Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Aim Arbor MI 48106-1346 USA 313/761-4700 800/521-0600

DISPOSITION OF PROFLAVINE AND ACRIFLAVINE IN RAINBOW TROUT

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the

Graduate School of The Ohio State University

Zhengrong Yu, B.S., M.S.

The Ohio State University

1996

Dissertation Committee:

Dr. William L. Hayton, Adviser Approved by

Dr. Kenneth K. Chan

Dr. William J. Collins Adviser' Dr. Richard H. Reuning College of Pharmacy UMI Number: 9710692

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeh Road Ann Arbor, MI 48103 DISPOSITION OF PROFLAVINE AND ACRIFLAVINE IN RAINBOW

TROUT

Zhengrong Yu, Ph.D.

The Ohio State University, 1996

Professor William L. Hayton, Adviser

Acriflavine (3,6-diamino-10-methyl acridine. A) and proflavine (3,6- diamlnoacridine, P) have potential for use in aquaculture as antiinfectives; knowledge of their disposition in fish would guide their effective use while avoiding human exposure. This dissertation describes: (1) an HPLC method for

A, P and their metabolites, (2) the pharmacokinetics of A and P in trout, (3) disposition of A and P in trout after water exposure, and (4) the metabolite profile of P in trout.

The HPLC method for the determination of A, P, and their metabolites in fish fluids and tissues involved solid-phase extraction for sample clean-up, and gradient, reversed-phase HPLC separation with UV detection.

The pharmacokinetic behaviors of A and P were comparable; their plasma concentration-time profiles appeared triexponential after intravascular bolus administration, and the terminal half-lives were 25.9 hr for P and 38.3 hr for A.

Large Vgg values indicated that they distributed extensively into tissues, possibly due to avid binding to DNA. Since A and P ionize at pH 7-8, their branchial elimination was low and their elimination was via metabolic and renal pathways.

Disposition of A and P was studied at various times after water exposure of the fish for 4 hr to 4 ppm ’^C-proflavine or ^^C-acriflavine. Immediately after exposure, the concentration of radioactivity in all tissues and fluids was below the exposure water concentration, except for bile. The rank order of residue concentration (high to low) was bile > kidney > liver > skin > plasma > muscle.

During the 8 day depuration period, the rank order of residue concentration in the tissues was the same as at the end of exposure. The relatively high concentrations of acriflavine in skin and kidney after 16 days of depuration were probably due to irreversible binding.

Metabolites observed from P-treated fish were characterized by acid hydrolysis, enzymatic treatment, UV-VIS absorption and mass spectra. The metabolites were determined to be 3-N-glucuronosyl proflavine (PG), 3-N- glucuronosyl, 6-N-acetyl proflavine (APG), and 3-N-acetyl proflavine (AP).

Metabolite identities were verified by chemical synthesis and HPLC co-elution.

An in vitro study confirmed this characterization and showed that APG can form from glucuronidation of AP or acétylation of PG. Finally, a mass balance study showed a 90% recovery of the administered dose, and supported the conclusion that other metabolic pathways were not significant.

iii To my family

IV ACKNOWLEDGMENTS

I would like to express my most sincere gratitude to my advisor, Dr.

William L. Hayton, for his invaluable instruction, encouragement and personal concern throughout this study.

With much sincerity, I wish to express my personal appreciation to Dr.

Kenneth K Chan for his help and advice on the metabolites analysis.

Appreciation goes to the other members of my advisory committee, Drs.

William J. Collins and Richard H. Reuning, for their comments, suggestions, and help.

The technical assistance of Mr. Andy Vick, Brian Kemmenoe and Tim He is gratefully acknowledged.

Further, I am thankful to Dr. Annamaria Szoke for her friendship and stimulating discussions.

Thanks go to my fellow graduate students, Kelli Clark, Erin Swope,

Kirsten Engelfried, Hong Mei and the graduate students in the College of

Pharmacy for their encouragement and friendship.

Finally, financial assistance from The Ohio State University, College of

Pharmacy, and the U.S. Food and Drug Administration Center for Veterinary

Medicine is gratefully acknowledged. VITA

October 13,1968 ...... Bom - Qiqihar, China

1991...... 8.S. in Pharmacy Beijing Medical University Beijing, China

1993...... M.S. in Chemistry Cleveland State University Cleveland, Ohio

1991 -1993...... Teaching Associate, Department of Chemistry, Cleveland State University

1993 -1996...... Teaching Associate, Graduate Research Associate, College of Pharmacy, The Ohio State University

PUBLICATIONS AND PRESENTATIONS

(1) "Pharmacokinetics of Cortisol in Rabbits", Z. Yu, B. S. Thesis, Beijing Medical University, Beijing, China, 1991.

(2) "An Ultra-sensitive Electrochemical Enzyme Immunoassay for Thyroid Stimulating Hormone in Human Serum", Z Yu, Y. Xu, and M. P. C. Ip, J. Pharm. and Biomed. Analysis 1994. 12: 787-93.

(3) "Adsorption Profiles of Active Antibodies on Negatively Charged Polystyrene Microtiter Plates", Y. Xu, Z. Yu, W. R. Heinemann and H. B. Halsall, Anal. Biochem., submitted. vi (4) "Heterogeneous Enzyme Immunoassay for Thyroid Stimulating Hormone in Human Serum by Flow Injection Amperometric Detection", Z. Yu, Y. Xu, and M. P. C. Ip, Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Atlanta, Georgia, March 8-12,1993

(5) "Separation of Proflavine and Acriflavine by Reverse Phase HPLC", Z Yu, A. Szoke, and W. L. Hayton, Annual meeting of Society of Environmental Toxicology and Chemistry, Denver, Colorado, October 30 - November 4,1994.

(6) "Pharmacokinetics of Proflavine in Rainbow Trout", Z Yu, R. Abbas, A. Vick, S. Doddapneni, and W. L Hayton, Annual meeting of Society of Toxicology and Chemistry, Denver, Colorado, October 30 - November 4,1994.

(7) "Separation of Proflavine and Acriflavine by Reverse Phase HPLC", Z Yu, A. Szoke, and W. L. Hayton, Annual meeting of American Association of Pharmaceutical Scientists, San Diego, California, November 6-10,1994.

(8) "Pharmacokinetics of Proflavine in Rainbow Trout", Z Yu, R. Abbas, A. Vick, S. Doddapaneni, and W. L. Hayton, Annual meeting of American Association of Pharmaceutical Scientists, San Diego, California, November 6-10, 1994.

(9) "Pharmacokinetics of Acriflavine in Rainbow Trout", W. L. Hayton, Z. Yu, and A. Vick, International Congress of Toxicology VII, Seattle, Washington, July 2-6, 1995.

(10) "Electrochemical Enzyme Immunoassay for Thyroid Stimulating Hormone in Human Serum", Y. Xu, Z. Yu, and M. P. C. Ip., the 22nd Annual Conference of the Federation of Analytical Chemistry and Spectroscopy Societies, Cincinnati, Ohio, October 15-20, 1995.

(11) "Pharmacokinetics of Acriflavine and Its Metabolite in Rainbow Trout", Z. Yu, A. Vick, and W. L. Hayton, Annual meeting of American Association of Pharmaceutical Scientists, Miami, Florida, November 5-9, 1995.

(12) "Metabolic Fate of Proflavine and Acriflavine in Rainbow Trout", Z Yu, W. L. Hayton and K. K. Chan, Annual meeting of Society of Toxicology, Anaheim, California, March 10-14, 1996.

(13) "Characterization of Proflavine Metabolites in Rainbow Trout", Z. Yu, W. L. Hayton and K K. Chan, Annual meeting of American Association of Pharmaceutical Scientists, Seattle, Washington, October 26-31,1996.

VII FIELDS OF STUDY

Major Field; Pharmacy

Studies in Analytical Method Development, Pharmacokinetics and Disposition, Drug Metabolism

VIII TABLE OF CONTENTS

Dedication ...... iv

Acknowledgments ...... v

Vita...... vi

List of T ables ...... xiii

List of Figures...... xvi

Chapters:

1. General introduction ...... 1

1.1 Background ...... 1

1.2 Problem statement ...... 6

1.3 Organization of dissertation ...... 7

1.4 References ...... 8

2. Analytical method development- specific and sensitive determination of proflavine and acriflavine in trout fluids and tissues by reverse phase HPLC...... 9

2.1 Abstract...... 9

2.2 Introduction ...... 10

2.3 Experimental ...... 11 2.3.1 Reagents and materials ...... 11 2.3.2 Instruments ...... 12 2.3.3 Sample extraction ...... 13

ix 2.3.4 Metabolite Isolation ...... 14 2.3.5 HPLC analysis ...... 15 2.3.6 Acriflavine purification ...... 16 2.3.7 Preparation of standard ...... 16 2.3.7 Validation ...... 17 2.4 Results and discussion ...... 18 2.4.1 Chromatography...... 18 2.4.2 Acriflavine purification ...... 19 2.4.2 Optimization of extraction ...... 19 2.4.3 Validation ...... 30 2.5 Conclusion ...... 30

2.6 References ...... 33

3. Pharmacokinetic study of acriflavine and proflavine after la bolus 35

3.1 Abstract...... 35

3.2 Introduction ...... 36

3.3 Experimental ...... 37 3.3.1 Reagents and materials ...... 37 3.3.2 Instruments ...... 38 3.3.3 Animal protocol ...... 39 3.3.4 Fish cannulatlon ...... 41 3.3.5 Dosing ...... 44 3.3.6 Sampling tim es...... 45 3.3.7 Sample preparation and analysis ...... 45 3.3.8 Stability study ...... 47 3.3.9 Plasma protein binding ...... 48 3.3.10 Pharmacokinetic parameter calculations ...... 49

3.4 Results and discussion ...... 53

3.5 References ...... 59

4. Disposition of proflavine and acriflavine In rainbow trout after water exposure ...... 62

4.1 Abstract...... 62 4.2 Introduction ...... 64

4.3 Experimental ...... 65 4.3.1 Reagents and materials ...... 65 4.3.2 Instruments ...... 65 4.3.3 Experimental animal ...... 65 4.3.4 Sample collection, preparation and analysis...... 67 4.3.5 Freezer stability study ...... 71 4.3.6 Pilot study ...... 77 4.3.7 Whole-body autoradiography ...... 83 4.3.8 Main study ...... 88

4.4 Results and discussion ...... 96 4.4.1 Freezer stability study ...... 96 4.4.2 Pilot study ...... 109 4.4.3 Autoradiography ...... 119 4.4.4 Main study ...... 145

5. Characterization of proflavine metabolites In rainbow trout ...... 168

5.1 Abstract...... 168

5.2 Introduction ...... 169

5.3 Experimental ...... 170 5.3.1 Reagents and materials ...... 170 5.3.2 Instruments ...... 171 5.3.3 Animal protocol ...... 171 5.3.4 Metabolite Isolation ...... 171 5.3.5 HPLC analysis ...... 172 5.3.6 Acid Incubation ...... 173 5.3.3 Enzyme Incubations ...... 174 5.3.7 Absorption spectra ...... 174 5.3.8 Mass spectrometry ...... 175 5.3.9 Chemical synthesis...... 175 5.3.10 In vitro metabolism ...... 176 5.3.11 Mass balance ...... 178

5.4 Results and discussion ...... 179 5.4.1 HPLC analysis ...... 179 5.4.2 Add Incubations ...... 180 xl 5.4.3 Enzyme treatment ...... 186 5.4.4 Absorption spectra ...... 186 5.4.5 Mass spectrometry ...... 187 5.4.6 Chemical synthesis...... 196 5.4.7 In vitro metabolism ...... 196 5.4.8 Mass balance ...... 198

5.5 Conclusion ...... 207

5.6 References ...... 211

Bibliography ...... 213

XII LIST OF TABLES

TABLE PAGE

1.1 Physicochemical properties and toxicity of proflavine and acriflavine ...... 3

2.1 Extraction recovery of proflavine in trout tissues and fluids ...... 29

2.2 Intraday coefficient of variation in the determination of proflavine and acriflavine in plasma ...... 30

2.3 Interday coefficients of variation in the determination of proflavine and acriflavine in plasma ...... 31

3.1 Water quality parameters for the holding aquarium on recovery of proflavine in trout tissues and fluids ...... 40

3.2 Stability of proflavine, acriflavine and their metabolites in plasma at -2 0 °C...... 55

3.3 Pharmacokinetic parameters of proflavine and acriflavine in rainbow trout after intravascular administration ...... 58

4.1 Serial numbers, body weights, and genders of the fish used for the freezer stability study of proflavine and acriflavine ...... 75

4.2 i4Q.profiavine and i‘*C-acriflavine radioactivity and concentration in the exposure water for the freezer stability study ...... 76

4.3 Conditions of exposure and depuration water during the pilot study of proflavine ...... 80

4.4 Exposure and depuration conditions during the pilot study of acriflavine ...... 81

4.5 The daily water temperatures (°C), ph values, and dissolved oxygen values during the period of autoradiography study of rainbow trout exposed to 4 ppm ^ Vacriflavine ...... 86

xiii 4.6 The daily water temperatures (°C), pH values, and dissolved oxygen values during the period of autoradiography study of rainbow trout exposed to 4 ppm ’^C-acriflavine ...... 87

4.7 Serial numbers, body weights, and genders of fish used to study the total residue depletion and metabolite profile of proflavine ...... 91

4.8 Amounts of unlabeled and '^C-proflavine used to prepare 45 liters of exposure solution for the study of proflavine total residue depletion and metabolite profile. Also shown is the nominal concentration of proflavine and the concentration of radioactivity ...... 92

4.9 Serial numbers, body weights, and genders of fish used to study the total residue depletion and metabolite profile of acriflavine ...... 95

4.10 Concentration during freezer storage of total radioactivity, proflavine, and two metabolites in multiple samples of tissues and fluids from rainbow trout exposed for 4 hr to 4 ppm ’“‘C-proflavine ...... 106

4.11 Concentration during freezer storage of total radioactivity (Fish AC6), and of acriflavine and three metabolites (Fish AC5) in multiple samples of tissues and fluids from rainbow trout exposed for 4 hr to 4 ppm ^'’C-acriflavine ...... 108

4.12 ■'^C-proflavine equivalent (pQ/g) in trout tissues at 0, 2, 4, and 8 days after water exposure to 4.00 ppm i^c-proflavine for 4 hr ...... 112

4.13 ^“C-acriflavine equivalent (ng/g) in trout tissues and fluids at 0 and 8 days after water exposure to 4.01 ppm ’^C-acriflavine for 4 hr ...... 118

4.14 Concentration of radioactivity in rainbow trout tissues and fluids at various times after 4 hr exposure to 4 ppm ’^C-acriflavine. Concentration values are in arbitrary units as determined from phosphor imaging of 40 pm thick sagittal sections ...... 136

4.15 Concentration (ppb) of radioactivity in tissues and fluids at various times during and after 4 hr water exposure of rainbow trout to 4 ppm ^“C-proflavine. Each mean value ± SD represents duplicate determinations for 3 male or 3 female fish; i.e., n = 6 for each gender and n = 12 for combined genders ...... 149

4.16 Concentration (nmole/g) of proflavine and its metabolite in kidney, liver, and bile of female and male rainbow trout at various times during and after 4 hr water exposure to 4 ppm ’"’C-proflavine. Each mean value ± SD represents duplicate determinations for 3

xiv male or 3 female fish; i.e., n = 6 for each gender and n = 12 for combined genders ...... 151

4.17 Concentration (ppb) of radioactivity In tissues and fluids at various times after 4 hr water exposure of rainbow trout to 4 ppm ’“C- acriflavine. Mean values ± SD represent the following; treatment I: duplicate determinations for 3 male and 3 female fish (n = 6 for each gender and n = 12 for combined genders); treatment II: duplicate determinations for 2 male and 3 female fish (n = 4 for male fish, n = 6 for female fish and n = 10 for combined genders); treatment III: duplicate determinations for 1 male and 5 female fish (n = 2 for male fish, n = 10 for female fish and n = 12 for combined genders) ...... 161

4.18 Concentration (ppb acriflavine equivalent) of acriflavine and its metabolites in fluids and tissues of female and male rainbow trout at various times after 4 hr water exposure to 4 ppm ^'’C-acriflavine. Mean values ± SD represent the following: treatment I: duplicate determinations for 3 male and 3 female fish (n = 6 for each gender and n = 12 for combined genders); treatment II: duplicate determinations for 2 male and 3 female fish (n = 4 for male fish, n = 6 for female fish and n = 10 for combined genders); treatment III: duplicate determinations for 1 male and 5 female fish (n = 2 for male fish, n = 10 for female fish and n = 12 for combined genders) ...... 164

4.19 Concentration of radioactivity in muscle, skin and fillet - composed of muscle and skin in their natural proportions - from individual male and female rainbow trout. (Average ± SD concentrations are the mean of 1 male and 1 female fish, 5 determinations per fillet and 2 determinations each of muscle and skin samples.) ...... 165

5.1 TOF-MS results and relative error ...... 205

5.2 The distribution of proflavine and its metabolites in mass balance study ...... 206

XV LIST OF FIGURES

FIGURE PAGE

2.1 Representative HPLC chromatogram of trout plasma spiked with 4.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 |im Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 19

2.2 Representative HPLC chromatogram of trout bile spiked with 4.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 |im Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 20

2.3 Representative HPLC chromatogram of trout liver spiked with 4.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 21

2.4 Representative HPLC chromatogram of trout muscle spiked with 4.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 22

2.5 Representative HPLC chromatogram of trout kidney spiked with 4.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 23

2.6 Representative HPLC chromatogram of trout skin spiked with 1.0 nmole of proflavine (P) and 2.5 nmole of acriflavine (A). Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 24

2.7 Representative HPLC chromatogram of proflavine (P) and its metabolite (Met I) extracted from the plasma of proflavine treated fish. Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 25

xvi 2.8 Representative HPLC chromatogram of proflavine (P) and its three metabolite (Met I, II and III) extracted from the liver of proflavine treated fish. Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 26

2.9 Representative HPLC chromatogram of acriflavine (A) and its two metabolite (Met I' and H') extracted from the plasma of acriflavine treated fish. Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 27

2.10 Representative HPLC chromatogram of acriflavine (A) and its three metabolite (Met I’, 11’, and III’) extracted from the liver of acriflavine treated fish. Column: 300 mm x 3.9 mm i.d., 4 pm Nova-Pak C-18 from Waters; flow rate: 0.8 ml/min; detection: UV 262 nm ...... 28

3.1 The diagram of a cannulated fish: the cannula allowed ia bolus injection and blood sampling from free-swimming fish, the aquarium water was kept at 12 °C and was aerated with an air stone, the aquarium water was flow-through with a flow rate of 500 ml/min (not shown) ...... 43

3.2 Three compartment model of proflavine and acriflavine plasma pharmacokinetics after ia injection, proflavine or acriflavine were administered into the central compartment (1 ) and they distributed reversibly into rapidly (2) and slowly (3) equilibrating peripheral compartments, the k symbols represent the intercompartmental rate constants, and elimination (kio) was from the central compartment ...... 51

3.3 Proflavine plasma concentration-time profile in rainbow trout after intraaterial administration of 2.0 mg/kg of ’'‘C-proflavine. This profile is from a single representative trout (wt. 360 g). Symbols represent the experimentally determined concentrations and the line represents the least squares fit of the three compartment model-based triexponential equation...... 56

3.4 Acriflavine plasma concentration-time profile in rainbow trout after intraaterial administration of 2.0 mg/kg of ’’‘C-acriflavine. This profile is from a single representative trout (wt. 450 g). Symbols represent the experimentally determined concentrations and the line represents the least squares fit of the three compartment model-based triexponential equation...... 57

XVII 4.1 Concentration-time profile of radioactivity in fluids and tissues, from rainbow trout exposed to 4 ppm ’^C-proflavine for 4 hr, during storage in a freezer at -20 °C. Concentrations are expressed as a percentage of the corresponding zero time values ...... 103

4.2 Concentration-time profile of radioactivity in fluids and tissues, from rainbow trout exposed to 4 ppm ^^C-acriflavine for 4 hr, during storage in a freezer at -20 °C. concentrations are expressed as a percentage of the corresponding zero time values ...... 104

4.3 ’“‘C-proflavine equivalent (pg/g) in trout tissues and fluids after various times of depuration that followed a 4 hr water exposure to 4 ppm ^^C-proflavine ...... 113

4.4 Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ^^C-proflavine. Sections were exposed on the phosphor screen for 28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. No depuration phase. KEY: BR, brain; GB, gall bladder; GL, gill; H, heart/blood; I, intestine; K, kidney; L, liver; M, muscle; 0, olfactory region; PC, pyloric caeca; SK, skin; SP, sp leen ...... 121

4.5 One day depuration. KEY: same as Figure 4.4, and IL, intestinal lumen ...... 123

4.6 Two days depuration. KEY: same as Figure 4.4, and AR, sectioning artifact; IL, intestinal lumen; LA, liver accumulation ...... 125

4.7 Four days depuration. KEY: same as Figure 4.4, and IL, intestinal lumen; LA, liver accumulation ...... 127

4.8 Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ‘'^c-acriflavine. Sections were exposed on the phosphor screen for 28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. No depuration phase. KEY: BR, brain; GB, gall bladder; GL, gill; H, heart/blood; I, intestine; K, kidney; L, liver; M, muscle; 0, olfactory region; PC, pyloric caeca; SK, skin; SP, sp leen ...... 137

4.9 Two days depuration. KEY: same as Figure 4.4, and IL, intestinal lumen ...... 139

XVIII 4.10 Four days depuration. KEY: same as Figure 4.4, and AR, sectioning artifact; IL, intestinal lumen; LA, liver accumulation ...... 141

4.11 Eight days depuration. KEY: same as Figure 4.4, and IL, intestinal lumen; LA, liver accumulation ...... 143

4.12 Concentration of radioactivity (proflavine equivalents) in rainbow trout tissues and fluids at various times after 4 hr water exposure of fish to 4 ppm ’^C-proflavine. Each point represents the mean from 3 male and 3 female fish and the bars represent ± SD ...... 152

4.13 Concentration of radioactivity (proflavine equivalents) in rainbow trout fillet with adhering skin at various times after 4 hr water exposure offish to 4 ppm ’^C-proflavine. Each point represents the mean from 1 male and 1 female fish and the bars represent ± SD. Also shown are the profiles for muscle and skin, as they appear in Fig. 4.12 ...... 153

4.14 Concentration of radioactivity (acriflavine equivalents) in rainbow trout tissues and fluids at various times after 4 hr water exposure of fish to 4 ppm ^^C-acriflavine. Each point represents the mean from 3 male and 3 female fish and the bars represent ± SD ...... 166

4.15 Concentration of radioactivity (acriflavine equivalents) in rainbow trout fillet with adhering skin at various times after 4 hr water exposure of fish to 4 ppm ’^C-acriflavine. Each point represents the mean from 1 male and 1 female fish and the bars represent ± SD. Also shown are the profiles for muscle and skin, as they appear in Fig. 4 .1 4 ...... 167

5.1 The scheme for the formation of proflavine metabolites from rainbow trout ...... 181

5.2 Typical HPLC chromatogram of proflavine metabolites extracted from plasma ...... 182

5.3 Typical HPLC chromatogram of proflavine metabolites extracted from liver ...... 183

5.4 Typical HPLC chromatogram of acriflavine metabolites extracted from plasm a ...... 184

5.5 Typical HPLC chromatogram of acriflavine metabolites extracted from liver ...... 185

XIX 5.6 Effect of enzymatic treatment on Met I: Met I was Incubated with p- glucuronidase for 24 hr, 87.3% of proflavine (P) was liberated ...... 188

5.7 Effect of enzymatic treatment on Met I: Control, Met I was incubated in buffer (with no p-glucuronidase) for 24 hr, 38.1% of proflavine (P) was liberated ...... 189

5.8 Effect of enzymatic treatment on Met I: Inhibition, Met I was incubated with p-glucuronidase and 20 mM saccharic acid 1,4- lactone for 24 hr, 40.6% of proflavine (P) was liberated ...... 190

5.9 Effect of enzymatic treatment on Met II; Met II was incubated with p-glucuronidase for 24 hr, 98.3% of Met III was liberated ...... 191

5.10 Effect of enzymatic treatment on Met II: Control, Met II was incubated in buffer (with no p-glucuronidase) for 24 hr, 38.0% of Met III was liberated ...... 192

5.11 Effect of enzymatic treatment on Met II: Inhibition, Met II was incubated with p-glucuronidase and 20 mM saccharic acid 1,4- lactone for 24 hr, 40.6% of Met III was liberated ...... 193

5.12 Absorption spectra of proflavine, (a): at pH 7.0, (b): at pH 12.0 ...... 194

5.13 Representative TOF-MS spectrum. Met II from fish liver. The mass of the MH* ion of met II was measured at 428.3 amu with 80 averaged laser shots. Sinapinic acid was used as the matrix. The mass of 207.7 amu and 225.3 amu was from the matrix and also acted as internal standard ...... 195

5.14 HPLC analysis of the synthetic proflavine glucuronide (PG): approximately 60% was PG, 30% was proflavine ...... 198

5.15 HPLC analysis of coelution of synthetic PG with Met I from trout liver extract: Mobile phase gradient to 25% of Mobile phase B ...... 199

5.16 HPLC analysis of coelution of synthetic PG with Met I from trout liver extract: Mobile phase gradient to 15% of Mobile phase B ...... 200

5.17 HPLC analysis of coelution of synthetic PG with Met I from trout liver extract: Mobile phase gradient to 10% of Mobile phase B ...... 201

5.18 Synthesis of proflavine metabolites in vitro: Formation of Met I, Met II, and Met III from proflavine ...... 202

XX 5.19 Synthesis of proflavine metabolites in vitro: Formation of Met II, from PG ...... 203

5.20 Synthesis of proflavine metabolites in vitro: Formation of Met II, from AP ...... 204

XXI CHAPTER 1

GENERAL INTRODUCTION

1.1. BACKGROUND

Acriflavine and proflavine are derivatives of acridine, a substance isolated from coal tar. The mixture was originally intended to be the 10-methyl component but it was discovered after its commercial development and marketing in the 1930's that 'bcriflavine” was in fact a mixture. In this dissertation, the term Acriflavine" will be used for 10-methyl-3,6- diaminoacridine, and the term iDroflavine" will be used for 3,6-diaminoacridine

(Scheme 1.1).

