Elsevier

Aquatic Toxicology

Volume 55, Issues 1–2, 1 November 2001, Pages 49-60
Aquatic Toxicology

Vitellogenin induction in painted turtle, Chrysemys picta, as a biomarker of exposure to environmental levels of estradiol

https://doi.org/10.1016/S0166-445X(01)00159-XGet rights and content

Abstract

Ponds within cattle farms often support turtle and fish populations and are impacted by manure runoff. Cattle excrete metabolized (glucuronide-conjugated) hormones in feces and urine into these ponds, and bacteria cleave the glucuronide metabolites to active steroids, which can be stable for several weeks in wastewater. The objectives of this study were to (1) assess levels of xenoestrogens found in ponds near livestock pastures; and (2) assess whether these levels of xenoestrogens induce vitellogenin (VTG) in painted turtles in the laboratory and field. We collected water twice, 6 weeks apart, and placed turtle traps weekly into two ponds, which receive runoff from beef cattle pastures, and into one pond with no cattle farm effluents. Water E2 levels were analyzed using C18 solid phase extraction disks and detected in a radioimmunoassay (RIA). Plasma was collected from painted turtles (Chrysemys picta) captured from these ponds and VTG levels were measured via enzyme linked immunosorbent assay (ELISA). Nine additional turtles were collected from a pond at the South Carolina Botanical Gardens, which receives no farm runoff, and were exposed in the laboratory to nominal concentrations of 0.15, 1.5, and 15 ng/l estradiol (static renewal) over a 28-day period, followed by 14 days in clean water. Plasma samples were taken weekly for VTG measurement via ELISA. Levels of free estradiol in the water column of farm ponds range from 0.05 to 1.80 ng/l, as measured by RIA, and up to 7.4 ng/l as measured by ER-β binding affinity. This is similar to what has been reported in streams receiving sewage treatment works (STW) effluents. In the laboratory, plasma VTG in male painted turtles could not be induced even at the high E2 dose (9.45 ng/l) after 28 days. In the field, VTG levels were induced only in females when compared with animals from the SC Botanical Gardens. Adult male turtles need to be primed with high doses of E2 prior to being able to respond to exogenous E2. Given that males would not typically be sensitized in the wild, environmentally relevant levels of E2 may not be sufficient to affect them. However, higher VTG levels in females could potentially change their reproductive fitness by altering egg size or by shifting energy allocations away from other survival needs. Long-term studies are needed to study potential impacts of VTG induction on female turtle reproductive success.

Introduction

Several wastewater treatment effluents have been implicated in endocrine disruption of fish using vitellogenin (VTG) as a biomarker of exposure (Harries et al., 1996, Harries et al., 1999, Jobling et al., 1998, Routledge et al., 1998). VTG is the precursor to egg yolk proteins in non-mammalian vertebrates and is synthesized in the liver and transported through the bloodstream to the ovaries, where it is taken up and cleaved into yolk proteins and incorporated into developing oocytes. Vitellogenesis is regulated by estrogen and, therefore, serves as an excellent model for studying estrogen-mimicking compounds in the environment.

Vitellogenesis has also been shown to be an estrogen specific and dose dependent response in turtles (Ho et al., 1981, Palmer and Palmer, 1995). Vitellogenesis is inducible in both sexes with administration of exogenous 17β-estradiol in the painted turtle (Chrysemys picta) (Ho et al., 1981). VTG in painted turtles is a relatively stable protein and appears to be composed of two polypeptides of equal size, which are not separable by electrophoresis, having a molecular weight between 210 and 220 kDa (Gapp et al., 1979). Palmer and Palmer (1995) showed that daily injections of estradiol at 1 μg/g for 7 days induced VTG in male red-eared slider turtles (Trachemys scripta).

