The multigenerational effects of water contamination and endocrine disrupting chemicals on the fitness of Drosophila melanogaster

Abstract Water pollution due to human activities produces sedimentation, excessive nutrients, and toxic chemicals, and this, in turn, has an effect on the normal endocrine functioning of living beings. Overall, water pollution may affect some components of the fitness of organisms (e.g., developmental time and fertility). Some toxic compounds found in polluted waters are known as endocrine disruptors (ED), and among these are nonhalogenated phenolic chemicals such as bisphenol A and nonylphenol. To evaluate the effect of nonhalogenated phenolic chemicals on the endocrine system, we subjected two generations (F0 and F1) of Drosophila melanogaster to different concentrations of ED. Specifically, treatments involved wastewater, which had the highest level of ED (bisphenol A and nonylphenol) and treated wastewater from a constructed Heliconia psittacorum wetland with horizontal subsurface water flow (He); the treated wastewater was the treatment with the lowest level of ED. We evaluated the development time from egg to pupa and from pupa to adult as well as fertility. The results show that for individuals exposed to treated wastewater, the developmental time from egg to pupae was shorter in individuals of the F1 generation than in the F0 generation. Additionally, the time from pupae to adult was longer for flies growing in the H. psittacorum treated wastewater. Furthermore, fertility was lower in the F1 generation than in the F0 generation. Although different concentrations of bisphenol A and nonylphenol had no significant effect on the components of fitness of D. melanogaster (developmental time and fertility), there was a trend across generations, likely as a result of selection imposed on the flies. It is possible that the flies developed different strategies to avoid the effects of the various environmental stressors.

nonhalogenated phenolic chemicals such as bisphenol A and nonylphenol. To evaluate the effect of nonhalogenated phenolic chemicals on the endocrine system, we subjected two generations (F0 and F1) of Drosophila melanogaster to different concentrations of ED. Specifically, treatments involved wastewater, which had the highest level of ED (bisphenol A and nonylphenol) and treated wastewater from a constructed Heliconia psittacorum wetland with horizontal subsurface water flow (He); the treated wastewater was the treatment with the lowest level of ED. We evaluated the development time from egg to pupa and from pupa to adult as well as fertility. The results show that for individuals exposed to treated wastewater, the developmental time from egg to pupae was shorter in individuals of the F1 generation than in the F0 generation. Additionally, the time from pupae to adult was longer for flies growing in the H. psittacorum treated wastewater. Furthermore, fertility was lower in the F1 generation than in the F0 generation. Although different concentrations of bisphenol A and nonylphenol had no significant effect on the components of fitness of D. melanogaster (developmental time and fertility), there was a trend across generations, likely as a result of selection imposed on the flies. It is possible that the flies developed different strategies to avoid the effects of the various environmental stressors.

K E Y W O R D S
contamination, development, Drosophila melanogaster, endocrine disrupting chemicals, fertility, fitness

