Poul Bjerregaard

Science and policy on endocrine disrupters must not be mixed: a reply to a “common sense” intervention by toxicology journal editors

The “common sense” intervention by toxicology journal editors regarding proposed European Union endocrine disrupter regulations ignores scientific evidence and well-established principles of chemical risk assessment. In this commentary, endocrine disrupter experts express their concerns about a recently published, and is in our considered opinion inaccurate and factually incorrect, editorial that has appeared in several journals in toxicology. Some of the shortcomings of the editorial are discussed in detail. We call for a better founded scientific debate which may help to overcome a polarisation of views detrimental to reaching a consensus about scientific foundations for endocrine disrupter regulation in the EU.

Vitellogenin induction by 17β-estradiol and 17α-ethinylestradiol in male zebrafish (Danio rerio)

Adult male zebrafish (Danio rerio) were exposed to 17beta-estradiol (E2) or 17alpha-ethinylestradiol (EE2) in flow-through systems for 8 days. This was done to compare the sensitivity of the estrogen inducible vitellogenin (Vtg) biomarker system of this proposed OECD test guideline species to other relevant test species. Vtg was quantified in whole body homogenate by a species-specific ELISA. Actual water concentrations of E2 and EE2 were quantified by LC-MS, with detection limits of 1.0 and 0.6 ng/l, respectively. Vtg induction (LOEC) occurred in whole body homogenate at actual water concentrations of 21 ng E2/l and 3.0 ng EE2/l, respectively. As an alternative to the ANOVA approach, the relationship between the percentage of responding fish (Vtg) and the external E2 or EE2 concentration was determined by logistic regression analysis. Based on the regression analysis, EC-values could be determined: EC10, EC50 and EC90 were 15.4, 41.2 and 67.1 ng E2/l, respectively and 0.92, 2.51 and 4.09 ng EE2/l, respectively. Comparisons of these response limits to corresponding values for rainbow trout (Oncorhynchus mykiss), fathead minnow (Pimephales promelas) and Japanese medaka (Oryzias latipes) revealed the zebrafish as a sensitive test species.

Development of an ELISA for vitellogenin in whole body homogenate of zebrafish (Danio rerio)

The yolk protein, lipovitellin (Lv) was purified from ovaries of mature female zebrafish (Danio rerio) by gel filtration and anion exchange chromatography. Polyclonal antibodies against Lv were raised in rabbits. Anti-Lv IgG was purified by affinity chromatography. SDS-PAGE followed by Western blotting was performed to analyse the specificity of the antibody and the immunological similarities between Lv and vitellogenin (Vtg). Anti-Lv IgG was used to develop a direct non-competitive sandwich ELISA to measure Vtg concentrations of whole body homogenate (WBH) in zebrafish. The intra- and interassay variabilities were 5.8% and 10.4%, respectively. The sensitivity was 0.2 ng Vtg x ml(-1) and the practical detection limit was 40 ng Vtg x g(-1) fish (wet weight). Adult male zebrafish were exposed to a nominal water concentration of 10 ng x l(-1) of ethinylestradiol (EE2) in a semi-static exposure system for 7 days. Compared with the control group, exposure to 10 ng EE2 x l(-1) induced a 200-fold increase in Vtg levels.

Cadmium accumulation in Littorina littorea, Mytilus edulis and Carcinus maenas: the influence of salinity and calcium ion concentrations

Accumulation of waterborne cadmium in Littorina littorea, Mytilus edulis and Carcinus maenas (collected in 1988 and 1989 around the island of Funen, Denmark) was investigated in a matrix of salinities (10 to 30‰) and calcium concentrations (2.9 to 8.9 mM Ca++). Cadmium accumulation rates in soft parts of L. littorina, soft parts and shells of M. edulis and whole bodies and exoskeletons of C. maenas decreased with increasing salinity. Changes in the calcium concentrations accounted for 72% of the ‘salinity effect’ on cadmium accumulation rates in L. littorina, whereas calcium concentrations had little or no effect on cadmium accumulation in M. edulis. Cadmium accumulation in the whole body of C. maenas was affected equally by calcium concentrations and total salinity, whereas accumulation in the exoskeleton was mainly affected by changes in total salinity. Individual variability in cadmium accumulation in the organs of C. maenas was greater than the variation attributable either to changes in ambient calcium concentrations or total salinity. An appreciable amount of the inter-individual variability in the cadmium accumulation in all three species was correlated with wet:dry weight ratios of the tissues and size of the organisms.

