Effects of paper mill effluents on the health and reproductive success of Largemouth Bass (Micropterus Salmoides)

MISSING IMAGE

Material Information

Title:
Effects of paper mill effluents on the health and reproductive success of Largemouth Bass (Micropterus Salmoides) field and laboratory studies.
Alternate title:
Field and laboratory studies
Physical Description:
ix, 268 leaves : ill. ; 29 cm.
Language:
English
Creator:
Sepulveda, Maria Soledad, 1967-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Bass -- physiology   ( mesh )
Estradiol -- analysis   ( mesh )
Estradiol -- metabolism   ( mesh )
Water Pollutants, Chemical -- analysis   ( mesh )
Industrial Waste -- toxicity   ( mesh )
Vitellogenin -- analysis   ( mesh )
Vitellogenin -- metabolism   ( mesh )
Steroids -- analysis   ( mesh )
Steroids -- metabolism   ( mesh )
Water Pollutants, Chemical -- toxicity   ( mesh )
Ecology   ( mesh )
Department of Veterinary Medicine thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Veterinary Medicine -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2000.
Bibliography:
Bibliography: leaves 251-268.
Statement of Responsibility:
by Maria Soledad Sepulveda.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002785036
oclc - 49792941
notis - ANR3200
System ID:
AA00011789:00001


This item is only available as the following downloads:


Full Text










EFFECTS OF PAPER MILL EFFLUENTS ON THE HEALTH AND
REPRODUCTIVE SUCCESS OF LARGEMOUTH BASS (MICROPTERUS
SALMOIDES): FIELD AND LABORATORY STUDIES













By

MARIA SOLEDAD SEPTiLVEDA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2000














ACKNOWLEDGMENTS


My deepest thanks to Dr. Timothy Gross for having accepted me as his graduate

student. He introduced me to the field of fish ecotoxicology and endocrine disruption,

and offered me the unique opportunity to work with an interdisciplinary research team at

the USGS-BRD Florida Caribbean Science Center Ecotoxicology Laboratory

(Gainesville, FL). Ecotoxicology staff members and friends Shane Ruessler, Carla

Wieser, Jon Wiebe, Nikki Kernaghan, Kelly McDonald, and Vincent and Lisa Centonze

assisted in innumerable tasks ranging from the treatment, collection and processing of

fish; analysis of plasma samples for reproductive hormones; to the conduction of

spawning studies. Lyn Day and Meri Nantz helped in the purchase of equipment and in

the day-to-day process of trying to get things done. My most sincere appreciation to all of

them. I will always remember my days at the "ecotoxicology lab"!

I would also like to thank my other committee members: Drs. Evan Gallagher and

Steve Roberts (Center for Environmental and Human Toxicology, College of Veterinary

Medicine, UF, Gainesville, FL); Dr. Trenton Schoeb (Department of Pathobiology,

College of Veterinary Medicine, UF, Gainesville, FL), and Dr. Nancy Denslow

(Department of Biochemistry and Molecular Biology, College of Medicine, UF,

Gainesville, FL). Their comments and suggestions greatly improved the quality of my

research and writing. Also, Dr. Schoeb's help in the interpretation of histology slides and








Dr. Gallagher's assistance in the conduction of EROD and other liver enzyme assays are

greatly acknowledged.

This dissertation would not have been possible without the support of the

Georgia-Pacific Corporation, Atlanta, GA. through a two-year grant awarded to Dr.

Timothy Gross. My deepest thanks to Stewart Holm, from Georgia-Pacific, for his help

in designing and overseeing this project. Also thanks to Myra Carpenter (Georgia-

Pacific, Palatka Operation, FL) who was responsible for the building and maintenance of

the treatment tank system in Palatka.

I also thank Karen Sheehy (Center for Environmental and Human Toxicology,

UF, Gainesville, FL) for her assistance in the EROD analyses, as well as Kevin Kroll and

Marjorie Chow (Center for Biotechnology, UF, Gainesville, FL) for conducting the 1998

vitellogenin analyses. I am specially thankful to Kevin, who kindly spent time training

me in the "art" of conducting ELISAs for the detection of vitellogenin in bass.

I would like to thank Bill Johnson and other staff from the Florida Game and

Freshwater Fish Commission (Fisheries Research Laboratory, Eustis, FL) for providing

boats and personnel for the collection of largemouth bass from the St. Johns River and for

aging the fish using otolith analyses. John Higman (St. Johns Water Management

District) provided invaluable information on fish and sediment chemical data from sites

along the lower St. Johns River. Jay Harrison and Galin Jones (Department of Statistics,

UF, Gainesville, FL) assisted in the statistical analyses.

Finally, with special recognition and love, I thank my husband, Hugo Ochoa, my

daughter Natalia Ochoa, and my mother Pura Luque, for their invaluable collaboration,









support, and most of all for their patience. I would not have been able to finish this

degree without them.















TABLE OF CONTENTS
page

ACKNOW LEDGM ENTS ...................................................................................................ii

ABSTRACT ........................................................................................................... viii

CHAPTERS
1 INTRODUCTION AND BACKGROUND ............................................................ 1
Introduction............................................................................................................ 1
General Aims of Ecotoxicological Studies......................................................... 2
Biomarkers of Exposure and Effects .................................................................. 2
Levels of Biological Organization ...................................................................... 3
Effects of Endocrine Disrupting Chemicals in W ildlife.....................................4...
The Pulp and Paper Industry..................................................................................... 5
Introduction...................................................................................................... 5
Pulp and Paper M manufacturing Process Sequence..............................................5...
Pollution Outputs ................................................................................................ 9
W astewater Treatment Technology .................................................................. 10
The Pulp and Paper Cluster Rules .................................................................... 11
Georgia-Pacific's Paper M ill Plant in Palatka, Florida........................................... 12
General Description .......................................................................................... 12
Ongoing Improvements .................................................................................... 13
Sublethal Physiological Effects of Pulp and Paper Mill Effluents on Fish ........... 13
General Health Effects...................................................................................... 14
Liver Health Effects.......................................................................................... 17
Reproductive Health Effects............................................................................. 22
The Largemouth Bass (M icropterus salmoides)..................................................... 27
General Description .......................................................................................... 27
Geographic Distribution ................................................................................... 28
Habitat and Range............................................................................................. 28
Growth and Feeding Habits .............................................................................. 29
Reproduction..................................................................................................... 30
Significance of this W ork ....................................................................................... 33
Organization of Dissertation................................................................................... 34
2 COMPARISON OF REPRODUCTIVE PARAMETERS FROM FLORIDA
LARGEMOUTH BASS (MICROPTERUS SALMOIDES FLORIDANUS)
SAMPLED FROM REFERENCE AND CONTAMINATED SITES IN THE
ST. JOHNS RIVER AND TRIBUTARIES........................................................... 36









Introduction ................................................................................................................ 36
M materials and M ethods ........................................................................................... 39
Sam pling Sites and Fish Collection.................................................................. 39
Chem ical Analysis from Fish Tissues .............................................................. 40
Bleeding, Necropsies, and Age D eterm ination................................................. 41
Reproductive Endpoints.................................................................................... 41
Liver EROD Activity........................................................................................ 46
Statistical Analyses........................................................................................... 47
Results........................................................................................................................ 48
Chemical Analysis from Sediments and Fish Tissues ......................................48
Physiological and Reproductive Endpoints ...................................................... 48
D iscussion.................................................................................................................. 53
3 IN VIVO ASSESSMENT ON THE REPRODUCTIVE EFFECTS OF
PAPER MILL EFFLUENTS ON LARGEMOUTH BASS................................... 91
Introduction.................................................................................. ... ..................... 91
M materials and M ethods ........................................................................................... 92
Anim als and Holding Facility........................................................................... 92
Effluent Characteristics......................................................... ............................ 93
Exposure Conditions......................................................................................... 94
Reproductive Endpoints.................................................................................... 94
Statistical Analyses........................................................................................... 95
Results............. ......... ............................................................................................ 96
Fem ales ................................................................................................................ 96
M ales.................................................................................................................... 98
Both Sexes .................................................................................................... 99
D iscussion................................................................................................................ 100
4 IMPACT OF PAPER MILL EFFLUENTS ON LARGEMOUTH BASS
HEALTH: FIELD AND LABORATORY STUDIES........................................... 128
Introduction.............................................................................................................. 128
M materials and M methods ............................................................................................ 129
Field Study..................................... ....................................................................129
Laboratory Study ................... .. ..................................................................... 134
Results...................................................................................................................... 136
Field Study.........................................................................................................136
Laboratory Study ................ ... ............................................... .. ...................... 137
Discussion ............................................................................................ .................... 138
5 EFFECTS OF PAPER MILL EFFLUENTS ON REPRODUCTIVE
SUCCESS OF LARGEM OUTH BA SS ............................. .................................. 165
Introduction.............................................................................................................. 165
M materials and M ethods .............................................. .............................................. 166
In Vivo Experim ent ....... ............................................. .. ................................ 166
Spawning Study ..................................... ......................................... ................... 169
Results...................................................................................................................... 173


vi









In Vivo Experiment ......................................................................................... 173
Spawning Study .............................................................................................. 178
Discussion.................................................................................. .... .................... 180
6 IN VITRO STEROIDOGENESIS BY GONADAL TISSUES FROM
FEMALE LARGEMOUTH BASS EXPOSED TO PAPER MILL
EFFLUENTS AND RESIN ACIDS .................................................................... 224
Introduction............................................................................. ......... ........... ....... 224
M materials and M ethods ......... ............................................................. ............ .... 225
Effluent Characteristics...................................................................................... 225
In Vivo Exposures ................... ... .................. .............. .................................. 226
In Vitro Gonadal Cultures ............................................ ............................... 227
Statistical Analyses.......................................... ................................................ .. 228
Results...................................................... .......................................................... 229
Experiment 1................................... ...................................... .................. .......... 229
Experiment 2......................................................................... .......................... 229
Experiment 3...................................................................................................... 230
Discussion................................................................ ............................................... 230
7 GENERAL CONCLUSIONS, ECOLOGICAL SIGNIFICANCE, AND
FUTURE RESEARCH NEEDS ....................................................................... .. .... 242
General Conclusions ........................................................................ ....................... 242
Field Studies ...................................................................................................... 242
In Vivo Studies ................................................................................................... 243
In Vitro Studies ............................................................................................... 245
Ecological Significance ........................................................................................... 245
Future Research Needs ............................................................................................ 247
Additional Field Studies ......................................... ........................................... 248
M esocosms Studies............................................................................................ 248
Evaluate Effects on Other Aquatic Organisms................. ............................ 249
Evaluate Biological Effects of M ill Improvements .............. ....................... 249
M echanistic Studies...........................................................................................250

REFERENCES ........................ ................................................. ................................. 248

BIOGRAPHICAL SKETCH ...................................................... .. ........... ........................ 268














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECTS OF PAPER MILL EFFLUENTS ON THE HEALTH AND REPRODUCTIVE
SUCCESS OF LARGEMOUTH BASS (MICROPTERUS SALMOIDES): FIELD AND
LABORATORY STUDIES

By

Maria S. Septilveda

August 2000

Chairman: Timothy S. Gross
Major Department: Veterinary Medicine

The effects of bleached kraft paper mill effluents (BKME) on the health and

reproduction of largemouth bass (Micropterus salmoides) were examined through field

and laboratory studies. During 1996/97 and 1998, bass were collected from both BKME-

exposed (located at different distances downstream from the effluents discharged by the

Palatka Paper Mill) and reference streams, and parameters compared across sites.

Although concentrations of sex steroids and vitellogenin and induction of

biotransformation enzymes were altered in bass from exposed streams, there were no

differences on gonad weights, fecundities, and age distributions across sites. Some health

endpoints were altered in bass collected from exposed streams, but these fell within

normal ranges and were probably not associated with detrimental health effects.

Laboratory studies that involved exposures of bass to different concentrations of

BKME (10, 20, 40, and 80%) for up to 56 days were conducted during the reproductive









seasons of 1998 and 1999. In contrast to what was observed in the field, bass exposed to

Palatka's BKME responded with changes at the biochemical-level (decline in sex steroids

and vitellogenin) that were usually translated into tissue/organ-level responses (declines

in gonad weights and retardation of gonad development). The majority of these responses

were observed after exposures to at least 20% BKME concentrations. These changes,

however, did not result in lower fecundities, egg sizes, or hatchabilities. Later

evaluations of fry numbers revealed significant negative effects of effluent exposure on

survivorship, with a threshold effluent concentration of 10%. This decline was probably

caused by an increased frequency of deformities coupled with alterations on growth. It

was hypothesized that these changes could have resulted from alterations in "egg quality"

due to chronic failure of parental reproductive systems after almost two months of

effluent exposures, and/or to acute embryo toxicity after translocation of persistent

organic compounds from the mother to the developing embryo.

Results from in vitro cultures showed significant declines in the production of 73-

estradiol by follicles collected from BKME-exposed females. These declines paralleled

changes in plasma 173-estradiol observed in females during the in vivo studies, and

suggested the direct action of chemicals) at the gonad level. There were no dose-

response changes associated with resin acid exposures, which would suggest the action of

chemicals other than resin acids as possible causative agents of the reproductive

alterations observed in BKME-exposed largemouth bass.














CHAPTER 1
INTRODUCTION AND BACKGROUND


Introduction


For over a century, men have used seas, lakes, rivers, and other sources of water

as final resting points for many industrial and agricultural contaminants. This chemical

release to the environment has prompted scientists around the world to evaluate the

potential effects of such pollutants on both human and ecosystem health. Since fish play

a fundamental role in aquatic ecosystems, they have been widely used as monitors of

environmental health and quality. Earlier studies on the effects of environmental

contaminants in fish and other wildlife were focused on examining and reporting obvious

and rather catastrophic responses, such as big kills or die-offs. Because stricter

environmental regulations nowadays have overall decreased the toxicity of chemicals that

are being released to the environment, acute lethality is no longer a likely response in

wild populations inhabiting contaminated areas. Long-term exposure to low

concentrations of pollutants, however, is still of major concern because it can seriously

affect the ability of individual animals to grow, survive to adulthood, and reproduce

leading to population number declines and eventually extinction. Measuring these subtle

responses in fish populations in response to environmental contaminants is the focus of

this dissertation.









The following literature review will first cover information regarding some

general aspects of ecotoxicological studies; a definition of biomarkers of exposure and

effects and of levels of biological organization; and a brief summary on the effects of

endocrine disrupting chemicals in wildlife. Since this dissertation evaluated the potential

effects of paper mill effluents, a general description of the pulp and paper industry and of

the sublethal effects of effluent exposure on fish will follow. Biological and life history

information regarding the study model used in this dissertation (the largemouth bass) is

presented next Finally this first chapter ends with a statement on the significance of this

work and with a description on the way this dissertation was organized.

General Aims of Ecotoxicological Studies

Ecotoxicology is an interdisciplinary science that integrates analytical,

toxicological, and ecological information to predict the fate and adverse effects of

chemicals on ecosystems (Brouwer et al. 1990). It is important to stress, however, that

this kind of information can only be obtained by conducting paired "field" and

"laboratory" studies. Whereas ecological and analytical information is mainly gathered

from field studies, toxicological data is mostly obtained through controlled laboratory

studies. Both types of studies are complementary in nature, and if well designed, should

help decrease the gap between cause and effect relationships and provide useful

information for developing ecological risk assessment models.

Biomarkers of Exposure and Effects

Biomarkers are quantifiable measures of either exposure or effects to

environmental stresses, such as environmental contaminants. The former indicate that

exposure to certain chemical (s) has occurred, but gives no information on potential









effects associated with such an exposure. An example of this type of biomarker would be

the induction of biotransformation enzymes (e.g., cytochrome P450 monooxygenases)

after exposure of fish to planar aromatic and halogenated hydrocarbons (Stegeman et al.

1992). A biomarker of effect, on the other hand, usually measures biochemical,

physiological, or histological adverse changes that result after exposure to certain

chemicals, for example, an increase in fry malformations after exposure of fish to

persistent compounds such as dioxins (Henry et al. 1997). The major goals of the

biomarker approach are to evaluate sublethal effects of chemicals; predict future trends

(i.e. serve as early warning indicators); monitor the distribution, changes, and persistence

of environmental pollutants; and whenever possible, establish cause and effects

relationships (Adams et al. 1989).