| INTRODUCTION
Human-mediated land-use change, species introductions, pollution, and ultimately climate change have caused an almost irreversible deterioration of natural environments (Dawson, Jackson, House, Prentice, & Mace, 2011;Hunter, 2007;Pereira et al., 2010). In particular, human alteration of natural environments has generated a massive amount of water pollution, which may contain a large amount of sediments, excess nutrients, and toxic chemicals (Kareiva & Marvier, 2011). Among the many toxic and carcinogenic pollutants, endocrine disruptors (ED) are known to affect the physiological pathways associated with reproduction and development of animals and plants (Guillette, 2006;Liao, Yen, & Wang, 2009;Qiu, Wang, & Zhou, 2013;Wang et al., 2015).
The molecular structure of ED allows them to behave like hormones or hormone precursors (Guillette & Gunderson, 2001;Lee & Choi, 2007). As such, these disruptors have been shown to affect not only different reproductive stages but also different developmental stages of vertebrates (Guillette & Edwards, 2005;Segner et al., 2003). For example, it has been shown that ED can cause infertility due to malformations of the reproductive tract (Guillette, 2006), impaired release of pheromones (Anway & Skinner, 2008;Colborn, Vom Saal, & Soto, 1993;Guillette & Edwards, 2005;Segner et al., 2003), premature release of progesterone and progestin and occasional premature birth (Guillette, 2006;Longnecker, Klebanoff, Zhou, & Brock, 2001), changes in the development of oocytes, thyroid abnormalities and changes during molting (Depledge & Billinghurst, 1999;Ha & Choi, 2008;Hutchinson, 2002;Planelló, Martínez-Guitarte, & Morcillo, 2008;Rodríguez, Medesani, & Fingerman, 2007), and impaired immune functioning (Martineau et al., 1988). Experiments involving ED are complicated due to the fact that EDs often naturally come from multiple sources and can generate additive effects. In particular, we used developmental time (from egg to pupae and from pupae to adult) and fertility as components of fitness, and bisphenol A (BPA) and nonylphenol (NP) were used simultaneously as ED. We obtained different concentrations of both BPA and NP by passing wastewater through different Horizontal Subsurface Flow Constructed systems. These ED were chosen for the experiments because they have characteristics (e.g., persistent, lipophilic, low vapor pressures) which facilitate dispersion in liquid environments; as such, BPA and NP have been known to contaminate groundwater reservoirs (Colborn et al., 1993;Kareiva & Marvier, 2011;Theis & Tomkin, 2012).
Although several studies have evaluated the effects of ED on animal fitness, most were focused on understanding the consequences of one endocrine disruptor at a time, and very few have looked at the combined effects of multiple ED on a given system. This is important as multiple ED are often found together in natural environments as a result of various sources of anthropogenic water pollution (Kareiva & Marvier, 2011;Theis & Tomkin, 2012). In addition, it is unknown whether organisms can evolve as a response to ED present in their environment. Given the complexity of treating contaminated waters together with the long term persistence of ED, changes in development and reproduction time might likely be the only solution for organisms that face permanent (and potentially increased) water contamination. Therefore, in this work we evaluated the impact that different concentrations of ED have on Drosophila development and reproduction.  (Zhi & Ji, 2012). Water from this treatment had a BPA concentration of 1.58 ± 0.85 μg/L (mean ± SD) and a NP concentration of 735.9 ± 283.6 μg/L (mean ± SD) which is 76% less BPA and 25.3% less NP than that of wastewater. (4) Treated wastewater from a gravel HSSF-CW that had a BPA concentration 2.64 ± 0.65 μg/L (mean ± SD) and a NP concentration of 1252.5 ± 540.62 μg/L (mean ± SD) which is 70% less BPA and 25% less NP than that of WW. (5) Distilled water (without EDCs and non-polluted sources).

| Developmental time
The 20 pairs of flies were left to reproduce in 10 bottles containing standard banana medium. After 4 days, adults were discarded and all the eggs from all the flies were collected and mixed. We randomly selected 800 eggs that were transferred to 40 bottles (Figure 2), which is eight bottles per experimental treatment, each with 20 eggs. All bottles were kept at 25 ± 3°C with a natural 12L:12D photoperiod.
Developmental time was estimated as the time in days it takes an egg to become pupae and the time it takes a pupae to become adult. Every day the number of pupae in each bottle was counted and, to prevent counting the same individual twice, a mark on the bottle was made. The number of emerging adults was counted on a daily basis until no more adults emerged. The pair that emerged first from each bottle and treatment was used for the fertility experiment.
The remaining flies from each experimental treatment were allowed to reproduce in 10 bottles with the corresponding experimental medium ( Figure 2). After 4 days, all eggs (i.e., F1 generation) from each treatment (160 each treatment) were collected and mixed in order to maintain randomness in each treatment. These were then transferred to eight new bottles, and developmental times were estimated as before.

| Fertility
The first emerging adult couple from each bottle from each treatment was allowed to reproduce in a new bottle with the corresponding experimental medium. After 9 days, the couple was discarded and the number of adults was counted until no more adults emerged. This was repeated for the F1 generation.