Trenbolone causes irreversible masculinization of zebrafish at environmentally relevant concentrations

Feminization of fish caused by certain estrogenic compounds e.g. 17 alpha-ethinylestradiol (EE2) has been shown to be partly reversible. So far it has not been studied if this applies for androgenic compounds too. The androgenic steroid trenbolone acetate (TbA) is used as growth promoter in beef cattle in the United States, South America, and Australia. TbA metabolites are stable in animal waste and have been detected in surface waters associated with feedlot areas and studies on both fish and mammals have demonstrated a strong androgenic effect of those metabolites. Zebrafish (Danio rerio) were exposed to environmentally relevant concentrations of the TbA metabolite 17beta-trenbolone from 0 to 60 days post-hatch (dph) and either sacrificed at 60 dph, transferred to clean water for 170 days or kept in exposure for 170 days. At 60 dph gonadal histology and vitellogenin analyses revealed all-male populations in groups exposed to 15.5 and 26.2 ng/L, and at 9.2 ng/L a skewed sex ratio towards males was observed. After the depuration period no sign of reversibility was observed. Environmentally relevant concentrations of 17beta-trenbolone cause a strong and irreversible masculinization of zebrafish and that raises concern about the effects of androgenic discharges in the aquatic environment. In addition this study also aids in understanding of the so far unknown sex determination process in zebrafish.

The Preservatives Ethyl‐, Propyl‐ and Butylparaben are Oestrogenic in an in vivo Fish Assay

The widely used phenolic preservatives ethylparaben, propylparaben, butylparaben and their common metabolite p-hydroxybenzoic acid were tested for their ability to evoke an oestrogenic response in vivo. Yolk protein induction in sexually immature rainbow trout was used as an oestrogen-specific endpoint after repeated injections of the compounds. All tested parabens were oestrogenic in doses between 100 and 300 mg/kg, while the metabolite showed no activity. Ethylparaben was found to be approximately sixty times weaker than propyl- and butylparaben which had oestrogenic potencies comparable to those previously found for bisphenol A.

In vivo estrogenic activity of branched and linear alkylphenols in rainbow trout (Oncorhynchus mykiss)

The in vivo estrogenic activity of the two branched alkylphenols, tert-octylphenol and technical nonylphenol, and the two linear isoforms, n-octylphenol and n-nonylphenol, was compared. The compounds were administered to juvenile rainbow trout (Oncorhynchus mykiss) by either intraperitoneal injection or water exposure and their estrogenic potential was evaluated by ELISA measurements of induced plasma vitellogenin. Intraperitoneal injections (50 mg/kg) of the two branched alkylphenols resulted in a significant vitellogenic response after 12 days whereas no significant induction was seen with the two linear isomers. Water exposure for 9 days to a nominal concentration of 150 micrograms/l of the alkylphenols elicited the same response pattern as seen for the injection experiment. Furthermore, in the present vitellogenin assay tert-octylphenol was giving a higher estrogenic response compared to technical nonylphenol using either of the two exposure routes.

The effect of 4-nonylphenol on the synthesis of vitellogenin in the flounder Platichthys flesus