Levels of Biological Organization

Exposure to environmental chemicals, however, usually leads to changes at

different levels of biological organization (i.e. molecule, organelle, cell, tissue, organ,

individual, population, community, and ecosystem). This means that when trying to

evaluate the effects of environmental contaminants on fish, a variety of responses at

several organizational levels are needed if biological and ecological meaningful results

are intended. In other words, no single method or index can provide all the necessary

information to understand the condition of a fish population or community. Indicators

that reflect conditions at lower organizational levels (such as molecular and biochemical

levels) respond relatively rapidly to stress and have high toxicological relevance; on the

other hand, indicators that reflect conditions at higher organizational levels (such as









organism and population-levels) respond more slowly and have less toxicological but

more ecological relevance (Adams et al. 1989).

Effects of Endocrine Disrupting Chemicals in Wildlife

Recently, a great deal ot of interest has arisen from the potential effects of

endocrine disrupting chemicals (EDCs) on wildlife and humans. In fact, endocrine-

disrupting effects of environmental contaminants have been observed or suspected in

almost all taxa, ranging from invertebrates gastropodss) to fish, reptiles, birds, and

mammals. At least 45 chemicals or their metabolites have been suggested as having

endocrine-modulating activity that could lead to adverse population-level effects in

wildlife (Colborn et al. 1993). Some of these chemicals include PCBs, DDT, dioxins,

furans, and heavy metals. In addition, there is recent evidence indicating that exposure of

fish to complex mixtures such as effluents discharged by sewage and paper mills, can also

lead to endocrine alterations (Matthiessen and Sumpter 1998). Although the exact

mechanism for the endocrine alterations is largely unknown, most of the concerns have

been towards the potential effects of estrogenic substances. These compounds are

capable of mimicking the action of steroid hormones such as estradiol, thus acting as

partial (weak) or complete estrogen agonists or antagonists (Matthiessen and Sumpter

1998). It is important to recognize, however, that estrogen mimicking is only one of the

many possible mechanisms of endocrine modulation. Besides affecting hormone action,

EDCs can also cause endocrine alterations through changes in biosynthesis, transport, and

metabolism of hormones.






5


The Pulp and Paper Industry


If no specific reference is given in the following text, it is assumed the

information was obtained from either Commission of the European Communities 1989,

US EPA 1995, Thompson and Graham 1997, Erickson et al. 1998, or US EPA Office of

Air Quality Planning and Standards 1998.

Introduction

It is estimated that each American consumes an average of approximately 300kg

of paper products each year, making the U.S. the largest worldwide producer of paper.

The approximately 555 pulp and paper mills in the U. S. manufacture wood pulp, primary

paper products (e.g. printing and writing papers and sanitary tissue), and paper board

products (e.g. container board and boxboard) mainly through the use of cellulose fibers

from timber. Pulp facilities are comprised of mills that produce only pulp (market pulp

facilities, 10% of the total), plants that manufacture paper from pulp produced elsewhere

(non-integrated facilities, 54%), and mills that produce both pulp and paper on-site

(integrated facilities, 36%).

Pulp and Paper Manufacturing Process Sequence

Presently, paper is made out of four basic sources of fiber: hardwood (such as oak,

maple, birch), softwood (pine, spruce, hemlock), recycled paper, and nonwood fibers

(cotton, hemp, flax). Making paper involves five basic steps: (1) fiber furnish

preparation and handling, which involves debarking, slashing, and chipping wood logs

and later screening chips and secondary fibers; (2) pulping, consisting of chemical, semi-

chemical, or mechanical breakdown of pulp into fibers; (3) pulp processing, which









removes impurities and cleans and thickens the pulp mixture; (4) bleaching pulp through

the addition of water and of different bleaching agents on a specific sequence; and finally

(5) stock preparation, which involves mixing, refining, and adding wet additives with

the objective of increasing the strength, gloss, and texture of the final paper product.

Because most of the pollutant releases associated with pulp and paper mills occur at the

pulping and bleaching stages, a detailed description of only these processes will follow.

Pulping techniques

One of the challenges of using wood as a source of pulp is that in addition to the

cellulose fibers, it contains other components (lignin, hemicellulose, and extractives such

as resins, turpentine, tall oil, and soap) that need to be removed for the production of

good-quality paper products. Although hemicellulose and extractives are generally easy

to remove, the removal of lignin is difficult and requires the implementation of some type

of pulping technique. The various methods of pulping can be classified as mechanical,

chemical, or a combination of the two.

The purpose of mechanical pulping is to physically tear the cellulose fibers from

the wood. The oldest method of mechanical pulping is groundwood pulping, and consists

of pressing blocks of wood against a rotating stone and later washing the fibers away

from the stone with water. More modem mechanical pulping techniques include refiner

mechanical pulping (RMP) and thermomechanical pulping (TMP). Because the pulp

produced by mechanical pulping is of low strength and quality, it is mainly used for short-

lived products like newspapers, catalogs and tissue. Mechanical pulping provides pulp

yields of over 90% and accounts for approximately 7% of pulp production in the U.S.









The objective of chemical pulping is to dissolve the lignin bonds holding the

cellulose fibers together. This is achieved by cooking/digesting the wood chips in

aqueous chemical solutions at elevated temperatures and pressures. The choice of

chemicals used in this cooking process, as well as the length of chemical treatment, are

important factors affecting the strength, appearance, and quality of the final paper

product. In contrast to mechanical pulping, chemical pulping produces long, strong and

stable fibers. The two major types of chemical pulping currently used in the U.S. are the

kraft and the sulfite processes. Presently, the kraft or sulfate process is clearly the most

popular method of chemical wood pulping. Its popularity stems from its ability to

produce a high strength pulp with low costs because chemicals are readily recovered and

reused. Lignin removal is high (up to 90%) which allows high levels of bleaching

without pulp degradation due to delignification. This process is also very flexible and can

be used with many types of raw materials. A downside of this technique is that it

produces a very dark brown pulp, which requires the use of extensive chemical bleaching

(see below). The kraft process uses a sodium-based alkaline pulping solution (liquor) that

consists of sodium hydroxide and sodium sulfide in 10% solution. This "white liquor" is

mixed with the wood chips in a digester, with the output products being wood fibers

(pulp) and a liquid that contains the dissolved lignin solids in solution with the pulping

chemicals ("black liquor"). The black liquor then undergoes a chemical recovery process

to regenerate white liquor for the first pulping step. The kraft process has a high pulp

yield (converts about 50% of input furnish into pulp) and produces a very strong pulp

used for manufacturing bags, wrapping paper, container boards, and towels.









The sulfite process uses a solution of sulfur dioxide and calcium bisulfite to

degrade the lignin bonds. It is usually restricted to softwood and non-resinous species of

furnish, and produces pulps that have less color than kraft pulps making the bleaching

process easier. This process is used for the manufacturing of products of average strength

and extreme brightness (such as toilet and facial tissues, napkins, and photographic

paper). In the U.S. chemical pulping accounts for about 60% of pulp production, and

approximately 95% of this is produced using the kraft process.

Semi-chemical pulping combines both chemical and mechanical treatment of

fibers. It consists of chemically treating the wood (using caustic soda, sulfite, or sulfide)

prior to mechanical defibrering. Yields and pulp quality can vary depending on the type

and extent of chemical pretreatment. The neutral sulfite semichemical pulping (NSSC) is

the most frequently used semichemical pulping method. Semichemical pulping allows

for the production of fibers of intermediate length and strength good for the

manufacturing of cardboard and paperboard. This technique accounts for 5% of pulp

production in the U.S.

Bleaching techniques

Bleaching is defined as a chemical process designed to increase the brightness of

the pulp. Bleached pulps create paper products that are whiter, brighter, and softer.

Approximately 50% of the paper products manufactured in the U.S. are bleached in some

fashion. A major factor determining the bleaching potential of a particular pulp is its

amount of lignin. Pulps with high lignin content (mechanical pulping) are difficult to

bleach, whereas chemical pulps can be bleached more efficiently due to their low lignin









content. Since most of the bleaching is done on chemical pulps, the following description

will be focused only on this type of bleaching.

Chemical pulps are bleached in bleach plants where the pulp is processed in

generally three to five stages of bleaching and washing. Bleaching stages generally

alternate between acid and alkaline conditions. In the acid phase, chemicals react with

lignin increasing the whiteness of the pulp, and later alkaline extraction dissolves

lignin/acid reaction products. The product is washed at the end to remove both chemical

solutions. Chemicals used in the bleaching process include hypochlorite (E), elemental

chlorine (C), and chlorine dioxide (D). Because bleaching of pulps with chlorine and

chlorine derivatives results in the production of chlorinated pollutants such as dioxins, a

recent major trend in the industry has been the reduction in both the types and amounts of

such chemicals used for pulp bleaching. In fact, many European mills have developed

bleaching processes that are totally chlorine free (TCF) and that use chemicals such as

ozone, oxygen, hydrogen peroxide, peracetic acid, and enzymes as bleaching agents. The

use of chlorine dioxide has also steadily increased relative to elemental chlorine due to its

reduction in the formation of chlorinated organic. Also, significant improvements have

been made to improve delignification in order to minimize dioxin formation while

reducing bleach chemical usage. Some of these delignification technologies include

extended delignification during kraft pulping, solvent pulping, and oxygen

delignification.

Pollution Outputs

The pulp and paper industry has historically been considered a major consumer of

natural resources and a significant contributor of pollutant discharges to the environment.









The stages of pulping and bleaching are considered the major sources of pollutant outputs

to air, water, and land, most of these being released by bleached kraft mills (effluents

released by these mills are referred to as Bleached Kraft pulp Mill Effluents or BKME).

The process of making pulp and paper is characterized by an intensive use of water.

Indeed, the pulp and paper industry is the largest industrial water user in the U. S., with

an average industry total discharge of 16 million m3/day of water. The main water

pollution concerns are total suspended solids (TSS), biological oxygen demand (BOD),

chemical oxygen demand (COD), total organic carbon (TOC), color, and turbidity. In

addition, toxicity concerns arise from the presence of chlorinated organic compounds

such as dioxins, furans, and others (collectively referred to as adsorbable organic halides

or AOX) after the chlorination sequence. Recently, additional concerns have arisen from

the potential toxic effects of natural components of wood (resin and fatty acids, and

phytosterols) on aquatic organisms.

Wastewater Treatment Technology

It was estimated that during 1993, the pulp and paper industry produced about 2

trillion pounds of waste. About 90% of this waste was managed on-site through recycling

(5% of the total), energy recovery (10%), or treatment (75%). Pulp and paper mill plants

in the U.S. operate treatment facilities (primary, secondary and/or tertiary) to remove

BOD, TSS, and other pollutants (such as AOX) before discharging their effluents into a

receiving waterway. Primary treatment mainly involves the mechanical removal of

suspended solid fibers through sedimentation. Secondary treatment relates to biological

degradation of effluents mainly through the use of aerated stabilization basins or

oxygenated activated sludge. Both methods are based on accelerating nature's process of









reducing wastes to carbon dioxide and water using aerobic microorganisms, which will

lead to significant reductions in BOD. Tertiary treatments use chemicals (such as ferric

and aluminum oxide) to help increase the quality of the effluent being released.

The Pulp and Paper Cluster Rules

The pulp and paper cluster rules were promulgated in 1998 by the U.S. EPA as a

way to simplify compliance by coordinating the regulation of industrial pollution. The

major goals of this coordinated regulator approach are to provide a greater protection of

human health and the environment; to reduce the costs of complying with wastewater and

air emission regulations; and to promote and facilitate pollution prevention. These rules

consist of regulations that specify both air emission standards (through the national

emission standards for hazardous air pollutants (NESHAP)) and water effluent discharges

(through the effluent limitations guidelines and standards, pretreatment standards, and

new source performance standards). In general, the NESHAP requires mills to collect

and control pulping and bleaching processes vent emissions and to eliminate the use of

certain bleaching chemicals. Effluent regulations include best management practices to

prevent leaks and spills of pulping liquor; specification of new analytical methods for 12

chlorinated phenolics pollutants and for AOX; and a voluntary advanced technology

program designed to encourage mills to install more pollution prevention technology than

required by regulations.









Georgia-Pacific's Paper Mill Plant in Palatka, Florida


General Description

The Palatka paper mill plant has been in operation since 1947. This mill has two

bleaching lines (40% product) and an unbleached line (60% product), which together

release an estimated 36 million gallons of effluent daily. Treated effluents are discharged

into Rice Creek, a small tributary of the St. Johns River. Rice Creek runs for about 5km

prior to its confluence with the St. Johns River. Because Rice Creek is a small tributary,

effluents can account for a large portion of its total flow (yearly average effluent

concentration is estimated to be around 60%, with a range of 50% to 97%) (Myra

Carpenter, personal communication). By the time effluents reach the St. Johns River,

concentrations have fallen below 10%. It should be noted however, that these

concentrations are higher when compared to the majority of paper mills in the U.S.,

where average effluent dilutions range from less than 1% to about 5%.

In this plant, the bleaching sequences for the bleach lines are CEHD and

CgodioEopHDp, where Cd = mixture of chlorine (C) and chlorine dioxide (d) in

proportions designated by subscripts; Eop = extraction with alkali and the addition of

elemental oxygen (o) and hydrogen peroxide (p); H = hypochlorite; and Dp = 100% d

substitution with the addition of p. The bleaching lines manufacture paper towels and

tissue paper, whereas the unbleached line produces mainly kraft bag and linerboard. The

wood furnish of this mill consists typically of 50% softwood species (mainly loblolly,

slash, sand, and pine) and 50% hardwood (mainly tupelo, gums, magnolia, and water

oaks). At the time of this study, effluents received secondary treatment, which consisted









of both anaerobic followed by aerobic biological degradation during a retention period of

40 days.

It should be noted that the bleaching sequence employed in the Palatka Operation

does not represent about 75% of the mills in the U. S. This is because of basic toxicity

concerns related to chlorinated phenolics and other chlorinated species, which prompted

most mills to eliminate elemental chlorine bleaching and replace it with an elemental

chlorine free (ECF) process using 100% chlorine dioxide in the first stage of bleaching.

Data has shown that conversion to ECF can reduce water quality concerns substantially.

Therefore, using the Palatka mill as a model of paper mill effluent effects should be

considered a "worst-case" scenario.

Ongoing Improvements

Georgia-Pacific has studied what mill improvements will be necessary to comply

with the U. S. EPA cluster rule promulgated in 1998. Some improvements that will be

implemented in the next few years include the use of chlorine dioxide bleaching instead

of elemental chlorine and also use of oxygen and hydrogen peroxide bleaching instead of

sodium hypochlorite. In addition, improvements of secondary treatment of effluents to

reduce BOD are currently underway.

Sublethal Physiological Effects of Pulp and Paper Mill Effluents on Fish


Over the past 25 years, considerable effort has been devoted to determining the

nature and extent of fish responses to pulp and paper mill effluents. The following is a

brief review of sublethal effects (ranging from the biochemical to the organism level)

measured in fish exposed to these complex effluents. For the purpose of this review,









responses have been grouped into the categories of general, liver, and reproductive health

effects.

General Health Effects

Growth

Growth can be considered as one of the ultimate indicators of health because it

integrates most of the biotic and abiotic variables acting on an organism (Goede and

Barton 1990). Laboratory experiments have shown that paper mill effluents can

negatively affect growth rates in fish (Warren et al. 1974). Similarly, Munkittrick et al.

(1991) reported that white suckers (Catostomus commersoni) collected from a site

receiving primary-treated BKME were shorter, lighter, and grew slower than fish from

reference sites. In contrast, Swanson et al. (1992) found no differences in growth rates

between contaminated and reference populations of longnose sucker (Catostomus

catostomus) and mountain whitefish (Prosopium williamsoni), and Servizi et al. (1992)

reported no effects of treated BKME on growth of Chinook salmon (Oncorhynchus

tshawytscha) in the laboratory.

Hematology

Hematology is defined as the study of blood and blood-forming or hematopoietic

tissues (primarily spleen and kidney in fish). Fish exposed to paper mill effluents may

respond by either decreasing (anemia) or increasing (polycythemia) several hematological

variables. Many field and laboratory studies have reported anemia in fish due to a decline

in the number of red blood cells and/or in hemoglobin concentrations after exposure to

BKME (Everall et al. 1991, Swanson et al. 1992, Khan et al. 1996, Soimasuo et al.