‘NH2 H2N

Proflavine Acriflavine 3.6-diaminoacridine 3.6-diamlno-10-methyl acridine

Scheme 1.1; Structure of proflavine and acriflavine. 1 For both acriflavine and proflavine, two absorption bands at 262 and about 450 nm are observed in the absorption spectra. For proflavine, the 450 nm band shifted to 395 nm at pH above 7.5. There was no comparable shift observed for acriflavine. The shift of this band for proflavine corresponds to the protonation of the ring nitrogen. The pKa of the ring nitrogen is above 12 for acriflavine, and is 9.65 for proflavine [1]. Because the resonance system favors the mono-ion charged proflavine or acriflavine, it is very difficult to protonate the amino groups. Therefore, only one pKa value is generally considered for proflavine and acriflavine. Although it is unusual to speak about an ionization constant for a quaternary amine, as in the case of acriflavine, the k- deficiency N-heterocycles of the acridine ring makes it subject to nucleophilic attack by the hydroxyl ion to form the carbinol (Scheme 1.2). The pKa of acriflavine characterizes this equilibrium. Both acriflavine and proflavine are slightly soluble in water. The aqueous solubility is 1:250 for acriflavine, 1:250 for proflavine hemisulfate salts, and 1:2000 for proflavine base [1]. Their physicochemical properties are summarized in Table 1.1.

NH2 H o N

Salt form Carbinol, base form

Scheme 1.2: The salt and the base forms of acriflavine. Property Proflavine Acriflavine (hemisulfate) (hydrochloride)

Aqueous Solubility 4.0 4.0 (gm/L)

pKa 9.65 > 12

Koctanol/water 0.7 <0.2

LD 50* 0.14 0.014 (gm/kg body wt.)

^max(logs) pH 12: 262(4.76), 395(4.25) 452(4.67) pH 7: 261(4.73), 444(4.59)

*Mouse via subcutaneous Injection.

Table 1.1: Physicochemical properties and toxicity of proflavine and acriflavine

[1].

Prior to modem methods of analysis, proflavine and acriflavine were quantified by the ferricyanide reaction, in which proflavine was first precipitated from the buffered solution by addition of an excess amount of ferricyanide. After filtration, the excess ferricyanide in the filtrate was determined by using hydroiodic acid, zinc sulfate, and sodium thiosulfate. An alternative method involved precipitation of the acridines as picrate salts, followed by back titration of the filtrate with methyleneblue. More recent methods have been based upon paper [2] and thin layer chromatography [3, 4], and capillary zone electrophoresis [5],

Proflavine and acriflavine appeared in the 1900's. Proflavine (Scheme 1.1) was manufactured by the condensation of m-pheneylenediamine and formic acid. Highly acidic proflavine hydrogen sulfate was introduced to the British Pharmacopoeia in 1932, and was replaced by the neutral sulfate (proflavine hemisulfate) after 1948. Acriflavine was listed in the British Pharmacopoeia in 1932. Both proflavine and acriflavine found early use as topical antiseptic agents for battle field wounds during World War I. It was found later that acriflavine was too caustic for treatment of human wounds, and that the neutral proflavine bore the same antibacterial activity. Therefore, acriflavine was dropped from the British Pharmacopoeia (1940) and replaced by proflavine for the treatment of wounds during World W ar II [1].

Proflavine and acriflavine are active against almost all the Gram-positive and Gram-negative organisms that commonly infect wounds. The mechanisms of action of proflavine and acriflavine were believed to involve the following;

(a) Competition between their cations and hydrogen ions for a vitally important anion in bacteria.

(b) Their cations bind with the vital anion, which makes the anion unusable.

4 Because of their toxicity, the use of proflavine and acriflavine in humans was discontinued. Toxicities observed include: tissue necrosis, anaphlactoid reactions, destruction of blood cells, depression of the central nervous system, cardiac arrest, and irreversible muscle paralysis. Autopsies of patients who died while receiving acriflavine therapy showed: liver atrophy, ulcerated duodenum, and acute fatty degeneration of the kidneys. It was subsequently discovered that proflavine and acriflavine bind to DNA and thus cause gene mutations. The i.v. lethal doses (rabbits) of proflavine and acriflavine are 11.1 and 7.3 mg per kilogram of body weight, respectively [6]. The subcutaneous LD 50 (mice) is 0.14 and 0.014 gm per kilogram of body weight for proflavine and acriflavine, respectively [1].

Extensive studies have been performed on the binding of proflavine and acriflavine with DNA. It is believed that two binding processes are involved. Type One is a first order, high affinity reaction that reaches equilibrium at one acridine molecule per 4 or 5 nucleotides. The acridine molecule intercalates between two layers of base pairs via hydrogen bond formation. Type Two is characterized by a high ratio of ligand to DNA. It is a low-affinity, high-capacity process that leads to the fixation of one acridine molecule per nucleotide. Proflavine and acriflavine preferably bind with the adenine-thymine base pair rather than with the guanine-cytosine pair [1].

Very few studies have been reported that deal with metabolism of the acridines. It was proposed that acridines might be oxidized by acridine dehydrogenase, which is present in liver of many mammals; however, amino

5 acridines or 9-substituted acridines were not expected to undergo this transformation [1]. No studies were found that dealt with metabolism of amino acridines.

1.2. PROBLEM STATEMENT

Proflavine and acriflavine have potential for use in aquaculture, as an antibacterial for the treatment of fish infection. Because consumption of proflavine and acriflavine treated fish is a potential hazard for humans, knowledge of their disposition in fish is essential to guide their effective use while avoiding subsequent exposure of consumers. As a consequence, the study of the disposition and total residue depletion of proflavine and acriflavine became our main objective. In order to accomplish this objective, supporting studies were needed. A sensitive and specific analytical method for the analysis proflavine, acriflavine and their metabolites in fish tissues and fluids was developed. This involved development of extraction methods. Moreover, to better understand the results of the disposition and total residue depletion studies, characterization of the pharmacokinetics and metabolism of P and A were desirable. The dissertation project became a study of the quantitative analysis, metabolism, pharmacokinetics, and residue distribution of acriflavine and proflavine in fish. 1.3. ORGANIZATION OF DISSERTATION

This dissertation consists of five chapters. Chapter 1 is the general introduction. The main body is in Chapter 2 to Chapter 5. In Chapter 2, a specific and sensitive reverse phase HPLC method for the determination of proflavine, acriflavine and their metabolites in rainbow trout tissues and fluids is described. Pharmacokinetic studies are described in Chapter 3. Chapter 4 consists of the disposition and total residue depletion study. In Chapter 5, the metabolites found in proflavine and acriflavine treated fish were investigated and characterized.

1.4. REFERENCES

1. Albert A. The acridines. St. Martins press, New York 1966; 295-310.

2. Fisher, A. Test of some antiseptic acridine dyes. Fresenius' Z. Anal

Chem 1975; 276(5):383-4.

3. Giebelmann R, Nagel s, Brunstein C, Scheibe E. Thin-layer

chromatographic detection of quaternary ammonium compounds of

toxicological-chemical relevance. Zentralbl Pharm, Pharmakother

Laboratoriumsdiagn 1976; 115(4):339-46.

4. Zhang G, Li G, Li Z. Determination of acriflavine by thin-layer

chromatographic densitometry. Yiyao Gongye 1986; 17(9):422-3. 5. Altria KD, Simpson CF. Analysis of some pharmaceuticals by high

voltage capillary zone electrophoresis. J. Pharm. Biomed. Anal. 1988;

6 (6 -8 ): 801-7.

6 . Oettingen WF. The therapeutic agents of quinoline groups 1933. The

Chemical Catalog Company, Inc., New York; 239-261.

8 CHAPTER 2

HPLC ANALYSIS OF PROFLAVINE. ACRIFLAVINE, AND THEIR

METABOLITES IN RAINBOW TROUT

2.1. ABSTRACT

Acriflavine (10-methyl-3,6 -diamino acridine, A) and proflavine {3,6- diamino acridine, P) have potential for use in aquaculture as antiinfectives. A sensitive and specific method was developed and validated for the determination of A, P, and their metabolites in fish fluids and tissues. A solid- phase extraction procedure was used for sample clean-up, and a reverse phase HPLC method with UV detection and a gradient mobile phase was developed for separation and quantification. The method allowed quantitative determination of P and A at 0.1 - 0.5 nmole/ml levels. The inter- and intraday variation was less than 5%, and the extraction recovery was above 90% for fluids, and above 58% for all the tissues. The assay method is suitable for studies of proflavine and acriflavine disposition, pharmacokinetics, and metabolism. 2.2. INTRODUCTION

A sensitive and specific method is important for the study of the disposition and total residue depletion, the pharmacokinetics, and the metabolic profiles of proflavine and acriflavine. The analytical methodology reported in the literature includes: paper [1] and thin layer chromatography [2, 3], and capillary zone electrophoresis [4]. These methods are not suitable because they are not specific, sensitive, or simplistic enough for these proposed studies.

Furthermore, no sample clean-up methods have been reported. Since the residues present in fish tissues and fluids might be fairly low after water exposure to proflavine or acriflavine, a very sensitive method was needed. A reverse isotope dilution analysis technique using ^^C-ring labeled proflavine or acriflavine, which is known one of the most sensitive detection technique, was considered for the disposition and residue depletion study. To determine the specific activity of proflavine or acriflavine after dilution, an HPLC method for their separation from radiolabeled metabolites was needed. Moreover, the physico-chemical nature of proflavine and acriflavine, and their potential polar metabolites made a reverse phase HPLC method very feasiable. However, little was known about their HPLC and extraction from tissues. HPLC requres that the analytes are isolated from the biological matrix and clean enough for

HPLC columns. Since proflavine and acriflavine are relatively hydrophilic, a

10 solid phase extraction method was proposed for sample clean-up. The binding of proflavine and acriflavine with DNA made low extraction recovery a potential problem. This problem was overcome by using a competitive salt effect [5].

This chapter describes a simple solid phase extraction as well as a sensitive reverse-phase HPLC method for the determination of proflavine, acriflavine, and their metabolites in rainbow trout.

2.3. EXPERIMENTAL

2.3.1. Reagents and Materials

Proflavine was purchased from Janssen Chimica (New Brunswick, NJ).

Acriflavine was purchased as a mixture of acriflavine (55%) and proflavine

(45%) from Pfaltz & Bauer, Inc. (Waterbury, CT), which was then purified [6 ];

HPLC, NMR, and elemental analysis showed an acriflavine purity >98%.

HPLC-grade acetonitrile, methanol and sodium chloride were obtained from

Fisher Scientific Company (Fair Lawn, NJ). Acetic acid and hydrochloric acid were purchased from EM Science (Gribbstown, NJ). Triethylamine was purchased from Aldrich Chemical Co. (Milwaukee, Wl). Sodium acetate was

11 obtained from J. T. Baker Chemical Company (Phillipsburg, NJ). LC-cyano

(LC-CN SPE) solid phase extraction tubes were purchased from Supelco

(Bellefonte, PA).

2.3.2. Instruments

The HPLC system (Beckman Instruments, Irvine, CA) was a Model 125 programmable binary gradient pump, a model 166 UV-VIS detector set at 262 nm, an injection valve supplied with a 100 pi sample loop, and an IBM model

55SX PC computer with a System Gold control system. The Omni-Mixer and

Micro-Homogenizer were obtained from Omni International (Waterburry, CT).

The Visiprep Vacuum Manifold was obtained from Supelco Inc. (Bellefonte,

PA), the centrifugal evaporator (Model SVC 200H) was obtained from Savant

Instruments, Inc. (Farmingdale, NY), and the refrigerated centrifuge (Model PR-

7000) was obtained from International Equipment Company (Needham Hts.,

MA).

12 2.3.3. HPLC Conditions

Samples were Injected onto a 4 um, 3.9 mm x 300 mm Nova-Pak C-18 column (Waters, Mllllpore Corporation, Milford, MA). The gradient mobile phase consisted of a mixture of solvent A (95 v/v% buffer, 5 v/v% acetonitrile) and solvent B (5 v/v% buffer, 95 v/v% acetonitrile). The buffer was water, acetic acid, and triethylamine (1000:15:4). The mobile phase gradient program was 100% A for 5 mln, and then over 13 mln, the mobile phase was linearly changed to 75% A and 25% 8 . At 23 mln, the mobile phase was returned to Its

Initial composition over 2 mln, and 4 mln was allowed for equilibration. The flow rate was 0.8 ml/mln. The assay was carried out at room temperature.

2.3.4. Sample Clean-up

Plasma: Proflavine or acriflavine standard was added to pooled clean plasma of rainbow trout. The mixture was vortexed and loaded onto a conditioned SPE cyanopropyl-bonded silica tube (Supelco LC-CN, 1 ml tube).

The tube was rinsed with 1 ml of water and allowed to run dry. The sample was eluted with a 500 pi solution of 1 % acetic add In methanol. The methanol was evaporated under a stream of nitrogen. The residue was dissolved In a

13 2 0 0 ni solution of 1 % acetic acid in methanol and the solution was filtered through a 22 n nylon filter (Alltech). The filtrate was combined with 200 nl of water and the solution was injected onto the HPLC.

Bile: Proflavine or acriflavine standard was added to an equal volume mixture of a 1% solution of acetic acid in methanol and trout bile. After being vortexed, the mixture was filtered though a nylon filter and injected onto the

HPLC.

Liver, muscle, kidney and skin samples: Samples of 0.3 to 1 gram were homogenized in 4 ml 0.1 M pH 4.4 sodium acetate buffer that contained 1 M

NaCI, and proflavine standard solution was added followed by 6 ml of methanol. The samples were shaken for 15 min. and centrifuged for 15 min. at

5000 rpm in an I EC Model PR-7000 centrifuge at 12°C. The supernatant of each sample was then transferred to a polypropylene tube, and the methanol was evaporated using a vacuum centrifugal evaporator. The sample was loaded onto an SPE tube and eluted with 2 ml of methanol (containing 1% acetic acid). The eluate was evaporated under nitrogen and the residue was dissolved in 200 |al of acidified methanol and filtered. The filtrate was diluted with 200 p.1 of water before it was injected onto the HPLC.

14 2.3.5. Metabolite Isolation

Trout were anesthetized with tricaine methanesulfonate (0.1 g/liter) and then were fitted with a cannula (28-G thin wall Teflon tubing (Zeus Industrial

Products, Rariton, NJ) in the dorsal aorta [7, 8 ], which allowed intraarterial injection and blood sampling in freely swimming fish. After 24 hr of recuperation, each fish received 1 0 mg/kg of proflavine or 6 mg/kg of acriflavine, that was administered lA as a 10 mg/ml solution for proflavine, 6 mg/ml solution for acriflavine in 80% v/v DMSO in water. Sixteen hours after dosing, blood was collected in a heparinized Ependorf tube, centrifuged, and the plasma portion was separated and stored in a - 2 0 °C freezer until analyzed by HPLC. Immediately after blood withdrawal, the fish was sacrificed, and bile and liver samples were collected and stored at -20°C. Sample clean-up procedures were described previously. Metabolites were separated by HPLC, the mobile phase component was evaporated either under a stream of nitrogen or in a vacuum centrifuge evaporator, and the purified metabolites were stored in a -20°C freezer.

15 2.3.6. Preparation of Standards

Stock solutions of proflavine and acriflavine were prepared in double distilled water. These solutions were stable for three months under refrigeration at 4°C. Working standard solutions were prepared on the day of analysis by dilution of the stock solution with double distilled water.

2.3.7. Validation

For separation of proflavine and its metabolites, liver samples from proflavine treated fish (via lA injection) were extracted and injected onto the

HPLC. Various mobile phase conditions were explored until base line separations of proflavine and its three metabolites (Met I, II, and III) were achieved. The same experimental procedures were performed to separate acriflavine and its metabolites. The developed HPLC conditions were also suitable for the separation of proflavine and acriflavine from different tissues and fluids. Plasma, bile, skin, muscle, kidney, and liver samples were spiked with a mixture of proflavine and acriflavine and were extracted. Blank tissue samples were also extracted and injected onto the HPLC using the same procedures to identify possible endogenous interference.

16 The extraction recovery of proflavine in plasma, bile, liver, kidney, muscle and skin was studied by addition of known amounts of proflavine to the tissue and fluid samples. Sample extraction procedures were the same as described in the Sample Clean-up Section. Acriflavine was used as an internal standard and was added before injection onto the HPLC. Recovery was calculated by comparison of the peak area ratio of proflavine to acriflavine with the nonextracted peak area ratio (Table 1).

Assay linearity was examined by preparation of proflavine and acriflavine standards in fish plasma over a concentration range of 1 to 1 0 0 nmole/ml.

Intraday accuracy and precision were determined by spiking pooled fish plasma with two concentrations of proflavine and acriflavine. Interday accuracy and precision were assessed from calibration curves that were prepared on five different days.

2.4. RESULTS AND DISCUSSION

Selective extraction and separation of proflavine and acriflavine were achieved from tissue and fluid samples by using reverse phase HPLC with a gradient mobile phase (Fig. 2.1-2.6). Satisfactory separation of proflavine, acriflavine and their metabolites extracted from plasma and liver samples was

17 achieved using the same HPLC conditions (Fig. 2.7-2.10). The apparent metabolites (Met I, II, and III for proflavine, Met I’, 11’, and III’ for acriflavine) were well separated and interfering substances from blank liver homogenate did not co-elute with the metabolites. Chromatographic separation was completed within 25 min. Chromatograms of blank extracts showed there were no endogenous interference peaks.

The standard curve was linear from 5 to 50 nmole/ml. Since metabolite standards were not available, their recovery was not measured. The recovery of proflavine was measured using acriflavine as an internal standard (Table 1).

The strong binding of proflavine and acriflavine to DNA was minimized by using a competitive salt effect [5]. The extraction buffer contained 1 M NaCI in 0.1 M acetate buffer (pH 4.4). NaCI improved the recovery due to competition between Na"" cations and the acridine molecules for the same binding sites on

DNA; a reasonable recovery was thereby obtained. Recovery exceeded 90% for plasma and bile, and was in the range of 50% to 80% for tissues. The intra- and interday coefficients of variation were less than 5% and 3%, respectively

(Tables 2 and 3).

18 2.5. CONCLUSION

A sensitive and specific HPLC method was developed for the separation and quantification of proflavine, acriflavine and their metabolites. Solid phase extraction was used to extract the analytes from trout tissues and fluids. The method proved to be sufficiently specific, accurate, sensitive and precise for use in pharmacokinetic and disposition studies of proflavine and acriflavine.

19 1.1

0.9

0.7 D <

c8 0.5 (0 f o (0 S i < 0.3

0.1

- 0.1 0 5 10 15 20 25

Time, mln

Figure 2.1. Representative HPLC chromatogram of fish plasma spiked with proflavine (P) and acriflavine ((A).

20 0.6 -,

g 0.4.

S c (0 -2 o en S < 0.2 -

0.0

- 0.1 0 5 10 15 20 25 Time, mln

Figure 2.2. Representative HPLC chromatogram of fish bile spiked with proflavine (P) and acriflavine (A).

21 2.0 -I

8c I

0.5 -

0.0 0 5 10 15 20 25 Time, min

Figure 2.3. Representative HPLC chromatogram of fish liver spiked with proflavine (P) and acriflavine (A).

22 2.0 1

< 8 c -eCD

0.8 -

0.5 -

0.2 -

- 0.1 0 5 10 15 20 25

Time, mln

Figure 2.4. Representative HPLC chromatogram of fish muscle spiked with proflavine (P) and acriflavine (A).

23 < 0.8 - 8 ■e

0.2 -

- 0.1 0 5 10 15 20 25 Time, mln

Figure 2.5. Representative HPLC chromatogram of fish kidney spiked with proflavine (P) and acriflavine (A).

24 0.5 -,

0.4 -

3 0.3 -

c8 (0 e

0.0 -

- 0.1 0 5 10 15 20 25 Time, min

Figure 2.6. Representative HPLC chromatogram of fish skin spiked with proflavine (P) and acriflavine (A).

25 1 .3 1

Met I 0.9-

< 0.7-

c8 •S 0.5 - S < 0.3 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 2.7. Representative HPLC chromatogram of proflavine (P) and its metabolite Met I extracted from plasma.

26 2.2 -,

Met II

3 < Met c8 CO -2 S <

0.6 -

Met

0.2 -

- 0.2 0 5 10 15 20 25 30

Retention Time, min

Figure 2.8. Representative HPLC chromatogram of proflavine (P) and its three metabolites, Met I, Met II, Met III extracted from liver.

27 0.9 1

0.7 -

Z) 0.5 - M etr

c8 CO £ S 0.3- <

- 0.1 0 5 10 15 20 25 30

Retention Time, min

Figure 2.9. Representative HPLC chromatogram of acriflavine (A) and its two metabolites, Met I' and Met II' extracted from plasma.

28 1.9 T

1.5

o 0.7-- < Met II' 0.3 --

- 0.1 0 5 10 15 20 25 Time, min

Figure 2.10. Representative HPLC chromatogram of acriflavine (A) and its three metabolites, Met I’, Met 11’, Met III’ extracted from liver.

29 Concentration Recovery(%) CV%

(nmole/ml) (mean ± SD. n=5)

Plasma 1 0 91.73 ±2.56 2.79

40 94.99 ±2.70 2.84

Bile 1 0 98.37 ±2.17 2 . 2 0

40 96.47 ±1.94 2 . 0 1

Muscle 1 0 75.15 ±1.88 2.50

40 79.15 ±4.00 5.06

Skin 1 0 70.21 ± 8.43 1 2 . 0

40 70.87±6.37 8.99

Liver 1 0 62.42 ±7.53 1 2 .1

40 64.48 ±2.64 4.09

Kidney 1 0 60.54 ±3.75 6.19

40 58.65 ±6.36 1 0 . 8

Table 2.1. Extraction recovery of proflavine in trout tissues and fluids

30 Concentration (nmole/ml) cv

Added Found (%)

(mean ± SD, n=5)

Proflavine 1 0 1.00 ±0.05 4.54

40 4.13 ±0.07 1.67

Acriflavine 1 0 0.99 ± 0.04 4.21

40 4.15 ±0.07 1.59

Table 2.2. Intraday coefficient of variation in the determination of proflavine and acriflavine in plasma

31 Date slope r range nmole/ml

Proflavine 6/24/95 49.219 0.9968 5 to 50 6/29/95 49.755 0.9995 5 to 50 7/4/95 49.321 0.9945 5 to 50 7/12/95 51.034 0.9969 5 to 50 7/13/95 48.917 0.9940 5 to 50

Average 49.649 SD 0.830 CV% 1.673

Acriflavine 6/24/95 51.249 0.9964 5 to 50 6/29/95 50.347 0.9997 5 to 50 7/4/95 51.900 0.9972 5 to 50 7/12/95 53.203 0.9967 5 to 50 7/13/95 50.753 0.9937 5 to 50

Average 51.590 SD 1.119 CV% 2.174

Table 2.3; Interday coefficients of variation in the determination of proflavine and acriflavine in plasma

32 2.6 REFERENCES:

1. Fisher, A. Test of some antiseptic acridine dyes. Fresenius' Z. Anal

Chem 1975; 276(5):383-4.

2. Giebelmann R, Nagel s, Brunstein C, Scheibe E. Thin-layer

chromatographic detection of quaternary ammonium compounds of

toxicological-chemical relevance. Zentralbl Pharm, Pharmakother

Laboratoriumsdiagn 1978; 115(4):339-46.

3. Zhang G, Li G, Li Z. Determination of acriflavine by thin-layer

chromatographic densitometry. Yiyao Gongye 1986; 17(9):422-3.

4. Altria KD, Simpson OF. Analysis of some pharmaceuticals by high

voltage capillary zone electrophoresis. J. Pharm. Biomed. Anal. 1988;

6 (6 -8 ): 801-7.

5. Schelhom T, Kretz S, Zimmermann HW. Reinvestigation of the binding of

proflavine to DNA. Is intercalation the dominant binding effect? Cell and

Mol Biol 1992. 38(4):345-365.

33 6 . Albert A. The acridines. St. Martins press, New York 1966; 346-7.

7. Hoeiton GF, Randall DJ. The effect of hypoxia upon the partial pressure

of gases in blood and water afferent and efferent to the gills of rainbow

trout. J Exp Biol 1967; 46: 317-22.

8 . A. R. Houston: Blood and circulation. In “Methods for Fish Biology” (C.

B. Schreck and P. B. Moyle, ed.), pp. 273-322, American Fisheries

Society, Bethesda, Maryland, 1990.

34 CHAPTER 3

PHARMACOKINETICS OF PROFLAVINE AND ACRIFLAVINE

IN RAINBOW TROUT

3.1. ABSTRACT

The plasma concentration-time profiles of ^'^C-proflavine (P) and ^'^C- acriflavine (A) were determined by reversed-phase HPLC with reverse-isotope dilution analysis after lA administration. Plasma protein binding, determined by equilibrium dialysis, was 80.9% for P and 89.0% for A. The pharmacokinetic behaviors of P and A were similar; their plasma concentration-time profiles appeared triexponential, and the terminal half-lives were 25.9 hr for P and 38.3 hr for A. A much longer half-life for a small fraction of the administered dose was possible, which might have resulted from entero-hepatic circulation. The apparent steady-state volume of distribution (Vss) and the total body clearance

(CIb) were 28.2 L/kg and 1.05 L/hr/kg for P, and 31.2 L/kg and 1.08 L/hr/kg for A, respectively. The very large Vss values indicated extensive distribution of both compounds into tissues, which may have resulted from their avid binding to

35 DNA. Since P and A are ionized at physiological pH, their gill epithelium permeability is low and their elimination is via metabolic and renal pathways. A large volume of distribution contributed to their long half life.

