Citation
Edocrine Disruption in the Fathead Minnow (Pimephales Promelas) following a series of upgrades to a wastewater treatment facility

Material Information

Title:
Edocrine Disruption in the Fathead Minnow (Pimephales Promelas) following a series of upgrades to a wastewater treatment facility
Creator:
Baroffio, Angelina Free ( author )
Language:
English
Physical Description:
1 electronic file (61 pages). : ;

Subjects

Subjects / Keywords:
Water -- Purification ( lcsh )
Minnows ( lcsh )
Fishes -- Endocrinology ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
This study aimed to characterize the impact of treatment infrastructure upgrades on the occurrence of endocrine disrupting chemicals (EDCs) discharged by a wastewater treatment plant (WWTP) effluent and their subsequent effects on fish endocrine function. This site has been evaluated before and after two major upgrades in wastewater treatment infrastructure, which were implemented in 2007 and 2012. Our study assessed the potential impacts on the Boulder Creek receiving water, as well as identified and evaluated the extent of estrogenic endocrine disruption in the native fathead minnow (Pimephales promelas) that may be occurring after the implementation of the 2012 upgrade. We conducted an integrative, 8-week, on-site, continuous-flow, exposure experiment using adult male fathead minnows to assess in vivo estrogenicity of the WWTP effluent water, relative to reference water and results from prior years (both pre- and post-upgrade). We collected data for a wide array of biological endpoints. Results for plasma vitellogenin concentrations and sperm development were emphasized. It was found that in vivo effluent estrogenicity following the 2012 upgrade was insignificant in comparison to pre-upgrade levels. However, the occurrence of an extreme flood event in the Boulder area resulted in the detection of some significant effluent estrogenicity, indicating that such conditions may impact the ability of WWTPs to effectively remove estrogenic EDCs from the effluent.
Thesis:
Thesis (M.S.)- University of Colorado Denver.
Bibliography:
Includes bibliographic references
System Details:
System requirements: Adobe Reader.
General Note:
Department of Integrative Biology
Statement of Responsibility:
by Angelina Free Barrofio.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
945637924 ( OCLC )
ocn945637924
Classification:
LD1193.L45 2015m B37 ( lcc )

Downloads

This item has the following downloads:


Full Text
ENDOCRINE DISRUPTION IN THE FATHEAD MINNOW
(PIMEPHALES PROMELAS) FOLLOWING A
SERIES OF UPGRADES TO A
WASTEWATER TREATMENT FACILITY
by
ANGELINA FREE BAROFFIO
B.S., Tulane University of Louisiana, 2012
A thesis submitted to the
Faculty of the Graduate School of the University of Colorado
in partial fulfillment of the requirements for the degree of
Master of Science
Integrative Biology Program
2015


2015
ANGELINA BAROFFIO
ALL RIGHTS RESERVED
11


This thesis for the Master of Science degree by
Angelina Free Baroffio
has been approved for the
Integrative Biology Program
by
Alan Vajda, Chair
Michael Greene
Kristen Keteles
October 30th, 2015
m


Baroffio, Angelina Free (M.S., Integrative Biology)
Endocrine disruption in the fathead minnow (Pimphalespromelas) following a series of
upgrades to a wastewater treatment facility.
Thesis directed by Assistant Professor Alan Vajda.
ABSTRACT
This study aimed to characterize the impact of treatment infrastructure upgrades
on the occurrence of endocrine disrupting chemicals (EDCs) discharged by a wastewater
treatment plant (WWTP) effluent and their subsequent effects on fish endocrine function.
This site has been evaluated before and after two major upgrades in wastewater treatment
infrastructure, which were implemented in 2007 and 2012. Our study assessed the
potential impacts on the Boulder Creek receiving water, as well as identified and
evaluated the extent of estrogenic endocrine disruption in the native fathead minnow
(Pimephales promelas) that may be occurring after the implementation of the 2012
upgrade. We conducted an integrative, 8-week, on-site, continuous-flow, exposure
experiment using adult male fathead minnows to assess in vivo estrogenicity of the
WWTP effluent water, relative to reference water and results from prior years (both pre-
and post-upgrade). We collected data for a wide array of biological endpoints. Results for
plasma vitellogenin concentrations and sperm development were emphasized. It was
found that in vivo effluent estrogenicity following the 2012 upgrade was insignificant in
comparison to pre-upgrade levels. However, the occurrence of an extreme flood event in
the Boulder area resulted in the detection of some significant effluent estrogenicity,
indicating that such conditions may impact the ability of WWTPs to effectively remove
estrogenic EDCs from the effluent.
IV


The form and content of this abstract are approved. I recommend its publication.
Approved: Alan Vajda


ACKNOWLEDGEMENTS
I would like to thank my advisor, Dr. Alan Vajda, and the members of my thesis
committee, Dr. Michael Greene and Dr. Kristen Keteles, for their help and patience as I
completed this thesis. I would also like to thank the University of Colorado at Denver and
Dr. Michael Wunder, USGS and Dr. Larry Barber, and all of those people who helped
make this research possible with all their hard work in our lab, including Chelsea Ladd,
Alexandra Harrison, Gary Broyles, Zia Faizi, Munira Lantz, Kendra Occhipinti, and
Reyna Griffin. Special thanks are also due to the City of Boulder, for allowing us to do
this work on the premises of their wastewater treatment plant. And, of course, all my
gratitude and love to my family and my friends.
vi


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION..............................................................1
Wastewater Treatment Plants: Engineering and Technology....................1
Endocrine Disruption and Endocrine-Disrupting Compounds....................1
Estrogens................................................................2
Estrogen exposure and reproductive function in aquatic ecosystems........4
Plasma Vitellogenin (Vtg)................................................5
Stream Background..........................................................7
Upgrades and Flood Events................................................7
Effluent source and contaminants.......................................9
Reproductive disruption in free-living organisms in WWTP effluent........9
Field Studies.........................................................10
Mobile Lab Studies...................................................11
Hypothesis..............................................................13
II. MATERIALS AM) METHODS.................................................. 17
Study Sites...............................................................17
On-Site Bioassay Laboratory.............................................17
Water Sampling and Chemical Analysis....................................18
Collection..........................................................18
Chemistry...........................................................19
Tissue Sampling and Analysis..........................................19
vii


Dissection
19
Scoring Secondary Sexual Characteristics.................................20
Quantitative monoclonal antibody detection: ELISA Assay..................21
Detecting Plasma Vitellogenin..........................................21
Histology................................................................22
Statistical Analysis.......................................................24
III. RESULTS....................................................................25
Plasma Vitellogenin...........................................................25
Somatic Development and Condition.............................................25
Secondary Sexual Characteristics..............................................26
Sperm Developmental Stages and Abundance......................................26
IV. DISCUSSION.................................................................32
Overview.....................................................................32
Hypothesis.................................................................32
Plasma vitellogenin....................................................33
Sperm abundance and development........................................35
Maleness...............................................................36
Condition factor.......................................................36
Synthesis....................................................................38
Summary......................................................................39
40
viii
References


LIST OF TABLES
TABLE
1. Secondary sexual characteristics and sperm abundance rating scales............21
2. One-way ANOVAs for Gonadosomatic Index, Condition Factor, Nuptial Tubercle
Number, Fish Length, and Plasma Vitellogenin Concentrations.................31
3.Statistical test results for secondary sexual characteristics: maleness, nuptial tubercle
prominence, and dorsal fat pad prominence...................................31
4. Statistical test results for sperm developmental stages by treatment..........31
IX


LIST OF FIGURES
FIGURE
1. Structures of common estrogenic compounds.....................................4
2. Physiological action of an estrogen in an oviparous vertebrate................5
3. Cellular responses to estrogens...............................................6
4. City of Boulder Wastewater Treatment Plant purification system................8
5. Measured and observed estrogenic effects in fish exposed on-site to Boulder WWTP
effluent, pre- and post-2007 upgrade and upstream Boulder Creek reference water
and a 50:50 effluent/reference mixture..................................10
6. Male fathead minnow {Pimephalespromelas)......................................11
7. Factors Contributing to Effluent Estrogenicity: where estrogenicity increases or
estrogenicity decreases......................................................12
8. Demasculinization and Feminization in the Fathead Minnow: a set of biological
endpoints....................................................................16
9. Boulder Creek, Colorado and the location of the COB WWTP......................17
10. Fish exposure mobile onsite at COB WWTP.......................................18
11. From left to right: reference ovary, intersex testis, and reference testis....23
12. Plasma VTG in male fathead minnow {Pimephalespromelas) in an initial control
group, reference groups, and effluent-exposed groups.........................27
13. Condition factor by week and treatment within that week.......................27
14. Gonadosomatic index score by week, and treatment within that week.............28


15. Length only, grouped by treatment: initial controls, reference individuals, and
effluent-treated fish............................................................28
16. Maleness index scores by treatment: initial controls, reference individuals, and
effluent-treated fish............................................................29
17. Percentage of male fish within a given sperm developmental stage, organized by
treatment........................................................................29
18. Sperm developmental stages by treatment: spermatogonia, spermatocytes, spermatids,
and spermatozoa..................................................................30
19. Plasma vitellogenin induction versus measured flow through the Boulder WWTP .. 34
XI


LIST OF ABBREVIATIONS
BPA Bisphenol-A
COB City of Boulder
EDC Endocrine Disrupting Compound
El Estrone
E2a 17a-Ethynylestradiol
E2P 17f-Ethynylestradiol
EE2 Ethinyl Estradiol
E3 Estriol
VTG Vitellogenin
WWTP Wastewater Treatment Plant


CHAPTERI
INTRODUCTION
Wastewater Treatment Plants: Engineering and Technology
Human management of wastewater has made significant strides in terms of
sustainability and effluent quality from wastewater treatment plants (WWTPs) by
employing numerous types of upgrades to traditional WWTP designs across the US and
around the world.1 The efficacy of WWTPs in removing small bio-active molecules in
the resulting effluent varies with treatment technology and has implications for
downstream human 2 and ecosystem health.3 Hydrologic variation, due to seasonality
and, in more extreme cases, drought 45 or flood 67 can impact WWTP plant efficiency89
and the dilution of WWTP effluents in total stream volume.10
Implementation of appropriate WWTP treatment technologies is complicated by
the often conflicting priorities of engineers, managers, environmental specialists and
other stakeholders, regulatory requirements,11 and the availability of funding for
research, and infrastructure upgrade and WWTP operation costs. I213
We evaluated major upgrades in WWTP disinfection processes to determine the
efficacy of such upgrades in minimizing the exposure of downstream wildlife populations
to estrogenic that is, interacting with estrogen receptors (ER) and eliciting a response at
the cellular, tissue, or behavioral levels endocrine-disrupting compounds (EDCs).
Endocrine Disruption and Endocrine-Disrupting Compounds
The concentration and composition of chemical contaminants in WWTP effluents
changes over time in response to a number of different factors.14 Most WWTP effluents
are composed of highly complex mixtures,15 as WWTPs process influent from a broad
range of domestic and industrial sources, such as cities, manufacturing plants, and even


animal feedlots.16 The complexity of these WWTP effluent chemical compositions varies
temporally 17 due to short- and long-term variation in chemical usage, anthropogenic land
use, WWTP design (ex. chlorine/decholorination versus UV light and advanced oxidation
disinfection treatments) and efficiency, hydrologic flow based on precipitation,10 and
population demography, as well as other variables.18 Commonly detected WWTP
effluents may be composed of estrogenic compounds,19 disinfection byproducts, and
pharmaceuticals12 and personal care products,1114 as well as heavy metals and industrial
or agricultural20 waste products.
The ecological effects of single compounds and environmentally relevant
mixtures have been investigated in numerous fish species, including Pimephales
promelas (the fathead minnow) and Catostomus commersoni (the white sucker). 21,22 A
preponderance of evidence implicates WWTP contaminants as having adverse effects on
reproduction (including demasculinization and feminization of males),3,23 organismal
fitness, n12 and short- and long-term population viability. 24 Bench-scale examinations
have shown adverse effects in fish exposed to effluent containing environmentally-
relevant concentrations of estrogenic compounds.21,25 Decreased reproduction and
viability often have broader implications for ecosystem health at large, as populations
may undergo an unsustainable loss of numbers over time. 22,26
Estrogens
For this investigation, the effects of estrogenic compounds on the physiology of
the study species are of primary interest. Estrogens may be excreted as products of
natural steroid synthesis in various human and animal tissues, or may be introduced into
the environment from a number of external sources. Estrogens excreted by humans and
2


other animals include estrone (El), estriol (E3), and 17P-estradiol (E2), as shown in
Figure 1.17"19,28 Plants and fungi also contribute estrogens to the environment, though
these compounds are mostly non-steroidal in nature. Structural differences in the
molecular structure of estrogenic compounds alter the affinity the molecule has for the
ER in organism tissues, meaning that estrogenicity and bioavailability differ between
compounds, as well as between species.29
Total contribution of both humans and livestock to receiving water estrogenicity
varies based on season, population density, precipitation, and changes in land use.20
Phytoestrogens estrogens produced by plants may be present in the environment, as
well as mycoestrogens produced by molds.30 Xenoestrogens, or synthetic estrogens, may
include steroidal and non-steroidal estrogens, and are commonly found in aquatic
environments. Among these are ethinylestradiol (EE2, shown in Figure 1) and mestranol
(MeEE2), as well as estrogen-like byproducts resulting from the production of pesticides
(ex. P-DDT, dieldrin, and methoxychlor), plastics and plasticizers (bisphenol A (BPA;
see Figure 1) and phthalates), and paints and detergents (ex. Tributyltin).17 The
breakdown of these same substances often results in novel estrogen-like compounds that
constitute a separate class of substances, including alkylphenolic compounds such as
nonylphenol and octylphenol.31,32 Chemical contaminants from industrial activity, such
as polychlorobiphenyls (PCB), polyaromatic hydrocarbons (PAH), and dioxins, may also
contribute to water estrogenicity.33,34
3


0
OM
nu
MO'
HO'
HO
Estrone (E1)
17(i estradiol (E2)
Estriol (E3)
HO
HO
OH
Ethinylestradiol (EE2)
Bisphenol A (BPA)
Figure 1. Structures of common estrogenic compounds.
Estrogen exposure and reproductive function in aquatic ecosystems
Disruption in reproductive function may be used to determine the presence of
many EDCs in the environment.35 The magnitude of disruption in the study organism,
parameterized by a number of biological endpoints, can give us information about the
concentration and occurrence of EDCs in that organisms environment. 3637 Reproductive
disruption may have implications for social interactions between individual organisms, 26
38 including predator avoidance by larval P. promelas,39'40 as well as broader effects on
the viability of populations in the short- and long-term. 41>42
Reproductive disruption in native fish populations has been observed downstream
from WWTP outfalls in Europe,3> 43 Asia,23 Australia,44 and North America. 2545 Such
disruption may manifest as an overall demasculinization of males, including decreased
gonadal volume, reduced or absent secondary sexual characteristics, reduced sperm
volume or viability, or skewing of sperm developmental ratios. 46,47 In P. promelas
4


specifically, Disruption may also be observed in the form of feminization of males,
indicated by elevated plasma vitellogenin concentrations and, in severe cases, sex
reversal or the appearance of intersex individuals within a population. 224548 The
contamination of river systems by EDCs derived from the discharge of WWTP effluents
mentioned, and has implications for agricultural and drinking water use downstream of
discharge outflows.21
Plasma Vitellogenin (Vtg)
The presence of elevated serum vitellogenin (Vtg) is a highly responsive
biomarker for estrogen exposure and endocrine disruption in oviparous vertebrates, such
as fish, as shown in Figure 2.50,51,52 In male fishes, endocrine disruption of primary
(gonadal intersex) and secondary sexual characteristics may coincide with high measured
plasma Vtg.53,54
Vtg is a phospholipoglyoprotein egg-yolking precursor protein 55 has been used as
can affect the viability and behavior of native fish populations,49 as previously
a biomarker for environmental estrogen
photoperiod
steroid hormones
amino acids
neurotransmitters
exposure in fish.54,5657 Experimental
validation of its use in predicting
estradiol-17fi
reproductive disruption in these animals
has been performed using monoclonal
antibody detection (quantitative ELISA
plasma transport
screens), 5859 quantitative mRNA PCR,
optical waveguide lightmode
Figure 2. Physiological action of an estrogen
(17f-estradiol) in an oviparous vertebrate. 54
5


spectroscopy immunosensor (OWLS) detection,60 and proteomics technologies. 61
Vitellogenin is a high density protein, bears both calcium (Ca) and zinc (Zn) 62
ligands, and serves as a precursor to vitellins (Vn), which function as energy reservoirs
for developing embryos.54 Vtg itself is produced in the livers of both male and female
oviparous animals, 6364 including monotremes. 65
Transport to ovary and
incorporation into oocytes
Figure 3. Cellular responses to estrogens. 54
In fish, sexually mature females synthesize 17P-estradiol (E2) in the gonadal
tissue. 66 E2 circulating in the serum is taken up at hepatocytes, at which point it binds to
the ER, and transcription of the Vtg gene is induced, as shown in Figure 3. 67 The Vtg
gene is also purportedly expressed in non-liver tissues in some species, especially the
skin and eye tissues.68
6


Stream Background
Upgrades and Flood Events
Starting in 2005, a series of long-term, integrated chemical and biological
investigations at the City of Boulder (COB) 75th Street WWTP provided evidence that the
activated sludge upgrade effectively reduced estrogenicity in effluent to a point where
physiological response of the study species P. promelas to estrogenic compounds was no
longer seen in Boulder Creek, downstream of the outflow for the plant. 21,45"69 In 2007,
this WWTP underwent an upgrade in disinfection treatment technology, from trickling
filter to activated sludge, 69 as shown in Figure 4 (A to B). In 2012, another upgrade was
conducted, replacing a chlorination disinfection process (Figure 4, A and B) with a UV
disinfection process (Figure 4, C). Bench-scale experiments suggest that chlorinated and
UV-treated effluents differ in estrogenicity,70,71 however, the impacts of a full-scale
upgrade in disinfection processes have not been evaluated previously at an operational-
scale WWTP.
7