1998). It has been postulated that declines in hemoglobin may result from increased

breakdown of red blood cells hemolysiss), since this phenomena has been induced in vitro

after exposure of red blood cells to resin acids (Bushnell et al. 1985). Increases in

hematocrit values probably due to disturbances in ion regulation and/or to stress-induced

polycythemias have also been reported in fish sampled downstream from paper mills

(Oikari et al. 1985, Hodson et al. 1992) and in fish exposed to chlorinated compounds

present in BKME (Bengtsson et al. 1988). In a field study on the effects of BKME

exposure on perch (Percafluviatilis), although there was a decline in hemoglobin

concentrations in polluted stations, this decline was associated with an increase in the

number of red blood cells (Larsson et al. 1988). These authors concluded that this

increased erythropoiesis was likely due to an increased oxygen demand as a response to

the high detoxification activity associated with exposure to these effluents. Several

studies, however, have found no effects on hematology of fish exposed to BKME. For

instance, chronic exposure (210 days) of Chinook salmon to treated BKME (0.3 to 4%

v/v) had no effect on hematocrit (Servizi et al. 1992), and Kennedy et al. (1995) found no

changes in hemoglobin in rainbow trout (Salmo gairdneri) exposed to resin acids for 24

hours. Finally, Swanson et al. (1992) reported no differences in packed cell volume

between fish sampled close to a paper mill area and reference fish.

Spleen histology

The spleen is one of the primary hematopoietic organs in fish, and thus

histological alterations in this tissue could explain some of the hematological changes

described above. Hemosiderosis (accumulation of hemosiderin, an endogenous pigment









that results after the breakdown of hemoglobin, within the spleen cells) has been observed

in spleens of fish naturally exposed to BKME (Khan et al. 1992, Mercer et al. 1997).

Osmoregulation

Structural changes in the gills have been observed in fish exposed to BKME,

which may lead to disruptions in oxygen diffusion and osmoregulatory functions. The

literature on this subject is confusing, however, since a wide array of electrolyte changes

(increases, decreases, and no effects) have been reported in fish exposed to BKME

(Oikari et al. 1988, Larsson et al. 1988, Lindstrom-Seppai and Oikari 1989, 1990,

Lehtinen et al. 1990, Everall et al. 1991, Swanson et al. 1992, Jeney et al. 1996).

Immune function

Exposure of fish to BKME may increase the circulating levels of corticosteroids

leading to immunological system disruptions, such as reductions in leuccorit and in

immunoglobulins (Jokinen et al. 1995, Soimasuo et al. 1995a, 1995b, Khan et al. 1996).

These changes in turn can result in an increased susceptibility to pathogens such as

bacteria and parasites. Kennedy et al. (1995) exposed juvenile trout to sublethal

concentrations of chlorinated resin acids for 24 hours and observed a reduced resistance

to infection by Aeromonas salmonicida. Several studies have also reported an increase in

the prevalence and intensity of infection with ecto and endoparasites in fish exposed to

pulp and paper effluents (Thulin et al. 1988, Axelsson and Norrgren 1991, Khan et al.

1992, 1994b).









Liver Health Effects

Because the liver is the primary organ for the biotransformation and excretion of

xenobiotics, the evaluation of its health and functioning is fundamental in any study on

the effects of environmental contaminants. Some of these measurements include analyses

of liver enzyme activities and evaluations of liver weights and histology. Biochemical

responses of fish to chemical stimuli have been studied extensively over the past years,

and the increase in monooxygenase enzyme activity (measured as ethoxyresorufin-O-

deethylase or EROD activities) in fish livers sampled downstream of BKME is a good

example of this. One of the major advantages of using biochemical responses as

indicators of contaminant exposure is their high sensitivity and rapid response time. A

major problem with this approach, however, is that the exact biological significance of

these changes for the functional integrity of the organism is poorly known (Thomas

1990). In addition, factors such as temperature, age, sex, and nutritional status of fish can

modify the activity of these detoxification enzymes, which could complicate the

interpretation of induction responses in fish (Jimenez and Stegeman 1990).

Liver enzymes

Cytochrome P450 refers to a family of enzymes involved in the biotransformation

of organic chemicals, resulting in either their activation to toxic metabolites or their

inactivation. P450 systems in fish are inducible by different types of endogenous and

exogenous compounds, a process that involves synthesis of new messenger RNA, and

thus of new enzyme protein (Stegeman et al. 1992). Since EROD activity is catalyzed by

P450 monooxygenases, an increase in EROD activity is indicative of P450 induction.

Measurements of EROD activity have been widely used as a biomarker for exposure of









fish to several groups of chemicals, including polychlorinated dibenzo-p-dioxins

(PCDDs) and dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs), polycyclic

aromatic hydrocarbons (PAHs), pesticides, metals, and natural biogenic substances.

Because BKME have been reported to contain EROD-inducing compounds, this

biomarker has played a major role in the study of fate and biological effects of paper mill

effluents. In general, researchers have reported background EROD activities in fish from

reference sites, with significant increases in areas close to pulp mill outfalls (Forlin et al.

1985, Lindstrom-SeppAi and Oikari 1989, Courtenay et al. 1993, Bankey et al. 1994,

Soimasuo et al. 1995b). Until recently, it was believed that the main inducers in mill

effluents were chlorinated persistent compounds (such as PCDDs and PCDFs) (Hodson

1996). However, new evidence suggests that enzymatic EROD induction also occurs in

fish exposed to unbleached effluents, and that the compounds) responsible for such

induction are not of the highly hydrophobic chlorinated type, but rather of the moderately

hydrophobic planar PAH-type form present as natural components of wood, and readily

metabolized by fish (Hodson 1996).

The effect of treated vs. untreated paper mill effluents on EROD activity in fish

has only recently been addressed and the results are so far inconclusive. Martel et al.

(1996) found that 17 of 46 primary and secondary-treated paper mill effluents did not

cause significant mixed-function oxygenase (MFO) responses in fish. In a later study,

Martel and Kovacs (1997) reported a significant increase in EROD activity in rainbow

trout exposed to primary-treated effluent compared to fish exposed to secondary-treated

effluents. In contrast, a field study revealed that EROD induction in wild fish was not

eliminated after the installation of a secondary treatment facility (Munkittrick et al.









1992a). Also, wild European carp (Cyprinus carpio) exposed to treated pulp mill

effluents had higher elevated hepatic EROD levels relative to reference fish (Ahokas et

al. 1994). In the latter study, EROD activity was strongly correlated with water AOX

levels, and poorly related with fish muscle and sediment extractable organic halogen

(EOX) levels. Gagn6 and Blaise (1993) also noted that EROD activity in rainbow trout

increased in fish exposed to sublethal concentrations of both primary and secondary-

treated effluents, but that the degree of increase was higher in the primary-treated exposed

group.

Phase II (conjugating) enzymes have also been studied in fish exposed to BKME.

The two most important conjugating enzymes studied include glutathione S-transferases

(GSTs) and UDP-glucoronosyltransferases (UDPGT). Cytoplasmic GSTs are a multi-

gene family of proteins that participate in detoxification processes by conjugating many

electrophilic compounds with glutathione (GSH) to produce more soluble and thus

excretable products (George and Buchanan 1989). Studies on the detoxification capacity

of effluent-exposed fish have reported both increases (Oikari et al. 1988) and declines

(Mather-Mihaich and Di Giulio 1991, Bucher et al. 1993) in hepatic GSH concentrations.

GST activity, on the other hand, has generally been found unaltered after exposure to

BKME (Soimasuo et al. 1995a, 1995b).

UDP-glucoronosyltransferase enzymes catalyze the transfer of glucoronyl groups

from uridine 5'-diphosphoglucuronate to many acceptors including PAHs and various

endogenous compounds (Stegeman et al. 1992). Field and laboratory studies on the

effects of BKME on fish have reported inductions, decreases, and no effects on UDPGT

activity (Forlin et al. 1985, Lindstrom-Seppai and Oikari 1988, Lindstrom-Seppii et al.









1989). Inhibitory effects on UDPGT activity (by up to 85%) have been observed in

rainbow trout after exposure to trichlorophenol, pentachlorophenol, and dehydroabietic

acid (all common components of paper mill effluents) (Andersson et al. 1988). This type

of inhibition can have important consequences, since it can not only reduce the ability of

UDPGT to metabolize xenobiotics but may also affect its role in metabolizing

endogenous compounds. This could explain the increase in bilirubin (a substrate of

UDPGT) in blood of BKME-exposed fish (Oikari and Nakari 1982).

Carbohydrate metabolism

Disturbances in carbohydrate metabolism have been observed in fish exposed to

BKME. It has been postulated that these effluents are capable of causing internal hypoxia

thorough gill damage (Davis 1973), which can lead to increased blood glucose levels and

depletion of liver glycogen. Exposure of coho salmon (Oncorhynchus kisutch) to an

effluent concentration equivalent to 0.8 of the 96-h LC50 produced an immediate

hyperglycemia, and after 48h of exposure liver glycogen concentrations had decreased to

almost zero (McLeay and Brown 1975). In another study, Oikari and Nakari (1982)

exposed trout to components of paper mill effluent for 11 days and observed an

exhaustion of liver glycogen reserves.

Blood glucose is one of the most commonly used parameters utilized for

measuring stress. The classic stress response involves an elevation of blood sugar in

response to the hormones adrenaline and cortisol. In rainbow trout exposed to

chlorinated phenolics and resin acids, plasma glucose concentrations were higher

compared to control fish, and concentrations remained high throughout the 40-day

experiment (Tana 1988). Hyperglycemia has also been reported from perch sampled









from an area contaminated with paper mill effluents (Andersson et al. 1988) and from

rainbow trout artificially exposed to chlorinated resin acids (Kennedy et al. 1995). Some

studies, however, have failed to detect changes in liver glycogen and/or blood glucose

concentrations in fish after exposure to BKME (Oikari et al. 1988, Swanson et al. 1992,

Soimasuo et al. 1998).

Liver histology

Livers of male bullheads (Cottus gobio) sampled close to an area affected by

paper mill effluents had a high incidence of fatty degeneration, fibrosis, necrosis, and

parasitism (Bucher et al. 1992). Khan et al.(1994a) also found that winter flounder

(Pleuronectes americanus) taken from areas contaminated with BKME had livers with

varying degrees of vacuolation and multifocal hemosiderosis. Similarly, in three-spined

stickleback (Gasterosteus aculeatus) chronically exposed to pulp mill effluents (5 1/2

months) several anomalies in the liver were observed (necrosis, nuclear pyknosis,

vacuolation, and fat accumulation) (Axelsson and Norrgren 1991). Servizi et al. (1992)

reported an increase in the incidence of hepatic granulomas in Chinook salmon

chronically exposed (for up to 210 days) to treated BKME. In contrast, Mather-Mihaich

and Di Giulio (1991) found no histopathological changes in liver of channel catfish

(Ictalurus punctatus) exposed to BKME for up to 14 days, and Mercer et al. (1997)

reported that 84% of cunner (Tautogolabrus adspersus) sampled at a reference site

showed evidence of vacuolation in the liver in contrast to 53% from the vicinity of the

paper mill.









Hepatosomatic index

Ratios of organ weight to body weight have been reported by several authors

when studying the effects of paper mill effluents on fish. The hepatosomatic index (HSI)

is calculated by dividing the weight of the liver by the body weight of the fish and

multiplying the resulting number by 100. Munkittrick et al. (1992a) reported that, after

the initiation of secondary treatment of BKME, liver weights of lake whitefish

(Coregonus clupeaformis) decreased by 37% in females and by over 50% in males

compared to values found at reference sites. Somatic indices were also decreased in

winter flounder inhabiting an inlet under the influence of pulp and paper mill effluent

(Khan et al. 1992). Similarly, Bucher et al. (1992) reported a decrease ("shrinkage") of

livers from bullheads sampled at the end of the low-water period in a river contaminated

with BKME. In this study, enlarged livers were found only following the high-water

period. The authors concluded that this change in liver weights was directly related to a

change in glycogen content in the hepatocytes (Bucher et al. 1992). On the other hand,

enlarged livers have also been reported in fish contaminated with paper mill effluents

(Larsson et al. 1988, Andersson et al. 1988). These authors attributed the increase in HSI

in BKME-exposed fish to proliferation of the endoplasmic reticulum, as well as to

increased fat accumulation.

Reproductive Health Effects

From recent studies in several fish species, there is substantial evidence that

exposure to paper mill effluents can result in reproductive alterations, such as delayed

sexual maturation, altered secondary sex characteristics, reduced gonad weights, decline

in the production of eggs and in their sizes, and decreased concentrations of sex steroids.









Although the compounds responsible for these effects have not yet been identified, it has

been hypothesized that these changes might be related to exposure to natural components

of wood (such as resin acids, sterols, and lignins), which have been reported to have weak

estrogenic activity (Van Der Kraak et al. 1998).

Results from studies on white sucker from Jackfish Bay, Canada, indicate that

several sites within the pituitary-gonadal-axis are affected after exposure to BKME. Fish

from exposed sites had significantly lower plasma levels of gonadotropin (GtH-I) and

showed depressed responsiveness of sex steroids and 17,20B-dihydroxy-4-pregnen-3-one

(17,208-P, a maturation-inducing steroid) after injections with gonadotropin releasing

hormone (GnRH) (Van Der Kraak et al. 1992). BKME-exposed fish also had lower

circulating levels of testosterone glucoronide, which would be suggestive of altered

peripheral steroid metabolism. Similarly to what was observed under in vivo conditions,

in vitro incubations of ovarian follicles collected from BKME-exposed females have also

shown reduced production of testosterone, 1713-estradiol, and 17,208-P 2 under basal and

human chorionic gonadotropin stimulated conditions (Van Der Kraak et al. 1992,

McMaster et al. 1995). The similarities between both types of studies would suggest that

reductions in plasma steroid levels in BKME-exposed fish from Jackfish Bay are mainly

due to alterations in ovarian steroid production.

Age at maturity

This parameter is defined as the age (in years) in which spawning first occurs. In

perch exposed to BKME, less than 50% of the potentially mature males had developed

gonads, compared to 80% at the reference site (Sandstr6m et al. 1988). Similarly, lake

whitefish and white sucker exposed to primary treated BKME exhibited delayed sexual









maturity relative to reference populations (Munkittrick et al. 1991, 1992a). In a study on

the effects of paper mill effluents on longnose sucker and mountain whitefish, Swanson et

al. (1992) found no differences in age at maturity between contaminated and reference

populations.

Secondary sex characteristics

These are traits that distinguish males from females but that are not responsible

for the production of gametes. Mature male white suckers naturally exposed to paper mill

effluents showed no evidence of secondary sexual characteristics (nuptial tubercles) in

relation to males sampled from a reference site (Munkittrick et al. 1991). Female

mosquitofish, Gambusia affinis, inhabiting a stream receiving paper mill effluents in

Florida were reported to be strongly masculinized showing both physical secondary sex

characteristics (fully developed gonopodium) and reproductive behavior of males

(Howell et al. 1980). More recently, masculinization of female fish has been identified

from an additional two species (least killifish, Heterandriaformosa and sailfin molly,

Poecilia latipinna) collected from Rice Creek, the stream receiving the effluents

discharged by the Palatka mill (Bortone and Cody 1999). Masculinization of female fish

has been attributed to the action of androgenic hormones that result from the

biotransformation of plant sterols (and also cholesterol and stigmasterol) by bacteria such

as Mycobacterium (Howell and Denton 1989).

Gonadosomatic index

Changes in gonad weights in relation to body weights (gonadosomatic indices or

GSIs) are routinely used as a way to assess reproductive effects in fish exposed to paper

mill effluents and other environmental contaminants. Several studies have reported









declines in GSIs in fish exposed to BKME (Larsson et al. 1988, Munkittrick et al. 1991,

1992a, 1994, Gagnon et al. 1994b, Gibbons et al. 1998). However, there is also evidence

to suggest that decreases in gonadal size in response to declines in sex steroids may not

always occur after exposures of fish to BKME (McMaster et al. 1996b), which would

indicate differences in reproductive responsiveness to contaminant exposure across

species.