3.2. INTRODUCTION

Pharmacokinetic models are important tools for the characterization of the bioaccumulation of organic chemicals in aquatic species, by separately characterizing the component process of uptake, distribution, metabolism and excretion [1-5]. Pharmacokinetic parameters are useful for comparison of chemicals, for the interpretation of disposition and depletion studies, and for the prediction of the fate of chemicals.

Currently, little pharmacokinetic information is available for P and A.

Knowledge of their pharmacokinetics may help in the characterization of their disposition in fish. Characterization of their biological half life, total body clearance and steady state volume of distribution would be useful for the determination of the withdrawal time that would avoid P and A residue in fish intended for human consumption. This chapter describes the pharmacokinetics of proflavine and acriflavine in rainbow trout after intraarterial administration.

36 3.3. EXPERIMENTAL

3.3.1. Reagents and Materials

P was purchased from Janssen Chlmica (New Brunswick, NJ). A was purchased as a mixture of A (55%) and P (45%) from Pfaltz & Bauer, Inc.

(Waterbury, CT), which was then purified. HPLC-grade acetonitrile, methanol, sodium chloride and scintillation cocktail (ScintiVerse) were obtained from Fisher

Scientific Company (Fair Lawn, NJ). Acetic acid and hydrochloric acid were purchased from EM Science (Gribbstown, NJ). Triethylamine was purchased from Aldrich Chemical Co. (Milwaukee, Wl). Sodium acetate was obtained from

J. T. Baker Chemical Company (Phillipsburg, NJ). LC-cyano solid phase extraction tubes were purchased from Supelco (Bellefonte, PA). Tricaine methanesulfonate (TMS) was obtained from Crescent Research Chemicals

(Phenix, AZ). Dimethyl sulfoxide (DMSO) was obtained from Aldrich Chemical

Company (Milwaukee, Wl).

’"‘C-Proflavine hemisulfate (C-| 3 H1 1 N3 1 /2 [H2 S 0 4 ]), was prepared by

DuPont New England Nuclear; 23.3 mCi/mmol; molecular weight, 258.29; the specific activity was 90.2 microcuries per mg or 200,300 dpm per microgram.

Unlabeled proflavine was from Janssen Chemica, Geel, Belgium; lot no.

61514/1. ^'*C-Acriflavine hydrochloride (C 14H16N3CI3 , 9-^"‘C-3,6-diamino-10-

37 methylacridinium chloride); 17.43 mCl/mmol; molecular weight, 332.66; the vendor labeled specific activity was 52.4 microcuries per mg or 116,300 dpm per microgram. The -labeled material (lot no. AT-107) was prepared by The

Bionetics Corp., National Center for Toxicological Research, Jefferson, AR

72079. Unlabeled acriflavine was purchased from Pfaltz and Bauer as the commercially available mixture of acriflavine (10-methyl-3,6-diaminoacridine) and proflavine (3,6-diaminoacridine). Acriflavine was isolated as described in

Chapter II. After recrystalization of the material, twice from water, the NMR spectrum and elemental analysis proved that the compound was acriflavine chloride: Carbon, Hydrogen, Nitrogen and Chlorine accounted for 94.5% of the total weight, and moisture was 1.15%. The rest was most likely 1 / 2 mole HgO, in the form of crystal water (3.36%) and inorganic material (0.99%). The HPLC chromatogram showed no peak other than acriflavine. When this study began, two minor peaks appeared in the chromatogram, which accounted for about 6 % of the total UV absorption. Therefore the acriflavine content of the unlabeled acriflavine was taken as 89%.

3.3.2. Instruments

Living Stream Aquarium was obtained from Frigid Units Inc. (Toledo, OH).

For other instruments, refer to Chapter II.

38 3.3.3. Animal protocol

Rainbow trout {Oncorhynchus mykiss), that weighed of 300 ~ 500 grams, were transported in aerated water within an hour from Fresh Water Fish Farms

(Urbana, Ohio). Trout were housed at 12 °C for a minimum of two weeks in a

1000 liter fiberglass aquarium. The water (Dechlorinated Columbus Municipal water) was aerated and was recirculated through a charcoal filter with an average recirculation time of 1.5 min. W ater was added continuously to the holding aquarium at a rate of 1 .5 liters/min. The aquarium was cleaned daily and water temperature, pH, and dissolved oxygen were measured daily; the acceptable ranges for these parameters were 11-13 °C, 6.5-8.0, and >9.6 mg/liter (> 80% saturation), respectively. The daily monitored values were within these ranges (Table 3.1). The vivarium was windowless and had an automatically timed 07:00 to 19:00 hr photoperiod. Fish were fed a commercially available (Rangen, Inc., Buhl, ID) soft, moist diet with a pellet size of 5/32". Fish were not fed on the day preceding their experimental use. Fish that received food during the pharmacokinetic study were fed 1 % of their body weight once every other day.

39 Parameter Analysis

Hardness (as CaCOa), mg/L 1 0

Nitrate - Nitrite, N mg/L 0.75

Phosphorus Total, P mg/L 0.16

Lead Total, Pb pg/L < 3

Mercury Total, Hg pg/L <0.5 pH 7.1

Dissolved Oxygen (% saturation) > 90%

Table 3.1 : Water quality parameters for the holding aquarium.

40 3.3.4 Fish Cannulatlon:

A rainbow trout fasted for 24 hours was transferred to a polyethylene cylinder that contained 8 liters of aerated 0 . 1 g/liter tricaine methanesulfonate solution for anesthesia. Temperature was maintained using ice that was placed outside the cylinder. The anesthetized fish was transferred ventral side up to a brace that was lined with a moistened paper towel, and a submersible pump was used to irrigate the gills with aerated TMS solution. A hole was drilled in the snout using a Dremel motor tool. Two sutures were placed on the dorsal surface of the oral cavity: one approximately midway between the snout and leading edge of the first gill arch and the second immediately posterior to the hole drilled into the snout. A third suture was placed on the dorsal side of the face beside the hole. A Teflon cannula (28-G thin wall Teflon tubing, Zeus Industrial

Products, Raritan, NJ) was attached to a 5/8” x 25 gauge needle and a 1 ml syringe filled with heparinized non-nutrient teleost saline [ 6 ]. This cannula was then inserted into the dorsal aorta [7,8] using a 2”, 18 gauge IV catheter. Prior to the insertion, the cannula tubing was filled with saline. The catheter was slowly inserted at the midline between the second and the third gill arches at approximately a 30° angle, until blood appeared in the tube. The center poke was removed and the tubing was threaded into the catheter and 7 cm into the artery. The catheter was removed, taking care not to dislodge the cannula. The cannula tubing was disconnected from the needle and threaded though the hole

41 drilled In the snout and then re-connected to the needle. The cannula was flushed with 0.5 ml of heparinized saline, and tied in place with the three sutures.

This cannula allowed lA bolus injection and blood sampling in the free-swimming fish. The fish was immediately transferred to a clean, anesthetic free, working aquarium that held 300 liters of water with a flow-through rate of 500 ml/min.

The fish was gently moved back and forth so that the gills would be irrigated with oxygenated water until operculum movement increased. The fish was allowed to recuperate for 24 hours before experiment (Figure 3.1).

42 12^0

Figure 3.1. The diagram of a cannulated fish: the cannula allowed lA bolus injection and blood sampling from free-swimming fish. The aquarium water was kept at 12 °C and was aerated with an air stone. The aquarium water was flow through with a flow rate of 500 ml/min (not shown).

43 3.3.5. Dosing

14C Labeled P or A was Injected Into the trout aorta, and reverse Isotope dilution analysis was used to quantify P and A In plasma samples. Because of their limited aqueous solubility, the dosing solution contained 80% (v/v) DMSO In water. Each trout received a 200 jil (2 mg/kg) bolus dose of P or A. The amount of each dose was accurately determined by LSC counting and was calculated

(Equation 3.1) from the specific activity:

DPMs * Vdose Dose (mg/kg) = ------(3.1) Vs * So * Wt

DPMs: Radioactivity (dpm) of the aliquot dosing solution

Vdose: Volume of the dosing solution

Vg: Volume of the aliquot for liquid scintillation counting

So: Specific activity, dpm/mg

Wt: Weight of fish, kg

Immediately after Injection, a new syringe filled with heparinized saline was attached and the cannula was rinsed first with the saline, and then three times by pulling blood Into the cannula and pushing It back Into the aorta.

44 Finally, the cannula was flushed by infusing 1 ml of heparinized saline. A control experiment showed that no residual drug remained in the cannula after dosing.

3.3.6. Sampling times

After dosing, serial blood samples were removed at various times (5, 10,

30 min, 1, 3, 4, 6 , 12, 24, 48, 72, 96, 120, 144, 168, 216, 264, 312, 384, 456, and 530 hr), which were selected based on a pilot study. The volume of blood removed was 20 to 200 |il at each time point in the first 10 days, 200 to 400 pi in the later days, and 2 ml at the last sampling time point. The total volume of blood removed did not exceed 15% [9] of estimated total blood volume of trout

(4% of body weight) [10] in the first ten days of each study. Thereafter, blood samples were removed once every two to three days to allow recovery from further blood removal. Each blood sample was collected in a heparinized

Ependorf tube. Plasma was collected after centrifugation and stored in a -20°C freezer until analyzed.

3.3.7. Sample preparation and analysis

Reverse isotope dilution analysis was used for the analysis of all plasma samples. Each plasma sample was weighed accurately on an electronic

45 balance. To each weighed plasma sample was added 5 nmole of proflavine or acriflavine cold spike, and the solution was vortexed. The sample was then loaded onto a conditioned 1 ml LC-CN SPE tube. The tube was rinsed with 1 ml of water and was allowed to drain. The sample was then eluted with 500 pi of a solution of 1 % acetic acid in methanol. The methanol was evaporated under a stream of nitrogen. To the residue was added 200 pi of a solution of 1 % acetic acid in methanol, and the mixture was filtered through a 2 2 pm pore size nylon filter (Alltech Associate, Inc., Deerfield, IL). The filtrate was mixed with 200 pi of water prior to injection onto the HPLC. One-minute fractions were collected from the HPLC. Each fraction was combined with 5 ml of scintillation cocktail and radioactivity was quantified by liquid scintintillation spectrometry (Beckman LSC

6000). The amount of proflavine or acriflavine in each plasma sample was calculated as follows:

W, Wx.p= ------(3.2) ((So/Sx)-1)*Ws

Wx, p: the amount of proflavine or acriflavine in per gram of sample, nmole/g Wfi the number of moles in the cold spike, nmole. Ws: the weight of plasma sample, g. So: the specific activity of proflavine or acriflavine used in the study, dpm/nmole.

46 Sx: the specific activity of proflavine or acriflavine after dilution with cold spike, dpm/nmole. It was determined by the

following calculation:

Sx — Adpm/Acompound

Acompound: the concentration of proflavine in the sample after the dilution with cold spike, as determined from the area under the compound peak on the HPLC chromatogram. Adpm: the amount of radioactivity in the proflavine or acriflavine peak that eluted from the HPLC.

3.3.8. Sample stability

This study characterized the stability of analytes in plasma samples that were stored frozen prior to their analysis. Two rainbow trout (one for each compound) received 1 mg/kg doses of ^'^C-proflavine or ^'^C-acriflavine. After 6 hr, as much blood as possible was removed, heparinized, and centrifuged. The plasma was divided into 24 Ependorf tubes, with 1 00 pi per tube, and the tubes were stored at -20°C. Total radioactivity and reverse isotope dilution HPLC analysis were performed on samples stored for various periods of time up to 15 days.

47 3.3.9. Plasma protein binding study

The binding of proflavine and acriflavine to plasma protein was measured

using an equilibrium dialysis apparatus (5-cell Equilibrium Dialyzer, Spectrum

Medical Industries, Inc., Los Angeles, CA) with regenerated cellulose

membranes with molecular weight cut off of 12-14,000 daltons (Spectrum

Medical Industries, Inc., Los Angeles, CA). Fresh trout plasma was pooled from several trout of mixed gender. Plasma that contained ’‘^C-proflavine or ^'^0- acriflavine and non-nutrient ringers solution [ 6 ] as dialysate were placed in Teflon cells that had 1 ml chambers. The radiochemical purity of labeled proflavine and acriflavine used in this study was higher than 98%. Moreover, preliminary studies showed that there was no decomposition of proflavine or acriflavine detected after this period of incubation. Therefore, radioactivity was used as the measure of the concentration of analytes present in each compartment.

Preliminary studies also showed that equilibrium was reached before 6 hours.

The cells were rotated at 10 rpm in a 12°C water bath for 6 hr. Subsequently, the final volume of each compartment was converted from its weight, which was measured on an electronic balance (sp. gr. = Ig/ml). An aliquot from each chamber was analyzed for radioactivity using a Beckman LSC 6000 scintillation counter. Volume shift was corrected by the method of Boudinot [11]. The fraction of proflavine or acriflavine bound to plasma proteins was calculated as follows:

48 D b F b = ------( 3 .3 ) Db + Dp

Where:

Db = (DT-DF)*(VeA/i) (3.4)

So:

(DT-Dp)*(VeA/i) Fb = ------(3.5) {(DT-DF)*VeA/p} + DF

Fb: the protein-bound fraction of P or A

Db: the protein-bound plasma concentration of P or A

Dt: the total plasma concentration of P or A at equilibrium

Dp: the concentration of P or A in the buffer chamber at equilibrium

Vg: the plasma volume at equilibrium

V;: the initial plasma volume

3.3.10. Pharmacokinetic Parameter Calculations

The PCNONLIN version 4.0 program (SCI Software, Lexington, Kentucky) was used to analyze the plasma concentration-time profiles of proflavine and

49 acriflavine after their lA bolus administration. Both two and three compartment models were examined. The criteria for model selection were: the sum of weighted squared deviations of the fitted equation, the pattern of residuals, the

AlC criteria, and visual evaluation of the goodness of fit [14]. Compared with the two compartment model, the three compartment model with elimination from the central compartment (Fig. 3.2) gave a better fit for both proflavine and acriflavine.

Elimination from the central compartment and first order kinetics were assum ed for both compounds [12,13].

50 lA Bolus I -21 —1-13 2 1 3 ki2 ---1 k -31 ?10

Figure 3.2. Three compartment model of proflavine and acriflavine plasma pharmacokinetics after lA injection. Proflavine or acriflavine were administered into the central compartment ( 1 ) and they distributed reversibly into rapidly ( 2 ) and slowly (3) equilibrating peripheral compartments. The k symbols represent the intercompartmental transfer rate constants, and elimination (kio) was from the central compartment.

51 Plasma concentration profiles after intraarterial administration of proflavine or acriflavine were fitted to the following three-compartment model- based equation:

Cp = Ae-“‘ + 6 0 '“* + Ce"^

where Cp was the drug concentration at time t, A, B and 0 were the three ordinate axis zero-time intercepts of the feathered lines of proflavine or acriflavine concentration time profile, and a, p, and y were the corresponding hybrid first-order disposition rate constants. The model rate constants were determined by:

cx + P + Y = kio + ki2 + kgi + ki3 + kai

aP + Py + ay = kiakai + kioKai + kiakai + kiaKai + kiokai + kaiKai

apy= kiakaikai + kiokaikai

3.4 RESULTS AND DISCUSSION:

Proflavine, acriflavine, and their metabolites were stable in fish plasma during freezer storage. During 15 days of storage, the changes of total radioactivity were within 2%, and the changes of concentrations of proflavine,

52 acriflavine and their metabolites were within 15% of the values determined

before storage, Table 3.2. All samples were analyzed within one week of

collection. The plasma protein binding at 12°C was 80.9% bound for proflavine,

89.0% bound for acriflavine.

Figures 3.3 and 3.4 show the plasma concentration-time course of proflavine and acriflavine after their lA administration. The plasma concentrations declined triexponentially. Mean values of ti/g, C Ib , and Vdss for proflavine were 25.9 hr, 1.05 L/hr/kg, and 28.2 L/kg, respectively, and for acriflavine were 38.3 hr, 1.08 L/hr/kg, and 31.2 L/kg, respectively.

The large volume of distribution indicated extensive distribution into the tissues; probably avid binding to DNA was involved [15]. The volumes of distribution of proflavine and acriflavine were similar.

The total body clearance for both proflavine and acriflavine were similar in magnitude to the cardiac output in trout, which would be consistent with elimination via the gills. Because they are Ionized at physiological pH, a low gill permeability would be expected, and water uptake experiments bear out this expectation. If this is the case, clearance across the gill epithelium would tend to be much lower than gill blood flow and an additional pathway would be required.

The fact that polar metabolites are prominent in plasma of both proflavine and

53 acriflavine treated fish suggested that metabolism also played a role In their elimination. Moreover, very high concentrations of proflavine and acriflavine were observed In kidney, and their renal clearance may also have been an

Important elimination pathway.

The long half life of proflavine and acriflavine was due to their large volume of distribution, even though they also had relatively high plasma clearance. Both proflavine and acriflavine showed a relatively long persistence

In fish tissues.

54 Time, day 0 2 4 7 11 15

T.R.A.-P, %= 100 99.56 98.30 98.50 99.43 98.38

HPLC-P, 100 101.4 113.6 102.3 106.1 108.6

HPLC-MP, 100 97.24 112.2 107.4 108.3 113.9

T.R.A.-A, 100 100.9 99.75 100.9 100.8 99.75

HPLC-A, %e 100 99.00 106.7 103.9 106.3 102.8

HPLC-MA, %f 100 99.77 104.9 95.33 104.3 102.9 a: % remaining of proflavine analyzed by total radioactivity. b: % proflavine remaining analyzed by HPLC. c: % proflavine metabolite remaining analyzed by HPLC. d: % remaining of acriflavine analyzed by total radioactivity. e: % acriflavine remaining analyzed by HPLC. f: % acriflavine metabolite remaining analyzed by HPLC.

Table 3.2. Stability of proflavine, acriflavine and their metabolites in plasma at

-20 °C.

55 I o E c c o 0.1 2 c (D Ü oC O 0.01

0.001

0.0001 0 50 100 150 200 Time, hr

Figure 3.3. Proflavine plasma concentration-time profile in rainbow trout after

intraarterial administration of 2.0 mg/kg of ^'‘C-proflavine. This profile is from a

single representative trout (Wt. 360 g). Symbols represent the experimentally

determined concentrations and the line represents the least squares fit of the three compartment model-based triexponential equation.

56 1 0 -,

_l I o E c c o 2 § 0.01 c o O

0.001

0.0001 0 50 100 150 200 250 300

Time, hr

Figure 3.4. Acriflavine plasma concentration-time profile in rainbow trout after intraarterial administration of 2.0 mg/kg of ^'^C-acriflavine. This profile is from a single representative trout (Wt. 450 g). Symbols represent the experimentally determined concentrations and the line represents the least squares fit of the three compartment model-based triexponential equation.

57 Proflavine (P) Acriflavine (A) Parameter

Mean ± SD® Mean ± SD®

V s s , L.kg-' 28.2 ± 7.20 31.2 ± 6.57

C Ib , L hr'^kg' 1.05 ±0.329 1.08 ± 0.375

ti/2 , hr 25.9 ± 4.09 38.3 ± 5.85

F free % 19.1 ±0.421 11.0 ±0.631 a: n=5

Table 3.3. Pharmacokinetic parameters of proflavine and acriflavine in rainbow trout after intravascular administration.

58 3.5 REFERENCES:

1. McKim JM, Schmieder P, Veith G. Absorption dynamics of organic

chemical transport across trout gills as related to octanol-water partition

coefficient. Toxicol AppI Pharmacol 1985; 77:1 -10.

2. Bungay PM, Dedrick RL, Guarino AM. Pharmacokinetic modeling of the

dogfish shark (Squalus acanthias): distribution and urinary and biliary

excretion of phenol red and its glucuronide. J Pharmacok Biopharm

1976; 4: 377-88.

3. Kaka JS, Hayton, WL. Temperature and surfactant dependence of

accumulation of 4-aminopyrine and ethanol in fish. J Pharm Sci 1978; 67:

1558-63.

4. Karara AH, Hayton WL. Pharmacokinetic model for uptake and

disposition of di-2-ethylhexyl phthala in sheepshead minnow (cyprynodon

variegatus). Aquat Toxic1984; 5, 181-95.

5. Schultz IR, Hayton WL. Toxicokinetics of trifluralin in rainbow trout.

Aquat Toxicol 1993; 26:287-306.

59 6. Olson KR. Distribution of flow and plasma skimming in isolated perfused

gills of three teleosts. J Exp Biol 1984; 109:97-108.

7. Hoelton GF, Randall DJ. The effect of hypoxia upon the partial pressure

of gasses in blood and water afferent and efferent to the gills of rainbow

trout. J Exp Biol 1967; 46: 317-22.

8. Houston AR. Blood and circulation. In: Methods for fish biology 1990,

edited by Schreck OB and Moyle PB. American Fisheries Society,

Betheda, MD, 273-322.

9. Hofman R, Lommel R. Effects of repeated blood sampling on some blood

parameters in fresh water fish. J Fish Biol 1984; 24: 245 - 251.

10. Duff DW, Fitzgerald D, Kullman D, Lipke DW, Ward J, Olson KR. Blood

volume and red cell space in tissues of the rainbow trout, Salmo gairdneri.

Comp Biochem & Physio 1987; 87(2): 393 - 398.

11. Boudinot FD, Jusko WJ. Fluid shifts and other factors affecting plasma

protein binding of prednisolone by equilibrium dialysis. J Pharm Sci

1984; 73:774-780.

60 12. Wagner JG. Fundamentals of clinical pharmacokinetics, 1st ed. Drug

Intelligence Pub., Inc. Hamilton, Illinois 1975.

13. Gibaldi M , Perrier D. Pharmacokinetics, 2nd ed., Marcel and Dekker,

Inc., New York/Basel. 1982; 92-94.

14. Boxenhaum HG, Riegelman A, Elashoff RM. Statistical estimations in

pharmacokinetics. J Pharmacok Biopharm 1974; 2:123-48.

15. Peacocke AR. The interaction of acridines with nucleic acids. In:

Acridine, edited by Acheson RM 1973, Interscience Publishers, New York,

pp. 723-758.

61 CHAPTER 4

DISPOSITION OF PROFLAVINE AND ACRIFLAVINE IN RAINBOW TROUT

AFTER WATER EXPOSURE

4.1. ABSTRACT

The study of proflavine and acriflavine total residue depletion and metabolite profile was designed to characterize the concentration-time profile of total radioactivity and of unchanged test compounds and their toxicologically significant metabolites in rainbow trout after water exposure of the fish to 4 ppm

’‘‘C-proflavine and to 4 ppm ’'‘C-acriflavine, respectively. Based on the pilot and autoradiography studies, the main residue depletion and metabolite profile protocol could be constrained to an 8 day depuration period for proflavine, and

16 days for acriflavine. For proflavine, at the end of the 4 hr exposure phase, the concentration of radioactivity in all tissues and fluids was below the exposure water concentration, with the exception of bile at 4 hr where the concentration of radioactivity averaged 5.3 ppm. The rank order of residue concentration at 4 hr was bile > liver > kidney > skin > plasma > muscle. The concentration in muscle averaged 18.7 ppb or 0.47% of the exposure water concentration. After 2 days

62 depuration, the highest concentration of residue by far was in the bile at 29.5 ppm; it increased to 39.3 ppm after 4 days depuration, and then declined after 8 days depuration to 21.2 ppm. Muscle residue was the lowest after 8 days depuration at 1.94 ppb or 0.05% of the concentration in the exposure water. For acriflavine, immediately after exposure (no depuration), bile showed the highest concentration of radioactivity, followed in descending order by liver, kidney, skin, plasma and muscle. After 8 days depuration, the concentration of residue was decreased most rapidly in muscle; to 2.56% of the 0-Day value. The 0-Day value was relatively low, and the edible tissue contained after 8 days of depuration 3.56 ± 0.550 ppb acriflavine. The rate of depuration during 8 days in the various tissues and fluids was: muscle > plasma > bile > liver > skin > kidney. Between 8 and 16 days of depuration the concentration decrease of radioactivity was significant in liver and in bile, decreasing to 0.652% and

0.742%, respectively, of the 0-Depuration values. The decrease in radioactivity was also pronounced in kidney during this period, from 36.6% of the 0 depuration value to 14.3%. Skin persisted after 16 days of depuration 24.6% of the amount taken up during the 4 hours of exposure. Whole body autoradiography showed a similar distribution pattern. At 1, 2, and 4 days

(proflavine) or 2, 4, and 8 days (acriflavine), the highest levels were in the gallbladder and intestinal lumen. From this study, we concluded that proflavine and acriflavine residue retained for a long period of time in trout.

63 4.2. INTRODUCTION

The toxicities of proflavine and acriflavine led to termination of their use in humans after the 1950s. More recently, however, they have shown potential for the treatment of bacterial infection and fish lice in aquaculture. As a consequence, consumption of proflavine or acriflavine treated fish may become a potential hazard for humans. The study of P and A disposition in rainbow trout may guide the proper use of proflavine and acriflavine treated fish. The total residue depletion and disposition study consisted of three parts; pilot study, autoradiography and main study. The objectives of these studies are:

Pilot Studv. To characterize the concentration-time profile of radioactivity in fluids and tissues of rainbow trout after 4 hr. water exposure to i4c-proflavine and ’"’C-acriflavine, respectively. The results of this study will guide the development of the protocol for the study of total residue depletion and metabolic profile of proflavine and acriflavine in rainbow trout.

Autoradiocraohv. To characterize the concentration-time profile of radioactivity in fluids and tissues of rainbow trout after 4 hr. water exposure to

^"C-proflavine and ’'‘C-acriflavine, respectively. The results of this study will guide the development of the protocol for the study of total residue depletion and metabolic profile of proflavine and acriflavine in rainbow trout.