(A) Preupgrade
Nitrification Chlorine
(B) Postupgrade
Activated Sludge
(c) Postupgrade
Activated Sludge
Tank 1
Figure 4. City of Boulder Wastewater Treatment Plant purification system: A.) prior to the 2007
upgrade; B.) 2007 upgrade; and C.) 2012 upgrade.
The purpose of this study was to evaluate the estrogenicity of the COB WWTP
effluent following the 2012 upgrade to UV light disinfection treatment processes. 69
Additionally, we evaluated the impact of the September 2013 1000-year flood on the
estrogenicity of the COB WWTP effluent and on Boulder Creek upstream from the
WWTP effluent outfall.
8


Effluent source and contaminants
Any compounds detected in the COB WWTP effluent are attributable to a
combination of four sources: industrial, commercial, residential, and agricultural
anthropogenic water use within the city of Boulder, Colorado; incidental or atmospheric
deposition of compounds, and the addition of compounds via the disinfection processes
utilized by the COB WWTP itself.
Reproductive disruption in free-living organisms in WWTP effluent
Previous studies indicated reproductive disruption of fish downstream of the
Boulder WWTP outfall. Investigation of reproductive disruption in fish downstream of
the plant was first conducted in 2001 and 2002,22 and again in the fall of 2003 and the
spring of 2004. 2L As shown in Figure 5, mobile lab investigations found a significant
reduction in effluent estrogenicity following the 2007 upgrade (A).
Mobile lab evaluations prior to the upgrade showed disruption in the form of
plasma vitellogenin induction (B), decreased nuptial tubercle number in fathead minnow
males (C), gonadal intersex, skewed sex ratios (17-21% males downstream versus 36-
46% upstream), reduced gonadosomatic index (GSI D), and disrupted ovarian and
testicular histopathology. 7221
9


Ref ESI 50:50 lEff
Date of Exposure Experiment
Date of Exposure Experiment
Figure 5. Measured and observed estrogenic effects in fish exposed on-site to Boulder WWTP
effluent (EFF), pre- and post-2007 upgrade and upstream Boulder Creek reference water (Ref)
and ci 50:50 EffRef mixture. A.) Average WWTP effluent estradiol equivalency quotient (EEq) as
a function of experimental exposure, based on weekly measurement of varied endocrine
disrupting compounds (EDCs); B.) relative plasma vitellogenin (VTG) concentrations
(normalized with respect to the mean Ref concentration) in adult male Pimepliales promelas
exposed to all Ref 50:50 Ref: Eff, and all Eff water for 28 days; C.) nuptial tubercle abundance in
adult male P. promelas exposed under the same conditions; D.) relative gonadosomcitic index
(GSI), normalized (see part B), exposed under the same conditions. Biological results are
expressed as the mean +/- the standard error of the mean (SEM). Bars with an asterisk indicate a
significant difference from the upstream Boulder Creek Ref exposure for the same year (Kruskal-
Wallis test, p < 0.05). n varies from 7-17. ND: not determined. 69
Field Studies
Prior to the activated sludge upgrade, reproductive disruption in native white
suckers (Catostomus commersoni) downstream from the COB WWTP was indicated by
the presence of gonadal intersex and female-biased sex ratios: 83% of individuals
10


collected downstream of the plant were female or intersex, compared to 45% female
upstream, and disrupted testicular and ovarian gametogenesis, as well as elevated plasma
vitellogenin in male fish was reported. 21 22
Mobile Lab Studies
On-site, flow-through experiments conducted following the upgrade to activated
sludge demonstrated an improved removal efficiency for a number of EDCs previously
measured in the WWTP effluent, particularly 17P-estradiol and estrone, as well as a
significantly reduced endocrine disruption in male P. promelas (Figure 6). 69
2.
*,'
Figure 6. Male fathead minnow (Pimcphalcs promelasj with nuptial tubercles (1) and dorsal
fat pad (2) indicated.
11


Figure 7. Factors Contributing to Effluent Estrogenicity: where estrogenicity increases (top
panel) or estrogenicity decreases (bottom panel).
12


Hypothesis
We hypothesized that the implementation of upgrades at the Boulder WWTP in
2012, operating in conjunction with the upgrades installed in 2007, would continue to
maintain acceptable levels of estrogenic compounds in the effluent produced by the
WWTP, barring extreme circumstances (in this case, a major flood event). More
specifically, we hypothesized that the cumulative effects of both upgrades would
counteract the addition of estrogenic compounds to the effluent, as detailed below. We
also hypothesized that a major climactic event, such as the 2013 Boulder flood, would
impact the WWTPs ability to maintain acceptable levels of estrogenic compounds in the
effluent.
1. Effects may be seen when chlorine is removed from the disinfection
process:
a. Removing the chlorination step could result in increased
estrogenicity in the effluent, as many chlorinated estrogenic
compounds have been shown to elicit weaker estrogenic responses
73 than their non-chlorinated counterparts in bench-scale
evaluations.74 74 75 These effects are detailed in Figure 7.
Effects observed in effluent-exposed male fish may take the
form of feminization (elevated plasma vitellogenin
concentrations) and demasculinization (reduced
prominence of the nuptial tubercles and dorsal fat pad,
decreased abundance of nuptial tubercles and spermatozoa
in the tubule, and/or altered sperm developmental stages,
GSI, and CF scores). 73,76
13


b. However, removal of the chlorination step may decrease effluent
estrogenicity attributable to certain compounds, including BP A,11
78 which may show increased binding affinity to the estrogen
receptor (ER) after chlorination. 7479
No evidence will be found to indicate demasculinization or
feminization of male P. promelas in effluent-exposed
individuals, as previously defined.46 The biological
ramifications for this can be seen in the series of
progressions shown in Figure 8.
2. Addition of a UV purification step could also affect the estrogenicity of
the effluent.
a. Adding a UV purification step might improve estrogen removal
from the effluent to a greater extent,51,80 and with fewer estrogenic
byproducts,50 than chlorination purification, as demonstrated in
bench-scale studies81 and experimental WWTPs. 5382
In this case, we would expect that no evidence will be
found to indicate demasculinization or feminization (as
described above) of male fish exposed to WWTP effluent,
83,84 as shown by Figure 8.
b. It has been suggested that some compounds, upon exposure to UV
light, may increase in estrogenicity.85
Demasculinzation and/or feminization of effluent-exposed
male fish may be observed.86
14


3. In the same month that this experiment was performed, a flood event of
rare volume and intensity affected Boulder and its surrounding areas.
a. We hypothesized that this flood event may have had an effect on
the COB WWTPs ability to continuously and efficiently remove
estrogens from the receiving water.
b. This might have manifested as higher levels of runoff or flow may
have deposited estrogenic compounds into the water that would
have otherwise remained in soils, on streets, or even on surfaces of
the plant itself. Additionally, many WWTP disinfection processes
rely on certain retention times that is, how long the water is in
the WWTP and volume.87 Activated sludge, for example, may
be unable to remove compounds if water passes over the substrate
too quickly, or at too great a depth.6 However, these additional
estrogenic compounds could very well be undetectable if water
volume was high enough to provide a dilution effect. This is
diagrammed in Figure 7.
If the flood event prevented the COB WWTP from
effectively removing estrogenic compounds from the
effluent, demasculinization and/or feminization may occur
in effluent-exposed individuals.
If the dilution effect is great enough, or if the WWTP is
able to continue normal operations under extreme flood
15


conditions, little to no demasculinizing or feminizing
effects will be observed.
Figure 8. Demasculinization and Feminization in the Fathead Minnow: a set of biological
endpoints.
16


CHAPTER II
MATERIALS AND METHODS
Study Sites
Boulder Creek, a tributary of the St. Vrain River, has been selected for this
investigation, as it has well-established
chemical, biological, and hydrological
baseline data in advance of a major
operational-scale upgrade in disinfection
processes.21,46 The study site for this
experiment is located at the 75th Street
Wastewater Treatment Facility (4049 N. 75th Street, Boulder, CO 80301; Figure 9). The
plant, which processes water output from Boulder, Colorado, is located in a rural
environment, surrounded by land used largely for agricultural purposes.
Effluent discharged from the city of Boulder WWTP comprises a large proportion
of total stream flow. This WWTP has an average discharge of 0.74 m3 s 1 (17 million
gallons/day) and can contribute from <10% of stream flow during high-flow conditions
(April-July) to >75% of stream flow during low-flow conditions (August-March)
(USGS). The Boulder WWTP uses a combined trickling filter/activated sludge treatment
process with nitrification/denitrification and chlorination/dechlorination.
On-Site Bioassay Laboratory
On-site fish exposure experiments have been conducted at Boulder Creek using a
mobile field laboratory developed for the site (Figure 10). As previously mentioned, the
laboratory is designed to conduct experiments under conditions of controlled
temperature, lighting, diet, aeration, and continuous flow,46 thus incorporating the
17
Figure 9. Boulder Creek, Colorado and the
location of the COB WWTP.


intrinsic variability in the composition of both the WWTP effluent and the receiving
stream. Minimizing variation in flow is especially vital during the high flow periods
characterizing the September flood. All surfaces in contact with test solutions are
composed of glass, stainless steel,
or Teflon. Water from Boulder
Creek and the COB WWTP
effluent were continuously pumped
through Teflon tubing to the
mobile lab using stainless steel
pumps, which empty into 200-L
Figure 10. Fish exposure mobile (FEM) onsite at COB stainless steel holding tanks
WWTP.
positioned above the laboratory.
The two water sources were thermally equilibrated, then flowed by gravity to stainless
steel splitter tanks that distribute the water to 10-L glass aquaria housing the minnows.
Throughout the experiments, water temperature was maintained at 22 1 C under fully
oxygenated (>85% saturation) flow-through conditions. Flow to individual aquaria was
maintained at 200 mL min"1, providing replacement of the water volume approximately
every 4 hours.
Water Sampling and Chemical Analysis
Collection
Water samples are to be collected weekly, for reference and effluent samples, at
Boulder Creek. During the mobile lab aquatic organism exposures on Boulder Creek,
weekly water grab samples will be collected starting at day-0 from the WWTP effluent
18


and the Boulder Creek upstream site inflows into the on-site laboratory to determine the
presence of EDCs.
Chemistry
Water samples were subsequently analyzed for estrogenic compounds using solid
phase extraction, derivatized with N-methyl-N-(trimethylsilyl) trifluoroacetamide
(MSTFA), and GC-MS.88 A variety of neutral endocrine disrupting chemicals, including
4-nonylphenol and 4-octylphenol, their ethoxylated (1-4) oligomers, and bisphenol A,
were also analyzed, using continuous liquid/liquid extraction (CLLE) followed by GC-
MS. 11 In all methods, surrogate standards were added to water samples prior to
extraction to evaluate method performance. Compound identification was based on
matching retention time ( 0.05 min) and ion ratios (3 ions 20%) against authentic
standards, and quantitation was based on isotope dilution and an external calibration
curve.
Tissue Sampling and Analysis
Dissection
Upon their arrival from Aquatic Biosystems (Ft. Collins, CO), 20 male fish were
sampled to constitute an initial control group. 15 males were sampled from reference
(Ref) and effluent (Eff) groups at weeks 4, 6, and 8. The fish were anesthetized by
immersion in ice water and wet weight (g) and length (mm) was recorded. In accordance
with established protocols (US-EPA 2002),89 the condition factor was calculated as
[body weight (g)/total length (mm3) x 100];90 gonadosomatic index was calculated as
[gonad weight (g)/Total Tissue Weight (g) x 100],91 Blood collected from the caudal
vein into heparinized capillary tubes was kept on ice until centrifuged for 5 minute at
19


4500 rpm (within 3 hours of collection). Hematocrit was recorded, and aliquots of plasma
frozen were stored at -40 C until assayed for vitellogenin, as described below.
Anesthetized fish were sacrificed by rapid decapitation across the dorsal aspect of
the spine, just below the head. Gonads were dissected and divided into two portions, and
any gross abnormalities noted. One portion was immediately frozen in a -80F freezer for
future analysis, and the other preserved in 10% neutral-buffered formalin until prepared
for histology. 92
Scoring Secondary Sexual Characteristics
Nuptial tubercle prominence (the degree of definition of the small rostral
protuberances found on sexually mature male fathead minnows) and dorsal fat pad
prominence (the degree of definition of the spongy tissue located along the dorsal portion
of the head and spine) were scored on a scale of 1-4, as shown in Table 1, in a modified
version of methods previously utilized by Parrott et al. and others. 9093 Nuptial tubercles
were also counted males develop no more than 16 tubercles, arranged in three rows.
Maleness the summation of nuptial tubercle prominence and dorsal fat pad prominence
scores was also calculated, using a modified version of the procedure proposed by
Parrott et al.,93 to provide a quantitative measure of demasculinization across secondary
sexual characteristics. The maleness index is especially useful in comparing gonadal size
between males of different sizes and maturational stages. Sperm abundance was also
rated, as shown in Table 1, and sperm developmental stages were evaluated. Sperm
develops through four stages: spermatogonia, spermatocyte, spermatid, and spermatozoa.
We determined the percent composition of each stage in each tubule and reported the
distribution of these developmental stages for each dissected fish.
20


Table 1. Secondary sexual characteristics and sperm abundance rating scales. 90
2 Sex. Char. Rating Scales 1 2 3 4
Nuptial Tubercles Tubercles Tubercles Tubercles
tubercles not visible visible as white discs prominent prominent and protruding sharply
Dorsal fat pad Not visible Soft discolored Spongy Dorsal hump
prominence tissue thickened tissue with spongy tissue
Sperm Sperm Sperm Sperm Sperm
abundance absent prominent in <25% of tubules (weak) prominent in 25- 75% of tubules (moderate) prominent in >75% of tubules (strong)
Quantitative monoclonal antibody detection: ELISA Assay
Detecting Plasma Vitellogenin
Aliquots of plasma frozen and stored at -80 C during each dissection period were
analyzed for Vtg by homologous enzyme-linked immunosorbent assay (ELISA) using an
anti-fathead minnow kit in accordance with the manufacturers protocol (Biosense,
Bergen, Norway).94,95 This assay uses specific vitellogenin-specific antibodies to
quantify vitellogenin in plasma samples. Pre-coated microplate wells utilize a specific
Capture antibody in both the standard and sample wells, and which bind to Vtg. Another
antibody, also specific to Vtg, is the designated detecting antibody, and is labelled with
the enzyme horseradish peroxidase (HRP). This detecting antibody is added to form a
sandwich of Vtg and antibodies. Subsequent enzyme activity is measured by the
addition of a substrate that produces a colored product. Absorbance, measured by
microplate reader, is directly proportional to Vtg concentration in the sample.
For this experiment, we used a single dilution (1:1000) to accommodate this
predicted range of Vtg concentrations, and to avoid possible matrix effects. 96 Matrix
effects are here defined as components of a sample, other than the analyte of interest,
21


which may confound the analysis of that sample and, therefore, the quality of the results
obtained via such analysis.
Dilution and washing buffers were prepared via standard procedures (Biosense,
Bergen, Norway).91 98 The standard curve and all samples were then diluted, added to the
microplate wells, duplicated across each plate, sealed, and incubated at room temperature
for 1.5 hours. Post-incubation, the plates were washed with washing buffer and detecting
antibody (diluted as previously described) was added to the wells. These were then
resealed and incubated again for 0.5 hours, again at room temperature. After this second
incubation, plates were washed again, and tempered TMB substrate solution (pre-
prepared according to standard procedures) was added to the wells. A third incubation
was performed this time, under aluminum foil, though still at room temperature for
twenty minutes. The reaction was then arrested with the addition of H2S04 to all wells, at
which time the plates were placed in the microplate reader to be read at a 450 nm
absorbance.
Histology
Hematoxylin and eosin (H&E) staining is used to clearly delineate tissue
structures.99 We prepared and utilized a set of H&E-stained slides (Figure 11) as
previously described to evaluate sperm abundance, and any gross abnormalities in the
gonadal tissues of male fathead minnows. Sperm abundance in particular is a sensitive
measure for reproductive disruption, as decreased abundance may occur even when other
measures or assays (such as PCNA and TUNEL) show low to no significant disruption. 72
22


Gonad and liver tissues were prepared for histology using standard dehydration
and embedding procedures. We embedded gonadal tissue in paraffin wax and sectioned it
into 5 pm ribbons, which we were subsequently glued and dried onto glass slides on a
Figure 11. From left to right: reference ovary, inter sex testis, and reference testis. Photo credit:
Dr. Alcm Vajdci.
heating element. Multiple ribbons containing different sections of the gonad were glued
onto a series of five glass slides per tissue to ensure that every slide contains multiple
sections of gonad from each individual sampled, and is representative of the gonad as a
whole, as opposed to one specific region. We produced five slides per tissue, and from
each set one slide was designated for hematoxylin and eosin (H&E).
Prior to staining, slides were dewaxed and rehydrated in three xylene solutions at
five minutes each. Slides were then be treated with two solutions each of 100% ethanol,
23


95% ethanol, and 70% ethanol for three minutes in each solution, and subsequently rested
in deionized water for 3 minutes. Following dewaxing and rehydration, slides were
placed in hematoxylin for 45 seconds, and rinsed in tap water until hematoxylin residues
were no longer detectable. Excess liquid was removed, and Permount was applied to
each. The slides were then topped with glass coverslips and air-dried for 24 hours in a
fume hood.
At least 10 cross-sections were evaluated from each gonad for determination of
sex, sperm abundance, intersex status, and any abnormalities.
Statistical Analysis
Scores and ratings for sperm abundance, sperm developmental ratios were
analyzed by Kruskall-Wallis tests. Secondary sexual characteristics, vitellogenin, GSI,
and CF were tested for homoscedasticity and analyzed by a series of one-way ANOVA
tests for effects of the site (reference versus effluent, control versus reference and
effluent, respectively) and sampling period (all weeks compared between one another and
the control). Vitellogenin concentration data were log-transformed prior to analysis. This
was in accordance with methods from previous work at the COB WWTP and accounted
for a lack of homoscedasticity in the data. Statistical significance was accepted &tp <
0.05. Biological results are expressed as a mean (standard error of the mean (SEM)) of
the back-transformed data where necessary.
24