Most studies investigating endocrine disruption in turtles have involved exposure in ovo to estrogenic contaminants such as hydroxylated PCB's, p,p′-DDE, and chlordane (Bergeron et al., 1994, Willingham and Crews, 1999, Willingham et al., 2000). Some studies on adults have also been done. Male Florida red-bellied turtles (Pseudemys rubiventris) collected from DDT contaminated Lake Apopka have a high incidence of ovotestes (Guillette et al., 1994). Increased incidence of deformities and unhatched common snapping turtle (Chelydra serpentina) eggs were found in Lakes Ontario and Erie associated with higher concentrations of PCB's, PCDF's, dibenzo-p-dioxins, organochlorine pesticides and their metabolites than those collected from control sites (Bishop et al., 1991). Although no single chemical could be implicated in this study, PCB's were strongly correlated with the deleterious effects (Bishop et al., 1991). Adult common snapping turtles having large body burdens of mixtures of organochlorines were found to have decreased sexual dimorphism (DeSolla et al., 1998). However, these animals showed no alterations in plasma E2 or T, when compared with those with lower body burdens, suggesting that the disruption occurred during sexual differentiation, not during adulthood (DeSolla et al., 1998).

Estrogenic compounds have been identified in human wastewater treatment effluents and surface waters at levels ranging from <0.2 to over 200 ng/l (Belfroid et al., 1999, Ternes et al., 1999a, Ternes et al., 1999b, Rodgers-Gray et al., 2000). Active estrogen compounds in surface waters receiving effluent from beef cattle farms have not been identified to date. Manure production in the US is estimated to be about 112 million tons each year (Copeland and Zinn, 1998). The presence of estrogenic compounds in ponds located on beef breeding farms is likely given the large numbers of pregnant cattle maintained at these facilities. Pregnant cattle excrete higher amounts of steroid hormone levels than non-pregnant female or male cattle (Mellin et al., 1966, Erb et al., 1968, Randel and Erb, 1971, Ferrell et al., 1983, Hoffmann et al., 1997). In many beef cattle operations, cattle are bred in April and calve in January. Studies describing the hormone excretion profile during pregnancy in cattle show a gradual increase in daily excreted estrogens to 0.0262 ng/mOsmol through the first 120 days, with an additional rise by 250 days gestation, peaking at 3.0475 ng/mOsmol through 2 days after parturition (Hoffmann et al., 1997, Randel and Erb, 1971). The majority of estrogens found in pregnant cattle urine are glucuronide or sulfate conjugated estrone, 17α- and 17β-estradiol. Free estrogens are excreted in feces, and show a similar increase with gestation, with 6.3 ng free E2/g feces per day at 120 days, up to 105.2 ng/g through parturition. As reference, cows can produce as much as 8 kg manure per 100 kg weight daily (Copeland and Zinn, 1998).

Desbrow et al. (1998) suggested that Escherichia coli in activated sludge were able to deconjugate human hormones, which are excreted as inactive glucuronide or sulfate conjugates in urine, producing active steroids. Bovine fecal steroids have also been shown to be stable for up to 3 weeks at 30°C, and up to 11 weeks at 5°C (Schlenker et al., 1999). Since steroid hormones are relatively stable compounds, and considering the daily input, they would be expected to persist in cattle farm ponds where resident wildlife would be exposed. Turtles can live in ponds impacted by manure, as was shown by Thomas et al. (1999) who studied turtles living in manure settling lagoons at Mississippi State University.

Painted turtles (Chrysemys picta) are common in South Carolina and their basic reproductive ecology and endocrinology, including the vitellogenic cycle, have been studied extensively (Callard et al., 1978, Gapp et al., 1979, Ho et al., 1981). Generally, the basal levels of circulating VTG in adult non-vitellogenic females (1-2 mg/ml) is present from December to February, followed by a peak in vitellogenesis resulting in maximum levels of 40 mg/ml VTG during two separate phases. The first phase is during the spring from March to June, and the second in the fall from October to November. Interestingly, the concentration of total plasma proteins is not altered, even during the vitellogenic periods, suggesting a possible reduction in hepatic synthesis of other circulating proteins. Corresponding plasma steroid hormone concentrations peak during the spring vitellogenic period, but not with the fall vitellogenic period (Gapp et al., 1979). This may be due to the absence of periovulatory progesterone peaks, or higher sensitivity of the liver to estrogen in the fall once the liver has been primed during the spring peak (Gapp et al., 1979).