| Statistical analyses
Two-way ANOVAs were used to evaluate how experimental treat-

| DISCUSSION
Water pollution as a result of anthropogenic activities (e.g., sediment, excess nutrients, and toxic chemicals) can negatively alter freshwater ecosystems. Endocrine disruptors (ED) are known to interfere in different ways with the normal synthesis, secretion, transport, and action of hormones (Guillette, 2006;Segner et al., 2003). Furthermore, some studies have indicated that the biodiversity of freshwater ecosystems is particularly vulnerable to ED (Cardinale et al., 2012;Kareiva & Marvier, 2011). Here, we evaluated the impact that dif- Although several studies have shown that developmental time in Drosophila species is affected by environmental stress (Castañeda & Nespolo, 2013;Sørensen & Loeschcke, 2004), there is no clear pattern regarding the effect of ED. While some studies have shown that developmental time increases from pupae to adult when flies are exposed to ED (Akins, Schroeder, Brower, & Aposhian, 1992;Cohn, Widzowski, & Cory-Slechta, 1992;Liu, Li, Prasifka, Jurenka, & Bonning, 2008), other studies report that developmental time from pupae to adult decreases (Atli, 2010;Memmi & Atlı, 2009). Our results do not provide any new insight regarding the consequences of exposure to ED on components of fitness measured here: Developmental time was not significantly influenced by wastewater. Furthermore, there was no consistent pattern (i.e., in treatment and/or generation) between developmental time from egg to pupae and developmental T A B L E 1 Summary of the statistical analyses evaluating differences in Drosophila melanogaster development and reproduction. Individuals were exposed to one of five experimental treatments over two generations (F0 and F1 shown that exposure to ED in invertebrates causes overall reduced growth rates and thus, overall increases in developmental time (Hill et al., 2002;Izumi, Yanagibori, Shigeno, & Sajiki, 2007;Marcial, Hagiwara, & Snell, 2003), and yet other studies have shown the opposite pattern increased growth rates and decreased development time (Weiner et al., 2014).
Wastewater contains other estrogenic EDs such as ethinelestradiol (EE2) and phthalate esters (e.g., dibutyl phthalate). The effect of these EDs could also affect the generation time of D. melanogaster (Memmi & Atlı, 2009) (Nunney, 1996;Roff, 2001). In Drosophila, selection for faster developmental rates, which is what we have indirectly carried out in order to obtain the F1 generation, usually results in F I G U R E 3 Response of Drosophila melanogaster exposed to five experimental treatments over two generations (F0 generation with black dots and F1 with gray dots). (a) Average (± SE) developmental time from egg to pupae for both generations. (b) Average (± SE) developmental time from pupae to adult for both generations. (c) Fertility is represented by the average (± SE) number of offspring produced during the first 9 days in each generation. The experimental treatments are shown on the x-axis, DW: Distilled water, He: Treated wastewater from a HSSF-CW with Heliconia psittacorum, Ph: Treated wastewater from a HSSF-CW with Phragmites australis, WG: Treated wastewater from a gravel HSSF-CW, WW: Wastewater the evolution of smaller body size as adults (Chippendale & Sorenson, 1997;Chippindale, Chu, & Rose, 1996). As body size is positively related to fertility (Roff, 2001), individuals that develop faster are smaller and thus have less offspring.
In conclusion, it is well-known that anthropogenic effects on ecosystems, such as contamination of water with toxic chemicals, such as EDs, have an effect on the fitness of individuals (Anway & Skinner, 2008;Colborn et al., 1993;Depledge & Billinghurst, 1999;Weiner et al., 2014). Despite this, the present study did not find a significant effect of EDs (i.e., BPA and NP) on the developmental time or fertility of two generations of D. melanogaster. Overall, more studies are needed to evaluate the generality of this finding. Studies of morphological variation and studies involving many generations will help to determine how D. melanogaster is affected by EDs.