CHAPTER III
RESULTS
Plasma Vitellogenin
Figure 12 shows an upward trend in the plasma vitellogenin (Vtg) of effluent-
exposed fish over this eight week sampling period. While there was no significant
difference in plasma Vtg between reference and effluent fish following four weeks of
exposure (Wk4). A significant difference can be seen between the effluent and reference
treatments during the post flood period in both week six and week eight (P-
value=0.000184). Vtg concentrations in the reference treatment did not differ significantly
from the initial control group (IC) at any sampling period.
Somatic Development and Condition
Figure 13 shows the condition factor (CF) scores for all fish groups, organized by
treatment. Condition factor was found to be significantly different (P -value=0.0007)
between the experimental and control groups when a one-way ANOVA was performed.
Figure 14 shows the normalized gonadosomatic index (GSI) scores for all fish
groups, organized by treatment. Evaluated with one-way ANOVA, GSI was found to be
significantly different between control and experimental groups (P=0.003342), which may
be attributable to natural growth and development of the fish after being acquired at 6
months of age. No statistically significant difference between reference and effluent-
exposed groups could be determined.
Figure 15 length and treatment week shows variation in fish body length over
the experimental period. No statistically significant difference was detected by analysis
with one-way ANOVA (P-value=0.1312460).
25


Secondary Sexual Characteristics
Table 3 shows the Kruskal-Wallis nonparametric test results for maleness (Figure
16), nuptial tubercle prominence, and dorsal fat pad prominence. No statistically
significant difference (P>0.05) was found between the IC and the reference, the IC and
the effluent, or the reference and effluent groups. Table 5 shows the one-way ANOVA
results for nuptial tubercle number (P-value=0.380518).
Sperm Developmental Stages and Abundance
Figure 17 shows the non-parametric distribution of sperm developmental stages
by treatment. In short, this figure shows the degree of development at each stage for each
treatment group as they compare to one another.
Essentially an expansion of figure 17, figure 18 provides a side-by-side comparison
of all four sperm developmental stages as they vary between treatments.
Table 4 displays the statistical analysis performed on sperm developmental stages
over the 8-week testing period. Of the four stages, only spermatocytes showed any
significant differences in effluent-exposed individuals as compared to the control group.
Table 5 displays the statistical analysis (one-way ANOVAs) performed on GSI,
CF, Vtg, NT #, and length over the 8-week testing period.
26


Plasma Vitellogenin
Figure 13. Plasma VTG (in nanograms per microliter) in male fathead minnow (Pimephales
promelas) in an initial control group (1C); reference groups for weeks 4 (RefWk4), 6 (RefWk6),
and 8 (RefWk8); and effluent-exposed groups for weeks 4 (EffWk4), 6 (EffWk6).
Condition Factor by Treatment

1C
RefWk4
E f f W k 4
RefWk6
E f f W k 6
RefWk8
E f f W k 8
T re a tm e n t
Figure 12. Condition factor (CF) by week and treatment within that week: initial controls (IC);
week four, reference (RefWk4); week four, effluent (EffWk4); week six, reference (RefWk6);
week six, effluent (EffWk6); week eight, reference (RefWk8): and week eight, effluent (EffWk8).
27


G o n a d o s o m atic Index
B y T re atm e n t
o
o
T re atm e n t
Figure 14. Gonadosomatic index (GS1) score by week, and treatment within that week:: initial-
controls (IC); week four, reference (RefWk4); week four, effluent (Eff Wk4); week six, reference
(RefWk6); week six, effluent (Eff Wk6); week eight, reference (RefWk8); and week eight, effluent
(EffWkS).
Fish Length by Treatment
80 1
70 -
60 -
50 -
A
A
iS
I I- ? S
V
A
40
S
^ ** ^ ** **
T re atm e n t
* IC
RefWk4
A E ff W k 4
RefWk6
E ff W k 6
RefWk8
E ff W k 8
Figure 15. Length (L, measured in mm) only, grouped by treatment: initial controls (IC),
reference individuals (Ref, and effluent-treated fish (Eff.
28


Maleness Index Scores by Week
A Eff W k8
Ref Wk8
Eff W k6
Ref Wk6
Eff W k4
Ref Wk4 IC
Week
Figure 16. Maleness index scores by treatment: initial controls (IC), reference individuals (Ref,
and effluent-treatedfish (Efif).
Figure 17. Percentage of male fish within a given sperm developmental stage, organized by
treatment: absent (1), weak (2), moderate (3), and strong (4): IC (initial controls), Wk 4 Re f
(Reference), Wk 4 Efif (Effluent), Wk 6 Ref/Eff and Wk 8 Ref/Eff.
29


Spermatogonia Occurence
by Treatm ent
Spermatocyte Occurence
by Treatment
Treatm ent
Spermatid Occurence
by Treatm ent
Spermatozoa Occurence
By Treatm ent
Figure 18. Sperm developmental stages by treatment: spermatogonia (A), spermatocytes (B),
spermatids (C), and spermatozoa (D).
30


Table 2. One-way ANOVAs for Gonadosomatic Index, Condition Factor, Nuptial Tubercle
Number, Fish Length, and Plasma Vitellogenin Concentrations.
GSI
P-value 0.2163
R squared 0.1049
Bartlett's Statistic 1.152
CF NT# Length and Mass Vtg
0.0001* 0.5599 0.1851 0.0002*
0.2413 0.02087 0.08346 0.2731
20.54* 5.934 6.723 178.6
Table 3. Kruskal-Wallis Statistics for secondary sexual characteristics: maleness, nuptial
tubercle prominence (NT Prom.), and dorsal fat pad prominence (DFP Prom.).
Secondary Sex. Char. Maleness NT Prom. DFP Prom.
P value 0.7430 0.3905 0.5424
Kruskal-Wallis Statistic 3.507 1.881 1.223
Table 4. Statistical test (Kruskal-Wallis) results for sperm developmental stages by treatment,
with indicating significance.
Spermatogonia Spermatocytes Spermatids Spermatozoa
P value 0.0918 0.0034* 0.1210 0.8674
31


CHAPTER IV
DISCUSSION
Overview
In this study, we investigated the environmental and toxicological implications of
potentially estrogenic WWTP effluents. This follow-up experiment is the latest in a series
of related work surrounding two upgrades to the COB WWTP. Our aim was to determine
the efficacy of these upgrades in minimizing the exposure of downstream wildlife
populations to estrogenic EDCs.
Hypothesis
Our hypothesis had a multi-part composition regarding the different components
of 2012 upgrade. In the context of effluent chlorination, we hypothesized that removing
the chlorination step in the disinfection process would result in decreased estrogenicity in
the effluent, as demonstrated by previous bench-scale7 75 studies, indicating a weaker physiological response to chlorinated estrogenic
compounds than unchlorinated forms. We tempered this assertion with evidence from
other studies, which have suggested that the removal of the chlorination step may also
introduce additional estrogenicity from certain compounds (ex. Bisphenol-A) and which
may show increased binding affinity to the ER when chlorinated.79 We also suggested
that adding a UV purification step could potentially improve estrogen removal from the
effluent to a greater extent, and with fewer estrogenic byproducts, than chlorination
purification, as demonstrated in bench-scale studies and experimental WWTPs.50- 85 This
assertion was also qualified by data that suggests that certain compounds may increase in
estrogenicity upon exposure to UV.86
32


In short, we hypothesized that removing the chlorination step could result in
feminization (elevated plasma vitellogenin concentrations) and demasculinization
(reduced prominence of the nuptial tubercles and dorsal fat pad, decreased abundance of
nuptial tubercles and spermatozoa in the tubule, and/or altered sperm developmental
stages, GSI, and CF scores) if chlorination was decreasing bioavailability of estrogenic
compounds. This hypothesis would also hold true if the addition of the UV purification
step induced increased binding affinity of estrogenic EDCs to the ER, or if the September
2013 flood event had a significant impact on the ability of the COB WWTP to remove
estrogenic compounds from the effluent. We asserted that if the flood prevented the
WWTP from effectively removing estrogenic compounds from the effluent,
demasculinization and/or feminization would occur in effluent-exposed individuals. If the
dilution from high water volume was great enough, however, or if the WWTP was able to
continue normal operations those conditions, no significant demasculinizing or
feminizing effects would be observed.
Plasma vitellogenin
The results for plasma vitellogenin were perhaps the most telling in this
investigation. Figure 12 shows an induction in plasma Vtg of effluent-exposed individuals
in weeks six and eight of the sampling period. Concentrations in the initial control group
and all the reference groups, and the effluence and reference groups from week four, were
relatively consistent across the sampling period. As seen in Figure 19, the difference
between plasma VTG concentrations in effluent versus reference treatment groups is
insignificant during the fourth week of sampling, but a significant difference can be seen
between the effluent and reference treatments in both week six and week eight right after
flow through the WWTP increased significantly over the mean flow measured through the
33


plant in the previous year (2012). This indicates that any effluent estrogenicity was
contributed by the flood event not the series of upgrades made to the WWTP.
Plasma Vitellogenin
Figure 19. Plasma vitellogenin induction (black and grey) versus measured flow (blue) through the
Boulder WWTP.
The induction of Vtg and the subsequence appearance of the protein in the
bloodstream of the organism can occur in as little as three days after estrogen exposure,
with plasma levels reaching a plateau at about 28 days if exposure remains relatively
constant. Once estrogen exposure has ended, plasma Vtg tends to decrease in concentration
after about three days, and the total time needed for male fish to clear their system of Vtg
is about seven days.100
34


Plasma Vtg concentrations decreased in week eight, which led us to conclude that
the exposure to estrogenic compounds in the effluent was not constant throughout the
sampling period, and dropped off after a period. This would be consistent with the drop off
in flow through in the WWTP after the cessation of the 2013 flood. It seems very likely
that the observed spike in effluent estrogenicity, which appears to have ended after week
six, coincided with increased hydrological flow surrounding the September 2013 flood
event (Figure 19). This would imply that effluent estrogenicity at the COB WWTP
following the 2012 upgrade can be attributed to sources outside of the WWTP itself in
this case, the flood and not to disinfection processes, or the upgrades therein, within the
WWTP, because effluent estrogenicity was not detected before the flood began or after it
ended.
Sperm abundance and development
Sperm abundance (Figure 17), and distribution of sperm throughout the four
developmental stages (Figure 18), appear largely unaffected by any effluent estrogenicity,
with one possible exception. Spermatocytes development appears to be significantly
different between effluent and reference/control groups in weeks six and eight, but no such
difference exists for any of the other stages we examined. Correlations have been found
between p estrogen receptor expression and spermatocyte apoptosis in a mouse model,101
so it is possible that a relationship between decreased ER expression and increased
spermatocyte representation exists. Spermatocytes also tend to last for an extended period
in the testis as a result of the long duration of first meiotic prophase it could be that an
increased spermatocyte representation is related to some disruption in meiotic division.102
35


It is possible that this increase detected in effluent-exposed fish could correlate
with exposure to an unknown EDC with effects on meiotic division or decreasing cell
apoptosis, although this would require additional testing to confirm. There was no
indication of deformation in the gonadal tissues examined histologically, and all other
stages failed to display any significant difference, so this significance in spermatocyte
development in flood effluent-exposed fish most likely has little to no effect on
reproductive function.
Maleness
All measures of maleness, including the maleness score (Figure 16), nuptial
tubercle number, nuptial tubercle prominence, and dorsal fat pad prominence showed no
statistically significant differences between control and treatment groups (Tables 1 and
2). This can be interpreted in one of two ways: first, that no significant estrogen exposure
occurred and, therefore, no alteration was made in these secondary sexual characteristics;
and second, that any estrogen exposure that may have occurred prior to or during the
sampling period wasnt adequate to elicit a change in these characteristics (as it might be
in fish from earlier developmental periods).
Condition factor
While statistically significant, the differences in condition factor (Figure 13)
between the initial control group and the experimental groups can most likely be
attributed to the growth and development of each individual fish over the 8 week long
testing period. Some increase in body size seems likely, as P. promelas only reaches
sexual maturity at 5-6 months of age, and may live as long as 2-3 years, during which
time they continue to gain mass and length;103 the fish used in this experiment were aged
6 months. This is especially probable given that condition factor, which represents the
36


ratio of body mass to body length, is closely related to gonadosomatic index, which
represents the ratio of gonad mass to body mass; GSI was not reported as statistically
significant between control and experimental groups when evaluated by the same
statistical measure, and to the same power (Figure 15). Therefore, we can assume that the
difference is in the length of the tested fish an increase which would represent the
natural growth and development of the organism, rather than any environmental variable.
Estrogens measured at this site over the past several years have showed low,
temporally variable quantities of estrogenic compounds in the effluent; however, further
assessment on the chemistry of the Boulder WWTP effluent during this particular
experimental period has yet to be fully assessed. Maleness appeared to remain largely
unaffected by estrogenic endocrine disruption, with results within the context of
maleness remaining much as they were between the 2007 and 2012 WWTP upgrade
periods. Due to statistically significant results for differences between the initial control
group and treatment groups in the plasma vitellogenin ELISA assay (to be further
discussed; see below), we can assume that estrogens in the effluent were not low enough
to leave the fish completely unscathed.
Because P. promelas seems to respond dramatically to estrogens at early
developmental stages, especially with respect to nuptial tubercle number,104 often to such
an extent that tubercles fail to develop at all. After early development, P. promelas has
been shown to respond to 10-100 nM E2 in a dose-dependent manner: nuptial tubercle
prominence appears to be the most sensitive measure, as they will atrophy at high doses,
while dorsal fat pads inconsistently atrophy at even greater doses.36
37


As effects were not seen in either nuptial tubercle or dorsal fat pad prominence,
we can deduce that the quantity of estrogenic compounds in the Boulder WWTP effluent
was lower than that seen prior to the 2007 upgrade, at which time secondary sexual
characteristics were significantly impacted.
It would appear, then, that effluent estrogenicity is in this case too low to elicit
lasting or population-wide effects on fish appearance or male distribution, as no
significant effects were detected for sperm abundance or secondary sexual characteristics.
This assertion would be strengthened by further investigation, which should include
behavioral and fecundity assays of both male and female organisms. No intersex was
detected, and no extreme morphologies were discovered, which implies that if the results
of the following tests (which were found to be statistically significant) can be attributed
to effluent estrogenicity, that estrogenicity is relatively low. Further, the exposure may
well have been acute, and possibly attributable to sources outside the WWTP itself
perhaps due to the excessive rainfall that occurred in the area during and around the
testing period. It should be noted that the major flood event was the largest recorded in
Boulder, Colorado since a comparable event in 1976, which itself resulted in less rainfall
than the 2013 event.
Synthesis
It can be concluded that effluent estrogenicity is low and temporally variable
following the 2012 upgrade. However, it would appear that some estrogenic compounds
are present in the effluent at higher levels than those seen after the 2007 upgrade. While
this slight uptick could be explained by increased ER binding affinity in compounds that
were previously chlorinated by the COB WWTP, or some compounds which have
38


increased in estrogenicity upon exposure to UV, such conclusions cannot be stated
definitively without further study at this site.
It should be stressed that any future studies would have to be conducted during
periods of drought as well as periods of typical flow, and with an eye to this study, in order
to formulate a clearer picture of how the COB WWTP operates from year to year, and with
significant variation. Increased flow, like that seen during September 2013, may well have
prevented the plant from maintaining consistent retention times, or it may have
overburdened the system as water bypassed the influent and flowed over the surface of the
plant itself. Such a flow may well have washed previously undetected EDCs into the water
column, resulting in an effluent of higher estrogenicity.
Summary
Based on our results, we conclude that the series of upgrades implemented at the
COB WWTP significantly decrease estrogenicity in the effluent the plant produces, and
reduce the effects of estrogenic EDCs on downstream fish populations.
We further conclude that the removal of the chlorination step in the disinfection
process, and its simultaneous replacement with UV treatment, had little to no effect on
increasing effluent estrogenicity after that upgrade was conducted in 2012. We also find
that high flow due to an extreme flood event may have had some effect on the WWTPs
ability to reduce effluent estrogenicity.
Despite these detected increases, we can conclude that current levels of estrogenic
compounds in the COB WWTP are low enough that no current danger is posed to
downstream fish population on the basis of reproductive or endocrine disruption.
39


REFERENCES
1. Liu, Z., Kanjo, Y. & Mizutani, S., Removal mechanisms for endocrine disrupting
compounds (EDCs) in wastewater treatment physical means, biodegradation, and
chemical advanced oxidation: a review. Sci. Total Environ, 407, 731-748 (2009).
2. National Research Council, Issues in Potable Reuse: The Viability of Augmenting
Drinking Water Supplies with Reclaimed Water (1998).
3. Jobling, S. etal, Altered Sexual Maturation and Gamete Production in Wild Roach
(Rutilus rutilus) Living in Rivers That REceive Treated Sewage Effluents. Biology
of Reproduction 66 (2), 272-281 (2002).
4. Head, B., Managing urban water crises: adaptive policy responses to drought in
Southeast Queensland, Australia. Ecology & Society 19 (2), 33 (2014).
5. National Climatic Data Center (NCDC), Drought January 2013 Report (2013).
6. Johnson, A. & Sumpter, J., Removal of endocrine-disrupting chemical sin activated
sludge treatment works. Environmental Science and Technology 35, 4697-4703
(2001).
7. National Weather Service, Hydrologic Information Center: Flood Loss Data,
Available at http://www.nws.noaa.gov/hic/.
8. Karl, T., Melillo, J. & Peterson, T., Global Climate Change Impacts in the United
States, 2009.
9. Strand Associates, Inc, Troubleshooting activated sludge: identifying and
mitigating impacts of drought conditions., presented at Central States Water
Environment Association (86th Annual Meeting), 2013-2014 (unpublished).
10. Barber, L. et al., Chemical loading into surface water along a hydrological,
biogeochemical, and land use gradient a holistic watershed approach.
Environmental Science and Technology 2006, 475-486 (2006).
11. Barber, L., Brown, G. & Zaugg, S., Potential endocrine disrupting organic
chemicals in treated municipal wastewater and river water., presented at In
Analysis of Environmental Endocrine Disruptors, Washington, DC, 2000
(unpublished).
40


