Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The chemistry and evaluation of porphyrin-based potential anti-cancer agents Posakony, Jeffrey Jerard 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-346129.pdf [ 14.86MB ]
Metadata
JSON: 831-1.0061519.json
JSON-LD: 831-1.0061519-ld.json
RDF/XML (Pretty): 831-1.0061519-rdf.xml
RDF/JSON: 831-1.0061519-rdf.json
Turtle: 831-1.0061519-turtle.txt
N-Triples: 831-1.0061519-rdf-ntriples.txt
Original Record: 831-1.0061519-source.json
Full Text
831-1.0061519-fulltext.txt
Citation
831-1.0061519.ris

Full Text

T H E C H E M I S T R Y A N D E V A L U A T I O N O F P O R P H Y R I N - B A S E D P O T E N T I A L A N T I - C A N C E R A G E N T S By JEFFREY JERARD POSAKONY B.Sc, Oregon State University, 1990 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May 1998 © Jeffrey J. Posakony, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract Towards the goal of synthesizing porphyrin-based radiosensitizers and hypoxia-selective cytotoxins (HSCs) leading to improved cancer treatment modalities, the chemistry of several porphyrin systems (diarylporphyrins, tetraarylporphyrins, and those based on protoporphyrin IX) was investigated. Porphyrin incorporation of nitroaromatic and heterocyclic-N-oxide moieties into porphyrins was the general objective, but additional areas were investigated. H H H H 1 3 The porphyrin precursor bis(2-pyrrolyl)methane (3) was obtained (84 %) by NaBFLt- or (80 %) LiAlH4- reduction of bis(2-pyrrolyl)thioketone (1), a substantial improvement over current methodologies. Raney Ni-reduction of 1 (-50 %) also produced the novel l,l',2,2'-tetrakis(2-pyrrolyl)ethane (4). Using 3 and an appropriate aldehyde, symmetrical diarylporphyrins incorporating pyridyl, oxidppyridyl-, nitrophenyl and phenyl substituents were synthesized. The highest yield was obtained for 5,15-diphenylporphyrin (25-40 %) which was subsequently meso-brominated and metallated (Zn); this complex was subsequently used in Stille-type Pd-catalyzed cross-coupling reaction attempts using a variety of organotin reagents. The reactivity of one such cross-coupling reaction product, [5,15-diphenyl-10,20-divinylporphyrinato]Zn(II) (DPhDVPZn, 77 %), reported previously, was explored via chemistry at the vinyl group, but most conditions attempted gave product mixtures in low yield. Reaction of DPhDVPZn with T l n i led to rapid porphyrin decomposition, but the demetallated vinyl analog, DPhDVP, reacted cleanly to afford 5,15-bis(2,2-dimethoxyethyl)-10,20-diphenylporphyrin (69 %). Condensation of pyrrole with one or two aldehydes in propionic acid (the Adler method) was used to synthesize several tetraarylporphyrins containing pyridyl or imidazolyl groups. The phenyl-pyridyl series was easiest to work with, while porphyrins containing imidazolyl groups were produced in low yield and were difficult to purify. Attempts to form porphyrins from heterocyclic, aromatic aldehydes via Lindsey conditions were generally unsuccessful and a Ill possible explanation is presented. Nitration of me5o-tettaMs(2-imidazolyl)porphyrin using several conditions was unsuccessful. Porphyrins containing one to four pyridyl groups were 'N-oxidized' with m-chloroperbenzoic acid to produce five novel (oxidopyridyl)porphyrins and seven porphyrin N-oxides. Sulfonation of the phenyl substituents 5-(l-oxido-4-pyridyl)- 10,15,20-triphenylporphyrin and c/5-5,10-bis(l-oxido-4-pyridyl)-10,20-diphenylporphyrin yielded their water-soluble derivatives, OPyTrSPhP and c-BOPyBSPhP, respectively. Some metallation reactions with Pt were pursued; tetrakis(4-sulfonatophenyl)porphyrin was readily metallated using K^PtCLi, but similar conditions with pyridyl-tris(4-sulfonatophenyl)porphyrin led to porphyrin decompositic OPyTrSPhP was metallated with K^PtCLi, but the resulting product evidendy had an externally bound Pt moiety, perhaps ligated through the oxidopyridyl group. ion. OPyTrSPhP o--*N o—c=o 21 (/ Vn-c-n-^ TirapPhTrPhP The product from tirapazamine (3-amino-l,2,4-benzotriazine-l,4-di-N-oxide) and triphosgene, 21, reacts like an isocyanate, and was used in reactions with carboxylic acids, alcohols, and amines to produce compounds (some new) incorporating tirapazamine. A tirapazamine-porphyrin conjugate (TirapPhTrPhP) was obtained from 21 and 5-(4-aminophenyl)-10,15,20-triphenylporphyrin. New substituents were introduced at the 8,13-positions of protoporphyrin IX dimethylester via the previously reported Tlm-vinyl oxidation product, and a selective deprotection of this compound's dimethyl acetal functionalities was developed. The deprotected product, dimethyl 3,7,12,17-tetramethyl-8,13-bis(2-oxoethyl)-porphyrin-2,18-dipropionate (BOEtPIXDME), was subsequently converted to derivatives incorporating aniline and C2F5CH2NH- groups via reductive IV amination chemistry; a -COOH substituted vinyl group was incorporated via a Knoevenagel reaction. Other reductive amination reactions with en, tirapazamine and NH4OAC met with limited success. Aspects of the successful reductive amination reactions are discussed. 0 2 N - < / 1ST HoC HoC H3COOC COOCH3 29 H3COOC C 0 0 C H 3 H 3 C 0 0 6 C 0 0 C H 3 BOEtPIXDME BNIMEtPIXDME The aldehyde moieties of BOEtPIXDME were reduced using NaBfii and the resulting bis(2-hydroxyethyl) product (BHEtPIXDME) was used to synthesize the 8,13-tosyl-, iodo- and bromoethyl derivatives. Novel porphyrins incorporating two 2-nitroimidazolyl (BNImEtPIXDME) or two phthalimido moieties (BPlEtPIXDME) were obtained in subsequent SN2 displacement reactions using the 8,13-bis(2-bromoethyl) derivative; elimination side-products were also observed (e.g. 29). Attempts to cleave the phthalimido groups of BPlEtPIXDME met with limited success. Acid-hydrolysis of BHEtPIXDME and BNImEtPIXDME yielded the carboxylic acid derivatives, BHEtPIX and BNImEtPIX. The solubility of BNImEtPLX was improved by formation of the 2,18-bis(L-aspartyl) amide. Attempts to incorporate other 2-nitroimidazoles and tirapazamine into derivatives of protoporphyrin IX were unsuccessful. Selected porphyrins were evaluated by cyclic voltammetry in DMF. Based on reported E1/2 and so called E 1 7 values, oxidopyridyl, nitrophenyl, tirapazamine, and 2-nitroimidazolyl substituent and porphyrin ring reduction potentials are assigned. The usefulness of porphyrins containing these substituents as radiosensitizers and HSCs is discussed by comparing these reduction potentials to those of known radiosensitizers and HSCs. In vitro assays for radiosensitization, hypoxia-selective toxicity, and photosensitization with Chinese hamster ovary cells were used to evaluate the potential of selected porphyrins incorporating oxidopyridyl (OPyTrSPhP), 2-nitroimidazolyl (BNImEtPIX) or tirapazamine (TirapPhTrPhP) substituents along with suitable 'reference' compounds. The cellular accumulation of these porphyrins was evaluated by UV-Vis spectroscopy or fluorescence microscopy. Because of the hydrophobic nature of some of these compounds, liposome-formulations were developed for three of the porphyrins whereas the others were evaluated as solutions in a-modified medium. Cremophor EL® emulsions of the hydrophobic porphyrins were also tested. The results of the assays are compared with available literature data. In general, the porphyrins were non-toxic, and they showed little radiosensitizing or photosensitizing ability; however, TirapPhTrPhP showed a modest radiation SER value (1.5). Some photosensitization was observed with OPyTrSPhP, but its effectiveness was poor in comparison to that of Photofrin II®; some evidence was found for protection from PDT-induced damaged by pyridine N-oxide. Based on the accumulation in cells measured by UV-Vis spectroscopy, BNImEtPIX was accumulated to the greatest degree, but, per microgram of porphyrin delivered, Photofrin n® was accumulated the most and the sulfonatophenyl porphyrins the least. The accumulation data for the liposome-formulated porphyrins obtained via UV-Vis measurements appear to conflict with those from the fluorescence microscopy; some possible explanations are discussed. Table of Contents Abstract ii Table of Contents vi List of Figures xvi List of Schemes xix List of Tables xix List of Abbreviations xxii Key to Chemical Compound Numbers and Abbreviations xxv Acknowledgements xl Chapter 1: Introduction to Porphyrins and Cancer Therapy 1.1 Background 1 1.2 Porphyrin Localization in Tumor Tissue 3 1.3 Tumor Hypoxia 4 1.4 Evaluation of Porphyrin-based Anti-cancer Agents 6 1.4.1 Radiosensitizer Mechanisms 6 1.4.2 Hypoxia-Selective Cytotoxin Mechanisms 8 1.4.3 Photodynamic Therapy (PDT) Mechanisms 9 1.4.3.1 Porphyrin-Nitroimidazole Interactions 12 1.5 Porphyrins and Ionizing Radiation 12 1.6 The Combination of PDT with Hypoxia-Selective Cytotoxins 14 1.7 Other Areas of Interest 16 1.7.1 Hypoxia Imaging 16 1.7.2 Pt-chelating groups 16 1.8 Thesis Preview: Synthesis and Evaluation of Porphyrin-based Anti-cancer Agents 18 vii 1.9 Conclusion 21 References for Chapter 1 22 Chapter 2: The Synthesis and Modification of Diarylporphyrins 2.1 Introduction 27 2.2 Experimental 28 2.2.1 Materials and Instrumentation 28 2.2.2 Porphyrin Precursors 29 2.2.2.1 Bis(2-pyrrolyl)thioketone (1) 30 2.2.2.2 Bis(2-pyrrolyl)ketone (2) 30 2.2.2.3 Bis(2-pyrrolyl)methane (3) 31 2.2.2.4 l,l',2,2'-Tetrakis(2-pyrrolyl)ethane (4) 33 2.2.2.5 (4-Pyridyl)bis(2-pyrrolyl)methane (5) 34 2.2.3 Diarylporphyrins 35 2.2.3.1 5,15-Diphenylporphyrin (DPhP) 35 2.2.3.2 5-Phenylporphyrin (PhP) 36 2.2.3.3 5,15-Bis(4-nitrophenyl)porphyrin (BNPhP) 36 2.2.3.4 5,15-Bis(4-pyridyl)porphyrin (BPyP) 37 2.2.3.5 5,15-Bis(l-oxido-4-pyridyl)porphyrin (BPyNOP) 37 2.2.3.6 3,7,13,17-Tetraethyl-2,8,12,18-tetramethyl-10,20-bis(4-pyridyl)porphyrin (TEtTMeBPyP) 38 2.2.3.7 Meso-linked DPhP Dimer 40 2.2.4 Preparation of Reagents for Pd-catalyzed Cross-coupling Reactions 40 2.2.4.1 5,15-Dibromo-10,20-diphenylporphyrin (DBrDPhP) 40 2.2.4.2 [5,15-Dibromo-10,20-diphenylporphyrinato]zinc(II) (DBrDPhPZn) 41 2.2.4.3 Tris(n-butyl)(l-imidazolyl)tin(IV) (10) 42 2.2.4.4 Tris(n-butyl)phenyltin(IV) (11) 4? 2.2.4.5 Diethyl 2-(tris(n-butyl)stannyl)-l,3-dipropanoate (12) 43 2.2.4.6 Tris(n-butyl)(4-pyridyl)tin(IV) (13) 43 2.2.4.7 Tris(n-butyl)vinyltin(IV) (14) 43 2.2.5 Cross-Coupling Experiments 44 2.2.5.1 (Tetrakis(phenyl)porphyrinato)zinc(II) (TPhPZn) 44 2.2.5.2 (5,15-Diphenyl-10,20-divinylporphyrinato)zinc(II) (DPhDVPZn) 45 2.2.5.3 Reaction of DBrDPhPZn with 10 46 2.2.5.4 Reaction of DBrDPhPZn with 12 46 2.2.5.5 Reaction of DBrDPhPZn with 13 47 2.2.5.6 Reaction of DBrDPhPZn with Hexamethylditin 47 2.2.6 Reactivity of DPhDVPZn 47 2.2.6.1 Treatment with HBr 47 2.2.6.2 Treatment with HBr and N-based Nucleophiles 48 2.2.6.3 5,15-Diphenyl-10,20-divinylporphyrin (DPhDVP) 50 2.2.6.4 Reactions with T l m : 5,15-Bis(2,2-dimethoxyethyl)-10,20-diphenylporphyrin (BDMEtDPhP) 51 2.2.6.5 Attempts at Conjugate Addition 52 2.2.6.6 Hydroboration-halogenation 54 2.2.6.7 Other Reactions 55 2.3 Results and Discussion 56 2.3.1 Porphyrin Precursors 56 2.3.2 Synthesis of DPhP c n 2.3.3 Other Diarylporphyrins 2.3.4 Derivitization of DPhP 2.3.5 Pd-catalyzed Cross-coupling Reactions 2.3.6 Denominated and Meso-Cl Porphyrins 2.3.7 Reactions with DPhDVPZn 71 2.4 Summary 78 References for Chapter 2 79 Chapter 3: The Synthesis and Modification of Tetraarylporphyrins 3.1 Introduction 85 3.2 Experimental 85 3.2.1 Porphyrin Precursors 86 3.2.1.1 4(5)-Hydroxymethylimidazole (19) 86 3.2.1.2 4(5)-Imidazolecarboxaldehyde (20) 87 3.2.2 Tetraarylporphyrins 87 3.2.2.1 Mixed-aldehyde Condensation 87 3.2.2.2 TPyP Via Lindsey Conditions 88 3.2.2.3 Tetrakis(2-imidazolyl)porphyrin (2-TImP) 89 3.2.2.4 Nitration of 2-TImP 90 3.2.2.5 Tetrakis(4(5)-imidazolyl)porphyrin (4-TImP) 91 3.2.2.6 Mixed-aldehyde Condensations with Pyrrole, Imidazolecarboxaldehyde and Pyridinecarboxaldehyde 92 3.2.3 N-Oxidations 93 3.2.3.1 5-(l-Oxido-4-pyridyl)-10,15,20-triphenyl-20)porphyrin (OPyTrPhP) 94 3.2.3.2 5-(l-Oxido-4-pyridyl)-10,15,20-triphenylporphyrin-21-oxide (OPyTrPhP-210), and -23-oxide (OPyTrPhP-230) 95 3.2.3.3 5,15-Bis(l-oxido-4-pyridyl)-10,20-diphenylporphyrin (t-BOPyDPhP) 95 3.2.3.4 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin (c-BOPyDPhP) 3.2.3.5 5,10-Bis( 1 -oxido-4-pyridyl)-15,20-diphenylporphyrin-21 -oxide (c-BOPyDPhP-210), -22-oxide (c-BOPyDPhP-220), and -24-oxide (c-BOPyDPhP-240) 96 3.2.3.6 5,10,15-Tris(l-oxido-4-pyridyl)-20-phenylporphyrin (TrOPyPhP) 97 3.2.3.7 5,10,15-Tris(l-oxido-4-pyridyl)-20-phenylporphyrin-21-oxide (TrOPyPhP-210), and -22-oxide (TrOPyPhP-220) 97 3.2.3.8 5,10,15,20-Tetrakis(l-oxido-4-pyridyl)porphyrin (TOPyP)... .98 3.2.3.9 Reaction of m-CPBA with TPhP 99 3.2.4 Sulfonations 106 3.2.4.1 Sodium 5,10,15,20-Tetrakis(4-sulfonatophenyl)porphyrin (TSPhP) 106 3.2.4.2 Sodium 5,10,15-Tris(4-sulfonatophenyl)-20-(4-pyridyl)porphyrin (PyTrSPhP) 106 3.2.4.3 Sodium 5-(l-Oxido-4-pyridyl)-10,15,20-tris(4-sulfonatophenyl)porphyrin (OPyTrSPhP) 106 3.2.4.4 Sodium 5,10-Bis(l-oxido-4-pyridyl)-15,20-bis(4-sulfonatophenyl)porphyrin (c-BOPyBSPhP) 107 3.2.5 Platination Reactions 107 3.2.5.1 Sodium [5,10,15,20-Tetrakis(4-sulfonatophenyl)porphyrinato]Pt(II) (TSPhPPt) 108 3.2.5.2 The Reaction of K 2PtCl4 with PyTrSPhP 108 3.2.5.3 The Reaction of K^PtCLi with OPyTrSPhP ("OPyTrSPhPPt"). 109 3.2.6 Chemistry of 3-Amino-1,2,4-benzotriazine-1,4-di-N-oxide (tirapazamine) I l l 3.2.6.1 2H-[l,2,4]Oxadiazolo[3,2-c][l,2,4]benzotriazin-2-one-5-oxide (21) I l l 3.2.6.2 3-N-Acetamido-l,2,4-benzotriazine-l,4-di-N-oxide (22) 112 3.2.6.3 Butyl (N-(l,4-dioxido-l,2,4-benzotriazin-3-yl)carbamate (23) 112 3.2.6.4 Reaction of 21 With Amines 112 3.2.6.5 Reaction of 21 with Porphyrins 113 3.2.7 Other Reactions Involving Pt and Ru Species 117 3.3 Results and Discussion 118 3.3.1 Tetra-arylporphyrin Syntheses 118 3.3.2 (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides 120 3.3.2.1 Synthesis and General Observations 121 3.3.2.2 X-ray Crystallography 122 3.3.2.3 iH-NMR Spectra 123 3.3.2.4 UV-Vis Spectroscopy 127 3.3.2.5 Infrared Spectroscopy 130 3.3.2.6 Mass Spectroscopy 130 3.3.3 Sulfonation Reactions 132 3.3.4 Platination Reactions 132 3.3.5 Tirapazamine Chemistry 133 3.4 Summary 136 References for Chapter 3 138 Chapter 4: Substituent Manipulation on Protoporphyrin IX 4.1 Introduction 142 4.2 Experimental 143 4.2.1 Dimethyl 8,13-Bis(2,2-dimethoxyethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BDMEtPIXDME) 143 4.2.2 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-oxoethyl)-porphyrin-2,18-dipropionate (BOEtPIXDME) 145 4.2.3 Reductive Amination Chemistry 146 4.2.3.1 Dimethyl 8,13-Bis(2-(N-anilino)ethyl)-3,7,12,17-teu-amethylporphyrin-2,18-dipropionate (BAnEtPIXDME) 148 4.2.3.2 Dimethyl 8,13-Bis(2-(N-3,3,3,2,2-pentafluoropropylamino)ethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BF5EtPIXDME) 147 4.2.3.3 Attempted Reductive Amination with Tirapazamine to Produce 26 149 4.2.3.4 Attempted Reductive Amination with NH40Ac to Produce 27 .150 4.2.3.5 Attempted Reductive Amination with en to Produce 28 151 4.2.4 Dimethyl 3,7,12,17-Tetramethylporphyrin-8,13-bis(£'/Z 2-propenoic acid)-2,18-dipropionate (BPrAPIXDME) and Tetramethyl 3,7,12,17-Tetramethylporphyrin-2,18-dipropionate-8,13-bis(£7Z2-propenoate) (BPrEPIXDME) 152 4.2.5 Dimethyl 8,13-Bis(2-hydroxyethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BHEtPIXDME) 152 4.2.6 8,13-Bis(2-hydroxyethyl)-3,7,12,17-tetramethyl-porphyrin-2,18-dipropionic acid (BHEtPIX) 153 4.2.7 Dimethyl 8,13-Bis(2-bromoethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BBrEtPlXDME) 153 4.2.8 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-(4-toluenesulfonyl)oxoethyl)-porphyrin-2,18-dipropionate (BTsEtPDCDME) 154 4.2.9 Dimethyl 8,13-Bis(2-iodoethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BIEtPlXDME) 155 4.2.10 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-(2-nitroimidazol-1 -yl)ethyl)-porphryin-2,18-dipropionate (BNImEtPlXDME) 155 4.2.11 3,7,12,17-tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yl)ethyl)-porphyrin-2,18-dipropionic acid (BNImEtPIX) 157 xni 4.2.12 Bis(N-L-Aspartyl) 3,7J2J7-Tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yl)ethyl)-porphyrin-2,18-dipropionamide (BNImEtPIXBNA) 158 4.2.13 Dimethyl 3,7,12,17-Tetramemyl-8J3-bis((2-phthalimido)ethyl)-porphyrin 2,18-dipropionate (BPlEtPIXDME) 160 4.2.14 Attempted Removal of Phthalimido Groups From BPlEtPIXDME to Produce 27 161 4.2.15 Other Reactions 162 4.2.15.1 Oxidation of BOEtPIXDME 162 4.2.15.2 Reaction of PIXDME with en to Produce 33 163 4.2.15.3 Attempted Synthesis of PIX- and PIXDME-Tirapazamine Conjugates 164 4.2.15.4 Attempted Synthesis of a PIX-EF5 Conjugate 165 4.2.15.5 Treatment of PIXDME with HBr and Imidazole 166 4.3 Results and Discussion 171 4.3.1 The Production and Reactions of BOEtPIXDME 172 4.3.1.1 Synthesis of BOEtPIXDME 172 4.3.1.2 Reductive amination 173 4.3.1.3 Synthesis of BPrAPIXDME and BPrEPIXDME 175 4.3.2 The Synthesis of BHEtPIXDME and Its Derivatives 176 4.3.2.1 BHEtPIXDME and BHEtPIX 176 4.3.2.2 BTsEtPIXDME, BIEtPIXDME, and BBrEtPIXDME 177 4.3.2.3 BNImEtPIXDME, BNImEtPIX, and BNImEtPIXBNA-(tBu)2 Ester 177 4.3.2.4 BPlEtPIXDME and Attempts at Phthalimide Cleavage 181 4.3.3 Other Reactions 184 4.4 Summary 185 References for Chapter 4 187 XIV Chapter 5: The Cyclic Voltammetry of Selected Porphyrins 5.1 Introduction 190 5.2 Experimental 190 5.3 Results and Discussion 192 5.3.1 Porphyrin Electrochemistry 192 5.3.2 General Observations 194 5.3.3 Interpretation of Results 194 5.3.4 Correlation with In Vitro Efficacy 196 5.4 Summary 199 References for Chapter 5 200 Chapter 6: In Vitro Evaluation of Selected Porphyrins 6.1 Introduction 202 6.2 Experimental 202 6.2.1 Materials and Methods 203 6.2.2 Preparation of Solutions and Formulations 205 6.2.2.1 Media Solutions 205 6.2.2.2 Liposomal Formulations 206 6.2.2.3 Other Formulations 207 6.2.3 Toxicity Assays 208 6.2.4 Radiosensitizing Assays 209 6.2.5 Photosensitizing Assays 210 6.2.6 Porphyrin Accumulation in Cells 211 6.2.7 Fluorescence Microscopy 212 6.3 Results and Discussion 213 6.3.1 Porphyrin Formulations 213 6.3.2 Toxicity 214 6.3.3 Radiosensitization 215 6.3.4 Photosensitization 218 6.3.5 Porphyrin Accumulation in Cells 222 6.3.6 Fluorescence Microscopy 223 6.3.7 Discussion of In Vitro Results 231 6.4 Summary 232 References for Chapter 6 233 Chapter 7: Conclusions and Future Work 7.1 General Remarks 237 7.2 Diarylporphyrins (Chapter 2) 237 7.3 Tetraarylporphyrins (Chapter 3) 239 7.3.1 Porphyrin Syntheses, Sulfonations and Metallations with Pt 239 7.3.2 (Oxidopyridyl)porphyrins 240 7.3.3 Tetra(nitroimidazolyl)porphyrins 241 7.3.4 Tirapazamine Chemistry 242 7.4 Protoporphyrin IX (PIX)-Based Chemistry (Chapter 4) 243 7.5 Development of Porphyrin-based Anti-cancer Agents 244 7.5.1 Results of the In Vitro Assays at BCCRC (Chapter 6) and Cyclic Voltammetry Studies (Chapter 5) 244 7.5.2 Suggestions for Improved Porphyrin-Based Hypoxia-Selective Cytotoxins and Radiosensitizers 245 7.5.3 Hypoxia-Imaging Agents and Pt-Porphyrins 247 7.6 Conclusion 247 References for Chapter 7 248 APPENDIX - -List of Figures Figure i . Porphyrin Position Numbering xxiv Figure 1.1. Schematic Representation of a Portion of a Hypoxic Tumor 5 Figure 1.2. Proposed Mechanisms for O2 Radiosensitization 6 Figure 1.3. Typical Results for X-ray Irradiation of Chinese Hamster Ovary Cells Under Aerobic (O2) or Hypoxic (N2) Conditions 7 Figure 1.4. Activation of a Hypoxia-Selective Cytotoxin (HSC), 8 Figure 1.5. Activation of a Photosensitizer and Subsequent Type I and Type II Photoprocesses 10 Figure 1.6. The Synthesis of DPhDVPZn 19 Figure 2.1. X-ray Crystal Structure of DBrDPhPZn»(THF)2 66 Figure 2.2. Proposed Structure for 10 .70 Figure 2.3. Summary of Reactions with DPhDVPZn 72 Figure 2.4. The 200 MHz iH-NMR-spectrum of BrEtDPhVP in CDCI3 73 Figure 2.5. The 200 MHz iH-NMR-spectrum of DPhDVP in CDCI3 74 Figure 2.6. Proposed Mechanism for the Reaction of HBr with DPhDVPZn 75 Figure 2.7. Conjugate Addition of en to a Vinyl Group on Protoporphyrin IX 77 Figure 3.1. Crystallization Setup for the (l-Oxido-4-pyridyl)porphyrins 94 Figure 3.2. X-ray Crystal Structure of OPyTrPhP 123 Figure 3.3 The Aromatic Region of the iH-NMR Spectra TrPhPyP and OPyTrPhP in C D C I 3 124 Figure 3.4. Mesomeric Forms of Pyridine-N-oxide 124 Figure 3.5. The Aromatic Region of the iH-NMR Spectra of OPyTrPhP-230 and -210 126 Figure 3.6. UV-Vis Spectrum of TPhPyP in CH2CI2 128 Figure 3.7. UV-Vis Spectrum of OPyTrPhP in CH2CI2 128 Figure 3.8. UV-Vis Spectrum of OPyTrPhP-230 in CH2CI2 129 XVII Figure 3.9 UV-Vis Spectrum of OPyTrPhPZn-230 in CH 2 C1 2 * 129 Figure 3.10. MALDI-TOF Mass Spectrum of f-BOPyDPhP 131 Figure 3.11. Low Resolution Mass Spectrum (EI) of OPyTrPhP-210 131 Figure 4.1. Products of the Reaction of BOEtPIXDME with CH2N2 173 Figure 4.2. The 400 MHz AH-NMR Spectrum of BNImEtPIXDME in CDCI3/TFA... 179 Figure 4.3. Normalized UV-Vis Spectra of BNImEtPIX in 0.05 M NaOH ( a q) and in 0.05 M NaOH(aq):MeOH (1:1) 180 Figure 4.4. The 300 MHz JH-NMR Spectrum of a Mixture of 31 and 32 in CDCI3. . . . 182 Figure 4.5. The 200 MHz iH-NMR Spectrum of BPlEtPDCDME in CDCI3 183 Figure 5.1. Cyclic Voltammetry Cell 191 Figure 5.2. The Cyclic Voltammogram of j-NPhDPhPyP Measured in DMF 196 Figure 6.1. In vitro Toxicity Assay 208 Figure 6.2. In vitro Radiosensitizing Assay 210 Figure 6.3. In vitro Photosensitizing Assay 211 Figure 6.4. Lack of Toxicity of Some Liposome-formulated Porphyrins in Aerobic CHO Cells 214 Figure 6.5. Toxicity in Hypoxic CHO Cells of BNImEtPIX and BHEtPIX Formulated as Cremophor EL® Emulsions 215 Figure 6.6. Lack of Radiosensitization by OPyTrSPhP in Hypoxic CHO Cells 216 Figure 6.7. Radiosensitization of Hypoxic CHO Cells by Liposome-formulated TirapPhTrPhP 217 Figure 6.8. Phototoxicity of OPyTrSPhP, PyTrSPhP, Pyridine N-oxide and Photofrin U® in CHO Cells 218 Figure 6.9. Results from the Accumulation Experiments 222 Figure 6.10. Fluorescence Microscopy Image of OPyTrSPhP-incubated CHO Cells using UV-Excitation 225 xviii Figure 6.11. Fluorescence Microscopy Image of OPyTrSPhP-incubated CHO Cells using Green-Excitation 226 Figure 6.12. Fluorescence Microscopy Image of the TirapPhTrPhP Liposome Formulation Using Green-Excitation 229 Figure 6.13. Fluorescence Microscopy Image of BHEtPIX-incubated CHO Cells using Green-Excitation 230 Figure 7.1. Potential Target Diarylporphyrins from BMEtDPhP 239 Figure 7.2. Proposed Scheme for the Synthesis of a Tetrakis(nitroimidazolyl)porphyrin.241 Figure 7.3. Synthesis of a Water-soluble Cationic Porphyrin from Deuteroporphyrin...246 Figure A . l . !H-2D-COSY spectrum of OPyTrSPhP (Section 3.2.4.3) 258 Figure A.2. iH^D-COSY Spectrum of BAnEtPIXDME (Section 4.2.3.1) 259 Figure A.3. !H-2D-COSY Spectrum of BF 5EtPIXDME (Section 4.2.3.2) 260 xix List of Schemes Scheme 2.1. Synthesis of TEtTMeBPyP 39 Scheme 2.2. Routes to Diarylporphyrins 56 Scheme 2.3. Synthesis of Bis(2-pyrrolyl)methane (3), its Precursors and Reaction Side-product 57 Scheme 2.4. Attempted Synthesis of a Mew-linked DPhP Dimer 64 Scheme 2.5. Synthesis of a Meso-linked Porphyrin from 15 64 Scheme 2.6. Proposed Mechanism for Pd-catalyzed Cross Coupling with DBrDPhPZn. 67 Scheme 3.1. The Reaction of Tirapazamine with CH3COOH in the Presence of 21 134 Scheme 4.1. General Synthesis Pathways for the PIX-based Porphyrin Series 144 Scheme 4.2. In situ Reductive Amination with NaBH3CN 173 Scheme 4.3. The Formation of a Porphyrin-dihydrazide 183 Scheme 5.1. Reduction Pathways of Porphyrins in Cyclic Voltammetry Experiments. 192 List of Tables Table i . Abbreviations Used in Porphyrin Nomenclature xxv Table ii. Key to Compound Numbers xxvi Table iii. Key to Porphyrin Abbreviations xxix Table 2.1. SpectroscopicaUy-Estimated Yields 61 Table 2.2. DCA Catalysis in the Synthesis of BPyP 63 Table 2.3. Summary of Cross-coupling Reactions with DBrDPhPZn 68 Table 3.1. Results of the Mixed-aldehyde Synthesis 88 Table 3.2. Conditions for Attempts at 2-TImP Nitration 91 Table 3.3. ! H-NMR Data for the (l-Oxido-4-pyridyl)porphyrins 100 Table 3.4. ifl-NMR Data for the Porphyrin-N-oxides Measured in CDCI3 100 Table 3.5. UV-Vis Data for the (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides 101 XX Table 3.6. Elemental Analyses for the (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides 102 Table 3.7. Infrared Spectroscopy Data for the (l-Oxido-4-pyridyl)porphyrins 103 Table 3.8. Infrared Spectroscopy Data for the Porphyrin-N-oxides 104 Table 3.9. Mass Spectroscopy Data for (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides 105 Table 3.10. iH-NMR Data for the Sulfonatophenylporphyrins and Pt-derivatives in DMSO-4> 109 Table 3.11. UV-Vis Data in H2O for the Sulfonatophenylporphyrins and Pt-derivatives 110 Table 3.12. Elemental Analyses for the Sulfonatophenylporphyrins and Pt-derivatives. 110 Table 3.13. ipI-NMR Data for the Benzotriazine-dioxide compounds 115 Table 3.14. Infrared Spectroscopy Data for the Benzotriazine-dioxide Compounds 116 Table 3.15. Mass Spectroscopy Data for the Benzotriazine-dioxide Compounds 117 Table 4.1. ifi-NMR Data for the R-PIXDME Porphyrins 167 Table 4.2. iPi-NMR Data for the R-PIX Porphyrins 168 Table 4.3. UV-Vis Data for the R-PLXDME and R-PIX Porphyrins 169 Table 4.4. Elemental Analyses for the R-PIXDME and R-PIX Porphyrins 170 Table 4.5. Mass Spectroscopy Data for the R-PIXDME Porphyrins and BNImEtPIXDNA-di-tBu 171 Table 5.1. Reduction Potentials of Reference Compounds and Correction Factor Calculation 191 Table 5.2. The Reduction Potentials of Selected Porphyrins and Reference Compounds 193 Table 5.3. Literature Data for One-electron Reduction Potentials (E1 7 ) of Selected Radiosensitizers and HSCs Measured by Pulse Radiolysis in Neutral Aqueous Media 198 Table 6.1. Summary of Performed In vitro Assays 203 Table 6.2. Liposome Size Distributions from One Preparation 213 xxi Table A. 1. Experimental Details for X-ray Crystal Structure of DBrDPhPZn(THF)2....249 Table A. LA. Crystal Data 249 Table A. LB. Intensity Measurements 250 Table A.1.C Structure Solution and Refinement 250 Table A.2. Atomic Coordinates for DBrDPhPZn(THF)2 (Section 2.3.4) 251 Table A.3. Bond Lengths for DBrDPhPZn(THF)2 (Section 2.3.4) 252 Table A.4. Bond Angles for DBrDPhZnP(THF)2 (Section 2.3.4) 253 Table A.5. Experimental Details for X-ray Crystal Structure of OPyTrPhP 254 Table A.5.A. Crystal Data 255 Table A.5.B. Intensity Measurements 255 Table A.5.C. Structure Solution and Refinement 255 Table A.6. Atomic Coordinates for OPyTrPhP (Section 3.3.2.2) 256 Table A.7. Bond Lengths for OPyTrPhP (Section 3.3.2.2) 256 Table A.8. Bond Angles for OPyTrPhP (Section 3.3.2.2) 257 List of Abbreviations Abbreviation aq. BCCRC Bu calc'd CHO DCA DCC DCI DDQ DMF DMSO dppf EI en EPC equiv. FAB Gy HBS HP HR-MS HSC iPr Meaning aqueous British Columbia Cancer Research Centre n-butyl calculated (used for elemental analysis and mass spectrometry results) Chinese hamster ovary cells dichloroacetic acid dicyclohexylcarbodiimide desorption chemical ionization 2,3-dichloro-5,6-dicyano-l,4-benzoquinone dimethylformamide dimethylsulfoxide diphenylphosphinoferrocenyl-electron impact ionization 1,2-diaminoethane (ethylenediamine) egg phosphatidylchohne (l-palmitoyl-2-oleoyl phosphatidylcholine) equivalent fast atom bombardment ionization Gray (1 joule/kg) HEPES buffered saline HEPES = [N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid] hematoporphyrin high resolution mass spectrometry hypoxia selective cytotoxin isopropyl xxiii IR infrared spectroscopy LAH lithium aluminum hydride lit. literature reference LR-MS low resolution mass spectrometry LSFMS low energy secondary ionization mass spectrometry m-CPBA meto-chloroperbenzoic acid MACE mono-L-aspartyl chlorin ee MALDI-TOF matrix assisted laser desorption ionization - time of flight (mass spectrometry) MoAb monoclonal antibody MWCO molecular weight cutoff (used to describe dialysis tubing) NBS N-bromosuccinimide NHE normal hydrogen electrode NHS- N-hydroxy succinimide (ester) NMM N-methyl-morpholine NMP N-methyl-2-pyrrolidinone NMR nuclear magnetic resonance (spectroscopy) OEP dianion of octaethylporphyrin OER oxygen enhancement ratio PBS phosphate buffer solution PDT photodynamic therapy PE plating efficiency (toxicity assays) PhA pheophorbide a RPtlc reverse phase thin layer chromatography SCE saturated calomel electrode SER sensitizer enhancement ratio SF surviving fraction (radiosensitization and photosensitization assays) xxiv STMT>BF4 0-(N-succinimidyl)-N,N,N',N'-tetramethyluronium tetrafluoroborate <Bu t-butyl TCA trichloroacetic acid TCQ 2,3,5,6-tetrachloro-l,4-benzoquinone TFA trifluoroacetic acid TFAA trifluoroacetic acid anhydride THF tetrahydrofuran tic thin layer chromatography TsCl 4-toluenesulfonylchloride TsOH p-toluenesulfonic acid UP uroporphyrin UV-Vis UV-Visible spectroscopy XXV Key to Chemical Compound Numbers and Abbreviations The porphyrin ring is numbered as shown in Figure i . The 5, 10, 15, and 20 positions are typically referred to as the raeso-positions; the 2, 3, 7, 8, 12, 13, 17, and 18-positions are typically referred to as the (3-pyrrolic positions. The inner pyrrolic Ns are numbered 21 through 24. 8 10 12 Figure i . Porphyrin Position Numbering. Numerous compounds were synthesized and used during the course of this thesis. In general, numbers were assigned to non-porphyrin compounds or mixtures of porphyrins (e.g. 29 and 30). Numbers not appearing in Table ii (e.g. 15 to 17) are assigned to compounds appearing in the text which were target compounds not attained, or compounds reported in the literature. In Table iii , combinations of letters were used to abbreviate names of porphyrins, such that they could be identified readily throughout the text even when they appear in more than one Chapter (e.g. BNImEtPIX appears in four different Chapters). To construct these abbreviations, the following functional group abbreviations were used (Table L). xxvi Table i. Abbreviations Used in Porphyrin Nomenclature Abbreviation Substituent Abbreviation Substituent porphyrin N-oxide 210 with 0 at 21 position NPh 4-nitrophenyl An anilino OEt oxoethyl B bis OPy l-oxido-4-pyridyl Br bromo P porphyrin c- cis Ph phenyl Cl chloro PIX a Protoporphyrin IX D di PI phmalimido DMED dimethyl ester Pr propyl DMEt 2-dimethoxyethyl SPh 4- sulfonatophenyl Et ethyl T tetrakis F 5 N H C H 2 C F 2 C F 3 t- trans Im imidazolyl THF tetrahydrofuran Me methyl Tr tri / tris M m 2-nitroimidazol-1 -yl V vinyl aProtoporphryin IX is typically abbreviated using PPDC, however PIX was chosen in the interest of keeping short the abbreviations used here. b D M E (dimethyl ester) is used only in conjunction with PIX. Two examples appear below. In the first, BNImEtPIXDME is based on protoporphyrin TX dimethyl ester, thus PIXDME forms the root. The substituents at the 8- and 13- positions define the other part of the name which precedes PIXDME. O2N-</ J N Dimethyl 3,7,12,17-tetramethyl-8,13-Bis(2-(2-nitroimidazol-l-yl)ethyl)porphryin-2,18-dipropionate The second example illustrates the naming of a tetraarylporphyrin N-oxide, OPyTrPhP-210. The letter P (for porphyrin) forms the root of tetraaryl- (and diaryl-) porphyrin abbreviations. The ring substituents precede P, according to the order in which they appear in the full name. Porphyrin-N-oxides are N-substituted, and hence require further numbering to distinguish the position of the oxido moiety. OPyTrPhP-210 5-(l-Oxido-4-pyridyl)-10,15,20-triphenylporphyrin-21-oxide Table ii. Key to Compound Numbers Number, Compound Structure and Name Number, Compound Structure and Name S O 1 2 H H Bis(2-pyrrolyl)thioketone H H Bis(2-pyrrolyl)ketone H H 3 H H Bis(2-pyrrolyl)methane 4 H H 1,1 ',2,2'-Tetrakis(2-pyrrolyl)ethane Table ii. Key to Compound Numbers (continued) N H H (4-Pyridyl)bis(2-pyrrolyl)methane 0 H Ethyl 4-ethyl-3,5-dimethyl-pyrrole-2-carboxylate O H 0 Ethyl 5-acetoxymethyl-4-ethyl-3-methyl-2-pyrrole carboxylate 0 H H O (5,5'-Bis(ethoxycarbonyl)-3,3'-diethyl-4,4'-dimethylbis(2-pyrrolyl)methane H H 3,3'-Diethyl-4,4'-dimethylbis(2-pyrrolyl)methane 10 N . N-Bu3Sn Tris(n-butyl)(l-imidazolyl)tin(rV) 11 ^ ~ ^ - S n B u 3 Tris(n-butyl)phenyltin(IV) Et0 2 C N 1 2 CH-SnBu 3 Et0 2 C' Diethyl 2-(tris(n-butyl)stannyl)-l,3-dipropanoate 13 [/~~^-SnBu 3 Tris(n-butyl)(4-pyridyl)tin(IV) 1 4 ^ - S n B u 3 Tris(n-butyl)vinyltin(IV) N 0 2 CH 2OH N0 2 4(5)-HydroxymethyUmidazole Picrate HOH 2C 1 9 >=\ N ^ N H 4(5)-Hydroxymethyhmidazole OHC 20 ^ N ^ N H 4(5)-Imidazolecarboxaldehyde o—c=o 2H-[l,2,4]Oxadiazolo[3,2-c] [ 1,2,4]benzotriazin-2-one-5-oxide xxix Table ii . Key to Compound Numbers (continued) O -6- H 3-N-Acetamido-1,2,4-benzotriazine-1,4-di-N-oxide 0 . _ M + 2 3 L A A c ^ ^ 6- H Butyl (N-(l,4-dioxido-1,2,4-benzotriazin-3-yl)carbamate f C H 3 \ N H N = = / ^ - N Y N 29 ( } N ° 2 / - N H N - \ H 3C-^JL^S S SJIS>- CH3 H3COOC COOCH3 Dimethyl 3,7,12,17-tetramethyl-13-(2-(2-nitroimidazol-1 -yl)ethyl)-8-vinyl-porphryin-2,18-dipropionate O2N-<' j L C H 3 V N H N=/ ^ H a C - ^ y i k ^ s V " - C H 3 H 3 C O O C C O O C H 3 Dimethyl 3,7,12,17-tetramethyl- 8- (2-(2-nitroimidazol-1 -yl)ethyl)-13-vinyl-porphryin-2,18-dipropionate v' C H 3 0 V J  0 H 3 C \ M s f " C H 3 H 3 C O O C C O O C H 3 Dimethyl tetramethyl-13-bis((2-phthalimido)ethyl)-3,7,12,17-8-vinyl-porphyrin-2,18-dipropionate L JC H 3 32 H 3 o ^ S r ^ v V N V N H N - \ H3C-Sy>!<rfSSiv>- C H 3 H 3 C O O C C O O C H 3 Dimethyl 3,7,12,17-tetramethyl-8-bis((2-phthalimido)ethyl)-13-vinyl-porphyrin-2,18-dipropionate XXX Table iii. Key to Porphyrin Abbreviations Abbreviation Structure and Name Abbreviation Structure and Name o HN / C H 3 BAnEtPIXDME j~m N=( "y\ / - N HN-\ H 3 c " \ ^ k > i * s I > ^ - - C H 3 HgCooc C O O C H 3 Dimethyl 8,13-bis(2-(N-anihno)ethyl)-3,7,12,17-tetramethylporphyrin 2,18-dipropionate S" 2* CH3 V NH N = / C H 2 B r BBrEtPIXDME ( } f~ N HN-\ N 3 C ' Y k A f ' C H 3 H 3 C O O C COOCH3 Dimethyl 8,13-bis(2-bromoethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate H 3 C 0 4 ^OCHa / C *H C H 3 BDMEtPlXDME V N H N=( ,CX / \ H OCH3 / - N H N - f H3COOC COOCH3 Dimethyl 8,13-bis(2,2-dimethoxyethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate R HN' / CH3 BF5EtPIXDME \ N H N _ / H3COOC COOCH3 R = CH2CF2CF3 Dimethyl 8,13-bis(2-(N-3,3,3,2,2-pentafluoropropylamino)ethyl)-3,7,12,17 -tetramethylp orphyrin-2,18-dipropionate C H 2 O H ( J C H 3 V N H N*/ CH2OH BHEtPDC ( } / - N HN-4 H 3 c - \ A ^ k ^ ~ C H 3 HOOC COOH 8,13-Bis(2-hydroxyethyl)-3,7,12,17-tetramethyl-porphyrin-2,18-dipropionic acid / C H ^ ° H CH3 V - N H N a / CH2OH BHEtPrXDME ( ) J—N WW-\ H 3 c ~VAs***»kA" C H 3 H3COOC COOCH3 Dimethyl 8,13-bis(2-hydroxyethyl)-3,7,12,17-tetramethyl-porphyrin-2,18-dipropionic acid xxxi Table iii. Key to Porphyrin Abbreviations (continued) C H 3 V N H N=/ CH2I BIEtPIXDME ( } f-N HN-A " s C ' y k A r CH3 H 3 C O O C C O O C H 3 Dimethyl 8,13-bis(2-iodoethyl)-3,7,12,17 -tetramethylp orphy rin-2,18-dipropionate H3CO <^T^T^S BDMEtDPhP H C _ f - N H N =<_ OCH3 HaCO 7 H N - Z ^CH V s J s / OCH3 5,15-bis(2-dimethoxyethyl)-10,20-diphenylporphyrin N - . OZN-4 j N ( CH3 BNImEtPIX H c - y V ^ r A ^ v /=» 3 " V N H N = ( ^ - N Y N / - N H N - \ "3°'\A^1^f~ C H 3 HOOC COOH 3,7,12,17-Tetramethyl-8,13-bis(2-(2-nitroimidazol-1-yl)ethy l)-porphyrin-2,18-dipropionic acid 0 2 N - ( ' j N ^ / CH3 BNImEtPIXBNA A - N H N=T V 1 J / - N H N - K H3C\j^g^~J^O\\ 3 0=C C = 0 HOOC COOH Bis(N-L-aspartyl) 3,7,12,17-tetramethyl-8,13-bis(2-(2-nitroimidazol-1 -yl)ethyl)-porphyrin-2,18-dipropionamide / C H 3 VNH N*/ Y BNImEtPIXBNA ( } NO2 -(tBu)2 7^N H N-( H a C - Y ^ t ^ - C H a o=c c=o ROOC^" H ^ C O O R r - C H H C S R = ' B u ROOC COOR Tetra(tertbutyl)bis(N-L-aspartate) 3,7,12,17-tetramethyl-8,13-bis(2-(2-nitroimidazol- l-yl)ethyl)-porphyrin-2,18-dipropionamide 0 2 N - ( ' \ N ^ BNImEtPIXDME ? CH3 V N H N=/ Y ( } / - N H N - \ H 3 C - ^ B ^ ^ ^ - C H a H3COOC COOCH3 Dimethyl 3,7,12,17-tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yl)ethyl) porphryin-2,18-dipropionate XXX11 Table iii. Key to Porphyrin Abbreviations (continued) N02 BNPhP V N H N=/ /~N HN-\ N02 5,15-bis(4-nitrophenyl)porphyrin 0 > CH3 BOEtPrXDME 3 ~VNH N=/~~V H 3 ° CH3 H3COOC C00CH3 Dimethyl 3,7,12,17-tetramethyl- 8,13-bis(2-oxoethyl)-porphyrin-2,18- • dipropionate 0 i N BOPyP Q [ H I W N i 0 5,15-bis( 1 -oxido-4-pyridyl)porphyrin R R = S03Na \ ^ c-BOPyBSPhP —* V N H I W + R - f \~L /r~i  N-° N + 6 -Sodium 5,10-bis( l-oxido-4-pyridyl)-15,20-bis(4-sulfonatophenyl)porphyrin V N H N=/ c-BOPyDPhP ( W ) — ( ; N - 6 — —' N+ 1 0-5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin .—. VNH N=/ + c-BOPyDPhP-210 W ~ l + N ; ° H N M _ ^ N _ 0 0" 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin-21-oxide xxxni Table ii . Key to Porphyrin Abbreviations (continued) c-BOPyDPhP-220 W \ N H \ t ( \ j N ' 5 /—. \=Nx NH-^ + c-BOPyDPhP-240 W ~ {  +'%0 M _ * N ~ ° /—NH N-\ 6-N + 1 O " 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin-22-oxide 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin-24-oxide 0 -N+ O .—. VNH N=/ f-BOPyDPhP ( W )—( ) !^ 0 -BPlEtPIXDME L^JL  3 y-NH N=( V _/ 0 H 3 C ' \ > ^ i v ^ CH3 H3COOC COOCH3 5,15-Bis(l-oxido-4-pyridyl)-10,20-diphenylporphyrin Dimethyl 3,7,12,17-tetramethyl-8,13-Bis((2-phthaUmido)ethyl)-porphyrin 2,18-dipropionate HOOC. MeOOC S CH3 BPrAPIXDME VNH N=/ ^ C O O H V N HN-/ HaC-^s^^s^CHa BPrEPIXDME "^'fiCXj^coau* / -N HN-/ H 3 C ~ V X ^ < > - C H 3 H3COOC COOCH3 H3COOC COOCH3 Dimethyl 3,7,12,17-tetramethylporphyrin-8,13-bis(£'/Z2-propenoic acid)-2,18-dipropionate Tetramethyl 3,7,12,17-tetramethylporphyrin-2,18-dipropanoate-8,13-bis(£7Z 2-propenoate) xxxiv Table iii. Key to Porphyrin Abbreviations (continued) BPyP V N H \ = / N 5,15-Bis(4-pyridyl)porphyrin BrEtDPhVP D U ^ V N H N*( /-N HN-\ 5-(2-Bromoethyl)-10,2-diphenyl-20-vinylporphyrin BrEtlmEtDPhP < * V " W ^ f\ V N H N=/ N _/ B r " M _ M / - N HN-\ R and 5 5-(2-Bromoethyl)-15-(l-(imidazol- l-yl)ethyt)-10,20-diphenylporphyrin R o' / CH3 BTsEtPDCDME H^"yJC^]j3*^o V N HN-f H 3 C CH3 H3COOC COOCH3 0 R = - f ^ O " c H 3 0 Dimethyl 3,7,12,17-tetramethyl-8,13-bis(2-(4-toluenesulfonyl)oxoethyl)-porphyrin-2,18-dipropionate CIDPhPZn V N X N=/ CI—(x Zn /) [5-Chloro-10,20-diphenylporphyrinato]Zn(II) CIDPhVP V N H N=/ Cl~v / - N HN-\ 5-Chloro-10,20-diphenyl-20-vinylporphyrin XXXV Table iii. Key to Porphyrin Abbreviations (continued) CIDPhVPZn V \ N*/ DBrDPhP V NH N*/ B R _ V J~Br hu HN-\ [5-Chloro-10,20-diphenyl-20-vinylporphyrinato]Zn(II) 5,15-Dibromo-10,20-diphenylporphyrin Q Q V N X N=S( DBrDPhPZn Br-6 a( }-Br DBrDPhPZn»THF2 V N X ^N-ZTHF Br-(\..-,ZrC'J- Br [5,15-Dibromo-10,20-diphenylporphyrinato]Zn(II) [5,15-Dibromo-10,20-diphenylporphyrinato]Zn(II)»(THF)2 DPhDVP . C N H DPhDVPZn V N N«7 ^ \ A JS / - N N-\ 5,15 -Diphenyl-10,20-divinylporphyrin [5,15-Diphenyl-10,20-divinylporphyrinato]Zn(II) Q Q DPhP V N H N«/ / - N HN-\ DPhPZn V N X ^ B / u Q 5,15 -Diphenylporphyrin [5,15-Diphenylporphyrinato]Zn(II) xxxvi Table iii. Key to Porphyrin Abbreviations (continued) OPyTrPhP 5-(l-Oxido-4-pyridyl)-10,15,20-triphenylporphyrin OPyTrPhP-210 5-(l-Oxido-4-pyridyl)-10,15,20-triphenylporphyrin-21-oxide -NH Ns OPyTrPhP-230 L H \ + A N / r b -N HN • 5-(l-Oxido-4-pyridyl)-10,15,20-triphenylporphyrin-23-oxide OPyTrSPhP R 6 NH N J~\-!f N HN ll 1 f * l ) R + -N - O Sodium 5-(l-oxido-4-pyridyl)-10,15,20-tris(4-sulfonatophenyl)porphyrin R = SC^Na I OPyTrSPhPPt Sodium [20-( 1 -oxido-4-pyridyl)-5,10,15 -tris(4-sulfonatophenyl)porphyrinato]Pt(II) PhP 5 -Phenylporphyrin XXXV11 Table iii. Key to Porphyrin Abbreviations (continued) / C H 3 H 3 c ~ / S r ^ v V ^ PDCDME >-NH Na< / - N H N - ^ H 3 C ~ \ X ^ ^ h CH3 H 3 C O O C C O O C H 3 Dimethyl 3,7,12,17-tetramethyl-8,13-divinyl-porphyrin-2,18-dipropionate R R = SOaNa C O PyTrSPhP ^ V N H N=( — \ 3 V J a ! / — R Sodium 5-(4-pyridyl)-10,15,20-tris(4-sulfonatophenyl)porphyrin N TEtTMeBPyP v ^ - ^ X J X / ^ N 3,7,13,17-Tetraethyl-2,8,12,18-tetramethyl- 10,20-bis(4-pyridyl)porphyrin /=\ N . N H N ")-NH N = 7 | M 2-TImP H N ^ N \=J Tetrakis(2-imidazolyl)porphyrin H 4-TImP / ^ U ^ M H Tetrakis(4(5)-imidazolyl)porphyrin TirapPhTrPhP O +VI=N " H ^ T A h n N i W 5-(4-N-(N'-(l,4-dioxido-l,2,4-benzotriazin-3-yl)aminocarbonyl) aminophenyl)-10,15,20-triphenylporphyrin xxxvni Table iii. Key to Porphyrin Abbreviations (continued) 0 -1 N+ TOPyP + V N H I O + o -5,10,15,20-Tetrakis(l-oxido-4-pyridyl)porphyrin TPhP _ . V N H N=/ OLJX^/ — 5,10,15,20-Tetraphenylporphyrin V N N N=/ .—. TPhPZn (J-L A J-\J [5,10,15,20-Tetraphenylporphyrinato]Zn(II) 0 -i N+ TrOPyPhP / V ^ T > + /—- V N H N=7 6 _ C M hi} d -5,10,15-Tris(l-oxido-4-pyridyl)-20-diphenylporphyrin o-i N+ TrOPyPhP-210 y ^ J L ^ + Y_, V N H N=7 . . 0 -5,10,15-Tris( 1 -oxido-4-pyridyl)-20-diphenylporphyrin-21-oxide 0 -N+ TrOPyPhP- c ^ V ^ S 220 + / V N H N=7 o-5,10,15-Tris(l-oxido-4-pyridyl)-20-diphenylporphyrin-22-oxide xxxix Table iii. Key to Porphyrin Abbreviations (continued) R R R = SOaNa | ^ TSPhP « T y V > R H C J K , H ~ > R R = S 0 3 N a y*^ TSPhPPt V N n r W ^ — ' / - N N - \ > — ' Q R R Sodium 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin Sodium [5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrinato] Pt(U) Acknowledgements I would like to thank my supervisors Brian James and Kirsten Skov for their guidance, support and patience over the years. Thanks also go to the departmental services, especially Lianne Diarge, Marietta Austria, Peter Borda, Steve Rak, Steve Rettig, and the people at mass spectrometry for their support. From the BC Cancer Research Centre, I thank Hans Adomat, and Haibo Zhou for help with just about everything involving the in vitro experiments; Mladen Korbelik, Lawrence Mayer, Ellen Wasan, Neal Poulin are also deserving of my gratitude for their help with PDT, liposomes, fluorescence microscopy and many useful discussions. From BRJ's lab, Ian "oh so loud" Baird (I didn't mean to knock it over Ian, honest), Matt "once a poocher" LePage (I need only a couple stir bars Matt, really) are to be commended for enduring my lab presence, and thanked for many good laughs and good companionship. I am indebted to Russell "double 1" Pratt for doing much of the grunt work last summer. Other members of the James group past and present: Don Yapp, James Ravensbergen, Terrance Wong, Ken MacFarlane, Yu Wang, Graham Cairns (and all the rest) deserve honorable mention. For companionship during many late thesis nights in the lab, I thank Ross Porter, Metallica and Our Lady Peace. I thank my parents and the rest of my family for all the love and support over the years; special thanks goes to my mother who edited much of this thesis before anyone else saw it. I am deeply grateful for these God-given talents with which I was able to accomplish this work. 1 Chapter 1 Introduction to Porphyrins and Cancer Therapy 1.1 Background Porphyrins and related macrocycles are ubiquitous in living systems. Hemin (or the reduced form, heme(Fen)), the most widely occurring porphyrin derivative in animal materials, is at the active site of proteins such as myoglobin and hemoglobin and enzymes such as tryptophan pyrrolase, many peroxidases, cytochrome P-450, b-type cytochromes, and catalase.1 Chlorins, dihydroporphyrin derivatives, are part of the chlorophyll photosynthetic system of green plants and vitamin B12 (not shown) is also a porphyrin-like macrocycle.2 Hemin Chlorophyll a HP Given their abundance in nature, it is no surprise that these pigments have captivated chemists and other scientists for more than a century. In 1867, Thudichum was the first to isolate a porphyrin (hematoporphyrin, HP) from natural materials, although the crystallization of hemoglobin was reported some 27 years earlier by Hiinefeld.3 In comparison, the preparation of P-unsubstituted tetraaryl and diarylporphyrins is a 'recent' development. In 1936, Rothemund reported the synthesis of tetraphenylporphyrin (TPhP) and other tetraarylporphyrins,4 and in 1968, Treibs and Haberle reported the preparation of diphenylporphyrin (DPhP).5 References on page 22 2 TPhP DPhP PIXDME The original goal of this project was to develop porphyrin chemistry for the purpose of preparing novel, porphyrin-based radiosensitizers. However, given the hypoxia-selective properties of some of the incorporated moieties (e.g. nitroimidazoles, heterocyclic N-oxides) the overall goal was broadened to include hypoxia-selective cytotoxins (HSCs). Also, as many porphyrins can photosensitize cells, these compounds were also investigated as photosensitizers. Finally, some steps were made to incorporate pentafluoroalkyl groups into porphyrins, which could form the basis for porphyrin-based hypoxia imaging agents. There are many reasons to pursue porphyrins incorporating radiosensitizers and HSCs. The reported tumor-locahzation properties of porphyrins (Section 1.2) might effectively deliver the compound to the tumor tissue. Photodynamic therapy (PDT) might be a viable treatment modality even in hypoxic areas of tumors because of porphyrin-nitroimidazole photoprocesses (Section 1.4.3.1). Also, a number of porphyrins and metalloporphyrins have been shown to be radiosensitizers (Section 1.5). Finally, the combination of PDT and selected HSCs has been shown to be more effective than either treatment alone (Section 1.6). In the course of this project, the synthesis and subsequent modification of tetraarylporphyrins and diarylporphyrins as well as substituent manipulation on PIXDME (a derivative from hemin) were explored in the pursuit of porphyrins incorporating radiosensitizers, HSCs and other potentially useful moieties. A preview of the thesis chapters appears in Section 1.8. References on page 22 3 1.2 Porphyrin Localization in Tumor Tissue One advantage of porphyrin-based cancer therapies is that tumor tissue has been reported to accumulate many porphyrins to a greater degree than many normal tissues; this phenomenon also occurs with other aromatic macrocycles (e.g. chlorins, phthalocyanines, and naphthalocyanines).6'7 It should be emphasized that photosensitizers are not accumulated exclusively by tumors; for example, the liver, kidney, and spleen typically accumulate more photosensitizer than does tumor tissue, whereas other tissues often accumulate photosensitizers to a lesser degree.8 These observations have been made in the context of photodynamic therapy (PDT), and thus references to 'photosensitizers' are made in this Section; however, this does not imply that porphyrins as radiosensitizers or HSCs do not accumulate in tumors. Cancerous cells in vitro do not accumulate photosensitizers to a significantly greater degree than do normal cells, so this selective accumulation must be linked to processes which occur in vivo . 7 ' 9 , 1 0 The mechanisms of tumor localization of photosensitizers are complex, and reviews by Moan and . Berg,7 Hamblin and Newman,9 Pass,10 and Ochsner8 have discussed this phenomenon. The biodistribution can vary with the lipophilicity and charge on the photosensitizer,6'7,11 and the kinetics of delivery of the photosensitizer can vary with the tumor type and photosensitizer delivery mode. 6' 7' 1 0 Low density lipoproteins (LDL), high density lipoproteins (HDL) and serum albumin are important blood carriers of photosensitizers.7"14 In general, highly lipophilic photosensitizers bind with HDL and LDL, whereas more hydrophilic photosensitizers bind with serum albumin or other serum proteins. LDLs can be incorporated into lysosomes in cells via receptor-mediated endocytosis and, because cancerous cells have an increased need for cholesterol for membrane synthesis, they may have a greater number of LDL receptors, thus incorporating increased amounts of photosensitizer.7"10 Albumin and HDLs are not internalized by receptor mediated endocytosis, and thus photosensitizers bound to these proteins must exchange with the plasma membrane to be incorporated into cells.10 On the other hand, strong binding to serum proteins can decrease the References on page 22 4 amount of photosensitizer available for cellular uptake and decrease the cellular retention of these compounds compared to serum-free media as shown by Korbelik in in vitro studies.13'15 Other important biological factors help to account for the localization phenomenon. Tumor-associated macrophages can accumulate large amounts of photosensitizers7'9'11'13'15 and have been implicated in host-immune response to the light treatment (see Section 1.4.3). Hydrophilic photosensitizers which are bound to serum albumin or globulins may reach tumors via their increased vascular permeability and localize in the tumor stroma.8'9'12 Compared to normal tissue, tumors often have greater numbers of lipid bodies which could accumulate hpophilic photosensitizers. Lower pH in a tumor may also contribute to photosensitizer localization.7,9*10 Also, "leaky" vasculature and poor lymphatic drainage in tumors, along with tumor necrosis and tumor-induced degenerative changes in the normal tissue it is invading, could contribute to increased concentrations of photosensitizers.9'13 In summary, the basis for tumor-accumulation of photosensitizers is complex, and depends on the composition of the photosensitizer and numerous processes in vivo. Nevertheless, that tumor tissue accumulates relatively larger amounts of photosensitizers is a tenet central to photodynamic therapy (PDT) and porphyrin-based anti-cancer therapies in general. 1.3 Tumor Hypoxia The presence of hypoxic cells is a feature of many cancers, particularly in tumors greater than 1 mm3. The degree of hypoxia within tumors varies according to the proximity of blood vessels in a rapidly growing tumor where the vasculature development does not keep up with cell proliferation.16"19 A schematic diagram of a hypoxic tumor is shown in Figure 1.1. As the distance from the capillaries increases, the [O2] decreases due to O2 consumption by the cells and limitations in O2 diffusion, thus yielding hypoxic regions at distances > -70 |im. 1 9 If the cells are too far from the capillaries for effective delivery of nutrients and O2, and for elimination of wastes, necrotic regions can result. Hypoxia can also be the result of intermittent interruption of blood flow because of irregularities and occlusions in the tumor vasculature and the high interstitial pressure of tumors with poorly developed lymphatic systems.7'16'18'20'21 References on page 22 5 Hypoxia can negatively influence the outcome of chemotherapy, presumably because of poor diffusion of drugs to hypoxic tissues and the reduced metabolic rate of hypoxic cells,16'23 and can decrease the effectiveness of radiotherapy (Section 1.4.1) and photodynamic therapy (Section 1.4.3). Much research has focused combining radiotherapy with hyperbaric oxygen treatment, and sensitizing these hypoxic cells to radiation via radiosensitizers (Sections 1.4.1 and 1.5).17'24 Subsequently, it has been recognized that hypoxia might be exploited in the treatment of cancer through the use of HSCs (Section 1.4.2) and that it could be more effective to eliminate hypoxic cells outright rather than sensitize them to radiation.17 More recently, tumor oxygen-status determinations via 02-selective Eppendorf probes or nitroaromatic hypoxia markers (including nitroimidazoles, Section 1.7.1) have received much attention.19'25"28 Figure 1.1. Schematic Representation of a Portion of a Hypoxic Tumor (from Farrell).22 Clearly, hypoxia is a problem in many cancers and much effort has been given to the evaluation and elimination of hypoxic cells within tumors. Porphyrins incorporating References on page 22 6 radiosensitizers, HSCs and other useful moieties might be useful in overcoming the problem of hypoxia. 1.4 Evaluation of Porphyrin-based Anti-cancer Agents 1.4.1 Radiosensitizer Mechanisms The most important subcellular target of ionizing radiation in cancer therapy is DNA, and O2 effectively sensitizes cells to ionizing radiation.19'24'29 Simple models to explain this phenomenon are outlined in Figure 1.2. In Path I, DNA radicals are generated by reaction with the products of water radiolysis (e.g. OH», indirect effect) and O2 reacts with these radicals and 'fixation' of the damage occurs. In Path II, DNA is ionized directly and O2 abstracts an electron from the radical species, resulting in a strand break.19'24'29 In either case, the damage to DNA is made permanent when O2 is present, and, if the damage is at a critical site and not repaired, successful cell procreation is blocked. (Path I) O H * 0 2 DNA-O2 adducts, strand breaks, etc. (damage 'fixation') (Path II) direct ionization electron migration electron abstraction 0 2 0 2 " radical decomposes strand break Figure 1.2. Proposed Mechanisms for O2 Radiosensitization (adapted from Ravensbergen).30 In the absence of O2, cells are approximately three times more resistant to radiation as seen in experiments where aerobic cells and hypoxic cells (produced by N 2 flow) were treated with X-rays (Figure 1.3). 1 9'2 2'2 3 The surviving fraction is a measure of the extent of cell kill, normalized References on page 22 7 to the plating efficiency at a dose of 0 Gy. The dose ratio to achieve the same survival in oxygenated and hypoxic cells is known as the oxygen enhancement ratio (OER), measured in Figure 1.3 at a surviving fraction of 0.1 (10 % of the cells were killed). To compensate for the relative radioresistance of hypoxic cells, compounds which mimic the effect of 0 2 , i.e. radiosensitizers, have been heavily researched.24 In the case of radiosensitizers, the dose ratio required to achieve the same survival level is known as the sensitizer enhancement ratio (SER); this is usually much smaller than the OER of ~3. 1.0 % 0.1 o r n ^ Dose in N 2 " f c M Dose in 0 2 for Surviving X Fraction of 0.1 \ 1 c s 0.01 0.001 15 _v i_ 20 25 Dose (Qy) -L I 30 Figure 1.3. Typical Results for X-ray Irradiation of Chinese Hamster Ovary Cells Under Aerobic (O2) or Hypoxic (N2) Conditions (from Farrell).22 Numerous nitroaromatics (e.g. misonidazole, other nitroimidazoles, nitrobenzenes, nitropyrroles, nitrofurans and nitrothiophenes),24'31,32 nitroxide free radicals (e.g. TAN, see below),33 aromatic heterocyclic N-oxides (e.g. tirapazamine),32 metals and metal complexes (e.g. Pt-amine, Co-amine, Ru-sulfoxide, Pt and Ru-nitroimidazole complexes),22,34"39 several porphyrins and metalloporphyrins (Section 1.5) and other compounds have been investigated as References on page 22 8 radiosensitizers. Of these classes, the nitroaromatics have been most thoroughly investigated. Their mechanisms of action are thought to mimic those of oxygen by abstracting electrons from ionized DNA or by forming DNA-adducts once the nitro groups have been reduced,1 8 , 2 4'3 2'4 0 and their activities are strongly correlated to their one-electron reduction potentials.32'41'42 Of the heterocyclic N-oxides, studies with tirapazamine indicate that the one-electron reduction product (the N-oxide radical anion) is the DNA-damaging species 4 2 , 4 3 Related mechanisms of action for nitroimidazoles and heterocyclic N-oxides are implicated in the cytotoxic activity of HSCs (next Section). 0 o_ v^ro/ NO2 9 6 -misonidazole TAN tirapazamine 1.4.2 Hypoxia-Selective Cytotoxin Mechanisms As with radiosensitization, the final HSC-action site is believed to be D N A . 1 8 , 2 4 , 4 2 4 4 However, in contrast to radiosensitizers, HSCs (in the less toxic, prodrug form) are activated by reductive cellular enzymes such as NADPH-cytochrome P-450 reductase, cytochrome P-450, xanthine oxidase (which reduces 0 2 to peroxide),45 aldehyde oxidase and DT-diaphorase.42,46 4 9 In the presence of O2, the reduced product (shown in Figure 1.4 as resulting from a one-electron process) reacts with O2 to yield superoxide and the regenerated prodrug. In hypoxic cells, the back reaction is unlikely to occur and the activated HSC can react with cellular targets to form cytotoxic lesions, thus providing a basis for the hypoxia-selective toxicity. Enzymatic Activation HSC HSC* (Prodrug) (Cytotoxic Species) o 2 - 0 2 Figure 1.4. Activation of a Hypoxia-Selective Cytotoxin (HSC) (from McClelland).50 References on page 22 9 HSCs have been based on nitroaromatics (e.g. misonidazole, metronidazole (flagyl), RSU 1069, nitroacridines, nitroquinolines, etc.),44'51 aromatic heterocyclic N-oxides (e.g. tirapazamine, RB90740, etc),42-44 quinones (e.g. mitomycin C, E09), 4 4 metal complexes (e.g. Co-nitrogen mustards 4 4 DNA-targeted Pt-, Rh-, and Ru-nitroaromatic complexes)35 and other compounds. It should be noted that many compounds which were developed as radiosensitizers also act as bioreductive drugs and vice-versa (e.g. misonidazole, RSU 1069, tirapazamine, etc.); as with radiosensitizers, the activity of many HSCs is strongly correlated with their one-electron reduction potentials.18'42'44 O O CH3 mitomycin C E09 1.4.3 Photodynamic Therapy (PDT) Mechanisms Molecular Level As described recently in reviews by Ochsner8 and Harriman,52 photosensitizers can induce cellular damage via formation of radical species or ^Oz, pathways known as Type I and Type II processes, respectively (see Figure 1.5). The first step in either process is light-activation (hv) of the photosensitizer to its excited singlet state (ip*) which can revert to the ground state via fluorescence or non-radiative processes, or undergo subsequent intersystem crossing (ISC) to References on page 22 10 yield the excited triplet state (3P*). The 3 P * state can undergo Type I or Type II processes in subsequent reactions, or revert to the ground state via phosphorescence (not shown). Light-Activation Subsequent Reaction Type I Type II Figure 1.5. Activation of a Photosensitizer and Subsequent Type I and Type II Photoprocesses; (the Jablonski diagram for Light-Activation was adapted from Streitweiser and Heathcock).53 In a Type I photoprocess, electron transfer occurs between the 3 P * and the substrate molecule(s) to produce the porphyrin radical cation (P ,+) and the radical anion of the substrate (S*~ ), or alternatively, and less commonly the porphyrin radical anion (P*-) and the substrate radical cation (S*+). Similar electron transfers can occur from the *P* state, but are much less likely to occur.8-52 Substrates for Type I photoprocesses include molecules such as cytosine, cysteine, guanine, histidine, tyrosine, tryptophan and methionine. In oxygenated environments, the majority of these radicals react rapidly with 0 2 to yield highly reactive superoxide, peroxide and hydroxyl radical species that subsequently oxidize nearby biomolecules. Type I processes may be particularly important for covalently-bound porphyrin-nitroimidazole compounds (Section 1.4.3.1). In Type II photoprocesses, 3 P* reacts with ground state oxygen (302) to produce singlet oxygen (102) (see Figure 1.5), which is known to impair cellular function via reaction with electrophilic biomaterials such as unsaturated lipids, proteins, nucleic acids, etc. to produce damage in cell membranes, lysosomes, mitochondria, and nuclei.8'52 The relative importance of Type I vs. Type II mechanisms in PDT depends on many factors including the photochemistry of References on page 22 11 the photosensitizer itself, the polarity of the environment in which 3 P * is generated, the availability of suitable Type I substrates and of O2 (le. the degree of hypoxia in a tumor), etc.8'52 In a heterogeneous system such as a living cell, distinction between Type I and Type II photoprocesses is not trivial.8 The importance of photothermal effects occurring during PDT has been emphasized by Harriman.52 Photothermal effects are the result of non-radiative relaxation of the excited states of photosensitizers (see Figure 1.5); if this is to be a significant mechanism of PDT, the photosensitizer should have a short lived excited singlet state (!p*) and minimal ISC to the 3 P * state. Cellular Level In contrast to radiotherapy and the use of many chemotherapeutic agents, the cytotoxicity of PDT is generally thought to affect cellular targets other than DNA, partly because of poor photosensitizer localization in the DNA, although PDT-mediated damage to nucleic acids and DNA repair enzymes has been observed.7'8'10 PDT causes significant damage to proteins and lipids in cellular membranes (e.g. plasma, nuclear, mitochondrial, lysosomal, and those of the Golgi apparatus and endoplasmic reticulum).7'10 Cationic photosensitizers tend to accumulate in the mitochondria and anionic photosensitizers in lysosomes; photo-induced damage to these regions within the cell can lead to disruption of cellular metabolic processes and release of lysosomal enzymes, respectively.7'8'54 However, the cytotoxic potential of more-lipophilic photosensitizers is presumably greater because of their greater uptake into cell membranes, which positions them for easy transfer across subcellular compartments.8 The products generated during PDT treatment can also severely damage proteins and enzymes.8 Organism Level In addition to direct cellular damage, damage to the tumor vasculature can disrupt blood flow thus inducing ischemia and hemorragic necrosis.7'9,11,55 Hypoxia may also be induced as a result of O2 consumption during PDT treatment.7-10'56 PDT can elicit a significant tumoricidal References on page 22 12 response from host macrophages present in the tumor and those which infiltrate the tumor after PDT treatment.7-9-55 1.4.3.1 Porphyrin-Nitroimidazole Interactions Photochemical studies with the nitroimidazoles misonidazole and metronidazole using photosensitizers such as hematoporphyrin (HP), uroporphyrin (UP), tetrakis(4-sulfonatophenyl)porphyrin and its Zn complex, and mono-L-aspartyl chlorin e6 (MACE) have given evidence for Type I photodynamic processes (see above), these yielding the radical cation of the photosensitizer and radical anion of metronidazole.57"59 In the presence of O2, evidence for the generation of a superoxide species (presumably generated by electron transfer from the nitroimidazole radical anion to O2) was presented. Even in the absence of O2, destruction of Type I substrates such as tryptophan, tyrosine and cysteine was observed. PDT experiments in the presence of nitroimidazoles in vitro and in vivo yielded mixed results (Sections 1.5.2 and 1.6), but the development of porphyrins with covalently attached nitroimidazoles in order to improve the effectiveness of PDT photosensitizers remains intriguing. COOH R UP chlorin e6 (R = OH) MACE (R = aspartyl) 1.5 Porphyrins and Ionizing Radiation Photodynamic Therapy Combined with Ionizing Radiation As reviewed by Moan and Berg, some studies combining PDT with radiotherapy have suggested a simple additive effect of the two therapies, whereas others have suggested synergistic effects which were the most pronounced when y-irradiation preceded PDT by about 24 h. 7 Tumor reoxygenation may play a role in the synergy. References on page 22 13 Porphyrins as Radiosensitizers Porphyrins have been reported in publications dating back to 1957 to modify the action of ionizing radiation; in some cases, radioprotection was found while in others radiosensitization was observed.7 In more recent studies, administration of hematoporphyrin derivatives and subsequent tumor irradiation using a ^ Co source showed increased cytotoxicity in vitro and reduced tumor volumes in transplanted tumors in mice, cats and dogs in vivo, in comparison with radiation therapy alone.60"62 O'Hara et al. reported SER values of 2.3 and 2.4 using Chinese hamster fibroblasts incubated with water-soluble cobalt porphyrins (TMPyPCo and TSPhPCo, see below) which were subsequently treated with ionizing radiation.63 Results from our laboratories using water-soluble porphyrins and metalloporphyrins have not demonstrated such high SER values in radiosensitizing experiments.30'64"66 The largest SER (1.22) in hypoxic Chinese Hamster Ovary (CHO) cells was observed using the cationic Co(III)-porphyrin BMPyBNPhPCo. 6 6 , 6 7 BMPyBNPhPCo KATD-F1 References on page 22 14 As mentioned earlier, porphyrins incorporating radiosensitizers (e.g. nitroimidazoles or heterocyclic N-oxides) might be part of an effective anti-cancer therapy (e.g. as radiosensitizers, HSCs and/or PDT agents) given the potential for tumor-localization of these compounds. Thus, it is not surprising that nitroimidazoleporphyrins have appeared in the literature. After this thesis work had begun, Sakata et al. patented the synthesis and use of covalently-bound nitroimidazole-metalloporphyrin complexes, including the Mn-porphyrin, KADT-F1. 6 8 Based on a dose response curve given in the Japanese patent, a modest SER value of 1.39 (measured at 1 % survival) was obtained using KATD-F1 (1.5 mM) in radiosensitizing experiments with hypoxic HeLa cells in vitro. In vivo assays with EMT-6 tumors in Balb/c mice showed a significant decrease in relative tumor volume using KADT-F1 compared to radiation alone. No other covalently-bound rutroimidazole-porphyrin compounds were found in literature searches. Based on these literature results, the development of new, porphyrin-based radiosensitizers seems promising (Section 1.8). 1.6 The Combination of PDT with Hypoxia-Selective Cytotoxins Because PDT can induce hypoxia via vasculature damage or O2 consumption (Section 1.4.3), combination of this treatment with HSCs looks encouraging. Several studies have been reported in the literature which combined PDT with HSCs; some examples include work by Chekulayev et al.,69 Adams et al.,10 Bremner et al.,11 Baas et al.,72 and Santus et al.72 Chekulayev et al. combined metronidazole and PDT using chlorin 0$ in transplanted Ehrlich ascites carcinomas in mice and a marked decrease in tumor size was observed in comparison to use of PDT or metronidazole alone.69 In in vitro experiments with the same cell line, enhanced cell kill was obtained with this combination in hypoxia (!), and the authors suggested the involvement of Type I photodynamic processes (see Section 1.4.3.1 and below). A similar enhancement of cell kill was observed in air; however, using hematoporphyrin derivative (HPD) in place of chlorin e$, a reduction in toxicity was observed which was attributed to a metronidazole-induced decrease in the population of excited triplet state HPD (3P*) molecules. References on page 22 15 Adams et al. described significant delay in regrowth of RIF murine sarcoma tumors treated with PDT using an aluminum sulfonated phthalocyanine in the presence of RSU 1069 or RB 6145 (see below) compared to treatment with PDT or the nitroimidazoles alone.70 The timing of light administration was an important factor, presumably because the distribution kinetics of phthalocyanine localization (vasculature vs. intracellular) affected the severity of induced hypoxia. Other work has not shown such dramatic results. Baas et al. and Bremner et al. studied the HSC tirapazamine in RIF-1 murine sarcomas in conjunction with PDT using Photofrin® and a sulfonated Al-phthalocyanine, respectively.71,72 Both groups reported only modest increases in tumor regrowth delay. In addition, Bremner et al. investigated five other nitroimidazole-, one quinone- and one heterocyclic N-oxide-based HSC using the same model. The other nitroimidazoles and heterocyclic N-oxides were also less effective than RSU 1069- or RB 6145-PDT combinations; however, the quinone (mitomycin-C) was nearly as effective. In in vitro experiments using EMT-6 mouse tumor cells, Santus et al. demonstrated that 2-nitroimidazoles effectively inhibited photosensitization by Photofrin® at low 0 2 concentrations; while under aerobic conditions, little effect was observed.73 They proposed that the Photofrin® triplet state (3P*) quenching by metronidazole (Section 1.4.3.1) became significant at low O2 concentrations and the putative Type I photoprocess products had a much narrower range of chemical reactivity than IO2. These results are similar to those of Chekulayev et al. (see above) who used a closely related photosensitizer, HPD. Thus, there is evidence to indicate that the importance of Type I photoprocesses depends on the photosensitizer. In summary, the combination of HSCs with PDT did not always improve the treatment results. The significandy improved outcomes (using RSU 1069 and RB 6145) were attributed to the hypoxia-selective toxicity of the nitroimidazoles under conditions of PDT-potentiated hypoxia and not to Type I photoprocesses. Nevertheless, the combination of PDT with HSCs is an area NQ, RB 6145 References on page 22 16 which has room for development and effective treatments may be realized via covalendy bound porphyrin-HSC conjugates. 1.7 Other Areas of Interest 1.7.1 Hypoxia Imaging Immunofluorescent imaging of tumor hypoxia is another area of interest. An example of the current methodology involves the use of EF5, a peralkylfluoronitroimidazole which forms cellular adducts under hypoxic conditions.25"27 Monoclonal antibodies (MoAb) have been developed which recognize these cellular adducts with high selectivity; as the MoAb is labelled with a fluorescent marker, [O2] can be quantified and O2 distribution in hypoxic tissues can be determined using fluorescence microscopy. In recent work in our group, several new fluoroalkylnitroimidazoles and their Ru-complexes have been synthesized and investigations into the specificity of the MoAbs for these derivatives and their complexes are currently in progress.74 Other derivatives based on porphyrins incorporating pentafluoroalkyl groups or EF5 itself might be effective hypoxia imaging agents given the tumor-localizing properties of many porphyrins (Sections 1.2 and 1.8). In addition, many porphyrins fluoresce and this may provide a means to image the tissues without employing immunochemical techniques. 1.7.2 Pt-chelating groups Because of the tumor-localizing properties of many porphyrins (Section 1.2), the combination of porphyrins with Pt-based anti-cancer agents (e.g. cisplatin) may yield effective chemotherapeutic and PDT agents. Numerous Pt-porphyrin complexes have appeared in the literature, including those with porphyrin ring-and porphyrin periphery-chelated Pt. Several water-f = \ O II EF 5 References on page 22 17 soluble Pt-porphyrin complexes (e.g. TMPyPPt, TrMPyNPhPPt, see below) have been synthesized and evaluated for their anti-cancer activity, but were shown to have limited effectiveness in in vitro experiments with CHO cells.30'67 In other work, Pasternack et al. demonstrated that TMPPyPPt intercalated into naked DNA. 7 5 However, in such complexes, Pt is chelated by the porphyrin ring and is unlikely to react with DNA in the same manner as cisplatin.22 In contrast, Pt coordinated at the porphyrin periphery may retain its tumoricidal properties. cisplatin TMPyPPt TrMPyNPhPPt BCyPIXPt H3C H3C R = HN""S^NH2 BenPIXPt Brunner et al. reported the preparation and evaluation (in vitro and in vivo) of numerous Pt-porphyrin complexes based on protoporphyrin DC derivatives (e.g. BCyPIXPt and BenPIXPt).76'77 BCyPIXPt (20 p:mol/kg) showed better anti-tumor activity than cisplatin (5 References on page 22 18 limol/kg) in P 388-leukemias and MXT(-) tumors in mice, and this might be attributable to increased [Pt], perhaps because of the tumor localization properties of BCyPIXPt. BenPIXPt (10 |j,M) showed poorer cytostatic activity than cisplatin (10 |LtM) in MDA-MB-231 cell cultures in vitro, but when combined with red-light irradiation, it was more toxic than the cisplatin treatment, and approached the toxicity observed with Photofrin® using red-light irradiation without cisplatin. Given these results, the development and investigation of other side-bound Pt-porphyrin complexes as anti-cancer agents is warranted. 1.8 Thesis Preview: Synthesis and Evaluation of Porphyrin-based Anti-cancer Agents Because of the tumor-localizing behavior of many porphyrins (Section 1.2) and the encouraging results of combining PDT and/or porphyrins with radiation therapy (Section 1.5) or HSCs (Section 1.6), the development of novel porphyrin-based anti-cancer agents and hypoxia imaging agents looks promising and worthwhile. Several approaches were used in this thesis work to synthesize porphyrin conjugates of nitroimidazoles, heterocyclic N-oxides, pentafluoroalkyl moieties and potential Pt-chelators. Development of novel porphyrin-based radiosensitizers was the original goal, but other aspects of these compounds were investigated based on inherent properties of the porphyrin (i.e. photosensitization) and of incorporated moieties (i.e. nitroimidazoles and heterocyclic N-oxides as HSCs). The Synthesis and Modification of Diarylporphyrins (Chapter 2) Diarylporphyrins are potentially useful starting materials for novel anti-cancer agents owing to the unsubstituted, reactive meso-positions to which numerous substituents can be attached. Much of the work in this chapter was based on the synthesis of novel diarylporphyrins and the improved synthesis of 5,15-diphenylporphyrin (DPhP). DPhP was further modified and used in Pd-catalyzed cross-coupling experiments to produce porphyrins such as [5,15-diphenyl-10,20-divinylporphyrinato]Zn(U) (DPhDVPZn, Figure 1.6). The reactivity of the raeso-vinyl groups was then investigated. References on page 22 19 PhCHO + + PhCHO + + DPhP DPhDVPZn Figure 1.6. The Synthesis of DPhDVPZn. The Synthesis and Modification of Tetraarylporphyrins (Chapter 3) The synthesis of tetraarylporphyrins via acid-catalyzed condensation of pyrrole and aromatic aldehydes is more straightforward than that of the diarylporphyrins, but the purification procedure is very chromatography-intensive when mixtures of unsymmetrical porphyrins are produced. Numerous tetraarylporphyrins were made using aromatic aldehydes based on imidazole, pyridine and benzene and combinations thereof. Heterocyclic N-oxides were introduced into the porphyrin via N-oxidation reactions of pyridyl porphyrins to produce a series of (oxidopyridyl)porphyrins (and their corresponding porphyrin N-oxides, e.g. OPyTrPhP-210, see below) and via coupling of tirapazamine to give an aminophenyl-substituted porphyrin (TirapPhTrPhP). Other porphyrin derivatizations and attempted derivatizations included sulfonation and nitration reactions, and metallation with platinum. OPyTrPhP-210 TirapPhTrPhP References on page 22 20 Substituent Manipulation on Protoporphyrin IX (Chapter 4) Protoporphyrin LX dimethylester (PIXDME) is a readily available starting material and the chemical manipulation of its substituents is well documented.78 In one such modification, the vinyl groups of PIXDME were oxidized using Tl(ffl) to yield the bis(dimethylacetal)porphyrin (BDMEtPIXDME),7 9 the porphyrin on which most of the chemistry in this Chapter was based. New substituents were introduced at the 8-and 13-positions via the bis(aldehyde) derivative (BOEtPIXDME) produced by selective acetal removal; subsequent reductive amination and Knoevenagel reactions using BOEtPIXDME were performed. Reduction of the aldehyde moieties, conversion to the 2-bromoethyl derivatives and subsequent SN 2 displacement reactions were used to produce the nitroimidazole-containing porphyrin BNImEtPIXDME and other porphyrins. H3COOC c o o c h b H3COOC COOCH3 H3C00C COOCH3 BDMEtPIXDME BOEtPIXDME BNImEtPIXDME The Cyclic Voltammetry of Selected Porphyrins (Chapter 5) The efficacy of radiosensitizers and HSCs is strongly correlated to their one-electron reduction potentials (Sections 1.4.1 and 1.4.2). Cyclic voltammetry was used to evaluate the redox properties of selected porphyrins and reference compounds; the usefulness of these porphyrins as radiosensitizers and HSCs is discussed by comparing their reduction potentials to those of known radiosensitizers and HSCs. In Vitro Evaluation of Selected Porphyrins (Chapter 6) Although the chemistry developed in this thesis is of prime significance, the evaluation of novel porphyrins in mammalian cells through in vitro experiments at the BC Cancer Research References on page 22 21 Centre (BCCRC) was also valuable. Selected porphyrins were evaluated for their accumulation by cells (fluorescence microscopy and UV-Vis spectroscopy studies), toxicity (including hypoxia-selective toxicity), radiosensitizing and photosensitizing characteristics. Because of the poor solubility of some of these porphyrins, some assays were carried out using Cremophor EL® emulsions or liposomal formulations of the porphyrins. Based on the results of the in vitro assays, the potential of these porphyrins as anti-cancer agents is discussed. Conclusions and Future Work (Chapter 7) Overall conclusions about the chemistry and characterization of the porphyrins are discussed and suggestions for future experiments are made. 1.9 Conclusion There are many reasons to investigate porphyrin-based radiosensitizers and HSCs, incorporating moieties such as nitroimidazoles and heterocyclic N-oxides. The localization behavior of many porphyrins might help to deliver these compounds to tumor tissue, and effective cancer treatment modalities using such porphyrins might include radiation therapy, hypoxia-selective cytotoxicity, PDT, or combinations thereof. References on page 22 22 References for Chapter 1 (1) DiNello, R. K.; Chang, C. K. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. I; pp 289-339. (2) Battersby, A. R.; McDonald, E. In Porphyrins and Metalloporphyrins; K. M . Smith, Ed.; Elsevier: Amsterdam, 1975; pp 61-122. (3) Drabkin, D. L. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. I; pp 29-83. (4) Kim, J. B.; Adler, A. D.; Longo, F. R. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. I; pp 85-100. (5) Treibs, A.; Haberle, N. Liebigs Ann. Chem. 1968, 718, 183-207. (6) Boyle, R.; Dolphin, D. Photochem. Photobiol. 1996, 64, 469-485. (7) Moan, J.; Berg, K. Photochem. Photobiol. 1992, 55, 931-948. (8) Ochsner, M . J. Photochem. Photobiol. B: Biol. 1997, 39, 1-18. (9) Hamblin, M . R.; Newman, E. L. J. Photochem. Photobiol. B: Biol. 1994, 23, 3-8. (10) Pass, H. J. Nat. Cancer Inst. 1993, 85, 443-456. (11) Maziere, J. C ; Morliere, P.; Santus, R. J. Photochem. Photobiol. B: Biol. 1991, 8, 351-360. (12) de Smidt, P. C ; Versluis, A. J.; van Berkel, J. C. Biochemistry. 1993, 32, 2916-2922. (13) Korbelik, M . Photochem. Photobiol. 1992, 56, 391-397. (14) Rosenberger, V.; Margalit, R. Photochem. Photobiol. 1993, 58, 627-630. (15) Korbelik, M . J. Photochem. Photobiol. B: Biol. 1992,12, 107-113. (16) Chaplin, D. J. Int. J. Radiat. One. Biol. Phys. 1992, 22, 685-687. (17) Brown, J. M. ; Giaccia, A. J. Int. J. Radiat. Biol. 1994, 65, 95-102. (18) Denny, W. A.; Wilson, W. R. J. Med. Chem 1986, 29, 879-887. (19) Hall, E. J. Radiobiology for the Radiologist; J.B. Lippincott: Philadelphia, 1994, pp. 133-152. 23 (20) Chaplin, D. J.; Durand, R. E.; Olive, P. L. Int. J. Radiat. One. Biol. Phys. 1986,12, 1279-1282. (21) Chaplin, D. J.; Olive, P. L.; Durand, R. E. Cancer Res. 1987, 47, 597-601. (22) Farrell, N. In Transition Metal Complexes as Drugs and Chemotherapeutic Agents; R. Ugo and B. R. James, Eds.; Kluwer Academic: Dordrecht, 1989; pp 183-207. (23) Chaplin, D. J.; Horsman, M . R.; Peters, C. E.; Trotter, M . J. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M. Fielden and P. Wardman, Eds.; Plenum: New York, 1990; pp 325-336. (24) Adams, G. E. Radiat. Res. 1992,132, 129-139. (25) Lord, E. M. ; Harwell, L.; Koch, C. J. Cancer Res. 1993, 53, 5721-5726. (26) Evans, S. M . ; Joiner, B.; Jenkins, W. T.; Laughlin, K. M . ; Lord, E. M . ; Koch, C. J. Br. J. Cancer 1995, 72, 875-882. (27) Koch, C. J.; Evans, S. M. ; Lord, E. M . Br. J. Cancer 1992, 72, 869-874. (28) Patrick, J.; Hodgkiss, R. J.; Jones, G. W.; Middleton, R. W.; Rami, H. K.; Wardman, P. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden and P. Wardman, Ed.; Plenum: New York, 1990; pp 249-262. (29) Powell, S.; McMillan, T. J. Radiother. Oncol. 1990,19, (30) Ravensbergen, J. R. M.Sc. Thesis, Univ. of British Columbia, 1993. (31) Fielden, E. M. ; Adams, G. E.; Cole, S.; Naylor, M . A.; O'Neill, P.; Stephens, M . A.; Stratford, I. J. Int. J. Radiat. One. Biol. Phys. 1992, 22, 707-711. (32) Wardman, P. Environ. Health Perspec. 1985, 64, 309-320. (33) Parker, E.; Skarsgard, L. D.; Emmerson, P. T. Radiat. Res. 1969, 38, 493-500. (34) Skov, K. A. Radiat. Res. 1987,112, 217-242. (35) Skov, K. A. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden and P. Wardman, Eds.; Plenum: New York, 1990; pp 263-273. (36) Chan, P. K.-L. Ph.D. Thesis, Univ. of British Columbia, 1988. 24 (37) Chan, P. K.-L.; James, B. R.; Frost, D. C ; Chan, P. K. H.; Hu, H.-L.; Skov, K. Can. J. Chem. 1989, 67, 508-518. (38) Yapp, D. T. T. Y. Ph.D. Thesis, Univ. of British Columbia, 1993. (39) Yapp, D. T. T.; Rettig, S. J.; James, B. R.; Skov, K. A. Inorg. Chem. 1997, 36, 5635-5645. (40) Greenstock, C. L. In The Chemistry of Functional Groups: Supplement F: The chemistry of amino, nitroso andnitro compounds and their derivatives, Parti; S. Patai, Ed.; John Wiley & Sons: Chichester, 1982. (41) Adams, G. E.; Flockhart, I. R.; Smithen, C. E.; Stratford, I. J.; Wardman, P.; Watts, M . E. Radiat. Res. 1976, 67, 9-20. (42) Wardman, P.; Priyadarsini, K. I.; Dennis, M . F.; Everett, S. A.; Naylor, M . A.; Patel, K. B.; Stratford, I. J.; Stratford, M . R. L.; Tracy, M. Br. J. Cancer 1996, 74, S70-S74. (43) Edwards, D. I.; Virk, N. S. Int. J. Radiat. One. Biol. Phys. 1992, 22, 677-679. (44) Denny, W. A.; Wilson, W. R.; Hay, M . P. Br. J. Cancer 1996, 74, S32-S38. (45) Cory, J. G. In Textbook of Biochemistry with Clinical Correlations; T. M . Devlin, Ed.; Wiley-Liss: New York, 1997; pp 502-503. (46) Naylor, M . A.; Adams, G. E.; Haigh, A.; Cole, S.; Jenner, T.; Robertson, N.; Siemann, D.; Stephens, M . A.; Stratford, I. J. Anti-Cancer Drugs 1995, 6, 259-269. (47) Walton, M . I.; Wolf, R. W.; Workman, P. Int. J. Radiat. One. Biol. Phys. 1989,16, 983-986. (48) Joseph, P.; Jaiswal, A. K.; Stobbe, C. C ; Chapman, J. D. Int. J. Radiat. One. Biol. Phys. 1994, 29, 351-355. (49) Fitzsimmons, S. A.; Lewis, A. D.; Riley, R. J.; Workman, P. Carcinogenesis 1994,15, 1503-1510. (50) McClelland, R. A. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden and P. Wardman, Eds.; Plenum: New York, 1990; pp 125-135. 25 (51) Denny, W. A.; Wilson, W. R.; Atwell, G. J.; Boyd, M. ; Pullen, S. M . ; Anderson, R. F. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden and P. Wardman, Eds.; Plenum: New York, 1990; pp 149-158. (52) Hairiman, A. In CRC Handbook of Organic Photochemistry and Photobiology; W. M . Horspool and P.-S. Song, Eds.; CRC Press: Boca Raton, 1995; pp 1374-1378. (53) Streitweiser, A.; Heathcock, C. H. Introduction to Organic Chemistry; 3rd ed.; Macmillan: New York, 1985, pp. 1104-1110. (54) Woodburn, K. W.; Vardaxis, N. J.; Hill, J. S.; Kaye, A. H.; Phillips, D. R. Photochem. Photobiol. 1991, 54, 725-732. (55) Korbelik, M. ; Matisic, J.; Krosl, G. SPIE Optical Methods for Tumor Treatment and Detection 1993,1881, 305-312. (56) Foster, T. H.; Gibson, S. L.; Gao, L.; Hilf, R. SPIE Optical Methods for Tumor Treatment and Detection 1992,1645, 105-114. (57) Kuznetsov, S. V.; Bazin, M. ; Santus, R. J. Photochem. Photobiol. A: Chem. 1997,103, 57-62. (58) Bazin, M. ; Santus, R. Photochem. Photobiol. 1986, 43, 235-242. (59) Bazin, M. ; Patterson, L. K.; Ronfard-Haret, J. C ; Santus, R. Photochem. Photobiol. 1988,48, 177-180. (60) Valuckas, K. P.; Didziapetriene, J.; Atkocius, V.; Gricuiete, L.; Luksiene, Z.; Zaleckaite, L.; Rotomskis, R.; Smilgevicius, V. SPIE Optical Methods for Tumor Treatment and Detection 1993,1881, 76-83. (61) Bloznelyte, L. SPIE Optical Methods for Tumor Treatment and Detection 1993,1881, 48-51. (62) Luksiene, Z.; Benediktas, J. SPIE 1993,1881, 67-75. (63) O'Hara, J. A.; Douple, E. B.; Abrams, M . J.; Picker, D. J.; Giandomenico, C. M. ; Vollano, J. F. Int. J. Radiat. One. Biol. Phys. 1989,16, 1049-1052. 26 (64) Meng, G. G.; James, B. R.; Skov, K. A.; Korbelik, M. Can. J. Chem. 1994, 72, 2447-2457. (65) Ware, C. M.Sc. Thesis, Univ. of British Columbia, 1994. (66) Meng, G. G. Ph.D. Thesis, Univ. of British Columbia, 1993. (67) James, B. R.; Meng, G. G.; Posakony, J. J.; Ravensbergen, J. A.; Ware, C. J.; Skov, K. A. Metal-Based Drugs 1996, 3, 85-89. (68) Sakata, I.; Nakajima, S.; Koshimizu, K.; Hiroyuki, T.; Yasushi, I. JP Patent 080 567 682 A2 960312 Heisei, 1994. (69) Chekulayev, V.; Shevchuk, I.; Mihkelsoo, V.; Kallikorm, A. SPIE Optical Methods for Tumor Treatment and Detection 1992,1645, 131-140. (70) Adams, G. E.; Stratford, I. J. Int. J. Radiat. Oncol. Biol. Phys. 1994, 29, 231-238. (71) Bremner, J. C. M. ; Bradley, J. K.; Adams, G. E.; Naylor, M . A.; Sansom, J. M. ; Stratford, I. J. Int. J. Radiat. One. Biol. Phys. 1994, 29, 329-332. (72) Baas, P.; Oppelaar, H.; Stavenuiter, M . Int. J. Radiat. One. Biol. Phys. 1993, 27, 665-670. (73) Santus, R.; Stobbe, C. C ; McPhee, M . S.; Chapman, J. D. Radiat. Res. 1992, 31-37. (74) Baird, I. R. Work in progress. (75) Pasternack, R. F.; Brigandi, R. A.; Abrams, M . J.; Williams, A. P.; Gibbs, E. J. Inorg. Chem. 1990, 29, 4483-4486. (76) Brunner, H.; Maiterth, F.; Treittinger, B. Chem. Ber. 1994,127, 2141-2149. (77) Brunner, H.; Obermeier, H.; Szeimies, R.-M. Chem. Ber. 1995,128, 173-181. (78) Smith, K. M. ; Cavaleiro, J. A. S. Heterocycles 1987, 26, 1947-1963. (79) Kenner, G. W.; McCombie, S. W.; Smith, K. M . Liebigs Ann. Chem. 1973, 1329-1338. 27 Chapter 2 The Synthesis and Modification of Diarylporphyrins 2.1 Introduction In recent years, P-unsubstituted diarylporphyrins and their metal complexes have been used in a variety of systems. Literature describing such systems has included: the synthesis of a porphyrin cyclodextrin complex,1 liquid crystal porphyrins,2"4 raesolinked porphyrin dimers, trimers and tetramers,5 strapped porphyrins for modeling heme systems6 and dinuclear metalloproteins,7 and dendritic porphyrins for modeling globular heme proteins.8'9 Diphenylporphyrin (DPhP), the simplest and most easily accessible 3-unsubstituted diarylporphyrin, was first reported in 1968 by Treibs and Haberle.10 An improved synthesis of DPhP and other |3-unsubstituted diphenylporphyrins from a variety of substituted benzaldehydes and bis(2-pyrrolyl)methane was later described by Manka and Lawrence.1 The derivatization of DPhP and its metal complexes has been the focus of several recent reports: the synthesis of novel meso-diphenylbenzochlorins,11 mesohalogenation (e.g. with Br, I) and nitration,12"14 the raesohalogen derivatives were further modified by metallation (e.g. with Zn, Ni, Cu) and subsequent Pd-catalyzed cross-coupling reactions with organotin or organozinc reagents14,15 and Heck alkynylations.12'13'16"18 These elegant Pd-catalyzed reactions made possible the synthesis of symmetrically substituted diphenylporphyrins incorporating a wide variety of meso-substitutents (e.g. alkyl, alkenyl, alkynyl, aromatic, and steroid groups) for purposes such as new photodynamic therapy (PDT) agents, the synthesis of supramolecular multiporphyrin-complexes, and non-linear optical applications. As described in Chapter 1, porphyrins incorporating heterocyclic N-oxides, nitroaromatic groups and ligands for Pt-chelation were considered desirable compounds for new cancer therapy agents. Similarly, sulfonatophenyl and N-methylpyridinium moieties have been incorporated into porphyrins19"23 to enhance their water-solubility, thus facilitating the biological assays performed at the B.C. Cancer Research Centre (BCCRC). To this end of incorporating these moieties into References on page 79 28 porphyrins, the chemistry of diarylporphyrins herein focuses on the synthesis of novel diarylporphyrins, improved synthesis of DPhP, introduction of new functional groups into the mew-positions of diphenylporphyrin via Pd-catalyzed cross-coupling methodologies,14 and derivatization of (5,15-diphenyl-10,20-divinylporphyrinato)zinc(U) (DPhDVPZn). The chemistry of the diarylporphyrins is considered the least successful area of porphyrin chemistry in this thesis work. With a few exceptions, the syntheses of new products in this area generally gave very low yields and numerous side-products. The hydroboration attempts with DPhDVPZn, borane and N-bromosuccinimide, for instance, gave at least 7 products, which were unstable, and even after isolation transformed to other products (Section 2.2.6.3). Nevertheless, some of the systems yielded interesting results. 2.2 Experimental 2.2.1 Materials and Instrumentation All solvents were reagent grade or better. When needed, dry CH2CI2 and dry THF were distilled from CaH2 and Na/K alloy, respectively. Solvents were degassed by standard freeze-pump-thaw procedures. Unless otherwise indicated, all commercial chemicals were used as purchased from Aldrich Chemical Co., Strem, or Sigma Chemical Co. Pyrrole was dried over and distilled from CaH2 or vacuum transferred before use. Aromatic aldehydes (e.g. benzaldehyde, 4-pyridine carboxaldehyde) and TFA were freshly distilled before use. NBS and TsCl were recrystallized before use. Reaction products were dried in vacuo at 78 °C overnight unless otherwise noted. To remove water from organic phases after aqueous washes, the solutions were stored -30 min over MgSC«4(anh) or MeCN was added and the water removed azeotropically when the solvent was removed under reduced pressure. The alumina used for chromatography was Fisher, activity I, and was deactivated by adding water to give a standard range of activities.24 Unless otherwise noted, the silica gel used in the flash chromatographies was Merck Silica Gel 60,230-400 mesh, whilst Rf-values were measured on Merck silica dc Al sheets (silica gel 60 F254). Preparative-tics of porphyrin free-References on page 79 29 bases were performed with Merck silica gel 60, 0.5 mm plates (without indicator, to avoid metallation of the products) and, for metalloporphyrins and other compounds, preparative-tics were performed with silica gel 60 F254, 1.0 mm plates. RPtlc was performed on RP-18 F254S coated, glass-backed plates (0.25 mm) from Merck. Some products were purified on a Harrison Research model 7924T Chromatotron (rotating disk dc) using silica or AI2O3 as the stationary phase. In general, the visualization of non-porphyrinic compounds was achieved with the aid of a UV lamp, but imidazoles and bispyrrolic compounds were visualized on tic plates by exposing the developed tics to iodine vapor and bromine vapor, respectively. The *H and 1 3 C NMR spectra were recorded on BRUKER AC-200, VARIAN XL-300, or BRUKER WH-400 instruments and J values are given in Hz. UV-Vis spectra were recorded on an HP8452A photo-diode array spectrophotometer (± 2 nm) and are reported as: (cone, for £ measurement, solvent) nm (login £). Molar absorptivity (8) values were measured for porphyrins determined to be pure by elemental analysis. Infrared spectra were recorded as KBr pellets (by compression of a mixture of finely ground KBr and the compound) or as a thin film on a KBr disk (from slow evaporation of a solution of the compound on the surface of a KBr disk) on an ALT Mattson Genesis Series FTIR instrument. Elemental analyses were performed by Mr. P. Borda of the departmental microanalytical laboratory on a Fisons CHN/O Analyzer, model 1108. The high and low resolution mass spectra were obtained at the departmental mass spectrometry service laboratories (G. Eigendorf, Director) on KRATOS MS50 (EI), KRATOS MS80 (DCI and GCMS), KRATOS Concept IIHQ (LSIMS) or BRUKER Biflex MALDI-TOF mass spectrometers. When analyzed with EI ionization techniques, the mass spectra of free-base porphyrins often showed a peak at 61 m/e higher (M+ -2H -HCU) than the parent mass due to a gas phase metallation reaction with Cu from the ionization source. 2.2.2 Porphyrin Precursors The syntheses of bis(2-pyrrolyl)methane (3) and l,l',2,2'-tetrakis(2-pyrrolyl)ethane (4) were reported recently as part of a collaborative effort with Dolphin's group;25 the work described References on page 79 30 improved syntheses of DPhP and its precursors. Al l procedures but the synthesis of 3 via Method JJ (which was performed by C. Bruckner) were carried out by the author. 2.2.2.1 Bis(2-pyrrolyl)thioketone (1) Based on the published literature procedure,26 pyrrole (15.4 raL, 222 mmol) in dry ether (150 mL) under a steady stream of N 2 was added dropwise to a solution of thiophosgene (8.2 mL, 108 mmol) in toluene at 0 °C in a 1 L, 3-neck round-bottom flask. After the addition was complete, the solution was stirred for 20 min at 0 °C; MeOH/H20 (8:2 350 mL) was added, and the solution was stirred for another 45 min. Silica (70-230 mesh, -15 g) was added, the solvent removed under reduced pressure, and the product mixture loaded onto a silica column (8 cm diam. x 15-20 cm) and eluted with CH2CI2 until the eluate contained no thioketone. The crude product (-13 g) was recrystalhzed in batches from EtOH/FfeO, but the recrystalhzation was complicated by the opacity of the solution. The crude thione was dissolved in a slight excess of hot EtOH, and distilled water was added at reflux temperature until the solution turned a lighter shade of brown/red; then more EtOH was added to redissolve the thione. The hot solution was filtered, and lustrous maroon needles/plates crystallized on cooling. Combining two syntheses of this scale yielded -25 g (-65%) of 1 (m.p. 96-99 °C, lit. 96-98 °C), which contained trace amounts of water, according to the iH-NMR spectrum. Tic of this product (silica/CH2Cl2) showed a trace of impurity material at the baseline. 1 was used without further purification in subsequent reactions. XH-NMR (200 MHz, CDCI3) 8 9.80 (s br, 2H, pyrrole N-#), (7.19 (m, 2H), 7.03 (m, 2H), 6.40 (m, 2H), pyrrole ring H). The iH-NMR data agree with those reported in the literature. 2.2.2.2 Bis(2-pyrrolyl)ketone (2) According to the procedures described in the literature,26 1 (4.8 g, 2.7 mmol) and NaOH (7 g) were dissolved in 95% aq. EtOH (600 mL) and the mixture was warmed to 40 °C. Then 5 % H2O2 (aq.) was added and the solution stirred for 10 min. Distilled H2O (300 mL) was added, the solution volume was reduced to -400 mL under reduced pressure and the resulting precipitate was filtered off and subsequendy recrystallized from distilled H 2 0 to yield 3.3 g (21 mmol, -75 %) of the title compound. iH-NMR (200 MHz, CDCI3) 8 10.05 (s br, 1.7H, m pyrrole N-fl), (7.09 (m, References on page 79 31 2H), 7.00 (m, 2H), 6.28 (m, 2H), pyrrole ring H); LR-MS (EI) m/e = 160 (82, M+), 94 (70, M+ -pyrrolyl). The mass spectrometry data match those reported in the literature, but the iH-NMR data were not reported previously. Yields varied from 50 to 85 %. The product was used without further purification to produce bis(2-pyrrolyl)methane (Section 2.2.2.3, Method IV). 2.2.2.3 Bis(2-pyrrolyl)methane (3) Method I Compound 1 (2.0 g, 11.3 mmol) was dissolved in EtOH (-50 raL) containing cone, aqueous NH3 or E13N (0.5 mL), and a 50 % aqueous slurry of Raney Ni (4-10 mL) was added; H2 was then bubbled through the solution. Hydrodesulfurization set in instantaneously and was, depending on the activity of the Raney Ni, completed within 20 min to 12 h. In some cases, when the Raney Ni was several years old, no conversion of the thione to dipyrromethane was observed, and tic gave evidence of starting material and a N i 2 + dipyrrothioketone coordination complex.27 In cases where tic revealed the presence of starting material after several hours of stirring, more Raney Ni was added until all thione was consumed, at which time the Raney Ni was filtered off through a short plug of Celite® and washed with EtOH until the rinsings contained no more 3. The combined filtrates were reduced in vacuo to an oil, taken up in CHCI3 and chromatographed on basic alumina (activity n/III)/CH2Cl2 (4 x 12 cm) or on silica gel/CHCl3. The first fraction was collected and evaporated to dryness to yield 0.805 g (49 %) of 3 as an off-white crystalline solid. Yields typically 40-60%. m.p. 73 °C (lit.28 73 °C); *H-NMR (200 MHz, CDCI3) 8 7.65 (br s, 2H, pyrrole N-H), (6.60 (dd, 2H), 6.15 (dd, 2H), 6.05 (m, 2H), pyrrole ring H), 3.95 (s, 2H, CH2)\ 1 3 C-NMR (50 MHz, CDCI3) 8 129.1, 117.4, 108.3, 106.5, 26.4; Analysis calc'd for C 9 H 1 0 N 2 : C, 73.95; H, 6.90; N, 19.16; found: C, 73.96; H, 6.70; N, 19.08. This method was scaled up to 9 g (51 mmol) of 1 with no loss of efficiency, producing 3.21 g (22 mmol, 43 %) of 3. The 1H-and 1 3 C-NMR data differ slighdy from those reported in the literature29 (measured in DMSO-cfc), but are in agreement with other mew-substituted bis(2-pyrrolyl)methanes.30 The products, following the bis(2-pyrrolyl)methane on the chromatography column from the reduction of 1, were eluted with CH2Cl2:MeOH (20:1) and returned to a Schlenk tube with References on page 79 32 more Raney Ni and base. After 1 h, however, no additional 3 was produced, suggesting that the materials are side-products and not intermediates. Method II: Compound 1 (200 mg, 1.14 mmol) was dissolved in dry THF (10 mL), and at 0°C under anhydrous conditions LAH (150 mg, 3.75 mmol) was added in portions to the orange solution over 2 h. The mixture was stirred for an additional 12 h at ambient temperature. The then yellow solution was quenched by the addition of Glauber's salt (Na2S04 • 10 H2O), diluted with CHCI3 (10 ml) and filtered through Celite®. The filtrates were evaporated and the oily residue was chromatographed on a short column (10 x 2 cm sihca/CHCl3). The first colorless fraction was collected to produce, after evaporation of the solvent in vacuo, 130 mg (80 % yield) of the analytically pure 3. Method III Compound 1 (1.0 g, 5.7 mmol) and KOH (0.2 g) were dissolved in 95% EtOH and N a B H 4 (1.0 g, 26.5 mmol) was added. The mixture was refluxed under N2 for 3 h or until no 1 remained, as indicated by tic. The pale orange solution was purged with N2 to help facilitate removal of sulfurous products while the solution was cooled to room temperature. Acetone (5 mL) was added, and once effervescence ceased, basic alumina (II/III) (~5 g) was added and the solvent removed. The product on alumina was then loaded onto a basic alumina(II/m) column (100 g, 10 cm x 3 cm) and eluted with CH2CI2 until the eluate contained only traces of 3. The solvent was removed to yield 0.695 g (84% yield) of analytically pure 3, m.p. 74.5-75.5 °C, which had a slight sulfurous odor. It was noticed that the reduction of the final traces of 1 proceeded very slowly, even with addition of more NaBFL} and further refluxing. Unconverted 1 co-eluted with the tail end of 3 and smelled strongly of sulfurous compounds. Attempts were made to identify these sulfurous byproducts. After a successful reduction had been completed, the volatile components of the reaction mixture were vacuum-transferred from the reaction flask and the mixture was analyzed by GCMS and iH-NMR. Two products were seen by GCMS; the first and second products had 'parent' peaks at 355 and 341; both had equally References on page 79 33 intense peaks at ' M + ' -88. iH-NMR yielded no useful information, as the solutions were too dilute. No further attempts were made to identify these compounds. Method IV Based on published procedures for related bis(2-pyrrolyl)ketones,31 2 (1.0 g, 6 mmol) was dissolved in dry THF (30 mL) under dry N 2 . To this solution was added BH3»THF (20 mL of a 1.0 M solution in THF, 20 mmol) and the reaction mixture was stirred for 15 min at room temperature. According to tic on silica/CH2Cl2, the reaction was incomplete; more BH3»THF solution (10 mL) was added and the reaction was stirred for another 10 min. Then MeOH (100 mL) was added slowly and the solution was subsequently refluxed for 45 min. The product was isolated by chromatography on silica/CH2Cl2 or steam distillation to yield 3 (0.3 g, -35%). 2.2.2.4 l,l',2,2-Tetrakis(2-pyrrolyI)ethane (4) All products remairting on the chromatography column after elution of 3 (Section 2.2.2.3, Method I) were washed off (5 % MeOH in CH2CI2). These fractions were evaporated in vacuo onto a few grams of silica and then initially purified on silica/(hexanes:Et20 3:2 to 1:1), the first three bands from this column being collected; no further purification was achieved with a second silica/(hexanes:Et20 3:2 to 1:1) column. Attempts were also made to separate the three components on a Chromatotron, loading as a THF solution, then drying and eluting with CH2CI2 or Et20/hexanes. Finally, the mixture was chromatographed on silica/(toluene:THF 20:1 to 10:1). Following residual 3, 4 eluted as a colorless fraction. Evaporation to dryness yielded 4 as an off-white to brown crystalline solid in 3 % yield (105 mg). m.p. 153-155° C, decomp.; R f = 0.78 (silica/CHCl3); 'H-NMR (200 MHz, CDCI3) 8 7.6 (br s, 4H, pyrrole N-H), (6.75 (m, 4H), 6.09 (dd, 4H, J 4.3), 5.88 (m, 4H), pyrrole ring H), 4.80 (s, 2H, CH-CH); ^C-NMR (75 MHz, CDCI3) 8 131.2, 117.3, 112.2, 108.5, 107.4, 42.6; LR-MS (+LSIMS) m/e = 289 (22, M+ -H), 222(30, M+ -H -pyrrolyl), 145 (100, M+ -bis(2-pyrrolyl)methyl); (EI 220°C) m/e = 290 (10, M+), 288 (20, M+ - 2H), 222 (12, M+ -pyrrolyl), 145 (100, M+ -bis(2-pyrrolyl)methyl); HR-MS (+LSIMS) calc'd for C 1 8 H 1 7 N 4 (-H): 289.14532, found 289.14439. Analysis calc'd for C 1 8 H 1 8 N 4 : C, 74.46; H, 6.25; N, 19.29; found: C, 74.34; H, 6.49; N, 19.15. In the solid state 4 References on page 79 34 exhibited noticeable darkening over several weeks, but still retained a clean ipi-NMR spectrum after several months storage in air in a tightiy capped vial. 2.2.2.5 (4-Pyridyl)bis(2-pyrrolyl)methane (5) The methods described by Lee and Lindsey30 and by Nagarkatti and Ashley32 were the basis for the conditions used here. Pyrrole (2.9 mL, 41.8 mmol) was purged with N 2 for 10 min and then acidified with anhydrous HCl(g) for ~5 min. Residual HC1 was purged from the reaction vessel by bubbling N 2 for several more min, and 4-pyridine carboxaldehyde (0.1 mL, 1.05 mmol) was added dropwise. [Alternatively, the pyrrole and aldehyde were combined and dry HC1 (g) was bubbled through the solution for ~5 min]. The solution turned deep red after the addition of the first drop of aldehyde. The mixture was stirred for 15 min and the excess pyrrole was removed under high vacuum to yield 600 mg of a product mixture (a 100% yield of pure 5 would be 0.273 g). The product mixture was dissolved in CH 2C1 2 (50 mL) and cone. NH4OH was added dropwise until the red color disappeared. The aqueous phase was washed several times with CH 2C1 2, the organic extracts dried over K 2C03 ; andthe solvent removed. The mixture was chromatographed on silica/(Et20:acetone:Et3N 3:1:0.01) and the eluate collected in fractions, combining those containing the desired product but which still contained some "low Rf" impurities. The mixture was then sublimed under high vacuum at 100-120 °C onto a water-cooled cold finger to yield 0.055 g (25% yield) of a tan solid, which still contained some low Rf impurities as seen by tic and subsequent Br 2 vapor exposure. Some unsublimed 5 remained in the bottom of the sublimation apparatus. Purer product, as judged by !H-NMR, was obtained when the product was first sublimed and then chromatographed on sihca/(CH2Cl2:MeOH:Et3N 100:2.5:1) and the eluate was collected in fractions. Low Rf side-products collected from this column are probably other oc-substituted pyrrolic compounds, as judged by iH-NMR. Rf = 0.55 (silica/(CH2Cl2:MeOH 50:1)); iH-NMR (200 MHz, CDCI3) 8 8.45 (dd, 2H, 2,6-pyridyl-tf), 8.05 (s br, 2H, pyrrole N-H), 7.12 (dd, 2H, 3,5-pyridyl-//), (6.72 (m, 2H), 6.15 (m, 2H), 5.89 (m, 2H), pyrrole ring-//). 5.45 (s, 1H, methyl-//); LR-MS (+LSIMS) m/e = 224 (40, M + 1 ) , 158 (30, M + 1 -pyrrolyl), 145 (25, M + 1 -pyridyl -H), 118 (100, M+1 -pyridyl -CH2N); HR-MS (+LSIMS) References on page 79 35 calc'd for C 1 4 H 1 4 N 3 : 224.11877, found 224.11798. The !H-NMR data differ with those reported in the literature32 (see Section 2.3.1). 2.2.3 Diarylporphyrins With the exception of TEtTMeBPyP (Section 2.2.3.6), the diarylporphyrin synthesis conditions described by Manka and Lawrence1 were the basis of the conditions used here. Chromatography was the usual method of purification; however, isolation of DPhP was optimized through chemical reduction of the TCQ to obviate chromatography. 2.2.3.1 5,15-Diphenylporphyrin (DPhP) In an oven-dried, 1-L, 3-neck flask purged with N 2 , 3 (0.294 g, 2.0 mmol) was dissolved in dry, degassed CH2CI2 (500 mL). To this solution were added benzaldehyde (215 |iL, 2.1 mmol) and TFA (30-35 p:L), and the contents were stirred under an N2 flow in the dark for 15 h. Then TCQ (1.5 g, 6.1 mmol) was added and the solution refluxed in air for 75 min. The mixture was then stirred with an excess of aq. Na2S20 4 and MeOH (100 mL) until all the TCQ was reduced, as indicated by tic (silica/cyclohexane:toluene 1:1). The organic phase was separated, the solvent removed, and the residue taken up in MeOH/H20 (95:5); the resulting suspension was filtered to isolate insoluble DPhP, which was then washed until the solid was free of dihydroxytetrachlorobenzene (Rf = 0.4, silica/CH2Cl2). The filtrant was taken up in CH2CI2 (100 mL) and the mixture filtered through a plug of alumina and a plug of silica to provide, after evaporation of the solvent and subsequent trituration with MeOH/H20 (95:5), DPhP in 42 % yield (0.195 g) as a dark purple, microcrystatline material. Yields were typically 25-40 %, but on occasion reached 50 %. iH-NMR (300 MHz, CDCI3) 8 10.31 (s, 2H, meso-H), (9.38 (d, 4H, J 4.5), 9.07 (d, 4H, J 4.5), (3-pyrrole-H), 8.27 (dd, 4H, 2,6-phenyl-/7), 7.80 (m, 6H, 3,4,5-phenyl-//), -3.12 (s br, 1H, pyrrole N-fl); 1 3 C-NMR (75 MHz, CDC1 3) 8 147.2, 145.2, 141.4, 134.8, 131.6, 131.0, 127.7, 127.0, 119.1, 105.2; UV-Vis ( l . l x l 0 " 6 M , CH 2C1 2) 406 (5.59), 502 (4.24), 536 (3.72), 574 (3.73), 630 (3.20) nm; HR-MS (EI) calc'd for C 3 2 H 2 2 N 4 : 462.1844, found 462.1823; Analysis calc'd for C 3 2 H 2 2 N 4 : C, 83.09; H, 4.79; N, 12.11; found: References on page 79 36 C, 82.96; H, 4.86; N, 11.96. The !H-NMR and UV-Vis data agree with those reported in the literature.1 This synthesis and characterization of DPhP were recently reported as part of a collaborative effort with Dolphin's group25; the 1 3C-NMR, molar absorptivity values and mass spectrometry data were not reported prior to this. The same procedure was repeated but using Fe(II)-phthalocyanine (0.114 g, 0.2 mmol) and TCQ (0.05 g, 0.2 mmol) as the oxidant,33 with the mixture being refluxed in air for 75 min. The product was pre-dried on silica (~2 g), flash chromatographed on silica/CH2Cl2, and subsequently precipitated from CH2CI2 with MeOH/H 20 (95:5). The yield was 0.082 g (-20 %) of brilliant purple flakes, which were free from TCQ by tic, and had characterization data identical to those of DPhP. The advantage of this method is the straightforward purification, although the yield is lower than by using TCQ alone. 2.2.3.2 5-PhenyIporphyrin (PhP) This porphyrin was produced in trace amounts in the synthesis of DPhP and was isolated by repeated preparative-tic of a sample of DPhP (Section 2.2.3.1) on silica/(toluene:cyclohexane 1:1) [Rf = 0.33, that of DPhP = 0.45]. iJT-NMR (300 MHz, CDCI3) mew-resonances could be seen at 8 10.31 and 10.25 and a pyrrole N-H signal at 8 -3.8, but the spectrum was otherwise too weak to reliably pick out other signals. UV-Vis (CH2C12) 400,496,528, 568, 630 nm; LR-MS (EI) m/e = 386 (100, M+), HR-MS (EI) calc'd for: 386.15314, found 386.15245. 2.2.3.3 5,15-Bis(4-nitrophenyl)porphyrin (BNPhP) Compound 3 (0.051 g, 0.35 mmol) and 4-nitrobenzaldehyde (0.053 g, 0.35 mmol) were dissolved in dry CH2CI2 (100 mL) under N2 and TFA (-8 |iL) was added to initiate the reaction. The mixture was stirred for 17.5 h in the dark under a slow stream of N 2 . TCQ (0.33 g, 1.3 mmol) was added and the solution refluxed for 45 min. Chromatography on silica/CHCl3 of the reaction mixture yielded TCQ as the first band, which overlapped with the main porphyrin band. A trace amount of some porphyrin (UV-Vis 400, 496, 532, 570, 622 nm) eluted before BNPhP, but was not investigated further. The BNPhP product streaked on the column and required > 250 mL of eluent for collection. The yield was -5 mg (5 %). Subsequent analyses were difficult, as References on page 79 37 the solubility of this porphyrin was very low. TFA was added in small percentages to facilitate the dissolution of the porphyrin. ifl-NMR (200 MHz, CDC13/TFA) 5 10.97 (s, 2H, meso-H), 9.56 (d, 4H, J 5, P-pyrrole-/7), 8.98 (d, 4H, J 5, fi-pyrrole-/T), 8.88 (d, 4H, J 9, 2,6-phenyl-#), 8.70 (d, 4H, J 9, 3,5-phenyl-//); UV-Vis (CH2C12) 294, 410, 502, 540, 576, 632 nm; LR-MS (EI) m/e = 552 (64, M+), 536 (5, M+ -O), 522 (45, M+ -NO), 506 (12, M+ -N0 2), 492 (100, M+ -2NO), 460 (12, M+ -2N02); HR-MS (EI) calc'd for C 3 2 H2oN 6 0 4 : 552.15460, found 552.15518. 2.2.3.4 5,15-Bis(4-pyridyl)porphyrin (BPyP) Compound 3 (0.025 g, 0.17 mmol), 4-pyridine carboxaldehyde (16 (iL, 0.17 mmol) and 1 drop of DCA were mixed in dry CH 2C1 2 (50 mL) under dry N 2 . After the mixture was stirred for 48 h, TCQ (0.15 g, 0.6 mmol) was added and the solution refluxed for 1 h. The porphyrin was isolated by chromatography on silica/(CH2Cl2:MeOH, 25:1). Isolated yield < 0.5 mg (< 1 %). !H-NMR (CDCI3/CD3OD) [BPyP had poor solubility in CDCI3, so CD3OD was added to help dissolve it.] 5 10.25 (s), 9.30 (d), 8.85 (d), 8.35 (d), 8.15 (d); the integrations did not match, as other impurities were present; UV-Vis (CHCl3/MeOH) 406, 500, 534, 574, 628 nm; LR-MS (EI) m/e = 525 (100, M+ -2H + Cu), 464 (65, M+); HR-MS (EI) calc'd for C 3 o H 2 0 N 6 : 464.17405, found 464.17496. BPyP had good solubility in pyridine-^, but when the ^ - N M R spectrum was obtained in this solvent, the resonances were swamped by pyridine signals because of the small amounts of porphyrin used. 2.2.3.5 5,15-Bis(l-oxido-4-pyridyl)porphyrin (BOPyP) With Manka and Lawrence's conditions1 BF3»MeOH, TFA, TCA and DCA were investigated as acid catalysts. To four separate reaction mixtures containing 4-pyridine carboxaldehyde N-oxide (0.023 g, 0.17 mmol) and 3 (0.025 g, 0.17 mmol) in dry CH 2C1 2 (50 mL) was added ~8 |iL of BF3»MeOH (~9 M in MeOH), TFA, TCA or DCA and the reaction mixtures stirred under N 2 for 48 h. TCQ (0.16 g, 0.65 mmol) was added to each reaction vessel and the solution refluxed for 1 h. The product mixture was purified by chromatography on silica/(Et20:CH2Cl2, 1:1) which removed any excess TCQ or non-References on page 79 38 polar pyrrolic products. The eluent was changed to CH2Cl2/MeOH (10:1) and two porphyrins were eluted, the less polar of the two being the major product, which was retained, while the minor product was discarded. Because of the low yield, the products from the TFA-, TCA- and BF3»MeOH- catalyzed reactions were combined and weighed together to give 6 mg of product (-4.5% average yield). See Table 2.1 (Section 2.3.3) for the relative yield of each reaction. The DCA-catalyzed reaction product was weighed separately to give only 0.6 mg, or 1.4% yield. 1 H -NMR (CDCI3/CD3OD) 5 10.29 (s, 2H, meso-H), (9.36 (d, 4H), 8.92 (d, 4H), f3-pyrrole-#), 8.54 (d, 4H, 2,6-(l-oxidopyridyl)-f0, 8.14 (d, 4H, 3,5-(l-oxidopyridyl)-#); UV-Vis (CHCl3:MeOH 20:1) 412, 504, 542, 580, 632 nm. LR-MS (MALDI-TOF) m/e = 497 (20, M + 1 ) , 481(30, M + 1 -O), 464 (30, M + 1 -20 -H). Attempts to analyze this product by MS (EI) showed only the deoxygenated product (464, M + ) . Other l-oxido-4-pyridylporphyrins had similar properties (see Chapter 3). With refluxing acetic acid Refluxing an acetic acid (2 mL) solution of 3 (0.025 g, 0.17 mmol) and 4-pyridine carboxaldehyde N-oxide (0.023 g, 0.17 mmol) for 45 min yielded 0.7 mg of the 5,15-bis(l-oxido-4-pyridyl)porphyrin (<2 % yield) after chromatography on silica/(CH2Ci2:MeOH 20:1). 2.2.3.6 3,7,13,17-Tetraethyl-2,8,12,18-tetramethyl-10,20-bis(4-pyridyOporphyrin (TEtTMeBPyP) Following the conditions described by Paine,34 ethyl 4-ethyl-3,5-dimethyl-pyrrole-2-carboxylate (6) was synthesized from ethyl acetoacetate and 3-ethyl-2,4 pentanedione35 (37% yield) (Scheme 2.1). The 5-methyl group on the pyrrole was subsequently oxidized with Pb(OAc)4 to yield the ethyl 5-acetoxymethyl-4-ethyl-3-methyl-2-pyrrole carboxylate (7) (45-70% yield) and then treated with HC1 in refluxing ethanol to produce (5,5'-bis(ethoxycarbonyl)-3,3'-diethyl-4,4'-dimethyl-2,2'-bispyrrolyl)methane (8)36 (11-47%). The ethoxycarbonyl groups were subsequently cleaved with KOH in refluxing EtOH/H.20 to yield 3,3'-diethyl-4,4'-dimethyl-2,2'-bispyrrolylmethane (9)37 (-5% yield, although the literature reports a quantitative yield.)38 9 was then mixed with one equiv. of 4-pyridine-carboxaldehyde in MeOH, and the mixture was treated References on page 79 39 with one equiv. of TsOH; the mixture was stirred for 24 h and then treated with TCQ to produce TEtTMeBPyP38 (Scheme 2.1). Some porphyrin was formed, but did not approach the reported yield of 67 %. The silica column used for purification (CH2Cl2:MeOH 20:1) had a relatively large amount of a blue compound at the top of the column, similar to that seen with the porphyrin synthesis involving 4-pyridine carboxaldehyde and strong acids (see Chapter 3). ^H-NMR (200 MHz, CDC13) 8 10.26 (s, 2H, meso-H), 9.02 (q, 4H, 2,6-pyridyl H), 8.04 (q, 4H, 3,5-pyridyl H), 4.00 (q, 8H, p-C/72-CH3), 2.51 (s, 12H, §-CH3), 1.76 (t, 12H, p-CH 2-C# 5), -2.35 (s br, <2 H, pyrrole N-//); UV-Vis (CH2Cl2/MeOH) 406, 506, 538, 574, 626 nm; LR-MS (EI) m/e = 632 (100, M + ) . The UV-Vis and mass spectrometry data agree well with those reported in the literature; the ^ - N M R data differ slightly from those in the literature, where the spectrum of TEtTMeBPyP was measured in CDCI3/CD3OD. o o TEtTMeBPyP Scheme 2.1. Synthesis of TEtTMeBPyP. Reaction conditions: (i) 1) NaN0 2, H 2 0 , CH3COOH 2) Zn; (ii) Pb(OAc)4, CH3COOH; (iii) EtOH, cone. HC1; (iv) NaOH, EtOH, H 2 0 , heat; (v) 1) MeOH, TsOH 2) TCQ. 2.2.3.7 Me so -linked DPhP Dimer A few attempts were made to synthesize a porphyrin dimer bound through the meso-References on page 79 40 positions (see Section 2.3.3). Under conditions similar to those used in DPhP syntheses (in CH2CI2 with TFA catalysis, Section 2.2.3.1, or in refluxing propionic acid) with 1:2:4 or 1:1:3 ratios of 4:3:benzaldehyde, DPhP was isolated in -20% yield. No evidence for a dimer was seen by UV-Vis or mass spectroscopies. The !H-NMR spectrum did contain some evidence for another product, including a weak meso-H signal (slightly downfield from and in a 1:3 ratio to that of DPhP), a weak downfield quartet at 8 9.45 in the P-pyrrole region and a new pyrrolic N-H signal (1:4 ratio to that of DPhP). These data do not match those reported by Osuka and Shimidzu for a meso-linked metalloporphyrin dimer in which the P-pyrrole protons adjacent to the meso-link are shifted upfield.5 In any case, the porphyrin dimer was not formed in large quantities. 2.2.4 Preparation of Reagents for Pd-catalyzed Cross-coupling Reactions DBrDPhPZn was prepared from DPhP according to the procedures described by DiMagno, et a l . 1 4 In general, all Sn reagents were prepared under dry N 2 in an oven-dried Schlenk tube. 2.2.4.1 5,15-Dibromo-10,20-diphenylporphyrin (DBrDPhP) DPhP (0.025 g, 0.054 mmol) was dissolved in CHCI3 (15 mL) and pyridine (50 |iL) was added to the solution. Then NBS (0.020 g 0.11 mmol) was added, and the solution was stirred for 15 min; acetone (10 mL) was then added and the solution stirred for another 15 min. The solvent was removed on a rotary evaporator, and the product washed with MeOH (-25 mL), to give 5,15-dibromo-10,20-diphenylporphyrin (22 mg, 82% yield). This reaction was scaled up to 1.1 g (0.8 mmol) of DPhP without loss of efficiency. The best results were obtained when the reaction was kept at 0 °C and the NBS was added in portions. iH-NMR (200 MHz, CDCI3) 5 (9.60 (d, 4H), 8.80 (d, 4H), P-pyrrole H), 8.10 (m, 4H, 2,6-phenyl H), 7.70 (m, 6H, 3,4,5-phenyl H), -2.75 (s, 1.5 H, pyrrole N-H); UV-Vis (CH2C12) 420, 486, 520, 556, 600, 658 nm. The UV-Vis and ^ - N M R spectra matched the reported values reported by DiMagno etal.u and the product was used without further purification in subsequent reactions. 2.2.4.2 [5,15-Dibromo-10,20-diphenylporphyrinato]zinc(II) ( D B r D P h P Z n ) From ZnCI 2 References on page 79 41 DBrDPhP (0.33 g, 0.53 mmol) and ZnCfe were added to freshly distilled DMF (15 mL) and the solution refluxed for 1 h, at which point UV-Vis spectroscopy showed that reaction was complete. The mixture was poured into 100 mL of distilled water and the porphyrin filtered off. The product mixture was chromatographed on silica/(hexanes:THF 2:1), this removing a more polar bright orange band (UV-Vis (hexanes:THF 1:1) 426, 450, 502, 540) from the desired porphyrin. The product was dried in vacuo overnight at 100 °C to yield 280 mg (77%) of DBrDPhPZn (see below). If the silica column was first eluted with CH2CI2 and then with hexanes:THF (2:1), there were two porphyrin products, which were followed by the bright orange band described above. The !H-NMR data and, with the exception of the missing absorption at 541 nm, the UV-Vis data of the first two products agreed with those reported in the literature.14 The difference in Rf values could be attributed to the presence of different axial ligands on the Zn (Section 2.3.4). From Zn(OAc)2 DBrDPhP (1.37 g, 2.2 mmol) was dissolved in CHCI3 (600 mL) and the solution was refluxed. To this was added Zn(OAc)2 (0.51 g, 2.3 mmol) as a solution in MeOH (125 mL). When the reaction was complete (UV-Vis spectroscopy), the solvent was removed and the product recrystallized from THF/heptane to yield 1.31 g (71.5 %) of lustrous purple crystals which were determined by X-ray crystallographic analysis to be the bis-THF adduct DBrDPhPZn«(THF)2 (Section 2.3.4). The opacity of the porphyrin solution complicated the recrystallization procedure. Washing the crystals with hexanes or subjecting them to vacuum at 78 °C removed the axial THF ligands, as shown by !H-NMR (loss of THF signals) and elemental analysis. Smaller reaction scales, up to 0.5 g of starting porphyrin, were worked up by pre-drying the product on silica and subsequent chromatography on silica/(hexanes:THF 2:1). Characterization data for DBrDPhPZn without THF ligands: iH-NMR (200 MHz, CDCI3) 5 (9.65 (d, 4H), 8.80 (d, 4H), p-pyrrole H), 8.15 (m, 4H, 2,6-phenyl H), 7.85 (m, 6H, 3,4,5 phenyl H)\ UV-Vis (THF) 426, 524, 564, 604 nm; Analysis calc'd for C3 2 Hi 8 Br 2 N 4 Zn: C, 56.21; H, 2.65; N, 8.19; found: C, 56.35; H, 2.72; References on page 79 42 N, 7.88. The !H-NMR data and, with the exception of the missing absorption at 541 nm, the UV-Vis data match well those reported in the literature.14 2.2.4.3 Tris(n-butyl)(l-imidazolyl)tin(IV) (10) According to a literature procedure,39 Bu3Sn(OMe) (0.85 mL, 3.0 mmol) was added to a Schlenk tube containing imidazole (0.20 g, 3.0 mmol) and the mixture was heated to 50-60 °C for 1 h. After a few minutes, the imidazole dissolved; the Schlenk was put under vacuum for 0.5 h with continued heating at -50 °C and a white material sublimed on the upper end of the tube. The product was taken up in warm, dry MeOH (2 mL) and the first crop of crystals grew when the solution cooled to room temperature. A second crop was isolated from dry MeOH (3-4 mL) on cooling to -40 °C for a few hours to give a total of 0.493 g (43 %) of 10. The 4- and 5-imidazole positions are not distinguished by 1 H - and 1 3C-NMR, perhaps because the tin is 5-coordinate in a polymeric structure.40 m.p. 63-65°C (lit.39 62-65 °C); iH-NMR (300 MHz, CDCI3) 8 7.37 (s, 1H, 2-imidazolyl-H), 6.93 (s, 2H, 4,5-imidazolyl H), (1.5 (m), 1.3 (m), 18H, overlapping signals of Sn-CH2-CH2-CH2-), 0.84 (t, 9H, -CH3); 1 3 C-NMR (50 MHz, CDCI3) 8 140.2, 124.2, 28.5, 27.2, 17.9, 13.6; Analysis calc'd for Ci5H 3oN 2Sn: C, 50.45; H, 8.47; N, 7.84; found: C, 50.64; H, 8.50; N, 7.96. The lH- and 1 3 C-NMR data were not reported previously. 2.2.4.4 Tris(n-butyl)phenyltin(IV) (11) PhBr (0.315 mL, 3 mmol) was added to dry THF (15 mL), and the solution cooled to -78 °C before n-BuLi (1.5 mL of 1.6 mM in hexanes, 2.5 mmol) was added. After a few minutes, the reaction was checked by tic (silica/hexanes) and a new product with an Rf lower than that for bromobenzene was evident. Then, Bu3SnCl (0.68 mL, 2.5 mmol) was added and the solution refluxed for 1 h to convert the in situ PhLi to 11, which appeared as a more polar product by tic. iH-NMR analysis of an aliquot of the mixture indicated three sets of multiplets in the aromatic region (8 7.3, 7.0 and 6.6) along with a multiplet at 8 3.5 and a series of resonances in the alkyl region (8 0.6-1.6). The solution was used without further purification or analysis. References on page 79 43 2.2.4.5 Diethyl 2-(tris(n-butyl)stannyl)-l,3-dipropanoate (12) Compound 12 was produced according to the general procedure for synthesis of Sn-enolates.41 A solution of diethylmalonate was treated with one equiv. of n-BuLi in dry THF at -78 °C, and Bu3SnCl (1 equiv.) was subsequently added; the solution was allowed to warm to room temperature. No reaction was indicated by tic on silica/(hexanes:Et20) after visualization with a solution of I2 in hexanes. However, as tin enolates are moisture-sensitive,41 diethylmalonate may have been regenerated when the sample was spotted on the tic and exposed to the atmosphere. This solution was used without further purification or analysis. 2.2.4.6 Tris(n-butyl)(4-pyridyl)tin(IV) (13) In situ synthesis of 13 was performed via L i halide exchange using a modified literature procedure.42 4-Bromopyridine»HCl was dissolved in dilute aq. NaOH, the mixture extracted with CH2CI2, dried over Na2C03 and the solvent removed on a rotary evaporator. The oily residue was taken up in dry THF, the solution cooled to -78 °C, and treated with 1 equiv. of n-BuLi when the solution turned deep red. The mixture was stirred for 1-2 h and 1 equiv. of Bu3SnCl was added; this discharged the red color, giving a clear yellow solution which was refluxed for 1 h. Alternatively, 4-bromopyridine»HCl salt was stirred in THF and 2 equiv. of n-BuLi were added at -78 °C before the addition of the tin reagent. The !H-NMR spectrum did not match the data reported in the literature42 (Section 2.3.5). The solution of 13 was used without further purification. 2.2.4.7 Tris(n-butyl)vinyltin(IV) (14) According to a literature procedure,43 vinyl magnesium bromide (37 mL, 1.0 M in THF, 37 mmol) was added to a solution of Bu3SnCl (5 mL, 18.4 mmol) in dry THF (5 mL). The solution was refluxed for 20 h, cooled to room temperature, and hydrolyzed with saturated aq. NH4CI. The solution and solids obtained were extracted several times with Et20; the organic fractions were dried over neutral AI2O3, and the product was distilled under reduced pressure to yield 2 mL (37 %) of 14. 1H-NMR (200 MHz, CDC13) 8 (6.42 (dd, 1H), 6.09 (dd, 1H), 5.60 (dd, 1H), vinyl-//), (1.4 (m), 1.3 (m), 0.83 (t), 27H, butyl-//); 1 3 C-NMR (50 MHz, CDCI3) 8 References on page 79 44 139.21, 133.61, 29.08, 27.27, 13.66, 9.29. Coupling to spin-active Sn nuclei was observed for the signals at 8 27.27 and 9.29. The ! H - and 1 3 C-NMR spectra were identical to those of a commercial sample of 14 from Aldrich. The product was used without further purification in the cross-coupling experiments. Alternatively, 14 was prepared in situ from vinyl magnesium bromide and Bu3SnCl, with the product having NMR spectra identical to those above. 2.2.5 Cross-Coupling Experiments In general, conditions described by DiMagno et a l . 1 4 were used. DBrDPhPZn and a catalytic amount of Pd(dppf)Cl2 or Pd(PPh3)4 were dissolved in dry, degassed THF in oven-dried glassware under dry, oxygen-free N 2 , treated with a solution of the appropriate organotin reagent, and the mixture refluxed. The reaction was monitored by tic and UV-Vis spectroscopy. Workups were performed by pouring the reaction mixture into water, extracting the product with CH2CI2, removing the solvent, washing the residue with hexanes, and purifying the products by chromatography. 2.2.5.1 (Tetrakis(phenyl)porphyrinato)zinc(II) (TPhPZn) To a solution of DBrDPhPZn (5.4 mg, 8 umol) and Pd(dppf)Cl2 (1 mg) in dry THF (5 mL) was added a freshly prepared solution of 11 (Section 2.2.4.4) (2 mL, -0.5 mmol). No reaction was apparent by UV-Vis spectroscopy or tic after stirring the mixture for several hours at room temperature. The reaction mixture was then refluxed overnight under N 2 and the porphyrin was converted nearly quantitatively to TPhPZn, which was isolated by column chromatography on silica/CH2Cl2. The product had the same Rf value, UV-Vis, and iH-NMR spectra as an authentic sample of TPhPZn 4 4 Two sets of weak doublets seen in the p-pyrrole region of the iH-NMR spectrum suggest a trace porphyrin co-product with a different meso-substituent, consistent with observations of small amounts of meso-chloro porphyrin isolated from later experiments with DPhDVPZn (see Sections 2.2.5.2 and 2.3.6). References on page 79 45 2.2.5.2 (5,15-Diphenyl-10,20-divinylporphyrinato)zinc(II) (DPhDVPZn) DBrDPhPZn (1.16 g, 1.53 mmol as a mixture of porphyrins with and without THF ligands, Section 2.2.4.2) was dissolved in dry THF (800 mL). 14 (4.5 mL, 15 mmol) and Pd(PPh3)4 (0.05 g, 0.04 mmol) were added and the solution refluxed for 48 h. To monitor the reaction, a sample was demetallated with HC1 or TFA, neutralized with Et3N, and checked by tic on sihca/(CH2Ct2:hexanes 1:1). No starting material was observed, but a minor co-product, which ran slightly ahead of the desired product on tic, remained even after 48 h of refluxing. After workup, the product was chromatographed on silica/(hexanes:CH2Cl2 3:1 to neat CH2CI2). The DBrDPhZnP was isolated in -77 % (0.68 g) yield along with a small percentage of the higher Rf co-product. Recrystallization from THF/heptane helped to remove the high Rf co-product. Pure DPhDVPZn yielded the following characterization data: Rf = 0.66 (silica/(hexanes:THF 2:1)); 1 H -NMR (300 MHz, CDCI3) 8 9.51 (d, 4H, p-pyrrole H), 9.19 (dd, 2H, -Cff-CH2), 8.92 (d, 4H, 0-pyrrole H), 8.15 (dd, 4H, 2,6-phenyl H), 7.75 (m, 6H, 3,4,5-phenyl H), (6.47 (dd, 2), 6.03 (dd, 2H), -CH-C//2); UV-Vis (CH2CI2) 422, 516, 556, 594 nm; LR-MS (EI) m/e 576 (100, M+), 549 (12, M+ -vinyl), 499 (25, M+ -phenyl), 472 (10, M+ -phenyl -vinyl), 288 (18, M++); analysis calc'd for C 3 6H 24N 4Zn«l/2H20: C, 73.66; H, 4.29; N, 9.54; found: C, 73.84; H, 4.13; N, 9.40. The less polar co-product (Rf = 0.73 on silica/(hexanes:THF 2:1)) was later determined to be (5-chloro-10,20-diphenyl-15-vinylporphyrinato)zinc(U) (CIDPhVPZn) (isolated by preparative-tic on silica/(hexanes:THF 2:1); LR-MS (EI) 584 (35, M+), 550 (15, M+ -Cl), UV-Vis (CH2CI2) 426, 526, 564, 610 nm). These data are consistent with those of products of subsequent reactions (Section 2.2.6.1) in which the DPhDVPZn/ClDPhVPZn product mixture is demetallated (see also Section 2.3.6). Pd(dppf)Cl2 was also used as a catalyst in the synthesis of DPhDVPZn, but the method was not reliable. In four separate experiments using Pd(dppf)Ct2, DPhDVPZn was produced in zero to near quantitative yield. Addition of a few mg of Cul with the aim of generating a Pd° species did not have a noticeable effect, the systems still giving very variable yields. References on page 79 46 2.2.5.3 Reaction of DBrDPhPZn with 10 D B r D P h P Z n , a 10 molar excess of 10, and a catalytic amount of Pd(PPh3)4 were dissolved in T H F and the solution refluxed for 24 h. No reaction was apparent by tic, so 1 equiv. (per DBrDPhPZn) of Pd(PPh3)4 was then added and the solution refluxed for another 24 h. A new, more polar product was seen and no starting material was observed by tic on silica/(THF:hexanes 2:1) and from a demetallated aliquot (TFA) on silica/(CH2Cl2:hexanes 1:1) (Rf of new product = 0.1, that of DBrDPhP = 0.5). The U V - V i s spectrum of a demetallated sample in T H F had changed from that of D B r D P h P (420, 486, 518, 552, 600, 656 nm) to 436, 530, 570, 606, 618 nm. The reaction was worked up by adding CH2CI2, washing with H2O, and treating the organic phase with T F A until the porphyrin product was (according to tic) demetallated. The excess T F A was then neutralized with Et3N, the organic phase washed again with water, and the solvent removed. The mixture was subsequently chromatographed by preparative-tic on Al203/(CH2Cl2:hexanes:MeOH 3:2:0.1) in the dark. The bands of the main product(s) stuck to the stationary phase between elutions, and the portion isolated was composed of at least 3 products (by tic on silica). The i H - N M R spectra of the isolated fractions gave no conclusive results; there was a very complex peak pattern in the aromatic region and evidence for inner N - H protons. This reaction was not investigated further. 2.2.5.4 Reaction of DBrDPhPZn with 12 A freshly prepared solution of 12, DBrDPhPZn, and a catalytic amount of Pd(dppf)Cl2 in T H F were stirred at room temperature, but no reaction was observed. However, refluxing the solution overnight yielded a new product with a slightly higher Rf than that of the starting porphyrin (Rf = 0.6, that of D B r D P h P Z n = 0.55 on siUca/(CH2Cl2:hexanes 1:1)), and thus refluxing was continued for 84 h ( U V - V i s (CH2Cl2/hexanes): reactant porphyrin 422, 520, 554, 592 nm; reaction solution 422, 558, 604 nm). The reaction products in the solution were purified by adding distilled H2O, extracting the product with CH2CI2 and doing repeated chromatography on silica/(CH2Cl2:hexanes 1:1). i H - N M R spectroscopy indicated new meso-U signals at 8 10.25 and 10.20 and several P-pyrrole resonances. Mass spectrometry indicated a mixture of (5,15-References on page 79 47 diphenyl porphyrinato)zinc(II) (DPhPZn) (m/e= 524 (100, M + ) , 262 (45, M++)) and (5-chloro-10,20-diphenylporphyrinato]zinc(II) (CIDPhPZn) (m/e= 558, (55, M+)). See Section 2.3.6. This reaction was not pursued further. 2.2.5.5 Reaction of DBrDPhPZn with 13 Refluxing a solution of DBrDPhPZn with 13 and a catalytic amount of Pd(dppf)Cl2 in THF showed no reaction by 'H-NMR or tic. 2.2.5.6 Reaction of DBrDPhPZn with Hexamethylditin According to published procedures for the trialkylstannylation of aryl halides,45'46 DBrDPhPZn (50 mg, 0.06 mmol), Me3SnSnMe3 (0.3 mL, 1.5 mmol), and a catalytic amount of Pd(PPh3)4 were dissolved in dry degassed THF. The solution was refluxed for 48 h at which point tic indicated a new product, although not all of the starting material had been converted. The new product was isolated by preparative-tic on sihca/(hexanes:THF 2:1) and had ^ - N M R and mass spectrometry data consistent with those of a mixture of DPhPZn and CIDPhPZn seen previously (Section 2.2.5.4). No evidence was found for a mes0-SnMe3-substituted porphyrin by mass and !H-NMR spectroscopies. This reaction was not pursued further. 2.2.6 Reactivity of DPhDVPZn 2.2.6.1 Treatment with H B r An excess of dry HBr (g) was added to the evacuated headspace of a Kontes valve-sealed NMR tube containing DPhDVPZn (3 mg) in dry CDCI3 (0.5 mL). The 'solution' rapidly turned green, although not all of the porphyrin had dissolved. The ^ - N M R spectrum showed many broad peaks. Few conclusions could be drawn except that some reaction had occurred at the vinyl groups, and the p-pyrrole groups integrated for fewer than 8 H which indicates that some reaction may have occurred at the P positions. In a similar reaction, excess HBr (g) was added via syringe through a septum to DPhDVPZn (10 mg) in dry CH2CI2 (10 mL) kept at -15 °C under N 2 in a sealed Schlenk tube. The solution was stirred for 15 min and the solvent removed under reduced pressure. To assist References on page 79 48 removal of HBr, an aliquot of dry CH2CI2 (10 mL) was added via syringe, subsequently removed under reduced pressure and the process repeated. The product was dissolved again in dry CH2CI2 (10 mL), the solution filtered through an anaerobic frit under dry N2 and the residue washed with dry CH 2Cl2 (10 mL). The solvent was removed from the deep green filtrate and the product dissolved in CDCI3. The iH-NMR spectrum showed only broad resonances with fewer than 8 P-pyrrole protons by integration and no sign of the vinyl groups. The spectrum "deteriorated" with time. Subsequent chromatography of the filtrate on basic Al203(U)/(CH2Cl2 to CH2CI2/THF 20:1) gave one major band of a product mixture that was further purified on sihca/(hexanes:THF 2:1) to give 5-chloro-10,20-diphenyl-15-vinylporphyrin and 5-(2-bromoethyl)-10,20-diphenyl-15-vinylporphyrin as judged by iH-NMR, mass, and UV-Vis spectroscopies (see Section 2.3.6). 5-Chloro-10,20-diphenyl-15-vinylporphyrin (CIDPhVP): iH-NMR (200 MHz, CDCI3) S (9.80 (d), 9.45 (d), 8.90 (2 doublets), p-pyrrole H), 8.20 (dd, 2,6-phenyl H), 7.75 (m, 3,4,5-phenyl H), -2.8 (s, pyrrole N-H) (but the integrations did not match); UV-Vis (CH2CI2) 420, 518, 556, 598, 654; LR-MS (EI) m/e = 522 (50, M+), 487 (10, M+ - CI), 445 (8, M+ -phenyl); HR-MS (EI) calc'd for C34H23N4CI: 522.16113, found 522.16007. 5-(2-Bromoethyl)-10,20-diphenyl-15-vinylporphyrin (BrEtDPhVP): iH-NMR (200 MHz, CDCI3) 5 9.35 (2 doublets, 4H, p-pyrrole H), 9.10 (dd, 1H, -Ctf-CH2), (8.85 (d, 2H), 8.80 (d, 2H), p-pyrrole H), 8.10 (dd, 4H, 2,6-phenyl H), 7.70 (m, 6H, 3,4,5-phenyl H), (6.47 (dd, 1H), 6.05 (dd, 1H), -CH-C// 2), 5.45 (t, 2H, -C// 2-CH 2Br), 4.33 (t, 2H, -CH2-C/72Br), -2.75 (s, 1.5H, pyrrole N-H); UV-Vis (CH2C12) 420, 518, 554, 598, 656 nm; LR-MS (EI) m/e= 596 (40, M+), 514 (100, M + -HBr), 82 (35, H^Br); HR-MS (EI) calc'd for C36H27N4 8 1Br: 596.13983, found 596.14053. 2.2.6.2 Treatment with HBr and N-based Nucleophiles Published hydrobromination procedures for P-vinyl chlorins47 were the basis for these experiments. To a solution of DPhDVPZn (-10 mg) in glacial acetic acid (1 mL) was added 45% HBr in acetic acid (2 mL) to yield a dark green solution, indicating the presence of a porphyrin dication. The solution was stirred for 2.5 h at room temperature, the solvent removed and a solution of excess amine (iPrNH2 or E12NH) in dry CH2CI2 (5 mL) was added. The solution References on page 79 49 immediately went from green to purple, and the solution was stirred for 1 h; the solvent was removed and the product mixture chromatographed on silica/(CH2Cl2'.MeOH 50:1 to 20:1). Alternatively, dry HBr (g) was bubbled for 20 min through a solution of the porphyrin (partially dissolved) in dry CH2CI2 (10 mL) under dry N2 at 0 or at -15 °C with stirring; this yielded fewer bands by tic. After 15 min, a dark precipitate appeared on the sides of the Schlenk and did not redissolve with the aid of an ultrasonic bath. The solvent was reduced to 2 mL and the amine (~10-fold excess) was added as a solution in CH2CI2. The solution was stirred for 1 h, the solvent removed and the product mixture chromatographed as described above. Typically, some high Rf material and some low Rf material were isolated, but most of the reaction mixture remained at the baseline. The less polar material had !H-NMR data similar to those of CIDPhVP and BrEtDPhVP (Section 2.2.6.1) and was not investigated further. At least 4 low Rf products were produced and these required MeOH in the eluent to elute on silica, which is consistent with the presence of amine groups. The !H-NMR spectra of these compounds were complex. The vinyl groups were absent, but several sets of doublets in the P-pyrrole region, which integrated for ~8 protons, indicated more than one product. Attempts to further purify the low Rf material on the Chromatotron were unsuccessful. Very small amounts of products were isolated from such reaction mixtures and their identities were not pursued further. In a slightly larger scale reaction, DPhDVPZn (0.03 g, 0.05 mmol) was treated with HBr (g) as described above, and subsequently with imidazole (0.037 g, 0.54 mmol). After the mixture was stirred for 15 min at 0 °C, MeOH (0.5 mL) was added to bring everything into solution and the mixture stirred for another 15 min; the solvent was removed under reduced pressure, and the product purified by preparative-dc on silica/(CH2Cl2/MeOH 200:1 to 20:1). There were two main products and a significant amount of baseline material, which was separated into numerous products. The major band (Rf = 0.4) was isolated from the silica to give 5.5 mg of shiny purple flakes of 5-(2-bromoemyl)-15-(l-(l-imidazolyl)ethyl)-10,20-diphenylporphyrin (BrEtlmEtDPhP) (16 %), while the minor band (Rf = 0.55) (< 1 mg) was not investigated further. 1H-NMR (400 MHz, CDCI3) 8 (9.41 (d, 2H, J 4.8), 9.06 (d, 2H, J 4.4), 8.89 (d, 2H, J 4.8), References on page 79 50 8.81 (m, 2H), p-pyrrole H), 8.14 (d, 4H, J 5.9, 2,6-phenyl H), 7.98 (s, 1H, 2-imidazolyl H), 7.83-7.72 (m, 7H, 3,4,5-phenyl H and -Q/-CH3), 7.13 (s br, 2H, 4,5-imidazolyl H), (5.47 (t, 2H, J 8.1), 4.37 (t, 2H, J 8.2) -CH2-QH2), 2.82 (d, 3H, J 6.8, -CH-Ctfj), (-2.66 (s), -2.71 (s), 2H, pyrrole N-H); UV-Vis (CH2C12) 306, 418, 488, 516, 550, 594, 650 nm; LR-MS (+LSIMS) m/e 663 (10, (79Br) M + 1 ) , 595 (100, (79Br) M+1 -imidazolyl); HR-MS (+LSIMS) calc'd for C 3 9 H 3 2 N 6 7 9 B r : 663.18646, found 663.18718; calc'd for C 3 6 H 2 8 N 4 7 9 B r : 595.14973, found 595.14994. The tic of BrEtlmEtDPhP stored for several months in a tightly capped vial revealed the presence of 4 other minor products, showing that decomposition had occurred. The presence of 2 pyrrole N-H peaks in the iH-NMR spectrum also supports the presence of decomposition products. 2.2.6.3 5,15-DiphenyI-10,20-divinylporphyrin (DPhDVP) In both reactions to be described, the sample of DPhDVPZn used contained a small percentage of (5-chloro-10,20-diphenyl-15-vinylporphyrinato)zinc(II) (CDPhVPZn) (Section 2.2.5.2). DPhDVPZn (30 mg, 0.05 mmol) was partially dissolved in CH 2C1 2 (10 mL) and TFA (0.5 mL) was added to the mixture. UV-Vis spectroscopy and tic indicated that demetallation had occurred within 5 min in near quantitative yield. Et3N (1 mL) was added, the solvent removed and the product purified by column chromatography on silica/(CH2Cl2:hexanes 1:1). Analysis for a sample containing only DPhDVP: iH-NMR (200 MHz, CDC13) 5 9.45 (d, 4H, P-pyrrole H), 9.2 (dd, 2H, -C//-CH 2), 8.87 (d, 4H, P-pyrrole H), 8.25 (m, 4H, 2,6-phenyl H), 7.80 (m, 6H, 3,4,5-phenyl H), (6.50 (dd, 2H), 6.10 (dd, 2H), -CH-CH2), -2.55 (s, 2H, pyrrole N-H); " C -NMR (125 MHz, CDCI3) 8 142.34, 137.32, 134.50, 131.38 (s br), 129.31 (s br), 127.72, 127.35, 126.68, 120.13, 117.28; UV-Vis (CH2C12) 420, 522, 560, 600, 668 nm; LR-MS (EI) m/e = 514 (100, M+), 488 (M+ -vinyl), 437 (M+ -phenyl); HR-MS (EI) calc'd for C 3 6 H 2 6 N 4 : 514.21576, found 514.21558; Analysis calc'd for C 3 6 H 2 6 N 4 : C, 84.02; H, 5.09; N, 10.89; found: C, 83.67; H, 4.81; N, 10.42. DPhDVP was sufficiently pure for use in a subsequent reaction with T1(N0 3) 3»3H 20 (Section 2.2.6.3). References on page 79 51 Demetallation of DPhDVPZn (60 mg, 0.1 mmol) using a biphasic mixture of CH2CI2/3 M HC1 (aq.) according to a general procedure for Zn-porphyrins48 yielded ~5 mg (0.01 mmol, 10%) after chromatography on sihca/(CH2Cl2:hexanes 1:1). Most of the porphyrin material remained at the baseline. Residual acid was present at the time of chromatography and may have reacted with the vinyl groups in the presence of silica. A small amount of CIDPhVP was also isolated which was consistent with previous observations on demetallation using HBr(g) in CH2CI2 (Section 2.2.6.1). 2.2.6.4 Reactions with T l I n : 5,15-Bis(2,2-dimethoxyethyI)-10,20-diphenylporphyrin (BDMEtDPhP) Based on the analagous reaction for PPIX 4 9 (see Chapter 4), DPhDVP (10 mg, 0.019 mmol) was dissolved in CH2CI2 (10 mL) and MeOH (2 mL), and the solution brought to reflux. To this was added T1(N03)3«3H20 (28.5 mg, 0.064 mmol, 3.3 equiv.) dissolved in MeOH (1-2 mL), when the solution went from purple to green, and then maroon within 10 min. S0 2 (g) was bubbled through the solution for a few minutes, 10 drops of cone. HC1 were added, and the solution was stirred at room temperature for 10 min. The solution was poured into water (150 mL), and the product was extracted with CH2CI2 (2 x 50 mL) and chromatographed by preparative-tic silica/(CH2Cl2:MeOH 100:1-2). The largest product band (Rf = 0.55) (of at least 6 other smaller bands) was isolated and the solvent removed. The product was dissolved in CH2CI2, and the solution filtered; the solvent was removed, the solid was washed with hexanes and then washed from the filter with CH2CI2. Evacuation yielded 8.5 mg (68.5 %) of a shiny, purple crystalline product. 1H-NMR (400 MHz, CDC13) 5 (9.50 (d, 4H, J 4.9), 8.88 (d, 4H, J 4.9), p-pyrrole H), 8.19 (dd poorly resolved, 4H, 2,6-phenyl H), 7.81 to 7.72 (m, 6H, 3,4,5-phenyl H), 5.34 (t, 2H, J 5.3, -CH 2-C//(OCH 3) 2), 5.25 (d, 4H, J 5.3, -C// 2-CH(OCH 3) 2), 3.31 (s, 12H, -CH 2-CH(OC// 5) 2), -2.79 (2H, pyrrole N-H); UV-Vis (4.7xl0-6 M , CH2C12) 416 (5.50), 486 (3.52), 516 (4.19), 548 (3.75), 592 (3.65), 648 (3.54) nm; LR-MS (+LSIMS) m/e = 639 (25, M+1); HR-MS (+LSIMS) calc'd for C ^ ^ N ^ 639.29611, found 639.29662; Analysis calc'd for : C, 75.21; H, 6.00; N, 8.77; found: C, 75.31; H, 5.88; N, 8.52. References on page 79 52 When as little as 1 equiv. T l (OOCCF3)3 was added to a solution of DPhDVPZn in CH2Cl2/MeOH at 0°C, the solution rapidly turned brown, and no porphyrin moieties remained according to UV-Vis spectroscopy. Other investigations with diphenylporphyrin (DPhP) indicated that with the addition of 1 equiv. of the T l m the porphyrin was rapidly metallated, as indicated by UV-Vis spectroscopy. Tic and UV-Vis spectroscopic analysis of a demetallated aliquot (HO) showed only DPhP. Only when another equivalent of T l m was added and the solution refluxed overnight did other reaction products appear, the identities of which were not pursued. 2.2.6.5 Attempts at Conjugate Addition With imidazole The reported conditions for acid-50 and base-catalyzed51 addition reactions of imidazole to alkenes were the basis for the conditions used here. DPhDVPZn (a few mg) and imidazole (2 to 16 equiv.) were dissolved in dry DMF with glacial acetic acid as the catalyst, in neat glacial acetic acid, or in trimethylbenzylammonium hydroxide (40% in MeOH) in THF; no reaction was seen by tic on silica/(CH2Cl2:hexanes:MeOH 1:1:~0.02) or by UV-Vis spectroscopy, even when heat was applied. With en When DPhDVPZn (0.055 g, -0.1 mmol) was dissolved in freshly distilled en (5 mL) and the mixture heated to reflux, tic on silica/CH2Cl2 indicated all the starting material was consumed within 1 h. The resulting product mixture was very polar, and moved slightly on silica/ with (CH2Cl2:MeOH 10:1) and other strong eluents such as THF and Et3N. Elution with CH2Cl2:en (50:1) moved the product as the stationary phase became saturated with en. The UV-Vis spectrum indicated that the product was still a metalloporphyrin (CH2Cl2/MeOH 430, 570, 610 nm), but attempts to get a mass spectrum via LSIMS and DCI were unsuccessful. The iH-NMR spectrum of the crude product indicated that the vinyl groups had reacted and the p-pyrrole region had become more complex. The P-region, which integrated for 7 protons, showed a mixture of doublets, rather than 2 doublets, indicating that some reaction may have occurred at these positions. The alkyl region was also complex. References on page 79 53 Reverse phase tic of a few milligrams of the product (silicaRP-is/ (H20:MeOH:MeCN:TFA 50:50:50:1)) yielded 3 poorly separated products, but there was insufficient product on the tic plate to get a satisfactory iH-NMR spectrum. Demetallation of the sample was achieved with cone. HC1 in EtOH/CH 2Cl 2 (UV-Vis (CH 2Cl 2:MeOH 1:1) 418, 516, 552, 592, 648), but chromatography on AI2O3 achieved no separation of products. Chromatography with CH 2C1 2 and CH 2Cl 2 :MeOH (40 to 20:1) on neutral or basic Al 203, polyamide or cellulose achieved little in the way of separation and purification. Attempts to derivatize free amino groups with TsCl (2 x excess) in CHCkj/pyridine yielded no observable reaction by dc even after the solution was refluxed for 1 h. As evidenced by UV-Vis spectroscopy, attempts to coordinate Pt to the compound were unsuccessful using PtCl 2 and Pt(COD)Cl2. The porphyrin product (soluble in CH 2Cl 2/MeOH) and the Pt starting materials (soluble in NH4OH (aq) or HC1 (aq) and CH 2C1 2, respectively) were not soluble in a common solvent mixture. With other amines Refluxing several milligrams of DPhDVPZn in neat iPr 2NH (b.p. 88-90 °C) gave no observable reaction by tic. Heating DPhDVPZn in neat PhCH 2 NH 2 to >100 °C showed no appreciable reaction by tic until an excess of trimethylbenzylammonium hydroxide (40 % in MeOH) was added, after which the starting material was consumed within 30 min. Chromatographic workup of this product showed > 3 compounds, but the major product showed only a very weak Soret band and this reaction was not pursued further. With NaCN and KO lBu During investigation of other nucleophiles as conjugate addition reagents, treatment of DPhDVPZn with excess NaCN in dry DMF at room temperature overnight and subsequent heating to 50°C for a few hours gave no appreciable reaction, as judged by UV-Vis spectroscopy or tic. Treatment of the porphyrin (12 mg) with excess KO lBu in THF at -50 °C for 1-2 h resulted in at least four products (dc), in addition to the baseline material, suggesting that the porphyrin had decomposed. The fractions were isolated by column chromatography, but there was not enough product to get an iH-NMR spectrum. UV-Vis spectroscopy indicated that the fractions were References on page 79 54 metalloporphyrins. Clearly this butoxide reaction leads to numerous products, which makes it unattractive as a way of derivatizing the bis-vinylporphyrin. 2.2.6.6 Hydroboration-halogenation According to the literature procedures for hydroboration-halogenation of alkenes,52 DPhDVPZn was treated with borane (0.37 equiv. per vinyl group), and subsequently MeOH, NaOAc and IC1 (1.0 equiv. per vinyl group) were added. Quenching the reaction with aq. sodium thiosulfate to destroy remaining IC1, followed by extraction with ether, drying over MgS04 and chromatography of the isolated residue on silica/(cyclohexane/CH2Cl2 1:1) showed at least 11 products. Some of the products were analyzed by !H-NMR which indicated that the vinyl groups had undergone reaction, but the integrations of the P-pyrrole region of the metalloporphyrins integrated for fewer than 8 protons. It is possible that the unsubstituted P-pyrrole positions had reacted with borane and/or IC1 to produce new products. The large number of products might be related to differing axial ligands on the Zn. In a similar reaction with NBS (1 equiv. per vinyl group) in place of the IC1, when the mixture was purified by preparative-tic on silica/(hexane:CH2Cl2:THF), at least 7 products were isolated from the reaction. Analysis of the products by !H-NMR and UV-visible spectroscopy showed that the products were metalloporphyrins in which the vinyl groups had reacted. Also, some reaction had occurred at the P-pyrrole positions, as the lower field P-pyrrole resonances generally integrated for much less than 4 protons and were often split into patterns more complex than doublets. Some of the products were rechromatographed to reveal that they had reacted further (presumably on the silica), each into at least 6 products. In the mass spectra of some of the fractions of intermediate polarity, there appeared a large peak at 524, which matches that for DPhPZn. Further, there were three or four cluster peaks, each 12 or 14 mass units apart, higher than the 524 peak, indicating successive losses of -CH2- groups from a parent compound. Also, fragment peaks were present that match with that of HBr (m/e- 80) and loss of CH2CH2CH2CH3 (m/e= 57). In the expected product, the vinyl groups would be replaced with two 2-bromoethyl groups; these groups are stable under the conditions of mass spectrometry (EI) (Sections 2.2.6.1 References on page 79 55 and 2.2.6.2). It seems unlikely then, that the 2-bromoethyl groups, if present, were the cause of instability of the products. It is not clear why this seemingly straightforward reaction was so problematic. Smith and Cavaleiro53 reported that hydroborations of vinyl groups on protoporphyrin IX afford very low yields, but did not elaborate why. This chemistry was not pursued further. 2.2.6.7 Other Reactions With N-bromosuccinimide Addition of NBS (1.8 equiv.) to DPhDVPZn in DMSO-d6 resulted in a decrease of the intensity of the vinyl and P-pyrrole resonances in the iH-NMR spectra, suggesting that NBS reacted both with the vinyl groups and with the P-pyrrole positions, as it does with diphenylporphyrin.14 This chemistry was not pursued further. With M e O H and acid In an attempt to add a methoxy group across the double bond, DPhDVPZn (a few mg) in MeOH/CH2Cl2 (1:5, ~7 mL) was treated with acid catalysts such as HC1 or pyridinium p-toluenesulfonate (PPTS).54 With 1 drop cone. HC1 some demetallation occurred, and a low Rf product appeared as observed by tic on sihca/(CH2Cl2:hexanes:MeOH 1:1:0.02). No further reaction was observed after heating the mixture for several hours, and the products were not isolated. With several milligrams of PPTS, a small amount of low Rf material appeared when heat was applied, but no further reaction occurred after refluxing overnight and addition of more PPTS. No further attempts were made. With m-CPBA and B r 2 Treatment of DPDVPZn with m-CPB A or bromine led to a degradation of the porphyrin as evidenced by iH-NMR. This is not surprising considering the reactivity of the P-positions toward electrophilic reagents (e.g. NBS). 1 4 This chemistry was not pursued further. References on page 79 56 2.3 Results and Discussion Although several conceivable routes to diarylporphyrins exist (Scheme 2.2), that shown in Pathway A was chosen because of its versatility. By simply varying the aldehyde, one can produce a variety of 5,15-diarylporphyrins. The other pathways require more extensive chemistry at the pyrrole units, making the synthesis of each porphyrin dependent on extensive manipulation of the pyrrolic precursors. o Scheme 2.2. Routes to Diarylporphyrins. 2.3.1 Porphyrin Precursors The synthesis of diarylporphyrins via Pathway A requires bis(2-pyrrolyl)methane, 3. Described as early as 1921 by Hess and Anselm 1921,55 3 became more accessible after its synthesis via Huang-Minion-10 or NaBH4-reduction28 was reported. Reduction of related bis(2-pyrrolyl)ketone compounds is often achieved via borane-reduction31 (Section 2.2.2.3; see Scheme 2.3). References on page 79 57 VII IV 4 s O N H III VI V Scheme 2.3. Synthesis of Bis(2-pyrrolyl)methane (3), its Precursors and Reaction Side-product. Reaction Conditions: (i) 1) EtMgBr, 2) COCI2; (ii) 1) BH3, 2) steam distillation, or 1) NH2NH2, KOH/ethylene glycol; (iii) 1) CSC12, Et 20,0°C, 1 min, 2) H 2 0 ; (iv) H 2 0 2 /KOH, MeOH; (v) 1) EtOH, trace amine, excess Raney-Ni, 1 bar H2, ~25°C, several h 2) chromatography; (vi) 1) THF, excess L A H , ~25°C, 12 h, 2) Glauber's salt, or 1) EtOH, KOH, excess NaBFLi, A, 2) chromatography; (vii) CH(OEt)3, catalytic TFA, 25 °C, 30 min. Compound 2 itself can be prepared according to an old procedure directly from pyrrole, activated as its Grignard salt, and phosgene56 (Scheme 2.3, (i)), from pyrrole-2-carbonyl chloride and pyrrole, also activated as its Grignard salt,57 or more conveniently from H202-oxidation of l 2 6 (Section 2.2.2.2). In this work, 3 was produced from pyrrole along Routes (iii), (iv) and (ii) in Scheme 2.3 in -20 % overall yield. Such moderate yields might be acceptable if 3 were the final target. To facilitate synthesis of gram-quantities of DPhP, however, it was desirable to circumvent this route. Through collaborative work with researchers in Dolphin's group,25 1 was found to be directly reduced to 3 (Section 2.2.2.3 and Scheme 2.3) via Raney Ni- (-50 %), LAH-(~80 %), or NaBH4-reduction (-85 %). Raney Ni is known to hydrodesulfurize nearly every C-S containing substrate58 and cleanly hydrodesulfurized 1 to produce 3 in -50 % yield. The reduction procedure produced a few side-products, one of which was isolated and identified as the novel l,l',2,2'-tetrakis(bis-2-pyrrolyl)ethane (4) (Section 2.2.2.4 and Scheme 2.3 (v)), which attests to the radical nature of this reaction. The other unidentified side-products had mass spectrometry and iH-NMR data References on page 79 58 consistent with a-substituted pyrrolic compounds, but the materials decomposed over several days and were not investigated further. L A H reduction of 1 produces 3 in high yield, but the highly reactive, pyrophoric nature of L A H makes this route unattractive for large scale synthesis. The lastly discovered NaBfL}-reduction provides 3 in high yield under mild conditions and is amenable to scale-up, which makes this the preferred method. One such NaBH4-reduction of 1 was worked up before completion and an intermediate, identified as ketone 2, was isolated. In fact, 2 was formed even without NaBFLi. It is not clear whether NaBH4 reacts directly with 1 to produce 3, or if 3 is produced solely via 2. It is interesting that 2 is produced at all under these conditions, as Clezy and Smythe reported the synthesis of 2 via H202-oxidation of l . 2 6 Regardless of the pathway, this route is convenient and produces 3 in high yields, making it attractive for the subsequent large scale synthesis of diarylporphyrins. Min Wang and Bruce recently reported a one-step synthesis of 3 by acid-catalyzed condensation of excess pyrrole with paraformaldehyde29 (Route (vii), Scheme 2.3). Although this direct route gives a reasonable -40 % yield, the reported work-up requires chromatographic separation of the product mixtures and the reaction has been described for the synthesis of 3 only on a 200 mg scale. Routes to 3 via the Raney Ni- or NaBFLj-induced hydrodesulfurization of 1 were better options for large scale synthesis of 3 in this thesis work. Bis(2-pyrrolyI)(4-pyridyl)methane (5) A report by Nagarkatti and Ashley32 described the synthesis of the bispyrrolylmethane (5) by condensation of 4-pyridine carboxaldehyde and pyrrole (1:2) in MeOH with HCl(g), but very Utile characterization data were reported. Similar conditions in my hands led to a red-brown tarry substance which contained (by tic) only traces of 5. Using only stoichiometric amounts of pyrrole, one would expect to form polymeric materials. A large excess of pyrrole was necessary to synthesize 5 here (Section 2.2.2.5). The characterization data of 5 match well those of similar meso-aryl bis(2-pyrrolyl)methanes described by Lee and Lindsey.30 References on page 79 59 (5) Conceptually, 5 is a useful precursor to BPyP, presumably making pathways B-E (Scheme 2.2) accessible. However, pathways C-E are unlikely to yield good results given the poor results in synthesizing DPhP from formylated bis(2-pyrrolyl)phenylmethane and its derivatives.25 5 was not investigated further as a precursor to BPyP. 2.3.2 Synthesis of DPhP Treibs and Haberle reported the first synthesis of DPhP in 3 % yield.10 Manka and Lawrence later synthesized a series of substituted diphenylporphyrins, including DPhP, in unusually high yields (up to 92%)1 and, although the conditions described in this paper were the basis of the syntheses of diarylporphyrins attempted here, such high yields were never attained. On occasion, the yield of DPhP reached -50 %, but typically were 25-40 %. When DPhP was isolated by chromatography on silica/CHCl3 without first chemically reducing the TCQ, TCQ and DPhP eluted together. If TCQ were present, it would give no signal in the !H-NMR, and this could explain the unusually high yields reported in the studies of Manka and Lawrence where no elemental analyses were reported.1 Separation could be achieved with chromatography on silica/(toluene:cyclohexane 2:1), but the porphyrin streaked noticeably because of low solubility and this method was less effective than chemical reduction of the TCQ. Attempts were made to correlate variations in concentration of the reactants (benzaldehyde, 3, and TFA) with the yield, but no correlation was found. In the course of this thesis work, this reaction was carried out -30 times, on a variety of scales. No controllable factors could be found that would reliably give the higher (-50%) yields; the yields did not vary in a predictable manner. In a series of six reactions over a one-week period, a successive decline of the yield from 55 to 14 % was observed. The presence of benzoic acid in the benzaldehyde could possibly References on page 79 60 account for the decrease in yield. Thus, the amount of benzoic acid in the benzaldehyde was determined by gas chromatography: in samples kept under N2, the aldehyde contained ~1 % benzoic acid, while samples that were repeatedly opened for use had up to 4 % benzoic acid. Recrystallization of DPhP from toluene and 2-3 drops of pyridine (used in the DPhP synthesis via pathway B, Scheme 2.2),25 toluene/MeOH, pyridine/THF, DMF, DMSO, and precipitation from DMF with MeOH gave poor results, as judged by dc and elemental analysis. DPhP had poor solubility in ethyl acetate, MeOH, EtOH, cyclohexane and MeCN. Final trituration with MeOH:H20 was necessary to get a satisfactory elemental analysis. Sometimes another porphyrin product that eluted faster than DPhP on the silica was seen in DPhP syntheses; the material (< 1%) is probably TPhP, which results from some acid-catalyzed dipyrromethane or porphyrinogen scrambling.30'59"63 No scrambling products were reported for related condensation reactions in propionic acid.6 4'6 5 Another scrambling product, PhP (Section 2.2.3.2), produced in minute amounts was isolated by repeated chromatography on silica/(cyclohexane:toluene 1:1), the material running slighdy behind DPHP. 2.3.3 Other Diarylporphyrins The yields of other diarylporphyrins were much lower than those for DPhP. The best yields obtained for BNPhP, BPyP and BPyNOP were ~5, < 1, and 4.5 %, respectively (Sections 2.2.3.3 to 2.2.3.5). Attempts to quantify the amount of porphyrinogen formed during these reactions by treating aliquots of the reaction mixture with a TCQ solution (0.01 M in toluene) and analyzing by UV-Vis spectroscopy66 gave inconsistent results. A more reliable method to estimate the relative yield without chromatographic workup was to analyze 50 |iL aliquots of the reaction mixture by UV-Vis spectroscopy at the end of the TCQ oxidation step. These relative yield estimates for selected synthetic reactions are reported as the 'Soret Intensity' in Table 2.1. References on page 79 61 Table 2.1. Spectroscopically-Estimated Yields Porphyrin Solvent / [3 and Aldehyde] raM / Reaction Volume (mL) Acid Catalyst Soret Intensity8 5,15-DPhP CH 2 Cl 2 /4 /500 TFA (30 pL) 0.22b 5,15-BNPhP CH 2C1 2 / 3.5 / 100 TFA (-10 |iL) 0.04b 5,15-BPyP CH 2 Cl 2 /3 .4 /50 DCA (-10 U L ) 0.006c 5,15-BPyNOP CH 2 Cl 2 /3 .4 /50 DCA (-10 ML) 0.01c 5,15-BPyNOP CH 2 Cl 2 /3 .4 /50 TCA or TFA (~ 10 uD 0.013c 5,15-BPyNOP CHC1 3 / 3.4 / 50 BF3»MeOH (-10 M L ) 0.024C 5,15-BPyNOP CHCI3/10/50 BF3»MeOH (2.6 eq.) 0.009b a The absorbance units per uL (AU/uL) of aliquots in a 1.6 mL cuvette (1 cm path length) of the reaction mixture refluxed with TCQ for 1 h. Because the solutions turned black when refluxed with TCQ and the absorbance background was high, the Soret AU were measured from the peak to the baseline on the higher wavelength side (e.g. 5,15-BPyP (405-438 nm), 5,15-BPyNOP (410-438 nm)). bReaction mixture stirred for -20 h before adding TCQ. cReaction mixture stirred for ~48 h before adding TCQ. Long reaction times may have been detrimental to the yield of the diarylporphyrins, as the kinetically trapped porphyrinogen intermediate can undergo a ring-opening reaction with water, and this could result in the formation of a polymeric, thermodynamically favoured product.67 5,15-Bis(4-nitrophenyl)porphyrin (BNPhP) Nitrophenylporphyrins were investigated extensively by Meng et a/.,19'22'68 James et al.10 and Ravensbergen21 in the development of porphyrin-based anti-cancer agents. These porphyrins were derived from the meso-phenyl/pyridyl-substituted porphyrin series (see Chapter 3). Several functional groups (e.g. pyridyl, N-methylpyridinium, phenyl, and sulfonato-phenyl) were incorporated into the nitro-porphyrins, but this approach was limited to derivatives of phenyl and pyridyl groups. Potential for substitution at the mew-positions and further manipulation of the nitro group (e.g. reduction and subsequent quarternization or coupling of other groups) made BNPhP an attractive target. However, because of the low yield and difficulty in handling BNPhP (low solubility), it was not a useful precursor for any new compounds that could be evaluated at BCCRC. References on page 79 62 The low solubility was circumvented by addition of a few percent TFA in the solvents used in characterization. Deactivation of the precursor aldehyde by the electron-withdrawing nitro group is likely the cause of the low yield, in comparison to that for DPhP. Again, a minor porphyrin side-product was observed in the synthesis of BNPhP. The source of the side-product could be acid-catalyzed scrambling of the porphyrinogen, as seen with DPhP, but the identity of this product was not pursued. One attempt was made to reduce the nitro group: treatment of BNPhP (0.04 g, 0.07 mmol) with SnCl2 in C H C I 3 , acetic acid and cone. H C 1 according to the described reduction conditions19,68 gave only traces of porphyrin in the final product, not yielding enough for an tfl-NMR spectrum. The open, reactive mew-positions may have reacted under these conditions, resulting in a low yield. Bis(4-pyridyl)porphyrins and BPyNOP Two bis(4-pyridyl) porphyrins were synthesized here: the fj-unsubstituted BPyP (Section 2.2.3.4) and the (3-alkyl-substituted TEtTMeBPyP (Section 2.2.3.6) It was initially hoped that these porphyrins would be useful precursors for meso-modiftcation and subsequent quarternization of the pyridyl groups to give cationic, water-soluble porphyrins. However, because of the low yield from the respective pyrrolic precursors (< 1%), and the fact that multistep synthesis was required for TEtTMeBPyP, these porphyrins were not pursued further as precursors to new porphyrins of interest for evaluation at BCCRC. Traces of BPyP were detected when a solution of 3 and 4-pyridine carboxaldehyde were refluxed for 1 h in acetic acid; DCA was the only acid catalyst suitable for synthesis of BPyP via Manka and Lawrence's conditions. Several DCA-catalyzed reaction conditions were investigated, as shown in Table 2.2. References on page 79 63 Table 2.2 DCA Catalysis in the Synthesis of BPyP [3 & Aldehyde] Vol. DCA / Reaction Vol. / Reaction Time Amount of Porphyrin 3.4 mM -lO^tL/50 mL/48h trace 4mM ~25|LtL/60mL/15h trace 4 m M a -100 M.L/70 mL/15h trace0 4.3 mM l m L / 6 0 mL/ 15 h none detected0 5.8 m M a ~80|iL/500 mL/48h several mg isolated 10 mM 0.5 eq. / 50 mL / 15 h tracec 10 mM 1.5 eq./50 mL / 15 h trace 10 mM 1.6 eq./225 mL/15h trace aUnreacted bis(2-pyrrolyl)methane detected before the addition of TCQ. bBlack precipitate formed. cOxidation via Fe(II)-phthalocyanine/TCQ.33 In acid-catalyzed conditions with TCA, TsOH, or large excesses of DCA, an initially hazy solution led with further stirring to a black-purple precipitate, similar to that seen for TPyP syntheses (Chapter 3). In these reactions very little, if any, porphyrin was detected after TCQ oxidation of the products, and residual dipyrromethane could be detected by dc. It is possible that the pyridinium salt formed with strong acids is activated to nucleophilic attack on the pyridyl ring. 6 9 Much stronger acid catalysts (TCA, TFA and BF3»MeOH) were successfully used to synthesize BPyNOP using Manka and Lawrence's conditions (Section 2.2.3.5). In fact, the yield was highest with BF3, the strongest acid catalyst. Perhaps the N-oxide group prevents the pyridine carboxaldehyde from participating in precipitate-forming side reactions. Again, modest yields (-5 %) made BPyNOP unattractive as a precursor to other porphyrins. Meso-linked diarylporphyrins Although not central to the goal of developing new anti-cancer based porphyrins, the novel l,r,2,2'-tetrakis(2-pyrrolyl)ethane (4) appeared to be an ideal synthon for the synthesis of a meso-linked porphyrin dimer as outlined in Scheme 2.4. References on page 79 64 Scheme 2.4. Attempted Synthesis of a Mew-linked DPhP Dimer. Several porphyrin dimers, trimers and higher oligimers, linked through a variety of functional groups, have been reported in the literature16'18'70"76. Recent examples have included porphyrins directly linked through the mew-carbon. In one system, metalloporphyrins were joined via Ag+ oxidation of open mew-sites.5 In another, a mew-linked porphyrin dimer and porphyrin trimer were produced from a condensation reaction with mew-formyldiarylporphyrin, mew-arylbis(2-pyrrolyl)methane, and a substituted benzaldehyde.77 Also, via a MacDonald-type ('2+2') coupling, tetrakis(5-formyl-2-pyrrolyl)ethene (15) and a P-substituted bis(2-pyrrolyl)methane78 were condensed under acidic conditions with subsequent oxidation with DDQ (Scheme 2.5) to form a mew-linked bis-porphyrin. Scheme 2.5. Synthesis of a Mew-linked Porphyrin from 15. Attempts to synthesize a porphyrin dimer analogous to that described by Khoury et al.18 using 4, 3, benzaldehyde and TFA produced only DPhP (Section 2.2.3.7, Scheme 2.4). As bis(2-pyrrolyl)methanes and porphyrinogens are known to undergo acid-catalyzed References on page 79 65 scrambling,30,59"62 4 may have simply decomposed into 3 under the given conditions. Although 15 is structurally similar to 4, the presence of the alkene may make 15 less prone to acid-catalyzed scrambling. In addition, the MacDonald-type coupling employed by Khoury et al. is a more direct route to a porphyrin dimer than that attempted here and would likely give a better yield. 2.3.4 Derivitization of DPhP [5,15-dibromo-10,20-diphenylporphyrinato]zinc(II) (DBrDPhPZn) When DBrDPhPZn was purified by column chromatography on sihca/(CH2Cl2 to hexanes:THF 2:1), three products were isolated (Section 2.2.4.2). The third, red band isolated from the metallation reaction with ZnCl2 had a substantial absorbance at 540 nm, which was not present in the other porphyrin fractions. DiMagno et al. reported that DBrDPhPZn had a peak in the UV-Vis spectrum at 541 nm,14 which indicates that their product was also contaminated with this compound. Its identity was not pursued further. The fact that the first two bands from the chromatography column had identical characterization data, but different Rf values, can be attributed to the presence of different axial ligands on the Zn. A review by Buchler in 1978 on metalloporphyrins reported that uncontrollable ligand exchange often occurs during chromatography and crystallization.79 At the time of this review, Zn-porphyrins were known to exhibit only 4- or 5-coordinate geometry; for example, in the presence of a ligand such as pyridine, the Zn typically assumes square-pyramidal geometry.80 Although the large size of the Zn(II) ion was thought to preclude Zn from occupying the mean porphyrin plane, thereby inhibiting coordination of a ligand at the 6th coordination site, Schauer et al. reported the first example of a 6-coordinate Zn-porphyrin, TPhPZn»(THF)2.81 Indeed, when DBrDPhPZn was recrystallized from THF/heptane, DBrDPhPZn«(THF)2 was isolated as lustrous cubes; its structure was determined by X-ray crystallography, as shown in Figure 2.1. References on page 79 66 C14 Figure 2.1. X-ray Crystal Structure of DBrDPhPZn»(THF)2. The THF ligands in this octahedral complex are weakly coordinated as evidenced by the relatively long Zn-0 bond length [2.400(3) A]; the Ol-Zn-Nl and 01-Zn-N2 bond angles are 90.6(1)°, and 91.3(1)°, respectively, indicating that the Zn essentially occupies the mean porphyrin plane. These results are similar to those of obtained for TPhPZn*(THF)281 and other more recently reported examples of [porphyrinato]Zn«(THF)2 complexes.18'82 2.3.5 Pd-catalyzed Cross-coupling Reactions Pd-catalyzed cross-coupling of organic halides and organotin reagents is a well established synthetic technique.83"85 This technique has been employed with a wide variety of organotin reagents and was the basis for the experiments used here for introduction of new substitutents to References on page 79 67 the porphyrin mew-positions: aminotrialkyltin complexes for amination,86"88 heterocycle-trialkyltin complexes for introduction of new function groups on aromatic heterocycles,89'90 trialkyltin enolates for introduction of esters,91,92 and hexaalkyldistannanes to form new C-SnR.3 complexes.45'93 Only recently, however, has this methodology been applied to porphyrins.14'82 Bu3Sn-Br Bu3Sn-R Scheme 2.6. Proposed Mechanism for Pd-catalyzed Cross Coupling with DBrDPhPZn. As described by DiMagno et al.,u cross-coupling reactions with DBrDPhPZn proceed via an oxidative-addition of the porphyrin halide to a Pd° center, transmetallation with the organotin reagent, and reductive-elimination. The cycle regenerates the Pd° center and incorporates a new substituent into the porphyrin structure (Scheme 2.6). The results of the Pd-catalyzed cross-coupling reactions are summarized in Table 2.3. References on page 79 68 Table 2.3. Summary of Cross-coupling Reactions with DBrDPhPZn Tin Reagent Catalyst Desired Product Obtained? Comments Tin Reagent / Experimental Section ^ 3 ~ S n B u 3 Pd(dppf)Cl2 yes 11 / 2.2.5.1 SnBu3 Pd(dppf)Cl2 sometimes 14 / 2.2.5.2. SnBu3 l=\ N- Bu3Sn Pd(PPh3)4 Pd(PPh3)4 yes no a new product formed with one 14 / 2.2.5.2 10 / 2.2.5.3 Etq,cx CH-SnBu3 EtCfcC' equiv. of Pd(PPh3)4 Pd(dppf)Cl2 noa denomination of porphyrin 12 / 2.2.5.4 \y~ s n B u a Pd(dppf)Cl2 noa 13 / 2.2.5.5 Me3Sn-SnMe3 Pd(PPh3)4 no debromination of porphyrin 7 2.2.5.6 aIt is unclear whether the lack of activity was related to the Pd(dppf)Cl2 or the lack of in situ formation of the organotin precursor. Pd-catalyzed cross-coupling is obviously a less practical route to TPhPZn than metallation of TPhP. It was useful, however, to investigate synthesis of TPhPZn to evaluate the viability of cross-coupling experiments. DPhDVPZn, described previously,82 was synthesized in order to introduce new functional groups via chemistry at the vinyl groups. Attempts with 10 and 13 were made to test the viability of cross-coupling with heterocyclic aromatics, with the eventual goal of coupling nitroimidazoles to the meso-positions. The tributyltin enolate of diethylmalonate (12) was investigated for incorporating diethylmalonate into a porphyrin; the esters could then be cleaved to produce potential ligands for Pt-chelation (16). Hexamethylditin was used in an attempt to make a general porphyrin-tin reagent (17) that could be subsequently coupled with a variety of other organic halides. References on page 79 69 HO-CX HO-C' O O • I CH CH ,C-OH "C-OH O O I  (H3C)aSn 16 17 Pd-catalysts DiMagno et al.u and others94'95 report Pd(dppf)Cl2 to be a better catalyst than Pd(PPh3)4 for cross-coupling, although the effect of the catalyst is often substrate-dependent. Pd(dppf)Cl2 was found to be an inconsistent catalyst under the conditions used in this thesis work. In four separate experiments to produce DPhDVPZn with catalytic amounts of Pd(dppf)Cl2, the desired porphyrin was produced in zero to near quantitative yield. The Pd in Pd(dppf)Cl2 is in the 2+ oxidation state and must be reduced before reacting with the meso-bromo porphyrin via oxidative addition. This reduction must occur with the TPhPZn synthesis, but the nature of the reductant is not clear. However, Pd° species are generated in-situ in solutions of Pd(U) salts and R3P (R = aryl, alky I).9 6'9 7 Also, it is unclear how the use of essentially identical conditions between experiments gave such variable yields. Reactions with Pd(PPh3)4 as the catalyst to produce DPhDVPZn were reliable, consistently producing the desired porphyrin in -80 % yield. D P h D V P Z n As the reactant (DBrDPhPZn) and product (DPhDVPZn) porphyrins have nearly identical UV-Vis spectra and tic Rf values on silica/(CH2Cl2:hexanes or THF:hexanes), it was difficult to determine when the reaction was complete in the solvents tried. The reaction was thus monitored by either working up an aliquot of the reaction mixture and analyzing by !H-NMR, or treating an aliquot of the mixture with aqueous acid (HC1 or TFA) to demetallate the porphyrins. The demetallated starting and product porphyrins have sufficiently different Rf values to be References on page 79 70 distinguishable. Residual alkyltin products could be removed by washing the product with hexanes,98 in which the porphyrin was insoluble. Inclusion of the demetallation step (HC1) in the workup procedure, and subsequent neutralization with dilute NaOH (aq) until the aqueous phase was pH = 8, led to substantial loss of product upon chromatographic workup on silica/(CH2Cl2/hexanes 1:2 to 1:1) and not all of the product was demetallated. Better results were obtained when DPhDVPZn was isolated first and then TFA was used to demetallate the porphyrin. Reaction with tris(n-butyl)(l-imidazolyl)tin(IV) There is some literature precedence for amination via Pd-catalyzed cross-coupling.86 8 8 In this thesis work, it is clear that under the reaction conditions using stoichiometric amounts of Pd(PPh3)4 some reaction occurred, but product(s) could not be characterized. One possibility is that the product was the porphyrin-bromo Pd complex, which could be formed under these conditions via the oxidative addition of DBrDPhZn to Pd(PPh3)4 (see Scheme 2.6). Luitjen and van der Kerk postulated that 10 assumes a polymeric structure39 (Figure 2.2). Such a polymeric configuration could make the imidazole unavailable for reaction under cross coupling conditions. If an (l-imidazolyl)porphyrin did form, the preparative procedure was not useful, considering the sensitivity of the resulting product. 2.3.6 Debrominated and Meso-C\ Porphyrins Debromination (hydrogenolysis) of DBrDPhPZn to yield DPhPZn and CIDPhPZn (discussed below) was observed for cross-coupling reactions of DBrDPhPZn with 12 or hexamethylditin. DiMagno et al. reported such denominations when a (3-Br porphyrin reacted with secondary or tertiary alkyl-Zn reagents under cross-coupling conditions with a catalytic Bu Bu Figure 2.2. Proposed Structure for 10. References on page 79 71 amount of Pd(dppf)Cl2-14 They suggested that the reaction occurred via P-hydride elimination once the alkyl group had been transferred to the Pd center (before reductive-elimination in Scheme 2.6). No (3-hydrogens are present in 12 or hexamethylditin (although a butyl group from the Sn center in 12 could be transferred), so this debromination likely occurred via a different mechanism. Meso-QX porphyrin products were observed in several cross-coupling experiments described here (see Sections 2.2.5.1, 2.2.5.2, 2.2.5.4, and 2.2.5.6). The occurrence of such products is perplexing, considering that the cross-coupling with tris(n-butyl)vinyltin(IV) (14) and hexamethylditin employed no CI sources. Considering the frequency of occurrence of these products, one likely source is the NBS. If the NBS used to synthesize DBrDPhP contained a small percentage of N-chlorosuccinimide, a fraction of the mew-sites would be chlorinated (the same sample of NBS was used for all brominations). Although alkyl-chlorides are reactive under similar Pd-catalyzed cross-coupling conditions, the use of aryl-bromides or -iodides is more typical.85 A meso-Cl porphyrin may not be reactive under the conditions employed here. 2.3.7 Reactions with DPhDVPZn It was initially hoped that the vinyl groups of DPhDVPZn would be easily derivatized to incorporate moieties such as nitroimidazoles or amines into the porphyrin. Olefins can be derivatized via a variety of conditions,99 but those investigated here were based on electrophilic-and nucleophilic-addition and oxidation. Of the conditions to which DPhDVPZn was exposed, only demetallation (Section 2.2.6.3) and subsequent TI111 oxidation of the vinyl groups (Section 2.2.6.4) proceeded smoothly. The results of the other experiments are summarized in Figure 2.3. References on page 79 72 rapid demetallation; significant loss of product (HC1) numerous products, difficult to purify; BrEtlmEtDPhP (-15 %) (from imidazole) decomposition 'anti-Markovnikov' addition (BrEtDPhVP) en, KO'Bu^Pr jNH or P h C H 2 N H 2 products(s) difficult to • characterize; no reaction with heat ' P r 2 N H or P h C H 2 N H 2 bromination at P-pyrrole positions, vinyl groups reacted No Reaction at least 5 products initially, each of which was unstable after isolation Figure 2.3. Summary of Reactions with DPhDVPZn. Treatment with H B r and HBr/nucleophiie When DPhDVPZn was treated with HBr, the 'anti-Markovnikov' vinyl addition product, BrEtDPhVP, was isolated (Section 2.2.6.1). The iH-NMR spectrum of BrEtDPhVP is shown in Figure 2.4. Markovnikov addition of HBr to a vinyl group is expected to yield the product with the Br bound to the most substituted carbon, the site of the most stable carbocation intermediate/transition state.100 One explanation for the occurrence of BrEtDPhVP is that pathway B is favored over pathway A (see Figure 2.6), as the intermediate carbocation/transition state in pathway A is destabilized by the adjacent porphyrin dication. References on page 79 73 r l Figure 2.4. The 200 MHz iH-NMR-spectrum of BrEtDPhVP in CDCI3; For comparison, the !H-NMR spectrum of DPhDVP is shown in Figure 2.5. References on page 79 74 Figure 2.5. The 200 MHz iH-NMR-spectrum of DPhDVP in CDCI3. References on page 79 75 Figure 2.6. Proposed Mechanism for the Reaction of HBr with DPhDVPZn. In the case of protoporphyrin IX, the vinyl groups are on the p-pyrrole positions and the addition products are the expected and highly reactive 1-bromoethyl groups.47 Mew-positions are more directly influenced by ring-current effects than the P-pyrrole positions, as judged from the iH-NMR chemical shifts of porphyrins with protons in the mew-positions. Therefore, vinyl groups in the mew-positions, as in DPhDVPZn, would be influenced more by the adjacent porphyrin dication than the P-vinyl groups in protoporphyrin IX, and would be more likely to react via Pathway B. When imidazole was added to the HBr/DPhDVPZn reaction mixture, substitution occurred via both pathways in the same porphyrin molecule as evidenced by the production of 5-(2-bromoethyl)-15-(l-(l-imidazolyl)ethyl)-10,20-diphenylporphyrin (BrEtlmEtDPhP) (see Section 2.2.6.2). Either Pathway B is not clearly favored over Pathway A or the observed 'anti-Markovnikov' addition is occurring via a radical process. References on page 79 BrEtlmEtDPhP Addition of i P r N H 2 , E t 2 N H or imidazole (discussed above) to the DPhDVPZn/HBr product yielded a mixture of products which were difficult to isolate and characterize. When the vinyl groups react with HBr via pathway A, the 1-ethyl position becomes a chiral center; thus enantiomers and diastereomers could account for the multitude of products. Also, the presence of atropisomers cannot be ruled out; these could further complicate the purification of the mixture. This chemistry to introduce new substituents on DPhDVPZn is less than ideal for production of suitable quantities of porphyrin products for evaluation at BCCRC and was not pursued further. Demetallation and reactions with T I 1 1 1 DPhDVPZn was cleanly demetallated using TFA, but demetallation with HC1 resulted in the loss of substantial amounts of the demetallated product during chromatographic purification. Presumably, the vinyl groups react with the silica in the presence of residual acid. Treatment of DPhDVPZn with as little as 1 equiv. of T l n i led to rapid decomposition of the porphyrin (Section 2.2.6.4). Barnett and co-workers reported that when OEPZn was treated with T l n i , oxyporphyrins were produced via the 7i-cation radical. 1 0 1 , 1 0 2 Evans et al. found that when TPhPZn was exposed to similar conditions the resulting product mixture contained 5,15-dihydroxy-5,10,15,20-tetraphenylporphodimethenes and some ring-opened products.103 It is likely that DPhDVPZn reacts with T l m to give similar products. However, once metallated with TI (via reaction of the free base porphyrin and Tl i n ) the porphyrin ring is protected from oxidation because of increased oxidation potentials of the H-porphyrin complex,1 0 1 , 1 0 2 but reaction can occur at substituents on the porphyrin periphery. This protective effect is clearly demonstrated in References on page 79 77 the efficient oxidation of the vinyl groups to dimethylacetals in the synthesis of BDMEtDPhP (Section 2.2.6.4). In principle, the chemistry described for the corresponding protoporphyrin IX oxidation product49'104 (see also Chapter 4) can also be applied to BDMEtDPhP. CH(OCH3)2 BDMEtDPhP Attempts at conjugate addition Vinyl groups activated by electron-withdrawing groups can undergo nucleophilic attack in a reaction known as conjugate addition,99 which was the basis for the experiments described in Section 2.2.6.5. For example, when protoporphyrin IX dimethylester is refluxed in ethylenediamine, a new, water-soluble product results, occurring via conjugate addition of en to the vinyl group1 0 5*1 0 6 (see Figure 2.7 and Chapter 4). Figure 2.7. Conjugate Addition of en to a Vinyl Group on Protoporphyrin IX. Ethylenediamine reacted rapidly with DPhDVPZn to yield a product that was very difficult to characterize. From the iH-NMR spectra, it was clear that the vinyl groups had reacted, but the identity of this product was never conclusively determined. It is perplexing that no significant reaction occurred with ip^NH and PhCH2NH2 (some reaction was observed once base was added References on page 79 78 to the PhCH2NH.2 reaction) as both have pK a values similar to that of en and both are good nucleophiles. Other nucleophiles (NaCN, KOlBu) did not cleanly yield any products. Imidazole has been reported to react in conjugate-addition fashion with a double bond in an acetic acid-catalyzed reaction with 2-vinyl pyridine50 and in a base-catalyzed reaction with a cyano-activated vinyl group51; no reaction was observed here under similar conditions. In the reaction of 2-vinyl pyridine described in the literature, some of the pyridine is protonated, thereby creating an electron withdrawing group which activates the vinyl group for conjugate addition. Such an activation site, however, does not exist in DPhDVPZn. In a reaction involving a cyano-activated vinyl group, the added base deprotonates the imidazole, which becomes a better nucleophile for the conjugate addition." It seems that the porphyrin does not sufficiently activate the vinyl groups for conjugate addition. Conjugate addition as a means to modify the vinyl groups was not pursued further. Hydroboration-halogenation and other reactions Treatment of DPhDVPZn with BH3»THF and subsequently IC1 or NBS led to a complex mixture of products which were difficult to isolate and characterize. Once isolated they continued to react to form other products (Section 2.2.6.6). Some reaction occurred at the vinyl groups and P-pyrrole positions when DPhDVPZn was treated with NBS. Mild conditions were employed in an attempt to add MeOH across the vinyl group using HC1 or PPTS catalysts, but no appreciable reaction was observed. Because of complications with these reactions, they were not investigated further for DPhDVPZn modification. 2.4 Summary Novel and efficient procedures for the production of bis(2-pyrrolyl)methane (3) and other porphyrin precursors were established. Synthesis of diarylporphyrins via Manka and Lawrence's conditions occurred in low yield, with the exception of DPhP. DPhP was produced in sizable quantities and was brominated, metallated, and subjected to Pd-catalyzed cross-coupling conditions. Only the cross-coupling reactions with Bu3PhSn (11) and Bu3(CH=CH2)Sn (14) References on page 79 79 were unequivocal successes. The origin of the meso-Cl and debrominated porphyrins from the cross-coupling reactions were discussed. DPhDVPZn was subjected to a variety of reaction conditions, with demetallation and the subsequent Tl 1 1 1 oxidation reaction of the vinyl groups giving good yields of the expected products. References for Chapter 2 (1) Manka, J. S.; Lawrence, D. S. Tetrahedron Lett. 1989, 30, 7341-7344. (2) Wang, Q. M . ; Bruce, D. W. Angew. Chem., Int. Ed. Engl. 1997, 36, 150-152. (3) Bruce, D. W.; Wali, M . A.; Wang, Q. M . J. Chem. Soc., Chem. Commun. 1994, 2089-2090. (4) Bruce, D. W.; Dunmur, D. A.; Santa, L. S.; Wali, M . A. J. Mater. Chem. 1992, 2, 363-366. (5) Osuka, A.; Shimidzu, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 135-137. (6) Jaquinod, L.; Kyritsakas, N.; Fischer, J.; Weiss, R. New J. Chem. 1995,19, 453-460. (7) Diebold, A.; Fischer, J.; Weiss, R. New J. Chem. 1996, 20, 959-970. (8) Dandliker, P. J.; Diederich, F.; Gisselbrecht, J.-P.; Louati, A.; Gross, M . Angew. Chem., Int. Ed. Engl. 1995, 34, 2725-2728. (9) Dandliker, P. J.; Diederich, F.; Gross, M. ; Knobler, C. B.; Louati, A.; Sanford, E. M . Angew. Chem., Int. Ed. Engl. 1994,53, 1739-1742. (10) Treibs, A.; Haberle, N. Liebigs Ann. Chem. 1968, 718, 183-207. (11) Boyle, R. W.; Dolphin, D. J. Chem. Soc., Chem. Commun. 1994, 2463-2464. (12) Boyle, R. W.; Johnson, C. K.; Dolphin, D. J. Chem. Soc, Chem. Commun. 1995, 527-528. (13) Arnold, D. P.; Bott, R. C ; Eldridge, H.; Elms, F. M. ; Smith, G.; Zojaji, M . Aust. J. Chem. 1997, 50, 495-503. (14) DiMagno, S. G.; Lin, V. S.-Y.; Therien, M . J. J. Org. Chem. 1993, 58, 5983-5993. 80 (15) Lin, V. S.-Y.; DiMagno, S. G.; Therien, M. J. Science 1994, 264, 1105-1111. (16) Lin, V. S.-Y.; Therien, M . J. Chem. Eur. J. 1995,1, 645-651. (17) LeCours, S. M . ; Guan, H.-W.; DiMagno, S. G.; Wang, C. H.; Therien, M . J. /. Am. Chem. Soc. 1996,118, 1497-1503. (18) LeCours, S. M. ; DiMagno, S. G.; Therien, M . J. J. Am. Chem. Soc. 1996,118, 11854-11864. (19) Meng, G. G.; James, B. R.; Skov, K. A.; Korbelik, M . Can. J. Chem. 1994, 72, 2447-2457. (20) James, B. R.; Meng, G. G.; Posakony, J. J.; Ravensbergen, J. A.; Ware, C. J.; Skov, K. A. Metal-Based Drugs 1996, 3, 85-89. (21) Ravensbergen, J. R. M.Sc. Thesis, Univ. of British Columbia, 1993. (22) Meng, G. G. Ph.D. Thesis, Univ. of British Columbia, 1992. (23) Ware, C. M.Sc. Thesis, Univ. of British Columbia, 1994. (24) Harris, D. C. Quantitative Chemical Analysis; W.H. Freeman: New York, 1987; p 644. (25) Bruckner, C ; Posakony, J. J.; Boyle, R. W.; Johnson, C. W.; Dolphin, D.; James, B. R. /. Porph. Phthal. 1997, In press, (26) Clezy, P. S.; Smythe, G. A. Aust. J. Chem. 1969, 22, 239-249. (27) Bruckner, C. Ph.D. Thesis, Univ. of British Columbia, 1996. (28) Chong, R.; Clezy, P. S.; Liepa, A. J.; Nichol, A. W. Aust. J. Chem. 1969, 22, 229-238. (29) Min Wang, Q.; Bruce, D. W. Synlett 1995, 1267-1268. (30) Lee, C.-H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427-11440. (31) Ballantine, J. A.; Jackson, A. H.; Kenner, G. W.; McGillivray, G. Tetrahedron Supplement 7 1966, 22, 241-259. (32) Nagarkatti, J. P.; Ashley, K. R. Synthesis 1974, 186-187. (33) Lindsey, J. S.; MacCrum, K. A.; Tyhonas, J. S.; Chuang, Y.-Y. J. Org. Chem. 1994, 59, 579-587. 81 (34) Paine III, J. B. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. 1; pp 101-234. (35) Johnson, A. W.; Markham, E.; Price, R.; Shaw, K. B. J. Chem. Soc. 1958, 4254-4257. (36) Johnson, A. W.; Kay, I. T.; Markham, E.; Price, R.; Shaw, K.B.J. Chem. Soc. 1959, 3416-3424. (37) Young, R.; Chang, C. K. J. Am. Chem. Soc. 1985,107, 898-909. (38) Anderson, H. L.; Hunter, C. A.; Mealt, M . N.; Sanders, J. K. M. J. Am. Chem. Soc. 1990,112, 5780-5789. (39) Luijten, J. G. A.; van der Kerk, G. J. M. Rec. Trav. Chim. 1963, 81, 1181-1188. (40) Janssen, M . J.; Luijten, J. G. A.; Van der Kerk, G. J. M . J. Organomet. Chem. 1964,1, 286-291. (41) Pereyre, M. ; Quintard, J.-P.; Rahm, A. Tin in Organic Synthesis; Butterworth: London, 1987, pp 11-12. (42) Mitchell, T. N. Org. Mag. Reson.1915, 7, 610-614. (43) Seyferth, D.; Stone, F. G. A. J. Chem. Soc. 1957, 515-517. (44) Barnett, G. H.; Hudson, M. F.; Smith, K. M . J. Chem. Soc. Perkin 11975, 1401-1403. (45) Azizian, H.; Eaborn, C ; Pidcock, A. J. Organomet. Chem. 1981, 215, 49-58. (46) Kosugi, M. ; Shimizu, K.; Ohtanin, A.; Migita, T. Chem. Lett. 1981, 829-830. (47) Pandey, R. K.; Bellnier, D. A.; Smith, K. M. ; Dougherty, T. J. Photochem. Photobiol 1991, 53, 65-72. (48) Buchler, J. W. In Porphyrins and Metalloporphyrins; K. M. Smith, Ed.; Elsevier: Amsterdam, 1975; pp 195-198. (49) Kenner, G. W.; McCombie, S. W.; Smith, K. M . Liebigs Ann. Chem. 1973, 1329-1338. (50) Profft, E.; Georgi, W. Liebigs Ann. Chem. 1961, 643, 136-144. (51) Efros, A. M . Zh. Org. Khim. 1966, 2, 178-179. (52) Kabalka, G. W.; Gooch III, E. E. J. Org. Chem. 1980,45, 3578-3580. (53) Smith, K. M. ; Cavaleiro, J. A. S. Heterocycles 1987, 26, 1947-1963. 82 (54) Miyashta, M ; Yoshikoshi, A.; Grieco, P. A. J. Org. Chem. 1977, 42, 3772-377'4. (55) Hess, K.; Anselm, F. Berichte 1921, 54, 2310-2326. (56) Fischer, H.; Orth, H. Die Chemie des Pyrrols; Akademische Verlagsgesellschaft: Leipzig, 1937; Vol. 1, p 335. (57) Rapoport, H.; Willson, C. D. J. Chem. Soc. 1962, 84, 630-635. (58) Caubere, P.; Coutrot, P. In Comprehensive Organic Synthesis; B. M. Trost and I. Fleming, Eds.; Pergamon Press: Oxford, 1991; Vol. 8; pp 835-847. (59) Lindsey, J. S.; Schreimann, I. C ; Hsu, H. C ; Kearney, P. C ; Marguerettaz, A. M . J. Org. Chem. 1987, 52, 827-836. (60) Wallace, D. M . ; Leung, S. H.; Senge, M . O.; Smith, K. M . J. Org. Chem. 1993, 58, 7245-7257. (61) Smith, K. M. ; Eivazi, F. J. Org. Chem. 1979, 44, 2491-2592. (62) Jackson, A. H.; Kenner, G. W.; Sach, G. S. J. Chem. Soc. (C) 1967, 2045-2059. (63) Clezy, P. S. Aust. J. Chem. 1991, 44, 1163-1193. (64) Mizutani, T.; Ema, T.; Tomita, T.; Kuroda, Y.; Ogoshi, H. J. Am. Chem. Soc. 1994, 116, 4240. (65) Ema, T.; Kuroda, Y.; Ogoshi, H. Tetrahedron Lett. 1991, 32, 4529-4532. (66) Lindsey, J. S.; Wagner, J. W. J. Org. Chem. 1989, 54, 828-836. (67) DiMagno, S. G.; Williams, R. A.; Therien, M . J. J. Org. Chem. 1994, 59, 6943-6948. (68) Meng, G. G.; James, B. R.; Skov, K. A. Can. J. Chem. 1994, 72, 1894-1909. (69) Joule, J. A.; Mills, K.; Smith, G. F. Heterocyclic Chemistry; Chapman & Hall: London, 1995, pp 95-100. (70) Lee, D. A.; Smith, K. M. /. Chem. Soc. Perkin Trans. 1 1997, 1215-1218. (71) Pandey, R. K.; Forsyth, T. P.; Gerzevske, K. R.; Lin, J. J.; Smith, K. M . Tetrahedron Lett. 1992,33, 5315-5318. (72) Dolphin, D.; Paine HI, J. B. Heterocycles 1981,16, 417-447. (73) Arnold, D. P.; Nitschinsk, L. J. Tetrahedron Lett. 1993, 34, 693-696. 83 (74) Arnold, D. P.; Heath, G. A. J. Am. Chem. Soc. 1993,115, 12197-12198. (75) Prathapan, S.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc. 1993, 115, 7519-7520. (76) Zhou, X.; Chan, K. S. J. Chem. Soc, Chem. Commun. 1994, 2493-2494. (77) Susumu, K.; Shimidzu, T.; Tanaka, K.; Segawa, H. Tetrahedron Lett. 1996, 37, 8399-8402. (78) Khoury, R. G.; Jaquinod, L.; Smith, K. M . Chem. Commun. 1997, 1057-1058. (79) Buehler, J. W. In The Porphryins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. 1; pp 389-483. (80) Goll, J. G.; Moore, K. T.; Ghosh, A.; Therien, M . J. J. Am. Chem. Soc. 1996,118, 8344-8354. (81) Schauer, C. K.; Anderson, O. P.; Eaton, S. S.; Eaton, G. R. Inorg. Chem. 1985, 24, 4082-4086. (82) DiMagno, S. G.; Lin, V. S.-Y.; Therien, M . J. J. Am. Chem. Soc. 1993,115, 2513-2515. (83) Heck, R. F. Palladium Reagents in Organic Syntheses; Academic Press: London, 1985. (84) Beletskaya, I. P. J. Organomet. Chem. 1983, 250, 551-564. (85) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508-524. (86) Kosugi, M. ; Kameyama, M.; Migita, T. Chem. Lett. 1983, 927-928. (87) Beller, M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1316-1317. (88) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem., Int. Ed. Engl. 1995, 34, 1348-1350. (89) Bailey, T. R. Tetrahedron Lett. 1986, 27, 440-4410. (90) Fujita, M. ; Oka, H.; Ogura, K. Tetrahedron Lett. 1995, 36, 5247-5250. (91) Kosugi, M. ; Suzuki, M. ; Hagiwara, I.; Katsuhisa, G.; Saitoh, k.; Migita, T. Chem. Lett. 1982, 939-940. (92) Kosugi, M. ; Hagiwara, I.; Sumiya, T.; Migita, T. Bull. Chem. Soc. Jpn. 1984, 57, 242-246. 84 (93) Kosugi, M. ; Ohya, T.; Migita, T. Bull. Chem. Soc. Jpn. 1983, 56, 3855-3856. (94) Hayashi, T.; Knishi, M. ; Kobori, Y.; Makoto, K.; Taiichi, H.; Hirotsu, K. J. Am. Chem. Soc. 1984,106, 158-163. (95) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987, pp 710-720. (96) Mandai, T.; Matsumoto, T.; Tsuji, J. Tetrahedron Lett. 1993, 34, 2513-2516. (97) Amatore, C ; Jutand, A.; M'Barki, M . A. Organometallics 1992,11, 3009-3013. (98) Clive, D. L. J.; Yang, W. J. Org. Chem. 1995, 60, 2607-2609. (99) Soil, H. In Alkene, Cycloalkene, Arylalkene; 4th ed.; E. Muller, Ed.; Georg Thieme Verlag: Stuttgart, 1972; Vol. V; pp 946-971. (100) Carey, F. A.; Sundberg, R. S. Advanced Organic Chemistry; Plenum Press: New York, 1993; Vol. A, pp 342-348. (101) Barnett, G. H.; Hudson, M . F.; McCombie, S. W.; Smith, K. M . J. Chem. Soc. Perkin I 1973, 691-696. (102) Barnett, G. H.; Evans, B.; Smith, K. M . Tetrahedron 1975, 31, 2711-2717. (103) Evans, B.; Smith, K. M. ; Cavaleiro, J. A. S. J. Chem. Soc. Perkin 11978, 768-773. (104) Brantley, S. E.; Gerlach, B.; Olmstead, M. M. ; Smith, K. M. Tetrahedron Lett. 1997, 38, 937-940. (105) Brunner, H.; Obermeier, H.; Szeimies, R.-M. Chem. Ber. 1995,128, 173-181. (106) Stein, T. P.; Plane, R. A. J. Am. Chem. Soc. 1969, 91, 607-610. 85 Chapter 3 The Synthesis and Modification of Tetraarylporphyrins 3.1 Introduction In contrast to the diarylporphyrins whose syntheses require an independent synthesis of a variety of bispyrrolic units (See Chapter 2), tetraarylporphyrins can conveniently be synthesized via condensation of pyrrole with an aromatic aldehyde. Refluxing a solution of pyrrole and an aromatic aldehyde in propionic acid (Adler method), or forming a porphyrinogen under acid-catalyzed equilibrium conditions with subsequent oxidation (Lindsey method), routinely gives symmetrically-substituted tetraarylporphyrins in yields from 20-40 %.1 Synthesis of unsymmetrical mew-tetraarylporphyrins via these methods can be problematic as the resulting mixtures of porphyrins are often difficult to separate by chromatography; however, directed syntheses (e.g. MacDonald "2+2" syntheses) of unsymmetrical porphyrins have been employed to circumvent the separation problem.2'3 With the exception of TPhP, the starting porphyrins described in this chapter all contained pyridyl or imidazolyl groups. Adler's method was used to synthesize the starting porphyrins, as poor results were obtained with pyridyl- or imidazolyl-aldehydes using Lindsey conditions. The porphyrins from a mixed aldehyde condensation (TPh4_nPynP; n = 0 to 4, Section 3.2.2.1) were used recently in our group as the basis for the preparation of novel porphyrins and metalloporphyrins incorporating pyridyl, N-methylpyridinium, sulfonato-, nitro-, and amino-phenyl groups.4"6 This series was the basis for the synthesis of new Pt-porphyrins (Section 3.2.5), a tirapazamine-porphyrin conjugate (Section 3.2.6), novel l-oxido-4-pyridyl porphyrins and porphyrin-N-oxides (Section 3.2.3). Also, several of these porphyrins were analyzed by cyclic voltammetry as part of their evaluation as radiosensitizers and hypoxia-selective cytotoxins (Chapter 5). 3.2 Experimental A description of the methods and materials used here can be found in Section 2.2.1. References on page 138 86 3.2.1 Porphyrin Precursors 3.2.1.1 4(5)-Hydroxymethylimidazole (19) Compound 19 was prepared according to a published method.7 Fructose (50 g, 0.26 mol), cupric carbonate CuCO3Cu(OH)2»H20 (120 g, 0.5 mol), cone. NFL4OH (400 mL), and formaldehyde (60 mL, 0.7 mol as a 37-40 % wt solution) were dissolved in distilled H 2 0 (750 mL) and the mixture was heated (60 °C) for 0.5 h. Then air was bubbled through the mixture kept at 75-80 °Cfov 2 h; the mixture was cooled and kept at 0 °C overnight. The grey-green precipitate was filtered off, resuspended in distilled H 2 0 (600 mL) and cone. HC1 (15 mL); H 2S (g) was bubbled through the suspension for 2.5 h and the resulting black precipitate was filtered off. To the filtrate was added picric acid (63 g, containing 35% wt. H20,0.18 mol) and the solution brought to reflux for 15 min. As the solution cooled, the crude picrate (18) crystallized as flat, yellow-green needles (70 g), m.p. 200 °C (lit. 203.5-206 °C) 7 which were recrystallized once from H 2 0 to yield 68 g of pure 18. ^H-NMR (200 MHz, DMSO-d<$) 8 14.2 (s br, 1H, N-H), 9.05 (s, 1H, 2-imidazolyl H), 8.60 (s, 2H, 3,5-phenyl H), 7.55 (s, 1H, 5-imidazolyl H), 5.6 (s br, < 1H, CH2-0#), 4.5 (s, 2H, -CH2-); the 2H-NMR data were not reported previously.) Compound 18 was then mixed with cone. HC1 (68 mL), distilled H 2 0 (180 mL) and toluene (340 mL), and the mixture was heated until 18 dissolved. The mixture was cooled and the aqueous layer was washed with toluene (4 x 200 mL); the organic fractions were combined, and the solvent removed to yield 19.5 g (63 % yield) of crude 19»HC1. 19»HC1 was recrystallized from EtOH and a second crop was collected after adding Et 20 to the mother liquor. The product was used without further purification. *H-NMR (300 MHz, DMSO-4 ) 8 9.04 (s, 1H, 2-imidazolyl H), 7.50 (s, 1H, 5-imidazolyl H), 5.7 (s br, < 1H, -CH20#), 4.50 (s, 2H, C#2OH). The ! H - N M R data for 19*HC1 matches those previously reported.8 Alternatively, a procedure adapted from Grimmet and Richards9 was used to circumvent the use of potentially explosive picric acid. The reaction was carried out as above and, after H 2S treatment and filtration, the filtrate was eluted down a column of Amberlite® IR-120(H) cation exchange resin. The column was then eluted with H 2 0 until the eluate was free from sugar related References on page 138 87 by-products of the reaction; 19 was eluted from the column with cone. NH4OH; the solvent was removed, and the product purified by column chromatography on sihca/(toluene:EtOH 1:1). The eluate was collected in fractions and those with the purest 19 (iH-NMR, tic) were used to produce 20 (Section 3.2.1.2). This procedure, although longer and more labour intensive than the picric acid route, did give the imidazole (19) free from HC1. iH-NMR (200 MHz, DMSO-cfe) 5 7.60 (s, 1H, 2-imidazolyl H), 6.85 (s, 1H, 5-imidazolyl H), 4.40 (s, 2H, C#2OH). The iH-NMR data were not reported previously. 3.2.1.2 4(5)-Imidazolecarboxaldehyde (20) Compound 20 was synthesized according to a literature procedure.10 19 (3.0 g, 30.8 mmol) and activated Mn02 n ' 1 2 (25 g) were mixed in 1,4-dioxane (80 mL) and the solution refluxed for 30 min; the MnC>> was filtered off and washed with warm 1,4-dioxane (250 mL). The solvent was removed from the filtrate on a rotary evaporator to yield 1.48 g (15.4 mmol, 50 %) of crude 20, which was used without further purification for the synthesis of tetrakis(4(5)-imidazolyl)porphyrin (4-TImP). A sample was recrystallized from H2O to yield analytically pure 20. m.p. 174-175 °C; iH-NMR (200 MHz, DMSO-dg) 8 13.0 (s br, < 1H, N-H), 9.75 (s, 1H, -CHO), (7.95 (s, 1H), 7.90 (s, 1H) 2,4-imidazolyl H). The iH-NMR and m.p. data agree well with literature values. This procedure was also used with 19*HC1 with similar results, but 19*HC1 had poorer solubility in 1,4-dioxan. 3.2.2 Tetraarylporphyrins 3.2.2.1 Mixed-aldehyde Condensation According to a literature method,4,13 benzaldehyde (6 mL, 0.059 mol), 4-pyridinecarboxaldehyde (8.0 mL, 0.084 mol) and pyrrole (2.15 mL, 0.127 mol) were mixed in propionic acid (500 mL), and the solution refluxed for lh. The solution was cooled and concentrated to 100 mL. Acetone (500 mL) was then added, and the crude porphyrin precipitate (4.5 g), which contained all 6 possible porphyrins, was filtered off. Portions of the crude mixture were purified by dissolving them in CHCI3 and chromatographing on silica/(CHCl3 to References on page 138 88 CHCI3:acetone 20:1) to yield TPhP and TrPhPyP as bands 1 and 2, respectively. The remaining porphyrins ((*-DPhBPyP), (c-DPhBPyP), (PhTrPyP), (TPyP)) could be separated on neutral Al203(in-IV), but better results were obtained with Florisil, eluting with CHCfj-.acetone (20:1 to 5:1) or CH2Cl2:MeOH (20:1 to 4:1). The products obtained gave characterization data identical to those reported previously.4,13 The yields are summarized in Table 3.1. Table 3.1 Results of the Mixed-aldehyde Synthesis Porphyrin Abbreviation Band Number / Yield 5,10,15,20-tetraphenylporphyrina TPhP 1/0.18 g (0.7 %) 5,10,15-triphenyl-20-(4-pyridyl)porphyrina TrPhPyP 2 / 0.79 g (3.0 %) 5,15-diphenyl-10,20-bis(4-pyridyl)porphyrinb ^-DPhBPyP 3 / 0.27 g (1.2 %) 5,10-diphenyl-15,20-bis(4-pyridyl)porphyrinD c-DPhBPyP 4 / 1.01 g (4.8 %) 5-phenyl-10,15,20-tris(4-pyridyl)porphyrinD PhTrPyP 5 / 1.43 g (4.2 %) 5,10,15,20-tetrakis(4-pyridyl)porphyrinb TPyP 6 / 0.60 g (2.5 %) ' isolated by chromatography on silica/(CHCl3:acetone 20:0 to 20:1). bIsolated by chromatography on Florisil/(CHCl3:acetone 20:1 to 5:1 or CH 2 Cl 2 :MeOH 20:1 to 4:1). 3.2.2.2 TPyP Via Lindsey Conditions The synthesis of TPyP via Lindsey conditions was investigated as a simple model for other porphyrins (e.g. BPyP, 4- and 2-TImP) incorporating heterocyclic aromatics. These conditions, however, did not produce significant amounts of TPyP. When a solution of pyrrole (10 mM) and 4-pyridinecarboxaldehyde (10 mM) in CHCI3, was treated with TCA, TFA or BF3»MeOH, a hazy, tan-coloured solution resulted. Continued stirring yielded a very polar, blue precipitate/sludge which stuck to the walls of the reaction flask. A similar precipiate was noted in the synthesis of BPyP (Section 2.3.3). The blue compound was soluble in MeOH (UV-Vis 600 nm broad) and was too polar to be chromatographed on AI2O3/ or References on page 138 89 silica/MeOH. Compared to the iH-NMR spectrum (DMSO-d<5) of 4-pyridinecarboxaldehyde, the new compound had a weak aldehyde signal (8 10.05) and multiplets (8 8.9, 8.0; each ~10x as intense as the aldehyde signal) in place of the pyridyl ring doublet of doublets. These results indicated that some reaction was occurring both at the aldehyde and pyridyl ring (See Section 3.3.1). The rate of the formation of the precipitate increased with increasing acid strength (i.e. TCA < TFA < BF^'MeOH) and acid concentration. One BF3-catalyzed reaction was monitored by UV-Vis spectroscopy after aliquots of the reaction mixture were treated with TCQ which oxidized the porphyrinogen formed in-situ to TPyP. 1 4 At 30 min, a weak Soret band (418 nm) was present; subsequent measurements showed this band weakening, and by 1.5 h no Soret band was seen. Some porphyrinogen was probably formed but disappeared as the reaction proceeded. Control reactions with 4-pyridinecarboxaldehyde and TFA in CHCI3 showed no reaction, indicating that formation of the blue compound required a pyrrole. In TPyP synthesis, attempts using other acid catalysts (p-TsOH, DCA, and propionic acid (5-20 mM)), evidence for bis(pyrrolyl)methanes was observed by TCQ oxidation to a bis(pyrrolyl)methene;15 UV-Vis (CHCI3) 470 nm), but no TPyP was formed when the solutions were treated with TCQ. 3.2.2.3 Tetrakis-(2-imidazolyl)porphyrin (2-TImP) According to the procedure described by Milgrom et al.,16 pyrrole (3.1 mL, 45 mmol) was added to a refluxing solution of 2-imidazolecarboxaldehyde (4.52 g, 47 mmol) in propionic acid (250 mL) and the solution kept at reflux for 1.5 h. At lower temperatures (-60 °C), the yield was much lower, presumably because of the poorer solubility of the aldehyde. The solution was allowed to cool and kept at ~5 °C overnight; 2-TImP was filtered off, washed with propionic acid (20 mL), CHCI3 (100 mL), acetone (100 mL), ethanol (50 mL), and methanol (100 mL) to yield 288 mg (4%) of crude porphyrin. The product was further purified by dissolving it in dilute HCl(aq.), neutralizing the solution (NaHC03), and filtering off the 2-TImP precipitate. R f = 0.12, silica/(CH2Cl2:MeOH:MeCN:TFA 80:20:1:0.5); iH-NMR (200 MHz, acetone-^TFA) 8 9.5 (s br, 8H, P-pyrrole H), 8.55 (s, 8H, 4,5-imidazolyl H); UV-Vis (0.01 N HC1 ( a q )) 412, 512, 582, References on page 138 90 630, 662; LR-MS (EI) m/e = 635 (1, M+ -2H +Cu), 574 (2, M+), 508 (1, M+ -imidazolyl), 68 (100, imidazole-1"1); HR-MS (EI) calc'd for C 3 2 H 2 2 N 1 2 : 574.20905, found 574.20948; Analysis calc'd for 0 ^ 2 2 ^ 2 * 0 . 5 TFA: C, 62.75; H, 3.59; N, 26.61; found: C, 62.60; H, 3.88; N, 25.36. The iH-NMR and UV-Vis data agree well with those reported in the literature. However, Milgrom et al. reported peaks at 8 1.28 and 1.26 which were assigned to the propionic acid of crystallization; these peaks were not observed using this workup procedure. For smaller scale reactions from which no porphyrin precipitated, an alternate, albeit lengthier, purification route to isolate 2-TImP was used. The propionic acid was removed under reduced pressure and the product initially purified by column chromatography on Al203(I)/(CH2Cl2:MeOH:NH40H 80:20:0.5), which removed significant amounts of black material. Then the solvent was removed and the product was washed with acetone and EtOH until the filtrate ran clear. Subsequent preparative-tic of the precipitate on stiica/(CH2Cl2:MeOH:MeCN:TFA 80:20:1:0.5) gave 2-TImP as a bright green band (porphyrin dication). The product was isolated from the silica with (CH2Cl2:MeOH:TFA 100:10:0.1), and the solvent was removed. The product was taken up in dilute HCl(aq.) and the solution then neutralized (NaHC03); the porphyrin precipitate was filtered off and rinsed with MeOH and EtOH. Then the product was washed from the filter with CH2CI2/TFA (100:1) and the solvent was removed; the product was then dissolved in dilute HCl(aq.), and the solution was neutralized (NaHC03) to precipitate the porphyrin. Attempts to synthesize 2-TImP via Lindsey-type conditions (BF3*MeOH catalysis in CHCI3 and TCQ oxidation)14 were unsuccessful, perhaps in part because of the poor solubility of 2-imidazolecarboxaldehyde in CHCI3. No 2-TImP was observed using MeOH as the solvent, TFA or TsOH as the acid catalyst, and air oxidation.17 3.2.2.4 Nitration of 2-TImP According to the conditions described by Meng et al. for the nitration of phenyl-substituted porphyrins 4 2-TImP (0.05 g) was dissolved in acetic acid: H2SO4: H N 0 3 (12 mL:4 mL:4 mL) and the mixture was stirred at room temperature overnight. No products were observed by RPtlc References on page 138 91 on RP-18-silica/ (H20:MeOH:MeCN:TFA 2:1:1:0.01). The mixture was then heated to -80 °C for 2-3 h and one product at a lower Rf than that of 2-TImP was observed. After the mixture was heated for another 12 h, at least 3 lower Rf products were evident in addition to much baseline material; however, after another 24 h of heating, no porphyrin remained according to tic and UV-Vis spectroscopy. Other attempts based on classical conditions and TFAA/nitrate salts18 to obtain a single product were unsuccessful (Table 3.2). Table 3.2. Conditions for Attempts at 2-TImP Nitration3 Solvent Nitration Reagent RPtlc analysis Other observations HN0 3/H 2S0 4(1:1) H N 0 3 several products and much baseline material some loss of porphyrin0 MeCN TFAA / excess NH4N0 3 baseline material only rapid bleaching of porphyrin0 CHC13 TFAA / excess NH4N0 3 not checked precipitate formed, nothing identified in iH-NMR spectrum MeCN orCHCl 3 TFAA / stoichiometric NH4N0 3 several products and much baseline material TFA TFAA/< 1 eq. NH4N0 3 several products new iH-NMR peaks 6- 5 ppm (CD3OD) a All reactions carried out at room temperature. b Assessed by UV-Vis spectroscopy. 3.2.2.5 Tetrakis(4(5)-imidazolyl)porphyrin (4-TImP) Based on the conditions reported by Huang et al., 19 4(5)-imidazolecarboxaldehyde (20) (1.08 g, 11.2 mmol) and pyrrole (0.7 mL, 10.1 mmol) were dissolved in glacial acetic acid (200 mL), and the solution was refluxed in air for 1 h. Huang et al. reported that a crystalline material precipitated from the mixture; subsequent washing with acetone and EtOH, and then chromatography on silica/(acetone to EtOH(95%)) gave 4-TImP in 20 % yield. However, in this thesis work no precipitate was found in the cooled solution. When the solution volume was reduced and acetone was added, no precipitate was formed (in contrast to the mixed aldehyde reaction, Section 3.2.2.1). Many stationary phase and eluent combinations were investigated for the isolation and purification of 4-TImP based on reported conditions used for the purification of References on page 138 92 imidazoles and nucletotide bases.20'21 Stationary phases such as silica, cellulose and AI2O3 were eluted with mixtures of up to 3 of the following solvents: acetic acid, H2O, iPrOH, n-BuOH, NH4OH, CHCI3, EtOH, MeOH, toluene, EtOAc, HCOOH, pyridine, and Et 20. Chromatography with relatively non-polar solvents (e.g. CH2C12:acetone) did not move the porphyrin on silica or alumina. The use of large quantities of MeOH or other polar solvents led to co-elution of some of the many side-products. The best results were obtained with chromatography on neutral Al203(IU-rV)/(CHCl3:MeOH:NH40H(COnc.) 80:25:1), when 4-TTmP eluted as a maroon band contaminated with AI2O3. The solution was filtered and the solvent removed to yield 0.085 g (6 % yield) of crude 4-TImP. Some of the column fractions grew small crystals, which were submitted for X-ray analysis, but were found to be unsuitable. iH-NMR (300 MHz, DMSO-sfo) 5 13.0 (s, ~3H, imidazolyl-N//), 9.20 (s, 8H, P-pyrrole H), 8.25 (s, 4H, 2-imidazolyl H), 7.95 (s, 4H, 4-imidazolyl H), -2.65 (s, ~1H, pyrrole N-fl); UV-Vis (acetic acid / EtOH) 416, 512, 548, 590, 648 nm. The characterization data agree favorably with those reported in the literature.19 One attempt was made to synthesize 4-TImP from 20 and pyrrole in dry MeOH with TFA as a catalyst and air oxidation,17 but no porphyrin was detected by UV-Vis when the reaction mixture was purged with O2. 3.2.2.6 Mixed-aldehyde Condensations with Pyrrole, Imidazolecarboxaldehyde and Pyridinecarboxaldehyde Under conditions similar to those described in Section 3.2.2.1, pyrrole, 4-pyridine carboxaldeyde, and 4(5)-imidazolecarboxaldehyde (2:1:1 ratio) were mixed in acetic acid and the solution refluxed for ~1 h. The reaction solution was then cooled; the volume of solvent was reduced under reduced pressure, and acetone was added; the mixture was filtered when most of the porphyrinic material remained in the filtrate. At least 4 porphyrin products were present in low yield according to tic (Al203/(CHCl3/MeOH 10:1)) and UV-Vis spectroscopy. Column chromatography on Al203(H)/ or Florisil/(CHCl3:MeOH 10:1) achieved little separation and only the least polar porphyrin (TPyP by ! H-NMR and UV-Vis spectroscopy) was cleanly isolated. References on page 138 93 Because of the low yields and difficulty in purifying these porphryins, their synthesis and characterization were not pursued further. An attempt was made using similar conditions with pyrrole, 4-pyridinecarboxaldehyde, and 2-imidazolecarboxaldehyde. Porphyrin products were produced in very low yield (UV-Vis) and the synthesis was not pursued further. The poor solubility of 2-imidazolecarboxaldehyde in propionic acid may be a factor in the poor results obtained here (see Section 3.3.1). 3.2.3 N-Oxidations Based on published procedures for the preparation of pyridine-N-oxide,22 excess ra-CPBA (60-85 %, the rest being benzoic acid and H2O) was added to a stirred solution of the starting porphyrin in CH2CI2 or CH2Cl2/MeOH at room temperature. Addition of 1-2 equiv. aliquots of m-CPB A gave the (oxidopyridyl)porphyrins, while a single addition of several equiv. gave the corresponding porphyrin N-oxides in low and variable yield. The reactions were monitored by tic on silica/(CH2Cl2:MeOH 100:1 to 10:1); Et 3N was added to the final reaction mixture, and the product was purified by chromatography after pre-adsorbing the product mixture on silica. With very polar eluent mixtures, metallation (with Zn from the fluorescent indicator in the silica) was observed to occur on the tic plate. In such cases, analytical tic plates without indicator were used. If the porphyrin-N-oxides were not produced or their isolation was not pursued, the product was purified by first removing the solvent under reduced pressure, and the residue then being washed with acetone. The resulting solid contained mostly the (oxidopyridyl)porphyrin, while the filtrate contained the corresponding porphyrin N-oxides and other reaction byproducts (e.g. large amounts of m-chlorobenzoic acid (m-CBA)). The excess m-CBA complicated the chromatography, and could be removed by washing the reaction mixture with dilute NaOH(aq.) In some cases, mixtures of porphyrins (e.g. c-DPhBPyP and PhTrPyP) were treated with ra-CPBA to give mixtures of (oxidopyridyl)porphyrins and the corresponding porphyrin-N-oxides. The mixtures were separated by chromatography, but no yield was determined for the minor components, as the molar ratio of the compounds was not determined prior to the reaction. References on page 138 94 Because of the polar solvent mixtures used in chromatography, the porphyrin samples were often contaminated with silica. To obtain analytically pure material, the porphyrin was allowed to crystallize slowly (days to weeks) from a CH2Ci2:MeOH (-10:1) solution in a vial placed in a chamber containing EtOH (see Fig. 3.1). Porphyrin in CH 2 Cl 2 :Me0H Figure 3.1. Crystallization Setup for the (l-Oxido-4-pyridyl)porphyrins. The (oxidopyridyl)porphyrins were deoxygenated under the conditions of EI or LSIMS mass spectrometry and had to be analyzed by MALDI-TOF mass spectrometry. The corresponding porphyrin-N-oxides were deoxygenated to a lesser degree, and in some cases the parent peaks were intense enough to be analyzed by HR-MS (EI). The elemental analysis, iH-NMR, UV-Vis, IR, and mass spectrometry data are summarized in Tables 3.3 to 3.9 found after Section 3.2.3.9. 3.2.3.1 5-(l-Oxido-4-pyridyl)-10,15,20-triphenyl-20)porphyrin (OPyTrPhP) TrPhPyP (0.223 g, 0.36 mmol) was dissolved in CH2CI2 (150 mL), and treated with ra-CPBA (3 x O.lg aliquots, ~5x excess, based on 85% m-CPBA) over the course of 1-2 h while the solution was stirred. Et 3N (20 mL) was added at the end of the reaction; the product was pre-adsorbed on silica and chromatographed on siUca/(CHCl3:pyridine 10:1), and the solvent was removed from the main product band to yield 0.128 g (56% yield) of OPyTrPhP. The reaction was scaled up to -0.5 g of TrPhPyP with similar results. Slow solvent evaporation from a References on page 138 95 CH2Cl2/MeOH solution of the porphyrin (Figure 3.1) yielded analytically pure crystals which were analyzed by X-ray crystallography (see Section 3.3.2.2). OPyTrPhP was also analyzed by cyclic voltammetry (see Chapter 5). 3.2.3.2 5-(l-Oxido-4-pyridyl)-10,15,20-triphenyIporphyrin-21-oxide (OPyTrPhP-210), and -23-oxide (OPyTrPhP-230) From reactions in which several equiv. of m-CPBA were added all at once to a solution of TrPhPyP in CFPiC^MeOH, two additional products were isolated (~5 % yield each) after OPyTrPhP was first isolated via column chromatography on silica/(CHCl3:pyridine 10:1). The 2 bands following that of OPyTrPhP were collected; the solvent was removed, and a second chromatography column silica/(CH2Cl2:THF:MeOH 100:5:2.5) was used to separate two isomers. The less polar band was determined to be OPyTrPhP-230, and the more polar band was OPyTrPhP-210. OPyTrPhP-230 OPyTrPhP-210 The porphyrin N-oxides metallated rapidly when mixed with Zn(OAc)2 in CF^CFi/MeOH to give OPyTrPhPZn-210 and OPyTrPhPZn-230 (see Table 3.5 for UV-Vis data). Attempts to grow crystals of the free base- or Zn-porphyrin-N-oxides were unsuccessful. 3.2.3.3 5,15-Bis(l-oxido-4-pyridyl)-10,20-diphenyIporphyrin ( * -BOPyDPhP) t-BOPyDPhP was prepared in the same manner as OPyTrPhP. m-CPBA was added in 1-2 equiv. aliquots to a solution of an impure sample of TrPhPyP containing a small amount of t-DPhBPyP. The products were purified by column chromatography on silica; OPyTrPhP was eluted from the column with CHCI3pyridine (10:1); f-BOPyDPhP was eluted with CHCI3pyridine References on page 138 96 (10:3). The yield was not determined. Because only small amounts of f-DPhBPyP were obtained from the mixed-aldehyde synthesis (Section 3.2.2.1), the synthesis of the corresponding porphyrin-N-oxide was not pursued. 3.2.3.4 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin ( c - B O P y D P h P ) c-BOPyDPhP was synthesized in the same manner as OPyTrPhP. To a solution of c-DPhBPyP (0.5 g, 0.8 mmol, a small percentage of PhTrPyP was also present) in CH2Cl2 :MeOH (125 mL:25 mL) was added m-CPBA (4 x -1.25 g ahquots, -30 excess based on 85 % m-CPBA), and the reaction monitored by tic. E13N (-10 mL) was added; the product was pre-adsorbed on silica and intially chromatographed on sihca/(CH2Ci2:pyridme:MeOH 20:1:2 to 20:1:3). The product bands were poorly resolved, possibly because of large amounts of residual m-CPBA and m-chlorobenzoic acid; the porphyrin-containing fractions were washed with dilute NaOH(aq). The product mixture was rechromatographed on silica/(CH2Cl2:pyridine:MeOH 10:1:2.5); c-BOPyDPhP eluted first (~0.27g, -50 % yield), followed by a mixture of porphyrin N-oxides (Section 3.2.3.5), and then a small amount of TrOPyPhP (Section 3.2.3.6). A c-BOPyDPhP sample of analytical purity was obtained by slow crystallization from a solution of the compound in CH 2 Cl 2 :MeOH (Figure 3.1). 3.2.3.5 5,10-Bis(l-oxido-4-pyridyl)-15,20-diphenylporphyrin-21-oxide (c-BOPyDPhP-210), -22-oxide (c-BOPyDPhP-220), and -24-oxide (c-BOPyDPhP-240) A mixture of porphyrin-N-oxides was produced (-30 mg, -5 %) when c-DPhBPyP was treated with a large excess of m-CPBA (Section 3.2.3.4). Attempts to separate the mixture on sihca/(CH2Cl2:pyridine:MeOH 20:1:2.5) were unsuccessful. There are three possible porphyrin-N-oxide isomers: c-BOPyDPhP-210, c-BOPyDPhP-220 and c-BOPyDPhP-240. References on page 138 97 o" 6- 6 _ c-BOPyDPhP-210 c-BOPyDPhP-220 c-BOPyDPhP-240 The iH-NMR data given below support the presence of at least two of these products; the two doublets (8 7.57 and 7.51) indicate the presence of c-BOPyDPhP-210; the singlet at 8 7.60 indicates the presence of either c-BOPyDPhP-240 or c-BOPyDPhP-220, or both. ! H-NMR (400 MHz, CDC13) 8 9.05 (d, J 5), 8.99 (overlapping ds, J 5), 8.93 (d, J 5), 8.91 (d, J 5), 8.85 (d, J 5), 8.74 (d, J 5), 8.65 (m), 8.22 (dd, Ji 6, J 2 1.5), 8.15 (m), 8.08 (d), 7.8 (m), 7.60 (s), 7.57 (d, J 6), 7.51 (d, J 6). Because the ratio of products was unknown, the !H-NMR signals were not assigned; general 8 values for the protons on the (oxidopyridyl)porphyrins and porphyrin-N-oxides can be found in Tables 3.3 and 3.4. 3.2.3.6 5,10,15-Tris(l-oxido-4-pyridyI)-20-phenylporphyrin (TrOPyPhP) TrOPyPhP was produced in the same manner as OPyTrPhP (Section 3.2.3.1). PhTrPyP was present in small amounts in the sample of c-DPhBPyP used in Section 3.2.3.4, and TrOPyPhP was isolated from the reaction mixture. The yield was not determined. A sample of analytical purity was obtained by slow crystallization from CH2Cl2:MeOH (Figure 3.1). 3.2.3.7 5,10,15-Tris(l-oxido-4-pyridyl)-20-phenylporphyrin-21-oxide (TrOPyPhP-210), and -22-oxide (TrOPyPhP-220) TrOPyPhP-210 and TrOPyPhP-220 were prepared in the same manner as the porphyrin-N-oxides of OPyTrPhP (Section 3.2.3.2). The product was purified by preparative tic on silica/(CH2Cl2:MeOH 10:1 to 5:1). Only ~1 mg of each porphyrin-N-oxide was obtained and the yield was not determined. References on page 138 TrOPyPhP-210 TrOPyPhP-220 3.2.3.8 5,10,15,20-Tetrakis(l-oxido-4-pyridyl)porphyrin (TOPyP) Oxidation of TPyP TOPyP was prepared in the same manner as OPyTrPhP. TPyP (0.35 g, 0.57 mmol) was dissolved in CH2Cl2/MeOH (150 mL:25 mL) and m-CPBA (6 x 0.5 g, -26 x excess, based on 85 % m-CPBA) was added in aliquots over 3 h, while the mixture was stirred. Tic showed at least 4 intermediates as the reaction progressed. Et3N (-20 mL) was added to the reaction mixture and the solvent was removed under reduced pressure. The residual solid was rinsed with MeOH, acetone and H2O and dried by drawing air through the precipitate. Further purification was achieved by dissolving the product in H20:TFA (20:1), neutralizing the solution (NaOH), and subsequently filtering off the TOPyP precipitate. Total yield 0.275 g (0.40 mmol, -70 %) of crude TOPyP. Analytically pure TOPyP was obtained by slow crystallization from CH2Cl2:MeOH (Figure 3.1). Alternatively, small amounts of the product could be chromatographed on Al203(m)/(CH2Cl2:MeOH 20:1). TOPyP was insoluble in CDCI3, and thus its ^H-NMR spectrum was measured in CDCI3/CD3OD. Acid-catalyzed condensation of pyrrole and 4-pyridinecarboxaldehyde N-oxide TOPyP was also prepared via acid-catalyzed condensation of pyrrole and 4-pyridinecarboxaldehyde N-oxide. Pyrrole (0.3 mL, 0.4 mmol) and 4-pyridinecarboxaldehyde N-oxide (0.05 g, 0.4 mmol) were dissolved in propionic acid (2 mL) and the solution was refluxed for -50 min. UV-Vis spectroscopy indicated that a porphyrin had formed (Soret 418 nm), but tic References on page 138 99 analysis of the mixture on silica/(CH2Cl2:MeOH) indicated numerous side-products and so further isolation was not attempted. Attempts to synthesize TOPyP with Lindsey-type conditions in CH2CI2 with pyrrole and 4-pyridinecarboxaldehyde (4 or 8 mM), using TFA or BF3«MeOH and subsequent oxidation with TCQ, yielded only traces of porphyrin by UV-Vis spectroscopy. Analysis of the reaction mixture by tic (elution and treatment of the dc strip with Br vapor) before TCQ was added suggested the presence of a bis(2-pyrrolyl)methane (see Section 2.2.1). As TOPyP was conveniendy prepared in large quantities via oxidation of TPyP, the acid-catalyzed synthesis was not pursued further. 3.2.3.9 Reaction of m-CPBA with TPhP When a solution of TPhP in CH2Cl2:MeOFJ was treated with excess m-CPBA, the porphyrin 'bleached out' as evidenced by UV-Vis spectroscopy, and no porphyrin-N-oxide product was detected by tic. When TPhP was a part of a mixture of porphyrins that was treated with m-CPBA, traces of a porphyrin material more polar than TPhP, but less polar than TrPhPyP was observed by tic. Because of the small quantities, this product was not isolated or characterized (see Section 3.3.2.1). References on page 138 100 Table 3.3. ^ - N M R Data for the (l-Oxido-4-pyridyl)porphyrinsa Porphyrin (MHz, solvent) p-Pyrroleb 2,6- / 3,5-OPyc 2,6- / 3,4,5-Ph Pyrrole N-H OPyTrPhP (200, 8.92 d(2), 8.84 s(4), 8.64 dd(2) / 8.18 d(6) / -2.82 s(~1.5) CDC1 3 ) 8.82 d(2) 8.11 dd(2) 7.76 m(9) t-BOPyDPhP (200, 8.94 d(4), 8.83 d(4) 8.62 dd(4) / 8.18 d(4) / -2.87 s(~1.5) CDCI3) 8.10 dd(4) 7.80 m(6) c-BOPyDPhP (300, 8.94 d(2), 8.93 s(2), 8.63 dd(4) / 8.18 d(4) / -2.85 s(~1.5) CDCI3) 8.86 s(2), 8.83 d(2) 8.10 dd(4) 7.78 m(6) TrOPyPhP (400, 8.95 d(2), 8.94 s(4), 8.63 dd(6) / 8.18 d(2) / -2.88 s(~1.5) CDCI3) 8.83 d(2) 8.10 dd(6) 7.8 m(3) TOPyP (400, 8.95 s br(8) 8.67 dd(8)/ none observed in CDCI3/CD3OD) 8.19 dd(8) CDCI3/CD3OD (See Section 3.3.2.3) aMeasured at room temperature; 5 in ppm, signal pattern (number of protons). bJ values typically ~5 Hz. Typically, Ji = 7 Hz, J 2 = 1 Hz. Table 3.4. !H-NMR Data for the Porphyrin-N-oxides Measured in CDCl3 a Porphyrin P-Pyrroleb 2,6- / 3,5- 2,6- / 3,4,5-Ph P-Pyrrole-NOd Pyrrole N - H OPyC OPyTrPhP-230 9.07 d(l), 9.00 d(D, 8.8 overlapping ds(2), 8.75 d(l), 8.66 d(l) 8.63 d(2) / 8.10 d(2) 8.5 m(6) / 7.9 m(9) 7.5 s(2) 1.6sbr(>2)e OPyTrPhP-210 8.97 overlapping ds(2), 8.91 d(l), 8.83 d(l), 8.65 m(2) under P-pyrrole signal at 8.65 / under 2,6-Ph signal at 8.17 8.23 d(2), 8.17 d(4)/ 7.8 m(9) 7.56 d(l), 7.50 (KD -1.7 s br(>2)e TrOPyPhP-210 9.06 overlapping ds(2), 8.92 d(l), 8.85 d(l), 8.73 m(2) 8.64 m(6) / 8.14 d(2);8.07 d(4) 8.22 d(2) / 7.83 m(3) 7.60 d(l), 7.54 d(l) N-H peak not assigned because of large H 2 0 peak. TrOPyPhP-220f 9.07 d, 9.00 d, 8.94 d, 8.92 d, 8.74 d, 8.63 d 8.6 m / 8.14 m, 8.05 d under 8.14 m/ 7.8 m 7.61 s N-H peak not assigned because of large H 2 0 peak. aMeasured at room temperature in C D C I 3 (400 MHz); 5 in ppm, signal pattern (number of protons). An H 2 0 signal was observed in all spectra bJ values typically -5 Hz. CJ values typically -7 Hz. dJ values typically ~6 Hz; Addition of D 2 0 changed these signals to broad multiplets. integration includes H 2 0 peak; the peak disappeared when D 2 0 was added, integrations did not match. References on page 138 101 Table 3.5. UV-Vis Data for the (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides Porphyrin (solvent) N-oxide Region (log £) Soret (log £) Q Bands (log £) OPyTrPhP3 (CH 2Cl 2:MeOH 20:1) 274 (4.32) 420 (5.60) 516 (4.35), 552 (3.95), 590 (3.80), 646 (3.67) f-BOPyDPhPb (CH 2Cl 2:MeOH 10:1) 274, 310 (weak) 426 516, 556, 602, 648 c-BOPyDPhPa (CH 2Cl 2 :MeOH 10:1) 272 (4.49) 422 (5.57) 518(4.30), 556 (4.00), 592 (3.85), 650 (3.59) TrOPyPhpa (CH 2Cl 2 :MeOH 10:1) 274 (4.61) 424 (5.65) 518 (4.33), 554 (4.07), 594 (3.87), 650 (3.65) TOPyP a (CH 2Cl 2:MeOH 10:1) 274 (4.66) 424 (5.63) 518 (4.3), 554 (4.03), 594 (3.85), 650 (3.56) OPyTrPhP-230c (CH2C12) 282 (4.30) 420 (5.14) 518 (3.77), 544 (3.76), 594 (3.85), 684 (3.41) OPyTrPhP-210 (CH2C12) 280 420 516, 548, 596, 684 OPyTrPhPZn-230d'e (CH 2Cl 2:MeOH 5:1) 272 (4.35), 322 (4.26) 436 (5.27) 574 (3.89), 622 (3.91) OPyTrPhPZn-210 (CH 2Cl 2:MeOH 5:1) 276, 330 438 574, 622 c-BOPyDPhP-O mixture (CH2C12) 280, 342 (weak) 422 514, 550, 594, 688 TrOPyPhP-210 (CH2C12) 283, 334 (weak) 424 518, 548, 600, 685 TrOPyPhP-220 (CH2C12) 280, 348 (weak) 424 518, 552, 598, 685 aSoret e determined at 2 x 10"6 M; £ of other bands determined at 2 x 10"5 M. bThe C, H and N elemental analyses were -0.5 % too low (see Table 3.6); £ values were not determined. cThe C, and N elemental analyses were -0.5 and 1.26 % too low, respectively (see Table 3.6). dSoret £ determined at 5 x 10"6 M; £ of other bands determined at 5 x 10"5 M. ePrepared from a known concentration of OPyTrPhP-230 and excess Zn(OAc)2. References on page 138 102 Table 3.6. Elemental Analyses for the (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides Porphyrin / Formula C % H % N % OPyTrPhP / C 4 3 H 2 9 N 5 0 Calculated: 81.53 4.63 11.09 Found: 81.75 4.88 11.10 OPyTrPhP-230 / Calculated: 77.58 4.69 10.52 C43H 2 9N 5 0 2 '1 H 2 0 Found: 77.12 4.65 9.26 OPyTrPhP-210 / not obtained C43H29N5O2 r-BOPyDPhP / Calculated: 73.67 4.71 12.27 C42H28N602-2 H 2 0 Found: 73.12 4.24 11.86 c-BOPyDPhP / Calculated: 75.66 4.54 12.60 C 4 2 H 2 8 N 6 0 2 - H 2 0 Found: 75.83 4.41 12.73 c-BOPyDPhP -0 mixture / Calculated: 72.93 4.52 12.60 C 4 2 H 2 8 N 6 0 3 -1 .5 H 2 0 Found: 72.71 4.46 12.73 TrOPyPhP / Calculated: 71.09 4.37 14.15 C 4 1 H 2 7 N 7 0 3 » 1 . 5 H 2 0 Found: 71.04 4.31 13.93 TrOPyPhP-210 / not obtained C 4 1 H 2 7 N 7 0 4 TrOPyPhP-220 / not obtained C 4 1 H 2 7 N 7 0 4 TOPyP / C 4 0 H 2 6 N 8 O 4 . 2 Calculated: 68.56 4.03 15.99 H 2 0 Found: 68.55 3.98 15.75 References on page 138 103 Table 3.7. Infrared Spectroscopy Data for the (l-Oxido-4-pyridyl)porphyrinsa OPyTrPhP ?-BOPyDPhP c-BOPyDPhP TrOPyPhP TOPyP 3469 w br 3552 s 3500 br m 3390 m br 3400 s br 3312 w 3468 s br 3306 m 3319 m 3200 s br 3414 vs 3090 m 3101 m 3115 s 3092 w 3234 m 3050 w 3101m 1654 w 1640 m 1654 m 1654 w 1559 w 1617 m 1561 w 1466 s 1466 m 1466 m 1466 s 1470 m 1438 m 1437 w 1437 w 1437 m 1434 w 1397 w 1402 w 1401 m 1402 m 1348 m 1348 w 1349 w 1349 w 1261 vs 1258 s 1259 vs 1249 vs 1241 vs 1167 s 1171 s 1169 m 1175 s 1175 s 1032 w 1035 w 1098 w 999 w 997 w 1034 w 966 s 967 w 967 m 969 m 970 w 795 s 801 w 799 m 835 m 832 m 728 s 731 w 729 w 799 m 617 m 699 s 701 w 730 m aMeasured as a KBr pellet. References on page 138 104 Table 3.8. infrared Spectroscopy Data for the Porphyrin-N-oxidesa OPyTrPhP-230 OPyTrPhP-210 c-BOPyDPhP-O mixture TrOPyPhP-210 TrOPyPhP-220 3400 w br 3400 w br 3415 m br 3419 w br 3100 w 3108 w 3077 m 3102 m 3102 w 3053 w 3052 w 3022 w 2922 m 1732 m 1703 w 2851 m r 1554 w 1713 m 1466 s 1473 s 1469 s 1468 s 1632 w 1440 s 1439 s 1435 m 1437 m 1556 w 1334 w 1334 w 1368 w 1376 w 1464 m 1304 w 1333 w 1347 w 1375 m 1260 s 1261 s 1257 vs 1242 vs 1238 s 1169 s 1169 s 1168 vs 1174 vs 1174 s 1126 w 1125 w 1125 w 1125 w 1072 w 1088 w 1077 w 1030 w 1074 w 1033 w 1000 w 1000 w 978 w 962 m 961 s 961 w 965 w 966 w 928 w 897 w 895 w 802 s 832 w 831 m 831 w 802 m 755 m 801m 803 m 801 w 754 w 735 m 722 w 701 vs 701 vs 702 m 701 w 643 w 686 m aMeasured as a thin film on a KBr disk. References on page 138 105 Table 3.9. Mass Spectrometry Data for (l-Oxido-4-pyridyl)porphyrins and Porphyrin-N-oxides Porphyrin LR-MS m/e (intensity %, assignment)3 HR-MS formula calc'd / found OPyTrPhPb 632 (88, M+1) 616(36,M + 1 -0) OPyTrPhP-210c 692 (2, M+ -0 -2H +Cu), 676 (5, M+ -20 -2H + Cu), 647 (4, M+), 631 (10, M+ -0), 615 (100, M+ -20) C 4 3 H 2 9 N 5 O 2 647.23212 / 647.23207 OPyTrPhP-230c 692 (1, M+ -O -2H + Cu), 677 (2, M+ -20 -2H +Cu), 647 (1, M+), 631 (20, M+ -0), 615 (100, M+ -20) C 4 3 H 2 9 N 5 O 2 647.23212 / 647.22965 C 4 3 H 2 9 N 5 0 631.23718 / 631.23701 r-BOPyDPhPb-d 650 (100, M+ 1 ) , 634 (62, M + 1 -0), 614 (19, M + 1 -20) c-BOPyDPhPb'd 649 (100, M+ 1 ) , 633 (86, M + 1 -0), 617 (27, M + 1 -20) c-BOPyDPhP-NO mixture0 677 (5, M+ -20 -2H + Cu), 693 (2, M+ -O -2H +Cu), 664 (1, M+), 648 (5, M+ -0), 632 (29, M+ -20), 616 (100, M+ -30) C 4 2 H 2 8 O 3 N 6 664.22229 / 664.21942 C 4 2 H 2 8 O 2 N 6 648.22736 / 648.22516 TrOPyPhPb'd 666 (87, M+ 1 ) , 650 (100, M + 1 -0), 634 (66, M + 1 -20), 618 (33, M + 1 -30) TrOPyPhP-210c 710 (1, M+ -20 -2H +Cu), 694 (7, M+ -30 -2H +Cu), 678 (8, M+ -40 -2H +Cu), 665 (1, M+ -0), 649 (11, M+ -20), 633 (80, M+ -30), 617 (100, M+ -40) C 4 i H 2 7 0 3 N 7 665.21753 / 665.21719 C 4 l H 2 7 0 3 N 7 649.22260 / 649.22210 TrOPyPhP-22-NOc 678 (20, M+ -40 -2H +Cu), 665 (2, M+ -0), 649 (9, M+ -20), 633 (10, M+ -30), 617 (100, M+ -40) C 4 i H 2 7 0 3 N 7 665.21753 /665.21511 C 4 i H 2 7 0 3 N 7 649.22260 / 649.22175 TOPyPb 683 (51, M+ 1 ) , 666 (84, M + 1 -0), 650 (99, M+1 -20), 635 (100, M + 1 -30), 619 (79, M + 1 -40) a M + 1 implies (M+H)+. bMALDI-TOF mass spectrometry. CEI mass spectrometry. dDimer peaks (5-10 % intensity, due to combinations of the species listed in the LR-MS column) are observed. The dimers are formed in the gas phase.23 References on page 138 106 3.2.4 Sulfonations The iH-NMR, UV-Vis and elemental analysis data for the sulfonated porphyrins appear in Tables 3.10 to 3.12, respectively, found after Section 3.2.5.3. Other characterization data appear in the individual sections. 3.2.4.1 Sodium 5,10,15,20-Tetrakis(4-sulfonatophenyl)porphyrin (TSPhP) Based on a literature procedure,4 TPhP (0.5 g, 0.8 mmol, Section 3.2.2.1) was dissolved in cone. H2SO4 (10 mL) and the solution was heated at 110 °C for 4 h. The reaction mixture was cooled with an ice-bath and cold, distilled H2O (50 mL) was added, with the solution's pH being adjusted to 7-8 with NaOH(aq) and Na2C03(aq). The solvent was removed on a rotary evaporator and the porphyrin washed from the residue with MeOH. Then the MeOH was removed and the porphyrin subsequently dialyzed in 1000 MWCO dialysis tubing in a beaker of distilled H2O to yield 0.335 g (-40 %) of TSPhP»10H2O as shiny purple flakes. The iH-NMR data matched those reported by Meng et al.,4 and the TSPhP was used without further purification. 3.2.4.2 Sodium 5,10,15-Tris(4-sulfonatophenyl)-20-(4-pyridyl)porphyrin (PyTrSPhP) TrPhPyP (0.2 g, 0.3 mmol, Section 3.2.2.1) was sulfonated, and the resulting product was isolated using the same conditions as used for TSPhP (Section 3.2.4.1) to yield 0.27 g (72%) of PyTrSPhP* 13 H2O. The characterization data agreed well with the reported values.5 3.2.4.3 Sodium 5-(l-Oxido-4-pyridyl)-10,15,20-tris(4-sulfonatophenyl)porphyrin (OPyTrSPhP) OPyTrPhP (0.535, 0.85 mmol, Section 3.2.3.1) was sulfonated, and the product was isolated using the same conditions as used for TSPhP (Section 3.2.4.1). The reaction was monitored by first neutralizing an aliquot of the mixture (using Na2C03) then analyzing by tic on sihca/(CH2Cl2/MeOH 100:1 to 5:1). After dialysis, the solvent was removed to yield 0.51 g (-53% yield) of OPyTrSPhP* 1 1 H 2 0 . IR (KBr pellet) 3400 vs br, 3180 vs br, 1655 m, 1632 m, 1398 s, 1221 s, 1182 s, 1123 s, 1037 s, 1010 s, 965 w, 857 w, 800 m, 738 m, 636 m crrr1. The References on page 138 107 iH-NMR signals were assigned with the help of a 1H-2D-COSY spectrum (see Appendix, Figure A.l) . An alternate purification procedure based on repeated precipitation of the porphyrin from a MeOH solution with acetone4 was unsuccessful in removing all of the remaining salts, as judged by elemental analysis. 3.2.4.4 Sodium 5,10-Bis(l-oxido-4-pyridyl)-15,20-bis(4-sulfonatophenyl)porphyrin (c-BOPyBSPhP) c-BOPyDPhP (0.10 g, 0.15 mmol, Section 3.2.3.3) was sulfonated in the same manner used for TPhP (Section 3.2.4.1). The reaction was monitored by tic as described in Section 3.2.4.3; an intermediate band appeared between that of the starting material and the baseline, and heating was continued ~ 0.5 h after the intermediate had disappeared (6 h total). After dialysis of the worked-up product, its !H-NMR spectrum revealed the presence of an impurity. Redissolving the product in cone. H2SO4 and heating at 100 °C for another 2 h with the same work-up procedure did not remove the impurity (i.e. it is not a mono-sulfonatophenyl product). Attempts to purify by column chromatography on silica/MeOH or by recrystallization from MeOH/acetone were unsuccessful. IR (KBr pellet) 3414 vs br, 1640 m, 1482 m, 1437 w, 1397 w, 1224 s, 1182 s, 1123 s, 1038 s, 1011 m, 967 m, 831 w, 800 m, 735 m, 636 s cm"1; LR-MS (+ LSDVIS) m/e = 875 (0.1, M + 1 +Na), 859 (0.1, M+1 +Na -O), 853 (0.1, M + 1 ) , 837 (0.1, M + 1 - O). The results of the elemental analysis (Table 3.12) indicate that the composition of the impurity does not differ significantly from that of c-BOPyBSPhP. Careful tic analysis of the sample of c-BOPyDPhP used here showed the presence of another compound with an Rf nearly identical to that of c-BOPyDPhP; the presence of the impurity was not evident in the iH-NMR spectrum (CDCI3). 3.2.5 Platination Reactions In general, the best results for platination were obtained when K 2 P t C l 4 was added to an aqueous, buffered solution of the porphyrin. The !H-NMR, UV-Vis and elemental analysis data appear in Tables 3.10 to 3.12. Other characterization data appear in the individual sections. References on page 138 108 3.2.5.1 Sodium [5,10,15,20-Tetrakis(4-sulfonatophenyl)porphyrinato]Pt(II) (TSPhPPt) TSPhP (0.052 g, 0.05 mmol, Section 3.2.4.1) and K2PtCl4 (0.05 g, 0.12 mmol)) were dissolved in distilled H2O (15 mL); the UV-Vis spectrum of this green solution showed the presence of the TSPhP porphyrin dianion (the porphyrin core is protonated in acidic media and is a dication; the overall charge on the molecule is 2-; Soret at 432 nm).64 The pH was adjusted to 3-4 with NaOAc (0.075 g, 0.5 mmol) and glacial acetic acid (35 ^tL, 0.5 mmol), and the mixture was heated to -100 °C, the reaction being monitored by UV-Vis. A new Soret band (396 nm) grew in intensity while those of the free-base and dianion species decreased in intensity. At -20 h, the reaction was complete; the reaction mixture was filtered, the filtrate reduced in volume to 5 mL, and the porphyrin dialyzed in 1000 MWCO dialysis tubing in a beaker of distilled H2O for 3 h. The product was then eluted down a cation exchange column (Na+ form) and the solvent removed to yield 0.024 g (38%) of TSPhPPt. The UV-Vis data differ slightly from those reported in the literature, measured in 9:1 (CH20H)2:FJ.20.24 The iH-NMR data were not reported previously. 3.2.5.2 The Reaction of K 2 P t C U with PyTrSPhP Several attempts were made with limited success to platinate PyTrSPhP with K^PtCU- No metallation occurred when the reactants were combined in refluxing DMSO or in warm DMF (-50 °C) in air. When the reactants were heated in DMF under Ar, no platination occurred, but a Pt mirror formed on the inside wall of the reaction flask. When the reactants (5x molar excess K2PCI4) were dissolved in H2O and the solution was heated (-60 °C), the reaction proceeded in a manner similar to that described for TSPhP (Section 3.2.5.1), as evidenced by UV-Vis. However, the reaction did not proceed past 75 % completion (> 66 h) and the porphyrin eventually decomposed according to UV-Vis spectroscopy. When formic acid (pH -3) or a NaOAc/HOAc buffer was added to the reactants in H2O and the solution refluxed, the porphyrin decomposed rapidly; decomposition was noticably slower at -50 °C. This chemistry was not pursued further. References on page 138 109 3.2.5.3 The Reaction of K 2 P t C l 4 with OPyTrSPhP ("OPyTrSPhPPt") Metallation occurred readily when K^PtCla (20 mg, 48 jimol) and OPyTrSPhP* 1IH2O (30 mg, 9 |imol) were dissolved in buffered H2O containing HOAc:NaOAc (75 |0,L:0.17g), and the solution was refluxed. After several hours the reaction was complete as evidenced by UV-Vis; the isolated residue was dialyzed in 1000 MWCO dialysis tubing in water, and the solvent was removed . MeOH was added to the residue, a black material was filtered off, and the filtrate was vacuum evaporated leaving the solid product. IR (KBr pellet) 3400 vs br, 3200 vs br, 1665 m, 1632 m, 1562 m, 1545 m, 1400 m, 1210 shoulder on 1181 s, 1125 s, 1036 s, 1008 s, 820 w, 796 w, 742 w, 641 w cm"1; LR-MS (MALDI-TOF) m/e = 1154 (30, M + 1 +2H20), 1138 (44, M + 1 +H20), 1116 (67, M + 1 ) , 1094 (81, M + 1 -Na), 1072 (58, M+1 -2Na), 1050 (100, M + 1 -3Na), based on M+ = C43H24N5Na30aS3Pt. Cluster peaks appeared in the MALDI-TOF spectrum, perhaps due to dimers (-2165 to -2265 m/e) and trimers (-3375 m/e) of the M + 1 and related fragment species. No peaks were seen in the + or -LSPMS mass spectra. The elemental analysis (Table 3.12) showed that a product other than the desired [5-(l-oxido-4-pyridyl)-10,15,20-tris(4-sulfonatophenyl)porphyrinato]platmum(n) (trisodium salt) had formed (see Section 3.3.4). Table 3.10. !H-NMR Data for the Sulfonatophenylporphyrins and Pt-derivatives in DMSO-<i<5a Porphyrin (MHz) P-Pyrroleb 2,6-/ 3,5-Pyc or OPyd 2,6-/ 3,5-Phe Pyrrole N-H TSPhP (300) 8.89 s(8) 8.22 d(8)/ -2.9 s br(2) 8.08 d(8) TSPhPPt (400) 8.75 s(8) 8.12 d(8)/ 8.02 d(8) PyTrSPhP (300) 8.86 m(8) 9.03 d(2) / 8.28 d(2) 8.18 d(8)/ -2.97 s br(1.5) 8.04 d(8) OPyTrSPhP 9.05 d(2), 8.88 8.63 d(2) / 8.63 d(2) 8.17 d(6)/ -2.96 s br(1.7) (200) d(2), 8.85 s(4) 8.05 d(6) "OPyTrSPhPPt" 8.95 d, 8.80 d, 8.61 d/8.21 d 8.12 d / (200)f 8.76 s 8.00 d c-BOPyBSPhP 9.10 s(2), 9.05 8.64 d(4) / 8.25 d(4) 8.18 d(4)/ -2.95 s br(1.6) (200)8 aA/f™„..,„j „. ..... d(2), 8.87 m(4) 8.05 d(4) aMeasured at room temperature; 5 in ppm, signal pattern (number of protons). bJ values typically 6 Hz. CJ values ~4 Hz. dJ values typically -7 Hz. eJ values typically ~8 Hz. integrations did not match. Slmpurity peaks appeared near 5 -7.8 and 8.4 (< 0.5 H each by integration). References on page 138 110 Table 3.11. UV-Vis Data in H2O for the Sulfonatophenylporphyrins and Pt-derivatives Porphyrin N-oxide Region (log 8) Soret (log 8) Q Bands (log 8) TSPhPa 411 (5.67) 513 (4.19), 549 (3.85), 577 (3.81), 630 (3.59) TSPhPPtb 398 (5.35) 508 (4.25), 536 (4.60) PyTrSPhPa 411 (5.62) 513 (4.18), 555 (3.83), 577 (3.81), 632 (3.51) OPyTrSPhPb 258 (4.32), 306 (4.27) 414 (5.59) 516 (4.14), 554 (3.84), 580 (3.81), 636 (3.57) "OPyTrSPhPPt" 398 508, 538 c-BOPyBSPhP 258, 306 418 520, 560, 586, 646 aTaken from Meng et al. be determined at 1.0 x 10~6 M . Table 3.12. Elemental Analyses for the Sulfonatophenylporphyrins and Pt-derivatives Porphyrin / Formula C % H % N % TSPhPPt/ calculated 41.32 2.44 4.38 C44H24N4Na40i2PtS4«3.5H20 found 41.58 2.69 4.35 PyTrSPhP / calculated 44.68 4.53 6.06 C43H26N5Na309S3'13H20 found 44.36 4.18 5.80 OPyTrSPhP/ calculated 45.46 4.26 6.16 C 43H 2 6N 5Na 3O 1 0S3.11H 2O found 45.72 4.11 6.12 "OPyTrSPhPPt" calculated 33.38 2.34 4.53 C 43H2 4N 5Na30 9S3Pf6H20«PtCl3Na a found 33.46 2.30 4.24 c-BOPyBSPhP/ calculated 54.02 3.78 9.00 C 4 2H 2 6N 6Na 20 8S 2.4.5H 20 found 54.28 3.66 8.80 aSee Section 3.3.4. References on page 138 I l l 3.2.6 Chemistry of 3-Amino-l,2,4-benzotriazine-l,4-di-N-oxide (Tirapazamine) The chemistry of tirapazamine and its reaction product with triphosgene (21, Section 3.2.6.1) was investigated for the purpose of attaching tirapazamine to a porphyrin. Thus the reactions of 21 (containing small amounts of tirapazamine) with acetic acid (Section 3.2.6.2), butanol (Section 3.2.6.3), and amines (Sections 3.2.6.4), were performed. Porphyrins with various functional groups were also exposed to 21 under similar conditions with mixed results (Sections 3.2.6.5 and 3.2.6.6). The ipI-NMR, IR, and mass spectrometry data for the resulting benzotriazine-dioxide compounds appear in Tables 3.13 to 3.15, respectively. Other characterization data appear in the individual sections. Chapter 4 also describes other attempts to synthesize tirapazamine-porphyrin conjugates. 3.2.6.1 2H-[l,2,4]Oxadiazolo[3,2-c][l,2,4]benzotriazin-2-one-5-oxide (21) o—c=o 21 Based on a similar reaction between phosgene and tirapazamine described by Seng and Ley,2 5 a suspension of tirapazamine (0.3 g, 1.7 mmol) in dry toluene (5 mL) was heated to 90 °C and triphosgene (0.173 g, 0.58 mmol) was added. Within 5 min the suspension turned from orange to yellow and tic indicated near quantitative conversion to 21 (Rf = 0.75 silica/(CH2Cl2:MeOH ~50:1)). When the solution was heated for ~2 h, it appeared from tic that some 21 had converted back to tirapazamine (Rf = 0.18). The reaction mixture was cooled, and the precipitate was filtered off and washed with hexanes to yield 0.324 g (1.59 mmol, 94% yield) of 21 as a yellow-brown solid, that contained a small amount (< 10 %) of tirapazamine (by tic). The product was used without further purification in subsequent reactions. 21 was reconverted to tirapazamine when exposed to H2O vapor or when co-spotted with H 2 0 on a tic plate, and tirapazamine was isolated in variable amounts from subsequent reactions involving 21. For a reaction of an amine with phosgene, an isocyanate is the expected product; however, no NCO peak References on page 138 112 was seen in the IR spectrum of the product and Seng and Ley proposed the following structure for 21. 3.2.6.2 3-N-Acetamido-l,2,4-benzotriazine-l,4-di-N-oxide (22) Compound 21 (-15 mg, containing a small percentage of tirapazamine) was dissolved in dry pyridine (1 mL), and an excess of glacial acetic acid (-0.75 mL) was added. The solution was stirred and heated at 60-80 °C for 2 days, at which point the solvent was removed and the product purified by repeated preparative-tic on silica/(CH2Cl2:MeOH -50:1). A bright yellow product (Rf = 0.55, m.p. 185 °C) between the bands of 21 and tirapazamine was isolated from other minor side-products. The yield was not determined. Presumably, the amide 22 (Table 3.13) was formed by reaction of tirapazamine and acetic acid, and the H2O co-product reacted with 21 to produce more tirapazamine (see Section 3.3.5). The synthesis of 22 by treating tirapazamine with acetic anhydride has been described (m.p. 200 °C, dec.),26'27 but few characterization data were given. 3.2.6.3 Butyl (N-(l,4-dioxido-l,2,4-benzotriazin-3-yl))carbamate (23) Compound 21 (several mg) was dissolved in pyridine (1 mL); n-butanol (1 mL) was added, and the mixture stirred and heated (90 °C) for 1 h. 23 (Table 3.13; Rf = 0.66, m.p. 220 °C, dec.) was isolated from residual 21 and tirapazamine by repeated preparative tic on silica/(CH2Cl2/MeOH -50:1). The yield was not determined. Other related carbamates have also been reported.27 3.2.6.4 Reaction of 21 With Amines Reaction with en To a solution of 21 (31 mg, 0.15 mmol) in dry DMF (0.5-0.75 mL) at 50-60 °C was added en (5 p:L, 0.075 mmol). The solution rapidly turned from orange to purple, but became orange-red within a few minutes. Monitoring by tic indicated that 21 was consumed and tirapazamine was present as well as a dark orange material at the baseline. No further change was noted with addition of more 21 and further heating. The solvent was removed under reduced pressure, the References on page 138 113 residue taken up in hot acetone and the product was filtered off as an orange precipitate (-30 mg; m.p. 210 °C, dec). Analysis calc'd for C 1 8 H 1 4 N 1 0 O 6 : C, 46.36; H, 3.03; N, 30.03; found: C, 42.43; H, 3.56; N, 27.02. No peaks apart from that of the benzotriazine di-N-oxide fragment were observed by LR-MS using DCI(+), EI, or LSFMS. The elemental analysis did not match the expected structure (24) (Table 3.13). The existence of two different !H-NMR methylene peaks (Table 3.13) indicated that the isolated compound is not symmetrical. Reaction with piperidine Piperidine (1 mL) was added to a solution of 21 (several mg) in pyridine (1 mL) at 90 °C; the solution rapidly turned from orange-yellow to deep purple, and tic indicated that no 21 remained. The solvent was removed under reduced pressure to yield a purple product, which turned yellow in acidic media (acetic acid). Chromatography on sihca/(CH2Cr2:MeOH 50:1) did not separate the yellow (produced in the acidic Si02 environment) and purple products. The product was not characterized further, but clearly 21 reacted rapidly with amines. Seng et al. described the reaction of 21 with several amines, including piperidine, but few characterization data were reported.26 The purple product was reported to be the piperidinium salt of 25, which was converted to its free base by treatment with acetic acid. 3.2.6.5 Reaction of 21 with Porphyrins With APhTrPhP 5-(4-N-(N'-(l,4-dioxido-l,2,4-benzotriazin-3-yl)aminocarbonyl) aminophenyl)-10,15,20-triphenylporphyrin (TirapPhTrPhP) To a solution of 5-(4-aminophenyl)-10,15,20-triphenylporphyrin (APhTrPhP)4 (31 mg, 0.05 mmol, from G. Meng) in dry DMF (3 mL) was added 21 (11 mg, 0.055 mmol, 1.12 eq), and the mixture was stirred at room temperature in the dark for 26 h. Tic [silica/(CH2Cl2:MeOH 25 References on page 138 114 100:2-3)] showed a new product (Rf = 0.4), some starting porphyrin, but little 21; so another 1.0 equiv. of 21 was added and the mixture was stirred another 24 h. The reaction mixture contained a trace of a porphyrin (with the same Rf (0.65) as APhTrPhP) which did not react in the presence of excess 21 at -65 °C. The solvent was removed under reduced pressure, and the desired product was isolated from several minor products by preparative-tic on sihca/(CH2Ct2:MeOH 100:2-3). The product was isolated from the tic plate, washed with hexanes, dissolved in CH2Ci2/MeOH (20:1) and the solution was washed with H2O in a separatory funnel. The solvent was removed to yield 20 mg of a purple powder (-50%). UV-Vis (2.4 x 10"6 M , CH 2 Cl 2 /MeOH (100:1)) corresponding to the tirapazamine moiety: 280 (log £: 4.65), 478 (4.26) nm, corresponding to the porphyrin core: 418 (5.81), 514 (4.54), 550 (4.28), 586 (4.16), 646 (4.03); Analysis calc'd for c 5 2 H 3 5 N 9 ° 3 * 0 - 5 H 2 0 : C, 74.10; H, 4.30; N, 14.96; found: C, 73.96; H, 4.18; N, 14.41. TirapPhTrPhP was also analyzed by cyclic voltammetry (see Chapter 5). With protoporphyrin IX No appreciable reaction was observed by tic when protoporphyrin IX was mixed with excess 21 in pyridine or DMF and the solution heated to 90 °C for 36 h. With B H E t P I X D M E A small amount of a product with higher Rf was observed by tic when BHEtPIXDME (several mg, Section 4.2.5) was mixed with 21 (6x molar excess) in pyridine, and the solution was heated to 90 °C for 24 h. Because of the small quantities involved, the product was not characterized further. Isolable amounts of this new product may be obtainable with larger excesses of 21. This chemistry was not pursued further. TirapPhTrPhP References on page 138 115 Table 3.13. iH-NMR Data for the Benzotriazine-dioxide compounds.3 Compound (MHz, solvent) -CONtf Benzotriazine Ring ffi Other Substituent Tkapazamine (400, acetone-d )^0 8.28tofd(2), 7.98tofd(2), 7.64tofd(2) 7.38 s br(2) -NH2 2 1 (400, acetone-^ 8.47 d(l), 8.27 t of d(l), 7.97 d ofd(l), 7.80tofd(l) 2 2 (300, DMSO-cfe) 10.88 s br(l) 8.33 d of d(2), 8.08 t(l), 7.85 t(D 2.28 (s, -COCH3) (1) 2 3 (400, acetone-fife) 9.56 s br(< 1) 8.44 d(2), 8.16 t(l), 7.92 t(l) 4.31 t(2, J 6), 1.75 m(2), 1.52 m(2), 1.00 t(3, J 7): butyl H 2 4 (400, D 2 0 ) e no peak seenf 8.44 d, 8.30 d, 8.20 t, 7.90 t 3.68 t(J 6), 3.26 t(J 6): -CH2-TirapPhTrPhP (400, CDC13) 9.71 s(l), 8.81 s(l) 7.27 t(l), 7.04 m(2), 6.63 t(l) (8.91 s br(4), 8.84 s(4), p-pyrrole H); 8.35 d(2, J 8, 2,6-RNHPh-W); 8.25 m(6, 2,6-Ph-#); 8.00 d(2, J 8, 3,5-RNHPh-W); 7.78 m(9, 3,4,5-Ph-#); -3.19 sbr(2, pyrrole N-H) aMeasured at room temperature; 5 in ppm, signal pattern (number of protons). bTypical J values for doublets or triplets 8-9 Hz; for a doublet or triplet of doublets Ji ~9, J2 ~1 Hz. cThe LH-NMR data are similar to those reported in the literature (recorded in TFA). 2 5 dThe ^ - N M R were not reported previously. eIntegrauons did not match. fN-H peak seen in DMSO-efe at 8 10.05. 23 24 References on page 138 116 Table 3.14. Infrared Spectroscopy Data for the Benzotriazine-dioxide Compounds Tirapazamine3 21b 22 b 23 b 24 b TirapPhTrPhPb 3411 w 3460 w br 3150 sbr 3218 wbr 3328 m 3312 w 3274 w br 3170 w 3102 w 3330 w br 3098 w 3067 w 2958 w 1814 vs 1725 w 1727 s 1709 m 1692 w 1671 vs 1598 s 1639 m 1596 w 1597 m 1594 vs 1541 vs 1541 s 1547 vs 1555 vs 1526 vs 1408 vs 1437 vw 1493 m 1458 w 1493 w 1405 m 1362 vs 1386 s 1404 vs 1403 s 1402 s 1332 m 1344 s 1354 s 1357 m 1354 m 1311 m 1226 m 1256 w 1332 m 1336 s 1225 m 1196 m 1194 w 1272 m 1270 w 1181 w 1105 vs 1149 m 1100m 1229 m 1233 w 964 m 1025 w 1110m 1199 m 1200 w 828 w 1173 w 1099 m 799 m 772 m 1080 w 727 m 723 w 734 m 785 w 701 m aMeasured as a thin film on a KBr disk. bMeasured as a KBr pellet. References on page 138 117 Table 3.15. Mass spectrometry Data for the Benzotriazine-dioxide Compounds Compound (technique) LR-MS m/e (intensity %, assignment)3 HR-MS formula calc'd / found 21 (+CI) 205 (30, M+1), 163 (100, M+1 - NCO), 147 (75, M+1 -NCO -0) C8H5N4O3 205.03616 / 205.03622 22 (EI)b 220 (1, M+), 204 (5, M+ -O), 162 (M+ - N H C O C H 3 ) C9H8O3N4 220.05965 / 220.05893 23 (+CI) 279 (55, M+1), 205 (25, M + 1 -butoxy), 189 (10, M+1 - butoxy -O) 173 (30, M+1 - butoxy -20) 163 (100, M+1 -NHCOOBu) C12H15N4O4 279.10934/279.10931 24 See Section 3.2.6.4 TirapPhTrPhP (+LSIMS) 834 (0.25, M+1), 818 (1, M+1 -O), 802 (0.25 M+1 -20), 656 (12, M+1 -tirapazamine) C52H36O3N9 834.29411 /834.29456 C52H36N9O2 818.29919/ 818.29953 a M + 1 indicates (M+H)+. bThe parent peak was not observed by +CI. 3.2.7 Other Reactions Involving Pt and Ru Species Reaction of TSPhPPt with H2O2 Buehler etal. have described the oxidation of [5,10,15,20-tetra(p-tolyl)porphyrinato]Pt(II) to dichloro[5,10,15,20-tetra(p-tolyl)porphyrinato]Pt(IV) with H2O2 in a biphasic mixture of H2O/HOAC and CHCI3. 2 8 However, when TSPhPPt was treated with H202( a q) in H2O, the porphyrin decomposed as evidenced by UV-Vis spectroscopy. Reaction of TSPhP with R u C l 3 » 3 H 2 0 No reaction was observed by UV-Vis when TSPhP was mixed with RuCl3»3H20 in H2O and the solution refluxed. Ruthenation of TSPhP occurs readily with [Ru(dmf)6][CF3S03]3 and the TSPhPRu(CO)2 thus produced can be photolyzed in the presence of DMSO to produce TSPhPRu(DMSO)2-29 References on page 138 118 Reaction of TSPhPRu(DMSO)2 with imidazole One possible method of incorporating nitroimidazoles into porphyrins is via coordination to a metalloporphyrin moiety such as TrSPhPRu' which has two available axial coordination sites. As a simple model, imidazole was investigated. TrSPhPRu(DMSO)2 (several mg, from C. Ware) was dissolved in MeOH and excess imidazole was added to the solution. The UV-Vis spectrum changed slightly (from 2irnax 412 and 516 nm to 414 and 508 nm) suggesting that ligand exchange had occurred. The reaction was also monitored by !H-NMR spectroscopy in D 2 0 , the f3-pyrrole region changing from 8 8.75 (s, 4H), 8.60 (s, 4H) to 8.50 (s, 8H), and the SPh region changing from 8 8.3 (m, 8H), 8.15 (d, 8H) to 8.30 (d, 4H), 8.15 (m, 12H). This chemistry, however, was not pursued further. 3.3 Results and Discussion 3.3.1 Tetraarylporphyrin Syntheses Adler conditions The syntheses of the precursor porphyrins in this chapter were performed by refluxing a solution of one or two aromatic aldehydes and pyrrole in propionic acid (Adler Method). The synthesis giving the highest overall yield (-16.5 %) was the mixed aldehyde condensation (Section 3.2.2.1). 2-TImP and 4-TImP were produced in lower yield (< 4 %), perhaps because of the decreased reactivity of imidazolecarboxaldehyde (compared to benzaldehyde), and their isolation was often problematic. In mixed aldehyde condensations involving imidazolecarboxaldehyde and pyridinecarboxaldehyde (Section 3.2.2.5), the product mixtures were difficult to purify by standard chromatographic methods. In retrospect, a mixed-aldehyde condensation with benzaldehyde, imidazolecarboxaldehyde and pyrrole might have been easier to work with. Along these lines, 5-(2-imidazolyl)-10,15,20-triphenylporphyrin was reported recently.30 Lindsey conditions When Lindsey conditions were applied to the synthesis of TPyP using TCA, TFA, or BF3»MeOH catalysis, TPyP was produced when aliquots of the reaction mixture were treated with TCQ, which showed that the porphyrinogen was forming. However, the porphyrinogen was References on page 138 119 eventually consumed by a side-reaction which produced a very polar, blue compound (Section 3.2.2.2). The rate of the side-reaction appeared to increase with increasing acid strength and concentration; with TsOH, DCA or propionic acid as catalyst, this side-reaction did not occur, but no TPyP formed either. A similar blue product was observed in the synthesis of another meso-pyridylporphyrin (Section 2.2.3.6). It has been noted that under Lindsey conditions, numerous heterocyclic aldehydes fail to give porphyrins but no explanation was suggested.1 N-substitution on the pyridine nitrogen (as in N-oxides, and alkyl or acyl pyridinium salts) greatiy activates the pyridine ring to nucleophilic addition at the a and y positions.31 Thus, in situ formation of a pyridinium salt via protonation of (or BF3 coordination to) the pyridine lone-pair under high [H+] and subsequent reaction (at the pyridine ring and aldehyde substituent) with pyrrole could explain the results obtained here. The side-reaction did not occur using weaker acids, perhaps because the pyridine was 'insufficiently' protonated. However, the weaker acids were not able to catalyze the formation of a porphyrinogen, and thus no TPyP was observed when TCQ was added. Lindsey-type conditions were tried for the syntheses of 2- or 4-TImP, but no porphyrin was produced (Sections 3.2.2.3 and 3.2.2.5). 2-TImP Nitration attempts Imidazole itself is easily nitrated at the 4(5)-position at room temperature in cone. HNO3 containing 1 % oleum.31 Thus, a direct route to nitroimidazole-substituted porphyrins appeared to be nitration of 2-TImP. However, in the presence of strong acids, the imidazole nitrogens and inner pyrrolic nitrogens of 2-TImP would be protonated, giving a 6+ charge for the entire molecule. Perhaps nitration, an electrophilic aromatic substitution, failed because the product of 2-UmP in acid is a highly electron-deficient species. No reaction was observed here using mild nitrating conditions based on those reported for phenyl substituted porphyrins.4 More vigorous and/or less acidic conditions based on in-situ formation of trifluoroacetyl nitrate18 bleached the porphyrin before a single product was achieved. As these routes were not particularly clean or useful, nitration of 2-TImP was not pursued further. References on page 138 120 3.3.2 (l-Oxido-4-pyridyI)porphyrins and Porphyrin-N-oxides As described in Chapter 1, porphyrins incorporating heterocyclic N-oxides may have potential as anti-cancer agents. Pyridine-N-oxide was easily "incorporated" via N-oxidation of the corresponding pyridyl groups (Section 3.2.3) of the porphyrins from the phenyl-pyridyl series (Section 3.2.2.1). (Oxido-pyridyl)porphyrins have not been reported in the literature. However, a related compound (BzOPyP(Bz)4), was obtained by Milgrom et al. from the aerobic oxidation of a (l-benzyl-4-pyridyl)porphyrin and this procedure led to further benzylation of the other macrocyclic nitrogens.32 o BzOPyP(Bz)4 OEP-0 Porphyrin N-oxides have appeared more frequentiy in the literature. Octaethylporphyrin N-oxide (OEP-O) was first prepared by Bonnet et al. by treating OEP with hypofluorous acid (in 64 % yield) or a peroxyacid (26 % using m-CPBA, to 68 % using permaleic acid).33'34 Also, in studies investigating the active site of cytochrome P-450 enzymes, metal complexes of OEP-0 (Ni, Mn, Co, Cu), 3 5" 3 8 an Fe tetramesitylporphyrin N-oxide complex (TMPFe-O),39"44 and a Ti tetraphenylporphyrin N-oxide complex (TPhPTi-O)45 have been reported. With OEP-O, the metal was introduced after N-oxidation, while TMPFe-0 was produced by treating TMPFe with m-CPBA, and TPhPTi-O was prepared by photolysis of peroxotitanium(rV)-tetraphenylporphyrin. TMP-0 was produced by oxidation of TMP or demetallation of TMPFe-0 with HCl-HOAc, 3 9 and TPhPTi-O spontaneously demetallated in an ethanolic solution to give TPhP-O.45 In the metal complexes, the oxygen remains on one face of the porphyrin but, in the free-base porphyrin N-References on page 138 121 oxides, the oxygen rapidly moves through the center of the porphyrin in a fluxional porphyrin inversion process.35 TMPFe-0 TPhPTi-0 3.3.2.1 Synthesis and General Observations The N-oxidations were sluggish at 0 °C, but proceeded readily at room temperature. The results of the syntheses depended on the amount and rate of addition of m-CPBA. If m-CPBA was added in small aliquots, only the (oxidopyridyl)porphyrins were obtained. However, if a large excess of m-CPBA was added all at once, the corresponding porphyrin N-oxides were obtained in low but variable yield (i.e. the pyridyl nitrogens were also oxidized). Using these same conditions with TPhP, no significant amount of TPhP-oxide was produced, and continued addition of m-CPBA only led to the degradation of the porphyrin. As an oxidopyridyl group is likely more electron-withdrawing than an unsubstituted phenyl group, the oxidopyridylporphyrins are more resistant to oxidative degradation, and thus isolable amounts of the (oxidopyridyl)porphyrin N-oxides were obtained. Acid catalyzed condensation of pyrrole and 4-pyridinecarboxaldehyde N-oxide did produce a porphyrin (Section 3.2.3.8). There were numerous side-products in the reaction (typical of porphyrin syntheses) and the product was not isolated, nor was it conclusively identified as TOPyP. TOPyP was much more conveniently prepared in large quantities by N-oxidation of TPyP. Recently, attempts have been made to synthesize the (oxidopyridyl)porphyrins in this series by acid-catalyzed condensation of pyrrole, benzaldehyde and 4-pyridinecarboxaldehyde N-References on page 138 122 oxide, but a significant percentage of the the oxidopyridyl groups were deoxygenated in refluxing propionic acid.46 Thermal deoxygenation is commonly observed with heterocyclic N-oxides,22 but no deoxygenation was observed in !H-NMR studies when a solid sample of OPyTrPhP was heated at 100 °C overnight in vacuo. Porphyrin N-oxides deoxygenate at much milder conditions,34 and were thus kept from heat extremes (i.e. below 30 °C) in this work. However, deoxygenation was observed in the mass spectra of the (oxidopyridyl)porphyrins and their N-oxides. The elemental analyses of these compounds, and the presence of H 2 0 peaks in their !H-NMR spectra were consistent with water being present, not surprising given that pyridine N-oxide is hygroscopic and freely soluble in water.22 The porphyin N-oxides demonstrated much greater solubility in weakly polar solvents (CH2CI2, C H C I 3 ) than their parent oxidopyridyl compounds, and were even soluble in MeOH. With the exception of OPyTrPhP which had good solubility in C H 2 C I 2 or C H C I 3 , the parent porphyrins were only slightly soluble in C H 2 C I 2 or C H C I 3 , and as the number of oxidopyridyl groups increased, increasing amounts of MeOH were required as a co-solvent to insure dissolution. For example, TOPyP was insoluble in C D C I 3 but dissolution (required for ^H-NMR measurement) occurred readily on addition of a few drops of C D 3 O D . The difference in solubility may be related to less aggregation in the porphyrin N-oxides because of steric effects of the macrocycle oxygens; also, these oxygens may well be involved in hydrogen bonding. 3.3.2.2 X-ray Crystallography Crystals of OPyTrPhP were obtained by slow solvent evaporation from a solution of the porphyrin in CH2Cl2:MeOH, and the X-ray analysis yielded the structure shown in Figure 3.2. The morphology of the porphyrin core and the orientation of the aryl groups do not differ significantly from those of TPhP.47 The porphyrin ring is nearly planar, and the dihedral angle of the oxidopyridyl group and the porphyrin plane is 78°. The N(2)-0(l) bond length (1.299(7) A) is significantly shorter than that in pyridine N-oxide (1.34 A), but lies within the range of N-O distances in a variety of substituted pyridine N-oxides.22'35 References on page 138 123 Figure 3.2 X-ray Crystal Structure of OPyTrPhP. 3.3.2.3 i H - N M R Spectra Oxidopyridyl groups iH-NMR spectroscopy was very useful in the characterization of the (oxidopyridyl)-porphyrins and enabled the porphyrin N-oxide isomers to be identified. Compared to the unoxidized pyridyl group, the proton signals of the 2,6-oxidopyridyl groups were shifted upfield by -0.4 ppm, while the 3,5-oxidopyridyl protons were shifted < 0.1 ppm (see Figure 3.3). Similar shift changes have been observed for pyridine N-oxide22 and have been rationalized in terms of additional electron density at the 2,4 and 6 positions resulting from mesomeric structures (Figure 3.4). The splitting patterns and approximate 8 values for the phenyl signals of the (oxidopyridyl)porphyrins correspond well to those of the parent compounds.4 References on page 138 124 > I I I • I I I I I I I I I I I I 9 4 9 2 9 0 8 .8 8 .6 8 .4 8 .2 8 .0 7.8 7.6 7 .4 PPM Figure 3.3. The Aromatic Region of the !H-NMR Spectra TrPhPyP and OPyTrPhP in CDCI3. • • 11 •• 11 o - o - o o o Figure 3.4. Mesomeric Forms of Pyridine-N-oxide. Pyrrole-N-oxides The P-pyrrole protons of the pyrrole-N-oxide of the porphyrin-N-oxides are shifted significantly upfield to 8 -7.5 (compared to those of the non-oxidized pyrrole rings, Figure 3.5, A -1.4 ppm). Similar 8 values were noted for the P-pyrrole-oxide positions in TMP-0 (8 7.49)39 and TPhPTi-0 (7.26).45 To explain this effect, a number of possiblities must be considered. A simple electron withdrawing effect by the oxygen group can be discounted as this give a downfteld References on page 138 125 shift for the P-pyrrole signals. Partial negative charges for mesomeric forms (cf. Figure 3.4) are a possibility, but these would likely include the whole macrocycle and such a local effect probably would not be seen. An out-of-plane tilt of the pyrrole N-oxide group could explain the upfield shift, as the P-pyrrole protons would then experience less of a deshielding effect from the induced ring current. However, in the solid state, the pyrrole-N-oxide ring in OEP-0 is tipped out of the porphyrin plane only 6.1°,36 which is similar to the tilt (6.6°) found in two of the pyrrole rings of TPhP.4 7 If the pyrrole N-oxide ring maintains such a small tilt in solution, such a large A8 is unlikely. However, in the dynamic porphyrin inversion process, in which the oxygen atom moves through the center of the porphyrin,35 the pyrrole groups may be tilted far enough for this loss of deshielding to occur. In support of this conclusion, reports in the literature describe a similar loss of deshielding for the methylene groups in the Ni(U) complex of OEP-O where a large pyrrole N-oxide tilt (38.3) was observed and inversion did not occur.35,36 A large tilt of the pyrrole N-oxide unit is also expected in TPhPTi-O where a AS of ~2 was observed.45 As in the corresponding pyridyl-phenyl- and (oxidopyridyl)-phenyl-porphyrin series, the P-pyrrole protons when situated between two identical mesosubstituents show a singlet, and two doublets when between two different meso-substituents.4 The rest of the P-pyrrole region is complex because of the inequivalence introduced by the proximity of the oxidopyrrole group and different mesosubsitutents. The inner N-H protons of the porphyrin N-oxides are assigned to the broad singlets at 8 -1.6 which disappeared when D 2 0 was added to the iH-NMR sample. Presumably, the induced ring current is slightly disrupted because of the out-of-plane pyrrole-N-oxide ring. The inner N-H signals of OEP-0 and TMP-0 appear at 8 0.835 and 8 1.75,39 respectively. References on page 138 126 p-pyrrole-NO p-pyrrole p-pyrrole-NO I I I I I I I 1 I I I | I I I | I | I | I | I 9 4 9 . 2 9 . 0 8 . 8 8 . 6 8 . 4 8 . 2 8 . 0 7 . 8 7 . 6 7 . 4 P P M Figure 3.5. The Aromatic Region of the ! H-NMR Spectra of OPyTrPhP-230 and -210. In the tH-NMR spectrum of TOPyP in CDCI3/CD3OD, the p%pyrrole protons appeared as broad singlets and no pyrrole N-H signals were present. This is a general phenomenon, and similar spectral results were obtained when D2O or CD3OD was added to the CDCI3 solutions of TrPhPyP, OPyTrPhP, ?-DPBPyP, f-BOPyDPhP, c-DPhBPyP, or c-BOPyDPhP. These results can be explained by 2H-exchange with the inner pyrrole N-H protons; the ^ -pyrrole protons are weakly coupled to the pyrrole N-H protons but rapid tautomerism prevents any observable sphtting at room temperature. With 2 H , the exchange is slow (kn/kD = 12.1 at 35 °C), 4 8 and coupling to the D-atoms broadens the (3-pyrrole signals. References on page 138 127 3.3.2.4 UV-Vis Spectroscopy Pyridine N-oxide displays a strong absorbance maximum (log 8 > 4) in the 255-285 nm region and a weaker (log £ ~2) absorbance maximum near 350 nm; in polar solvents, the longer wavelength band is blue-shifted and can be hidden by the shorter wavlength absorption maximum.22 The shorter wavelength absorption maximum appears at -275 nm in the spectra of the (oxidopyridyl)porphyrins and the corresponding porphyrin N-oxides (see Table 3.5); also, the corresponding longer wavelength band is sometimes observed in the -310 to 350 nm region. However, the "background" absorbance from the porphyrin itself is fairly high in this region, and the less intense bands are difficult to pick out (see Figures 3.6 and 3.7). With the exception of the UV region, the spectra of the (oxidopyridyl)porphyrins are similar to those of free-base tetraarylporphyrins. The presence of the Soret bands in the spectra of the porphyrin N-oxides suggests that the macrocycle's conjugation is uninterrupted;34 however, the Soret bands are broadened and the Q bands are not as intense as those in the parent (oxidopyridyl)porphyrins (Figure 3.8). The Soret band and Q-bands in the UV-Vis spectra of OPyTrPhPZn-210 and -230 are similarly broad and diminished in intensity, but the peaks at 320-330 nm are more prominent (see Figure 3.9). A peak in the UV-Vis spectrum solely due to an electronic transition of the pyrrole N-oxide group of the porphyrin N-oxides has not been described in the literature, and is not evident here. References on page 138 128 Wavelength (nm) Figure 3.6. UV-Vis Spectrum of TPhPyP (2.3 x l O 6 M) in CH2CI2. D < < 1 . 0 0 - /I 4 2 0 0.75 - 2 7 4 / 5 1 6 0 . 5 0 - A / \ 5 5 2 0.25 - 1 A \ \ } \ 5 9 0 I / W \ 6 4 6 0 . 0 0 - 1 1 1 1 o o O o o o 10 o o o o Wavelength (nm) Figure 3.7. UV-Vis Spectrum of OPyTrPhP (2.6 x l O 6 M) in CH 2C1 2 . References on page 138 129 < 1 1 I.OO ^ 0.75 H 0.50 0.25 H 0.00 Wavenlength (nm) Figure 3.8. UV-Vis Spectrum of OPyTrPhP-230 (-7.6 x 10"6 M) in CH2CI2. (At -2.8 x 10"5 M, an additional, small peak is observed at 518 nm). Wavelength (nm) Figure 3.9 UV-Vis Spectrum of OPyTrPhPZn-230 (-1.0 x 10"5 M) in CH 2C1 2 . References on page 138 130 3.3.2.5 Infrared Spectroscopy The N-oxide moiety exhibits a intense, characteristic band in the 1200-1350 cm - 1 region which is attributed to an N-0 stretch,22 and the IR spectra of the (oxidopyridyl)porphyrins and their corresponding N-oxides show an intense band in the 1238 to 1261 cm"1 region due to the pyridyl N-oxide stretch (see Tables 3.7 and 3.8). Porphyrin N-oxide stretches for OEP-0 and TMP-O appear at 126534 and 127339 cm - 1, respectively. Presumably the bands due to the porphyrin N-oxide stretches in the compounds synthesized here appear in a similar region and are masked by the pyridine N-oxide stretches or vice versa. The other spectral bands are not assigned, but assignments of the common peaks of the porphyrin ring structure can be found in the literature.49 3.3.2.6 Mass Spectrometry The oxidopyridyl groups were deoxygenated using the usual ionization techniques (EI, LSEVIS), and no molecular ion peak was seen. It was necessary to resort to MALDI-TOF mass spectrometry (using a dihydroxybenzoic acid matrix) to see the parent peaks (Figure 3.10). However, some of the parent peaks of the porphyrin N-oxides were observed under EI ionization conditions (see Figure 3.11), and HR-MS analyses were obtained for these species. References on page 138 131 0.018 H 0.014 'in §5 0.012 .> 0.010 4—» CO cu o.ooa H 0.002 o.ooo M +1 6 4 9 .7 M + 1 -0 6 3 3 M + 1 -20 o-N + O -dimer 12 8 1 .3 Figure 3.10. MALDI-TOF Mass Spectrum of *-BOPyDPhP. 100 90 >N 80^ w 70 § 60^ -c5 50^ | 40-3 a> 30^ * 20 10^ 0 M + - 2 0 615 Relative M + - 2 H - 2 0 +Cu M + - 0 631 I ntensity x 2 0 677 M" 633 600 '20 A i'I'I i Hi', I' M + -2H-0+Cu 692 \ 4^41 650 m/z 700 Figure 3.11. Low Resolution Mass Spectrum (EI) of OPyTrPhP-210. References on page 138 132 3.3.3 Sulfonation Reactions Sulfonation of the phenyl groups of TPhP, TrPhPyP, OPyTrPhP and c-BOPyDPhP was achieved readily in hot cone. H2SO4. Like the pyridyl subsituent, the oxidopyridyl group was not sulfonated under these conditions and the N-oxide was retained. The pyridine nitrogen (or oxygen in the oxidopyridyl group) is protonated under the sulfonation conditions producing a positive charge which makes the group unreactive towards electrophilic addition. When the sulfonated porphyrins were isolated by solvent removal from an aqueous solution, they retained several mole equiv. of water as judged by elemental analysis. This trend in the sulfonated porphyrins has been noted previously by our group;5 less water was retained when the sulfonated porphyrins were precipitated from MeOH with acetone. As noted in Section 3.2.4.4, c-BOPyBSPhP was judged by ^ - N M R to be impure, and the impurity was not removed to a significant degree by chromatography or attempts at recrystallization. The precursor porphyrin c-BOPyDPhP contained an impurity (by dc), which was probably another porphyrin obtained from the oxidation reaction of c-DPhBPyP (Section 3.2.3.4). However, a suitable elemental analysis was obtained for the formulation c-BOPyBSPhP»4.5 H2O, indicating that the impurity had a composition similar to that of c-BOPyDSPhP. 3.3.4 Platination Reactions Water-soluble platinum porphyrins have appeared only recently in the literature. Pasternack etal. described the interaction of DNA with [5,10,15,20-tetrakis(N-methylpyridinium) porphyrinato]Pt(II) (TMPyPPt).50 In other work, our group has reported the synthesis of cationic, water-soluble Pt-porphyrins containing nitroaromatic groups and described their evaluated in vitro anti-cancer efficacy.6,51 Only one report of an anionic, water-soluble porphyrin (TSPhPPt) has appeared, but it was characterized only by UV-Vis spectroscopy.24 As part of the early experimental work for this thesis, the syntheses of TSPhPPt and PyTrSPhPPt were investigated. The platination of TSPhP was easily achieved by heating an aqueous, buffered solution of TSPhP and K^PtCLj; without the buffer, the highly acidic References on page 138 133 environment produced by the hydrolysis of K^PtCLi52 seemed to decrease the reaction rate. When similar reaction conditions were used with PyTrSPhP, some platination occurred when no buffer was used, and rapid decomposition was observed when the buffer was added! The presence of the pyridyl group clearly interferes with the platination reaction. The pyridyl groups on porphyrins have been used recently as ligands in coordination complexes of Pd and Pt 5 3 ' 5 4 and this is likely the case here. Other attempts to platinate PyTrSPhP were unsuccessful (Section 3.2.5.2). The N-oxide group in OPyTrSPhP was considered a potential "protecting group" for the pyridine which would allow the platination of the porphyrin to proceed. Indeed, the platination was achieved with conditions similar to those used to produce TSPhPPt. However, once the product was isolated, the characterization data showed the compound to be impure or to be of a different identity. The !H-NMR integrations did not match, no N-oxide band was observed in the generally broadened UV-Vis spectrum, and formulas with only added water did not agree with the elemental analysis. A formula that does match the results well includes the externally coordinated Pt moiety PtCl3Na, but it is impossible to identify conclusively the compound from elemental analysis alone. Some evidence for Pt coordination to the N-oxide was found in the IR spectrum: the N-oxide stretch in "OPyTrSPhPPt" was decreased in intensity compared to that in OPyTrSPhP, and appeared as a shoulder (-1210) on the peak at 1181 cm - 1, which is consistent with Pt-O-pyridine coordination.55,56 3.3.5 Tirapazamine Chemistry Tirapazamine and several of its derivatives have been investigated as anti-microbial agents and hypoxia selective cytotoxins.26'57"60 Thus, incorporation of tirapazamine into a porphyrin appeared to be a possible route to an effective anti-cancer agent (Chapter 1). As tirapazamine was not successfully coupled to protoporphyrin IX via peptide-based chemistry, SN2, or reductive amination reactions (Chapter 4), the reactive 21 species appeared to be a plausible reagent for forming a porphyrin-tirapazamine conjugate. Seng and Ley 2 5 synthesized 21 from phosgene and tirapazamine in a 4 h reaction. However, phosgene is a highly toxic gas and is difficult to handle, and so its use was circumvented by using triphosgene, a crystalline compound introduced by References on page 138 134 Eckert and Foster in 1987.61 Using Seng and Ley's conditions but using triphosgene in place of phosgene, the reaction was surprisingly rapid and produced 21 in high yield (-95%), but the product contained a small amount of tirapazamine. o . O OCNI o - o — c = o Tirapazamine 21 The formula for 21 was postulated by Seng and Ley, presumably because C=0 (1814 cnr1) and C=N (1541 cm-1) stretches were observed in the IR spectrum (see Table 3.14); bands typical of isocyanates (2273-2000 cm - 1) 6 2 were not observed. Despite its cyclic structure, 22 reacts like an isocyanate and produces ureas and carbamides when mixed with amines and alcohols, respectively, undergoing a ring-opening process.25,26 Also typical of isocyanates, 21 reacted with water (even on the dc plate) to regenerate tirapazamine. Reactivity of 21 The reactivity of 21 was investigated with the intent of employing similar conditions with porphyrins containing carboxylic acid, alcohol, and amino functional groups . When 21, containing a small amount of tirapazamine, was mixed with CH3COOH in pyridine at 60-80 °C, amide 22 was produced. As isocyanates are unreactive towards carboxylic acids, 22 was probably produced by a reaction of tirapazamine with CH3COOH; the H2O thus produced could then be consumed by 21, to produce more tirapazamine (Scheme 3.1). H 2 O + 2 1 * ~ tirapazamine ?- / ?-N+ y ^ k + ,| J + C H 3 C O O H S » rX J 9 O - o - H tirapazamine 22 Scheme 3.1. The Reaction of Tirapazamine with CH3COOH in the Presence of 21. References on page 138 135 21 reacted smoothly with butanol in pyridine at 90 °C to produce the novel butyl carbamide 23 (Section 3.2.6.3). Seng et al. obtained 25 via its piperidinium salt from a reaction of 21 and piperidine, but few characterization data were reported.26 Similar results were obtained here when 21 was mixed with piperidine, but the product was not isolated (Section 3.2.6.4). The reaction of 21 with en proceeded rapidly. A product other than the expected 24 was obtained but its identity was not established. O . 23 o. 25 24 Reactions of 21 with porphyrins APhTrPhP reacted readily with 21 to produce TirapPhTrPhP in -50 % isolated yield. IR peaks due to N-oxide stretches were present that were not observed in aminophenyl porphyrins.63 Peaks attributed to tirapazamine appeared in the UV-Vis spectrum, but these were found in an area of relatively high "background" absorbance from the porphyrin core (Section 3.2.6.5). TirapPhTrPhP was analyzed by cyclic voltammetry (see Chapter 5) and evaluated in vitro (see Chapter 6). TirapPhTrPhP BHEtPIXDME References on page 138 136 No reaction was observed when a solution (pyridine or DMF) of 21 and protoporphyrin DC (containing free COOH groups) was refluxed. BHEtPIXDME contains primary alcohols, but its reaction with 21 was sluggish. Only a small amount of a new product was produced from this reaction with prolonged refluxing of the reaction solution (see Section 3.2.6.6). In principle, a large excess of 21 could be mixed with BHEtPIXDME to obtain enough porphyrin carbamate for in vitro evaluation, but tirapazamine, the precursor for 21, is currently difficult to obtain (Section 7.3.4). 3.4 Summary Several tetraarylporphyrins (except TPhP) containing pyridyl or imidazolyl groups were synthesized by acid-catalyzed condensation of pyrrole with one or two aldehydes in propionic acid (the Adler method). The phenyl-pyridyl series was easiest to work with, whereas porphyrins containing imidazolyl groups were produced in low yield and were difficult to purify. Nitration of imidazolyl porphyrins using several conditions were unsuccessful. Attempts to form porphyrins from heterocyclic, aromatic aldehydes via Lindsey conditions met with little success, and a possible explanation is presented. Porphyrins containing one to four pyridyl groups were 'N-oxidized' with m-CPBA to produce five novel (oxidopyridyl)porphyrins and seven porphyrin N-oxides (including three from the c-BOPyBPhP-O mixture). Only three free-base pophryin N-oxides have been reported in the literature (OEP-O, TMP-0 and TPhP-O), and thus the number of known porphyrin N-oxides has been tripled. The (oxidopyridyl)porphyrins and porphyrin N-oxides were well characterized and an X-ray crystal structure of OPyTrPhP was obtained. Sulfonation of TPhP, TrPhPyP, OPyTrPhP and c-BOPyDPhP was achieved readily under standard condtions with no loss of oxygen from the oxidopyridyl groups. TSPhP reacted with K^PtCU to produce TSPhPPt, but similar conditions with PyTrSPhP met with little success. OPyTrSPhP was metallated with K^PtCLi, but the resulting product evidently had an externally bound Pt moiety, perhaps ligated through the oxidopyridyl group. References on page 138 137 The product 21 from tirapazamine and triphosgene reacts like an isocyanate, and was used in reactions with carboxylic acids, alcohols, and amines to produce compounds (some new) incorporating tirapazamine. In the reactions with porphyrins bearing these functional groups, the best results were obtained using APhTrPhP to produce the novel TirapPhTrPhP. References on page 138 138 References for Chapter 3 (1) Lindsey, J. S. In Metalloporphyrins Catalyzed Oxidations; F. Montanari and L. Casella, Eds.; Kluwer Academic Publishers: Dortdrecht, 1994; pp. 49-86. (2) Wallace, D. M . ; Leung, S. H.; Senge, M . O.; Smith, K. M . J. Org. Chem. 1993, 58, 7245-7257. (3) Wijesekera, T. P.; Dolphin, D. In Metalloporphyrins in Catalytic Oxidations; R. A. Sheldon, Ed.; Marcel Dekker: New York, 1994; pp. 193-240. (4) Meng, G. G.; James, B. R.; Skov, K. A. Can. J. Chem. 1994, 72, 1894-1909. (5) Meng, G. G.; James, B. R.; Skov, K. A.; Korbelik, M. Can. J. Chem. 1994, 72, 2447-2457. (6) James, B. R.; Meng, G. G.; Posakony, J. J.; Ravensbergen, J. A.; Ware, C. J.; Skov, K. A. Metal-Based Drugs 1996, 3, 85-89. (7) Totter, J. R.; Darby, W. J. Org. Synth. Coll. Vol. Ill 1955, 460-462. (8) Griffith, R. K.; DiPeitro, R. A. Synthesis 1983, 576. (9) Grimmet, M . R.; Richards, E. L. J. Chem. Soc. 1965, 3751-3754. (10) Battersby, A. R.; Nicoletti, M. ; Staunton, J.; Vleggaar, R. J. Chem. Soc, Perkin Trans. 1 1980, 43-51. (11) Attenburrow, J.; Cameron, A. F. B.; Chapman, J. H.; Evans, R. H.; Hems, B. A.; Jansen, A. B. A.; Walker, T. J. Chem. Soc. 1952, 1904. (12) Mancera, O.; Rosenkranz, G.; Sondheimer, F. J. Chem. Soc. 1953, 2189-2191. (13) Little, R. G.; Anton, J. A.; Loach, P. A.; Ibers, J. A. J. Heterocycl. Chem. 1975,12, 343. (14) Lindsey, J. S.; Wagner, J. W. J. Org. Chem. 1989, 54, 828-836. (15) Gossauer, A.; Engel, J. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. 2; pp. 197-253. (16) Milgrom, L. R.; Bone, S.; Bruce, D. W.; Macdonald, M . P. J. Mol. Elec. 1991, 7, 95-100. 139 (17) Gunter, M . J.; Robinson, B. C. Aust. J. Chem. 1990, 43, 1839-1860. (18) Crivello, J. V. J. Org. Chem. 1981, 46, 3056-3060. (19) Huang, S.; Ding, L.; Wang, X.; Li , G. Youji Huaxue 1987, 3, 189-192; CA 107: 167640. (20) Grimmet, M . R.; Richards, E. L. J. Chromatog. 1965,18, 605-608. (21) Grimmet, M . R.; Richards, E. L. J. Chromatog. 1965, 20, 171-173. (22) Albini, A.; Pietra, S. Heterocyclic N-oxides; CRC Press: Boca Raton, 1991, pp. 7-25. (23) Eigendorf, G., University of British Columbia, Personal Communication, 1997. (24) Blinova, I. A.; Vasil'ev, V. V.; Shagisultanova, G. A. Russ. J. Inorg. Chem. 1994, 39, 253-257. (25) Seng, F.; Ley, K. Angew. Chem. Int. Ed. Eng. 1972,11, 1009-1010. Mason, J. C ; Tennant, G. J. Chem. Soc. (B) 1970, 911-916. (26) Seng, F.; Ley, K.; Metzger, K. G. U.S. Patent 3 957 779, 1976. (27) Neunhoeffer, H.; Wiley, P. F. Chemistry of 1,2,3-Triazines and 1,2,4-Triazines, Tetrazines, and Pentazines; John Wiley & Sons: New York, 1978; Vol. 33, pp. 1335. (28) Buehler, J. W.; Kiong-Lam, L.; Stoppa, H. Z. Naturforsch. 1980, 35 b, 433-438. (29) Ware, C. M.Sc. Thesis, University of British Columbia, 1994. (30) Zhu, H.-J.; Wang, X.-Y.; L i , G.-N. Youji Huaxue 1991,11, 103-105; CA 114: 239273. (31) Joule, J. A.; Mills, K.; Smith, G. F. Heterocyclic Chemistry; Chapman & Hall: London, 1995, pp. 374. (32) Milgrom, L. R.; Hill, J. P.; Dempsey, P. J. F. Tetrahedron 1994, 47, 13477-13484. (33) Bonnet, R.; Ridge, R. J. Chem. Soc. Chem. Commun. 1978, 310-311. (34) Andrews, L. E.; Bonnet, R.; Ridge, R.; Appelman, E. H. J. Chem. Soc. Perkin Trans. I 1983,1983, 103-107. (35) Balch, A. L.; Y.-W., C ; Olmstead, M . ; Renner, M . W. J. Am. Chem. Soc. 1985,107, 2393-2398; and references therein. 140 (36) Balch, A. L.; Chan, Y.-W.; Olmstead, M. M . J. Am. Chem. Soc. 1985,107, 6510-6514. (37) Balch, A. L.; Chan, Y. W. Inorg. Chim. Acta 1986,115, L45-L46. (38) Arasasingham, R. D.; Balch, A. L.; Olmstead, M. M. ; Renner, M . W. Inorg. Chem. 1987,26, 3562-3568. (39) Groves, J. T.; Watanabe, Y. J. Am. Chem. Soc. 1986,108, 7836-7837. (40) Groves, J. T.; Watanabe, Y. J. Am. Chem. Soc. 1988,110, 8443-8452. (41) Watanabe, Y.; Tekehira, K.; Shimizu, M. ; Hayakawa, T.; Orita, H.; Kaise, M . J. Chem. Soc. Chem. Commun. 1990, 1262-1264. (42) Tsurumaki, H.; Watanabe, Y.; Morishima, I. J. Am. Chem. Soc. 1993,115, 11784-11788. (43) Mizutani, Y.; Watanabe, Y. ; Kitagawa, T. J. Am. Chem. Soc. 1994,116, 3439-3441. (44) Rachlewicz, K.; Latos-Grazynski, L. Inorg. Chem. 1996, 35, 1136-1147. (45) Hoshino, M . ; Yamamoto, K.; Lillis, J. P.; Chijimatsu, T.; Uzawa, J. Inorg. Chem. 1993, 32, 5002-5003. (46) Pratt, R. C. University of British Columbia, Personal Communication, 1997. (47) Silvers, S. J.; Tulinsky, A. J. Am. Chem. Soc. 1967, 89, 3331-3337. (48) Scheer, H.; Katz, H. In Porphyrins and Metalloporphyrins; K. M . Smith, Ed.; Elsevier: Amsterdam, 1975; pp. 399-524. Storm, C. B.; Teklu, Y. J. Am. Chem. Soc. 1972, 94, 1745-1747. (49) Alben, J. O. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. Il l ; pp. 323-346. (50) Pasternack, R. F.; Brigandi, R. A.; Abrams, M . J.; Williams, A. P.; Gibbs, E. J. Inorg. Chem. 1990, 29, 4483-4486. (51) Ravensbergen, J. R. M.Sc. Thesis, University of British Columbia, 1993. (52) Sanders, C. I.; Martin, D. S. J. J. Am. Chem. Soc. 1961, 83, 807-810. (53) Drain, C. M. ; Lehn, J.-M. J. Chem. Soc, Chem. Commun. 1994, 2313-2315. (54) Yuan, Ff.; Thomas, L.; Woo, L. K. Inorg. Chem. 1996, 35, 2808-2817. 141 (55) Garvey, R. G.; Nelson, J. H.; Ragsdale, R. O. Coord. Chem. Rev. 1968, 3, 375-407. (56) Orchin, M. ; Schmidt, P. J. Coord. Chem. Rev. 1968, 3, 345-373. (57) Seng, F.; Ley, K.; Metzger, K. G. U.S. Patent 4 027 022, 1977. (58) Seng, F.; Ley, K.; Hamburger, B.; Franz, B. U.S. Patent 3 991 189, 1976. (59) Zeman, E. M . ; Baker, M . A.; Lemmon, M . J.; Pearson, C. I.; Adams, J. A.; Brown, J. M . ; Lee, W. W.; Tracy, M . Int. J. Radiation Oncology Biol. Phys. 1989,16, 977-981. (60) Minchinton, A. I.; Lemmon, M . J.; Tracy, M. ; Pollart, D. J.; Martinez, A. P.; Tosto, L. M. ; Brown, J. M . Int. J. Radiation Oncology Biol. Phys. 1992, 22, 701-705. (61) Eckert, H.; Foster, B. Angew. Chem., Int. Ed. Engl. 1987, 26, 894. (62) Silverstein, R. M. ; Bassler, G. C ; Morrill, T. C. Spectrometry Identification of Organic Compounds; John Wiley & Sons: New York, 1991, pp. 419. (63) Meng, G. G. Ph.D. Thesis, University of British Columbia, 1993. (64) Smith, K. M . In Porphyrins and Metalloporphyrins; K. M. Smith, Ed.; Elsevier: Amsterdam, 1975; pp. 3-28. 142 Chapter 4 Substituent Manipulation on Protoporphyrin IX 4.1 Introduction Because they are ubiquitous in living organisms, naturally occurring porphyrins have been widely used as starting materials for the synthesis of new compounds. For example, about 100 mg of the Fe-complex of protoporphryin IX (PLX), hemin, can be isolated from the hemoglobin in 100 mL of blood. Hemin is also the prosthetic group of myoglobin, a number of enzymes including many peroxidases, some cytochromes, catalase, tryptophan pyrrolase, and many other proteins.1 The literature describing the synthesis and derivitization of naturally occurring porphyrins is quite extensive and will not be reviewed here. Related to the development of anti-cancer agents, some derivatives of PIX, notably hematoporphyrin (HP), Photofrin®, and a Benzoporphyrin Derivative (BPD), have been investigated as photodynamic therapy agents in cancer therapy.2'3 PLX was the starting material for one porphyrin incorporating a nitroimidazole,4 but no other examples of porphyrins incorporating radiosensitizers or hypoxia-selective cytotoxins were found. In this thesis work, PIXDME was viewed as a suitable precursor to porphyrins incorporating nitroimidazoles, heterocyclic N-oxides and other 'useful' substituents (see Chapter 1). PIXDME is commercially available and there is a substantial database describing the interconversion of functional groups of PIXDME.1-5 The majority of the chemistry in this chapter describes the derivatization of the substituents in the 8,13-positions of PIXDME via the product of Tlin-oxidation of the vinyl groups. The derivatization at these positions was investigated as the 2,18-propionic ester groups could be subsequently cleaved to improve the solubility of the compounds for the in vitro tests at BCCRC. Novel porphyrins were synthesized via reductive amination chemistry, a Knoevenagel reaction, SN2 reactions, peptide chemistry, and general porphyrin synthetic procedures described in the literature. References on page 187 143 4.2 Experimental A general description of the methods and materials used here can be found in Section 2.2.1. PIXDME was obtained from Dr. D. Dolphin's group. C2F5CH2NH2*HC1 was a generous gift from Dr. C. Koch, University of Pennsylvania. The iH-NMR, UV-Vis, mass spectrometry data, and elemental analyses of the major products described here, are summarized in Tables 4.1-4.5, which can be found after Section 4.2.15. Other characterization data are reported in the individual sections. The synthesis pathways and conditions used here are oudined in Scheme 4.1. 4.2.1 Dimethyl 8,13-Bis(2,2-dimethoxyethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate ( B D M E t P I X D M E ) According to the conditions described by Kenner et al.,6 PIXDME (0.71 g, 1.2 mmol) was dissolved in CH2Cl2:MeOH (210 mL, 6:1), and the solution was stirred and refluxed. A solution of T1(N03)3*3H20 (1.76 g, 4.0 mmol, 3.3 equiv.) in MeOH (-10 mL) was added and the mixture was refluxed for 10 min. The heat was removed, the solution was purged with S0 2 for 5 min, cone. HC1 (3 mL) was added, and the solution was stirred for 5 min. Then CH2CI2 (250 mL) was added, and the mixture extracted with distilled H2O (3 x 300 mL) in order to keep the water-soluble Tl salts separate for later disposal with the UBC Waste Management Facility. The organic phase was dried and the solvent was removed under reduced pressure. The residue was flash chromatographed on silica/(CH2Ci2:THF 100:5) to yield 0.75 g (87 %) of crude BDMEtPLXBDME which was pure enough for subsequent reactions. Purer samples were obtained by subsequent chromatography on silica/(CH2Cl2:MeOH 50:1) or by recrystallization. The tH-NMR data matched those reported in the literature.6 Equally suitable results were obtained using Tl(OOCCF3)3 in place of T1(N03)3»3H20. References on page 187 144 PIXDME 1 » BDMEtPIXDME " ( ° r m ) » BOEtPIXDME v i > BAnEtPIXDME BHEtPIX iv (or v) BHEtPIXDME Vll l BF 5EtPIXDME BPrAPIXDME xvi BBrEtPIXDME BTsEtPIXDME BPrEPIXDME IX Xll XVll BNImEtPIXDME Xll l BPhEtPIXDME BIEtPIXDME BNImEtPIX XIV BNIMEtPIXBNA-diTiu ester Scheme 4.1. General Synthesis Pathways for the PLX-based Porphyrin Series. Reaction Conditions: (With the exceptions of procedures i i , i i i , x, xii, and xiii, subsequent chromatography was required) (i) 1) T1(N03)3 (3.3 equiv.), 2) SO2, HC1; (ii) TFA, 5 min; (iii) THF/H20/HCl/heat; (iv) NaBFLi; (v) 1) NaBFLi, 2) MeOH:H 2 S0 4 (19:1); (vi) excess aniline, HC1, NaBH 3CN, (vii) excess C2F 5CH2NH2«HC1, N M M or Et3N, NaBH 3 CN; (viii) malonic acid, piperidine, pyridine, heat; (ix) CH2N2; (x) 25 % HCl( a q); (xi) SOBr2; (xii) excess 2-nitroimidazole, iPr2EtN; (xiii) 25 % HCl( a q); (xiv) 1) STMl>BF4, Et3N, 2) ditBu-aspartate; (xv) excess K»phthalimide, Et3N; (xvi) excess TsCl, pyridine; (xvii) excess Nal, heat. References on page 187 145 4.2.2 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-oxoethyl)-porphyrin-2,18-dipropionate (BOEtPIXDME) Deprotection using T F A In a typical reaction that followed a literature procedure for dimethylacetal deprotection,7 BDMEtPIXDME (50 mg, 0.07 mmol) was dissolved in TFA (~2 mL) and the solution was stirred for 5 min. The solution was carefully added to dilute NaHC03(aq) and the product was extracted into CH2CI2 (50-100 mL). The solvent was dried and subsequently removed under reduced pressure. The residue was redissolved in CH2CI2, the mixture was filtered and the solvent was then removed from the filtrate. The resulting BOEtPIXDME was used without further purification in subsequent reactions. A sample for characterization was prepared by eluting a CH2CI2 solution of this porphyrin down a short AI2O3CV) column. Rf = 0.55 (silica/(GH2Cl2:MeOH 50:1)); IR (thin film, cm"1) 3400 w, 3310 m, 2951 s, 2918 s, 2584 m, 1733 vs, 1438 s, 1164 s, 1107 s, 734 s, 676 m. The !H-NMR data (in acetone-^, Table 4.1) are similar to those reported in the literature (measured in CDCI3).8 The parent peak in the mass spectrum (M + , 622; Table 4.5) did not match that reported by Kahl et al. (M + , 582) but is consistent with the formula reported in their elemental analysis.9 The UV-Vis and IR data were not reported previously. BOEtPIXDME is a precursor to BHEtPIXDME (Section 4.2.5) and was used in reductive amination reactions (Section 4.2.3). Deprotection using HC1 According to the procedure described by Kenner et al.,6 BDMEtPIXDME (300 mg, 0.42 mmol) was dissolved in THF:H 20 (100 mL:3 mL) and the solution was heated to reflux. Then cone. HC1 (1.5 mL) was added, the mixture was refluxed for 5 min and poured into water (150 mL), and the product was then extracted with CH2CI2. The solvent was dried in the usual manner and removed under reduced pressure. The bis(oxoethyl) product was used in subsequent reactions (Sections 4.2.5 and 4.2.13.1) without further purification. No characterization data have been reported for BOEtPIXDME prepared in this manner. References on page 187 146 These 'HC1 conditions' tended to cleave the ester functionalities to give carboxylic acids. Attempts to purify the product by chromatography on Al203(V) were unsuccessful as it stuck to the stationary phase, consistent with observations made by Kenner et al. In an attempt to re-esterify the isolated product, CH2N2 was added to a solution of the porphyrin in THF. Tic analysis on silica/(CH2Cl2:MeOH 50:1) of this reaction showed near quantitative conversion to a new product, which was slightly more polar than BDMEtPIXDME. Attempts to purify the product by column chromatography led to decomposition or transformation into other products; iH-NMR spectroscopy of the 'purified' product suggested a product mixture and the mass spectrum (LSIMS, EI) showed numerous species, -14 mass units apart, with 'parent' peaks of 697 and 622 for FAB and EI, respectively. Kahl et al.9 performed a similar CH2N2 reaction with BOEtPIXDME (obtained by treating BDMEtPIXDME with formic acid, which left the esters intact) and generated a mixture of ketones and epoxides (see Section 4.3.1.1). Other deprotection attempts Before the TFA deprotection was found, several other selective deprotection methods were investigated. Transacetalation of BDMEtPIXDME with TFA was studied by iH-NMR in acetone-d$, and was partially successful. Over a few days, a -CHO signal (5 10.27) increased in intensity at the expense of the -C//-(OCH3)2 proton signal (5 5.12), but as the reaction neared completion (-1 week), noticeable decomposition had occurred. When transacetylation was attempted in refluxing acetone, the iH-NMR spectrum showed that significant decomposition had occurred. Treating BDMEtPIXDME with PPTS in acetone10 was less effective than with TFA in acetone. No reaction was observed (by tic) when BDMEtPIXDME was mixed with wet Si02 in CH2CI2 with or without oxalic acid,11 or Amberlite IR-120 in acetone/water.12 Other literature conditions employing SnCl2»2H20 in CH2CI2 1 3 or L 1 B F 4 in MeCN 1 4 were also unsuccessful. 4.2.3 Reductive Amination Chemistry Based on general reductive amination conditions reported in the literature,15 a solution of BOEtPIXDME (usually from TFA deprotection of BDMEtPIXDME) and an amine (1.2 to 12 equiv. per aldehyde) in CH2Cl2:MeOH (1:1 or 1:2) was adjusted with 5N HC1 in MeOH (or with References on page 187 147 N-methylmorpholine, Section 4.2.3.2) to pH ~7. To this solution was added NaBFljCN (a few mg) and the mixture was stirred. Reduction of the imine, formed in-situ from the amine and BOEtPIXDME, proceeded rapidly. However, rapid reduction of unreacted aldehydes in BOEtPIXDME also occurred. The products were purified by chromatography. 4.2.3.1 Dimethyl 8,13-Bis(2-(N-anilino)ethyl)-3,7,12,17-tetramethylporphyrin-2.1 H-dipropionate (BAnEtPIXDME) Via T F A deprotection of B D M E t P I X D M E BDMEtPIXDME (50 mg, 0.07 mmol) was deprotected with TFA as described in Section 4.2.2. The resulting BOEtPIXDME was dissolved in dry MeOH (10 mL) and dry CH2CI2 (5 mL). To this solution was added aniline (75 |iL, 0.85 mmol), 5 N HC1 in MeOH (55 j_tL, 0.28 mmol), and finally NaBH3CN (~2 mg), and the solution was stirred overnight. The product mixture was washed with dilute NaHC03(aq), extracted with CH2CI2 and subsequently chromatographed by preparative-tic on silica/(CHCl3/MeOH 200:1). The main product band (Rf = 0.8) yielded 0.015 g (27 %) of BAnEtPIXDME. Another, more polar product was present, but its identity was not pursued. Overnight stirring was then found to be unnecessary, as the reduction of the imine and aldehyde (see also Section 4.2.3.2) occurred rapidly. Via HC1 deprotection of B D M E t P I X D M E Only traces of BAnEtPIXDME were isolated from a mixture of BOEtPIXDME (formed via HC1 deprotection of BDMEtPIXDME), aniline (~8x excess per aldehyde) and NaHC03 in refluxing benzene:CHCl3 (1:1) with subsequent NaBfLi-reduction. The yield of BAnEtPIXDME improved when a reesterification step (dissolution in Me0H:H2S04 19:1 and stirring for 4 h, or treatment of the residue with CH2N2) was included after NaBELi reduction and filtering off the NaHC03. Attempts to isolate the imine from the reaction mixture before NaBFLt was added were unusuccessful. References on page 187 148 4.2.3.2 Dimethyl 8,13-Bis(2-(N-3,3,3,2,2-pentafluoropropylamino)ethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate ( B F 5 E t P I X D M E ) BDMEtPIXDME (100 mg, 0.14 mmol) was deprotected with TFA (Section 4.2.2) and the resulting BOEtPIXDME was dissolved in dry MeOH (7 mL) and dry CH2CI2 (7 mL). To this solution was added C2F5CH2NH2»HC1 (65 mg, 0.33 mmol), N-methylmorpholine (25 uL, 0.23 mmol), and finally NaBHsCN (2-3 mg). A new product was present (Rf = 0.65; silica/(CH2Cl2:MeOH 200:1)) when the reaction progress was checked several hours later, but no further reaction was observed by tic when the mixture was stirred overnight. The product mixture was washed with dilute NaHC03(aq) and the residue chromatographed by preparative-tic on sihca/(CH2Cl2:MeOH 200:1). The main product band yielded 10 mg (31 %) of BF5EtPIXDME. 19F-NMR (188.3 MHz, CDC13) 5 -7.33 (s), -44.69 (s). A second, minor product also had resonances in the 1 9F-NMR spectrum, but its identity was not investigated further. Attempts to carry out this reaction in THF gave poor yields of BFsEtPrXDME, possibly because of the poor solubility of C2F5CH2NH2*HC1 in THF. When the reaction was carried out in MeOH alone, only traces of BFsEtPIXDME were produced. Optimization attempts and general observations During attempts to optimize this reaction using different solvent mixtures (e.g. CH2CI2, MeOH, THF), BOEtPLXDME was found to be reduced to BHEtPIXDME by NaBH 3CN (with or without C2F5CH2NH2#HC1) even in the presence of a large excess of base. The sensitivity of BOEtPIXDME is unusual, as aldehydes are normally not so easily reduced by NaBH3CN (see Section 4.3.1.2). In these optimization reactions, if the addition of NaBH3CN was delayed by ~1 h, BF5EtPLXDME was detected by tic, while no such product was detected in reactions to which 3A molecular sieves were added (see below). BOEtPIXDME had poor solubility in MeOH. However, when C2F5CH2NH2«HC1 was added, the porphyrin dissolved readily and UV-Vis spectroscopy indicated predominant formation of the porphyrin dication. ^ F s C ^ M ^ ' H C l alone in MeOH yielded a weakly acidic solution (pH 4-6), and subsequent addition of 3A molecular sieves to this mixture made the solution more References on page 187 149 basic (pH 8-10)). Attempts were made to moderate the pH, as the imine formation rate is pH-dependent (Section 4.3.1.2): addition of E13N regenerated the free-base spectrum, but adding 3A molecular sieves gave a mixture of the dication and free-base. When imidazole was added with the molecular sieves to act as a buffer (pH ~7), no B F 5 E 1 P I X D M E was produced. Other attempts to synthesize B F s E t P I X D M E No B F 5 E 1 P I X D M E was formed when BBrEtPIXDME (Section 4.2.7) or BTsEtPEXDME (Section 4.2.8) was mixed with with 6x excess of C2F5CH2NH2»HC1 and -P^EtN in NMP. No reaction was observed (by tic) when C2F5CH2NH2 (isolated from its HC1 salt by treatment with dilute aqueous base, extraction into CH2CI2 and drying over MgS04) was mixed with BBrEtPIXDME or BTsEtPEXDME in NMP at 90 °C. 4.2.3.3 Attempted Reductive Amination with Tirapazamine to Produce 26 Via B O E t P I X D M E Attempts to apply the conditions used for BAnEtPIXDME (Section 4.2.3.1) to synthesize the PIXDME-tirapazamine conjugate (26) by using tirapazamine in place of aniline were unsuccessful. Traces of BHEtPIXDME and a product that was consistent with 3-amino-1,2,4-benzotriazine-l(or 4-)-oxide by tH-NMR and mass spectrometry16 was isolated from the reaction mixture. References on page 187 150 Via B B r E t P I X D M E BBrEtPIXDME (Section 4.2.7) was mixed with tirapazamine in NMP and the mixture was heated to 80 °C. Tic did not show any new porphyrinic compound, but one new product might be 3-amino-l,2,4-benzotriazine-l(or 4-)-oxide.16 Using jPr2EtN, a non-nucleophilic base, as in the synthesis of BNImEtPIXDME (Section 4.2.10), might have improved the result, but this chemistry was not pursued further. 4.2.3.4 Attempted Reductive Amination with N H 4 O A C to Produce 27 Based on chemistry with tirapazamine (Section 3.2.6), 27 looked to be a useful precursor to a porphyrin-(tirapazamine)2 conjugate. Based on the procedure described in Section 4.2.3.1, BOEtPIXDME, prepared from BDMEtPIXDME (50 mg, 0.07 mmol), was dissolved in MeOH:CH2Cl2 (10 ml: 5 mL). To this solution was added NH4OAC (0.13 g, 1.7 mmol) and the mixture was stirred for 1 h. When NaBH3CN (~1 mg) was added to an aliquot of the reaction mixture, tic on silica//CH2Cl2:MeOH:Et3N 50:1:1) showed the presence of a new product which was more polar than BHEtPIXDME. More NH4OAC (0.43 g, 5.6 mmol) was added to try to shift the aldehyde/imine equilibrium in favor of the imine (see Section 4.3.1.2), and a porphyrinic precipitate formed. This was filtered off, washed with MeOH and analyzed by iH-NMR, which in CDCI3 or DMSO-J6 showed no porphyrin signals. UV-Vis spectroscopy showed a broad Soret and weak Q bands (CH2C12 : 394, 502, 534, 570, 622), but dissolution of the precipitate in CH2Cl2:MeOH (10:1) or THF and subsequent solvent removal showed no 27 in the resulting residue by mass spectrometry. References on page 187 151 With use of the same procedure, but without the second addition of NH4OAC, NaBFJ^CN was added just as a precipitation began. Purification of the product by preparative-tic on Al203/(CH2Cl2:MeOH:Et3N 100:1:1) yielded two products, neither of which gave conclusive evidence (iH-NMR, tic, or MS) for 27 (see Section 4.3.1.2), and this chemistry was not pursued further. 4.2.3.5 Attempted Reductive Amination with en to Produce 28 Attempts were made to synthesize 28, a potentially good Pt-chelator, via reductive amination. Using the conditions outlined in Section 4.2.3.1, BOEtPIXDME, produced from BDMEtPIXDME (100 mg, 0.14 mmol), was mixed with en (0.112 mL, 1.7 mmol) and 5N HC1 in MeOH (220 p:L, 1.1 mmol) in dry THF (10-15 mL). The mixture was stirred for 20 min; NaBH3CN (0.015 g) was then added, and the mixture was stirred for 0.5 h. A tic control on silica/(CH2Cl2:MeOH 50:1) showed no BOEtPIXDME, and most of the product remained at the baseline. The product was washed with dilute NaHC03 (aq) and extracted into CH2Ci2. The product was too polar to be effectively chromatographed on silica, but better results were obtained with chromatography on Al203/(CH2Cl2:MeOH 50:1). The signals in the iH-NMR spectrum were generally broad (see data) and no molecular ion peak could be detected by mass spectrometry (EI or +LSIMS). In the iH-NMR (200 MHz, acetone-d6) the integrations did not match [8 10.1 (s br), 4.3 (s br), 3.8 (s), 3.6 (s br), 3.5 (s), 3.3 (s br), 2.9 (s br), 1.3 (s), 0.9 (s), -3.9 (s); UV-Vis (CH2CI2) 400, 502, 536, 572, 624 nm]. The polarity of the compound is perhaps similar to that H3COOC COOCH3 28 References on page 187 152 expected for 28, but the characterization data are lacking. Similar results were obtained when BBrEtPIXDME was mixed with en. This chemistry was not pursued further. 4.2.4 Dimethyl 3 , 7 , 1 2 , 1 7 - T e t r a m e t h y l p o r p h y r i n - 8 , 1 3 - b i s ( £ ; / Z 2-propenoic acid)-2,18-dipropionate (BPrAPIXDME) and Tetramethyl 3,7,12,17-Tetramethylporphyrin-2,18-dipropionate-8,13-bis(i?/Z 2-propenoate) ( B P r E P I X D M E ) Based on conditions used with 8,13-diformyl-PIX,17 BOEtPIXDME (6 mg, 0.01 mmol, formed via TFA deprotection, Section 4.2.2), malonic acid (100 mg, 1 mmol) and piperidine (-50 |j,L, -0.5 mmol) were dissolved in dry pyridine (2 mL) and the mixture was heated to 95 °C and stirred. After 25 min, tic on silca/(CH2Cl2:MeOH 50:1) showed near quantitative conversion to a new product (Rf = 0.15) (for BOEtPIXDME, Rf = 0.55). The heat was removed, and the solvent was removed under reduced pressure to yield a sticky residue, which was dissolved in CH2CI2 (20 mL) and washed with distilled H2O; MeOH was added to the organic phase, the mixture was filtered, and the product (in the filtrate) was chromatographed by preparative-tic on sihca/(CH2Cl2:MeOH 50:1) to give BPrAPIXDME, presumably as a mixture of E and Z isomers. Because of the small amounts of materials involved, the yield was not determined. IR (KBr pellet, cm-1) 3461 m, 3917 w, 1732 s, 1609 w, 1581 w, 1436 w, 1383 w, 1199 m, 1156 m, 1023 m. The product was soluble in dilute NaOH(aq), and in CH2Cl2:MeOH (10:1). The !H-NMR spectrum (DMSO-cfo) of this product showed vinyl resonances at 5 8.2 and 7.3, but cleaner spectra were obtained when the product was treated with CH2N2, and rechromatographed to produce the tetramethyl ester BPrEPIXDME (see Table 4.1). Poorer results were obtained when the reaction was scaled up, as the reaction did not go to completion and other products were formed. 4.2.5 Dimethyl 8,13-Bis(2-hydroxyethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BHEtPIXDME) Via T F A deprotection BOEtPLXDME (0.1 g, 0.14 mmol), produced by treating BDMEtPIXDME with TFA (Section 4.2.2), was dissolved in CH2Cl2:MeOH (80 mL:20 mL), and to the solution was added References on page 187 153 NaBH4 (0.3 g) dissolved in MeOH (20 mL). Tic on silica/(CH2Cl2:MeOH 50:1) indicated that the reaction was complete after 10 min. Acetic acid was then added added dropwise until effervescence ceased, and the product was pre-dried on a few grams of AI2O3 and chromatographed on Al203(in)/(CH2Cl2:MeOH 50:1) to yield 66 mg (75 %) of BHEtPIXDME. The tH-NMR and other characterization data are similar to those reported by Carr et a/.18 The other characterization data match those reported by Carr et a/.18 The chromatography step could be eliminated and the crude product treated directiy with SOBr2 to produce BBrEtPIXDME (Section 4.2.7). Via HC1 deprotection These conditions were based the original preparation of BHEtPIXDME from BOEtPIXDME described by Kenner et al.6 BDMEtPIXDME (0.3 g, 0.42 mmol) was deprotected with HC1 (Section 4.2.2). The product was dissolved in CH 2 Cl 2 :MeOH (125 mL, 4:1) and to the solution was added NaBEL} (1 g) in MeOH (10 mL). The carboxylic acid groups were re-esterified by stirring the product in MeOH:H2S04 (75 mL, 19:1) for 5 h. After the acid was neutralized (using NH4OH) and the product was extracted into CH 2C1 2, the ester was chromatographed on Al203(V)/(CH2Cl2:MeOH 100:1) to yield 0.18 g (70 %) of BHEtPIXDME. The characterization data were identical to those obtained via the TFA deprotection method described above. 4.2.6 8,13-Bis(2-hydroxyethyl)-3,7,12,17-tetramethyl-porphyrin-2,18-dipropionic acid (BHEtPIX) A sample of BHEtPIXDME was dissolved in 25% HCl(aq) (75 mL), and the solution was stirred in the dark for 4 h at room temperature and then neutralized with NaOH (aq) and NaHC03. The solution was brought to pH 11-12 using NaOH and then gradually acidified to pH 4.6 using dilute HCl(aq) when the porphyrin precipitated. BHEtPIX was filtered off and rinsed with distilled H 2 0 (5 mL). The synthesis of BHEtPIX has been described previously, but few characterization data were given.1 9'2 0 References on page 187 154 4.2.7 Dimethyl 8,13-Bis(2-bromoethyl)-3,7,12,17-tetramethyIporphyrin-2,18-dipropionate (BBrEtPIXDME) BDMEtPIXDME (2.0 g, 2.8 mmol)) was treated with TFA and subsequently NaBFL* to produce crude BHEtPIXDME (Section 4.2.5). Based on the conditions described by Kenner et al.,6 the product was dissolved in CH2CI2 (350 mL) and DMF (20 mL), and to this solution was added anhydrous K2CO3 (25 g) and SOBr2 (8.6 mL). The mixture was stirred at room temperature in the dark for 3 h and poured into aqueous NaHC03/K2C03, and the product was then extracted into CH2CI2. The solvent was removed under reduced pressure, and the residue was chromatographed twice on Al203(III)/CH2Cl2 and subsequently precipitated from CH2CI2 (50 mL) with MeOH:H 20 (250 mL 21:4). The product was filtered off to yield 1.27 g (60%) of BBrEtPIXDME as a red-purple solid, which was 95-99% pure by UV-Vis spectroscopy6 and was used without further purification. The !H-NMR, UV-Vis and mass spectrometry data were identical to those reported by Kenner et al. Alternatively, BBrEtPIXDME was prepared in -75 % yield from purified BHEtPIXDME using similar conditions. 4.2.8 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-(4-toluenesulfonyl)oxoethyl)-porphyrin-2,18-dipropionate ( B T s E t P I X D M E ) Based on procedures used for the tosylation of alcohols on porphyrins,21 BHEtPIXDME (25 mg, 0.04 mmol), TsCl (0.15 g, 0.8 mmol) and dry pyridine (1 mL) were dissolved in dry CH2CI2 (25 mL), and the mixture was stirred in the dark at room temperature overnight. Tic analysis on Al203/(CH2Cl2 100:1) indicated two new products: Rf = 0.95 (the required BTs product) and Rf = 0.7 (presumably a mono tosylate product). More TsCl (25 mg) was then added and the mixture was stirred for another 48 h, when tic indicated only the BTs product. The solvent was removed under reduced pressure, and the product was partially purified by chromatography on Al203(ni)/(CH2C12:THF 20:1). The product smelled slightly of pyridine and TsCl was detectable by tic; no yield was determined because of these impurities. Attempts to purify the product further by chromatography on Al203(V)/CH2Cl2 led to two new, less-polar products. Their !H-NMR spectra were very similar to that of BTsEtPIXDME with the exception of the Ts References on page 187 155 region, which indicated loss of one and two Ts groups for the second least and least polar products, respectively. Mass spectrometry indicated that the new products were the Cl-ethyl / Ts-ethyl (+LSIMS, m/z = 799, 100, M + 1 ) and bis(Cl-ethyl) (EI, m/z = 662, 100, M+) species. The source of the CI that displaced the Ts group is not obvious (see Section 4.3.2.2). Crude BTsEtPIXDME was used without further purification in subsequent reactions (Sections 4.2.10 and 4.2.3.2). 4.2.9 Dimethyl 8,13-Bis(2-iodoethyl)-3,7,12,17-tetramethylporphyrin-2,18-dipropionate (BIEtPIXDME) Conversion of the TsEt group to the IEt group were based on Tovey's reaction conditions.21 BTsEtPIXDME was prepared from BHEtPIXDME (0.025 g, 0.04 mmol) (Section 4.2.8) without chromatographic purification; the solvent was removed and the product was dissolved in MeCN (10 mL). To this solution was added Nal (0.12 g, 0.8 mmol) and the mixture was refluxed for 5 h. Tic analysis on silica/(CH2Cl2:MeOH 100:1) indicated high yield conversion to BIEtPIXDME (Rf = 0.38). Analagous to the tosylation of BHEtPIXDME, an intermediate polarity porphyrin (Rf = 0.28) was detected that was consistent with it being a monoiodo-monotosylate species. The solvent was removed and the crude product was washed with MeOH/H20 (1:1, 50 mL); the yield was not determined. A portion of the product was filtered through a short plug of silica with CH2Cl2/MeOH (100:1) and was analyzed by iH-NMR, UV-Vis and mass spectrometry. BIEtPIXDME was used without purification in attempts to synthesize BNImEtPIXDME (Section 4.2.10). 4.2.10 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yl)ethyl)-porphryin-2,18-dipropionate (BNImEtPIXDME) BBrEtPIXDME (0.80 g, 1.1 mmol, Section 4.2.7), 2-nitroimidazole (1.2 g, 11 mmol), and ip^EtN (0.400 mL) were dissolved in NMP (40 mL) and the mixture was heated to -100 °C with stirring. To monitor the reaction, the solvent was removed from an aliquot of the reaction mixture and the residue was analyzed by tic on silica/(CH2Cl2-.MeOH -50:1). After 4 h, two new products (Rf = 0.63, Rf = 0.75) and a small amount of a product with the same Rf (0.95) as References on page 187 156 BBrEtPIXDME were present. No further reaction was observed when additional ip^EtN was added and the mixture was heated for another 8 h. The reaction mixture was cooled, poured into distilled H2O (100 mL) and the product was filtered off. Purification was achieved by partially dissolving the product mixture in hot THF (100 mL) and hot-filtering. BNImEtPIXDME was poorly soluble and remained on the filter while the other products were found in the filtrate. During each hot filtration, some BNImEtPIXDME was lost in the filtrate, but was recovered by reducing the volume of the THF and hot-filtering again. Yield was 0.39 g (44 %). A sample for characterization was prepared by hot filtering from THF (2x). Because BNImEtPIXDME exhibited poor solubility in CDCI3 or acetone-^, TFA was added to the samples for NMR measurements (see Table 4.1). IR (KBr pellet, cm"1) 1735 s (COOMe), 1655 w, 1530 m, 1482 s (N02-imidazolyl), 1402 m, 1357 s (iV02-imidazolyl), 1271 w, 1163 w, 837 w. DMF was also a suitable solvent for this reaction, but no BNImEtPIXDME was formed when the reaction was carried out in CH2Cl2/MeOH or THF/MeOH. The reaction did not proceed without ip^EtN, but a large molar excess of ip^EtN did not give good yields (by tic). Chromatographic purification of BNImEtPIXDME was possible on a small scale (-25 mg), but was impractical on a larger scale because of its poor solubility. The side-product (Rf = 0.75) was isolated from the filtrate of the THF wash by preparative-tic on silica/(CH2Cl2:MeOH 100:1) and was determined to be a mixture of two isomers, each with one vinyl and one (2-(2-nitroimidazol-l-yl)ethyl) group (29 and 30). Compounds 29 and 30 in solution reacted in the presence of air and light to produce green, more polar products according to tic, similar to the chemistry seen with PIXDME.1 References on page 187 157 0 2 N-<' J Other Attempts Before the conditions described above were found, other conditions for the introduction of 2-nitroimidazole to the porphyrin core using BTsEtPIXDME and BIEtPIXDME were investigated. Imidazole alkylations reported in the literature have been carried out in DMF at 100 °C in the presence of K2CO3. 2 2 However, when such conditions were used with crude BTsEtPIXDME or BEtPIXDME, elimination, which regenerated the vinyl groups (as evidenced by tH-NMR, MS), and decomposition were observed; very little, if any, BNImEtPIXDME was produced. Switching to CSCO3 and lowering the temperature (50 °C) did not significantly improve the results. BTsEtPIXDME and BIEtPIXDME were not pursued further as starting materials, as they took longer to prepare than BBrEtPIXDME and were sometimes difficult to purify. Based on the literature preparation of a Cs-imidazole adduct,23 a "Cs-2-nitroimidazole adduct" was prepared by mixing 2-nitroimidazole with 1 equiv. CsOH in water and removing the solvent. Traces of BNImEtPIXDME were isolated when BBrEtPIXDME was mixed with the "Cs-2-nitroimidazole adduct" in NMP and the mixture was heated to -50 °C. This route was not pursued further. 4.2.11 3,7,12,17-tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yl)ethyl)-porphyrin-2,18-dipropionic acid (BNImEtPIX) BNImEtPIX (150 mg, 0.18 mmol) was produced from BNImPIXDME in the same manner that BHEtPIX was formed from its diester (Section 4.2.6 ). However, when the solution was neutralized with NaOH(aq) and NaHC03, the porphyrin precipitated and was filtered off. The References on page 187 158 product was rinsed through the filter with distilled H2O (see Section 4.3.2.3) when some some porphyrinic material remained on the filter. The filtrate (~pH 8) was brought to ~pH 11 with NaOH(aq.) and the porphyrin was precipitated by gradual acidification to pH 4.6. BNImEtPIX was filtered off, rinsed with distilled H2O and then MeOH. (In solutions of high ionic strength, such as after acid neutralization, the porphyrin is insoluble even as its 2Na+ salt. The 2Na+ salt was soluble in distilled water but, when in the COOH form, the porphyrin was insoluble in water and MeOH.) Yield was 0.108 g (73 %) of BNImEtPIX* IH2O. When the UV-Vis spectrum was obtained in 0.05 M NaOH, two Soret peaks were seen, possibly resulting from aggregation; a single Soret peak was observed with 10% aqueous Scintigest as solvent (see Sections 4.3.2.3 and 6.2.1). Poor elemental analyses (low C, H, and N) were obtained when dialysis (using 1000 MWCO tubing) was used to remove excess salts after the initial neutralization of the acid. 4.2.12 Bis(N-L-Aspartyl) 3,7,12,17-Tetramethyl-8,13-bis(2-(2-nitroimidazol-l-yI)ethyl)-porphyrin-2,18-dipropionamide (BNImEtPIXBNA) Because of the poor solubility of BNImEtPIX in media used in the biological in vitro experiments (see Chapter 6), the coupling of two aspartic acid groups to this compound was investigated. Similar chemistry has been used to synthesize the water-soluble photosensitizer MACE derived from Chlorophyll a.24 Formation of BNImEtPIXBNA-( t Bu) 2 ester Based on published peptide coupling techniques,25,26 BNImEtPIX (30 mg, 0.04 mmol) and Et3N (12 ^.L, 0.09 mmol) were dissolved in dry DMF (1 mL). To this solution was added STMU'BFa (a reagent used to produce the -NHS esters, 25 mg, 0.082 mmol); the dark solution rapidly turned hazy and was stirred in the dark for 20 h. Tic on silica/(CH2Cl2:MeOH 50:1) indicated high yield conversion to a new product. Et3N (15 |iL, 0.11 mmol) and di^u-aspartate^HCl (23 mg, 0.08 mmol) were added, and the mixture was stirred for 6 h in the dark; tic indicated one main product (Rf = 0.55) and other minor side-products. The solvent was removed under reduced pressure, and the product was chromatographed 2x by preparative-tic on silica/(CH2Cl2:MeOH 50:1) to give 24 mg (52 %) of BNImEtPIXBNA-(tBu)2 ester. !H-NMR References on page 187 159 (400 MHz, CDCI3) 8 (10.35 (s, 1H), 9.85 (s, 1H), 9.42 (s, 1H), 9.12 (s br, 1H, meso-/7), 8.00 (s, 1H), 7.38 (d, 1H), 7.23 (d, obscured by solvent peak), 7.16 (q, 4H), 6.66 (d, 2H), 6.23 (s, 1H), 5.86 (s, 1H), 4.76 (q, 2H), 4.47 (t, 6H), 4.03 (s br, 2H), 3.63 (s, 4H), 3.54 (s, 4H), (3.26 (m), 3.18 (s), 12 H, $-CH3), 2.93 (s, 2H), 2.87 (t poorly resolved, 6H), 2.78 (s, 2H), 2.33 (s, 4H), (1.28 (s, 9H), 1.22 (s, 9H), 1.03 (s, 9H), 0.96 (s, 9H), *Bu-CH3), -4.09 (s, 2H, pyrrole N-H) The iH-NMR spectra of these lBu-aspartyl derivatized porphyrins are often complex27'28 and only limited assignments are made here: however, the integrations for the pyrrole N-H, meso-H, lBu-CH3 and $-CH3 signals were correct and the spectrum looked clean. Cleavage of the tBu-ester BNImEtPIXBNA-(tBu)2 ester (20 mg, 0.016 mmol) was dissolved in TFA (3 mL); the mixture was stirred in the dark for 3 h, and the solvent was removed under reduced pressure. Analysis of this product by iH-NMR indicated that the lBu groups had been removed. The product was then dissolved in distilled H2O, and the solution was adjusted to pH 5 with dilute HCl(aq), and filtered. After more HCl(aq) was added to the filtrate, the porphyrin precipitated (at pH 2.8) and was filtered off to give 9 mg of a product which analyzed well (C, H) for BNImEtPIXBNA'HCl or «2H20 (-55 %). Some of the product was lost in the workup. LR-MS (+LSIMS) of this product did not show the parent peak. BNImEtPIXBNA had good solubility in an acetone:H20 (1:1) or in dilute aqueous base. Analysis by RPtlc on silica(RP-18)/(water:MeOH:MeCN:TFA 1.5:1:1:0.01) showed one main product and five minor products which might explain the somewhat low N in the elemental analysis (see Table 4.4). Better results might have been obtained if BNImEtPIXBNA-(tBu)2 had been more rigorously purified before dissolution in TFA. Because of the low overall yield from BNImEtPIX, this chemistry was not pursued further, and BNImEfPIXDNA was not evaluated at BCCRC. iH-NMR (200 MHz, DMSO-^/with traces of TFA) 8 10.3 (s), 10.2 (s), 10.1 (s br), 8.4 (s br), 7.3 (s), 6.9 (d), 5.1 (s br), (4.5 s v br), 4.3 (s v br), 4.0 (s v br), 3.7 (s br), 3.5 (s br), 3.1 (s br), -4.0 (s br); the relative integrations were not correct and thus assignments were not made. References on page 187 160 4.2.13 Dimethyl 3,7,12,17-Tetramethyl-8,13-bis((2-phthalimido)ethyl)-porphyrin 2,18-dipropionate ( B P l E t P I X D M E ) BPlEtPIXDME looked to be a good precursor to 27 (see Section 4.2.3.4), which could be used to couple tirapazamine to a porphyin via chemistry similar to that described in Section 3.2.6.5. However, attempts to cleave the phthalimido groups to generate the compound with 'free' -NH2 moieties met with limited success (see Section 4.2.14). Synthesis of B P l E t P I X D M E BBrEtPIXDME (0.10 g, 0.14 mmol) and potassium phthalimide (0.5 g, 2.7 mmol) were dissolved in dry DMSO (6 mL), and the mixture was heated and stirred for 2 h at 50 °C. Tic analysis on sihca/(CH2Cl2-.MeOH 100:0.5) indicated one main product (Rf = 0.6) and another less polar side-product (Rf = 0.8), similar to behavior seen in the synthesis of BNImEtPIXDME. The mixture was poured into dilute NFLjOAqaq), extracted into CH 2Cl2, and the solvent was then removed. The residue was dissolved in CH2CI2 (10-15 mL) and precipitated with MeOH:H20 (19:1) to help remove residual phthalimide; the porphyrinic material was filtered off and subsequendy purified by preparative-dc on silica/(CH2Cl2/MeOH 100:1). The main product was precipitated again with MeOH:H20 (19:1); the product was filtered off and washed from the filter with CH2CI2, and the solvent was removed to yield 42.5 mg (34%) of BPlEtPIXDME. The higher Rf band was determined to be a mixture of the 13-(2-(N-phthalimidoethyl))-8-vinyl- and 8-(2-(N-phthalimidoethyl))-13-vinyl porphyrins (31 and 32, respectively). References on page 187 161 4.2.14 Attempted Removal of Phthalimido Groups From BPlEtPIXDME to Produce 27 With H 2 N N H 2 » H 2 0 A common method for removal of the phthahmido group is via hydrazinolysis.29 Typically, the phthalimide is refluxed with ^ N N T ^ ' r ^ O in a suitable solvent (e.g. EtOH) and the amine salt of phthalhydrazide is produced. When BPlEtPIXDME was stirred at room temperature or heated with ^ N M - ^ ' ^ O m CH2Cl2/MeOH, the starting material was consumed (as evidenced by dc on silica/(CH2Cl2:MeOH 50:1)). A moderately polar intermediate was first formed and subsequently consumed to yield at least two very polar products (Rf ~ 0.05). Also, large amounts of a white precipitate formed, perhaps an alkylation product of hydrazine and CH2CI2. 3 0 The white solid was filtered off, the solution was washed with water and the solvent was removed. Mass spectrometry (+LS1MS) indicated the presence of 27 (C36H45N6O4:625 (7, M + 1 )) . Some of the mixture was soluble in DMSO-6% but no phthahmido groups were present in this component (by iH-NMR); however, the remainder of the product, soluble in CDCI3, still retained phthahmido groups. When a solution of the product mixture and 21 (a derivative of tirapazamine which reacts with amines to form urea moieties, Section 3.2.6.1) in DMF was heated and stirred, there was no obvious reaction (by tic). Either the resulting product was too polar to be analyzed by dc, or only small amounts of 27 were present to begin with. When the reaction was performed in MeOH, the white precipitate did not form and a similar low Rf product mixture was obtained that retained the phthalimido groups (by iH-NMR). Because the mixtures were too polar to be chromatographed and the desired product was not reliably formed, this chemistry was not pursued further. With 6N HCl(aq) Based on conditions described in the literature,29 BPlEtPIXDME (several mg) was added to 6N HC1 (aq) and the mixture was stirred for 5 h at room temperature. Some of the porphyrin dissolved, but much remained as a pink precipitate (probably a porphyrin dication salt) which was removed by filtration. The filtrate was neutralized (NaOH(aq)) and the resulting red-brown References on page 187 162 precipitate, that was filtered off, was insoluble in aqueous solution (pH 11.5 to 2) and had poor solubility in many solvents (CH2Cl2/MeOH, DMSO, DMF, pyridine). The !H-NMR spectrum of the product in DMSO-J5 showed a new peak at 8 1.8 that was coupled to signals in the methylene region, indicating the presence of the desired -CH2CH2NH2 group; however, the resonances in the phthalimide region remained. As esters are converted to carboxylic acids under these conditions (Section 4.2.11), the product was stirred in MeOH:H2SC«4 (25 mL, 19:1) for 4 h in attempts to resynthesize the esters (Section 4.2.5). However, according to !H-NMR spectroscopy and mass spectrometry, these conditions simply regenerated BPlEtPIXDME; presumably the phthalimido groups were not cleaved. This chemistry was not pursued further. With i P r N H 2 Based on conditions using a primary amine to cleave phthalimido groups,31 BPlEtPIXDME was taken as a suspension in EtOH/iPrNH2 (13:7), and the mixture was stirred for 3 days in the dark. The porphyrin product was in solution, and tic on silica/(CH2Cl2:MeOH 50:1) indicated near quantitative conversion to a more polar product which was isolated by preparative-tic. New resonances appeared in the aromatic and alkyl regions of the ^H-NMR spectrum but phthalimide resonances were still present. A 2D-1H-COSY spectrum did not help in product identification. Mass spectrometry (+LSIMS) showed a M + 1 peak at 1031, which differs from that of 27 ( M + 1 625). The identity of this compound and this chemistry were not pursued further. 4.2.15 Other Reactions 4.2.15.1 Oxidation of B O E t P I X D M E Attempts to oxidize the aldehyde groups of BOEtPIXDME to carboxylic acids were generally unsuccessful. When BOEtPIXDME was mixed with m-CPBA, bleaching of the porphyrin was observed, and with Mg monoperoxyphthalate»6H20 no reaction was observed (by tic). References on page 187 163 4.2.15.2 Reaction of P I X D M E with en to Produce 33 33 Synthesis of 33 Stein and Plane proposed 33 as the reaction product of PIXDME with en.3 2 According to the described conditions, PIXDME (50 mg, 0.08 mmol) was dissolved in en (3 mL) and the solution was refluxed for 1 h. The solvent was removed under reduced pressure to yield a very hygroscropic, water-soluble product, whose iH-NMR spectrum showed very broad peaks. The integrations were not quantitative, but it was clear that the vinyl groups were no longer present, consistent with the observations of Stein and Plane. A portion of the product was dialyzed in 1000 MWCO dialysis tubing in distilled water, dried at 100 °C in vacuo, and handled under dry N 2 . UV-Vis (H20) 372, 506, 538, 568, 620 nm; Analysis calc'd for C 4 2 H 6 2 N 1 2 0 2 : C, 65.77; H, 8.15; N, 21.91; found: C, 47.24; H, 6.70; N, 20.97. Calculating the elemental formula with several equiv. of H 2 0 does not give better agreement. No peaks were seen in the mass spectrum (+ or -LSIMS). When excess Zn(OAc)2 was added to an aqueous solution of this product, UV-Vis studies showed rapid metallation, the 'kmax data for the free-base and metallated product (in H 2 0 , 406, 502, 538, 574 nm) agreeing well with those reported.32 A repeat reaction was monitored by RPtlc on silica(RP-18)/(H20:MeCN:MeOH:TFA 10:10:10:0.1). After 30 min, three major products were isolated from the tic plate and analyzed by iH-NMR. The two lower Rf (less polar) product bands showed vinyl signals in their iH-NMR spectra, while the higher Rf (more polar) product had no vinyl groups, or signals at 8 7-8 seen in References on page 187 164 the final product (see below). As the reaction progressed, the lower Rf products diminished and the relative yield of the higher Rf product grew (RPtlc), which suggested that the lower Rf bands were intermediates. However, continued refluxing did not remove minor side-products and the final product was not pure by RPUc. Nevertheless, a sample of the mixture was purified by RPtlc, and the major product band was scraped from silica(RP-18) and the product eluted with H20:TFA (50:1). !H-NMR (300 MHz, DMSO-d6) integrations didn't match, 5 10.3 (s), 9.9 (s), 8.3 (s br), 7.9 (s br), 7.35 (s), 7.15 (s), 7.0 (s), 4.5 (s br), 4.4 (s br), 3.8 (s), 3.75 to 3.65 (four s), 3.5 (s), 3.3 (s), 3.25 (s), 3.15 (s), 2.85 (s), -3.9 (s) ; lU NMR (200 MHz, DMSO-dg/TFA) 5 10.45 (s br), 8.25 (s br), 7.95 (s), 7.80 (s br), 4.45 (s br), 3.7 (s br), 3.4 (s br), 3.25 (s br), 3.1 (s br), 2.8 (s br). The product gave no mass spectral peaks, and its identity was not pursued further (see Section 4.3.2.4). Reaction of 33 with K2PtCl4 When a sample of the 47.24 % C product was mixed with K2PCI4 in distilled H2O, a brown precipitate rapidly formed. The material had poor solubility in many solvents (H2O, MeOH, DMSO, DMF) and was not analyzed further. The precipitate did not form and no change was observed in the UV-Vis spectrum, if H Q or NH4OH was added to the solution before the K2PtCl4. This chemistry was not pursued further. 4.2.15.3 Attempted Synthesis of PIX- and PIXDME-Tirapazamine Conjugates 34 Attempts were made to prepare a PLX-tirapazamine conjugate (34) via an amide bond through the 2,18-propionic acid groups of PIX and the -NH2 group of tirapazamine. PIX was R R References on page 187 165 prepared according to the literature procedure by treating PIXDME with 2N KOHrTHF, neutralizing the solution with HC1, and filtering off the PIX. 3 3 Via the isobutylformate ester The conditions described here were based on similar chemistry described by Traylor et al. for coupling imidazole moieties to the 2,18-propionic acid group via an amide bond.34 Mixing PLX with 2.1 equiv. of isobutyl chlorofonnate (in place of the lBu-COCl used by Traylor et al.) in dry pyridine and, subsequently, tirapazamine did not give a PIX-tirapazamine conjugate. Rather, PLX-di('Bu)ester was isolated from the reaction, presumably produced via decarboxylation of the isobutylformate ester formed in-situ.. iH-NMR (200 MHz, acetone-d6) 5 ((10.04 (s), 10.03 (s), 3H), 9.97 (s, 1H) meso-//), 8.31 (m, 2H, -C/Y=CH2), 6.35 (dd, 2H, -CH=C/72), 6.15 (dd, 2H, -CH=C#2), 4.34 (t, 4H, porphyrin-C#2-), 3.81 (d, 4H, CH2), (3.61 (s), 3.59 (s), 3.55 (s), 12H (beta-Ctfj)), 1.74 (m, 2H, CH2-C/7-(CH3)2), 0.69 (d, 12H, CH 2-CH-(C// 5) 2), -4.0 (s, 2H, pyrrole N-H); UV-Vis (acetone) 404, 504, 538, 574, 630, 668 nm; LR-MS (+FAB) m/e 675 (100, M + 1 ) . This chemistry was not pursued further. Via the N-hydroxysuccinimidyl (NHS) ester The peptide coupling chemistry described in Section 4.2.12 was employed here. When PIX was mixed with STMU»BF4 (2.2 eq) in pyridine, DMSO or DMF (with 2.1 equiv. of triethylamine in DMF and DMSO) under anhydrous conditions, a new product appeared by tic, presumably the NHS ester. However, no further reaction was observed after tirapazamine was added (2.5 eq. per porphyrin) even with subsequent heating. 4.2.15.4 Attempted Synthesis of a P I X - E F 5 Conjugate o References on page 187 166 Attempts were made to synthesize a porphyrin-(EF5)2 conjugate (35). BBrEtPIXDME (several mg) was mixed with ip^EtN and an excess of EF5 in NMP, and the solution was heated to 80 °C. Numerous products (-20) were observed by tic, but the largest product bands did not exhibit 1 9 F-NMR signals. In another attempt based on conditions used to alkylate amides,35 EF5 was pre-treated with 1 equiv. KO'Bu in dry THF (to deprotonate the amide N-H); the solvent was removed and the residue was mixed with BBrEtPIXDME (0.6 equiv.) in DMSO. The major product from this reaction was PIXDME, formed via an elimination reaction, which was consistent with similar reactions described by Kenner et al.6 Two other, minor, more polar products also contained vinyl groups, and one did have signals in the 1 9F-NMR spectrum, but because of the low yields involved this chemistry was not pursued further. 4.2.15.5 Treatment of PIXDME with HBr and Imidazole Published hydrobromination procedures for derivitization of (3-vinyl chlorins36 were the basis for these conditions. PIXDME (17 mg, 0.03 mmol) was dissolved in HBr(45 %)/HOAc (2 mL); the mixture was stirred in the dark for 3 h under N2, and the solvent was removed under reduced pressure. The residue was dissolved in dry CH2CI2 (5 mL) and imidazole (43 mg, 0.6 mmol) was then added. After 1 h, tic showed at least 10 products. Mass spectrometry (+LSIMS) indicated that the mixture contained a small amount of a monoimidazolyl monovinyl species (m/e = 658 (5, M + 1 )) and large amounts of PIXDME. The mixture was purified by preparative-tic on silica/(CH2Cl2:MeOH 25:1). A few of the products were analyzed by iH-NMR which showed that elimination to regenerate the vinyl moiety had occurred. Repeating this procedure with 2-nitroimidazole gave similar results. These results are also similar to those seen with DPhDVPZn (Section 2.2.6.2), and again this chemistry was not pursued further. References on page 187 167 Table 4.1. iH-NMR Data for the R-PKDME Porphyrins3 Porphryin meso H 8,13-/? (or 8,13-CH 2 CH 2 f l )b 2,1S-CH2CH2- (& S,13-CH2CH2-)h -COOC//5, 3,7,12,17-C//j Pyrrole N-H BDMEtPIXDME 10.14 s(l) 5.12 q(2, J 5, -CH- 4.32 m(4), (300 MHz, acetone-d^ ) BOEtPIXDME [TFA deprotection] (300 MHz, acetone-dtf) BAnEtPIXDMEe (400 MHz, acetone-45) BF5EtPIXDMEe> f (400 MHz, CDCI3) BPrEPIXDME (200 MHz, CDCI3) BHEtPIXDME (300 MHz, CDCI3) BBrEtPIXDME (200 MHz, CDCI3) BTsEtPIXDME (200 MHz, CDCI3) BIEtPIXDME (200 MHz, C D C I 3 ) 10.04 s(l) (9.98 s, 9.97 s,(2)) 10.04 s(l), (9.91 s, 9.89 s, (2)), 9.79 s(l) 10.19 s(l), 10.09 s(l), 10.03 s(l), 9.99 s(l) 10.02 s(l), 10.00 s(l), (9.89 s, 9.88 s, (2)) (10.10 s, 10.09 s, 10.01 s, 9.99 s, (4)) (10.12 s, 10.10 s, 10.09 s, (4)) (10.07 s 10.05 s,(2)) (9.96 s, 9.95 s,(2)) 10.11 s(l), 10.01 s(l), 9.90 s(l), 9.83 s(l) 9.87 s(l), (9.80 s, 9.76 s, 9.73 s, (3)) (OCH 3) 2 3.27 m(4) (4.16 d(2, J 5), 4.08 d(2, J 5), -CH2-CH-) (3.45 s(6), 3.42 s(6) CH-OCH3) (10.20 s, 10.16 s, (2), CH2-C//O), 5.01 d(4, CH2-CHO) (7.22 t of ts(4, Ji 7, h 2), 6.85 d(4, J 8), 6.69 t(2, J 7), aniUne ring H), (5.24sbr(1.6 aniline N-H) -NW-Cf/2-C2F5 under 8 3.52 (7.90 m(2), 3.56 m(2), vinyl-W), (3.71 d(4), -CH2-), (3.92 s(6), OCH3) -OH not observed (7.18 2 ds(4), 6.16 m(4), Ts-ring H), (1.26 s, 1.23 s, (6, Ts-Ctf?)) 4.28 m(4), 3.25 m(4) (4.36 m(4), 3.28 m(4), -CH2CH2_ COOCH 3) 4.26 m(4, -CH2. CH2-NH-) 4.00 s br(4, -CH 2 -C//2-NH-), ((4.29 t(J 8), 4.24 t(J 8), (4)), (3.25 t(J 8), 3.21 t(J 8), (4)) C/ / 2 C^ 2 COOCH 3 ) (4.15 t(2, J 7), 4.06 t(2, J 7), -CH2. CH2-NH-) (-CH2CW2-NH- is under 8 3.52) 4.37 t(4, J 7), 3.27 t(4, J 7) 4.45-4.34 m (12), 3.28 t(4) 4.56, 4.39, 4.17 all overlapping ts(4), 3.27 t(4) 4.83 m(4), 4.41 m(8), 3.30 t(4) 4.45 m(4), 4.22 m(4), 3.81 m(4), 3.27 t(4) 3.62 s(3), 3.60 s(3), 3.56 s(3), 3.54 s(3), 3.49 s(3), (3.45 s(3) under 8,13-OMe) 3.60 s, 3.51 s, 3.50 s, 3.40 s, (18)) (3.62, 3.61 s, 3.60 s, (9)), 3.54 s(3), 3.46 s(3), 3.40 s(3) 3.62 s(3), 3.60 s(3), 3.52 s over complex pattern 8 3.57-3.50 (11, includes -CH2. NH- CH2), (3.48 s, 3.47 s, (6)), 3.45 s(3) (3.64 s, 3.60 s, 3.52 s,(18)) (3.66 s, 3.64 s, 3.63 s, (18)) (3.65 s, 3.63 s, 3.62 s, 3.61 s, 3.60 s, (18)) (3.67 s, 3.65 s, 3.64 s, 3.60 s, (12)), (3.52 s, 3.49 s, (6)) (3.52 s, 3.51 s, 3.49 s, 3.47 s, 3.45 s, 3.33 s, (18)) -4.07 s(1.5) -4.39 s(1.75) -3.97 s(1.5) -4.24 s(1.5) -3.80 s(2) -3.8 s(1.5) -3.84 s(1.5) -3.91 s(2) -3.8 s(1.5) References on page 187 168 Table 4.1. iH-NMR Data for the R-PIXDME Porphyrins (continued)3 Porphryin meso H 8,13-J?(or8,13-CH 2 CH 2 /? )b 2,18-C//2C#2- ( & S,l3-CH2CH2-)b -COOC//3, 3,7,12,17-C#3 Pyrrole N-H BNImEtPIXDME (400 MHz, CDC1 3)C 10.15 s, 10.06 s, 9.77 s, 9.76 s 6.67 s, 6.58 s, 5.89 s, 6.25 s (4,5-imidazolyl-//)d 5.09 t, 4.95 s br, 4.83 m, 4.52 t, 4.42 m, 4.36 s br, 3.29 tofds 3.67 s, 3.65 s, 3.54 s, 3.61 s, 3.43 s, 3.37 s -3.1 s br(2) BNImEtPIXDME (400 MHz, CDCI3/TFA) 11.09 s(l), 10.61 s(l), 10.58 s(l), 10.48 s(l) 6.73 s(l), 6.54 s(l), 6.30 s(l), 6.01 s(l) (4,5-imidazolyl-W)d 4.97 m(4), 4.48 m(4), 4.25 m(4), 3.19 m(4) 3.68 s(6) 3.57 s(3), 3.54 s(3), 3.47 s(3), 3.43 s(3) -3.1 s br(2) 2 9 / 3 0 mixture (300 MHz, CDCl 3 ) d 10.00 s, 9.90 s, 9.59 s, 8.75 s 7.24 d, 7.16 d (4,5-imidazolyl-//) [(8.14 q, 6.31 d, 6.15 d) vinyl] 4.30 m, 3.25 m 3.68 s 3.53 s, 3.50 s, 3.35 s -4.25 s BPlEtPIXDME (300 MHz, CDCI3) (10.21 s, 10.19 s,(2)) (10.06 s, 10.05 s,(2)) (7.80 m(4), 7.60 m(4), (phthalimido//)) 4.47-4.32 m(12), 3.27 t(4) (3.73 s, 3.69 s, 3.66 s, 3.65 s, 3.64 s, 3.62 s, (18)) -3.86 s(1.75) 3 1 / 3 2 mixture (300 MHz, CDCI3) 10.02-9.95 8s(4) (7.77 m(2), 7.55 m(2), phthalimido H) (8.25-8.12 overlapping ds(l), 4.32 m(4), vinyl) (4.15 m, 4.03 m, (4)),3.24q(4) (3.69-3.44 9s(18) -4.04 s(1.75) aMeasured at room temperature; 5 in ppm, signal pattern (number of protons, coupling constant when appropriate). bThe bracketed assignments are included when the substituents at the 8- and 13-positions are (2-/?)CH2CH2-. integrations did not match. dSinglets have been observed for other imidazole-containing porphyrins (see Section 4.3.2.3). eSpectrum assigned with the data from a 2D-1H-COSY spectrum (see Appendix). fSee Section 4.2.3.2 for the 1 9F-NMR data. Table 4.2. iH-NMR Data for the R-PIX Porphyrins3 Porphryin -COOH/ meso H 8,13-CH2CH2-fl 2,18-C//2C#2- & 8,13-C#2C#2-R 3,7,12,17-C#j Pyrrole N-H BHEtPIX 12.25 s br(l) / 5.14 s br(2) 4.36 t(4), 4.24 s 3.64 s(12) -3.99 s(2) (400 MHz, 10.30 s(l) (-CK2-OH) br(8), 3.19 t(4) DMSO-45) (10.23 s, 10.22 s, (3)) BNImEtPIX 12.5 s br(1.5) / 7.37 s(l), 4.90 t br (4), (3.58 s, 3.55 s -4.23 s(2) (400 MHz, 10.26 s(l), 7.36 s(l), 4.22 t br(4), (6)) DMSO-fifc) 10.03 s(l), 6.95 s(l), 4.32 s br(4), (3.30 s, 3.27 s 9.80 s(l), 6.90 s(l) 3.18 m(4) (6)) 9.75 s(l) (4,5-imidazolyl) BNImEtPIX not observed / 7.04 d(l, J 1), 5.26 s br(4), (3.80 s, 3.77 s, -4.0 s(2) (300 MHz, 11.38 s(l), 6.98 d(l, J 1), 4.96 s br(4), 3.62, s 3.57 s acetone-dg/TFA) 11.11 s(l), 6.75 d(l, J 1), 4.55 s br(4), (12)) (10.98 s, 10.93 6.68 d(l, J 1) 3.22 s br(4) s,(2)) (4,5-imidazolyl) aMeasured at room temperature; 5 in ppm, signal pattern (number of protons, coupling constant when appropriate). References on page 187 169 Table 4.3. UV-Vis Data for the R-PIXDME and R-PIX Porphyrins3 Porphyrin (solvent) Soret (log 8) Q Bands (log 8) BDMEtPIXDME (CH2C12) 400 498, 532, 568, 622 BOEtPIXDME (CH2C12) 400 (broad) 500, 534, 570, 622 BAnEtPIXDME (3.6 uM, CH2C12) 406 (5.57) 498 (4.41), 534 (4.28), 566 (4.27), 620 (3.89) BF5EtPIXDME (CH2C12) 400 498, 532, 568, 622 BPrAPIXDME (0.05 M NaOH ( a q )) 380 br 514, 552, 574, 630 BPrAPIXDME (CH 2Cl 2:MeOH 10:1) 406 506, 542, 574, 630 BHEtPIXDME (CH2C12) 400 498, 532, 568, 620 BHEtPIX ( 5.1 xlO"6 M, 0.05 M NaOH ( a q )) 372 (5.03, broad) 504 (3.83), 538 (3.70), 564 (3.61), 616 (3.32) BHEtPIX (1.1 uM, 10 % aq. Scintigest) 398 (5.25) 498 (4.17), 532 (4.01), 566 (3.88), 620 (3.69) BHEtPIX (9.9 uM, MeOH) 396 (5.01) 496 (3.97), 530 (3.79), 568 (3.62), 620 (3.28) BBrEtPIXDME (CH2C12) 400 500, 534, 568, 622 BTsEtPIXDME (CH2C12) 400 498, 532, 568, 622 BIEtPIXDME (CH2C12) 400 500, 534, 568, 622 BNImEtPIXDME (1.6 uM CH2C12) 400 (5.66) 498 (4.58), 532 (4.40), 568 (4.24), 622 (4.07), (328 (4.94) N02-Im) BNImEtPIX (1.2 pM, 10 % aq. Scintigest) 400 (5.21) 498 (4.19), 534 (4.00), 568 (3.85), 620 (3.66), 326 (4.55 N02-Im) BNImEtPIXBNA (NMP) 402 498, 532, 568, 622 BPlEtPIXDME (0.87 uM CH2C12) 400 (5.25) 498 (4.17), 532 (3.99), 568 (3.85), 622 (3.68) (230 (4.76) phthahmido) aIn general, when e values were measured, the Q bands' e values were determined at lOx the concentration listed. References on page 187 170 Table 4.4. Elemental Analyses for the R-PIXDME and R-PfX Porphyrins Porphyrin / Formula c% H % N % BAnEtPIXDME Calculated: 73.35 6.80 10.69 C 4 8 H 5 2 N 6 O 4 . 0 . 5 H 2 0 Found: 73.40 6.86 10.42 BHEtPIX Calculated: 67.20 6.47 9.22 C3 4 H 3 8 N 4 O 6 .0 .5 H 2 0 Found: 67.17 6.28 9.04 BNImEtPIXDME Calculated: 61.08 5.49 16.96 C 4 2H44N 1 0 O 8 ' l /2H 2 O Found: 61.11 5.36 16.81 BNImEtPIX Calculated: 59.55 5.25 17.36 C4oH4oNio08'lH20 Found: 59.40 5.18 17.30 BNImEtPDCBNA Calculated: 54.62 4.87 15.92 C 4 8 H 5 0 N 1 2 O 1 4 - H C l (54.65) (5.16) (15.93) (C48H 5 0 N 1 2 O 1 4 .2H 2 O) Found: 54.81 4.97 14.78 BPlEtPIXDME Calculated: 70.57 5.47 9.50 C 5 2 H 4 8 N 6 0 8 Found: 70.32 5.59 9.31 References on page 187 171 Table 4.5. Mass Spectrometry Data for the R-PIXDME Porphyrins and BNImEtPIXDNA-(tBu)2 Porphyrin (technique) LR-MS m/e (intensity %, assignment)3 HR-MS formula calc'd / found BOEtPIXDME (EI) 622 (100, M + ) C 3 6 H 3 8 N 4 0 6 622.27911 /622.28040 BAnEtPIXDME (+LSIMS) 777 (100, M + 1 ) not obtained BF5EtPIXDME (+LSIMS) 889 (100, M + 1 ) C 4 2 H 4 7 O 4 N 6 F 1 0 889.34991 / 889.35096 BPrAPIXDME (+LSIMS) 707 (100, M + 1 ) not obtained BPrEPIXDME (+LSIMS) 735 (100, M + 1 ) not obtained BHEtPIXDME (+LSIMS) 627 (100, M + 1 ) not obtained BBrEtPIXDME (+LSIMS) 753 (100, M + 1 ) not obtained BTsEtPIXDME (+LSIMS) 935 (100, M + 1 ) C 5 0 H 5 5 N 4 ° 1 0 S 2 935.33597 / 935.33261 BIEtPIXDME (+LSIMS) 847 (100, M + 1 ) C 3 6 H 4 1 I 2 N 4 0 4 847.12174 / 847.12294 BNImEtPIXDME (+LSIMS) 817 (0.3, M + 1 ) not obtained 2 9 / 3 0 mixture (+LSIMS) 704 (25, M + 1 ) C 3 9 H 4 2 0 6 N 7 704.31966 / 704.31925 BNImEtPIX (+LSIMS) 789 (1, M + 1 ) not obtained BNImEtPIXBNA-(tBu)2 (+LSIMS) 1244 (7, M + 1 ) C 6 4 H 8 3 N 1 2 ° 1 6 1243.61669 / 1243.61593 BPlEtPIXDME (+LSIMS) 885 (25, M + 1 ) not obtained 3 1 / 3 2 mixture (+LSIMS) 738 (100, M + 1 ) not obtained a M + 1 indicates (M+H)+. 4.3 Results and Discussion The chemistry in this chapter was founded on the work of Kenner et al. who were pursuing a convenient protecting group for vinyl moieties on porphyrins and phlorins.6 Of the compounds they reported, BDMEtPIXDME, BOEtPIXDME, BHEtPIXDME, and BBrEtPIXDME were used References on page 187 172 here. Gram quantities of BDMEtPIXDME were prepared from PIXDME via high yield T l i n -oxidation of the 8,13-vinyl groups. BOEtPIXDME, prepared by cleaving the acetal of BDMEtPIXDME, was used in reductive amination reactions with reagent-dependent results, and a Knoevenagel reaction (condensation of the in situ anion of malonic acid with the aldehyde moieties of BOEtPIXDME) was used to introduce a carboxylic acid-substituted alkene. BHEtPIXDME was used as a precursor to BHEtPIX, BBrEtPIXDME, BTsETPIXDME, and BIEtPIXDME which were investigated as reagents for SN2-type reactions. Because of its accessiblity, BBrEtPIXDME was chosen as the precursor to porphyrins incorporating nitroimidzole and phthalimide groups. The original goals of synthesizing porphyrins containing nitroimidazoles and pentafluoropropylamino groups were realized here. However, the syntheses of porphyrins incorporating tirapazamine and en groups (for exocyclic Pt-chelation) were less successful. Of the compounds synthesized here, only BHEtPIX and BNImPIX were evaluated in in vitro experiments at BCCRC (see Chapter 6). The synthesis pathways and conditions used here were presented in Scheme 4.1 (Section 4.2). 4.3.1 The Production and Reactions of B O E t P I X D M E 4.3.1.1 Synthesis of B O E t P I X D M E BOEtPIXDME was first produced by Kenner et al. via HC1 deprotection of BDMEtPIXDME in refluxing THF.6 However, these vigorous conditions tended to cleave the propionate ester groups as well, which complicated subsequent reactions because a reprotection step was required (Section 4.2.2). Later publications reported selective deprotection of the acetal groups of BDMEtPIXDME via transacetalation with TsOH in acetone8 and by treatment with HCOOH. 9 When this compound was investigated in this thesis work, I was unaware of Snow and Smith's method,8 and the method of Kahl et al. 9 had not been published. Several literature methods of acetal-deprotection were investigated before the TFA-deprotection was found. TFA-deprotection is an attractive method, as the reaction is complete in 5 min (as opposed to 2 h for transacetalation or HCOOH treatment), and large quantities of BOEtPIXDME are easily prepared. References on page 187 173 The reduction of this compound to BHEtPIXDME occurred in approximately the same yield (-75 %) as reported by Kenner et al.,6 but was lower than that reported by Kahl et al. (91 %).9 The product(s) formed via HCl-deprotection of BDMEtPIXDME were treated with CH2N2 to esterify the free -COOH groups, but a mixture of products was obtained and these were difficult to characterize. Kahl et al. had treated BOEtPIXDME with CH2N2 and obtained a mixture of epoxides and ketones (see Figure 4.1), and that such a mixture was likely obtained here. o R = A H,C CH3 or O - C — C H 2 H H3COOC COOCH3 Figure 4.1. Products of the Reaction of BOEtPIXDME with CH2N2. 4.3.1.2 Reductive amination Reductive amination is a general method of selective alkylation of amines via an imine formed from an aldehyde and the amine; the imine is typically reduced, often in situ, with a borohydride. With NaBH3CN, amines or aldehydes with otherwise reduction-sensitive functional groups (e.g. -CN, epoxides, or nitroxide radicals) can often be used.15'37 Reductive amination chemistry was applied to BOEtPLXDME as a means to introduce useful amine-containing compounds at the 8- and 13-positions of PIXDME (see Scheme 4.2). R-NH 2 + H 2 0 R-N=< NaBH 3CN R ' ^ H = BOEtPIXDME R' NaBH 3CN H R-N-C-R' H H Rate^  = k[H+] [aldehyde] [amine] BHEtPIXDME Scheme 4.2. In situ Reductive Amination with NaBH3CN. References on page 187 174 The synthesis of BAnEtPIXDME (Section 4.2.3.1) was investigated in order to evaluate the viability of reductive aminations with BOEtPIXDME. The best results were obtained when the BOEtPIXDME was produced via TFA-deprotection of the bis-acetal, and the imine was formed and reduced in situ. B F 5 E 1 P I X D M E (Section 4.2.3.2) was synthesized because of our group's interest in hypoxia probes incorporating fluorinated side-chains (see Chapter 1). Reductive amination with tirapazamine to form 26 was unsuccessful (Section 4.2.3.3), as only BHEtPIXDME and reduced tirapazamine were isolated from the reaction mixture. The reductive amination product of en with BOEtPIXDME (Section 4.2.3.4) was very polar, as would be expected for 28, but the product was difficult to characterize and was not pursued further. Compound 28 might have been a suitable ligand for exocyclic chelation of Pt (see also Section 4.3.3). The two obviously successful reductive aminations to make BAnEtPIXDME and BF5E1PIXDME gave rather low yields (-30 %), which may be related to the fact that BOEtPIXDME was reduced to BHEtPIXDME in a competing reaction under the reaction conditions (Scheme 4.2). Based on the work of Borch et al.,15 aldehydes and ketones are rapidly reduced by N a B H 3 C N at pH 3-4 (the reduction consumes H +), but the reduction is negliglible in the pH 6-7 range. It was thus surprising that BOEtPIXDME was so readily reduced, even in the presence of excess E13N (Section 4.2.3.2). The yield-determining factor for the reductive aminations with BOEtPIXDME is probably the position of the aldehyde-imine equilibrium (i) (Scheme 4.2). In the synthesis of BAnEtPIXDME, attempts to remove water via azeotropic distillation from benzene/CHCl3 did not give better yields, although this synthesis was performed with BOEtPIXDME formed from HCl-deprotection of the bis-acetal. A 6-fold excess of aniline per aldehyde was used with the TFA-deprotected BOEtPIXDME, which should have shifted equilibrium (i) (Scheme 4.2) to the right. Perhaps the N a B H 3 C N was added and reduction occurred before large amounts of imine could form. At a moderate pH, the rate of imine formation is dependent on [H+], [aldehyde], and [amine] (Scheme 4.2).38 However, at higher acid concentrations, the rate decreases because References on page 187 175 protonation decreases the amount of the free amine present. In this thesis work, the reaction pH was not monitored, but was kept about neutral by adding HC1 (for BAnEtPIXDME) or N M M (for BF5EtPLXDME), based on the literature methods.15 Attempts to optimize the reaction by adding 3A molecular sieves to remove water, thus shifting equilibrium (i) to the right, did not lead to higher yields. The molecular sieves decreased the pH of the solution (Section 4.2.3.2), and perhaps slowed the rate of the imine formation. Better results might be obtained if larger excesses of amines were added and the reaction mixture were stirred longer before adding the NaBH3CN, thus giving equilibrium (i) sufficient time to establish; also, the pH must be adequately controlled to ensure rapid formation of the imine. Finally, 3A molecular sieves might be employed if the reaction scale were large enough such that the sieves would have a minimal influence on the pH. The use of large amounts of C2F5CH2NH2*HC1 might not be practical, as it is not commercially available. However, the amine hydrochloride can be obtained in -45 % yield from the commercially available pentafluoropropionic acid.39 4.3.1.3 Synthesis of BPrAPIXDME and BPrEPIXDME C H 2 C F 2 C F 3 36 E F 5 An (F5)2-porphyrin conjugate (36) could be realized via an amide link as in EF5 itself and it also seemed possible to complete the synthesis of EF5 on the side-chain as in 35 (Section 4.2.15.4). However, as the 2,18-propionate esters would be later cleaved to COOH groups for testing the compounds at BCCRC, free COOH groups at the 8,13-positions were desirable. Direct oxidation of the aldehydes of BOEtPIXDME to the carboxylic acids via m-CPBA or Mg References on page 187 176 monoperoxyphthalate,6H20 was unsuccessful (Section 4.2.15.1), but the Knoevenagel reaction appeared to be a viable alternative. Indeed, when malonic acid, piperidine and BOEtPIXDME were mixed in pyridine and the mixture was heated, BPrAPIXDME was produced, presumably as a mixture of E and Z isomers. However, when the synthesis was scaled up, the yield was much lower and the isolation of BPrAPIXDME was more difficult. As BPrAPIXDME has two -COOH groups, very polar eluent mixtures were needed for chromatography and this complicated its isolation. For the sake of characterization, BPrAPIXDME was converted to BPrEPIXDME with CH2N2 but, as large quantities of BPrAPIXDME were not readily available, further derivitization was not investigated. 4.3.2 The Synthesis of B H E t P I X D M E and Its Derivatives 4.3.2.1 B H E t P I X D M E and BHEtPIX BHEtPIXDME was first described by Carr et al. as part of the total synthesis of PIXDME, 1 8 and was later reported by Kenner et al. and Kahl et al. who investigated vinyl protection groups6 and the reactivity of BOEtPIXDME9, respectively. Simple ether and glycoside analogues of BHEtPIX have been investigated as photosensitizing agents that are chemically pure alternatives to hematoporphyrin derivatives, which exist as a mixture of enantiomers and diastereomers.19'20 BHEtPIXDME was the starting material for the rest of the compounds shown in Scheme 4.1 (Section 4.2), and its synthesis was greatly simplified by TFA-deprotection of BDMEtPIXDME, as subsequent reprotection of cleaved esters was not required (Section 4.2.5). BHEtPIX, obtained via ester cleavage of BHEtPIXDME, was isolated by gradual acidification (to pH = 4.6) of a solution of this porphyrin in aqueous base; this effects protonation of the carboxylic acid groups and avoids protonation of the inner pyrrolic N atoms, which would tend to redissolve the porphyrin (Section 4.2.6). BHEtPIX was used in in vitro tests at BCCRC (see Chapter 6). References on page 187 177 4.3.2.2 B T s E t P I X D M E , BIEtPIXDME, and B B r E t P I X D M E Tosylation of the hydroxyethyl groups was considered a means to provide an easily displaceable substituent for subsequent SN2-type reactions with nitroimidazoles. The Ts-group was particularly labile, being converted, to the corresponding 2-chloroethyl derivative when BTsEtPIXDME was chromatographed, the chlorine presumably coming from the hydrolysis products of residual TsCl during chromatography. Similar chloro compounds have also been obtained when a bis(2-hydroxyethyl)phlorin (a precursor to PIXDME) was mixed with mesyl chloride.18 Crude BTsEtPIXDME was readily converted to BIEtPLXDME using Tovey's conditions.21 The syntheses of BTsEtPIXDME and BIEtPIXDME were time consuming, and the compounds were sometimes difficult to purify; they offered no significant advantage over BBrEtPIXDME and were not pursued further. In the reactions with nucleophiles, some fraction of BTsEtPIXDME, BIEtPIXDME and BBrEtPIXDME underwent elimination reactions to regenerate the vinyl groups (Sections 4.2.10, 4.2.13, and 4.2.15.4). These results were not surprising, as such haloalkyl and Ts-alkyl functionalities have been used as vinyl-protection groups, the vinyl moiety being regenerated when the compounds are heated in the presence of base.6 No reaction was observed when these porphyrins were mixed with C2F5CH2NH2 or tirapazamine (Sections 4.2.3.2 and 4.2.15.3), perhaps because of the poor nucleophilicity of these compounds. The reaction of BBrEtPIXDME with en did produce a very polar compound (Section 4.2.15.2) but, as in the reduction amination reaction with en (Section 4.2.3.5), the product was difficult to characterize and the chemistry was not pursued further. Crude BHEtPIXDME was used to produce gram quantities of BBrEtPIXDME (Section 4.2.7), the precursor to BNImEtPIXDME and BPlEtPIXDME. 4.3.2.3 BNImEtPIXDME, BNImEtPIX, and BNImEtPIXBNA-( t Bu) 2 Ester BNImEtPIXDME was synthesized from BBrEtPIXDME and 2-nitroimidazole in 44 % yield (Section 4.2.10). BNImEtPIXDME had very poor solubility in many solvents, which complicated its chromatographic purification. However, the low solubility was an advantage when purifying via hot THF extraction. Solublity was also a problem when obtaining ! H-NMR spectra. References on page 187 178 The integrations of the signals from a spectrum obtained in CDCI3 were not quantitative (Table 4.1) and the 5 values for the 8,13-CH2CH2- and imidazolyl-signals varied noticeably with concentration of the material, suggesting that aggregation occurred;40 in CDCI3/TFA, the integration values in the iH-NMR spectrum were self-consistent (see Figure 4.2). The IR-spectrum of BNImEtPIXDME shows strong resonances at 1482 and 1357 cm - 1 which correspond well to the reported asymmetric and symmetric stretches of N 0 2 groups, respectively.41 BNImEtPIX was produced in -75 % yield by treating BNImEtPIXDME with 25 % HC1 and was isolated in the same manner as BHEtPIXDME, by gradual acidification of a solution of BNImPLX in dilute aqueous base (Section 4.2.11). Although BNImEtPIX had reasonable solubility in DMSO and DMF, its solubility was poor in aqueous base, in contrast to BHEtPIX (see Chapter 6). In solutions of high ionic strength (i.e. after neutralization of the HC1), BNImEtPIX precipitated even though the carboxylic acids were in the form of their sodium salt at the pH used. Poor elemental analyses were obtained when the precipitated compound was redissolved in distilled water and isolated following dialysis. References on page 187 179 Figure 4.2. The 400 MHz !H-NMR Spectrum of BNImEtPLXDME in CDCI3/TFA. When BNImEtPIX was dissolved in 0.05 M NaOH, UV-Vis spectroscopy showed two Soret peaks and deviations from Beer's law were observed, which are evidence for aggregation.42 However, a single Soret peak was observed when MeOH was added, indicating that the aggregates were.now dispersed (Figure 4.3). Aliquots of a stock solution in 0.05 M NaOH added to PBS showed similar two-Soret behavior (spectra not shown), and when added to a +/- media (Section 6.2.1) used in the in vitro tests at BCCRC, the porphyrin precipitated (see Chapter 6). No evidence for aggregation was observed in 10 % aq. Scintigest, the solvent mixture used for References on page 187 180 digesting cells in the in vitro experiments (Section 6.2.1), and thus the £ values were also measured in this solvent mixture. o o o o o o o o o o cn "^ j- ir, \o r-Wavelength (nm) Figure 4.3. Normalized UV-Vis Spectra of BNImEtPIX in 0.05 M NaOH ( a q ) (—), and in 0.05 M NaOH(aq):MeOH (1:1) (-, the concentration was -0.5 of that in 0.05 M NaOH(aq)). The absorption scale is arbitrary. In an attempt to improve the solubility of BNImEtPIX in the media used in the in vitro experiments, the carboxylic acid groups were activated as their -NHS esters and then the groups were amidated with L-aspartic acid-(lBu)2 ester (Section 4.2.12). The resulting BNImEtPIXBNA-(lBu)2 ester was much more soluble in CDCI3 than BNImEtPIXDME and analyzed well by mass spectrometry. The lBu esters were cleaved in TFA and the resulting BNImEtPIXBNA had good solubility in dilute aqueous base and was partially soluble in PBS. In contrast to BNImEtPIX, the L-aspartate derivative remained in solution in a +/- media (Section 6.2.1) up to 100 [iM (prepared by dissolving BNImPIXBNA in dilute aqueous base, adding this solution to a +/- media and adjusting the pH to -7). However, the BNImPIXBNA was impure by RPtlc, despite the lBu ester having been purified twice by preparative-tic. As the overall yield to the BNA compound from References on page 187 181 BNImEtPIX was rather low (-25 %) and the quantities of BNImEtPIX were limited, this route was not pursued further, but it was clear that the solubility of BNImEfPIXBNA in culture media was much improved over that of BNImEtPIX. 4.3.2.4 BP lE tPIXDME and Attempts at Phthalimide Cleavage Two phthahmido groups were incorporated into the porphyrin via an SN2 reaction of BBrEtPIXDME and K»phthahmide in -35 % yield (Section 4.2.13). As in other reactions with BBrEtPIXDME (see above), elimination was a competing side-reaction. The iH-NMR spectrum of the product mixture of mono-elimination side-products (31 and 32) appears in Figure 4.4. For reference, the iH-NMR spectrum of BPlEtPIXDME appears in Figure 4.5. Attempts to cleave the phthahmide group via hydrazinolylsis met with limited success (Section 4.2.14). Some evidence for 27 (Section 4.2.3.4) was obtained (e.g. LR-MS, iH-NMR), but its synthesis was complicated by the polar nature of the products which were difficult to purify by chromatography. The 2,18-propionate esters of similar porphyrins are known to react with H2NNH2»H20 to produce a dihydrazide43-44 (see Scheme 4.3). If present here, such hydrazides and phthalimidohydrazide salts (from phthalimido cleavage) may have contributed to the complex product mixtures obtained. Use of 6 N HC1 did cleave the propionate esters of BPlEtPIXDME and may have cleaved the phthalimide groups, but attempts to reesterify the product yielded only BPlEtPIXDME. Primary amines have also been used to cleave phthalimide groups,31 but when BPlEtPIXDME was mixed with iPrNH2 a product was isolated that was not 27 and its identity was not pursued. Because of the low yield from BBrEtPIXDME and the difficulties encountered in cleaving the phthalimide, this chemistry was not pursued further. References on page 187 182 Figure 4.4. The 300 MHz !H-NMR Spectrum of a Mixture of 31 and 32 in CDCI3. References on page 187 183 H,c H,C H3COOC COOCH3 BPhEtPIXDME phthaloyl-// meso-// solvent / 8,13-C//2C//2-R, 2,18-C//2CH2-COOMe -COOC//3, 3,7,12,17-C//? 2,18-CH2C//2-COOMe T 1 I 1 1 1 1 1 1 • ' • I • l I l 1 • 1 J 1 1 1 1 J I 1 1 1 p 10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 -4.0 PPM Figure 4.5. The 200 MHz iH-NMR Spectrum of BPlEtPIXDME in CDCI3. H 2 NNH 2 «H2p H,C H3COOC H-,C Heat + 2 M e O H COOCH3 BjNHNOC CONHNH2 Scheme 4.3. The Formation of a Porphyrin-dihydrazide. References on page 187 184 4.3.3 Other Reactions The reaction of P I X D M E with en Stein and Plane first reported the synthesis of 33 (Section 4.2.15.2) from the reaction of PIXDME with refluxing en; they described the synthesis of several metal complexes of 33 (e.g. with Ag(U), Sn(IV), Cu(II), In(III), and Mn(U)) and investigated the kinetics of metallation of 33 with Zn. 3 2 Reported later were the kinetics of metallation of 33 with Cu and subsequent transmetallation with Zn, 4 5 aspects of dimerization of 33 and its metal complexes (e.g. with Zn, Cu, Ni, Mn) 4 6 4 9 the use of a photoreduced form of 33Sn(IV) as a co-catalyst in Pt-catalyzed hydrogen production from H2O, 5 0 and the evaluation of 33 as a photosensitizer51,52. Brunner et al. recently reported a Pt-complex of 33 in which the Pt-is bound through one of the 8,13-alkyl-en chains.53 Despite numerous reports on 33, few characterization data have appeared. Stein and Plane reported an elemental analysis (for C38H54Ni202*8H20, which does not match the empirical structure for 33!), and noted the absence of vinyls in the !H-NMR and IR spectra, while an amide resonance (1636 cm-1) appeared in the IR spectrum.32 White and Plane later reported only the C/N ratio in an elemental analysis of the HC1 salt of 33 as a measure of purity (C42H62Ni202*8HCl), and stated that "the UV-Vis, ipI-NMR, IR, and elemental analyses were consistent with formulation 33" 4 7 In this thesis work, 33 was produced using Stein and Plane's conditions but was very difficult to characterize, and this could explain the paucity of characterization data in the literature. The vinyl signals were absent from the Ifl-NMR spectrum, but the other signals were very broad and no assignments could be made. No peaks were observed by mass spectrometry; the elemental analysis did not match the expected formula, and RPtlc analysis indicated more than one product. The reaction was repeated and monitored by RPtlc which showed that at least two vinyl-containing intermediates were produced and consumed. Pure 33 could probably be obtained via preparative reverse-phase chromatography, but this route would be impractical given the cost of reverse-phase chromatographic media (~$400 per 100 g). References on page 187 185 In an effort to make a Pt-derivative, 33 was mixed with excess K^PtCLj. in H2O, and a brown precipitate formed. However, no further analyses of this precipitate were performed because of its poor solubility. Brunner and co-workers reported the synthesis of a Pt-complex of 33 (33Pt) which precipitated from a MeOH:H20 (20:1) solution maintained at pH ~6;5 3 they reported that 33Pt was water-soluble, but only IR data were reported. Compound 33Pt compared favorably with Photofrin® in PDT tests. Attempts to synthesize Tirapazamine- and EFs-porphyrin conjugates In an attempt to synthesize a PLX-tirapazamine conjugate (34, Section 4.2.15.3), the 2,18-propionic acid groups were activated as their isobutyl formate or NHS esters. In pyridine, the formate ester was decarboxylated and the resulting isobutyl ester did not react with tirapazamine. The NHS-ester of PIX also did not react with tirapazamine. Also, no reaction was observed when tirapazamine was mixed with BBrEtPIXDME in NMP in an attempt to synthesize an 8,13-subsituted tirapazamine conjugate (26, Section 4.2.3.3). Attempts to synthesize a porphyrin-EFs conjugate (35) through the amide-N of EF5 were unsuccessful (Section 4.2.15.4). When EF5 was mixed with BBrEtPIXDME and ip^EfN in NMP and the solution was heated, at least 20 products were seen by tic. When the amide N-H was deprotonated first with KO lBu, elimination was the dominant reaction. This chemistry was not pursued further. Reaction of P I X D M E with HBr and imidazole When PIXDME was treated with HBr/HOAc, and subsequently imidazole, at least 10 products were seen by tic (Section 4.2.15.5). Most of the products retained vinyl groups and some mass spectrometry evidence was obtained for a monovinyl, monoimidazolyl product. These results are similar to those observed with DPhDVPZn (Section 2.2.6.2), and this chemistry was not pursued further. 4.4 Summary New substituents were introduced at the 8,13-positions of PIXDME via the Tlin-oxidation of the vinyl groups to produce BDMEtPIXDME. A selective deprotection of the dimethyl acetal References on page 187 186 functionality of BDMEtPIXDME was developed. The deprotected product, BOEtPIXDME, was subsequently converted to derivatives incorporating aniline and C2F5CH2NH- groups via reductive amination chemistry; a -COOH substituted vinyl group was incorporated via a Knoevenagel reaction. Aspects of the successful reductive amination reactions are discussed. Other reductive amination reactions with en, tirapazamine and NH4OAC met with limited success. Reduction of BOEtPIXDME was carried out according to literature procedures and the resulting BHEtPIXDME was used to synthesize the Ts-, I- and Br-ethyl derivatives. BBrEtPIXDME was used to produce novel porphyrins incorporating 2-nitroimidazolyl and phthalimido groups. The phthahmido groups of BPlEtPIXDME were not cleanly cleaved. BNImEtPIX and BHEtPIX were produced by acid-hydrolysis of their parent esters. BNImEtPIX was further derivatized by forming the bis-L-aspartyl amide to yield a product with greater solubility in tissue culture media than the precursor. Attempts to incorporate EF5, and tirapazamine into derivatives of PIX were unsuccessful. References on page 187 187 References for Chapter 4 (1) DiNello, R. K.; Chang, C. K. In The Porphyrins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. I; pp 289-339. (2) Pandey, R. K.; Majchrzycki, D. F.; Smith, K. M. ; Dougherty, T. J. SPIE 1989,1065, 165-174. (3) Shiau, F.-Y.; Pandey, R. K.; Dougherty, T. J.; Smith, K. M . SPIE 1991,1426, 331-339. (4) Sakata, I.; Nakajima, S.; Koshimizu, K.; Hiroyuki, T.; Yasushi, I. JP Patent 080 567 682 A2 960312 Heisei, 1994. (5) Smith, K. M. ; Cavaleiro, J. A. S. Heterocycles 1987, 26, 1947-1963. (6) Kenner, G. W.; McCombie, S. W.; Smith, K. M . Liebigs Ann. Chem. 1973, 1329-1338. (7) Tufariello, J. J.; Winzenberg, K. Tet. Letters 1986, 27, 1645-1648. (8) Snow, K. M. ; Smith, K. M. J. Org. Chem. 1989, 54, 3270-3281. (9) Kahl, S. B.; Schaeck, J. J.; Koo, M.-S. J. Org. Chem. 1997, 62, 1875-1880. (10) Sterzycki, R. Synthesis 1979, 724-725. (11) Huet, F.; Lechevallier, A.; Pellet, M. ; Conia, J. M. Synthesis 1978, 63-65. (12) Coppola, G. M . Synthesis 1984, 1021-1023. (13) Ford, K. L.; Roskamp, E. J. J. Org. Chem. 1993, 58, 4142-4143. (14) Lipshutz, B. H.; Harvey, D. F. Synthetic Comm. 1982,12, 267-277. (15) Borch, R. F.; Bernstein, M . D.; Durst, H. D. J. Am. Chem. Soc. 1971, 93, 2897-2904. (16) Baird, I. R. University of British Columbia, personal communication, 1997. (17) Sparatore, F.; Mauzerall, D. J. Org. Chem. 1960, 25, 1073-1076. (18) Carr, R. P.; Jackson, A. H.; Kenner, G. W.; Sach, G. S. J. Chem. Soc. C 1971, 487-502. (19) Fulling, G.; Schroder, D.; Franck, B. Angew. Chem., Int. Ed. Engl. 1989, 28, 1519-1521. 188 (20) Pvimington, C ; R0nnestad, A.; Moan, J.; Lawson, A. M . Int. J. Biochem. 1992, 24, 951-957. (21) Tovey, A. N. M.Sc. Thesis, University of British Columbia, 1994. (22) Hay, M . P.; Lee, H. H.; Wilson, W. R.; Roberts, P. B.; Denny, W. A. J. Med. Chem. 1995,38, 1928-1941. (23) Begtrup, M. ; Larsen, P. Acta. Chem. Scand. 1990, 44, 1050-1070. (24) Aizawa, K.; Kuroiwa, Y. US Patent 5 567 409, 1996. (25) Bannwarth, W.; Schmidt, D.; Stallard, R. L.; Hornung, C ; Knorr, R.; Mtiller, F. Helv. Chim. Acta 1988, 71, 2085-2099. (26) Knorr, R.; Trzeciak, A.; Bannwarth, W.; Gillessen, D. Tetrahedron Lett. 1989, 30, 1927-1930. (27) Sternberg, E. The Univ. of British Columbia, Personal Communication, 1997. (28) Smith, K. M. ; Lee, S.-J. H. US Patent 5 330 741, 1994. (29) Gibson, M . S.; Bradshaw, R. W. Angew. Chem., Int. Ed. Engl. 1968, 7, 919-930. (30) Smith, P. A. S. Derivatives of Hydrazine and Other Hydronitrogens Having N-N Bonds; Benjamin/Cummings: London, 1983, Chapter 1. (31) Motawia, M . S.; Wengel, J.; Abdel-Begid, A. E.-S.; Pedersen, E. B. Synthesis 1989, 384-387. (32) Stein, T. P.; Plane, R. A. J. Am. Chem. Soc. 1969, 91, 607-610. (33) Fuhrhop, J.-H.; Smith, K. M . In Porphyrins and Metalloporphyrins; K. M . Smith, Ed.; Elsevier: Amsterdam, 1975; pp 757-869. (34) Traylor, T. G.; Chang, C. K.; Geibel, J.; Berzinis, A.; Mincey, T.; Cannon, J. J. Am. Chem. Soc. 1979,101, 6716-6731. (35) Nefzi, A.; Ostresh, J. M. ; Meyer, J.-P.; Houghten, R. A. Tetrahedron Lett. 1997, 38, 931-934. (36) Pandey, R. K.; Bellnier, D. A.; Smith, K. M. ; Dougherty, T. J. Photochemistry and Photobiology 1991, 53, 65-72. 189 (37) Lane, C. F. Synthesis 1974, 135-146. (38) Streitweiser, A.; Heathcock, C. H. Introduction to Organic Chemistry; 3rd ed.; Macmillan: New York, 1985. (39) Husted, D. R.; Ahlbrecht, A. H. J. Chem. Soc. 1953, 75, 1605-1608. (40) Scheer, H.; Katz, H. In Porphyrins and Metalloporphyrins; K. M . Smith, Ed.; Elsevier: Amsterdam, 1975; pp 399-524. (41) Silverstein, R. M . ; Bassler, G. C ; Morrill, T. C. Spectrometry Identification of Organic Compounds; John Wiley & Sons: New York, 1991, pp. 419. (42) White, W. I. In The Porphryins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. 5; pp 303-339. (43) Baker, E. W.; Ruccia, M . ; Corwin, A. H. Anal. Biochem. 1964, 8, 5*03-518. (44) Fisher, H.; Haarer, E.; Stadler, F. Z. Physiol. Chem. 1936, 241, 209-219. (45) Das, R. R. J. Inorg. Nucl. Chem. 1972, 34, 1263-1269. (46) Das, R. R.; Pasternack, R. F.; Plane, R. A. J. Am. Chem. Soc. 1970, 92, 3312-3316. (47) White, W. I.; Plane, R. A. Bioinorg. Chem. 1974, 4, 21-35. (48) Das, R. R.; Plane, R. A. J. Inorg. Nucl. Chem. 1975, 37, 147-151. (49) Das, R. R. J. Inorg. Nucl. Chem. 1975, 37, 153-157. (50) Fuhrhop, J.-FL; Kriiger, W.; David, H. H. Liebigs Ann. Chem. 1983, 204-210. (51) Kessel, D. Biochemistry 1977,16, 3443-3449. (52) Kohn, K.; Kessel, D. Biochem. Pharmacol. 1979, 28, 2465-2470. (53) Brunner, H.; Obermeier, H.; Szeimies, R.-M. Chem. Ber. 1995,128, 173-181. 190 Chapter 5 The Cyclic Voltammetry of Selected Porphyrins 5.1 Introduction Investigations of nitroaromatics and heterocyclic-N-oxides as radiosensitizers and hypoxia-selective cytotoxins (HSCs) have clearly demonstrated a correlation between the efficacy of these compounds and the one-electron reduction potential of the nitro or N-oxide moiety.1"6 In the case of radiosensitizers, reduction occurs via radicals generated by ionizing radiation on the DNA structure (thus making the DNA damage permanent); and the reduction products may react further to produce DNA adducts, further damaging the DNA (see Chapter l ) . 2 , 7 It should be noted that many compounds which were developed as radiosensitizers also act as HSCs, and vice versa. The difference between radiosensitizers and HSCs lies in the source of electrons performing the reduction. With HSCs drugs, cellular reductive enzymes (e.g. cytochrome p450 reductase, and the xanthine oxidase system that involves 2e-redox processes,8 etc.) are thought to be the reducing agents;5'9 the reduced species (or subsequent reduction or fragmentation products) then react with DNA to produce cytotoxic lesions.2 If the reduction potential is too negative (ignoring other pharmacological effects), insufficient quantities of reactive species will be produced. If the reduction potential is too positive (i.e. approaches that for reduction of 0 2 to superoxide), loss of hypoxia selectivity is observed.2'6 Given this biological importance of reduction potentials, it was of interest to evaluate them for selected porphyrins from this work and for the nitrophenylporphyrins synthesized by G. Meng.10 5.2 Experimental The experiments were performed using a Bipotentiostat Model AFCBP1 (Pine Instrument Co.) using PineChem 2.0 software. Generally, a solution of the porphyrin, or reference compound (1 mM) and BU4NCIO4 (0.1 M) in anhydrous degassed DMF was prepared under N2 in the cell shown in Figure 5.1, and the potential was scanned down at 100 mV/s. OPyTrPhP did not completely dissolve in DMF, its concentration being < 5.3 x 10"4 M . The working and counter References on page 200 191 electrodes were Pt wire, and the reference electrode was Ag wire. Once the correct potential range to observe all the expected reductions was found, a fresh solution of the porphyrin and electrolyte was prepared and the scan performed again. If necessary, a third solution was prepared and analyzed, and the results of the runs were averaged. The reduction potentials (E1/2) were obtained from the average of the potentials of the reduction and oxidation peaks [(Ep.c. + Ep.a.)/2].n Pt Wire Electrode 11 cm 15 cm Figure 5.1. Cyclic Voltammetry Cell (Made by Steve Rak, UBC Department of Chemistry). Reproducible reduction potentials (within 0.03 V) were obtained for the ferrocenium/ferrocene and TPhP/TPhP" redox couples (Table 5.1). The SCE is a common practical reference electrode used in porphyrin redox chemistry and comparisons have been made with this reference for the reduction potentials of radiosensitizers and HSCs measured by other methods (see Section 5.3). Therefore, to reference the reduction potentials vs. SCE, 0.19 V was subtracted from the reduction potentials measured versus Ag wire (see Table 5.1). Table 5.1. Reduction Potentials (Volts) of Reference Compounds and Correction Factor Calculation. Redox Couple Ered vs. SCE Efed vs. vs Ag (Correction Factor) Ferrrocenium / Ferrocene 0.48 (ref. 12) 0.67 0.19 TPhP / TPhP- -1.05 (ref. 13) -0.86 0.19 References on page 200 192 5.3 Results and Discussion 5.3.1 Porphyrin Electrochemistry Porphyrins and metalloporphyrins exhibit a rich electrochemistry. Under reductive conditions in aprotic media (DMSO, DMF), free-base porphyrins exhibit two, reversible one-electron reductions to generate the porphyrin radical anion and dianion, respectively.14 In this thesis work, when nitroaromatic or heterocyclic N-oxide substituents are present, an additional reduction peak is observed (see Scheme 5.1). In addition to the pathways outlined in Scheme 5.1, the possibility exists that the reduction potential of a substituent would fall between the two reduction potentials Ei/2(1) and Ei/2(2). The reduction potentials of the selected porphyrins are summarized in Table 5.2. Ei/2(1) + e" Pathway A: p ^ free-base porphyrin Ei/2(2) + e" radical dianion V dianion with anion \ r e d u c e d substituent — P^S Pathway B: E1/2(S) + e" PS PS" Ei/2(1) + e" Ei/2(2) + e" PS ^ free-base porphyrin with substituent (S) porphyrin with reduced substituent radical anion with reduced substituent - e P2"S" dianion with reduced substituent Scheme 5.1. Reduction Pathways of Porphyrins in Cyclic Voltammetry Experiments. Pathway A is valid for porphyrins not containing easily reducible substituents, or containing substituents with reduction potentials more negative than that of the porphyrin dianion. Pathway B is valid for porphyrins containing substituents with reduction potentials less negative than that of the porphyrin. In general Ei/2(1) - Ei/2(2) is 0.43 ± 0.03 V . 1 4 It is assumed reduction of the porphyrin core and substituents are mutually independent. References on page 200 193 Table 5.2. The Reduction Potentials (measured in Volts in DMF, referenced vs. SCE) of Selected Porphyrins and Reference Compounds (cf. Scheme 5.1); tw = this work Porphyrin El/2(1) Ei/2(2) El/2(S) AEi/ 2(l-2) Ref. TPhP -1.05 -1.52 not observed 0.47 tw TPhP -1.05 (-1.08) -1.47 (-1.52) 0.42 (0.44) 13 (14) PyTrPhP -1.04 -1.50 not observed 0.46 tw PyTrPhP -1.04 -1.48 not observed 0.44 14 OPyTrPhP3 -0.94 -1.41 -1.91 to 2.09b>c not apphcable tw pyridine-N-oxide not applicable not apphcable -2.20b>d not apphcable tw pyridine-N-oxide not apphcable not apphcable -2.lb,e not apphcable 15 ?-NPhDPhPyPa>f -0.97 -1.61 not observed, see text 0.64 tw c-NPhDPhPyPa.f -0.97 -1.55 not observed, see text 0.58 tw c/f-NPhDPhPyPa -0.998 14 nitrobenzene not apphcable not apphcable -1.15n not apphcable tw TirapPhTrPhP -1.11 -1.56 -0.78,b -0.94b 0.45 tw tirapazamine not apphcable not apphcable -0.61 b -0.84b not apphcable tw tirapazamine not apphcable not apphcable -1.08n not apphcable tw BNImEtPlXDMEa -1.35 -1.77 -1.06 0.42 tw MPDCDMEa -1.34 -1.73 not observed 0.39 16 2-nitroimidazole not apphcable not apphcable -0.84b not apphcable tw 2-nitroimidazole not apphcable not apphcable -1.04h not apphcable tw Structure is shown below. bOnly the reduction peak was observed (irreversible reduction), and thus this value is not a reliable measure of E1/2. cThe values were inconsistent between successive runs. dMeasured using a scan rate of 2500 mVVsecond. eMeasured using a scan rate of 10 V/second. Synthesized by G. Meng. 1 0 ^Estimated from results obtained by Worthington et al.14 ^Estimated from data correlating E 1 7 values (the one-electron reduction potential measured by pulse radiolysis in aqueous solutions at pH 7, see Section 5.3.4) and E1/2 by cyclic voltammetry in D M F . 4 ' 5 ' 1 7 References on page 200 194 OPyTrPhP ?-NPhDPhPyP c-NPhDPhPyP TirapPhTrPhP BNImEtPIXDME MPIXDME 5.3.2 General Observations In general, the difference between the anodic and cathodic peak potentials was between 70-120 mV at this scan rate (100 mV/s) (see Figure 5.2) and slower rates were not investigated. A reversible redox couple should have a peak potential difference of 59 mV, and thus these reductions were presumably quasi-reversible at this scan rate.11 With the exception of the nitrophenylporphyrins, the differences between the two E1/2 values (AEi/2(l-2)) are within the reported range (0.43 ± 0.03 V) . 1 3 - 1 4 5.3.3 Interpretation of Results The one- and two-electron reduction potentials of TPhP and PyTrPhP agree well with those reported in the literature. PyTrPhP was analyzed in order to provide a reference for comparison with OPyTrPhP, c- and f-NPhDPhPyP. OPyTrPhP had poor solubility in DMF and the References on page 200 195 appearance of the voltammogram was generally poor; thus, there is uncertainty within the reduction potential of the oxidopyridyl group. Based on this work and the literature data, however, the oxidopyridyl moiety is reduced at a potential more negative than those of the porphyrin core and the reductions within a oxidopyridyl-substituted porphyrin would follow Pathway A (Scheme 5.1). The possibility exists that the porphyrin Ei/2(1) and oxidopyridyl reductions are not mutually independent. The first reduction potential of the nitrophenylporphyrins (c- and f-NPhDPhPyP) was observed at -0.97 V. This value matches well with that predicted by Worthington et al. who compiled the reduction potentials of seventy-five free-base porphyrins and generated 'partial potential' values for a variety of substitutents, which provides the means to predict the Ei/2(1) of unsymmetrically substituted porphyrins.14 In this thesis work, no separate peak was observed for the nitrophenyl moiety (see Figure 5.2). Based on the expected reduction potential for nitrobenzene (Table 5.2), it is likely that the E1/2 of the nitrophenyl moiety and Ey2(l) overlap and are not distinguished. The AEi/2(l-2) is greater than 0.43 V, and the current observed for the Ei/2(1) value is about twice that in the second Ei/2(2) measurement; these results support the conclusion that the nitrophenyl reduction is incorporated within the first reduction peak.11 The possibility exists that the porphyrin Ei/2(1) and nitrophenyl reductions are not mutually independent. The voltammogram of TirapPhTrPhP showed two small peaks which appeared before the Ei/2(1) on the cathodic sweep, but no corresponding oxidation peaks were observed on the anodic sweep. Similar results were observed with tirapazamine in this work, and the radical anion of tirapazamine is known to be unstable and highly reactive.5 Therefore, it is reasonable to conclude that the reduction follows Pathway B (Scheme 5.1) and to assign these peaks to the N-oxide groups(s) of the tirapazamine substituent. Again, the reduction of the tirapazamine substituent and the porphyrin core may not occur independently. References on page 200 196 50' 25 4 ^ 0-I 3 u -25 H -5(H -75-(N -0.91 -1.56 El/2( 2) _ — ^ — / ^ Scan direction / El/2(D / / -1.65 -1.03 o Potential (V) vs. SCE Figure 5.2. The Cyclic Voltammogram of ?-NPhDPhPyP Measured in DMF. The first reduction in the voltammogram of BNImEfPLXDME is assigned to the 2-nitroimidazolyl substituent as the second two reductions match well those of a similar compound, mesoporphyrin IX dimethyl ester (MPIXDME). In this case, the reduction more clearly follows Pathway B and the reductions of the nitroimidazole moiety and porphyrin ring are independent. Calculating E1/2 for 2-nitroimidazole itself was difficult because no oxidation peak was observed on the anodic sweep, but the value was estimated from the literature data and it corresponds well with the peak assigned to the 2-nitroimidazolyl substituent. 5.3.4 Correlation with In Vitro Efficacy One of the most useful parameters to predict the efficacy of radiosensitizers or HSCs is the one-electron reduction potential, E1 7 , of the nitro or N-oxide moieties, which is usually measured by pulse radiolysis in neutral, buffered, aqueous media, similar to physiological conditions.4 This reduction is, in effect, the 'trigger' which activates the cytotoxic aspects of the molecule and is the key to selective cytotoxicity. However, the kinetic and mechanistic aspects of subsequent reactions, in addition to pharmacological properties of the prodrugs and their reduced species, are References on page 200 197 also important.3 For nitroaromatic compounds, the ideal reduction potential for the first reduction appears to be in the range -0.30 to -0.45 V vs. NHE. The reduction potentials measured by cyclic voltammetry in DMF (E1/2) cannot be directly compared to E1 7 values, but a linear correlation has been observed between the two and can be used to approximately convert from E1/2 to E1 7 and vice versa.1 7 , 1 8 The E1 7 values for several radiosensitizers and HSCs (and selected structures) appear in Table 5.3. The compound with the most positive reduction potential (i.e. easiest to reduce), nitracrine, is a potent HSC in vitro. However, the reduction potential is sufficiently high such that the nitro group is rapidly reduced via metabolic processes in vivo. Because of this and because of its strong affinity for DNA, nitracrine did not diffuse well into hypoxic areas and was not suitable for use in vivo.19 Metronidazole, 2-nitroimidazole, and misonidazole, compounds with moderate E1 7 values, are effective radiosensitizers and/or HSCs and are currently being, or have been, clinically investigated as anti-cancer agents.1'3 Although the E 1 7 values for porphyrins incorporating heterocyclic N-oxides and nitroaromatic groups were not directly determined, some correlation between these values and those of known radiosensitizers and HSCs (Table 5.3) can be made. The compound with the most negative reduction potential (i.e. most difficult to reduce) is nitrobenzene (-0.49 V), which is beyond the optimal potential for usefulness as a radiosensitizer or HSC. The Ei/2(1) of c- or t-NPhDPhPyP is ca. 0.18 V less negative than that of nitrobenzene (Table 5.2), which indicates that these porphyrins might be effective radiosensitizers or HSCs. However, some water-soluble nitrophenylporphyrins were not effective radiosensitizers (their reduction potentials were not measured),20 and other factors such as aggregation in aqueous media and the uncertainty in correlating the E1/2 and E 1 7 values must be considered. References on page 200 198 Table 5.3. Literature Data for One-electron Reduction Potentials (E1 7) of Selected Radiosensitizers and HSCs Measured by Pulse Radiolysis in Neutral Aqueous Media. Compound E i 7 vs. NHE (V) nitrobenzene -0.490a NO, r=< V ~ ~ O H CH3 (metronidazole) -0.486a CO N+ NH2 6 -(tirapazamine) -0.45b 2-nitroimidazole -0.418a /=\ OH N Y N ->-^- ' 0 M e NOZ (misonidazole) Me -0.389a -0.297C (nitracrine) aData from Wardman. bData from Wardman et al. cData from Lee et al.19 Porphyrins incorporating heterocyclic N-oxides (i.e. OPyTrSPhP and TirapPhTrPhP) also showed no radiosensitizing or hypoxia-selective activity (see Chapter 6). Judging from the voltammetry results, the reduction potential of the oxidopyridyl moiety is not within a range to be useful as a radiosensitizer or HSC. In contrast, the reduction potential of the tirapazamine substituent in TirapPhTrPhP appeared to be in the proper range. However, this porphyrin had to be evaluated as a liposomal formulation (Section 6.2.2.2), which may have compromised its effectiveness as a radiosensitizer or HSC. Based on the E1/2 alone, BNImEtPIXDME shows References on page 200 199 promise as a radiosensitizer or HSC. However, this porphyrin was difficult to evaluate because of its poor solubility (see Chapter 6). 5.4 Summary Several porphyrins were evaluated by cyclic voltammetry in DMF. Based on reported E1/2 and E1 7 values, the Ei/2(1), Ei/2(2) and Ei/2(S) values of the oxidopyridyl, nitrophenyl, tirapazamine, and 2-nitroimidazolyl substituents are assigned. The usefulness of porphyrins containing these substituents as anti-cancer agents is discussed by comparing these reduction potentials to those of known radiosensitizers and HSCs and in vitro evaluation results for the porphyrins. References on page 200 200 References for Chapter 5 (1) Adams, G. E.; Flockhart, I. R.; Smithen, C. E.; Stratford, I. J.; Wardman, P.; Watts, M . E. Rod. Research 1976, 67, 9-20. (2) Denny, W. A.; Wilson, W. R. J. Med. Chem 1986,29, 879-887. (3) Denny, W. A.; Wilson, W. R.; Hay, M . P. Br. J. Cancer 1996, 74, S32-S38. (4) Wardman, P. Environ. Health Perspec. 1985, 64, 309-320. (5) Wardman, P.; Priyadarsini, K. I.; Dennis, M . F.; Everett, S. A.; Naylor, M . A.; Patel, K. B.; Stratford, I. J.; Stratford, M . R. L.; Tracy, M. Br. J. Cancer 1996, 74, S70-S74. (6) Zeman, E. M . ; Maker, M . A.; Lemmon, M. J.; Pearson, C. I.; Adams, J. A.; Brown, J. M . ; Lee, W. W.; Tracy, M . Int. J. Rod. One. Biol. Phys. 1989,16, 977-981. (7) Adams, G. E. Rad. Res. 1992,132, 129-139. (8) Cory, J. G. In Textbook of Biochemistry with Clinical Correlations; T. M . Devlin, Ed.; Wiley-Liss: New York, 1997; pp 502-503. (9) Naylor, M . A.; Adams, G. E.; Haigh, A.; Cole, S.; Jenner, T.; Robertson, N.; Siemann, D.; Stephens, M . A.; Stratford, I. J. Anti-Cancer Drugs 1995, 6, 259-269. (10) Meng, G. G.; James, B. R.; Skov, K. A. Can. J. Chem. 1994, 72, 1894-1909. (11) Heinze, J. Angew. Chem., Int. Ed. Engl. 1984, 23, 831-847. (12) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877-910. (13) Davis, D. G. In The Porphryins; D. Dolphin, Ed.; Academic Press: New York, 1978; Vol. 5;pp 127-152. (14) Worthington, P.; Hambright, P.; Williams, R. F. X.; Reid, J.; Burnham, C ; Shamim, A.; Turay, J.; Bell, D. M. ; Kirkland, R.; Little, R. G.; Datta-Gupta, N.; Eisner, U. J. Inorg. Biochem. 1980,12, 281-291. (15) Miyazaki, H.; Matsuhisa, Y.; Kubota, T. Bull. Chem. Soc. Jpn. 1981, 54, 3850-3853. (16) Peychal-Heiling, G.; Wilson, G. S. Anal. Chem. 1971, 43, 545-556. (17) Breccia, A.; Berrilli, G. Int. J. Radiat. Biol. 1979, 36, 85-89. (18) Tercel, M . ; Wilson, W. R.; Denny, W. A. J. Med. Chem. 1993, 36, 2578-2579. 201 (19) Lee, H. H.; Wilson, W. R.; Ferry, D. M. ; van Zijl, P.; Pullen, S. M. ; Denny, W. A. J. Med. Chem. 1996, 39, 2508-2517. (20) James, B. R.; Meng, G. G.; Posakony, J. J.; Ravensbergen, J. A.; Ware, C. J.; Skov, K. A. Metal-Based Drugs 1996, 3, 85-89. 202 Chapter 6 In Vitro Evaluation of Selected Porphyrins 6.1 Introduction A feature of this thesis work was the opportunity to gain expertise in both synthetic chemistry and the biological screening of potential anti-cancer agents, facilitated by the ongoing collaboration between Prof. James and Dr. Skov (see below). The original goal of this part of the project was the development of porphyrin-based radiosensitizers (Section 1.5). However, because some of the moieties incorporated into the porphyrins (e.g. nitroimidazoles and heterocyclic N-oxides) can act as hypoxia-selective cytotoxins (HSCs, Section 1.4.2),1 the hypoxia-selective properties of the porphyrins were also evaluated. In addition, many porphyrins can act as photosensitizers in photodynamic therapy (PDT, Section 1.4.3);2"4 thus, it was logical to evaluate these compounds for their photosensitizing abilities. The experiments described this Chapter involved collaboration with several departments of the BCCRC: the Skov Lab (toxicity and accumulation), Medical Biophysics (radiosensitization), Photobiology (photosensitization), Advanced Therapeutics (liposomal formulations), and Cancer Imaging (fluorescence microscopy). Previous graduate students who have taken part in this collaboration between Prof. James and Dr. Skov have investigated metal (e.g. Pt and Ru) complexes of nitroimidazoles,5'6 Ru-sulfoxide complexes7'8 and numerous water-soluble porphyrins and metalloporphyrins.9"11 Although the compounds and complexes investigated to date5'7'9"12 have not been as effective as current treatment modalities, significant advances have been made in these areas of chemistry and the experience gained in such interdisciplinary work has been invaluable. 6.2 Experimental Several porphyrins and reference compounds were evaluated for their toxicity (in air and N2), radiosensitizing (in N2) and photosensitizing (in O2) abilities toward CHO cells. A summary of the assays performed in this work is found in Table 6.1. OPyTrSPhP, BNImEtPIX, and TirapPhTrPhP were chosen because of the interest in nitroimidazole- and heterocyclic N-oxide-References on page 233 203 containing porphyrins (see Chapter 1). The remaining compounds (PyTrSPhP, pyridine N-oxide, BHEtPIX, APhTrPhP, and Photofrin®) were evaluated to provide a data base against which the in vitro results of OPyTrSPhP, BNImEtPIX and TirapPhTrPhP could be compared. Table 6.1. Summary of Performed In-vitro Assays3 Compound Toxicity Radio-sensitization Photo-sensitization Cell Accumulation Type of Formulation OPyTrSPhP 0.1 to 2.5 mM (<*+/-) 0.3 - 1.0 mM («+/-) 0.3 - 0.5 mM (a+/- or a-/-) yes a-mediumb PyTrSPhP c 0.3 - 1.0 mM («+/-) 0.5 mM (a+/- or a-/-) yes a-mediumb pyridine N-oxide not tested not tested 0.5 mM (a+/- or a-/-) no a-mediumb Photofrin® not tested not tested 2.5 p:g/mL (a+/- or a-/-) yes a-mediumb BHEtPIX 0.05 - 0.2 mM (a+/- or a-/-) 0.2 mM (cc+/- or a-/-) -33 [iM (a-/-) yes a-medium,b Cremophor EL® BNImEtPLXd -30 (iM (a-/-) -30 pJvl (a-/-) -30 [iM (a-/-) yes liposomes, Cremophor EL® APhTrPhP -30 | iM (a-/-) -30 (iM (a-/-) -30 | iM (a-/-) yes liposomes TirapPhTrPhP -22 p:M (a-/-) -22 pJvl (a-/-) -22 (iM (a-/-) yes liposomes aThe values in the toxicity, radiosensitization and photosensitization columns refer to the concentrations at which the compounds were evaluated; the type of a-modified media used for incubation with the cells appears in the parentheses. bMedia refers to a-modified media described in Section 6.2.1. Tested previously (ref. 6). dBNImEtPIX was insoluble in media formulations. 6.2.1 Materials and Methods All media, buffer solutions and drug solutions (including Cremophor EL® emulsions) were sterilized by filtration through a 0.22 pirn filter (Nalgene) before cell suspensions were added. The References on page 233 204 liposomal formulations were not sterilized after preparation, but were prepared with sterile solutions. Media for biological tests Eagle's minimum essential medium (Gibco) formed the basis of the a-modified media used for the maintenance and incubation of the CHO cells. To make a-/- medium, one packet of a-medium powder and 10,000 units of penicillin/streptomycin antibiotic (Gibco) were added to 10 L of doubly distilled water and the solution was stirred for 2 h at room temperature. Preparation of oc+/+ medium involved the addition of fetal bovine serum (Gibco) to a final concentration of 10 % and NaHC03 (20 g /10 L) to the a-/- solution before filtration. Then the pH was adjusted to 7.3 with 4 M NaOH(aq). Preparation of oc+/- medium was accomplished by adding 10 % (v/v) of fetal bovine serum to a-/- medium buffered with 10 mM HEPES. Al l media were stored at 4 °C. To prepare the phosphate buffer saline (PBS) solution, NaCl (160 g), KC1 (4 g), Na2HP04 (23 g) and KH2PO4 (4 g) were dissolved in distilled water (20 L). The solution was stored at room temperature but was cooled to 4 °C before use. A solution of methylene blue (2 g dissolved in 1 L H2O) was used for fixing and staining of cells after incubation. Solutions for the accumulation measurements were made by solubilizing the cells in 10 % ScintiGest(aq) (a tissue solubilizer composed of cetyl trimethylammomum bromide, MeOH and KOH, Fisher Scientific Co.). Cell handling Aseptic techniques were used for working with the CHO cells in order to avoid contamination. The cells were typically grown in suspension in a+/+ medium in bottles with magnetic stir bars maintained at 37 °C in an "Incu-cover" incubator (Associated Biomedic Systems) under air:C02 (19:1). The CHO cells showed a doubling time of-13 h, and these spinner cultures were diluted with fresh medium to a cell concentration of -1 x 105 cells/mL on a daily basis to maintain exponential growth. References on page 233 205 General equipment and techniques To avoid inadvertent photosensitization during incubation of the cells with the porphyrins, the vessels used in the assays were kept from light exposure by wrapping them with aluminum foil and performing all operations (e.g. sampling, centrifuging, plating) under subdued lighting conditions. The cells were isolated from suspension by centrifugation (Sorvall RC-3, 600 RPM, usually 8-10 min, maintained at 4 °C). Cell concentrations (cells/mL) were measured using a "Coulter Cell Counter" (Coulter Electronic Inc.). Tray incubators (National Inc.) were maintained at 37 °C with an air:CC«2 (19:1) gas flow to buffer pH to 7.3. The N 2 used to produce hypoxic conditions was 02-free grade (Linde Specialty Gas, Union Carbide). Sterile, plastic Petri dishes and centrifuge tubes (Falcon) were used for all experiments. The pH measurements were made with a Orion Model 420A pH meter, which was calibrated with pH 7.00 and 10.00 buffer solutions before use. 6.2.2 Preparation of Solutions and Formulations 6.2.2.1 Media Solutions OPyTrSPhP, PyTrSPhP, and pyridine N-oxide had good water-solubility and fresh stock solutions (~2 mM) were prepared by dissolving the compounds in the appropriate medium (a+/- or a -/-) on the same day as the experiment. Solutions of up to 0.2 mM BHEtPIX were prepared by dissolving the porphyrin in a minimum of dilute NaOH; the appropriate medium (oc+/- or a -/-) was added, and the pH of the solution was adjusted to -7.3 with dilute HC1. Solutions of pyridine N-oxide and Photofrin® (donated by Dr. Mladen Korbelik) were prepared in a similar manner. Attempts to prepare solutions of BNImEtPIX in this manner were unsuccessful, as the porphyrin precipitated when the oc-medium was added to the solution. However, solutions of up to 0.5 mM of BNImEtPIX could be prepared in PBS in a similar manner, but the UV-Vis spectrum of this solution indicated significant aggregation (see Chapter 4). References on page 233 206 6.2.2.2 Liposomal Formulations The liposomal formulations were prepared on the day of the experiment either by Method I or Method II. The mean diameter of the vesicles was measured by quasi-elastic light scattering on a Nicomp Model 270 submicron particle analyzer. Method I The preparation of a liposomal formulation of 5-(4-aminophenyl)-10,15,20-trisphenylpophryin (APhTrPhP) was based on a general technique.13 APhTrPhP (2 mg) and egg phosphatidylcholine (EPC, Northern Lipids) (40 mg) were dissolved in CHCI3; the solvent was then evaporated by a stream of N2 and the sample placed under vacuum for 1-2 h. Then sterilized 5 % dextrose (4 mL) was added and the sample was vortexed in order to hydrate the lipid film. Finally, the solution/suspension was passed through an extruder (Lipex Biomembranes, Vancouver, BC) using a 0.4 p:m polycarbonate membrane (2 times) and then a 0.1 |im polycarbonate membrane (10 times) (membrane source: Poretics brand, AMD Manufacturing, Mississauga, ON). When the flow slowed considerably, usually after the second or third extrusion, the 0.1 (im membranes were replaced with fresh ones. The pressures required for extruding were generally 500 and up to 1600 psi for the 0.4 and 0.1 p:m filters, respectively. The concentration of the porphyrin was determined by extracting the porphyrin from an aliquot of the formulation with CHCI3/E1OH (10:1), diluting 10-fold with CHCI3 (now 100:1 CHCl3:EtOH) and measuring the UV-Visible spectrum. Based on the reported molar absorptivity coefficient of the Soret band (395,000 in C H C I 3 ) , 1 4 the concentration of APhTrPhP in the formulation was 0.8 mM. Method II The preparation of stock liposomal formulations of TirapPhTrPhP, BNImEtPIX and a control (without a porphyrin) was accomplished by the method of Mayer.15 TirapPhTrPhP (2.5 mg) and EPC (50 mg) were dissolved in NMP (0.5 mL) with gentie heating, vortexing and sonication. The TirapPhTrPhP solution (150 p:L) was added to a sterile solution of 5% dextrose (4 mL) while the dextrose solution was vortexed, this yielding a -225 p:M stock solution. A control References on page 233 207 stock liposome formulation was prepared by dissolving EPC (50 mg) into NMP (0.5 mL), and 300 (iL of this solution were added to 4 mL of HEPES buffered saline (HBS). BNImEtPIX was first converted to its sodium salt "BNlmEtPIXNa" by dissolving BNImEtPIX (-15 mg) in HC1 (aq) and then neutralizing the solution with dilute NaOH(aq) and NaHCC«3(aq). The resulting porphyrin precipitate was filtered off and rinsed with 1-2 mL of distilled H2O to remove residual NaHC03 (some porphyrin was lost in the filtrate). "BNlmEtPIXNa" was washed from the filter with distilled water (-15 mL) into a flask, and the solvent was removed under reduced pressure. "BNlmEtPIXNa" (2.5 mg) and EPC (50 mg) were dissolved in NMP (0.5 mL) as described above. Then 200 p:L of the NMP solution were added to HBS (4 mL) while the HBS solution was vortexed, this producing a -300 [iM stock solution. 6.2.2.3 Other Formulations Cremophor E L ® emulsions Preparation of the Cremophor EL® emulsions was based on published procedures.16 BNImEtPIX (0.9 mg, 1.1 |imol) was dissolved in -0.5 mL of DMF and to this solution was added Cremophor EL® (2.2 mL) and propylene glycol (0.2 mL); the solution was mixed thoroughly by stirring, gende warming and sonication. The DMF was removed under reduced pressure; the resulting oil was emulsified in a-/- medium (20 mL) by vortexing and sonication, and the pH was adjusted to 7.3 with dilute NaOH(aq). Some of the porphyrin precipitated; this complicated the subsequent filtration/sterihzation. The final concentration of BNImEtPLX was determined to be -9 pJVl by UV-Vis spectroscopy. In addition, one toxicity experiment was performed using the above procedure, but substituting sterile PBS for the a-/- medium; the final concentration of BNImEtPIX was 0.5 mM. A Cremophor EL® emulsion with BHEtPIX (-50 (0,M), and a control emulsion without porphyrin, were prepared in a similar manner. 10% D M S O formulation Preliminary reports on the photosensitizer benzoporphyrin derivative (BPD) utilized 10% DMSO to solubilize the compound for in vitro assays.17 However, when BNImEtPIX was dissolved in DMSO and diluted with a-/- medium, the porphyrin precipitated from solution and the References on page 233 208 solution could not be filtered. When PBS was the solvent, the solubility was greater and a 0.5 mM solution was prepared; BNImEtPIX (9 mg, 0.01 mmol) was dissolved in DMSO (2 mL), and to this solution was added PBS (20 mL). 6.2.3 Toxicity Assays The toxicity assays were performed under 1 atmosphere of air or N2 according to a standard protocol (see Figure 6.1).18 A sterile solution or formulation of the porphyrin to be tested (Section 6.2.2) was added to a+/- or a-/- medium to yield a total of 9 mL of solution/suspension. Controls without the porphyrin, including Cremophor EL® and liposomal formulations (Method II), were treated in a similar manner. The 'toxicity vessels' were shaken in a Labline instruments "Orbit Shaker Bath", maintained at 37 °C in a warm room. Humidified air or N2, for oxic and hypoxic conditions, respectively, was passed over the solutions for 1 h before adding the cells, and was maintained throughout the experiment. Approximately 2 x 106 cells, harvested from the spinner culture, were added to each of the vessels as suspensions in ~ 1 mL of a-/- or a+/-medium. porphyrin stock solution O O0O 0 CHO cells wrap in Al foil air orN2 flow incubate @ 37 °C sample at ' 0,1,2, 3 h plate incubate 1 week count colonies (37 °C, 19:1 air:C02) centrifuge Figure 6.1. In vitro Toxicity Assay. The control vessels were sampled (1 mL) at 0, 1, 2 and 3 h after addition of the cells, while the porphyrin-containing samples were taken at 1, 2, and 3 h. When pipetting from the hypoxic vessels, a small amount of gas was taken from the vessels before sampling to avoid introduction of References on page 233 209 air into the vessels. Each 1 mL sample was diluted in a-/- medium (9 mL); the mixtures were vortexed and centrifuged; the supernatant was decanted and the cells were resuspended in oc-/-medium (10 mL). Aliquots (10, 100, and 1000 p:L) from these cell suspensions were plated in duplicate into separate 5 cm Petri dishes which contained a+/+ medium (5 mL). The concentration of cells was determined using a Coulter Counter. The dishes were incubated for 1 week, when the colonies were stained with methylene blue and counted. The toxicity is expressed in terms of the plating efficiency (PE = number of colonies / number of cells plated) on a log scale as a function of time (see Section 6.3.2). 6.2.4 Radiosensitizing Assays The radiosensitizing assays were based on literature procedures.18 A sterile solution or formulation of the porphyrin to be tested (Section 6.2.2) was added to a+l- or a-l- medium (total 14 mL) in a glass "duck" irradiation vessel, and the mixture was stirred magnetically (see Figure 6.2). Controls without the porphyrin, including liposomal formulations (Method II), were treated in a similar manner. The "ducks" were kept at 37 °C and humidified N 2 was passed over the solutions for ~1 h. Then 3 x 106 cells (suspended in 1 mL of medium) were added to each of the vessels, and the mixtures were stirred for 1 h maintaining the temperature and N 2 flow. The "ducks" were then immersed in ice and a zero dose sample (1 mL) was taken from each "duck". The "duck" was then irradiated with an X-ray source (Phillips, 250 kV, 0.5 mm Cu) in 5 Gy increments up to 25 Gy, with 1 mL samples being taken after each 5 Gy dose. Radiation doses were measured using a Precision Electrometer (Victoreen, Model 500). N 2 flow was continued and precautions were taken to ehminate air contamination during sampling which would sensitize the cells to radiation. Each 1 mL sample was diluted in a+l- medium (9 mL), and the mixtures were vortexed and centrifuged; the supernatant was decanted and the cells were resuspended in oc-/- medium (10 mL). The plating, cell counting, incubation and colony counting followed the procedures described in Section 6.2.3, but the aliquots for plating were 10, 20, 50, 200, 500 and 1000 [ih for the 0, 5, 10, 15, 20 and 25 Gy samples, respectively. References on page 233 210 The results of the experiment are expressed in a dose-response curve by plotting the surviving fraction (SF = PE at dose D / PE at zero dose) (on a log scale) as a function of dose (in Gy) (see Section 6.3.3). The effectiveness of a sensitizer is expressed by its sensitizer enhancement ratio (SER = dose at 0.01 SF for control / dose at 0.01 SF for 'drug'). porphyrin stock — solution centrifuge irradiate (X-rays, 5, 10, 15,20, 25 Gy) Figure 6.2. In vitro Radiosensitizing Assay. 6.2.5 Photosensitizing Assays The photosensitizing assays were based on procedures described in the literature (see Figure 6.3).1 1 , 1 9'2 0 Solutions or liposomal formulations of the porphyrins (19 mL, Section 6.2.2) and 4 x 106 cells (in 1 mL) were added to the toxicity flasks and shaken in air for 1 h as described in Section 6.2.3. The contents of each flask were then transferred to centrifuge tubes and the cells were separated by centrifugation and washed three times with cold, sterile PBS (10 mL each, one wash = resuspension in cold PBS by a vortexing, centrifuging, and decanting of the supernatant). Each set of cells was suspended in ~ 1 mL of cold PBS kept at O °C. The irradiation treatment (630 ±10 nm) was dehvered from a tunable light source equipped with a 1 kW xenon bulb (Model A 5000; Photon Technology International, Inc.) through a 5-mm core diameter liquid light guide (1000A; Luminex, Munich, Germany). Before each experiment, References on page 233 211 the power output of the light source was measured, and the dose rate (typically 1 J/cm2 / 70 s) was calculated for a 19 mm diameter circle (-diameter of the drop on the Petri dish, see Figure 6.3). Al l cell-suspension transfers and irradiations were carried out in a dimly lit room to avoid inadvertent light exposure. For each cell suspension, a zero-dose sample was taken (100 |iL) and added to a+l- medium (10 mL). Then for each dose (0.2, 0.5, 1, 2, 5 and 10 J/cm2), 120 uL of the cell suspension were placed on a Petri dish and then placed in the circle of light for the required time to achieve the desired dose. After irradiation, 100 (iL of the cell suspension were removed from the drop on the Petri dish and added to a+l- medium (10 mL); these suspensions were centrifuged and the cells were resuspended in a+l- medium (10 mL), plated and counted as described for the toxicity assay. The results of the experiment are expressed as a dose-response curve of SF (Section 6.2.4) versus dose (in J/cm2). irradiate (630 nm, e.g. 4.5 mW/cmz) Figure 6.3. In vitro Photosensitizing Assay. 6.2.6 Porphyrin Accumulation in Cells Accumulation measurements using the cells from the toxicity, radiosensitizing and photosensitizing experiments were based on literature procedures, but UV-Vis spectroscopy was References on page 233 212 used to measure drug content in place of fluorescence spectroscopy;21 the latter spectroscopy was also used, but the high porphyrin concentrations permitted measurement by UV-Vis spectroscopy, which was more convenient. After completion of the toxicity or radiosensitizing assays, each of the remaining cell suspensions from the primary experiment vessels was centrifuged. The supernatant was decanted, and the cells were washed three times with PBS (the cells from the photosensitizing experiments had already been washed this way). A sample of the cells from each suspension was counted prior to the last centrifugation, after which the remaining cells were solubilized (digested) by adding 10 % ScintiGest in distilled H2O (5 mL); the containers were left in the dark at 37 °C for 36-48 h. The accumulation of the porphyrin into the cells was determined as p:mol of porphyrin per million cells (see Section 6.3.5). 6.2.7 Fluorescence Microscopy Procedures to visualize the localization of the porphyrins within the cells were adapted from those of Matthews et al.21 Approximately 5 x 104 cells (in -100 \\L a-/- medium) were deposited onto a sterilized microscope slide (two for each porphyrin solution or formulation) contained in a Petri dish, and the dishes were incubated at 37 °C for 15 min to allow cell-adhesion to occur. Then solutions of OPyTrSPhP (0.5 mM) or BHEtPIX (0.1 mM) in a-/- medium, or BNImEtPLX (30 MM), APhTrPhP (30 |iM) or TirapPhTrPhP (22 pjvl) as liposome formulations in a-/- medium, were added to each Petri dish such that the slides were completely submerged, and the dishes were incubated 1 h at 37 °C. One slide from each dish was removed, washed 3x with cold PBS and placed in a solution of Hoechst 33258 (1 x 10"6 M) for 1 h at 0 °C. Then the stained slide and the unstained, duplicate slide were washed 3x with cold PBS, covered with a cover slip, and kept at 0 °C prior to analysis by fluorescence microscopy. Each slide was mounted on a Cytosavant image cytometry platform (Oncometrics Imaging Corp., Vancouver, B.C.) which is capable of acquisition in both fluorescence and absorbance configurations. Several photos were taken in the fluorescence mode from different locations on the slide to observe different cell clusters. Fluorescence from the chromophores localized in the cells was observed when the slides were irradiated with UV light (330-380 nm excites both the Hoechst References on page 233 213 and porphyrin chromophores, with emission measured at > 420 nm) and in the green region (510-560 nm, which excites only the porphyrin chromophore, with emission measured at > 590 nm). The integration procedure required for the green-excitation was ~6x longer than with the UV excitation. The fluorescence data are presented in Section 6.3.6. 6.3 Results and Discussion 6.3.1 Porphyrin Formulations The evaluation of the porphyrins, with the exception of the water-soluble sulfonatophenyl porphyrins, was hindered by their poor solubility in aqueous media. Similar difficulties with other lipophilic, aggregating macrocycles have been circumvented by formulating them as Cremophor EL® emulsions16'23"25 or incorporating them into liposomes.3'26"30 In this thesis work, BHEtPIX was evaluated as a Cremophor EL® formulation for one toxicity assay and in a-medium solutions for the remaining assays. BNImEtPIX, APhTrPhP and TirapPhTrPhP were evaluated as liposomal formulations. Experiments were also performed with BNImEtPIX as Cremophor EL® emulsions in a -/- and PBS. The hposomes were prepared and their average diameter measured on the same day of the experiment studying toxicity, radiosensitization, or photosensitization. Vesicle diameters from one experiment are summarized in Table 6.2 and are typical of those observed in the other preparations. Method I (used for APhTrPhP only) consistently gave small diameter vesicles, and Method II (used for TirapPhTrPhP, BNImEtPIX and control) yielded much larger vesicles and much broader distributions. Table 6.2. Liposome Size Distributions (in nm) from One Preparation. Formulation Preparation Method Mean diameter Standard deviation EPC blank in HBS n 859 734 APhTrPhP i 118 49 TirapPhTrPhP n 712 528 BNImEtPIX n 1147 983 References on page 233 214 6.3.2 Toxicity Media solutions and liposomal formulations In general, the porphyrins at the concentrations tested were non-toxic to CHO cells in air and in N 2 . Typical results are presented in Figure 6.4. The results with OPyTrSPhP (not shown) parallel those reported earlier for similar compounds (e.g. PyTrSPhP);12'20 no corresponding systems of tirapazamine- or nitroimidazole-containing porphyrins encapsulated in liposomes have been reported in the literature. The lack of toxicity, including hypoxia-selective toxicity, for compounds such as BNImEtPIX and TirapPhTrPhP suggests that the nitroimidazole and tirapazamine moieties were not enzymatically reduced or that the 'activated' form did not reach the DNA, the presumed site of hypoxia selective toxicity (Section 1.4.2). " J U3 a o.i H 0.01 • Control (no porphyrin) Liposome Control (NMP liposome formulation) APhTrPhP formuladon — & TirapPhTrPhP formulation BNImEtPIX formulation TIME (h) Figure 6.4. Lack of Toxicity of Some Liposome-formulated Porphyrins in Aerobic CHO Cells. When precautions were not taken to minimize light exposure during the toxicity experiments, some light-induced toxicity was observed with cells that were incubated with OPyTrSPhP. The results of the photosensitizing assays with OPyTrSPhP are presented in Section 6.3.4. References on page 233 215 Cremophor E L ® emulsions The assays with Cremophor EL® emulsions of BNImEtPIX and BHEtPIX in a -/-medium showed some toxicity (see Figure 6.5). Cremophor EL® is known to be biologically active and can inhibit mitosis,31'32 but the reason for the increased toxicity of Cremophor EL® emulsions of BNImEtPIX and BHEtPIX compared to that of the Cremophor EL® control is not known. When the toxicity of BNImEtPIX was evaluated in PBS, no viable cells were present after 1 h incubation in the controls or in the porphyrin-containing solutions; this is not surprising, as cell-incubation at 37 °C in PBS is known to be toxic.33 Because of the inherent toxicity of the Cremophor EL®, and the fact that no cell accumulation of the porphyrins was observed in this medium (data not shown), further tests using Cremophor EL® emulsions were not pursued. o . H OH 0 .01 -J 0.001 8 O - A -N 2 Control (no porphyrin) Cremophor EL® control BNImEtPIX / Cremophor EL® emulsion BHEtPIX / Cremophor EL® emulsion TIME (h) Figure 6.5. Toxicity in Hypoxic CHO Cells of BNImEtPIX and BHEtPIX Formulated as Cremophor EL® Emulsions. 6.3.3 Radiosensitization In general, the evaluated porphyrins at the concentrations tested did not radiosensitize hypoxic CHO cells. Typical results (e.g. for OPyTrSPhP) are presented in Figure 6.6. The slight deviations from the control curve are not significant,33 and generally these results parallel those References on page 233 216 observed with other porphyrins and metalloporphyrins.11'12'20 From these results and the cyclic voltammetry data presented in Chapter 5 (Section 5.3.4), it is concluded that the pyridine N-oxide moiety is a poor radiosensitizer or it does not localize at DNA, the target of radiation therapy (see Section 6.3.7). 1 -[ 0.1 -0.01 -— • Control (no porphyrin) ""^S -<> 333 um — •-•>— 500 urn 1000 um •9 0 001 -0 • i 1 5 10 15 Dose (Gy) • i 20 25 Figure 6.6. Lack of Radiosensitization by OPyTrSPhP in Hypoxic CHO Cells. One exception was TirapPhTrPhP which showed a modest SER value of -1.5 (Figure 6.7), but these results are from only one experiment and the phenomenon needs to be confirmed, considering that TirapPhTrPhP showed no toxicity, including hypoxia-selective toxicity (Figure 6.4). However, radiosensitizers are not always HSCs (see Section 6.3.7) Tirapazamine itself has been shown to be an effective radiosensitizer in vivo with reported SER values of 1.5 to 3.O.34 However, these values cannot be compared directly to those of TirapPhTrPhP because of the differences in experimental procedures, as Brown and Lemmon34 irradiated tumors in mice in vivo, and either measured delay in regrowth or subsequently excised the tumors and the cells plated in a clonogenic assay in vitro. References on page 233 217 0.01 A o . H TirapPhTrPhP 0.001 o o o in Dose (Gy) Figure 6.7. Radiosensitization of Hypoxic CHO Cells by Liposome-formulated TirapPhTrPhP. That no radiosensitization was observed with BNImEtPIX suggests this compound does not reach the DNA target, even though measurable amounts of the porphyrin were accumulated by the cells (see Section 6.3.5). Nevertheless, radiosensitization by nitroimidazoleoporphyrins is promising; Sakata et al. reported radiosensitization in vitro and in vivo by porphyrins and metalloporphyrins containing nitroimidazole moieties (e.g. KATD-F1 showed an SER of 1.39 in HeLa cells in vitro).35 Suggestions for improving the aqueous solubility of porphyrins incorporating a nitroimidazole moiety to produce effective radiosensitizers are presented in Chapter 7. HOOC COOH KATD-F1 References on page 233 218 6.3.4 Photosensitization Sulfonatophenyl porphyrins Of the porphyrins synthesized in this work, only OPyTrSPhP and PyTrSPhP demonstrated phototoxicity under the conditions used here. The results of the assays with OPyTrSPhP, PyTrSPhP, pyridine N-oxide and Photofrin® are presented in Figure 6.8. At doses > 0.5 J/cm2 and > 5 J/cm2 (data not shown), no viable cells remained from those incubated with Photofrin® and the sulfonated porphyrins, respectively. Pyridine N-oxide alone (control) showed no photo-induced toxicity in doses of up to 10 J/cm2. OPyTrSPhP was a less effective photosensitizer than PyTrSPhP, but the photosensitizing effectiveness of PyTrSPhP decreased when an equivalent amount of pyridine N-oxide was added. These results suggest a protective effect of pyridine N-oxide, but further experiments would be required to substantiate this hypothesis. 10 -| 1 0.01-1 \ [ • p y T r S P P 1 \ O PyTrSPhP + pyridine N-oxide - - "A- - - Photofrin ® \ ~-~fU--~ pyridine N-oxide 0.001 -I $ 1 1 , 0 0.5 1 1.5 2 Dose (J/cm )^ Figure 6.8. Phototoxicity of OPyTrSPhP (0.5 mM, -0.57 mg/mL), PyTrSPhP (0.5 mM, -0.58 mg/mL), Pyridine N-oxide (0.5 mM) and Photofrin® (2.5 ng/mL) in CHO Cells. References on page 233 219 The results obtained here with Photofrin® match well those expected under these conditions;19 however, in comparison PyTrSPhP and OPyTrSPhP were not nearly as effective. Using similar sulfonatophenyl porphyrins (TSPhP and 5-phenyl-10,15,20-tris(4-sulfonatophenyl)porphyrin (PhTrSPhP)), Kessel et al. have described photosensitizing experiments with Murine leukemia L1210 cells and achieved similar results to those obtained here (low toxicity using relatively high porphyrin concentrations).36 Differences between Photofrin® and the sulfonated porphyrins is unlikely to be due to large differences in *02 production, as TSPhP produces *02 more efficiently than Photofrin® in PBS with or without EPC liposomes, although their efficiencies are comparable in the presence of bovine serum albumin.37 It is more likely that Photofrin® and the sulfonated porphyrins exhibit different cellular accumulation and localization properties (Sections 6.3.6 and 6.3.7). In vitro and in vivo experiments using PhTrSPhP and TSPhP have shown that these porphyrins bind well to serum albumin36 and accumulate to a high degree in the extracellular tumor stroma.2 However, because these sulfonatophenyl porphyrins have not been as effective as other photosensitizers3 and because of the results in this thesis work, further investigation of OPyTrSPhP as a photosensitizer is probably not warranted. BHEtPIX, BNImEtPIX, APhTrPhP and TirapPhTrPhP The remaining porphyrins, BHEtPIX, BNImEtPIX, APhTrPhP and TirapPhTrPhP, did not appreciably photosensitize CHO cells under the conditions used in this thesis work (data not shown) despite having been accumulated by the cells to a greater degree than OPyTrSPhP or PyTrSPhP (see Section 6.3.5). Upon initial inspection, the poor photosensitizing ability of BHEtPIX is puzzling considering its similarity to Photofrin® in structure and mode of delivery, but Photofrin® is a mixture of products and not all of the components are equally effective at sensitizing cells.3 , 3 8 A better comparison for BHEtPIX can be made with hematoporphyrin (HP, below) given their similarity in composition. Woodburn and co-workers have shown that HP is a relatively poor photosensitizer of C6 glioma and V79 Chinese Hamster lung fibroblast cells in vitro}9 More effective photosensitizers References on page 233 220 based on PLX-derivatives were less polar than HP (based on their octanol/PBS partition coefficients) and had tertiary amino groups which are protonated (cationic) at physiological pH. Woodburn et al. evaluated the tumor and tissue distribution of the same porphyrin series in vivo and, although the most promising porphyrins in terms of selective tumor accumulation were also cationic at physiological pH, the best candidates were not the same as those identified by their in vitro photosensitizing ability and toxicity.40 These apparendy contradictory results illustrate the difficulty in predicting the in vivo effectiveness of photosensitizers from data obtained in vitro.41 Based on the behavior of HP reported in the literature and the results obtained in this thesis work, BHEtPIX is not a promising photosensitizer. Also, the inital hypothesis of comparing the in vitro results of BNImEtPIX with those of BHEtPIX is called into question based on the wide variability of photosensitizing results with side-chain variations.39 HOOC COOH HOOC COOH HOOC COOH HP PIX BHEtPIX Many examples of liposome-bound photosensitizers using numerous formulations and cell lines have appeared in the literature,3'27'29 ,30'42"45 most of them being applied to studies in vivo. However, in in vitro studies of mitochondria photosensitization by HP and protoporphyin (PIX), Ricchelli et al. showed that delivery via dipalmitoyl phosphatidylcholine (DPPC) liposomes increased the photosensitizing effectiveness of PIX, but decreased that of H P . 3 0 , 4 4 Addition of 20 % cardiolipin to the DPPC liposomes increased the effectiveness of PIX further. In other in vitro studies, Reich et al. demonstrated that tetramethylhematoporphyrin delivered in DPPC liposomes effectively photosensitized human bladder carcinoma cells.29 Toledano and Kimel studied a series of liposome-bound porphycene photosensitizers in vitro and the efficiency of the photosensitizer References on page 233 221 varied according to its location in the liposome; those 'buried' inside the hposome bilayer were not efficient photosensitizers.45 Thus, based on the literature reports, liposome delivery of photosensitizers can be effective in vitro, but the photosensitization efficiency may depend on the composition of the photosensitizer and the liposome formulation. That the hposome-bound porphyrins in this work (APhTrPhP, TirapPhTrPhP, BNImEtPIX) were not effective photosensitizers might be related to accumulation, localization, and aggregation phenomena (Sections 6.3.5 and 6.3.6). The effectiveness of photosensitizers has been correlated with several physico-chemical parameters (e.g. octanol-water partition coefficient, aggregation properties, efficiency of 102 production, etc.).2 ,39'41 The evaluation of these parameters was not addressed in this thesis work, but such information would have been useful had the evaluated porphyrins shown promise as photosensitizers. Nitroimidazoles and PDT Nitroimidazoles have been reported to interact with chlorins and porphyrins to quench both the singlet and triplet excited states of these compounds.46"49 Triplet state quenching by nitroimidazoles yields the porphyrin or chlorin radical cation and the nitroimidazole radical anion, and indeed this type of interaction has been proposed as a mechanism for oxygen-independent Type I photodynamic processes, which could possibly make PDT effective for treating hypoxic tumors (see Chapter l ) . 5 0 However, in vitro studies with EMT-6 mouse tumor cells have suggested that the Type I photodynamic processes are not the dominant mode of action and that nitroimidazoles effectively inhibit photosensitization by Photofrin® at low oxygen tensions.51 In contrast, other in vitro and in vivo studies using a nitroimidazole-chlorin-e6 combination with Ehrlich carcinoma cells demonstrated an increased phototoxicity under reduced oxygen tensions.52 Also, such PDT-nitroimidazole combinations may be effective even if they do not act via Type I photodynamic processes. Other in vivo studies have shown enhanced tumor response using such a combination, and the enhancement was attributed to the hypoxia-selective toxicity of nitroimidazoles in PDT-potentiated tumor ischemia.53,54 Thus, despite poor photosensitizing results with BNImEtPIX here, the investigation of nitroimidazoleporphyrins as photosensitizers References on page 233 222 (and/or radiosensitizers and HSCs) remains of interest. Suggestions for improving the solubility of such compounds are presented in Chapter 7. 6.3.5 Porphyrin Accumulation in Cells The accumulation of the porphyrins by the cells was measured to help evaluate their potential as anti-cancer agents. The results in this section describe overall accumulation and appear in Figure 6.9. Because Photofrin® is known to be a mixture of compounds,3 the number of 'Hmol' of porphyrin moieties for this compound was estimated using the £ value for BHEtPIX. 0.0125 Porphyrin Incubation Concentration Figure 6.9. Results from the Accumulation Experiments (* = liposomal formulation); n = the number of samples with one porphyrin system. According to these results, BNImEtPIX was accumulated by the cells more than any other porphyrin at the given concentrations, although clearly Photofrin® was accumulated to the greatest References on page 233 223 extent and the sulfonatophenyl porphyrins to the lowest extent per |ig of porphyrin delivered. The liposomal formulations appear to be effective delivery vehicles; however, the possibility exists that some fraction of the liposomes aggregated with the cells and was not 'washed out' with the PBS washes (Section 6.2.6), and this would give falsely large accumulation values (see Sections 6.3.6 and 6.3.7). Korbelik and Hung quantitatively evaluated Photofrin® accumulation and the effect of human plasma proteins in an HeLa cell line; in cells incubated with 40 p:g/mL in the absence of serum, the accumulation value was -2.9 |ig/million cells.21 Using the results presented in Figure 6.9, -3.9 and -4.7 |ig/million cells of BHEtPIX and Photofrin® (using the molecular weight of BHEtPIX as an estimate of the mass per chromophore) were accumulated by the CHO cells, respectively. The slightly higher accumlation values obtained here might be related to differences in cell hne, experimental procedure and the inaccuracy of using the £ and molecular weight of BHEtPIX to calculate the Photofrin® values. Numerous examples of hposome-delivered porphyrins and other photosensitizers have been reported in the literature,3'27'29'30'42"45 but accumulation is typically quantified only for in vivo and not in vitro studies. Thus, it is difficult to compare directly the accumulation values obtained here to those in the literature. Nevertheless, liposomes are an effective means of delivery of hydrophobic porphyrins. The large accumulation value of 100 |imol / million cells for PyTrSPhP (incubation concentration = 100 pJvI) obtained by Meng and Meng et al. using HT-29 cells 1 1 , 2 0 is erroneous, because this number is > 50 times the total number of moles in the incubation solution! A mathematical error or perhaps hmitations of fluorescence spectroscopy using porphyrin calibration solutions of high concentration (0.1 to 5 mM) is likely to blame for this error. 6.3.6 Fluorescence Microscopy Fluorescence microscopy has been used for monitoring and evaluating the accumulation and subcellular localization of porphyrins within cells.3 6'3 8-5 5"6 0 In this thesis work, fluorescence References on page 233 224 microscopy was used as a qualitative measure of porphyrin accumulation to compliment the results of Section 6.3.5. As described in Section 6.2.7, the UV-excitation induces fluorescence in both the Hoechst stain and the porphyrin, whereas the green-excitation induces only porphyrin fluorescence. Under the microscope, the light emitted by the fluorescing Hoechst stain was light-blue as located in the DNA 6 1 and that by the porphyrins was red. The intensity of the fluorescence of the porphyrin was much lower than that of the Hoechst stain and longer integration times were needed to capture the image with green-excitation (or with UV-irradiation of cells which were not exposed to the Hoechst stain). In general, the auto-fluorescence of the cells without dye was minimal in comparison to that stained with Hoechst or a porphyrin. OPyTrSPhP Images of CHO cells after incubation with OPyTrSPhP via UV-excitation and green-excitation appear in Figures 6.10 and 6.11, respectively (same field). Fluorescence from OPyTrSPhP is observed in the plasma membrane and in bright dots throughout the cytoplasm which may indicate lysosomal localization. Fluorescence from OPyTrSPhP is also observed to a lesser degree in the cytoplasm and the nucleus. These results are generally consistent with those reported by Berg et al, who studied the accumulation and subcellular localization of TSPhP (incubation concentration: 75 |ig/mL) in a human cervix carcinoma cell line (NHIK 3025).56'57 Kessel et al. reported that PhTrSPhP and TSPhP were found at 'pin-point' loci on the plasma membranes in Murine leukemia L1210 cells.36 Such loci are also observed in the fluorescence microscopy image of OPyTrSPhP (Figure 6.11). References on page 233 225 A 1 # Hi 1 i t • Figure 6.10. Fluorescence Microscopy Image of OPyTrSPhP (0.5 mM)-Incubated CHO Cells using UV-Excitation. References on page 233 226 Figure 6.11. Fluorescence Microscopy Image of OPyTrSPhP (0.5 mM)-Incubated CHO Cells using Green-Excitation. References on page 233 227 APhTrPhP, TirapPhTrPhP, BNImEtPIX A few cellular localization studies using liposome-delivered photosensitizers have been reported in the literature. Using fluorescence microscopy, Reich and co-workers studied the accumulation of tetxamemylhematoporphyrin delivered in DPPC liposomes by bladder carcinoma cells.29 They observed porphyrin fluorescence in the cytoplasm, plasma and nuclear membranes, but no fluorescence was observed from the nucleus. In another report, Poretz et al. used density gradient centrifugation to show that free pheophorbide a (PhA) was concentrated in the plasma membrane of human bladder tumor cells, whereas liposome-delivered PhA was concentrated in the lysosomes when incubated with the cells at 4 °C. However, when the liposome-delivered PhA was incubated with the cells at 37 °C, the liposomes fused with the plasma membrane resulting in preferential PhA accumulation at this site. Based on these studies, liposome-delivery can be an effective means to deliver photosensitizers in vitro. o O-C C 0 2 M e OH PhA In the liposome-delivered porphyrins evaluated here, fluorescence microscopy (green excitation) showed faint cell-outlines in the TirapPhTrPhP-incubated cells (not shown) and a noticeable background fluorescence in the form of dots, a pattern that matches that seen for the liposomal formulation of TirapPhTrPhP (Figure 6.12.). Similar results were seen for APhTrPhP-incubated cells, but individual liposomes were not observed, only a general background fluorescence being seen. The differences in size and size-distribution between the TirapPhTrPhP and APhTrPhP formulations (Section 6.3.2) are consistent with these observations, as the TirapPhTrPhP formulation has vesicles large enough to be seen by microscopy, whereas the References on page 233 228 APhTrPhP formulation does not.33 With BNImEtPLX-incubated cells, no significant fluorescence was observed using green excitation. In all cases, the Hoechst stain was observed in the nucleus. The poor cellular accumulation of these porphyrins evidenced by fluorescence microscopy appears to contradict the results presented in Section 6.3.5. One possible explanation stems from hposome aggregation when the cells were prepared for the accumulation measurements (Section 6.3.5). Also, if the porphyrins are highly aggregated once they have been accumulated by the CHO cells, the fluorescence could be quenched and no signal would be observed.37 This is likely the case with BNImEtPIX, as this porphyrin was poorly soluble in many solvent combinations (Section 4.3.2.3); however, as a dilute solution in NMP in which aggregation should be minimal, fluorescence at -625 nm was observed using excitation at 400 nm. B H E t P I X The fluorescence microscopy image of BHEtPTX-incubated cells clearly shows accumulation of this porphyrin with most of the material appearing in the cell membrane along with a diffuse cytoplasmic distribution (Figure 6.13). According to Woodburn et al, the structurally similar hematoporphyrin demonstrated a diffuse cytoplasmic and lysosomal localization,38 which generally agrees with the results obtained here. In cells with no Hoechst stain, the fluorescence disappeared after 10 s UV-excitation indicating bleaching of the chromophore.29,49'62 References on page 233 229 Figure 6.12. Fluorescence Microscopy Image of the TirapPhTrPhP Liposome Formulation (225 p:M) Using Green-Excitation. References on page 233 230 Figure 6.13. Fluorescence Microscopy Image of BHEtPIX-Incubated CHO Cells (100 | iM, no Hoechst Stain) using Green-Excitation. References on page 233 231 6.3.7 Discussion of In Vitro Results Radiosensitizers and hypoxia-selective cytotoxins (HSC) act by damaging DNA via 'fixation' of damage caused by ionizing radiation or via activation by endogenous enzymes, respectively (see Chapter l). 1 Some compounds {e.g. misonidazole, tirapazamine) are effective as both radiosensitizers and HSCs, but this is not always the case. o_ T O H N + N H 2 N0 2 O -misonidazole tirapazamine BNImEtPIX and TirapPhTrPhP both contain moieties that could act as either radiosensitizers or HSCs. Thus, based on their poor efficacy in vitro, it can be concluded that the active moieties did not localize at the DNA (lack of radiosensitization), they were not enzymatically reduced, or did not reach the DNA in their activated form (ineffectual HSCs). The method of delivery (liposomal formulation) may have been a factor in this as the fluorescence microscopy studies indicated that these porphyrins were not accumulated well by the cells (Section 6.3.6). The water-soluble OPyTrSPhP was considered a promising compound because similar sulfonatophenyl porphyrins have been shown to be accumulated by cells in vitro.11'36'56'51 However, the cyclic voltammetry results (Chapter 5) indicated that the reduction potential of the pyridine N-oxide group hes outside the useful potential range for HSCs and radiosensitizers. The other porphyrins also showed no significant radiosensitizing or hypoxia-selective activity, but such activity was not expected, as they did not incorporate known bioactive moieties. PDT incurs extensive damage at the cellular membranes, including the plasma, mitochondrial and nuclear membranes; also, liberated lysosomal enzymes have been implicated as being part of the cytotoxic effects of PDT (Section 1.4.3).50 Localization at the DNA does not appear to be particularly important for effectiveness as a photosensitizer.2 Fluorescence from OPyTrSPhP was observed in the membranes and possibly in the lysosomes of CHO cells References on page 233 232 incubated with this porphyrin (Section 6.3.6); OPyTrSPhP did photosensitize the CHO cells, but the pyridine N-oxide group appeared to mitigate the photoinduced toxicity. OPyTrSPhP and PyTrSPhP, the two porphyrins with obvious photosensitizing ability, were not nearly as effective as Photofrin® as photosensitizers, and further investigation of these two porphyrins as photosensitizers is not warranted. The other porphyrins did not show significant photosensitizing ability, which might have been related to their lack of accumulation by CHO cells. 6.4 Summary In vitro assays with CHO cells were used to evaluate the potential of OPyTrSPhP, TirapPhTrPhP, and BNImEtPIX, along with their 'reference' compounds PyTrSPhP, APhTrPhP, and BHEtPIX, as HSCs, radiosensitizers and photosensitizers, and the results were compared to the literature data. For the photosensitizing assays, two reference compounds (Photofrin® and pyridine N-oxide) were also investigated. The cellular accumulation of these porphyrins was evaluated by UV-Vis spectroscopy and that of selected porphyrins by fluorescence microscopy. In order to perform the assays, APhTrPhP, TirapPhTrPhP, and BNImEtPIX were formulated in liposomes, whereas the other compounds were evaluated as solutions in a-modified medium. Toxicity assays were also performed with Cremophor EL® emulsions of BNImEtPIX and BHEtPIX. In general, the porphyrins were non-toxic, and they showed little radiosensitizing or photosensitizing ability. However, TirapPhTrPhP showed a modest radiation SER value (1.5) and some photosensitization was observed with OPyTrSPhP and PyTrSPhP. The effectiveness of OPyTrSPhP and PyTrSPhP as photosensitizers was poor in comparison to that of Photofrin®. Some evidence was seen for protection from PDT-induced damaged by pyridine N-oxide. With the exception of TirapPhTrPhP, which demonstrated radiosensitizing properties, none of the evaluated porphyrins from this thesis work probably warrants further investigation as an anti-cancer agent. Based on the accumulation in cells measured by UV-Vis spectroscopy, BNImEtPIX was accumulated to the greatest degree but, per microgram of porphyrin delivered, Photofrin® was References on page 233 233 accumulated the most and the sulfonatophenyl porphyrins the least. The accumulation data for the liposome-formulated porphyrins obtained via UV-Vis measurements appear to conflict with those from the fluorescence microscopy, but the results could be explained by liposome aggregation during cell preparation. References for Chapter 6 (1) Denny, W. A.; Wilson, W. R.; Hay, M . P. Br. J. Cancer 1996, 74, S32-S38. (2) Pass, H. J. Nat. Cancer Inst. 1993, 85, 443-456. (3) Boyle, R.; Dolphin, D. Photochem. Photobiol. 1996, 64, 469-485. (4) Moan, J.; Berg, K. Photochem. Photobiol. 1992, 55, 931-948. (5) Chan, P. K.-L. Ph.D. Thesis, Univ. of British Columbia, 1988. (6) Chan, P. K.-L.; James, B. R.; Frost, D. C ; Chan, P. K. H.; Hu, H.-L.; Skov, K. Can. J. Chem. 1989, 67, 508-518. (7) Yapp, D. T. T. Y. Ph.D. Thesis, Univ. of British Columbia, 1993. (8) Yapp, D. T. T.; Rettig, S. J.; James, B. R.; Skov, K. A. Inorg. Chem. 1997, 36, 5635-5645. (9) Ravensbergen, J. R. M.Sc. Thesis, Univ. of British Columbia, 1993. (10) Ware, C. M.Sc. Thesis, Univ. of British Columbia, 1994. (11) Meng, G. G. Ph.D. Thesis, Univ. of British Columbia, 1993. (12) James, B. R.; Meng, G. G.; Posakony, J. J.; Ravensbergen, J. A.; Ware, C. J.; Skov, K. A. Metal-Based Drugs 1996, 3, 85-89. (13) Hope, M. J.; Bally, M . B.; Mayer, L. D.; Janoff, A. S.; Cullis, P. R. Chem. and Phys. of Lipids 1986,40, 89-107. (14) Meng, G. G.; James, B. R.; Skov, K. A. Can. J. Chem. 1994, 72, 1894-1909. (15) Mayer, L. B.C. Cancer Research Centre, 1997, personal communication. (16) Brasseur, N.; Ouellet, R.; Lewis, K.; Ptter, W. R.; van Lier, J. E. Photochem, Photobiol. 1995,62, 1058-1065. 234 (17) Richter, A. M. ; Kelly, B.; Chow, J.; Liu, D. J.; Towers, G. H. N.; Dolphin, D.; Levy, J. G. J. Nat. Cancer Inst. 1987, 79, 1327-1332. (18) Moore, B. A.; Palcic, B.; Skarsgard, L. D. Radiat. Res. 1976, 67, 459-473. (19) Korbelik* M. British Columbia Cancer Research Centre, Personal Communication, 1997. (20) Meng, G. G.; James, B. R.; Skov, K. A.; Korbelik, M. Can. J. Chem. 1994, 72, 2447-2457. (21) Korbelik, M. ; Hung, J. Photochem. Photobiol. 1991, 53, 501-510. (22) Matthews, J.; Adomat, H.; Farrell, N.; King, P.; Koch, C ; Lord, E.; Palcic, B.; Poulin, N.; Sangulin, J.; Skov, K. Br. J. Cancer 1996, 74, S200-S203. (23) Biolo, R.; Jori, G.; Soncin, M. ; Rihter, B.; Kenney, M . E.; Rodgers, M . A. Photochem. Photobiol. 1996, 63, 224-228. (24) Woodburn, K.; Chang, C. K.; Lee, S.; Henderson, B.; Kessel, D. Photochem. Photobiol. 1994,60, 154-159. (25) Kongshaur, M. ; Moan, J.; Cheng, L. S.; Morgan, A. R. Int. J. Biochem. Cell biol. 1995, 27, 481-492. (26) Greenbaum, N. L.; Kappas, A. Pharmacol. 1994, 49, 215-225. (27) Cannon, J. B. J. Pharm. Sci. 1993, 82, 435-446. (28) Ehrenberg, B.; Gross, E.; Lavi, A.; Johnson, F. M. ; Malik, Z. SPIE Optical Methods for Tumor Treatment and Early Diagnosis: Mechanisms and Techniques 1991,1426, 244-251. (29) Reich, E.; Bachor, B.; Miller, K.; Konig, K.; Hautmann, R. SPIE Optical Methods for Tumor Treatment and Detection 1993,1881, 126-140. (30) Ricchelli, F.; Jori, G.; Moreno, G.; Vinzens, F.; Salet, C. J. Photochem. Photobiol. B.-Biol. 1990, 6, 69-77. (31) Liebmann, J.; Cook, J. A.; Mitchell, J. B. Lancet 1993, 342, 1428. (32) Dorr, R. T. Ann. Pharm. 1994, 28, S11-S14. (33) Adomat, H. B.C. Cancer Research Centre, 1997, personal communication. 235 (34) Brown, J. M . ; Lemmon, M. J. Cancer Res. 1990, 50, 7745-7749. (35) Sakata, L; Nakajima, S.; Koshimizu, K.; Hiroyuki, T.; Yasushi, I. JP Patent 080 567 682 A2 960312 Heisei, 1994. (36) Kessel, D.; Thompson, P.; Saatio, K.; Nantwi, K. D. Photochem. Photobiol. 1987, 45, 787-790. (37) Gottfried, V.; Peled, D.; Winkelman, J. W.; Kimel, S. Photochem. Photobiol. 1988, 48, 157-163. (38) Woodburn, K. W.; Vardaxis, N. J.; Hill, J. S.; Kaye, A. H.; Phillips, D. R. Photochem. Photobiol. 1991, 54, 725-732. (39) Woodburn, K. W.; Vardaxis, N. J.; Hill, J. S.; Kaye, A. H.; Reiss, J. A.; Phillips, D. R. Photochem. Photobiol. 1992, 55, 697-704. (40) Woodburn, K. W.; Stylli, S.; Hill, J. S.; Kaye, A. H.; Reiss, J. A.; Phillips, D. R. Br. J. Cancer 1992, 65, 321-328. (41) Kessel, D. SPIE Photodynamic Therapy Mechanisms 1989, 7065, 130-137. (42) Poretz, R. D.; Vucenik, I.; Bergstrom, L.; Segelman, A.; Sigel, J., G.; Chernomorsky, S. SPIE Photodynamic Therapy: Mechanisms 1989,1065, 197-203. (43) Ricchelli, F.; Gobbo, S.; Jori, G.; Salet, C.; Moreno, G. Eur. J. Biochem. 1995, 233, 165-170. (44) Ricchelli, F.; Gobbo, S.; Jori, G.; Moreno, G.; Vinzens, F.; Salet, C. Photochem. Photobiol. 1993, 58, 53-58. (45) Toledano, H.; Kimel, E. S. J. Photochem. Photobiol. B: Biol. 1998, 42, 20-27. (46) Bazin, M. ; Santus, R. Photochem. Photobiol. 1986, 43, I'iS-lAl. (47) Bazin, M . ; Patterson, L. K.; Ronfard-Haret, J. C ; Santus, R. Photochem. Photobiol. 1988,48, 177-180. (48) Kuznetsov, S. V.; Bazin, M. ; Santus, R. J. Photochem. Photobiol. A: Chem. 1997,103, 57-62. (49) Spikes, J. D.; Bommer, J. C. Photochem. Photobiol. 1993, 58, 346-350. 236 (50) Ochsner, M . J. Photochem. Photobiol. B: Biol. 1997, 39, 1-18. (51) Santus, R.; Stobbe, C. C ; McPhee, M . S.; Chapman, J. D. Radiat. Res. 1992, 31-37. (52) Chekulayev, V.; Shevchuk, I.; Mihkelsoo, V.; Kallikorm, A. SPIE Optical Methods for Tumor Treatment and Detection 1992,1645, 131-140. (53) Adams, G. E.; Stratford, I. J. Int. J. Radiat. Oncol. Biol. Phys. 1994, 29, 231-238. (54) Bremner, J. C. M . ; Adams, G. E.; Pearson, J. K.; Sansom, J. M. ; Stratford, I. J.; Bedwell, J.; Bown, S. G.; MacRobert, A. J.; Phillips, D. Br. J. Cancer 1992, 66, 1070-1076. (55) Ruck, A.; Strauss, W.; Schneckenburger, H.; Steiner, R. SPIE Optical Methods for Tumor Treatment and Detection 1992,1645, 284-291. (56) Berg, K.; Western, A.; Bommer, J. C ; Moan, J. Photochem. Photobiol. 1990, 52, 481-487. (57) Berg, K.; Madslien, K.; Bommer, J. C ; Oftebro, R.; Winkelman, J. W.; Moan, J. Photochem. Photobiol. 1991, 53, 203-210. (58) Trepte, O.; Rokahr, I.; Andersson-Engels, S.; Carlsson, K. J. Microsc. 1994,176, 238-244. (59) Yan He, X.; Jacques, S. L.; Gofstein, G. SPIE Optical Methods for Tumor Treatment and Detection 1992,1645, 205-215. (60) Karagianis, G.; Hill, J. S.; Stylli, S. S.; Kaye, A. H.; Varadaxis, N. J.; Reiss, J. A.; Phillips, D. R. Br. J. Cancer 1996, 73, 514-521. (61) Daxhelet, G. A.; Coene, M . M. ; Hoet, P. P.; Cocito, C. G. Anal. Biochem. 1989, 779, 401-403. (62) Spikes, J. D. Photochem. Photobiol. 1992, 55, 797-808. 237 Chapter 7 Conclusions and Future Work 7.1 General Remarks This chapter is intended to highlight both more successful and less successful areas of this thesis work, and to make suggestions for future investigations. For more detailed information, the reader is referred to the individual Chapters. In addition, general conclusions regarding the results of development of porphyrin-based anti-cancer agents are made. 7.2 Diarylporphyrins (Chapter 2) Although the preparation of diarylporphyrins and their precursors (e.g. bis(2-pyrrolylmethane), (3)) was notably improved over previously reported methodologies, the chemistry investigated in this area was not as successful as was anticipated. Five diarylporphyrins, three being new, were synthesized via acid catalyzed condensation of (3) and an aromatic aldehyde However, with the exception of DPhP, the yields were so low that only small amounts of these porphyrins could be generated practically, thus limiting their usefulness as precursors to anti-cancer agents. From DPhP, the previously reported DBrDPhZn1 was synthesized and used in several Pd-catalyzed cross-coupling reactions; however, only the cross-coupling reactions to produce TPhPZn (a trial reaction for this known compound) and DPhDVPZn (also previously reported)1 were unequivocal successes. This Pd-catalyzed cross-coupling methodology for derivitization of DBrDPhPZn shows promise for the introduction of other moieties, but the organotin precursors need to be studied and characterized more thoroughly. Br 3 DPhP DBrDPhPZn References on page 248 238 DPhDVPZn BDMEtDPhP The reactivity of DPhDVPZn was explored, but reactions with this complex generally gave very low yields and numerous side-products. However, once demetallated (DPhDVP), the vinyl groups were efficiently oxidized by T l m to produce BMEtDPhP; this was probably the most successful of the porphyrin-based reactions (Section 2.2.6.4). Using the amount of DPhDVPZn synthesized in this thesis work (~1 g) as a guide, BMEtDPhP would be readily available in quantities of several hundred milligrams. This, in principle, opens up the preparation of numerous new porphyrins using the chemistry described by Kenner et al.,2 Kahl et al.,3 and that described in Chapter 4 to produce the bis(aldehyde)-, bis(2-hydroxyethyl)-, bis(2-(2-nitroimidazol-l-yl)ethyl)-, bis(anilino)-, and other derivatives (Figure 7.1). However, given the decreased yields of multistep syntheses and the questionable utility of such derivatives as anti-cancer agents (other porphyrins look more promising, Sections 7.4 and 7.5), improved cancer therapeutic agents would not likely be the impetus for synthesizing these compounds. References on page 248 239 Figure 7.1. Potential Target Diarylporphyrins from BMEtDPhP (Based on Chemistry Described in Chapter 4). 7.3 Tetraarylporphyrins (Chapter 3) 7.3.1 Porphyrin Syntheses, Sulfonations and Metallations with Pt Condensation of pyrrole with one or two aldehydes in propionic acid (the Adler method) was used to synthesize several tetraarylporphyrins containing pyridyl or imidazolyl groups. The References on page 248 240 phenyl-pyridyl series was easiest to work with, whereas porphyrins containing imidazolyl groups were produced in low yield and were difficult to purify. Attempts to form porphyrins from heterocyclic, aromatic aldehydes via Lindsey conditions were generally unsuccessful. Water-soluble porphyrins (e.g. TSPhP, PyTrSPhP, OPyTrSPhP and c-BOPyBSPhP, none of these being shown here), were produced by sulfonation using standard conditions and no difficulties were encountered with the oxidopyridyl groups (Section 7.3.2). Some metallation reactions with Pt were pursued; TSPhP reacted with K^PtCLi to produce TSPhPPt, but similar conditions with PyTrSPhP led to porphyrin decomposition. The metallation of OPyTrSPhP is discussed below. OPyTrSPhP was evaluated in in vitro studies at BCCRC (see Section 7.5). 7.3.2 (Oxidopyridyl)porphyrins The N-oxidation of pyridylporphyrins proved to be a rich area of chemistry. Although the (oxidopyridyl)porphyrins (5 new compounds) were obtained in reasonable yield from their pyridylporphyrin precursors (often -50-60 %), the yield of the corresponding porphyrin N-oxides was low (7 new compounds, each < 5 % yield). Only three free-base porphyrin N-oxides have been reported in the literature (see Section 3.3.2), and thus the number of known porphyrin N-oxides has been tripled. Andrews et al. used hypofluorous acid or permaleic acid to produce octaethylporphyrin N-oxide in 64 and 68 % yields, respectively;4 thus, better yields of the tetraarylporphyrin N-oxides might be obtained using an oxidant, different to that used here (m-CPBA). Then, having sufficient quantities of these porphyrin N-oxides, their coordination chemistry could be explored. This might yield some interesting results, given that the oxidopyridyl moiety can also act as a ligand.5,6 On a related topic, the reaction product of K2P1CI4 with OPyTrSPhP was not conclusively identified. Metallation of the porphyrin definitely occurred, but coordination of the oxidopyridyl group to an external Pt-moiety seems likely. Some investigations into the coordination chemistry of (oxidopyridyl)porphyrins have been carried out in this laboratory by Pratt,7 but additional work is needed in this area. Finally, the bis-and tris-(oxidopyridyl)porphyrins (e.g. r-BOPyDPhP and References on page 248 241 TrOPyPhP) could be sulfonated to yield ?-BOPyBSPhP and TrOPySPhP, thus completing the series of sulfonated (oxidopyridyl)porphyrins. 9- 9-o- o-?-BOPyBSPhP TrOPySPhP 7.3.3 Tetra(nitroimidazolyl)porphyrins As suggested in Chapter 1, porphyrins incorporating nitroimidazoles might be useful anti-cancer agents. However, attempts to produce such a porphyrin via nitration of 2-TImP yielded mostly decomposition products. A potential route to tetrakis(nitroimidazol-2-yl)porphyrin is the acid-catalyzed condensation of l-methyl-4-nitro-2-imidazolecarboxaldehyde (reported by Smithen and Hardy)8 and pyrrole (Figure 7.2). However, as the yield of the tetrakis-imidazole porphyrins was generally low (e.g. ~4 % for 2-TTmP), this condensation might give even poorer yields because of the electron-withdrawing (i.e. deactivating) nature of the nitro group. 0 2 N Figure 7.2. Proposed Scheme for the Synthesis of a Tetrakis(nitroimidazolyl)porphyrin. References on page 248 242 7.3.4 Tirapazamine Chemistry As porphyrins incorporating hypoxia-selective cytotoxins (HSCs) might be effective anti-cancer agents (Chapter 1), the chemistry of tirapazamine was investigated. When tirapazamine is treated with phosgene (4 h at 90 °C in toluene), 21 is produced in 84% yield;9 however, substituting triphosgene (a crystalline material and thus easy to handle) in place of phosgene, 21 is produced in near quantitative yield in just a few minutes (Section 3.2.6.1). tirapazamine 21 TirapPhTrPhP Compound 21 reacts like an isocyanate and was found to react the most rapidly with amines; thus porphyrin-incorporation of tirapazamine was achieved by mixing 21 with an aminophenylporphyrin to yield TirapPhTrPhP (above). Reactions of 21 with other porphyrins were investigated, but were generally unsuccessful. In principle, the synthesis of numerous tirapazamine-porphyrin conjugates is possible using this chemistry (Chapter 4 and Section 7.5). Some limitations of the scope of this chemistry should be noted; currently, tirapazamine is available commercially but is quite expensive ($180 /100 mg, Sigma 1997) and the patent rights are held by Sinoft-Winthrop. In-house synthesis of the chemical is possible, but the first-reported methods10 yielded only small amounts of tirapazamine.11 Recent, more efficient methods require the use of Na 2NCN 9 which is, at present, unavailable commercially and its synthesis is achieved only with very vigorous conditions (i.e. molten Na at -300 °C under an Ar atmosphere with subsequent distillation of the Na from the mixture).12 References on page 248 243 7.4 Protoporphyrin IX (PIX)-Based Chemistry (Chapter 4) BDMEtPIXDME BOEtPIXDME The Tlm-oxidation product of PIXDME, BDMEtPIXDME, was the starting material for several areas of chemistry (Chapter 4); several new compounds (16, including regioisomers) and numerous known compounds were synthesized via this starting material. Selective acetal deprotection of BDMEtPIXDME was achieved using TFA (an improvement over literature procedures) to yield BOEtPIXDME, which was subsequently used in reduction, reductive amination, and Knoevenagel reactions. The reduction and subsequent reaction with SOBr2 yielded BBrEtPIXDME,2 which was then used in SN2 reactions to produce compounds such as BNImEtPIXDME (Section 4.2.10) and BPlEtPXDME (Section 4.2.13). Elimination products (i.e. re-formation the vinyl group) was common feature in these SN2 reactions. The free-acid derivative of BNImEtPIXDME, BNImEtPIX, was used in the in vitro assays (see below). BBrEtPIXDME BNImEtPIXDME BPlEtPIXDME References on page 248 244 B A n E t P r X D M E B F 5 E 1 P I X D M E Reductive amination was used to produce BAnEtPLXDME and BFsEtPIXDME, and reactions using other amines were also investigated (Section 4.2.3). Among unexpected results, the aldehyde moieties of BOEtPIXDME were easily reduced under reductive amination reactions using NaBFl3(CN), even at high pH, which complicated these reactions. As reductive amination remains an attractive method for introduction of novel functional groups, investigation of the reaction optimization (i.e. pH effects, and adjustment of the imine-amine equilibrium, see also Section 4.3.1.2) is warranted. Some suggestions for future PIX-based chemistry are described below. 7.5 Development of Porphyrin-based Anti-cancer Agents 7.5.1 Results of the In Vitro Assays at B C C R C (Chapter 6) and Cyclic Voltammetry Studies (Chapter 5) In general, the evaluated porphyrins incorporating heterocyclic N-oxides and nitroimidazoles (OPyTrSPhP, TirapPhTrPhP, and BNImEtPIX) and their 'reference' compounds (PyTrSPhP, APhTrPhP, and BHEtPIX) showed little in vitro activity. These compounds were not radiosensitizers or hypoxia-selective cytotoxins (HSCs) in mammalian (Chinese hamster ovary, CHO) cells, with the exception of TirapPhTrPhP which had a modest sensitivity enhancement ratio (SER) value (1.5). The results of photosensitizing experiments showed that OPyTrSPhP and References on page 248 245 PyTrSPhP were poor photosensitizers in vitro compared to Photofrin U®, and that the other evaluated porphyrins did not appreciably photosensitize CHO cells. Among the challenges faced in evaluating these porphyrins was the poor water-solubility of some of these compounds. Solubilization using Cremophor EL® emulsions was investigated, but this was unsuitable because of inherent Cremophor EL® toxicity and poor cellular accumulation of the porphyrins. Thus, liposomal formulations of BNImEtPLX, TirapPhTrPhP, and APhTrPhP were developed. The accumulation of these porphyrins by CHO cells was evaluated by first incubating the cells with the appropriate porphyrin-formulation and then cell-digestion with subsequent UV-Vis spectroscopy, or fluorescence microscopy. Although the UV-Vis method indicated reasonable cellular accumulation of all the evaluated porphyrins, fluorescence microscopy suggested that the liposome-formulated porphyrins were not incorporated into the cells. Some suggestions were made to explain these apparently contradictory results (Sections 6.3.5 to 6.3.7). The cyclic voltammetry studies (Chapter 5) showed that the nitrophenyl-, nitroimidazolyl-, and tirapazamine-substituted porphyrins had reduction potentials similar to those of known radiosensitizers and HSCs; thus the poor results in vitro were not likely related to the redox behavior of these porphyrins. The (oxidopyridyl)porphyrins do not warrant further investigation in this respect, as these moieties are too difficult to reduce. 7.5.2 Suggestions for Improved Porphyrin-Based Hypoxia-Selective Cytotoxins and Radiosensitizers Although the combination of PDT with HSCs has shown promise (Section 1.6), it remains to be proven that HSCs covalently bound to porphyrins would give improved results, as the photosensitizer concentrations used for PDT are generally much lower than those of the HSC. For example, Bremner etal. used a 1:20 ratio of sulfonated Al-phthalocyanine:RSU 1069 (a DNA-binding nitroimidazole) in a murine sarcoma tumor model.13 Nevertheless, the development of nitroimidazole- and tirapazamine-porphyrin conjugates remains of interest. References on page 248 246 O . T O H N + N H 2 N O 2 6-RSU 1069 tirapazamine The efficacy of nitroimidazolyl- or tirapazamine-porphyrin conjugates might be improved (for the in vitro evaluations at least) if more hydrophilic derivatives could be used. Recent work by Pandey et al. might provide the means to such an end;14 they described the electrophilic alkylamination of deuteroporphyrin using CH2=N(Me)2l with subsequent alkylation of the amino group leading to a dicationic, water-soluble porphyrin (see Figure 7.3). Deuteroporphyrin Figure 7.3. Synthesis of a Water-soluble Cationic Porphyrin from Deuteroporphyrin. Similar chemistry in combination with peptide coupling chemistry (Chapter 4) and tirapazamine chemistry (Chapter 3) might be used to prepare water-soluble nitroimidazole- and tirapazamine porphyrins (e.g., 37 and 38, p 247). Alkylation of the imidazole and benzotriazine moieties might occur in the final step, but even the tertiary amines (i.e. before alkylation, not shown) might be soluble enough to be evaluated in vitro, as the amines would likely be protonated at physiological pH. References on page 248 247 7.5.3 Hypoxia-Imaging Agents and Pt-Porphyrins Much more work would be needed to develop a porphyrin-based hypoxia-imaging agent (Section 1.7.1). A fluoroalkyl-substituted porphyrin (BF5E1PLXDME) was produced by reductive amination using C2F5CH2NH2 and BOEtPIXDME, but large quantities of this porphyrin are not readily available via this route (Section 4.3.1.2). Perhaps other fluoroalkyl-substituted porphyrins (e.g. 39 and 40, synthesized via chemistry described above) would be effective hypoxia imaging agents. If fluorescence from the porphyrin were intense enough (i.e. not quenched because of aggregation) the porphyrin core could act as the fluorescent marker, in which case the need for monoclonal antibodies and the fluoroalkyl groups would be eliminated (Section 1.7.1). Numerous platinum porphyrins with externally coordinated Pt-moieties, including ones based on PIX and tetraarylporphyrins, have been reported by Brunner et al.15,16 Their work included in vitro and in vivo evaluation of their compounds and effectively covered this area of interest, as described in the Introduction (Section 1.7.2). 7.6 Conclusion A broad range of chemistry was used and developed in this thesis work, including classical porphyrin syntheses and substituent manipulation, general organic chemistry, Pd-catalyzed cross coupling reactions, some organometallic chemistry and coordination chemistry. Also, significant steps were taken in the areas of development and evaluation of porphyrin-based anti-cancer agents. Much was learned in the course of writing this dissertation, not the least of which is the value of knowing the literature well when approaching the chemistry (or biology) laboratory bench. References on page 248 248 References for Chapter 7 (1) DiMagno, S. G.; Lin, V. S.-Y.; Therien, M . J. J. Org. Chem. 1993, 58, 5983-5993. (2) Kenner, G. W.; McCombie, S. W.; Smith, K. M . Liebigs Ann. Chem. 1973, 1329-1338. (3) Kahl, S. B.; Schaeck, J. J.; Koo, M.-S. J. Org. Chem. 1997, 62, 1875-1880. (4) Andrews, L. E.; Bonnet, R.; Ridge, R.; Appelman, E. H. J. Chem. Soc. Perkin Trans. I 1983,1983, 103-107. (5) Garvey, R. G.; Nelson, J. H.; Ragsdale, R. O. Coord. Chem. Rev. 1968, 3, 375-407. (6) Orchin, M. ; Schmidt, P. J. Coord. Chem. Rev. 1968, 3, 345-373. (7) Pratt, R. C. University of British Columbia, B.Sc. Undergraduate Thesis, 1998. (8) Smithen, C. E.; Hardy, C. R. In Advanced Topics on Radiosensitizers of Hypoxic Cells; A . Breccia, C. Rimondi and G. E. Adams, Eds.; Plenum: New York, 1982; pp 1-47. (9) Seng, F.; Ley, K. Angew. Chem. Int. Ed. Eng. 1972,11, 1009-1010. (10) Mason, J. C ; Tennant, G. J. Chem. Soc. B 1970, 911-916. (11) Baird, I. R. University of British Columbia, personal communication, 1997. (12) Harper, A.; Hubberstey, P. J. Chem. Res. (S) 1989, 194-195. (13) Bremner, J. C. M . ; Adams, G. E.; Pearson, J. K.; Sansom, J. M. ; Stratford, I. J.; Bedwell, L; Bown, S. G.; MacRobert, A. J.; Phillips, D. Br. J. Cancer 1992, 66, 1070-1076. (14) Pandey, R. K.; Shiau, F.-Y.; Smith, N. W.; Dougherty, T. J.; Smith, K. M . Tetrahedron 1992, 48, 7591-7600. (15) Brunner, H.; Maiterth, F.; Treittinger, B. Chem. Ber. 1994,127, 2141-2149. (16) Brunner, H.; Obermeier, H.; Szeimies, R.-M. Chem. Ber. 1995,128, 173-181. 249 Appendix Table A . l . Experimental Details for X-ray Crystal Structure of DBrDPhPZn(THF)2 (Section 2.3.4). Table A. 1. A. Crystal Data. Empirical Formula C4oH23Br2N402Zn Formula Weight 827.92 Crystal Color, Habit purple, prism Crystal Dimensions 0.20 x 0.40 x 0.45 mm Crystal System monoclinic Lattice Type Primitive No. of Reflections Used for Unit Cell 25 (91.6 - 107.9°) Determination (20 range) Omega Scan Peak Width at Half-height 0.38° Lattice Parameters a = 10.905(1) A b- 13.515(1) A c = 12.640(1) A P= 112.993(7)° V = 1714.9(3) A3 Space Group P2i/c (#14) Z value 2 1.603 c/cm3 FfjOO 836.00 |l(CuK(X) 40.88 cm-1 250 Table A. L B . Intensity Measurements. Diffractometer Rigaku AFC6S Radiation CuKoc (X = 1.54178 A) graphite monochromated Take off Angle 6.0° Detector Aperture 6.0 mm horizontal 6.0 mm vertical Crystal to Detector Distance 285 mm Temperature 21 °C Scan Type co-20 Scan Rate 32° / min (in co) (up to 9 scans) Scan Width (0.94 + 0.20 tan 0)° 2©max 155° No. of Reflections Measured Total: 3971 Unique: 3791 (R,nr = 0.031) Corrections Lorentz-polarization Absorption (trans, factors: 0.646 - 1.000) Decay (38.61 % dechne) Secondary Extinction (coefficient: 5.29(15) x 10"6) Table A.l .C. Structure Solution and Refinement. Structure Solution Direct Methods Refinement FuU-matrix least-squares Function Minimized Zco(IFol - IFCI)2 Least Squares Weights unit weights p-factor 0.000 Anomalous Dispersion All non-hydrogen atoms No. Observations (I>3a(I)) 2667 No. Variables 224 Reflection/Parameter Ratio 11.91 Residuals: R; R w 0.035 ; 0.035 Goodness of Fit Indicator 1.13 Max Shift / Error in Final Cycle 0.001 Maximum peak in Final Diff. Map 0.25 e- / A3 Minimum peak in Final Diff. Map -0.49 e- / A3 25 1 Table A.2. Atomic Coordinates for DBrDPhPZn(THF)2 (Section 2.3.4). atom X y z Br(l) 0.69901(5) 0.13694(3) 0.53957(4) Zn(l) 0.50000 0.50000 0.50000 0(1) 0.6815(3) 0.5391(2) 0.6774(2) N(l) 0.6322(3) 0.4304(2) 0.4439(2) N(2) 0.4737(3) 0.3703(2) 0.5730(2) C(l) 0.6999(3) 0.4734(3) 0.3846(3) C(2) 0.7854(4) 0.4000(3) 0.3661(3) C(3) 0.7690(4) 0.3151(3) 0.4132(3) C(4) 0.6729(3) 0.3340(3) 0.4633(3) C(5) 0.6273(3) 0.2671(3) 0.5241(3) C(6) 0.5387(3) 0.2827(3) 0.5777(3) C(7) 0.5020(4) 0.2131(3) 0.6458(3) C(8) 0.4156(4) 0.2584(3) 0.6819(3) C(9) 0.3976(3) 0.3573(3) 0.6364(3) C(10) 0.3133(3) 0.4289(3) 0.6535(3) C(l l ) 0.2298(3) 0.3956(3) 0.7168(3) C(12) 0.2850(4) 0.3818(3) 0.8342(3) C(13) 0.2071(5) 0.3457(4) 0.8903(4) C(14) 0.0762(5) 0.3253(3) 0.8310(4) C(15) 0.0203(4) 0.3395(4) 0.7149(4) C(16) 0.0963(4) 0.3750(3) 0.6578(3) C(17) 0.7035(5) 0.4876(4) 0.7801(4) C(18) 0.8370(5) 0.5122(4) 0.8641(4) C(19) 0.8966(5) 0.5777(4) 0.8034(5) C(20) 0.8029(5) 0.5726(4) 0.6794(4) 252 Table A.3. Bond Lengths (A) for DBrDPhPZn(THF)2 (Section 2.3.4). atom atom distance atom atom distance Br(l) C(5) 1.904(4) Zn(l) 0(1) 2.400(3) Zn(l) N(l) 2.065(3) Zn(l) N(2) 2.052(3) O(l) C(17) 1.408(5) 0(1) C(20) 1.390(5) N(l) C(l) 1.370(4) N(l) C(4) 1.367(4) N(2) C(6) 1.370(4) N(2) C(9) 1.372(4) C(l) C(2) 1.442(5) C(l) C(10)* 1.393(5) C(2) C(3) 1.337(5) C(3) C(4) 1.442(5) C(4) C(5) 1.398(5) C(5) C(6) 1.396(5) C(6) C(7) 1.434(5) C(7) C(8) 1.344(5) C(8) C(9) 1.438(5) C(9) C(10) 1.408(5) C(10) C(l l ) 1.498(5) C(l l ) C(12) 1.379(5) C( l l ) C(16) 1.380(5) C(12) C(13) 1.3.91(6) C(13) C(14) 1.357(6) C(14) C(15) 1.365(6) C(15) C(16) 1.380(6) C(17) C(18) 1.466(6) C(18) C(19) 1.477(7) C(19) C(20) 1.501(6) Symmetry operation: 1-x, 1-y, 1-z. 253 Table A.4. Bond Angles (°) for DBrDPhZnP(THF)2 (Section 2.3.4). atom atom atom angle atom atom atom angle 0(1) Zn(l) 0(1)* 180.0 0(1) Zn(l) N(l) 89.4(1) 0(1) Zn(l) N(l)* 90.6(1) 0(1) Zn(l) N(2) 88.7(1) 0(1) Zn(l) N(2)* 91.3(1) N(l) Zn(l) N(l)* 180.0 N(l) Zn(l) N(2) 90.2(1) N(l) Zn(l) N(2)* 89.8(1) N(2) Zn(l) N(2)* 180.0 Zn(l) 0(1) C(17) 122.2(3) Zn(l) O(l) C(20) 121.4(3) C(17) 0(1) C(20) 109.0 Zn(l) N(l) C(l) 126.3(2) Zn(l) N(l) C(4) 126.5(2) C(l) N(l) C(4) 107.2(3) Zn(l) N(2) C(6) 126.6(2) Zn(l) N(2) C(9) 126.5(2) C(6) N(2) G(9) 106.5(3) N(l) C(l) C(2) 108.5(3) N(l) C(l) C(10)* 125.9(3) C(2) C(l) C(10)* 125.6(3) C(l) C(2) C(3) 108.1(3) C(2) C(3) C(4) 106.9(3) N(l) C(4) C(3) 109.4(3) N(l) C(4) C(5) 123.6(3) C(3) C(4) C(5) 127.0(3) Br(l) C(5) C(4) 115.7(2) Br(l) C(5) C(6) 115.1(3) C(4) C(5) C(6) 129.2(3) N(2) C(6) C(5) 123.7(3) N(2) C(6) C(7) 109.4(3) C(5) C(6) C(7) 126.9(3) C(6) C(7) C(8) 107.6(3) C(7) C(8) C(9) 107.0(3) N(2) C(9) C(8) 109.5(3) N(2) C(9) C(10) 125.5(3) C(8) C(9) C(10) 125.0(3) C(l)* C(10) C(9) 125.7(3) C(l)* C(10) C(l l ) 117.9(3) C(9) C(10) C( l l ) 116.4(3) C(10) C( l l ) C(12) 121.0(3) C(10) C(l l ) C(16) 120.3(3) C(12) C( l l ) C(16) 118.7(3) C(l l ) C(12) C(13) 119.9(4) C(12) C(13) C(14) 120.7(4) C(13) C(14) C(15) 119.8(4) C(14) C(15) C(16) 120.2(4) C(l l ) C(16) C(15) 120.6(4) Symmetry operation: l-x, 1-y, l-z. 254 Table A.5. Experimental Details for X-ray Crystal Structure of OPyTrPhP (Section 3.3.2.2). Table A.5.A. Crystal Data. Empirical Formula C 4 3 H 2 9 N 5 O Formula Weight 631.73 Crystal Color, Habit purple, irregular Crystal Dimensions 0.20 x 0.35 x 0.45 mm Crystal System tetragonal Lattice Type I-centered No. of Reflections Used for Unit Cell 25 (55.7 - 93.2°) Determination (20 range) Omega Scan Peak Width at Half-height 0.44° Lattice Parameters a = 15.174(1) A c = 13.709(3) A V = 3156.3(9) A3 Space Group I42d (#122) Z value 4 1.329 g/cm3 FfJOO 1320.00 |l(CuKa) 6.41 cm - 1 255 Table A.5.B. Intensity Measurements. Diffractometer Rigaku AFC6S Radiation CuKa (k = 1.54178 A)graphite monochromated Take off Angle 6.0° Detector Aperture 6.0 mm horizontal 6.0 mm vertical Crystal to Detector Distance 285 mm Voltage, Current 45 kV, 25 mA Temperature 21 °C Scan Type co-20 Scan Rate 32° / min (in co) (up to 9 scans) Scan Width (1.00 - 0.20 tan 0) ° 2©max 155° No. of Reflections Measured Total: 998 Corrections Lorentz-polarization Absorption (trans, factors: 0.867 - 1.000) Secondary Extinction (coefficient: 1.54(5) x 10"6) Table A.5.C. Structure Solution and Refinement. Structure Solution Direct Methods (SIR92) Refinement Full-matrix least-squares Function Minimized Zco(IF0l - IFCI)2 Least Squares Weights CO = (l/(o2(F0) = [ac2(F0) + (p2/4)F02]-l p-factor 0.000 Anomalous Dispersion AU non-hydrogen atoms No. Observations (I>3a(I)) 685 No. Variables 143 Reflection/Parameter Ratio 4.79 Residuals: R ;R W 0.031 ; 0.026 Goodness of Fit Indicator 2.20 Max Shift / Error in Final Cycle 0.0008 Maximum peak in Final Diff. Map 0.09 e- / A3 Minimum peak in Final Diff. Map -0.07 e- / A3 256 Table A.6. Atomic Coordinates for OPyTrPhP (Section 3.3.2.2). atom X y z 0(1) 0.5590(4) 0.3378(6) 0.1417(9) N(l) 0.0680(1) 0.3847(1) 0.2468(2) N(2) 0.4789 0.3672 0.1491 C(l) 0.0344(2) 0.3027(2) 0.2670(2) C(2) 0.1059(2) 0.2406(2) 0.2692(2) C(3) 0.1804(2) 0.2851(2) 0.2478(3) C(4) 0.1574(2) 0.3759(2) 0.2349(2) C(5) 0.2170(2) 0.4450(2) 0.2217(2) C(6) 0.3096(2) 0.4202(2) 0.1961(2) C(7) 0.3305(2) 0.3948(2) 0.1022(3) C(8) 0.4147(2) 0.3695(3) 0.0789(3) C(9) 0.4789(2) 0.3672(2) 0.1491(3) C(10) 0.4600(2) 0.3928(3) 0.2407(3) C( l l ) 0.3762(2) 0.4195(3) 0.2645(3) Table A.7. Bond Lengths (A) for OPyTrPhP (Section 3.3.2.2). atom atom distance atom atom distance O(l) N(2) 1.299(7) N(l) C(l) 1.372(3) N(l) C(4) 1.374(3) N(2) C(8) 1.369(4) N(2) C(10) 1.346(4) C(l) C(2) 1.438(4) C(l) C(5)' 1.398(4) C(2) C(3) 1.349(4) C(3) C(4) 1.433(4) C(4) C(5) 1.396(3) C(5) C(6) 1.496(4) C(6) C(7) 1.381(4) C(6) C(l l ) 1.378(4) C(7) C(8) 1.372(4) C(9) C(10) 1.346(5) C(10) C(l l ) 1.375(4) *Symmetry operations: (') - 1/2 +y, 1/2 - x, 1/2 - z(") 1/2 - y, 111 + x, 1/2 z. 257 Table A.8. Bond Angles (°)* for OPyTrPhP (Section 3.3.2.2). atom atom atom angle atom atom atom angle C(l) N(l) C(4) 107.7(2) 0(1) N(2) C(8) 128.3(6) 0(1) N(2) C(10) 111.8(5) C(8) N(2) C(10) 119.8(2) N(l) C(l) C(2) 108.6(2) N(l) C(l) C(5)' 125.2(2) C(2) C(l) C(5)' 126.1(2) C(l) C(2) C(3) 107.4(3) C(2) C(3) C(4) 107.7(2) N(l) C(4) C(3) 108.6(2) N(l) C(4) C(5) 125.7(2) C(3) C(4) C(5) 125.5(2) C(l) C(5) C(4) 125.1(2) C(l) C(5) C(6) 118.1(2) C(4) C(5) C(6) 116.8(2) C(5) C(6) C(7) 120.3(3) C(5) C(6) C(l l ) 122.1(3) C(7) C(6) C(l l ) 117.7(3) C(6) C(7) C(8) 120.6(3) C(9) C(8) C(7) 120.4(3) C(8) C(9) C(10) 119.8(3) C(9) C(10) C( l l ) 120.2(3) C(6) C( l l ) C(10) 121.3(4) *Symmetry operations: (') - 1/2 +y, 1/2 - x, 1/2 - z(") 1/2 - y, 1/2 + x, 111 - z. 258 PPH Figure A . l . !H-2D-C0SY Spectrum of OPyTrSPhP (Section 3.2.4.3). 259 260 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061519/manifest

Comment

Related Items