Chang, Chapman, Travers, Tuttle

Effects of 17α-Ethinyl Estradiol on the Reproductive Success of Freshwater Fish

Katie Chang

Lucas Chapman

Jackie Travers

Alea Tuttle

Executive Summary

Over the last fifteen years, the scientific community has become increasingly aware that synthetic estrogens are present in sewage treatment effluent (Tenance, 2007). These synthetic estrogens are highly potent sexual hormones that have been connected to adverse effects on freshwater fish species, even in concentrations as low as partspertrillion (Hoffman et al., 2003). The common effects of synthetic estrogens on fish, specifically 17α-ethinylestradiol (hereafter referred to as EE2), are altered development of sex organs, production of egg yolk and proteins in males, loss of DNA protein integrity, decreased immune cell count, and a change in the ability of fish to break down other toxic pollutants. Other effects have also been connected to the presence of EE2 in freshwater fish populations, including the alteration of external color and adverse behavioral changes (Tenance, 2007). These adverse effects on freshwater fish can lead to the loss of reproductive success and the crash of entire populations (Kidd et al., 2007). However, it should be noted that the effects of EE2 can differ among fish species depending on the concentration, duration, and timing of exposure. It has also been shown that when synthetic estrogens are removed from aquatic environments, fish populations are able to rebound (CBC, 2008). Therefore, understanding the effects of EE2 on freshwater fish species is vital if we are to continue to support healthy fish populations for the purpose of ecosystem services, sport, and consumption.

The goal of this project is to investigate the changes in reproductive success as a result of exposure to EE2 on freshwater fish. This report consolidates some of the most recent and scientifically significant research on the topic. This paper begins with an explanation of EE2, how it breaks down in the environment, and its environmentally active concentrations. Safe levels (the lowest observable adverse effects level and the no observable adverse effects level) of EE2 in the environment are also addressed. Finally, this paper poses a brief discussion on alternative methods of birth control to EE2 based birth control pills and how they may reduce the amount of synthetic estrogen excreted into waste water effluent, while still serving as an effective method of birth control.

Problem statement

EE2, distributed in freshwater systems as a result of waste water effluent discharge, has negative effects on the reproductive success of freshwater fish.

Background

The most common source of EE2 is birth control pills (Johnson and Williams, 2004). The first birth control pill, Enovid, was introduced in the United States by Searle Pharmaceuticals in 1954, and was approved as a method of contraception by the United States Food and Drug Administration in 1960. Enovid was originally thought by scientists to contain solely synthetic progestin, however, synthetic estrogen that was also being manufactured by Searle Pharmaceuticals was mistakenly contaminating the Enovid pills. Once the synthetic estrogen contamination was discovered and removed from the pills, researchers determined that the estrogen component of the pill was what rendered it effective. Enovid first contained 10mg (1.0*107ng) of synthetic estrogen per pill (PBS, 2002). Synthetic estrogen concentrations in the variety of today’s more than 40 choices of birth control pills currently range from 2.0*104ng to 3.5*104ng per pill (USFDA, 2008).

Today, more than 18 million women in the United States are currently using birth control pills as a form of contraception (Johns Hopkins, 2000). This number does not include the millions of women and men in the United States that are also using EE2 as a form of hormone replacement therapy. In 1995, scientists became concerned that EE2 entering freshwater systems from waste water treatment plants could have adverse impacts on freshwater fish. It has since been determined that EE2 causes a variety of reproductive effects in both male and female freshwater fish, and Johnson and Williams concluded that steroid estrogens, predominately EE2, are the most potent endocrine disrupters in waste water effluent (Johnson and Williams, 2004).

Goal

The intention of this project was to research how EE2 released from waste water treatment effluent puts freshwater fish at risk for reproductive failure and possible population collapse.

Objectives

There are several components to this review: First, identify the potential for EE2 to reach compromising concentrations in freshwater bodies. Second, clearly define some assessment endpoints (the attributes associated with freshwater fish that is affected by exposure to synthetic estrogen). Note that the affected attribute may shift at differing exposure levels among different species, ranging from no effect to death. Third, identify the lowest observable adverse effects level (LOAEL) and the no observable adverse effects level (NOAEL) as defined in the literature available.

