Evaluation of in vitro assays for determination of estrogenic activity in the environment

Evaluation of in vitro assays for determination of estrogenic activity in the environment

1 Summary (English)

Recent studies in a number of countries have shown that the aquatic environment can possess estrogenic activity capable of influencing the fauna. (Xeno)estrogens are believed to reach the aquatic environment mainly by means of municipal and industrial sewage outfalls. However, agricultural drainage may also be a route for (xeno)estrogens to enter the aquatic system. Numerous natural and anthropogenic substances are known to exhibit estrogenic activity. In the aquatic environment, estrogenic activity has primarily been ascribed to the natural steroids, 17b-estradiol (E2), estrone (E1) and estriol (E3), and the synthetic estrogen, ethinylestradiol (EE2), used in contraceptives. To a lesser extent xenoestrogenic chemicals, such as alkylphenols and bisphenol A, may also contribute to the estrogenic activity in the aquatic environment.
In vitro assays measure the total estrogenic activity of an environmental water sample, regardless of which compounds are responsible for the activity. The total estrogenic activity in the sample is then compared to the activity of the natural estrogen, E2, and expressed as estradiol equivalents (EEQ). A number of studies employing in vitro assays have demonstrated the estrogenic activity of wastewater and surface water in various countries. Total estrogenic activity (expressed as EEQ values) of sewage treatment plant influents have been reported to be 0.6-153 nanograms per litre. In the effluents, EEQ values are usually below 25 nanograms per litre, although values of up to about 150 nanograms per litre have been reported in the USA. In surface water, the EEQ values found are generally from below 1 nanogram up to 15 nanograms per litre, although values of up to about 80 nanograms per litre have been reported in one study. The EEQ levels found in some aquatic systems are sufficient to cause estrogenic effects in fish in laboratory experiments.

Several in vitro assays have been developed to assess the estrogenic activity of single compounds or complex mixtures. Each assay measures different end points at different levels of biological complexity of estrogen action. Most assays fall into one of three categories: 1) estrogen receptor (ER) competitive ligand binding assays that measure the binding affinity of a chemical for the ER; 2) cell proliferation assays that measure the increase in cell number of estrogen sensitive cells (E-screen); and 3) reporter gene assays that measure ER binding-dependent transcriptional and translational activity. No single in vitro assay can be regarded as ideal for assessing the estrogenic activity of wastewater and surface water. They all have their advantages and limitations.

Most ER binding assays quantifies the ability of a test compound to compete with radiolabelled E2 for binding to the ER. The sample is added along with an excess of radiolabelled E2 to isolated ERs whereupon the amount of unbound radioactivity is measured. ER binding assays are fast. However, they are significantly less sensitive than the other in vitro assays. In addition, binding assays are not easily amenable to automation, thereby limiting their utility as a screening tool. Furthermore, ER binding assays require specialised laboratory facilities because of the radioactive substances. Finally, the binding of a substance to the ER is only indicative that it may act as a xenoestrogen; ER binding may be a poor predictor of more complex in vitro and in vivo responses.

In the E-screen assay, proliferation of human MCF-7 breast cancer cells as a response to estrogen is measured. The E-screen is based on the following three premises: (i) factors in serum added to the medium inhibit the proliferation of MCF-7 cells, (ii) estrogens induce cell proliferation by negating this inhibitory effect, and (iii) non-estrogenic substances do not neutralize the inhibitory signal present in serum. However, it has been shown that the E-screen may not be as estrogen specific as assumed, since a range of non-estrogenic substances has been found to influence the proliferation of MCF-7 cells, at least in some cell lines. In addition, considerable inter-laboratory variability has been observed in test results from the E-screen. Furthermore, the E-screen is more time consuming than the other assays and is thus considered impractical for extensive monitoring studies.

Reporter gene assays are based on the ability of a compound to stimulate ER-dependent transcriptional activity. Reporter gene assays are carried out with genetically engineered human cancer cells or yeast cells tranfected with estrogen response elements (ERE) linked to a reporter gene. In the human-based reporter gene assays (ER-CALUX, MVLN cell assay and chimeric receptor/reporter gene assays) the reporter gene codes for luciferase and in the yeast-based reporter gene assay (YES) the reporter gene codes for b-galactosidase. Yeast cells are further transfected with the DNA sequence for the human ER, since yeast cells do not possess endogenous ER.

In reporter gene assays, the sample is added to the transfected cells. Estrogenic substances that enter the cells binds to the ER, which becomes activated and binds to the EREs. This biding initiates the expression of the reporter gene and thereby the synthesis of the enzyme. An appropriate substrate in the incubation medium is metabolized by the newly synthesized enzyme, resulting in the production of an easily detected product. The mammalian-based reporter gene assays have the major drawback, compared to the yeast-based assay, that mammalian cells are more difficult and expensive to cultivate, and are more susceptible to cytotoxic effects. The simplicity of the YES assay is a distinct advantage, as the product of the reporter gene is secreted in the medium and no cell lysis is required. In comparing the YES assay with the mammalian-based reporter gene assays, however, differences in responses to (xeno)estrogens and anti-estrogens are evident. Firstly, a difference in the sensitivity is observed between the two mammalian-based endogenous receptor/reporter gene assays (ER-CALUX and MVLN cell assay) and the YES assay, demonstrating that the former can detect (xeno)estrogens at lower concentrations. Secondly, a difference in response to anti-estrogens is found between the mammalian-based reporter gene assays and the YES assay , as the latter does not consistently detect anti-estrogenic activity, but sometimes identifies it as agonistic. This “limitation”, which the YES assay has in common with ER binding assays, could be considered an advantage if all one is interested in is detecting compounds that interacts with the ER and elicit a response, thus having potential endocrine disrupting effects. From this point of view, the mammalian-based reporter gene assays may actually underestimate the actual estrogenic potential of a complex water sample.
A main problem in the utilization of in vitro assays to analyse aquatic environmental samples is the presence of inhibitory/cytotoxic compounds. Yeast assays may perform better for monitoring of environmental samples, as these samples are frequently contaminated with substances other than (xeno)estrogens interfering with the growth and viability of animal cells, but not with yeast cells.

Reporter gene assays seem to be a suitable choice for monitoring environmental matrices for estrogenic activity. The final choice of which reporter gene assay to employ (mammalian-based or yeast-based) depends on the importance of lower detection limit versus the importance of ease of use and lower costs.

Significant advantages of in vitro assays over chemical analyses are that no unknown components with estrogenic activity are overlooked and that any combination effects are taken into account in the analysis. Chemical analysis of all compounds with potential estrogenic activity would be very costly and unknown estrogenic compounds, including metabolites, may still be present in environmental matrices. By a combination of the two types of analysis it is possible both to assess the estrogenic activity in a sample and to (partly) identify and quantify the compounds responsible for the estrogenic activity.

The advantages of in vitro assays over in vivo assays include lower costs and time consumption as well as sparing of experimental animals. However, in vitro assays do not always reliably predict the results of in vivo assays and should not be used alone for full assessment of potential estrogenic hazards in the aquatic system. In vitro assays usually possess minimal metabolic capabilities. As a result, extrapolation from in vitro to in vivo systems can lead to false negatives for compounds that are bioactivated, and overestimates of potency for compounds readily degraded in vivo. In addition, bioavailability, cross talk between biological pathways and the complex processes of uptake, binding to carrier proteins, transport, targeting, disposition and excretion of compounds in whole animals are not taken into account in the in vitro assays. Furthermore, it should be kept in mind that there are estrogenic effects that are based on mechanisms different from receptor binding, e.g. interferences with hormone synthesis and metabolism. Environmental samples should therefore also be tested for their estrogenic activity in relevant in vivo tests, such as vitellogenin induction or gonadal effects in fish.