The goals of our study were to (1) assess the levels of xenoestrogens in ponds receiving effluent from beef cattle farms, and (2) to test whether environmentally relevant concentrations of E2 are sufficient to induce the estrogen specific protein VTG in painted turtles, both in lab exposed and in field-caught turtles.

Section snippets

Study sites

The study sites were located on the Clemson University Simpson Agricultural Station, near Pendleton, SC (Fig. 1). The station contains several ponds receiving effluents from up to 320 beef cattle. Upper Pond (UP) receives effluent from a herd of beef cattle (approx. 100 adult cattle and their calves). This herd does not have direct access to the pond since they are retained by an electric fence 5 m directly upslope from the north and eastern edges of UP. Lower Pond (LP) is located downslope

Steroid extraction

Average recovery of 3H-17β-estradiol from filtered pond water matrix was 87.8%. Average recovery of a surrogate standard solution containing 100 pg E2 extracted from 250 ml well water was 86.7%. Intra and interassay variation for RIAs were 7.50 and 6.95%, respectively. For the parallelism studies, r2>0.98 for each assay. These results collectively demonstrate that the methods used met all criteria necessary to validate a radioimmunoassay for steroid hormones analyzed from our sample type.

Radioimmunoassay (RIA) on pond water extracts

E2

Discussion

Levels of estradiol quantified from cattle impacted ponds in this study fell within levels reported for sewage treatment plant effluents (Belfroid et al., 1999, Ternes et al., 1999a, Ternes et al., 1999b, Rodgers-Gray et al., 2000). The RIA methods used in this study were sensitive and have been validated for the extraction procedure used. The low levels of E2 detected at the BG pond may be excreted by the wildlife species present, such as resident ducks and other aquatic fauna (e.g. fish and

Conclusions

In conclusion, E2 levels in beef cattle farm ponds are within the range reported in streams receiving STW effluents, up to 1.80 ng/l via RIA and 7.4 ng/l via ER-β binding assays. We have shown that male and juvenile C. picta do not have elevated VTG under these levels of E2, either in the laboratory or in the field, but that females are affected with higher VTG plasma levels. These higher VTG levels may alter reproductive fitness of populations living in these ponds by changing energy

Acknowledgements

Funding was provided by a Clemson University Research Grant to E. Oberdorster. Field assistance provided by Jacqueline Brynda was very much appreciated. Thanks to Dr R. Brent Thomas and Dr Charles D. Rice for comments on earlier drafts of this manuscript, as well as technical assistance on our ELISAs by Dr Rice. We express our thanks to Dr Brent Palmer (University of Kentucky) for sharing his anti-turtle VTG antibody, Dr Kyle Selcer (Duquesne University) for screening several anti-turtle VTG

References (32)

  • F.R. Cagle

    A system for marking turtles for future identification

    Copeia

    (1939)
  • F.R. Cagle

    Observations on the life cycles of painted turtles (Genus Chrysemys)

    Am. Midl. Nat.

    (1954)
  • Copeland, C., Zinn, J., 1998. Animal waste management and the environment: background for current issues. Congressional...
  • C. Desbrow et al.

    Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening

    Environ. Sci. Technol.

    (1998)
  • S.R. DeSolla et al.

    Impact of organochlorine contamination on levels of sex hormones and external morphology of common snapping turtles (Chelydra serpentina serpentina) in Ontario, Canada

    Environ. Health Perspect.

    (1998)
  • C.L. Ferrell et al.

    Blood flow, steroid secretion and nutrient uptake of the gravid bovine uterus and fetus

    J. Anim. Sci.

    (1983)
  • Cited by (119)

    View all citing articles on Scopus
    View full text