Dose–response effects of the putative estrogenic compound 4-nonylphenol (4-NP) on the de novo synthesis of vitellogenin were investigated in the flounder Platichthys flesus. Male flounders received intraperitoneal injections during a period of 2 weeks with 10, 50, 100, 150 and 200 μg g−1 week−1 of 4-NP. Controls received the peanut oil vehicle only. The dose–response effects of the treatment were investigated on different plasma indicators of vitellogenin, alkali-labile protein phosphorous, total calcium and protein. Increased concentrations of these indicated a dose–response induced synthesis of vitellogenin in the liver of the 4-NP-treated male flounders. Vitellogenin from plasma of estradiol-treated male flounders was purified by gel filtration followed by ion-exchange chromatography and subsequently subjected to native gel electrophoresis. The purified vitellogenin fraction appeared as a high molecular protein band of 540 kDa in the electropherograms. The presence of vitellogenin in the plasma of the 4-NP-treated fish could then be directly detected by the native gel electrophoresis method. A high molecular weight vitellogenin band from plasma of the 4-NP-treated males appeared in the electropherograms in exactly the same position as the purified vitellogenin and vitellogenin in plasma of estradiol-treated male flounders. No vitellogenin band could be detected in the plasma of control males. The dose–response effect of the 4-NP-treatment was also reflected by an increased thickness of the vitellogenin band of the electropherograms by increasing doses of the 4-NP. De novo induction of vitellogenin synthesis in the liver was also indicated by increases of the hepatosomatic indices and of hepatic total RNA in the 4-NP-treated fish. No changes could be observed in the hepatic DNA concentrations. A significant increase of plasma GPT-concentrations indicated that 4-NP also elicited toxic effects on the fish. This was further indicated by a high mortality in the group treated with the highest dose of 4-NP (200 μg g−1). No effect could be observed on the gonadosomatic indices of the treated fish when compared to the peanut oil-injected controls. The present experiments were carried out towards the end of the reproductive season with testicular size observed to be near maximal in the control group. This is the most likely reason why no effect of the 4-NP injections could be observed on the testicular morphology or growth in the present experiments.

Effect of copper on ion- and osmoregulation in the shore crab Carcinus maenas

The uptake and effect of dissolved copper on regulation of hemolymph osmolality and Na+, K+, Cl-, Ca++, and Mg++ concentrations in the shore crab Carcinus maenas (L.) were determined at 400 mOsm (=14‰ S) ambient salinity. One mg Cu l-1 resulted in 50% mortality in 11 to 22 d; the highest sensitivity was observed around the moulting period. 0.25 and 0.5 mg Cu l-1 were not lethal after a onemonth exposure. Ten, 1, and 0.5 mg Cu l-1 altered regulation of osmolality, Na+, K+, and Cl- concentrations, while 0.25 mg Cu l-1 did not. Exposure to 1 mg Cu l-1 reduced hemolymph osmolality and Na+, K+, and Cl- concentrations to 80 to 90% of controls within 4 to 6 d and no further reduction was observed during a 21-d exposure. The effect of various copper concentrations on these four parameters were almost identical and the highest sensitivity was observed around the moulting period. Hemolymph calcium levels increased 20 to 80% in crabs exposed to ≧1 mg Cu l-1, while they decreased 20 to 30% in crabs exposed to ≦0.5 mg Cu l-1. Prolonged exposure to copper caused 20 to 70% reductions in hemolymph magnesium levels. Crabs exposed to 0.5 mg Cu l-1 for 29 d accumulated copper in hepatopancreas, gills, carapace, heart, testes, and hypodermis, but not in muscles and hemolymph. Increased copper levels in crabs exposed to 0.25 mg Cu l-1 were observed in hepatopancreas, gills, and carapace only. The present results suggest that effects of copper on ion regulatory processes in part explain why the toxicity of copper towards euryhaline invertebrates increases at low salinities.

Estrogenicity of xenobiotics in rainbow trout (Oncorhynchus mykiss) using in vivo synthesis of vitellogenin as a biomarker

Abstract The estrogenicity of several xenobiotics was evaluated using in vivo vitellogenin (Vtg) synthesis in immature rainbow trout as a biomarker. 17β-estradiol, DES and ethinyl estradiol were tested as positive controls. The xenobiotic compounds tested were technical nonylphenol, bisphenol A, butylbenzylphthalate (BBP) and dibutylphthalate (DBF). Measurements of the Vtg concentration was performed with a direct sandwich ELISA. 17β-estradiol, DES and ethinyl estradiol caused up to 100 000-fold increases in the Vtg-levels. Nonylphenol and bisphenol A had the highest estrogenic potency of the xenobiotics increasing the vitellogenin level approximately 300-fold while BBP was only weakly estrogenic, increasing the concentration about 3 times. DPB did not raise the vitellogenin contration above the detection limit.