Approach

Information was collected for this project through literature, such as scientific journals, peer reviewed articles, and textbooks. Scientific databases were used when searching for relevant articles, including: Web of Science, Environment Complete, LexisNexis, Science Direct, and Wildlife & Ecology Studies Worldwide. Google Scholar was also used when full text articles could not be accessed. Search words included, but were not limited to: “synthetic estrogen”, “ethinyl estradiol”, “EE2”, “fish”, “reproduction”, “birth control”, “sewage”, and “waste water”. Appropriate articles were collected, read, and summarized for discussion.

The group worked both together and individually to achieve the objectives and ultimately reach the goal. The project was divided into two parts; three weeks to address the three main components of research and one week to compile the information and write the report. During each of the first three weeks, each member was responsible for researching the appropriate subject and providing the findings at group meetings. Rotations were taken so that each member had a chance to act as the group leader. Group leaders were responsible for organizing meeting times and locations, facilitating the meeting, and for taking notes (that were shared with the rest of the group). During each meeting, discussions revolved around sharing the progress of individual research and further, unanswered questions yet to be researched. In the final week, group members worked in compiling and reviewing work. All group members were also responsible for clear and frequent communication to the entire group on the progress of the project.

Findings

The following sections of this paper will discuss the sources of EE2, the pathway of EE2 in the environment, and the effects of EE2 on freshwater fish. Unfortunately, some authors found conflicting results and some made claims about “environmentally relevant concentrations” that may have been unfounded. These discrepancies may stem from the life history characteristics of varying species used in experimentation. NOAEL concentrations for one species may be above the LOAEL concentrations for a different species, thus results cannot be considered conclusive for all fish species.

Source of 17α-Ethinyl Estradiol and Exposure

Pharmaceuticals in the Human Body

A portion of EE2 is metabolized within the body and subsequently released as CO2. The rest of the compound is either excreted in the feces or secreted in the urine (Figure 1). During this process some EE2 is metabolized by the liver, a process through which a sulfate or glucuronide attaches to the molecule at the location of a hydroxyl (-OH) group. Alternatively, the molecule is oxidized (Zava et al., 1997; Johnson and Williams, 2004). The fraction of EE2 excreted from the body differs depending on the dose and whether the user is male or female (Mao et al., 2004). Johnson and Williams reported that, on average, 27% of ingested EE2 leaves the body in urine, and 30% in feces as either EE2 or as a glucuronide or sulphate attachment. The conjugate forms of EE2 secreted in bile are almost completely deconjugated by natural intestinal flora, so that the fraction of EE2 expelled by feces is primarily in its free (original) form (Johnson and Williams, 2004). Most of the EE2 excreted in urine is in the glucuronide form, according to Zava et al., who also reported that 19% of the daily EE2 dose was recovered in the urine using a radioactive tracker (Zava et al., 1997).

Figure 1. Metabolism of Ethinylestradiol in the Humans (Johnson and Williams, 2004)

Forms of Estrogen in Sewage Treatment Influent and Effluent

Freshwater fish exposed to EE2 are additionally exposed to a variety of natural estrogens found in waste water treatment plant effluent (WWTPE). EE2 is more resistant to biodegradation than natural estrogens; however, EE2 is susceptible to photodegradation. The half-life of EE2 as observed under laboratory conditions is about 10 days (Jürgens et al., 2002). Although, EE2 is the most potent endocrine disrupting estrogen, an assessment of individual estrogen compoundsmay result in an underestimation of risk, as was shown by Sumpter et al. (2006).

The concentration of EE2 in influent and effluent are highly dependent on the proportion of the population using birth control pills. The bulk of information in the literature is region specific; however Johnson and Williams provide a mathematical model for determining the daily concentration of estrogens entering a waste water treatment plant (WWTP) serving a given population (Johnson and Williams, 2004). This model considers the following factors:

  • the different sections of population excreting certain quantities of estrogen (such as women in different age groups, pregnant women, or women taking birth control),
  • the conjugation of estrogen in human metabolism,
  • the deconjugation and transformation of estrogens en route to the WWTP,
  • the dose of EE2 and proportion transformed in the body.

It was explained earlier that the primary form of estrogen in feces is deconjugated, so the model does not account for roughly thirty percent of estrogen as glucuronide or sulfate. It also assumes that EE2 is not transformed during transit to the WWTP due to its low potential for biodegradation. As was mentioned, the concentration of estrogen in WWTPE is highly variable and complex. EE2 has resistance to breakdown in primary treatment and activated sludge, but due to its hydrophobicity it tends to bind with sludge, colloids, and organic substances so that 80% may become passively retained in the treatment system. Performance results for WWTPs are inconsistent, ranging from failure to remove EE2 to 80-90% removal. However, data collected may not be accurate as the detection limits of the hardware used to measure EE2 are close to the suspected environmental levels (Sumpter et al., 2001). The water systems at highest risk are located downstream from a WWTP that serve a densely populated area and/or one that has outdated or inadequate treatment methods. (Sumpter et al., 2006)

One example, is a highly polluted system is that of Taipei, Taiwan, where the average surface water concentration of EE2 in the Dan-Shui River as measured by Chen et al. (2007) was 15ng/L, with treatment effluent observed to be as high as 26ng/L. In this case, the population density of Taipei, Taiwan is 9649 persons/km2 and 47% of total discharge from the watershed is untreated municipal and livestock waste water. On the other hand, Desbrow et al. (1998) reported that EE2 was largely below detection limit in a survey of 7 WWTPEs discharging into British rivers. EE2 was detected in only three of seven cases, where 0.2-7ng/L effluent concentrations were discovered.

Environmental Levels

Estrogens, both synthetic and natural, are not currently targeted during the waste water treatment process. Studies have shown that environmental concentrations of estrogens are typically in the low parts per trillion (5ng/L or 5pptr). However, merely measuring concentrations of these endocrine disrupting compounds (EDCs) is insufficient to express their true toxicological potency. The environmental concentrations of synthetic estrogens, when expressed as E2 (natural estrogen) equivalent concentrations (i.e. the summation of concentrations of individual compounds after adjusting for the estrogenic potency relative to E2), can be thought of as being 17 and 147ng E2/L in effluent and surface waters respectively (Kidd et al., 2007).

Sumpter et al. (2006) have modeled mixture effects and risks associated EDCs present at a watershed scale. Utilizing the GREAT-ER hydrological model, they have developed a method to predict environmental concentrations, evaluate the risk posed to certain populations within a watershed, and account for changes in risk that may occur as a result of different environmental conditions, including different EDC mixtures.

Responses to EE2 exposure

Acute vs. Chronic Effects

To understand the differences in acute and chronic exposure to EE2, studies involving partial and full life-cycle exposures have been reviewed. This is especially important since some fish species grow from egg to mature adult in a matter of months, while others take years to fully reach maturity. Therefore it is imperative to perform a variety of endpoint tests (for instance growth and reproductive characteristics) for different lengths of time and different phases in the life cycle of the fish.

One study by Schäfers et al.(2007), addressed these concerns of timing and duration of exposure very effectively. Two generations of zebrafish were observed when exposed to four concentrations of EE2 (0.05, 0.28, 1.7, and 10ng/L) over varying durations of time (partial life-cycle, full life-cycle, and two-generation exposure). Schäfers et al. found that the duration of exposure seemed to be crucial for the ability of the zebrafish to recover from an exposure. For example, as shown in Figure 2 and 3, reproduction was completely unsuccessful in zebrafish exposed for a full life cycle at a concentration of 9.3ng/L, but those exposed to a partial life cycle of the same concentration were still able to recover, once EE2 exposure was discontinued. (Schäfers et al., 2007)

Figure 2. Exposure of first generation of fish to EE2 until 75 days postfertilization (partial life-cycle). Fertility and fecundity are in mean percent success. (Schäfers et al., 2007)

Figure 3. Exposure of first generation fish to EE2 to maturity (full life-cycle). (Schäfers et al., 2007)

Nash et al. presented similar findings in that exposure of zebrafish to high concentrations of EE2,which was found to be acutely toxic. After a laboratory exposure of zebrafish to an EE2 concentration of 50ng/L, Nash et al. observed a time-related reduction in egg production and egg viability. After 10 days of exposure to 50ng/L EE2, there was complete reproductive failure among the zebrafish population. In addition to reproductive failure, the zebrafish experienced 35% mortality and all surviving fish exhibited a wide range of adverse health effects. (Nash et al., 2004)

Nash et al. performed another experiment where it was found that not only duration, but also the timing of EE2 exposure in the life cycle influences reproductive success. Two generations of zebrafish were exposed to EE2 concentrations of 0.5 and 5.0ng/L 14 days after their egg production life-stage began for a total of 40 days. In the original generation there were no observed effects on egg production and egg viability as a result of the concentrations applied. (Nash et al., 2004)

When the offspring of this original generation were exposed to the same concentrations, the results varied greatly. The offspring of the original zebrafish population were collected at two different times; once at 15 days post initial EE2 exposure, and again at 40 days post initial EE2 exposure. The offspring collected at 15 days after their parents’ initial EE2 exposure were reared to 100 hours post fertilization, and the offspring collected at 40 days after their parents’ initial EE2 exposure were reared for their entire life-span (310 days). In both groups of offspring there were observable adverse reproductive effects caused by the exposure to 0.5ng/L EE2. There were no observed reductions in egg production, but the number of nonviable eggs produced by these two groups was more than twice that of the control (Nash et al., 2004). When offspring collected at 15 days and 40 days after their parents’ initial exposure to EE2 were exposed to 5.0ng/L, the zebrafish collected at 15 days exhibited a decreases in egg production that ranging between 42-45% lower than the control’s total egg production. The offspring collected at 40 days that underwent life-long exposure to 5.0ng/L exhibited complete reproductive failure. Nashet al., found not one viable egg produced by the more than 12,000 fish exposed to 5.0ng/L EE2 in this generation after 310 days of 5.0ng/L EE2 exposure (Nash et al., 2004). Based on these results, exposure to a low, environmentally relevant concentration of EE2 does have chronic effects on the reproductive success of zebrafish.

In the aforementioned experiment by C. Schäfers et al., the LOAEL (LOEC) was calculated for both the first and second generations of zebrafish. In the first generation, the lowest observed effect concentrations (LOEC) that influenced the growth, maturity, and reproduction was 1.1ng/L. The LOEC for the second zebrafish generation was calculated to be 2ng/L. When these two values were compared, calculations showed that the LOEC reduced fecundity and fertility by 23% and 45% (respectively) in the first generation and by 83% and 98% in the second generation (respectively). There was also a 12 day and 47 day delay in spawning for first and second generation respectively (Schäfers et al., 2007). Similar results for LOEC were found by J.L. Parrott et al. in their study with fathead minnows. Male fish exposed to concentrations between 0.32 and 23ng/L were found to be negatively affected by concentrations of EE2 that are similar to or lower than those found in most waste water discharge. Also, responses were seen at concentrations of 1ng/L of EE2, which are environmentally relevant. (Parrott et al., 2004)

It should be noted that LOEC and NOEC -- in the context of EE2 exposure -- are difficult to assign. This is due to the fact that these levels as observed by tests on specific fish species,and thus cannot be applied to other species.

Feminization of Fish

There are many studies ofEDCs that document the feminization of male fish when exposed to EE2. Parrott et al. experimented on fathead minnows and noted that emasculinization occurred atan EE2 exposure as low as 0.96ng/L. A decrease in male sex characteristics was also found. Exposure to high concentrations negatively impacted females and males, shown by a reduced gonadosomatic index (the ratio of a fishes eggs/sperm to its body weight) at 150 days post hatch in fish exposed to >3.5ng/L EE2. (Parrott et al., 2004)

A common misconception is that reproductive failure is due to a gender imbalance caused by EE2 exposure during egg development (with predominantly female populations). Some studies have shown these skewed sex ratios in EE2 exposed fish, however, Nash et al. believes that this is false data. They suggest that this reported skew is due to testicular tissue that appears as though it is ovarian tissue. They also suggest that this mistake can be avoided by measuring vitellogenin (VTG) concentrations in blood samples from affected fish. The presence of VTG concentrations in fish blood is a way to determine gender of a fish. VTG is a yolk protein that is expressed by female fish and dormant in male fish. Concentrations of VTG in female fish are normally high and normally undetectable in male fish. The presence of VTG in the blood of male fish is often used as a biomarker of endocrine disruption, particularly to exposure of EE2. As seen in Figure 2, true females have 1,092 ± 106μg VTG/mL, while intersex male concentrations are observed at 1.8 ± 1.1μg VTG/mL. (Nash et al., 2004)