12. Kolpin, D. etal., Pharmaceuticals, hormones, and other organic wastewater
contaminants in U.S. streams, 1999-2000 A national reconnaissance.
Environmental Science and Technology 36, 1202-1211 (2002).
13. Ternes, T., Joss, A. & Siegrist, H., Scrutinizing pharmaceuticals and personal care
products in wastewater treatment. Environmental Science and Technology 38,
393A-399A (2004).
14. Sim, W., Lee, J. & Oh, J., Occurence and fate of pharmaceuticals in wastewater
treatment plants and rivers in Korea. Environmental Pollution 158 (5), 1938-1947
(2010).
15. Kloepfer, A., Jekel, M. & Reemtsma, T., Occurrence, sources, and fate of
benxotiazoles in municipal wastewater treatment plants. Environmental Science
and Technology 39 (10), 3792-3798 (2005).
16. VanNuijs, A. eta/., Spatial and temporal variations in the occurence of cocaine
and benzoylecgonine in waste- and surface water from Belgium and removal
during wastewater treatment. Water Research 43 (5), 1341-1349 (2009).
17. Hohne, C. & Puttmann, W., Occurrence and temporal variations of the
xenoestrogens bisphelo A, 4-tert-octylphenol, and tech. 4-nonylphenol in two
German wastewater treatment plants. Environmental Science and Pollution
Research 15, 405-416 (2008).
18. Johnson, A. & Williams, R., A model to estimate influent and effluent
concentrations of estradiol, estrone, and ethinylestradiol at sewage treatment works.
Environmental Science and Technology 38, 3649-3658 (2004).
19. DAscenzo, G. eta/., Fate of natrual estrogen conjugates in municipal sewage
transport and treatment facilities. Science of the Total Environment 302, 199-209
(2003).
20. Lange, I. et al., Sex hormones originating from different livestock production
systems: fate and potential disrupting activity in the environment. Analytica
ChimicaActa 473 (1-2), 27-37 (2002).
21. Vajda, A., Lopez, E., Woodling, J., Maldonado, T. & Norris, D., Reproductive
disruption and intersex in white suckers (Catostomus commersoni) downstream of
a Colorado wastewater treatment plant. Integrative and Comparative Biology 44
(6), 655 (2004).
41


22. Woodling, J., Lopez, E., Maldonado, T., Norris, D. & Vajda, A., Intersex and other
reproductive disruption of fish in wastewater effluent-dominated Colorado streams.
Comparative Biochemistry and Physiology, Part C: Toxicology and Pharmacology
144(1), 10-15 (2006).
23. Ma, T., Wan, X., Huang, Q., Wang, Z. & Liu, J., Biomarker responses and
reproductive toxicity of effluent from a Chinese large sewage treatment plant in
Japanese medaka (Oryzias latipes). Chemosphere 59 (2), 281-288 (2005).
24. Schwindt, A., Winkelman, D., Keteles, K., Murphy, M. & Vajda, A., An
environmental oestrogen disrupts fish population dynamics through direct and
transgenerational effects on survival and fecundity. Journal of Applied Ecology 51
(3), 582-591 (2014).
25. Tetreault, G. et al., Intersex and reproductive impairment of wild fish exposed to
multiple municipal wastewater discharges. Aquatic Toxicology 104 (3-4), 278-290
(2011).
26. Shenoy, K. & Crowley, P., Endocrine disruption of male mating signals: ecological
and evolutionary implications. Functional Ecology 25 (3), 433-448 (2010).
27. Kumar, A. et al., Treated effluent in the aqautic environment: impact assessment of
endocrine disrupting chemicals. Water for a Healthy Country Flagship Report
Series (2012).
28. Johnson, A., Belfroid, A. & Di Corcia, A., Estimating steroid oestrogen inputs into
activated sludge treatment works and observations on their removal from the
effluent. Science of the Total Environment 256, 163-173 (2000).
29. De Sa Salomao, A., Estrogenicity and Genotoxicity Detection in Different
Contaminated Waters. Human and Ecological Risk Assessment (2014).
30. Murkies, A., Wilcox, G. & Davies, S., Clinical review 92 phytoestrogens. The
Journal of Clinical Endocrinology and Metabolism 83, 297-303 (1998).
31. Combalbert, S. & Hernandez-Raquet, G., Occurrence, fate, and biodegradation of
estrogens in sewage and manure. Applied Microbiology and Biotechnology 86 (6),
1671-1692 (2010).
42


32. Lai, K., Johnson, K., Scrimshaw, M. & Lester, J., Binding of waterborne steroid
estrogens to solid phases in river and estuarine systems. Environmental Science and
Technology 34, 3890-3894 (2000).
33. Andersen, H. et al., Comparison of short-term estrogenicity tests for identification
of hormone-disrupting chemicals. Environmental Health Perspectives 107, 89-108
(1999).
34. Datson, G., Cook, J. & Kaylock, R., Uncertainties for endocrine disruptors: our
view on progress. Toxicological Science 74, 245-252 (2003).
35. Scholz, S. & Mayer, I., Molecular biomarkers of endocrine disruption in small
model fish. Science of the Total Environment 293 (1-2), 57-70 (2008).
36. Miles-Richardson, S. et al., Effects of waterbirne exposure of 17B-estradiol on
secondary sex characteristics and gonads of fathead minnows (Pimephales
promelas). Aquatic Toxicology 47 (2), 129-145 (1999).
37. Vajda, A. & Norris, D., Endocrine-active chemicals (EACs) in fishes. Hormones
and Reproduction of Vertebrates: Fishes 1, 245-264 (2011).
38. Norris, D., Vajda, A., Barber, L. & Schoenfuss, H., Impacts of neuroactive and
estrogenic chemicals in wastewater effluents on behavior and reproduction in
freshwater fishes. Integrative and Comparative Biology 52, E130 (2012).
39. McGee, M. et al., Predator avoidance performance of larval fathead minnows
(Pimphales promelas) following short-term exposure to estrogen mixtures. Aquatic
Toxicology 91 (4), 355-361 (2009).
40. Painter, M. et al., Antidepressants at environmentally relevant concentrations affect
predator avoidance behavior of larval fathead minnows (Pimephales promelas).
Environmental Toxicology and Chemistry 28 (12), 2677-2684 (2009).
41. Kime, D. & Nash, J., Gamete viability as an indicator of reproductive endocrine
disruption in fish. Science of the Total Environment 233 (1-3), 123-129 (1999).
42. Bickley, L. et al., Interactive effects of inbreeding and endocrine disruption on
reproduction in a model laboratory fish. Evolutionary Applications 6 (2), 279-289
(2013).
43. Svenson, A., Removal of estrogenicity in Swedish municipal sewage treatment
plants. Water Research 37 (18), 4433-4443 (2003).
43


44. Vajda, A. et al., Integrated assessment of wastewater treatment plant effluent
estrogenicity in the Upper Murray River, Australia, using the native Murray
rainbowfish (Melanotaenia fluviatilis). Environmental Toxicology and Chemistry
(2015).
45. Vajda, A., Bolden, A., Woodling, J., Barber, L. & Norris, D., Reproductive
Disruption by Estrogenic Wastewater Effluents. Endocrine Reviews 31 (3) (2010).
46. Vajda, A. et al., Demasculinization of male fish by wastewater treatment plant
effluent. Aquatic Toxicology 103 (93-94), 213-221 (2011).
47. Kidd, K., McGree, M., Winkelman, D., Vieira, N. & Vajda, A., Reproductive
failure of the red shiner (Cyprinella lutrensis) after exposure to an exogenous
estrogen. Candaian Journal of Fisheries and Aquatic Sciences 67 (11), 1730-1743
(2010).
48. Gray, J., Barber, L., Vajda, A. & Norris, D., Detection of Environmental Estrogens
and Concurrent Observation of Feminization of Fathead Minnows in Waste Water
Effluent, presented at 5th International Conference on Pharmaceuticals and
Endocrine Disrupting Chemicals in Water, Costa Mesa, CA, 2006 (unpublished).
49. Salierno, J. & Kane, A., 17alpha-ethinylestradiol alters reproductive behaviors,
circulating hormones, and sexual morphology in male fathead minnows
(Pimephales promelas). Environmental Toxicology and Chemistry 28 (5), 953-961
(2009).
50. Chen, P. et al., Biological assessment of bisphenol A degradation in water
following direct photolysis and UV advanced oxication. Chemosphere 65, 1094-
1102 (2006).
51. Rosenfeldt, E. & Linden, K., Degradataion of Endocrine Disrupting Chemicals
Bisphenol A, Ethinyl Estradiol, and Estradiol during UV Photolysis and Advanced
Oxidation Processes. Environmental Science and Technology 38, 5476-5483
(2004).
52. Hepell, S., Denslow, N., Folmar, L. & Sullivan, C., Universal assay of vitellogenin
as a biomarker for environmental estrogens. Environmental Health Perspectives
103 (7), 9-15 (1995).
53. Chen, P., Rosenfeldt, E., Kullman, S., Hinton, D. & Linden, K., Biological
assessments of a mixture of endocrine disruptors at environmentally relevant
44


concentrations in water following UV/H202 oxidation. Science of the Total
Environment 376, 18-26 (2007).
54. Denslow, N., Chow, M., Kroll, K. & Green, L., Vitellogenin as a biomarker of
exposure for estrogen or estrogen mimics. Ecotoxicology 8, 385-398 (1999).
55. Arukwe, A. & Goksoyr, A., Eggshell and egg yolk proteins in fish: hepatic proteins
for the next generation. Comparative Hepatology 2 (4) (2003).
56. Cheek, A. eta/., Experimental evaluation of Vitellogenin as a Predictive Biomarker
for Reproductive Disruption. Environmental Health Perspectives 109 (7), 681-690
(2001).
57. Hansen, P. et al., Vitellogenin a biomarker for endocrine disruptors. Trends in
Analytical Chemistry 17 (7), 448-451 (1998).
58. Sumpter, J. & Jobling, S., Vitellogenesis as a biomarker for estrogenic
contamination of the aquatic environment. Environmental Health Perspectives 103
(7), 173-178 (1995).
59. Garcia-Reyero, N. et al., Else of vitellogenin mRNA as a biomarker for endocrine
disruption in feral and culture fish. Analytical andBioanalytical Chemistry 278,
670-675 (2004).
60. Adanyi, N. et al., Optical waveguide lightmode spectroscopy immunosenor for
detection of carp vitellogenin. Sensors and Actuators B: Chemical 176, 932-939
(2013).
61. Om, A., Jasmani, S., Ismail, N., Yeong, S. & Abol-Munafi, A., Application
MALDI TOF on protein identification of vitellogenin in giant grouper (Epinphelus
lanceolatus). Fish Physiology and Biochemistry 39 (5), 1277-1286 (2013).
62. Montorzi, M., Falchuk, K. & Vallee, B., Xenopus laevis vitellogenin is a zinc
protein. Biochemical and Biophysical Research Communications 200, 1407-1413
(1994).
63. Wallace, R., in Developmental Biology: A Comprehensive Synthesis, edited by
Browder, L. (Plenum Press, New York, 1985), Vol. 1, pp. 127-177.
64. Short, S., Yang, G., Kille, P. & Ford, A., Vitellogenin is not an appropriate
biomarker of feminization in a Crustacean. Aquatic Toxicology 153, 89-97 (2014).
45


65. Brawand, D., Wahli, W. & Kaessmann, H., Loss of Egg Yolk Genes in Mammals
and the Origin of Lactation and Placentation. PLoS Biology 6 (3), 63 (2008).
66. Nagler, J., Ruby, S., Idler, D. & So, Y., Serum phosphoprotein phosphorus and
calcium levels as reproductive indicators of vitellogenin in highly vitellogenic
female and estradiol-injected immature rainbow trout (Onchorhynchus mykiss).
Canadian Journal of Zoology 65, 2421-2425 (1987).
67. Nilsen, B. et al., Development of quantitative vitellogenin-ELISAs for fish test
species used in endocrine disruptor screening. Analytical andBioanalytical
Chemistry 387, 621-633 (2004).
68. Zhong, L. et al., Distribution of vitellognin in zebrafish (Danio rerio) tissues for
biomarker analysis. Aquatic Toxicology 149, 1-7 (2014).
69. Barber, L., Vajda, A., Douville, C., Norris, D. & Writer, J., Fish Endocrine
Disruption Responses to a Major Wastewater Treatment Facility Upgrade.
Environmental Science and Technology 46 (4), 2121-2131 (2012).
70. Bourgin, M. etal., Chlorination of bisphenol-A: Non-targeted screening for the
identification of transformation products and assessment of estrogenicity in
generated water. Chemosphere 93, 2814-2822 (2013).
71. Korshin, G., Kim, J. & Gan, L., Comparative study of reactions of endocrine
disruptors bisphenol A and diesthylstilbestrol in electrochemical treatment and
chlorination. Water Research 40 (5), 1070-1078 (2006).
72. Vajda, A. etal., Reproductive Disruption in Fish Downstream from an Estrogenic
Wastewater Effluent. Environmental Science and Technology 42 (9), 3407-3414
(2008).
73. Lee, B. etal., Effects of chlorine on the decrease of estrogenic chemicals. Water
Research 38 (3), 733-739 (2004).
74. Hu, J., Cheng, S., Aizawa, T., Terao, Y. & Kunikane, S., Products of Aqueous
Chlorination of 17B-Estradiol and Their Estrogenic Activities. Environmental
Science and Technology 37, 5665-5670 (2003).
75. Schiliro, T., Pignata, C., Rovere, R., Fea, E. & Gilli, G., The endocrine disrupting
activity of surface waters and of wastewater treatment plant effluents in relation to
chlorination. Chemosphere 75, 335-340 (2009).
46


76. Bolden, A., Vajda, A. & Larber, L., Reproductivedisruption of fishes by endocrine-
active wastewater effluent. Integrative and Comparative Biology 50, El3 (2010).
77. Alum, A., Yoon, Y., Westerhoff, P. & Abbaszadegan, M., Oxidation of Bisphenol
A, 17B-Estradiol, and 17a-Ethynyl Estradiol and Byproduct Estrogenicity.
Environmtnal Toxicology 19 (3), 257-264 (2004).
78. Gallard, EL, Leclerq, A. & Croue, J., Chlorination of bisphenol-A: kinetics and by-
product formation. Chemosphere 56 (5), 465-473 (2004).
79. Hu, J., Aizawa, T. & Ookubo, S., Products of Aqueous Chlorination of Bisphenol
A and Their Estrogenic Activity. Environmental Science and Technology 36, 1980-
1987 (2002).
80. Coleman, H., Routledge, E., Sumpter, J., Eggins, B. & Byrne, J., Rapid loss of
estrogenicity of steroid estrogens by UVA photolysis and photocatalysis over an
immobilized titanium dioxide catalyst. Water Research 38 (14-15), 3233-3240
(2004).
81. Zhang, Y., Zhou, J. & Ning, B., Photodegradation of estrone and 17B-estradiol in
water. Water Research 41, 19-26 (2007).
82. Rosenfeldt, E., Chen, P., Kullman, S. & Linden, K., Destruction of estrogenic
activity in water using UV advanced oxidation. Science of the Total Environment
377, 105-113 (2007).
83. Li Puma, G., Puddu, V., Tsang, H., Gora, A. & Toepfer, B., Photocatalytic
oxidation of multicomponent mixtures of estrogens (estrone (El), 17B-estradiol
(E2), 17a-ethynylestradiol (EE2) and estriol (E3)) under UVA and UVC radiation:
Photon absorption, quantam yields and rate constants independent of photon
absorp. Applied Catalysis B: Environment 99 (3-4), 388-397 (2010).
84. Maniero, M., Bila, D. & Dezotti, M., Degradation and estrogenic activity removal
of 17B-estradiol and 17a-ethinylestradiol by ozonation and 03/H202. Science of
the Total Environment 407 (1), 105-115 (2008).
85. Zhang, Z. eta/., Kinetic degradation model and estrogenicity changes of EE2 (17a-
ethinylestradiol) in aqueous solution by UV and UV/H202 technology. Journal of
Hazardous Materials 181, 1127-1133 (2010).
47


86. Atkinson, S. et al., Environmental Factors Affecting Ultraviolet Photodegradation
Rates and Estrogenicity of Estrone and Ethinylestradiol in Natural Waters. Archives
of Environmental Contamination and Toxicology 60 (1), 1-7 (2011).
87. Anderson, H. et al., Assessment of the importance of sorption for steroid estrogens
removal during activated sludge treatment. Chemosphere 61, 139-146 (2005).
88. Furlong, E. T., Vaught, D. G., Merten, L. M., Foreman, W. T. & Gates, P. M., US
Geological Survey Report No. 95-719, 1995.
89. Ankley, G., Jensen, K., Kahl, M., Korte, J. & Makynen, E., Description and
evaluation of a short-term reproduction test with the fathead minnow (Pimephales
promelas). Environmental Toxicology and Chemistry 20 (6), 1276-1290 (2001).
90. Parrott, J. & Blunt, B., Life-cycle exposure of fathead minnows (Pimephales
promelas) to an ethinylestradiol concentration below 1 ng/L reduces egg
fertilization success and demasculinizes males. Environmental Toxicology 20 (2),
131-141 (2005).
91. Latrouite, D. & Claude, S., Essai de culture suspendue de la coquille Saint Jacques
Pecten maximus influence de la transplantation sur le development gonadique,
presented at International Council for the Exploration of the Sea (ICES),
Copenhagen, Denmark, 1979 (unpublished).
92. Presnell, J. & Schireibman, M., Humason's Animal Tissue Techniques (Johns
Hopkins University Press, Baltimore, Maryland, 1997).
93. Parrott, J., Wood, C., Boutot, P. & Dunn, S., Changes in growth and secondary sex
characteristics of fathead minnows exposed to bleached sulfite mill effluent.
Environmental Toxicology and Chemistry 22 (12), 2908 (2003).
94. Eidem, J., Kleivdal, H. & Goksoyr, A. (Biosense, 2005).
95. Eidem, J. K. H. & Goksoyr, A. (Biosense, 2005).
96. Eurachem, The Fiteness for Purpose of Analytical Methods, Available at
http://www.eurachem.ul.pt/auides/valid.pdf (1998).
97. AO AC, Single Laboratory Validation of Analytical Methods for Dietary
Supplements., presented at International Training Course, Atlanta, GA, 2003
(unpublished).
48


