-‘! .3 13 a.‘ Fifi-i?" . "VU'V‘frV'. I w~V—'——T‘——r'— 2 “‘Iuh" _ fill? W ..I -v . l‘: .1 c Eh.” ” ’ ". ‘ ( My, . . _ .W:"':2;- ' .l I :5 ' - .rW .1 I .; 4: I ..,H lett} .J' I' -54.”; ' I m .; , 4 >1, l ' "5‘81"" ’0‘ ‘ ‘ _ |"| ‘- - ’6‘“. I ‘ I .: I .._. 5‘ 7 Tan» _ <" ‘fi ”“3; . on“, ., II ‘. > ‘r' " .r.'.$_... WI: ‘ I ”-413: ~I-n . . ~= —-~ A LI .xL-F E 4)“ ~>Iv )3 2" W541" .‘. I ."ll .I.“ Ac 335475 ““ gn'az‘. .I . l I? ‘ I. Z . . . '. VI, :2: ‘1 ‘ ‘l 3—1;- er; ‘2 : .vpgf'dj'} ' '. “lag-:1; t .1", nun. V .u. If ,f _ ‘1‘ ’. :.a3 4' . ‘l Q '.~.0 . k . q r :5 II V. 4: ' :éfw.‘ . -h .113." “:3.” 3-? . 1' :u‘ 1 L ,- 93“. 2 o; 3 «12‘ .. v. -' "- fr; I574:— .- .3 -‘ E? 1" - -- ._. If." .2,- - ““3“ c 4"} or...“ .—. I - I.v 1,. '44 I I . :1 ' ix 3r “ :rfi,.?.’.f.:,." a‘:;".fl:f. W! 1 . I -1 ' ‘I I 1,2: ‘v;; ‘ 'w‘“: .‘r— ‘ J‘- ' J. , , " ,2. v ‘Tfiul’ 'JY}C'-.JL "v’ ' «JJ 41%“ r'.’,“:‘-;a" ' "r (‘ . 1‘ I ."I ' r3? "1‘ ”Pi-ab", . . ".¥:j " _ “3f” ‘.‘ I” "‘ :' - t3 afifiiflr Eff... '1'. 4 u , F31"? "tab” .0 I . 'c 1; “,7; r fmq ”.1 ' I. LIT mgmzm ‘ M‘ I . ‘ . 'J r: . I II ‘1 I ll .4’ w'w‘lu ' jg m Ii}. .' "I' P 3? 1%: ' fiIimu..f'i.‘ "£qu filly ,1' .,'II ulllLJ'JI Ff. n “1%,; ; . ‘ '4‘!" “A? "37")9 1‘ “it... 7" II J‘I’p‘ 7T". $3 '6'; I'L. I" ” v2,“ “J THESIS This is to certify that the thesis entitled AMMONIA DETOXIFICATION AND HISTOPATHOLOGY IN CHANNEL CATFISH FROM AN IN ggzy SEWAGE TREATMENT EFFLUENT BIOASSAY presented by Stephen Van Mitz has been accepted towards fulfillment of the requirements for MASTER OLSQENCE—degree in Eisheries_& Wildlife Major professox DateaZ/_Z__I ~( /7J:3 0-7639 MS U i: an Affirmative Action/Equal Opportunity Institution MSU RETURNING MATERIALS: Place in book drop to LJBRARJES remove this checkout from “ your record. FINES will r be charged if book is returned after the date stamped below. 133?":334 1‘71~ (f \=-/ 'r‘*“w 300 I‘&Q.=l‘, ”5° 1311111 AMMONIA DETOXIFICATION AND HISTOPATHOLOGY IN CHANNEL CATFISH FROM AN IN_SITU SEWAGE TREATMENT EFFLUENT BIOASSAY By Stephen Van Mitz A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1984 Copyright by STEPHEN VAN MITZ T984 ABSTRACT AMMONIA DETOXIFICATION AND HISTOPATHOLOGY IN CHANNEL CATFISH FROM AN IN_SITU SEWAGE TREATMENT EFFLUENT BIOASSAY by Stephen Van Mitz A l7-day jn_situ bioassay was conducted to determine the effects of the Anthony Ragnone Nastewater Treatment Plant effluent on caged, juvenile, channel catfish. Acute, 100% mortality occurred at study sites 300 and 500 m downstream from the outfall, probably from the mean total residual chlorine concentrations of 0.24 and 0.30 mg/l, respectively. Growth rate, measured as wet weight gain, was not significantly different among fish from locations 600 m upstream (FRl) or 9.00 km downstream (FRS) from the outfall. Histopathological effects were observed in the gill tissues from the FR5 fish. Neither the glutamate dehydrogenase specific activity in liver tissue, nor the brain free concentrations of glutamate, glutamine, or a-ketoglutarate were significantly different among the FRl, FR5 and laboratory refer- ence fish. Thus, these biochemical analyses were not useful indica— tors of ig_situ ammonia exposure in catfish from a sewage treatment effluent biomonitoring study in which mean unionized ammonia concen- trations were 0.032 mg/l or less. To my wife, Lisa, and to my Parents ACKNOWLEDGEMENTS I would first like to thank Mr. Jack Wuycheck of the Michigan Department of Natural Resources (DNR) Environmental Services Division for granting me the opportunity to participate in the DNR's Anthony Ragnone Wastewater Treatment Plant Bioassay Project of July Zl- August 6, l982, thereby making this research effort possible. I also wish to thank DNR aquatic biologist, Bonnie White, for her suggestions during the planning phase of the project. My special thanks go to DNR water quality technician, Bill Erickson, for his assistance in: l) tagging and routine checking of the fish, 2) providing transportation to and from the study reach, 3) collecting water quality data and samples, and 4) providing DNR literature (Ni previous Flint River studies. I wish to express my gratitude to my major professor, Dr. John Giesy, for his technical assistance and moral support throughout the course of this study. I would also like to thank the other members of my committee for their individual contributions. Specifically, my appreciation goes to Dr. D. Garling for suggesting pertinent litera- ture, and to Dr. P. 0. Fromm for his general advice and comments re- garding the histology of the test fish gill and liver tissues. Dr. K. Jamison instructed me in the operation of a Zeiss Photo- microscope. I thank Dr. A. Trapp for his assessment of histopatholo- gical damage in gill and liver tissues from representative samples of iv my test fish. I am grateful to the faculty and residents of the Michigan State University Veterinary Diagnostic Laboratory for per- mitting me to use their Department's light microscopes and Zeiss Photomicroscope. Dr. T. Isleib and Dr. C. E. Cress provided valuable assistance in the statistical analysis of my data. I thank Dr. P. Muzzall for his identification of parasites in the test fish gill tissues. I wish to express my appreciation to Jay Gooch for providing access to and instruction in the use of his laboratory's spectrophoto- meter, ultracentrifuge, motorized tissue grinder, and sonicator. I am grateful to Nelson Herron for tutoring me in the use of his spectro- fluorometer. I wish to acknowledge collectively the numerous other graduate students of the Pesticide Research Center whose suggestions and advice provided me with insights and encouragement during difficult times. I would like to thank Diane Hummel for typing the final two drafts of this thesis. Finally, I want to thank my wife, Lisa, for her continuous support, patience and faith in what I was trying to accom- plish. TABLE OF CONTENTS Page LIST OF TABLES --------------------------------------------------- viii LIST OF FIGURES -------------------------------------------------- x INTRODUCTION ----------------------------------------------------- 1 MATERIALS AND METHODS -------------------------------------------- 6 Fish -------------------------------------------------------- 6 Cages ------------------------------------------------------- 7 Study Sites ------------------------------------------------- 7 Anthony Ragnone WWTP ---------------------------------------- 7 Field Exposure ---------------------------------------------- 10 Water Quality Analyses -------------------------------------- ll Tissue Sample Preservation ---------------------------------- ll Histological Procedures ------------------------------------- 13 Glutamate Dehydrogenase Assay ------------------------------- 15 Protein ----------------------------------------------------- 16 Brain Tissue Assays ----------------------------------------- l6 Deproteinization --------------------------------------- l6 Glutamate ---------------------------------------------- l7 Glutamine ---------------------------------------------- l7 a-Ketoglutarate Assay ---------------------------------- 18 Experimental Design and Statistical Analyses ---------------- 18 RESULTS AND DISCUSSION ------------------------------------------- 20 Water Quality ----------------------------------------------- 20 Vandalism --------------------------------------------------- 20 Columnaris Outbreak ----------------------------------------- 24 Mortality --------------------------------------------------- 24 Growth ------------------------------------------------------ 28 Histopathology ---------------------------------------------- 