Fecundity and egg size

There are relatively few studies on the effects of BKME on egg parameters, and

the results from these studies are conflicting. Many field and laboratory studies have

reported declines in fecundities of several fish species after exposure to paper mill

effluents (Landner et al. 1985, Munkittrick et al. 1991, Gagnon et al. 1994b, 1995,

Kovacs et al. 1995). Fecundities, however, were not altered after exposures to BKME in

several other field (Karis et al. 1991, Swanson et al. 1992, Adams et al. 1992) and

laboratory studies (Kovacs et al. 1996). Sandstram et al. (1988) reported that developing

eggs from female perch sampled close to a bleachery outlet were smaller and more

irregular in shape compared to controls. In lake whitefish naturally exposed to BKME,

Munkittrick et al. (1992a) reported that even though females had a higher fecundity

compared to females from a reference site, these eggs were smaller. McMaster et al.

(1991) also reported reduced egg size in white sucker females exposed to BKME.

Sex steroids

The most important reproductive hormones in teleost fish are testosterone, 11-

ketotestorene, and 178-estradiol. They are produced by the gonads and their

measurement in plasma is a good indicator of reproductive status, seasonality, and









gonadal function. One of the most consistent findings in studies that have focused on the

effects of BKME on reproductive parameters of fish is a decline in the concentration of

sex steroids in plasma of exposed animals. BKME-exposed white suckers from Jackfish

Bay, Lake Superior show decreased concentrations of several sex steroid hormones

(testosterone, 11 -ketotestosterone 176-estradiol, and 17, 20f3-dihydroxy-4-pregnen-3-one)

(Munkittrick et al. 1991, McMaster et al. 1995, 1996b). Declines in steroid

concentrations have also been documented in longnose sucker and lake whitefish from

Jackfish Bay (Munkittrick et al. 1992a, McMaster et al. 1996b), in white sucker at other

mills (Hodson et al. 1992, Munkittrick et al. 1994, Gagnon et al. 1994a), and in other fish

species sampled elsewhere (Adams et al. 1992, McMaster et al. 1996b). The

consequences of these similar endocrine alterations to whole animal reproductive fitness

and population dynamics, however, have varied greatly between species. For example,

longnose sucker exposed to BKME show no organism responses other than an altered age

distribution, whereas white sucker and lake whitefish show decreased gonadal sizes,

secondary sexual characteristics, and egg sizes, and increased age to maturity (McMaster

et al. 1996b). In a review of whole organism responses of fish exposed to different kinds

of mill effluents (including unbleached pulps), 48% of the populations studied had

increased condition factors, 80% showed increased age to sexual maturation, and reduced

gonadal size was reported in 58% of the studies (Sandstr6m 1996). These observations

provide evidence for species differences in susceptibility to BKME, but also show the

inherent difficulty when trying to compare biological responses in fish populations

inhabiting highly different environments and exposed to complex mixtures likely to vary

in chemical composition.









Vitellogenin

Vitellogenesis involves the synthesis of vitellogenin by the liver, its uptake by

growing oocytes, and its storage as yolk to serve as source of food for the developing

embryos. There is little information on the effects of BKME on plasma vitellogenin

concentrations. Plant sterols, such as j3-sitosterol, which are commonly found in pulp

mill effluents, are estrogenic compounds known to bind in vitro to rainbow trout hepatic

estrogen receptors (Tremblay and Van Der Kraak 1998) and can induce vitellogenin

synthesis in male goldfish (Carassius carassius) (MacLatchy and Van der Kraak 1995).

The Largemouth Bass (Micropterus salmoides)


This section outlines some important biological information regarding the fish

species used as the study model during the course of this dissertation. Except for the

information presented under general description and geographical distribution, this

summary is focused on presenting data related to the Florida subspecies of largemouth

bass.

General Description

There are two recognized subspecies of largemouth bass: the Florida (M.

salmoides floridanus) and the northern (M. salmoides salmoides) subspecies.

Morphologically, they are both very similar with the major differences being a larger

number of scale rows and pyloric caeca in the Florida subspecies (Bailey and Hubbs

1949). Largemouth bass are characterized by having a robust and compressed body that

can measure up to 876 mm in length, and a long head and mouth that contains brush-like

teeth on both jaws and that extend to cover the palatines, vomer and pharynx (Hardy









1978). Their pigmentation is dark brown to olive with a silvery sheen and a conspicuous

black lateral stripe that fades with age (Chew 1974).

Geographic Distribution

In North America, the northern largemouth bass has a wide geographic

distribution ranging from Mexico to southern Canada (Hardy 1978). In the U. S., they

can be found in many states including Virginia, West Virginia, Texas, Oklahoma, Kansas,

Nebraska, Iowa, Minnesota, Wisconsin, North Dakota, New York, Pennsylvania, Ohio,

Maryland, and Florida (Hardy 1978). Although M. salmoidesfloridanus is endemic to

Florida, in northern areas of the state its range overlaps with that of M. salmoides

salmoides (Bailey and Hubbs 1949).

Habitat and Range

In Florida, largemouth bass can be found in most habitats (lakes, ponds, bayous,

marshes, sloughs, impoundments, rivers, and creeks), except some low-oxygen boils.

They occur over all types of substrates (mud, muck, organic debris, sand, clay, and

gravel) and depths, but prefer shallow, vegetated areas (water lilies, cattails, and pond

weed) (Chew 1974, Hardy 1978). Known water quality parameter ranges under field

conditions for the species are: 0.56 35.00C for temperature (with an optimum of 26.6 -

27.70C); 4.7 11.0 for pH; and a maximum salinity of 32.1 ppt (Hardy 1978).

Information on the range of movement of Florida largemouth bass suggests the

presence of both mobile and sedentary populations. Snyder et al. (1986) reported that

38% of the bass marked and released in the lower St. Johns River were recaptured in the

same area are as tagged, and of the remaining 62%, 44% had moved a distance of less









than 2km. In another study, 84% of specimens tagged moved less than 8km, with a

maximum distance of 20km (cited by Hardy 1978).

Growth and Feeding Habits

Growth rates in Florida largemouth bass are higher when compared to their

northern counterpart, probably due to both intrinsic factors as well as to the more

favorable environment and extended growing season present in southern latitudes

(Clugston 1964). In this respect, fry growth is directly related to water temperature, with

minimum and maximum growth rates at temperatures below 17.5 and above 25C,

respectively. In addition, growth rates are known to vary with age (are highest during the

first two years and decreases after fish reach sexual maturity) and season (are highest in

summer and fall and lowest in winter and spring) (Chew 1974). For example, in bass

from Lake Weir, Florida, growth rates from hatching to 1 year of age, and from ages 1 to

2, were estimated at 0.383mm/day and 0.255mm/day, respectively (Chew 1974).

There is a well documented shift in the diet of largemouth bass in relation to age.

Approximately 40 hours after they leave the nest, larvae feed almost entirely on

crustaceans (mainly copepods and cladocerans) (Kramer and Smith 1960). By the time

bass reach 50mm, insects and fish are the next prey items to become incorporated in the

diet. Studying the food habits of largemouth bass inhabiting the St. Johns River, Mclane

(1949) also demonstrated a progressive change in diet, from zooplankton and macro-

invertebrates (cladocerans, decapods, and insects (larvae, pupae, and nymphs)) in fry and

juveniles, to almost exclusively fishes in adults.









Reproduction

Sexual maturity

In Florida, largemouth bass reach sexual maturity at a size of 250mm, i.e. within

the first year of age. Minimum size at maturity has been reported at 140mm. A restricted

growing season in northern climates precludes attainment of sexual maturity in one year,

and extends it to 2-4 years of age instead (Hardy 1978).

Spawning

Florida largemouth bass are capable of successfully spawning anywhere from

mid-November through August, with peaks in February and March (Clugston 1966).

Induction of spawning is mainly triggered by increases in water temperature during the

spring. Spawning usually occurs near dusk or dawn, with maximum activity at water

temperatures between 20 and 24C and no spawning has been observed at temperatures

below 180C and above 270C (Clugston 1966).

Nests are built in 0.6 to 1.2m of water, although depths can range from 10cm to

over 2m (Carr 1942). Males of the species are in charge of constructing the nests usually

through excavation of substrate, although in some instances no nest is prepared and eggs

are deposited directly on aquatic vegetation. Nests are often built in open areas in

association with various aquatic plants and on different types of substrates ranging from

fibrous organic debris to bare sand (Kramer and Smith 1960). Nests measure from 30.5

to 152.4cm, are round, and are located between 1.2 and 6.4m offshore at spacing intervals

of 1.8 to 2.1m (Carr 1942, Hardy 1978). Careful spacing of nests is related to the strong

territorial behavior exhibited by males during the spawning season. Fecundity is highly








variable ranging from 2,000 to 145,000 eggs, and appears to be directly related to age and

condition of the fish, as well as to environmental parameters such as water temperature

(Chew 1974, Hardy 1978).

Reproductive cycles

As the spawning season approaches, the percent of gonad weight to total body

weight (the gonadosomatic index or GSI) increases. In Florida largemouth bass (age

class I) reared in Texas, GSIs peaked in March in females (to about 5%), and at a slighter

later date (late April) in males (to about 0.9%) (Rosenblum et al. 1994). In northern

largemouth bass from Tennessee the peak in female GSIs occurred later, between mid

April (age class II, to about 5%) and mid May (age class I, to about 5.5%) (Adams and

McLean 1985). Seasonal changes in organ somatic indices of M. salmoidesfloridanus

(GSI and hepatosomatic index or HSI) have also been observed in bass sampled from

lakes in central Florida, with highest values in January and February (Timothy Gross,

unpublished data). Seasonal cycles of sex steroids and vitellogenin have also been

studied in females from Lake Woodruff, Florida (Timothy Gross, unpublished data). This

study reports peaks in 173-estradiol, testosterone, and 11-ketotestosterone in February at

concentrations of 3,892, 2,167, and 971 pg/mL, respectively. Vitellogenesis, on the other

hand, begins in September and peaks in January with values of 6.3mg/mL of vitellogenin

in plasma (Timothy Gross and Nancy Denslow, unpublished data). As with other teleost

species, the predominant sex steroids in female and male largemouth bass during the

reproductive season are 17p-estradiol and 11-ketotstosterone, respectively, and

vitellogenin in females is usually found at concentrations that are about 12 times the

values reported in males.









Eggs and fry

The eggs of the largemouth bass are spherical or oval, adhesive and demersal.

They are light yellow to orange, and contain one large oil globule that measures 0.34 -

0.54mm in diameter and that persists throughout the entire embryonic and larval stages

until the yolk is completely reabsorbed (Chew 1974, Hardy 1978). Unfertilized eggs

measure between 0.75 1.7mm in diameter, whereas fertilized eggs are larger measuring

between 1.3 and 1.95mm (Hardy 1978). After deposition, eggs always lie with the oil

globule uppermost, and water-harden within 15 minutes (Carr 1942). Embryonic

development can be summarized as follows: first mitotic divisions of the ova begin about

one hour after fertilization; blastula stage is reached at 3.25hr; blastoderm at 5.25hr;

gastrula at 14hr; early embryo at 21.5hr; and late embryo stage occurs at 37hr post-

fertilization (Chew 1974). Under laboratory conditions, hatching has been reported to

begin between 45 and 47hr after fertilization (Carr 1942, Chew 1974). Larvae are highly

active about 77hrs post-fertilization, and can swim off the bottom about 73hrs later. The

mouth is fully formed after 167hrs, and initial feeding has been observed after 206hrs, or

approximately on the 8t day, although the yolk is not yet fully absorbed at this time

(Chew 1974). This sequence of events agrees with what has been reported to occur under

field conditions. For example, Carr (1942) found that in Lake Bivans Arm, Florida, eggs

hatched 50 to 60hrs after fertilization, and larvae began to leave the bottom 4 days after

hatching. In this study, regular feeding and schooling was recorded on the 8h day. In

another study, rise from the nest was reported to occur at fry lengths of 5.92 6.31mm

(average of 6.16mm) (Kramer and Smith 1960). Time of rising from the nest is inversely








correlated with temperature (from 7.2 days at temperatures between 13 and 18C, to 6.0

days at temperatures ranging from 21 to 240C) (Hardy 1978).

Survival

Kramer and Smith (1962) found that the success of a year-class of largemouth

bass was determined within the first two weeks of spawning. They concluded that the

major mortality factors at this time were related to water temperature and wind (major

drops in water temperature and strong winds caused high mortalities). They also found

that food availability, predation, and fecundity of the spawning stock did not play a major

role in the observed mortalities.

Significance of this Work


Field experiments provide data that is easily related with the natural occurring

studied phenomena (external validity) while laboratory experiments provide the settings

to adequately control the effects of confounding variables (internal validity). Because of

this trade-off between external and internal validity the effects of BKME on health and

reproduction of largemouth bass were studied through paired laboratory and field studies.

The primary objective of the field studies was to determine whether reproductive

and health parameters were altered in fish populations inhabiting streams contaminated

with BKME. The primary objective of the experimental studies was to determine dose-

related effects in captive fish exposed to paper mill effluents. For these experiments,

effluent concentrations were calculated based upon the current range of environmental

concentrations of effluent reported for Rice Creek and the St. Johns River (10 90%).

Exposure periods were designed to reflect multiple endpoints throughout the reproductive









season. Exposure of BKME at concentrations likely to be encountered by free-ranging

fish over a given period is fundamental for any risk assessment. In addition, most toxic

effects are based on lethality. However, sublethal effects, such as effects on growth and

development, liver function, and other physiological parameters are equally if not more

important when trying to determine the effects of BKME on fish. In addition, an

understanding of the sublethal effects of BKME on the reproductive physiology of adult

fish is essential for evaluating the impact of these and other environmental contaminants

at a population level.

Organization of Dissertation


This dissertation evaluates the effects of BKME on health and reproduction of

largemouth bass. Specifically, the objectives of this work are as follows:

* To compare reproductive parameters of Florida largemouth bass sampled from

reference and BKME-exposed sites along the St. Johns River (Chapter 2).

* To expose adult largemouth bass to different concentrations of BKME for various

lengths of time and evaluate effects on reproductive physiology (Chapter 3).

* To evaluate the effects of BKME exposure on health parameters of largemouth bass

through the conduction of both field and laboratory studies (Chapter 4)

* To expose adult largemouth bass to different concentrations of BKME for various

lengths of time and evaluate effects on reproductive physiology and success (Chapter

5).

* To examine the effects of BKME and resin acids on the steroidogenic capacity of

isolated ovarian follicles (Chapter 6).






35


* To integrate these findings, evaluate their ecological significance, and propose areas

of future research needs (Chapter 7).














CHAPTER 2
COMPARISON OF REPRODUCTIVE PARAMETERS FROM FLORIDA
LARGEMOUTH BASS (MICROPTERUS SALMOIDES FLORIDANUS) SAMPLED
FROM REFERENCE AND CONTAMINATED SITES IN THE ST. JOHNS RIVER
AND TRIBUTARIES


Introduction


Over the past decade, several Canadian and Scandinavian studies have focused on

the effects of bleached kraft pulp mill effluent (BKME) on multiple biochemical and

physiological parameters of fish. From these studies, some of the most meaningful

responses have been related to altered reproductive function. Specifically, fish exposed

to BKME have lower circulating concentrations of reproductive hormones (testosterone,

11-ketotestosterone, and 17B-estradiol), reduced gonadal growth, increased age to sexual

maturation, smaller eggs, and reduced expression of secondary sex characteristics when

compared to fish from reference sites (Sandstrim et al. 1988, Larsson et al. 1988,

Andersson et al. 1988, Munkittrick et al. 1991, 1992b, 1994, McMaster et al. 1991).

Detailed endocrine laboratory studies have demonstrated that the pituitary-gonadal

axis of fish is affected by exposure to BKME, including decreased circulating

concentrations of gonadotropin (GtH) and sex steroids, depressed responsiveness of

gonadal steroidogenesis to gonadotropin-releasing hormone (GnRH), and altered

peripheral metabolism of sex steroids (Van Der Kraak et al. 1992). In addition,

McMaster et al. (1995) reported a reduced conversion of testosterone to 17B-estradiol,

indicating a reduced level of aromatase in BKME-exposed follicles during early








vitellogenic stages. In another study, steroid synthesis by ovarian follicles from BKME-

exposed and non-exposed female white suckers (Catostomus commersoni) was similar,

suggesting that the origin of different steroid concentrations in wild BKME-exposed fish

is external to the gonad (Gagnon et al. 1994b).