98. IUPAC Technical Report, Harmonized guidelines for single-laboratory validation
of methods of analysis. Pure Applied Chemistry 74 (5), 835-855 (2002).
99. Blazer, V., Histopathological assessment of gonadal tissue in wild fishes. Fish
Physiology and Biochemistry 26 (85), 101 (2002).
100. Schmid, T., Gonzalez-Valero, J., Rufli, H. & Dietrich, D., Determination of
vitellogenin kinetics in male fathead minnows (Pimephales promelas). Toxicology
Letters 131 (1-2), 65-74 (2002).
101. Selva, D. etal, Estrogen Receptor B Expression and Apoptosis of Spermatocytes
of Mice Overexpressing a Rat Androgen-Binding Protein Transgene. Biology of
Reproduction 71 (5), 1461-1468 (2004).
102. Schulz, R. etal., Spermatogenesisin fish. Fish Reproduction 165 (3), 290-411
(2010).
103. Werner, R., Freshwater Fishes of the Northwestern United States: A Field Guide
(Syracuse University Press, Syracuse, NY, 2004).
104. Bogers, R. et al., Development of chronic tests for endocrine active chemicals: Part
2: An extended fish early-life stage test with an androgenic chemical in the fathead
minnows (Pimephales promelas). Aquatic Toxicology 80 (2), 119-130 (2006).
105. Nilsen, B. et al., Development of quantitative vitellogenin-ELISAs for fish test
species used in endocrine disruptor screening. Analytical andBioanalytical
Chemistry 378, 621-633 (2004).
106. Blaise, C., Gagne, F., Salazar, S., Trottier, S. & Hansen, P., Experimentally-
induced feminization of freshwater mussels after long-term exposure to a municipal
effluent. Fresenius Environmental Bulletin 12, 865-870 (2003).
49


Full Text

PAGE 1

ENDOCRINE DISRUPTION IN THE FATHEAD MINNOW ( PIMEPHALES PROMELAS ) FOLLOWING A SERIES OF UPGRADES TO A WASTEWATER TREATMENT FACILITY by ANGELINA FREE BAROFFIO B.S., Tulane University of Louisiana, 2012 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of M aster of Science Integrative Biology Program 2015

PAGE 2

ii 2015 ANGELINA BAROFFIO ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Science degree by Angelina Free Baroffio has been approved for the Integrative Biology Program by Alan Vajda, Chair Michael Greene Kristen Keteles October 30 th 2015

PAGE 4

iv Baroffio, Angelina Free (M.S., Integrative Biology) Endocrine disruption in the fathead minnow ( Pimphales promelas ) following a series of upgrades to a wastewater treatment facility. Thesis directed by Assistant Professor Alan Vajda. ABSTRACT This study a imed to characterize the impact of treatment infrastructure upgrades on the occurrence of endocrine disrupting chemicals (EDCs) discharged by a wastewater treatment plant (WWTP) effluent and their subsequent effects on fish endocrine function. This site ha s been evaluated before and after two major upgrades in wastewater treatment infrastructure, which were implemented in 2007 and 2012. Our study assessed the potential impacts on the Boulder Creek receiving water, as well as identified and evaluated the ext ent of estrogenic endocrine disruption in the native fathead minnow ( Pimephales promelas ) that may be occurring after the implementation of the 2012 upgrade. We conducted an integrative, 8 week, on site, continuous flow, exposure experiment using adult mal e fathead minnows to assess in vivo estrogenicity of the WWTP effluent water, relative to reference water and results from prior years (both pre and post upgrade). We collected data for a wide array of biological endpoints. Results for plasma vitellogenin concentrations and sperm developme nt were emphasized. It was found that in vivo effluent estrogenicity following the 2012 upgrade was insignificant in comparison to pre upgrade levels. However, the occurrence of an extreme flood event in the Boulder area resulted in the detection of some s ignificant effluent estrogenicity, indicating that such conditions may impact the ability of WWTPs to effectively remove estrogenic EDCs from the effluent.

PAGE 5

v The form and content of this abstract are approved. I recommend its publication. Approved: Alan Vajd a

PAGE 6

vi ACKWNOWLEDGEMENTS I would like to thank my advisor, Dr. Alan Vajda, and the members of my thesis committee, Dr. Michael Greene and Dr. Kristen Keteles, for their help and patience as I completed this thesis. I would also like to thank the University of Colorado at Denver and Dr. Michael Wunder, USGS and Dr. Larry Barber, and all of those people who helped make this research possible with all their hard w ork in our lab, including Chelsea Ladd, Alex andra Harrison, Gary Broyles, Zia Faizi, Munira Lantz, Kendra Occhipinti, and Reyna Griffin Special thanks are also due to the City of Boulder, for allowing us to do this work on the premises of their wastewater treatment plant. And, of course, all my gratitude and love to my family and my friends.

PAGE 7

vii TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ........... 1 Wastewater Treatment Plants: Engineering and Technology ................................ ......... 1 Endocrine Disruption and Endocrine Disrupting Compounds ................................ ....... 1 Estrogens ................................ ................................ ................................ ..................... 2 Estrogen exposure and reproductive function in aquatic ecosystems ......................... 4 Plasma Vitellogenin (Vtg) ................................ ................................ ........................... 5 Stream Background ................................ ................................ ................................ ......... 7 Upgrades and Flood Events ................................ ................................ ......................... 7 Effluent source and contaminants ................................ ................................ ............ 9 Reproductive disruption in free living organisms in WWTP effluent ........................ 9 Field Studies ................................ ................................ ................................ ........... 10 Mobile Lab Studies ................................ ................................ ................................ 11 Hypothesis ................................ ................................ ................................ ................. 13 II. MATERIALS AND METHODS ................................ ................................ ................. 17 Study Sites ................................ ................................ ................................ ..................... 17 On Site Bioassay Laboratory ................................ ................................ ..................... 17 Water Sampling and Chemical Analysis ................................ ................................ ... 18 Collection ................................ ................................ ................................ ........... 18 Chemistry ................................ ................................ ................................ ........... 19 Tissue Sampling and Analysis ................................ ................................ ............... 19

PAGE 8

viii Dissection ................................ ................................ ................................ ........... 19 Scoring Secondary Sexual Characteristics ................................ ............................. 20 Quantitative monoclonal antibody detection: ELISA Assay ................................ 21 Detecting Plasma Vitellogenin ................................ ................................ ........... 21 Histology ................................ ................................ ................................ ................ 22 Statistical Analysis ................................ ................................ ................................ .... 24 III. RESULTS ................................ ................................ ................................ ................... 25 Plasma Vitellogenin ................................ ................................ ................................ ...... 25 Somatic Development and Condition ................................ ................................ ........... 25 Secondary Sexual Characteristics ................................ ................................ ................. 26 Sperm Developmental Stages and Abundance ................................ .............................. 26 IV. DISCUSSION ................................ ................................ ................................ ............. 32 Overview ................................ ................................ ................................ ....................... 32 Hypothesis ................................ ................................ ................................ ................. 32 Plasma vitellogenin ................................ ................................ ............................ 33 Sperm abundance and development ................................ ................................ ... 35 Maleness ................................ ................................ ................................ ............. 36 Condition factor ................................ ................................ ................................ .. 36 Synthesis ................................ ................................ ................................ ........................ 38 Summary ................................ ................................ ................................ ....................... 39 References ................................ ................................ ................................ ......................... 40

PAGE 9

ix LIST OF TABLES TABLE 1. Secondary sexual characteristics and sperm abundance rating scales. ........................ 21 2. One way ANOVAs for Gonadosomatic I ndex, Condition Factor, Nuptial Tubercle Number, Fish Length, and Plasma Vitellogenin Concentrations. ............................. 31 3.Statistical test results for secondary sexual characteristics: maleness, nuptial tubercle prominence, and dorsal fat pad prominence ................................ ............................ 31 4. Statistical test results for sperm developmental stages by treatment. ........................... 31

PAGE 10

x LIST OF FIGURES FIGURE 1. Structures of common estrogenic compounds. ................................ ............................... 4 2. Phys iological action of an estrogen in an oviparous vertebrate. ................................ ..... 5 3. Cel lular responses to estrogens. ................................ ................................ ...................... 6 4. City of Boulder Wastewater Treatment Plant purification system. ................................ 8 5. Measured and observed estrogenic effects in fish exposed on site to Boulder WWTP effluent pre and post 2007 upgrade and upstream Bou lder Creek reference water and a 50:50 effluent/reference mixture ................................ ................................ .... 10 6. Male fathead minnow ( Pimephales promelas ). ................................ ............................ 11 7. Factors Contributing to Effluent Estrogenicity: where estrogenicity increases or estrogenicity decreases ................................ ................................ ............................. 12 8. Demasculinization and Feminization in the Fathead Minnow: a set of biological endpoints. ................................ ................................ ................................ ................... 16 9. Boulder Creek, Colorado and the location of the COB WWTP. ................................ .. 17 10. Fish exposure mobile onsite at COB WWTP. ................................ ............................ 18 11. From left to right: reference ovary, inters ex testis, and reference testis .................... 23 12. Plasma VTG in male fathead minnow ( Pimephales promelas ) in an initial control group reference groups and effluent exposed groups. ................................ ............ 27 13 Condition factor by week and treatment within that week. ................................ ........ 27 14. Gonadosomatic index score by week, and treatment within that week ..................... 28

PAGE 11

xi 15. Length only, grouped by treatment: initial controls, reference individuals and effluent treated fish ................................ ................................ ................................ .. 28 16. Maleness index scores by treatment: initial controls reference individuals and effluent treated fish ................................ ................................ ................................ .. 29 17. Percentage of male fish within a given sperm developmental stage, organized by treatment. ................................ ................................ ................................ ................... 29 18. Sperm developmental stages by treatment: spermatogonia, spermatocytes, spermatids, and spermatozoa ................................ ................................ ................................ ....... 30 19 Plasma vitellogenin induction versus measured flow through the Boulder WWTP .. 34

PAGE 12

xii LIST OF ABBRIEVIATIONS BPA Bisphenol A COB City of Boulder EDC Endocrine Disrupting Compound E1 Estrone Ethynylestradiol Ethynylestradiol EE2 Ethinyl Estradiol E3 Estriol VTG Vitellogenin WWTP Wastewater Treatment Plant

PAGE 13

1 CHAPTER I INTRODUCTION Wastewater Treatment Plants: Engineering and Technology Human management of wastewater has made significant strides in terms of sustainability and effluent quality from wastewater treatment plants (WWTPs) by employing numerous types of upgrades to traditional WWTP designs across the US and around the world. 1 The efficacy of WWTPs in removing small bio active molecules in the resulting effluent varies with treatment technology and has implications for downstream human 2 and ecosystem health 3 Hydrologic variation, due to seasonality and, in more extreme cases, drought 4 5 or flood 6 7 can impact WWTP plant efficiency 8 9 and the dilution of WWTP effluents in total stream volume 10 Implementation of appropriate WWTP treatment technologies is complicated by the often conflicting priorities of engineers, managers environmental specialists and other stakeholders, regulatory requirements, 11 and the availability of funding for research, and infrastructure upgrade and WWTP operation costs. 12 13 We evaluated major upgrades in WWTP disinfection processes to determine the efficacy of such upgrades in minimizing the exposure of downstream wildlife populations to estrogenic that is, interacting with estrogen receptors (ER) and eliciting a response at the cellular, tissue, or beha vioral levels endocrine disrupting compounds ( EDCs ). Endocrine Disruption and Endocrine Disrupting Compounds The concentration and composition of chemical contaminants in WWTP effluents changes over time in response to a number of different factors. 14 Most WWTP effluents are composed of highly complex mixtures, 15 as WWTPs process influent from a broad range of domestic and industrial sources such as cities, manufacturing plants, and even

PAGE 14

2 animal feedlots 16 The complexity of these WWTP effluent chemical compositions varies temporally 17 due to short and long term variation in chemical usage anthropogenic land use, WWTP design (ex. c hlorine /decholorination versus UV light and advanced oxidation disinfection treatments) and efficiency, hydrologic flow based on precipitation 10 and population demography as well as other variables. 18 Commonly detected WWTP effluents may be composed of estrogenic compounds 19 disinfection byproducts, and pharmaceuticals 12 and personal care products, 13 14 as well as heavy metals and industrial or agricultural 20 waste products T he ecological effects of single compounds and environmentally relevant mixtures have been investigated in numerous fish species including Pimephales promelas (the fathead minnow) and Catostomus commersoni (the white sucker) 21 22 A preponderance of evidence implicates WWTP contaminants as having adverse effects on reproduction (including demasculinization and feminization of males) 3 23 organism al fitness, 11 12 and short and long term population viability 24 Bench scale examinations have shown adverse effects in fish exposed to effluent containing environmentally relevant concentrations of estrogenic compounds 21, 25 Decreased reproduction and viability often have broader implications for ecosys tem health at large as populations may undergo an unsustainable loss of numbers over time 22 26 Estrogens For this investigation, the effects of estrogenic compounds on the physiology of the study species are of primary interest. Estrogens may be excreted as products of natural steroid synthesis in various human and animal tissues or may be introduced into the environment from a number of external sources. Estrogens excreted by humans and

PAGE 15

3 other animals i nclude estrone (E1), estriol (E3), and estradiol (E2) as shown in Figure 1 17 19 28 Plants and fungi also contribute estrogens to the environment, though these compounds are mostly non steroidal in nature. Structural differences in the molecular structure of estrogenic compounds alter the affinity the molecule has for the ER in organism tissues, meaning that estrogenicity and bioavailability differ between compounds as well as between species 29 Total contribution of both humans and livestock to receiving water estrogenicity varies based on season, population density, precipitation, and changes in land use. 20 Phytoestrogens estrogens produced by plants may be present in the environment, as well as mycoestrogens produced by molds. 30 Xenoestrogens, or synthetic estrogens may include steroidal and non steroidal estrogens, and are commonly found in aquatic environments Among these are ethinylestradiol (EE2 shown in Figure 1 ) and mestranol (MeEE2), as well as estrogen like byproducts resulting from the production of pesticides (ex. P DDT, dieldrin, and methoxychlor), plastics and plasticizers (bisphenol A (BPA ; see Figure 1 ) and phthalates), and paints and detergents (ex. Tri butyltin). 17 The breakdown of these same substances often result s in novel estrogen like compounds that constitute a separate class of substances including alkylphenolic compounds such as nonylphenol and octylphenol 31 32 Chemical contaminants from industrial activity such as polychlorobiphenyls (PCB), polyaromatic hydrocarbons (PAH), and dioxins may also contribute to water estrogenicity. 33 34

PAGE 16

4 Figure 1 Structures of common estrogenic compounds. Estrogen exposure and reproductive function in aquatic ecosystems Disruption in reproductive function may be u sed to determine the presence of many EDCs in the environment. 35 T he magnitude of disruption in the study organism parameterized by a number of biological endpoints, can gi ve us information about the concentration 36 37 Reproductive disruption may have implications for social interactions between individual organisms, 26 38 including predator avoidance by larval P. promelas 39 40 as well as broader effects on the viability of populations in the short and long term. 41 42 Reproductive disruption in native fish populations has been observed downstream from WWTP outfalls in Europe, 3 43 Asia, 23 Australia, 44 and North America. 25 45 Such disruption may manifest as an overall demasculinization of males, including decreased gonadal volume, reduced or absent secondary sexual characteristics, reduced sperm volume or viability, or skewin g of sperm developmental ratios. 46 47 In P. promelas

PAGE 17

5 specifically, Disruption may also be observed in the form of f eminization of males indicated by elevated plasma vitelloge nin concentrations and, in severe cases, sex reversal or the appearance of intersex individuals within a population. 22 45 48 The contamination of river systems by EDCs derived from the discharge of WWTP efflu ents can affect the viability and behavior of native fish populations 49 as previously mentioned, and has implications for agricultural and drinking water use d ownstream of discharge outflows. 21 Plasma Vitellogenin (Vtg) The presence of elevated serum vitellogenin (Vtg) is a highly responsive biomarker for estrogen exposure and endocrine disruption in oviparous vertebrates, such as fish as shown in Figure 2 50 51 52 In male fishes, endocrine disruption of primary (gonadal intersex) and secondary sexual c haracteristics may coincide with high measured plasma Vtg. 53 54 Vtg is a phospholipoglyoprotein egg yolking precursor protein 55 has been used as a biomarker for environmental estrogen exposure in fish. 54 56 57 Experimental validation of its use in predicting reproductive disruption in these animals has been performed using monoclonal antibody detection (quantitative ELISA screens), 58 59 quantitative mRNA PCR, optical waveguide lightmode Figure 2 Physiological action of an estrogen estradiol) in an oviparous vertebrate. 54

PAGE 18

6 spectroscopy immunosensor ( OWLS) detection, 60 and proteomics technologies. 61 Vitellogenin is a high density protein, bears both calcium (Ca) and zinc (Zn) 62 ligands, and serves as a precursor to vitellins (Vn), which function as energy reservoirs for developing embryos. 54 Vtg itself is produced in the livers of both male and female oviparous animals, 63 64 including monotremes. 65 Figure 3 Cellular responses to estrogens. 54 estradiol (E2) in the gonadal tissue. 66 E2 circulating in the serum is taken up at hepatocytes, at which point it binds to the ER, and transcription of the Vtg gene is induced, as shown in Figure 3 67 The Vtg gene is also purportedly expressed in non liver tissues in some species, especially the skin and eye tissues. 68

PAGE 19

7 Stream Background Upgrades and Flood Events Starting in 2005, a series of long term, integrated chemical and biological investigations at t he City of Boulder (COB) 75 th Street WWTP provided evidence that the activated sludge upgra de effectively reduced estrogeni city in effluent to a point where physiological response of the study species P promelas to estrogenic co mpounds was no longer seen in Boulder Creek, downstream of the outflow for the plant. 21 45 69 In 2007 this WWTP underwent an upgrade in disinfection treatment technology, from trickling filter to activated sludge, 69 as shown in Figure 4 (A to B). In 2012, another upgrade was conducted, replacing a chlorin ation disinfection process (Figure 4 A and B) with a UV disinfection process (Figure 4 C). Bench scale experiments suggest that chlorinated and UV treated effluents differ in estrogenicity, 70 71 however, the impacts of a full scale upgrade in disinfection processes have not been evaluated previously at an operational scale WWTP.

