Viewing the 3.5-11.0 Min

INFORMATION TO USERS

The most advanced technology has been used to photo graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer.

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 copyrighted material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are re produced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge.

Photographs included in the original manuscript have been reproduced xerographically in this copy. 35 mm slides or 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.

Accessing theUMI World's Information since 1938

300 North Zeeb Road, Ann Arbor Ml 48106-1346 USA Order Number 8824619

Studies on the metabolism of merbarone in man

Supko, Jeffrey G., Ph.D.

The Ohio State University, 1988

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48'06 PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages______

2. Colored illustrations, paper or print______

3. Photographs with dark background_____

4. Illustrations are poor copy______

5. Pages with black marks, not original copy ^

6. Print shows through as there is text on both sides of page______

7. Indistinct, broken or small print on several pages

8. Print exceeds margin requirements______

9. Tightly bound copy with print lost in spine______

10. Computer printout pages with indistinct print______

11. Page(s)______lacking when material received, and not available from school or author.

12. Page(s) seem to be missing in numbering only as text follows.

13. Two pages num bered . Text follows.

14. Curling and wrinkled pages 'S

15. Dissertation contains pages with print at a slant, filmed as received

16. Other Studios on the, Metabolism of Merbarono in Han

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Jeffrey G. Supko, B.S.

•jV Vr Vr «sV Vr

The Ohio State University

1988

Dissertation Committee: Approved by

Louis Malspeis, Ph.D.

Alfred E. Staubus, Ph.D

Robert E. Notari, Ph.D.

Louis Malspeis, Advi Division of Pharmaceutics and Pharmaceutical Chemistry College of Pharmacy DF.DTCATION

To my wife and parents.

ii ACKNOWLEDGEMENTS

I wish to express my most sincere appreciation to Dr. Louis Malspeis

for his invaluable guidance throughout the course of my graduate studies. He is truly a preeminent scientist and has been a great

inspiration to continually enhance my knowledge and research proficiency. Gratitude is also expressed to Drs. Robert E. Notari and

Alfred E. Staubus who have significantly contributed to my development as a graduate student. Furthermore, it has not been forgotten that through his earnest recruiting efforts, Dr. Notari first intoduced me to both the discipline of Pharmaceutical Chemistry and my wife, for which I am most appreciative. The technical assistance of Mr. Michael T. Fox and Ms. Margaret E. Lyon is gratefully acknowledged. VITA

September 23, 1960 Born - Hackensack, New Jersey

December, 1983 B.S. cum laude - Double Major Chemistry and Chemistry/Business University of Scranton

Sept. 1983 - Aug. 1987 Graduate Teaching Assistant

Sept. 1987 - present Berlex Fellow College of Pharmacy The Ohio State University Columbus, Ohio

Publications

"Pharmacokinetics and Metabolism of Merbarone (NSC 336628) in Patients." L. Malspeis, A. E. Staubus, J. G. Supko, M. R. Grever and E. H. Kraut. Proc. Am. Assoc. Cancer Res., 28, 228 (1987).

"Spectral Characterization and Quantitation of the Principal Metabolites of Merbarone (NSC 336628) in the Urine of Patients." J. G. Supko, M. R. Grever, S. P. Balcerzak, E. H. Kraut, A. E. Staubus and L. Malspeis. Proc. Am. Assoc. Cancer Res., 29, 189 (1988).

"Merbarone (NSC 336628) Enhances the Urinary Excretion of Uric Acid of Cancer Patients." M. R. Grever, E. H. Kraut, J. G. Supko, R. W. Trewyn, A. E. Staubus and L. Malspeis. Proc. Am. Assoc. Cancer Res., 29, 189, (1988).

Field of Study

Major Field Pharmaceutics and Pharmaceutical Chemistry

iv TABLE OF CONTENTS

PAGE

DEDICATION ...... ii

ACKNOWLEDGEMENTS ...... iii

VITA ...... iv

LIST OF TABLES ...... viii

LIST OF FIGURES ...... xi

INTRODUCTION ...... 1

CHAPTER

I. CHARACTERIZATION OF MERBARONE METABOLITES IN MAN ...... 3

A. Introduction ...... 3

B. Materials and Methods...... 6

1. Patient Urine and Plasma Samples ...... 6 2. Chemicals and Materials ...... 6 3. Reagent and Solvent Purification ...... 8 4. Instrumentation ...... 9 5. High Performance Liquid Chromatography ...... 10

a. Apparatus ...... 10 b. Sample Preparation ...... 10 c. Chromatographic Conditions ...... 10

6 . Chemical Syntheses ...... 10 7. Extraction of Metabolites from Patient Urine ...... 15 8 . Acetylation of Phenolic Metabolites Isolated from Urine ...... 16

C. Results and Discussion ...... 17

1. The Metabolism of Functionally Related Xenobiotics 17 2. Synthesis of Potential Oxidative Metabolites ...... 19 3. Metabolite Characterization ...... 22

v 2. Synthesis of Potential Oxidative Metabolites ...... 19 3. Metabolite Characterization ...... 22

D. References ...... 33

II. ANALYSIS OF MERBARONE AND TTS METABOLITES IN HUMAN URINE BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ...... 110

A. Introduction ...... 110

B. Materials and Methods ...... Ill

1. Reagents and Chemicals ...... Ill 2. Apparatus ...... 112 3. Standard Solutions ...... 113 4. Optimization of Reaction Conditions ...... 114 5. Sample Preparation ...... 115 6 . Chomatographic Conditions ...... 116 7. Quantitation ...... 116 8 . Relative Recovery ...... 116 9. Stability of the Acetate Derivatives ...... 116

C. Results and Discussion ...... 117

D. References ...... 128

III. EXCRETION OF MERBARONE AND ITS MAJOR METABOLITES IN HUMAN URINE ...... 166

A. Introduction ...... 166

B. Materials and Methods ...... 169

1. Acquisition of Human Urine and Plasma Specimens ... 169 2. Analytical Method ...... 169 3. Plasma Sample Preparation ...... 169

C. Results and Discussion ...... 170

D. References ...... 175

IV. XANTHINE OXIDASE INHIBITION STUDIES ...... 190

A. Introduction ...... 190

B. Materials and Methods ...... 191

1. Reagents and Chemicals ...... 191 2. Apparatus ...... 192 3. Stock Solutions ...... 192 4. Standard Solutions ...... 193 5. Spectrophotometer Parameters ...... 194

vi C. Results and Discussion ...... 196

D. References ...... 203

BIBLIOGRAPHY ...... 212

vii I.TST OF TABLES

Reaction yields and appearance of the synthesized 5-carboxanilide barbiturates |1| and |3|...... 37

HPLC retention and UV spectral data of the 5-carboxanilide barbiturates I and 3. Tsocratic, C^g reverse-phase chroma tography was performed at ambient temperature employing a mobile phase of 25% methanol (v/v), 0.1 M acetate buffer, 40 mM magnesium sulfate and 1 mM SOS at pH 5.5; flow rate 1.0 mL/min...... 38

^11 NMR spectral data of the 5-cnrboxani.li.de barbiturates _1 and 3 ...... 39

1 3 C NMR chemical shifts (6 ) for the 5-carboxanilide barbiturates J. and 3 ...... 40

Electron impact mass spectral data of the synthesized 5-carboxanilide barbiturates 1 and 3 ...... 41

Chemical ionization mass spectral data of the synthesized 5-carboxanilide barbiturates 1 and 3 ...... 42

Elemental analyses for the 5-cnrhoxani1 ide barbiturates.... 42

Conditions for the HPLC analysis of merbarone and related compounds...... 43

Sample preparation for HPLC...... 44

Quasi-molecular ions in the chemical ionization mass spectrum of the urine isolate...... 45

Quasi-molecular ions in the chemical ionization mass spectrum of the acetylated urine isolate...... 45

Exchangeable proton assignments in the 500 MHz NMR spectrum of the urine isolate...... 46

viii 1.13 Aromatic proton assignments in the 500 MHz NMR spectrum of the urine isolate...... 46

2.1 Optimization of the reaction conditions for the acylation of 4'-hydroxy analogs of merbarone in aqueous solution with acetic anhydride and potassium carbonate...... 129

2.2 Urine sample preparation for the assay of merbarone and its metabolites...... 130

2.3 Conditions for the HPLC analysis of merbarone and its metabolites in urine...... 131

2.4 Backcalculated concentrations for 4'-0H-2-oxo-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 4'-0H-2-oxo-MB using 1 /peak area ratio squared as the weighting factor, during a 3.5 month period...... 132

2.5 Backcalculated concentrations for 4'-0H-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 4'-0H-MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period. 133

2.6 Backcalculated concentrations for 2-oxo-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 2-oxo-MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period. 134

2.7 Backcalculated concentrations for MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the con centration of added MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period...... 135

2.8 Relative recovery and reproducibility of the analytical method for 4'-0H-2-oxo-MB in human urine...... 136

2.9 Relative recovery and reproducibility of the analytical method for 4'-0H-MB in human urine...... 136

2.10 Relative recovery and reproducibility of the analytical method for 2-oxo-MB in human urine...... 137

ix 2.11 Relative recovery and reproducibility of the analytical method for MB in human urine...... 137

2.12 Stability of the acetate derivatives of phenolic merbarone metabolites in the assay solution at ambient temperature... 138

3.1 Cumulative urinary excretion of MB and its metabolites during 7 days by patients on the 24 h continuous i.v. infusion daily x 5 schedule...... 177

3.2 Cumulative urinary excretion of MB and its metabolites during 7 days by patients on the 2 h i.v. infusion daily x 5 schedule...... 178

4.1 Reaction mixture composition, conditions and spectrophoto meter parameters used for the determination of the initial fractional inhibition of xanthine oxidase...... 2C5

4.2 Reaction mixture composition, conditions and spectrophoto meter parameters used for the acquisition of time courses for the conversion of xanthine to uric acid by xanthine oxidase...... 206

4.3 Beer's Law curve for the absorbance of uric acid determined under conditions similar to that of the xanthine oxidase inhibition experiments...... 207

4.4 Mean initial fractional inhibition of xanthine oxidase by allopurinol, MB, 4'-0H-MB, 2-oxo-MB and 4'-0H-2-oxo-MB under the conditions specified in Table 4.1...... 208

x LIST OF FIGURES

FIGURE PAGE

1.1 The chemical structure of merbarone...... 47

1.2 Some xenobiotics that are functionally related to MB . 47

1.3 The synthesis of 4-acetoxypheny1 isocyanate 6 ...... 48

1.4 Chromatogram of patient plasma obtained on day 5 during the 24 h civ of 1000 mg/m^/d x 5 MB (8 . 8 min) determined with a diodtf'-array detector at 306 n m ...... 49

1.5 Three-dimensional chromatogram from 0-15 min of patient plasma obtained on day 5 during treatment with MB, 1000 mg/m /d x 5. Chromatographic peaks: 7.4 min; MB, 8 . 8 min. 50

1.6 Plot of the three-dimensional chromatogram from 6-11 min with enhanced resolution of the patient plasma sample obtained on day 5 during the infusion of MB, 1000 mg/m /d x 5. Chromatographic peaks: metabolite, 7.4 min; MB, 8 . 8 m i n ...... 51

1.7 Three-dimensional chromatogram from 0-15 min of pre-dose patient plasma showing the absence of constituents which would interfere with the detection of MB and the metabolite eluting at 7.4 m i n ...... 52

1.8 Chromatogram of 24 h pooled patient urines obtained on day 5 during the 24 h civ of MB at 1000 mg/m /d x 5 determined with a diode-array detector at 306 nm. Chromatographic peaks: metabolites, 1.3, 1.7, 3.6 and 7.5 min; MB, 8.9 m i n ...... 53

1.9 Three-dimensional chromatogram from 0-15 min truncated at 100 mAU of 24 h pooled patient urines obtained on day 5 during the infusion of MB at 1000 mg/m /d x 5. Chromatographic peaks: metabolites, 1.3, 1.7, 3.6 and 7.5 min; MB, 8.9 m i n ...... 54

xi 1 . 10 Plot of the three-dimensional chromatogram from 6-11 min at greater sensitivity for the 24 h pooled patient urines obtained on day 5 during treatment. Chromatographic peaks: metabolite, 7.4 min; MB, 8 . 8 m i n ...... 55

1.11 Three-dimensional chromatogram from 0-15 min truncated at 100 mAU of patient urine obtained prior to dosing...... 56

1 . 12 Plot of the three-dimensional chromatogram from 0-2.5 min at 1000 mAU resolution for 24 h pooled patient urines obtained on day 5 during the infusion of MB, 1000 mg/m /d x 5. Chromatographic peaks: urinary constituents, 0.8 min; metabolites, 1.3 and 1.7 m i n ...... 57

1. 13 Plot of the three-dimensional chromatogram from 0-2.5 min at 1000 mAU resolution for the pre-dose patient urine sample. Chromatographic peaks: urinary constituents, 0.8 m i n ...... 58

1. 14 Procedure for the extraction of merbarone and its metabolites from patient urine...... 59

1. 15 Chromatogram of the urine isolate determined at 293 nm with a diode-array detector. Chromatographic peaks:... metabo lites, 1.3, 1.7, 4.0 and 7.8 min; MB, 9.2 m i n ...... 60

1 . 16 Three-dimensional chromatogram from 0-15 min of the urine isolate. Chromatographic peaks: metabolites, 1.3, 1.7, 4.0 and 7.8 min; MB, 9.2 m i n...... 61

1. 17 Plot of the 1.0-2.25 min region of the urine isolate three- dimensional chromatogram. Chromatographic peaks:...metabo lites, 1.3 and 1.7 m i n...... 62

1 . 18 Chromatogram of synthesized 4 1-0H-2-oxo-MB (1.2 min) determined at 293 nm ...... 63

1. 19 Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 4'-0H-2-oxo-MB and the metabolite eluting at 1.3 min in the urine isolate...... 64

1.20 Chromatogram of synthesized 4'-0H-MB (1.7 min) determined at 293 n m ...... 65

1.21 Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 4'-0H-MB and the metabolite eluting at 1.7 min in the urine isolate...... 66

xii 1.22 Chromatogram of synthesized 2-oxo-MB (7.8 min) determined at 293 n m ...... 67

1.23 Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 2-oxo-MB and the metabolite eluting at 7.8 min in the urine isolate...... 68

1.24 Chromatogram of synthesized MB (9.1 min) determined at 293 n m ...... 69

1.25 Overlay of the UV spectra determined at the apexes of chromatographic peaks of the compound eluting at 9.2 min (I) in the urine isolate and synthesized MB (II)...... 70

1.26 Overlay of the UV spectra determined at the apexes of chro matographic peaks of synthesized MB (I) and 4'-0Me-MB (II). 71

1.27 Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 2-oxo-MB (I) and 4'-0Me-2-oxo-MB (II)...... 72

1.28 Chromatogram of the urine isolate treated with AC 2 O-K 2 CO 3 for phenolic acetylation determined at 293 nm with a diode- array detector. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4'-0Ac-MB, 6.2 min; 2-oxo-MB, 7.9 min; MB, 9.2 m i n ...... 73

1.29 Three-dimensional chromatogram from 0-15 min of the urine isolate treated with AC 2 O-K 2 CO 3 . Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4'-0Ac-MB, 6.2 min; 2-oxo-MB 7.9 min; MB, 9.2 m i n ...... 74

1.30 Plot of the three-dimensional chromatogram from 3.5-11.0 min with enhanced resolution for the AC 2 O-K 2 CO 3 treated urine isolate. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5; 4' -OAc-MB, 6.2; 2-oxo-MB, 7.9; MB, 9.2 m i n ...... 75

1.31 Overlay of the UV spectra determined at the upslope (4.36 min), apex (4.56 min) and downs lope (4.76 min) for the peak eluting at 4.6 min in the chromatogram of AC2 O-K 2 CO 3 treated urine isolate...... 76

1.32 Overlay of the UV spectra determined at the upslope (5.90 min), apex (6.17 min) and downslope (6.40 min) for the peak eluting at 6 . 2 min in the chromatogram of Ac^O-I^COg treated urine isolate...... 77

1.33 Overlay of the UV spectra determined at the upslope (7.61 min), apex (7.85 min) and downslope (8.14 min) for the peak eluting at 7.9 min in the chromatogram of AC2 O-K 2 CO 3 treated urine isolate...... 78

xiii 1.34 Chromatogram of synthesized 4 1 -OAc-2-oxo-MB (4.3 min) determined at 293 n m ...... 79

1.35 Overlay of the UV spectra determined at the apexes of chromatographic peaks of the compound eluting at 4.5 min for the AC2 O-K 2 CO 3 treated urine isolate and synthesized 4'-OAc-2-oxo-MB...... 80

1.36 Chromatogram of synthesized 4'-0Ac-MB (6.0 min) determined at 293 n m ...... 81

1.37 Overlay of the UV spectra determined at the apexes of chromatographic peaks of the compound eluting at 6 . 2 min for the AC2 O-K 2 COU treated urine isolate and synthesized 4' -OAc-MB...... 82

1.38 Electron impact mass spectrum of synthesized MB ...... 83

1.39 Electron impact mass spectrum of synthesized 2-oxo-MB...... 84

1.40 Electron impact mass spectrum of synthesized 4'-0H-MB...... 85

1.41 Electron impact mass spectrum of synthesized 4 f-0H-2-oxo- MB ...... 86

1.42 Electron impact mass spectrum of the urine isolate. 87

1.43 Chemical ionization (CH^) mass spectrum of the urine isolate...... 8 8

1.44 Chemical ionization (CH^) mass spectrum of synthesized MB.. 89

1.45 Chemical ionization (CH^) mass spectrum of synthesized 2-oxo-MB...... 90

1.46 Chemical ionization (CU. ) mass spectrum of synthesized 4 ' -OH-MB...... * ...... 91

1.47 Chemical ionization (CH^) mass spectrum of synthesized 4'-OH-2-oxo-MB...... 92

1.48 Electron impact mass spectrum of the AC 2 O-K 2 CO 3 treated urine isolate...... 93

1.49 Electron impact (70 eV) mass spectrum of synthesized 4' -0Ac2-oxo-MB...... 94

1.50 Electron impact (70 eV) mass spectrum of synthesized 4 ' -OAc-MB...... 95

xiv 1.51 Chemical ionization (Cli^) mass spectrum of the AC2 O-K 2 CO3 treated urine isolate...... 96

1.52 Chemical ionization (CH^) mass spectrum of synthesized 4' -0Ac-2-oxo-MB...... 97

1.53 Chemical ionization (CH^) mass spectrum of synthesized 4 ' -OAc-MB...... 98

1.54 The 500 MHz NMR of synthesized MB from 0.0-14.0 ppm 99

1.55 Plot of the 500 MHz *H NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized MB ...... 100

1.56 The 500 MHz *H NMR of synthesized 2-oxo-MB from 0.0-14.0 p p m ...... 101

1.57 Plot of the 500 MHz ^H NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized 2-oxo-MB.. 102

1.58 The 500 MHz ^H NMR of synthesized 4'-0H-MB from 0.0-14.0 p p m ...... 103

1.59 Plot of the 500 MHz *H NMR from 6 .6-7.7 ppm showing an expansion of the aromatic region for synthesized 4'-0H-MB.. 104

1.60 The 500 MHz *H NMR of synthesized 4'-0H-2-oxo-MB from 0.0-14.0 ppm...... 105

1.61 Plot of the 500 MHz *H NMR from 6 .6-7.7 ppm showing an expansion of the aromatic region for synthesized 4'-OH-2-oxo-MB...... 106

1.62 The urine isolate 500 MHz ^H NMR spectrum from 0.0-14.0 p p m ...... 107

1.63 Plot of the urine isolate 500 MHz *H NMR spectral region from 6.6-14.0 ppm with assignments for the exchangeable protons...... 108

1.64 Plot of the urine isolate 500 MHz ^H NMR spectrum from 6.6-7.7 ppm showing an expansion of the aromatic region and peak assignments...... 109

2.1 Chromatogram at 293 nm of 24 h pooled patient urine, collected on day 5 during the 24 h civ administration of MB, 1000 mg/m^/d x 5. Chromatographic peaks: 4'-0H-MB, 1.7 min; 2-oxo-MB, 7.6 min; MB, 8.9 m i n ...... 139

xv 2.2 Chromatogram at 293 nm of a spiked aqueous sample contain ing synthetic 4'-0H-2-oxo-MB, 1.4 min; 4'-0H-MB, 1.8 min; 2-oxo-MB, 7.9 min; and MB, 9.2 m i n ...... 140

2.3 Liquid chromatograms of 100 yg/mL spiked aqueous solutions. (A) 4 ' -0H-MB, 1.8 min; (B) 4'-0Ac-MB, 5.3 min; (C) 4'-0H-MB ( lOOyL) alkalinized with 1.0 M KOH (100 yL) and injected 5 min after the addition of AC2 O (25 y L ) ...... 141

2.4 Liquid chromatograms of aqueous 4'-0H-MB (100 yg/mL) treated with an equivalent volume of aqueous base (100 yL) and 25 yL of AC 2 O. (A) Sodium bicarbonate, 1.0 M (B) potassium carbonate, 1.0 M. Chromatographic peaks: 4'-0H- MB, 1.8 min; 4 ’-OAc-MB, 4.8-5. 1 m i n ...... 142

2.5 Chromatograms of patient urine collected during the infu sion MB. (A) Assayed directly; (B) injected after treating with an equivalent volume of 1.0 M l^COo (100 yL) and 25 yL of Ac^O. The indicated peaks are: (1) 4'-0Ac-2-oxo-MB; (2) 4 -OAc-MB; (3) 2-oxo-MB; (4) MB ...... 143

2.6 Chromatograms of derivatized patient urine samples deter mined under the same conditions with the exception of the methanol content of the mobile phase. (A) 25% (v,v) meth anol; (B) 30% (v,v) methanol; (C) 35% (v,v) methanol. The indicated peaks are: (1) 4'-0Ac-2-oxo-MB; (2) 4'-OAc-MB; (3) 2-oxo-MB; (4) M B ...... 144

2.7 Chromatogram at 293 nm of an aqueous sample, spiked with MB and its major urinary metabolites, determined with diode- array detection after acetylation of the phenolic components with AC2 O-K 2 CO 3 . Chromatographic peaks 4 ' -0Ac-2-oxo-MB, 4.5 min; 4 -OAc-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 m i n ...... 145

2.8 Three-dimensional chromatogram of the spiked aqueous sample containing MB and its major urinary metabolites treated with AC2 O-K2 CO 0 . Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4 -OAc-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 m i n ...... 146

2.9 Plot of the three-dimensional chromatogram from 3.0-11.0 min with enhanced resolution of the spiked sample treated with AC 2 O-K2 CO 0 . Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4 -OAc-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 m i n ...... 147

2.10 Three-dimensional chromatogram of the spiked aqueous sample containing MB and its major urinary metabolites assayed directly. Chromatographic peaks: 4 1 -0H-2-oxo-MB, 1.4 min; 4'-0Ac-MB, 1.8 min; 2-oxo-MB, 7.9 min; MB, 9.2 mi n ...... 148

xv i 2.11 Chromatograms of drug-free urine spiked with synthetic samples of (A) 4'-OAc-2 -oxo-MB (5.2 min) and 4'-OAc-MB (7.1 min) assayed directly and (B) 4'-OH-2-oxo-MB and 4'-OH-MB chromatographed after dorivatization. The peak at 15.4 min is the internal standard. The concentration of each analyte was 6 yg/mL...... 149

2.12 Overlaid UV spectra determined at the apex of the chroma tographic peak of synthetic 4'-OAc-2-oxo-MB and the peak at 4.5 min in the sample spiked with MB and metabolites after acetylation...... 150

2.13 Overlaid UV spectra determined at the apex of the chroma tographic peak of synthetic 4'-OAc-MB and the peak at 6.1 min in the sample spiked with MB and metabolites after acetylation...... 151

2.14 The 250 MHz NMR spectrum of the reaction product formed upon treating 4'-0H-2-oxo-MB with AC 2 O-K2 CO 2 ...... 152

2.15 The 250 MHz *H NMR spectrum of the reaction product formed upon treating 4'-OH-MB with AC 2 O-K2 CO 2 ...... 153

2.16 Overlaid UV spectra determined at the apex of the chroma tographic peak of reference 2-oxo-MB and the peak at 7.7 min in the sample spiked with MB and metabolites after acetylation...... 154

2. 17 Overlaid UV spectra determined at the apex of the chroma tographic peak of reference MB and the peak at 9.0 min in the sample spiked with MB and metabolites after acetyla tion...... 155

2.18 Three-dimensional chromatogram truncated at 300 mAU of day 5 pooled urine, collected during the continuous infusion of MB at 1000 mg/m /d x 5, treated with AC 2 O-K 2 CO 2 . Chromatographic peaks: 4 1 -0Ac-2-oxo-MB, 4.4 min; 4'-OAc- MB, 5.9 min; 2-oxo-MB, 7.6 min; MB, 8 . 8 m i n ...... 156

2.19 Plot of the three-dimensional chromatogram from 3.0-11.0 min of the day 5 urine sample treated with AC 2 O-K2 CO 2 at greater sensitivity. Chromatographic peaks: 4'-0Ac-2- oxo-MB, 4.4 min; 4'-OAc-MB, 5.9 min; 2-oxo-MB, 7.6 min; MB, 8 . 8 m i n ...... 157

2.20 Three-dimensional chromatogram truncated at 300 mAU of pre dose urine treated with AC 2 O-K 2 CO 3 ...... 158

xvii 2.21 Liquid chromatograms of standard curve samples prepared by spiking drug-free urine with the four analytes and assayed according to the der iva t. i za t i on procedure. The concentra tions of the analytes were essentially equivalent in the standards. (A) Analyte and internal standard-free urine; (B) analytes, 0.25 yg/mL; (C) analytes, 1.0 yg/mL. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.8 min; 4'-OAc- MB, 6.5 min; 2-oxo-MB, 8.1 min; MB, 9.6 min; internal standard, 14.6 m i n ...... 159

2.22 Liquid chromatograms of standard curve samples prepared by spiking drug-free urine with the four analytes and assayed according to the derivatization procedure. The concentra tions of the analytes were essentially equivalent in the standards. (A) analytes, 2.0 yg/mL; (B) analytes, 6.0 yg/mL; (C) analytes, 10.0 yg/mL. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.8 min; 4 1 -OAc-MB, 6.5 min; 2-oxo-MB, 8.1 min; MB, 9.6 min; internal standard, 14.6 m i n ...... 160

2.23 Chromatograms of patient urine obtained on day 5 during the 2 h civ of MB, 190 mg/m^/d x 5. (A) Sample chromatographed directly; (B) sample chromatographed following derivatization. Chromatographic peaks: 4 1 -0Ac-2-oxo-MB, 4.9 min; 4 ' -OAc-MB, 6 . 6 min; 2-oxo-MB, 8.4 min; MB, 10.0 min; internal standard (3*-F-MB), 15.2 min ...... 161

2.24 Calibration curve for 4'-0H-2-oxo-MB in human urine: slope, 0.3645; y-intercept, 0.0523; r, 0.9992...... 162

2.25 Calibration curve for 4'-OH-MB in human urine: slope, 0.3198; y-intercept, 0.0335; r, 0.9989...... 163

2.26 Calibration curve for 2-oxo-MB in human urine: slope, 0.3195; y-intercept, 0.0219; r, 0.9993...... 164

2.27 Calibration curve for MB in human urine: slope, 0.3426; y- intercept, -0.0023; r, 0.9998...... 165

3.1 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 192 mg/m /d x 5 to patient J. S. by 24 h civ. Key: (•) 4 ' -0H-2-oxo-MB; (+) 4 ' -OH-MB; (*) 2-oxo-MB; (o) M B ...... 179

3.2 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 317 mg/m^/d x 5 to patient J. P. by 24 h civ. Key: (•) 4'-OH-2-oxo-MB; (+) 4 1 -OH-MB; (*) 2-oxo-MB; (o) MB ...... 180

xviii 3.3 Plot of the dally excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 560 mg/m^/d x 5 to patient C. F.. by 24 h civ. Key: (•) 4 ' -0H-2-oxo-MB; ( + ) 4 ' -OH-MB; (*) 2-oxo-MB; (o) MB ...... 181

3.4 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 1000 mg/m /d x 5 to patient W. K. by 24 h civ. Key: (•) 4'-0H-2-oxo-MB; (+) 4 ' -OH-MB; (*) 2-oxo-MB; (o) M B ...... 182

3.5 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 1250 mg/m /d x 5 to patient J. W. by 24 h civ. Key: (•) 4 ' -0H-2-oxo-MB; (+) 4 ' -OH-MB; (*) 2-oxo-MB; (o) M B ...... 183

3.6 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 150 mg/m /d x 5 to patient V. P. by 2 h civ. Key: (•) 4 1 -0H-2-oxo-MB; (+) 4 1 -OH-MB; (*) 2-oxo-MB; (o) MB ...... 184

3.7 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 253 mg/m /d x 5 to patient E. H. by 2 h civ. Key: (•) 4' -0H-2-oxo-MB; (+) 4 ' -OH-MB; (*) 2-oxo-MB; (o) M B ...... 185

3.8 Plot of the daily excretion rate versus the urinary collec tion interval midpoint for MB and its metabolites during and after the administration of 336 mg/m /d x 5 to patient W. N. by 2 h civ. Key: (•) 4' -0H-2-oxo-MB; (+) 4 ' -OH-MB; (*) 2-oxo-MB; (o) MB ...... 186

3.9 Three-dimensional chromatogram from 0-15 min, truncated at 40 mAU, of patient plasma obtained on day 5 during the 24 h civ infusion of MB, 1000 mg/m^/d x 5, assayed according to the acetylation procedure. Chromatographic peaks: 4 '-OAc- MB, 5.9 min; 2-oxo-MB, 7.5 min;.. MB, 8 . 8 m i n ...... 187

3.10 Plot of the three-dimensional chromatogram from 3-11 min with enhanced resolution of the patient plasma sample obtained on day 5 during the infusion of MB, 1000 mg/m /d x 5, assayed according to the acetylation procedure. Chromatographic peaks: 4 1 -OAc-MB, 5.9 min; 2-oxo-MB, 7.5 min; MB, 8 . 8 m i n ...... 188

xix 3.11 Three-dimensional chromatogram from 0-15 min, truncated at 40 mAU, of pre-dose patient plasma assayed after treatment with AC 2 O-K2 CO 3 showing the absence of constituents which would interfere with analytes eluting from 3-11 min ...... 189

4.1 Plot of absorbance at 295 nm as a function of uric acid concentration under conditions similar to that of the xanthine oxidase inhibition experiments. Solutions were buffered to pH 7.8 with 0.05 M potassium phosphate (y =■ 0.14) containing 0.01 mM EDTA. The best fit line was obtained from linear regression employing a weighting factor of inverse absorbance squared: s’ope, 11,535 1/M; y-intercept, -0.004664; r. 1.0000...... 208

4.2 Plot of initial fracuional inhibition of xanthine oxidase by allopurinol. IiB, 4'-OH-MB, 2-oxo-MB and 4 1 -0H-2-oxo-MB as a function of concentration. The reaction conditions are described in Table 4.1. Key: (x) allopurinol; (o) 4'-OH-2-oxo-MB; ( + ) 4'-OH-MB; (*) 2-oxo-MB; (•) MB ...... 209

4.3 Profiles for the initial fractional inhibition of xanthine oxidase by MB, 4'-OH-MB, 2-oxo-MB and 4'-0H-2-oxo-MB. The reaction conditions are described in Table 4.1. Key: (o) 4'-OH-2-oxo-MB; (*) 2-oxo-MB; ( + ) 4 ’-OH-MB; (•) MB ...... 210

4.4 Time courses obtained over 45 min for the change in absor bance at 295 nm due to the xanthine oxidase catalyzed con version of xanthine to uric acid. Profiles are shown rep resenting reactions run in the absence of inhibitor, and in the presence of 0.01 mM allopurinol, MB, 4'-OH-MB, 2-oxo-MB and 4'-OH-2-oxo-MB. The reaction conditions are described in Table 4 . 1 ...... 211

xx INTRODUCTION1

The general objective of the project was to obtain information

pertaining to the metabolism of merbarone in man. Merbarone is a new

investigational anticancer agent which was recently introduced into phase I clinical trials. This structurally interesting compound belongs

to a class of barbituric acid derivatives that have not been the object of much attention in the past. Aside from information derived in preclinical studies of the drug, the biotransformation of compounds of this type has not been examined.

The dissertation is organized into four chapters. The introduction to each chapter contains a statement regarding the nature of the problem being examined and the goals of the study. References, tables and figures are appended at the end of each chapter. The following summarizes the studies undertaken in this project.

Urinary excretion of the unchanged drug is only a minor route of elimination in cancer patients treated with merbarone. Although plasma levels of metabolites are exceedingly low during and after dosing, the drug is subject to extensive biotransformation, as indicated by the relatively high levels of metabolites excreted in urine. The three major metabolites isolated from the urine of patients receiving merbarone were identified as 4'-hydroxy-2 -oxo-desthiomerbarone,

4'-hydroxymerbarone, and 2-oxo-desthiomerbarone by HPLC, UV 2

spectroscopy, NMR spectroscopy, mass spectrometry and chemical

synthesis.

To determine the significance of metabolite excretion as a route of

elimination, a method was developed to simultaneously assay merbarone

and the three characterized metabolites in urine. The polarity of the

hydroxylated metabolites precluded the use of the existing HPLC method

for the drug. Accordingly, a procedure was developed based on selective

phenolic acetylation in alkaline aqueous solution, resulting in

increased retentions for these compounds upon chromatography.

The assay method was used to quantitate the daily urinary excretion

of merbarone and the three metabolites during the 5 day infusion and 2

days post-infusion for patients in the phase I trials. Typically, 25%

of the dose was accounted for by the excretion of these compounds,

indicating that major routes of elimination remain to be elucidated.

It was observed that hypouricemfa commonly resulted in patients

subsequent to treatment with merbarone. Studies were therefore

undertaken to examine the inhibition of xanthine oxidase by the drug and

its metabolites. The compounds were found to be extremely weak

inhibitors of the enzyme ijn v_it.ro, suggesting that uricosuric affects associated with the drug and/or its metabolites may be operative.

* Some of the abbreviations used are: MB, merbarone; 2-oxo-MB, 2-oxo- desthiomerbarone; 4'-OH-MB, 4'-hydroxymerbarone; 4'-0H-2-oxo-MB, 4'-hydroxy-2 -oxo-desthiomerbarone; acetic anhydride, AC 2 O; civ, continuous intravenous infusion. CHAPTER I

CHARACTERIZATION OF MERBARONE METABOLITES IN NAN

A. Introduction

Merbarone (1,2,3,4-tetrahydro-6 -hydroxy-4-oxo-2-thioxo-phenyl-5-py-

rimidinecarboxamide; NSC 33662R) is one of the few barbituric acid der

ivatives that exhibit antitumor activity. The structure of merbarone is

shown in Fig. 1.1. The compound was synthesized in the research

laboratories of Uniroyal Ltd. (Guelph, Ontario, Canada) and found to

exhibit antitumor activity in a prescreening test conducted by the

National Cancer Institute (NCI) (1). Following a favorable preclinical evaluation of the compounds potential as a chemotherapeutic agent (2 ),

Phase I clinical trials were initiated at The Ohio State University

Cancer Center (OSU) and subsequently at the Memorial Sloan-Kettering

Cancer Center.

The activity of merbarone against several murine tumor models, based on percent increased life span values, has been reported (1). The greatest therapeutic efficacy was observed against i.p. and s.c. implanted L1210 leukemia, with administration of the drug either local or distant to the implantation site. However, i.p. administration was significantly less effective following intracranial implantation. Good activity was also demonstrated against the other i.p. implanted tumors, namely P388 leukemia, B16 melanoma and M5076 sarcoma. A preliminary investigation on the mechanism of merbarone's

antitumor activity provided predominantly negative information (3).

Results indicate that merbarone is unlikely to function as an alkylating

agent, respiratory inhibitor or classical antimetabolite. It does not

appear to bind or intercalate with DNA, and DNA-protein crosslinking was

also not apparent. However, significant numbers of dose related DNA

single strand breaks were observed that are probably not protein

associated. Furthermore, merbarone may be capable of generating oxygen-

centered radicals in the presence of NADPH in microsomal systems (4).

Prior to the screening of merbarone, more than 700 barbiturate

derivatives were tested for antitumor activity in the NCI screening

program (5). None of the compounds were found active against the L1210

leukemia or B16 melanoma systems. The earliest apparent effort to

systematically screen barbiturates for antitumor activity was reported

in 1960 by the group of T. Ukita (6 ). Among the compounds tested, the

most promising potential chemotherapeutic was thought to be 5-phenyl-

carbamoylbarbituric acid, the 2-oxo analog of merbarone. Excellent in

vivo activity against implanted Ehlich ascites carcinoma in mice was demonstrated when the compound was administered i.p.

Interest in the antitumor properties of barbiturate derivatives by this group evolved from structure activity studies of some antibiotics

(7,8). It was demonstrated that the so-called tricarbonylmethane group, or 2-carbonyl derivative of a 1,3-dione, was the essential structural feature for the antibacterial effects of a number of compounds, including usnic acid, citrinin, and the tetracyclines. Compounds with an exocyclic carbamoyl group appeared to be more effective than other carbonyl der ivnt 1 v p r . In particular, plio.ny 1 carbamoyl derivatives were

observed to be exceptionally potent in many instances. Antitumor

studies by this group were begun by examining a series of acyclic

tricarbonylmethane compounds (9). The results of in vitro screening

were promising, prompting an investigation of similarly substituted

barbiturates. The barbiturate ring was selected as the parent molecule

as it was postulated that the cyclic ureido structure would increase

affinity for tumor cell components, resulting in greater activity.

Although it appears that 5-carbamoylbarbiturates were shown to be

antitumor leads of good potential, reports on further investigations of

these compounds are notably absent in the literature, until their

rediscovery with merbarone. The only related study was an NCI funded

examination of the antitumor activity of 5-carboxaldehyde derivatives of

2 -thiobarbituric acid, for which screening tests were negative in all

cases ( 1 0 ).

Limited metabolism studies were conducted in our laboratories in

association with the preclinjcal pharmacokinetics (2). A glucuronide conjugate of merbarone was detected in dog plasma and a polar metabolite, tentatively identified as 4 1 -hydroxymerbarone, was excreted

in the urine together with the parent drug. Metabolites were not detectable in mouse plasma, although the 41-hydroxy metabolite was found in urine (1 1 ).

During the course of the phase I trials conducted at OSU, several significant peaks were observed in the liquid chromatograms of patient urine obtained during and after the administration of merbarone which were thought to be metabolites of the drug. The relatively low concentration of merbarone in patient urine indicated that urinary

excretion of the unchanged drug is a minor route of elimination.

However, preliminary evidence was obtained suggesting that the

concentration of urinary metabolites greatly exceeded that of the drug,

although significant levels of metabolites were not observed in plasma

samples. Therefore, an effort was made to isolate and identify the

urinary metabolites of merbarone in man.

B. Materials and Methods

1. Patient Urine and Plasma Samples

Urine and plasma samples were obtained from patients

participating in the phase I program at OSU who were treated with

merbarone according to the 120 h continuous infusion schedule. Pooled

24 h urines and plasma samples were acquired on day 5 of the infusion

from patients receiving doses of 192 and 1000 mg/m /d. The urine was

stored at -20°C as collected. Plasma, separated from blood by

centrifugation for 10 min at 1800 x g, was immediately frozen and stored

at -20°C.

2. Chemicals and Materials

A reference sample of merbarone (NSC 336628, Lot// PC0618) was

obtained from Dr. J. A. R. Mead, Developmental Therapy Program, Division

of Cancer Treatment, NCI. 4-Acetoxybenzoic acid was purchased from

Lancaster Synthesis Ltd. (Windham, NH). Acetic acid and ammonium

acetate were analyzed reagents, and the triethylamine was Baker grade

(J. T. Baker Chemical Co., Phillipsburg, NJ). Barbituric acid,

4,6-dihydroxy-2-mercaptopyrimidine, 4 1-methoxyphenyl isocyanate, phenyl 7

isocyanate and calcium hydride were obtained from the Aldrich Chemical

Co. (Milwaukee, WI). Dodecyl sodium sulfate was purchased from Eastman

Kodak Co. (Rochester, NY). Acetic anhydride, hydrochloric acid and

phosphorous pentoxide were certified A. C. S. reagents; reagent grade

thionyl chloride and purified sodium azide were used (Fisher Scientific

Co., Fairlawn, NJ). The carbon tetrachloride, chloroform, n-hexane,

magnesium sulfate heptahydrate, anhydrous potassium carbonate and

potassium hydroxide pellets were analytical reagents (Mallinckrodt,

Paris, KY). Ominisolv grade acetone, dioxane and methanol were obtained

from EM Science (Cherry Hill, NJ). The dimethylsulfoxide was reagent

grade (MCB, East Rutherford, NJ). Deionized double distilled water was passed through a 0.2 ym nylon - 6 6 filter (Rainin Instrument Co., Woburn,

MA) before use. High purity dry grade nitrogen (Union Carbide, Danbury,

CT) and prepurified argon (AGA Gas, Cleveland, OH) were used.

Dimethylsulfoxide-dg (99.5% D min.) (Wilmad Glass Co., Buena, NJ), gold

label chloroform-d (99.8 atom% D) and NMR grade tetramethylsilane

(Aldrich) were used. All chemicals were used without further purification except as noted.

Purified amberlite XAD-2 polystyrene resin (20/60 mesh size) was purchased from Alltech Associates (Deerfield, IL). Whatman qualitative

#1 and hardened if50 and if54 filter paper were obtained through Thomas

Scientific (Philadelphia, PA). Davison Chemical molecular sieves were used (Baltimore, MD). 8

3. Reagent and Solvent Purification (12,13)

Acetone. Potassium permanganate was added to refluxing acetone until a violet color persisted. The solution was cooled, dried over anhydrous calcium sulfate, filtered and fractionated in an atmosphere of nitrogen. The acetone (bp 56-57°C) was stored over type 4A molecular sieves under argon.

Carbon tetrachloride. Dry solvent was prepared by fractional distillation in an atmosphere of nitrogen. The carbon tetrachloride (bp

76-77°C) was stored over type 5A molecular sieves under argon.

Dioxane. One liter of dioxane, 100 mL of water and concentrated

HC1 (14 mL) were refluxed for 12 h in a one neck round-bottomed flask

(2000 mL) while a slow stream of nitrogen was bubbled through the solution. Potassium hydroxide pellets were added to the solution, after cooling, with shaking until they no longer dissolved. The upper dioxane layer was decanted and allowed to stand over potassium hydroxide ( 1 0 g) for 12 h. After decanting again, the dioxane was refluxed for 12 h over calcium hydride (10 g) and distilled. Excess sodium metal was added to the solvent in the receiving flask, which was fitted with a distilling head for maintenance of an inert atmosphere. The anhydrous, peroxide- free dioxane was kept under reflux in the presence of sodium under nitrogen.

n-Hexane. The solvent (200 mL) was dried by refluxing over calcium hydride (2 g) with stirring for 12 h under an atmosphere of nitrogen. The fraction boiling at 69°C was collected by distillation from the calcium hydride and stored over sodium metal under argon. 9

Thlonyl chloride. Triphenyl phosphite (16 mL) was added

dropwlse over 30 min to thlonyl chloride (100 mL) while stirring

magnetically. The mixture was fractionally distilled through a Vigreaux

column connected to a reflux distilling head equipped with a drierite

drying tube. Colorless thionyl chloride was collected over the boiling

range 75-76°C and stored under argon.

Triethylamine. Two hundred milliliters of triethylamine was

allowed to stand over KOH pellets (10 g) for approximately 12 h,

decanted, and distilled in the presence of phenyl isocyanate (5 mL).

The reagent was stored over type 3A molecular sieves under argon.

4. Instrumentation

Fourier transform NMR spectra were recorded at 250 MHz on an

IBM NR-250 FT-NMR. An IBM AF-270 spectrometer was used to acquire

NMR spectra at 6 8 MHz. This instrument was also used for the

acquisition of 270 MHz *H NMR spectra as a function of temperature. The multinuclear Bruker AM-500 spectrometer at The Ohio State University

Chemical Instrument Center was employed to obtain *H NMR spectra at 500

MHz.

Electron impact (70 eV) and chemical ionization (CH^) mass spectra were acquired with a Finnigan 4021 GC/MS instrument by direct probe sample introduction at the Biomedical Branch of the Chemical Instrument

Center. Infrared spectra were obtained with a Beckman IR 4230 and a

Beckman model 3600 spectrophotometer was used to record UV absorption spectra of solutions in matched 1 . 0 cm cuvettes. Melting points were determined with a Thomas Hoover Capillary Melting Point Apparatus

(Arthur H. Thomas Co., Philadelphia, PA). Elemental analyses were 10

determined by Galbraith Laboratories (Knoxville, TN) .

2 5. High Performance Liquid Chromatography

a. Apparatus

Qualitative HPLC was performed using a 421A system

controller interfaced to 114M solvent delivery modules (Beckman

Instruments, Berkeley, CA) and a model 7125 syringe loading sample

injector fitted with a 500 yL sample loop (Rheodyne, Cotati, CA). UV

spectra during chromatography were acquired every 1600 ms with a

Hewlett-Packard (Palo Alto, CA) model 1040M diode-array detector and

processed with HP 79995A operating software and HP 79997A color-vieto

software on an HP 79994A analytical workstation. The system was

equipped with a 3.9 mm x 15 cm stainless steel column packed with 4 ym

Nova-Pak C^g (Waters Associates, Milford, MA), with a 0.5 ym post

injector filter (Rheodyne).

b. Sample Preparation

Plasma samples were assayed with prior precipitation of

protein by adding 250 yL of methanol-DMSO (85:15, v/v) to 50 yL of

plasma in a 1.5 mL polypropylene microcentrifuge tube. The tube was

mixed by vortexing for 0.5 min then centrifuged at 12,000 x g for 10

min. After diluting 2 0 0 yL of the supernatant with mobile phase less

methanol, 100 yL was injected onto the column.

o The HPLC method for the detection of merbarone, including the preparation of plasma samples by protein precipitation, was developed by Dr. L. Malspeis (14) and used in these studies with minor modifications. 11

Urine samples were prepared by adding DMSO (25 pL) to 100 pL of

urine in a microcentrifuge tube and diluting with mobile phase (435 pL).

The tube was mixed by vortexing and centrifuged at 12,000 x g for 5 min.

The injection volume was 200 pL.

Reactions involving merbarone and related compounds were monitored

by HPLC as follows. An aliquot (25 pL) of the reaction mixture was

added to DMS0 (1.0 mL). After further dilution with mobile phase to

achieve an absorbance of approximately 0.5 at 293 nm, 100 pL was loaded onto the column.

Synthetically prepared samples of 5-carboxani1ide barbiturates were chromatographed upon diluting 5 pL of a DMSO stock solution (1.0 mg/mL) with mobile phase (500 yL) and injecting 200 pL onto the column.

Solutions of the purified urinary metabolite extract were prepared by adding 25 pL of a 1.0 mg/mL solution in DMSO to mobile phase (500 pL).

The injection volume was 250 pL.

c. Chromatographic Conditions

The mobile phase consisted of methanol (25%, v/v), 67 mM ammonium acetate, 33 mM acetic acid, 40 mM magnesium sulfate and 1 mM dodecyl sodium sulfate. Degasification was effected by placing the solution in an ultrasonic bath for 15 min. The flow rate was 1.0 mL/min and the temperature ambient.

6 . Chemical Syntheses

4-Acetoxybenzoyl chloride (4). A one-neck round bottom flask

(100 mL) was fitted with a Graham condenser and drierite drying tube without the use of silicone grease. The outlet from the drying tube was directed to a gas trap (12). Adsorbed moisture was removed by flaming 12

the apparatus.

4-Acetoxybenzoic acid (18.01 g, 0.1 mol) and purified thionyl

chloride (14.6 mL, 0.2 mol) were introduced into the reaction vessel.

The mixture was heated on a steam bath, gently at first, for 4 h. The

reaction mixture was cooled to ambient temperature and transferred to a

50 ml. flask containing a magnetic stir bar for vacuum distillation.

After removal of the excess thionyl chloride, further distillation

afforded 17.5 g (8 8 %) of a colorless solid: bp 134-136°C (5 mm Hg), mp

29-30°C (lit. (15), bp 145-146°C (12 mm H g ) , mp 29°C); UV (n-hexane)

Xmax 2 5 3 nm (E = 1 7 > 100 ± 1 2 0 M _ 1cm-1); IR (CC14 ) 1172 (C=0, ester),

1745 (C=0, acid chloride), 1159,1190 (C-0) cm"1; NMR (CDCI3 ) 6 2.34

(s, 3 H ) , 6 7.22 (d, J (c *-er peak spacing) = 8 . 8 6 Hz, 2 H), 6 8.16 (d,

J (outer peak spacing) = 8 . 8 6 Hz, 2 H ) ; MS (El), m/e (relative

intensity), 200 (M+2, 1.9), 198 (M, 6 .6 ), 163 (24.3), 156 (13.7), 121

(100); MS (CI-CH4), m/e (relative intensity), 201 (MH+2, 28.2), 199 (MH,

89.6), 163 (M-35, 100); Anal. Calcd for CgHyClC^: C, 54.53; H, 3.55; Cl,

17.35; 0, 24.17. Found: C, 54.25; H, 3.35; Cl, 18.08; 0, 24.94.

4-Acetoxybenzazide (5). Sodium azide (6.50 g, 0.10 mol) was dissolved in water (50 mL) by magnetic stirring in a round-bottomed flask (250 mL) placed in a circulating water bath maintained at 20-25°C.

Over a period of 1 h, 4-acetoxybenzoyl chloride (4, 16.0 g, 81 mmol) in dry acetone (25 mL) was added dropwise with stirring to the solution, resulting in immediate precipitation of the product. Thirty minutes after addition was complete, water (50 mL) was added and stirring continued for 30 min. The precipitate was collected by vacuum filtration, washed with water and dried overnight in an evacuated 13 desiccator charged with drier! to.. The pure white solid was used without

further purification: yield 15.7 g (95%); mp 73-74°C (dec. with evolution of gas); UV (n-hexane) ^max nm = 22,800 ± 200

M _1cm_1); IR (CC14 ) 2136 (N=N=N, sym), 1770 (C=0, ester), 1694 (C=0, azlde), 1246 (N=N=N, asym), 1195,1158 (C-0) cm-1; !H NMR (CDC13) 6 2.33

(s, 3 H), 6 7.20 (d, J (outer peak spacing) = 8.84 Hz, 2 H),6 8.06 (d,

J (outer peak spacing) = 8.84 Hz, 2 H); MS (El), m/e (relative intensity), 205 (M, 30.6), 163 (85.0), 135 (59.7), 121 (100), 107

(59.0), 93 (19.5), 79 (37.8); MS (CI-CH4), m/e (relative Intensity), 206

(MH, 62.6), 178 (MH-28, 100), 163 (84.6); Anal. Calcd for CgH 7N 30 3 : C,

52.69; H, 3.44; N, 20.48; 0, 23.39. Found: C, 54.64; H, 3.41; N, 19.34;

0, 23.78.

4-Acetoxyphenyl Isocyanate (6 ). A round-bottomed flask (100 mL) fitted with a Graham condenser was flamed to remove adsorbed moisture while purging with argon. The argon atmosphere was maintained in the vessel throughout the course of the reaction. 4-Acetoxybenzazide (5,

15.0 g, 73 mmol) and anhydrous, peroxide-free dioxane (25 mL) were introduced into the flask, which was placed in an oil bath at 110°C and magnetically stirred. Dissolution of the azide occurred readily, with evolution of gas, which continued for approximately 1 h. The reaction mixture was allowed to cool after heating for 2 h. The flask was fitted for vacuum distillation, and after removal of the solvent, the isocyanate 6 was afforded as a colorless liquid: yield 12.0 g (93%); bp

118-119°C (5 mm Hg); UV (n-hexane) Xmflx 231 nm (e = 19,600 ± 120

M-1cm-1) ; IR (CC14 ) 2260(w), 2279(s) (N=C=0), 1767 (C=0, ester),

1194,1159 (C-0) cm'1; XH NMR (CDClj) 6 2.29 (s, 3 H), 5 7.07 (tn, Ph, 4 14

H); MS (El), m/e (relative intensity), 177 (M, 10.9), 135 (100), 107

(20.5), 79 (6.2), 52 (12.0); MS (CI-CH4), m/e (relative intensity), 178

(MH, 100); Anal. Calcd for C 9 H?N03: C, 61.02; H, 3.98; N, 7.91; 0,

27.09. Found: C, 60.51; H, 3.87; N, 7.57; 0, 27.02.

Synthesis of 2 -(N-Phenylcarboxamido)-1 .3 -dlone3 1. General Pro

cedure. A three neck round-bottomed flask (50 mL) was fitted with an

inlet for dry argon, a Graham condenser, and a pressure equalized drop

funnel. The glassware was thoroughly dried arid an argon atmosphere

provided during the reaction. The 3-diketone 2 (7.0 mmol) was

introduced into the flask as a solid and dissolved in anhydrous

peroxide-free dioxane (25 mL) with magnetic stirring at 50-55°C. To the

rapidly stirred solution, triethylamine (1.0 mL, 6.9 mmol) in dioxane

(2.5 mL) was added dropwise over 15 min, forming a suspension of

the 3“diketone-NEt3 salt. A solution of the isocyanate (7.05 mmol) in dioxane (5.0 mL) was added dropwise over 15 min to the stirred suspension. The bath temperature was increased to 80-85°C and stirring continued for 3 h.

Decomposition of unreacted isocyanate was effected by pouring the hot reaction mixture into HC1 (150 mL, 0.1 M) with vigorous stirring.

After 30 min, the precipitated product was recovered by vacuum filtration through Whatman #54 hardened paper, washed liberally with water and dried dried overnight in a vacuum desiccator charged with drierite. The product was recrystallized from dioxane-methanol (two crops) and dried at 101°C in vacuo (5-10 mm Hg) over ?2 ® 5 • 15

Hydrolysis of 2-N-(4'-Acetoxyphenylcarboxamldo)-1.3-dlones 3.

General Procedure. Aqueous KOH (20 mL, 2 N) was added to the phenyl

acetate le-f (3.2 mmol) and stirred at ambient temperature for 45 min,

giving rise to an opaque, purple colored solution. Slow addition of the

mixture to ice-cold HC1 (80 mL, 1 M) with vigorous stirring resulting in

precipitation of the product, which was recovered by vacuum filtration

through Whatman #50 paper. Initial drying overnight in a vacuum

desiccator charged with drierite was followed by drying in vacuo (5-10 mm Hg) at 101°C over ^2^5'

7. Extraction of Metabolites from Patient Urine

The metabolites were isolated from urine by the solid phase extraction technique employing an amberlite XAD-2 column. The column was prepared by washing the resin (75 g, 200 mL) with three volumes of methanol. After degassing by sonnication in the presence of methanol, the resin was introduced into a glass chromatographic column (3.25 cm

i.d. x 40 cm) as a slurry and washed with three bed volumes of water.

Urine was thawed, brought to ambient temperature-, filtered through

Whatman qualitative paper, and gradually transferred onto the column.

The flow rate was adjusted to 2-3 mL/min and eluant was collected in 100 mL fractions. After 1000 mL of urine had passed through, the column was washed with water (500 mL) and eluted with 500 mL of dioxane-methanol

(1:1, v/v). All eluant fractions were qualitatively assayed by HPLC upon dilution of an aliquot (100 pL) with mobile phase (400 pL) and directly injecting 200 pL onto the analytical column. The column was regenerated by successive 500 mL washes with methanol and water. 16

Removal of the solvent mixture from the pooled dloxane-methanol

fractions under vacuum at 40°C on a rotary evaporator afforded a viscous

brown residue, which was transferred to a test tube as a slurry.

Removal of the remaining solvent by evaporation under a stream of

nitrogen at 40°C and drying over P 2 O 5 vacuo afforded 60 mg of a brown

solid.

Purification was achieved by dissolving the dry residue in aqueous

KOH (25 mL, 0.1 M), filtering the solution through Whatman #54 hardened

paper, and precipitating the merbarone metabolites by acidification with

6 M HC1 to approximately pH 2. After pelleting by centrifugation and

removal of the supernatant, the precipitate was washed with water until the washings were neutral and devoid of chloride ions, as indicated by ethanolic silver nitrate. The metabolite mixture was dried under a stream of nitrogen at 40°C prior to drying in vacuo (5-10 mm Hg) at

101°C over P 2 O 5 • The light brown solid was washed with chloroform until the solvent remained colorless.

8 . Acetylatlon of Phenolic Metabolites Isolated from Urine

In a siliconized screw top test tube, the purified metabolite mixture (15 mg) was dissolved in DMSO (1.5 mL) and alkalinized with aqueous potassium carbonate (2.0 mL, 0.5 M). Immediately after adding acetic anhydride (100 yL, 1.1 mmol), the tube was capped and rapidly mixed for 60 s on a vortex mixer. The reaction mixture was then added to HC1 (2 mL, 1 M) dropwise with stirring. The precipitate was recovered by vacuum filtration through Whatman #54 paper, washed with water, and dried in vacuo (5-10 mm Hg) at 101°C over P 2 ^ 5 - 17

C . Results and Discussion

1. The Metabolism of Functionally Related Xenobiotlcs

A search of the literature has shown that, aside from studies

conducted in our laboratories, an investigation of the biotransformation

of 5-carboxanilide barbiturates has not been previously undertaken.

However, consideration of the primary metabolic routes of functionally

related xenobiotics, such as the 5,5-disubstituted thiobarbiturates,

phenobarbital and acetanilide (Fig. 1.2), enables potential metabolites

of merbarone to be predicted.

The conversion of thiobarbiturates to the corresponding oxybarbiturates by oxidative desulfurization is well documented, and much of the earlier work has been reviewed (16). Non-quantitative studies have shown that desulfurated metabolites were detected in the urine of patients treated with thiopental, thiobutabarbital and methitural. Several quantitative investigations have also been undertaken. In a particularly well conducted study by Bush, et al.

(17), barbital was assayed in the urine of patients who were administered i.v. thiobarbitaI. Their results indicate that 5-7% of the dose was excreted in the urine as barbital. Since barbital is quantitatively eliminated by urinary excretion of the unchanged drug, this is indicative of the extent of desulfurization.

Aromatic ring hydroxylation is an important metabolic pathway for phenobarbital and related compounds (18-23) and acetanilide (24). The major metabolite of phenobarital arising from this route is the p- hydroxy derivative, accounting for 15-35% of the dose excreted in the urine. Metabolic p-hydroxylation of acetanilide gives rise to 18 acetaminophen, which is further metabolized.

A minor urinary metabolite of phenobaribital in man was found to be the 5-(3,4-dihydroxy-1,5-cyclohexadien-1-yl) derivative (19). Since dihydrodiol metabolites have been characterized for a number of xenobiotics with a monosubstituted benzene ring, this is thought to be a general route of drug metabolism (19,23).

The major urinary metabolite of phenobarbital has been characterized as l-(fl-D-glucopyranosyl)-phenobarbital, accounting for 30% of the total dose which was administered (22). An N-glucoside conjugate of amobarbital has also been identified (25) which was incorrectly characterized as an N-hydroxy derivative in a previous report (26).

Therefore, although N-glucoside formation is extremely uncommon, it has been suggested that N-glucosidation may be an important pathway of barbiturate metabolism in man. These conjugates are apparently not cleaved by B-glucuronidase (2 1 ,2 2 ) or acid hydrolysis (2 0 ) under the usual conditions.

Approximately one-half of the p-hydroxyphenobarbital which is excreted in the urine of man is conjugated to glucuronic acid

(18,20,21,23), while in the dog it is completely conjugated (19).

Although the conjugate in human urine was initially thought to be the sulfate ester (18), subsequent studies have conclusively demonstrated that it is a glucuronide (20,21). Similarly, acetaminophen is largely excreted in the urine as glucuronide and sulfate conjugates (27).

It is also possible that thiobarbiturates may be subject to conjugation through the sulfur atom with glucuronic acid. Although an example of this type of metabolite for a thiobarbiturate has not been 19

reported, the S-glucuronidation of a thiolar.tam has been demonstrated

(28).

2. Synthesis of Potential Oxidative Metabolites

Merbarone (la) may be viewed as an N-substituted carboxamido

derivative of the 3-diketone 2-thiobarbituric acid (2a). Compounds of

this type are readily prepared from the reaction of an isocyanate with

an acyclic or cyclic 1,3-dione, yielding the C-substituted addition

product (6,29-32) (eq 1.1). The reaction is typically performed in dry,

aprotic solvents in the presence of a basic catalyst such as pyridine or

triethylamine, however, the alkali metal salt of the 0 -diketone may be

used directly.

(eq 1 .1 )

Although the synthesis of la has not been published, 5-(N- phenylcarboxamido)-barbituric acid (lb), the 2 -oxo-desthio analog of la. was prepared by refluxing equimolar quantities of barbituric acid (2 b)

and phenyl isocyanate in dry pyridine (6 ). Since both la and 2a were

found to be insoluble in pyridine, this particular method was not

suitable for the synthesis of the thiobarbiturate derivatives. However, anhydrous, peroxide-free dioxane was found to be a satisfactory solvent

for the reaction when triethylamine was employed as the catalyst. 20

Accordingly, a general procedure Cor the synthesis of merbarone and

related compounds ^1 in high purity and excellent yield has been

developed (eq 1.2).

NEt3, Dloxana H N ^ (eq 1 .2 )

H R 2 1

As a consequence of the reactivity of isocyanates (33), the direct

preparation of phenolic derivatives by this method is not possible. p-

Hydroxyphenobarbital has been synthesized by the classical route of nitration, reduction and diazotization (18). Nitration of phenobarbital produces a mixture of the isomeric para and m-nitro derivatives, which must be separated prior to subsequent reactions. Due to the physical properties of La, a satisfactory separation of the corresponding derivatives would probably not be easily achieved. However, since

4-nitrophenyl isocyanate is commercially available, the pure p-nitro derivative could be readily prepared. Nevertheless, the reactivity of

2 -thiobarbiturates precludes the use of this synthetic approach without protection of the thiocarbonyl group.

Alternately, an isocyanate with a functionally protected phenolic oxygen could be utilized to prepare 5-carboxanilide barbiturates 1^ from which the p-hydroxy derivatives 3 would be afforded upon cleavage of the protecting group. The only suitable commercially available reagent was found to be 4-methoxyphenyl isocyanate, with which the 4'-methoxy 21

analogs Lc and Id worn prepared. Howevor, cleavage of the methyl group

by the usual methods (34) could not be effected without at least partial

degradation of the thiobarbiturate derivative.

Since merbarone was found to be stable under moderate alkaline

conditions, a phenolic protecting group readily cleaved by mild

saponification, such as an acetate, was desirable. However, as

mentioned previously, isocyanates of this type were not commercially

available. Therefore, the synthesis of 4-acetoxyphenyl isocyanate (6 )

was undertaken.

The preparation of 6 by the phosgenation of methyl 4-aminobenzoate

has been described in the patent literature (32). However, due to the

hazards associated with handling phosgene, 6 was prepared by an approach

involving acyl azide pyrolysis, as shown in Fig. 1.3, according to

standard procedures (33). 4-Acetoxybenzoyl chloride (4) was initially

prepared as a result of the reaction between the corresponding benzoic

acid and thionyl chloride (15). Treatment of 4 with sodium azide

afforded 4-acetoxybenzazide (5) (35), which was thermally decomposed in

anhydrous dioxane to give 6 (36,37).

The 4'-hydroxy-5-carboxanilide barbiturates 3a and 3b were

successfully prepared in high purity upon hydrolysis of the

corresponding 4'-acetoxyphenyl derivatives le and 1J (eq 1.3), which were synthesized by the general method previously described employing

the isocyanate 6 . As judged by HPLC, the hydrolysis of the acetyl protecting group in 2 N KOH at ambient temperature was complete within

45 min. The purity of the phenolic derivatives, which were isolated by precipitation upon acidifying the reaction mixture, was such that 22

further purification was not necessary. The physical properties and

analytical data for the synthesized 5-carboxani1ide barbiturates 1 and 3

are summarized in Tables 1.1-1.7.

OAc OH OH OH

HN HN 2 N KOH , , u , 1 _ . H (e

H H le-f

3 3. Metabolite Characterization

The metabolites of MB were characterized upon isolation from urine by several complimentary techniques, including HPLC, UV spectroscopy, NMR spectroscopy, mass spectrometry, derivatization and chemical synthesis. The HPLC method developed for assaying MB in plasma

(14) was used qualitatively in these studies with minor modifications.

The sample preparation procedures and chromatographic conditions are summarized in Tables 1.8 and 1.9. UV detection at 306 nm was employed in the analytical method since the drug possesses a reasonably strong chromophore, exhibiting an absorption maximum (^max) at that wavelength in the mobile phase.

Further advantage was taken of this property, which is characteristic of barbiturates in general, in the utilization of diode- array spectrophotometric LC detection in these studies. This powerful

3 For a listing of the 5-carboxanilide barbiturate abbreviations used in this section, see page 2 or Table 1.1. 23

technique enables the rapid acquisition of UV/VTS absorption spectra

during chromatography, from which three-dimensional chromatograms (i.e.,

absorbance-time-wavelength) may be constructed. From these

representations, detailed information pertaining to sample composition,

spectral characteristics of eluting components, and peak homogeneity may

be readily ascertained. The effort required to derive similar

information employing traditional chromatographic techniques would be

substantial.

A typical chromatogram determined at 306 nm of patient plasma, o obtained on day 5 during the infusion of MB at the 1000 mg/m /d dose

level is shown in Fig. 1.4. The drug elutes at 8 . 8 min and is preceded

by a minor peak eluting near 7.4 min and plasma constituents eluting

shortly after the solvent front. The peak due to the compound eluting

at 7.4 min appears comparatively larger in the three-dimensional

chromatogram of this sample (Fig. 1.5.), indicating that the

chromatogram determined at 306 nm does not adequately convey the

relative concentration of this component. As shown in a plot of the

region from 6-11 min at greater sensitivity (Fig. 1.6), this is a consequence of significant differences in the spectral characteristics of these compounds. The component at 7.4 min exhibits a Xmax at 278 nm with a shoulder near 250 nm, while MB has maxima at 306 and 235 nm.

Since peaks are not present in this region of the three-dimensional chromatogram of plasma obtained before dosing (Fig. 1.7), the component eluting at 7.4 min appears to be a metabolite.

The concentration of this metabolite relative to the drug is significantly greater in urine than plasma, as demonstrated in the 1 0 0 24 mAUFS chromatogram at 306 nm of the day 5 pooled urines obtained from the same patient (Fig. 1.8). The three-dimensional chromatogram (Fig.

1.9), which has also been truncated at 100 mAU, indicates that the concentration of MB is actually much less than the metabolite at 7.5 min. These peaks are shown in greater detail in the plot of the 6-11 min region of the chromatogram with enhanced sensitivity (Fig. 1.10).

Several other peaks due to components eluting in the presence of urinary constituents near the solvent front appear at approximately 1.3,

1.7 and 3.6 min in Fig. 1.8. The characteristics of these peaks are difficult to discern in the three-dimensional chromatogram (Fig. 1.9) due to their relative concentrations and proximity to urinary constituents. However, these components may also be metabolites, since the peaks are not present in the chromatogram of pre-dose urine (Fig.

1.11). This was further demonstrated by plotting the region from 0-2.5 min in the chromatograms of urine obtained during treatment and pre-dose urine at 1000 mAUFS, as shown In Figs. 1.12 and 1.13, respectively. Fig

1.12 also shows that the metabolite at 1.7 min has a Xmflx near 306 nm, while that of the compound eluting at 1.3 min is approximately 280 nm.

Spectral details of these peaks at wavelengths below 250 nm were obscured by endogenous urinary constituents. The concentration of the metabolite at 3.6 min was insufficient to afford an absorption spectrum without significant distortion.

It is therefore suggested that at least four components attributable to metabolites are present in the urine of patients treated with merbarone. The metabolites are presumably more polar than the parent drug since they elute prior to it upon reverse-phase chromatography. 25

Efforts wore directed toward isolating the metabolites from urine to

provide sufficient quantities of the purified components for definitive

characterization. The solid phase extraction technique used for the

isolation of the drug and metabolites is illustrated in Fig 1.14.

Analysis of the effluent from the extraction column by HPLC indicated

that the metabolites were efficiently extracted from urine and remained

adsorbed to the resin during the aqueous wash to remove polar urinary

constituents. Desorption was effected with a methanol-dioxane (1:1,

v/v) solvent system, which afforded a crude mixture of the drug, metabolites and retained urinary constituents.

The separation and purification of metabolite mixtures obtained in this manner is typically achieved by additional chromatographic techniques, such as TLC, column chromatography or preparative HPLC.

However, exhaustive attempts to separate and isolate the MB metabolites from the crude mixture were met without success. The methods which were found to be unsatisfactory include TLC, normal and reverse-phase HPLC, normal phase and anion exchange column chromatography, and various techniques following derivatizations. Therefore, the metabolites were characterized in the mixture after removal of the residual urinary constituents. This was largely achieved upon precipitation of the alkali soluble material by acidification and washing with chloroform.

The chromatogram determined at 293 nm of the purified metabolite mixture (Fig. 1.15), otherwise referred to as the urine isolate, shows peaks due to MB (9.2 min) and the four metabolites. The metabolite peak at 1.3 min, which was almost completely obscured by urinary components

(Fig. 1.8), is reasonably well resolved, indicative of the extent to 26

which the sample had been purified. Since the absorbances of these

compounds are similar at 293 nm, the peak areas are representative of

their relative concentrations, as substantiated by the three-dimensional

chromatogram (Fig. 1.16). The sample is primarily composed of the

metabolites eluting at 7.8, 1.7 and 1.3 min, listed according to

decreasing concentration. Merbarone and the metabolite eluting at 4.0

min, which is poorly visualized at this sensitivity, are minor

components. The low wavelength absorbance at the initial region of the

chromatogram is largely attributable to the solvent front.

The characteristics of the peaks due to the most polar metabolites

are more evident in an expansion of the region of the chromatogram from

1.0-2.25 min (Fig. 1.17). The absorption spectrum of the metabolite at

1.7 min is similar to MB, while the spectra of the metabolites eluting

at 1.3 and 7.8 min are similar, eluding to structural similarities in

the chromophores of these respective compounds.

Several of the synthesized analogs of MB were found to have

retention times and absorbance spectra that were extremely similar to

the three major metabolites. The retention of 4'-0H-2-oxo-MB was

similar to the metabolite eluting at 1.3 min (Fig. 1.18). Furthermore,

an overlay of the UV spectra obtained at the apexes of the chromatographic peaks were found to be nearly identical (Fig. 1.19).

The greater absorbance at lower wavelengths in the spectrum of the metabolite relative to the synthetic material is due to the elution of residual impurities in the urine isolate. The metabolites eluting at

1.7 and 7.8 min were tentatively identified as 4'-0H-MB and 2-oxo-MB, respectively, by analogous comparisons (Figs. 1.20-1.23). A retention 27

corresponding to the minor metabolite (4.0 min) was not exhibited by any

of the synthesized 5-oarboxaniUde barbiturates.

An overlay of the spectra determined at the apex of the peak in the

chromatogram of synthetic MB (Fig. 1.24) and that for the 9.2 min peak

of the urine isolate is shown in Fig. 1.25. The apparent poor

correlation in the region of the spectra below the Xmax is due to the

low concentration of this component in the urine isolate.

During the course of these experiments, it was observed that the

4'-methoxy derivatives of MB and 2-oxo-MB were chromatographically

indistinguishable from the parent compounds. Although catechols are

known to undergo methylation (38), this is not a likely metabolic route

for MB. As shown in overlays of spectra obtained from chromatographic peaks for synthetic samples (Figs. 1.26 and 1.27), the 4'-methoxy derivatives can be differentiated from the parent compounds since the absorption band of the 4 1-methoxy derivative is significantly broader, although the maxima are similar. Therefore, the properties of components eluting at approximately 7.8 and 9.2 min in the chromatograms of samples obtained from patients treated with MB are not consistent with 4 '-methoxy derivatives.

If the metabolites eluting at 1.3 and 1.7 min in the urine isolate are in fact p-hydroxylated compounds, acylation of the phenol group should increase their retentions. Phenolic acetylation in aqueous solution can be rapidly and quantitatively effected with acetic anhydride upon alkalinization (39, also see Chapter II of this report).

As anticipated, chromatography of the urine isolate following treatment with potassium carbonate and acetic anhydride revealed the presence of 28

two new components eluting at 4.5 end 6.2 min, while the peaks at 1.3

and 1.7 min were absent (Fig. 1.2R and 1.29). Viewing the 3.5-11.0 min

region of the three-dimensional chromatogram at increased sensitivity

shows that the spectral characteristics of the derivatives are similar

to their precursors (Fig. 1.30). Homogeneity of the derivative peaks,

and the peak at 7.8 min, was demonstrated by overlaying the UV spectra

determined on the ups lope, apex and downs lope of the chromatographic

peak (Figs. 1.31-1.33). Furthermore, synthetically prepared 4'-0Ac-2-

oxo-MB and 4 'tOAc -MB have retentions and UV spectra that are essentially

identical to the respective metabolite derivatives eluting at 4.5 and

6.2 min (Figs. 1.34-1.37).

Mass spectra and ^11 NMR spectra at 500 MHz of the urine isolate and

synthetic samples have been acquired to further substantiate the

tentative metabolite characterizations. The El mass spectra of

authentic samples of MB and the three proposed metabolites (Figs.

1.38-1.41, Table 1.5) are relatively uncomplicated, showing rather

intense molecular ion peaks and a characteristic fragmentation pattern.

The most important fragmentation reaction involves cleavage of the

exocyclic C-N bond with rearrangement, and elimination of the neutral

5-ketene barbiturates, to give base peaks attributable to aniline (m/e

93) for the derivatives containing monosubstituted benzene and p- hydroxyaniline (m/e 109) for the 41-hydroxy analogs. Simple o-cleavage of this bond involving loss of the amino moiety forms the corresponding acylium ions RC0+ giving rise to low intensity peaks at m/e 155 for the oxybarbiturates and m/e 171 for the 2-thiobarbiturate derivatives.

Aside from the ion at m/e 69, which is present in the spectra of the 29

2-thiobarbiturates as wp.11 as thp 2-oxo analogs, very few peaks can be

attributed to fragmentations of the pyrimidine ring.

The urine isolate El mass spectrum (Fig. 1.42) is consistent with

the proposed composition of the sample. The peaks at m/e 247 and 279

correspond to the respective molecular ions of 2-oxo-MB and 4'-0H-MB.

The most intense peaks in the spectrum are at m/e 93 and 109.

Structurally significant fragment ions are also shown at m/e 69, 155 and

171. The CI-CH^ mass spectrum of the urine isolate (Fig. 1.43) confirms

that molecular species with approximate masses of 247, 263 and 279 are

present in the sample. The usual pattern of quasi-molecular ions at m/e

M+l, M+29 and M+41 are shown in the appropriate relative intensities

with respect to the M+l ion (Table 1.10). The corresponding CI-CH^ mass

spectra for the synthetically prepared compounds are also shown (Figs.

1.44-1.47) for comparison. For each of these compounds, the base peak

appears as the MH+ ion and fragmentation is minimal. The relative

intensities of the (M+29) and (M+41) ions range from 14-21% and 3-5%,

respectively.

The presence of 4 1 -0H-2-oxo-MB in the sample cannot be unambiguously

ascertained from this mass spectral data. Its molecular ion is

indistinguishable from that of MB in low resolution El mass spectra, since both appear at m/e 263. Likewise, their Cl mass spectral quasi- molecular ion patterns are similar. Furthermore, there are no fragment

ions unique to 4'-0H-2-oxo-MB that are not also found in the spectra of the other components in the sample.

However, as previously demonstrated, the two metabolites believed to be 4'-hydroxy derivatives are subject to selective acetylation, 30

apparently without modification of the non-phenolic compounds.

Therefore, a method was presented to distinguish 4'-0H-2-oxo-MB from MB

by low resolution mass spectrometry. Accordingly, the El and Cl mass

spectra of the material isolated from the metabolite mixture upon

treatment with aqueous potassium carbonate and acetic anhydride were

determined.

The El mass spectrum (Fig. 1.48) shows weak, but observable, peaks

consistent with the molecular ions of 4 1 -0Ac-2-oxo-MB at m/e 305 and

4'-0Ac-MB at m/e 321. Esters typically provide very weak parent ions in

El mass spectra determined at 70 eV, as demonstrated in the El spectra

of authentic 4 ' -0Ac-2-oxo-MB and 4'-0Ac-MB (Figs. 1.49 and 1.50). The

base peaks, which occur at m/e 109 in both spectra, arise from a

rearrangement of the acetoxy moiety with elimination of the neutral

ketene molecule. The resultant fragment ions, corresponding to the

phenolic species at m/e 263 and 279, undergo additional fragmentation to

produce the 4-hydroxyaniline ion at m/e 109, as previously described.

Therefore, the loss of ions at masses corresponding to the phenolic

species is not observed in the El mass spectrum of the acetylated urine

isolate.

However, the Cl mass spectrum of the treated sample clearly demonstrates the presence of acetoxy derivatives (Fig. 1.51). The pertinent spectral data is presented in Table 1.11. As shown in the corresponding spectra of the synthetic compounds (Figs. 1.52 and 1.53), the molecular ion (MH+) of the esters is greatly enhanced, and the higher mass ions are also observed. The intensity of the peak at m/e

280, attributed to the 4'-0H-MB MH+ ion, in the treated metabolite 31

mixture is quite low, while fin intense peak is shown at m/e 322.

Analogously, the intense peak at m/e 306 is consistent with the MH+ ion

of 4'-0Ac-2-oxo-MB formed from acetylation of the phenolic metabolite.

To confirm the structural assignments of the metabolites, the 500

MHz NMR spectra was acquired for the metabolite mixture. The

corresponding spectra determined for the synthetically prepared

compounds are shown in Figs. 1.54-1.61, for which the peak assignments

are analogous to those presented in Table 1.3.

The NMR spectrum from 0.0-14.0 ppm of the urine isolate is shown

in Fig. 1.62. Since all of the resonances for the 5-carboxanilide

barbiturate protons occur at chemical shifts greater than 6 ppm, the peaks at higher field primarily result from impurities of endogenous origin. Although it may not be apparent in this particular

representation of the spectrum, resolution is sufficient to identify every significant peak in the region downfield from 6 ppm, including the aromatic resonances, from chemical shift comparisons with the reference spectra.

The assignments that have been made for the exchangeable protons are shown in the plot of the 6.0-14.0 ppm region (Fig. 1.63). The correlation between the chemical shifts of peaks from components in the urine isolate and reference samples was found to be excellent (Table

1.12). The partially resolved peaks near 9.55 ppm are consistent with the phenolic protons of 4'-0H-MB and 4'-0H-2-oxo-MB. The peak intensities for the exocylic amido protons between 1 1 . 2 and 1 1 . 6 ppm correlate well with the relative concentrations of the respective components, as indicated by HPLC. 32

An expansion of the aromatic region of the spectrum is shown in Fig.

1.64. Two sets of AA'BB 1 coupling patterns, characteristic of the p-

substituted phenyl moiety, and the monosubstituted benzene AA'BB'C

coupling pattern are clearly recognizable. Again, the chemical shift

correspondence of these peaks with those determined from reference

samples of 2-oxo-MB, 4'-0H-MB and 4'-0H-2-oxo-MB is excellent (Table

1.13). As a result of its low concentration, the aromatic resonances of

MB were completely obscured by the metabolites.

These studies conclusively demonstrate that merbarone is extensively

metabolized in man, even though the levels of the metabolites in plasma

are exceedingly low throughout the course of administration. The three major urinary metabolites of the drug have been definitively characterized as 5-(N-phenyl-carboxamido)-barbituric acid, 5-(N-4'-hy- droxyphenylcarboxamido)-barbituric acid and 5-(N-4'-hydroxyphenylcar- boxamido)-2-thiobarbituric acid. Structures of the metabolites were elucidated by UV spectroscopy during HPLC, NMR spectroscopy, El and

Cl mass spectrometry, and synthesis of the proposed compounds. Several minor urinary metabolites remain to be identified. The low levels of these metabolites in plasma suggest that their rates of excretion exceed their rates of formation. Furthermore, since the relative concentration of metabolites in urine is much greater than merbarone, metabolite renal clearance is apparently greater than that of the drug. 33

D. References

1. Brewer, A. D. , Minatelli, J. A., Plowman, J. , Pauli, K. D., and Narayaman, V.L. 5-(N-phenylcarboxamido)-2-thiobarbituric acid (NSC 336628), a novel potential antitumor agent. Blochem. Pharmacol., 34: 2047-2050, 1985.

2. Glover, A., Chun, H. G. , Kleinman, L. M. , Cooney, D. A., Plowman, J., Grleshaber, C. K., Malspeis, L., and Leyland-Jones, B. Merbarone: an antitumor agent entering clinical trials. Investigational New Drugs, 137-143, 1987.

3. Cooney, D. A., Covey, J. M., Kang, G. J., Dalai, M., McMahon, J.B., and Johns, D.G. Initial mechanistic studies with merbarone (NSC 336628). Biochem. Pharmacol., 34: 3395-3398, 1985.

4. Cooney, J. M. , Covey, M. , Dalai, J. , Plowman, J. Kensler, T. W. , Sinha, B., and Johns, D. G. Mechanistic studies with merbarone (MB; NSC-336628). (Abstract) Proc. Am. Assoc. Cancer Res., 27: 276, 1986.

5. Driscoll, J. S., Nannette, R. M. , Quinn, F. R. , Davignon, J.P., Ing, R., Abbott, B. J., Congleton, G., and Dudeck, L. Psychotropic drugs as potential antitumor agents: A selective screening study. Cancer Treat. Rep., 62: 45-74, 1978.

6 . Ukita, T., Kato, Y., Hori, M., and Nishizawa, H. On the anti-tumor activity of nitrogenous cyclic (5-diketones. Chem. Pharm. Bull., 8 : 1021-1028, 1960.

7. Ukita, T. , and Arakawa, K. Antibacterial properties of compounds containing the tricarbonylmethane group. IX. Effects of nitro- geneous radicals in the side-chains of 3-acyl-4-hydroxycoumarins. Chem. Pharm. Bull., 1: 255-260, 1953.

8 . Ukita, T. , and Arakawa, K. Antibacterial activity of compounds possessing a tricarbonylmethane group. XI. Synthesis of a compound having structure analogous to A-ring of tetracycline. Chem. Pharm. Bull., 5: 535-538, 1957.

9. Ukita, T., and Mizuno, D. In vitro screening of tricarbonylmethane and related compounds for their anti-tumor effect by cylinder agar plate (CAP) method. Chem. Pharm. Bull., 8 : 1016-1020, 1960. 34

10. Wiley, R. H., Hussung, K. F., Hobbs, W. E., and Huh, S. Hydroxy-2-thiopyrimidine-5-carboxaldehyde derivatives in cancer chemotherapy. J. Med. Chem., 7: 3.58-359, 1964.

11. Malspeis, L. , Lyon, M. E. , DeSouza, J. J. V., and Staubus, A. E. Preclinical pharmacology studies of merbarone (NSC 336628). (Abstract) Proc. A. Assoc. Cancer Res., 27: 418, 1986.

12. Vogel, A. Vogel's Textbook of Practical Organic Chemistry, 4th ed., New York, NY: Longman, Inc., 1978.

13. Fieser, L. F., and Fieser, M. Reagents for Organic Synthesis, New York, NY: John Wiley and Sons, Inc., 1967.

14. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March, 1986.

15. Schoenenberger, H., Holheu-Eckardt, J. and Baman, E. Antimycotics. III. Comparative studies of the preparation of isomeric hydroxybenzamides. Arzneimittel. Forsch., 14: 324-328, 1964.

16. Mark, L. C. Metabolism of barbiturates in man. Clin. Pharmacol. Therap., 4: 504-530, 1963.

17. Bush, M. T. , Mazel, P., and Chambers, J. The metabolic fate of thiobarbiturates: thiobarbital in man. J. Pharmacol. Exp. Ther., 134: 110-116, 1961.

18. Butler, T. C. The metabolic hydroxylation of phenobarbital. J. Am. Pharm. Assoc., 116: 326-336, 1956.

19. Harvey, D. J., Glazener, L., Stratton, C., Newlin, J., Hill, R. M., and Horning, M. G. Detection of a 5-(3,4-dihydroxy-l,5-cyclohexadien-l-yl) metabolite of phenobarbital and mephobarbital in rat, guinea pig and human. Res. Commun. Chem. Path. Pharmacol., 3: 557-565, 1972.

20. Kallberg, N. , Agurell, S., Ericsson, 0., Bucht, E. , Jailing, B., and Borcus, L.O. Quantitation of phenobarbital and its main metabolites in human urine. Europ. J. Clin. Pharmacol., 9: 161-168, 1975.

21. Whyte, M. P., and Dekaban, A. S. Metabolic fate of phenobarbital. Drug Metab. Disp., 5: 63-70, 1977.

22. Tang, B. K., Kalow, W., and Grey, A. A. Metabolic fate of phenobarbital in man: N-glucoside formation. Drug Metab. Disp., 7: 315-318, 1979. 35

23. Hooper, W. D. , Kunge, H. F., and F.adie, M. J. Qualitative and quant i tative studies of mot.hylphenobarbi ta 1 metabolism in man. Drug Metab. Disp., 9: 381-385, 1981.

24. Leibman, K. C. , and Ortiz, F,. Further studies of metyrapone effects upon anilide hydroxylation. Drug Metab. Disp., 3: 507-512, 1975.

25. Tang, B. K. , Kalow, W. , and Grey, A. A. Amobarbital metabolism In man: N-glucoside formation. Res. Gommun. Chem. Pathol. Pharmacol. 21: 45-53, 1978.

26. Tang, B. K. , Inuba, T. , and Kalow, W. N-Hydroxyamobarbital: The second major metabolite of amobarbital in man. Drug Metab. Dispos., 3: 479-486, 1975.

27. Baler, J. R., Sadler, P. J., Nicholson, J. K., and Timbrell, J. A. Urinary excretion of acetaminophen and its metabolites as studied by proton nmr spectroscopy. Clin. Chem., 30: 1631-1636, 1984.

28. Clapp, J. W. A new metabolic pathway for a sulfonamide group. J. Biol. Chem., 223: 207-214, 1956.

29. Rogers, N. A. J. , and Smith, H. 2-Acylcyclohexane-1:3-diones. Part II. 2-Formyl-, 2-propionyl-, 2-isobutyryl-, and 2-phenyl- carbamoyl-cyclohexane-1 :3-dione, and their conversion into phenanthridines. J. Chem. Soc., 341-346, 1955.

30. Richter, R. and Ulrich, H. 5-Acy1-7-(N-arylcarbamoyl)-4,6 -dioxo- 2,3,3a,4,5,6 -tetrahydrooxa(thia)zo1o-[2,3-c]pyrimidines and 3-[N- arylcarbamoy1]-2,4-dihydroxyquinolines f rom 2-methy1oxa(thia)zol- ine and aryl isocyanates. J. Org. Chem., 44: 4877-4880, 1979.

31. Arbuzov, B. A., and Zubova, N. N. Addition of aliphatic and aromatic acyl isocyanates to unsaturated compounds. Synthesis, 433-450, 1982.

32. Diehl, R. E., Schriber, M. S., and Kantor, S. Dioxocyclohexanecar- boxanilide insecticides and acaricides. U.S. Patent 3,801,630, April 2, 1974.

33. March, J. Advanced Organic Chemistry, 2nd ed., New York, NY: McGraw-Hill Book Company, 1977.

34. Greene, T. W. Protective Groups in Organic Synthesis, New York, NY: John Wiley and Sons, Inc., 1981.

35. Munch-Peterson, J. m-Nitrobenzazide. Org. Syntheses Coll. Vol. IV: 715-717, 1963. 36

36. Newman, N. A., Lee Jr., S. H. , and Garrett, A. B. Solvent effect in the Curtius rearrangement of hen7.a7.tde. J. Am. Chem Soc. , 69: 113-116, 1947.

37. Allen, C. F. H., and Bell, A. Undecyl isocyanate. Org. Syntheses Coll. Vol. Ill: 846-847, 1955.

38. Gibson, G. G. , and Sket, P. Introduction to Drug Metabolism, New York, NY: Chapman and Hall, Inc., 1986.

39. Chattaway, F. D. Acetylation in aqueous alkaline solution. J. Chem. Soc., 2495-2496, 1931. 37

Table 1.1 Reaction yields and appearance of the synthesized 5-carboxanilide barbiturates 1 and 3.

Compound Yield ,a % no. abbrev X R precip recryst m p , °C Appearance

la MB 95 79 off-white s H >30° lb 2-oxo-MB 0 H 94 85 295-6 colorless lc 4 ' -OMe-MB s och 3 99 93 >300 off-white Id 4 r-0Me-2-oxo-MB 0 OCH 3 90 78 >300 colorless le 4 ' -OAc-MB s OCOCH 3 97 84 >300 off-white If 4'-0Ac-2-oxo-MB 0 OCOCH-j 93 80 >300 colorless 3a 4 ' -OH-MB s OH 96 >300 It brown 3b 4'-OH-2-oxo-MB 0 OH 95 >300 It purple

a Yields determined for the product as isolated by precipitation and after recrystallization. b lit. (6 ); mp 293-294°C. 38

Table 1.2 HPLC retention and UV spectral data of the 5-carboxanilide barbiturates 1 and 3. Isocratic, C^g reverse-phase chromatography was performed at ambient temperature employing a mobile phase of 25% methanol (v/v), 0.1 M acetate buffer, 40 mM magnesium sulfate and 1 mM SDS at pH 5.5; flow rate 1.0 mL/min.

no. tfa (min) Xmaxb

la 9,. 1 306, 235 lb 7,.8 278, 250 (sh) lc 8 ,.9 308, 240 Id 7.,0 279, 254 (sh) le 6 ..0 308, 240 If 4. 3 278, 253 (sh) 3a 1 .7 308, 239 3b 1 .2 279, 255 (sh)

a Retention data refer to a particular column. Absorption spectra were determined at chromatographic peak apexes; sh, shoulder inflection. 39

Table 1.3 NMR spectral data of the 5-carboxanilide barbiturates 1 and 3.a

Chemical shift.k 6 Phenol Exocyclic Ureido no. ch3 0-H N-H N-H

la 11.428 (s) 12.978 (bs) lb 11.534 (s) 11.70 (vbs) lc 3. 761 (s) 11.305 (3 ) 12.881 (bs) Id 3. 759 (s) 11.390 (s) 11.58 (vbs) le 2.270 (s) 11.407 (s) 12.974 (bs) If 2.268 (s) 11.511 (s) 11.70 (vbs) 3a 9.579 (s) 11.262 (s) 12.840 (bs) 3b 9.538 (s) 11.327 (s) 11.53 (vbs)

chemical shift .0 6 J.d Hz no. para (1 H) meta (2 H) ortho (2 H) m-p o-m

la 7.225 (t) 7.417 (t) 7.551 (d) 7.39 8.54 lb 7.208 (t) 7.407 (t) 7.540 (d) 7.38 8 . 15 lc 6.974 (d) 7.459 (d) 9.01 Id 6.966 (d) 7.437 (d) 9.00 le 7.175 (d) 7.582 (d) 8.85 If 7.164 (d) 7.561 (d) 8 . 8 8 3a 6.790 (d) 7.328 (d) 8.84 3b 6.777 (d) 7.296 (d) 8.82

a All spectra were measured at 250 MHz for 13.3 mg/mL solutions in DMSO-dg, employing tetramethylsilane as the internal standard, at 25°C; s, singlet; bs, broad singlet; vbs, very broad singlet; d, doublet; t, triplet. Protons exchangeable with D 2 O. c Multiplet centroid; aromatic resonances were well resolved and nearly first order in appearance. Outer peak spacing. Table 1.4 NMR chemical shifts (6 ) for the 5-carboxanilide barbiturates 1 and 3.a

no. 2 4,6 5 7

la 174.87 168.24 168.15 83.63 164.92 lb 148.5 168. 7 80. 1 168.7 lc 174.93 167.88 167.79 83. 19 164.90 Id 148.50 168.35 79.87 168.35 le 174.90 168.24 168. 14 83.65 164.89 If 148.42 168.69 80.29 168.69 3a 175.01 167.75 83.05 164.91 3b 148.44 168.20 79.60 168.20

no. 1 ' 2 ' ,6' 3' ,5' 4' ch3 0C0R

la 135.62 121.24 129.11 125.38 lb 135.6 1 2 1 . 2 129. 1 125.2 lc 128.27 123.07 114.27 156.92 55. 22 Id 128.47 122.99 114.25 156.76 55.21 le 133.12 122.43 122.37 147.58 2 0 . 6 6 169.00 If 133.31 122.34 122.30 147.40 2 0 . 6 6 169.03 3a 126.73 123.22 115.54 155.29 3b 126.79 123.07 115.43 155.04

a All spectra were recorded at 6 8 MHz in DMSO-dg with chemical shifts reported in ppm relative to external tetramethylsilane. 41

Table 1.5 Electron impact mass spectral data of the synthesized 5-carboxanilide barbiturates 1 and 3.

Parent Base ion Deak no. m/e RAa m/e Major ions, m/e (RA) la 263 34. 7 93 171(4.9), 170(6.9), 119(1.3), 92(4.4), 77(5.8), 69(13.5), 66(9.4), 65(6.8) lb 247 27. 7 93 155(3.7), 154(1.1), 119(1.1), 92(4.0), 77(4.1), 69(10.2), 66(9.3), 65(8.0) lc 293 46.9 123 171(6.8), 170(3.3), 149(2.7), 1 2 2 (6 .6 ), 108(72.9), 92(2.0) 80(8.6) 69(18.0)

Id 277 37.6 123 155(3.7), 154(0.7), 149(5.5), 122(5.3), 108(84.3), 92(1.8), 80(9.6), 69(13.0) le 321 4.6 109 279(24.1), 171(5.9), 170(2.1), 135(1.2), 108(15.1), 82(8.9), 81(3.4), 69(17.5)

If 305 8.4 109 263(68.5), 155(4.5), 154(0.7), 135(1.5), 108(10.5), 81(4.0), 80(8.7), 69(12.8)

3a 279 27.5 109 171(5.5), 170(2.2), 135(1.7), 108(11.7), 81(6.5), 80(11.1), 69(17.8)

3b 263 19.2 109 155(3.3), 154(2.7), 135(1.5), 108(10.5), 93(0.5), 81(7.7), 80(12.7), 69(12.8)

a Percent relative abundance. 42

Table 1.6 Chemical ionization mass spectral data of the synthesized 5-carboxanilide barbiturates 1^ and 3.a

(M+l)+ (M+29)+ (M+41)+

no. m/e RAb m/e RA m/e RA

la 264 1 0 0 292 17.64 304 5.46 lb 248 1 0 0 276 15. 14 288 4.14 le 322 1 0 0 350 20.33 362 4.68 If 306 1 0 0 334 16. 18 346 3.90 3a 280 1 0 0 308 18.06 320 4.54 3b 264 1 0 0 292 13.83 304 3.30

a Methane reagent gas. Percent relative abundance.

Table 1.7 Elemental analyses for the 5-carboxanilide barbiturates.

Composition , % Empirical no. formula MWCHN 0 S ,

la 263.28 Calcd 50. 18 1 2 . 18 C 11H 9N 3°3S 3.45 15.96 18.23 Found 49.98 3.30 15.75 18. 72 12.49

lb Calcd 3.67 C 11H9N 3°4 247.52 53.38 16.98 25.86 Found 53. 19 3.63 16.77 25. 73

lc 49. 14 3. 78 10.93 C 12H 11N 3°4S 293.31 Calcd 14.33 21.82 Found 48.99 3. 73 14.37 22.23 11. 19

Id 277.55 Calcd 51.93 3.99 15. 14 28.82 C 12H 11N 3°5 Found 51.85 3.84 14.88 28.71

le 321.31 Calcd 48.60 3.45 13.08 24.90 9.98 C13H11N3°5S Found 48. 73 3.33 12.94 25.15 1 0 . 2 2 if 305.25 Calcd 51. 15 3.63 13.77 31.45 C 13H 11N 3°6 Found 51.34 3.84 14.88 31. 70

3a 11.48 C 11H 9N 3°4S 279.28 Calcd 47. 31 3.25 15.05 22.92 Found 46. 80 3. 33 14.85 23.60 11. 76

3b 263.22 Calcd 50. 19 3.45 15.96 30.39 C 11H9N 3°5 Found 49.88 3.59 16.01 30.69 Table 1.8 Conditions for the HPLC analysis of merbarone and related compounds.

LC: Isocratic

Column: Nova-Pak C^g, 4 ym (3.9 mm x 15 cm)

Precolumn: None

Mobile phase: 25% MeOH, 67 mM NH, -Ac, 33 mM AcOH, 40 mM MgS04 , 1 mM £>DS; pH 5.5

Flow rate: 1.0 mL/min

Detection: UV absorption with diode-array Table 1.9 Sample preparation for HPLC.

Plasma Samples Plasma: 50 pL

Protein precipitation: MeOH-DMSO (85:15, v/v), 250 pL

Centrifugation: 1 2 . 0 0 0 x g, 1 0 min

Supernatant: 2 0 0 pi.

Diluent: Mobile phase (0% MeOH), 235 pL

Injection volume: 100 pL

Urine Samples Urine: 100 pL

Solvent: DMS0, 25 pL

Diluent: Mobile phase, 435 pL

Centrifugation: 12.000 x g, 5 min

Injection volume: 200 pL

Solid Samples Stock solution: 1.0 mg/mL in DMSO

Diluent: Mobile phase

Concentration: Synthetic samples - 10 pg/mL Urine isolate - 50 pg/mL

Volume assayed: 100 pL

Assay method: Urinary sample procedure Table 1.10 Quasi-molecular ions in the chemical ionization mass spectrum of the urine isolate.®

(M+l)+ (M+29)+ (M+41)+

Compoundb m/e RAC m/e RA m/e RA

1 248 1 0 0 . 0 276 13.6 288 3.2 2,3 264 31.4 292 3.0 304 0.8 4 280 18.9 308 2.3 320 0.7

® Methane reagent gas. b 1, 2-oxo-MB; 2, 4'-0H-2-oxo-MB; 3, MB; 4, 4'-0H-MB. c Percent relative abundance.

Table 1.11 Quasi-molecular ions in the chemical ionization mass spectrum of the acetylated urine isolate.®

(M+l)+ (M+29)+ (M+41)+

Compound m/e RAC m/e RA m/e RA

1 248 100.0 276 16.7 288 4.7 3 264 64.6 292 10.3 304 2.4 5 306 65.6 334 8.9 346 2 . 0 6 322 50.0 350 8 . 2 362 1.5

Methane reagent gas. b 1, 2-oxo-MB; 3, MB; 5, 4'-0Ac-2-oxo-MB; 6 , 4 ’-0Ac-MB. ° Percent relative abundance. 46

Table 1.12 Exchangeable proton assignments in the 500 MHz NMR spectrum of the urine isolate. 3

Chemical shift. Hz Assignment urine reference isolate sample0 proton compound

6408.0 6401.0 thiolactam, N-H 4 ' -0H-MB 5992.0 5977.0 thiolactam, N-H 2-oxo-MB 5765.7 5766.6 2° amide, N-H 2-oxo-MB 5717.3 5714.3 2° amide, N-H MB 5666.0 5666.4 2° amide, N-H 4'-OH-2-oxo-MB 5629.0 5630.4 2° amide, N-H 4f-0H-MB 4775.3 4772.3 phenol, 0-H 4'-OH-MB 4756.3 4755.3 phenol, 0-H 4'-OH-2-oxo-MB

a Spectra were measured at 25°C in DMSO-d^. k Chemical shift relative to tetramethylsilane. ° Chemical shifts in the 500 MHz ^H NMR of synthesized compounds acquired under the same conditions.

Table 1.13 Aromatic proton assignments in the 500 MHz *H NMR spectrum of the urine isolate.

Chemical Shift, Hz Assignment urine reference ------isolate sample proton compound

3768. 1 3767.9 a 2-oxo-MB 3760.5 3760.1 a 2-oxo-MB 3712.1 3711.6 m 2-oxo-MB 3704.5 3704.0 m 2-oxo-MB 3696.3 3695.8 m 2-oxo-MB 3666.8 3667.1 0 4 ' -OH-MB 3657.8 3658.3 0 4 ' -OH-MB 3654.3 3654.7 0 4'-OH-2-oxo-MB 3645.3 3645.8 0 4'-OH-2-oxo-MB 3613.1 3612.5 P 2-oxo-MB 3605.8 3605. 1 P 2-oxo-MB 3598.4 3597.8 P 2-oxo-MB 3401.1 3400.3 m 4'-OH-MB 3392.3 3391.4 m 4 ’-OH-MB 3396.6 3396.8 m 4 1 -OH-2-oxo-MB 3387.7 3388.0 m 4'-OH-2-oxo-MB 47

Fig. 1.1. The chemical structure of merbarone.

5,5-disubstituted phenobarbital thiobarbiLurie acid

3hc"''C v ‘‘n H

acetanilide

Fig. 1.2. Some xenobiotics that are functionally related to merbarone. 48

□Ac OAc

SOCI

reflux, 4h

Dloxana

Fig. 1.3. The synthesis of 4-acetoxypheny1 isocyanate 6 . 1 0 0 1

90-

8 0 -

70-

60- 50- 40:

30:

20:

10:

9 Fig. 1.4. Chromatogram of patient plasma obtained on day 5 during the 24 h civ of 1000 mg/m“/d x 5 MB (8.8 min) determined with a diode-array detector at 306 run.

•S' VO Fit* ■ * 13iMB8B088.D Sulval > 343 Tilt i 23 HPI040M

Fig. 1.5. Three-dimensional chromatogram from 0-15 min of patient plasma obtained on day 5 during treatment with MB, 1000 mg/m^/d x 5. Chromatographic peaks: 7.4 min; MB, 8.8 min. La O ri !• « #13 t MBS 8 08 6 . D SwWel • 39 Tl It • 33 HP 1 04 QM

4QO

Wtvtltngth Cnm3 Tim# C m1n 3

Fig. 1.6. Plot of the three-dimensional chromatogram from 6-11 min with enhanced resolution of the patient plasma sample obtained on day 5 during the infusion of MB, 1000 mg/m^/d x 5. Chromatographic peaks: metabolite, 7.4 min; MB, 8.8 min. F1 1 I ♦13iMB88088.D Swivel i 345 Tilt « 25 H P 104 BM

850

400 1 4

T 1 mi» Cmln] Neva length Cnm]

Fig. 1.7. Three-dimensional chromatogram from 0-15 min of pre-dose patient plasma showing the absence of constituents which would interfere with the detection of MB and the metabolite eluting at 7.4 min. 1 . 69 1 00-1

90:

80:

70:

60:

50: 40:

30: 7 . 55 8 . 86 2 0 : 3 . 56

I**'!*** TI T* T* T* 1 i TTi--1--1--T1---1--T*--1--T1 I TTI | I TTI I T| I 0 2 4 6 8 10 12 14 T i me (mi n . )

Fig. 1.8. Chromatogram of 24 h pooled patient urines obtained on day 5 during civ of MB at 1000 mg/m /d x 5 determined with a diode-array detector at 306 nm. Chromatographic peaks: metabolites, 1.3, 1.7, 3.6 and 7.5 min; MB, 8.9 min.

L n Fig. 1.9. Three-dimensional chromatogram from 0-15 min truncated at 100 mAU of 24 h pooled patient urines obtained on day 5 during the infusion of MB at 1000 mg/m^/d x 5. Chromatographic peaks: metabolites, 1.3, 1.7, 3.6 and 7.5 min; MB, 8.9 min. 33 35 MP1B40M

i a a . s

e . 9

H » v 1 »na*h Cnm3

Fig. 1.10. Plot of the three-dimensional chromatogram from 6-11 min at greater sensitivity for the 24 h pooled patient urines obtained on day 5 during treatment. Chromatographic peaks: metabolite, 7.5 min; MB, 8.9 min. F 1 1 i i ♦ 13 IMB00081 .D Swlvo1 I 345 TlIt i 25 HP1040M

890

1 2

T 1 m Cnin3 U v I ongth Cnm3

Fig. 1.11. Three-dimensional chromatogram from 0-15 min truncated at 100 mAU of patient urine obtained prior to dosing. Fig. 1.12. Plot of the three-dimensional chromatogram from 0-2.5 min at 1000 mAU resolution for 24 h pooled patient urines obtained on day 5 during the infusion of MB, 1000 mg/m /d x 5. Chromatographic peaks: urinary constituents, 0.8 min; metabolites, 1.3 and 1.7 min. Wavs 1angth C n m 3 T I in* Cm 1 n 3

Fig. 1.13. Plot of the three-dimensional chromatogram from 0-2.5 min at 1000 mAU resolution for the pre-dose patient urine sample. Chromatographic peaks: urinary constituents, 0.8 min. 59

Pooled raw urine (1000 mL)

Filtration

XAD-2 column (2-3 mL/min)

Urine effluent Aqueous wash MeOH-Dioxane (1:1, v/v) Discarded 500 mL 500 mL Discarded

Evaporate

Residue

Dry: 101°C, P2C>5, 5 mm Hg

Crude Mixture (approx. 60 mg): Drug Metabolites Urinary Constituents

Fig. 1.14. Procedure for the extraction of merbarone and its metabolites from patient urine. 1 . G 8

8 0 - 70- T5 0 r— — 60- rd o tn 50- 30 □ 40" CE E 30-

20 -

1 0 -

0 H 0 2 G 8 10 12 144 T i me (min. )

Fig- 1.15. Chromatogram of the urine isolate determined at 293 nm with a diode-array detector Chromatographic peaks: metabolites, 1.3, 1.7, 4.0 and 7.8 min; MB, 9.2 min. F" 1 !• a 413 IMB8BQ43 . D Suj 1 1 a 29 T1 1 t t 39 HP104 0M

4BB

H » v l « n a*h Cnm3______Tim* Cmlnl

Fig. 1.16. Three-dimensional chromatogram from 0-15 min of the urine isolate. Chromatographic peaks: metabolites, 1.3, 1.7, 4.0 and 7.8 min; MB, 9.2 min. Fl '• • *>3 IHBBBQ43.D GuJ 1 ' • 33 Tilt , ^ HP1040M

a.a

in] T I n» Cm I n 3

Fig. 1.17. Plot of the 1.0-2.25 min region of che urine isolate three dimensional chromatogram. Chromatographic peaks: metabolites, 1.3 and 1.7 min. mflU Sc a 1ed i. .8 Crmtga o snhszd 0--x-B 12 i) eemnd t 9 nm. 293 at determined min) (1.2 -0H-2-oxo-MB ' 4 synthesized of Chromatogram 1.18. Fig. d 0 0 1 30- 50- 80- 90- 40- 60- 701 : 0 1 2 6 1 1 14 12 10 8 6 4 2 0 1. 23 i m CTi me m i n . ) yteie 4'-H2ooM ad h mtblt euig t . mn n h uie isolate. urine the in min 1.3 at eluting metabolite the and -0H-2-oxo-MB ' 4 synthesized mflU Seale d Fig. 1.19. Overlay of the the of Overlay 1.19. Fig. 30" 70- 9 01 20 40" 10 " " 240 260 UV spectra determined at the apexes of chromatographic peaks peaks chromatographic of apexes the at determined spectra 300280 320 4 400 340 6 380 360 of 1 . 69 1001

90-

80-

60-

50- 40- 30-

1 0 1

0 H 10 12 14

Fig. 1.20. Chromatogram of synthesized 4'-0H-MB (1.7 min) determined at 293 nm.

ON Ln yteie 4-HM ad h mtblt euig t . mn n h uie isolate. urine the in min 1.7 at eluting metabolite the and 4'-0H-MB synthesized mflU Sc a 1e d Fig- 1.21. Overlay of the UV spectra determined at the apexes of chromatographic peaks peaks chromatographic of apexes the at determined spectra UV the of Overlay 1.21. Fig- d 0 0 1 90: 70: 80: 50: 80: 20 40: 0-1 : 4 6 280 260 240 ae1nt (nm) Wave 1ength 0 320 300 340 380380 400 of 7 . 82 1 0 0 i

90- 80-

70- 80-

50- 40- 30-

20 -

0 2 4 B 8 1 0 1 2 1 4 T i me (min. )

Fig. 1 .22. Chromatogram of synthesized 2-oxo-MB (7.8 min) determined at 293 nm. 1001

90: 80~

Ti 701 Q) 6 0i (0 0 01 5 0“

D 40“ CE E 301

201

101

0 1 220 240 2G0 280 300 340 3 G 0 380320 400 Wavelength (nm)

Fig. 1.23. Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 2-oxo-MB and the metabolite eluting at 7.8 min in the urine isolate. mRU Seale d i. .4 Crmtga o snhszd B 91 i) eemnd t 9 nm. 293 at determined min) (9.1 MB synthesized of Chromatogram 1.24. Fig. i 0 1 0 80i 901 50i 701 30i 401 20i 1 1 0 y\. 9 . 06 1 21 0 1 4 O' o opud ltn a 92 i () n h uie slt ad yteie M (II). MB synthesized and isolate urine the in (I) min 9.2 at eluting compound mRU Sc a 1ed i. .S Oely f h U seta eemnd t h aee o crmtgahc ek o the of peaks chromatographic of apexes the at determined spectra UV the of Overlay 1.2S. Fig. 4 0 0 1 70- 80 90 501 GO: 301 40i 20i 101 4 0 4 260 240 II 8 300 280 aeegh Cnm) Wavelength 320 340 380 380 400 I MB 1001 I I 4 ' —O M e —MB

0 0

4 0“ IIII 30:

20:

1 0 -

220 240 260 200 300 320 340 360 300 400

Fig. 1.26. Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized MB (I) and 4'-0Me-MB (II). 2 —0X0—MB 4 ' OMe—2—0X0—MB

901

80- II II 701

60i

50i

401

301

201

1 0 i

0-1 220 240 260 280 300 320 360 380340 400

Fig. 1.27. Overlay of the UV spectra determined at the apexes of chromatographic peaks of synthesized 2-oxo-MB (I) and 4 1-OMe-2-oxo-MB (II). 7 . 86 1 0 0 d

901

80“

70:

80-

5 0-

4 0“ 5 3 30“

20 “

1 0 “

0 4 0 2 4 B 8 10 12 14 T 1 me (min. )

Fig. 1.28. Chromatogram of the urine isolate treated with AC 2O-K2CO 2 for phenolic acetylation determined at 293 nm with a diode-array detector. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4'-0Ac-MB, 6.2 min; 2-oxo-MB, 7.9 min; MB, 9.2 min. FI19 • 411iMB0B037•D Bui 1 va 1 a 18 T I U a 20 HP104 0M

380

Hi Cnml

Fig. 1.29. Three'dimensional chromatogram from 0-15 min of the urine isolate treated with AC2 O-K2 CO 2 . Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4'-OAc- MB, 6.2 min; 2-oxo-MB, 7.9 min; MB, 9.2 min. as

830

H*v§1»ngth Cnm3

Fig. 1.30. Plot of the three-dimensional chromatogram from 3.5-11.0 min with enhanced resolution for the AC 2 O-K 2 CO 2 treated urine isolate. Chromatographic peaks: 41 -0Ac-2-oxo-MB, 4.5 min; 4'-0Ac-MB, 6.2 min; 2-oxo-MB, 7.9 min; MB, 9.2 min. 1 0 0 H

90:

80"

7 0- •0 0 601

□ 401 n E 30i

201

1 0 1

0-1 240 260 280 300 320340 360 380 400 Nave 1ength (nm)

Fig. 1.31. Overlay of the UV spectra determined at the ups lope (4.36 min), apex (4.56 min) and downslope (4.76 min) for the peak eluting at 4.6 min in the chromatogram of AC 2O-K2CO 3 treated urine isolate. 1 0 0 -d

90:

80-:

70:

60:

50:

40:

30:

20 :

1 0 :

0-1 220 240 260 280 300 320 340 360 380 400 Nave 1ength (nm)

Fig. 1.32. Overlay of the UV spectra determined at the ups lope (5.90 min), apex (6.17 min) and downs lope (6.40 min) for the peak eluting at 6.2 min in the chromatogram of Ac 20-K2C02 treated urine isolate. onlp (.4 i) o te ek ltn a 79 i i te hoaorm f C A of chromatogram the in min 7.9 at eluting peak the for min) (8.14 downslope isolate. mRU Sc a 1e d i. .3 Oely f h U seta eemnd t h usoe 76 mn, px 78 mn and min) (7.85 apex min), (7.61 upslope the at determined spectra UV the of Overlay 1.33. Fig. 1

50: : 0 B 7 90- 0 0 30- 40: 80: 0 2 1

01 0 0-1 d : : 2 240 220 260 aeegh (nm) Wavelength 280 300 320 340 6 8 400 380 360 2 O-K 2 O C 3 rae urine treated 30 mRU Sc a 1e d i. .4 Crmtga o snhszd 4 synthesized of Chromatogram 1.34. Fig. 1 004 801 901 " B 0 70- 301 501 20i 1

01 4.34 1 0c2ooM (. mn dtrie a 23 nm. 293 at determined min) (4.3 -0Ac-2-oxo-MB 1 0 12 14 O 1 0 0 d

90-

80"

7 0- -0 0) 80- ft) 0 0) 50-

D a: 401 E 301

20i

1 0 1

0-1 240 260 280 300 320 340 380380 400 Wavelength Cnm)

Fig. 1.35. Overlay of the UV spectra determined at the apexes of chromatographic peaks of the compound eluting at 4.5 min for the Ac20-K9C03 treated urine isolate and synthesized 4 1-0Ac-2-oxo-MB. mflU Sc a 1 e d .6 Crmtga o snhszd '0cM (. mn dtrie a 23 nm. 293 at determined min) (6.0 4'-0Ac-MB synthesized of Chromatogram 1.36. . g i F 1 00d 0 0 80: 90: 40: 50- 60: 7 0- 2 10

0 0-1 : : 0 2 ie m ) . n (mi Time 5 . 96 6 8 1 0 1 2 1 44 1 00d

90-

601

501

401

201

1 0 1

0-1 220 240 260 280 300 320 360 380340 400

Fig. 1.37. Overlay of the UV spectra determined at the apexes of chromatographic peaks of the compound eluting at 6.2 min for the AC 2O-K 2CO 2 treated urine isolate and synthesized 4*-0Ac-MB.

00 NJ 83

9 3 .9 100.01

50.0 -

263.0

69.0

77. 51. 3,1.9-,217,,?. , | ,277^8,2^.-8 250 31

Fig. 1.38. Electron impact mass spectrum of synthesized MB. 84

so.

247

151

Fig. 1.39. Electron impact mass spectrum of synthesized 2-oxo-MB. 85

se.

279

n/E

Fig. 1.40. Electron impact mass spectrum of synthesized 4'-0H-MB. Fig. 1.41. Electron impact mass spectrum of synthesized '-0H-2-oxo-MB. 87

100.0 93.0

50.0- 109.0

69.0

55,0 .0 45i:,^iiiiiii.i ill-- 'lli'l-■*!?.■* II.'3?-8 - W - i l7l 185.1 231.1 tl/E 60 80 X 120 140 160 180 220 240

50.

247.1

260 280 360 380

Fig. 1.42. Electron impact mass spectrum of the urine isolate. 88

50.0 -

94.1

110.1 129.0 0 T ■i, 1..... 1 ... ff: U .flfcU. ■ ISW .'fl I tt/E 10 60 100 120 140 160 180 220 1O0.0-I 24?'1

50.0-

264.0

280.0

1 . . . A . 'i •"■I... tVE 240MO 260 280 3M 321 341 361 381 4^0 420

Fig. 1.43. Chemical ionization (CH^) mass spectrum of the urine isolate. 89

100.0 2S4.0

5 0.0-

292.0

199.0

304.0 171.0 ,2 ?*‘ 139.0 17[ 2 1 1 .0 270.0 » I ■ Fp*1 1■ Ii •' "I*i •» ji i|i i-— . m . -rSt, , . i— .— ,— .— | — i | ■ | I ■ iTgaenr l l •— I • I — 1 I a C IS* 2*t 25250$ m

Fig. 1.44. Chemical ionization (CH^) mass spectrum of synthesized MB. 100.0n 248.0

50.0

276.1

12?.0 155.0 173.0 i 262.0 L t t 1i—i* i i ■ •!—■—i—■—r-J—i—■—r=-'— i ■ i r—] '“ i • i - ‘ rtr-H"1 i ■ | '— r -1— i—' 1— I— crij— >— r~i— I— I ■■ li*— I— I— '— r n/E 108 150 200 250

Fig. 1.45. Chemical ionization (CH^) mass spectrum of synthesized 2-oxo-MB.

vO o 91

279.9

119.9

170.9 248.9 138.1

159 359

Fig. 1.46. Chemical ionization (CH4) mass spectrum of synthesized 4'-0H-MB. 92

2M .0

50.0-

110.0 292.1

3 M . I 278.0 ii t9. ,L,L, tVl X 159 2M 250 2

Fig. 1.47. Chemical ionization (CH^) mass spectrum of synthesized 4'-OH-2-oxo-MB. 93

100.0 93.0

50.0 109.0

69.0

III ■il.lt.M ,13g.0 H?-», ( »7t.0 l8g.e Kf., 100 120 141 160 180 2M 220 240 [00.01

50.0-

24;

279.1

(VE

Fig. 1.48. Electron impact mass spectrum of the AC 2O-K 2CO 3 treated urine isolate. 94

263.0

5 0 .0 -

68.9

154.9 305.

91.9 L'9,;ig.g ,211;9 , „ rvE » 2S0

Fig. 1.49. Electron impact (70 eV) mass spectrum of synthesized 4'-0Ac-2-oxo-MB. 95

109.0

5 0 .0 -

279.0

60.9

80.9 59.9 178.9 T i 1» ■' 3 1 1 if ■ <1 r ii* i , ■ W yl I - ■ tt /I 5« X 151 250

Fig. 1.50. Electron impact (70 eV) mass spectrum of synthesized 4 ' -OAc-MB. 50.0- 152.0 129.0 94.0

110.0

213.0 120 MO 160 180 220 248.0

306.0

276.1

260 280 380

Fig. 1.51 Chemical ionization (CH^) mass spectrum of the AC 2O K 2CO 3 treated urine isolate. 100.0-1 306.0

50.0-

152.0 33 .1

89.1 n ?'0 120.0 .18M 248.1 il 291.9 , 320.0 , 34?'1 ■ i 1 I ■ 3 ■ i ■ Ti ■ .| ■ 1=1 i—1— 1=1 ■ * fc* ■ i 1 I 1 i 1 i 1 i " r ' M ■' 115. i—■—i ■—r-1— | :ci| i i i i i—ct]—T r—i— i—i—>—r fl/E 150 200 250 300 350

Fig. 1.52. Chemical ionization (CH^) mass spectrum of synthesized 41-0Ac-2-oxo-MB. 322.0 100 .0 -i

50.0-

152.0 350.1

110.0 171 0 199*8 290.0 89.0 J J29. H i i a sr J .d ia4: -nlixf'f. ?i,<.g, c j— & >*, i i—l i t=> I -*■—* *- r~-*-r p " i=j i ’” i tl/E 150 200 250 300 350

Fig. 1.53. Chemical ionization (CH^) mass spectrum of synthesized 4'-0Ac-MB.

vO oo J l

”T lil.U t p m

Fig. 1.54. The 500 MHz NMR of synthesized MB from 0.0-14.0 ppm.

NO NO 100

____J

PPM

Fig. 1.55. Plot of the 500 MHz NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized MB. r “nr^ T "n"*" • i • ■ ~~T~m 15.0 it.tt 10.d 7.0 C. 0 1.4 2.4

Fig. 1.56. The 500 MHz NMR of synthesized 2-oxo-MB from 0.0-14.0 ppm. 101 Fig. 1.57. Plot of the 500 MHz 1H NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized 2-oxo-MB. ______A_

a l — • ■ • |- — i ' I...... I ' ' '— '— ■ " | ■ i ' ■■ ' |...... -)« . r----| ■—-.. — I I ‘ ' I ' ' '—' ,,p.I...... I- H . 0 1 Z. 0 II. 0 10.H 0.0 0.0 7.0 0.0 0.0 «.0 5.0 2.0 1.0 PPH

Fig. 1.58. The 500 MHz iH NMR of synthesized 4 1-OH-MB from 0.0-14.0 ppm. 104

i------1------1------1------1------,------r------r------1------r------r- 7.6 7.5 7.* 7.3 7.2 7.1 7.0 6.9 6.8 6.7 PPM

Fig. 1.59. Plot of the 500 MHz *H NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized 4'-0H-MB. JLJl

I I . i) *

Fig. 1.60. The 500 MHz NMR of synthesized 4'-0H-2-oxo-MB from 0.0-14.0 ppm. 105 106

PPH

Fig. 1.61. Plot of the 500 MHz NMR from 6.6-7.7 ppm showing an expansion of the aromatic region for synthesized 4'-0H-2-oxo-MB. r ,'T- -r 13.4 II. * ?. * CH PPM

Fig. 1.62. The urine isolate 500 MHz NMR spectrum from 0.0-14.0 ppm. 107 MB 2-oxo-MB 4'-0H-MB 4'-OH-2-oxo-MB

IS.I I2.S II.I II. 5 II. I PM

Fig. 1.63. Plot of the urine isolate 500 MHz NMR spectral region from 6.6-14.0 ppm with assignments for the exchangeable protons. 108 109

II II 2-oxo-MB 4'-OH-MB 4'-OH-2-oxo-MB

7.6 7.5 7.3 7.2 7.0 6.9 6. 3 G.7 PPM

Fig. 1.64. Plot of the urine isolate 500 MHz XH NMR spectrum from 6 .6-7.7 ppm showing an expansion of the aromatic region and peak assignments. CHAPTER II

ANALYSIS OF MERBARONE AND ITS METABOLITES IN URINE BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY

A. Introduction

The three primary metabolites of merbarone (MB) which are excreted

in the urine of patients were identified in Chapter I as

4'-hydroxy-2-oxo-desthiomerbarone (41-0H-2-oxo-MB), 4'-hydroxymerbarone

(4'-0H-MB), and 2-oxo-desthiomerbarone (2-oxo-MB). Quantitative

information regarding the concentrations of these compounds in patient

urine was desired for evaluating the significance of metabolite

formation and urinary excretion as a route of elimination.

The polarity of the hydroxylated metabolites precluded the direct

use of the HPLC method which was developed for MB, as these compounds were not sufficiently resolved from endogenous urinary constituents

eluting near the solvent front. In order to permit the simultaneous determination of MB and the major metabolites in urine, a method was developed in which the phenolic metabolites in the sample were selectively acetylated. The method was based upon the rapid and quantitative acetylation of phenolic compounds in aqueous alkaline solution with acetic anhydride (1). The derivatization procedure did not affect MB nor 2-oxo-MB and therefore, was specific for the

4'-hydroxylated metabolites. The acetylated derivatives were separated

110 Ill

from Interfering components, the drug and 2-oxo-MB upon HPLC under the

conditions of the method previously developed for MB (2) with only

slight modification of the mobile phase.

In this work, an HPLC method is described for the concurrent assay

of MB and its three oxidative metabolites in a single urine specimen.

After the addition of an internal standard and a simple pre-injection

derivatization procedure in which the phenolic metabolites were

acetylated, the urine was diluted with the mobile phase containing

additional methanol and an aliquot was injected onto the chromatographic

column. Chromatography was carried out under isocratic conditions and

the four compounds were monitored by UV detection at 293 nm. Employing

100 yL of urine, the lowest concentration on the standard curve for each

compound was 0.25 yg/mL. The method was applied to the quantitation of

the drug and the three principal urinary metabolites in the urine of

patients treated according to the two dosing schedules employed in the

phase I evaluation of MB. The urinary excretion of these compounds is

described in Chapter III.

B. Materials and Methods

1. Reagents and Chemicals

Synthetic samples of merbarone (NSC 336628) and its principal

urinary metabolites, 2-oxo-2-desthiomerbarone (NSC 366236), 4'-hydroxy-

merbarone (NSC 380960) and 4'-hydroxy-2-oxo-2-desthiomerbarone were

synthesized as previously described in Chapter I. The internal

standard, 3'-F-merbarone (NSC 372106) was obtained from Dr. K. D. Pauli,

Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, 112

Division of Cancer Treatment, NCI.

The ammonium acetate, acetic acid and anhydrous potassium carbonate

were analyzed reagents (J. T. Baker Chemical Co., Phillipsburg, NJ).

The methanol was glass distilled (Omnisolv grad?, E.M. Science, Cherry

Hill, NJ). Reagent grade dimethylsulfoxide, DMSO, (MCB, East

Rutherford, NJ), analytical reagent magnesium sulfate heptahydrate

(Mallinkrodt, Paris, KY) and certified A. C. S. acetic anhydride

(Fisher, Fairlawn, NJ) were used. Dodecyl sodium sulfate was purchased

from Eastman Kodak Co. (Rochester, NY). The water was double distilled

and deionized and filtered through a 0.2 ym Nylon-66 filter (Rainin

Instrument Co., Woburn, MA) before use.

2. Apparatus

Chromatography was performed using a model 114M pump (Beckman

Instruments, Berkeley, CA), a WISP 712 automatic injector (Waters

Associates, Milford, MA) and a model 3390A recording integrator

(Hewlett-Packard, Avondale, PA). A variable wavelength Spectroflow 783

programmable absorbance detector (ABT Analytical, Kratos Division,

Ramsey, NJ) containing a 12 yL flow cell (path length 8 mm) was used to monitor UV absorption.

The system was equipped with a stainless steel 3.9 mm x 15 cm

column, packed with 4 ym Nova-Pak Cjg (Waters Associates, Milford, MA), and with a 0.5 ym post-injection filter (Rainin Instrument Co.). Since resolution was unacceptable when a precolumn was incorporated in the system, the analytical column was routinely regenerated after approximately 100 samples were assayed, by the following procedure. The column was reversed and flushed at 1.0 mL/rain with 300-500 mL of water 113

followed by methanol (300-500 ml.), and then methanol-water (25:75, v/v)

for one hour. The analytical mobile phase was then passed through the

column until equilibration was achieved, which required approximately 8

hours at 1.0 mL/min.

Samples were centrifuged at 12,000 x g in an Microfuge E microcen

trifuge (Beckman Instruments). UV spectra during chromatography were

obtained with a second analytical system configured as previously

described (Chapter I).

3. Standard Solutions

Stock solutions of the four analytes, 4'-0H-2-oxo-MB (I), 4'-0H-

MB (II), 2-oxo-MB (III) and MB in DMSO were prepared to provide a

concentration of approximately 1.0 mg/mL for each compound. The

individual compounds, 5.0 mg, were weighed on a Cahn 25 electrobalance,

transferred to the same 5 mL volumetric flask and dissolved in DMSO.

Stock solutions of the internal standard in DMSO were prepared to

afford a concentration of approximately O.t mg/mL. The internal standard, 3'-F-MB (1.0 mg, 3.56 pmol), was weighed on the electrobalance and dissolved in DMSO (10.0 ml.) in a volumetric flask.

Dilution of 800 pL of this solution to 5.0 mL with DMSO in a volumetric flask provided the solution (16.0 yg/mL) used in the assay.

Urine standards were prepared by pipetting drug-free urine into 1.5 mL polypropylene microcentrifuge tubes (VWR Scientific, Inc., San

Francisco, CA) and adding varying volumes of the analyte stock solution.

Further dilution of these standards with drug-free urine afforded the range of concentrations employed. Urine standards were prepared using the following volumes: 114

Analyte Ur ine Tube Urine stock soln standard no. (mL) (VL) (mL)

1 0.5 5 0 2 0.5 4 0 3 0.5 3 0 4 1.0 4 0 5 0.5 0 0.5 (tube //3) 6 0.5 0 0.5 (tube //4) 7 0.5 0 0.5 (tube #5) 8 0.5 0 0.5 (tube it6) 9 0.5 0 0

The resultant concentration of each analyte in the urine standards are summarized below for a stock containing I, 1.010 mg/mL; II, 1.008 mg/mL;

III, 1.007 mg/mL; MB, 1.012 mg/mL.

Analyte concentration, yg/mL Tube no. I II III MB

1 10.022 10.004 9.992 10.044 2 8.018 8.003 7.994 8.035 3 6.025 6.014 6.007 6.038 4 4.025 4.018 4.013 4.034 5 2.012 2.009 2.006 2.017 6 1.006 1.004 1.003 1.008 7 0.503 0.502 0.502 0.504 8 0.252 0.251 0.251 0.252 9 0.0 0.0 0.0 0.0

4. Optimization of Reaction Conditions

Aliquots of the 8.0 yg/mL urine standard were prepared and assayed as described in the sections below with the following exceptions: (1) the effect of potassium carbonate concentration was 115

evaluated by employing solutions of the reagent which ranged from 0-3.0

M; (2) for each alkali concentration, the volume of acetic anhydride was

varied from 0-25 yL. Samples for each set of reaction conditions were

prepared in duplicate. The analyte to internal standard peak area

ratios were calculated for the four components. If the individual peak

area ratios deviated from the mean of the duplicate runs by more than

10%, a third trial was performed.

5. Sample Preparation

To 100 yL of urine in a 1.5 mL polypropylene microcentrifuge

tube 25 yL of the internal standard stock solution was added. Aqueous

potassium carbonate (25 yL, 1.2 M) and acetic anhydride (10 yL) were

added to the tubes and immediately mixed by vortexing for 1.0 min.

After approximately 5 min, 400 yL of diluent consisting of a solution of

methanol (35%, v/v) with ammonium acetate (67 m M ) , acetic acid (33 m M ) ,

MgSO^^l^O (40 mM) and sodium dodecyl sulfate (1 mM) was added. The

tube was again mixed by vortexing and then centrifuged for 10 min in a

microcentrifuge at 12,000 x g. The solution was transferred to a 250 yL

insert, which was placed in the automatic sampler. The injection volume

was 30 yL.

For urine samples containing concentrations exceeding the range of

the standard curve, an aliquot of the sample was diluted with drug-free urine to make the total volume 100 yL. 116

6. Chromatographic Conditions

The chromatography was isocratic, employing a mobile phase

consisting of methanol-aqueous phase (25:75, v/v) containing ammonium

acetate (67 mM), acetic acid (33 mM), magnesium sulfate (40 mM), and

sodium dodecyl sulfate (1 mM). Degassing was effected by placing the

solution in an ultrasonic bath for 15 min prior to chromatography. The

flow rate was 1.0 mL/min. The wavelength used for detection was 293 nm

(5 nm bandwidth) with an integrator threshold of 2 and 0.2 cm/min chart

speed.

7. Quantitation

The peak area ratios of each analyte to the internal standard

were plotted against the analyte concentration. The best fit straight

line was determined by least squares regression using a weighting factor

of the inverse peak area ratio squared to calculate the slope, y-

intercept and correlation coefficient (r). The concentration of the

analytes in the unknown samples were calculated using the results of the

corresponding regression analysis.

8. Relative Recovery

The relative recovery of each analyte was calculated from the backcalculated concentrations.

9. Stability of the Acetate Derivatives

The stability of the acetate derivatives of 4'-0H-2-oxo-MB and

4'-0H-MB in the assay solution was examined at ambient temperature for a period of 24 hours. Samples of the 8.0 yg/raL urine standard were individually prepared in duplicate as described above. The solutions 117

were allowed to stand in the automatic sampler until injected at

intervals of 0.5, 1.0, 2.0, 4.0, 8.0 and 24.0 h subsequent to diluent

addition.

C. Results and Discussion

Potentially, the HPLC method developed for the determination of

merbarone in plasma (2) could also be used to assay the concentrations

of the desulfurated metabolite, 2-oxo-MB, in urine. A chromatogram

determined at 293 nm of patient urine, collected on day 5 during p treatment with 1000 mg/m /d x 5 of MB by the 24 h civ schedule,

illustrates that peaks due to the drug (8.9 min) and 2-oxo-MB (7.6 min)

are sufficiently resolved to permit quantitation (Fig. 2.1). Both

compounds appear to be well separated from urinary constituents, as

substantiated by viewing the three-dimensional chromatograms of this

sample (Figs. 1.9 and 1.10). Furthermore, interfering components at

these retention times are not present in a three-dimensional

chromatogram of drug-free urine obtained from the same patient (Fig.

1.11).

When synthetic samples of the three oxidative metabolites had been prepared, a chromatogram determined at 293 nm of an aqueous solution

containing these compounds and the drug demonstrated that the polar hydroxylated metabolites, 4'-0H-2-oxo-MB and 4'-0H-MB, elute at

approximately 1.4 and 1.8 min, respectively (Fig. 2.2). Comparison with

Fig. 2.1 indicates that the hydroxy compounds have similar retentions to those of UV absorbing urinary constituents. The endogenous compounds

interfere with the quantitation of 41-OH-MB and almost completely mask 118

the presence of 4 1-0H-2-oxo-MB. This w a s also clearly demonstrated by

the three-dimensional chromatogram of the patient urine sample, plotted

with truncation at 100 mAU (Fig. 1.9). Three peaks were observed when

the initial region of the chromatogram was viewed without truncation at

1000 mAUFS, with retentions of 0.8, 1.3 and 1.7 min (Fig. 1.12). The

latter peak is due to 4'-0H-MB. Comparison with a similar plot of the

chromatogram for patient pre-dose urine (Fig. 1.13) indicated that the

peak at 0.8 min arises from urinary constituents and that the peak at

1.3 min in the sample obtained during treatment, which was not present

in the pre-dose specimen, is due to the polar metabolite identified as

4'-OH-2-oxo-MB.

Studies were therefore undertaken to develop a method to

simultaneously assay MB and the three metabolites in urine. Earlier

investigations showed that changes in the composition of the mobile

phase and/or the use of two columns in tandem were unsatisfactory.

Gradient elution was not considered attractive due to the excessive time

required for re-equilibration of the system. However, in studies

associated with merbarone metabolite characterization (Chapter I), it was observed that acetate esters of the 4'-hydroxy metabolites exhibited

retention times which were significantly longer than that of the free phenolic compounds. Accordingly, the development of a liquid

chromatographic assay based upon the prior conversion of the polar hydroxy metabolites to the less polar esters was undertaken.

The rapid and quantitative acetylation of phenols in alkaline

aqueous solution (eq 2.1), as described by Chattaway (1), presented a convenient method to derivatize the metabolites directly, without 119

pre1iminnry extraction from urine. In the reported procedure, phenolic

compounds were dissolved in an aqueous solution containing a 50% molar

excess of sodium or potassium hydroxide, which was sufficient to leave

the medium slightly alkaline upon completion of the reaction. A 25%

molar excess of acetic anhydride (AC2 O) was then added to the ice cold

solution and the mixture immediately shaken. The reaction was

apparently instantaneous and quantitative.

OH 0 — c — ch3

ACpO (eq 2.1) 0 °. 1001

Preliminary experiments to assess the applicability of this

procedure were conducted utilizing an aqueous solution of synthetic

4'-0H-MB. Chromatography was performed according to the conditions of

the MB assay, which were essentially the same as listed in Table 2.3,

except that the mobile phase contained 30% (v/v) methanol and the wavelength of detection was 306 nm. Chromatograms of authentic 4'-0H-MB

and 4'-0Ac-MB obtained upon direct injection of their aqueous solutions are shown in Fig. 2.3. The phenolic compound eluted immediately after the solvent front at 1.8 min and the acetate derivative had a retention time of 5.3 min. As illustrated in the chromatogram determined after alkalinizing the 4'-0H-MB solution with an equal volume of 1.0 M K0H prior to the addition of AcjO (Fig. 2.3C), phenolic analogs of MB are subject to acetylation under these conditions. However, the derivative 120 peak at 5.8 min was badly tailed, which may be the result of introducing the alkaline solution directly into the HPLC system, since precipitation of insoluble magnesium hydroxide occurred upon dilution of the reaction mixture with mobile phase. When the alkali concentration was decreased by 10-fold, the reaction did not go to completion. Therefore, without neutralization prior to chromatography, potassium hydroxide was not a satisfactory alkalinizing agent for this particular application of the reaction.

For this reason, the effectiveness of other bases to catalyze the acetylation of 4'-0H-MB by this method was examined. Although distortion of the derivative peak did not result when the reaction was performed in the presence of sodium bicarbonate, complete conversion of the substrate was not achieved (Fig. 2.4A). Potassium carbonate was found to be an acceptable reagent in all respects, as shown in a chromatogram determined by injecting the reaction mixture directly onto the column several minutes after treating an aqueous 4'-0H-MB solution with the base and AC 2 O (Fig. 2.4B).

The applicability of this procedure for concurrently assaying the drug and its major urinary metabolites was initially demonstrated by similarly treating urine obtained from a patient during the infusion of

MB. A comparison of the chromatograms determined before and after derivatization, shown in Fig. 2.5, suggested that the phenolic compounds were converted to the corresponding esters without affecting 2-oxo-MB or

MB. Inasmuch as the peaks for the components of interest were relatively well resolved from each other and urinary constituents, sample clean-up procedures were apparently not necessary. 121

The analytical mobile phase developed for the plasma assay was used

in these studies with modification only of the methanol strength, which

was systematically varied to optimize the resolution of the four

chromatographic peaks. Chromatography of derivatized urine obtained

from a patient during dosing was performed with mobile phases containing

25, 30 and 35% (v/v) methanol at a flow rate of 1.0 mL/min after

allowing the system to equilibrate (Fig. 2.6). Visual inspection,

supported by estimation of the chromatographic parameters, indicated

that the best overall separation was achieved with a mobile phase

containing 25% methanol. Significant improvements in resolution were

not observed as the flow rate was varied.

Optimization of the reaction conditions was achieved by varying the

amounts of I^COg and AC 2 O added to drug-free urine spiked with

4'-0H-2-OXO-MD, 4'-0H-MB, 2-oxo-MB and MB at ambient temperature. The

solutions were assayed by HPLC upon dilution with mobile phase and

internal standard 5 min after the reaction was initiated by the addition

of AC 2 O. The analytical data are presented in Table 2.1. No apparent

trend was evident in the 2-oxo-MB and MB peak area ratios for the 21 assayed sets of reaction conditions. The mean and standard deviation

for the 2-oxo-MB peak area ratios was 2.5506 ± 0.1065 and 2.6123 ±

0.1248 for MB. Therefore, neither of these compounds were affected by the reaction.

Optimal conditions for derivatization resulted when the alkalinizing reagent concentration was 1.2 M with 5-10 yL of AC2 O. A maximum in the peak area ratio of the acetate derivatives occurred as the AC2 O volume was increased for reactions in which the concentration of added I^COg 122 was the same, but less than 1.5 M. Prncipitat1 on resulted upon addition of the mobile phase diluent when 1.5 and 3.0 M K2C03 solutions were used, except for runs at the former alkali concentration with 25 yL of

AC2O.

The data may be qualitatively interpreted by considering that the anhydride is subject to two competing reactions in these systems. Base catalyzed attack by the phenolic compound affords the acetate ester and an equivalent of acetic acid (eq 2.2). However, the requisite alkaline conditions result in appreciable hydrolysis of the anhydride (eq 2.3).

Neutralization of the acetic acid formed in both reactions consumes hydroxide ions, and the pH of the solution decreases. The desired reaction will be inhibited if the rate of pH reduction greatly exceeds that of esterification, resulting in less than quantitative derivatization of the phenolic compounds.

R-Ph-OH + CH3-§-0-§-CH3 ---- ► R-Ph-0-§-CH3 + CH-j-P-OH (eq 2.2)

CH3 -P-0-P-CH3 + H 20------+ 2 CH3-P-0H (eq 2.3)

Additional evidence to support this interpretation was obtained by measuring the pH of these solutions. With the exception of the blank run, for which neither alkali nor Ac20 were added, the initial pH of the solutions, measured prior to anhydride addition, was at least 10. The pH of reaction mixtures ranged from 4-6 approximately 5 min subsequent to the addition of Ac20 for runs in which the concentration of added

K2C03 was 0.06-1.2 M. For the runs with higher alkali concentration, 123

the pH had not significantly diminished, and precipitation was observed

upon addition of mobile phase.

The extent of phenolic acetylation under the optimum reaction

conditions was evaluated by similarly treating an aqueous reference

mixture of MB and the three metabolites, which permitted peaks due to

the 4*-hydroxy analogs to be detected in the absence of interfering

urinary constituents. Chromatography with diode-array detection was

performed employing the mobile phase containing 25% (v/v) methanol. The

procedure resulted in four well-resolved peaks, each of which was

readily detected at 293 nm (Fig. 2.7). The three-dimensional

chromatogram of the derivatized sample (Fig. 2.8), and a segment of the

chromatogram from 3-11 min with enhanced sensitivity (Fig. 2.9), further

demonstrate the resolution of the peaks due to the derivatives.

Comparison with the three-dimensional chromatogram of an underivatized

reference sample (Fig 2.10) indicated that the procedure resulted in

essentially quantitative conversion of the phenolic compounds to the

corresponding esters.

Identical capacity factors were determined, respectively, for the

resultant derivatives of 4'-0H-2-oxo-MB and 4'-0H-MB and synthetic

samples of 4'-0Ac-2-oxo-MB and 4'-0Ac-MB (Fig. 2.11). The similarity

between the UV spectrum of each derivative and the corresponding

synthetic compound was demonstrated by overlaying the spectra determined

at the apexes of the chromatographic peaks (Figs. 2.12 and 2.13).

Furthermore, the NMR spectra of material isolated upon treating the synthetic 4'-hydroxycarboxanilide barbiturates with AC 2 O-K2 CO2 revealed that the derivatives were indeed the 4'-acetoxy analogs (Figs. 2.14 and 124

2.15). Again, neither MB nor 2-oxo-MB were affected by the

derivatization reaction, as illustrated In overlays of UV spectra

corresponding to their peaks in the chromatogram following

derivatization with those in the untreated sample (Figs. 2.16 and 2.17).

The three-dimensional chromatogram of a derivatized specimen of 24 h pooled patient urine collected on day 5 during the 120 h civ of MB at

1000 mg/m /d (Fig. 2.18), and a section of the chromatogram displaying

the peaks of interest (Fig. 2.19), demonstrate the applicability of the

acylation procedure to patient urine. Peaks due to the esters, 2-oxo-MB and the parent drug were resolved, and comparison with a similarly treated pre-dose urine sample revealed that the region in which these compounds elute is essentially free of Interfering peaks (Fig. 2.20).

The urine sample derivatization procedure and chromatography are summarized in Tables 2.2 and 2.3. The Internal standard used in the plasma assay, 31-F-merbarone, was also used for the assay of urine.

Chromatograms of the urine standards and patient samples were obtained using a variable wavelength detector set at 293 nm and an integrator for measuring the areas under the chromatographic peaks. The wavelength was selected as intermediate between the Xmax in the mobile phase of the UV absorption bands of the 2-thio compounds (MB and 4'-0H-MB) at 306 nm and the 2-oxo-desthio compounds (4'-0H-2-oxo-MB and 2-oxo-MB) at 278 nm.

Chromatograms of several standard curve samples after derivatization are shown in Figs. 2.1 and 2.2. Components interfering with the internal standard, which usually eluted near 14.7 min, were not observed in drug-free urine samples. Urine specimens obtained during and after infusion of the drug showed no interfering metabolites in this region. 125

Chromatograms of n 24 h pooled urine sample, collected on day 5 from a 2 patient treated with 190 mg/m /d x 5 of MB according to the 2 h civ

dosing schedule, resulting from direct injection and after

derivatization, illustrate that the hydroxylated metabolites can be

quantitated subsequent to acetylation (Fig. 2.23).

Figs. 2.24-2.27 represent typical calibration curves for

4'-OH-2-oxo-MB, 4'-0H-MB, 2-oxo-MB, and MB, respectively, in which the

ratio of the peak area of each compound to that of the internal standard

is proportionate to the analyte concentration. The best fit lines of

the calibration curves having a range of concentrations from 0.25 to

10.0 yg/mL were obtained by linear regression analysis employing a

weighting factor of inverse peak area ratio squared. Thq correlation

coefficients were generally greater than 0.998.

Urine standards, prepared on 10 days during a 3.5 month period from

single stock solutions containing each of the metabolites and MB, were

used for calibration curves employing the same analytical column. The

backcalculated concentrations obtained for these standards are shown in

Tables 2.4-2.7. The coefficients of variation ranged from 3.36 to 8.94%

for 4'-OH-2-oxo-MB, 2.31 to 8.80% for 4'-0H-MB, 2.81 to 10.86% for

2-oxo-MB, and 2.86 to 6.24% for MB. The analytical method is considered

to be reproducible inasmuch as these backcalculated concentrations were obtained during an extended period. The coefficient of variation of the slopes of the 10 linear regression curves were 4.36% for 4'-0H-2-oxo-MB,

4.94% for 4'-0H-MB, 5.34% for 2-oxo-MB, and 7.45% for MB. Accordingly, the assay proved to be consistent during the period. 126

For a given column, the. within run retention times of the drug,

metabolites and internal standard generally did not vary greatly (<5%).

The analytical column was not protected by a precolumn Cjg cartridge

since peak resolution was deleteriously affected. The only protection

employed was a 0.5 ym post-lnjectlon filter. A procedure was adopted

for regularly flushing and regenerating the analytical column. The

flushing method was routinely used during the assay of patient urine

samples where 3 columns were alternately employed for the chromatography

of about 1500 samples without significant loss of peak resolution.

The relative recovery of each of the metabolites and MB from urine

and the reproducibility of the assay for the range of concentrations of

the standard curves determined during the 3.5 month period are

summarized in Tables 2.8-2.11. The recovery ranged from 97.8 to 106.0%

for 4'-OH-2-oxo-MB, 96.9 to 103.7% for 4'-0H-MB, 98.4 to 103.0% for

2-oxo-MB, and 98.3 to 103.8% for MB of the amount of each of the

compounds added to the urine samples.

The volume of urine used in the assay was 100 yL and with these

samples, the lowest concentration of the standard curve was 0.25 yg/mL.

The lower limit of the calibration curve proved to be adequate for all

of the patient urine specimens examined, and therefore, no effort was

made to enhance the sensitivity. Approximately 10 h were required to

assay a typical series of patient urines in duplicate together with a

standard curve, since the chromatographic run time was 20 min. As

substantiated by the data in Table 2.12, the stability of the acetate

derivatives in the processed sample solution was sufficient to permit overnight assaying with an automatic injector. In summary, this HPLC nssny permits the quantitatIon of merbarone and its three principal urinary metabolites in a single human urine specimen. The method employs derivatization of the hydroxylated metabolites in order to separate these components from endogenous interfering constituents. The lowest concentration on the calibration curve for each of the compounds was 0.25 pg/mL, which was found to be adequate for the quantitation of the compounds in the urine obtained from all of the patients in the phase I trial. The method was shown to be specific and reproducible. 128

D. References

1. Chattaway, F. D. Acetylatlon in aqueous alkaline solutions. J. Chem. Soc., 2495-2496, 1931.

2. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March, 1986. 129

Table 2.1 Optimization of the reaction conditions for the acylation of 4'-hydroxy analogs of mnrharone in aqueous solution with acetic anhydride and potassium carbonate.

Peak area ratio® KoCOob AC 2 O CM) (PL) I II III MB

0.00 0 2.4820 2.5208

0.06 0 2.5189 2.5618 0.06 1 2.0497 1.6998 2.5359 2.4697 0.06 5 1.2550 1.0057 2.7148 2.5819 0.06 10 1.4364 1. 1530 2.7343 2.5722 0.06 25 0.7976 0.5651 2.8314 2.5185

0.60 0 2.5220 2.6322 0.60 1 2.7954 2.2003 2.4926 2.3971 0.60 5 2.8821 2.2945 2.6045 2.4986 0.60 10 2.8899 2.3860 2.6878 2.6585 0.60 20 2.3426 2.0506 2.4702 2.6536 0.60 25 2.6239 2.0947 2.5366 2.4753

1.20 0 2.4235 2.4442 1.20 1 2.2849 1.9769 2.4622 2.6052 1.20 5 3.0152 2.4864 2.5599 2.7393 1. 20 10 2.8985 2.5604 2.5456 2.8294 1.20 15 2.7230 2.4021 2.4580 2.6934 1.20 20 2.6909 2.3907 2.4675 2.7138 1.20 25 2.7349 2.4638 2.5384 2.8431

1.50c 25 2.8412 2.3317 2.4606 2.7245 CL OJ 0 0

Mean 2.5506 2.6123

SDe 0.1065 0.1248

C V f 4. 18 4. 78

® I, 4'-0Ac-2-oxo-MB; II, 4'-0Ac-MB; III, 2-oxo-MB. Concentration of the alkali solution added to the sample. c For runs with less than 25 pL of AC2 O, precipitation occurred upon dilution with mobile phase; the samples were not assayed. Precipitation occurred upon dilution with mobile phase for all runs; the samples were not assayed. e Standard deviation (N = 21). ^ Coefficient of variation, percent. Table 2.2 Urine sample preparation for the assay of merbarone and Its metabolites.

Urine: 100 yL

Internal standard: 3'-F-Merbarone (NSC 372106); 16.0 yg/mL In DMSO, 25 yL

Alkalinlzlng reagent: K 2C03 (1.2 M), 25 yL

Acetylating reagent: Acetic anhydride, 10 yL

Mixing: Vortex, 1 min

Temperature: Ambient

Diluent: Methanol (35%, v/v)- Ammonium acetate (67 mM)- Acetic acid (33 mM)- MgS04.7H20 (40 mM)- Sodium dodecyl sulfate (1 mM)

Centrifugation: 12,000 x g, 5 mln

Volume injected: 30 yL Table 2.3 Conditions for the HPLC analysis of merbarone and its metabolites in urine.

LC: Isocratic

Column: Nova-Pak Cjg, 4 Jim (3.9 mm x 15 cm)

Precolumn: None

Mobile phase: Methanol (25%, v/v)- Ammonium acetate (67 mM)- Acetic acid (33 mM)- MgS04 .7H20 (40 mM)- Sodium dodecyl sulfate (1 mM) pH 5.5

Flow rate: 1.0 mL/min

Detection: UV absorption 293 nm, 0.005 AUFS

Lowest conc. on std. curve: 0.25 ug/mL Table 2.4 fiackcalculated concentrations for 4'-0H-2-oxo-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 4 1-0H-2-oxo-MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period.

Concentration, ug/oL

Day 10.0220 8.0180 6.0250 4.0250 2.0120 1.0060 0.5030 0.2520 slope y-int corra

1 10.4890 7.8448 6.0482 4.1124 1.7966 1.2416 0.4457 0.2668 0.3742 0.0988 0.9943 2 9.6428 7.7291 6.3890 3.9727 2.1343 1.0342 0.4733 0.2571 0.3830 0.0916 0.9986 3 10.2855 7.9812 5.9723 4.1078 1.9904 0.9757 0.5031 0.2547 0.3623 0.0337 0.9998 4 9.6752 8.8138 5.9087 3.6597 2.0476 1.1260 0.4999 0.2459 0.4167 0.0250 0.9968 5 9.3682 8.3995 5.7401 3.9446 2.0843 1.1955 0.4931 0.2379 0.3679 0.0671 0.9965 0 9.5920 7.9159 6.5290 3.9929 2.0908 1.0051 0.4718 0.2601 0.3709 0.0074 0.9985 7 9.9407 7.4614 6.2040 4.0305 1.9914 1.0942 0.5260 0.2401 0.3823 0.0267 0.9984 8 9.9547 7.8961 6.3149 3.9277 1.9996 0.9902 0.5319 0.2452 0.3834 0.0346 0.9994 9 9.8831 7.8602 6.1152 4.2095 1.9951 1.0284 0.4687 0.2633 0.3645 0.0523 0.9994 10 9.7993 8.1520 6.0770 4.0919 2.0295 0.9677 0.5072 0.2534 0.3968 0.0499 0.9997

N 10 10 10 10 10 10 10 10 10 10 10

Mean 9.8631 8.0054 6. 1299 4.0050 2.0160 1.0659 0.4921 0.2525 0.3802 0.0487 0.9981

SDb 0.3319 0.3756 0.2361 0.1499 0.0916 0.0953 0.0271 0.0098 0.0166 0.0017

CVc 3.36 4.69 3.85 3. 74 4.54 8.94 5.51 3.89 4.36 0. 18

a Correlation coefficient. b Standard deviation. 0 Coefficient of variation, percent. Table 2.5 Backcalculated concentrations for 4'-OH-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 4 1-OH-MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period.

Concentration, wg/mL

Day 8.0032 6.0143 4.0175 2.0088 1.0044 0.5022 0.2511 slope y-int corra10.0040

1 9.9880 8.0559 6.2019 4.0856 1.8524 1.0857 0.4752 0.2588 0.3073 0.0472 0.9984 2 9.7489 8.0513 6.4844 4.0398 2.1655 0.9505 0.4642 0.2610 0.3288 -0.0223 0.9974 3 10.3525 8.4606 6.0728 4.0329 1.8394 0.9405 0.5563 0.2471 0.2918 0.0314 0.9975 4 10.1276 8.7058 6.2513 3.7597 1.8944 1.0902 0.4663 0.2585 0.3204 -0.0130 0.9969 5 10.0167 8.8603 5.9895 3.9358 1.9633 1.0213 0.4665 0.2592 0.3111 -0.0187 0.9980 6 9.5634 7.8257 6.4719 3.8564 2.0357 0.9868 0.5599 0.2425 0.3116 -0.0073 0.9977 7 10.1156 7.4927 6.2144 3.9267 1.8405 1.1681 0.5622 0.2374 0.2993 -0.0021 0.9951 8 10.0319 8.1100 6.3437 3.9186 1.9752 0.9448 0.5333 0.2488 0.3202 -0.0075 0.9988 9 10.0696 7.9871 6.0769 4.1332 1.9400 1.0437 0.4620 0.2653 0.3198 0.0335 0.9989 10 10.2768 8.4790 6.2832 4.0184 1.9653 0.9351 0.4809 0.2596 0.3475 -0.0188 0.9984

N 10 10 10 10 10 10 10 10 10 10 10

Mean 10.0291 8.2028 6.2390 3.9707 1.9472 1.0167 0.5027 0.2538 0.3158 0.0022 0.9977

SDb 0.2314 0.4182 0.1652 0.1125 0.1009 0.0794 0.0443 0.0092 0.0156 0.0011

CVc 2.31 5. 10 2.65 2.83 5. 18 7.81 8.80 3.62 4.94 0. 11

a Correlation coefficient. b Standard deviation. c Coefficient of variation, percent. Table 2.6 Backcalculated concentrations for 2-oxo-MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added 2-oxo-MB 1/peak area ratio squared as the weighting factor, during a 3.5 month period.

Concentration, yg/mL

Day 9.9922 7.9937 6.0072 4.0127 2.0064 1.0032 0.5016 0.2508 slope y-int corra

1 9.7717 7.6163 5.9278 4.0582 1.8734 1.2205 0.5893 0.2201 0.3264 0.0437 0.9939 2 9.7232 7.8923 6.4861 3.9393 2.1824 0.9765 0.4306 0.2900 0.3136 0.0576 0.9960 3 10.4072 8.1515 6.0030 4.1390 1.8299 0.9786 0.5245 0.2516 0.3129 0.0595 0.9986 4 9.7008 8.7178 5.9197 3.6396 2.0262 1.1027 0.5113 0.2441 0.3683 0.0095 0.9972 5 9.7539 8.6750 5.8725 3.8902 1.9788 1.0645 0.4802 0.2545 0.3159 0.0205 0.9986 6 9.7960 8.0001 6.6296 3.9546 2.1206 0.8948 0.4931 0.2646 0.3226 0.0341 0.9971 7 10.3252 7.6114 6.3272 4.0329 1.9787 1.1814 0.4176 0.2791 0.3235 0.0060 0.9921 8 10.3359 8.1383 6.4521 3.9929 1.9593 0.8998 0.5057 0.2602 0.3230 0.0154 0.9979 9 9.9035 8.0239 6.0376 4.0964 1.9233 1.0614 0.4831 0.2540 0.3195 0.0219 0.9993 10 9.9804 8.3072 6.1651 4.0127 1.9717 0.9530 0.4993 0.2554 0.3481 0.0061 0.9994

N 10 10 10 10 10 10 10 10 10 10 10

Mean 9.9698 3.1134 6.1821 3.9756 1.9844 1.0333 0.4935 0.2574 0.3274 0.0274 0.9970

SDb 0.2800 0.3785 0.2727 0. 1395 0.1056 0.1122 0.0479 0.0189 0.0175 0.0024

CVc 2.81 4.66 4.41 3.51 5. 32 10.86 9. 70 7.35 5.34 0. 24

a Correlation coefficient. b Standard deviation. ° Coefficient of variation, percent. Table 2.7 Backcalculated concentrations for MB urine standards. Data was obtained through linear regression analysis of the peak area ratio of the analyte to internal standard vs the concentration of added MB using 1/peak area ratio squared as the weighting factor, during a 3.5 month period.

Concentration, pg/mL

Day 10.0436 8.0349 6.0382 4.0335 2.0168 1.0084 0.5042 0.2521 slope y-int corra

1 9.6934 7.9407 6.0977 4.0872 1.8809 1.1026 0.5410 0.2436 0.3445 -0.0090 0.9979 2 9.7026 7.9795 6.5204 3.9691 2.1301 0.9984 0.4677 0.2619 0.3708 -0.0052 0.9982 3 10.1986 8.5408 6.1428 4.1256 1.8440 1.0266 0.4700 0.2668 0.3006 0.0177 0.9979 4 9.8947 8.5622 6.1863 3.7244 1.9637 1.0231 0.5358 0.2479 0.3386 -0.0236 0.9983 5 10.0101 8.8499 5.9517 3.9238 1.9567 1.0484 0.4703 0.2616 0.3243 0.0051 0.9981 6 9.8191 8.0185 6.6790 3.9799 2.0314 0.9789 0.4802 0.2599 0.3331 -0.0053 0.9985 7 10.5821 7.7008 6.4216 4.0505 1.9522 1.0616 0.4534 0.2660 0.3106 -0.0058 0.9971 8 10.1980 8.1909 6.3354 3.9335 1.9797 0.9631 0.5088 0.2540 0.3341 -0.0086 0.9994 9 9.9478 8.0938 6.1084 4.1528 1.9760 0.9821 0.5090 0.2527 0.3426 -0.0023 0.9998 10 10.3343 8.4930 6.2573 3.9640 1.9122 0.9514 0.5175 0.2530 0.3849 -0.0267 0.9937

N 10 10 10 10 10 10 10 10 10 10 10

Mean 10.0381 8.2370 6.2701 3.9911 1.9627 1.0136 0.4954 0.2567 0.3384 -0.0064 0.9984

SDb 0.2875 0.3576 0.2205 0.1235 0.0794 0.0478 0.0309 0.0077 0.0252 0.0008

CVc 2.86 4.34 3.51 3. 10 4.04 4.71 6. 24 3.01 7.45 0.08

A Correlation coefficient. b Standard deviation. ° Coefficient of variation, percent. 136

Table 2.8 Relative recovery and reproducibility of the analytical method for 4'-0H-2-oxo-MB in human urine.

Amount Number of Mean amount Percent added replicates found recovery CVa (yg/mL) N (yg/mL)

0.2520 10 0.2525 100.2 3.89 0.5030 10 0.4921 97.8 5.51 1.0060 10 1.0659 106.0 8.94 2.0120 10 2.0160 100. 2 4.54 4.0250 10 4.0050 99.5 3. 74 6.0250 10 6. 1299 101. 7 3.85 8.0180 10 8.0054 99.8 4.69 10.0220 10 9.8631 98.4 3.36

a Coefficient of variation, percent.

Table 2 .9 Relative recovery and reproducibility of the analytical method for 4' -OH-MB in human urine.

Amount Number of Mean amount Percent added replicates found recovery CVa (yg/mL) N (yg/mL)

0.2511 10 0.2538 101. 1 3.62 0.5022* 10 0.5027 100. 1 8.80 1.0044 10 1.0167 101.2 7.81 2.0088 10 1.9472 96.9 5. 18 4.0175 10 3.9707 98.8 2.83 6.0143 10 6.2390 103. 7 2.65 8.0032 10 8.2028 102.5 5. 10 10.0040 10 10.0291 100. 3 2.31

a Coefficient of variation, percent 137

Table 2.10 Relative recovery and reproducibility of the analytical method for 2-oxo-MB In human urine.

Amount Number of Mean amount Percent added replicates found recovery CVa (yg/mL) N (yg/mL)

0.2508 10 0.2574 102.6 7.35 0.5016 10 0.4935 98.4 9. 70 1.0032 10 1.0333 103.0 10.86 2.0064 10 1.9844 100.2 5.32 4.0127 10 3.9756 98.9 3.51 6.0072 10 6.1821 102.9 4.41 7.9937 10 8.1134 101.5 4.66 9.9922 10 9.9698 99.8 2.81

a Coefficient of variation, percent.

Table 2 .11 Relative recovery and reproducibility of the analytical method for MB in human urine.

Amount Number of Mean amount Percent added replicates found recovery CVa (yg/mL) N (yg/mL)

0.2521 10 0.2567 101.8 3.01 0.5042 10 0.4954 98.3 6.24 1.0084 10 1.0136 100.5 4.71 2.0168 10 1.9627 97.3 4.04 4.0335 10 3.9911 98.9 3.10 6.0382 10 6.2701 103.8 3.51 8.0349 10 8.2370 102.5 4.34 10.0436 10 10.0381 99.9 2.86

a Coefficient of variation, percent. 138

Table 2.12 Stability of the acetate derivatives of phenolic merbarone metabolites in the assay solution at ambient temperature.a

Concentrat ion, b yg/mL Time (h) I II III MB

0.5 7.9050 8.1102 7.8408 8.0667 1.0 7.5460 8.0372 7.6369 8.0337 2.0 7.0209 7.3574 6.8683 7.3254 4.0 7.3400 8.0633 7.4791 8.0208 8.0 7.5040 7.2847 7.4190 7.3547 24.0 7.3564 7.4372 7.5947 7.6126

Mean 7.4454 7.7150 7.4731 7.7357

SD° 0.2912 0.3928 0.3303 0.3488

CVd 3.91 5.09 4.42 4.51

Drug-free urine containing 4'-0H-2-oxo-MB (8.018 yg/mL), 4'-OH-MB (8.003 pg/mL), 2-oxo-MB (7.994 Pg/mL) and MB (8.035 pg/mL) were assayed by HPLC according to the derivatization procedure with injection at the indicated time intervals subsequent to sample preparation. I, 4'-0Ac-2-oxo-MB; II, 4'-0Ac-MB; III, 2-oxo-MB. Standard deviation (N = 6). Coefficient of variation, percent. 60-i

50-

40-

30-

20 -

10 - 8 . 86

0 H 0 2 4 6 8 1 0 12 1 4 T i me (mi n. )

Fig. 2.1. Chromatogram at 293 nm of 24 h pooled patient urine, collected on day 5 during the 24 h civ administration of MB, 1000 mg/m /d x 5. Chromatographic peaks: 4'-OH-MB, 1.7 min; 2-oxo-MB, 7.6 min; MB, 8.9 min. 1.4 min; 4 1-OH-MB, 1.8 min; 2-oxo-MB, 7.9 rain; and MB, 9.2 min. 9.2 rain; 7.9 MB, and 2-oxo-MB, min; 1.8 1-OH-MB, 4 min; 1.4 mRU Sc a 1e d Fig. 2.2. Chromatogram at 293 nm of a spiked aqueous sample containing synthetic 4'-0H-2-oxo-MB, synthetic containing sample aqueous spiked a of nm 293 at Chromatogram 2.2. Fig. 1 1 0 0 60- 80- 30 40 50- 90- 1

0 1 - 0 - 0 1.81 2 ie (min. ) Time 6 8 10 1

144 0 4 1 'O-B lOL aklnzd ih . M O (0 y) and yL) (100 KOH M 1.0 with alkallnized (lOOyL) 4'-OH-MB ouin. A 4-HM, . mn () '0cM, . mn (C) min; 5.3 4'-0Ac-MB, (B) min; 1.8 4'-0H-MB, (A) solutions. injected 5 min after the addition of AC of addition the after min 5 injected START i. .. iud hoftgfm o 10 gm sie aqueous spiked yg/mL 100 chromfltogrflms of Liquid 2.3. Fig. 10 > r>

CO CO

141 Q_ c: O c o I CO CO Fig. 2.4. Liquid chromatograms of aqueous 4'-OH-MB (100 yg/mL) treated with an equivalent volume of aqueous base (100 yL) and 25 yL of AC2O. (A) Sodium bicarbonate, 1.0 M (B) potassium carbonate, 1.0 M. Chromatographic peaks: 4'-OH-MB, 1.8 min; 4'-0Ac-MB, 4.8-5.1 min. 143

u CO

c ■ C I J" Q_ O' Q. O CO c o

Fig. 2.5. Chromatograms of patient urine collected during the infusion of MB. (A) Assayed directly; (B) injected after treating with an equivalent volume of 1.0 M I^COo (100 pL) and 25 pL of AcoO. The Indicated peaks are: (1) 4 ' -0Ac-2-oxo-MB; (2) 4'-0Ac-MB; (3) 2-oxo-MB; (4) MB. 144

iO-O B OSM

oo u : co

c • 0_ Q. o_ o O o CO CO CO co CO CO

Fig. 2.6. Chromatograms of derivatized patient urine samples determined under the same conditions with the exception of the methanol content of the mobile phase. (A) 25% (v,v) methanol; (B) 30% (v,v) methanol; (C) 35% (v,v) methanol. The indicated peaks are: (1) 4 ' -0Ac-2-oxo-MB; (2) 4'-0Ac-MB; (3) 2-oxo-MB; (4) MB. S . 0 5 1 00d

90: 9 . 03 80 70- 60- 50 40- 30-

10 -

0 H 0 24 6 8 10 12 14 T i me (min. )

Fig. 2.7. Chromatogram at 293 nm of an aqueous sample, spiked with MB and its major urinary metabolites, determined with diode-array detection after acetylation of the phenolic components with AC 2 O-K 2 CO2 . Chromatographic peaks: 4*-0Ac-2-oxo-MB, 4.5 min; 4 1 -OAc-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 min. Fit* I ♦ 13 » MBS a 03 3 . D Siu 1 va 1 i 13 T 1 1 * i 23 HR 104 0M

4 0 0

W « v l » n a t h CniwJ______T I mm C m I n 3

Fig. 2.8. Three-dimensional chromatogram of the spiked aqueous sample containing MB and its major urinary metabolites treated with AC 2 O-K 2 CO 2 . Chromatographic peaks: 4*-0Ac-2-oxo-MB, 4.5 min; 4'-0Ac-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 min. FI la • • 1 3 I M B B B B 3 S . D 8u>1v»1 3 3 H P 1 B 4 B M

H « v 1 angth Cr»m3 T t mm CmIn 3

Fig. 2.9. Plot of the three-dimensional chromatogram from 3.0-11.0 min with enhanced resolution of the spiked sample treated with AC 2 O-K2 CO 3 - Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.5 min; 4'-OAc-MB, 6.1 min; 2-oxo-MB, 7.7 min; MB, 9.0 min. Sui 1 vo 1 29 HP104QM

890

390

Cnm3

Fig. 2.10. Three-dimensional chromatogram of the spiked aqueous sample containing MB and its major urinary metabolites assayed directly. Chromatographic peaks: 4 1 -OH-2-oxo-MB, 1.4 min; 4'-OAc-MB, 1.8 min; 2-oxo-MB, 7.9 min; MB, 9.2 min. 149

A B

rs ro

roa\ iri

ir>

Fig. 2.11. Chromatograms of drug-free urine spiked with synthetic samples of (A) 4'-0Ac-2-oxo-MB (5.2 min) and 4'-OAc-MB (7.1 rain) assayed directly and (B) 4'-0H-2-oxo-MB and 4'-OH-MB chromatographed after derivatization. The peak at 15.4 rain is the internal standard. The concentration of each analyte was 6 yg/mL. 1001 90:

80-

7 0" 73 (D

ft) O m 50:

D 40: CE E 30:

20:

1 0:

0 4 220 240 260 280 300 320 340 3 G 0 380 400 Wavelength (nm)

Fig. 2.12. Overlaid UV spectra determined at the apex of the chromatographic peak of synthetic 4' -0Ac-2-oxo-MB and the peak at 4.5 min in the sample spiked with MB and metabolites after acetylation. OcM ad h pa a 61 i i te ape pkdwt M ad eaoie atr acetylation. after metabolites and MB with spiked sample the in min 6.1 at peak the and -OAc-MB ' 4 mflU Sc a 1ed i. .3 Oeli U seta eemnd t h ae o te hoaorpi pa o synthetic of peak chromatographic the of apex the at determined spectra UV Overlaid 2.13. Fig. 100 901 601 80- 70- 50 30 40 20 0-1 d 2 20 6 20 0 320 300 280 260 240 220 ae1nt Cnm) Nave 1ength 340 6 380 360 400 B ih C A with MB :m T £ C * a l i. .4 Te 5 Mz M setu o te ecin rdc fre uo tetn 4'-0H-2-oxo- treating upon formed product reaction the of spectrum NMR MHz 250 The 2.14. Fig. 2 O-K 20 10 00 9.0 10.0 11.0 12.0 2 l 1 O C

1 I 3 ...... I ...... I N) Fig. 2.15. The 250 MHz NMR spectrum of the reaction product formed upon treating 4 1 -0H-MB with AC2O-K2CO2. 153 1 0 0 1

90-

80"

701

601

50i

401

301

201

101

0 4 220 240 260 280 300 320 360 380 400340 Navelength (nm)

Fig. 2.16. Overlaid UV spectra determined at the apex of the chromatographic peak of reference 2-oxo-MB and the peak at 7.7 min in the sample spiked with MB and metabolites after acetylation. 1 00d

90- 80- 70: -0 (D 60: id cno 50- □a: 40: E 30

0 4 220 240 260 280 300 320 340 360 380 400 Wavelength (nm)

Fig. 2.17. Overlaid UV spectra determined at the apex of the chromatographic peak of reference MB and the peak at 9.0 min in the sample spiked with MB and metabolites after acetylation. 3-45 2 5

9 0 0

9 8 0

10 12

Cm I n3

Fig. 2.18. Three-dimensional chromatogram truncated at 300 mAU of day 5 pooled urine, collected during the continuous infusion of MB at 1000 mg/m /d x 5, treated with A C 2O-K 2CO 2 . Chromatographic peaks: 4 1-0Ac-2-oxo-MB, 4.4 min; 4'-0Ac-MB, 5.9 min; 2-oxo-MB, 7.6 min; MB, 8.8 min. rlit I ♦13•MBB80?a.D S u j 1 ve I i 335 T 1 I t i 3 0 HP 1040M

Fig. 2.19. Plot of the three-dimensional chromatogram from 3.0-11.0 min of the day 5 urine sample treated with AC 2O-K 2CO 3 at greater sensitivity. Chromatographic peaks: 4* -0Ac-2-oxo-MB, 4.4 min; 4'-0Ac-MB, 5.9 min; 2-oxo-MB, 7.6 min; MB, 8.8 min. 345 25 H FM0 4 0 M

350

1 0

T I Cmtn] H » v 1 tngth Cnm3

Fig. 2 .20. Three-dimensional chromatogram truncated at 300 mAU of pre-dose urine treated with AC 2O-K 2CO.J. A B C

U.I L . CO CO

Fig. 2.21. Liquid chromatograms of standard curve samples prepared by spiking drug-free urine with the four analytes and assayed according to the derivatization procedure. The concentrations of the analytes were essentially equivalent in the standards. (A) Analyte and internal standard- free urine; (B) analytes, 0.25 ug/mL (C) analytes, 1.0 yg/mL. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.8 min; 4'-0Ac-MB, 6.5 min; 2-oxo-MB, 8.1 rain; MB, 9.6 min; internal standard, 14.6 min. B K K S S 5 C * r vrl ooo*i

Fig. 2.22. Liquid chromatograms of standard curve samples prepared by spiking drug-free urine with the four analytes and assayed according to the derivatization procedure. The concentrations of the analytes were essentially equivalent in the standards. (A) analytes, 2.0 yg/mL; (B) analytes, 6.0 pg/mL; (C) analytes, 10.0 yg/mL. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.8 min; 4'-0Ac-MB, 6.5 min; 2-oxo-MB, 8.1 min; MB, 9.6 min; internal standard, 14.6 min.

Ov O 161

eg no voro vooo

co O** PO ir> m

ro a> ir> l . CO

Fig. 2.23. Chromatograms of patient urine obtained on day 5 during the 2 h civ of MB, 190 mg/m^/d x 5. (A) Sample chromatographed directly; (B) sample chromatographed following derivatization. Chromatographic peaks: 4'-0Ac-2-oxo-MB, 4.9 min; 4'-0Ac-MB, 6 . 6 min; 2-oxo-MB, 8.4 min; MB, 10.0 min; internal standard (3'-F-MB), 15.2 min. .65 yitret 002; , 0.9992. r, 0.0523; y-intercept, 0.3645; AREA RATIO (4‘-OH-2-oxo-MB/IS) 3.5 2.5 3.0 4.0 2.0 0.5 0.0 i. .4 Clbain uv fr 0--x-B n ua uie slope, urine: human in -0H-2-oxo-MB ' 4 for curve Calibration 2.24. Fig. 0 2 3 C O N C E N T R A T I O N(u g /m L ) 4 5 6 9 7 8 10 necp, .35 r 0.9989. r, 0.0335; intercept, AREA RATIO (4'-OH-MB/IS) 3.0 3.5 2.5 4.0 2.0 0.5 0.0 i. .5 Clbain uv fr '0-B n ua uie soe 039; y- 0.3198; slope, urine: human in 4'-0H-MB for curve Calibration 2.25. Fig. 0 2 3 C O N C E N T R A T I O N(u g /m L ) 4 5 6 7 8 9 10 necp, .29 r 0.9993. r, 0.0219; intercept, AREA RATIO (2-oxo-MB/IS) 3.0 2.0 2.5 3.5 4.0 0.5 0.0 i. .6 Clbain uv fr -x-B n ua uie soe 039; y- 0.3195; slope, urine: human in 2-oxo-MB for curve Calibration 2.26. Fig. 0 2 3 C O N C E N T R A T I O N(u g /m L ) 4 5 6 7 8 9 10 11 necp, 002; , 0.9998. r, -0.0023; intercept, AREA RATIO (MB/IS) 3.0 3.5 4.0 2.0 2.5 0.5 0.0 i. .7 Clbain uv fr B n ua uie soe 032; y- 0.3426; slope, urine: human in MB for curve Calibration 2.27. Fig. C O N C E N T R A T I O N(u g /m L )

11 165 CHAPTER III

EXCRETION OF MERBARONE AND ITS HAJOR METABOLITES IN HUMAN URINE

A. Introduction

The analytical method developed for the simultaneous assay of

merbarone and the three principal urinary metabolites, 4'-0H-2-oxo-MB,

4 f-0H-MB and 2-oxo-MB, was employed to quantitate the urinary excretion

of these compounds for patients in the phase I evaluation of MB

conducted at The Ohio State University. Two dosing schedules were

adopted for these clinical trials, a 5 day 24 h continuous infusion via

a central i.v. catheter and a daily 2 h peripheral venous infusion for 5

days (1). At present, 20 patients have been treated with 96-1250

o 2 mg/m /d of MB on the 24 h civ regimen and 150-447 mg/m /d has been

administered to 16 patients by the 2 h civ schedule. Accordingly,

information regarding drug and metabolite excretion as a function of dose in a reasonably sized patient population could be accrued.

The primary objective of this study was to quantitatively assess the significance of the renal excretion of MB and its three characterized metabolites in the overall elimination of the drug from man. In a preclinical investigation of 2-^C-merbarone disposition in the dog, the cumulative urinary radioactivity for 30 h following administration by i.v. bolus accounted for only 23.7% of the dose (2). Biliary excretion of radiolabeled MB was shown to be significant in the mouse after dosing

166 167

by I.v. injection (3). Similar studios to characterize the disposition

of MB in man following a radiolabeled dose of the drug have not been

conducted.

Information acquired from the study presently being considered may

be of interest in some additional respects. For example, systematic

investigations pertaining to dose dependencies on the metabolism of

thiobarbiturates by desulfurization and aromatic hydroxvlation of

phenobarbital have apparently not been undertaken. Both of these

oxidative pathways are operative in the biotrnnsformation of MB.

The aromatic hydroxylation of phenobarbital has been shown to be a detoxification mechanism (4). The advent of toxicities in hepatically

impaired patients, at dose levels tolerated in normal subjects, was associated with an accumulation of the drug in plasma and decreased

formation of p-hydroxyphenobarbital. However, in the MB phase I trials, dose dependency for total body clearance or any of the other pharmacokinetic parameters was not evident over the dose range studied

(1).

The primary toxicity in patients treated with MB on the 24 h civ schedule has been renal toxicity, as demonstrated by an elevation in serum creatinine (1). In preclinical studies with the rat, among the organ systems affected, dose-related toxicities were detected in the renal, hepatic and cardiovascular systems (5). It is not inconceivable that some of the toxic responses may be associated with the metabolism of merbarone. Formation of the metabolite 4'-0H-MB is thought to be a detoxification mechanism, since the compound is apparently not cytotoxic

(5). However, the reactivity of the p-hydroxycarboxanilide moiety 168 present In this compound suggests that further transformation to toxic metabolites is possible, in a manner analogous to that of acetaminophen.

The acute toxicities of acetaminophen in the rat include capillary- venous congestion in most organs and extensive hepatic necrosis associated with degenerative changes in the kidneys, such as tubular nephritis (6 ). Large overdoses of acetaminophen can produce fatal hepatic necrosis in man (7). These affects result from the formation of a toxic intermediate, presumably N-acetyl-p-benzoquinoneimine, in a quantitatively minor metabolic route (8-14). At low doses of acetaminophen, glutathione protects the liver from damage by reacting with the quinoneimine, forming a conjugate which is excreted into the urine (15-17). However, at higher doses glutathione becomes depleted and the reactive intermediate covalently binds with macromolecules, resulting in the observed toxicities.

Therefore, quantitative information pertaining to MB biotransformation may prove to be beneficial toward understanding the observed and potential toxicities of the drug. Elucidation of the relative importance of the major oxidative metabolic pathways, as evaluated by urinary metabolite excretion, is of interest due to the formation of an active metabolite (i.e., 2-oxo-MB) as opposed to inactive and potentially toxic phenolic metabolites. Furthermore, the amount of the dose unaccounted for by excretion of MB and these metabolites will be indicative of the significance of alternate routes of elimination, which may include additional metabolic pathways. 169

B . Materials and Methods

1. Acqulstion of Human Urinejnd_ P1 asma Specimens

Urine and plasma samples were obtained from patients

participating In the phase I trials of merbarone. The protocol for the

administration of MB, acquisition of blood and collection of urine has

been reported (1). Each patient gave informed consent before receiving

the drug.

Prior to dosing, 50 mL of urine was obtained. Following MB

administration, 10 mL of urine was removed from each void and the

remainder was pooled to obtain a 24 h urine collection. A 50 mL aliquot

of the pooled urine was removed. Urine was collected for 72 h after the

initiation of dosing. The time, pH and volume of each void were

recorded. All urine samples were stored at -20°C until assayed.

2. Analytical Method

The concentration of MB and the metabolites 4'-0H-2-oxo-MB,

4'-0H-MB and 2-oxo-MB in the urine of patients was determined according

to the HPLC method described In Chapter II. The aliquots obtained from

each 24 h pooled urine collection were assayed in duplicate. If the

individual concentrations deviated from the mean of the duplicate assays by more than 10 %, the sample was re-assayed.

3. Plasma Sample Preparation

Plasma samples were assayed with prior precipitation of protein by adding 250 yL of methanol-DMSO (85:15, v/v) to 50 yL of plasma in a

1.5 mL polypropylene microcentrifuge tube. The tube was mixed by vortexing for 0.5 min then centrifuged at 12,000 x g for 10 min.

Aqueous potassium carbonate (200 yL, 0.42 M) was added to 200 yL of the 170

supernatant In a microcentrifuge tube. The mixture was treated with

acetic anhydride (35 yL) and immediately mixed by vortexing for 60 s.

The sample was allowed to stand for 5 min, centrifuged at 12,000 x g for

5 min, and 100 yL injected onto the column. methanol, 100 yL was

Injected onto the column. Diode-array detection was employed as

previously described.

C. Results and Discussion

The urinary excretion time courses of MB and the three oxidative

metabolites were determined for 7 days following initiation of the

infusion. The analytical results for 11 patients on the 24 h civ daily

x 5 schedule (Table 3.1) and for 13 patients on the 2 h civ daily x 5

schedule (Table 3.2) give the total cumulative amount, relative to the

dose on a molar basis, of each compound excreted into the urine.

Preliminary studies suggest that glucuronide and sulfate conjugates of

the drug and its characterized metabolites are not present in the urine

or plasma of patients. Therefore, the data in Tables 3.1 and 3.2 are

considered to be indicative of the overall extent to which these

compounds are excreted.

For patients on the 24 h schedule, the mean cumulative urinary excretion of the parent drug during the 7 days was 0.9 ± 0.52% of the dose, and for patients on the 2 hour schedule it was 1.91 ± 1.13%.

Thus, very little of the administered dose is excreted in the urine as the unchanged drug. Accordingly, renal clearance of the unchanged drug

is a very minor component of the total body clearance. 171

The total percentage of the dose excreted Into the urine during 7

days as MB and the three oxidative metabolites was very similar for the

patients on the two schedules. The mean percentage of the dose excreted

In the urine as parent drug plus the three metabolites was 25.20 ± 7.42%

for patients on the 24 h dosing schedule and 27.55 ± 6.44% for the 2 h

schedule. Therefore, metabolites represent the major fraction of the

compounds which are excreted In the urine.

The two principal urinary metabolites are 2-oxo-MB and 4’-0H-MB.

For the patients on the 24 h schedule, the mean percent of the dose

excreted as 4'-0H-MB (14.29 ± 6.57%) exceeded that of 2-oxo-MB (7.35 ±

3.46%). However, it should be noted that in one of the patients, the

cumulative excretion of the 2 -oxo metabolite was more than twice that of

4'-0H-MB and in several patients, the amounts of the two metabolites

excreted were similar. In comparison, for patients on the 2 h regimen,

the mean percent of the dose excreted as 4'-0H-MB (11.83 ± 3.81%) was

similar to that of 2-oxo-MB (11.11 ± 5.52%). At present, it is unclear whether there is a difference related to dose or schedule in the

cumulative urinary excretion of these two metabolites, since the number

of subjects is limited and there is large inter-subject variability.

On both schedules, the mean cumulative excretion of the third metabolite, 4'-0H-2-oxo-MB, was less than that of the others; 2.66 ±

1.80% for the patients on the 24 h schedule and 2.70 ± 1.83% for the patients on the 2 h schedule. With the exception of one patient on each schedule where the cumulative percent excreted was 7.22 and 7.44%, the cumulative urinary excretion of 4'-0H-2-oxo-MB for most subjects was small. 172

Representative plots of the daily urinary excretion rate for MB and each of the three metabolites as function of the urine collection interval midpoint are shown at several dose levels for patients on the

24 h civ (Figs 3. 1-3.5) and 2 h civ (Figs. 3.6-3.8) schedules. In general, the excretion rate of MB was typically more or less uniform from day 2 until the end of the infusion. The excretion rates of the two principal urinary metabolites, 2-oxo-MB and A'-OH-MB, also appeared to be relatively uniform during this period. However, a peak in the excretion rate profile for 4 1 -0H-2-oxo-MB was generally not observed until somewhat later during the infusion.

The oxidative metabolites, which are so prominent in the chromatograms of patient urine, are either not detectable or present in exceedingly low concentrations in patient plasma, as demonstrated by the three-dimensional chromatograms shown in Figs. 3.9-3.11. On the other hand, the levels of the parent drug are high in plasma but not in urine.

Evidently, the renal clearance of the oxidative metabolites is greater than that of the drug.

In preclinical pharmacology studies of MB in Beagle dogs and CD2F1 mice, high levels of metabolites were present in the chromatograms of dog plasma which were not seen in chromatograms of mouse plasma (2 ).

Liquid chromatograms of both dog and mouse urine showed the presence of the parent drug and 4'-0H-MB as a major urinary metabolite. Very low levels of 2-oxo-MB were also detected in the chromatograms of dog urine.

Furthermore, treatment of dog plasma with 3-glucuronidase indicated the presence of a glucuronide conjugate of MB. 173

Whole-body autoradiography studies in CD2F1 mice after I.v.

injection of radiolabeled MB showed that the radiolabel is rapidly taken

up by the liver and kidney (3). Biliary excretion was seen as early as

3 min after dosing and significant excretion of radioactivity into the

bile had occurred by 30 min.

In dogs, only a small fraction of the dose appeared to be excreted

in the urine as the unchanged drug. The extensive metabolism of MB was

illustrated using the radiolabeled compound, 2 - -merbarone (2). After

an i.v. bolus dose (29.4 mg/kg, 187 v

parent drug measured by HPLC and the levels of total radioactivity were monitored to 30 h after dosing. Metabolites were present in the plasma at 10 min and peak levels were seen at 4 h after dosing. The levels of metabolites were 4-fold higher than that of the parent drug 4 h subsequent to dosing and 40-fold higher after 30 h. Thus, while the plasma levels of MB declined approximately 250-fold, higher levels of metabolites were sustained. The radioactivity in the urine collected during the 30 h contained 23.7% of the dose (2).

The present studies have shown that the urinary excretion of merbarone and the three oxidative metabolites accounts for approximately one-fourth of the dose administered to humans. Similar to man, in the dog plasma levels of these metabolites were also low compared to the parent drug. Thus, the observed high levels of radiolabeled compounds in dog plasma must correlate mainly to substances which were not identified. These unknown metabolites in the dog might also be present in patients and thereby in part account for the remaining mass balance of the administered dose. On the other hand, the unidentified compounds could be unique to the dog and the remaining fraction of the dose in man might be due to biliary-fecal elimination. Further work is needed to characterize the unidentified metabolites in dog plasma and then to confirm their presence or absence in man. 175

D. References

1. Malspeis, L. Study of the clinical pharmacology of antitumor drugs. Volume I: merbarone (NSC 336628). Volume II: amonafide (NSC 308847). Report to the National Cancer Institute, March, 1988.

2. Malspeis, L. , Lyon, M. E. , DeSouza, J. J. V., and Staubus, A. E. Preclinical pharmacology studies of merbarone (NSC 336628). (Abstract) Proc. Am. Assoc. Cancer Res., 27: 402, 1986.

3. Kemmenoe, B. H., and Malspeis, L. Distribution of 2-^C-merbarone in mice by autoradiography of whole-body cryosections. Cancer Res., 47: 1135-1142, 1987.

4. Kutt, H. Winters, W. Schermnn, R., and McDowell, F. Diphenylhydan- toin and phenobarbital toxicity - The role of liver disease. Arch Neurol., 11: 649-656, 1964.

5. Glover, A., Chun, H. G. , Kleinman, L. M. , Cooney, D. A., Plowman, J., Grieshaber, C. K., Malspeis, L., and Leyland-Jones, B. Merbarone: an antitumor agent entering clinical trials. Investigational New Drugs, 137-143, 1987.

6 . Boyd, E. M. , and Bereczky, G. M. Liver necrosis from paracetamol. Brit. J. Pharmacol., 26: 606-614, 1966.

7. Mitchell, J. R., Jollow, D. J., Gillette, J. R., and Brodie, B. B. Drug metabolism as a cause of drug toxicity. Drug Metab. Disp., 1: 418-423, 1973.

8 . Potter, W. Z., Thorgeirsson, S. S., Jollow, D. J., and Mitchell, J. R. Acetaminophen-induced hepatic necrosis. V. Correlation of hepatic necrosis, covalent binding and glutathione depletion in hamsters. Pharmacology, 12: 129-143, 1974.

9. Mitchell, J. R., Jollow, D. J. Potter, W. Z. Davis, D. C., Gillette, J. R. , and Brodie, B. B. Acetaminophen-induced hepatic necrosis. I. Role of drug metabolism. J. Pharmacol. Exp. Therap., 187: 185-194, 1973.

10. Hinson, J. A., Mitchell, J. R. , and Jollow, D. J. Microsomal N- hydroxylation of p-chloroacetanilide. Mol. Pharmacol., 11: 462-469, 1975. 176

11. Hinson, J. A., Nelson, S. D., and Mitchell, J. R. Studies on the microsomal formation of arylating metabolites of acetaminophen and phenacetin. Mol. Pharmacol., 13: 625-633, 1977.

12. Miner, D. J. and Kissinger, P. T. Evidence for the involvement of N-acetyl-p-quinoneimine in acetaminophen metabolism. Biochem. Pharmacol., 28: 3285-3290, 1979.

13. Hinson. J. A., Pohl, L. R. , and Gillette, J. R. N-hydroxyacet- aminophen: a microsomal metabolite of N-hydroxyphenacetin but apparently not of acetaminophen. Life Sciences, 24: 2133-2138, 1979.

14. deVries, J. Hepatotoxic metabolic activation of paracetamol and its derivatives phenacetin and benorilate: oxygenation or electron transfer? Biochem. Pharmacol., 30: 399-402, 1981.

15. Jollow, D. J. , Thorgeirsson, S. S., Potter, W. Z. , Hashimoto, M. , and Mitchell, J. R. Acetaminophen-induced hepatic necrosis. VI. Metabolic disposition of toxic and nontoxic doses of acetaminophen. Pharmacology, 12: 251-271, 1974.

16. Andrews, R. S., Bond, C. C., Burnett, J., Saunders, A., and Watson, K. Isolation and identification of paracetamol metabolites. J. Int. Med. Res., 4: 34-39, 1976.

17. Moldeus, P., Jones, D.P., Ormstad, K., and Orrenius, S. Formation and metabolism of a glutathione-S-conjugate in isolated rat liver and kidney cells. Biochem. Biophys. Res. Commun., 83: 195-200, 1978. 177

Table 3.1 Cumulative urinary excretion of merbarone and Its metabolites during 7 days by patients on the 24 h continuous I.v. dally x 5 schedule.

% Dose excreted® Dose ------Patient (mg/m /d) I II III MB Total

J.B. 144 1 ,. 12 7 . 27 6 .25 1 ,.47 16 . 11 L.F. 144 7,.22 11 . 8 8 9 .41 0 ,.27 28 .78 J.S. 192 1 ,.07 5 .42 11 . 10 0 ..24 17 .83 J.P. 317 2 ,.28 24 .37 7 .81 1 ..05 35 .51 N.Y. 317 1 ..69 13 .36 1 1 ,.43 1 ..87 28 .35 C.E. 560 3.. 36 25 .80 6 . 16 1 .,07 36..40 F.W. 560 3., 70 18..81 4.,41 0 .6 8 27..60 L.E. 1 0 0 0 2 .44 16..28 1 .,30 0 .22 2 0 ..24 W.K. 1 0 0 0 1 .35 1 2 .,66 4. 98 0 .89 19. 8 8 C.T. 1 0 0 0 3. 54 1 2 ., 78 1 2 .56 1 .18 30. 06 J.W. 1250 1 .50 8 .55 5. 45 0 .94 16. 44

Mean 2 .6 6 14. 29 7. 35 0 .90 25. 2 0

SDb 1 .80 6 .57 3. 46 0 .52 7. 42

CVc 67. 47 45. 94 47. 01 00 29 29. 44

® I, 4'-OH-2-oxo-MB; II, 4'-0H-MB; III, 2-oxo-MB. Standard deviation (N = 11). c Coefficient of variation, percent. 178

Table 3.2 Cumulative urinary excretion of merharone and its metabolites during 7 days by patients on the 2 h continuous i.v. daily x 5 schedule.

7o Dose excreted® DOSe ------

Patient (mg/m2 /d) I II III MB Total

H.F. 150 2 .56 10 .61 8 .59 1.71 23 .47 J.P. 150 2 .83 12 .82 2 0 .96 2.45 39 .06 V.P. 150 0 .91 7 .99 8 .98 2.57 20 .45 P.B. 190 2 ,.01 15,.34 1 2 ..30 2 . 0 2 31..67 P.K. 190 1 ,.65 17,.24 6 . 32 2.95 28.. 16 J. A. 253 2 ,.96 6 ,.11 15 .49 2.71 27..27 S.C. 253 2 ..49 14.,08 2 1 ..51 1.41 39..49 E.H. 253 1 .,45 6 .,33 14..56 1.82 24., 16 M. J. 336 3.,55 17., 73 8 ..84 0. 78 30..90 J.S. 336 7..44 1 2 .03 7.,91 0.43 27. 81 P.T. 336 4. 78 9. 0 1 3. 93 0.49 18. 21 W.N. 336 2 .40 13. 71 8 .44 1. 17 25. 72 G.C. 447 0 .60 1 0 .79 6 .64 4.44 2 2 .47

Mean 2 .70 1 1 .83 1 1 .11 1.91 27. 55

SDb 1 .83 3. 81 5. 52 1. 13 6 .44

CVc 67. 87 32. 21 49. 67 59. 10 23. 38

® I, 4'-0H-2-oxo-MB; II, 4'-0H-MB; III, 2-oxo-MB. Standard deviation (N = 13). c Coefficient of variation, percent. k_ -C cn E

Id cr< z o b ££ O X Ld y— cr < z

-2

0 48 72 9624 120 144 168 192 Tmid (hr)

Fig. 3.1. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 192 mg/m^/d x 5 to patient J. S. by 24 h civ. Key: (•) 4'-0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2-oxo-MB; (o) MB. CD

______I______I______I______I______I______I______0 24 48 72 96 120 144 168 Tmid (hr)

Fig. 3.2. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 317 mg/m^/d x 5 to patient J. P. by 24 h civ. Key: (•) 4 ' -0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2-oxo-MB; (o) MB. Ld

0 24 48 96 120 144 168 19272 Tmid (hr)

Fig. 3.3. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 560 mg/m^/d x 5 to patient C. E. by 24 h civ. Key: (•) 4 ' -0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2 -oxo-MB; (o) MB. 181 CT 1 0 ’

I—

X Ld

<10

-2 0 24 48 72 96 120 144 168 192 Tmid (hr)

Fig. 3.4. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 1000 mg/m /d x 5 to patient W. K. by 24 h civ. Key: (•) 4 1 -0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2-oxo-MB; (o) MB. Ld

-3 0 24 48 72 96 120 144 168 192 Tmid (hr)

Fig. 3.5. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 1250 mg/m /d x 5 to patient J. W. by 24 h civ. Key: (•) 4'-0H-2-oxo-MB; (+) 4 ' -OH- MB; (*) 2-oxo-MB; (o) MB. 10’ t i------1------r

-C oi 10° E

Ld cr

1 10” i— Ld cc o x Ld >-

W ~ 3 J______1______I______I______L 0 24 48 72 96 120 144 168 Tmid (hr)

Fig. 3.6. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 150 mg/m /d x 5 to patient V. P. by 2 h civ. Key: (•) 4 1 -0H-2-oxo-MB; (+) 4'-0H~ MB; (*) 2-oxo-MB; (o) MB. 1 0 ’

CP £

1 0 ° Ld (— < cr

Ld a: o x Ld 10-1 >- a: < z q : 3

10-2 0 24 48 72 96 120 144 168 Tmid (hr)

Fig. 3.7. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 253 mg/m^/d x 5 to patient E. H. by 2 h civ. Key: (•) 4 1 -0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2 -oxo-MB; (o) MB. 10'

cn E

1 0 ° < Od

Cd o X Ld 10-1 Od>- < z Cd Z>

1 0 ' 0 24 48 72 96 1 2 0 144 168 Tmid (hr)

Fig. 3.8. Plot of the daily excretion rate versus the urinary collection interval midpoint for MB and its metabolites during and after the administration of 336 mg/m^/d x 5 to patient W. N. by 2 h civ. Key: (•) 4 1 -0H-2-oxo-MB; (+) 4'-0H- MB; (*) 2-oxo-MB; (o) MB. Fig. 3.9. Three-dimensional chromatogram from 0-15 min, truncated at 40 mAU, of patient plasma obtained on day 5 during the 24 h civ infusion of MB, 1000 mg/m /d x 5, assayed according to the acetylation procedure. Chromatographic peaks: 4'-0Ac- MB, 5.9 min; 2-oxo-MB, 7.5 min; MB, 8.8 min. as as

tsa

Mivttngth CnmJ

Fig. 3.10. Plot of the three-dimensional chromatogram from 3-11 min with enhanced resolution of the patient plasma sample obtained on day 5 during the infusion of MB, 1000 mg/m /d x 5, assayed according to the acetylation procedure. Chromatographic peaks: 4'-0Ac-MB, 5.9 min; 2-oxo-MB, 7.5 min; MB, 8 . 8 min. Fig. 3.11. Three-dimensional chromatogram from 0-15 min, truncated at 40 mAU, of pre-dose patient plasma assayed after treatment with AcjO-^CO^ showing the absence of constituents which would interfere with analytes eluting from 3-11 min. CHAPTER IV

XANTHINE OXTDASE INHIBITION STUDIES

A. Introduction

Hypouricemia has been observed in patients treated with merbarone during the course of phase. I trials at both The Ohio State University

and Memorial Sloan-Kettering Cancer Center. This could result from xanthine oxidase inhibition by the. drug and potentially its metabolites, or uricosuric effects induced by the excretion of these compounds. The present study was undertaken to compare merbarone and the three major urinary metabolites with allopurinol as xanthjne oxidase inhibitors.

Xanthine oxidase is a meta1loflavoprotein which catalyzes the two electron oxidation of the purines hypoxanthine (HX) and xanthine (X) to uric acid (UA) with the stoichiometric formation of hydrogen peroxide

(1) (eqs 4.1 and 4.2).

HX + H 20 + 0 2 ► X + H 2 0 2 (4.1)

X + H 20 + 0 2 ► UA + H 2 0 2 (4.2)

The mechanism of action of the enzyme is quite complex and has been elucidated following several decades of investigation (1-8). The oxidation of hypoxanthine and xanthine by xanthine oxidase is inhibited

190 191

by structural analogs of these purines In addition to a number of other

compounds (1,9). Allopurinol (4-hydroxypyrazo1o[3,4,-d]pyrimidine) is a

particularly potent inhibitor of the enzyme both iji vitro and in vivo,

for which the mechanism of inhibition has been determined (10-14).

Since it is currently the drug of choice for the treatment of

hyperuricemia, allopurinol is the comparative standard for xanthine

oxidase inhibitors. Therefore, similar reaction conditions and

analytical methods were employed for examining the affect of merbarone

(MB) and its three primary metabolites, 4 1 -hydroxy-2-oxo-

desthiomerbarone (4 1 -0H-2-oxo-MB), 4'-hydroxymerbarone (4'-0H-MB) and

2-oxo-desthiomerbarone (2-oxo-MB), on xanthine oxidase activity as those

that have been reported in allopurinol inhibition studies.

B. Materials and Methods

1• Reagents and Chemicals

The xanthine oxidase was grade TII from buttermilk (Sigma

Chemical Co., St. Louis, MO). Xanthine and uric acid were Sigma grade;

allopurinol was obtained from the Aldrich Chemical Co. (Milwaukee, WI).

Merbarone, 2-oxo-desthiomerbarone, 41-hyrdroxymerbarone and 4'-hydrox- y-2-oxo-desthiomerbarone were prepared as described in Chapter I.

Monobasic potassium phosphate was a Baker analyzed reagent (J. T.

Baker Chemical Co., Phillipsburg, NJ); dibasic potassium phosphate trihydrate was an analytical reagent (Mallinckrodt, Inc., Paris, KY).

A. C. S. reagent disodium dihydrogen ethylenediaminetetraacetate dihydrate (EDTA) was obtained from G. Frederick Smith Chemical Co.

(Columbus, OH) and certified A. C. S. potassium hydroxide pellets were 192

purchased from Fisher Scientific (Fair Lawn, NJ) . The other reagents

were previously described. All chemicals were used without additional

purification.

2. Apparatus

UV absorption was monitored with a Beckman UV 5260

Spectrophotometer (Beckman Instruments, Irvine, CA) using matched 10 mm

rectangular quartz cuvettes. A Beckman Model 4500 Digital pH Meter

equipped with a Ross Combination pH Electrode (Orion Research,

Cambridge, MA) was used for pH measurements. Milligram quantities of

material were weighed with a Cahn 25 Automatic Electrobalance (Cahn

Instruments, Cerritus, CA). Class A volumetric flasks were used for the

preparation of stock solutions.

3. Stock Solutions

Stock solutions of allopurinol, MB and the metabolites (5 mL)

were prepared by dissolving the appropriate quantities of material in

0.1 M K0H (1.0 mL) and diluting with water to afford concentrations of

approximately 3 mM. The xanthine stock solution (10 mM) was prepared by

dissolving 7.6 mg in 0.1 M K0H (1.0 mL) and diluting to 5 mL with water.

Uric acid (8.4 mg) was dissolved in 0.1 M KOH (1.5 mL) and diluted to 5

mL with water to give a 10 mM stock solution.

The xanthine oxidase stock solution (0.33 units/mL) was prepared

daily by pipeting 20 yL of the reagent into 3.0 mL phosphate buffer,

0.05 M (y 0.14), pH 7.8 containing 0.01 mM EDTA. The buffer solution was made from 0.02 M KH2 P04 (25 m L ) , 0.09 M K 2 HP04 (50 mL) and 1 mM EDTA

(1.0 mL) and diluting to 100 mL. Buffer without added enzyme was used 193

for the enzyme blank.

4. Stand ard SolutIons

Standard solutions of allopurinol and the potential Inhibitors were prepared from dilutions of the stock solutions (tube //l).

Solutions were pipetted Into siliconized glass screw cap test tubes to afford the range of concentrations as tabulated below. A series of

Inhibitor blank solutions was similarly prepared by substituting 0 . 0 2 M

KOH for the indicated volumes of inhibitor standards.

Tube Allopurinol KOH, 10 mM no. standard, pi. (Pi.) Cone. (M)

1 -- 3.00 x 10' 3 2 1 0 0 (tube //l) 900 3.00 x 10"4 3 50 (tube HI) 950 1.50 x 10-4 4 10 0 (tube ii2) 900 3.00 x 10" 5 5 100 (tube in) 900 1.50 x 10-5 6 100 (tube #4) 900 3.00 x 10‘6 7 100 (tube ii6 ) 900 3.00 x 10" 7 8 - 500 0 . 0

Tube Inhibitor KOH, 10 mM __ Concentration.8 mM no. standard, pL (pr.) I ~ II III MB

1 -_ 3.046 2.971 3.000 2.999 2 750 (tube HI) 250 2. 285 2 . 228 2.250 2.249 3 500 (tube HI) 500 1.523 1.486 1.500 1.500 4 300 (tube HI) 700 0.914 0.891 0.900 0.900 5 1 0 0 (tube HI) 900 0. 305 0. 297 0.300 0. 300 6 1 0 0 (tube ii5) 900 0.0305 0.0297 0.0300 0.0300 7 - 500 0 . 0 0 . 0 0 . 0 0 . 0

a I, 4'-OH-2-oxo-MB; II, 4 ’-0H-MB; III, 2-oxo-MB. 194

The xanthine standard was prepared such that addition of the enzyme

stock and inhibitor standard solutions ( 2 0 0 yL) to 2.8 mL afforded a pH

7.8 solution containing phosphate buffer (0.05 M), EDTA (0.01 mM) and

xanthine (0.1072 mM). The solution was prepared by pipetting 1.072 mM

xanthine (535 yL) and 1 mM EDTA (518 yT.) stock solutions into a 50 mL

volumetric flask containing 0.02 M K^PO^ (16.8 mL) and 0.09 M F^HPO^

(25.0 mL) and diluting to the mark with water.

Uric acid standards were prepared in volumetric flasks from

dilutions of the stock solution with 1 mM KOH to afford a range of

concentrations as tabulated below.

Tube Stock Uric acid Final Uric acid no. solution, yL standard, mL volume, mL cone., mM

1 1500 - 5.0 3.003 2 1 2 0 0 - 5.0 2.403 3 930 - 5.0 1.862 ' 4 650 - 5.0 1.301 5 360 - 5.0 0. 721 6 75 - 5.0 0. 150 7 - 1 . 0 (tube //6 ) 2 . 0 0.0751

5. Spectrophotometer Parameters

Xanthine oxidase activity was measured according to the method of Avis, Bergel and Bray by spectrophotometrically monitoring uric acid

formation at 295 nm at 25°C (15). Initial rate studies employed a 0.1

AUFS span and 1 in/min chart speed. Traces of the reaction time course

acquired over 45 min were obtained with a 1.0 AUFS span and 0.2 in/min

chart speed. A period of 4 s and 1 mm slit width were utilized. 195

6. Spectrophotometric Linearity

Absorbanco-concentration H non r ity under the conditions employed

was demonstrated by measuring the absorbance at 295 nm (^2 9 5 ) f°r a

series of uric acid solutions. The reference cuvette contained

phosphate buffer as prepared for dilution of the enzyme. Sample

solutions were prepared by pipetting uric acid standards (100 yL) into

buffer (2.9 mb) in a 10 mm cuvette to afford a range of concentrations

from 2.503 x 10- 6 M to 1.001 x 10- 4 M (Table A.3).

The span was adjusted to the maximum scale expansion permitted for

the sample, and the instrument was calibrated and zeroed prior to the

addition of uric acid solution to the sample cell. Absorbances were

calculated from traces of the recorder deflection from baseline.

7. Enzyme Activity Assay

Prior to each run, the instrument was calibrated and zeroed with the sample cuvette containing the buffered xanthine standard (2.8 mL),

inhibitor standard or blank ( 1 0 0 yl.) and enzyme blank (100 yL) solutions. All measurements were made against a reference sample prepared similarly using inhibitor blank in place of the standard solution. The enzyme and inhibitor were preincubated by addition of the inhibitor standard or blank (150 yL) to enzyme stock solution (150 yL) in a polypropylene microcentrifuge tube and stirred for approximately 5 s on a vortex mixer. After I min had elapsed, the reaction was initiated by pipetting an aliquot (200 yL) of the mixture into the sample cuvette containing fresh xanthine solution (2.8 mL) with subsequent activation of the spectrophotometer recorder. The cuvette was stoppered, mixed by inverting twice and placed in the cell 196

compartment. The time which elapsed from inflation of the reaction

until the spectrophotometer stabilized was approximately 20-25 s. The

absorbance was continuously recorded until it exceeded 0.1 AU in the

initial rate studies. Initial rates were calculated from the slope

(AA2 9 5 /Atmin) of the initial linear region of the profiles, which

comprised no more than 5% of the theoretical absorbance change for

complete reaction.

8 . Fractional Inhibition

The initial fractional inhibition as a function of inhibitor

concentration was calculated from values of the initial rate determined

in the presence and absence of allopurinol and the compounds being

evaluated. Three replicate reactions for each inhibitor concentration were run in succession, followed by a single run employing the

corresponding inhibitor blank. Mean values of the initial rate for the

reaction in the presence (v^) and absence (v) of the inhibitor were used to calculate Initial fractional inhibition (i) as defined by

i = (v - v 1)/v (4.3)

C. Results and Discussion

An extensive body of literature concerning the kinetics and mechanism of reactions catalyzed by xanthine oxidase has been generated.

However, comparisons of the data presented by different workers is often made difficult due to a lack of uniformity in the conditions under which studies were performed. This is of particular concern since the 197

complexities of xanthine oxidase reactions are manifested in fln extreme

sensitivity to the conditions employed.

For example, the steady state parameters of the reaction are highly

pH dependent (6,16,17). Vmflx increases with pH, being particularly

sensitive in the region 5 to 8 , until a plateau is reached near pH 8.5

and Km exhibits a minimum at pH 7. Similarly, a pH dependence of would be be anticipated. In addition, the magnitude of the absorbance

change associated with the formation of uric acid, which is usually measured to monitor the reaction, is also pH dependent (16,17).

Xanthine oxidase is subject to significant substrate inhibition even at

relatively low initial substrate concentrations (3,6,8,18).

Furthermore, enzyme activity is rapidly degraded if precautions are not taken to exclude transition metal ions from the system (2,3).

In general, there are two approaches to the design of enzyme inhibition studies, the underlying difference being the amount of information which is obtained (19,20). In the classical and more extensive experiment, initial rates are measured as both the initial substrate and inhibitor concentration are varied. A series of

Lineweaver-Burk plots is constructed, from which Km, Vmflx, and the mechanism of inhibition may be derived. The parameter K^, or enzyme- inhibitor dissociation constant, is a quantitative indicator of inhibitor potency.

Alternately, a less intensive experimental approach examines the dependence of initial rate on inhibitor concentration at a single substrate concentration. The data provide an initial fractional inhibition profile, from which the 50 percent inhibitory concentration 198

of the inhibitor (I5 0 ) may bn ost. fmutod. The parameter T^q is a measure

of the relative potency of an inhibitor and has been utilized for

comparative studies.

Since the objective of the present study was to compare the in vitro

inhibitory effects of merbarone and its major urinary metabolites on

xanthine oxidase to that of allopurinol, the latter approach was taken.

Flexibility to expand the study to include additional substrate

concentrations depending on the results of the initial experiments was

retained.

Xanthine oxidase activity has been determined in numerous buffer

systems differing in composition and pH, employing both xanthine and

hypoxanthine as the substrate. However, in studies conducted since

1960, there is a definite trend toward the use of a reaction medium

composed of 0.05 M potassium phosphate buffer, pH 7.8, containing 0.01

mM EDTA, air-equilibrated at 25°C (3,8,10-13,21). Xanthine is used

almost exclusively as the substrate (2-6,10-18,21) with adoption of the

assay method developed by Avis, et al. (15), in which the absorbance

change at 295 nm is monitored. The enzyme concentration in the

reactions is commonly that which results in an initial rate of

approximately 0.03 AU/min for 0.1 mM xanthine. These conditions and methods were used in the present studies, since they were employed in most of the allopurinol inhibition work (10-13). It should be noted that the concentrations and conditions used in iji vitro enzymatic studies are not meant to reflect the intracellular environment ( 1 ,1 2 ).

However, the in vitro potency of inhibitors is usually assumed to afford a relative indication of potential ijt vivo activity. 199

The Inhibition mechanism of xanthine oxidase by allopurinol is

itself rather complex, and merits some comment with regard to

experimental methodology. Inactivation results from the enzymatic

conversion of allopurinol, a substrate of xanthine oxidase, to

alloxanthine (4,6 -dihydroxypyrazolo[3,4-d]pyrlmidine) by 6 -hydroxyla-

tion. The alloxanthine molecule remains tightly associated with the

enzyme, trapping it in a partially reduced state (10,14). The kinetic

behavior of allopurinol has resulted, therefore, in its classification

as a stoichiometric or tight-binding inhibitor ( 1 0 ,1 1 ).

A detailed theoretical development of the kinetics and properties of

tight-binding inhibitors has heen presented (19). Thus, without

unnecessary elaboration, it will be merely stated that the time course

and initial rate of an enzymatic reaction in the presence of a tight-

binding inhibitor depends on the order of addition of components.

Therefore, the enzyme and tight-binding inhibitor are commonly

preincubated for a specific time period, with initiation of the reaction

by adding the preincubated solution to the substrate (10,11,14,19,22).

Preincubation, and the order of addition of components does not affect

the kinetics of enzymatic reactions in the presence of classical

inhibitors.

The experimental conditions which were used for the determination of

initial fractional inhibition and reaction time courses are summarized

in Tables 4.1 and 4.2. The spectrophotometer slit width and period were

adjusted to minimize noise to an acceptable level in baseline recorder

traces with the span at 0.1 AUFS. The measured A 2 gg for uric acid solutions employing these parameters are compiled in Table 4.3. Fig. 200

4.1 shows a plot of A 2 9 5 against uric acid concentration. The bp.st fit

line of the curve wns obtained by linear regression analysis employing a

weighting factor of I/A2 9 5 squared. Excellent conformity to Beer's Law

throughout the concentration range is apparent, as substantiated by the

correlation coefficient ( 1 .0 0 0 0 ) and a y-intercept that is not

significantly different from zero.

Values for the mean initial fractional inhibition determined for

allopurinol, merbarone and its major urinary metabolites are tabulated

in Table 4.4. The potency of allopurinol as a xanthine oxidase

inhibitor is clearly demonstrated in the plot of initial fractional

inhibition against allopurinol concentration (Fig. 4.2). The

characteristic sigmoidal curve approaches complete inhibition (i = 1 )

asymptotically following a sharp rise in the vicinity of I^q, which

occurs at an allopurinol concentration less than 10 yM. The initial

fractional inhibition profiles obtained for the carboxanilide barbiturates (Fig. 4.3) suggest that the Inhibitory effects demonstrated by these compounds are relatively weak In comparison. Fifty percent

inhibition of the reaction was not even achieved at the highest concentration examined, a level approaching the the solubility of merbarone. Whereas 0.1 mM allopurinol resulted in 99% inhibition, the react.ion was only inhibited 14 to 29 percent by merbarone and its metabolites. Since the degree of inhibition attained by the carboxanilide barbiturates at this concentration was similar to that for

0.5 yM allopurinol, these compounds are apparently at least two orders of magnitude less effective than allopurinol as inhibitors of xanthine oxidase. 201

This is further demonstrated in the reaction time courses which were

obtained over a period of 45 min (Fig. 4.4). Tn the presence of the

carboxanilide barbiturates at 0.01 mM, the profiles are not appreciably

different from that obtained in the absence of inhibitors. However, the

effectiveness of 0.01 mM allopurinol on Inhibiting the reaction is quite

evident.

These initial studies indicate that merbarone and its major urinary

metabolites are exceedingly weak Inhibitors of xanthine oxidase.

Therefore, additional experiments to generate data required for determining the kinetic parameters K , Vmflx Kj were not performed.

Furthermore, a reasonably strong argument may be developed to support the contention that these compounds are not enzyme inhibitors in the usual context of the term.

During the development of an analytical method for merbarone, it was observed that the compound had a disturbingly high affinity to adsorb from aqueous solution onto a wide variety of surfaces (23). Since the compound is also highly bound to plasma protein, it is reasonable to assume that it will similarly adsorb to other proteins, such as xanthine oxidase. In addition, these compounds do not appear to be substrates of xanthine oxidase, although reaction of the 2 -thiobarbiturate analogs with the hydrogen peroxide produced during the hydroxylation of xanthine could result in desulfurization to the corresponding oxybarbiturates

(24). Therefore, the weak, concentration dependent inhibitory effect of merbarone and the structural analogs on xanthine oxidase may result from the adsorption of these compounds on the enzyme in a manner which is not site specific. However, an inhibitor is usually considered to be a 202

compound that exerts its effect by a specific Interaction with an enzyme

(20).

These studies, therefore, support the view that the hypouricemia

occurring in patients upon dosing with merbarone is not due to xanthine oxidase inhibition, but primarily results from uricosuric effects

associated with metabolite excretion. This is further substantiated by

a consideration of pharmacokinetic information. The steady state plasma concentration of merbarone in patients treated with the 5 day 24 h continuous i.v. infusion regimen ranged from 0.02 to 0.3 mM. Similar concentrations were employed in the in vijtro enzyme inhibition experiments, in which only weak inhibition was observed. Since merbarone is highly bound to plasma proteins, the concentration of free drug available to interact with endogenous xanthine oxidase would be much lower. In addition, the plasma concentration of the merbarone metabolites are extremely low throughout the course of merbarone administration. Therefore, inhibition of xanthine oxidase by these compounds to the degree required to effect the observed decrease in plasma uric acid levels is unlikely. However, as demonstrated earlier in this report, a reasonably large fraction of the total dose is excreted into the urine in the form of metabolized drug. Since these compounds, including merbarone, are weak organic acids which exist largely in their dissociated forms under physiological conditions, an effect on enhancing the excretion of uric acid is possible. 203

D. References

1. Bray, R. C. Xanthine oxidase. In: Boyer, P. 0. (Ed) The Enzymes, Vol. 12, 3rd ed., pp. 300-387, New York, NY: Academic Press, 1975.

2. Bergel, F., and Bray, R. C. The chemistry of xanthine oxidase. Biochem. J., 73: 182-192, 1959.

3. Fridovich, I., and Handler, P. Xanthine oxidase. J. Biol. Chem., 237: 916-921, 1962.

4. Olson, J. S. , Ballou, D. P. , Palmer, G. , and Massey, V. The mechanism of action of xanthine oxidase. J. Biol. Chem., 294: 4363-4382, 1974.

5. Massey, V., Brumby, P. F,. , Komai, H. Studies on milk xanthine oxidase. J. Biol. Chem., 244: 1682-1691, 1969.

6 . Olson, J. S. , Ballou, D. P. , Palmer, G. , and Massey, V. The reaction of xanthine oxidase with molecular oxygen. J. Biol. Chem., 249: 4350-4362, 1974.

7. Bray, R. C. Xanthine oxidase. In: Boyer, P. 0., Lardy, H. , Myrback, K. (Eds) The Enzymes, Vol. 7, 2nd ed. pp 533-556, New York, NY: Academic Press, 1963.

8 . Fridovich, I., and Handler, P. Xanthine oxidase. J. Biol. Chem., 233: 1581-1585, 1958.

9. Roussos, G. G. Xanthine oxidase from bovine small intestine. In: Colowick, S. P. and Kaplan, N. 0. (Eds) Methods in Enzymology, Vol. XII, pp. 5-16, New York, NY: Academic Press, 1967.

10. Spector, T. , and Johns, D. G. Stoichiometric inhibition of reduced xanthine oxidase by hydroxypyrazolo[3,4-d]pyrimidines. J. Biol. Chem. 245: 5079-5085, 1970.

11. Cha, S., Agarwal, R. P., and Parks, R. E. , Jr. Tight-binding inhibitors-II. Biochem. Pharmacol., 24: 2187-2197, 1975.

12. Johns, D. G. , Spector, T. , and Robins, R. K. Studies on the mode of oxidation of pyrazolo(3,4-d)pyrimidine by aldehyde oxidase and xanthine oxidase. Biochem. Pharmacol., 18: 2371-2383, 1969. 204

13. Spector, T. and Johns, D. G. 4-Hydroxypyrazolo( 3,4-d) pyrimidine as a substrate for xanthine oxidase: I.oss of conventional substrate activity with catalytic cycling of the enzyme. Biochem. Biophys. Res. Communs., 38: 583-589, 1970.

14. Massey, V., Komal, H., and Palmer, G. On the mechanism of inactivation of xanthine oxidase by alluropurinol and other pyrazolo[3,4-d]pyrimidines. J. Biol. Chem., 245: 2837-2844, 1970.

15. Avis, P. G. , Bergel, F. , and Bray, R. C. Cellular constituents. The chemistry of xanthine oxidase. Part 1. The preparation of a crystalline xanthine oxidase from cow's milk. J. Chem. Soc. 1100-1105, 1955.

16. Fridovich, I. Synergistic inhibition of xanthine oxidase by guianidinium plus thiocyanate. Arch. Biochem. Biophys., 109: 511-515, 1965.

17. Krebs, E. G. and Norris, F,. R. The competitive inhibition of xanthine oxidation by xanthopter1n . Arch. Biochem., 24: 49-54, 1949.

18. Hoffstee, B. H. J. On the mechanism of Inhibition of xanthine oxidase by the substrate xanthine. J. Biol Chem., 216: 235-244, 1955.

19. Cha. S. Tight-binding inhibitors - I. Biochem. Pharmacol., 24: 2177-2185, 1975.

20. Laidler, K. J., and Bunting, P. S. The Chemical Kinetics of Enzyme Action, 2nd ed., Oxford, UK: Carendon Press, 1973.

21. Fridovich, I. Competitive inhibition of myoglobin of the reduction of cytochrome c by xanthine oxidase. J. Biol. Chem., 237: 584-586, 1962.

22. Williams, J. W. , and Morrison, J. F. The kinetics of reversible tight-binding inhibition. In: Colowick, S. P. , and Kaplan, N. 0. (Eds) Methods in Enzymology, Vol. 63, Part A, pp. 437-467, New York, NY: Academic Press, 1979.

23. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March 31, 1988.

24. Bush, M. T. , Hazel, P., and Chambers, J. The metabolic fate of thiobarbiturates: thiobarbitol in man. J. Pharmacol. Exp. Therap., 134: 110-116, 1961. Table. 4.1 Reaction mixtu re composition, conditions and spectrophotometer parameters used for the determination of the initial fractional inhibition of xanthine oxidase.

Xanthine: 1.00 x 10- 4 M

Xanthine oxidase: 0.011 units/mL

Inhibitors: 0 .0 - 1 . 0 0 x 1 0 - 4 M

Na 2 -EDTA: 1 . 0 0 x 1 0 - 5 M

Potassium phosphate 0.05 M (y = 0.14), buffer: pH 7.8 (± 0.02)

Temperature: 25°C

Enzyme-Inhibitor 1 min preincubation time:

Spectrophotometer 295 nm, 0.1 AUFS, parameters: 1 mm slit width

Table 4.2 Reaction mixture composition, condi tions and spectrophotometer parameters used for the acquisition of time courses for the conversion of xanthine to uric acid by xanthine oxidase.

Xanthine: 1 . 0 0 x IQ' 4 M

Xanthine oxidase: 0 . 0 1 1 units /mL

Inhibitors: 1 . 0 0 x 1 0 - 5 M

Na 2 -EDTA: I. 0 0 x 1 0 " 5 M

Potassium phosphate 0.05 M (y = 0.14), buffer: pH 7. 8 (± 0 .0 2 )

Temperature: 25°C

Enzyme-Inhibitor 1 min preincubation time:

Spectrophotometer 295 nm, 1 . 0 AUFS, parameters: 1 mm s lit width Table 4.3 Beer's Law curve for the absorbance of uric acid determined under conditions similar to that of the xanthine oxidase Inhibition experiments.

Sample Span Uric acid no. (AUFS) cone A b • ,a M 295

1 2 1 . 0 0 1 x 1 0 " 4 1. 142 2 1 8.009 x 1 0 " 5 0.914 3 1 4.207 x 1 0 " 5 0. 718 4 1 4.338 x 1 0 " 5 0.498 5 0.5 2.403 x 1 0 ‘ 5 0.270 6 0 . 1 5.005 x 1 0 " 6 0.0538 7 0.05 2.503x 1 0 ' 6 0.0241 8 0.05 0 . 0 0 0 0 . 0 0 0 0

a Solutions were buffered to pH 7.8 with 0.05 M potassium phosphate (y = 0.14) containing 0.01 mM EDTA. b Absorbances at 295 nm were determined from recorder traces employing n 1 mm slit width and 4 s period. 207

Table 4.4 Mean Initial fractional inhibition of xanthine oxidase by allopurinol, MB, 4 ’-0H-MB, 2-oxo-MB and 4'-0H-2-oxo- MB under the conditions specified in Table 4.1.

Allopurinol Cone., M ia CVb

1 . 0 0 0 x 1 0 " 4 0.994 50.03 1 . 0 0 0 x 1 0 “ 5 0.951 24.55 5.000 x 10" 6 0.928 7.76 1 . 0 0 0 x 1 0 ‘ 6 0.543 12.97 5.000 x 10' 7 0.243 3.02 1 . 0 0 0 x 1 0 ‘ 7 0.070 2.51 1 . 0 0 0 x 1 0 ' 8 0. 064 8 . 72

Merbarone 4'-0H-MB Cone,* » M i CV Cone,■ » M i CV

9.997 X 1 0 - 5 0. 173 1.37 9.903 X 1 0 “ 5 0.287 4.42 7.497 X 1 0 ' 5 0. 131 1. 72 7.427 X 1 0 ' 5 0.254 2.93 5.000 X 1 0 ' 5 0.080 2.95 4.953 X 1 0 ' 5 0 . 2 1 2 3.82 3.000 X 1 0 ' 5 0.053 3.69 2.970 X 1 0 ' 5 0. 164 3. 13 1 . 0 0 0 X 10'1 0 . 0 1 2 4.84 9. 900 X 1 0 ' 6 0.072 3.71 1 . 0 0 0 X 1 0 ' 6 0.009 4. 16 9.900 X 1 0 ' 7 0.052 4.32

2-oxo-MB 4 -OH-2-oxo-MB Cone., M i CV Cone., M i CV

1 . 0 0 0 x 1 0 - 4 0. 141 5.06 1.015 x 10" 4 0.271 7.91 7.500 x 10“ 5 0.082 4.89 7.617 x 10- 5 0.227 7.87 5.000 x 10' 5 0.066 5.11 5.077 x 10“ 5 0. 134 8 . 0 2 3.000 x 10' 5 0.035 5.32 3.047 x 10' 5 0. 107 8 . 19 1 . 0 0 0 x 1 0 ' 5 0.032 5.08 1.017 x 10' 5 0.052 8.87

a Mean initial fractional inhibition. Coefficient of variation, percent (N = 3). 0.8

E c cnm CM < 0.4

0.2

0.0 0 2 3 4 5 6 7 8 9 10 URIC ACID CONC. (10**-5 M) Fig. 4.1. Plot of absorbance at 295 nm as a function of uric acid concentration under conditions similar to that of the xanthine oxidase inhibition experiments. Solutions were buffered to pH 7.8 with 0.05 M potassium phosphate (y = 0.14) containing 0.01 mM EDTA. The best fit line, obtained from linear 208 regression employing a weighting factor of inverse absorbance squared: slope = 11,535 1/M; y-intercept, -0.004644; r, 1.0000. 100

80 o o

* 70 z o I— go | 50

< 40 o i— o cr< Li_ 20

-• -7 10-8 -5 INHIBITOR CONC. (M)

Fig. 4.2. Plot of initial fractional inhibition of xanthine oxidase by allopurinol, MB, 4'-OH-MB, 2-oxo-MB and 4'-0H-2-oxo-MB as a function of concentration. The reaction conditions are described in Table 4.1. Key: (x) allopurinol; (o) 4 ' -0H-2-oxo-MB; (+) 4'-OH-MB; (*) 2-oxo-MB; (•) MB. 209 25 o o

* z o t— CD X z _] < z o h- o < (X Li_

10~5 INHIBITOR CONC. (M)

Fig. 4.3. Profiles for the initial fractional inhibition of xanthine oxidase by MB, 4'-OH-MB, 2-oxo-MB and 4 1-0H-2-oxo-MB. The reaction conditions are described in Table 4.1. Key: (o) 4 1-0H-2-oxo-MB; (*) 2-oxo-MB; (+) 4 ’-OH-MB; (•)

MB. 210 CURVES FROM TOP DOWN 0.9 2-OXO-MB 4'-0H-2-0X0-MB NO INHIBITOR 0.8 MB V -OH-MB ALLOPURINOL 0.7

^ 0.6 c m 05 cn u -° CNI < < 0.3

0.2

0.0 0 5 10 1520 30 40 45 TIME (m in) Fig. 4.4. Time courses obtained over 45 min for the change in absorbance at 295 nm due to the xanthine oxidase catalyzed conversion of xanthine to uric acid. Profiles are shown representing reactions run in the absence of inhibitor, and in the presence of 0.01 mM allopurinol, MB, 4*-OH-MB, 2-oxo-MB and 4 ’-0H-2-oxo-MB. The reaction conditions are described in Table 4.1. BIBLIOGRAPHY

CHAPTER I

1. Brewer, A. D. , Minatelli, J. A., Plowman, J. , Pauli, K. D. , and Narayaman, V.L. 5-(N-phenylcarboxamido)-2-thiobarbituric acid (NSC 336628), a novel potential antitumor agent. Biochem. Pharmacol., 34: 2047-2050, 1985.

2. Glover, A., Chun, H. G. , Kleinman, L. M. , Cooney, D. A., Plowman, J., Grieshaber, C. K., Malspeis, L., and Leyland-Jones, B. Merbarone: an antitumor agent entering clinical trials. Investigational New Drugs, 137-143, 1987.

3. Cooney, D. A., Covey, J. M., Kang, G. J., Dalai, M., McMahon, J.B., and Johns, D.G. Initial mechanistic studies with merbarone (NSC 336628). Biochem. Pharmacol., 34: 3395-3398, 1985.

6 . Ukita, T., Kato, Y., Hori, M., and Nishizawa, H. On the anti-tumor activity of nitrogenous cyclic 3-diketones. Chem. Pharm. Bull., 8 : 1021-1028, 1960.

212 9. Ukita, T., and Mizuno, D. In vitro screening of tricarbonylmethane and related compounds for their anti-tumor effect by cylinder agar plate (CAP) method. Chem. Pharm. Bull., 8 : 1016-1020, 1960.

10. Wiley, R. H., Hussung, K. F . , Hobbs, W. E., and Huh, S. Hydroxy-2-thiopyrimidine-5-carboxaldehyde derivatives in cancer chemotherapy. J. Med. Chem., 7: 358-359, 1964.

11. Malspeis, L. , Lyon, M. E. , DeSouza, J. J. V. , and Staubus, A. E. Preclinical pharmacology studies of merbarone (NSC 336628). (Abstract) Proc. A. Assoc. Cancer Res., 27: 418, 1986.

12. Vogel, A. Vogel's Textbook of Practical Organic Chemistry, 4th ed., New York, NY: Longman, Inc., 1978.

13. Fieser, L. F., and Fieser, M. Reagents for Organic Synthesis, New York, NY: John Wiley and Sons, Inc., 1967.

14. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March, 1986.

15. Schoenenberger, H., Holheu-Eckardt, J. and Baman, E. Antimycotics. III. Comparative studies of the preparation of isomeric hydroxybenzamides. Arzneimittel. Forsch., 14: 324-328, 1964.

16. Mark, L. C. Metabolism of barbiturates in man. Clin. Pharmacol. Therap., 4: 504-530, 1963.

17. Bush, M. T. , Mazel, P., and Chambers, J. The metabolic fate of thiobarbiturates: thiobarbita1 in man. J. Pharmacol. Exp. Ther., 134: 110-116, 1961.

18. Butler, T. C. The metabolic hydroxy 1ation of phenobarbital. J. Am. Pharm. Assoc., 116: 326-336, 1956.

19. Harvey, D. J. , Glazener, I.. , Stratton, C., Newlin, J. , Hill, R. M. , and Horning, M. G. Detection of a 5-(3,4-dihydroxy-l,5-cyclohexadien-l-yl) metabolite of phenobarbital and mephobarbital in rat, guinea pig and human. Res. Commun. Chem. Path. Pharmacol., 3: 557-565, 1972.

20. Kallberg, N. , Agurell, S., Ericsson, 0., Bucht, E. , Jailing, B., and Borcus, L.0. Quantitation of phenobarbital and its main metabolites in human urine. Europ. J. Clin. Pharmacol., 9: 161-168, 1975.

21. Whyte, M. P., and Dekaban, A. S. Metabolic fate of phenobarbital. Drug Metab. Disp., 5: 63-70, 1977.

213 22. Tang, B. K., Kalow, W. , and Grey, A. A. Metabolic fate of phenobarbital in man: N-glucoside formation. Drug Metab. Disp. , 7: 315-318, 1979.

23. Hooper, W. D. , Kunge, H. F.., and Eadio, M. J. Qualitative and quantitative studies of methylphnnobarbita 1 metabolism in man. . Drug Metab. Disp., 9: 381-385, 1981.

24. Leibman, K. C. , and Ortiz, E. Further studies of metyrapone effects upon anilide hydroxylation. Drug Metab. Disp., 3: 507-512, 1975.

25. Tang, B. K. , Kalow, W. , and Grey, A. A. Amobarbital metabolism in man: N-glucoside formation. Res. Commun. Chem. Pathol. Pharmacol. 21: 45-53, 1978.

26. Tang, B. K. , Inuba, T. , and Kalow, W. N-Hydroxyamobarbital: The second major metabolite of amobarbital in man. Drug Metab. Dispos., 3: 479-486, 1975.

27. Baler, J. R., Sadler, P. J. , Nicholson, J. K., and Timbrell, J. A. Urinary excretion of acetaminophen and its metabolites as studied by proton nmr spectroscopy. Clin. Chem., 30: 1631-1636, 1984.

28. Clapp, J. W. A new metabolic pathway for a sulfonamide group. J. Biol. Chem., 223: 207-214, 1956.

29. Rogers, N. A. J., and Smith, H. 2-Acylcyclohexane-1:3-diones. Part II. 2-Formyl-, 2-propionyl-, 2-isobutyryl-, and 2-phenyl- carbamoy1 -cyclohexane-1 :3-dione, and their conversion into phenanthridines. J. Chem. Soc., 341-346, 1955.

30. Richter, R. and Ulrich, H. 5-Acyl-7-(N-arylcarbamoyl)-4,6 -dioxo- 2,3,3a,4,5,6 -tetrahydrooxa(thia)zolo-[2,3-c]pyrimidines and 3-[ N- ary 1 carbamoyl]-2 ,4-dihydroxyqui noli nes from 2 -methyloxa(thia)zol- ine and aryl isocyanates. J. Org. Chem., 44: 4877-4880, 1979.

31. Arbuzov, B. A., and Zubova, N. N. Addition of aliphatic and aromatic acyl isocyanates to unsaturated compounds. Synthesis, 433-450, 1982.

32. Diehl, R. E., Schriber, M. S., and Kantor, S. Dioxocyclohexanecar- boxanilide insecticides and acaricides. U.S. Patent 3,801,630, April 2, 1974.

33. March, J. Advanced Organic Chemistry, 2nd ed. , New York, NY: McGraw-Hill Book Company, 1977.

214 34. Greene, T. W. Protective Groups in Organic Synthesis, New York, NY: John Wiley and Sons, Tno., 198!.

35. Munch-Peterson , J. m-Ni t. robenznz i He. Org. Syntheses Coll. Vol. IV: 715-717, 1963.

36. Newman, N. A., Lee Jr., S. H. , and Garrett, A. B. Solvent effect in the Curtius rearrangement of benzazide. J. Am. Chem Soc., 69: 113-116, 1947.

37. Allen, C. F. H., and Bell, A. Undecyl isocyanate. Org. Syntheses Coll. Vol. Ill: 846-847, 1955.

38. Gibson, G. G., and Sket, P. Introduction to Drug Metabolism, New York, NY: Chapman and Hall, Inc., 1986.

39. Chattaway, F. D. Acetylation in aqueous alkaline solution. J. Chem. Soc., 2495-2496, 1931.

CHAPTER TI

1. Chattaway, F. D. Acetylation in aqueous alkaline solutions. J. Chem. Soc., 2495-2496, 1931.

2. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March, 1986.

CHAPTER III

1. Malspeis, L. Study of the clinical pharmacology of antitumor drugs. Volume I: merbarone (NSC 336628). Volume IT: amonafide (NSC 308847). Report to the National Cancer Institute, March, 1988.

2. Malspeis, L. , Lyon, M. E. , DeSouza, J. J. V., and Staubus, A. E. Preclinical pharmacology studies of merbarone (NSC 336628). (Abstract) Proc. Am. Assoc. Cancer Res., 27: 402, 1986.

3. Kemmenoe, B. H., and Malspeis, L. Distribution of 2-^C-merbarone in mice by autoradiography of whole-body cryosections. Cancer Res., 47: 1135-1142, 1987.

215 4. Kutt, H. Winters, W. Scherman, R., and McDowell, F. Diphenylhydan- toin and phenobarbital toxicity - The role of liver disease. Arch Neurol., 11: 649-656, 1964.

5. Glover, A., Chun, H. G. , Klelnman, L. M. , Cooney, D. A., Plowman, J. , Grieshaber, C. K. , Malspeis, 1.., and Ley land-Jones, B. Merbarone: an antitumor agent entering clinical trials. Investigational New Drugs, 137-143, 1987.

6 . Boyd, E. M., and Bereczky, G. M. Liver necrosis from paracetamol. Brit. J. Pharmacol., 26: 606-614, 1966.

7. Mitchell, J. R., Jollow, D. J., Gillette, J. R., and Brodie, B. B. Drug metabolism as a cause of drug toxicity. Drug Metab. Disp., 1: 418-423, 1973.

9. Mitchell, J. R., Jollow, D. J. Potter, W. Z. Davis, D. C., Gillette, J. R., and Brodie, B. B. Acetaminophen-induced hepatic necrosis. I. Role of drug metaboljsm. J. Pharmacol. Exp. Therap., 187: 185-194, 1973.

10. Hinson, J. A., Mitchell, J. R. , and Jollow, D. J. Microsomal N- hydroxylation of p-chloroacetanilide. Mol. Pharmacol., 11: 462-469, 1975.

11. Hinson, J. A., Nelson, S. D. , and Mitchell, J. R. Studies on the microsomal formation of arylating metabolites of acetaminophen and phenacetin. Mol. Pharmacol., 13: 625-633, 1977.

12. Miner, D. J. and Kissinger, P. T. Evidence for the involvement of N-acety 1-p-quinoneimine in acetaminophen metabolism. Biochem. Pharmacol., 28: 3285-3290, 1979.

13. Hinson. J. A., Pohl, L. R. , and Gillette, J. R. N-hydroxyacet- aminophen: a microsomal metabolite of N-hydroxyphenacetin but apparently not of acetaminophen. Life Sciences, 24: 2133-2138, 1979.

14. deVries, J. Hepatotoxic metabolic activation of paracetamol and its derivatives phenacetin and benorilate: oxygenation or electron transfer? Biochem. Pharmacol., 30: 399-402, 1981.

15. Jollow, D. J. , Thorgeirsson, S. S., Potter, W. Z., Hashimoto, M. , and Mitchell, J. R. Acetaminophen-induced hepatic necrosis. VI. Metabolic disposition of toxic and nontoxic doses of acetaminophen. Pharmacology, 12: 251-271, 1974.

216 16. Andrews, R. S., Bond, C. C., Burnett, J., Saunders, A., and Watson, K. Isolation and identification of paracetamol metabolites. J. Int. Med. Res., 4: 34-39, 1976.

17. Moldeus, P., Jones, D.P., Drmstad, K., and Orrenius, S. Formation and metabolism of a glutathione-S-conjugate in isolated rat liver and kidney cells. Biochem. Biophys. Res. Commun., 83: 195-200, 1978.

CHAPTER IV

1. Bray, R. C. Xanthine oxidase. In: Boyer, P. 0. (Ed) The Enzymes, Vol. 12, 3rd ed., pp. 300-387, New York, NY: Academic Press, 1975.

2. Bergel, F. , and Bray, R. C. The chemistry of xanthine oxidase. Biochem. J., 73: 182-192, 1959.

3. Fridovich, I., and Handler, P. Xanthine oxidase. J. Biol. Chem., 237: 916-921, 1962.

4. Olson, J. S., Ballou, D. P., Palmer, G. , and Massey, V. The mechanism of action of xanthine oxidase. J. Biol. Chem., 294: 4363-4382, 1974.

5. Massey, V., Brumby, P. E. , Komai, H. Studies on milk xanthine oxidase. J. Biol. Chem., 244: 1682-1691, 1969.

6 . Olson, J. S., Ballou, I). P., Palmer, G. , and Massey, V. The reaction of xanthine oxidase with molecular oxygen. J. Biol. Chem., 249: 4350-4362, 1974.

7. Bray, R. C. Xanthine oxidase. In: Boyer, P. 0., Lardy, H. , Myrback, K. (Eds) The Enzymes, Vol. 7, 2nd ed. pp 533-556, New York, NY: Academic Press, 1963.

8 . Fridovich, I., and Handler, P. Xanthine oxidase. J. Biol. Chem., 233: 1581-1585, 1958.

9. Roussos, G. G. Xanthine oxidase from bovine small intestine. In: Colowick, S. P. and Kaplan, N. 0 . (Eds) Methods in Enzymology, Vol. XII, pp. 5-16, New York, NY: Academic Press, 1967.

10. Spector, T., and Johns, D. G. Stoichiometric inhibition of reduced xanthine oxidase by hydroxypyrazolo[3,4-d]pyrimidines. J. Biol. Chem. 245: 5079-5085, 1970.

217 11. Cha, S., Agarwal, R. P., and Parks, R. F,., Jr. Tight-binding inhibitors-11. Biochem. Pharmacol., 24: 2187-2197, 1975.

12. Johns, D. G. , Spector, T. , and Robins, R. K. Studies on the mode of oxidation of pyrazolo(3,4-d)pyrimidine by aldehyde oxidase and xanthine oxidase. Biochem. Pharmacol., 18: 2371-2383, 1969.

13. Spector, T. and Johns, D. G. 4-Hydroxypyrazolo(3,4-d) pyrimidine as a substrate for xanthine oxidase: Loss of conventional substrate activity with catalytic cycling of the enzyme. Biochem. Biophys. Res. Communs., 38: 583-589, 1970.

14. Massey, V. , Komal, H., and Palmer, G. On the mechanism of inactivation of xanthine oxidase by alluropurinol and other pyrazolo[3,4-d]pyrimidines. J. Biol. Chem., 245: 2837-2844, 1970.

15. Avis, P. G. , Bergel, F. , and Bray, R. C. Cellular constituents. The chemistry of xanthine oxidase. Part I. The preparation of a crystalline xanthine oxidase from cow's milk. J. Chem. Soc. 1100-1105, 1955.

16. Fridovich, I. Synergistic inhibition of xanthine oxidase by guianidinium plus thiocyanate. Arch. Biochem. Biophys., 109: 511-515, 1965.

17. Krebs, E. G. and Norris, E. R. The competitive inhibition of xanthine oxidation by xanthopterin. Arch. Biochem., 24: 49-54, 1949.

18. Hoffstee, B. H. J. On the mechanism of inhibition of xanthine oxidase by the substrate xanthine. J. Biol Chem., 216: 235-244, 1955.

19. Cha. S. Tight-binding inhibitors - I. Biochem. Pharmacol., 24: 2177-2185, 1975.

20. Laidler, K. J., and Bunting, P. S. The Chemical Kinetics of Enzyme Action, 2nd ed., Oxford, UK: Carendon Press, 1973.

21. Fridovich, I. Competitive inhibition of myoglobin of the reduction of cytochrome c by xanthine oxidase. J. Biol. Chem., 237: 584-586, 1962.

22. Williams, J. W. , and Morrison, J. F. The kinetics of reversible tight-binding inhibition. In: Colowick, S. P., and Kaplan, N. 0. (Eds) Methods in Enzymology, Vol. 63, Part A, pp.437-467, New York, NY: Academic Press, 1979.

218 23. Malspeis, L., and Staubus, A. E. Assay development and preclinical pharmacology studies with merbarone (NSC 336628). Report to the National Cancer Institute, March 31, 1988.

24. Bush, M. T. , Hazel, P., and Chambers, J. The metabolic fate of thiobarbiturates: thiobarbitol In man. J. Pharmacol. Exp. Therap., 134: 110-116, 1961.

219 STUDIES ON THE METABOLISM OF MERBARONE IN MAN

Jeffrey G. Supko, Ph.D.

The Ohio State University

Professor Louis Malspies, Adviser

Merbarone (1,2,3,4-tetrahydro-6 -hydroxy-4-oxo-2-thioxo-pheny1-5-py- rimidinecarboxamide; NSC 336628) is a new investigational anticancer agent which was recently introduced into phase I clinical trials. This structurally interesting compound belongs to a class of barbituric acid derivatives that have not been the object of much attention in the past.

Aside from information derived in precl Inical studies of the drug, the biotransformation of compounds of this type has not been examined.

The major urinary metabolites of merbarone were identified from

HPLC, UV spectroscopy, NMR spectroscopy and mass spectrometry as

4'-hydroxy-2 -oxo-desthiomerbarone, 4'-hydroxymerbarone and 2 -oxo- desthiomerbarone. These compounds have been prepared by chemical synthesis.

An HPLC method has been developed for the concurrent quantitation of of merbarone and its major metabolites in a single urine aliquot. The daily urinary excretion of merbarone and its metabolites was monitored for 24 patients treated with merbarone according to the 24 h continuous i.v. infusion (civ) daily x 5 and 2 h civ daily x 5 schedules under

evaluation In the phase I trials. During and for 2 days subsequent to

treatment, 20-40% of the dose was excreted in the urine with only

0.2-2.8 % excreted as parent drug. The total percentage of the dose

excereted in the urine was similar for the patients on the two

schedules. The two principal urinary metabolites were 2-oxo- desthiomerbarone and 4'-hydroxymerbarone. At present, it is unclear whether there is a difference related to dose or schedule in the cumulative urinary excretion of these two metabolites, since the number of subjects was limited and the inter-subject variability was relatively

large. On both schedules, the mean cumulative excretion of

4 1 -hydroxy-2 -oxo-desthiomerbarone was significantly less than that of the other matabolites.

The excretion rates of the drug and the two major metabolites were relatively uniform from days 2 to 5, while that of the third metabolite typically did not peak until somewhat later during the infusion. Since the levels of the metabolites In plasma were extremely low relative to that of merbarone, the renal clearance of the metabolites appears to be significantly greater than that of the parent drug.

Hypouricemia was observed in all patients treated with merbarone.

Preliminary studies demonstrated that merbarone and its metabolites were extremely weak inhibitors of xanthine oxidase in vitro, suggesting that uricosuric affects associated with the excretion of the drug and its metabolites may be operative.