Effects of paper mill effluents on the reproductive physiology of fish have also

been documented in the St. Johns River, Florida. Female mosquitofish, Gambusia

affinis, inhabiting a stream receiving paper mill effluents were strongly masculinized

showing both physical secondary sexual characteristics (fully developed gonopodium)

and reproductive behavior of males (Howell et al. 1980). Masculinization of female

mosquitofish has also been reported from laboratory studies after exposures to B-

sitosterol, a plant sterol byproduct of wood delignification (Denton and Howell 1989).

More recently, masculinization of female fish has been identified from an additional two

species (least killifish, Heterandriaformosa and sailfin molly, Poecilia latipinna)

collected from paper mill effluent-receiving streams (Bortone and Cody 1999).

Since 1947, the St. Johns River has received effluents from a paper mill plant

located in Palatka (Figure 2.1). This mill has two bleaching lines (40% product) and an

unbleached line (60% product), which together release an estimated 36 million gallons of

effluent daily. In this plant, the bleaching sequences for the bleach lines are CEHD and

C9dioEopHDp (where Cd = mixture of chlorine (C) and chlorine dioxide (d) in

proportions designated by subscripts; Eop = extraction with alkali and the addition of

elemental oxygen (o) and hydrogen peroxide (p); H = hypochlorite; and Dp = 100% d

substitution with the addition of p). The bleaching lines manufacture paper towels and

tissue paper, whereas the unbleached line produces mainly kraft bag and linerboard. The








wood furnish of this mill consists typically of 50% softwood species (mainly loblolly,

slash, sand, and pine) and 50% hardwood (mainly tupelo, gums, magnolia, and water

oaks). At the time of this study, effluents received secondary treatment, which consisted

of both anaerobic followed by aerobic biological degradation after a retention period of

40 days.

The objectives of this study were, first, to conduct a preliminary seasonal survey

to examine the reproductive physiology of populations of Florida largemouth bass

(Micropterus salmoidesfloridanus) sampled at increasing distances from a paper mill

discharge area in relation to a reference site. For this part of the study, bass were sampled

prior (September 1996) and during (February 1997) the spawning seasons. Since some

reproductive alterations were observed in this preliminary survey, the second main

objective was to further evaluate the possible impacts) of environmental exposure to

paper mill effluents on the reproductive physiology of this species by increasing the

number of reproductive endpoints measured as well as the number of sites sampled. This

second phase of the study was restricted to sampling during the reproductive season

(March 1998). Parameters measured in these studies included body weights, lengths,

condition factors, hepatic 7-ethoxyresorufin O-deethylase (EROD) activity, liver weights,

gonad weights and histology, concentrations of vitellogenin and sex steroids

(testosterone, 11-ketotestosterone, 17B-estradiol) and number and size of mature eggs in

females.








Materials and Methods


Sampling Sites and Fish Collection

Field sampling was divided in two phases (see Table 2.1 for a summary of the

sampling methodology employed). The first phase included the sampling of 100

largemouth bass (70 females and 30 males) during September 1996 (pre-spawning

season) by electroshocking from four sites within the St. Johns River (Figure 2.1).

Mainstream sites included a reference site located 40km upstream from the effluent

discharge (Welaka), and three exposed sites located at increasing distance from the

discharge (Palatka, Green Cove, and Julington Creek, at 3, 40, and 55km from the

discharge, respectively). An additional 84 bass (36 females and 48 males) were sampled

from the same sites during February 997 (spawning season).

The second phase of the study was conducted during the spawning season (March)

1998. A total of 61 females and 53 males were collected by electroshocking from six

sites (-20 site) within the St. Johns River (mainstream) and its tributaries (small creeks)

(Figure 2.1). Areas sampled included two tributary reference sites: Cedar Creek

located approximately 25km downstream from the mill and Etonia Creek which is the

primary water source for the mill and is located about 100-200 m upstream from the

effluent discharge, and tributary exposed site Rice Creek, a small tributary stream

(about 5 km in length) receiving the direct discharge from the mill. Fish were also

sampled from two mainstream reference sites: Welaka and Dunn's Creek (the latter

located 18km upstream from effluent discharge), and from mainstream exposed site

Palatka, which receives the direct discharge from tributary Rice Creek. The estimated








paper mill effluent concentration in exposed sites Rice Creek and Palatka averages 60%

and less than 10%, respectively (Georgia-Pacific Corporation, personal communication).

However, water flow in Rice Creek is tidally influenced, so that during periods of low

flow mill effluents can account for up to 90% of the total flow (Schell et al. 1993).

Reference sites were matched to exposed sites in most characteristics, except presence of

effluent. In order to minimize the variation in parameters measured in relation to timing

of reproductive season, all fish within each site were collected within an average of four

hours, and all sites were sampled in a 1-week period. Rice Creek was the only exception

to this strict sampling protocol, where it was necessary to collect largemouth bass on

three different occasions over a two-week period to achieve adequate numbers.

Chemical Analysis from Fish Tissues

With the objective of chemically characterizing some of the sites used in this

study, fish tissues were collected and analyzed for up to 113 trace organic and 20 trace

metal contaminants. For this analysis, livers from 5 females were collected from Welaka,

Palatka, Green Cove, and Julington Creek during September 1996 and February 1997.

Livers from the 1997 collection were pooled and run as one sample. Sample collections,

laboratory analysis and quality control procedures were carried out as previously

described (St. Johns Water Management District 1998). In brief, for determination of

organic, samples were serially extracted using dichloromethane and then analyzed

through gas chromatography/mass spectrometry (GC/MS) for determination of polycyclic

aromatic hydrocarbons (PAHs) and phthalates or through GC/electron capture detection

for analysis of polychlorinated biphenyls (PCBs), hexachlorocyclohexanes (BHCs),

pesticides dichlorodiphenyltrichloroethanee and derivatives, DDTs), and other chlorinated









compounds. For metals, samples were digested with a mixture of nitric and hydrofluoric

acids and concentrations were measured either by graphite furnace atomic absorption

spectroscopy or inductively coupled plasma/mass spectrometry. Contamination data is

reported on a dry weight basis, and was not corrected for lipid content, nor percent

recoveries.

Bleeding, Necropsies, and Age Determination

Fish were weighed using a portable digital scale to the nearest 0.1g and body

length measured (total length, from the tip of the mouth to the tip of the tail) to the

nearest millimeter. Condition factor was calculated as K = weight/length3 x 100. Blood

was collected in the field from the caudal vein using 3mL syringes and 1.5 inch, 20G

needles. Blood samples were transferred to 5mL-heparinized vacutainers and kept on

ice until centrifugation for 10min at 1,100 x g. Plasma was pipetted into 2mL cryotubes

and stored at -80C until analyzed. After bleeding, fish were euthanized with a blow to

the head, and a complete necropsy performed. Gonads and livers were excised, weighed

for the determination of organosomatic indices, and a section preserved in Notox for

histological evaluation as explained below. Fish collected during 1998 were decapitated

for the removal of sagittal otoliths, which were used for the determination of age as

described in Crawford et al. (1989).

Reproductive Endpoints

Analysis of sex steroid hormones

Plasma samples from largemouth bass were analyzed for testosterone (only during

1996/97 sampling), 11-ketotestosterone and 17B-estradiol (all fish in the study) using a









radioimmunoassay (RIA) technique. The following is a description of the methodology

for 17B-estradiol and 11-ketotestosterone; the method for testosterone determination is

similar. First, 50uL of plasma were extracted twice with 5mL of diethyl ether before RIA

analysis. Samples were then analyzed in duplicate for both hormones and corrected for

extraction efficiencies of 92 and 86 % for 178-estradiol and 11-ketotestosterone,

respectively. Standard curves were prepared in buffer with known concentrations of

radioinert 17B-estradiol (ICN Biomedicals, Costa Mesa, CA, USA) or 11-ketotestosterone

(Sigma Chemical, St. Louis, MO, USA) (1, 5, 10, 25, 50, 100, 250, 500, and 1,000pg).

The minimum concentration detectable was 6.4 pg/mL for 178-estradiol and 8.1 pg/mL

for 11-ketotestosterone. All plasma samples were assayed in duplicate, and interassay

variability was < 10% for each steroid. Values are reported as pg/mL of plasma.

Cross-reactivities of 17B-estradiol antiserum (produced and characterized by T. S.

Gross, University of Florida) with other steroids were as follows: 11.2% for estrone, 1.7%

for estriol, and < 1% for 17a-estradiol and androstenedione. 11-ketotestosterone

antiserum cross-reacted with: testosterone (9.7 %), a-dihydrotestosterone (3.7 %), and

with androstenedione (< 1 %). A pooled sample (approximately 275pg of 17B-

estradiol/mL and 220pg of 11-ketotestosterone/mL) was assayed serially in 10, 20, 30, 40,

and 50uL volumes (final volume of 50uL with charcoal-stripped plasma). The resulting

inhibition curves were parallel to the respective standard curve.

Analysis of vitellogenin

Vitellogenin concentrations in plasma of largemouth bass were quantified by

Direct Enzyme-Linked Immunosorbent Assay (ELISA). First, vitellogenin from

largemouth bass was purified by anion exchange chromatography (LMB VTG 102396B),








and its protein concentration determined by the Bradford method (Bradford 1976) for use

as a standard. The monoclonal antibody, Mab 3G2 Ascites 109 AB (produced by the

Hybrydoma Core, University of Florida) was used in the ELISA assay. This antibody

reacts with high specificity and sensitivity to largemouth bass vitellogenin, with little or

no cross-reaction with other plasma proteins.

Plasma samples were diluted from 1:200 (male samples) to 1:10,000 (female

samples) in phosphate buffer saline azide (PBSZ 0.15 M NaCi, 10 nM phoshapte, 0.02%

NaN3, pH 7.2) with aprotinin (10 KIU/mL), 50uL was added in triplicate to microtitre

plate wells and incubated overnight at 4C in a humidified chamber. Plates were then

washed with PBSZ plus tween (PBSTZ, 0.05% Tween-20), blocked with 360 ul/well of

blocking buffer (1% bovine serum albumin (BSA) and 10mM Tris BSTZ) for 2 hours at

room temperature, and washed again with PBSTZ. Purified monoclonal antibody was

diluted with blocking buffer with aprotinin to 3 ug/mL for male runs and to 0.1 ug/mL for

female runs, coated onto 96-well microtitre plates (50 uULwell), and stored overnight at

4C in a humified chamber. The next day plates were washed with Tris BSTZ, incubated

with 50 uL/well polyclonal biotinylated goat mouse anti-vitellogenin IgG antibody (H +

L) (Pierce, Rockford, IL, USA), diluted to 1:1,000 with blocking buffer, and incubated

for 1 hour at room temperature. Plates were then washed with Tris BSTZ, and incubated

with 50 uULwell of strep-avidin-alkaline phosphatase, diluted to 1:1,000 with blocking

buffer, for 1 hour at room temperature. After a final wash with Tris BSTZ, 100 uL/well

of p-nitro phenyl phosphate in carbonate buffer (pH 9.6) was added to each well and

incubated at room temperature in the dark for 30 minutes. The intensity of yellow color

that developed was quantified at 405nm with and automated ELISA reader (Spectra Max









250, Molecular Devices, Sunnyvale, CA, USA). Vitellogenin concentrations were

calculated from standard curves after subtracting the small blank value (around 0.2 A405

nm) of a nonspecific color reaction with male control plasma.

Standard curves were constructed by adding serial dilutions of purified

largemouth bass vitellogenin (0 mg/mL to 0.001 mg/mL) to male control plasma and

processed the same way as samples. Male control plasma was made from a pool of

plasma from fish collected at an uncontaminated site, which was shown by Direct ELISA

and Western Blot analysis to have no vitellogenin. Each assay was run with a positive

control with a known vitellogenin concentration, to test for interassay and intra-assay

variation. Samples were rerun if the coefficient of variation between triplicates exceeded

10%. Standard curves fit by quadratic regression were used to calculate vitellogenin

concentration, with R2 values usually between 0.95 and 0.99. The minimum

concentration detectable in this assay is of 0.001 mg/mL. Values are reported as mg/mL

of plasma.

Gonadosomatic (GSIs) and hepatosomatic (HSIs) indices

Gonads and livers (without gall bladder) were excised from each fish and weighed

using a portable scale to the nearest 0.01g. Somatic indices were calculated by dividing

the weight of the organ by the weight of the fish and multiplying the resulting number by

100.

Number and size of mature eggs

The total number of mature eggs was estimated in bass sampled during 1998 by

collecting a subsample of follicles (approximate wet weight of 100mg) along a mid-

section of one of the ovaries. The sample was preserved in Notox @ and the number of









mature eggs (defined as tan- yellow to brown-yellow eggs of over 0.6mm in diameter)

counted under a dissecting scope. To obtain the total number of mature eggs in both

ovaries, the number of mature eggs in the subsample was multiplied by the weight of both

ovaries and the resulting number divided by the weight of the subsample. The mean size

or diameter of mature eggs was calculated by lining 10 eggs on a ruler under a dissecting

microscope and dividing the resulting number by 10. This procedure was done twice, so

that 20 eggs were measured from each sample of follicles.

Histopathology

During the 1998 study, samples of gonads were collected and preserved in

Notox for histological evaluation. Testes were cut longitudinally and ovaries were cut

transversally. Tissue samples were then embedded in paraffin, sectioned at 5um,

mounted on glass slides, air dried and stained with Mayer's hematoxylin and eosin

(H&E). Ovaries were classified into four stages of sexual maturation: undeveloped

(stage 1, mostly primary oocytes at various stages of previtellogenic growth);

previtellogenic (stage 2, primary and secondary oocytes, no vitellogenic oocytes); early

vitellogenic (stage 3, some vitellogenic oocytes of different sizes, with few to moderate

amount of vitelline granules, and few to no fully developed eggs); and late vitellogenic

(stage 4, most of the oocytes contained numerous vitelline granules). In addition, the

number of atretic follicles was counted in each histologic section of ovaries. Testes were

classified into three stages of sexual maturation: low to no spermatogenic activity (stage

1, thin germinal epithelium, scattered spermatogenic activity); moderate spermatogenic

activity (stage 2, thick germinal epithelium, diffuse to moderate proliferation and









maturation of sperm); and high spermatogenic activity (stage 3, thick germinal

epithelium, high proliferation and maturation of sperm).

Liver EROD Activity

For this assay, buffers, substrates, and cofactors were purchased from Sigma

Chemical, St. Louis, MO, USA. Snap frozen livers from fish collected during 1998 were

cut with a hammer and chisel so as to collect a total of approximately 250mg of tissue.

Samples were homogenized with 3 volumes (v/w) of homogenizing buffer consisting of

10mM TRIS (pH 7.4), 250mM sucrose, ImM EDTA, 0.2mM dithiothrietol, and 0.1mM

phenylmethylsulfonyl fluoride. Samples were homogenized for approximately 10sec, and

the homogenates were centrifuged at 8000 x g for 10 min. The 10,000g supernatant (S-9

fractions) containing microsomal enzymes was isolated by centrifugation at 12,000 x g

for 20 min and stored at -80C until analyzed. S-9 proteins were assayed by the BioRad

protein assay kit (Richmond, CA, USA) using bovine serum albumin as a standard. Liver

samples were kept ice-cold (4C) throughout. Hepatic EROD kinetic activity was

measured in triplicate in the S-9 fractions using a Spectromax Fmax 96 fluorescent

microplate reader at an excitation wavelength of 544nm and emission at 590nm. For this

reaction, 5uL of enzyme (S-9) were mixed with 195uL assay buffer (0.1 m NaPO4, pH

7.8) and 5uL substrate (100uM ethoxyresorufin in methanol). The reaction was started by

adding 5uL NADPH, and the fluorescence change recorded for 2 min at 30C. Negative

controls were run in the absence of enzyme and positive controls in the presence of rat

microsomes. Reaction linearity was demonstrated over the course of the reactions. A

resorufin standard curve (0, 1, 2.5, 5, 10, and 20pmol) was run for each set of samples.

EROD reaction rates were determined by dividing the rate of change in fluorescence per









minute by the slope of the resorufin standard curve. Results are expressed as pmol of

resorufin formed/min/mg microsomal protein.