64 Main Study. To characterize the concentration-time profile of total

radioactivity and of unchanged proflavine, acriflavine, and their metabolites in

plasma, bile, muscle, skin, liver, and kidney of 0.5-1.0 kGm male and female

rainbow trout at various times during and after water exposure of the fish to 4

ppm ^4c-proflavine and "'^C-acriflavine for 4 hr, respectively.

4.3. EXPERIMENTAL

4.3.1. Reagents and Materials

Refer to Chapters 2 and 3.

4.3.2. Instruments

Cryo-Microtome, Model 450 MR, was obtained from PMV (Switzerland).

Storage phosphor screens was obtained from Molecular Dynamics (Sunnyvale,

CA).

For other instruments, refer to Chapters 2 and 3.

4.3.3. Experimental Animal

A. Animal Care

Rainbow trout {Oncorhynchus mykiss) were obtained from London Fish

Hatchery (London, Ohio); the average ± SD body weight of the fish was 601 ±

65 122 g, range 409-836g. Trout were housed with other trout at 12 °C for a minimum of two weeks in a fiberglass aquarium that contained 1000 liters of dechlorinated water. This aquarium water was continuously aerated, and filtered through a charcoal filter with an average recirculation time of 1.5 min.

Dechlorinated Columbus Municipal water was added continuously to the holding aquarium at a rate of 1.5 liters/min. The aquarium was cleaned daily and water temperature, pH, and dissolved oxygen were measured daily; the acceptable ranges for these parameters were 11 ~ 13 °C, 6.5 ~ 8.0, and > 9.6 mG/liter

(> 80% saturation), respectively. Our daily monitored values were within these ranges (Table 3.1). The vivarium was windowless and had an automatically timed photoperiod that was set at 07:00 to 19:00. Fish were fed a commercially available soft, moist food obtained from Rangen Inc. (Buhl, ID) with a pellet size of 5/32”. Fish were not fed on the day preceding their experimental use. Fish receiving food during the pharmacokinetic study were fed 1% of their body weight once every two days. The fish used had no signs of disease, e.g., lethargy, erratic swimming, skin lesions, eroded or reddish fins.

8. Randomization

Fish that met the weight criterion were chosen by the grab method. The order of administration of treatments (exposure and depuration times) was varied in a random fashion. Fish were identified with a serially numbered and color- coded tag.

6 6 c . Bias Control

For each treatment, three apparent male and three apparent female fish

were taken from the holding aquarium, tagged with a serial numbered tag, and

exposed to the 4.00 mG/L proflavine or 4.00 mG/L acriflavine solution. The tag

remained with each fish until necropsy and the tag number was entered on the

necropsy data sheet. The gender was confirmed at necropsy. Male and female

fish were exposed together to maximize the possibility for detection of gender

differences.

4.3.4. Sample collection, preparation and analysis

A. Sample collection

After fish were transferred into the exposure solution, the time was

recorded. After 4 hr in the exposure solution, the fish were removed, and

rinsed/anesthetized in a chipped ice-100 mG/L tricane methanesulfonate

solution mixture. Immediately after removal of fish from the exposure solution,

samples of exposure water were taken for determination of total radioactivity by

liquid scintillation counting and for analysis by HPLC. The clock time that fish were removed was recorded. Then blood was collected from each fish via the

dorsal aorta, through the roof of the mouth, and the fish was sacrificed with a

67 blow to the head. The blood sample was centrifuged and the plasma was

collected. Samples of bile, muscle, liver, kidney, and abdominal skin were taken

from the carcass of each fish as quickly as possible after it was removed from

the exposure water or depuration aquarium and killed. The fish weight was

recorded and the gender was determined and recorded after the abdominal

cavity was opened and the gonads inspected.

B. Sample storage

Samples for the metabolite profile study were weighed, rapidly frozen in

liquid nitrogen and placed into individual, labeled storage vials. These vials

were stored in a freezer at -20°C. A second identical set of tissue and fluid

samples were prepared and stored for use in case of assay problems with

particular samples. The weight of each sample was recorded.

C. HPLC analysis

Samples were prepared as described in Chapter 2. During

homogenization, 5 to 50 nmoles of a proflavine cold spike was added to each

sample. For HPLC, each one minute fraction was collected and mixed with 4.5

mL of scintillation cocktail. These samples were analyzed for radioactivity using

a scintillation counter (Beckman LSC 6000). The amount of proflavine or acriflavine in each sample was calculated as described in Section 3.2.7. The amount of proflavine or acriflavine metabolites was calculated as follows;

68 VVx, p Adpm, m Wx.m = ------(4.1) Adpm, p

Adpm. m: the amount of radioactivity in the metabolite peak eluting

from the HPLC.

Adpm. p: the amount of radioactivity in the parent compound peak

eluting from the HPLC.

Wx, m: the amount of proflavine metabolite per gram of sample,

nmole/g

D. Determination of total radioactivity by digestion analysis

White muscle, skin, liver, and kidnev. The tissues were cut into small pieces and 0.2 - 1.2 g samples were weighed and added to polyethylene scintillation vials. The samples were either used promptly or stored at -20°C.

For solublization, 2 - 3 mL of tissue solubilizer (Solvable, ) was added to the fresh or thawed samples, which were incubated on a horizontal shaker. After 5-

6 hours of shaking, 50 - 100 pL of 30% hydrogen peroxide was added to the dark colored suspensions (liver, kidney, skin) and the samples were shaken for an additional 15 to 18 h. After digestion and decolorization, the samples were neutralized by addition of 30 pL of glacial acetic acid per mL Solvable. The samples were then shaken for 10 -15 min, mixed with 16 mL scintillation cocktail

69 (ScintiVerse E, Fisher Scientific Company, Fair Lawn, NJ), and counted in the

Beckman 60010 scintillation counter. For measurement of background counts,

2 mL Solvable, 60 |iL acetic acid, and 16 mL ScintiVerse E in the 20 mL

polyethylene vial were used.

Plasma and Bile. The samples were counted without digestion. Since

only 10 - 100 |j,L of bile was usually taken and 18 mL of cocktail was added,

decolorization was not necessary.

E. Statistical method

For each treatment, the average concentration of proflavine, the

metabolite, the sum of proflavine and metabolite, and the total radioactivity was

calculated for each tissue and fluid. Average values were calculated for males

and females separately, and for males and females combined.

In most instances, duplicate concentrations were determined on separate samples from each fish and all values were averaged. The standard deviation of each average was calculated, with the n value generally being 6 for males, 6 for females and 12 for the combined average.

Average and standard deviation values were calculated using spreadsheet software (Microsoft EXCEL v.5.0) and the AVERAGE and STDEV

70 functions therein. These values were also calculated using the SYSTAT

software package; Identical values were obtained using each software.

Differences between males and females were examined using the SYSTAT

statistical analysis program package, SYSTAT, Inc., 1800 Sherman Ave.,

Evanston, IL 60210-3793 (SYSTAT for Windows, Version 5 Edition). The

Multiple General Linear Hypothesis (MGLH) - General Linear Model section of

SYSTAT was used. Due to apparent lack of normal distribution of values for the dependent variable (concentration of total radioactivity, proflavine and Its

metabolites), the logarithms of the values were used as the dependent variable, and the Independent variables were tissue, gender, exposure time and depuration time. The same procedures were also performed for the treatment of acriflavine.

4.3.5. Freezer stability study

This study was designed to determine the stability during various periods of storage of total radioactivity, proflavine and Its metabolites, and acriflavine and Its metabolites (the test compounds) In samples of rainbow trout fluids and tissues after water exposure of fish to 4 ppm for 4 hr and prior to digestion analysis and HPLC analysis. It was possible that during sample storage prior to assay, analyte(s) of interest may have degraded and that as a result, the amounts measured during the

71 assay would not be an accurate measure of the amounts that were present at the time that the samples were removed from the fish.

A. Preparation and Administration of Test Article

Proflavine, dechlorinated water, 11.7 liters, was placed into a 5-gallon

(18.7 liter) rectangular, insulated container (Nalgene) having dimensions LI 2" X

W10" X H9". For acriflavine, dechlorinated water, 14.7 liters, was placed into a

10-gallon (38 liter) rectangular, glass container having dimensions L20" X W12"

X HI2". The water temperature was at 12.2-12.8 °C during the 4 hr exposure.

For the exposure solution, 42.00 mG unlabeled and 6.00 mG labeled proflavine,

54.5 mG unlabeled (89% pure) and 11.2 mG labeled acriflavine were first dissolved in 300 mL dechlorinated water, respectively. This solution was added to the 11.7 liters of water for proflavine, 14.7 liters of water for acriflavine, to make a nominal total concentration of 4.00 ppm and 100,000 dpm/mL. After addition of the 300 mL proflavine or acriflavine solution, the exposure solution was stirred for 5 min. The solution was sampled and assayed by HPLC and by direct liquid scintillation counting to ensure that all the proflavine or acriflavine was in solution. Each of these solutions was used serially to expose two fish for

4 hr. One fish was used to characterize the stability of proflavine or acriflavine and their corresponding metabolites. The other fish was used to characterize the stability of the total radioactivity content of sample. The first fish of each study was placed into the solution as soon as possible after the stirring. The

72 solution was gently aerated using a gas dispersion stone. The profile of

radiolabeled species was characterized by injection onto the HPLC of 0.1 mL of exposure solution that had been diluted 1:1 with double distilled water, measurement of UV absorption and collection of 1 minute fractions for determination of radioactivity.

B. Sample collection and analysis

Samples for HPLC analysis. The plasma was divided into ten 0.25 mL samples for proflavine, seven 0.4 mL samples for acriflavine. The kidney was removed, and the different functional regions were mixed and the mixture was divided into 6-8 equal sized samples. Ten samples each of white muscle, bile, abdominal skin, and liver were also collected. Each sample was placed into a tared storage vial. The weights of each sample were determined and recorded using an analytical balance. One set of duplicate samples was for immediate analysis by HPLC and the others were frozen in liquid N 2 and the vials were stored at -20°C. Duplicate samples were thawed and analyzed by HPLC after 1,

2, 4, and 8 weeks storage for proflavine, and after 1, 2 and 5 weeks storage for acriflavine; the procedures described in the previous section were used.

Samples for digestion analysis. The tissues were cut into small pieces (about 2 mm by 2 mm), on a Parafilm sheet, mixed thoroughly, and the mixture was divided into approximately equal sized portions (0.8-1.1 g for white muscle, G.4-0.6 g for liver, 0.2-

73 0.4 g for kidney and skin). Plasma (0.5-0.6 g) and bile (0.05-0.1 g) were also divided into equal sized portions. The portions were placed into tared and labeled plastic 20 mL liquid scintillation vials and the weight of tissue in each vial was determined. A total of 10 portions of each tissue was taken. One set of duplicate samples was assayed as soon as possible for total radioactivity by digestion analysis, following the procedures described in the previous section. The remaining samples were frozen and stored at -20 °C for future assay. Periodically the samples of plasma, bile, white muscle, kidney, abdominal skin, and liver were removed from the freezer and the duplicate assays performed on the fresh samples were repeated. The following schedule was followed: 1, 2, 4, and 8 weeks for proflavine, 1, 2, 5, and 8 weeks for acriflavine. The 8 week storage was added since the samples were available.

74 Serial Depuration Weight,

Number Time, days grams Gender

PF20 0 446 M

PF21 0 546 F

ACS 0 510 F

AC6 0 442 F

Table 4.1. Serial numbers, body weights, and genders of the fish used for the freezer stability study of proflavine and acriflavine.

75 Sampling time Radioactivity Concentration Sp. Activity

dpm/mL ppm dpm/pg

Proflavine:

Before fish PF20 100,400 4.08 20,8SS

After fish PF20 118,700 3.84 22,937

After fish PF21 112,100 4.07 21,70S

Acriflavine:

Before fish ACS 103,4S6 3.80 22,1S2

After fish ACS 10S,776 3.80 21,904

After fish ACS 102,909 3.80 21,927

Table 4.2. ’^C-Proflavine and ’'‘C-acrlflavine radioactivity and concentration in the exposure water for the freezer stability study.

76 4.3.6. Pilot Study

A. Preparation and Administration of Test Article.

Proflavine. Dechlorinated water (29.7 liters) was placed into a 30 gallon

(114 liter) cylindrical polyethylene (Nalgene) container having dimensions 18" dia by 30" high. The water temperature was maintained at 12 °C. For each exposure solution, unlabeled and labeled proflavine were first dissolved in 300 mL dechlorinated water. This solution was added to the 29.7 liters of water to make a nominal total concentration of 4.00 ppm and 50,000 or 100,000 dpm/mL.

After addition of the 300 mL proflavine solution, the exposure solution was stirred for 10-15 min. The solution was sampled and assayed by HPLC and by direct liquid scintillation counting initially and after each pair of fish was removed. The fish were placed into the solution as soon as possible after the

10-15 min. of stirring. The solution was gently aerated using a gas dispersion stone.

Fish were randomly selected by the grab method, identified with a serially numbered, color-coded tag, and exposed to the proflavine solution. Four groups of two fish each were studied according to the following regime. Two fish were exposed simultaneously in the same solution and each solution was used for two groups consecutively. Each group of fish was exposed to the 4.00 mG/liter solution for 4 hr.

77 They were then placed into proflavine-free water for 0, 2, 4, or 8 days before they were sacrificed and samples removed.

Each fish was transferred into the exposure solution and the time was

recorded. After 4 hr in the exposure solution, each fish was removed, and either transferred to a rectangular Living Stream aquarium containing 400 liters of clean water, or rinsed/anesthetized in a chipped ice-100 mG/L tricaine methanesulfonate solution mixture. The 400 liter depuration aquarium was operated in a flow-through mode at a flow of about one liter per minute. The daily water temperatures (°C), pH and dissolved oxygen (ppm) values are listed in Table 4.3.

Acriflavine. For each fish studied, 29.7 liters of dechlorinated water was placed into a 30 gallon (114 liter) cylindrical container having dimensions 12”dia by 30” high. The water temperature was maintained at 12° ±1 °C during the 4 hr exposure.

To prepare the exposure solution, 109.0 mG unlabeled (89% purity) and

21.57 mG labeled acriflavine were first dissolved in 300 mL dechlorinated water.

This solution was added to the 29.7 liters of water to make a nominal total concentration of 4.00 ppm and 100,000 dpm/mL. After addition of the 300 mL acriflavine solution, the exposure solution was stirred for 5 min. The solution

78 was sampled and assayed by HPLC and by direct liquid scintillation counting to

ensure that all the acriflavine was in solution. The fish were placed into the

solution as soon as possible after the stirring. The solution was gently aerated

using a gas dispersion stone. Fish were randomly selected by the grab method,

identified with a serially numbered, color-coded tag, and exposed to the

acriflavine solution.

Fish were transferred into the exposure solution and the time was

recorded. Two fish were exposed simultaneously and the exposure solution was used twice to consecutively expose four fish. After 4 hr in the exposure solution, the fish were removed. The first two fish exposed were transferred to a rectangular Living Stream aquarium that contained 400 liters of clean water, and they were allowed 8 days depuration (Group II, 8 day depuration). The 400 liter depuration aquarium was operated in a flow-through mode at a flow of about one liter per minute. The daily water temperatures (°C), pH values, and dissolved oxygen values are shown in Table 4.4.

79 Group Day Temperature pH Dissolved O 2 °C ppm

1 exposure 12.3 6.8 saturated

II exposure 12.3 6.8 saturated initial 12.5 6.8 10.9 1 12.4 6.8 11.0 2 12.1 11.1 lll&IV exposure 12.8 6.8 saturated

initial 12.3 6.8 11.0 1 12.4 6.8 11.0 2 12.3 6.8 10.9 3 12.3 6.8 11.0 4 12.3 6.8 11.1 5 12.3 6.9 11.0 6 12.3 6.8 11.0 7 12.3 6.8 11.1 8 12.2 6.8 11.1

Table 4.3 Conditions of exposure and depuration water during the pilot study of proflavine.

80 Day Temperature pH Disolved 0% °C ppm initial 12.2 7.1 9.9

1 12.1 7.0 9.8

2 12.1 7.0 9.9

3 12.2 7.1 9.8

4 12.0 7.0 9.9

5 12.1 6.9 9.9

6 12.2 7.0 9.8

7 12.1 7.0 9.8

8 12.0 6.9 9.9

Table 4.4. Exposure and depuration conditions during the pilot study of acriflavine.

81 B. Collection of Samples

For both test compounds, immediately after removal of the fish from the exposure solution, samples of exposure water were taken for determination of total radioactivity by liquid scintillation counting and for analysis by HPLC. After each fish was removed from the exposure or depuration aquarium, it was rinsed/anesthetized in a chipped ice-100 mG/L tricaine methanesulfonate solution mixture. Blood was taken from the fish via the dorsal aorta, through the roof of the mouth, and it was then sacrificed with a blow to the head. The blood sample was centrifuged and the plasma was collected. The fish was weighed and the gender was determined. Samples of the bile, white muscle, skin, and liver were then removed. The samples were weighed into scintillation vials for analysis of total radioactivity. Tissue and plasma samples were solubilized using commercial tissue solubilizer and total radioactivity was determined by liquid scintillation counting, according to the procedures described in Section

4.3.4.

82 4.3.7. Whole-body autoradiography

A. Preparation and Administration of Test Article.

For each fish studied, 30 liters of exposure solution with a nominal total concentration of 4.00 ppm, 200,000 dpm/mL for proflavine and 4.00 ppm,

300,000 dpm/mL for acriflavine were prepared. The solution was sampled and assayed by HPLC and by direct liquid scintillation counting to ensure that all the proflavine was in solution. The fish were placed into the solution as soon as possible after the stirring. The solution was gently aerated using a gas dispersion stone. Fish were randomly selected by the grab method, identified with a serially numbered, color-coded tag, and exposed to the proflavine or acriflavine solution. Fish were transferred into the exposure solution in pairs and the time was recorded. After 4 hr in the exposure solution, each fish was removed, and either transferred to a rectangular Living Stream aquarium containing 400 liters of clean water, or rinsed/anesthetized in a chipped Ice-100 mG/L tricaine methanesulfonate solution mixture. The first pair of fish exposed was assigned to 0 and 4 days depuration; the second pair was assigned to 1 and 2 days depuration. The 400 liter depuration aquarium was operated in a flow-through mode at a flow of about one liter per minute. The daily water temperatures (°C), pH values, and dissolved oxygen values are listed in Table

4.5 and 4.6.

83 B. Sample collection and analysis

Whole body sagittal sections of 40 pm thickness were prepared using the

PMV Model 450 MP Cryo-Microtome from three levels through the carcass. The gender of the fish was determined and recorded during the sectioning process.

The organs of interest were obtained within three section levels.

0. Preparation of Autoradiograms

The 40 pm sagittal sections were collected upon Scotch-type tape and freeze-dried within the cryomicrotome chamber. The temperature was raised to ambient temperature and the sections were apposed to Molecular Dynamics storage phosphor screens, for the detection of i^C-beta emissions. The screens

84 in their cassettes were shielded from external radiation sources such as cosmic radiation by placing them inside a lead storage chamber having 5 cm thick walls.

The phosphorimager digitized the level of fluorescence emitted by the phosphor screen, which emited phosphorescence in proportion to the concentration of radioactivity in the sagittal section that overlayed the screen. The digital output from discrete areas was used to provide a quantitative estimate (relative) of total residue in the various tissues and organs.

85 Day Temperature pH Dissolved O 2

°C ppm

initial 12.3 6.8 11.2

1 12.3 6.8 11.1

2 12.3 6.8 11.1

3 12.4 6.8 11.1

4 12.3 6.8 11.1

Table 4.5. The daily water temperatures (°C), pH values, and dissolved oxygen values during the period of autoradiography study of rainbow trout exposed to 4 ppm ^‘‘C-acriflavine.

86 Day Temperature pH Dissolved 0%

°C ppm

initial 12.2 7.0 10.0

1 12.3 7.0 9.9

2 12.2 7.1 10.0

3 12.3 7.0 9.9

4 12.2 7.0 10.0

5 12.2 7.0 10.0

6 12.3 7.1 9.9

7 12.2 7.0 10.0

8 12.1 6.9 10.1

Table 4.6. The daily water temperatures (°C), pH values, and dissolved oxygen values during the period of autoradiography study of rainbow trout exposed to 4 ppm ^^C-acriflavine.

87 4.3.8 Main study:

A. Preparation and Administration of Test Article.

Proflavine. Fish were randomly selected by the grab method, tagged, and exposed to the proflavine solution. Five groups of six fish each were studied according to the following regime. Three fish were exposed simultaneously in the same solution and each solution was used twice for two groups of three fish.

For Treatment I and II, where the time from initial exposure to collection of samples was short, the exposure regime was modified to three groups of fish to more accurately manage the sample collection. The gender and weight of fish used are listed in Table 4.7. The nominal concentration of the exposure solution for each treatment is listed in Table 4.8.

Treatment I Exposed to the 4.00 mG/L solution for 2 hr, and sacrificed and

samples removed.

Treatment II Exposed to the 4.00 mG/L solution for 4 hr, and sacrificed and

samples removed.

Treatment III Exposed to the 4.00 mG/L solution for 4 hr, placed into

proflavine free water for 2 days, and sacrificed and samples

removed.

88 Treatment IV Exposed to the 4.00 mG/L solution for 4 hr, placed into

proflavine free water for 4 days and sacrificed and samples

removed.

Treatment V Exposed to the 4.00 mG/L solution for 4 hr, placed into

proflavine free water for 8 days and sacrificed and samples

removed.

89 Treatment Exposure Depuration Fish ID # Weight Time, hr Time, days Gender grams 1 2 0 PF22 M 639 2 0 PF23 M 637 2 0 PF24 F 731 2 0 PF25 F 629 2 0 PF26 F 809 2 0 PF27 M 802

II 4 0 PF34 M 610 4 0 PF35 M 836 4 0 PF36 F 655 4 0 PF37 M 633 4 0 PF38 F 603 4 0 PF39 F 747

III 4 2 PF28 F 815 4 2 PF29 M 530 4 2 PF30 F 655 4 2 PF31 M 531 4 2 PF32 F 532 4 2 PF33 M 517

IV 4 4 PF40 M 540 4 4 PF41 F 601 4 4 PF42 F 535 4 4 PF43 M 580 4 4 PF44 F 691 4 4 PF45 M 502

Table 4.7 to be continued.

90 Table 4.7 (continued)

4 8 PF46 F 409 4 8 PF47 F 470 4 8 PF48 M 440 4 8 PF49 M 471 4 8 PF50 F 432 4 8 PF51 M 440 Mean 601 SD 122

Table 4.7. Serial numbers, body weights, and genders offish used to study the total residue depletion and metabolite profile of proflavine.

91 Treatment Proflavine hemisulfate, mG Concentration

(nominal)

unlabeled labeled total mG/L dpm/ mL

1 168.75 11.25 180 4.00 50,000

II 168.75 11.25 180 4.00 50,000

III 168.75 11.25 180 4.00 50,000

IV 157.5 22.5 180 4.00 100,000

V 157.5 22.5 180 4.00 100,000

Table 4.8. Amounts of unlabeled and ’^C-proflavlne used to prepare 45 liters of exposure solution for the study of proflavine total residue depletion and metabolite profile. Also shown is the nominal concentration of proflavine and the concentration of radioactivity.

92 Acriflavine. Dechlorinated water (44.7 liters) was placed into a 30-gallon

(114 liter) cylindrical container (Nalgene) having dimensions 18”dia by 30"high.

The water temperature was maintained at 12° ± 1° C by immersion of the container in a thermostatted water bath. For each exposure solution, unlabeled and labeled acriflavine were first dissolved in 300 mL dechlorinated water. This solution was added to the 44.7 liters of water to make a nominal concentration of

4 ppm and 100,000 or 150,000 dpm/mL.

Fish were randomly selected by the grab method, tagged, and exposed to the acriflavine solution (Table 4.9). Two groups of six fish were studied according to the following regime. Three fish were exposed simultaneously in the same solution and each solution was used twice (Treatments I and II) for two groups of three fish and three times for three groups of three fish (Treatment III).

In Treatment II one fish died after 3 hours of depuration. In Treatment III, because of the long depuration time, more fish were introduced into the exposure in order to have 6 healthy fish after 16 days of depuration. From the first group of three fish one male was found dead after 1 hour of exposure, and one male fish died from the sam e group after 2 days of depuration. Thus 7 fish were available after 16 days of depuration, the last one from the third group was not used for the study. Since by the end of this study only few male fish were available and unfortunately two of them died, only one male fish survived for

Treatment III.

93 Treatment I Exposed to the 4 mG/L solution for 4 hr. and sacrificed and

samples removed.

Treatment II Exposed to the 4 mG/L solution for 4 hr, placed into acriflavine-

free water for 8 days, and sacrificed and samples removed.

Treatment III Exposed to the 4 mG/L solution for 4 hr, placed into acriflavine-

free water for 16 days, and sacrificed and samples removed.

B. Collection, Storage and Analysis of Samples

For both test compounds, after fish were transferred into the exposure

solution, the time was recorded. After 4 hr in the exposure solution, the fish

were removed, and rinsed/anesthetized in a chipped ice-100 mG/L tricaine

methanesulfonate solution mixture or transferred to the depuration tank. After 2

to 16 days depuration the fish were removed, and rinsed/anesthetized in a

chipped ice-100 mG/L tricaine methanesulfonate solution mixture. Immediately

after removal offish from the exposure solution, samples of exposure water were taken for determination of total radioactivity by liquid scintillation counting and for analysis by HPLC. The clock times that fish were removed from the exposure solution and the depuration tank were recorded. After anesthesia blood was collected from each fish via the dorsal aorta, through the roof of the mouth, and the fish was sacrificed with a blow to the head. The blood sample was centrifuged and the plasma was collected.

94 Treatment Depuration Fish ID # Gender Weight Time, days G 1 0 AC22 M 403 0 AC23 F 459 0 AC24 F 508 0 AC25 F 393 0 AC26 M 438 0 AC27 M 465 II 8 AC17 M 448 8 AC18 F 420 8 AC19 F 461 8 AC20 F 410 8 AC21 M 409 III 16 AC28 F 456 16 AC29 M 417 16 AC30 F 470 16 AC31 F 428 16 AC32 F 420 16 AC33 F 518 Mean 443 SD 35.5

Table 4.9. Serial numbers, body weights, and genders of fish used to study the total residue depletion and metabolite profile of acriflavine.

95 The fish weight was determined and recorded. The gender was confirmed and

recorded after visual inspection of the viscera. Samples of bile, kidney, muscle, abdominal skin, and liver were removed. Samples of plasma, bile, kidney, muscle, abdominal skin, and liver for analysis by HPLC were placed into tared polyethylene storage vials, the fresh sample weight was determined using an analytical balance and recorded. Each sample was frozen in liquid Nz and then transferred to a freezer at -20 “0. Fillets from 1 male and 1 female fish were taken from each treatment, and stored at -20 °C until analyzed for total radioactivity.

4.4. Results and Discussion

4.3.1. Freezer stability study

A. Proflavine

Total Radioactivity (Fish PF-20L The total radioactivity in all tissues and fluids studied appeared to be stable for up to 8 weeks of storage (Table 4.10 and

Figure 4.1). There was some fluctuation over time, which could have been due to sample-to-sample variability in the initial concentration of radioactivity for the solid tissues (muscle, liver, skin), but not for the fluids (plasma and bile) nor for the kidney. The kidney is relatively fluid after its collection and it was minced prior to separation into samples. The plasma and bile were collected as a single

9 6 large sample that was subdivided prior to freezing. Samples were weighed when they were inserted into the storage vials. The variability apparent in Table

4.10 for total radioactivity may have been due in part to chemilumenescence, which can be a problem with the digestion procedure used. The liquid scintillation counter was equipped with a chemilumenescence detector and when chemilumenescence was detected, the liquid scintillation vials were stored in the dark and recounted. This was not done for the kidney sample after 2 weeks storage, which explains the large peak in apparent radioactivity. Unfortunately, the liquid scintillation vials for these samples were discarded and it was not possible to recount them. While there was some fluctuation over time in the concentration of radioactivity, there was no evidence that radioactivity was lost over 8 weeks storage. The data shown in Figure 4.1 for each tissue generally fall along a horizontal line and it would appear that total radioactivity was stable for considerably longer than the 8 week period of study.

Proflavine and Metabolite (Fish PF-211. Proflavine in the tissues studied showed varying degrees of fluctuation (Table 4.10). In plasma its concentration appeared to decline from 0 to 2 weeks; it then appeared to increase. It is unlikely that this represents the true behavior and the results probably derive from the relatively low concentration of proflavine present in the plasma, and the difficulty of obtaining good analytical results when the concentration of analyte is at the limit of detection. I interpret the data as indicative of stability in plasma

97 over the 8 day period. In bile, muscle, skin, and, kidney, the concentration of

proflavine appeared to decline rather abruptly during the first week of storage

and then to remain unchanged for the duration of the storage period; note that the final determination for muscle was well below the 4 week value and there

may be a problem with its stability after 4 weeks. The abrupt decline from 0 to 2 weeks was puzzling; there was a possibility that these time zero samples were contaminated with radioactivity during a nitrogen evaporation step and that the time 0 values were elevated as a result. In liver, proflavine concentration showed random fluctuation but there was no evidence that the concentration declined over the period of storage. I concluded from these results that proflavine appeared to be stable during freezer storage for up to 8 weeks in plasma, bile, skin, kidney, and liver. Muscle samples, however, should not be stored longer than 4 weeks, due to the possible decline in its concentration that was apparent in Table 4.10.

Metabolite 1 is a conjugate of proflavine and there was preliminary evidence that it was a saccharide conjugate, which was identified later as a glucuronide (Chapter 5). It was present in plasma at a much higher concentration than the parent after IV administration of proflavine (data not shown). After water exposure, however, this metabolite was at the limit of detection in plasma, but it appeared to be stable. Table 4.10. It was also present in bile, liver and kidney and in these tissues its concentration appeared

98 not to decline with time over the 8 week period of freezer storage. Metabolite 1

was not observed in muscle or in skin, except at the limit of detection in the

samples that were stored two weeks.

Metabolite 2 was not normally observed in fresh samples. It eluted from

the HPLC after proflavine. It formed after storage, and primarily during the first

week, in liver and kidney. Table 4.10. Therefore, Metabolite 2 was probably a

product of proflavine decomposition; e.g., oxidization. Liver and kidney samples

therefore, should not be stored more than a week before HPLC analysis.

In summary, radioactivity associated with proflavine residue in tissues from rainbow trout that were exposed to 4 ppm ’^C-proflavine was stable for up to 8 weeks of storage at -20 “C. Proflavine and an endogenously formed metabolite of proflavine were also stable over the 8 weeks period in all tissues, except muscle where the 8 week value was well below the 4 week value. This may have been an artifact as it hinged on a single analysis (duplicate samples), but until further studies are carried out, muscle samples should not be stored longer than 2 weeks prior to their assay.

B. Acriflavine

The total radioactivity in all tissues and bile appeared to be stable during

8 weeks of storage (kidney was studied for only 5 weeks). Table 4.11 and Figure

9 9 4.2. Plasma showed an apparent 13% decrease in total radioactivity after 5 weeks of storage. There was some fluctuation over time, which may have resulted from sample-to-sample variability in the concentration of radioactivity for the solid tissues. The recovery versus time profiles shown in Figure 4.2 for each tissue and bile fell generally along a horizontal line and it appeared that total radioactivity in the fluids and tissues studied was stable for considerably longer than the 8 week period of study.

The concentration of acriflavine and the other radiolabeled species in the tissues and fluids showed varying degrees of fluctuation, Table 4.11. A possible reason for these fluctuations was that the concentrations of acriflavine and other species approached the limits of their detection by the HPLC method. This was the case with the concentrations in plasma, muscle, and skin, and with the concentrations of non-acriflavine species in liver and kidney. The concentration of acriflavine approached the detection limit in liver and some bile samples. The concentrations of the non-acriflavine species in bile were also close to their detection limits. Only the acriflavine concentration in kidney was well above the limit of detection. For the solid tissue samples, another source of variability was in the sample-to-sample variability in the initial concentration of each analyte, and sample-to-sample variability in the stability of the various analytes.

100 The ’^C-labeled acriflavine available for this study had a radiochemical purity of 88%; a purer form was unavailable and attempts at further purification led to unacceptable losses. The '^C-label was associated with acriflavine and with two other species designated Unk 1 and Link 2. The distribution of radioactivity among these species was calculated from the radioactivity in each fraction compared with the sum of the radioactivity in all fractions. For example,

HPLC analysis of the starting solution showed that 4,206 dpm was associated with the acriflavine peak, and that 24 and 210 dpm were associated with the peaks for Unk 1 and Unk 2. The sum of the radioactivity in this sample was

4440 dpm, and 94.7% was acriflavine. Unk 1 (structure not identified) was more polar than was acriflavine and it eluted 3 minutes before acriflavine. Unk 2

(structure not identified) was less polar than acriflavine, with a retention time 3 minutes longer than acriflavine. The presence of the two minor radiolabeled species was also detected in the stored unlabeled acriflavine. The exposure solution contained 0.54% Unk 1 and 4.73% Unk 2 before fish AC 5 was exposed. Unk 2 was not found in kidney, but was 3.56% of the radioactivity in the fresh liver and 9.24% of radioactivity in the fresh bile. In bile the ratio of Unk

2 did not change significantly during the 8 week period of storage, but in liver there was a slight increase in Unk 2. Unk 1 was present in larger relative concentration in bile and liver, than in the exposure solution, which could have resulted either from the formation of this metabolite in these tissues or from its preferential accumulation by these tissues. Metabolite 1 was present only in

101 liver Immediately after exposure, and It was found In bile samples analyzed after

2 and 8 weeks of freezer storage.

The following conclusions can be drawn. Plasma, muscle and probably skin cannot be analyzed for metabolite profile due to the low accumulation of acrlflavlne In the tissues. Kidney can be stored for at least 5 weeks without a significant change In the concentration of acrlflavlne. Liver and bile samples should be analyzed within 2 weeks’ storage to avoid changes In the composition of residue. Skin samples can be stored for 8 weeks without any decomposition.

Radioactivity associated with acrlflavlne residue In tissues from a rainbow trout that was exposed to 4 ppm ^^C-acrlflavlne was stable for up to 8 weeks

(kidney studied for 5 weeks) of sample storage at -20 “C. Acrlflavlne determined by HPLC approached the limit of detection In plasma and muscle and Its stability

In these matrices could not be characterized. The concentration of acrlflavlne In kidney was stable over the 5 weeks storage period. Liver and bile samples should be analyzed within two weeks to avoid significant changes In the concentrations of acrlflavlne and Its metabolites. Skin samples had no detectable concentrations of non-acrlflavlne residue, and acrlflavlne appeared to be stable In the skin samples for the 8 week storage period.

1 0 2 2 0 0 1

skin 160 - o ' kidney I (D plasma O) CO 80 - bile liver I muscle <0 0. 40 -

0 24 6 8 Storage Time, weeks

Figure 4.1. Concentration-time profile of radioactivity in fluids and tissues, from rainbow trout exposed to 4 ppm ^'’C-proflavine for 4 hr, during storage in a freezer at -20 °C. Concentrations are expressed as a percentage of the corresponding zero time values.

103 2 0 0 1

160 - liver 0 1 120- kidney skln o 0} c 80 - plasma g bile muscle 40 -

0 24 6 8 Storage Time, weeks

Figure 4.2. Concentration-time profile of radioactivity in fluids and tissues, from

rainbow trout exposed to 4 ppm ’'‘C-acriflavine for 4 hr, during storage in a freezer at -20 “C. Concentrations are expressed as a percentage of the

corresponding zero time values.

104 Time ■ Proflavine Metab. 1 Metab.2 Sum week ppb nmole/G nmol/g nmol/G nmol/G

Plasma 0 36.2 0.0601 0 0 0.0601 1 38.4 0 0 0 0 2 38.6 0.0291 0.0442 0 0.0733 4 36.3 0.0424 0.0249 0 0.0673 8 37.1 0.108 0.0675 0 0.176 Bile 0 3,950 37.3 17.3 0 54.6 1 3,880 25.2 34.6 0 59.8 2 4,090 29.7 17.4 0 47.1 4 3,270 19.9 14.1 0 34.0 8 3,980 22.3 17.5 3.11 42.9 Muscle 0 9.04 0.143 0 0 0.143 1 8.91 0.0342 0 0 0.0342 2 10.6 0.0294 0 0 0.0294 4 9.97 0.0458 0 0 0.0458 8 10.1 0.00654 0 0 0.00654 Skin 0 106 1.17 0 0 1.17 1 113 0.408 0 0 0.408 2 118 0.410 0.0349 0 0.456 4 121 0.361 0 0 0.361 8 105 0.430 0 0 0.430

To be continued.

105 Table 4.10 (continued)

Liver 0 597 1 550 3.01 1.47 0.398 4.48 2 670 3.62 1.77 0.420 5.79 4 641 2.74 1.17 0.421 4.33 8 592 1.76 0.804 0.465 2.99 Kidney 2.52 1.69 4.68 0 483 0 2 735 3.05 0.255 0.893 3.31 8 541 2.48 0.265 0.311 3.64 2.38 0.183 0 2.87 'Each value represents the mean of 2 determinations; a value of 0 indicates that the

concentration was below the limit of detection.

Table 4.10. Concentration® during freezer storage of total radioactivity, proflavine, and two metabolites in multiple samples of tissues and fluids from rainbow trout exposed for 4 hr to 4 ppm ^“C-proflavine

106 Tissue/Fluid Time Acriflavine Metab 1 Link 1 Unk2 Sum week ppb ppb ppb ppb ppb ppb

Plasma 0 18.7 1.60 0 0 0 1.60 1 17.1 3.75 0 0 0 3.75 2 19.6 0 0 0 0 0 5 15.7 na" na^ na" na" na" 8 15.3 na^ na" na** na** na" Bile 0 2059 399 0 261 67.2 727 1 1832 191 0 416 55.7 663 2 1916 266 23.5 271 74.8 636 5 1935 215 0 224 212 651 8 1895 95.6 59.5 240 76.9 472 Muscle 0 5.36 4.43 0 0 0 4.43 1 5.07 0 0 0 0 0 2 5.33 0 0 0 0 0 5 4.33 0 0 0 0 0 8 4.88 0 0 0 0 0 Skin 0 67.5 7.34 0 0 0 7.34 1 55.9 16.0 0 0 0 16.0 2 60.7 3.07 0 0 0 3.07 5 71.4 10.0 0 0 0 10.0 8 76.6 6.38 0 0 0 6.38

To be continue d

107 Table 4.11 (continued)

Liver 0 426 27.3 6.92 3.04 1.38 38.7 1 409 10.0 8.16 2.63 0 20.8 2 383 35.2 1.62 2.43 0 39.2 5 388 19.1 2.17 6.28 2.17 29.7 8 419 7.61 0 7.39 2.39 17.4 Kidney 0 593 202 0 2.90 0 205 1 591 156 7.00 0 0 163 2 659 117 0 0 0 117 5" 609 168 0 0 0 168 “Each value represents the mean of two determinations; a value of 0 indicates that the concentration was below the limit of detection. "No samples were available for metabolite profile analysis of plasma "No samples were available for 8 weeks because of the small size of kidney

Table 4.11. Concentration® during freezer storage of total radioactivity (Fish

ACS), and of acriflavine and three metabolites (Fish ACS) in multiple samples of tissues and fluids from rainbow trout exposed for 4 hr to 4 ppm ^^C-acriflavine

108 4.4.2 Pilot Study:

A. Proflavine

The concentration-time profile of radioactivity in plasma, bile, liver, skin, and muscle of rainbow trout exposed to 4 ppm ‘•'^C-proflavine was characterized.

The concentration of residue in the tissues after the 4 hour water exposure was generally a small fraction of the concentration in the exposure water: plasma

2%, muscle 1%, skin 12%, and liver 34%. However, the residue concentration in the bile was more than 9 times the water concentration. Distribution to the tissues was apparently relatively rapid as the concentration-time profile of residue in the tissues and plasma declined in parallel. Figure 4.3. This was not the case in bile where the concentration of residue initially increased and then seemed to remain at the Day-0 value.

Bile showed the highest concentration of radioactivity, followed in descending order by liver, skin, plasma, and muscle. By 4 days, the concentration of residue was between 4.4% and 12% of the corresponding 0-

Day value in liver, skin, plasma, and muscle; the concentration did not decline further between 4 and 8 days depuration in these tissues and plasma. The concentration in bile increased on Day 2 to 2.60 times the Day-0 value and it then declined by Day 8 to near the Day-0 value. Figure 4.3. The residue concentration in edible tissues (plasma, skin, muscle) was below 0.1 ppm by 8

109 days depuration, and in liver the residue concentration approached 0.1 ppm

(0.163 ppm). Thus, it appeared that the main residue depletion and metabolite

profile protocol could be constrained to an 8 day depuration period. The

concentration of residue in the tissues after the 4 hour water exposure was

generally a small fraction of the concentration in the exposure water: plasma

residue in the tissues and plasma declined in parallel.

110 Depuration Plasma Muscle Skin Liver Bile Fish ID Time (Days)

PF5 0 0.0789 0.0421 0.839 1.414 21.8 0.0782 0.0445 0.725 1.463 22.6 0.0802 0.0416 1.167 1.414 22.0 PF6 0 0.0715 0.0341 0.145 1.314 51.3 0.0717 0.0348 0.166 1.317 50.8 0.0701 0.0327 0.151 1.320 47.9

Mean 0.0751 0.0383 0.478 1.374 36.1 SD 0.0045 0.0050 0.426 0.064 15.3 Percentage of 0 day 100% 100% 100% 100% 100%

PF7 2 0.0177 0.0065 0.161 0.370 128. 0.0178 0.0068 0.165 0.333 127. 0.0172 0.0067 0.160 0.313 124. PF8 2 0.0175 0.0079 0.183 0.585 61.8 0.0174 0.0081 0.165 0.614 79.2 0.0171 0.0077 0.168 0.613 76.9

Mean 0.0174 0.0073 0.167 0.471 94.0 SD 0.0003 0.0008 0.009 0.147 29.9 Percentage of 0 day 23.2% 19.1% 34.9% 34.3% 260.4%

Table 4.12 to be continued.

I l l Table 4.12 (continue).

PF11 4 0.00411 0.00158 0.0596 0.0916 10.4 0.00367 0.00163 0.0554 0.0968 10.9 0.00408 0.00190 0.0569 0.0887 10.3

Mean 0.00395 0.00170 0.0573 0.0924 10.5 SD 0.00024 0.00017 0.0021 0.0041 0.32 Percentage of 0 day 5.2% 4.4% 12% 6.7% 29.1%

PF9 8 0.00631 0.00261 0.0563 0.208 10.4 0.00698 0.00239 0.0935 0.222 11.5 0.00612 0.00239 0.0720 0.213 11.7 PF12 8 0.00322 0.00142 0.0653 0.136 56.5 0.00303 0.00169 0.0681 0.098 55.3 0.00310 0.00138 0.0664 0.103 56.9

Mean 0.0048 0.00198 0.0703 0.163 33.7 SD 0.0018 0.00055 0.0125 0.058 24.7 Percentage of 0 day 6.4% 5.2% 14.7% 11.9% 93.4%

Table 4.12. "'^C-proflavlne equivalent (pg/g) in trout tissues at 0, 2, 4, and 8 days after water exposure to 4.00 ppm i^C-proflavine for 4 hr.

112 100

bile 10 I c I § liver E 0.1 Ü skin

0.014 muscle

plasma

0.001 J 0 2 4 6 8 Depuration Time, days

Figure 4.3. ’^C-proflavine equivalent (pg/g) in trout tissues and fluids after various times of depuration that followed a 4 hr water exposure to 4 ppm '^C- proflavine.

113 B. Acriflavine

The acriflavine total radioactivity and the metabolite profile in the exposure solution before and after exposure of two groups with two rainbow trout in each group were analyzed. From HPLC analysis, two unknown species other than acriflavine were observed. Species Unk 1 (not identified) was more polar than acriflavine and eluted 3 minutes before acriflavine did. Species Unk 2 (not identified) was less polar than acriflavine, with an elution time that was 3 minutes after acriflavine. The total concentration of radioactivity did not change during the exposure of the two group offish serially.

Table 4.13 shows the ’^C-acriflavine equivalent (nG/G) in trout tissues and fluids at 0 and 8 days after water exposure to 4.01 ppm ^'’C-acriflavine for 4 hours. Immediately after exposure (no depuration) bile showed the highest concentration of radioactivity, followed in descending order by kidney, liver, skin, plasma, and muscle. After 8 days depuration, the concentration of residue was decreased only in plasma and liver, to 4.37% and 18.8% respectively of the 0 day value. The concentration of radioactivity decreased slightly over 8 days in skin to an average of 76.6% of the 0 day value, but the average residue values in skin in both groups showed high standard deviations. Depletion from skin will be carefully followed in the main study, since it is part of the edible fillet.

Radioactivity did not decrease during the 8 days of depuration in muscle and kidney (95.7% and 102.3% respectively of the 0 day value). In bile the average

114 increase was 352.6% after 8 days. The concentration of radioactivity in bile depended not only on the amount of radioactivity present but also on the volume of the bile. The volume of bile was not recorded (it is technically not always feasible). Thus the 16 times larger concentration of radioactivity in fish AC11 compared with fish AC10 was not necessary due to a larger amount of radioactivity in the bile.

C. Comparison of proflavine and acriflavine

Compared with proflavine, the presence of acriflavine residue was lower in all tissues after similar exposure conditions. At the end of the four hour exposure period, the ratio of the acriflavine residue concentration (Table 4.13) to the proflavine residue concentration (Table 4.12) was 1/9 in bile, 1/8 in liver, 1/7 in muscle, 1/6 in plasma, 1/5 in kidney and 1/3 in skin. The lower residue concentration for acriflavine may be due to slower absorption as a result of the quaternary ammonium group, which confers a formal positive charge on the molecule. Proflavine is weakly basic and while it would exist in the exposure solution predominantly in the ionized form, there would be a significant concentration of the electrically neutral form, which would be expected to penetrate absorbing membranes more rapidly than the protonated form of proflavine and the cationic acriflavine.

115 Proflavine and acriflavine residue levels in plasma and liver showed similar depuration rates. After 8 days, the residue of proflavine in plasma had declined to 8.21% of its initial value and the residue of acriflavine declined to

4.37% of its initial value. In liver, proflavine residue was 12.8% after 8 days compared with 18.8% for acriflavine. In kidney and bile, residue levels were higher at 8 days than at the end of the exposure for both compounds. Kidney for proflavine was 158% of its initial value after 8 days, compared with 102% for acriflavine. In bile after 8 days, 404% of the initial value was observed for proflavine compared with 351.6% for acriflavine. Proflavine was also more rapidly eliminated from skin during the 8 day depuration than was acriflavine:

39.6% of the 0 day depuration value remained versus 76.6% for acriflavine. The largest and most significant difference between the two compounds was their depletion from muscle. During 8 days of depuration 10% of the initial concentration of proflavine remained in muscle, compared with 95.7% of the initial acriflavine. This preliminary observation was more fully studied in the main acriflavine study, which used a larger number offish and both genders.

As a guide to design the main study of total residue depletion and metabolic profile of acriflavine, this pilot study indicates that a relatively high concentration of radioactivity will be needed in the exposure water to achieve detectable tissue concentrations of residue. A concentration of 100,000 to

200,000 dpm/mL would be the minimum concentration requirement. The long

116 elimination half life indicates that the main study should be designed with a relatively long depuration phase. It may not be possible to fully characterize the depuration phase in all tissues without using a several month depuration period.

Acriflavine was poorly taken up from exposure water over a 4 hour exposure period. The concentration of residue in plasma was only 0.22% of the water concentration, and in muscle it was 0.06% of the exposure water concentration. The distribution profile after the 4 hr exposure was in descending order; bile > kidney > liver > skin > plasma > muscle. The residue level declined during an 8 day depuration period in plasma and liver; muscle, skin, and kidney levels were similar to the 0 day depuration values, and residue in bile was increased compared with the 0 day value. Compared with proflavine, acriflavine was more slowly accumulated, had a similar residue distribution profile, was more slowly eliminated, and the depletion of residue from muscle was much slower for acriflavine than for proflavine.

117 Depuration Plasma Muscle Skin Liver Kidney Bile

Fish ID Time (Days)

ACS 0 8.05 2.21 39.9 120 171 644 8.06 2.48 40.5 98.1 190 622 AC9 8.56 2.96 65.8 155 257 556 10.1 2.64 71.1 152 143 572

Mean 8.69 2.57 54.3 131.2 190.1 599 SD 0.90 0.32 20.0 31.3 14.3 48.5 Percentage of 0 day 100% 100% 100% 100% 100% 100%

AGIO 8 0.15 2.53 20.3 17.8 172 245 0.99 1.93 14.8 17.0 166 249 AC11 0.36 2.64 64.3 30.4 222 4046 0.0 2.75 67.0 33.1 223 3882

Mean 0.38 2.46 41.6 24.6 195.6 2106 SD 0.28 0.33 34.0 10.1 38.2 2628 Percentage of 0 day 4.37% 95.7% 76.6% 18.8% 102.9% 351.6%

Table 4.13. ^'‘C-acriflavine equivalent (nG/G) in trout tissues and fluids at 0 and 8 days after water exposure to 4.01 ppm ’'’C-acriflavine for 4 hr. 4.4.3. Whole-body autoradiography

A. Proflavine

The autoradiograms were consistent with the time course of radioactivity distribution determined in the pilot studies by digestion analysis of radioactivity.

After four hours exposure to 4.00 ppm proflavine and no depuration, the highest concentration of radioactivity was present in the gallbladder, Figure 4.4. Very high levels were also observed in the kidney, liver, upper intestinal lumen, and gills. Concentration of residue in the olfactory region was also observed. Lower levels were present in the spleen, heart/blood, and lower intestinal lumen. The muscle contained detectable but relatively low levels and the level within the central nervous system was below the detection limits of this technique. After one day of depuration the levels within the kidney, liver, and gills had declined significantly. Figure 4.5. The level within the trunk portion of the kidney was somewhat higher than that in the head portion. Extremely high levels remained within the gall bladder and intestinal lumen. Significant levels were present within the spleen; levels within the heart/blood and muscle were near background. After two days depuration (Figure 4.6) the distribution was similar to that observed at one day depuration and the levels of radioactivity had declined only slightly. The gallbladder in this trout contained extremely high levels of radioactivity as did the intestinal lumen. An area of concentration remained in the olfactory region. The distribution pattern at 1 and 2 days was

119 also observed after 4 days depuration, Figure 4.7. The highest levels were in the gall bladder and intestinal lumen. The concentration within the kidney of this fish remained relatively high, about half the level observed in the 0 depuration fish. The liver and gills showed low levels of residue, and residue was also apparent in the olfactory region. The similar pattern of distribution of radioactivity observed in the 2 and 4 day depuration fish was consistent with the relatively long half life for residue that was observed in the total residue depletion pilot study.

1 2 0 Figures 4.4. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm '•^C-proflavine. Sections were exposed on the phosphor screen for 28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY: BR, brain; GB, gall bladder; GL, gill; H, heart/blood; I, intestine; K, kidney; L, liver; M, muscle; 0, olfactory region; PC, pyloric caeca; SK, skin; SR. spleen.

121 Figure 4.4

/ V - ‘ /

: S '. : À '

. 9

122 Figure 4.5. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm i"^C-proflavine and depuration in clean water for one day. Sections were exposed on the phosphor screen for

28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY: same as Figure 4.4, and IL, intestinal lumen.

123 figure ^

124 Figure 4.6. Autoradiograms of sagittal sections from 3 levels In rainbow trout after water exposure for 4 hr to an Initial 4.0 ppm '*'*C-proflavlne and depuration

In clean water for two days. Sections were exposed on the phosphor screen for

28 days and the same print range (5-250) was used for the four figures. The relative Intensity of radioactivity is directly comparable among the figures. KEY: same as Figure 4.4, and AR, sectioning artifact; IL, Intestinal lumen; LA, liver accumulation.

125 (O'c (D 45k b)

GL SP

to a> ' ..,v : # ') -V:.

- i Figure 4.7. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ‘'^C-proflavine and depuration in clean water for four days. Sections were exposed on the phosphor screen for

28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY; same as Figure 4.4, and IL, intestinal lumen; LA, liver accumulation.

127 Figure 4.7.

/I)-

h .*4

128 B. Acriflavine

Whole-body autoradiograms of sagittal sections from rainbow trout were

prepared after water exposure to 4 ppm ''^C-acrlflavlne and were permitted 0, 2,

4 and 8 days depuration In acrlflavlne-free water (Figs. 4.8-4.11 ).

After four hours exposure to 4 ppm acriflavine and no depuration, the

highest concentration of radioactivity was present In the gall bladder, Figure 4.8.

Very high levels were also observed In the kidney, liver, pyloric caeca, upper GI tract, and gills. Relatively high concentrations of residue were observed In the skin, heart/blood, and spleen. Concentration of residue In the olfactory region was also observed. The olfactory cavity communicates directly with the external solution and radioactivity In this region may be due to surface adsorption or to adherent exposure solution. Lower levels of residue were present In the lower

Intestine, brain, and muscle. After two days of depuration the levels of radioactivity had generally declined. Figure 4.9, except for the Gl lumen where the concentration of radioactivity was higher than observed In the non-depurated fish (Figure 4.8). The level within the trunk portion of the kidney was somewhat higher than that In the head portion. High levels remained within the gall bladder, kidney, and Intestinal lumen. Low levels were present within the liver, gill, skin, spleen and olfactory region; levels within the heart/blood, muscle, and brain were near background. The concentration of residue In the skin was significantly higher than In the muscle.

129 After four days depuration (Figure 4.10) the distribution was similar to that observed at two days depuration and the levels of radioactivity had declined only slightly. The gallbladder in this trout contained high levels of radioactivity as did the pyloric caeca and intestinal lumen. The skin concentration of radioactivity was relatively low. In this trout, an uneven distribution pattern within the liver was observed. Some high spots of radioactivity (artifacts) due to transport of radioactivity by the microtome knife during sectioning were observed in muscle

(AR, Fig. 4.10). The uneven distribution in the liver was not thought to be an artifact, however, as the same uneven distribution was apparent in Fig. 4.11 (8 days depuration). Uneven distribution of residue in the livers of the 0 and 2 day depuration fish (Figs. 4.8 and 4.9) was not observed. This fish was a female and a well developed egg mass was apparent in the abdominal cavity. Residue was not observed in the egg mass.

The distribution pattern at 2 and 4 days was also observed after 8 days depuration. Figure 4.11. The highest levels were in the gall bladder and distal intestinal lumen. The concentration within the kidney of this fish remained high.

Skin levels were similar to those observed at 2 and 4 days depuration, and the gills showed low levels of residue. As with the 4 day depuration fish uneven distribution of hepatic residue was apparent. A low concentration of residue, not readily visible in Figure 4.11, was apparent in the gonads of this male fish.

130 Table 4.14 shows the concentration of radioactivity in trout tissues and organs based on the digital output of the phosphohmager; the values shown are arbitrary values that may be used to compare relative concentrations of radioactivity among the tissues. The figures can only show a limited dynamic range. The dynamic range chosen for the figures was 5-250 arbitrary units, which means that intensity values above 250 were indicated as black in the autoradiograms. All regions with intensity greater than 250 show the same relative concentration of radioactivity in the figures. From Table 4.14, it is apparent that these regions of high concentration of radioactivity have different concentrations.

The autoradiograms in Figures 4.8-4.11 together with the relative concentrations of radioactivity in Table 4.14 indicate that biliary excretion is a dominant pathway for elimination of acriflavine residue from trout. Persistent high levels of radioactivity within the lower Gl tract apparent in Figures 4.9-4.11 suggest the possibility of enterhepatic cycling of residue. If residue were not cycling, lower levels of radioactivity would be expected after 8 days depuration as the fish were fed during the depuration period and the intestinal residue would likely be eliminated in the feces.

The similar pattern of distribution of radioactivity observed in the 2, 4 and

8 day depuration fish was consistent with the relatively long half life for residue

131 that was observed in the total residue depletion pilot. Both studies identified the same tissues of high accumulation of residue. The autoradiograms from the zero-depuration fish (Fig. 4.8) tended to show a higher initial average body concentration of residue than would have been suggested from the results of the digestion analysis; it was expected that the concentration of radioactivity in muscle, blood, and brain would have been too low to detect. Interindividual variability and the small number of fish involved in the two studies may explain this variance. Another possible explanation is that different salt forms of acriflavine were used. The pilot study used ^'‘C-acriflavine hydroxide that was purified by NCTR, Jefferson, AR, from the DuPont/NEN preparative reaction mixture. The Autoradiography study and the main study, however, used the ^'’C- acriflavine hydrochloride prepared by Alex Tomazic of Bionetics, NCTR,

Jefferson AR. This difference in salt form should not have contributed to any difference in the residue accumulation, distribution, or elimination behavior in fish, however, since the particular anion(s) involved in the acriflavine-anion complex and their degree of dissociation in solution would be determined by the anion composition of the exposure solution. While the anion of the source materials differed, the same anion (chloride) was added to the exposure solution with the unlabeled acriflavine and the water used to make the exposure solution would have had the same anion profile for the two source materials. It would generally be expected that monovalent ion pairs would fully dissociate in

132 aqueous solution and that uptake would involve only the acriflavine cation; i.e., that electrically neutral ion pairs would not exist in the exposure solution.

Overall, in the case of no depuration, the distribution of residue (highest to lowest) was: gall bladder » kidney = liver > upper Gl tract = pyloric caeca = gill > blood = skin » brain > muscle. Residue concentrations had declined markedly after 2 days depuration while the residue concentration within the Gl lumen was considerably increased, apparently due to biliary excretion.

The residue concentration in edible tissues was low after 2 days depuration.

Residue concentrations were higher in skin than muscle. Kidney, gall bladder bile, and lower Gl lumen were the regions showing the highest and most persistent concentrations of residue. The general pattern of distribution of residue after acriflavine exposure was similar to that observed after proflavine exposure.

C. Comparison of acriflavine with proflavine

The general pattern of distribution of residue after acriflavine exposure was similar to that observed after proflavine exposure. Muscle, blood and brain levels in the acriflavine section are about twice the proflavine residue level, although both compounds exhibited a similar overall distribution profiles.

Comparisons of autoradiograms from 2 and 4 day depurated fish showed that the profiles were similar for the two compounds (Figs 4.6 & 4.7 and Figs 4.9 &

133 4.10). In the depurated fish the acriflavine residue levels tended to be higher than with proflavine, and areas of the acriflavine liver (Figs 4.10 & 4.11) showed a non-uniform distribution of radioactivity.

134 Depuration Time, days Tissue 2 4 8

Gall bladder 22,643 4,917 17,979 677 f% of 0 depuration value) (100) (21.7) (79.4) (2.99) Gl lumen, lower 133 6,188 4,039 1,157 (% ofO depuration value) (100) (4.653) (3.037) (870) Gl lumen, upper Ind. pyloric caeca 1,450 3,507 2,373 277 (% of 0 depuration value) (100) (242) (164) (19.1) Kidney, trunk end 2,193 456 194 696 (% of 0 depuration value) (100) (20.8) (8.85) (31.7) Kidney, head end 1,377 289 118 322 (% of 0 depuration value) (100) (21.0) (8.57) (23.4) Liver 1,847 96 424 130 (% of 0 depuration value) (100) (5.2) (23.0) (7.04) Gill 1,340 89 51 34 (% of 0 depuration value) (100) (6.6) (3.8) (2.5) Spleen 826 66 na 22 (% of 0 depuration value) (100) (8.0) (2.7) Skin 386 87 61 62 (% ofO depuration value) (100) (23) (16) (16) Heart (muscle) 344 30 8 8

(% of 0 depuration value) (100) (8.7) (2) r2)

Table 4. 14 to be continued.

135 Table 4.14 (continue)

Blood 417 15 4 5

{Vo of 0 depuration value) (100) (3.6) (1) (1) Muscle (skeletal) 38 2 1 1

(% of 0 depuration value) (100) (5) (3) (3) Brain 72 5 1 1

(% of 0 depuration value) (100) (7) (1) (1) na, not available; not in sections examined.

Table 4.14. Concentration of radioactivity in rainbow trout tissues and fluids at various times after 4 hr exposure to 4 ppm ’‘’C-acriflavine. Concentration values are in arbitrary units as determined from phosphor imaging of 40 |im thick sagittal sections.

136 Figures 4.8. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ‘'^C-acrifiavine. Sections were exposed on the phosphor screen for 28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY: BR, brain; GB. gall bladder; GL, gill; H, heart/blood; I, intestine; K, kidney; L, liver; M, muscle; O, olfactory region; PC, pyloric caeca; SK, skin; SP, spleen.

137 Figure 4.8.

138 Figure 4.9. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ‘'^c-acriflavine and depuration in clean water for two days. Sections were exposed on the phosphor screen for

28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY; same as Figure 4.8, and IL, intestinal lumen.

139 Figure 4.9.

O-/.

0 0 '

140 Figure 4.10. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm '•^c-acriflavine and depuration in clean water for four days. Sections were exposed on the phosphor screen for

141 Figure 4.10.

«■

-ÿ S ' t

142 Figure 4.11. Autoradiograms of sagittal sections from 2 levels in rainbow trout after water exposure for 4 hr to an initial 4.0 ppm ‘''^C-acriflavine and depuration in clean water for eight days. Sections were exposed on the phosphor screen for 28 days and the same print range (5-250) was used for the four figures. The relative intensity of radioactivity is directly comparable among the figures. KEY: same as Figure 4.8, and IL, intestinal lumen; LA, liver accumulation.

143 Figure 4.11.

-Ë cc_ < ' §

144 4.4.4 Main Study:

A. Proflavine

The concentration of proflavine in the exposure water before and after the

4 hr exposure of each group of fish did not decline with repeated use of the solutions, and no detectable concentration of metabolite was observed.

Total Radioactivitv. The concentration of total radioactivity in the fluids and tissues for the various treatments is shown in Table 4.15 and Figure 4.12.

Similar values were observed after 2 and 4 hr exposure to 4 ppm proflavine in all tissues and fluids, except bile, where the concentration after 4 hr was more than twice the 2 hr value. At the end of the 4 hr exposure phase, the concentration of radioactivity in all tissues and fluids was below the exposure water concentration, with the exception of bile at 4 hr where the concentration of radioactivity averaged 5.3 ppm. The rank order of residue concentration at 4 hr

(high to low) was bile > liver > kidney > skin > plasma > muscle. The concentration in muscle averaged 18.7 ppb or 0.47% of the exposure water concentration.

After 2 days depuration, the highest concentration of residue by far was in the bile at 29.5 ppm; it increased to 39.3 ppm after 4 days depuration, and then declined after 8 days depuration to 21.2 ppm. During the 8 day depuration

145 period, the rank order of residue concentration in the tissues was the same as at the end of exposure except that kidney and liver were reversed, with the level in kidney being greater than in liver. Muscle residue was the lowest after 8 days depuration at 1.94 ppb or 0.05% of the concentration in the exposure water. The concentration-time profile of residue in fillet with adhering skin is shown in

Figure 4.13, along with the profiles for skin and muscle separately. The fillet profile lay between the profiles of the component parts, as expected, and closer to the muscle which comprises a larger fraction of the fillet than does the skin.

At 2 hours depuration the concentrations of radioactivity in fillet was higher than expected when compared with the concentration in muscle. It must be remembered, however, that the muscle values in the figure are average values from six fish. Fillets were taken from one female and one male fish from each treatment group randomly. In Treatment III (2 day depuration) the average radioactivity concentration was 4.95 ppb (Table 4.15) for muscle. The radioactivity concentration in the muscle of the female fish (PF28) taken for fillet analysis was 10.4 ppb. The statistical analysis of the total residue data found no influence of gender on the concentration of radioactivity in the fluids and tissues

(p = 0.621).

Proflavine and Its Metabolite. The level of radioactivity was sufficiently high in kidney, liver, and bile to subject these samples to HPLC analysis. Table

4.16. The metabolite was observed in liver and bile samples after most

146 treatments and in kidney only after 4 days depuration. In liver and kidney, the concentration of metabolite was lower than that of proflavine, while in bile the metabolite concentration was higher than that of the parent. Statistical analysis found no influence of gender on the concentration of proflavine in the tissues and fluids in which it was measured (p = 0.681) and no influence on the concentration of the metabolite (p = 0.342).

After 4 hr water exposure of rainbow trout to proflavine, the concentration of proflavine residue was generally lower than the water concentration of proflavine in all tissues and fluids examined except bile. Bile showed residue concentrations after 4 days depuration that were about 10 times the water concentration. The bile residue was primarily the metabolite with the only other component detected being proflavine itself. During depuration, the kidney showed the second highest concentration, which was about 25% to 50% of the exposure water concentration over the 8 day depuration period. The kidney residue was only proflavine during uptake and after 2 and 8 days depuration.

After 4 days depuration, about 28% of the kidney residue was metabolite. The concentration of residue and the concentration of proflavine and its metabolite declined slowly over the 8 day depuration period.

147 Time F, female White M, male Plasma Muscle Skin Kidney Liver Bile Uptake 2 hr F 43.2 ±15.2 15.6 ±3.24 153 ±77.8 519 ±178 739 ±192 2,540 ±1,920 M 63.3 ± 35.0 27.1 ±27.8 152 ±108 1,300 ±358 1,060 ±588 1,150 ±365 mean 53.3 ± 27.8 21.3 ±19.8 152 ±89.9 911 ± 490 900 ± 450 1,840 ±1,510

4 hr F 48.2 ± 5.58 17.3 ±4.02 170 ±82.7 856 ±148 895 ± 35.8 3,200 ±1,180 M 56.7 ± 29.2 20.2 ± 8.31 140 ±90.0 895 ± 245 1,104 ±294 7,280 ± 5,280 00 mean 52.5 ±20.5 18.7 ±6.41 155 ±83.9 875 ±194 1,000 ±227 5,240 ±4,220

Depuration 2 day F 13.3 ±4.38 6.15 ±3.43 210±121 2,170 ±983 623 ± 254 27,000 ±12,600 M 8.65 ±1.97 3.76 ±1.44 151 ±42.0 1,340 ±632 412 ±165 31,900 ±18,900 mean 11.0 ±4.05 4.96 ±2.80 181 ±91.7 1,760 ±899 517 ±232 29,500 ±15,500

Table 4.15 to be continued. Table 4.15 (continue)

4 day F 12.1 ±5.10 5.09±1.18 146 ±41.7 1,450±635 353±204 42,000±15,000 M 9.20 ±3.22 5.67 ±2.02 112 ±54.0 1,250 ± 550 376 ±245 36,600 ±11,000 mean 10.7 ±4.34 5.38 ±1.61 129 ±49.3 1,350 ± 576 364 ±215 39,300 ±12,800

8 day F 5.23 ±3.56 2.47 ±1.27 62.4 ± 4.57 1,340 ±393 128 ±14.8 20,600 ± 20,900 ^ M 3.70 ±0.651 1.59 ±0.558 60.9 ±26.9 1,410 ±934 128 ±50.4 21,600 ± 15,800 co mean 4.31 ± 2.26b 1.94 ±0.958 61.5 ±20.2 1,380 ±734 128 ±38.5 21,200 ± 16,900

Table 4.15. Concentration (ppb) of radioactivity in tissues and fluids at various times during and after 4 hr >vater

exposure of rainbow trout to 4 ppm ’^C-proflavine. Each mean value ± SD represents duplicate determinations for 3

male or 3 female fish; i.e., n = 6 for each gender and n = 12 for combined genders. Sample Gender Proflavine Metabolite Sum Time male; F, female nmole/G nmol/g nmol/G Kidney 2 hr F 0.425 + 0.211 O' 0.425 ± 0.211 M 0.657 ±0.371 O' 0.657 ± 0.371

mean 0.531 ± 0.304 0.531 ± 0.304

4 hr F 0.775 ±0.250 O' 0.775 ±0.250 M 1.18 ±0.500 O' 1.18 ±0.500

mean 0.975 ± 0.431 O' 0.975 ±0.431

2 day F 0.424 ± 0.0994 O' 0.424 ± 0.0994 M 0.408 ±0.168 O' 0.408 ±0.168

mean 0.416 ±0.125 O' 0.416 ±0.125

4 day F 1.09 ±0.410 0.421 ± 0.141 1.51 ±0.545 M 0.794 ±0.303 0.341 ±0.165 1.14 ±0.393

mean 0.942 ±0.377 0.381 ±0.152 1.32 ±0.494

8 day F 0.817 ±0.393” O' 0.817 ±0.393' M 1.93 ±1.87 O' 1.93 ±1.87

mean 1.48 ±1.52” 1.48 ±1.52”

Liver 2 hr F 0.661 ± 0.353 0.225 ±0.126 0.886 ± 0.420 M 0.920 ± 0.684 0.478 ±0.324 1.40 ± 0.971

mean 0.791 ± 0.537 0.351 ± 0.269 1.14 ±0.762

4 hr F 1.31 ±0.403 0.923 ±0.177 2.23 ±0.476 M 1.19 ±0.509 0.803 ± 0.409 1.99 ±0.906

mean 1.25 ±0.442 0.863 ±0.307 2.11 ±0.701

Table 4.16 to be continued.

150 Table 4.16 (continue)

4 day F 0.216 ±0.166 0.245 ±0.198 0.461 ± 0.360 M 0.294 ±0.163 0.344 ±0.193 0.638 ± 0.323

mean 0.255 ±0.162 0.294 ±0.194 0.550 ± 0.339

Bile

4 hr F 2.37 ±0.361 3.54 ±1.40 5.91 ±1.70 M 9.47 ±6.53 12.1 ±3.97 21.6 ±10.3

mean 5.92 ±5.72 7.81 ± 5.34 13.7 ±10.8

2 day F 18.7 ±17.0 38.0 ±18.5 56.7 ±34.8 M 30.3 ±24.8 75.2 ± 37.0 106 ±59.6

mean 24.5 ±21.2 56.6 ± 34.0 81.1 ±53.1

4 day F 28.8 ± 9.27 76.2 ±31.2 105 ±40.1 M 29.3 ± 11.9 59.6 ±21.5 88.8 ±30.1

mean 29.0 ± 10.2 67.9 ± 27.0 96.9 ± 34.8

8 day F 10.7 ±12.1 17.0 ±19.8 27.8 ±31.9 M 12.0 ±9.48 32.1 ±23.9 44.0 ± 33.0

mean 11.3 ±10.4 24.6 ±22.3 35.9 ±32.1 value of 0 indicates that the concentration was below the limit of detection. ‘’PF50 is excluded because of its extra high value.

Table 4.16. Concentration (nmole/G) of proflavine and its metabolite in kidney, liver, and bile of female and male rainbow trout at various times during and after 4 hr water exposure to 4 ppm ^^C-proflavine. Each mean value ± SD represents duplicate determinations for 3 male or 3 female fish; i.e., n = 6 for each gender and n = 12 for combined genders.

151 100000 1 bile

10000 -

kidney Q. § 1000 -

liver !c o 100 - Ü skin

10 - plasma

muscle r 1 2 4 8

Depuration Time, days

Figure 4.12. Concentration of radioactivity (proflavine equivalents) in rainbow trout tissues and fluids at various times after 4 hr water exposure offish to 4 ppm

’^C-proflavine. Each point represents the mean from 3 male and 3 female fish and the bars represent ± SD.

152 100000 -,

10000 -

Q. § 1000 - 1 § OC skin ü 100 -

10 - filet

muscle r T 1 2 4 6

Depuration Time, days

Figure 4.13. Concentration of radioactivity (proflavine equivalents) in rainbow

trout fillet with adhering skin at various times after 4 hr water exposure of fish to

4 ppm ’^C-proflavine. Each point represents the mean from 1 male and 1 female fish and the bars represent ± SD. Also shown are the profiles for muscle and

skin, as they appear in Fig. 4.12.

153 B. Acriflavine.

Studies of the acriflavine profile in the exposure solutions before and after exposure of the rainbow trout showed that two unidentified species were present other than acriflavine (97.7%). The two unidentified species were named as Link

1 (0.3%) and Unk 2 (2.0%) as described previously. The total concentration of radioactivity and the concentration of different species did not change appreciably during the exposure of the fish.

Table 4.17 and Figure 4.14 show the ^^C-acriflavine equivalent (nG/G) in trout tissues and fluids at 0, 8, and 16 days after water exposure to 4 ppm ’'’C- acriflavine for 4 hours. Immediately after exposure (no depuration), bile showed the highest concentration of radioactivity, followed in descending order by liver, kidney, skin, plasma and muscle. After 8 days depuration, the concentration of residue was decreased most rapidly in muscle; to 2.56% of the 0 day value. The

0 day value was relatively low, and the edible tissue contained after 8 days of depuration 3.56 ± 0.550 ppb acriflavine. The rate of depuration during 8 days in the various tissues and fluids was: muscle > plasma > bile > liver > skin > kidney. The skin of trout is part of the fillet. Its depuration of acriflavine was relatively slow, which might affect the minimum depuration time required before consumption.

154 Between 8 and 16 days of depuration the concentration decrease of radioactivity was significant in liver and in bile, decreasing to 0.652% and

0.742%, respectively, of the 0 depuration values. In plasma and white muscle the already low levels decreased to 1.17% and 2.06% of the 0 depuration concentration. The decrease in radioactivity was also pronounced in kidney during this period, from 36.6% of the 0 depuration value to 14.3%. Skin retained after 16 days of depuration 24.6% of the amount taken up during the 4 hours of exposure. The relatively high concentrations of acriflavine in skin and kidney after 16 days of depuration were probably due to irreversible binding. The amount of radioactivity that was not extracted from homogenates was not routinely determined. However, in some cases the radioactivity of the residual homogenates was measured. In skin homogenates 97.4% ± 3.02 of the total radioactivity was not extracted. (The total radioactivity in the skin of Fish AC-28,

-29 and -32 was larger than 1000 dpm/sample, and they were therefore extracted for metabolite profile determination.) In kidney samples of the 16 day depuration group 77.2 ± 32.6%, of the radioactivity was not extracted. The high standard deviation was probably due to the structurally diverse nature of this tissue, and although the kidney was minced and thoroughly mixed, the samples from the same fish contained different amounts of acriflavine. The liver of fish

AC28 contained more than 1000 dpm/sample, and 32.6% of the total activity remained in the homogenate after extraction, which indicated that irreversible binding of acriflavine was less in liver than in kidney and skin.

155 Statistical analysis showed no gender associated differences for the concentration of total residue, acriflavine and its various metabolites, with the exception of Metabolite 1 (Met 1) and the fillet (Tables 4.18 and 4.19).

The data in Tables 4.18 and 4.19 do not show pronounced gender differences for Met 1 and for the fillet total radioactivity. There is a tendency for female fish to show higher levels than males for some tissues. Given the small number of fish studied for the fillet determinations, the borderline statistical significance is possibly not reproducible; it seems to arise from a relatively large concentration of residue in the female fish from Treatment I. Met 1 is a minor metabolite and it was not observed in most samples from Treatments II and III.

When only the data from Treatment I were examined, the gender associated difference for Met 1 was also found (p = 0.018). Plasma showed the most pronounced gender difference for Met 1, Table 4.18.

Comparison of the concentration of radioactivity in the tissues and fluids with the concentration in the exposure solution (nominal 4 ppm acriflavine concentration) the following values were obtained; After exposure the concentration in plasma was 5%, in muscle 3.5%, and in skin 9.5% of that in the exposure water. In kidney it was 1.7 times, in liver 3.3 times, and in bile 8.2 times higher than the concentration in the exposure solution. After 8 days of

156 depuration all tissues and fluids contained less than 4 ppm concentration of total residue.

The level of radioactivity was sufficiently high for HPLC analysis in all

tissues and fluids after exposure (0 day depuration), and it was also high

enough in kidney, liver and bile after 8 days of depuration. After 16 days of

depuration, the following samples had sufficient radioactivity to perform HPLC

analysis: the kidney of all six fish, skin of three fish (AC-28, -29, and -32), liver

and bile of one fish (AC28 and AC32, respectively). The HPLC results are

summarized in Table 4.18. After exposure, peaks of radioactivity that

corresponded to four different metabolites were observed in all tissues and

fluids. Two metabolites were more polar than acriflavine, with retention times 6

and 3 minutes prior to acriflavine (Met 1 and Unk 1 ). Two metabolites (Unk 3

and Unk 2) were less polar than acriflavine, eluting 3 and 5 minutes after the

parent compound. Unk 1 and Unk 2 appeared in the exposure solution and

seemed to be absorbed along with acriflavine. Unk 3 was probably a

metabolite of Unk 2 that was formed by the fish. Unk 3 was not observed in the

later acriflavine metabolic profile study.

Unk 1 was predominant in plasma from female fish, and Unk 2

predominant in plasma from male fish. Muscle, skin and kidney of both genders

had similar metabolite profiles, with Unk 2 and Unk 3 being the dominant

species. Liver and bile contained high concentrations of radioactivity, which

157 was almost evenly distributed among the parent compound and the different

metabolites. The peaks of radioactivity in these samples overlapped; they were

partially separated, but no interpeak fraction had a background count rate. In

bile at 0 day depuration, the peaks of radioactivity of Met 1 and Unk 2

overlapped.

After 8 days the most polar metabolite (Met 1 ) had disappeared from liver and kidney, and the concentration of Unk 1 was low in kidney (ca. 4% of that of acriflavine) and relatively low in liver (ca. 30% of that of acriflavine). The order of depuration rate of the different compounds was similar in kidney and liver; no

Met 1 was found after 8 days, Unk 1 decreased to 4% and 0.62% respectively of its 0 day depuration value, and the acriflavine concentration decreased to

11% and 2.5% respectively of its 0 day depuration value. In kidney, the concentrations of Unk 2 and Unk 3 were a higher fraction of the residue than they were at 0 day depuration. Unk 2 residue disappeared somewhat faster from liver than did acriflavine (1.34 % of the 0 day value) and there was no Unk

3 in liver after 8 days. In bile after 8 days the acriflavine concentration was

3.8% of the 0 day value. The amount of the more polar Met 1 and Unk 1 increased to 115% of the 0 day value, and Unk 2 increased to 140%. Unk 3 in bile was 84% of the 0 day value.

After 16 days of depuration, only kidney showed a measurable metabolite profile. Polar metabolites Unk 1 and Unk 2 were not found or found only in

158 traces. The distribution of total radioactivity among acriflavine, Unk 2 and Unk 3 was about the same as after 8 days depuration. The radioactivity in the skin was irreversibly bound, and no metabolite profile could be determined. The liver of the one fish studied contained 71% of the total activity as Unk 2, and the remainder was evenly distributed between acriflavine and Unk 1. The relative amounts of acriflavine and Unk 2 decreased during the last depuration period, and the less polar Unk 2 was accumulating in liver. After 16 days of depuration, the bile of one fish (AC32) had sufficient radioactivity for measurement of a metabolite profile. Unk 1 was the major component in this sample, with 78.9% of the sum of the metabolites. The other components found after 8 days of depuration were either not present after 16 days (Met 1 and Unk 3) or their relative abundance had decreased: acriflavine from 15.3% to 10.1% and Unk 2 from 27.1% to 10.9%. This conclusion is tentative since the 16 day depuration group values are from only one fish, compared with an average of six fish in the

8 day depuration group.

Fillets from one female and one male of each treatment were measured for total radioactivity. Table 4.19 and Figure 4.15. The average concentration of radioactivity after 16 days of depuration in fillets was 5.16% of the 0 depuration value and 0.224% of the 4 ppm exposure solution. The data for fillets agree with the findings for muscle and skin: at 0 depuration the concentration was higher in females and after 8 days it was higher in males. From each group the fillet of one male and one female fish was measured for concentration of total

159 radioactivity, and Table 4.19 shows the data for the fillets along with the

corresponding concentration data of muscle and skin from the same individual

fish.

After 4 hr exposure to 4 ppm acriflavine, the total residue concentration

was relatively low in plasma, white muscle and skin (0.14-0.38 ppm) and

relatively high in kidney (6.86 ppm), liver (13.3 ppm) and bile (32.9 ppm). After 8

days depuration, total residue in plasma, muscle, bile, and liver was 3.32%,

2.56%, 8.64%, and 11.6% of the respective 0 depuration values; skin and kidney were 32.9% and 36.6% of their 0 depuration values. After 16 days depuration,

the residue concentration declined to 2% or less in plasma, muscle, liver, and

bile while skin and kidney retained 24.6% and 14.3% of their 0 depuration values. Four metabolites of acriflavine were observed using HPLC. At 0 depuration, acriflavine was the dominant species in plasma, muscle, skin and

kidney in contrast to liver and bile, which had one of the metabolites as the largest component. After 8 days depuration, metabolite profiles were observable only in kidney, liver, and bile; acriflavine was present but was only the dominant species in liver. After 16 days depuration, acriflavine was the dominant species only in kidney. Gender-associated differences in total residue, acriflavine, each metabolite, and the sum of acriflavine and its four metabolites were not observed

(p > 0.05), with the exceptions of one minor metabolite and the total residue found in fillet.

160 Depuration F, female Plasma White Skin Kidney Liver Bile

Time, days M, male Muscle

0 F 245 ±131 171 ±103 422 ±187 7817 ±4407 14600 ±8952 34588 ± 32265 M 162 ±92.2 108 ±80.3 344 ±148 5912 ±4316 11992 ±11938 31201 ±21736 mean 204 ±116 139 ±93.7 383 ± 166 6864 ± 4225 13296 ±10140 32895 ± 26287

8 F 6.55 ±2.57 3.31 ± 0.534 123 ±29.2 1829 ±427 1185 ±1598 2183 ±1386 M 7.13 ±2.22 3.94 ± 0.333 130 ±4.21 3540 ±1513 2074 ±2146 3830 ±1725 mean 6.78 ± 2.34 3.56 ± 0.550 126 ±22.1 2513 ±1282 1540 ±1780 2842 ±1669 Percentage of 0 day 3.32% 2.56% 32.9% 36.6% 11.6% 8.64% O) 16 F 2.71 ±2.73 2.88 ± 0.909 91.5 ±73.5 1012 ±665 91.6 ±96.5 258 ±171 M 0.769 ±0.193 2.81 ±0.212 107.3 ±9.68 828 ± 240 61.8 ±14.5 146 ±6.90 mean 2.38 ±2.58 2.87 ± 0.825 94.1 ±66.8 981 ±610 86.7 ± 88.2 240 ±161 Percentage of 0 day 1.17% 2.06% 24.6% 14.3% 0.652% 0.730%

Table 4.17. Concentration (ppb) of radioactivity in tissues and fluids at various times after 4 hr water exposure of rainbow trout to 4 ppm ^'’C-acriflavine. Mean values ± SD represent the following: Treatment I: Duplicate determinations for 3 male and 3 female fish (n = 6 for each gender and n = 12 for combined genders); Treatment II: Duplicate determinations for 2 male and 3 female fish (n = 4 for male fish, n = 6 for female fish and n = 10 for combined genders); Treatment III: Duplicate determinations for 1 male and 5 female fish (n = 2 for male fish, n = 10 for female fish and n = 12 for combined genders). Tissue/ Depur. Gender AC Met 1 Unk 1 Unk 2 Unk 3 Sum Fluid Time M, male ppb ppb ppb ppb ppb ppb Days F, female

Plasma 0 F 55.0 ± 34.8 9.25 ± 5.94 40.8± 26.9 17.7 ±10.8 9.59 ± 3.70 132 ±66.6 M 39.4 ±18.4 1.42 ±1.66 8.41 ± 3.02 25.7 ± 16.9 11.6 ±5.71 86.5 ±35.1 mean 42.6 ± 23.9 4.89 ± 5.90 20.9 ± 22.4 22.0 ±14.8 10.3 ±4.78 100.7 ±49.5

White 0 F 75.1 ±50.3 0.00 ± 0.00 5.94 ± 3.87 50.8 ±15.9 32.7 ±21.2 165 ±89.1 Muscle M 30.8 ±12.0 1.20 ± 5.49 ± 4.03 23.4 ±12.2 14.9 ±6.99 75.8 ± 35.8 0.959 mean 47.5 ± 39.2 0.653 ± 5.32 ± 3.69 34.1 ±17.4 21.6 ±16.8 109 ±73.4 s 0.923 Skin 0 F 146 ±87.4 1.60 ±3.92 10.1 ±15.0 70.6 ± 32.9 62.7 ± 28.6 291 ± 165 M 102 ±65.2 3.12 ±2.69 11.4 ±5.44 51.1 ±27.9 37.5 ± 16.4 205 ±108 mean 107 ±51.0 1.70 ±2.50 8.23 ± 6.69 54.7 ± 23.2 44.4 ±17.6 216 ±88.5

Kidney 0 F 2199 ± 54.4 ±21.5 300 ±149 2142 ±1176 951 ±511 5648 ± 2549 1110 M 1154 ±371 37.7 ±12.9 203 ±156 1326 ±880 599 ±315 3319 ±1478 mean 1693 ± 46.6 ±19.8 266 ±154 1576 ±973 729 ± 435 4310 ±2360 1005

Table 4.18 to be continued. Table 4.18 (continue).

8 F 225 ±116 0.00 ± 0.00 6.52 ±7.17 422 ± 84.7 99.9 ± 24.0 753 ±166 M 281 ±48.1 0.00 0.00 17.7 ±20.5 287 ±110 122 ±62.9 708 ±149 mean 251 ±101 0.00 ± 0.00 10.6 ±15.0 354 ±111 115 ±40.0 730 ±161

16 F 97.1 ±90.3 0.00 ± 0.00 3.63 ±7.90 95.4 ± 77.3 54.0 ± 57.7 253 ±195 M 139 ±23.8 0.00 0.00 0.00 ± 0.00 124 ±34.3 39.8 ± 56.2 303 ±1.88 mean 104 ±83.7 0.00 ± 0.00 3.03 ± 7.28 100 ±71.5 54.2 ± 55.3 262 ±177

Liver 0 F 2492 ± 1833 ±972 2929 ± 1612 ±987 1637 ±976 10502 ±6220 S 1503 1855 M 1987 ± 1828± 2583 ± 1241± 1013 1202 ±1240 8841 ± 8309 1583 1732 2783 mean 2097 ± 1738± 2540 ± 1348 ±979 1301 ±1057 9024 ± 7011 1481 1363 2239 8 F 25.6 ± 48.3 0.00 ± 0.00 7.55 ±18.5 13.0 ±31.9 0.00 ± 0.00 46.1 ±98.6 M 78.5 ±100 0.00 ± 0.00 24.3 ± 30.2 21.1 ±23.0 0.00 ± 0.00 124 ±151 mean 51.9 ±75.9 0.00 ± 0.00 15.9 ±24.7 18.1 ±28.6 0.00 ± 0.00 85.8 ± 125

16 F" 6.91 ± 0.59 0.00 ± 0.00 5.19 ±7.34 29.9 ±42.2 0.00 ± 0.00 42.0 ± 49.0

Bile 0 F 18848± 16796± 0.00 ± 0.00“ 6219 ±3635 3191 ±2668 45053 ±41692 20893 15948“

Table 4.18 to be continued. Table 4.18 (continue). M 12489± 15045± 0.00 ± 0.00“ 5885 ± 2524 2349 ±2140 35767 ±21957 10027 10397= mean 12547± 13820± 0.00 ± 0.00= 6016 ±3132 2420 ±2109 34803 ± 26851 12332 11130= 8 F 2668 ± 5607 ± 7419 ± 6999 ±4315 1887 ±2589 24581 ± 17275 1847 2268 6999 M 7046 ± 8177 ± 10037± 9398 ± 4484 1769 ±784 36426 ±12997 6119 2896 2206 mean 4767 ± 6870 ± 9030 ± 8431 ± 4289 2044 ±1988 31143 ±15954 4535 2778 5560 2 16 F° 63.3 ± 7.07 0.00 ± 0.00 490 ± 13.7 67.8 ± 25.2 0.00 ± 0.00 621 ±46.0 Radioactive peaks for Met 1 and Unk 1 collapsed b. Data only from Fish#28, since the other fish did not contain enough radioactivity for measuring metabolite profile Data only from Fish#32, since the other fish did not contain enough radioactivity for measuring metabolite profile

Table 4.18. Concentration (ppb acriflavine equivalent) of acriflavine and its metabolites in fluids and tissues of female and male rainbow trout at various times after 4 hr water exposure to 4 ppm ^'’C-acriflavine. Mean values ± SD represent the following: Treatment I: Duplicate determinations for 3 male and 3 female fish (n = 6 for each gender and n = 12 for combined genders); Treatment II: Duplicate determinations for 2 male and 3 female fish (n = 4 for male fish, n = 6 for female fish and n = 10 for combined genders); Treatment III: Duplicate determinations for 1 male and 5 female fish (n = 2 for male fish, n = 10 for female fish and n = 12 for combined genders) and see footnotes. Treatment Gender Concentration Concentration Concentration (Fish ID) Fillet Muscle Skin ppb ppb ppb

1 M (AC26) 88.7 ±2.23 63.0 ±0.126 226 ± 0.384

F (AC24) 260 ± 14.2 178 ±2.21 385 ±14.9

mean 174 ±90.5 121 ±81.3 306± 112

II M(AC17) 31.1 ±3.86 4.16 ±0.046 130 ±3.93

F(AC18) 15.7 ±0.974 3.64 ± 0.333 114 ±24.7

mean 23.4 ± 8.55 3.90 ± 0.368 122 ±11.3

III M (AC29) 6.11 ±0.528 2.81 ±0.212 107 ±9.68

F (AC28) 11.8 ±0.624 3.66 ± 0.371 205 ± 41.4

mean 8.97 ± 3.07 3.24 ± 0.601 156 ±69.3

Table 4.19. Concentration of radioactivity in muscle, skin and fillet - composed of muscle and skin in their natural proportions - from individual male and female rainbow trout. (Average ± SD concentrations are the mean of 1 male and 1 female fish, 5 determinations per fillet and 2 determinations each of muscle and skin samples.)

165 100000 1 bile

10000 - kidney

Q. liver Q. 1000 - g 1 I skin o 100 ■ o plasma

muscle

10 ■

1 -* 0 2 4 6 8 10 12 14 16 Depuration Time, days

Figure 4.14. Concentration of radioactivity (acriflavine equivalents) in rainbow trout tissues and fluids at various times after 4 hr water exposure of fish to 4 ppm

’^C-acriflavine. Each point represents the mean from 3 male and 3 female fish and the bars represent ± SD.

166 1000 1

skin I 100 - §

filet Io Ü 10 ■

muscle

1 T 2 4 6 8 10 12 14 16 Depuration Time, days

Figure 4.15. Concentration of radioactivity (acriflavine equivalents) in rainbow trout fillet with adhering skin at various times after 4 hr water exposure of fish to

4 ppm ’^C-acriflavine. Each point represents the mean from 1 male and 1 female fish and the bars represent ± SD. Also shown are the profiles for muscle and skin, as they appear in Fig. 4.14.

167 CHAPTER 5

CHARACTERIZATION OF PROFLAVINE METABOLITES IN RAINBOW TROUT

5.1. ABSTRACT

Proflavine has potential for use as an antiinfectlve in fish and its metabolism by rainbow trout was therefore studied. After their separation by reverse-phase HPLC with uv detection at 262 nm, three metabolites were found in liver, and one metabolite was found in plasma 14 hr after lA bolus administration of 10 mg/kg of proflavine. Treatment with hydrochloric acid converted the three metabolites to proflavine, which suggested that the metabolites were proflavine conjugates. Treatment with p-glucuronidase and saccharic acid 1,4-lactone, a specific p-glucuronidase inhibitor, revealed that two metabolites were proflavine glucuronides. For determination of uv-vis absorption and mass spectra, HPLC-purified metabolites were isolated from liver. Data from these experiments suggested that the proflavine metabolites were 3-N-glucuronosyl proflavine (PG), 3-N-glucuronosyl, 6-N-acetyl proflavine

168 (APG), and 3-N-acetyl proflavine (AP). The identities of the metabolites were

verified by chemical synthesis. When compared with synthetic PG and AP, the

two metabolites isolated from trout had the same molecular weight as

determined by matrix-assisted, laser desorption ionization (MALDI), time-of-flight

mass spectrometry. Additionally, they co-eluted on HPLC under different mobile

phase conditions. The in vitro incubation with liver subcellular preparations

confirmed this characterization and provided the evidence that APG can be

formed from glucuronidation of AP or acétylation of PG. Finally, mass balance

study showed a 90% recovery of the administered dose.

5.2. INTRODUCTION

Proflavine (3,6-diamino acridine), an acridine derivative, was used as a topical anti-infective in the two World Wars. While proflavine is no longer approved for human use because of its toxicity, it has a potential to treat lice in fish. Proflavine is a symmetrical compound and the acridine ring system could be oxidized by acridine dehydrogenase to form acridone [1]. Conjugation at the two amino groups by a number of endogenous substances, such as glucuronic acid, sulfate, and amino acids is also possible. Additionally, the acridine ring nitrogen is vulnerable to conjugation.

169 The objective of the present study was to explore the metabolism of proflavine by fish. We characterized the metabolites of proflavine to better understand its biochemical fate, pharmacokinetics, and total residue depletion in proflavine exposed fish. Proflavine was administered to rainbow trout via lA bolus injection. After isolation and purification of proflavine metabolites, acid hydrolysis, enzymatic treatment, absorption spectra, TOF-MS and chemical synthesis were used to identify the metabolites. Finally, an in vitro study and mass balance were used to elucidate the metabolic pathways.

This chapter emphasized the methodology used to identify the metabolites of proflavine from fish. From the mass spectra, acriflavine had similar metabolites with proflavine. Because of the limited quantity of acriflavine metabolites, not all the experiments used for proflavine metabolites identification were performed for acriflavine.

5.3. EXPERIMENTAL

5.3.1. Reagents and Materials

3-glucuronidase, saccharic acid 1,4-lactone, D-glucuronic acid, UDP- glucuronic acid, and acetyl coenzyme A were purchased from Sigma Chemical

Co. (St. Louis, MO). BCA protein assay kits were from Pierce (Rockford, IL)

170 For other materials and reagents, refer to Chapters 1 to 4.

5.3.2. Instruments

Time of flight mass spectrometer Kompact MALDI III was obtained from

Kratos Analytical (Ramsey, New Jersy). Spectrophotometer, KONTRON

UVIKON 860 was obtained from KONTRON Instruments (Everett, MA).

Centrifuge Model J2-21 and Ultracentrifuge Model L7-65 were obtained from

Beckman Instruments (Palo Alto, CA).

For other instruments, refer to Chapters 1 to 4.

5.3.3. Metabolite Isolation

Details of the animal protocol and fish cannulation procedure were presented in Chapters III and IV. Trout were anesthetized with tricaine methanesulfonate (0.1 g/liter) and then were fitted with a cannula (28-G thin wall

Teflon tubing, Zeus Industrial Products, Rariton, NJ) in the dorsal aorta [2, 3], which allowed lA bolus injection and blood sampling from free-swimming fish.

171 After 24 hr of recuperation, each fish was administered a single lA injection of proflavine (10 mg of proflavine/kg body weight, 10 mg/ml in 80% v/v DMSO and

20% v/v water). Sixteen hours after dosing, blood samples were removed from the cannula. The blood was collected in a heparinized Ependorf tube, centrifuged, and the plasma portion was separated and stored in a -20°C freezer until analyzed by HPLC. After blood withdrawal, the fish was sacrificed immediately. Liver samples were then collected and stored in a -20°C freezer until analysis.

Sample clean-up procedures were described in Chapter 2. Metabolites were separated by HPLC; and several fractions of each metabolite were collected. The mobile phase component was evaporated under a stream of nitrogen or using a vacuum centrifuge evaporator. The purified metabolites were stored in a -20°C freezer.

5.3.4. HPLC Analysis

The HPLC system consisted of a Model 125 programmable binary gradient pump. Model 166 UV-VIS detector set at 262 nm, Pharmacia Model

FRAC-100 fraction collector, IBM model 55SX PC computer, and System Gold control system (Beckman Instruments, Irvine, CA). Samples were introduced

172 into a 100 (il sample loop. Separation was achieved on a 4 nm Nova-Pak C-18

column, 3.9 mm x 300 mm (Waters, Millipore Corporation, Milford, MA) and

eluted using the following gradient mobile phase: initial, from 5 min of 100%

system A; then a 8 min linear gradient to 75% system A (24% system B); then hold 75% system A for 5 min before returning to 100% system A. The flow rate was 0.8 ml/min. Mobile phase A consisted of 93.23% v/v of water, 5.00% v/v of acetonitrile, 1.40% v/v of acetic acid, and 0.37% v/v of triethylamine. Mobile phase B consisted of 4.91 % v/v of water, 95.00% v/v of acetonitrile, 0.07% v/v of acetic acid, and 0.02% v/v of triethylamine.

5.3.5. Acid Incubations

Metabolites were dissolved in water. To a 200 jil aliquot of each solution was added 10 nl of IN HOI (final acid concentration: 0.05N). The solution was incubated for 30 min at 45°C and the reaction was terminated by an addition of

10 (il of IN NaOH. An aliquot of the solution was then injected onto the HPLC.

173 5.3.6. Enzyme Incubations

Metabolites were dissolved in 0.1 M sodium phosphate buffer (pH 6.8).

To a 200 III aliquot of each metabolite solution was added 100 pi of 13- glucuronidase solution (1000 units/ml in sodium phosphate buffer) and 100 pi of the sodium phosphate buffer. After the mixture was incubated at 37°C for 24 hr, the reaction was terminated by processing the sample using SPE tubes. The eluate was analyzed by the previously described HPLC procedure. A control experiment in which 100 pi of the sodium phosphate buffer was added to the metabolite solution instead of the enzyme solution using the same procedure as described above was also performed. For the p-glucuronidase inhibition experiment, 100 pi of 100 mM saccharic acid 1,4-lactone was added prior to addition of the p-glucuronidase solution into the enzymatic reaction mixture [4].

5.3.7. Absorption Spectra

The extracted metabolites were dissolved in 0.1 M sodium phosphate buffer at pH 7.0 and pH 12; pH was adjusted using NaOH. A 0.5 ml aliquot of each solution was placed in a cuvette and the UV-VIS spectrum between 200 nm and 600 nm was measured.

174 5.3.8. Mass Spectrometry

The extracted metabolites (about 50 to 100 pmole) were reconstituted with water and loaded onto a 20-well sample slide for TOF-MS analysis. After the samples were dried, and 3 jil of saturated sinapinic acid in 30:70 VA/ of acetonitrile:water solution was added to each sample well. After drying with a stream of air, the slide was placed into the sample compartment of the TOF-MS.

The positive ion mode was selected and a laser power of 60 to 80 mV was used to obtain the molecular ion peak.

5.3.9. Chemical Synthesis

Acetvloroflavine (AP): Sodium acetate, 10 mg, was dissolved in 5 ml 0.1 mM proflavine aqueous solution. To the solution was added 200 \x\ of acetic anhydride and the mixture was incubated at 60°C for 2 hr. After dilution, a 100 nl aliquot was injected onto the HPLC. The AP fraction was collected and the mobile phase component was evaporated under a stream of nitrogen. The residue was subjected to TOF-MS analysis as before.

175 Proflavine alucuronide (PG): D-glucuronic acid (0.64 g, 0.0033 mole) and proflavine (0.37 g, 0.0015 mole) were placed in a round-bottom 200 ml flask equipped with a reflux condenser and a magnet stirrer. A mixture of 13.5 ml acetone and 1.5 ml of water was added with vigorous stirring. The suspension was heated to 45°C, and then a 0.1 ml aliquot of concentrated hydrochloric acid was added slowly. After 10 min, the reaction mixture was placed into an ice bath for an hour, the precipitate was isolated, and then dissolved in 3 ml of water.

The solution was mixed with 20 ml of acetone, and the precipitate was filtered and washed with 2 ml of ethylacetate and 1 ml of ether. Approximately 50 mg of product was obtained [5]. Approximately 1 ng of this product was dissolved in water and the solution was injected onto the HPLC. The PG fraction was collected and the mobile phase component was evaporated under a stream of nitrogen. The residue was subjected to TOF-MS analysis as before. It was also mixed with the metabolites extracted from liver and the mixture was injected onto the HPLC at three different HPLC mobile phase conditions: mobile phase B gradient up to 10%, 15% and 25%.

5.3.10. In vitro Metabolism of Proflavine

Preparation of the liver subcellular fraction. Fresh trout liver was minced into small pieces, which were mixed with 10 mM potassium phosphate buffer (pH

176 7.4), that contained 1.15% (w/v) potassium chloride. The tissue was then

homogenized on ice with the Omni Mixer Homogenizer. The resultant homogenate contained 0.2 ~ 0.25 g of wet tissue per ml of buffer; it was centrifuged at SOOOg, 4°C for 20 min. The supernatant (the S9 fraction) was separated from the pellet. One aliquot of the S9 fraction was stored at -80 °C.

Another aliquot of the S9 fraction was used for the preparation of the cytosol and the microsomal fraction. After the S9 fraction was centrifuged further at

105,000g, 4°C for 60 min, the supernatant (cytosol fraction) was obtained. The microsomal fraction was obtained by reconstituting the pellet in 6 ml of KCI- phosphate buffer [6]. Both the cytosol and the microsomal fraction were stored at -80°C. Protein concentrations of the S9, cytosol, and microsomal fractions were determined by the Bicinchoninic Acid (BCA) method [7].

Formation of AP. PG. and APG from proflavine. To a 0.2 ml aliquot of the

S9 fraction was added 0.4 ml of 15 mM Tris-HCI buffer (pH 7.4) containing 100 nM of proflavine, 0.2 ml of the same Tris-HCI buffer containing 3 mM of acetyl

Co-A, and 0.2 ml of the sam e Tris-HCI buffer containing 5 mM of UDP-glucuronic acid. The mixture was incubated at 12°C for 3 hours. Two controls were also prepared. Control One was the complete mixture without acetyl Co-A and UDP- glucuronic acid. Control Two was the complete mixture without the S9 fraction.

The controls and the sample were incubated similarly. The reaction was

177 terminated by SPE. All samples were analyzed by the HPLC procedures as described previously.

Formation of APG from PG. To a 0.2 ml aliquot of the cytosol fraction was added 0.4 ml of 15 mM Tris-HCI buffer (pH 7.4) containing 100 nM of PG and 0.2 ml of the same Tris-HCI buffer containing 3 mM of acetyl Co-A. The mixture was incubated at 12°C for 3 hours. The reaction was terminated by the SPE extraction. The sample was analyzed by the HPLC procedures as described previously.

Formation of APG from AP. To a 0.2 ml aliquot of the microsomal fraction was added 0.4 ml of 15 mM Tris-HCI buffer (pH 7.4) containing 100 nM of AP and 0.2 ml of the same Tris-HCI buffer containing 5 mM of UDP-glucuronic acid.

The mixture was incubated at 12°C for 3 hours. The reaction was terminated by

SPE, and the SPE eluate was analyzed by the HPLC procedures as described previously.

5.3.11. Mass Balance Study

Two Trout were anesthetized with tricaine methanesulfonate (0.1 g/liter) and then were fitted with a cannula. After 24 hr of recuperation, each fish was

178 transferred into a 50 liter glass aquarium which contained 36 liters of water.

Each fish was then administered a single lA injection of proflavine (80 pmole of proflavine/kg body weight, -20 mg/ml in 80% v/v DMSO and 20% v/v water).

Sixteen hours after dosing, both fish were removed from the aquarium and were frozen immediately. Water samples of each aquarium were collected before and after dosing, and right after the fish was removed. Each fish was cut into 5-7 pieces and was homogenized in a five-liter blender by adding equal weight of

0.1 M pH 4.4 sodium acetate buffer that contained 1 M NaCI. Ten aliquot of samples were collected for each homogenate. Five samples were digested immediately for total radioactivity analysis. The other five samples were stored in a -20°C freezer until analyzed by HPLC. The procedures for digestion and

HPLC analysis were the same as described in Chapters II to IV. Water samples were also analyzed by total radioactivity counting and HPLC.

5.4. RESULTS

5.4.1. HPLC Analysis

The HPLC chromatograms of plasma and liver samples for proflavine treated fish are shown in Figure 5.2 and 5.3. One metabolite was found in plasma and three metabolites were found in liver. According to their retention

179 behaviors on HPLC. these three metabolites found in liver were arbitrarily named Met I, Met II, and Met III; only Met I was found in plasma. Met I and Met

II had retention times shorter than proflavine, which suggested that they were more polar than proflavine, while the retention time of Met III was similar to that of proflavine. The HPLC chromatograms of plasma and liver samples for acriflavine treated fish are shown in Figure 5.4 and 5.5. Three metabolites were found in plasma and liver. According to their retention behaviors on HPLC, these three metabolites were arbitrarily named Met I', Met 11’, and Met III". Met I’ and Met II’ had retention times shorter than acriflavine, which suggested that they were more polar than acriflavine, while the retention time of Met III’ was the same as that of proflavine.

5.4.2. Acid Incubation

After acid treatment, proflavine was released from all three metabolites.

Almost 100% of Met I was converted to proflavine and 80% of Met III was converted to proflavine. Interestingly, acid hydrolysis of Met II gave not only proflavine, but also a compound with the same retention time as Met III. This fraction was collected for MS analysis. The results showed that the [M+lf peak was at m/z 252.5, which was the same as that of Met III.

180 OH OH

3-N-li-D-glucuronosyl proflavine (Met I, PG)

HO OH .A NH OH

Proflavine (3,6-diamino acridine) 3-N-B-D-glucuronosyl-6-N-acetylprollavine (Met n,A PG)

3-N-acetylproflavine (Met m, AP)

Figure 5,1. The scheme for the formation of proflavine metabolites from rainbow trout. Met I 0.9 -

< S c CO -e

0.3 -

- 0.1 0 510 15 20 25

Retention Time, min

Figure 5.2. Typical HPLC chromatogram of proflavine metabolites extracted

from plasma.

182 2.2 n

Met

< Met c8 (0 € 8 5 ^ 0.6 -

Met I

0.2 ■

- 0.2 0 510 15 20 25 30

Retention Time, min

Figure 5.3. Typical HPLC chromatogram of proflavine metabolites extracted from liver.

183 2.2 1

Met

? Met sc I o (0 a ^ 0.6 -

Met

0.2 -

- 0.2 0 5 10 15 20 25 30

Retention Time, min

Figure 5.4. Typical HPLC chromatogram of acriflavine metabolites extracted

from plasma.

184 1.9 T

1.5 -

< 8 c (0 € o 0 .7 " < Met II' 0.3 --

- 0.1 0 5 10 15 20 25 Time, min

Figure 5.5. Typical HPLC chromatogram of acriflavine metabolites extracted from liver.

185 5.4.3. Enzyme Incubation

After 24 hr Incubation with p-giucuronidase, 87% of Met I converted to proflavine. However, In a control experiment without enzyme, 38% of Met I generated proflavine (Figure 5.6 and 5.7). In the presence of 20 mM saccharic acid 1,4-lactone, a specific p-glucuronldase Inhibitor, the enzyme reactivity was almost totally Inhibited (Figure 5.8). Similar results were obtained for the enzyme treatment of Met II. After 24 hr Incubation with p-glucuronldase, 98% of

Met II was converted to Met III (Figure 5.9). However, In the control experiment where there was no enzyme, 38% of Met II was converted to Met III (Figure

5.10). The enzyme reactivity was also totally Inhibited In the presence of 20 mM saccharic acid 1,4-lactone (Figure 5.11).

5.4.4. Absorption Spectra

The absorption spectra of the three metabolites and proflavine showed two main absorption bands; Xma% 262 nm and 454 nm at pH 7.0 (Figure 5.12a).

For all metabolites, the 454 nm band shifted to 395 nm at pH 12.0, as did proflavine (Figure 5.12b). This shift corresponded to Ionization of the acridine ring (pKa = 9.6). When the acridine ring Is protonated, the low energy band Is at

454 nm, which shifts to 395 nm when It Is not protonated [8]. Because the ring

186 nitrogen appeared to protonate at low pH for all the metabolites, none of the metabolites involved substitution on the acridine ring nitrogen.

5.4.5. Mass Spectrometry

The TOF-MS spectrum of Met I gave an MH* ion at m/z 386.3, which essentially matched the calculated MH* value of proflavine glucuronide at m/z

386.4. Met II gave an MH* ion at m/z 428.3 whereas the calculated MH"" value of acetylproflavine glucuronide was at m/z 428.4. Met III gave an MH* ion at m/z

252.5, essentially the same as 252.3 calculated for acetylproflavine. The TOF-

MS spectrum of Met I’ gave an MH"” ion at m/z 400.4, which essentially matched the calculated MH"^ value of acriflavine glucuronide at m/z 400.4. Met II’ gave an

MH* ion at m/z 442.5 whereas the calculated MH* value of acetylacrifiavine glucuronide was at m/z 442.5. In all cases, the observed values were within less than 0.1% of the calculated values (Table 5.1). A representative TOF mass spectrum is shown in Figure 5.13.

187 0.5 1

0.4 -

0.3 - 3 s I 0.2 - 8 < Met I

0.0 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.6. Effect of enzymatic treatment on Met I: Met I was incubated with 3-glucuronidase for 24 hr, 87.3% of proflavine (P) was liberated.

188 0 .9 -I

0.7-

Met I < 0.5 - 8 -e S ^ 0.3-

- 0.1 0 5 10 15 20 25

Retention Time, min

Figure 5.7. Effect of enzymatic treatment on Met I; Control, Met I was

Incubated in buffer (with no p-glucuronidase) for 24 hr, 38.1% of proflavine (P) was liberated.

189 0.4 1

0.3 -

Met I

D 0.2 - < 8 (0C o 0.1 - Ui <

0.0 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.8. Effect of enzymatic treatment on Met I: Inhibition, Met I was

Incubated in with p-glucuronidase and 20 nM saccharic acid 1,4-lactone

for 24 hr, 40.6% of proflavine (P) was liberated.

190 0.4 1

0.3 -

< 0.2 - Met cs CO € g <

0.0 -

- 0.1 510 150 20 25

Retention Time, min

Figure 5.9. Effect of enzymatic treatment on Met II: Met II was Incubated

with P-glucuronidase for 24 hr, 98.3% of Met III was liberated.

191 0.7-1

Met 0.5 -

D <

0.3 - S Met <

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.10. Effect of enzymatic treatment on Met II: Control, Met II was

Incubated in buffer (with no p-glucuronidase) for 24 hr, 38.0% of Met III

was liberated.

192 0.6 -I

0.5 - Met 0.4 -

< 8 0.3 - c I 8 0.2 - Met

0.0 -

- 0.1 5 10 15 20 250

Retention Time, min

Figure 5.11. Effect of enzymatic treatment on Met II: Inhibition, Met II

was incubated in with p-glucuronidase and 20 nM saccharic acid 1,4-

lactone for 24 hr, 40.6% of Met III was liberated.

193 0.8

0.6 8 c CO € 0.4 I 0.2

200 250 300 350 400 450 500 550 600 W ave Length, nm

0.6 -1 (b)

0.5 -

8 0.4 c (0 0.3 -

I 0.2 -

0.1 -

0.0 -I

200 250 300 350 400 450 500 550 600 Wave Length, nm

Figure 5.12. Absorption spectra of proflavine, (a): at pH 7.0, (b); at pH 12.0.

194 %feM. we-i

«-

TO-

#0 M T .7

4 0 10 M 10- 100.7 0 vJWvAvnJ t—'—I—' • ■ t~ 4 f\tyi,, ^ r|ii

Figure 5.13. Representative TOF-MS spectrum, Met II from fish liver. The mass

of the MH* ion of Met II was measured at 428.3 amu with 80 averaged laser

shots. Sinapinic acid was used as the matrix. The mass of 207.7 amu and

225.3 amu was from the matrix and also acted as internal standard.

195 5.4.6. Chemical Synthesis

Acetylproflavine. After 2 hr acétylation reaction, approximately 80% of

proflavine was converted to acetylproflavine. Upon HPLC analysis, the synthetic

acetylproflavine gave the same retention time as Met III isolated from fish. The

identity was also verified by TOF-MS, which gave a MH* peak at m/z 252.5

(calculated 252.30) (Table 5.1).

Proflavine glucuronide. Approximately 50 mg of crude product was obtained, which appeared as reddish-brown powder. Upon HPLC analysis, the crude product contained 60% proflavine glucuronide and 30% proflavine (Figure

5.14). The coelution experiment showed that at three different mobile phase compositions, the synthetic proflavine glucuronide co-eluted with Met I isolated from fish liver (Figure 5.15-5.17). The molecular weight of the synthetic PG was determined by TOF-MS, which gave a MH^ peak at m/z 386.3 (calculated 386.4)

(Table 5.1).

5.4.7. In vitro Metabolism

The protein concentration was measured using the BCA method and using bovine serum albumin (BSA) as the standard. The protein concentrations

196 of the S9. cytosol, and microsomal fraction were found to be 20.7, 13.5, and 19.5

mg/ml, respectively. In the presence of liver S9 fraction, UDP-GA, and acetyl

Co-A, all three metabolites were produced with AP being the most abundant

(Figure 5.18). However, no metabolites were found for the two control

experiments, which Indicated that both the liver subcellular fraction and the high energy substrate (UDP-GA and acetyl Co-A) were essential for the in vitro

proflavine metabolism. APG was found to be generated from AP by glucuronldatlon and from PG by acétylation (Figure 5.19 and 5.20).

5.4.8. Mass Balance Study

From total radioactivity analysis, 99.9% of the total administered dose was recovered; 92.0% was from the fish and 7.86% was from the water. The total proflavine residues observed In fish and In water were 37.4 ± 1.37 nmole/g

(n = 10) and 60.8 ± 9.57 pmole/ml, respectively. The relative abundance of proflavine and its distribution was examined by HPLC analysis. The results showed that an average of 90.5% recovery of residues resulted from the sum of

PG, APG, AP and P. The percentage distribution of each metabolite and proflavine was listed on Table 5.2. Proflavine and all the three metabolites were also observed In water. However, they were not quantified because of the limit of detection.

197 0.6 1

0.5 -

0.4 - PGgys

< 0.3 - 8 c I 0.2 -

0.0 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.14. HPLC analysis of the synthetic proflavine glucuronide

(PG): approximately 60% was PG, 30% was proflavine.

198 0.9 -

0.7 - < 8 c ■eCO 0.5 - S < 0.3 -

0.1

- 0.1 0 10 155 20 25 Retention Time, min

Figure 5.15. HPLC chromatograms of co-elution of synthetic PG with

Met I from trout liver extract; mobile phase gradient to 25% of mobile

phase B.

199 0.4 1

0.3 ■

Met § . 0.2 - 8 c nCO S <

0.0 - LA

- 0.1 5 10 15 200 25 Retention Time, min

Figure 5.16. HPLC chromatograms of co-elution of synthetic PG with

Met I from trout liver extract: mobile phase gradient to 15% of mobile

phase B.

2 0 0 0.4 1

0.3 -

Met < 0.2 -

0.0

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.17. HPLC chromatograms of co-elution of synthetic PG with

Met I from trout liver extract; mobile phase gradient to 10% of mobile

phase B.

2 0 1 0.6 n

0.5 -

0.4 ■ Z) < 8 0.3 - (0c € S 0.2 - Met <

Met I

0.0 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.18. Synthesis of proflavine metabolites in vitro: formation of

Met I, Met II, and Met III from proflavine.

2 0 2 0.7 1

0.5 - Met

c8 (D 0.3 - ■e 8 <

- 0.1 0 105 15 20 25 Retention Time, min

Figure 5.19. Synthesis of proflavine metabolites in vitro: formation of

Met II from PG.

203 AP 0.9 -

0.7 - < 8 0.5 - Met II

0.3 -

- 0.1 0 5 10 15 20 25 Retention Time, min

Figure 5.20. Synthesis of proflavine metabolites in vitro; formation of

Met II from AP.

204 ID Molecular Mass Error Observed % Calculated

AP, synthesis 252.5 252.30 0.079

PG. synthesis 386.3 386.39 0.079

Met III from Met II 252.5 252.30 0.079

hydrolysis

Met 1 386.3 386.39 0.023

Met III 428.3 428.43 0.030

Met III 252.5 252.30 0.079

Met r 400.4 400.41 0.002

Met ir 442.5 442.45 0.011

Table 5.1. TOF-MS result and relative error.

205 Time Fish MB1 Fish MB2

water water

T = 0 hr (total, nmole) 0 (total, nmole) 0

Fish Fish

(total, nmole) 29305 (total, nmole) 24323

Water Water

(total, nmole) 2512 (total, nmole) 1868

Fish Fish

T = 14 hr (total, nmole) 27307 (total, nmole) 22094

P ± SD, % 50.5 ± 3.26 P ± SD, % 39.5 ± 3.39

Met 1 ± SD, % 15.6 ±0.874 Met 1 ± SD, % 12.5 ±1.29

Met II ± SD, % 26.16 ±2.27 Metll±SD, % 21.7 ±2.42

Metlll ± SD, % 7.68 ± 0.601 Metlll ±SD, % 15.7 ±1.64

total ± SD, % 86.3 ± 6.07 total ± SD, % 94.7 ±1.89

Table 5.2. The distribution of proflavine and its metabolites in mass balance study.

206 5.5. DISCUSSION

From the HPLC analysis, two polar and one less polar metabolites were observed from proflavine treated fish. Acid hydrolysis of all the metabolites released proflavine, which suggested that all the metabolites were proflavine conjugates. Moreover, only proflavine was released from Met I and Met III, but both proflavine and Met III were released from Met II after acid hydrolysis. This suggested that Met I and Met III were proflavine monoconjugates, and that Met II was a proflavine diconjugate that shared one common ligand with Met III.

Known xenobiotic conjugates found in fish include those formed with glucose, sulfate, amino acids, glutathione, acetate, and glucuronic acid [9].

Chemical synthesis of proflavine glucoside and comparison of its HPLC retention time with those of the metabolites revealed that the glucoside was not the isolated metabolite. Treatment of the metabolites with sulfatase did not release proflavine. Amino acid conjugation was not likely because proflavine does not have a carboxylic group, which is perquisite for amino acid conjugation.

Glutathione conjugation was also not possible because proflavine does not have a electrophilic center, which is essential for glutathione conjugation. Among the usual conjugates, that left only glucuronidation and acétylation as candidates for proflavine metabolism.

207 Treatment of the metabolites with p-giucuronidase and its specific inhibitor, saccharic acid 1,4-lactone provided evidence that Met I and Met II were proflavine glucuronides. That Met I was converted to proflavine after this treatment suggested that Met I was proflavine conjugated with glucuronic acid.

However, Met II was converted to Met III after enzymatic treatment, which indicated that Met II was a proflavine diconjugate in which one ligand was glucuronic acid and the other was the same as the ligand of Met III. Possible conjugation sites of proflavine included the 3- and 6-amino groups. The ring nitrogen probably remained intact in all metabolites, based on the pH dependence of their absorption spectra.

TOF-MS analysis of three metabolites identified Met I as proflavine glucuronide (3-N-glucuronosyl proflavine, PG), Met II was identified as acetylproflavine glucuronide (3-N-glucuronosyl, 6-N-acetyl proflavine, APG), and

Met III was identified as acetylproflavine (3-N-acetyl proflavine, AP). Therefore,

Met I and Met III were proflavine monoconjugates, and Met II was a proflavine diconjugate. These conclusions were supported by the results of acid hydrolysis and enzymatic treatments.

The identity of the metabolites was further verified by chemical synthesis of Met I and Met III. Based on HPLC and TOF-MS analysis, the synthetic AP was indistinguishable from Met III, and the synthetic PG was indistinguishable

208 from Met I. The mass spectrum of Met III gave a molecular ion Identical to that

of synthetic AP, indicating that Met III was AP. Proflavine glucuronide from

chemical synthesis was found to coelute with Met I at three different mobile

phase compositions, and gave the same molecular weight by TOF-MS analysis,

indicating that Met I was PG.

These three metabolites were also formed after in vitro incubation of

proflavine with trout liver preparations in the presence of UDP-GA and Acetyl

Co~A. Met II (APG) was formed from Met I (PG) by acétylation, and from Met III

(AP) by glucuronidation. Therefore, Met II can be formed from either pathways

(Figure 5.1), i.e., glucuronidation of P followed by acétylation of PG, or

acétylation of P followed by glucuronidation of AP.

Mass balance study showed a 90.5% recovery from the sum of PG, APG,

AP and P. The unrecovered 9.5% residues may be a tightly bound fraction that was not extracted, and a result of errors from analysis. The possibility that the

unrecovered fraction came from unidentified metabolites still exists.

In summary, the combined results of this study provided unequivocal evidence that the three metabolites of proflavine found in rainbow trout following

intra-arterial administration are 3-N-glucuronosyl proflavine (Met I, PG), 3-N- glucuronosyl proflavine 3-N-glucuronosyl (Met II, APG), and 6-N-acetyl

209 proflavine (Met III, AP). Moreover, The pathways of metabolite formation were also elucidated. The two metabolites found in acriflavine treated fish were acriflavine glucuronide and acetylacriflavine glucuronide.

2 1 0 5.6. REFERENCES

1. Albert A: Vertebrate pharmacology and toxicology of acridlnes. Action on

enzymes, carcinogenicity and anticancerogenicity. In “The Acridines:

Their Preparation, Physical, Chemical, and Biological Properties and

Uses”, pp. 504-519, St. Martin’s Press, New York, 1966.

2. Hoelton GF and Randall DG: The effect of hypoxia upon the partial

pressure of gases in blood and water afferent and efferent to the gills of

rainbow trout. J. Exp. Biol. 46, 317-322 (1967).

3. Houston AR: Blood and circulation. In “Methods for Fish Biology”

(Schreck CB and Moyle PB, éd.), pp. 273-322, American Fisheries

Society, Bethesda, Maryland, 1990.

4. Lewy GA: Baicalinase, a plant p-glucuronidase. Biochem. J. 58, 462-9

(1954).

5. Kovacs A, Liptak A, Nanasi P, Janossy L, Csemus I, Erdei J, Kaszab I,

Polya K, and Neszmelyi A: Novel aminoacridine-a- and/or -p- (D)-N-

glycoside and aminoacridine-a- and/or -p-(L)-N-glycoside derivatives.

Canadian Patent. No. 1201117 (1984).

2 1 1 6. Rodrigues AD, Ferrero JL, Amann MT, Rotert GA, Cepa SP, Surber BW,

Machinist JM, Tich NR, Sullivan JP, Garvey DS, Fitzgerald M and Americ

SP: The in vitro hepatic metabolism of ABT-418, a cholinergic channel

activitor, in rats, dogs, cynomolgus monkeys, and humans. Drug

Metab. and Dispos. 22, 788-798 (1994).

7. Morton RE and Evans TA: Modification of the bicinchoninic acid protein

assay to eliminate lipid interference in determining lipoprotein protein

content. Anai Biochem. 204,332-4(1992).

8. Acheson RM and Orgel LE: Visible and ultraviolet absorption spectra of

acridine and its derivatives. In “The Chemistry of Heterocyclic

Compounds: Acridines”. Interscience Publishers, Inc., New York, 1956.

9. Sieber SM and Adamson RH: The metabolism of xenobiotics by fish. In

“Drug Metabolism - from Microbe to Man” (Parke DV and Smith RL, ed.),

pp. 233-246. Taylor & Francis Ltd, London, Great Britain, 1977.

2 1 2 BIBILIOGRAPHY

Acheson RM and Orgel LE; Visible and ultraviolet absorption spectra of acridine and its derivatives. In “The Chemistry of Heterocyclic Compounds: Acridines”. Interscience Publishers, Inc., New York, 1956.

Albert A: Vertebrate pharmacology and toxicology of acridines. Action on enzymes, carcinogenicity and anticancerogenicity. In “The Acridines: Their Preparation, Physical, Chemical, and Biological Properties and Uses”, pp. 504- 519, St. Martin’s Press, New York, 1966.

Altria KD, Simpson CF. Analysis of some pharmaceuticals by high voltage capillary zone electrophoresis. J. Pharm. Biomed. Anal. 1988; 6(6-8): 801-7.

Boudinot FD, Jusko WJ. Fluid shifts and other factors affecting plasma protein binding of prednisolone by equilibrium dialysis. J Pharm Sci 1984; 73:774-780.

Boxenhaum HG, Riegelman A, Elashoff RM. Statistical estimations in pharmacokinetics. J Pharmacok Biopharm 1974; 2:123-48.

Bungay PM, Dedrick RL, Guarino AM. Pharmacokinetic modeling of the dogfish shark (Squalus acanthias): distribution and urinary and biliary excretion of phenol red and its glucuronide. J Pharmacok Biopharm 1976; 4: 377-88.

Duff DW, Fitzgerald D, Kullman D, Lipke DW, Ward J, Olson KR. Blood volume and red cell space in tissues of the rainbow trout, Salmo gairdneri. Comp Biochem & Physio 1987; 87(2): 393 - 398.

Fisher, A. Test of some antiseptic acridine dyes. Fresenius' Z. Anal Chem 1975; 276(5):383-4.

Gibaldi M , Perrier D. Pharmacokinetics, 2nd ed., Marcel and Dekker, Inc., New York/Basel. 1982; 92-94.

213 Giebelmann R, Nagel s, Brunstein C, Scheibe E. Thin-layer chromatographic detection of quaternary ammonium compounds of toxicological-chemical relevance. Zentralbl Pharm, Pharmakother Laboratoriumsdiagn 1976; 115(4)1339-46.

Hoelton GF, Randall DJ. The effect of hypoxia upon the partial pressure of gasses in blood and water afferent and efferent to the gills of rainbow trout. J Exp Biol 1967; 46: 317-22.

Hofman R, Lommel R. Effects of repeated blood sampling on some blood parameters in fresh water fish. J Fish Biol 1984; 24: 245 - 251.

Houston AR. Blood and circulation. In: Methods for fish biology 1990, edited by Schreck CB and Moyle PB. American Fisheries Society, Betheda, MD, 273- 322.

Kaka JS, Hayton, WL. Temperature and surfactant dependence of accumulation of 4-aminopyrine and ethanol in fish. J Pharm Sci 1978; 67:1558-63.

Karara AH, Hayton WL. Pharmacokinetic model for uptake and disposition of di- 2-ethylhexyl phthala in sheepshead minnow (cyprynodon variegatus). Aquat Toxic1984; 5, 181-95.

Kovacs A, Liptak A, Nanasi P, Janossy L, Csemus I, Erdei J, Kaszab I, Polya K, and Neszmelyi A: Novel aminoacridine-a- and/or -p- (D)-N-glycoside and aminoacridine-a- and/or -p-(L)-N-glycoside derivatives. Canadian Patent. No. 1201117(1984).

Lewy GA: Baicalinase, a plant p-glucuronidase. Biochem. J. 58, 462-9 (1954).

McKim JM, Schmieder P, Veith G. Absorption dynamics of organic chemical transport across trout gills as related to octanol-water partition coefficient. Toxicol AppI Pharmacol 1985; 77: 1-10.

Morton RE and Evans TA: Modification of the bicinchoninic acid protein assay to eliminate lipid interference in determining lipoprotein protein content. Anal Biochem. 204, 332-4 (1992).

Oettingen WF. The therapeutic agents of quinoline groups 1933. The Chemical Catalog Company, Inc., New York; 239-261.

Olson KR. Distribution of flow and plasma skimming in isolated perfused gills of three teleosts. J Exp Biol 1984; 109:97-108.

214 Peacocks AR. The interaction of acridines with nucleic acids. In: Acridine, edited by Acheson RM 1973, Interscience Publishers, New York, pp. 723-758.

Rodrigues AD, Ferrero JL, Amann MT, Rotert GA, Cepa SP, Surber BW, Machinist JM, Tich NR, Sullivan JP, Garvey DS, Fitzgerald M and Americ SP: The in vitro hepatic metabolism of ABT-418, a cholinergic channel activitor, in rats, dogs, cynomolgus monkeys, and humans. Drug Metab. and Dispos. 22, 788-798 (1994).

Schelhom T, Kretz S, Zimmermann HW. Reinvestigation of the binding of proflavine to DNA. Is intercalation the dominant binding effect? Cell and Mol Biol 1992. 38(4):345-365.

Schultz IR, Hayton WL. Toxicokinetics of trifluralin in rainbow trout. Aquat Toxicol 1993; 26:287-306.

Sieber SM and Adamson RH: The metabolism of xenobiotics by fish. In “Drug Metabolism - from Microbe to Man” (Parke DV and Smith RL, ed.), pp. 233-246. Taylor & Francis Ltd, London, Great Britain, 1977.

Wagner JG. Fundamentals of clinical pharmacokinetics, 1st ed. Drug Intelligence Pub., Inc. Hamilton, Illinois 1975.

Zhang G, Li G, Li Z. Determination of acriflavine by thin-layer chromatographic densitometry. Yiyao Gongye 1986; 17(9):422-3.

215