females

Females

Study A, 1999 Study B, 1999

56 110

54 93

65 15

63 17

Study C, 2000 Study D, 2001 Study E, 2002

117 54 40

105 28 35

70 45 39

63 24 17

Remark Source Unpublished Unpublished Solecki et al. (2001) Current study Unpublished

cal signs, body weight effects) in adult birds and did not impair reproductive success up to the highest concentration of nearly 500 ppm. This was mainly evidenced by the lack of doserelated changes in such parameters as laying and hatchling rate, percentage of cracked eggs, and survival and body weight development of the chicks during the 14-day growing period, although limited transfer of vinclozolin or its metabolites into the eggs has been confirmed, and accordingly, exposure of the embryos in ovo must be assumed. Unfortunately, the generally low fertility rate, only about 70% of the total number of eggs set, prevented the setting of a clear-cut NOAEL. Once more, the advantage of including a pretreatment period of at least 2 weeks has become apparent. Even if no statistical comparison is performed as in our study, these data facilitate a more reliable interpretation of the study results. For the assumption of true test compound effects, alterations in reproduction-related endpoints should have been observed only in the administration period, not during pretreatment. With regard to a possible effect of the test compound on sex distribution among quail chicks, no definite conclusion can be drawn. Although our findings suggest that the administration of vinclozolin to Japanese quail might increase the amount of morphologically female chicks, this is not likely to have actually occurred against the background of our small historical database. A total of three studies from 1999 and 2000 clearly points to an unusual ratio between male and females chicks in the control group of the current experiment and contradicts the assumption of a possible impact of vinclozolin on quantitative sex ratio similar to that observed in rats, whereas a different pattern between pretreatment and administration period does not allow conclusive assessment of a fourth and more recent

one. Thus, we consider the skew distribution toward the male sex among the chicks from all groups during the pretreatment phase and from the control group during the treatment period rather a chance finding, although the occurrence of a substancerelated effect in the current study cannot be completely excluded. One of our main objectives was to continue the prevalidation of additional methods and endpoints that are under consideration for possible inclusion in reproduction studies in birds. Histopathology provided evidence of a substance-related increase in the incidence of histological alterations in the testis at both dietary levels, suggesting that this method, because of its high sensitivity, might be suitable for detection of subtle effects. However, the number of animals (six per sex and group) was rather small and should be extended in future studies to support more reliable evaluation. Counting of spermatids revealed a statistically significant reduction at the highest concentration. Since the (generally low) percentage of fertile eggs was not further impaired by treatment in any of the groups, an adverse effect of this decrease on reproduction is not to be assumed. It is well known that even a strong reduction in the number of spermatids is tolerated by rats or rabbits without a loss in fertility, whereas the same endpoint is considered a much more critical one in men (Working 1988). To our knowledge, there is no information available to what extent the spermatid count can be altered in bird species until fertility is impaired. Comparison of the plasma concentrations of four (male quail) or five (females) hormones did not show statistically significant intergroup differences. As in many animal species, most hormones exhibit a large interindividual variability also in quail. Therefore, it appears difficult to achieve meaningful and reliable results without including a much higher number of birds, which is not feasible for ethical and economical reasons and would surely exceed the resources of most testing facilities. According to our current experience, testosterone in male quail and estradiol in females seem to be the most appropriate candidates for future diagnostic use (Selzsam et al., in preparation). Thus, for all additional parameters investigated, questions remain open and further research is needed. However, our results provide further evidence that the inclusion of new endpoints may enhance the potential of a study to detect even weak effects. If only the “classical” avian reproduction endpoints are taken into account, our findings might suggest a lower sensitivity of the Japanese quail compared to other bird species when exposed to vinclozolin. In the bobwhite quail, the laying rate was diminished at the highest dietary level of 250 ppm. An increase in embryonic deaths resulted in a lower hatchling rate. Furthermore, chick survival during the first 2 weeks after hatch was affected. The next-lower concentration, 125 ppm, was considered the NOAEL for reproductive effects in this species. In the mallard duck, the hatchling rate was decreased at 250 ppm due to higher embryonic mortality. Again, the NOAEL was established at 125 ppm (Anonymous 1997). These findings were not confirmed in the Japanese quail, neither in the current experiment nor in a previous study in our laboratory (Chahoud et al. 2000, unpublished). However, the results of examining additional endpoints (spermatid count, histopathology) provide some evidence that the assumption of

Vinclozolin and Reproductive Success in the Quail

a generally lower vulnerability of this species to vinclozolin is not justified.

Acknowledgments. This reproduction study was funded by the German Federal Department of Environment, Nature Protection, and Nuclear Safety (“Umweltforschungsplan,” Forschungsvorhaben 46 04366) and by the former German Federal Institute for Health Protection of Consumers and Veterinary Medicine. We are indebted to Mrs. Barac, Mrs. Ladwig, Mrs. Massow, Mrs. Schwarck, Mrs. Stu¨ rje, Mrs. Woelffel, and Mrs. Zarnecke for animal care and reliable technical assistance.

References Anonymous (1997) Vinclozolin (monograph pepared in the context of the inclusion of the active substance in Annex I of the Council Directive 91/414/EEC). Rapporteur Member State: France. European Commission, Peer Review Programme Bennett JK, Ringer RK, Bennett RS, William BA, Humphrey PE (1988) Comparison of breaking strength and shell thickness as evaluators of eggshell quality. Environ Toxicol Chem 7:351–357 Chahoud I, Niemann L, Solecki R (2000) Pru¨ fung reproduktionstoxikologischer Parameter an der Japanischen Wachtel (Coturnix coturnix japonica). Wirkstoff: Vinclozolin (Unpublished report No. 01180399 to the German Ministry of Environment, Nature

533

Protection and Nuclear Safety. Available from the authors, in German) Chapin RE, Stevens JT, Hughes CL, Kelce WR, Hess RA, Daston GP (1996) Endocrine modulation of reproduction. Symposium overview. Fundam Appl Toxicol 29:1–17 Dalsenter PR, Faqi AS, Webb J, Merker H-J, Chahoud I (1996) Reproductive toxicity and tissue concentrations of lindane in adult male rats. Hum Exp Toxicol 15:406 – 410 Faqi AS, Klug A, Merker H-J, Chahoud I (1997) Ganciclovir induces reproductive hazards in male rats after short-term exposure. Hum Exp Toxicol 16:505–511 Gray LE, Ostry JS, Kelce WR (1994) Developmental effects of an environmental antiandrogen: The fungicide vinclozolin alters sex differentiation of the male rat. Toxicol Appl Pharmacol 129: 46 –52 Hellwig J, van Ravenzwaay B, Mayer M, Gembardt C (2000) Pre- and postnatal oral toxicity of vinclozolin in Wistar and Long-Evans rats. Regul Toxicol Pharmacol 32:42–50 OECD (Organisation for Economic Cooperation and Development) (2000) OECD guideline for testing of chemicals. Proposal for a new OECD test guideline, “Avian Reproduction Toxicity in the Japanese Quail or Northern Bobwhite.” OECD Internet home page Solecki R, Niemann L, Gericke C, Chahoud I (2001) Dietary administration of dimethoate to the Japanese quail: Reproductive effects and successful hatchability of eggs. Bull Environ Contam Toxicol 67:807– 814 Working PK (1988) Male reproductive toxicology: Comparison of the human to animal models. Environ Health Perspect 77:37– 44