Statistical Analyses

Pairwise comparisons were conducted using a two-way analysis of covariance

(ANCOVA) (PROC GLM, SAS Institute 1988) within sexes and years (1996-97 and

1998) to test for differences in the dependent variables between sites. Data sets that did

not meet the criteria of normality and homogeneity of variance (PROC UNIVARIATE)

were log or arcsin transformed. For the 1996-97 data set, season (spawning or non-

spawning) was used as the second cofactor and body weight was used as the covariate,

whereas for the 1998 fish, type of stream (tributary or mainstream) was used as the

second cofactor and age was used as the covariate. For the 1996-97 data, exposed sites

Palatka, Green Cove, and Julington Creek were compared against the reference Welaka,

while for the 1998 study, exposed sites Palatka and Rice Creek were compared to the

reference sites Welaka and Dunn's Creek and Cedar and Etonia Creeks, respectively. If

the ANCOVA showed a significant site effect, a Dunnett's multiple comparison test was

used to examine which exposed site(s) differed from the reference. Regressions between

gonad and liver weight to body weight were compared among reference and exposed sites

after combining data from spawning seasons 1997 and 1998. In addition, regressions of

liver EROD activity and several reproductive parameters measured in females collected

during 1998 are also presented. The frequency distributions of different gonadal

developmental stages were compared between sites using a X2 Test (PROC FREQ). For

purposes of statistical comparisons, ovaries and testes were classified as either low to









moderate (stages 1 and 2 for both sexes) or high gametogenesis (stages 3 for males, and 3

and 4 for females). Statistical significance was assessed at p < 0.05.

Results


Chemical Analysis from Fish Tissues

A summary of fish chemical data is presented in Table 2.2. With the exception of

BHCs, the sum of organic measured in fish tissues appeared highest in bass collected

from exposed site Palatka when compared to fish from reference site Welaka (increases

ranged from 1.5 to 7-fold). There was also an overall trend for a decline in organic

chemicals in fish from exposed sites Green Cove and Julington Creek in relation to fish

from Palatka, with several groups being lower (low molecular PAHs and BHCs) or

comparable (chlorinated benzenes and other chlorinated pesticides) to values found in

reference fish (Table 2.2). Metals were more variable across sites, with highest mean

concentrations mainly found in bass from either Julington Creek (Ag, As, Cr, Cu, Zn) or

Welaka (Cd, Hg, Pb, Se, Tn).

Physiological and Reproductive Endpoints

1996-97 field study

A summary of several physiological and reproductive parameters measured from

largemouth bass sampled along the St. Johns River during September 1996 (pre-

spawning) and February 1997 (spawning) is presented in Table 2.3. Females from Green

Cove and Julington Creek were smaller and lighter when compared to females from the

reference site (Welaka), whereas males did not differ between sites. For both sexes, body

weights, lengths and condition factor did not differ across seasons. There were









differences in gonad weights and GSIs across sites. Palatka and Julington Creek females

had gonad weights and GSIs that were approximately half of those reported from Welaka.

Males from Palatka and Julington Creek had lower and higher gonad weights and GSIs,

respectively when compared to males from the reference site.

For both sexes, plasma concentrations of testosterone were not affected by site,

but decreased from September to February (from a mean of 397 to 338 pg/mL in females,

and from 404 to 207 pg/mL in males) (Figure 2.2). Females from Green Cove had almost

twice the concentration of 11-ketotestosterone when compared to the reference (for both

sampling periods), and this hormone increased from an average of 267 pg/mL in pre-

spawning bass to 448 pg/mL in fish sampled during the spawning season (Figure 2.3).

Pre-spawning males from Palatka and Julington Creek had slightly lower concentrations

of 11-ketotestosterone when compared to the reference stream, but when sampled during

the spawning season Welaka males had concentrations of 11-ketotestosterone that were

over twice of that found in males from exposed sites (Figure 2.3). There were seasonal

changes in the concentration of 11-ketotestosterone in males from all sites except Green

Cove (increased from a mean of 319 pg/mL in September to 628 pg/mL in February). For

both sampling periods, 178-estradiol was lower in Palatka females, and the concentration

of this hormone increased from September to February (mean = 325 to 927 pg/mL)

(Figure 2.4). In contrast, plasma concentrations of 178-estradiol increased in pre-

spawned males from all sites and in spawned males from exposed Palatka and Green

Cove sites in relation to the reference. Seasonal effects were observed only in males

collected from Palatka and Julington Creek (Figure 2.4). During the spawning season,

the ratio of 17B-estradiol to 11-ketotestosterone (E/ 11-KT) decreased from a mean of 4.2









in females from the reference site to 1.8 in bass from exposed sites (Figure 2.5). In

contrast, E/1 1-KT ratios increased in males from all sites (except for Julington Creek

males sampled during February) from 0.36 to 1.22 due to a decline in 1 1-ketotestosterone

and an increase in 178-estradiol. Vitellogenin concentrations were approximately 17

times lower in spawning females from Palatka and Green Cove in relation to the

reference (mean = 0.42 and 7.0 mg/mL, respectively) (Figure 2.6). Although Julington

Creek females had about half the concentration of this protein when compared to bass

from Welaka, this difference was not significant. Seasonal changes in vitellogenin

concentrations were observed only in females sampled from Welaka and Julington Creek

(from September to February it increased over 1000-fold from 0.005 to 5.42 mg/mL).

Although male bass had comparable concentrations of vitellogenin across sites, as with

females, plasma concentrations of this protein increased from a mean of 0.003 mg/mL in

males sampled during September, to a mean of 0.006 mg/mL in males collected during

February (Figure 2.6).

1998 field study

A summary of several physiological and reproductive parameters measured from

largemouth bass sampled along the St. Johns River during March 1998 is presented in

Table 2.4. Age was affected only at tributary Rice Creek, where females were

significantly younger when compared to the reference tributary sites. There were no other

differences in the remaining parameters among reference and exposed sites. Females

sampled from tributaries, however, were older, heavier and longer than females collected

from mainstream sites (4.1 vs. 3.4 years; 1041 vs. 738g; and 41 vs. 37cm, respectively).

Although liver weights were higher in females collected from tributary streams when









compared to mainstream values (12.1 vs. 8.3g), GSIs were lower in these females (3.4 vs.

2.4%). Males from tributaries, on the other hand, had higher condition factors, and higher

HSIs and liver weights in relation to mainstream stations (1.6 vs. 1.4; 1.4 vs. 0.91%, and

8.4 vs. 5.6g, respectively), but lower GSIs (0.35 vs. 0.47%).

Female largemouth bass sampled from a small creek receiving the direct discharge

from the mill (Rice Creek) had a 5-fold increase in EROD activity compared to bass

sampled from reference streams (Figure 2.7). Males had EROD activities that were about

twice as high as females (mean = 6.7 vs. 3.0 pmol resorufin/mg/min for males and

females, respectively), but that did not differ across sites. Vitellogenin concentrations

were not affected by site of collection, and averaged 0.66 mg/mL in females and 0.08

mg/mL in males (Figure 2.7). Concentrations of 11-ketotestosterone and 178-estradiol in

females from exposed sites were about half and 1/3, respectively of those from the

reference station (Figure 2.8). In males, 178-estradiol did not change between exposed

and reference sites, but 11-ketotestosterone decreased in tributaries and mainstream

exposed sites in relation to controls. E/ 11-KT ratios were decreased and increased in

females and males, respectively from exposed stations (Figure 2.9). Males from exposed

tributaries also had higher E/ 11-KT ratios when compared to males sampled from

exposed mainstream sites (2.2 vs. 0.65). Although females from Rice Creek tended to

produce fewer and smaller eggs when compared to females from reference Cedar and

Etonia Creeks, this difference was not significant (Figure 2.10). In contrast, and despite

the decline in sex steroid concentrations observed in females from Palatka, this group

produced almost as twice as many eggs in relation to bass from reference Welaka and

Dunn's Creek. The size of these eggs, however, was similar between the two types of









mainstream sites (Figure 2.10). Overall, females collected from tributary sites had lower

fecundities and egg sizes than females from mainstream sites (mean = 14,856 vs. 17,017

eggs, and mean = 0.75 vs. 0.99mm). Differences in reproductive parameters between

reference and exposed sites occurred despite the fact that there were no differences in

reproductive development as measured in histological sections of ovaries and testes

(Figure 2.11). There was, however, an effect of type of stream on ovarian development,

with a higher proportion of females from mainstream sites having ovaries with a high

degree of oogenesis when compared to tributary sites (62 vs. 38%). The number of

atretic follicles/histological section did not differ among reference and exposed sites

(Figure 2.12), but was significantly increased in females from tributary sites when

compared to bass from mainstream areas (mean = 9.2 vs. 3.5). In addition, the level of

ovarian development was negatively correlated to the number of atretic follicles (R = -

0.55, p = 0.0001). Regression analyses between reproductive parameters and liver EROD

activity in female largemouth bass from exposed tributary and mainstream sites is

presented in Figure 2.13. Vitellogenin, logarithm of gonad weight, GSI, fecundity, and

egg size were inversely related to liver EROD activity.

All years of field studies

Regression lines between body lengths and organ weights versus body weight for

spawning bass from reference and exposed sites collected during 1997 and 1998 are

presented in Figure 2.14. When all spawning fish were combined, there were no

differences among sites in the linear regressions relating body weight to body length and

gonad and liver weight (i.e. all slopes were parallel). The size distribution of females and

males during the spawning sample (1997 and 1998 combined) was shifted towards fish









that ranged in length from 31 to 49cm (91% females and 93% males fell in this size

range), and was similar among reference and exposed sites (Figure 2.15).

Discussion


Although one of the main objectives of this study was to evaluate potential

impacts of BKME on populations of largemouth bass, during the first year of study fish

were also sampled at considerable distances from the mill discharge (Green Cove and

Julington Creek located at 40 and 55km from the mill, respectively). Female bass

sampled from Palatka (closest to the mill) and Green Cove had lower concentrations of

17B-estradiol, vitellogenin, lower E/l 1-KT ratios, and lower GSIs in relation to the

reference site. Females from Julington Creek, however, only showed lower E/ 11-KT

ratios. Males from Palatka and Green Cove showed higher increases in 178-estradiol and

in E/11-KT ratios, but comparable declines in 11-ketotestosterone in relation to males

from Julington Creek, and GSIs were decreased only in Palatka males. These results

indicate a geographical trend in reproductive effects, with decreasing changes as the St.

Johns River flows north. In addition, the presence of reproductive alterations in fish

sampled at a considerable distance from the mill discharge would suggest exposure to

chemicals other than BKME. In this respect, there is considerable evidence showing

endocrine alterations in fish due to exposure to different groups of chemicals not

necessarily related to BKME (such as metals, halogenated aromatic hydrocarbons, and

chlorinated hydrocarbons) (Kime 1995, Heath 1995b, Giesy and Snyder 1998).

Except for higher values of some organic and metals in fish tissues from Palatka

and Julington Creek, the chemical data available was not powerful enough to detect clear









trends across sites. This was probably related to the small number of samples analyzed,

which resulted in a high degree of variation. Although no fish tissues from the site

closest to the discharge (Rice Creek) were available for chemical analysis in this study,

there is indication that this stream is being impacted as a result of effluent discharge.

Schell et al. (1993) reported up to 52.8 ppt of 2,3,7,8-tetrachlorodibenzo-p-dioxin

(TCDD) in sediments collected from Rice Creek, with significant declines in the

concentration of this chemical at the confluence with the St. Johns River (6.8 ppt at Site

3). These authors also reported concentrations of TCDD in liver and gonads of

largemouth bass (range 1.8 8.8), bowfin, Amia calva (11.2 46.1), and brown bullhead

catfish, Ictalurus nebulosus (1.8 2.8) collected from Rice Creek. Another indication of

the presence of effluent chemicals in Rice Creek comes from the work by Quinn (2000)

who found a distinct gradient in the waterborne concentrations of three resin acids

(dehydroabietic, abietic, and isopimaric acids) with highest concentrations at the site of

effluent discharge and non-detectable levels at the confluence of Rice Creek with the St.

Johns River (Palatka site).

Measurements of EROD activity have been widely used as a biomarker for

exposure of fish to several groups of chemicals, including polychlorinated dibenzo-p-

dioxins (PCDDs) and dibenzofurans (PCDFs), PCBs, PAHs, pesticides, metals, and

natural biogenic substances. Because BKME are known to contain EROD-inducing

compounds, this biomarker has played a major role in the study of fate and biological

effects of paper mill effluent discharges. In general, researchers have reported low EROD

activities in fish from reference sites, with significant increases in areas close to pulp mill

outfalls (Forlin et al. 1985, Lindstrom-Seppai and Oikari 1989, Courtenay et al. 1993,









Bankey et al. 1994, Soimasuo et al. 1995b). Until recently, it was believed that the main

inducers in mill effluents were chlorinated persistent compounds (such as PCDDs and

PCDFs) (Hodson 1996). However, new evidence suggests that enzymatic EROD

induction also occurs in fish exposed to unbleached effluents, and that the compounds)

responsible for such induction are not of the highly hydrophobic chlorinated type, but

rather of the moderately hydrophobic planar PAH-type form present as natural

components of wood, and readily metabolized by fish (Hodson 1996).

In the present study, EROD activity was induced in females from the site closest

to the mill outfall (Rice Creek) when compared to the reference (8.4 vs. 1.7

pmol/min/mg, about a 5-fold induction). There was also a gradient of induction from

Rice Creek to its confluence with the St. Johns River (a distance of only 3km), with

females from this latter site having EROD activities that were comparable to values from

reference areas (1.2 pmol/min/mg). This rapid fall in mixed-function oxygenase (MFO)

activity would suggest that compounds capable of causing enzyme induction are present

in high enough concentrations only in water and/or sediments from Rice Creek, and that

by the time they reach the St. Johns River they are diluted enough for EROD activities to

fall to background levels. There is also evidence for a decline in total organic content

(TOC) in sediments from mainstream Palatka in relation to Rice Creek (from 36 to 25

mg/g) (Schell et al. 1993). This decline in TOC could imply that, upon entering this

creek, lipophilic organic contaminants would associate at higher levels with organic-rich

sediments present in Rice Creek with lower associations in sediments from the

mainstream.









EROD activities in males were more variable across sites, and although they

appeared higher in bass sampled from Rice Creek and Palatka (average of 9.1

pmol/min/mg) when compared to males from reference sites (average of 5.4

pmol/min/mg), these differences were not statistically significant. The uniformity of

hepatic MFO induction in male largemouth bass from reference and exposed sites would

suggest a higher degree of mobility compared to females. Information on the range of

movement of largemouth bass is restricted to the work by Snyder et al. (1986) who

reported that 38% of the bass marked and released in the lower St. Johns River were

recaptured in the same are as tagged, and of the remaining 62%, 44% had moved a

distance of less than 2km. Unfortunately, the study did not evaluate differences in range

between sexes.

The degree of EROD activity was different between female and male largemouth

bass, with males having approximately twice the activities seen in females. This

differential activity level could be reflecting differential body burdens of organic

contaminants due to sex differences in lipid concentrations and/or food habits. A more

plausible explanation for sex differences in enzyme activity, however, is that EROD

levels are known to be reduced or even eliminated in female fish during reproduction

(Elskus et al. 1989, Larsen et al. 1992, Stegeman and Hahn 1994). Although the exact

mechanism of this regulation remains unclear, the suppression is thought to be related to

higher increases in 178-estradiol concentrations in spawning females in relation to males.

If a similar pattern occurs in spawning largemouth bass, it is likely that we

underestimated EROD basal activities in this study. In this respect, EROD basal

activities of adult bass in this study are much lower than the values reported in immature









largemouth bass exposed to up to 8% BKME for 263 days (which resulted in up to 800

pmol/min/mg of activity) (Bankey et al. 1994), although the induction levels were similar

(up to six-fold). This large difference in absolute EROD values is probably related to

having worked with adult spawning bass as opposed to juvenile fish. On the other hand,

induction levels in females from Rice Creek are comparable to values reported by Haasch

et al. (1993) in bass (age not reported) exposed in the laboratory to B-naphthoflavone for

up to 4 days, and to caged bass exposed to waters from a PAH and PCB-contaminated

river for up to 7 days (overall inductions of 4.4 and 6, respectively).

An area of intense research has been to try to relate increases in EROD activity

with changes in physiological and reproductive endpoints. Studies on the effects of paper

mill effluents on fish have provided most of the information available on this subject.