PAGE 20

8 Figure 4 City of Boulder Wastewater Treatment Plant purification system: A.) prior to the 2007 upgrade; B.) 2007 upgrade; and C.) 2012 upgrade. The purpose of this study was to evaluate the estrogenicity of the COB WWTP effluent following the 2012 upgrade to UV light disinfection treatment processes. 69 estrogenicity of the COB WWTP effluent and on Boulder Creek upstream from the WWTP effluent outfall.

PAGE 21

9 Effluent source and contaminants Any compounds detected in the COB WWTP effluent are attributable to a combination of four sources: industrial, commercial, residential, and agricultural anthropogenic water use within the city of Boulder, Colorado; incidental or atmospheric deposition of compounds, and the addition of compounds via the disinfection processes utilized by the COB WWTP itself. Reproductive disruption in fr ee living organisms in WWTP effluent Previous studies indicated reproductive disruption of fish downstream of the Boulder WWTP outfall. Investigation of reproductive disruption in fish downstream of the plant was first conducted in 2001 and 2002, 22 and again in the fall of 2003 and the spring of 2004. 21 As shown in Figure 5 mobile lab investigations found a significant reduction in effluent estrogenicity following the 2007 upgrade (A). Mobile lab evaluations prior to the upgrade showed disruption in the form of plasma vitellogenin induction (B ), decreased nuptial tubercle number in fathead minnow males (C), gonadal intersex, skewed sex ratios (17 21% males downstream versus 36 46% upstream), reduced gon adosomatic index (GSI D), and disrupted ovarian and testicular histopathology. 72 21

PAGE 22

10 Figure 5 Measured and observed estrogenic effects in fish exposed on site to Boulder WWTP effluent (EFF), pre and post 2007 upgrade and upstream Boulder Creek reference water (Ref) and a 50:50 Eff:Ref mixture. A.) Average WWTP effluent estradiol equivalency quot ient (EEq) as a function of experimental exposure, based on weekly measurement of varied endocrine disrupting compounds B.) relative plasma vitellogenin (VTG) concentrations (normalized with respect to the mean Ref concentration) in adult male Pi mephales promelas exposed to all Ref, 50:50 Ref:Eff, and all Eff water for 28 days; C.) nuptial tubercle abundance in adult male P. promelas exposed under the same conditions; D.) relative gonadosomatic index (GSI), normalized (see part B), exposed under t he same conditions. Biological results are expressed as the mean +/ the standard error of the mean (SEM). Bars with an asterisk indicate a significant difference from the upstream Boulder Creek Ref exposure for the same year (Kruskal Wallis test, p < 0.05 ). n varies from 7 17. ND: not determined. 69 Field Studies Prior to the activated sludge upgrade, reproductive disruption in native white suckers (Catostomus commersoni) downstream from the COB WWTP was indicated by the presence of gonadal intersex and female biased sex ratios: 83% of individuals

PAGE 23

11 collected downstream of the plant were female or intersex compared to 45% female upstream, and disrupted testicular and ovarian gametogenesis, as well as elevated plasma vitellogenin in male fish was reported. 21 22 Mobile Lab Studies On site, flow through experiments conducted following the upgrade to activated sludge demonstrated an improved removal efficiency for a number of EDCs previously estradiol and estrone, as well as a significantly reduced endocrine disruption in male P. p romelas (Figure 6 ). 69 Figure 6 Male fathead minnow ( Pimephales promelas ) with nuptial tubercles (1) and dorsal fat pad (2) indicated. 1. 2

PAGE 24

12 Figure 7 Factors Contributing to Effluent Estrogenicity: where estrogenicity increases (top panel) or estrogenicity decreases (bottom panel). Increased Effluent Estrogenicity Some compounds are more bioactive when unchlorinated or when exposed to UV. Removal of chlorination step Addition of UV Step Additional compounds may be added within the plant. Flood Decreased Effluent Estrogenicity Some compounds are less bioactive when chlorinated, or when exposed to UV. Removal of chlorination step Addition of UV step Any added compounds have too small a concentration to increase overall estrogenicity. Flood

PAGE 25

13 Hypothesis We hypothesize d that the implementation of upgrades at the Boulder WWTP in 2012, operating in conjunction with the upgrades installed in 2007, would continue to maintain acceptable levels of estrogenic compounds in the effluent produced by the WWTP, barring extreme circumstan ces (in this case, a major flood event). More specifically, we hypothesized that the cumulative effects of both upgrades would counteract the addition of estrogenic compounds to the effluent, as detailed below. We also hypothesized that a major climactic e vent, such as the 2013 Boulder flood, would effluent. 1. Effects may be seen when chlorine is removed from the disinfection process: a. Removing the chlorination step could re sult in increased estrog enicity i n the effluent, as many chlorinated estrogenic compounds have been shown to elicit weaker estrogenic responses 73 than their non chlorinated counterparts in bench scale evaluations. 70 74 75 These effects are detailed in Figure 7. Effects observed in effluent e xposed male fish may take the form of feminization (elevated plasma vitellogenin concentrations) and demasculinization (reduced prominence of the nuptial tubercles and dorsal fat pad, decreased abundance of nuptial tubercles and spermatozoa in the tubule, and/or altered sperm developmental stages, GSI, and CF scores) 73 76

PAGE 26

14 b. However, r emoval of the chlorination step may decrease effluent estrogenicity attributable to certain compounds, including BPA, 77 78 which may show increased binding affinity to the estrogen receptor (ER) after chlorination. 74 79 No evidence will be found to indicate demasculinization or feminization of male P promelas in effluent exposed individuals as previously defined. 46 The biological ramifications for this can be seen in the series of progressions shown in Figure 8. 2. Additio n of a UV purification step could also affect the estrogenicity of the effluent. a. A dding a UV purification step might improve estrogen removal from the eff luent to a greater extent, 51 80 and with fewer estrogenic byproducts, 50 than chlorination purification, as demonstrated in bench scale studies 81 and experimental WWTPs. 53 82 In this case, we would expect that no evidence will be found to indicate demasculinization or feminization (as described above) of male fish exposed to WWTP effluent 83 84 as shown by Figure 8. b. It has been suggested that some compounds, upon exposure to UV light may increase in estrogenicity. 85 Demasculinzation and/or feminization of effluent exposed male fish may be observed. 86

PAGE 27

15 3. In the same month that this experiment was performed, a flood event of rare volume and intensity affected Boulder and its surrounding areas. a. We hypothesized that this flood event may have had an effect on ove estrogens from the receiving water. b. This might have manifested as higher levels of runoff or flow may have deposited estrogenic compounds into the water that would have otherwise remained in soils, on streets, or even on surfaces of the plant itself. A dditionally, many WWTP disinfection processes rely on certain retention times that is, how long the water is in the WWTP and volume. 87 Activated sludge, for example, may be unable to remove compounds if water passes over the substrate too quickly, or at too great a depth. 6 However, these additional estrogenic compounds could very well be undetectable if wate r volume was high enough to provide a dilution effect. This is diagrammed in Figure 7. If the flood event prevented the COB WWTP from effectively removing estrogenic compounds from the effluent, demasculinization and/or feminization may occur in effluent e xposed individuals. If the dilution effect is great enough, or if the WWTP is able to continue normal operations under extreme flood

PAGE 28

16 conditions, little to no demasculinizing or feminizing effects will be observed. Figure 8 Demasculinization and Feminization in the Fathead Minnow: a set of biological endpoints. Exposure to Estrogenic Compounds Demasculinization Effects on Secondary Sexual Characteristics Fewer Nuptial Tubercles Decreased Prominence of Nuptial Tubercles, Dorsal Fat Pad Effects on Sperm Abundance and Development Reduced Sperm Abundance Skewed Sperm Developmental Ratios Effects on Gonadosomatic Index, Condition Factor Reduced GSI and CF Feminization Induction of Vitellogenin Elevated Plasma Vitellogenin Concentrations

PAGE 29

17 CHAPTER II MATERIALS AND METHODS Study Sites Boulder Creek, a tribut ary of the St. Vrain River h as been selected for this investigation, as it has well established chemical, biological, and hydrological baseline data in advance of a major operational scale upgrade in disinfection processes. 2 1 46 The study site for this experiment is located at the 75 th Street Wastewater Treatment Facility (4049 N. 75 th Street, Boulder, CO 80301 ; Figure 9 ). The plant, which processes water output from Boulder, Colorado, is located in a rural environment, surrounded by land used largely for agricultural purposes. E ffluent discharged from the city of Boulder WWTP comprises a large proportion of total stream flow. This WWTP has an average discharge of 0.74 m 3 s (17 million gallons/day) and can contribute from <10% of stream flow during high flow conditions (USGS). The Boulder WWTP uses a combined trickling filter/activated sludge treatment process with nitrification/denitrification a nd chlorination/dechlorination. On Site B ioassay L aboratory On sit e fish exposure experiments have been conducted at Boulder Creek using a mobile field laboratory developed for the site (Figure 10) As previously mentioned, the laboratory is designed to conduct experiments under conditions of controlled temperature, lighting, diet, aeration, and continuous flow 46 thus incorporating the Figure 9 Boulder Creek, Colorado and the location of the COB WWTP.

PAGE 30

18 intrinsic variability in the composition of both the WWTP effluent and the receiving stream. Minimizing variation in flow i s especially vital during the high flow periods characterizing the September flood. All surfaces i n contact with test solutions a re composed of glass, stainless steel, or Teflon. Water from Boulder Cr eek and the CO B WWTP effluent were continuously pumped through Teflon tubing to the mobile lab using stai nless steel pumps, which empty into 200 L stainless steel holding tanks positioned above the labo ratory. The two water sources we re the rmally equilibrated, then flowed by gravity to stainless steel splitter tanks that distribute the water to 10 L glass aquaria housing the minnows. Throughout the e xperi ments, water temperature was maintained at 22 1 C under fully oxygenated (>85% satura tion) flow through conditions Flo w to individual aquaria was maintained at 200 mL min 1 providing replacement of the water volume approximately every 4 hours. Water Sampling and Chemical Analysis Collection Water samples are to be collected weekly, for reference and effluent samples, at Boulder Creek. During the mobile lab aquatic organism exposures on Boulder Creek, weekly water grab samples will be collected starting at day 0 from the WWTP effluent Figure 10 Fish exposure mobile (FEM) onsite at COB WWTP.

PAGE 31

19 and the Boulder Creek upstream site inflows into the on site laboratory to determine the presence of EDCs. Chemistry Water s amples were subsequently analyzed for estrogenic compounds using solid phase extraction, derivatized with N methyl N (trimethylsilyl) trifl u oroacetamide (MSTFA), and GC MS. 88 A variety of neutral endocrine disrupting chemicals including 4 nonylphenol and 4 octylphenol, their ethoxylated (1 4) oligomers, and bisphenol A were also analyzed, using continuous liquid/liquid extraction (CLLE) followed by GC MS 11 In all m ethods, surrogate standards were added to water samples prior to extraction to evaluate method performance. Compound identific ation was based on ma tching retention time ( 0.05 min) and ion ratios (3 ions 20%) against authentic standards, and quantitation was based on isotope dilution and an external calibration curve. Tissue Sampling and Analysis Dissection Upon their arrival from Aquatic Biosystems (Ft. Collins, CO), 20 male fish were sampled to constitute an initial control group. 15 males were sampled from reference (Ref) and effluent (Eff) groups at weeks 4, 6, and 8. The f ish were anesthetized by immersi on in ice water and wet we ight (g) and length (mm) was re cord ed. In accor dance with establish ed protocols (US EPA 2002) 89 t he condition factor was calculated as [body weight (g)/total length (mm 3 ) 100] ; 90 gonadosomatic index was calculated as [gonad weight (g)/Total Tissue Weight (g) 100] 91 Blood collected from the caudal vein into heparinized capillary tube s was kept on ice until centrifuged for 5 minute at

PAGE 32

20 4500 rpm (within 3 hours of collection). Hematocrit was recorded, a nd aliquots of plasma frozen were sayed for vitellogenin, as described below. Anesthetized fish we re sacrificed by r apid decapitation across the dorsal aspect of the spine, jus t below the head. Gonads were dissected and divided into two portions, a nd any gross abnormal ities noted. One portion was immediately frozen in a 80 F freezer for future analysis, and the other p reserved in 10% neutral buffered formalin until prepared for histology 92 Scoring Secondary Sexual Characteristics Nuptial tubercle prominence (the degree of definition of the small rostral protuberances found on sexually mature male fathead minnows) and dorsal fat pad prominence (the degree of definition of the spongy tissue located along the dorsal portion of the head and spine) were scored on a sca le of 1 4, as shown in Table 1 in a modified version of methods previously utilized by Parrott et al. and others. 90 93 Nuptial tubercles were also counted males develop no more than 16 tubercles, arranged in three rows. Maleness the summation of nuptial tubercle prominence and dorsal fat pad prominence scores was also calculated, using a modified version of the p rocedure proposed b y Parrott et al., 93 to provide a quantitative measure of demasculinization across secondary sexual characteristics. The maleness index is especially useful in comparing gonadal size between males of differen t sizes and maturational stages. Sperm abundance was also rated, as shown in Table 1 and sperm developmental stages were evaluated. Sperm develops through four stages: spermatogonia, spermatocyte, spermatid, and spermatozoa. We determined the percent comp osition of each stage in each tubule and reported the distribution of these developmental stages for each dissected fish.

PAGE 33

21 Table 1 Secondary sexual characteristics and sperm abundance rating scales. 90 2 Sex. Char. Rating Scales 1 2 3 4 Nuptial tubercles Tubercles not visible Tubercles visible as white discs Tubercles prominent Tubercles prominent and protruding sharply Dorsal fat pad prominence Not visible Soft discolored tissue Spongy thickened tissue Dorsal hump with spongy tissue Sperm abundance Sperm absent Sperm prominent in <25% of tubules (weak) Sperm prominent in 25 75 % of tubules (moderate) Sperm prominent in > 75% of tubules (strong) Quantitative monoclonal antibody detection: ELISA Assay Detecting Plasma Vitellogenin Aliquots of plasma frozen and stored at 8 0 C during each dissection period were analyzed for Vtg by homologous enzyme linked immunosorbent assay (ELISA) using an anti Bergen, Norway). 94 95 This assay uses specific vitellogenin specific antibodies to quantify vitellogenin in plasma samples. Pre coated microplate wells utilize a specific Capture antibody in both the standard and sample wells, and which bind to Vtg. Another antibody, also spec ific to Vtg is the designated d etecting antibody, and is labelled with the enzyme hors eradish peroxidase (HRP). This d etecting antibody is added to form a addition of a substrate that produces a colored product. Absorbance measured by microplate reader, is directly proportional to Vtg concent ration in the sample. For this experiment, we used a single dilution (1:1000) to accommodate this predicted range of Vtg concentrations, and to avoid possible matrix effects. 96 Matrix effects are here defined as components of a sample, other than the analyte of interest,

PAGE 34

22 which may confound the analysis of that sample and, therefore, the q uality of the results obtained via such analysis. Dilution and washing buffers we re prepared via standard procedures ( Biosense, Bergen, Norway ). 97 98 The standard curve and all samples we re then diluted, added to the microplate wells, duplicated across each plate, sealed, and incubated at room temperature for 1.5 hour s. Post incubation, the plates we re washed with w ashing buffer and d etecting antibody (diluted as pre viously described) was added to the wells. These we re then resealed and incubated again for 0.5 hours, again at room temperature. After this second incubation, plates we re washed again, and tempered TMB substrate solution (pre prepared according to standard proced ures) was added to the wells. A third incubation wa s performed this time, under aluminum foil, though still at room temperature fo r twenty minutes. The reaction wa s then arrested with the addition of to all w ells, at which time the plates were placed in the microplate reader to be read at a 450 nm absorbance. Histology Hematoxylin and eosin (H&E) staining is used to clearly delineate tissue structures. 99 We prepared and utilized a set of H&E stained slides (Figure 11) as previously described to evaluate sperm abundance and any gross abnormalities in the gonadal tissues of male fa thead minnows. Sperm abundance in particular is a sensitive measure for reproductive disruption, as decreased abundance may occur even when other measures or assays (such as PCNA and TUNEL) show low to no significant disruption. 72

PAGE 35

23 Gonad and liver tissues were prepared for histology using standard dehydration and embedd ing procedures. We embedded gonadal tissue in paraffin wax and sectioned it heating element. Multiple ribbons containing different sections of the gonad were glued onto a series of five g lass slides per tissue to ensure that every slide contains multiple sections of gonad from each individual sampled, and is representative of the gonad as a whole, as opposed to one specific region. We produced five slides per tis sue, and from each set o ne slide was designated for hematoxylin and eosin (H&E). P rior to staining, slides were dewaxed and rehydrated in three xylene solutions a t five minutes each. Slides were then be treated with two solutions each of 100% ethanol, Figure 11 From left to right: reference ovary, intersex testis, and reference testis. Photo credit: Dr. Alan Vajda.

PAGE 36

24 95% ethanol, and 70% ethanol for three minut es in each solution, and subsequently rested in deionized water for 3 minutes. F ollowing dewaxing and rehydration, slides were placed in hematoxylin for 45 seconds, and rinsed in tap water until hematoxylin residues were no longer detectable. Excess liquid was removed, and Permount was applied to each. The slides were then topped with glass coverslips and air dried fo r 24 hours in a fume hood. At least 10 cross sections were evaluated from each gonad for determination of sex, sperm abundance, intersex status, and any abnormalities. Statistical Analysis Scores and ratings for sperm abundance, sperm developmental ratios w ere analyzed by Kruskall Wallis tests. Secondary sexual characteristics, vitellogenin, GSI, and CF were tested for homoscedasticity and analyzed b y a series of one way ANOVA tests for effects of the site (reference versus effluent, control versus reference and effluent, respectively) and sampling period (all weeks compared between one another and the control) Vitellogenin concentr ation data were log transformed prior to analysis. This was in accordance with methods from previous work at the COB WWTP and accounted for a lack of homosc edasticity in the data. Statistical significance was accepted at p < 0.05. Biological results are expressed as a mean (standard error of the mean (SEM) ) of the back transformed data where necessary.