28 Gill --------------------------------------------------- 28 Liver -------------------------------------------------- 37 Liver Glutamate Dehydrogenase ------------------------------- 45 vi TABLE OF CONTENTS (continued) Page RESULTS AND DISCUSSION (continued) Brain Metabolites ------------------------------------------- 47 Glutamate and Glutamine Concentration ------------------ 47 a-Ketoglutarate ---------------------------------------- 57 SUMMARY AND CONCLUSIONS ------------------------------------------ 60 RECOMMENDATIONS -------------------------------------------------- 64 APPENDIX A; Water Quality Laboratory Analyses -------------------- 67 APPENDIX B; Statistical Results ---------------------------------- 70 LIST OF REFERENCES ----------------------------------------------- 77 vii Table 10 LIST OF TABLES Measurement methods for on-site water quality parame- ters performed during the Flint River Bioassay study-~- Dehydrating and embedding schedule for fish gill and liver tissues. Process performed on an autotechnicon model 2A tissue processor ------------------------------ Water quality characteristics at the in-stream loca- tions and Ragnone NWTP effluent during the Flint River bioassay study ----------------------------------------- Heavy metals and organic compounds analysis results from the Ragnone NWTP effluent and each in-stream study site --------------------------------------------------- Initial and final total group weights and mean weight per fish, as well as mean growth rate for the FRI and FRS location fish -------------------------------------- Mean liver tissue GDH specific activities in the FRl, FRS, and LR fish samples ------------------------------- Mean concentrations of glutatmate and g1utamine, as well as ratio of glutamine/glutamate from brain tissue of FR], FR5, and LR sample fish ------------------------ Literature values for brain tissue concentrations of glutamate and glutamine, and ratio, in fish and the rat Mean brain a-KGA concentration in fish from the FRl, FRS, and LR locations .................................. Literature values of brain a-KGA concentrations in rats given I-P. injected NH4-Ac solutions ................... viii Page 12 14 21 23 29 46 48 50 58 59 LIST OF TABLES (continued) Table A1 A2 Bl 82 B3 B4 B5 B6 B7 Page List of the water quality analyses performed in the laboratory by the Michigan Environmental Laboratory---- 65 Schedule for obtaining samples for laboratory analysis of water quality of the Ragnone effluent and at each in-stream study site during the Flint River bioassay study-----------------+ -------------------------------- 66 Coefficients of expected mean squares (EMS) for the 3-way nested ANOVA of liver GDH activity in channel catfish from the Flint River bioassay study ------------ 70 Three-way nested ANOVA of liver GDH specific activity in channel catfish from the Flint River bioassay study- 71 Coefficients of EMS for the 3-way nested ANOVA for brain glutamate, glutamine, a-KGA concentrations, as well as glutamine/glutamate ratio in channel catfish from the Flint River bioassay study -------------------- 72 Three-way nested ANOVA of brain glutamate concentration in channel catfish from the Flint River bioassay study- 73 Three-way nested ANOVA of brain glutamine concentration in channel catfish from the Flint River bioassay study- 74 Three-way nested ANOVA for brain ratio of glutamine/ glutamate in channel catfish from the Flint River bio- assay study -------------------------------------------- 75 Three-way nested ANOVA of brain a-KGA concentration in channel catfish from the Flint River bioassay study---- 76 ix Figure 10 11 LIST OF FIGURES Page Two biochemical reactions responsible for ammonia de- toxification in animals -------------------------------- 4 Map of the Flint River bioassay study reach ------------ 9 An FRl location fish with severe Columnaris infection, resulting in skin necrosis and exposure of underlying muscle tissue ------------------------------------------ 25 Photomicrograph of a skin scrape from the lesion on the FRl fish pictured above -------------------------------- 25 Photomicrograph of a laboratory reference (LR) fish gill section ------------------------------------------- 31 Photomicrographic close-up of an LR fish gill section, showing the absence of histopathological damage -------- 31 Photomicrographic close-up of a gill section from a FRl upstream control fish, showing the hyperplasia and hy- pertrophy in the respiratory epithelium ---------------- 32 Photomicrograph of a gill section from a FR5 location fish --------------------------------------------------- 33 Photomicrographic close-up of a gill section from a FR5 location fish, showing extensive hyperplasia of the respiratory epithelium --------------------------------- 33 Photomicrograph of a gill section from a FR5 location fish, showing severe edema in the secondary lamellae, resulting in separation of the respiratory epithelium from the pillar cells ---------------------------------- 35 Photomicrographic close-up of a gill section from a FR5 location fish, showing two blood-filled aneurysms in the secondary lamellae --------------------------------- 35 LIST OF FIGURES (continued) Figure 12 13 14 15 16 17 18 19 20 21 22 Page Photomicrograph of a liver section from a FRl location fish, showing the severe vacuolation of the hepatocytes 38 Photomicrograph of a liver section from a FR5 location fish, showing severe vacuolation ----------------------- 38 Photomicrograph of a liver section from a FRI location fish, showing less vacuolated and a greater number of nucleated hepatocytes ---------------------------------- 39 Photomicrograph of a liver section from a FR5 location fish, showing the reduced vacuolation and greater num- ber of nucleated hepatocytes --------------------------- 39 Photomicrograph of a severely vacuolated liver section from a FRI location fish, showing the negative lipid staining results after staining with Oil Red 0 and Harris hematoxylin ------------------------------------- 4O Photomicrograph of a dense liver section from a FRl lo- cation fish, showing the negative lipid staining re- sults after staining with Oil Red 0 and Harris hema- toxylin ------------------------------------------------ 4o Photomicrograph of a severely vacuolated liver section from a FRl location fish, showing the intense, positive glycogen staining results after staining with Best's Carmine and Harris hematoxylin ------------------------- 41 Photomicrograph of a dense liver section from a FRI lo- cation fish, showing the positive, but less intense glycogen staining results after staining with Best's Carmine and Harris hematoxylin ------------------------- 41 Photomicrograph of a FRl liver section pretreated with human saliva (amylase) for l min before fixation and staining with Best's Carmine and Harris hematoxylin---- 42 Photomicrograph of a FRI liver section from the same fish as in Figure 20, only untreated with human saliva- 42 Proposed mechanism and set of biochemical reactions for the small compartment glutamate synthesis during hyper- ammonemia in channel catfish --------------------------- 54 xi INTRODUCTION Chlorinated sewage treatment effluents can affect stream fish in a variety of ways, including species diversity reduction (Tsai, I973), behavioral reactions (Fava and Tsai, 1976), and toxic effects (Bausch gt_gl,, 1971; EIFAC, 1973; Saalfeld, 1975; DeKraker, 1978; Lubinski and Sparks, 1981). However, it is difficult to predict the in situ .