The overall conclusion from these studies is that there appears to be no clear relationship

between decreased titers of steroid hormones and other reproductive alterations and

increased hepatic EROD induction (Munkittrick et al. 1992b, 1994, Gagnon et al. 1994a,

1994b, Soimasuo et al. 1998). Preliminary results from the present study, however,

indicate a significant inverse relationship between many reproductive endpoints measured

in female bass (vitellogenin, GSI, log gonad weight, fecundity and egg size) and EROD

activity (see Figure 2.13). The consistency of this negative relationship is suggestive of

an association between antiestrogenic effects and EROD induction in largemouth bass.

English sole (Parophrys vetulus) exposed naturally to Puget Sound sediments

contaminated with PCBs and PAHs, also showed significant correlations between

chemical exposure, MFO induction, and reduced concentrations of plasma 17B-estradiol

(Johnson et al. 1988). In this respect, mechanistic in vitro studies have shown a negative









relationship between a compound's antiestrogenicity and it's ability to induce cytochrome

P450-dependant monooxygenase (CYP1A) proteins (Anderson et al. 1996). These

authors also reported depression of estrogen receptor binding capacity in relation to

increased EROD activity in juvenile rainbow trout (Oncorhynchus mykiss) injected with

50 mg/kg of p-naphthoflavone. These findings indicate alterations in the affinity of 178-

estradiol for the estrogen receptor, probably due to Ah-receptor mediated changes in the

phosphorylation state of the estrogen receptor (Anderson et al. 1996). Since CYP1A in

fish does not participate in the catabolism of 178-estradiol, it is unlikely that increases in

EROD activities are related to increases in the oxidative metabolism of this hormone

(Snowberger and Stegeman 1987). It is clear that additional studies are needed for a

better understanding on the involvement of EROD enzymatic activity on endocrine

modulation and its potential deleterious effects on fish reproduction.

Many studies on the effects of paper mill effluents, report a concomitant increase

in HSIs in fish with high EROD activities (Larsson et al. 1988, Munkittrick et al. 1992a,

1994, Bankey et al. 1994, Huuskonen and Lindstr6m-Seppli 1995). Although we did

observe an increase in EROD activity in at least some of the fish analyzed, this induction

was not associated with increases in liver size. The absence of an increase in HSIs in the

present study could have been related to the timing of fish sampling. Bass were collected

during the reproductive season (March), and it is well known that the weight of the liver

in this species undergoes seasonal variations, with highest values in winter and spring

(December-April) and lowest values in the summer months (Adams and McLean 1985).

Thus, any increases in liver weight due to enhanced activity of xenobiotic

biotransformation enzymes would have been masked by the physiological increases that









are associated with reproductive status (this is particularly true in the case of females due

to increases in the synthesis of vitellogenin). Lack of increases in liver weight after

exposures to pulp mill effluents have also been reported in longnose sucker (Catostomus

catostomus) (Munkittrick et al. 1992), trout-perch (Percopsis omiscomaycus) (Gibbons et

al. 1998), and whitefish (Coregonus muksun) (Lindstrom-Seppdi and Oikari 1989).

There were similarities, but also differences in the reproductive responses of bass

from the 1996/97 and 1998 studies (Table 2.5). For both years of study, female bass

collected from reference sites had higher concentrations of 17B-estradiol (between 42 and

64% higher) whereas males from these sites had higher concentrations of 11-

ketotestosterone in relation to fish from exposed sites (between 46 and 67% higher).

Concentrations of plasma testosterone appeared less sensitive than those of 178-estradiol

and 11-ketotestosterone, with changes only across seasons but not sites. Year differences

included declines in vitellogenin in females and in GSIs (both sexes), and increases in

178-estradiol in males from exposed sites only during the February 1997 sampling. In

addition, concentrations of 11-ketotestosterone in spawning females from exposed sites

behaved oppositely across years, with increases during 1997 and declines in 1998 (see

Table 2.5).

Because bass were collected from both mainstream and tributary streams during

1998 but only from the former in 1997, comparisons among years are most appropriate if

restricted to responses of mainstream spawning fish. An explanation for the differences

in reproductive responses between years could be related to the timing of sampling.

During 1998 mainstream bass were sampled a month later in the reproductive season

(March as opposed to February in 1997), which resulted in higher GSIs and plasma








concentration of sex steroids in fish from both sexes in relation to 1997. The only

reproductive parameter that did not follow this same trend was vitellogenin in females,

which was lower in 1998. This is not surprising since plasma concentrations of this

protein are known to decline by over 50% from February to March (from about 4 to 2

mg/mL) in female bass (Timothy Gross and Nacy Denslow, unpublished data). It is

possible then, that in order to detect small differences in GSIs and vitellogenin between

clean and contaminated streams bass need to be sampled earlier during the reproductive

season. In addition, because of the high dynamism of the hydric system under study, it is

expected that year differences in chemical composition may occur through time (Durell et

al. 1998). This coupled with the fact that largemouth bass are likely to move some

distance from year to year (Snyder et al. 1986) could result in a differential degree of

exposure, and thus on differences on the physiological responses being measured.

Despite the differential response in sex steroids among years, there was an overall

similar decline (31 58%) and increase (61 66%) in E/ 11-KT ratios in females and

males, respectively for both years of study. Throughout this study, E/11-KT ratios were

above 3 (4.2 in 1997 and 3.4 in 1998) in plasma of spawning female bass from reference

sites and below 0.5 in males (0.3 in 1997 and 0.4 in 1998). E/11-KT ratios above 3 are

expected in healthy reproductively active female bass because of a predominance in 178-

estradiol over 11-ketotestosterone. Conversely, low E/ 11-KT ratios in males indicate a

higher proportion of 11-ketotestosterone over 17B-estradiol. An imbalance between the

concentration of these two sex steroids, as that observed in fish collected from exposed

streams (E/l 1-KT ratios of 1.8 and 2.4 in females from 1997 and 1998, and of 0.9 and 1.3

in males from 1997 and 1998) may be indicative of endocrine disruption.








In the present study, season and type of stream affected some of the reproductive

parameters measured. From the 1996/97 seasonal study it was apparent that although

changes in sex steroids between sites were observed as early as 5 months prior to

spawning (September), these were most evident in spawning fish (February). On the

other hand, results from the 1998 study would suggest that fish collected from tributary

streams were at a slighter earlier reproductive stage than animals from the mainstream.

For example, even though females from tributary streams tended to be older and heavier,

they also had lower GSIs, fecundities, and egg sizes in relation to females from

mainstream sites. Similarly, males from tributary streams had higher condition factors,

but lower GSIs. In both sexes, liver weights were also increased in bass collected from

tributary stations, which would suggest that at the time of sampling these fish had

allocated less body reserves into reproduction in comparison to mainstream bass. These

reproductive differences were corroborated histologically in females, with tributary fish

having a lower frequency of ovaries with high degree of oogenesis and an increase in the

number of atretic follicles. It is worth noting, however, that despite the reproductive

changes just mentioned concentrations of sex steroids and vitellogenin did not differ

among bass from mainstream and tributary streams in either sex. For future studies then,

it would be of importance to take into consideration the seasonal fluctuations in

reproductive parameters, as well as the effect of type of stream on the reproductive

physiology of this species of fish.

There was a lack of consistency on the effects of environmental contaminants on

GSIs. During 1997, GSIs were decreased in females from exposed sites Palatka and

Green Cove, but not in females from exposed Julington Creek. Although the decline in









178-estradiol was comparable across females from exposed sites (42% decline overall),

females from Julington Creek (site furthest away from the mill discharge) had

vitellogenin concentrations that were not significantly different in relation to females

from the reference site. A higher concentration of vitellogenin in females from this site

could explain a lack of a decline in GSIs. During this same year, GSIs in males were

lowered only in fish collected from the site closest to the mill discharge (Palatka), despite

a similar decline in 11-ketotestosterone across exposed sites (46% decline overall).

Similarly to what was observed with females, however, males from Julington Creek had

increased GSIs, which could be related to a lack of an increase in 17B-estradiol in these

males in relation to males from Palatka and Green Cove. In the 1998 field study, a

significant decrease (42 to 67%) in reproductive hormones was observed in females and

males from exposed sites. Gonadosomatic indices, however, were either slightly

decreased in the case of males (19%, although not statistically significant) or not affected

at all in the case of females. Despite some declines in GSIs in fish from exposed sites

during the spawning season of 1997, when data from both years was combined, there was

a clear lack of a relationship between GSIs and site of collection (Figure 2.14). These

results suggest that variations in gonad weight are probably more related to variations in

local environmental conditions rather than to contaminant exposure, and/or that higher

declines in sex steroids and vitellogenin are necessary before declines in GSIs are

observed. Several studies have reported declines in GSIs in fish exposed to BKME

(Larsson et al. 1988, Munkittrick et al. 1991, 1992a, 1994, Gagnon et al. 1994b, Gibbons

et al. 1998). However, there is also evidence to suggest that decreases in gonadal size in

response to declines in sex steroids may not always occur after exposures of fish to









BKME (McMaster et al. 1996b), which would indicate differences in reproductive

responsiveness to contaminant exposure across species.

The present study is one of a few to evaluate changes in vitellogenin

concentrations in relation to exposure to paper mill effluents. Although there was a

decline in the plasma concentration of this protein in females from exposed sites when

compared to reference streams during both years of study (94 and 56% declines for 1997

and 1998, respectively), this decline was significant only in 1997. Higher declines in

vitellogenin in females sampled during February 1997 could explain the concomitant

decrease in GSIs and the lack of such a decline in March 1998. Declines in vitellogenin

and 178-estradiol in females from contaminated streams would suggest exposure to

antiestrogenic compounds.

The presence of vitellogenin in plasma of male largemouth was a consistent

finding in this study. The concentrations of vitellogenin found in males, however, were

an average of about 1/12 of those found in females, and although increased from

September to February, its high variability precluded the detection of any differences

between exposed and reference sites. Finding detectable concentrations of vitellogenin in

plasma of male fish has generally been considered as a sign of endocrine disruption

associated with exposure to estrogenic compounds (Sumpter and Jobling 1995). Plant

sterols, such as 3-sitosterol, which are commonly found in pulp mill effluents, are

estrogenic compounds known to bind in vitro to rainbow trout hepatic estrogen receptors

(Tremblay and Van Der Kraak 1998) and induce vitellogenin synthesis in male goldfish

(Carassius carassius) (MacLatchy and Van der Kraak 1995). Although we can not rule

out the possibility that male bass sampled closest to the mill discharge could have been









exposed to the estrogenic effects of plant sterols, several reports have documented low

background concentrations of vitellogenin in different species of male fish, and have

regarded such concentrations as physiologically normal (Copeland and Thomas 1988,

Ding et al. 1989). The results from the present study, consistently showing low

concentrations of vitellogenin in male largemouth bass sampled from both clean and

contaminated streams, further supports this hypothesis, and cautions the use of just the

presence of this protein in plasma of males as a definitive sign of endocrine disruption.

During the 1998 study, gonads were evaluated histologically to ensure that

observed differences in concentrations of sex steroids and vitellogenin between fish from

exposed and reference sites were not caused by different stages of sexual maturity.

Within each type of stream, the results showed similar stages of ovarian and testicular

development among bass collected from all three sites. In addition, the similarity in the

number of atretic follicles in ovaries from females from clean and contaminated streams

and the absence of any noticeable lesions in the testes examined would suggest that

alterations in sex steroid concentrations in bass from exposed sites (declines in 17B-

estradiol and 11-ketotestosterone in females and males, respectively) were probably not

enough to cause damage to gonadal tissue.

A rather unexpected finding in this study was that females from Palatka had

ovaries that contained more eggs in comparison to females from the reference sites and

females from Rice Creek. These eggs were also larger, but only when compared to fish

from Rice Creek. Females sampled from the Palatka site also had larger gonads (4%

body weight vs. 3% in females from reference sites), although these differences were not

statistically significant. These results occurred despite the fact that Palatka females had









less than half the concentrations of 176-estradiol in relation to females from the reference

sites (611 vs. 1513 pg/mL). Increases in fecundity that were coupled with declines in

plasma sex steroids have also been reported in lake whitefish (Coregonus clupeaformis)

females exposed to BKME (Munkittrick et al. 1992a). In contrast to what was observed

in the present study, however, lake whitefish females also had lower gonad weights,

which resulted in smaller egg sizes in comparison to reference females. White suckers

from metal-contaminated sites have also shown increased fecundities (Munkittrick and

Dixon 1989), but these females also had reduced spawning and the hatching rate was

decreased because of egg shell-thinning problems. Increases in fecundity and fertility

have also been observed in certain fish populations, and have been interpreted as an

adaptation to high levels of pollution (by securing the production of more offspring these

populations will eventually lead to the formation of a more pollutant resistant strain of the

species) (Kime 1995). At this point, the origin of the increased fecundities in females

from Palatka, as well as the population-level effects that could be associated with such a

change remain unknown.

One of the most consistent findings in studies that have focused on the effects of

BKME on reproductive parameters of fish is a decline in the concentration of sex steroids

in plasma of exposed animals. BKME-exposed white suckers from Jackfish Bay, Lake

Superior show decreased concentrations of several sex steroid hormones (testosterone,

11-ketotestosterone 178-estradiol, and 17, 2013-dihydroxy-4-pregnen-3-one) (Portt et al.

1991, McMaster et al. 1995, 1996b). Declines in steroid concentrations have also been

documented in longnose sucker and lake whitefish from Jackfish Bay (Munkittrick et al.

1992a, McMaster et al. 1996b), in white sucker at other mills (Hodson et al. 1992,









Munkittrick et al. 1994, Gagnon et al. 1994a), and in other fish species sampled

elsewhere (Adams et al. 1992, McMaster et al. 1996b). The consequences of these

similar endocrine alterations to whole animal reproductive fitness and population

dynamics, however, have varied greatly between species. For example, longnose sucker

exposed to BKME show no organism responses other than an altered age distribution,

whereas white sucker and lake whitefish show decreased gonadal sizes, secondary sexual

characteristics, and egg sizes, and increased age to maturity (McMaster et al. 1996b). In a

review of whole organism responses of fish exposed to different kinds of mill effluents

(including unbleached pulps), 48% of the populations studied had increased condition

factors, 80% showed increased age to sexual maturation, and reduced gonadal size was

reported in 58% of the studies (Sandstrom 1996). These observations provide evidence

for species differences in susceptibility to BKME, but also show the inherent difficulty

when trying to compare biological responses in fish populations inhabiting highly

different environments and exposed to complex mixtures likely to vary in chemical

composition. In this respect, except for a decline in GSIs during the first year of study,

the results from our field study suggest that decreased hormone concentrations in

response to paper mill effluent exposure may not always be associated with obvious

reproductive impairment, such as reduction in gonad weight and fecundity. Although this

study was not designed to evaluate potential population-level effects, preliminary analysis

would indicate no effects on age distributions and growth between exposed and reference

populations of largemouth bass. The absence of organism-level responses is probably not

related to a lack of sensitivity, since laboratory in vivo experiments on the impacts of

BKME on the reproductive performance of largemouth have shown effects on gonad









weights and other measures of reproductive success (see Chapters 3 and 5). It seems

more likely to suspect an insufficient exposure to BKME in the populations of

largemouth bass sampled closest to the mill outfall (Rice Creek and Palatka). Although

effluent concentrations are high in Rice Creek, the scarcity of bass in this stream would

indicate absence of adequate prey and/or nesting substrate, thus making this area

unsuitable for long-term residency. Fish from mainstream Palatka, on the other hand, are

being exposed to a highly diluted effluent (less than 10% v/v) because of the high water

flow present in the St. Johns River.

In summary, although lower and potentially sensitive levels of biological

organization (biochemical and physiological) were altered in largemouth bass from

contaminated streams, these changes were not necessarily related to impacts at higher and

less sensitive levels of organization (organ, organism, and population). It is clear then

that additional studies are needed to further evaluate the possible impact of such

endocrine changes in populations of Florida largemouth bass. Finally, future field study

designs should incorporate the capability for testing relationships between chemical

exposure and biological responses and should be accompanied by controlled laboratory

studies that explore dose-response relationships to better interpret the data generated.










68

0 0



oS o o8 0 0 0 8
M PoQ UCQo w




cn 0 c -
0 c 0 r.
PC ;0- w-
h-sg a .a = c a ^ O
I ~ U~ mS0^i m^



0 S ,.2 r"A5 '
0 0

> o
C C
m .S h g 'Sca









0.0
^~~~ B 5rU C B o o 0 .a


Co












a f ; *6 *
a) a 4) o 1- ; c .4



















z 0C
U co a oo c
CCo
4) o g g














a !? &p
a -5 -






CoCa
C Co


44)
0'
C.2o -; 1i S
Co B>" c<** -





0 40.