PAGE 37

25 CHAPTER III RESULTS Plasma Vitellogenin Figure 1 2 shows an upward trend in the plasma vitellogenin (V tg ) of effluent exposed fish over this eight week sampling period. While there was no significant difference in plasma V tg between reference and effluent fish following four weeks of exposure (Wk4). A significant difference can be seen between the effluent and reference treatments during the post flood period in both week six and week eight ( P value = 0.000184) V tg c oncentrations in the reference treatment did not differ significantly from the initial control group (IC) at any sampling period Somatic Development and Condition Figure 1 3 shows the condition factor (CF) scores for all fish groups, organized by treatment. Condition factor was found to be significantly different (P value = 0.0007) between the experimental and control groups when a one way ANOVA was performed. Figure 1 4 shows the normalized gonadosomatic index (GSI) scores for all fish groups, organiz e d by treatment. Evaluated with one way ANOVA, GSI was found to be significantly different between control and experimental groups (P=0.003342) which may be attributable to natural growth and development of the fish after being acquired at 6 months of age No statisticall y significant difference between reference and effluent exposed groups could be determined. Figure 1 5 length and treatment week shows variation in fish body length over the experimental period. No statistically significant difference was detected by a nalysis with one way ANOVA (P value= 0.1312460)

PAGE 38

26 Secondary Sexual Characteristics Table 3 shows the Kruskal Wallis nonparametric test results for maleness (Figure 1 6 ) nuptial tubercle prominence, and dorsal fat pad prominence. No statistically significant difference (P>0.05) was found between the IC and the reference, the IC and the effluent, or the reference and effluent groups. Table 5 shows the one way ANOVA results for nuptial tubercle number (P value=0 .380518). Sperm Developmental Stages and Abundance Figure 1 7 shows the non parametric distribution of sperm developmental stages by treatment. In short, this figure sho ws the degree of development at each stage for each treatment group as they compare to one another. Essentially an expansion of figure 1 7, figure 18 provides a side by side comparison of all four sperm developmental stages as they vary between treatments. Table 4 displays the statistical analysis performed on sperm developmental stages over the 8 week testing period. Of the four stages, only spermatocytes showed any significant differences in effluent exposed individuals as compared to the contr ol group. Table 5 displays the statistical analysis ( one way ANOVAs) performed on GSI, CF, Vtg, NT #, and length over the 8 week testing period.

PAGE 39

27 Figure 13 Plasma VTG (in nanograms per microliter) in male fathead minnow (Pimephales promelas) in an initial control group (IC); reference groups for weeks 4 (Ref Wk4), 6 (Ref Wk6), and 8 (Ref Wk8); and effluent exposed groups for weeks 4 (Eff Wk 4), 6 (Eff Wk6). Figure 12 Condition factor (CF) by week and treatment within that week: initial controls (IC); week four, reference (Ref Wk4); week four, effluent (Eff Wk4); week six, reference (Ref Wk6); week six, effluent (Eff Wk6); week eight, reference (Ref Wk8); and week eight, effluent (Eff Wk8).

PAGE 40

28 Figure 14 Gonadosomatic index (GSI) score by week, and treatment within that week:: initial controls (IC); week four, reference (Ref Wk4); week four, effluent (Eff Wk4); week six, reference (Ref Wk6); week six, effluent (Eff Wk6); week eight, reference (Ref Wk8); and week eight, effluent (Eff Wk8). Figure 15 Length (L, measured in mm) only, grouped by treatment: initial controls (IC), reference individuals (Ref), and effluent treated fish (Eff).

PAGE 41

29 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage of Male Fish Treatment by Week Sperm Abundance Strong Moderate Weak Absent Figure 16 Maleness index scores by treatment: initial controls (IC), reference individuals (Ref), and effluent treated fish (Eff). Figure 17 Percentage of male fish within a given sperm developmental stage, organized by treatment: absent (1), weak (2), moderate (3), and strong (4); IC (initial controls), Wk 4 Ref (Reference), Wk 4 Eff (Effluent), Wk 6 Ref/Eff, and W k 8 Ref/Eff.

PAGE 42

30 Figure 18 Sperm developmental stages by treatment: spermatogonia (A), spermatocytes (B), spermatids (C), and spermatozoa (D).

PAGE 43

31 Table 2 One way ANOVAs for Gonadosomatic Index, Condition Factor, Nuptial Tubercle Number, Fish Length, and Plasma Vitellogenin Concentrations. GSI CF NT # Length and Mass Vtg P value 0.2163 0.0001* 0.5599 0.1851 0.0002* R squared 0.1049 0.2413 0.02087 0.08346 0.2731 Bartlett's Statistic 1.152 20.54* 5.934 6.723 178.6 Table 3 Kruskal Wallis Statistics for secondary sexual characteristics: maleness, nuptial tubercle prominence (NT Prom.), and dorsal fat pad prominence (DFP Prom.). Secondary Sex. Char. Maleness NT Prom. DFP Prom. P value 0.7430 0.3905 0.5424 Kruskal Wallis Statistic 3.507 1.881 1.223 Table 4 Statistical test (Kruskal Wallis) results for sperm developmental stages by treatment, with indicating significance. Spermatogonia Spermatocytes Spermatids Spermatozoa P value 0.0918 0.0034* 0.1210 0.8674

PAGE 44

32 CHAPTER IV DISCUSSION Overview In this study, we investigated the environmental and toxicological implications of potentially estrogenic WWTP effluents. This follow up experiment is the latest in a series of related work surrounding two upgrades to the COB WWTP. Our aim was to determine the efficacy of t hese upgrades in minimizing the exposure of downstream wildlife populations to estrogenic EDCs. Hypothesis Our hypothesis had a mu lti part composition regarding the different componen ts of 2012 upgrade In the context of effluent chlorination, we hypothesized that r em oving the chlorination step in the disinfection process would result in decreased estrog enicity in the effluent, as de monstrated by previous bench scale 70 73 74 and operational plant scale 75 studies indicating a weaker physiological response to chlorinated estrog enic compounds than unchlorin ated forms. We tempered this assertion with evidence from other studies, which have suggested that the removal of the chlorination step may also introduce additional estrogenicity f rom certain compounds (ex. Bisphenol A) and which may show increased binding affini t y to the ER when chlorinated. 79 We also suggested that a d ding a UV purification step could potentially improve estrogen removal from the effluent to a greater extent, an d with fewer estrogenic byproducts, than chlorination purification, as demonstrated in bench scale studies and experimental WWTPs. 50 85 This assertion was also qualified by data that suggests that certain compou nds may increase in estrogenicity upon exposure to UV. 86

PAGE 45

33 In short, we hypothesized that removing the chlorination step could result in feminization (elevated plasma vitellogenin concentrations) and demasculinization (reduced prominence of the nuptial tubercles and dorsal fat pad, decreased abundance of nuptial tubercles and spermatozoa in the tubule, and/or altered sperm developmental stages, GSI, and CF scores) if chlorination was decreasing bioavailability of estrogenic compounds. This hypothesis would also hold true if the addition of the UV purification step induced increased binding affinity of estrogenic EDCs to the ER, or if the September 2013 f lood event had a significant impact on the ability of the COB WWTP to remove estrogenic compounds from the effluent. We asserted that if the flood prevented the WWTP from effectively removing estrogenic compounds from the effluent, demasculinization and/or feminization would occur in effluent exposed individuals. If the dilution from high water volume was great enough, however, or if the WWTP was able to continue normal operations those conditions, no significant demasculinizing or feminizing effects would be observed. Plasma vitellogenin The r esults for plasma vitellogenin we re perhaps the most telling in this investigation. Figure 12 shows an induction in plasma V tg of effluent exposed individuals in weeks six and eight of the sampling period. Concentrations in the initial control group and all the reference groups, and the effluence and reference groups from week four, were relatively consistent across the sampling period. As seen in Figure 19, t he difference between plasma VTG concentrations in effluent versus reference treatment groups is insignificant during the fourth week of sampling, but a significant difference can be seen between the effluent and reference treatments in both week six and wee k eight right after flow through the WWTP increased significantly over the mean flow measured through the

PAGE 46

34 plant in the previous year (2012) This indicates that any effluent estrogenicity was contributed by the flood event not the series of upgrades ma de to the WWTP Figure 19 Plasma vitellogenin induction (black and grey) versus measured flow (blue) through the Boulder WWTP The induction of V tg and the subsequence appearance of the protein in the bloodstream of the organism can occur in as little as three days after estrogen exposure, with plasma levels reaching a plateau at about 28 days if exposure remains relatively constant. Once estrogen ex posure has ended, plasma Vtg tends to decrease in concentration after about three days, and the total time needed for male fish to clear their system of Vtg is about seven days. 100

PAGE 47

35 P lasma Vtg concentrations decrease d in week eight, which led us to conclude that the exposure to estrogenic compounds in the effluent was not constant throughout the sampling period and dropped off after a period. This would be consistent with the drop off in flow through in the WWTP after the cessation of the 2013 flood It seems ver y likely that the observed spike in effluent estrogenicity which appears to have ended after week six, coincided with increased hydrological flow surrounding the September 2013 flood event (Figure 19) This would imply that effluent estrogenicity at the COB WWTP following the 2012 upgrade can be attributed to sources outside of the WWTP itself in this case, the flood and not to disinfection processes, or the upgrades therein, within the WWTP becau se effluent estrogenicity was not detected before the flood began or after it ended. Sperm abundance and development Sperm abundance ( Figure 1 7 ), and distribution of sperm throughout the four developmental stages ( Figure 1 8 ), appear largely unaffected by any effluent estrogenicity, with one possible exception. Spermatocytes development appears to be significantly different between effluent and reference/control groups in weeks six and eight but no such difference exists for any of the other stages we examined. C orrelation s have been found between and spermatocyte apoptosis in a mouse model, 101 so it is possible that a relationship between decreased ER expression and increased spermatocyte representation exists. Spermatocytes also tend to last for an extended period in the testis as a result of the long duration of first meiotic prophase it could be that an increased spermatocyte representation is related to some disruption in meiotic division. 102

PAGE 48

36 It is possible that this increase detected in effluent exposed fish could correlate with exposure to an unknown EDC with effects on meiotic division or decreasing cell apoptosis al though this would require additional testing to confirm. There was no in dication of deformation in the gonadal ti ssues examined histologically, and all other stages failed to display any significant difference, so this significance in spermatocyte development in flood effluent exposed fish most likely has little to no effect o n reproductive function. Maleness All measures of maleness, including the maleness score ( Figure 1 6 ), nuptial tubercle number, nuptial tubercle prominence, and dorsal fat pad prominence showed no statistically significant differences between control and treatment groups ( Tables 1 and 2 ). This can be interpreted in one of two ways: first, that no significant estrogen exposure occurred and, therefore, no alteration was made in these secondary sexual chara cteristics; and second, that any estrogen exposure that may have occurred prior to or during the in fish from earlier developmental periods). Condition factor While statistically significant, the differences in condition factor ( Figure 13 ) between the initial control group and the experimental groups can most likely be attributed to the growth and development of each individual fish over the 8 week long testing period. Some increase in body size seems likely, as P. promelas only reaches sexual maturity at 5 6 months of age, and may live as long as 2 3 years, during which time they continue to gain mass and length; 103 the fish used in this experiment were aged 6 months. This is especially probable given th at condition factor, which represents the

PAGE 49

37 ratio of body mass to body length, is closely related to gonadosomatic index which represents the ratio of gonad mass to body mass; GSI was not reported as statistically significant between control and experimental groups when evaluated by the same statistical measure, and to the same power ( Figure 1 5 ) Therefore, we can assume that the difference is in the length of the tested fish an increase which would represent the natural growth and development of the organism, rather than any environmental variable. Estrogens measured at this site over the past several years have showed low, temporally variable quantities of estrogenic compoun ds in the effluent; however, further assessment on the chemistry of the Boulder WWTP effluent during this particular experimental period has yet to be fully assessed. Maleness appeared to remain largely unaffected by estrogenic endocrine disruption, with r esults within the context of periods. Due to statistically significant results for differences between the initial control group and treatment groups in the plasma vitellogenin E LISA assay (to be further discussed; see below), we can assume that estrogens in the effluent were not low enough to leave the fish completely unscathed. Because P. promelas seems to respond dramatically to estrogens at early developmental stages, especially with respect to nuptial tubercle number, 104 often to such an extent that tubercles fail to develop at all. After early development, P. prome las has been shown to respond to 10 100 nM E2 in a dose dependent manner: nuptial tubercle prominence appears to be the most sensitive measure, as they will atrophy at high doses, while dorsal fat pads inconsistently atrophy at even greater doses. 36

PAGE 50

38 As effects were not seen in either nuptial tubercle or dorsal fat pad prominence, we can deduce that the quantity of estrogenic compounds in the Boulder WWTP effluent was lower than that seen prior to the 2007 upgrade, at which time secondary sexual characteristics were significantly impacted. It would appear, then, that effluent estrogenicity is in this case too low to eli cit lasting or population wide e ffects on fish appearance or male distribution as no significant effects were detected for sperm abundance or secondary sexual characteristics. This assertion would be strengthened by further investigation, which should include behavioral and fecundity assays of both male and female organisms No intersex was detected, and no e xtreme morphologies were discovered, which implies that if the results of the following tests (which were found to be statistically significant) can be attributed to effluent estrogenicity, that estrogenicity is relatively low. Further, the exposure may we ll have been acute, and possibly attributable to sources outside the WWTP itself perhaps due to the excessive rainfall that occurred in the area during and around the testing period. It should be noted that the major flood event was the largest recorded in Boulder, Colorado since a comparable event in 1976, which itself resulted in le ss rainfall than the 2013 event. Synthesis It can be concluded that effluent estrogenicity is low and temporally variable following the 2012 upgrade. However, it would appear that some estrogenic compounds are present in the effluent at higher levels than those seen after the 2007 upgrade. While th is slight uptick could be explained by increased ER binding affinity in compounds that were previously chlorinated by the CO B WWTP, or some compounds which have

PAGE 51

39 increased in estrogenicity upon exposure to UV such conclusions cannot be stated definitively without further study at this site. It should be stressed that any future studies would have to be conducted during periods of drought as well as periods of typical flow, and with an eye to this study, in order to formulate a clearer picture of how the CO B WWTP operates from year to year, and with significant variation. Increased flow like that seen during September 2013, may well have prevented the plant from maintaining consistent retention times, or it may have overburdened the system as water bypassed the influent and flowed over the surface of the plant itself. Such a flow may well have washed previously undetected EDCs into the water column, resulting in an effluent of higher estrogenicity. Summary Based on our results, we conclude that the series o f upgrades implemented at the COB WWTP significantly decrease estrogenicity in the effluent the plant produces, and reduce the effects of estrogenic EDCs on downstream fish populations. We further conclude that the removal of the chlorination step in the disinfection process, and its simultaneous replacement with UV treatment, had little to no effect on increasing effluent estrogenicity after that upgrade was conducted in 2012. We also fi nd ability to reduce effluent estrogenicity. Despite these detected increases, we can conclude that current levels of estrogenic compounds in the COB WWTP are low enough t hat no current danger is posed to downstream fish population on the basis of reproductive or endocrine disruption.

PAGE 52

40 REFERENCES 1. Liu, Z., Kanjo, Y. & Mizutani, S., Removal mechanisms for endocrine disrupting chemical advanced oxidation: a review. Sci. Total Environ 2. National Research Council, Issues in Potable Reuse: The Viability of Augmenting Drinking Water Supplies with Reclaimed Water (1998). 3. Jobling, S. et al. Altered Sexual Maturation and Gamete Production in Wild Roach (Rutilus rutilus) Living in Rivers That REceive Treated Sewage Effluent s. Biology of Reproduction 66 (2), 272 281 (2002). 4. Head, B., Managing urban water crises: adaptive policy responses to drought in Southeast Queensland, Australia. Ecology & Society 19 (2), 33 (2014). 5. National Climatic Data Center (NCDC), Drought January 2013 Report (2013). 6. Johnson, A. & Sumpter, J., Removal of endocrine disrupting chemical sin activated sludge treatment works. Environmental Science and Technology 35 4697 4703 (2001). 7. National Weather Service, Hydrologic Information Center : Flood Loss Data, Available at http://www.nws.noaa.gov/hic/. 8. Karl, T., Melillo, J. & Peterson, T., Global Climate Change Impacts in the United States, 2009. 9. Strand Associates, Inc, Troubleshooting activated sludge: identifying and mitigating impacts of drought conditions. presented at Central States Water Environment Association (86th Annual Meeting), 2013 2014 (unpublished). 10. Barber, L. et al. Chemical loading into surface water along a hydro logical, biogeochemical, and land use gradient a holistic watershed approach. Environmental Science and Technology 2006 475 486 (2006). 11. Barber, L., Brown, G. & Zaugg, S., Potential endocrine disrupting organic chemicals in treated municipal wastewa ter and river water. presented at In Analysis of Environmental Endocrine Disruptors, Washington, DC, 2000 (unpublished).

PAGE 53

41 12. Kolpin, D. et al. Pharmaceuticals, hormones, and other organic wastewater A national r econnaissance. Environmental Science and Technology 36 1202 1211 (2002). 13. Ternes, T., Joss, A. & Siegrist, H., Scrutinizing pharmaceuticals and personal care products in wastewater treatment. Environmental Science and Technology 38 393A 399A (2004). 14. Sim, W., Lee, J. & Oh, J., Occurence and fate of pharmaceuticals in wastewater treatment plants and rivers in Korea. Environmental Pollution 158 (5), 1938 1947 (2010). 15. Kloepfer, A., Jekel, M. & Reemtsma, T., Occurrence, sources, and fate of benxotiazoles in municipal wastewater treatment plants. Environmental Science and Technology 39 (10), 3792 3798 (2005). 16. Van Nuijs, A. et al. Spatial and temporal variations in the occurence of cocaine and benzoylecgonine in waste and surface water from Belgium and removal during wastewater treatment. Water Research 43 (5), 1341 1349 (2009). 17. Hohne, C. & Puttmann, W., Occurrence and tempo ral variations of the xenoestrogens bisphelo A, 4 tert octylphenol, and tech. 4 nonylphenol in two German wastewater treatment plants. Environmental Science and Pollution Research 15 405 416 (2008). 18. Johnson, A. & Williams, R., A model to estimate inf luent and effluent concentrations of estradiol, estrone, and ethinylestradiol at sewage treatment works. Environmental Science and Technology 38 3649 3658 (2004). 19. D'Ascenzo, G. et al. Fate of natrual estrogen conjugates in municipal sewage transport and treatment facilities. Science of the Total Environment 302 199 209 (2003). 20. Lange, I. et al. Sex hormones originating from different livestock production systems: fate and potential disrupting activity in the environment. Analytica Chimica Acta 473 (1 2), 27 37 (2002). 21. Vajda, A., Lopez, E., Woodling, J., Maldonado, T. & Norris, D., Reproductive disruption and intersex in white suckers (Catostomus commersoni) downstream of a Colorado wastewater treatment plant. Integrative and Comparative Bio logy 44 (6), 655 (2004).