— effects of such effluents upon stream fish for two reasons. First, chlorinated sewage effluents are typically comprised of a complex mix- ture of toxicants, thus allowing synergistic or antagonistic inter- actions among the toxicants. Most toxicity data, however, are ob- tained from tests using pure compounds (Weber, 1981). Secondly, the dynamic nature of the biotic and abiotic environment in a stream can affect the toxicant concentrations, as well as the effects they elicit. For example, changes in the water level affect toxicant con- centrations, whereas changes in temperature, pH, dissolved oxygen (00), and/or other factors may directly or indirectly influence the effect a toxicant elicits in an organism. Thus, there is a need for site-specific in situ field experiments designed to evaluate the toxicological effects of wastewater effluents on aquatic organisms under prevailing environmental conditions. Field studies are espe- cially needed to determine the effects of pollution on organisms at metabolic and cellular levels, including histOpathological effects (Weber, 1981). 2 Although the toxicity of a mixed effluent to aquatic organisms can be evaluated by laboratory or ig_situ bioassays, it is often difficult to determine the relative toxicity of specific effluent constituents. This is critical, however, when adjusting treatment processes to correct effluent toxicity. For example, a reduction in unionized ammonia (UIA) concentrations would be ineffective if the observed toxicity were due to excessive chlorine. There is a need to be able to determine which of the multiplicity of possible toxicants is most responsible for toxicity, so that the most effective and cost efficient abatement plans can be instituted. In the case of the Anthony Ragnone Wastewater Treatment Plant (WWTP), UIA concentrations in the effluent and river were great enough that changes in the treatment process were being considered to reduce the UIA concentra- tions in the effluent. Thus, there was a need for an ig_§jtu_eff1u- ent biomonitoring test to evaluate ammonia-specific toxicological responses in fish below the Ragnone outfall, Ammonia is a major toxic component of sewage effluents (EIFAC, 1973; Tsai, 1975; Thurston gt_al,, 1981). It is also the primary form of nitrogenous waste resulting from nitrogen metabolism in ammonotelic freshwater teleosts (Hillaby and Randall, 1979). UIA is very toxic to fish, although its mechanism of toxicity is not yet fully understood. Fish must possess mechanisms to quickly eliminate or detoxify internal UIA. Under most conditions, endogenous UIA in ammonotelic teleosts is eliminated through the gills via passive diffusion. This process is possible because of the small molecular size and lipid solubility of UIA. However, laboratory studies have demonstrated two ammonia 3 detoxification reactions which fix ammonia into less toxic organic compounds (Berl gt_al,, 1962a; Cooper _t,_l,, 1979; Subcommittee on Ammonia, 1979) (Figure 1, Equations 1 and 2). The linking of ammonium ion to a-ketoglutarate (a-KGA), catalyzed by glutamate dehydrogenase (GDH) E.C.l.4.l.3, to form glutamate (Figure 1, Equation 1) occurs predominately in the liver, but also takes place in the brain (Wu, 1963). At physiological pH, the equilibrium constant (Keq) for this reaction is approximately 6x10M , thus strongly favoring the reductive amination of a-KGA to glutamate. The amination of glutamate to gluta- mine (Figure 1, Equation 2) takes place only in the brain (Wu, 1963), and the Ke for this reaction favors the formation of glutamine (Sub- q committee on Ammonia, 1979). A question of interest to aquatic toxicologists is whether sub- strate or product concentrations, or enzyme activity changes can be monitored as indicators of ammonia exposure or toxicity. Laboratory studies have indicated that an increase in brain glutamine concentra- tion is a useful indicator of ammonia exposure (Berl gt_gl,, 1962a; Levi gt_al,, 1974; Arillo gt_al,, 1981). However, I am unaware of any literature on the evaluation of either of the above mentioned reactions in fish from in situ sewage effluent studies. Aquatic toxicologists would possess a powerful diagnostic tool if measurable changes in fishes' 1) concentrations of substrates or products, or 2) enzyme acti- vities associated with either of these reactions were demonstrated to occur as a result of jn_§jtu_exposure to sewage treatment effluent containing toxic concentrations of ammonia. In order to address these needs, I conducted an 1n_situ, caged- fish bioassay near the Anthony Ragnone WWTP on the Flint River at Equation 1 C00' C00- C-H +H N-C-H (:412 + Nicotinamide + NH I ‘1" 3 C—H + NAD+ H 0 _ 2 . 4 2 2 C-0 Adenine GDH C-H2 COO__ Dinucleotide C00_ a—Ketoglutarate NADH glutamate Equation 2 C00' C00- +H NC-H Adenosine +H NC-H Adenosine 3 C-H2 + Triphosphate + NH + “"“;‘ 3 C-H2 + Diphosphate + Pi C-H2 glutamine C-H COO__ synthetase O=C-NR2 glutamate ATP glutamine ADP Figure 1. Two biochemical reactions responsible for ammonia detoxi- fication in animals. Micl vei' of: 5 Michigan Department of Natural Resources (DNR) Water Quality Sur- veillance personnel. My specific goals were to determine the effects of the Ragnone WWTP effluent on the test fishes': 1) growth and survival; 2) gill and liver histopathology; 3) liver glutamate dehy- drogenase (GDH, E.C. 1.4.1.3) enzyme activity; 4) free glutamate and glutamine concentrations and the glutamine/glutamate ratio in brain tissue; and 5) brain concentrations of a-KGA. Le PL me Ill 1119 MATERIALS AND METHODS Fish The channel catfish (Ictalurus punctatus) was chosen as the test organism because of the literature available on the physiology of ammonia metabolism and toxicity of ammonia for this species. Addi- tionally, channel catfish are a typical warmwater stream fish occur- ring in the Flint River (Odin, 1981). Juvenile channel catfish were obtained from Aquatic Control Inc., in Seymore, Indiana, 12 days before commencement of the field exposure. The fish were kept in a 300 l fiberglass flow-through tank at the DNR's Water Quality Surveillance Bioassay Unit Headquarters facility in Lansing, Michigan. Fish in the laboratory were fed a daily ration of Purina Catfish ChowR at the rate of 3% body weight per day. The manufacturer's guaranteed analysis stated that the protein content of the Catfish Chow was at least 36%. One week prior to commencement of the field exposure, tissue samples were obtained from 10 fish to serve as laboratory references, and will henceforth be referred to as the LR sample. Five days before the field exposure began, the remaining fish were weighed (mean wt 23.1 i_0.6 gm S.E., n=112) and tagged. Numbered tags consisted of 5 mm diameter, colored-plastic discs which were attached to the base of the dorsal spine with heavy-duty nylon thread. The fishes' gain in t1 dl‘ 7 wet weight at the completion of the exposure was used as an index of the sublethal effects on fish growth. Cages The cages were constructed of 19 mm-thick plywood, approximately 0.76 m x 0.76 m x 0.61 m. Each cage had a hinged, latchable top and 6 mm mesh, galvanized hardware cloth on two adjacent sides. Cages were secured in position by wiring them between two metal fence posts. Each cage was positioned such that one of the hardware cloth sides faced into the current, thus allowing constant flow through the cage. An additional metal fence post was placed approximately 5 m directly upstream from each cage to act as a weed catcher. The cages were placed into the Flint River on July 19, 1982, two days before the introduction of the fish. StudygSites Five study sites, hereafter designated FRl-FRS, were located along an approximately 9 km stretch of the Flint River near the Ragnone WWTP at Montrose, Michigan (Figure 2). Site selections were based upon accessability and the on-site total ammonia concentrations such that the chosen sites would provide a gradient of UIA concentration expo- sures. A control site (FRl) was located upstream from the Ragnone WWTP outfall, and four sites (FR2-FR5) were located downstream from the Ragnone outfall (Figure 2). Anthony Ragnone WWTP The Anthony Ragnone WWTP serves a major portion of the suburban area within Genesee County outside the City of Flint. The Ragnone Figure 2. Map of the Flint River Bioassay study reach. The direc- tions and approximate distances of the study sites from the Ragnone WWTP outfall were: FRl, 600 m upstream; FR2, 300 m downstream; FR3, 500 m downstream; FR4, 3.75 km downstream; FR5, 9.00 km downstream. ——>z SAGINAW CO. aznzs‘tt‘fiéi ”””””” PEG 1km MONTROSE .// McKlNLEY ROA D -1---——---~-—~ 1 ( \ Mantras; / Figure 2 ANTHONY RAGNONE WWTP lO facility is a conventional, activated sludge WWTP which uses ferric chloride for phosphorus removal, and has a design flow of 76,000 m3/ day (Roycraft and Buda, 1979). Wastewater is treated with ferric» chloride as it enters the main pump station. After primary settling, activated sludge aeration, secondary clarification and chlorine dis- infection, the effluent is discharged to the Flint River through out- fall 250035 (001) (Rock and Erickson, 1980). The most significant industrial pollutants in the effluent come from chrome plating opera- tions (Rock and Gauthier, 1978). Field Exposure The l7-day field exposure was from July 21-August 6, 1982. On day 1, approximately 100 test fish were transported to the study reach R aerated tanks. At each site, 15 fish were by truck in mini-cool selected and their tag numbers recorded. The fish were then carried to the cage in a 18.9 1 plastic bucket. The study sites were visited on days 3, 6, 8, 10, 13 and 15 to: 1) check on the condition of the cages and remove accumulated debris; 2) check on the condition of the fish and record any mortality; 3) feed the fish; and 4) perform on-site water quality measurements and collect water samples for laboratory analysis. The fish were fed Purina Catfish ChowR. On days 1, 6, 8, 13 and 15, the fish were given a ration equivalent to 6% of their body weight. On days 3 and 10, the ration was increased to 9% of the body weight. On day 17, the test fish from the FRl and FR5 sites were sacrificed, and appropriate tissue samples were obtained and preserved in the field. On-site water quality measurements and water samples for laboratory analysis were also obtained. 11 Water Quality Analyses Temperature, pH, 00, and total residual chlorine (TRC) were measured on-site (Table l). A YSI Model 33$-C-T Meter was used to measure conductivity in water samples collected that same day. All water quality analyses performed in the laboratory were conducted by the DNR's Environmental Services Division Environmental Laboratory (Appendix A, Table A1) in Lansing, Michigan, according to their methods (Appendix A). Water samples for analysis by the DNR's Environ- mental Laboratory were collected (Appendix A, Table A2) in polyethy- lene bottles, fixed and stored on ice according to the methods of Rock and Erickson (1980), and delivered to the Environmental Laboratory within several hours after collection. Tissue Samples Preservation Each fish was rapidly weighed to the nearest 0.1 gm in a tared, 1000 ml glass beaker. Weighing was performed on an Ohaus platform scale. The liver was quickly excised and the gall bladder removed. A portion of the liver for general histopathological examination was placed into Bouin's fixative. A piece of liver, approximately 3 mm3 , was placed into OCTR embedding medium and frozen in liquid nitrogen-cooled 2- methylbutane. The remaining liver was freeze-clamped in a liquid nitro- gen-cooled, aluminum tissue smasher in order to halt enzyme activity. The brain was then excised and freeze-clamped. The left gill arches, for general histological examination, were removed to Bouin's fixative. Gross observations were made on the carcass, which was then tagged and placed into buffered, 10% formalin. All freeze-clamped tissues were stored in liquid nitrogen. 12 TABLE 1 Measurement Methods for 0n-site Water Quality Parameters Performed During the Flint River Bioassay Study Measurement Method Temperature Hand-held thermometer pH Sargent-Welch Model PBL Portable pH Meter DO YSI Model 54 Oxygen Meter TRca DPDb Titrimetric aTotal residual chlorine (C12 + HClO + ClO- + mono-di-tri- chloramine). bN,N-diethyl-p-phenylene-diamine. 13 Histological Procedures Gill and liver tissues for routine histological examination were dehydrated in an ethanol (ETOH) series, cleared in xylene, and embedded in ParaplastR. using an Autotechnicon Model 2A Tissue Processor (Table 2). Tissue sections (5 um) were cut on an American Optical microtome, mounted and stained with Harris hematoxylin and Eosin (H and E) accor- ding to the method of Humason (1972). Pro-TexxR was used as the mountant. Unfixed, frozen tissue sections (10 um) for qualitative glycogen staining were cut on an IEC cryostat, mounted, and fixed in Carnoy's Fluid for 15 min at 4°C. The sections were then cleared in absolute ETOH for 15 min to remove any chloroform. The sections were hydrated and stained in Best's Carmine and Harris hematoxylin, according to a modified method of Humason (1972). The sections were not coated with nitrocellulose or celloidin, and Harris hematoxylin was substituted for Mayer hematoxylin. Sections were stained in Harris hematoxylin for 5 min, followed by a 5 min rinse in tap water. The sections were then stained for 15 min in Best's Carmine, followed by Best's differentiator for l min. Lastly, the sections were dehydrated, cleared, and mounted in Pro-TexxR. Unfixed, frozen tissue sections (15 pm) for qualitative lipid staining were cut on an IEC cryostat and mounted. Staining was with Oil Red 0 and Harris hematoxylin according to the method of Disbrey and Rack (1970). Photomicrographs of tissues were taken with a Zeiss Photomicroscope. Black and white photographs were shot using Kodak Panatomic-X high grain film. Color photographs were shot using Kodak Ektachrome 160 Tungsten film. 14 TABLE 2 Dehydrating and Embedding Schedule for Fish Gill and Liver Tissues. Process Performed on an Autotechnicon Model 2A Tissue Processor Step Tissue Treatment Duration 1 50% ETOH* 2 h 2 70% ETOH 2 h 3 lst 95% ETOH 2 h 4 2nd 95% ETOH 2 h 5 lst 100% ETOH 2 h 6 2nd 100% ETOH 2 h 7 lst xylene 2 h 8 2nd xylene 2 h 9 lst Paraplast 2 h 10 2nd Paraplast 2 h *ETOH = % ethanol (v/v) 15 Glutamate Dehydrogenase Assay Liver GDH activity was determined according to a modified method of Wilson (1973). Each frozen liver sample was weighed to the nearest 0.001 gm and placed in ice-cold 0.2 M potassium phosphate buffer at pH 7.6. The volumes were adjusted so as to obtain approximately 100 mg wet wt tissue/ml buffer solution. Samples were homogenized with a motor-driven TRl-R Instruments Model K43 teflon tissue grinder and then sonicated for two min at 130-140 watts with a Braunsonic 1510 Sonica- tor. The homogenate was centrifuged for 30 min at 14,000 rpm at 4°C. The supernatant was removed and used directly for the GDH analysis. The assay mixture contained 1.85 ml of 0.2 M potassium phosphate buffer, pH 7.6; 0.15 ml of 3 M NH4C1 in the phosphate buffer; 0.6 m1 of 0.45 mM NADH in the phosphate buffer; 0.2 ml of 0.167 M potassium a-KGA in the phosphate buffer; 0.2 ml of GDH standard or supernatant. Potassium a-KGA, NADH, and GDH stock solution Type II (703 Units/m1) were obtained from Sigma Chemical Co., St. Louis. All other reagents were analytical grade. GDH activity was assayed kinetically. Oxidation of NADH in the GDH catalyzed reaction was recorded on a Beckman chart recorder as the decrease in absorbance at 340 nm, and the rate of loss was proportional to the amount of GDH present. The line segment between 30-90 sec after addition of GDH standard or supernatant to the reaction cuvette was used to calculate the slope. GDH activity was calculated from a GDH standard curve, plotting slope versus Units GDH. A Unit is defined as the amount of GDH that converts 1 pmole of a-KGA to glutamate per min at 25°C. Results were expressed as Units GDH/mg protein. 16 Protein Soluble protein was determined by a modification of the method described by Bergmeyer (1974). Folin-Ciocalteu Reagent (Sigma Chemical Co., St. Louis) was diluted 1:1 with deionized water. Bovine serum albumin (Sigma Chemical Co.) was used as the standard. Absorbance was measured at 660 nm on a Varian 634 dual beam Spectrophotometer. Pro- tein determination was performed on a 40-fold diluted portion of the GDH assay supernatant. Brain Tissue Assays Deproteinization. Brain tissue was deproteinized according to a modification of the method of Dagley (1974). Each brain was rapidly weighed and ground to a powder in a stainless steel tissue grinder cooled in liquid nitrogen. Five hundred pl of 2 N HClO4 was frozen and ground in a separate stainless steel tissue grinder cooled in liquid nitrogen. The two frozen powders were combined in a plastic centrifuge tube and kept on dry ice. The centrifuge tube was then allowed to reach 4°C and 1.5-2.0 ml of ice-cold deionized water was added. The centrifuge tube was then allowed to stand for 10 min before the con- tents were homogenized for 60 sec with a motorized teflon tissue grinder. The homogenate was centrifuged at 3000 g for 15 min at 0°C. The supernatant was pipetted off, neutralized with 2 N KHC03, and centrifuged again under the same conditions as the first time. Half of the final supernatant was stored at -20°C for the glutamate and gluta- mine assays. The remaining supernatant for the a-KGA assay was stored on dry ice. All reagents were analytical grade. l7 Glutamate. Glutamate concentrations were determined based on a method similar to Witt (1974). The principle of this enzymatic spec- trophotometric assay is that the reduction of 3-acetylpyridine adenine dinucleotide (APAD) to APADH is proportional to the amount of glutamate present. The amount of APADH formed during the assay was determined by measuring the increase in absorbance at 363 nm on a Varian 634 Spec- trophotometer, and comparing to a standard curve. The assay mixture contained 0.1 ml of standard or sample extract; 0.5 m1 of 66 mM phos- phate buffer, pH 8.2; 0.05 ml of 6.5 mM APAD solution; 0.330 ml of deionized water; 10 ul GDH stock solution. Glutamate, GDH stock solu- tion Type II (703 Units/ml), and APAD were obtained from Sigma Chemical Co. Glutamine. Glutamine concentrations were determined by the method of Lund (1974). The assay is based on the enzymatic hydrolysis of glutamine with purified glutaminase (E.C. 3.5.1.2) to yield gluta- mate, which is then quantitated spectrophotometrically according to the method of Witt (1974). The tissue deproteinization procedure de- scribed by Lund (1974) was not followed. The assay mixture consisted of 0.4 ml of 0.5 M acetate buffer, pH 5.0; 0.1 ml of 20 mM hydroxyl- amine; 0.1 m1 of 1 mM glutaminase in the acetate buffer; 0.05 ml of glutamine standard or 0.3 ml sample extract; deionized H20 to make 1.0 ml. Glutamine, hydroxylamine, and glutaminase Type II were obtained from Sigma Chemical Co. All other reagents were analytical grade. The glutaminase contained glutamate decarboxylase. Thus, 0.1 ml of 20 mM hydroxylamine was added to the hydrolysis reaction mixture to act as a powerful inhibitor of glutamate decarboxylase. 18 a-Ketoglutarate Assay. a-KGA was assayed according to the method of Narins and Passonneau (1974). The basis for this fluorometric assay is that the decrease in fluorescence resulting from the oxidation of NADH is stochiometrically proportional to the amount of a-KGA present. The analyses were performed on an Aminco SPF 500 Spectrofluorometer. Maximum fluorescence was obtained at excitation wavelength 346 nm, and emission wavelength 464 nm. The assay mixture consisted of 1 ml of reagent mixture; 0.1 ml standard or sample extract; 10 p1 of GDH stock solution. The reagent mixture contained 5 m1 of 1.0 M phosphate buffer, pH 6.8; 4 ml of l M NH4-Ac; 0.1 ml 0.1 M EDTA; 0.1 mi 0.1 M ADP; 0.2 mi of 2.56 mM NADH in 0.1 M carbonate buffer, pH 10.6; deionized H20 to make 100 m1. GDH stock solution Type II (703 Units/ml), NADH, ADP, and a-KGA (free acid) were obtained from Sigma Chemical Co. Other reagents were analytical grade. Sample concentrations were determined from an a-KGA standard curve. Experimental Design and Statistical Analyses This experiment was a completely randomized design. "Treatments" were actually the site-specific water conditions to which the test fish were exposed at the FRl and FR5 locations. Thus, "treatment" differ- ences were actually determined as the differences among the FRl, FR5, and LR location fish. Hereafter, "location" will represent the FRl, FR5, and LR groups of fish, and will be synonomous with "treatment". Differences between the FRl and FR5 mean wet weight gain per fish were evaluated by a Student's t-test. Analysis of liver GDH activity, brain concentrations of glutamate, glutamine and a-KGA, and gluta- mine/glutamate ratio, all involved unbalanced 3-way nested designs. 19 The 3-way nested structure was imposed because in each of the four analyses only a small portion of the total number of samples could be analyzed in one day. Thus, analysis day was nested in location, fish were nested in analysis day, and two determinations were made per fish. Reduction of data, analysis of variance (ANOVA), and calculation of the coefficients of the expected mean squares (EMS) were obtained by using the Statistical Analysis System (SAS) 79, PROC NESTED routine (Helwig and Council, 1979). Satterthwaite's (1946) approximate test procedure was used to synthesize mean squares and degrees of freedom for the F- test of significant location and analysis day effects. RESULTS AND DISCUSSION Water‘guality Mean temperature, pH, hardness, alkalinity, TOC, and total dis- solved solids showed little variation among the 5 study sites (Table 3). However, Fe and total cyanide concentrations, and C00 were pro- gressively greater at each site downstream from the outfall. In con- trast, DO, TRC, total ammonia, and UIA concentrations decreased at each site downstream from the outfall (Table 3). Total ammonia, UIA, and TRC concentrations were slightly greater at the FR3 site than at the FR2 site. This was probably due to insufficient mixing of the effluent plume with the river water at the FR2 site. Heavy metals and several classes of organic compounds were below detection limits (Table 4). Water quality results will be discussed more thoroughly throughout the remainder of this section, particularly in relation to observed biological effects in the fish. Vandalism The cage at the FR4 site was discovered vandalized on day 12, and contained no test fish. The vandalism resulted in the unfortunate loss of much biological information about the test fish at the FR4 site. No replacement fish were put in the vandalized cage because it was physically damaged. 20 m. H.335» 21 ....mo xvapm ammmoowm sm>wm peep; as» mcwczo pcmzpwwm mp3: mcocmmm use mco_umoo4 succumtcm on» we muwumwswaumsmsu xHWszo swam: m m4m0.05). Smith and Piper (1975) observed no effect on the growth of rainbow trout exposed to 0.010 and 0.016 mg/l UIA for 4 months; however, growth was significantly reduced after 6 and 12 months exposure to 0.017 mg/l UIA. Similarly, Robinette (1976) reported that the growth of channel catfish fingerlings cultured for 4 weeks at 0.01 and 0.06 mg/l UIA were not significantly different from controls. However, significant dif- ferences in growth were observed between experimental and control fish at 0.12 and 0.13 mg/l UIA after 4 weeks. These results support my conclusion that the duration of the present study was too brief to adquately assess the effects on fish growth rates. Histopathology Gill. Gill sections from all of the fish from the FRl, FR5, and LR samples were examined. Gill sections from 20-35% of the fish from each of the three locations contained external parasites, or internal cysts in the filaments or secondary lamellae. The number of cysts per 29 TABLE 5 Initial and Final Total Group Weights and Mean Weight per Fish, as well as Mean Growth Rate for the FRl and FR5 Location Fish s2 = Sample variance. N = 15. Location FRl FR5 Initial Final Initial Final Total group wt 354.0 368.9 360.8 392.2 Mean wt per fish 23.6 24.6 24.1 26.1 (S.E.) 4.5 6.0 4.6 4.9 s2 18.7 33.6 19.9 22.6 Growth rate (gm/fish/day) 0.06 0.12 30 gill arch ranged from 1-15. Dr. Patrick Muzzall, Assistant Professor in the Department of Natural Science at Michigan State University, identified the cysts as Protozoans, Genera Myxosporidea or Micro- sporidea (?). The parasites and cysts did not appear to cause any inflammatory or necrotic response in surrounding tissues, except for slight hyperplasia immediately surrounding a few of the cysts fused between two secondary lamellae. According to Dr. Muzzall, the light degree of infestation of the parasites probably resulted in negli- gible effects to the fish. Gill sections from the LR sample fish exhibited no hyperplasia or hypertrophy of the filaments, and the secondary lamellae were long and slender (Figures 5 and 6). The LR fish gill sections resembled those from normal, healthy channel catfish (Grizzle and Rogers, 1976). Gill sections from approximately 40% of the FRl upstream control sample fish did have hyperplasia and hypertrophy of the respiratory epithelium and the cells at the base of the secondary lamellae (Figure 7). One blood-filled aneurysm was also observed. Gill sec- tions from the remainder of the FRl fish showed no histopathological damage. In marked contrast, the gill sections of all of the FR5 fish had one or more moderate to severe histopathological effects. The most prevalent of these effects, observed in all 15 FR5 fish, were severe hyperplasia and moderate hypertrophy of the respiratory epithelium and cells at the base of the secondary lamellae (Figures 8 and 9). Clubbing and fusion of the secondary lamellae were observed in 60% of the FR5 fish. Edema, resulting in separation of the respiratory 31 Figure 5. Photomicrograph of a laboratory reference (LR) fish gill section. Note the long, slender secondary lamellae, and the absence of cellular hypertrophy or hyperplasia. H and E. X120. '8. Y 5 1:14 f»; ‘k‘I L r ' Figure 6. Photomicrographic close— up of an LR fish gill section, showing the absence of histopathological damage. H and E. X480. 32 I:' —. .."J:,'.V',.- '9; I)" - :r \t ' A}. )f'lfi .‘w ”pk-fl) ‘b‘ Figure 7. Photomicrographic close-up of a gill section from a FRl upstream control fish, showing the hyperplasia and hypertrophy in the respiratory epithelium. H and E. X480. 33 Figure 8. Photomicrograph of‘a gill section from a FR5 location fish. Note the severe hyperplasia and moderate hypertrophy in the respiratory epithelium. H and E. X120. ‘5 ~ ' Figure 9. Photomicrographic close—up of a gill section from a FR5 location fish, showing extensive hyperplasia of the respiratory epithelium. H and E. X480. 34 epithelium from the pillar cells in the secondary lamellae (Figure 10), was observed in 33% of the FR5 fish. Multiple, blood-filled aneurysms (Figure 11) were detected in the gill sections from two FR5 fish. As summarized by Smart (1976), ammonia, as well as many other toxicants such as heavy metals, detergents, organics and acids, are capable of producing histopathological changes in fish gill tissues. Numerous researchers have demonstrated in the laboratory that ammonia exposure may cause gill histopathological changes in fish. Flis (1968) exposed carp(0yprinus carpio) for 14 days to fluctuating UIA concentra- tions ranging from 0.036 to 2.382 mg/l, and observed severe destruc- tion of the cellular structure, formation of blood-filled aneurysms, and separation of the respiratory epithelium. Burrows (1964) observed extensive hyperplasia of the gill epithelium in chinook salmon (9229f rhynchus tshawytscha) after 6 wk exposure to 0.006 mg/l UIA. Larmoyeax and Piper (1973) reported fusion of gill lamellae, occasional fusion of gill filaments, edematous lamellae, and aneurysms in certain lamellae in chinook salmon exposed to a maximum UIA concentration of 0.014 mg/l for 8 months. Smith and Piper (1975) observed only mild, scattered hypertrophy of gill epithelium in rainbow trout exposed to 0.017 mg/l UIA for 4 months. However, the authors observed severe hyperplasia of the gill epithelium, and fusion of the lamellae, after 6 to 12 months of exposure. Robinette (1976) observed hyperplasia in the gill epi- thelium of channel catfish exposed to 0.010-0.130 mg/l UIA for 4 weeks. Smart (1976) reported no gill histopathological changes in rainbow trout exposed to acutely lethal ammonia concentrations. However, 35 51 w l 212;: .1! 2'” ,. ‘ I . g, f. ., ~ . V Figure 10. Photomicrograph of a gill section from a FR5 location fish, showing severe edema in the secondary lamellae, resulting in separation of the respiratory epithelium from the pillar cells. H and E. X120. igure ll. Photomicrographic close-up of a gill section from a FR5 location fish, showing two blood-filled aneurysms in the secon- dary lamellae. H and E. X480. Sm 30 36 Smart did observe swollen, rounded secondary lamellae in approximately 30% of the rainbow trout exposed to O.250-0.300 mg/l UIA for 14 days. Contrary to the preceding results, Mitchell and Cech (1983) re- ported no evidence of gill histopathology in either adult or juvenile channel catfish exposed to 0.530 and 0.401 mg/l UIA for 12 and 9 weeks, respectively. However, the authors saw severe gill hyperplasia in the juveniles after 5 weeks when the fish were exposed to the same UIA concentrations combined with monochloramine in concentrations up to 0.070 mg/l. It should be noted that in comparison to my study, the Mitchell and Cech (1983) juvenile catfish ammonia exposure was 4-fold longer in duration, and was conducted at an average UIA concentration more than lZ-fold greater. Yet, the authors reported no significant gill histo- pathological damage, and concluded that the histopathology observed in gill tissues from other ammonia exposure studies (Burrows, 1964; Larmoyeax and Piper, 1973; Robinette, 1976) may have been caused by either chlorine exposure or elevated levels of bacteria or particulate matter, rather than from direct effects of ammonia. It is difficult to accurately assess the biological significance of the gill histopathological changes observed in this study. The gill histopathology could not be correlated with any deleterious effects on growth or survival. However, the short duration of the exposure may have masked the expression of these effects which might have occurred, given sufficient time. In any case, there is evidence in the litera- ture, that histopathological changes such as those observed in this study, increase the susceptibility of fish to bacterial gill disease (Burrows, 1964; Smith and Piper, 1975; Smart, 1976). 51 DC Vd Drj to 37 Liygr, Certain liver sections from both the FRl and FR5 location fish were extremely vacuolated (Figures 12 and 13). Liver sections from 38% of the FRl fish were severely vacuolated, in contrast to 60% of the FR5 fish. The remaining fish from both locations had liver tissue sections that were considerably less vacuolated and possessed a greater number of nucleated hepatocytes (Figures 14 and 15). These less vacuolated liver sections will hereafter be referred to as "dense". All of the frozen, unfixed FR5 fish livers were inadvertently dis- carded before any sample sections were obtained. Frozen, unfixed tissue sections of the FRl fish livers were stained with Oil Red 0 lipid stain. No differences were observed in the qualitative lipid staining among severely vacuolated and dense liver sections (Figures 16 and 17). Additional frozen, tissue sections from the FRl fish livers were stained with glycogen-specific, Best's Carmine stain. Qualitative, positive, glycogen staining appeared to be more intense in the severely vacuolated liver sections, compared to the dense liver sections (Figures 18 and 19). Liver tissue sections exposed to human saliva (amylase) for l min before fixation and staining, exhibited negative glycogen staining, whereas liver sections from the same fish but not exposed to saliva exhibited positive glycogen staining (Figures 20 and 21). Vacuolation in fish liver cells may result from at least three primary processes, namely: 1) fat infiltration; 2) glycogen storage; 3) lytic necrosis. Specific staining procedures are usually performed to indicate the presence of lipid or glycogen. Lytic necrosis is 38 I Figure 12. Photomicrograph of a liver section from a FRl location fish, showing the severe vacuolation of the hepatocytes. Note the reduction in the number of nuclei and the absence of cytoplasmic de- tail. H and E. X480. Figure'13. Photomicrographkof a liver section from a location fish, showing severe vacuolation. Note the similarity to the FRl fish liver pictured in Figure 12. H and E. X480. fl, dwgb‘if’i I! 5‘ all..." Illlllllll i l 39 .‘O 49 ,. k‘ we ' 5&5: ;' f'; ,- { Photomicrograph of a liver section from a FRl location fish, showing less vacuolated and a greater number of nucleated hepatocytes. Note the increased cytoplasmic detail. H and E. X480. igure 5. Photomicrograph of a liver section from a FR5 location fish, showing the reduced vacuolation and greater number of nucleated hepatocytes. H and E. X480. 40 Figure 16. Photomicrograph of a severely vacuolated liver section from a FRl location fish, showing the negative lipid staining results after staining with Oil Red 0 and Harris hematoxylin. X480. Figure 17. Photomicrograph of a dense liver section from a FRl loca- tion fish, showing the negative lipid staining results after staining with Oil Red 0 and Harris hematoxylin. X480. 41 Figure 18. Photomicrograph of a severely vacuolated liver section from a FRl location fish, showing the intense, positive glycogen staining results after staining with Best's Carmine and Harris hema- toxylin. X313. [I . . .. I . ‘ , r ‘ ,6 , Figure 19. Photomicrograph of a dense liver section from a FRl loca- tion fish, showing the positive, but less intense glycogen staining results after staining with Best's Carmine and Harris hematoxylin. X313 42 o _ . _ b “ ‘ Figure 20. Photomicrograph of a FRl liver section pretreated with human saliva (amylase) for l min before fixation and staining with Best's Carmine and Harris hematoxylin. X480. Figure 21. Photomicrograph of a FRl liver section from the same fish as in Figure 20, only untreated with human saliva. Note the positive glycogen staining. Best's Carmine and Harris hematoxylin. X480. 43 more difficult to verify, and is typically diagnosed following negative lipid and glycogen staining results. The lack of positive lipid staining in either the severely vacuo- lated or dense FRl liver sections provides evidence that the vacuola- tion did not represent fat infiltration. Fatty change in fish liver may be associated with nutritional factors, as well as with exposure to chemical stressors, particularly organic xenobiotics (Couch, 1975). The water quality analysis data from the present study were consistent with the negative lipid staining observed in the fish livers, as PCB, phenols, chlorinated hydrocarbons and purgable aromatic hydrocarbons were all below detection limits in the water samples. Fatty changes in the FR5 fish livers as a result of starvation would be illogical be- cause weight gain was not significantly different among the FR5 and FRl fish. H and E stained liver sections from well-fed channel catfish con- tain a large degree of vacuolation which results from increased glyco- gen content (Grizzle and Rogers, 1976). The pattern of vacuolation produced from glycogen storage, closely resembles the severe vacuola- tion observed in the livers from fish in my study. Qualitative differ- ential glycogen staining intensity among the severely vacuolated and dense liver sections, coupled with negative glycogen staining in the sections after amylase exposure, strongly suggests that the vacuolation represented glycogen stores within the hepatocytes. Thus, I conclude that the severely vacuolated liver sections contained substantial quantities of glycogen. In N 44 The pattern of vacuolation in the H and E stained liver sections superficially resembled lytic necrosis. Enlargement of the cell, with a loss of cytoplasmic detail and staining quality are characteristics of lytic necrosis. Additionally, there is a reduction in the number of nuclei, and those present tend to remain centrally located within the cell. These features were evident in the vacuolated FRl liver sections, (Figures 12 and 13). However, my results from the glycogen staining procedures support the conclusion that the vacuolation in the FRl fish livers represented glycogen stores. It is unknown whether the FR5 severely vacuolated liver sections would have showed negative lipid and positive glycogen staining. The unfortunate discarding of the frozen FR5 livers before sections were cut, prevented the obtaining of any glycogen or lipid staining data from FR5 fish livers. However, vacuolation in the FR5 fish livers due to increased glycogen storage would not seem unlikely, because Tsai (1975) mentioned three studies in which fish growth rates increased below sewage treatment plants. Increased productivity below the sewage outfalls may provide additional food for the fish, resulting in greater glycogen deposition. In contrast, Buckley gt_gl, (1979) reported that liver glycogen increased in fish chronically exposed to ammonia solutions, but de- creased in fish exposed to "30% unchlorinated secondary-treated domes- tic sewage" [§1£J. The authors concluded that the glycogenesis ob- served from ammonia solution exposure was attributable to the factors of exposure time and strength of stimulus [sjgfl (concentration). The authors also concluded that the decrease in glycogen in fish exposed to the 30% sewage was not attributable to starvation, and factors other .—lo 45 than ammonia caused the hypermetabolic state that resulted in decreased energy stores. Liver Glutamate Dehydrogenase Mean liver tissue GDH specific activities among fish from the three locations ranged from 0.092-0.103 Units/mg protein (Table 6), and were not significantly different (P>0.05, 3-way Nested ANOVA, Appendix B, Tables B1 and B2). In comparison, Wilson (1973) reported liver GDH specific activities of 0.0294 and 0.0358 Units/mg protein in native and cultured channel catfish, respectively. Korsgaard (1982) reported 0.0307 Units/mg protein in the marine Zoarces viviparus. Iwata gt_§l, (1981) determined the specific activity of GDH ranged from approxi- mately 0.2-0.6 Units/mg protein in liver of the mudskipper (Perioph- thalmus cantonensis) adjusted to different salinities or kept out of water. In a study on the effects of dietary dieldrin to rainbow trout, Mehrle and Bloomfield (1974) stated that the liver GDH specific acti- vities ranged from 0.20-0.36 Units/mg protein. The liver GDH specific activities obtained in this study were within the range of values obtained by other researchers on fish GDH activities, although my results were slightly greater than those re- ported by Wilson (1973) for channel catfish. The fact that almost identical mean GDH activities were observed in fish livers from the three locations, in spite of significantly different UIA concentrations (Student's t-test, P<0.001), indicates that this enzyme is not a useful 1 situ indicator of ammonia exposure in channel catfish when mean UIA concentrations were 0.032 mg/l or less. 46' TABLE 6 Mean Liver Tissue GDH Specific Activities in the FRl FR5, and LR Fish Samples Activity is in Units/mg protein. Results are the mean : S.E. Location GDH Activity FRl 0.095:0.003 n=28 FR5 0.092:0.003 n=30 LR 0.103:0.0l3 n=20 C8! fig Sen 47 There are several possible explanations forthe lack of signifi- cant differences among liver GDH activities from the three locations. The first explanation is that even if channel catfish are exposed to UIA, liver tissue GDH activity is not induced. I know of no literature describing the effect of direct ammonia exposure on liver GDH activity in fish. However, Mehrle and Bloomfield (1974) reported that liver GDH activity in rainbow trout increased significantly after chronic dietary ingestion of dieldrin. The authors suggested that the increased liver GDH activity was in response to increased endogenous ammonia, which resulted from decreased ammonia detoxification in the brain. In another study, Iwata gt_gl, (1981) stated that liver GDH activity increased in the mudskipper when it was kept out of water or exposed to osmotic shock. The authors attributed this to free amino acid synthe- sis as a method of removing the resulting excess internal ammonia. A second possible explanation for the similar GDH specific acti- vities among the three locations is that the UIA concentrations were not sufficiently great to affect GDH activity. The present body of literature on the role of GDH in ammonia detoxification is insufficient to evaluate the metabolic role of this enzyme in animal tissues (Storey eta” 1978). Brain Metabolites Glutamate and Glutamine Concentration. Mean brain tissue con- centrations of glutamate and glutamine, were determined and the glutamine/glutamate ratios were calculated for the FRl, FR5, and LR fish (Table 7). No significant differences among locations were ob- served in the mean concentrations of glutamate or glutamine, or in 48 TABLE 7 Mean Concentrations of Glutamate and Glutamine, as well as Ratio of Glutamine/Glutamate from Brain Tissue of FRl, FR5, and LR Sample Fish Concentrations are in pmoles/g wet wt. Location FRl FR5 LR Glutamate X :_SE 3.04:0.29 3.03:0.29 2.02:0.15 (range) (0.70-6.34) (1.46-6.30) (0.76-2.86) n 30 28 20 Glutamine X :_SE 5.76:0.29 4.60:0.37 lO.96:1.22 (range) (3.00-8.53) (0.66-9.08) (l.06-21.80) n 28 27 20 Ratio X i SE 2.53:0.29 1.58:0.08 6.93:1.26 (range) (0.99-6.35) (0.45-2.29) (0.40-21.00) n 28 27 20 49 their ratio in brain tissues (P>0.05, 3-way Nested ANOVA, Appendix B, Tables B3-B6). The glutamate and glutamine concentrations observed in this study were similar to those reported previously for channel catfish, carp and rainbow trout (Table 8). The mean glutamate and glutamine concen- trations in the LR sample fish (2.02 and 10.96 pmoles/g wet wt, re- spectively) were similar to the corresponding concentrations in rainbow trout brain (2.67 and 8.80 pmoles/g wet wt, respectively) after 24 h exposure to 0.500 mg/l UIA (Arillo gt_gl,, 1981). Additionally, the FR5 location fish in this study were exposed to an average UIA concen- \, tration of 0.032 mg/l, and contained mean brain glutamate and glutamine concentrations of 3.03 and 4.60 pmoles/g wet wt. These results were similar to the brain glutamate and glutamine concentrations in rainbow trout (4.74 and 4.80 pmoles/g wet wt) exposed for 48 hr to 0.020 mg/l UIA (Arillo gt_gl,, 1981). Although the mean brain glutamate and glutamine concentrations in the fish from this study were similar to the concentrations reported by other researchers, my results raise two questions which need to be addressed in order to evaluate the utility of fish brain glutamate and glutamine concentrations as ig_situ indicators of ammonia exposure. The first question is, "why were the glutamate concentrations nearly identical among the three location groups of fish?" Secondly, "why were the glutamine concentrations also not significantly different among the three location groups of fish?" The fact that mean brain glutamate concentrations were nearly identical among locations may not be unusual. Levi gt_al, (1974) re- ported that while brain glutamine concentration increased lO-fold in 50 mcwEmuapm u cpo mumsmua—m n span a .comuowwcw .a.~ swede awe m mwmxpmcm muwponopms mu: xvon ox son uwm mean :a o: "muwmg mzz ze mm.o an ; wetem so» we: msamoaxme 11.11. --- em.w m.__ 11mo_oee o_ omoe Ankm_v ._6 So mewxzez --- Fs.e e.~_ _oLSeoo epo>cw was» mmwuaum :H .mmammwa a; pm: m\mmpoEn cw msm mcowpmsucmocou Sam on“ nee 5mm; cw .owumm use .mcwsmuapo ecu mumsmazpw mo mcowumspcoucou mammwp :wmsm do mmspm> mssgmcmpm4 w u4m0.05, 3-way Nested ANOVA, Appendix B, Tables B3 and B7). I am not aware of any data on brain o-KGA concentrations in fish. For comparison, I have summarized the results from several studies on brain o-KGA concentrations in rats which were administered i.p. injec- tions of NH4-Ac (Table 10). The mean brain o-KGA concentrations in the fish from this study were approximately 3-7-fold less than the concen- trations reported in the rat studies. I knew a_prigri that evaluating the o-KGA results from this study in regards to o-KGA as a potential indicator of ammonia exposure would be hampered by the lack of data on o-KGA brain concentrations in fish, and the paucity of studies on the effects of subchronic ammonia expo- sure on brain o-KGA concentrations in fish or mammals. It is inter- esting, though, that the mean brain o-KGA concentrations in the FRl and FR5 location fish were nearly identical, even though the mean UIA concentrations were significantly different (P<0.001). This suggests that brain o-KGA concentration is not an effective indicator 0f.ifl situ ammonia exposure at mean UIA concentrations of 0.032 mg/l or less. 58 TABLE 9 Mean Brain o-KGA Concentration in Fish from the FRl, FR5, and LR Locations Concentrations are umoles/g wet wt‘: S.E. Location o-KGA Concentration FRl 0.03:0.01 (range) (0.01-0.12) n=26 FR5 0.02:0.004 (range) (0.01-0.06) n=28 LR 0.03:0.001 (range) (0.01-0.07) n=l9 59 11.11. eom.o m o.op Amem_v ._6 So meexso: wo~.o m _osoeoo NmP.o m_ m.e Fmp.o m_ N.m omp.o m_ e.N A_Nopv ohmowm nee epoeeewz _m_.o m_ Poeueou ,Pmo.o o.mp News.o m.~ m.s .11.11 memo.o o.mp Auemfiv .Fe 8 xoeoem mmmo.o m.~ Foaeeou - Aeeev eoeoooeeH AGS\mo_o v mucmcwwmm cowumsucmocou wo mama :? meowumsucmucou assumcmuwg op m4m