ON allu/ '' ^ ^l | |t
ON~b
4)- -9I I~




o s. 1- I- U-^ 1 ^ s

c4i 4) rUn n t O K)^'' o


i. O- C IC Co SS o'












0.o
-Q
L
O
















to
~i|


















4.c
ro ca


go

r .







o QL|

60 01
4) II
e6



.c 3
S|
4-.4)Q
C x1 go














C a



0 M
0 -a

li2 >
gt *
11-1
No '-

Ug .O


00

1-4


'-e 00
r-i
^-4


O r- oo r--oo
n o r-


w

a
F4 -4Q()




asper
cu
: )



0 0

2 u, S 0 -0
0 C 0 4
CQ'seQrQoo'


tn
^ i r a C ri 'R oo
- 01 C) 0
(-- -^ od r-S








00
O0 0o 0 0oqC1 O
0 M N Co 00+ C 00


o
0

v
I
0
Ii8



-
fkon


-'" s



Si0
0 a C) Cu -


c0 C14 (- -0 0 00 00
ooocncooo




















0000'.00<
0 00 0 8- 0





CoIoto4 or-
(6 -5 06 i vi 6~ 0 6-














W/) tn IC 0~ s


uo
13
Sa a

ei-



O -









,70
,6 N

6 0
p,= -S4
6 0 =









ut o "a.




CS '1 0 -, .0 .


Q 0 o Ni ,'








Soz0" 4-)
w M0 r- .
o so i

4.) ..















Sr-g












a
x o-


0 4) --4.)
'i P oOl U












O
4)

so *
0.4
- ^ ^
-H -H+ +1 +1




e4 -4 -

0
o












Noo c4 0 0 0
vi -00 r' 0








0 +1 +1+1 +1 +1 +1 +1 +1 H+I + +l +
4 '-4








(U
0 c) Nr *)
os"Q/ss o r '


0 +1 -H +1 +1 +1 +1 +1 -H -H 44 +1 +1



no enVC- (- Wen



NON 'NO





S++1+1+1 +1+1+1+1 +1+1+1+1
-4 -. 14 4 -l






-4)
0

m 0 It 0 t

41 +1+1+-H +1 +1 +1 +1 +1 +1+1 +1
'-, U ,







Co W -0 efq 00 r- W)c -' tol0 -
0 0 ) 't OO o
m COi 2OMo
*+1 + +1 +1 +1 +1 +1 +1 -H +1 -H -H






W1e W) 0 m) *0)
O hn1tf.( o in ti r~C t- C r o t-: c i k; 06 _
m- *5 0cns cn cn < m M I















ot C> m4 00 (ON 4 o 0 o
.;i "0*'^^ ^ ^ -










rA fu N c



co~4 0 MJ 4) -,
I- a 1.4
.mi 42 0'a "-s o ~ '- 4- r4 s 't d cn u
6 pS >>3 n 2- >o c 0 q o o o
Uc 0 0 U^ 0S q m Uso ^: c od
=- Vo W -0w
'^^ cco














o en
0 0
6 d
+1 +1








0 c
-H +1

cn





-

+1 +1
-o







i-4
+1 +1
00o 0


W- 0-41-4





oo 0

-s0




0a-'
(3 U o 4)
= w 0
C'e s 4) CL


.-

4-1



00
*S3









co4
C

I







*. I D4

00 S
&0 "C




*a2







cd
o





U -
'0 *So a
,11 -











u h o73

* ogrlg d o o d o
'-U+l +l +1
+1 +1 +1 +1



o100 Q -s

a6
-a ff +1 o+ ++1


W- +1 +1 +1 +1


i_ cg an MO c
o^ 'O ^ 0o o o an c
0 ed +1 +1
S 6 +1 +1 + o
0 40 N C
(3 < 0 0 e

C)I 'O '
+1 +1 + + + +1
0 en +1 +1 a '
o C CA m











~s 03 -a- 1o
0J- 0 0 0 r 0 00
H + +1 +1 +1 +1 +1



o .t o oo o o oI




0ge W




> 2 +i ai
c5 l > gU ,a oo so I -
S8+1 +1 +1 +1 -+1 +1
Nl Cl 00 Cl 00 0\
40. 0 4 Cl ON a 0
0 C "0 \10 '0 00
U"4'
~ 1U


a +1 +1 +1 +1 +1 +1

I _'t I 0 C



Cl 'n M* Cl r n o oo -
o IM cd X
4 6O 0 6 t N 6
0 00+1 + +1 +1 +1 1



'.0 ar feg 3 -
4 ~ ~ ~ ~ V c:W 0 t ) I






-a 4|a m
(U 'U


GO~



^-Hl- |i |^ ll
,v U,
75j) l 4) 1.. W1~ j |is


rAg~ C, (U 4m. & u&S us

Ps 'i~ Q4.ps~u ^SS U










U





~
.~ .~ '.~5


& -p


- u


S3 -
as a
S Q S


-*-
d e .
o -C-
0





l- .
*^1






so il



o o >
*-o -

*a *s 's

o3 o -
e3m o I
CU O '.

















0 0 0 0 0 0 0 0 P0o 0
1 +


40

0

U





0









4-
-o


















0









0
&=
0

















o




0
- -


i a 0


s












1-
WEx


a g g o ,o g E ^ ^ Et
t* 1 + 1




















oo
0
o-0 a


r- A
lm m u >s
iU, iUIill .


*8 ~ ~ ~ ~ 4 4) g 00' UI^S^^Ig
m S^S'^5 iMOE'


+











I I


I + +





















a)0

*w

la1



0 *










Ca
^a)
3



























bo
.CM



'3 (j:











t: a
o S





42
a)
Bi a)






.-

>e





4-


.C 2
&o


0 *



EE
*M









700
600


W
0


500
400
300 -
200
100

600

500-

400 -

300

200 -

100

0


I Females


Males

T


Figure 2.2. Mean SEM testosterone concentrations in largemouth bass sampled during
pre-spawning (September 1996) and spawning (February 1997) seasons. Numbers inside
histograms indicate sample sizes (n). There were no differences in relation to reference
site (Welaka) (ANCOVA).


I I Spawning


-8


I


-1
151


v


O-
ENO cdim
C '0.4


Csee"


I IPre-Spawning









I I Spawning


800


0u
a,
0





0
<-
U,
0


700
600
500
400
300
200
100

1200

1000

800

600

400

200

0


8


Females








1M
12 *





Males


nT
8



CO


Figure 2.3. Mean SEM 11-ketotestosterone concentrations in largemouth bass sampled
during pre-spawning (September 1996) and spawning (February 1997) seasons. Numbers
inside histograms indicate sample sizes (n). Asterisks indicate differences in relation to
reference site (Welaka) (ANCOVA, Dunnett's multiple comparison test; a = 0.05).


Pre-Spawning


12


I?


A^s











1750

1500
1250
1000
750
500
250
0
600

500

400

300

200


100


I I Spawning


Figure 2.4. Mean SEM 178-estradiol concentrations in largemouth bass sampled
during pre-spawning (September 1996) and spawning (February 1997) seasons. Numbers
inside histograms indicate sample sizes (n). Asterisks indicate differences in relation to
reference site (Welaka) (ANCOVA, Dunnett's multiple comparison test; a = 0.05).


Pre-Spawning









I I Spawning


5

4

3

2

1

0

2.5

2.0

1.5

1.0

0.5

0.0


*
a


Females








S l 9



Males





W. i


v^


Figure 2.5. Mean SEM ratio of 178-estradiol to 11-ketotestosterone (E/11-KT) in
largemouth bass sampled during pre-spawning (September 1996) and spawning (February
1997) seasons. Numbers inside histograms indicate sample sizes (n). Asterisks indicate
differences in relation to reference site (Welaka) (ANCOVA, Dunnett's multiple
comparison test; a = 0.05).


mmd


l, -l


4\ev


Pre-Spawning








I I Spawning


1

0.020

0.016

0.012

0.008

0.004

0.000


Figure 2.6. Mean SEM vitellogenin concentrations in largemouth bass sampled during
pre-spawning (September 1996) and spawning (February 1997) seasons. Numbers inside
histograms indicate sample sizes (n). Asterisks indicate differences in relation to
reference site (Welaka) (ANCOVA, Dunnett's multiple comparison test; a = 0.05).


Pre-Spawning









16
14
12
10
8
6
4
2
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0


I I Mainstream


Figure 2.7. Mean SEM EROD and vitellogenin concentrations in largemouth bass
sampled along the St. Johns River during the spawning season (March) of 1998. Fish
were collected from tributaries or mainstream sites. Numbers inside histograms indicate
sample sizes (n). Asterisks indicate differences of exposed tributary (Rice Creek) and
mainstream (Palatka) sites in relation to reference streams (Cedar and Etonia Creeks for
tributaries or Welaka and Dunn's Creek for mainstream sites) (ANCOVA, Dunnett's
multiple comparison test; a = 0.05).


S Tributaries








I I Mainstream


0


1800

1500

1200

900

600

300

2100

1800

1500

1200

900

600

300


Figure 2.8. Mean SEM 11-ketotestosterone and 178-estradiol concentrations in
largemouth bass sampled along the St. Johns River during the spawning season (March)
of 1998. Fish were collected from tributaries or mainstream sites. Numbers inside
histograms indicate sample sizes (n). Asterisks indicate differences of exposed tributary
(Rice Creek) and mainstream (Palatka) sites in relation to reference streams (Cedar and
Etonia Creeks for tributaries or Welaka and Dunn's Creek for mainstream sites)
(ANCOVA, Dunnett's multiple comparison test; a = 0.05).


Tributaries










I I Mainstream


Figure 2.9. Mean SEM ratio of 17B-estradiol to 11-ketotestosterone (E/1 1-KT) in
largemouth bass sampled along the St. Johns River during the spawning season (March)
of 1998. Fish were collected from tributaries or mainstream sites. Numbers inside
histograms indicate sample sizes (n). Asterisks indicate differences of exposed tributary
(Rice Creek) and mainstream (Palatka) sites in relation to reference streams (Cedar and
Etonia Creeks for tributaries or Welaka and Dunn's Creek for mainstream sites)
(ANCOVA, Dunnett's multiple comparison test; a = 0.05).


Tributaries









I 1 Mainstream


35000
30000
25000
20000
15000
10000
5000
0


1.5

1.2

0.9

0.6

0.3

0.0


Figure 2.10. Mean SEM fecundity and egg size in female largemouth bass sampled
along the St. Johns River during the spawning season (March) of 1998. Fish were
collected from tributaries or mainstream sites. Numbers inside histograms indicate
sample sizes (n). Asterisks indicate differences of exposed tributary (Rice Creek) and
mainstream (Palatka) sites in relation to reference streams (Cedar and Etonia Creeks for
tributaries or Welaka and Dunn's Creek for mainstream sites) (ANCOVA, Dunnett's
multiple comparison test; a = 0.05).


= Tributaries










100 -

S80 -

60
34
o 40
0


20

0

100

S80-

60-
(-U

40-

20 -

0


aI


Females




16


Males
18

I,


11




_


20






i


13







U


6







rl-'


X2 = 0.39, p = N.S.

W Tributary Low/Mod
Oogenesis

Tributary High
Oogenesis

Mainstream Low/Mod


Oogenesis


X2 = 0.04,p = N.S.

|L Tributary Low/Mod
Spermatogenesis
| Tributary High
Spermatogenesis

Mainstream Low/Mod
Spermatogenesis
Mainstream High
Spermatogenesis


7 hi
CFO


Figure 2.11. Differences on the frequency of ovarian and testes development (Chi-square
Test) in exposed largemouth bass in relation to reference. Fish were sampled along the
St. Johns River (tributaries and mainstream sites) during the spawning season (March) of
1998. Numbers on top of histograms indicate sample sizes (n).


I


I












20


16-


12-


8-
8


4


Tributaries Mainstream
-U


Figure 2.12. Number of atretic follicles in hitological sections of ovaries from exposed
(tributary Rice Creek and mainstream Palatka) largemouth bass in relation to reference
(tributaries Cedar and Etonia Creeks, and mainstream Dunn's Creek and Welaka).
Numbers inside of histograms indicates sample sizes (n). There were no differences in
this parameter between exposed and reference sites (ANCOVA, p > 0.05).


I I Mainstream


-11 0


I I


F ]Tributaries










.0

-4 ?
o&


3
S 2
O 1
0
0 0
o -1
10
8
6
4
2
0
-2

36000
S24000
12000
0


0.12
0.08
a 0.04
0.00
-0.04


Y = 10.24 9.95(X), r2 = 0.40, n = 13, p = 0.02






Y = 10.49 5.23(X), r = 0.35, n = 13,p = 0.03

-S




Y = 10.50 1.71(X), r = 0.52, n = 13,p = 0.005
-






Y = 9.88 0.0003(X), r2 = 0.50, n = 13,p = 0.006



.53, n 13, 0.005


Y = 11.44 77(X), r = 0.53, n = 13,p = 0.005


- I


EROD (pmol resorufin/mg/min)



Figure 2.13. Regression analyses between several reproductive parameters and EROD
activity in female largemouth bass from exposed Rice Creek and Palatka sites sampled
during the spawning season 1998.


Q












0 Reference Sites


Females


0 Exposed Sites


55

50

C45

40


60
0






CO


35

30
8
7
6
5
4
3
2
1
0
05
20


10


0 1000 2000 3000


Males


0I
0 1000 2000


Body Weight (g)


Body Weight (g)


Figure 2.14. Regression lines between body length, and gonad and liver weights versus
body weight, by sex. For gonad weight, data from spawning season 1997 was combined
with that from spawning season 1998. For 1998, mainstream and tributary sites were
combined. Liver weights were measured only during 1998. = reference sites (Cedar
and Etonia Creeks, Welaka, and Dunn's Creek); o = exposed sites (Rice Creek, Palatka,
Green Cove, and Julington Creek).


Co




0
0



2-.
1

1
Co

Io


Y = 27 + 0.01(X), r2 = 0.77, = 99,p = 0.0001


0


oo-

0 0


Y = -0.001 + 0.04(X), r2 0.59, a = 99, p = .M01


o








S*
*
0
0


0 0 0


3000


I-I










Spawning Females


24 -

20 -

16-
-

12 -

Z 8-

4-


24-

20 -

16-

S12-

S8-

4


I I Exposed
SReference


9 44 49 54 59


Total Length (cm)

Figure 2.15. Length-frequency distributions of largemouth bass, by sex. Data from
spawning season 1997 was combined with that from spawning season 1998. For 1998,
mainstream and tributary sites were combined. Reference sites (Cedar and Etonia Creeks,
Welaka, and Dunn's Creek); exposed sites (Rice Creek, Palatka, Green Cove, and
Julington Creek).


Spawning Males
















29 34 ;


_ ____ ____ ____ ____ ____ _____














CHAPTER 3
IN VIVO ASSESSMENT ON THE REPRODUCTIVE EFFECTS OF PAPER MILL
EFFLUENTS ON LARGEMOUTH BASS


Introduction


Results from field studies have indicated altered reproductive biomarkers for

largemouth bass (Micropterus salmoides) sampled downstream from a paper mill in

Florida. Fish inhabiting streams with effluent discharges had lower circulating levels of

sex steroids (11-ketotestosterone and 173-estradiol) and showed increased mixed-

function oxygenase (MFO) activity. These biochemical changes, however, were not

necessarily related to impacts at higher and less sensitive levels of organization (organ,

organism, and population). Additional evidence showing possible endocrine alterations

as a result of exposure to this effluent comes from work by Bortone and Cody (1999).

These authors reported masculinization (evidenced by gonopodial development) of

females from three poeciliid species inhabiting Rice Creek, the stream receiving the direct

discharge from the mill.

It has become apparent from several studies that both field and laboratory

approaches are essential for data interpretation. Although field studies are important and

necessary because they provide ecological relevance, they are subject to many limitations.

For example, there is great uncertainty about exposures (doses, lengths, and routes) of

free-ranging fish, which can seriously hinder data interpretation. In addition, the effects

of environmental factors other than the degree of contamination on the parameters being




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EDT4YPF28_JRZEMN INGEST_TIME 2012-08-30T13:53:45Z PACKAGE AA00011789_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES