USF Libraries
USF Digital Collections

N-thiolated β-lactams


Material Information

N-thiolated β-lactams chemistry, SAR and intracellular target of a novel class of antimicrobial and anticancer agents
Physical Description:
Heldreth, Bart Allan
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:


Subjects / Keywords:
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


ABSTRACT: N-Thiolated β-lactams (1) represent a promising new group of compounds with potent inhibition effects on bacteria, like Bacillus anthracis and methicillin resistant Staphylococcus aureus, and onco-systems, like breast cancer and leukemia. Originally developed as part of a synthetic pathway to bicyclic lactams, N-thiolated β-lactams have been shown in this laboratory to possess intriguing biological activities. The antibacterial activities of this new class of agents rely on novel structural features unlike those of any existing family of β-lactam drugs. The lactams seem to exert their effects intracellularly, requiring passage of the bioactive species through the cellular membrane, rather than acting extracellularly on cell wall components in the manner of penicillin and related antibiotics.The lipophilic nature of these molecules, which lack the polar side chain functionality of all other microbially-active β-lactams, suggests the compounds do not target the penicillin binding proteins within bacterial membranes but instead pass through these membranes. The biological target of these compounds has been investigated. The most active members of this β-lactam class appear to be those bearing a small branched alkyl chain on the sulfur atom. The effects of stereochemistry, branching and chain length of the sulfur group on bioactivities were studied. This dissertation is divided into six chapters. A review of organosulfur anti-infectives is discussed in Chapter 1. The types of existing antibiotics and their modes of action will be discussed in Chapter 2. The synthesis of these novel agents is discussed in Chapter 3. A structure-activity relationship of these lactam analogues is discussed in Chapter 4.And Chapters 5 and 6 demonstrate a novel mode of action and biological target for these drugs using techniques which include target identification, metabolic effects, and reactivity kinetics.
Thesis (Ph.D.)--University of South Florida, 2004.
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Bart Allan Heldreth.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 249 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
notis - AJU6772
usfldc doi - E14-SFE0000511
usfldc handle - e14.511
System ID:

This item is only available as the following downloads:

Full Text


N -Thiolated -Lactams: Chemistry, SAR and Intracellular Target of a Novel Class of Antimicrobial and Anticancer Agents by Bart Allan Heldreth A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Edward Turos, Ph.D. Greg Baker, Ph.D. Randy Larsen, Ph.D. Julie Harmon, Ph.D. Date of Approval: November 12, 2004 Keywords: Antifungal, Anticancer, Antibiotic, Sulfur, MRSA Copyright 2004 Bart Allan Heldreth


i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v LIST OF ABBREVIATIONS viii LIST OF SPECTRA x ABSTRACT xii CHAPTER ONE INVESTIGATIONS IN ORGANOSULFUR ANTI-INFECTIVES 1 1.1.1 General Introduction 1-2 1.1.2 Classes of Biologically-A ctive Organosulfur Compounds 2-4 1.1.3 Relationship Between Reactivity of Organosulfur Compounds and Microbiological Activity 4-7 1.2.1 The Key Biological Targets of Orga nosulfur Drugs 7-8 1.2.2 Glutathione-based Systems 8-9 1.2.3 Coenzyme A-based Systems 9-10 1.2.4 Thioredoxin-based Systems 10-12 1.2.5 Mycothiol-based Systems 12-13 1.2.6 Other Cellular Targets 13-14 1.3.0 Antibacterial Organosulfur Compounds 1.3.1 Introduction 14-17 1.3.2 Thiols 17-22 1.3.3 Disulfides 22-27 1.3.4 Trisulfides 27-29 1.3.5 Pentasulfides 30 1.3.6 N -Thiolated Compounds 31-32 1.4.0 Antifungal Organosulfur Compounds 1.4.1 Introduction 32-34 1.4.2 Thiols 34-35 1.4.3 Sulfides 35-36 1.4.4 Disulfides 37-38 1.4.5 Trisulfides 38-39 1.4.6 Polysulfides 39-40 1.5.0 Antiviral Organosulfur Compounds 1.5.1 Introduction 41-42 1.5.2 Thiols 42-43


ii 1.5.3 Disulfides 43-47 1.6.0 Antiparasitics 1.6.1 Introduction 47-49 1.6.2 Thiols 50-51 1.6.3 Sulfides 51 1.6.4 Pentasulfides 52 1.7.1 Conclusions 52-54 CHAPTER TWO DEVELOPMENT OF ANTI-MRSA AND ANTI-ANTHRAX ANTIBIOTICS 55 2.1.1 Introduction 55-58 2.2.1 -Lactam Antibiotics: Anti-MRSA 58-61 2.3.1 Vancomycin and the Peptide Antibiotics 62-63 2.4.1 DNA and Protein Synthesis Inhi biting Antibiotics 64-65 2.5.1 Anthrax 65-66 2.6.1 Conclusions 66 CHAPTER THREE N -ORGANOTHIOLATED -LACTAMS: SYNTHESIS 67 3.1.1 A Brief History 67-68 3.2.0 Synthesis Towards an SAR 3.2.1 Introduction 68-70 3.2.2 Synthetic Strategy 70-71 3.2.3 Acid Chloride Synthesis 72 3.2.4 Imine Synthesis 72-73 3.2.5 Staudinger Coupling 73-75 3.2.6 Dearylation 75-76 3.2.7 N-Substitution: Basis for the Phthalimide Transfer Reagent 76-82 3.3.1 Asymmetric Synthesis 82-90 3.4.1 Oxidation 90-93 3.5.1 An N-Resinthiolated -Lactam 93-96 3.6.1 A Fluorescence System 96-100 CHAPTER FOUR N -ORGANOTHIOLATED -LACTAMS: STRUCTURE ACTIVITY RELATIONSHIPS 101 4.1.1 Introduction: Finding Stru cture Activity Relationships 101 4.1.2 Initial Antibacterial Screening 101-102 4.2.1 MRSA Activities 102-115 4.3.1 Anthrax 116-118 CHAPTER FIVE NORGANOTHIOLATED -LACTAMS: MODE OF ACTION 119 5.1.1 Introduction 119 5.2.1 Chemical Interaction 119-122 5.3.1 Loci 122-125 5.4.1 Sugar Uptake 125-126


iii 5.5.1 Identity of the Intracellular Target 127-130 5.6.1 CoA-Antibacterial Effects 130-133 CHAPTER SIX N -ORGANOTHIOLATED -LACTAMS: OTHER BIOLOGICAL ACTIVITIES AND CONCLUSIONS 134 6.1.1 Fungi 134-136 6.2.1 Neoplasmic Systems 136-140 6.3.1 Conclusions and Future Directions 140 CHAPTER SEVEN MATERIALS AND METHODS 141 7.1 Synthetic Procedures 7.1.2 Preparation of Imines 141 7.1.3 Preparation of Acid Chlorides 142 7.1.4 Preparation of N -4-Anisyl Azetidin-2-ones 143 7.1.5 Dearylation of N-Anisyl Azetidin-2-ones 143 7.1.6 Preparation of Phthalimide-based Sulf ur Transfer Reagents 145 7.1.7 Preparation of N-Thiolated Azetidin-2ones 148 7.1.8 Preparation of Thioesters 152 7.1.9 Preparation of N-Sulfoxylated -Lactams 152 7.1.10 Preparation of N-Sulfonylated -Lactams 154 7.1.11 Preparation of a -Lactam Sulfonic Acid 154 7.1.12 De-Acetylation of a C 3 Acetoxy Substituted -Lactam 155 7.1.13 Re-Acetylation of a C 3 Hydroxy Substituted -Lactam 155 7.1.14 Dansylation 156 7.2 Microbiological Test Procedures 156 7.2.1 Antimicrobial Susceptibility Testing 158 7.2.2 MIC Calculations 158 7.2.4 Glucose Uptake / Respiration Study 159 7.2.5 Resin / Lysate Exposure 163 7.2.6 HPLC Experiments 163 CHAPTER EIGHT SPECTRA 164-223 REFERENCES 224-234 ABOUT THE AUTHOR End Page


iv LIST OF TABLES Table 1.1. Antibacterial Activities of Allicin 16 Table 1.2. Values (M) of Fluoroand Mercapto -Substituted Pyrimidines 20 Table 1.3. MICs of Quinoline Deriva tives 21 Table 1.4. Bioactivities of Bismuth-Coordina ted Thiols Against Me thicillin-resistant S. aureus 22 Table 1.5. Antibacterial Activities of Ajoene 23 Table 1.6. Antibacterial Activity of Representative Enediynes, 10 and 11, versus Penicillin G 29 Table 1.7. Antibacterial Activities of Li ssoclinotoxin A 30 Table 1.8. Antifungal Activities of Ajoene 33 Table 1.9. Antifungal Activities of Allicin 34 Table 1.10. Antifungal Activities of Thione Derivatives 38 Table 1.11. Enediyne Antifungal Ac tivities. 39 Table 1.12. Isothiazole Anti-Poliovirus Activities 45 Table 1.13. Growth Inhibition (IC 50 ) Activities of Dasuansu (Diallyl Trisulfide ( 17)) Against Parasites 48 Table 1.14. Antiprotazoal Activ ity of Garlic Extracts. 49 Table 1.15. Anti-Malaria Activ ities of Lissoclinotoxin A and a Few Clinical Standards 52 Table 3.1 Effect of -Lactam Substitutions on Antibacterial Activities 70 Table 4.1 N -Alkylthiolated -Lactam Vs. Standard Antibiotics 105 Table 4.2 N -Alkylthiolated -Lactam Vs. Standard Antibiotics 108 Table 4.3 Effects of Chain Branching and Cyclic Substituents 112 Table 4.4 Effects of Sulfur Oxidation St ate on Bio-Activity 115 Table 4.5 N -Alkylthiolated -Lactams Vs. Anthrax 118 Table 4.6 N-sec Butylthiolated -Lactam Vs. Cipro Against Bacillus Species 119 Table 5.1 Lactam Effect on Glucose Uptake 126 Table 5.2 Amino Acids Stirred with N -Thiolated Resin 128 Table 6.1 N -Methylthiolated -Lactams Vs. Candida species 135 Table 6.2 Candida MICs 136 Table 7.1 Standard Curve 162


v LIST OF FIGURES Figure 1.1. Elemental Sulfur S 8 1 Figure 1.2. Bicyclic -Lactams and Sulfa Drugs 2 Figure 1.3. Thiosugars and Thionucleosid es 3 Figure 1.4. Sulfur Mustard 4 Figure 1.5. Reactions of Thiol Compounds with Electrophiles 5 Figure 1.6. Oxidation Reactions of Disulfides 5 Figure 1.7. Calichemicin, an Enediyne Trisulfide 6 Figure 1.8. Reactions of Disulfide Compounds with Nucleophiles 6 Figure 1.9. Namenamicin, a Novel Enediyne 7 Figure 1.10. Coenzyme A Redox System 10 Figure 1.11. Potential Drug Interactio ns with Thioredoxins 12 Figure 1.12. Mycothiol 13 Figure 1.13. ATP 14 Figure 1.14. Potential Mode of Action of Allicin 16 Figure 1.15. Natural Antibiotics from Garlic 17 Figure 1.16. S. aureus -Active Thiodiazoles 18 Figure 1.17. Mercaptotriazoles, Precursors to the Thiodiazoles 18 Figure 1.18. 5-Fluoro-2-deoxyuridine ( 24a) and 5-Thio Analogue 19 Figure 1.19. 5-Fluorouracil ( 25a) and 5-Thio Analogue 19 Figure 1.20. Example of a Boron-Complexed Benzot hiazoline Antibacterial 19 Figure 1.21. 5,7-Dichloroquinoline Thiol and Meta llated Derivatives 21 Figure 1.22. Bismuth-Coordinated Thio l 22 Figure 1.23. Gliotoxin, an Epipolyt hiodioxopiperazine (ETP) Antibacterial 24 Figure 1.24. Cyclic Polysulfide Natural Products 24 Figure 1.25. Leinamycins Mode of Action 25 Figure 1.26. Psammaplin A 26 Figure 1.27. Biaprasin 26 Figure 1.28 Psammaplin D 27 Figure 1.29. Citorellamine 27 Figure 1.30. 4-Dioxo-1,2,4,6-tetrathiepan e 27 Figure 1.31. AscidianTrisulfide and Dith iane 28 Figure 1.32. Proposed Mechanism of Enediyne Bioactivation 29 Figure 1.33. Lissoclinotoxin 30 Figure 1.34. N-Sulfenylated-Lactam 31 Figure 1.35. BIT 31 Figure 1.36. Lansoprazole and in vivo Sulfenamide 32 Figure 1.37. Mercaptotriazole Antifungals 35 Figure 1.38. Naphthoquinone Sulfides Antifungal Compounds 36 Figure 1.39. Core Structure of the Thiarubine Group of Antifungals 37


vi Figure 1.40. Core Structure of the Dithiole-Thi one Group of Antifungals 38 Figure 1.41. Lissoclinotoxin D 40 Figure 1.42. S-Acetylcysteamines Derivatives 42 Figure 1.43. Thiocyanatopyrimidine Nucleoside 43 Figure 1.44. Antiarenavirus Disulfides 44 Figure 1.45. Macrocyclic Disulfide anti-HIV Agent 44 Figure 1.46. 5,5-Diphenyl-3,3-diisothiazole Disulfide and Thiol Antipoliovirus Agents 45 Figure 1.47. Dideoxynucleotides Thiryl Radi cal-based Reverse Transcriptase Inhibitors 46 Figure 1.48. Mercaptouridine Thiryl Radical-based Reverse Transc riptase Inhibitor 46 Figure 1.49. Nucleosidic Disulfide anti-HIV Agent 47 Figure 1.50. Proposed Mechanism of Sulfur-bearing Nucleotide-based Reverse Transcriptase Inhibition 47 Figure 1.51. Antiprotazoal Agent T-Cadinthiol 50 Figure 1.52. Typanocidals Cymelasan and 2Melarsenyl 51 Figure 1.53. A Methylthiobenzimidazole Fasc iolicide 51 Figure 2.1. Penicillin-analogues 55 Figure 2.2. Common Classes of -Lactam Antibiotics 56 Figure 2.3. Increase of S. aureus Resistance, as Reported by the Centers for Disease Control (CDC) 57 Figure 2.4. Open ring Penicillin Biosyntheti c Intermediate 59 Figure 2.5. Annulation and Epimerization of Open ring Lactam 59 Figure 2.6. Ring Expansion and Hydroxyla tion to Deacetyl Cephalosporin C 60 Figure 2.7. Penicillins Inte raction with PBPs 60 Figure 2.8. N -Thiolated -Lactam Antibiotics 61 Figure 2.9. -Lactamase Ring Opening of Penicillin 61 Figure 2.10. Vancomycin 63 Figure 2.11. Teicoplanin, an Example of a Se mi-synthetic Analogue of Vancomycin 63 Figure 2.12. Novobiocin 64 Figure 2.13. Linezolid 65 Figure 2.14. Ciprofilaxin 66 Figure 3.1. Discovery of Novel N -Substituted Monocyclic -Lactams 68 Figure 3.2. -Lactam Substitutions 68 Figure 3.3. Retrosynthetic Analysis 70 Figure 3.4. Standard Synthetic Overvi ew 71 Figure 3.5. Acid Chlorination 72 Figure 3.6. Imine Formation 73 Figure 3.7. Possible Staudinger Coupling Mechanisms 75 Figure 3.8. Dearylation 76 Figure 3.9. Failed Direct N -Organothiolation Method 77 Figure 3.10. Failed Transfer N -Organothiolation Method 78 Figure 3.11. The N -Organothiolation Methods 78 Figure 3.12. Phthalimide Synthesis 79


vii Figure 3.13. Alternative Phtha limide Synthesis Routes 80 Figure 3.14. Unsuccessful Phthalimide-based Transfers 81 Figure 3.15. Initial Enzymatic Scheme Towards Enantiomerically Pure Thiols 83 Figure 3.16. Nicotine Based Scheme Toward s Enantiomerically Pure Thiols 84 Figure 3.17. Second Enzymatic Scheme Towa rds Enantiomerically Pure Thiols 85 Figure 3.18. Successful Synthesis of an N-Thiolated -Lactam from an Alcohol 86 Figure 3.19. Successful Synthesis of Enantio merically Pure N-sec-Butylthiolated -Lactam from Alcohols 87 Figure 3.20. Successful Synthesis of Enantiomerically Pure -Lactams 88 Figure 3.21. Dearylation of p-Methoxyphenyl Substituted -Lactams 89 Figure 3.22. Completion of Full Stereochemical Control 90 Figure 3.23. First Ineffectual Pathway to Oxidized Sulfur Substitutions 90 Figure 3.24. Second Ineffectua l Method of Sulfoxyl a nd Sulfonyl Transfer 91 Figure 3.25. Route to Sulfoxides and Sulfone 92 Figure 3.26. Route to Sulfonic Acid and Salt 92 Figure 3.27. Accomplished Sulfur Substitutions 93 Figure 3.28. Attempted Direct Thiolation 94 Figure 3.29. Successful Synthesis of an N-Resinthiolated -Lactam 96 Figure 3.30. Attempted Synthesis of a Model FRET System 97 Figure 3.31. Methylation of Fluorescein 98 Figure 3.32. Thiolation of Fluorescein 99 Figure 3.33. Fluoresceinylthiolation of the Phthalimide Reagent 100 Figure 3.34. Fluoresceiny lthiolation of the -Lactam 100 Figure 3.35. Coumarin / Din itrophenyl FRET Paired -Lactam 100 Figure 4.1. Kirby-Bauer Disk Diffusion Antibacterial Testing 102 Figure 4.2. Initial Evidence to a Dependence on Sulfur 103 Figure 4.3. Initial Variations in the Sulfur Sidechain Oxidation State 104 Figure 4.4. Relationship of C 3 Side Length to Antibacterial Activities 106 Figure 4.5. Comparison of N-sec-Butylthiolated -Lactam to Clinical Standards 110 Figure 4.6. Attempt at Hart Meth od of N-Organothiolation 113 Figure 4.7. Differences Amongst Diastereomers 116 Figure 4.8. Kirby-Bauer Scr eening of N-Thiolated -Lactams Against Anthrax 117 Figure 5.1. Nucleophilic Attack at C 2 120 Figure 5.2. Nucleophilic Attack at the Alpha-carbon of Sulfur Sidechain 121 Figure 5.3. Nucleophilic A ttack at Sulfur 121 Figure 5.4. Chemical Reactivity of N-Thiolated -Lactams 122 Figure 5.5. FRET Pair Concept 124 Figure 5.6. Testing of the L actam-thiolated Resin 128 Figure 5.7. Thiol-CoA Adduct 193 129 Figure 5.8. Effect of Glutathi one on Growth Inhibition by N -Thiolated -Lactams 133 Figure 6.1. N-Thiolated -Lactam Anti-Cancer Mode of Action 137 Figure 6.2. Apoptotic Effects of N-Thiolated -Lactams 139 Figure 6.3. Trend of An ticancer Activity 140


viii ABBREVIATIONS = alpha ATP = adenosine triphosphate ATPase = adenosine triphosphate synthase = beta BisBAL = bismuth-2,3-dimercaptopropanol BisEDT = bismuth-1,2-ethanedithiol BisPYR = bismuth-pyrithione BisTOL = bismuth-3,4-dimercaptotoluene Bis ME = bismuth-2-mercaptoethanol Bn = benzyl BTs = bismuth thiols o C = degrees Celsius 13 C = carbon 13 CAN = ceric ammonium nitrate CMV = cytomegalovirus CoA = coenzyme A CoADR = coenzyme A disulfide reductase Cys = cysteine = delta or chemical shift DIAD = diisopropyl azodicarboxylate DIBAL = diisobutylaluminum hydride DID = 5,5-diphenyl-3,3-d iisothiazole disulfide dNTP = deoxyribonucleoside triphosphate Dsb = disulfide bridge forming enzyme EC = effective concentration Et 3 N = triethylamine ETP = epipoly(thiodioxopiperazine) FRET = Forster Resonance Energy Transfer 1 H = proton 1 H NMR= proton nuclear magnetic resonance HIV = human immunodeficiency virus HIV-RT = human immu nodeficiency virus reverse transcription HPLC = high pressure liquid chromatography Hz = hertz IR = infrared J = coupling constant JUNV = Junin (agent of Argentine hemorrhagic fever) LAH = lithium aluminum hydride m CPBA = meta -chloroperoxybenzoic acid


ix MDM = monocyte-derived macrophages MEA = 2-mercaptoethylamine (cysteamine) MHz = megahertz MIC = minimum inhibitory concentration g = micrograms M = micromolar mM = millimolar MRSA = methicillin-resistant Staphylococcus aureus MRSE = methicillin-resistant Staphylococcus epidermidis NAC = N -acetyl-L-cysteine ng = nanogram Ph = phenyl ppm = parts per million S 8 = elemental sulfur TCBZ = triclabendazole TCRV = Tacaribe TLC = thin layer chromatography trx = thioredoxin VRE = Vancomycin Resistant enterococci


x LIST OF SPECTRA Spectrum 8.01 165 Spectrum 8.02 166 Spectrum 8.03 167 Spectrum 8.04 168 Spectrum 8.05 169 Spectrum 8.06 170 Spectrum 8.07 171 Spectrum 8.08 172 Spectrum 8.09 173 Spectrum 8.10 174 Spectrum 8.11 175 Spectrum 8.12 176 Spectrum 8.13 177 Spectrum 8.14 178 Spectrum 8.15 179 Spectrum 8.16 180 Spectrum 8.17 181 Spectrum 8.18 182 Spectrum 8.19 183 Spectrum 8.20 184 Spectrum 8.21 185 Spectrum 8.22 186 Spectrum 8.23 187 Spectrum 8.24 188 Spectrum 8.25 189 Spectrum 8.26 190 Spectrum 8.27 191 Spectrum 8.28 192 Spectrum 8.29 193 Spectrum 8.30 194 Spectrum 8.31 195 Spectrum 8.32 196 Spectrum 8.33 197 Spectrum 8.34 198 Spectrum 8.35 199 Spectrum 8.36 200 Spectrum 8.37 201 Spectrum 8.38 202 Spectrum 8.39 203


xi Spectrum 8.40 204 Spectrum 8.41 205 Spectrum 8.42 206 Spectrum 8.43 207 Spectrum 8.44 208 Spectrum 8.45 209 Spectrum 8.46 210 Spectrum 8.47 211 Spectrum 8.48 212 Spectrum 8.49 213 Spectrum 8.50 214 Spectrum 8.51 215 Spectrum 8.52 216 Spectrum 8.53 217 Spectrum 8.54 218 Spectrum 8.55 219 Spectrum 8.56 220 Spectrum 8.57 221 Spectrum 8.58 222 Spectrum 8.59 223


xii N -Thiolated -Lactams: Chemistry, SAR and Intracellular Target of a Novel Class of Antimicrobial and Anticancer Agents Bart Allan Heldreth ABSTRACT N -Thiolated -lactams ( 1) represent a promising new group of compounds with potent inhibition effects on bacteria, like Bacillus anthracis and methicillin resistant Staphylococcus aureus, and onco-systems, like breast cancer and leukemia. Originally developed as part of a synthetic pathway to bicyclic lactams, N -thiolated -lactams have been shown in this laboratory to possess intr iguing biological activities. The antibacterial activities of this new class of agents rely on novel structural features unlike those of any existing family of -lactam drugs. The lactams seem to exert their effects intracellularly, requiring passage of the bioactive species th rough the cellular membrane, rather than acting extracellularly on cell wall components in the manner of penicillin and related antibiotics. The lipophilic nature of these molecules, which lack the polar side chain functionality of all ot her microbially-active -lactams, suggests the compounds do not target the penicillin binding proteins within bacterial membranes but instead pass through these membranes. The biological target of these compounds has been investigated. The most active members of this -lactam class appear to be those bearing a small branched alkyl chain on the sulfur atom. The effects of stereochemistr y, branching and chain length of the sulfur group on bioactivities were studi ed. This dissertation is divided into six chapters. A review of organosul fur anti-infectives is discussed in Chapter 1. The types of existing antibiotics and their modes of actio n will be discussed in Chapter 2. The


synthesis of these novel agents is discussed in Chapter 3. A structure-activity relationship of these lactam analogues is discussed in Chapter 4. And Chapters 5 and 6 demonstrate a novel mode of action and biological target for these drugs using techniques which include target identification, metabolic effects, and reactivity kinetics. N O O S Cl R' R (1) xiii


CHAPTER ONE INVESTIGATIONS IN ORGANOSULFUR ANTI-INFECTIVES An all too often overlooked group of anti-infective compounds are those whose biological activity is based on sulfur functionalities. As a prelude to my research on N-thiolated -lactams, this chapter provides a review of organosulfur anti-infectives. 1.1.1 General Introduction As far back as 1000 B.C. elemental sulfur (S 8 ) (1) was used, at the very least, as a pesticide. In 1824 S 8 was shown to treat peach mildew [1] and over the centuries, S 8 has successfully been applied as a fungicide to protect a wide variety of plants against fungal infestation. The current rationale for these antifungal properties is that S 8 is absorbed by the spores of the fungi, and converted within the fungi to toxic hydrogen sulfide, a compound which could have deleterious effects on mitochondrial respiration. [1] Although the mode of S 8 s antifungal action is not fully defined, there is no doubt of its dependency on the reaction of sulfur with a biological target. SSSSSSSS 1 Figure (1.1). Elemental Sulfur S 8 The use of organosulfur compounds to control the onset or progression of infectious diseases in humans and animals also has its roots dating back to early times, when ancient Egyptians recognized the potent medicinal effects of naturally occurring 1


organosulfur substances from leeks. In this report we summarize the types of molecules and reaction mechanisms associated with the anti-infective properties of various organosulfur substances. 1.1.2 Classes of Biologically-Active Organosulfur Compounds Organic compounds that contain sulfur cover an extraordinary range of chemical structures and reactivities. Many of these have biological activity. In the simplest case, the presence of one or more sulfur atoms in a biologically active molecule may not actually give the compound its biological effects, but rather may act as a non-participant in a side chain residue or an innocuous constituency of the molecular framework. A few prime examples are the bicyclic beta-lactams, including penicillins (2), cephalosporins (3), and penems (4), and the sulfa drugs (5), whose anti-infective activity is not related to sulfur-centered events (Figure 1.2). N S O RCONH CO2H N S O CO2H X RCONH N S OH R CO2H O HN S NH O O R' R 2345 Figure (1.2). Bicyclic -Lactams and Sulfa Drugs. Alternatively, the presence of a sulfur atom in a bioactive molecule may exert a more definitive, yet subtle, effect on its biological activity, such as the case of the 2


thiosugars (6a-c) and thionucleosides (7). These sulfur analogues of the natural sugars and nucleosides act as inhibitors of glycosidases and reverse transcriptases, respectively. [2-8] It has been documented that these properties are due to the conformational and stereoelectronic changes brought about by the replacement of oxygen with sulfur in the heterocyclic ring. However, even in these molecules, the sulfur atom does not, per se, play a central role in the reaction of the molecule with a biological entity. S OH HO HO N NH R O O S S S OH OH OH OH OH OH OH HO OH H3C HO OH HO HO HO (6a)(6b)(6c)7D-5-ThioglucoseL-5-ThiofucoseD-5-Thiomannose Figure (1.3). Thiosugars and Thionucleosides. In many other cases, however, the bioactivity of an organosulfur compound may be directly attributable to the reactivity of the sulfur center, which is certainly the case for the sulfur mustards (8). Here, the sulfur atom is responsible for activating the compound toward nucleophilic attack by displacing a chlorine atom. Once activated, the sulfur atom is then a potent electrophile for attack by a biological nucleophile (Figure 1.4). 3


Cl S Cl Cl S Cl S Nu Nu 8 Figure (1.4). Sulfur Mustard. In many instances, it is the sulfur center itself which is the site of chemical attachment of the biomolecules to its target. What follows in this review is a discussion of the antibacterial, antifungal, antiviral, and antiparasitic activities of organosulfur compounds whose mode of action depends on the reactivity of one or more sulfur atoms in that molecule with a biological target. 1.1.3 Relationship Between Reactivity of Organosulfur Compounds and Microbiological Activity. Biologically active organosulfur compounds can interfere with the processes associated with human infections and disease through a number of common pathways, depending on the type of sulfur functionality that is present and the nature of its reactivity. For instance, thiols are relatively reactive groups which can act as either powerful nucleophiles or reducing agents, depending on the nature of the electrophile (E + ), the thiol substituent (R), and the local environment in which the reaction occurs (equations 1 and 2). 4


ER-S-E + H (1) R-SH+ R-S + E + H (2) Figure (1.5). Reactions of Thiol Compounds with Electrophiles. Disulfides, on the other hand, behave as electrophilic reactants in the presence of various thiophilic nucleophiles, resulting in the heterolytic cleavage of the sulfur-sulfur bond (equation 3). Intracellular nucleophiles such as glutathione (9) and thioredoxins (trxs), for example, can react rapidly with diand trisulfide compounds to disrupt cellular stasis (Figure 1.6). [9,10] RS S R SSNHNH2HOOONHHOOO R Glutathione / mixed disulfide adductHO2+ H2O2O2H2O GlutathioneSHNHNH2HOOONHHOOO 9 Figure (1.6). Oxidation Reactions of Disulfides. Alternatively, disulfides can act as electrophilic oxidants (equation 4), which can lead to the generation of superoxide and hydrogen peroxide within the cell via redox pathways, thereby affecting oxidative conditions in and around the cell (Figure 1.8). These two pathways, represented by equations 3 and 4, are generally competing within the cell, and may both lead to cell death. 5


A number of other sulfur functionalities act in similar fashion to either thiols or disulfides and thus serve as masked versions of these reactive groups. Bismuth thiols (Figure 1.21), for instance, have greater potency, increased selectivity, reduced toxicity, and better stability than the free thiol, while trisulfides, such as that found in calichemicin (10), and certain sulfenamides, behave mechanistically similarly to disulfides. Some trisulfide compounds may also result in the generation of a destructive radical species, such as the case of calichemicin (10) and esperamicins (11). [11-16] O O S O O O HN O O O O HO CH3 CH3 HO NH H3C O OH HO O I CH3 O O CH3 CH3 CH3 H3C SSSCH3 O H3COOCHN OH H H3C 10 Figure (1.7). Calichemicin, an Enediyne Trisulfide. NuR-S-Nu + SR (3) +Nu + SR + RS (4)R-S-S-R Figure (1.8). Reactions of Disulfide Compounds with Nucleophiles. 6


O O O O CH3 HO HN O CH3 SSSCH3 O H3COOCHN OH H H3CS O HO O H3C S HO H3C CH3 H3C 11 Figure (1.9). Namenamicin, a Novel Enediyne. The focus of this review centers on the structures and modes of action of those agents whose anti-infective properties have been proven, or postulated, to be dependent on the sulfur atom as the primary site of biochemical reactivity. This list includes those already mentioned, thiols, disulfides, trisulfides, as well as related moieties such as sulfenamides, thiosulfinates and thiosulfonates. Compounds that contain groups such as sulfenic acids, sulfoxides and more highly oxidized compounds like sulfinic acids, sulfones, sulfonic acids, sulfamates and sulfate esters are not discussed here since their activities are rarely directly dependent on the sulfur atoms. Those compounds that are only dependent on sulfur for such properties as architecture, polarity, stability or solubility are also not included here. 1.2.1 The Key Biological Targets of Organosulfur Drugs Over the course of time, living cells have developed defenses against the harmful effect of biological oxidants, or disinfectants. The thiol-disulfide redox equilibrium in cells is central to this natural defense mechanism, and thus serves as a potentially valuable target of sulfur-based antibiotics. Nature protects microbes from oxidative stress 7


8 by maintaining high thiol:disulfi de ratios, typically 19:1 or higher. Disruption of this redox system can alter many vital cellular activ ities such as regulation of protein activity, regeneration of enzymatic co factors and reductases, like ri bonucleotide reductase, and a host of other processes where an antioxidant is required. [17] 1.2.2 Glutathione-based Systems One of the most importan t thiol/disulfide dyads f ound in many cell types is glutathione (9) ( -L-glutamyl-L-cysteinylglycine), a tripeptide assembled exclusively in the cell. Glutathione is produced biologically via the glutaredoxin enzymatic pathway in two steps: 1) formation of the glutamyl-cys teine adduct and 2) s ubsequent attachment addition of glycine to the c-terminus. The equi librium between free thiol and disulfide, at stasis, is typically maintained at a cy toplasmic thiol concentration of around 90%. Although intracellular glutathione concentrations are notoriously difficult to measure, and can vary throughout the life cycle of a cell, cell lines with high intracellular glutathione levels are generally less susceptible to damage by organosulfur drugs. [18] Serving a primary role as an antioxidant, glutathione is extremely effective in scavenging reactive free radicals, electrophiles and other destruct ive oxidants in the cytoplasm. Glutathione can react directly with various drugs to deactivate them before they are able to inflict irreversible damage, while also serving to re activate enzymes that have been inhibited as mixed disulfides formed between a drug an d enzyme. In response to this protective mechanism afforded by glutathione, nature has cleverly designed prodrug molecules such as the enediyne trisulfides and anthracyclines which can interact dire ctly with glutathione as a way to be biochemically activated within the target cell. Inhi bition of glutathiones


9 antioxidant abilities can be induced by formation of glutat hione-S-S-drug mixed disulfides. The rate of thiol cleavage of th e sulfur-sulfur bond of the glutathione disulfide is directly proportional to twice th e protic acidity of the free thiol. [19] 1.2.3 Coenzyme A-based Systems In the past it was commonplace to assume the presence of glutathione in all organisms. Glutathione, however, is not enti rely ubiquitous. Some bacteria are totally devoid of glutathione, but in its place have some other thiol/disulfide-based redox system. S. aureus, for example, does not generate or utilize glutathi one at all, but rather produces millimolar levels of the nucleosidic entity, coenzyme A ( 12) (CoA). [20] At stasis, the bacterium maintains a ratio of thiol:d isulfide of about 95:5, via coenzyme A disulfide reductase (CoADR) (Figure 1.10). This is extraordinarily specific in its ability to reduce CoA disulfide. Mixed disulfides formed between CoA and glutathione, or other thi ophilic agents, are typically unable to be reduced by CoADR. [20,21] Therefore, anti-infective compounds that can form CoADR resistant mixed disulfides with CoA can offer an e ffective mode of inhibition against S. aureus Indeed, glutathione is an inhibitor of the redox cycle of CoA and is detrimental to the growth of this bacterium.


O N OH N NN H2N O P O O OP O O OP HO O OHO O HN HN O HS 12O N OH N NN H2N O P O O OP O O OP HO O OHO O HN HN O S AntioxidantActivityCoADR 2 Figure (1.10). Coenzyme A Redox System. 1.2.4 Thioredoxin-based Systems Another very important group of native cellular thiols / disulfides are the thioredoxins (trxs) and a related subfamily, the disulfide bridge forming enzymes (Dsbs), which span through the bacterial membrane connecting the cytoplasm with the periplasm. The characteristic Cys-X-X-Cys motif in the structures of the trxs is highly conserved in many bacteria. These dithiols undergo reversible oxidation and can quickly react with non-native thiols or disulfides. Before a drug even passes through the membrane of a bacterium however, a subfamily of thioredoxins, the Dsbs, could potentially intervene. Formation of mixed disulfides between the trxs and the organothio compound can thereby inhibit enzymatic reduction and thus can shut down cellular function. Figure 11 illustrates the cellular function of trxs and Dsbs. Dsbs, for the most part, inhabit the cytoplasmic membrane with exposed (Cys-X-X-Cys) functionalities on both the cytoplasmic and periplasmic surfaces. Dsbs are involved in electron transport across the membrane and serve as a signaling mechanism, 10


11 communicating the oxidation state of the cytopl asmic trxs with the oxidation state on the periplasmic side. Disulfide formation involving trx motifs on either side of the membrane can cause disruption of the Dsbs electron tran sport abilities, affecting a host of cellular processes such as respiration and cytochrome syntheses. Although there is evidence that Dsb repair enzymes exist [9,10] their effectiveness and versatil ity against non-native thiols or sulfides has not been charted. So, these extraordinarily re active, native groups represent significant potential as drug targets. Especially of interest is that human thioredoxin has a greater distance between cy steine residues, Cys-X-X-X-X-Cys, which could allow for development of inactivators which are specific for the bacterial thioredoxins. [22]


Figure (1.11). Potential Drug Interactions with Thioredoxins. (a) Normal function of disulfide bridge forming enzymes (Dsbs), allowing electron flow between Dsbs and thioredoxins (trxs). (b) Drug molecules with thiol or disulfide functionalities can bind irreversibly with the Dsb, on either the cytoplasmic or periplasmic side, and thereby inhibit communication across the bacterial membrane. (c)As well, direct, irreversible, binding of an organosulfur molecule to free thioredoxins (trx) can also shut down these important redox mechanisms. 1.2.5 Mycothiol-based Systems Actinomycetes also do not produce glutathione, but generate mycothiol (13) instead as their primary antioxidant. [23] First discovered in a species of Streptomyces, and then identified in Mycobacterium bovis, mycothiol has since been found to be prevalent only amongst actinomycetes and is produced in high levels by mycobacteria. 12


Consequently, these cellular antioxidants have great potential as targets for anti-tuberculosis agents. O O HN HS NH O H O OH OH OH OH OH OH OH HO 13 Figure (1.12). Mycothiol (13). 1.2.6 Other Cellular Targets Another potential intracellular thiol target is adenosine triphosphate synthase (ATPase). ATPase is the enzyme responsible for regeneration of the all-important energy source of cells, adenosine triphosphate (ATP) (14). ATPase, which sits on the mitochondria, contains a reactive, free thiol which is paramount to activity. Inhibition of ATPase, therefore, can be achieved by formation of a mixed disulfide. One final target of note is the phosphoenol pyruvate phosphotransferase system which is involved in the sugar uptake of some bacteria. This system employs an enzyme whose structure is moderately conserved between species, carries a molecular weight of approximately 70,000, and is extremely vulnerable to attack by free thiols. [24] 13


O N OH OH N NN H2N O P O O OP O O OP -O O O14 Figure (1.13). ATP 1.3.0 Antibacterial Organosulfur Compounds 1.3.1 Introduction Bacterial cells differ greatly from human and animal cells, thus providing a number of targets that are specific to bacterial infection, but not healthy tissues. A major focus to antibacterial development has been on finding compounds that can interrupt bacterial cell wall biosynthesis without inducing harm to human cells. In the case of Gram-negative bacteria there is also an additional cell membrane on the exterior of the cell wall which can block the entry of some polar antibacterial agents. The remaining classical targets of antibiotics are tetrahydrofolate, DNA, RNA and protein synthesis. However, the search is on for novel bacterial targets that can damage bacteria through non-classical pathways. For thousands of years homeopathic medicine extracts, and oils from leeks such as garlic, onions, shallots, chives and scallions, have been successfully used to treat bacterial infections. [25] Early experimentation found that thoroughly dried formulations were totally ineffective, thus pointing to volatile reagents as the active anti-infective 14


15 agents. We now know that the responsible compounds in these remedies are low molecular weight organosulfur compounds. [24,25] Today, extracts are still used abundantly. In China, Dasuansu (diallyl trisulfide) is a popular commercial product for the treatmen t of bacterial, fungal and parasitic infections. Garlic-based remedies spawned the initial interest of medicinal sulfur chemistry and continue to be actively investigated. [26] In 1944 a rather smelly extract was determ ined to be a primary agent responsible for garlics wide range of antibacterial properties. [21] Three years later the first structural proof of a sulfur containing garlic extract wa s determined, by chemical synthesis, to be a thiosulfinate. This component was named Allicin ( 15) (originating from Alluim sativum the Latin name for garlic). Allicin has been shown to have potent in vitro antibiotic activity against Escherichia coli, Staphylococcus aureus, Streptococcus pyrogenes, Proteus mirabilis, Pseudomonas aeruginosa, Acinetobacter baumanii, and Klebsiella pneumonie (Table 1.1). [27,28] The likely mode of action of Allicin in volves reaction of the thiosulfinate with a cellular thiol to produce a mixed disulfide. [29] The identity of the native thiol has not been definitively proven, but is likely to be a thioredoxin or glutaredoxin.


S S O S S SH +15 Figure (1.14). Potential Mode of Action of Allicin. Table 1.1. Antibacterial Activities of Allicin(15). [28] Bacterial Strain LD50 (g/ml) Escherichia coli 15 E. coli (multidrug resistant) 15 Staphylococcus aureus 12 S. aureus (methicillin resistant) 12 Streptococcus pyogenes 3 Streptococcus hemolyticus >100 Proteus mirabilis 15 P. mirabilis (multidrug resistant) >30 Pseudomonas aeruginosa 15 P. aeruginosa (multidrug resistant) >100 Acinetobacter baumanii 15 Klebsiella pneumoniae 8 Enterococcus faecium >100 Although Allicin was the first sulfur-containing extract identified from garlic, it is certainly not the only one: other isolates with in vitro antibacterial activities include diallyl disulfide (16), diallyl trisulfide (17), Ajoene (18), diallyl sulfide (19), S-allylmercaptocysteine (20), and S-allyl-L-cysteine (21). [25,27-31] 16


S S S S S S S S S O S S NH2 COOH S COOH NH2 212016171819 Figure (1.15). Natural Antibiotics from Garlic. Many of these substances have also been tested for in vivo activities, including the diallyl sulfide and diallyl disulfide, which have potent activity against murine methicillin-resistant Staphylococcus aureus (MRSA) infections. [32] The mechanism by which they exert their bacteriostatic effects has not been fully defined but undoubtedly relies on the sulfur functionalities. 1.3.2 Thiols Thiol-bearing enzymes, such as glutaredoxin and thioredoxin, often lose some or all of their activity in the presence of non-native (to the cell) thiols like allyl mercaptan. To that effect, thiol-bearing therapeutics are often effective at shutting down enzymatic pathways regulated by these proteins. Although there are a large number of small, natural product thiols now known to possess antibiotic activities, there are only a few synthetically-derived thiols that have been shown to have antibacterial properties. A trio of thiodiazole aromatics, 22a-c, are reported to have minimum inhibitory concentrations 17


(MICs) against Staphylococcus aureus in the 31-62 g/ml range, but impose no effect on E. coli. [33] The reasons for this are not known. Compounds 23a-c, synthetic precursors to 22a-c, possess much lower antibiotic activities, with MICs above 125 g/ml for Staphylococcus aureus. NNS NN SH X 22a = 2-OCH322b = 4-Br 22c = 4-Cl Figure (1.16). S. aureus-Active Thiodiazoles. NNN SH NH2 X 23a = 2-OCH323b = 4-Br 23c = 4-Cl Figure (1.17). Mercaptotriazoles, Precursors to the Thiodiazoles. Some mercaptopyrimidines have also been shown to have antibiotic activities. [34] Biological evaluation of 5-mercapto-2-deoxyuridine (24b) in comparison to a known antibiotic, 5-fluoro analogue (24a), has been reported. Although weaker than the 5-fluoro analogue, the mercapto analogue shows selectivity to Lactobacillus leichmannii over Lactobacillus arabinosus and Streptococcus faecalis, which is not observed for the fluoro analogue (Table 1.2). As well, there is a synergistic effect when 5-mercapto-2-deoxyuridine is used in tandem with the fluoro analogue, indicating that the two compounds exert different modes of action in their inhibition of Lactobacillus leichmannii. The potency of the drug is displayed by an IC 50 of 0.06 M. Interestingly, for the 5-mercaptouracil, this selectivity seems to be reversed, with L. faecalis being the most greatly affected. Similar properties are seen with uracil analogues 25. 18


HOOHONONHO 24a X = F24b X = SHX Figure (1.18). 5-Fluoro-2-deoxyuridine (24a) and 5-Thio Analogue (24b). 25a X = F25b X = SHHNONHOX Figure (1.19). 5-Fluorouracil (25a) and 5-Thio Analogue (25b). OBO NS S 26 Figure (1.20). Example of a Boron-Complexed Benzothiazoline Antibacterial. Nisin Z is an antimicrobial peptide (lantibiotic), comprised of 34 amino acids with 5 free thiols, that effectively inhibits growth of various gram-positive bacteria. [36] It exerts its antimicrobial activity by permeabilizing the cytoplasmic membrane of target bacteria. This leads to the release of small cytoplasmic compounds, depolarization of the membrane potential, and ultimately cell death. The free thiols along the backbone of this peptide are required for the antibiotic properties, since blockage or removal of these thiols completely destroys activity. The reasons for this are not yet fully understood. 19


20 Table 1.2. IC 50 Values (M) of Fluoroan d Mercapto-Substituted Pyrimidines (24a,b and 25a,b). [34] Compound L.leichmannii L. arabinosus L. faecalis 5-fluorodeoxyuridine ( 24a) 0.03 0.001 0.00001 5-mercaptodeoxyuridine ( 24b ) 0.06 4 0.1 5-fluorouracil ( 25a) 0.04 0.2 0.00005 5-mercaptouracil ( 25b ) 30 700 3 As mentioned, certain groups of enzymes with in bacteria cells require free thiols for their activity, which can in some cases be inhibite d by non-native thiols. These proteins are especially susceptible to inac tivation by metal thiol complexes, such as boron-complexed benzothiazolines (26), which form mixed disulfides with thiolcontaining proteins to bl ock the metabolic pathway. [35] Compounds 26 have shown promising in vitro activities against Escherichia coli, Staphylococcus aureus, Klebsiella aerogenous and Pseudomonas cepacicola The levels of biocidal activity such complexes have against different microor ganisms depend primarily on the permeability of the cells. Bearing the idea that free thiols can be responsible for biological activity, it is important to realize the potential for these r eagents to dimerize and thus lower, or even abolish, their in vivo activity. Metal-thiol coordinations can prevent these dimerizations without destroying activity of the thiol. To prove this, one report compared the antib acterial effects of the sodium thiolate, free thiol, and tin thiolate of 5,7-dichloroquinone ( 27a-c). [37,38] Presumably, disulfide formation would be more likely for the sodium thiolate than free thiol, and more for the free thiol than for the tin th iolate. The three compounds were evaluated against four bacteria: Pseudomonas aeruginosa, Escherichia coli Staphylococcus aureus and Bacillus


cereus (Table 1.3). The sodium thiolate displayed no antibacterial activity, while the free thiol showed MICs in the 100 to 200 g/ml range and the tin thiolate had MICs in the 37.5 to 50 g/ml range. This suggests that spontaneous dimerization of thiol antibacterials may be a limiting effect which can be overcome by metal coordination. N Cl Cl S X 27a = Na27b = H27c = Sn Figure (1.21). 5,7-Dichloroquinoline Thiol and Metallated Derivatives. It seems that the tin entity is not responsible for this activity, but merely preserves the thiolate from disulfide formation. This has been demonstrated conclusively in the use of tin thiol derivatives for antifungal treatment (vide infra). Table 1.3. MICs of Quinoline Derivatives 27a-c. Compound S.aureus E.coli 27a >128 >128 27b 103 102.5 27c 50 37.5 Similarly, some simple bismuth-coordinated thiols (BTs) have been reported to possess potent antibacterial properties. [39] Bismuth-2,3-mercaptopropanol (BisBAL) (28), bismuth-3,4-dimercaptotoluene (BisTOL) (29), bismuth-1,2-ethanedithiol (BisEDT) (30), and bismuth-pyrithione (BisPYR) (31) have been shown to effectively inhibit the growth of MRSA and methicillin-resistant Staphylococcus epidermidis (MRSE), with MICs in the single digits and low MBCs (Table 1.4). 21


OH SH SH Bi CH3 SH SH Bi HO SH SH BiN SH OH Bi31282930 Figure (1.22). Bismuth-Coordinated Thiols. Table 1.4. Bioactivities of Bismuth-Coordinated Thiols Against Methicillin-resistant S. aureus. Compound MIC MBC BisBAL (28) 8.3 18 BisTOL (29) 4.4 9 BisEDT (30) 2.4 73 BisPyr (31) 6.7 14 As well, these bismuth thiols prevent growth of bacterial biofilms on coated indwelling medical devices, such as catheters and intravascular lines. Synthesis of BTs is straightforward, simply requiring the addition of bismuth nitrate to the appropriate thiol. [40] 1.3.3 Disulfides As therapeutic agents, disulfides usually serve one of two purposes: 1) as inactive structural components of biomolecules, or 2) as biological oxidants. In the latter role, disulfides are particularly prone to react with thiols to give biologically-inert mixed disulfide adducts. 22


23 Many of the leek extracts, like Ajoene ( 18 ), create the same type of mixed disulfide products with native thiols that are seen with the corresponding thiol drugs. [31] This leads to effective an tibacterial activities such as those shown in Table 1.5. Table 1.5. Antibacterial Acti vities of Ajoene (18). Bacterial Strain MIC (g/ml) Bacillus cereus 4 Bacillus subtilis 4 Staphylococcus aureus 16 Mycobacterium smegmatis 4 Mycobacterium pheli 14 Micrococcus luteus 136 Lactobacillus plantarum 19 Streptococcus spp. 56 Streptomyces griseus 4 Escherichia coli 116 Klebsiella pneumoniae 152 Pseudomonas aeruginosa >500 Xanthomonas maltophilia 118 Epipoly-thiodioxopiperazines (ETPs) are a class of fungal metabolites of Candida, Thermoascus and Penicillium, to name a few, that possess characteristic bridged disulfide piperazine dione six-membered rings. These antibiotics only inhibit Gram-negative bacteria, which is likely rela ted to their outer membrane permeability, and are prone to nucleophilic attack on their el ectrophilic disulfide bridge. Gliotoxin ( 32) and related ETPs are reported to act as oxidants by at least two different pathways: 1) generation of superoxide and hydrogen peroxide via glutathione redox cycling, and 2) sulfenylation of native thiols of certain pr oteins to make catalytically defunct mixed


disulfides. [41,42] Which sulfur center in gliotoxin is the site of enzymatic attack is still unclear but it is likely that both may be involved. N N O O CH3 S S OH H O H3C N N O O CH3 SH S OH H O H3C S Enzyme Enzyme-SH+ 32 Figure (1.23). Gliotoxin, an Epipolythiodioxopiperazine (ETP) Antibacterial. Although somewhat rare in nature, some cyclic polysulfides (33-35) occur in a few fungi and aquatic organic organisms. [43,44] The intermittent carbon linkages in these molecules distinguish these analogues from S 8 and enhance solubility in non-aqueous environments. SSSS SS SSS SSSS 33 34 35 Figure (1.24). Cyclic Polysulfide Natural Products. These cyclic systems contain reactive disulfide bridges that most likely behave like those of other disulfide antibiotics. These compounds have been found to inhibit Staphylococcus aureus, Streptococcus faecium, Escherichia coli, Klebsiella sp., Proteus mirabilis and Pseudomonas aeruginosa. 24


Similarly, the natural product Leinamycin (36) has potent inhibition activity, MIC = 0.03 g/ml, against Bacillus subtilus. [45] Leinamycin has an intriguing mechanism of action. It is believed that Leinamycin reacts with cellular thiols to form, in a few additional intramolecular steps, an episulfenium cation which can, in turn, alkylate DNA. [46-48] As a DNA alkylater, Leinamycin blocks cellular replication in bacteria. O SN NH O SS H H OH O O OH O SN NH O SSSR H H OH OO OH O SN NH O SO H H OH O OH O SN NH O H H OH S OH CO2 DNA-Nu: O SN NH O H H OH S OH CO2 N N HN O H2N DNA 36RS Figure (1.25) Leinamycins Mode of Action Isolated from both a species of Psammaplysilla and Thorectopsamma xana, a trio of closely related disulfides were discovered with noteworthy antibacterial activity. [49] Psammaplin A (37) and the dimer, Bisaprasin (38), along with Psammaplin D (39) all demonstrated growth inhibition of Staphylococcus aureus and Bacillus subtilis. Psammaplin D, interestingly, also shows activity against the Gram-negative bacterium Trichophyton mentagraphytes. [50] 25


Isolated from a tunicate, Polycitorella mariae, a novel disulfide termed Citorellamine (40) has been reported to have significant antimicrobial activity. [51] Originally assigned the structure of a sulfide, the disulfide Citorellamine demonstrates potent antibacterial activity against Staphylococcus aureus, Bacillus subtilis and Escherichia coli as well as cytotoxicity in some cancer cell lines. [52] Disulfides with interesting antibacterial activities also occur in proteins specifically synthesized by life forms for defense. These antibiotics are found most commonly in marine sources, and usually contain two neighboring cysteine residues. One such antibacterial protein, from a marine decapod, displayed potent inhibition of Planococcus citreus, Planococcus kocurii, Aerococcus viridans, and Micrococcus luteus and an extraordinarily strong resistance to heat damage. [53] HN S S NH HO Br N HO O N OH O Br OH 37 Figure (1.26). Psammaplin A (37). HN S S NH HO Br N HO O N OH O Br OH HN S S NH Br HO N HO O O N OH OH Br 38 Figure (27). Bisaprasin (38) 26


OH Br N NH O S S HN O O CH3 HO 39 Figure (1.28). Psammaplin D (39). NH NH S S Br HN NH Br 40 Figure (1.29). Citorellamine (40). 1.3.4 Trisulfides Trisulfides found in garlic, such as diallyl trisulfide and allyl methyl trisulfide, can act as antibacterial agents in the same way as disulfides, but with greater efficacy. The antibacterial activity of some cyclic polysulfides can be attributed to a trisulfide bridge, as much as previous examples owe their activity to a disulfide bridge. [43,44] 4-Dioxo-1,2,4,6-tetrathiepane (41), an extract from the red alga Chondria californica, has potent antibacterial activity against Vibrio anguillarium, the causative agent of a tropical fish disease. SSSS O O 41 Figure (1.30). 4-Dioxo-1,2,4,6-tetrathiepane. 27


Enantiomeric marine natural products are fairly uncommon, especially those with two chiral centers. However, a pair of Enantiomeric trisulfide alkaloids (42) from the New Zealand ascidian Hypsistozoa fasmeriana have been identified and shown to possess weak (4 mm zone of inhibition by a 120 g/disk), but identical, activities against Bacillus subtilis and weak (10% inhibition at 12.5g/ml) activities against Mycobacterium tuberculosis. [54] Similar Ascidian natural products without the trisulfide functionality (43), but with a 1,3-dithiane instead, are completely devoid of antimicrobial activity. SSS NHN OH O HO H3C 42SS HN HNN N O HO HO O OH HO CH3 43 Figure (1.31). Ascidian Trisulfide and Dithiane. Also part of the trisulfide family of antibiotics are the enediyne trisulfide antitumor antibiotics, Calichemicin (10), natural product of Micromonospora echinospora, and Namenamicin (11), Esperamicins natural product of Actinmadura verrucosospora, Although these antibiotics are not currently used to treat infection, in anticancer systems these compounds have been shown to undergo an intricate cascade of intramolecular reactions initiated by glutathione attack, resulting in an intensely reactive radical species. In the first step of this process, reaction of the trisulfide group with glutathione generates a free thiolate anion which in turn undergoes a Michael addition 28


across the ,-saturated ketone. This subtle change in hybridization of the carbons allows the enediyne to undergo Bergman cycloaromatization, producing the phenylene diradical. [13] These diradicals are believed to cleave DNA by sequentially stripping off hydrogen atoms along the minor grove of the double helix. Even though their mechanism of antibacterial action has not been proven to be the same as that of the anticancer mechanism (Figure 36), it is likely that the trisulfide moiety is involved as an electrophilic reactant with cellular thiols (Table 1.6). O SS S CH3 NH HO O O CH3 O Sugar GSH O HS NH HO O O CH3 O Sugar O NH HO O O CH3 O Sugar S O NH HO O O CH3 O Sugar S DNA damagingdiradical Figure (1.32). Proposed Mechanism of Enediyne Bioactivation. Table 1.6. Antibacterial Activity of Representative Enediynes, 10 and 11, versus Penicillin G. [11,12] MIC(g/ml) Bacteria Calicheamicin Namenamicin Penicillin G Bacillus subtilis 0.00005 0.03 0.25 Staphylococcus aureus 0.000001 0.001 0.015 Enterococcus faecium 0.00012 0.03 128 Escherichia coli 0.12 0.12 32 Klebsiella pneumoniae 0.25 0.06 128 29


1.3.5 Pentasulfides Within the Didemnidae or tunicate family, Lissoclinumi species are rich sources of organosulfur antibiotics. One such isolate Lissoclinotoxin A (44), demonstrated potent growth inhibition of a number of bacteria, including S. aureus, Streptococcus faecalis, Cirrobacter species, Klebsiella species, E. coli, Enterobacter species, Serratia species, Salmonella species, Pseudomonas aeruginosa, Acinetobacter, and Proteus species (Table 1.7). [55] SSSSS O H3C HO NH2 44 Figure (1.33). Lissoclinotoxin A Table 1.7. Antibacterial Activities of Lissoclinotoxin A (44). Bacteria MIC 44 (g/ml) Cefotaxim S. aureus 0.08 0.15 1.2 10 S. faecalis 0.3 0.6 0.6 2.5 Cirrobacter spp. 0.3 0.6 0.08 10 Klesiella spp. 0.3 0.6 0.01 0.15 E. coli 0.15 0.6 0.005 5 Enterobacter spp. 0.3 0.6 0.08 40 Serratia spp. 0.3 0.6 0.08 2.5 Salmonella spp. 0.3 0.6 0.08 3 P. aeruginosa 2.5 10 2.5 40 Acinetobacter 0.3 10 Proteus spp. 0.15 0.6 0.005 0.3 30


1.3.6 N-Thiolated Compounds N-Sulfenylated monocyclic -lactams (45) are another class of sulfur-containing antibacterial compounds recently discovered to have an unusual mode of action, where the N-S functionality may interact with cellular thiols in the same fashion as a disulfide. [56] Despite the presence of a -lactam ring, the mode of action of these N-thiolated compounds is totally different to that of the penicillins and other beta-lactam drugs which act as cell wall biosynthesis inhibitors. It is believed that an intracellular thiol attacks the sulfur atom to form a mixed disulfide which in turn causes inhibition of bacterial growth. These thiolated lactams show a narrow range of antibacterial properties, including Staphylococcus strains such as MRSA and S. epidermidis. Full details on the synthesis and biological activities is covered in the following chapters. N O O R2 S CH3 R1 45 Figure (1.34). N-Sulfenylated--Lactam Another sulfenamide, 1,2-benzoisothiazolin-3-one (BIT) (46), has shown weak activity against Staphylococcus aureus, with an MIC around 100 g/ml. [57] SNHO 46 Figure (1.35). BIT 31


BIT has been shown to inhibit the action of a number of intracellular thiols such as glutathione and ATPase. The mode of action has been linked to an inhibitory effect on cellular respiration upon metabolic uptake. Lansoprazole (47), a drug originally designed as a gastric acid pump inhibitor, has been shown to rearrange in the acidic environment of the stomach to a sulfenamide (48), which is an inhibitor of Helicobacter pylori. [58,59] H. pylori is considered to be a main culprit in the cause of gastric ulcers and therefore lansoprazole serves double duty as an antibiotic and an acid production reducer. Even though an intermediate sulfenic acid derivative of lansoprazole also displays anti-pylori activity, the sulfenamide affords fast action with an MIC value of 10 g/ml. NN S N CH3 F F F O H+ NN S N CH3 F F F 47 48 Figure (1.36). Lansoprazole and in vivo Sulfenamide. 1.4.0 Antifungal Organosulfur Compounds 1.4.1 Introduction In addition to their antibacterial properties many organosulfur compounds have fungistatic or fungicidal activity. Being eukaryotes, fungi comprise a separate group of 32


microorganisms, having a membrane bound nucleus, a more extensive endoplasmic reticulum and mitochondria. Compared to bacteria, fungi have a very different cell wall which is built from a complex set of constituents including chitin (aminoglucans), polyuronids, galactosamine, melanin and various lipids. Fungal DNA is found isolated within the nucleus instead of dispersed throughout the cytoplasm, as in bacteria. Fungal infections are usually isolated to the dermis or mucosal membranes (superficial mycosis). However, with the increasing prevalence of immune deficiency diseases, like HIV-AIDS, development of treatments for internal and systemic fungal infections (deep mycosis) are now of significant concern. Although antifungal activity attributed to garlic extracts dates back to 1936, the first garlic isolate to display antifungal activity, Ajoene (Fig. (18)) against Aspergillus, was not reported until 1987. [60] Since then a number of other antifungal activities have been found for Ajoene, such as Candida and Paracoccidiodes (Table 1.8). [31] Table 1.8. Antifungal Activities of Ajoene (18). Fungal Strain MIC (g/ml) Candida albicans 13 Hanseniaspora valbyensis 11 Pichia anomala 11 Schizosaccharomyces pombe 5.5 Saccharomyces cerevisiae 12 Other garlic-derived natural products that exhibit potent antifungal activities include diallyl trisulfide (against Cryptococcal meningitis) and Allicin (against species of 33


Candida, Cryptococcus, Trichophyton, Epidermophyton, and Microsporum, with MICs as low as 1.57 g/ml)(Table 1.9). [28,32] The mode of action of these antifungal agents has yet to be fully elucidated, however, they are believed to function as sulfenylating agents, as they do in bacteria. Table 1.9. Antifungal Activities of Allicin (15). Fun g al Strain MIC ( g /ml ) Candida albican s 0.3 C.albicans ( clinical isolate ) 0.8 C. neoforman s 0.3 C. p ara p silosi s 0.15 C. tro p icali s 0.3 C. krusei 0.3 Torulo p sis g labrata 0.3 T. g labrata ( clinical isolate ) 1.9 1.4.2 Thiols Due to the proliferation of thiol-bearing enzymes in a large majority of life forms, it is to be expected that anti-infective thiol compounds could be found in nature that can successfully inhibit the growth of fungi by formation of mixed disulfides. Consequently, some of the thiols examined earlier in this report for antibacterial activity are also effective antifungal reagents. Closely related derivatives (49a-c) of the antibacterial mercaptotriazoles inhibit Candida albicans and Saccharomyces cerevisiae, with MIC values in the range of 12.5 to 61 g/ml. [34] Both the thiol and amine groups are believed to be required for antifungal activity, since the thiodiazole aromatics 22a-c (which lack the amino moiety) and their precursors 23a-c have no antifungal properties. 34


NNN SH NH2 X 49a = 3-Br49b = 4-CH349c =4-OCH3 Figure (1.37). Mercaptotriazole Antifungals (49). The unsymmetrical boron-complexed benzothiazolines, BisBAL, BisTOL, BisEDT, BisPYR and BisME, whose antibacterial properties were previously discussed (Table 1.4), all display weak antifungal activities, with MICs just above 200 g/ml against Macrophomina phaseolina, Fusarium oxysporum and Aspergillus niger. [31] Comparatively, the marketed non-sulfur-containing drug, Bavistin, exhibits MIC values of approximately 100 g/ml each. Likewise, 5,7-dichloroquinoline-8-thiol and its sodium and tin thiolates (Figure 1.21) have been examined against fungi. [28] As in the case of the antibacterials, the stannous thiolate showed the best antifungal activities, with MICs against Candida albicans and Saccharomyces cerevisiae of 50 g/ml, while the free thiol and sodium thiolate displayed MICs of 128 and greater, respectively. To substantiate the claim that this is an antifungal effect of the thiol and not strictly a function of tin toxicity, another lab compared a relatively inactive control group of tin-2-thionaphthalenes to their free thiols and discovered almost equal antifungal activities. As well, the tin-thiolated compound was shown to be no more toxic (lethal dose = LD 50 >100 g/ml) than the free thiol itself. [29] 1.4.3 Sulfides Although the mechanism of action of the Ajoene organosulfide, 18, is still under active investigation, it is likely that antifungal activity is the result of induced cell wall 35


damage, since morphological changes in fungal cells treated with these compounds have been observed via scanning and transmission electron microscopy. [54] Six newly reported sulfide-bearing 1,4-naphthoquinones (50-55) have been found to display good to potent activities against Candida albicans, Cryptococcus neoformans, Sporothrix schenckii, Trichophyton mentagraphytes, Aspergillus fumigatus and Microsporum cannis, with MICs ranging from less than 0.78 to 50 g/ml, a marked improvement over their respective oxygen counterparts [60] As of yet, the mechanism of their activity is not understood, but may in fact be related to the oxidative potential of the naphthoquinone ring system. O O S 545153505255 O S CH3 O O O O S(CH2)2CO2H OH SN N Cl OH O O SN N O O SN N OH O O Figure (1.38) Naphthoquinone Sulfides Antifungal Compounds. 36


1.4.4 Disulfides Disulfides are commonly found as toxins produced by fungi [43] however, two groups of disulfides having antifungal activity have been reported. [62-64] The first group of compounds contains some of the most potent antifungal agents ever known, the thiarubines (56), with MICs in the ng/ml range. [62] Isolated as natural products from the Compositae (Asteraceae) family of plants, the thiarubines display activities against Cryptococcus neoformans, Aspergillus fumigatus, Candida albicans and other species of Candida. SS R' R 56 Figure (1.39). Core Structure of the Thiarubine Group of Antifungals. The tunicate-generated disulfide, Citorellamine, not only possesses strong antibacterial activity, but is also very active against Saccharomyces cereviae and mildly active towards Pseudomonas aeruginosa (Figure 1.31). [51,52] The other group of disulfides, the 1,2-dithiole-3-thiones (57a-j), is very interesting because it consists of compounds that could possibly behave as a disulfide, sulfide, or thiol in its activity. [63] The compounds are fungicidal against Candida albicans, Candida tropicalis, Cryptococcus neoformans, Sacharomyces cerevisiae, Aspergillus fumigatus, Aspergillus flavus, Aspergillus niger, Microsporum cannis, Microsporis gypseum, Epidermophyton floccosum, Trichophyton rubrum and Trichophyton mentagraphytes. The main activity is due to the disulfide-thione functionality. Although, the pendant sulfide (thioether) is not strictly required for activity [64] its presence, as compared to alkyl or aryl groups, 37


increases activity greatly. After lengthening the sulfur sidechain beyond ethyl, an inverse relationship between chain length and antifungal activity begins to develop. However, against all of the fungi tested the benzylthio analogue (57i) maintained the greatest potency, with MICs ranging from 0.7 to 6.25 g/ml (Table 1.10). SS S R S 57a R = CH2CH357b R = CH2CH2CH357c R = (CH2)3CH357d R = (CH2)5CH357e R = (CH2)9CH357f R = (CH2)11CH357g R = CH2C(CH3)357h R = Phenyl57i R = Benzyl57j R = Cyclopentyl Figure (1.40). Core Structure of the Dithiole-Thione Group of Antifungals. Table 1.10. Antifungal Activities of Thione Derivatives 57. Fungi 57a 57b 57c 57d 57e 57f 57g 57h 57i 57j C. albicans 50 25 125 250 >250 >250 >250 50 6.25 100 C. tropicalis 50 25 125 125 125 >250 >250 12.5 6.25 62.5 C. neofomans 62.5 12.5 50 62.5 125 125 250 6.25 6.25 50 S. cerevisiase 50 25 >250 >250 >250 >250 >250 12.5 6.25 >250 A. fumigatus 50 6.25 50 250 >250 >250 >250 25 6.25 125 A. flavus 125 25 125 >250 >250 >250 >250 25 6.25 250 A. niger 125 12.5 125 250 >250 >250 >250 25 3.12 250 M. Canis 6.25 6.25 6.25 250 >250 >250 >250 3.12 0.7 3.12 M. gypseum 12.5 12.5 12.5 250 >250 >250 >250 25 0.7 6.25 E. floccosum 12.5 12.5 12.5 50 >250 250 250 6.25 0.7 6.25 T. rubrum 12.5 12.5 12.5 50 >250 250 >250 6.25 0.7 6.25 T. mentagrop. 12.5 12.5 12.5 250 >250 250 >250 25 3.12 12.5 1.4.5 Trisulfides Although not currently used in clinical settings, enediyne antitumor antibiotics also display potent antifungal activities. Candida albicans, Ustilago maydis, 38


Sacchromyces cerisiae, and Neurospora crassa are all inhibited by Calichemicin and Namenamicin with MICs below 1 g/ml (Table 1.11). [65] Their mechanism of action is presumably similar to that of their antibacterial and anticancer properties as DNA cleaving agents triggered by S-thiolation of glutathione or a related thiophile. Table 1.11. Enediyne Antifungal Activities MIC (g/ml) Fungus Calicheamicin Namenamicin C. albicans 0.03 0.25 U. maydis 0.001 0.004 S. cerevisiae 0.008 0.06 N. crassa 0.06 0.25 Isolated from an ascidian, trans-5-hydroxy-4-(4-hydroxy-3-methoxyphenyl)-4-(2-imidazolyl)-1,2,3trisulfide, 42, also displays moderate antifungal activity against Candida albicans. [48] 1.4.6 Polysulfides Lissoclinotoxins A (44) (Figure 1.34) and D (58), pyridoacridine alkaloids from ascidians, both display potent antifungal activities against Candida albicans and Trichosporon mentagraphytes, with cell-mediated immunity levels (CMI) of 40 and 20 g/ml, respectively. More moderate activities (CMIs greater than 40 g/ml) were observed against several other fungi. [66] Pyridoacridines have been shown to intercalate DNA and in some cases inhibit topoisomerase II. [66] 39


SSSS O H3C HO NH2 O CH3 OH NH2 58 Figure (1.41). Lissoclinotoxin D. 1.5.0 Antiviral Organosulfur Compounds 1.5.1 Introduction For many years the use of antiviral agents in the clinical setting was a dream yet to be realized. Now there are a plethora of researchers studying viral pathways, discovering new drugs, and delineating novel modes of action. Viruses differ immensely from all other classes of infective agents. With no real cell to speak of, a virus consists of simply a shell or hard protein coating that encapsulates viral DNA or RNA. Classical modes of anti-viral action include the incorporation of false DNA building blocks, which leads to a blockage of replication, inhibition of the virally-induced DNA polymerase (which can be done with some selectivity in relation to the endogenous enzyme) inhibition of reverse transcriptase (a virus specific enzyme), inhibition of viral protein synthesis, and interference with the uncoating process, by which viruses release their genetic material. Within this context, organosulfur compounds may play an increasingly important role. A recent report shows conclusively that cytomegalovirus (CMV), a member of the herpes virus family which can induce mononucleosis-like 40


41 symptoms in immunocompromised patients, contains free thiol groups which are paramount to infectivity. [67] It has been found that if these groups are blocked as disulfides the virus is unable to infect, while deblocking the disulf ide back to the free thiol returns full infectivity. Potentially other viruses may contain similar thiols important to their ability to infect, and therein lies the possibility to design novel organosulfur compounds for use as antiviral agents. In the search for nove l nucleosides with AZT (3azido-3-deoxythymidine) like ac tivities, a variety of thionucleosides have been examined. Here, a sulfur atom occupies the site of the ribose ring oxygen, allowing the molecules to be introduced during reverse transcription into viral DNA. A good review of the synthesis and biological evaluation of antiviral thionucleosi de analogues has been published by Wnuk. [68] Although the antiviral prope rties of garlic natural products are not as highly appreciated as their antibacterial effects, Ajoene, Allicin, allyl methythiosulfinate, and methyl allylthiosulfinate (Figure 1.16 ) and Figure 1.18) have all been reported to have antiviral activity. [26,69,70] Specifically, these organosulfur compounds show detectable inhibitory activity against herpes simplex virus type 1, herpes simplex virus type 2, parainfluenza virus type 3, vaccinia viru s, vesicular stomatitis virus, and human rhinovirus type 2. [70] Ajoene and allyldisulfide are also active against human immunodeficiency virus (HIV), apparently by inhibiting integrin-d ependent processes. [26] Integrins are a family of cytokines that pr ovide costimulatory signals to T cells and protect them from abnormal cell death. In AIDS patients the action of integrins is


impaired by the viral consumption of these T cell protectors. It is interesting to note, however, that alliin and S-allyl cysteine have no antiviral activity. 1.5.2 Thiols Intracellular redox activity plays an integral part in regulating replication and infectivity of viruses. [71] The cellular thiol, glutathione, itself has purported in vitro and in vivo anti-influenza activity. As levels of glutathione are depleted in the oral, nasal and upper airways, susceptibility to viral infection is enhanced. Decreased intracellular glutathione levels are also implicated in HIV, and methods to increase glutathione production have been proposed as a means to stem these infections. [72] N-Acetyl-L-cysteine (NAC) and 2-mercaptoethylamine (MEA) have been shown to strongly increase glutathione levels in various cell lines. [73] Several new N-(N-acetylL-cysteinyl)-S-acetylcysteamine derivatives (59-61) have also been reported to actively release NAC and MEA, which in turn strongly bolsters glutathione levels. H3C HN NH S R S H O R1 O O O H3C HN NH S R HS H O O O H3C HN NH SH HS H O O 596061 Figure (1.42). S-Acetylcysteamines Derivatives These compounds display EC 90 (effective concentration for 90% inhibition of virus yields) ranging from 80 to 380 M against HIV in human monocyte-derived macrophages (MDM). Organosulfur compounds 60 and 61 with small chain thioester moieties are the most active analogues. S-Acylated derivatives are believed to have 42


increased activity by 1) having a protected thiol and 2) increasing lipophilicity relative to the free thiol. Another way reported to protect a free thiol is by use of a thiocyanate, which is likely reduced in vivo to the thiol through an equilibrium exchange with a native thiol such as glutathione. [74] Examples of this include a group of thiocyanatopyrimidine nucleosides (62), which are reported to display reasonable activity (EC 75 = 100 M) against vaccinia virus replication in HeLa cells. NN SCN O O H R NN SH O O H R R'SH 62 Figure (1.43). Thiocyanatopyrimidine Nucleoside (R = Ribofuranosyl Nucleoside.) 1.5.3 Disulfides As discussed in relation to antibacterials, organodisulfides can act as in vivo oxidants of thiols. It is likely that disulfides, and their related thiosulfonates, behave in the same manner in terms of their activity in viral systems. A group of aromatic disulfides and a thiosulfonate (63-68) have been reported with potent antiviral activities against the arenaviruses Junin (JUNV), agent of Argentine hemorrhagic fever, and Tacaribe (TCRV). [75] The disulfides and thiosulfonate displayed 50% effective concentration (EC 50 ) values, the concentration where 50% of the virus yield is eliminated, ranging from 3.6 to 100 M towards these microbes. This is at least ten times lower than the concentrations needed to induce cytotoxic effects. 43


S S S S S S N S S SS OH HO O O SN S S NHNCOCHPhCl2 Cl2PhHCOCNHN NHCH(NH2)NH HN(H2N)HCHN N O2N NO2 NS 636567646668 Figure (1.44). Antiarenavirus Disulfides. A large focus of preclinical and clinical development of anti-HIV drugs is in protease inhibition. However, other processes are certainly important to viral infectivity and replication. Metabolic pathways of infected cells, such as precursor protein processing, have been shown to be inhibited by a macrocyclic disulfide, 7-methyl-6,7,8,9-tetrahydrodibenzo[c,k][1,2,6,9]-dithiadiazacyclododecine-5,10-dione (69). [76] This compound displays an EC 50 of 0.05 g/ml against HIV-infected macrophages. Compared to the current standard AZT that has an EC 50 of 0.004. With a different mode of action from AZT however, the disulfide acts synergistically with AZT when tested in vitro, and could potentially be used in tandem. HN HN O O S S CH3 69 Figure (1.45). Macrocyclic Disulfide anti-HIV Agent. 44


Another disulfide with promising antiviral properties is 5,5-diphenyl-3,3-diisothiazole disulfide (DID) (70a). [77] DID induces potent inhibition of plaque-infected cells derived from invasion of poliovirus type 1, with an IC 50 of 0.35 M. Cytoxicity to healthy human cells was also examined, and no adverse effects were observed with uninfected cell cultures at 50 M concentration of DID. This agent is believed to inhibit an enzyme associated with RNA synthesis. 50% cytotoxic concentrations (CC 50 ), the concentrations where normal cell proliferation is inhibited by 50%, were more that 200 times higher that the IC 50 s, illustrating the exquisite selectivity this compound has for the viral infected cells. [77] The reduced form, thiol 70b, has almost the same activity and selectivity as 70a (Table 1.12). SN S S NS SN SH 70a70b Figure (1.46). 5,5-Diphenyl-3,3-diisothiazole Disulfide and Thiol Antipoliovirus Agents. Table 1.12. Isothiazole Anti-Poliovirus Activities. Compound IC 50 M CC 50 M Selectivity CC 50 /IC 50 70a 0.35 89.28 255 70b 0.42 90.75 216 45


HIV reverse transcription (HIV-RT) and deoxyribonucleoside triphosphate (dNTP) synthesis are paramount to viral replication, and thus are prime inhibition targets for anti-HIV therapy. [78] A number of sulfur bearing nucleotide HIV-RT inhibitors, which have similar effects to AZT, include 3-mercapto-2,3-dideoxynucleotides (71) and 2-deoxy-2-mercaptouridine-5-diphosphate (72). [79-81] These nucleosides serve to transfer a radical to a thiol of the transcriptase as shown in Figure 1.50. A new pyrimidine nucleoside disulfide (73) has been synthesized and shown to inhibit both HIV-RT and dNTP. The disulfide also has an EC 50 of 10 M and an IC 50 of 25 M against proliferation of human T-lymphocyte cells. Very interestingly, the corresponding thiol derivative had no activity at all. The mechanism of 73 is thought to involve release of the thiryl radical, RSS, as the primary active species (Figure 1.50). O OX HS P O O OP O P -O O O OOX = Adenine Cytosine Guanine Thymine71 Figure (1.47). Dideoxynucleotides Thiryl Radical-based Reverse Transcriptase Inhibitors. O ON HO P O O OP -O O OY HN O O Y = SH SSC3H772 Figure (1.48). Mercaptouridine Thiryl Radical-based Reverse Transcriptase Inhibitors. 46


O N HO SSCH3 NH H3C O O 73 Figure (1.49). Nucleosidic Disulfide anti-HIV Agent. O Base RO OH SH SH S. H O Base RO OH SH S S. H SH S O Base RO OH SH S SH S O Base RO O SH S SH S. O2 Figure (1.50). Proposed Mechanism of Sulfur-bearing Nucleotide-based Reverse Transcriptase Inhibition Gliotoxin (Figure 1.23), although a somewhat non-specific toxin, is a potent inhibitor of poliovirus. [41] Although Gliotoxin has a cytotoxicity which is too high to be a useful clinical agent, the compounds antiviral effects warrant further investigation in an effort to find an analogue with lower toxicity. 1.6.0 Antiparasitic Organosulfur Compounds 1.6.1 Introduction As with all three of the previously discussed types of infections, bacterial, fungal and viral, garlic extracts have been known the longest to exhibit antiparasitic properties. Generally, the most serious of these infections involve invasion of the intestinal tract or other internal organs. Since these microbes generally have low intracellular glutathione or 47


trypanothione concentrations they are proposed to have a higher sensitivity toward sulfur anthelmintics compared to mammalian cells. Allicin (Figure 1.16) has been tested against a number of protozoan parasites, Giardia lamblia, Leishmania major, Leptomonas colosoma and Crithidia fasciculata, and determined to have MIC 50 values equal to about 30 g/ml. Toxicity levels toward tissue-cultured mammalian cells of Allicin is above 100 M. [69] Allicin has also been reported to display potent inhibition of Entamoeba histolytica, with complete inhibition at 30 g/ml and an IC 90 of only 5 g/ml. [82, 83] Diallyl trisulfide, 17, which Allicin has been shown to degrade to in situ, also shows potent inhibition of Entamoeba histolytica with an IC 50 equal to 14 g/ml. [27] Diallyl trisulfide displays the same MICs against Giardia lamblia and Trypanosoma species (Table 1.13). [84] Table 1.13. Growth Inhibition (IC 50 ) Activities of Dasuansu (Diallyl Trisulfide (17)) Against Parasites. [26] Parasite IC 50 (g/ml) Trypanosoma brucei brucei 2.5 T. b. gambiense 1.8 T. b. rhosesiense 2.9 T. evansi 0.8 T. congolense 5.5 T. equiperdum 1.2 Entamoeba histolytica 59 Giardia lamblia 14 48


Ajoene (Figure 1.18) inhibits the proliferation of both epimastigotes and amastigotes of Trypanosoma cruzi, the causative agent of Chagas' disease. [84] Ajoene alters the composition of phospholipids, most likely through a sulfenylation that inhibits the phosphatidylcholine biosynthesis and cell proliferation. A 40 M concentration of Ajoene is all that is required to eradicate the parasite, in amastigote form, from host cells in 96 hours, while an 80 M concentration is enough to immediately inhibit growth in the epimastigote form. Ajoene is also cytocidal to epimastigotes at 100 M within 24 hours. As well, an entire host of garlic extracts, methyl propyl sulfide, allyl methylsulfide, diallyl sulfide, dimethyl sulfide, diallyl disulfide, dimethyl disulfide, methyl propyl disulfide, allyl mercaptan and dipropyl disulfide (some in Figure 1.16), have been tested separately against Giardia intestinalis, displaying IC 50 values ranging from 100 to 1300 g/ml (Table 1.14). [85] Table 1.14. Antiprotazoal Activity of Garlic Extracts. S IC50=250S IC50=550S IC50=1300S IC50=1300S S IC50=100S S IC50=200S S IC50=300 S S IC50=450SH IC50=37 49


1.6.2 Thiols A novel sesquiterpene named T-cadinthiol (74), also shows significant antiparasitic properties. [86,87] This terpenoid metabolite, with four fixed stereocenters, displays activity towards cultured Plasmodium falciparum, a species of malaria, with an IC 50 of 3.6 g/ml. The mode of action of this agent is still unexplored, however, and a number of derivatives of 74, including the corresponding alcohol analogue, were tested and shown to have absolutely no biological activity. The thiol group appears to be essential for anti-parasitic effects. H H H3C SH 74 Figure (1.51). Antiprotazoal Agent T-Cadinthiol. Until the availability of Cymelarsan (75) (melarsamine hydrochloride), treatment of African trypanosomiasis had relied heavily on antiviral agents which have a number of strong side effects. A successful attempt to increase the activity of this drug, by derivatizing the thiol groups to get the level of affinity between arsenic and sulfur atoms optimal for biological activity, has been reported. [88] The most active thiol system is the propane-1,3-dithiol (76) (2-melarsenyl). In fact, 2-merlarsenyl is twice as potent as Cymelasan against Trypanosoma brucei brucei strains (0.025 versus 0.05 M concentration to terminate all growth in 1 hour). It is believed that in aqueous solution Cymelasan is in equilibrium with the hydrolyzed oxide form (melarsen oxide), which has lost one thiol group and thus half the activity. [89] 50


NNN NH H2N H2N As S S NH2 NH2 NNN NH H2N H2N As SS 75 76 Figure (1.52). Typanocidals Cymelasan and 2-Melarsenyl. 1.6.3 Sulfides Fasciolosis (Fasciola hepatica) is a serious parasitic disease in humans and livestock. Few new anti-fasciolitic compounds have been marketed since triclabendazole (TCBZ), a benzimidazole used routinely in veterinary medicine since 1983 and for human use, in some regions, since 1989, was patented in 1978. [90] A new bioactive derivative of TCBZ, 5-chloro-2-methylthio-6-(1-naphthyloxy)-1H-benzimidazole (77) has been recently discovered. [91] While this analogue has an effective dose of 15 mg/kg with 100% effectivity, the marketed TCBZ displays a 5 to 10 mg/kg effective dose against Fasciola hepatica. The mechanism of action of this analogue has not been elucidated, and the role of the sulfide moiety is not known. O NHN Cl S CH3 77 Figure (1.53). A Methylthiobenzimidazole Fasciolicide 51


52 1.6.4 Pentasulfide Another anti-malarial ag ent, Lissoclinotoxin A ( 45), has demonstrated potent activity towards a resistant strain of the parasite Plasmodium falciparum. [55] Lissoclinotoxin A is intermediate in activity compared to the usual antimalarials quinine, mefloquine, halofantrine and chloroquine, with an IC 50 of 296 nM (Table 1.15). Although not yet defined conclusively, its mechanism is likely similar to that described for diand trisulfide antibacterials. Table 1.15. Anti-Malarial Activities of Lissoclinotoxin A and a Few Clinical Standards. Anti-Malaria Agent IC 50 (nM) Lissoclinotoxin A ( 45) 296 Quinine 350 Mefloquine 40 Halofantrine 2 Chloroquine 580 1.7.1 Conclusions & Future Prospects Garlic formulations have been used fo r thousands of years for the treatment of many types of infections. We now know that the attributable anti-infective effects are caused by the active organosulfur species within. The failure to commercialization these compounds most likely stems from their volatil ity and instability. However, other active sulfur reagents have been discovered, or synthesized, which have similar or better biological activities with good stabilities and ve ry low vapor pressures. Numerous sulfur compounds are yet to be evaluated for biological activity. The precise targets and modes


53 of action of many sulfur reagen ts are still unexamined. It has been proposed that the thioldisulfide redox metabolisms of infectious or ganisms might serve as a potentially valuable target for development of new anti-infectives. [11] For decades, no new targets have been discovered and brought to bear in clinical usage, even though these systems are essential for pathogenic growth and viab ility. The rapid procession of resistance to available antimicrobials, which target essentially three types of cell processes (cell wall synthesis, protein synthesis, and DNA synthesis), is depleting the current arsena l of antibiotics that remain effective. Often, these redox syst ems are extraordinar ily divergent from mammalian physiology and therefore provide ta rgets where selectivity should be very easy to come by. Stripping these infectious agen ts of their natural antioxidants will leave them wide open to the stresses of the extern al environment, or cause them to invest significant resources to develop resistan ce mechanisms for survival. Therefore compounds that recognize these systems should be excellent drug candidates. The genes for these systems, in a number of micr obes, have been cloned and sequenced. Imaginative, novel designs and development of new structural cla sses of anti-infective agents, with the intention of affecting these th iol-disulfide redox systems, is a rational but nearly unexplored avenue of drug discovery and is the fo cus of this dissertation. An introduction of classical antibiotics follows in Chap ter 2. The detailed organic synthetic methodology used to generate this new class of -lactams is provided in Chapter 3. An explanation of how N -thiolated -lactams have been found to be active organosulfur agents against bacteria, such as Staphylococcus and Bacillus, fungi such as


54 Candida, and even neoplasmic systems, such as those related to leukemia and cancerous tumors, then follows in Chapters 4 through 6.


CHAPTER TWO DEVELOPMENT OF ANTI-MRSA AND ANTI-ANTHRAX ANTIBIOTICS 2.1.1 Introduction Staphylococcus aureus, a species of bacteria which is often referred to as "staph", can live harmlessly on many skin surfaces, especially around portals such as the nose, mouth, genitals, and rectum. But when the skin is punctured or broken for any reason, S. aureus can enter the wound and cause an infection. S. aureus can cause folliculitis, boils, scalded skin syndrome, impetigo, toxic shock syndrome, cellulitis, and other types of infections. After the discovery of penicillin (78) by Alexander Fleming in 1928, the elucidation of a crystal structure by Hodgekins and Rodgers-Low, and the scale-up production processes of Flory and Chain, S. aureus infections became very treatable by penicillin, and eventually by a variety of penicillin analogues 78-82. N S O HN CH3 CH3 CO2H O N S O HN CH3 CH3 CO2H O O O H3C CH3 N S O HN CH3 CH3 CO2H O N S O HN CH3 CH3 CO2H O NH2 NH2 HO N S O HN CH3 CH3 CO2H O CO2H 78Penicillin G79Methicillin80Ampicillin81Amoxicillin82Carbenicillin Figure (2.1) Penicillin analogues. 55


During the 1960s and 70s a number of other -lactam antibiotic classes, with the ability to inhibit growth of S. aureus as well as other bacteria, were either discovered from natural sources or generated via synthetic manipulations (2-4,83-86). N S O CO2H RCONH N O CO2H N S O CO2H N O RCONH S CO2H X R R OH OH 23483N O O OH CO2H N O RCONH R HO N O SO3 R RCONH 848586penicillinscephalosporinspenamscarbapenamsclavulanic acidsmonobactamsnocardicins Figure (2.2) Common Classes of -Lactam Antibiotics. Unfortunately, the efficacy of these drugs has steadily declined since their initial conception, due to the ability of S. aureus to acquire resistance mechanisms. Methicillin, a standard lactam for measuring resistance, resistant Staphylococcus aureus (MRSA), is 56


a multi-drug resistant bacteria whose immunity to antibiotics was first discovered in 1961 (only one year after clinical introduction of methicillin). [93] The percentage of S. aureus strains with resistance to standard -lactam antibiotics has doubled over the past fourteen years, from 30% to 60% as seen in Figure 2.3. 01020304050607019891991199319951997199920012003YearPercent ResistanceProportion of S. aureus NosocomialInfections Resistant to Oxacillin (MRSA) Among Intensive Care Unit Patients,1989-2003**Source: NNIS System, data for 2003 are incomplete Figure (2.3) Increase of S. aureus Resistance, as Reported by the Centers for Disease Control (CDC). The classical method of MRSA infection is acquisition during unrelated treatments in a hospital setting. For example, a patient may visit a hospital for a puncture wound and become infected with MRSA. With low drug efficacies against these infections, it often becomes quite difficult to treat MRSA. Even other types of antibiotics, such as Vancomycin, considered the last line of defense against MRSA, are now exhibiting lower efficacy levels due to acquired resistance. 57


58 Recent headlines are providing additional urgency to develop new antibiotics for the treatment of resistant infections, as re ports have come out th at MRSA is no longer constrained to the hospital setting. This y ear a large number of the MRSA infections reported included infections acquired in the community. A group of novel drugs, whose mode of action are not affected by current resistance pathways, are therefore in high demand as this scourge is becoming untreatable. Bacillus anthracis (Anthrax) has also received much attention, in recent years, as a matter of national security. Currently the most effective antibiotic fo r this infection is Ciprofloxacin (Cipro). Generating more, equa lly potent, defenses against this potential terrorist weapon is certainly an appropriate goal in light of the evolution of drug resistance seen in other bacterial systems. This chapter will focus on the development of antibiotic agents for MRSA and Anthrax, a nd on investigations into their modes of action. 2.2.1 -Lactam Antibiotics: Anti-MRSA Penicillins and Cephalosporins have disputab ly been the most clinically important group of compounds discovered to date. Extensive investigations have led to a clear picture of biosynthesis and an tibacterial modes of action. The biosynthetic pathway of penicillin begins with the condensation of Laminoadipic acid (87), L-cysteine ( 88) and L-valine (89 ) via -(L-aminoadipyl)-Lcysteinyl-D-valine (ACV) synthetase to fo rm a tripeptide intermediate ( 89) with epimerization at the -carbon of the valine residue.


H2N OH O HS H2N OH O CH3 H3C H H HO OH O NH2 O H N SH N O HO O H NH2 O O OH ACVsynthase 87888990 Figure (2.4) Open ring Penicillin Biosynthetic Intermediate. Next, is the annulation of the bicyclic core skeleton to isopenicillin N (IPN) (91), by what else but IPN synthase, followed by epimerization of the -aminoadipoyl -carbon, by IPN epimerase, to provide penicillin N. N S N HO O NH2 O H H H O O OH penicillin NIPN synthaseIPN epimerase 9190 Figure (2.5) Annulation and Epimerization of Open ring Lactam. Need a cephalosporin? Ring expansion and hydroxylation of penicillin N, by deacetoxycephalosporin C/deacetyl cephalosporin C (DAOC/DAC) synthase(s), effectually produces deacetyl cephalosporin C (92) from penicillin N. 59


N S N HO O NH2 O H H H O O OH N N HO O NH2 O H H H O S O OH DAOC/DACsynthase(s) deacetyl cephalosporin C9291 Figure (2.6) Ring Expansion and Hydroxylation to Deacetyl Cephalosporin C. Penicillins are well known for there mode of action involving the coupling with penicillin binding proteins (PBPs). In this manner, penicillin antibiotics kill bacteria by disrupting cell wall synthesis and as effected cells attempt to expand they become deformed or even lysed. Simply speaking, penicillin binds to a serine residue of the transpeptidase enzyme (the primary enzyme of cell wall crosslinking), irreversibly, and inhibits that enzyme from being involved in cell wall crosslinking. Like a broken link in a chain, this results in a malformed peptidoglycan network and thus deformed or lysed cells. Without this strong crosslinked network, ordinary osmotic pressure within the cells can cause them to rupture. N S N -O O NH3+ O H H H O O ONH O O O H N S N -O O NH3+ O H H H O O ONH O O O H Figure (2.7) Penicillins Interaction with PBPs. Due to widespread over usage, -lactam antibiotics continue to succumb to developed resistance. Researchers have had great success synthesizing new lactam 60


derivatives to combat resistance. However, none of these is immune to the cell defense enzyme -lactamase, except the N-thiolated -lactams (93) (Chapters 3 through 6). N O O S Cl R' R 93 Figure 2.8 N-Thiolated -Lactam Antibiotics. Despite the continuous increase in resistance to multiple analogs of cephalosporins and penicillins, these drugs remain the most abundantly prescribed class of antibiotics. Primary resistance to these antibiotics is related to cleavage of the -lactam ring by a specialized bacterial defense enzyme that is released extracellularly, -lactamase. -Lactamase interacts with the lactam ring in much the same way as PBPs do, by breaking open the four membered ring. This ring opened form does not inhibit cell wall synthesis or disrupt any other cell function. So, in effect, the drug is deactivated by formation of the lactam-lactamase adduct, before it can reach the transpeptidase target. Except in the case of N-thiolated -lactams, this weakness is ubiquitous amongst -lactam antibiotics due to the extreme lability of this strained ring system. Again, the N-thiolated -lactams are unaffected by -lactamase and are discussed in greater detail in the following chapters. N S N -O O NH3+ O H H H O O ON S N -O O NH3+ O H H H O O OHOH -Lactamase 61 Figure (2.9) -Lactamase Ring Opening of Penicillin.


62 2.3.1 Vancomycin and the Peptide Antibiotics Vancomycin ( 94) is the most successful clinical treatment for MRSA infections thus far. Like the standard -lactam family, this antibiotic acts by interrupting cell wall synthesis, albeit in a diffe rent way. Part of a family of peptide antibiotics, the glycopeptides, Vancomycin is a potent an tibacterial agent which, unlike ordinary lactams, is impervious to -lactamase defenses. However, bacteria with specific defenses that can deactivate Vancomycin have develo ped over the course of time. The first of these was from a strain of enterococci, now termed Vancomycin resistant enterococci (VRE). The resistance has since jumped, via a plasmid, to S. aureus, alerting a need for new antibiotics to fight this cross-resistan ce. Synthetic modifications have successfully defeated VRE and Vancomycin resistant strains of S. aureus (VRSA), however development of strains with resistance to these analogues is bound to occur. Other peptide antibiotics, depsipetides and lipopeptid es, are reported to di splay potent activities against VRSA, but are, as of ye t, not clinically available.


O O O O NH HN O NH O O HN O NH O NH2 O NH2 OH OH HO H3C HO H3N CH3 OH OH OH Cl OH Cl O CH3 -O -O O 94 Figure (2.10) Vancomycin. O O O HN NH HN O NH O O HN O NH OH HO OH OH OH Cl Cl O O -O O O O O NH3 HO HO O OH OH OH HO O NH HO OH OH O 95 Figure (2.11) Teicoplanin, an Example of a Semi-synthetic Analogue of Vancomycin. 63


2.4.1 DNA and Protein Synthesis Inhibiting Antibiotics Although DNA synthesis inhibition or the cleavage of fully formed DNA are accepted targets of antibacterial exploitation, there currently are no clinically available analogues. One group of promising DNA synthesis inhibitors are the coumarins. These type II topoisomerase inhibitors have shown great selectivity in targeting DNA gyrase [96] X-ray crystallographic analysis of a 24-kDa N-terminal fragment of DNA gyrase with bound novobiocin (96), a typical coumarin, reveals that the antibiotic binds competitively at the ATP site and that the aminocoumarin bicyclic ring is the scaffold for presenting the L-noviosyl sugar moiety to interact with the gyrase. [97] O O HN O CH3 H3C O O OH O H2N CH3 OH O OH H3C 96 Figure (2.12) Novobiocin. Linezolid (97, trade name Zyvox), an oxazolidinone, is a fairly recent addition to the clinically available arsenal of antibiotics known to affect S. aureus. Most, oxazolidinones are protein synthesis inhibitors by irreversible binding to ribosomes. Their bacteriostatic activities are most pronounced amongst Gram-positive bacteria. 64


ON O HN H3C O N O F 97 Figure (2.13) Linezolid. 2.5.1 Anthrax Bacillus anthracis, Anthrax, was once a common problem amongst people and cattle. With the advent of antibiotics like penicillin, doxycycline, and amoxicillin this scourge was all but wiped out. Unfortunately, Anthrax has now become a weapon which is popular amongst terrorists. While, these antibiotics can be very effective against susceptible strains, a better defense has been developed, Ciprofilaxin (98). Ciprofilaxin retains potent activity against most strains of anthracis and even acts effectively as a prophylactic treatment for suspected exposures. However, resistant strains are emerging. A nalidixic acid-resistant strain of anthracis has already been discovered with a 10-fold lower susceptibility to Ciprofilaxin than other strains. [98] Therefore, it is equally important to find antibiotics with a novel mode of action which impedes induction of resistance amongst these bacteria. 65


N O OH O N F HN 98 Figure (2.14) Ciprofilaxin. 2.6.1 Conclusions For a number of years, pharmaceutical companies all but abandoned the search for new antibiotics, out of complacency. Now, as the advent of resistant bacteria is on a steady climb, researchers, industrial and academic, are coming back into this area of research. While new derivatives of currently available drugs seem to be a quick-fix, agents with entirely novel modes of action must be discovered to establish a more permanent solution to the ever-present problem of antibiotic resistance. 66


67 CHAPTER THREE N -ORGANOTHIOLATED -LACTAMS: SYNTHESIS 3.1.1 A Brief History Since Alexander Flemings di scovery of penicillin in 1928, -lactam antibiotics have developed into the most profound me dical discovery to date. Though, it was not until 1940 that Florey and Chain showed penicillins value as a potent systemic antibacterial agent. Fortunately, this discove ry came just in time to be of use during World War II, thus saving many lives. Althoug h bacterial resistance was already rearing its ugly head just a year after penicillins introduction, it was not until the 1960s that a multitude of -lactam analogues, to defeat these resistant microbes, were either discovered from natural products or developed synthetically, as discussed in Chapter 2. Most structurally related to the N -thiolated -lactam project is the first monocyclic class of -lactam antibiotics, the monobactams (dis cussed in chapter 2). Mechanistically however, these agents are drastically different from N -thiolated -lactams. The N thiolated -lactams are truly a novel class of an tibiotics, separate from penicillins, cephalosporins, carbapenams, penams a nd monobactams, not on ly by structural differences but also in their mechanisms of antibacterial activity (as will be shown in chapter 5). In earlier studies an N -sulfenylated -lactam ( 99) was isolated on synthetic route to the generation of novel isopenems ( 100). [99] Testing this intermediate against a panel of


microbes revealed potent inhibition of an array of bacteria, including MRSA. This sparked our laboratorys interest in studying N-thiolated monocyclic -lactams. N O S O H3C CH3 I2CH2Cl240oCN S O I O H3C 99100 Figure (3.1) Discovery of Novel N-Substituted Monocyclic -lactams. 3.2.0 Synthesis Towards an SAR 3.2.1 Introduction With five possible sites of substitution on the lactam ring, two at C 3 (R 2 & R 3 ), two at C 4 (R 4 & R 5 ) and one at N 1 (R 1 ), a large variety of variations could be explored as long as they are synthetically feasible (Figure 3.2). N O R1 R5 R2 R3 R4 C3C4 C2 Figure (3.2) -Lactam Substitutions Initial structure-activity studies focused on determining the importance of the C 4 substituent. [100] Although monocyclic -lactams have antimicrobial activity with their monocyclic form, certain amido functional groups are required for their activity. However, substitutions at either C 3 or C 4 on these N-thiolated monocyclic -lactams demonstrated no prominent effect on antimicrobial activity. [100] Activities were different 68


69 for various substitutions at these loci, but the absence or presence of saturation levels or functional groups did not turn activity on or off like a switch. So, for all intense purposes, variation of substituents at sites R 2 through R 5 does not radically a ffect antibacterial activity. For example, cha nging the substitution of R 2 from a methoxy group to an acetoxy group has almost no effect. As long as there is some electronegative group at R 2 or R 3 activity does not seem to change drasti cally. In fact, with no substituent at C 3 antibacterial activity still persists. As well, a variety of different halo-substituted aromatic functionalities R 5 affords very little flux in antib acterial activity. As long as an unsaturated group or electrone gative element exists at C 4 antibacterial activity changes very little. Lipophilicity can assist N -thiolated -lactams though, as demonstrated by better bioactivities for de rivatives having longer C 3 side chains. [100] However, total absence of this side chain results in a compound which stil l maintains potent biological activities. This led me to believe that activity had a greater dependence on some other structural motif than these side chains. Substitution at R 1 indeed, is paramount to activity (Table 3.1). Without an appropriate functiona l group at this position biological activity is non-existent. The focus of this chapte r will be the synthesis of varied N -substituted lactams for the SAR which follows in the next chapter.


70N O R1 R2 R3 R 1 R 2 R 3 Antibacterial Activity S-CH 3 O-CH 3 Alkyl +++ S-CH 3 O-CH 3 Aryl +++ S-CH 3 O-CH 3 Alkenyl +++ S-CH 3 O-CH 3 Alkynyl +++ S-CH 3 O-CH 3 Propargylic +++ S-CH 3 O-CH 3 Ester +++ S-CH 3 O-C(O)CH 3 Aryl +++ S-CH 3 O-Ph Aryl +++ S-CH 3 H Aryl +++ S-CH 3 S-Bn Alkynyl +++ S-CH 3 O-C(O)CH 3 Alkynyl +++ S-CH 3 Phthalimido Alkynyl +++ H O-CH 3 Alkynyl --Cl O-CH 3 Alkynyl --Se-Ph O-CH 3 Alkynyl --+++ = Active --= Inactive Table 3.1 Effect of -Lactam Substitutions on Antibacterial Activities. 3.2.2 Synthetic Strategy The synthesis of these N-thiolated -lactams, like that in Figure 3.3, is comprised of two major steps. First, the annulation of the -lactam ring from an acid chloride and imine. Secondly, N-thiolation (or sulfenylation) of the -lactam nitrogen. N O O SR Cl H3C N O O H Cl H3C N O O Cl H3C O CH3 Cl + Figure (3.3) Retrosynthetic Analysis Closer inspection of the standard synthetic steps towards N-thiolated -lactam syntheses reveals a few additional nuances (Figure 3.4). First is the generation of an appropriate acid chloride. Variation of the acid sidechain will directly translate to a


variation in the identity of the functional groups at C 3 Next, a simple synthesis of an imine is performed by the addition of para-anisidine with an aldehyde of choice. Here, as well, choice of the starting material translates directly into an -lactam substitution pattern, for whatever side groups are attached to the chosen imine will become the substituents at C 4 With these coupling agents in hand, the aforementioned annulation occurs by a Staudinger type coupling to generated an N-protected -lactam. Deprotection (or dearylation) is induced to generate the N-protio -lactam. Finally, N-thiolation occurs readily with an appropriate transfer reagent. Greater detail follows in the subsequent sections. N O O H Cl H3C N O O SR Cl H3C N O O Cl H3C O CH3 N O O Cl H3C O CH3 Cl +H2N Cl O CH3 H O + O O H3C OH SOCl2R3NCAN N O O S R N O O H Cl-S-RH-S-R N O O Br o r I m i n e S y n t h e s i s S t a u d i n g e r C y c l i z a t i o n D e a r y l a t i o n / D e p r o t e c t i o n P h t h a l i m i d e T r a n s f e r R e a g e n t S y n t h e s i s N T h i o l a t i o n A c i d C h l o r i d e Figure (3.4) Standard Synthetic Overview 71


3.2.2 Acid Chloride Synthesis Although some of the necessary acid chlorides were available for purchase, the low purity and moderately high expense made the synthesis of these reagents more attractive. The synthesis of these acid chlorides is simple and straight forward. Since the free acids were inexpensive and of relatively high purity upon purchase, they were used without further purification. The activating agent, thionyl chloride, was always racked with impurities such as sulfates and sulfoxides. A two step distillation process, with the use of linseed oil, proceeded to produce fairly pure thionyl chloride. This was then immediately used after purification. In most cases the resulting acid chloride required purification via one last fractional distillation. O O H3C Cl O O H3C OH SOCl2101102 Figure (3.5) Acid Chlorination 3.2.3 Imine Synthesis N-(2-Chlorobenzylidene)-4-methoxybenzenamine (105) was easily prepared via the condensation of 2-chlorobenzaldehyde (103) and 4-anisidine (104). Both reagents, however, required purification before use, even when newly purchased. The amine (105) was recrystallized from water and dried under reduced pressure. Aldehyde (103) was purified via atmospheric distillation. Originally the reaction was commenced by dissolution of 4-anisidine in freshly distilled methylene chloride, followed by pipette 72


addition of 2-chlorobenzaldehyde. It was later discovered that the presence of the solvent was completely unnecessary. Later reactions were performed by simple addition of (6) to (5). In the solvent free methodology a significant portion of released thermal energy could be observed. Camphorsulfonic was added in a few instances to trigger the condensation, however this was usually unnecessary. In either case, long, sharp, bright yellow crystals of product were formed along with an equivalent amount of water. A spot to spot conversion could also be followed by thin layer chromatography (TLC), but reaction completion was rarely of doubt. Proton nuclear magnetic resonance ( 1 HNMR) spectroscopy was used to confirm complete formation of the imine. O H Cl O NH2 CH3 103104 Cl N H O CH3 105 Figure (3.6) Imine Formation 3.2.4 Staudinger Coupling The Staudinger coupling is a formal [2+2] cycloaddition process of an acid chloride (102) and a Schiff base, imine (105). The Staudinger coupling is the most prevalent method of monocyclic -lactam formation and it is successfully used to prepare N-protected lactam (106). In this case, a ketene-mediated cycloaddition is believed to be on of the possible mechanisms of annulation, versus a direct acylation of the imine. Ketene formation is initiated via deprotonation by a Bronsted-Lowry base, triethylamine or diisopropylethylamine (Hunigs base). This solution is normally refluxed in freshly 73


74 distilled methylene chloride for 12 to 24 hours and closely monitored by TLC. The product obtained always demonstrated the cis relative stereochemistry, although this is not the case under modified condi tions. In the case of Figure 4.3 the imine used is always of the E -configuration which ultimately leads to the cis formation if the cyclization is truly [2+2]. It is possible, for the cases where a trans substituted lactam is formed via a Staudinger coupling, that either a Z -imine was used or a pote nt nucleophile can interrupt the cyclization and allow free rotation of the imine unsaturat ion. All of these steps are part of an equilibrium, therefore it is also possible that the lactam ring may reopen and interconvert between forms. Regardless of the mechanistic pathway, this cyclization always provided a racemic mixture of cis diastereomers. Exploration of the properties of each separate diastereomer is detailed in later sections.


N H O CH3 105C O R N O O R O Cl H3C 106O Cl N H O CH3 105O R N O O R O Cl H3C 106Cl Cl O N O O R O CH3 Cl Cl N O R O Cl H3C O Cl Ketene PathwayAcylation Pathway Figure (3.7) Possible Staudinger Coupling Mechanisms 3.2.5 Dearylation A procedure was reported in the literature which effectively deprotects lactam (106) by oxidative cleavage of the N-methoxyphenyl carbon-nitrogen bond. [87] In this case, ceric ammonium nitrate (CAN) is used as the oxidant in an aqueous acetonitrile 75


solution. CAN initiates a radical mechanism which results in a di-hydroxylated, pre-quinone form. Washing with a mild reagent, such as sodium bisulfite, drives the reaction to completion with generation of the protonated lactam (107) and one equivalent of benzoquinone (108). In small scale reactions, less than 2 grams, dearylation with three equivalents of CAN proceeded, at zero degrees Celsius, in moderate yields within 10 minutes. Larger scale reactions also proceeded within 10 minutes, however yields suffered significantly at these larger quantities. Once the reaction was determined to be complete, by TLC, the reaction mixture was poured into water, extracted with ethyl acetate, washed with sodium bisulfite and sodium bicarbonate, dried over magnesium sulfate and concentrated via rotary evaporation. Attempts to increase yields with extended reaction times, variations in temperatures or different stiochiometric ratios failed to provide any benefit. N O O R O Cl CH3 N O O H R Cl O O CANCH3CN-H2O106107108 Figure (3.8) Dearylation. 3.2.6 N-Substitution: Basis for the Phthalimide Transfer Reagent 76 A series of sulfenyl, sulfinyl, and sulfonyl analogues was synthesized using N-thiolating reagents and some direct modifications. Extremely important to the thiolation of the monocyclic -lactams is an organothio-transfer reagent, the N-thiolated phthalimide. More direct methods failed to succeed (Figure 3.10). What seemed like the


most direct method was to first generate the sulfenyl chloride which could then be attacked by the deprotonated -lactam nitrogen. However, this resulted in opening of the lactam ring, as crude 1 HNMR spectra show absolutely no lactam ring protons. The most likely culprit for opening the ring in this case is the ejected chloride anion. Once the sulfur is installed, it is very plausible that the chloride anion could act as a nucleophile, attacking the carbonyl of the ring and forcing the ring to cleave between the carbonyl-carbon and nitrogen bond, although this was never isolated. Regardless of conditions, this pathway yielded no product, yet consumed all of the N-protio lactam starting material. A gentler method was then attempted where the sulfenyl chloride was first reacted with deprotonated succinimide to generate the N-thiolated succinimide. This reagent transfers the organothiol group to the lactam with extremely poor yields, below usable limits. However, no lysis of the lactam ring occurred, making this pathway somewhat better than direct introduction using sulfenyl chlorides. N O O H H3C Cl N O O S H3C Cl R 108109 Cl-S-R MeO HN Cl O S R Cl ClFigure (3.9) Failed Direct N-Organothiolation Method. 77


N O O H H3C Cl N O O S H3C Cl R 108109 N O O S R 110 Figure (3.10) Failed Transfer N-Organothiolation Method. The best method for sulfenylating the lactam-nitrogen was via another transfer reagent 111 made from phthalimide. A number of organothiol groups could be transferred to the lactam with this type of reagent, whether by using a tertiary amine or carbonate base. N O O H H3C Cl N O O S H3C Cl R N O O S R 108111109 Et3NorDIPEA N O O H H3C Cl N O O S H3C Cl R N O O S R 108111109 K2CO3Sonication Figure (3.11) N-Organothiolation Methods. Synthetic pathways where a sulfur functionality is already installed on the imine prior to annulation may be plausible for some trityl systems, however this method was 78


completely ineffective here. [101] Although often problematic, thio-transfer seems to be the best method, as shown in Figure 3.11. Most thiols can be chlorinated in a straight forward manner by dissolution in a benzene solution which contains an equimolar quantity of dissolved chlorine gas. This sulfenyl chloride solution can then be added, in situ, to a preformed slurry of Hunigs base and phthalimide in ice cold benzene. As the reaction proceeds and the thiolated phthalimide 111 is formed, the mixture dissolves into a clear solution. Unreacted phthalimide 112 remains as a solid and can be easily filtered off. After aqueous workup with solutions of bicarbonate, bisulfite and saturated salt, the dried and concentrated solid can be easily purified via recrystallization from ethanol. For cases which did not proceed smoothly via this route, greater success could be found via reaction of the thiol with N-bromo phthalimide reagent 113. This method brought higher yields to reactions with longer chain thiols. N O O S R NH O O R-S-Cl DIPEA N O O Br R-S-HDIPEA or NaH 111112113 Figure (3.12) Phthalimide Synthesis. The yields for the synthesis of N-thiolated phthalimide reagents bearing large aryl substituents were increased by using potassium carbonate and sonication rather than Hunigs base or sodium hydride. Likewise, some N-thio phthalimides could be formed by first oxidizing the thiol to the thiosulfone via mCBPA, which is more easily attacked by phthalimide. Alkyl and aryl substituents were attainable through these routes, however, 79


great resistance was confronted with all attempts to install sulfurs of different oxidation states or hetero-groups on the thiol side chain. N O O S R NH O O R-S-H K2CO3Sonication N O O S R NH O O RS S R O O 112112111111BuLiTHF-78oC Figure (3.13) Alternative Phthalimide Synthesis Routes. Synthetically, the N-octylthiolated 109e lactam was extraordinarily challenging at the phthalimide transfer reagent synthesis step. More than 10 sets of conditions were attempted before any N-octylthiophthalimide was observed, and even in this one case the yield was extremely poor, as the equilibrium between octylthiol and 1,2-dioctyl disulfide heavily favored the disulfide. The yield was improved by mCPBA oxidation of the disulfide to the thiosulfonate, making the chlorination and addition to phthalimide easier. The N-butylthio lactam 109d was generated quite simply using N-bromophthalimide with n-butylthiol and triethylamine, where the other three analogues were synthesized in the highest yields from the chlorinated thiol via addition of chlorine gas to the thiols, followed by addition of phthalimide and a tertiary amine base. Also, transfer of organothio-groups from the phthalimide to the lactam was seriously hindered by steric bulk. Although phthalimide transfer reagent synthesis was achieved for highly hindered organothio systems, anything larger or more branched 80


proved difficult to transfer to the lactam by this method. sec-Butylthio and tert-butylthio groups were successfully installed, but with difficulty. Phthalimide reagents which were either unsuccessfully prepared, as with the branched alkyl chains, or were prepared but unsuccessfully used to make the corresponding -lactam derivatives, as with the heteroatom containing sidechains, are shown in Figure 3.14. N O O S OH N O O S O CH3 N O O S S CH3 N O O S C N N O O S Cl Cl Cl N O O S N N O O S N N O O S O N O O S NNNN N O O S CH3 O N O O S CH3 O N O O S CH3 O N O O S O N O O S O CH3 O N O O S O O CH3 N O O S CH3 O N O O S CH3 O O N O O S CH3 CH3 N O O S CH3 CH3 N O O S CH3 CH3 N O O S CH3 CH3 CH3 N O O S CH3 CH3 N O O S CH3 CH3 H3C N O O S CH3 H3C CH3 H3C CH3 Figure (3.14) Unsuccessful Phthalimide-based Transfers. 81


82 As described, most of the N -organothio phthalimides were prepared via chlorination of the corresponding thiol or disulf ide to generate the sulfenyl chloride in situ and then added dropwise into a ch illed flask containing phthalimide, and triethylamine or Hunigs base. After purification by flash chromatography the phthalimide reagents (111 ) were then added to an equimolar quantities of N -protio lactam (108) and an amine base, usually Hunigs ba se or triethylamine, and refluxed in dry methylene chloride or be nzene for 1 to 10 days. After aqueous workup with solutions of bicarbonate, bisulfite and saturated salt the resulting phthalimide was removed via trituration with chloroform. Residual or ganothiolated phthali mide was removed, painstakingly, via column chromatography. 3.3.1 Asymmetric Synthesis Of course not all of the desired thiols or disulfides needed for these studies were commercially available. For example, when this project was initiated enantiomerically pure thiols were unavailable and enantiomeri cally pure alcohol pr ecursors were cost prohibitive. So, the first method attempted to generate enantiomeri cally pure thiols involved enzymatic resolution of esters 114 via Lipase PS-30, a gene rous gift from the Amano enzyme company. In a slightly less than pH 7 buffered so lution, complete 50:50 resolution of racemic esters could be accomp lished in about a week. After filtration through Celite, the alcohol 115 and unaffected ester 116 were separated via column chromatography. Mitsunobu coupling of 115 and acetylated 116, however, to afford 117 and 118 did not proceed so smoothly. Steric hi ndrance was a serious problem for these secondary centers, forcing the crude yield to be quite low. Though a similar literature


procedure exists for the thioester formation starting from 2-nonanol, the shorter chain analogues proved to be much more difficult to purify due to their volatility. Reduction of the thioester to the free thiol (119 and 120) also proved problematic. While some conversion occurred, the volatility of the thiol product made purification impossible (Figure 3.16). HS O O Lipase PS-30BufferO O O H + 1) LAH, Et2O2) DIAD, Ph3P, Thiolacetic acidDIAD, Ph3P, Thiolacetic acidS O S O LAH, Et2OHS Really poor yields.Really poor yields &Extremely volatile114115116117118119120 Figure (3.15) Initial Enzymatic Scheme Towards Enantiomerically Pure Thiols. Next, a pathway was tried which involved a chirality inducing, nicotine based rearrangement process. [95] The first step of this process involved the hydride assisted attack of racemic 2-butanol 121 on carbon disulfide, which in turn activated attack on methyl iodide to form dithioate 122. Rearrangement of this sec-butyloxy-dithioate 122 to the sec-butylthio-dithioate 125 proceeded via a dioxy-(S)-(-)-nicotine reagent 124 (from oxidation of 123 via mCPBA). Unfortunately, the following reduction step suffered 83


similar problems of low yields. Purification was again made impossible by the extreme volatility of the product thiol 119. Extremely volatileHO NaH, CS2MeI, DMSOO S S S O S N N CH3 O O H Hexanes N N CH3 H mCPBAchloroformHS H2NCH2CH2OH (S)-(-)-Nicotine121122124123125119 Figure (3.16) Nicotine Based Scheme Towards Enantiomerically Pure Thiols. At this point, attempts to obtain enantiomerically pure sec-butylthiol were abandoned in favor of a less volatile system such as 2-phenylethylthiol. Beginning with the acetic ester 126, Lipase PS-30 was used again for enzymatic resolution. This worked very well. After filtration and concentration, purification simply involved the separation of the two layers, alcohol 127 and unaffected acetic ester 128. Unfortunately, this is as far as this pathway would proceed, for the subsequent Mitsunobu substitution on the alcohol would not take place at any appreciable level to generate enough product (129 and 130) for purification. 84


O O Lipase PS-30BufferO O O H + 1) LAH, Et2O2) DIAD, Ph3P, Thiolacetic acidDIAD, Ph3P, Thiolacetic acidS O S O No ProductObtained 126127128129130 Figure (3.17) Second Enzymatic Scheme Towards Enantiomerically Pure Thiols. The ability to generate the enantiomerically pure alcohol 127 suggested that it may provide a way to produce the enantiomerically pure thiol. This pathway was first tested with a racemic mixture of 2-phenylethanol (131), which is easier to handle than the more volatile, more soluble 2-butanol (Figure 3.18). Chloride substitution of the alcohol with thionyl chloride was straightforward and proceeded smoothly. This reaction was assumed to go through an S N 2 process to afford the stereochemically-inverted chloride 132. Thioesterification also proceeded smoothly by a second nucleophilic substitution with thiolacetic acid to generate thioester 133. Next, a one-pot, two-step procedure produced the N-thiolated phthalimide 134, without any free thiol intermediate. [84] The first step involves the sulfuryl chloride-induced oxidative cleavage of 133 to generate the sulfenyl chloride. In step two, this sulfenyl chloride solution is added to a preformed slurry of phthalimide and Hunigs base to generate the N-thiolated phthalimide 134. With 85

PAGE 100

this transfer reagent in hand, production of the N-(1-phenylethyl)thiolated -lactam 135 proceeded smoothly by addition of one equivalent of the N-protio -lactam 108 and three equivalents of triethylamine in a benzene reflux. Adduct 135 was obtained as an equimolar mixture of racemic diastereomers. O H SOCl2Cl Thiolacetic AcidEt3N, benzeneS O 1)SO2Cl2, CCl42)Phthalimide Hunigs' Base benzeneS N O O N O O S H3C Cl N O O H3C Cl H Et3Nbenzene131132133134135108 Figure (3.18) Successful Synthesis of an N-Thiolated -Lactam from an Alcohol. With the absence of a free thiol in this synthesis, this appeared to be an effective method for generation of the enantiomerically pure sec-butylthio -lactams. At about this time, the R and S sec-butanols became available at a reduced cost and were therefore purchased, instead of being independently synthesized. Alcohols in hand, each one was chlorinated, thioesterified, and eventually converted to the N-thio lactams via the route developed in Figure 3.19 for the R enantiomer (136-140). The process was repeated to generate the S enantiomer (141-145). 86

PAGE 101

O H SOCl2Cl Thiolacetic AcidEt3N, benzeneS O 1)SO2Cl2, CCl42)Phthalimide Hunigs' Base benzeneS N O O N O O S H3C Cl N O O H3C Cl H Et3Nbenzene108 136(R)137(S)138(R)139(R)140(R) Figure (3.19) Enantiospecific Synthesis of Enantiomerically Pure N-sec-Butylthio -Lactams from Optically-pure Alcohols. To prepare enantiomerically and diastereomerically pure lactams for more detailed bioassays it was necessary to couple an enantiomerically pure version of lactam 108 and 139. Fortunately, this proceeded smoothly utilizing the same enzymatic resolution scheme described above. [103] To provide a handle for resolution, the C 3 acetylated -lactam 146 was synthesized via the standard procedures previously described, simply starting with acetoxyacetyl chloride instead of methoxyacetyl chloride. As developed earlier through a collaboration with Dr. Bisht (University of South Florida), Lipase PS-30 efficiently hydrolyzed only one enantiomer 147 from the racemic mixture of acetoxy-lactams, leaving the other enantiomer 148 unaffected (with greater than 97% ees). These two products were then separated via column chromatography. The 87

PAGE 102

hydrolyzed enantiomer, 147, was re-acetylated by acetic anhydride in pyridine to give (+) 149. [104] N O O O O H3C N O HO O H3C N O O O O H3C +Lipase PS-30Buffer N O O O O H3C Ac2OPyr(+ )-cis-146147(-) 148(+) 149 Figure (3.20) Successful Synthesis of Enantiomerically Pure -Lactams. With both lactam enantiomers 148 and 149 prepared, what followed was the N-dearylation with ceric ammonium nitrate to produce enantiomerically pure N-protio -lactams 150 and 151, respectively. 88

PAGE 103

N O O O O H3C N O O O O H3C N O O H O N O O H O CANCH3CN / H2OCANCH3CN / H2O(-) 148(+) 149150151 Figure (3.21) Dearylation of Enantiomeric -Lactams 150 and 151. By applying both R and S sec-butylthio phthalimde transfer reagents 142 and 143 to each of these -lactam enantiomers 50 and 51, four separate stereoisomers were produced (152-155). Lactams 152 and 155 are enantiomeric with each other, as are compounds 153 and 154. These four compounds were tested independently for antibacterial activities. 89

PAGE 104

N O O H O N O O S O N O O S O N O O S O N O O S O 152153154155150 or 151 N O O S 142 or 143Hunig's BaseBenzeneReflux Figure (3.22) Synthesis of Lactams 152-155. 3.4.1 Oxidation For comparison of the N-sulfenyl lactams just described, more highly oxidized sulfur side chain analogues were also needed for biological screening. The structures of these compounds are shown below (Figures 3.25 and 3.26). Unfortunately, the preparation of these derivatives by thiolation of the N-protio lactam was made difficult by the inability to synthesize the necessary N-thio phthalimide transfer reagents, such as 156. N O O S R O N O O H R-S(O)-Cl3oAmine112156 Figure (3.23) Attempted Preparation of N-Sulfinyl Phthalimide Oxidation of the phthalimide sulfenamide was successful, but transfer of the sulfinyl group to the lactam did not work under the usual conditions (refluxing dichloromethane, Hunigs base and N-protio -lactam) (Figure 3.24). 90

PAGE 105

N O O S R N O O H H3C Cl N O O S H3C Cl R 108157 O 156 N O O S R mCPBAorH2O2111O Figure (3.24) Second Ineffectual Method of Sulfoxyl and Sulfonyl Transfer The N-sulfinyl and N-sulfonyl -lactams were instead prepared from the N-sulfenyl lactams via oxidation (Figure 3.25). There were two sets of oxidation conditions that were successful for this procedure, depending on the organothio substituent. One is by mCPBA in ether, and the other is by 30% hydrogen peroxide in acetic acid. In the case of the N-cyclohexylthio lactam 109i, only the peroxide method worked to give the N-sulfinyl product 157b, whereas mCPBA cleaved the sulfur-nitrogen bond. Re-subjection of this N-sulfinyl compound to the hydrogen peroxide conditions afforded N-sulfonyl product 158b in a nearly quantitative yield. This pathway also worked well for the generation of N-phenylsulfinyl 157a and N-phenylsulfonyl 158a -lactams from the N-phenylthio -lactam 109j. In the other case examined, where the sulfur sidechain is the sec-butyl in 109g, peroxide was completely ineffectual. Here, mCPBA allowed for the conversion of the sulfenamide to the N-sulfinyl lactam 157c; however further oxidation to 91

PAGE 106

the sulfonamide 158c with additional mCPBA could not be accomplished. Instead, cleavage of the nitrogen-sulfur bond was again observed. N O O S H3C Cl N O O S H3C Cl R R O N O O S H3C Cl R O O H2O2 / Acetic AcidormCPBA / etherH2O2 / Acetic Acid109157a R = Phenyl157b R = Cyclohexyl157c R = sec-Butyl158a R = Phenyl158b R = Cyclohexyl158c R = secButyl Figure (3.25) Oxidative Routes to N-Sulfinyl and N-Sulfonyl Lactams. N-Sulfonic acids and sulfonate salts were formed in a completely different manner than above. This turned out to be the only direct method of installing a sulfur sidechain on the lactam nitrogen. A sulfur trioxide / pyridine / DMF transfer solution was used [105] This pathway directly provides the potassium N-sulfonate salt which can be converted by ion exchange to the tetrabutylammonium salt 159. Eluting the tetrabutylammonium salt 159 down a standard silica gel chromatography column gave the sulfonic acid derivative 160 in good overall yield. N O O H H3C Cl N O O SO3H H3C Cl 1) SO3 DMF2) TBAB Column N O O SO3 H3C Cl NBut4 109159160 Figure (3.26) Route to N-Sulfonate and N-Sulfonic Acid Lactams. 92

PAGE 107

N O S H3CO Cl CH3 N O S H3CO Cl CH3 N O S H3CO Cl CH3 N O S H3CO Cl CH3 N O S H3CO Cl CH3 N O S H3CO Cl CH3 CH3 N O S H3CO Cl CH3 CH3 N O S H3CO Cl CH3 CH3 H3C N O S H3CO CH3 N O S H3CO CH3 CH3 CH3 N O S H3CO Cl N O S H3CO Cl N O S H3CO Cl N O S H3CO Cl CH3 N O S H3CO Cl N O S H3CO Cl N O S H3CO Cl O N O S H3CO Cl O O N O S H3CO Cl O N O S H3CO Cl O O N O S H3CO CH3 CH3 N O S H3CO CH3 CH3 N O S H3CO Cl CH3 CH3 O N O S H3CO Cl N O SO3H H3CO Cl 109a109b109c109d109e109f109g157c109h60109i109j109k109l109m157b157a152154109n158b158a153155165 Figure (3.27) Organothiolated -Lactams Synthesized. 3.5.1 An N-Resinthiolated -Lactam For the purpose of trying to identify the intracellular targets of these novel antibacterial agents, as to be discussed in chapter 5, a lactam-bound polymer resin 165 was synthesized. The hypothesis is that an intracellular target could be physically captured by the appropriate resin. The first of two methods examined in this study 93

PAGE 108

involved Merrifields resin (161) as a solid support on which the lactam could be attached through an N-thiolation process. This chloro benzyl material was subjected to thiolation via sodium hydrosulfide in water with cetyltriethylammonium bromide. Since this commercially available resin, and the expected product, is 2% crosslinked with divinylbenzene, it is completely insoluble in the available solvents, thus making NMR verification of the product difficult without a solid phase NMR probe. However, a standard chemical test for the presence of free thiols, sodium nitroprusside, demonstrated the total absence of the expected thiolation product (162). Infrared spectroscopy also indicated the absence of the S-H stretch that would be expected around 2500 cm -1 for 162. Potentially, 162 could dimerize to form a disulfide but no disulfide S-S stretching peaks were observed in the expected region around 500 cm -1 Cl SH NaSH 161162 Figure (3.28) Attempted Direct Thiolation of Merrifields Resin Fortunately, another, although less direct, method of resin thiolation was successful. With the insolubility of this resin, workups at each stage of the synthesis were simple and consisted of a set of filtrations with boiling solvent washings. The first step of the procedure was to thioesterify Merrifields resin (161) with thiolacetic acid and triethylamine, a method similar to that previously used for the synthesis of chiral thiolated lactams. [102] This provided thioester 163, which was confirmed by a single 94

PAGE 109

95 carbonyl stretch which was apparent in the in frared spectrum. Next, a two-step, one-pot procedure converted thioester 163 to the N -thiolated phthalimide transfer reagent 164 This analogue was also confirmed by infrared spectroscopy through the appearance of two carbonyl stretching peaks representing the phthalimide carbonyls. The first step involved the addition of sulfuryl chloride in carbon tetrachloride to the thioester 163 to generate the in situ the sulfenyl chloride. This solution was then added to a benzene solution of phthalimide and Hunigs base. Hunigs base proved to be the highest yielding choice of amine bases, over triethylamine and pyridine, for this deprotonation. With the synthesis of this N -resinthiolated phthalimi de transfer reagent 164 complete, the production of the N -resinthiolated -lactam 165 followed through the standard transfer conditions of mixing 164 with N -protio -lactam 107 in the presence of Hunigs base. The reaction was done in benzene and refl uxed for 24 hours. After completion, thorough washing of the resulting resin commenced with a myriad of solvents until no solutes were detectable by TLC or NMR. However, exposure of the resin to diisobutyl aluminum hydride (DIBAL) resulted in the release of a nearly quantitative yield of the N -protio lactam 107, which was observed in the 1 H NMR spectrum of the post-reaction washings. As well, infrared examination of resin 165 demonstrated only one carbonyl stretching indicating the absence of the phthalimide a nd providing evidence to formation of the resin-lactam adduct.

PAGE 110

Cl S Thiolacetic acidBenzeneEt3N0oC O S N 1)SO2Cl2 / CCl42) Phthalimide DIPEA O O S N O O CH3 Cl NH-lactamDIPEA, benzeneMerrifield's Peptide Resin(2% crosslinked with divinylbenzene) 161163165164 Figure (3.29) Successful Synthesis of an N-Resinthiolated -Lactam. 3.6.1 A Fluorescence System Another synthetic target, proposed for experiments to probe the basis for biological activity of these lactams, consisted of a Forster Resonance Energy Transfer (FRET) quenching pair, to be discussed in further detail in chapter 5. The model system involved the placement of N-naphthylthio and C 3 -dansyl groups onto the -lactam ring, where dansyl is a fluorescence quenching accepter for naphthalene. The first attempt to synthesize this compound, 77, involved deacetylation of -lactam 166 with potassium hydroxide in methanol to give alcohol 167, which was then deprotonated with sodium hydride and coupled with dansyl chloride. This product, 168, was isolated but could not 96

PAGE 111

be further derivatized to the N-naphthylthiolated -lactam. Instead, all attempts to install this thiol group with phthalimide transfer reagent 169 resulted in lactam ring cleavage. Therefore, another route was proposed in which the thiol moiety would be installed first, producing lactam 170. The dansyl substituent would then be introduced to produce FRET product 171. However, while the naphthylthiol substituent was installed smoothly, the addition of dansyl chloride resulted, again, in lactam ring cleavage. Thus, 171 could not be prepared via either route shown in Figure 3.30. N HO O H Cl S N O O N HO O Cl S N O O Cl S S O O N Dansyl ChlorideNaHN O O H Cl O KOHMethanol Et3N Dansyl ChlorideNaHN O O Cl H S O O N S N O O Et3N 16616769170168169171 Figure (3.30) Attempted Synthesis of a Model FRET System. Another FRET paired system was then targeted with the fluorescein analogue 172 and 4-N,N-dimethylaminoazobenzene-4sulfonyl (dabsyl), donor-acceptor pair. This first step of this process involved the esterification of this carboxylic acid and etherification of the alcohol with methyl iodide in an acetonitrile slurry of potassium carbonate, which proceeded but with a very small yield. 97

PAGE 112

O H3CO O OCH3 O Cl Cl 173 O HO O OH O Cl Cl 172 MeI, K2CO3MeOH Figure (3.31) Methylation of Fluorescein. From 173, two routes were attempted to yield the free thiol 175. The first route was meant to utilize the successful processes of past experiments by first generating the thioacetate 174, which would then be saponified to the thiol. However, thioesterification of the chloro precursor 173 completely failed. Secondly, a very direct method of arylthiolation through the use of sodium hydrosulfide in water, with the phase transfer catalyst cetyltriethylammonium bromide (CTAB), was carried out, but did not generate product. [106] Finally, a literature procedure for thiolating aryl halides via sodium thiophosphate dodecahydrate directly from 173 delivered about a 50% yield of 176. [107] Selectivity was not as high as hoped though, and a mixture of thiolation products 176a and 176b from substitution of either of the two chloride groups and 176c, from displacement of both chlorides, was obtained. This mixture appeared as one spot by thin layer chromatography and was therefore used as a mixture without attempts at separation. 98

PAGE 113

NaSHCTABH2O100oC O HO O OH O R1 R2 O H3CO O OCH3 O Cl Cl O H3CO O OCH3 O S Cl O O SH Et3NNa3SPO3H20 / MeOH O H3CO O OCH3 O HS Cl 173174175176a R1 = SH, R2 = Cl176b R1 = Cl, R2 = SH176c R1 = SH, R2 = SH O HO O OH O Cl Cl 172 Figure (3.32) Thiolation of Fluorescein. This mixture was then subjected to the standard procedures of chlorination, followed by reaction of the in situ generated sulfenyl chlorides with phthalimide, to give N-thio phthalimdes in a about a 40% yield of 177a-c. This mixture was then subjected directly to reaction with the N-protio -lactam 107 in the presence of triethylamine in refluxing benzene. Visible in the crude 1 H NMR spectrum were two doublets (J = 5.0 Hz) for the two -lactam protons. As well, these doublets were significantly shifted downfield from the doublet and triplet positions of the N-protio -lactam 107. Unfortunately, this product could not be purified. The only recoverable lactam following column 99

PAGE 114

chromatography was the starting material, 107. Without the ability to cleanly transfer the N-fluoresceinylthio group to the -lactam, there was no point in pursuing the C 3 dabsylated lactam. O HO O OH O R1 R2 176a R1 = SH, R2 = Cl176b R1 = Cl, R2 = SH176c R1 = SH, R2 = SH 1) Cl2, benzene2) Phthalimide Hunig's Base Benzene O HO O OH O R1 R2 177a R1 = S-Phthalimide, R2 = Cl177b R1 = Cl, R2 = S-Phthalimide177c R1 = S-Phthalimide, R2 = S-Phthalimide Figure (3.33) Fluoresceinylthiolation of the Phthalimide Reagent. O HO O OH O R1 R2 O HO O OH O R1 R2 177a R1 = S-Phthalimide, R2 = Cl177b R1 = Cl, R2 = S-Phthalimide177c R1 = S-Phthalimide, R2 = S-Phthalimide178 R1 = S-Lactam or Cl R2 = S-Lactam or ClN O O H H3C Cl Et3N Benzene Reflux107 Figure (3.34) Fluoresceinylthiolation of the -Lactam. Future work should probe the viability of a coumarin/dinitrophenyl FRET paired -lactam, which may encounter fewer synthetic problems. OO S N O O H3C O2N NO2 100 Figure (3.35) Coumarin / Dinitrophenyl FRET Paired -Lactam.

PAGE 115

101 CHAPTER FOUR N -ORGANOTHIOLATED -LACTAMS: STRUCTURE ACTIVITY RELATIONSHIPS 4.1.1 Introduction: Finding Struct ure Activity Relationships One of the original goal s of this project, in esse nce, was to understand the minimum structural requirements of N -thiolated -lactam antibiotics to sustain antimicrobial activity. In this chapter, the testing of compounds prepared in chapter 3, and the effects that sulfur si de chains have on activity of the lactams is presented. 4.1.2 Initial Antibacterial Screening The initial method for antib acterial screening was ex ecuted by the Kirby-Bauer disk diffusion method. In this procedur e, a cultured bacterium such as Staphylococcus aureus was streaked out (inoculated) on an agar filled Petri dish to encourage a homogeneous coat of growth across the surface of the plate. Next, 6 mm blank paper disks impregnated with 20 g of the chosen -lactam analogue (or control agent such as penicillin G or Vancomycin) were arrange d on the streaked plate, spaced with an appropriate distance as to not interfere with each other. These plates were then covered and incubated at 37 o C for 24 hours. The expected result for antibacterial agents was to see what is called a halo e ffect, or zone of growth i nhibition, around the impregnated disk. This halo is due to th e inhibition of bacterial growth elicited by the antibacterial agent. Demonstrated in Figure 4.1 is such an assay comparing the effects of penicillin G, (to the left), Vancomycin (at the top), an in active analogue (to the right) and a very active

PAGE 116

N-thiolated -lactam (at the bottom). The larger these halos, or zones of inhibition, the more potent the antibacterial activity. In this example, the bacterium was a methicillin-resistant strain of Staphylococcus aureus, MRSA. It is quite clear from this test that vancomycin is more potent than penicillin, but the N-thiolated -lactams are stronger still. Figure (4.1) Antibacterial Testing by Kirby-Bauer Disk Diffusion. 4.2.1 MRSA Activities As mentioned, substitution on the nitrogen center of the -lactam ring is key to activity. Early in this program the importance of a sulfur group directly bonded to nitrogen quickly became quite evident. N-Protio and N-para-methoxyphenyl lactams 179 and 180, as well as others having an N-phenylselenyl, N-chloro, and N-benzyl groups, were chosen as a first round of N-substituents (Figure 4.2). These analogues proved (179102

PAGE 117

183) to have no activity against a panel of bacterial species. However, an additional sulfenyl analogue, the N-phenylthiolated -lactam 184, demonstrated potent activity against a variety of bacteria. N O O H H3C N O O Cl H3C N O O Se H3C N O O S H3C InactiveInactiveInactiveActiveN O O H3C O CH3 InactiveN O O H3C Inactive179180181182183184 Figure (4.2) Different N-Substituted -Lactam Analogues Tested. Although activity persisted with variation in the identity of the sulfur sidechain from S-alkyl to S-aryl, changes in the oxidation state of the sulfur center was extraordinarily detrimental to activity (Figure 4.3). Thus, disulfide 185 and methylsulfonyl lactam 186 were devoid of activity. N-Benzylthio and N-methylthio lactams, 187 and 188, were about equally as potent. These early studies led us to believe that any increase in the oxidation state of the sulfur substituents on these N-thiolated 103

PAGE 118

lactams completely obliterates antibacterial activity. However, later results with different types of sulfur-substituted derivatives showed that some S-oxidized forms are tolerated. N O O S H3C InactiveN O O S CH3 N O O S H3C InactiveCH3 O O N O O S H3C Active N O O S H3C ActiveCH3 185186187188 Figure (4.3) Initial Variations in the Sulfur Sidechain Oxidation State. To further evaluate the influence of the N-thio moiety on antibacterial activity an SAR study of alkyl-thiolated -lactams was conducted, the synthesis of which is described in chapter 3. The N-methythio -lactam 109a was shown to be a potent inhibitor of MRSA strains, with an average zone of inhibition of about 28 mm (Table 4.1). Comparatively, penicillin G demonstrates a zone of less than 15 mm against the same strains. And Vancomycin (94), the most potent clinical antibiotic for MRSA, only ranks a zone of about 18 mm. The next step was to extend the chain of methylthio compound 109a by one methylene unit to the N-ethylthio -lactam 109b. This change raised the anti-MRSA activity to 31 mm. Further elongation of the straight chain, however, only demonstrated a steady decline in zone of inhibition sizes. As can be seen in Table 4.1, S-propyl109c, S-butyl109d, and S-octyl109e moieties had diminished activities compared to that of the N-ethylthiolated -lactam 109b. 104

PAGE 119

Table 4.1 N-Alkylthiolated Lactams Vs Standard Antibiotics 05101520253035MRSAMRSAMRSAMRSAMRSAMRSAMRSA NSHNOOCO2H n=0 n=1n=2n=3n=7 NOMeOS(CH2)nCH3Cl Black bars = MRSA strains, Gray bars = susceptible S. aureus Vancomycin Examining these bioactivity trends, the notion may arise that this diminishing effect on antibacterial activity as a function of S-alkyl chain length may simply be proportional to the analogues ability (or inability) to diffuse across the agar surface. Increasing S-alkyl chain length could certainly inhibit migration from the disk, leading to decreasing zone sizes as an artifact of the experiment. This exact problem arose upon the search for an SAR between different alkyl chain lengths at C 3 [87] In that case longer chain lengths also show decreased antibacterial activities via Kirby-Bauer screening. However, upon further investigation by methods 105

PAGE 120

which are independent of diffusability requirements, the trend in bioactivities was actually shown to be reversed to that of Kirby-Bauer testing. In other words, the longer C 3 sidechain analogues were actually more active than their shorter chain counterparts, and the trend in diminishing activities as measured by zone of inhibition testing was indeed an artifact of the method (Figure 4.4). Figure (4.4) Relationship of C 3 Side Length to Antibacterial Activities. [87] The first step taken to determine if a similar scenario is being created with the sulfur sidechain was to repeat the Kirby-Bauer disk diffusion experiments by a similar method, well diffusion. This methodology involves burrowing a 6 mm cylindrical well into the agar after streaking out the bacteria, and adding 20l of a 1g/l solution of the analogues into the wells. The plates were then incubated as usual for 24 hours. The reason for doing the agar diffusion assays this way was to see if there was a problem of 106

PAGE 121

107 the lactams not being able to diffuse off of the cellulose disk. By adding the lactam solution to the wells, there could be no interf erence due to physical adsorption onto the surface of the disks. These results, however, were essentially the same. Overall, the zone of inhibition values were slightly larger than with disk diffusion, but the relative trends in bioactivity were unchanged. Believing that the sudden drop in activity for the long chain lactam analogues could still be an artifact of agar plate te sting, a different type of experiment was performed to measure minimum inhibitory con centration, or MIC, values in broth media. MIC determination is an accepted means for assessing antimicrobial activity of compounds. This method is an aqueous phase technique, where bacteria, nutrients (like agar, called Mueller-Hinton br oth), and antibiotic candidates are all suspended in solution together. The minimum amount of compound required to complete ly inhibit visible bacterial growth (100% growth inhibition) is the MIC value for that compound. The lower the MIC is, the stronger is the drugs bioactivity. This method is without concerns of diffusion since the drug is dispersed homogeneously throughout the experimental vessel. Bacterial growth is examined optically for changes in opacity due to increasing cell counts, and is considered to be very accurate (done in triplicate). Indeed, these experiments further substantiated that the incr ease in sulfur chain length leads to a steady increase in MIC values, and thus to a drop in bioactivity. For example, the N -ethylthio lactam 109b has an MIC value against MRSA of only 8 g/ml, while N -octylthio lactam 109e has an MIC of 64 g/ml (Table 4.2). So, a shorter sulfur sidechain must either react better with the cel lular target or positively influence delivery to the target.

PAGE 122

The differences in reactivity between analogues of different bioactivity are explored more deeply in chapter 5. Along with the MIC testing, it is quite convenient to do another test to determine the minimum bactericidal concentration (MBC). MBC screening quickly determines if a compound is bacteriostatic or bactericidal. Classical -lactam antibiotics are bactericidal, however, of the N-thiolated -lactams which were tested for an MBC all were shown to be bacteriostatic. Each MIC sample that displayed 100% inhibition of bacterial growth was streaked out onto an agar plate, allowed to incubate for an additional 24 hours, and yet displayed a positive growth of bacteria. This MBC evidence again points to a mode of action for these compounds differing from that of penicillin. Table 4.2 N-Alkylthiolated Lactams Vs Standard Antibiotics 05101520253035MRSAMRSAMRSAMRSAMRSAMRSAMRSA NSHNOOCO2H n=0 n=1n=2n=3n=7 NOMeOS(CH2)nCH3ClMIC = 8g/mlMIC = 64g/mlMIC = 16g/ml Vancomycin Gray bars = MSSA, Black bars = MRSA strains 108

PAGE 123

109 Of course, linear alkyl chains are not the only possibility for the sulfur substituents. Investigating the effects of br anching with these chains, or aromaticity, should provide some evidence for, or against, a steric effect on bioa ctivity. The first study involves a simple variation of the ethylth io substituent with a longer, branched N isopropyl group 109f This change in structure led to an increase in the anti-MRSA activity, from 31 mm for ethyl to 33 mm (Tab le 4.3). Steric hind rance around sulfur seems to be ineffectual, or to even enhan ce growth inhibition. Further lengthening of the alkyl chain, to sec-butyl 109g, further increased potency to 40 mm. This is greater than twice the zone size of the same quantity (20 g) of vancomycin (94)! MIC testing concurred with the potency of this compound, yielding a value of less than 0.5 g/ml for 100% growth inhibiti on (Figure 4.5). A alternative way to compare the relative potencies of the N-sec-butylthiol analogue 109g to that of other compouns is by Kirby-Bauer te sting against MRSA, as shown in Fig. 4.5. This plate contains four disk s loaded with different test compounds. To the right of the plate is a disk containing 20 g of an inactive compound To the left is a disk impregnated with 20 g of penicillin G ( 78 ), which produces a very small zone of inhibition. At the top of the pl ate is another disk impregnate d with 20 g of vancomycin ( 94), a non-lactam antibiotic considered the last line of defense against MRSA. Lastly, at the bottom of the plat e is a disk containing N-sec-butylthio -lactam 109g. The differences in bioactivit ies are clearly visible.

PAGE 124

Figure (4.5) Comparison of N-sec-Butylthio -Lactam 109g to Clinical Standards. The next step was to further hinder the sulfur center of these lactams by synthesizing and testing the N-tert-butylthio analogue 109h. The synthesis of this compound was described in chapter 3, and turned out to be a difficult task. Fortunately, a small amount was synthesized for testing. Surprisingly, this exchange of N-sec-butyl to N-tert-butyl nearly obliterated anti-MRSA activity, with a zone size being reduced down to 10 mm. The reason for this appreciable drop in bioactivity in going from N-sec-butylthio to N-tert-butylthio is not entirely clear. Sterically, this change certainly creates greater bulk directly around the sulfur center, but electronically there is also a greater induction of electrons into to the sulfur atom. Both factors may decrease electrophilicity toward cellular thiophiles. However, if steric bulk alone was influencing the lability of 110

PAGE 125

the N-S bond, then one must wonder why the thiol transfer process occurred from phthalimide to lactam occurred so readily. Another series of N-thiolated lactams to examine was that of cyclic and aromatic systems. To this end, N-cyclohexyl, N-phenyl and N-benzyl -lactams 109i-k, synthesized as shown in chapter 3, proved valuable. Table 4.3 compares the bioactivity of these three compounds to the branched lactams 109f-h. All three of the cyclic variants showed lower activity that the sec-butyl and isopropyl analogues, and were roughly equal to each other in activity. Thus, the presence of cyclic side chains on sulfur diminished anti-MRSA activity to some degree. Table 4.3. Gray bars = MSSA, Black bars = MRSA strains 111

PAGE 126

Attempts to synthesize more complicated branched sidechain derivatives for further studies were unsuccessful, all examples of which are shown in Figure 3.29. In these cases, even the phthalimide transfer reagents were difficult to synthesize. This, of course, is likely due to major steric hindrances afforded by bulky constituents. The next objective was to study some N-thiolated lactams bearing heteroatom-containing side chains on sulfur. A series of these were attempted without success as shown in Figure 3.27. Attempts to prepare these compounds failed. In most cases the synthesis of the requisite phthalimide transfer reagent occurred without any problems, such as the case of a N-p-methoxyphenylthio analogue. However, transfer of the thio moiety onto the lactam never occurred in a useful yield after purification. Other methods of N-thiolation also failed. Direct N-thiolation with a sulfenyl chloride always resulted in either recovery of the starting N-protio lactam 107 or in cleavage of the lactam ring as shown in Figure 3.9. Installing the thiol groups onto the imine prior to the Staudinger cyclization, as suggested in the literature [101] also proved ineffective for the synthesis of these compounds (Figure 4.6). O O N O SCPh3 Cl 1)LDA / THF2) N SCPh3 Cl 189190191 Figure (4.6) Attempt to Prepare N-Tritylthio Lactams from N-Tritylthioimines. 112

PAGE 127

113 Next, compounds having additional hetero atoms (oxygen) directly on the sulfur center were examined. As shown in Figure 4.3, initial results demonstrated no activity for lactams bearing a sulfur at a greater oxidation state than th e sulfenyl systems discussed thus far. Further investigation with so me additional analogues, however, seemed warranted. Thus, the synthesis of N -cyclohexyl lactams 109i 157b and 158b and N sulfonic acid analogue 160 described in Chapter 3 enabled a comparison of their bioactivities. For these four compounds, the N -sulfenyl and N -sulfinyl lactams exhibited similar bioactivities, while the N -sulfonyl lactam was appreciably less active, and the N sulfonic acid was devoid of bioactivity. This di ffers from the previous observations that Soxidation destroys activity. Logi cally, the most active analogue, N sec-butylthiol lactam 109g should be the analogue to oxidize and test comparatively for antibacterial effects. However, oxidation of 109g with hydrogen peroxide (Chapt er 3) led to cleavage of the nitrogen-sulfur bond. Fortunately though, oxidation of 109g to N sec -butylsulfinyl 157c analogue could be achieved with slig htly less than one equivalent of meta chloroperoxybenzoic acid ( m CPBA). Further oxidation w ith another equivalent of m CPBA did not facilitate the synthesis N sec -butylsulfonyl analogue however. Instead, the lactam ring was deannulated and the same results occurred with the use of other oxidants under a variet y of conditions. Unlike N -cyclohexylsulfinyl analogue 157b this N sec-butylsulfinyl compound 157c retained only weak antibiotic activity as evidenced by a zone size ~ 25 percent the diameter of that produced by the N -sulfenyl lactam.

PAGE 128

Table 4.4 Black bars = MRSA strains, Gray bars = susceptible S. aureus Thus far, all of the -lactam antibacterial activities discussed have been prepared and tested as racemic mixtures. The bioactivity of many drugs depend heavily on stereochemistry. So, it seemed very important to examine these chiral N-thiolated -lactams more closely with respect to their relative and absolute stereochemistry. Even though no cytotoxic effects have been demonstrated for these antibiotics [113] perhaps one or more analogues might be more potent than the others. To examine these possible stereochemical effects the N-sec-butylthio -lactam 109g was chosen as the test candidate for its three chiral centers, and potency. Unremarkably, there was very little difference between each of the four stereoisomers 114

PAGE 129

152-155. However, what is curious and potentially meaningful is that cis up stereoisomers 152-153 have equal potencies, which are greater than that of their -lactam enantiomers cis down stereoisomers 154-155. Also of note is that variations in -lactam ring stereochemistry have no effect on bioactivity when the N-organothio substituent is not chiral, such as N-methylthio. [100] N H3CO O S 152N H3CO O S 155N H3CO O S 153N H3CO O S 154 Figure (4.7) Dependency of Anti-MRSA Activity on Lactam Stereochemistry. So it seems that the overall configuration of N-thiolated -lactams is important to bioactivity, perhaps for delivery to a target or to avoid aggregation of the drug, but stereochemistry of the most important substituent, the N-organothio group, is unimportant. 115

PAGE 130

4.3.1 Anthrax In addition to Staphylococcus species, the N-thiolated -lactams have potent antibacterial effects on Bacillus species. Initial experiments with Bacillus anthracis showed that the same SAR as that for MRSA seems to hold up for B. anthracis. The analogues with the most potent activities against MRSA, such as N-sec-butylthio -lactam 109g, are still these most potent against B. anthracis. So, it seems very likely that whatever is causing activity of the lactams against MRSA is the same for B. anthracis. This will be further explored and discussed in chapter 5. Preliminary studies of N-AlkylthiolatedLactams vsAnthrax NOMeOSCl42mm zone of inhibition againstBacillus anthracis Figure (4.8) Kirby-Bauer Screening of N-Thiolated -Lactams Against Anthrax. Table 4.5 shows a sampling of structures and respective activities of N-thiolated -lactams against Bacillus anthracis, the causative agent of anthrax. The standard clinical antibiotic of choice to treat anthrax infections, Ciprofloxacin, has a zone of inhibition of 116

PAGE 131

40 mm by agar disk diffusion. This is slightly less than that of the sec-butyl analogue 109g. Table 4.5 N-AlkylthiolatedLactams vsAnthrax 051015202530354045Zones of Inhibition (mm) NOMeOSClNOMeOSCl O O NOMeOSCl O NOMeOSClOH O O NOHOSCl NOMeOSCl Ciprofloxacin, the current drug of choice to treat Bacillus anthracis, demonstrates a zone of inhibition of 40mm. Anthrax was not the only Bacillus species tested, however. B. cereus, B. coagulans, B. globigii, B. megaterium, B. subtilis, and B. thuringenesis were also inhibited by N-sec-butylthiol -lactam 109g, with zones of comparable size to those produced by Ciprofloxacin (Table 4.6). As well, the same trends seen for anthracis hold true for the remainder of these six species of Bacilli. These shared trends seem to point to a common mode of antibacterial action of the N-thiolated -lactams in Staphylococcus and Bacillus, which will be explored further in chapter 5. 117

PAGE 132

Table 4.6 118

PAGE 133

119 CHAPTER FIVE N -ORGANOTHIOLATED -LACTAMS: MODE OF ACTION AND AN INTRACELLULAR TARGET 5.1.1 Introduction As discussed in chapters 1 and 2, classical -lactam antibiotics inhibit bacterial growth by interrupting cell wall crosslinking. Studies completed in this lab have shown that cell wall crosslinking inhibi tion is not the m ode of action of N -thiolated -lactams. [18] Evidence to this fact includes no changes in cellular morphologies, via examination by scanning electron microscopy of cells treated with N -thiolated -lactams, and no change in cell wall density as determined by Gram -staining. Drastic alterations in morphology and cell wall thickness occur for the same cells treated with penicillins. This means that for the first time a -lactam antibiotic has a mode of action not directly related to cell wall synthesis. The purpose of the work discus sed in this chapter was to find out where in the cell these drugs are going, what are th ey interacting with and how do these interactions produce inhibitive effects. 5.2.1 Chemical Interaction Since it seems apparent from the SAR studi es that the sulfur sidechain of these N thiolated -lactams is paramount to antibacterial activity, and that ot her evidence from our laboratories suggests that these lactams are impervious to radicals, there are really only three logical reaction mechanisms to c onsider in terms of the compounds biological mode of action. The first pathway is that some bio-nucleophile attacks the carbonyl carbon, C 2 of the lactam, thus opening the ring (Fi gure 5.1). This pathway is identical to

PAGE 134

the fate of all other -lactam antibiotics whose mode of action involves attack at that very same site by a serine residue of the transpeptidase enzyme. However, logic would dictate that if this was the site of attack, stronger electron-withdrawing groups on the lactam nitrogen would be expected to increase the electrophilicity, and thus reactivity, of the lactam ring. Thus, although increased reactivity does not always mean increased activity an N-sulfonyl lactam should be more active than an N-sulfinyl, which should, in turn, be more active than the N-sulfenyl derivatives. In actuality though, the precise reverse trend in bioactivities is seen. If this was not enough to rule out this pathway, it is important to point out that the N-protio -lactam can be isolated un-cleaved after treatment with bacterial culture media. NOMeOSRCl NuN-OMeOSRClNu Activity should decrease for N-substituent = SO3H > SO2R > SOR > SRReverse is observed SR > SOR > SO2R >SO3H Figure (5.1) Nucleophilic Attack at C 2 The second potential site of attack on the lactam by a biological nucleophile is the first carbon of the sulfur sidechain (Figure 5.2). If this site was the point of nucleophilic attack, then it would be expected that less hindered groups like methyl, as compared to ethyl or even isopropyl, on sulfur would be the most reactive. However, the exact reverse trend in bioactivity is seen. The isopropyl-substituted system was one of the most active analogues, certainly more active than the ethyl and methyl analogues. 120

PAGE 135

NOMeOSRCl Nu-NOMeOS-Cl + Nu-RActivity should decrease for R = CH3 > CH2CH3 > CH(CH3)2We observe the reverse R = CH3 < CH2CH3 < CH(CH3)2 Figure (5.2) Nucleophilic Attack at the Alpha-carbon of Sulfur Sidechain. The third, and final, site of attack to consider is the sulfur center itself (Figure 5.3), which seems very likely. First of all, the sulfur atom has proven to be absolutely required for antibacterial activity of these compounds and therefore attack at that site is possible. Secondly, attack at the sulfur center, and thus ejection of the lactam ring intact, fits well with the finding that the intact N-protio lactam is recovered cleanly from culture media. Shorter sulfur sidechains mean lower lipophilicity and therefore a better chance for the lactam molecule to pass completely through the cellular membrane into the cytoplasm to interact with some intracellular constituent. NOMeOSRCl NuN-OMeOCl + Nu-S-RMost Probable Mechanism Figure (5.3) Nucleophilic Attack at Sulfur. Also shown previously in this lab and in others (Shah/ Cama), is the extreme reactivity of these N-thiolated -lactams toward free thiols, such as 2-mercaptopyridine (Figure 5.4). [99] This shows a selectivity of thiophilic nucleophiles for this electrophilic 121

PAGE 136

site, and points to the likelihood of attack here by some intracellular thiol nucleophile. N-Thiolated -lactams are otherwise quite stable towards radicals, acidic conditions, and weakly basic conditions. NHOMeOCl +NOMeOSRCl N HS S S R Figure (5.4) Chemical Reactivity of N-Thiolated -Lactams to Thiophilic Reagents. So the questions that remain are: where does the biological action occur, what is the intracellular target, and how does this interaction confer antibacterial activity? Attempts to answer these questions are what follows in this chapter. 5.3.1 Loci The first experiments done in this lab by Dr. Timothy Long to determine where these drugs may end up in a bacterial cell involved radiolabeling. [108] For the lowest cost and smallest number of hot synthetic steps, the N-methylthio -lactam was synthesized with all three methyl hydrogens swapped for tritium labels. Unfortunately, the results from this experiment did not show where the drug went, only where it did not go. These lactams were not detected in any appreciable levels in cellular fractions contatining DNA, RNA, or proteins. At the time, low molecular weight targets were not considered and quite possibly the radio-tag was thrown out with the cellular bath water. Experiments 122

PAGE 137

123 were underway, at the time of this disserta tions completion, to us e radiolabeling to detect small molecule targets in aqueous fractions. A different method to follow the drugs pa thway into the cell was considered to measure fluorescence uptake. The first idea wa s to attach a fluorophore as part of the sulfur sidechain, to trace the path of this fluorophore through the bacterial cell membrane and into the cell. To see precisely the lo cation within the cell where the lactam may interact with its target, a Forster Resonance Energy Transfer (FRET) pair was built into the N -thiolated -lactam framework. The principle w ith FRET is that two functional moieties are installed in the framework wh ere the fluorescence wavelength of one, the donor, would overlap with the absorption wavele ngth of the other, th e acceptor. In this way the fluorescence of the donor would be que nched by the acceptor. Also of note is that this FRET phenomenon is greatly distan ce-dependent. Therefore, as the donor and acceptor are separated, the quenching effect e xponentially decreases and the donor lights up as its fluorescence is released. If a donor like a naphthyl group, is incorporated in the sulfur sidechain and the acceptor is attached to the lactam moiety, like a dansyl group, perhaps at C 3 or C 4 then once the sulfur-nitrogen bond is cleaved fluorescence would be emitted (Figure 5.5) as it moved away from the lactam. So the model system was a napthyl-dansyl FRET pair 171.

PAGE 138

Probing the Mechanism of Action via ForsterResonance Energy Transfer SO O N Donor AcceptorNaphthylDansyl N O O Cl S S O O N Potential Action of N-Thiolated-Lactams N O O Cl S S O O N Nu: S Nu Figure (5.5) FRET Pair Concept. In practice, however, there were significant problems. Unforeseen synthetic challenges thwarted efforts to synthesize this FRET pair 171 impossible. Even after 124

PAGE 139

125 changing the order of steps fo r introducing the two fluoresce nt side chains, lactam ring decomposition still occurred, preventing fo rmation achievement of the final FRET product. As well, this was only, at best, a model system as the donor fluorophore (naphthyl) is simply not red-shifted enough to be of use for cell structures. Naphthalenes fluorescence wavelength is fairly blue ( emission = 380 nm), as is the auto-fluorescence of bacteria. So, the small, faint blue light of this fluorophore would be difficult to observe amongst the large intense blue backgr ound of the cell. The compound and the background could, theoretically, be differentia ted based on fluorescence lifetimes, but no real-time imaging would be possible. Thus, another FRET candidate, with the appropriate fluorescence wavelength, was selected for synthesis, a fluor esceinyl / dabsyl paired compound 178. However, initial attempts tp prepare this compound were also unsuccessful fo r reasons noted above for 178. In the future, it may be possible to over come the difficulties of the synthesis, and perhaps use a similar procedure to make coumarin / dinitrophenyl FRET paired analogues. 5.4.1 Sugar Uptake Another sulfenamide, 1,2-benzoisothiazolin-3-one (BIT) ( 46 ), has shown weak activity against Staphylococcus aureus with an MIC around 100 g/ml [57] as discussed in Chapter 1. It has been sugge sted to interact with thiols of biological importance as part of its mode of action, such as glutthione. Particularly, however, th e molecule has been shown to inhibit the glucose uptake path way of staph. This pathway, called the phosphoenol pyruvate phosphotransferase system is responsible for the uptake and

PAGE 140

phosphorylation of sugars, such as glucose, for metabolism. Inhibition of this pathway starves the cell from nutrient absorption. Importantly, the important phosphotransferase is highly thiophilic and sharply inhibited by free thiols. Indeed, 46 is known to form an adduct with this enzyme an inhibit the glucose uptake of Staphylococcus aureus by 96%. It is thus evident that this molecule ring opens across the nitrogen-sulfur bond to reveal a free thiol for bioactivity. Since N-thiolated -lactams have been shown to interact with thiols there was reasonable suspicion that they may interact with this sugar uptake system in a similar pattern. Using a fluorescence based monitoring kit, glucose uptake of Staphylococcus aureus cultures were monitor with and without the presence of N-sec-butylthio -lactam 109g, as well as a series of positive an negative controls. As seen in Table 5.1, N-thiolated -lactams do not appear to inhibit this pathway, in light of the marginal decrease in sugar uptake. Apparently, these drugs are effecting a different system. Table 5.1 5.5.1 Identity of the Intracellular Target 126

PAGE 141

127 Since radiolabeling and fluorescence tagging methods has thus far failed to produce insightful data on the N -thiolated -lactams mode of action, another novel method of for identifying the in tracellular target(s) was inves tigated. In this experiment the goal was to synthesize a polymer resin onto whic h is covalently bonded an N thiolated -lactam through the sulfur moiety. In this way, it may be possible to use this lactam-conjugated resin to capture the cellular target. The plan was to expose the contents of bacterial cells to the lactam-conjugated resin, and upon reaction of the cellular target with the N -thio moiety, the nucleophile would then be covalently bonded to the resin as glutathione is shown to do in Figure 5.6. On ce the nucleophile is bonded to the resin, all impurities could be washed away. The nucleophile or intracellular target, could then be chemically cleaved from the resin, further purified if necessary, and characterized. This indeed proved to be possible. As described in chapter 3, the synthesis of N thiolated lactam resin 165 proceeded smoothly to generate a crosslinked N polystyrenylthio -lactam resin (Figure 5.6). To test this resins abi lities to capture thiophilic nucleophiles, it was fi rst subjected to glutathione 9, a common intracellular thiol, under pseudo-physiological conditions. The washings of this resin were examine by HPLC and 1 H NMR for organic compounds, and the N -protio -lactam 108 was thus observed. This experiment successf ully produced a resin-glutathione adduct 192 which could be chemically cleaved by a number of conditions, like triphenyl phosphine / water, or DIBAL / ether, 1 H NMR confirmed the recovery of the glutathione 9 after cleavage from the resin.

PAGE 142

S N O O CH3 Cl S OHNHOOSHNNH2OHOO Glutathione PPh3 / H2OOHNHOOHSHNNH2OHOO +N O O CH3 Cl H 1651081929 Figure (5.6) Testing of the Lactam-thiolated Resin. The resin was also tested in an aqueous solution containing equimolar amounts of 18 different amino acids including glutathione 9 and cysteine, all at 2 millimolar concentrations, with this DMSO swelled resin. After thorough washings, this adduct was then cleaved with DIBAL in ether to afford glutathione 9 only. Table 5.2 Amino Acids Stirred with N-Thiolated Resin L-cysteine glutathione L-lysine L-isoleucine L-methionine L-tyrosine L-asparagine L-glutamine L-proline L-histidine L-threonine L-tryptophan D-threonine L-valine L-alanine L-serine DL-phenylalanine DL--3,4-dihydroxyphenylalanine The real goal of these experiments was, of course, to see if the resin could extract targets from the intracellular contents of cells. This was completed by first cultivating a large number of Staphylococcus aureus cells, then concentrated them from their growth media before washing and lysing them via sonication. After centrifugation, the lysate was 128

PAGE 143

filtered to obtain only the soluble contents of the cells. These lysates were then shaken with a significant portion of the resin for 24 hours at 37 o C. The resin was then thoroughly washed and chemically cleaved with DIBAL to obtain the putative target species. One and only one compound was captured by this method, Coenzyme A (CoA) (12). A simple high pressure liquid chromatography (HPLC) experiment (traces shown as spectra 8.51-8.59 on pages 213-221) was performed to demonstrate the formation of an organothio-coenzyme A adduct 193. This was quite simple, once the proper eluent was discovered. There was an immediate generation of two new, less polar peaks which were representative of the adduct 193 and the resultant N-protio -lactam. NN NN O O P O O O P O O O HN HN S NH2 OH O O O OH P O O O S 193 Figure (5.7) Thiol-CoA Adduct 193 129 These experiments supported the claim that exposure of the S. aureus lysate to the N-thiolated -lactam would generate the same type of disulfide adduct. This was also proven to be so. A relationship between the rate of adduct formation and the antibacterial potency of various N-thiolated -lactams was explored. However, either there is no relationship between these properties, or the rate of thiol reaction between coenzyme A and the lactam is very fast. Different N-organothio -lactams, ethyl, phenyl and sec-butyl, were separately combined with coenzyme A all at 10 M in a solution of aqueous, pH 7 phosphate buffer and DMSO, and immediately injected into the HPLC. For each

PAGE 144

130 sample, it appears that the adducts is formed instantaneously under these experimental conditions, with no observable differences in the reactivity of the different N -thiolated lactams with coenzyme A. Several questions thus remain. Do different N -thiolated lactams react more readily with the target ? Is coenzyme A the only target? Does the organosulfur substituent server to enhance activity of the lactam by promoting its stability, reactivity, or delivery? Are the disu lfide adducts formed between the lactam and different thiols stable to reductases? Do the lactams of different bioactivities have different stabilities to non-target thiophiles? 5.6.1 CoA-Antibacterial Effects To counteract the effects of oxidative stre ss, cells have developed an important defense mechanism: a thiol / disulfide redox buffer consisting of small molecules and proteins with redox-active thiol moieties, disu lfides and disulfide re ductases (also called thiol-disulfide oxidoreductases) Through redox regulation of di fferent target proteins and small molecules, disulfide reductases control diverse cellular functions including apoptosis, cell proliferation, pr otein folding, oxidative stress and signal transduction, by constantly keeping a fresh s upply of free thiols. These antioxidants are particularly susceptible to attack by foreign thiophilic agents. At stasis, the reductases keep the equilibrium between thiol and disulfide at over 90% thiol. The most well known of these is the glutathione / glutathi one reductase system. Classically, glutathione was thought to be the ubiquitous thiol involve d in the thiol / disulfide re dox metabolism of all life and thus was the thiol of choice for the model experiments with the N -resinthiolated -lactam experiments.

PAGE 145

131 So, how does this tie in with Coenzyme A ( 9) ? CoA is a fairly ubiquitous chaperone enzyme used, by defi nition, to assist in enzymatic processes such as renaturing and folding. CoAs role in Staphylococcus aureus however, has recently been found to be much more profound. [109] The exciting part of th is discovery is that S. aureus has been found to neither produce nor utilize glutathione. Instead, S. aureus utilizes CoA, which it generates in millimolar quantities, and coenzyme A reductase (CoADR) as its primary thiol / disulfide re dox metabolism. Therefore, S. aureus is highly dependant on this system to prevent oxidat ion of important cellular syst ems, and any disruption, say by N -thiolated -lactams, of this metabolism could result in significant detrimental effects for this bacterium. Of equal significance, it is important to realize that CoADR is extraordinarily selective for CoA-CoA disulfides. [110] It has been shown that CoADR is incapable of reducing a mixed disulfide between CoA and gl utathione. It is th erefore likely that CoADR is incapable at reducing a mixed disulfide formed between CoA and the organothiolate procured from an N -thiolated -lactam. In this way, N -thiolated -lactams can possibly inhibit this redox cy cle and, eventually downstream, inhibit cell growth. It is also possible that the enzyme, CoADR, itself forms an adduct with the thiol from these lactams and is thereby irreversibly deactivate d. Since the enzyme only bears one thiol, on a cysteine residue (Cys43), which is needed for enzymatic activity [110] blockage of this site could quickly shut down the entire redox metabolism. There is additional evidence to support this mode of action. The bacteria that are affected by N -thiolated -lactams, such as multiple species of Staphylococcus and

PAGE 146

132 Bacillus each use CoA for their primary redox metabolism. Bacteria which are resistant to N -thiolated -lactams, such as Mycobacterium spp., Streptococcus spp., Enterococcus spp., and Escherichia coli, are know to utilize other thiols like mycothiol or glutathione for their redox cycling. The reductases for th ese systems are not as selective for their homo-disulfide and therefore may simply re duce any mixed disulfides that are formed. Human cells also use glutathione and this may be why, amongst other reasons, these compounds are non-cytotoxi c to healthy tissue. These lactams indeed react well with othe r thiols such as glutathione. This was shown in the case of the lactam-bound solid phase. As well, thin layer chromatography experiments show immediate formation of the Nprotio -lactam, when any organothiolated -lactam is mixed in solution with the free glutathione thiol at M concentrations, roughly equal to that known to be present in cytoplasmic fluid. Regardless of the identity of the organothiosubstituent, the reaction is instantaneous. So it seems that the trends in biol ogical activity can be based on neither the reactivity of the thiolate with the target, CoA, or the ability to not react with othe r antioxidants such as glutathione. Given glutathiones reac tivity toward different Nthiolated -lactams seems to be the equivalent, the question arises as to what effect glutathione levels in a cell correspond to sensitivity of those cells to N -thiolated lactams. Glutathi one demonstrates a greater ability to inhibit the potency of less active N -thiolated analogues as shown in Figure 5.7. As can be seen, three lactams were exposed to zone of inhibition testing, which resulted in the less active analogues, 109a and 109b having greatly dimini shed zones, while a

PAGE 147

more active analogue, like 109g, was less affected. So although, all tested N-thiolated analogues have a very high rate of reaction with free thiols, like glutathione, biological interaction is not a question of rate. Certain structural difference, the same ones that make an N-thiolated -lactam more bioactive, bestow a level of defense against free media thiols such as the glutathione distributed in these experiments. A B Picture A) Each well (small dark circles) was filled with 20 g of the shown drugs plus 20 g of glutathione. Picture B) Each well was filled with 20 g of the shown drugs plus 50 g of glutathione. In both cases, incubation was for 24 hours. Figure (5.8) Effect of Glutathione on Growth Inhibition by N-Thiolated -Lactams 133

PAGE 148

134 CHAPTER SIX BEYOND BACTERIA: OTHER BI OLOGICAL ACTIVITIES AND CONCLUSIONS 6.1.1 Fungi As mentioned in chapter 1, fungi, which are eukaryotes, comprise a separate group of microorganisms, having a memb rane bound nucleus, a more extensive endoplasmic reticulum and mitochondria. Comp ared to bacteria, fungi have a mostly chitin-based cell wall, instead of a peptidogly can like that in Gram-positive bacteria. This is why fungi are completely unaffected by traditional -lactam drugs like penicillin and other peptidoglycan ce ll wall crosslinking inhibitors lik e Vancomycin. As well, fungal DNA is found isolated within the nucleus in stead of dispersed throughout the cytoplasm as in bacteria. Fungal infections are usually constrained to only the dermis or mucosal membranes (superficial mycosis). However, ju st as resistant bacterial infections, like MRSA, are becoming a problem with the incr easing prevalence of immune deficiency diseases like HIV-AIDS, development of treatments for internal and systemic fungal infections (deep mycosis) are of great importance. As discussed in chapter 5, N -thiolated -lactams have a completely different mode of action from that of traditional -lactam antibiotics, which li kely involves the inhibition of a cellular redox metabolism. Fungi, are also known to rely on th iol / disulfide redox metabolisms to defend themselves from over-oxidation and theref ore are potentially susceptible to drugs that inhibit these systems.

PAGE 149

To determine if these N-thiolated -lactams can serve as antifungal agents, Dr. Timothy Long and Marci Culbreath tested N-methylthio lactam 109a against a panel of eight different species of Candida, including C. albicans, C. tropicalis, C. glabrata, C. kefyr, C. krusei, C. lusitaniae, C. parapsilosis and C. utilis. [111] The antifungal screening method closely mimicked the Kirby-Bauer method used for antibacterial screenings. Compared to a standard clinical antifungal agent, Clotrimazole, the N-thiolated -lactams, specifically 109a, faired quite well. For some species Clotrimazole was more potent, but in others the N-thiolated -lactam was more potent (Fig. (6.1)). Table 6.1 135

PAGE 150

136 This is certainly a departure from standard -lactam antibiotics. To verify this data MIC testing was also performed (Tab le 6.2). MIC values obtained after 24 hours were in the 10 to 15 g/ml range, with some even lower. Table 6.2 MICs of Lactam 109a Against Candida Species. Candida sp. MIC 24 hrs. MIC 48 hrs. C. albicans <5 g/ml <5 g/ml C. tropicalis 10-15 g/ml 30-35 g/ml C. glabrata 10-15 g/ml 10-15 g/ml C. kefyr 10-15 g/ml 35-40 g/ml C. lusitaniae 10-15 g/ml 15-20 g/ml C. parapsilosis <5 g/ml <5 g/ml C. utilis 10-15 g/ml 15-20 g/ml This verifies, yet again, that the mode of action of these N -thiolated -lactams is completely novel. The identities of the thiol / disulfide redox metabolisms of these fungi have thus far been assumed to be glutathione based. In light of these results, it would be very interesting to conclusively determin e the true redox metabolism of those fungi which are inhibited by N -thiolated -lactams versus those fungi which are not. If there is a strong line of difference betw een inhibited fungi thiols a nd uninhibited fungi thiols, a potentially important avenue to the design of new antifungal agents would be identified. 6.2.1 Neoplasmic Systems As remarkable as it is that these N -thiolated -lactam antibiotics are antifungal agents, it is perhaps even more intrigui ng that they are also anti-cancer agents. [111] This is the first time that a -lactam antibiotic has been show n to possess any anti-cancer properties. In leukemic Jurkat T-cells, N -thiolated -lactams have demonstrated the ability to induce DNA damage and inhibit DNA replication. Eventual ly, downstream this

PAGE 151

cascades to p38 mitogen-activated protein kinase activation, S-phase arrest, and apoptotic cell death. This apoptotic program was also induced in human leukemia, breast, prostate and head-and-neck cancer cell lines. Figure (6.1) N-Thiolated -Lactam Anti-Cancer Mode of Action. [110] Still unknown is how do N-thiolated -lactams induce DNA damage in tumor cells? Is there a direct interaction between these compounds and DNA? Our labs have 137

PAGE 152

138 shown that these compounds are non-cytotoxic at five times the concentration needed to induce DNA damage and inhibit DNA replication in leukemic Jurkat T cells within 2 hours [110] If the lactams damage DNA directly then why do they not affect the DNA of healthy human cells? What difference between healthy cells and cancer cells can these lactams differentiate between? A major differe nce between these two types of cells is intracellular glutathione concentration. It has been shown that tumor cells generate much higher concentrations of glutathione that do non-can cerous cells. Much like an infection, tumor cells are constantly fighting off oxidation, and by evidence of these increased glutathione levels, appear to be doing so more than non-cancerous cells. So, in essence, tumor cell redox metabolisms may be mo re susceptible to inhibition by N -thiolated lactams. As well, the influx roads into tumo r cells are known to be quite different from other types of cells. Much is yet to be l earned about these lactam s mode of action, however the exciting results of induced a poptosis (Figure 6.2) are not diminished.

PAGE 153

N-AlkylthiolatedLactams vsCancer Cell Lines Apoptotic Cells Figure (6.2) Apoptotic Effects of N-Thiolated -Lactams. [110] In cancer cell lines, there is also a pattern of activity seen amongst sulfur sidechain analogues that is similar to that previously described in Chapter 4 for anti-MRSA activity. In the case of these cancer cell lines, longer chain lactams have lower potencies, as shown with compounds 109a, 109b and 109d in Figure 6.3. (This is demonstrated by %Caspase 3 activity values of 100%, 40% and 30% respectively.) However, the more potent antibacterial lactams, like 109g, are yet to be tested against these cell lines, and lactams with an oxidized sulfur, like N-cyclohexylsulfinyl -lactam 109i, have recently shown decreased in vitro growth of breast cancer cell lines by more than 50%. 139

PAGE 154

Increased Chain Length NOMeOSClCH3 NOMeOSCl NOMeOSCl CH3 CH3 Decreased Anticancer Activities Figure (6.3) Trend of Anticancer Activity. 6.3 Conclusions and Future Directions The rapid procession of resistance to available antimicrobials is depleting the current arsenal of antibiotics that remain effective against infections. For decades, no new antibacterial targets have been discovered and brought to bear in clinical usage. Numerous sulfur compounds are yet to be evaluated for biological activity, and precise targets and modes of action of many sulfur reagents are still unexamined. It has been proposed that the thiol-disulfide redox metabolisms of infectious organisms might serve as a potentially valuable target for development of new anti-infectives. Often, these redox systems are extraordinarily divergent from healthy mammalian physiology and therefore provide targets where selectivity could be very high. Therefore, N-thiolated -lactams, with their novel mode of action, are excellent drug candidates, as shown here, for the treatment of cancer, as well as drug-resistant bacterial and fungal infections. Further analogue synthesis does not appear to be warranted, however, with the potential of a novel target across a number of systems, further biochemical and in vivo investigations are definitely worth exploring. It is the hope of this author that this work may contribute to the development of the thiol-disulfide redox as a target of rational drug design. 140

PAGE 155

141 CHAPTER SEVEN MATERIALS AND METHODS 7.1 Synthetic Procedures All chemicals required for the synthesis of N -thiolated -lactams were purchased from one of the following sources: Sigma Al drich, Fisher Scientific Acros Organics, TCI Organic Chemicals, or Lancaster Research Ch emicals. Most were used without further purification. Solvents were obtained from Fi sher Scientific. Products were purified by flash chromatography were done with either J.T.Baker or Whatman flash chromatography silica gels (40 m). NMR spectra were recorded in either CDCl 3 or D 2 O as indicated 7.1.2 Preparation of Imines N -Anisylimine (105): p -Anisidine, regardless of the s ource company, was always impure upon receipt and therefore required purificatio n by recrystallization from water. (1eq.) 11.20g (0.0797 moles) of o-chlorobenzaldehyde (103) and 11.00g (0.0894 moles) of recrystallized p-anisidine ( 104) were stirred together neat, at room temperature, open to the atmosphere, for a couple of minutes. Th e solid product was then dissolved in dry dichloromethane, dried over magnesium sulfate, and concentrated via rotary evaporation. The imine product was quite pure at this stage, but was further purified by recrystallization from metha nol prior to further use.

PAGE 156

O H Cl O NH2 CH3 103104 Cl N H O CH3 105 (E)-N-(2-Chlorobenzylidene)-4-methoxybenzenimine (105): yellow solid; mp 51-52 o C; 1 H NMR (250 MHz, CDCl, CDCl 3 ): 8.94 (s, 1H), 7.4-6.9 (m, 8H), 3.85 (s, 3H). 7.1.3 Preparation of Acid Chlorides Methoxyacetyl chloride (102): Although this acid chloride was available for purchase, the price was moderately high and the purity was very poor. Thus, it was preferable to synthesized it fresh. Purification of the thionyl is achieved through distillations with quinoline and then linseed oil as directed in the literature procedure: Rigby, Chem. Ind., 1969, 1508. To a dry round bottom flask 30.00g (0.335 moles) of methoxyacetic acid (3) was added, followed by dropwise addition of 39.58g (0.335 moles) of thionyl chloride at 0 o C. The mixture was allowed to stir and warm to room temperature overnight. The solution was then distilled, open to the atmosphere from 111 to 114 o C (literature value = 112-113 o C), to obtain pure product. O O H3C Cl O O H3C OH SOCl2101102 142

PAGE 157

7.1.4 Preparation of N-4-Anisyl Azetidin-2-ones (-Lactams) N-4-Anisyl azetidin-2-ones (106): In a round bottom flask, 6.00g (0.0245 moles) of N-Anisylimine 7 and 7.5g (0.0742 moles) of triethyl amine were dissolved in minimal amount of dry dichloromethane and cooled to 0 o C. A solution of 2.81g (0.0260 moles) of methoxyacetyl chloride in an equal volume of dry dichloromethane was added dropwise with stirring. The solution was then heated to reflux, which was maintained overnight. After cooling, the solution was poured into an equal volume of water, extracted thrice with the dichloromethane, dried over magnesium sulfate and concentrated via rotary evaporation. Column chromatography was used for purification, eluting with dichloromethane or a mixture of hexanes and ethyl acetate. Cl N H O CH3 105 Cl O O R 102N O O R O Cl H3C 106 ()-(3S,4R)-4-(2-chlorophenyl)-3-methoxy-1-(4-methoxyphenyl)azetidin-2-one (106): white solid; mp 183-184 C 1 H NMR (250 MHz, CDCl3): 7.43 (d, 1H, J = 7.4 Hz), 7.29-7.19 (m, 5H), 6.80 (d, 2H, J = 9.0 Hz), 5.61 (d, 1H, J = 4.8 Hz), 4.89 (d, 1H, J = 4.8 Hz), 3.73 (s, 3H), 3.27 (s, 3H). 7.1.5 Dearylation of N-Anisyl Azetidin-2-ones N-Protio Azetidin-2-ones (107a): In a round bottom flask 1.00g (0.0035 moles) of N-anisyl azetidin-2-one 106 was dissolved in a minimal amount of acetonitrile at 0 o C. A solution of 5.77g (0.01053 moles) of ceric ammonium nitrate, dissolved in a minimal 143

PAGE 158

amount of water, was added dropwise via addition funnel. The reaction was monitored via thin layer chromatography (TLC) and was usually complete within 10 minutes. The reaction mixture was then poured into three volume equivalents of water and extracted thrice with ethyl acetate. The organic extracts were washed sequentially with a 5% sodium bicarbonate solution, a 1% sodium bisulfate solution, and a saturated sodium chloride solution, then dried over magnesium sulfate and concentrated via rotary evaporation. The -lactam product was purified via column chromatography with a mixture of ethyl acetate and hexanes. N O O R O Cl CH3 N O O H R Cl O O CANCH3CN-H2O106107108 ()-(3S,4R)-4-(2-Chlorophenyl)-3-methoxyazetidin-2-one (107a): white solid; mp 93-95 o C; 1 H NMR (250 MHz, CDCl3): 7.34 (m, 4H), 6.55 (bs, 1H), 5.27 (d, 1H, J = 4.6Hz), 4.85 (d, 1H, J = 4.6Hz), 3.29 (s, 3H). ()-(3S,4R)-3-Methoxy-4-phenylazetidin-2-one (107b): white solid; mp 130-131 o C; 1 H NMR (250 MHz, CDCl3): 7.35 (m, 5H), 6.24 (bs, 1H), 5.30 (d, 1H, J = 4.6 Hz), 4.87 (dd, 1H, J = 3.4, 1.2 Hz), 3.30 (s, 3H). ()-(3S,4R)-3-Acetoxy-4-(2-chlorophenyl)azetidin-2-one (107c): white solid; mp 100-101 o C; 1 H NMR (250 MHz, CDCl3): 7.37 (m, 4H), 5.30 (d, 1H, J = 4.6Hz), 4.87 (t, 1H, J = 3.8Hz), 3.29 (s, 3H), 1.59 (s, 3H). 144

PAGE 159

7.1.6 Preparation of Phthalimide-based Sulfur Transfer Reagents (111) Method 1: In dry benzene, chlorine gas was bubbled until the increase in weight equaled 0.93g (0.01313 moles). This was then added to a round bottom flask containing 1.00g (0.01313 moles) 2-propanethiol in a minimal quantity of benzene and then sealed at 0 o C with stirring for 1 hour. The solution was then added dropwise to a second round bottom flask containing a slurry of 1.93g (0.01313 moles) phthalimide (110) and 2.89g (0.02232 moles) of Hunigs base and allowed to reach room temperature over 2 hours with stirring. The solution was then poured into water, extracted thrice with benzene, dried over magnesium sulfate and concentrated via rotary evaporation. This often generated a pure product. When unreacted phthalimide remained, triteration with chloroform dissolved the product nicely, leaving the starting material to be filtered off. When present, other impurities were removed by recrystallization with methanol or column chromatography. N O O S R NH O O R-S-Cl DIPEA111110 2-(Methylthio)isoindoline-1,3-dione (111a): white solid; 1 H NMR (250 MHz, CDCl3): mp 178-180 o C; 7.82 (m, 4H), 3.77 (s, 3H). 2-(Ethylthio)isoindoline-1,3-dione (111b): white solid; mp 158-160 o C; 1 H NMR (250 MHz, CDCl3): 7.86 (m, 4H), 2.92 (q, 2H, J = 7.34 Hz), 1.56 (t, 3H, J = 8.1 Hz). 2-(Propylthio)isoindoline-1,3-dione (111c): white solid; 77-78 o C; 1 H NMR (250 MHz, CDCl3): 7.87 (m, 4H), 2.86 (t, 2H, J = 7.3 Hz), 1.62 (m, 2H), 1.04 (t, 3H, J = 7.3 Hz) 2-(Butylthio)isoindoline-1,3-dione (111d): white solid; mp 189-190 o C; 1 H NMR (250 MHz, CDCl3): 7.8 (m, 4H), 2.85 (t, 2H, J = 7.4 Hz), 1.61 (m, 2H), 1.42 (m, 2H), 0.91 (t, 3H, J = 5.0Hz). 145

PAGE 160

146 2-(Isopropylthio)isoindoline-1,3-dione (111f): white solid; mp 61-62 o C; 1 H NMR (250 MHz, CDCl3): 7.8-7.5 (m, 4H), 4.9 (m, 1H), 2.17 (d, 6H, J = 6.9Hz). (+ )-2-( sec -Butylthio)isoindoline-1,3-dione (111g): white solid; mp 43-45 o C; 1 H NMR (250 MHz, CDCl3): 7.83 (m, 4H), 3.21 (m, 1H), 1.55 (m, 2H), 1.24 (d, 3H, J = 6.8 Hz), 1.04 (t, 3H, J = 7.3 Hz). 2-(Cyclohexylthio)isoindoline-1,3-dione (111i): white solid; mp 92-94 o C; 1 H NMR (250 MHz, CDCl3): 7.80 (m, 4H), 3.05 (m, 1H), 1.90-1.25 (m, 11H). 2-(Phenylthio)isoindolin e-1,3-dione (111j): white solid; mp 149-155 o C; 1 H NMR (250 MHz, CDCl3): 8.0-6.8 (m, 9H). 2-(Benzylthio)isoindoline-1,3-dione (111k): white solid; mp 165-168 o C; 1 H NMR (250 MHz, CDCl3): 7.9-7.5 (m, 4H), 7.31 (m, 5H), 1.75 (s, 2H). (+ )-2-(1-Phenylethylthio)isoindoline-1,3-dione (111l): white solid; mp 100-104 o C; 1 H NMR (250 MHz, CDCl3): 7.20 (m, 9H), 3.67 (q, 1H, J = 6.2), 1.76 (d, 3H, J = 5.0). 2-(2-Napthylthio)isoindoline-1,3-dione (111m): white solid; mp 120-128 o C; 1 H NMR (250 MHz, CDCl3): 7.65 (m, 11H). Method 2: In dry benzene 0.1g (0.0004424 moles) of N -bromophthalimide was dissolved in a minimal quantity of benzene and stoppered at 0 o C with stirring. The solution was then added dropwise to a second round botto m flask containing a solution of 0.034g (0.0004424 moles) of 1-butnaethi ol and 0.06g (0.0004424 moles) of triethyl amine. The mixture was allowed to achieve room temper ature over 2 hours with stirring. The solution was then poured into an equa l volume of water, extracted th rice with equa l volumes of benzene, dried over magnesium sulfate and c oncentrated via rotary evaporation. This often generated a fairly pure product. Unreact ed phthalimide was removed via triteration with chloroform which dissolv ed the product nicely, leaving the starting material to be

PAGE 161

147 filtered off. Also present, were other impurities which were removed via recrystallization with methanol and column chromatography. 2-(Octylthio)isoindoline-1,3-dione (111e): white solid; mp 55-58 o C; 1 H NMR (250 MHz, CDCl 3 ): 7.88 (m, 4H), 2.91 (t, 2H, J = 7.3 Hz), 1.76-1.28 (m, 12H), 0.89 (t, 3H, J = 6.4 Hz). 2-( tert -Butylthio)isoindoline-1,3-dione (111h): white solid; mp 130-133 o C; 1 H NMR (250 MHz, CDCl3): 7.87 (m, 4H), 1.36 (s, 9H). Method 3: In a round bottom flask was added 0.0061g (0.0000272 moles) of (anthracen10-yl)methanethiol, 0.0035g (0.0000354 mole s) potassium carbonate, and 0.0040g (0.0000272 moles) phthalimide in 5ml of reagent grade acet one. The vessel was sealed and the liquid meniscus was submerged below an active sonication bath for 24 hours. The solution was concentrated via rotary ev aporation and re-di ssolved in 10ml of dichloromethane. The solution was washed wi th and equal volume of water, dried over magnesium sulfate and concentr ated via rotary evaporation. The product yields were very low and purification was not attempted for loss of all product. 2-((Anthracen-10-yl)methylthio)isoindoline-1,3-dione (111n): white solid; 1 H NMR (250 MHz, CDCl3): 8.38-7.17 (m, 13H), 3.12 (s, 2H). Method 4: In a round bottom flask was added th e thioester dissolved in a minimal quantity of carbon tetrachl oride and cooled to 0 o C with stirring. A so lution of sulfuryl chloride in an equal volume of carbon tetrachloride was then added dropwise and the solution was allowed to come to room temperature with stirring over 30 minutes. The solution was then added dropwise into a slu rry of phthalimide and Hunigs base in an

PAGE 162

148 equal volume of carbon tetrachloride cooled to 0 o C with stirring. After the mixture was allowed to warm to room temperature over 3 hours with stirring, 40 ml of water was then added and the mixture was then allowed to s tir at high speed for 20 min. The mixture was filtered and the solid washed with 40 ml more of water. The solid was then triterated with chloroform and the dissolved product was dr ied over magnesium sulfate and concentrated via rotary evaporation. 2-(7-(Methyl2-(2-chloro-6-methoxy-3-oxo-3H-xanthen-9-yl)benzoate)thio)isoindoline-1,3-dione (111o): white solid; mp 140-146 o C; 1 H NMR (250 MHz, CDCl3): 9.2-7.3 (m, 9H), 6.05 (s, 1H), 5.35 (s, 1H), 3.8 (s, 3H), 3.05 (s, 3H). 2-(Polystyrenylthio)isoindoline-1,3-dione (164): light brown powder; IR 3015 (aromatic C-H stretch), 2980 (aliphatic C-H stretch), 1490, 1450 (C=O stretches). 7.1.7 Preparation of N -Thiolated Azetidin-2-ones N -Thiolated azetidin-2-ones (109j): In a round bottom fl ask was added 0.005g (0.0000236 moles) N -protio -lactam 107 0.0067g (0.000236 moles) of N phenylthiolated phthalimide a nd 0.0119g (0.000118 moles) of triethylamine, in a minmal quantity of dichloromethane. The solution was refluxed and followed via TLC. Reaction was complete after 12 hours. After cooling, the solution was poured into an equal volume of water, washed with aqueous solutions of 5% sodium bicarbonate, 1% sodium bisulfate, and saturated sodium chloride. The extracts were then dried over magnesium sulfate and concentrated via rotary evaporation. Phth alimide was removed via triteration with chloroform and the remaining impurities were removed via column chromatography, eluting with either dichloromethane.

PAGE 163

N O O H H3C Cl N O O S H3C Cl R N O O S R 107111109 Et3NorDIPEA ()-(3S,4R)-4-(2-Chlorophenyl)-3-methoxy-N-(methylthio)azetidin-2-one (109a) white crystal; mp 71-73 C; 1 H NMR (250 MHz, CDCl3): 7.35 (d, 1H, J = 7.4 Hz), 7.24 (m, 3H), 5.29 (d, 1H, J = 4.9 Hz), 4.80 (d, 1H, J = 4.9 Hz), 3.16 (s, 3H), 2.40 (s, 3H); 13 C NMR (63 MHz, CDCl3): 170.4, 133.8, 131.4, 129.6, 128.9, 126.8, 86.7, 62.7, 58.9, 21.8. (+ )-(3S,4R)-4-(2-Chlorophenyl)-N-ethylthio-3-methoxyazetidin-2-one (109b): white solid; mp 68-70 o C; 1 H NMR (250MHz, CDCl3) 7.34 (4H, m), 5.33 (1H, d, J = 5.0 Hz), 4.88 (1H, d, J = 5.0 Hz), 3.20 (3H, s), 1.35 (2H, q, J = 10.0 Hz), 0.92 (3H, t, J = 6.8 Hz); 13 C NMR (63 MHz, CDCl3): 172.5, 142.1, 137.5, 130.8, 128.8, 128.4, 87.5, 70.5, 46.7, 34.4, 22.0. ()-(3S,4R)-4-(2-Chlorophenyl)-3-methoxy-N-(propylthio)azetidin-2-one (109c) colorless oil; 1 H NMR (250 MHz, CDCl3): 7.6-7.2 (m, 4H), 5.45 (d, 1H, J = 4.9Hz), 5.09 (d, 1H, J = 4.9Hz), 3.21 (s, 3H), 2.49 (m, 2H), 1.55 (m, 2H), 0.82 (t, 3H, J = 6.9Hz); 13 C NMR (63 MHz, CDCl3): 171.0, 133.6, 128.8, 128.3, 86.3, 66.9, 58.3, 38.2, 30.8, 21.5, 13.6. (+ )-(3S,4R)-N-Butylthio-4-(2-chlorophenyl)-3-methoxyazetidin-2-one (109d): light yellow, viscous oil, 1 H NMR (250MHz, CDCl3) 7.30 (4H, m), 5.35 (1H, d, J = 4.9Hz), 4.89 (1H, d, J = 4.9Hz), 3.23 (3H, s), 2.85 (2H, t, J = 7.0), 1.60 (4H, m), 0.91 (3H, t, J = 7.3Hz); 13 C NMR (63 MHz, CDCl3): 176.5, 134.0, 132.1, 129.7, 129.0, 126.9, 86.8, 64.8, 58.1, 48.0, 12.3. (+ )-(3S,4R)-4-(2-Chlorophenyl)-N-octylthio-3-methoxyazetidin-2-one (109e): white solid, 1 H NMR (250MHz, CDCl3) 7.35 (4H, m), 5.35 (1H, d, J = 4.9 Hz), 4.89 (1H, d, J = 4.9 Hz), 3.23 (1H, m), 2.77 (2H m), 1.64-1.26 (12H, m), 0.88 (t, 3H, J = 6.2 Hz); 13 C NMR (63 MHz, CDCl3): 167.8, 132.5, 131.1, 127.5, 127.0, 80.3, 52.7, 43.7, 28.0, 26.0, 24.5, 22.3, 18.9. 149

PAGE 164

150 (+ )-(3S,4R)-4-(2-Chlorophenyl)-N -isopropylthio-3-methoxyazetidin-2-one (109f) : light yellow paste, mp 30-38 o C; 1 H NMR (250MHz, CDCl3) 7.34 (4H, m), 5.39 (1H, d, J = 4.9 Hz), 4.95 (1H, d, J = 4.9 Hz), 3.27 (1H, m), 3.25 (3H, s), 1.27 (6H, d, J = 5.7Hz); 13 C NMR (63 MHz, CDCl3): 171.3, 133.2, 128.9, 127.0, 87.1, 65.2, 59.7, 49.4, 34.0, 30.1, 28.5, 25.8, 21.5. (+ )-(3S,4R)N sec -Butylthio-4-(2-chlorophenyl)-3-methoxyazetidin-2-one (109g) : light yellow, viscous oil, + cis mixture. 1 H NMR (250MHz, CDCl3) 7.4 (1H, d, J = 7.4 Hz), 7.3 (3H, m), 5.3 (1H, d, J = 4.7 Hz), 4.9 (1H, d, J = 4.8 Hz), 3.2 (3H, s), 3.0 (1H, m), 1.48 (1H, m), 1.2 (3H, dd, J = 6.8, 4.9 Hz), 0.94 (3H, q, J = 6.0 Hz); 13 C NMR (63 MHz) 171.0, 133.8, 131.4, 129.5, 128.9, 126.8, 86.3, 64.1, 58.8, 48.1, 28.1, 19.0, 18.6, 11.1. (+ )-(3S,4R)N tert -Butylthio-4-(2-chlorophenyl)-3-methoxyazetidin-2-one (109h) : white solid, 1 H NMR (250MHz, CDCl3) 7.34 (4H, m), 5.50 (1H, d, J = 4.8 Hz), 4.99 (1H, d, J = 4.8 Hz), 3.25 (3H, s), 1.35 (9H, s); 13 C NMR (63 MHz, CDCl3): 166.0, 133.0, 129.0, 88.0, 65.0, 60.0, 44.0, 34.0, 31.0, 27.0, 19.0. (+ )-(3S,4R)-4-(2-Chlorophenyl)-N -cyclohexylthio-3-methoxyazetidin-2-one (109i) : brown solid; mp 7879 o C; 1 H NMR (250MHz, CDCl3) 7.34 (4H, m), 5.35 (1H, d, J = 4.9 Hz), 4.93 (1H, d, J = 4.9 Hz), 3.24 (3H, s), 3.05 (1H, m), 2.01-1.41. (10H, m); 13 C NMR (63 MHz, CDCl3): 171.3, 133.5, 128.9, 128.7, 128.3, 85.2, 67.6, 58.3, 49.5, 32.2, 30.9, 25.6, 25.4. (+ )-(3S,4R)-4-(2-Chlorophenyl)-3-methoxyN -phenylthioazetidin-2-one (109j): light yellow, oily solid, mp 60-62 o C 1 H NMR (250MHz, CDCl3) 7.34 (9H, m), 5.35 (1H, d, J = 5.0Hz), 4.91 (1H, d, J = 5.0Hz), 3.25 (3H, s); 13 C NMR (63 MHz, CDCl3): 172.0, 137.0, 135.0, 134.0, 132.0, 130.0, 129.0, 128.0, 127.0, 125.0, 82.0, 59.0, 48.0. (+ )-(3S,4R)N -Benzylthio-4-(2-chlorophenyl)-3-methoxyazetidin-2-one (109k) : light yellow, oily solid; mp 68-70 o C; 1 H NMR (250MHz, CDCl3) 7.22 (9H, m), 5.30 (1H, d, J = 4.7 Hz), 4.55 (1H, d, J = 4.8 Hz), 3.16 (3H, s), 1.25 (3H, s); 13 C NMR (63 MHz, CDCl3): 179.0, 142.0, 138.0, 136.0, 135.0, 134.0, 132.0, 130.0, 128.0, 126.0, 88.0, 59.0, 48.0, 40.0.

PAGE 165

151 (+ )-(3S,4R)-4-(2-Chlorophenyl)-3-methoxyN -(1-phenylethylthio)-azetidin-2-one (109l): white solid; mp 65-67 o C; 1 H NMR (250MHz, CDCl3) 7.56 (9H, m), 5.25 (1H, d, J = 4.2 Hz), 4.84 (1H, d, J = 4.2 Hz), 3.27 (3H, s), 3.01 (2H, m), 0.93 (3H, d, J = 7.1 Hz); 13 C NMR (63 MHz, CDCl3): 168.9, 164.1, 139.6, 138.3, 134.5, 133.5, 118.5, 79.9, 51.4, 43.7, 40.8, 29.3. (+ )-(3S,4R)-4-(2-Chlorophenyl)-3-hydroxy -N -naphthylthioazetidin-2-one (109m1) : white solid; 1 H NMR (250MHz, CDCl3) 7.9-7.2 (11H, m), 5.3 (2H, m); 13 C NMR (63 MHz, CDCl3): 172.0, 142.0, 135.0, 133.0, 132.0, 131.0, 129.0, 128.0, 127.0, 126.0, 125.0, 124.0, 123.0, 122.0, 81.0, 58.0. (+ )-(3S,4R)-4-(2-Chlorophenyl)-3-methoxyN -napthylthioazetidin-2-one (109m) : yellow solid, 1 H NMR (250MHz, CDCl3) 8.13-7.19 (11H, m), 5.35 (1H, d, J = 5.0 Hz), 4.91 (1H, d, J = 5.0 Hz), 3.24 (3H, s); 13 C NMR (63 MHz, CDCl3): 170.0, 142.0, 130.0, 134.0, 132.0, 131.0, 130.0, 129.0, 128.0, 127.0, 126.0, 125.0, 123.0, 121.0, 82.0, 58.0, 55.0. (+ )-(3S,4R)N -((Anthracen-10-yl)methylthio)-4-(2-chlo rophenyl)-3-methoxyaze tidin-2-one (109n) : pink solid, 1 H NMR (250MHz, CDCl3) 7.48-7.19 (13H, m), 5.25 (1H, d, J = 4.8 Hz), 4.81 (1H, d, J = 4.8 Hz), 4.078 (2H, s), 3.25 (3H, s). (+)-(3S,4R)N -(R)(sec -Butylthio)-3-methoxy-4-phenylazetidin-2-one (152) : white oil; 1 H NMR (250MHz, CDCl3) 7.29 (m, 5H), 5.94 (d, 1H, J = 4.9 Hz), 5.01 (d, 1H, J = 4.9), 2.99 (m, 1H), 1.47 (m, 2H), 1.28 (d, 3H, J = 6.8 Hz), 0.94 (t, 3H, J = 7.4 Hz). (-)-(3S,4R)N -(R)( sec -Butylthio)-3-methoxy-4-phenylazetidin-2-one (153) : white oil; 1 H NMR (500MHz, CDCl3) 7.30 (m, 5H), 5.87 (d, 1H, J = 3.5 Hz), 4.94 (d, 1H, J = 4.0 Hz), 2.90 (m, 1H), 1.621.40 (m, 2H), 1.12 (d, 3H, J = 6.5 Hz), 0.90 (t, 3H, J = 8.0 Hz). (+)-(3R,4S)N -(S)(sec -Butylthio)-3-methoxy-4-phenylazetidin-2-one (154): white oil; 1 H NMR (250MHz, CDCl3) 7.32 (m, 5H), 5.94 (d, 1H, J = 4.9 Hz), 5.01 (d, 1H, J = 4.9 Hz), 2.97 (m, 1H), 1.56 (m, 2H), 1.18 (d, 3H, J = 6.9 Hz), 0.97 (t, 3H, J = 7.4 Hz). (-)-(3R,4S)N -(S)( sec -Butylthio)-3-methoxy-4-phenylazetidin-2-one (155): white oil; 1 H NMR (250MHz, CDCl3) 7.29 (m, 5H), 5.94 (d, 1H, J = 4.9 Hz), 5.01 (d, 1H, J = 4.9 Hz), 2.99 (m, 1H), 1.51 (m, 2H), 1.28 (d, 3H, J = 6.75 Hz), 0.94 (t, 3H, 7.4 Hz).

PAGE 166

152 7.1.8 Preparation of Thioesters Method 1 (Mitsunobu Reaction): In a round bottom flask was added 9ml (0.04506 moles) of diisopropyl azodicaroxylate (DIAD) and 12.16g (0.04506 moles) of triphenylphosphine were combined in THF at 0 o C, under nitrogen with stirring. This solution was allow to stir for 30 min, or until a white precipitate formed. Then a solution of 3.43g (0.04506 moles) thiolacetic acid and 3.25g (0.0225 moles) of 2-nonanol in a minimal quantity of THF was added carefull y, dropwise. After work-up it was very difficult to remove all of the tri phenylphosphine as it shared a close R f value with the desired products. Method 2 (Substitution): In a round bottom flask was added (1 eq.) an alkyl halide in a minimal amount of benzen e under nitrogen, at 0 o C with stirring. To this, a solution, of (1eq.) triethylamine and (1 eq.) thiolacetic acid in enough benzene, was added dropwise. The solution was allowed to come to room temperature over 3 hours with stirring. The reaction mixture was then poured into an eq uivalent portion of wa ter, extracted with benzene, dried over magnesium sulfate and conc entrated via rotary evaporator. Continued exposure to a stream of nitr ogen successfully removed a ny remaining triethylamine. 7.1.9 Preparation of NSulfoxylated -Lactams Method 1 H 2 O 2 (157b): Inspired via Prinzbach and Netscher, Synthesis 1987, 683-688. In a round bottom flask was added 0.0135g (0.0000415 moles) an N -cyclohexylthiolated -lactam in a minimal amount of glacial acetic acid at 0 o C with stirring. To this, a solution, of 0.008g (0.0000415 moles) of 30% hydroge n peroxide in few drops of glacial

PAGE 167

153 acetic acid, was added dropwise. The solution wa s allowed to come to room temperature over 3 hours with stirring. The reaction mixtur e was then poured into an equivalent portion of water, extracted with three equivale nt volumes of benzene, washed with water until the washings were at neutral pH, as observed via pH paper, dried over magnesium sulfate and concentrated via rotary evaporation. (+ )-(3S,4R)N -Cyclohexylsulfinyl-4-(2-chloro phenyl)-3-methoxyazetidin-2-one (157b) : white oil, 1 H NMR (250MHz, CDCl3) 7.41 (8H, m), 5.85 (1H, d, J = 6.0Hz), 5.71 (1H, d, J = 5.6), 4.94 (1H, d, J = 6.0Hz), 4.91 (1H, d J = 6.0Hz), 3.75 (2H, m), 3.29 (3H, s), 3.23 (3H, s), 2.72 (2H, m), 2.2-0.80 (20H, m); 13 C NMR (63 MHz, CDCl3): 171.4, 133.6, 130.9, 130.8, 129.3, 123.5, 83.4, 61.6, 53.2, 50.5, 37.4, 29.9, 25.6. (+ )-(3S,4R)N -Phenylsulfinyl-4-(2 -chlorophenyl)-3-methoxyazetidin-2-one (157a) : white oil, 1 H NMR (250MHz, CDCl3) 7.30 (m, 9H), 5.18 (bs, 1H), 4.80 (bs, 1H), 3.25 (bs, 1H); 13 C NMR (63 MHz, CDCl3): 168.9, 164.1, 131.4, 128.1, 131.5, 118.0, 113.9, 108.8, 79.7, 54.7, 43.9. Method 2 m CPBA: In a round bottom flask was a dded 0.20 g (0.00085 moles) an N sec butylthiolated -lactam in a minimal amount of diethyl ether at 0 o C with stirring. To this, a solution, of 0.14 g (0.00080 moles) meta -chloroperoxybenzoic acid in a minimal amount of diethyl ether, was added dropwise. The solution wa s monitored via thin layer chromatography. The reaction mixture was then poured into an equivalent portion of water, extracted thrice with equivalent volum es of benzene, washed twice with water, dried over magnesium sulfate and concentrated via rota ry evaporation.

PAGE 168

154 (+ )-(3S,4R)N sec -Butylsulfinyl-4-(2-chlorophe nyl)-3-methoxyazetidin-2-one (157c) : dark brown paste, mp 35-43 o C; 1 H NMR (250 MHz, CDCl3) 7.32 (4H, m), 5.38 (1H, d, J = 4.9 Hz), 4.94 (1H, d, J = 4.9 Hz), 3.11 (2H, m), 2.11 (3H, s), 1.75-0.96 (8H, m); 13 C NMR (63 MHz, CDCl3): 171.2, 149.8, 131.4, 129.5, 129.0, 126.8, 86.5, 74.1, 58.8, 48.1, 28.1, 19.0, 14.6, 11.2. 7.1.10 Preparation of NSulfonylated -Lactams Method: Identical to 7.1.9 Method 1. (+ )-(3S,4R)N -Cyclohexylsulfonyl-4-(2-chlorophenyl)-3-methoxyazetidin-2-one (158b) : white, oily solid; mp 152-158 o C; 1 H NMR (250 MHz, CDCl3) 7.47 (m, 4H), 5.79 (d, 1H, J = 5.0 Hz), 4.97 (d, 1H, J = 4.6 Hz), 3.29 (s, 3H), 3.12 (m, 1H), 2.20-0.86 (m, 10H); 13 C NMR (63 MHz, CDCl3): 174.9, 143.8, 127.9, 125.8, 120.1, 118.5, 84.6, 61.9, 52.0, 45.2, 40.1, 34.9, 33.6. (+ )-(3S,4R)N -Phenylsulfonyl-4-(2-chlorophe nyl)-3-methoxyazetidin-2-one (158a) : white oil, 1 H NMR (250 MHz, CDCl3) 7.96-7.16 (9H, m), 5.70 (1H, d, J = 5.6 Hz), 4.82 (1H, d, J = 5.6 Hz), 3.35 (3H, s) 13 C NMR (63 MHz, CDCl3): 168.9, 164.1, 144.4, 139.5, 134.2, 128 .0, 125.7, 119.0, 79.9, 54.5, 43.8. 7.1.11 Preparation of a -Lactam Sulfonic Acid Method (160): Literature Procedure Cimarusti, C. Tetrahedron 1983, 39, 2577. In a round bottom flask a solution 0.01g (0.472 mmol) of N -protio -lactam in 2 ml of freshly distilled dichloromethane and 2 ml of dry DMF was stirred with 0.014g (0.00944 moles) of 50% sulfur trioxide-pyrid ine under nitrog en at room temperature for 2 hours. The solution was concentrated via rotary evaporat or. The resultant salt was ion exchanged for a potassium salt. Application of this salt to column chromatography afforded the necessary protonation to afford the sulfonic acid in a pure form.

PAGE 169

155 (+ )-(3S,4R)-2-(2-Chlorophenyl)-3-methoxy-4-oxoazetidine-1-sulfonic acid (160) : dark brown oil, 1 H NMR (250MHz, CDCl3) 7.32 (4H, m), 5.69 (1H, d, J = 4.9 Hz), 4.98 (1H, d, J = 4.9 Hz), 3.32 (2H, m), 0.89 (1H, s); 13 C NMR (63 MHz, CDCl3): 179.9, 145.2, 137.7, 131.0, 122.7, 83.4. 55.2. 43.3. 7.1.12 De-Acetylation of a C 3 Acetoxy Substituted -Lactam Method: To a round bottom flask 0.184g (0.0329 moles) of potassium hydroxide was dissolved in a minimal quantity of methanol. To this solution was added a solution of 0.787g (0.0329 moles) of (+)-(3S,4R)-4-(2-chlor ophenyl)-3-acetoxy-1-azetidin-2-one in enough 0 o C. The solution was allowed to warm to room temperature over night with stirring. The solution was then concentrated via rotary evaporation, re-dissolved in methylene chloride, washed with water until the washing were neutra l pH via pH paper, dried over magnesium sulfate, and concentrated via rota ry evaporation. (+ )-(3S,4R)-4-(2-Chlorophenyl)-3-hydroxyN -azetidin-2-one (108OH): white solid; mp 204-205 o C; 1 H NMR (250 MHz, CDCl3): 7.30 (m, 4H), 6.82 (s, 1H), 5.64 (d, 1H, J = 5.2 Hz), 5.34 (d, 1H, J = 4.9 Hz), 3.79 (s, 3H). 7.1.13 Re-Acetylation of a C 3 Hydroxy Substituted -Lactam. Method: To a round bottom flask was added 10.0ml (excess) of pyridine and 0.119ml (0.0001283moles) of acetic anhydride and chilled to 0 o C. To this solution was added 0.0345g (0.0001283 moles) of (-)-(3S,4R)-4-(2-Chlorophenyl)-3-hydroxyN -(4methoxyphenyl)azetidin-2-one. The resultant solution was stirred for 1 hour at 0 o C and was allowed to stir overnigh t at room temperature, then poured into i ce water. The mixture was filtered to yield a pure white solid.

PAGE 170

156 (-)-(3S,4R)-3-Acetoxy-4-(2-chlorophenyl)N -(4-methoxyphenyl)azetidin-2-one (p152): white solid; mp 165-167 o C; 1 H NMR (250 MHz, CDCl3): 7.30 (m, 8H), 5.94 (d, 1H, J = 4.9 Hz), 5.34 (d, 1H, J = 4.9 Hz), 3.76 (s, 3H), 1.68 (s, 3H). 7.1.14 Dansylation Method: To a round bottom flask was added 0.005 g (0.0261 mmol) to a minimal quantity of tetrahydrofuran for dissolution. The flask was then chilled to 0 o C with stirring. To this was added 0.002 (0.0522 mmol) of sodium hydride, pre-dissolved in aminimal quantity of tetrahydrofuran. This solution was allowed to stir for 30 minutes. This solution was then added to 0.007 (0.0261 mmol) of dansyl chloride, pre-dissolved in a minimal quantity of tetrahydrofuran. This mi xture was allowed to stir for an additional 30 minutes. The remaining sodium hydride was quenched with slightly wet methanol. The reaction mixture was then poured into an equal volume of wate r and extracted thrice with equal portions of tetrahydrofuran. The organic layers were combined and washed with an equal volume of water. The organi c extracts were then dried over magnesium sulfate and concentrated via rotary evaporation. 2-(2-Chlorophenyl)-4-oxoazetidin-3-yl 5-(d imethylamino)naphthalene-1-sulfonate: yellow solid; mp 135-140 o C; 1 H NMR (250 MHz, CDCl3): 8.48 (b, 1H), 8.25 (b, 1H), 8.05 (b, 1H), 7.52 (b, 1H), 7.40 (b, 1H), 7.30 (b, 2H), 6.90 (b, 1H), 6.75 (b, 1H), 6.60 (b, 1H), 5.25 (b, 1H), 4.80 (b, 1H), 2.85 (s, 6H). 7.2 Microbiological Test Procedures The following bacteria were used for the antimicrobial evaluation of N -thiolated -lactams: Bacillus anthracis (Sterne strain), Bacillus cereus (ATCC 14579), Bacillus coagulans (USF 546), Bacillus globigii (Department of Defense Reagents Program),

PAGE 171

157 Bacillus megaterium (ATCC 14581), Bacillus subtilis (19569), Bacillus thuringensis (ATCC 10792), Bacteroides fragalis (obtained from Smith-Kline Laboratory), Candida albicans (clinical isolate), Candida tropicalis (clinical isolate), Enterobacter cloace (environmental isolate, USF510), Enterococcus gallinarium (ATCC 49573), Enterococcus faecalis (ATCC 19433), Enterococcus casseliflavus (ATCC 700327), Enterococcus durans (ATCC 6056), Enterococcus avirum (ATCC 14025), Enterococcus saccharolyticus (ATCC 43076), Escherichia coli (ATCC 23590), Haemophilus influenzae ( USF 561), Klebsiella pneumoniae (USF 512), Lactococcus lactis (ATCC 11454), Listeria monocytogenes (ATCC 19115), Micrococcus luteus (environmental isolate, USF681), Niesserria gonnorheae (obtained from the Tampa Branch State Laboratory, -lactamase positive, USF 662), Pseudomonas aeruginosa (ATCC 15442), Salmonella typhimurium (obtained from University of S outh Florida Medical Clinic, USF 515), Serratia marcescens (ATCC 29634), Staphylococcus aureus USF525 (ATCC 25923) Staphylococcus aureus USF652-658 (obtained from Lakeland Regional Medical Center, -lactamase positive), Staphylococcus epidermidis (environmental isolate, USF528), Staphylococcus saprophyticus (ATCC 35552), Staphylococcus simulans (ATCC 11631), Staphylococcus capitis (ATCC 35661), Staphylococcus cohnii (ATCC 35662), Staphylococcus lentus (ATCC 700403), Staphylococcus lugdunensis (ATCC 700328), Staphylococcus xylosus (ATCC 29971), Streptococcus pyrogenes, Streptococcus agalactiae Vibrio cholerae (biotype E1 Tor Ogawa, cholera toxin positive, CDC E5906),

PAGE 172

158 7.2.1 Antimicrobial Susceptibility Testing Culture preparation: From a freezer stock in tryptic soy broth (Difco Laboratories, Detroit, MI) and 20% glycerol, a culture of each organism was grown on tryptic soy agar (TSA) plates (Becton-Dickinson Laboratories, Cockeysville, MD) at 37C for 24 hours. A 108 suspension was then made in steril e phosphate buffered saline (pH 7.2) and swabbed across fresh TSA plates. Disc method: From each 1mg/ml stock solution in dimethyl sulfoxide (DMSO), sterile 6mm paper discs (Becton-Dickinson Laborator ies, Cockeysville, MD) were impregnated with 20 l of the test compounds. At this con centration, the microliter quantity is equivalent to the micrograms in solution. The discs were allowed to dry in a biohazard safety hood then placed onto the inoculated TSA plates. The plates were incubated for 24 hours at 37C and the antimicrobial susceptib ilities were determined by measuring the zones of growth inhibition around each disc. Well method: A 108 standardized cell count suspension was then made in sterile phosphate buffered saline (pH 7.2) and swabbed across fresh TSA plates. Circular wells (6 mm in diameter) were cut into the inoculated plates and 20 L of a 1 mg/ml stock solution of the test lactam in dimethylsulfoxi de (DMSO) was pipetted into the wells. The plates were incubated for 24 hours at 37C and the antimicrobial susceptibilities were determined by measuring the zones of growth inhibition around each well. 7.2.2 MIC Calculations Media preparation: The minimum inhibitory concentrations were determined by the agar plate dilution (need reference). The te st media were prepared in 24 well plates

PAGE 173

159 (Costar 3524, Cambridge, MA) by adding a know n concentration of the test drug in DMSO together with a solution of Mu eller-Hinton II agar (Becton-Dickinson Laboratories, Cockeysville, MD) for a total vol ume of 1 ml in each well. Calculations of the overall concentration of antibiotic in the wells were standardized by measuring from a 1mg/ml stock solution of the test drug. At this concentration, the micr oliter quantity is equivalent to the micrograms in solution. The amount of agar soluti on added to the wells was determined by adding to the quantity of test drug in each well to give a combined volume of 1 ml. Following preparation of th e well plates, the media were allowed to solidify at room temperature fo r 24 hours before inoculation. Inoculation: From an 24 hour culture of each organism on tryptic soy agar (TSA) plates (Becton-Dickinson Laboratories, Cockeysville, MD), the Staphylococcal strains were grown overnight in 5 ml of tr yptic soy broth (Difco Laborator ies, Detroit, MI) at 37C. One microliter of each culture was then applied to the appropriate well of agar and incubated at 37C overnight. After 24 hr, th e MICs were determined by examining the wells for growth. 7.2.4 Glucose Uptake / Respiration Study A fresh 200 ml culture of an 106 cfu/ml suspension of S. aureus (ATCC 25923) was precipitated via centrifugation. The so lid contents were separated from the supernatant and washed with a phosphate buffer solution (P BS). The mixture was then precipitated again and the supernatant removed. The cells were resuspended in 10 ml of PBS to an approximate concentration of 5 X 10 9 cells/ml. This is the bacteria stock

PAGE 174

160 solution. The solutions of 1 M horseradish pe roxidase (HRP), and glucose oxidase were prepared. 4mL of 5X buffer was added to 16 mL of DI water to generate a 1X buffer. Then, 1ml of 1X buffer was used to dissolve 5.9 mg of glucose. Then 50 L of this solution was diluted in 3950L of 1X buffe r. This is the glucose stock solution. Next, 1mg of lactam was dissolved in 10m L of DMSO/water. This is the lactam stock solution. 1.2 mg of glutathione was a dded to 10 mL of water. This is the glutathione stock solution. Then 1.2 mg of erythromycin was dissolved in 10 ml DMSO/water. This is the erythromycin stoc k solution. Finally, 1.2 mg of dithiothritol (DTT) was dissolved in 10ml DMSO/water. This is the DTT stock solution. The experimental tubes were then set up as follows. Experiment tubes: 1)Tube 1. Maximum Glucose Uptake. Add 1mL of Bacteria Stock Solution Add 303L of Glucose Stock Solution Add 3.99mL of buffer. 2)Tube 2. Just Sugar. Add 303L of Glucose Stock Solution Add 4.99mL of buffer. 3)Tube 3. Sugar with Lactam. Add 303L of Glucose Stock Solution Add 1mL of Lactam Stock Solution Add 3.99mL of buffer. 4)Tube 4. Test 0.5 MIC (0.25g/mL) Add 303L of Glucose Stock Solution Add 0.250mL of Lactam Stock Solution Add 1mL of Bacteria Stock Solution Add 3.74mL of buffer.

PAGE 175

161 5)Tube 5. Test 1.0 MIC (0.5g/mL) Add 303L of Glucose Stock Solution Add 0.50mL of Lactam Stock Solution Add 1mL of Bacteria Stock Solution Add 3.49mL of buffer. 6)Tube 6. Test 2.0 MIC (1.0g/mL) Add 303L of Glucose Stock Solution Add 1.0mL of Lactam Stock Solution Add 1mL of Bacteria Stock Solution Add 2.74mL of buffer. 7)Tube 7. Test 4.0 MIC (2.0g/mL) Add 303L of Glucose Stock Solution Add 2.0mL of Lactam Stock Solution Add 1mL of Bacteria Stock Solution Add 1.74mL of buffer. 8)Tube 8. Just Buffer. Add 5mL of buffer 9)Tube 9. Glutathione Add 303L of Glucose Stock Solution Add 1mL of Bacteria Stock Solution Add 1mL of Glutathione Stock Solution Add 2.74mL of buffer. 10)Tube 10. Erythromycin Add 303L of Glucose Stock Solution Add 1mL of Bacteria Stock Solution Add 1mL of Erythromycin Stock Solution Add 2.74mL of buffer. 11)Tube 11. DTT Add 303L of Glucose Stock Solution Add 1mL of Bacteria Stock Solution Add 1mL of DTT Stock Solution Add 2.74mL of buffer. 12)Tube 12. Erythromycin Blank Add 303L of Glucose Stock Solution Add 1mL of Doxycyclin Stock Solution Add 3.74mL of buffer.

PAGE 176

162 Once these were prepared, all 12 tube s were incubated with shaking at 37 o C for 30 minutes. A standard curve was prepared as at the concentrations shown in table 7.1. Table 7.1 Standard Curve Well Plate Final Concentrations of Sugar Well (micromolar) 1 0 2 3 3 6 4 9 5 12 6 15 7 18 8 21 9 24 10 27 11 30 12 33 Each tube was then precipitated via centrifugation and the supernatant was filtered with 0.2m cellulose nitrate membra ne. 50L of each supernatant was pipetted into three wells each. The Amplex Red stock was prepared by dissolv ing the contents of the vial of Amplex Red reagent in 60 L of DMSO. A Working Solution of fluorophore was prepared by mixing 1) 50L of Amplex Red Stock, 2) 100L of HRP, 3) 100L of Glucose Oxidase, and 4) 4.75mL 1X Buffer. Then 50L of this Working Solution was added to each well. The entire plate of we lls was then incubated in a drawer for 30 minutes. Fluorescence was then measured via a fluorometer.

PAGE 177

163 7.2.5 Resin / Lysate Exposure 1 liter of Staphylococcus aureus USF849 was cultured at 37 o C for 24 hours. The culture was then centrifuged and washed. The resultant pellet was washed with PBS buffer and then resuspended in 5 ml of PBS bu ffer. The cells were then sonicated in an icebath for 30 minutes total, stopping every 5 minutes to check for overheating. The lysed cells were then centrifuged and the lysate ex tracted. The lysate was centrifuged again and the lystate extracted from the solids. Filtering the lysate yielded a slightly opaque yellow solution. 0.25226g of Lactam Thiolated Resin ( 12 ) was swelled in 0.5ml of DMSO and added the lysate solution. This mixture was then setup for 200 rpm shaking for 24 hours. Next, the lysate was filtered and repeatedly washed solid with boiling water and boiling ethanol. The solid was dried and 196mg of post lysate exposed resin material was collected which was a light brown mixture of amorphous and crystalline solid (looked like sand). After DIBAL cleavage, the solid was repeatedly freezed-dried with 100% deuterium oxide. Spectrum 8.49 was observed be aring stark similarity to Conezyme A. 7.2.6 HPLC Experiments All HPLC experiments (Spectra 8.51 th rough 8.59) were done using a Shimadzu LC-8A HPLC through a analytical reverse phase column to a Shimadzu SPD-10A UVVIS Detector at a 2ml/min flow rate. 10 M solutions of standards: coenzyme A, N protio -lactam, N -ethylthiolated -lactam, N sec-butylthiolated -lactam, phenylthiolated -lactam and the ethylthio-coenzyme A disulfide adduct were prepared. As well, mixtures of each thiolated -lactam with coenzyme A were also produced. A cell lysate of S. aureus prepared as above, was prepared and mixed with a equal volume of a

PAGE 178

164 20 M solution of N -thiolated -lactam. Each mixture of coenzyme A and lactam provided two new peaks, apparen tly the adduct a nd the resultant N -protio -lactam. The mixture of lysate and lactam appeared to present the same adduct and N -protio -lactam peaks. A reverse phase column was used w ith a eluent equal to 90% acetonitrile 10% water.

PAGE 179

CHAPTER EIGHT SPECTRA Spectrum 8.01: 1 H NMR (250 MHz, CDCl3) of imine 105: Cl N H O CH3 105 165

PAGE 180

Spectrum 8.02: 1 H NMR (250 MHz, CDCl3) of -lactam 106: N O O H3C O Cl H3C 106 166

PAGE 181

Spectrum 8.03: 1 H NMR (250 MHz, CDCl3) of -lactam 107: N H O O H3C Cl 107 167

PAGE 182

Spectrum 8.04: 1 H NMR (250 MHz, CDCl3) of -lactam 107b: N H O O H3C 107b 168

PAGE 183

Spectrum 8.05: 1 H NMR (250 MHz, CDCl3) of phthalimide 111a: N O O S CH3 111a 169

PAGE 184

Spectrum 8.06: 1 H NMR (250 MHz, CDCl3) of phthalimide 111b: N O O S 111bCH3 170

PAGE 185

Spectrum 8.07: 1 H NMR (250 MHz, CDCl3) of phthalimide 111c: NS CH3 O O 111c 171

PAGE 186

Spectrum 8.08: 1 H NMR (250 MHz, CDCl3) of phthalimide 111d: NS O O CH3 111d 172

PAGE 187

Spectrum 8.09: 1 H NMR (250 MHz, CDCl3) of phthalimide 111e: NS O O CH3 111e 173

PAGE 188

Spectrum 8.10: 1 H NMR (250 MHz, CDCl3) of phthalimide 111f: NS CH3 CH3 O O 111f 174

PAGE 189

Spectrum 8.11: 1 H NMR (250 MHz, CDCl3) of phthalimide 111g: NS CH3 O O CH3 111g 175

PAGE 190

Spectrum 8.12: 1 H NMR (250 MHz, CDCl3) of phthalimide 111h: N O O S CH3 CH3 CH3 111h 176

PAGE 191

Spectrum 8.13: 1 H NMR (250 MHz, CDCl3) of phthalimide 111i: NS O O 111i 177

PAGE 192

Spectrum 8.14: 1 H NMR (250 MHz, CDCl3) of phthalimide 111j: NS O O 111j 178

PAGE 193

Spectrum 8.15: 1 H NMR (250 MHz, CDCl3) of phthalimide 111k: NS O O 111k 179

PAGE 194

Spectrum 8.16: 1 H NMR (250 MHz, CDCl3) of phthalimide 111l: NS CH3 O O 111l 180

PAGE 195

Spectrum 8.17: 1 H NMR (250 MHz, CDCl3) of phthalimide 111m: NS O O 111m 181

PAGE 196

Spectrum 8.18: 1 H NMR (250 MHz, CDCl3) of phthalimide 111n: NS O O 111n 182

PAGE 197

Spectrum 8.19: 1 H NMR (250 MHz, CDCl3) of phthalimide 111o: NOOSOOOOOH3CClH3C 111o 183

PAGE 198

Spectrum 8.20: FTIR of phthalimide resin 164: S N O O 164 184

PAGE 199

Spectrum 8.21: 1 H NMR (250 MHz, CDCl3) of lactam 109a: N O O S CH3 H3C Cl 109a 185

PAGE 200

Spectrum 8.22: 1 H NMR (250 MHz, CDCl3) of lactam 109b: N O O H3C S CH3 Cl 109b 186

PAGE 201

Spectrum 8.23: 1 H NMR (250 MHz, CDCl3) of lactam 109c: N O O H3C S Cl CH3 109c 187

PAGE 202

Spectrum 8.24: 1 H NMR (250 MHz, CDCl3) of lactam 109d: N O O S H3C Cl CH3 109d 188

PAGE 203

Spectrum 8.25: 1 H NMR (250 MHz, CDCl3) of lactam 109e: N O O H3C S Cl CH3 109e 189

PAGE 204

Spectrum 8.26: 1 H NMR (250 MHz, CDCl3) of lactam 109f: N O O S H3C Cl CH3CH3 109f 190

PAGE 205

Spectrum 8.27: 1 H NMR (250 MHz, CDCl3) of lactam 109g: N O O S CH3 CH3 H3C Cl 109g 191

PAGE 206

Spectrum 8.28: 1 H NMR (250 MHz, CDCl3) of lactam 109h: N O O H3C S CH3 Cl CH3 H3C 109h 192

PAGE 207

Spectrum 8.29: 1 H NMR (250 MHz, CDCl3) of lactam 109i: N O O S H3C Cl 109i 193

PAGE 208

Spectrum 8.30: 1 H NMR (250 MHz, CDCl3) of lactam 109j: N O O H3C S Cl 109j 194

PAGE 209

Spectrum 8.31: 1 H NMR (250 MHz, CDCl3) of lactam 109k: N O O S H3C Cl 109k 195

PAGE 210

Spectrum 8.32: 1 H NMR (250 MHz, CDCl3) of lactam 109l: N O O S H3C Cl CH3 109l 196

PAGE 211

Spectrum 8.33: 1 H NMR (250 MHz, CDCl3) of lactam 109m1: N O HO S Cl 109m1 197

PAGE 212

Spectrum 8.34: 1 H NMR (250 MHz, CDCl3) of lactam 109m: N O O S H3C Cl 109m 198

PAGE 213

Spectrum 8.35: 1 H NMR (250 MHz, CDCl3) of lactam 109n: N O O S H3C Cl 109n 199

PAGE 214

Spectrum 8.36: 1 H NMR (250 MHz, CDCl3) of (R)(+) lactam 152: N O O S H3C CH3 CH3 152 200

PAGE 215

Spectrum 8.37: 1 H NMR (500 MHz, CDCl3) of (R)(-) lactam 153: N O O S H3C CH3 CH3 153 201

PAGE 216

Spectrum 8.38: 1 H NMR (250 MHz, CDCl3) of (S)(+) lactam 154: N O O S H3C CH3 CH3 154 202

PAGE 217

Spectrum 8.39 1 H NMR (250 MHz, CDCl3) of (S)(-) lactam 155: N O O S H3C CH3 CH3 155 203

PAGE 218

Spectrum 8.40: 1 H NMR (250 MHz, CDCl3) of lactam 157b: N O O S H3C O Cl 157b 204

PAGE 219

Spectrum 8.41: 1 H NMR (250 MHz, CDCl3) of lactam 157a: N O O S H3C Cl O 157a 205

PAGE 220

Spectrum 8.42: 1 H NMR (250 MHz, CDCl3) of lactam 157c: N O O S H3C CH3CH3 O 157c 206

PAGE 221

Spectrum 8.43: 1 H NMR (250 MHz, CDCl3) of lactam 158b: N O O S H3C Cl O O 158b 207

PAGE 222

Spectrum 8.44: 1 H NMR (250 MHz, CDCl3) of lactam 158a: N O O S H3C Cl O O 158a 208

PAGE 223

Spectrum 8.45: 1 H NMR (250 MHz, CDCl3) of lactam 160: N O O S H3C HO Cl O O 160 209

PAGE 224

Spectrum 8.46: FTIR of lactam resin 165: S N O O CH3 Cl 165 210

PAGE 225

Spectrum 8.47: 1 H NMR (250 MHz, CDCl3) of lactam 108OH: N O HO Cl H 108OH 211

PAGE 226

Spectrum 8.48: 1 H NMR (250 MHz, CDCl3) of lactam p152: N O O H3C O O CH3 p152 212

PAGE 227

Spectrum 8.49: 1 H NMR (250 MHz, CDCl3) of lactam 168: N O O H S Cl O O N 168 213

PAGE 228

Spectrum 8.50: 1 H NMR (250 MHz, CDCl3) of Coenzyme A Standard 9 and Resin-Lysate Extract 193: NN NN O O P O O O P O O O HN HN HS NH2 OH O O O OH P O O O CoA 214

PAGE 229

Spectrum 8.51: HPLC of Coenzyme A 9: Coenzyme A Disulfide Coenzyme A 215

PAGE 230

Spectrum 8.52: HPLC N-Protio -Lactam 107: N O O H3C H Cl 216

PAGE 231

Spectrum 8.53: HPLC of ethylthiolated -lactam 109b: N O O H3C S CH3 Cl 217

PAGE 232

Spectrum 8.54: HPLC Purified Thioethyl Adduct: NN NN O O P O O O P O O O HN HN S NH2 OH O O O OH P O O O S 218

PAGE 233

Spectrum 8.55: HPLC Immediately after mixture of Ethylthiolated -Lactam and Coenzyme A: N O O H3C H Cl Adduct 193 219

PAGE 234

Spectrum 8.56: HPLC Immediately after mixture of sec-Butylthiolated -Lactam and Coenzyme A: N O O H3C H Cl Adduct 193 220

PAGE 235

Spectrum 8.57: HPLC Immediately after mixture of Phenylthiolated -Lactam and Coenzyme A: N O O H3C H Cl Adduct 193 221

PAGE 236

Spectrum 8.58: HPLC Immediately after mixture of ethylthiolated -Lactam and S. aureus lysate: N O O H3C H Cl Adduct 193 222

PAGE 237

Spectrum 8.59: HPLC S. aureus lysate: 223

PAGE 238

224 REFERENCES 1) Review: The Oldest Fungicide and Newest Phytoale xin a Reappraisal of the Fungitoxicity of Elemental Su lphur, Williams, J.; Cooper, R. Plant Pathology, 2004, 53, 263. 2) Derivatives of 5-Thio-D-glucose, Feather, M.S.; Whistler, R.L. Tetrahedron Letters, 1962 15, 667-668. 3) Enzyme-Catalyzed Aldol Condensation for Asymmetric Synthesis of Azasugars: Synthesis, Evaluation, and M odeling of Glycosidase Inhibitors, Kajimoto, T.; Liu, K.K.; Pederson, R.L.; Zhong, Z.; Ichikawa, Y.; Porco, J.A.; Wong, C. Journal of the American Chemical Society 1991, 113, 6187-6196. 4) Sulfoxide and Sulfone Derivatives of D-Xylothiopyranose, Whistler, R.L.; Van Es, T.; Rowell, R.M. Journal of Organic Chemistry 1965, 30, 27192721. 5) -Thio-D-ribopyranose. Part IV. Su lfoxides and Sulfones, Clayton, C.J.; Hughes, N.A.; Inch, T.D. Carbohydrate Research 1975, 45, 55-64. 6) Inhibition of Cellular Transport Pr ocesses by 5-Thio-D-glucopyranose, Whistler, R.L.; Lake, W.C. Biochemical Journal 1972, 130, 919-925. 7) Synthesis of 5-Thio-L-fucose and its Inhibitory Effect of Fucosidase, Hashimoto, H.; Fujimori, T.; Yuasa, H. Journal of Carbohydrate Chemistry 1990, 9, 683-694. 8) -Thio-D-mannose from the Marine Sponge Clathria pyramida (Lendenfeld). The First Example of a Naturally Occurring 5-Thiosugar, Capon, R.J.; MacLeod, J.K. Journal of the Chemical Society, Chemical Communications, 1987, 15, 1200-1. 9) Roles of Thiol-Redox Pathways in Bacteria, Ritz, D.; Beckwith, J. Annual Review of Microbiology. 2001, 55, 21. 10) Transfer of Electrons Across the Cytoplasmic Membrane by DsbD, a Membrane Protein Involved in Thio l-Disulphide Exchange and Protein Folding in the Bacterial Periplasm, Chung, J.; Chen, T.; Missiakas, D. Molecular Microbiology, 2000, 35, 1099-1109.

PAGE 239

225 11) Calichemicins, a Novel Family of Antitumor Antibiotics. 1. Chemistry and Partial Structure of Calicheamicin 1 I , Lee, M.D.; Dunne, T.S.; Siegel, M.M.; Chang, C.C.; Morton, G.O.; Borders, D.B. Journal of the American Chemical Society 1987 109, 3464-3466. 12) Namenamicin, a New Enediyne Antitumor Antibiotic from the Marine Ascidian Polysyncraton lithostrotum , McDonald, L.; Capson, T.; Krishnamurthy, G.; Ding, W.; Ellestad, G.; Bernan, V.; Maiese, W.; Lassota, P.; Discafani, C.; Kramer, R.; Ireland, C. Journal of the American Chemical Society 1996 118, 10898-10899. 13) Calichemicins, a Novel Family of Antitumor Antibiotics. 2. Chemistry and Structure of Calicheamicin 1 I , Lee, M.D.; Dunne, T.S.; Siegel, M.M.; Chang, C.C.; Morton, G.O.; Borders, D.B.; McGahren, W.J. Journal of the American Chemical Society 1987, 109, 3466-3468. 14) Total Synthesis of Calicheamicin 1 I 1. Synthesis of the Oligosaccharide Fragment, Groneberg, R.D.; Miyazaki, T.; Stylianides, N.A.; Schulze, T.J.; Stahl, W.; Schreiner, E.P.; Suzuki, T.; Iwabuchi, Y.; Smith, A.L.; Nicolaou, K.C. Journal of the American Chemical Society 1993, 115, 7593-7611. 15) Total Synthesis of Calicheamicin 1 I 2. Development of an Enantioselective Route to (-)-Calicheamicinone, Smit h, A.L.; Pitsinos, E.N.; Hwang, C.K.; Mizuno, Y.; Saimoto, H.; Scarlato, G. R.; Suzuki, T.; Nicolaou, K.C. Journal of the American Chemical Society 1993, 115, 7612-7624. 16) Total Synthesis of Calicheamicin 1 I 3. The Final Stages, Nicolaou, K.C.; Hummel, C.W.; Nakada, M.; Shibayama, K.; Pitsinos, E.N.; Saimoto, H.; Mizuno, Y.; Baldenius, K.U.; Smith, A.L. Journal of the American Chemical Society 1993 115, 7625-7635. 17) Bacterial Thiol-disulfide Redox Meta bolism: A New Target for Rational Drug Design, Cohen, G.; Borovok, I.; Uzie l, O.; Schreiber, R.; Aharonowitz, Y. Periodicum Biologorum 2001, 103, 153-155. 18) N -Thiolated -Lactams: Chemistry and Biology of a Novel Class of Antimicrobial Agents for MRSA, Long, T.E., Ph.D. Dissertation, University of South Florida, 2003, 65 19) Rate Constants and Equilibrium Constants for Thiol-Disulfide Interchange Reactions Involving Oxidized Glutathione , Szajewski, R.; Whitesides, G. Journal of the American Chemical Society 1980 10, 2011-2026.

PAGE 240

226 20) Coenzyme A Disulfide Reductase, the Primary Low Molecular Weight Disulfide Reductase from Staphylococcus aureus, delCardayre, S.; Stock, K.; Newton, G.; Fahey, R.; Davies, J. Journal of Biological Chemistry 1998, 273, 5744-5751. 21) Coenzyme A Thiosulfonate (Coenzyme A Disulfide-S,S-Dioxide), an Affinity Analog of Coenzyme A, Mish imura, J.; Mitchell, T.; Hill, K.; Collier, G. Journal of Biological Chemistry 1982 257, 14896-14902. 22) Energy Coupling in Microbial Transport, Hamilton, W. Advances in Microbial Physiology 1975, 12, 1-53. 23) Distribution of Thiols in Microorganisms: Mycothiol Is a Major Thiol in Most Actinomycetes, Newton, G.; Arnold, K.; Price, M.; Sherrill, C.; delCardayre, S.; Aharonowitz, Y.; Cohen, G.; Davies, J.; Fahey, R.; Davis, C. Journal of Bacteriology 1996, 178, 1990-1995. 24) The Bacterial Phospho enolpyruvate : Sugar Phosphotransferase System, Roseman, S.; Postma, P.W. Biochimica et Biophysica Acta 1976, 457, 213257. 25) Antimicrobial Properties of Allium sativum (garlic), Harris, J.C.; Cottrell, S.L.; Lloyd, D. Applied Microbiology and Biotechnology 2001 57, 282-286. 26) Antiparasitic Activity of Diallyl Tr isulfide (Dasuansu) on Human and Animal Pathogenic Protozoa (Trypanosoma sp., Entamoeba histolytica and Giardia lamblia ) in vitro , Lun, Z.R.; Burri, C.; Menzinger, M.; Kaminsky, R. Annales de la Societe Belge de Medecine Tropicale 1994, 74, 51-59. 27) The Organosulfur Chemistry of the Genus Allium Implications for the Organic Chemistry of Sulfur, Block, R. Angewandte Chemie 1992, 31, 1135-1178. 28) Allicin, the Antibacterial Principle of Allium sativum Isolation, Physical Properties and Antibacterial Action, Cavallito, C.; Bailey, J.H. Journal of the American Chemical Society, 1944, 66, 1944-1952. 29) The Mode of Action of Allicin: Trapping of Radicals and Interaction with Thiol Containing Proteins, Rabinkov, A.; Miron, T.; Konstantinovski, L.; Wilchek, M.; Mirelman, D.; Weiner, L. Biochimica et Biophysica Acta 1998, 1379, 233-244. 30) HPLC of Allicin and Other Thiosulfin ates in Garlic Clove Homogenates, Lawson, L.D.; Wood, S.G.; Hughes, B.G. Planta Medica 1991, 57, 263-270.

PAGE 241

227 31) Inhibition of Microbial Growth by Ajoene, a Sulfur-Containing Compound Derived from Garlic, Naganawa, R.; Iwata, N.; Keiko, I.; Fukuda, H.; Tsuchiyoshi, F.; Suzuki, A. Applied and Environmental Microbiology 1996, 62, 4238-4242. 32) Garlic Extract and Two Diallyl Sulphides Inhibit Methicillin-Resistant Staphylococcus aureus Infection in BALB/cA Mi ce, Tsao, S.; Cheng-chin, H.; Yin, M. Journal of Antimicrobial Chemotherapy 2003, 52, 974-980. 33) Antimicrobial Activity of Some Thiol-Containing Heterocycles, Ghannoum, M.A.; Eweiss, N.F.; Bahajaj, A.A.; Qureshi, M.A. Microbios 1983, 37, 151-159. 34) -Mercaptodeoxyuridine Its Enzymatic Synthesis and Mode of Action in Microbiological Systems, Baranski, K.; Bardos, T.J.; Bloch, A.; Kalman, T.I. Biochemical Pharmacology 1969, 18, 347-358. 35) Unsymmetrical Borole Complexes as Biocides: Synthetic, Structural and Biological Aspects, Saxena, C.; Singh, R.V. Journal of Inorganic Biochemistry 1995, 57, 209-218. 36) Engineering a Disulfide Bond and Free Thiols in the Lantibiotic Nisin Z, Kraaij, C.; Breukink, E.; Rollema, H.; Bongers, R.; Kosters, H.; Kruijff, B.; Kuipers, O. European Journal of Biochem istry, 2000, 267, 901-909. 37) Synthesis and Antimicrobial Activity of 5,7-Dichloroqui noline-8-thiol and Its Derivatives, Journal of Medicinal Chemistry 1970, 13, 1008-1009. 38) Synthesis, Characterizati on and Biological Activity of Organotin Derivatives of 2-Thionaphthalene, Including the Crystal Structures of (Naphthalenethiolato-S)triphenyltin(IV) and Bis(naphthalenethiolatoS)dimethyltin(IV), Applied Organometallic Chemistry 1997 11, 47. 39) Activities of Bismuth Thiols agai nst Staphylococci and Staphylococcal Biofilms, Domenico, P.; Baldassarrii, L.; Schoch, P.E.; Kaehler, K.; Sasatsu, M.; Cunha, B.A. Antimicrobial Agents and Chemotherapy 2001, 45, 14171421. 40) The Structurally Flexible Bicyclic Bis(2-hydroxyethanethiolato)bismuth(III) Complex: A Model for Asymmetric Monoa nionic Chelation of Bismuth(III), Agocs, L.; Briand, G.G.; Burford, N.; Cameron, T.S.; Kwiatkowski, W.; Robertson, K.N. Inorganic Chemistry 1997, 36, 2855-2860. 41) Gliotoxin and Related Epipolythiodi oxopiperazines, Waring, P.; Beaver, J. General Pharmacology 1996, 27, 1311-1316.

PAGE 242

228 42) Total Synthesis of Gliotoxin, Dehydrogliotoxin and Hyalodendrin, Fukuyama, T.; Nakatsuka, S.; Kishi, Y. Tetrahedron 1981, 37, 2045-2078. 43) Cyclic Polysulphides From Parikia speciosa , Gmelin, R.; Susilo, R.; Fenwick, G.R. Phytochemistry, 1981 20, 2521-2523. 44) Cyclic Polysulfides from the Red Alga Chondria californica , Faulkner, D.J.; Wratten, S.J. Journal of Organic Chemistry 1976, 41, 2465-2467. 45) Isolation, Structure, and Synthesis of Lenthionine and Its Analogs, Morita, K.; Kobayashi, S. Chemical & Pharmaceutical Bulletin 1967, 15, 988-993. 46) DNA Strand Scission by the Novel Antitumor Antibiotic Leinamycin, Hara, M.; Saitoh, Y.; Nakano, H. Biochemistry, 1990, 29, 5676-5681. 47) Activation of Leinamycin by Thio ls: A Theoretical Study, Breydo, L.; Gates, K.S. Journal of Organic Chemistry 2002 67, 9054-9060. 48) Total Synthesis of (+)-Leinamycin, Fukuyama, T.; Kanda, Y. Journal of the American Chemical Society 1993, 115, 8451-8452. 49) Two Bromotyrosine-Cysteine De rived Metabolites From a Sponge, Rodriguez, A.; Akee, R.; Scheuer, P. Tetrahedron Letters, 1987, 4989-4992. 50) Isolation of Psammaplin A 11-Sulfate and Bisaprasin 11-Sulfate From the Marine Sponge Aplysinella rhax, Pham, N.; Butler, M.; Quinn, R. Journal of Natural Products 2000, 63, 393-395. 51) A Revised Structure for the Marine Bromoindole Derivative Citorellamine, Moriarty, R.; Roll, D.; Ku, Y.; Nelson, C.; Ireland, C. Tetrahedron Letters, 1987, 28, 749-752. 52) Citorellamine, a New Bromoindole Derivative From Polycitorella mariae , Roll, D.; Ireland, C. Tetrahedron Letters, 1985, 26, 4303-4306. 53) Purification and character ization of a cysteine-rich 11.5-kDa antibacterial protein from the granular haemocytes of the shore crab, Carcinus maenas, Relf, J.; Chisholm, J.; Kemp, G.; Smith, V. Eur. J. Biochemistry, 1999, 264, 350-357. 54) Enantiomeric 1,2,3-Trithiane-Cont aining Alkaloids and Two New 1,3Dithiane Alkaloids From New Zealand Ascidians, Pearce, A.; Babcock, R.; Battershill, C.; Lambert, G.; Copp, B. Journal of Organic Chemistry 2001 66, 8257-8259.

PAGE 243

229 55) Lissoclinotoxins: Antibiotic Polysulf ur Derivatives From the Tunicate Lissoclinum perforatum Revised Structure of Li ssoclinotoxin A, Litaudon, M.; Trigalo, F.; Martin, M. ; Frappier, F.; Guyot, M. Tetrahedron, 1994, 50, 5323-5334. 56) N-Thiolated -Lactams, Long, T.E.; Turos, E. Current Medicinal Chemistry Anti-Infective Agents 2002, 1, 251-268. 57) The Mode of Action of 1,2-Benzisothiazolin-3-one on Staphylococcus aureus , Fuller, S.; Denyer, S.; Hugo, W.; Pemberton, D.; Woodcock, P.; Buckley, A. Letters in Applied Microbiology 1985, 1, 13-15. 58) In vitro Activity of Lansoprazole Against Helicobacter pylori , Figuraa, N.; Crabtreeb, J.; Dattiloc, M. Journal of Antimicrobial Chemotherapy 1997 39, 585. 59) Review Article: The Control of Gastric Acid and Helicobacter pylori Eradication, Sachs, G.; Sh in, J.; Munsun, K; Vagin, O.; Lambrecht, N.; Scott, D.; Weeks, D.; Melchers, K. Alimentary Pharmacology & Therapeutics 2000, 14, 1383-1401. 60) Antifungal Activity of Ajoene Derived From Garlic, Yoshida, S.; Kasuga, S.; Hayashi, N.; Ushiroguchi, T.; Matsuura, H.; Nakagawa, S. Applied Environmental Microbiology 1987, 53, 615-617. 61) Design, Synthesis and Evaluation of Novel 1,4-Naphthoquinone Derivatives as Antifungal and Anticancer Agents, Tandon, V.K.; Chhor, R.B.; Singh, R.V.; Rai, S.; Yadav, D.B. Bioorganic & Medicinal Chemistry Letters 2004, 14, 1079-1083. 62) In vitro-in vivo Antifungal Evaluation and Stru cture-Activity Relationships of 3H-1,2-Dithiole-3-thi one Derivatives, Giannini, F.A.; Aimar, M.L.; Sortino, M.; Gomez, R.; Sturniollo, A.; Juarez, A.; Zacchino, S.; de Rossi, R.H.; Enriz, R.D. IL FARMACO 2004, In Press. 63) Systemic Antifungal Agents Against AIDS -Related Opportunistic Infections: Current Status and Emerging Drugs in Development, Ablordeppey, S.Y.; Fan, P.; Ablordeppey, J.H.; Mardenborough, L. Current Medicinal Chemistry 1999, 6, 1151-1195. 64) One-pot Synthesis of 3H-1,2-Dith iole-3-thione Derivatives from Dithiolmalonic Esters, Aimar, M.L.; Kreiker, J.; de Rossi, R.H. Tetrahedron Letters, 2002 43, 1947.

PAGE 244

230 65) Espermicins, a Class of Potent Antitumor Antibiotics: Mechanism of Action, Long, B.H.; Golik, J.; Foren za, S.; Ward, B.; Rehfuss, R.; Dabrowiak, J.C.; Catino, J.J.; Musial, S.T.; Brookshire, K.W.; Doyle, T.W. Proceedings of the National Academy of Sciences, 1989, 86, 2-6. 66) Five New Alkaloids From the Tropical Ascidian, Lissoclinum sp. Lissoclinotoxin A is Chiral, Searle, P.; Molinski, T. Journal of Organic Chemistry, 1994, 59, 6600-6605. 67) Free Thiol Groups are Essential for In fectivity of Human Cytomegalovirus, Mirazimi, A.; Mousavi-Jazi, M.; Sundqvist, V.; Svensson, L. Journal of General Virology 1999 80, 2861-2865. 68) Sulfurand Seleno-Sugar Modified Nu cleosides. Synthesis, Chemical Transformations and Biological Properties, Wnuk, S.F. Tetrahedron 1993, 49, 9877-9936. 69) Antimicrobial Properties of Allicin fr om Garlic, Ankri, S.; Mirelman, D. Microbes and Infection 1999, 2, 125-129. 70) In vitro Virucidal Effects of Allium sativum (Garlic) Extract and Compounds, Weber, N.D.; Andersen, D.O.; North, J.A.; Murray, B.K.; Lawson, L.D.; Hughes, B.G. Planta Medica 1992 8, 417-423. 71) Inhibition of Influenza In fection by Glutathione, Cai, J.; Chen, Y.; Seth, S.; Furukawa, S.; Compans, R.W.; Jones, D.P. Free Radical Biology & Medicine 2003, 34, 928-936. 72) Synthesis and Biological Evalua tion in Human Monocyte-Derived Macrophages of N -(N -Acetyl L-cysteinyl)S -acetylcysteamine Analogues with Potent Antioxidant and Anti-HIV Activities, Oiry, J.; Mialocq, P.; Puy, J.; Fretier, P.; Dereuddre-Bosquet, N.; Dormont, D.; Imbach, J.; Clayette, P. Journal of Medicinal Chemistry 2004, 47, 1789-1795. 73) NAC/MEA Conjugate: a New Potent Antioxidant Which Increases the GSH Levels in Various Cell Lines, Oiry, J.; Mialocq, P.; Puy, J.Y.; Fretier, P.; Clayette, P.; Dormont, D.; Imbach, J.L. Bioorganic & Medicinal Chemistry Letters, 2001 11, 1189-1191. 74) Synthesis, Chemistry, and Biological Activity of 5-Thiocyanatopyrimidine Nucleosides as Potential Masked Thiols, Nagamachi, T.; Fourrey, J.; Torrence, P.F.; Waters, J.A.; Witkop, B. Journal of Medicinal Chemistry 1974, 17, 403-406.

PAGE 245

231 75) Mode of Inactivation of Arenaviruses by Disulfide-Based Compounds, Garcia, C.C.; Candurra, N.A.; Damonte, E.B.; Antiviral Research 2002, 55, 437-446. 76) Anti-HIV Activity and Mechanism of Action of Macrocyclic Diamide SRRSB3, Mahmood, N.; Jhaumeer-Lauloo, S.; Sampson, J.; Houghton, P.J. Journal of Pharmacy and Pharmacology 1998, 50, 1339-1342. 77) Antipoliovirus Activity of Isothiazole Derivatives: Mode of Action of 5,5Diphenyl-3,3-diisothiazole Disulfide (DID), Garozzo, A.; Pinizzotto, M.R.; Guerrera, F.; Tempera, G.; Castro, A.; Geremia, E. Archives of Virology 1994, 135, 1-11. 78) In vitro Antiviral Activity of Four Isothiaz ole Derivatives Against Poliovirus Type 1, Pinizzotto, M.R.; Garozzo, A. ; Guerrera, F.; Castro, A.; LaRosa, M.G.; Furneri, P.M.; Geremia, E. Antiviral Research 1992, 19, 29-41. 79) Deoxyribonucleoside 2or 3Mixed Disulfides: Prodrugs To Target Ribonucleotide Reductase and/or To Inhibit HIV Reverse Transcription, Roy, B.; Chambert, S.; Lepoivre, M.; Aube rtin, A.; Balzarini, J.; Decout, J. Journal of Medicinal Chemistry 2003, 46, 2565-2568. 80) -Mercapto-2,3-dideoxynucleotides ar e Highly Effective Terminators of DNA Synthesis Catalyzed by HIV Reverse Transcriptase, Yuzhakov, A.A.; Chidgeavadze, Z.G.; Beabealashvilli, R. FEBS Letters, 1992, 306, 185-188. 81) Inactivation of Escherichia coli Ribonucleotide Reducta se by 2-Deoxy-2mercaptouridine 5-Diphosphate. Electr on Paramagnetic Resonance Evidence for a Transient Protein Perthiyl Radical, Coves, J.; de Fallois, L.; Pape, L.; Decout, J.; Fontecave, M. Biochemistry 1996, 35, 8595-8602. 82) Inhibition of Growth of Entamoeba histolytica by Allicin, the Active Principle of Garlic Extract ( Allium sativum ), Mirelman, D.; Monheit, D.; Varon, S. Journal of Infectious Disease 1987, 156, 243-244. 83) Allicin From Garlic Strongly Inhibits Cysteine Proteinases and Cytopathic Effects of Entamoeba histolytica , Ankri, A.; Rabinkov, A.; Wilchek, M.; Mirelman, D. Antimicrobial Agents and Chemotherapy 1997 10, 2286-2288. 84) Inhibition of Phosphatidylcholine Bios ynthesis and Cell Proliferation in Trypanosoma cruzi by Ajoene, an Antiplatelet Compound Isolated From Garlic, Urbina, J.A.; Marchan, E.; L azardi, K.; Visbal, G.; Apitz-Castro, R.; Gil, F.; Aguirre, T.; Piras, M.; Piras, R. Biochemical Pharmacology 1993 45, 2381-2387.

PAGE 246

232 85) Contribution of D ithiol Ligands to in vitro and in vivo Trypanocidal Activities of Dithiaarsanes and Inves tigation of Ligand Exchange in an Aqueous Solution, Loiseau, P.M.; Lubert, P.; Wolf, J. Antimicrobial Agents and Chemotherapy, 2000 44, 2954-2961. 86) The Microaerophilic Flagellate Giardia intestinalis: Allium sativum (Garlic) is an Effective Antigiardial, Harris, J.C.; Plummer, S.; Turner, M.P.; Lloyd, D. Microbiology, 2000, 146, 3119-3127. 87) Anthelmintic Composition, Gallay, J.; Kuehne, M.; Meyer, A.; Rechsteiner, O.; Schellenbaum, M. German Patent DE 78-2815621 19780411, 1978. 88) New and Unusual Sesquiterp enes: Kelsone, Prespatane, Epi-gurjunene, and T-Cadinthiol, From th e Tropical Marine Sponge Cymbastela hooperi, Konig, G.; Wright, A. Journal of Organic Chemistry 1997, 62, 3837-3840. 89) Synthesis and in vitro Antihelmintic Properties of Some New Dithiaarsanes, Loiseau, P.M.; Lubert, P.; Wolf, J.G. Arzneimittel-Forschung 1999, 49, 944950. 90) Synthesis and Fasciolicidal Activi ty of 5-Chloro-2-methylthio-6-(1naphthyloxy)-1 H -benzimidazole, Hernandez-Campos, A.; Ibarra-Velarde, F.; Vera-Montenegro, Y.; Rivera-Fernandez, N.; Castillo, R. Chemical & Pharmaceutical Bulletin 2002, 50, 649-652. 91) Biologically Oriented Organic Su lfur Chemistry. 14. Antiinflammatory Properties of Some Aryl Sulfides, Sulf oxides, and Sulfones, Brannigan, L.H.; Hodge, R.B.; Field, L. Journal of Medicinal Chemistry 1976 19, 798-802. 92) Biologically Active Sulphur Compounds of Plant Origin, Haq, K.; Ali, M. Sulphur in Plants 2003, 375-386. 93) The Coming Plague, Garrett, L., Penguin Books: New York, USA 1994. 94) Oxidative N -Dearylation of 2-Azetidinones. P -Anisidine as a Source of Azetidinone Nitrogen, Kronentha l, D.; Han, C.; Taylor, M. Journal of Organic Chemistry, 1982 47, 2765-2768. 95) Enantioselective Synthesis of Thio ls by Thione-Thiol Rearrangement Catalyzed by Optically Active Pyridine N-Oxides, Diana, M.; Marchetti, M.; Melloni, G. Tetrahedron: Asymmetry 1995, 6, 1175-1179. 96) The Interaction Between Coumarin Drugs and DNA Gyrase, Maxwell, A. Molecular Microbiology, 1993, 9, 681-686.

PAGE 247

233 97) Mass Spectrometric Characterization of a Three-enzyme Tandem Reaction for the Assembly and Modification of the Novobiocin Skeleton, Pi, N.; Meyers, C.L.F.; Pacholec, M.; Walsh, C.T.; Leary, J.A. Proceedings of the National Academy of Sciences 2004, 101, 10041. 98) Antibiotic Susceptibility of Isolates of Bacillus anthracis a Bacterial Pathogen with the Potential to Be Used in Biowarfare, Jones, M.E.; Goguen, J.; Critchley, I.A.; Draghi, D.C.; Karlowsky, J.A.; Sahm, D.F.; Porschen, R.; Patra, G.; DelVecchio, V.G. Clinical Microbiology and Infection 2003, 9, 984-986. 99) Studies on Nonconventi onally Fused Bicyclic -Lactams, Ren, X.; Konaklieva, M.; Shi, H.; Dickey, S.; Lim, D.; Gonzalez, J.; Turos, E. Journal of Organic Chemistry 1998, 63, 8898-8917. 100) Synthesis and Structure-Activ ity Studies of N-Thiolated -Lactam Antibacterials: Synthesis of Novel Oxazolidinone-Lactam Hybrids and the Efforts Toward the Formal Total Synt hesis of (+)and (-)-Cytoxazone and (+)and (-)-epi-Cytoxazone, Coates, C. Ph.D. Dissertation, University of South Florida 2004. 101) -Lactams from Esters and Sulfenimines: A New Route to Monobactams, Burnett, D.; Hart, D.; Liu, J. Journal of Organic Chemistry 1986, 51, 19291930. 102) A General Method for the Enantiospeci fic Synthesis of Optically Active Aliphatic Sulfenyl Chlorides and Thi ophthalimides, Cevasco, G.; Narisano, E.; Thea. S.; Tetrahedron: Asymmetry 1990, 1, 141-142. 103) Lipase-catalyzed Resolution of 4-Aryl-substituted -Lactams: Effect of Substitution on the 4-Aryl Ring, Carr, J.; Al-Azemi, T.; Long, T.; Shim, J.; Coates, C.; Turos, E.; Bisht, K. Tetrahedron, 2003, 59, 9147-9160. 104) Synthesis and Induction of Apoptos is in B Cell Chronic Leukemia by Diosgenyl 2-Amino-2-deoxyD -glucopyranoside Hydrochloride and Its Derivatives, Myszka, H.; Bednarczyk, D.; Najder, M.; Kaca, W. Carbohydrate Research 2003, 338, 133-141. 105) -Alkylated Monobactams Chiral Synthesis and Antibacterial Activity, Cimarusti, C.; Bonner, D.; Chang, H.; Fritz, A.; Floyd, D.; Kissick, T.; Koster, W.; Kronenthal, D.; Massa, F.; Mueller, R.; Pluscec, J.; Slusarchyk, W.; Sykes, R.; Taylor, M.; Weaver, E. Tetrahedron 1983, 39, 2577-2589.

PAGE 248

234 106) Catalysis by Surfactants of Nucleophilic Displacement Reactions of Primary Alkyl Halides with Solid-phase Nucleophile Salts, Jursic, B. Journal of Chemical Research, Synopses, 1989, 4, 104-105. 107) A Facile, High-yielding Method for the Conversion of Halides to Mercaptans, Bieniarz, C.; Cornwell, M. Tetrahedron Letters 1993, 34, 939942. 108) Coenzyme A Disulfide Reductase, the Primary Low Molecular Weight Disulfide Reductase from Staphylococcus aureus, Davies, J.; Fahey, R.; Newton, G.; Stock, K.; delCardayre, S. Journal of Biological Chemistry 1998, 273, 5744-5751. 109) Coenzyme A-Disulfide Reductase from Staphylococcus aureus: Evidence for Asymmetric Behavior on Interaction w ith Pyridine Nucleotides, Luba, J.; Charrier, V.; Claiborne, A. Biochemistry, 1999, 38, 2725-2737. 110) A Novel -Lactam Antibiotic Activates Tu mor Cell Apoptotic Program by Inducing DNA Damage, Smith, S.; Kazi, A.; Smith, L.; Long, T.; Heldreth, B.; Turos, E. Dou, Q. Molecular Pharmocology, 2002, 61, 1348-1358. 111) N -Thiolated -Lactams: Chemistry and Biology of a Novel Class of Antibacterial Agents for MRSA, Long, T. Dissertation University of South Florida, 2003

PAGE 249

ABOUT THE AUTHOR Bart Heldreth received his bachelors degree in chemistry at Kent State University, Kent, Ohio. After marrying Amy, he pursued a doctorate in chemistry in the synthetic laboratory of Professor Edward Turos. Bart continues to follow his interest in organic synthetic chemistry, as well as chemical biology.

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200421Ka 4500
controlfield tag 006 m d s
007 cr bnu
008 041209s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000511
9 035
b SE
1 100
Heldreth, Bart Allan.
0 245
N-thiolated -lactams
h [electronic resource] :
chemistry, SAR and intracellular target of a novel class of antimicrobial and anticancer agents /
by Bart Allan Heldreth.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ph.D.)--University of South Florida, 2004.
Includes bibliographical references.
Includes vita.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 249 pages.
ABSTRACT: N-Thiolated -lactams (1) represent a promising new group of compounds with potent inhibition effects on bacteria, like Bacillus anthracis and methicillin resistant Staphylococcus aureus, and onco-systems, like breast cancer and leukemia. Originally developed as part of a synthetic pathway to bicyclic lactams, N-thiolated -lactams have been shown in this laboratory to possess intriguing biological activities. The antibacterial activities of this new class of agents rely on novel structural features unlike those of any existing family of -lactam drugs. The lactams seem to exert their effects intracellularly, requiring passage of the bioactive species through the cellular membrane, rather than acting extracellularly on cell wall components in the manner of penicillin and related antibiotics.The lipophilic nature of these molecules, which lack the polar side chain functionality of all other microbially-active -lactams, suggests the compounds do not target the penicillin binding proteins within bacterial membranes but instead pass through these membranes. The biological target of these compounds has been investigated. The most active members of this -lactam class appear to be those bearing a small branched alkyl chain on the sulfur atom. The effects of stereochemistry, branching and chain length of the sulfur group on bioactivities were studied. This dissertation is divided into six chapters. A review of organosulfur anti-infectives is discussed in Chapter 1. The types of existing antibiotics and their modes of action will be discussed in Chapter 2. The synthesis of these novel agents is discussed in Chapter 3. A structure-activity relationship of these lactam analogues is discussed in Chapter 4.And Chapters 5 and 6 demonstrate a novel mode of action and biological target for these drugs using techniques which include target identification, metabolic effects, and reactivity kinetics.
Adviser: Turos, Edward.
Dissertations, Academic
x Chemistry
t USF Electronic Theses and Dissertations.
nkt 1/14/05
4 856