Article
pubs.acs.org/est
Additional Treatment of Wastewater Reduces Endocrine Disruption
in Wild FishA Comparative Study of Tertiary and Advanced Treatments
Alice Baynes,*,† Christopher Green,† Elizabeth Nicol,† Nicola Beresford,† Rakesh Kanda,‡ Alan Henshaw,§
John Churchley,∥ and Susan Jobling†
†
Institute for the Environment, Brunel University, Uxbridge, Middlesex, UB8 3PH, U.K.
Severn Trent Services, Britten Road, Reading, Berkshire, RG2 0AU, U.K.
§
Environment Agency National Coarse Fish Farm, Moor Lane, Calverton, Nottingham NG14 6F2, U.K.
∥
WatStech, The Technology Centre, Wolverhampton Science Park, Wolverhampton, WV10 9RU, U.K.
‡
S Supporting Information
*
ABSTRACT: Steroid estrogens are thought to be the major cause of feminization
(intersex) in wild fish. Widely used wastewater treatment technologies are not effective at
removing these contaminants to concentrations thought to be required to protect aquatic
wildlife. A number of advanced treatment processes have been proposed to reduce the
concentrations of estrogens entering the environment. Before investment is made in such
processes, it is imperative that we compare their efficacy in terms of removal of steroid
estrogens and their feminizing effects with other treatment options. This study assessed
both steroid removal and intersex induction in adult and early life stage fish (roach,
Rutilus rutilus). Roach were exposed directly to either secondary (activated sludge process
(ASP)), tertiary (sand filtrated (SF)), or advanced (chlorine dioxide (ClO2), granular
activated charcoal (GAC)) treated effluents for six months. Surprisingly, both the
advanced GAC and tertiary SF treatments (but not the ClO2 treatment) significantly
removed the intersex induction associated with the ASP effluent; this was not predicted by the steroid estrogen measurements,
which were higher in the tertiary SF than either the GAC or the ClO2. Therefore our study highlights the importance of using
both biological and chemical analysis when assessing new treatment technologies.
■
INTRODUCTION
Estrogenic chemicals in effluents discharged from wastewater
treatment works (WwTWs) are thought to be the major cause
of feminization observed in wild fish populations worldwide
(e.g., refs 1−3). A wealth of evidence for the occurrence of an
intersex condition, those fish with both male and female
gametes and/or mismatched gametes and reproductive ducts,
has been gathered for the gonochoristic (single sex) cyprinid
species Rutilus rutilus (roach) inhabiting rivers contaminated
with treated sewage effluent in the U.K.4,5 In severely intersex
roach, male gamete production, milt (semen) volume, and milt
density are reduced.6,7 In vitro analysis has shown wild intersex
fish also have lower sperm counts, motility, and fertilization
success when compared to nonfeminized fish,7 leading to the
hypothesis that reproduction might be compromised in these
individuals. Recently, this has been demonstrated in competitive breeding experiments showing intersex severity is negatively correlated with reproductive success.8
A growing number of studies have shown that both natural
steroid estrogens, 17β-estradiol (E2) and estrone (E1), and the
potent synthetic estrogen, 17α-ethinylestradiol (EE2), can
feminize male fish from a range of species, including Oryzias
latipes (medaka; e.g., 9) Pimephales promelas (fathead minnow;
e.g.,10), and Danio rerio (zebrafish; e.g., 11) at concentrations
similar to those measured in WwTW effluents (i.e., low ng/L).
© 2012 American Chemical Society
Many steroids, such as estrogens, enter the wastewater system
(via urine) as water-soluble glucuronide or sulfate conjugates12;
these conjugates are less biologically active (compared to “free”
steroids). However, microbial activity in the sewers and in the
WwTW themselves cause deconjugation, and therefore release
active free estrogens into the wastewater system (e.g., 13).
Fortunately, biological activities within WwTW break down the
estrogens further to varying degrees, depending on the type of
treatment and their operating efficiency.
A number of different WwTW systems are employed
depending on the region, location, and/or population served.
The majority have primary treatment involving settlement and
screening of the raw sewage, however, this treatment only
removes a small percentage of the steroid estrogens (e.g., 14).
The secondary treatments, such as biological filters (BFs) and
activated sludge process (ASP), are far more effective at removing natural steroid estrogens. Nitrifying ASP is considered to be
the most effective of the secondary treatments at removing
steroid estrogens,15 with removal rates as high as 97.8% for
E116 and 97% for E2.17 Whereas BFs have removal rates of
Received:
Revised:
Accepted:
Published:
5565
December 21, 2011
April 11, 2012
April 13, 2012
April 13, 2012
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
∼80% and are less effective at removing E1 than E2.17,18 Both
ASP and BF have much lower removal rates of EE2, which,
although present at much lower concentrations, has higher
estrogenic activity in fish than the natural steroid estrogens.16,18,19 Some WwTW then have additional tertiary
treatment such as sand filters or reed-beds that can provide
further steroid removal.17,20 However, it is now well-known that
low biologically active concentrations of estrogens do continue
to be measured in effluents at the point of discharge. The
Environment Agency (EA) of England and Wales has suggested
a tentative predicted no-effect concentration (PNEC) of 1 ng/L
total estrogens to protect aquatic wildlife21 and more recently
Caldwell et al.22 have calculated a PNEC of 0.35 ng/L EE2.
Exposure threshold values/PNEC of 0.1 ng/L EE2 have been
used for risk assessment in Australia and Norway23,24 and an
environmental quality guideline has been set at 0.5 ng/L EE2 in
British Columbia.25 Therefore further investigation and investment in the refining of WwTW effluents to remove estrogens
more completely will be required to meet these standards. To
this end in the UK, an Endocrine Disruption Demonstration
Programme (EDDP) ran from 2005−2010 in order to assess
the efficacy of various sewage treatment technologies for
removing environmental estrogens.26
Advanced treatment technologies offering particular promise
for estrogen removal include adsorption onto granular activated
carbon (GAC)27 and chemical transformation using ozonation
(O3)28 or chlorine dioxide (ClO2) treatment.29 There is,
however, a lack of knowledge as to whether these advanced
technologies will actually reduce the biological effects of
effluents on wild fish living in rivers receiving these effluents.
Moreover, these are expensive options both in terms of their
carbon footprint and their financial costs. The majority of
studies carried out to investigate this problem have been
analytical chemistry studies, measuring only removal of specific
chemicals, particularly steroid estrogens. A small number of
studies have assessed fish biomarker responses such as
vitellogenin (VTG) induction/gene expression or secondary
sexual characteristics (SSC). For example, O3 advanced
treatment has been assessed in pilot plants in China, Sweden,
the U.K, and Germany;30−33 all studies found a reduction in
estrogenic effects post O3 treatment compared to the standard
treated effluent. However, two studies found VTG induction to
be still elevated compared to the control30,32 and interestingly
one study found estrogenic effects, in terms of VTG gene
expression, to be reinstated after further treatment by moving
biofilm reactor.31 Additionally, Stalter et al.34 found O3 treated
effluent to be toxic in a number of in vivo and in vitro tests,
although this toxicity was reduced by additional sand filtration
of the O3 effluent. In the U.K. Filby et al.32 also assessed
advanced treatment by GAC and ClO2 on steroid removal and
biomarker responses (VTG and SSC) in the fathead minnow.
In this study chemical analysis found ClO2 treatment to be the
most effective at removing steroid estrogens and both GAC and
ClO2 significantly reduced estrogenic biomarker responses.
To date, no attempts have been made to examine the longterm effects of these “improved” effluents on sexual development and, in particular, intersex induction of wild fish. Indeed,
no single study has combined chemical and biological
assessment of estrogen removal with the economic costs of
these processes. This is essential for determining whether the
investment required to install and maintain advanced treatment
processes in WwTW would be justified or whether existing
treatment technologies may be sufficient.
Therefore, there are two main aims of this study:
1. To assess the efficacy of advanced wastewater treatments,
compared to each other and to standard and tertiary
treated effluents, in their ability to mitigate feminization
of a native fish species and to reduce the steroid
estrogens present in the effluent.
2 To assess the relative costs of these different treatment
options.
Using this integrated approach a comparison of the proposed
advanced treatments with the secondary and tertiary treatments
currently employed could be undertaken.
In this study (using the same UK effluent as that tested in the
study of Filby et al.32) we have directly compared the
feminizing effects of ASP effluent (prior to sand filtration)
and sand filtered (SF) effluent with that of two different
advanced treatments, GAC and ClO2 (chosen as the most
effective advanced treatments from Filby et al.32), and also with
river and tap water (Figure 1). Our in vivo analyses included
Figure 1. Diagrammatic representation of the different treatment
streams tested. Fish (bottom of diagram) represent the adult and ELS
fish tanks. River treatment: river water was abstracted above the
WwTW outfall. Activated sludge process treatment: ASP effluent was
fed to the fish tanks prior to sand filtrations. Sand filter treatment: sand
filtered effluent (post ASP). Both the granular activated carbon (GAC)
and chlorine dioxide (ClO2) fed into the fish tanks directly after
treatment. The potable water was fed into the potable water control
(PWC) tanks.
the biomarker response (VTG induction), as well as
feminization of the gonads in long-term exposures covering
both adult and early life stages (ELS) of the cyprinid fish, the
roach, chosen because this is the wild fish species in which
intersexuality is known to be widespread throughout UK and
European rivers.
■
MATERIALS AND METHODS
Test Effluent. The experiments were conducted at a full
scale working WwTWs (treating primarily domestic effluent
(>90%) with a total population equivalent (pe) of ∼60 500)
with the addition of pilot plants for the advanced treatments.
The main WwTW had average sewage flows of 17.3 ML/d with
dry weather flows of 12.9 ML/d and maximum sewage flows of
28.0 ML/d.
1. The main sewage treatment plant comprised screening,
primary sedimentation, and nitrifying ASP (with ferrous
chloride addition for phosphate removal), followed by
sand filtration. Solid retention time (SRT) was 10−15
days and hydraulic retention time (HRT) was 11 h.
5566
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
In addition to the analytical chemistry, samples were also
analyzed using the yeast estrogen screen (YES) assay, as
described by Routledge and Sumpter,36 to determine total
estrogenic activity. Twenty-four-hour composites were also
taken for indicators of water quality (ammonia, nitrate, nitrite,
suspended solids (SS), chemical oxygen demand (COD),
biological oxygen demand (BOD), and total organic carbon
(TOC). See Supporting Information (Tables S1 and S2) for
details of physical−chemical properties of the effluent streams,
the potable water, and river water.
Biological Sampling and Analyses. After 3 months and
6 months exposure, blood plasma were taken from the adult
male roach for VTG measurement and measured using ELISA
as described previously.37 At the end of the tests, all fish were
humanely sacrificed by terminal anesthesia with benzocaine as
approved by the U.K. Home Office (Animals (Scientific
Procedures) Act 1986) and measured for fork length and wet
weight. Plasma VTG, gonad weight (for Gonad Somatic Index
(GSI)) and histology were recorded for adult fish as described
previously.4,38,39 Histological analysis was conducted on both
ELS and adult roach. Intersex indices were assigned to intersex
fish according to Jobling et al.4
Statistical Analyses. All data are expressed as means ± SD.
Data on estrogenicity were compared using repeated measures
one-way ANOVA, paired t-test or Friedman test for repeated
measures followed by Kruskal−Wallis for nonparametric data.
Biometric data were compared using one-way ANOVA
followed by least significant difference or Kruskal−Wallis test
followed by Mann−Whitney U test if nonparametric. Data on
biomarker responses were compared to the PWC using oneway ANOVA followed by Tukey’s post hoc test (VTG data
were log10 transformed prior to analysis). Data on intersex
induction were compared using Fishers exact test. Data on
intersex index (adults only) were compared using Kruskal−
Wallis test followed by Mann−Whitney U test. Statistical
significance was accepted at P < 0.05 for all tests.
Wastewater Treatment Cost Analysis. The full details of
the cost analysis of the different tertiary and advanced
wastewater treatments are detailed in the WRc report in the
Supporting Information (SI Section 2). Briefly the capital,
operating, and capital maintenance costs for SF alone or SF
plus O3, ClO2, or GAC were estimated and combined to
calculate the whole life cost for each treatment over a period
of 20 years. The WRc’s TR61 Version 10 Cost and Carbon
estimating software was used to estimate costs for capital
maintenance for the civil, mechanical, and electrical components of the treatment trains. Costs were determined for three
population sizes: 50 000, 100 000, and 250 000 pe.
2. The advanced treatments tested were chlorine dioxide
addition (1 mg/L applied dose) and GAC adsorption
(20 min empty bed contact time) operated as pilot plants
run in parallel each treating 200−1000 m3/day of the
main plant sand filter effluent (2500 pe each).
Test Organisms. Both adult and juvenile roach were
obtained from the Environment Agency’s National Coarse Fish
Farm (Calverton, Nottinghamshire, UK). These fish were bred
and raised in pure borehole water. Both adult and juvenile fish
were transported from Calverton in large fish poly bags
supplied with oxygen. Fish bags were floated in their respective
tanks for 1 h until water temperatures were ambient, and tank
water was gradually added to the bags before the fish were
released into the tanks.
■
EXPERIMENTAL DESIGN
Experiment 1: Adult Roach Assay. Following a preliminary study in 2008, in which the estrogenicity and potency
of the effluent was confirmed (by vitellogenic induction in adult
male roach), groups of adult male roach (40 per treatment)
were again housed in large 1000 L tanks and exposed to potable
water control (PWC), 100% river water, 100% effluent treated
by ASP, or 100% ASP effluent further treated by sand filtration
(SF), ClO2, or GAC. These roach were exposed for 6 months
from July to December 2009 and fed Carpco crumble excellent
(size 0.5−0.8) ad libitum, adjusted on the basis of uneaten feed.
Experiment 2: Early Life Stage (ELS) Sexual Differentiation Assay. Groups of 200 roach fry (30 days post
hatch) were placed into large 300 L tanks above the adult
exposure tanks and received the same effluent stream for each
treatment as the adults. Fish were exposed for six months to a
PWC, river water, 100% effluent after ASP, or after additional
sand filtration (SF), GAC, or ClO2 treatments (as for the
adults). Following the results of this exposure, a subset of ELS
fish exposed to PWC, river water, ASP, and SF were further
exposed for an additional six months (12 months total) to
further assess the effects of long-term exposure. Roach fry were
fed ZM fish food (pellet sizes 100−300) ad libitum, adjusted on
the basis of uneaten feed.
Both adult and juvenile fish were maintained in water
with flow rates of 6 and 3 L/min, respectively, to provide an
average of 8 and 14 tank renewals every 24 h. Tanks were
gently aerated at the surface to maintain dissolved oxygen
concentrations.
Analytical Chemistry. Analytical chemistry was carried out
weekly (24 sampling points in total) to determine steroid
estrogen (E1, E2, and EE2) concentrations with a limit of
detection (LOD) of 0.05 ng/L. Each of these weekly samples
was made up of 24 hourly samples “fixed” with copper(II)
nitrate and hydrochloric acid (Cu/HCl) to prevent biodegradation of steroids (for method see Supporting Information).
Samples were analyzed for steroid estrogens (estrone, 17βestradiol, and 17α-ethynylestradiol) using the LC-MS/MS
method as described elsewhere.16,35 Combined E2 equivalents
(EEQ) were calculated from the analytical chemistry using the
EA’s21 combined EEQ equation
■
RESULTS
Exposure Analyses: Analytical Chemistry of the
Treatment Waters. The estrogenic properties of the various
effluent streams are illustrated in Table 1 (and in Figure S1
of the Supporting Information). EEQ calculated from the
analytical chemistry found the most potent test stream was the
ASP effluent >SF > river = GAC > ClO2 > PWC. On average
the PWC had the lowest concentrations of the three steroid
estrogens measured (Table 1). YES analysis also showed
maximum estrogenic activity in the ASP treatment, and lower
levels were seen in the SF and river water. The GAC, ClO2, and
potable were all below the limit of quantification (Figure S2).
Removal rates for total estrogenicity (EEQ) for the different
tertiary or advanced treatments were 35.5% for SF, 72.4% for
EEQ = (E1 × 0.3) + (E2 × 1) + (EE2 × 10)
i.e., E1 was considered to be a third as potent as E2, whereas
EE2 was considered to be ten times as potent as E2. For
samples which were below the limit of detection (LOD), half
the LOD was used (i.e., 0.025 ng/L) for the calculations.
5567
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
River water, river; sand filtered effluent, SF; activated sludge process effluent, ASP; granular activated carbon treated effluent, GAC; chlorine dioxide treated effluent, ClO2; potable water, potable.
All concentrations in ng/L. Limit of detection (LOD) for all three steroid estrogens (E1, E2, and EE2) was 0.05 ng/L. Total estrogen equivalent EEQ was calculated from the analytical chemistry
EEQ = ((E1 × 0.3) + (E2 × 1) + (EE2 × 10)). When steroids were measured at LOD, half the LOD was used for the calculations, i.e. 0.025. Letters (a−e) signify statistical similarity (P < 0.05) between
treatments.
GAC, and 78% for ClO2. Specifically, removal rates ranged
from 64 to 87.7% for E1, 59.7−93.4% for E2, and 22.6−68.9%
for EE2 when compared to the ASP effluent. Overall, for the
three steroid estrogens analyzed, the ClO2 treatment had the
greatest removal effect of the advanced treatments tested.
Concentrations of ammonia, nitrite, and nitrate where
highest in the ASP treatment and lowest in the PWC; values
were significantly higher than PWC for all effluent streams (SF,
ASP, GAC, and ClO2). GAC, but not SF or ClO2, significantly
reduced the ammonia concentration compared to ASP (Table S4).
Tank conditions such as flow rate, pH, water temperature, and
dissolved oxygen are reported in Tables S1 and S2.
Experiment 1. Biological Effects of Treatment
Streams on Adult Roach. Over the six months of exposure,
adult survival was highest in the GAC and PWC treatments
(97.5%) and lowest in the ClO2 treatment (82.5%). The length
and weight of the adult male fish were not significantly different
among treatment groups. However, the GSI was significantly
lower in the ASP and SF effluents compared to the PWC
(Figure S3).
After three months exposure to the effluents, plasma VTG
was significantly elevated in the ASP treated fish compared to
the PWC and the other treatment groups (see Figure 2B). After
a further three months exposure, plasma VTG was 17.1-fold
higher in the SF effluent and 58.8-fold higher in the ASP
effluent than the PWC (see Figure 2B). There was no significant induction seen in the GAC (2.3-fold higher) or ClO2
(0.8-fold) treatments compared to the PWC. Regression analysis
found mean plasma VTG were significantly (P = 0.005)
positively correlated (R2 = 0.888) with mean calculated EEQ
from the chemical analysis.
Adult fish from the PWC, GAC, and SF treated effluent
showed no induction of intersex (ovotestis). In contrast, both
the ASP and the ClO2 exposures had significant induction of
intersex when compared to the PWC (Figure 2A). Therefore,
there was no significant correlation between percentage intersex
induction and EEQ (R2 = 0.25). The ASP exposed fish had a
significantly higher intersex index (= severity of the condition)
compared to the ClO2 treated fish (1.1 and 0.1, respectively).
A small (but not significant) number of the roach from the river
water exposure were also intersex; these fish had an intersex
index between that of the ClO2 and the ASP exposed fish (0.3).
Experiment 2. Biological Effects of Treatment
Streams on ELS Roach. After six months of exposure, the
survival of the ELS roach was highest in the SF effluent (97%)
and lowest in the ASP effluent (45%); GAC, ClO2, and PWC
also had survival rates below 54% (Figure S4). Somatic growth
was similar in all treatments, except for in the juvenile roach
from the river water treatment, which were significantly smaller
than those from other treatments (Figure S5).
ELS roach from the ASP treatment had the highest percentage of intersex (see Figure 3), whereas roach from the
PWC, river water, ClO2, and GAC treatments showed no
intersex. After a further six months exposure (PWC, river water,
SF, and ASP effluents only) it was found that there were no
“normal” males in the ASP treatment; only intersex and female
fish were found (Figure S7). Conversely, the level of intersex
found in the SF effluent remained at a low percentage similar to
that seen after only six months exposure compared with no ELS
intersex (male gonad with ovarian cavity or ovotestis) fish in
the river water or PWC.
Wastewater Treatment Cost Analysis. Treatment of
ASP by SF was the cheapest option with capital costs of
a
2.26
0.20
0.30
0.77
11.00
1.00
0.21
2.63
13.80
0.71
15.3
maximum
1.60
0.28
0.33
LOD
LOD
LOD
LOD
0.13
LOD
0.28
0.49
LOD
LOD
LOD
0.06
0.31
LOD
0.34
1.83
0.15
0.35
0.15
median
minimum
LOD
LOD
EEQ
2.50 ± 2.92 c
EE2
0.21 ± 0.19 b
E2
0.13 ± 0.31 a
E1
1.06 ± 3.04 b
GAC
treatment
Article
LOD
EEQ
0.55 ± 0.49 a
EE2
0.06 ± 0.03 a
potable
E2
0.08 ± 0.07 a
E1
0.18 ± 0.17 a
EEQ
1.95 ± 2.94 b
EE2
0.19 ± 0.27 b
ClO2
E2
0.08 ± 0.04 a
E1
0.49 ± 0.58 b
0.79
8.25
5.85
0.48
0.43
4.77
maximum
steroid ng/L
mean ± SD
2.00
7.67
21.80
1.87
2.89
19.4
11.10
0.96
0.12
0.50
0.38
LOD
0.16
6.29
5.54
1.88
0.16
0.44
0.18
LOD
0.77
2.13
3.41
0.37
LOD
0.11
0.08
minimum
LOD
1.44
median
0.18
EEQ
EE2
0.60 ± 0.39 d
0.62 ± 0.73 c
E2
E1
7.37 ± 5.09 e
5.84 ± 2.70 d
EEQ
EE2
0.46 ± 0.24 d
0.22 ± 0.21 b
E2
E1
2.97 ± 1.96 d
2.57 ± 1.65 c
EEQ
EE2
0.18 ± 0.15 c
mean ± SD
E2
E1
1.85 ± 1.10 c
steroid
ng/L
0.19 ± 0.11 b
ASP
SF
river
treatment
Table 1. Analytical Chemistry Results for Estrone (E1), 17β-Estradiol (E2), and 17α-Ethinylestradiol (EE2) over the Six Month Fish Exposure Perioda
9.05 ± 4.85 e
Environmental Science & Technology
5568
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
Figure 2. (A) Percentage induction of intersex in adult male roach exposed to different treatment streams for six months. Gray area represents
histologically male fish, hatched area represents intersex fish (male with oocytes in testis). Number (n) equals the number of fish analyzed in each
treatment. (B) Vitellogenin (VTG) induction in adult male roach exposed for three (black bars) and six (gray bars) months to different treatment
streams. (C−E) Photomicrographs of adult male roach testis after six months exposure to different treatment streams: (C) normal male testis from
sand filtered (SF) effluent; (D) intersex male from activated sludge process (ASP) effluent; (E) intersex male from chlorine dioxide treated effluent
(ClO2). (A and B) River, river water; SF, sand filtered effluent; ASP, activated sludge process effluent; GAC, granular activated carbon treated
effluent; ClO2, chlorine dioxide treated effluent; potable, potable water control. (B) VTG measured in ng/mL of blood plasma. Stars indicate
significant induction compared to the potable water control. (C−E) Arrows point to oocytes within testicular tissue. Scale bar 100 μm. Histological
sections of 5 μm, stained with Haematoxylin and Eosin.
£6,607,000 and yearly operational costs of £73,600. All the
advanced treatments required the ASP effluent to be treated
with SF. Although we did not assess O3 in terms of biological
effects (due to its lower efficacy in previous studies32) it was
found to be the next cheapest treatment option, followed by
ClO2. These treatments had additional (i.e., on top of SF)
capital costs of £1,300,000 and £2,200,000 and yearly
operational cost of £100,000 and £310,000 respectively. GAC
was found to be the most expensive of the advanced treatments
assessed, with additional capital costs of £8,500,000 and yearly
operational costs of £800,000. All the costs shown here (both
advanced and tertiary SF) were calculated for a 250,000 p.e.
plant. For the whole cost analysis report see SI Section 2.
percentage removal of estrogens similar to, or more effective
than, the other large scale UK GAC project (also part of the
UK EDDP) of 81% removal of E1, 68% removal of E2, and
50% removal of EE2.40 Over the six month exposure period, the
GAC plant generally showed good removal of steroid estrogens
when compared to that seen in the ASP and SF effluents.
However, on 14 of the 24 sampling occasions EE2 was measured above 0.1 ng/L, the proposed PNEC,21 and on 17
sampling occasions the GAC effluent had a combined EEQ
above the proposed total estrogen PNEC of 1 ng/L.21 On the
last sampling occasion even higher concentrations were
measured for all three estrogens (almost double the ASP
concentration for E1), suggesting the GAC may have become
saturated at this point and had begun releasing steroids back
into the effluent stream. In comparison, the ClO2 treatment was
generally more effective at removing steroid estrogens.
■
DISCUSSION AND CONCLUSIONS
Removal of Estrogens by Tertiary and Advanced
Treatment Processes. The GAC treatment produced
5569
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
Figure 3. (A) Percentage induction of intersex in early life stage (ELS) roach exposed to different treatment streams for six months. (B−E)
Photomicrographs of early life stage (ELS) roach after six (B−D) and twelve (E) months exposure to different treatment streams. (B) male testis
from potable water control (PWC); (C) female ovary from PWC, intersex male (normal testis with ovarian cavity) from activated sludge (ASP)
treated effluent; and (E) intersex male testis with ovarian cavity and testicular dysgenesis from ASP effluent after twelve months exposure. (A) River,
river water; SF, sand filtered effluent; ASP, activated sludge process effluent; GAC, granular activated carbon treated effluent; ClO2, chlorine dioxide
treated effluent; potable, potable water control. Gray area represents histologically male fish, black area represents histologically female fish, hatched
area represent intersex fish (male testis with ovarian cavity), and white area represents fish of undetermined sex. Fish of undetermined sex were
mainly due to the undifferentiated nature of the gonad at time of sampling, which could be due to the small size of the fish (see Figure S5 in
Supporting Information). Number (n) equals the number of fish analyzed in each treatment. (B−E) OC arrow; ovarian cavity. Scale bar 100 μm.
Histological sections of 5 μm, stained with Haematoxylin and Eosin.
ClO2 treatment compared to ASP or SF conflicts with the
significant induction of intersex (although not VTG) seen in
this treatment. Although ClO2 treatment employs oxidation as
its main route of chemical transformation it is possible that
chlorination could also have occurred. A few chlorination
byproducts have been found to be estrogenic in vitro. For
instance, one of the chlorination byproducts of EE2 (4-chloroEE2) has a binding affinity similar to its parent compound48
and one of Bisphenol A’s byproducts has increased activity in
MCF-7 cells.49 Indeed, in a study by Hu et al.,50 chlorinating
effluent, longer contact time (from 10 to 30 min), or higher
total organic carbon (TOC, humic acid) increased the
estrogenic activity (E-SCREEN) to above nonchlorinated
levels. However, due to the complex nature of effluents,
estrogens may not be the only factor in biological activity. In a
study by An et al.51 post-chlorinated effluent was found to
significantly induce VTG response in zebrafish when compared
to prechlorinated effluent and the control water; the authors
suggested this was due to a reduction in polycyclic aromatic
hydrocarbons (PAHs) in the post-chlorinated effluent.51
However, in our study, ClO2 treatment did not significantly
induce VTG or YES activity or reduce GSI, perhaps suggesting
some other (non-estrogenic) driver for intersex in this
treatment group. One possible explanation is antiandrogenic
factors. Intersex has been experimentally induced using the
antiandrogen flutamide in medaka52 without significant VTG
induction. Antiandrogenic activity (measured using in vitro
bioassays) has been reported in UK sewage effluents,53 and the
removal of this activity has been documented using in vitro
bioassays in GAC and O3 treated effluents,54,55 but, to the
authors knowledge, not yet in ClO2 treated effluents. However,
in the study of Filby et al.32 the same ClO2 treated effluent used
in our study did not inhibit male secondary sexual characteristics (fatpad and tubercles) in fathead minnows, suggesting the
However, this treatment technology also had a combined EEQ
above 1 ng/L for 33% of the sampling occasions.
Comparison of VTG Induction. Average VTG induction
generally mirrored total steroid EEQ measured in the analytical
chemistry. The adult male VTG concentrations in the ASP
treatment were similar to those published for adult roach
experimentally exposed to 100% sewage effluent for 4 months
(∼100 000 ng VTG/mL)41 but lower than those published for
roach exposed to 4 ng/L EE2 for ∼18 months post hatch
(∼2 000 000 ng VTG/mL).42 However, the baseline (PWC)
VTG concentrations were higher than would be expected when
compared to male roach from low/nonpolluted sites (<300 ng
VTG/mL)4,5,7,43 or dilution water/potable water controls from
other roach studies.41,42 Therefore we questioned whether
these higher than expected VTG baseline levels may be due to
the fish’s diet, as has been demonstrated recently by Beresford
et al.44 in fathead minnows. However, both the Carpco crumble
and the ZM diets were tested in the same manner as described
by Beresford et al.44 and no estrogenic activity was evident. The
PWC did have measurable concentrations (i.e., above LOD)
of estrogens. However, it is important to highlight that water
samples for chemical analysis were taken directly from the fish
tanks themselves, not from the potable input, and therefore
low levels of natural estrogens may have come from the fish.
Indeed, in a comprehensive review of measuring fish steroid
excretion, Scott and Ellis45 highlight that fish both excrete
steroids freely into water and also easily reuptake them.45 This
suggests the higher than expected VTG baseline results were
linked to aquatic rather than dietary exposure. Notwithstanding
this, it is important to note that male VTG induction is only a
biomarker of estrogen exposure rather than a direct predictor of
impaired reproductive health.46,47
ClO2 Treated Effluent and Intersex Induction. The
apparent high removal efficiency of steroid estrogens in the
5570
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
potable, ClO2, or GAC. Interestingly, in a 21 day fathead
minnow pair breeding assay, Filby et al.32 found significantly
reduced egg laying when fish were exposed to GAC treated
effluent compared to the control; this egg reduction was not
thought to be due to estrogenic effects. These data perhaps
suggest that advanced treatment of wastewater may also reduce
beneficial components of biologically treated wastewater and
that water could be “too clean” for wildlife to thrive. Unlike
laboratory conditions, sewage effluent and river water have
extremely complex chemistry and biological activity which
means chemicals may act in a synergistic, additive, or antagonistic manner. Therefore data from single chemical laboratory
exposures may not be entirely representative of real world
conditions; this should be an important consideration when
setting environmental quality standards (EQS), due to the
significant costs involved in producing effluents of such high
quality.
Cost Comparison for Tertiary and Advanced Treatments. Cost analyses of sewage treatments have rarely been
conducted in the literature, however Jones et al.59 employed
similar techniques to look at advanced treatments different
from those compared in our research. Of the advanced treatments assessed in this experiment, GAC was the only one
which completely removed intersex induction. Notwithstanding
this, GAC is both expensive and carbon intensive as a treatment
technology. During the UK EDDP the capital and running
costs of GAC, ClO2, and O3 treatments were calculated and
compared. Whole life financial costs and carbon dioxide (CO2)
release were found to be highest for GAC (£5.53 pe/yr and
>13 kg CO2/pe/yr) followed by ClO2 (£2.91 pe/yr and
<6 kg/CO2/pe/yr) and then O3 (£2.25 pe/yr and <3 kg/CO2/
pe/yr).26 However, sand filtration also removed significant
feminization of ASP effluent on both ELS and adult roach
(preventing male oviduct and oocyte formation). Unfortunately, no cost comparison was made with the SF tertiary
treatment during the EDDP.26 Therefore we commissioned the
comparison detailed in the SI Section 2. This found SF alone
cost £1.69 pe/yr, around a third of the total GAC cost. These
cost comparisons are obviously indicative rather than definitive
when considering wastewater treatment worldwide. However,
they do highlight that the addition of SF to secondary treated
effluent could be a cost-effective way of mitigating feminization
of wild fish, especially in locations where the cost of advanced
sewage treatments is prohibitive.
effluent was not antiandrogenic. Therefore, the mechanism for
intersex induction in this treatment is yet to be elucidated. The
possible impacts of ClO2 treatment on fish should be further
investigated if ClO2 is to be introduced as an advanced wastewater
treatment method.
Ovotestis Can Be Induced in Adult Roach. This is the
first study to suggest adult male roach are sensitive to ovotestis
induction (in only six months), indicating the period of
testicular regrowth (following the annual spawning even) is a
“window” in development when roach are sensitive to
disruption. Previous studies by Rodgers-Gray et al.41 showed
that intersex in adult male roach is not induced when the testes
are fully mature, further supporting the existence of a sensitive
window. This may also explain the field observations of Jobling
et al.4,6 showing that older fish were more feminized than
younger fish at the same site. This phenomenon of increasing
feminization was also seen by Lange et al.56 in roach exposed to
100% effluent for 3.5 years, after which time all of the fish were
histologically female compared to only 31% of males having
ovotestis after 1 year of exposure.56 No sewage exposure study,
so far, has reported ovotestis in ELS roach less than 1 year
old,38,56 again suggesting it is the post-spawning regrowth
period which is most sensitive to ovotestis induction.
Interestingly the occurrence of oocyte induction in the adult
males and feminized sperm ducts in juveniles were not always in
the same treatments. Although both conditions occurred at a
high frequency in the ASP treated effluent, no feminization of
sperm ducts was found in the ClO2 treatment (which had a
significant occurrence of ovotestis), or in the river water (which
also had a low incidence of ovotestis). Conversely, a low level of
feminized ducts was seen in the juveniles from the SF effluent
whereas no ovotestis induction was seen in adult males from this
treatment. These data suggest different mechanisms or perhaps
even different chemical drivers for these conditions. Statistical
modeling by Jobling et al.57 has suggested that antiandrogenic
activity (measured by a yeast androgen screen (YAS) and
frequently found in WwTW effluents53,58) is significantly
correlated with the frequency of feminized ducts seen in wild
roach, whereas ovotestis occurrence was more strongly correlated
to the fish’s age and estrogen (E1/nonylphenol) exposure.57
Therefore, more research may be required to further understand
the likely multicausal nature of intersex induction in wild fish.
Regulatory Implications. During the six month exposure,
the GAC’s mean (2.6 ng/L) and median (1.62 ng/L) EEQ
would have failed the proposed combined estrogen PNEC
of 1 ng/L.21 However, no intersex was induced in this treatment. The majority of studies considered when deriving PNEC
use reproductive end points rather than intersex induction.
However, in a long-term roach developmental study to EE2,
probably the best studied of the estrogens, Lange et al.42 found
no observable effect on intersex or reproduction at 0.3 ng/L EE2
measured over the exposure (1 ng/L nominal); this value is
similar to Caldwell et al.22 EE2 PNEC of 0.35 ng/L. Both of
these EE2 values fit with the lack of intersex induction in the
GAC (mean EE2 0.21 ± 0.19 ng/L), but not with the low level
of adult intersex observed in the river water (mean 0.18 ±
0.15 ng/L) exposed fish. Indeed, no significant difference in
calculated EEQs was found between the river water and the
GAC treatments, highlighting the difficulty of predicting effects
of chemicals in complex mixtures. Indeed it was also notable
that although the SF effluent had significantly higher levels of
estrogens (and low induction of intersex) juvenile survival was
much higher compared to the estrogen reduced conditions of
■
ASSOCIATED CONTENT
S Supporting Information
*
Section 1: supplementary experimental section detailing
composite water sampling; tables containing tank conditions
(flow rates, DO, pH, temperature); graphs of analytical chemistry E1, E2, EE2 and total estrogens (EEQ) and average YES
results; water quality data for ammonia, nitrite and nitrate,
suspended solids, BOD, COD and TOC; adult male roach
GSI; percentage survival, average lengths and weights of ELS
roach after six and twelve months; percentage intersex of ELS
fish after twelve months. Section 2: WRc report entitled −
Comparison of cost of Tertiary Sand Filtration for EDC
removal with Other Processes. This material is available free
of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected].
5571
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
Notes
estrogens in carbonaceous and nitrifying activated sludge processes.
Chemosphere 2010, 81, 1−6.
(16) Kanda, R.; Churchley, J. Removal of endocrine disrupting
compounds during conventional wastewater treatment. Environ.
Technol. 2008, 29, 315−323.
(17) Ifelebuegu, A. O. The fate and behavior of selected endocrine
disrupting chemicals in full scale wastewater and sludge treatment unit
processes. Int. J. Environ. Sci. Technol. 2011, 8, 245−254.
(18) Johnson, A.; Tanaka, H.; Okayasu, Y.; Suzuki, Y. Estrogen
content and relative performance of Japanese and British sewage
treatment plants and their potential impact on endocrine disruption.
Environ. Sci. 2007, 14, 319−329.
(19) Thorpe, K. L.; Cummings, R. I.; Hutchinson, T. H.; Scholze, M.;
Brighty, G. C.; Sumpter, J. P.; Tyler, C. R. Relative potencies and
combination effects of steroidal estrogens in fish. Environ. Sci. Technol.
2003, 37, 1142−1149.
(20) Matsui, S.; Takigami, H.; Matsuda, T.; Taniguchi, N.; Adachi, J.;
Kawami, H.; Shimizu, Y. Estrogen and estrogen mimics contamination
in water and the role of sewage treatment. Water Sci. Technol. 2000, 42,
173−179.
(21) Young, W. F.; Whitehouse, P.; Johnson, I.; Sorokin, N. Proposed
Predicted-No-Effect-Concentrations (PNECs) for Natural and Synthetic
Steroid Oestrogens in Surface Waters; EA R&DTechnical Report P2T04/1; 2004; EA5098.
(22) Caldwell, D. J.; Mastrocco, F.; Hutchinson, T. H.; Laenge, R.;
Heijerick, D.; Janssen, C.; Anderson, P. D.; Sumpter, J. P. Derivation
of an aquatic predicted no-effect concentration for the synthetic
hormone, 17 alpha-ethinyl estradiol. Environ. Sci. Technol. 2008, 42,
7046−7054.
(23) Cao, Q.; Yu, Q.; Connell, D. W. Fate simulation and risk
assessment of endocrine disrupting chemicals in a reservoir receiving
recycled wastewater RID B-8657-2008. Sci. Total Environ. 2010, 408,
6243−6250.
(24) Grung, M.; Kallqvist, T.; Sakshaug, S.; Skurtveit, S.; Thomas, K.
V. Environmental assessment of Norwegian priority pharmaceuticals
based on the EMEA guideline. Ecotoxicol. Environ. Saf. 2008, 71, 328−
340.
(25) Nagpal, N. K.; Meays, C. L. Water Quality Guidelines for
Pharmaceutically-active Compounds (PhACs) 17 a-ethinylestradiol
(EE2) Overview Report; 2009. www.env.gov.bc.ca/wat/wq/
BCguidelines/...EE2/PhACs-EE2-overview.pdf.
(26) Butwell, A. J.; Harman, M.; Johnson, I.; Rockett, L.; Silvil, D.
Assessment of the Performance of Wastewater Treatment Works in
Removing Oestrogenic Substances; 2010; UKWIR Report 10/TX/04/17.
(27) Zhang, Y. P.; Zhou, J. L. Removal of estrone and 17 betaestradiol from water by adsorption. Water Res. 2005, 39, 3991−4003.
(28) Hansen, K. M. S.; Andersen, H. R.; Ledin, A. Ozonation of
estrogenic chemicals in biologically treated sewage. Water Sci. Technol.
2010, 62, 649−657.
(29) Lee, Y.; Escher, B. I.; Von Gunten, U. Efficient removal of
estrogenic activity during oxidative treatment of waters containing
steroid estrogens. Environ. Sci. Technol. 2008, 42, 6333−6339.
(30) An, L.; Hu, J.; Yang, M. Evaluation of estrogenicity of sewage
effluent and reclaimed water using vitellogenin as a biomarker. Environ.
Toxicol. Chem. 2008, 27, 154−158.
(31) Gunnarsson, L.; Adolfsson-Erici, M.; Bjorlenius, B.; Rutgersson,
C.; Forlin, L.; Larsson, D. G. J. Comparison of six different sewage
treatment processes-Reduction of estrogenic substances and effects on
gene expression in exposed male fish RID E-5243-2011. Sci. Total
Environ. 2009, 407, 5235−5242.
(32) Filby, A. L.; Shears, J. A.; Drage, B. E.; Churchley, J. H.; Tyler,
C. R. Effects of advanced treatments of wastewater effluents on
estrogenic and reproductive health impacts in fish. Environ. Sci.
Technol. 2010, 44, 4348−4354.
(33) Stalter, D.; Magdeburg, A.; Weil, M.; Knacker, T.; Oehlmann, J.
Toxication or detoxication? In vivo toxicity assessment of ozonation as
advanced wastewater treatment with the rainbow trout. Water Res.
2010, 44, 439−448.
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
This project was funded by Severn Trent Water and Brunel
University. We thank all those who provided logistical and
technical support to the project at Severn Trent Water, Severn
Trent Services, Calverton Fish Farm, and the Institute for the
Environment, Brunel University.
(1) Kavanagh, R. J.; Balch, G. C.; Kiparissis, Y.; Niimi, A. J.; Sherry, J.;
Tinson, C.; Metcalfe, C. D. Endocrine disruption and altered gonadal
development in white perch (Morone americana) from the lower Great
Lakes region. Environ. Health Perspect. 2004, 112, 898−902.
(2) Tyler, C. R.; Jobling, S. Roach, sex, and gender-bending
chemicals: The feminization of wild fish in English rivers. Bioscience
2008, 58, 1051−1059.
(3) Bjerregaard, L. B.; Madsen, A. H.; Korsgaard, B.; Bjerregaard, P.
Gonad histology and vitellogenin concentrations in brown trout
(Salmo trutta) from Danish streams impacted by sewage effluent.
Ecotoxicology 2006, 15, 315−327.
(4) Jobling, S.; Williams, R.; Johnson, A.; Taylor, A.; Gross-Sorokin,
M.; Nolan, M.; Tyler, C. R.; van Aerle, R.; Santos, E.; Brighty, G. C.
Predicted exposures to steroid estrogens in UK rivers correlate with
widespread sexual disruption in wild fish populations. Environ. Health
Perspect. 2006, 114, 32−39.
(5) Jobling, S.; Nolan, M.; Tyler, C. R.; Brighty, G. C.; Sumpter, J. P.
Widespread sexual disruption in wild fish. Environ. Sci. Technol. 1998,
32, 2498−2506.
(6) Jobling, S.; Beresford, N.; Nolan, M.; Rodgers-Gray, T.; Brighty,
G. C.; Sumpter, J. P.; Tyler, C. R. Altered sexual maturation and
gamete production in wild roach (Rutilus rutilus) living in rivers that
receive treated sewage effluents. Biol. Reprod. 2002, 66, 272−281.
(7) Jobling, S.; Coey, S.; Whitmore, J. G.; Kime, D. E.; Van Look, K.
J. W.; McAllister, B. G.; Beresford, N.; Henshaw, A. C.; Brighty, G. C.;
Tyler, C. R.; Sumpter, J. P. Wild intersex roach (Rutilus rutilus) have
reduced fertility. Biol. Reprod. 2002, 67, 515−524.
(8) Harris, C. A.; Hamilton, P. B.; Runnalls, T. J.; Vinciotti, V.;
Henshaw, A.; Hodgson, D.; Coe, T. S.; Jobling, S.; Tyler, C. R.;
Sumpter, J. P. The consequences of feminization in breeding groups of
wild fish. Environ. Health Perspect. 2011, 119, 306−311.
(9) Metcalfe, C. D.; Metcalfe, T. L.; Kiparissis, Y.; Koenig, B. G.;
Khan, C.; Hughes, R. J.; Croley, T. R.; March, R. E.; Potter, T.
Estrogenic potency of chemicals detected in sewage treatment plant
effluents as determined by in vivo assays with Japanese medaka
(Oryzias latipes). Environ. Toxicol. Chem. 2001, 20, 297−308.
(10) Parrott, J. L.; Wood, C. S. Fathead minnow lifecycle tests for
detection of endocrine-disrupting substances in effluents. Water Qual.
Res. J. Can. 2002, 37, 651−667.
(11) Orn, S.; Holbech, H.; Madsen, T. H.; Norrgren, L.; Petersen, G.
I. Gonad development and vitellogenin production in zebrafish (Danio
rerio) exposed to ethinylestradiol and methyltestosterone. Aquat.
Toxicol. 2003, 65, 397−411.
(12) Johnson, A. C.; Williams, R. J. A model to estimate influent and
effluent concentrations of estradiol, estrone, and ethinylestradiol at
sewage treatment works. Environ. Sci. Technol. 2004, 38, 3649−3658.
(13) Kumar, V.; Nakada, N.; Yasojima, M.; Yamashita, N.; Johnson,
A. C.; Tanaka, H. The arrival and discharge of conjugated estrogens
from a range of different sewage treatment plants in the UK.
Chemosphere 2011, 82, 1124−1128.
(14) Gabet-Giraud, V.; Miege, C.; Choubert, J. M.; Ruel, S. M.;
Coquery, M. Occurrence and removal of estrogens and beta blockers
by various processes in wastewater treatment plants. Sci. Total Environ.
2010, 408, 4257−4269.
(15) McAdam, E. J.; Bagnall, J. P.; Koh, Y. K. K.; Chiu, T. Y.; Pollard,
S.; Scrimshaw, M. D.; Lester, J. N.; Cartmell, E. Removal of steroid
5572
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573
Environmental Science & Technology
Article
development and reproduction in medaka (Oryzias latipes). Mar.
Environ. Res. 2006, 62, S253−S257.
(53) Johnson, I.; Hetheridge, M.; Tyler, C. R. Assessment of (anti-)
oestrogenic and (anti-) androgenic activities of f inal eff luents f rom sewage
treatment works; 2007; Science Report SC020118/SR; http://
publications.environment-agency.gov.uk/pdf/SCHO0207BMAX-e-e.
pdf.
(54) Grover, D. P.; Balaam, J.; Pacitto, S.; Readman, J. W.; White, S.;
Zhou, J. L. Endocrine disrupting activities in sewage effluent and river
water determined by chemical analysis and in vitro assay in the context
of granular activated carbon upgrade. Chemosphere 2011, 84, 1512−
1520.
(55) Stalter, D.; Magdeburg, A.; Wagner, M.; Oehlmann, J.
Ozonation and activated carbon treatment of sewage effluents:
Removal of endocrine activity and cytotoxicity. Water Res. 2011, 45,
1015−1024.
(56) Lange, A.; Paull, G. C.; Hamilton, P. B.; Iguchi, T.; Tyler, C. R.
Implications of Persistent Exposure to Treated Wastewater Effluent for
Breeding in Wild Roach (Rutilus rutilus) Populations. Environ. Sci.
Technol. 2011, 45, 1673−1679.
(57) Jobling, S.; Burn, R. W.; Thorpe, K.; Williams, R.; Tyler, C.
Statistical modelling suggests that anti-androgens in effluents from
wastewater treatment works contribute to widespread sexual
disruption in fish living in English rivers. Environ. Health Perspect.
2009, 117, 797−802.
(58) Sohoni, P.; Sumpter, J. P. Several environmental oestrogens are
also anti-androgens. J. Endocrinol. 1998, 158, 327−339.
(59) Jones, O. A. H.; Green, P. G.; Voulvoulis, N.; Lester, J. N.
Questioning the excessive use of advanced treatment to remove
organic micropollutants from wastewater. Environ. Sci. Technol. 2007,
41, 5085−5089.
(34) Stalter, D.; Magdeburg, A.; Oehlmann, J. Comparative toxicity
assessment of ozone and activated carbon treated sewage effluents
using an in vivo test battery. Water Res. 2010, 44, 2610−2620.
(35) Standing Committee of Analysts. The determination of steroid
estrogens in environmental and waste waters using chromatography
and mass spectrometry. Methods for the Examination of Waters and
Associated Materials; 2008;http://www.environment-agency.gov.uk/
research/commercial/32874.aspx.
(36) Routledge, E. J.; Sumpter, J. P. Estrogenic activity of surfactants
and some of their degradation products assessed using a recombinant
yeast screen. Environ. Toxicol. Chem. 1996, 15, 241−248.
(37) Tyler, C.; van Aerle, R.; Hutchinson, T.; Maddix, S.; Trip, H. An
in vivo testing system for endocrine disruptors in fish early life stages
using induction of vitellogenin RID C-5148-2009. Environ. Toxicol.
Chem. 1999, 18, 337−347.
(38) Rodgers-Gray, T. P.; Jobling, S.; Kelly, C.; Morris, S.; Brighty, G.
C.; Waldock, M. J.; Sumpter, J. P.; Tyler, C. R. Exposure of juvenile
roach (Rutilus rutilus) to treated sewage effluent induces dosedependent and persistent disruption in gonadal duct development.
Environ. Sci. Technol. 2001, 35, 462−470.
(39) Nolan, M.; Jobling, S.; Brighty, G. C.; Sumpter, J. P.; Tyler, C.
R. A histological description of intersexuality in the roach. J. Fish Biol.
2001, 58, 160−176.
(40) Grover, D. P.; Zhou, J. L.; Frickers, P. E.; Readman, J. W.
Improved removal of estrogenic and pharmaceutical compounds in
sewage effluent by full scale granular activated carbon: Impact on
receiving river water. J. Hazard. Mater. 2011, 185, 1005−1011.
(41) Rodgers-Gray, T. P.; Jobling, S.; Morris, S.; Kelly, C.; Kirby, S.;
Janbakhsh, A.; Harries, J. E.; Waldock, M. J.; Sumpter, J. P.; Tyler, C.
R. Long-term temporal changes in the estrogenic composition of
treated sewage effluent and its biological effects on fish. Environ. Sci.
Technol. 2000, 34, 1521−1528.
(42) Lange, A.; Paull, G. C.; Coe, T. S.; Katsu, Y.; Urushitani, H.;
Iguchi, T.; Tyler, C. R. Sexual reprogramming and estrogenic
sensitization in wild fish exposed to ethinylestradiol. Environ. Sci.
Technol. 2009, 43, 1219−1225.
(43) Geraudie, P.; Gerbron, M.; Hill, E.; Minier, C. Roach (Rutilus
rutilus) reproductive cycle: A study of biochemical and histological
parameters in a low contaminated site. Fish Physiol. Biochem. 2010, 36,
767−777.
(44) Beresford, N.; Brian, J. V.; Runnalls, T. J.; Sumpter, J. P.;
Jobling, S. Estrogenic activity of tropical fish food can alter baseline
vitellogenin concentrations in male fathead minnow (Pimephales
promelas). Environ. Toxicol. Chem. 2011, 30, 1139−1145.
(45) Scott, A. P.; Ellis, T. Measurement of fish steroids in water - A
review. Gen. Comp. Endrocrinol. 2007, 153, 392−400.
(46) Hutchinson, T. H.; Ankley, G. T.; Segner, H.; Tyler, C. R.
Screening and testing for endocrine disruption in fish - Biomarkers as
“signposts”, not “traffic lights”, in risk assessment. Environ. Health
Perspect. 2006, 114, 106−114.
(47) Bosker, T.; Munkittrick, K. R.; MacLatchy, D. L. Challenges and
opportunities with the use of biomarkers to predict reproductive
impairment in fishes exposed to endocrine disrupting substances.
Aquat. Toxicol. 2010, 100, 9−16.
(48) Moriyama, K.; Matsufuji, H.; Chino, M.; Takeda, M.
Identification and behavior of reaction products formed by
chlorination of ethynylestradiol. Chemosphere 2004, 55, 839−847.
(49) Liu, J. H.; Carr, S.; Rinaldi, K.; Chandler, W. Screening
estrogenic oxidized by-products by combining ER binding and
ultrafiltration. Environ. Toxicol. Pharmacol. 2005, 20, 269−278.
(50) Hu, J. Y.; Chen, X.; Jin, X.; Tan, X. L. Effect of chlorination on
estrogenicity in chlorinated treated effluent. Drinking Water Treatment,
Supply and Management in Asia (IWA-ASPIRE 2005) 2006, 6, 185−
191.
(51) An, L.; Hu, J.; Yang, M.; Jin, F.; Du, Q.; Ke, Z. Enhanced
vitellogenin induction of secondary effluents by chlorination. Bull.
Environ. Contam. Toxicol. 2006, 77, 67−73.
(52) Kang, I. J.; Hano, T.; Oshima, Y.; Yokota, H.; Tsuruda, Y.;
Shimasaki, Y.; Honjo, T. Anti-androgen flutamide affects gonadal
5573
dx.doi.org/10.1021/es204590d | Environ. Sci. Technol. 2012, 46, 5565−5573