Impairments of Endocrine Functions: Causes and Consequences - Ecological Biomarkers: Indicators of Ecotoxicological Effects - page 206

effects from infra- to supraindividual effects were observed. An increase in testosterone
was reported downstream of the discharges, and increases and decreases in 17β-estradiol
levels according to site. Reproduction was significantly impacted downstream from the dis-
charges with 60-70% decrease of embryos without shells after 3-4 weeks' exposure (Gust et
al. 2010a). Intersex is commonly observed in bivalves Scrobicularia plana living in European
estuaries. However, at this stage, no clear link may be established between intersex and
populational effects (Langston et al. 2007; Gomes et al. 2009; Fossi Tankoua et al. 2012).
8.8 Conclusions
Caliman and Gavrilescu (2009) and Vandenberg et al. (2012) have highlighted a number
of discrepancies between classic toxicology assumptions and the specificity of EDCs.
Contrary to dose-response curves showing that the greater the exposure to a chemical,
the greater the health effect, some EDCs exhibit effects only at dilute or large concen-
trations yielding U-shaped or J-shaped dose-response curves. Generally, a chemical can-
not have opposite effects, whereas EDCs such as dioxin exhibit an estrogen-like effect in
fetal organisms but an antiestrogen effect in adults. As also admitted for carcinogenic
compounds, most hormones in the bloodstream occur in dilute concentrations, and even
very dilute levels of endocrine-disrupting chemicals can prove dangerous, suggesting the
absence of any threshold effect.
Understanding the gap between the endocrine potencies of chemical mixtures such as
released in effluents and their real impact in the wild is an outcome that is highly desir-
able. Even if in vitro assays present some limitations in terms of estimation of the activities
of estrogens or androgens released into the aquatic environment, they are able to give a
good overview of the impact of effluent in certain cases, as demonstrated for a wild popu-
lation of gudgeon Gobio gobio (Nadzialek et al. 2010). According to these authors, it could
be interesting to include in vitro assays more in laboratory studies to establish direct cor-
relations, so the actual significance of estrogenicity screening for predicting in vivo effects
can be assessed. They conclude that an integrated picture of the estrogenicity of a given
compartment can only be achieved if chemical analysis, estrogenic activity screening, and
in vivo research in the field are combined.
Ankley et al. (2010) have defined an adverse outcome pathway (AOP) as “a conceptual con-
struct that portrays existing knowledge concerning the linkage between a direct molecular
initiating event (e.g., a molecular interaction between a xenobiotic and a specific biomol-
ecule) and an adverse outcome at a biological level of organization relevant to risk assess-
ment.” This concept is very relevant to describe the case of EDC effects (Figure 8.5). An AOP
involves cascading effects observed at different levels of biological organization. If the links
between these different levels are well established from a mechanistic point of view, or even
if they are based on strong correlations, population models may be used to formalize these
links, with an interesting potential for predictive risk assessment of chemicals in a regula-
tory context (Ankley et al. 2010; Galic et al. 2010). Despite effects of EDCs on sexual behavior
(courtship, territorial behavior) being well documented (Chapter 10), no clear links have
been established with population effects. Thus, it is necessary to study in depth the relation-
ships between reproductive behavior, reproductive success, and mechanisms controlling
population dynamics. In the case of EDCs, models have already been used to predict the
fate of fish populations (Miller and Ankley 2004; Brown et al. 2005; Miller et al. 2007).