PAGE 54

42 22. Woodling, J., Lopez, E., Maldonado, T., Norris, D. & Vajda, A., Intersex and other reproductive disruption of fish in wastewater effluent dominated Colorado streams. Comparative Biochemistry and Physiology, Part C: Toxicology and Pharmacology 144 (1), 10 15 (2006). 23. Ma, T., Wan, X., Huang, Q., Wang, Z. & Liu, J., Biomarker responses and reproductive toxicity of effluent from a Chinese large sewage treatment plant in Japanese medaka (Oryzias latipes). Chemosphere 59 (2), 281 288 (2005). 24. Schwindt, A., Winkelman, D., Keteles, K., Murphy, M. & Vajda, A., An environmental oestrogen disrupts fish population dynamics through direct and transgenerational effects on survival and fecundity. Journal of Applied Ecology 51 (3), 582 591 (2014). 25. Tetreault, G. et al. Intersex and reproductive impairment of wild fish exposed to multiple municipal wastewater discharges. Aquatic Toxicology 104 (3 4), 278 290 (2011). 26. Shenoy, K. & Crowley, P., Endocrine disruption of male mating signals: ecological and evolutionary implications. Functional Ecology 25 (3), 433 448 (2010). 27. Kumar, A. et al. Treated effluent in the aqautic environment: impact assessment of endocrine disrupting chemicals. Water for a Healthy Country Flags hip Report Series (2012). 28. Johnson, A., Belfroid, A. & Di Corcia, A., Estimating steroid oestrogen inputs into activated sludge treatment works and observations on their removal from the effluent. Science of the Total Environment 256 163 173 (2000). 29. De Sa Salomao, A., Estrogenicity and Genotoxicity Detection in Different Contaminated Waters. Human and Ecological Risk Assessment (2014). 30. Murkies, A., Wilcox, G. & Davies, S., Clinical review 92 phytoestrogens. The Journal of Clinical Endocrinology and Metabolism 83 297 303 (1998). 31. Combalbert, S. & Hernandez Raquet, G., Occurrence, fate, and biodegradation of estrogens in sewage and manure. Applied Microbiology and Biotechnology 86 (6), 1671 1692 (2010).

PAGE 55

43 3 2. Lai, K., Johnson, K., Scrimshaw, M. & Lester, J., Binding of waterborne steroid estrogens to solid phases in river and estuarine systems. Environmental Science and Technology 34 3890 3894 (2000). 33. Andersen, H. et al. Comparison of short term estro genicity tests for identification of hormone disrupting chemicals. Environmental Health Perspectives 107 89 108 (1999). 34. Datson, G., Cook, J. & Kaylock, R., Uncertainties for endocrine disruptors: our view on progress. Toxicological Science 74 245 25 2 (2003). 35. Scholz, S. & Mayer, I., Molecular biomarkers of endocrine disruption in small model fish. Science of the Total Environment 293 (1 2), 57 70 (2008). 36. Miles Richardson, S. et al. Effects of waterbirne exposure of 17B estradiol on secondary sex characteristics and gonads of fathead minnows (Pimephales promelas). Aquatic Toxicology 47 (2), 129 145 (1999). 37. Vajda, A. & Norris, D., Endocrine active chemicals (EACs) in fishes. Hormones and Reproduction of Vertebrates: Fishes 1 245 264 (2011). 38. Norris, D., Vajda, A., Barber, L. & Schoenfuss, H., Impacts of neuroactive and estrogenic chemicals in wastewater effluents on behavior and reproduction in freshwater fishes. Integrat ive and Comparative Biology 52 E130 (2012). 39. McGee, M. et al. Predator avoidance performance of larval fathead minnows (Pimphales promelas) following short term exposure to estrogen mixtures. Aquatic Toxicology 91 (4), 355 361 (2009). 40. Painter, M et al. Antidepressants at environmentally relevant concentrations affect predator avoidance behavior of larval fathead minnows (Pimephales promelas). Environmental Toxicology and Chemistry 28 (12), 2677 2684 (2009). 41. Kime, D. & Nash, J., Gamete viab ility as an indicator of reproductive endocrine disruption in fish. Science of the Total Environment 233 (1 3), 123 129 (1999). 42. Bickley, L. et al. Interactive effects of inbreeding and endocrine disruption on reproduction in a model laboratory fish. Evolutionary Applications 6 (2), 279 289 (2013). 43. Svenson, A., Removal of estrogenicity in Swedish municipal sewage treatment plants. Water Research 37 (18), 4433 4443 (2003).

PAGE 56

44 44. Vajda, A. et al. Integrated assessment of wastewater treatment plant effluent estrogenicity in the Upper Murray River, Australia, using the native Murray rainbowfish (Melanotaenia fluviatilis). Environmental Toxicology and Chemistry (2015). 45. Vajda, A., Bolden, A., Wo odling, J., Barber, L. & Norris, D., Reproductive Disruption by Estrogenic Wastewater Effluents. Endocrine Reviews 31 (3) (2010). 46. Vajda, A. et al. Demasculinization of male fish by wastewater treatment plant effluent. Aquatic Toxicology 103 (93 94), 213 221 (2011). 47. Kidd, K., McGree, M., Winkelman, D., Vieira, N. & Vajda, A., Reproductive failure of the red shiner (Cyprinella lutrensis) after exposure to an exogenous estrogen. Candaian Journal of Fisheries and Aquatic Sciences 67 (11), 1730 1743 ( 2010). 48. Gray, J., Barber, L., Vajda, A. & Norris, D., Detection of Environmental Estrogens and Concurrent Observation of Feminization of Fathead Minnows in Waste Water Effluent presented at 5th International Conference on Pharmaceuticals and Endocrine Disrupting Chemicals in Water, Costa Mesa, CA, 2006 (unpublished). 49. Salierno, J. & Kane, A., 17alpha ethinylestradiol alters reproductive behaviors, circulating hormones, and sexual morphology in male fathead minnows (Pimephales promelas). Environment al Toxicology and Chemistry 28 (5), 953 961 (2009). 50. Chen, P. et al. Biological assessment of bisphenol A degradation in water following direct photolysis and UV advanced oxication. Chemosphere 65 1094 1102 (2006). 51. Rosenfeldt, E. & Linden, K., Degradataion of Endocrine Disrupting Chemicals Bisphenol A, Ethinyl Estradiol, and Estradiol during UV Photolysis and Advanced Oxidation Processes. Environmental Science and Technology 38 5476 5483 (2004). 52. Hepell, S., Denslow, N., Folmar L. & Sullivan, C., Universal assay of vitellogenin as a biomarker for environmental estrogens. Environmental Health Perspectives 103 (7), 9 15 (1995). 53. Chen, P., Rosenfeldt, E., Kullman, S., Hinton, D. & Linden, K., Biological assessments of a mixtur e of endocrine disruptors at environmentally relevant

PAGE 57

45 concentrations in water following UV/H2O2 oxidation. Science of the Total Environment 376 18 26 (2007). 54. Denslow, N., Chow, M., Kroll, K. & Green, L., Vitellogenin as a biomarker of exposure for es trogen or estrogen mimics. Ecotoxicology 8 385 398 (1999). 55. Arukwe, A. & Goksoyr, A., Eggshell and egg yolk proteins in fish: hepatic proteins for the next generation. Comparative Hepatology 2 (4) (2003). 56. Cheek, A. et al. Experimental evaluation of Vitellogenin as a Predictive Biomarker for Reproductive Disruption. Environmental Health Perspectives 109 (7), 681 690 (2001). 57. Hansen, P. et al. Vitellogenin a biomarker for endocrine disruptors. Trends in Analytical Chemistry 17 (7), 448 451 (1998). 58. Sumpter, J. & Jobling, S., Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environmental Health Perspectives 103 (7), 173 178 (1995). 59. Garcia Reyero, N. et al. Use of vitellogenin mRNA as a biomarker for endocrine disruption in feral and culture fish. Analytical and Bioanalytical Chemistry 278 670 675 (2004). 60. Adanyi, N. et al. Optical waveguide lightmode spectroscopy immunosenor for detection of carp vit ellogenin. Sensors and Actuators B: Chemical 176 932 939 (2013). 61. Om, A., Jasmani, S., Ismail, N., Yeong, S. & Abol Munafi, A., Application MALDI TOF on protein identification of vitellogenin in giant grouper (Epinphelus lanceolatus). Fish Physiology and Biochemistry 39 (5), 1277 1286 (2013). 62. Montorzi, M., Falchuk, K. & Vallee, B., Xenopus laevis vitellogenin is a zinc protein. Biochemical and Biophysical Research Communications 200 1407 1413 (1994). 63. Wallace, R., in Developmental Biology: A Comprehensive Synthesis edited by Browder, L. (Plenum Press, New York, 1985), Vol. 1, pp. 127 177. 64. Short, S., Yang, G., Kille, P. & Ford, A., Vitellogenin is not an appropriate biomarker of feminization in a Crustacean. Aquatic Toxicology 153 89 97 (2014).

PAGE 58

46 65. Brawand, D., Wahli, W. & Kaessmann, H., Loss of Egg Yolk Genes in Mammals and the Origin of Lactation and Placentation. PLoS Biology 6 (3), 63 (2008). 66. Nagler, J., Ruby, S., Idler, D. & So, Y., Serum phosphoprotein phosphorus and calcium levels as reproductive indicators of vitellogenin in highly vitellogenic female and estradiol injected immature rainbow trout (Onchorhynchus mykiss). Canadian Journal of Zoology 65 2421 2425 (1987). 67 Nilsen, B. et al. Development of quantitative vitellogenin ELISAs for fish test species used in endocrine disruptor screening. Analytical and Bioanalytical Chemistry 387 621 633 (2004). 68. Zhong, L. et al. Distribution of vitellognin in zebrafish (D anio rerio) tissues for biomarker analysis. Aquatic Toxicology 149 1 7 (2014). 69. Barber, L., Vajda, A., Douville, C., Norris, D. & Writer, J., Fish Endocrine Disruption Responses to a Major Wastewater Treatment Facility Upgrade. Environmental Science a nd Technology 46 (4), 2121 2131 (2012). 70. Bourgin, M. et al. Chlorination of bisphenol A: Non targeted screening for the identification of transformation products and assessment of estrogenicity in generated water. Chemosphere 93 2814 2822 (2013). 71. Korshin, G., Kim, J. & Gan, L., Comparative study of reactions of endocrine disruptors bisphenol A and diesthylstilbestrol in electrochemical treatment and chlorination. Water Research 40 (5), 1070 1078 (2006). 72. Vajda, A. et al. Reproductive Disruption in Fish Downstream from an Estrogenic Wastewater Effluent. Environmental Science and Technology 42 (9), 3407 3414 (2008). 73. Lee, B. et al. Effects of chlorine on the decrease of estrogenic chemicals. Water Research 38 (3), 733 739 (2004). 74. Hu, J., Cheng, S., Aizawa, T., Terao, Y. & Kunikane, S., Products of Aqueous Chlorination of 17B Estradiol and Their Estrogenic Activities. Environmental Science and Technology 37 5665 5670 (2003). 75. Schiliro, T., Pignata, C., Rovere, R., Fea, E. & Gilli, G., The endocrine disrupting activity of surface waters and of wastewater treatment plant effluents in relation to chlorination. Chemosphere 75 335 340 (2009).

PAGE 59

47 76. Bolden, A., Vajda, A. & Larber, L., Reproductivedisruption of fishes by endocrine active wastewater effluent. Integrative and Comparative Biology 50 E13 (2010). 77. Alum, A., Yoon, Y., Westerhoff, P. & Abbaszadegan, M., Oxidation of Bisphenol A, 17B Estradiol, and 17a Ethynyl Estradiol and Byproduct Estrogenicity. Environmtnal Toxicology 19 (3), 257 264 (2004). 78. Gallard, H., Leclerq, A. & Croue, J., Chlorination of bisphenol A: kinetics and by product formation. Chemosphere 56 (5), 465 473 (2004). 79. Hu, J., Aizawa, T. & Ookubo, S., Products of Aqueous Chlorination of Bisphenol A and Their Estrogenic Activity. Environmental Science and Technology 36 1980 1987 (2002). 80. Coleman, H., Routledge, E., Sumpter, J., Eggins, B. & Byrne, J., Rapid loss of estrogenicity of ster oid estrogens by UVA photolysis and photocatalysis over an immobilized titanium dioxide catalyst. Water Research 38 (14 15), 3233 3240 (2004). 81. Zhang, Y., Zhou, J. & Ning, B., Photodegradation of estrone and 17B estradiol in water. Water Research 41 1 9 26 (2007). 82. Rosenfeldt, E., Chen, P., Kullman, S. & Linden, K., Destruction of estrogenic activity in water using UV advanced oxidation. Science of the Total Environment 377 105 113 (2007). 83. Li Puma, G., Puddu, V., Tsang, H., Gora, A. & Toepfer, B., Photocatalytic oxidation of multicomponent mixtures of estrogens (estrone (E1), 17B estradiol (E2), 17a ethynylestradiol (EE2) and estriol (E3)) under UVA and UVC radiation: Photon absorption, quantam yields and rate constants independent of photon ab sorp. Applied Catalysis B: Environment 99 (3 4), 388 397 (2010). 84. Maniero, M., Bila, D. & Dezotti, M., Degradation and estrogenic activity removal of 17B estradiol and 17a ethinylestradiol by ozonation and O3/H2O2. Science of the Total Environment 407 (1), 105 115 (2008). 85. Zhang, Z. et al. Kinetic degradation model and estrogenicity changes of EE2 (17a ethinylestradiol) in aqueous solution by UV and UV/H2O2 technology. Journal of Hazardous Materials 181 1127 1133 (2010).

PAGE 60

48 86. Atkinson, S. et al. Environmental Factors Affecting Ultraviolet Photodegradation Rates and Estrogenicity of Estrone and Ethinylestradiol in Natural Waters. Archives of Environmental Contamination and Toxicology 60 (1), 1 7 (2011). 87. Anderson, H. et al. Assessment of the importance of sorption for steroid estrogens removal during activated sludge treatment. Chemosphere 61 139 146 (2005). 88. Furlong, E. T., Vaught, D. G., Merten, L. M., Foreman, W. T. & Gates, P. M., US Geological Survey Report No. 95 719, 1995. 89. An kley, G., Jensen, K., Kahl, M., Korte, J. & Makynen, E., Description and evaluation of a short term reproduction test with the fathead minnow (Pimephales promelas). Environmental Toxicology and Chemistry 20 (6), 1276 1290 (2001). 90. Parrott, J. & Blunt, B., Life cycle exposure of fathead minnows (Pimephales promelas) to an ethinylestradiol concentration below 1 ng/L reduces egg fertilization success and demasculinizes males. Environmental Toxicology 20 (2), 131 141 (2005). 91. Latrouite, D. & Claude, S., Essai de culture suspendue de la coquille Saint Jacques Pecten maximus influence de la transplantation sur le development gonadique presented at International Council for the Exploration of the Sea (ICES), Copenhagen, Denmark, 1979 (unpublish ed). 92. Presnell, J. & Schireibman, M., Humason's Animal Tissue Techniques (Johns Hopkins University Press, Baltimore, Maryland, 1997). 93. Parrott, J., Wood, C., Boutot, P. & Dunn, S., Changes in growth and secondary sex characteristics of fathead minnows exposed to bleached sulfite mill effluent. Environmental Toxicology and Chemistry 22 (12), 2908 (2003). 94. Eidem, J., Kleivdal, H. & Goksoyr, A. (Biosense, 2005). 95. Eide m, J. K. H. & Goksyr, A. (Biosense, 2005). 96. Eurachem, The Fiteness for Purpose of Analytical Methods, Available at http://www.eurachem.ul.pt/guides/valid.pdf (1998). 97. AOAC, Single Laborato ry Validation of Analytical Methods for Dietary Supplements. presented at International Training Course, Atlanta, GA, 2003 (unpublished).

PAGE 61

49 98. IUPAC Technical Report, Harmonized guidelines for single laboratory validation of methods of analysis. Pure Applied Chemistry 74 (5), 835 855 (2002). 99. Blazer, V., Histopathological assessment of gonadal tissue in wild fishes. Fish Physiology and Biochemistry 26 (85), 101 (2002). 100. Schmid, T., Gonzalez Valero, J., Rufli, H. & Dietrich, D., Determination of vitellogenin kinetics in male fathead minnows (Pimephales promelas). Toxicology Letters 131 (1 2), 65 74 (2002). 101. Selva, D. et al. Estrogen Receptor B Expression and Apoptosis of Spermatocytes of Mice Overexpressing a Rat Androgen Binding Protein Transgene. Biology of Reproduction 71 (5), 1461 1468 (2004). 102. Schulz, R. et al. Spermatogenesisin fish. Fish Reproduction 165 (3), 290 411 (2010). 103. Werner, R., Freshwater Fishes of the Northwestern United States: A F ield Guide (Syracuse University Press, Syracuse, NY, 2004). 104. Bogers, R. et al. Development of chronic tests for endocrine active chemicals: Part 2: An extended fish early life stage test with an androgenic chemical in the fathead minnows (Pimephales promelas). Aquatic Toxicology 80 (2), 119 130 (2006). 105. Nilsen, B. et al. Development of quantitative vitellogenin ELISAs for fish test species used in endocrine disruptor screening. Analytical and Bioanalytical Chemistry 378 621 633 (2004). 106. Blaise, C., Gagne, F., Salazar, S., Trottier, S. & Hansen, P., Experimentally induced feminization of freshwater mussels after long term exposure to a municipal effluent. Fresenius Environmental Bulletin 12 865 870 (2003).